Quasispecies: Concept and Implications for Virology (Current Topics in Microbiology and Immunology)

299 Current Topics in Microbiology and Immunology

Editors R.W. Compans, Atlanta/Georgia M.D. Cooper, Birmingham/Alabama T. Honjo, Kyoto · H. Koprowski, Philadelphia/Pennsylvania F. Melchers, Basel · M.B.A. Oldstone, La Jolla/California S. Olsnes, Oslo · M. Potter, Bethesda/Maryland P.K. Vogt, La Jolla/California · H. Wagner, Munich

E. Domingo (Ed.)

Quasispecies: Concept and Implications for Virology

With 44 Figures and 7 Tables

123

Esteban Domingo Centro de Biología Molecular “Servero Ochoa” (CSIC-UAM) Universidad Autónoma de Madrid Cantoblanco 28049 Madrid Spain e-mail: [email protected]

Cover illustration: Amino acid residues in the parvovirus MVMi capsid selected during evolutionary processes in mice. Highlighted in colours are residues of the spike whose replacement conferred the MAR phenotype (red), residues involved in primary receptor recognition (green), and residues conferring hematotropism (yellow). All other capsid residues are colored blue and shown as wireframe, except for the white-space-filling model of the residues defining the spike at the three-fold axis of symmetry.The structure of MVMi viral capsid is shown by the program RasMol (Sayle and Milner-White 1995) and the MVMi coordinates (1MVM) deposited in the PDB (Agbandje-McKenna et al. 1998).

Library of Congress Catalog Number 72-152360 ISSN 0070-217X ISBN-10 3-540-26395-0 Springer Berlin Heidelberg New York ISBN-13 978-3-540-26395-1 Springer Berlin Heidelberg New York This work is subject to copyright. All rights reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilm or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September, 9, 1965, in its current version, and permission for use must always be obtained from Springer-Verlag. Violations are liable for prosecution under the German Copyright Law. Springer is a part of Springer Science+Business Media springeronline.com © Springer-Verlag Berlin Heidelberg 2006 Printed in Germany The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Product liability: The publisher cannot guarantee the accuracy of any information about dosage and application contained in this book. In every individual case the user must check such information by consulting the relevant literature. Editor: Simon Rallison, Heidelberg Desk editor: Anne Clauss, Heidelberg Production editor: Nadja Kroke, Leipzig Cover design: design & production GmbH, Heidelberg Typesetting: LE-TEX Jelonek, Schmidt & Vöckler GbR, Leipzig Printed on acid-free paper SPIN 11501312 27/3150/YL – 5 4 3 2 1 0

Dedication

Many observations on the great potential for phenotypic change of RNA viruses were made during the twentieth century, and John Holland reviews them in the closing chapter of this volume. There is a very remarkable precedent that concerns the noted Catalan virologist Jordi Casals, a pioneer of virology research who sadly passed away last year. Born Jordi Casals i Ariet in Viladrau, Girona on May 15, 1911 he died in New York City on February 10, 2004 after a productive and distinguished career in the United States. Jordi Casals left Spain in 1936 at the onset of the civil war and occupied positions at Rockefeller Institute for Medical Research in New York, Cornell University Medical College, Rockefeller Foundation, Yale University, and Mount Sinai School of Medicine (see a bibliographical note by Charles Calisher (2004) Arch Virol 149:1264-1266). A survivor of Lassa fever, he was a devoted, meticulous and observant scientist interested in virus classification. He was particularly concerned by confounding antigenic cross-reactions among viruses and viral diversity within a virus group. In a letter addressed to Charles Calisher in 1968 Jordi Casals wrote “A virus or species is a cluster of different individualities grouped around and resembling a prototype or model, rather than a number of strains all identical with a prototype.” (I am indebted to Charles Calisher for sharing this information). Jordi Casals was honored at the recent annual meeting of the Virology Group of the Catalan Society of Biology (Societat Catalana de Biologia) held in Barcelona on 25 October 2004, with a biographical note read by Dr. Albert Bosch. This volume on quasispecies is dedicated to his memory. Esteban Domingo

Preface

High mutation rates and quasispecies dynamics are essential features of RNA viruses. Continuous genetic variation and selection of virus subpopulations in the course of RNA virus replication are intimately related to viral disease mechanisms. Experience has taught that the adaptive potential of viruses must be taken into consideration in designing preventive and therapeutic antiviral strategies. The central topics of this volume are the origins of the quasispecies concept, and the implications of quasispecies dynamics for viral populations. It includes chapters that emphasize general concepts (quasispecies, sequence space, fitness, error catastrophe, lethal defection, adaptive responses, population bottlenecks, etc.) and chapters that deal with population dynamics of specific viruses such as Picornaviruses, Pestiviruses, Arenaviruses, Arboviruses, human immunodeficiency virus, plant viruses and some DNA viruses that display features of RNA genetics. In particular, implications of quasispecies dynamics in vivo are dealt with in several chapters. I thank all authors who have contributed their time and expertise to produce a markedly transdisciplinary volume that should provide a stimulus for future research. Hopefully, as sometimes happens with mutant swarms, the entire volume will be more “fit” than the sum of its chapters! I am also deeply indebted to Dr. Michael B.A. Oldstone for his generous invitation to serve as guest editor for a volume of this prestigious CTMI series. The Springer-Verlag staff and Lucia Horrillo of Centro de Biología Molecular “Severo Ochoa” were decisive to successfully completing this volume. Madrid, April 2005

Esteban Domingo

List of Contents

What Is a Quasispecies? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. K. Biebricher and M. Eigen

1

Quasispecies in Time-Dependent Environments . . . . . . . . . . . . . . . . . . . . . . . 33 C. O. Wilke, R. Forster, and I. S. Novella Viruses as Quasispecies: Biological Implications . . . . . . . . . . . . . . . . . . . . . . . 51 E. Domingo, V. Martín, C. Perales, A. Grande-Pérez, J. García-Arriaza, and A. Arias Virus Fitness: Concept, Quantification, and Application to HIV Population Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 M. E. Quiñones-Mateu and E. J. Arts Population Bottlenecks in Quasispecies Dynamics . . . . . . . . . . . . . . . . . . . . . . 141 C. Escarmís, E. Lázaro, and S. C. Manrubia Evolutionary Dynamics of HIV-1 and the Control of AIDS . . . . . . . . . . . . . . . . 171 J. I. Mullins and M. A. Jensen Evolution of Virulence in Picornaviruses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 S. Tracy, N. M. Chapman, K. M. Drescher, K. Kono, and W. Tapprich Molecular Mechanisms of Poliovirus Variation and Evolution . . . . . . . . . . . . . . 211 V. I. Agol Hepatitis C Virus Population Dynamics During Infection . . . . . . . . . . . . . . . . . 261 J.-M. Pawlotsky Evolutionary Influences in Arboviral Disease . . . . . . . . . . . . . . . . . . . . . . . . . 285 S. C. Weaver Arenavirus Diversity and Evolution:Quasispecies In Vivo . . . . . . . . . . . . . . . . . 315 N. Sevilla and J. C. de la Torre

X

List of Contents

Mutant Clouds and Occupation of Sequence Space in Plant RNA Viruses . . . . . . 337 M. J. Roossinck and W. L. Schneider Parvovirus Variation for Disease: A Difference with RNA Viruses? . . . . . . . . . . . 349 A. López-Bueno, L. P. Villarreal, and J. M. Almendral Transitions in Understanding of RNA Viruses: A Historical Perspective . . . . . . . 371 J. J. Holland Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 403

List of Contributors (Addresses stated at the beginning of respective chapters)

Agol, V. I. 211 Almendral, J. M. 349 Arias, A. 51 Arts, E. J. 83 Biebricher, C. K. 1 Chapman, N. M. 193 de la Torre, J. C. 315 Domingo, E. 51 Drescher, K. M. 193

López-Bueno, A. 349 Manrubia, S. C. 141 Martín, V. 51 Mullins, J. I. 171 Novella, I. S. 33 Pawlotsky, J.-M. 261 Perales, C. 51 Quiñones-Mateu, M. E. 83

Eigen, M. 1 Escarmís, C. 141 Forster, R. 33 García-Arriaza, J. 51 Grande-Pérez, A. 51

Roossinck, M. J. 337 Schneider, W. L. 337 Sevilla, N. 315

Holland, J. J. 371

Tapprich, W. 193 Tracy, S. 193

Jensen, M. A. 171

Villarreal, L. P. 349

Kono, K. 193 Lázaro, E. 141

Weaver, S. C. 285 Wilke, C. O. 33

CTMI (2006) 299:1–31 c Springer-Verlag Berlin Heidelberg 2006


What Is a Quasispecies? C. K. Biebricher (u) · M. Eigen Max Planck Institute for Biophysical Chemistry, Am Fassberg, 37077 Göttingen, Germany [email protected]

1

Species and Quasispecies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2

2

Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2

3

Selection of the Fittest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5

4

Mutation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

7

5

Sequence Heterogeneity in Virus Populations . . . . . . . . . . . . . . . . . . . . 11

6

The Sequence Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

7

The Quasispecies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

8

The Error Threshold . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

9 9.1 9.2 9.3

Evolutionary Biotechnology . . . . . . . . . Principles . . . . . . . . . . . . . . . . . . . . . . Protein Design . . . . . . . . . . . . . . . . . . . Selection of nucleic acids with a function

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

21 21 23 24

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Abstract The concept of the quasispecies as a society formed from a clone of an asexually reproducing organism is reviewed. A broad spectrum of mutants is generated that compete one with another. Eventually a steady state is formed where each mutant type is represented according to its fitness and its formation by mutation. This quasispecies has a defined wild type sequence, which is the weighted average of all genotypes present. The quasispecies concept has been shown to affect the pathway of evolution and has been studied on RNA viruses which have a particularly high mutation rate. They (and possibly the majority of other species) operate close to the error threshold that allows maximum exploration of sequence space while conserving the information content of the genotype. The consequences of the quasispecies concept for the new ‘evolutionary technology’ are discussed.

2

C. K. Biebricher · M. Eigen

1 Species and Quasispecies The taxonomic classification of living organisms by Carl von Linné (1707– 1778) was a milestone in biology. According to the view of his time, the living world was divided into different species that preexisted since the creation of the world and would persist until its end. Linné determined the systematic kinship of species by similarities in their anatomical build. A species was considered a society of individuals that are able to generate fertile offspring. This definition remained valid after the view of invariable species was shattered: Since Darwin it has been accepted that species can go extinct and new ones are formed by diversification. In the first half of the twentieth century, this view took on a theoretical fundament with the Mendelian view of biological species and the Neodarwinistic synthetic theory. With the advent of the molecular biology in the second half of the twentieth century, interest focussed on organisms that have a much simpler, ‘vegetative’ propagation mechanism, like Prokarya. Obviously, the biological concept of species stated above does not apply to them, even less to forms of life such as viruses that have no cellular organisms. There is no gene pool to select from and no continuous shuffling of gene alleles. Nevertheless, these forms of life have been classified with reasonable results. Obviously, the reproducing individuals in prokaryotic taxons and virus populations must lead to a society that shares many properties with the ‘classical’ species; it is called a ‘quasispecies’. Its emergence and its properties are described in this article.

2 Growth The purpose of a theory is not to describe how processes occur in nature in detail, but rather to understand why certain regularities can be observed. Theory is no alternative to experiment, but theory and experiment support one another, theory by interpreting experimental results and suggesting further meaningful experiments, which then can be used to refine the theory. However, theory needs reduction to the most important parameters while neglecting less important contributions. It is necessary to find suitable experimental conditions where complexity is low enough to verify theoretical predictions. The different reproduction mechanisms among the superkingdoms in nature cause difficulties; hence one must find the fundamental property of reproduction. Fisher [85], in deriving the laws of population dynamics of

What Is a Quasispecies?

3

a Mendelian population, was fully aware of this difficulty. He assumed that the population contains a sufficient number of males to fertilize all females, regarded only the female part of the population and obtained Malthusian, i.e. exponential, growth. The potential for exponential growth is an inherent property of life, even though it is rarely observed in nature because growth is rapidly limited by a shortage of resources. Indeed, processes such as leavening, where exponential spread can be observed, have always served as a metaphor for life. Exponential growth can be easily observed in the superkingdom of Prokarya. For further considerations, horizontal gene transfer phenomena such as transformation, infection, transfection, and conjugation should be disregarded. An experimental system that allows the study of exponential growth is the growth of a bacterial culture in a defined nutrient medium under controlled conditions. Bacteria grow by metabolizing nutrients from the medium, and divide after a certain time τ to two daughter cells. While the τ values vary from one bacterium to another, the average τ values for large populations of a species can be determined with high precision; the population density can be preferentially derived from macroscopic properties of the population, e.g. by turbidity measurements. After the time t, the initial concentration of bacteria has grown to c(t) = c(t

= 0) · eAt

(1)

where the growth rate A = ln 2t |τ (Table 1 lists the symbols for the parameters used in this article). Dynamical processes are best described by differential equations. The rate of population increase is directly proportional to the number of parental cells and their specific growth rate parameter: dc|dt

= Ac .

(2)

In experimental measurements, this equation is only strictly observed at large dilution. At higher concentrations, the consumption of resources, in particular of oxygen, slow down bacterial growth until the growth curve eventually levels in to a maximal concentration. While this behaviour can be described by the logistical equation, we shall use for our arguments the simple form. A specially designed apparatus, the chemostat, was invented to avoid large concentration increases by pumping in fresh medium at the constant rate Φ and removing a balancing volume of bacterial culture [93]. The dynamical equation then becomes dc|dt

= (A − Φ)c .

(3)

If A − Φ > 0, the population increases, if A − Φ < 0, the population decreases. When A − Φ = 0, the bacterial concentration should be constant,

4

C. K. Biebricher · M. Eigen

Table 1 Parameters Growth rate parameter of type i
¯ = k Ak xk ) Average growth rate parameter of the population (A Mortality or decomposition rate parameter of type i Excess reproduction rate of type i (Ei = Ai − Di )
Average excess reproduction rate of the population (E¯ = k Ek xk ) Number of generations Probability of precise reproduction of sequence i (Qii = q¯ νi ) Probability of producing type i while replicating sequence k Rate parameter of precise (excess) production of sequence i: (Wii = Qii Ai − Di ) Rate parameter of production of sequence i by erroneous copying of sequence k (Wik = Qik Ak ) ci Population density of type i dik The Hamming distance between two sequences i and k, i.e., the number of positions at which both genomes differ. q¯ Average fidelity of single digit reproduction (insertion of correct nucleotide) 1 − q¯ Probability of single error production
xi Fraction of type i in the total population (type frequency) (xi = ci | k ck ) x˜i type frequency of i in the evolution steady state F i Fitness value of species i; Fi = Ai |Am εi Average error rate per sequence i (εi = νi (1 − q¯ )) νi Genome length of type i σ¯ m Average superiority of master sequence over competitor mutants (¯σm = Am | (Dm + E¯ k
=m ), requiring σm >1) τ Duplication time of a bacterium Φ Flux rate of growth medium in reactor Ai ¯ A Di Ei E¯ N Qii Qik Wii Wik

but measurements show irreproducible changes in the bacterial concentration. Experiments with finite populations never observe deterministic laws exactly: inevitable statistical fluctuations of the A values lead to concentration fluctuations that are not compensated because A is independent of the concentration. Another experimental device, the turbidostat, keeps the turbidity of the bacterial culture and thus the bacterial concentration constant. A nonbiotic system of studying exponential growth is RNA replication [111] in a cell-free system, which is even capable of Darwinian evolution [86]. An RNA strand serving as template replicates, producing a complementary replica or minus strand, which can itself replicate again to produce plus strands. This cross-catalysis leads to exponential growth of the RNA [8, 9]. Direct measurement of the exponential growth is difficult,

What Is a Quasispecies?

5

Fig. 1 Growth profile of replicating RNA [121]. A mixture containing 0.75 mM ATP, CTP, GTP and UTP, 1 µM RNA polymerase from Escherichia coli and 3 µM thiazole orange was inoculated with RNA species EcorpG at different concentrations and the fluorescence of the probes was measured in a multichannel fluorimeter at an excitation wavelength of 488 nm and an emission wavelength of 515 mm.

because exponential growth requires that the replicase enzyme that catalyses the replication is present in large excess of the growing RNA. As in measuring the bacterial population with turbidity, only populations above a boundary give a measurable signal. The solution is that one measures growth at different initial concentrations (Fig. 1). If the population is diluted by the factor Fdil , the growth curve appears to be shifted on the time axis by the difference ∆t and one calculates the replication rate to be A = ln Fdil |∆t [9, 121] (Fig. 1).

3 Selection of the Fittest Populations are rarely composed of absolutely identical individuals, but contain usually types that can be distinguished by certain criteria. In the past, the criteria were obtained by visual inspection, e.g. Mendel distinguished the types of his pea population by the color of the flowers and the shape

6

C. K. Biebricher · M. Eigen

of the seeds. In the first half of the twentieth century, a large number of different variants of the fruit fly Drosophila melanogaster were investigated that were distinguished by the visual differences in the appearance at the larval or imago stages. Bacterial types can be distinguished by inspection of the colonies they form on solid nutrient media. Types classified by such criteria are called phenotypes. In contrast to our lack of quantitative predictions of evolutionary events, the molecular basis of transmitting phenotypic traits to offspring is well understood: The information needed for morphogenesis and function is encoded into the genotype, the nucleotide sequence in each organism’s genome. The key step in reproduction is replication of the genome. The enormously complex process of decoding the genotypic program resulting in a phenotype with certain properties is called expression. Let us assume that the different types in a population ignore one another and that each individuum grows as if it were alone, by observing the growth of different types, e.g. species, of bacteria in a medium that contains nutrients
in vast excess. Each of them is present at a relative concentration xi = ci | k ck and grows with the specific fecundity Ai . As in the previous examples, the total concentration of bacteria shall be kept constant by removing the
excess production of offspring caused by the average growth of the types A¯ = k Ak xk by appropriately diluting the medium. The relative population of each type then changes with the rate dxi (t)|dt

¯ = {Ai − A(t)}x i (t) .

(4)

¯ Note that equation (4) is nonlinear because A(t) is time-dependent. Types ¯ whose growth rate exceeds the average are enriched while types with Ai < A(t) are depleted in the population. It is obvious that the value for Ai of a type i is a good measure of its ‘fitness’ in Darwin’s sense. The enrichment of fitter types in the population raises the average growth rate, and more and more types fall below the average, until eventually the population reaches a maximum growth rate where only the master type m, the type with the maximal fitness, survives: ¯ → Amax A(t)

xm → 1

xi
=m → 0 .

(5)

Natural selection is thus indeed an immediate consequence of autocatalytic reproduction [36]. The following conditions must be fulfilled: 1. One type cannot be converted into another, e.g. because the types belong to different species. 2. The growth of one species is not influenced by the presence of another. 3. The selection proceeds far from chemical equilibrium.

What Is a Quasispecies?

7

In reality, populations do not change solely by differences in fecundity. Mortality rates Di also influence the composition since types with higher mortality also get depleted in the population. In this case, fitness is correlated with the excess reproduction rate Ei = Ai − Di . Not only maximizing the reproduction rate, but also minimizing the mortality rate is then important for selection. A practical application of this selection process is the preparation of pure bacterial cultures from a natural specimen by the use of selective media. In nature, the selection process we described is much more complicated. Fecundity and mortality rates are often not time-independent. Under realistic conditions, say in an ecosystem, the types compete for nutrients, or interact one with another by mutually influencing their fecundities and mortalities, most conspicuously in a predator–prey relationship. The fecundity and mortality rates of one species depend on the presence of other types and change with time. The situation is further complicated by variations of the environmental conditions. Calculation of selection values or fitness values is then very difficult. When working under defined conditions with the RNA replication system described above, it was possible to predict precise selection values. At low RNA concentrations, with nucleoside triphosphate substrates and replicase in vast excess, the RNA species with the highest growth rate is selected as described above. When the RNA concentration reaches the replicase concentration, however, the growth characteristic as well as the selection drastically changes. Since an RNA strand must bind a replicase molecule for replication, exponential increase of the RNA concentration is suddenly replaced by linear growth when the template concentration reaches the enzyme concentration. Free RNA strands, both of plus and minus polarity, accumulate, which can react one with the other by formation of a double helix that is unable to replicate. With the onset of the linear phase it is no longer the RNA species with the highest growth rate that wins, but rather the one with the highest rate of replicase binding. At still higher RNA concentrations, minimizing the loss rate by double strand formation also becomes important. Eventually, often a stable ecosystem is formed where different RNA species occupy constant parts of the population. The rather complicated but quite instructive calculation of the selection values can be found in the literature [12, 13].

4 Mutation Selection among preformed and invariable types alone does not suffice for Darwinian evolution. Evolution needs the formation of new types by a process

8

C. K. Biebricher · M. Eigen

called mutation. While in Mendelian populations the offspring usually is of a different type than the parents because the different parental gene alleles are reshuffled, ‘vegetatively’ growing populations such as bacteria and viruses have been assumed to produce identical offspring. New types are created by mutation, i.e. by misincorporation of nucleotides during the replication process. Evolutionary innovation is driven in all biological species by mutation. It is straightforward to introduce mutation rates Wji describing the rate of producing type j during reproduction of type i. However, in many cases mutation rates are too low to be realistically described by deterministic rate coefficients, even in large populations. When writing down an equation with rate coefficients, one has to be aware that the parameters are merely averaged probabilities. Before further conclusions can be drawn, one has to estimate whether there is a reasonable probability that the pertaining mutation will take place at all. How do we modify the equation to take mutation into account? Reproduction can proceed with fidelity, producing offspring of the parental type (Wii ) or it can erroneously produce offspring of another type (Wji , i
= j). The production rates of each type i are composed by the rates of fidelity reproduction of the same type and the rates of erroneous reproduction from other types [32]. The equation considering selection and mutation then is  Wik xk (t) (6) dxi |dt = {Wii − E¯ (t)}xi (t) + k
=i

where E¯ (t) is the average excess growth rate of the total population. If the mutation terms become negligible, i.e. at high fidelity reproduction, equation (6) is converted to the selection equation (5). The reproduction rate with fidelity can also be written as Wii = Qii Ai − Di , where Qik , the mutation probability is the probability of producing type i in a reproduction round of type k. When Di can be neglected, which can often be obtained by providing suitable environmental conditions, then equation (6) converts to  ¯ dxi |dt = {Qii Ai − A(t)}x Qik Ak xk (t) . (7) i (t) + k
=i


There we have two mutation terms: (i) the mutational gain k
=i Qik Ak xk , producing type
i by replication of other types, and (ii) the mutational loss (1 − Qii )Ai xi = i
=k Qki Ai xi , producing mutants in reproduction. How can Qik values be measured? With the sequencing machines available today, it is in principle possible to screen the genotypes of the offspring of a single reproduction cycle, but, since mutations are very rare, millions of sequence determinations would be needed to get values with some statistical significance. In the past, one had to work with phenotypic markers.

What Is a Quasispecies?

9

The challenge is determining a small number of mutants in a vast excess of wild type. Luria and Delbrück [84] have invented an ingenious method to measure mutations by observing the formation of mutants resistant to phage T1 (type r) in a sensitive wild type (type s) population. Since the wild type population has to be created, one needs a conditional lethal mutation, i.e. under one set of conditions, the ‘permissive’ one, the bacteria grow normally while under ‘nonpermissive’ conditions growth is inhibited. Mutation can be observed only under permissive conditions. Since the initial population already contains mutants and since more than a single reproduction round takes place, the accumulation of mutants with time has to be observed. When performing several independent experiments, the absolute number of resistant bacteria in the offspring was found to scatter widely, depending on whether the mutation was formed early or late during growth, showing the stochastic and undirected nature of the mutation. From the kinetic profiles of mutant accumulation after further growth, one can directly calculate Qrs values, where r is the viable, resistant and s the lethal, sensitive phenotype. The above equation can be simplified. Mutations are rare (Qss ≈ 1), wild type bacteria are in large excess (xs ≈ 1) and the growth of the total population ¯ ≈ As ), and we obtain depends only on the wild type (A dxr |dt

= {Ar − As }xr (t) + Qrs Ar .

(8)

When the mutation is entirely neutral, the first term cancels and the linear increase of the type frequency gives the mutation rate Wrs = Qrs Ar [93]. Otherwise one observes an exponential and a linear growth component and one can determine the mutation probability and, from the exponential part, the selection rate Ar − As . Generally, we can introduce classical fitness values Fi for a type i by dividing its growth rate by the rate of the predominant wild type w and divide the time by the growth rate to obtain the generation number N. The equation, which is valid only for a population dominated by the wild type where the contributions from other mutants to the mutational gain are negligible, then reads dxi |dN

= {Fi − 1}xi + Qiw .

(9)

Simplifications introduced by experimental conditions always limit the validity of the interpretation. When the time periods for observing the accumulation of phenotypic mutations was extended to several weeks by working in the turbidostat [31], non-reproducible, erratic patterns appeared, comprising periods where the relative mutant increased linearly as described above, but also periods where the relative mutant concentration decreased. How can

10

C. K. Biebricher · M. Eigen

the periods of decrease be understood? Obviously, the model chosen for the analysis was too simplistic: there were not only two types in the solution. Other mutations not changing the phenotype also occurred and, if one happened to be advantageous, it was enriched in the population and eventually replaced the old wild type together with its mutant spectrum. A new mutant spectrum was then built around the new wild type. Furthermore, the measured mutation probabilities were rather high since a degenerative mutation has been chosen. Resistance to phage T1 is simply achieved by abolishing the phage receptor, which is not needed by the host cell under these conditions. Hundreds of different mutational events may lead to destruction of a gene, causing a specific phenotypic change while only one (or a very few if pseudorevertants are possible) leads to the restoration of a lost gene. The found mutation probability was thus the sum of many specific mutation probabilities that led to the same mutant phenotype. Benzer [5, 6] succeeded in adapting this technique to map mutations within a gene of bacteriophage T4 by measuring the probabilities to restore wild type function by recombination after double infections. A surprisingly precise map of the gene was obtained. Two mutants not able to recombine to wild type were judged to be identical, and Benzer thus also obtained type frequencies. Benzer noted that type frequencies scattered for different loci, and called the loci with particularly high type frequencies ‘hot spots’. This name suggests that type frequencies are particularly enhanced at loci with high mutation rates, but this conclusion is not justified because Benzer’s measurements were taken from a snapshot of the mutant distribution at a certain time. Other factors also contribute to the type frequency, most notably error propagation by replication of the mutant genomes. As previously seen, only a kinetic study of the mutant spectrum can clarify the contributions of mutation and error propagation. Measurements of error rates of replicating enzymes by determining the increase of the number of revertants from a lethal phage mutant in vitro [42, 43, 45, 79, 78] restricted the growth to one replication round to avoid error propagation. Strictly seen, these error rates only apply to the specific mutation tested in the experiment, because error rates are not uniformly distributed within the sequence. Average enzyme fidelities q¯ , i.e. the probability of inserting the correct nucleotide in the incorporation of a single nucleotide, and average error rates 1 − q¯ , i.e. the probability of producing a mismatch in incorporating a nucleotide can be estimated only by statistical analysis of several such experiments. With an average fidelity of q¯ , the probability of obtaining a precise copy of a sequence with the chain length ν is calculated to be Qii = q¯ ν [37]. This suggests that at sufficiently large chain lengths, the probability of making a precise copy can become small. While the probabilities of

What Is a Quasispecies?

11

making specific mutations are always small in comparison to the probability of making correct offspring, the summation of the huge number of mutants may add up to a majority, causing Qii 1. Though Eq. (10) is an approximation, relation (11) generally holds surprisingly well, even for wide-spread quasispecies distributions. In all studied cases with single-stranded RNA viruses, the critical product νm (1 − qm ) turned out to be in the vicinity of 1. Therefore, σ¯ m must clearly exceed 1 to obtain a natural logarithm ln σ¯ m that is not too far from 1. In contrast, σ¯ m -values in Mendelian species are near unity, because large parts of the genome do not contribute to the immediate survival under the pertaining conditions. For these species, the average error rate per sequence i, εi , is very small and the term ln σ¯ m can be approximated by ε. For the cases studied, ε turned out to be between 10−2 and 10−3 .

What Is a Quasispecies?

21

The disintegration of information at the error threshold behaves like a firstorder phase transition [113, 114, 34]. In a numerical simulation carried out by Swetina and Schuster [113], where all mutants were assumed to have a uniform fitness of one tenth of the master, the relative population of the master x˜m dropped with increasing error rate to quite small values, until, at the error threshold, an instability occurred and all types in the population, including the master, became equally populated. While a defined wild type sequence could be determined below the error threshold, it suddenly disappeared at the error threshold. The chosen example of a single master sequence surrounded by uniformly less fit mutants is quite instructive, but unrealistic in nature. In reality, much of the sequence space cannot be accessed because nonviable mutants do not produce progeny. Hence, at the error threshold, the mutant distribution cannot evaporate into the whole sequence space. Instead, the cloud spreads out so much into the lethal area that fewer and fewer progeny are produced, resulting in the eventual annihilation of the population. This has been shown to happen with viral populations after the error rate was artificially raised by the addition of mutagenic drugs [80, 109, 23, 35, 50] (see E. Domingo et al, this volume, for discussion of error catastrophe as an antiviral strategy).

9 Evolutionary Biotechnology 9.1 Principles A new research field of increasing impact is evolutionary biotechnology. It is an application of the Darwinian principles of generation of diversity and selection with the aim of deriving novel molecules with desired properties [65]. The latter are improved by further cycles of amplification and selection (Fig. 3), whereby natural selection is replaced by artificial selection. Two general strategies can be chosen: rational and irrational design [15, 116, 92]. Rational design is a well-tested method that is suitable for well-understood systems. It is frequently used to modify protein properties [75, 44, 102] by site-directed, not random mutations. It is evolutionary in the sense that it involves evaluation and selection of successful mutants. Exploration of fitness is restricted to the mountain itself and its close surroundings. The irrational design is a gunshot method: we try to obtain success by playing a lottery game. The advantage of this method is that it does not depend on prior knowledge of how to achieve a desired function. The whole

22

C. K. Biebricher · M. Eigen

Fig. 3 General flow diagram for evolution experiments using artificial selection by a biassed process.

sequence space is used for exploration. This approach is very old: nearly all information on food and drugs accumulated in the history of mankind has been gained using this method, even though essentially by serendipity. A systematic exploration requires high-throughput methods and the invention of efficient screening methods. The extension to chemically synthesized polymers composed by suitable nonbiotic chemical residues (combinatorial chemistry [18, 77, 83]) is also new, as is the application of evolution strategies to technical processes [99]. Unfortunately, the expectation of success drops sharply with increasing complexity. Once a positive result has been obtained, the neighbourhood of its locus in sequence space is explored. For most applications, both strategies have severe limitations. A mixed approach is better: shotgunning is restricted to areas in sequence space with

What Is a Quasispecies?

23

better chances of achieving success, using whatever information is available. Furthermore, the lottery has fortunately not only one major win but many minor wins. In further evolutionary optimization, it is then possible to climb to the highest hill on the mountain. Suitable strategies can reduce the number of trials and reduce the risk of getting trapped in local optima [47], which is inevitable whenever the fitness landscape contains several, well-isolated mountains [20]. Instrumental for any shotgunning method to work are highly efficient screening methods able to spot successful types in a large number of blanks and a procedure to amplify them. Except for nucleic acids (see below), amplification is usually achieved by the synthesis of a new population with more sequential constraints. 9.2 Protein Design A typical mixed approach is the exploration of the fitness landscape around the mountainous part several steps deep into sequence space. These loci are already too far apart to be hit by a mutational event in a population of finite size. Mountainous parts of the fitness landscape are quite rugged. Many single mutations are lethal and lead to a drop into a precipice with no rescue. At other positions, the mountains can send ridges far into the sequence space. This part of the sequence is less stringent and the cloud becomes thin and large, while at the parts of the sequence critical for function the cloud becomes quite compact: lethal mutations interrupt transgression to regions that are farther apart. Exploration of the wider surroundings of the mountain in the fitness landscape is possible by artificially raising the error rate, e.g. by randomizing critical parts of the sequence [58, 60, 30]. Isolated peaks that are unlikely to be reached by natural mutation can thus be reached by jumping over broad canyons of lethal mutations. Such experiments [59, 89, 30] have revealed surprising extensions of the fitness landscape farther out in sequence space. Apparently, mountains in the fitness landscape are more extended than the corresponding mountains in the population landscape, in particular at limited population sizes. A fairly thorough exploration of the fitness landscape is only possible if oligomers of rather short chain lengths are used. Oligomers are poor in function, however. Binding of an antigen to an antibody is often determined by only a few amino acids of the antigen. Libraries of chemically synthesized oligopeptides [100] were thus screened for binding to certain monoclonal antibodies [90]. Often, several isolated mountains in the fitness landscape have been found: oligopeptides unrelated to the amino acid sequence of the

24

C. K. Biebricher · M. Eigen

antigenic part of the antigen were found to bind to monoclonal antibodies nearly as efficiently [68]. In some cases, it was possible to find ridges in the fitness landscape connecting one peak with another [57]; in other cases, the peaks appeared to be well isolated. An ingenious rational approach to create proteins with entirely new functions is the development of catalytic antibodies [73, 105, 74, 106]. The immune system of mammals provides an immunoglobulin repertoire as high as 107 specificities, which can be raised by another three orders of magnitude during the process of affinity maturation. The trick was to synthesize chemically stable analogues that structurally resemble a transition state at an intermediate distance from the chemical reaction that is desired to be catalyzed and use them as a hapten to provoke an in vivo immune response in an animal. Several monoclonal antibodies produced from hybridomas screened for binding the hapten were able to catalyze the desired reaction by several orders of magnitude. While the successful antibodies did not reach the efficiency nor the specificity of optimized enzymes, they were highly stereospecific. This method was limited to a few reactions. A broader application range has been the screening of large libraries of antibodies for a function, e.g. by replacing the hybridoma technique by a display approach. Screening is particularly efficient when the protein is displayed on the surface of a phage [97, 108, 110, 46] because the phage population not only provides clones of a certain protein phenotype, but also carries the corresponding genotypic information to reproduce and to amplify it. However, natural antibodies that contain four rather large polypeptide chains are nearly impossible to display. Single-chain antibody analogues were used [51, 98, 56] that have a conserved part responsible mainly for the structure and a variable, partly randomized part for isolating a binding pocket. The display techniques have been extended to molecular systems such as ribosome display [55, 1] and mRNA display [123, 76]. Since the bottleneck of transformation is avoided, much larger libraries are possible. Single-chain antibodies were also replaced by other protein alternatives that display a variable region on a rigid scaffold [7]. 9.3 Selection of nucleic acids with a function Display techniques limit the library to about 109 types. Much larger libraries of up to 1015 types can be created with nucleic acids. Furthermore, efficient in vitro amplification procedures for DNA [88, 104] and RNA [69, 52] are available. Selection from randomized or partially randomized RNA sequences, followed by iterative cycles of amplification and selection, has facilitated the design of nucleic acids with new properties, e.g., RNA variants that grow with

What Is a Quasispecies?

25

certain RNA polymerases [10, 121], ‘aptamers’ that bind ligands with high affinity and specificity [117, 41, 91, 72, 119, 71, 63] or ‘ribozymes’ [125, 53, 48] that catalyze certain chemical reactions, e.g. alkylation [122, 120], RNA ligation [40, 21], amino acid transfer [61, 81], and polynucleotide kinase activity [82]. Evolutionary optimization of natural ribozymes or extending their catalytic repertoire to similar reactions by coupling the wanted activity with a selective advantage has also been highly successful [3, 24]. Similar experiments with DNA single strands [16] have shown that even DNA strands can catalyze specific reactions. However, the catalytic efficiency of nucleic acids does not reach those of proteins due to the lack of side groups that allow efficient acid-base or redox catalysis. Ribozymic self-incorporation of a coenzyme [17] has opened the possibility of introducing residues capable of redox reactions. One of the current challenges of evolutionary biotechnology is designing procedures for coupling nucleic acid amplification with efficient in vitro expression into protein [112] followed by purification of the protein and evaluation of its function.

References 1. 2.

3. 4. 5. 6. 7.

8. 9.

Amstutz P, Forrer P, Zahnd C, Pluckthun A (2001) In vitro display techniques: novel developments and applications. Curr Op Biotechnol 12:400–405 Batschelet E, Domingo E, Weissmann C (1976) The proportion of revertant and mutant phage in a growing population, as a function of mutation and growth rate. Gene 1:27–32 Beaudry AA, Joyce GF (1992) Directed evolution of an RNA enzyme. Science 257:635–641 Beekwilder MJ (1996) Secondary structure of the RNA genome of bacteriophage Qβ. Dissertation, University of Leiden, the Netherlands Benzer S (1959) On the topology of the genetic fine structure. Proc Natl Acad Sci U S A 45:1607–1620 Benzer S (1961) On the topography of the genetic fine structure. Proc Natl Acad Sci U S A 47:403–415 Beste G, Schmidt FS, Stibora T, Skerra A (1999) Small antibody-like proteins with prescribed ligand specificities derived from the lipocalin fold. Proc Natl Acad Sci U S A 96:1898–1903 Biebricher CK (1983) Darwinian selection of RNA molecules in vitro. Evol Biol 16:1–52 Biebricher CK (1987) Replication and evolution of short-chained RNA species replicated with Qβ replicase. Cold Spring Harbor Symp Quant Biol 52:299–306

26

C. K. Biebricher · M. Eigen

10. Biebricher CK, Orgel LE (1973) An RNA that multiplies indefinitely with DNAdependent RNA polymerase: selection from a random copolymer. Proc Natl Acad Sci U S A 70:934–938 11. Biebricher CK, Eigen M (1988) Kinetics of RNA replication by Qβ replicase. In: Holland JJ, Ahlquist R, Domingo E (eds) RNA genetics. Vol I RNA-directed Virus Replication. CRC Press, Boca Raton FL, pp 1–21 12. Biebricher CK, Gardiner WC, Eigen M (1985) Kinetics of RNA replication: competition and selection among self-replicating species. Biochemistry 23:6550–6560 13. Biebricher CK, Gardiner WC, Eigen M (1991) Quantitative analysis of selection and mutation in self-replicating RNA. In: Peliti L (ed) Biologically inspired physics. Plenum Press, New York, pp 317–337 14. Billeter MA, Dahlberg JE, Goodman HM, Hindley J, Weissmann C (1969) Sequence of the first 175 nucleotides from the 5 terminus of Qβ RNA synthesized in vitro. Nature 224:1083–1086 15. Breaker RR, Joyce GF (1994) Inventing and improving ribozyme function: rational design versus iterative selection methods. Trends Biotechnol 12:268–275 16. Breaker RR, Joyce GF (1995) A DNA enzyme with Mg2+ -dependent RNA phosphoesterase activity. Chem Biol 2:655–660 17. Breaker RR, Joyce GF (1995) Self-incorporation of coenzymes by ribozymes. J Mol Evol 40:551–558 18. Brenner S, Lerner RA (1992) Encoded combinatorial chemistry. Proc Natl Acad Sci U S A 89:5381–5383 19. Brion P, Westhof E (1997) Hierarchy and dynamics of RNA folding. Annu Rev Biophys Biomol Struct 26:113–137 20. Carothers JM, Oestreich SC, Davis JH, Szostak JW (2004) Informational complexity and functional activity of RNA structures. J Am Chem Soc 126:5130–5137 21. Chapman KB, Szostak JW (1995) Isolation of a ribozyme with 5 → 5 ligase activity. Chem Biol 2:325–333 22. Crick FHC (1968) The origin of the genetic code. J Mol Biol 38:367–379 23. Crotty S, Cameron CE, Andino R (2001) RNA virus error catastrophe: direct molecular test by using ribavirin. Proc Natl Acad Sci U S A 98:6895–6900 24. Dai XC, Demesmaeker A, Joyce GF (1995) Cleavage of an amide bond by a ribozyme. Science 267:237–240 25. Domingo E, Sabo D, Taniguchi T, Weissmann C (1978) Nucleotide sequence heterogeneity of an RNA phage population. Cell 13:735–744 26. Domingo E, Biebricher CK, Eigen M, Holland JJ (2001) Quasispecies and RNA virus evolution: principles and consequences. Landes Bioscience, Georgetown TX 27. Drake JW (1993) Rates of spontaneous mutation among RNA viruses. Proc Natl Acad Sci U S A 90:4171–4175 28. Drake JW, Holland JJ (1999) Mutation rates among RNA viruses. Proc Natl Acad Sci U S A 96:13910–13913 29. Duarte E, Clarke D, Moya A, Domingo E, Holland J (1992) Rapid fitness losses in mammalian RNA virus clones due to Muller’s ratchet. Proc Natl Acad Sci U S A 89:6015–6019 30. Dube DK, Black ME, Munir KM, Loeb LA (1993) Selection of new biologically active molecules from random nucleotide sequences. Gene 137:41–47 31. Dykhuizen DE, Hartl DL (1983) Selection in chemostats. Microbiol Rev 47:150–168

What Is a Quasispecies?

27

32. Eigen M (1971) Self-organisation of matter and the evolution of biological macromolecules. Naturwissenschaften 58:465–523 33. Eigen M (1986) The physics of molecular evolution. Chemica scripta 26B:13–26 34. Eigen M (2000) Natural selection: a phase transition? Biophys Chem 85:101–123 35. Eigen M (2002) Error catastrophe and antiviral strategy. Proc Natl Acad Sci U S A 99:13374-13376 36. Eigen M, Biebricher CK (1988) Sequence space and quasispecies distribution. In: Domingo E, Holland JJ, Ahlquist R (eds) RNA Genetics. Vol III. Variability of RNA genomes. CRC Press, Boca Raton FL, pp 211–245 37. Eigen M, Schuster P (1979) The hypercycle – a principle of natural self-organization. Springer, Berlin New York Heidelberg 38. Eigen M, Winkler-Oswatitsch R (1981) Transfer-RNA: an early gene. Naturwissenschaften 68:282–292 39. Eigen M, McCaskill J, Schuster P (1989) The molecular quasispecies. Adv Chem Phys 75:149–263 40. Ekland EH, Szostak JW, Bartel DP (1995) Structurally complex and highly active RNA ligases derived from random RNA sequences. Science 269:364–370 41. Ellington AD, Szostak JW (1990) In vitro selection of RNA molecules that bind to specific ligands. Nature 346:818–822 42. Fersht AR (1976) Fidelity of replication of phage ΦX174 by DNA polymerase III holoenzyme: spontaneous mutation by misincorporation. Proc Natl Acad Sci U S A 76:4946–4950 43. Fersht AR (1979) DNA polymerase accuracy and spontaneous mutation rates frequency of purine-purine purine-pyrimidine and pyrimidine-purine mismatches during replication. Proc Natl Acad Sci U S A 76:4946–4950 44. Fersht AR (1987) Dissection of the structure and activity of the tyrosyl-tRNAsynthetase by site-directed mutagenesis. Biochemistry 26:8031–8037 45. Fersht AR, Knill-Jones JW (1981) DNA polymerase accuracy and spontaneous mutation rates: frequencies of purine:purine purine:pyrimidine and pyrimidine:pyrimidine mismatches during DNA replication. Proc Natl Acad Sci U S A 78:4251–4255 46. Forrer P, Jung S, Pluckthun A (1999) Beyond binding: using phage display to select for structure, folding and enzymatic activity in proteins. Curr Op Struct Biol 9:514–520 47. Forrest S (1993) Genetic algorithms: principles of natural selection applied to computation. Science 261:872–878 48. Forster AC, Symons RH (1987) Self-cleavage of plus and minus RNAs of a virusoid and a structural model for the active sites. Cell 49:211–220 49. Gould SJ, Eldredge N (1977) Punctuated equilibria: the tempo and mode of evolution revisited. Palaeobiol 3:115–151 50. Grande-Pérez A, Sierra S, Castro MG, Domingo E, Lowenstein PR (2002) Molecular indetermination in the transition to error catastrophe: systematic elimination of lymphocytic choriomeningitis virus through mutagenesis does not correlate linearly with large increases in mutant spectrum complexity. Proc Natl Acad Sci U S A 99:12938–12943 51. Griffiths AD (1993) Building an in vitro immune system – human antibodies without immunization from phage display libraries. Ann Biol Clin 51:554–554

28

C. K. Biebricher · M. Eigen

52. Guatelli JC, Whitfield KM, Kwoh DY, Barringer KE, Richman DD, Gingeras TR (1990) Isothermal, in vitro amplification of nucleic acids by a multienzyme reaction modeled after retroviral replication. Proc Natl Acad Sci U S A 87:1874–1878 53. Guerrier-Takada C, Gardiner K, Marsh T, Pace N, Altman S (1983) The RNA moiety of ribonuclease P is the catalytic subunit of the enzyme. Cell 35:849–857 54. Hamming RW (1980) Coding and information theory. Prentice Hall Inc. Englewood Cliffs NJ 55. Hanes J, Plückthun A (1997) In vitro selection and evolution of functional proteins by using ribosome display. Proc Natl Acad Sci U S A 94:4937–4942 56. Hanes J, Jermutus L, Weber-Bornhauser S, Bosshard HR, Plückthun A (1998) Ribosome display efficiently selects and evolves high-affinity antibodies in vitro from immune libraries. Proc Natl Acad Sci U S A 95:14130–14135 57. Hoffmüller U, Knaute T, Hahn M, Höhne W, Schneider-Mergener J, Kramer A (2000) Evolutionary transition pathways for changing peptide ligand specificity and structure. EMBO J 19:4866–4874 58. Horwitz MS, Loeb LA (1986) Promoters selected from random DNA sequences. Proc Natl Acad Sci U S A 83:7405-7409 59. Horwitz MS, Loeb LA (1988) DNA sequences of random origin as probes of Escherichia coli promoter architecture. J Biol Chem 263:14724–14731 60. Horwitz MS, Dube DK, Loeb LA (1989) Selection of new biological activities from random nucleotide sequences: evolutionary and practical considerations. Genome 31:112–117 61. Illanga-Sekare M, Sanchez G, Nickles T, Yarus M (1995) Aminoacyl RNA synthesis catalyzed by an RNA. Science 267:643–647 62. Jenkins GM, Worobey M, Woelk CH, Holmes EC (2001) Evidence for the nonquasispecies evolution of RNA viruses. Mol Biol Evol 18:987–994 63. Jiang L, Suri AK, Fiala R, Patel DJ (1996) Saccharide-RNA recognition in an aminoglycoside antibiotic-RNA aptamer complex. Chem Biol 4:35–50 64. Jones BL, Enns RH, Rangnekar SS (1976) On the theory of selection of coupled macromolecular systems. Bull Math Biol 38:15–28 65. Kauffman SA (1992) Applied molecular evolution. J Theor Biol 157:1–7 66. Kimura M (1983) The neutral theory of molecular evolution. Cambridge University Press, Cambridge 67. Klovins J, Tsareva NA, de Smit MH, Berzins V, van Duin J (1997) Rapid evolution of translational control mechanisms in RNA genomes. J Mol Biol 265:372–384 68. Kramer A, Keitel T, Winkler K, Stöcklein W, Höhne W, Schneider-Mergener J (1999) Molecular basis for the binding promiscuity of an anti-p24 (HIV) monoclonal antibody. Cell 91:799–809 69. Kwoh DY, Davis GR, Whitfield KM, Chapelle H, DiMichele L, Gingeras TR (1989) Transcription-based amplification system and detection of amplified immunodeficiency virus type 1 with a bead-based sandwich hybridization format. Proc Natl Acad Sci U S A 86:1173–1177 70. Lake JA (1987) Prokaryotes and archaebacteria are not monophyletic: Rateinvariant analysis of rRNA genes indicate that eukaryotes and eocytes form a monophyletic taxon. Cold Spring Harbor Symp Quant Biol 52:839–846 71. Lato SM, Boles AR, Ellington AD (1995) In vitro selection of RNA lectins – using combinatorial chemistry to interpret ribozyme evolution. Chem Biol 2:291–303

What Is a Quasispecies?

29

72. Lauhon C, Szostak J (1993) In vitro selection of RNA molecules that specifically bind riboflavin. FASEB J 7:A1087 73. Lerner RA, Tramontaur A (1988) Catalytic Antibodies. Sci Am 258:42–50 74. Lerner RA, Benkovic SJ, Schultz PG (1991) At the crossroads of chemistry and immunology: catalytic antibodies. Science 252:659–667 75. Leatherbarrow RJ, Fersht AR (1986) Protein engineering. Protein Eng 1:7–16 76. Lipovsek D, Plückthun A (2004) In-vitro protein evolution by ribosome display and mRNA display. J Immunol Meth 290:51–67 77. Liscamp RMJ (1994) Opportunities of new chemical libraries. Angew Chem Int Ed Engl 33:633–636 78. Loeb LA, Kunkel TA (1982) Fidelity of DNA synthesis. Annu Rev Biochem 51:428-497 79. Loeb LA, Weymouth LA, Kunkel TA, Gopinathan KP, Beckman RA, Dube DK (1978) On the fidelity of DNA replication. Cold Spring Harbor Symp Quant Biol 43:921–927 80. Loeb LA, Essigman JM, Kazazi F, Zhang F, Rose KD, Mullins JI (1999) Lethal mutagenesis of HIV with mutagenic nucleotide analogs. Proc Natl Acad Sci U S A 96:1492–1497 81. Lohse PA, Szostak JW (1996) Ribozyme-catalyzed amino acid transfer reactions. Nature 381:442–444 82. Lorsch JR, Szostak JW (1995) In vitro evolution of polynucleotide kinase ribozymes. FASEB J 9:A1422 83. Lowe G (1995) Combinatorial chemistry. Chem Soc Rev 24:309–317 84. Luria SE, Delbrueck M (1943) Mutations of bacteria from virus sensitivity to virus resistance. Genetics 28:486–491 85. Maynard Smith J (1989) Evolutionary genetics. Oxford University Press, Oxford 86. Mills DR, Peterson RL, Spiegelman S (1967) An extracellular Darwinian experiment with a self-duplicating nucleic acid molecule. Proc Natl Acad Sci U S A 58:217–224 87. Muller HJ (1964) The relation or recombination to mutational advance. Mutat Res 1:2–9 88. Mullis K, Faloona F, Scharf S, Saiki R, Horn G, Erlich H (1986) Specific enzymatic amplification of DNA in vitro: the polymerase chain reaction. Cold Spring Harbor Symp Quant Biol 51:263–273 89. Munir KM, French DC, Loeb LA (1993) Thymidine kinase mutants obtained by random sequence selection. Proc Natl Acad Sci U S A 90:4012–4016 90. Nakamura GR, Reynolds ME, Chen YM, Starovasnik MA, Lowman HB (2002) Stable “zeta” peptides that act as potent antagonists of the high affinity IgE receptor. Proc Natl Acad Sci U S A 99:1303–1308 91. Niewlandt P, Decker D , Gold L (1995) In vitro selection of RNA ligands to substances. Biochemistry 34:5651–5659 92. Nixon AE, Firestine SM (2000) Rational and “irrational” design of proteins and their use in biotechnology. IUBMB Life 49:181–187 93. Novick A, Szilard L (1950) Experiments with the chemostat on spontaneous mutations of bacteria. Proc Natl Acad Sci U SA 36:708–719 94. Olsthoorn RCL, van Duin J (1996) Evolutionary reconstruction of a hairpin deleted from the genome of an RNA virus. Proc Natl Acad Sci U S A 93:12256–12261

30

C. K. Biebricher · M. Eigen

95. Olsthoorn RCL, Licis N, van Duin J (1994) Leeway and constraints in the forced evolution of a regulatory RNA helix. EMBO J 13:2660-2668 96. Ort´ın J, Nájera R, Lopez C, Davila M, Domingo E (1980) Genetic variability of Hong Kong (H3N2) influenza viruses: spontaneous mutations and their location in the viral genome. Gene 11:319–331 97. Parmley SF, Smith GP (1988) Antibody-selectable filamentous fd phage vectors – affinity purification of target genes. Gene 73:305–318 98. Posner B, Smiley J, Lee I, Benkovic S (1994) Catalytic antibodies: Perusing combinatorial libraries. Trends Biochem Sci 19:145–150 99. Rechenberg I (1973) Evolutionsstrategie. Problemata Formann-Holzboog, Stuttgart-Bad Cannstatt 100. Reineke U, Kramer A, Schneider-Morgener J (1999) Antigen sequence- and library-based mapping of linear and discontinuous protein-protein interaction sites by spot synthesis. Curr Top Microbiol Immunol 243:23–36 101. Rohde N, Daum H, Biebricher CK (1995) The mutant distribution of an RNA species replicated by Qβ replicase. J Mol Biol 249:754–762 102. Russell AJ, Fersht AR (1987) Rational modification of enzyme catalysis by engineering surface charge. Nature 328:496–500 103. Sanger R, Brownlee GG, Barrell BG (1965) A two-dimensional fractionation procedure for radioactive nucleotides. J Mol Biol 13:378–398 104. Saiki RK, Gelfand DH, Stoffel S, Scharf SJ, Higuchi R, Horn GT, Mullis KB, Erlich HA (1988) Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239:487–491 105. Schultz PG (1989) Catalytic antibodies. Angew Chem Int Ed 28:1283–1295 106. Schultz PG, Lerner RA (1993) Antibody catalysis of difficult chemical transformations. Acc Chem Res 26:391–395 107. Schuster P (1986) The physical basis of molecular evolution. Chemica scripta 26B:27–41 108. Scott JK, Smith GP (1990) Searching for peptide ligands with an epitopic library. Science 249:386–390 109. Sierra S, Davila M, Lowenstein PR, Domingo E (2000) Response of foot-andmouth-disease virus to increased mutagenesis: influence of viral load and fitness in loss of infectivity. J Virol 74:8316–8323 110. Smith GP, Scott JK (1993) Libraries of peptides and proteins displayed on filamentous phage. Meth Enzymol 217:228–257 111. Spiegelman S, Haruna, Holland IB, Beaudreau G, Mills DR (1965) The synthesis of a self-propagating and infectious nucleic acid with a purified enzyme. Proc Natl Acad Sci U S A 54:919–927 112. Stiege W, Erdmann VA (1995) The potentials of the in vitro biosynthesis system. J Biotechnol 41:81–90 113. Swetina J, Schuster P (1982) Self-replication with errors. A model for polynucleotide replication. Biophys Chem 16:329–345 114. Tarazona P (1992) Error thresholds for molecular quasispecies as phase transitions: from single landscapes to spin-glass models. Phys Rev E 45:6038–6050 115. Thompson CL, McBride JL (1974) On Eigen’s theory of the self-organization of matter and the evolution of biological macromolecules. Math Biosci 21:127–142

What Is a Quasispecies?

31

116. Tobin MB, Gustafsson C, Huisman GW (2000) Directed evolution: the ‘rational’ basis for ‘irrational’ design. Curr Op Struct Biol 10:421–427 117. Tuerk C, Gold L (1990) Systematic evolution of ligands by exponential enrichment: RNA ligands to bacteriophage T4 DNA polymerase. Science 249:505–510 118. Van Meerten D, Girard G, van Duin J (2001) Translational control by delayed RNA folding: identification of the kinetic trap. RNA 7:483–494 119. Wallis MG, von Ahsen U, Schroeder R, Famulok M (1995) A novel RNA structure for neomycin recognition. Chem Biol 2:543–552 120. Wecker M, Smith D, Gold L (1996) In vitro selection of a novel catalytic RNA – characterization of a sulfur alkylation reaction and interaction with a small peptide. RNA 2:982–994 121. Wettich A, Biebricher CK (2001) RNA species that replicate with DNA-dependent RNA polymerase from Escherichia coli. Biochemistry 40:3308–3315 122. Wilson C, Szostak JW (1995) In vitro evolution of a self-alkylating ribozyme. Nature 374:777–782 123. Wilson DS, Keefe AD, Szostak JW (2001) The use of mRNA display to select high-affinity protein-binding peptides. Proc Natl Acad Sci U S A 98:3750–3755 124. Woese CR, Fox GE (1977) The concept of cellular evolution. J Mol Evol 10:1–6 125. Zaug AJ, Cech TR (1986) The intervening sequence RNA of Tetrahymena is an enzyme. Science 231:470–475

CTMI (2006) 299:33–50 c Springer-Verlag Berlin Heidelberg 2006


Quasispecies in Time-Dependent Environments C. O. Wilke1 (u) · R. Forster2 · I. S. Novella3 1 Section of Integrative

Biology and Center for Computational Biology and Bioinformatics, University of Texas at Austin, Austin, TX 78712, USA [email protected] 2 Digital Life Laboratory, California Institute of Technology, MC 136–93, Pasadena, CA 91125, USA 3 Department of Medical Microbiology and Immunology, Medical University Ohio, 3055 Arlington Ave., Toledo, OH 43616, USA

1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

2

Virus Evolution in Time-Dependent Fitness Landscapes . . . . . . . . . . . . 35

3

Adaptation to Two Alternating Hosts . . . . . . . . . . . . . . . . . . . . . . . . . . 37

4

Adaptation of Mutation Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

5

Co-evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

6

Discussion and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Abstract In recent years, quasispecies theory in time-dependent (that is, dynamically changing) environments has made dramatic progress. Several groups have addressed questions such as how the time scale of the changes affect viral adaptation and quasispecies formation, how environmental changes affect the optimal mutation rate, or how virus and host co-evolve. Here, we review these recent developments, and give a nonmathematical introduction to the most important concepts and results of quasispecies theory in time-dependent environments. We also compare the theoretical results with results from evolution experiments that expose viruses to successive regimes of replication in two or more different hosts.

1 Introduction From the point of view of a virus, the world is in a state of constant change. For sustained existence, the virus has to move from host organism to host organism, and each new host differs somewhat from the previous one. Some viruses

34

C. O. Wilke et al.

even change their host species on a regular basis. For example, arboviruses alternate between infections in arthropods and infections in vertebrates. Within a host, the world is not necessarily constant either. Viruses infecting higher organisms find themselves under constant attack from an adaptive immune system, whose ability to fight any particular virus mutant grows dramatically after exposure to the mutant. Fortunately—for the virus, that is—the large class of RNA viruses is more than prepared for this world of change. Their high replication rates, combined with exceptionally high mutation rates, allow them to explore vast numbers of mutations, which leads to rapid adaptation in the face of environmental variation. For example, it has been estimated that HIV-1 produces all possible single- and double-point mutants and a good fraction of the triple mutants in every single patient every day (Perelson et al. 1997). The theoretical framework in which RNA viruses are typically modeled is quasispecies theory, which was put forward by Eigen and co-workers in the 1970s (Eigen and Schuster 1979; Eigen et al. 1988, 1989) (see also chapter by Biebricher and Eigen, this volume). Quasispecies theory describes how a virus population reacts to the two forces of selection and mutation, and how these two forces counterbalance each other. Mathematically, quasispecies theory is equivalent to the theory of mutation–selection balance, which has been developed parallel to quasispecies theory in the population-genetics literature (Bürger 2000). Eigen and co-workers (Eigen and Schuster 1979; Eigen et al. 1988, 1989) defined the quasispecies as the stable mutant configuration that arises when mutation and selection are in perfect equilibrium. With this definition, the theory seemed to apply only to static environments and an infinite population size. Indeed, it has been argued that since viral populations are finite, quasispecies theory may not realistically reflect their behavior, because perfect equilibrium between mutation and selection can never be achieved in the real world (Jenkins et al. 2001; Holmes and Moya 2002). However, a number of theoretical studies have shown that understanding the infinite-population dynamics is crucial for developing a more detailed finite-population model (van Nimwegen et al. 1999a; Wilke 2001a) and that all the effects predicted by quasispecies theory can in principle also be observed in finite populations (Nowak and Schuster 1989; van Nimwegen et al. 1999b; Wilke et al. 2001b, Wilke 2001b). In recent years, substantial progress has been made in extending quasispecies theory to dynamic environments. This work can be subdivided into two groups, studies on virus evolution in time-dependent fitness landscapes (Nilsson and Snoad 2000, 2002a, 2002b, Wilke et al. 2001a; Wilke and Ronnewinkel 2001) and studies on the interaction between virus and the host’s

Quasispecies in Time-Dependent Environments

35

immune system, or virus–host co-evolution in general (Kamp and Bornholdt 2002; Kamp et al. 2002; Brumer and Shakhnovich 2004). In this chapter, we put most emphasis on the former type of studies, but will also give a brief overview on the latter. We discuss the relationship of theory to experimental results in the individual sections where appropriate.

2 Virus Evolution in Time-Dependent Fitness Landscapes In this section, we consider situations in which the virus’s environment changes due to external factors but is not in return influenced by any of the virus’s adaptive moves. Throughout this chapter, we use for this case the terminology of “adaptation to a time-dependent fitness landscape.” As an example, consider a virus that is grown alternatingly on different hosts (see also the next section). From a theoretical point of view, it is particularly interesting to study changes in the fitness landscape that are periodic (that is, situations in which the environment changes over and over again in exactly the same pattern), because the theory that has been developed for static fitness landscapes can— after a suitable mathematical transformation—to a large extent be directly applied to periodic fitness landscapes (Wilke et al. 2001a). A number of exact results and simple principles have been derived for periodic fitness landscapes. We will review these results in the following paragraphs, and will consider nonperiodic changes briefly at the end of this section. Before we can talk in depth about periodically changing (or oscillating) fitness landscapes, we have to introduce some terminology. We refer to the time it takes for the fitness landscape to go through one cycle of changes as the oscillation period. For example, if the fitness landscape alternates between environment A for 100 generations and environment B for 100 generations, then the oscillation period is 200 generations. We denote the oscillation period by T. The inverse of the oscillation period, 1/T, is called the oscillation frequency. The shorter the oscillation period, the higher the oscillation frequency. We refer to a special point in time within one oscillation period as the phase. We typically express the phase in fractions of one complete oscillation cycle. For example, in the above example, the switch from A to B occurs at phase 1/2, and the switch from B to A occurs at phase 1, which by definition is equivalent to phase 0. One important result for oscillating fitness landscapes is that as long as we look at the system under study only at a single phase (say, for example, at the beginning of the oscillation, phase = 0), the virus population behaves as

36

C. O. Wilke et al.

if the fitness landscape were static (Wilke et al. 2001a). After an initial transient period, the virus population develops a stable distribution of mutants, a quasispecies. The only difference from a static fitness landscape is that this equilibrium depends on the phase. At phase 1/2, the equilibrium distribution of mutants may be totally different from the distribution at phase 0, but at any given phase, the equilibrium distribution will always be the same. In theory, the population reaches this equilibrium only after infinitely many oscillation cycles have passed, but in practice, more than a couple of hundred cycles are rarely necessary to obtain equilibration, and frequently equilibration occurs after only a handful of cycles. We can also calculate accurately what happens when the oscillations are very fast or very slow. The general result is the following: under very fast changes, the population behaves as if the fitness landscape did not change at all, but instead was the average of the changing landscape (Wilke et al. 2001a; Wilke and Ronnewinkel 2001). That is, if for example two traits are alternatingly selected for, then above a certain oscillation frequency, the population will simply acquire both traits at the same time. For very slowly changing landscapes, the population essentially does not realize that the environment is changing, and climbs the local peak that best represents the current environment. If the environment changes slowly but continuously, then the population will track this local optimum, and always stay optimally adapted (Nilsson and Snoad 2002b). If, however, the environment suddenly requires a completely different adaptation, then the population may be thrown off track and suddenly find itself hopelessly ill-adapted and unable to cope with the new environment (Ronnewinkel et al. 2001). The behavior at intermediate oscillation frequencies interpolates between the two extremes. As the oscillation frequency goes up, the population ceases to track the changes accurately, and behaves more as if it were only experiencing the average fitness landscape. An interesting phenomenon that occurs at intermediate oscillation frequencies is the phase shift. The term “phase shift” means that there is a substantial lag between the point in time at which a change in the fitness landscape takes place and the point in time at which the population shows response to this change. If we measure this lag in units of the oscillation period, then the lag goes to zero as the oscillation period increases. However, when the oscillation period decreases, then the phase shift can assume substantial values, as the population is constantly trying to keep up with the changes in the landscape (Nilsson and Snoad 2002b). To summarize, we can compare the evolutionary dynamics in a periodic fitness landscape with the effect of a low-pass filter on an audio signal (Wilke et al. 2001a; Nilsson and Snoad 2002b): A low-pass filter replaces high-frequency signals with their average intensity, and lets low-frequency signals pass unal-

Quasispecies in Time-Dependent Environments

37

tered. Moreover, for intermediate- to high-frequency signals, a low-pass filter introduces a time lag (phase shift) that depends on the frequency of the signal. The general predictions for a fitness landscape whose changes are aperiodic or irregular are very similar to those for a periodically changing fitness landscape. Again, evolution essentially acts as a low-pass filter on the changing fitness landscape. The population will not be able to follow changes in the fitness landscape that happen on a short time scale, and will see only the average effect of these changes. However, the population will typically be able to track a slowly changing landscape accurately and stay optimally adapted. One significant difference from periodic fitness landscapes is the case of very fast, irregular changes. In this case, it can happen that the population loses track of the advantageous regions in the fitness landscape altogether and becomes completely ill adapted. This loss of adaptation can be considered as the dynamic equivalent of the classic (static) error threshold. The static error threshold sets an upper bound to the maximum mutation rate that a population can sustain without losing all genetic information (Eigen and Schuster 1979). The dynamic error threshold differs from the classic error threshold in that it sets a lower bound on the minimum mutation rate with which a population can still follow the changes in the fitness landscape (Nilsson and Snoad 2000).

3 Adaptation to Two Alternating Hosts As an example for a periodic fitness landscape, we consider here the situation where a virus population has to adapt to two alternating hosts. Such a situation is common, for example in arboviruses, which alternate between rounds of replication in arthropod vectors and rounds of replication in vertebrate hosts. One question that is frequently asked in this context is whether the virus can adapt to both hosts at the same time, or whether adaptation to one host is typically accompanied by loss of adaptation to the other host. We can study this question in the framework of a very simple model. Assume that a virus strain has one gene that is advantageous in host one, and neutral in host two, and a second gene that has the reversed properties. We refer to this strain as the divided strain. Further, assume that there is also a second virus strain which has a single gene that conveys the advantageous functions of each of the two genes in the corresponding hosts. We refer to this strain as the fused strain. In addition, both the divided and the fused strain can suffer from point mutations, and do so at the same rate per site. In the case of the divided strain, a single point mutation will affect either of the two genes (and thus the

38

C. O. Wilke et al.

two functions) separately, whereas in the fused strain, point mutations affect both functions simultaneously. Note that while we talk about separate genes throughout this chapter, this model can also apply to separate functions carried out by a single gene. In this case, the divided strain corresponds to a gene that can adapt to either function, but not to both at the same time, whereas the fused strain corresponds to a gene that can adapt to both functions at the same time. These differences in the strains can be caused by pleiotropic effects. Throughout this section, we consider populations of finite size N, whereas the general results presented in the previous section are derived from infinitepopulation theory. By and large, these general results also apply to finite populations. The main difference is that (a) in finite populations, genotypes can truly go extinct, whereas all genotypes are always present in infinite populations, and (b) that the quasispecies dynamic becomes a stochastic process, subject to fluctuations and chance events. Figure 1 demonstrates how the divided strain will evolve if it is grown alternatingly for 30 generations, each on one of the two hosts. When the divided strain grows on host one, the second gene is not under selection, and therefore slowly deteriorates due to mutation accumulation. When the virus is transferred to the second host, then gene two comes under selection and rebounds quickly, while gene one starts to deteriorate. Figure 2 illustrates this dynamic by visualizing the mutant cloud that forms at different points in time. When gene one is under selection, then only a small number of sequences

Fig. 1 Frequencies of error-free genes as a function of time. Parameters are oscillation period T = 60, per-site mutation rate u = 0.02, length of a single gene of the divided strain ldiv = 5, selective advantage of functional gene s = 1, population size N = 1,000

Quasispecies in Time-Dependent Environments

39

Fig. 2 Population structure of either the divided or fused strain at various time points. Gray levels indicate the fraction of sequences at the given mutational distance from the respective error-free sequence. Parameters are oscillation period T = 60, per-site mutation rate u = 0.02, length of a single gene of the divided strain ldiv = 5, length of the fused gene lfused = 8, selective advantage of functional gene s = 1, infinite population size

acquire mutations in this gene, while at the same time the majority of the population acquires mutations in gene two. The image reverses when gene two starts to come under selection. Then, only a small number of sequences acquire mutations in gene two, but almost all sequences acquire mutations in gene one. Note that the fused strain, grown under the same conditions, fully retains both functions at all times, and accumulates only a small number of mutations in any case (Fig. 2). If we compete the divided with the fused strain in a changing environment, which one will prevail? There is no simple answer to this question. The answer depends on a number of factors, the two most important of which are the time scale on which the changes in the environment occur and whether mutation pressure creates any additional trade-offs between the divided strain and the fused type. For example, if the environment changes very quickly (such as when the host changes every generation), then the divided strain will maintain both functions at all times, just as the fused type does. Therefore, in this case the only difference between the two strains is the extent to which they suffer from mutation pressure. If we assume that their per-site mutation rate u is the same, then the mutation pressure translates into the number of sites in the genome that are under selection. The strain with the larger number of sites under selection will suffer from a higher mutational pressure and lose against the other strain (Wilke and Adami 2003). Thus, if the fused gene can perform both functions with fewer sites under selection than the two genes of the divided strain, then the fused strain has an advantage. A similar argument applies when the environment changes very slowly. However, in this case the

40

C. O. Wilke et al.

selection pressure is different, because the virus has sufficient time to adapt fully to one host before it is exposed again to the other host. During this long period of exposure to a single host, the decisive factor is now whether the single gene of the divided strain that is needed for this host suffers from a higher mutational pressure than the fused gene. It is probably reasonable to assume that the two genes of the divided strain each have fewer selective sites than the single fused gene, but that the fused gene has fewer selective sites than the two former genes together. Under this

Fig. 3 Fused strain invades a population of the divided strain. A single fused genotype is introduced into the population at t = 0 and reintroduced at later times whenever all fused genotypes have been lost from the population. The 25th introduced genotype goes to fixation. Parameters are oscillation period T = 60, per-site mutation rate u = 0.006, length of a single gene of the divided strain ldiv = 5, length of the fused gene lfused = 8, selective advantage of functional gene s = 1, population size N = 1,000

Quasispecies in Time-Dependent Environments

41

assumption, the fused strain will be advantageous when the environment changes quickly, while the divided strain will be advantageous when the environment changes slowly. Figure 3 shows how the fused strain can invade a population of the divided strain if the environmental changes occur on a sufficiently fast time scale, while Fig. 4 shows how the divided strain can invade a population of the fused strain if the environment changes slowly. In experiments with digital organisms (self-replicating computer programs that mutate and evolve; Wilke and Adami 2002) that were adapting

Fig. 4 Divided strain invades a population of the fused strain. A single divided genotype is introduced into the population at t = 0, and reintroduced at later times whenever all divided genotypes have been lost from the population. The 31st introduced genotype goes to fixation. Parameters are oscillation period T = 600, per-site mutation rate u = 0.006, length of a single gene of the divided strain ldiv = 5, length of the fused gene lfused = 8, selective advantage of functional gene s = 1, population size N = 1,000

42

C. O. Wilke et al.

to a periodically changing fitness landscape, trajectories very similar to the one shown in Fig. 3 were observed for moderately to rapidly changing fitness landscapes, but not for very slowly changing fitness landscapes (Li and Wilke 2004). These observations imply that genes or part of genes can indeed evolve to be fused or divided, even if the system under study does not a priori provide fused or divided genes. In the digital organisms, separation or fusion of genes was not imposed from the outside, unlike in the case of the simple toy model discussed in this section, and evolved presumably through changes in the amount of epistatic interactions among the different parts of the digital organisms’ genomes. Several groups have experimentally tested the effect of replication in two different hosts on the evolution of viral quasispecies. Alternating acute replication of vesicular stomatitis virus (VSV), eastern equine encephalitis virus (EEEV), and Dengue virus in cells of insect and mammalian origin led to fitness increase in both cell types (Novella 1999; Weaver et al. 1999; Chen et al. 2003). The difference in behavior between VSV and EEEV is that for VSV, insect and mammalian cells represent similar fitness landscapes for replication, whereas for EEEV a trade off can be observed: adaptation to one cell type results in loss of viral fitness in the other. Similar results are observed for adaptation to cells of insect and avian origin. Viruses that grow alternatingly on these two cell types produce adaptations to both environments in spite of trade-offs (Cooper and Scott 2001). However, while acute insect replication does not involve trade-offs for acute mammalian replication, persistent replication of VSV in insect cells does result in trade-offs for replication in mammalian cells (Novella et al. 1995), and evolution during alternation between acute mammalian replication and persistent insect replication is dominated by the persistent phase (Zárate and Novella 2004). Additional work has been done using different types of mammalian cells (BHK-21, MDCK, and HeLa) as hosts for VSV replication. In these experiments, VSV consistently adapted to both of the alternating cell types independently of the existence of tradeoffs and independently of whether the host switch occurred periodically or randomly (Turner and Elena 2000). All these experimental results are in qualitative agreement with the theory, in that they show that there is no simple rule of how viruses behave that are grown alternatingly in two different hosts. They may either adapt to both hosts at the same time or adapt to one host at the expense of adaptation to the other. Which of the two cases occurs depends on the time spent in each of the two hosts as well as on the details of the fitness landscape in the two hosts.

Quasispecies in Time-Dependent Environments

43

4 Adaptation of Mutation Rate In the context of time-dependent environments, it is natural to ask what the optimal mutation rate is for the viral population. In a static environment, once a virus has reached the global optimum in the fitness landscape, all mutations are deleterious. In this case, it pays for the virus to reduce the mutation rate as much as possible, as a lower mutation rate translated into a lower mutational load—and thus higher average fitness—for the offspring virions. In a timedependent environment, on the other hand, a zero or near-zero mutation rate is detrimental. A virus needs a sufficiently high mutation rate in order to be able to follow changes in the fitness landscape. It is clear that the mutation rate at which the virus is optimally capable of tracking the fitness landscape will depend on the speed and nature of the changes in the fitness landscape. Nilsson and Snoad (2002a) studied a simple model of a viral quasispecies that has to track a moving fitness peak, and calculated the optimal mutation rate in this model. The model assumptions are the following: the fitness peak remains static for a time interval of length τ1 . Then it moves to a position nearby in the sequence space. The distance from the old position to the new position is k1 point mutations. The peak remains static at the new position for a time interval of length τ2 , until it moves to another nearby position, this time with distance k2 , and so on. The lengths of the intervals τi and the distances ki need not be identical from shift to shift. Nilsson and Snoad found that the optimal mutation rate could be expressed simply as µopt = k/(στ), where k is the average mutational distance that the peak moves, σ is the mean replication rate of the virus on the peak, and τ is the average time between peak shifts. We can rewrite Nilsson and Snoad’s result such that its interpretation becomes more intuitive. Note that 1/σ is the average time the virus needs to replicate, and that therefore στ is the average time between peak shifts measured in virus generations. Thus Nilsson and Snoad’s result states that the virus can optimally adapt to the changing environment if its genomic mutation rate per generation is the mean distance between peaks divided by the number of generations between successive peak shifts. This result makes intuitive sense: with this mutation rate, the majority of mutants in the virus population will, by the next peak shift, have accumulated exactly the right number of mutations to be located at the position of the new peak. It is interesting to note that for k = 1, we recover a result that Kimura had already derived in a much simpler model in 1967 (Kimura 1967). Note that Nilsson and Snoad’s result is very different from the optimal mutation rate of an ill-adapted virus in a static environment, that is, of a virus that

44

C. O. Wilke et al.

can still discover many advantageous mutations even when the environment does not change. In this case, the optimal mutation rate is given by the mean effect of deleterious mutations (Orr 2000). A mutation rate of this magnitude optimally balances the beneficial effect of new advantageous mutations with the deleterious effect of too high a mutational load. It is not clear whether Orr’s theory or Nilsson and Snoad’s theory is more relevant to RNA viruses. The answer depends on the time scale on which changes happen, and on how easy it is for a virus to reach optimal adaptation after a change in the fitness landscape has taken place. If changes happen only occasionally, and a number of beneficial mutations need to be discovered to reach optimal adaptation to the new environment, then Orr’s theory is probably more relevant. If on the other hand, changes happen regularly, and only one or two mutations are needed to reach optimal adaptation to the new environment, then Nilsson and Snoad’s theory should apply.

5 Co-evolution In the preceding sections, we have discussed virus evolution in fitness landscapes that change because of external forces. By “external forces” we mean that the virus has to adapt to environmental changes, but the environment is not influenced by the adaptive moves of the virus. In many cases, however, the virus imposes a time-dependent fitness landscape on the environment just as much as the environment imposes a time-dependent fitness landscape on the virus. Typical examples are the interaction between virus and immune system, and co-evolution of virus and host in general. This section is not meant as an in-depth review of these areas, but rather as an entry point that should help the reader locate the relevant literature in this field. Kamp and co-workers (Kamp and Bornholdt 2002; Kamp et al. 2002) have extended earlier work on a virus population that is tracking a moving optimum (Nilsson and Snoad 2000) to the case where the moving optimum is defined by an adaptive immune system, which itself is tracking the virus population (see also Kamp 2003 for a review). Interestingly, in this case not only does the virus population form a quasispecies, but also the immune system can be interpreted as a quasispecies: under B cell proliferation and somatic hypermutation, a mutant cloud of B cells develops that behaves similarly to a viral quasispecies. Kamp and Bornholdt found that both the viral quasispecies and the B cell quasispecies suffer from the classical error threshold if their mutation rates are too high. In addition, the viral quasispecies also suffers from a dynamic threshold if its mutation rate is too low in comparison

Quasispecies in Time-Dependent Environments

45

to that of the B cells. In this case, the virus cannot escape rapidly enough from the immune system, and is driven to extinction. Kamp and co-workers also calculated the optimal mutation rates for both the immune system and the virus and found that both results compare favorably with experimental observations (Kamp and Bornholdt 2002; Kamp et al. 2002). The optimal mutation rate for the virus has a particularly simple and intuitive interpretation: the mutation rate should be such that each viral epitope acquires on average exactly one mutation within the time-frame in which the immune system reacts to that epitope. (Note that this result is essentially identical to the one of Kimura 1967 discussed in the preceding section.) Brumer and Shaknovich (2004) have extended this work to study the effect of semiconservative replication (where both template and copied sequence acquire mutations, as is the case in replication of double-stranded DNA) vs conservative replication (where mutations arise only in the copied sequence, as is the case with most RNA-based viruses). They have found that the optimal viral mutation rate is largely unaffected by the mode of replication, but that the model dynamics away from this optimum differ substantially with the mode of replication. Brumer and Shaknovich suggest that their result might explain why DNA-based viruses tend to have much smaller mutation rates than RNA-based viruses. In general, a substantial amount of modeling work on the interaction between viruses and immune systems has been done by theoretical immunologists. While many of the basic immunological models do not explicitly consider evolution, they provide a natural starting point for more realistic models that do include evolutionary effects. Good introductions to theoretical immunology and its relation to virus dynamics and virus evolution are the book by Nowak and May (2000) and a recent review article by Perelson (2002). Co-evolution of viruses and host cells has been studied experimentally in several viral species, particularly during persistent infection. Infection of BHK-21 cells with foot-and-mouth-disease virus can select cells of increased resistance, which in turn selects viral populations of increased virulence for nonselected host cells (de la Torre et al. 1988; Hernández et al. 1994). Similar results have been reported for experiments with persistent infection of poliovirus (Borzakian 1992), reovirus (Ahmed et al. 1981; Dermody et al. 1993), and mouse hepatitis virus (Chen and Baric 1996). As a consequence, persistently infected cells are less susceptible to viral infection, and this susceptibility tends to correlate with the ability (or lack thereof) of the virus to enter the cells (Borzakian 1992; Dermody et al. 1993; Chen and Baric 1996). Another model of co-evolution that has been studied extensively both theoretically and experimentally is that of defective interfering particles (DIPs)

46

C. O. Wilke et al.

and their full-length wild-type (wt) counterparts. DIPs are mutant viruses with large deletions in their genomes. Because of these deletions, DIPs cannot produce all the proteins needed for replication, but they can use these proteins from co-infecting full-length wt helper viruses. Several authors have studied the dynamics of DIPs and their associated helper viruses theoretically (Bangham and Kirkwood 1990; Szathmáry 1992; Kirkwood and Bangham 1994; Frank 2000). DIPs can accumulate when the multiplicity of infection (MOI) is high, for two reasons: First, at high MOI, DIPs frequently co-infect cells together with wt, and thus have access to the wt proteins that they need for replication. Second, DIPs have a smaller genome size than wt, and thus can replicate faster than wt and use up a larger share of wt protein than the wt does itself. The simplest models of this system predict that DIPs and wt will evolve toward a stable equilibrium, and that a DIP-resistant mutant can occur which will replace the DIP–wt system (Szathmáry 1992). More elaborate models that take into account the changing size of the virus population, for example under recurrent passaging, show that fluctuating levels of DIPs and wt are more realistic, and that the DIP–wt system can spontaneously go to extinction (Bangham and Kirkwood 1990; Kirkwood and Bangham 1994; Frank 2000). For example, extinction happens when a severe bottleneck removes all wt particles from the virus population. While DIPs have been identified in many viral species (reviewed in Holland 1991), persistent VSV infection of mammalian cells is the system of DIP-driven evolution that is characterized in most detail. DIPs accumulate during infection of mammalian cells at high MOI, but not at low MOI. Eventually, DIP-resistant full-length mutant genomes appear, and rise to fixation as they replace both wt and DIPs. These mutants are typically characterized by extensive rearrangement of the genomic termini (O’Hara et al. 1984). If viral replication continues at high MOI, then new DIPs arise that can use the full-length mutant genomes as helper viruses. Thus starts a second phase of interference. Such cycles of the emergence of interfering DIPs followed by the emergence of resistant full-length genomes can proceed for long periods of time (DePolo et al. 1987).

6 Discussion and Conclusions While the first 25 years of quasispecies theory were dominated by studies of static fitness landscapes, in recent years several groups have started to analyze quasispecies dynamic in time-dependent environments. It is comforting to see that by and large, the results derived for static fitness landscapes apply—

Quasispecies in Time-Dependent Environments

47

with only minor modifications—to dynamic fitness landscapes as well. There is no need to replace quasispecies theory with some other theory just because viruses do not live in perfectly stable environments. On the contrary, the theory for periodically changing fitness landscapes, for example, is actually mathematically equivalent to that of a static fitness landscape, and can be analyzed with exactly the same tools. However, dynamic fitness landscapes do of course introduce new effects that cannot be observed in static fitness landscapes. The most dramatic such effect is probably the dynamic error threshold, discovered by Nilsson and Snoad (2000), which puts a lower bound on the minimum mutation rate at which a virus population is still capable of following the changes in the landscape. This dynamic error threshold, in combination with the classical error threshold known from static landscapes, restricts the useful range of mutation rates to a fairly narrow region, and it has been suggested that the mutation rates of several RNA viruses fall right into this region (Kamp and Bornholdt 2002; Kamp et al. 2002). A particularly useful way to describe the overall behavior of an evolving population on a dynamic fitness landscape is to say that the population acts like a low-pass filter. The slower the changes in the landscape, the more accurately the population will track them, and the smaller the time lag from the change in the landscape until the population reacts to it. The faster the changes, on the other hand, the less accurately the population will track them, and the larger the time lag until the population reacts. In the limit of very fast changes, the population will stop to track the changes completely, and simply adapt to the time-averaged fitness landscape. This review has shown that we have a fairly good understanding of quasispecies theory in time-dependent environments, at least on a broad, qualitative level. However, as always, the devil is in the detail, and there remain many open questions. Most importantly, the theoretical models tend to be based on simplified fitness landscapes, in which a single gene sequence provides full function, and any mutation to that sequence destroys or at least severely inhibits the function. However, we know that mutations can come with a broad spectrum of effects, from fully neutral to somewhat deleterious to lethal. In a changing environment, this spectrum may change over time as well, for example as a function of the virus host. One of the main goals of future work in this area must be to determine the detailed structure of realistic fitness landscapes and how exactly they change over time. To realize this goal, close collaboration between theoretical and experimental groups will be necessary. Acknowledgements AI45686.

Work at the Medical College of Ohio is funded by NIH grant

48

C. O. Wilke et al.

References Ahmed R, Canning WM, Kauffman RS, Sharpe AH, Hallum JV, Fields BN (1981) Role of the host cell in persistent viral infection—coevolution of L cells and reovirus during persistent infection. Cell 25:325–332 Bangham CRM, Kirkwood TBL (1990) Defective interfering particles: effects in modulating virus growth and persistence. Virology 179:821–826 Borzakian S, Couderc T, Barbier Y, Attal G, Pelletier I, Colberegarapin F (1992) Persistent poliovirus infection – establishment and maintenance involve distinct mechanisms. Virology 186:398–408 Brumer Y, Shakhnovich EI (2004) Host-parasite coevolution and optimal mutation rates for semiconservative quasispecies. Phys Rev E Stat Nonlin Soft Matter Phys 69:061909 Bürger R (2000) The mathematical theory of selection, recombination, and mutation. Wiley, Chichester, UK Chen W, Baric RS (1996) Molecular anatomy of mouse hepatitis virus persistence: Coevolution of increased host cell resistance and virus virulence. J Virol 70:3947– 3960 Chen W-J, Wu H-R, Chiou S-S (2003) E/NS1 modifications of dengue 2 virus after serial passages in mammalian and/or mosquito cells. Intervirology 46:289–295 Cooper LA, Scott TW (2001) Differential evolution of eastern equine encephalitis virus populations in response to host cell type. Genetics 157:1403–1412 De la Torre JC, Martínez-Salas E, Diez J, Villaverde A, Gebauer F, Rocha E, Dávila M, Domingo E (1988) Coevolution of cells and viruses in a persistent infection of foot-and-mouth disease virus in cell culture. J Virol 62:2050–2058 DePolo NJ, Giachetti C, Holland JJ (1987) Continuing coevolution of virus and defective interfering particles and of viral genome sequences during undiluted passages – virus mutants exhibiting nearly complete resistance to formerly dominant defective interfering particles. J Virol 61:454–464 Dermody TS, Nibert ML, Wetzel JD, Tong X, Fields BN (1993) Cells and viruses with mutations affecting viral entry are selected during persistent infections of L cells with mammalian reoviruses. J Virol 67:2055–2063 Eigen M, Schuster P (1979) The hypercycle—principle a of natural self-organization. Springer, Berlin Heidelberg New York Eigen M, McCaskill J, Schuster P (1988) Molecular quasi-species. Phys J Chem 92:6881– 6891 Eigen M, McCaskill J, Schuster P (1989) The molecular quasi-species. Adv Chem Phys 75:149–263 Frank SA (2000) Within-host spatial dynamics of viruses and defective interfering particles. Theor J Biol 205:279–290 Holland JJ (1991) Defective viral genomes. In Fields B, Knipe D (eds) Fundamental virology. Raven Press, New York, pp 151–165 Holmes EC, Moya A (2002) Is the quasispecies concept relevant to RNA viruses? J Virol J 76:460–462 Jenkins GM, Worobey M, Woelk CH, Holmes EC (2001) Evidence for the nonquasispecies evolution of RNA viruses. Mol Biol Evol 18:987–994

Quasispecies in Time-Dependent Environments

49

Kamp C (2003) A quasispecies approach to viral evolution in the context of an adaptive immune system. Microbes Infect 5:1397–1405 Kamp C, Bornholdt S (2002) Co-evolution of quasispecies: B-cell mutation rates maximize viral error catastrophes. Phys Rev Lett 88:068104 Kamp C, Wilke CO, Adami C, Bornholdt S (2002) Viral evolution under the pressure of an adaptive immune system—optimal mutation rates for viral escape. Complexity 8:28–33 Kimura M (1967) On the evolutionary adjustment of spontaneous mutation rates. Genet Res 9:23–34 Kirkwood TB, Bangham CR (1994) Cycles, chaos, and evolution in virus cultures: a model of defective interfering particles. Proc Natl Acad Sci U S A 91:8685–8689 Li Y, Wilke CO (2004) Digital evolution in time-dependent fitness landscapes. Artificial Life 10:123–134 Martín Hernández MAM, Carrillo EC, Sevilla N, Domingo E (1994) Rapid cell variation can determine the establishment of a persistent viral infection. Proc Natl Acad Sci U S A 91:3705–3709 Nilsson M, Snoad N (2000) Error thresholds on dynamic fitness landscapes. Phys Rev Lett 84:191–194 Nilsson M, Snoad N (2002a) Optimal mutation rates in dynamic environments. Bull Math Biol 64:1033–1043 Nilsson M, Snoad N (2002b) Quasispecies evolution on a fitness landscape with a fluctuating peak. Phys Rev E 65:031901 Novella IS, Clarke DK, Quer J, Duarte EA, Lee CH, Weaver SC, Elena SF, Moya A, Domingo E, Holland JJ (1995) Extreme fitness differences in mammalian and insect hosts after continuous replication of vesicular stomatitis virus in sandfly cells. J Virol 69:6805–6809 Novella IS, Hershey CL, Escarmís C, Domingo E, Holland J (1999) Lack of evolutionary stasis during alternating replication of an arbovirus in insect and mammalian cells. J Mol Biol 287:459–465 Nowak M, Schuster P (1989) Error thresholds of replication in finite populations— mutation frequencies and the onset of Muller’s ratchet. J Theor Biol 137:375–395 Nowak MA, May RM (2000) Virus dynamics. Oxford University Press, Oxford O’Hara PJ, Nichol ST, Horodyski FM, Holland JJ (1984) Vesicular stomatitis virus defective interfering particles can contain extensive genomic sequence rearrangements and base substitutions. Cell 36:915–924 Orr HA (2000) The rate of adaptation in asexuals. Genetics 155:961–968 Perelson AS (2002) Modelling viral and immune system dynamics. Nature Rev Immunol 2:28–36 Perelson AS, Essunger P, Ho DD (1997) Dynamics of HIV-1 CD4+ lymphocytes in vivo. AIDS 11:S17–S24 Ronnewinkel C, Wilke CO, Martinetz T (2001) Genetic algorithms in time-dependent environments. In Kallel L, Naudts B, Rogers A (eds) Theoretical aspects of evolutionary computing. Springer, Berlin Heidelberg New York, pp 261–285 Szathmáry E (1992) Natural selection and dynamical coexistence of defective and complementing virus segments. J Theor Biol 157:383–406 Turner PE, Elena SF (2000) Cost of host radiation in an RNA virus. Genetics 156:1465– 1470

50

C. O. Wilke et al.

Van Nimwegen E, Crutchfield JP, Mitchell M (1999a) Statistical dynamics of the royal road genetic algorithm. Theoretical Computer Science 229:41–102 Van Nimwegen E, Crutchfield JP, Huynen M (1999b) Neutral evolution of mutational robustness. Proc Natl Acad Sci U S A 96:9716–9720 Weaver SC, Brault AC, Kang W, Holland JJ (1999) Genetic and fitness changes accompanying adaptation of an arbovirus to vertebrate and invertebrate cells. J Virol 73:4316–4326 Wilke CO (2001a) Adaptive evolution on neutral networks. Bull Math Biol 63:715–730 Wilke CO (2001b) Selection for fitness versus selection for robustness in RNA secondary structure folding. Evolution 55:2412–2420 Wilke CO, Ronnewinkel C (2001) Dynamic fitness landscapes: expansions for small mutation rates. Physica A 290:475–490 Wilke CO, Adami C (2002) The biology of digital organisms. Trends Ecol Evol 17:528– 532 Wilke CO, Adami C (2003) Evolution of mutational robustness. Mut Res 522:3–11 Wilke CO, Ronnewinkel C, Martinetz T (2001a) Dynamic fitness landscapes in molecular evolution. Phys Rep 349:395–446 Wilke CO, Wang JL, Ofria C, Lenski RE, Adami C (2001b) Evolution of digital organisms at high mutation rate leads to survival of the flattest. Nature 412:331–333 Zárate S, Novella IS (2004) Vesicular stomatitis virus evolution during alternation between persistent infection in insect cells and acute infection in mammalian cells is dominated by the persistence phase. J Virol 78:12236–12242

CTMI (2006) 299:51–82 c Springer-Verlag Berlin Heidelberg 2006


Viruses as Quasispecies: Biological Implications E. Domingo (u) · V. Martín · C. Perales · A. Grande-Pérez · J. García-Arriaza · A. Arias Centro de Biología Molecular “Severo Ochoa” (CSIC-UAM), Universidad Autónoma de Madrid, Cantoblanco, 28049 Madrid, Spain [email protected]

1

Introduction: How a Theory Met Reality and Vice Versa . . . . . . . . . . . . . 52

2

Molecular Characterization of RNA Virus Populations . . . . . . . . . . . . . 55

3

The Importance of Being Mutant Spectra . . . . . . . . . . . . . . . . . . . . . . . 58

4 4.1

Suppressive Effects of Mutant Spectra . . . . . . . . . . . . . . . . . . . . . . . . . 61 Consequences for Fitness Determinations . . . . . . . . . . . . . . . . . . . . . . . 64

5

Interactions Within Mutant Spectra: A Proposal . . . . . . . . . . . . . . . . . . 65

6 6.1

Additional Roles for Minorities: Memory in Viral Quasispecies . . . . . . . 69 Memory and Long-Term Evolutionary History: A Distinction . . . . . . . . . 71

7

Virus Entry into Error Catastrophe: An Antiviral Strategy Derived from Quasispecies Dynamics . . . . . . . . . . 72

8

Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

Abstract During viral infections, the complex and dynamic distributions of variants, termed viral quasispecies, play a key role in the adaptability of viruses to changing environments and the fate of the population as a whole. Mutant spectra are continuously and avoidably generated during RNA genome replication, and they are not just a by-product of error-prone replication, devoid of biological relevance. On the contrary, current evidence indicates that mutant spectra contribute to viral pathogenesis, can modulate the expression of phenotypic traits by subpopulations of viruses, can include memory genomes that reflect the past evolutionary history of the viral lineage, and, furthermore, can participate in viral extinction through lethal mutagenesis. Also, mutant spectra are the target on which selection and random drift act to shape the long-term evolution of viruses. The biological relevance of mutant spectra is the central topic of this chapter.

52

E. Domingo et al.

1 Introduction: How a Theory Met Reality and Vice Versa The historical overview of evolutionary virology by John Holland in the closing chapter of this volume documents the many indications of genetic and phenotypic instability of RNA viruses, and the growing suspicion over the second half of the twentieth century that something was fundamentally different between RNA genetics and DNA genetics. The first evidence that an RNA virus consisted of distributions of mutant genomes, and that the wild type existed only as a weighted average of many different sequences, was obtained by sampling genomic sequences of biological clones of bacteriophage Qβ by Charles Weissmann and colleagues in Zurich (review in Holland, this volume). Weissmann presented these experimental results to Manfred Eigen and his colleagues at a Max Planck Institute meeting in Klosters (Switzerland) in 1978. Eigen had developed the first theoretical treatment of a system that replicated with limited fidelity, so that the replication process regularly generated mutant copies of the templates (Eigen 1971). This theory, later to become known as quasispecies theory (Eigen and Schuster 1979), was developed to understand self-organization and adaptability of early life forms on earth. In seeing the experimental results on phage Qβ, Eigen exclaimed “Quasispecies in reality!” and this represented the beginning of a productive interaction between theoretical biophysics and experimental virology (further information on these early encounters is given in Domingo et al. 1995). Quasispecies theory is covered in the chapters by Biebricher and Eigen and by Wilke et al. in this volume, and here we review biological implications of quasispecies dynamics for RNA viruses. The extended definition of quasispecies currently used by virologists is the following: “Viral quasispecies are dynamic distributions of non-identical but closely related mutant and recombinant viral genomes subjected to a continuous process of genetic variation, competition and selection, and which act as a unit of selection” (Domingo 1999). This definition incorporates general principles of Darwinian evolution, whose effects, in the case of viruses, can be observed within short time periods (usually days and weeks) both in natural hosts, and in model systems such as alternative hosts in vivo or in cell cultures. Short-term evolution of viruses underlies virus adaptation to compartments within infected organisms, that may contribute to time-dependent changes of disease symptoms and to disease progression. This is particularly evident with RNA viruses, and several examples are discussed in this volume, and others are regularly reported in the current literature on virology. Some terms (such as “fitness”, “environment”, etc.) used in this and other chapters, and that have been approached mainly with viral infections in cell culture, may seem abstract and distant

Viruses as Quasispecies: Biological Implications

53

from the real world of viral diseases. Yet they are not. Fitness values express a replication capacity, and competitive growth occurs in infected hosts when variant forms of the same virus meet to penetrate a compartment of a differentiated organism, or to replicate in the same cell subset. Environment may mean specific cell types or cell assemblies in tissues and organs where virus replication takes place, or a set of physiological conditions that may affect virion stability or susceptibility of cells to viral variants. The main steps involved in middle-term and long-term RNA virus evolution have some resemblance with some of the steps that determine shortterm evolution in quasispecies dynamics (Fig. 1), despite their occurring with very different space-time scales. The triggering event in these evolutionary episodes is reproduction with genetic variation. Then positive and negative selection, together with random sampling (drift) that take place within a host, or between host individuals, shape the genetic composition of the virus. Reproductive success can be quantified as relative fitness values (QuiñonesMateu and Arts, this volume) or with epidemiological parameters such as the basic reproductive rate (or ratio) (Ro), defined as the average number of secondary cases resulting from the introduction of a single infected case into a susceptible population. Ro is a general parameter (which can be applied to demography of any type of biological entity) that can predict the long-term epidemic spread of a pathogenic agent (reviews concerning application to viral infections in Nowak and May 2000 and Woolhouse 2004). Both fitness and Ro values capture average values of fundamentally heterogeneous entities, and such averages may vary as an infection progresses or an epidemic spreads, thus adding an additional level of complexity to the interpretation of the reproductive success of a virus. While the reproductive ratio is generally used in epidemiological investigations, fitness finds its application in the comparison of the relative replication capacity of variant viruses within a single host or in cell culture. As indicated by Biebricher (1999), evolutionary success depends on two components of the phenotype: those that determine survival and those that determine the rate of production of viable progeny, and the combination of the two is what we call fitness. Virologists have adapted growth-competition experiments to measure the relative capacity of two viruses to produce infectious progeny, thus providing estimates of relative fitness values under a given set of environmental conditions (Holland et al. 1991). Fitness can be measured in other ways (DeFilippis and Villarreal 2000), and it is virtually impossible to design a measurement that captures in full the potential evolutionary success of viruses replicating in their host organisms (Biebricher 1999; DeFilippis and Villarreal 2000). Despite limitations in the measurements and significance of fitness values for RNA viruses, variations in relative fitness, based on

54

E. Domingo et al.

a

b

Fig. 1a,b Schematic representation of viral quasispecies in an infected host. Viral genomes are represented as horizontal lines, and mutations as symbols on the lines. a Upon infection with an RNA virus (even with a single particle, as depicted here, enlarged about 106 times), viral replication leads to a mutant spectrum of related genomes, termed viral quasispecies. Nucleotide sequencing of the ensemble produces a consensus sequence which includes in each of its positions (nucleotide or amino acid) the residue found most frequently in the corresponding position of the distribution (mutant spectrum). b Different mutant distributions are found in different infected organs or at different sites of the same organ. As further discussed in the text, in real infections multiple mutant spectra that can amount to a large number of replicating (or potentially replicating) genomes (up to 109 or even 1012 per infected individual) provide highly dynamic mutant repertoire

Viruses as Quasispecies: Biological Implications

55

viral yields in cell culture, have been immensely powerful in characterizing the population dynamics of RNA viruses (see references in the reviews by Domingo and Holland 1997; Quiñones-Mateu and Arts 2002; Novella 2003; and the chapters by Quiñones-Mateu and Arts and Escarmís et al., this volume). As usually performed, fitness values incorporate differences in replication capacity (both of viruses in isolation and when co-replicating in the same cells) at all stages of the viral life cycle. Restricting fitness measurements to individual events of the replication cycle (RNA synthesis, viral-specific protein synthesis, virion assembly, etc.) would entail additional ambiguities (i.e. RNA replication may be coupled to translation, translation to assembly, etc.). Quiñones-Mateu and Arts in the next chapter of this volume discuss several important implications of fitness measurements based on viral yields. Figure 2 shows a schematic view of quasispecies dynamics in relation to population size and fitness variations. But there are additional implications of the links between quasispecies theory and virus population dynamics.

2 Molecular Characterization of RNA Virus Populations Since RNA viruses replicate as complex mutant distributions (Figs. 1 and 2), determination of the consensus nucleotide sequence (or the consensus amino acid sequence of encoded proteins) provides very fragmentary information of the genetic composition and of the evolutionary potential of a virus population. Analyses of individual genomic sequences of mutant spectra can be achieved by two alternative procedures. One is to isolate virus either from individual plaques developed on cell monolayers or from an infection following end-point dilution. Viral RNA is then subjected to reverse transcriptionpolymerase chain reaction (RT-PCR) amplification and nucleotide sequencing. This procedure leads a sequence that cannot be influenced by possible misincorporations introduced during the RT-PCR amplification (that can arise due to the limited copying fidelity of the enzymes used in the amplification). In sequence screening of biological clones, a bias may occur that favours representation of genomes that are more infectious (produce early cytopathology or larger plaques) in the particular cell line or primary culture chosen. A second means to characterize a virus population is to subject total RNA extracted from the biological specimen of interest to RT-PCR, followed by molecular cloning and sequencing of DNA of individual molecular clones. This procedure does not depend on infectivity of the viral RNA in a cell culture system. Here a bias may arise from the low fidelity of the enzymes used in the RT-PCR, which may result in an overestimate of the nucleotide sequence

56

E. Domingo et al.

Fig. 2 Scheme of quasispecies dynamics and fitness variations. Top: Viral genomes are depicted as horizontal lines and mutations as symbols on the lines (see also Fig. 1). Small arrows indicate genetic bottlenecks experimentally realized as plaque-to-plaque transfers. Serial bottleneck events lead to accumulation of mutations in the consensus sequence. The large arrow represents replication of the quasispecies without population size limitations. Bottom: Fitness (F) variation associated with passage regime. Bottleneck events lead to a decreases in fitness; however, at low fitness values a fluctuating pattern with a constant average fitness is observed (Escarmís et al., this volume). Large population passages result in a fitness increase that may or may not result in a modification of the consensus sequence. Again, at high fitness, a fluctuating pattern of fitness values is observed, presumably reflecting a limitation exerted by the population size on the capacity for fitness increase (Novella et al. 1995, 1999). (Figure adapted from Domingo 1999, with permission)

heterogeneity. This can be solved by using high-fidelity polymerases, correct amplification reaction conditions (adequate pH and ionic composition, unbiased concentrations of chemically intact nucleoside triphosphates, etc.), and control experiments to determine a basal error rate for the system (Arias et al. 2001). Another bias may come from insufficient initial template RNA so that several sequences that originate from the same viral template RNA may be represented among the clones analyzed. This will generally result in an underestimate of the nucleotide sequence heterogeneity. Some simple calculations have been applied to derive the expected number of independent sequences (those that will represent different templates in the viral population) from the

Viruses as Quasispecies: Biological Implications

57

last dilution of the template RNA that yields positive RT-PCR amplification (Airaksinen et al. 2003). Our rule of thumb is that when a 1:100 dilution of the template yields positive RT-PCR amplification, we proceed with the cloning and sequencing of the amplification product of the undiluted template. When the amount of template is limiting, alternative procedures (such as multiple parallel amplifications prior to cloning and sequencing) must be sought (Airaksinen et al. 2003). It is a common practice to align at least 20 genomic (or subgenomic) sequences and to determine the mutation frequency and Shannon entropy (Table 1). In most situations, 20–100 sequences represent a tiny minority of Table 1 The characterization of mutant spectra of viral quasispecies 1. Mutation frequency Definition: The proportion of mutant nucleotides in a genome population. Mutant frequency may refer to the proportion of genomes harbouring a specific mutation. Calculation: Determine the total number of mutations (counting repeated mutations only once) relative to the consensus (defined by the same set of sequences) and divide by the total number of nucleotides that have been sequenced (i.e. 10,000 when the same 500 nucleotides of 20 individual genomes have been determined). In some cases, it may be interesting for statistical reasons to count all mutations found (counting repeated mutations as many times as they occur), yielding a maximum mutation frequency. 2. Shannon entropy Definition: The proportion of different genomes in a mutant distribution
Calculation: Normalized Shannon entropy, Sn = − [ i (pi × lnpi )] | ln N, in which pi is the proportion of each different sequence of the mutant spectrum, and N the total number of sequences compared. 3. Genetic distance and Hamming distance Definitions: Genetic distance is the number of mutations that distinguish any two sequences from the population. The average for all possible pairs reflects the genetic complexity. Hamming distance is the number of mutated positions in a genome with respect to the best adapted sequence (the most abundant) in a genome distribution. For quasispecies analysis, a list of Hamming distances (or the average for the population) can characterize the complexity of the ensemble. Calculation: Align sequences and divide the number of mismatched positions between any two sequences by the number of identical positions. Matrices of pair-wise genetic distances are used to determine phylogenetic trees. For distantly related sequences (rarely occurring within a quasispecies), multiple sequence alignments using CLUSTAL W and adequate scoring of gaps may be needed. Based on Eigen 1992; Volkenstein 1994; Domingo 1996; Mount 2004; Domingo et al. 2005.

58

E. Domingo et al.

the total number of viral genomes in a biological specimen, and therefore this procedure must be regarded as a very crude sampling of a viral population. In some cases, a selective agent may permit the screening of minority components of mutant spectra. For example, in studies on quasispecies memory (described in Sect. 6) analyses of mutant spectra were based on the repertoire of mutants resistant to a monoclonal antibody that neutralized the dominant FMDV population, but did not neutralize the portion of the mutant spectra of interest (Ruiz-Jarabo et al. 2002, 2003b). Despite these limitations, determination of nucleotide sequence heterogeneities in virus populations using correct reagents and adequate controls has consistently documented that most RNA viruses (and also some DNA viruses) consist of complex mutant spectra, with an average number of 1–100 mutations per genome (Sect. 3). The degree of heterogeneity within a viral population will be influenced by the fidelity of the replication machinery, the distance (in rounds of replication) from a clonal origin, constraints for variation (both at the RNA and protein levels), and intervening bottlenecks in the evolutionary history, among other influences. Scientists should be cautious in attributing to high-fidelity RNA replication what may be a standard error level together with negative selection, which maintains an invariant consensus nucleotide sequence. The literature contains several cases of such likely misinterpretation of data. It is not surprising that comparisons of different virus–host systems have led to exceedingly broad ranges of genetic diversity within mutant spectra. This should not blur the common underlying influences and the general biological relevance of population complexity even if seemingly modest in terms of mutations per individual genome. This is justified in the next section.

3 The Importance of Being Mutant Spectra The relevance of viruses replicating as mutant spectra is intuitively obvious since any individual mutant genome of the ensemble can potentially differ in behaviour from other individuals or from the ensemble of genomes. The importance of such a population structure is strengthened by considering five parameters that characterize a mutant distribution: average number of mutations per genome, virus population size, genome length, mutations needed for a phenotypic change, and virus fecundity (Table 2). Despite all cellular organisms being highly polymorphic genetically (in that distinct alleles from a gene are represented among individuals of one biological species), the level of heterogeneity of RNA virus populations confers a much greater adaptability

Viruses as Quasispecies: Biological Implications

59

than the levels of polymorphism estimated for cells. This is a consequence of parameters 2 and 3 (listed in Table 2): as an example, a viral genome of 10,000 nucleotides has a total of 3 × 104 possible single mutants (disregarding fitness effects of mutations), which is a figure well below the population size of many natural virus populations. In contrast, the total number of possible single mutants for a mammalian genome is about 1010 , well above the population size of mammalian species. That is, the capacity to explore sequence space (a concept discussed in the chapter by Biebricher and Eigen in this volume, which refers to the total number of possible sequences available to a genetic system, reviewed by Eigen and Biebricher 1988) is far greater for viruses than for cellular organisms. This confers adaptability to viruses (with amply recognized biological implications) and renders viruses suitable experimental systems for probing evolutionary concepts (see Domingo et al. 2001; Flint et al. 2004, and other chapters of this volume). One of the critical parameters in viral quasispecies is the number of mutations in an RNA virus that is needed for a phenotypic change in the virus (Table 2). Indeed, if a relevant phenotypic change (for example, a modification in host cell tropism, resistance to neutralizing antibodies or to an antiviral agent, etc.) depended on the occurrence of 50–100 mutations in a viral genome (to invent a simple example), then the quasispecies nature of RNA viruses (with the characteristic parameters we measure today; Table 2) would

Table 2 Some important parameters that influence the adaptability of viral quasispecies 1. Average number of mutations per genome in a mutant spectrum Generally it amounts to an average of 1–100 mutations per genome. (See text for reasons for broad range). 2. Virus population size Variable, but very high upper limits. An acutely infected organism may include 109 –1012 viral particles at any given time. Even a single viral plaque on a cell’s monolayer can yield 103 –1010 particles. 3. Genome length 3 kb–32 kb 4. Mutations needed for a phenotypic change Many recorded adaptive changes depend on one or a few mutations (see text). 5. Fecundity Variable. Average of < 1–106 particles per cell have been measured. High fecundity promotes quasispecies dynamics, as discussed in several chapters of this volume. Based on Domingo et al. 2001 and references therein.

60

E. Domingo et al.

be largely irrelevant for short-term adaptation. This is because in the exploration of sequence space the probability of arriving at a point corresponding to a mutant with the required 50–100 mutations is exceedingly low (10–200 to 10–400 !). Even if we relax the requirement to multiple combinations of mutations, it can be estimated from the Poisson distribution that the probability of genomes with any 50 mutations in a population of 1011 individuals is about 10–44 genomes (on a purely statistical basis, ignoring fitness effects; additional representative figures can be found in Domingo 1997). As evidenced by many experimental results with RNA viruses of virtually all families that have been examined, one or a few amino acid replacements are sufficient to modify a biologically relevant feature of a virus, as amply recorded in the literature (review in Domingo et al. 2001; examples of single mutations associated with epidemiologically relevant adaptations of arboviruses are given in the chapter by Weaver in this volume). Because phenotypically relevant mutations can be frequently represented in mutant spectra, they have the potential to become dominant (command the development of new mutant distributions) in response to environmental demands, despite the modulating effects of mutant spectra described in Sect. 4. Multiple constraints for variability have been evidenced in viral genomes (Simmonds and Smith 1999; Simmonds et al. 2004). Therefore, many variants with one or few mutations that may confer biologically relevant, adaptive traits may not be represented in a mutant spectrum (under a set of environmental conditions) due to fitness costs. It is impressive that despite such constraints (for example, genome-scale, ordered RNA structures in hepaciviruses; Simmonds et al. 2004), the level of heterogeneity and adaptive capacity of viral populations are the ones we record with unfailing continuity with any virus we examine in some detail. A highly significant example is the occurrence of mutants with one or a few amino acid substitutions that confer decreased sensitivity to antiviral inhibitors. This is a general phenomenon – documented with many viruses since the 1960s both in vivo and in cell culture – which complicates enormously the treatment of viral disease (a recent overview can be found in Domingo 2003; see also previous versions published in Progress in Drug Research for a historical account). It is not the case that “suddenly” a virus strain “appears” that is resistant to an inhibitor. Resistant mutants are generated as components of mutant spectra and then may become dominant when virus replication occurs in the presence of the inhibitor. Perhaps the most dramatic example has been the selection over the years of mutants of human immunodeficiency virus type 1 (HIV-1) resistant (or with decreased sensitivity) to the antiretroviral inhibitors (targeted mainly to reverse transcriptase, protease or surface structures involved in virus fusion or cell recognition) used in clinical practice (see the chapter by Mullins and Jensen, this volume). Both experi-

Viruses as Quasispecies: Biological Implications

61

mental analyses of HIV-1 populations (Coffin 1995; Nájera et al. 1995) and theoretical predictions (Ribeiro et al. 1998; Gerrish and Garcia-Lerma 2003) suggest that such mutants may preexist in HIV-1 populations, even when they have not been exposed to the inhibitors. In addition, resistant mutants that may occupy very low frequency levels in HIV-1 mutant spectra in the absence of the inhibitor may be raised to higher frequency levels in the presence of the inhibitor. The recently described phenomenon of memory in viral quasispecies, including memory subpopulations of HIV-1 in vivo (Sect. 6), supports even further a role of genome subpopulations in the response of viruses to inhibitors (Domingo et al. 2003). The fitness costs of inhibitor-resistance mutations, as well as relative fitness values of wild-type and resistant mutants in the absence and the presence of the inhibitor will influence the kinetics and degree of dominance of inhibitor-resistant mutants (treated in the chapter by Quiñones-Mateu and Arts, this volume). Paradoxically, the presence of a mutant with a specific phenotypic trait may not be sufficient to guarantee its dominance even when a selective constraint favours that phenotypic trait. The reason for this important phenomenon again has to do with the presence of a mutant spectrum surrounding the relevant mutant, and this is discussed next.

4 Suppressive Effects of Mutant Spectra A mutant virus potentially capable of becoming dominant in an evolving viral quasispecies (either because of its high fitness or because it harbours a selectable trait) may remain as a minority in the population, depending on the nature of the mutant spectrum in which it is immersed. This concept was first documented with a numerical example of two master sequences of small size differing in fitness by 10%, replicating near the error threshold (see the chapters by Biebricher and Eigen and by Wilke et al., this volume, and Sect. 7 in this chapter). Interestingly, the inferior mutant outgrew the fitter one when the inferior mutant was surrounded by 50 closely related mutants that were somewhat better adapted than the mutants that surrounded the fitter master. This and other numerical simulations (Swetina and Schuster 1982; review in Eigen and Biebricher 1988) suggest a strong influence of the mutant spectrum on the behaviour of any particular variant and supports the view that the target of selection is not a single species, but rather the distribution of the quasispecies as a whole (Eigen and Biebricher 1988; Perales et al. 2005). The prediction that the mutant spectrum can affect the dominance of an individual mutant has found experimental confirmations both in vivo and in

62

E. Domingo et al.

cell culture (Table 3). The first evidence was obtained using vesicular stomatitis virus (VSV) (de la Torre and Holland 1990). A mutant spectrum of VSV suppressed replication of a mutant of superior fitness than the ensemble, unless the mutant was present above a certain frequency in the population. In another instance of remarkable practical relevance, it was found that in anti-poliomyelitis vaccine preparations, the dominant attenuated poliovirus (PV) suppressed neurological disease associated with minority virulent PV, unless the latter was present above a minimal concentration in the vaccine (Chumakov et al. 1991). Some strains of the arenavirus lymphocytic choriomeningitis virus (LCMV) (see the chapter by Sevilla and de la Torre, this volume) induce a hormone-deficiency syndrome in mice (reviewed by Oldstone 2002). Remarkably, some co-infecting nonpathogenic strains of LCMV suppressed expression of pathogenic strains in such a way that no disease was manifested (Teng et al. 1996). Studies with foot-and-mouth disease virus (FMDV) have provided two additional examples (recent reviews on FMDV in Rowlands 2003; Sobrino and Domingo 2004; Mahy 2005). FMDV serially passaged in BHK-21 cells in the presence of polyclonal antibodies directed to a specific antigenic site of the virus, generated a complex mutant spectrum of antigenic variants of low relative fitness. Such mutant spectra included biological clones that manifested high fitness when separated from the mutant cloud (Borrego et al. 1993). In the transition to error catastrophe of FMDV (Sect. 7), pre-extinction viral RNA (which is defined as viral RNA extracted from the population that, in the serial passages in the presence of mutagens, precedes the one in which virus extinction occurs) interferes with infectious RNA (González-López et al. 2004). Interference was documented by co-electroporation of cells with two RNAs and it was exerted by pre-extinction RNA but was not exerted by a defective FMDV RNA (containing an in-frame deletion), unrelated viral and nonviral RNAs, or the same pre-extinction RNA reduced in size by chemical treatment. It was proposed that due to mutations in pre-extinction viral RNA, abnormal expression patterns and expression of abnormal viral proteins may jeopardize the replication capacity of coexisting infectious RNA (González-López et al. 2004). In support of this proposal, FMDV 3Ds (polymerases) harbouring deleterious mutations have been identified in mutagenized populations (Sierra et al. 2000; Arias et al. 2005; see also Sect. 5 for additional information on possible mechanisms of suppressive effects of mutant spectra). An alternative model, termed positive clonal interference, taken from bacterial population genetics, was proposed to explain dominance of one VSV clone over another (Miralles et al. 1999). In this model, the basis for the interference is competition among genomes carrying advantageous mutations during replication as large populations. In addition to a technical problem

Viruses as Quasispecies: Biological Implications

63

derived from the likely presence of defective interfering (DI) particles in the co-infections carried out to test the model (see Domingo 2003 for a more detailed discussion of this problem), recent results with FMDV show that suppression does not require comparable fitness of the two competing populations (González-López et al. 2005), in agreement with a suppressive effect exerted by a low fitness mutant distribution on genomes displaying higher fitness (Sect. 5). The suppression of individual genomes by mutant spectra (Table 3) confers additional biological relevance to population bottleneck events that may intervene during virus replication and evolution. In addition to effects on virus population dynamics (see the chapter by Escarmís et al., this volume), and in the relationship between virulence and transmission mode (Bergstrom et al. 1999), bottlenecks may release a portion of mutant spectrum from the complete cloud that may modulate the behaviour of minority genomes (Fig. 3). Also, some contradictory results seen in the literature on the frequency of deleterious mutations or epistatic effects of mutations in RNA viruses may in part derive from the fact that it is not possible to “freeze” a mutant to remain as the “same” individual in the course of replication: a cloud forms immediately and unavoidably. Thus, what initially was “a” genome with one or two mutations of interest will soon become a “distribution” of genomes

Table 3 Modulating effects of mutant spectra on individual virus variants Cell culture Mutant spectrum of vesicular stomatitis virus (VSV) suppresses a VSV variant of superior fitness (de la Torre and Holland 1990). Low fitness antibody-escape population of foot-and-mouth disease virus (FMDV) suppresses individual antigenic variants displaying high fitness (Borrego et al. 1993). Pre-extinction FMDV RNA interferes with infectious FMDV RNA (González-López et al. 2004). Defective lymphocytic choriomeningitis virus (LCMV) contribute to suppression of infectivity and extinction of the virus (Grande-Pérez et al. 2005). In vivo Attenuated poliovirus (PV) can suppress neurological disease in monkeys, associated with virulent PV (Chumakov et al. 1991). Nonpathogenic LCMV can suppress manifestation of the growth hormone-deficiency syndrome in mice, associated with pathogenic LCMV strains (Teng et al. 1996).

64

E. Domingo et al.

Fig. 3 Scheme of the effect of a population bottleneck on the replication of an individual viral genome. Lines represent a complex mutant spectrum in which individual mutations have not been indicated. An individual genome characterized by two specific mutations (dot and rectangle) is surrounded by a mutant spectrum, which suppresses its replication by the biochemical mechanisms discussed in the text (see also Fig. 4). A population bottleneck represented by the shaded rectangle relieves the individual genome of part of the suppressing ensemble, thereby facilitating expression of the individual genome, which may increase its frequency (right). Experimental evidence and references are given in the text

with additional mutations that may affect the behaviour of the genomes harbouring the initial mutations. Careful monitoring of progeny sequences and mutant spectrum complexities is needed. It is tempting to suggest that some of the uncertainties that appear to characterize virus evolution in vivo (for example, that an antibody-, CTL- or inhibitor-resistant variants may raise to dominance or not) may relate to suppressive effects of mutant spectra perturbed by intra-host bottleneck events of different intensities, prompted by the replicative characteristics of the virus in its permanent interaction with of the host. As further penetration into the fine population structure of viruses in vivo becomes technically feasible, it may be possible to elucidate some of these issues. 4.1 Consequences for Fitness Determinations In the course of fitness determinations [usually by growth competition between the virus to be tested and a reference virus in a defined biological

Viruses as Quasispecies: Biological Implications

65

environment (see the chapter by Quiñones-Mateu and Arts, this volume)] suppressive effects of mutant spectra may also influence the outcome of the competition. Indeed, as discussed in Sect. 1, the mutation rates during RNA genome replication and retrotranscription render extremely unlikely the sustained synthesis of genomes without mutations (Drake and Holland 1999) (although, interestingly, evolution appears to have adjusted mutation rates to permit the presence and survival of the nonmutated class when the latter are fitter than mutant progeny). Thus, despite the fact that during fitness determination a competition is established that should result in dominance of the most fit mutant distributions, modulating effects of mutant spectra of the type discussed in the previous section cannot be excluded, and they have been recently evidenced in the extreme case of fitness determinations of pre-extinction FMDV populations (Gonzalez-Lopez et al. 2005). The converse effect, that is, complementation among mutants (Moreno et al. 1997; Agol 2002), can also occur at the stage of fitness determination in which the two competing ensembles co-infect the same cells, and this effect has been modeled (Wilke et al. 2004) (see the chapter by Wilke et al., this volume). Taking relative virus yield in a given environment as a measure of fitness is justified because first, no virus genome ever exists as a “single nucleotide sequence”; and second, when it exists (as in the very beginning of a formation of a plaque on a cell monolayer) it rapidly becomes a mutant spectrum in the first infected cell. Thus, initial RNA replication rates are fictitious for these reasons and others summarized in Sect. 1, supporting the adequacy of fitness values given as relative viral yields (see Sect. 1 and the chapter by Quiñones-Mateu and Arts, this volume).

5 Interactions Within Mutant Spectra: A Proposal It is a rooted tradition, not only among virologists but also in some schools of population genetics, to view the individual as the standard unit of selection, although this is a debated subject (as a general discussion of this topic see Lewontin 1970; Williams 1992). In this section, we argue that since RNA viruses consist of mutant spectra, and components of the spectra may (and often do) deviate genotypically and phenotypically from other components of the same spectrum, the behaviour of an individual may (and often will) be influenced by surrounding individuals that we call “the ensemble” (extreme cases of suppression of individual variants are discussed in Sect. 4). We further suggest that this feature of viral quasispecies is consistent with many observations made in classical viral genetics.

66

E. Domingo et al.

Viral interference provides a useful introduction to our argument. Interference referred initially to the inhibition of multiplication exerted by one virus on an entirely different virus entering the same cell (see reviews in White and Fenner 1986; Youngner and Whitaker-Dowling 1999; Condit 2001). This led to the distinction between homologous and heterologous interference, depending on the relatedness of the two viruses under analysis. Interference acquired its major impetus with the discovery of defective, noninfectious genomes (generally deletion mutants) that interfered with the corresponding standard virus and mediated the establishment of persistent infections of some RNA viruses (see reviews in Huang and Baltimore 1970; Perrault 1981; Holland 1990). The interfering genome, whether infectious or noninfectious, often has a higher affinity than the standard (interfered) virus for a viral or host protein (or other host component). Here a conceptual overlap with the concept of fitness exists (García-Arriaza et al. 2005). Most fitness measurements involve a phase of intracellular competition between the two viruses to be tested. This is because even in competitions started at low MOI there will be a second round of cell infection that, depending on the initial yield, may involve high MOI. This point has been discussed by (Novella 2004) regarding complementation, and it applies to fitness assays as well. When intracellular competition occurs, the difference in progeny production may have the same molecular basis as interference. This has been manifested in recent studies that have shown a fitness-dependent interference by a bipartite version of the FMDV genome (García-Arriaza et al. 2004, 2005). We arrive at the dilemma that either fitness values should be determined avoiding any type of intracellular competition, or it must be accepted that fitness differences (generally measured for mutants of the same virus) are influenced by intracellular competition events akin to those underlying viral interference. Intracellular competition could be avoided either by each virus infecting cells on separate culture dishes or animal hosts, or by restricting replication to the initial round of progeny production following infection at low MOI; both requirements pose technical problems. Interference may be mediated by interferon, which evokes a general antiviral state in a cell, as well as by a number of interactions between the interfering and interfered viruses (White and Fenner 1986; Youngner and Whitaker-Dowling 1999; Condit 2001; Agol 2002). Concerning possible interference between mutants of the same virus (the form that may be most relevant to quasispecies behaviour), interference may be associated with multiple, different mutants which, even if present individually at low frequency, may have collectively the effect of a dominant-negative mutant. The latter may act through different mechanisms: for example, in the “rotten apple” hypoth-

Viruses as Quasispecies: Biological Implications

67

esis, a defective polypeptide produced by the mutant may enter a multimeric complex, inhibiting its activity. In the “road-block” hypothesis, a defective function (protein or regulatory region) may sequester a factor (or a site) required for replication of both viruses. In the “all things in moderation” hypothesis, imbalances in gene dosage may lead to decreases in efficiency of steps in the life cycle of both viruses. In the “direct competition” hypothesis, both viruses compete for some viral or host factor. According to the “attractive genome” hypothesis, the dominant-negative virus may possess sites that bind required factors, and such sites may be more abundant or show higher affinity than the sites in the wild-type virus. These different hypotheses have been taken from the summary by Youngner and Whitaker-Dowling (1999), and we propose that at least some of these mechanisms may underlie modulating effects observed within mutant spectra of viral quasispecies (Fig. 4). This was proposed to explain the interfering activity of pre-extinction FMDV RNA (González-López et al. 2004) and it emphasizes increasing evidence of the critical role that defective genomes may have in dictating virus behaviour (Grande-Pérez, et al., 2005; see also the chapter by Escarmís et al., this volume). Within mutant spectra, multiple potentially dominant-negative mutants (when present at high frequency) may have a synergistic effect in preventing standard virus replication by one or several of the proposed mechanisms (Youngner and Whitaker-Dowling 1999). These negative effects on replication of individual mutants may be favoured by the multifunctional nature of viral proteins (both in mature and precursor forms), increasingly recognized for many viruses (Cornell et al. 2004; Flint et al. 2004). Parallel concepts can be applied both to modulating effects of mutant spectra (Table 4) and to the transition to error catastrophe (González-López et al. 2004; Grande-Pérez et al., in press; see also Sect. 7). Mutants with complementing or interfering capacity have been referred to as cooperators or defectors, respectively, during processes of competition inside a host (Turner and Chao 1998; Novella 2004; Novella et al. 2004). What we know of quasispecies’ capacity to attain different fitness levels suggests that the same altered viral protein may be advantageous or disadvantageous depending on the dominant genomes in the mutant distribution (Fig. 4). Thus, our proposal is that a mutant spectrum of a replicating viral quasispecies constitutes a “genetic microcosmos” in which interactions among components of a mutant spectrum may include effects similar to those previously characterized between well defined mutants of a virus or between two different viruses. What is observed to occur “between populations” is now extended to interactions among “components of a population”. Experiments are now in progress to further test the validity of this proposal.

68

E. Domingo et al.

Fig. 4a–c Scheme of the effect of a trans-acting viral gene product on the replicative efficacy of a viral quasispecies. A Three forms of the same viral gene (1, 2, 3; shaded box in genome; symbols represent mutations) encode a protein with decreasing biological activity (white to dark grey) that acts as a homopolymeric hexamer (needed to produce viral progeny). B In a mutant spectrum in which 1 and 2 are expressed, increasing proportions of gene 2 will result in a decrease in activity of the hexameric protein, and, consequently of viral yield. C In a mutant spectrum in which 2 and 3 are expressed, increasing proportions of gene 3 will result in a decrease in activity of the hexameric protein, and, consequently of viral yield. Note that the same viral subunit can act as a negative modulator (B) or a positive modulator (C) depending on the activity of the accompanying subunit. In real viral quasispecies, modulating effects are far more complex than illustrated here because a range of different activities in trans-acting products may be present, several viral and host gene products may participate in heteropolymeric complexes, viral proteins are often multifunctional, and regulatory signals on the viral RNA may be involved

Viruses as Quasispecies: Biological Implications

69

6 Additional Roles for Minorities: Memory in Viral Quasispecies Viral quasispecies may possess in their mutant spectra minority genomes that reflect those that were dominant at an earlier phase of quasispecies evolution. This has been shown experimentally with FMDV in cell culture (Ruiz-Jarabo et al. 2000, 2002; Arias et al. 2001, 2004; Baranowski et al. 2003) and HIV-1 in vivo (Briones et al. 2003; Briones et al., personal communication). The experiments with FMDV took advantage of two well-studied classes of mutants: an antigenic variant selected with a neutralizing mAb and a very unusual variant that contained in its genome an internal polyadenylate tract, generated upon repeated plaque-to-plaque passage of FMDV clones (Escarmís et al. 1996; see also the chapter by Escarmís et al., this volume). When populations dominated by either one of these mutants were serially passaged in cell culture, revertant viruses with the wild-type sequence (that is, without the amino acid substitution that led to antigenic variation and absence of the internal polyadenylate tract, respectively) became dominant. However, genomes with the initially dominant markers did remain in the mutant spectrum not at a basal level accounted solely by mutational pressure, but at levels 10- to 100fold higher. A scheme of implementation of memory is shown in Fig. 5 and the main features of quasispecies memory are summarized in Table 4. Two types of memory were distinguished during HIV-1 infections in vivo. One was the “replicative” memory as defined with FMDV, and the other was a “cellular” or “anatomical” memory derived from the retroviral life cycle with Table 4 Main features of quasispecies memory Replicative memory Memory genomes derive from genomes that were dominant at an early phase in the evolution of the same viral population. Memory genomes are present at higher frequencies (in the cases studied 10 to 100 times) than expected from mere mutational pressure exerted on parental genomes. Memory genomes are lost upon subjecting virus to a population bottleneck. Memory genomes increase in fitness in parallel with the dominant genomes, as replication proceeds (Red Queen hypothesis). Memory genomes may decrease in frequency as virus replicates. Cellular or anatomical memory Found when viral reservoirs (displaying no replication or limited replication) are activated and contribute to the composition of an evolving quasispecies. Based on Ruiz-Jarabo et al. 2000, 2002, 2003b; Briones et al. 2003; Arias et al. 2004.

70

E. Domingo et al.

Fig. 5 Schematic representation of implementation, maintenance and loss of memory genomes in a viral quasispecies. Genomes of the mutant spectra are depicted as horizontal lines, and mutations as symbols on the lines. The mutant spectra have been divided in three levels: level 1 includes those genomes found most frequently in the mutant distribution; level 2 includes a frequency class potentially occupied by memory genomes; and level 3 gathers the less abundant genomes that occur as a result of mutational pressure and the competitive rating of all components of the mutant spectrum. Initially genomes with the genetic marker A belong to level 3. When genomes with marker A are selected, they occupy the entire quasispecies. Upon further replication, when genomes with A display a selective disadvantage with respect to genomes lacking A, genomes with the A marker will eventually become undetectable in the consensus sequence of the quasispecies. However, genomes with marker A may remain as memory genomes in level 2 because in the course of selection their fitness increased (upper right mutant distribution). When bottleneck passages (of sufficient intensity to exclude level 2 genomes) intervene, memory is lost and genomes with A occupy again level 3 as in the initial mutant distribution (bottom right mutant distribution). In real viral quasispecies, mutants are ranked to form many frequency levels (perhaps a continuum of frequency levels) rather than 3, as assumed here for simplification. Experimental evidence and references on implementation, maintenance and decay of quasispecies memory in cell culture and in vivo are given in the text and in Table 4. (Based on Domingo 2000, with permission)

integration of proviral DNA in cellular DNA, which results in reservoirs that may re-emerge at a later stage of infection (Briones et al. 2003; Briones et al., personal communication). Reemergence (or raise to dominance) of ancestral

Viruses as Quasispecies: Biological Implications

71

(minority) genomes in vivo has been documented with several viral infections (Borrow et al. 1997; Wyatt et al. 1998; Karlsson et al. 1999; Briones et al. 2000; Lau et al. 2000; García-Lerma et al. 2001; Imamichi et al. 2001; Kijak et al. 2002; Charpentier et al. 2004), providing additional support for the concept of memory in viruses both in cell culture and in vivo. The experiments to investigate the possible presence of memory in viral quasispecies (Ruiz-Jarabo et al. 2000) were motivated by features of virus evolution that fit those of complex adaptive systems (see reviews in Gell-Mann 1994; Frank 1996; Solé and Goodwin 2000). A characteristic of such systems is that they vary, yet they have continuity in the form of some built-in (heritable) component, and are endowed with mechanisms to respond to external stimuli. There is a continuous interplay between the components of the system and the interacting environment. Examples of complex adaptive systems are neurological activity in memory and learning, and the immune system of vertebrates in which long-lived memory T cells may be expanded when the organism is again exposed to an antigen identical (or related) to the one that triggered the initial response. Such potential to act quickly in response to an environmental demand has a functional parallel in the presence of memory genomes in viral quasispecies. Indeed, memory in viral quasispecies may serve the adaptive scheme of viruses by providing a molecular reservoir capable of reacting swiftly to a selective constraint which has been previously experienced by the same population, provided no bottlenecks intervened (Table 4). All evidence suggests that immune or other internal physiological responses in hosts are neither uniform nor continuous, and that memory may help confronting discontinuous or fluctuating challenges that require dominance of related viral genome subsets. Minority genomes present at higher than basal levels, associated with mutational pressure, need not be related only to memory. They may occur as a consequence of the fitness values of randomly generated variants that, although not sufficient to drive the genome subset to dominance, may allow the subset to remain at levels similar to those associated with memory. All these possibilities provide an incentive to develop methods to diagnose the presence of minority genomes in viral populations in vivo, for example for a more personalized and adequate antiviral treatment plan (i.e. to avoid antiretroviral inhibitors for which mutations that contribute to resistance are present at memory levels) (reviewed in Domingo et al. 2003). 6.1 Memory and Long-Term Evolutionary History: A Distinction There are two concepts related to memory in viruses that should be distinguished. One is a form of long-term memory, not erased by population

72

E. Domingo et al.

bottlenecks, that has gradually shaped the genetic structure and replication strategy of viruses. A present-day virus can be viewed as a “reservoir of memories” of all influences that have conformed its present organization and biological potential, in constant co-evolution with hosts that themselves are an “outcome of history”. In contrast, the quasispecies memory discussed in previous paragraphs is an immediate, short-term consequence of the recent evolutionary history of a virus, independently of the historical events that gradually shaped its current configuration.

7 Virus Entry into Error Catastrophe: An Antiviral Strategy Derived from Quasispecies Dynamics Theoretical studies summarized in the chapters by Biebricher and Eigen and by Wilke et al., this volume, led to the very significant prediction that there must be a limit to the error rate that a replicating system can tolerate if the genetic information is to be maintained (Swetina and Schuster 1982; Eigen and Biebricher 1988; Nowak and Schuster 1989; Eigen 2002). This limit is termed the error threshold and its value depends on the complexity of the genetic information conveyed by the system. Here the term “complexity” means the information encoded by a genome (regulatory regions and number of different open-reading frames), which must be distinguished from the number of different sequences for any regulatory region or open-reading frame represented in a population of genomes; the latter is often referred to as complexity of the mutant spectrum or quasispecies complexity. For RNA viruses, which generally encode little or no redundant information, complexity (in its first meaning) can be equated with genome length. It may be expected that the most complex RNA viruses (such as the coronaviruses, with 27–32 kb genomes) may display higher average copying fidelities than the least complex RNA genomes. In this respect, it will be interesting to establish whether a domain found in the SARS and other coronavirus genomes, which corresponds to a nuclease activity, can provide a proofreading–repair function during coronavirus replication (Snijder et al. 2003). The interactions among components of mutant spectra (Sect. 5) also provide a biochemical basis for impairing viral replication to the point of leading to the information meltdown typical of error catastrophe. Interestingly, the proposed negative interactions (for example, among trans-acting products as discussed in Sect. 5 and depicted schematically in Fig. 4) also favour a more likely replicative collapse when the number of open-reading frames is high (higher complexity in the sense of number of genes encoding proteins). Un-

Viruses as Quasispecies: Biological Implications

73

der mutational pressure, more gene products can be deficient and contribute to replicative incompetence. This is a biochemical counterpart of the genetic basis of error catastrophe. Measurements of LCMV RNA and infectivity levels during enhanced mutagenesis of steady-state persistent infections of the virus in cell culture have defined two viral extinction pathways. One of them occurs at low mutagenic activities, and it is mediated by defective genomes which drive the infective genomes towards extinction; both experimental results and a computational model support the lethal defection model of virus extinction, with a prominent role of the mutant spectrum (Grande-Pérez et al. 2005). There is considerable experimental evidence that RNA viruses replicate with an error rate which is close to the error threshold for maintenance of genetic information. In several viral systems, it has been documented that an increase in mutation rate by added chemical mutagens results in decreases in infectivity, and in some cases it may lead to virus extinction (Holland et al. 1990; Lee et al. 1997; Loeb et al. 1999; Crotty et al. 2000, 2001; Loeb and Mullins 2000; Sierra et al. 2000; Lanford et al. 2001; Pariente et al. 2001; Contreras et al. 2002; Grande-Pérez et al. 2002; Airaksinen et al. 2003; Ruiz-Jarabo et al. 2003a; Severson et al. 2003; reviews in Eigen 2002; Anderson et al. 2004; Domingo 2005b). Two lines of study are particularly promising regarding a possible clinical application of error catastrophe. One concerns the mechanism of antiviral activity of the nucleoside analogue ribavirin (1-β-D-ribofuranosyl-1, 2, 3-triazole-3-carboxamide; Rib), a licensed drug extensively used in clinical practice, which shows antiviral activity against a number of DNA and RNA viruses (Snell 2001; see also Airaksinen et al. 2003 and references therein). Rib is phosphorylated by cellular enzymes to its mono-, di- and triphosphate forms and it exerts its antiviral activity through different mechanisms. One of the mechanisms unveiled by Crotty and colleagues, working with poliovirus, is enhanced mutagenesis as a result of incorporation of ribavirin monophosphate (Rib MP) into poliovirus RNA by the viral polymerase (Crotty et al. 2000, 2001; Graci and Cameron 2002). It has also been shown that Rib MP can be incorporated by the hepatitis C virus (HCV) and coxsackievirus B3 polymerase (Maag et al. 2001; Vo et al. 2003; Freistadt et al. 2004). This raises the intriguing possibility that benefits of Rib (alone but mainly in combination with interferon α or its derivatives) for treatment of chronic HCV infections (McHutchison et al. 1998, among many other studies) may be partly due to its mutagenic activity (see also the chapter by Pawtlosky, this volume, for alternative mechanisms of Rib action on HCV infection). If this were established (by many ongoing studies), it would mean that the principle of virus entry into error catastrophe as an antiviral activity would already have been successfully documented without experts being aware of the underlying mechanisms.

74

E. Domingo et al.

A second line of promising research was launched by de la Torre and his collaborators by showing that the base analogue 5-fluorouracil (FU) prevented the establishment of a persistent lymphocytic choriomeningitis infection in mice (Ruiz-Jarabo et al. 2003a). FU is mutagenic for a number of RNA viruses (Pringle 1970; Eastman and Blair 1985; Sierra et al. 2000 and references therein) and is used to treat some types of human cancer (Parker and Cheng 1990). The demonstration that it can be a potent antiviral agent in vivo constitutes a proof-of-principle that is extremely encouraging for the prospect of a clinical application of virus entry into error catastrophe (for recent reviews of error catastrophe as an antiviral strategy, see Anderson et al. 2004 and Domingo 2005b). This recent field of research illustrates how decisive is to cultivate basic research in its many facets to arrive at potential practical applications, often in unpredictable ways.

8 Concluding Remarks The studies summarized in this and other chapters of this book emphasize two growing concepts in the studies of infectious diseases that apply not only to viral infections, but also to bacterial, fungal and parasitic infections, as well as to cancer cells. One is the increasing experimental evidence in support of the remarkable statement by Theodosius Dobzhansky, written three decades ago, that “Nothing in biology makes sense except in the light of evolution” (Dobzhansky 1973). Dealing with evolutionary concepts is not rooted in the tradition of many schools of biochemistry and molecular biology. However, recent developments point to a decisive contribution of evolutionary events as part of the interactions between hosts and infectious agents. Justification of this statement includes the recognition of pathogen variation and adaptability in disease emergence and reemergence (Smolinski et al. 2003), as well as in pathogenesis (this volume and many references included in it). Evolutionary events underlie selection of viruses resistant to antiviral agents, and of bacteria resistant to antibiotics (for example, Welsh 2003). The list could go on for any pathogen whose replication cycles last orders of magnitude less than those of the hosts they infect (see the overview of molecular mechanisms of pathogen adaptation in Domingo 2005a). A concept that originated in physics and that is gradually penetrating many fields of science (including biology and infectious diseases) is complexity and emergence (Cowan et al. 1994; Solé and Goodwin 2000). What this concept entails is that certain phenomena whose occurrence depends on interactions among many individual influences cannot be accounted for by the sum of

Viruses as Quasispecies: Biological Implications

75

the individual contributions. Disease emergence offers an example: although we know (or strongly suspect with a reasonable basis) that human mobility, demographic changes and urbanization favour human-to-human pathogen transmission, that many agents are potentially zoonotic and that in many areas of the world there is close contact between humans and animals, it was not possible to predict the AIDS epidemic. At most, it can be predicted that because of (at least) thirteen different factors (Smolinski et al. 2003) (each one including, in turn, multiple sub-factors!) it is very likely that in the coming years new infectious diseases will emerge. Can studies on complexity be of use to infectious diseases? Probably, but the way such help can be translated into practical results is hardly graspable at the moment. This is not so, however, with regard to considering evolutionary concepts in designing antiviral strategies, since important successes (for example, highly active antiretroviral treatments to control HIV infections, or the prospects of lethal mutagenesis as an antiviral strategy) have been a consequence of understanding and trying to counter the evolutionary potential of pathogens. The need of a holistic view of biology with a focus on evolution, emergence and complexity has been emphasized by biologists working in areas other than infectious diseases, such as general evolution and developmental biology (Woese 2004). A point has been reached that encourages a transdisciplinary approach to problems of biological complexity, including viral quasispecies. The challenge is to obtain detailed molecular information provided by current biochemistry and molecular biology and to integrate it with the concepts of complexity and evolution, with the aim of providing a more complete picture of the problems we face. Acknowledgements Work was supported by grants BMC 2001–1823-C02–01, CAM (08.2/0046/2000, 08.2/0015/2001.1) GR/SAL/0172/2004 to Genetrix, S.L., EU, and Fundación Ramón Areces. AG-P was supported by a postdoctoral contract from CAM and a contract Ramón y Cajal from MCyT, JG-A was supported by a predoctoral fellowship from MEyC, and AA by a predoctoral fellowship from CAM and an I3P contract from CSIC.

References Agol VI (2002) Picornavirus genetics: an overview. In Semler BL, Wimmer E (eds) Molecular biology of picornaviruses. American Society for Microbiology, Washington DC, pp 269–284 Airaksinen A, Pariente N, Menendez-Arias L, Domingo E (2003) Curing of foot-andmouth disease virus from persistently infected cells by ribavirin involves enhanced mutagenesis. Virology 311:339–349

76

E. Domingo et al.

Anderson JP, Daifuku R, Loeb LA (2004) Viral error catastrophe by mutagenic nucleosides. Annu Rev Microbiol 58:183–205 Arias A, Agudo R, Ferrer-Orta C, Pérez-Luque R, Airaksinen A, Brocchi E, Domingo E, Verdaguer N, Escarmis C (2005) Mutant viral polymerase in the transition of virus to error catastrophe identifies a critical site for RNA binding. J Mol Biol (in press) Arias A, Lázaro E, Escarmís C, Domingo E (2001) Molecular intermediates of fitness gain of an RNA virus: characterization of a mutant spectrum by biological and molecular cloning. J Gen Virol 82:1049–1060 Arias A, Ruiz-Jarabo CM, Escarmis C, Domingo E (2004) Fitness increase of memory genomes in a viral quasispecies. J Mol Biol 339:405–412 Baranowski E, Ruíz-Jarabo CM, Pariente N, Verdaguer N, Domingo E (2003) Evolution of cell recognition by viruses: a source of biological novelty with medical implications. Adv Virus Res 62:19–111 Bergstrom CT, McElhany P, Real LA (1999) Transmission bottlenecks as determinants of virulence in rapidly evolving pathogens. Proc Natl Acad Sci U S A 96:5095–5100 Biebricher CK (1999) Mutation, competition and selection as measured with small RNA molecules. In Domingo E, Webster RG, Holland JJ (eds) Origin and evolution of viruses. Academic Press, San Diego, pp 65–85 Borrego B, Novella IS, Giralt E, Andreu D, Domingo E (1993) Distinct repertoire of antigenic variants of foot-and-mouth disease virus in the presence or absence of immune selection. J Virol 67:6071–6079 Borrow P, Lewicki H, Wei X, Horwitz MS, Peffer N, Meyers H, Nelson JA, Gairin JE, Hahn BH, Oldstone MB, Shaw GM (1997) Antiviral pressure exerted by HIV-1specific cytotoxic T lymphocytes (CTLs) during primary infection demonstrated by rapid selection of CTL escape virus. Nat Med 3:205–211 Briones C, Mas A, Gómez-Mariano G, Altisent C, Menéndez-Arias L, Soriano V, Domingo E (2000) Dynamics of dominance of a dipeptide insertion in reverse transcriptase of HIV-1 from patients subjected to prolonged therapy. Virus Res 66:13–26 Briones C, Domingo E, Molina-París C (2003) Memory in retroviral quasispecies: experimental evidence and theoretical model for human immunodeficiency virus. J Mol Biol 331:213–229 Charpentier C, Dwyer DE, Mammano F, Lecossier D, Clavel F, Hance AJ (2004) Role of minority populations of human immunodeficiency virus type 1 in the evolution of viral resistance to protease inhibitors. J Virol 78:4234–4247 Chumakov KM, Powers LB, Noonan KE, Roninson IB, Levenbook IS (1991) Correlation between amount of virus with altered nucleotide sequence and the monkey test for acceptability of oral poliovirus vaccine. Proc Natl Acad Sci U S A 88:199–203 Coffin JM (1995) HIV population dynamics in vivo: implications for genetic variation, pathogenesis, and therapy. Science 267:483–489 Condit C (2001) Principles of virology. In Knipe DM, Howley PM (eds) Fields virology. Lippincot Williams and Wilkins, Philadelphia, pp 811–815 Contreras AM, Hiasa Y, He W, Terella A, Schmidt EV, Chung RT (2002) Viral RNA mutations are region specific and increased by ribavirin in a full-length hepatitis C virus replication system. J Virol 76:8505–8517

Viruses as Quasispecies: Biological Implications

77

Cornell CT, Brunner JE, Semler BL (2004) Differential rescue of poliovirus RNA replication functions by genetically modified RNA polymerase precursors. J Virol 78:13007–13018 Cowan GA, Pines D, Meltzer D (eds) (1994) Complexity. Metaphors, models and reality. The Advanced Book Program, Jacob Way. Addison-Wesley Publishing, Reading, MA Crotty S, Maag D, Arnold JJ, Zhong W, Lau JYN, Hong Z, Andino R, Cameron CE (2000) The broad-spectrum antiviral ribonucleotide, ribavirin, is an RNA virus mutagen. Nat Med 6:1375–1379 De la Torre JC, Holland JJ (1990) RNA virus quasispecies populations can suppress vastly superior mutant progeny. J Virol 64:6278–6281 Crotty S, Cameron CE, Andino R (2001) RNA virus error catastrophe: direct molecular test by using ribavirin. Proc Natl Acad Sci U S A 98:6895–6900 DeFilippis VR, Villarreal LP (2000) An introduction to the evolutionary ecology of viruses. In Hurst CJ (ed) Viral ecology. Academic Press, San Diego, pp 125–208 Dobzhansky T (1973) Nothing in biology makes sense except in the light of evolution. Am Biol Teacher 35:125–129 Domingo E (1996) Biological significance of viral quasispecies. Viral Hepatitis Reviews 2:247–261 Domingo E (1997) RNA virus evolution, population dynamics, and nutritional status. Biol Trace Elem Res 56:23–30 Domingo E (1999) Quasispecies. In Granoff A, Webster RG (eds) Encyclopedia of virology. Academic Press, London, pp 1431–1436 Domingo E (2000) Viruses at the edge of adaptation. Virology 270:251–253 Domingo E (2003) Quasispecies and the development of new antiviral strategies. Prog Drug Res 60:133–158 Domingo E (2005a) Microbial evolution and emerging diseases. In Power C, Johnson RT (eds) Emerging neurological infections. Marcel Dekker, New York Domingo E (2005b) Virus entry into error catastrophe as a new antiviral strategy. 107:115–228 Domingo E, Holland JJ (1997) RNA virus mutations and fitness for survival. Annu Rev Microbiol 51:151–178 Domingo E, Holland JJ, Biebricher C, Eigen M (1995) Quasispecies: the concept and the word. In Gibbs A, Calisher C, García-Arenal F (eds) Molecular evolution of the viruses. Cambridge University Press, Cambridge, pp 171–180 Domingo E, Biebricher C, Eigen M, Holland JJ (2001) Quasispecies and RNA virus evolution: principles and consequences. Austin, Landes Bioscience Domingo E, Ruíz-Jarabo CM, Arias A, Molina-Paris C, Briones C, Baranowski E, Escarmís C (2003) Detection and biological implications of genetic memory in viral quasispecies. In Matsumori A (ed) Cardiomyopathies and heart failure: biomolecular, infectious and immune mechanisms. Kluwer, London, pp 259–276 Domingo E, Brun A, Núñez JI, Cristina J, Briones C, Escarmís C (2006) Genomics of viruses. In Hacker J, Dobrindt U (eds) Pathogenomics: Genome Analysis of Pathogenic Microbes. WILEY-VCH Verlag GmbH and Co.KGaA, Weinheim, 369–388 Drake JW, Holland JJ (1999) Mutation rates among RNA viruses. Proc Natl Acad Sci U S A 96:13910–13913

78

E. Domingo et al.

Eastman PS, Blair CD (1985) Temperature-sensitive mutants of Japanese encephalitis virus. J Virol 55:611–616 Eigen M (1971) Self-organization of matter and the evolution of biological macromolecules. Naturwissenschaften 58:465–523 Eigen M (1992) Steps towards life. Oxford University Press, Oxford Eigen M (2002) Error catastrophe and antiviral strategy. Proc Natl Acad Sci U S A 99:13374–13376 Eigen M, Schuster P (1979) The hypercycle. A principle of natural self-organization. Springer, Berlin Heidelberg New York Eigen M, Biebricher CK (1988) Sequence space and quasispecies distribution. In Domingo E, Ahlquist P Holland JJ (eds) RNA genetics. CRC Press, Boca Raton, FL, pp 211–245 Escarmís C, Dávila M, Charpentier N, Bracho A, Moya A, Domingo E (1996) Genetic lesions associated with Muller’s ratchet in an RNA virus. J Mol Biol 264:255–267 Flint SJ, Enquist LW, Racaniello VR, Skalka AM (2004) Principles of virology. Molecular biology, pathogenesis, and control of animal viruses. ASM Press, Washington, DC Frank SA (1996) The design of natural and artificial adaptive systems. In Rose MR, Lauder GV (eds) Adaptation. Academic, San Diego, pp 451–505 Freistadt MS, Meades GD, Cameron CE (2004) Lethal mutagens: broad-spectrum antivirals with limited potential for development of resistance? Drug Resist Updat 7:19–24 García-Lerma JG, Nidtha S, Blumoff K, Weinstock H, Heneine W (2001) Increased ability for selection of zidovudine resistance in a distinct class of wild-type HIV-1 from drug-naive persons. Proc Natl Acad Sci U S A 98:13907–13912 García-Arriaza J, Manrubia SC, Toja M, Domingo E, Escarmís C (2004) Evolutionary transition toward defective RNAs that are infectious by complementation. J Virol 78:11678–11685 García-Arriaza J, Domingo E, Escarmís C (2005) A segmented form of foot-andmouth disease virus interferes with standard virus: a link between interference and competitive fitness. Virology 335:155–164 Gell-Mann M (1994) Complex adaptive systems. In Cowan GA, Pines D Meltzer D (eds) Complexity metaphors, models and reality. Wesley, Reading, MA, pp 17–45 Gerrish PJ, Garcia-Lerma JG (2003) Mutation rate and the efficacy of antimicrobial drug treatment. The Lancet Infect Dis 3:28–32 González-López C, Gómez-Mariano G, Escarmís C, Domingo E (2005) Invariant aphthovirus consensus nucleotide sequence in the transition to error catastrophe. Infect Genet Evol 5:366–374 González-López C, Arias A, Pariente N, Gómez-Mariano G, Domingo E (2004) Preextinction viral RNA can interfere with infectivity. J Virol 78:3319–3324 Graci JD, Cameron CE (2002) Quasispecies, error catastrophe, and the antiviral activity of ribavirin. Virology 298:175–180 Grande-Pérez A, Sierra S, Castro MG, Domingo E, Lowenstein PR (2002) Molecular indetermination in the transition to error catastrophe: systematic elimination of lymphocytic choriomeningitis virus through mutagenesis does not correlate linearly with large increases in mutant spectrum complexity. Proc Natl Acad Sci U S A 99:12938–12943

Viruses as Quasispecies: Biological Implications

79

Grande-Pérez A, Lázaro E, Lowenstein P, Domingo E, Manrubia SC (2005) Suppression of viral infectivity through lethal defection. Proc Natl Acad Sci U S A 102:4448– 4452 Grande-Pérez A, Gómez-Mariano G, Lowenstein P, Domingo E (2005) Mutagenesisinduced, large fitness variations with an invariant arenavirus consensus genomic nucleotide sequence. J Virol 79:10451–10459 Holland JJ (1990) Defective viral genomes. In Fields BM, Knipe DM (eds) Virology. Raven, New York, pp 151–165 Holland JJ, Domingo E, de la Torre JC, Steinhauer DA (1990) Mutation frequencies at defined single codon sites in vesicular stomatitis virus and poliovirus can be increased only slightly by chemical mutagenesis. J Virol 64:3960–3962 Holland JJ, de la Torre JC, Clarke DK, Duarte E (1991) Quantitation of relative fitness and great adaptability of clonal populations of RNA viruses. J Virol 65:2960–2967 Huang AS, Baltimore D (1970) Defective viral particles and viral disease processes. Nature 226:325–327 Imamichi H, Crandall KA, Natarajan V, Jiang MK, Dewar RL, Berg S, Gaddam A, Bosche M, Metcalf JA, Davey RT Jr, Lane HC (2001) Human immunodeficiency virus type 1 quasi species that rebound after discontinuation of highly active antiretroviral therapy are similar to the viral quasi species present before initiation of therapy. J Infect Dis 183:36–50 Karlsson AC, Gaines H, Sallberg M, Lindback S, Sonnerborg A (1999) Reappearance of founder virus sequence in human immunodeficiency virus type 1-infected patients. J Virol 73:6191–6196 Kijak GH, Simon V, Balfe P, Vanderhoeven J, Pampuro SE, Zala C, Ochoa C, Cahn P, Markowitz M, Salomon H (2002) Origin of human immunodeficiency virus type 1 quasispecies emerging after antiretroviral treatment interruption in patients with therapeutic failure. J Virol 76:7000–7009 Lanford RE, Chavez D, Guerra B, Lau JY, Hong Z, Brasky KM, Beames B (2001) Ribavirin induces error-prone replication of GB virus B in primary tamarin hepatocytes. J Virol 75:8074–8081 Lau DT, Khokhar MF, Doo E, Ghany MG, Herion D, Park Y, Kleiner DE, Schmid P, Condreay LD, Gauthier J, Kuhns MC, Liang TJ, Hoofnagle JH (2000) Long-term therapy of chronic hepatitis B with lamivudine. Hepatology 32:828–834 Lee CH, Gilbertson DL, Novella IS, Huerta R, Domingo E, Holland JJ (1997) Negative effects of chemical mutagenesis on the adaptive behavior of vesicular stomatitis virus. J Virol 71:3636–3640 Lewontin RC (1970) The units of selection. Annu Rev Ecol System 1:1–18 Loeb LA, Mullins JI (2000) Lethal mutagenesis of HIV by mutagenic ribonucleoside analogs. AIDS Res Hum Retroviruses 13:1–3 Loeb LA, Essigmann JM, Kazazi F, Zhang J, Rose KD, Mullins JI (1999) Lethal mutagenesis of HIV with mutagenic nucleoside analogs. Proc Natl Acad Sci U S A 96:1492–1497 Maag D, Castro C, Hong Z, Cameron CE (2001) Hepatitis C virus RNA-dependent RNA polymerase (NS5B) as a mediator of the antiviral activity of ribavirin. J Biol Chem 276:46094–46098 Mahy BWJ (ed) (2005) Foot-and-mouth disease virus. Curr Topics Microbiol Immunol 288

80

E. Domingo et al.

McHutchison JG, Gordon SC, Schiff ER, Shiffman ML, Lee WM, Rustgi VK, Goodman ZD, Ling MH, Cort S, Albrecht JK (1998) Interferon alfa-2b alone or in combination with ribavirin as initial treatment for chronic hepatitis C. Hepatitis Interventional Therapy Group. N Engl J Med 339:1485–1492 Miralles R, Gerrish PJ, Moya A, Elena SF (1999) Clonal interference and the evolution of RNA viruses. Science 285:1745–1747 Moreno IM, Malpica JM, Rodriguez-Cerezo E, Garcia-Arenal F (1997) A mutation in tomato aspermy cucumovirus that abolishes cell-to-cell movement is maintained to high levels in the viral RNA population by complementation. J Virol 71:9157– 9162 Mount DW (2004) Bioinformatics. sequence and genome analysis. Cold Spring Harbor Laboratory Press, Cold Spring Harbor Nájera I, Holguín A, Quiñones-Mateu ME, Muñoz-Fernández MA, Nájera R, LópezGalíndez C, Domingo E (1995) Pol gene quasispecies of human immunodeficiency virus: mutations associated with drug resistance in virus from patients undergoing no drug therapy. J Virol 69:23–31 Novella IS (2003) Contributions of vesicular stomatitis virus to the understanding of RNA virus evolution. Curr Opin Microbiol 6:399–405 Novella IS (2004) Negative effect of genetic bottlenecks on the adaptability of vesicular stomatitis virus. J Mol Biol 336:61–67 Novella IS, Elena SF, Moya A, Domingo E, Holland JJ (1995) Size of genetic bottlenecks leading to virus fitness loss is determined by mean initial population fitness. J Virol 69:2869–2872 Novella IS, Quer J, Domingo E, Holland JJ (1999) Exponential fitness gains of RNA virus populations are limited by bottleneck effects. J Virol 73:1668–1671 Novella IS, Reissig DD, Wilke CO (2004) Density-dependent selection in vesicular stomatitis virus. J Virol 78:5799–5804 Nowak M, Schuster P (1989) Error thresholds of replication in finite populations mutation frequencies and the onset of Muller’s ratchet. J Theor Biol 137:375–395 Nowak MA, May RM (2000) Virus dynamics. Mathematical principles of immunology and virology. Oxford University Press, New York Oldstone MBA (2002) Biology and pathogenesis of lymphocytic choriomeningitis virus infection. Curr Top Microbiol Immunol 263:83–117 Pariente N, Sierra S, Lowenstein PR, Domingo E (2001) Efficient virus extinction by combinations of a mutagen and antiviral inhibitors. J Virol 75:9723–9730 Parker WB, Cheng YC (1990) Metabolism and mechanism of action of 5-fluorouracil. Pharmacol Ther 48:381–395 Perales C, Martin V, Ruiz-Jarabo CM, Domingo E (2005) Monitoring sequence space as a test for the target of selection in viruses. J Mol Biol 345:451–459 Perrault J (1981) Origin and replication of defective interfering particles. Curr Top Microbiol Immunol 93:151–207 Pringle CR (1970) Genetic characteristics of conditional lethal mutants of vesicular stomatitis virus induced by 5-fluorouracil, 5-azacytidine, and ethyl methane sulfonate. J Virol 5:559–567

Viruses as Quasispecies: Biological Implications

81

Quiñones-Mateu ME, Arts EJ (2002) HIV-1 fitness: implications for drug resistance, disease progression, and global epidemic evolution. In Theoretical Biology and Biophysics Group. Kuiken C, Foley B, Hahn BH, Marx P, McCutchan FE, Mellors J, Mullins JI, Sodroski J, Wolinsky S Korber BT (eds) HIV-1 sequence compendium. Los Alamos National Laboratory, Los Alamos NM, pp 134–170 Ribeiro RM, Bonhoeffer S, Nowak MA (1998) The frequency of resistant mutant virus before antiviral therapy. AIDS 26:461–465 Rowlands DJ (ed) (2003) Foot-and-mouth disease. Virus Res 91:1–161 Ruiz-Jarabo CM, Arias A, Baranowski E, Escarmís C, Domingo E (2000) Memory in viral quasispecies. J Virol 74:3543–3547 Ruiz-Jarabo CM, Arias A, Molina-París C, Briones C, Baranowski E, Escarmís C, Domingo E (2002) Duration and fitness dependence of quasispecies memory. J Mol Biol 315:285–296 Ruiz-Jarabo CM, Ly C, Domingo E, de la Torre JC (2003a) Lethal mutagenesis of the prototypic arenavirus lymphocytic choriomeningitis virus (LCMV). Virology 308:37–47 Ruiz-Jarabo CM, Miller E, Gómez-Mariano G, Domingo E (2003b) Synchronous loss of quasispecies memory in parallel viral lineages: a deterministic feature of viral quasispecies. J Mol Biol 333:553–563 Severson WE, Schmaljohn CS, Javadian A, Jonsson CB (2003) Ribavirin causes error catastrophe during Hantaan virus replication. J Virol 77:481–488 Sierra S, Dávila M, Lowenstein PR, Domingo E (2000) Response of foot-and-mouth disease virus to increased mutagenesis. Influence of viral load and fitness in loss of infectivity. J Virol 74:8316–8323 Simmonds P, Smith DB (1999) Structural constraints on RNA virus evolution. J Virol 73:5787–5794 Simmonds P, Tuplin A, Evans DJ (2004) Detection of genome-scale ordered RNA structure (GORS) in genomes of positive-stranded RNA viruses: implications for virus evolution and host persistence. RNA 10:1337–1351 Smolinski MS, Hamburg MA, Lederberg J (eds) (2003) Microbial threats to health. Emergence, detection and response. The National Academies Press, Washington, DC Snell NJ (2001) Ribavirin—current status of a broad spectrum antiviral agent. Expert Opin Pharmacother 2:1317–1324 Snijder EJ, Bredenbeek PJ, Dobbe JC, Thiel V, Ziebuhr J, Poon LL, Guan Y, Rozanov M, Spaan WJ, Gorbalenya AE (2003) Unique and conserved features of genome and proteome of SARS-coronavirus, an early split-off from the coronavirus group 2 lineage. J Mol Biol 331:991–1004 Sobrino F, Domingo E (eds) (2004) Foot-and-mouth disease: current perspectives. Horizon Bioscience. Wymondham, UK Solé R, Goodwin B (2000) Signs of life. how complexity pervedes biology. Basic Books, New York Swetina J, Schuster P (1982) Self-replication with errors. A model for polynucleotide replication. Biophys Chem 16:329–345 Teng MN, Oldstone MB, de la Torre JC (1996) Suppression of lymphocytic choriomeningitis virus-induced growth hormone deficiency syndrome by diseasenegative virus variants. Virology 223:113–119

82

E. Domingo et al.

Turner PE, Chao L (1998) Sex and the evolution of intrahost competition in RNA virus phi6. Genetics 150:523–532 Vo NV, Young KC, Lai MMC (2003) Mutagenic and inhibitory effects of ribavirin on hepatitis C virus RNA polymerase. Biochemistry 42:10462–10471 Volkenstein MV (1994) Physical approaches to biological evolution. Springer-Verlag, Berlin Heidelberg New York Welsh C (2003) Antibiotics. Actions, origins, resistance. ASM Press, Washington, DC White DO, Fenner FJ (1986) Medical virology. Academic, Orlando, FL Wilke CO, Reissig DD, Novella IS (2004) Replication at periodically changing multiplicity of infection promotes stable coexistence of competing viral populations. Evolution Int J Org Evolution 58:900–905 Williams GC (1992) Natural selection. Domains, levels and challenges. Oxford University Press, New York Woese CR (2004) A new biology for a new century. Microbiol Mol Biol Reviews 68:173– 186 Woolhouse MEJ (2004) Mathematical models of the epidemiology and control of foot-and-mouth disease. In Sobrino F, Domingo E (eds) Foot and mouth disease current perspectives. Horizon Bioscience, Wymondham UK, pp 355–381 Wyatt CA, Andrus L, Brotman B, Huang F, Lee DH, Prince AM (1998) Immunity in chimpanzees chronically infected with hepatitis C virus: role of minor quasispecies in reinfection. J Virol 72:1725–1730 Youngner JS, Whitaker-Dowling P (1999) Interference. In Granoff A, Webster RG (eds) Encyclopedia of virology. Academic Press, San Diego, pp 850–854

CTMI (2006) 299:83–140 c Springer-Verlag Berlin Heidelberg 2006


Virus Fitness: Concept, Quantification, and Application to HIV Population Dynamics M. E. Quiñones-Mateu1 (u) · E. J. Arts2 1 Department of Molecular Genetics, Section Virology, Lerner Research Institute,

Cleveland Clinic Foundation, 9500 Euclid Avenue/NN10, Cleveland, OH 44195, USA [email protected] 2 Division of Infectious Diseases, Department of Medicine, Case Western Reserve University, 10900 Euclid Avenue, Cleveland, OH 44106, USA

1 1.1 1.2

Concepts in Viral Fitness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 Lessons from Other RNA Viruses and Application to the HIV Model . . . . 85

2 2.1 2.1.1 2.1.2 2.1.3

Quantification of Viral Fitness . . . . . . . Methods Used to Quantify HIV Fitness . Viral Growth Kinetics . . . . . . . . . . . . . Single-Cycle Infections . . . . . . . . . . . . Growth Competition Experiments . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

87 89 90 90 91

3 3.1 3.1.1 3.1.2 3.1.3 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9

Application to HIV Population Dynamics . . . . . . . . . . . Replicative Fitness of Drug-Resistant Viruses . . . . . . . . Viruses Resistant to Protease Inhibitors . . . . . . . . . . . . Viruses Resistant to Reverse Transcriptase Inhibitors . . . Viruses Resistant to Other Antiretroviral Inhibitors . . . . Clinical Implications of HIV Fitness . . . . . . . . . . . . . . . Fitness of the Transmitted HIV Isolate . . . . . . . . . . . . . HIV Fitness During Disease Progression . . . . . . . . . . . . HIV Fitness and Immune Response . . . . . . . . . . . . . . . HIV Diversity and Fitness . . . . . . . . . . . . . . . . . . . . . . A Model for HIV Fitness During Disease . . . . . . . . . . . . Are There Intersubtype HIV Variations in Viral Fitness? . Is the Fitness and Pathogenicity of HIV Decreasing? . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

92 94 95 96 97 98 99 103 106 109 113 116 118

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

Abstract Viral fitness has been broadly studied during the past three decades, mainly to test evolutionary models and population theories difficult to analyze and interpret with more complex organisms. More recent studies, however, are focused in the role of fitness on viral transmission, pathogenesis, and drug resistance. Here, we used human immunodeficiency virus (HIV) as one of the most relevant models to evaluate the importance of viral quasispecies and fitness in HIV evolution, population dynamics, disease progression, and potential clinical implications.

84

M. E. Quiñones-Mateu · E. J. Arts

1 Concepts in Viral Fitness Fitness of RNA viruses has been a topic of study for several decades. As outlined below, sentinel studies on the fitness of foot-and-mouth disease virus (FMDV), vesicular stomatitis virus (VSV), and bacteriophages have addressed and tested basic evolutionary and fitness theories such as Red Queen Dynamics (Clarke et al. 1994; Novella et al. 1995a; Domingo et al. 1996; Sole et al. 1999), Muller’s ratchet (Chao 1990; Duarte et al. 1994; Escarmis et al. 1998; Domingo et al. 1996; Bergstrom et al. 1999), the competitive exclusion principle (Clarke et al. 1994; Domingo et al. 1996; Sole et al. 1999; Anderson and May 1996), Fisher’s geometric model (Burch and Chao 1999), Wright’s fitness landscape (Burch and Chao 1999), and even evolutionary game theory (Turner and Chao 1999). The preference in utilizing RNA viruses, obligate, nonliving parasites as compare to complex or even other simple organisms is the short generation time and high mutation frequencies, which can be controlled by defining the host environment. In contrast to these detailed fitness studies testing evolutionary models, research on the impact of viral fitness on the etiological disease and pathogenesis is limited to a handful of studies. Even these virus studies can be segregated into those focused on fitness during vertical transmission between individual multicellular hosts (e.g., in the human population) and fitness within a host spreading through susceptible cells, which may be analogous to horizontal transmission (as defined by evolutionary theory) and spread. Here, we will use one of the most studied RNA viruses in the last 20 years, human immunodeficiency virus (HIV), as a model to illustrate the complex relationship between viral quasispecies and fitness. 1.1 Definition Fitness is a parameter defining the replicative adaptation of an organism to its environment (reviewed in Domingo and Holland 1997; Domingo et al. 1999). Survival of the fittest is the concept that drives evolution in a complex population. Within a given viral quasispecies, each clone has a fitness value, representative of those viral properties (e.g., activity and stability) undergoing selection in that particular environment. During viral replication within a defined microenvironment, different genomes encode virus that replicate at high rates, continually mutate, but generally remain under the same selective pressures (Domingo et al. 1999). Positive (Darwinian) selection implies that one or more members of the quasispecies are better suited to a given environment, whereas negative selection eliminates unfit variants (Domingo et al.

Virus Fitness: Concept, Quantification, and Application

85

1996, 1999; Domingo and Holland 1997). In the case of HIV, each individual member of the quasispecies has an intrinsic growth rate, known as replicative fitness. This concept can be extended to the swarm of variants that constitute HIV isolates. For example, a broad range of replication capacities within a population of viruses with no phenotypic or genotypic evidence of drug resistance (wild-type viruses) have been reported (Wrin et al. 2001; Barbour et al. 2004). Therefore, the term “viral fitness” has emerged to describe the fact that, under defined environmental conditions, different HIV isolates may have different replicative fitness. Selective pressure in the form of drug therapy often leads to dramatic shifts in the quasispecies distribution, as a virus that was poorly fit in the presence of drug can rapidly emerge as the most fit in the presence of drug (Coffin 1995; Domingo and Holland 1997; Domingo et al. 1997, 1999). During this in vivo selection, several drug-resistant variants may emerge and compete for dominance. These resistant isolates will pass through the drug-induced bottleneck and initiate a new quasispecies distribution that will again be governed by replication efficiency, now in the presence of drugs (Coffin 1995; Loveday and Hill 1995). The factors determining which virus population eventually wins are not well defined. However, both host factors and chance likely contribute to the tremendous variability in evolutionary pathways that HIV may take when confronted with drug pressure. 1.2 Lessons from Other RNA Viruses and Application to the HIV Model Much can be learned from the evolution of other RNA viruses and applied to HIV. In terms of basic evolutionary theory derived from and tested with RNA viruses, HIV follows many of the basic models, as described below and in other chapters of this book. Unfortunately, much of the ground-breaking theoretical work on transmission mechanisms/rates and subsequent disease progression caused by pathogens (Cooper et al. 2002; Anderson and May 1986; Lenski and May 1994; Boots and Sasaki 1999) have been supported by few empirical studies (Weaver et al. 1999; Novella et al. 1995c; Messenger et al. 1999; Elena 2001; Bergstrom et al. 1999). In terms of the pathogenesis as influenced by viral fitness, HIV is quite different than most RNA viruses in that acute infection leads to chronic disease and in all cases, eventual mortality. Although mortality is common during acute disease of many RNA viruses (Knipe and Howley 2001), the only other RNA viruses that cause significant mortality due to prolonged chronic disease may be hepatitis B and C viruses (Knipe and Howley 2001). Even in the case of hepatitis B and C infection, progressive disease is generally associated with cirrhosis or hepatocellular carcinoma, which are only indirectly affected by virus destruction and loads (Knipe

86

M. E. Quiñones-Mateu · E. J. Arts

and Howley 2001). Even simian immunodeficiency virus (SIV) infection of various primates (Hirsch 2004), infections by genetically engineered SIV–HIV hybrids (Etemad-Moghadam et al. 2002; Joag 2000), or lentiviral infections of non-primates (Leroux et al. 2004) are relatively poor animal models for HIV pathogenesis in humans, let alone HIV evolution and fitness during disease. As discussed below, the models of HIV evolution and fitness were derived in part from experimental models and evolutionary theories derived from various parasites. As summarized previously (Elena 2001), these models and theories include the study of genetic drift on the accumulation of deleterious mutations (Chao 1990; Duarte et al. 1992), the fixation of beneficial mutations (Burch and Chao 1999; Miralles et al. 1999a), the extent of competition among coexisting genotypes (Clarke et al. 1994), the role of population patchiness and gene flux (Miralles et al. 1999b), the units of selection (Miralles et al. 1997), the existence of frequency-dependent selection mechanisms (Miralles et al. 2000), the different effect of vertical and horizontal transmissions (Messenger et al. 1999; Elena et al. 2001; Bergstrom et al. 1999), the cooperation between viruses (Turner and Chao 1999), or the cost of host radiation (Turner and Elena 2000). Virulence is typically defined as the rate in host mortality as a consequence of infection (Bull 1994), which can be further refined to reproduction rate and pathogenic potential of the parasite (Bremermann and Pickering 1983). In contrast, a parasite’s fitness is dependent on its survival and adaptability in a given environment. Hence, there is often confusion between the principles of virulence and fitness when applied to the interaction and survival of both parasite and host. Ewald, Anderson, May and others suggest that a pathogen will reduce its reproductive success/rates and virulence to ensure survival when host population size, density, and reproduction rates are limited (Ewald 1994; Anderson and May 1992). In terms of a pathogen spreading through a host population, higher fitness of a parasite is often defined by a reduction in reproduction and virulence. In contrast, the fitness of virus within a host is not always analogous to other pathogens. Viruses are obligate parasites that require a living cell for reproduction and survival. Thus, higher fitness within a host is dependent on mechanisms that enhance spread between susceptible host cells such as improved replication efficiency and increased transmission efficiency. For many viral infections, there are large susceptible cell populations, which are at high densities and quickly replenished. However, fitness within a host in contrast to fitness in the host population is subjected to different bottlenecks in the form of innate antiviral and acquired host immune responses. Survival and increasing fitness is dependent on immune avoidance (Goulder and Watkins 2004). There is often a balance between the selection of viral genotypes “resistant” with typically lower replication efficiencies and simply increased replication rates that can offset the viral clearance by host

Virus Fitness: Concept, Quantification, and Application

87

responses (Troyer and Arts, in press). However, this delicate balance between a host and viral pathogen resulting in prolonged but pathogenic infection is almost a unique attribute of HIV. Most viruses rapidly kill their hosts (e.g., Ebola and Hanta viruses; Mahanty and Bray 2004), are eliminated by an effective immune responses (e.g., influenza; Knipe and Howley 2001), lie dormant or hide in compartments relatively inaccessible to host antiviral responses (e.g., various Herpesviruses; Knipe and Howley 2001), or even establish an innocuous, sometimes symbiotic relationship with its host (e.g., SIV and GBV; Hirsch 2004; Stapleton 2003). Thus, no virus–host system appears to be a good model for HIV disease, but different stages in the HIV disease may parallel that of other host pathogens. Even in HIV research, a topic that is often overlooked is the impact of fitness on HIV transmission, disease progression, evolution, and prevalence in the human population. This review will outline the effect of drug resistance on HIV fitness, which has been the focus of several hundred studies. However, the vast majority of HIV infections worldwide are not being treated with conventional antiretroviral therapies. The rate of new infections continues to increase, people progress to AIDS and die, and yet little is still known about the phenotypic differences between the heterogeneous etiological agent. Recent studies suggest that the nature of the virus itself, and not solely manifestations of host factors and the immune response is contributing to HIV disease progression (Quiñones-Mateu et al. 2000). Numerous studies have also shown that there are possible differences in interpatient and intersubtype HIV fitness (Quiñones-Mateu et al. 2000; Kanki et al. 1999; Soto-Ramirez et al. 1996; Ball et al. 2003). However, these fitness studies are plagued by the same difficulties and shortcomings found in research on the fitness of drug-resistant HIV strains. What is an appropriate assay to measure ex vivo fitness? Are primary isolates or pseudotyped HIV clones most suitable for accessing differences in fitness? How do we compare ex vivo HIV replicative capacity to viral fitness within a patient or even within the human population? Although this field remains controversial, recent studies aided by new technologies are now making progress in answering some of the questions related to HIV fitness and AIDS (Ball et al. 2003; Quiñones-Mateu et al. 2000; Kanki et al. 1999; Soto-Ramirez et al. 1996).

2 Quantification of Viral Fitness How do we measure, in vitro, a parameter that by definition is determined by the interaction between viral quasispecies and a particular environment that

88

M. E. Quiñones-Mateu · E. J. Arts

changes many times? This very same question has plagued many virologists for the last 30 years. Many different approaches have been used to address this problem, each one of them with their intrinsic advantages and limitations. Mono- or single-infection assays used to quantify viral growth in vitro are among the simplest methods to estimate viral replication capacity. However, growth competition experiments designed to measure the capacity of two viruses to produce infectious progeny are the current gold standard to estimate relative fitness values on a predetermined environment (Holland et al. 1991; Domingo et al. 1999). This method allows the analysis of the contribution of every aspect of the viral life cycle (from entry into the host cell to the release of a new virus particle) to overall viral fitness. Furthermore, as described below, viral competitions are able to discern and quantify small, but highly significant, differences on fitness values. Understandably, some studies have tried to simplify the evaluation of viral fitness, severely limiting the analysis to single events within the viral cycle. Although these studies have greatly contributed to our understanding of the role of individual events of the replication cycle to overall viral fitness (reviewed in Quiñones-Mateu and Arts 2001), the capacity of an RNA virus to successfully replicate on a given setting may well involve an entanglement of events which, as a whole, define the fitness of the virus in a particular set of environmental conditions. As described to this point, viral fitness determinations using in vitro growth competition experiments have been based on the quantification of virus production after multiple rounds of replication (reviewed in Domingo and Holland 1997; Domingo et al. 1999), which guarantee the inclusion of every single phase of the viral cycle on each fitness measurement (similar to what should be observed in vivo). Recently, this well-established methodology has been challenged using individual viral clones to estimate their fitness based on parameters such as intracellular replication rate, cellular viral yield, time spent by a viral particle on a single infection, and the rate at which viruses diffuse in the medium (Cuevas et al. 2005). This interesting exercise resembles many studies conducted using HIV as a model where viral entry, RNA reverse transcription into DNA, viral integration and maturation have all been individually analyzed and correlated to overall HIV fitness (reviewed in Quiñones-Mateu and Arts 2001). However, as in the case with HIV, fitness measurements of single-steps in the viral life cycle hardly mimic viral fitness as a whole. Cuevas et al. (2005) suggested that fitness correlates with the intrinsic growth rate but not with viral yield. To our knowledge, both parameters are unavoidably linked. Furthermore, although this method could be useful to analyze fitness values of individual members within a viral quasispecies, the correlation between viral fitness and viral yield in actual in vivo situations is evident; that is, viral quasispecies better adapted to growth in

Virus Fitness: Concept, Quantification, and Application

89

a given host (more fit) will produce a higher number of viruses, which at the end is one of the most important parameters associated with viral evolution and pathogenesis (Quiñones-Mateu et al. 2000). 2.1 Methods Used to Quantify HIV Fitness Viral fitness is closely linked to a particular environment (e.g., a human host offers a variety of cell types and microenvironments to the infecting HIV, resulting in various selective pressures). Thus, it is logical to think that the optimal method to assess viral fitness is in vivo, by examining HIV kinetics in plasma (Goudsmit et al. 1996). However, host-to-host comparisons of in vivo HIV fitness are very difficult to evaluate because of differences in host genetics and immune response. Therefore, in vivo HIV fitness studies are limited to the emergence of specific quasispecies or drug-resistant mutants, and cannot determine the impact of specific substitutions on replicative fitness. In contrast, ex vivo fitness assays of primary HIV isolates are independent of host-specific variables and focus on replication efficiency in standardized systems. Multiple methods have been employed to measure HIV replicative fitness in vitro, including 1. The catalytic activity of HIV enzymes 2. Viral production in monoinfected cultures 3. Infectious virions:virus particle ratios 4. Single-cycle infection assays 5. Growth competition experiments (reviewed in Clavel et al. 2000; Nijhuis et al. 2001; Quiñones-Mateu and Arts 2001, 2002) However, there is also little continuity between in vitro studies, creating specific ambiguities in our understanding of HIV fitness. Many different factors will affect (and thus must be controlled) HIV fitness measurements. Figure 1summarizes the steps and variables that need to be taken into consideration for the analyses of HIV fitness. Most in vitro fitness studies are performed with virus pseudotyped with HIV genes (e.g., PR, RT, ENV, GagPol protease cleavage sites), often PCR-amplified directly from patient plasma or cells (Kellam and Larder 1994; Shi and Mellors 1997; Hertogs et al. 1998; Martinez-Picado et al. 1999; Robinson et al. 2000; Bleiber et al. 2001). One of the main advantages of these recombinant viruses is the ease and flexibility of directly comparing replicative fitness and drug sensitivity of various target genes within a neutral HIV backbone (Mammano et al. 1998; Croteau et al.

90

M. E. Quiñones-Mateu · E. J. Arts

Fig. 1 Steps in designing an HIV fitness assay. A compilation of the most important factors to be considered in the development of methods to quantify HIV fitness. EIA, enzyme immunoassay; RT, reverse transcriptase; PCR, polymerase chain reaction; MOI, multiplicity of infection; TCID50, tissue culture infectious doses

1997; Sharma and Crumpacker 1997, 1999; Martinez-Picado et al. 1999, 2000; Imamichi et al. 2000). However, there is always the possibility that by looking at the “snapshot” (recombinant viruses) we may be missing the overall replicative fitness given by complete HIV isolates. Here, we briefly describe three of the most accepted methods to determine HIV replicative fitness in vitro. 2.1.1 Viral Growth Kinetics This approach utilizes biochemical markers, such as p24 antigen concentration or reverse transcriptase activity, to measure viral growth in monoinfections (Borman et al. 1996; Doyon et al. 1996; Sharma and Crumpacker 1997; Mammano et al. 1998; Deeks et al. 2001). Differences in the replication kinetics of HIV mutants can be compared in parallel infections. Such assays, however, can only discern gross changes in replicative fitness and cannot accurately define the impact of subtle genetic changes on the replication rates of HIV isolates. In general, direct competition between two different viruses is a more accurate and sensitive assay to detect minute fitness differences (Holland et al. 1991; Quiñones-Mateu et al. 2000; Weber et al. 2003b). 2.1.2 Single-Cycle Infections Recombinant viruses are generated using patient-derived HIV sequences (generally the protease, reverse transcriptase, and Gag-Pol cleavage sites),

Virus Fitness: Concept, Quantification, and Application

91

which are cloned into a replication-deficient HIV plasmid vector, usually HIVNL4–3 or HIVHXB2 . This technique is useful for assessing the effect of specific drug-resistance mutations on viral replicative fitness, since it eliminates the confounding effects of mutations or polymorphisms located in other genomic regions. These vectors are also engineered to contain an indicator gene such as luciferase or β-galactosidase to allow quantification of viral replication. Activity of the reporter gene is determined after a single round of viral replication, following co-transfection with the test or control vector and a murine leukemia virus envelope expression vector. This method of measuring viral replication has been mastered by ViroLogic, Inc. (Petropoulos et al. 2000; Wrin et al. 2001; Deeks et al. 2001) adapting the PhenoSense assay used in drug-susceptibility assays. Although results obtained with this system have been shown to correlate well with in vivo measures (Weber et al. 2003b; Prado et al. 2002), it is also limited by the use of recombinant viruses and assessment of only specific steps in the replication cycle. Furthermore, this single-cycle infection assay cannot be used effectively in growth competition experiments. 2.1.3 Growth Competition Experiments Head-on competitions in cell culture between two viral isolates of the same species provides the internal control lacking in monoinfections. Growth competition experiments involve co-infection of a cell culture by two different HIV isolates. Following exposure of cells to a viral mixture, the proportion of viruses after several viral passages is compared with that in the initial mixture (Holland et al. 1991). Although laborious assays, growth competitions provide a more accurate measurement of viral replicative fitness by detecting the effect of small differences in replication rates and are generally reproducible in the same cell culture environment. Many different approaches have been developed to measure dual virus production in growth competition experiments. Most methods rely on laborious point mutation assays or on the sequencing of a large number of clones (Croteau et al. 1997; Sharma and Crumpacker 1997; Harrigan et al. 1998; Martinez-Picado et al. 1999, 2000; Imamichi et al. 2000). However, new studies use more rapid techniques to estimate the frequency of the two viruses in the competition: heteroduplex tracking assay (Yuste et al. 1999; Quiñones-Mateu et al. 2000; Nelson et al. 2000; Resch et al. 2001), realtime PCR (De Ronde et al. 2001; Lu and Kuritzkes 2001; Weber et al. 2003b), or recombinant viruses with reporter genes (Lu and Kuritzkes 2001; Zhang et al. 2004; Neumann et al. 2005; Weber and Quiñones-Mateu, unpublished results).

92

M. E. Quiñones-Mateu · E. J. Arts

What is the best method to measure HIV replicative fitness? Most studies analyzing the effects of drug-resistance mutations on HIV replicative fitness have used recombinant viruses based on the same genetic background (Mammano et al. 1998; Croteau et al. 1997; Sharma and Crumpacker 1997, 1999; Martinez-Picado et al. 1999, 2000; Imamichi et al. 2000). Although useful in correlating amino acid substitutions with alterations in replicative fitness without the effects from mutations outside the targeted viral gene, a great diversity between these assays hampers even comparisons of fitness values obtained using the same drug-resistant mutants cloned into the same HIV backgrounds. In a recent review article, we outlined discrepancies in the replicative fitness values of various viruses analyzed in several independent studies (Quiñones-Mateu and Arts 2002). Contradictory results were obtained depending on the assay used in each study, which may be due to differences in methodology (e.g., mono-, single-cycle, and competitive infections) or the use of various genetic backgrounds (e.g., primary HIV isolates or laboratory strains such as HIVNL4–3 and HIVHXB2 ) (Fig. 2). It is clear that a consensus on the experimental approach used to measure fitness has to be established to allow a direct qualitative and quantitative comparison between different studies.

3 Application to HIV Population Dynamics It is difficult to durably control HIV replication with currently available therapies, in part because of the extraordinary capacity of the virus to develop resistance to antiretroviral drugs (the theoretical basis are discussed in the chapters by Biebricher and Eigen, Wilke et al., and Domingo et al., this volume). All currently available classes of antiretroviral therapy, i.e., protease, reverse transcriptase, and fusion inhibitors, can select for mutations within the target gene that confers high-level drug resistance. Indeed, the capacity of a novel compound to select for a mutation is now used as evidence that the experimental agent has anti-HIV activity. Although these mutations result in an increased ability of the variant to replicate in the presence of drug, they often reduce the in vitro replicative fitness of the virus compared with wild-type strains (in the absence of drug pressure). Thus, in the absence of antiretroviral therapy, strains containing drug-resistance mutations have a reduced replicative fitness compared to the wild-type (wt) quasispecies within the population (Coffin 1995). Selective pressure in the form of drug therapy leads to dramatic shifts in the quasispecies distribution and fitness of those mutants with decreased sensitivity to the respective antiretrovirals (Coffin

Virus Fitness: Concept, Quantification, and Application

93

Fig. 2 Effects on viral replicative fitness of single mutations in genes targeted by antiretroviral drugs. Summary of integrase, protease, reverse transcriptase, and envelope mutations associated with resistance to the corresponding antiretroviral inhibitor. Red and black arrows represent a decrease or increase in viral replicative fitness, respectively, compared with the wild-type virus in the absence of antiretroviral drugs. Amino acids substitutions in blue do not affect viral replicate fitness (i.e., replicative fitness of the mutant virus is comparable to the wild-type strain used in that particular experiment). (This is a modified and updated version of data summarized in Quiñones-Mateu and Arts 2001)

94

M. E. Quiñones-Mateu · E. J. Arts

1995; Domingo and Holland 1997; Domingo et al. 1997, 1999). During this in vivo selection, several drug-resistant variants will emerge and compete for dominance. These resistant isolates will pass through the drug-induced bottleneck and initiate a new quasispecies distribution that will be governed again by replication efficiency (Coffin 1995; Loveday and Hill 1995). Many studies have examined the potential relationship of HIV replicative fitness with plasma viral load, drug resistance, and disease progression (reviewed in Quiñones-Mateu and Arts 2001, 2002; Nijhuis et al. 2001; Buhler et al. 2001). Here, we will describe the implications of viral fitness on different aspects of HIV drug resistance, pathogenesis, transmission, and evolution. 3.1 Replicative Fitness of Drug-Resistant Viruses Resistance to antiretroviral drugs is widespread in the Western hemisphere (Hirsch et al. 1998) and transmission of drug-resistant viruses is an increasing problem (Little et al. 1999, 2002; Tang and Pillay 2004). Currently, the nineteen antiretroviral drugs licensed in the United States fall into four classes: nucleoside reverse transcriptase inhibitors (NRTIs), non-nucleoside reverse transcriptase inhibitors (NNRTIs), protease inhibitors (PIs), and fusion inhibitors (FIs). Resistance mutations have been observed for all four classes of antiretroviral drugs (Deeks 2003; Gallant et al. 2003), including mutations in the gp41 envelope gene associated with resistance to enfuvirtide (EFV), the first in the new class of viral FI (Derdeyn et al. 2000; Menendez-Arias and Este 2004; Wei et al. 2002; De Ronde et al. 2001). Drug-resistance-associated mutations can either directly decrease the capacity of antiviral drug to inhibit HIV replication or compensate for the reduced replicative fitness associated with drug-resistance mutations. Both therefore contribute to increasing the relative fitness of the virus to replicate in the presence of drug. Such mutations are referred to as primary and secondary mutations (Quiñones-Mateu and Arts 2001). In general, primary resistance mutations involve amino acids located at the active site of the enzyme to which a given drug is directed (e.g., protease or reverse transcriptase). These mutations confer direct resistance to the drug, and often come at the price of compromised viral replication in the absence of drug pressure (reviewed in Nijhuis et al. 2001; QuiñonesMateu and Arts 2001). Secondary mutations (i.e., compensatory mutations) are selected and emerge under continued drug pressure. These mutations may also be located at the enzyme active site or its substrate cleavage site. Accumulation of compensatory mutations is assumed to restore the enzymatic activity of the drug-targeted resistant enzyme, usually leading to a rebound in fitness (Hirsch et al. 1998; Nijhuis et al. 2001; Berkhout 1999; Clavel et al.

Virus Fitness: Concept, Quantification, and Application

95

2000). It is important to note that the rate of appearance of specific mutations is often highly dependent on the baseline sequence and the sequential selection of “de novo” compensatory mutations that contribute to viral fitness (Weber et al. 2003a 2003b; Rose et al. 1996; Precious et al. 2000). Here, we will briefly describe the replicative fitness of viruses resistant to PI, RTI, and other antiretroviral inhibitors. Figure 2 summarizes the effect on replicative fitness of single amino acid substitutions associated with resistance to the most common antiretroviral drugs. 3.1.1 Viruses Resistant to Protease Inhibitors HIV protease is the enzyme responsible for cleavage of the viral Gag and Gag-Pol polyprotein into mature structural proteins and enzymes found in the infectious virion (Park and Morrow 1993; Miller 2001). Seven HIV protease inhibitors have been approved to date in the United States, i.e., amprenavir (APV), indinavir (IDV), lopinavir (LPV), nelfinavir (NFV), ritonavir (RTV), saquinavir (SQV), and atazanavir (ATV). Numerous primary and/or secondary mutations have been associated with HIV resistance to PI (Miller 2001; Shafer et al. 2000). Primary drug-resistant substitutions rarely dominate the quasispecies in PI-naïve HIV-infected individuals (Lech et al. 1996), suggesting they confer a selective disadvantage to the virus (Kozal et al. 1996; Shafer et al. 2000). For most of the protease inhibitors, primary PI-resistant mutations cluster near the active site of the enzyme, reducing both catalytic activity and viral fitness (Borman et al. 1996; Croteau et al. 1997; Mammano et al. 1998; Miller 2001). Secondary mutations within the protease gene compensate for the impairment on HIV replication by helping the enzyme to adapt to the primary changes in the active site (Borman et al. 1996; Ho et al. 1994; Nijhuis et al. 1998, 1999; Eastman et al. 1998; Mammano et al. 1998; Rose et al. 1996). In addition, increased PI-resistance is often associated with substitutions in the protease cleavage sites (gag and pol genes) (Doyon et al. 1996; Zhang et al. 1997; Miller 2001; Clavel et al. 2000; Nijhuis et al. 2001), providing better peptide substrates for the mutated protease and compensating for the loss in viral fitness (Doyon et al. 1996; Mammano et al. 1998; Zennou et al. 1998; Clavel et al. 2000; Nijhuis et al. 2001; Tamiya et al. 2004; Myint et al. 2004). Perhaps the most dramatic example of the impact of PI resistance on viral replicative fitness in a single amino acid substitution at position 30 in the protease gene (D30N) (Martinez-Picado et al. 1999; Gonzalez et al. 2004), which confers resistance to nelfinavir. Interestingly, discordant responses have occurred in HIV-infected patients treated with PI (i.e., high viral loads and sustained CD4+ T cell counts) (Deeks et al. 2000), suggesting a pos-

96

M. E. Quiñones-Mateu · E. J. Arts

sible preservation of CD4+ T cell counts in patients failing PI-based therapy and a beneficial effect of HIV strains with impaired fitness (see Sect. 3.2). 3.1.2 Viruses Resistant to Reverse Transcriptase Inhibitors NRTIs compete for binding to RT with the native deoxynucleoside triphosphates (dNTPs), can be incorporated into elongating HIV DNA, and result in chain termination (Arts and Wainberg 1996). To date, six NRTIs have been approved for therapy in the United States, i.e., zidovudine (AZT), didanosine (ddI), zalcitabine (ddC), stavudine (d4T), lamivudine (3TC), and abacavir (ABC). Tenofovir (TDF), a nucleotide-RT inhibitor, has recently received approval from the FDA. NRTIs were the first class of antiretrovirals to be approved for anti-HIV therapy (Loveday 2001). Thus, it is not surprising that some of the first studies showing the effect of drug-resistance mutations on viral replication fitness were related to AZT (Goudsmit et al. 1997; GarciaLerma et al. 2000, 2004; Harrigan et al. 1998; Arts et al. 1998). Other studies followed, describing the effects of many NRTI-resistance mutations in viral replicative fitness (Quiñones-Mateu and Arts 2001). The M184V mutation, which confers resistance to 3TC, has a marked effect on reducing RT processivity and replicative fitness (Larder et al. 1995; Back et al. 1996; Keulen et al. 1997; Yoshimura et al. 1999; Sharma and Crumpacker 1999; Frost et al. 2000; Devereux et al. 2001; Picchio et al. 2000; Lu and Kuritzkes 2001; MartinezPicado et al. 2001). Similarly, the K65R mutation (associated with resistance to TDF, ddC, ddI, d4T, and ABC) (Gu et al. 1994; Zhang et al. 1994; Winston et al. 2002; Margot et al. 2002) reduces HIV replicative fitness (White et al. 2002; Deval et al. 2004; Weber et al. 2005). Nonnucleoside reverse transcriptase inhibitors (NNRTIs) are noncompetitive or uncompetitive inhibitors that bind to a hydrophobic pocket adjacent to the polymerase active site of RT (Deeks 2001). This binding inhibits DNA polymerization by an allosteric change of the polymerase active site (Esnouf et al. 1995). Mutations in this hydrophobic binding pocket are rapidly selected during NNRTI therapy (Deeks 2001; Shafer et al. 2000). Three NNRTIs are currently approved for antiretroviral therapy in the United States: delavirdine (DLV), efavirenz (EFV) and nevirapine (NVP). Presence of NNRTI resistance mutations results in a slight decrease in RT polymerase activities relative to RNase H activity of the enzyme (Gerondelis et al. 1999; Archer et al. 2000). In contrast to NRTI, single-point NNRTI-resistant mutations such as 103N or 181C have limited effects on viral fitness but confer a high level resistance and persist in the absence of drug pressure (Imamichi et al. 2001; Archer et al. 2000; Iglesias-Ussel et al. 2002; Collins et al. 2004). Unlike other antiretroviral

Virus Fitness: Concept, Quantification, and Application

97

drugs, the genetic barrier to NNRTI resistance is minimal and replicative fit-resistant mutants are found at high frequencies in HIV quasispecies populations in the absence of therapy (Havlir et al. 1996; Demeter et al. 2000; Dueweke et al. 1993; Iglesias-Ussel et al. 2002). In addition to individuals, mutations conferring resistance to particular RT inhibitors, accumulation of multiple mutations associated with multidrugresistant (MDR) viruses may occur in heavily antiretroviral-experienced individuals. For example, the 151M mutation is thought to be the first MDR mutation selected in RT, followed by 75I, 77L, 116Y, and then 62V (also known as the MDR-151 complex) (Shirasaka et al. 1995; Shafer et al. 2000). Interestingly, specific combinations of these mutations actually appear more fit than wild-type viruses in the absence of drug (Garcia-Lerma et al. 2000; Kosalaraksa et al. 1999; Maeda et al. 1998). Another example is the MDR-69 complex, which confer high-level resistance to multiple RT inhibitors (Hirsch et al. 2000). Decreased viral replicative fitness in the absence of drugs is quite evident in HIV clones harboring these mutations in RT (Quiñones-Mateu et al. 2002; Prado et al. 2004). 3.1.3 Viruses Resistant to Other Antiretroviral Inhibitors Novel drugs in development target different steps in the virus life cycle, including host cell entry and viral integration. The most promising drugs are those that interfere with entry into host cells, either by inhibiting gp120 binding to the chemokine receptors CCR5 or CXCR4 (Cocchi et al. 1995; Simmons et al. 1997) or the fusion of the viral and cellular membranes (Kilby et al. 1998). We have shown that the rate of host cell entry appears to control fitness in wild-type viruses (Ball et al. 2003; Rangel et al. 2003). Thus, it is quite conceivable that decreased replicative fitness is also linked to increase resistance to these entry inhibitors. For example, we found that resistance to AMD3100, a bicyclam compound that binds to CXCR4, has a negative impact on HIV replication (Armand-Ugon et al. 2003). Another class of entry inhibitors was designed to block virus–cell fusion and target conserved fusion domains in gp41. Included in this class of inhibitors are T-20 (Enfuvirtide), T1249, C34, 5helix, and IQN-17 (Chan et al. 1997; Eckert et al. 1999; Greenberg et al. 2001; Kilby et al. 1998; Root et al. 2001). Enfurvirtide, the only fusion inhibitor approved to treat patients to date, is a synthetic peptide analog that binds to the alpha helix bundle region in the gp41 transmembrane domain, preventing formation of a hairpin structure necessary for membrane fusion. Although enfuvirtide-resistant recombinant viruses carrying mutations in the gp41 coding region (i.e., D36S, I37T, V38M/A, N42T/D, or N43K/S) have been

98

M. E. Quiñones-Mateu · E. J. Arts

shown to have diminished replicative fitness (Lu et al. 2004), recently data from our and other groups (using enfuvirtide-resistant HIV isolates) showed that pre-existent “compensatory” mutations within the env gene appear to guarantee a minimal effect of these mutations on HIV replicative fitness (Menzo et al. 2004; Weber and Quiñones-Mateu, unpublished results). Further studies using viruses resistant to this and other entry inhibitors are necessary to clarify the effect of drug-resistance mutations in the env gene over HIV fitness. Finally, integrase inhibitors (e.g., diketo acid analogs) are the most recent anti-HIV class of drugs (Fikkert et al. 2003; Hazuda et al. 2004a, 2004b; Svarovskaia et al. 2004). Resistance to these compounds is related to specific mutations in the integrase active site (i.e., 153Y, 66I, 154I, and 155S), which also impair enzymatic function in vitro (Hazuda et al. 2004a; Fikkert et al. 2003). As expected, increasing levels of resistance to integrase inhibitors is associated with significant loss of viral replicative fitness (Fikkert et al. 2003; Lee and Robinson 2004). 3.2 Clinical Implications of HIV Fitness A clinical paradox has emerged with the use of antiretroviral therapy: sustained CD4+ T cell counts despite the presence of relatively high viral loads. This “discordant” response has been attributed to a reduced viral replicative fitness following development of resistance to antiretroviral drugs (Deeks et al. 2001; Deeks 2003; Bates et al. 2003; Barbour et al. 2004). Although the immunologic benefits of suppressive antiretroviral therapies are greatest among persons in whom plasma HIV RNA levels are suppressed below the limits of detection, incomplete suppression of HIV replication with the emergence of viruses resistant to antiretroviral drugs seems to provide some degree of immunologic benefit. This would suggest that resistant viruses are less “pathogenic” in terms of inducing immune dysfunction than are wildtype viruses. It is not certain whether the diminished “pathogenicity” of drug-resistant viruses is related to diminished replicative fitness and a lower level of HIV replication or whether other viral pathogenesis factors determine the diminished ability of these strains to cause and maintain immune deficiency. Also, we still do not know if there is a correlation between replicative fitness and the ability of drug-resistant variants to be transmitted (i.e., the actual transmission event and not the ability of the newly transmitted virus to replicate in the new host). If so, it would mean that drug-resistant viruses have lower transmission fitness than wild-type strains, an issue still being debated (Leigh Brown et al. 2003; Simon et al. 2003; Gandhi et al. 2003). It is clear then that further fitness studies on HIV drug-resistant viruses could lead

Virus Fitness: Concept, Quantification, and Application

99

us to understand the potential of certain mutants to be inhibited by salvage therapy and should help to explain the apparent diminished pathogenicity and/or transmission of these isolates. 3.3 Fitness of the Transmitted HIV Isolate During primary HIV infection, individuals may be exposed to varying amounts of virus depending on the mode of transmission (Zhu et al. 1993, 1996 Richardson et al. 2003; Derdeyn et al. 2004). Sexual and vertical transmission generally involves small amounts of virus and infected cells, whereas a bolus dose of virus may be transmitted during direct blood-to-blood contact, e.g., contaminated needles or blood transfusions (Wolinsky et al. 1992; Zhu et al. 1993; Overbaugh et al. 1999). Regardless of the vertical transmission route, a significant genetic bottleneck reduces the HIV population size from donor to recipient (Zhu et al. 1993, 1996). The virus population recovered from primary HIV infections is more homogeneous with a narrow genetic distribution of quasispecies as compared to the donor HIV quasispecies. How this reduction in genetic diversity affects virus replication efficiency (or fitness) is not known, but fitness is dramatically decreased in other RNA virus systems when a stringent bottleneck is applied (Domingo and Holland 1997). As described below, quasispecies theory and Fisher’s geometric model suggest that larger, more diverse virus populations can evolve to higher fitness and that progressive, adaptive mutations must accumulate to override the deleterious changes. However, severe bottlenecks during each transmission event may have an effect of “turning back” the evolutionary clock by reducing genetic diversity, population size, and relative replicative fitness. As will be discussed, gains in viral fitness during HIV disease progression may be lost with each transmission event. Transmission models have been tested experimentally using FMDV, VSV, bacteriophage, and even HIV (Lazaro et al. 2003; Escarmis et al. 1996; Duarte et al. 1992; Yuste et al. 1999, 2000; Novella and Ebendick-Corp 2004; Elena et al. 1996; Burch and Chao 2004; Ruiz-Jarabo et al. 2003) (see chapter by Escarmís et al., this volume). Genetic bottlenecks can be created in vitro by reducing or maintaining small viral population sizes for each serial passage in tissue culture (Lazaro et al. 2003; Escarmis et al. 1996; Novella and Ebendick-Corp 2004; Elena et al. 1996). Alternatively, other studies have alternated different target cells for infection, resulting in stringent bottlenecks and immediate selection for tropic clones (Ruiz-Jarabo et al. 2003), which may be analogous to the tight genetic constraints observed in arthropod-borne viruses (Holmes 2003). The selective pressures applied in these experimental systems

100

M. E. Quiñones-Mateu · E. J. Arts

have reduced quasispecies diversity and significantly decreased fitness of the transmitted virus. Finally the most stringent bottleneck is the direct serial plaque-to-plaque transfer of virus (Lazaro et al. 2003). Although these experiments are not analogous to an in vivo transmission event, a plaque-to-plaque transfer involves a strong bottleneck, prevents the accumulation of most adaptive mutations, and leads to dramatic decreases in fitness. Interestingly, even in these extremes the virus is not brought to extinction but retains low-level fitness (Lazaro et al. 2003). Bergstrom et al. (1999) have described the consequences of vertical transmission bottlenecks on the evolution of pathogens by using both analytical and simulation methods. They demonstrated that severe bottlenecks reduce the virulence of a pathogen due to stochastic loss of fitness, a process analogous to Muller’s ratchet. They also contrast the differences between horizontal and vertical transmission, which vary dramatically in terms of more severe bottlenecks in the latter. When a pathogen is spread by vertical transmission, as is the case with HIV, the pathogen strives to reduce virulence in the interest of ensuring host survival and reproductive success. As described below, there may be evidence that HIV is actually evolving to reduce its virulence/fitness during the current pandemic (Ball et al. 2003). When the inoculum size is large, changes in the virus population due to sampling effects become insignificant and the genetic distribution of clones in the donor and recipient quasispecies will be similar during transmission. When inoculum size is small due to severe bottlenecks, reduction in fitness may be simply be due to sampling error from the population of clones in the donor. As described in Fig. 1, the transmission bottleneck with HIV may be most severe in cases of vertical transmission (Zhu et al. 1993, 1996). Most infections through heterosexual vertical transmission reduce the diversity in the donor quasispecies (population as many as 1012 clones = 10,000 nt genome × 0.001 substitutions/nt × 1010 virus particles/day) (Ho et al. 1995; Perelson et al. 1996; Wei et al. 1995; Preston et al. 1988; Mansky and Temin 1995) to only a few clones infecting the recipient. The selective factors creating the bottleneck in HIV transmission are diverse and likely include: 1. Host factors such as innate immune response 2. Density of target cells at the site of infection 3. Number of transmitted virions 4. The structure of transmitted viral quasispecies. Quasispecies distribution and size of the transmitted pool may have a significant effect on fitness of the infecting isolate and subsequent disease progres-

Virus Fitness: Concept, Quantification, and Application

101

sion (Domingo and Holland 1997; Novella et al. 1995b). The viral properties selected during initial infection may not be the same attributes necessary for efficient dissemination and rapid turnover during acute disease. A combination of these viral characteristics may only be found in exposures with a significant load of diverse HIV quasispecies. Finally, environmental differences (pH, target cells, mucosal composition) at the site of exposure may not only affect the efficiency of transmission but also fitness of the infecting isolates (Overbaugh et al. 1999; Blauvelt et al. 2000). For example, there appear to be phenotypic and genotypic differences between HIV variants infecting men and women following heterosexual contact (Overbaugh et al. 1999). Recently, Richardson et al. (2003) have shown that the diversity of the HIV population responsible for primary infection of women is related to the severity of subsequent disease, i.e., higher HIV-1 genetic diversity at transmission resulted in faster disease progression. Based on models by Bergstrom, Domingo, Chao, Holland and others, the greater sampling of clones from the donor HIV quasispecies would result in a greater chance of infection by high-fitness clones, which could expand more rapidly in the recipient HIV quasispecies (see, however, alternative possibilities in the chapters by Domingo et al. and Escarmís et al., this volume). Although there may be an element of chance in the expansion of a particular HIV clone, phenotypic selection does occur in nearly every HIV infection. There are two basic phenotypes of HIV that have different host cell tropisms. The SI HIV isolates, which can form syncytia in infected T cell cultures, utilize the CXCR4 chemokine receptor in addition to CD4 for host cell entry. The nonsyncytium-inducing (NSI) isolates utilize CCR5 as a co-receptor for entry. SI/X4 HIV isolates can dominate the quasispecies but only in late HIV disease. Regardless of HIV-1 phenotype in the donor, only the NSI/R5 variant is vertically transmitted to a recipient (reviewed in Fenyo et al. 2000). Preferential transmission of NSI/R5 over SI/X4 HIV isolates is contradictory to increased turnover of SI/X4 HIV over NSI/R5 isolates in culture (see Sect. 3.4) (Tersmette et al. 1988; Bjorndal et al. 1997). Thus, efficiency or fitness of transmission appears not to be related to the HIV replicative capacity measured in normal cell culture. As described below, this fitness dichotomy is not restricted to co-receptor usage but is also observed in transmission of different NSI/R5 HIV isolates in the human population (Blackard et al. 2001). Although in vivo findings suggest that NSI/R5 HIV isolates may out-compete the SI/X4 variants at the site of primary infection, one report suggests that the NSI/R5 isolates only predominate after a temporary expansion of SI/X4 HIV isolates is quenched by an activated immune response (Cornelissen et al. 1995). However, this observation is difficult to reconcile with the finding that humans who are homozygous for a deletion in the CCR5 gene (i.e., lack CCR5 on

102

M. E. Quiñones-Mateu · E. J. Arts

any cell surface) are typically resistant to HIV infection (O’Brien and Moore 2000; Dean et al. 1996b). To date, the universal factors (i.e., found in almost every human host) involved in the selection of NSI/R5 HIV isolates during transmission and asymptomatic disease are not well defined. Langerhans cells (LCs) are found embedded in mucosa (e.g., vaginal mucosa) and may be the first cell targets for primary heterosexual transmission (Blauvelt et al. 2000; Braathen et al. 1987; Soto-Ramirez et al. 1996). LCs may play a role in NSI/R5 HIV selection since CCR5 and not CXCR4 is preferentially expressed in situ and in the absence of external stimuli (Blauvelt et al. 2000). A recent report describes increased replication of an NSI/R5 (HIVBal ) over an SI/X4 (HIVIII-B ) isolate in LCs embedded in skin-derived explants even though the opposite is true in PBMC cultures or other permissive cell lines (Blauvelt et al. 2000; Soto-Ramirez et al. 1996). Aside from the preferential infection by NSI/R5 isolates, there is little evidence supporting the notion that specific clones from the donor quasispecies are selected during transmission. It is conceivable that one NSI/R5 clone may be more adept than others at infecting LCs or other primary target cells at the site of transmission. However, more recent results suggest the contrary. In competition experiments with a subtype B and C primary HIV isolate, the subtype B isolate dominated over the C strain in PBMC cultures regardless of the genetic background of the human donor (Ball et al. 2003). In contrast, these two viruses competed equally in skin explant cultures where LCs were the only target cell (Ball et al. 2003). Fitness differences derived from these two primary human cultures

Fig. 3 A model for increasing HIV-1 replicative fitness and genetic diversity during disease progression

Virus Fitness: Concept, Quantification, and Application

103

may be attributable to how these viruses compete. Relative HIV replicative capacity appears to be directly related to the efficiency of host cell entry. Cells such as LCs or other members of the dendritic cell lineage have the ability to trap virus on the cell surface through various adhesion molecules, including DC sign (Geijtenbeek et al. 2000; McDonald et al. 2003). Trapping of virus on the cell surface may negate the rate-limiting step of host cell entry (Ball et al. 2003; Rangel et al. 2003; Marozsan et al. 2005) and thus, dampen differences in fitness (Ball et al. 2003). The net result of these findings would be that all infectious NSI/R5 HIV clones had an equal opportunity to establish infection in the recipient. Of course, as indicated in Fig. 3, the extreme bottleneck and random sampling of the quasispecies would favor infection by clones falling under the peak of normal fitness distribution. How fitness increases and how this fitness distribution expands during disease will be discussed in the following sections. 3.4 HIV Fitness During Disease Progression Although HIV is the etiological agent of AIDS, it is often assumed that phenotypic characteristics and replication efficiency (ex vivo fitness) of the infecting, wild-type HIV isolates has little impact on the rate of disease progression. In general, acute HIV infection is followed by a chronic and progressing disease resulting in immunodeficiency and acquisition of various life-threatening opportunistic infections. Multiple studies have tried to compare both host and viral factors, with HIV pathogenesis and progression to AIDS. The strongest correlates of HIV disease progression include various host immunological and genetic factors: 1. A strong HIV-specific CD8+ cytotoxic lymphocyte throughout disease 2. Retention and proliferation of HIV-specific CD4+ lymphocyte response following acute infection (Pontesilli et al. 1998; Dyer et al. 1999; Cao et al. 1995; Pantaleo et al. 1995; Montefiori et al. 1996; Carotenuto et al. 1998; Rosenberg et al. 1997) 3. Some polymorphisms in the CCR5, RANTES, and SDF-1 promoter regions or altered expression of the chemokines (Dean et al. 1996a; Berger et al. 1999) 4. Specific HLA class I genes (De Maria and Moretta 2000) However, there are several general observations suggesting that HIV fitness may play an important role in disease progression, i.e., (a) HIV load is the best marker to date for disease progression (Mellors et al. 1996), (b) treatment

104

M. E. Quiñones-Mateu · E. J. Arts

with highly active antiretroviral therapy (HAART) may reduce viral loads to undetectable levels and delays progression to AIDS indefinitely, and (c) emergence of drug-resistant isolates and rebound in viral load results in resumption of normal disease progression. In addition to these general observations, there are also three clear examples that HIV phenotype can affect disease progression. First, the faster replicating SI/X4 HIV isolates are generally isolated during AIDS or in late HIV disease, whereas the slower replicating, NSI/R5 strains generally predominate during asymptomatic stages (Asjö et al. 1986; Tersmette et al. 1988; Bjorndal et al. 1997). Several studies have now shown that appearance of SI/X4 isolates does coincide with rapid decline in CD4 cells, a burst in viral load, and the onset of AIDS. However, SI/X4 isolates are inconsistently isolated in late stages of disease and are not a prerequisite for progression or AIDS (Cheng-Mayer et al. 1988; Schuitemaker et al. 1992; Fenyo et al. 2000). Using a mathematical model, Wodarz and Nowak (Wodarz and Nowak 1998) suggested that the evolution of SI/X4 strains depends on the fitness landscape of the HIV quasispecies. Variation in intrapatient HIV evolution and the fitness landscape would explain why SI/X4 viruses only appear in half of HIV-infected individuals. Recently, the dogma that all SI/X4 isolates are more fit in cell culture than NSI/R5 isolates has been challenged by competing several NSI/R5–SI/X4 pairs in PBMC cultures (Quiñones-Mateu et al. 2000). Although most SI/X4 were more fit in cell culture, NSI/R5 isolates from rapid progressors could outcompete SI/X4 isolates from long-term survivors or even SI/X4 isolates from patients displaying typical HIV progression (Quiñones-Mateu et al. 2000). Aside from these changes in co-receptor usage/cell tropism, evidence that HIV genetic alterations could affect disease progression was clearly demonstrated in a few long-term nonprogressor patients (LTNPs) shown to harbor HIV strains with nef deletions (Deacon et al. 1995; Kirchhoff et al. 1995). Independently of these studies, Daniel et al. (1992) had generated several HIV clones with similar nef deletions, all of which were replication defective in PBMC cultures. Several of the LTNP infected with nef -deleted viruses eventually progressed to AIDS (after >10 years of asymptomatic disease) (Brambilla et al. 1999). Changes in HIV fitness during this time period have not been published but would provide valuable information on the possible accumulation of compensatory mutations. It is important to note that a similar rebound in HIV fitness does occur during the emergence of drug-resistant mutations, albeit on a shorter time scale. In these LTNPs now progressing to AIDS, the Red Queen Hypothesis would support the notion that slow but continual replication of HIV would expand the quasispecies diversity and lead to increased fitness. Only genetic bottlenecks due to changes in selective pressure could reverse this trend (i.e., Muller’s ratchet). Although the impact of HIV fitness

Virus Fitness: Concept, Quantification, and Application

105

on disease progression is still not well understood, studies on SIV pathogenesis have provided valuable information about the disease process that leads to simian AIDS (reviewed in Whetter et al. 1999). Using the SIV model, Kimata et al. (1999) showed that antigenic and cytopathic properties of the SIV strain and not phenotype (e.g., SI vs NSI) predict fitness in the host. Emerging SIV variants have increased replicative capacity over the SIV strain used in the initial infection. Furthermore, infectious dose and virulence (viral fitness) of the initial inoculum did influence viral load during disease progression in SIV-infected macaques (Holterman et al. 2000). As described above, several viral parameters such as infectious dose, route of infection, and viral fitness may be contributing to the clinical course of HIV infection (Tersmette and Miedema 1990; Coffin 1986; Zhu et al. 1993; Lukashov and Goudsmit 1997). Subtle differences in viral fitness may also contribute to HIV pathogenesis and even overwhelm an HIV-specific immune response. In an attempt to correlate ex vivo HIV replicative capacity and disease progression, several studies have compared the replication differences of primary HIV isolates using monoinfections of PBMCs or a CD4+ tumor cell line (Blaak et al. 1998; Tersmette et al. 1988; Schuitemaker et al. 1992; Fenyo et al. 1989). Variations in the microenvironment of tissue culture systems, the lack of an internal control, and the possibility of a high virus titer saturating the target cells, all lead to inconsistent results with monoinfections. Moreover, a moderate fitness cannot be accurately measured in monoinfections because of the inherent variability between cultures. This may explain why this assay has only characterized drastic defects in the replication kinetics of HIV isolates from long-term nonprogressors (LTNPs). Several studies have examined the replication kinetics of HIV isolated from patients with atypical progression (Blaak et al. 1998; Fenyo et al. 1989; Schuitemaker et al. 1992). Using a monoinfection assay with primary HIV isolates, Blaak et al. (1998) showed that some LTNPs harbored NSI isolates with slow replication kinetics. This observation has been supported by other studies by this group (Kimata et al. 1999; Kwa et al. 2003). Until recently, there have been few studies that have compared the replication kinetics of primary HIV isolates using competition assays, which have become standard with most other RNA virus systems. A study published in 2000 described competition of HIV isolates from long-term survivors (LTSs) and typical progressors (PRO) with four reference HIV strains (i.e., primary isolates) to determine a relative replicative fitness in primary human CD4+ PBMCs (Quiñones-Mateu et al. 2000). PRO HIV isolates out-competed the reference strains in growth competition experiments, while the opposite was observed for HIV isolated from longterm survivors. These results suggest that regardless of the viral phenotype (NSI or SI), HIV isolates from long-term survivors were less fit than strains

106

M. E. Quiñones-Mateu · E. J. Arts

obtained from progressors (Quiñones-Mateu et al. 2000). The relative fitness values of each LTS and PRO isolate also showed strong correlations with viral RNA load (Quiñones-Mateu et al. 2000). This initial study on ex vivo HIV fitness in LTS or PRO patients has now been expanded to include at least four consecutive HIV isolates from each of ten patients studied over a 2- to 5-year period (Troyer et al. 2005). In each case, the relative fitness of the primary HIV isolates increased during disease and as a direct correlate of increasing viral load and decreasing CD4 cell counts (Troyer et al. 2005). These data support earlier findings that the fitness of the infecting HIV isolate may predict the clinical course of disease (Lukashov and Goudsmit 1997). As described above, the diversity of the infecting HIV isolate at acute infection did appear to correlate with subsequent disease progression (Richardson et al. 2003; Derdeyn et al. 2004), suggesting that a greater sampling of the donor quasispecies may increase the chances of transmitting higher-fitness clones. In a current model (Fig. 3), decreased genetic diversity in donor quasispecies and low infectious dose results in transmission of an HIV clone of reduced fitness. 3.5 HIV Fitness and Immune Response Rosenberg et al. (1997) have suggested that rapid depletion of CD4+ cells during acute/early infection results in an irreplaceable loss of HIV-specific T helper cells. In contrast, proliferation and retention of HIV-specific T helper cells during early disease is associated with slower disease progression (Rosenberg et al. 1997). Thus, it is quite conceivable that infection with an isolate of poor replicative capacity may lead to limited depletion of CD4+ cells and retention of HIV-specific T helper cells. Slower disease progression may then be due to a combination of an enhanced HIV-specific immune response and reduced HIV fitness. It is important to note that infection with an HIV clone outside of the normal distribution of quasispecies would be rare, which could explain the low frequency of atypical progression (e.g., LTS or rapid progressors). Considering most individuals would be infected with a virus of “modest” replicative fitness, HIV-specific CD4+ T cell help would persist, albeit at lower levels than in response to other pathogens that do not directly target CD4+ cells. This HIV-specific CD4+ T cell help could then stimulate class I and II restricted responses, increase anti-HIV antibody production, some of which may be neutralizing, and activate of HIV-specific CD8+ (or CD4+ ) cytotoxic T lymphocytes. This hypothesis brings up the “chicken or egg” scenario since strong CD4+ T cell responses can also stimulate HIVspecific CTL activity, which in turn can reduce HIV fitness by creating strong genetic bottlenecks and selecting for “less” fit CTL escape clones.

Virus Fitness: Concept, Quantification, and Application

107

HIV-specific CTLs are thought to be at least partly responsible for initial immunological control of HIV viremia following acute infection. Continual viral replication following the establishment of viral load set point may be related to virus “escape” from this CTL pressure. To understand the direct relationship between anti-HIV CTL activity and immune avoidance by virus, it is important to first define the immune response to HIV and second, provide an analogy for immune escape by comparing genetic/fitness changes associated with the emergence of HIV drug resistance. There are three class I genes: HLA-A, B, and C. Polymorphic alleles for each gene expand the breath of antigen presentation by simply providing the possibility of six different coding genes for MHC class I. Early studies suggest that homozygosity in these loci will result in a more limited CTL response to fewer epitopes, which may in turn lead to faster disease progression. In contrast, heterozygosity in combination with specific alleles at each loci (e.g., HLA-B57 or HLA-B27) can result in apparent control of viremia and even long-term nonprogression to AIDS (Carrington et al. 1999; Kaslow et al. 1996; Migueles et al. 2000; Gao et al. 2001). The presentation of specific HIV peptides in association with MHC class I varies and appears to be related to protein/peptide processing efficiency as well as the avidity/affinity of MHC I for the peptide. Dominant vs weaker epitope responses are likely due to multiple aspects of antigen presentation, including abundance of specific viral proteins in infected cells, peptide processing, MHC I-peptide interaction, and subsequent binding to specific T cell receptors. Regardless, dominant epitope responses during infection appear most responsible in controlling viremia (Santra et al. 2002; Goulder et al. 2001). Furthermore, several studies have now shown that the breadth and specificities of HIV-specific CD8+ T cells identified during acute infection may be distinct from those during chronic infection (Altfeld et al. 2001; Dalod et al. 1999; Cao et al. 2003). Thus, it is not surprising that HIV isolates that predominate at early infection often develop mutations in dominant CTL epitopes. Goulder and Watkins (2004) provide a thorough review on the emergence of CTL escape mutations in HIV as well as describe the escape mechanisms encoded by these mutations. HIV escape to a specific CTL response was first described by Phillips et al. in 1991 (Phillips et al. 1991) but was previously well documented in other virus infections (Aebischer et al. 1991; White et al. 1990). CTL epitope mapping of the HIV proteins for specific HLA genetic backgrounds has helped to identify various CTL escape mutations during the course of disease (Korber et al. 1997b). New technologies such as ELISPOT can then assess if a specific mutation within an epitope actually confers “resistance” or escape to the specific CTL response (Goulder and Watkins 2004). By employing these techniques, a large volume of literature has identified various

108

M. E. Quiñones-Mateu · E. J. Arts

escape mutations. Escape mutations that may have the greatest impact on disease progression typically emerge during early infection rather than chronic disease (Altfeld et al. 2001; Dalod et al. 1999; Cao et al. 2003). It is hypothesized that the emergence and possibly the timing of early escape mutations may be linked to (a) the strength of the CTL responses for specific epitopes and/or (b) with the cost of escape mutations on replicative fitness. Certainly, many putative epitopes and escape mutations are actually missed due to limitations of epitope mapping and the use of consensus peptides to generate this map. It is unreasonable to map the epitopes in all HIV-infected patients using the homologous HIV peptide libraries. The few studies that have used this approach have confirmed studies in SIV/macaque models (Evans et al. 1999; Allen et al. 2000; O’Connor et al. 2002; Barouch et al. 2002) and have described early escape in CTL epitopes found in accessory proteins (e.g., Tat, Vpr, and Nef) whereas escape in later time points were observed in structural genes (e.g., Gag) (Cao et al. 2003; Addo et al. 2001). As outlined by Goulder and Watkins (2004), it is tempting to speculate that epitopes encoded by the most conserved and functionally constrained HIV sequences may take longer to escape and may require the accumulation of compensatory mutations. For example, the emergence of an escape mutation in the conserved KK10 epitope (R264K) in gag appear in HLA-B27 restricted humans with the emergence of compensatory mutations (L268 M) (Kelleher et al. 2001). It is assumed that these mutations come at great fitness cost to the virus, but unfortunately, these hypotheses have not been substantiated with empirical data that actually test the replicative fitness of the virus prior to and after the appearance of escape mutations. HIV-specific CTL responses are likely responsible for a genetic bottleneck, which ultimately results in selection of CTL escape variants. However, the strength of this bottleneck is not well understood in terms of reducing quasispecies diversity or replicative fitness. In some ways, introduction of combination antiretroviral treatment may be analogous to an HIV-specific CTL response. Both are targeting multiple epitopes or gene functions. Both reduce viral load and may slow disease progression. However, only combination antiretroviral treatment reduces viral load to undetectable levels. In fact, many antiretroviral drugs such as non-nucleoside RT inhibitors and protease inhibitors can effectively lower viral loads to a greater extent than a CTL response directed at multiple epitopes (Chouquet et al. 2002; Goulder et al. 2001). Thus, it is likely that antiretroviral drugs induce a greater genetic bottleneck than CTL pressure. In addition, there is discrepancy in the nature of the target HIV gene or sequence. CTL epitopes are selected by protease processing of the viral proteins and by virtue of affinity/avidity of the various HIV peptides for MHC class II molecules. In contrast, antiretroviral drugs were selected to target specific enzymatic steps in viral replication. Resistance

Virus Fitness: Concept, Quantification, and Application

109

to ARVs is typically conferred by mutations that reduce drug binding and are in relatively conserved regions of the protein (e.g., RT or PR) that contribute to enzymatic function (Larder 2001; Arts and Wainberg 1996). Thus, it is not surprising that resistant mutations have a significant fitness cost (QuiñonesMateu and Arts 2002). Mutations that emerge in CTL epitopes during early HIV infection may or may not fall in conserved sites or regions important for function. It is possible that a high proportion of these mutations may not have a cost to overall fitness that is comparable to drug-resistant mutations. As described earlier, this topic has not been explored in experimental studies. Many studies on fitness theory using other RNA viruses indicate that horizontal transmission in the absence of strong bottlenecks will permit the accumulation of adaptive mutations (Burch and Chao 1999, 2000; Burch et al. 2003). Again, experimental approaches have not been employed to address this topic. Do escape mutations in CTL epitopes revert back to the wild-type sequence? Are compensatory mutations adequate to regain fitness and stabilize this genotype? When applying Wright’s fitness landscape to this scenario, it is possible that the valley of fitness loss due to a CTL escape mutation may be sufficiently shallow to permit a rapid climb to the adjacent fitness ridge with only a few compensatory mutations. Thus, these original CTL escape mutations would likely not revert later in disease progression. In more conserved and functionally constrained sites such as (a) in HLA-B27-restricted CTL KK10 epitope in gag (R264K) (Goulder et al. 1997; Kelleher et al. 2001) or (b) in the RT of 3TC-resistant virus (M184V) (Frost et al. 2000; Newstein and Desrosiers 2001), escape or resistance may result in a deeper valley of fitness loss, which would require a longer time for the accumulation of compensatory mutations to stabilize the escape genotype, i.e., analogous to a slow climb up to the fitness ridge. In these cases, reduced CTL pressure in late disease stages or removal of antiretroviral drug pressure would result in a reversion of the escape/resistance mutation to a wild-type sequence and possibly a jump back to original fitness ridge. 3.6 HIV Diversity and Fitness Genetic diversity in HIV is often overlooked as a parameter of disease progression due to complications in interpretation. However, the great propensity of HIV to mutate and recombine undoubtedly shapes the fitness of virus in the context of replication kinetics, host tropism, and immune evasion. Most RNA viruses have a polymerase capable of high mutation rates and yet diversity remains relatively low within an individual host infection (Moya et al. 2004; Simmonds 2004). This low genetic diversity may be one reason for resolution

110

M. E. Quiñones-Mateu · E. J. Arts

of these acute RNA virus infections by host immune response. As described earlier, not all mutations result in adaptation. In the presence of constant selective pressure, deleterious mutations may accumulate to aid in viral clearance. How these deleterious mutations interact and affect fitness is the basis of many studies in evolutionary biology and population genetics (Kondrashov and Crow 1991; Partridge and Barton 1993; Barton and Charlesworth 1998). Epistasis is defined by the association between mutations on fitness and may be dependent on the total abundance of deleterious and adaptive mutations in a genome. Independent relationships between mutation would result in a linear, additive/subtractive interaction between mutations (multiplicative; Fig. 4; Burch et al. 2003). Synergistic or “negative” epistasis is somewhat analogous to Muller’s ratchet and suggests that the accumulation of deleterious mutations relative to total mutations during selective pressure would lead to continual log decreases in viral fitness (Burch et al. 2003). However, “negative” epistasis also applies to the theoretical case of two independent, beneficial mutations resulting in less than additive gains in fitness when linked through mechanisms such as recombination. This theory of synergistic or negative epistasis was recently challenged by experimental data showing that continual plaque-to-plaque transfers of FMDV does not result in steady but rather fluctuating decline in fitness (Lazaro et al. 2003). In these cases, compensatory adaptation (restoration of fitness) may occur faster under synergistic epistasis because it allows for increased impact of each compensatory mutation

Fig. 4 Hypothetical fitness effects of increasing deleterious mutation number. (Adapted from Burch et al. 2003).

Virus Fitness: Concept, Quantification, and Application

111

(Moore et al. 2000). In conservation biology, negative or synergistic epistasis plays a role in determining the rate of fitness loss via genetic drift in small populations (Kondrashov 1994), but recent studies suggest that these general theories on epistasis do not always apply in pathogen evolution. In studies by Elena and Lenski (1997) and (de Visser et al. 1997) where Escherichia coli or Aspergillus niger were mutagenized, some mutation combinations showed negative epistasis, whereas others displayed positive epistasis. On average, no epistasis was evident but it is important to note that the worst mutational combinations could have been missed due to rapid extinction of the low fit variants. Epistatic theory was also examined in RNA viruses through the use attenuated viral vaccine (Burch et al. 2003), FMDV (Elena 1999), and RNA bacteriophage s6 (Burch and Chao 2004) in stringent bottlenecks, and ribavirin as a mutagen (Crotty et al. 2001). In two of these studies (Burch et al. 2003; Elena 1999), there was a log-linear relationship between mutation number and fitness, suggesting the lack of epistasis. The trend toward positive epistasis using various RNA viruses treated with ribavirin was not significant (Crotty et al. 2001). Only a recent study using s6 showed positive epistasis as inferred by the observation that low fitness genotypes were less sensitive to deleterious mutations (Burch and Chao 2004). In retroviruses, recombination due the diploid nature of the virion can reshuffle genetic material and can be viewed as analogous to sexual reproduction in higher organisms. It is generally understood that sex creates mutational combinations of very low fitness that are more quickly eliminated by selection (Eshel and Feldman 1970; Kondrashov 1988; Agrawal 2001; Siller 2001), but this theory may not be applicable for RNA virus systems (Novella et al. 2004; Froissart et al. 2004). The genetic complexity of the quasispecies and the high frequency of sex could contribute to rapid adaptation of viral clones due to recombination of two or more adaptive/compensatory mutations to overcome any deleterious changes. Recent studies by Bonhoeffer et al. (2004) and Sanjuan et al. (2004) have tested the epistasis theories by utilizing HIV1, capable of recombination and vesicular stomatitis virus, which can only evolve through point mutations and not recombination (respectively). The epistasis study on HIV-1 involved fitness and sequence data on 9466 HIV-1 isolates clearly described a slightly positive epistasis. Although these results are strongly supported by an extensive database, the population of sequences are skewed to one HIV-1 coding region (protease and reverse transcriptase) and derived specifically from patients under drug selection targeting these gene products (Bonhoeffer et al. 2004). Drug-resistant mutations that would lead to increased fitness in the presence of drug typically result in reduced fitness in its absence. Nonetheless, it is also likely that recombination could promote the inclusion of compensatory mutations and that any antagonism

112

M. E. Quiñones-Mateu · E. J. Arts

between negative mutations and recombination (evidence of “negative” epistasis) might simply be lost due to competitive elimination of these less fit clones in the viral population. Positive epistasis may imply that HIV-1 is gaining replicative fitness during the epidemic. However, this observation is controversial based on the tendency of pathogens to attenuate over time. Of course, the majority of studies described above test the impact of mutations on fitness in very controlled in vitro systems. The direct effect of neutral or positive epistasis in pathogen evolution and fitness within the complex host infection is poorly understood. In the face of strong selection pressure, virus must rapidly adapt to survive. Fitness of an obligate parasite, such as HIV, is defined by how these properties affect fecundity (or replication) and survival within a particular host environment. In general, fitness gains are dependent on increasing population size and genotypic complexity, commonly termed Red Queen dynamics (RQD). RQD for RNA viruses was best demonstrated by a study by Novella et al. (1995b) where they showed that repeated transfers of large VSV populations in cell culture resulted in exponential fitness gains. When it was later shown that HIV populations also increase in diversity and size during disease progression, it was natural to assume that this may be associated with concomitant increases in fitness. However, the forces shaping the HIV genetic variation during the course of chronic disease are complex. In the case of the envelope gene, genetic variation is likely the result of random genetic drift as well as selection by the immune response (both humoral and CTL) and by the infection of various host cell types with different co-receptors for entry. Following escape or adaptation to a new environment due to CTL pressure or host cell differences, the virus can accumulate mutations that enhance replicative fitness and possibly become fixed in the population. Four studies have now tested whether random drift or natural selection is shaping the HIV population during disease. Using coalescent theory-based methods and assuming neutrality, low population size, no recombination, and no population growth, Leigh Brown (1997) suggested that random genetic drift may be more important than natural selection in shaping the HIV populations. This hypothesis was supported by Seo et al. (2002), who used a maximum-likelihood approach but also assumed no population growth or recombination. In contrast, natural selection during intrapatient HIV evolution was supported by the linkage disequilibrium test for large population sizes applied by Rouzine and Coffin (1999). Unfortunately, these studies were limited by the number of HIV env sequences from sequential sampling of many patients. The empirical data from Shankarappa et al. (1999), which clearly showed increasing HIV diversity during asymptomatic disease, was employed by this same group to test natural selection vs random drift during disease (Shriner et al. 2004). By employing Tajima’s D test, Fu and Li’s D test,

Virus Fitness: Concept, Quantification, and Application

113

and the test of recurrent mutations, they discovered HIV genetic variation during disease is due mostly to random genetic drift with brief episodes of natural selection. These analyses did not support the possibility of recurring selective sweeps affecting env during chronic infection and that natural selection only played a minor role in driving diversity (Shriner et al. 2004). This result was somewhat surprising considering the HIV quasispecies continues to escape anti-HIV humoral/antibody response throughout chronic disease (Richman et al. 2003; Wei et al. 2003). If humoral responses did invoke strong selective pressure, there should have been continual or more frequent episodes of purifying selection in env due to escape. If fitness increases are due to distinct mutations that can be fixed in the population, there is also a dichotomy in the lack of natural selection and the observation that HIV replicative fitness and diversity both increase during disease (Troyer and Arts, in press). However, aside from the distinct mutations responsible switch in co-receptor usage, it appears that multiple combinations of different linked mutations may be responsible for increases in fitness during disease. Because of high recombination frequencies, these combinations that enhance fitness may never be fixed in the population. These new revelations in HIV genetic diversity during disease progression hark back to earlier studies that proposed the model of antigen diversity threshold in HIV disease (Nowak et al. 1991). In this model, the dynamic interaction between viral diversity and the human immune system suggests a threshold for HIV diversity, below which the immune system is able to regulate viral population growth but above which the virus population induces the collapse of the CD4+ lymphocyte population. As described below, this model could be modified to suggest that increases in replicative fitness as a function of increasing diversity (Troyer and Arts, in press) may be the key in the destruction of CD4+ T cell subset and escalating progression to AIDS. 3.7 A Model for HIV Fitness During Disease Based on the findings and assumptions in the above sections, a model of fitness during HIV disease progression can be formulated. This theoretical model was established on several fundamental observations, namely 1. HIV transmission is accompanied by a stringent bottleneck that only permits infection by few clones 2. Initial CTL pressure creates another genetic bottleneck for the selection of CTL escape mutants that have lower replicative fitness 3. Eventual escape from dominant CTL pressure promotes a slow expansion of the HIV population size and diversity

114

M. E. Quiñones-Mateu · E. J. Arts

4. HIV fitness steadily increases during disease 5. Exhaustion of susceptible cells/host destruction leads to decreases in diversity and possible attenuation. Figure 5 represents the distribution of HIV clones in the donor quasispecies. Following prolonged infection, the donor has accumulated a population of clones, each with a defined frequency and fitness. The normal distribution is likely to continually shift to the left (increasing fitness) as new high-fitness clones emerge and then dominate the population. Low-fitness clones may not be completely eliminated from the population as described in the quasispecies theory and the total fitness of the infecting HIV isolate is equal to the area under this curve. During transmission to the recipient (Fig. 5), the transmitted virus undergoes a stringent bottleneck, which would sample the normal distribution of clones in the donor. Based on the frequency of clones, most patients would be infected with clones of “moderate” fitness.

Fig. 5 A model for changes in HIV-1 population diversity and fitness during transmission

Virus Fitness: Concept, Quantification, and Application

115

However, increased sampling due to a weak bottleneck at transmission may result in higher genetic diversity in the recipient and more frequent sampling of the clones of higher and lower fitness. During the initial infection, the higher-fitness clones may then out-compete the less fit HIV clones, ravage the CD4 T cell population, and even decrease T cell-mediated immunity. This hypothetical scenario, albeit rare, could lead to rapid progression to AIDS. Aside from innate immunity, HIV populations rapidly expand in the absence of direct selection and loads can reach greater than 108 copies/ml of plasma following acute infection. Introduction of T cell-mediated immunity directed toward HIV will create another bottleneck, limiting quasispecies expansion and selecting for escape mutants with lower fitness (Fig. 5). The expanded clones at acute infection will be forced through repeated bottlenecks due to sequential escape from CTL pressure. Coupled with this escape, clones can select for compensatory mutations that may partly restore fitness (Fig. 5). However, Muller’s ratchet suggests that the accumulation of these deleterious mutations (CTL escape mutations) will override most compensatory mutations. This oscillation between fitness loss and slight fitness gain likely continues for at least 1 year and may subside with the advent of viral load set point. In other words, HIV may reach trough levels due to continual depletion by immune responses and the accumulation of deleterious mutations to escape the overall CTL response directed at multiple CTL epitopes. Following CTL escape, the greatest selective pressure on the virus population may be the HIV-specific humoral response. In most studies, it is clear that the humoral/antibody response is less efficient in controlling viremia than CTL response. Escape from humoral response is almost immediate and continuous throughout disease progression and, based on several models, is not sufficient to induce prolonged selection on the population. The major dominant epitope, the hypervariable V3 loop of the HIV envelope glycoprotein, is extremely accommodating to genetic drift, suggesting that humoral escape may come at minimal fitness cost. During this stage of disease progression, it is possible that increases in HIV diversity (due to random drift) could exceed the accumulation of deleterious mutations to escape CTL pressure (to possibly weaker CTL pressure), humoral responses, or adapt to changing host cell environments. Thus, minimal bottlenecks could allow the HIV quasispecies to expand in total number and diversity. Thus, concomitant fitness increases may suggest that Red Queen dynamics is in play. In panel C-III, the fitness and diversity would continue to increase until the susceptible cell population and density (CCR5+ /CD4+ cells) was exhausted. A new genetic bottleneck may coincide with the selection of clones in the quasispecies that can infect new susceptible host cell populations, i.e., CXCR/CD4-expressing cells. This bottleneck would reduce diversity and possibly decrease fitness. One discrepancy

116

M. E. Quiñones-Mateu · E. J. Arts

to this theory is the actual increased replicative capacity of CXCR4-tropic HIV isolates as compared to CCR5-tropic strains. However, it is still not clear as to when this co-receptor switch may occur during disease. Experimental studies are underway to test if the transition between CCR5 and CXCR4 co-receptor usage is accompanied by a decrease in fitness, which is obscured by the mixed HIV population until late in disease. 3.8 Are There Intersubtype HIV Variations in Viral Fitness? Divergent interpatient HIV evolution coupled with new introductions into susceptible human populations has led to the current HIV diversification from original zoonotic jumps in central Africa (Hahn et al. 2000; Korber et al. 2000). Although a founder effect results in further HIV spread and divergent evolution in different regions in the world, the phylogenies of most non-African HIV strains can be traced to central African isolates (Hahn et al. 2000; Korber et al. 2000). At least two to three separate zoonotic jumps from chimpanzees into humans led to the disproportionate spread of HIV groups M (main), O (outlier), and N (non-M/non-O) (Hahn et al. 2000; Gao et al. 1999; Kuiken et al. 1999). Increased fitness has likely played a key role in the predominance and extreme variation of HIV group M over group N or O isolates. Phylogenetic and recombination analyses further subdivided the HIV group M into nine subtypes (A, B, C, D, F, G, H, J, and K) and fourteen circulating recombinant forms (CRF) (reviewed in Quiñones-Mateu and Arts 1999; Peeters 2000). These subtypes and CRFs are unequally distributed across the globe, e.g., subtype B in the Americas and Europe whereas A, C, and CRF02-AG are the most prevalent clades in Africa (Quiñones-Mateu and Arts 1999; Peeters 2000). To date, CRF have been identified in nearly every region of the world where two or more subtypes co-circulate and may account for over 10% of new HIV infections (Peeters 2000; Korber et al. 1997a; Quiñones-Mateu and Arts 1999). The proportions of subtypes in defined populations are not stable but are in constant flux due to new introductions of HIV subtypes, changes in human behavior, therapeutic intervention, mode of transmission, and possibly subtype fitness. Over the past decade, there has been a considerable shift in the epicenter of the HIV epidemic from sub-Saharan Africa to Southern Africa, India, and Southeast Asia (Essex 1999; Kalish et al. 1995; Santiago et al. 1998; Lole et al. 1999). Subtype C has now emerged as the predominant clade in the world due to these regional pandemics and accounts for at least half of all infections worldwide (Hu et al. 1996; Rodenburg et al. 2001). Although subtype B likely preceded subtype C as a founder clade in India and China,

Virus Fitness: Concept, Quantification, and Application

117

most of the new infections in these countries are attributable to subtype C isolates or intersubtype B/C recombinants (Rodenburg et al. 2001; Piyasirisilp et al. 2000). Similar trends have been observed in Kenya, Tanzania, and South America (e.g., Brazil and Argentina) (Quinn 1996; Essex 1999; Lole et al. 1999; Hu et al. 1996; Van Harmelen et al. 1999). This rapid insurgence of subtype C may be due to a founder effect or to differences in intersubtype fitness. Any difference in ex vivo fitness likely reflects genotypic or phenotypic variations between subtypes. For example, most subtype C isolates appear to have an extra or third NFkB element in the LTR as compared to the two sites found in most subtype, which augment transcription in the presence or absence of HIV Tat protein (van Harmelen et al. 2001; Rodenburg et al. 2001; Hunt and Tiemessen 2000). A subsequent study on HIV protease activity showed increased cleavage of peptide substrates by HIV subtype C vs subtype B protease (Velazquez-Campoy et al. 2001). These phenotypic data suggest that dominance of subtype C in the HIV epidemic could be due to increased replicative capacity of subtype C isolates over other HIV subtype isolates. In addition to these phenotypic differences, subtype C isolates rarely switch from a NSI/R5 phenotype to an SI/CXCR4 phenotype during late disease (Cecilia et al. 2000). However, a recent study has compared the fitness of nine subtype B and six subtype C HIV isolates in a pair-wise competition experiment in PBMC culture (Ball et al. 2003). The relative fitness values were not statistically different in pair-wise competitions involving isolates of the same subtype (intrasubtype B or C competitions). In contrast, almost all of the subtype C isolates were out-competed by the subtype B isolates in the pair-wise competitions. In contrast to previous reports with recombinant HIV clones containing fragments of the subtype C genome, these data suggest that subtype C isolates are likely less fit than subtype B isolates (Ball et al. 2003). The poor relative fitness of subtype C isolates can also fit into the model for subtype C dominance in the epidemic. Decreased fitness is linked to slower disease progression, which could result in increased transmission time (Quiñones-Mateu et al. 2000). A recent study has now ranked the fitness order of most human lentiviruses by performing pair-wise fitness comparisons with isolates of different types (HIV-1 vs HIV-2), HIV-1 groups (M vs O), and subtypes (A, B, C, and D) (Arien et al., personal communication). Regardless of the donor human PBMCs used for these competitions, a definitive fitness rank order was established: HIV-1 group M (subtypes A, B, and D) > group M-subtype C > HIV-2 >> HIV-1 group O. This order in replicative fitness was also observed when virus pairs were added to human dendritic cells then co-cultured with primary, quiescent T cells, which is model for HIV-1 transmission. These results suggest that reduced replicative and transmission fitness may be contributing to the low

118

M. E. Quiñones-Mateu · E. J. Arts

prevalence and limited geographical spread of HIV-2 and group O HIV-1 in the human population. 3.9 Is the Fitness and Pathogenicity of HIV Decreasing? It is apparent that some simian immunodeficiency viruses (SIV) have become highly adapted to their host species over an extended period of time, resulting in asymptomatic, nonpathogenic infections (Whetter et al. 1999). Based on the subtype C fitness results, is it possible that HIV is evolving to an attenuated state? This hypothesis was proposed by Temin more than 10 years ago (Temin 1989). Rapid attenuation of virus was first observed in the myxoma virus infection of Australian rabbits (Ewald 1993). Several introductions of this highly lethal virus resulted in a decrease of nearly half of the rabbit population followed by rapid evolution of a less virulent (fit) virus (Fenner and Kessler 1994). When a virus “jumps” and infects a new species, an increase in

Fig. 6 Schema of the relationship between transmission and pathogenesis fitness and its effect in HIV-1 evolution. More virulent (red) viruses would rapidly kill their hosts, reducing the time for transmission and permitting the expansion of less virulent (blue) viruses that are still capable of establishing new infections. These less pathogenic and more transmissible viruses will then out-compete the more virulent viruses, at the population level

Virus Fitness: Concept, Quantification, and Application

119

the initial virulence can only be supported in minor pandemics (e.g., Ebola pandemics). It is evident that to survive, viruses must continue to propagate in a living host and that a low/attenuated level of pathogenesis represents a trade-off between virulence and transmissibility. Ewald (1993) suggested that more virulent viruses would rapidly kill their hosts, reducing the time for transmission and permitting the expansion of less virulent viruses that were still capable of establishing new infections. In the case of HIV, the continual spread of this lethal virus is a consequence of long asymptomatic but transmissible periods following initial infection. Individuals infected with a more pathogenic (high-replication) strain will progress faster to HIV disease, decreasing the probability of viral transmission. In contrast, attenuated HIV strains (i.e., lower replicative capacity) would in theory delay disease progression and increase the likelihood of transmission (Fig. 6). Based on this theory, changes in HIV pathogenesis in the population may not be easily identified by common correlates of disease progression (e.g., viral load and CD4 cell counts). For example, the nonpathogenic infection of Sooty mangabey monkeys with SIV (Rey-Cuille et al. 1998) results in extremely high viral load, while maintaining transmissibility in the absence of symptomatic disease. Thus, the higher viral loads observed in patients infected with subtypes B or A as compared to subtypes C or D, respectively may be unrelated to the effect of subtype on disease progression. Acknowledgements M.E.Q-M is supported by research grants NIH-HL-67610, NIHDE-015510, NIH-AI-36219 (Center for AIDS Research at Case Western Reserve University), and unrestricted research grants from Agouron Pharmaceuticals, Inc. E.J.A. is supported by research grants NIH-HD-0-3310-502-02 and NIH-AI-AI49170.

References Addo MM, Altfeld M, Rosenberg ES, Eldridge RL, Philips MN, Habeeb K, Khatri A, Brander C, Robbins GK, Mazzara GP, Goulder PJ, Walker BD (2001) The HIV-1 regulatory proteins Tat and Rev are frequently targeted by cytotoxic T lymphocytes derived from HIV-1-infected individuals. Proc Natl Acad Sci U S A 98:1781–1786 Aebischer T, Moskophidis D, Rohrer UH, Zinkernagel RM, Hengartner H (1991) In vitro selection of lymphocytic choriomeningitis virus escape mutants by cytotoxic T lymphocytes. Proc Natl Acad Sci U S A 88:11047–11051 Agrawal AF (2001) Sexual selection and the maintenance of sexual reproduction. Nature (London) 411:692–695 Allen TM, O’Connor DH, Jing P, Dzuris JL, Mothe BR, Vogel TU, Dunphy E, Liebl ME, Emerson C, Wilson N, Kunstman KJ, Wang X, Allison DB, Hughes AL, Desrosiers RC, Altman JD, Wolinsky SM, Sette A, Watkins DI (2000) Tat-specific cytotoxic T lymphocytes select for SIV escape variants during resolution of primary viraemia. Nature (London) 407:386–390

120

M. E. Quiñones-Mateu · E. J. Arts

Altfeld M, Rosenberg ES, Shankarappa R, Mukherjee JS, Hecht FM, Eldridge RL, Addo MM, Poon SH, Phillips MN, Robbins GK, Sax PE, Boswell S, Kahn JO, Brander C, Goulder PJ, Levy JA, Mullins JI, Walker BD (2001) Cellular immune responses and viral diversity in individuals treated during acute and early HIV-1 infection. J Exp Med 193:169–180 Anderson RM, May RM (1986) The invasion, persistence and spread of infectious diseases within animal and plant communities. Philos Trans R Soc Lond B Biol Sci 314:533–570 Anderson RM, May RM (1992) Infectious diseases in humans: dynamics and control. Oxford University Press, Oxford Anderson RM, May RM (1996) The population biology of the interaction between HIV-1 and HIV-2: coexistence or competitive exclusion? AIDS 10:1663–1673 Archer RH, Dykes C, Gerondelis P, Lloyd A, Fay P, Reichman RC, Bambara RA, Demeter LM (2000) Mutants of human immunodeficiency virus type 1 (HIV-1) reverse transcriptase resistant to nonnucleoside reverse transcriptase inhibitors demonstrate altered rates of RNase H cleavage that correlate with HIV-1 replication fitness in cell culture. J Virol 74:8390–8401 Armand-Ugon M, Quiñones-Mateu ME, Gutierrez A, Barretina J, Blanco J, Schols D, De Clercq E, Clotet B, Este JA (2003) Reduced fitness of HIV-1 resistant to CXCR4 antagonists. Antivir Ther 8:1–8 Arts EJ, Wainberg MA (1996) Mechanisms of nucleoside analog antiviral activity and resistance during human immunodeficiency virus reverse transcription. Antimicrob Agents Chemother 40:527–540 Arts EJ, Quiñones-Mateu ME, Albright JL, Marois JP, Hough C, Gu Z, Wainberg MA (1998) 3 -Azido-3 -deoxythymidine (AZT) mediates cross-resistance to nucleoside analogs in the case of AZT-resistant human immunodeficiency virus type 1 variants. J Virol 72:4858–4865 Asjö B, Morfeldt-Manson L, Albert J, Biberfeld G, Karlsson A, Lidman K, Fenyö EM (1986) Replicative capacity of human immunodeficiency virus from patients with varying severity of HIV infection. Lancet ii:660–662 Back NK, Nijhuis M, Keulen W, Boucher CA, Oude Essink BO, van Kuilenburg AB, van Gennip AH, Berkhout B (1996) Reduced replication of 3TC-resistant HIV-1 variants in primary cells due to a processivity defect of the reverse transcriptase enzyme. EMBO J 15:4040–4049 Ball SC, Abraha A, Collins KR, Marozsan AJ, Baird H, Quiñones-Mateu ME, PennNicholson A, Murray M, Richard N, Lobritz M, Zimmerman PA, Kawamura T, Blauvelt A, Arts EJ (2003) Comparing the ex vivo fitness of CCR5-tropic human immunodeficiency virus type 1 isolates of subtypes B and C. J Virol 77:1021–1038 Barbour JD, Hecht FM, Wrin T, Segal MR, Ramstead CA, Liegler TJ, Busch MP, Petropoulos CJ, Hellmann NS, Kahn JO, Grant RM (2004) Higher CD4+ T cell counts associated with low viral pol replication capacity among treatment-naive adults in early HIV-1 infection. J Infect Dis 190:251–256 Barouch DH, Kunstman J, Kuroda MJ, Schmitz JE, Santra S, Peyerl FW, Krivulka GR, Beaudry K, Lifton MA, Gorgone DA, Montefiori DC, Lewis MG, Wolinsky SM, Letvin NL (2002) Eventual AIDS vaccine failure in a rhesus monkey by viral escape from cytotoxic T lymphocytes. Nature (London) 415:335–339

Virus Fitness: Concept, Quantification, and Application

121

Barton NH, Charlesworth B (1998) Why sex and recombination? Science 281:1986– 1990 Bates M, Wrin T, Huang W, Petropoulos C, Hellmann N (2003) Practical applications of viral fitness in clinical practice. Curr Opin Infect Dis 16:11–18 Berger EA, Murphy PM, Farber JM (1999) Chemokine receptors as HIV-1 coreceptors: roles in viral entry, tropism, and disease. Annu Rev Immunol 17:657–700 Bergstrom CT, McElhany P, Real LA (1999) Transmission bottlenecks as determinants of virulence in rapidly evolving pathogens. Proc Natl Acad Sci U S A 96:5095–5100 Berkhout B (1999) HIV-1 evolution under pressure of protease inhibitors: climbing the stairs of viral fitness. J Biomed Sci 6:298–305 Bjorndal A, Deng H, Jansson M, Fiore JR, Colognesi C, Karlsson A, Albert J, Scarlatti G, Littman DR, Fenyo EM (1997) Coreceptor usage of primary human immunodeficiency virus type 1 isolates varies according to biological phenotype. J Virol 71:7478–7487 Blaak H, Brouwer M, Ran LJ, de Wolf F, Schuitemaker H (1998) In vitro replication kinetics of human immunodeficiency virus type 1 (HIV-1) variants in relation to virus load in long-term survivors of HIV-1 infection. J Infect Dis 177:600–610 Blackard JT, Renjifo B, Fawzi W, Hertzmark E, Msamanga G, Mwakagile D, Hunter D, Spiegelman D, Sharghi N, Kagoma C, Essex M (2001) HIV-1 LTR subtype and perinatal transmission. Virology 287:261–265 Blauvelt A, Glushakova S, Margolis LB (2000) HIV-infected human Langerhans cells transmit infection to human lymphoid tissue ex vivo. AIDS 14:647–651 Bleiber G, Munoz M, Ciuffi A, Meylan P, Telenti A (2001) Individual contributions of mutant protease and reverse transcriptase to viral infectivity, replication, and protein maturation of antiretroviral drug-resistant human immunodeficiency virus type 1. J Virol 75:3291–3300 Bonhoeffer S, Chappey C, Parkin NT, Whitcomb JM, Petropoulos CJ (2004) Evidence for positive epistasis in HIV-1. Science 306:1547–1550 Boots M, Sasaki A (1999) ‘Small worlds’ and the evolution of virulence: infection occurs locally and at a distance. Proc R Soc Lond B Biol Sci 266:1933–1938 Borman AM, Paulous S, Clavel F (1996) Resistance of human immunodeficiency virus type 1 to protease inhibitors: selection of resistance mutations in the presence and absence of the drug. J Gen Virol 77:419–426 Braathen LR, Ramirez G, Kunze RO, Gelderblom H (1987) Langerhans cells as primary target cells for HIV infection. Lancet 2:1094 Brambilla A, Turchetto L, Gatti A, Bovolenta C, Veglia F, Santagostino E, Gringeri A, Clementi M, Poli G, Bagnarelli P, Vicenzi E (1999) Defective nef alleles in a cohort of hemophiliacs with progressing and nonprogressing HIV-1 infection. Virology 259:349–368 Bremermann HJ, Pickering J (1983) A game-theoretical model of parasite virulence. J Theor Biol 100:411–426 Buhler B, Lin YC, Morris G, Olson AJ, Wong CH, Richman DD, Elder JH, Torbett BE (2001) Viral evolution in response to the broad-based retroviral protease inhibitor TL-3. J Virol 75:9502–9508 Bull JJ (1994) Virulence. Evolution 48:1423–1437 Burch CL, Chao L (1999) Evolution by small steps and rugged landscapes in the RNA virus phi6. Genetics 151:921–927

122

M. E. Quiñones-Mateu · E. J. Arts

Burch CL, Chao L (2000) Evolvability of an RNA virus is determined by its mutational neighbourhood. Nature (London) 406:625–628 Burch CL, Chao L (2004) Epistasis and its relationship to canalization in the RNA virus phi 6. Genetics 167:559–567 Burch CL, Turner PE, Hanley KA (2003) Patterns of epistasis in RNA viruses: a review of the evidence from vaccine design. J Evol Biol 16:1223–1235 Cao J, McNevin J, Holte S, Fink L, Corey L, McElrath MJ (2003) Comprehensive analysis of human immunodeficiency virus type 1 (HIV-1)-specific gamma interferonsecreting CD8+ T cells in primary HIV-1 infection. J Virol 77:6867–6878 Cao Y, Qin L, Zhang L, Safrit J, Ho DD (1995) Virologic and immunologic characterization of long-term survivors of human immunodeficiency virus type 1 infection. N Eng J Med 332:201–208 Carotenuto P, Looij D, Keldermans L, de Wolf F, Goudsmit J (1998) Neutralizing antibodies are positively associated with CD4+ T-cell counts and T-cell function in long-term AIDS-free infection. AIDS 12:1591–1600 Carrington M, Nelson GW, Martin MP, Kissner T, Vlahov D, Goedert JJ, Kaslow R, Buchbinder S, Hoots K, O’Brien SJ (1999) HLA and HIV-1: heterozygote advantage and B*35-Cw*04 disadvantage. Science 283:1748–1752 Cecilia D, Kulkarni SS, Tripathy SP, Gangakhedkar RR, Paranjape RS, Gadkari DA (2000) Absence of coreceptor switch with disease progression in human immunodeficiency virus infections in India. Virology 271:253–258 Chan DC, Fass D, Berger JM, Kim PS (1997) Core structure of gp41 from the HIV envelope glycoprotein. Cell 89:263–273 Chao L (1990) Fitness of RNA virus decreased by Muller’s ratchet [see comments]. Nature (London) 348:454–455 Cheng-Mayer C, Seto D, Tateno M, Levy JA (1988) Biologic features of HIV-1 that correlate with virulence in the host. Science 240:80–82 Chouquet C, Autran B, Gomard E, Bouley JM, Calvez V, Katlama C, Costagliola D, Riviere Y (2002) Correlation between breadth of memory HIV-specific cytotoxic T cells, viral load and disease progression in HIV infection. AIDS 16:2399–2407 Clarke DK, Duarte EA, Elena SF, Moya A, Domingo E, Holland J (1994) The red queen reigns in the kingdom of RNA viruses. Proc Natl Acad Sci U S A 91:4821–4824 Clavel F, Race E, Mammano F (2000) HIV drug resistance and viral fitness. Adv Pharmacol 49:41–66 Cocchi F, DeVico AL, Garzino-Demo A, Arya SK, Gallo RC, Lusso P (1995) Identification of RANTES, MIP-1 alpha, and MIP-1 beta as the major HIV-suppressive factors produced by CD8+ T cells [see comments]. Science 270:1811–1815 Coffin JM (1986) Genetic variation in AIDS viruses. Cell 46:1–14 Coffin JM (1995) HIV population dynamics in vivo: implications for genetic variation, pathogenesis, and therapy. Science 267:483–489 Collins JA, Thompson MG, Paintsil E, Ricketts M, Gedzior J, Alexander L (2004) Competitive fitness of nevirapine-resistant human immunodeficiency virus type 1 mutants. J Virol 78:603–611 Cooper VS, Reiskind MH, Miller JA, Shelton KA, Walther BA, Elkinton JS, Ewald PW (2002) Timing of transmission and the evolution of virulence of an insect virus. Proc R Soc Lond B Biol Sci 269:1161–1165

Virus Fitness: Concept, Quantification, and Application

123

Cornelissen M, Mulder-Kampinga G, Veenstra J, Zorgdrager F, Kuiken C, Hartman S, Dekker J, van Der HL, Sol C, Coutinho R (1995) Syncytium-inducing (SI) phenotype suppression at seroconversion after intramuscular inoculation of a nonsyncytium-inducing/SI phenotypically mixed human immunodeficiency virus population. J Virol 69:1810–1818 Croteau G, Doyon L, Thibeault D, McKercher G, Pilote L, Lamarre D (1997) Impaired fitness of human immunodeficiency virus type 1 variants with high-level resistance to protease inhibitors. J Virol 71:1089–1096 Crotty S, Cameron CE, Andino R (2001) RNA virus error catastrophe: direct molecular test by using ribavirin. Proc Natl Acad Sci U S A 98:6895–6900 Cuevas JM, Moya A, Sanjuan R (2005) Following the very initial growth of biological RNA viral clones. J Gen Virol 86:435–443 Dalod M, Dupuis M, Deschemin JC, Goujard C, Deveau C, Meyer L, Ngo N, Rouzioux C, Guillet JG, Delfraissy JF, Sinet M, Venet A (1999) Weak anti-HIV CD8(+) T-cell effector activity in HIV primary infection. J Clin Invest 104:1431–1439 Daniel MD, Kirchhoff F, Czajak SC, Sehgal PK, Desrosiers RC (1992) Protective effects of a live attenuated SIV vaccine with a deletion in the nef gene [see comments]. Science 258:1938–1941 De Maria A, Moretta L (2000) HLA-class I-specific inhibitory receptors in HIV-1 infection. Hum Immunol 61:74–81 De Ronde A, van Dooren M, van Der HL, Bouwhuis D, de Rooij E, van Gemen B, de Boer R, Goudsmit J (2001) Establishment of new transmissible and drugsensitive human immunodeficiency virus type 1 wild types due to transmission of nucleoside analogue-resistant virus. J Virol 75:595–602 De Visser JA, Hoekstra RF, van de Ende H (1997) Test of interaction between genetic markers that affect fitness in Aspergillus niger. Evolution 51:1499–1505 Deacon NJ, Tsykin A, Solomon A, Smith K, Ludford-Menting M, Hooker DJ, McPhee DA, Greenway AL, Ellet A, Chatfield C, Lawson VA, Crowe S, Maerz A, Sonza S, Learmont J, Sullivan JS, Cunningham A, Dwyer D, Dowton D, Mills J (1995) Genomic structure of an attenuated quasispecies of HIV-1 from a blood transfusion donor and recipients. Science 270:988–991 Dean M, Carrington M, Winkler C, Huttley GA, Smith MW, Allikmets R, Goedert JJ, Buchbinder SP, Vittinghoff E, Gomperts E, Donfield S, Vlahov D, Kaslow R, Saah A, Rinaldo C, Detels R, O’Brien SJ (1996b) Genetic restriction of HIV-1 infection and progression to AIDS by a deletion allele of the CKR5 structural gene. Hemophilia Growth and Development Study, Multicenter AIDS Cohort Study, Multicenter Hemophilia Cohort Study, San Francisco City Cohort, ALIVE Study. Science 273:1856–1862 Deeks SG (2001) International perspectives on antiretroviral resistance. Nonnucleoside reverse transcriptase inhibitor resistance. J Acquir Immune Defic Syndr 26 Suppl 1:S25–S33 Deeks SG (2003) Treatment of antiretroviral-drug-resistant HIV-1 infection. Lancet 362:2002–2011 Deeks SG, Barbour JD, Martin JN, Swanson MS, Grant RM (2000) Sustained CD4+ T cell response after virologic failure of protease inhibitor-based regimens in patients with human immunodeficiency virus infection. J Infect Dis 181:946–953

124

M. E. Quiñones-Mateu · E. J. Arts

Deeks SG, Wrin T, Liegler T, Hoh R, Hayden M, Barbour JD, Hellmann NS, Petropoulos CJ, McCune JM, Hellerstein MK, Grant RM (2001) Virologic and immunologic consequences of discontinuing combination antiretroviral-drug therapy in HIVinfected patients with detectable viremia. N Engl J Med 344:472–480 Demeter LM, Shafer RW, Meehan PM, Holden-Wiltse J, Fischl MA, Freimuth WW, Para MF, Reichman RC (2000) Delavirdine susceptibilities and associated reverse transcriptase mutations in human immunodeficiency virus type 1 isolates from patients in a phase I/II trial of delavirdine monotherapy (ACTG 260). Antimicrob Agents Chemother 44:794–797 Derdeyn CA, Decker JM, Sfakianos JN, Wu X, O’Brien WA, Ratner L, Kappes JC, Shaw GM, Hunter E (2000) Sensitivity of human immunodeficiency virus type 1 to the fusion inhibitor T-20 is modulated by coreceptor specificity defined by the V3 loop of gp120. J Virol 74:8358–8367 Derdeyn CA, Decker JM, Bibollet-Ruche F, Mokili JL, Muldoon M, Denham SA, Heil ML, Kasolo F, Musonda R, Hahn BH, Shaw GM, Korber BT, Allen S, Hunter E (2004) Envelope-constrained neutralization-sensitive HIV-1 after heterosexual transmission. Science 303:2019–2022 Deval J, White KL, Miller MD, Parkin NT, Courcambeck J, Halfon P, Selmi B, Boretto J, Canard B (2004) Mechanistic basis for reduced viral and enzymatic fitness of HIV-1 reverse transcriptase containing both K65R and M184V mutations. J Biol Chem 279:509–516 Devereux HL, Emery VC, Johnson MA, Loveday C (2001) Replicative fitness in vivo of HIV-1 variants with multiple drug resistance-associated mutations. J Med Virol 65:218–224 Domingo E, Holland JJ (1997) RNA virus mutations and fitness for survival. Annu Rev Microbiol 51:151–178 Domingo E, Escarmís C, Sevilla N, Moya A, Elena SF, Quer J, Novella IS, Holland JJ (1996) Basic concepts in RNA virus evolution. FASEB J 10:859–864 Domingo E, Menendez-Arias L, Quiñones-Mateu ME, Holguin A, Gutierrez-Rivas M, Martinez MA, Quer J, Novella IS, Holland JJ (1997) Viral quasispecies and the problem of vaccine-escape and drug-resistant mutants. Prog Drug Res 48:99–128 Domingo E, Escarmis C, Menendez-Arias L, Holland J (1999) Viral quasispecies and fitness variations. In: Domingo E, Webster R, Holland J (eds) Origin and evolution of viruses. Academic Press, San Diego, pp. 141–161 Doyon L, Croteau G, Thibeault D, Poulin F, Pilote L, Lamarre D (1996) Second locus involved in human immunodeficiency virus type 1 resistance to protease inhibitors. J Virol 70:3763–3769 Duarte E, Clarke D, Moya A, Domingo E, Holland J (1992) Rapid fitness losses in mammalian RNA virus clones due to Muller’s ratchet. Proc Natl Acad Sci U S A 89:6015–6019 Duarte EA, Novella IS, Weaver SC, Domingo E, Wain-Hobson S, Clarke DK, Moya A, Elena SF, de la Torre JC, Holland JJ (1994) RNA virus quasispecies: significance for viral disease and epidemiology. Infect Agents Dis 3:201–214 Dueweke TJ, Pushkarskaya T, Poppe SM, Swaney SM, Zhao JQ, Chen IS, Stevenson M, Tarpley WG (1993) A mutation in reverse transcriptase of bis(heteroaryl)piperazine-resistant human immunodeficiency virus type 1 that confers increased sensitivity to other nonnucleoside inhibitors. Proc Natl Acad Sci U S A 90:4713–4717

Virus Fitness: Concept, Quantification, and Application

125

Dyer WB, Ogg GS, Demoitie MA, Jin X, Geczy AF, Rowland-Jones SL, McMichael AJ, Nixon DF, Sullivan JS (1999) Strong human immunodeficiency virus (HIV)specific cytotoxic T-lymphocyte activity in Sydney Blood Bank Cohort patients infected with nef-defective HIV type 1. J Virol 73:436–443 Eastman PS, Mittler J, Kelso R, Gee C, Boyer E, Kolberg J, Urdea M, Leonard JM, Norbeck DW, Mo H, Markowitz M (1998) Genotypic changes in human immunodeficiency virus type 1 associated with loss of suppression of plasma viral RNA levels in subjects treated with ritonavir (Norvir) monotherapy. J Virol 72:5154– 5164 Eckert DM, Malashkevich VN, Hong LH, Carr PA, Kim PS (1999) Inhibiting HIV-1 entry: discovery of D-peptide inhibitors that target the gp41 coiled-coil pocket. Cell 99:103–115 Elena SF (1999) Little evidence for synergism among deleterious mutations in a nonsegmented RNA virus. J Mol Evol 49:703–707 Elena SF (2001) Evolutionary history conditions the timing of transmission in vesicular stomatitis virus. Infect Genet Evol 1:151–159 Elena SF, Lenski RE (1997) Test of synergistic interactions among deleterious mutations in bacteria. Nature (London) 390:395–398 Elena SF, Gonzalez-Candelas F, Novella IS, Duarte EA, Clarke DK, Domingo E, Holland JJ, Moya A (1996) Evolution of fitness in experimental populations of vesicular stomatitis virus. Genetics 142:673–679 Elena SF, Sanjuan R, Borderia AV, Turner PE (2001) Transmission bottlenecks and the evolution of fitness in rapidly evolving RNA viruses. Infect Genet Evol 1:41–48 Escarmis C, Davila M, Charpentier N, Bracho A, Moya A, Domingo E (1996) Genetic lesions associated with Muller’s ratchet in an RNA virus. J Mol Biol 264:255–267 Escarmis C, Carrillo EC, Ferrer M, Arriaza JF, Lopez N, Tami C, Verdaguer N, Domingo E, Franze-Fernandez MT (1998) Rapid selection in modified BHK21 cells of a foot-and-mouth disease virus variant showing alterations in cell tropism. J Virol 72:10171–10179 Eshel I, Feldman MW (1970) On the evolutionary effect of recombination. Theor Popul Biol 1:88–100 Esnouf R, Ren J, Ross C, Jones Y, Stammers D, Stuart D (1995) Mechanism of inhibition of HIV-1 reverse transcriptase by non-nucleoside inhibitors. Nat Struct Biol 2:303– 308 Essex M (1999) Human immunodeficiency viruses in the developing world. Adv Virus Res 53:71–88 Etemad-Moghadam B, Rhone D, Steenbeke T, Sun Y, Manola J, Gelman R, Fanton JW, Racz P, Tenner-Racz K, Axthelm MK, Letvin NL, Sodroski J (2002) Understanding the basis of CD4(+) T-cell depletion in macaques infected by a simian-human immunodeficiency virus. Vaccine 20:1934–1937 Evans DT, O’Connor DH, Jing P, Dzuris JL, Sidney J, da Silva J, Allen TM, Horton H, Venham JE, Rudersdorf RA, Vogel T, Pauza CD, Bontrop RE, DeMars R, Sette A, Hughes AL, Watkins DI (1999) Virus-specific cytotoxic T-lymphocyte responses select for amino-acid variation in simian immunodeficiency virus Env and Nef. Nat Med 5:1270–1276 Ewald PW (1993) The evolution of virulence. Sci Am 268:86–93 Ewald PW (1994) Evolution of infectious diseases. Oxford University Press, New York

126

M. E. Quiñones-Mateu · E. J. Arts

Fenner F, Kessler H (1994) Evolution of the poxviruses, including the coevolution of virus and host myxomatosis. In: Morse SS (ed) The evolutionary biology of viruses. Raven Press, New York, pp 273–292 Fenyo EM, Albert J, Asjo B (1989) Replicative capacity, cytopathic effect and cell tropism of HIV. AIDS 3 Suppl 1:S5–S12 Fenyo E, Schuitemaker H, Asjö B, McKeating J, Sattentau Q (2000) EC concerted action HIV variability. The history of HIV-1 biological phenotypes. Past, present and future. Review in 1997 Human Retroviruses and AIDS Compendium Fikkert V, Van Maele B, Vercammen J, Hantson A, Van Remoortel B, Michiels M, Gurnari C, Pannecouque C, De Maeyer M, Engelborghs Y, De Clercq E, Debyser Z, Witvrouw M (2003) Development of resistance against diketo derivatives of human immunodeficiency virus type 1 by progressive accumulation of integrase mutations. J Virol 77:11459–11470 Froissart R, Wilke CO, Montville R, Remold SK, Chao L, Turner PE (2004) Co-infection weakens selection against epistatic mutations in RNA viruses. Genetics 168:9–19 Frost SD, Nijhuis M, Schuurman R, Boucher CA, Brown AJ (2000) Evolution of lamivudine resistance in human immunodeficiency virus type 1-infected individuals: the relative roles of drift and selection. J Virol 74:6262–6268 Gallant JE, Gerondelis PZ, Wainberg MA, Shulman NS, Haubrich RH, St Clair M, Lanier ER, Hellmann NS, Richman DD (2003) Nucleoside and nucleotide analogue reverse transcriptase inhibitors: a clinical review of antiretroviral resistance. Antivir Ther 8:489–506 Gandhi RT, Wurcel A, Rosenberg ES, Johnston MN, Hellmann N, Bates M, Hirsch MS, Walker BD (2003) Progressive reversion of human immunodeficiency virus type 1 resistance mutations in vivo after transmission of a multiply drug-resistant virus. Clin Infect Dis 37:1693–1698 Gao F, Bailes E, Robertson DL, Chen Y, Rodenburg CM, Michael SF, Cummins LB, Arthur LO, Peeters M, Shaw GM, Sharp PM, Hahn BH (1999) Origin of HIV-1 in the chimpanzee Pan troglodytes troglodytes [see comments]. Nature (London) 397:436–441 Gao X, Nelson GW, Karacki P, Martin MP, Phair J, Kaslow R, Goedert JJ, Buchbinder S, Hoots K, Vlahov D, O’Brien SJ, Carrington M (2001) Effect of a single amino acid change in MHC class I molecules on the rate of progression to AIDS. N Engl J Med 344:1668–1675 Garcia-Lerma JG, Gerrish PJ, Wright AC, Qari SH, Heneine W (2000) Evidence of a role for the Q151L mutation and the viral background in development of multiple dideoxynucleoside-resistant human immunodeficiency virus type 1. J Virol 74:9339–9346 Garcia-Lerma JG, MacInnes H, Bennett D, Weinstock H, Heneine W (2004) Transmitted human immunodeficiency virus type 1 carrying the D67N or K219Q/E mutation evolves rapidly to zidovudine resistance in vitro and shows a high replicative fitness in the presence of zidovudine. J Virol 78:7545–7552 Geijtenbeek TB, Torensma R, van Vliet SJ, van Duijnhoven GC, Adema GJ, van Kooyk Y, Figdor CG (2000) Identification of DC-SIGN, a novel dendritic cell-specific ICAM3 receptor that supports primary immune responses. Cell 100:575–585

Virus Fitness: Concept, Quantification, and Application

127

Gerondelis P, Archer RH, Palaniappan C, Reichman RC, Fay PJ, Bambara RA, Demeter LM (1999) The P236L delavirdine-resistant human immunodeficiency virus type 1 mutant is replication defective and demonstrates alterations in both RNA 5 -end- and DNA 3 -end-directed RNase H activities. J Virol 73:5803–5813 Gonzalez LM, Brindeiro RM, Aguiar RS, Pereira HS, Abreu CM, Soares MA, Tanuri A (2004) Impact of nelfinavir resistance mutations on in vitro phenotype, fitness, and replication capacity of human immunodeficiency virus type 1 with subtype B and C proteases. Antimicrob Agents Chemother 48:3552–3555 Goudsmit J, De Ronde A, Ho DD, Perelson AS (1996) Human immunodeficiency virus fitness in vivo: calculations based on a single zidovudine resistance mutation at codon 215 of reverse transcriptase. J Virol 70:5662–5664 Goudsmit J, De Ronde A, de Rooij E, de Boer R (1997) Broad spectrum of in vivo fitness of human immunodeficiency virus type 1 subpopulations differing at reverse transcriptase codons 41 and 215. J Virol 71:4479–4484 Goulder PJ, Watkins DI (2004) HIV and SIV CTL escape: implications for vaccine design. Nat Rev Immunol 4:630–640 Goulder PJ, Phillips RE, Colbert RA, McAdam S, Ogg G, Nowak MA, Giangrande P, Luzzi G, Morgan B, Edwards A, McMichael AJ, Rowland-Jones S (1997) Late escape from an immunodominant cytotoxic T-lymphocyte response associated with progression to AIDS. Nat Med 3:212–217 Goulder PJ, Altfeld MA, Rosenberg ES, Nguyen T, Tang Y, Eldridge RL, Addo MM, He S, Mukherjee JS, Phillips MN, Bunce M, Kalams SA, Sekaly RP, Walker BD, Brander C (2001) Substantial differences in specificity of HIV-specific cytotoxic T cells in acute and chronic HIV infection. J Exp Med 193:181–194 Greenberg M, McDanal C, Stanfield-Oakley S, Jin L, Trembaly C, Sista P, Hirsch M, Matthews T (2001) Virus sensitivity to T-20 and T-1249 is independent of coreceptor usage. 8th Conference on Retroviruses and Opportunistic Infections Gu Z, Gao Q, Fang H, Salomon H, Parniak MA, Goldberg E, Cameron J, Wainberg MA (1994) Identification of a mutation at codon 65 in the IKKK motif of reverse transcriptase that encodes human immunodeficiency virus resistance to 2 ,3 -dideoxycytidine and 2 ,3 -dideoxy-3 -thiacytidine. Antimicrob Agents Chemother 38:275–281 Hahn BH, Shaw GM, De Cock KM, Sharp PM (2000) AIDS as a zoonosis: scientific and public health implications. Science 287:607–614 Harrigan PR, Bloor S, Larder BA (1998) Relative replicative fitness of zidovudineresistant human immunodeficiency virus type 1 isolates in vitro. J Virol 72:3773– 3778 Havlir DV, Eastman S, Gamst A, Richman DD (1996) Nevirapine-resistant human immunodeficiency virus: kinetics of replication and estimated prevalence in untreated patients. J Virol 70:7894–7899 Hazuda DJ, Anthony NJ, Gomez RP, Jolly SM, Wai JS, Zhuang L, Fisher TE, Embrey M, Guare JP Jr, Egbertson MS, Vacca JP, Huff JR, Felock PJ, Witmer MV, Stillmock KA, Danovich R, Grobler J, Miller MD, Espeseth AS, Jin L, Chen IW, Lin JH, Kassahun K, Ellis JD, Wong BK, Xu W, Pearson PG, Schleif WA, Cortese R, Emini E, Summa V, Holloway MK, Young SD (2004a) A naphthyridine carboxamide provides evidence for discordant resistance between mechanistically identical inhibitors of HIV-1 integrase. Proc Natl Acad Sci U S A 101:11233–11238

128

M. E. Quiñones-Mateu · E. J. Arts

Hazuda DJ, Young SD, Guare JP, Anthony NJ, Gomez RP, Wai JS, Vacca JP, Handt L, Motzel SL, Klein HJ, Dornadula G, Danovich RM, Witmer MV, Wilson KA, Tussey L, Schleif WA, Gabryelski LS, Jin L, Miller MD, Casimiro DR, Emini EA, Shiver JW (2004b) Integrase inhibitors and cellular immunity suppress retroviral replication in rhesus macaques. Science 305:528–532 Hertogs K, de Bethune MP, Miller V, Ivens T, Schel P, Van Cauwenberge A, Van Den EC, Van Gerwen V, Azijn H, Van Houtte M, Peeters F, Staszewski S, Conant M, Bloor S, Kemp S, Larder B, Pauwels R (1998) A rapid method for simultaneous detection of phenotypic resistance to inhibitors of protease and reverse transcriptase in recombinant human immunodeficiency virus type 1 isolates from patients treated with antiretroviral drugs. Antimicrob Agents Chemother 42:269–276 Hirsch MS, Conway B, D’aquila RT, Johnson VA, Brun-Vezinet F, Clotet B, Demeter LM, Hammer SM, Jacobsen DM, Kuritzkes DR, Loveday C, Mellors JW, Vella S, Richman DD (1998) Antiretroviral drug resistance testing in adults with HIV infection: implications for clinical management. International AIDS Society–USA Panel. JAMA 279:1984–1991 Hirsch MS, Brun-Vezinet F, D’aquila RT, Hammer SM, Johnson VA, Kuritzkes DR, Loveday C, Mellors JW, Clotet B, Conway B, Demeter LM, Vella S, Jacobsen DM, Richman DD (2000) Antiretroviral drug resistance testing in adult HIV-1 infection: recommendations of an International AIDS Society-USA Panel. JAMA 283:2417–2426 Hirsch VM (2004) What can natural infection of African monkeys with simian immunodeficiency virus tell us about the pathogenesis of AIDS? AIDS Rev 6:40–53 Ho DD, Toyoshima T, Mo H, Kempf DJ, Norbeck D, Chen CM, Wideburg NE, Burt SK, Erickson JW, Singh MK (1994) Characterization of human immunodeficiency virus type 1 variants with increased resistance to a C2-symmetric protease inhibitor. J Virol 68:2016–2020 Ho DD, Neumann AU, Perelson AS, Chen W, Leonard JM, Markowitz M (1995) Rapid turnover of plasma virions and CD4 lymphocytes in HIV-1 infection. Nature (London) 373:123–126 Holland JJ, de la Torre JC, Clarke DK, Duarte E (1991) Quantitation of relative fitness and great adaptability of clonal populations of RNA viruses. J Virol 65:2960–2967 Holmes EC (2003) Patterns of intra- and interhost nonsynonymous variation reveal strong purifying selection in dengue virus. J Virol 77:11296–11298 Holterman L, Niphuis H, Koornstra W, Dubbes R, ten Haaft P, Heeney JL (2000) The rate of progression to AIDS is independent of virus dose in simian immunodeficiency virus-infected macaques. J Gen Virol 81:1719–1726 Hu DJ, Dondero TJ, Rayfield MA, George R, Schochetman G, Jaffe HW, Luo C, Kalish ML, Weniger BG, Pau C, Schable CA, Curran JW (1996) The emerging genetic diversity of HIV. The importance of global surveillance for diagnostics, research, and prevention. JAMA 275:210–216 Hunt G, Tiemessen CT (2000) Occurrence of additional NF-kappaB-binding motifs in the long terminal repeat region of South African HIV type 1 subtype C isolates. AIDS Res Hum Retroviruses 16:305–306 Iglesias-Ussel MD, Casado C, Yuste E, Olivares I, Lopez-Galindez C (2002) In vitro analysis of HIV-1 resistance to nevirapine and fitness determination of resistant variants. J Gen Virol 83:93–101

Virus Fitness: Concept, Quantification, and Application

129

Imamichi T, Berg SC, Imamichi H, Lopez JC, Metcalf JA, Falloon J, Lane HC (2000) Relative replication fitness of a high-level 3 -azido-3 -deoxythymidine-resistant variant of human immunodeficiency virus type 1 possessing an amino acid deletion at codon 67 and a novel substitution (Thr→Gly) at codon 69. J Virol 74:10958–10964 Imamichi T, Murphy MA, Imamichi H, Lane HC (2001) Amino acid deletion at codon 67 and Thr-to-Gly change at codon 69 of human immunodeficiency virus type 1 reverse transcriptase confer novel drug resistance profiles. J Virol 75:3988–3992 Joag SV (2000) Primate models of AIDS. Microbes Infect 2:223–229 Kalish ML, Baldwin A, Raktham S, Wasi C, Luo C-C, Schochetman G, Mastro TD, Young N, Vanichseni S, Rübsamen-Waigmann H, von Briesen H, Mullins JI, Delwart E, Herring B, Esparza J, Heyward WL, Osmanov S (1995) The evolving molecular epidemiology of HIV-1 envelope subtypes in injecting drug users in Bangkok, Thailand: implications for HIV vaccine trials. AIDS 9:851–857 Kanki PJ, Hamel DJ, Sankale JL, Hsieh C, Thior I, Barin F, Woodcock SA, GueyeNdiaye A, Zhang E, Montano M, Siby T, Marlink R, NDoye I, Essex ME, Mboup S (1999) Human immunodeficiency virus type 1 subtypes differ in disease progression. J Infect Dis 179:68–73 Kaslow RA, Carrington M, Apple R, Park L, Munoz A, Saah AJ, Goedert JJ, Winkler C, O’Brien SJ, Rinaldo C, Detels R, Blattner W, Phair J, Erlich H, Mann DL (1996) Influence of combinations of human major histocompatibility complex genes on the course of HIV-1 infection. Nat Med 2:405–411 Kellam P, Larder BA (1994) Recombinant virus assay: a rapid, phenotypic assay for assessment of drug susceptibility of human immunodeficiency virus type 1 isolates. Antimicrob Agents Chemother 38:23–30 Kelleher AD, Long C, Holmes EC, Allen RL, Wilson J, Conlon C, Workman C, Shaunak S, Olson K, Goulder P, Brander C, Ogg G, Sullivan JS, Dyer W, Jones I, McMichael AJ, Rowland-Jones S, Phillips RE (2001) Clustered mutations in HIV-1 gag are consistently required for escape from HLA-B27-restricted cytotoxic T lymphocyte responses. J Exp Med 193:375–386 Keulen W, Nijhuis M, Schuurman R, Berkhout B, Boucher C (1997) Reverse transcriptase fidelity and HIV-1 variation. Science 275:229, author reply 230–231 Kilby JM, Hopkins S, Venetta TM, DiMassimo B, Cloud GA, Lee JY, Alldredge L, Hunter E, Lambert D, Bolognesi D, Matthews T, Johnson MR, Nowak MA, Shaw GM, Saag MS (1998) Potent suppression of HIV-1 replication in humans by T-20, a peptide inhibitor of gp41-mediated virus entry. Nat Med 4:1302–1307 Kimata JT, Kuller L, Anderson DB, Dailey P, Overbaugh J (1999) Emerging cytopathic and antigenic simian immunodeficiency virus variants influence AIDS progression. Nat Med 5:535–541 Kirchhoff F, Greenough TC, Brettler DB, Sullivan JL, Desrosiers RC (1995) Brief report: absence of intact nef sequences in a long-term survivor with nonprogressive HIV-1 infection. N Engl J Med 332:228–232 Knipe DM, Howley PM (2001) Fields virology, 4th edn. Lippincott Williams and Wilkins, Philadelphia Kondrashov AS (1988) Deleterious mutations and the evolution of sexual reproduction. Nature (London) 336:435–440 Kondrashov AS (1994) Muller’s ratchet under epistatic selection. Genetics 136:1469– 1473

130

M. E. Quiñones-Mateu · E. J. Arts

Kondrashov AS, Crow JF (1991) Haploidy or diploidy: which is better? Nature (London) 351:314–315 Korber B, Hahn B, Foley B, Mellors J, Leitner T, Myers G, McCutchan F, Kuiken C (1997a) Human retroviruses and AIDS 1997. Theoretical Biology and Biophysics Group, Los Alamos National Laboratory, Los Alamos, NM Korber B, Hahn B, Foley B, Mellors JW, Leitner T, Myers G, McCutchan F, Kuiken C (1997b) Human retroviruses and AIDS. A compilation and analysis of nucleic acid and amino acid sequences. Theoretical Biology and Byophysics Group, Los Alamos National Laboratory, Los Alamos, NM Korber B, Muldoon M, Theiler J, Gao F, Gupta R, Lapedes A, Hahn BH, Wolinsky S, Bhattacharya T (2000) Timing the ancestor of the HIV-1 pandemic strains [see comments]. Science 288:1789–1796 Kosalaraksa P, Kavlick MF, Maroun V, Le R, Mitsuya H (1999) Comparative fitness of multi-dideoxynucleoside-resistant human immunodeficiency virus type 1 (HIV-1) in an in vitro competitive HIV-1 replication assay. J Virol 73:5356–5363 Kozal MJ, Shah N, Shen N, Yang R, Fucini R, Merigan TC, Richman DD, Morris D, Hubbell E, Chee M, Gingeras TR (1996) Extensive polymorphisms observed in HIV-1 clade B protease gene using high-density oligonucleotide arrays. Nat Med 2:753–759 Kuiken CL, Foley B, Hahn B, Korber B, McCutchan F, Marx P, Mellors J, Mullins JI, Sodroski J, Wolinsky S (1999) Human retroviruses and AIDS 1999: a compilation and analysis of nucleic acid and amino acid sequences. Theoretical Biology and Biophysics Group, Los Alamos National Laboratory, Los Alamos, NM Kwa D, Vingerhoed J, Boeser B, Schuitemaker H (2003) Increased in vitro cytopathicity of CC chemokine receptor 5-restricted human immunodeficiency virus type 1 primary isolates correlates with a progressive clinical course of infection. J Infect Dis 187:1397–1403 Larder B (2001) Mechanisms of HIV-1 drug resistance. AIDS 15 Suppl 5:S27–S34 Larder BA, Kemp SD, Harrigan PR (1995) Potential mechanism for sustained antiretroviral efficacy of AZT-3TC combination therapy. Science 269:696–699 Lazaro E, Escarmis C, Perez-Mercader J, Manrubia SC, Domingo E (2003) Resistance of virus to extinction on bottleneck passages: study of a decaying and fluctuating pattern of fitness loss. Proc Natl Acad Sci U S A 100:10830–10835 Lech WJ, Wang G, Li Yang Y, Chee Y, Dorman K, McCrae D, Lazzeroni LC, Erickson JW, Sinsheimer JS, Kaplan AH (1996) In vivo sequence diversity of the protease of human immunodeficiency virus type 1: presence of protease inhibitor-resistant variants in untreated subjects. J Virol 70:2038–2043 Lee DJ, Robinson WE Jr (2004) Human immunodeficiency virus type 1 (HIV-1) integrase: resistance to diketo acid integrase inhibitors impairs HIV-1 replication and integration and confers cross-resistance to L-chicoric acid. J Virol 78:5835–5847 Leigh-Brown AJ (1997) Analysis of HIV-1 env gene sequences reveals evidence for a low effective number in the viral population. Proc Natl Acad Sci U S A 94:1862–1865 Leigh Brown AJ, Frost SD, Mathews WC, Dawson K, Hellmann NS, Daar ES, Richman DD, Little SJ (2003) Transmission fitness of drug-resistant human immunodeficiency virus and the prevalence of resistance in the antiretroviral-treated population. J Infect Dis 187:683–686

Virus Fitness: Concept, Quantification, and Application

131

Lenski RE, May RM (1994) The evolution of virulence in parasites and pathogens: reconciliation between two competing hypotheses. J Theor Biol 169:253–265 Leroux C, Cadore JL, Montelaro RC (2004) Equine infectious anemia virus (EIAV): what has HIV’s country cousin got to tell us? Vet Res 35:485–512 Little SJ, Daar ES, D’aquila RT, Keiser PH, Connick E, Whitcomb JM, Hellmann NS, Petropoulos CJ, Sutton L, Pitt JA, Rosenberg ES, Koup RA, Walker BD, Richman DD (1999) Reduced antiretroviral drug susceptibility among patients with primary HIV infection [see comments]. JAMA 282:1142–1149 Little SJ, Holte S, Routy JP, Daar ES, Markowitz M, Collier AC, Koup RA, Mellors JW, Connick E, Conway B, Kilby M, Wang L, Whitcomb JM, Hellmann NS, Richman DD (2002) Antiretroviral-drug resistance among patients recently infected with HIV. N Engl J Med 347:385–394 Lole KS, Bollinger RC, Paranjape RS, Gadkari D, Kulkarni SS, Novak NG, Ingersoll R, Sheppard HW, Ray SC (1999) Full-length human immunodeficiency virus type 1 genomes from subtype C-infected seroconverters in India, with evidence of intersubtype recombination. J Virol 73:152–160 Loveday C (2001) International perspectives on antiretroviral resistance. Nucleoside reverse transcriptase inhibitor resistance. J Acquir Immune Defic Syndr 26 Suppl 1:S10–S24 Loveday C, Hill A (1995) Prediction of progression to AIDS with serum HIV-1 RNA and CD4 count. Lancet 345:790–791 Lu J, Kuritzkes DR (2001) A novel recombinant marker virus assay for comparing the relative fitness of hiv-1 reverse transcriptase variants. J Acquir Immune Defic Syndr 27:7–13 Lu J, Sista P, Giguel F, Greenberg M, Kuritzkes DR (2004) Relative replicative fitness of human immunodeficiency virus type 1 mutants resistant to enfuvirtide (T-20). J Virol 78:4628–4637 Lukashov V, Goudsmit J (1997) Founder virus population related to route of virus transmission: a determinant of intrahost human immunodeficiency virus type 1 evolution? J Virol 71:2023–2030 Maeda Y, Venzon DJ, Mitsuya H (1998) Altered drug sensitivity, fitness, and evolution of human immunodeficiency virus type 1 with pol gene mutations conferring multi-dideoxynucleoside resistance. J Infect Dis 177:1207–1213 Mahanty S, Bray M (2004) Pathogenesis of filoviral haemorrhagic fevers. Lancet Infect Dis 4:487–498 Mammano F, Petit C, Clavel F (1998) Resistance-associated loss of viral fitness in human immunodeficiency virus type 1: phenotypic analysis of protease and gag coevolution in protease inhibitor-treated patients. J Virol 72:7632–7637 Mansky LM, Temin HM (1995) Lower in vivo mutation rate of human immunodeficiency virus type 1 than that predicted from the fidelity of purified reverse transcriptase. J Virol 69:5087–5094 Margot NA, Isaacson E, McGowan I, Cheng AK, Schooley RT, Miller MD (2002) Genotypic and phenotypic analyses of HIV-1 in antiretroviral-experienced patients treated with tenofovir DF. AIDS 16:1227–1235 Marozsan AJ, Moore D, Lobritz M, Fraundorf E, Abraha A, Reeves JD, Arts EJ (2005) Differences in the fitness of two diverse wild-type HIV-1 isolates are related to the efficiency of cell binding and entry. J Virol 79:7121–7124

132

M. E. Quiñones-Mateu · E. J. Arts

Martinez-Picado J, Savara AV, Shi L, Sutton L, D’aquila RT (2000) Fitness of human immunodeficiency virus type 1 protease inhibitor-selected single mutants. Virology 275:318–322 Martinez-Picado J, Savara AV, Sutton L, D’aquila RT (1999) Replicative fitness of protease inhibitor-resistant mutants of human immunodeficiency virus type 1. J Virol 73:3744–3752 Martinez-Picado J, Morales-Lopetegui K, Wrin T, Garcia-Prado J, Frost SD, Petropoulos CJ, Clotet B, Ruiz L (2001) Selection of the M184V mutation during repetitive cycles of structured antiretroviral treatment interruptions. 5th International Workshop on HIV Drug Resistance and Treatment Strategies, Scottsdale, Arizona June 2001, Abstract 36 McDonald D, Wu L, Bohks SM, KewalRamani VN, Unutmaz D, Hope TJ (2003) Recruitment of HIV and its receptors to dendritic cell–T cell junctions. Science 300:1295–1297 Mellors JW, Rinaldo CR Jr, Gupta P, White RM, Todd JA, Kingsley LA (1996) Prognosis in HIV-1 infection predicted by the quantity of virus in plasma [see comments] [published erratum appears in Science 1997 Jan 3; 275(5296):14]. Science 272:1167–1170 Menendez-Arias L, Este JA (2004) HIV-resistance to viral entry inhibitors. Curr Pharm Des 10:1845–1860 Menzo S, Castagna A, Monachetti A, Hasson H, Danise A, Carini E, Bagnarelli P, Lazzarin A, Clementi M (2004) Genotype and phenotype patterns of human immunodeficiency virus type 1 resistance to enfuvirtide during long-term treatment. Antimicrob Agents Chemother 48:3253–3259 Messenger SL, Molineux IJ, Bull JJ (1999) Virulence evolution in a virus obeys a tradeoff. Proc R Soc Lond B Biol Sci 266:397–404 Migueles SA, Sabbaghian MS, Shupert WL, Bettinotti MP, Marincola FM, Martino L, Hallahan CW, Selig SM, Schwartz D, Sullivan J, Connors M (2000) HLA B*5701 is highly associated with restriction of virus replication in a subgroup of HIVinfected long term nonprogressors. Proc Natl Acad Sci U S A 97:2709–2714 Miller V (2001) International perspectives on antiretroviral resistance. Resistance to protease inhibitors. J Acquir Immune Defic Syndr 26 Suppl 1:S34–S50 Miralles R, Moya A, Elena SF (1997) Is group selection a factor modulating the virulence of RNA viruses? Genet Res 69:165–172 Miralles R, Gerrish PJ, Moya A, Elena SF (1999a) Clonal interference and the evolution of RNA viruses. Science 285:1745–1747 Miralles R, Moya A, Elena SF (1999b) Effect of population patchiness and migration rates on the adaptation and divergence of vesicular stomatitis virus quasispecies populations. J Gen Virol 80:2051–2059 Miralles R, Moya A, Elena SF (2000) Diminishing returns of population size in the rate of RNA virus adaptation. J Virol 74:3566–3571 Montefiori DC, Pantaleo G, Fink LM, Zhou JT, Zhou JY, Bilska M, Miralles GD, Fauci AS (1996) Neutralizing and infection-enhancing antibody responses to human immunodeficiency virus type 1 in long-term nonprogressors. J Infect Dis 173:60–67 Moore FB, Rozen DE, Lenski RE (2000) Pervasive compensatory adaptation in Escherichia coli. Proc R Soc Lond B Biol Sci 267:515–522

Virus Fitness: Concept, Quantification, and Application

133

Moya A, Holmes EC, Gonzalez-Candelas F (2004) The population genetics and evolutionary epidemiology of RNA viruses. Nat Rev Microbiol 2:279–288 Myint L, Matsuda M, Matsuda Z, Yokomaku Y, Chiba T, Okano A, Yamada K, Sugiura W (2004) Gag non-cleavage site mutations contribute to full recovery of viral fitness in protease inhibitor-resistant human immunodeficiency virus type 1. Antimicrob Agents Chemother 48:444–452 Nelson JA, Baribaud F, Edwards T, Swanstrom R (2000) Patterns of changes in human immunodeficiency virus type 1 V3 sequence populations late in infection. J Virol 74:8494–8501 Neumann T, Hagmann I, Lohrengel S, Heil ML, Derdeyn CA, Krausslich HG, Dittmar MT (2005) T20-insensitive HIV-1 from naive patients exhibits high viral fitness in a novel dual-color competition assay on primary cells. Virology 333:251–262 Newstein MC, Desrosiers RC (2001) Effects of reverse-transcriptase mutations M184V and E89G on simian immunodeficiency virus in Rhesus monkeys. J Infect Dis 184:1262–1267 Nijhuis M, Boucher CA, Schipper P, Leitner T, Schuurman R, Albert J (1998) Stochastic processes strongly influence HIV-1 evolution during suboptimal proteaseinhibitor therapy. Proc Natl Acad Sci U S A 95:14441–14446 Nijhuis M, Schuurman R, de Jong D, Erickson J, Gustchina E, Albert J, Schipper P, Gulnik S, Boucher CA (1999) Increased fitness of drug resistant HIV-1 protease as a result of acquisition of compensatory mutations during suboptimal therapy. AIDS 13:2349–2359 Nijhuis M, Deeks S, Boucher C (2001) Implications of antiretroviral resistance on viral fitness. Curr Opin Infect Dis 14:23–28 Novella IS, Duarte EA, Elena SF, Moya A, Domingo E, Holland JJ (1995a) Exponential increases of RNA virus fitness during large population transmissions. Proc Natl Acad Sci U S A 92:5841–5844 Novella IS, Duarte EA, Elena SF, Moya A, Domingo E, Holland JJ (1995b) Exponential increases of RNA virus fitness during large population transmissions. Proc Natl Acad Sci U S A 92:5841–5844 Novella IS, Ebendick-Corp (2004) Molecular basis of fitness loss and fitness recovery in vesicular stomatitis virus. J Mol Biol 342:1423–1430 Novella IS, Elena SF, Moya A, Domingo E, Holland JJ (1995c) Size of genetic bottlenecks leading to virus fitness loss is determined by mean initial population fitness. J Virol 69:2869–2872 Novella IS, Ball LA, Wertz GW (2004) Fitness analyses of vesicular stomatitis strains with rearranged genomes reveal replicative disadvantages. J Virol 78:9837–9841 Nowak MA, Anderson RM, McLean AR, Wolfs TFW, Goudsmit J, May RM (1991) Antigenic diversity thresholds and the development of AIDS. Science 254:963–969 O’Brien SJ, Moore JP (2000) The effect of genetic variation in chemokines and their receptors on HIV transmission and progression to AIDS. Immunol Rev 177:99–111 O’Connor DH, Allen TM, Vogel TU, Jing P, DeSouza IP, Dodds E, Dunphy EJ, Melsaether C, Mothe B, Yamamoto H, Horton H, Wilson N, Hughes AL, Watkins DI (2002) Acute phase cytotoxic T lymphocyte escape is a hallmark of simian immunodeficiency virus infection. Nat Med 8:493–499

134

M. E. Quiñones-Mateu · E. J. Arts

Overbaugh J, Kreiss J, Poss M, Lewis P, Mostad S, John G, Nduati R, Mbori-Ngacha D, Martin H Jr, Richardson B, Jackson S, Neilson J, Long EM, Panteleeff D, Welch M, Rakwar J, Jackson D, Chohan B, Lavreys L, Mandaliya K, Ndinya-Achola J, Bwayo J (1999) Studies of human immunodeficiency virus type 1 mucosal viral shedding and transmission in Kenya. J Infect Dis 179 [Suppl 3]:S401–S404 Pantaleo G, Menzo S, Vaccarezza M, Graziosi C, Cohen OJ, Demarest JF, Montefiori D, Orenstein JM, Fox C, Schrager LK, Margolick JB, Buchbinder S, Giorgi JV, Fauci AS (1995) Studies in subjects with long-term nonprogressive human immunodeficiency virus infection. N Eng J Med 332:209–216 Park J, Morrow CD (1993) Mutations in the protease gene of human immunodeficiency virus type 1 affect release and stability of virus particles. Virology 194:843–850 Partridge L, Barton NH (1993) Optimality, mutation and the evolution of ageing. Nature (London) 362:305–311 Peeters M (2000) Recombinant HIV sequences: their role in the global epidemic. In: Kuiken C, Foley B, Hahn B, Korber B, McCutchan F, Marx P, Mellors J, Mullins JI, Sodroski J, Wolinsky S (eds) Human retroviruses and AIDS 2000. Theoretical Biology and Biophysics Group, Los Alamos National Laboratory, Los Alamos, NM, pp I-39–I-54 Perelson AS, Neumann AU, Markowitz M, Leonard JM, Ho DD (1996) HIV-1 dynamics in vivo: virion clearance rate, infected cell life-span, and viral generation time. Science 271:1582–1586 Petropoulos CJ, Parkin NT, Limoli KL, Lie YS, Wrin T, Huang W, Tian H, Smith D, Winslow GA, Capon DJ, Whitcomb JM (2000) A novel phenotypic drug susceptibility assay for human immunodeficiency virus type 1. Antimicrob Agents Chemother 44:920–928 Phillips RE, Rowland-Jones S, Nixon DF, Gotch FM, Edwards JP, Ogunlesi AO, Elvin JG, Rothbard JA, Bangham CR, Rizza CR (1991) Human immunodeficiency virus genetic variation that can escape cytotoxic T cell recognition. Nature (London) 354:453–459 Picchio GR, Valdez H, Sabbe R, Landay AL, Kuritzkes DR, Lederman MM, Mosier DE (2000) Altered viral fitness of HIV-1 following failure of protease inhibitor- based therapy. J Acquir Immune Defic Syndr 25:289–295 Piyasirisilp S, McCutchan FE, Carr JK, Sanders-Buell E, Liu W, Chen J, Wagner R, Wolf H, Shao Y, Lai S, Beyrer C, Yu XF (2000) A recent outbreak of human immunodeficiency virus type 1 infection in southern China was initiated by two highly homogeneous, geographically separated strains, circulating recombinant form AE and a novel BC recombinant. J Virol 74:11286–11295 Pontesilli O, Klein MR, Kerkhof-Garde SR, Pakker NG, de Wolf F, Schuitemaker H, Miedema F (1998) Longitudinal analysis of human immunodeficiency virus type 1-specific cytotoxic T lymphocyte responses: a predominant gag-specific response is associated with nonprogressive infection. J Infect Dis 178:1008–1018 Prado JG, Wrin T, Beauchaine J, Ruiz L, Petropoulos CJ, Frost SD, Clotet B, D’aquila RT, Martinez-Picado J (2002) Amprenavir-resistant HIV-1 exhibits lopinavir crossresistance and reduced replication capacity. AIDS 16:1009–1017

Virus Fitness: Concept, Quantification, and Application

135

Prado JG, Franco S, Matamoros T, Ruiz L, Clotet B, Menendez-Arias L, Martinez MA, Martinez-Picado J (2004) Relative replication fitness of multi-nucleoside analogue-resistant HIV-1 strains bearing a dipeptide insertion in the fingers subdomain of the reverse transcriptase and mutations at codons 67 and 215. Virology 326:103–112 Precious HM, Gunthard HF, Wong JK, D’aquila RT, Johnson VA, Kuritzkes DR, Richman DD, Leigh Brown AJ (2000) Multiple sites in HIV-1 reverse transcriptase associated with virological response to combination therapy. AIDS 14:31–36 Preston BD, Poiesz BJ, Loeb LA (1988) Fidelity of HIV-1 reverse transcriptase. Science 242:1168–1171 Quinn TC (1996) Global burden of the HIV pandemic. Lancet 348:99–106 Quiñones-Mateu ME, Arts EJ (1999) Recombination in HIV-1: update and implications. AIDS Reviews 1:89–100 Quiñones-Mateu ME, Arts EJ (2001) HIV-1 fitness: implications for drug resistance, disease progression, and global epidemic evolution. In: Kuiken C, Foley B, Hahn B, Marx P, McCutchan F, Mellors J, Wolinsky S, Korber B (eds) HIV Sequence Compendium 2001. Theoretical Biology and Biophysics Group, Los Alamos National Laboratory, Los Alamos, NM, pp 134–170 Quiñones-Mateu ME, Arts EJ (2002) Fitness of drug resistant HIV-1: methodology and clinical implications. Drug Resist Updat 5:224–233 Quiñones-Mateu ME, Ball SC, Marozsan AJ, Torre VS, Albright JL, Vanham G, van der Groen G, Colebunders RL, Arts EJ (2000) A dual infection/competition assay shows a correlation between ex vivo human immunodeficiency virus type 1 fitness and disease progression. J Virol 74:9222–9233 Quiñones-Mateu ME, Tadele M, Parera M, Mas A, Weber J, Rangel HR, Chakraborty B, Clotet B, Domingo E, Menendez-Arias L, Martinez MA (2002) Insertions in the reverse transcriptase increase both drug resistance and viral fitness in a human immunodeficiency virus type 1 isolate harboring the multi-nucleoside reverse transcriptase inhibitor resistance 69 insertion complex mutation. J Virol 76:10546–10552 Rangel HR, Weber J, Chakraborty B, Gutierrez A, Marotta ML, Mirza M, Kiser P, Martinez MA, Este JA, Quiñones-Mateu ME (2003) Role of the human immunodeficiency virus type 1 envelope gene in viral fitness. J Virol 77:9069–9073 Resch W, Parkin N, Stuelke EL, Watkins T, Swanstrom R (2001) A multiple-site-specific heteroduplex tracking assay as a tool for the study of viral population dynamics. Proc Natl Acad Sci U S A 98:176–181 Rey-Cuille MA, Berthier JL, Bomsel-Demontoy MC, Chaduc Y, Montagnier L, Hovanessian AG, Chakrabarti LA (1998) Simian immunodeficiency virus replicates to high levels in sooty mangabeys without inducing disease. J Virol 72:3872–3886 Richardson BA, Mbori-Ngacha D, Lavreys L, John-Stewart GC, Nduati R, Panteleeff DD, Emery S, Kreiss JK, Overbaugh J (2003) Comparison of human immunodeficiency virus type 1 viral loads in Kenyan women, men, and infants during primary and early infection. J Virol 77:7120–7123 Richman DD, Wrin T, Little SJ, Petropoulos CJ (2003) Rapid evolution of the neutralizing antibody response to HIV type 1 infection. Proc Natl Acad Sci U S A 100:4144–4149

136

M. E. Quiñones-Mateu · E. J. Arts

Robinson LH, Myers RE, Snowden BW, Tisdale M, Blair ED (2000) HIV type 1 protease cleavage site mutations and viral fitness: implications for drug susceptibility phenotyping assays. AIDS Res Hum Retroviruses 16:1149–1156 Rodenburg CM, Li Y, Trask SA, Chen Y, Decker J, Robertson DL, Kalish ML, Shaw GM, Allen S, Hahn BH, Gao F (2001) Near full-length clones and reference sequences for subtype C isolates of HIV type 1 from three different continents. AIDS Res Hum Retroviruses 17:161–168 Root MJ, Kay MS, Kim PS (2001) Protein design of an HIV-1 entry inhibitor. Science 291:884–888 Rose RE, Gong YF, Greytok JA, Bechtold CM, Terry BJ, Robinson BS, Alam M, Colonno RJ, Lin PF (1996) Human immunodeficiency virus type 1 viral background plays a major role in development of resistance to protease inhibitors. Proc Natl Acad Sci U S A 93:1648–1653 Rosenberg ES, Billingsley JM, Caliendo AM, Boswell SL, Sax PE, Kalams SA, Walker BD (1997) Vigorous HIV-1-specific CD4+ T cell responses associated with control of viremia [see comments]. Science 278:1447–1450 Rouzine IM, Coffin JM (1999) Linkage disequilibrium test implies a large effective population number for HIV in vivo. Proc Natl Acad Sci U S A 96:10758–10763 Ruiz-Jarabo CM, Miller E, Gomez-Mariano G, Domingo E (2003) Synchronous loss of quasispecies memory in parallel viral lineages: a deterministic feature of viral quasispecies. J Mol Biol 333:553–563 Sanjuan R, Moya A, Elena SF (2004) The contribution of epistasis to the architecture of fitness in an RNA virus. Proc Natl Acad Sci U S A 101:15376–15379 Santiago ML, Santiago EG, Hafalla JC, Manalo MA, Orantia L, Cajimat MN, Martin C, Cuaresma C, Dominguez CE, Borromeo ME, De Groot AS, Flanigan TP, Carpenter CC, Mayer KH, Ramirez BL (1998) Molecular epidemiology of HIV-1 infection in the Philippines, 1985 to 1997: transmission of subtypes B and E and potential emergence of subtypes C and F. J Acquir Immune Defic Syndr Hum Retrovirol 18:260–269 Santra S, Schmitz JE, Kuroda MJ, Lifton MA, Nickerson CE, Lord CI, Pal R, Franchini G, Letvin NL (2002) Recombinant canarypox vaccine-elicited CTL specific for dominant and subdominant simian immunodeficiency virus epitopes in rhesus monkeys. J Immunol 168:1847–1853 Schuitemaker H, Koot M, Kootstra NA, Dercksen MW, de Goede RE, van Steenwijk RP, Lange JM, Schattenkerk JK, Miedema F, Tersmette M (1992) Biological phenotype of human immunodeficiency virus type 1 clones at different stages of infection: progression of disease is associated with a shift from monocytotropic to T-celltropic virus population. J Virol 66:1354–1360 Seo TK, Thorne JL, Hasegawa M, Kishino H (2002) Estimation of effective population size of HIV-1 within a host: a pseudomaximum-likelihood approach. Genetics 160:1283–1293 Shafer RW, Dupnik K, Winters MA, Eshleman SH (2000) A guide to HIV-1 reverse transcriptase and protease sequencing for drug resistance studies. HIV Sequence Compendium 2000.

Virus Fitness: Concept, Quantification, and Application

137

Shankarappa R, Margolick JB, Gange SJ, Rodrigo AG, Upchurch D, Farzadegan H, Gupta P, Rinaldo CR, Learn GH, He X, Huang XL, Mullins JI (1999) Consistent viral evolutionary changes associated with the progression of human immunodeficiency virus type 1 infection. J Virol 73:10489–10502 Sharma PL, Crumpacker CS (1997) Attenuated replication of human immunodeficiency virus type 1 with a didanosine-selected reverse transcriptase mutation [published erratum appears in J Virol 1998; 72:3504]. J Virol 71:8846–8851 Sharma PL, Crumpacker CS (1999) Decreased processivity of human immunodeficiency virus type 1 reverse transcriptase (RT) containing didanosine-selected mutation Leu74Val: a comparative analysis of RT variants Leu74Val and lamivudineselected Met184Val. J Virol 73:8448–8456 Shi C, Mellors JW (1997) A recombinant retroviral system for rapid in vivo analysis of human immunodeficiency virus type 1 susceptibility to reverse transcriptase inhibitors. Antimicrob Agents Chemother 41:2781–2785 Shirasaka T, Kavlick MF, Ueno T, Gao WY, Kojima E, Alcaide ML, Chokekijchai S, Roy BM, Arnold E, Yarchoan R (1995) Emergence of human immunodeficiency virus type 1 variants with resistance to multiple dideoxynucleosides in patients receiving therapy with dideoxynucleosides. Proc Natl Acad Sci U S A 92:2398–2402 Shriner D, Shankarappa R, Jensen MA, Nickle DC, Mittler JE, Margolick JB, Mullins JI (2004) Influence of random genetic drift on human immunodeficiency virus type 1 env evolution during chronic infection. Genetics 166:1155–1164 Siller S (2001) Sexual selection and the maintenance of sex. Nature (London) 411:689– 692 Simmonds P (2004) Genetic diversity and evolution of hepatitis C virus—15 years on. J Gen Virol 85:3173–3188 Simmons G, Clapham PR, Picard L, Offord RE, Rosenkilde MM, Schwartz TW, Buser R, Wells TN, Proudfoot AE (1997) Potent inhibition of HIV-1 infectivity in macrophages and lymphocytes by a novel CCR5 antagonist. Science 276:276–279 Simon V, Padte N, Murray D, Vanderhoeven J, Wrin T, Parkin N, Di Mascio M, Markowitz M (2003) Infectivity and replication capacity of drug-resistant human immunodeficiency virus type 1 variants isolated during primary infection. J Virol 77:7736–7745 Sole RV, Ferrer R, Gonzalez-Garcia I, Quer J, Domingo E (1999) Red queen dynamics, competition and critical points in a model of RNA virus quasispecies. J Theor Biol 198:47–59 Soto-Ramirez LE, Renjifo B, McLane MF, Marlink R, O’Hara C, Sutthent R, Wasi C, Vithayasai P, Vithayasai V, Apichartpiyakul C, Auewarakul P, Pena Cruz V, Chui DS, Osathanondh R, Mayer K, Lee TH, Essex M (1996) HIV-1 Langerhans’ cell tropism associated with heterosexual transmission of HIV. Science 271:1291–1293 Stapleton JT (2003) GB virus type C/hepatitis G virus. Semin Liver Dis 23:137–148 Svarovskaia ES, Barr R, Zhang X, Pais GC, Marchand C, Pommier Y, Burke TR Jr, Pathak VK (2004) Azido-containing diketo acid derivatives inhibit human immunodeficiency virus type 1 integrase in vivo and influence the frequency of deletions at two-long-terminal-repeat-circle junctions. J Virol 78:3210–3222

138

M. E. Quiñones-Mateu · E. J. Arts

Tamiya S, Mardy S, Kavlick MF, Yoshimura K, Mistuya H (2004) Amino acid insertions near Gag cleavage sites restore the otherwise compromised replication of human immunodeficiency virus type 1 variants resistant to protease inhibitors. J Virol 78:12030–12040 Tang JW, Pillay D (2004) Transmission of HIV-1 drug resistance. J Clin Virol 30:1–10 Temin HM (1989) Is HIV unique or merely different? J Acquir Immune Defic Syndr 2:1–9 Tersmette M, de Goede RE, Al BJ, Winkel IN, Gruters RA, Cuypers HT, Huisman HG, Miedema F (1988) Differential syncytium-inducing capacity of human immunodeficiency virus isolates: frequent detection of syncytium-inducing isolates in patients with acquired immunodeficiency syndrome (AIDS) and AIDS-related complex. J Virol 62:2026–2032 Tersmette M, Miedema F (1990) Interactions between HIV and the host immune system in the pathogenesis of AIDS. AIDS 4:S57–S66 Troyer R, Collins KR, Abraha A, Fraundorf E, Moore D, Krizan RW, Toossi Z, Colebunders RL, Jensen MA, Mullins JI, Vanham G, Arts EJ (2005) Impact of ex vivo HIV-1 fitness and genetic diversity on disease progression. J Virol 79 (in press) Turner PE, Chao L (1999) Prisoner’s dilemma in an RNA virus. Nature (London) 398:441–443 Turner PE, Elena SF (2000) Cost of host radiation in an RNA virus. Genetics 156:1465– 1470 Van Harmelen JH, van der RE, Loubser AS, York D, Madurai S, Lyons S, Wood R, Williamson C (1999) A predominantly HIV type 1 subtype C-restricted epidemic in South African urban populations. AIDS Res Hum Retroviruses 15:395–398 Van Harmelen J, Williamson C, Kim B, Morris L, Carr J, Karim SS, McCutchan F (2001) Characterization of full-length HIV type 1 subtype C sequences from South Africa. AIDS Res Hum Retroviruses 17:1527–1531 Velazquez-Campoy A, Todd MJ, Vega S, Freire E (2001) Catalytic efficiency and vitality of HIV-1 proteases from African viral subtypes. Proc Natl Acad Sci U S A 98:6062– 6067 Weaver SC, Brault AC, Kang W, Holland JJ (1999) Genetic and fitness changes accompanying adaptation of an arbovirus to vertebrate and invertebrate cells. J Virol 73:4316–4326 Weber J, Rangel HR, Chakraborty B, Marotta ML, Valdez H, Fransen K, Florence E, Connick E, Smith K, Colebunders R, Landay A, Kuritzkes DR, Lederman MM, Vanham G, Quiñones-Mateu ME (2003a) Role of baseline pol genotype in HIV-1 fitness evolution. JAIDS 33:448–460 Weber J, Rangel HR, Chakraborty B, Tadele M, Martinez MA, Martinez-Picado J, Marotta ML, Mirza M, Ruiz L, Clotet B, Wrin T, Petropoulos CJ, QuiñonesMateu ME (2003b) A novel TaqMan real-time PCR assay to estimate ex vivo human immunodeficiency virus type 1 fitness in the era of multi-target (pol and env) antiretroviral therapy. J Gen Virol 84:2217–2228 Weber J, Chakraborty B, Weberova J, Miller MD, Quiñones-Mateu ME (2005) Diminished replicative fitness of primary human immunodeficiency virus type 1 isolates harboring the K65R mutation. J Clin Microbiol 43:1395–1400

Virus Fitness: Concept, Quantification, and Application

139

Wei X, Ghosh SK, Taylor ME, Johnson VA, Emini EA, Deutsch P, Lifson JD, Bonhoeffer S, Nowak MA, Hahn BH (1995) Viral dynamics in human immunodeficiency virus type 1 infection. Nature (London) 373:117–122 Wei X, Decker JM, Liu H, Zhang Z, Arani RB, Kilby JM, Saag MS, Wu X, Shaw GM, Kappes JC (2002) Emergence of resistant human immunodeficiency virus type 1 in patients receiving fusion inhibitor (T-20) monotherapy. Antimicrob Agents Chemother 46:1896–1905 Wei X, Decker JM, Wang S, Hui H, Kappes JC, Wu X, Salazar-Gonzalez JF, Salazar MG, Kilby JM, Saag MS, Komarova NL, Nowak MA, Hahn BH, Kwong PD, Shaw GM (2003) Antibody neutralization and escape by HIV-1. Nature (London) 422:307– 312 Whetter LE, Ojukwu IC, Novembre FJ, Dewhurst S (1999) Pathogenesis of simian immunodeficiency virus infection. J Gen Virol 80:1557–1568 White HD, Robbins MD, Green WR (1990) Mechanism of escape of endogenous murine leukemia virus emv-14 from recognition by anti-AKR/Gross virus cytolytic T lymphocytes. J Virol 64:2608–2619 White KL, Margot NA, Wrin T, Petropoulos CJ, Miller MD, Naeger LK (2002) Molecular mechanisms of resistance to human immunodeficiency virus type 1 with reverse transcriptase mutations K65R and K65R+M184V and their effects on enzyme function and viral replication capacity. Antimicrob Agents Chemother 46:3437– 3446 Winston A, Mandalia S, Pillay D, Gazzard B, Pozniak A (2002) The prevalence and determinants of the K65R mutation in HIV-1 reverse transcriptase in tenofovirnaive patients. AIDS 16:2087–2089 Wodarz D, Nowak MA (1998) The effect of different immune responses on the evolution of virulent CXCR4-tropic HIV. Proc R Soc Lond B Biol Sci 265:2149–2158 Wolinsky SM, Wike CM, Korber BT, Hutto C, Parks WP, Rosenblum LL, Kunstman KJ, Furtado MR, Munoz JL (1992) Selective transmission of human immunodeficiency virus type-1 variants from mothers to infants [see comments]. Science 255:1134–1137 Wrin T, Gamarnik A, Whitehurst N, Beauchaine J, Whitcomb JM, Hellmann NS, Petropoulos CJ (2001) Natural variation of replication capacity measurements in drug-naive/susceptible HIV-1. 5th International Workshop on HIV Drug Resistance and Treatment Strategies, Scottsdale, Arizona June 2001, Abstract 24: Yoshimura K, Feldman R, Kodama E, Kavlick MF, Qiu YL, Zemlicka J, Mitsuya H (1999) In vitro induction of human immunodeficiency virus type 1 variants resistant to phosphoralaninate prodrugs of Z-methylenecyclopropane nucleoside analogues. Antimicrob Agents Chemother 43:2479–2483 Yuste E, Sanchez-Palomino S, Casado C, Domingo E, Lopez-Galindez C (1999) Drastic fitness loss in human immunodeficiency virus type 1 upon serial bottleneck events. J Virol 73:2745–2751 Yuste E, Lopez-Galindez C, Domingo E (2000) Unusual distribution of mutations associated with serial bottleneck passages of human immunodeficiency virus type 1. J Virol 74:9546–9552

140

M. E. Quiñones-Mateu · E. J. Arts

Zennou V, Mammano F, Paulous S, Mathez D, Clavel F (1998) Loss of viral fitness associated with multiple Gag and Gag-Pol processing defects in human immunodeficiency virus type 1 variants selected for resistance to protease inhibitors in vivo. J Virol 72:3300–3306 Zhang D, Caliendo AM, Eron JJ, DeVore KM, Kaplan JC, Hirsch MS, D’aquila RT (1994) Resistance to 2 ,3 -dideoxycytidine conferred by a mutation in codon 65 of the human immunodeficiency virus type 1 reverse transcriptase. Antimicrob Agents Chemother 38:282–287 Zhang H, Zhou Y, Alcock C, Kiefer T, Monie D, Siliciano J, Li Q, Pham P, Cofrancesco J, Persaud D, Siliciano RF (2004) Novel single-cell-level phenotypic assay for residual drug susceptibility and reduced replication capacity of drug-resistant human immunodeficiency virus type 1. J Virol 78:1718–1729 Zhang YM, Imamichi H, Imamichi T, Lane HC, Falloon J, Vasudevachari MB, Salzman NP (1997) Drug resistance during indinavir therapy is caused by mutations in the protease gene and in its Gag substrate cleavage sites. J Virol 71:6662–6670 Zhu T, Mo H, Wang N, Nam DS, Cao Y, Koup RA, Ho DD (1993) Genotypic and phenotypic characterization of HIV-1 patients with primary infection. Science 261:1179–1181 Zhu T, Wang N, Carr A, Nam DS, Moor-Jankowski R, Cooper DA, Ho DD (1996) Genetic characterization of human immunodeficiency virus type 1 in blood and genital secretions: evidence for viral compartmentalization and selection during sexual transmission. J Virol 70:3098–3107

CTMI (2006) 299:141–170 c Springer-Verlag Berlin Heidelberg 2006


Population Bottlenecks in Quasispecies Dynamics C. Escarmís1 (u) · E. Lázaro2 · S. C. Manrubia2 1 Centro de Biología Molecular “Severo Ochoa” (CSIC-UAM),

Universidad Autónoma de Madrid, 28049 Cantoblanco, Madrid, Spain [email protected] 2 Centro de Astrobiología (CSIC-INTA), Ctra. de Ajalvir km. 4, 28850 Torrejón de Ardoz, Madrid, Spain

1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142

2

Theoretical Approaches to Mutation and Selection Processes . . . . . . . . . 143

3

Mathematical Models Including Compensatory Mutations . . . . . . . . . . . 146

4

Experimental Effects of Population Bottlenecks . . . . . . . . . . . . . . . . . . 149

5

Bottlenecks and Muller’s Ratchet in RNA Viruses . . . . . . . . . . . . . . . . . 151

6

Sequence Changes Accompanying Loss of Viral Fitness Due to Genetic Bottlenecks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154

7

Recovery of Viral Fitness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158

8

Epistatic Interactions Among Mutations. The Role of Sex and Recombination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

9

Adaptability of Viral Populations with a Long Bottleneck History . . . . . . 163

10

Final Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166

Abstract The characteristics of natural populations result from different stochastic and deterministic processes that include reproduction with error, selection, and genetic drift. In particular, population fluctuations constitute a stochastic process that may play a very relevant role in shaping the structure of populations. For example, it is expected that small asexual populations will accumulate mutations at a higher rate than larger ones. As a consequence, in any population the fixation of mutations is accelerated when environmental conditions cause population bottlenecks. Bottlenecks have been relatively frequent in the history of life and it is generally accepted that they are highly relevant for speciation. Although population bottlenecks can occur in any species, their effects are more noticeable in organisms that form large and heterogeneous

142

C. Escarmís et al.

populations, such as RNA viral quasispecies. Bottlenecks can also positively select and isolate particles that still keep the ability to infect cells from a disorganized population created by crossing the error threshold.

1 Introduction The characteristics of natural populations result from different stochastic and deterministic processes that include reproduction with error, selection, and genetic drift. All these processes make possible the adaptation to changing environmental conditions. In particular, population fluctuations constitute a stochastic process that may play a very relevant role in shaping the structure of populations. For example, it is expected that small asexual populations will accumulate mutations at a higher rate than larger ones. As a consequence, in any population the fixation of mutations is accelerated when environmental conditions cause population bottlenecks. Then the size of the population, and thus its genetic diversity, is strongly reduced and selected through processes that do not depend on fitness. Any mutation present in this small number of individuals will be transmitted to most of their progeny. Bottlenecks have been relatively frequent in the history of life and it is generally accepted that they are highly relevant for speciation. Although population bottlenecks can occur in any species, their effects are more noticeable in organisms which form large and heterogeneous populations, such as RNA viral quasispecies. The high mutation rate of RNA viruses has as a consequence that each new individual generated has an average of 1 mutation relative to the consensus sequence (Batschelet et al. 1976; Drake and Holland 1999). Differences in the replicative ability of coexisting types, the continuous generation of new mutations, and the action of selective mechanisms, eventually define a complex population of interrelated individuals. If the environment is stable enough, the distribution of phenotypes attains a mutation-selection equilibrium characteristic of the environment where the population evolves. In terms of Wright’s fitness landscapes, the quasispecies occupies a fraction of the genotype space where survival of the population as a whole is optimized. In addition to a number of different genomes representing the phenotype with maximal fitness, there are many other variants coexisting in the population. The very structure of viral quasispecies permits the presence of minority genomes which explore regions of the genotype space far from the optimum fitness peak. The structure of the genotype space that is occupied by a quasispecies has not been studied in depth experimentally, since it would require sequencing a very large number of genomes. However, it is known that virus

Population Bottlenecks in Quasispecies Dynamics

143

isolates which have evolved under different adaptive pressures have different genotype structures and explore diverse regions of the fitness landscape, even in cases where they share the same consensus sequence. Given appropriate conditions, such as strong environmental changes or bottlenecks, genomes that are present at a low amount can be selected to generate a new quasispecies with suitable properties, and thus guarantee the persistence of the population. We have strong evidence that the formation of self-sustained and highly heterogeneous populations is an intrinsic feature of RNA viruses that strongly conditions their adaptability, robustness, and evolutionary dynamics. This review is structured as follows. In Sects. 2 and 3, we discuss a number of theoretical approaches relevant to mutation and selection processes in heterogeneous populations. We start with classical predictions on the effect of Muller’s ratchet and present more recent mathematical models which already take into explicit account the presence of bottlenecks. In Sect. 4, experiments and observations of bottlenecks in different organisms are reviewed, and Sect. 5 is entirely devoted to empirical results with RNA viruses. In that section, it is shown that periodic bottlenecks applied to optimized populations result in decreases in fitness. Changes in genomic sequences accompanying the fitness losses are discussed in Sect. 6. Section 7 presents a number of mechanisms through which viral fitness can be recovered. In particular, stationary states of fitness can be expected if compensatory mutations occur. The role of epistatic interactions in halting the progressive loss of fitness and experimental and theoretical observations on interactions between mutations are reviewed in Sect. 8. Finally, Sect. 9 discusses how a long history of in vitro bottlenecks can affect the adaptability of viral populations. We conclude with some final remarks in the last section.

2 Theoretical Approaches to Mutation and Selection Processes In the early days of population genetics, little was known about the actual molecular mechanisms driving evolution and adaptation of populations. The first attempts to formalize the process of appearance of variants and the eventual fixation or disappearance of the mutation involved were carried out by Haldane, Fisher, and Wright, in a series of works of the highest relevance which settled the basis of theoretical population genetics (Haldane 1924; 1927; Fisher 1922, 1930; Wright 1931, 1939). One of the main concerns at the time regarded the fate of mutations in asexual populations. Since the appearance of change is an unavoidable effect of reproduction, mutations would necessarily accu-

144

C. Escarmís et al.

mulate in the population if mechanisms such as recombination or sex, which could eliminate them, were absent. This process was studied, among others, by Muller (1964), who compared the unavoidable accumulation of mutations in asexual populations to the clicks of a ratchet mechanism. However, and contrary to some current interpretations, Muller did not equate accumulation of mutations to degradation of the population. In his own words: Under conditions where only stability of type is needed, a nonrecombining population does not actually degenerate as a result of an excess of mutation over selection, after the usual equilibrium between these pressures is reached. However, a kind of irreversible ratchet mechanism exists in the non-recombining species (unlike the recombining ones) that prevents selection, even if intensified, from reducing the mutational loads below the lightest that were in existence when the intensified selection started, whereas, contrariwise, “drift” and what might be called “selective noise” must allow occasional slips of the lightest loads in the direction of increased weight. In asexual populations, deleterious mutations can be fixed, particularly if the population size is small (Crow and Kimura 1970). Eventually, an asexual population of size N subjected to a mutation rate u and a selection coefficient s against deleterious mutations attains a deterministic mutation–selection equilibrium. At that point, there is a fixed amount of genomes nm with m mutations. The equilibrium distribution of the population was calculated by Kimura and Maruyama (1966) and reads nm

=N

exp (−u|s)  u m . m! s

(1)

As long as N is large enough, there will be a significant amount of wild-type (mutation free) genomes, n0

= N exp (−u|s) .

(2)

However if N is small or the coefficient u|s is sufficiently large, the number of genomes in the mutation-free class can be small enough that population fluctuations cause the disappearance of all of the individuals in that class. If this happens, the ratchet has clicked once and the least loaded class corresponds now to individuals carrying one mutation. Analogously, the one-mutation class can disappear due to population fluctuations, and the ratchet clicks again. This model implicitly assumes that reversions (an exceedingly rare process implying in this framework a change from the class m to the class m − 1) are the only mechanism able to produce an increase in fitness. If the

Population Bottlenecks in Quasispecies Dynamics

145

highest fitness class is identified with the mutation free class and is assigned value 1, the fitness w(m) of individuals with m mutations will be w(m) = (1 − s)m .

(3)

This defines a multiplicative fitness landscape for the mutated genomes and implicitly assumes that each mutation affects fitness independently (epistatic effects are not included). Nowadays, most interpretations of Muller’s ratchet equate accumulation of mutations with fitness loss. The rationale behind this identification comes from presupposing that the mean genotypic value of the population is close to the optimum. Indeed, if one accepts that populations are well adapted to their natural environment, any further change, that is, any new mutation, should have a deleterious effect on their fitness with high probability. Comparatively, advantageous changes would be much rarer, to the point that they can be ignored. The first formal study of Muller’s ratchet under the previous assumptions, and using the results of Kimura and Maruyama (Eq. (1)) was carried out by Haigh (1978) on a model proposed by Felsenstein (1974). He concluded that the speed of the ratchet (the time between two successive clicks or losses of the least-loaded class) depended chiefly on n0 (see also Bell 1988a). However, more recent analyses (Stephan et al. 1993; Gordo and Charlesworth 2000) reveal that the speed of the ratchet depends not only on the number of individuals in the least mutated class, but also on the parameters u and s. Seen from a different viewpoint, this is equivalent to saying that it is the whole structure of the population that determines the fate of each and every class of mutants present. Following the initial studies of Felsenstein and Haigh, other authors analysed and discussed different properties of Muller’s ratchet under similar assumptions and proposed recombination as the most effective mechanism in order to arrest the ratchet and thus recover high-fitness (or mutation-free) genomes (Maynard Smith 1978; Kondrashov 1982). Epistatic effects among mutations where also considered as a possible mechanism capable of accelerating or slowing down the speed of the ratchet. Synergistic epistasis refers to the situation where the joint effect of two deleterious (or beneficial) mutations is larger in absolute value than the sum of the individual effects, and antagonistic epistasis occurs when two mutations in the same genome change fitness in an amount smaller than the sum of their individual effects. Charlesworth et al. (1993) observed that under weak antagonistic epistasis, the time between successive clicks of the ratchet grows, though the rate of fitness decline is barely affected (as in previous models, it is assumed that all mutations have a negative effect on fitness). However, under stronger epistatic selection the ratchet can be effectively halted, such that, in practice,

146

C. Escarmís et al.

a finite population can survive almost indefinitely (Kondrashov 1994). In contrast, Colato and Fontanari (2001) studied the speed of the ratchet in a population subjected to bottlenecks (with a time long enough between bottlenecks such that mutation–selection equilibrium could be attained), and introduced a fitness function of the form α

w(m) = (1 − s)m .

(4)

The parameter α is the epistasis parameter: synergistic and antagonistic epistasis is described by α > 1 and α < 1, respectively, and Eq. (3) (absence of epistasis) is recovered for α = 1. While, in all cases studied, antagonistic epistasis slows down the ratchet and synergistic epistasis accelerates it, other effects, such as a decrease of the population size facilitated by the steady accumulation of mutations, can notably accelerate the extinction of the population due to mutational meltdown (Lynch and Gabriel 1990).

3 Mathematical Models Including Compensatory Mutations The studies reviewed up to now describe the theoretical understanding of the operation of Muller’s ratchet in large populations of approximately constant size, with the exception of Colato and Fontanari’s work. There were earlier attempts to include population bottlenecks in mathematical models. Those first studies were motivated by the form of transmission of mitochondrial genomes in mammals, which have no recombination and mostly undergo monoparental inheritance. With a simple model devised to represent such a situation, Bergstrom and Pritchard (1998) demonstrated that “rather than hastening genetic degradation, a bottleneck may be essential in maintaining mitochondrial genetic quality over evolutionary time.” This result is in contrast with most previous theoretical expectations, which predicted that mutations would accumulate more easily in small populations, such that they would consequently suffer from faster degradation than larger ones. However, the large amount of asexual, nonrecombining species subjected to strong population bottlenecks in their natural environments which, despite of them, seemed to keep a high average fitness, turned the attention to other forms in which the continued degeneration predicted by most models could be compensated. The occurrence of compensatory mutations (not reversions) which could have phenotypic effects and even outweigh the effect of the accumulated deleterious mutations has been considered (Wagner and Gabriel 1990; Bergstrom et al. 1999; Lázaro et al. 2002; Rouzine et al. 2003; Bachtrog and Gordo 2004). Wagner and Gabriel (1990) review the assumptions and

Population Bottlenecks in Quasispecies Dynamics

147

implications of previous models of Muller’s ratchet and suggest that compensatory mutations can be as effective as recombination in halting the deleterious effect of the ratchet. Interestingly, they separate effects in the genotype from changes in the phenotype, a relevant difference not considered in most previous theoretical approaches. This distinction is especially relevant in the context of quasispecies, since the mutation–selection equilibrium is defined as the stationary distribution of fitness values selected in the given environment. This equilibrium does not fix the consensus sequence, which would surely be affected by (quasi-) neutral drift. Starting with a high-fitness variant, Wagner and Gabriel observed decreases in fitness for a number of generations due to the action of the ratchet. However, at a certain point compensatory mutations stop the effect of deleterious mutations and the average fitness of the population keeps an average value with small fluctuations around it. Hence, the evolution of the population displays a biphasic behavior, reaching a statistically stationary state after a transient state where fitness decreases, thus avoiding extinction. Actually, in all models where compensatory mutations are explicitly included, it is possible to recover higher fitness states in the population, and to escape extinction. As a side effect, biphasic evolution is observed whenever the initial fitness of the population is far from the average stationary fitness at the corresponding mutation–selection equilibrium. This is so even when bottlenecks, and thus strong fluctuations in the population size – leading to an enhanced mutation fixation rate – are periodically applied. Aiming at understanding the relationship between the mode of transmission of a pathogen and its virulence, Bergstrom and co-workers (Bergstrom et al. 1999) proposed a simple model where the population was structured in several fitness classes. A down-mutation (occurring with probability p) moved the offspring of a genome from class w to class w − 1, while an upmutation increased fitness from w to w + 1 (this happening with probability q). By means of numerical simulations, it was shown that vertically transmitted pathogens suffer decreases in fitness much stronger than when transmission is horizontal. A similar model was numerically (Lázaro et al. 2002) and analytically (Manrubia et al. 2003) studied in order to understand the appearance of biphasic behavior and large fluctuations in the viral yield observed in an in vitro system for viral evolution (Escarmís et al. 2002). An additional parameter of this model is the development time between bottlenecks, which determines the distance to the mutation–selection equilibrium (which would only be attained after very many generations), and also constrains the degree of heterogeneity of the population before the bottleneck. This model has a number of differences when compared to classical descriptions of Muller’s ratchet, where only deleterious mutations were taken into account. First, the fitness of a sequence with m mutations depends now

148

C. Escarmís et al.

on its history, that is to say, in the number of compensatory and deleterious mutations it has suffered. On the average, w(m) = 1 − m(p − q)s ,

(5)

since a fraction p of the accumulated mutations are deleterious and a fraction q are compensatory or beneficial, both modifying fitness in the same amount according to the model. However, the actual value of fitness of an individual carrying m mutations derived from an original sequence of fitness w0 can take any value between w0 − ms and w0 + ms (with a minimum at 1|s and a maximum at 1), corresponding to all the mutations being deleterious or all being beneficial, respectively. By construction, the selection coefficient in this model is 1 (6) s= , F where F is the number of different fitness classes considered. If all mutations are deleterious, the fitness of the genome is w(m) = 1 − ms ,

(7)

which corresponds to a first-order expansion of Eq. (3). In this model, all mutations have the same effect on fitness regardless of the fitness state of the genome suffering the mutation. Though other models assume that mutations decrease or increase fitness in a fixed percent, and not in a fixed amount, it is possible, under certain conditions, to demonstrate that at least the changes in fitness due to beneficial mutations act in an additive way (Orr 2003). Some models, including beneficial mutations, have tackled the problem of fixation of an advantageous mutant in this new context. Peck (1994) argued that mutations with a positive effect on fitness could appear in stable environments because no species is likely to be perfectly adapted. In changing environments, beneficial mutations should not be that rare at all. Still, he discussed that a beneficial mutation (a ruby) appearing in a background of deleterious mutations (the rubbish) would have little chance of being fixed in an asexual population. As in previous analysis, Peck’s results supported the great advantage of sex and recombination in promoting adaptation, but, indirectly, also pointed to the relevance of the population structure (through the amount and distribution of deleterious mutations, in that case) in determining the fate of a mutant. The fixation of higher fitness variants in the framework of quasispecies has been explored further (Wilke 2003), with the conclusion that the chances for an advantageous mutant to be eventually fixed in the quasispecies can be only accurately estimated if the fitness of all potential members of the invading quasispecies is known.

Population Bottlenecks in Quasispecies Dynamics

149

The equilibrium distribution in Eq. (1) is attained after a long development time, for large enough populations, and only if all mutations have the same effect on fitness. Most populations, even in carefully controlled experiments, are expected to be out of the equilibrium described by Eq. (1). In particular, bottlenecks condition the structure of the population and its phenotypic characteristics in nontrivial ways. Following a bottleneck, both the frequency and variance of the population’s genetic composition change (Zhang et al. 2004). If bottlenecks are regularly applied such that the development time of the population between bottlenecks is much shorter than the time required to achieve mutation–selection equilibrium, the (out-of-equilibrium) composition of the population in the presence of deleterious and beneficial mutations can be analytically calculated (Manrubia et al. 2003). Several of the models reviewed have been designed in order to better understand the dynamics of in vitro experiments. These will be described in the forthcoming sections, where further results from the models discussed here will be presented.

4 Experimental Effects of Population Bottlenecks A population is well adapted to a given environment when the processes of competition and selection have acted for a time long enough to reach an equilibrium close to the optimum. Nevertheless, mutations are inherent to replication and, in any well-adapted population, many of them will have a deleterious effect on fitness. In a situation where only a small number of individuals found a new population, the genetic diversity is reduced and mutations are transmitted to most of the descendants. Under these circumstances, the chances that the least mutated sequence is lost increase notably. The process of mutation accumulation is particularly effective when the mutation rate is high, as in RNA viruses. Attending to theoretical predictions derived from models where compensatory mutations were not taken into account, it was expected that the repetition of serial bottlenecks could lead to the extinction of the population. Early theories on population evolution left open many questions that could be solved only experimentally, as for instance the actual role that sex and recombination play as mechanisms able to counteract the ratchet mechanism. Other debated questions concerned the way in which the progressive accumulation of mutations affect fitness: do mutations have a simple multiplicative effect? Are epistatic interactions important in the process? Is the effect of a mutation the same irrespective of other mutations accumulated in the genome? Can beneficial or compensatory mutations really

150

C. Escarmís et al.

outweigh the effects of the ratchet? In the forthcoming sections, we will discuss the experimental studies carried out to answer these important questions. The empirical implementation of Muller’s ratchet in RNA viruses has been carried out through the classical protocol of plaque-to-plaque transfers of clonal populations (Fig. 1). Each transfer starts with a single viral genome. To this end, the starting population is appropriately diluted and plated under a semisolid agar layer. The virus contained in a single plaque of the progeny is isolated, diluted again, and plated a second time. A second plaque from the latter progeny is isolated and the process is serially repeated. The effect of Muller’s ratchet following this procedure was experimentally observed for the first time by Lin Chao with the phage φ6 (Chao 1990). Since then, it has been studied in several microorganisms, such as the phage MS2 (de la Peña et al. 2000), several animal viruses, such as vesicular stomatitis virus (VSV) (Duarte et al. 1992), foot-and-mouth disease virus (FMDV) (Escarmís et al. 1996), and human immunodeficiency virus (HIV) (Yuste et al. 1999). All these viruses were subjected to plaque-to-plaque transfers, and it was documented that this transmission regime resulted in losses of viral fitness. Although in this review we will refer to RNA viruses, the effect of bottlenecks on evolution has also been studied in DNA viruses, such as phage T7 (Bull et al. 2003), where strikingly similar conclusions have been reached. In that case, the

Fig. 1 Scheme of the plaque-to-plaque protocol. At each transfer a lytic plaque is generated from a single viral particle. The progeny present in the plaque is resuspended, diluted and plated to generate the next plaque (top). The consensus sequence of the population at each transfer carries the mutations from the parental genome (bottom)

Population Bottlenecks in Quasispecies Dynamics

151

authors increased the mutation rate of the DNA phage through the addition of a mutagen during growth. Evidence of the operation of Muller’s ratchet has also been obtained in studies with protozoa (Bell 1988b), bacteria (Andersson and Hughes 1996; Kibota and Lynx 1996; Funchain et al. 2000), yeast (Zeyl et al. 2001), and multicellular organisms (Fry et al. 1999; Vassilieva et al. 2000). The issue of whether a bottleneck occurred among our human ancestors is a question that is strongly debated by geneticists and anthropologists (Ayala et al. 1994; Harpending and Rogers 2000).

5 Bottlenecks and Muller’s Ratchet in RNA Viruses The experiments carried out by Chao (1990) with the tripartite bacteriophage φ6 involved 40 consecutive plaque-to-plaque transfers, which led to a significant decrease in mean fitness. No extinction of infectivity was observed. The experimental design precluded reassortment of the three genome segments of the phage while allowing the comparison of the relative replication rates of wild-type phage vs repeatedly bottlenecked clones. In this way, sex (understood as encapsidation in the same particle of genomes from different viruses), as a mechanism able to repair genetic lesions was excluded. In subsequent studies carried out by the same group (Froissart et al. 2004), it was suggested that the mutational load could be purged faster in the absence of co-infection, thus hinting at the possibility that complementation diminishes the benefits of sex. Duarte et al. (1992) studied the relative fitness of VSV, the prototype model for negative-stranded RNA viruses, when subjected to serial bottlenecks. VSV has a single nonsegmented genome and does not experience recombination at a detectable level. The assay to determine fitness was direct competition between the parental clone and genetically marked monoclonal antibodyresistant (MAR) clones. They found that after only 20 plaque-to-plaque transfers, the viral clones displayed wide variations in fitness. The overall trend was towards mean fitness reduction, in agreement with the classical interpretation of Muller’s ratchet. Again, no extinctions of infectivity were observed. An additional study concerning the accumulation of mutations through bottlenecks in the same virus has been carried out by Elena and Moya (1999), who theoretically calculated a deleterious mutation rate of 1.2 mutations per genome and generation, with a mean fitness effect of −0.39% per generation. The average decrease in fitness per transfer may also change depending on the fitness of the viral clones at the beginning of the experiment, as Elena and co-workers have shown. High-fitness clones lose fitness faster than clones

152

C. Escarmís et al.

with lower fitness (Elena et al. 1996): a well-adapted clone can accumulate deleterious mutations more easily than a clone that is less well adapted. The opposite is also true, since low-fitness clones are comparatively more prone to acquiring advantageous mutations and thus to improving fitness. In the study conducted by de la Peña and collaborators (de la Peña et al. 2000), four MS2 clones were subjected to 20 serial bottlenecks. Although the number of transfers was rather low, they found a sharp decrease in fitness (about 96% at the end of the experiment). They observed that in three of the clones under study, fitness decreased at a constant rate as the number of transfers progressed. One clone lost fitness at a rate significantly higher during the first ten transfers than in the last ten transfers, thus showing biphasic behavior in its evolution. As discussed in the sections devoted to theoretical models, the decrease in fitness should stop after a period of variable length whenever compensatory mutations are included (Wagner and Gabriel 1990; Lázaro et al. 2002). However, in plaque-to-plaque transfers carried out with HIV-1 (Yuste et al. 1999), only four out of ten clones could produce viable progeny after 15 plaque-to-plaque transfers, and three of the four survivors showed important decreases in fitness. This represents a relevant difference with other viral systems, where extinctions were not observed. The most extensive study on the effect of repeated bottlenecks on fitness evolution and mutation accumulation has been carried out by Escarmís and co-workers (Escarmís et al. 1996; 1999; 2002; Lázaro et al. 2002; 2003) using the animal pathogen FMDV, a positive-strand RNA virus of the Picornaviridae family that, in contrast to VSV, can undergo recombination. The main objective was to determine whether successive reductions in the population size could lead viral populations to extinction. The titer of virus in the plaques (PFU/plaque) at the successive transfers was taken as a measure of its fitness. If the titer of each plaque is represented as a function of the number of plaque transfers, biphasic dynamics are observed (Fig. 2). After a period where the plaque titer decreases exponentially, a stationary state with large fluctuations around a constant average fitness value is reached. A detailed analysis of the statistical properties of fluctuations in the viral yield was performed, with the conclusion that the expected yield at the stationary state follows a Weibull distribution (Lázaro et al. 2003). By means of a simple model (Lázaro et al. 2002; Manrubia et al. 2003), it was shown that this type of function is to be expected if an exponential growth of the population (affected by both deleterious and compensatory mutations) takes place between successive bottlenecks. The results obtained with FMDV sharply contrast with the high number of extinctions observed when HIV-1 clones were subjected to successive bottlenecks. This can be ascribed to the number of replication rounds that takes place during the development of the HIV-1 plaques. This number is probably

Population Bottlenecks in Quasispecies Dynamics

153

Fig. 2 Titer of successive plaques of a clone subjected to plaque transfers

lower than in other viruses and, therefore, the possibilities of competition and selection inside HIV-1 plaques are more limited than in the case of FMDV. Even a single plaque is formed by a heterogeneous genome population, and the degree of optimization of this population depends on the number of replication cycles undergone by the virus (Manrubia et al. 2005). Mutations accumulated steadily in the consensus nucleotide sequence in the fitness decrease phase as well as in the constant average fitness phase (Escarmís et al. 2002). These results are unexpected in the light of classical Muller’s ratchet interpretations. This might indicate that, although mutations always occur at the same rate, their nature and effects can vary depending on the number and type of mutations previously accumulated in the genome. Selection may act with different strengths depending on the average fitness of the population, since genomes with low fitness are less tolerant to the introduction of additional deleterious mutations. The average rate of fitness decline (determined as the slope of the straight line obtained when the logarithm of fitness is represented as a function of the transfer number) in the study conducted with the phage MS2 is similar to that observed in the exponential phase of fitness losses in the case of FMDV (0.16 for MS2 and 0.12 for FMDV). The similarity between both results is striking, especially considering that, in the case of the phage MS2, fitness was deter-

154

C. Escarmís et al.

mined by quantifying genotypes (using the Northern blot technique) instead of phenotypes (counting the number of plaque-forming units for FMDV). It could be very interesting to study whether the documented resistance of bottlenecked FMDV clones to extinction (Escarmís et al. 2002, Lázaro et al. 2003) is a generic fact or a particular property of this virus. Testing this would require subjecting other viruses to a comparably large number of plaque-toplaque transfers. There are many viral systems that have already experienced dramatic fitness losses. It could be checked if those clones become extinct after subjecting them to additional bottlenecks. The characteristics of the state attained by the virus after many bottlenecks depend on the number of viral particles that generate each new population. The effect of different bottleneck sizes on fitness changes has been studied experimentally (Novella et al. 1995a). Novella and co-workers used clones of different fitness values as starting populations. They observed that the effective size of a genetic bottleneck causing fitness loss is larger when the fitness of the parental population increases. For example, for starting virus populations with low fitness, population transfers of five-clone-to-five-clone passages resulted in a fitness increase. However, when a parental population with high fitness was transferred, 30-clone-to-30-clone passages were required simply to maintain fitness values. This result can be explained by the probability of sampling variants of lower, equal or greater fitness than the fitness of the original population. As a consequence of this dependence between initial fitness and size of the bottleneck required to observe Muller’s ratchet effect, Novella et al. (1999) suggested that bottleneck effects limited exponential fitness gains of RNA virus populations passaged at a high population concentration. In this regime, viral fitness reaches a plateau at which stochastic fitness variations were observed. In their view, transmission through a large number of particles may represent a bottleneck at this high fitness value and might limit further increases in fitness. Their results are in agreement with a simple model where it has been shown that, at the stationary state, the average fitness of the population is determined by the size of the population bottleneck, and does not depend on the initial fitness of the population (Lázaro et al. 2002).

6 Sequence Changes Accompanying Loss of Viral Fitness Due to Genetic Bottlenecks Several genomes of viruses subjected to serial plaque transfers have been sequenced to investigate the genetic changes that accompany fitness losses. In general, in all the viruses analyzed, unusual mutations have been found. This

Population Bottlenecks in Quasispecies Dynamics

155

result indicates that, during the development of a plaque, negative selection has insufficient time to act, that mutants that would be otherwise eliminated by competition with fitter genomes remain in the population and have a chance to be selected for the next transfer. In a study with HIV (Yuste et al. 1999, 2000), the analysis of the mutations present in several clones revealed a predominance of transitions over transversions. One of the clones analyzed displayed a remarkably high mutation frequency (28 mutations accumulated after 15 transfers) and showed a very high abundance of G → A transitions uniformly distributed along the genome. This had been previously observed in natural isolates of HIV-1. The remaining mutations were distributed in an unusual way. There was a statistically significant higher accumulation of mutations in gag and the first third of the genome, compared to env, which appeared to be the most conserved region in all the clones studied. This is in sharp contrast to the mutation distribution observed in natural HIV isolates and in large population passages of HIV-1 clones subjected to positive selection. In these populations, gag and pol are more conserved than env. The reduced action of purifying selection during plaque development can cause the differences observed. However, the low number of mutations in env remains puzzling. A possible explanation might be that mutation rates are not the same in different genomic regions, and that the replication of gag is more error prone. In a study carried out with VSV (Novella and Ebendick-Corpus 2004) the mutations accumulated during the operation of Muller’s ratchet were found in both coding and noncoding regions. Mutations accumulated at a rate of 1.92 × 10−5 per nucleotide and per transfer. Transitions were 2.1-fold more abundant than transversions, as observed in many viral systems. Nonsynonymous mutations were also more frequent than synonymous mutations. However, if the number of mutations of each type is corrected by the number of synonymous (ds) and nonsynonymous (dn) sites in the genome, a different result is obtained: dn|ds = 0.48. Several strains accumulated a high number of mutations in the N open reading frame, contrasting with the conservation of this region in populations under positive selection or in natural isolates of the virus. This indicates again that genomic regions that appear to be very conserved in different viruses can mutate at a rate comparable to the remainder of the genome, though they may experience stronger constraints. Negative selection eliminates most of the mutants in these conserved regions, and mutations are not observed unless negative selection is reduced, as happens when bottlenecks occur. That is to say, bottlenecks allow isolating and maintaining genomes that would be eliminated in other situations. The changes observed in FMDV qualitatively differed from those fixed during large population passages (Escarmís et al. 1996; 2002). The dominance

156

C. Escarmís et al.

of nonsynonymous mutations associated with large population passages diminished upon repeated plaque transfers. The percentage of nonsynonymous mutations ranged from 39% to 48% (Escarmís et al. 1996), although this number is not normalized with respect to the total number of synonymous and nonsynonymous sites in the genome. Of the substituted residues in capsid proteins VP1, VP2 and VP3, 33% were internal (exposed neither in the outer nor in the inner surfaces of the protein shell, as determined by their accessibility to a probe of 0.3 nm radius; Lea et al. 1994). In contrast, only 4% of the substitutions found in viruses subjected to large population passages affected internal amino acid residues. Around 50% of the clones subjected to 30 serial plaque transfers acquired different extensions of five adenosine residues preceding the uridine of the second AUG initiation codon (see Fig. 3) (Escarmís et al. 1996, 2002). The phenotypic effect of this internal poly A elongation is not known, but its presence is deleterious for the virus. From data in Fig. 4, it can be deduced that there is an inverse linear correlation (r = −0.77) between the logarithm of plaque titer and the length of the poly A tract (p < 0.001, Student’s t test) (Escarmís et al. 2002). To predict possible phenotypic effects of the fixed mutations in the plaqueto-plaque transfers, each mutation was analyzed with regard to potential effects on RNA and protein structure and function, as well as the variability of the residues according to the FMDV sequences available. Mutations that affected conserved residues and therefore could account for the loss of fitness observed in the plaque transfers were disrupting a pseudoknot, changing the Y of VPg2 responsible for binding to the RNA rendering VPg2 inactive; a mutation in the S fragment that destabilizes its secondary structure; amino acid replacements in capsid proteins VP1 and VP4; and amino acid replacements in nonstructural proteins 2C, 3A and Vpg1 (for a detailed map of

Fig. 3 Scheme of the genome of foot-and-mouth disease virus. Location of the elongated internal oligoadenylate preceding the second initiating AUG. The number of As preceding the AUG in the wt is four

Population Bottlenecks in Quasispecies Dynamics

157

Fig. 4 Dependence of the titer of a plaque on the length of the internal elongated oligoadenylate present in the genome. The length of the oligoadenylate was determined by amplifying the RNA (RT-PCR) in the presence of 35 S-dATP with two primers flanking the oligoadenylate. The amplification products were electrophoresed through a 6% polyacrylamide sequencing gel. The heterogeneity of the oligoadenylate gives rise to a number of bands. The position in the gel of the product obtained from the wt genome (total of 5 A s) is marked with an arrow on the right. The determination was done for four clones. Clones 3a, 3b and 3c are three clones isolated in plaque transfer number 3 of clone 2.Clones 4a and 4b are two clones isolated in plaque transfer 4 of clone 4. These clones were transferred, giving rise to clones 5a and 5b

FMDV, see Mason et al. 2003). However, confirmation of the adverse effects of these mutations on virus viability would require their separate inclusion in a cDNA clone, growth-competition experiments between viruses with and without each mutation and depending on the outcome, the monitoring of nucleotide sequences of the progeny viruses. No back mutations (reversions) were observed in the plaque transfers. Two FMDV clones were subjected to prolonged plaque-to-plaque transfers and the genomic sequences were determined at different plaque transfers. The comparison of the sequences shown in Fig. 5 revealed a new mechanism of clonal diversification by mutation clustering. It consists of the accumulation of mutations in some genomic regions, with localized mutation frequencies significantly larger than those observed in the whole genome. Mutation clustering was demonstrated by dividing the genome into 250 nucleotide-long fragments with a 200-base overlap and comparing the mutation frequency of

158

C. Escarmís et al.

each fragment to that of the whole genome. In this way, three regions with a significantly enhanced mutation frequency were identified: the region of the L protein, the 2A2B region, and the 3A3B region. The mechanism by which this clustering is produced is unknown. Three possible mechanisms have been suggested (Escarmís et al. 2002): local variation of the copying fidelity of the viral replicase due to the presence of certain mutations; production of compensatory mutations close to the triggering ones; some mutations making a particular genomic region more tolerant to neighboring mutations. Weak evidence of mutation clustering has been obtained at some loci of FMDV genomes subjected to large population passages, but clustering is less obvious, presumably due to a more active negative selection. In successive plaque transfers of FMDV, mutations accumulated at a rate of about 0.3 nucleotides per plaque transfer, even with constant average fitness. The mutation frequency attained after 130 plaque-to-plaque transfers was 4.8 × 10−3 changes per nucleotide. The rate of accumulation of mutations during large population passages of FMDV clones in liquid culture medium was 0.15 mutation per passage. It is difficult to compare the number of genome copying rounds during the development of a plaque from 1 to 102 –104 PFU (the range observed for FMDV clones) with the number of rounds in one passage in liquid culture medium (generally a tenfold increase in PFU). Nevertheless, the important selective difference between the two regimes (plaque-to-plaque transfers and large population passages) appears to be the nature and location of the mutations rather than the rate of fixation (Escarmís et al. 2002).

7 Recovery of Viral Fitness A large number of experimental studies have been devoted to understanding the mechanisms that permit recovery of fitness in bottlenecked viruses. Some experiments have analyzed the transfer regimes that cause fitness gains. More recent studies have also investigated the number and type of mutations appearing during the process. As discussed, large population passages of RNA viruses usually result in an increase in relative fitness due to competition and selection of genomes within the quasispecies. (Novella et al. 1995b; Weaver et al. 1999). This transmission regime is also effective to recover fitness in viruses that have been debilitated by the accumulation of mutations through bottlenecks (Clarke et al. 1993; Escarmís et al. 1999; Novella and Ebendick-Corpus 2004). One study (Duarte et al. 1993) showed that two consecutive large population passages intercalated between bottlenecks were not enough to avoid the

Population Bottlenecks in Quasispecies Dynamics

159

Fig. 5 Location of the mutations found in the genome of two FMDV clones subjected to serial plaque-to-plaque transfers. In each panel, the top horizontal line is a scheme of the FMDV genome, in which the main regulatory regions and encoded proteins are indicated. The lines below the genome indicate nonsynonymous (ns) and synonymous (s) mutations present in virus of plaque transfers 0, 50 and 100 for clone a, and 0, 50, 78 and 84 for clone b. The bottom line indicates the size of the FMDV genome. The letter a indicates the presence of a mixture of two nucleotides not seen in subsequent transfers. b Represents a mixture of two nucleotides, and the mutant residues became dominant upon multiplication of the virus in liquid culture medium

160

C. Escarmís et al.

effect of the ratchet. These experiments were carried out with viruses of high to moderately low fitness. Other studies (Elena et al. 1998) have explored the evolution of VSV subclones derived from a very debilitated clone obtained after 20 successive bottlenecks. All the subclones analyzed gained fitness under different transmission regimes. As pointed out above, a possible explanation for this discrepancy is that, although mutation rates should be the same for all the clones, the fraction of beneficial mutations is larger in a clone with lower fitness. There is a higher chance of improving a debilitated virus than a well-adapted one. This reveals that the assumption that mutations have a multiplicative effect on fitness, as assumed in most theoretical models used to study the accumulation of mutations in genomic sequences, is probably too simplistic. The recovery of the fitness of a debilitated FMDV clone through large population passages and the mutations accompanying this recovery were also studied (Escarmís et al. 1999). The debilitated clone included the internal polyadenylate extension and a number of additional mutations scattered throughout the genome. Three different pathways, regarding the internal polyadenylate tract, were followed by several subclones to recover fitness: (a) a true reversion to yield the wild-type sequence; (b) a shortening of the internal polyadenylate tract; and (c) a deletion of 69 residues spanning the site of the internal polyadenylate tract. These results indicate that an RNA virus can find multiple pathways to reach alternative high-fitness peaks on the fitness landscape. In large population passages, only a few reversions were observed (Escarmís et al. 1999). This low number of reversions is also true for bacteria. It has been documented that the deleterious effects of a resistance mutation in Salmonella typhimurium is compensated by a variety of new mutations, and only four out of 81 independent lineages contained true streptomycin-sensitive revertants (Maisnier-Patin et al. 2002). Novella and Ebendick-Corpus (2004) also analyzed the genetic changes involved in the recovery of the negative effect of Muller’s ratchet by large population passages. They found that fitness increases were associated with both reversions and compensatory mutations. A significant level of fitness increase was observed in several bottlenecked strains with no obvious changes in the consensus sequence. The structure of the ensemble of genomes composing the quasispecies must have been altered in some way, confirming that the consensus sequence represents only a fraction of the genomic information contained in the quasispecies. Three of these strains which did not alter the consensus sequence were subjected to additional large population passages. At passage 20, each population further increased its fitness and a single mutation was observed in each case. None of the recovered strains reverted to a wild-type sequence. This is consistent with the observation that bottlenecks

Population Bottlenecks in Quasispecies Dynamics

161

force a genome to constitute a new quasispecies at a different position of the genomic landscape, where the adaptive value of a given mutation might vary. Once more, this speaks of the enormous adaptive capacity of RNA viruses: a single viral clone is able, at each passage, to generate an optimized though different quasispecies under the action of mutation and selection. Yuste et al. (2005) showed that a few mutations were sufficient to meditate fitness recovery of HIV clones that had been subjected to plaque-to-plaque transfers (Yuste et al. 1999; Yuste et al. 2000). Of all mutations observed in different clones, 25% were reversions, and 12 out of 20 mutations were located in the primer-binding-site loop for initiation of reverse transcription. In addition to the fitness gains observed in very debilitated viruses when subjected to massive passages, other studies reveal the possibility of gaining fitness between bottlenecks. Elena et al. (1998) observed that several subclones could regain fitness after having been subjected to additional bottlenecks. The authors explained this result on the basis of the heterogeneity of the viral population inside a plaque. It had been generally accepted that transfers through small population sizes precluded the appearance of advantageous mutations, but this study demonstrated the possibility of the generation of beneficial mutants that could be selected for the next transfer due to its faster replication. The possibilities of observing fitness rise after a bottleneck increase when the virus is highly debilitated. This is what happens in the stationary state reached by FMDV clones (Escarmís et al. 2002). In this case, the large number of bottlenecks experienced by the virus possibly leads to the minimal average fitness permitted under the transmission regime of plaque-to-plaque transfers. However, the virus still possesses an enormous capacity to recover from the negative effect of the serial bottlenecks, probably due to the selection of mutants with advantageous mutations. RNA viruses are robust enough to find solutions to outweigh the negative effect of the accumulated mutations. The transmission regime of plaque-to-plaque transfers introduces a positive bias, since it selects virus able to form plaques, and thus isolates genomes with a particular combination of mutations such that the virus is viable. It is likely that many genomes become extinct during the development of a plaque due to the accumulation of deleterious mutations, but epistatic interactions and compensatory mutations might permit the appearance of genomes which still maintain the ability to generate a progeny. As shown in Fig. 4, debilitated FMDV clones (Escarmís et al. 2002) can recover fitness in a single plaque transfer. In most clones shown in Fig. 4, the recovery of viral fitness in consecutive transfers was accomplished by reduction of the length of the elongated poly A tract. In one clone, the recovery of plaque titer was caused by a change of adenosine to guanosine in the internal poly A tract. These results show the great plasticity of RNA genomes.

162

C. Escarmís et al.

As a general conclusion, fitness in highly debilitated viruses can be recovered through massive passages, which permit selection processes to act efficiently. In addition to this generally accepted mechanism, fitness can also be recovered at consecutive bottlenecks. Our interpretation is that, when a virus has reached a low fitness state, the chances of selecting a virus with a compensatory mutation for the next transfer are enhanced. This permits sudden gains in fitness and leads to the fluctuating pattern in viral yield observed in FMDV (Lázaro et al. 2003, Manrubia et al. 2003).

8 Epistatic Interactions Among Mutations. The Role of Sex and Recombination There have been few experimental studies devoted to the analysis of epistatic effects among mutations. In a recent study (Sanjuán et al. 2004), the effect of a number of mutations was compared when they were present as single mutations or in pairs in the same genome. Between two deleterious mutations, interactions were mainly antagonistic, meaning that their combined effect is significantly smaller than expected under a simple multiplicative model. This fact can partly account for the nonlinear fitness loss in the study carried out with FMDV clones, although this explains neither the arrest of the population around a mean fitness value (and thus its resistance to extinction) nor the strong fluctuations observed. In addition, it has to be considered that, under the plaque-to-plaque transmission regime, many mutations are accumulated, and the effect of their interactions on fitness is much more difficult to quantify than in a genome with only two mutations. In the same study of Sanjuán and co-workers, antagonistic epistasis between beneficial mutations was also found. These results imply that recombination and sex in RNA viruses would not necessarily result in an immediate adaptive benefit. Other experiments on epistatic interactions after mutation accumulation in RNA viruses have used bacteriophage φ6 (Burch and Chao 2004) and FMDV (Elena 1999). The former study found that viral genomes with low fitness were less sensitive to deleterious mutations, pointing to antagonistic epistasis. The latter study failed to find epistasis. It would be very interesting to continue the study of Sanjuán et al. (2004) with genomes containing both types of mutations, deleterious and advantageous. We have reviewed some theoretical models which essentially demonstrate that antagonistic epistasis slows down Muller’s ratchet, while synergistic epistasis accelerates it (Charlesworth 1993; Kondrashov 1994; Colato and Fontanari 2001). A more complete understanding of the role of mutations and their

Population Bottlenecks in Quasispecies Dynamics

163

interactions in RNA genomes requires the explicit introduction of (at least) the secondary structure of the molecules. Few studies have been devoted to this question to date, though the results obtained are extremely interesting. Analysis of the population structure in sequence space by explicitly considering the secondary structure of RNA sequences of 76 nucleotides reveals that the sequences represented in the population are clustered (that is to say, form groups differing in one or two point mutations) around a small number of variants (which differ in several to many point mutations), all of them folding in the same secondary structure (Huynen et al. 1996). The fraction of the sequence space explored by a finite population strongly conditions its adaptability to new environments (represented in the model by a different secondary structure). Algorithms for folding RNA into its secondary structure predict that the fitness landscape is extremely rugged and that there are extended connected networks of sequences with identical structure. When secondary structure is taken into account, a clear prevalence of antagonistic epistasis in RNA secondary structure folding is observed, together with the appearance of mutations at a long distance in the sequence with compensatory effects (Wilke et al. 2003).

9 Adaptability of Viral Populations with a Long Bottleneck History It has been pointed several times throughout this review that systematic bottlenecks force an ensemble of viral clones to move through the genotypic (and phenotypic) landscape as the number of transfers increases. These viral populations continue accumulating mutations at a constant rate while eventually maintaining a constant average fitness value. However, an open question is whether the mutational load has consequences in the adaptability of these populations. A VSV strain obtained after successive bottlenecks showing a fitness equal to that of the wild type, always lost in long-term competition experiments with the wild type (Quer et al. 1996; 2001). The accumulation of mutations, as a consequence of the repeated bottlenecks, has been thought to be the reason for this lower adaptability. In order to test this explanation, an additional experiment was carried out. Several populations were subjected to repeated bottlenecks and allowed to achieve a relative fitness equal to one through additional large population passages (Novella 2004). These populations were competed over 79 passages with the wild-type virus, and they were shown unable to out-compete the wild type, apparently reflecting a lower adaptive ability. A striking observation is that some clones experienced a period of

164

C. Escarmís et al.

higher fitness than the wild type but, as the number of passages progressed, and in all the cases analyzed, the wild type performed eventually better than the bottlenecked virus. The authors of this study argue that mutant populations have a lower beneficial mutation rate than the wild type (Novella 2004). It is likely that a full analysis of the structure of the new quasispecies generated after the bottlenecks can give a more complete answer to this question. Maybe it is the structure of the quasispecies what confers higher or lower adaptability to a population, as theoretical studies suggest (Huynen et al. 1996; Wilke et al. 2001). It would also be very interesting to compare the adaptive ability to a new environment of clones that, having the same fitness, differ in the number of accumulated mutations. These experiments are nowadays possible, since the research of Escarmís and co-workers has produced a number of viral clones with these characteristics. It might be that the bottlenecked clones display lower adaptability, as shown by Novella (2004), when forced to compete with the wild type in an environment where the latter has been optimized. However, this does not imply that a similar result would be found if both viruses compete in a new environment.

10 Final Remarks Many of the results shown in this review indicate a great resistance of RNA viruses to extinction through the accumulation of mutations upon repeated bottlenecks. However, there are many experiments documenting extinction in viruses experiencing an enhanced replication error rate due to the action of mutagens (Sierra et al. 2000; Pariente et al. 2001; Crotty et al. 2001; Grande-Pérez et al. 2002; Severson et al. 2003). This result agrees with early molecular evolution theories (Eigen and Schuster 1979; Swetina and Schuster 1982; Eigen and Biebricher 1988, Eigen 2002) postulating that viral replication operates very close to the error threshold. When this limit is crossed, the population enters into “error catastrophe”, and the genetic information is lost. Near this threshold, RNA viruses can maintain a population structure organized in quasispecies, with maximal variability, and from which fittest viruses in a given environment can be rapidly selected. Nevertheless, when this limit is crossed the population disorganizes and forms an ensemble of random sequences unable to maintain the genetic information. Bottlenecks can positively select and isolate particles that still keep the ability to infect cells. Thanks to the bottleneck, these minority particles are separated from the unstructured group of mutants and allowed to generate a new popula-

Population Bottlenecks in Quasispecies Dynamics

165

tion without the interfering effect of a highly complex quasispecies (see also chapter by Domingo et al., this volume). It is, however, difficult to assess the relevance that bottlenecks can have in vivo. In nature, population bottlenecks are very frequent during the transmission of many viruses and may affect the severity of disease outbreaks. In the case of vertically transmitted viruses, bottlenecks may be particularly severe, since only a small amount of viruses are able to cross the barriers necessary to infect the embryo. In addition, competition among genomes can only take place inside the infected host, so the action of selection is highly reduced. Bottlenecks may also be frequent in horizontally transmitted viruses and during the intra-organ transmissions inside an infected organism. Under horizontal transfer, each virus can infect every susceptible individual of the host population, and competition can also happen at the inter-host level (Wilson et al. 1992; Chao et al. 2000). In HIV, newly infected individuals typically contain a smaller diversity of sequences than long-term infected patients, suggesting that, initially, a low number of viral particles was transmitted (Nowak et al. 1991; Pang et al. 1992). Bottlenecks also occur in many diseases transmitted via respiratory droplets, because most droplets only contain one or a few virions (Artenstein et al. 1966; Gerone et al. 1966). Population bottlenecks have been suggested to be partly responsible for the observed evolution of poliovirus in an individual (Hovi et al. 2004) and have been invoked as an explanation for the extinction and rapid emergence of strains of dengue 3 virus during an interepidemic period (Wittke et al. 2002). Theoretically, it was demonstrated that while fitness weakly declines under horizontal transmission of viruses, vertically transmitted viruses are affected by dramatic decreases in fitness, since this form of transmission involves frequent bottlenecks (Bergstrom et al. 1999). This provides an alternative explanation to Ewald’s theory (Ewald 1987), which claims that vertically transmitted viruses evolve to lower virulence because reproduction of the pathogen is limited by the reproductive success of the host. Population bottlenecks are one out of many factors determining the structure of a viral quasispecies and its characteristics. Further analysis, both theoretical and experimental, is required in order to disentangle the relative roles played by the different mechanisms involved in the evolution and adaptation of viral populations. Mutation rates are probably not independent of the transmission mode used by viruses in nature. The optimum mutation rate for a certain virus may be selected according to the characteristics of the environment where the viral population evolves and adapts (Earl and Deem 2004). In particular, this optimum may depend on the frequency of transmission through bottlenecks and on the degree of intra- and inter-host optimization that takes place before the next bottleneck occurs. These parameters acting in close concert confer viral

166

C. Escarmís et al.

populations their plasticity and robustness and prevent an easy extinction of viruses in nature.

References Andersson DI, Hughes D (1996) Muller’s ratchet decreases fitness of a DNA-based microbe. Proc Natl Acad Sci U S A 93:906–907 Artenstein M, Miller WS (1966) Air sampling for respiratory disease agents in army recruits. Bacteriol Rev 30:571–572 Ayala FJ, Escalante A, O’Huigin C, Klein J (1994) Molecular genetics of speciation and human origins. Proc Natl Acad Sci U S A 91:6787–6794 Bachtrog D, Gordo I (2004) Adaptive evolution of asexual populations under Muller’s ratchet. Evolution 58:1403–1413 Batschelet E, Domingo E, Weissmann C (1976) The proportion of revertant and mutant phage in a growing population, as a function of mutation and growth rate. Gene 1:27–32 Bell G (1988a) Recombination and the immortality of the germ line. J Evol Biol 1:67–82 Bell G (1988b) Sex and death in protozoa. The history of an obsession. Cambridge University Press, Cambridge Bergstrom CT, Pritchard JK (1998) Germline bottlenecks and the evolutionary maintenance of mitochondrial genomes. Genetics 149:2135–2146 Bergstrom CT, McElhany P, Real LA (1999) Transmission bottlenecks as determinants of virulence in rapidly evolving pathogens. Proc Natl Acad Sci U S A 96:5095–5100 Bull JJ, Badgett Mr, Rokyta D, Molineux IJ (2003) Experimental evolution yields hundreds of mutations in a functional viral genome. J Mol Evol 57:241–248 Burch CL, Chao L (2004) Epistasis and its relationship to canalization in the RNA virus phi 6. Genetics 167:559–567 Chao L (1990) Fitness of RNA virus decreased by Muller’s ratchet. Nature 348:454–455 Chao L, Hanley KA, Burch CL, Dahlberg C, Turner PE (2000) Kin selection and parasite evolution: higher and lower virulence with hard and soft selection. Quart Rev Biol 75:261–275 Charlesworth D, Morgan MT, Charlesworth B (1993) Mutation accumulation in finite outbreeding and inbreeding populations. Genet Res 61:39–56 Clarke D, Duarte E, Moya A, Elena S, Domingo E, Holland JJ (1993) Genetic bottlenecks and population passages cause profound fitness differences in RNA viruses. J Virol 67:222–228 Colato A, Fontanari JF (2001) Soluble model for the accumulation of mutations in asexual populations. Phys Rev Lett 87:238102 Crotty S, Cameron CE, Andino R (2001) RNA virus error catastrophe: direct molecular test by using ribavirin. Proc Natl Acad Sci U S A 98:6895–6900 Crow JF, Kimura M (1970) An introduction to population genetics theory. Harper & Row, New York De la Peña M, Elena SF, Moya A (2000) Effect of deleterious mutation-accumulation of the fitness of RNA bacteriophage MS2. Evolution 54:686–691

Population Bottlenecks in Quasispecies Dynamics

167

Drake JW, Holland JJ (1999) Mutation rates among RNA viruses. Proc Natl Acad Sci U S A 96:13910–13913 Duarte E, Clarke D, Moya A, Domingo E, Holland J (1992) Rapid fitness losses in mammalian RNA virus clones due to Muller’s ratchet. Proc Natl Acad Sci U S A 89:6015–6019 Duarte EA, Clarke DK, Elena SF, Moya A, Domingo E, Holland JJ (1993) Manytrillionfold amplification of single RNA virus particles fails to overcome the Muller’s ratchet effect. J Virol 67:3620–3623 Earl DJ, Deem MW (2004) Evolvability is a selectable trait. Proc Natl Acad Sci U S A 101:11531–11536 Eigen M (2002) Error catastrophe and antiviral strategy. Proc Natl Acad Sci U S A 99:13374–13376 Eigen M, Biebricher CK (1988) Sequence space and quasispecies distribution. In: E Domingo, JJ Holland, Ahlquist P (eds) Variability of RNA genomes, 1st edn. Vol III. CRC Press, Boca Raton, pp 211–245 Eigen M, Schuster P (1979). The hypercycle: a principle of natural self-organization. Springer-Verlag, Berlin Heidelberg New York Elena SF (1999) Little evidence for synergism among deleterious mutations in a nonsegmented RNA virus. J Mol Evol 49:703–707 Elena SF, Moya A (1999) Rate of deleterious mutation and the distribution of its effects on fitness in vesicular stomatitis virus. J Evol Biol 12:1078–1088 Elena SF, González-Candelas F, Novella IS, Duarte EA, Clarke DK, Domingo E, Holland JJ, Moya A (1996) Evolution of fitness in experimental populations of vesicular stomatitis virus. Genetics 142:673–679 Elena SF, Dávila M, Novella IS, Holland JJ, Domingo E, Moya A (1998) Evolutionary dynamics of fitness recovery from the debilitating effects of Muller’s ratchet. Evolution 52:309–314 Escarmís C, Dávila M, Charpentier N, Bracho A, Moya A, Domingo E (1996) Genetic lesions associated with Muller’s ratchet in an RNA virus. J Mol Biol 264:255–267 Escarmís C, Dávila M, Domingo E (1999) Multiple molecular pathways for fitness recovery of an RNA virus debilitated by operation of Muller’s ratchet. J Mol Biol 28:495–505 Escarmís C, Gómez-Mariano G, Dávila M, Lázaro E, Domingo E (2002) Resistance to extinction of low fitness virus subjected to plaque-to-plaque transfers: diversification by mutation clustering. J Mol Biol 315:647–661 Ewald PW (1987) Transmission modes and evolution of the parasitism-mutualism continuum. Ann N Y Acad Sci 503:295–306 Felsenstein J (1974) Evolutionary advantage of recombination. Genetics 78:737–756 Fisher RA (1922) On the dominance ratio. Proc R Soc Edin 42:321–341 Fisher RA (1930) The genetical theory of natural selection, 2nd edn. Clarendon, New York Froissart R, Wilke CO, Montville R, Remold SK, Chao L, Turner PE (2004) Co-infection weakens selection against epistatic mutations in RNA viruses. Genetics 168:9–19 Fry JD, Keightley PD, Heinsihn SL, Nuzhdin SV (1999) New estimates of the rates and effects of mildly deleterious mutation in Drosophila melanogaster. Proc Natl Acad Sci U S A 96:574–579

168

C. Escarmís et al.

Funchain P, Yeung A, Stewart JL, Lin R, Slupska MM, Miller JH (2000) The consequences of growth of a mutator strain of Escherichia coli as measured by loss of function among multiple gene targets and loss of fitness. Genetics 154:959–970 Gerone PJ, Couch RB, Ketter GV, Douglas RG, Derrenbacher EB, Knight V (1966) Assessment of experimental and natural viral aerosols. Bacteriol Rev 30:576–588 Gordo I, Charlesworth B (2000) The degeneration of asexual haploid populations and the speed of Muller’s ratchet. Genetics 154:1379–1387 Grande-Pérez A, Sierra S, Castro MG, Domingo E, Lowenstein PR (2002) Molecular indetermination in the transition to error catastrophe: systematic elimination of lymphocytic choriomeningitis virus through mutagenesis does not correlate linearly with large increases in mutant spectrum complexity. Proc Natl Acad Sci U S A 99:12938–12943 Haigh J (1978) Accumulation of deleterious genes in a population – Muller’s ratchet. Theor Pop Biol 14:251–267 Haldane JBS (1924) A mathematical theory of natural and artificial selection. Part I. Trans Camb Phil Soc 23:19–41 Haldane JBS (1927) A mathematical theory of natural and artificial selection. V. Selection and mutation. Proc Camb Phil Soc 23:838–844 Harpending H, Rogers A (2000) Genetic perspectives on human origin and speciation. Annu Rev Genomics Human Genet 1:361–385 Hovi T, Lindholm N, Savolainen C, Stenvik M, Burns C (2004) Evolution of wild-type 1 poliovirus in two healthy siblings excreting the virus over a period of 6 months. J Gen Virol 85:369–377 Huynen MA, Stadler PF, Fontana W (1996) Smoothness within ruggedness: the role of neutrality in adaptation. Proc Natl Acad Sci U S A 93:397–401 Kibota TT, Lynx M (1996) Estimate of the genomic mutation rate deleterious to overall fitness in E. coli. Nature 381:694–696 Kimura M, Maruyama T (1966) The mutational load with epistatic gene interactions in fitness. Genetics 54:1337–1351 Kondrashov AS (1982) Selection against harmful mutations in large sexual and asexual populations. Genet Res 40:325–332 Kondrashov AS (1994) Muller’s ratchet under epistatic selection. Genetics 136:1469– 1473 Lázaro E, Escarmís C, Domingo E, Manrubia SC (2002) Modeling viral genome fitness evolution associated with serial bottleneck events: evidence of stationary states of fitness. J Virol 76:8675–8681 Lázaro E, Escarmis C, Pérez-Mercader J, Manrubia SC, Domingo E (2003) Resistance of virus to extinction on bottleneck passages: study of a decaying and fluctuating pattern of fitness loss. Proc Natl Acad Sci U S A 100:10830–10835 Lea S, Hernández J, Blakemore W, Bricchi E, Curry S, Domingo E, Fry E, Ghazaleh RA, King A, Newman J, Stuart D, Mateu MG (1994) The structure and antigenicity of a type C foot-and-mouth disease virus. Structure 2:123–139 Lynch M, Gabriel W (1990) Mutation load and the survival of small populations. Evolution 44:1725–1737 Maisnier-Patin S, Berg OG, Liljas L, Andersson DI (2002) Compensatory adaptation to the deleterious effect of antibiotic resistance in Salmonella typhimurium. Mol Microbiol 46:355–366

Population Bottlenecks in Quasispecies Dynamics

169

Manrubia SC, Lázaro E, Pérez-Mercader J, Escarmís C, Domingo E (2003) Fitness distributions in exponentially growing asexual populations. Phys Rev Lett 90:188102 Manrubia SC, Escarmís C, Domingo E, Lázaro E (2005) High mutation rates, bottlenecks, and robustness of RNA viral quasispecies. Gene 347:273–282 Mason PW, Grubman MJ, Baxt B (2003) Molecular basis of pathogenesis of FMDV. Virus Res 91:9–32 Maynard Smith J (1978) The evolution of sex. Cambridge University Press, Cambridge Muller HJ (1964) The relation of recombination to mutational advance. Mut Res 1:2–9 Novella IS (2004) Negative effect of genetic bottlenecks on the adaptability of vesicular stomatitis virus. J Mol Biol 336:61–67 Novella IS, Ebendick-Corpus BE (2004) Molecular basis of fitness loss and fitness recovery in vesicular stomatitis virus. J Mol Biol 342:1423–1430 Novella IS, Elena SF, Moya A, Domingo E, Holland JJ (1995a) Size of genetic bottlenecks leading to virus fitness loss is determined by mean initial population fitness. J Virol 69:2869–2872 Novella IS, Duarte EA, Elena SF, Moya A, Domingo E, Holland JJ (1995b) Exponential increases of RNA virus fitness during large population transmissions. Proc Natl Acad Sci U S A 92:5841–5844 Novella IS, Quer J, Domingo E, Holland JJ (1999) Exponential fitness gains of RNA virus populations are limited by bottleneck effects. J Virol 73:1668–1671 Nowak MA, Anderson, RM, Mc Lean AR, Wolfs TF, Goudsmit J, May RM (1991) Antigenic diversity thresholds and the development of AIDS. Science 254:963– 969 Orr HA (2003) The distribution of fitness effects among beneficial mutations. Genetics 163:1519–1526 Pang S, Schlesinger Y, Daar ES, Moudgil T, Ho DD, Chen ISY (1992) Rapid generation of sequence variation during primary HIV-infection. AIDS 6:453–460 Pariente N, Airaksinen A, Domingo E (2003) Mutagenesis versus inhibition in the efficiency of extinction of foot-and-mouth disease virus. J Virol 77:7131–7138 Peck JR (1994) A ruby in the rubbish: beneficial mutations, deleterious mutations and the evolution of sex. Genetics 137:597–606 Quer J, Huerta R, Novella IS, Tsimring L, Domingo E, Holland JJ (1996) Reproducible nonlinear population dynamics and critical points during replicative competitions of RNA virus quasispecies. J Mol Biol 264:465–471 Quer J, Hershey CL, Domingo E, Holland JJ, Novella IS (2001) Contingent neutrality in competing viral populations. J Virol 75:7315–7320 Rouzine IM, Wakeley J, Coffin JM (2003) The solitary wave of asexual evolution. Proc Natl Acad Sci U S A 100:587–592 Sanjuán R, Moya A, Elena SF (2004) The contribution of epistasis to the architecture of fitness in an RNA virus. Proc Natl Acad Sci U S A 101:15376–15379 Severson WE, Schmaljohn CS, Javadian A, Jonsson CB (2003) Ribavirin causes error catastrophe during Hantaan virus replication. J Virol 77:481–488 Sierra S, Dávila M, Lowenstein PR, Domingo E (2000) Response of foot-and-mouth disease virus to increased mutagenesis. Influence of viral load and fitness in loss of infectivity J Virol 74:8316–8323

170

C. Escarmís et al.

Stephan W, Chao L, Smale J (1993) The advance of Muller’s ratchet in a haploid asexual population: approximate solution based on diffusion theory. Genet Res 61:225–232 Swetina J, Schuster P (1982) Self-replication with errors. A model for polynucleotide replication. Biophys Chem 16:329–345 Vassilieva LL, Hook AM, Lynch M (2000) The fitness effects of spontaneous mutations in Caenorhabditis elegans. Evolution 54:1234–1246 Wagner GP, Gabriel W (1990) Quantitative variation in finite parthenogenetic populations: what stops Muller’s ratchet in the absence of recombination? Evolution 44:715–731 Weaver SC, Brault AC, Kang W, Holland JJ (1999) Genetic and fitness changes accompanying adaptation of an arbovirus to vertebrate and invertebrate cells. J Virol 73:4316–4326 Wilke CO (2003) Probability of fixation of an advantageous mutant in a viral quasispecies. Genetics 163:467–474 Wilke CO, Wang JL, Ofria C, Lenski RE, Adami C (2001) Evolution of digital organisms at high mutation rates leads to survival of the flattest. Nature 412:331–333 Wilke CO, Lenski RE, Adami C (2003) Compensatory mutations cause excess of antagonistic epistasis in RNA secondary structure folding. BMC Evol Biol 3:3 Wilson DS, Pollock GB, Dugatkin LA (1992) Can altruism evolve in purely viscous populations? Evol Ecol 6:331–341 Wittke V, Robb TE, Thu HM, Nisalak A, Nimmanniyya S, Kalayanrooj S, Vaughn DW, Endy TP, Holmes EC, Aaskov JG (2002) Extinction and rapid emergence of strains of dengue 3 virus during an interepidemic period. Virology 301:148–156 Wright S (1931) Evolution in Mendelian populations. Genetics 16:97–159 Wright S (1939) Statistical genetics in relation to evolution. Paris: Hermann Yuste E, López-Galindez C, Domingo E (2000) Unusual distribution of mutations associated with serial bottleneck passages of human immunodeficiency virus type 1. J Virol 74:9546–9552 Yuste E, Sánchez-Palomino S, Casado C, Domingo E, López-Galíndez C (1999) Drastic fitness loss in human immunodeficiency virus type 1 upon serial bottleneck events. J Virol 73:2745–2751 Yuste E, Bordería AV, Domingo E, López-Galíndez C (2005) Few mutations in the 5 leader region mediate fitness recovery of debilitated human immunodeficiency type 1 viruses. J Virol 79:5421–5427 Zeyl C, Mizesko M, de Visser JA (2001) Mutational meltdown in laboratory yeast populations. Evolution 55:909–917 Zhang X-S, Wang J, Hill WG (2004) Redistribution of gene frequency and changes of genetic variation following a bottleneck in population size. Genetics 167:1475– 1492

CTMI (2006) 299:171–192 c Springer-Verlag Berlin Heidelberg 2006


Evolutionary Dynamics of HIV-1 and the Control of AIDS J. I. Mullins (u) · M. A. Jensen Departments of Microbiology, University of Washington School of Medicine, Seattle, WA 98195–8070, USA [email protected]

1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171

2

The Natural History of HIV-1 Diversity . . . . . . . . . . . . . . . . . . . . . . . . 172

3

HIV Diversity and Pathogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176

4

HIV Diversity, Latency, and Drug Treatment . . . . . . . . . . . . . . . . . . . . . 179

5

HIV Diversity and Immune Response . . . . . . . . . . . . . . . . . . . . . . . . . . 181

6

Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184 Abstract Human immunodeficiency viruses (HIV) have exhibited an extraordinary capacity for genetic change, exploring new evolutionary space after each transmission to a new host. This presents a great challenge to the prevention and management of HIV-1 infection. At the same time, the relentless diversification of HIV-1, developing as it does under the constraints imposed by the human immune system and other selective forces, contains within it information useful for understanding HIV epidemiology and pathogenesis. Comparing the sheer mutational potential of HIV with actual data representing viral lineages that can survive selection suggests that HIV does not have unlimited capacity for change. Rather, clinical and bioinformatic data suggest that, even in the most diverse gene of the most highly variable organism, natural selection places severe limits on the portion of amino acid sequence space that ensures viability. This suggests some optimism for those attempting to identify sets of antigens that can generate effective humoral and cellular immune responses against HIV.

1 Introduction Since our initial recognition of type 1 human immunodeficiency viruses (HIV-1), these viruses have exhibited an extraordinary capacity for genetic change, exploring new evolutionary space after each transmission to a new

172

J. I. Mullins · M. A. Jensen

host. This presents a great challenge to the prevention and management of HIV-1 infection. At the same time, the relentless diversification of HIV-1, developing as it does under the constraints imposed by the human immune system and other selective forces, contains within it information useful for understanding HIV epidemiology and pathogenesis. At a point of considerable pessimism over the prospects for an effective vaccine (Desrosiers 2004), HIV-1 nonetheless provides enough information, as sequence variation arising in response to genetically varying hosts, to potentially allow the rational design of vaccine antigens that might finally moderate the spread of the virus, and slow the development of AIDS.

2 The Natural History of HIV-1 Diversity The rate of progression of HIV-1 infection to AIDS is highly variable among individuals, developing an average of 9–10 years after infection in most adults in the absence of antiretroviral therapy (Muñoz et al. 1997). The progress of disease does follow a general course, however. This can be illustrated by temporal changes in two of the strongest correlates of disease: virus load, defined by the quantification of viral RNA molecules in peripheral blood plasma; and the number of circulating CD4+ and CD8+ T lymphocytes (Fig. 1a). During acute infection, viral load peaks at up to 6–8 log/ml after about 2 weeks (Daar et al. 1991), while CD4+ cell numbers drop precipitously. This period of rapid viral replication is sometimes accompanied by a syndrome of nonspecific flu-like symptoms, the occurrence of which is related to faster disease progression (Schacker et al. 1998). Viral replication comes under control as signs of immune responses are detected (Cooper et al. 1987; Koup et al. 1994), and typically levels off to between 3–5 log/ml, while CD4 counts recover, often to normal levels (Ho et al. 1985). Subsequently, viral load and total T cell count can remain more or less constant, reflecting a period of T cell homeostasis in which CD4+ counts are often in decline while CD8 counts rise proportionately (Margolick et al. 1994). T cell homeostasis can persist for a period of up to a decade or more, during which the individual may exhibit few overt symptoms of illness. The level at which viral load is maintained during this early, asymptomatic period, known as the set-point, is highly predictive of the length of the asymptomatic period and AIDS-free survival (Mellors et al. 1997) (Fig. 1a). Ultimately, however, CD4+ cell levels in the blood decline severely, viral loads increase, CD8+ cell levels drop (Margolick and al. 1995), and clinical AIDS develops (Fig. 1a).

Evolutionary Dynamics of HIV-1 and the Control of AIDS

173

HIV genetic diversification within the host also follows a course that includes common milestones (Fig. 1b). The behavior of the envelope gene has been most thoroughly studied. Very low levels of diversity are typically found early following seroconversion in men and women infected through sex and IV drug abuse (Zhang et al. 1993; Delwart et al. 1994; Shpaer et al. 1994; Grobler et al. 2004). However, a substantial fraction of women infected with HIV-1 clades A and D have been reported to have higher levels of diversity early in infection (Long et al. 2000; Sagar et al. 2003). After this bottleneck, variation in env expands at a linear rate of approximately 1% per year in untreated individuals (Shankarappa et al. 1999). Diversity levels then stabilize, or actually shrink, at a point that precedes catastrophic immune decline (Fig. 1b). Divergence from the initially sampled sequence population proceeds apace with diversification, generally continuing beyond the point at which diversity peaks, but may flatten out late in infection (Shankarappa et al. 1999) (Fig. 1b). Recent studies have added perspective to the viral population changes found early in infection and revealed some of the selective forces shaping those populations. Over the first 3 months of infection, variability within the env gene in a diverse inoculum is winnowed down almost completely in men infected with clade B (Learn et al. 2002) (Fig. 1c). Interestingly, this homogenization has been observed in env, but not in gag, suggesting stronger selective pressure to be acting on env early in infection (Zhang et al. 1993; Learn et al. 2002). Then, over the first 3 years of infection, the viral population changes seem to reflect opposing forces (Herbeck et al., personal communication). The consistently paced forward evolution (Shankarappa et al. 1999) most visibly results from the effect of mutations leading to immunologic escape from cytotoxic T lymphocyte (CTL) responses (Allen et al. 2004; Leslie et al. 2004; Liu et al., personal communication). However, env sequences evolve toward a node at the root of the phylogenetic clade of viral sequences within the patient, referred to the most recent common ancestor (MRCA). This continues for 2–3 years during which some ancestral features are recovered (Herbeck et al., personal communication). As a result, the net viral divergence relative to other HIV strains is reversed or at least balances the forward evolution (Fig. 1c). This recovery of ancestral features appears to reflect loss of immunologic escape mutations and attendant restoration of fitness lost as a result of these escapes in the previous host (Herbeck et al., personal communication). The amino acid changes that define this recovery are at least in part defined by differences in donor and recipient HLA types (Moore et al. 2002; Kiepiela et al. 2004; Tang et al. 2004). HLA molecules encoded by distinct alleles have a restricted capacity to present short antigenic peptides to T cells. This limits the epitopes that can be recognized in each host. Changes in glycosylation patterns associated with

174

J. I. Mullins · M. A. Jensen

escape from host antibody responses also occur early in infection (Derdeyn et al. 2004) and may also be associated with “resetting” the viral population. Despite the large census viral population size that develops during acute HIV-1 infection, these virus particles are produced from a small effective

Evolutionary Dynamics of HIV-1 and the Control of AIDS

175

Fig. 1a–c Predictors for the development of AIDS and death in HIV-1 infected men from the United States. a Classical patterns of CD4 cell and viral load properties over infection are shown. Viral load and survival data taken from Mellors et al. 1997. Shaded rectangles at the left and right sides of the figure delineate the periods of acute HIV infection and AIDS, respectively. b Consistent viral evolutionary trends in the development of AIDS. (Data taken from Shankarappa et al. 1999). c Recent refinements of our understanding of the viral evolutionary trends in HIV infection. These include early homogenization of viral env sequences (Learn et al. 2002), evolutionary convergence to population ancestor sequence (Herbeck et al., personal communication), and the appearance of dual tropic, R5X4 HIV-1 strains prior to fulminate outgrowth of X4 strains (Jensen et al. 2003; Jensen and van ’t Wout 2003)

population, on the order of 103 producer cells (Leigh-Brown 1997; Achaz et al. 2004; Shriner et al. 2004b). This low effective population size implies that relatively strong selection is required to influence genetic diversity, e.g., selection applied by the immune system, changes in host cell target identity, distribution and susceptibility over infection, or the administration of drug treatment. The reverse implication is that such forces are overcoming equally strong selection to maintain the amino acids that were originally present at high frequency, evidently because these strongly influence viral replicative fitness (Quinones-Mateu et al. 2000; Quinones-Mateu and Art 2002). This can result in the rapid reversion of drug-resistance mutations or outgrowth of sensitive virus upon treatment hiatus (Birk et al. 2001; Deeks et al. 2001; Frenkel and Mullins 2001), and as described above, of immune escape mutations. These estimates may in fact be too high, since recombination between

176

J. I. Mullins · M. A. Jensen

viral genomes are estimated to be greater than the rate of fixation of point mutations (Rhodes et al. 2003; Shriner et al. 2004a). Thus, recombination is becoming recognized as a force that greatly enhances apparent viral diversity (Liu et al. 2002; Shriner et al. 2004a), which facilitates development of antigenic diversity and drug resistance (Bretscher et al. 2004). Development of diversity, at least in env, seems to proceed rather uniformly through time when considered as a bulk property, i.e., as nucleotide polymorphisms averaged over all sites, and when compared to the initially infecting strain (Shankarappa et al. 1999) or the donor’s virus (Liu et al., personal communication). Particular changes arise stochastically, and to the extent that they initiate or modulate disease processes, the random appearance of such mutations may play a role (Shriner et al. 2004b), along with host genetic limiting factors (O’Brien and Nelson 2004), in explaining the variability of HIV disease. We discuss some of these possibilities in the next sections.

3 HIV Diversity and Pathogenesis Can the variation in the rate of progression to AIDS be explained in any part by the development of viral genetic diversity? The answer appears to be yes, as the increase in env diversity has been correlated with disease milestones (Shankarappa et al. 1999) (Fig. 1b) and is a predictor of in vitro competitive fitness of primary HIV isolates (Quinones-Mateu and Arts 2002; Arts and Quinones-Mateu 2003). Further, the inhibition of HIV replication by antiretroviral therapy (ART) interrupts both disease progression and the accumulation of viral mutations (Gunthard et al. 1999; Zhu et al. 2000, 2002; Frenkel et al. 2003). Plus, there is substantial data in the literature indicating that viruses with differing co-receptor specificity play a critical role in the disease process. HIV-1 virus isolates were initially placed into as few as two phenotypic categories defined in vitro: non-syncytium-inducing (NSI) or syncytiuminducing (SI) in CD4+ T cell lines. Ultimately, the difference between the NSI and SI phenotypes was shown to be due largely to the differential use of chemokine receptors as co-receptors for viral entry. NSI viruses predominantly use CCR5, while SI viruses can use CCR5 and CXCR4, or CXCR4 exclusively (Björndal et al. 1997). Additional, minor co-receptors are also known to be used for cell entry (Karlsson et al. 2003). HIV-1 present at primary infection use the CCR5 co-receptor (R5 virus) approximately 90% of the time (Zhang et al. 1993; Zhu et al. 1993; van ’t Wout et al. 1994). A substantial proportion of individuals then go on to develop virus that uses the CXCR4

Evolutionary Dynamics of HIV-1 and the Control of AIDS

177

co-receptor (X4 virus). These X4 viruses are associated with accelerated CD4 decline and more rapid progression of HIV-1 disease (Koot et al. 1993). Little is known about the mechanisms by which X4 viruses come to predominate in infection. For example, it is not known whether X4 emergence is a primary pathogenic event or is secondary to some other event, i.e., whether the virus itself causes accelerated disease progression or another event causes the acceleration, perhaps then permitting X4 outgrowth. This could in turn lead to further acceleration of disease, as observed in the FeLV-AIDS model (Overbaugh et al. 1988; Mullins et al. 1991). Another important unanswered question is whether X4 viruses arise multiple times during the course of disease, and if so, why they do not become dominant whenever they emerge. There is also uncertainty about the frequency with which X4 viruses are associated with the development of AIDS. Phenotypic studies suggest that 50%–60% of progressing subjects acquire X4/SI virus (Tersmette et al. 1988, 1989; Koot et al. 1993). However, a detailed longitudinal genotypic study has noted the occurrence, often transient, of viruses of predicted X4 phenotype in nine out of nine individuals (Shankarappa et al. 1999). The latter work suggests that previous estimates were low and that culturable X4 virus may be lost by the time AIDS develops. Certain mutations, particularly in the V3 loop of env, are strongly associated with the X4 phenotype; in particular, basic amino acids at V3 positions 11 and 25 very frequently distinguish primary X4 from R5 viruses (Fouchier et al. 1992), with positions 24 and 27 implicated in some cases (de Jong et al. 1992; Milich et al. 1993). Many genetic changes not yet examined virologically are also likely to influence co-receptor usage (Resch et al. 2001; Hoffman et al. 2002; Jensen et al. 2003). However, a question pertinent to unraveling the cause vs effect role of X4 virus in disease is whether the appearance of these basic amino acids is necessary or sufficient to lead to the outgrowth of X4 virus, or whether instead a more gradual process of mutation accumulation takes place. Viruses that can use both co-receptors (R5X4 viruses) are known to arise around the time of R5 to X4 transition (Björndal et al. 1997; Connor et al. 1997; van Rij et al. 2000). We have shown that these dual tropic viral V3 sequences are intermediates along the evolutionary path to X4 formation (Figs. 1c and 2) using a position-specific site matrix scoring method (PSSM) (Jensen et al. 2003; Jensen and van ’t Wout 2003; Brumme et al. 2004). PSSMs are used to detect nonrandom distributions of amino acids at adjacent sites associated with empirically determined groupings of sequences. A PSSM facilitates comparison of the residues of a sequence fragment to a group of aligned sequences known to have the desired property, using background genetic variation as a baseline comparison or null model. The comparison leads

178

J. I. Mullins · M. A. Jensen

to a score that can be interpreted as a likelihood that the sequence fragment has the property of interest. The PSSM score can also act as a continuous indicator of X4 evolution, thus, by scoring reconstructed ancestors of the sampled viruses we demonstrated that the progression from low-scoring (R5-like) to

Fig. 2 Phylogenetic reconstructions of HIV-1 env sequences for subjects 2 (left), and 3 (right); from Shankarappa et al. 1999. Color of tip symbols indicate the years post seroconversion each sample was obtained. Color of nodes and branches reflect PSSM score of the reconstructed ancestors or the sample (tip) V3 sequences; blue/cooler colors are lower (R5-like), red/warmer colors are higher (X-like), as indicated by the scale. The extreme scores varied among subjects, so that the color scales differ among the trees. For each tree, however, light green is fixed at –5, approximately intermediate between the R5 and X4 cutoffs. Branches are colored according to the score of the sequence of the branch’s right-hand node. These figures are also available as appendices in association with Jensen et al. (2003)

Evolutionary Dynamics of HIV-1 and the Control of AIDS

179

high-scoring (X4-like) virus was typically gradual, transiting through R5X4 intermediates. Interestingly, the loss of putative X4 virus at the later stages of infection was found to occur by two different types of changes in the viral populations: extinction of the X4 lineage with outgrowth of pre-existing R5 lineages, and further evolution of the X4 lineage leading to recovery of R5 properties (Fig. 2). Co-variation among V3 sites is well-known and has been observed in subtype B (Korber et al. 1993; Bickel et al. 1996). The fact that sites can be considered independently, as with the PSSM, for the purposes of phenotype prediction suggests that certain cis combinations of amino acids are strongly selected against, in most individuals, so that these combinations are rarely seen in the data. This may explain, at least in part, the dearth of X4 viruses found early in infection. The gradual evolution of X4 viruses also suggests that R5 viruses are indeed selectively transmitted (Shpaer et al. 1994; Rodrigo 1997).

4 HIV Diversity, Latency, and Drug Treatment The most immediately practical implication of HIV evolution is the potential it creates for rapid development of resistance to antiretroviral drugs (Chen et al. 2005). HIV’s capacity for acquiring genetic resistance to antiviral drugs thwarted attempts to durably suppress viral infection until the advent of highly active antiretroviral therapy (HAART) (Palella et al. 1998). As multiple drug therapy has become the standard of clinical care, combination therapy is now referred to simply as antiretroviral therapy, or ART. Under such regimens, viral replication is suppressed to a degree such that the accumulation of multiple resistance mutations in a single viral lineage becomes extremely rare. As long as the patient continues therapy and viral load is controlled, disease can be delayed, possibly indefinitely. A continuing advantage of maintaining therapy, even in the presence of drug resistance, has been noted (Deeks et al. 2000), probably owing to a diminished replicative capacity of the resistant virus (Frenkel and Mullins 2001). Early dynamical models of proviral load under therapy suggested that the virus would eventually be cleared from patients (Perelson et al. 1996; Wein et al. 1997; Ho 1998). Unfortunately, it later became clear that ART could not achieve complete clearance of the virus (Chun et al. 1997; Finzi et al. 1999; Siliciano et al. 2003). Investigators realized that the decay of provirus, that is, viral DNA integrated into the genome of peripheral blood mononuclear cells (PBMCs), had two phases; one of rapid decline as viral load dropped to the level

180

J. I. Mullins · M. A. Jensen

of detection, and a subsequent phase of slow decline (Wei et al. 1995; Perelson et al. 1996). Whether a biphasic model is required to explain the data has been questioned (Holte et al., personal communication). Certain latently infected immune cells, including not only T cells but also macrophages and monocytes, can act as long-term reservoirs for the maintenance of HIV provirus. Estimates of the half-life of the so-called T cell latent reservoir in the peripheral blood vary widely, from 6 to 44 months or longer (Siliciano et al. 2003). The rapid evolution of drug resistance and the dynamics of the latent reservoir thus interact to create a difficult clinical problem. Even when replication of virus has been essentially stopped by therapy (Zhu et al. 2002; Frenkel et al. 2003), latent reservoirs or small pockets of replicating virus can seed viral regrowth upon therapy interruption at any time in the patient’s life (Schrager and D’Souza 1998). Since the threat of replication, and therefore resistance, never vanishes in most if not all patients (Hatano et al. 2000; Ruiz et al. 2000), they must adhere to the drug regimen with diligence. In addition, viral variants already generated over the course of infection, including those possessing resistance mutations, can be archived in the latent reservoir, escaping drug and immune surveillance, but with the capacity to later emerge. Reservoirs essentially sample the viral population over time, so that old viruses remain as a fifth column, though the onslaught of new variants is stemmed by ART (Nickle et al. 2003a, 2003b). Even under therapy, there remains a risk of developing multiple drug-resistant lineages of virus, through mutation or through recombination (Shriner et al. 2004a), during reverse transcription of singly-resistant or replication-defective and archived proviruses within the same cell. In the face of these evolutionary processes, ART treatment of HIVinfected individuals remains a subtly difficult task (Domingo et al. 1997), and treatment failure is frequent (Deeks et al. 2003). The persistence and evolutionary capacity of HIV means that most patients will have to remain on antiretroviral drugs for life. This presents another dilemma for the physician: the most effective drugs also have toxic or dysregulating side effects, and are not well tolerated over the long term by some individuals. Protease inhibitors, in particular, are associated with high levels of lipidemia (Jain et al. 2001; Mallon et al. 2002). One strategy for alleviating these effects is that of structured treatment interruption (STI) (Dybul et al. 2001; Miller 2001). In STI, patients who meet certain criteria for viral control under continuously administered therapy cease taking antiretroviral drugs for specified intervals, alternating with intervals of normal therapy. Some individuals who are treated with ART experience extended (months to years) periods of undetectable viral load upon structured treatment interruption (Rosenberg et al. 2000; Altfeld and Walker 2001; Oxenius and Hirschel 2003). In general, evidence of control has been more pronounced in indi-

Evolutionary Dynamics of HIV-1 and the Control of AIDS

181

viduals treated very early in their infection (Di Mascio et al. 2004), and this control is in most if not all cases of limited durability (Kaufmann et al. 2004). Surprisingly, neither the well-studied latent reservoir in peripheral blood mononuclear cells, nor the lymph node are major sources of the rebounding virus (Chun et al. 2000; Malaspina et al. 2002). Re-emergence of virus appears to result from multiple, as yet unidentified reservoirs (Kijak et al. 2002; Dybul et al. 2003), possibly including monocytes (Zhu et al. 2002) and tissue macrophages (Laco et al., personal communication). Several known drug-resistance mutations have been shown to reduce replicative fitness in vitro (Goudsmit et al. 1996; Nijhuis et al. 2001; Brenner et al. 2002; Kijak et al. 2002; Quinones-Mateu and Arts 2002; Menzo et al. 2003; Chen et al. 2004). This suggests that cyclical release of drug pressure that occurs in STI might result in an evolutionary clearance of drug resistance. The danger in this approach lies in possibly facilitating both the archiving of resistant virus, and in the development of compensatory mutations to achieve both resistance and a measure of recovered replication fitness (Tremblay et al. 2003). Compensatory mutations would reduce the selective pressure against resistance mutations, and so can lead to a mutational irreversibility of resistance, a phenomenon frequently observed in antibiotic resistance in bacteria.

5 HIV Diversity and Immune Response A prophylactic HIV vaccine is the holy grail of AIDS research, and a feeling that it is just as unattainable has grown in the research community (Desrosiers 2004). Two arguments fueling this pessimism are that: (a) natural immune responses rarely control HIV infection, and (b) immune responses to preexisting HIV infection do not preclude superinfection by other HIV strains (Altfeld et al. 2002; Jost et al. 2002; Yang et al. 2005). HIV also has active mechanisms for attenuating HIV-specific immune responses (Casartelli et al. 2003; Stoddart et al. 2003; Tanaka et al. 2003). However, a major hurdle for the immune system, and the one that vaccines are designed to overcome, is mustering specific responses to HIV’s constantly diversifying spectrum of antigens. There are many examples of human immune control of HIV over 2 decades. Although the mechanisms of protection are not well understood, the existence of these cases is reason for cautious optimism in the continuing vaccine research effort. Nonetheless, attenuation of viremia and disease course, as often occurs in macaque/SIV vaccine studies, and the expected attendant decrease in transmission rate, has replaced sterilizing immunity as the proximate goal of current HIV vaccine efforts (Mullins 1997).

182

J. I. Mullins · M. A. Jensen

About 5%–10% of individuals infected with HIV-1 as adults progress to AIDS in under 5 years (Muñoz et al. 1997) and are referred to as rapid progressors. In those infected as neonates, who have expanded CD4+ cell numbers yet immature immune responses, there is a higher incidence of rapid disease progression (Aldrovandi 1999). Individuals who seroconvert but who do not experience CD4+ decline and do not progress to disease even after spans of 15–20 years, the so-called long-term nonprogressors (LTNP), constitute approximately 5% of HIV-positive individuals (Muñoz et al. 1997). A more extreme example of virologic control is found in highly exposed but persistently seronegative individuals, some of which harbor extremely low levels of viral DNA in their peripheral blood cells, presumably reflecting abortive HIV infection since these sequences do not detectably evolve over periods of years (Zhu et al. 1999, 2003, 2004). There are also individuals who are substantially resistant to HIV infection. These include individuals with homozygous defects in the CCR5 co-receptor gene for R5 viruses (Huang et al. 1996; Michael et al. 1997), a few of which have become infected with X4 viruses (O’Brien et al. 1997; Theodorou et al. 1997). In the absence of recognized CCR5 or other relevant cellular gene defects, certain female sex workers (FSWs) (Fowke et al. 1996) and other high-risk individuals who are repeatedly exposed to different strains of HIV do not seroconvert. As well, some HIV-negative members of discordant couples do not acquire HIV infection for years, if at all (Bienzle et al. 2000). The instances in which immune control fails in LTNP or highly exposed seronegative individuals suggest that not only within-host evolution, but also standing, epidemiological diversity within the population of infected individuals plays a critical role in the persistence and spread of HIV. Interestingly, whereas constant exposure to multiple prevalent HIV strains may have afforded protection to some highly exposed FSWs, follow-up study showed that FSWs were more likely to become infected by their long-term partners after stopping work in the sex trade (Kaul et al. 2001). Any HIV vaccine then, evidently must not only deliver antigen effectively and augment immune response to it, but also generate broad responses that can confront standing diversity. To address the problem of HIV diversity, successful vaccines may require, in some form, not only a multiplicity of epitopes in several genes, but also a representation of the genetic diversity within those epitopes. The fear is that the number of epitopes necessary to station the immune system at all the evolutionary exits would be astronomical. Again, we may turn for ideas, if not comfort, to the evidence that HIV evolution is constrained by selection on replication to a relatively small subset of the vast number of sequence possibilities. On the host side, we know that variability in correlates of disease

Evolutionary Dynamics of HIV-1 and the Control of AIDS

183

progression, such as viral load set point, can be accounted for in significant measure by differences in HLA type between individuals (Moore et al. 2002; Trachtenberg et al. 2003; Kiepiela et al. 2004). If the potential for viral escape were limited only by the generation of mutants, there would be little difference in the progress or outcome of disease between HLA types. Viral genetic data highlight these constraints even more clearly. Results correlating standing genetic diversity to the positions of CTL epitopes indicate that selection on replicative efficiency acts strongly to counter immune pressure, in that conserved regions of the HIV genome are richest in known epitopes, while variable regions are low in known epitopes, and contain a high proportion of amino acids that cannot serve as epitope anchor residues (Yusim et al. 2002). Longitudinal patterns of HIV evolution are also compelling. Reversion of CTL escape mutations have now been described, both in SIV (Friedrich et al. 2004) and HIV (Allen et al. 2004; Friedrich et al. 2004; Leslie et al. 2004; Liu et al., personal communication) infections. Escape mutants from restricted epitopes that arose within donor individuals having a given HLA type revert to a susceptible epitope sequence in recipients with different HLA alleles. Reversion to a recognizable epitope in the absence of targeted immune pressure may reflect a common pattern of HIV evolution occurring upon transmission. In studies of env evolution early in infection, we have observed both overall declines in diversity compared to the inoculating population prior to seroconversion (Learn et al. 2002), as well as a tendency for the earliest viruses to be more divergent from the most recent common ancestor (MRCA) than some later viral populations (Herbeck et al., personal communication). The combination of a broad pattern of early decline and convergence and a specific mechanism of reversion suggest, circumstantially, strong sequence constraints that are important to viral reproduction across patients. This “sawtooth” evolution from donor to recipient, if it proves to be general, also suggests that the development of compensatory mutations that improve replicative fitness in an escaped genetic background may be rare, in contrast to the circumstances surrounding drug resistance. The immediately foregoing findings provide impetus to the development of vaccine immunogens that embody ancestral or consensus features of viruses circulating in a given population (for review, see Mullins et al. 2004).

6 Discussion There is no doubt that HIV is capable of rapid evolution, with severe clinical consequences. But comparing the sheer mutational potential of HIV with

184

J. I. Mullins · M. A. Jensen

actual data representing viral lineages that can survive selection, suggests that HIV does not have unlimited capacity for change. Each individual whose infection progresses from R5 to X4 virus represents an independent run of sequence evolution. In spite of high inherent variability in the process of mutation, these independently evolving populations attain a common phenotypic endpoint with extraordinary repeatability. Clinical and bioinformatic data suggest that, even in the most diverse gene of the most highly variable organism, natural selection places severe limits on the portion of amino acid sequence space that ensures viability. This suggests some optimism for those attempting to identify sets of antigens that can generate effective humoral and cellular immune responses against HIV.

References Achaz G, Palmer S, Kearney M, Maldarelli F, Mellors JW, Coffin JM, Wakeley J (2004) A robust measure of HIV-1 population turnover within chronically infected individuals. Mol Biol Evol 21:1902–1912 Aldrovandi GM (1999) Natural history of pediatric HIV disease. Lippincott Williams and Wilkins, Philadelphia Allen TM, Altfeld M, Yu XG, O’Sullivan KM, Lichterfeld M, Le Gall S, John M, Mothe BR, Lee PK, Kalife ET, Cohen DE, Freedberg KA, Strick DA, Johnston MN, Sette A, Rosenberg ES, Mallal SA, Goulder PJ, Brander C, Walker BD (2004) Selection, transmission, and reversion of an antigen-processing cytotoxic T-lymphocyte escape mutation in human immunodeficiency virus type 1 infection. J Virol 78:7069–7078 Altfeld M, Walker BD (2001) Less is more? STI in acute and chronic HIV-1 infection. Nat Med 7:881–884 Altfeld M, Allen TM, Yu XG, Johnston MN, Agrawal D, Korber BT, Montefiori DC, O’Connor DH, Davis BT, Lee PK, Maier EL, Harlow J, Goulder PJ, Brander C, Rosenberg ES, Walker BD (2002) HIV-1 superinfection despite broad CD8+ T-cell responses containing replication of the primary virus. Nature 420:434–439 Arts EJ, Quinones-Mateu ME (2003) Sorting out the complexities of HIV-1 fitness. Aids 17:780–781 Bickel PJ, Cosman PC, Olshen RA, Spector PC, Rodrigo AR, Mullins JI (1996) Covariability of V3 loop amino acids. AIDS Res Human Retrovir 12:1401–1411 Bienzle D, MacDonald KS, Smaill FM, Kovacs C, Baqi M, Courssaris B, Luscher MA, Walmsley SL, Rosenthal KL (2000) Factors contributing to the lack of human immunodeficiency virus type 1 (HIV-1) transmission in HIV-1-discordant partners. J Infect Dis 182:123–132 Birk M, Svedhem V, Sonnerborg A (2001) Kinetics of HIV-1 RNA and resistanceassociated mutations after cessation of antiretroviral combination therapy. Aids 15:1359–1368

Evolutionary Dynamics of HIV-1 and the Control of AIDS

185

Björndal A, Deng H, Jansson M, Fiore JR, Colognesi C, Karlsson A, Albert J, Scarlatti G, Littman DR, Fenyö EM (1997) Coreceptor usage of primary human immunodeficiency virus type 1 isolates varies according to biological phenotype. J Virol 71:7478–7487 Brenner BG, Routy JP, Petrella M, Moisi D, Oliveira M, Detorio M, Spira B, Essabag V, Conway B, Lalonde R, Sekaly RP, Wainberg MA (2002) Persistence and fitness of multidrug-resistant human immunodeficiency virus type 1 acquired in primary infection. J Virol 76:1753–1761 Bretscher MT, Althaus CL, Muller V, Bonhoeffer S (2004) Recombination in HIV and the evolution of drug resistance: for better or for worse? Bioessays 26:180–188 Brumme ZL, Dong WWY, Yip B, Wynhoven B, Hoffman NG, Swanstrom R, Jensen MA, Mullins JI, Hogg RS, Montaner JSG, Harrigan PR (2004) Clinical and immunological impact of HIV envelope V3 sequence variation after starting initial triple antiretroviral therapy. Aids 18:F1–F9 Casartelli N, Di Matteo G, Potesta M, Rossi P, Doria M (2003) CD4 and major histocompatibility complex class I downregulation by the human immunodeficiency virus type 1 nef protein in pediatric AIDS progression. J Virol 77:11536–11545 Chen R, Quinones-Mateu ME, Mansky LM (2004) Drug resistance, virus fitness and HIV-1 mutagenesis. Curr Pharm Des 10:4065–4070 Chen R, Quinones-Mateu ME, Mansky LM (2005) HIV-1 mutagenesis during antiretroviral therapy: implications for successful drug treatment. Front Biosci 10:743–750 Chun TW, Stuyver L, Mizell SB, Ehler LA, Mican JA, Baseler M, Lloyd AL, Nowak MA, Fauci AS (1997) Presence of an inducible HIV-1 latent reservoir during highly active antiretroviral therapy. Proc Natl Acad Sci U S A 94:13193–13197 Chun TW, Davey RT Jr, Ostrowski M, Shawn Justement J, Engel D, Mullins JI, Fauci AS (2000) Relationship between pre-existing viral reservoirs and the re-emergence of plasma viremia after discontinuation of highly active anti-retroviral therapy. Nat Med 6:757–761 Connor RI, Sheridan KE, Ceradini D, Choe S, Landau NR (1997) Change in coreceptor use correlates with disease progression in HIV-1-infected individuals. J Exp Med 185:621–628 Cooper DA, Imrie AA, Penney R (1987) Antibody response to human immunodeficiency virus after primary infection. J Infect Dis 155:1113–1118 Daar ES, Moudgil T, Meyer RD, Ho DD (1991) Transient high levels of viremia in patients with primary HIV-1 infection. N Engl J Med 324:961–964 De Jong JJ, Goudsmit J, Keulen W, Klaver B, Krone W, Tersmette M, de Ronde A (1992) Human immunodeficiency virus type 1 clones chimeric for the envelope V3 domain differ in syncytium formation and replication capacity. J Virol 66:757–765 Deeks SG, Barbour JD, Martin JN, Swanson MS, Grant RM (2000) Sustained CD4+ T cell response after virologic failure of protease inhibitor-based regimens in patients with human immunodeficiency virus infection. J Infect Dis 181:946–953 Deeks SG, Wrin T, Liegler T, Hoh R, Hayden M, Barbour JD, Hellmann NS, Petropoulos CJ, McCune JM, Hellerstein MK, Grant RM (2001) Virologic and immunologic consequences of discontinuing combination antiretroviral-drug therapy in HIVinfected patients with detectable viremia. N Engl J Med 344:472–480

186

J. I. Mullins · M. A. Jensen

Deeks SG, Grant RM, Wrin T, Paxinos EE, Liegler T, Hoh R, Martin JN, Petropoulos CJ (2003) Persistence of drug-resistant HIV-1 after a structured treatment interruption and its impact on treatment response. Aids 17:361–370 Delwart EL, Sheppard HW, Walker BD, Goudsmit J, Mullins JI (1994) Human immunodeficiency virus type 1 evolution in vivo tracked by DNA heteroduplex mobility assays. J Virol 68:6672–6683 Derdeyn CA, Decker JM, Bibollet-Ruche F, Mokili JL, Muldoon M, Denham SA, Heil ML, Kasolo F, Musonda R, Hahn BH, Shaw GM, Korber BT, Allen S, Hunter E (2004) Envelope-constrained neutralization-sensitive HIV-1 after heterosexual transmission. Science 303:2019–2022 Desrosiers RC (2004) Prospects for an AIDS vaccine. Nat Med 10:221–223 Di Mascio M, Markowitz M, Louie M, Hurley A, Hogan C, Simon V, Follmann D, Ho DD, Perelson AS (2004) Dynamics of intermittent viremia during highly active antiretroviral therapy in patients who initiate therapy during chronic versus acute and early human immunodeficiency virus type 1 infection. J Virol 78:10566–10573 Domingo E, Menendez-Arias L, Quinones-Mateu ME, Holguin A, Gutierrez-Rivas M, Martinez MA, Quer J, Novella IS, Holland JJ (1997) Viral quasispecies and the problem of vaccine-escape and drug-resistant mutants. Prog Drug Res 48:99–128 Dybul M, Chun TW, Yoder C, Hidalgo B, Belson M, Hertogs K, Larder B, Dewar RL, Fox CH, Hallahan CW, Justement JS, Migueles SA, Metcalf JA, Davey RT, Daucher M, Pandya P, Baseler M, Ward DJ, Fauci AS (2001) Short-cycle structured intermittent treatment of chronic HIV infection with highly active antiretroviral therapy: effects on virologic, immunologic, and toxicity parameters. Proc Natl Acad Sci U S A 98:15161–15166 Dybul M, Daucher M, Jensen MA, Hallahan CW, Chun TW, Belson M, Hidalgo B, Nickel DC, Yoder C, Metcalf JA, Davey RT, Ehler L, Kress-Rock D, Nies-Kraske E, Liu S, Mullins JI, Fauci AS (2003) Genetic characterization of rebounding human immunodeficiency virus type 1 in plasma during multiple interruptions of highly active antiretroviral therapy. J Virol 77:3229–3237 Finzi D, Blankson J, Siliciano JD, Margolick JB, Chadwick K, Pierson T, Smith K, Lisziewicz J, Lori F, Flexner C, Quinn TC, Chaisson RE, Rosenberg E, Walker B, Gange S, Gallant J, Siliciano RF (1999) Latent infection of CD4+ T cells provides a mechanism for lifelong persistence of HIV-1, even in patients on effective combination therapy. Nat Med 5:512–517 Fouchier RAM, Groenink M, Kootstra NA, Tersmette M, Huisman HG, Miedema F, Schuitemaker H (1992) Phenotype-associated sequence variation in the third variable domain (V3) of the human immunodeficiency virus type 1 gp120 molecule. J Virol 66:3183–3187 Fowke KR, Nagelkerke NJ, Kimani J, Simonsen JN, Anzala AO, Bwayo JJ, MacDonald KS, Ngugi EN, Plummer FA (1996) Resistance to HIV-1 infection among persistently seronegative prostitutes in Nairobi, Kenya. Lancet 348:1347–1351 Frenkel LM, Mullins JI (2001) Should patients with drug-resistant HIV-1 continue to receive antiretroviral Therapy? N Engl J Med 344:520–522 Frenkel LM, Wang Y, Learn GH, McKernan JL, Ellis GM, Mohan KM, Holte SE, De Vange SM, Pawluk DM, Melvin AJ, Lewis PF, Heath LM, Beck IA, Mahalanabis M, Naugler WE, Tobin NH, Mullins JI (2003) Multiple viral genetic analyses detecting low-level HIV-1 replication during effective HAART. J Virol 77:5721–5730

Evolutionary Dynamics of HIV-1 and the Control of AIDS

187

Friedrich TC, Dodds EJ, Yant LJ, Vojnov L, Rudersdorf R, Cullen C, Evans DT, Desrosiers RC, Mothe BR, Sidney J, Sette A, Kunstman K, Wolinsky S, Piatak M, Lifson J, Hughes AL, Wilson N, O’Connor DH, Watkins DI (2004) Reversion of CTL escape-variant immunodeficiency viruses in vivo. Nat Med 10:275–281 Goudsmit J, De Ronde A, Ho DD, Perelson AS (1996) Human immunodeficiency virus fitness in vivo: calculations based on a single zidovudine resistance mutation at codon 215 of reverse transcriptase. J Virol 70:5662–5664 Grobler J, Gray CM, Rademeyer C, Seoighe C, Ramjee G, Abdool Karim S, Morris L, Williamson C (2004) Incidence of HIV-1 dual infection and its association with increased viral load set point in a cohort of HIV-1 subtype C-infected female sex workers. J Infect Dis 190:1355–1359 Gunthard HF, Frost SD, Leigh-Brown AJ, Ignacio CC, Kee K, Perelson AS, Spina CA, Havlir DV, Hezareh M, Looney DJ, Richman DD, Wong JK (1999) Evolution of envelope sequences of human immunodeficiency virus type 1 in cellular reservoirs in the setting of potent antiviral therapy. J Virol 73:9404–9412 Hatano H, Vogel S, Yoder C, Metcalf JA, Dewar R, Davey RT Jr, Polis MA (2000) Pre-HAART HIV burden approximates post-HAART viral levels following interruption of therapy in patients with sustained viral suppression. Aids 14:1357–1363 Ho DD (1998) Toward HIV eradication or remission: the tasks ahead. Science 280:1866– 1867 Ho DD, Sarngadharan MG, Resnick L, Dimarzoveronese F, Rota TR, Hirsch MS (1985) Primary human T-lymphotropic virus type III infection. Ann Intern Med 103:880– 883 Hoffman NG, Seillier-Moiseiwitsch F, Ahn J, Walker JM, Swanstrom R (2002) Variability in the human immunodeficiency virus type 1 gp120 Env protein linked to phenotype-associated changes in the V3 loop. J Virol 76:3852–3864 Huang Y, Paxton WA, Wolinsky SM, Neumann AU, Zhang L, He T, Kang S, Ceradini D, Jin Z, Yazdanbakhsh K, Kunstman K, Erickson D, Dragon E, Landau NR, Phair J, Ho DD, Koup RA (1996) The role of a mutant CCR5 allele in HIV-1 transmission and disease progression [see comments]. Nat Med 2:1240–1243 Jain RG, Furfine ES, Pedneault L, White AJ, Lenhard JM (2001) Metabolic complications associated with antiretroviral therapy. Antiviral Res 51:151–177 Jensen MA, van ’t Wout AB (2003) Predicting HIV-1 coreceptor usage with sequence analysis. AIDS Rev 5:104–112 Jensen MA, Li F-S, van ’t Wout AB, Nickle DC, Shriner D, He H, McLaughlin S, Shankarappa R, Margolick JB, Mullins JI (2003) Improved coreceptor usage prediction and genotypic monitoring of R5-to-X4 transition by motif analysis of human immunodeficiency virus type 1 env V3 loop sequences. J Virol 77:13376– 13388 Jost S, Bernard MC, Kaiser L, Yerly S, Hirschel B, Samri A, Autran B, Goh LE, Perrin L (2002) A patient with HIV-1 superinfection. N Engl J Med 347:731–736 Karlsson I, Antonsson L, Shi Y, Karlsson A, Albert J, Leitner T, Olde B, Owman C, Fenyo EM (2003) HIV biological variability unveiled: frequent isolations and chimeric receptors reveal unprecedented variation of coreceptor use. Aids 17:2561–2569

188

J. I. Mullins · M. A. Jensen

Kaufmann DE, Lichterfeld M, Altfeld M, Addo MM, Johnston MN, Lee PK, Wagner BS, Kalife ET, Strick D, Rosenberg ES, Walker BD (2004) Limited durability of viral control following treated acute HIV infection. Plos Med 1:e36 Kaul R, Rowland-Jones SL, Kimani J, Dong T, Yang HB, Kiama P, Rostron T, Njagi E, Bwayo JJ, MacDonald KS, McMichael AJ, Plummer FA (2001) Late seroconversion in HIV-resistant Nairobi prostitutes despite pre-existing HIV-specific CD8+ responses. J Clin Invest 107:341–349 Kiepiela P, Leslie AJ, Honeyborne I, Ramduth D, Thobakgale C, Chetty S, Rathnavalu P, Moore C, Pfafferott KJ, Hilton L, Zimbwa P, Moore S, Allen T, Brander C, Addo MM, Altfeld M, James I, Mallal S, Bunce M, Barber LD, Szinger J, Day C, Klenerman P, Mullins J, Korber B, Coovadia HM, Walker BD, Goulder PJ (2004) Dominant influence of HLA-B in mediating the potential co-evolution of HIV and HLA. Nature 432:769–775 Kijak GH, Simon V, Balfe P, Vanderhoeven J, Pampuro SE, Zala C, Ochoa C, Cahn P, Markowitz M, Salomon H (2002) Origin of human immunodeficiency virus type 1 quasispecies emerging after antiretroviral treatment interruption in patients with therapeutic failure. J Virol 76:7000–7009 Koot M, Keet I, Vos A, deGoede R, Roos M, Coutinho R, Miedema F, Schellekens P, Tersmette M (1993) Prognostic value of HIV-1 syncytium-inducing phenotype for rate of CD4+ cell depletion and progression to AIDS. Ann Intern Med 118:681–688 Korber BT, Farber RM, Wolpert DH, Lapedes AS (1993) Covariation of mutations in the V3 loop of human immunodeficiency virus type 1 envelope protein: an information theoretic analysis. Proc Natl Acad Sci U S A 90:7176–7180 Koup RA, Safrit JT, Cao Y, Andrews CA, McLeod G, Borkowsky W, Farthing C, Ho DD (1994) Temporal association of cellular immune responses with the initial control of viremia in primary human immunodeficiency virus type 1 syndrome. J Virol 68:4650–4655 Learn GH, Muthui D, Brodie SJ, Zhu T, Diem K, Mullins JI, Corey L (2002) Virus population homogenization following acute human immunodeficiency virus type 1 infection. J Virol 76:11953–11959 Leigh-Brown AJ (1997) Analysis of HIV-1 env gene sequences reveals evidence for a low effective number in the viral population. Proc Natl Acad Sci U S A 94:1862–1865 Leslie AJ, Pfafferott KJ, Chetty P, Draenert R, Addo MM, Feeney M, Tang Y, Holmes EC, Allen T, Prado JG, Altfeld M, Brander C, Dixon C, Ramduth D, Jeena P, Thomas SA, John AS, Roach TA, Kupfer B, Luzzi G, Edwards A, Taylor G, Lyall H, TudorWilliams G, Novelli V, Martinez-Picado J, Kiepiela P, Walker BD, Goulder PJ (2004) HIV evolution: CTL escape mutation and reversion after transmission. Nat Med 10:282–289 Liu S-L, Mittler JE, Nickel DC, Mulvania TM, Shriner D, Rodrigo AG, Kosloff B, He X, Corey L, Mullins JI (2002) Selection for human immunodeficiency virus type 1 recombinants in a patient with rapid progression to AIDS. J Virol 76:10674–10684 Long EM, Martin HL Jr, Kreiss JK, Rainwater SM, Lavreys L, Jackson DJ, Rakwar J, Mandaliya K, Overbaugh J (2000) Gender differences in HIV-1 diversity at time of infection. Nat Med 6:71–75

Evolutionary Dynamics of HIV-1 and the Control of AIDS

189

Malaspina A, Moir S, Nickle DC, Donoghue ET, Ogwaro KM, Ehler LA, Liu S, Mican JA, Dybul M, Chun TW, Mullins JI, Fauci AS (2002) Human immunodeficiency virus type 1 bound to B cells: relationship to virus replicating in CD4(+) T cells and circulating in plasma. J Virol 76:8855–8863 Mallon PW, Carr A, Cooper DA (2002) HIV-associated lipodystrophy: description, pathogenesis, and molecular pathways. Curr Diab Rep 2:116–124 Margolick JB, Donnenberg AD, Munoz A (1994) T lymphocyte homeostasis after HIV seroconversion. J Acquir Immune Defic Syndr 7:415–416 Margolick JB, Munoz A, Donnenberg AD et al (1995) Failure of T-cell homeostasis preceding AIDS in HIV-1 infection. Nat Med 1:674–680 Mellors JW, Munoz A, Giorgi JV, Margolick JB, Tassoni CJ, Gupta P, Kingsley LA, Todd JA, Saah AJ, Detels R, Phair JP, Rinaldo CR Jr (1997) Plasma viral load and CD4+ lymphocytes as prognostic markers of HIV-1 infection. Ann Intern Med 126:946–954 Menzo S, Monachetti A, Balotta C, Corvasce S, Rusconi S, Paolucci S, Baldanti F, Bagnarelli P, Clementi M (2003) Processivity and drug-dependence of HIV-1 protease: determinants of viral fitness in variants resistant to protease inhibitors. Aids 17:663–671 Michael NL, Chang G, Louie LG, Mascola JR, Dondero D, Birx DL, Sheppard HW (1997) The role of viral phenotype and CCR-5 gene defects in HIV-1 transmission and disease progression. Nat Med 3:338–340 Milich L, Margolin B, Swanstrom R (1993) V3 loop of the human immunodeficiency virus type 1 Env protein: interpreting sequence variability. J Virol 67:5623–5634 Miller V (2001) Structured treatment interruptions in antiretroviral management of HIV-1. Curr Opin Infect Dis 14:29–37 Moore CB, John M, James IR, Christiansen FT, Witt CS, Mallal SA (2002) Evidence of HIV-1 adaptation to HLA-restricted immune responses at a population level. Science 296:1439–1443 Mullins JI (1997) Curtailing the AIDS pandemic (letter). Science 276:1955–1957 Mullins JI, Hoover EA, Quackenbush SL, Donahue PR (1991) Disease progression and viral genome variants in experimental feline leukemia virus-induced immunodeficiency syndrome (published erratum appears in J Acquir Immune Defic Syndr 1991;4:925). J Acquir Immune Defic Syndr 4:547–557 Mullins JI, Nickle DC, Heath L, Rodrigo AG, Learn GH (2004) Immunogen sequence: the fourth tier of AIDS vaccine design. Expert Rev Vaccines 3 [Suppl 1]: S151–S159 Muñoz A, Sabin CA, Phillips AN (1997) The incubation period of AIDS. AIDS 11:S69– S76 Nickle DC, Jensen MA, Shriner D, Brodie SJ, Frenkel LM, Mittler JE, Mullins JI (2003a) Evolutionary indicators of human immunodeficiency virus type 1 reservoirs and compartments. J Virol 77:5540–5546 Nickle DC, Shriner D, Mittler JE, Frenkel LM, Mullins JI (2003b) Importance and detection of virus reservoirs and compartments of HIV infection. Curr Opin Microbiol 6:410–416 Nijhuis M, Deeks S, Boucher C (2001) Implications of antiretroviral resistance on viral fitness. Curr Opin Infect Dis 14:23–28 O’Brien SJ, Nelson GW (2004) Human genes that limit AIDS. Nat Genet 36:565–574

190

J. I. Mullins · M. A. Jensen

O’Brien TR, Winkler C, Dean M, Nelson JA, Carrington M, Michael NL, White GC 2nd (1997) HIV-1 infection in a man homozygous for CCR5 delta 32. Lancet 349:1219 Overbaugh J, Donahue PR, Quackenbush SL, Hoover EA, Mullins JI (1988) Molecular cloning of a feline leukemia virus that induces fatal immunodeficiency disease in cats. Science 239:906–910 Oxenius A, Hirschel B (2003) Structured treatment interruptions in HIV infection: benefit or disappointment? Expert Rev Anti Infect Ther 1:129–139 Palella FJ Jr, Delaney KM, Moorman AC, Loveless MO, Fuhrer J, Satten GA, Aschman DJ, Holmberg SD (1998) Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med 338:853–860 Perelson AS, Neumann AU, Markowitz M, Leonard JM, Ho DD (1996) HIV-1 dynamics in vivo: virion clearance rate, infected cell life-span, and viral generation time. Science 271:1582–1586 Quinones-Mateu ME, Arts EJ (2002) HIV-1 fitness: implications for drug resistance, disease progression, and global epidemic evolution. In: Kuiken C, Foley B, Hahn B, Marx P, McCutchan F, Mellors J, Wolinsky S, Korber B (eds) HIV Sequence Compendium 2001. Theoretical Biology and Biophysics Group, Los Alamos National Laboratory, Los Alamos, NM, pp 134–170 Quinones-Mateu ME, Ball SC, Marozsan AJ, Torre VS, Albright JL, Vanham G, van Der Groen G, Colebunders RL, Arts EJ (2000) A dual infection/competition assay shows a correlation between ex vivo human immunodeficiency virus type 1 fitness and disease progression. J Virol 74:9222–9233 Resch W, Hoffman N, Swanstrom R (2001) Improved success of phenotype prediction of the human immunodeficiency virus type 1 from envelope variable loop 3 sequence using neural networks. Virology 288:51–62 Rhodes T, Wargo H, Hu WS (2003) High rates of human immunodeficiency virus type 1 recombination: near-random segregation of markers one kilobase apart in one round of viral replication. J Virol 77:11193–11200 Rodrigo AG (1997) Dynamics of syncytium-inducing and non-syncytium-inducing type 1 human immunodeficiency viruses during primary infection. AIDS Res Human Retrovir 13:1447–1451 Rosenberg ES, Altfeld M, Poon SH, Phillips MN, Wilkes BM, Eldridge RL, Robbins GK, D’Aquila RT, Goulder PJ, Walker BD (2000) Immune control of HIV-1 after early treatment of acute infection. Nature 407:523–526 Ruiz L, Martinez-Picado J, Romeu J, Paredes R, Zayat MK, Marfil S, Negredo E, Sirera G, Tural C, Clotet B (2000) Structured treatment interruption in chronically HIV-1 infected patients after long-term viral suppression. Aids 14:397–403 Sagar M, Lavreys L, Baeten JM, Richardson BA, Mandaliya K, Chohan BH, Kreiss JK, Overbaugh J (2003) Infection with multiple human immunodeficiency virus type 1 variants is associated with faster disease progression. J Virol 77:12921–12926 Schacker TW, Hughes JP, Shea T, Coombs RW, Corey L (1998) Biological and virologic characteristics of primary HIV infection. Ann Intern Med 128:613–620 Schrager LK, D’Souza MP (1998) Cellular and anatomical reservoirs of HIV-1 in patients receiving potent antiretroviral combination therapy. JAMA 280:67–71

Evolutionary Dynamics of HIV-1 and the Control of AIDS

191

Shankarappa R, Margolick JB, Gange SJ, Rodrigo AG, Upchurch D, Farzadegan H, Gupta P, Rinaldo CR, Learn GH, He X, Huang XL, Mullins JI (1999) Consistent viral evolutionary changes associated with the progression of human immunodeficiency virus type 1 infection. J Virol 73:10489–10502 Shpaer EG, Delwart EL, Kuiken CL, Goudsmit J, Bachmann MH, Mullins JI (1994) Conserved V3 loop sequences and transmission of human immunodeficiency virus type 1. AIDS Res Hum Retroviruses 10:1679–1684 Shriner D, Rodrigo AG, Nickle DC, Mullins JI (2004a) Pervasive genomic recombination of HIV-1 in vivo. Genetics 167:1573–1583 Shriner D, Shankarappa R, Jensen MA, Nickle DC, Mittler JE, Margolick JB, Mullins JI (2004b) Influence of random genetic drift on HIV-1 env evolution during chronic infection. Genetics 166:1155–1164 Siliciano JD, Kajdas J, Finzi D, Quinn TC, Chadwick K, Margolick JB, Kovacs C, Gange SJ, Siliciano RF (2003) Long-term follow-up studies confirm the stability of the latent reservoir for HIV-1 in resting CD4(+) T cells. Nat Med 9:727–728 Stoddart CA, Geleziunas R, Ferrell S, Linquist-Stepps V, Moreno ME, Bare C, Xu W, Yonemoto W, Bresnahan PA, McCune JM, Greene WC (2003) Human immunodeficiency virus type 1 nef-mediated downregulation of CD4 correlates with nef enhancement of viral pathogenesis. J Virol 77:2124–2133 Tanaka M, Ueno T, Nakahara T, Sasaki K, Ishimoto A, Sakai H (2003) Downregulation of CD4 is required for maintenance of viral infectivity of HIV-1. Virology 311:316– 325 Tang J, Tang S, Lobashevsky E, Zulu I, Aldrovandi G, Allen S, Kaslow RA (2004) HLA allele sharing and HIV type 1 viremia in seroconverting Zambians with known transmitting partners. AIDS Res Hum Retroviruses 20:19–25 Tersmette M, de Goede REY, Al BJM, Winkel IN, Gruters RA, Cuypers HT, Huisman HG, Miedema F (1988) Differential syncytium-inducing capacity of HIV isolates: frequent detection of syncytium-inducing isolates in patients with AIDS and AIDS-related complex. J Virol 62:2026–2032 Tersmette M, Gruters RA, de Wolf F, de Goede RE, Lange JM, Schellekens PT, Goudsmit J, Huisman HG, Miedema F (1989) Evidence for a role of virulent human immunodeficiency virus (HIV) variants in the pathogenesis of acquired immunodeficiency syndrome: studies on sequential HIV isolates. J Virol 63:2118–2125 Theodorou I, Meyer L, Magierowska M, Katlama C, Rouzioux C (1997) HIV-1 infection in an individual homozygous for CCR5 delta 32. Seroco Study Group. Lancet 349:1219–1220 Trachtenberg E, Korber B, Sollars C, Kepler TB, Hraber PT, Hayes E, Funkhouser R, Fugate M, Theiler J, Hsu YS, Kunstman K, Wu S, Phair J, Erlich H, Wolinsky S (2003) Advantage of rare HLA supertype in HIV disease progression. Nat Med 9:928–935 Tremblay CL, Hicks JL, Sutton L, Giguel F, Flynn T, Johnston M, Sax PE, Walker BD, Hirsch MS, Rosenberg ES, D’Aquila RT (2003) Antiretroviral resistance associated with supervised treatment interruptions in treated acute HIV infection. Aids 17:1086–1089

192

J. I. Mullins · M. A. Jensen

Van ’t Wout AB, Kootstra NA, Mulder-Kampinga GA, Albrecht-van Lent N, Scherpbier HJ, Veenstra J, Boer K, Coutinho RA, Miedema F, Schuitemaker H (1994) Macrophage-tropic variants initiate HIV-1 infection after sexual, parenteral, and vertical transmission. J Clin Invest 94:2060–2067 Van Rij RP, Blaak H, Visser JA, Brouwer M, Rientsma R, Broersen S, de Roda Husman AM, Schuitemaker H (2000) Differential coreceptor expression allows for independent evolution of non-syncytium-inducing and syncytium-inducing HIV-1. J Clin Invest 106:1039–1052 Wei X, Ghosh SK, Taylor ME, Johnson VA, Emini EA, Deutsch P, Lifson JD, Bonhoeffer S, Nowak MA, Hahn BH, Saag MS, Shaw GM (1995) Viral dynamics in human immunodeficiency virus type 1 infection. Nature 373:117–122 Wein LM, D’Amato RM, Perelson AS (1997) Mathematical analysis of antiviral therapy aimed at HIV-1 eradication or maintenance of low viral loads. J Theor Biol 192:81–98 Yang OO, Daar ES, Jamieson BD, Balamurugan A, Smith DM, Pitt JA, Petropoulos CJ, Richman DD, Little SJ, Brown AJ (2005) Human immunodeficiency virus type 1 clade B superinfection: evidence for differential immune containment of distinct clade B strains. J Virol 79:860–868 Yusim K, Kesmir C, Gaschen B, Addo MM, Altfeld M, Brunak S, Chigaev A, Detours V, Korber BT (2002) Clustering patterns of cytotoxic T-lymphocyte epitopes in human immunodeficiency virus type 1 (HIV-1) proteins reveal imprints of immune evasion on HIV-1 global variation. J Virol 76:8757–8768 Zhang LQ, MacKenzie P, Cleland A, Holmes EC, Leigh Brown AJ, Simmonds P (1993) Selection for specific sequences in the external envelope protein of HIV-1 upon primary infection. J Virol 67:3345–3356 Zhu T, Mo H, Wang N, Nam DS, Cao Y, Koup RA, Ho DD (1993) Genotypic and phenotypic characterization of HIV-1 in patients with primary infection. Science 261:1179–1181 Zhu T, Corey L, Akridge R, Change Y, Feng F, Kim J, Alef C, McElroy A, Mullins J, McElrath J (1999) Evidence for HIV-1 latent infection in exposed seronegative individuals. 6th Conference on Retroviruses and Opportunistic Infections, Chicago Zhu T, Brodie S, Mullins JI, Corey L (2000) Reservoirs of viral persistence after HAART therapy of primary HIV. International meeting of the Institute of Human Virology, Baltimore Zhu T, Muthui D, Holte S, Nickle D, Feng F, Brodie S, Hwangbo Y, Mullins JI, Corey L (2002) Evidence for human immunodeficiency virus type 1 replication in vivo in CD14(+) monocytes and its potential role as a source of virus in patients on highly active antiretroviral therapy. J Virol 76:707–716 Zhu T, Corey L, Hwangbo Y, Lee JM, Learn GH, Mullins JI, McElrath MJ (2003) Persistence of extraordinarily low levels of genetic homogeneous human immunodeficiency virus type 1 in exposed seronegative individuals. J Virol 77:6108–6116 Zhu T, Hu S-L, Feng F, Polacino P, Liu H, Hwangbo Y, Learn GH, Mullins JI, Corey L (2004) Persistence of low levels of simian immunodeficiency virus in macaques that were transiently viremic by conventional testing. Virology 323:208–219

CTMI (2006) 299:193–209 c Springer-Verlag Berlin Heidelberg 2006


Evolution of Virulence in Picornaviruses S. Tracy1 (u) · N. M. Chapman1 · K. M. Drescher2 · K. Kono1 · W. Tapprich3 1 Department of Pathology and Microbiology, University of Nebraska Medical Center,

Omaha, NE 68198-6495, USA [email protected] 2 Department of Medical Microbiology and Immunology, Creighton University School of Medicine, 2500 California Plaza, Omaha, NE 68178, USA 3 Department of Biology, University of Nebraska at Omaha, Omaha, NE 68182, USA

1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193

2

Cardioviruses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195

3

Enteroviruses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197

4

Enterovirus Virulence: Poliovirus Attenuation . . . . . . . . . . . . . . . . . . . 199

5

Musings on Picornavirus Virulence Evolution . . . . . . . . . . . . . . . . . . . . 199

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 Abstract The Picornaviridae encompass many positive-strand RNA viruses, all of which share a generally similar genome design and capsid structure, but which induce quite diverse diseases in humans and other animals. Picornavirus strains of the same serotype have been shown to express different virulence (or pathogenic) phenotypes when studied in animal models, demonstrating that key elements of pathogenesis reside in the viral genome. However, the genetics that determine the virulence phenotype of any picornavirus are poorly understood. Picornaviruses do not have virulence genes per se, but the design of the capsid and how it interacts with the virus receptor expressed on the host cell surface, specific sequences within the nontranslated regions of the viral genome, as well as coding sequences that result in different protein sequences may all have a part in determining the virulence phenotype. Virulence may be better understood as a continuum from an apparent inability to induce disease to the ability to cause severe pathogenic changes. Ultimately, the ability of a picornavirus to induce disease depends upon viral genetics and how they are modulated by the host environment.

1 Introduction The picornavirus family consists of numerous positive-strand RNA viruses that induce diverse diseases in humans and other animals (Rueckert 1996).

194

S. Tracy et al.

Among the most well-known picornaviruses are the polioviruses (PV; genus Enterovirus; Pallansch and Roos 2001), causative agents of the widespread epidemics of poliomyelitis in the twentieth century, but which are being eradicated worldwide (Minor 2003; Dowdle et al. 2004); foot-and-mouth disease virus (FMDV; genus Aphthovirus; Domingo et al. 2003), which causes chronic disease of cattle and other ungulates that restricts the widespread export of meat products from these regions, with associated severe economic impact; human rhinoviruses (HRV; genus Rhinovirus; Savolainen et al. 2003), a common and prominent causative agent of the irritating common cold; and the group B coxsackieviruses (CVB; genus Enterovirus; Tracy et al. 1997), which induce numerous serious diseases including myocarditis, pancreatitis, and possibly are involved in the etiology of insulin-dependent diabetes mellitus (Drescher et al. 2004). Picornaviral genomes share a common design: a single-strand positive sense RNA encodes a single open reading frame, which is flanked on either end by nontranslated regions (NTR) and a terminal 3 poly (A) tract. Variations on this familial theme help to differentiate the genera (Rueckert 1996). The genomic RNA is encapsidated within a naked protein shell formed in an icosahedral shape and composed of four capsid proteins. The reader is directed for further detail to the regularly updated website at the Institute for Animal Health in the UK [http://www.iah.bbsrc.ac.uk/virus/Picornaviridae/]. A significant body of work has defined the genetics of the artificiallyattenuated PV Sabin vaccine strains (Racaniello 1988; Minor 1992, 1993), has demonstrated the diminution of encephalomyocarditis and Mengo virus virulence due to deletion of the 5 NTR poly (C) tract (Osorio et al. 1996; Martin et al. 1996), and has identified various mutations that reduce or ablate CVB virulence in mice (Tracy et al. 1996; Chapman et al. 1997). Yet, despite the clinical and economic havoc that picornavirus infections wreak upon society, surprisingly little is understood of the viral genetics that influence the expression of a virulence phenotype. A working definition of virulence is the ability of a specific strain of virus to induce disease. Of course, the induction of disease in animals is not solely driven by the genetics of the infecting virus strain, but clearly, this plays a crucial triggering role: viruses initiate the disease and naturally occurring virus strains can be isolated that are more or less virulent when other parameters are held constant. These other parameters include host species, host genetic background, host immune system status, gender, age, route of infection, stress placed on the host and so forth. Only picornavirus genetics are examined in this review, with the intent of understanding why some virus strains become more virulent than others.

Evolution of Virulence in Picornaviruses

195

2 Cardioviruses The two TMEV (Theiler’s murine encephalomyelitis virus) subgroups, GDVII and TO (Lorch et al. 1981), are classified on their ability to cause disease in the central nervous system following intracerebral (i.c.) inoculation. GDVII strains (FA, GDVII) are highly neurovirulent in mice, usually resulting in death within 1 week (Lorch et al. 1981; Tsunoda et al. 1996). The TO strains (DA, BeAn) induce acute encephalitis, which resolves within 10 days in immunocompetent mice (Lorch et al. 1981; Daniels et al. 1952; Lipton and Dal Canto 1979); a persistent demyelinating disease can occur, which is largely dependent on the H-2 haplotype (Rodriguez and David 1985). The mean lethal dose (LD50 ) for the TMEV subgroups is very different but appears to be host strain-independent: for example, the LD50 of GDVII is 1 PFU in SJL mice when given i.c., while the LD50 of the DA strain is greater than 106 . The amino acid sequences of TMEV strains are 95% identical (Pevear et al. 1987, 1988). Replacing the region of the less virulent DA genome that encompasses 1B-2C (containing both capsid and nonstructural proteins) with the corresponding region from the GDVII strain results in a virus that is more neurovirulent than the parental DA virus while the virus generated by replacing the 5 terminus through protein 2C induces neurovirulence similar to GDVII (Fu et al. 1990), suggesting the virulence is influenced by the 5 NTR as well as capsid and nonstructural proteins. Several studies have focused on analyzing receptor-capsid interactions between the two subgroups as an approach to understanding virulence, which have led to two hypotheses regarding receptor use and viral virulence. The two TMEV subgroups may bind the same receptor but in different regions—GDVII strains bind to a protein moiety (Zhou et al. 1997), while TO strains bind both the protein and sialic acid (Fotiadis et al. 1991; Jnaoui and Michiels 1999; Zhou et al. 2000). Alternatively, unique receptors or co-receptors have been proposed (Jnaoui et al. 1999). TMEV is presumed to bind to its receptor through interactions in the pit which surrounds the fivefold axis on the capsid surface (Luo et al. 1992; Grant et al. 1992). Near the pit’s rim, four loop structures exist perpendicular to the surface of the virus: capsid protein 1D contains loop 1 and loop 2 (McCright et al. 1999); puff A and puff B are contained in capsid protein 1B (Wada et al. 1998). Puff B can bind sialic acid and mutations in puff B can affect sialic acid binding, viral persistence, and pathogenicity (Wada et al. 1998; Tsunoda et al. 2001). Interactions between these surface loops may influence viral biology such as the movement of TMEV from the gray matter to the white matter (Tsunoda et al. 2001): a DA virus strain containing both the 1D loop 2 of GDVII and a puff B mutation replicated to lower titers during the acute

196

S. Tracy et al.

encephalitic disease phase relative to the parental DA strain and mice inoculated with the chimeric TMEV strain did not progress to the demyelinating phase of the disease (Tsunoda et al. 2001). Receptor interactions as determinants of virulence phenotypes have also been suggested for encephalomyocarditis virus (EMCV), a cardiovirus. The EMCV strain M, which has a much greater LD50 , induces diabetes in susceptible strains of mice and replicates in the pancreatic islets (Ross et al. 1975). Plaque isolation of strains from EMCV M identified a nondiabetogenic B strain and a diabetogenic D strain (Yoon et al. 1980) which differed from each other by just 14 nucleotides (Bae et al. 1989). Nucleotide 3155 (VP1 aa152) determined whether Ala was expressed (diabetogenic) or Thr (not diabetogenic) (Bae and Yoon 1993). Other amino acid residues at this position were also nondiabetogenic (Jun et al. 1997). This may be due to changes in receptor interaction, as this residue maps to the “pit” of the mengovirus capsid, which is suggested to be as important for receptor interaction (Luo et al. 1987). The EMCV diabetogenic variation, which occurred in a site encoding a region of the capsid predicted to have receptor interactions, echoes TMEV studies in which capsid protein variation may affect the molecular details of receptor interactions, thereby helping to determine the type of disease generated. Genomic RNAs of EMCV and Mengo virus (another cardiovirus) have a long poly (C) tract situated about 150 nucleotides from the 5 terminus. Studies of attenuation in EMCV strains have shown that the truncation of this tract can diminish the LD50 in mice; similarly, poly (C) truncation or elimination greatly diminishes virulence in Mengo virus. Most EMCV (and Mengo) strains are lethal for adult mice in a short period of time at very low dose. Due to instability of such tracts in bacterial plasmid clones, infectious EMCV and Mengo virus genomes were fortuitously generated with truncated poly (C) tracts; these were found to replicate with nearly normal virulence in HeLa cells (Duke and Palmenberg 1989). However, in mice, pathogenicity in these recombinant Mengo strains was significantly reduced or ablated (Duke et al. 1990). Other studies demonstrated that the shortened poly (C) tract limited the extent of replication in murine hematopoietic cell lines (Martin et al. 2000). Despite the close similarity of these cardioviruses, the degree of attenuation due to poly (C) truncation is greater for Mengo than for EMCV: mutational studies using EMCV demonstrate the tract must be decreased to fewer than nine nucleotides for attenuation (Hahn and Palmenberg 1995), while a natural EMCV strain from an aborted swine fetus with a poly (C) tract of 16 pyrimidine residues was virulent (LaRue et al. 2003). The isolation of a virulent EMCV strain with a short poly (C) tract 16 nt in length indicates such strains may exist in the wild; more isolates must be characterized in

Evolution of Virulence in Picornaviruses

197

order to determine the common variations of poly pyrimidine tract length that are naturally permitted.

3 Enteroviruses In mice, a CVB strain may induce either (a) no disease (avirulent), (b) pancreatitis (inflammation of the acinar or exocrine pancreas tissue, leaving insulin and glucagon producing islets of Langerhans intact), or (c) pancreatitis and myocarditis (inflammation of the heart muscle) (Tracy et al. 2000). Myocarditis without preceding pancreatitis has not been observed, suggesting a link between the induction of pancreatitis and subsequent myocarditis. Cardiovirulent viruses replicate to high titers and can persist in mice for weeks to months (Tam et al. 1994; Kandolf et al. 1999; Tracy et al. 2000, 2002). Most isolates are pancreovirulent; pancreatitis is induced by all strains characterized to date, with the exception of rare, naturally occurring avirulent strains, suggesting that pancreas is a natural target for CVB (inferred also from clinical observations; e.g., Ursing 1973; Arnesjo et al. 1976; Imrie et al. 1977; Kennedy et al. 1986; Lal et al. 1988). Strains that induce just pancreatitis (no myocarditis, although virus replicates in the heart) replicate to lower titers than cardiovirulent strains in both mouse pancreas and heart. Relatively few studies have examined naturally occurring (or wild type) genetics of an enterovirus virulence phenotype. Wild type is equated with a wild or clinical isolate: e.g., isolated from sewage or from stool or tissue and of low passage history in the laboratory. This definition differentiates between other, laboratory-modified strains, such as antibody escape mutants or genetically modified strains. The site that determines a CVB3 cardiovirulent phenotype was mapped using three different CVB3 strains (Dunn et al. 2000): (a) a cloned cDNA copy of a cardiovirulent strain’s genome (CVB3/20; Tracy et al. 1992)], (b) CVB3/AS, a cardiovirulent strain (Tracy et al. 2000), and (c) CVB3/CO, a pancreovirulent strain that causes pancreatitis but not myocarditis in mice (Tracy et al. 2000). Chimeric infectious cDNA genomes were created on the CVB3/20 background using amplified sequences from either CVB3/AS or CVB3/CO, and tested for pathogenicity in mice (Dunn et al. 2000). This identified the 5 NTR as the primary region that determines a cardiovirulent phenotype, work which led to finer mapping within the 5 NTR (Dunn et al. 2003), which identified a region, contained within nt88–181, as a key determinant of cardiovirulence. The region between nt88 and 181 encloses a predicted stem-loop structure (stem loop II or SLII). Replacing the CVB3/20 SLII with that from CVB3/CO resulted

198

S. Tracy et al.

in a strain that did not induce myocarditis; replacement of this domain with that from CVB3/AS or CVB3/20 in the same construct restored cardiovirulence (Dunn et al. 2003). Domain II (SLII) structures predicted using MFOLD (Le and Zuker 1990; Zuker 2003) suggested that the secondary structure for CVB3/CO may vary significantly from very similar SLII secondary structure of CVB3/20 and CVB3/AS (despite differences in primary structure for the latter strains). The mechanism by which changes in this structure invoke different virulence phenotypes is not known. The pig enterovirus SVDV (swine vesicular disease virus; Knowles and McCauley 1997), is related to human CVB5 (Zhang et al. 1993) and most likely derived from a CVB5 infection of swine (Brown et al. 1973). Using a virulent SVDV strain from a clinically sick pig and an avirulent strain from a healthy pig, and a series of intratypic SVDV chimeric infectious cDNA copies, the genetics underlying the lesion-producing phenotype was mapped to capsid protein 1D and protease 2A, identifying two amino acid changes (1D, aa132, and 2Apro, aa20) (Kanno et al. 1998, 1999, 2001). Mutation of the 2Apro site to the avirulent genotype rapidly reverted to virulent type after inoculation into pigs. The mechanism by which these changes influenced the phenotype was not determined. Because work on CVB3 has demonstrated the 5 NTR to be a key arbiter of murine cardiovirulence with no apparent influence of the coding region or 3 NTR (Dunn et al. 2003; S. Tracy, unpublished data, below), the SVDV results suggest the possibility that different enteroviruses may determine virulence phenotypes through different genetic mechanisms. Indeed, similar results to those derived in the SVDV model have been obtained using mouse-adapted strains of CVB4 to map the determinants of pancreovirulence (pancreatitis) (Ramsingh and Collins 1995; Ramsingh 1997; Ramsingh et al. 1999; Halim and Ramsingh 2000): in this model, two amino acid changes in the capsid proteins 1D and 1A play primary roles. A clinical isolate and tissue-culture-derived strains of CVB1 (Tam et al. 1994, 2003; Tam and Messner 1997, 1999) have been employed to identify sites impacting the induction of chronic inflammatory myopathy (skeletal myositis) in mice; here, the impact of specific sites is less clear, with five different sites having been mapped that influence virulence. It would appear that despite differences in serotype classification, the type and site of pathogenic outcome, the expression of a virulence phenotype is influenced by similar genomic regions. Cumulatively, the data from the various CVB or SVDV studies suggest that specific diseases are induced by one or more different sites in the viral genome. So far, there have been no cross-talk studies to determine the impact of, for example, the 5 NTR on the expression of cardiac or pancreatic virulence phenotypes in other CVB serotypes using diverse strains. Until this is accomplished, it will be difficult to integrate current knowledge regarding

Evolution of Virulence in Picornaviruses

199

how these different enteroviruses determine a virulence phenotype. This itself is contingent upon a prior, clear description of the naturally occurring molecular genetics that determine any specific virulence phenotype. As mentioned below, the alacrity with which the CVB normally and regularly recombine (Oberste et al. 2004) may make this task somewhat more formidable than it might otherwise be.

4 Enterovirus Virulence: Poliovirus Attenuation The Sabin PV vaccine strains were derived by repeatedly passaging neurovirulent PV in culture until inoculation of primates demonstrated that the strains were avirulent (reviewed in Sabin and Boulger 1973). PV3 attenuation, as well as its subsequent reversion to neurovirulence upon replication in the gut, correlated primarily with a single nucleotide change in the 5 NTR (Evans et al. 1985). That a single transition could accomplish such a significant biologic change initially met with skepticism; nonetheless, subsequent work (reviewed in Racaniello 1988) demonstrated similar sites of attenuation in the 5 NTRs of PV1 (Omata et al. 1986; Kawamura et al. 1989) and PV2 (Macadam et al. 1991, 1993). These mutations (in domain V of the 5 NTR) cause structural changes (Malnou et al. 2004) in the IRES that reduce binding of the translation initiation factor eIF4G (Ochs et al. 2002). Comparison of neurovirulent to avirulent PV strains in cultures of neural origin demonstrated that artificially attenuated vaccine strains were less efficient at translating viral proteins than neurovirulent strains (Svitkin et al. 1985; La Monica and Racaniello 1989). Other sites have been demonstrated to play lesser roles in attenuation; e.g., an amino acid change in the capsid protein VP3 which, in the vaccine strain, partially destabilizes the viral capsid (Racaniello 1988). Only early, nonmolecular studies (Sabin 1955) refer to different strains of wild-type PV that exhibit different virulence phenotypes. With the effort to eradicate PV as a cause of human misery worldwide, the naturally occurring genetics that modulated PV neurovirulence may never be determined, pointing out the need for the use of other enterovirus models to study this aspect of enteroviral biology.

5 Musings on Picornavirus Virulence Evolution At present, the picornaviral genetics that translates to a specific virulence phenotype remains inadequately described, although efforts are beginning

200

S. Tracy et al.

to clarify the issue in specific systems. Consequently, we may only speculate based upon the limited information available on the topic. Picornaviruses (and viruses in general) often cause obvious or clinically defined disease in their hosts because virus replication often significantly disturbs the body’s homeostasis. Yet even with severely pathogenic viruses, the rate of induction of severe disease is low. For example, PV infections were greatly feared during epidemics, and rightly so; nonetheless, it is estimated that about 1/100 PV infections or fewer led to paralysis, 10% of which were lethal while most cases either resolved with some residual paralysis or resolved completely (Minor 2003). The extent of viral disease (subclinical or readily apparent symptoms) and its outcome (getting well, chronic persistence of symptoms, or death) are related to the impact that a specific virus strain has upon the host. The host’s contribution to this equation is huge and variable. Understanding the whole process focuses initially upon the trigger for the disease process—the virus—a discrete unit which can be isolated, studied, and its impact upon the host assessed. Work with the CVB has demonstrated that distinct virulence phenotypes are associated with CVB strains; genetic analysis, through the use of chimeric intratypic genome construction, demonstrates that virulence phenotypes can be mapped (Tu et al. 1995; Dunn et al. 2000, 2003). This may seem intuitive, particularly with the wealth of studies that have defined the molecular mechanisms underlying artificially attenuated picornavirus strains. However, the small picornavirus genome is highly tuned to function efficiently: what may seem to be minor, artificially induced variations in the genome can have enormous impact upon the biology of the parental virus. The elegant pioneering work by Evans and colleagues (Evans et al. 1985), which defined a single nucleotide transition in the 5 NTR of PV3 as a key attenuating site for the PV3 Sabin vaccine, demonstrated this point. But this is different from considering how Nature has fine-tuned picornaviral genomes to function optimally within a given host and more specifically, within an organ in that host wherein disease can be induced. Picornaviruses do not have virulence genes per se but have evolved specific genomic regions that play key roles in whether or not they can cause disease. The best evidence to date implicates the 5 NTR and the capsid protein coding regions as having the greatest role in determining virulence phenotypes: the nonstructural proteins and 3 NTR appear to have much less input. The poly (C) tract of Mengo virus is not lost willingly, yet artificial reduction or elimination of the poly (C) tract has shown that the virus can function well without it; furthermore, these murine viruses do not reacquire the poly (C) tract (Palmenberg and Osorio 1994) during replication in mice. Nonetheless, while shortening the poly (C) tract of the closely related encephalomyocarditis

Evolution of Virulence in Picornaviruses

201

virus (EMCV) causes diminished virulence, these artificially created EMCV strains remain more virulent in mice than Mengo strains with similarly truncated homopolymer tracts (Hahn et al. 1995), indicating that the biology resulting from such induced changes must be considered in the context of the specific viral genome. The internal ribosomal entry site (IRES) employed by all picornaviruses in translational initiation (Belsham and Sonenberg 1996) may differentially affect virus replication depending upon the cell type in which the virus is replicating (Borman et al. 1997; Roberts et al. 1998). The cardiovirus IRES is highly efficient in promoting translation, not requiring shut-off of host cell capped mRNA synthesis like enteroviruses in HeLa cells. The SLII in the 5 NTR of CVB3, and likely other CVB genotypes, appears at present to play a key role in determining whether a CVB strain induces myocarditis in mice (Dunn et al. 2003). The selection of a stable, virulent, picornavirus quasispecies population during replication within the host is influenced by numerous factors. One is the high rate of nucleotide misincorporation without an editing function that occurs with the picornaviral RNA-dependent RNA polymerase (Ward et al. 1988; Domingo et al. 1996). This poor fidelity introduces about one misincorporation per replication event in a picornavirus-sized genome of 7,000–8,000 nucleotides. Another factor is the promiscuous genetic recombination that occurs among related species (Agol et al. 1999; Lindberg et al. 2003; Oberste et al. 2004). Recombination between two viral genomes permits rapid, stochastic, saltatory jumps into different and potentially exploitable regions of sequence space. Finally, one must consider the genome of the infecting virus strain: how close is the genome to that which is necessary to become a virulent genome? For example, the rapid evolution of the Sabin PV3 vaccine strain to neurovirulence during replication in human vaccinees involved a single nucleotide change and occurred within days after vaccination (Evans et al. 1985). Replacement of the short region encompassed by domain II in the CVB3 5 NTR in a myocarditic CVB3 strain with the domain II (SLII) from another CVB3 strain unable to induce myocarditis resulted in a nonmyocarditic phenotype (Dunn et al. 2003). Again, though the host itself plays a colossal role in this effort in terms of species, age, gender, and so forth, the initiating agent and the endgame itself in the process remains the viral genome. Current work in our laboratories suggests that the rate of virus replication plays a key role in the ability of a CVB strain to induce disease, whether pancreatitis or myocarditis. This is further influenced by what appears to be a requirement for a specific domain II sequence to assume a discrete higherorder structure. While the ability to induce pancreatitis alone appears to be closely linked to the rate of CVB3 replication, the mechanism must also invoke rapidly attaining a threshold virus load with concurrent cytopathic

202

S. Tracy et al.

changes. The ability to induce inflammatory heart disease in the mouse is related to a specific RNA structure, which suggests a more complicated dance, one between the viral RNA structure and host cell components that differentially recognize the virus RNA structure. Although other work with different CVB systems implies a necessity for presentation of a specific antigen to the T cell repertoire of the mouse host in order to induce disease (Ramsingh et al. 1989, 1995; Halim et al. 2000), the potential impact of replication rate (which appears to be largely mediated by the 5 NTR) using different, wildtype isolates has not been examined. The molecular genetic grounds for PV Sabin strain attenuation have been ascribed to a single attenuating mutation in domain V in the 5 NTR (Evans et al. 1985; Racaniello 1988; Minor 1992; Racaniello and Ren 1996), a change that alters the replication rate of Sabin strains relative to wild-type strains in cells of neural origin (Svitkin et al. 1985; La Monica et al. 1989). It would have been illuminating to have conducted experiments with diverse wild-type PV strains to determine whether less virulent strains also replicated more slowly in specific environments than virulent strains. Our current working hypothesis is that picornavirus virulence represents the coming together of specific genetic components that promulgate virion stability and receptor-binding site formation for efficient transmission, and rapid virus replication in primary and key secondary sites within the host that establish a pathogenic state despite the innate immune system and before full activation of the adaptive immune response. A better definition of virulence than that with which we began this discussion might then be one that takes into account the level of disease induced by a specific virus following a standard, environmentally reasonable dose, perhaps calling that normal. Thus, virulence might be understood as a continuum, from the causation of no apparent disease (rare) through inducing normal disease states (common) to the induction of extremely severe or lethal infection (relatively rare). This definition may be more practical from the standpoint of viral biology. Sequence space offers any virus a wealth of opportunities to explore, a great number of which may be potentially visited on any given day due to the short replication cycle times, polymerase infidelity, and genomic recombinational events. Assuming access to the right host (the meaning of which almost escapes definition but would include considerations of immune status, age, gender, and species as key components), a specific minor population of virus might rapidly assume ascendancy. For example, although we have characterized some CVB strains capable of inducing severe myocarditis and pancreatitis in mice at low doses, the great majority of CVB3 strains cause only pancreatitis at fairly high doses (1–5 × 105 TCID50 or more per mouse). The more virulent strains that affect the mouse heart also replicate to several logs higher titer in mice. This

Evolution of Virulence in Picornaviruses

203

can be recapitulated as well in primary cultures of human cells, suggesting this is not a mouse-specific artifact. Assuming that the experimentally infected mouse approximates the human condition, we have wondered why most CVB strains are not highly virulent (able to cause serious disease with nearly every infectious event). What are the constraints upon the continuing evolution of the CVB genome that keep the dominant viral phenotype from establishing itself as highly virulent? We can speculate that all that should be necessary for this to occur is replication to high titers. This apparently is accomplished by the more virulent CVB strains, those able to induce myocarditis in mice, yet they are not the most commonly isolated (Tracy et al. 2000). Assuming that pancreovirulence is normal for CVB and cardiovirulence is the exception, as it appears to be, does cardiovirulence evolve within the host following infection? The optimal setting might be imagined to be in an infant who lacks an established immune repertoire and whose mother cannot provide protective antiviral immunity during gestation and nursing. Preliminary work in this laboratory has shown that inoculation of scid mice, which lack functional T and B cells, with different nonmyocarditic CVB3 strains, does not result in rapid death (as when inoculated with myocarditic strains), these mice live for months before suddenly dying with massive myocarditis (S. Tracy, unpublished data). This observation suggests that cardiovirulence may evolve within a host but requires significant time, at least in a mouse. If CVB cardiovirulence evolves as a new dominant quasispecies in a host, why does it not establish itself as the dominant phenotype elsewhere? We speculate that the answer may lie in the difference between the level of immune system development of infants on one hand and children to adults on the other hand. Evolution of a cardiovirulent CVB strain in an infant may be the route by which such a phenotype normally appears and viral myocarditis can spread within a hospital or enjoy a limited local circulation in society. While an infant lacking a functional adaptive immune system or pre-existing immunity could shed such a virus in stool and urine, viral spread within older children through adults could be expected to be limited both by pre-existing antiviral immunity as well as other factors. If the evolution of such a virulent phenotype requires a specialized immunocompromised host (such as a neonate), we can hypothesize that this could also be expected to significantly limit exposure of the population to such viral strains, thereby helping to make them relatively rare. The low-level persistence of memory genomes in a viral quasispecies (Arias et al. 2004) would be consistent as well with this hypothesis. In summary, the ability of a picornavirus to induce disease rests upon both viral genetics and the host environment. Viral genetics are key to the process, but are modulated by diverse host factors, any of which can be variable

204

S. Tracy et al.

upon transmission between outbred individuals. In the majority of cases, the picornavirus genome represents a master sequence (Domingo et al. 1996) that is well-adapted to the specific host: this adaptation occurs continually as a function of polymerase error and recombination events. Entry into the new host effects new pressures upon any specific viral genome, which may have the result of permitting disease or not, as well as determining how efficient the next cycle of infection to the next host becomes. Studies in relatively well-controlled host environments such as cell cultures or inbred mice make possible mapping of specific viral genetic regions that are key to expression of a specific virulence phenotype. However, it is important to test findings with other, naturally occurring viral strains to determine their broad applicability for these have evolved in the uncontrolled ‘outside’ world.

References Agol VI, Paul AV, Wimmer E (1999) Paradoxes of the replication of picornaviral genomes. Virus Res 62:129–147 Arias A, Ruiz-Jarabo C, Escarmis C, Domingo E (2004) Fitness increase of memory genomes in a viral quasispecies. J Mol Biol 339:405–412 Arnesjo B, Eden T, Ihse I, Nordenfelt E, Ursing B (1976) Enterovirus infections in acute pancreatitis—a possible etiological connection. Scand J Gastroenterol 11:645–649 Bae Y, Yoon J (1993) Determination of diabetogenicity attributable to a single amino acid, Ala776, on the polyprotein of encephalomyocarditis virus. Diabetes 42:435– 443 Bae Y, Eun M, Yoon J (1989) Genomic differences between the diabetogenic and nondiabetogenic variants of encephalomyocarditis virus. Virology 170:282–287 Belsham GJ, Sonenberg N (1996) RNA-protein interactions in regulation of picornavirus RNA translation. Microbiol Rev 60:499–511 Borman A, Le Mercier P, Girard M, Kean K (1997) Comparison of picornaviral IRESdriven internal initiation of translation in cultured cells of different origins. Nucleic Acids Res 25:925–932 Brown F, Talbot P, Burrows R (1973) Antigenic differences between isolates of swine vesicular disease virus and their relationship to coxsackie B5 virus. Nature 245:315–316 Chapman N, Ramsingh A, Tracy S (1997) Genetics of coxsackievirus virulence. Curr Top Microbiol Immunol 223:227–258 Daniels JB, Pappenheimer AM, Richardson S (1952) Observations on encephalomyelitis of mice (DA strain) J Exp Med 96:517–530 Domingo E, Escarmis C, Sevilla N, Moya A, Elena SF, Quer J, Novella IS, Holland JJ (1996) Basic concepts in RNA virus evolution. FASEB J 10:859–864 Domingo E, Escarmis C, Baranowski E, Ruiz-Jarabo C, Carrillo E, Nunez J, Sobrino F (2003) Evolution of foot-and-mouth disease virus. Virus Res 91:47–63 Dowdle WR, De Gourville E, Kew OM, Pallansch M, Wood D (2004) Polio eradication: the OPV paradox. Rev Med Virol 13:277–291

Evolution of Virulence in Picornaviruses

205

Drescher KM, Kono K, Bopegamage S, Carson SD, Tracy S (2004) Coxsackievirus B3 infection and type 1 diabetes development in NOD mice: insulitis determines susceptibility of pancreatic islets to virus infection. Virology 329:381–394 Duke G, Palmenberg AC (1989) Cloning and synthesis of infectious cardiovirus RNAs containing short, discrete poly (C) tracts. J Virol 63:1822–1826 Duke G, Osorio J, Palmenberg A (1990) Attenuation of Mengo virus through genetic engineering of the 5 noncoding poly (C) tract. Nature 343:474–476 Dunn G, Bradrick S, Chapman N, Tracy S, Romero J (2003) The stem loop II within the 5 nontranslated region of clinical coxsackievirus B3 genomes determines cardiovirulence phenotype in a murine model. J Infect Dis 15:1552–1561 Dunn JJ, Chapman NM, Tracy S, Romero JR (2000) Natural genetics of cardiovirulence in coxsackievirus B3 clinical isolates: localization to the 5 non-translated region. J Virol 74:4787–4794 Evans D, Dunn G, Minor PD, Schild GC, Cann AJ, Stanway G, Almond JW, Currey K, Maizel J (1985) Increased neurovirulence associated with a single nucleotide change in a noncoding region of the Sabin type 3 poliovaccine genome. Nature 314:548–550 Fotiadis C, Kilpatrick DR, Lipton HL (1991) Comparison of the binding characteristics to BHK-21 cells of viruses representing the two Theiler’s virus neurovirulence groups. Virology 182:365–370 Fu J, Rodriguez M, Roos RP (1990) Strains from both Theiler’s virus subgroups encode a determinant for demyelination. J Virol 64:6345–6348 Grant RA, Filman DJ, Fujinami RS, Icenogle JP, Hogle JM (1992) Three-dimensional structure of Theiler’s virus. Proc Nat Acad Sci U S A 89:2061–2065 Hahn H, Palmenberg A (1995) Encephalomyocarditis viruses with short poly (C) tracts are more virulent than their mengovirus counterparts. J Virol 69:2697–2699 Halim S, Ramsingh AI (2000) A point mutation in VP1 of coxsackievirus B4 alters antigenicity. Virology 269:86–94 Imrie CW, Ferguson J, Sommerville R (1977) Coxsackie and mumpsvirus infection in a prospective study of acute pancreatitis. Gut 18:53–56 Jnaoui K, Michiels T (1999) Analysis of cellular mutants resistant to Theiler’s virus infection: differential infection of L929 cells by persistent and neurovirulent strains. J Virol 73:7248–7254 Jun H, Kang Y, Yoon J, Notkins A (1997) Gain or loss of diabetogenicity resulting from a single point mutation in recombinant encephalomyocarditis virus. J Virol 71:9782–9785 Kandolf R, Sauter M, Aepinus C, Schnorr J, Selinka H, Klingel K (1999) Mechanisms and consequences of enterovirus persistence in cardiac myocytes and cells of the immune system. Virus Res 62:149–158 Kanno T, Inoue T, Mackay D, Kitching P, Yamaguchi S, Shirai J (1998) Viruses produced from complementary DNA of virulent and avirulent strains of swine vesicular disease virus retain the in vivo and in vitro characteristics of the parental strains. Arch Virol 143:1055–1062 Kanno T, Mackay D, Inoue T, Wilsden G, Yamakawa M, Yamazoe R, Yamaguchi S, Shirai J, Kitching P, Murakami Y (1999) Mapping the genetic determinants of pathogenicity and plaque phenotype in swine vesicular disease virus. J Virol 73:2710–2716

206

S. Tracy et al.

Kanno T, Mackay D, Wilsden G, Kitching P (2001) Virulence of swine vesicular disease virus is determined at two amino acids in capsid protein VP1 and A protease. Virus Res 80:101–107 Kawamura N, Kohara M, Abe S, Komatsu T, Tago K, Arita M, Nomoto A (1989) Determinants in the 5 noncoding region of poliovirus Sabin 1 RNA that influence the attenuation phenotype. J Virol 63:1302–1309 Kennedy JD, Talbot IC, Tanner MS (1986) Severe pancreatitis and fatty liver progressing to cirrhosis associated with coxsackie B4 infection in a three year old with alpha1-antitrypsin deficiency. Acta Paediatr Scand 75:336–339 Knowles N, McCauley JW (1997) Coxsackievirus B5 and the relationship to swine vesicular disease virus. Curr Top Microbiol Immunol 223:153–168 La Monica N, Racaniello VR (1989) Differences in replication of attenuated and neurovirulent polioviruses in human neuroblastoma cell line SH-SYY J Virol 63:2357– 2360 Lal S, Fowler D, Losasso C, Berg G (1988) Coxsackie virus-induced acute pancreatitis in a long-term dialysis patient. Am J Kidney Dis 11:434–436 LaRue R, Myers S, Brewer L, Shaw D, Brown C, Seal B, Njenga M (2003) A wildtype porcine encephalomyocarditis virus containing a short poly (C) tract is pathogenic to mice, pigs, and cynomolgus macaques. J Virol 77:9136–9146 Le S, Zuker M (1990) Common structures of the 5 non-coding RNA in enteroviruses and rhinoviruses. J Mol Biol 216:729–741 Lindberg A, Andersoon P, Savolainen C, Mulders MN, Hovi T (2003) Evolution of the genome of human enterovirus B: incongruence between phylogenies of the VP1 and CD regions indicates frequent recombination within the species. J Gen Virol 84:1223–1235 Lipton HL, Dal Canto MC (1979) The TO strains of Theiler’s viruses cause ‘slow-virus like’ infection in mice. Ann Neurol 6:25–28 Lorch Y, Freidmann A, Lipton HL, Kotler M (1981) Theiler’s murine encephalomyelitis cirrus group includes two distinct genetic subgroups that differ pathologically and biologically. J Virol 40:560–567 Luo M, Vriend G, Kamer G, Minor I, Arnold E, Rossmann MG, Boege U, Scraba DG, Duke G, Palmenberg A (1987) The atomic structure of Mengo virus at 3.0 A resolution. Science 235:182–191 Luo M, He C, Toth KS, Zhang CX, Lipton HL (1992) Three-dimensional structure of Theiler’s murine encephalomyelitis virus (BeAn strain) Proc Nat Acad Sci U S A 89:2409–2413 Macadam AJ, Pollard SR, Ferguson G, Dunn G, Skuce R, Almond J, Minor PD (1991) The 5 noncoding region of the type 2 poliovirus vaccine strain contains determinants of attenuation and temperature sensitivity. Virology 181:451–458 Macadam AJ, Pollard SR, Ferguson G, Skuce R, Wood D, Almond JW, Minor PD (1993) Genetic basis of attenuation of the Sabin type 2 vaccine strain of poliovirus in primates. Virology 192:18–26 Malnou C, Werner A, Borman AM, Westhof E, Kean K (2004) Effects of vaccine strain mutations in domain V of the internal ribosome entry segment compared in the wild type poliovirus type 1 context. J Biol Chem 279:10261–10269

Evolution of Virulence in Picornaviruses

207

Martin L, Duke G, Osorio J, Hall DJ, Palmenberg A (1996) Mutational analysis of the mengovirus poly (C) tract and surrounding heteropolymeric sequences. J Virol 70:2027–2031 Martin L, Neal Z, McBride M, Palmenberg A (2000) Mengovirus and encephalomyocarditis virus poly (C) tract lengths can affect virus growth in murine cell culture. J Virol 74:3074–3081 McCright IJ, Tsunoda I, Whitby FG, Fujinami RS (1999) Theiler’s viruses with mutations in loop I of VP1 lead to altered trophism and pathogenesis. J Virol 73:2814– 2824 Minor PD (1992) The molecular biology of poliovaccines. J Gen Virol 73:3065–3077 Minor PD (1993) Attenuation and reversion of the Sabin vaccine strains of poliovirus. Dev Biol Stand 78:17–26 Minor PD (2003) Polio vaccines and the cessation of vaccination. Exp Rev Vaccines 2:99–104 Oberste MS, Penaranda S, Pallansch M (2004) RNA recombination plays a major role in genomic change during circulation of coxsackie B viruses. J Virol 78:2948–2955 Ochs K, Zeller A, Saleh L, Bassili G, Song Y, Sonntag A, Niepmann M (2002) Impaired binding of standard initiation factors mediates poliovirus translation attenuation. J Virol 77:115–122 Omata T, Kohara S, Kuge S, Komatsu T, Abe S, Semler B, Kameda A, Itoh H, Arita M, Wimmer E, Nomoto A (1986) Genetic analysis of the attenuation phenotype of poliovirus type. J Virol 58:348–358 Osorio J, Martin L, Palmenberg A (1996) The immunogenic and pathogenic potential of short poly (C) tract Mengo viruses. Virology 223:344–350 Pallansch M, Roos RP (2001) Enteroviruses: polioviruses, coxsackieviruses, echoviruses, and newer enteroviruses. In: Knipe DM, Howley PM (eds) Fields virology. Lipincott Williams and Wilkins, Philadelphia, pp 723–776 Palmenberg A, Osorio J (1994) Cardioviral poly (C) tracts and viral pathogenesis. Arch Virol Suppl 9:67–77 Pevear DC, Calendoff M, Rozhon E, Lipton HL (1987) Analysis of the complete nucleotide sequences of the picornavirus Theiler’s murine encephalomyelitis virus indicates that it is closely related to cardioviruses. J Virol 61:1507–1516 Pevear DC, Borkowski J, Calendoff M, Oh CK, Ostrowski B, Lipton HL (1988) Insights into Theiler’s virus neurovirulence based on genomic comparison of the neurovirulent GDVII and less virulent BeAn strains. Virology 165:1–12 Racaniello V (1988) Poliovirus neurovirulence. Adv Virus Res 34:217–246 Racaniello VR, Ren R (1996) Poliovirus biology and pathogenesis. Curr Top Microbiol Immunol 206:305–325 Ramsingh AI (1997) Coxsackievirus and pancreatitis. Front Biosci 2:53–62 Ramsingh AI, Collins DN (1995) A point mutation in the VP4 coding sequence of coxsackievirus B4 influences virulence. J Virol 69:7278–7281 Ramsingh A, Salck J, Silkworth M, Hixson A (1989) Severity of disease induced by a pancreotropic coxsackie B4 virus correlates with the HKq locus of the major histocompatibility complex. Virus Res 14:347–358 Ramsingh AI, Lee WT, Collins DN, Armstrong LE (1999) T cells contribute to disease severity during coxsackievirus B4 infection. J Virol 73:3080–3086

208

S. Tracy et al.

Roberts L, Seamons R, Belsham GJ (1998) Recognition of picornavirus internal ribosome entry sites within cells; influence of cellular and viral proteins. RNA 4:520–529 Rodriguez M, David CS (1985) Demyelination induced by Theiler’s virus: influence of the H-2 haplotype. J Immunol 135:2145–2148 Ross ME, Onodera T, Hayashi K, Notkins A (1975) Virus-induced diabetes mellitus. V Biological differences between the M variant and other strains of encephalomyocarditis virus. Infect Immun Infect Immun 12:1224–1226 Rueckert RR (1996) Picornaviridae: the viruses and their replication. In Fields BN, Knipe DM, Howley PM (eds), Virology. Raven Press, New York, pp 609–654 Sabin AB (1955) Characteristics and genetic potentialities of experimentally produced and naturally occurring variants of poliomyelitis virus. Ann N Y Acad Sci 61:924– 939 Sabin AB, Boulger LR (1973) History of Sabin attenuated poliovirus oral live vaccine strains. J Biol Stand 1:115–118 Savolainen C, Blomqvist S, Hovi T (2003) Human rhinoviruses. Paediatr Respir Rev 4:91–98 Svitkin Y, Maslova S, Agol V (1985) The genomes of attenuated and virulent poliovirus strains differ in their in vitro translation efficiencies. Virology 147:243–252 Tam P, Messner R (1997) Coxsackievirus-induced chronic inflammatory myopathy: virus variants distinguish between acute cytopathic effects and pathogenesis of chronic disease. Virology 233:199–209 Tam P, Messner R (1999) Molecular mechanisms of coxsackievirus persistence in chronic inflammatory myopathy: viral RNA persists through formation of a double-stranded complex without associated genomic mutations or evolution. J Virol 73:10113–10121 Tam PE, Schmidt A, Ytterberg S, Messner R (1994) Duration of virus persistence and its relationship to inflammation in the chronic phase of coxsackievirus B1-induced murine polymyositis. J Lab Clin Med 123:346–356 Tam PE, Weberse-Sanders M, Messner R (2003) Multiple viral determinants mediate myopathogenicity in coxsackievirus B1-induced chronic inflammatory myopathy. J Virol 77:11849–11854 Tracy S, Chapman N, Tu Z (1992) Coxsackievirus B3 from an infectious cDNA copy of the genome is cardiovirulent in mice. Arch Virol 122:399–409 Tracy S, Chapman NM, Romero J, Ramsingh AI (1996) Genetics of coxsackievirus B cardiovirulence and inflammatory heart muscle disease. Trends Microbiol 4:175– 179 Tracy S, Chapman N, Mahy B (1997) The coxsackie B viruses. Curr Top Microbiol Immunol 223 Tracy S, Hofling K, Pirruccello S, Lane PH, Reyna SM, Gauntt C (2000) Group B coxsackievirus myocarditis and pancreatitis in mice: connection between viral virulence phenotypes. J Med Virol 62:70–81 Tracy S, Drescher KM, Chapman NM, Kim K-S, Carson SD, Pirruccello S, Lane PH, Romero JR, Leser JS (2002) Toward testing the hypothesis that group B coxsackieviruses (CVB) trigger insulin-dependent diabetes: Inoculating nonobese diabetic mice with CVB markedly lowers diabetes incidence. J Virol 76:12097– 12111

Evolution of Virulence in Picornaviruses

209

Tsunoda I, Iwasaki Y, Terunuma H, Sako K, Fujinami RS (1996) A comparative study of acute and chronic diseases induced by two subgroups of Theiler’s murine encephalomyelitis virus. Acta Neuropathol (Berl) 91:595–602 Tsunoda I, Wada Y, Libbey JE, Cannon TS, Whitby FG, Fujinami RS (2001) Prolonged gray matter disease without demyelination caused by Theiler’s murine encephalomyelitis virus with a mutation in VP2 puff B J Virol 75:7494–7505 Tu Z, Chapman N, Hufnagel G, Tracy S, Romero JR, Barry WH, Currey K, Shapiro B (1995) The cardiovirulent phenotype of coxsackievirus B3 is determined at a single site in the genomic 5 non-translated region. J Virol 69:4607–4618 Ursing B (1973) Acute pancreatitis in coxsackie B infection. BMJ 3:524–525 Wada Y, McCright IJ, Whitby FG, Tsunoda I, Fujinami RS (1998) Replacement of loop II of VP-1 of the DA strain with loop II of the GDVII strain of Theiler’s murine encephalomyelitis virus alters neurovirulence, viral persistence and demyelination. J Virol 72:7557–7562 Ward C, Stokes M, Flanegan JB (1988) Direct measurement of the poliovirus RNA polymerase error frequency in vitro. J Virol 62:558–562 Yoon J, McClintock P, Onodera T, Notkins A (1980) Virus-induced diabetes mellitus. XVIII. Inhibition by a nondiabetogenic variant of encephalomyocarditis virus. J Exp Med 152:878–892 Zhang G, Wilsden G, Knowles NJ, McCauley JW (1993) Complete nucleotide sequence of a coxsackie B5 virus and its relationship to swine vesicular disease virus. J Gen Virol 74:945–953 Zhou L, Lin X, Green TJ, Lipton HL, Luo M (1997) Role of sialyloligosaccharide binding in Theiler’s virus persistence. J Virol 71:9701–9712 Zhou L, Luo Y, Wu Y, Tsao J, Luo M (2000) Sialylation of the host receptor may modulate entry of the demyelinating persistent Theiler’s virus. J Virol 74:1929–1937 Zuker M (2003) Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res 31:3406–3415

CTMI (2006) 299:211–259 c Springer-Verlag Berlin Heidelberg 2006


Molecular Mechanisms of Poliovirus Variation and Evolution V. I. Agol 1, 2 (u) 1 M.P. Chumakov Institute

of Poliomyelitis and Viral Encephalitides, Russian Academy of Medical Sciences, 142782, Russia [email protected] 2 M.V. Lomonosov Moscow State University, Moscow 119899, Russia

1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213

2

Poliovirus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213

3 3.1 3.1.1 3.1.2 3.1.3 3.2 3.2.1 3.2.2

Origin of Genetic Variability Replication Errors . . . . . . . Point Mutations . . . . . . . . . Slippage . . . . . . . . . . . . . . Rearrangements . . . . . . . . . Nonreplicative Alterations . . Point Mutations . . . . . . . . . Recombination . . . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

214 214 215 216 216 218 218 218

4 4.1 4.2 4.2.1 4.2.2 4.2.3

Variability, Fitness and Phenotype . Limits of Variability . . . . . . . . . . . The Space of Allowed Sequences . . Coding Sequences . . . . . . . . . . . . RNA cis-Elements . . . . . . . . . . . . . Intragenomic Interactions . . . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

219 219 220 220 221 224

5 5.1 5.2 5.3

Mutation Fixation . . . . . Rate of Mutation Fixation Selection . . . . . . . . . . . . Bottlenecking . . . . . . . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

225 225 225 226

6

Changes Occurring in Tissue Culture . . . . . . . . . . . . . . . . . . . . . . . . . . 227

7 7.1 7.1.1 7.1.2 7.2 7.3 7.4 7.5

Evolution in the Organism . . . . . . . . . . Fitness Changes . . . . . . . . . . . . . . . . . . Elimination of Adverse Mutations . . . . . Recombination . . . . . . . . . . . . . . . . . . . Antigenic Changes . . . . . . . . . . . . . . . . Host Range Changes . . . . . . . . . . . . . . . Potential to Establish Persistent Infection Neutral Mutations . . . . . . . . . . . . . . . .

. . . .

. . . .

. . . . . . . .

. . . .

. . . . . . . .

. . . .

. . . . . . . .

. . . .

. . . . . . . .

. . . .

. . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

231 231 231 232 233 235 235 235

212

V. I. Agol

7.6 7.7

Co-existence of Different Lineages . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236 Rate of Evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236

8 8.1 8.2 8.3 8.4 8.5 8.6

Evolution in Human Populations . Dissemination Factors . . . . . . . . Antigenic Variability . . . . . . . . . . Pattern of Mutation Accumulation Contribution of Recombination . . Rate of Evolution . . . . . . . . . . . . Geographical Constraints . . . . . .

9 9.1 9.2

Relationships Between Poliovirus and Other Picornaviruses . . . . . . . . . 242 Engineered Viable Chimeric Polioviruses . . . . . . . . . . . . . . . . . . . . . . . 243 Natural Recombination Between Poliovirus and Other Picornaviruses . . . 243

10

A Few Remarks on Poliovirus Macroevolution . . . . . . . . . . . . . . . . . . . 244

11

Towards the Eradication of Poliovirus . . . . . . . . . . . . . . . . . . . . . . . . . 245

12

Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

238 238 239 239 240 241 242

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247 Abstract Replication of poliovirus RNA is accomplished by the error-prone viral RNAdependent RNA polymerase and hence is accompanied by numerous mutations. In addition, genetic errors may be introduced by nonreplicative mechanisms. Resulting variability is manifested by point mutations and genomic rearrangements (e.g., deletions, insertions and recombination). After description of basic mechanisms underlying this variability, the review focuses on regularities of poliovirus evolution (mutation fixation) in tissue cultures, human organisms and populations. Abbreviations IRES OPV ORF RdRP UTR

Internal ribosome entry site Oral polio vaccine Open reading frame RNA-dependent RNA polymerase Untranslated region

Molecular Mechanisms of Poliovirus Variation and Evolution

213

1 Introduction Poliovirus is a representative of the Picornaviridae family. The molecular biology of picornaviruses is amongst the best-studied branches of fundamental virology (Semler and Wimmer 2002; Racaniello 2002). On the other hand, picornaviruses are etiological agents of important human and animal diseases such as poliomyelitis, hepatitis A, foot-and-mouth disease and a number of others. Understanding the evolutionary aspects is crucial for the both parts of picornavirology: fundamental and medical/veterinarian. There are several excellent recent reviews on picornavirus evolution (Domingo et al. 1999; 2002a; Gromeier et al. 1999). Here, we shall focus specifically on variability and evolution of poliovirus. After a brief description of the basic mechanisms underlying viral genome variability, we shall analyze regularities of evolution of poliovirus in tissue cultures, organisms and human populations. Additional references may be found in our recent related review (Agol 2002a).

2 Poliovirus Poliovirus is a small virus encapsidating within an icosahedral capsid a single 7.4-kb RNA molecule of positive (mRNA-like) polarity. It is a human pathogen, the causative agent of poliomyelitis (Gromeier and Nomoto 2002). Poliovirus genome has an organization typical of other picornaviruses and, in particular, of enteroviruses (Fig. 1). It consists of a single open reading frame (ORF) flanked by 5 and 3 untranslated regions (UTR), approximately 740 and 70 nt long, respectively. The ORF encodes a polyprotein of approximately 2,200 residues, which is eventually processed by viral proteolytic activities into 11 “mature” polypeptides, with some of the processing intermediates serving as distinct functional units. Four of the final polypeptides (VP1–4), corresponding to the N-terminal region of the polyprotein, are structural components of the viral capsid, whereas the remainder polypeptides are involved in the replication of the viral genome, proteolytic processing of the polyprotein, and a variety of functions directly or indirectly ensuring efficient viral reproduction (Agol 2002b). The viral RNA also possesses a variety of cis-acting elements. The best studied among them are replicative elements oriL (known also as the cloverleaf element) and oriR in the 5 UTR and 3 UTR, and cre (oriI) within the region encoding nonstructural protein 2C (Paul 2002) as well as a translational ciselement, internal ribosome entry site (IRES) in the 5 UTR, responsible for the

214

V. I. Agol

Fig. 1 A schematic representation of the poliovirus genome. The 7.4-kb RNA molecule contains a single ORF, which encodes the polyprotein that is co-translationally and post-translationally processed into mature proteins. The ORF is composed of three major sections: P1–P3. Four capsid proteins (VP1–VP4) are encoded in P1, whereas P2 and P3 code for proteins involved in polyproteins processing (proteases 2A and 3C) and genome replication (RNA-dependent RNA polymerase 3D, the primer protein VPg, and proteins with poorly defined specific functions, 2A, 2B and 3A). The ORF is flanked by 5 and 3 untranslated regions, terminating with VPg and poly(A), respectively

cap-independent internal initiation of translation of the viral RNA (Ehrenfeld and Teterina 2002). Traditionally, and according to the current official taxonomy, three serotypes of poliovirus constitute a separate species within the Enterovirus genus. This classification seems warranted from the clinical and pathogenetic points of views, because polioviruses may induce a characteristic disease and, compared to other enteroviruses, exploit CD155 as a cellular receptor (Rieder and Wimmer 2002). However, with regards to the genomic segment outside of the capsid-encoding region there is no clear borderline between polioviruses and a subset of the so-called cluster C enteroviruses, which includes a number of coxsackie A viruses (CAV) (Gromeier et al. 1999; Brown et al. 2003). Therefore, the proposal to re-classify these viruses and consider polioviruses as serotypes of a larger species (Brown et al. 2003) makes sense. We shall return to this problem later on.

3 Origin of Genetic Variability Variability of the poliovirus genome is a fundamental property underlying evolution of the virus, its pathogenic and epidemiological characteristics, as well as a major tool elucidating the genotype–phenotype relationships. Two distinct groups of mechanisms, replicative and nonreplicative, are responsible for the covalent alterations of the viral RNA. 3.1 Replication Errors Copying the viral RNA is accomplished with the assistance of other viral and host proteins by the product of viral gene 3Dpol , the RNA-dependent RNA

Molecular Mechanisms of Poliovirus Variation and Evolution

215

polymerase (RdRP). The enzyme is intrinsically capable of at least two types of errors, incorporation of incorrect nucleotides and premature termination, which may or may not be associated with template switching. Misincorporation results in point mutations, whereas template switching may generate intra- or intermolecular rearrangements. Slippage or shuttering, resulting in repeated copying of a nucleotide or oligonucleotide, while not well documented in the case of poliovirus RdRP, may also be added to the list of potential replicative errors. 3.1.1 Point Mutations Determination of the level of nucleotide misincorporation by using purified preparations of the enzyme yielded values in the range of 5 × 10−3 −10−5 (Ward et al. 1988; Wells et al. 2001; Arnold and Cameron 2004), with transitions occurring approximately ten times more frequently than transversions. However, one never can be sure that in vitro conditions used for such assays mimic adequately the real in vivo environment during the synthesis of viral RNA. Nevertheless, values within the same range, e.g., roughly 2 × 10−4 , have been obtained in genetic experiments where the error frequency was assayed by a phenotypic change (loss of dependence of growth on the presence of guanidine.HCl) associated with a single point mutation (de la Torre et al. 1990), and also upon sequencing multiple independent clones of the same viral harvest (Crotty et al. 2001). In this respect, the poliovirus RdRP does not appreciably differ from other picornavirus RdRPs (Sierra et al. 2000) or RdRPs of many other RNA-containing viruses (Drake and Holland 1999). Nevertheless, the level of poliovirus RdRP infidelity is not something absolutely constant. By selecting poliovirus variants resistant to antiviral drug ribavirin, it was possible to identify a single point mutation in the viral polymerase, which conferred a significant increase in fidelity (Pfeiffer and Kirkegaard 2003; Castro et al. 2005). An enhanced RdRP fidelity is accompanied with a decrease in the viral fitness, at least under certain conditions (Vignuzzi et al. 2005; Pfeiffer and Kirkegaard 2005). In particular, poliovirus with a more precise RdRP is less adaptable and competitive and is more attenuated. Some indirect evidence suggests that RdRPs of certain RNA viruses may be several orders of magnitude more accurate (Pugachev et al. 2004). Thus, the extent of faithfulness of the viral replication seems to be an evolutionary acquired property. An important corollary of the existing level of variability is that each progeny RNA molecule synthesized by the poliovirus RdRP harbors on average about one nucleotide difference compared to its template. In a population

216

V. I. Agol

containing 1–3 × 108 PFU (which may be produced by as few as 104 –105 cells), every double mutation is expected to be present (Crotty and Andino 2002). 3.1.2 Slippage Many RdRPs are capable of repeated copying of a single template nucleotide by a kind of slippage or shuttering, thereby generating a homopolymeric stretch, e.g., poly (A), or insertion of a single additional nucleotide. This property is an essential component of the transcription strategy of negative-strand RNA viruses. For the poliovirus RdRP, we are aware of no solid facts supporting this capacity, although some in vitro observation may be interpreted as evidence for such a slippage (Arnold and Cameron 1999). A similar explanation may be offered for changes in the length of homo-oligomeric stretches or nucleotide insertions occurring upon passages of certain poliovirus mutants (Gmyl et al., 1999). An alteration of internal poly (A) length was also reported for foot-andmouth disease virus (FMDV; Escarmis et al. 2002), supporting that slippage may be a common property of picornaviral RdRPs. It can be speculated that not only single nucleotides, but also short oligonucleotides may be used as templates for reiterated copying, resulting in generation of short tandem repeats. 3.1.3 Rearrangements A direct demonstration of the capacity of poliovirus RdRP to change template in in vitro experiments was provided only recently (Arnold and Cameron 1999; Chetverin et al. 2005). However, it was for long assumed that a variety of intra- and intermolecular rearrangements of the full-length poliovirus RNA observed in vivo are caused by a similar mechanism. A major class of poliovirus RNA rearrangements thought to be accomplished intramolecularly is represented by the so-called defective-interfering (DI) RNA species. Such molecules tend to accumulate upon high multiplicity passages and usually have extended, variable-length deletions within the region encoding capsid proteins (Kuge et al. 1986). The mechanism of generation of DI RNAs is unknown. According to one of the models, the termini of the fragments to be joined are brought in close proximity to each other by base pairing with an unrelated “guiding” sequence, while the region to be deleted is looped out and skipped by the viral RdRP (Kuge et al. 1986). Short deletions and insertions not infrequently found in RNAs of natural (cf., Toyoda et al. 1984) and laboratory (cf., Pilipenko et al. 1992a; Gmyl

Molecular Mechanisms of Poliovirus Variation and Evolution

217

et al. 1993) polioviruses could also be considered as originating, at least in part, through intramolecular template switching (Pilipenko et al. 1995). Several mechanisms may theoretically be responsible for such switching: (a) skipping of a few template nucleotides by the sliding elongation complex; (b) skipping, also without dissociation of the elongation complex, of longer looped-out nucleotide stretches within an appropriate secondary structure; and (c) dissociation of the complex and landing of the 3 -end of the nascent RNA strand and RdRP on a new site of the same template. It should be noted, however, that there is no experimental evidence for the existence of the two former mechanisms. With regards to the third one, there is no proof that it can occur in cis. In other words, we do not know whether the RdRP switches between different loci of the same template or between different, though identical, RNA template molecules. The two latter types of rearrangements may share common mechanisms. A classical form of the intermolecular RNA rearrangement is genetic recombination, first discovered (Ledinko 1963) and extensively studied in poliovirus (reviewed in Agol 1997). It may occur between the same serotypes and between different serotypes, though much less frequently (Tolskaya et al. 1983). In the framework of the template switch model, there are several major questions to be answered. The first one concerns the nature of the template, positive or negative RNA, between which poliovirus RdRP jumps. It is generally accepted that the switch occurs predominantly, or exclusively, during the (–)RNA strand synthesis, with viral (+)RNA molecules, present in excess and in single-stranded form, serving as primary and secondary templates (Kirkegaard and Baltimore 1986). An important and unsolved question concerns conditions favoring premature termination of elongation of the nascent chain on the primary template. It was proposed that termination might be facilitated by secondary structure elements (Romanova et al. 1986) and a high A-U content of the template (King 1988), but both these proposals are based on indirect evidence. It may be noted that the level of processivity of viral and cellular DNA-dependent RNA polymerases and of RdRPs of negativestrand RNA viruses may be controlled by interactions with protein “factors”. If a similar mechanism operates in poliovirus RNA synthesis, it would affect the frequency of recombination. Natural recombinants are predominantly homologous, preserving the consensus structures of the UTRs and ORF, with very rare alterations in the area of crossovers. A major factor contributing to such a precision is certainly selection for viable viruses, although it was reported that recombination could be predominantly precise even under nonselective conditions (Jarvis and Kirkegaard 1992). However, even if selection is a major factor responsible for precision of crossovers, a relatively high frequency of homologous recombi-

218

V. I. Agol

nation forces one to think about the mechanism potentially able to facilitate the correct choice of the landing site on the acceptor template. The precise landing is likely determined by complementarity of the 3 terminal sequence of the nascent strand to a homologous region of the secondary template. However, such complementary stretches in intertypic poliovirus recombinants may be composed of only a few nucleotides (Kirkegaard and Baltimore 1986; Romanova et al. 1986; Tolskaya et al. 1986; Jarvis and Kirkegaard 1992). The viral RNA should have many such sites, and the correct choice between them is an obvious problem. Bringing together homologous regions of the primary and secondary templates may be facilitated by the formation of heteroduplexes involving elements of secondary structures (Romanova et al. 1986; Tolskaya et al. 1986), but direct confirmation of such a mechanism has yet to be provided. 3.2 Nonreplicative Alterations 3.2.1 Point Mutations Practically no attention has as yet been paid to a feasible nonreplicative mechanism of point mutation generation. Enzyme-promoted RNA base deamination serves essential function(s) in several viral systems. Adenosine deaminases acting on RNA (ADARs) are involved in editing hepatitis delta transcripts (Wong and Lazinski 2002) and possibly some other RNA viruses (Bass 2002; Martinez and Melero 2002; Bishop et al. 2004). It is unknown whether poliovirus-specific RNA species, and in particular double-stranded replicative form, may be used by ADARs as substrates. However, such a possibility should not be ignored. Nonenzymatic deamination of nucleotides cannot be excluded either. 3.2.2 Recombination Recently discovered nonreplicative RNA recombination (Chetverina et al. 1999; Gmyl et al. 1999) may also contribute to the variability of viral genomes. Fragments of poliovirus RNA devoid of replicative and/or translational capacity, being transfected into susceptible cells, are able to generate viable viral progeny. The fragments can be joined in different ways. The 5 - or 3 terminal nucleotide of a fragment may be inserted into an internal position of a partner RNA molecule resulting in the acquisition, by the recombinant genome, of the former fragment in its entirety. Alternatively, internal crossovers may be

Molecular Mechanisms of Poliovirus Variation and Evolution

219

accomplished between internal positions of overlapping portions of two fragments. Both imprecise and precise crossovers could be generated in the both ways (Gmyl et al. 1999, 2003). The distribution of crossovers does not appear to be random, and some clustering (hot spots) is evident. Nevertheless, a high degree of promiscuity in the location of crossovers is also obvious. There is no currently available tool to judge what type of recombination, replicative or nonreplicative, is responsible for a particular crossover between two poliovirus genomes under natural conditions. A similar mechanism may also provide for insertion of host sequences into the viral genome (unpublished observations).

4 Variability, Fitness and Phenotype Given such a level of viral RNA variability, what are its biological consequences in terms of general fitness and specific phenotype? To discuss this point, we must first briefly consider the extent of phenotypic variability of polioviruses. 4.1 Limits of Variability Variability of biological properties of poliovirus (reviewed in Agol 2002a) may concern its sensitivity/resistance to antibodies, inhibitors, changed temperature and other environmental conditions, the pathogenicity and, to some extent, the host range (capacity to infect certain species of cells or organisms). In addition, virus isolates may differ from one another by their general fitness, as judged by the amount of infectious progeny per infected cell, the time-course of reproduction, and competitive strength. What are the theoretical limits of this variability? In other words, what phenotypic changes would allow us to state that the altered virus is no longer a poliovirus? This important question may have several answers. One of them may define poliovirus as a virus whose RNA has a primary structure (especially within the polyprotein ORF) consistent with being a picornavirus genome, whose virions have a poliovirus-like architecture and use CD155 as a cellular receptor. By this definition, if a poliovirus is modified, by accumulated mutations or engineering, in such a way that it will be unable to infect a cell by using CD155 as a receptor, it should no longer be considered a poliovirus.

220

V. I. Agol

4.2 The Space of Allowed Sequences In view of such a variability of poliovirus phenotypes, it is not surprising that the level of observed genetic variability is also rather high. A very rough idea of this level can be derived simply from inspection of aligned RNA sequences of the natural poliovirus isolates. Although this alignment has not yet been analyzed thoroughly enough, it is clear that the virus may find itself quite comfortable in spite of a huge variability of its genomic nucleotide sequences. What is then the space of permitted mutations in poliovirus RNA compatible with the viability, fitness and its very identity as the poliovirus genome? Although definitive answer to this question is lacking, some empirical data may be relevant. 4.2.1 Coding Sequences Considering the coding sequences, we should first distinguish between synonymous and nonsynonymous (missense) nucleotide substitutions. At the first approximation, the former may be considered neutral. However, there are important exceptions: cis-acting RNA signals mapping to the ORF should be mentioned first. Currently, only one such signal is well characterized, it is cre (oriI), which serves as a template for uridylylation of VPg by the viral RdRP and is represented by a hairpin-like secondary structure within the sequence encoding protein 2C (Goodfellow et al. 2000; reviewed in Paul 2002). Obviously, synonymous mutations, if they affect essential residues of cre, could be detrimental to virus viability. However, some other synonymous positions within the poliovirus polyprotein ORF appear to be significantly more conserved than others (Gavrilin et al. 2000; and unpublished observations). Several hypothetical explanations could be offered for this phenomenon, including biases in the codon usage or requirements in specific or nonspecific secondary structures in certain regions of the viral RNA, but none of these explanations is as yet supported by direct evidence. For obvious reasons, nonsynonymous sites exhibit less flexibility. Certain positions, e.g., corresponding to the catalytic residues of viral enzymes or the cleavage sites recognized by viral proteases, are very strictly conserved. A high degree of conservation is also characteristic for amino acid residues involved in maintenance of specific folding of structural and nonstructural proteins. Nevertheless, many nonsynonymous positions exhibit a significant promiscuity, especially with regards to the so-called conservative substitutions (e.g. replacements of a hydrophobic residue by another hydrophobic residue of a similar size, etc.).

Molecular Mechanisms of Poliovirus Variation and Evolution

221

It should be noted, however, that experimental substitution of certain highly conserved amino acid residues would not necessarily lead to a detectable phenotypic change. For example, poliovirus protease 3C has a conserved ThrGlyLys (TGK) motif, thought to be involved in interaction with hairpin d of the cloverleaf-like replicative element oriL in the 5 UTR (Andino et al. 1993). However, randomization of the TGK-coding sequence of the viral RNA yields numerous viable mutants, including large-plaque-forming ones, in which either T or K, or both (but not G) were replaced by unrelated amino acid residues (D. Bakhmutov et al., unpublished observations). 4.2.2 RNA cis-Elements The poliovirus IRES contains several secondary structure elements, which are highly conserved not only between different polioviruses but also within a broad group of picornaviruses, including enteroviruses and rhinoviruses (Pilipenko et al. 1989). Mutations, which do not alter the conserved IRES structure, are generally not accompanied by phenotypic changes, while destabilization of this structure or alterations at some other critical positions may result in cell-specific translation deficiency (Svitkin et al. 1985, 1988; Haller et al. 1996), temperature sensitivity (Macadam et al. 1991), altered host range (La Monica and Racaniello 1989) and attenuation of neurovirulence (Evans et al. 1985; Skinner et al. 1989). Phenotypic changes caused by an IRES mutation may be reversed by second-site mutations (pseudoreversions) restoring the altered structure (Skinner et al. 1989; Muzychenko et al. 1991). The IRES function requires its interactions with ribosomes and a variety of host proteins (e.g., conventional and IRES-specific initiation factors; cf. Pestova et al. 2001; Ehrenfeld and Teterina 2002). Thus, in addition to the necessity of maintaining the overall spatial structure, variability of the poliovirus IRES is restricted by the demand to retain its capacity for numerous specific RNA–protein interactions. There is a translational cis-element in poliovirus (and other picornaviruses) RNA, which may or may not be considered a part of the IRES. It is represented by a combination of an oligopyrimidine stretch just downstream of the so-called attenuation hairpin (stem-loop V) of the IRES and an AUG triplet, which in the poliovirus genome is cryptic (noninitiating) and precedes the polyprotein ORF by approximately 150 nt. The optimal distance between these two elements in poliovirus RNA is 22 nt, and an increase or decrease in this distance is accompanied by lowering the translation efficiency and growth potential (Pilipenko et al. 1992a; Gmyl et al. 1993).

222

V. I. Agol

Thus, the repertoire of mutations in poliovirus IRES is relatively limited. However, its complete replacement by an IRES of rhinovirus (Gromeier et al. 1996) or structurally unrelated IRESes of encephalomyocarditis virus (Alexander et al. 1994; Rohll et al. 1994) or even by a non-picornavirus (hepatitis C virus) IRES (Lu and Wimmer 1994) still produced viable polioviruses, usually with somewhat altered biological properties, e.g. small plaque phenotype and attenuated neurovirulence. This suggests that the IRES-containing segment of poliovirus RNA hardly has any essential functions other than promoting cap-independent initiation of translation. An approximately 100-nt-long spacer between the last conserved hairpin (stem-loop VI or domain E) and the initiator AUG codon, which has no clear known functions, exhibits, in wild-type polioviruses, a high level of sequence variability but significant conservation with respect to the length (Poyry et al. 1992). Despite this length conservation, the entire spacer and some upstream nucleotides could be removed without dramatic alteration of the in vitro phenotype, though sometimes with a marked decrease in neurovirulence (Kuge and Nomoto 1987; Iizuka et al. 1989; Pilipenko et al. 1992a; Gmyl et al. 1993; Slobodskaya et al. 1996). The 5 terminal, roughly 100-nt-long piece of the poliovirus (as well as other enteroviruses and rhinoviruses) RNA folds into a conserved secondary structure, called cloverleaf or oriL, involved in initiation of synthesis of (+) and (–) RNA strands (Lyons et al. 2001; Paul 2002) as well as in control of initiation translation (Ehrenfeld and Teterina 2002). These functions require formation of ribonucleoprotein complexes with a variety of viral and host proteins and hence conservation of signals specifically recognized by appropriate ligands. Nevertheless, at least some of these presumably highly structurally restricted and evolutionary conserved elements, such as hairpin d, could be modified in a variety of ways without loss of the viral viability (D. Bakhmutov et al., unpublished observations). The 5 terminal portions of the RNAs of picornaviruses other than entero- and rhinoviruses possess no analogous cloverleaf structure, and we are aware of no successful attempts to use them for the replacement of the poliovirus oriL. The 3 terminal replicative element, oriR, of poliovirus 3 UTR RNA exhibits quite paradoxical properties. On the one hand, it has a conserved secondary structure composed of two hairpins, the mutual orientation of which is stabilized by tertiary kissing interactions between the loops of these hairpins. Destabilization of this complex architecture is accompanied by a marked impairment or complete inhibition of viral RNA replication, indicating that oriR is an important replicative element (Pilipenko et al. 1996; see also Melchers et al. 2000 for related observations with coxsackievirus B3). Nevertheless, the poliovirus 3 UTR may be replaced by structurally unrelated rhinovirus 3 UTR

Molecular Mechanisms of Poliovirus Variation and Evolution

223

(Rohll et al. 1995) or even completely deleted (Todd et al. 1997a) without the loss of viral viability and with a relatively mild defect in viral RNA replication in HeLa cells. Thus, poliovirus oriR may be considered a representative of a distinct class of control elements, deletion of which is accompanied with a much lesser functional deficiency than small structural modifications (Agol et al. 1999). Such paradoxical properties of oriR may hypothetically have a simple explanation. One may assume that the 3 terminus of the poliovirus RNA template becomes accessible to the viral RdRP only after interaction of oriR with appropriate viral and/or host proteins. Alteration of the oriR structure diminishes or eliminates its capacity to form such RNP complexes. However, oriR deletion may perhaps (partially) unmask the template 3 terminus, making RNP formation unnecessary. The situation with poliovirus oriR is also illuminating in another respect. While removal of oriR, as stated above, only moderately affects the efficiency of poliovirus reproduction in HeLa cells, the in vitro and in vivo growth potential of the truncated genome in cells of neural origin is suppressed dramatically, i.e., by several orders of magnitude (Brown et al. 2004). It would be important to know whether this deficiency also extends to the virus growth in human gastrointestinal tract, the natural niche of the virus. In any case, cell or tissue specificity of viral functions may be an important factor in limiting the space of allowed sequences of a given RNA element. The third replicative cis-element, cre or oriI, is a hairpin structure within the 2C-coding sequence (Goodfellow et al. 2000) functioning as a template for VPg uridylylation (Paul et al. 2000). VPg-pU-pU generated in this reaction is used for the initiation of viral positive RNA strands (Murray and Barton 2003; Morasco et al. 2003; Goodfellow et al. 2003b); the role of cre and credependent uridylylation in the initiation of negative RNA strand appears more controversial (van Ooij et al. 2005). Elucidation of the structural requirements of cre is hampered by the location of this element within the 2C-coding sequence. This problem may be obviated by transposing cre into a promiscuous region of 5 UTR, a procedure which does not interfere with the cre function (Yin et al. 2003; Goodfellow et al. 2003a). It turned out that only a few residues within the loop and the upper stem of poliovirus cre are essential for VPg uridylylation. It is appropriate to note that this reaction may occur, though less efficiently, in a cre-independent mode by using poly (A) as a template (Paul et al. 2000), but the physiological relevance of this reaction remains to be elucidated. Thus, on the one hand, the RNA cis-elements involved in translation and replication of poliovirus genome exhibit a high degree of conservation, suggesting that their natural structures confer a significant selective advantage. On the other hand, markedly altered or even completely unrelated RNA struc-

224

V. I. Agol

tures also appear to be able to fulfill many, if not all, of the essential functions of these elements. It cannot be ruled out that there exist additional, not yet characterized RNA cis-elements involved in other viral functions, e.g., encapsidation. 4.2.3 Intragenomic Interactions Until this point, consideration of constraints on the variability of coding sequences and RNA cis-elements did not take into account a circumstance of fundamental significance: the participants of the viral reproductive machinery do not work as separate entities but rather form a highly interactive (in a sense, even inseparable) community. The functions needed to ensure viral reproduction are highly coordinated in time and intracellular space, and the structure of the relevant viral functionaries, be it protein or RNA motifs, should be adapted for such coordination. In simpler words, viral translation, replication and other functions involve numerous interactions between virus-specific macromolecules (protein–protein, protein–RNA, RNA–RNA). The necessity of preserving these interactions is a major factor limiting the extent of the viral RNA variability. As mentioned above, the cloverleaf replicative cis-element oriL contains a stem-loop structure, called hairpin d, which is fairly well conserved among enteroviruses and rhinoviruses (Rivera et al. 1988). Genetic and biochemical experiments strongly suggest that the loop of this hairpin interacts with the 3C moiety of the virus-specific precursor protein 3CD and that impaired RNA synthesis (and viral reproduction) caused by a mutation in domain d could be suppressed by an alteration of the 3C structure (Andino et al. 1990; 1993). Therefore, the variability and evolution of these two elements are expected to be coupled, at least to some extent. Indeed, simultaneous randomization of nucleotides in the loop of domain d and the region encoding the relevant part of 3C protein resulted in a larger repertoire of mutations found in these regions of rescued viruses, as compared to the viruses rescued after transfection with RNAs harboring the individually randomized elements (D. Bakhmutov et al., unpublished observations). Importantly, the constraints on the 3C (or 3CD) structure are imposed by the necessity of its functional interaction not only with oriL, but also with other viral RNA cis-elements, such as cre (oriI) (Yin et al. 2003; Yang et al. 2004), and proteins, such as 3AB (Molla et al. 1994; Xiang et al. 1995a), as well as the necessity to perform its essential role in the polyprotein processing. Functional interdependence between IRES and protease 2A was also reported (Macadam et al. 1994; Rowe et al. 2000). Other viral proteins similarly participate in multiple interactions (Cuconati et al.

Molecular Mechanisms of Poliovirus Variation and Evolution

225

1998; Xiang et al. 1998), and maintenance of these interactions obviously requires either a high level of genomic conservation or concerted changes at different genomic loci. There are examples of functional interaction between diverse RNA ciselements, e.g., between oriL (the cloverleaf), on the one hand, and cre and oriR, on the other (Lyons et al. 2001; Paul 2002), or between IRES and oriR (Dobrikova et al. 2003). Such interactions often involve participation of protein bridges. All these complicated cross-talks certainly affect the phenotypic significance of relevant mutations. Summing up, we can conclude that there are numerous restrictions on the variability of coding and noncoding nucleotides in poliovirus RNA. Nevertheless, the remaining space of allowed mutations appears to be extremely high. The balance between the allowed and forbidden is, however, not clear.

5 Mutation Fixation Above, we have considered molecular mechanisms generating poliovirus variability and some limitations making certain mutations incompatible with, or at least detrimental to, viral viability. What factors result in fixation of a given mutation? This section will present a brief overview of the problem, while a more detailed description of appropriate factors will be given in Sects. 6–8. 5.1 Rate of Mutation Fixation As indicated above, picornavirus genome replication is intrinsically errorprone, and this is reflected in a relatively rapid natural evolution of the viral RNA. It is widely accepted that approximately 1%–2% of nucleotides in the VP1-coding region of natural poliovirus lineages change during 1 year of circulation (Kew et al. 1995), although the rate may fluctuate significantly (Hovi et al. 2004). Moreover, different parts of the poliovirus RNA may evolve at nonuniform rates, and the rate of mutation accumulation is not necessarily linear with time (Gavrilin et al. 2000). The exact reason(s) for the deviation from the molecular clock remains unknown, although different hypotheses may be put forward (cf., Sect. 7.6). 5.2 Selection Theoretically, a mutation may confer selective advantage or disadvantage or be neutral with regards to viral fitness. In practice, hardly any examples of

226

V. I. Agol

advantageous mutations in naturally circulating wild-type polioviruses have been documented. Experimental broadening, by appropriate selection, of the very strict (primate) host range of the virus was reported (Li and Habel 1951; Roca-Garcia et al. 1953; Couderc et al. 1993; Colston and Racaniello 1995; Jia et al. 1999), although, happily, no natural poliovirus host-switch has ever been observed. A more common phenomenon consists in antigenic changes in capsid proteins possibly helping escape from immunological pressure either in vitro (Emini et al. 1982; Minor et al. 1983) or in vivo (Minor et al. 1986; Huovilainen et al. 1987). The biological relevance of this type of poliovirus variability is in fact not as clear as it may appear (see Sect. 7.2). There are countless ways to generate viral variants with a decreased reproductive potential. Variants, in which the phenotype of such fitness-decreasing mutations (e.g., of attenuating mutations of the Sabin strains) is fully or partially suppressed, can usually be readily selected upon viral passages. The suppression may result from either true reversions or second-site mutations (pseudoreversions), illustrating the enormous plasticity and robustness of the viral genome. Acquisition of a novel advantageous phenotype in the above cases is obviously due to the selection of appropriate winners from preexisting variants in intrinsically vastly heterogeneous viral populations. It is clear that the genome of a variant carrying an advantageous mutation may harbor additional nucleotide substitutions with no or minor effects on the viral fitness. Such passenger mutations, even if they are fitness-decreasing, exhibit increased chances to be fixed in the viral progeny. The probability of fixation of a given adaptive mutation in a viral population depends, in addition to other factors, on the previous history of this population. In other words, viral populations have a kind of genetic memory (Domingo et al. 2002a, 2002b). This is evidenced by a more efficient capacity of a population to adapt to certain conditions if it had already experienced a similar adaptation in its previous evolution, even when the relevant mutations cannot be revealed by the genome sequencing. The appropriate variants may exist as a tiny minority for some time but, in the absence of selection, not indefinitely (Ruiz-Jarabo et al. 2003). Although appropriate experiments with poliovirus were not reported, its memory (or, more accurately, the memory of its populations) should hardly be weaker. 5.3 Bottlenecking As already mentioned, not only advantageous mutations are fixed. The major mechanism responsible for the fixation of neutral and even detrimental mutations consists in random sampling of a clone or a small population from

Molecular Mechanisms of Poliovirus Variation and Evolution

227

a large heterogeneous viral population. This mechanism is known as bottlenecking. In vitro, the bottlenecking may be readily achieved by plaque cloning, especially if this procedure is performed repeatedly. Detailed studies of the outcome of in vitro bottlenecking were conducted with another picornavirus, FMDV (Escarmis et al. 1996, 2002; Lazaro et al. 2003). With regards to poliovirus, relevant results were reported by Crotty et al. (2001), who demonstrated that independent viral clones harbored an apparently random distribution of mutations. Bottlenecking likely also occurs during natural poliovirus infection, although this important point was not studied thoroughly enough. The biological consequences of bottlenecking depend on a variety of factors, e.g., the size and heterogeneity of the donor population, the number of infectious virions effectively transferred to the recipient host and others. Since the probability of harboring a disadvantageous, as compared to advantageous, mutation is obviously higher, successive bottlenecking may result in a steady decrease in viral fitness. Theoretically, the viral lineage may even be completely extinguished. This phenomenon is called the Muller’s ratchet. Studies with FMDV, however, demonstrate that during such evolution, stochastic fitness-increasing mutations may improve the phenotype, and the virus lineages are more robust than one could expect (Escarmis et al. 1996, 2002; Lazaro et al. 2003).

6 Changes Occurring in Tissue Culture There is a vague point associated with the very first step of infection of a cultured cell with poliovirus. It is known that a tiny minority of physical viral particles (e.g., in the order of 1%) does initiate effective infection. What does it mean? It could be, though seems not very likely, that the overwhelming majority of virions are somehow damaged during preparation of the inoculate. Another possibility is that the majority of virions contain dead (or very sick) genomes as a result of error-prone replication. Intuitively, it seems that this factor, though plausible, could hardly inactivate 99% of progeny RNA. The possibility of erroneously assembled virions should not be completely ruled out. A significant loss of infectivity could occur during the first steps of virus–cell interaction, because not all adsorbed virions could undergo proper uncoating (Crowell and Landau 1983). Finally, it is probable that not all viral RNA molecules properly uncoated are able to start infection, and this may depend on the properties of not only of the virus but also of host cells. All these possibilities are evidently not mutually exclusive, but their

228

V. I. Agol

individual contribution, if any, is completely unknown. Nevertheless, they may contribute to the selection of specific variants of the viral genomes, especially if the multiplicity of infection is low. Wild-type polioviruses are remarkably stable upon in vitro cultivation under relatively constant conditions. This notion stems from a very close similarity, if not identity, of the primary structure of RNA of different lineages of such model strains as, for example, Mahoney, determined in different laboratories, even though these lineages might have experienced a quite long independent history, sometimes even involving plaque cloning, i.e. bottlenecking events (cf., Tolskaya et al. 1987). The situation is markedly changed, however, if either the virus harbors some debilitating mutations or a completely fit variant is cultivated upon suboptimal conditions. In response to selective pressure, the viral genome exhibits a striking plasticity. Such instability has readily been observed upon passages of Sabin strains, a real problem during the production of the oral polio vaccine (OPV). These strains were derived by extensive selection procedures involving plaque cloning (Sabin and Boulger 1973; Minor and Almond 2002). Their fitness was certainly decreased during the selection and therefore the vaccine strains are especially prone to acquiring adaptive fitness-increasing mutations. Even the standard, presumably authentic, Sabin strains used by different producers may differ from each other (Rezapkin et al. 1999). Passages of any of the three Sabin strains resulted in consistent accumulation of specific mutations in a cell type-dependent and temperature-dependent way (Rezapkin et al. 1994, 1995; Taffs et al. 1995). Some of these mutations, particularly in the 5 UTR, represented reversions to the wild-type progenitors and are associated with increased neurovirulence. It should be kept in mind that mutations detectable by sequencing do not fully reflect the genetic make-up of a viral population. The oligonucleotide-based microarray technique is able to reveal more subtle sequence heterogeneity (Cherkasova et al. 2003). A more systematic use of this technique promises to provide deeper insight into the regularities of poliovirus evolution in vitro. Other, even much more severe, alterations of the poliovirus genome may also be readily suppressed or alleviated by mutations accumulated in vitro. Thus, the viral RNA with a specific 5 UTR deletion could not generate the authentic progeny, but the deleterious effect of this deletion could be suppressed by at least three different types of spontaneous second-site mutations, substitutions, deletions, and insertions (Pilipenko et al. 1992a; Gmyl et al. 1993). Such viral RNA incapable of faithful reproduction, but generating infectious progeny with altered genomes, were called by us quasi-infectious. (Gmyl et al. 1993).

Molecular Mechanisms of Poliovirus Variation and Evolution

229

Plasticity of the viral genome manifests itself in a readiness with which resistance to various viral inhibitors can be developed. A famous example is the appearance of mutants with altered sensitivity to millimolar concentrations of guanidine.HCl. The ability of this simple compound to inhibit reproduction of poliovirus (Rightsel et al. 1961) and the ease with which mutants resistant to, or dependent on, this drug could be selected (Melnick et al. 1961; Loddo et al. 1962) have been known for more than 40 years (although they are still not understood in molecular terms, despite the knowledge of the target viral protein, 2C, and the nature of the responsible mutations—see Pincus et al. 1987; Tolskaya et al. 1994). Guanidine-resistant and guanidinedependent variants can readily be obtained from plaque-cloned populations and, remarkably, the altered sensitivity to the drug in some of them is due to two specific neighboring nucleotide substitutions. This is a nice illustration of the validity of calculations demonstrating the feasibility of the occurrence of any combination of two mutations in a relatively small viral population (see Sect. 3.1.1). Likewise, development of resistance to other viral inhibitors could readily be observed (see, for example, Pfeiffer and Kirkegaard 2003). An illuminating recent example of this sort is rapid selection of mutants resistant to the inhibitory action of short interfering RNA (siRNA) due to accumulation of substitutions impairing complementarity between such RNAs and target viral sequences (Gitlin et al. 2005). Poliovirus mutants resistant to the drugs that target cellular (rather than viral) functions important for viral reproduction can also be readily selected. For example, mutants resistant to brefeldin A, known to impair intracellular membrane trafficking, were reported (Crotty et al. 2004). The resistance was mapped to two cooperating mutations in viral proteins 2C and 3A. Important insights into poliovirus variability came from the experiments in which changes in the antigenic epitopes in response to the presence of antibodies were investigated. As already mentioned, such changes develop in the presence of monoclonal antibodies and result in the appearance of so-called escape mutants (Emini et al. 1982; Minor et al. 1983). While these and similar subsequent studies helped characterize the major antigenic sites of poliovirus and demonstrated that these epitopes can in principle be changed without loss of viability (Minor 1990), two negative results are particularly important in the framework of the present discussion: no poliovirus mutants resistant to polyclonal virus-specific antibodies have ever been selected, nor was conversion from one serotype to another serotype observed under the antibody pressure. These observations suggest that the capsid elements responsible for the antigenic properties serve, in addition, some essential function(s). One such function is well known and consists in interaction with the cellular poliovirus receptor (CD155) required for binding of the virus and reorga-

230

V. I. Agol

nization of the structure of its capsid, an essential step in the intracellular release of the viral RNA (Hogle and Racaniello 2002). It was also reported that alterations of the antigenic structure might affect virion assembly (Reynolds et al. 1992). Combinations of these and possible other structural constraints allow poliovirus to exist in three and only three stable serotypes (Wimmer et al. 1993). We are aware of no reports on adaptation of poliovirus to cultured cells lacking the CD155 receptor, although mutants able to use modified CD155 (nonfunctional with the wild-type poliovirus) have been selected (Colston and Racaniello 1995). Impaired interaction with the poor receptor appeared to be compensated by increased lability of the mutant capsid (Wien et al. 1997). Again, this indicates that certain parts of the poliovirus capsid are highly conserved because the virions can be productively uncoated only by a very specialized receptor. However, as shown below, the relevant restriction is perhaps not absolute. A special case of poliovirus variability in tissue culture is its evolution towards establishing a persistent infection (reviewed in Colbère-Garapin et al. 2002). The mechanisms responsible for this phenomenon may differ, but most often persistence is established in cells with an impaired poliovirus receptor (Pavio et al. 2000). Such cells may be selected due to the elimination (as a result of lytic infection) of cells exhibiting unimpaired receptor functions (ColbèreGarapin et al. 1989). The virus is nevertheless reproduced in a proportion of survived cells, and the host cells and the virus may co-evolve towards a kind of equilibrium during serial passages of the infected cultures (Borzakian et al. 1992). Viral evolution consists in accumulation of a limited number of mutations, predominantly in the capsid proteins (Calvez et al. 1993), and these mutations change the efficiency of productive viral interaction with both modified and unmodified receptors (Pavio et al. 2000; Pelletier et al. 2003). Remarkably, capsid structures selected during persistent infection may confer to poliovirus the capacity to infect mice (Couderc et al. 1994). The persistence of poliovirus may also be associated with some intracellular peculiarities caused, for example, by a specific differentiation status of the host cell (Benton et al. 1996). It would be interesting to investigate possible changes in the viral RNA during such long-term persistence. Covalent rearrangements of poliovirus RNA, such as DI-RNA, which can be observed in vitro under certain conditions have already been briefly discussed (Sect. 3.1.3). Summing up observations related to the variability of poliovirus in tissue culture, we see a somewhat paradoxical picture. On the one hand, the viral genome exhibits a high level of conservation, provided the environmental conditions are relatively constant. On the other hand, there is a tremendous

Molecular Mechanisms of Poliovirus Variation and Evolution

231

potential for variability, with the genome retaining its viability in spite of diverse, sometimes dramatic, alterations. This property is reflected in a remarkable plasticity, allowing the virus to adapt to a variety of environmental changes and to survive in the case of serious injuries. Nevertheless, this potential for variability is explicitly exploited primarily when there is no other way to survive. In practice, this means conditions that profoundly decrease the size of viral population, shrinking it to a few or even a single genome, due to either restrictive and damaging conditions or simply mechanistic dilution. Otherwise, the negative, purifying selection will effectively remove innovations from the more conformist RNA species because they are likely to confer a decreased fitness. This makes bottlenecking, if present, a major contributing factor to the viral variability in tissue culture.

7 Evolution in the Organism Factors affecting poliovirus evolution in an organism differ from those operating in tissue culture in several aspects, the most important of which are variability of potential host cells and existence of diverse mechanisms of innate and adaptive immune defense. There is a good and, in a sense, unique opportunity to follow regularities of poliovirus evolution within an organism. Millions of children are given the same live virus in the form of OPV, and changes, if any, occurring during its reproduction can readily be detected in the excreted virus. Although this experimental system has obvious limitations, such as rather short-term virus excretion (usually, a few weeks) and others (see below), it did provide a good deal of important data. Few cases of long-term vaccine virus excretors, such as immunocompromised persons, represent another source of relevant information. Animal models, e.g., mice expressing the human poliovirus receptor, have so far proved to be of more limited significance in this respect. 7.1 Fitness Changes 7.1.1 Elimination of Adverse Mutations As already mentioned, derivation of the Sabin strains had involved intensive selection, including multiple bottlenecking (Sabin and Boulger 1973; Minor and Almond 2002) and not surprisingly had been accompanied by acquisition of several fitness-decreasing properties, such as a smaller size of plaques,

232

V. I. Agol

thermosensitivity of reproduction (ts or so-called rct40– marker) and others. Some of these deficiencies are likely to be general, but others appear to be host cell-specific. For example, vaccine viruses exhibit a lower growth potential in cultured cells of neural origin (La Monica and Racaniello 1989; Agol et al. 1989) apparently caused by a cell-specific translation deficiency (Svitkin et al. 1985, 1988; Haller et al. 1995), which in fact reflects in vivo attenuation. The nonspecific adverse mutations are readily and rapidly eliminated in the vaccine recipients by negative, purifying selection. The excreted vaccinederived viruses usually regain the large-plaque phenotype and the capacity to grow at 40 °C. An often observed increase in their neurovirulence might be caused by this general fitness improvement but also by the cell-specific selective pressure. Thus, loss of specific attenuating mutations is a common feature of the viruses excreted by healthy vaccines (Minor and Almond 2002). This may mean that the intracellular environment (e.g., availability of hostspecific translation factors) in the cells supporting poliovirus replication in the gastrointestinal tract may be similar to that in neurons. 7.1.2 Recombination Less clear are the mechanisms underlying the fact that recipients of trivalent (containing three serotypes) OPV excrete virus populations enriched with intertypic recombinants, which become detectable, and sometimes even preponderant, very soon after the vaccination (Cammack et al. 1988; Cuervo et al. 2001). This fact may suggest that such recombinants have some selective advantage, which theoretically may stem largely from two circumstances. Recombination may be a tool to get rid of vaccine-specific fitness-decreasing mutations, some of which map to different parts of the viral genome in different Sabin strains. In this regard, it is interesting that distribution of recombination crossovers differs among the Sabin strains of different serotypes (Cuervo et al. 2001; E.A. Korotkova et al., in preparation), not alternatively, recombination may help eliminate deleterious mutations introduced by the error-prone replication (Agol et al. 2001). It would be important to know whether this putative advantage of intertypic recombinants is, at least in part, responsible for the preponderance of recombinants among the polioviruses isolated from the vaccine-associated paralytic poliomyelitis (VAPP) cases (Lipskaya et al. 1991; Furione et al. 1993; Georgescu et al. 1997a). It may be assumed that intratypic crossovers should occur at least not less frequently than intertypic ones, but the latter, firstly, are much more readily detectable using the existing techniques and, secondly, may confer more selective advantage.

Molecular Mechanisms of Poliovirus Variation and Evolution

233

Some evidence for the existence of intratypic crossovers will be considered in Sect. 7.5. All the above genetic changes occurred quite rapidly, during the first weeks of reproduction of the Sabin strains in the organism of healthy vaccines. There is no compelling evidence that additional fitness-increasing mutations accumulate during long-term poliovirus persistence in immunodeficient patients, although the reason(s) why such patients may develop explicit paralytic illness only after years of virus persistence (see for example, Kew et al. 1998) remains unknown. Indeed, poliovirus may persist in such subjects for over 20 years without any signs of poliomyelitis (MacLennan et al. 2004). On the other hand, VAPP may develop, though extremely rarely, in recently vaccinated children or their contacts (see, for example, Nkowane et al. 1987). Thus, either the viral mutations primarily responsible for the development of VAPP have yet to be discovered or, more likely, physiological and/or genetic peculiarities of the host organisms are the most important contributing factor. Likewise, we are aware of no evidence that the fitness of wild-type polioviruses may be enhanced during their growth in an individual organism, though some evolution of these viruses under such conditions has been documented (see below). 7.2 Antigenic Changes Antigenic changes in the Sabin viruses excreted by healthy vaccines are known for long but more precise characterization of these changes (Minor et al. 1986) became possible after identification of individual epitopes, primarily by using monoclonal antibodies. It turned out that only certain of the known epitopes are affected, and their changes are markedly serotype-specific. For example, the so-called site 3B located in VP3 capsid protein (and in particular lysine at position 60) is most often changed in excreted Sabin 1 isolates (M. L. Yakovenko et al., submitted). It should be noted that possession of K60 in VP3 by the Sabin strain is unique among type 1 polioviruses. It is not known whether the tendency to substitute this residue is primarily caused by immune pressure or by selection for a better interaction with poliovirus receptor. Since there is a significant overlapping between antigenic epitopes and receptorrecognizing amino acid residues, this uncertainty holds also for several other observed epitope alterations. This issue is worthy of further experimental investigation. Nevertheless, rapid alteration of antigenic epitopes suggests that it may confer some selective advantage. However, no known natural antigenic changes would make poliovirus fully non-neutralizable by antibodies

234

V. I. Agol

contained in human immune sera or by polyclonal antibodies raised by immunization of animals with complete virus, although human sera may exhibit somewhat lesser neutralizing power against derivatives of the Sabin strains as compared to the Sabin strains themselves (Shulman et al. 2000b; Horie et al. 2002; Martin et al. 2002; Blomqvist et al. 2004; M.L. Yakovenko et al., submitted). Antigenic drift, as revealed by reactivity with monoclonal antibodies and selected human sera, has also been observed in wild-type polioviruses during their reproduction in the individual organisms (Huovilainen et al. 1987; Kinnunen et al. 1990; Hovi et al. 2004). In part, these alterations may again be caused by the selection for optimal interaction with the cellular receptors. In any case, the driving force for this evolution is not clear. One may guess that during poliovirus’s long history, it had countless encounters with antibodies targeted to specific epitopes as well as with appropriate receptors, and therefore “had done” everything it could to adapt well to its only host, humans. On the other hand, the spectrum of antibodies caused by the natural herd immunity in the prevaccination era was not necessarily identical to that found in the populations immunized with either Sabin or Salk vaccines (Hovi 1989). Thus, the advent of the vaccination era could possibly affect evolution of wild-type polioviruses. Different hypothetical explanations for the observed antigenic drift may be considered. According to one, this drift is of no adaptive significance but merely reflects the fluctuating character of viral population caused by its quasispecies nature and random selection of diverse variants by bottlenecking events. On the other hand, such bottlenecking may result in selection of less fit viruses, and, in this case, the observed changes may be caused by viral attempts to regain its fitness. Individual host peculiarities may additionally contribute to the antigenic variability (see below). In any case, this type of evolution can hardly select polioviruses with grossly altered antigenic properties or changed receptor specificity, which could have an increased epidemiological significance. However, the situation may dramatically change in the future (Sect. 16). While considering antigenic changes occurring in poliovirus during its reproduction in human gut, it is appropriate to mention trypsin susceptibility of VP1 of certain viral strains (Icenogle et al. 1986; Roivainen and Hovi 1987). Although this is a phenotypic rather than genetic alteration, it may affect genetic variability of the virus because the proteolytic cleavage of the capsid protein may diminish susceptibility of the virus to antibodies as well as affect its growth potential and hence be a selectable factor (Hovi 1989).

Molecular Mechanisms of Poliovirus Variation and Evolution

235

7.3 Host Range Changes Since the gene encoding the human poliovirus receptors is polymorphic, and relevant mutations appear to affect the efficiency of receptor interaction with poliovirus (Saunderson et al. 2004), it could be assumed that this factor may also influence evolution of the viral capsid proteins. In particular, it may contribute to the variability of amino acid residues within and around antigenic sites, as discussed in the preceding section. However, no studies of the relationships between the structure of specific alleles of poliovirus receptor, on the one hand, and viral capsid, on the other, have been reported. Nor are we aware of acquisition of the ability to infect inappropriate (i.e., lacking CD155) human cells even after long-term reproduction in immunodeficient patients. On the other hand, it was reported that mouse-virulent variants (i.e., able to circumvent the absence of a bona fide CD155) might be isolated from mice infected with poliovirus, which normally is devoid of mouse pathogenicity (Couderc et al. 1993; Jia et al. 1999). These variants appeared to exhibit tropism to different mouse neural cells due to modified capsid proteins. The basis of the altered host range is the ability to undergo uncoating upon interaction with mouse proteins, which mimic functions of the human poliovirus receptor. 7.4 Potential to Establish Persistent Infection It has recently been reported that long-term replication of Sabin virus in the gut of an immunodeficient person may be accompanied by acquisition of the capacity to induce persistent infection in cultures cells (Labadie et al. 2004). It is tempting to assume that this phenotypic change reflects the viral ability to persist in the organism. Several mutations were detected in the capsid proteins and there were some alterations in the early steps of the virus–cell interaction, but the mechanism underlying the enhanced potential to trigger persistent infection and the biological relevance of this phenomenon have yet to be elucidated. 7.5 Neutral Mutations While the majority of mutations detected in polioviruses excreted early after OPV vaccination have an adaptive character, mutations accumulating during more prolonged viral growth in immunodeficient persons are predominantly synonymous (Bellmunt et al. 1999; Martin et al. 2000; Gavrilin et al. 2000)

236

V. I. Agol

and hence assumed to be to a significant extent neutral (although, admittedly, effects of these mutations on viral fitness have never been investigated). As could be expected, transitions were observed much more frequently than transversions. Why is growth in the gut accompanied with accumulation of synonymous mutations to a much higher extent compared to the growth in tissue culture? A likely, though hypothetical, explanation consists in postulating the existence of multiple bottlenecking events during viral reproduction in the gastrointestinal tract (Gavrilin et al. 2000; see also Georgescu et al. 1997b). Taking into account a relatively low virus content in the feces, it can be proposed that either few susceptible cells are available at each given moment or only a tiny minority of susceptible cells is effectively infected. If so, virus reproduction in the gut may be to some extent likened to the propagation of a virus by consecutive plaque (or small-population-size) passages. 7.6 Co-existence of Different Lineages The very fact of excretion of intertypic recombinants by OPV vaccinees (see above) is unequivocal evidence for the possibility that more than one poliovirus lineage exists in an organism. However, this is a special case caused by the trivalent nature of this vaccine. Several independent homotypic viral lineages arising from the same parents or resulting from multiple infections may also coexist in the same individual, as evidenced by lineage-specific mutation fixation and/or differences in the patterns of crossover location (Georgescu et al. 1994, 1997b; Kew et al. 1998). The lineages may enter into genetic exchanges with each other. Thus, the character of mutations accumulated in polioviruses isolated during a relatively short period (80 days) of excretion by an immune-compromised patient suggests that the isolates most likely originated from at least two independent infections with type 2 Sabin virus, which underwent, in addition to a crossover with Sabin 1, intratypic recombination within the capsid protein-encoding region (Cherkasova et al. 2005). Co-existence of more than one lineage of wild-type poliovirus in the same organism was also reported (Hovi et al. 2004). 7.7 Rate of Evolution Long-term excretion of OPV derivatives, e.g., by immunocompromised persons, provides a good opportunity to estimate the rate of poliovirus evolution

Molecular Mechanisms of Poliovirus Variation and Evolution

237

in an organism. Such estimations provide insights into the evolutionary mechanisms, on the one hand, and may be used for evaluation of the viral age (i.e., the time elapsed after the divergence of a given virus from its predecessor), on the other. It was already mentioned that a number of adaptive mutations in the vaccine virus accumulate shortly after the onset of viral reproduction in the gastrointestinal tract. These adaptive mutations do not obey the molecular clock hypothesis and should not be taken into account for the estimation of the “intrinsic” evolution rate, which should be based on the time course of fixation of synonymous mutations. Indeed, the majority of subsequently accumulating mutations are synonymous (Bellmunt et al. 1999; Martin et al. 2000; Gavrilin et al. 2000; Cherkasova et al., 2002), in line with the assumption that the adaptive selection is not the major driving force of poliovirus evolution. The synonymous mutations are fixed in a more or less linear manner. However, the rates of evolution of different portions of the same poliovirus genome may differ markedly, and the regions more rapidly evolving in one lineage may turn out to evolve much more slowly in another one. The uneven rate of mutation accumulation in different parts of the viral RNA remains unexplained. It may be caused by, for example, intratypic recombination, especially in the case of concurrently replicating lineages of different ages (Cherkasova et al. 2005). In addition, some synonymous mutations may not be neutral and therefore may be selected for or against. Thus, certain synonymous sites may be less prone for changes because they play a role in maintenance of either the secondary structure of functionally important RNA cis-acting elements or serve to retain a ratio of synonymous codons optimal for translation or post-translational protein folding (Gavrilin et al. 2000). Synonymous substitution may accumulate nonlinearly with time. The most likely explanation of this phenomenon is again co-existence of competing lineages. It seems feasible that separate lineages, even originating from the same parent, may accumulate mutations at different rates. It could be so, if rates of viral reproduction in different niches of the organism are different because of differences in duration of either the single infectious cycle or intervals between the cycles. Fluctuating, and sometimes unusually high, rates of evolution and coexistence of distinct lineages were also observed in wild-type poliovirus during its relatively prolonged multiplication in an individual organism (Hovi et al. 2004). Theoretically, uneven rates of evolution of the entire viral RNA or its parts may be due to the local or general changes in the fidelity of viral RNA polymerase. Thus far, no supporting evidence for such a possibility has been

238

V. I. Agol

reported but, as already mentioned, a single amino acid substitution in the polymerase encoding sequence may change the enzyme fidelity several-fold (Pfeiffer and Kirkegaard 2003; Castro et al. 2005). Despite all these uncertainties, the proportion of changed synonymous sites may be used for the estimation of the approximate age of the virus. Usually, the VP1-coding sequence is used for such estimations. This region is evolved at a rate of around 3 × 10−2 synonymous substitutions per synonymous site per year (Bellmunt et al. 1999; Gavrilin et al. 2000; Martin et al. 2004). However, it should be kept in mind that the rate of mutation fixation in the VP1-coding region may significantly differ from that in regions encoding other capsid proteins (Cherkasova et al. 2005).

8 Evolution in Human Populations Several additional factors influence evolution of poliovirus in human populations as compared to its evolution in individual organisms. The most important of them is that, in the former case, the human-to-human virus transfer usually involves a relatively small, and not necessarily representative, portion of a viral population. Thus, multiple successive bottlenecking events are a general feature of such evolution. Variability of hosts with their individual immune response and perhaps a variable intracellular environment may also be contributing factors. 8.1 Dissemination Factors To evolve in a population, poliovirus should spread among susceptible human subjects. The efficiency of this spread depends on several factors: level and duration of excretion by infected persons, environmental stability of the virus, climate, hygiene conditions, level of population immunity, efficiency of “takes” by the new recipient, etc. (Fine and Carneiro 1999). In general, the vaccine-derived viruses are less prone to transmission compared to the wild-type virus, but the transmission potential of the former is high enough, as judged by its ability to infect vaccinee’s contacts and trigger overt epidemics (Kew et al. 2002, 2004; Yang et al. 2003; Shimizu et al. 2004) or to spread without accompanying paralytic cases (Korotkova et al. 2003) in populations with a low level of immunity. There is evidence even for cryptic spread of vaccine-derived viruses in apparently adequately immunized populations (Cherkasova et al. 2002; 2003).

Molecular Mechanisms of Poliovirus Variation and Evolution

239

8.2 Antigenic Variability In the framework of the present review, it is important to know whether the character of poliovirus spread among humans results in adaptive antigenic changes, which may potentially endow the virus with a higher potential to overcome the herd or vaccination-induced immunity. As far as the circulating OPV-derived viruses are concerned, they exhibit generally a similar, if not identical, spectrum of substitutions in the antigenic epitopes compared to the isolates from vaccinees. As an exceptional case, a deletion of two amino acid residues in antigenic site 1 of VP1 of Sabin 1 virus was observed (Mulders et al. 1999). A decrease in the level of neutralizing activity of human sera (capable of neutralizing Sabin viruses) towards antigenically changed virus is not infrequent (Shulman et al. 2000b; Horie et al. 2002; Martin et al. 2002; Blomqvist et al. 2004; M.L. Yakovenko et al., submitted), but the complete loss of neutralizing power of some of such sera (having a low anti-Sabin titer) was observed only with highly diverged OPV derivatives, e.g., the virus with an estimated age of independent evolution of approximately 10 years (Blomqvist et al. 2004). Interestingly, wild-type polioviruses isolated during epidemics exhibited a significant variability in the antigenic epitopes and sometimes a decreased neutralizability with sera of humans immunized with either OPV or IPV (Huovilainen et al. 1987, 1988; Hovi 1989; Deshpande and Dave 1992; Fiore et al. 1998). However, no clear indications for directed evolution towards selection of non-neutralizable or even poorly neutralizable variants was observed. Therefore, the driving force(s) for the observed variability remains uncertain. As discussed above, it may include immunological pressure, especially against more readily yielding epitopes, optimization of interaction with receptors or even merely chance bottlenecking. 8.3 Pattern of Mutation Accumulation There is an interesting problem that has not yet been adequately explored in spite of the availability of the necessary data. A great number of independent OPV-derived genomes have been partially sequenced and there is also substantial information about the full-genome sequences of such RNAs. Some of the substitutions, e.g., reversions of attenuating mutations in noncoding and coding sequences as well as alterations of antigenic epitopes, are very common among the isolates. But are there any other regularities of the evolution of OPV derivatives? The preponderance of synonymous substitutions in highly evolved lineages (cf., Gavrilin et al. 2004; Blomqvist et al. 2004) and their

240

V. I. Agol

predominantly random character (K. Chumakov, personal communication) suggest that a major quantitative contribution to the mutation accumulation comes from neutral (or nearly neutral) evolution, based primarily on bottlenecking events. However, a more detailed study may perhaps unravel certain elements which are either especially prone to variability or, on the contrary, are unusually conserved. A similar analysis of evolutionary trends of wild-type polioviruses is more difficult in view of absence of knowledge of the parental genomic structure. Nevertheless, related strains differ largely at synonymous sites (Gavrilin et al. 2000; Hovi et al. 2004) and no specific pattern of missense substitutions was reported except for above-mentioned frequent variations of antigenic epitopes and hypervariability of a 5 UTR region preceding the translation initiator codon (Poyry et al. 1992). 8.4 Contribution of Recombination The majority of circulating OPV derivatives exhibiting some divergence in the capsid-coding region from the original Sabin strains turned out to be recombinants. This is true of Sabin viruses belonging to all the three serotypes, and each serotype may recombine with any other Sabin serotype (Cuervo et al. 2001; Cherkasova et al. 2002, 2003), the abundance of such recombinants and distribution of crossovers along the genome being serotype-specific (Cuervo et al. 2001; E.A. Korotkova et al., in preparation). The intertypic recombination between Sabin-derived strains can readily be detected by specially devised oligonucleotide-based assay (Cherkasova et al. 2003; Korotkova et al. 2003). Recombinants of Sabin-derived viruses with wild-type polioviruses were also reported (Georgescu et al. 1995; Guillot et al. 2000; Liu et al. 2000, 2003; Kew et al. 2002; Yang et al. 2003; Korotkova et al. 2003). Such interpretation of the sequencing results is fully justified only in the cases when the wild-type partner or its close relative is known. Otherwise, due to a close similarity of the sequences encoding nonstructural proteins, and especially viral RNA polymerase 3Dpol , in polioviruses and other enteroviruses belonging to the so-called C cluster (Gromeier et al. 1999; Brown et al. 2003), it is impossible to state whether the recombination involved a wild-type poliovirus or another representative of this enterovirus cluster. Recombination between wild-type polioviruses was also reported (Dahourou et al. 2002). Again, one may be sure of poliovirus origin of only the capsid-coding region of such recombinants.

Molecular Mechanisms of Poliovirus Variation and Evolution

241

The key question of whether the recombination in the above cases confers some selective advantage, and in particular, contributes to the enhanced transmissibility and pathogenicity, is still open. Obviously, elimination, by recombination, of attenuating mutations residing in the relevant genomic region of Sabin viruses should have adaptive significance. However, such adverse mutations are likely to be rapidly removed also by negative (purifying) selection. Biological relevance, if any, of the acquisition of nonstructural proteins from nonpolio enteroviruses is unknown. It cannot be ruled out that recombination by itself confers little, if any, advantage, being merely a very common by-product of concurrent reproduction of different enteroviruses in the same gastrointestinal tract. The occurrence of overt diseases after a period of circulation of the recombinant also should not necessarily be interpreted as the acquisition of an enhanced pathogenicity. It may merely be due to the necessity to infect, by a virus with a relatively low pathogenicity, a significant number of susceptible persons to cause paralytic symptoms. It should be kept in mind that even wild-type poliovirus, though usually considered as highly pathogenic on the basis of experiments with primates and cultured cells, exhibits relatively very mild pathogenic potential under natural conditions, inflicting the paralytic disease in less than 1% of infected nonimmune persons. 8.5 Rate of Evolution The first estimate of the rate of natural poliovirus evolution, based on very crude data suggesting that approximately 1% of nucleotides in poliovirus RNA are changed over 1 year of circulation (Nottay et al. 1981) proved to be correct. For practical reasons (e.g., the availability of sequences), mutations in the VP1-coding region are often used for calculations of the rate, and, as discussed above, the proportion of changed synonymous sites (or, in a simpler version, the 3rd codon positions) is taken into the account. The values on the order of 3 × 10−2 per year appear to reflect the evolution rate of persisting and circulating OPV derivatives as well as of wild-type polioviruses (Gavrilin et al., 2000). It is obvious that the rate of mutation accumulation should depend not only on intrinsic factors (discussed above) but also on such extrinsic parameters as the number of bottlenecking effects (human-to-human transfer), the amount of infectious virus transferred during such events, the number of contact people receiving the virus from its excretors, etc. In turn, these factors are influenced by the immune status of human population, socioeconomic conditions, etc. Thus, it would not be surprising that the rate of mutation

242

V. I. Agol

fixation during the evolutionary history of poliovirus has not been and is not now constant. The local fluctuations of this rate also could be expected. Nevertheless, the similarity of the rates of fixation of synonymous mutations in poliovirus lineages evolving in an individual organism and in human populations (Gavrilin et al. 2000) is striking. It may suggest, somewhat unexpectedly, that the frequencies of bottlenecking events in these different situations do not differ very much from one another. 8.6 Geographical Constraints Wild-type polioviruses exhibit such a clear tendency for geographical clustering of related viruses (Rico-Hesse et al. 1987; Kew et al. 1995) that, for such genomic clusters, the term “geotype” (geographical genotypes; Lipskaya et al., 1995) was proposed. The knowledge of specific signatures of the geotypes is helpful for the identification of the origin of viruses causing particular poliomyelitis outbreaks, such as in Finland in 1984–1985 (Poyry et al. 1990), Israel in 1987–1988 (Shulman et al. 2000a) or Albania and neighboring countries in 1996 (Fiore et al. 1996). Apart from its epidemiological significance, the genome variability within a geotype provides information about the regularities of wild-type poliovirus evolution. The outbreaks may be caused by a single transfer (bottleneck) of an endemic virus to a novel susceptible population, thus generating a monophyletic group of viral genomes. Endemic geotypes with no known history also constitute sets of genomes amenable for comparative studies. Attempts to consider the geotypes from a general evolutionary perspective have been undertaken (cf., Hovi 1989; Gavrilin et al. 2000). It seems, however, that the useful information has not yet been fully extracted from the existing data.

9 Relationships Between Poliovirus and Other Picornaviruses Insights into the evolution of poliovirus can be provided by understanding its relationships with picornaviruses, in general, and other enteroviruses, in particular. There are two major sources of the relevant information: studies on mutual compatibility of parts of picornavirus genomes through engineering of genetic chimeras between poliovirus and other picornaviruses, on the one hand, and studies on natural relationships between these viruses, on the other.

Molecular Mechanisms of Poliovirus Variation and Evolution

243

9.1 Engineered Viable Chimeric Polioviruses The first interspecies chimeric poliovirus was engineered by Semler et al. (1986) who replaced an IRES-containing portion of the poliovirus 5 UTR by a homologous segment of the coxsackievirus B3 RNA. The chimera proved to be infectious, though exhibiting a ts phenotype. In contrast, the replacement of the nearly entire poliovirus 5 UTR by its coxsackievirus counterpart produced a virus phenotypically very similar to that of the wild type (Johnson and Semler 1988). Subsequent studies demonstrated that various portions of the poliovirus RNA could be replaced by the relevant segments of the genome of other picornaviruses without loss of viability but, quite often, with a general or cell-specific decrease in fitness. Thus, both replicative and translational cis-elements of poliovirus 5 UTR (in their entirety or in part) could be exchanged for sequences derived not only from other enteroviruses but also from rhinoviruses (Xiang et al. 1995b; Gromeier et al. 1996; Rieder et al. 2003), cardioviruses (Alexander et al. 1994; Rohll et al. 1994), and even from nonpicornavirus, hepatitis C virus (Lu and Wimmer 1996). Other rhinovirus replicative cis-acting elements, such as cre (Yin et al. 2003) and oriR (Rohll et al. 1995) also proved to be functional in the context of the poliovirus genome. Replacements of certain poliovirus proteins, e.g., 2A (Lu et al. 1995) and VPg (Paul et al. 2003; Cheney et al. 2003), by their counterparts from other picornaviruses may also be possible without loss of viral viability. Likewise, parts of poliovirus RNA can be used to substitute appropriate portions of the genome of other picornaviruses to generate viable viruses (Zell et al. 1995; Todd et al. 1997b; van Kuppeveld et al. 1997; Chapman et al. 2000). However, some other poliovirus proteins, such as 3C and 3CD, exhibit a higher specificity and could not be so readily replaced by homologous proteins of other picornaviruses (Dewalt et al. 1989; Bell et al. 1999; Cornell et al. 2004). Thus, there is a significant, though limited, compatibility between portions of poliovirus genome and RNAs of other picornaviruses. A very important question is whether this potential is actually realized under natural conditions. 9.2 Natural Recombination Between Poliovirus and Other Picornaviruses The answer to the above question is most likely positive, even though evidence for it is indirect. To document natural recombination between poliovirus and other picornaviruses, one should demonstrate a close similarity between RNA portions of a putative recombinant and of a poliovirus, on the one hand, and a known nonpoliovirus partner, on the other. However, the only known

244

V. I. Agol

reliable signature of the bona fide poliovirus is its capsid, whereas the untranslated and nonstructural portions of the poliovirus genome cannot safely be distinguished from those of representatives of cluster C enteroviruses, especially those of CAV 11, 17, 20, and, in the 3CD-encoding region, also of CAV 13, 21, and 24 (Brown et al. 2003). Plausible candidates to polio–nonpolio recombinants are those OPV-derived viruses, which were isolated during recent poliomyelitis outbreaks in inadequately immunized territories (Kew et al. 2002, 2004; Rousset et al. 2003; Shimizu et al. 2004). These viruses exhibit a reliable relatedness to their OPV progenitor in the capsid-encoding part of the genome, whereas the origin of the noncoding part is uncertain. The major argument that they represent polio–nonpolio rather than OPV–wild-type polio recombinant is the absence of detectable circulation of wild-type poliovirus on the territories where these viruses appeared to originate and evolve. In the cases when even this, admittedly indirect, argument cannot be invoked, it would be safer to designate such viruses as recombinants between an OPV derivative and unidentified member of the cluster C enterovirus (keeping in mind that this cluster includes polioviruses as well). The similarity between nonstructural proteins of poliovirus and cluster C CAVs makes plausible the hypothesis that the relevant genes form, and evolve as, a common pool. We are aware of no documented natural recombinants between polioviruses and enteroviruses not belonging to cluster C, let alone other picornaviruses, but the above-mentioned data on engineered chimeras allow one to guess that such intergeneric recombination is in principle possible.

10 A Few Remarks on Poliovirus Macroevolution Sharing several important features, picornaviruses appear to form a monophyletic group. What is the place of polioviruses in this group? As already mentioned, the three poliovirus serotypes constitute a subset within the cluster C of human enteroviruses. Separations of polioviruses from other members of this group were based on the peculiarities of the capsid structure related to its adaptation to the specific receptor, CD155. No marked genomic rearrangements appeared to accompany this separation. It was speculated that polioviruses may not constitute a monophyletic group within this cluster but rather they had evolved more than once from the common trunk (Gromeier et al. 1999). If so, it would be interesting to know whether the separation into three distinct serotypes was a result of immunoselection or rather the origin from different, though related, predecessors.

Molecular Mechanisms of Poliovirus Variation and Evolution

245

At the preceding evolutionary nodes, separation of cluster C from other clusters of human enteroviruses (e.g., D, A and B) should take place. A more substantial divergence of RNA domains occurred at this step. Thus, 3 UTR (oriR) of coxsackie B viruses is some 30 nt longer and has an additional conserved hairpin compared to its poliovirus counterpart (Pilipenko et al. 1992b). At an even earlier bifurcation, when enteroviruses were separated from rhinoviruses, the former group of viruses appeared to undergo, among other changes, an extended duplication in the 5 UTR, which resulted in generation of a long spacer between the IRES and polyprotein ORF (Pilipenko et al. 1990). A more detailed discussion of the possible evolution of picornaviruses was presented by Gromeier et al. (1999).

11 Towards the Eradication of Poliovirus Although the previous evolutionary history of poliovirus is still a matter of speculation, we are likely witnessing its last step. In 1988, the World Health Organization started a program aimed at the global eradication of poliomyelitis with the goal of stopping circulation of wild-type polioviruses (Dowdle and Cochi 2002; Kew and Pallansch 2002). This program, based on the immunization with OPV as well as IPV (inactivated polio vaccine), was a tremendous success, even though it has not achieved its goal by the proposed deadline in 2000. Indeed, the incidence of paralytic poliomyelitis has decreased by two orders of magnitude and wild-type poliovirus remains endemic only in a few countries in Africa and Asia (Minor 2004). The detailed discussion of the achievements and setbacks of the eradication program is beyond the scope of this review, but one of its aspects is highly relevant. Let us assume that the wild-type poliovirus circulation is stopped. Should vaccination, and in particular OPV use be discontinued? This vaccination is very costly, and in the absence of the infectious agent may seem senseless, if not harmful because of the known ability of OPV to cause, admittedly very rarely, vaccine-associated disease (Nkowane et al. 1987). The situation, however, is not so straightforward (Dowdle et al. 2003). In the absence of vaccination, the proportion of nonimmune persons (primarily, children) will rapidly grow, generating a rich soil for the second advent of the virus (Korotkova et al. 2003). This re-emerging virus may come from a variety of sources, such as laboratories and industry or long-term virus excretors, i.e., immunodeficient persons (Dowdle et al. 2003). Vaccine-derived polioviruses are also able to persist and evolve for years, even in apparently adequately immunized populations (Shulman et al. 2000b; Cherkasova et al. 2002, 2003; Blomqvist et al. 2004).

246

V. I. Agol

Still another possibility should not be discounted. In the absence of immune defense, a repetition of the ancient evolutionary event may occur, the adaptation of a representative of cluster C enterovirus to the human poliovirus receptor, CD155 (Rieder et al. 2001), thereby creating a neopoliovirus. In view of these and some other concerns, it seems reasonable to continue OPV or IPV usage even after the eradication of wild-type circulation, until alternative means for poliovirus prevention are available, e.g. new types of polio vaccines are developed (Korotkova et al. 2003; Fine et al. 2004; Minor 2004).

12 Concluding Remarks A vast amount of data has accumulated regarding the variability and evolution of poliovirus. If the mechanisms underlying variability of the virus are more or less clear, at least in general terms, our comprehension of the regularities of poliovirus evolution is much more vague. It seems that immune pressure, a major factor in the evolution of many other viruses, is hardly currently playing a decisive role in poliovirus evolution. One cannot even be sure that the existence of three poliovirus serotypes was due to immunoselection. Nor is there evidence that adaptation to new receptors or to novel ecological niches is occurring in the contemporary polioviruses, at least to such an extent as to generate phenotypically distinct lineages. The current knowledge appears to fit best the model according to which the major contributions to natural evolution of poliovirus come from two sources: genetic drift caused by diverse bottlenecking events and negative (purifying) selection. There is a high probability that we shall see the end of poliovirus evolution due to the successful implementation the eradication program. Nevertheless, for those of us who devoted considerable time and effort to study this beloved agent, there may be some condolence in the fact that a significant part of the poliovirus genome (that encoding nonstructural proteins and untranslated cis-elements) will continue to live, as a component of cluster C enteroviruses, even after the virus’s death. On the other hand, enteroviruses themselves, as any other biological species, are expected to have a limited lifespan and eventually will either disappear completely or be transformed in something distinctly different. Acknowledgements The original works from the author’s lab are currently supported by grants from INTAS, the Russian Foundation for Basic Research, the Ministry of Industry, Science and Technology as well as the Program for Support of Scientific Schools.

Molecular Mechanisms of Poliovirus Variation and Evolution

247

References Agol VI (1997) Recombination and other genomic rearrangements in picornaviruses. Semin Virol 8:1–9 Agol VI (2002a) Poliovirus genetics: an overview. In: Semler BL, Wimmer E (eds) Molecular biology of picornaviruses. ASM Press, Washington, pp 269–284 Agol VI (2002b) Poliovirus genome: an overview. In: Semler BL, Wimmer E (eds) Molecular biology of picornaviruses. ASM Press, Washington, pp 127–148 Agol VI, Drozdov SG, Ivannikova TA, Kolesnikova MS, Korolev MB, Tolskaya EA (1989) Restricted growth of attenuated poliovirus strains in cultured cells of a human neuroblastoma. J Virol 63:4034–4038 Agol VI, Paul AV, Wimmer E (1999) Paradoxes of the replication of picornaviral genomes. Virus Res 62:129–147 Agol VI, Belov GA, Cherkasova EA, Gavrilin GV, Kolesnikova MS, Romanova LI, Tolskaya EA (2001) Some problems of molecular biology of poliovirus infection relevant to pathogenesis, viral spread and evolution. Dev Biol (Basel) 105:43–50 Alexander L, Lu HH, Wimmer E (1994) Polioviruses containing picornavirus type 1 and/or type 2 internal ribosomal entry site elements: genetic hybrids and the expression of a foreign gene. Proc Nat Acad Sci U S A 91:1406–1410 Andino R, Rieckhof GE, Baltimore D (1990) A functional ribonucleoprotein complex forms around the 5 end of poliovirus RNA. Cell 63:369–380 Andino R, Rieckhof GE, Achacoso PL, Baltimore D (1993) Poliovirus RNA synthesis utilizes an RNP complex formed around the 5 -end of viral RNA. EMBO J 12:3587– 3598 Arnold JJ, Cameron CE (1999) Poliovirus RNA-dependent RNA polymerase (Dpol) is sufficient for template switching in vitro. J Biol Chem 274:2706–2716 Arnold JJ, Cameron CE (2004) Poliovirus RNA-dependent RNA polymerase (Dpol ): pre-steady-state kinetic analysis of ribonucleotide incorporation in the presence of Mg2+ . Biochemistry 43:5126–5137 Bass BL (2002) RNA editing by adenosine deaminases that act on RNA. Annu Rev Biochem 71:817–846 Bell YC, Semler BL, Ehrenfeld E (1999) Requirements for RNA replication of a poliovirus replicon by coxsackievirus B3 RNA polymerase. J Virol 73:9413–9421 Bellmunt A, May G, Zell R, Pring-Akerblom P, Verhagen W, Heim A (1999) Evolution of poliovirus type 1 during 5.5 years of prolonged enteral replication in an immunodeficient patient. Virology 265:178–184 Benton PA, Barrett DJ, Matts RL, Lloyd RE (1996) The outcome of poliovirus infections in K562 cells is cytolytic rather than persistent after hemin-induced differentiation. J Virol 70:5525–5532 Bishop KN, Holmes RK, Sheehy AM, Malim MH (2004) APOBEC-mediated editing of viral RNA. Science 305:645 Blomqvist S, Savolainen C, Laine P, Hirttio P, Lamminsalo E, Penttila E, Joks S, Roivainen M, Hovi T (2004) Characterization of a highly evolved vaccine-derived poliovirus type 3 isolated from sewage in Estonia. J Virol 78:4876–4883 Borzakian S, Couderc T, Barbier Y, Attal G, Pelletier I, Colbère-Garapin F (1992) Persistent poliovirus infection: establishment and maintenance involve distinct mechanisms. Virology 186:398–408

248

V. I. Agol

Brown B, Oberste MS, Maher K, Pallansch MA (2003) Complete genomic sequencing shows that polioviruses and members of human enterovirus species C are closely related in the noncapsid coding region. J Virol 77:8973–8984 Brown DM, Kauder SE, Cornell CT, Jang GM, Racaniello VR, Semler BL (2004) Celldependent role for the poliovirus 3 noncoding region in positive-strand RNA synthesis. J Virol 78:1344–1351 Calvez V, Pelletier I, Borzakian S, Colbère-Garapin F (1993) Identification of a region of the poliovirus genome involved in persistent infection of HEp-2 cells. J Virol 67:4432–4435 Cammack N, Phillips A, Dunn G, Patel V, Minor PD (1988) Intertypic genomic rearrangements of poliovirus strains in vaccinees. Virology 167:507–514 Castro C, Arnold JJ, Cameron CE (2005) Incorporation fidelity of the viral RNAdependent RNA polymerase: a kinetic, thermodynamic and structural perspective. Virus Res 107:141–149 Chapman NM, Ragland A, Leser JS, Hofling K, Willian S, Semler BL, Tracy S (2000) A group B coxsackievirus/poliovirus 5 nontranslated region chimera can act as an attenuated vaccine strain in mice. J Virol 74:4047–4056 Cheney IW, Naim S, Shim JH, Reinhardt M, Pai B, Wu JZ, Hong Z, Zhong W (2003) Viability of poliovirus/rhinovirus VPg chimeric viruses and identification of an amino acid residue in the VPg gene critical for viral RNA replication. J Virol 77:7434–7443 Cherkasova EA, Korotkova EA, Yakovenko ML, Ivanova OE, Eremeeva TP, Chumakov KM, Agol VI (2002) Long-term circulation of vaccine-derived poliovirus that causes paralytic disease. J Virol 76:6791–6799 Cherkasova EA, Laassri M, Chizhikov V, Korotkova EA, Dragunsky E, Agol VI, Chumakov KM (2003) Microarray analysis of evolution of RNA viruses: evidence of circulation of virulent highly divergent vaccine-derived polioviruses. Proc Natl Acad Sci U S A 100:9398–9403 Cherkasova EA, Yakovenko ML, Rezapkin GV, Korotkova EA, Ivanova OE, Eremeeva TP, Krasnoproshina LI, Romanenkova NI, Rozaeva NR, Sirota L, Agol VI, Chumakov KM (2005) Spread of vaccine-derived poliovirus from a paralytic case in immunodeficient child: an insight into natural evolution of oral polio vaccine. J Virol 79:1062–1070 Chetverina HV, Demidenko AA, Ugarov VI, Chetverin AB (1999) Spontaneous rearrangements in RNA sequences. FEBS Lett 450:89–94 Chetverin AB, Kopein DS, Chetverina HV, Demidenko AA, Ugarov VI (2005) Viral RNA-directed RNA polymerase use diverse mechanisms to promote recombination between RNA molecules. J Biol Chem 280:8748–8755 Colbère-Garapin F, Christodoulou C, Crainic R, Peletier I (1989) Persistent poliovirus infection in human neuroblastoma cells. Proc Natl Acad Sci U S A 86:7590–7594 Colbère-Garapin F, Peletier I, Ouzilou L (2002) In: Semler BL, Wimmer E (eds) Molecular biology of picornaviruses. ASM Press, Washington, pp 437–448 Colston EM, Racaniello VR (1995) Poliovirus variants selected on mutant receptorexpressing cells identify capsid residues that expand receptor recognition. J Virol 69:4823–4829

Molecular Mechanisms of Poliovirus Variation and Evolution

249

Cornell CT, Perera R, Brunner JE, Semler BL (2004) Strand-specific RNA synthesis determinants in the RNA-dependent RNA polymerase of poliovirus. J Virol 78:4397–4407 Couderc T, Hogle J, Le Blay H, Horaud F, Blondel B (1993) Molecular characterization of mouse-virulent poliovirus type 1 Mahoney mutants: involvement of residues of polypeptides VP1 and VP2 located on the inner surface of the capsid protein shell. J Virol 67:3808–3817 Couderc T, Guedo N, Calvez V, Pelletier I, Hogle J, Colbère-Garapin F, Blondel B (1994) Substitutions in the capsids of poliovirus mutants selected in human neuroblastoma cells confer on the Mahoney type 1 strain a phenotype neurovirulent in mice. J Virol 68:8386–8391 Crotty S, Andino R (2002) Implications of high RNA virus mutation rates: lethal mutagenesis and the antiviral drug ribavirin. Microb Infect 4:1301–1307 Crotty S, Cameron CE, Andino R (2001) RNA virus error catastrophe: direct molecular test by using ribavirin. Proc Natl Acad Sci U S A 98:6895–6990 Crotty S, Saleh MC, Gitlin L, Beske O, Andino R (2004) The poliovirus replication machinery can escape inhibition by an antiviral drug that targets a host cell protein. J Virol 78:3378–3386 Crowell RL, Landau BJ (1983) Receptors in the initiation of picornavirus infection. In: Fraenkel-Conrat H, Wagner RR (eds) Comprehensive virology. Vol. 1 Plenum Press, New York, pp 1–42 Cuconati A, Xiang W, Lasher F, Pfister T, Wimmer E (1998) A protein linkage map of the P2 nonstructural proteins of poliovirus. J Virol 72:1297–1307 Cuervo NS, Guillot S, Romanenkova N, Cochi SL, Combiescu M, Aubert-Combiescu A, Seghier M, Caro V, Crainic R, Delpeyroux F (2001) Genomic features of intertypic recombinant Sabin poliovirus strains excreted by primary vaccinees. J Virol 75:5740–5751 Dahourou G, Guillot S, Le Gall O, Crainic R (2002) Genetic recombination in wild-type poliovirus. J Gen Virol 83:3103–3110 De la Torre JC, Wimmer E, Holland JJ (1990) Very high frequency of reversion to guanidine resistance in clonal pools of guanidine-dependent type 1 poliovirus. J Virol 64:664–671 Deshpande JM, Dave KH (1992) Antibody response of children immunized with poliovaccines: an evaluation using two strains of poliovirus type 1. Indian J Med Res 95:216–220 Dewalt PG, Lawson MA, Colonno RJ, Semler BL (1989) Chimeric picornavirus polyproteins demonstrate a common C proteinase substrate specificity. J Virol 63:3444– 3452 Dobrikova E, Florez P, Bradrick S, Gromeier M (2003) Activity of a type 1 picornavirus internal ribosomal entry site is determined by sequences within the 3 nontranslated region. Proc Natl Acad Sci U S A. 100:15125–15130 Domingo E, Escarmís C, Menéndez-Arias L, Holland JJ (1999) Viral quasi-species and fitness variations. In: Domingo E, Webster, RG, Holland JJ (eds) Origin and evolution of viruses. Academic Press, San Diego, pp 141–161

250

V. I. Agol

Domingo E, Baranowski E, Escarmis C, Sobrino F, Holland JJ (2002a) Error frequencies of picornavirus RNA polymerases: evolutionary implications for virus populations. In: Semler BL, Wimmer E (eds) Molecular biology of picornaviruses. ASM Press, Washington, pp 285–298 Domingo E, Ruiz-Jarabo CM, Sierra S, Arias A, Pariente N, Baranowski E, Escarmis C (2002b) Emergence and selection of RNA virus variants: memory and extinction. Virus Res 82:39–44 Dowdle WR, Cochi SL (2002) Global eradication of poliovirus: history and rationale. In: Semler BL, Wimmer E (eds) Molecular biology of picornaviruses. ASM Press, Washington, pp 473–480 Dowdle WR, De Gourville E, Kew OM, Pallansch MA, Wood DJ (2003) Polio eradication: the OPV paradox. Rev Med Virol 13:277–291 Drake JW, Holland JJ (1999) Mutation rates among RNA viruses. Proc Natl Acad Sci U S A 96:13910–1393 Ehrenfeld E, Teterina NL (2002) Initiation of translation of picornavirus RNAs: structure and function of the internal ribosome entry site. In: Semler BL, Wimmer E (eds) Molecular biology of picornaviruses. ASM Press, Washington, DC, pp 159169 Emini EA, Jameson BA, Lewis AJ, Larsen GR, Wimmer E (1982) Poliovirus neutralization epitopes: analysis and localization with neutralizing monoclonal antibodies. J Virol 43:997–1005 Escarmis C, Davila M, Charpentier N, Bracho A, Moya A, Domingo E (1996) Genetic lesions associated with Muller’s ratchet in an RNA virus. J Mol Biol 264:255–267 Escarmis C, Gomez-Mariano G, Davila M, Lazaro E, Domingo E (2002) Resistance to extinction of low fitness virus subjected to plaque-to-plaque transfers: diversification by mutation clustering. J Mol Biol 315:647–661 Evans DM, Dunn G, Minor PD, Schild GC, Cann AJ, Stanway G, Almond JW, Currey K, Maizel JV Jr (1985) Increased neurovirulence associated with a single nucleotide change in a noncoding region of the Sabin type 3 poliovaccine genome. Nature 314:548–550 Fine PE, Carneiro IA (1999) Transmissibility and persistence of oral polio vaccine viruses: implications for the global poliomyelitis eradication initiative. Am J Epidemiol 150:1001–1021 Fine PE, Oblapenko G, Sutter RW (2004) Polio control after certification: major issues outstanding. Bull WHO 82:47–52 Fiore L, Genovese D, Diamanti E, Catone S, Ridolfi B, Ibrahimi B, Konomi R, van der Avoort HG, Hovi T, Crainic R, Simeoni P, Amato C (1998) Antigenic and molecular characterization of wild type 1 poliovirus causing outbreaks of poliomyelitis in Albania and neighboring countries in 1996. J Clin Microbiol 36:1912–1918 Furione M, Guillot S, Otelea D, Balanant J, Candrea A, Crainic R (1993) Polioviruses with natural recombinant genomes isolated from vaccine-associated paralytic poliomyelitis. Virology 196:199–208 Gavrilin GV, Cherkasova EA, Lipskaya GY, Kew OM, Agol VI (2000) Evolution of circulating wild poliovirus and of vaccine-derived poliovirus in an immunodeficient patient: a unifying model. J Virol 74:7381–7390

Molecular Mechanisms of Poliovirus Variation and Evolution

251

Georgescu MM, Delpeyroux F, Tardy-Panit M, Balanant J, Combiescu M, Combiescu AA, Guillot S, Crainic R (1994) High diversity of poliovirus strains isolated from the central nervous system from patients with vaccine-associated paralytic poliomyelitis. J Virol 68:8089–8101 Georgescu MM, Delpeyroux F, Crainic R (1995) Tripartite genome organization of a natural type 2 vaccine/nonvaccine recombinant poliovirus. J Gen Virol 76:2343– 2348 Georgescu MM, Balanant J, Macadam A, Otelea D, Combiescu M, Combiescu AA, Crainic R, Delpeyroux F (1997a) Evolution of the Sabin type 1 poliovirus in humans: characterization of strains isolated from patients with vaccine-associated paralytic poliomyelitis. J Virol 71:7758–7768 Georgescu MM, Balant M, Ozden S, Crainic R (1997b) Random selection—a model for poliovirus infection of the central nervous system. J Gen Virol 78:1819–1828 Gitlin L, Stone JK, Andino R (2005) Poliovirus escape from RNA interference: short interfering RNA-target recognition and implications for therapeutic approaches. J Virol 79:1027–1035 Gmyl AP, Pilipenko EV, Maslova SV, Belov GA, Agol VI (1993) Functional and genetic plasticities of the poliovirus genome: quasi-infectious RNAs modified in the 5 untranslated region yield a variety of pseudorevertants. J Virol 67:6309–6316 Gmyl AP, Belousov EV, Maslova SV, Khitrina EV, Agol VI (1999) Nonreplicative RNA recombination in poliovirus. J Virol 73:8958–8965 Gmyl AP, Korshenko SA, Belousov EV, Khitrina EV, Chetverin AB, Agol VI (2003) Nonreplicative homologous RNA recombination: promiscuous joining of RNA pieces? RNA 9:1221–1231 Goodfellow I, Chaudhry Y, Richardson A, Meredith J, Almond JW, Barclay W, Evans DJ (2000) Identification of a cis-acting replication element within the poliovirus coding region. J Virol 74:4590–4600 Goodfellow IG, Kerrigan D, Evans DJ (2003a) Structure and function analysis of the poliovirus cis-acting replication element (CRE) RNA 9:124–137 Goodfellow IG, Polacek C, Andino R, Evans DJ (2003b) The poliovirus C cis-acting replication element-mediated uridylylation of VPg is not required for synthesis of negative-sense genomes. J Gen Virol 84:2359–2363 Gromeier M, Nomoto A (2002) Determinants of poliovirus pathogenesis. In: Semler BL, Wimmer E (eds) Molecular biology of picornaviruses. ASM Press, Washington, pp 367–379 Gromeier M, Alexander L, Wimmer E (1996) Internal ribosomal entry site substitution eliminates neurovirulence in inter-generic poliovirus recombinants. Proc Natl Acad Sci U S A 93:2370–2375 Gromeier M, Wimmer E, Gorbalelenya AE (1999) Genetics, pathogenesis and evolution of picornaviruses, In: Domingo E, Webster RG, Holland JJ (eds) Origin and evolution of Viruses. Academic Press, San Diego, pp 287–343 Guillot S, Caro V, Cuervo N, Korotkova E, Combiescu M, Persu A, Aubert-Combiescu A, Delpeyroux F, Crainic R (2000) Natural genetic exchanges between vaccine and wild poliovirus strains in humans. J Virol 74:8434–8443 Haller AA, Stewart SR, Semler BL (1996) Attenuation stem-loop lesions in the 5 noncoding region of poliovirus RNA: neuronal cell-specific translation defects. J Virol 70:1467–1474

252

V. I. Agol

Hogle JM, Racaniello VR (2002) Poliovirus receptors and cell entry. In: Semler BL, Wimmer E (eds) Molecular biology of picornaviruses. ASM Press, Washington, pp 71–83 Horie H, Yoshida H, Matsuura K, Miyazawa M, Wakabayashi K, Nomoto A, Hashizume S (2002) Isolation of vaccine-derived type 1 polioviruses displaying similar properties to virulent wild strain Mahoney from sewage in Japan. J Med Virol 68:445–451 Hovi T (1989) The outbreak of poliomyelitis in Finland in 1984–1985: significance of antigenic variation of type 3 polioviruses and site specificity of antibody responses in antipolio immunizations. Adv Virus Res 37:243–275 Hovi T, Lindholm N, Savolainen C, Stenvik M, Burns C (2004) Evolution of wild-type 1 poliovirus in two healthy siblings excreting the virus over a period of 6 months. J Gen Virol 85:369–377 Huovilainen A, Hovi T, Kinnunen L, Takkinen K, Ferguson M, Minor P (1987) Evolution of poliovirus during an outbreak: sequential type 3 poliovirus isolates from several persons show shifts of neutralization determinants. J Gen Virol 68:1373–1378 Huovilainen A, Kinnunen L, Ferguson M, Hovi T (1988) Antigenic variation among 173 strains of type 3 poliovirus isolated in Finland during the 1984 to 1985 outbreak. J Gen Virol 69:1941–1848 Icenogle JP, Minor PD, Ferguson M, Hogle JM (1986) Modulation of humoral response to a 12-amino-acid site on the poliovirus virion. J Virol 60:297–301 Iizuka N, Kohara M, Hagino-Yamagishi K, Abe S, Komatsu T, Tago K, Arita M, Nomoto A (1989) Construction of less neurovirulent polioviruses by introducing deletions into the 5 noncoding sequence of the genome. J Virol 63:5354–5363 Jarvis TC, Kirkegaard K (1992) Poliovirus RNA recombination: mechanistic studies in the absence of selection. EMBO J 11:3135–3145 Jia Q, Ohka S, Iwasaki K, Tohyama K, Nomoto A (1999) Isolation and molecular characterization of a poliovirus type 1 mutant that replicates in the spinal cords of mice. J Virol 73:6041–6047 Johnson VH, Semler BL (1988) Defined recombinants of poliovirus and coxsackievirus: sequence-specific deletions and functional substitutions in the 5 -noncoding regions of viral RNAs. Virology 162:47–57 Kew OM, Pallansch MA (2002) The mechanism of poliovirus eradication. In: Semler BL, Wimmer E (eds) Molecular biology of picornaviruses. ASM Press, Washington, pp 481–491 Kew OM, Mulders MN, Lipskaya GY, da Silva EE, Pallansch MA (1995) Molecular epidemiology of polioviruses. Semin Virol 6:401–414 Kew OM, Sutter RW, Nottay BK, McDonough MJ, Prevots DR, Quick L, Pallansch MA (1998) Prolonged replication of a type 1 vaccine-derived poliovirus in an immunodeficient patient. J Clin Microbiol 36:2893–2899 Kew O, Morris-Glasgow V, Landaverde M, Burns C, Shaw J, Garib Z, Andre J, Blackman E, Freeman CJ, Jorba J, Sutter R, Tambini G, Venczel L, Pedreira C, Laender F, Shimizu H, Yoneyama T, Miyamura T, van der Avoort H, Oberste MS, Kilpatrick D, Cochi S, Pallansch M, de Quadros C (2002) Outbreak of poliomyelitis in Hispaniola associated with circulating type 1 vaccine-derived poliovirus. Science 296:356–359

Molecular Mechanisms of Poliovirus Variation and Evolution

253

Kew OM, Wright PF, Agol VI, Delpeyroux F, Shimizu H, Nathanson N, Pallansch MA (2004) Circulating vaccine-derived polioviruses: current state of knowledge. Bull WHO 82:16–23 King AM (1988) Preferred sites of recombination in poliovirus RNA: an analysis of 40 intertypic cross-over sequences. Nucleic Acids Res 16:11705–1723 Kinnunen L, Huovilainen A, Poyry T, Hovi T (1990) Rapid molecular evolution of wild type 3 poliovirus during infection in individual hosts. J Gen Virol 71:317–324 Kirkegaard K, Baltimore D (1986) The mechanism of RNA recombination in poliovirus. Cell 47:433–443 Korotkova EA, Park R, Cherkasova EA, Lipskaya GY, Chumakov KM, Feldman EV, Kew OM, Agol VI (2003) Retrospective analysis of a local cessation of vaccination against poliomyelitis: a possible scenario for the future. J Virol 77:12460–12465 Kuge S, Nomoto A (1987) Construction of viable deletion and insertion mutants of the Sabin strain of type 1 poliovirus: function of the 5 noncoding sequence in viral replication. J Virol 61:1478–1487 Kuge S, Saito I, Nomoto A (1986) Primary structure of poliovirus defective-interfering particle genomes and possible generation mechanisms of the particles. J Mol Biol 192:473–487 Labadie K, Pelletier I, Saulnier A, Martin J, Colbère-Garapin F (2004) Poliovirus mutants excreted by a chronically infected hypogammaglobulinemic patient establish persistent infections in human intestinal cells. Virology 318:66–78 La Monica N, Racaniello VR (1989) Differences in replication of attenuated and neurovirulent polioviruses in human neuroblastoma cell line SH-SYY. J Virol 63:2357– 2360 Lazaro E, Escarmis C, Perez-Mercader J, Manrubia SC, Domingo E (2003) Resistance of virus to extinction on bottleneck passages: study of a decaying and fluctuating pattern of fitness loss. Proc Natl Acad Sci U S A 100:10830–10835 Ledinko N (1963) Genetic recombination with poliovirus type studies of crosses between a normal horse serum-resistant mutant and several guanidine-resistant mutants of the same strain. Virology 20:107–119 Li CP, Habel K (1951) Adaptation of Leon strain of poliomyelitis virus to mice. Proc Soc Exp Biol Med 78:233–238 Lipskaya GY, Muzychenko AR, Kutitova OK, Maslova SV, Equestre M, Drozdov SG, Perez Bercoff R, Agol VI (1991) Frequent isolation of intertypic poliovirus recombinants with serotype 2 specificity from vaccine-associated polio cases. J Med Virol 35:290–296 Lipskaya GY, Chervonskaya EA, Belova GI, Maslova SV, Kutateladze TN, Drozdov SG, Mulders M, Pallansch MA, Kew OM, Agol VI (1995) Geographical genotypes (geotypes) of poliovirus case isolates from the Former Soviet Union: relatedness to other known poliovirus genotypes. J Gen Virol 76:1687–1699 Liu H-M, Zheng D-P, Zhang L-B, Oberste MS, Pallansch MA, Kew OM (2000) Molecular evolution of a type 1 wild-vaccine poliovirus recombinant during widespread circulation in China. J Virol 74:11153–11161 Liu H-M, Zheng D-P, Zhang L-B, Oberste MS, Kew OM, Pallansch MA (2003) Serial recombination during circulation of type 1 wild-vaccine recombinant polioviruses in China. J Virol 77:10994–11005

254

V. I. Agol

Loddo B, Ferrari W, Spaneda A, Brotzu G (1962) In vitro guanidine-resistance and guanidine-dependence of poliovirus. Experientia 18:518–519 Lu HH, Wimmer E (1996) Poliovirus chimeras replicating under the translational control of genetic elements of hepatitis C virus reveal unusual properties of the internal ribosomal entry site of hepatitis C virus. Proc Natl Acad Sci U S A 93:1412–1417 Lu HH, Li X, Cuconati A, Wimmer E (1995) Analysis of picornavirus 2A(pro) proteins: separation of proteinase from translation and replication functions. J Virol 69:7445–7452 Lyons T, Murray KE, Roberts AW, Barton DJ (2001) Poliovirus 5 -terminal cloverleaf RNA is required in cis for VPg uridylylation and the initiation of negative-strand RNA synthesis. J Virol 75:10696–10708 Macadam AJ, Pollard SR, Ferguson G, Dunn G, Skuce R, Almond JW, Minor PD (1991) The 5 noncoding region of the type 2 poliovirus vaccine strain contains determinants of attenuation and temperature sensitivity. Virology 181:451–458 Macadam AJ, Ferguson G, Fleming T, Stone DM, Almond JW, Minor PD (1994) Role for poliovirus protease A in cap independent translation. EMBO J 13:924–947 MacLennan C, Dunn G, Huissoon AP, Kumararatne DS, Martin J, O’Leary P, Thompson RA, Osman H, Wood P, Minor P, Wood DJ, Pillay D (2004) Failure to clear persistent vaccine-derived neurovirulent poliovirus infection in an immunodeficient man. Lancet 363:1509–1513 Martin J, Dunn G, Hull R, Patel V, Minor P (2000) Evolution of the Sabin strain of type 3 poliovirus in an immunodeficient patient during the entire 637-day period of virus excretion. J Virol 74:3001–3010 Martin J, Samoilovich E, Dunn G, Lackenby A, Feldman E, Heath A, Svirchevskaya E, Cooper G, Yermalovich M, Minor PD (2002) Isolation of an intertypic poliovirus capsid recombinant from a child with vaccine-associated paralytic poliomyelitis. J Virol 76:10921–10928 Martin J, Odoom K, Tuite G, Dunn G, Hopewell N, Cooper G, Fitzharris C, Butler K, Hall WW, Minor PD (2004) Long-term excretion of vaccine-derived poliovirus by a healthy child. J Virol 78:13839–13847 Martinez I, Melero JA (2002) A model for the generation of multiple A to G transitions in the human respiratory syncytial virus genome: predicted RNA secondary structures as substrates for adenosine deaminases that act on RNA. J Gen Virol 83:1445–1455 Melchers WJ, Bakkers JM, Bruins Slot HJ, Galama JM, Agol VI, Pilipenko EV (2000) Cross-talk between orientation-dependent recognition determinants of a complex control RNA element, the enterovirus oriR. RNA 6:976–987 Melnick JL, Crowther D, Barrera-Oro J (1961) Rapid development of drug resistant mutants of poliovirus. Science 134:557 Minor PD (1990) Antigenic structure of picornaviruses. Curr Top Microbiol Immunol 161:121–154 Minor PD (2004) Polio eradication, cessation of vaccination and re-emergence of disease. Nat Rev Microbiol 2:473–482 Minor PD, Almond JW (2002) Poliovirus vaccines: molecular biology and immune response. In: Semler BL, Wimmer E (eds) Molecular biology of picornaviruses. ASM Press, Washington, pp 381–390

Molecular Mechanisms of Poliovirus Variation and Evolution

255

Minor PD, Schild GC, Bootman J, Evans DM, Ferguson M, Reeve P, Spitz M, Stanway G, Cann AJ, Hauptmann R, Clarke LD, Mountford RC, Almond JW (1983) Location and primary structure of a major antigenic site for poliovirus neutralization. Nature 301:674–679 Minor PD, John AM, Ferguson M, Icenogle JP (1986) Antigenic and molecular evolution of the vaccine strain of type 3 poliovirus during the period of excretion by a primary vaccinee. J Gen Virol 67:693–706 Molla A, Harris KS, Paul AV, Shin SH, Mugavero J, Wimmer E (1994) Stimulation of poliovirus proteinase Cpro-related proteolysis by the genome-linked protein VPg and its precursor AB. J Biol Chem 269:27015–27020 Morasco BJ, Sharma N, Parilla J, Flanegan JB (2003) Poliovirus cre(C)-dependent synthesis of VPgpUpU is required for positive- but not negative-strand RNA synthesis. J Virol 77:5136–5144 Mulders MN, Reimerink JH, Stenvik M, Alaeddinoglu I, van der Avoort HG, Hovi T, Koopmans MP (1999) A Sabin vaccine-derived field isolate of poliovirus type 1 displaying aberrant phenotypic and genetic features, including a deletion in antigenic site. J Gen Virol 80:907–916 Murray KE, Barton DJ (2003) Poliovirus CRE-dependent VPg uridylylation is required for positive-strand RNA synthesis but not for negative-strand RNA synthesis. J Virol 77:4739–4750 Muzychenko AR, Lipskaya GY, Maslova SV, Svitkin YV, Pilipenko EV, Nottay BK, Kew OM, Agol VI (1991) Coupled mutations in the 5 -untranslated region of the Sabin poliovirus strains during in vivo passages: structural and functional implications. Virus Res 21:111–122 Nkowane BU, Wassilak SG, Orenstein WA (1987) Vaccine-associated paralytic poliomyelitis in the United States: 1973 through 1984. JAMA 257:1335–1340 Nottay BK, Kew OM, Hatch MH, Heyward JT, Obijeski JF (1981) Molecular variation of type 1 vaccine-related and wild polioviruses during replication in humans. Virology 108:405–423 Paul AV (2002) Possible unifying mechanisms of picornavirus genome replication. In: Semler BL, Wimmer E (eds) Molecular biology of picornaviruses. ASM Press, Washington, pp 227–246 Paul AV, Rieder E, Kim DW, van Boom JH, Wimmer E (2000) Identification of an RNA hairpin in poliovirus RNA that serves as the primary template in the in vitro uridylylation of VPg. J Virol 74:10359–10370 Paul AV, Peters J, Mugavero J, Yin J, van Boom JH, Wimmer E (2003) Biochemical and genetic studies of the VPg uridylylation reaction catalyzed by the RNA polymerase of poliovirus. J Virol 77:891–904 Pavio N, Couderc T, Girard S, Sgro JY, Blondel B, Colbère-Garapin F (2000) Expression of mutated poliovirus receptors in human neuroblastoma cells persistently infected with poliovirus. Virology 274:331–342 Pelletier I, Ouzilou L, Arita M, Nomoto A, Colbère-Garapin F (2003) Characterization of the poliovirus 14S particle: new insights into poliovirus uncoating. Virology 305:55–65 Pestova TV, Kolupaeva VG, Lomakin IB, Pilipenko EV, Shatsky IN, Agol VI, Hellen CU (2001) Molecular mechanisms of translation initiation in eukaryotes. Proc Natl Acad Sci U S A 98:7029–7036

256

V. I. Agol

Pfeiffer JK, Kirkegaard K (2003) A single mutation in poliovirus RNA-dependent RNA polymerase confers resistance to mutagenic nucleotide analogs via increased fidelity. Proc Natl Acad Sci U S A 100:7289–7294 Pfeiffer JK, Kirkegaard K (2005) Increased fidelity reduces poliovirus fitness and virulence under selective pressure in mice. PLoS Pathog 1:e11 Pilipenko EV, Blinov VM, Romanova LI, Sinyakov AN, Maslova SV, Agol VI (1989) Conserved structural domains in the 5 -untranslated region of picornaviral genomes: an analysis of the segment controlling translation and neurovirulence. Virology 168:201–209 Pilipenko EV, Blinov VM, Agol VI (1990) Gross rearrangements within the 5 -untranslated region of the picornaviral genomes. Nucl Acids Res 18:3371–3375 Pilipenko EV, Gmyl AP, Maslova SV, Svitkin YV, Sinyakov AN, Agol VI (1992a) A prokaryotic-like cis-element in the cap-independent internal initiation of translation on picornavirus RNA. Cell 68:119–131 Pilipenko EV, Maslova SV, Sinyakov AN, Agol VI (1992b) Toward identification of cisacting elements involved in the replication of enterovirus and rhinovirus RNAs: a proposal for the existence of tRNA-like terminal structures. Nucl Acids Res 20:1739–1745 Pilipenko EV, Gmyl AP, Agol VI (1995) A model for rearrangements in RNA genomes. Nucleic Acids Res 23:1870–1875 Pilipenko EV, Poperechny KV, Maslova SV, Melchers WJG, Bruins Slot HJ, Agol VI (1996) Cis-element, oriR, involved in the initiation of (–) strand poliovirus RNA: a quasi-globular multi-domain RNA structure maintained by tertiary (“kissing”) interaction. EMBO J 19:5428–5436 Pincus SE, Rohl H, Wimmer E (1987) Guanidine-dependent mutants of poliovirus: identification of three classes with different growth requirements. Virology 157:83–88 Poyry T, Kinnunen L, Kapsenberg J, Kew O, Hovi T (1990) Type 3 poliovirus/Finland/1984 is genetically related to common Mediterranean strains. J Gen Virol 71:2535–2541 Poyry T, Kinnunen L, Hovi T (1992) Genetic variation in vivo and proposed functional domains of the 5 noncoding region of poliovirus RNA. J Virol 66:5313–5319 Pugachev KV, Guirakhoo F, Ocran SW, Mitchell F, Parsons M, Penal C, Girakhoo S, Pougatcheva SO, Arroyo J, Trent DW, Monath TP (2004) High fidelity of yellow fever virus RNA polymerase. J Virol 78:1032–1038 Racaniello VR (2002) Picornaviridae: the viruses and their replication. In: Knipe DM, Howley PM (eds) Fields virology, vol 1. Lippincott Williams & Wilkins, Philadelphia. pp 685–722 Reynolds C, Bimby D, Chow M (1992) Folding and processing of capsid precursor P1 is kinetically retarded in neutralization site B mutants of poliovirus. J Virol 66:1641–1648 Rezapkin GV, Chumakov KM, Lu Z, Ran Y, Dragunsky EM, Levenbook IS (1994) Microevolution of Sabin 1 strain in vitro and genetic stability of oral poliovirus vaccine. Virology 202:370–378 Rezapkin GV, Norwood LP, Taffs RE, Dragunsky EM, Levenbook IS, Chumakov KM (1995) Microevolution of type 3 Sabin strain of poliovirus in cell cultures and its implications for oral poliovirus vaccine quality control. Virology 211:377–384

Molecular Mechanisms of Poliovirus Variation and Evolution

257

Rezapkin GV, Douthitt M, Dragunsky E, Chumakov KM (1999) Reevaluation of nucleotide sequences of wild-type and attenuated polioviruses of type Virus Res 65:111–119 Rico-Hesse R, Pallansch MA, Nottay BK, Kew OM (1987) Geographic distribution of wild poliovirus type 1 genotypes. Virology 160:311–322 Rieder E, Wimmer E (2002) Cellular receptors of picornaviruses: an overwiev. In: Semler BL, Wimmer E (eds) Molecular biology of picornaviruses. ASM Press, Washington, pp 61–70 Rieder E, Gorbalenya AE, Xiao C, He Y, Baker TS, Kuhn RJ, Rossmann MG, Wimmer E (2001) Will the polio niche remain vacant? Dev Biol (Basel) 105:111–122 Rieder E, Xiang W, Paul A, Wimmer E (2003) Analysis of the cloverleaf element in a human rhinovirus type 14/poliovirus chimera: correlation of subdomain D structure, ternary protein complex formation and virus replication. J Gen Virol 84:2203–2216 Rightsel WA, Dice JR, McAlpine RJ, Timm EA, McLean IW, Dixon GJ, Schabel FM (1961) Antiviral effect of guanidine. Science 134:558–559 Rivera VM, Welsh JD, Maizel J (1988) Comparative sequence analysis of the 5 noncoding region of the enteroviruses and rhinoviruses. Virology 165:42–50 Roca-Garcia M, Moyer AW, Cox HR (1953) Poliomyelitis. Propagation of MEF-1 strain of poliomyelitis virus in developing chick embryo by yolk sac inoculation. Proc Soc Exp Biol Med 81:519–525 Rohll JB, Percy N, Ley R, Evans DJ, Almond JW, Barclay WS (1994) The 5 -untranslated regions of picornavirus RNAs contain independent functional domains essential for RNA replication and translation. J Virol 68:4384–4391 Rohll JB, Moon DH, Evans DJ, Almond JW (1995) The 3 untranslated region of picornavirus RNA: features required for efficient genome replication. J Virol 69:7835–7844 Roivainen M, Hovi T (1987) Intestinal trypsin can significantly modify antigenic properties of polioviruses: implications for the use of inactivated poliovirus vaccine. J Virol 61:3749–3753 Romanova LI, Blinov VM, Tolskaya EA, Viktorova EG, Kolesnikova MS, Guseva EA, Agol VI (1986) The primary structure of crossover regions of intertypic poliovirus recombinants: a model of recombination between RNA genomes. Virology 155:202–213 Rousset D, Rakoto-Andrianarivelo M, Razafindratsimandresy R, Randriamanalina B, Guillot S, Balanant J, Mauclere P, Delpeyroux F (2003) Recombinant vaccinederived poliovirus in Madagascar. Emerg Infect Dis 9:885–887 Rowe A, Ferguson GL, Minor PD, Macadam AJ (2000) Coding changes in the poliovirus protease A compensate for 5 NCR domain V disruptions in a cell-specific manner. Virology 269:284–293 Ruiz-Jarabo CM, Miller E, Gomez-Mariano G, Domingo E (2003) Synchronous loss of quasispecies memory in parallel viral lineages: a deterministic feature of viral quasispecies. J Mol Biol 333:553–563 Sabin AB, Boulger LR (1973) History of Sabin attenuated poliovirus oral live vaccine. J Biol Stand 1:115–118 Saunderson R, Yu B, Trent RJ, Pamphlett R (2004) A polymorphism in the poliovirus receptor gene differs in motor neuron disease. Neuroreport 15:383–386

258

V. I. Agol

Semler BL, Wimmer E (eds) (2002) Molecular biology of picornaviruses. ASM Press, Washington Semler BL, Johnson VH, Tracy S (1986) A chimeric plasmid from cDNA clones of poliovirus and coxsackievirus produces a recombinant virus that is temperaturesensitive. Proc Natl Acad Sci U S A 83:1777–1781 Shimizu H, Thorley B, Paladin FJ, Brussen KA, Stambos V, Yuen L, Utama A, Tano Y, Arita M, Yoshida H, Yoneyama T, Benegas A, Roesel S, Pallansch M, Kew O, Miyamura T (2004) Circulation of type 1 vaccine-derived poliovirus in the Philippines in 2001. J Virol 78:13512–13521 Shulman LM, Handsher R, Yang C-F, Yang S-J, Manor J, Vonsover A, Grossman Z, Pallansch M, Mendelson E, Kew OM (2000a) Resolution of the pathways of poliovirus Type 1 transmission during an outbreak. J Clin Micorobiol 38:945–952 Shulman LM, Manor Y, Handsher R, Delpeyroux F, McDonough MJ, Halmut T, Silberstein I, Alfandari J, Quay J, Fisher T, Robinov J, Kew OM, Crainic R, Mendelson E (2000b) Molecular and antigenic characterization of a highly evolved derivative of the type 2 oral poliovaccine strain isolated from sewage in Israel. J Clin Microbiol 38:3729–3734 Sierra S, Dávila M, Lowenstein PR, Domingo E (2000) Response of foot-and-mouth disease virus to increased mutagenesis: influence of viral load and fitness in loss of infectivity. J Virol 74:8316–8323 Skinner MA, Racaniello V, Dunn G, Cooper J, Minor PD, Almond JW (1989) New model for the secondary structure of the 5 non-coding RNA of poliovirus is supported by biochemical and genetic data that also show that RNA secondary structure is important in neurovirulence. J Mol Biol 207:379–392 Slobodskaya OR, Gmyl AP, Maslova SV, Tolskaya EA, Viktorova EG, Agol VI (1996) Poliovirus neurovirulence correlates with the presence of a cryptic AUG upstream of the initiator codon. Virology 221:141–150 Svitkin YV, Maslova SV, Agol VI (1985) The genomes of attenuated and virulent poliovirus strains differ in their in vitro translation efficiencies. Virology 147:243– 252 Svitkin YV, Pestova TV, Maslova SV, Agol VI (1988) Point mutations modify the response of poliovirus RNA to a translation initiation factor: a comparison of neurovirulent and attenuated strains. Virology 166:394–404 Taffs RE, Chumakov KM, Rezapkin GV, Lu Z, Douthitt M, Dragunsky EM, Levenbook IS (1995) Genetic stability and mutant selection in Sabin 2 strain of oral poliovirus vaccine grown under different cell culture conditions. Virology 209:366–373 Todd S, Towner JS, Brown DM, Semler BL (1997a) Replication-competent picornaviruses with complete genomic RNA 3 noncoding region deletions. J Virol 71:8868–8874 Todd S, Towner JS, Semler BL (1997b) Translation and replication properties of the human rhinovirus genome in vivo and in vitro. Virology 229:90–97 Tolskaya EA, Romanova LI, Kolesnikova MS, Agol VI (1983) Intertypic recombination in poliovirus: genetic and biochemical studies. Virology 124:121–132 Tolskaya EA, Romanova LI, Blinov VM, Viktorova EG, Sinyakov AN, Kolesnikova MS, Agol VI (1987) Studies on the recombination between RNA genomes of poliovirus: the primary structure and nonrandom distribution of crossover regions in the genomes of intertypic poliovirus recombinants. Virology 161:54–62

Molecular Mechanisms of Poliovirus Variation and Evolution

259

Tolskaya EA, Romanova LI, Kolesnikova MS, Gmyl AP, Gorbalenya AE, Agol VI (1994) Genetic studies on the NTP-binding pattern containing C protein of poliovirus: possible mechanism of guanidine effect on C function and evidence for importance of C oligomerization. J Mol Biol 236:1310–1323 van Kuppeveld FJ, van den Hurk PJ, van der Vliet W, Galama JM, Melchers WJ (1997) Chimeric coxsackie B3 virus genomes that express hybrid coxsackieviruspoliovirus B proteins: functional dissection of structural domains involved in RNA replication. J Gen Virol 78:1833–1840 van Ooij MJM, Vogt DA, Paaul A, Castro C, Kuijpers J, van Kuppeveld FJM, Cameron CE, Wimmer E, Andino R, Melchers WJG (2005) Structural and functional characterization of the Coxsackievirus B3 CRE(2C): role of CRE(2C) in negative and positive strand RNA synthesis. J Gene Virol (in press) Vignuzzi M, Stone JK, Andino A (2005) Ribavirin and lethal mutagenesis of poliovirus: molecular mechanisms, resistance and biological implications. Virus Res 107:173– 181 Ward CD, Stokes MA, Flanegan JB (1988) Direct measurement of the poliovirus RNA polymerase error frequency in vitro. J Virol 62:558–562 Wells VR, Plotch SJ, DeStefano JJ (2001) Determination of the mutation rate of poliovirus RNA-dependent RNA polymerase. Virus Res 74:119–132 Wien MW, Curry S, Filman DJ, Hogle JM (1997) Structural studies of poliovirus mutants that overcome receptor defects. Nat Struct Biol 4:666–674 Wimmer E, Hellen CUT, Cao X (1993) Genetics of poliovirus. Annu Rev Genet 27:353– 436 Wong SK, Lazinski DW (2002) Replicating hepatitis delta virus RNA is edited in the nucleus by the small form of ADAR1. Proc Natl Acad Sci U S A 99:15118–15123 Xiang W, Cuconati A, Paul AV, Cao X, Wimmer E (1995a) Molecular dissection of the multifunctional poliovirus RNA-binding protein AB. RNA 1:892–904 Xiang W, Harris KS, Alexander L, Wimmer E (1995b) Interaction between the 5 terminal cloverleaf and AB/CDpro of poliovirus is essential for RNA replication. J Virol 69:3658–3667 Xiang W, Cuconati A, Hope D, Kirkegaard K, Wimmer E (1998) Complete protein linkage map of poliovirus P3 proteins: interaction of polymerase Dpol with VPg and with genetic variants of AB. J Virol 72:6732–6741 Yang CF, Naguib T, Yang SJ, Nasr E, Jorba J, Ahmed N, Campagnoli R, van der Avoort H, Shimizu H, Yoneyama T, Miyamura T, Pallansch M, Kew O (2003) Circulation of endemic type 2 vaccine-derived poliovirus in Egypt from 1983 to 1993. J Virol 77:366–377 Yang Y, Rijnbrand R, Watowich S, Lemon SM (2004) Genetic evidence for an interaction between a picornaviral cis-acting RNA replication element and CD protein. J Biol Chem 279:12659–12667 Yin J, Paul AV, Wimmer E, Rieder E (2003) Functional dissection of a poliovirus cis-acting replication element [PV-cre(C)]: analysis of single- and dual-cre viral genomes and proteins that bind specifically to PV-cre RNA. J Virol 77:5152–5166 Zell R, Klingel K, Sauter M, Fortmuller U, Kandolf R (1995) Coxsackieviral proteins functionally recognize the polioviral cloverleaf structure of the 5 -NTR of a chimeric enterovirus RNA: influence of species-specific host cell factors on virus growth. Virus Res 39:87–103

CTMI (2006) 299:261–284 c Springer-Verlag Berlin Heidelberg 2006


Hepatitis C Virus Population Dynamics During Infection J.-M. Pawlotsky (u) Service de Virologie, Hôpital Henri Mondor, 51 avenue du Maréchal de Lattre de Tassigny, 94010 Créteil, France [email protected]

1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262

2 2.1 2.2 2.3 2.4

Hepatitis C Virus Genetic Variability . . . . . . . . . . Mechanisms of HCV Replication . . . . . . . . . . . . . HCV Genetic Variability . . . . . . . . . . . . . . . . . . . Emergence and Diversification of HCV Genotypes . Quasispecies Distribution of HCV Genomes . . . . .

. . . . .

. . . . .

262 262 263 264 264

3 3.1 3.2 3.3 3.3.1 3.3.2 3.3.3 3.3.4

Evolution and Constraints on HCV Genomes . . . . . . . . . . . . . . . . . . . Constraints on HCV Genetic Evolution in Immune-Free Environments . Constraints on HCV Evolution in Experimentally Infected Chimpanzees Constraints on HCV Genetic Evolution in Natural Human Infection . . . Variable Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conserved Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conservation of Variable Regions, Variability of Conserved Regions . . . Functional Consequences of the HCV Quasispecies Distribution . . . . . .

. . . . . . . .

265 265 266 266 267 268 268 270

4 4.1 4.2 4.3 4.4

Dynamics of HCV Quasispecies . . . . . . . . . . . . . . . . . . . HCV Transmission . . . . . . . . . . . . . . . . . . . . . . . . . . . . Dynamics of HCV Quasispecies During Acute Infection . . Dynamics of HCV Quasispecies During Chronic Infection Dynamics of HCV Quasispecies During Antiviral Therapy

. . . . .

273 273 273 275 275

5

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277 Abstract Hepatitis C virus (HCV) behaves as an evolving viral quasispecies in its continuously changing environment. The study of HCV quasispecies population dynamics in experimental models and infected patients can provide useful information on factors involved in the HCV life cycle and pathogenicity. HCV quasispecies variability also has therapeutic implications, as the continuous generation and selection of fitter or truly resistant variants can allow the virus to escape control by antiviral drugs.

262

J.-M. Pawlotsky

1 Introduction Hepatitis C virus (HCV), like many other RNA viruses and viruses using RNA as a replication intermediate, circulates in infected individuals as a quasispecies, i.e., a mixture of genetically distinct but closely related variants (Martell et al. 1992; Weiner et al. 1991). This is principally due to the properties of the viral RNA-dependent RNA polymerase (RdRp), the enzyme that ensures viral replication, which cannot correct nucleotide misincorporations that occur randomly during replication. At any given point of the infection, the variant populations are at a steady state that depends closely on the replicative environment. The quasispecies distribution of HCV populations is often overlooked in clinical and experimental studies. Yet the study of HCV quasispecies population dynamics in experimental models and infected patients can provide useful information on factors involved in the HCV life cycle and pathogenicity. For instance, HCV quasispecies variability appears to have a subtle regulatory influence on viral protein and RNA functions that could play an important role in chronicity. The fact that HCV behaves as a quasispecies and has very rapid replicative kinetics also has therapeutic implications, as the continuous generation and selection of fitter or truly resistant variants can allow the virus to escape control by antiviral drugs. This chapter reviews HCV quasispecies variability and dynamics, and discusses the principal pathophysiological and clinical implications.

2 Hepatitis C Virus Genetic Variability 2.1 Mechanisms of HCV Replication HCV mainly replicates in hepatocytes, the most abundant liver cells. HCV replication is catalyzed by RdRp, a 68-kDa viral protein encoded by the nonstructural (NS) genomic region 5B located at the 3 end of the open reading frame (Lohman et al. 1997; Behrens et al. 1996). The three-dimensional structure of RdRp was recently resolved by means of X-ray crystallography(Bressanelli et al. 1999; Lesburg et al. 1999; Love et al. 2003; Wang 2003). In infected cells, RdRp localizes near the perinuclear membranes, where it associates with other nonstructural viral proteins and cellular factors to form a replication complex (Penin et al. 2004). The replication complex takes the form of a membranous web in Huh7 cell lines harboring HCV replicons, i.e.,

Hepatitis C Virus Population Dynamics During Infection

263

modified HCV genomes that self-replicate in nonproductive cell cultures (Egger et al. 2002; Gosert et al. 2003; Moradpour et al. 2003). During replication, RdRp interacts with the 3 end of the viral RNA, the hairpin structure of which may serve as the initiator oligonucleotide for RNA synthesis (Cheng et al. 1999). The polymerization reaction leads to the synthesis of complementary RNA strands. Negative-strand RNAs, the replication intermediates, are synthesized with genomic RNAs as templates, whereas positive-strand RNAs are synthesized with negative-strand RNAs as templates and are subsequently encapsidated into new virions or used as messenger RNAs for viral protein synthesis. 2.2 HCV Genetic Variability HCV genetic variability results from a typical Darwinian evolutionary process in which continuous diversification of viral populations leads to competition among them and to continuous selection of the fittest variants by the environment in which the virus replicates. This phenomenon is favored by the rapid replication of HCV (producing on average 1012 virions per day in infected adults [Neumann et al. 1998]), the large viral population sizes, and high mutation rates related to the low fidelity of RdRp (Duarte et al. 1994; Domingo 1998). Indeed, HCV RdRp has a high error rate, with misincorporation frequencies averaging about 10–4 –10–5 per copied nucleotide, meaning that each progeny viral RNA synthesized in an infected cell contains, on average, approximately one mutation per RNA molecule (Domingo 1996). As the viral RdRp is devoid of 3 -5 proofreading exonuclease activity and other postreplicative repair mechanisms, the incorporated substitutions are not corrected. Mutations incorporated into negative RNA strands probably have a greater impact, because they can be quickly transmitted to a large number of positive-strand progeny RNAs. A substantial proportion of newly synthesized viral genomes are defective, because of highly deleterious genetic lesions. In contrast, nonlethal mutations that accumulate during viral replication are transmitted to progeny viruses and confer selective advantages or disadvantages, depending on the replicative environment (Duarte et al. 1994; Domingo 1996, 1998). The continual appearance and selection of new mutations has two important consequences: (a) the selection of variant strains during evolution in geographically or epidemiologically distinct populations, which is responsible for progressive diversification of viral genotypes and subtypes (Simmonds 1995, 1998, 2001); and (b) the quasispecies distribution of the virus population in a given individual (Martell et al. 1992; Weiner et al. 1991).

264

J.-M. Pawlotsky

2.3 Emergence and Diversification of HCV Genotypes Phylogenetic analyses of partial or full-length sequences of HCV strains isolated in various regions of the world has led to the identification of six main HCV genotypes, designated types 1–6 (Robertson et al. 1998). A seventh type was recently identified (Innogenetics, unpublished data). The HCV types comprise a large number of subtypes, identified by lower-case letters (1a, 1b, etc.) (Robertson et al. 1998). The types differ by between 31% and 34% of their nucleotide sequences and approximately 30% of their amino acid sequences, whereas the subtypes differ by between 20% and 23% of their nucleotide sequences, with marked differences in particular genomic regions (Robertson et al. 1998). Variants isolated principally in Southeast Asia were found to have a divergence level between types and subtypes (Tokita et al. 1994, 1995, 1996). They were incorporated into their phylogenetically closest type (all type 6, except for one type 3 strain) (Robertson et al. 1998). The different HCV genotypes are strongly associated with particular routes of transmission (Pawlotsky et al. 1995) and are major determinants of the response to antiviral combination therapy with interferon (IFN)-α and ribavirin (Manns et al. 2001; Fried et al. 2002; Hadziyannis et al. 2004). HCV genotyping is currently performed prior to initiating therapy, in order to tailor the dose of ribavirin and the duration and frequency of monitoring to the individual patient Consensus Panel 2002). 2.4 Quasispecies Distribution of HCV Genomes HCV, like many other RNA viruses, does not circulate in infected individuals as a homogeneous population of identical viral particles, but as a pool of genetically distinct but closely related variants referred to collectively as a quasispecies (Martell et al. 1992; Weiner et al. 1991). The quasispecies nature of HCV confers a significant survival advantage, as the simultaneous presence of multiple variant genomes and the high rate at which new variants are generated allow rapid selection of mutants better suited to new environmental conditions (Duarte et al. 1994; Domingo 1996, 1998). Contradictory selection forces are involved, including: (a) conservatory constraints on HCV RNA or protein sequences, related to the need to conserve their functional properties and to ensure virus survival; (b) positive pressures toward change, principally related to humoral and cellular host immune responses. Environmental changes occur frequently during the course of HCV infection. They can be spontaneous, related to complex metabolic interactions in the host, or

Hepatitis C Virus Population Dynamics During Infection

265

triggered by external factors such as intercurrent infections, drug intake, and antiviral treatments.

3 Evolution and Constraints on HCV Genomes Nucleotide mutations may occur at any position on the HCV genome during replication, as a result of the low fidelity of the RdRp. Nevertheless, different regions are subject to different selection constraints depending on the functional role of the RNA or the encoded protein and the presence of target motifs for selection pressure by host immune responses or, possibly, interactions with cellular proteins. 3.1 Constraints on HCV Genetic Evolution in Immune-Free Environments In the absence of immunological selection pressure, HCV should theoretically accumulate nucleotide mutations at random positions during replication, and the newly generated variants should be selected principally, if not solely, on the basis of the intrinsic replicative advantages or disadvantages these mutations confer. In the replicon system, an in vitro model of HCV replication in a transfected Huh7 cell line, mutations indeed accumulate at a high rate at random positions (Lohmann et al. 1999, 2001; Blight et al. 2000; Pietschmann et al. 2001; Krieger et al. 2001, Cheney et al. 2002). Mutations appear to be better tolerated than during human infection, because there is no pressure to produce infectious virions in this system. Long-term replication of the replicon system leads to the selection of adaptive mutations at various genomic positions. These adaptive mutations confer a selective advantage, because they are associated with increased replication capacities in vitro (Lohmann et al. 1999; Blight et al. 2000; Krieger et al. 2001, Cheney et al. 2002). Nevertheless, adaptive mutations in the replicon system were shown to be lethal for HCV replication in chimpanzees after intrahepatic transfection of infectious cDNAs into which they were inserted (Bukh et al. 2002). This suggests that survival advantages or disadvantages conferred by specific substitutions are strongly dependent on the environment in which the virus replicates. These findings also indicate that one should be cautious when discussing the clinical relevance of data obtained in vitro or in models in which the replicative environment differs substantially from the infected human host. We recently showed that HCV replicates in primary cultures of normal human hepatocytes infected in vitro, the closest model to natural human

266

J.-M. Pawlotsky

infection Castet et al. 2002. In this system, HCV replication is transient and is characterized by random accumulation of nucleotide substitutions on the HCV genome (Castet et al. 2002). Interestingly, IFN-α, the principal molecule used to treat chronic HCV infection, significantly reduces the rate at which mutations accumulate, as a result of RdRp inhibition (Castet et al. 2002). In this nonproductive model, conservative constraints on HCV genomic evolution appear to be weak, and no selection of amino acid mutations conferring a survival advantage was observed in the absence of immune pressure (Castet et al. 2002). 3.2 Constraints on HCV Evolution in Experimentally Infected Chimpanzees HCV quasispecies changes in the E2 envelope glycoprotein gene have been compared in acutely infected humans and chimpanzees. HCV sequence diversity was significantly lower in chimpanzees than in humans, and serial passage in chimpanzees were not associated with changes in the sequence of the major variants, nonsynonymous mutations per nonsynonymous site being less frequent in chimpanzees than in humans (Ray et al. 2000; Sugitani and Shikata 1998). In addition, we recently showed in chimpanzees that partially controlled virus replication for several weeks prior to developing chronic infection, that the viral quasispecies that reappeared after downregulation was essentially identical to the initial quasispecies (Prince et al. 2004). Together, these findings indicate that HCV quasispecies are subjected to substantially weaker selection pressures in chimpanzees than during natural human infection. In addition, the low rate of accumulation of amino acid substitutions on the HCV genome after intrahepatic transfection of infectious HCV clones in chimpanzees (Bukh 2004) suggests that HCV is subjected to strong conservative constraints on its protein sequences, likely related to the function of these proteins in the HCV life cycle and the onset of chronicity. 3.3 Constraints on HCV Genetic Evolution in Natural Human Infection In the infected human host, conservatory constraints on the HCV genome and/or protein sequences, and positive pressures toward change (principally related to the immune response), differ according to the genomic region. As a result, it has been possible to define conserved and variable regions in the HCV genome. Variable regions are constrained to some extent by the need to conserve the essential biological functions of the proteins, whereas conserved regions may tolerate a certain number of changes that play a role in the regulation of RNA structure or protein functions.

Hepatitis C Virus Population Dynamics During Infection

267

3.3.1 Variable Regions Certain coding regions are variable, i.e., highly tolerant of amino acid changes. This is the case of a 27-amino-acid stretch located at the N-terminus of the E2 envelope glycoprotein, named the hypervariable region 1 (HVR1) (Weiner et al. 1991). HVR1 varies considerably among genotypes and also among different strains of the same genotype, and is subject to frequent amino acid changes during the acute infection. In the chronic stage of infection, HVR1 quasispecies have a high genetic complexity and evolve rapidly, both spontaneously and under the influence of environmental factors such as antiviral treatments (Pawlotsky et al. 1998a, 1999; Enomoto et al. 1994, 1995; Nagasaka et al. 1996; Polyak et al. 1998; Shindo et al. 1996; Yeh et al. 1996). We observed profound HVR1 changes during IFN-α treatment of patients with chronic

Fig. 1 Phylogenetic tree (amino acid sequences) of hypervariable region 1 (HVR1) quasispecies in a patient studied before IFN-α treatment, during treatment (months 3 and 6) and after treatment, showing distinctive clustering of quasispecies variants isolated at different time points and reflecting significant genetic evolution of the HVR1 during the study period. The phylogenetic reconstruction is a neighbor-joining tree with bootstrap proportions of more than 500 of 1,000 bootstrap replicates shown at appropriate branch points. (Adapted from Pawlotsky et al. 1999, with permission)

268

J.-M. Pawlotsky

hepatitis C, which appeared to result from strong positive pressure toward change, probably stimulated by the treatment itself (see example in Fig. 1) (Pawlotsky et al. 1999). As HVR1 is a target for neutralizing responses to HCV, it is likely that changes in HVR1 are at least partly driven by neutralizing antibody responses. Other variable regions include hypervariable region 2 (HVR2) located downstream of the HVR1 in the E2 envelope glycoprotein of HCV genotype 1 (Kato et al. 1992; Saito et al. 1996), and the V3 region located in the C-terminal part of NS5A protein (Nousbaum et al. 2000). It is unclear whether these regions are also subjected to particularly strong selection pressure from host responses or simply tolerate a larger number of substitutions. 3.3.2 Conserved Regions Conserved regions can only tolerate a few sporadic amino acid changes at specific positions. This is the case of the NS5A protein sequence (excluding the V3 region) (Pawlotsky et al. 1998b), NS3 proteinase (Soler et al., unpublished data), NS3 helicase, and RdRp (Pawlotsky et al., unpublished data). These proteins have limited natural amino acid variability, in order to conserve their functional properties in vivo. In the corresponding genome regions there are few if any constraints on the encoded nucleotide sequence, and synonymous mutations accumulate rapidly during replication and are transmitted, whereas nonsynonymous mutations are negatively selected when they occur (Pawlotsky et al. 1998b). Some noncoding regions of the genome, such as the functional parts of the 5 and the 3 untranslated regions, are also highly conserved. This is the case of the internal ribosome entry site (IRES) structure, which spans most of the HCV 5 noncoding region and the first nucleotides of the coreencoding region (Cuceanu et al. 2001; Smith et al. 1995). Like other genomic regions, the IRES has a quasispecies distribution but only certain nucleotide positions tolerate mutations (mainly those located in unpaired regions) (Soler et al. 2002). Mutations in paired regions are rare, and are often associated with compensatory mutations that preserve base-pairing and the stem loop structure of the IRES in order to conserve its translational properties (Soler et al. 2002). 3.3.3 Conservation of Variable Regions, Variability of Conserved Regions We recently showed that despite its apparent amino acid variability, the physicochemical characteristics of HVR1 residues are conserved at most positions

Hepatitis C Virus Population Dynamics During Infection

269

in different HCV genotypes and different strains of a given genotype, and even among quasispecies variants isolated from the same individual at single and multiple time points of infection (Penin et al. 2001). A consensus HVR1 sequence has been defined on the basis of the hydrophobic characteristics of HVR1 residues in all HVR1 sequences so far submitted to GenBank (Fig. 2) (Penin et al. 2001). This consensus sequence was shown to apply to all of the HVR1 sequences included in GenBank after its publication (F. Penin, unpublished data). Thus, despite of its tolerance of amino acid changes, HVR1 appears to be subject to strong conservatory constraints with regard to the physicochemical properties of the residues, pointing to strong conservation of its conformation and to a significant role in the virus life cycle (Penin et al. 2001). However, HVR1 might not be absolutely mandatory for the virus to be infectious, as infectious clones lacking HVR1 can chronically infect chimpanzees (Forns et al. 2000), albeit mildly (Forns et al. 2000). HVR1 has been conserved during HCV evolution in humans, and is present in all human HCV isolates (Penin et al. 2001). HVR1 thus appears to confer a significant survival advantage, yet its precise role in the viral life cycle is unknown. We recently suggested that HVR1 might be involved in facilitating HCV infection of target cells, a specific property of primate bloods, in cooperation with

Fig. 2 HVR1 consensus hydropathic pattern established from 1,382 HVR1 sequences: o, n, i, and v refer to the hydrophobic, neutral, hydrophilic, and variable positions, respectively; G is a fully conserved glycine residue. The + signs indicate positions often occupied by basic residues. (Adapted from Penin et al. 2001, with permission)

270

J.-M. Pawlotsky

the cellular scavenger receptor B1 (a candidate component of the receptor complex) and circulating high-density lipoproteins (Bartosch et al., in press). The fact that other variable regions, such as the NS5A V3 region, are always present suggests that they too could play a role in the virus life cycle (Nousbaum et al. 2000). Conversely, highly constrained proteins such as NS3 proteinase, NS5A, and RdRp tolerate amino acid changes at certain positions. These changes are sporadic and mainly involve residues located at the protein surface (Soler et al., unpublished data; Pawlotsky et al., unpublished data) Pawlotsky et al. 1998b). Such changes therefore do not alter the functionally vital three-dimensional structure of the hydrophobic core. They could, however, modify viral protein interactions with cellular partners and thus be involved in the pathophysiology of HCV-related disease. Substitutions located within or close to critical functional sites are very rare. We recently observed minor quasispecies variants bearing mutations within the catalytic site or the zinc-binding site of NS3 serine proteinase (Soler et al., unpublished data). It is unclear whether these sequences encode functional NS3 proteinases or belong to replicationdeficient virions encapsidated through cooperation with infectious genomes. The recent demonstration that NS3 proteinase and RdRp inhibitors select replication-competent HCV replicons bearing mutations in these drugs’ target sites (Lu et al. 2004; Lin et al. 2004; Nguyen et al. 2003; Migliaccio et al. 2003) suggests that such variants exist as minor quasispecies variants and can be selected, at least in vitro, on the basis of their improved fitness in the presence of the drugs. It is thus very likely that the use of such antiviral drugs in vivo will select resistant variants bearing substitutions in functionally important sites. Conserved functional RNA structures can also tolerate a degree of variation. Indeed, even the 5 noncoding region, by far the most strongly conserved HCV RNA region, has a quasispecies distribution. A minority of mutations occur at positions where they may disrupt base pairing and alter the stability of the IRES stem loop (Soler et al. 2002). Although the selection of nucleotide substitutions conferring resistance to specific IRES inhibitors has never been observed (Soler et al. 2004), this is very likely to occur if such IRES inhibitors reach the clinical development phase. 3.3.4 Functional Consequences of the HCV Quasispecies Distribution In vivo and in vitro experiments have shown that amino acid and nucleotide sequence differences can be associated with quantitatively different functions of viral proteins and RNAs, respectively. Extrahepatic sites of HCV replication have been identified, including certain subsets of peripheral blood mononu-

Hepatitis C Virus Population Dynamics During Infection

271

clear cells and dendritic cells (Shimizu et al. 1992; Shimizu and Yoshikura 1994; Lerat et al. 1998; Negro and Levrero 1998; Goutagny et al. 2003). Interestingly, quasispecies variants isolated from different body compartments of the same individual, such as the liver, general circulation, and peripheral blood mononuclear cells, have slightly different sequences (Afonso et al. 1999; Cabot et al. 2000; Navas et al. 1998). This compartmentalization of quasispecies variants probably results from differences in tissue tropism. The observed compartmentalization of HVR1 sequences (Afonso et al. 1999; Cabot et al. 2000; Navas et al. 1998; Ducoulombier et al. 2004) suggests that HVR1 plays a role in virus-cell recognition and entry. The fact that different IRESs have quantitatively different translational properties in different cell types suggests that this cellular tropism could also be related to specific interactions between the viral genome or viral proteins and cell type-specific intracellular factors (Laporte et al. 2000, 2003). Distinct variants originating from the same viral quasispecies mixture can also have quantitatively different functional properties within the same cell type. Different levels of IFN induction by individual components of vesicular stomatitis virus quasispecies have been reported (Marcus et al. 1998). Using the functional model of HCV NS5A transcriptional activation in the yeast Saccharomyces cerevisiae, we showed that NS5A quasispecies variants isolated from infected patients induce different levels of transcriptional activation depending on the charge of the residues (and possibly minor conformational changes) in the quasispecies variant sequence (Fig. 3) (Pellerin et al. 2004). This suggests that the accumulation of mutations on HCV genomes during replication randomly generates variant proteins with quantitatively different functional properties (Pellerin et al. 2004). The fact that each new variant protein is initially produced in a single infected hepatocyte as a result of a random substitution and then may or may not spread throughout the liver (depending on the replication capacities of the variant virus) points to cellular compartmentalization of virus–host interactions during chronic infection (Pellerin et al. 2004). The same apparently applies to the accumulation of nucleotide mutations in functional RNA structures such as the HCV IRES. We indeed showed that single nucleotide substitutions scattered throughout the IRES (but not evenly distributed in the different domains of the IRES) of variants originating from various patients’ quasispecies were sufficient to significantly reduce translation efficiency relative to a prototype IRES (Paulous et al., personal communication). Similarly, different quasispecies variants exhibited consistently different susceptibility to the stimulatory or inhibitory action of the X region of the HCV 3 noncoding segment on IRES-mediated translation (Paulous et al., personal communication). Overall, these functional features of quasispecies-distributed viruses, which probably apply to any functional

272 J.-M. Pawlotsky

Fig. 3 Transcriptional activation measured in vitro by HCV NS5A variants isolated from a patient before and after various IFN-α-based treatment courses over a 5-year period. The quantitative results are shown in relative units by comparison with a laboratory standard. At each time point, the most frequent quasispecies variant (major variant) is shown in black and is followed from top to bottom by the other variants isolated from the same quasispecies mixture. Each bar represents the mean±SD of three experiments performed in duplicate

Hepatitis C Virus Population Dynamics During Infection

273

protein or genomic region, could play an important role in various aspects of the viral life cycle and pathogenicity.

4 Dynamics of HCV Quasispecies 4.1 HCV Transmission Humans become infected by direct contact of their blood with the blood of an infected individual. Whatever the route of transmission (the most frequent being blood transfusion in the past and intravenous drug use and nosocomial transmission today), the inoculum contains a fraction of the circulating quasispecies of the source patient, i.e., a mixture of distinct quasispecies populations in various relative amounts. Infection in the recipient patient is characterized by selection of certain variants among those present in the inoculum. Such selection may take place during viral uptake by liver cells, replication in hepatocytes, and/or later, when replication interacts with the adaptive immune response. The genetic background of the infected subject and his/her immune responses to viral replication therefore play a major role in the selection events that lead to a steady-state replicative quasispecies in the chronic phase of infection. The recipient’s quasispecies is thus genetically related but consistently different from that of the source patient, as shown in experimentally infected chimpanzees, babies born to HCV-infected mothers, recipients of bone marrow from HCV-infected donors, and individuals infected through occupational or iatrogenic exposure, and also in well-documented cases of blood product-transmitted hepatitis C (Prince et al. 2004; Wyatt et al. 1998; Power et al. 1995; Murakami et al. 2000; Ni et al. 1997; Kudo et al. 1997. The quasispecies distribution of HCV therefore favors the spread of infection, because it ensures that the spectrum of potentially infectious variants includes variants capable of giving rise to a new infection in most infected recipients, whatever their target organs or cells, their genetic background, or the state of their immune system. 4.2 Dynamics of HCV Quasispecies During Acute Infection HCV infection persists in 50%–85% of cases (Consensus Panel 2002). Whether or not the quasispecies distribution of HCV populations plays a role in this persistence is unclear. Persistent infection has been reported to be associated with higher quasispecies genetic complexity (i.e., a larger number of distinct

274

J.-M. Pawlotsky

variants in the quasispecies) during acute infection compared to subjects who spontaneously clear the infection (Ray et al. 2002). According to the escape theory, continuous generation of new variants capable of escaping host defenses at least partly explains how the infection becomes chronic. Acute resolving hepatitis was recently reported to be associated with a relative evolutionary stasis of HVR1 quasispecies (a target of anti-HCV neutralizing responses), whereas progression toward chronicity correlated with significant HVR1 genetic evolution (Farci et al. 2000). It was unclear from this study whether persistent HCV infection was the result of immune escape or diversification and subsequent selection of quasispecies variants, favored by high levels of replication resulting from the failure of the host to control the infection. More recent findings argue against an important role of viral escape in the establishment of persistent infection. We found that chronic infection occurred in bone marrow recipients who received their graft from an HCV-infected donor in the absence of amino acid changes in HVR1 and in regions encoding cytotoxic epitopes (as compared to the donor’s quasispecies), during the several months of aplasia that follow transplantation. In contrast, immune restoration was accompanied by shifts in the virus populations (essentially in HVR1), that likely occurred in response to humoral responses and did not play a role in the establishment of an already established persistent infection (Hézode et al., unpublished data). We also studied a cohort of acutely infected patients infected during a dual-source outbreak in their hemodialysis center (Lavillette et al. 2005). Serial measurements of neutralizing responses during the course of acute hepatitis with an assay based on infectious retroviral pseudoparticles expressing folded HCV envelope glycoproteins showed that strong neutralizing responses were associated in some cases with the selection of escaped HVR1 variants, and that chronic infection occurred in patients whose quasispecies harbored no HVR1 amino acid changes, regardless of whether a neutralizing response was mounted (Lavillette et al. 2005). Qualitative and quantitative differences in cellular immune responses, including both T-helper 1 (Th1) and cytotoxic T cell responses, are now considered to be the main determinants of HCV persistence (Diepolder et al. 1995, 1997; Racanelli and Rehermann 2003; Pawlotsky 2004a). In this model, viral escape would be the consequence rather than the cause of chronic infection established in the presence of an adapted immune response. It is likely, however, that if the quasispecies distribution of HCV populations does not play an important role in establishing chronic infection, it is crucial for its maintenance. Continuous generation of new variants allows the virus to adapt to any changes in the replicative environment and to subtly regulate viral protein and RNA functions. This probably explains why spontaneous resolution of chronic infection is rare or even perhaps nonexistent.

Hepatitis C Virus Population Dynamics During Infection

275

4.3 Dynamics of HCV Quasispecies During Chronic Infection Discordant results have been reported regarding a possible relationship between HCV quasispecies genetic complexity or diversity and the characteristics and severity of HCV-related liver disease, the occurrence of cirrhosis and hepatocellular carcinoma (the main complications of chronic hepatitis C), and the occurrence and severity of HCV-related extrahepatic immunological disorders (Naito et al. 1995; Koizumi et al. 1995; Gonzalez-Peralta et al. 1996; Yuki et al. 1997; Hayashi et al. 1997; Lopez-Labrador et al. 1999; Aiyama et al. 1996). These discrepancies appear to be due mainly to technical limitations and difficulties in establishing causal links. Because chronic viral replication is associated with the accumulation of mutations and genetic diversification of the quasispecies over time (genetic drift), one would expect to find a significant relationship between the duration of infection (itself associated with HCV-related liver disease severity) and the genetic complexity and diversity of HCV quasispecies. However, successive shifts in viral populations may gradually eliminate a large number of variants, which are replaced by newly generated and positively selected variants. In addition, only a minority of variants replicate at sufficiently high levels to circulate in amounts adequate for their isolation from peripheral blood using current techniques, including those based on the generation of numerous molecular clones. More sensitive techniques, and methods for studying intrahepatic HCV quasispecies, might help to resolve this intriguing issue. 4.4 Dynamics of HCV Quasispecies During Antiviral Therapy The current standard of care for chronic hepatitis C is the combination of pegylated IFN-α (an IFN-α molecule linked to a polyethylene glycol molecule, permitting a single weekly administration) and ribavirin, a guanosine analog with a weak antiviral effect in vivo that accelerates infected cell clearance in the presence of IFN-α (Consensus Panel 2002). This combination therapy cures the infection in approximately 50% of treated patients Manns et al. 2001; Fried et al. 2002; Hadziyannis et al. 2004). The remaining patients have ongoing viral replication and progressive liver disease. In patients receiving IFN-α-based therapy, we showed that HCV quasispecies genetic complexity before therapy was an independent predictor of a sustained virological response (Pawlotsky et al. 1998a). Only patients with low HCV complexity, i.e., a small quasispecies sequence repertoire, appear to be able to achieve sustained HCV clearance after therapy. In contrast, patients with a large quasispecies sequence repertoire are very unlikely to have a sustained virological response, probably

276

J.-M. Pawlotsky

because there is a higher chance that one or several pretreatment variants will escape the action of IFN-α effectors and will proliferate (Pawlotsky et al. 1998a). Successful therapy, i.e., permanent viral clearance, is preceded by a drastic fall in the number of quasispecies variant populations (Farci et al. 2002). In contrast, treatment failure is almost always associated with significant changes in the composition of the HCV quasispecies (Pawlotsky et al. 1998b, 1999). These changes are evolutionary and are characterized by successive shifts in the virus populations. They appear to be related to the drastic changes in the replicative environment induced by IFN-α administration and its subsequent withdrawal (Pawlotsky et al. 1998b, 1999). Together, these results point to the induction of strong pressures during treatment, related to IFN-α-induced host responses, that may lead either to viral clearance (no more “fit” variants) or to the selection of fit quasispecies variants (treatment failure). Interestingly, the selected variants are not intrinsically IFN-α-resistant, i.e., do not bear specific sequences or motifs that confer intrinsic resistance to the antiviral action of IFN-α. Retreatment of the same patient with the same therapeutic schedule can occasionally lead to sustained viral clearance (Pawlotsky et al. 1999). A recent study suggested that long-term ribavirin administration could select a Phe-to-Tyr substitution at position 415 in the RdRp of HCV genotype 1a that would confer reduced sensitivity to ribavirin’s antiviral action in vitro (Young et al. 2003). This substitution, however, appears to be a relatively frequent polymorphism of the viral RdRp, and selection of ribavirin-resistant variants is unlikely, given the weak antiviral effect of ribavirin (Dev et al., unpublished data). Ribavirin’s putative mechanisms of action involves accelerated clearance of infected cells during potent inhibition of viral replication by IFN-α. The current preclinical and early clinical development of specific HCV inhibitors targeting the IRES or viral proteins such as NS3 proteinase, NS3 helicase, and NS5B RdRp raises the issue of viral resistance to these agents (Pawlotsky 2004b; Pawlotsky and McHutchison 2004). Selection of resistant variants is foreseeable because the targets of these new drugs exhibit natural variability. We recently isolated minor quasispecies variants bearing mutations within the catalytic site and the zinc-binding site of the HCV NS3 serine protease (Soler et al., unpublished data), potential target sites for new therapies. Variants bearing amino acid substitutions conferring resistance to novel protease and RdRp inhibitors have been selected by such drugs in Huh7 cell lines harboring HCV replicons (Lu et al. 2004; Lin et al. 2004; Nguyen et al. 2003; Migliaccio et al. 2003). Thus, specific resistance to HCV inhibitors in clinical practice is highly likely, because of the HCV quasispecies distribution, the large viral population sizes, and the very rapid kinetics of HCV

Hepatitis C Virus Population Dynamics During Infection

277

replication. Preventive approaches based on potent combinations of drugs with different viral targets will no doubt be needed (Pawlotsky 2004b).

5 Conclusion HCV behaves as an evolving viral quasispecies in its continuously changing environment. The dynamics of quasispecies populations must therefore be taken into account in the HCV life cycle and pathogenicity. Although a qualitatively and quantitatively inappropriate immune response appears to be the key determinant of chronicity, the quasispecies distribution of HCV is probably important in maintaining the chronic infection. This has important consequences for the treatment of what is the only curable chronic human viral infection, because it allows the virus to escape control by antiviral therapy and may delay or impair the clearance of infected cells. For these reasons, tools permitting quasispecies analysis must be used in mechanistic studies of HCV-related disease and therapy. The ultimate research goal is to study HCV quasispecies dynamics and evolution within the individual host cell. Acknowledgements The author would like to thank all of his past and present collaborators involved in the study of HCV genetic variability. The author’s laboratory received grants from Institut National de la Santé et de la Recherche Médicale (INSERM), Agence nationale de Recherche sur le SIDA et les Hépatites Virales (ANRS), Association pour la Recherche sur le Cancer (ARC), Ligue Nationale contre le Cancer, and European Commission 5th and 6th Framework Programmes.

References Afonso AM, Jiang J, Penin F, Tareau C, Samuel D, Petit MA, Bismuth H, Dussaix E, Feray C (1999) Nonrandom distribution of hepatitis C virus quasispecies in plasma and peripheral blood mononuclear cell subsets. J Virol 73:9213–9221 Aiyama T, Yoshioka K, Okumura A, Takayanagi M, Iwata K, Ishikawa T, Kakumu S (1996) Sequence analysis of hypervariable region of hepatitis C virus (HCV) associated with immune complex in patients with chronic HCV infection. J Infect Dis 174:1316–1320 Bartosch B, Verney G, Dreux M, Donot P, Morice Y, Penin F, Pawlotsky JM, Lavillette D, Cosset FL (2005) An interplay between the hypervariable region 1 of the HCV E2 glycoprotein, the scavenger receptor B1 and HDL promotes both enhancement of infection and protection against neutralizing antibodies. J Virol, in press Behrens SE, Tomei L, De Francesco R (1996) Identification and properties of the RNA-dependent RNA polymerase of hepatitis C virus. EMBO J 15:12–22

278

J.-M. Pawlotsky

Blight KJ, Kolykhalov AA, Rice CM (2000) Efficient initiation of HCV RNA replication in cell culture. Science 290:1972–1974 Bressanelli S, Tomei L, Roussel A, Incitti I, Vitale RL, Mathieu M, De Francesco R, Rey FA (1999) Crystal structure of the RNA-dependent RNA polymerase of hepatitis C virus. Proc Natl Acad Sci U S A 96:13034–13039 Bukh J (2004) A critical role for the chimpanzee model in the study of hepatitis C. Hepatology 39:1469–1475 Bukh J, Pietschmann T, Lohmann V, Krieger N, Faulk K, Engle RE, Govindarajan S, Shapiro M, St Claire M, Bartenschlager R (2002) Mutations that permit efficient replication of hepatitis C virus RNA in Huh-7 cells prevent productive replication in chimpanzees. Proc Natl Acad Sci U S A 99:14416–14421 Cabot B, Martell M, Esteban JI, Sauleda S, Otero T, Esteban R, Guardia J, Gomez J (2000) Nucleotide and amino acid complexity of hepatitis C virus quasispecies in serum and liver. J Virol 74:805–811 Castet V, Fournier C, Soulier A, Brillet R, Coste J, Larrey D, Dhumeaux D, Maurel P, Pawlotsky JM (2002) Alpha interferon inhibits hepatitis C virus replication in primary human hepatocytes infected in vitro. J Virol 76:8189–8199 Cheney IW, Naim S, Lai VC, Dempsey S, Bellows D, Walker MP, Shim JH, Horscroft N, Hong Z, Zhong W (2002) Mutations in NS5B polymerase of hepatitis C virus: impacts on in vitro enzymatic activity and viral RNA replication in the subgenomic replicon cell culture. Virology 297:298–306 Cheng JC, Chang MF, Chang SC (1999) Specific interaction between the hepatitis C virus NS5B RNA polymerase and the 3 end of the viral RNA. J Virol 73:7044–7049 Consensus Panel (2002) National Institutes of Health Consensus Development Conference Statement: Management of hepatitis C: 2002–June 10–12, 2002. Hepatology 36:S3–S20 Cuceanu NM, Tuplin A, Simmonds P (2001) Evolutionarily conserved RNA secondary structures in coding and non-coding sequences at the 3 end of the hepatitis G virus/GB-virus C genome. J Gen Virol 82:713–722 Diepolder HM, Zachoval R, Hoffmann RM, Wierenga EA, Santantonio T, Jung MC, Eichenlaub D, Pape GR (1995) Possible mechanism involving T-lymphocyte response to non-structural protein 3 in viral clearance in acute hepatitis C virus infection. Lancet 346:1006–1007 Diepolder HM, Gerlach JT, Zachoval R, Hoffmann RM, Jung MC, Wierenga EA, Scholz S, Santantonio T, Houghton M, Southwood S, Sette A, Pape GR (1997) Immunodominant CD4+ T-cell epitope within nonstructural protein 3 in acute hepatitis C virus infection. J Virol 71:6011–6019 Domingo E (1996) Biological significance of viral quasispecies. Viral Hepatitis Rev 2:247–261 Domingo E (1998) Quasispecies and the implications for virus persistence and escape. Clin Diagn Virol 10:97–101 Duarte EA, Novella IS, Weaver SC, Domingo E, Wain-Hobson S, Clarke DK, Moya A, Elena SF, de la Torre JC, Holland JJ (1994) RNA virus quasispecies: significance for viral disease and epidemiology. Infect Agents Dis 3:201–214 Ducoulombier D, Roque-Afonso AM, Di Liberto G, Penin F, Kara R, Richard Y, Dussaix E, Feray C (2004) Frequent compartmentalization of hepatitis C virus variants in circulating B cells and monocytes. Hepatology 39:817–825

Hepatitis C Virus Population Dynamics During Infection

279

Enomoto N, Kurosaki M, Tanaka Y, Marumo F, Sato C (1994) Fluctuation of hepatitis C virus quasispecies in persistent infection and interferon treatment revealed by single-strand conformation polymorphism analysis. J Gen Virol 75:1361–1369 Enomoto N, Sakuma I, Asahina Y, Kurosaki M, Murakami T, Yamamoto C, Izumi N, Marumo F, Sato C (1995) Comparison of full-length sequences of interferonsensitive and resistant hepatitis C virus 1b. Sensitivity to interferon is conferred by amino acid substitutions in the NS5A region. J Clin Invest 96:224–230 Farci P, Shimoda A, Coiana A, Diaz G, Peddis G, Melpolder JC, Strazzera A, Chien DY, Munoz SJ, Balestrieri A, Purcell RH, Alter HJ (2000) The outcome of acute hepatitis C predicted by the evolution of the viral quasispecies. Science 288:339–344 Farci P, Strazzera R, Alter HJ, Farci S, Degioannis D, Coiana A, Peddis G, Usai F, Serra G, Chessa L, Diaz G, Balestrieri A, Purcell RH (2002) Early changes in hepatitis C viral quasispecies during interferon therapy predict the therapeutic outcome. Proc Natl Acad Sci U S A 99:3081–3086 Forns X, Thimme R, Govindarajan S, Emerson SU, Purcell RH, Chisari FV, Bukh J (2000) Hepatitis C virus lacking the hypervariable region 1 of the second envelope protein is infectious and causes acute resolving or persistent infection in chimpanzees. Proc Natl Acad Sci U S A 97:13318–13323 Fried MW, Shiffman ML, Reddy KR, Smith C, Marinos G, Goncales FL Jr, Haussinger D, Diago M, Carosi G, Dhumeaux D, Craxi A, Lin A, Hoffman J, Yu J (2002) Peginterferon alfa-2a plus ribavirin for chronic hepatitis C virus infection. N Engl J Med 347:975–982 Gonzalez-Peralta RP, Qian K, She JY, Davis GL, Ohno T, Mizokami M, Lau JY (1996) Clinical implications of viral quasispecies heterogeneity in chronic hepatitis C. J Med Virol 49:242–247 Gosert R, Egger D, Lohmann V, Bartenschlager R, Blum HE, Bienz K, Moradpour D (2003) Identification of the hepatitis C virus RNA replication complex in Huh-7 cells harboring subgenomic replicons. J Virol 77:5487–5492 Goutagny N, Fatmi A, De Ledinghen V, Penin F, Couzigou P, Inchauspe G, Bain C (2003) Evidence of viral replication in circulating dendritic cells during hepatitis C virus infection. J Infect Dis 187:1951–1958 Hadziyannis SJ, Sette H Jr, Morgan TR, Balan V, Diago M, Marcellin P, Ramadori G, Bodenheimer H Jr, Bernstein D, Rizzetto M, Zeuzem S, Pockros PJ, Lin A, Ackrill AM (2004) Peginterferon-alpha2a and ribavirin combination therapy in chronic hepatitis C: a randomized study of treatment duration and ribavirin dose. Ann Intern Med 140:346–355 Hayashi J, Kishihara Y, Yamaji K, Furusyo N, Yamamoto T, Pae Y, Etoh Y, Ikematsu H, Kashiwagi S (1997) Hepatitis C viral quasispecies and liver damage in patients with chronic hepatitis C virus infection. Hepatology 25:697–701 Kato N, Ootsuyama Y, Ohkoshi S, Nakazawa T, Sekiya H, Hijikata M, Shimotohno K (1992) Characterization of hypervariable regions in the putative envelope protein of hepatitis C virus. Biochem Biophys Res Commun 189:119–127 Koizumi K, Enomoto N, Kurosaki M, Murakami T, Izumi N, Marumo F, Sato C (1995) Diversity of quasispecies in various disease stages of chronic hepatitis C virus infection and its significance in interferon treatment. Hepatology 22:30–35 Krieger N, Lohmann V, Bartenschlager R (2001) Enhancement of hepatitis C virus RNA replication by cell culture-adaptive mutations. J Virol 75:4614–4624

280

J.-M. Pawlotsky

Kudo T, Yanase Y, Ohshiro M, Yamamoto M, Morita M, Shibata M, Morishima T (1997) Analysis of mother-to-infant transmission of hepatitis C virus: quasispecies nature and buoyant densities of maternal virus populations. J Med Virol 51:225–230 Laporte J, Malet I, Andrieu T, Thibault V, Toulme JJ, Wychowski C, Pawlotsky JM, Huraux JM, Agut H, Cahour A (2000) Comparative analysis of translation efficiencies of hepatitis C virus 5 untranslated regions among intraindividual quasispecies present in chronic infection: opposite behaviors depending on cell type. J Virol 74:10827–10833 Laporte J, Bain C, Maurel P, Inchauspe G, Agut H, Cahour A (2003) Differential distribution and internal translation efficiency of hepatitis C virus quasispecies present in dendritic and liver cells. Blood 101:52–57 Lavillette D, Morice Y, Germanidis G, Donot P, Soulier A, Pagkalos E, Sakellariou G, Intrator L, Bartosch B, Pawlotsky JM, Cosset FL (2005) Human serum facilitates HCV infection and neutralizing responses inversely correlate with viral replication kinetics at the acute phase of hepatitis C. J Virol 79:6023–6034 Lerat H, Rumin S, Habersetzer F, Berby F, Trabaud MA, Trepo C, Inchauspe G (1998) In vivo tropism of hepatitis C virus genomic sequences in hematopoietic cells: influence of viral load, viral genotype, and cell phenotype. Blood 91:3841–3849 Lesburg CA, Cable MB, Ferrari E, Hong Z, Mannarino AF, Weber PC (1999) Crystal structure of the RNA-dependent RNA polymerase from hepatitis C virus reveals a fully encircled active site. Nat Struct Biol 6:937–943 Lin C, Lin K, Luong YP, Rao BG, Wei YY, Brennan DL, Fulghum JR, Hsiao HM, Ma S, Maxwell JP, Cottrell KM, Perni RB, Gates CA, Kwong AD (2004) In vitro resistance studies of hepatitis C virus serine protease inhibitors, VX-950 and BILN 2061: structural analysis indicates different resistance mechanisms. J Biol Chem 279:17508–17514 Lohmann V, Korner F, Herian U, Bartenschlager R (1997) Biochemical properties of hepatitis C virus NS5B RNA-dependent RNA polymerase and identification of amino acid sequence motifs essential for enzymatic activity. J Virol 71:8416–8428 Lohmann V, Korner F, Koch J, Herian U, Theilmann L, Bartenschlager R (1999) Replication of subgenomic hepatitis C virus RNAs in a hepatoma cell line. Science 285:110–113 Lohmann V, Korner F, Dobierzewska A, Bartenschlager R (2001) Mutations in hepatitis C virus RNAs conferring cell culture adaptation. J Virol 75:1437–1449 Lopez-Labrador FX, Ampurdanes S, Gimenez-Barcons M, Guilera M, Costa J, Jimenez de Anta MT, Sanchez-Tapias JM, Rodes J, Saiz JC (1999) Relationship of the genomic complexity of hepatitis C virus with liver disease severity and response to interferon in patients with chronic HCV genotype 1b infection. Hepatology 29:897–903 Love RA, Parge HE, Yu X, Hickey MJ, Diehl W, Gao J, Wriggers H, Ekker A, Wang L, Thomson JA, Dragovich PS, Fuhrman SA (2003) Crystallographic identification of a noncompetitive inhibitor binding site on the hepatitis C virus NS5B RNA polymerase enzyme. J Virol 77:7575–7581 Lu L, Pilot-Matias TJ, Stewart KD, Randolph JT, Pithawalla R, He W, Huang PP, Klein LL, Mo H, Molla A (2004) Mutations conferring resistance to a potent hepatitis C virus serine protease inhibitor in vitro. Antimicrob Agents Chemother 48:2260–2266

Hepatitis C Virus Population Dynamics During Infection

281

Manns MP, McHutchison JG, Gordon SC, Rustgi VK, Shiffman M, Reindollar R, Goodman ZD, Koury K, Ling M, Albrecht JK (2001) Peginterferon alfa-2b plus ribavirin compared with interferon alfa-2b plus ribavirin for initial treatment of chronic hepatitis C: a randomised trial. Lancet 358:958–965 Marcus PI, Rodriguez LL, Sekellick MJ (1998) Interferon induction as a quasispecies marker of vesicular stomatitis virus populations. J Virol 72:542–549 Martell M, Esteban JI, Quer J, Genesca J, Weiner A, Esteban R, Guardia J, Gomez J (1992) Hepatitis C virus (HCV) circulates as a population of different but closely related genomes: quasispecies nature of HCV genome distribution. J Virol 66:3225–3229 Migliaccio G, Tomassini JE, Carroll SS, Tomei L, Altamura S, Bhat B, Bartholomew L, Bosserman MR, Ceccacci A, Colwell LF, Cortese R, De Francesco R, Eldrup AB, Getty KL, Hou XS, LaFemina RL, Ludmerer SW, MacCoss M, McMasters DR, Stahlhut MW, Olsen DB, Hazuda DJ, Flores OA (2003) Characterization of resistance to non-obligate chain-terminating ribonucleoside analogs that inhibit hepatitis C virus replication in vitro. J Biol Chem 278:49164–49170 Moradpour D, Gosert R, Egger D, Penin F, Blum HE, Bienz K (2003) Membrane association of hepatitis C virus nonstructural proteins and identification of the membrane alteration that harbors the viral replication complex. Antiviral Res 60:103–109 Murakami J, Okamoto M, Miyata H, Nagata I, Shiraki K, Hino S (2000) Evolution in the hypervariable region of hepatitis C virus in infants after vertical transmission. Pediatr Res 48:450–456 Nagasaka A, Hige S, Tsunematsu I, Yoshida J, Sasaki Y, Matsushima T, Asaka M (1996) Changes in hepatitis C virus quasispecies and density populations in patients before and after interferon therapy. J Med Virol 50:214–220 Naito M, Hayashi N, Moribe T, Hagiwara H, Mita E, Kanazawa Y, Kasahara A, Fusamoto H, Kamada T (1995) Hepatitis C viral quasispecies in hepatitis C virus carriers with normal liver enzymes and patients with type C chronic liver disease. Hepatology 22:407–412 Navas S, Martin J, Quiroga JA, Castillo I, Carreno V (1998) Genetic diversity and tissue compartmentalization of the hepatitis C virus genome in blood mononuclear cells, liver, and serum from chronic hepatitis C patients. J Virol 72:1640–1646 Negro F, Levrero M (1998) Does the hepatitis C virus replicate in cells of the hematopoietic lineage? Hepatology 28:261–264 Neumann AU, Lam NP, Dahari H, Gretch DR, Wiley TE, Layden TJ, Perelson AS (1998) Hepatitis C viral dynamics in vivo and the antiviral efficacy of interferon-alpha therapy. Science 282:103–107 Nguyen TT, Gates AT, Gutshall LL, Johnston VK, Gu B, Duffy KJ, Sarisky RT (2003) Resistance profile of a hepatitis C virus RNA-dependent RNA polymerase benzothiadiazine inhibitor. Antimicrob Agents Chemother 47:3525–3530 Ni YH, Chang MH, Chen PJ, Lin HH, Hsu HY (1997) Evolution of hepatitis C virus quasispecies in mothers and infants infected through mother-to-infant transmission. J Hepatol 26:967–974 Nousbaum J, Polyak SJ, Ray SC, Sullivan DG, Larson AM, Carithers RL Jr, Gretch DR (2000) Prospective characterization of full-length hepatitis C virus NS5A quasispecies during induction and combination antiviral therapy. J Virol 74:9028–9038

282

J.-M. Pawlotsky

Pawlotsky JM (2004a) Hepatitis C: it’s a long way to new therapy, it’s a long way to go... Gastroenterology 127:1629–1632 Pawlotsky JM (2004b) Pathophysiology of hepatitis C virus infection and related liver disease. Trends Microbiol 12:96–102 Pawlotsky JM, McHutchison JG (2004) Hepatitis C. Development of new drugs and clinical trials: promises and pitfalls. Summary of an AASLD hepatitis single topic conference, Chicago, IL, February 27–March 1, 2004. Hepatology 39:554–567 Pawlotsky JM, Pellerin M, Bouvier M, Roudot-Thoraval F, Germanidis G, Bastie A, Darthuy F, Remire J, Soussy CJ, Dhumeaux D (1998a) Genetic complexity of the hypervariable region 1 (HVR1) of hepatitis C virus (HCV): influence on the characteristics of the infection and responses to interferon alfa therapy in patients with chronic hepatitis C. J Med Virol 54:256–264 Pawlotsky JM, Germanidis G, Neumann AU, Pellerin M, Frainais PO, Dhumeaux D (1998b) Interferon resistance of hepatitis C virus genotype 1b: relationship to nonstructural 5A gene quasispecies mutations. J Virol 72:2795–2805 Pawlotsky JM, Germanidis G, Frainais PO, Bouvier M, Soulier A, Pellerin M, Dhumeaux D (1999) Evolution of the hepatitis C virus second envelope protein hypervariable region in chronically infected patients receiving alpha interferon therapy. J Virol 73:6490–6499 Pawlotsky JM, Tsakiris L, Roudot-Thoraval F, Pellet C, Stuyver L, Duval J, Dhumeaux D (1995) Relationship between hepatitis C virus genotypes and sources of infection in patients with chronic hepatitis C. J Infect Dis 171:1607–1610 Pellerin M, Lopez-Aguirre Y, Penin F, Dhumeaux D, Pawlotsky JM (2004) Hepatitis C virus quasispecies variability modulates nonstructural protein 5A transcriptional activation, pointing to cellular compartmentalization of virus–host interactions. J Virol 78:4617–4627 Penin F, Combet C, Germanidis G, Frainais PO, Deleage G, Pawlotsky JM (2001) Conservation of the conformation and positive charges of hepatitis C virus E2 envelope glycoprotein hypervariable region 1 points to a role in cell attachment. J Virol 75:5703–5710 Penin F, Dubuisson J, Rey FA, Moradpour D, Pawlotsky JM (2004) Structural biology of hepatitis C virus. Hepatology 39:5–19 Pietschmann T, Lohmann V, Rutter G, Kurpanek K, Bartenschlager R (2001) Characterization of cell lines carrying self-replicating hepatitis C virus RNAs. J Virol 75:1252–1264 Polyak SJ, McArdle S, Liu SL, Sullivan DG, Chung M, Hofgartner WT, Carithers RL Jr, McMahon BJ, Mullins JI, Corey L, Gretch DR (1998) Evolution of hepatitis C virus quasispecies in hypervariable region 1 and the putative interferon sensitivity-determining region during interferon therapy and natural infection. J Virol 72:4288–4296 Power JP, Lawlor E, Davidson F, Holmes EC, Yap PL, Simmonds P (1995) Molecular epidemiology of an outbreak of infection with hepatitis C virus in recipients of anti-D immunoglobulin. Lancet 345:1211–1213 Prince AM, Pawlotsky JM, Soulier A, Tobler L, Brotman B, Pfahler W, Lee DH, Li L, Shata MT (2004) Hepatitis C virus replication kinetics in chimpanzees with selflimited and chronic infections. J Viral Hepat 11:236–242

Hepatitis C Virus Population Dynamics During Infection

283

Racanelli V, Rehermann B (2003) Hepatitis C virus infection: when silence is deception. Trends Immunol 24:456–464 Ray SC, Wang YM, Laeyendecker O, Ticehurst JR, Villano SA, Thomas DL (1999) Acute hepatitis C virus structural gene sequences as predictors of persistent viremia: hypervariable region 1 as a decoy. J Virol 73:2938–2946 Ray SC, Mao Q, Lanford RE, Bassett S, Laeyendecker O, Wang YM, Thomas DL (2000) Hypervariable region 1 sequence stability during hepatitis C virus replication in chimpanzees. J Virol 74:3058–3066 Robertson B, Myers G, Howard C, Brettin T, Bukh J, Gaschen B, Gojobori T, Maertens G, Mizokami M, Nainan O, Netesov S, Nishioka K, Shin IT, Simmonds P, Smith D, Stuyver L, Weiner A (1998) Classification, nomenclature, and database development for hepatitis C virus (HCV) and related viruses: proposals for standardization. Arch Virol 143:2493–2503 Saito S, Kato N, Hijikata M, Gunji T, Itabashi M, Kondo M, Tanaka K, Shimotohno K (1996) Comparison of hypervariable regions (HVR1 and HVR2) in positiveand negative-stranded hepatitis C virus RNA in cancerous and non-cancerous liver tissue, peripheral blood mononuclear cells and serum from a patient with hepatocellular carcinoma. Int J Cancer 67:199–203 Shimizu YK, Yoshikura H (1994) Multicycle infection of hepatitis C virus in cell culture and inhibition by alpha and beta interferons. J Virol 68:8406–8408 Shimizu YK, Iwamoto A, Hijikata M, Purcell RH, Yoshikura H (1992) Evidence for in vitro replication of hepatitis C virus genome in a human T-cell line. Proc Natl Acad Sci U S A 89:5477–5481 Shindo M, Hamada K, Koya S, Arai K, Sokawa Y, Okuno T (1996) The clinical significance of changes in genetic heterogeneity of the hypervariable region 1 in chronic hepatitis C with interferon therapy. Hepatology 24:1018–1023 Simmonds P (1995) Variability of hepatitis C virus. Hepatology 21:570–583 Simmonds P (1998) Variability of the hepatitis C virus genome. Curr Stud Hematol Blood Transfus 62:38–63 Simmonds P (2001) The origin and evolution of hepatitis viruses in humans. J Gen Virol 82:693–712 Smith DB, Mellor J, Jarvis LM, Davidson F, Kolberg J, Urdea M, Yap PL, Simmonds P (1995) Variation of the hepatitis C virus 5 non-coding region: implications for secondary structure, virus detection and typing. J Gen Virol 76:1749–1761 Soler M, Pellerin M, Malnou CE, Dhumeaux D, Kean KM, Pawlotsky JM (2002) Quasispecies heterogeneity and constraints on the evolution of the 5 non coding region of hepatitis C virus (HCV). Relationship with HCV resistance to interferon-alfa therapy. Virology 298:160–173 Soler M, McHutchison JG, Kwoh TJ, Dorr FA, Pawlotsky JM (2004) Virological effects of ISIS 14803, an antisense oligonucleotide inhibitor of hepatitis C virus (HCV) internal ribosome entry site (IRES), on HCV IRES in chronic hepatitis C patients and examination into the potential role of primary and secondary HCV resistance in the outcome of treatment. Antiviral Ther 9:953–968 Sugitani M, Shikata T (1998) Comparison of amino acid sequences in hypervariable region-1 of hepatitis C virus clones between human inocula and the infected chimpanzee sera. Virus Res 56:177–182

284

J.-M. Pawlotsky

Tokita H, Okamoto H, Tsuda F, Song P, Nakata S, Chosa T, Iizuka H, Mishiro S, Miyakawa Y, Mayumi M (1994) Hepatitis C virus variants from Vietnam are classifiable into the seventh, eighth, and ninth major genetic groups. Proc Natl Acad Sci U S A 91:11022–11026 Tokita H, Okamoto H, Luengrojanakul P, Vareesangthip K, Chainuvati T, Iizuka H, Tsuda F, Miyakawa Y, Mayumi M (1995) Hepatitis C virus variants from Thailand classifiable into five novel genotypes in the sixth (6b), seventh (7c, 7d) and ninth (9b, 9c) major genetic groups. J Gen Virol 76:2329–2335 Tokita H, Okamoto H, Iizuka H, Kishimoto J, Tsuda F, Lesmana LA, Miyakawa Y, Mayumi M (1996) Hepatitis C virus variants from Jakarta, Indonesia classifiable into novel genotypes in the second (2e and 2f), tenth (10a) and eleventh (11a) genetic groups. J Gen Virol 77:293–301 Wang M, Ng KK, Cherney MM, Chan L, Yannopoulos CG, Bedard J, Morin N, NguyenBa N, Alaoui-Ismaili MH, Bethell RC, James MN (2003) Non-nucleoside analogue inhibitors bind to an allosteric site on HCV NS5B polymerase. Crystal structures and mechanism of inhibition. J Biol Chem 278:9489–9495 Weiner AJ, Brauer MJ, Rosenblatt J, Richman KH, Tung J, Crawford K, Bonino F, Saracco G, Choo QL, Houghton M (1991) Variable and hypervariable domains are found in the regions of HCV corresponding to the flavivirus envelope and NS1 proteins and the pestivirus envelope glycoproteins. Virology 180:842–848 Wyatt CA, Andrus L, Brotman B, Huang F, Lee DH, Prince AM (1998) Immunity in chimpanzees chronically infected with hepatitis C virus: role of minor quasispecies in reinfection. J Virol 72:1725–1730 Yeh BI, Han KH, Oh SH, Kim HS, Hong SH, Kim YS (1996) Nucleotide sequence variation in the hypervariable region of the hepatitis C virus in the sera of chronic hepatitis C patients undergoing controlled interferon-alpha therapy. J Med Virol 49:95–102 Young KC, Lindsay KL, Lee KJ, Liu WC, He JW, Milstein SL, Lai MM (2003) Identification of a ribavirin-resistant NS5B mutation of hepatitis C virus during ribavirin monotherapy. Hepatology 38:869–878 Yuki N, Hayashi N, Moribe T, Matsushita Y, Tabata T, Inoue T, Kanazawa Y, Ohkawa K, Kasahara A, Fusamoto H, Kamada T (1997) Relation of disease activity during chronic hepatitis C infection to complexity of hypervariable region 1 quasispecies. Hepatology 25:439–444

CTMI (2006) 299:285–314 c Springer-Verlag Berlin Heidelberg 2006


Evolutionary Influences in Arboviral Disease S. C. Weaver (u) Center for Biodefense and Emerging Infectious Diseases and Department of Pathology, University of Texas Medical Branch, Galveston, TX 77555-0609, USA [email protected]

1 Evolution and Systematics of the Arboviruses . 1.1 RNA Viruses as Arboviruses . . . . . . . . . . . . . 1.2 The Alphaviruses . . . . . . . . . . . . . . . . . . . . . 1.2.1 Evolution of the Alphaviruses . . . . . . . . . . . . 1.2.1.1 Relationships Within the Genus . . . . . . . . . . . 1.2.1.2 Patterns of Host Utilization . . . . . . . . . . . . . . 1.2.1.3 Rates of Evolution . . . . . . . . . . . . . . . . . . . . . 1.3 The Flaviviruses . . . . . . . . . . . . . . . . . . . . . . 1.3.1 Relationships Within the Flavivirus Genus . . . 1.4 Evolution of the Flaviviruses . . . . . . . . . . . . . 1.4.1 Patterns of Host Utilization . . . . . . . . . . . . . . 1.4.2 Rates of Evolution . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

286 286 287 289 289 290 292 293 293 293 293 296

2

Recombination and Reassortment . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297

3 3.1 3.2 3.3

Emergence Mechanisms of Arboviral Diseases Direct Spillover . . . . . . . . . . . . . . . . . . . . . . Secondary Amplification . . . . . . . . . . . . . . . . Humans as Arboviral Amplification Hosts . . .

4 4.1

Experimental Approaches to the Study of Arbovirus Evolution . Effect of the Alternating Host Cycle on Arbovirus Genetic and Phenotypic Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . Adaptation of RNA Viruses to New Hosts and Host Cells . . . . . Constraints of the Arbovirus Transmission Cycle on Adaptation to New Hosts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

4.2 4.3

5 5.1 5.2

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

298 298 298 300

. . . . . . . 302 . . . . . . . 302 . . . . . . . 303 . . . . . . . 304

Future Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305 Genetic and Phenotypic Stability of Arboviruses . . . . . . . . . . . . . . . . . . 305 Host Switching by Arboviruses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308

Abstract Arthropod-borne viruses (arboviruses) generally require horizontal transmission by arthropod vectors among vertebrate hosts for their natural maintenance. This requirement for alternate replication in disparate hosts places unusual evolutionary constraints on these viruses, which have probably limited the evolution of

286

S. C. Weaver

arboviruses to only a few families of RNA viruses (Togaviridae, Flaviviridae, Bunyaviridae, Rhabdoviridae, Reoviridae, and Orthomyxoviridae) and a single DNA virus. Phylogenetic studies have suggested the dominance of purifying selection in the evolution of arboviruses, consistent with constraints imposed by differing replication environments and requirements in arthropod and vertebrate hosts. Molecular genetic studies of alphaviruses and flaviviruses have also identified several mutations that effect differentially the replication in vertebrate and mosquito cells, consistent with the view that arboviruses must adopt compromise fitness characteristics for each host. More recently, evidence of positive selection has also been obtained from these studies. However, experimental model systems employing arthropod and vertebrate cell cultures have yielded conflicting conclusions on the effect of alternating host infections, with host specialization inconsistently resulting in fitness gains or losses in the bypassed host cells. Further studies using in vivo systems to study experimental arbovirus evolution are critical to understanding and predicting disease emergence, which often results from virus adaptation to new vectors or amplification hosts. Reverse genetic technologies that are now available for most arbovirus groups should be exploited to test assumptions and hypotheses derived from retrospective phylogenetic approaches.

1 Evolution and Systematics of the Arboviruses Arthropod-borne viruses (arboviruses) comprise a taxonomically diverse group with similar ecology and maintenance mechanisms. Although several of these viruses can be maintained in their arthropod hosts alone via transovarial transmission and some generate persistent infection of vertebrates, most if not all of these viruses require occasional or frequent horizontal transmission among vertebrate hosts by biological vectors, in which replication must occur. Therefore, arboviruses must acquire and retain fitness for replication in disparate vertebrate and invertebrate hosts. This fundamental difference with respect to most animal RNA and nearly all animal DNA viruses, which tend to specialize, on certain taxa of vertebrates, arthropods or other animals, presents unique evolutionary challenges along with many advantages of vector transmission such as high mobility and the lack of a need to be shed into bodily secretions. These challenges have probably greatly influenced the evolution of vector transmission by limiting it to only a few families of RNA viruses and a single taxon of DNA viruses. 1.1 RNA Viruses as Arboviruses The vast majority of arboviruses are classified into only a few families and genera of RNA viruses: the alphaviruses (one of two genera) in the family Togaviridae; the flaviviruses (one of three genera) in the family Flaviviridae;

Evolutionary Influences in Arboviral Disease

287

the bunyaviruses, nairoviruses, and phleboviruses (three of five genera) in the family Bunyaviridae; the orbiviruses (one of nine genera) in the family Reoviridae; the vesiculoviruses (one of six genera) in the family Rhabdoviridae; and the thogotoviruses (one of four genera) in the family Orthomyxoviridae. The only DNA arbovirus known is African swine fever virus (Asfarviridae: Asfarvirus) (Karabatsos 1985; Calisher and Karabatsos 1988; van Regenmortel et al. 2000); this lack of DNA arboviruses suggests that the greater genetic plasticity and higher mutation rates exhibited by RNA viruses (Holland and Domingo 1998) facilitate their ability to replicate alternately in disparate vertebrate and invertebrate hosts. Arboviruses cause a wide range of diseases in humans and domestic animals. However, there is relatively little evidence of severe disease in reservoir hosts; most of the apparent disease caused by arboviruses involves humans, equines and other ungulates, and other domestic animals representing deadend infections that do not exert long-term evolutionary pressures. The lack of apparent disease in many reservoir hosts may reflect selection for resistance by populations exposed for long time periods to infection and/or selection for attenuation of arboviruses in these species. These competing hypotheses are difficult to evaluate experimentally aside from the use of model cell culture systems (see below). However, the recent introduction of West Nile virus into North America provides a unique opportunity to observe these hypothetical evolutionary pressures on an arbovirus in vivo, in nature (Weaver and Barrett 2004). Retrospective evolutionary studies of two of the major groups of arboviruses, the alphaviruses and flaviviruses, and the diseases they cause are briefly reviewed below. 1.2 The Alphaviruses The Togaviridae is the only virus family comprised almost exclusively of arboviruses. Aside from Rubella virus (the sole member of the genus Rubivirus) and two alphaviruses with no known vector (southern elephant seal virus and salmon pancreas disease virus), all togaviruses are mosquito-borne viruses in the genus Alphavirus (Weaver et al. 2000). In humans and domestic animals, alphaviruses cause a spectrum of disease ranging from inapparent to highly pathogenic syndromes including arthralgia accompanied by rash, and severe, often fatal encephalitis (Griffin 2001; Tsai et al. 2002). The most important causes of severe morbidity and mortality include the New World members Venezuelan (VEEV), eastern (EEEV) and western equine encephalitis virus (WEEV), etiologic agents of encephalitis in humans and equines, and Old World alphaviruses that cause a severe but self-limiting arthralgia and rash

288

S. C. Weaver

syndrome, including Ross River, Chikungunya, and o’nyong-nyong viruses. Epidemiological studies suggest that the latter viruses can use humans as amplification hosts during some outbreaks; otherwise, the alphaviruses generally use birds or small mammals as reservoir and amplification hosts, with humans and domestic animals representing dead-end infections. However, a notable exception is VEEV, which exploits equines as highly efficient amplification hosts, resulting in explosive and widespread epidemics

Fig. 1 Phylogenetic tree of all species and major lineages of alphaviruses derived from E1 envelope glycoprotein sequences. Subtypes are written in parentheses after virus names. Reservoir hosts and vectors are listed after viruses. New World viruses are printed in bold and underlined. Open circle indicates virus introductions from the Old to New Worlds, and closed circle indicates introductions from the New to Old Worlds; hashed circles indicate introductions with ambiguous directionality. Dashed line represents the recombination event that led to the ancestor of WEE, Highlands J and Ft. Morgan viruses. The tree was drawn using the neighbor joining program with the HKY distance formula using PAUP 4.0. Similar topologies were produced using maximum parsimony and Bayesian methods

Evolutionary Influences in Arboviral Disease

289

Fig. 2 Phylogenetic trees of VEE complex alphaviruses derived from structural polyprotein amino acid sequences using the neighbor joining program, and of their mosquito vectors derived from ribosomal ITS-2 DNA sequences (Navarro and Weaver 2004). Discordance in the topologies indicates a lack of co-speciation of the viruses with their enzootic mosquito vectors

(Weaver et al. 2004b). Outbreaks of VEE appear to involve adaptation of equine-avirulent, sylvatic enzootic strains for equine replication, involving small numbers of envelope glycoprotein gene mutations (Figs. 1, 2). Adaptation to new mosquito vectors, also involving envelope glycoprotein amino acid changes, appears to mediate some but not all outbreaks as well (Brault et al. 2002b, 2004; Ortiz and Weaver 2004, Brault et al. 2004). 1.2.1 Evolution of the Alphaviruses 1.2.1.1 Relationships Within the Genus Comprehensive phylogenetic analyses of the genus alphavirus have been used to elucidate patterns of evolution and epidemiology (Powers et al. 2001).

290

S. C. Weaver

Although Rubella virus is clearly closely related to the alphaviruses based on genome organization and functions of the major proteins, sequence divergence is extensive and cannot be demonstrated statistically aside from conserved motifs in the nonstructural proteins. Sequence analyses also have demonstrated homology among the nonstructural proteins of alphaviruses and those of several plant virus groups with dissimilar genome organizations, indicating a process of modular evolution leading to these groups (Strauss and Strauss 1994). The alphaviruses, with no known vectors, are the most divergent members of the genus and, although rooted trees are inappropriate due to the lack of a closely related outgroup for the alphaviruses, probably represent a basal clade (Fig. 1). The distribution of these fish and seal viruses in both the Old and New Worlds provides no information on ancestral distributions to estimate the geographic origin of the mosquito-borne members of the genus. Serocomplexes of alphaviruses first defined by antigenic cross-reactivity (Calisher and Karabatsos 1988) generally correspond to clades defined by phylogenetic studies (Fig. 1). These include the Old World Semliki Forest and New World VEE and EEE complexes. The WEE complex represents a geographically and pathologically diverse group including the new World WEEV, Ft. Morgan (FMV), and highlands J viruses (HJV) some of which cause equine and/or human encephalitis, the Sindbis-like viruses including Whataroa and Sindbis (SINV) from the New World, and Aura from the New World, which can cause a human arthralgia syndrome. The dichotomy in disease syndromes and distribution of the WEE complex viruses is most easily explained by an ancient recombination event between a SINV-like virus and the ancestor of the WEE-HJV-FMV group, followed by introduction of a descendant of the SINV ancestor into the Old World (Fig. 1; see below) (Hahn et al. 1988; Weaver et al. 1997). 1.2.1.2 Patterns of Host Utilization Examination of host relationships in the alphavirus tree (Fig. 1) also suggests patterns of host switching and a lack of co-speciation of the viruses with their hosts and vectors. Vector species and genera vary widely within virus clades, serocomplexes and even species, with only a few exceptions: (a)the VEE complex viruses probably use exclusively members of the Spissipes section (a group of only 23 species) within the subgenus Culex (Melanoconion) as enzootic vectors. However, some if not all relationships among the vector species (Navarro and Weaver 2004) are discordant with virus relationships, indicating a lack of co-speciation (Fig. 2); (b) with the exception of the North

Evolutionary Influences in Arboviral Disease

291

American strains of EEEV, all of the EEEV and VEE complex lineages also appear to use these Culex (Melanoconion) vectors, suggesting that either genetic or ecological constraints limit vector switching to closely related mosquitoes. The almost complete lack of alphavirus vectors outside of the mosquito family (Culicidae), including lack of evidence for an important role of ticks, which are vectors of several other arbovirus taxa, suggests similar constraints for the arboviral alphaviruses as a whole. Alphaviruses use a wide variety of mammalian and avian vertebrate hosts for their maintenance reservoir hosts (Fig. 1). In contrast to their relationships with vectors, where a given alphavirus typically uses one or a few mosquito species as primary vectors, individual alphavirus species and lineages may use several different vertebrates simultaneously; for example, EEEV infects a variety of passeriform birds in enzootic swamp habitats of North America, many of which generate viremia sufficient for horizontal transmission by the highly susceptible and ornithophilic enzootic vector, Culiseta melanura (Scott and Weaver 1989). Although an important role in maintenance has not been established for many groups, alphaviruses like EEEV infect an extremely diverse group of vertebrates, including birds, mammals, amphibians, and reptiles. The wider vertebrate host range of the alphaviruses compared to the range of their hematophagous arthropod vectors suggests greater potential for reservoir than vector host switching during the course of evolution and disease emergence. Studies described below have begun to test this and related hypotheses experimentally. The uniformity in vector taxa (mosquitoes) used by the alphaviruses is also observed in other arbovirus taxa, and contrasts with the wide range of vertebrate hosts that serve as reservoirs and amplification hosts, typically including both birds and mammals. This pattern suggests that adaptation to different vectors, such as other biting flies or ticks, is genetically difficult, and/or that arboviruses have evolved as generalists for their vertebrate hosts but specialists with respect to their vectors. However, the specificity for vectors is often manifested only at the level of midgut infection, and most alphaviruses replicate in most mosquitoes following intrathoracic inoculation, which is analogous to infection of a vertebrate via a vector bite or needle. Better understanding of the interactions between host factors and arboviruses during infection and replication is needed to understand differences in vertebrate and vector host specificity. The taxa used as reservoir hosts appear to strongly influence the genetic structure of the alphaviral populations (Mackenzie et al. 1995; Weaver 1995). Those viruses that use avian hosts, such as EEEV (Brault et al. 1999), HJV (Cilnis et al. 1996), WEEV (Weaver et al. 1997), Barmah Forest virus (Poidinger et al. 1997), and SINV (Norder et al. 1996; Sammels et al. 1999) appear to

292

S. C. Weaver

evolve within a small numbers of broadly distributed lineages, presumably reflecting efficient dispersal by birds. In the case of SINV, lineage replacement may occur in Australia (Mackenzie et al. 1995). In contrast, the alphaviruses that use mammals with limited dispersal, such as Ross River (Sammels et al. 1995), chikungunya (Powers et al. 2000), and most of the VEE complex viruses (Powers et al. 2001), evolve within a greater number of geographically limited lineages, reflecting very limited dispersal ability. Presumably, efficient dispersal acts to constrain lineage diversity by resulting in frequent mixing of populations and elimination of less fit populations via competition. 1.2.1.3 Rates of Evolution Studies on the rates of sequence evolution in alphaviruses have yielded estimates that generally fall below those of single host taxon, non-arthropodborne RNA viruses (Weaver et al. 1992). These estimates as well as analyses of sequence change obtained from phylogenies, which emphasize the preponderance of synonymous substitutions, suggest that strong purifying selection dominates alphavirus evolution. Even the 26S subgenomic promoter sequence that is conserved but includes a few differences among alphaviruses shows no evidence of adaptive substitutions, and most of the promoter sequences are interchangeable between SINV and other species (Hertz and Huang 1992). Studies of genetic diversity within alphavirus populations indicate a quasispecies distribution of genetic variants similar to that exhibited by single-host RNA viruses, suggesting that mutation frequencies are not the explanation for the genetic stability observed (Weaver et al. 1993). The requirement for alternate replication in disparate hosts is a possible explanation for this phenomenon and for the slow rate of sequence change (see below). The time scales of evolution for the alphavirus genus as well as other arbovirus taxa have also been estimated using various genetic formulas and sequence evolution models coupled with phylogenetics. These analyses have typically yielded estimates on the order of thousands of years for divergence of arbovirus groups from common ancestors. However, the strong evidence of co-speciation of certain rodent-borne viruses with their reservoir hosts, including some in the family Bunyaviridae, which includes many arboviruses, indicate a time scale of tens of millions of years for these RNA virus groups (Morzunov et al. 1998). The bunyaviruses and arenaviruses exhibit genetic diversity comparable to or in some cases greater than those exhibited by arboviruses. Therefore, reconciliation of time scales derived from co-speciation evidence coupled with the fossil record, vs phylogenetic techniques, which differ by several orders of magnitude for divergence of these groups, is

Evolutionary Influences in Arboviral Disease

293

problematic. One possible explanation is that the phylogenetic methods are not yet capable of accurately compensating for multiple substitutions of nucleotides and variation in the rates of substitution among nucleotide sites (Holmes 2003). Assumptions that most synonymous nucleotide sites in RNA viral genomes are subject to little or no selection, and therefore exhibit little variability in substitution rates, may be invalid due to genome-scale, ordered RNA structures that can now be identified using improved computational tools (Simmonds et al. 2004). These structures need to be examined experimentally to determine their influence on RNA virus evolution. 1.3 The Flaviviruses The genus Flavivirus comprises a highly diverse group of both vector-borne and non-vector-borne viruses distributed nearly worldwide (Gould et al. 2003). Included in this taxon are important causes of human encephalitis such as Japanese (JEV) and tick-borne encephalitis viruses (TBEV), yellow fever virus (YFV), which is among the most virulent human pathogens and remains an important cause of mortality in Africa and South America, and dengue viruses (DENV), the leading arboviral causes of morbidity and mortality. In addition to their overall greater diversity compared to the alphaviruses, the flaviviruses exhibit a wider range of transmission cycles and vectors; some flaviviruses have no known vector, and large monophyletic groups use either mosquitoes or ticks as vectors (Fig. 3). 1.3.1 Relationships Within the Flavivirus Genus The flaviviruses comprise one genus in the Family Flaviviridae. The other genera, Pestivirus and Hepacivirus, are non-vector-borne animal viruses. Within the genus Flaviviruses, four major clades of viruses include non-vector-borne, tick-borne, and two mosquito-borne groups. Like the alphaviruses, the flaviviruses are distributed nearly worldwide except for in Antarctica. They also infect a wide range of vertebrates and arthropod vectors, including ticks, which are not considered important vectors of alphaviruses. 1.4 Evolution of the Flaviviruses 1.4.1 Patterns of Host Utilization Phylogenetic studies have identified interesting differences in evolutionary patterns among the four flavivirus groups mentioned above. Greater genetic

294

S. C. Weaver

Fig. 3 Phylogenetic tree of the flaviviruses derived from partial NS5 sequences. Subtypes are written in parentheses after virus names. New World viruses are printed in bold and underlined. The tree was drawn using Bayesian methods and similar topologies were produced using maximum parsimony and neighbor joining. Numbers indicate bootstrap values for major clades to the right

Evolutionary Influences in Arboviral Disease

295

conservation in the tick-borne than mosquito-borne groups has suggested that different selective constraints operate during the evolution of these two groups (Shiu et al. 1991). The tick-borne viruses appear to have evolved in a progressive, clinal pattern from east to west across Asia and Europe (Zanotto et al. 1995), while the mosquito-borne flaviviruses have evolved in a more discontinuous manner, probably in several different regions of the world. The mosquito-borne viruses tend to exhibit relatively long time periods between lineage divergence, suggesting a “boom and bust” pattern of intense diversification followed by extinction of many lineages (Zanotto et al. 1996). The best examples of this pattern are DENV, which appear to be undergoing a period of rapid radiation (Holmes and Twiddy 2003a). Detailed maximum likelihood analyses of DENV isolates to analyze rates of synonymous vs nonsynonymous substitution suggest that different genotypes or lineages experience different selective pressures, including positive selection on some amino acid sites implicated in virulence and transmissibility (Twiddy et al. 2002a). Amino acid positions subject to weak, positive selection were also identified in the envelope glycoprotein of some but not all DENV serotypes. The majority of these amino acid sites were located in, or near to, putative T or B cell epitopes, suggesting immune selection, as well as in the NS2B and NS5 genes of DENV-2 (Twiddy et al. 2002b). These kinds of studies implying positive selection should be followed up with reverse genetic validation of fitness effects in mosquito vectors or surrogate model systems for human infection. As for the alphaviruses, the mobility of the reservoir hosts appears to have a strong influence on the population structure and evolution of flaviviruses. Those that use birds as reservoir hosts, like Japanese (Solomon et al. 2003), St. Louis (Kramer and Chandler 2001), and Murray Valley encephalitis viruses (Lobigs et al. 1988), as well as West Nile viruses (Beasley et al. 2003) evolve within broadly distributed lineages that exhibit genetic stability (Mackenzie et al. 1995). Flaviviruses with mammalian hosts exhibiting more limited dispersal, such as yellow fever virus, which uses nonhuman primate reservoir hosts, tend to be partitioned into smaller, geographically delineated populations (Bryant et al. 2003). However, the dengue viruses, which are perhaps the most mobile arboviruses due to the extensive and rapid travel behavior of human reservoir hosts, exhibit complex patterns of evolution within multiple lineages that are frequently introduced into new locations and also appear to undergo local extinctions (Holmes and Twiddy 2003; Thu et al. 2004). Genetic studies suggest that population shifts and replacements may be selected by adaptive mutations in the DENV nonstructural proteins (Bennett et al. 2003) and in cytotoxic T cell epitopes (Hughes 2001). Fitness for transmission may be responsible for some of these population changes; evidence from Ae. aegypti susceptibility studies suggests that an Asian genotype that has recently

296

S. C. Weaver

colonized the New World is more infectious than the American genotype it is replacing in some locations (Armstrong and Rico-Hesse 2003). This change in the distribution of DENV genotypes has critical public health implications because the Asian genotype is more likely to cause hemorrhagic disease (Watts et al. 1999). Analyses of flavivirus phylogenies also indicate considerable plasticity in their relationships with vertebrate hosts, and less plasticity in vector usage. Although tick- and mosquito-borne flaviviruses are occasionally isolated from mosquitoes and ticks, respectively, it appears that their principal vectors are very stable taxonomically within these groups (Gould et al. 2003). Even within the mosquito-borne clades, generic vector relations are relatively stable, with the hemorrhagic viruses mainly using Aedes spp. and the encephalitic members relying principally on Culex spp. (Fig. 3). Of particular interest in flavivirus evolution is the presence of a large group of animal viruses with no known arthropod vectors (Fig. 3). This group appears to have diverged early during the evolution of the flavivirus genus and may represent the ancestral phenotype. Another smaller group of bat viruses comprised of Yokose, Entebbe bat, and Sokoluk viruses appears to have lost the need for vector transmission secondarily (Gould et al. 2003). These nonvector-borne flaviviruses represent an ideal system to study the effect of vector transmission on arbovirus evolution because they share basic replication strategies and genetics with the vector-borne members of the genus. 1.4.2 Rates of Evolution Like the alphaviruses, estimates of flavivirus evolutionary rates are generally below those of single host animal RNA viruses. Also like the alphaviruses, the detection of diverse quasispecies populations within naturally infected mosquitoes and human hosts (Lin et al. 2004) suggests that mutation frequencies are comparable to those of other RNA viruses. The tick-borne viruses appear to evolve approximately two to three times more slowly than the mosquito-borne flaviviruses, probably the result of persistent infections of ticks for longer time periods than those of mosquitoes, a result of the prolonged tick life cycle (Gould et al. 2003). Nonviremic transmission of some tick-borne arboviruses (Jones et al. 1997) may result in nearly all replication occurring in the tick vector rather than the vertebrate host, compounding the effect of the tick reproductive cycle in slowing rates of sequence change. Based on phylogenetic trees, time scale estimates for flavivirus evolution have been estimated at 5,000–10,000 years since a common ancestor (Zanotto et al. 1996). However, as explained above, these time estimates rely on

Evolutionary Influences in Arboviral Disease

297

corrections for multiple substitutions of nucleotides and estimates of rate variation across nucleotide sites that may be unreliable. The recent report that flavivirus sequences are found in the DNA genomes of mosquitoes, probably the result of endogenous reverse transcriptase activity (Crochu et al. 2004), suggests a possible mechanism for arboviral sequence stability that would not be detected using phylogenetic methods. These flavivirus DNA sequences are apparently transcribed by mosquito cells, and cellular genes are generally conserved in sequence by high-fidelity DNA replication and proofreading. Therefore, recombination between these mosquito cell transcripts and RNA from an infecting flavivirus could result in restoration of ancestral viral RNA sequences via recombination.

2 Recombination and Reassortment As described above, evidence of recombination within the alphavirus genus is limited to the WEE complex, but the possibility of recombinants between more closely related viruses or strains has received little attention. The most likely venue for an alphavirus recombination event is difficult to predict; both mosquitoes and vertebrate hosts exhibit superinfection exclusion of sequential infection by closely related alphaviruses (Karpf et al. 1997). However, exclusion is not immediate, so sequential infection of a vertebrate within a few hours by multiple mosquito bites, or sequential infection of a mosquito via multiple, partial blood meals from two different viremic hosts could result in a dual infection. Like the alphaviruses, there is evidence of recombination from sequence and phylogenetic studies of DENV (Holmes and Twiddy 2003). Recombinant viruses as well as both parents have been detected within an infected mosquito (Craig et al. 2003). However, this recombination appears to be intraspecific (intraserotype) and there is no evidence of recombination between different flaviviruses comparable to the origins of the alphavirus WEEV as described above. The abundance of recombination in DENV may reflect the propensity for its principal vector, Aedes aegypti, to take multiple, partial blood meals from several different human hosts and to rely on blood as a carbohydrate nutritional source, rather than on plant nectars like most other mosquitoes (Harrington et al. 2001). Multiple feeding may increase the chances of dual infections in both mosquitoes (from biting more than one viremic human during a short time period) and in humans (from receiving multiple Ae. aegypti bites during a short time period due to this vector’s endophilic resting and feeding behavior, and its peridomestic larval habitats.

298

S. C. Weaver

Reassortment of gene segments has been shown to occur extensively within the family Bunyaviridae, and occurs efficiently in dually infected mosquitoes when the two different viruses are ingested within 2 days (Borucki et al. 1999). Reassortant bluetongue viruses can be detected in Culicoides variipennis that ingest two different strains within 5 days of each other, while superinfection exclusion prevents reassortment by day 7 (el Hussein et al. 1989). A recombinant Orthobunyavirus (family Bunyaviridae) was recently characterized from hemorrhagic fever cases during an East African epidemic. This virus, Ngari virus, a reassortant with S and L segments derived from Bunyamwera virus and an M segment from an unidentified member of the genus, demonstrates the public health importance of arbovirus reassortment (Gerrard et al. 2004).

3 Emergence Mechanisms of Arboviral Diseases 3.1 Direct Spillover The vast majority of arboviral diseases are zoonotic, with primary, enzootic transmission cycles involving wild animals and with humans and domestic animals representing tangential or dead-end infections that do not influence the long-term evolution of the pathogen. The simplest mechanism of infection is direct “spillover,” whereby enzootic transmission in the vicinity of humans or domestic animals, or the epizootic amplification of a virus due to favorable ecological conditions such as large vector populations following rainfall, lead to direct, tangential transmission (Fig. 4). This can result from a wide host range of the enzootic vector, including both reservoir hosts and humans or domestic animals, such as transmission of West Nile virus from birds to humans by the principal enzootic vector in north America, Culex pipiens (Turell et al. 2002). However, arboviruses such as EEEV that utilize vectors with narrow host ranges, such as Culiseta melanura, which feeds almost exclusively on birds, may rely on bridge vectors that bite both birds and humans for spillover. The mosquito host range of an arbovirus and the host preferences of its vector can therefore have a strong influence on arboviral disease. 3.2 Secondary Amplification The development of domestic animals has provided some arboviruses with the opportunity to undergo secondary amplification to increase levels of circulation and the probability of spillover to humans or domestic animals.

Evolutionary Influences in Arboviral Disease

299

Fig. 4 Cartoon showing mechanisms of human infection by zoonotic arboviruses. At the center is a typical enzootic cycle involving avian, rodent, or nonhuman primates as reservoir and/or amplification hosts and mosquito vectors. Humans become infected via direct spillover when they enter enzootic habitats and/or when amplification results in high levels of circulation in their proximity. Transmission to humans may involve the enzootic vector or bridge vectors with broader host preferences including humans. At the right, secondary amplification involving domestic animals can increase circulation around humans, increasing their chance of infection via spillover. Examples include Rift Valley fever, Japanese and Venezuelan equine encephalitis virus (VEEV). In the case of VEEV, mutations that enhance equine viremia mediate secondary equine amplification. At the left, dengue, yellow fever, and chikungunya viruses can use humans directly for amplification, resulting in urban epidemic cycles and massive outbreaks. In the case of dengue viruses, humans also serve as reservoir hosts

Good examples include JEV, which infects pigs and chickens living in close proximity to humans in many parts of Asia, resulting in local amplification and transmission to humans by mosquitoes that do not necessarily include the avian enzootic reservoir hosts among their preferred blood sources (Endy and Nisalak 2002). Another is Rift Valley fever virus, which amplifies itself in secondary cycles involving cattle, sheep, and other ungulates (Bouloy 2001). These viruses rely on wide vertebrate host ranges, as well as on susceptible vectors that feed on several hosts, to cause human disease via spillover from secondary amplification cycles. There is no evidence that adaptation

300

S. C. Weaver

is required for most of these secondary amplification cycles, i.e., most or all wild-type strains can readily infect these animals. A more complex form of secondary amplification is epitomized by VEEV, the most important alphaviral pathogen of the New World. The VEEV strains that undergo sustained, continuous transmission and long-term evolution are the enzootic variants that circulate primarily in sylvatic or swamp habits, where they utilize rodents as reservoir hosts and specialize almost exclusively on vectors in the Culex (Melanoconion) subgenus (Weaver et al. 2004b). The reservoir hosts from enzootic regions generate viremia sufficient to infect the mosquito vectors yet generally develop no detectable disease. The enzootic VEEV infect people, horses, bovines, and a wide range of other hosts via spillover, with humans suffering severe febrile disease that can be fatal. Horses and bovines living near enzootic habitats become infected but develop little or no disease. The limited dispersal of the Culex (Melanoconion) vectors generally limits disease resulting from direct spillover to locations close to forest or swamp habitats (Mendez et al. 2001). The “silent” sylvatic VEEV cycle is occasionally expanded into new habitats when mutations allow the virus to expand its host range and undergo secondary amplification, resulting in explosive equine epizootics and epidemics (Weaver et al. 2004b). Mutations in the E2 envelope glycoprotein mediate two critical adaptation events: (a) enzootic strains are selected for the generation of high titer equine viremia, which inadvertently (with respect to selection) results in equine virulence (Greene et al. 2005; Weaver et al. 2004a). Recent studies (SCW, unpublished) indicate that a single point mutation can mediate adaptation for equine viremia; (b) in some cases selection for enhanced infection of mosquito vectors that populate agricultural settings results in enhanced transmission among equine amplification hosts and humans. Adaptation to epizootic mosquito vectors can also involve as little as one mutation in the E2 protein (Brault et al. 2002a, 2004). The efficiency of VEEV in achieving dramatic host range changes with minor genetic changes epitomizes the threats naturally imposed by RNA viruses as emerging pathogens. The dramatic effect of importation of equines to the New World on VEE emergence also underscores the ability of arboviruses to exploit anthropogenic changes in unpredictable ways. 3.3 Humans as Arboviral Amplification Hosts A few arboviruses including DENV, the most important human pathogens, have exploited host range changes to the fullest extent to cause human disease by adapting to humans as reservoir and amplification hosts. Several

Evolutionary Influences in Arboviral Disease

301

arboviruses such as Ross River and chikungunya probably use humans as temporary amplification hosts during epidemic conditions, but there is no evidence that adaptation is involved. Chikungunya is particularly successful at exploiting human amplification because it uses a highly anthropophilic vector, Ae. aegypti (Woodall 2001). This mosquito itself underwent an evolutionary recent adaptation from the ancestral, sylvatic form found in West Africa, Ae. aegypti formosus (Tabachnick and Powell 1979). The derived form Ae. aegypti aegypti now lives in close contact with people in urban settings by relying on artificial water containers for its larval habitats, becoming endophilic to increase contact with people, and relying on blood (instead of plant carbohydrates) for its energetic needs (Harrington et al. 2001). In many respects, DENV are the ultimate human arboviral pathogens. The ancestral forms are sylvatic strains that continue to circulate in sylvatic habitats of West Africa and Asia. These strains utilize sylvatic treehole mosquitoes as vectors and nonhuman primates as reservoir hosts (Rudnick 1984; Diallo et al. 2003). Phylogenetic studies indicate that hundreds to thousands of years ago, the four DENV serotypes each underwent ecological and host range changes to establish peridomestic and later urban transmission cycles (Wang et al. 2000; Holmes and Twiddy 2003). These endemic and epidemic DENV strains use humans as their sole reservoir hosts and peridomestic mosquitoes as vectors to cause a huge burden of human disease in the tropics and subtropics. Experimental studies indicate that the ancestral, sylvatic DENV-2 strains underwent adaptation to increase their ability to infect the peridomestic vectors, Ae. aegypti and Ae. albopictus (Moncayo et al. 2004). Adaptation to human reservoir hosts may have been equally critical to human dengue emergence. In addition, the partial cross-protectivity exhibited among the four DENV serotypes may have allowed for the co-circulation of closely related DENV strains leading to immune enhancement, which can result in severe hemorrhagic forms of disease (Ferguson et al. 1999). A more complete understanding of the molecular determinants of host range changes responsible for the emergence of arboviruses like VEEV and DENV are critical to anticipating future disease trends and designing public health interventions. For example, several candidate DENV vaccines offer the hope of DEN eradication because humans are the only reservoir hosts for the strains circulating in most locations. However, predicting the ability of sylvatic DENV strains to re-emerge will depend on a more thorough understanding of human pathogenesis following sylvatic strain infection, and characterization of the genetic changes required to adapt to humans and peridomestic vectors. If only a small number of mutations is required, sylvatic DENV strains in Asia and Africa will represent a readily available source of new urban DENV emergence for the foreseeable future.

302

S. C. Weaver

4 Experimental Approaches to the Study of Arbovirus Evolution The ecological and phylogenetic approaches for studying arboviral evolution and disease emergence summarized above have revealed important host and vector associations and their evolutionary trends. They have also been used to generate mechanistic hypotheses that can be tested experimentally for optimal evaluation. For example, the slow rates of arbovirus evolution and strong evidence for purifying selection revealed by genetic studies suggest that the required alteration of vertebrate and invertebrate host infections imposed by most arbovirus transmission cycles constrains their evolution. In other words, viruses with adaptive mutations for the vertebrate host may impose fitness tradeoffs for infection and transmission by vectors, and vice versa. The identification of several alphavirus (Strauss and Strauss 1994; Schlesinger and Schlesinger 2001) and flavivirus (Lindenbach and Rice 2001) mutants with vector- or vertebrate-host-specific phenotypes and restrictions supports this hypothesis. One approach to studying the roles of vector and vertebrate hosts on arbovirus stability has been to assess the viral genetic diversity of populations within each host. Studies of EEEV populations in naturally infected birds and mosquitoes showed no evidence of differences in genetic diversity that would assign a greater constraining selective force to either host (Weaver et al. 1993). However, greater genetic diversity has been identified in humans naturally infected with DENV-3 than in Ae. aegypti either naturally or experimentally infected, suggesting that the mosquito vector constrains DENV evolution (Lin et al. 2004). This constraint may simply reflect the smaller population sizes and lesser amount of viral replication in mosquitoes than in vertebrate hosts. Recent advances in nucleic acid amplification and sequencing should be exploited to further assess the heterogeneity of arbovirus populations in vertebrate and vector hosts. 4.1 Effect of the Alternating Host Cycle on Arbovirus Genetic and Phenotypic Stability The alternating host life cycle of arboviruses and their genetic stability in nature suggest that alternating host replication may constrain adaptation, as explained above. Experimental validation of arbovirus genetic stability was first provided by studies of La Crosse virus during horizontal (oral infection of Ae. triseriatus mosquitoes) and vertical (transovarial transmission in mosquitoes) transmission (Baldridge et al. 1989). No RNA sequence changes were detected by RNA oligonucleotide fingerprinting in any of the passages,

Evolutionary Influences in Arboviral Disease

303

corroborating genetic stability in nature. Similar studies examining transovarial transmission of Toscana virus also revealed no genetic changes during over 12 sandfly generations during a 2-year time period (Bilsel et al. 1988). The effect of arbovirus adaptation to different hosts and cells was examined by Taylor and Marshall using the alphavirus Ross River virus (Taylor and Marshall 1975a). Serial passage in cell cultures or mice was followed by virulence testing. Passage in cell cultures depressed virulence, while mouse passage raised the level of virulence in a step-wise manner. Biological clones from both the original virus population and the 10th passage in mice were heterogeneous with respect to virulence, indicative of a quasispecies distribution. Most interesting was the finding that alternate passage between Ae. aegypti mosquitoes and mice resulted in no detectable change in virulence of two different wild-type virus strains (Taylor and Marshall 1975b). The authors speculated that the conservation of initial virulence by alternating mosquito-mouse passages could be related to the fact that Ae. aegypti can only be infected when fed on mice at the time of peak viremia, when a subpopulation of higher virulence is not present in high enough infectivity to be represented in the mosquito’s blood meal (Taylor and Marshall 1975b). 4.2 Adaptation of RNA Viruses to New Hosts and Host Cells The evolution of host range changes and host/vector alternation has been studied in several RNA viruses, only a few of which are arboviruses. Pioneering studies by Holland and colleagues with vesicular stomatitis virus (VSV) and other RNA viruses demonstrated high mutation frequencies, which allow for potentially rapid evolution (Holland et al. 1982), and the ability to rapidly adapt to new vertebrate cell lines as evidenced by dramatic increases in fitness (Holland et al. 1991). These experiments also revealed that such adaptation was often cell-specific, with fitness losses resulting in cells that were not subject to serial passages. Later, adaptation of VSV to sandfly cells was shown to reduce fitness for replication in vertebrate cells or in mouse brains, consistent with the host-specific nature of adaptation (Novella et al. 1995). Although not an arbovirus, evolution of the RNA virus mouse hepatitis virus has been studied by Baric and colleagues (Baric et al. 1997), who attempted simultaneous adaptation of mouse hepatitis virus to mixed cell cultures containing progressively increasing concentrations of nonpermissive hamster cells and decreasing concentrations of permissive murine cells. Variant, polytropic viruses with expanded host cell ranges were generated that replicated efficiently not only in hamster and murine cells, but also in human and nonhuman primate cell lines. However, porcine and feline cells

304

S. C. Weaver

were not efficiently infected. One derived polytropic variant was an RNA recombinant. Positive selection that appeared to be episodic occurred in the spike glycoprotein genes to allow for interspecies transfer (Baric et al. 1997). 4.3 Constraints of the Arbovirus Transmission Cycle on Adaptation to New Hosts More direct evidence for the effect of host alteration on arboviral adaptation came from studies of VSV and alphaviruses. Llewellyn et al. showed that a natural sandfly isolate of VSV replicates more efficiently in sandfly cells than isolates of mammalian origin. When VSV was passaged alternately in sand fly and hamster cells, or allowed to specialize on one cell type through serial passages, fitness increases were observed in all cases (Novella et al. 1999). The most surprising finding was that VSV replicating exclusively in hamster cells also increased its fitness in sandfly cells, indicating that specialization did not result in cell-specific adaptation. Similar results demonstrating host range expansion following selection for replication in a single cell type have also been obtained for other, non-arthropod-borne RNA viruses (Ruiz-Jarabo et al. 2004). The above studies with VSV suggested that arboviruses do not necessarily compromise their fitness by adapting to both vertebrate and invertebrate hosts. The number of mutations accumulated during alternated cell culture passages was similar or larger than that observed in VSV populations allowed to specialize, arguing against the hypothesis that the alternating cycle constrains rates of sequence change. Studies with the alphavirus EEEV yielded different results and conclusions (Weaver et al. 1999). In this case, specialization on vertebrate cells resulted in fitness losses for mosquito cells, and vice versa. However, viruses forced to alternate achieved comparable fitness increases in both cell types to the specialists, contradictory to the hypothesis that alternation constrains adaptation by arboviruses. However, rates of sequence change were lower in the alternating passage series, supporting the hypothesis that host alteration constrains evolutionary rates. Similar results with EEEV were also obtained using avian and mosquito cells (Cooper and Scott 2001) and Greene et al. (in press) obtained comparable result with SINV. Other studies with alphaviruses have focused on the subgenomic promoter and its response to selective pressure for adaptation to hosts. To determine if promoter utilization varies in vertebrate vs mosquito cells, Hertz and Huang (1995b) passaged SINV containing a library of different promoter sequences in hamster and mosquito cells. Selection was faster and more rapid in mosquito cells, which selected a smaller number of promoter sequences. Extensive

Evolutionary Influences in Arboviral Disease

305

passaging of the viral libraries in hamster cells led to a promoter consensus sequence that increasingly resembled the wild type, suggesting that the wildtype and similar sequences are optimal for promoter function in hamster cells (Hertz and Huang 1995a); similar results were obtained from mosquito cell passage (Hertz and Huang 1995b). These studies suggest that SINV makes little or no evolutionary compromise in maintaining the ability to replicate alternately in the two disparate host organisms. The inconsistencies in the results described above from different studies suggest limitations in the cell culture model systems used to study arbovirus evolution. In addition, artifactual adaptation events mediated by binding of some arboviruses to unnatural receptors such as glycosaminoglycans (Byrnes and Griffin 1998; Klimstra et al. 1998; Hilgard and Stockert 2000), which are found on the surface of both vertebrate and invertebrate cells, suggest that selection conditions in vivo might yield different results and conclusions. In vivo studies of the effect of natural transmission cycles on arbovirus evolution are extremely limited. Preliminary studies in our laboratory have yielded results that differ dramatically from in vitro model systems. When three different alphaviruses were introduced into laboratory transmission cycles involving unnatural hosts and vectors, high degrees of genetic stability were invariably observed, with no amino acid substitutions detected following ten cycles. No fitness changes could be detected in either the mosquito or vertebrate hosts (SCW, unpublished). However, specialization of VEEV for replication in hamsters without mosquito transmission resulted in a rapid gain in fitness, with faster viremia appearance and higher peak titers (Brault 2001). These results contrast with those from the same viruses using the cell culture model systems, and are completely consistent with the hypothesis that the alternating host cycle of arboviruses constrains their evolutionary rates and ability to adapt to new hosts.

5 Future Studies 5.1 Genetic and Phenotypic Stability of Arboviruses The genetic and phenotypic stability indicated by phylogenetic and experimental studies of arboviruses reviewed above has important public implications. As reviewed above, many experimental studies indicate the great capacity of RNA viruses to increase their virulence for vertebrate hosts, while others suggest that the alternating host transmission cycle inhibits such phe-

306

S. C. Weaver

notypic changes. Unfortunately, several aspects of vector-borne transmission cycles have not been examined to determine their effect on arbovirus stability. One aspect of virus transmission deserving more attention is population size, which has profound effects on the evolution of any organism (see chapters by Wilke et al. and by Escarmís et al., this volume). Large population sizes favor efficient natural selection; in the case of arboviruses, phylogenetic studies indicate that purifying selection acts to maintain phenotypic traits. However, the exact phenotypes under selection have not been determined comprehensively. The amino acid sequences of arboviral proteins are clearly one target of such purifying selection, as indicated by the overwhelming preponderance of synonymous mutations in arboviral genomes. However, in addition to conserved, cis-acting sequences in arboviral genomes that are under selection for primary RNA sequence and secondary RNA structure, genome-scale ordered RNA structures have been identified in some arboviruses (Simmonds et al. 2004). Such structures could introduce additional constraints on RNA virus evolution, and could also confound phylogenetic methods for estimating the ages of virus lineages due to violations of assumptions related to heterogeneity in mutation rates across nucleotide sites. While large population sizes can suppress rapid evolution of RNA viruses, small population sizes can lead to rapid genetic and phenotypic change. In the most extreme example, genetic bottlenecks can result in inefficient natural selection and rapid genetic drift (see also the chapter by Escarmís et al., this volume). This can result in the random fixation of mutations that can be deleterious for an organism without sufficient recombinatorial capacity or opportunity, resulting in progressive fitness declines via Muller’s ratchet; such effects have been demonstrated for the arboviruses VSV (Duarte et al. 1992) and EEEV (Weaver et al. 1999). Genetic drift can also facilitate the sampling of novel phenotypes, which cannot be selected in a step-wise fashion due to the lack of intermediate genotypes with improved fitness and the complexity of the selective landscape. A small population size could therefore be essential to allowing certain mutants to persist in nature. When bluetongue virus was placed in a laboratory transmission cycle involving Culicoides sonorensis vectors and sheep or calves, individual gene segments appeared to evolve independently by genetic drift in a host-specific fashion (Bonneau et al. 2001). In one case, a unique variant was randomly ingested by C. sonorensis insects that fed on a low titer blood meal, representing a genetic bottleneck, thereby fixing this new genotype by a founder effect. Additional studies of this kind are needed to assess the effects of blood meal titers on introducing bottlenecks leading to founder effects and genetic drift. Another example of the importance of drift may be the recombinant ancestor of WEEV; sequence analyses of WEEV (Hahn et al. 1988), and adap-

Evolutionary Influences in Arboviral Disease

307

tation experiments with artificially derived alphavirus recombinants (Lopez et al. 1994) indicate that adaptive mutations in the cytoplasmic tail of the E2 protein are necessary for efficient interactions with a heterologous capsid protein. Because the recombination event that generated the WEEV ancestor involved heterologous capsid and E2 genes, these results suggest that the original recombinant WEEV-like ancestor replicated inefficiently, and a population bottleneck and genetic drift are possible explanations for its initial persistence before adaptive mutations mediated more efficient replication. These examples indicate a need to better understand the effects of the vector-borne transmission cycle on arbovirus population sizes. Unfortunately, little is known about these sizes. Viral titers in insect vectors typically reach 105–7 infectious units, but the amount transmitted is usually much smaller. Estimates of mean saliva titers vary from approximately 40 to 200,000 infectious units, but mosquitoes frequently transmit far less virus (Chamberlain et al. 1954; Ross 1955; Lamotte 1960; Collins 1963; Hurlbut 1966; Gubler and Rosen 1976; Smith et al. 2005; Weaver et al. 1990; Vanlandingham et al. 2004). Viral populations within vertebrate reservoir or amplification hosts are generally very high compared to those in the vector. However, even less is known about the number of infectious virus particles that initiate the mosquito infection by entering and replicating in midgut epithelial cells. Better quantitative data on these population sizes within hosts and during transmission are needed to assess their influence on the evolution of arboviruses. 5.2 Host Switching by Arboviruses Despite the evidence above indicating that arboviruses may be constrained in their ability to adapt to new hosts, the phylogenetic studies described above indicate that host switching is a common event during long-term arbovirus evolution. Positive selection suggested by some phylogenetic studies may reflect past adaptive events or those in progress, but for the most part have not been validated experimentally. A fundamental question requiring additional experimental studies is whether these host-switching events require adaptation or simply take advantage of pre-existing, coincidental fitness for a new host. Experiments to answer this question are now possible using reverse genetic approaches now available for many arboviruses, and are critical for predicting the emergence of future arboviral diseases. For example, many zoonotic arboviruses circulate in tropical forest habitats that are being rapidly eliminated for logging and agriculture. The elimination of these natural habitats coupled with the rapid expansion of tropical urban populations is undoubtedly placing selective pressures on arboviruses to adapt for

308

S. C. Weaver

human-to-human transmission. The end results could be more DENV-like transmission cycles exploiting humans as reservoir and amplification hosts, with devastating public health consequences. Retrospective studies to determine the genetic determinants of adaptation to new hosts by DENV and other arboviruses will be invaluable in predicting the likelihood of additional arboviral urbanization and disease emergence.

References Armstrong PM, Rico-Hesse R (2003) Efficiency of dengue serotype 2 virus strains to infect and disseminate in Aedes aegypti. Am J Trop Med Hyg 68:539–544 Baldridge GD, Beaty BJ, Hewlett MJ (1989) Genomic stability of La Crosse virus during vertical and horizontal transmission. Arch Virol 108:89–99 Baric RS, Yount B, Hensley L, Peel SA, Chen W (1997) Episodic evolution mediates interspecies transfer of a murine coronavirus. J Virol 71:1946–1955 Beasley DW, Davis CT, Guzman H, Vanlandingham DL, Travassos da Rosa AP, Parsons RE, Higgs S, Tesh RB, Barrett AD (2003) Limited evolution of West Nile virus has occurred during its southwesterly spread in the United States. Virology 309:190–195 Bennett SN, Holmes EC, Chirivella M, Rodriguez DM, Beltran M, Vorndam V, Gubler DJ, McMillan WO (2003) Selection-driven evolution of emergent dengue virus. Mol Biol Evol 20:1650–1658 Bilsel PA, Tesh RB, Nichol ST (1988) RNA genome stability of Toscana virus during serial transovarial transmission in the sandfly Phlebotomus perniciosus. Virus Res 11:87–94 Bonneau KR, Mullens BA, MacLachlan NJ (2001) Occurrence of genetic drift and founder effect during quasispecies evolution of the VP2 and NS3/NS3A genes of bluetongue virus upon passage between sheep, cattle, and Culicoides sonorensis. J Virol 75:8298–8305 Borucki MK, Chandler LJ, Parker BM, Blair CD, Beaty BJ (1999) Bunyavirus superinfection and segment reassortment in transovarially infected mosquitoes. J Gen Virol 80:3173–3179 Bouloy M (2001) Rift Valley fever virus. In: Service MW (ed) The Encyclopedia of arthropod-transmitted Infections. CAB International, Wallingford, UK, pp 426– 434 Brault AC (2001) Genetic analysis of epizootic venezuelan equine encephalitis virus emergence mechanisms. In: Pathology. University of Texas Medical Branch, Galveston, Texas, p 318 Brault AC, Powers AM, Chavez CL, Lopez RN, Cachon MF, Gutierrez LF, Kang W, Tesh RB, Shope RE, Weaver SC (1999) Genetic and antigenic diversity among eastern equine encephalitis viruses from North, Central, and South America. Am J Trop Med Hyg 61:579–586 Brault AC, Powers AM, Holmes EC, Woelk CH, Weaver SC (2002a) Positively charged amino acid substitutions in the E2 envelope glycoprotein are associated with the emergence of Venezuelan equine encephalitis virus. J Virol 76:1718–1730

Evolutionary Influences in Arboviral Disease

309

Brault AC, Powers AM, Weaver SC (2002b) Vector infection determinants of Venezuelan equine encephalitis virus reside within the E2 envelope glycoprotein. J Virol 76:6387–6392 Brault AC, Powers AM, Ortiz D, Estrada-Franco JG, Navarro-Lopez R, Weaver SC (2004) Venezuelan equine encephalitis emergence: enhanced vector infection from a single amino acid substitution in the envelope glycoprotein. Proc Natl Acad Sci U S A 101:11344–11349 Bryant J, Wang H, Cabezas C, Ramirez G, Watts D, Russell K, Barrett A (2003) Enzootic transmission of yellow fever virus in Peru. Emerg Infect Dis 9:926–933 Byrnes AP, Griffin DE (1998) Binding of Sindbis virus to cell surface heparan sulfate. J Virol 72:7349–7356 Calisher CH, Karabatsos N (1988) Arbovirus serogroups: definition and geographic distribution. In: Monath TP (ed) The arboviruses: epidemiology and ecology. Vol. I. CRC Press, Boca Raton, FL, pp 19–57 Chamberlain RW, Kissling RE, Sikes RK (1954) Studies on the North American arthropod-borne encephalitides. VII. Estimation of amount of eastern equine encephalitis virus inoculated by infected Aedes aegypti. Am J Hyg 60:286–291 Cilnis MJ, Kang W, Weaver SC (1996) Genetic conservation of Highlands J viruses. Virology 218:343–351 Collins WE (1963) Transmission of Semliki Forest virus by Anopheles albimanus using membrane feeding techniques. Mosq News 23:96–99 Cooper LA, Scott TW (2001) Differential evolution of eastern equine encephalitis virus populations in response to host cell type. Genetics 157:1403–1412 Craig S, Thu HM, Lowry K, Wang XF, Holmes EC, Aaskov J (2003) Diverse dengue type 2 virus populations contain recombinant and both parental viruses in a single mosquito host. J Virol 77:4463–4467 Crochu S, Cook S, Attoui H, Charrel RN, De Chesse R, Belhouchet M, Lemasson JJ, de Micco P, de Lamballerie X (2004) Sequences of flavivirus-related RNA viruses persist in DNA form integrated in the genome of Aedes spp. mosquitoes. J Gen Virol 85:1971–1980 Diallo M, Ba Y, Sall AA, Diop OM, Ndione JA, Mondo M, Girault L, Mathiot C (2003) Amplification of the sylvatic cycle of dengue virus type 2, Senegal, 1999–2000: entomologic findings and epidemiologic considerations. Emerg Infect Dis 9:362–367 Duarte E, Clarke D, Moya A, Domingo E, Holland J (1992) Rapid fitness losses in mammalian RNA virus clones due to Muller’s ratchet. Proc Natl Acad Sci U S A 89:6015–6019 El Hussein A, Ramig RF, Holbrook FR, Beaty BJ (1989) Asynchronous mixed infection of Culicoides variipennis with bluetongue virus serotypes 10 and 17. J Gen Virol 70:3355–3362 Endy TP, Nisalak A (2002) Japanese encephalitis virus: ecology and epidemiology. Curr Top Microbiol Immunol 267:11–48 Ferguson N, Anderson R, Gupta S (1999) The effect of antibody-dependent enhancement on the transmission dynamics and persistence of multiple-strain pathogens. Proc Natl Acad Sci U S A 96:790–794 Gerrard SR, Li L, Barrett AD, Nichol ST (2004) Ngari virus is a Bunyamwera virus reassortant that can be associated with large outbreaks of hemorrhagic fever in Africa. J Virol 78:8922–8926

310

S. C. Weaver

Gould EA, de Lamballerie X, Zanotto PM, Holmes EC (2003) Origins, evolution, and vector/host coadaptations within the genus Flavivirus. Adv Virus Res 59:277–314 Greene IP, Paessler S, Austgen L, Anishchenko M, Brault AC, Bowen RA, Weaver SC (2005) Envelope glycoprotein mutations mediate equine amplification and virulence of epizootic Venezuelan equine encephalitis virus. J Virol 79: 9128–9133 Greene IP, Wang E, Deardorff ER, Milleron R, Domingo E, Weaver SC (2005) Effect of alternating passage on adaptation of Sindbis virus to vertebrate and invertebrate cells. J Virol (in press) Griffin DE (2001) Alphaviruses. In: Knipe DM, Howley PM (eds) Fields’ virology, 4th edn. Lippincott, Williams and Wilkins, New York, pp 917–962 Gubler DJ, Rosen L (1976) A simple technique for demonstrating transmission of dengue virus by mosquitoes without the use of vertebrate hosts. Am J Trop Med Hyg 25:146–150 Hahn CS, Lustig S, Strauss EG, Strauss JH (1988) Western equine encephalitis virus is a recombinant virus. Proc Natl Acad Sci U S A 85:5997–6001 Harrington LC, Edman JD, Scott TW (2001) Why do female Aedes aegypti (Diptera: Culicidae) feed preferentially and frequently on human blood? J Med Entomol 38:411–422 Hertz JM, Huang HV (1992) Utilization of heterologous alphavirus junction sequences as promoters by Sindbis virus. J Virol 66:857–864 Hertz JM, Huang HV (1995a) Evolution of the Sindbis virus subgenomic mRNA promoter in cultured cells. J Virol 69:7768–7774 Hertz JM, Huang HV (1995b) Host-dependent evolution of the Sindbis virus promoter for subgenomic mRNA synthesis. J Virol 69:7775–7781 Hilgard P, Stockert R (2000) Heparan sulfate proteoglycans initiate dengue virus infection of hepatocytes. Hepatology 32:1069–1077 Holland J, Domingo E (1998) Origin and evolution of viruses. Virus Genes 16:13–21 Holland JJ, Spindler K, Horodyski F, Grabau E, Nichol S, VandePol S (1982) Rapid evolution of RNA genomes. Science 215:1577–1585 Holland JJ, de la Torre JC, Clarke DK, Duarte E (1991) Quantitation of relative fitness and great adaptability of clonal populations of RNA viruses. J Virol 65:2960–2967 Holmes EC (2003) Molecular clocks and the puzzle of RNA virus origins. J Virol 77:3893–3897 Holmes EC, Twiddy SS (2003) The origin, emergence and evolutionary genetics of dengue virus. Infect Genet Evol 3:19–28 Hughes AL (2001) Evolutionary change of predicted cytotoxic T cell epitopes of dengue virus. Infect Genet Evol 1:123–130 Hurlbut HS (1966) Mosquito salivation and virus transmission. Am J Trop Med Hyg 15:989–993 Jones LD, Gaunt M, Hails RS, Laurenson K, Hudson PJ, Reid H, Henbest P, Gould EA (1997) Transmission of louping ill virus between infected and uninfected ticks co-feeding on mountain hares. Med Vet Entomol 11:172–176 Karabatsos N (1985) International catalogue of arboviruses. Am Soc Trop Med Hyg, San Antonio Karpf AR, Lenches E, Strauss EG, Strauss JH, Brown DT (1997) Superinfection exclusion of alphaviruses in three mosquito cell lines persistently infected with Sindbis virus. J Virol 71:7119–7123.

Evolutionary Influences in Arboviral Disease

311

Klimstra WB, Ryman KD, Johnston RE (1998) Adaptation of Sindbis virus to BHK cells selects for use of heparan sulfate as an attachment receptor. J Virol 72:7357–7366 Kramer LD, Chandler LJ (2001) Phylogenetic analysis of the envelope gene of St. Louis encephalitis virus. Arch Virol 146:2341–2355 Lamotte LC Jr (1960) Japanese B encephalitis virus in the organs of infected mosquitoes. Am J Hyg 72:73–87 Lin SR, Hsieh SC, Yueh YY, Lin TH, Chao DY, Chen WJ, King CC, Wang WK (2004) Study of sequence variation of dengue type 3 virus in naturally infected mosquitoes and human hosts: implications for transmission and evolution. J Virol 78:12717–12721 Lindenbach BD, Rice CM (2001) Flaviviridae: the viruses and their replication. In: Knipe DM, Howley PM (eds) Fields’ virology, 4th edn. Lippincott, Williams and Wilkins, New York, pp 991–1041 Llewellyn ZN, Salman MD, Pauszek S, Rodriguez LL (2002) Growth and molecular evolution of vesicular stomatitis serotype New Jersey in cells derived from its natural insect-host: evidence for natural adaptation. Virus Res 89: 65–73 Lobigs M, Marshall ID, Weir RC, Dalgarno L (1988) Murray Valley encephalitis virus field strains from Australia and Papua New Guinea: studies on the sequence of the major envelope protein gene and virulence for mice. Virology 165:245–255 Lopez S, Yao JS, Kuhn RJ, Strauss EG, Strauss JH (1994) Nucleocapsid-glycoprotein interactions required for assembly of alphaviruses. J Virol 68:1316–1323 Mackenzie JS, Poidinger M, Lindsay MD, Hall RA, Sammels LM (1995) Molecular epidemiology and evolution of mosquito-borne flaviviruses and alphaviruses enzootic in Australia. Virus Genes 11:225–237 Mendez W, Liria J, Navarro JC, Garcia CZ, Freier JE, Salas R, Weaver SC, Barrera R (2001) Spatial dispersion of adult mosquitoes (Diptera: Culicidae) in a sylvatic focus of Venezuelan equine encephalitis virus. J Med Entomol 38:813–821 Moncayo AC, Fernandez Z, Diallo M, Ortiz D, Sall A, Hartman S, Davis CT, Coffey LL, Mathiot CC, Tesh RB, Weaver SC (2004) Dengue emergence and adaptation to peridomestic mosquitoes. Emerg Infect Dis 10:1790–1796 Morzunov SP, Rowe JE, Ksiazek TG, Peters CJ, St. Jeor SC, Nichol ST (1998) Genetic analysis of the diversity and origin of hantaviruses in Peromyscus leucopus mice in North America. J Virol 72:57–64 Navarro JC, Weaver SC (2004) Molecular phylogeny of the Vomerifer and Pedroi groups in the Spissipes section of the subgenus Culex (Melanoconion). J Med Entomol 41:575–581 Norder H, Lundstrom JO, Kozuch O, Magnius LO (1996) Genetic relatedness of Sindbis virus strains from Europe, Middle East, and Africa. Virology 222:440–445 Novella IS, Clarke DK, Quer J, Duarte EA, Lee CH, Weaver SC, Elena SF, Moya A, Domingo E, Holland JJ (1995) Extreme fitness differences in mammalian and insect hosts after continuous replication of vesicular stomatitis virus in sandfly cells. J Virol 69:6805–6809 Novella IS, Hershey CL, Escarmis C, Domingo E, Holland JJ (1999) Lack of evolutionary stasis during alternating replication of an arbovirus in insect and mammalian cells. J Mol Biol 287:459–465 Ortiz DI, Weaver SC (2004) Susceptibility of Ochlerotatus taeniorhynchus (Diptera: Culicidae) to infection with epizootic (subtype IC) and enzootic (subtype ID) Venezuelan Equine encephalitis viruses: evidence for epizootic strain adaptation. J Med Entomol 41:987–993

312

S. C. Weaver

Poidinger M, Roy S, Hall RA, Turley PJ, Scherret JH, Lindsay MD, Broom AK, Mackenzie JS (1997) Genetic stability among temporally and geographically diverse isolates of Barmah Forest virus. Am J Trop Med Hyg 57:230–234 Powers AM, Brault AC, Tesh RB, Weaver SC (2000) Re-emergence of Chikungunya and O’nyong-nyong viruses: evidence for distinct geographical lineages and distant evolutionary relationships. J Gen Virol 81:471–479 Powers AM, Brault AC, Shirako Y, Strauss EG, Kang W, Strauss JH, Weaver SC (2001) Evolutionary relationships and systematics of the alphaviruses. J Virol 75:10118– 10131 Ross RW (1955) A laboratory technique for studying the insect transmission of animal viruses, employing a bat-wing membrane, demonstrated with two African viruses. In: Annual Report, Virus Research Institute, Entebbe, Uganda, pp 192–200 Rudnick A (1984) The ecology of the dengue virus complex in Peninsular Malaysia. In: Pang T, Pathmanathan R (eds) Proceedings of the International Conference on Dengue/DHF. University of Malaysia Press, Kuala Lumpur, p 7 Ruiz-Jarabo CM, Pariente N, Baranowski E, Davila M, Gomez-Mariano G, Domingo E (2004) Expansion of host-cell tropism of foot-and-mouth disease virus despite replication in a constant environment. J Gen Virol 85:2289–2297 Sammels LM, Coelen RJ, Lindsay MD, Mackenzie JS (1995) Geographic distribution and evolution of Ross River virus in Australia and the Pacific Islands. Virology 212:20–29 Sammels LM, Lindsay MD, Poidinger M, Coelen RJ, Mackenzie JS (1999) Geographic distribution and evolution of Sindbis virus in Australia. J Gen Virol 80:739–748 Schlesinger S, Schlesinger MJ (2001) Togaviridae: The viruses and their replication. In: Howley PM (ed) Fields’ virology, 4th edn. Lippincott, Williams and Wilkins, New York, pp 895–916 Scott TW, Weaver SC (1989) Eastern equine encephalomyelitis virus: epidemiology and evolution of mosquito transmission. Adv Virus Res 37:277–328 Shiu SY, Ayres MD, Gould EA (1991) Genomic sequence of the structural proteins of louping ill virus: comparative analysis with tick-borne encephalitis virus. Virology 180:411–415 Simmonds P, Tuplin A, Evans DJ (2004) Detection of genome-scale ordered RNA structure (GORS) in genomes of positive-stranded RNA viruses: implications for virus evolution and host persistence. RNA 10:1337–1351 Smith DR, Carrara AS, Aguilar PV, Weaver SC (2005) Evaluation of methods to assess transmission potential of Venezuelan equine encephalitis virus by mosquitoes and estimation of mosquito saliva titers. Am J Trop Med Hyg 73:33–39 Solomon T, Ni H, Beasley DW, Ekkelenkamp M, Cardosa MJ, Barrett AD (2003) Origin and evolution of Japanese encephalitis virus in southeast Asia. J Virol 77:3091– 3098 Strauss JH, Strauss EG (1994) The alphaviruses: gene expression, replication, and evolution. Microbiol Rev 58:491–562 Tabachnick WJ, Powell JR (1979) A world-wide survey of genetic variation in the yellow fever mosquito, Aedes aegypti. Genet Res 34:215–229 Taylor WP, Marshall ID (1975a) Adaptation studies with Ross River virus: laboratory mice and cell cultures. J Gen Virol 28:59–72

Evolutionary Influences in Arboviral Disease

313

Taylor WP, Marshall ID (1975b) Adaptation studies with Ross River virus: retention of field level virulence. J Gen Virol 28:73–83 Thu HM, Lowry K, Myint TT, Shwe TN, Han AM, Khin KK, Thant KZ, Thein S, Aaskov J (2004) Myanmar dengue outbreak associated with displacement of serotypes 2, 3, and 4 by dengue 1. Emerg Infect Dis 10:593–597 Tsai TF, Weaver SC, Monath TP (2002) Alphaviruses. In: Richman DD, Whitley RJ, Hayden FG (eds) Clinical virology. ASM Press, Washington, DC, pp 1177–1210 Turell MJ, Sardelis MR, O’Guinn ML, Dohm DJ (2002) Potential vectors of West Nile virus in North America. Curr Top Microbiol Immunol 267:241–252 Twiddy SS, Farrar JJ, Vinh Chau N, Wills B, Gould EA, Gritsun T, Lloyd G, Holmes EC (2002a) Phylogenetic relationships and differential selection pressures among genotypes of dengue-2 virus. Virology 298:63–72 Twiddy SS, Woelk CH, Holmes EC (2002b) Phylogenetic evidence for adaptive evolution of dengue viruses in nature. J Gen Virol 83:1679–1689 Van Regenmortel MHV, Fauquet CM, Bishop DHL, Carstens EB, Estes MK, Lemon SM, Maniloff J, Mayo MA, McGeogh DJ, Pringle CR, Wickner RB (eds) (2000) Virus taxonomy. Classification and nomenclature of viruses. Seventh report of the International Committee on Taxonomy of Viruses. Academic Press, San Diego Vanlandingham DL, Schneider BS, Klingler K, Fair J, Beasley D, Huang J, Hamilton P, Higgs S (2004) Real-time reverse transcriptase-polymerase chain reaction quantification of West Nile virus transmitted by Culex Pipiens Quinquefasciatus. Am J Trop Med Hyg 71:120–123 Wang E, Ni H, Xu R, Barrett AD, Watowich SJ, Gubler DJ, Weaver SC (2000) Evolutionary relationships of endemic/epidemic and sylvatic dengue viruses. J Virol 74:3227– 3234 Watts DM, Porter KR, Putvatana P, Vasquez B, Calampa C, Hayes CG, Halstead SB (1999) Failure of secondary infection with American genotype dengue 2 to cause dengue haemorrhagic fever. Lancet 354:1431–1434 Weaver SC (1995) Evolution of alphaviruses. In: Gibbs AJ, Calisher CH, Garcia-Arenal F (eds) Molecular basis of virus evolution. Cambridge University Press, Cambridge, pp 501–530 Weaver SC, Barrett AD (2004) Transmission cycles, host range, evolution and emergence of arboviral disease. Nat Rev Microbiol 2:789–801 Weaver SC, Scott TW, Lorenz LH (1990) Patterns of eastern equine encephalomyelitis virus infection in Culiseta melanura (Diptera: Culicidae). J Med Entomol 27:878– 891 Weaver SC, Rico-Hesse R, Scott TW (1992) Genetic diversity and slow rates of evolution in New World alphaviruses. Curr Topics Microbiol Immunol 176:99–117 Weaver SC, Bellew LA, Gousset L, Repik PM, Scott TW, Holland JJ (1993) Diversity within natural populations of eastern equine encephalomyelitis virus. Virology 195:700–709 Weaver SC, Kang W, Shirako Y, Rumenapf T, Strauss EG, Strauss JH (1997) Recombinational history and molecular evolution of western equine encephalomyelitis complex alphaviruses. J Virol 71:613–623 Weaver SC, Brault AC, Kang W, Holland JJ (1999) Genetic and fitness changes accompanying adaptation of an arbovirus to vertebrate and invertebrate cells. J Virol 73:4316–4326

314

S. C. Weaver

Weaver SC, Dalgarno L, Frey TK, Huang HV, Kinney RM, Rice CM, Roehrig JT, Shope RE, Strauss EG (2000) Family Togaviridae. In: van Regenmortel MHV, Fauquet CM, Bishop DHL, Carstens EB, Estes MK, Lemon SM, Maniloff J, Mayo MA, McGeogh DJ, Pringle CR, Wickner RB (eds) Virus taxonomy. Classification and nomenclature of viruses. Seventh report of the International Committee on Taxonomy of Viruses. Academic Press, San Diego, pp 879–889 Weaver SC, Anishchenko M, Bowen R, Brault AC, Estrada-Franco JG, Fernandez Z, Greene I, Ortiz D, Paessler S, Powers AM (2004a) Genetic determinants of Venezuelan equine encephalitis emergence. Arch Virol Suppl:43–64 Weaver SC, Ferro C, Barrera R, Boshell J, Navarro JC (2004b) Venezuelan equine encephalitis. Annu Rev Entomol 49:141–174 Woodall J (2001) Chikungunya virus. In: Service MW (ed) The encyclopedia of arthropod-transmitted infections. CAB International, Wallingford, UK, pp 115– 119 Zanotto PM, Gao GF, Gritsun T, Marin MS, Jiang WR, Venugopal K, Reid HW, Gould EA (1995) An arbovirus cline across the northern hemisphere. Virology 210:152–159 Zanotto PM, Gould EA, Gao GF, Harvey PH, Holmes EC (1996) Population dynamics of flaviviruses revealed by molecular phylogenies. Proc Natl Acad Sci U S A 93:548–553

CTMI (2006) 299:315–335 c Springer-Verlag Berlin Heidelberg 2006


Arenavirus Diversity and Evolution: Quasispecies In Vivo N. Sevilla1 (u) · J. C. de la Torre2 1 Centro de Biología Molecular Severo Ochoa, Universidad Autónoma de Madrid,

Campus de Cantoblanco, 28049 Madrid, Spain [email protected] 2 Department of Neuropharmacology, The Scripps Research Institute, IMM-6 10550 North Torrey Pines Road, La Jolla, CA 92037, USA

1 1.1 1.2 1.3

Genetic Variability of Arenaviruses . . . . . . . . . . . . . . . . . . . . . . . . . . Arenavirus Error-Prone Replication: Mutation, Recombination, and Reassortment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Quasispecies Swarms and Adaptive Value of Arenavirus Mutant Spectra Mutation Frequencies in Arenavirus Populations . . . . . . . . . . . . . . . . .

Contribution of Arenavirus Genetic Variability to Persistence and Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Overview of LCMV Biology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Selection of LCMV Variants In Vivo: Role in Viral Pathogenesis . . . . . . 2.2.1 Selection of Immunosuppressive Variants in LCMV Persistently Infected Mice . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.2 Contribution of LCMV Variants to Virally Induced Growth Hormone Deficiency Syndrome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. 316 . 317 . 319 . 320

2

3

. 321 . 321 . 322 . 323 . 326

Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 330 Abstract Arenaviruses exist as viral quasispecies due to the high mutation rates of the low-fidelity viral RNA-dependent RNA polymerase (RdRp). This genomic heterogeneity is advantageous to the population, allowing for adaptation to rapidly changing environments that present varying types and degrees of selective pressure. The significant variation in biological properties observed among lymphocytic choriomeningitis virus (LCMV) strains, the prototypic arenavirus, indicates to what extent a quasispecies dynamics may play a role in arenavirus adaptability and pathogenesis. Several aspects of arenavirus variability and its contribution to pathogenesis will be discussed.

316

N. Sevilla · J. C. Torre

1 Genetic Variability of Arenaviruses The family Arenaviridae comprises two distinct complexes: the LCMV-Lassa complex, which includes the Old World arenaviruses, and the Tacaribe virus (TACV) complex, which includes all known New World arenaviruses (Fig. 1). Early sequence analysis of arenaviruses revealed a significant degree of genetic stability with amino acid sequence homologies of 90%–95% among different strains of the prototypic arenavirus lymphocytic choriomeningitis virus (LCMV), whereas significantly higher levels of genetic diversity (37%– 56%) were observed for homologous proteins of different arenavirus species (Fulhorst et al. 1996; Southern 1996). More recently, genetic studies with a variety of arenavirus field isolates, including Lassa (Bowen et al. 2000), Junin (Garcia et al. 2000), Guanarito (GTO) (Weaver et al. 2000), Pirital (PIR) (Fulhorst et al. 1999), and Whitewater Arroyo (Fulhorst et al. 2001) have revealed a high degree of genetic variation among geographical and temporal isolates of the same virus species. Notably, a remarkably high level of genetic

Fig. 1 Phylogenic organization of Arenaviridae and electron micrograph of the prototypic Arenavirus LCMV

Arenavirus Diversity and Evolution: Quasispecies In Vivo

317

divergence (26% and 16% at the nt and amino acid level, respectively) has been documented among PIR isolates within very small geographic regions (Weaver et al. 2001) The substantial degree of inter- and intra-species genetic variation among arenaviruses appear to have important biological correlates, as suggested by the significant variation in biological properties observed among LCMV strains. In addition, early studies recognized the influence of the passage history on the virus biological properties (Hotchin et al. 1971; Hotchin et al. 1975; Hotchin and Sikora 1973). Thus, early mouse brain passages of LCMV were associated with immunologic tolerance and low mortality in newborn mice. In contrast, late mouse brain passages of LCMV lacked the toleranceinducing capacity and caused high mortality. Likewise, neonatal infection of certain mouse strains with LCMV strains Armstrong (ARM) and E-350 induced growth hormone deficiency syndrome (GHDS) associated with severe hypoglycemia, which frequently resulted in the death of the infected mice. In contrast, infection of the same mouse strains with LCMV strains WE and Traub did not cause GHDS. These striking phenotypic differences were associated with the ability of LCMV ARM and E-350, but not WE and Traub, to replicate at high levels in the GH-producing cells in the anterior pituitary (Oldstone et al. 1985). Moreover, LCMV variants with distinct phenotypes are often selected in different organs of infected mice. Thus, most isolates from the central nervous system (CNS) of mice persistently infected since birth with ARM tend to produce acute infection in adult mice, whereas isolates from the spleen of the same mice exhibit frequently an immunosuppressive phenotype that favors the establishment of persistence (Ahmed and Oldstone 1988). Such dramatic phenotypic differences among genetically closely related LCMV isolates indicate that a few amino acid replacements in LCMV proteins could suffice to produce important alterations in the virus biological properties. This, in turn, raises the question as to what extent quasispecies dynamics may play a role in arenavirus adaptability and pathogenesis. 1.1 Arenavirus Error-Prone Replication: Mutation, Recombination, and Reassortment The lack of proofreading–repair activities in viral RNA-dependent RNA polymerases (RdRp) is a key determinant of the observed high mutation rates for riboviruses. Thus, mutation rates during RNA genome replication are in the range of 10–3 –10–5 substitutions per nucleotide copied (Drake and Holland 1999; Holland 1992). This means that ribovirus genomes are highly unlikely to undergo two rounds of template copying without introducing an incorrect

318

N. Sevilla · J. C. Torre

nucleotide into the product. As with other RdRp, the LCMV L polymerase does not contain a recognizable proofreading exonuclease domain. However, it cannot be excluded that the Arenavirus replicase complex could include some cellular subunit with a capacity for error correction. In addition to mutation, both RNA recombination and genome reassortment can contribute to the genetic variability of arenaviruses. Recombination is particularly relevant for some positive-strand RNA viruses including picornaviruses and coronaviruses. Recombination can facilitate the generation of divergent genome combinations to produce large evolutionary jumps, as well as the rescue of fit genomes from debilitated mutated parents. In contrast, recombination in negative-strand (NS) RNA viruses, including arenaviruses, occurs at frequencies several orders of magnitude lower than in most positivestrand RNA viruses. Moreover, experimental evidence from both laboratory setting experiments and field molecular epidemiology indicates that recombination between two NS ribonucleoproteins (RNPs) is a very rare event, suggesting that perhaps only intramolecular recombination plays a significant role in the evolution of NS RNA viruses. Nevertheless, natural evidence for genetic recombination between two TACV complex members has been reported as the origin whitewater arroyo virus, the most recently discovered arenavirus (Charrel et al. 2001; Fulhorst et al. 1996). The segmented nature of the arenavirus genome facilitates the generation of reassortant viruses during co-infection of the same cell by two distinguishable parental viruses. In cell culture, reassortant viruses can be generated very efficiently. In some cases, their frequency is close to the theoretical predicted maximum of 50%. However, even among genetically closely related viruses, some particular combinations appear to be excluded. Hence, some specific reassortants are never recovered (Teng et al. 1996a). It is more difficult, however, to assess the contribution of segment reassortment to Arenavirus biology and evolution in the field. As with recombinant viruses, the generation of such reassortants would require co-infection of the same rodent host with two different arenaviruses. Dual infection of the same host may be highly restricted due to homotypic viral interference, which can be readily demonstrated with several arenaviruses including LCMV (Welsh and Pfau 1972) and TV (Gimenez and Compans 1980). This phenomenon is not strictly strain-specific, as illustrated by the existence of interference among different pairs of LCMV strains. Heterotypic interference between arenaviruses has been occasionally reported, and its degree appears to correlate with the genetic relationship of the viruses (Damonte et al. 1983; Welsh and Pfau 1972). This superinfection exclusion phenomenon could contribute to explaining the observed population partitioning in the field, resulting in the maintenance of independent evolutionary lineage of the

Arenavirus Diversity and Evolution: Quasispecies In Vivo

319

same strain within a small geographic range (Weaver et al. 2001). Nonetheless, it is plausible that in some instances arenavirus reassortants occur and they can display pathogenic potential, which is not manifested by any of the two parental viruses (Riviere 1987; Riviere et al. 1986; Riviere and Oldstone 1986). 1.2 Quasispecies Swarms and Adaptive Value of Arenavirus Mutant Spectra RNA viruses, even clonal populations, are organized as dynamic mutant distributions termed viral quasispecies (Domingo et al. 2001a; Domingo and Holland 1997). In each infected cell, replicating genomes compete for whatever resources might be limiting (nucleotide substrates, ribosomes, membrane sites, etc.) and needed to complete the virus life cycle. RNAs replicating in two separate cells do not enter into competition and could be regarded as separate quasispecies in separate compartments. However, virus progenies from different cells or tissues may eventually meet in a common compartment (e.g., a new organ such as the brain, or the blood stream), where they may then compete. Error-prone replication might contribute to spatial and temporal heterogeneities in RNA genome populations. Thus for LCMV, selection of organ-specific variants has been extensively documented (Ahmed and Oldstone 1988; Ahmed et al. 1984a). Viral quasispecies contain a plethora of variants, which allows these quasispecies to display different phenotypes in response to environmental demands. The influence of mutant distributions on the biology of RNA viruses is supported by demonstration of the presence and selection of variants in quasispecies evolving in vivo as well as evidence that mutant spectrum complexity may have a predictive value with regard to the response to antiviral treatments and to shifts from acute into chronic phases of viral disease. The presence of biologically significant variants within quasispecies is well documented by the presence of HIV variants resistant to viral inhibitors in patients who have not been subjected to therapy with the inhibitors, and some of these variants are likely to contribute to viral persistence and pathogenesis (Domingo 2003a, 2003b, Domingo et al. 2001a, 2001b; Domingo and Holland 1997). Although a great majority of mutations harbored in mutant swarms of RNA virus populations may not have either short- or long-term selective value, the evolution of viral quasispecies can, however, be adaptive and influence viral pathogenesis (Domingo 2001a, 2003a; Domingo and Holland 1997). Increasing evidence indicates that the complexity of mutant spectra influences the virus pathogenic potential, response to treatments and its ability to establish chronicity (Domingo 2003a, 2003b; Domingo et al. 2001b). Moreover,

320

N. Sevilla · J. C. Torre

it seems plausible that parameters that can be quantified in viral quasispecies could be used as surrogate markers to anticipate the virus responses to treatment and its potential for chronicity (Farci et al. 2000; Pawlotsky et al. 1998). 1.3 Mutation Frequencies in Arenavirus Populations No systematic measurements of mutant spectrum complexity have been reported for arenavirus populations. However, CTL escape mutants were readily selected in T cell receptor (TCR) transgenic mice infected with high doses (106 PFU) of LCMV (Pircher et al. 1990). Escape mutants included single amino acid replacements in T cell epitopes. CTLs from infected, transgenic mice lysed target cells coated with peptides representing the wild-type T cell epitopes, but not cells coated with variant peptides. Observations in cell culture also suggest high frequencies of CTL escape mutants. The latter were readily selected when B6-SV40 primary fibroblasts were infected, under CTL selective pressure, at a multiplicity of infection (MOI) of 1 or 10–3 , but not 10–5 , suggesting that escape mutants either pre-existed in the LCMV population or occurred early in the course of one replication round in the infected cells (Aebischer et al. 1991). Although CTLs play a key role in the control of LCMV infection, a neutralizing antibody response may be involved in the control of viremia and viral clearance in the absence of CD8+ T cells (Ciurea et al. 2001). High doses of LCMV-WE led to an enhanced antibody response, and control of virus replication in CD8–/– mice early on during infection. Nonetheless, neutralizing-resistant LCMV mutants emerged after weeks of CTL absence. The mutants included amino acid substitutions within GP1, and manifested a strong tendency to persist in mice. In addition to amino acid replacements, silent mutations (synonymous, not leading to amino acid substitutions) were found at frequencies of 3×10–4 substitutions per nucleotide, which would argue for considerable complexity of LCMV in these infected mice. Recent studies exploring the use of lethal mutagenesis to combat arenavirus infections have provided more direct estimates of the error frequency associated with RNA synthesis mediated by the Arenavirus polymerase (GrandePerez et al. 2002; Ruiz-Jarabo et al. 2003). The observed mutation frequencies of 2.6×10–4 to 5.5×10–4 for the NP, GP, and L regions analyzed are consistent with previous findings on genetic heterogeneity of LCMV (Sevilla et al. 2002) and other arenaviruses (Bowen et al. 1996, 1997, 2000; Charrel et al. 2003; Fulhorst et al. 2001; Weaver et al. 2000, 2001), and also within the range commonly observed for riboviruses (Domingo et al. 2001a).

Arenavirus Diversity and Evolution: Quasispecies In Vivo

321

2 Contribution of Arenavirus Genetic Variability to Persistence and Disease 2.1 Overview of LCMV Biology The prototypic Arenavirus LCMV has been used extensively as a model system for the study of viral persistence and pathogenesis. LCMV can infect its natural host, the mouse, either acutely or persistently. The outcome of the infection depends on both host and viral determinants, including the genetics, age, and immune status of the host, as well as route and dose of the virus inoculum (Buchmeier et al. 2001; Buchmeier and Zajac 1999). Investigations using this viral model have been central to defining basic virologic and immunologic concepts including the major histocompatibility complex (MHC) restriction of T cell recognition, tolerance, immunological memory, immune-mediated pathology, as well as the strategies by which viruses evade the host immune responses (Buchmeier et al. 2001; Buchmeier and Zajac 1999). Studies of LCMV virus–host interaction have also uncovered the ability of noncytolytic persistent viruses to induce disease by disrupting specialized functions of infected cells, revealing a new way by which viruses do harm in the absence of the classic hallmarks of cytolysis and inflammation (de la Torre and Oldstone 1996, 1989, 1991, 1998, 2002; Oldstone et al. 1982). LCMV is an enveloped virus with a bisegmented negative strand (NS) RNA genome (Buchmeier et al. 2001; Meyer et al. 2002) (Fig. 2). Each of the twogenome segments, designated L (~7.2 kb) and S (~3.4 kb), expresses two viral gene products using an ambisense coding strategy. The S RNA directs the synthesis of the nucleoprotein, NP, (ca 63 kDa) and the GP-C enveloped glycoprotein precursor. NP, encoded in antigenome polarity, is the most abundant viral protein and encapsidates viral genomes and antigenomic replicative intermediates. GP-C, encoded in genome polarity, is post-translationally cleaved by the cellular subtilase S1P into mature viral glycoproteins, GP-1 (40–46 kDa) and GP-2 (35 kDa) (Beyer et al. 2003; Pinschewer et al. 2003). Noncovalently associated GP1/GP2 complexes make up the spikes on the virion envelope, and mediate virus interaction with the host cell receptor (Cao et al. 1998; Kunz et al. 2002). The L segment codes for the virus RNA-dependent RNA polymerase (RdRp) (L, ~200 kDa) (Salvato et al. 1989) and a small (11 kDa) RING finger protein called Z (Salvato 1993; Salvato et al. 1992; Salvato and Shimomaye 1989), which functions as the arenavirus counterpart of the matrix protein found in many NS RNA viruses (Perez et al. 2003). Additional roles of Z in the Arenavirus life cycle have been proposed based on its interaction with several host cell proteins and its ability to inhibit RNA synthesis mediated

322

N. Sevilla · J. C. Torre

Fig. 2 Schematic representation of the Arenavirus genome organization and virion structure

by the LCMV polymerase (Borden et al. 1998a, 1998b, Campbell Dwyer et al. 2000; Cornu and de la Torre 2001, 2002). Closely related variants of the same LCMV strain can display remarkable phenotypic differences in vivo, including those related to disease manifestations. We will next discuss two different LCMV–host interactions that illustrate how the generation and selection of specific variants during the natural course of LCMV infection can significantly influence the outcome of the infection. 2.2 Selection of LCMV Variants In Vivo: Role in Viral Pathogenesis When immunocompetent adult mice are injected with LCMV, they generate a marked immune response to eliminate the infectious agent. Virus clearance is mediated by major histocompatibility complex (MHC) class-I restricted CD8+ antiviral cytotoxic T lymphocytes (CTL) (Borrow and Oldstone 1997; Buchmeier and Zajac 1999; Oldstone 2002). In contrast, mice infected neonatally or in uterus with LCMV become persistently infected for life. This persistent infection is due to the infection of thymic cells and the specific removal (negative selection) of lymphocytes with potential responsiveness to LCMV (Borrow and Oldstone 1997; Buchmeier and Zajac 1999; Oldstone 2002). This

Arenavirus Diversity and Evolution: Quasispecies In Vivo

323

long-term persistent infection allows the generation of viral variants that have specific growth advantages in certain tissues and can display phenotypic differences in vivo, including those related to disease manifestations. Several members of the arenavirus family use α-dystroglycan (α-DG) as a cellular receptor, including the Old World arenaviruses LCMV, Lassa fever virus (LFV), and Mobala as well as Clade C New World arenaviruses (Kunz et al. 2002). This receptor, α-DG, is a ubiquitously expressed, highly versatile cell surface receptor that provides a molecular link between the extracellular matrix (ECM) and the actin-based cytoskeleton and plays a critical role in cell-mediated assembly of basement membranes. Differences in binding properties to α-DG have been documented among arenaviruses, and such differences have been linked to differences in virus-host interactions, including tropism and pathogenesis, which is discussed below. 2.2.1 Selection of Immunosuppressive Variants in LCMV Persistently Infected Mice In the course of persistent infections of mice with LCMV, distinct viral variants can be isolated from the brain and lymphoid tissue (Ahmed et al. 1984b). These viral variants have different biological properties that correlate with the type of tissue from which they are isolated. One type of variant predominates in the CNS (Evans et al. 1994), whereas the other type predominates in lymphocytes and macrophages of the immune system (Fig. 3) (Ahmed and Oldstone 1988; Ahmed et al. 1984a). Most of the CNS isolates are similar to the parental Armstrong (Arm) strain used to infect the mice and induce a potent virusspecific CTL responses in adult mice, in which the infection is cleared within 2 weeks. In contrast, the majority of the isolates derived from the lymphoid tissue cause chronic infections in adult mice associated with absence of LCMVspecific CTL response, absence of CTL response to other viruses, and absence to antibody responses to soluble and foreign antigens (Ahmed and Oldstone 1988; Ahmed et al. 1984a). The prototypic member of the immunosuppressive variants is clone 13 (Cl 13), which was derived from the spleen of a mouse persistently infected from birth with the nonsuppressive LCMV Arm isolate. Sequence analysis revealed only two amino acid changes in Cl 13 compared with the parental Arm, F260L in the viral glycoprotein-1 (GP1) (Salvato et al. 1988), and K1079Q in the viral polymerase encoded by the L gene (Salvato et al. 1991). Injection of mice with reassortant viruses between Cl13 and Arm showed that only reassortants containing the S segment (GP1) that came from Arm exhibited high levels of LCMV-specific CTL and the infection was cleared in 2 weeks (Matloubian et al. 1993). These findings indicate that the immunosuppressive potential of Cl 13 maps to the S segment, with strong

324

N. Sevilla · J. C. Torre

evidence for an association of amino acid substitution F260L in GP1 with the immunosuppressive phenotype (Fig. 3). The emergence of cell-specific viral variants can be explained by the fact that when a viral infection occurs in the whole animal, the various organs and cell types present in the body provide a rich milieu for the selection of viral variants. During long-term persistence in carrier mice with continuous virus replication, and given the high mutation rate of RNA viruses, LCMV variants that have a growth advantage in certain cell types are likely to emerge. In addition, the contribution of the selective

Fig. 3 Selection of viral variants from mice persistently infected at birth with LCMV Arm. Variants isolated from a variety of immune cells (lymphoid tissue), when inoculated intravenously into adult immunocompetent mice, fails to generate a sufficient primary day-7 CTL response to clear viral infection resulting in a persistent viral infection (immunosuppressive variants). In contrast, those variants isolated from the CNS, upon inoculation of adult immunocompetent mice, generate a robust CTL response and clear the virus infection (nonimmunosuppressive). The immunosuppressive variants bind at high affinity to α-DG immobilized on membranes while the nonimmunosuppressive variants bind at low affinity to α-DG. The difference in GP amino acid residues is a single amino acid at position 260, having a Leu the immunosuppressive variants as opposed to the nonimmunosuppressive with a Phe at that site

Arenavirus Diversity and Evolution: Quasispecies In Vivo

325

pressure of the host immune system has to be taken into consideration. In order to evaluate all these components, the selection of a large number of LCMV variants in vivo has been documented (Sevilla et al. 2000). The vast majority of viruses isolated from lymphocytes and macrophages of CD4 knock out (ko), perforin ko, and TNF-α ko mice persistently infected with Arm showed an immunosuppressive phenotype when inoculated i.v. into adult immunocompetent mice. In contrast, all the CNS viral isolates from the same mice exhibited a nonimmunosuppressive phenotype. Selection of immunosuppressive and nonimmunosuppressive variants occurred at similar frequencies in normal and immunocompromised mice, indicating that it is highly unlikely that the host immune response represents a main driving force in this selection process. Sequence analysis of a large number of these variants consistently showed an amino acid exchange F260L or F260I in GP1 of the immunosuppressive viruses. As the selection of such viral variants is influenced in a critical way by the host tissue, it may involve, among other factors, virus–receptor interactions (Sevilla et al. 2000). In fact, these viruses showed a 200- to 500-fold enhanced binding to α-DG and a high dependence on α-DG for the infection of cells (Kunz et al. 2002). The analysis of the anatomic distribution of immunosuppressive variants in the spleen of adult immunocompetent mice 3 days after infection showed a different tropism for both group of variants (Fig. 4) (Sevilla et al. 2000). Cl13-like viruses localized exclusively in cells of the marginal zone and white pulp of spleen, whereas Arm-like viruses localized primarily in cells within the red pulp. This distinct distribution in spleen indicated that different subsets of cells were infected by the immunosuppressive and the nonimmunosuppressive viruses. The identification of cells in the spleen infected by LCMV revealed two subsets mainly infected by immunosuppressive viruses, CD11c- and DEC-205 positive dendritic cells (Fig. 4) (Sevilla et al. 2000). The major cell population expressing α-DG in spleen is CD11c+ cells. All these data together suggest that those viral variants with increased binding receptor affinity might be selected within the replicating quasispecies to bind to CD11c+ and DEC-205+ cells. The selective pressure is likely based on the interaction of the virus in initiating the dysfunction of the host antigen presentation, aborting the necessary host immune responses required to terminate the viral infection (Sevilla et al. 2004). The differential tropism of immunosuppressive (high-affinity binders) and nonimmunosuppressive (low-affinity binders) LCMV variants may relate to their ability to compete with host-derived ligands of α-DG. Variants with high-affinity receptor binding may out-compete host-derived ligands of αDG, such as Arm, and may be blocked by host-derived ligands. Given the highly complex tissue-specific binding pattern of dystroglycan, the ability of

326

N. Sevilla · J. C. Torre

Fig. 4a,b Differential tropism of LCMV immunosuppressive and nonimmunosuppresive variants in vivo. a Viral nucleic acid sequences of these variants located the immunosuppressive to the white pulp of spleen and the nonimmunosuppressive to the red pulp. b Replication of immunosuppressive variants in CD11c+ and DEC-205+ dendritic cells (up to 80% of such cells are infected) is shown. By comparison,