Advances in Insect Physiology, Volume 25

Advances in Insect Physiology

Volume 25

This Page Intentionally Left Blank

Advances in Insect Physiology edited by

P. D. EVANS Department of Zoology, The University Cambridge, England

Volume 25

ACADEMIC PRESS Harcourt Brace & Company, Publishers London San Diego New York Sydney Toronto Tokyo

Boston

ACADEMIC PRESS LIMITED 24-28 Oval Road London NW17DX United States Edition published by ACADEMIC PRESS INC. San Diego, CA 92101 This book is printed on acid-free paper Copyright 01994 by ACADEMIC PRESS LIMITED

All Rights Reserved

No part of this book may be reproduced in any form by photostat, microfilm, or any other means, without written permission from the publishers A catalogue record for this book is available from the British Library ISBN 0-1 2-0242257 Typeset by Keyset Composition, Colchester, Essex Printed and bound in Great Britain by TJ Press Ltd, Padstow, Cornwall

Contents Contributors Advances in Insect Virology L. A. KING, R. D. POSSEE, D. S. HUGHES, A. E. ATKINSON, C. P. PALMER, S. A. MARLOW, J. M. PICKERING, K. A. JOYCE, A. M. LAWRIE, D. P. MILLER, D. J. BEADLE Genetic Mechanisms of Early Neurogenesis in Drosophila melanogaster J. A. CAMPOS-ORTEGA

vii

1

75

Molecular Biology of the Honeybee

R. F. A. MORITZ

105

Information Processing in the Insect Ocellar System: Comparative Approaches t o the Evolution of Visual Processing and Neural Circuits M. MlZUNAMl

151

Allatostatins: Identification, Primary Structures, Functions and Distribution 6. STAY, S. S. TOBE, W. G. BENDENA

267

Index

339

Color Plates are located between pages 152 and 153.

This Page Intentionally Left Blank

Contributors A. E. Atkinson

School of Biological and Molecular Sciences, Oxford Brookes University, Gipsy Lane Campus, Oxford OX3 OBP, U K D. J. Beadle

School of Biological and Molecular Sciences, Oxford Brookes University, Gipsy Lane Campus, Oxford OX3 OBP, UK

W. G. Bendena Department of Biology, University of Iowa, Iowa City, ZA 52242-1324, USA J. A. Carnpos-Ortega

Institut fur Entwicklungsbiologie, Universitat zu Koln, 0-50931 Koln, Germany D. S. Hughes

School of Biological and Molecular Sciences, Oxford Brookes University, Gipsy Lane Campus, Oxford OX3 OBP, UK K. A. Joyce

School of Biological and Molecular Sciences, Oxford Brookes University, Gipsy Lane Campus, Oxford OX3 OBP, U K L. A. King

School of Biological and Molecular Sciences, Oxford Brookes University, Gipsy Lane Campus, Oxford OX3 OBP, UK A. M. Lawrie

School of Biological and Molecular Sciences, Oxford Brookes University, Gipsy Lane Campus, Oxford OX3 OBP, U K

S. A. Marlow

School of Biological and Molecular Sciences, Oxford Brookes University, Gipsy Lane Campus, Oxford OX3 OBP, U K Current address: Royal Free Medical School, UK D. P. Miller

School of Biological and Molecular Sciences, Oxford Brookes University, Gipsy Lane Campus, Oxford OX3 OBP, U K M . Mizunami

Laboratory of Neuro-Cybernetics, Research Institute for Electronic Science, Hokkaido University, Sapporo 060, Japan R. F. A. Moritz

Institut fur Biologie, Technische Universitat Berlin, Franklinstr 28/29, 10587 Berlin, Germany C. P. Palmer

School of Biological and Molecular Sciences, Oxford Brookes University, Gipsy Lane Campus, Oxford OX3 OBP, U K Current address: Department of Microbiology, University of Reading, UK J. M. Pickering

School of Biological and Molecular Sciences, Oxford Brookes University, Gipsy Lane Campus, Oxford OX3 OBP, UK R. D. Possee

NERC Institute of Virology & Environmental Microbiology, Mansfield Road, Oxford, U K B. Stay

Department of Biology, University of Iowa, Iowa City, IA 52242-1324, U S A S. S. Tobe

Department of Zoology, University of Toronto, Toronto, Canada

Advances in Insect Virology Linda A. King,a Robert D. Possee! David S. Hughes: Allan E. Atkinson,c Christopher P. Palmer,d Susan A. Marlow,8 Jason M . Pickering,f Kirsti A. Joyce,a Alison M. Lawrie,a Davin P. MilleP and David J. Beadlea a School

of Biological and Molecular Sciences, Oxford Brookes University, Oxford, UK NERC Institute of Virology and Environmental Microbiology, Mansfield Road, Oxford, UK Current address: Department of Biology, University of California, San Diego, California, USA Current address: Department of Microbiology, University of Reading, Reading, UK Current address: Royal Free Hospital Medical School, London, UK 'Current address: St Mary's Hospital Medical School, London, UK Correspondence to L. A. King, School of Biological and Molecular Science, Oxford Brookes University, Gipsy Lane Campus, Oxford OX3 OBe UK

1 Introduction 2 2 Baculoviruses 2 2.1 Introduction 2 2.2 Isolation and host range 2 2.3 Structure and classification 3 2.4 Baculovirus replication in vivo 5 2.5 Transmission of baculoviruses between hosts 6 2.6 Baculovirus replication in vitro 9 2.7 Biological control of insect pests 15 2.8 Baculovirus expresssion vectors 22 3 Entomopoxviruses 29 3.1 Isolation and host range 29 3.2 Structure and classification 30 3.3 Replication cycle in insects 33 3.4 Molecular studies 34 3.5 Replication in vitro 36 3.6 Biological control 38 4 Iridescent viruses 38 4.1 Classification, isolation and host range 38 4.2 Virus structure 39 4.3 Replication cycle 40 4.4 Molecular studies 42 4.5 Biological control 42 5 RNA viruses of insects 43 5.1 Introduction 43 ADVANCES IN INSECT PHYSIOLOGY VOL 25 ISBN &124)24225-7

0

Copyright 1994 Academic Press Limited All rights ofreproducnon in any form reserved

2

L. A. KING e t a / .

5.2 Nodaviridae 43 5.3 Tetraviridae (Nudaurelia p virus group) 48 5.4 Picornaviridae 50 References 54

1 Introduction

This chapter describes current advances that have been made in a number of aspects of insect virology in the past few years. As in all such reviews, the baculoviruses feature most dominantly, as it is this group of insect viruses that forms the focus of attention for the majority of insect virologists and molecular biologists. However, a number of recent advances have been made in our understanding of the replication of other insect viruses, notably the entomopoxviruses. This chapter does not aim to provide a comprehensive treatise on every aspect of insect virology, and at the beginning of each section the reader is referred to other recent review articles for further information as necessary. 2 Baculoviruses

2.1

INTRODUCTION

This section provides an overview of baculovirus biology and molecular biology, and focuses on progress in these areas. In particular, we describe advances in the use of baculoviruses as expression vectors of foreign genes, in the development of genetically modified insecticides, vertical transmission, and aspects of interactions with host cells including the phenomenon of apoptosis. Other recent reviews provide detailed accounts of the use of baculoviruses as insecticides (Entwistle and Evans, 1985; Podgewaite, 1985; Bilimoria, 1986; Huber, 1986; Possee et al., 1993; Vlak, 1993a,b), of their biology (Granados, 1980; Mazzone, 1985; Granados and Williams, 1986; Kelly, 1987; Volkman and Keddie, 1990; Cory, 1993), molecular biology (Blissard and Rohrmann, 1990) and their use as expression vectors (Luckow and Summers, 1988; Miller, 1988; Maeda, 1989a; Atkinson et al., 1990; Bishop and Possee, 1990; Fraser, 1989; King and Possee, 1992; O’Reilly er al., 1992). 2.2

ISOLATION AND HOST RANGE

Baculoviruses have only been isolated from invertebrates. Most examples have been found in insect species, but there are some reports of baculoviruses which are pathogenic for crustacea (Anderson and Prior, 1992). Baculovirus infections have been described in over 600 species of insects

ADVANCES IN INSECT VIROLOGY

3

including Lepidoptera (butterflies and moths), Hymenoptera (sawflies and wasps), Diptera (flies), Coleoptera (beetles) and Trichoptera (caddis flies) (Martignoni and lwai, 1986). The diseases that they can produce in insect pest populations have made them an obvious choice for use as biological insecticides (Entwistle and Evans, 1985). It is only in the last 10 years that the development of the baculovirus expression vector system has overshadowed, but not replaced, the original reason for studying these viruses. 2.3

STRUCTURE AND CLASSIFICATION

Baculoviruses have a large, double-stranded, covalently closed, circular DNA genome of between 88 and 200 kbp (Arif, 1986). This is associated with a highly basic (arginine-rich) protein of 6.5 kDa (Tweeten et al., 1980; Kelly et al., 1983; Wilson et al., 1987; Russell and Rohrmann, 1990b; Maeda et al., 1991). The DNA-protein complex is contained by a rod-shaped nucleocapsid comprising a 39 kDa capsid protein (Pearson et a f . , 1988; Blissard et al., 1989; Thiem and Miller, 1989) and an 87 kDa capsid protein (Mueller et al., 1990). Other structural components almost certainly remain to be identified. The size of the virus genome determines the length of the nucleocapsid, which may be 200-400 nm. The width remains constant at about 36 nm (Fraser, 1986). One or more nucleocapsids are further packaged within a single lipoprotein envelope to form the virus particle or virion. These structures may be occluded within a crystalline matrix, referred to as a polyhedron or granule (see below). Baculoviruses used to be classified into three subgroups (Matthews, 1982). This system has recently been superseded by classification into two subfamilies, the Eubaculovirinae and the Nudibaculovirinae (Table 1) (Francki et al., 1991). The former contains the nuclear polyhedrosis virus (NPV) and granulosis virus (GV) genera. The NPV genus is further subdivided into two subgenera: multiple nucleocapsids per envelope (MNPV) and single nucleocapsids per envelope (SNPV). The Nudibaculovirinae comprises one genus, the non-occluded baculoviruses, which do not produce a crystalline protein matrix around the virus particles. The Autographa californica (Ac) NPV is the type species of the subgenus MNPV and the Bombyx mori (Bm) NPV is the type species of the subgenus SNPV. In each case, several multiply or singly enveloped virions are embedded in a proteinaceous occlusion body or polyhedron. Polyhedra consist largely of a single protein (polyhedrin) of about 30 kDa (Hooft van Iddekinge et al., 1983; reviewed by Rohrmann, 1986, 1992), and form in the nucleus of infected cells. Polyhedra are large structures, ranging in size from 1 to 1 5 p m in diameter with an outer envelope which appears to confer additional strength and protection (Vlak et a f . , 1988; Williams et al., 1989). It was suggested that the polyhedrin envelope (PE) was carbohydrate in

L. A. KING et a/.

4

TABLE 1 Classification of the Baculoviridae ~

English vernacular name Baculoviruses

Nuclear polyhedrosis virus (NPV) Multiple nucleocapsids per envelope (MNPV) Single nucleocapsids per envelope Granulosis viruses (GV) Non-occluded baculoviruses Non-occluded baculoviruses (NOB)

Taxonomic status (international name)

~

Type species

Family (Baculoviridae) Subfamily (Eubaculovirinae) Genus Subgenus

Autographa californica (Ac) MNPV

Subgenus

Bombyx mori (Bm) SNPV

Genus Subfamily (Nudibuculovirinae) Genus

Plodia interpunctella (Pi) G V Heliothis zea (Hz) NOB

nature (Minion et al., 1979). More recent evidence supports the view that there is a protein component in the PE (Whitt and Manning, 1988). Gombart et al. (1989) mapped a gene encoding the PE protein in the AcMNPV and Orgyia pseudotsugata OpMNPV genomes. Studies with immunoelectron microscopy have shown that the PE protein is a major component of the PE (Russell and Rohrmann, 1990a). The PE is sensitive to protease, suggesting that protein forms a major part of the structure (Russell and Rohrmann, 1990a). Virions that have been released from polyhedra are known as polyhedra-derived virus (PDV), whereas virions that are released from cells without occlusion are called extracellular virus (ECV). The type species of the granulosis virus genus is the Plodia interpunctella GV. Granulosis viruses contain one virion (singly enveloped nucleocapsid) per virus occlusion body or granule. Granulin, the major granule protein, is similar in function to polyhedrin (Longworth et al., 1972; Akiyoshi et al., 1985; Chakerian et al., 1985). The type species of the non-occluded baculoviruses (NOB) is Heliothis zea NOB. These viruses are composed of singly enveloped nucleocapsids, but as the name suggests, they are not further packaged into occlusion bodies. Baculoviruses are usually named after the host from which they were isolated. While this svstem is convenient, it ignores the genetic relatedness

ADVANCES IN INSECT VIROLOGY

5

of viruses from different species. For example, the baculovirus isolated from the alfalfa looper was designated Aurugruphu culifurnica (Ac) MNPV (Vail et al., 1973); however, baculoviruses which are almost identical to AcMNPV have been found in Trichuplusia ni, Galleria mellunella (Smith and Summers, 1979) and Rachiplusia uu (Summers and Anderson, 1973). 2.4

BACULOVIRUS REPLICATIONI N V I V O

The larval or caterpillar stage of the insect life cycle is the most susceptible to infection with polyhedra or granules. Polyhedra are ingested when the insect feeds on contaminated diet and dissolve in the alkaline environment of the mid-gut to release the virus particles. After negotiating the peritrophic membrane lining the gut, the virus lipoprotein envelope then fuses with the plasma membrane of the gut wall cells and liberates nucleocapsids into the cytoplasm. The nucleocapsids serve to transport the virus DNA to the nucleus of the cell; it is unclear whether the nucleocapsid enters the nucleus or merely ‘injects’ the virus DNA via a nuclear pore. The processes of virus gene expression have been most thoroughly examined using cultures of insect cells maintained in vitro and infected with NPVs. What was apparent from the earliest studies on virus replication in vivo was the fact that baculoviruses produce two distinct structural forms in a biphasic replication cycle. In the infected gut cells, nucleocapsids are formed by about 8 h post-infection (hpi) and begin to bud through the nuclear membrane by 12 hpi, thus acquiring a lipid envelope. This membrane appears to be ‘lost’ in the cytoplasm, but the nucleocapsid gains another as it buds through the plasma membrane. In the course of this latter process, it also acquires a virus-encoded glycoprotein of 67 kDa (gp67; Blissard and Rohrmann, 1989; Whitford et al., 1989). This protein most probably serves to attach the budded virus to other susceptible cells within the insect; in cell culture the budded virus is 1000-fold more infectious than virus particles released from polyhedra, which lack gp67 (Volkman et al., 1976; Keddie and Volkman, 1985). The budded, or extracellular virus (ECV) is released into the haemolymph to infect other cells and disseminate infection throughout the insect; affected tissues include fat bodies, nerve cells and haemocytes. The cells infected in the second round of virus replication in the insect larva also produce ECV, but in addition occlude virus particles within polyhedra, in the nucleus. The virus particles occluded within polyhedra, which are genetically identical with the ECV, obtain their lipid envelope de nuvu within the nucleus and lack the gp67 found in the budded virus phenotype. The accumulation of polyhedra within the insect proceeds until the host consists almost entirely of a bag of virus. In the terminal stages of infection the insect liquifies and thus releases polyhedra which can infect other insects. Volkrnan and Keddie (1990) aptly described the infected insect at

L. A. KING et a/.

6

this stage as an ‘amorphous puddle’. Recent studies have demonstrated that a virus-encoded chitinase has a role in this process (Hawtin, 1993). Deletion of the chitinase gene from AcMNPV abolished the liquefaction process seen in the latter stages of virus infection. However, to date, the chitinase gene has only been found in AcMNPV or closely related viruses. It remains to be seen how baculoviruses lacking the chitinase gene can also cause liquefaction of virus-infected larvae.

2.5

TRANSMISSION OF BACULOVIRUSES BETWEEN HOSTS

Baculoviruses must infect a succession of susceptible host insects to ensure the maintenance of the virus population. In the most simple case, occluded virus released from a dead insect is almost immediately consumed by an uninfected insect larva and the replication cycle is reinitiated. Inevitably, the process is disrupted when host insects are unavailable. This may occur when the numbers of insects in an area decline due to the effects of the virus outbreak or, as in temperate climates, the host species overwinters as pupae (e.g. Panolis flammea) or eggs (e.g. Neodiprion sertifer). Occluded virus can overwinter on plant surfaces and in the soil. Carruthers et a f . (1988) showed that during fieldwork studies on the pine beauty moth P. Pammea, significant quantities of viable NPV inclusion bodies persisted on pine foliage, after a spray-initiated epizootic, and were still present at the hatch of the following year’s generation. Virus can also remain associated with the insects throughout the winter period. In a continuation of the study reported by Carruthers et a f . (1988), further work at the same site, 1 year after initial spraying, indicated a low-level persistent NPV infection in both the P. jlammea larvae and in the resulting overwintering pupae (Cory and Entwistle, 1990). Vertical transmission of baculoviruses, from parents to offspring, is probably a vital factor in the host-virus relationship. Fuxa et a f . (1992) have shown that polyhedra could be observed in larvae, pupae and adults of Spodoptera frugiperda whose parents had survived exposure to the S . frugiperda NPV. These polyhedra were isolated from the F, insects and fed to first instars. The polyhedra isolated from F1 adult insects did not cause further infection in the first instar larvae. Electron microscopy indicated that the non-infectious polyhedra did not contain virions. From these observations it could be seen that when the S. frugiperda NPV is vertically transmitted, the infections in the F1 adults are non-lethal and are manifested as non-infectious, empty polyhedra. Transmission of the virus between generations is thought to involve both infectious virus remaining viable on the outside of the egg and also the viral genome inside the egg existing in a persistent state (Longworth and Cunningham, 1968; Shapiro and Robertson, 1987). Recent work by Murray and Elkington (1989) has suggested that Lymanfriu dispar NPV is transmit-

ADVANCES IN INSECT VIROLOGY

7

ted between generations primarily via external contamination of eggs. Burand et af. (1992) reported the use of the polymerase chain reaction and short-wave UV irridiation to detect baculovirus DNA sequences from viral occlusion bodies contaminating the surface of gypsy moth eggs. They found that virus contaminating the surface of eggs is the primary source of inoculum for newly hatched larvae. When these larvae die, virus is released, providing a source of inoculum for subsequent infections (Murray and Elkington, 1989). There have been many reports of sporadic outbreaks of NPV infections in natural populations of insects that appear to be caused by the activation of latent or occult NPV or cytoplasmic (C)PV infections (Longworth and Cunningham, 1968; Evans and Harrap, 1982). Latency provides another mechanism for the transgeneration transmission of virus from one generation to another. It has been suggested that such infections can be activated by stress factors (Smith, 1963; Longworth and Cunningham, 1968) such as overcrowding and the consequent lack of food (Steinhaus and Dineen, 1960). This phenomenon occurs frequently in nature, as in the case of mass infestations of the gypsy moth, Porthetria dispar. The CPV of the alfalfa caterpillar, Cofias eurytheme, was discovered in insects subjected to the stress of overcrowding (Steinhaus and Dineen, 1960). Increased temperature and UV radiation may also activate occult virus. Hukuhara (1962) showed the effect of increased temperature on the induction of polyhedrosis by the CPV of the silkworm, Bombyx mori. Unsuitable diet, or an abrupt change of food plant, has in some situations, caused the activation of occult viruses. David and Gardiner (1966) reported the onset of polyhedrosis in Pieris brassicae by a granulosis virus following changes of population density, food and temperature. Biever and Wilkinson (1978) showed that when larvae of P. rapae were reared at 25°C on dehydrated diet (ca 50% moisture), the granulosis virus of P . rapae was activated. Himeno et af. (1973) investigated the effect of various temperature treatments on the infectivity of the nuclear polyhedrosis virus of silkworm larvae. From their results they suggest the virulent virus is bound with biochemical substances which change the virus into an inactive, non-infectious form. These substances, it is suggested, should be present in sufficient quantity to maintain a physiological inactive state. Such virus complexes would be sensitive to thermal shock. By keeping insects at low temperature, the viruses become free from the complexes in the insect, and can replicate. The ingestion of foreign NPV has also been shown to activate latent CPV and NPV infections (Smith, 1963). McKinley et af. (1981) demonstrated that feeding insect larvae (Spodoptera fittorufis)with heterologous NPV (Spodoptera exempta NPV, Hefiothis armigera NPV and S . frugiperda NPV) resulted in the activation of a latent virus resembling the homologous S. fittorafis NPV, rather than replication of the virus used as the inoculum. In another study, the polyhedra of Adoxophyes orana (SNPV) and Barathra

8

L. A. KING et a/.

brassicae (MNPV) were used in cross-inoculation experiments of A . orana and B. brassicae larvae, which were each suspected to harbour latent NPV infections (Ponsen and de Jong, 1964). Subsequent analysis of the DNA obtained from virus harvested from infected insects indicated that the two NPVs were not cross-infective but had activated latent virus in each host which resembled the homologous pathogen (Jurkovicova, 1979). A recent report by Hughes et al. (1993) found that a culture of Mamestra brassicae insects was found to harbour a latent baculovirus infection. The latent virus was activated by feeding the M . brassicae larvae with either the closely related P . fIarnmea NPV, or the distantly related AcMNPV. Restriction fragment profiles of the activated virus DNA showed that it was very closely related, if not identical, to M . brassicae NPV. Further studies using polymerase chain reaction (PCR) amplification of polyhedrin gene sequences demonstrated that the latent virus was present throughout the life cycle of the insect; eggs, larvae, pupae and adults. Using PCR analysis of DNA isolated from dissected tissues of fourth instar larvae, latent virus sequences were only detected in the fat body. Cell lines established from the isolated fat body tissue were also shown to harbour the latent virus sequences. Preliminary experiments using these latent virus-containing cell lines have demonstrated a use in further studies to elucidate the mechanisms of baculovirus latency and virus activation. There have also been reports of baculovirus-like particles in the reproductive tracts of female braconid wasps. Virus-like particles have been found in specific regions of the reproductive tracts of seven different species of wasp, all parasitoids of the tobacco hornworm, Heliothis virescens. The particles are nuclear in origin and are suggested to be related to baculoviruses on the basis of structural homologies (Stoltz et a[., 1976). Other baculovirus-like particles have been found persistently infecting an insect cell line derived from Heliothis zea (Granados et al., 1978; Kelly et al., 1981). Attempts to infect these cells with other baculoviruses induced the replication of the baculovirus-like particle. Despite these reports, the state of the latent or occult virus and the mechanism of activation remain unknown. The role of foreign polyhedra in the activation of latent virus is interesting, as previous studies by Grace (1962) and Longworth and Cunningham (1968) have suggested that polyhedra or polyhedrin itself, rather than the virons contained within, may be the important factor in the activation process. These suggestions were supported by Kelly et al. (1981) who found that the baculovirus-like particles persistently infecting an H . zea cell line could be induced by inoculation with inactivated baculovirus preparations. These reports suggest that polyhedra may have a more significant role than merely acting as a protective coat for the virus between hosts. Activation of the latent A . orana NPV with polyhedra of B. brassicae CPV has also indicated that the activating agent does not have to be a component of an NPV (Jurkovicova, 1979).

ADVANCES IN INSECT VIROLOGY

2.6

9

BACULOVIRUS REPLICATION IN VITRO

Cell lines which support the replication of AcMNPV have been derived from S. frugiperda (Sf) (Fall army worm) pupal ovarian tissue (Vaughn et a f . , 1977) or ovaries from adult Trichoplusia ni (cabbage looper) (Hink, 1970) and M. brassicae (cabbage moth) (King et al., 1991). Although other baculoviruses replicate in cell culture, few match the efficiency of the AcMNPV-S. frugiperda cell combination. The study of baculovirus replication in vitro has greatly simplified experiments to understand the kinetics of virus gene expression and replication. It should be noted, however, that after repeated passage in cell culture, baculoviruses may suffer insertions of host cell DNA and transposable elements within its genome (reviewed by Blissard and Rohrmann, 1990). A consequence of one such insertion within a gene encoding a 25 kDa protein, is the production of viruses which yield few polyhedra (FP phenotype) (Beames and Summers, 1988, 1989, 1990). Clearly, baculovirus genomes, in common with those of other viruses, are subject to mutations which may produce altered phenotypes. The biphasic production of ECV and polyhedra observed in insect larvae is also found in cell culture and, in general, the processes are similar. In this section more details are provided of the molecular events accompanying virus replication. The budded or ECV form of AcMNPV enters insect cells in culture by the process of adsorptive endocytosis. The nucleocapsids serve to translocate the DNA to the nucleus of the cell where virus replication is initiated. Baculovirus genes are expressed in a regulated fashion in infected insect cells. For convenience, and to match the effects of inhibitors of virus replication, virus gene expression is divided into three phases, early, late and very late (Friesen and Miller, 1986). In general, the expression level attained in each succeeding phase is higher than that of the preceding one. 2.6.1 Early gene expression Early gene expression is sometimes further subdivided into immediate-early and delayed-early phases. Immediate-early (IE) genes have been defined as those genes which can be transcribed in the presence of inhibitors of protein synthesis (e.g. cycloheximide) (Kelly and Lescott, 1981; Guarino and Summers, 1986a,b), indicating that other virus proteins are not required for their activation. Several examples have been described for AcMNPV. These include IE-1 (Guarino and Summers, 1986a), IE-N (Carson et a f . , 1988), PE38 (Krappa and Knebel-Morsdorf, 1991; Krappa et al., 1992) and ME53 (Knebel-Morsdorf et al., 1993). A spliced version of IE-1 has also been detected very early in AcMNPV-infected cells (Chisolm and Henner, 1988), and has been designated IE-0. Their classification as immediate-early genes was supported by the fact that copies of these genes, inserted into plasmids, were transcriptionally active after transfection into uninfected insect cells (Guarino and Summers, 1986a; Carson et al., 1988, 1991).

10

L. A. KING e t a / .

Delayed early genes were originally identified using cycloheximide and other inhibitors of protein synthesis in virus-infected cells (Kelly and Lescott, 1981). When cells were treated with these inhibitors and then returned to normal growth conditions, a distinctive pattern of protein expression was observed. Some proteins were expressed immediately after the inhibitor was removed, while others were produced after a delay; hence the division between immediate-early and delayed-early gene products. These results were extended in more recent experiments where immediateearly gene products have been found to transactivate the expression of delayed-early genes after transfection of uninfected insect cells (Guarino and Summers, 1986a; Carson et al., 1988). The 39K delayed-early gene is first detected in infected cells between 3 and 6 hpi. The chloramphenicol acetyl transferase (CAT) gene coding sequences were inserted, in frame, with a truncated AcMNPV 39K gene in a recombinant plasmid. Transfection of insect cells with this construct did not result in detectable CAT activity. When a plasmid containing the IE-1 gene was co-transfected with the 39K-CAT construct, however, significant quantities of CAT enzyme activity were produced (Guarino and Summers, 1986a). It was concluded from these experiments that the immediate-early class of gene products is required in the virus infection to transactivate the delayed-early genes. It has also been demonstrated that IE-N can augment the transactivation of 39K in combination with IE-1, but not when transfected with 39K alone (Carson et al., 1988). More recent results, however, do not support the immediate-early/ delayed-early subdivision. Studies with uninfected insect cell nuclear extracts confirm that both immediate-early and delayed-early genes are transcribed in vitro (Hoopes and Rohrmann, 1991; Glocker et al., 1992). This suggests that virus-encoded transcription factors are not required for the transcription of genes such as 39K. 2.6.2 Late genes The second major class of virus genes expressed in infected cells coincides with the onset of virus DNA replication at about 6hpi. If virus DNA synthesis is inhibited with aphidicolin, the late genes are not transcribed (Miller et al., 1981; Wang and Kelly, 1983). The virus encodes a DNA polymerase gene, the transcription of which is also inhibited by aphidicolin (Tomalski et al., 1988). Virus genes which are expressed during this phase include those encoding structural elements of the virus particles, e.g. the basic protein (Wilson e f al., 1987), the 39K capsid protein (Blissard et al., 1989; Thiem and Miller, 1989), the p87 capsid protein (Mueller et al., 1990) and the virus membrane glycoprotein (gp67) (Whitford et al., 1989; Blissard and Rohrmann, 1991). Transcription of these genes is probably mediated by the action of an unusual RNA polymerase which is a-amanitin resistant and

ADVANCES IN INSECT VIROLOGY

11

induced in cells late in infection (Grula et al., 1981; Fuchs ef al., 1983). It is unclear whether this enzyme is virus encoded or simply a modification of a host cell RNA polymerase. It does, however, have a different subunit composition to host RNA polymerases (Yang et al., 1991). Interestingly, the delayed-early 39K gene promoter is also active in the late phase of baculovirus gene expression. By virtue of a shift in the position of transcription initiation to a late gene promoter start site about 25 nucleotides upstream from the early mRNA start site, the gene is expressed at least until 18 hpi (Guarino and Summers, 1986a). 2.6.3 Very late genes There is some debate as to whether this class of genes should be regarded as a division separate from the preceding late genes. The principal reason for designating a third class is that they are transcribed in the period when the virus is assembling occlusion bodies within the nucleus of the infected cell, from about 15 hpi. The very late gene products include the polyhedrin protein, which forms the matrix of the occlusion body, and the p10 protein, which most probably has a role in polyhedra formation (Vlak et al., 1988; Williams et al., 1989; Russell et al., 1987). The p10 protein forms a crystalline matrix in the infected cell nucleus that is associated with polyhedra formation, although it does not form part of the mature polyhedron. Another justification might be to define very late genes as those which play no role in the formation of infectious virus particles; the polyhedrin and p10 genes may be deleted from the virus genome without affecting virion (ECV) production (Smith et al., 1983b; Vlak et al., 1988). These two very late genes have certainly been the major focus for development of baculovirus expression vectors, since their promoters are extremely efficient and can result in their combined proteins accounting for up to 50% of the total cell protein mass in the terminal stages of infection. 2.6.4 Baculovirus gene promoters Our increasing knowledge of baculovirus gene promoters has paralleled the development of the expression vector system. Many studies which have reported new expression vectors have added information concerning the nature of the promoters. The interest in the use of the very late gene promoters (polyhedrin and p10) as expression systems has focused attention on these structures. Transcription is initiated at a TAAG motif located 50 (AcMNPV polyhedrin; Howard er al., 1986) or 70 (AcMNPV p10; Kuzio et al., 1984) nucleotides upstream from the respective translation initiation codons. This motif is found in all of the baculovirus late and verv late gene

12

L. A. KING et a/.

promoters identified to date (Rohrmann, 1986; Blissard and Rohrmann, 1990). The importance of the 5' non-coding leader sequence of the polyhedrin gene was suggested by early results from Smith et al. (1985), which showed that removing part of this sequence decreased the level of expression. These results were extended in a more detailed study (Matsuura et al., 1987) where progressive deletions were made between the translation initiation codon and the transcription start site (i.e. 3' to 5'). These data showed that the length of the 5' non-coding leader could be correlated directly with the level of expression. The seven nucleotides before the ATG codon were found to be particularly important for maximum promoter activity. A series of linker-scan mutations in the 5' non-coding leader also confirmed the importance of this region (Rankin et al., 1988; Ooi et a l . , 1989). Furthermore, these latter studies also demonstrated that replacing eight nucleotides spanning the T A A G motif with a synthetic oligonucleotide linker resulted in a 2000-fold decrease in promoter activity. The transcription start site was further dissected in a study by Gearing and Possee (1990), where point mutations were introduced into the 11 nucleotides spanning the T A A G motif. Alteration of the TAAG sequence abolished promoter activity, while changes in the flanking regions only resulted in a small decrease in expression. The T A A G motif appears to be a universal feature of all the late and very late baculovirus gene promoters; it also serves as the late transcription start site in the delayed early 39K gene promoter (for review, see Blissard and Rohrmann, 1990). The region upstream from the transcription start site has also been analysed using deletion mutants (5' to 3') (Possee and Howard, 1987). It was concluded that between 7 and 20 nucleotides were required for maximum promoter activity. Interestingly, Ooi et al. (1989) reported that inserting a synthetic linker between 12 and 22 nucleotides upstream from the TAAG sequence increased the levels of steady-state mRNA by up to 50%. Both studies confirm that the sequences upstream from the mRNA start site are relatively unimportant. Similar results have been obtained with the AcMNPV p10 gene promoter (Weyer and Possee, 1988, 1989). The conclusion from these studies is generally the same as with the studies of the polyhedrin promoter. The p10 promoter consists of about 100 nucleotides extending upstream from the translation initiation codon. There is an absolute requirement for the 5' non-translated leader sequence and about 30 nucleotides upstream from the transcription start site. The p10 promoter is fully functional when located in a heterologous location within the virus genome, namely upstream of the polyhedrin gene (Weyer et af., 1990). Recently, advances have been made in our understanding of the transcriptional control mechanisms involved in the regulation of the late and very late genes. Glocker et al. (1993) have shown that nuclear extracts prepared

ADVANCES IN INSECT VIROLOGY

13

from AcMNPV-infected cells can support accurate transcription from several late and very late gene promoters including p39, 39K, p10 and polyhedrin. In vitro transcription of the 39K promoter was resistant to high concentrations of a-amanitin and tagetitoxin, which suggested that neither RNA polymerase I1 nor 111 was responsible for baculovirus late gene transcription. Furthermore, Passarelli and Miller (1993a) have identified three genes that play a crucial role in late (p39) and very late (polyhedrin) gene transcription. Using a method based on subtraction of clones from an AcMNPV genomic library, which is able to transactivate promoters of reporter plasmids in transient expression assays, the three genes identified were IE-1, IE-N and LEF-2. IE-1 was found to be necessary but insufficient for expression from the p39 and polyhedrin gene promoters. The presence of IE-N increased expression from both gene promoters but the third gene, LEF-2 (late gene expression factor 2), was found to be specifically required for expression from the late and very late gene promoters. Two other late gene expression factors have since been identified and characterized, LEF-1 (Passarelli and Miller, 1993b) and LEF-3 (Li et al., 1993). Other virus gene promoters have been less well characterized at the primary sequence level. The IE-1 gene is transcribed from two mRNA start sites, implying that it has two promoters (Chisholm and Henner, 1988). The first of these structures appears to be functional between 0 and 2 hpi and produces a spliced transcript of 2.1 kb. The 5' end of this transcript maps to a position about 4 k b p upstream of the mRNA start site of the second, 1.9 kb transcript. The 1.9 kb transcript reaches its steady-state level 30 min after the virus adsorption period and maintains this level throughout the infection (Chisholm and Henner, 1988). The 39K gene promoter also utilizes two mRNA start sites in the delayed-early and late phases of gene expression (see above). The DNA polymerase gene also has two start sites for mRNA initiation, but these are both active in the early phase of gene expression (Tomalski et a l . , 1988). Another interesting feature of immediate-earlyldelayed-early gene expression is the role of enhancer elements. In AcMNPV these consist of five regions of homologous repeats (hr 1-hr5) containing repeated sequences with multiple EcoRI sites (Cochran and Faulkner, 1983; Guarino et al., 1986). When linked in cis with immediate-early and delayed-early genes, these sequences may enhance transcription by up to 3000-fold (Guarino and Summers, 1986b; Nissen and Friesen, 1989). The role, if any, of these enhancer elements in late and very late gene transcription remains to be elucidated. 2.6.5 DNA replication Relatively little is understood about the DNA replication of AcMNPV, although the structure and functional organization of the genome has been

14

L. A. KING et a/.

well characterized (reviewed by Kool and Vlak, 1993). A few genes have been identified which are thought to be involved in the process, including a DNA polymerase (Tomalski et al., 1988), a proliferating cell nuclear antigen-like protein (O’Reilly et al., 1989), and a helicase (Lu and Carstens, 1991). Attempts to isolate putative origins of DNA replication have proved difficult (Blinov, 1984). However, more recently there has been some progress in this area and several groups have reported the identification of putative origins of replication on the AcMNPV genome (Pearson et al., 1992; Kool et al., 1993). In the report by Kool and colleagues, use was made of defective AcMNPV genomes that lacked considerable portions of the genome (up to 43%), yet still retained the ability to replicate in cells in vitro. Three separate regions were retained in the mutant genomes and two of these were identified as containing putative origins of replication. The two regions were mapped to the HindIIIB fragment between map units 50.1 and 53.2, and to the HindIIIQ fragment between map units 87.2 and 88.9. Transfection of AcMNPV-infected S. frugiperda cells with plasmids containing these sequences resulted in the amplification of the plasmids, as demonstrated by DpnI sensitivity assays. 2.6.6 Apoptosis Apoptosis or programmed cell death is recognized in vertebrate cells. A similar phenomenon has been reported in certain insect cells infected with an AcMNPV mutant (vAcAnh)(Clem et al., 1991; Crook et al., 1993). This virus was isolated because of its production of small plaques lacking polyhedra in S. frugiperda cells. Subsequent tests in three cell lines demonstrated that it caused premature death of S. frugiperda, and B. mori but not T. ni cells. The mutation in vAcAnh was mapped within the virus genome and the appropriate region of virus DNA sequenced. When compared with parental virus DNA sequence, a deletion of 754 nucleotides was noted, which would foreshorten the native p35 protein by 132 amino acids. Inactivation of p35 in the wild-type virus confirmed its role in preventing apoptosis. These data suggest that apoptosis may enable certain insects to overcome virus infection. Kamita et al. (1993) have also described a similar phenomenon in cells infected with the BmNPV. More recently, Clem and Miller (1993) have shown that viral gene expression is abnormal in S. frugiperda cells infected with the p35-mutant AcMNPV, with a delay in the expression of both early and late genes and a lack of expression of the very late genes. The infectivity of the p35-mutant for S. frugiperda larvae was about 1000-fold lower than wild type or revertant viruses. In contrast, the replication and infectivity of the p35-mutant in T. ni cells or larvae was equivalent to wild-type or revertant viruses (Clem et al., 1991). Thus it would appear that a host apoptotic response provides protection against

ADVANCES IN INSECT VIROLOGY

15

viral infection at the organismal level, and that the p35 gene may constitute a host range determinant for AcMNPV infection (Clem and Miller, 1993). Crook er al. (1993) have identified a gene from the Cydia pomonella granulosis virus that was able to rescue wild-type infection from S. frugiperda cells infected with the p35-mutant AcMNPV. This gene, named iap (inhibitor of apoptosis), had no significant homology to the AcMNPV p35 gene but contained a zinc finger-like motif which has been found in other genes with the potential to regulate apoptosis. Both the p35 and iap genes were able to block apoptosis induced by actinomycin D , indicating that these genes act by blocking cellular apoptosis rather than by preventing viral stimulation of apoptosis. 2.7

BIOLOGICALCONTROL OF INSECT PESTS

2.7.1 History offield control The detailed studies of baculovirus biology and molecular genetics described above were predated by numerous examples of the use of baculoviruses as biological control agents of insect pests. It is outside the scope of this section to provide comprehensive descriptions of each one; the reader is referred to reviews by Podgewaite (1985), Entwistle and Evans (1985), Huber (1986) and Evans and Entwistle (1987) for detailed accounts of this topic. One of the most spectacular successes has been the control of the velvetbean caterpillar, Anticarsia gemmatalis, with the A . gemmuralis (Ag) NPV (Moscardi and Correa Ferreira, 1985; Moscardi, 1989; P. Zanotta, personal communication). A long-term programme to use this virus in insect control programmes has resulted in the annual spraying of 2 million hectares of soybeans (P. Zanotto, personal communication). An accurate description of insect pest control with baculoviruses is also complicated by the fact that many field programmes are not documented in the scientific literature (J. Cory, personal communication). In developing countries, farmers and other agriculturalists are more concerned with controlling the insect pests than conducting and reporting Western-style field trials. 2.7.2 Advantages and disadvantages The principal advantage of baculoviruses as insecticides is their narrow host range. The Baculoviridae are specific for invertebrates. Each virus isolated from insects has a narrow host range enabling the targeting of particular pest species. Conversely, this may be considered a disadvantage, since more than one virus may need to be applied to a crop to effect control of different insect pests. Chemicals can be used to control multiple insect species, although beneficial species may also be affected by these agents. Despite the lack of a direct effect, it has been suggested that the decimation of a pest

16

L. A. KING et ar

population with baculoviruses could seriously affect dependent predator! such as insect parasitoids (Hochberg, 1991). In subsequent years, then would be insufficient predators to provide natural control of the pest insect permitting an explosion in their numbers with inevitable crop damage While it is difficult to refute this argument, the same hypothesis holds gooc for chemical insecticide application. The use of chemical insecticides is frequently associated with the development of resistance in the target species. To date, the same phenomenon has not been documented in situations where baculoviruses have been used for insect pest control. Although it is considered that mas5 resistance to baculoviruses is unlikely to develop (Kirschbaum, 1985), this conclusion has yet to stand the test of time and it would be unwise to assume that insect populations will remain unresponsive to severe selection pressure. Although baculovirus infection of insect larvae often results in death, the virus can coexist with the host without producing symptoms (see below). Chemical insecticides are cheap and easy to produce. Baculoviruses require a biological system, the insect larva, to provide large quantities of infectious material for field application. Large-scale production by this method is labour intensive, which in developed countries is expensive. The maintenance of virus-free insect cultures is also problematic unless one is careful to separate the virus production facility from the stock culture. Baculoviruses may be stored as frozen, or freeze-dried preparations. The latter is particularly advantageous in areas lacking facilities for chilled storage. The most frequently cited disadvantage of baculoviruses as insecticides is the long period, relative to chemical agents, after application of the virus and death of the target insect. This may be as long as 7-10 days, depending on temperature and initial virus dose. In this period, insect larvae continue to feed, resulting in damage to the crop. In contrast, chemicals are usually effective after a matter of hours and almost instantly stop the insects from feeding. Before baculoviruses can seriously challenge the market domination of chemical insecticides their speed of action must be improved. 2.7.3 Genetic modification of baculovirus insecticides

The most promising approach to solving the problem of delay in killing the insect pests with baculovirus insecticides involves the insertion of foreign DNA encoding insect-specific hormones, enzymes or toxins into the virus genome, under the control of a strong virus gene promoter. Several examples have been reported in the literature, each describing varying degrees of success. Water balance in insects is influenced by diuretic and anti-diuretic

ADVANCES IN INSECT VIROLOGY

17

hormones (Maddrell, 1986). To test whether this balance might be affected by elevated levels of a diuretic hormone (DH), Maeda (1989b), inserted a sequence encoding the Manduca sexta (tobacco hornworm) diuretic hormone into the B. mori NPV genome, under the control of the polyhedrin gene promoter. A signal peptide coding region from the Drosophila melanogaster (fruit fly) cuticle protein was fused in frame with the DH coding sequence to effect secretion from virus-infected cells. The polyhedrin-negative virus was used to infect B. mori larvae by injection directly into the haemocoel. It was reported that after infection, a 30% decrease in haemolymph volume was noted compared with mock-infected or unmodified virus-infected larvae. Insect larvae infected with the recombinant virus died one day earlier than controls. The production of D H in insect larvae was inferred from analysis of mRNA isolated from virusinfected insects. A reverse phase column was used to isolate D H from virus-infected insects; this material was assayed using newly emerged Pieris r a p e (small white butterfly) adults (Kataoka et al., 1989). The small size of the D H and complicated assay procedures required to detect biologically active material, make it difficult to assess the effectiveness of inserting such coding regions into the virus genome. Another complex physiological process amenable to disruption by genetically modified baculoviruses is that of metamorphosis. Large amounts of juvenile hormone esterase (JHE) are produced by the insect larva in the last instar, prior to pupation. This enzyme hydrolyses the chemically stable, conjugated methyl ester of juvenile hormone (JH) to the J H acid (Hammock, 1985). A decrease in JH titre in the last larval instar is associated with a cessation of feeding and metamorphosis (de Kort and Granger, 1981). These data, together with the fact that inhibition of J H E produces overgrown insect larvae (Sparks and Hammock, 1980), suggested that elevated levels of JHE in early larval instars would inhibit feeding and cause premature pupation (Hammock et al., 1990). A cDNA clone encoding J H E was prepared from fat body mRNA isolated from H. virescens (tobacco budworm) (Hanzlik et al., 1989). This was inserted into a baculovirus transfer vector, under the control of the AcNPV polyhedrin gene promoter, and used to produce a polyhedrin-negative recombinant virus (Hammock et al., 1990). The virus produced active JHE in cell culture. In virus-infected neonate T. ni, active J H E was also produced and was accompanied by a reduction in feeding. The feeding inhibition was not observed when later larval instars were used. This may have been due to the production of insufficient JHE. The levels of J H E recorded in the haemolymph of virus-infected insects were only 10% of those normally measured in the last larval instar, prior to ecdysis. The virus-induced J H E may have been insufficient to overcome J H biosynthesis, or other control mechanisms as yet undefined. The enzyme is also very unstable in vivo when produced by the recombinant virus, or in its natural form. The baculovirus also produces an

18

L. A. KING e t a / .

ecdysteroid UDP-glucosyl transferase (O'Reilly and Miller, 1989) which may reduce the effects of JHE. The lack of effect of the recombinant J H E on later larval instars is a disadvantage, since it is unlikely that neonates could be targeted consistently. However, it is likely that the effectiveness of JHE will be improved as more is elucidated of its function in the natural insect. The most significant improvements in the effectiveness of baculoviruses have been achieved by inserting insect-specific toxin genes into the virus genome. In two similar studies (Martens er al., 1990; Merryweather et al., 1990) the Bacillus thuringiensis delta endotoxin gene was inserted into the baculovirus genome. This bacterial toxin is produced as a 130 kDa protoxin, in a crystalline form. When ingested by larvae, the protoxin dissolves in the alkaline midgut and is cleaved to an active 62 kDa component (reviewed by Hofte and Whiteley, 1989). This toxin binds to cells lining the gut wall and is thought to introduce pores into the plasma membranes, causing disruption of the osmotic balance and cell lysis. The immediate effect of this process is a cessation of feeding (Heimpel and Angus, 1959). The DNA encoding the B. thuringiensis delta endotoxin was inserted into AcMNPV in lieu of the polyhedrin gene coding region, but under the control of the polyhedrin gene promoter (Martens et al., 1990; Merryweather et al., 1990). A polyhedrin-positive AcMNPV containing the bacterial sequence was produced by inserting the endotoxin coding region under the control of the virus p10 gene promoter (Merryweather et al., 1990). Polyhedrin-negative and polyhedrin-positive viruses produced biologically active endotoxin, as evidenced by a reduction in feeding when diet was contaminated with recombinant virus-infected cell extracts. Although functional toxin was produced in virus-infected cells, this was not accompanied by a reduction in lethal dose (LD,,) when purified recombinant virus polyhedra were assayed in T. ni larvae. Further tests showed that toxin was produced in the virus-infected insects, but it is likely that this material could not access the midgut for proteolytic cleavage and conversion to the active form. Provision of a signal peptide sequence to a truncated form of the toxin does result in some secretion of toxin in virus-infected cells in culture, but has no effect on the biological activity of the virus (A. T . Merryweather and R. D. Possee, unpublished data). While the use of the B. thuringiensis delta endotoxin remains attractive, its production by baculoviruses in insect larvae has yet to show any significant advantage over the unmodified virus. This is unfortunate, since the B. thuringiensis itself is used as a biological control agent and is widely accepted as a safe formulation. The most successful examples of genetic modification of baculoviruses have utilized insect-specific neurotoxin genes. Stewart et al. (1991) inserted a copy of the Androctonus australis (North African scorpion) insect-specific neurotoxin coding region into a polyhedrin-positive AcMNPV, under the control of the p10 gene promoter. This toxin affects the sodium conductance

ADVANCES IN INSECT VIROLOGY

19

of neurones, producing a presynaptic excitatory event, leading to paralysis and death (Walter et al., 1976; Teitelbaum et al., 1979). When produced by the recombinant virus in T. ni larvae, the insects consumed 50% less diet in comparison with unmodified virus-infected controls (Stewart et al., 1991). In other tests, the LDSowas reduced slightly and there was a 25% decrease in the survival time (ST,,). These effects were only observed if the neurotoxin was produced as a fusion protein with the AcNPV glycoprotein (gp) 67 coding sequence (Whitford et al., 1989). Clearly, the neurotoxin must enter the secretory pathway of the cell to attain full biological activity. Other studies using the same scorpion toxin have reported essentially similar results (Maeda et al., 1991; McCutchen et al., 1991). Another insect-specific neurotoxin gene used to enhance the effectiveness of baculoviruses was isolated from Pyernotes tritici (mite) by Tomalski and Miller (1991). This was inserted into the AcMNPV genome under the control of the polyhedrin promoter to derive a polyhedrin-negative virus (Tomalski and Miller, 1991). A polyhedrin-positive virus was also produced by placing the neurotoxin coding region upstream of the native polyhedrin gene, under the control of a hybrid gene promoter comprising elements of both late and very late transcription units (Tomalski and Miller, 1992). Both reports recorded that feeding the recombinant viruses to T. ni larvae resulted in significantly earlier mortality, reduced feeding and weight gains and paralysis. Undoubtedly, the isolation of other insect-specific toxins will lead to the development of recombinant baculoviruses with even better insecticidal activities than those reported by Stewart et al. (1991) and Tomalski and Miller (1991, 1992). It is worth considering, however, that as recombinant viruses which have very rapid effects on the insect host are produced, it will become problematic to amplify those same viruses to levels sufficient for field application. Stewart et a[. (1991) recorded a decrease in virus yield from insects infected with the AcNPV containing the scorpion toxin gene. This decrease was not sufficient to prevent production of enough virus for laboratory use and a future small-scale field trial but it signalled that attention should be directed to this phenomenon. It may be necessary to regulate the production of the foreign protein in the virus amplification stage and then release the control in the field application. Alternatively, insect hosts which are resistant to the effects of the toxin may be used as a virus production facility. 2.7.4

Other techniques for improving baculovirus insecticides

Before the introduction of techniques for the genetic modification of baculoviruses various strategies were evoked to improve the effectiveness of the virus preparation. Despite the protection afforded the virus particles by the occlusion body, solar ultraviolet light can rapidly inactivate virus

20

L. A. KING e t a / .

infectivity, particularly in tropical climates. Brassel and Benz (1979) selected a strain of the codling moth GV with improved resistance to ultraviolet light. In laboratory tests this virus was 5.6 times more resistant to artificial ultraviolet irradiation and retained infectivity twice as long in the field. Wood et al. (1981) conducted studies to increase the virulence of AcMNPV by replication of wild-type virus in the presence of 2-aminopurine. The lethal time 50 of a mutant, designated HOB, was significantly shorter than the parental virus. Fifth instar larvae infected with HOB gained weight at a lower rate than the unmodified virus. Hughes et al. (1983) compared the time-mortality response of H . zea to 14 isolates of H . zea NPV. Those isolates with genomes producing identical profiles in agarose gels after digestion with restriction endonucleases, did not differ significantly in the time required to kill the insect host. Isolates producing dissimilar restriction endonuclease cleavage patterns had significantly different STSovalues. That baculovirus isolates do not consist of genetically homogeneous populations has been confirmed in various studies. For example, Smith and Crook (1988) were able to separate eight distinct genotypes from Arfogeia rapae larvae by low mortality dose infections and three genotypes by the same method using L. dispar larvae. The biological activity of these viruses was not assessed. Weitzman et al. (1992) characterized two variants (PfNPV(A) and (B)) of P . flammea NPV isolated from a population of wild-type virus. These variants had similar restriction endonuclease cleavage patterns, but displayed interesting properties when propagated together in two different insect hosts. In P . flammea larvae, PfNPV(B) formed at least 80% of the virus population. In M . brassicae larvae, PfNPV(A) predominated and attained 80% of the population after only three successive passages. This suggests that, according to the host insect used to amplify the viruses, a ceratin genotype has an advantage. The significance of this for field control programmes is that propagation of a mixed genotype virus population in one host insect for later application to control another species in the field may not derive the most efficient variant. 2.7.5 Assessing the safety of genetically modified baculovirus insecticides Empirically, there is no difference in assessing the safety of unmodified or genetically modified baculovirus insecticides. Inevitably, the proposal to use the altered virus insecticides has provoked additional concerns for various parties. These issues have been discussed at length in previous publications (Bishop et al., 1988; Bishop, 1989; Possee et al., 1992, 1993; Vlak, 1993a,b) and will not be considered in the same detail here. The question most often raised is whether genetically modified baculoviruses have an altered (expanded) host range. Our knowledge of what determines baculovirus host range is very limited and so this must remain an important consideration. Prior to field release experiments, those insect

ADVANCES IN INSECT VIROLOGY

21

species most likely to be present at the site of the trial should be challenged with the virus in the laboratory to determine the outcome. These data may not be so easy to interpret, since it is unclear whether insects requiring very high doses of virus to initiate infection would encounter the same levels of virus in the field. A further concern is whether genetically modified baculoviruses might recombine with indigenous baculoviruses after co-infection of insect larvae and transfer the foreign gene. For this to occur, the two viruses would have to be present in the same cell within the insect. Recombination frequencies are affected by DNA similarity, so co-infection of insect cells would not guarantee the generation of stable virus recombinants. Recombination has been demonstrated to occur between recombinant and unmodified AcMNPVs after coinfection of T. ni larvae (A. T. Merryweather and R. D. Possee, unpublished data). The long-term persistence of the modified baculovirus in the environment has also to be considered. The baculovirus occlusion bodies may persist in soil for many years (Evans and Harrap, 1982). On leaf surfaces, the virus is more susceptible to inactivation by solar ultraviolet radiation. In the short term, field experiments should probably include the option of disinfection of the soil to prevent the formation of a reservoir of infectious viruses. 2.7.6 Past field release experiments The considerable number of laboratory studies to assess the feasibility of improving baculovirus insecticides using genetic engineering techniques has not been matched by subsequent field trials to monitor their effectiveness in the environment. This is understandable, considering the cost of field trials and the trepidation with which such experiments are viewed by scientists and the public alike. Each field trial involves several years of planning, only one experiment can usually be accomplished in a season and the results require careful interpretation afterwards. The first field trial with a genetically modified baculovirus was conducted in 1986 in Oxford (Bishop, 1986; Bishop et al., 1988). A genetic marker was inserted into a non-coding region of the AcMNPV genome, downstream of the polyhedrin coding region. This marker served to distinguish the virus from parental virus and other, unrelated baculoviruses. The virus was used to infect insect larvae in the laboratory. These individuals were then placed onto sugar beet plants within a netted field facility to examine persistence of the genetically modified virus. The virus-infected larvae died within a 7-day period. Marked virus could be recovered from the site up to 6 months later, whereupon the plants were cleared and the soil disinfected. Later experiments, between 1987 and 1989, used AcNPV mutants lacking the polyhedrin gene, but containing another unique genetic marker or a functional lac2 gene from Ewherichia coli. These experiments served to demonstrate that

22

L. A. KING et a/.

viruses lacking the protection of the polyhedrin-based occlusion body cannot persist in the environment for significant periods. They suggest that if a baculovirus insecticide could be modified to package virus particles within polyhedra in the laboratory, but only produced polyhedrin-negative viruses in the field, this would constitute a very safe biological insecticide. The only other field trial with a genetically modified baculovirus has also made use of an AcMNPV mutant deficient in polyhedra production. Hamblin et al. (1990) described the process of co-occlusion of a polyhedrinnegative virus by an unmodified AcMNPV after co-infection of insect cells in culture or in larvae with the two viruses. After several passages in insects, the polyhedrin-negative virus was gradually lost from the population. This study has recently been extended to the field, with spray application of virus to a cabbage plot artificially infested with T. ni larvae (H. A. Wood, personal communication). Monitoring of the site is continuing over a number of years to determine the levels of polyhedrin-negative virus. 2.7.7 Future experiments The field release experiments with genetically modified baculoviruses performed to date have utilized viruses harbouring relatively simple, innocuous changes to the virus genome. These changes were not intended to improve the efficacy of the virus insecticide. Future field tests will involve the use of viruses containing foreign genes encloding insect-specific hormones, enzymes or toxins (see Section 2.7.3). These viruses are known to kill the insect host more rapidly and to reduce the feeding damage to the plant. Laboratory tests must be confirmed by authentic field trials if the early promise shown by these agents is to be confirmed. 2.8

BACULOVIRUS EXPRESSION VECTORS

The expression of foreign genes in insect cells using baculovirus expression vectors is now a well-established technology that is widely used by many investigators (reviewed by Luckow and Summers, 1988; Miller, 1988, 1993; Cameron et al., 1989; Maeda, 1989a; Atkinson et al., 1990; Bishop and Possee, 1990; Possee et al., 1990; King e f al., 1992; King and Possee, 1992; O’Reilly et al., 1992). There are several advantages in the use of baculoviruses instead of other eukaryotic expression systems: the expression vectors do not require a helper virus system; the viruses are non-pathogenic to vertebrates and plants; and no oncogenic or transforming elements are employed. It has also been demonstrated that insect cells will perform a wide range of post-translational processing events that are often required for biological activity of a recombinant protein. One other advantage which has not been well documented is the application of the baculovirus expression vector system to the study of insect

ADVANCES I N INSECT VIROLOGY

23

proteins. This increases the possibility that the protein of interest will retain the range of properties associated with its native counterpart. This review will describe some of the more recent advances in baculovirus expression vector technology.

2.8.1 The development of baculovirus expression vectors Most baculovirus expression vectors make use of the very late polyhedrin gene promoter of AcMNPV to drive the expression of the foreign gene in virus-infected S. frugiperda cells (Smith et al., 1983b). The polyhedrin gene promoter is very strong and can result in the production of very large quantities of the foreign protein in virus-infected cells. An alternative expression vector system is based on the homologous gene in the B. mori (silkworm) NPV, but this has not found such wide application as AcMNPV vector system (Maeda et al., 1985; Maeda, 1989a; Iatrou and Meidinger, 1990). The polyhedrin gene of AcMNPV has been mapped to the EcoRI ‘I’ fragment of the virus genome (Smith ei al., 1983a) and sequenced (Hooft van Iddekinge et al., 1983). It was subsequently demonstrated that the polyhedrin gene could be deleted, preventing formation of virus occlusion bodies, without affecting the production of ECV (Smith et al., 1983b). This also provided a useful phenotypic marker for recognizing recombinant viruses in plaque assays. These features of the polyhedrin gene, including a strong promoter, redundant function of the protein and easily recognizable phenotype, made it an attractive proposition for development as a high level eukaryotic expression system. A complication, however, was the fact that the AcMNPV genome is quite large (approx. 133 kbp), and difficult to modify by conventional ligation with foreign DNA. Consequently, indirect methods are used to modify the baculovirus genome. These involve construction of bacterial plasmids, or transfer vectors, containing portions of the AcMNPV genome, which encompass and flank the polyhedrin gene. The EcoRI ‘I’ fragment has formed the basis for most AcMNPV transfer vectors (Possee et al., 1991). Within a transfer vector, the polyhedrin gene coding sequences are either partially or completely deleted and replaced with a recognition site for a restriction endonuclease. The polyhedrin gene promoter and transcription termination signals are left intact. The foreign gene coding region, with its own translation initiation signal, is inserted into the transfer vector into the correct orientation. The recombinant virus is produced by co-transfection of insect cells with the transfer vector and purified ‘wild-type’ AcMNPV DNA. The foreign gene is inserted into the AcMNPV genome by homologous recombination between identical sequences flanking the native polyhedrin gene in the virus and the foreign DNA in the transfer vector. The net result is the production of a virus which is able to replicate in insect cells but lacks the ability to

24

L. A. KING et a/.

produce polyhedra. The virus progeny from the co-transfection are then titrated in a plaque assay in S. frugiperda cells and polyhedrin-negative plaques identified by visual inspection with the aid of a microscope. The success of the AcMNPV expression vector system has resulted in the derivation of many alternative transfer vectors for introducing foreign coding regions into the baculovirus genome. For the inexperienced user of the system it can be difficult to choose an appropriate plasmid. It is advisable to consult the two practical guides to the baculovirus expression system currently available (King and Possee, 1992; O’Reilly et al., 1992) for guidance in selecting a transfer vector. Briefly, transfer vectors may have varying amounts of virus sequence flanking the foreign coding region; it is often easier to make insertions in the smaller vectors. The bacterial plasmids may have the M13 intergenic region, facilitating production of singlestranded DNA in bacteria for sequencing and rapid mutagenesis. Finally, more than one restriction enzyme site may be available for the insertion of the foreign DNA. This renders some transfer vectors compatible with DNA fragments possessing a variety of cohesive or blunt ends. Although most foreign genes are expressed using the polyhedrin genebased vectors, other baculovirus gene promoters may be used. The p10 gene coding region may also be removed from the virus genome without affecting either ECV or polyhedra production (Vlak et al., 1988; Williams er al., 1989; Weyer et al., 1990). The p10 gene product is, like polyhedrin, non-essential in the production of virus particles (Vlak et al., 1988). Transfer vectors analogous in structure to the polyhedrin gene-based vectors described above are constructed, prior to the insertion of the foreign coding region under the control of the p10 gene promoter. Recombinant viruses are produced by co-transfection of insect cells with plasmid and virus DNA. A complication when using the p10 gene locus to insert foreign DNA into the virus genome is that, unlike polyhedrin, the p10 protein does not produce a recognizable phenotype in virus-infected cells. Fortunately, recent advances in recombinant virus selection have solved this problem (see Section 2.8.3). Foreign genes are inserted directly under the p10 promoter by co-transfection of insect cells with a p10 transfer vector and purified wild-type AcMNPV DNA. Other AcMNPV gene promoters may also be used for the expression of foreign genes in virus-infected cells. Several late gene promoters, while not as active as the very late polyhedrin and p10 gene promoters, can produce useful quantities of recombinant material. One problem encountered when using the late gene promoters is that they are normally associated with genes responsible for producing virus particle structural proteins. In consequence, replacement of the native virus coding sequence with a foreign coding region is unlikely to result in the production of a viable recombinant virus. For example, the AcMNPV basic or arginine-rich protein is associated with virus DNA within the nucleocapsid. In order to use this virus gene promoter as an

ADVANCES IN INSECT VIROLOGY

25

expression system, it was necessary to insert a copy of the promoter in lieu of the polyhedrin gene promoter (Hill-Perkins and Possee, 1990). Production of bacterial P-galactosidase by the basic protein gene expression vectors was significantly lower than when the polyhedrin gene promoter was used, but initiated from about 8 hpi. This was 6 h before the polyhedrin gene promoter attained a reasonable level of activity. Other late gene promoters used in a similar manner include the 39K capsid protein gene promoter (Thiem and Miller, 1989) and the gp67 gene promoter (A. T. Merryweather and R. D. Possee, unpublished data). The earlier production of foreign proteins in virus-infected cells may be advantageous if they require extensive post-translational modifications (see Section 2.8.4).

2.8.2 The development of multiple expression vectors More than one foreign protein may be produced in insect cells simply by co-infection with two or more recombinant viruses. This approach was used to express simultaneously three influenza virus proteins (St Angelo et al., 1987). A more efficient and reproducible way to achieve co-expression, however, is to insert each foreign gene into the same recombinant virus. Emery and Bishop (1987) inserted a copy of the polyhedrin gene promoter and putative transcription termination signals, upstream of the native polyhedrin gene. This enabled the expression of a foreign gene in addition to the virus polyhedrin. Subsequent modification of this vector permitted the co-expression of the hepatitis B virus surface and core antigens in baculovirus-infected cells (Takehara et al., 1988). Each foreign sequence was placed under the control of the native or duplicated polyhedrin gene promoter. Similar expression vectors were derived by using a combination of the polyhedrin and p10 gene promoters (Weyer and Possee, 1991). A copy of the p10 gene promoter was inserted upstream of the polyhedrin gene promoter. The influenza virus haernagglutinin o r neuraminidase gene was placed under the control of each promoter and co-synthesis achieved in recombinant virus-infected cells. Baculovirus expression vectors are not limited to the production of two foreign proteins in insect cells. The synthesis of non-infectious virus-like particles of bluetongue virus by the simultaneous expression of four structural proteins, by the co-infection of two dual recombinant baculoviruses has been described (French et al., 1990). Five bluetongue virus structural proteins have been co-expressed within the same cell by coinfection of two dual recombinants and one single recombinant virus (Loudon and Roy, 1991). The baculovirus expression system, therefore, is ideally suited to the study of protein-protein interactions.

26

L. A. KING eta/.

2.8.3 Improved methods for the selection of recombinant baculoviruses Foreign genes have been inserted into the AcMNPV genome at the polyhedrin locus by co-transfection of insect cells with the appropriate transfer vector and purified wild-type AcMNPV DNA. Recombinant virus is separated from parental virus by titration in a plaque assay and selection of polyhedrin-negative plaques after visual inspection. This method, while very simple, has presented problems to many users of the system. Polyhedrinnegative plaques can be difficult to identify, necessitating many hours of tedious searching. Various methods have been developed to simplify the recombinant virus selection procedure. The virus progeny from a co-transfection may be titrated in a microtitre plate containing insect cells. After a suitable period, the medium is removed from each well and retained. DNA is extracted from the cells remaining in the well and the presence of the foreign gene sequences monitored using simple dot blot hybridization techniques. Those wells eliciting positive results provide the means to select the stored medium samples for further analysis. These samples usually contain enhanced numbers of polyhedrinnegative viruses, which may be more easily identified in subsequent plaque assays. Vialard et al. (1990a) produced polyhedrin gene-based transfer vectors with a copy of the bacterial P-galactosidase gene, under the control of a baculovirus gene promoter, inserted upstream of the polyhedrin gene promoter used for foreign gene expression. Recombination between the transfer vector and AcMNPV DNA produced recombinant viruses with both P-galactosidase and the foreign gene, facilitating selection of blue plaques in a titration stained with X-gal. A similar approach has been used to effect recombinant virus selection when inserting foreign genes at the p10 gene locus (Vlak et al., 1988). In a radically different approach, the poor infectivity of linearized virus DNA has been used as a way of selecting recombinant viruses. Kitts ef af. (1990) observed that when AcMNPV DNA, linearized at the polyhedrin gene locus (using Bsu36I), was co-transfected with a plasmid transfer vector, 30% of the virus progeny contained the foreign gene. The transfer vector appeared to ‘rescue’ the linear virus DNA by recircularization. Fortuitously, the E. coli P-galactosidase coding region contains the appropriate restriction enzyme site (Bsu361). Recombinant virus DNA containing this sequence at the polyhedrin or p10 gene loci may be linearized with Bsu361, mixed with the appropriate transfer vector and used to co-transfect insect cells. Progeny virus, containing the foreign DNA in place of the P-galactosidase sequences, may be selected as colourless plaques in the presence of X-gal at a frequency of about 30%. This system, using linearized virus DNA, has been improved by inserting additional Bsrr36I sites in the virus DNA harbouring the P-galactosidase

ADVANCES IN INSECT VIROLOGY

27

coding region under the control of the polyhedrin gene promoter. The first was inserted within a non-essential virus gene of unknown function upstream of the polyhedrin gene promoter (ORF 603; Gearing and Possee, 1990). The second was inserted within an essential gene, also of unknown function (ORF 1629; Possee et a f . , 1991), downstream of the P-galactosidase coding region. The sequence of the second Bsu36I recognition site was designed to preserve the coding region of ORF 1629. This virus was designated BacPAK6 (Kitts and Possee, 1993). Digestion of the virus DNA with Bsu36I removes part of ORF 603, the polyhedrin gene promoter, the P-galactosidase coding region and part of ORF 1629. When the Bsu361digested BacPAK6 virus DNA was used to transfect insect cells, very low recoveries of infectious virus were attained. Recircularization of linear BacPAK6 DNA produces a crippled virus which is unable to produce infectious virus in insect cells. If the same DNA was co-transfected with a plasmid transfer vector, however, virus yields were enhanced considerably with nearly 100% recovery of recombinant virus. Very few parental (BacPAK6) viruses are evident in the first round of plaque purification, enabling rapid isolation of recombinant viruses. The co-transfection of linear BacPAK6 DNA with the transfer vector serves to repair the partial deletion in ORF 1629 and permit the production of infectious virus.

2.8.4 Post-translational processing in insect cells Post-translational processing plays an important role in determining the biological activity of a recombinant protein. The range of post-translational processing events in insect cells have been widely documented. These include glycosylation, phosphorylation, proteolysis, ADP-ribosylation, sulphation, acylation and disulphide bond formation (reviewed by Miller, 1988; Luckow and Summers, 1988; Maeda, 1989a, Atkinson et af., 1990; Bishop and Possee, 1990; King et al., 1992; King and Possee, 1992; O’Reilly et a f . , 1992). Of the post-translational processing events listed, insect cells appear to possess a different glycosylaction pattern to that associated with other systems. It is known that the glycosylation sites utilized in insect cells are the same as in mammalian cells (Hsieh and Robbins, 1984), i.e. asparagine residues, and that the most common type of glycosylation is N-linked and can be inhibited by tunicamycin (Kelly, 1982). The principal difference is in the nature of the oligosaccharides added to these sites. Insect cells appear to lack galactose and sialic acid transferases and trim the oligosaccharide to a short core containing mannose, whereas in mammalian cells the central core is extensively modified (Kuroda et al., 1990). Jarvis et al. (1990a) performed a study on the role of glycosylation in the transport of recombinant glycoproteins through the secretory pathway of insect cells following treatment with tunicamycin or castanospermine (an inhibitor of the initial steps of N-linked oligosaccharide processing). They demonstrated that

28

L. A. KING eta/.

tunicamycin treatment inhibits the secretion or cell-surface expression of some but not all glycoproteins. Castanospermine treatment did not inhibit the secretion or cell-surface expression of any of the glycoproteins tested. This suggests the possible role of N-glycosylation but not oligosaccharide processing in the transport of some recombinant proteins through the secretory pathway of insect cells. The differences in glycosylation appear to have no effect on the overall function of the recombinant protein. It has been shown that foreign proteins synthesized in insect cells can be targeted to the nucleus (Forstova et a f . , 1989), the cytoplasm (Jeang et al., 1987), to the cell surface (Possee, 1986; Greenfield et al., 1988; Atkinson et al., 1992) or for secretion (Jarvis and Summers, 1989). The baculovirus expression vector system has been used to express proteins from viral, fungal, plant, protozoan, invertebrate and vertebrate origins. Historically, two main cell lines have been used for the baculovirus expression system, S. frugiperda IPLB (Sf21) (Vaughn ei a f . , 1977) and Sf9 (Smith et a f . , 1983b). More recently, it has been documented that other insect cell lines may also be useful. These include T. ni (Tn368) and T, ni High Five (TnSBl-4) (Invitrogen), M . brassicae (Mb) (King et a f . , 1991) and Estigmene acrea (A. L. Lawrie and L. A. King, unpublished data). In some situations, e.g. the expression of recombinant neurotransmitter receptors, Sf9 cells are used in preference to Sf21 cells, as the electrophysiological techniques employed to study these receptors proved difficult using Sf21 cells (King et al., 1992). As mentioned above, the processing of glycoproteins in insect cells results in the addition of mainly mannose-rich side chains which are not further trimmed to form complex oligosaccharides (Kuroda et al., 1990). This observation produces recurrent criticism of the baculovirus expression system because the recombinant material is not identical to the native protein. The glycosylation patterns of influenza haemagglutinin have been compared in Sf9, Tn368 and Esiigmene acrea (Ea) cells (Klenk ei al., 1992). It was shown that the majority of the side chains attached to the haemagglutinin in the Sf9 and Tn368 cells were processed from oligomannosidic to truncated trimannosyl cores. These results were consistent with those of Kuroda et a f . (1990). However, in the Ea cells it appeared that the trimannosyl cores were elongated by the addition of N-acetylglucosamine. Such processing of complex oligosaccharides has not been reported in any other insect cell line to date. The secretion of recombinant human urokinase has also been compared in different insect cell lines: Sf21, Mb, Ea and Tn368 (A. M. Lawrie and L. A. King, unpublished data). In this study the Tn368 and Mb cell lines produced significantly higher yields of intracellular and secreted urokinase than did the Sf21 or Ea cells. Interestingly, secretion of urokinase in the Ea cells was not detected until 48 hpi, whereas in the other cell lines tested, it was apparent in the culture medium after 18-24 hpi.

ADVANCES IN INSECT VIROLOGY

29

2.8,6 Expression of foreign genes in insect cell lines Baculovirus vectors can only produce transient expression of foreign genes in insect cells because the virus infection results in the death of the host cells, after about 72 hpi. This may partly account for the reason that membrane-targeted and secreted proteins are relatively poorly synthesized in recombinant virus-infected cells, as undoubtedly the secretory pathway is compromised very late in the infection cycle. A recent study involved the expression of the bovine GABAA receptor a1- and pl-subunits in insect cells and their subsequent analysis using electrophysiological techniques (Atkinson et al., 1992). It was found that the efficiency of patch clamping to the insect cell plasma membrane was greatly reduced following virus infection. Therefore, in some cases it may be advantageous to use a stable expression system utilizing non-infected cells. It has been demonstrated that the insect cell can be made to continuously express foreign genes by making use of the AcMNPV IE-1 gene promoter incorporated into the insect cell genome (Jarvis et al., 1990b). Since the IE-1 gene promoter is transcriptionally active in the absence of any other viral gene product (see Section 2.6.1), it can be used to express foreign genes in transformed insect cell lines, thus allowing the foreign gene to be continuously synthesized. The stable and continuous expression of tissue plasminogen activator and E. coli /3-galactosidase (Jarvis et al., 1990b) have been described, although the levels of foreign protein synthesized were much reduced compared with those obtained via recombinant virus-infected cells. The expression of functional, GABA-gated homo-oligomeric GABAA receptors, by the integration of the bovine GABAA receptor pl-subunit cDNA under control of the IE-1 gene promoter, has also been described (Joyce et al., 1993).

3 Entomopoxviruses 3.1

ISOLATION A N D HOST R A N G E

Poxviruses in insects were originally described by Vago (1963) and since then entomopoxviruses (EPVs) have been isolated from over 60 different insect species in widespread geographical locations (Arif, 1984). EPVs have been found in all four economically important insect orders, Lepidoptera, Coleoptera, Diptera, and Orthoptera. Compared with several other groups of insect viruses, very little is known about their biochemical and biophysical nature or about the details of their replication cycle (Arif and Kurstak, 1991). In general, EPV infections of Coleopteran species exhibit an exceptionally prolonged course of disease development which may be as long as 30-40

L. A. KING et a/.

30

weeks depending on the temperature (Hurpin and Vago, 1963; Milner and Lutton, 1975). Development of the disease in the cockchafer Mefolontha melofontha (Mm) (Scarabaeidae) is slow, with larvae dying 5-6 months after inoculation. The external symptoms are not very marked; affected larvae are sluggish and lose turgidity. Internally, the adipose tissue and blood cells are primarily affected (Amargier et a f . , 1964; Bergoin et a f . , 1969). All larval stages, particularly the first two instars, are susceptible to infection (Roberts, 1970), and normal development of the insect during the pupal and imaginal moults is arrested. In Lepidopteran EPV infections, the course of disease is relatively short, usually less than 3 weeks. Infected larvae may become lethargic and lose coordination and mobility during the late stages of infection. Death of E. acrea larvae infected with Amsacta moorei (Am) EPV is frequently preceded by paralysis of the abdomen and by regurgitation or defecation of fluid containing virus. The primary site of virus multiplication is in the cytoplasm of fat body cells, although mid-intestinal cells, hypodermis, muscle cells, tracheoblasts, haemocytes and ganglion connective tissue are also affected (Roberts and Granados, 1968). The infection of Lepidopteran pupae by some EPVs has been reported (Retnakaran and Bird, 1972; Sutter, 1972), however, infections of the adult stage are not known. EPV-infected grasshoppers (e .g. Melanoguin sanguinipes, Ms) exhibit a general torpor, take longer to develop and show a high rate of mortality. The bodies of heavily infected nymphs are frequently distended with protruding cervical membranes due to the accumulation of virus in the fat body. The fat body of M . sanguinipes appears to be the only tissue affected by MsEPV. The outward signs of disease in the midge Chironomus luridus (CI) are striking; the entire body exhibits irregular whitish spots caused by massive accumulations of virus occlusion bodies in the fat tissue. Infected fourth instar larvae are of normal size and as active as healthy larvae, but usually die before the next moult (Gotz et al., 1969). The ClEPV appears to be polytrophic, affecting various tissues including fat bodies, haemocytes, epidermis, oenocytes, imaginal discs of the legs and genital organs, muscles, nerve cells, and the intestinal tract (Huger et a f . , 1970). Other chironomid EPVs show a high degree of tissue specificity and are known to infect only fat body and haemocytes (Stoltz and Summers, 1972). Entomopoxvirus-like particles have also been found in three species of bumble bee (Clark, 1982), mosquitoes and other water-borne insects (Lebdeva and Zelenko, 1972). 3.2

STRUCTURE A N D CLASSIFICATION

The EPVs present morphological and physicochemical characteristics that are similar to those of the orthopoxviruses of vertebrates, however they have one distinguishing feature, the virus particles are occluded in a

ADVANCES IN INSECT VIROLOGY

31

TABLE 2 Classification of the entomopoxviruses English vernacular name En tomopoxviruses (EPV) Poxvirus of Coleoptera Poxvirus of Lepidoptera and Orthoptera Poxvirus of Diptera

Taxonomic status (international name) Subfamily (Entomopoxvirinae) Probable Genus (Entomopoxvirus A) Probable Genus (Entomopoxvirus B) Probable Genus (Entomopoxvirus C)

Type species

Melolontha melolontha ( M m ) EPV Amsacta moorei (Am) EPV Chironomus luridus (CI) EPV

paracrystalline matrix, the spheroid. Spheroids appear to be analogous to the baculovirus polyhedra. Because of these similarities, the Entomopoxvirinae have been classified as a subfamily in the Foxviridae family (Francki et al., 1991). The Entomopoxvirinae has been subdivided into three probable genera, as described in Table 2. Historically, an EPV has been named after the host insect from which it was originally isolated. Like the orthopoxviruses, the EPV virion is brick-shaped or oval with sizes ranging from 150 to 470nm long, and from 165 to 300nm wide. Negatively stained virus particles exhibit a folded outer membrane to give the appearance of a mulberry-like surface (Westwood et al., 1964). These spherical folds vary in size depending on the virus species and measure approximately 40 nm for AmEPV and 22 nm for MmEPV (Granados and Roberts, 1970; Bergoin et al., 1971). In cross-section or when the negative stain penetrates the particle, the virion is shown to contain an electrondense core surrounded by a multilayer membrane. EPVs from Orthopteran and Lepidopteran hosts generally contain a cylindrical core and two lateral bodies, while those infecting Dipteran hosts contain a biconcave core and two well-developed lateral bodies. EPVs from Coleoptera contain a unilaterally concave core and one lateral body located in the cavity of the core (Granados and Roberts, 1970; Bergoin and Dales, 1971; Stoltz and Summers, 1972; Granados, 1973a). Three forms of EPV have been identified in virus-infected cells: non-occluded intracellular virus, extracellular released virus and occluded virus. These forms are described in more detail in Section 3.5. Virus occlusion bodies, or spheroids, are spherical or oval in shape and have a paracrystalline matrix which is primarily composed of a single polypeptide, spheroidin. Spheroids and spheroidin appear to be analogous to the polyhedra and polyhedrin protein, respectively, of the baculoviruses (see Section 2.1). Spheroids vary in size from 5 to 24 pm in diameter and contain embedded virus particles arranged in a random or radial orientation. The number of virions occluded is variable, and depends

32

L. A. KING e t a / .

on the host species and even the cell type within the same host. The size, form and distribution of the spheroids also depends on the species of EPV. They range in shape from nearly spherical (Caleopteran hosts) to nearly ellipsoidal (Lepidopteran hosts). Generally, in the infected tissues of Coleoptera, one or more large spheroids develop per cell but in the Lepidoptera, numerous smaller spheroids are found in each cell. The amino acid content of the spheroids of several different EPVs (including AmEPV, E. acrea (Ea)EPV, MmEPV and MsEPV) has been determined (Bergoin et af., 1970; Langridge and Roberts, 1982). It was demonstrated that the acidic amino acids (aspartic and glutamic acids) and the basic amino acids (lysine and arginine) were present in approximately equimolar amounts. The sulphur-containing amino acids, cysteine and methionine, in AmEPV, EaEPV and MsEPV constituted 996, 8.1% and 3.7% of the total amino acids, respectively. From these data, it is likely that the need for disulphide bond-reducing agents during the alkali dissolution of spheroids is due to the formation of disulphide bonds between sulphurcontaining amino acids in the paracrystalline matrix of the spheroid structure. Virions have been reported to contain between 24 and 40 polypeptides, depending on the virus isolate (Bergoin and Roberts, 1971; Langridge and Roberts, 1982), with sizes ranging from 12 to 250 kDa (Bilimoria and Arif, 1980). This represents approximately 38% of the total coding capacity of the genome. Virion cores of a Choristoneura sp. EPV, isolated by treatment with the non-ionic detergent NP-40, have been shown to contain only one major protein VP59 (Bilimoria and Arif, 1980). Four enzymatic activities have been associated with the virion particles of AmEPV: a nucleotide phosphohydrolase (Pogo et af. 1971), acidic and neutral DNAases (Pogo et al., 1971), and a DNA-dependent RNA polymerase (McCarthy et a f . , 1974). An endogenous alkaline proteolytic activity has also been reported to be associated with occlusion bodies isolated from insect larvae. Spheroidin was degraded from a 102 kDa protein to a 52 kDa polypeptide and eventually into smaller polypeptides when the spheroids were dissolved in alkali (Bilimoria and Arif, 1979). Tissue culture-derived spheroids did not exhibit any alkaline protease activity (Langridge and Roberts, 1982). The alkaline protease appears to be similar to the enzyme associated with the matrix protein of baculovirus polyhedra (Epstein and Thoma, 1975; Zummer and Faulkner, 1979). The EPV genome consists of a linear, dsDNA molecule (Gotz et a f . , 1969; Granados and Roberts, 1970; McCarthy et af., 1974) which constitutes about 5% of the viral particle. The genome of AmEPV has been shown to have terminal hairpin loops (Hall and Hink, 1990), that are similar to those found in the orthopoxviruses, and is reported to be 225 kbp by pulse field gel electrophoresis (Hall and Moyer, 1991). The G + C content of EPV DNA is low, varying from 17% to 27% (Arif,

ADVANCES IN INSECT VIROLOGY

33

1976; Langridge et al., 1977), when compared with the G + C content of orthopoxviruses, which ranges from 32.5% to 39% (Joklik, 1962). The significance of this low G + C content is not known. 3.3

REPLICATION CYCLE IN INSECTS

Relatively detailed electron microscopic studies on the replication of EPVs in larvae have been carried out. Since these studies were the result of asynchronous infections, however, the sequence of morphogenic events has been deduced using the vertebrate poxviruses as a model. Most studies have been carried out with AmEPV infections of E. acrea larvae. Infection begins when larvae ingest viral occlusion bodies which dissolve in the alkaline environment of the gut to release virus particles. Granados (1973b) observed that AmEPV was first detected in the gut lumen 1-2 h after per 0s inoculation of E. acrea larvae. The viral envelope then fuses with the plasma membrane of microvilli and subsequently the viral core and lateral bodies enter the cell cytoplasm. This appears to be the normal mechanism of entry into cells, however viropexis has also been demonstrated when larvae receive an intrahaemocoelic injection of virus (Devauchelle er al., 1971). Following uncoating and a period of latency, cytoplasmic foci consisting of either electron-dense amorphous material (type I viroplasm), or aggregates of granular material interspersed with spherical vesicles (type I1 viroplasm) begin to appear in infected cells. The first recognizable viral structures are incomplete crescent-like shells or membranes appearing at the periphery of the virogenic stroma. These membranes develop and eventually enclose a mass of electron-dense material. Electron micrographs have shown that these immature particles consist of an inner trilaminar structure of unit membrane and a spicule coat (Stoltz and Summers, 1972). In type I1 viroplasms, crescent or arch-like envelopes are present in association with fibrillar material containing a large number of vesicles (Bergoin et a f . , 1969). The incomplete viral envelopes progressively close, and in the process, engulf the granulated material which eventually condenses to give the appearance of immature particles found with type I viroplasms. The material inside the particles begins to differentiate and a viral nucleoid forms as a condensed mass. As the nucleoid structure differentiates further into a mature core surrounded by three-layered membrane, the particle begins to assume a more rectangular shape with a concomitant loss of the outer layer of spicules. The lateral bodies also assume a more recognizable structural form. Later the outer membrane is modified by folding to give the appearance of a beaded mulberry-like structure (Devauchelle er al., 1971; Stoltz and Summers, 1972; Granados, 1973a; Bird, 1974; Bergoin et al., 1968a,b, 1969, 1971). The most predominant cytopathic feature of EPV-infected insects is the

L. A. KING e t a / .

34

formation of the large, ovoid occlusion bodies or spheroids within the cytoplasm of infected cells (see Section 3.2). It is likely that the spheroids serve a similar function to the baculovirus polyhedra in protecting virion infectivity during horizontal transmission outside the insect. In some EPV-infected cells, other homogeneous protein structures (spindles), which show a crystalline lattice in thin sections, have been identified. Spindles are smaller than spheroids, measure between 0.5 and 12 p m (Vago and Bergoin, 1968), and are devoid of virions; their function is unknown. The spindle protein is antigenically distinct from that of the virions or spheroids, and in MmEPV the spindle protein has been estimated to have a molecular mass of 50 kDa species (Gauthier et a f . , 1992). 3.4

MOLECULAR STUDIES

A restriction enzyme map has been generated for the AmEPV genome using the enzymes HindIII, BamHI and EcoRI (Hall and Hink, 1990), and showed no similarity to restriction enzyme maps of the vertebrate poxviruses. Extensive genomic heterogeneity was detected in the restriction endonuclease cleavage patterns of DNA from five EPVs (EaEPV, MsEPV, Othnonius batesi (Ob)EPV, CbEPV and AmEPV) and vaccinia virus, strain WR (Langridge, 1984). The first gene to be identified and sequenced from an EPV was the putative spheroidin gene from CbEPV (Yuen et a f . , 1990). The CbEPV spheroidin protein has been reported to be 100 kDa by SDS-PAGE analysis, however the predicted size of the protein following sequencing of the gene was found to be 37 kDa. The authors explained this disparity by suggesting that the spheroidin protein might be glycosylated to form a 50 kDa protein which forms dimers to give the expected molecular mass of 100 kDa. Analysis of the gene sequence suggested that a signal peptide was present at the 5' end and that the gene possessed homology to a baculovirus protein associated with occlusion bodies (Vialard et al., 1990b). A 100 bp sequence upstream of the putative transcription start site was inserted into a vaccinia virus expression vector and the sequence was found to act as an efficient promoter in a recombinant virus (Pearson et a f . , 1991). The spheroidin gene of the AmEPV was sequenced by Hall and Moyer (1991) and surprisingly had little significant homology with the sequence of the CbEPV spheroidin gene. The AmEPV spheroidin gene was identified as a 3.0kbp open reading frame (ORF) potentially encoding a protein of 114.8 kDa, which agreed well with the predicted molecular mass of 115 kDa from SDS-PAGE of purified occlusion bodies. The third spheroidin gene to be sequenced was from MmEPV (Sanz et al., 1992), and this study identified an O R F encoding 942 amino acids, corresponding to a potential polypeptide of 109 kDa. This agreed with the molecular weight of approximately 100 kDa from SDS-PAGE. This gene showed more than 40% amino acid

ADVANCES IN INSECT VIROLOGY

35

homology with the AmEPV, but no homology to the reported CbEPV spheroidin gene sequences. More recently, the spindle (fusolin) gene of MmEPV has been sequenced (Gauthier et al., 1992) and had an unexpected 55.6% amino acid homology to the CbEPV ‘spheroidin’ gene. In a recent study, Hall and Moyer (1993) reported that amino acid sequence data, generated from peptide fragments of purified spheroidin from CbEPV occlusion bodies, differed from that predicted from the reported DNA sequence (Yuen et al., 1990). The new data, however, showed over 80% homology to the predicted amino acid sequence of the AmEPV spheroidin gene (Hall and Moyer, 1993). The sequence data indicate that upstream of the AmEPV and MmEPV spheroidin gene ATG initiation codons is a very A + T rich region containing a TAAATG consensus motif, similar to that found in many late promoters of the vertebrate poxviruses. This sequence was shown by primer extension analysis to be the site of transcription initiation. Primer extension studies also demonstrated that the spheroidin gene mRNA contained 5‘ poiy(A) sequences typical of vertebrate poxvirus late transcripts (Hall and Moyer, 1991). The spindle (fusolin) gene of MmEPV has been identified and sequenced (Gauthier et al., 1992). PAGE analysis of purified spindles indicated a molecular mass of 50 kDa. The spindle gene was shown to encode an O R F consisting of 401 codons. The 5’ region of the gene was shown to act as an early promoter when inserted into a recombinant vaccinia virus. Antibodies generated to the MmEPV spindle protein did not bind to protein extracted from E. acrea insects infected with AmEPV, however they did bind to a 37 kDa protein isolated from CbEPV-infected insects. This suggests that while CbEPV and MmEPV both produce spindle proteins, they are not found in AmEPV-infected insects or cells (Gauthier et al., 1992). A thymidine kinase (TK) gene from AmEPV has been identified and sequenced (Gruidl et al., 1992). Analysis of the data revealed an ORF of 182 amino acids, encoding a polypeptide of 21.2 kDa. Amino acid homology comparisons indicated that the gene was most closely related to the TK genes of poxviruses (45%) and less so to the TK genes of vertebrates (40%). The TK from African swine fever virus (ASF) showed the least homology (31.4%) to the AmEPV TK gene, suggesting that these two viruses are not closely related, although ASF shares some biological features with poxviruses, and both ASF and AmEPV can replicate within arthropod hosts. More recently, Lytvyn et a f . (1992) reported the identification and sequencing of TK genes from C. furniferana (Cf)EPV, CbEPV and AmEPV. The three EPV TK genes were shown to be related, exhibiting 63.2% identity and 9.9% similarity at the amino acid level; only 36.7% identity and 13.6% similarity was observed when the EPV sequences were compared with the TK gene of vaccinia virus. Other partial sequence data for the AmEPV genome have identified a

L. A. KING e t a / .

36

potential nucleoside triphosphate phosphohydrolase I (NPH I) gene which exhibits close homology to the vaccinia NTPase and to the NPH I of CbEPV (Yuen et a f . , 1991). Immediately upstream of the AmEPV spheroidin gene, an ORF has been identified that has significant homology to the capripoxvirus HM3 ORF (Hall and Moyer, 1991). 3.5

REPLICATION I N VITRRO

At present only three EPVs have been reported to replicate in continuous insect cell lines; all have been isolated from Lepidopteran species and, therefore, no cell lines currently exist for the study of the replication of EPVs from Orthopteran, Coleopteran and Dipteran hosts. The ArnEPV has been shown to replicate in a number of cell lines, including E. acrea (EAA-BTI), L. dispar (LD-SDZl), L . dispar (LD-65Z), S . frugiperda (IPLB-Sf21), H . zea (IPLB-1075) and B. mori (Quiot et al., 1975; Granados, 1981; Langridge, 1983a,c; Goodwin et af., 1990; Marlow et al., 1992, 1993). The EAA-BTI cell line was established from primary cultures of E. acrea haemocytes (Granados and Naughton, 1975). Various strains of the L. dispar cell line IPLB-LD-65 (listed in Goodwin et a f . , 1978) have been shown to be capable of supporting the complete replication cycle of AmEPV. Of the different IPLD-65 strains, the IPLB-LD-652 line was reported to contain the highest percentage of cells supporting virus replication. A complex, serum-free medium has been developed for the replication of AmEPV in this cell line. Interestingly, it was found that the sterol concentrations used in serum-free media developed for baculovirus replication studies were not sufficient to support the full replication of AmEPV (Goodwin et a f . , 1990), suggesting that EPV replication makes greater nutritional demands on infected host cells during the replication than do baculoviruses. In addition to the studies on AmEPV replication in v i m , Pseudafetia separata (Ps)EPV has been demonstrated to replicate in two Lepidopteran cell lines isolated from P. separata and B. mori larvae (Hukuhara et af., 1990). A tentative sequence of viral replication events was proposed that was similar to that made by Granados and Roberts (1970) and Devauchelle et af. (1971) for ArnEPV. Adoxphyes orana (Ao)EPV has been shown to replicate in a cell line derived from newborn larvae of A . orana fasciata and Hornona magnanirna (Sato, 1989). Most of the studies on EPV replication in vitro have, however, been performed with AmEPV in either the EAA-BTI or LD-652 cell lines. In EAA-BTI cells, AmEPV DNA synthesis can first be detected between 6 and 12 hpi, with a period of rapid DNA synthesis from 12 to 24 hpi (Langridge, 1983a). The synthesis of viral structural proteins begins at about

ADVANCES IN INSECT VIROLOGY

37

18 hpi with progeny virus detectable after 20 hpi. Extracellular virus and occluded virus are first detected at about 18 hpi and the biosynthesis of virus structural proteins increases rapidly from 18 to 34 hpi. The viral replication cycle resembles that of other poxviruses, except for the appearance of occluded virus late in the infection. Virus-induced rounding of EAA-BTI cells is one of the early features of AmEPV infection, and suggests that virus infection results in a reorganization of the host cell cytoskeleton. In a recent study using immunofluorescent staining of the tubulin and f-actin components of the cytoskeleton, Marlow et al. (1992) demonstrated that the microtubules begin to depofymerize between 12 and 24 hpi, and by 48-72 hpi have further contracted to form a reduced network around the main cell body. At the same time, the f-actin components become rearranged to form distinct foci and microspikes. By 96 hpi, both the tubulin and f-actin are detected in areas of the cytoplasm associated with virus assembly, and by 120 hpi, after the formation of spheroids, the tubulin and f-actin are reduced to sparse patches over the cell surface. Depolymerization of the microtubules by colchicine corresponded to the virus-mediated effects, and virus replication was shown to be unimpeded by colchicineinduced depolymerization. Treatment of the cells with aphidicolin and cycloheximide indicated that the effects on the cytoskeleton in virus-infected cells may have been mediated by both early and late genes (Marlow et al., 1992). Observations on the replication of AmEPV in LD-652 cells revealed similar cytoplasmic events as those observed in virus-infected E. acrea larvae (see Section 3.3) and in the BTI-EAA cell line (Goodwin et al., 1990). However, an absence of type I viroplasms was noted and this may be due to the fact that L. dispar (Lymantriidae) is an alien host species (e.g. the larvae of this species are only susceptible by intrahaemocoelic inoculation), the normal host being A . moorei and the experimental host being E. acrea (both Arctiidae). In EAA-BTI cells we found that mature virus particles were predominantly occluded into spheroids or were found as the non-occluded, intraceilular form; very little virus was released into the culture media. The latter observation probably explains the difficulty in obtaining reliable plaque assays with the EAA-BTI cell line (S. A. Marlow and L. A. King, unpublished observations). A few, limited experiments have been undertaken to discern the ability of EPVs to replicate in vertebrate cells. In murine L-929 cells inoculated with AmEPV, no virus-induced proteins were detected at 37°C using [35S]methionine pulse-labelling. However, in E. acre0 (EAA-BTI) cells inoculated with vaccinia virus, an increase in protein production was detected by ELISA using antiserum raised against purified vaccinia virus (Langridge, 1983b).

L. A. KING eta/.

38

3.6

BIOLOGICALCONTROL

Because of their similarities to the vertebrate poxviruses, few serious studies have reported the use of EPVs as biological control agents. In one study, three viruses, Wiseana nuclear polyhedrosis virus (WNPV) , Wiseana EPV (WEPV) and Wiseana granulosis virus (WGV) that infect the insects W. cervinafa, W. umbraculafa and W. signafa respectively, were compared for their effectiveness as control agents. WNPV, and to a lesser extent WEPV, were found to be effective in controlling their respective hosts to subeconomic levels (Crawford and Kalmakoff, 1975). Aerial application of an EPV and NPV against the spruce budworm at Chapleau, Ontario (Cunningham and McPhee, 1973) indicated that the level of insect control was greater than that achieved by just applying the NPV.

4 Iridescent viruses

The Iridoviridae are a group of large, icosahedral dsDNA viruses that replicate in the cytoplasmic compartment of infected cells (Vaughn, 1979). The viruses infect both vertebrates (amphibians and pleuronect fish) and invertebrates (insects, nematodes and crustaceans) (Kelly, 1985); this review will be confined to current advances made in our understanding of the insect-specific iridoviruses (IVs). 4.1

CLASSIFICATION,ISOLATIONA N D HOST RANGE

The invertebrate IVs are divided into two genera, the small (120-140 nm) IVs or Iridovirus genus and the large (180-200nm) IVs or Chloriridovirus genus (Table 3 ) . The type species of the small IV genus is Chilo iridescent virus type 6 (CIV) which infects the rice stem borer (Chilo suppressah) (Fukaya and Nasu, 1966; Devauchelle et al., 1985), and that of the large IV genus is mosquito iridescent virus type 3 (MIV) which was first isolated from Aedes faeniorhynchus (Clark ef al., 1965). Insect IV infections are characterized by a blue-green or lilac iridescence caused by the formation of crystalline arrays of virus particles in the host tissue (Smith, 1967). Historically, it is this opalesence that has been used as a diagnostic feature of IV infections (Tinsley and Kelly, 1970). In common with most insect viruses, IVs have been named according to the insect host of origin and the sequence of isolation, thus Tipulu IV type 1 (TIV) was the first IV to be discovered (Xeros, 1954; Smith, 1955), and to date about 32 different IVs have been isolated from insects in at least three different orders: Lepidoptera, Diptera and Cdeoptera (Kelly, 1985; Ward and Kalmakoff, 1991;

ADVANCES IN INSECT VIROLOGY

39

TABLE 3 Classification of the Iridoviridae English vernacular name Small iridescent insect viruses Large iridescent insect viruses Frog viruses

Taxonomic status (international name) Genus (Iridovirus) Genus (Chloriridovirus Genus (Ranavirus)

Members (type species first) Chilo iridescent virus (CIV) Insect IVs 1, 2, 6, 9, 10, 16-32 Mosquito iridescent virus (MIV) Insect IVs 3-5, 7, 8, 11-15 Frog virus 3 Frog viruses 1. 2 , 5-24, L2. L4, L5 Tadpole edema virus LT1-4, T6-T20 (Newts) T21 (Xenopus)

Francki et al., 1991). No in vivo replication of insect IVs has been demonstrated in vertebrates (Kelly and Robertson, 1973). Often just one, or a few, obviously infected insects have been recovered from populations of many thousands of apparently healthy individuals (Kelly, 1985). This low frequency of overt infection in the environment contrasts with the high infectivity of virus particles when injected into the haemocoel, and has led to the suggestion that natural transmission of the virus may occur through wounds, by cannibalism or via parasitic nematodes (Ward and Kalmakoff, 1991).

4.2

VIRUS STRUCTURE

Much of the information on the structure of IVs has come from studies on two members, CIV and TIV, and more recently from studies on IV22 (isolated from blackflies) (Devauchelle el al., 1985; Tajbakhsh and Seligy, 1989). Virus particles have an electron-dense core surrounded by an internal lipid-protein envelope and an outer icosahedral shell (Williams and Smith, 1958; Stoltz, 1973; Vaughn, 1979; Orange-Balange and Devauchelle, 1982a,b). This outer protein shell renders the insect IVs resistant to ether (Francki et al., 1991). The dense core contains the dsDNA genome, which in the case of CIV, is packed in a chromatin-like structure with six DNA binding proteins (Cerutti and Devauchelle, 1985). The DNA comprises 11-18% of the virus particle by weight (Vaughn, 1979). The internal lipid membrane is distinct in composition from host cell membranes, is approximately 4 n m thick, and is closely associated with the outer capsid shell (Kelly and Vance, 1973; Stoltz, 1973; Orange-Balange and Devauchelle,

L. A. KING eta/.

40

1982a,b; Kelly, 1985). Protein complexes have been shown to extend throughout the single unit membrane, attaching the core polypeptides to those in the outer capsid, thus probably contributing to the stable nature of these viruses (stable at pH 3-10 and at 4°C for several years) (Klump et al., 1983; Cerutti and Devauchelle, 1985). The outer capsid is icosahedral in shape (Williams and Smith, 1958) and is composed of a lattice of hexagonally packed subunits (Wrigley, 1969). Surface fibres of 3-5 nm in diameter and up to 150nm in length (or microfibrillar fringe) have been reported on the surface of some IVs, including MIV, CIV and IV29 (Willison and Cocking, 1972; Kelly, 1985). It has been suggested that these fibres might be responsible for the interplanar spacings in IV crystals (Kelly and Robertson, 1973). The number of polypeptides associated with the virus particle appears to vary with the method used to prepare purified virions. For CIV, the number has been estimated at 19, ranging from 10 to 213kDa, with the major 65 kDa capsid polypeptide (Kelly and Tinsley, 1972) accounting for 40% of the total virion mass (Moore and Kelly, 1980). The IV22 and TIV capsid proteins have an estimated mass of 48-50 kDa, again comprising about 40% of the total virion mass (Kelly, 1985; Cameron, 1990). Studies by Krell and Lee (1974) have indicated that none of the viral structural proteins are glycosylated. Several enzyme activities have been associated with IV particles including RNA polymerase, nucleotide phosphohydrolase, protein kinase and an alkaline protease (Kelly and Tinsley, 1973; Monnier and Devauchelle, 1980; Farara and Attias, 1983, 1986). The latter may be a larval contaminant as it has not been detected in virus propagated in v i m (Farara and Attias, 1983, 1986). The virus genome consists of one molecule of linear dsDNA with a molecular weight ranging from 100 to 160x lo6 Da (small IVs; Bellet and Inman, 1967; Delius et al., 1984; Ward and Kalmakoff, 1991) and from 160 to 185 x lo6 Da (large IVs; Kelly, 1981). The DNA of two insect IVs (CIV and IVS), in common with the vertebrate IVs, have been shown to be circularly permuted and to have direct terminal repeats accounting for up to 12% (CIV) and 25% (WIV) of the total genome (Delius et al., 1984; Ward and Kalmakoff, 1991). Tajbakhsh and Seligy (1989) have also reported that the TIV genome consists of the major linear dsDNA component and up to three other smaller DNA fragments; the significance of these smaller fragments is uncertain. 4.3

REPLICATION CYCLE

Very little is known about the cellular events that occur in insect IV-infected cells, and surprisingly little is known about the pathway of infection in insect larvae. The latter is probably because of the difficulties in establishing the mechanism of entry of the virus; the initial site of a natural infection is still

ADVANCES IN INSECT VIROLOGY

41

unclear (Kelly, 1985; Ward and Kalmakoff, 1991). After initial infection, the virus spreads throughout the insect tissues (especially the epidermis and fat body) and results in a massive accumulation of virus; 25% by weight of a larva infected with TIV (Williams and Smith, 1957). Within 7-10 days of infection larvae generally become flaccid, with death occurring soon afterwards. Much of what is known about the insect IV replication cycle has come from in vitro studies from either TIV infections in cultured E. acrea or Galleria mellonella haemocytes (Yule and Lee, 1973; Krell and Lee, 1974), or from CIV infections of Choristoneura fumiferana or Trichoplusia ni cells (McIntosh and Kimura, 1974; Cerutti et al., 1981). The gross cytopathological effects of IV infection are characterized by syncytia formation, the marked shut-off of host cell macromolecular (DNA, RNA and protein) synthesis and an increase in cellular DNA and RNA polymerases and thymidine kinase activities (Cerutti and Devauchelle, 1980, 1982; Devauchelle et al., 1985). A virogenic stroma appears in the cytoplasm that contains fine fibrils of viral DNA and proteins (Yule and Lee, 1973). More recently, Cameron (1990) identified three putative phases of protein synthesis in Spodoptera frugiperda cells infected with IV22; these were designated immediate early (until 3 hpi), early (3-9 hpi) and late (from 12hpi). One late polypeptide was shown to migrate with the IV22 major capsid protein. Very little has been published on the molecular aspects of insect IV replication. Instead, comparisons are always drawn with the more extensively studied replication cycle of the vertebrate IV, frog virus 3 (FV3); however, the relevance of the FV3 replication cycle to the insect IVs is not known (Kelly, 1985; Ward and Kalmakoff, 1991). The replication of FV3 has been recently reviewed and will only be briefly summarized here (Murti et al., 1985; Francki et al., 1991). Virus uptake into cells is by pinocytosis and incorporation into phagocytic vesicles, in which the virus particle is uncoated. Replication of FV3 requires a functional host cell nucleus (this requirement has not been conclusively ascertained for insect IVs) in which initial virus DNA replication occurs. After replication, the new DNA molecules migrate to the cytoplasm where further DNA copies are made. The switch to cytoplasmic DNA replication coincides with the synthesis of a virus-encoded protein. In the cytoplasm, the DNA molecules form long concatemers which are necessary for the regeneration of the DNA ends. Early mRNA transcripts are also synthesized in the nucleus using a virus-modified RNA polymerase, leading to early protein synthesis in the first 5 h after infection. The late proteins are produced after 8 h infection and are thought to be associated with DNA replication. Following DNA replication, in the later stages of infection virus release is by budding or lysis. Virus that buds from the cell surface acquires a plasma- or endoplasmic reticulum-derived envelope, although most virus appears to remain cell associated and unenveloped virions are infectious.

42

4.4

L. A. KING et a/. MOLECULAR STUDIES

As mentioned above, there have been few studies on the insect IVs at the molecular level. Physical (restriction endonuclease) maps have been generated for the DNAs of CIV (Soltau et al., 1987) and IV9 (Wiseana spp., WIV) (Ward and Kalmakoff, 1987, 1991), and Fischer et al. (1989) have reported the formation of a complete genome library for CIV, which should form the basis for further studies to examine the structural and functional properties of the IV genome. In addition, DNA-DNA hybridization techniques are now being used for the detection of new IV isolates (Ward and Kalmakoff, 1991) and comparison of restriction endonuclease (RE) profiles can be used to separate and distinguish different IV strains, as has been extensively documented for other insect DNA viruses such as the baculoviruses. The use of these techniques should improve the taxonomy of the IVs, since at present the typing of these viruses is based only upon iridescence, host of origin and date of isolation (Xeros, 1954; Fowler and Robertson, 1972; Batson et al., 1976). Hybridization techniques may also be used in future to study the routes of infection, secondary hosts and virus transmission. This should provide a more reliable indication of infection than the present methods which rely primarily on visual iridescence of infected insects. Three insect IV capsid proteins genes have been identified and sequenced: IV22 (Cameron, 1990), TIV (Tajbakhsh et al., 1990) and CIV (Smith et al., 1993). The predicted molecular mass of the capsid protein of IV22 from the sequence data, 51.9 kDa, is slightly higher than that predicted from SDS-PAGE analyses (49 kDa). In an earlier study, Moore and Kelly (1980) showed that the major capsid protein of three other insect IVs had an N-terminal proline and, therefore, it has been suggested that the capsid of IV22 may be post-translationally processed to give the apparent molecular mass identified on gels (Cameron, 1990). Comparison of the amino acid sequences of the capsid proteins of TIV and IV22 showed that 442 of the first 451 amino acids were identical, however, because of an insertion of an additional 44 bp in the IV22 sequence, the C-terminal sequences had no similarities (Cameron, 1990). A high degree (64.7%) of amino acid sequence identity was reported between the capsid protein of CIV and TIV/IV22 (Smith et al., 1993). It will be interesting to determine if other, more distantly related IVs, also have a conserved capsid protein gene.

4.5

BIOLOGICAL CONTROL

Insect IVs have not been generally considered as viable biological control agents, although they have attracted some interest as a potential means to control some Dipteran pests, an insect group not usually susceptible to the more commonly used baculovirus insecticides. In particular, there has been

ADVANCES IN INSECT VIROLOGY

43

interest in IV22, isolated from the larvae of a blackfly (Sirnufiurn spp.), as blackflies are major vectors of human disease in tropical countries, notably river blindness (T. Williams, personal communication). Part of the reluctance to consider IVs as control agents is the lack of fundamental understanding of their life cycle, in particular how they infect their hosts. Some studies have shown that infection rates in the laboratory may reach levels of 70-90% (Chapman et af., 1971) and it is difficult to correlate this with the proposed mechanisms that exist in the field (i.e. wounding, cannibalism or parasitic inoculation). Use of DNA hybridization techniques to identify virus-infected insects, rather than visual iridescence, has recently been used in one study to show that 100% of G. melfoneffalarvae were susceptible to infection with WIV; none of the larvae showed signs of iridescence (Ward and Kalmakoff, 1991). Thus, IV infections in natural populations may be higher than suggested in the early literature, where infection rates were based solely on the identification of iridescence. Other studies have shown that massive virus proliferation is not necessary for IVs to cause insect death (Sieburth and Carner, 1987), which may also help to explain the lack of iridescence in many infected larvae. These studies, together with the advances that are being made in the molecular characterization of the IV genome, which may permit strain selection or genetic manipulation, suggest that the IVs may in future be considered as serious candidates for the biological control of pest species.

5 RNA viruses of insects 5.1

INTRODUCTION

There have recently been a number of excellent reviews on the three families of small insect viruses that have an ssRNA genome: Nodaviridue (Garzon and Charpentier, 1991; Hendry, 1991), Tetraviridae (Moore, 1991a; Reinganum, 1991), and Picornaviridae (Moore, 1991b; Moore and Eley, 1991). This section will, therefore, include the general classification and structure of these viruses (see Table 4), but will be primarily confined to a discussion of the more recent advances in our understanding of the replication of these groups of viruses. In addition, there is one group of segmented dsRNA viruses that infect insects; these are the cytoplasmic polyhedrosis viruses. 5.2

NODAVlRIDAE

The Nodaviridue are non-enveloped, icosahedral insect pathogenic viruses, which contain a bipartite RNA genome encapsidated within a single virion (Matthews, 1982). The single-stranded RNAs are messenger-sense and both

44

L. A. KING e t a / .

are required for infectivity (Newman and Brown, 1973; Gallagher et al., 1983). The virus particles have a diameter of 3 0 n m and the sizes of the major coat protein and the two genomic RNA species, RNAl and RNA2, are 40 kDa, 1.1 x 10' Da and 0.47 x 10' Da, respectively (Garzon and Charpentier, 1991; Hendry, 1991). As a family they have been more thoroughly studied than any of the other small insect riboviruses, and their initial popularity probably arose because their type member, Nodamura virus, was found to infect mammals as well as insects (Scherer and Hurlbut, 1967). Interest eventually spread to other nodaviruses when it was discovered they were easier to amplify in tissue culture, and were as a consequence more amenable to molecular analyses. A number of significant advances in nodavirus research have previously been reviewed (Garzon and Charpentier, 1991; Hendry, 1991) including the respective roles of RNAs 1 and 2 together with their complete nucleotide sequences (Friesen and Rueckert, 1981, 1982, 1984; Gallagher et al., 1983; Dasgupta et al., 1984); the first production of an infectious mRNA transcript from a cloned cDNA (of insect or animal RNA virus) (Dasmahapatra et al., 1986); and the elucidation of the atomic structure of the viral capsid at 3.0 A resolution. 5.2.1 Isolation and host range Nodamura virus was first isolated from mosquitoes and was initially classified as an arbovirus. Later, on the basis of physicochemical properties, this was changed to membership of the picornaviruses (Murphy et al., 1970). However, when it was demonstrated that the virus had a bipartite genome, the virus was classified in a new family, the Nodaviruses (Newman and Brown, 1973, 1977). Six viruses have been classified as members of the Nodaviridae (Table 4); all have been isolated from insects (Lepidoptera, Coleoptera and Diptera) but none appear to be host specific. Nodamura virus is the only virus known to infect other animals (e.g. mice) as well as insects. Little is known about the spread of nodavirus infections in natural insect populations, as each virus was isolated from individual insects in a single location (Hendry, 1991). For example, Nodamura virus was isolated from female Culex tritaeniorhynchus mosquitoes trapped at Nodamura, near Tokyo, Japan in 1956; black beetle virus was isolated from Heteronychus arator collected near Helensville in New Zealand in 1974 (Longworth and Archibald, 1975); and Flock House virus was isolated from the New Zealand grass grub, Costelytra zealandica, collected at Flock House in New Zealand in 1980 (Dearing et al., 1980). 5.2.2 Structure and classification The six viruses which have been classified by the International Committee on Taxonomy for Viruses (ICTV) as belonging to the family Nodaviridae

ADVANCES IN INSECT VIROLOGY

45

TABLE 4 R N A viruses of insects" Family (Genus)

Species

Nodaviridae

Nodarnuara virus (NMV) Black beetle virus (BBV) Flock House virus (FHV) Boolarra virus (BoV) Gypsy moth virus (GMV) Manawata virus (MwV)

Tetraviridae (Nudaurelia p virus)

Nudaurelia p virus (NdpV) Antheraea /3 virus (AnPV) Darna p virus (DrpV) Thosea G , virus (ThPV) Philosamia /3 virus (PhPV) Dasychira p virus (DspV) Trichoplusia /3 virus (TrPV)

Picornaviridae

Cricket paralysis virus (CrPV) Drosophila C virus (DCV) Gonometu virus (GV)

Reoviridae (Cytoplasmic polyhedrosis virus, CPV)

Bombyx mori CPV (type species; type 1) lnachis io CPV (type 2 ) S. exempta CPV (type 3 ) Actias selene CPV (type 4) T. ni CPV (type 5 ) Biston betularia CPV (type 6) Triphena pronuba NPV (type 7) Abraxas grossulariata CPV (type 8) Agrotis segetum CPV (type 9) Aporophytla iutulenta CPV (type 10) S. exigua CPV (type 11) S. exempta CPV (type 12)

"Francki et ul 1991: Garzon and Charpentier. 1991; Hendry, 1991; Moore. 1991a,b; Moore and Eley, 1991; Reinganum. 1991.

L. A. KING eta/.

46

are Nodamura virus (NV), black beetle virus (BBV) (Longworth and Carey, 1976), Flock House virus (FHV) (Dearing et al., 1980); Boolarra virus (BoV) (Reinganum et a f . , 1985), gypsy moth virus (GMV) (Reavy et al., 1982), and Manawata virus (MwV) (see Table 4). Serology has been the most convenient method for identifying nodaviruses, although none of the six identified in Table 4 cross-react with all the others. In addition, nodaviruses can be diagnosed by their particle diameter (about 30 nm), size of the major coat protein (about 40 kDa) and the presence of two RNA species (about 0.5 and 1.0 X lo6 Da). On the basis of such comparisons of morphological and biochemical properties, Mori et al. (1992) have proposed that a new virus, designated Striped Jack Nervous Necrosis Virus (SJNNV), be included in the Nodaviridae. SJNNV is non-enveloped with a bipartite, positive-sense genome and has a virion diameter of approximately 25 nm. The genomic RNAs have molecular masses of 1.01 X lo6 Da (RNA1) and 0.49 x lo6 Da (RNA2), and neither have a poly(A) tail at the 3' terminus. There are two viral structural proteins with molecular masses of 42 and 40 kDa, and RNAs 1 and 2 direct the synthesis of proteins of 100 kDa and 42 kDa, respectively, in reticulocyte Iysates. The physicochemical properties of all members of the nodavirus family have been summarized by Hendry (1991). Briefly, virions are 29-31 nm in diameter based on electron microscopy measurements. BBV has also been measured by small-angle X-ray scattering (Hosur et al., 1984) and this has given an accurate sizin of 31.2 nm. The structure of the BBV capsid has been determined at 3 resolution using X-ray diffraction of virus crystals (Hosur et al., 1984). The data obtained from this study have been reviewed (Hendry, 1991) and indicated that the BBV capsid is composed of 180 identical protomers arranged in a T = 3 icosahedral symmetry (Hosur et al., 1984). Each protomer was shown to consist of 407 amino acids; either as the alpha precursor polypeptide or as the two cleavage products, beta and gamma (see below). Using the information obtained from sequencing of the genome and predicted protein sizes, the virion particle weight for BBV is 9.4 X lo6 Da. The same data indicate that, for BBV, the RNA content comprises 16% of the virion mass. Virions are resistant to both ether and chloroform, suggesting that they do not contain any lipid. Neither FHV, BBV nor NV are affected by acidic pH treatments but heating to 5040°C appears to result in loss of infectivity of NV and FHV (Murphy et al., 1970; Scotti et al., 1983). Using BBV as the best studied example, the nodavirus capsid consists of a major (beta, 39 kDa) and two minor protein species (gamma, 4.5 kDa; and a precursor, alpha, 44kDa) (Friesen and Rueckert, 1981). The alpha precursor is cleaved between an asparagine residue at position 363 and an alanine residue at 364, to generate beta and gamma (Dasgupta et al., 1984). There is considerable variation in the reported number and sizes of the minor proteins of the other nodaviruses. It is only in the case of BBV, which

1

ADVANCES IN INSECT VIROLOGY

47

has been sequenced, that the relationship between alpha, beta and gamma has been firmly established (Dasgupta et al., 1984).

5.2.3 The bipartite RNA genome Both nodaviral RNA species are messenger-sense, single-stranded and both are necessary for infectivity (Newman and Brown, 1973; Gallagher et al., 1983). The BBV RNAs have been shown to have typical methylated guanine cap structures at t h e 5' terminus, but the 3' terminus is thought to be modified so as to prevent polyadenylation. The larger of the two RNAs (RNA1) has a molecular mass of about 1.1 X 10' (3106 bases (b) for BBV; Kaesberg, 1987) and encodes two viral proteins, one of which is known to be a viral replicase. The smaller RNA2 molecule has a molecular mass of about 0.47 X lo6 (1399 b for BBV; Kaesberg, 1987) and encodes the viral coat protein precursor (alpha). Computer analysis using the nucleotide sequence data of BBV predicts a stable and distinctive secondary structure at both the 3' and 5' ends of RNA2 (Dasgupta et a!., 1984). The mechanism by which each of the genomic RNAs is selected and packaged into the same virion is unknown. Recently, however, a report by Zhong et a f . (1992) indicated that a 32-base region of the RNA2 (bases 186-217) of FHV appears to be important in this process. Analysis predicts that the RNA is folded in this region into a stem-loop structure with a 5-base loop and a 13 base-pair bulged stem.

5.2.4 Molecular studies The demonstration that RNA transcripts of BBV cDNA are infectious in cultured Drosophilu cells (Dasmahapatra et al., 1986) allows the possibility of modifying the nodaviral genome for analysis by recombinant DNA techniques. Dasmahapatra er al. (1987) devised a cell-free expression system by constructing a plasmid containing a translation initiation signal from the 5' non-coding region of BBV. They inserted the 3C protease coding region of Coxsackievirus B3 (CVB3) next to the ribosome binding sequences and initiator A U G site of the nodavirus. Transcripts of this plasmid directed the efficient synthesis of an active protease. More recently, Dasmahapatra et al. (1991) demonstrated that using this system, a biologically active protease is synthesized which possesses both cis and trans processing capabilities. This in v i m synthesized protease is analogous to the native 3C produced by CVB3-infected HeLa cells, and antibody prepared against the native protease cross reacts with the in vitro protease. Using the translational initiation signal from BBV RNA1, the authors have also expressed the CVB3 capsid precursor and part of the P2 region in vitro. Additionally, they report that the capsid precursor is cleaved, between IC (VP3) and 1D (VPI), by the proteolytic activity of in vitro synthesized 3C in trans.

,

L. A. KING e t a / .

48

The amplification of RNA that is a result of RNA replication has been found to occur naturally only in RNA viruses. Nodavirus cDNA sequences have recently been used in an attempt to harness this power for the amplification of heterologous mRNAs (Ball, 1992). The authors have expressed both an RNA replicase and its corresponding RNA templates in functional form, using a vaccinia virus-bacteriophage T7 RNA polymerase vector. Plasmids were constructed which contained in 5' to 3' order: a T7 promoter; a full length cDNA encoding either the RNA replicase (RNAl) or the coat protein (RNA2) of FHV; a cDNA encoding the self-cleaving ribozyme of satellite tobacco ringspot virus; and a T7 transcriptional terminator. RNAs were produced, both in vivo and in vitro, by T7 RNA polymerase with sizes that closely resembled those of the two authentic genomic RNAs (RNA1 and 2). In baby hamster kidney (BHK) cells that expressed authentic FHV RNA replicase, the RNA2 transcripts were accurately replicated. More importantly, the RNAl transcripts directed the synthesis of an enzyme that could replicate not only authentic virion-derived FHV RNA, but also the plasmid-derived transcripts themselves. Under the latter conditions, replicative amplification of the RNA transcripts ensued and resulted in a high rate of synthesis of the encoded products. This successful expression from a DNA vector of the complex biological process of RNA replication will greatly facilitate studies of its mechanism and is a major step towards the goal of achieving RNA replication for mRNA amplification.

5.3

P

T E T R A V I R I D A E ( N C J D A C J R E L I A VIRUS GROUP)

The Tetraviridae, also referred to as the Nudaurefia P virus family after the type member of the group, are relatively important small insect riboviruses that naturally regulate a number of pest species of Lepidoptera. There is, however, no tissue culture system available to support their replication, and so they are not widely studied. Past work on these viruses has been restricted to physicochemical, in vivo pathogenicity and replication studies using in vitro translation systems (Hendry, 1991; Moore, 1991a).

5.3.1 Classijication, isolation and host range This family of insect viruses comprises seven members (see Table 4) that have all been isolated from pest species of Lepidoptera. All members replicate exclusively in insect hosts. The type member is Nudaurefia capensis p virus (NPV) which was originally identified from the pine emperor moth, N . cytharea capensis (Struthers and Hendry, 1974; du Plessis et a f . , 1991). Using antisera raised against purified virions, five other members of this family have been identified, as shown in Table 4. The final member,

ADVANCES IN INSECT VIROLOGY

49

Nudaurelia capensis virus (NV), was also identified from the pine emperor moth, but showed no serological relationship to NPV.

5.3.2 Virion structure The virions of the NPV are non-enveloped, icosahedrons with a diameter of 3540 nm and contain a single-stranded RNA genome with a molecular mass of about 1.8 x lo6 Da. The virion capsid contains a single protein, which has a molecular mass of 60-68 kDa. The biophysical properties of the individual group members have recently been reviewed by Moore (1991a) and Olson et nl. (1990). Olson et al. (1990) have recently reconstructed the three-dimensional structure of Nudaurelia P virus (NPV), to a resolution of 3.2nm, using images of frozen-hydrated virions. The model of a distinctly icosahedral capsid (with 240 copies of a single 61 kDa subunit) and T = 4 symmetry compares well with what was previously observed with negative staining using electron microscopy. The authors state that analyses of the density maps, volume estimates and model building experiments, indicate that each subunit consists of two domains. Each large cylindrical domain (40 kDa; 4 X 4 nm) associates with two other large domains in neighbouring subunits to form a Y-shaped trimeric aggregate in the outer capsid surface. Four trimers come together to make each of the 20 planar faces of the icosahedron. The small domains (21 kDa; 13-16.5 nm) are presumed to be associated with the inside of the virion to make a contiguous, non-spherical shell. The small ssRNA genome is loosely packed inside the capsid with a low average density. Similar physical characteristics were reported for NV, following virus crystallization and X-ray diffraction studies at 2.8 8, resolution (Cavarelli et al., 1991).

5.3.3 Replication and molecular studies Very little is known about the replication cycle or genetic organization of this group of viruses, mainly due to the lack of suitable cell culture systems capable of supporting virus replication. Electron microscopy and ELISA studies have indicated that replication in vivo occurs, at least initially, in the cytoplasm of fore- and midgut cells (Reavy and Moore, 1982). In vitro replication studies have been confined to translation of the viral RNA species in rabbit reticulocyte lysates. A number of putative polypeptides have been identified in this way, including a potential capsid precursor and polymerase, although it was not possible in these experiments to draw definite conclusions (Reavy and Moore, 1984; King et al., 1984). In a recent study, Agrawal and Johnson (1992) report the sequencing and analysis of the second RNA species of NV. It was found to consist of 2448 b and contained one long ORF encoding a 644 amino acid capsid protein

L. A. KING eta/.

50

precursor (70kDa). The 5' and 3' ends contained 366 and 150 b of non-coding region, respectively. The capsid protein precursor was shown to start at the second AUG initiation codon. The authors identified a putative asparagine/phenylalanine cleavage site in the precursor protein, which yields the previously identified capsid protein of 62 kDa, and a smaller polypeptide of 8 kDa that was also discovered by the authors in mature virus particles (Agrawal and Johnson, 1992). du Plessis et a!. (1991) reported that the replicative form of NPV is a dsRNA molecule. They found that larvae of the pine emperor moth consistently contained a ds species of RNA with the expected size for a ds replicative form. In Northern blots, it hybridized with a '*P-labelled vRNA probe, whereas other smaller dsRNAs did not, and cell extracts from non-infected larvae contain no dsRNA molecules. 5.3.4 Biological control All members of this family of viruses have been isolated from pest species of Lepidoptera and, to date, analyses indicate that replication is confined to insect hosts, making these viruses potential candidates for biological control programmes (Moore, 1991a). Darna trima is an important pest of several crops, including coconut and oil palms in Southeast Asia, and Thoseu asigna is a pest of oil palms in Malaysia. The viruses isolated from N . cytherea capensis and D. trima have been shown to cause large reductions in insect populations, affecting up to 90% of the larvae, and the latter virus has been used successfully in biological control programmes (Moore, 1991a). 5.4

PICORNAVIRIDA E

There are more than 30 insect viruses which have been proposed as members of the Picornuviridae, but only three have been studied in enough detail to be classified as such by the International Committee on Taxonomy for Viruses. The three that have been assigned to this group, although not to specific genera, are cricket paralysis virus (CrPV), Drosophila C virus (DCV) and Gonometa virus (GV). Of these, replicating isolates exist only for CrPV and DCV, and so most of the existing information about insect picornaviruses comes from work performed with these two viruses (Moore, 1991b). 5.4.1 Isolation, classification and host range CrPV was first identified during a mass breeding programme of the Australian field cricket Teleogryllus oceanicus. Early instar nymphs became paralysed and eventually died (Reinganum et al., 1970; Reinganum, 1973). When virus particles were injected into crickets, death occurred within 3

ADVANCES IN INSECT VIROLOGY

51

days. DCV was first identified in a laboratory colony of the fruit fly Drosophila melanogaster (reviewed in Moore, 1991b) and was shown to be pathogenic for the flies. GV was isolated by Harrap et al. (1966) from the insect Gonomefa podocarpi, which at the time was a serious pest of exotic pines in Uganda. GV proved to be an extremely effective biocontrol agent for this insect species (Longworth ef a f . , 1973). One problem with GV is that no laboratory stocks of the virus exist, so it is difficult to include this virus in any classification system (Moore, 1991b). A number of other picorna-like viruses have more recently been isolated from a variety of different insects (Thomas-Orillard, 1988; see Table 4). Kawino virus was isolated from Mansonia uniformis mosquitoes in Kenya (Pudney ef al., 1978). The virus was also shown to replicate in three mosquito cell lines at 28°C and had most of the physicochemical properties associated with picornaviruses, with the exception that the 3' end of the RNA genome appeared not be be polyadenylated. The host ranges of CrPV and DCV have been evaluated by the experimental infection of purified virions into a range of hosts and by the serological identification of related viruses in natural insect populations. The results of these tests have recently been reviewed by Moore (1991b). However, as both procedures are open to criticism, the natural host range of these viruses remains to be determined. Fediere et al. (1990) have described a new picorna-like virus from the oil palm pest Lafoia viridissima (Parasa viridissima]. Virions of this new virus have a diameter of 30 nm which is composed of four proteins, two major (30 and 31 kDa) and two minor. The genome was shown to comprise a single molecule of ssRNA with a molecular mass of 2.9 x lo6 Da. In a comparative in vitro study of picornaviruses in Drosophila cells, Plus (1989) has suggested that three major conclusions can be reached about DCV isolated from Drosophila and CrPV isolated from different Grylid and Lepidopteran populations: they are both true picornaviruses but are unique in their properties; they are serologically related but are appearing increasingly different as research continues; and that DCV isolated from insect populations and screened on Drusophita cell lines form a homogeneous group, whereas CrPV isolates from five different insect species do not. Despite the fact that the latter were screened on Drosophila cell lines, they were found to be divided into three distinct host range/ serological groups. As previously mentioned, there are many picorna-like viruses that are not well enough characterized to be classified as members of the Picornaviridae. Two of these viruses, aphid lethal paralysis virus (ALPV) and Rhopalosiphum padi virus (RhPY), have recently been the focus of a number of studies and may now be closer to being assigned family status. Williamson et al. (1989) published an account of comparisons made between a South African strain of RhPV and an isolate from Illinois, USA. Both viruses were

52

L. A. KING e t a /

serologically related but there were differences in their capsid protein molecular weights and virion buoyant densities. The South African isolate was found to have two more capsid proteins than was previously thought, and the size of the RNA, which contains a poly(A) tract, was found to be 10 kb. The virions were stable at pH 5.0-8.4, degraded quickly below pH 3.0 and the virions lost genomic RNA when heated to 56°C. From this evidence, together with the results of radio-iodination and Western blotting of capsid proteins, the authors proposed that RhPV be classified as a picornavirus. Support for the inclusion of both RhPV and ALPV into this family is given in the report by Williamson and Rybicki (1989), which details a comparative study on the cell-free translation of the genomic RNAs of both viruses. Studies have also been performed, using nucleic acid in situ hybridization, to identify and localize actively replicating ALPV in whole body sections of aphids (Hatfill et al., 1990). Virus was detected in the stomach and intestinal epithelia of infected Rh. padi aphids and, in the advanced stages of infection, virus spread throughout the protocerebrum. In another report, the pathogenicity, host specificity and tissue specificity of RhPV were studied by inoculation of virus-free clones of five aphid species with gradient-purified virus (Gildow and D’Arcy, 1990). Only two out of the five test species of insects became infected by RhPV, and upon ultrastructural examination, virus was only visualized in the posterior region of the midgut and the hindgut. The authors suggested that viral entry was via endocytosis and that infection resulted in progressive loss of cytoplasmic organelles and the development of membrane vesicles. Virions accumulated only in the cytoplasm and were subsequently released into the gut lumen and haemolymph. The results confirm the pathogenicity of RhPV and suggest a high degree of host and tissue specificity for the virus. 5.4.2 Virion structure In common with their mammalian counterparts, insect picornaviruses have non-enveloped, icosahedral virions which contain a single-stranded, messenger-sense RNA genome (2.5-3.0 x lo6 Da). Virus particles are about 27nm in diameter and the vRNA has been shown to possess a genomelinked protein (VpG; approximately 3.9 x lo3 Da) at the 5’ end and a poly(A) tract at the 3‘ end (King et al., 1987; Moore et al., 1987; King and Moore, 1988). The capsid is composed of 60 copies each of four viral proteins (VP1-4, although CrPV appears to have only VP1-3), and these have molecular masses of 31-35, 30-34, 28-30 and 5.5-13.5 kDa, respectively (Moore, 1991b). As with the Enterovirus genus of the Picornaviridue, insect picornavirus particles are acid stable; with the vertebrate viruses this protects the virus in the gut. The significance of this finding for the insect picornaviruses is uncertain as the insect gut is alkaline.

ADVANCES IN INSECT VIROLOGY

53

5.4.3 Virus replication and molecular studies Most of the information on the replication and molecular biology of insect picornaviruses has come from in vitro studies of CrPV infection in insect cell lines. CrPV will replicate in a number of cell lines including D. mefanogasfer, T. ni, L. dispar and A. aegyptii (Reinganum, 1973; Scotti, 1976). Most of the replication studies have utilized Scheider's Drosophifa DL1 cells, as these produce a reliable cytopathic effect which can be used to titrate the virus by plaque assay (Scotti et af., 1981). Both CrPV and DCV replicate to high titres in the DL1 cells (Moore et af., 1981a; Moore and Pullin, 1982), with the CrPV infection being more lytic and DCV more cell associated. About 20 virus-specified polypeptides have been identified in CrPVinfected DLl cells using SDS-PAGE (Moore et a f . , 1982). Moore and co-workers have shown that virion structural protein synthesis, from high molecular weight precursors, is similar to that observed for mammalian picornaviruses (Moore et al., 1980, 1981a). A putative VPO, the precursor of VP2 and VP4 in mammalian picornaviruses, was identified in both CrPVand DCV-infected cells. With DCV, VPO was shown to be the precursor of VP2 and VP4 by pulse-chase experiments, but with CrPV, there was no evidence for VP4 (Moore et al., 1980, 1981b). Pactinomycin mapping studies have demonstrated that the order of capsid proteins in the CrPV polyprotein is VPO, VP3, VPl and that these are encoded at the 5' terminus of the genome (Reavy and Moore, 1982). The CrPV genome has a poly(A) tail and a 5' linked VPg (King and Moore, 1988). Sequence data generated from the 3' terminal 1600 bases of the CrPV genome confirmed the presence of the poly(A) tail, a 3' non-coding region and an ORF that encodes a potential RNA polymerase (King et al., 1987). 5.4.4 Biological control There are relatively few reports of the use of small RNA viruses of insects as biological control agents. As mentioned above, GV has been used successfully in Uganda to control G. podocarpi. There have been two recent reports describing the potential use of the insect picornviruses as control agents. Manousis and Moore (1987) demonstrated that CrPV caused a high mortality when tested against Dacus ofeae, a serious pest of olive plantations, and Fediere et al. (1990) reported that different doses of a picorna-like virus sprayed onto an industrial oil palm plantation infested by P. viridissima gave good control of the pest. After a period of 1 week, a dose-related mortality gradient ranging from 11% to 61% was obtained. Two weeks after spraying, the mortality of larvae in treated plots reached 92% and during the next generation the numbers of larvae were very low.

54

L. A. KING eta/.

References Agrawal, D. K. and Johnson, J . E. (1992). Sequence analysis of the capsid protein of Nudaurelia capensis omega virus, an insect virus with T equals 4 icosahedral symmetry. Virology 190, 806-814. Akiyoshi, D.. Chakerian, R., Rohrmann, G . F., Nesson, M. H. and Beaudrea, G. S. (1985). Cloning and sequencing of the granulin gene from the Trichoplusia ni granulosis virus. Virology 141, 328-332. Amargier, A , , Vago, C. and Meynadier, G. (1964). Etude histopathologique de I’evolution de la virose a fuseaux chez le Coleoptere Melolontha rnelolontha. Mikroskopie 19, 309-315. Anderson, I. G. and Prior, H. C. (1992). Baculovirus infections in the mud crab Scylla serrata, and a freshwater crayfish, Cherax quadricarinolus, from Australia. J . Invert. Pathol. 60, 265-273. Arif, B. M. (1976). Isolation of an entomopoxvirus and characterization of its DNA. Virology 69, 626-634. Arif, B. M. (1984). The entomopoxviruses. Adv. Virus Res. 29, 195-213. Arif, B. M. (1986). The structure of the viral genome. I n “The Molecular Biology of Baculovirmes’‘ (Eds W Doerfler and i? Bohrn), pp. 21-29, Sprri-lger-Verlag, Berlin. Arif, B. M. and Kurstak, E. (1991). The entomopoxviruses. I n “Viruses of Invertebrates” (Ed. E. Kurstak), pp. 179-195, Dekker, New York. Atkinson, A. E., Weitzman, M. D., Obosi, L., Beadle, D. J. and King, L. A. (1990). Baculoviruses as vectors for foreign gene expression in insect cells. Pest. Sci. 28, 215-224. Atkinson, A. E . , Bermudez, I . , Darlison, M. G., Barnard, A. E., Earley, F. P. G., Possee, R. D., Beadle, D. .I.and King, L. A. (1992). Assembly of functional GABAA receptors in insect cells using baculovirus expression vectors. NeuroReport 3, 597400. Ball, L. A. (1992). Cellular expression of a functional nodavirus RNA replicon from vaccinia virus vectors. J. Virol. 66, 2335-2345. Batson, B. S., Johnson, M. R. L . , Arnold, M. K. and Kelly, D. C. (1976). An iridescent virus from Sirnuliurn sp. (Diptera Simuliidae) in Wales. J. Invert. Pathol. 27, 133-135. Beames, B. and Summers, M. D. (1988). Comparisons of host cell DNA insertions and altered transcription at the site of insertions in few polyhedra baculovirus mutants. Virology 162, 206-220. Beames, B. and Summers, M. D. (1989). Location and nucleotide sequence of the 25K protein missing from baculovirus few polyhedra (FP) mutants. Virology 168, 344-353. Beames, B. and Summers, M. D. (1990). Sequence comparison of cellular and viral copies of host cell DNA insertions found in Autographa californica nuclear polyhedrosis virus. Virology 174, 354-363. Bellet, A. J. D. and Inman, R. B. (1967). Some properties of deoxyribonucleic acid preparations from Chifo, Sericesthis and Tipula iridescent viruses. 1. Mol. Biol. 25, 425-532. Bergoin, M. and Dales, S. (1971). Comparative observations on poxviruses of invertebrates and vertebrates. In “Comparative Virology” (Eds K. Maramorosch, and E. Kurstak), pp. 169-205, Academic Press, New York. Bergoin, M. and Roberts, D. W. (1971). Enzymes associated with an insect poxvirus. Virology 43, 306-309. Bergoin, M., Devauchelle, G . , Duthoit, J. L. and Vago, C. (1968a). Etude au

ADVANCES IN INSECT VIROLOGY

55

microscope electronique des inclusions de la virose a fuseaux des coleopteres. C. R. Acad. Sci. 266D, 2126-2128. Bergoin, M., Devauchelle, G . and Vago, C. (1968b). Observations au microscope electronique sur le developpement du virus de la maladie a fuseaux du coleoptere Melolontha melolontha, L. C. R. Acad. Sci. Ser. D267, 382-385. Bergoin, M . , Devauchelle, G. and Vago, C. (1969). Electron microscopy study of the pox-like virus of Melolontha melolontha L.: Virus morphogenesis. Arch. Ges. Virusforsch 28, 285-302. Bergoin, M., Veyrunes, J . C. and Scalia, R. (1970). Isolation and amino acid composition of the inclusions of Melolontha melolontha poxvirus. Virology 40, 760-763. Bergoin, M., Devauchelle, G. and Vago, C. (1971). Electron microscopy study of Melontha poxvirus: The fine structure of occluded virions. Virology 43, 453467. Biever, K . D. and Wilkinson, J . D. (1978). A stress induced granulosis virus of Pieris brassicae. Environ. Entomol. 7, 572-573. Bilimoria, S. L. (1986). Taxonomy and identification of baculoviruses. In “The Biology of Baculoviruses, Vol I , Biological Properties and Molecular Biology” (Eds R. R. Granados and B. A. Federici), pp. 37-59. CRC Press, Boca Raton, Florida. Bilimoria, S. L. and Arif, B. M. (1979). Subunit protein and alkaline protease of entomopoxvirus spheroids. Virology 96, 596-603. Bilimoria, S. L. and Arif, B. M. (1980). Structural polypeptides of Choristoneura biennis entomopoxvirus. Virology 104, 253-257. Bird, E. T. (1974). The development of spindle inclusions of Choristoneura fumiferana (Lepidoptera: Torticidae) infected with entomopox virus. J. Invert. Pathol. 23, 325-332. Bishop, D. H. L. (1986). UK release of a genetically marked virus. Nature, London 323, 496. Bishop, D. H. L. (1989). Genetically engineered viral insecticides: a progress report 1986-1989. Pest. Sci. 27, 173-189. Bishop, D. H. L. and Possee, R. D. (1990). Baculovirus expression vectors. Adv. Gene Technol. 1, 55-72. Bishop, D. H. L., Entwistle, P. F., Cameron, I. R., Allen, C. J. and Possee, R. D. (1988). Field trials with genetically engineered baculovirus insecticides. In “The Release of Genetically Engineered Micro-organisms” (Eds M. Sussman, C. H. Collins, F. A. Skinner and D. E. Stewart-Tull), pp. 143-179, Academic Press, New York and London. Blinov, V. M. (1984). Nucleotide sequence of the Galleria mellonella nuclear polyhedrosis virus origin of DNA replication. FEBS Letters 161, 254-258. Blissard, G. W. and Rohrmann, G. F. (1989). Location, sequence, transcriptional mapping and temporal expression of the gp64 envelope glycoprotein gene of the Orgyia pseudotsugata multicapsid nuclear polyhedrosis virus. Virology 170, 537-555. Blissard, G. W. and Rohrmann, G. F. (1990). Baculovirus diversity and molecular biology. Annu. Rev. Entomol. 35, 127-155. Blissard, G. W. and Rohrmann, G. F. (1991). Baculovirus gp64 gene expression: analysis of sequences modulating early transcription and transactivation by IE-1. J . Virol. 65, 5820-5827. Blissard, G. W., Quant-Russell, R. L., Rohrmann, G. F. and Beaudreau, G. S. (1989). Nucleotide sequence, transcriptional mapping and temporal expression of the gene encoding p39, a major structural protein of the multicapsid nuclear polyhedrosis virus of Orgyia pseudotsugata. Virology 168, 354-362. Brassel, J. and Benz, G. (1979). Selection of a strain of the granulosis virus of the

56

L. A. KING eta/.

codling moth with improved resistance against artificial ultraviolet radiation and sunlight. J. Invert. Pathol. 33, 358-363. Burand, J . P., Horton, H. M., Retnasami, S. and Elkington, J . S. (1992). The use of polymerase chain reaction and shortwave UV irradiation to detect baculovirus DNA on the surface of gypsy moth eggs. J. Virol. Methods 36, 141-150. Cameron, I. R. (1990). Identification and characterisation of the gene encoding the major structural protein of insect iridescent virus type 22. Virology 178, 3542. Cameron, I. R., Possee, R. D. and Bishop, D. H. L. (1989). Insect cell culture technology in baculovirus expression systems. Trends Biotechnol. 7, 66-70. Carruthers, W. R., Cory, J. S. and Entwistle, P. F. (1988). Recovery of pine beauty moth (Panolis flamrnea) nuclear polyhedrosis virus from pine foliage. J. Invert. Pathol. 52, 21-32. Carson, D. D., Guarino, L. A. and Summers, M. D. (1988). Functional mapping of an AcNPV immediate early gene which augments expression of the IE-1 trans-activated 39K gene. Virology 162, 444-451. Carson, D. D., Summers, M. D . and Guarino, L. A. (1991). Transient expression of the Autographa californica nuclear polyhedrosis virus immediate-early gene, IE-N, is regulated by three viral elements. J. Virol. 65, 945-951. Cavarelli, J . , Bomu, W., Liljas, L., Kim, S., Minor, W., Munshi. S., Muchmore, S., Schmidt, J.. Johnson, J. and Hendry, D. A . (1991). Crystallisation and preliminary structure analysis of an insect virus with T equal to 4 quasi-symmetry: Nudaurelia capensis o virus. Acta Crystallography 47( l ) , 23-29. Cerutti, M. and Devauchelle, G. (1980). Inhibition of macromolecular synthesis in cells infected with an invertebrate virus (iridovirus type 6 or CIV). Arch. Virol. 63, 297-303. Cerutti, M. and Devauchelle, G . (1982). Isolation and reconstitution of Chilo iridescent virus membrane. Arch. Virol. 74, 145-155. Cerutti, M. and Devauchelle, G . (1985). Characterisation and localisation of CIV polypeptides. Virology 145, 123-131. Cerutti, M., Guerillon, J., Arella, M. and Devauchelle, G. (1981). La replication de I’iridovirus de type 6 (CIV) dans differentes lignees cellulaires. CR Seances Academy of Sciences [III] 292, 791-802. Chakerian, R.. Rohrmann, G . F. and Beaudreau, G. S. (1985). The nucleotide sequence of the Pieris brassicae granulosis virus granulin gene. J. Gen. Virol. 66, 1263-1269. Chapman, H. C., Clark, T. B., Anthony, D. W. and Glenn, F. E . (1971). An iridescent virus from the larvae of Corethralla brakelyi (Diptera: Chaoboridae) in Louisiana. J . Invert. Pathol. 18, 284-287. Chisholm, G. E. and Henner, D. J . (1988). Multiple early transcripts and splicing of the Aurographa cali,fornica nuclear polyhedrosis virus IE-1 gene. J. Virol. 62, 3 193-3200. Clark, T. B. (1982). Entomopox-like particles in three species of bumble bees. J. Invert. Pathol. 39, 119-122. Clark, T. B., Kellen, W. R. and Lum, P. T. M. (1965). A mosquito iridescent virus (MIV) from Aedes taeniorhynchus (Weidermann). J. Invert. Pathol. 7, 519-524. Clem, R. J. and Miller, L. K. (1993). Apoptosis reduces both the in vitro and the in vivo infectivity of baculoviruses. J. Virol. 67, 313g3738. Clem, R. J . , Fechheimer, M. and Miller, L. K. (1991). Prevention of apoptosis by a baculovirus gene during infection of insect cells. Science 254, 1388-1390. Cochran, M. A. and Faulkner, P. (1983). Location of homologous DNA sequences interspersed at five regions in the baculovirus AcMNPV genome. J. Virol. 45, 961-970. Cory, J. S. (1993). Biology and ecology of baculoviruses. In “Opportunities for

ADVANCES IN INSECT VIROLOGY

57

Molecular Biology in Crop Production” (Eds D. J . Beadle, D. H. L. Bishop, L. G. Copping and D. W. Hollomon), pp. 3-10. BCPC Monograph 55. Cory, J. S. and Entwistle, P. F. (1990). The effect of time of spray application on infection of the pine beauty moth, Panolis frammea (Den. and Schiff.) (Lep., Noctuidae), with nuclear polyhedrosis virus. J . Appl. Entomol. 110, 235-241. Crawford, A. M. and Kalmakoff, J . (197-5). A host-virus interaction in a pasture habitat. J. Invert. Pathol. 29, 81-87. Crook, N. E., Clem, R. J. and Miller, L. K. (1993). An apoptosis-inhibiting baculovirus gene with a zinc finger-like motif. J . Virol. 67, 2168-2174. Cunningham, J. C. and McPhee, J . R. (1973). Aerial application of entomopoxvirus and nuclear polyhedrosis virus against the spruce budworm at Chapleau Ontario. Report of the Insect Pathology Research Institute, Canada Forest Service, Sault Ste. Marie, Ontario. Dasgupta, R., Ghosh, A., Dasmahapatra, B., Guarino, L. A. and Kaesberg, P. (1984). Primary and secondary structure of Black Beetle virus RNA2 the genomic messenger for BBV coat protein precursor. Nucl. Acids Res. 12, 7215-7223. Dasmahapatra, B., Dasgupta, R., Saunders, K . , Selling, B., Gallagher, T. and Kaesburg, P . (1986). Proc. Nad. Acad. Scc. 83, 63-66. Dasmahapatra, B., Rozhon, E. and Schwartz, J. (1987). Cell free expression of Coxsackievirus 3C protease from a cDNA clone which contains the translational initiation signal of an insect virus. In “Proceedings of the VIIth International Congress of Virology”, Edmonton, Alberta, p. 135. Dasmahapatra, B., Rozhon, E., Hart, A , , Cox S., Tracy, S. and Schwartz, J. (1991). Cell-free expression of the coxsackie 3C protease using the translational initiation signal of an insect virus RNA and its characterisation. Virus Res. 20, 237-249. David, W. A. L. and Gardiner, B. 0. C. (1966). Persistence of a granulosis virus of Pierk brassicae on cabbage leaves. J . Invert. Pathol. 8, 180-183. Dearing, S. C., Scotti, P. D., Wigley, P. J . and Dhana, S. D. (1980). A small RNA virus isolated from the grass grub. Costelyfra zealandica (Coleoptera: Scarabaeidae). N. Z.1.Zool. 7, 267-269. de Kort, C.A.D. and Granger, N. A. (1981). Regulation of the juvenile hormone titre. Annu. Rev. Entomol. 26, 1-28. Delius, H . , Darai, G. and Flugel, R. M . (1984). DNA analysis of the insect iridescent virus type 6: evidence for circular permutation and terminal redundancy. J . Virol. 49, 609-614. Devauchelle, G., Bergoin, M. and Vago, C. (1971). J. Ultrastruct. Res. 37, 301-321. Devauchelle, G., Attias, J., Monnier, C., Barray, S., Cerutti, M., Guerillon, J. and Orange-Balange, N. (1985). Chilo iridescent virus. Curr. Top. Microbiol. Immunol. 116, 37-51. duPlessis, D. A., Mokhosi, G. and Hendry, D. A. (1991). Cell free translation and identification and the replicative form of Nudaurelia beta virus RNA. J . Gen. Virol. 12, 267-274. Emery, V. C. and Bishop, D. H. L. (1987). The development of multiple expression vectors for high level synthesis of eukaryotic proteins: expression of LCMV-N and AcNPV polyhedrin protein by a recombinant baculovirus. Protein Eng. 1, 359-366. Entwistle, P. F. and Evans, H. F. (1985). Viral control. In “Comprehensive Insect Physiology, Biochemistry and Pharmacology” (Eds G. A. Kerkut and L. I. Gilbert), pp. 347-412, Pergamon Press, Oxford. Epstein, D. A. and Thoma, J . A. (1975). Alkaline protease activity associated with the matrix protein of a virus infecting the cabbage looper. Biochem. Biophys. Res. Commun. 62. 478-484.

58

L. A. KING eta/.

Evans, H. F. and Entwistle, P. F. (1987). Viral diseases. In “Epizootiology of insect diseases” (Eds J. R. Fuxa and Y. Tanada), pp. 257-322. Wiley, New York. Evans, H. F. and Harrap, K. A. (1982). Persistence of insect viruses. In “Virus Persistence” (Eds B. W. J . Mahy, A. C. Minson and G. K. Darby), pp. 57-96. SGM Symposium, Vol. 33, Cambridge University Press, Cambridge. Farara, T. and Attias, J . (1983). Alkaline protease activity associated with iridescent virus type 6. F E M S Lett. 20, 351-355. Farara, T. and Attias. J . (1986). Further characterisation of an alkaline protease activity associated with iridescent virus type 6. Arch. Virol. 67, 307-314. Fediere, G., Philippe, R., Veyrunes, J. C. and Monsarrat. P. (1990). Biological control of the oil palm pest Latoia viridissima. Entomophaga 35, 347-354. Fischer, M., Schnitzler, P., Delius, H., Rosen-Wolff, A. and Darai, G. (1989). Molecular biology of insect iridescent virus type 6. In “Molecular Biology of Iridoviruses” (Ed. G. Darai), Developments in Molecular Virology. pp. 47-82. Kluwer Academic Publishers. Boston. Forstova, J., Krauzewicz, N. and Griffin, B. E. (1989). Expression of biologically active middle T antigen of polyoma virus from recombinant baculovirus. Nucl. Acids Res. 17, 1427-1443. Fowler, M. and Robertson, J . S. (1972). Iridescent virus infection in field populations of Wiseana cervinata (Lepid0ptera:Hepialidae) and Witlesia sp. (Lepidoptera:Pyralidae) in New Zealand. J. Invert. Pathol. 19, 154-157. Francki, R. I . B., Faquet. C. M., Knudson, D. L. and Brown, F. (1991). Classification and nomenclature of viruses. In “The Fifth ReDort of the International Committee on Taxonomy of Viruses.” Arch. Virol. Suppl. 2. pp. 117-1 23. Fraser, M. J. (1986). Ultrastructural observations of virion maturation in Autographa californica nuclear polyhedrosis virus infected Spodoptera frugiperda cell cultures. J . Ultrastruct. Mol. Struct. Res. 95, 189-1 95. Fraser, M. J . (1989). Expression of eucaryotic genes in insect cell cultures. In Vitro Cell. Dev. Biol. 25, 225-235. French, T. J.. Marshall, J . J . A. and Roy, P. (1990). Assembly of double-shelled, viruslike particles of bluetongue virus by the simultaneous expression of four structural proteins. J. Virol. 64, 5695-5700. Friesen, P. D. and Miller. L. K. (1986). The regulation of baculovirus gene expression. In “The Molecular Biology of Baculoviruses” (Eds W. Doerfler and P. Bohm). pp. 31-49. Springer-Verlag, Berlin. Friesen, P. D. and Rueckert, R. R. (1981). Synthesis of black beetle virus proteins in cultured Drosophilu cells: differential expression of RNAs 1 and 2. J. Virol. 37, 876-886. Friesen, P. D. and Rueckert, R. R. (1982). Black beetle virus: messenger for protein B is a subgenomic vital RNA. J. Virol. 42, 986-995. Friesen, P. D. and Rueckert, R. R. (1984). Early and late functions in a bipartite RNA virus: evidence for translational control by competition between vital RNAs. J. Virol. 49, 116-124. Fuchs, L. Y . , Woods. M. S. and Weaver, R. F. (1983). Viral transcription during Autogropha californica nuclear polyhedrosis virus infection: a novel RNA polymerase induced in infected Spodoptera frugiperda cells. J . Virol. 48, 641-646. Fukaya, M. and Nasu, S. (1966). A Chilo iridescent virus (CIV) from the stem borer, Chilo cervinata (Lepidoptera: Pyralidae). Appl. Entomol. Zool. 1, 69-72. Fuxa, J . R., Weidner. E. H. and Richter, A. R. (1992). Polyhedra without virions in a vertically transmitted nuclear polyhedrosis virus. J. Invert. Pathol. 60, 53-58. Gallagher, T . M., Friesen, P. D. and Rueckert, R. R. (1983). Autonomous

ADVANCES IN INSECT VIROLOGY

59

replication and expression of RNA 1 from Black Beetle virus. J. Virol. 46, 481489. Garzon, S. and Charpentier. G . (1991). Nodaviridae. In “Atlas of Invertebrate Viruses” (Eds J . R. Adams and J. R. Bonami), pp. 351-370, C R C Press, Boca Raton. Gauthier, L., Veyrunes. J. C., Cousserans, F. and Bergoin, M. (1992). Abstracts: 9th International Conference on Poxviruses and Iridoviruses, Les Diablerets, Switzerland. Gearing, K. L. and Possee, R. D . (1990). Functional analysis of a 603 nucleotide open reading frame upstream of the polyhedrin gene of Autographa californica nuclear polyhedrosis virus. J. Gen. Virol. 71, 251-262. Gildow, F. E. and D’Arcy, C. J . (1990). Cytopathology and experimental host range of Rhopalosiphum padi virus a small isometric R N A virus infecting cereal grain aphids. J . Invert. Pathol. 55, 245-257. Glocker, B., Hoopes. R. R. and Rohrmann, G. F. (1992). fn vitro transactivation of baculovirus early genes by nuclear extracts from Arctographa californica nuclear polyhedrosis virus-infected Spodoptera frugiperda cells. J . Virol. 66, 3476-3484. Glocker, B., Hoopes, R . R., Hodges, L. and Rohrmann, G . F. (1993). In vitro transcription from baculovirus late gene promoters: Accurate mRNA initiation by nuclear extracts prepared from infected Spodoptera frugiperdu cells. J . Virol. 67, 377 1-3776. Gombart, A . F., Pearson. M. N., Rohrmann. G . F. and Beaudreau, G . S. (1989). A baculovirus polyhedral envelope-associated protein: genetic location, nucleotide sequence and immunocytochemical characterization. Viro/ogy 169. 182-193. Goodwin, R. H., Tompkins, G . J . and McCauley. P. (1978). Gypsy moth cell lines divergent in viral susceptibility. I, Culture and identification. In Vitro 14, 485-494. Goodwin, R. H . , Adams, J. R. and Shapiro, M. (1990). Replication of the entomopoxvirus from Amsacfa moorei in serum-free cultures of a gypsy moth cell line. J. Invert. Pathol. 56. 19(k205. Gotz, P., Huger, A . M. and Kreig, A . (1969). Uber ein insektenpathogenes virus aus der gruppe der pockenviren. Naturwissenschuften 56, 145. Grace, T. D. C. (1962). The development of a cytoplasmic polyhedrosis in insect cells grown in vitro. Virology 18. 33-42. Granados, R. R. (1973a). Insect poxviruses: Pathology, morphology, and development. Miscellaneous Publications of the Entomology Society of America 9, 73-94. Granados, R. R. (1973b). Entry of an insect poxvirus by fusion of the virus envelope with the host cell membrane. Virology 5 2 , 305-309. Granados, R . R. (1980). Infectivity and mode of action of baculoviruses. Biotechnology and Bioengineering 23, 1377-1405. Granados, R. R. (1981). Entomopoxvirus infections in insects. In “Pathogenesis of Invertebrate Microbial Diseases” (Ed. E. W. Davidson), pp. 101-126. Allanheld, New Jersey. Granados, R. and Roberts, D . (1970). Electron microscopy of a poxlike virus infecting an invertebrate host. Virology 40, 23G243. Granados, R. R. and Naughton, M. (1975). Development of Amsacta moorei entomopoxviruses in ovarian and hemocyte cultures from Estigmene acrea larvae. Intervirology 5 , 62-68. Granados, R. R. and Williams, K. A . (1986). In vitro infection and replication of baculoviruses. In “The Biology of Baculoviruses, Vol I , Biological Properties and Molecular Biology” (Eds R . R. Granados and B. A . Federici). pp. 89-108, C R C Press, Boca Raton. Florida.

60

L. A. KING eta/.

Granados, R. R., Nguyen, T. and Cato, B. (1978). An insect cell line persistently infected with a baculovirus-like particle. lntervirology 10, 309-317. Greenfield, C . , Patel, G . , Clark, S . . Jones, N . and Waterfield, M. D . (1988). Expression of the human EGF receptor with ligand-stimulatable kinase activity in insect cells using a baculovirus vector. EMBO Journal 7 , 139-146. Gruidl, M. E . , Hall, R. L. and Moyer, R . W. (1992). Mapping and molecular characterization of a functional thymidine kinase from Amsacta moorei. Virology 186, 507-516. Grula, M. A . , Buller, P. L. and Weaver, R . F. (1981). a-amanitin-resistant viral R N A synthesis in nuclei isolated from nuclear polyhedrosis virus-infected Heliothis zea larvae and Spodoptera frugiperda cells. J . Virol. 38, 916-921. Guarino, L. A . and Summers, M. D . (1986a). Functional mapping of a tramactivating gene required for expression of a baculovirus delayed-early gene. J. Virol. 57, 563-571. Guarino, L. A . and Summers, M. D . (1986b). Interspersed homologous DNA of Autographa californica nuclear polyhedrosis virus enhances delayed-early gene expression. J . Virol. 60, 215-223. Guarino, L. A . , Gonzalez, M. A . and Summers, M. D. (1986). Complete sequence and enhancer function of the homologous D N A regions of Autographa californica nuclear polyhedrosis virus. J . Virol. 60, 224-229. Hall, R. L. and Hink, W. F. (1990). Physical mapping and field inversion gel electrophoresis of Amsacta moorei entomopoxvirus DNA. Arch. Virol. 110, 77-90, Hall, R . L. and Moyer, R. W. (1991). Identification, cloning, and sequencing of a fragment of Amsacta moorei entomopoxvirus D N A containing the spheroidin gene and three vaccinia virus-related open reading frames. J . Virol. 65, 6516-6527. Hall, R. L. and Moyer, R . W. (1993). Identification of an Amsacta spheroidin-like protein within the occlusion bodies of Choristoneura entomopoxviruses. Virology 192, 179-187. Hamblin, M., van Beek, N . A. M., Hughes, P. R . and Wood, H . A . (1990). Co-occlusion and persistence of a baculovirus mutant lacking the polyhedrin gene. Applied and Environmental Microbiology 56, 3051-3062. Hammock, B. D . (1985). Regulation of the juvenile hormone titre: degradation. In “Comprehensive Insect Physiology, Biochemistry and Pharmacology” (Eds G. A. Kerkut and L. I. Gilbert), pp. 431472. Pergamon Press, New York. Hammock, B. D . , Bonning, B. C., Possee, R. D . , Hanzlik, T. N. and Maeda, S . (1990). Expression and effects of the juvenile hormone esterase in a baculovirus vector. Nature, London 344, 458-461. Hanzlik, T. N . , Abdel-Aal, Y. A . I., Harshman, L. G. and Hammock, B. D . (1989). Isolation and sequencing of the cDNA clones coding for juvenile esterase from Heliothis virescens. J . Biol. Chem. 264, 12419-1 2425. Harrap, K. A . , Longworth. J. F., Tinsley, T. W. and Brown, K . W. (1966). A noninclusion virus of Gonometa podocarpi (Lepidoptera: Lasiocampidae). J. Invert. Pathol. 8, 27c272. Hatfill, S. J., Williamson, C., Kirby, R. and Von Wechmar. M. B. (1990). Identification and localisation of aphid lethal paralysis virus particles in thin tissue sections of the Rhopalosiphum padi aphid by in situ nucleic acid hybridisation. J. Invert. Pathol. 55, 265-271. Hawtin, R. E. (1993). The chitinase if Autographa californica nuclear polyhedrosis virus. PhD Thesis, Oxford Brookes University. Heimpel, A . M. and Angus, T. A . (1959). The site of action of crystalliferous bacteria in lepidopteran larvae. J . Insect Pathol. 1, 152-170.

ADVANCES IN INSECT VIROLOGY

61

Hendry, D. A. (1991). Nodaviridae of Invertebrates. In “Viruses of Invertebrates” (Ed. E. Kurstak), pp. 227-276, Marcel Dekker. New York. Hill-Perkins, M. S. and Possee, R. D. (1990). A baculovirus expression vector derived from the basic protein promoter of Autographa californica nuclear polyhedrosis virus. J . Gen. Virol. 71, 971-976. Himeno, M., Matsubara, F. and Hayashiya, K. (1973). The occult virus of nuclear polyhedrosis of the silkworm larvae. J . Invert. Pathol. 22, 292-295. Hink, W. F. (1970). Established insect cell line from the cabbage looper, Trichoplusia ni. Nature 226, 266467. Hoehberg, M. E. (1991). Extra-host interactions between a braconid endoparasitoid, Apanteles glomeratus, and a baculovirus for larvae of Pieris brassicae. J . Anim. Eco~.60, 65-77. Hofte, H. and Whiteley, H. R. (1989). Insecticidal crystal proteins of Bacillus thuringiensis. Microbiol. Rev. 53, 242-255. Hooft van Iddekinge, B. J . L., Smith, G. E. and Summers, M. D. (1983). Nucleotide sequence of the polyhedrin gene of Autographa californica nuclear polyhedrosis virus. Virology 131, 561-565. Hoopes, R. R. and Rohrmann, G. F. (1991) In vitro transcription of baculovirus immediate early genes: accurate mRNA initiation by nuclear extracts from both insect cells and human cells. Proc. Natl. Acad. Sci. USA 88, 4513-4517. Hosur, M. V . , Schmidt, T., Tucker, C., Johnson, J , E., Selling, B. H. and Rueckert, R. R. (1984). Black beetle virus+-crystallisation and particle symmetry. Virology 133, 119-127. Howard, S. C., Ayres, M. D. and Possee, R. D. (1986). Mapping the 5‘ and 3’ ends of Autographa californica nuclear polyhedrosis virus polyhedrin mRNA. Virus Res. 5 , 109-119. Hsieh, P. and Robbins, P. W. (1984). Regulation of asparagine-linked oligosaccharide processing. J . Biol. Chem. 259, 2375-2382. Huber, J. (1986). Use of baculoviruses in pest management programs. In “The Biology of Baculoviruses Vol. 11: Practical Application for Insect Control” (Eds R. R. Granados and B. A. Federici). pp. 181-202. CRC Press, Boca Raton, Florida. Huger, A. M., Kreig, A , , Emschermann, P. and Gotz, P. (1970). Further studies on Polypoxvirus chironomi, an insect virus of the pox group isolated from the midge Chironomus luridus. J . Invert. Pathol. 15, 253-261. Hughes, D. S . , Possee, R. D. and King, L. A. (1993). Activation and detection of a latent baculovirus resembling Mamestra brassicae nuclear polyhedrosis virus. Virology 194, 608-615. Hughes, P. R., Gettig, R. R. and McCarthy, W. J. (1983). Comparison of the time-mortality response of Heliothis zea to 14 isolates of Heliothis nuclear polyhedrosis virus. J . Invert. Pathol. 41, 246261. Hukuhara, T. (1962). Generation to generation transmission of the cytoplasmic polyhedrosis virus of the silkworm, Bombyx mori (Linnaeus). J . Invert. Pathol. 4, 32-35. Hukuhara, T., Xu, J. and Yano, K . (1990). Replication of an Entomopoxvirus in two Lepidopteran cell lines. J . Invert. Pathol. 56, 222-232. Hurpin, B. and Vago, C. (1963). Une maladie a inclusions cytoplasmiques fusiformes chez le Coleoptere Melolontha melolontha. Rev. Pathol. Vegetale Entomoi. Agr. 42, 115-117. Iatrou, K. and Meidinger, R. G. (1990). Tissue-specific expression of silkmoth chorion genes in vivo using Bombyx mori nuclear polyhedrosis virus as a transducing vector. Proc. Natl Acad. Sci., USA 87, 3650-3654. Jarvis, D. L. and Summers, M D. (1989). Glycosylation and secretion of human

62

L. A. KING eta/.

tissue plasminogen activator in recombinant baculovirus-infected cells. Mol. Cell. Biol. 9, 214-233. Jarvis. D. L., Oker-Blom, C. and Summers, M. D. (1990a). Role of glycosylation in the transport of recombinant glycoproteins through the secretory pathway of Lepidopteran insect cells. J . Cell. Biochem. 42, 181-191. Jarvis, D. L. Fleming, J-A. G. W . , Kovacs, G . R. Summers, M. D. and Guarino, L. A. (1990b). Use of early baculovirus promoters for continuous expression and efficient processing of foreign gene products in stably transformed Lepidopteran cells. BiolTechnology 8, 950-955. Jeang, K-T., Giam, C-Z.. Nerenberg, M. and Khoury, G . (1987). Abundant synthesis of functional human T-cell leukaemia virus type I p40” protein in eucaryotic cells by using a baculovirus expression vector. J . Virol. 61, 708-713. Joklik, W. K. (1962). Some properties of poxvirus deoxyribonucleic acid. J. Mol. Biol. 5, 265-274. Joyce, K. A,. Atkinson, A. E., Bermudez, I . , Beadle. D . J. and King, L. A. (1993). Synthesis of functional GABAA receptors in stable insect cell lines. FEBS Left. 335, 61-64. Jurkovicova, M . (1979). Activation of latent virus infections in larvae of Adoxophyes orana (Lepidoptera: Tortricidae) and Barathra brassicae (Lepidoptera: NOCtuidae) by foreign polyhedra. J. Invert. Pathol. 34, 213-223. Kaesburg. P. (1987). Organisation of pipartite insect virus genomes: the genome of black beetle virus. In “The Molecular Biology of the Positive Strand RNA Viruses” (Eds D. J . Rowlands. M. A. Mayo and B. W . J . Mahy), pp. 207-218. Academic Press. New York. Kamito, S. G., Majima, K. and Maeda. S. (1993). Identification and characterisation of the p35 gene of Bombyx mori nuclear polyhedrosis virus that prevents virus-induced apoptosis. J. Virol. 67, 455-463. Kataoka. H . , Troetschler, R. G.. Li, J. P.. Kramer, S. J., Carney, R. L. and Schooley, D. A. (1989). Isolations and identification of a diuretic hormone from the tobacco hornworm Manduca sexta. Proc. Natl. Acad. Sci. USA 86,2976-2950. Keddie, B. A . and Volkman, L. E. (1985). Infectivity difference between the two phenotypes of Autographa californica nuclear polyhedrosis virus: importance of the 64K envelope glycoprotein. J . Gen. Virol. 66, 1195-1200. Kelly, D. C. (1981). Non-occluded viruses. In “Pathogenesis of Invertebrate Microbial Diseases” (Ed. E. W. Davidson), pp. 3 9 4 0 , Allanheld Osmun. Kelly. D. C. (1982). Baculovirus replication. J. Gen. Virol. 63, 1-13. Kelly. D. C. (1985). Insect iridescent viruses. Curr. Top. Microbiol. Immunol. 116, 23-35. Kelly, D. C. (1987). Baculovirus pathogenesis. In “Molecular Basis of Virus Disease” (Eds W . C. Russell and J. W. Almond), pp. 289-302. Cambridge University Press, Cambridge. Kelly, D. C. and Robertson. J. S. (1973). Icosahedral cytoplasmic deoxyriboviruses. J . Gen. Virol. 20, 17-41. Kelly, D . C. and Tinsley, T. W. (1972). The proteins of iridescent virus types 2 and 6. J. Invert. Pathol. 19, 273-275. Kelly. D. C. and Tinsley, T. W. (1973). Ribonucleic acid polymerase activity associated with particles of iridescent virus types 2 and 6. J. Invert. Pathol. 22, 199-202. Kelly, D. C. and Vance, D. E. (1973). The lipid content of two iridoviruses. J. Gen. Virol. 21, 417-423. Kelly, D. C. and Lescott, T. (1981). Baculovirus replication: protein synthesis in Spodoptera frugiperda cells infected with Trichoplusia ni nuclear polyhedrosis virus. Microbiologica 4, 35-57.

ADVANCES IN INSECT VIROLOGY

63

Kelly, D. C., Lescott, T., Ayres, M. D.. Carey, D., Coutts, A. and Harrap, K. A. (1981). Induction of a non-occluded baculovirus persistently infecting Heliothis zea cells by Heliothis armigera and Trichoplusia ni nuclear polyhedrosis viruses. Virology 112, 174-189. Kelly, D. C., Brown, D. A., Ayres, M. D., Allen, C. J. and Walker, I. 0. (1983). Properties of the major nucleocapsid protein of Heliorhis zea singly enveloped nuclear polyhedrosis virus. J . Gen. Virol. 64,399-408. King, L. A. and Moore, N. F. (1988). Evidence for the presence of a genome linked protein in two insect picornaviruses, Cricket paralysis virus and Drosophila C virus. FEMS Lett. 50, 41-44. King, L. A. and Possee, R. D. (1992). “The Baculovirus Expression System. A Laboratory Guide.” Chapman and Hall, London, UK. King, L. A., Pullin, J. S. K. and Moore, N . F. (1984). Characterisation of a picornavirus isolated from a tumorous blood cell line of Drosophila melanogaster. Microbios 26, 121-127. King, L. A . , Pullin. J. S. K., Stanway, C . , Almond, J. W. and Moore, N. F. (1987). Cloning of the genome of Cricket paralysis virus and sequence of the 3’ end. Virus Res. 6, 331-344. King, L. A., Mann, S. G., Lawrie. A. M. and Mulshaw, S. H. (19Y1). Replication of wild-type and recombinant Autographa californica nuclear polyhedrosis virus in a cell line derived from Mamestra brassicae. Virus Res. 19, 93-104. King, L. A . , Mann, S. G., Marlow, S. A., Bermudez. I., Lawrie, A. M., Obosi, L. A. and Atkinson, A. E. (1992). Alternative insect cell lines for baculovirus expression vectors. In “Baculovirus and Recombinant Protein Production Processes” (Eds J. M . Vlak, E.-J. Schlaeger and A. R. Bernard), pp. 157-165, Editiones Roche. Basel. Kirschbaum, J. B. (1985). Potential implication of genetic engineering and other biotechnologies to insect control. Annu. Rev. Enrornol. 30, 51-70. Kitts, P. A. and Possee, R. D. (1993). A method for producing recombinant baculovirus expression vectors at high frequency. BiolTechniques 14. 810-817. Kitts, P. A , , Ayres. M. A. and Possee, R. D. (1990). Linearization of baculovirus DNA enhances the recovery of recombinant virus expression vectors. Nucl. Acids Res. 18, 566775672, Klenk, H.-D., Cramer, A , , Wagner, R.. Groner, A. and Kretzschmar, E. (1992). Processing of influenza virus hemagglutinin in insect cells: variations depending on the host and on co-expression with other viral proteins. I n “Baculoviruses and Protein Product Processes” (Eds J . M. Vlak. E.-J. Schlaeger and A. R. Bernard), pp. 166174, Editiones Roche, Basel. Klump, H., Beaumais, J. and Devauchelle, G. (1983). Structural and thermodynamic investigation of the Chilo iridescent virus (iridovirus type 6). Arch. Virol. 75, 269-276. Knebel-Morsdorf, D., Kremer, A. and Jahnel, F. (1993). Baculovirus gene ME53, which contains a putative zinc finger motif, is one of the major early-transcribed genes. J . Virol. 67, 753-758. Kool, M. and Vlak. J. M. (1993). The structural and functional organisation of the Aurographa californica nuclear polyhedrosis virus genome. Arch. Virol. 130, 1-16. Kool, M., van den Berg. P. M. M., Tramper, J . , Goldbach, R. W. and Vlak, J . M. (1993). Location of two putative origins of DNA replication of Autographa californica nuclear polyhedrosis virus. Virology 192, 94-101. Krappa, R. and Knebel-Morsdorf. D. (1991). Identification of the very early transcribed baculovirus gene PE-38. J . Virol. 65, 805-812. Krappa, R., Bohn-Krappa. A , , Jahnel, F., Doerfler. W. and Knebel-Morsdorf, D.

64

L. A. KING eta/.

(1992). Differential factor binding at the promoter of early baculovirus gene PE38 during viral infection: GATA motif is recognised by an insect protein. J. Virol. 66. 3494-3503. Krell. P. and Lee, P. E. (1974). Polypeptides in Tipula iridescent virus and in TIV-infected haemocytes of Galleria mellonella (L.) larvae. Virology 60, 315-326. Kuroda, K., Geyer, H.. Geyer, R., Doerfler, W. and Klenk, H-D. (1990). The oligosaccharides of influenza virus haemagglutinin expressed in insect cells by a baculovirus vector. Virology 174. 418-429. Kuzio. J.. Rohel. D. Z., Curry, C. J . , Krebs, A , , Carstens, E. B. and Faulkner, P. (1984). Nucleotide sequence of the p10 polypeptide gene of Autographa Californica nuclear polyhedrosis virus. Virology 139, 414-418. Langridge, W. (19834. Partial characterization of DNA from five Entomopoxviruses. J . invert. Pathol. 42, 369-375. Langridge, W. H. R . (1983b). Detection of Amsacta moorei Entomopoxvirus and Vaccinia virus proteins in cell cultures restrictive for poxvirus multiplication. J. Invert. Pathol. 42, 77-82. Langridge, W. H. R. (1983~).Virus DNA replication and protein synthesis in Amsacta moorei entomopoxvirus-infected Estigmene acrea cells. J . Invert. Pathol. 41,34\-349.

Langridge, W. H. R. (1984). Detection of DNA base sequence homology between entomopoxviruses isolated from Lepidoptera and Orthoptera. J . Invert. Pathol. 43. 4146. Langridge. W. H. R. and Roberts, D. W. (1982). Structural proteins of Amsacta moorei, Euxoa auxiliaris and Melanoplus sanguinpes, entomopoxviruses. J. Invert. Pathol. 39, 346353. Langridge. W. H. R., Bozarth, R. F. and Roberts, D. W. (1977). The base composition of entomopoxvirus DNA. Virology 76, 616-620. Lebdeva, 0. P. and Zelenko. A. P. (1972). Virus-like formations in larvae of Aedes and Culex mosquitoes. Med. Parazitof. Parazit. Bolezni 41, 490492. Li, Y . . Passareli, L. and Miller, L. K. (1993). Identification, sequence and transcriptional mapping of fef-3, a baculovirus gene involved in late and very late gene expression. J . Virol. 67, 5260-5268. Longworth. J . F. and Archibald, R. D. (1975). A virus of black beetle, Heteronychur arator (F.) (Coleoptera: Scarabaeidae). N . Z . J . Zool. 2, 233-236. Longworth, J. F. and Carey. G. P. (1976). A small RNA virus with a divided genome from Heteronychus arator (F.) (Coleoptera: Scarabaeidae). J . Gen. Virol. 33, 3140. Longworth, J . F. and Cunningham, J . C. (1968). The activation of occult nuclear-polyhedrosis viruses b; foreign nuclear polyhedra. J. Invert. Pathol. 10, 36 1-367. Longworth, J . F.. Robertson, J . S. and Payne, C. C. (1972). The purification and properties of inclusion body protein of the granulosis virus of Pieris brassicae. J. invert. Pathol. 19, 42-50. Longworth. J . F., Payne, C. C. and MacLeod, R. (1973). Studies on a virus isolated from Gonometa podocarpi (Lepidoptera: Lasiocampidae). J . Gen. Virol. 18, 119-125. Loudon, P. T. and Roy, P. (1991). Assembly of five bluetongue virus proteins expressed by recombinant baculoviruses: inclusion of the largest protein VP1 in the core and virus-like particles. Virology 180, 798-802. Lu, A. and Carstens. E. B. (1991). Nucleotide sequence of a gene essential for viral DNA replication in the baculovirus Autographa californica nuclear polyhedrosis virus. Virolog,v 181. 336-347.

ADVANCES IN INSECT VIROLOGY

65

Luckow, V. A. and Summers, M. D. (1988). Trends in the development of baculovirus expression vectors. BiolTechnology 6, 47-55. Lytvyn, V., Fortin, Y., Banville, M., Arif. B. and Richardson, C. (1092). Comparison of the thymidine kinase genes from three entomopoxviruses. J . Gen. Virol. 73, 3235-3240. Maddrell, S. (1986). In “Insect Neurochemistry and Neurophysiology” (Eds A. B. Borkovec and D. B. Gelman). pp. 35-74. Marcel Dekker. New York. Maeda, S. (1989a). Expression of foreign genes in insects using baculovirus vectors. Annu. Rev. Entomol. 34, 351-372. Maeda, S . (1989b). Increased insecticidal effect by a recombinant baculovirus carrying a synthetic diuretic hormone. Biochim. Biophys. Actu Res. Comrnun. 165, 1177-1183. Maeda, S., Kawai, T., Obinata, M.. Fujiwara, H . , Horiuchi, T., Saeki, Y . , Sato, Y. and Furusawa, M. (1985). Production of human a-interferon in silkworm using a baculovirus vector. Nature, London 315, 592-594. Maeda, S., Kamita, S. G. and Kataoka. H. (1991). The basic DNA-binding protein of Bombyx mori nuclear polyhedrosis virus: the existence of an additional arginine repeat. Virology 180, 807-810. Manousis, T. and Moore, N. F. (1987). Cricket Paralysis virus, a potential control agent for the olive fruit fly. Dacus oleae. Appl. Environ. Microbiol. 53. 142-. Marlow, S. A . , Palmer. C. P. and King, L. A. (1992). Cytopathic effects of Amsacru moorei entomopoxvirus infection on the cytoskeleton of Estigmene ucrea cells. Virus Res. 26, 41-55. Marlow, S. A., Billam, L. J., Palmer, C. P. and King, L. A. (1993). Replication and morphogenesis of Amsacfa moorei entomopoxvirus in cultured cells of Estigmene ucreu (salt marsh caterpillar). J . Gen. Virol. 74, 1457-1461. Martens, J. W. M.. Honee, G., Zuidema, D., van Lent, J . W. M.. Visser, B. and Vlak, J. M. (1990). Insecticidal activity of a bacterial crystal protein expressed by a recombinant baculovirus in insect cells. Appl. Environ. Microbiol. 56. 27642770. Martignoni, M. E. and Iwai, P. J. (1986) “A Catalogue of Viral Diseases of Insects, Mites and Ticks,” 4th edn. USDA Forest Service PNW-195. Washington, DC. Matsuura, Y . , Possee, R. D., Overton, H. A. and Bishop, D. H. L. (1987). Baculovirus expression vectors: the requirements for high level expression of proteins, including glycoproteins. J . Gen. Virol. 68, 1233-1250. Matthews, R. E. F. (1982). Classification and nomenclature of viruses. Intervirology 17, 1-200. Mazzone, H. M. (1985). Pathology associated with baculovirus infection. In “Viral Insecticides for Biological Control” (Eds K. Maramorosch and K. E. Sherman), pp. 81-120, Academic Press, London. McCarthy, W. J., Granados, R. R. and Roberts, D. W. (1974). Isolation and characterization of entomopox virions from virus-containing inclusions of Amsactu moorei (Lepidoptera: Arctiidae). Virology 59, 5940. McCutchen, B. F., Choudary, P. V., Crenshaw, R., Maddox, D., Kamita, S. G., Palekar, N., Volrath. S., Fowler, E., Hammock, B. D. and Maeda, S. (1991). Biotechnology 9, 848-852. McIntosh, A. H. and Kimura, M. (1974). Replication of the insect Chilo iridescent virus (CIV) in a poikilothermic vertebrate cell line. Intervirology 4, 257-267. McKinley, D. J . Brown, D. A., Payne, C. C. and Harrap. K. A. (1981). Cross infectivity and activation studies with four baculoviruses. Entomophuga 26, 79-90. Merryweather, A. T., Weyer, U., Harris, M. P. G., Hirst, M., Booth, T. and Possee. R . D. (1990). Construction of genetically engineered baculovirus insecti-

66

L. A. KING eta/.

cides containing the Bacillus thuringiensis susp. kurstaki HD-73 delta endotoxin. J. Gen. Virol. 71, 1535-1544. Miller, L. K. (1988). Baculoviruses as gene expression vectors. Annu. Rev. Microbiol. 42, 177-199. Miller, L. K. (1993). Baculoviruses. High-level expression in insect cells. Curr. Opin. Genet. Dev. 3, 97-101. Miller, L. K., Jewell, J . E. and Browne, D. (1981). Baculovirus induction of a DNA polymerase. J. Virol. 46. 275-278. Milner, R. J. and Lutton, G. G. (1975). The pathogenicity of an entomopoxvirus from Othonius batesi and its possible use as a control agent. Entomophaga 20, 2 13-220, Minion, F. C.. Coons, L. B. and Broom, J . R. (1979). Characterization of the polyhedral envelope of the nuclear polyhedrosis virus of Heliothis virescens. J. Invert. Pathol. 34, 303-307. Monnier. C. and Devauchelle. G. (1980). Enzyme activities associated with an invertebrate iridovirus: protein kinase activity associated with iridescent virus type 6 (Chilo iridescent virus). J. Virol. 35, 444-450. Moore, N. F. (1991a). The Nudaurelia p family of insect viruses. In “Viruses of Invertebrates” (Ed. E. Kurstak). pp. 277-286, Marcel Dekker, New York. Moore, N. F. (1991b). Identification, pathology, structure, and replication of insect picornaviruses. In “Viruses of Invertebrates” (Ed. E. Kurstak), pp. 287-300, Marcel Dekker, New York. Moore. N. F. and Eley. S. M. (1991). Picornaviridae: Picornaviruses of Invertebrates. In “Atlas of lnvertebrake Viruses” (Eds J . R. Adams and J . R. Bonami), pp. 371-386. CRC Press, Boca Raton. Moore, N. F. and Kelly, D. C. (1980). A comparative study of the polypeptides of three iridescent viruses by N-terminal analysis, amino acid analysis and surface labelling. J. Invert. Pathol. 36, 415-422. Moore. N . F. and Pullin, J. S. K. (1982). Plaque purification of cricket paralysis virus using an agar overlay on Drosophila cells. .I. Invert. Pathol. 39, 10-14. Moore, N. F., Kearns. A. and Pullin, J. S. K. (1980). Characterisation of Cricket paralysis virus-induced polypeptides in Drosophila cells. J. Virol. 33, 1-9. Moore, N. F., Pullin, J. S. K., Crump, W. A . L., Reavy. B. and Greenwood, L. K. (1981a). Comparison of two insect picornaviruses, Drosophila C and cricket paralysis viruses. Microbiologica 4, 359-370. Moore, N. F., Reavy, B. and Pullin, J. S. K . (1981b). Processingof Cricket paralysis virus induced polypeptides in Drosophila cells: production of high molecular weight polypeptides by treatment with iodoacetamide. Arch. Virol. 68, 1-8. Moore, N. F., Pullin. J. S. K . , Crump, W. A. L. and Plus, N. (1982). The proteins expressed by different isolates of Drosophila C virus. Arch. Virol. 74. 21-30. Moore, N. F., King, L. A. and Pullin. J. S. K. (1987). Insect Picornaviruses. In “The Molecular Biology of the Positive Strand RNA Viruses” (Eds D. Rowlands. M. Mayo and B. Mahy), pp. 67-74. Academic Press. New York. Mori. K. I . , Nakai, T.. Muroga, K.. Arimoto. M., Mushiake, K. and Furusawa, I. (1992). Properties of a new virus belonging to Nodaviridae found in larval striped jack (Pseudocarunx dentex) with nervous necrosis. Virology 187, 368-371. Moscardi. F. (1989). Use of viruses for pest control in Brazil: the case of the nuclear polyhedrosis virus of the soybean caterpillar, Anticarsia gemmatalis. Mem. Inst. Oswaldo Cruz 84, 51-56. Moscardi, F. and Correa Ferreira. B. S. (1988). Biological control of the soybean caterpillar. In “World Research Conference 111: Proceedings” (Ed. R. Shibles), pp. 703-71 1, Westview Press, London. Mueller, R., Pearson. M. N.. Russell, R. I. Q. and Rohrmann. G. F. (1990). A

ADVANCES IN INSECT VIROLOGY

67

capsid-associated protein of the multicapsid nuclear polyhedrosis virus of Orgyia pseudotsugata: genetic location, sequence, transcriptional mapping and immunocytochemical characterisation. Virology 176. 133-144. Murphy, F. A., Scherer, W. F.. Harrison. A . K., Dunne, M. W. and Gary, G. W. (1970). Characterisation of nodamura virus, an arthropod transmissible picornavirus. Virology 40,1008-1021. Murray, K. D. and Elkington, J . S. (1989). Environmental contamination of egg masses as a major component of transgenerational transmission of gypsy moth nuclear polyhedrosis virus (LdMNPV). J . Invert. Pathol. 53, 324-334. Murti, K. G., Goorha, R . and Granoff, A . (1985). An unusual replication strategy of an animal iridovirus. Adv. Virus Res. 30. 1-15. Newman, J. F. E . and Brown, F. (1973). Evidence for a divided genome in Nodamura virus, an arthropod borne picornavirus. J. Gen. Virol. 21. 371-384. Newman, J. F. E. and Brown, F. (1977). Further physicochemical characterisation of nodamura virus. Evidence that the divided genome occurs in a single component.

J. Gen. Virol. 38, 8SY5. Nissen, M . S. and Friesen, P. D . (1989). Molecular analysis of the transcriptional regulatory region of an early baculovirus gene. J . Virol. 63, 493-503. Olson, N. H . , Baker, T. S., Johnson, J . E. and Hendry. D. A . (1990). The three dimensional structure of frozen hydrated Nuduurelia capensis beta virus, a T equals 4 insect virus. J . Struct. B i d . 105, 11 1-122. Ooi, B. G . , Rankin, C . and Miller. L. K. (1989). Downstream sequences augment transcription from the essential initiation site of a baculovirus polyhedrin gene. J . Mol. Biol. 210. 721-736. Orange-Balange. N . and Devauchelle, G . (1982a). Lipid composition of an iridescent virus type 6 (CJV]. Arch. Virol. 73, 36S367. Orange-Balange, N. and Devauchelle, G. (1982b). Effect of infection with iridovirus type 6 on invertebrate cell phospholipids. Annals of Virology (Insfirure Pasteur) 133E, 205-214. O’Reilly, D . R. and Miller, L. K. (1989). A baculovirus blocks insect molting by producing ecdysteroid UDP-glucosyl transferase. Science 245, 1110-1 112. OReilly, D. R . . Crawford. A . M. and Miller, L. K. (1989). Viral proliferating cell nuclear antigen. Nature 337, 606. O’Reilly, D . R.. Miller, L. K. and Luckow. V. A . (1992). “Baculovirus Expression Vectors. A Laboratory Manual.” W. H . Freeman and Company, New York. Passarelli, A . L. and Miller, L. K . (lYY3a). Three baculovirus genes involved in late and very late gene expression: iel, ien and lef2. J . Virol. 67, 2149-2158. Passarelli, A. L. and Miller. L. K. (1093b). Identification and characterisation of left, a baculovirus gene involved in late and very late gene expression. J . Virol. 67, 3481-3488. Pearson, A . , Richardson, C. and Yuen. L. (1991). The 5’ noncoding region sequence of the Choristoneura hiennis Entomopoxvirus spheroidin gene functions as an efficient late promoter in the mammalian vaccinia expression system. Virology 180, 561-566. Pearson, M. N., Russell, R. L., Rohrmann, G . F. and Beaudreau, G . S. (1988). P39, a major baculovirus structural protein: immunocytochemical characterization and genetic location. Virology 167, 407-413. Pearson, M. N . , Bjornson, R.. Pearson, G. and Rohrmann, G . (1992). The Autograph californica baculovirus genome: evidence for multiple replication origins. Science 257, 1382-1384. Plus, N. (1989). Comparative study of picornaviruses in Drosophila cells in vitro. In “Invertebrate Cell System Applications” (Ed. J . Mitsuhashi), pp. 6 3 4 8 . CRC Press, Boca Raton.

68

L. A. KING eta/.

Podgewaite, J. D. (1985). Strategies for field use of baculoviruses. In “Viral Insecticides for Biological Control” (Ed. K. Maramorosch and K. E. Sherman), pp. 775-797. Academic Press, London. Pogo, B. G. T.. Bergoin. M., Dales, S. and Roberts, D. W. (1971). Enzymes associated with an insect poxvirus. Virology 43. 306-309. Ponsen, M. B. and de Jong, D. J. (1964). On a nuclear and cytoplasmic polyhedrosis of Adoxophyes reticulana HB. Entomophaga 9(3), 253-255. Possee, R. D. (1986). Cell-surface expression of influenza virus haemagglutinin in insect cells using a baculovirus vector. Virus Res. 5, 43-59. Possee, R. D. and Howard, S. C. (1987). Analysis of the polyhedrin gene promoter of the Autographa californica nuclear polyhedrosis virus. Nucleic Acids Res. 15, 10233-10248. Possee, R. D., Weyer, U. and King, L. A. (1990). Recombinant antigen production using baculovirus expression vectors. In “Control of Virus Diseases” (Eds N. J. Dimmock, P. D. Griffiths and C. R. Madely), pp. 53-76. Society General Microbiology Symposium Vol. 45. SGM, Cambridge University Press, Cambridge. Possee, R. D., Sun, T.-P., Howard, S. C., Ayres. M. D., Hill-Perkins, M. and Gearing, K. L. (199 1). Nucleotide sequence of the Autographa californica nuclear polyhedrosis virus 9.4 kbp EcoRI-I and -R (polyhedrin gene) region. Virology 185, 229-24 1. Possee, R. D., King, L. A.. Weitzman, M. D., Mann, S. G . , Hughes, D. S., Cameron, I. R., Hirst, M. L. and Bishop D. H. L. (1992). Progress in the genetic modification and field-release of baculovirus insecticides. In “The Release of Genetically Modified Microorganisms-REGEM 2”. (Eds D. E. S. Stewart-Tull and M. Sussman). Plenum Press, New York. Possee, R. D., Hirst, M., Jones, L. D. and Bishop, D. H. L. (1993) i n “Opportunities for Molecular Biology in Crop Production” (Eds D. J . Beadle, D. H. L. Bishop, L. G. Copping and D. W. Hollomon). pp. 23-36. BCPC Monograph 55. Pudney, M., Newman, J. F. E. and Brown, F. (1978). Characterisation of Kawino virus; an entero-like virus isolated from the mosquito Mansonia uniformis (Diptera: Culicidae). J . Gen. Virol. 40, 433441. Quiot, J. M., Bergoin, M. and Vago, C. (1975). Developpement et pathogenese d’un entomopoxvirus en culture cellulaire de lepidoptere. C. R. Acad. Sci., Serte D 280, 2273-2275. Rankin, C.. Ooi, B. G. and Miller, L. K. (1988). Eight base pairs encompassing the transcriptional start point are the major determinant for baculovirus polyhedrin gene expression. Gene 70. 39-49, Reavy, B. and Moore, N . F. (1982). The replication of small RNA-containing viruses of insects. Microbiologica 5, 63-84. Reavy, B. and Moore, N. F. (1984). In vitro translation of the genome of a small RNA virus from Trichoptusia ni. J . invert. Pathol. 44,244-248. Reavy, B., Crump, W. A. L., Greenwood, L. K. and Moore, N. F. (1982). Characterisation of a bipartite genome ribovirus isolated from Lymantria ninayi. Microbiologica 5, 305-319. Reinganum. C. (1973). Studies o n a non-occluded virus of the field crickets Telleogryllus oceanicus and T . comrnodus. MSc Thesis, Monash University, Melbourne, Australia. Reinganum, C. (1991). Tetraviridae. In “Atlas of Invertebrate Viruses’’ (Eds J. R. Adams and J . R. Bonami), pp. 387-392, CRC Press, Boca Raton. Reinganum, C., O’Loughlin, G. T. and Hogan, T. W. (1970). A non-occluded virus

ADVANCES IN INSECT VIROLOGY

69

of field crickets Telleogryllus oceanicus and T. cornmodus (Orthoptera: Gryllidae). J . Invert. Pathol. 16, 216220. Reinganum, C., Bashiruddin, J. B. and Cross, G. F. (1985). Boolarra virus: a member of the Nodaviridae isolated from Oncopera intricoides (Lepidoptera: Hapialidae). Intervirology 24, 10-17. Retnakaran, A. and Bird, F. T. (1972). Apparent hormone imbalance syndrome caused by an insect virus. J . Invert. Pathol. 20, 358-360. Roberts, D. W. (1970). Poxvirus of Arctiid larvae: Infectivity for greater wax moth. Bacteriol. Proc. 70th Meeting 151, 177. Roberts, D. W. and Granados, R. R . (1968). A poxlike virus from Amsacta moorei (Lepidoptera: Arctiidae). J . Invert. Pathol. 12, 141-143. Rohrmann, G. F. (1986). Polyhedrin structure. J . Gen. Virol. 67, 1499-1513. Rohrmann, G. F. (1992). Baculovirus structural proteins. J . Gen. Virol. 73, 749-761. Russell, R. L. Q. and Rohrmann, G. F. (1990a). A baculovirus polyhedron envelope protein: immunogold localization in infected cells and mature polyhedra. Virology 174, 177-184. Russell, R. L. 0. and Rohrmann, G. F. (1990b). The p6.5 gene region of a nuclear polyhedrosis virus of Orgyia pseudotsugata: DNA sequence and transcriptional analysis of four late genes. J . Gen. Virol. 71, 551-560. Russell, R. L. Q . , Pearson, M. N., Rohrmann, G. F. and Beaudreau, G. S. (1987). Characterization of baculovirus p10 synthesis using monoclonal antibodies. Virology 160, 9-19. Sanz, P., Veyrunes, J. C., Cousserans, F. and Bergoin, M. (1992). Abstracts, 9th International Conference on Poxviruses and Iridoviruses, Les Diablerets, Switzerland. Sato, T. (1989). Establishment of eight lines from newborn larvae of tortricids and characteristics including susceptibility to insect viruses. Bulletin of the Fruit Tree Research Station, Series A . (Yatabe) 16, 115-132. Scherer, W. F. and Hurlbut, H. S. (1967). Nodamura virus from Japan: a new and unusual arbovirus resistant to diethyl ether and chloroform. A m . J . Epidemiol. 86, 271-285. Scotti, P. D. (1976). Cricket paralysis virus replicates in cultured Drosophila cells. Intervirology 6, 333-342. Scotti, P. D., Longworth, J. F. E., Plus, N., Croizier, G. and Reinganum, C. (1981). The biology and ecology of strains of an insect small RNA virus complex. A d v . Virus Res. 26, 117-143. Scotti, P. D., Dearing, S. and Mossop, D. W. (1983). Flock house virus: a nodavirus isolated from Costelytra zealandica (White) (Coleoptera: Scarabaeidae). Arch. Virol. 75, 181-189. Shapiro, M. and Robertson, J. L. (1987). Yield and activity of gypsy moth (Lepidoptera: Lymantriidae) nuclear polyhedrosis virus recovered from survivors of viral challenge. J . Econ. Entomol. 80, 901-905. Sieburth, P. J. and Carner, G. R. (1987). Infectivity of an iridescent virus for larvae of Anticarsia gemmatalis (Lepidoptera: Noctuidae). J. Invert. Pathol. 49, 49-51. Smith, G. E. and Summers, M. D. (1979). Restriction maps of five Autographa californica MNPV variants, Trichoplusia ni MNPV, and Galleria mellonella MNPV DNAs with endonucleases SmaI, KpnI, BamHI, S a d , XhoI and EcoRI. J . Virol. 30, 828-838. Smith, G. E., Vlak, J. M. and Summers, M. D. (1983a). Physical analysis of Autogrupha californica nuclear polyhedrosis virus transcripts for polyhedrin and 10,000-molecular-weight protein. J . Virol. 45, 215-225. Smith, G. E., Fraser, M. J. and Summers, M. D. (1983b). Molecular engineering of

70

L. A. KING eta/.

the Autographa californica nuclear polyhedrosis virus genome: deletion mutations within the polyhedrin gene. J . Virol. 46, 584-593. Smith, G. E., Ju, G., Ericson, B. L., Moschera, J . , Lahm, H., Chizzonite, R. and Summers, M. D. (1985). Modification and secretion of human interleukin 2 produced in insect cells infected with a baculovirus expression vector. Proc. Nod. Acad. Sci. USA 82, 8404-8408. Smith, G. E., J u , G., Ericson, B. L.. Moschera, J . , Lahm, H., Chizzonite, R., Stohwasser, R . , Raab. K., Schnitzler, P., Janssen. W. and Darai, G. (1993). Identification of the gene encoding the major capsid protein of insect iridescent virus type 6 by polymerase chain reaction. J . Gen. Virol. 74. 873-879. Smith, I . R. L. and Crook, N . E. (1988). In vivo isolation of baculovirus genotypes. Virology 166, 240-244. Smith, K . M . (1955). What is a virus‘? Nature 175. 12-21. Smith, K. M. (1963). The cytoplasmic virus diseases. In “Insect Pathology: an Advanced Treatise” (Ed. E. A . Steinhaus). Vol. 1 , pp. 457-497. Academic Press, New York. Smith, K. M. (1967). “Insect Virology.” Academic Press, New York. Soltau, J. B., Fischer, M., Schnitzler, P., Scholz. J. and Darai, G. (1987). Characterisation of the genome of insect iridescent virus type 6 by physical mapping. J . Gen. Virol. 68, 2717-2722. Sparks, T . C . and Hammock, B. D. (1980). Comparative inhibition of the juvenile hormone esterase from Trichoplusia ni, Tenebrio monitor, and Musca domestica. Pesticide Biochem. Physiol. 14, 29G302. St Angelo, C., Smith, G. E . , Summers, M. D. and Krug, R. M. (1987). Two of the three influenza viral polymerase proteins expressed by using baculovirus vectors form a complex in insect cells. J . Virol. 61. 361-365. Steinhaus. E . A . and Dineen, J. P. (1960). Observations on the role of stress in a granulosis of the variegated cutworm. J . Insect Pathol. 2, 55-65. Stewart, L. M. D., Hirst, M., Lopez-Ferber, M., Merryweather, A . T., Cayley, P. J. and Possee. R . D. (1991). Construction of an improved baculovirus insecticide containing an insect-specific toxin gene. Nature 352. 85-88. Stoltz, D. B. (1973). The structure of icosahedral cytoplasmic deoxyriboviruses. An alternative model. J . Ultrastruct. Res. 43, 58-74. Stoltz, D . B. and Summers, M. D. (1972). Observations on the morphogenesis and structure of a hemocytic poxvirus in the midge Cliironomus attenuatus. J. Ultrastruct. Res. 40, 58 1-598. Stoltz, D . B . , Vinson, S. B. and MacKinnon, E. A. (1976). Baculovirus-like particles in the reproductive tracts of female parasitoid wasps. Can. J . Microbiol. 22(7), 1013-1 023. Struthers, J. K . and Hendry, D . A . (1974). Studies on the protein and nucleic acid components of Nudaurelia capensis beta virus. J . Gen. Virol. 22, 255-362. Summers, M. D . and Anderson, D. L. (1973). Characterization of nuclear polyhedrosis virus DNAs. J . Virol. 12, 13361346. Sutter, G. (1972). A poxvirus of the army cutworm. J . Invert. Pathol. 19, 375-382. Tajbakhsh, S. and Seligy. V. L. (1989). Molecular biology of Tipula iridescent virus. In “Molecular Biology of Iridoviruses, Developments in Molecular Virology” (Ed. G. Darai), pp. 98-132, Kluwer Academic Publishers, Boston. Tajbakhsh. S . , Lee, P. E., Watson, D . C. and Seligy. V. L. (1990). Molecular cloning. characterisation and expression of the Tipula iridescent virus capsid gene. J . Virol. 64,125-136. Takehara, K.. Ireland, D. and Bishop. D . H. L. (1988). Co-expression of the hepatitis B surface antigen and core antigens using baculovirus multiple expression vectors. J . Gen. Virol. 69. 2763-2777.

ADVANCES IN INSECT VIROLOGY

71

Teitelbaum, Z., Lazarovici, P . and Zlotkin, E. (1979). Selective binding of the scorpion venom insect toxin to insect nervous tissue. Insect Biochem. 9, 343-346. Thiem, S. M. and Miller, L. K. (1989). Identification, sequence and transcriptional mapping of the major capsid protein gene of the baculovirus Autographa californica nuclear polyhedrosis virus. J . Virol. 63, 2008-2018. Thomas-Orillard, M. (1988). Interaction between a picornavirus and a wild population of Drosophila melanogaster. Oecologia 75, 516 5 2 0 . Tinsley, T. W. and Kelly, D. C. (1970). An interim nomenclature system for the iridescent group of insect viruses. J . Invert. Pathol. 16, 470-472. Tomalski, M. D. and Miller, L. K. (1991). Insect paralysis by baculovirus-mediated expression of a mite-neurotoxin gene. Nature 352, 82-85. Tomalski, M. D. and Miller, L. K. (1992). Expression of the paralytic neurotoxin to improve baculoviruses as biopesticides. BioiTechnology 10, 545-549. Tomalski, M. D., Wu, J. and Miller, L. K. (1988). The location, sequence, transcription and regulation of a baculovirus DNA polymerase. Virology 167, 591-600. Tweeten, K. A., Bulla, L. A. and Consigli, R. A. (1980). Characterisation of an extremely basic protein derived from granulosis virus nucleocapsids. J . Virol. 33, 866-876. Vago, C. (1963). A new type of insect virus. J . Invert. Pathol. 5 , 275-276. Vago, C. and Bergoin, M. (1968). Viruses of invertebrates. Adv. Virus Res. 13, 247-303. Vail, P. V., Jay, D. L. and Hink, W. F. (1973). Replication and infectivity of the nuclear polyhedrosis virus of the Alfalfa looper Autographa californica, produced in cells grown in vitro. J . Invert. Pathol. 22, 231-237. Vaughn, J. L. (1979). Virus and rickettsia1 diseases. In “Insect Diseases Vol I” (Ed. G . E. Cantwell), pp. 4 9 4 5 , Marcel Dekker, New York. Vaughn, J. L., Goodwin, R. H., Thompkins, G. J. and McCawley, P. (1977). The establishment of two cell lines from the insect Spodoptera frugiperda (Lepidoptera: Noctuidae). In Vitro 13, 213-217. Vialard, J., Lalumiere, M., Vernet, T., Briedis, D., Alkhatib, G., Henning, D., Levin, D. and Richardson, C. (1990a). Synthesis of the membrane fusion and hemagglutinin proteins of measles virus, using a novel baculovirus vector containing the P-galactosidase gene. J . Virol. 64, 37-50. Vialard. J. E., Yuen. L. and Richardson, C. D. (1990b). Identification and characterization of a baculovirus occlusion body glycoprotein which resembles spheroidin, an entomopoxvirus protein. J . Virol. 64,5804-5811. Vlak, J. (1993a) Genetic engineering of baculoviruses. In “Opportunities for Molecular Biology in Crop Production” (Eds D. J. Beadle, D. H. L. Bishop, L. G. Copping and D. W. Hollomon), pp. 11-22, BCPC Monograph 55. Vlak, J. M. (1993b). Genetic engineering of baculovirus insecticides for insect control. In “Molecular Approaches to Pure and Applied Entomology” (Eds M. J. Whitten and J. G. Oakeshott), pp. 9@127, Springer-Verlag, New York. Vlak, J. M., Klinkenberg, F. A , , Zaal, K. J. M., Usmany, M., Klinge-Roode, E. C., Geervliet, J. B. F., Roosien, J. and van Lent, J. W. M. (1988). Functional studies on the p10 gene of Autographa californica nuclear polyhedrosis virus using a recombinant expressing a pl0-P-galactosidase fusion gene. J . Gen. Virol. 69, 765-776. Volkman, L. E. and Keddie, B. A. (1990). Nuclear polyhedrosis virus pathogenesis. Semin. Virol. I , 249-256. Volkman, L. E., Summers, M. D. and Hsieh, C. H. (1976). Occluded and non-occluded nuclear polyhedrosis virus grown in Trichoplusia ni: comparative neutralization, comparative infectivity, and in vitro growth studies. J . Virol. 19, 82C832

72

L. A. KING eta/.

Walter, C . , Zlotkin. E. and Rathmeyer, W. (1976). Action of different toxins from the scorpion Androctonus australis on a locust nerve-muscle preparation. J. Insect Physiol. 22, 1187-1 194. Wang, X . and Kelly, D. C. (1983). Baculovirus replication: purification and identification of Trichoplusia ni nuclear polyhedrosis virus-induced DNA polymerase. J. Gen. Virol. 64, 2229-2236. Ward, V. K. and Kalmakoff, J. (1987). Physical mapping of the D N A genome of insect iridescent virus type 9 from Wiseana spp. larvae. Virology 160, 507-510. Ward, V. K. and Kalmakoff, J . (1991). Invertebrate iridoviridae. In “Viruses of Invertebrates” (Ed. E. Kurstaki), pp. 197-226, Marcel-Dekker, New York. Weitzman, M. D., Possee, R . D. and King, L. A . (1992). Characterisation of two variants of Panolis flammea multiple nucleocapsid nuclear polyhedrosis virus. J. Gen. Virol. 7, 1881-1886. Westwood, J. C . N., Harris, W. J . , Zwartouw, H. T . , Titmus, D . H. J. and Appleyard, G. (1964). Studies on the structure of vaccinia virus. J . Gen. Microbiol. 34. 67-78. Weyer, U.and Possee, R. D. (1988). Functional analysis of the p10 gene 5’ leader sequence of the Autographa californica nuclear polyhedrosis virus. Nucleic Acids Res. 16, 3635-3653. Weyer, U.and Possee, R. D. (1989). Analysis of the promoter of the Autographa californica nuclear polyhedrosis virus p10 gene. J. Gen. Virol. 70, 203-208. Weyer, U. and Possee, R. D . (1991). A baculovirus dual expression vector derived from the Autographa californica nuclear polyhedrosis virus polyhedrin and p10 promoters: co-expression of two influenza virus genes in insect cells. J . Gen. Virol. 72, 2967-2974. Weyer, U . , Knight, S. and Possee, R. D. (1990). Analysis of very late expression by Autographa californica nuclear polyhedrosis virus and the further development of multiple expression vectors. J. Gen. Virol. 71, 1525-1534. Whitford, M., Stewart, S., Kuzio, J. and Faulkner. P. (1989). Identification and sequence analysis of a gene encoding gp67, an abundant envelope glycoprotein of the baculovirus Autographa californica nuclear polyhedrosis virus. J. Virol. 63, 1393-1 399. Whitt, M. A. and Manning, J. S. (1988). A phosphorylated 34-kDa protein and a subpopulation of polyhedrin are thiol linked to the carbohydrate layer surrounding a baculovirus occlusion body. Virology 163, 33-42. Williams, G. V., Rohel, D. Z . , Kuzio, J. and Faulkner, P. (1989). A cytopathological investigation of Autographa californica nuclear polyhedrosis virus p10 gene function using insertion/deletion mutants. J. Gen. Virol. 70. 187-202. Williams, R. C.-and Smith, K. M. (1957). A crystallizable insect virus. Nature 179, 119-1 20. Williams, R. C . and Smith, K. M. (1958). The polyhedral form of the Tipula iridescent virus. Biochim. Biophys. Acta 28. 464-469. Williamson, C. and Rybicki, E. P. (1989). A comparative study on the cell-free translation of the genomic RNAs of two aphid picorna-like viruses. Arch. Virol. 109, 59-70. Williamson, C . , Von Wechmar. M. B. and Rybicki, E. P. (1989). Further characterisat ion of Rhopalosiphum padi virus of aphids and comparison of isolates from South Africa and Illinois. J. Invert. Pathol. 54, 85-96. Willison, J. H. M. and Cocking, E. C. (1972). Frozen fractured viruses: a study of virus structure using freeze etching. J. Microsc. 95, 397411. Wilson, M. E . . Mainprize, T . H., Friesen, P. D. and Miller, L. K. (1987). Location, transcription and sequence of a baculovirus gene encoding a small arginine-rich polypeptide. J. Virol. 61, 661-666.

ADVANCES IN INSECT VIROLOGY

73

Wood, H. A . , Hughes, P. R., Johnson, L. B . and Langridge, W . H. R. (1981). Increased virulence of Autographa californica nuclear polyhedrosis virus by mutagenesis. J. Invert. Pathol. 38, 236-241. Wrigley, N. G. (1969). A n electron microscopy study of the structure of Sericesthis iridescent virus. J . Gen. Virol. 5, 129-134. Xeros, N . (1954). A second virus disease of the leatherjacket, Tipula paludosa. Nature 174, 562-565. Yang, C . L., Stetler, D . A. and Weaver, R. F. (1991). Structural comparison of the Autographa californica nuclear polyhedrosis virus-induced RNA polymerase and the three nuclear RNA polymerases from the host, Spodoptera frugiperda. Virus Res. 20, 251-264. Yuen, L . , Dionne, J . , Arif, B . and Richardson, C. (1990). Identification and sequencing of the spheroidin gene of Choristoneura biennis entomopoxvirus. Virology 175, 427-433. Yuen, L., Noiseux, M. and Gomes, M. (1991) D N A sequence of the nucleoside triphosphate phosphohydrolase I (NPH I ) of the Choristoneura biennis entomopoxvirus. Virology 182, 40-06. Yule, B. G. and Lee, P. E. (1973). A cytological and immunological study of Tipula iridescent virus-infected Galleria mellonella larvae haemocytes. Virology 51, 409423. Zhong, W., Dasgupta, R . and Rueckert, R. (1992). Evidence that the packaging system for nodaviral RNA2 is a bulged stem-loop. Proc. Natl. Acad. Sci. USA 23. 11146-11150. Zummer, M. and Faulkner, P. (1979). Absence of protease in baculovirus polyhedra grown in vitro. J. Invert. Pathol. 33, 383-384.

This Page Intentionally Left Blank

Genetic Mechanisms of Early Neurogenesis in Drosophila rn elanogaster Jose A. Campos-Ortega Institut fur Entwicklirngshiologie. Uriiver.sitat zu Kiilri, 0-50931 Koln. Germany

1 Introduction 75 2 Cellular basis of neurogenesis 76 3 The segregation of neuroblasts 77 4 Cell interactions in the neuroectoderm 79 5 Neuralizing signals and intrinsic factors 80 6 Genetics of neurogenesis 80 7 The neurogenic genes are functionally interrelated 84 8 Physical interactions of Notch and Delta 85 9 The epidermal decision is controlled by the E(sPL)-C 87 10 The neural decision is controlled by the proneural genes 88 11 Interactions between neurogenic and proneural genes 89 12 Conclusions 90 Acknowledgements 95 References 96

1 Introduction

In insects, the cells of the central nervous system (CNS) are generated by the proliferation of progenitor cells called neuroblasts which develop from a special neurogenic region of the ectoderm. In Drosophila melanogasfer,the neuroectoderm consists of two different parts, the ventral neuroectoderm (VNE), from which the ventral cord and the suboesophageal ganglion will develop, and the procephalic neuroectoderm, from which the brain hemispheres emerge (Poulson, 1950; Hartenstein and Campos-Ortega, 1984; Technau and Campos-Ortega, 1985). Both regions give rise to neural progenitor cells; however, whereas the cells of the VNE have to decide between developing either as neuroblasts or as epidermoblasts (progenitor cells of the epidermis), there is no clear picture as to how the procephalic neuroectoderm is organized and how neuroblasts develop from this region (Hartenstein and Carnpos-Ortega, 1984; Technau and Campos-Ortega, ADVANCES IN INSECT PHYSIOLOGY VOL. 25 ISBN &1?424?25-7

CopvrrRhr 0f9Y4 Academic Press Lintired A / / righrs of repruducrion i n any form reserved

76

JOSE A. CAMPOS-ORTEGA

1985; Jiirgens et al. , 1986; Stiittem and Campos-Ortega, 1991). Accordingly, this review will deal with the VNE. A hypothesis has been recently formulated to account for the development of progenitor cells of sensory organs within the epidermis of Drosophila (Ghysen and Dambly-Chaudiere, 1989; see Jan and Jan, 1990; Ghysen et al., 1993, for reviews). Appropriately modified, this hypothesis can also be applied to the segregation of neuroblasts and epidermoblasts in the VNE. The hypothesis proposes a sequence of three steps to explain the segregation of epidermal and central neural progenitor cells. In the first step, all cells of the VNE acquire the capability to develop as neuroblasts, whereby contiguous groups of about four to five cells, so-called proneural clusters, are each enabled to give rise to a particular type of neuroblast. In the second step, one cell in each group is singled out by intervening neuralizing signals and segregates into the space between ectoderm and mesoderm to develop as a particular type of neuroblast. In the third step, the neuroblast sends inhibitory, epidermalizing signals to the surrounding cells preventing them from following a neural fate and permitting them to assume an epidermal fate. Therefore, three operations are included in this scheme: one confers upon the VNE cells the capability to produce neural or epidermal progenies, another permits the separation of the two classes of progenitor cells, and the third specifies particular types of neuroblasts and epidermoblasts. The first two of these operations are brought about by the participation of a large number of gene products, functionally interconnected to form a complex network. Major constituents of this network are two groups of regulatory proteins, which allow the development of the neural and epidermal cell fates, respectively, and a group of membraneassociated proteins which serve to pass the regulatory signals between neighbouring cells in the neuroectoderm and transduce them to the nuclei of the interacting cells. 2 Cellular basis of neurogenesis

In wild-type Drosophila melanogaster, the neuroectoderm becomes morphologically manifest during the initial phase of germ band elongation in stage 7 (stages of embryogenesis according to Campos-Ortega and Hartenstein, 1985). Within the segmented germ band, the ectodermal layer becomes subdivided into a lateral part, comprising small cylindrical cells, and a medial part with large cuboidal cells (Hartenstein and Campos-Ortega, 1984; see Fig. 1). The lateral part will differentiate during later stages into the dorsal epidermis, with its annexes, whereas the medial part is the VNE itself, from which the ventral cord and the ventral epidermis will develop (Technau and Campos-Ortega, 1985). Virtually all cells of the VNE, i.e. 100 rows of mediolaterally arranged cells on either side of the midline, with

NEUROGENESIS IN DROSOPHlLA

77

approximately nine cells in each row, enlarge to become conspicuously different from the cells of the dorsal epidermal anlage. After the neuroblasts have segregated from the VNE into the interior of the embryo, the cells remaining on the outside shrink. In striking contrast to the case in the fruit fly, only single cells of the W E within groups comprising several cells are reported to enlarge during early neurogenesis in grasshoppers; the enlarged cells are the prospective neuroblasts themselves, which will segregate from the remaining cells (Bate, 1976, 1982; Doe and Goodman, 1985a). 3 The segregation of neuroblasts

In Drosophila, segregation of neuroblasts lasts for approximately 3 h and is discontinuous, proceeding in discrete pulses which give rise to different subpopulations of neuroblasts. Originally, three pulses, which give rise to SI, SII and SIII neuroblasts, were distinguished (Hartenstein and CamposOrtega, 1984). However, Doe (1992), using molecular markers, has distinguished two additional pulses of segregation which give rise to two additional subpopulations of neuroblasts, SIV and SV. Single cells among the large cells of the VNE undergo conspicuous shape changes, leave the ectoderm and move internally to form the neural primordium. The cells immediately adjacent to each of the SI neuroblasts establish relationships with them by means of long basal processes, which transiently surround the segregated neuroblast forming a sort of sheath. The ensheathing processes are later retracted and the prospective epidermal progenitor cells diminish in size. Therefore, there is sufficient contact between the VNE cells both before and during lineage segregation to enable cellular interactions to take place. After their segregation, the neuroblasts form a monolayer between ectoderm and mesoderm (Fig. 1). Due to their sequential segregation and subsequent behaviour, various subpopulations of neuroblasts can be distinguished from each other on the basis of their size and location within the array (Hartenstein and Campos-Ortega, 1984; Hartenstein ef af., 1987; Doe et al., 1988; JimCnez and Campos-Ortega, 1990; Cui and Doe, 1992; Doe, 1992). Doe (1992) has reported on markers that permit reliable identification of individual neuroblasts, even when in late stages the pattern becomes more complex. Midway through stage 11, another group of neural progenitor cells develops in each segment from the mesectodermal cells, that comprises the unpaired median neuroblast (MNB) and a number of small, paired cells called midline precursor cells (MPs; Bate, 1976; Thomas et af., 1984; Klambt et af., 1991). All of these midline cells in each segmental group form an ovoid cluster which straddles the boundary between adjacent neuromeres. The first maps of neuroblasts in insects (Bate, 1976; Hartenstein and

JOSE A. CAMPOS-ORTEGA

FIG. 1 Illustration of the morphological modifications of the ectoderm accompanying neuroblast segregation. The dark shadowing in all drawings indicates neuroblasts and their progeny, as well as ectodermal cells with capabilities to produce neuroblasts. Epidermal cells are shown in lighter shadowing. (A) One row of neuroectodermal ( W E ) cells, from the midline to the dorsal epidermal anlage (DEA), is shown. All cells of the VNE enlarge considerably in stage 8 and become conspicuously different from the cells of the DEA. (B) Segregation of most of the SI neuroblasts (SI NB) takes place from median and lateral regions of the W E . Here, single cells undergo conspicuous shape changes to leave the outer layer and move

NEUROGENESIS IN DROSOPHILA

79

Campos-Ortega, 1984) were obtained by reconstructing serial sections. The advent of antibody staining techniques (Hartenstein et al., 1987; Doe et al., 1988; JimCnez and Campos-Ortega, 1990; Martin-Bermudo et al., 1991; Skeath and Carroll, 1992; Doe, 1992) and other markers, like lacZ expressing ‘enhancer trap’ lines (O’Kane and Gehring, 1987; Bier et al., 1989; Bellen et al., 1989; Campos-Ortega and Haenlin, 1992; Cui and Doe, 1992; Doe, 1992), has permitted a more accurate mapping of the embryonic neuroblasts. The most complete map of neuroblasts in the Drosophila embryo has been presented by Doe (1992), who describes an approximate number of 30 neuroblasts per hemisegment, including a glioblast that generates longitudinal glia. Klambt et al. (1991), in a careful analysis of the midline progenitor cell clusters using a number of ‘enhancer trap’ lines, describe eight cells per segmental midline cluster, i.e. one median neuroblast, three neuronal precursors, three glial precursors and one cell whose identity is not well established. 4 Cell interactions in the neuroectoderm

Two pieces of evidence indicate that the decision of the VNE cells to adopt the epidermal or the neural fate is mediated by cell-cell interactions. First, laser ablation experiments carried out in grasshoppers show that the cells remaining in the VNE after the neuroblasts have segregated are not firmly committed to their fate (Taghert el al., 1984; Doe and Goodman, 1985b). Under normal circumstances, these cells would develop as epidermoblasts; however, in the conditions of the experiment they may adopt a neural fate instead. These results provide the best evidence in support of the idea that

internally, where they become arranged in two longitudinal stripes. Notice that the VNE cells immediately adjacent to each of the SI NB establish intimate contact with the latter cells by means of ensheathing processes. Regulatory signals are assumed to be sent preceding the segregation. (C) SII neuroblasts segregate mainly from intermediate regions of the VNE, where the cells enlarge again and single cells separate from the germ layer. The processes which ensheathed the SI neuroblasts are retracted and the prospective epidermal progenitor cells diminish in size. Ectodermal cells are shown pale when they have lost the capability to produce neuroblasts and have entered the epidermal pathway. (D) Most SIII and subsequent neuroblasts (SIV and SV, according to Doe, 1992) segregate from the median cells and, again in this case, there is a conspicuous enlargement of the ectodermal cells that precedes the separation of the neuroblasts from the germ layer. (E) Once all the neuroblasts have segregated, the cells remaining on the outside shrink to form the epidermal sheath. Immediately after segregation, neuroblasts round up and begin to divide asymmetrically to produce ganglion mother cells (GMC) and neurones (or ganglion cells, GC). Since neuroblasts decrease in size after divisions, SI can be distinguished from SII and from SIII NB on the basis of their smaller size.

80

JOSE A. CAMPOS-ORTEGA

the prospective epidermoblasts are inhibited by the neuroblasts from adopting the neural fate (Doe and Goodman, 1985b). Second, results of cell transplantations in Drosophila suggest that regulatory signals in this species pass between the ectodermal cells and that these signals are involved in the cells’ commitment to one of the two developmental fates (Technau and Campos-Ortega, 1986a, 1987; Technau, 1986, 1987; Technau et al., 1988; Becker and Technau, 1990; Stiittem and Campos-Ortega, 1991). These experiments suggest that two types of signals, epidermalizing and neuralizing, operate during lineage segregation. The epidermalizing signals are required to inhibit the neural fate in the cells of the proneural cluster destined to become epidermoblasts; the neuralizing signals to implement the primary neural fate. 5 Neuralizing signals and intrinsic factors

Intrinsic properties of the neuroectodermal cells are of paramount importance in deciding whether any given cell can adopt a neural fate upon its transplantation in the neuroectoderm. For example, proctodeal cells (Stuttem and Campos-Ortega, 1991), endodermal (Technau and CamposOrtega, 1986b) and, probably, also mesodermal cells (Beer et al., 1987) do not develop neural fates upon transplantation into the VNE, but produce proctodeal, endodermal and mesodermal lineages. Two different factors appear thus to be involved in the neural decision of ectodermal cells. One is a particular environment that permits neurogenesis; the other factor is the competence of the transplanted cells to respond to whatever signals might be provided by the environment and permit neurogenesis; this competence is present in cells of the VNE, the DEA or the foregut anlage, but absent from cells of the proctodeal anlage. Therefore, intrinsic properties of the transplanted ectodermal cells as, for example, particular transcription factors, may confer upon them a bias toward adoption of the neural fate and facilitate a higher degree of autonomy in the choice. A recent study by Luer and Technau (1992) shows that DEA cells have intrinsic epidermogenic capabilities, which confer upon them a clear bias to develop as epidermoblasts in the absence of cell interactions. The authors have cultivated individual cells from well-defined regions of the ectoderm and observed what kind of progenies they form. They found a gradient of epidermogenic capabilities in the ectoderm, decreasing from dorsal to ventral, and another gradient of neurogenic capabilities in the opposite direction. 6 Genetics of neurogenesis

In Drosophila melanogaster, the correct separation of neuroblasts and epidermoblasts is controlled by two groups of genes (Table l), the so-called

TABLE 1

Genes involved in early neurogenesis of Drosophila melanoaoster

Genes

Loss-of-function phenotype

Proneural genes AS-C Neural hypoplasia vnd

da

Neural hypo p 1asi a Neural hypoplasia

Neurogenic genes N Neural hyperplasia

Gene product

Possible function

bHLH

Regulation of transcription

?

3

bHLH

Regulation of transcription

Transmembrane EGF-like repeats

Adhesion signal receptor?

Relevant references JimCnez and Campos-Ortega (1979, 1987, 1990), White (1980), Dambly-Chaudikre and Ghysen (1987), Villares and Cabrera (1987), Alonso and Cabrera (1988) White (1980), Jimenez and Campos-Ortega (1987, 1990) Caudy et al. (1988a), Caudy et al. (1988b)

Poulson (1937), Wharton et al. (1985), Kidd et al. (1986), Fehon et al. (1990). Heitzler and Simpson (1991) Lehmann et al. (1981). Vassin et al. (1987), Kopczynski et al. (1988). Haenlin et al. (1990), Fehon et al. (1990), Heitzler and Simpson (1991) Lehmann et al. (1983). Klambt et al. (1989). Knust et al. (1992). Schrons et al. (1992) Lehmann et al. (1981), Smoller et al. (1990)

DI

Neural hyperplasia

Transmembrane EGF-like repeats

Adhesion signal source?

E(sPL)-C

Neural hyperplasia Neural h yperplasia Neural hyperplasia

bHLH

Regulation of transcription

Nuclear protein

3

Homeobox bacterial repressors Transmembrane

Regulation of transcription?

Lehmann et al. (1981). Boulianne et al. (1991)

Channel? Transporter?

Lehmann et al. (1981), Rao et al. (1990)

3

Bourouis et al. (1989, 1990)

?

Knust et al. (1987b), Hartley et al. (1988), Delidakis et al. (1991), Schrons et al. (1992)

mum neu

big brain shaggy groucho

Neural h yperplasia Neural hyperplasia Neural hyperplasia

Serine-threonine kinase Nuclear protein

82

JOSE A. CAMPOS-ORTEGA

neurogenic genes (Poulson, 1937; Lehmann et al., 1981, 1983) on the one hand, and the so-called proneural genes (Ghysen and Dambly-Chaudiire, 1989, 1990; Romani et al., 1989), i.e. the various members of the achaete-scute complex (AS-C), the locus ventral nervous system condensation defective (vnd), and daughterless, and probably other as yet unidentified genes, on the other hand (Jimenez and Campos-Ortega, 1979, 1987, 1990; White, 1980; White et al., 1983; Caudy et al., 1988a; Brand and CamposOrtega, 1988). Neurogenic and proneural genes are currently referred to as though they form two different groups, and such a distinction is justified by the phenotypes of their mutations (to be discussed below). However, I would like to emphasize that the products of all of these genes are apparently involved in the same complex genetic network and together contribute to the process of separation of neural and epidermal cell progenitors. Since the functions of the AS-C, vnd and daughterless promote neural development, they have been generically called “proneural genes” (Ghysen and Dambly-Chaudiere, 1989, 1990; Romani et a f . , 1989). Poulson (1937) called Notch (the first neurogenic gene discovered) a “neurogenic” gene following the convention in Drosophila genetics of naming a gene according to the phenotype of the mutation that leads to its discovery. Accordingly, other genes which cause the same phenotype as Notch have also been called neurogenic. Loss of function of any of the neurogenic genes causes most ectodermal cells to develop as neuroblasts (Fig. 2) (Lehmann et al., 1981, 1983; Jiminez and Campos-Ortega, 1982). Neuralization of the ectoderm of neurogenic mutants follows the pattern of neuroblast segregation in the wild-type and proceeds in pulses (Campos-Ortega and Haenlin, 1992). In the mutants, all the VNE cells from which neuroblasts normally segregate at each pulse take on neural fate until, in mid-stage 11, all cells in the VNE have adopted neural fate. Regions from which sensory organs develop exhibit also a high proportion of neural cells (Hartenstein and Campos-Ortega, 1986; Ghysen and Dambly-Chaudiere, 1990; Goriely et al., 1991). Therefore, neurogenesis is initiated in these mutants by a much higher number of neuroblasts and sensory organ mother cells than in the wild-type and, consequently, the mutant embryos die with a highly hyperplasic central and peripheral nervous system and lacking ventral and cephalic epidermis. Thus, the wild-type functions of the neurogenic genes are formally required to suppress neural development of a large fraction of ectodermal cells and allow them to develop as epidermoblasts. The complete loss of the shaggy function leads to all the embryonic cells taking up neural fate; intermediate phenotypes are obtained when maternal or zygotic gene expression is affected (Bourouis et al., 1989; Sirnpson and Carteret, 1989). The molecular nature of the shaggy product, a putative serine-threonine kinase (Bourouis et al., 1990), together with the phenotype of its mutants, strongly suggests a role for this gene in the transmission of

NEUROGENESIS IN DROSOPHILA

a3

Neuroblast Epidermoblast

Neuroblast

Neuroblast

- Notch Delta

Epidermoblast Epidermoblast or cell death FIG.2 (A) Shows two interacting cells of the VNE in the wild-type; the white ovals represent the cell nuclei. The proneural and the neurogenic genes, including the genes of the E(sPL)-C,encode the proteins of a regulatory signal chain which allows the cells to develop either as neuroblasts or as epidermoblasts. The proneural genes are required to regulate the genetic activity of the neuroblasts, those of the E(sPL)-C to give neuroectodermal cells access to epidermal development. (B) Mutation of a neurogenic gene results in the development of all neuroectodermal cells as neuroblasts. This is probably due to derepression of the proneural genes in the cells which would have normally developed as epidermoblasts. (C) Mutation in the proneural genes results in either the development of additional neuroectodermal cells as epidermoblasts, at the expense of neuroblasts, or in cell death. This is probably due to inactivation of the proneural genes in the cells which would have normally developed as neuroblasts.

84

JOSE A. CAMPOS-ORTEGA

signals between neuroectodermal cells. However, there is as yet no further evidence to support this function. Embryos homozygous for loss-of-function mutations in the proneural genes exhibit a highly hypoplasic CNS and severe defects in the PNS (Fig. 2; Jimenez and Campos-Ortega, 1979, 1987, 1990; White, 1980; DamblyChaudiere and Ghysen, 1987; Ghysen and Dambly-Chaudiere, 1988; Caudy et al., 1988a). The origin of the neural hypoplasic defects of such mutants is complex. Embryos lacking the AS-C or vnd initiate neurogenesis with less than the normal complement of neuroblasts: 2&25% of all neuroblasts are missing (Jimenez and Campos-Ortega, 1990); mutants lacking both the AS-C and vnd lack up to 50% of all neuroblasts. In addition, during later stages, large numbers of cells degenerate within the neural primordium of all these mutants (JimCnez and Campos-Ortega, 1979, 1990; White, 1980; Brand and Campos-Ortega, 1988). However, the CNS of these mutants still contains a significant number of neurones, indicating that still other genes are necessary for neuroblast commitment. Since the populations of neuroblasts affected by mutations in the AS-C and vnd do not seem to overlap significantly, these genes, and probably other as yet unidentified ones, may each control the development of particular sets of neuroblasts. Brand and Campos-Ortega (1988) found that the epidermal sheath of neurogenic mutants is larger when they also lack the AS-C; this is suggestive evidence to support the notion that at least some of the cells which fail to develop as neuroblasts in the mutants develop as epidermoblasts instead. The participation of the gene daughterless in neuroblast commitment and segregation is not yet clear. The complement of neuroblasts in daughterlessmutants is initially normal; however, most of these cells die early in embryonic development and mutants show obvious hypoplasic defects in the CNS (Jimenez and Campos-Ortega, 1990; Brand and Campos-Ortega, submitted). However, daughterless- mutants completely lack the peripheral nervous system, due to death of the progenitor cells of the sensory organs (Caudy et al., 1988a; Brand and Campos-Ortega, 1988; Vassin et al., 1993). 7 The neurogenic genes are functionally interrelated

There is abundant evidence from various kinds of genetic analyses to support the assumption that the neurogenic loci are functionally interrelated (Campos-Ortega et a f . , 1984; Vassin et al., 1985; de la Concha et al., 1988; Shepard et af., 1989; Brand and Campos-Ortega, 1990); Godt et al., 1991; Xu et al)., 1990. Insofar as their participation in the segregation of neuroblasts from epidermoblasts is concerned, neurogenic loci were found to be functionally linked in a chain of epistatic relationships, in which the E(sPL)-C was found to be the last link; big brain was found to act independently of the others (Vassin el al., 1985; de la Concha et al., 1988).

NEUROGENESIS IN DROSOPHILA

85

Hence, the function of each of these genes appears to be dependent on that of another member of the group and, consequently, the function of the entire chain is perturbed if any of the links is missing. Recent data on the distribution of transcripts from neurogenic genes in neurogenic mutants indicate that some of the interrelationships between the neurogenic genes are likely to reflect transcriptional regulation (Godt et al., 1991).

8 Physical interactions of Notch and Delta The participation of Notch and Delta in cell communication processes is strongly suggested by the primary structure of the encoded proteins. The Notch protein, as deduced from sequences of genomic and cDNA clones (Wharton et al., 1985; Kidd et al., 1986), shows features typical of a transmembrane protein. Indeed, antibodies raised against different parts of the Notch protein confirm its location in the cell membrane (Kidd et al., 1989; Johansen et al., 1989). A striking feature of the extracellular domain is a tandem array of 36 cysteine-rich motifs each of about 40 amino acids with similarity to the epidermal growth factor (EGF) and other proteins of vertebrates and invertebrates. The Notch protein is, like the Notch RNA (Hartley et al., 1987), ubiquitously distributed during early developmental stages, although there are quantitative differences between cells suggesting a role for Notch in epidermogenesis (Johansen et al., 1989; IOdd et al., 1989; Fehon et a f . , 1991; see also Hoppe and Greenspan, 1986, 1990). Delta is also a transmembrane protein, with a hydrophobic signal sequence and a membrane-spanning domain (Vassin et al., 1987; Kopczynski et al., 1988). The extracellular domain contains nine EGF-like repeats, arranged in tandem like those in the Notch protein. The Delta protein includes 833 residues (Haenlin et al., 1990). Transcription of Delta is spatially regulated; however, during neuroblast segregation Delta RNA is apparently present in all VNE cells (Vassin et al., 1987; Kopczynski and Muskavitch, 1989; Haenlin et al., 1990). Recent data on the distribution of the Delta protein (Kooh et al., 1993) corroborate the RNA findings and show no evident asymmetric distribution within the neuroectoderm. The EGF-like repeats and other motifs in the extracellular domains of Delta and Notch may well represent essential parts of the cell communication pathway by mediating direct protein-protein interactions. Genetic mosaic analyses indicate that the products of Notch and Delta cannot diffuse (Dietrich and Campos-Ortega, 1984; Hoppe and Greenspan, 1986; de Celis et al., 1991; Heitzler and Simpson, 1991). Various pieces of experimental evidence support the view that the EGF-like repeats of Notch mediate protein-protein interactions. First, mutations at the Notch locus, i.e. split and several Abruptex alleles, each differ from the Notch wild-type protein by single amino acid exchanges in specific, distantly separated EGF-like

86

JOSE A. CAMPOS-ORTEGA

repeats, i.e. 14th with respect to split, and 24th, 25th, 27th and 29th with respect to Abruptex alleles (Hartley et af., 1987; Kelley et al., 1987). Two alleles of Delta have been recovered which suppress the phenotype of the split mutation but which do not modify the phenotype of Abruptex alleles (Brand and Campos-Ortega, 1990); that is to say, this is an allele-specific suppression of the split phenotype in both cases. These alleles have been sequenced and both are associated with single amino acid exchanges in the 4th and 9th EGF-like repeat respectively (Lieber et af., 1992). This is altogether strong evidence to support the contention of direct interactions between Notch and Delta. Two different pieces of biochemical evidence provide support for a direct association of Notch and Delta and show that, indeed, both proteins may mediate cell adhesion. Transient expression of both proteins under the control of a metallothionein promoter (Fehon et al., 1990; Rebay et al., 1991), or expression of clones comprising the coding regions of Notch or Delta under the control of a heat shock promoter after stable transfection (Lieber et af., 1992), confers upon Schneider cells adhesive properties which they normally lack. The adhesivity is cell concentration dependent and heterophilic, in that cells expressing Notch adhere only to Delta- but not to other Notch-expressing cells (Lieber et al., 1992). Schneider cells expressing either split or an Abruptex variant (AxR) are still capable of adhering to cells expressing one of the Delta proteins that suppress split (Deltasup’) with the same kinetics as cells expressing wild-type forms of Notch and Delta. However, the mutant proteins cannot compete with the corresponding wild-type proteins in cell aggregation assays, suggesting that their adhesivity is somehow impaired (Lieber et al., 1992). Notch-mediated cell adhesion has been found to depend on EGF-like repeats 11th and 12th (Rebay et af., 1991). Yet, the region of interaction between both proteins appears to be much larger. Suppression of split by Deltasup’ is not compatible with a mechanism based on cell adhesion in vivo, since no reversion of the diminished binding activity of split is observed after aggregation with cells expressing the suppressor protein. This result suggests that intracellular signalling by spfit relevant to compound eye development is mediated by specific parts of the protein, different from those required to mediate adhesion (Lieber et af., 1992). The experimental evidence thus indicates that Notch and Delta are directly associated at the membrane of the neuroectodermal cells. Data on the distribution of the proteins (Fehon et af., 1991; Kooh et af., 1993) confirm their co-localization in assumedly interacting cells. One possible mode of direct interaction between Notch and Delta at the cell membrane involves a receptor-ligand relationship. The available evidence points to Notch as a receptor and Delta as the source of regulatory signals required for epidermogenesis (Dietrich and Campos-Ortega, 1984; Hoppe and

NEUROGENESIS IN DROSOPHILA

a7

Greenspan, 1986, 1990; Technau and Campos-Ortega, 1987; de Celis et af., 1991; Heitzler and Simpson, 1991, 1993).

8 The epidermal decision is controlled by the E(sPL)-C The genetic analysis shows that the functions encoded by the E(sPL)-C are epistatic over those of the other neurogenic genes. This is compatible with the assumption that this locus is responsible for the epidermalizing function i.e. inhibition of neurogenesis, ascribed to the whole group of the neurogenic genes (Vassin et al., 1985; Knust et a f . , 1987b; de la Concha et al., 1988). Rather than being the function of a single gene, the epidermal decision of the VNE cells is controlled by contributions from seven partially redundant, ancestrally related genes, i.e. H L H - m a , H L H - m y , H L H - m P , H L H - d , H L H - m 5 , H L H - m 7 , and E(spf), which constitute the E(sPL)-C (Knust et al., 1987b, 1992; Klambt et af., 1989; Schrons et a f . , 1992). Causal relationships between a small deletion present in the m8 transcription unit of the E(spf)D mutant and the enhancement of split, were established by P element-mediated transformation using mutant and wild-type genes (Klambt etal., 1989; Tietze et a f . , 1992). Genetic analyses have demonstrated functional redundancy among the members of the E(sPL)-C (Ziemer et a f . , 1988; Schrons et a f . , 1992). This redundancy is apparently due to two factors. First, during SI and SII neuroblast segregation, transcripts from all genes of the complex but HLH-m3 exhibit virtually identical spatial distributions. The distribution of these RNAs is overlapping and, immediately before the segregation of SI and SII neuroblasts takes place, it matches remarkably well the regions of the VNE from which first the SI neuroblasts and later the SII neuroblasts Will segregate (Knust et af., 1987b, 1992). Second, sequencing of genomic and cDNA clones has shown a high degree of sequence similarity in the proteins encoded by H L H - m a , H L H - m y , H L H - m P , H L H - i d , HLH-m.5, HLH-m7 and E(sp1) (Klambt et al., 1989; Knust et al., 1992). All seven are members of the bHLH family of transcriptional regulators. The transcription unit m9-ml0, which is located in the immediate neighbourhood of the E(sPL)-C (Knust et af., 1987, 1992), corresponds to groucho (Preiss et a f . , 1988; Schrons et a f . , 1992), a mutation associated with various bristle defects (Knust ef al., 1987b). The putative groucho protein is nuclear (Delidakis et af., 1991) and contains a repeated motif which is similar to another motif in the p-subunit of transducin (Hartley et al., 1988), as well as in the product of a cell cycle gene (CDC4; Yochem and Byers, 1987) and in a component of the spliceosome (PRP4; Balroques and Abelson, 1989), both of yeast. Tietze et al. (1992) have shown that the grouch; phenotype itself is due to a lesion in m9-mlO and that an insertion

88

JOSE A. CAMPOS-ORTEGA

of a middle repetitive DNA element present in transcription unit m9/m10of the E(spl)D strain causes a hypomorphic, visible mutation in the groucho gene. groucho is in fact a neurogenic gene with a prominent maternal component of gene expression, which happens to be located in the neighbourhood of the E(sPL)-C (Schrons et al., 1992). 10 The neural decision is controlled by the proneural genes

The AS-C includes four genes, achaete, scute, lethal of scute and asense, the names being derived from the phenotypic effects of their mutations on bristle development and viability (Muller, 1935; Garcia-Bellido, 1979; Ghysen and Dambly-Chaudiere, 1988). The spatial distribution of achaete, scute and lethal of scute transcripts (Cabrera et al., 1987; Romani et al., 1987; Brand and Campos-Ortega, 1988; Cabrera, 1990; Martin-Bermudo et al., 1991; Skeath and Carroll, 1992; Ruiz Gomez and Ghysen, 1993) is in principle similar for all three and shows a high degree of correlation with the processes of neuroblast segregation and development of sensory organs, i.e. those processes in which, from the analysis of mutants, we know the functions of the genes to be required. During early neurogenesis, the three transcripts are expressed in partially overlapping clusters of cells within the VNE. One or two cells from among these clusters will delaminate as neuroblasts. After the segregation has occurred, RNA of these genes remains detectable in the neuroblasts for some time. This pattern of transcription is suggestive of a role for the AS-C genes in neuroblast commitment. Cabrera (1990) and Martin-Bermudo et al. (1991) have raised antibodies against lethal of scute, and Skeath and Carroll (1991, 1992) against achaete and scute, and they find a similar correlation between accumulation of protein and neuroblast segregation. Not all the neuroblasts express similar amounts of lethal of scute, suggesting that not all the neuroblasts require this gene product for their commitment (MartinBermudo et al., 1991; see also Jimenez and Campos-Ortega, 1990). Causal relationships between expression of proneural genes and commitment of VNE cells as neuroblasts are indicated by genetic studies (JimCnez and Campos-Ortega, 1987, 1990). As mentioned above, AS-C- mutants have 20-25% fewer neuroblasts than the wild-type; similar findings have been made with vnd mutants. Apparently, the AS-C and vnd control the commitment of non-overlapping populations of neuroblasts, since AS-Cand vnd- double mutants show additive effects and lack roughly 50% of all neuroblasts. In addition, increasing the number of copies of the AS-C genes and of vnd by using duplications of the region leads to development of additional neuroectodermal cells as neuroblasts at the expense of epidermoblasts. Therefore, the complementary phenotypes caused by decreasing and increasing the number of copies and, presumably, the amount

NEUROGENESIS IN DROSOPHILA

89

of product of the proneural genes, strongly support the hypothesis that these products are responsible for the development of neuroectodermal cells as neuroblasts. A similar role for these genes has been proposed with respect to the commitment of the progenitor cells of sensory organs. It has been shown that, contrary to the situation with the neuroblasts, the function of achaete, scute and asense is prominent in sensory organ development of the larva and imago, whereas lethal of scute is apparently dispensable for this process (Garcia-Bellido and Santamaria, 1978; Garcia-Bellido, 1979; Dambly-Chaudiere and Ghysen, 1987; Ghysen and Dambly-Chaudiere, 1988, 1989; Ghysen and O’Kane, 1989; Bodmer et al., 1989; Jan and Jan, 1990). Sequence analyses have shown that the proteins encoded by achaete, scute, lethal of scute and asense are similar to each other and contain the bHLH domain, involved in transcription activation (Villares and Cabrera, 1987; Alonso and Cabrera, 1988; Gonzalez et af., 1989). The daughterless locus has been found to encode a bHLH protein as well (Caudy et af., 1988b; Cronmiller ef al., 1989). Murre et al. (1989a,b) showed that proteins with the HLH motif are able to form homodimers and heterodimers and bind to specific DNA sequences. These findings strongly support the contention that all these proteins function in vivo as transcriptional regulators. A high degree of specificity and complexity in the regulatory functions of the corresponding genes may thus be achieved through the combination of different proteins to form heterodimcrs. Genetic interactions between some of the AS-C genes and daughterless during sensory organ development have been recently described (Dambly-Chaudiere et al., 1988), suggesting that these genes are involved in closely related functions (Jan and Jan, 1990). Thus, the finding (Murre et al., 1989b) that lethal of scute and daughterless may form DNA binding heterodimers corroborates the observations of and the inferences drawn from the genetic analysis. 11 Interactions between neurogenic and proneural genes

Evidence has been obtained that the AS-C genes and daughterless are functionally interconnected with the neurogenic genes within the same genetic network (Brand and Campos-Ortega, 1988, 1990; Cabrera, 1990; Skeath and Carroll, 1992; Hinz et a f . , 1994; Haenlin et al., 1994; Kramatschek and Campos-Ortega, 1994; Oellers et al., 1994. First, the severity of the neurogenic phenotype of double mutants for neurogenic genes and deletions of the AS-C or daughterfess is considerably reduced, compared with the phenotype of the same neurogenic mutation alone, i.e. without the concomitant presence of a AS-C- or daughterless mutation. AS-C- mutations were found to be epistatic over the neurogenic mutations, suggesting that the function of the AS-C genes follows that of the neurogenic genes in the functional chain (Brand and Campos-Ortega, 1988).

90

JOSE A. CAMPOS-ORTEGA

The distribution of products from t h e AS-C genes in neurogenic mutants suggests that at least some of the interactions between neurogenic and AS-C genes are likely to involve regulation of the transcription of these genes. Changes in the pattern of transcription of the genes lethal of scute and achuete have been observed in embryos carrying any of several neurogenic mutations (Brand and Campos-Ortega, 1988; Skeath and Carroll, 1992; Ruiz-Gomez and Ghysen. 1993). In these embryos, the early expression of lethal of scute and achaete is indistinguishable from the wild-type. However, already at the beginning of neuroblast segregation, RNA is found in larger clusters than in the wild-type probably due to the fact that neuroblasts do not segregate from the clusters. In the wild-type, a restriction of lethal of scute and uchaete transcription occurs from an initial group of several neuroectodermal cells to one, or a few neuroblasts, as they segregate from the epidermoblasts; in neurogenic mutants, this restriction fails to occur. Cabrera ( 1990) has recently presented similar evidence with respect to lethal o f s c u t e , and Skeath and Carroll (1992) with respect to achaete in that the number of neuroectodermal cells expressing these proteins is higher in neurogenic mutants than in the wild-type. All these observations indicate that the neurogenic genes define the normal expression of AS-C genes in that they suppress the transcription (or the accumulation) of RNA from at least lethal of scute and achaete in some of the VNE cells. Molecular evidence, both in v i m and in vivu, for interactions between proneural and neurogenic genes, derives from the analysis of the promoters of Delta (Haenlin et al., 1994) and of HLH-m5 and E(sp1) (Kramatschek and Campos-Ortega, 1994; Oellers et ul., 1994, as well as from the ectopic expression of lethal of scute (Hinz et al.. 1994). These studies have revealed that proneural proteins activate transcription of Delta, HLH-m5 and E(spl) by means of binding to multiple sites distributed throughout their promoters. The activation of E(sPL.)-C genes by proneural proteins within the VNE is a n early event. initiating the transcription of these genes in this particular region. Thc assumed function of this activation process will be discussed below. Activation o f transcription of Delta. which encodes the presumptive epidermalizing signal molecule, strongly suggests that it serves the function of increasing the efficacy of lateral inhibition in the prospective neuroblast during its singling out from the proneural cluster.

12 Conclusions

The emergence of neuroblasts and epidermoblasts from an undifferentiated anlage is the result of a rather complex set of operations. At the beginning of this review, I proposed a hypothesis in which these operations were arranged in three consecutive steps: acquisition of competence to assume the neural fate; selection of single cells from groups of a few cells to take on

NEUROGENESIS IN DROSOPHILA

91

the fate of particular neuroblasts; suppression of the primary neural fate by lateral inhibition of the remaining cells of the groups to enable their epidermal development. The competence to take on neural fate is likely to depend on the presence of products of the proneural genes in the neuroectodermal cells. Neural competence in the VNE exhibits a clear positional context, in that groups of four to five cells are enabled to give rise to particular types of neuroblasts depending on their position (Taghert et af., 1984; Doe and Goodman. 1985b; Doe et af., 1988; Skeath et af., 1992). These groups are comparable to the so-called proneural cell clusters in sensory organ development (Ghysen and Dambly-Chaudiere, 1989; Romani etaf., 1989; Simpson, 1990; Jan and Jan, 1990). However, the postulated proneural clusters of the VNE are probably in immediate contiguity to each other, such that eventually the entire VNE receives neural competence and can become neuralized in neurogenic mutants, whereas the clusters in the epidermis, from which progenitor cells of sensory organs will arise, may be widely spaced. In fact, both the phenotype of neurogenic mutations and the results of transplanting cells from the anterolateral ectoderm suggest that the neural competence extends outside of the VNE. Neural competence and specification of neuroblasts and epidermoblasts are probably two different operations which may require the functions of different genes (Skeath et af., 1992; see Rodriguez et af., 1990, for sensory organ specification). In the second step, one cell from each proneural cluster has to be singled out to become a neuroblast. The important element in this process seems to be the presence of a critical amount of proneural proteins, but it is unknown how this comes to be since all cells of the group are assumed to have initially comparable amounts. Oscillations in the content of proneural gene products, which would cause the predominance of proneural gene products in one cell and its entry into the neural pathway. may occur in a cellautonomous manner. Cubas and Modolell (1992) and Van Doren et a/. (1992) have presented evidence that within the epidermal proneural clusters, progenitor cells of the sensory organs develop in regions of minimal concentration of extramacrochaetae, a negative regulator of proneural genes (Ellis et al., 1990; Garreil and Modolell, 1990). Higher concentration of extramacrochaetae in a given cell would down-regulate proneural proteins and impede neural development in that cell. Unfortunately, there is no evidence for a participation of extramacrochaetae in neuroblast commitment. Here, the predominance of proneural gene products in one cell could be the consequence of information conveyed to the prospective neuroblast by way of its relations with its neighbours. With respect to bristle development, Heitzler and Simpson’s (1991, 1993) results suggest that the amount of Notch product present in a cell has an influence on its fate, less Notch than normal leading to neural, more Notch to epidermal development. Struhl er af. (1993) have recently obtained evidence that truncated forms of the Notch protein, which contain only the intracellular domain, are

92

JOSE A. CAMPOS-ORTEGA

constitutively active. These observations are compatible with a function of Notch as a ligand-activated receptor in neuroblast segregation. Finally, as far as the third step is concerned, the data show convincingly that cell interactions are somehow involved in development of the epidermoblasts, although they do not yet allow inferences about when the salient cellular interactions take place. Evidence suggests that the decision of one cell from each proneural group to become a neuroblast is followed by cell communication to impede neurogenesis in the remaining cells of the groups and permit in this way their epidermal commitment (Doe and Goodman, 198%; Hartenstein and Campos-Ortega, 1984). However, as with the neuroblasts, epidermogenesis may also be initiated autonomously and reinforced by lateral inhibition derived from the neuroblasts. In fact, observations of Luer and Technau (1902) on the in vitro behaviour of single cells from defined ectodermal regions indicate that epidermogenesis can in part be the result of a cell autonomous process. Cell communication could be required, for example, to suppress expression of proneural genes in the cells normally developing as epidermoblasts, for in neurogenic mutants the domains of expression of lethal of scute and achaete do not become restricted to single cells, as in the wild-type, but extend to all the cells of a proneural cluster. The results of transplanting E(sPL)-C- cells into the VNE of the wild-type were interpreted to mean that the E(sPL)-C encodes functions related to the reception of epidermalizing signals. The proteins encoded by HLH-mG, HLH-my. HLH-mP, HLH-m5, HLH-m7, and E(spl), with the basic DNA binding and the HLH dimerization motif (Klambt et al., 1980; Knust et al., 1992), are indeed compatible with a function at the side of the receptor. Therefore, the failure of E(sPL)-C- cells to develop epidermal fates following transplantation into the wild-type VNE was most probably due to the lack of proteins allowing epidermal development of the transplanted cells. Granted a regulation of the specific genetic activities of the epidermoblasts by the products of the E(sPL)-C, this regulation may occur by direct binding to DNA; indeed, the HLH-m5 and E(sp1) proteins are capable of binding to specific DNA sequences in vitro (Tietze et al., 1992; Oellers ef al., 1994). Or, alternatively, regulation may occur by heterodimer formation with other bHLH proteins, impeding their activity in this way. I t seems improbable that proteins encoded by the E(sPL)-C activate directly transcription of the ‘realizators’ (Garcia-Bellido, 1975) of the epidermal pathway, i.e. the genes whose products eventually make the epidermal cells, since the complete deletion of the E(sri>)-C still permits epidermal development in the dorsal-most embryonic regions (Knust et al, l087a). I would like to propose that the E(sPL)-Cregulates epidermogenesis indirectly via the proneural genes (Fig. 3). My assumption here is that epidermogenesis is a constitutive developmental pathway that is followed by any cell without intervening genetic regulation. In the cells destined to

NEUROGENESIS IN DROSOPHILA A

93

realisators, epilermis (-1

E(SFL) -c-

proneural

f- genes (-)

realisators, CNS

x

B

realisators epidermis

genes realisators, CNS 4 Neuroblasts

C

FIG. 3 A formal genetic model of the interrelationships between proneural and E(sPL)-C genes during neuro-epidermal lineage segregation. (A) Proneural and E(sPL)-C genes are thought to functionally inactivate each other; proneural genes activate the neural and repress the epidermal ‘realizator’ genes. If the balance between both groups of regulatory genes is perturbed, proneural (B) or E(sPL)-C genes (C) will predominate in the corresponding cell, leading to its development as a neuroblast or as an epidermoblast, respectively. ’

become neuroblasts, proneural gene products would suppress the epidermal ‘realizator’ genes; in the other cells, the proneural gene products would be suppressed by the products of the E(sPL)-C. I already mentioned how the neurogenic phenotype of double mutants AS-C- and, for example, E(sPL)C-, is reduced compared with that of the E(sPL)-C- mutants alone (Brand and Campos-Ortega, 1988). This observation suggests that the AS-C, as well as the other proneural genes, suppress the epidermal ‘realizator’ genes: the deletion of the AS-C eliminates this suppressive effect and permits the development of epidermis. I further assume that the proneural gene products activate in addition the neural ‘realizator’ genes and thus permit neural development of the neuroblasts. This assumption is also supported by the reduction of the neurogenic phenotype of AS-C- and E(sPL)-C- double mutant embryos. Activation and suppression here are meant operationally, without presupposing any particular molecular mechanism. Cell determination in the VNE can thus be envisaged as the result of a delicate balance

94

JOSE A. CAMPOS-ORTEGA

between the functional activity of the proteins encoded by the proneural genes and the E(sPL)-C. Both groups of proteins are assumed to be capable of reciprocally regulating each other, as suggested by the opposite effects, neural hyperplasia or hypoplasia, of duplications and deletions of the proneural and of the E(sPI.)-C genes, discussed in the previous sections. In fact, the complexity of both gene complexes and the structure and function of the bHLH proteins they encode, point to a great potential for interrelationships, of which very little is known. This model is based exclusively on genetics and there is as yet little molecular evidence to support it. The proposed regulation can be achieved at the transcriptional, at the post-transcriptional, or at both levels. There is both in vifro and in vivo evidence that the proneural proteins activate the transcription of the E(sPL)-Cgenes E(spl) and HLH-m5 (Kramatschek and Campos-Ortega, 1994; Hinz et id., 1994). This activation appears to be required to allow products of the E(sPL)-C genes to be present within the same VNE cells in which products of the proneural genes occur, in order for them to be able to compete with each other. In v i m studies using cell transfection assays (Oellers et ul.. 1994) have shown that addition of HLH-m5 and E(spl) to a specific DNA sequence in the E(sp1) promoter (N-box) reduces the transcriptional activation mediated by heterodimers between lethal of scute and daughterless. Therefore, the proteins encoded by the E(sPL)-Cmay act indirectly as negative regulators of transcription of proneural genes. The results of these interrelationships are likely to be either the predominance of the proneural proteins or their functional suppression in any particular VNE cell, leading to its commitment to one of the fates. In any cell in which proneural products predominate, one first step on the neural pathway would consist of the synthesis of the molecular machinery that establishes communication with ncighhouring cclls (Fig. 4). This initial communication WOUHperm2 reception of neurakzing sf&-zX? and fhus reinforce and stabilize the neural decision. A second step for the same cell on its way to neurogenesis would be the synthesis of the molecular apparatus for transmission of epidermalizing signals. for instance formation of functional Delta protein, in order to permit lateral inhibition to occur. Delta transcription can indeed be activated in the VNE by proneural proteins (Kunisch et ul., submitted). Initially, this activation takes place in all cells of the proneural cluster, but is likely to be dependent on the concentration of proneural proteins. Hence, this would lead to an increase of the amount of inhibitory signal in the cell that has initiated neurogenesis, to the consecutive inhibition of the surrounding cells of the cluster and the reinforcement of the neural decision in the neuroblast. Conversely, the behaviour of those cells of the competence group which follow the epidermal pathway is likely to reflect the functional suppression of the proneural proteins, and consequently the initiation of the constitutive

95

NEUROGENESIS IN DROSOPHILA Pioneural i3eiies act i ' J d t P t r a n s c r i p t i o n of E(SPL)-(. genes

111i i ~ u i r ? i ~ t (ni ~ ~ i ~ r

f ' r > m p p t i t i o nb e t w e e n E I S P L ) - C and p r o n e u r a l g p n e s

1(+' e p i d e r m a l is i n g

FIG. 4 Sequence of events assumed to occur during lineage segregation. The squares represent cells within a proneural cluster. Refer to text for further details.

developmental pathway, with the synthesis of receptor for epidermalizing signals in order to reinforce and stabilize the epidermal decision. Acknowledgements

I would like to thank the members of my laboratory for support, and Elisabeth Knust, Paul Hardy and Thomas Klein for constructive criticisms on the manuscript. The research reported here was supported by several grants of the Deutsche Forschungsgemeinschaft (DFG, SFB 243) and the Fonds der Chemischen Industrie.

96

JOSE A. CAMPOS-ORTEGA

References Alonso. M. C . and Cabrera, C.V. (1988). The uchaete-scute gene complex of Drosophilu melunogaster comprises four homologous genes. E M B O J. 7, 25852501. Balroques, J . and Abelson, J. N . (1989). PRP4: a protein of the yeast U4iU6 small nuclear ribonucleoprotein particle. Mol. Cell. B i d . 195, 233-245. Bate. C . M. (1976). Embryogenesis of an insect nervous system. I . A map of the thoracic and abdominal neuroblasts in Locitsta rnigrutoriu. J . Embryol. Exp. Morphol. 35. 107-123. Bate. C . M. (1982). Development of the neuroepithelium. Neurosci. Res. Bull. 20, 803-806. Becker. T. and Technau. G . M. (1990). Single cell transplantation reveals interspecific cell communication in Drosophila chimeras. Development 109. 821-832. Bccr. J . , Technau, G . M.and Campos-Ortega, J. A . (1987). Lineage analysis of transplanted individual cells in embryos of Drosophila melanogaster. IV. Commitment and proliferative capabilities of mesodermal cells. Roux’s Arch. Dev. Biol. 196, 222-230. Bellen. H. J.. O’Kane. C. J . , Wilson. C., Grossniklaus, U., Pearson, R . K. and Gehring. W. J. ( 1989). P-element-mediated enhancer detection: a versatile method to study development in Drosophilu. Genes Dev. 3 . 1288-1300. Bier. E . , VBssin, H . , Sepherd, S., Lee, K.. McCall, K., Barbel, S., Ackerman, L., Carretto. Uemura, T.. Grell, E., Jan. L. Y . and Jan. J. N. (1989). Searching for pattern and mutation in the Drrxnphilu genome with a P-lacZ vector. Genes Dev. 3 , 1273-1287. Bodmer. R., Carretto. R. and Jan, Y . N. (1989). Neurogenesis of the peripheral nervous system in Drosophila embryos: DNA replication patterns and cell lincages. Neuron 3 . 21-32. Boulianne. G . L.. de la Concha, A . . Campos-Ortega. J. A , . Jan. L. Y . and Jan Y . N . (1991). The Drosophila neurogenic gene neuralized encodes a novel protein which is expressed in precursors of larval and adult neurones. E M B O J. 10, 2975-2983. Bourouis. M., Heitzler. P.. El Messal, M. and Simpson, P. (1989). Mutant Drosophilu embryos in which all cells adopt a neural fate. Nature 341, 442444. Bourouis. M.. Moore. P., Ruel, L.. Grau. Y . . Heitzler, P. and Simpson. P. (1990). An early embryonic product of the gene shaggy encodes a serinehhreonine protein kinase related to the CDC281cdc2f subfamily. E M B O J. 9, 2877-2884. Brand. M. and Campos-Ortega, J. A . (1988) Two groups of interrelated genes regulate early neurogenesis in Drosophilu melanogaster. Roux’s Arch. Dev. Biol. 197. 457470. Brand. M. and Campos-Ortega, J. A . (1990). Second site modifiers of the split mutation of Notch define genes involved in neurogenesis in Drosophila melanogusrer. Roux’s Arch. Dev. Biol. 198, 275-285. Brand. M. and Campos-Ortega ( 1094). The role of daughterless during development of the embryonic and adult nervous system of Drosophila melanogaster. submitted. Cabrera. C. V. (1990). Lateral inhibition and cell fate during neurogenesis in Drosophilu: the interactions between scute. Notch and Delta. Development 109, 733-742. Cabrera. C . V.. Martinez-Arias. A . and Bate, M. (1087). The expression of three members of the uchaere-scute gene complex correlates with neuroblast segregation in Drosophilu. Cell 50. 425433.

NEUROGENESIS IN DROSOPHlLA

97

Campos-Ortega. J. A . and Haenlin, M. (1992). Regulatory signals and signal molecules in early neurogenesis of Drosophila melanogaster. Roux’s Arch. Dev. Biol. in press. Campos-Ortega, J. A . and Hartenstein, V. (1985). ”The Embryonic Development of Drosophila melanogaster.” Springer Verlag, Berlin. Campos-Ortega, J. A.. Lehmann. R . , Jimenez, F. and Dietrich, U. (1984). A genetic analysis of early neurogenesis in Drosophila. In “Organizing Principles of Neural Development” (Ed. S. C. Sharma). pp. 129-144. Plenum Press, New York. Caudy. M., Grell, E. H., Dambly-Chaudiere, C.. Ghysen, A , . Jan, L. Y. and Jan, Y. N. (1988a). The maternal sex determination gene daughterless has zygotic activity necessary for the formation of peripheral neurons in Drosophila. Genes Dev. 2. 843-852. Caudy, M.. Vissin. H., Brand. M., Tuma, R., Jan, L. Y., Jan, Y. N. (198%). daughterless, a gene essential for both neurogenesis and sex determination in Drosophila. has sequence similarities to m y and the achaete-scute complex. Cell 55, 1061-1067. Cronmiller, C.. Schedl, P. and Cline. T . W. (1989). Molecular characterization of daughterless, a Drosophila sex determination gene with multiple roles in development. Genes Dev. 2. 1666-1676. Cubas, P. and Modolell, J. (1992) The extramacrochaetae gene provides information for sensory organ patterning. E M B O J. 11. 3385-3393. Cui, X. and Doe, C. Q. (1992). ming is expressed in neuroblasts sublineages and regulates gene expression in the Drosophila central nervous system. Development 116, 943-952. Dambly-Chaudiere. C. and Ghysen. A . (1987). Independent subpatterns of sense organs require independent genes of the achaete-scute complex in Drosophila larvae. Genes Dev. 1, 297-306. Dambly-Chaudiere, C., Ghysen, A , , Jan, L. Y. and Jan, Y. N. (1988). The determination of sense organs in Drosophila: interactions of scute with daughterless. Roux’s Arch. Dev. Biol. 197, 419-423. de Celis, J., Mari-Beffa, M. and Garcia-Bellido, A. (1991) Cell autonomous role of Notch, an epidermal growth factor homolog. in sensory organ differentiation in Drosophila. Proc. Natl. Acad. Sci. USA 88, 632436. de la Concha. A . . Dietrich, U . . Weigel. D. and Campos-Ortega. J. A . (1988). Functional interactions of neurogenic genes of Drosophila melanogaster. Genetics 118, 499-508. Delidakis, C.. Preiss, A , , Hartley. D . and Artavanis-Tsakonas. S. (1991). Two genetically distinct functions reside within the Enhancer of split locus of Drosophila melanogaster. Genetics 129, 803-823. Dietrich. U . and Campos-Ortega, J. A . (1984). The expression of neurogenic loci in imaginal epidermal cells of Drosophila melanogaster. J. Neurogenef. 1, 315-332. Doe, C. Q. (1992). Molecular markers for identified neuroblasts and ganglion mother cells in the Drosophila central nervous system. Development 116. 855-863. Doe, C. Q. and Goodman, C. S. (1985a). Early events in insect neurogenesis. I . Development and segmental differences in the pattern of neuronal precursor cells. Dev. B i d . 111, 193-205. Doe. C. Q. and Goodman. C. S. (1985b). Early events in insect neurogenesis. 11. The role of cell interactions and cell lineages in the determination of neuronal precursor cells. Dev. Biol. 111, 20&219. Doe, C. Q.. Hiromi, Y., Gehring, W. J. and Goodman C . S. (1988). Expression and

98

JOSE A. CAMPOS-ORTEGA

function o f the segmentation gene firshi-turuzit during Drosophila neurogenesis. Sciencc 239, 17(k175. Ellis, H. M.. Spann. D. R. and Posakony. J. W. (1990) extramacrochaetae, a negative regulator of sensory organ development in Drosophilu, detines a new class o f helix-loop-helix proteins. Cell 61. 27-38. Fehon, R. G . . Kooh. P. J.. Rebay, I.. Regan. C. L., Xu. T., Muskavitch. M. A. T. and Artavanis-Tsakonas. S. (1990). Molecular interactions between the protein products of the neurogenic loci Notch and Delta. two EGF-homologous genes in Drosoplzilri. Cdl 62. 523-533. Fehon, R. G . . Johansen, K . . Rebay, I. and Artavanis-Tsakonas, S. (1991). Complex cellular and subcellular regulation of Notch expression during embryonic and imaginal development of Drosophila: implications for Notch function J . Cell Biol. 113, 657-669. Garcia-Bcllido, A . (1975). Genetic control of wing disc development in Drosophila. In “Cell Patterning” Ciba Foundation Symposium 29 (Ed. by S. Brenner), pp. 161-178. Elsevier. Excerpta Medica. North Holland. Assoc. Amsterdam, Oxford, New York. Garcia-Bellido, A . (1979). Genetic analysis o f the acliric>te-.mite system of Drosophilu melanoguster. Genetics 91. 491-520. Garcia-Bellido, A . and Santamaria. P. ( 1978). Developmental analysis of the nchaetc.-scure system of Drosophilu rnelanogaster. Genetics 88. 469-486. Garrell. J. and Modolell, J . ( 1990). The Drosophilu extranzacrocliuetae locus, an antagonist of proneural genes that, like these genes, encodes a helix-loop-helix protein. Crll61, 39-48. Ghysen. A . and Damhly-Chaudiere. C . (1988). From DNA t o form: the achaetescute complex. Genes Dev. 2. 495-501. Ghysen, A. and Dambly-Chaudiere. C . ( 1989). Genesis of the Drosophila peripheral nervous system. Trend in Genetics 5. 251-255. Ghysen. A . and Dambly-Chaudiere, C. (1990). Early events in the development of Drosophila peripheral nervous system. J . Plivsiol. (Paris) 84, 1 1-20. Ghysen, A . and O’Kane, C. (1989). Detection of enhancer-like elements in the genome of Drosophilu. Develoiiment 105, 35-52. Ghysen. A . . Dambly-Chaudiere. C.. Jan. L. Y. and Jan. Y.-N (1993). Cell interactions and gene interactions in peripheral neurogenesis. Genes Dev. 7, 723-733. Godt, D.. Schrons, H . . Guth, S. and Camps-Ortega, J. A . (1991). The distribution o f transcripts of neurogenic genes in neurogenic mutants of Drosophila melanogaster. J . Neurogeriet. 7. 241-252. Gonzilez. F., Romani, S., Cubas. P.. Modolell. J. and Campuzano. S. (1989). Molecular analysis of userise. a member of the uchuete-scure complex of Drosophila rnelanogristrr. and its novel role in optic lobe development. E M B O J. 8. 3553-3562. Goriely. A . , Dumont. N.. Dambly-Chaudiere. C. and Ghysen. A . (1991). The determination of sense organs in Drosophilri. effect of the neurogenic mutations in the embryo. Drvelopmenr 113, 1395-1404. Haenlin. M., Kramatschek. B. and Campos-Ortega, J. A. (1990). The pattern of transcription of the neurogenic gene Delta of Drosophilu nielanogaster. Developrnent 110, 905-914. Haenlin, M., Kunisch, M., Kramatschek. B. and Campos-Ortega. J. A . (1994). Genomic regions regulating expression of the Drosophila neurogenic gene Delta. Meclz. Dev. (in press). Hartenstein. V. and Campos-Ortega, J . A . (1984). Early neurogenesis in wildtype Drosophila melunogrister. Roux’s Arch. Dev. B i d . 193, 308-325.

NEUROGENESIS IN DROSOPHILA

99

Hartenstein, V. and Campos-Ortega. J. A . (1986). The peripheral nervous system o f mutants of early neurogenesis in Drosophila melanopster. ROILX'S Arch. Dev. Biol. 195. 210-221. Hartenstein. V., Rudloff, E. and Campos-Ortega, J. A . (1987). The pattern o f proliferation of the neuroblasts in the wild-type embryo of Drosophila melunogasrer. Roux's Arch. Dev. Biol. 196. 473485. Hartley, D. A.. Xu. T. and Artavanis-Tsakonas, S. (1987). The embryonic expression of the Notch locus o f Drosophifa melanogaster and the implications o f point mutations in the extracellular EGF-like domain of the predicted protein. EMBO J . 6. 3407-3417. Hartley. D. A . , Preiss. A . and Artavanis-Tsakonas. S. (1988). A deduced gene product from the Drosophilu neurogenic locus Enhancer of split shows homology to mammalian G-protein p subunit. Cell 55. 785-795. Heitzler, P. and Simpson, P. (1991). The choice of cell fate in the epidermis of Drosophila. Cell 64. 1083-1092. Heitzler, P. and Simpson. P. (1993). Altered epidermal growth factor-like sequences provide evidence for a role o f Notch as a receptor in cell fate decisions. Development 117. 1 1 13-1 123. Hinz, U.. Giebel. B. and Campos-Ortega. J. A . (1994). The basic-helix-loop-helix domain of the Drosophilu lethal of scute protein is sufficient for proneural function and activates neurogenic genes. Cell 76, 77-87. Hoppe, P. E . and Greenspan, R. J. (1986). Local function of the Notch gene for embryonic ectodermal choice in Drosophila. Cell 46, 773-783. Hoppe, P. E. and Grecnspan. R . J . (1990). The Notch locus of Drosophila is required in epidermal cells for epidermal development. Development 109. 875-885. Jan, Y. N. and Jan. L. Y. (1990). Genes required for specifying cell fates in Drosophila embryonic sensory nervous system. Trends Neurosci. 13, 493498. Jimenez. F. and Campos-Ortega. J . A . (1979). A region of the Drosophilu genome necessary for CNS development. Nature 282. 310k312. Jimenez, F. and Campos-Ortega. J . A . (1982) Maternal effects of zygotic mutants affecting early neurogenesis in Drosophila. Roux's Arch. Dev. Biol. 191. 191-201. Jimenez. F. and Campos-Ortega, J . A . (1987). Genes in subdivision 1B of the Drosophila melanoguster X-chromosome and their influence on neural development. 1. Neurogenet. 4, 179-200. Jimenez. F. and Campos-Ortcga. J . A . (1990). Defective neuroblast commitment in mutants of the achaete-scute complex and adjacent genes of Drosophila melanogaster. Neuron 5. 81-89. Johansen. K. M . . Fehon. R. G. and Artavanis-Tsakonas, S. (1989) The Notch gene product is a glycoprotein expressed o n the cell surface of both epidermal and neuronal precursor cells during Drosophila development. J . Cell. Biol. 109. 2427-2440, Jiirgens, G.. Lehmann. R.. Schardin. M. and Niisslein-Volhard, C. (1986). Segmental organisation of the head in the embryo of Drosophila melanogaster. A blastoderm fate map of the cuticle structures of the larval head. Roux's Arch. Dev. Biol. 195. 359-377. Kelley. M. R., Kidd, S., Deutsch. W. A. and Young, M . W. (1987). Mutations altering the structure of epidermal growth factor-like coding sequences at the Drosophilu Notch locus. Cel! 51, 539-548. Kidd. S.. Kelley, M. R. and Young. M. W . (1986). Sequence of the Notch locus of Drosophilu mdonugaster: Relationship o f the encoded protein to mammalian clotting and growth factors. Mol. Cell. Biol. 6. 3094-3108. Kidd. S.. Baylies, M. K., Gasic. G . P. and Young, M. W. (1989) Structure and

100

JOSE A. CAMPOS-ORTEGA

distribution of the Notch protein in developing Drosophila. Genes Dev. 3, 11 13-1 129. Klambt, C., Knust, E., Tietze, K. and Campos-Ortega, J. A . (1989). Closely related transcripts encoded by the neurogenic gene complex Enhancer of split of Drosophila melanogaster. EMBO J . 8, 203-210. Klambt, C., Jacobs, J. R. and Goodman, C. S. (1991). The midline of the Drosophila central nervous system. A model for the genetic analysis of cell fate, cell migration, and growth cone guidance. Cell 64, 801-815. Knust, E . , Bremer, K. A., Vassin, H . , Ziemer, A , . TepaB, U. and Campos-Ortega, J. A . (1987a). The Enhancer of split locus and neurogenesis in Drosophila melanogaster. Dev. Biol. 122, 262-273. Knust, E.. Tietze, K. and Campos-Ortega, J. A . (1987b). Molecular analysis of the neurogenic locus Enhancer of split of Drosophila melanogaster. EMBO J . 6, 4113-4123. Knust, E . , Schrons, H., Grawe, F. and Campos-Ortega, J. A . (1992) Seven genes of the Enhancer of split complex of Drosophila melanogaster encode helix-loop-helix proteins. Genetics 132, 505-518. Kooh, P. J . , Fehon, R. and Muskavitch, M. A . T . (1993). Implications of dynamic patterns of Delta and Notch expression for cellular interactions during a Drosophila development. Development 117, 493-507. Kopczynski, C. C. and Muskavitch, M. A. T. (1989) Complex spatio-temporal accumulation of alternative transcripts from the neurogenic gene Delta during Drosophila embryogenesis. Development 107, 623436. Kopczynski. C. C., Alton, A . K., Fechtel, K.. Kooh, P. J. and Muskavitch, M. A. T. (1988). Delta, a Drosophila neurogenic gene, is transcriptionally complex and encodes a protein related to blood coagulation factors and epidermal growth factor of vertebrates. Genes Dev. 2, 1723-1735. Kramatschek, B. and Campos-Ortega, J . A . (1994) Neuroectodermal transcription of the Drosophila neurogenic genes E(sp1) and HLH-m5 is regulated by proneural and neurogenic trans-activators. Development 120, 815-826. Kunisch, M., Haenlin, M. and Campos-Ortega, J. A . (1994) Lateral inhibition mediated by the Drosophila neurogenic gene Delta is activated by proteins encoded by proneural genes. (submitted) Lehmann, R., Dietrich, U., Jimenez, F. and Campos-Ortega, J. A . (1981). Mutations of early neurogenesis in Drosophila. Roux’s Arch. Dev. Biol. 190, 226229. Lehmann, R., Jimenez. F.. Dietrich. U . and Campos-Ortega, J . A . (1983). On the phenotype and development of mutants of early neurogenesis in Drosophila melanogaster. Roux’s Arch. Dev. Biol. 192, 62-74. Lieber, T., Alcamo, E.. Hassel, B . , Krane, J. F., Campos-Ortega. J . A . and Young, M. W. (1992). Single amino acid substitutions in EGF-like elements of the Notch and Delta proteins modify Drosophila development and depress cell adhesion in vitro. Neuron 9, 847-859. Luer, K. and Technau, G. M. (1992). Primary culture of single ectodermal precursors of Drosophila reveals a dorsoventral prepattern of intrinsic neurogenic and epidermogenic capabilities at the early gastrula stage. Development 116, 377-385. Martin-Bermudo, M. D . , Martinez, C . and Jimenez. F. (1991). Distribution and function of the lethal of scute gene product during early neurogenesis in Drosophila. Development 113, 445454. Muller. H. J. (1935). The origination of chromatin deficiencies as minute deletions subject to insertion elsewhere. Genetica. 17, 237-252.

NEUROGENESIS IN DROSOPHILA

101

Murre, C., Schonleber McCaw, P. and Baltimore, D. (1989a). The amphipathic helix-loop-helix: a new DNA-binding and dimerization motif in immunoglobulin enhancer binding, daughterless, MyoD and myc proteins. Cell 56, 777-783. Murre, C., Sehonleber McCaw, P . , Vaessin, H., Caudy, M., Jan, L. Y., Jan, Y. N . , Cabrera, C. V., Buskin, J. N., Hauschka. S. D., Lassar, A. B., Weintraub, H. and Baltimore, D. (1989b). Interactions between heterologous helix-loop-helix proteins generate complexes that bind specifically to a common DNA sequence. Cell 58, 537-544. Oellers, N . , Dehio, M. and Knust, E. (1994). Basic-helix-loop-helix proteins of the Enhancer of split complex of Drosophila negatively interfere with transcriptional activation by proneural proteins. Mol. Gen. Genet. (in press). O’Kane, C. and Gehring, W. J. (1987) Detection in situ of genomic regulatory elements in Drosophila. Proc. Natl. Acad. Sci. USA. 84, 9123-9127. Poulson, D. F. (1937). Chromosomal deficiencies and embryonic development of Drosophila melunogaster. Proc. Natl. Acad. Sci. USA 23. 133-137. Poulson, D. F. (1950). Histogenesis, organogenesis and differentiation in the embryo of Drosophilu melunogaster. In “Biology of Drosophila” (Ed. M. Demerec), pp. 268-274. Preiss, A., Hartley, D. A. and Artavanis-Tsakonas, S. (1988). The molecular genetics of Enhancer of split, a gene required for embyonic neural development in Drosophilu. EMBO J . 7,3917-3928. Rao, Y . , Jan, L. Y. and Jan, Y. N. (1990). Similarity of the product of the Drosophilu neurogenic gene big brain to transmembrane channel proteins. Nature 345, 163-167. Rebay, I . , Fleming, R. J . , Fehon, R. G., Cherbas, L., Cherbas, P. and ArtavanisTsakonas, S. (1991). Specific repeats of Notch mediate interactions with Delta and Serrate: implications for Notch as a multifunctional receptor. Cell 67. 687-699. Rodriguez, I., Hernindez, R., Modolell, J . and Ruiz-Gomez, M. (1990). Competence to develop sensory organs is temporally and spatially regulated in Drosophila epidermal primordia. EMBO J . 9, 3583-3592. Romani, S.. Campuzano. S. and Modolell, J . (1987). The uchaete-scute complex is expressed in neurogenic regions of Drosophila embryos. EMBO J . 6, 2085-2092. Romani, S . , Campuzano. S . , Macagno, E. R. and Modolell, J. (1989). Expression o f uchaete and scute genes in Drosophila imaginal discs and their function in sensory organ development. Genes Dev. 3, 997-1007. Ruiz-Gomez, M. and Ghysen, A. (1993). The expression and role of a proneural gene, achaete, in the development of the larval nervous system of Drosophila. E M B O J . 12, 1121-1 130. Schrons, H.. Knust. E. and Campos-Ortega. J. A. (1992) The Enhancer of split complex and adjacent genes in the 96F region of Drosophilu melunogaster are required for segregation of neural and epidermal progenitor cells. Genetics 132, 481-503. Shepard, S. B., Broverman. S. A. and Muskavitch, M. A. T. (1989). A tripartite interaction among alleles of Notch, Delta and Enhancer of split during imaginal development of Drosophila melanogaster. Genetics 122, 429-438. Simpson, P. (1990). Lateral inhibition and the development of the sensory bristles of the adult peripheral nervous system of Drosophila. Development 109, 509-519. Simpson, P. and Carteret, C. (1989). A study of shaggy reveals spatial domains of expression of achaete-scute alleles on the thorax of Drosophila. Development 106. 57-66. Skeath. J . B. and Carroll, S. B. (1991). Regulation of achuete-scute gene expression

JOSE A. CAMPOS-ORTEGA

102

and sensory organ pattern formation in the Drosophilri wing. Get1t.s Dev. 5 , 9 x 4 9 9 5.

Skeath, J . B. and Carroll. S. B. (1992). Regulation of proneural gene expression and cell fate during neuroblast segregation in the Drosophila embryo. Developmenr 114. 930-946. Skeath. J . B., Panganiban. G.. Selcgue, J . and Carroll. S. B. (1992). Gene regulation in two dimensions: the proneural rickric.re and sc'iite genes are controlled by combinations (if axis-patterning genes through a common intergenic control region. Genes D ~ L6, , . 260&2619. Smoller. D.. Friedel. C.. Schmid. A., Bettler. D.. Lam. L. and Yedvobnich, B. ( 1990). The Drosopfzila neurogenic gene musterrnind encodes a nuclear protein unusually rich in amino acid homopolymers. Genes & Dew/. 4, 1688-1 700. Struhl. G . , Fitzgerald. K. and Greenwald, I. (1993). Intrinsic activity of the Lin-iz and Notch intracellular domains in vivo. Cell 74. 331-345. Stiitten. I. and Camps-Ortega, J . A. (1901).Cell commitment and cell interactions in the ectoderm of Drosophilri melrrnoguster. Development. Suppl. 2, 3946. 7'aghert. P. H.. Doe. C. 0 . and Goodman. C. S. (1984).Cell determination and reguhtion during development of neuroblasts and neurones in grasshopper embryos. Nritiire 307, 163-165. Technau. G. M . (1986). Lineage analysis of transplanted individual cells in embryos of Drosophilci tnelririogasrcr. I. The method. Roiix's Arch. Dev. Biol. 19.5, 389-398. Technau, G . M . (1087).A single cell approach to problems of cell lineage and commitment during embryogenesis of Drosoplzila melunogaster. Development 100, 1-12. Technau. G . M . and Campos-Ortega. J . A. (1985). Fate mapping in wildtype Drosophilrr melutiogrisfer, I1 Injections of horseradish peroxidase in cells of the early gastrula stage. Roiix's Arch. D P V .Riol. 194, 196212. Technau. G. M . and Campos-Ortega, J . A. (IYX6a).Lineage analysis of transplanted individual cells in embryos of Drosophilu melanogusrer. 11. Commitment and proliferative capabilities of neural and epidermal cell progenitors. Roux's Arch. Dev. Biol. 195. 445454. Technau. G. M. and Campos-Ortega. J . A. (1986b).Lineage analysis of transplanted individual cclls in embryos of Drosophilci melrinogcister. 111. Commitment and proliferative capabilities of pole cells and midgut progenitors. Roux's Arch. Dev.

Biol. 19.5. 389498. Technau, G . M . and Camps-Ortega. J . A . (1987). Cell autonomy of expression of neurogenic genes of Drosophilu rtieluriogrister. Proc. Nritl. A u I ~ .Sci. U S A , 84, 45003504. Technau. G . M.. Becker, T. and Canipos-Ortega, J . A . (1988). Reversible commitment of neural and epidermal progenitor cells during embryogenesis of Drosophilu melritioguster. Roiix's Arch. Dev. B i d . , 197. 413-418. Thomas. J . B . . Bastiani, M . J . , Bate, M. and Goodman. C . (1984). From grasshopper to Drosophilu: ii common plan for neuronal development. Nature 310, 203-207. Tietze, K., Oellers, N . and Knust. E. (1992) Enhancer of .split'). ;I dominant mutation of Drosophila and its use in the study o f functional domains of a helix-loop-helix protein. Proc. Narl. A c d . Sci. U S A 89. 61524156. Van Doren, M . , Powell, P. A., Pasternak, D . , Singson, A. and Posakony, J . W . ( 1992). Spatial regulation of proneural gene activity: auto and cross-activation of rdiciere is antagonized by extr(It7iucrocharrie. Genes Dev. 6, 2592-2605. Viissin, H.. Vielmetter. J . and Campos-Ortcgz. J . A . (1985). Genetic interactions in early ne u roge ncs is o f Drosophilti tnelrinogri.srer. J . Neii rogcn . 2. 29 1-308.

NEUROGENESIS IN DROSOPHlLA

103

Vassin, H., Bremer. K. A.. Knust. E. and Campos-Ortega. J. A. (1087). The neurogenic locus Deltrr of Ilrosophilu nirlritiogmster is expressed in neurogenic territories and encodes ii putative transmembrane protein with EGF-like repeats. EMBO J. 6. 343 1-3440. Vassin. H.. Brand. M.. Jan. L. and Jan. Y. N . (1094). daughterless is essential for neuronal precursor differentiation but not for initiation of neuronal precursor formation in Drosophila embryos. Development (in press). Villares, R. and Cabrera. C. V . (1087). The achuere-scirfe gene complex of Drosophila melrrriogastcr: conservcd domains in a subset of genes required for neurogenesis and their homology to tnyc. Cdl 50. 415424. Wharton. K. A . , Johansen. K . M . , Xu. T. and Artavanis-Tsakonas. S. (lY85). Nucleotide sequence from the neurogenic locus Notch implies a gene product that shares homology with proteins containing EGF-like repeats. Cell 43, 567-581. White, K. ( 1980). Defective neural development i n Drosophila melanogusfer embryos deficient for the tip of the X-chromosome. Dev. B i d . 80. 322-344. White. K . , DeCelles. N . L. and Enlow. T. C. (1083). Genetic and developmental analysis of the locus t w r l in Drosophilri niel(rnogaster. Genetics 104. 433488. Xu, T., Rebay. 1.. Fleming. R. J.. Scottgale, T. N . . Artavanis-Tsakonas. S. (1090). The Notch locus and the genetic circuitry involved in early Drosophila neurogenesis. Genes Dev. 4. 4641175. Yochem, J. and Byers. B . (1087). Structural comparison of the yeast cell dimension cycle gene CDC4 and a related pseudogene J. Mol. B i d . 195. 233-234. Ziemer. A.. Tietze, K . , Knust, E . und Campos-Ortega, J. A. (1988). Genetic analysis o f Enhancer qf’ .split. a locus involved in neurogenesis in Drosophila melanogaster. Genetics 119, 63-74.

This Page Intentionally Left Blank

Molecular Biology of the Honeybee Robin F.A. Moritz lnstitut fur Biologie, Technische Universitat Berlin, Franklinstr 28/29, 10587 Berlin, Germany

1 Introduction 105 2 Genes and sequences

107 2.1 Nuclear genes 107 3 Mitochondrial genome 114 3.1 Mitochondrial genes 116 3.2 Non-coding sequences and length variation 117 4 Gene activity in embryonic development 124 5 Population variability 125 5.1 Nuclear DNA markers 125 5.2 Mitochondrial DNA markers 129 6 Molecular evolution and biogeography 130 6.1 Molecular phylogeny of bees 130 6.2 Genetic variability among honeybee species 131 6.3 Molecular variability within species 133 7 Outlook 140 Acknowledgements 141 References 144

1 Introduction

The biological rules that govern the honeybee colony have fascinated scientists since Aristotle. The bustling activities of the thousands of workers are seemingly at random and yet in the end a unified and apparently coordinated system is achieved. The honeybee riddle of chaos on the one hand and yet pattern on the other was to occupy generations of scientists after Aristotle. It took a long time to grab some pieces of the great puzzle and understand the basics of communication in honeybee colonies. It was the great pioneers in the field of bee research who unravelled the mystery of identifying the dance language of the worker bee (von Frisch, 1965) and of queen control through pheromones (Butler, 1973). In the second half of this ADVANCES IN INSECI'PHYSIOLOGY VOL. 25 ISBN lL1?4l?1??5-7

106

R. F. A. MORITZ

century bee research was dominated by studies focusing o n behavioural ecology and communication under the influence of Karl von Frisch's and Martin Lindauer's epochal work on honeybees. However. the focus on behavioural ecology was not always so strong. and in fact honeybees also entered the arena of modern research as a genetic model system. Mendel conducted experiments with bees but unfortunately he failed to achieve controlled matings in netted cages and could not produce any F2 offspring which he required to confirm his theories based on the pea experiments (Johannson, 1980). Real breakthroughs were achieved by Dzierzon (1845) who found that drones are parthenogenetically produced. Petrunkewitsch (1903) and Nachtsheim (1912, 1914) used honeybees in their studies to elucidate the mode of fertilization and chromosome duplication. Honeybees were among the prime genetic model systems used until Castle and Morgan introduced Drosoplzila to the field of genetics in the early 20th century. Yet after the wide acceptance of fruit flies as the genetic test organism, honeybee genetics became a very specialized field and even the introduction of artificial insemination by Laidlaw (1944) and Mackensen (1947) could not stop the decline of honeybee genetics. The honeybee system was difficult to handle, had a slow generation cycle and an annoyingly large number of chromosomes ( n = 16) which made linkage studies tedious. Controlled matings could only be achieved through artificial insemination, and the generation cycle was too long to generate a rapid series of high-quality publications, an important basis for successful research and the recruiting of funds. Maintenance costs were high and other organisms offered better, swifter, and less expensive ways to solve pending questions. One of the last strongholds of honeybee genetics was in behavioural genetics, but after the work of Rothenbuhler (1964) on the genetics of hygienic behaviour of worker honeybees, a depressing silence also broke out in international scientific journals in this field. Clearly. the molecular revolution of genetics took place leaving the honeybees aside. The molecular genetics of Drosophila melanogaster boomed while bee geneticists felt more attracted to applied breeding research. Only recently, after a surprisingly long abstinence of several decades, are geneticists slowly re-entering the scene in honeybee research, trying to catch up with modern techniques and addressing new and (more important) unique questions that can only be solved by using honeybees as a test system. Although honeybees as an animal model for basic genetics have their pitfalls, they are far from being useless in basic genetical research. In fact, they have several unique features that make them more attractive than any other genetical test system available in higher organisms. Particularly important are the following traits for genetical research: 1. Male haploidy allows the study of gene expression in haplotypes. This is important for both selection experiments and gene mapping studies.

MOLECULAR BIOLOGY OF THE HONEYBEE

107

2. The rich behavioural repertoire and the social organization makes it a primc system in behavioural genetics. 3. The slow embryonic development offers plenty of opportunities to study gene regulation and expression during early development. These are just a few characteristics which make the honeybee profitable for basic genetical research. Since honeybees are of significant economical and ecological importance, there is obviously also a compelling need to understand their population and breeding genetics. Here also molecular techniques are aiding us in understanding processes in natural and artificial selection and the undcrlying genetic mechanisms. Today the stage is set for rapid progress in molecular honeybee genetics, since we can capitalize on the very detailed knowledgc obtained from Drosophila. Cloned DNA probes are available which enable us to rapidly map the honeybee genome; sequenced genes allow us a better understanding of gene regulation and evolution. In addition, mitochondria1 DNA is used in population genetic studies revealing the dynamics of natural selection in feral honeybee populations. Thus, molecular genetics of honeybees can both incrcase our understanding of basic genetic mechanisms and improve our knowledge of honeybee specific genetic problems.

2 Genes and sequences

2.1

NUCLEAR GENES

2.1.1 Elongation factor I The search for genes in honeybees has been limited and only a few genes have been isolated and sequenced so far. The most productive way of isolating genes has been through hybridization of honeybee DNA to known Drosophila genes. Walldorf and Hovemann (1990) studied a DNA coding for one of the three proteins ( a , p , y ) of the cytoplasmic elongation factor 1 (EF-1). EF-la catalyses the transport of the aminoacyl-tRNA to the 80s ribosome. In Drosophila melanogaster two independent genes (F1 and F2) code for EF-la (Hovemann et al., 1988). Walldorf and Hovemann (1990) found the EF-la gene of honeybees to be closely related to the corresponding coding region in Drosophila melanogaster. They cloned two fragments of 1.0kb and 1.1 kb. respectively, which revealed as much as 77% sequence homology to the EF-la F1 Drosophila reading frame. The same degree of homology was found for another elongation factor gene (EF-lcu, F2). Through this high sequence conservation Walldorf and Hovemann (1990) were able to locate the translation start and stop sequence in the gene (nucleotides 365 and 2121, respectively) and they also found two introns

108

R. F. A. MORlT o p e n reading f r a m e

intron

FIG. 1 Putative structure of the E F - l a gene in Apis melliferu. The base pair numbering is adopted from Walldorf and Hovernann (1990).

similar to the Drosophila EF-la F2gene (Fig. 1). Walldorf and Hovemann (1990) suggested that the Apis EF-la gene evolved from a common ancestral =-type gene rather than the F1 gene which is free of introns. The isolated EF-la sequence of Apis codes for 461 amino acids with a calculated mass of 50.5 kDa. It is unclear whether Apis has more than one EF-la gene. In a southern hybridization of the Drosophila probe on an EcoRI digest of Apis mellifera DNA, Walldorf and Hovemann (1990) found a weak second signal at about 9.4 kb and they could not exclude the possibility of a second EF-la gene in honeybees. The high homologies between Drosophila and Apis are not necessarily surprising since the elongation factor is known to be encoded in an extremely conservative gene region with only little variance among different taxa. Only two more variable amino acid sites appear, at positions 186 and 315, respectively, by adding the Apis sequence to the already known sequence variability among eight different species (Walldorf and Hovemann, 1990). 2.1.2 Segmentation genes Fleig et a/. (1988) studied the homologies of several homeobox genes of Drosophila rnelanogaster and Apis rnellifera. They constructed a genomic library of Apis mellifera and identified a homeobox DNA sequence in a cloned 500 bp Cla-Sull fragment of Apis rnellifera showing 82% homology to the Drosophila gene Dfd, and coding for the identical amino acid sequence. They termed the honeybee gene H42 and found additional homologies that extended beyond the 5' and 3' ends of the box. They also argued that the position of the intron may be at an identical site in both honeybees and fruit flies. In further work, Walldorf et a/. (1989) compared other homeobox genes of Apis rnellifera and Drosophila rnelanogusfer in more detail. Drosophila DNA probes containing Antennupedia ( A n f p ) ,fushi tarazu ( f t z ) , sex combs

109

MOLECULAR BIOLOGY OF THE HONEYBEE

TABLE 1 Sequenced homeobox genes in Apis mellifera

Drosophila probe abdominal-A Antennapedia Deformed fushi tarazu engrailed" invected"

muscle segment homeobox Sex combs reduced W-13

Honeybee clone

% amino acid similarity in the homeobox

E30 E6O E30 E60 H17

96.7 98.3 100 not detected 96.7 91.7 81.7 83.3 96.7 (with 5.4 kb intron)

H55 H40

98.3 54.2

H15 H90 H42 -

~~

T h e sequence similarity between eti and inv is smaller between than within species. It is therefore not possible to assign E30 and E60 to either gene.

reduced (Scr) and Deformed ( D f d ) hybridized to EcoRl digests of total genomic DNA of Apis rneffifera (Table 1). Although frz did yield hybridization signals in genomic honeybee DNA (Fleig et al., 1988), the authors were unable to detect a homologous conserved sequence in the honeybee. Screening their library with f t z , the only clones that were reisolated were those that had already been isolated with the Antp probe. Walldorf et al. (1989) concluded that, given the honeybee has an ftz homologous gene, it must have considerably diverged from the Drosophila gene. Alternatively, they suggest that the gene is completely lacking and other loci perform the ftz function. The genes of the engrailed class (en = engrailed and inv = invected) appeared to be conservative. Two clones, ,560 and E30, were highly homologous, yet it was not possible to assign the two Drosophila genes to the two honeybee clones because sequence divergence between the two genes within the species was less than between the species. The homology of the homeobox in clone H40 showed only 55.2% similarity to the Drosophila W-13 gene. No homeobox has been found in Drosophila with strong homologies to H40. Sommer et al. (1992) studied segmentation genes with the Cys2-His2 zinc-finger DNA binding motif. They found evolutionary conserved patterns between Drosophila and Apis for hunchback, Kriippel, and snail (Fig. 2). Particularly, Kriippel showed high similarities between both species with a 94.6% amino acid homology, and the authors suggest that it might be duplicated in the honeybee. It appeared that those amino acids which are believed to have a direct contact with the bases in the DNA are completely conserved. The escargot sequence from Drosophila had a higher homology

110

R. F. A. MORlTi

hunchback

Kiiippel

snail

FIG. 2 Sequence alignment of finger fragments obtained from the genes of Drosophilu (top) and Apis (bottom). Identical amino acids are denoted with a dash, a deletion with a dot. The stars indicate those amino acids that are believed to be in direct contact with the bases in the DNA. For the snail gene (c) also the escargot sequence is plotted to reveal the sequence similarity to the Apis sequence.

to thc Apis fragment (91.4%) than had the paralog snail sequence (75.6%) which was initially used to isolate the Apis DNA (Sommer et ul., 1992).

2.1.3 Genes coding for honeybee venom compounds There are also few studies focusing on honeybee specific genes. These mainly include DNA regions coding for enzymes and peptides of bee venom (Table 2 ) which have been analysed because their composition is known from various detailed biochemical studies (Habermann and Jentsch, 1967; Shipolini er al., 1971; Bachmeyer et al., 1972; Habermann, 1972; Gauldie et ul., 1976, 1978; Suchanek et al., 1978; Kreil et al., 1980). Melittin is the main lytic peptide of the honeybee venom and is found in both queens and workers. Like most peptides it is initially synthesized as a larger precursor which is then proteolytically cleaved to the final product (Kreil, 1990). Typical of the melittin precursor is the pro-region which is composed of an amino acid sequence in which every other position is either alanine or

MOLECULAR BIOLOGY OF THE HONEYBEE

111

TABLE 2 Composition ot t h e worker honeybee venom in 54 dry weight Enzymes Phospholipase A? Hyaluronidase Acid Phosphatase

l(klS% 2 (7,

Peptides Mclittin A pa mi n MCD-peptidc Secapin T e rt i a pi n

45-605? 2-3 %' 2 %' 1%

r :._ . 50 n

c

0

-0.2

n

x

0.2 a

0.8

,

0 35 0 wovelength

- nm

,

4 50 550 wovelength , n m

\, 650

MAKOTO MlZUNAMl

168

and moth, however, showed that separate spectral mechanisms were needed to explain their UV-green ocellar response (Goldsmith and Ruck, 19.58; Chappell and DeVoe, 197.5; Eaton, 1976; Pappas and Eaton, 1977; Yamazaki and Yamashita, 1991). Intracellular studies on dragonfly photoreceptors established that UV-green sensitivity resulted from two pigments, but it was not possible to state whether these were located in one cell or in different cells coupled electrically (Chappell and DeVoe, 1975). In the compound eye, there are typically three types of colour receptors, UV, blue and green receptors, which are characterized by relatively narrow spectral tuning curves (Fig. 4C). The colour signals are processed by higher order neurones which receive excitatory inputs from one type of colour receptor and inhibitory inputs from another type, thus encoding colour contrast (Kien and Menzel, 1977). It appears that different strategies have been adopted in the ocelli and compound eyes to process signals from multiple pigment systems. In the compound eye, signals from different colour receptors are subtracted to enhance colour contrast, but in the ocelli, signals from different pigment systems are summed to enhance sensitivity.

3.6

POLARIZATION SENSITIVITY

Polarization sensitivity has been rarely examined for insect ocelli, but there is a report which shows that ocellar photoreceptors of the desert ant Catagfyphis are sensitive to the plane of polarized light. Mote and Wehner (1980) found that all ocellar receptors of the desert ant demonstrated polarization sensitivity ratios of about four. This is comparable with the polarization sensitivity of the compound eyes which ranges from 1.5 to 6. The ocellar photoreceptors are UV sensitive and thus are similar to UV photoreceptors of the compound eyes in terms of spectral and polarization sensitivities. Further behavioural studies show that Cutaglyphis can utilize the ocelli to detect the celestial polarization pattern (Fent and Wehner, 1985) as will be discussed in Section 4.6. The structural and neural basis of polarized light detection in Cutaglyphis remains to be clarified. 3.7

SUMMARY

The high absolute sensitivity and high speed of signal transmission are possibly the most advantageous features of insect ocelli over the compound eyes, which are attained by abandoning spatial resolution. Ocellar Lneurones are characterized by high absolute sensitivity, a broad spectral tuning curve with a peak at UV, and a large receptive field. These characteristics are most suited to detect contrast between the sky and land for stability control in flight, when the ocelli are directed horizontally in a normal flight posture (Wilson, 1978a). The high absolute sensitivity of ocelli is also suited to other visual behaviours in low light conditions.

THE DIVERSITY OF INSECT OCELLAR SYSTEMS

169

4 Behavioural roles of ocelli

4.1

OCELLI AS A STIMULATORY ORGAN

There is a theory which suggests that some sensory organs have, in addition to their specific sensory function, an essential function in maintaining the central excitatory state by sending sustained stimuli to the central nervous system and so raising the non-specific reactivity of the organism (Wolsky, 1930). This interesting but unproved (Bullock and Horridge, 1965) function is termed a general (or unspecific) stimulatory function as opposed to a specific sensory function. Wolsky (1930, 1933) noted the loss of tonus and of speed of walking in several insects when the ocelli were occluded and claimed that the ocelli are examples of stimulatory organs. A number of subsequent behavioural studies have also claimed that the ocelli have a general stimulatory role. None of these studies, however, proved that ocelli do function in this manner, as has been pointed out by Goodman (1970). A question which remains to be solved, concerning the concept of a general stimulatory role, is whether or not insects possess a system analogous to reticular activating systems in vertebrates, as pointed out by Bullock and Horridge (1965). In vertebrates, the reticular formation of the brain stem plays a role in producing arousal and sleep states of the animal and thus producing a circadian activity rhythm. The activity of the reticular formation is regulated by specific sensory pathways, called the reticular activating system, thus sensory inputs to the system result in changes in states of arousal (Shepherd, 1988). If insects possess a system comparable to the reticular activating system, and ocellar and other sensory inputs regulate the state of arousal by affecting the activating system, the once abandoned concept of a general stimulatory function may be revived in a refined fashion. This awaits future examination.

4.2

CONTRIBUTIONTO PHOTOTACTICORIENTATION

Most earlier behavioural studies of ocelli have been carried out with regard to their contribution to phototaxis. It has been concluded that the ocelli alone are rarely responsible for producing phototactic orientation, but rather that the ocelli appear to contribute to positive phototactic orientation mediated by the compound eyes, since ocellar occulution results in a decrease in the accuracy of such orientation (Cornwell, 1955; Cassier, 1965; Jander and Barry, 1968; Schricker, 1965; Meyer, 1978; reviewed by Goodman, 1970). In flies, locusts and crickets, the ocelli complement the compound eyes only under low levels of ambient illumination (Cornwell, 1955; Jander and Barry, 1968). In the fly Drosophilu, the contribution of ocelli and three types of photoreceptors, R1-6, R7 and R8, of compound eyes to positive phototaxis have been studied by utilizing receptor-deficient

MAKOTO MlZUNAMl

170

mutants (Fischbach and Reichert, 1978; Hu and Stark, 1977, 1980; Miller et al., 1981). All three photoreceptor types in the compound eyes evoke positive phototaxis. The sensitivity of the input from R1-6 receptors is higher than that from R7 and R8 receptors. In bright light, R7 dominates positive phototaxis and the presence of ocellar receptors facilitates R7 input (Hu and Stark, 1980). In dim light, R1-6 dominates positive phototaxis. Interestingly, strains with R1-6 do not exhibit positive phototaxis in dim light without ocelli, demonstrating strong facilitatory effects of the ocelli. Earlier studies of phototaxis were too limited to fully understand the behavioural roles of ocelli. First, phototaxis represents only the simplest form of a rich variety of visual responses insects exhibit in a natural environment. Furthermore, some earlier experiments on phototaxis were made in an extreme light condition which insects rarely encounter in a natural environment. Thus, it is not very surprising that these earlier studies did not contribute much to the clarification of the direct behavioural role of ocelli. Most of our knowledge on the behavioural roles of ocelli are based on recent experiments carried out in more natural behavioural situations and under natural light conditions.

4.3

VISUAL COURSE CONTROL I N FLIGHT

Many insects have a tendency to turn their back towards the centre of brightness during flight and walking, to keep their course straight (reviewed by Wehner, 1981). This reaction is referred to as the dorsal light response. A way to attain a dorsal light response is to monitor position and movement of the horizon relative to the body. Hesse (1908) argued that the ocelli are most suited to detect movement of the horizon for stability control in flight, and Wilson (1978a) reformulated Hesse’s argument based on new evidence. Wilson (1978a) argued that locust ocelli have a large receptive field directed horizontally, providing the animal with heavily blurred neural images of the skyline, where unwanted information about structural details is eliminated. Ocellar sensitivity to UV facilitates horizon detection since the contrast between bright sky and dark ground is highest in UV. The high speed of signal detection and transmission in the ocellar system is ideal for rapid course control. Pitch and roll deviation of the flight course are independently detectable by the combination of signals from three ocelli. A roll (turning around the long axis of the body) will cause no change in signal from the median ocellus but will tend to cause a decrease of illumination in one lateral ocellus and an increase in the other lateral ocellus. Detection of pitch could be achieved through measurement of the output of the median ocellus . This hypothesis has received support from behavioural studies. In dragonflies, Stange and Howard (1979) and Stange (1981) observed that

THE DIVERSITY OF INSECT OCELLAR SYSTEMS

171

stimulation of ocelli can evoke a steering response. The stimulation of the median ocellus evoked a head movement around a pitch axis, and the stimulation of the lateral ocellus evoked head movements around the roll axis. In locusts, Goodman (1965) noted the ocellar contribution to stability control in flight and Taylor (1981a,b) further examined the contribution of compound eyes and ocelli to flight steering (Fig. 5 ) . Locusts were tethered inside a simulated horizon display (Fig. 5A,B). Rotation of the horizon elicited following motions of the animal’s head and rudder-like movements of the body during flight. Head and steering body motions were still elicited after either the compound eyes, or the ocelli, were surgically ablated. Head motion after the cautery of the compound eyes (Fig. 5C,E,F) suggested that the ocelli may function synergistically with the compound eyes to (a) minimize the delays of visual responses and (b) augment visual responses at a dim light condition and when n o sharp horizontal border is present. Taylor (1981a) noted that hoverflies (Eristalis renax, Syrphidae) and damselflies (Argia vivida) also followed horizon rotators with head motion after their compound eyes had been ablated. Kastberger (1990) suggested that the honeybee ocelli help to control the flight course. Although it is well established that one of the principal functional roles of ocelli is to detect the horizon for flight stabilization in some insects, ocelli of some other insects seem not to participate in the stability control in flight. In blowfly, Cafliphora, simulated roll of an artificial horizon evokes little ocelli-mediated head movements (Hengstenberg, 1984, 1993) and likewise, changes of the brightness in the visual field of the median and one lateral ocellus elicit only a weak steering response (Schuppe and Hengstenberg, 1993). The ocelli of blowfly are directed dorsally, thus they may not be suited for horizon detection. Tomioka and Yamaguchi (1980) also concluded that ocelli play little role in posture control during flight in the night-flying cricket, Gryllus bimaculafus. It is difficult to suppose that well-developed and specialized ocelli of the cockroach Peripfanera (see Section 2.2) have been evolved to attain better flight steering for this weak flier. Therefore, functional roles of ocelli differ among insects: to contribute to flight steering is a principal function in some insects, but not in others.

4.4

ORIENTATION TOWARD EDGES

Walking houseflies can use ocelli for orientation. Wehrhahn (1984) tested the orientation of walking houseflies with blinded compound eyes, and found that by using only their ocelli, flies orient toward edges and relatively small bright objects situated in the frontal equatorial part of the visual field (Fig. 6). This finding may indicate that the fly ocelli resolve coarse visual images, a possibility which needs to be pursued by future examination.

MAKOTO MlZUNAMl

172

pickups for head position transducer

Head Dosi tion

Head position

Horizon position

E

-

50"

Head position

L 7 J l2.0Js - L n l

-

Head oosition

LLeft-side illuminated - - -

- - - - -

Kiglit

0.5 s

FIG. 5 (A) The artificial horizon apparatus used to evoke rapid rolling motions in a locust. A locust was suspended inside the apparatus with its long axis coaxial to the horizon tubes. The blackened hemisphere was rotated by an axle connected to a chart recorder pen motor. The small spheres are capacitative pickups used to sense the position of the animal's head. A 40 kHz electrical signal was applied to a fine wire within the glass wand shown attached to the locust's head. Comparison of the signals induced in wires leading from the two spheres allowed determination of head position about the roll axis. (B) Apparatus used to study pitch axis motions of the horizon. Similar to (A), except that the animal is mounted transversely in the illuminated tube. (C-F) Responses to visual stimulation of the ocelli. Except for (D), all locusts have compound eyes disconnected by section of the optic lobes leaving the ocelli intact. Unless noted otherwise, downward rotation signifies clockwise rotation with respect to the animal. 1000 lux. (C) Head rolling response to horizon rotation. After three cycles of horizon oscillation, the display was rotated 180" counterclockwise (i.e. dark above, illuminated below). (D) A similar experiment to that shown in (C), except with all eyes intact. After five cycles of motion, the horizon was inverted. (E) Head-rolling response of a non-flying locust to horizon rotation (no wind applied). (F) Head rolling response of a flying locust to alternate illumination of the two sides of the animal's head by a pair of mechanically switched light sources subtending 5". Illuminance of the animal's head was approximately 75 lux. From Taylor (1981a).

THE DIVERSITY OF INSECT OCELLAR SYSTEMS

173

FIG. 6 (a) Frequency distribution of the orientation of freely walking flies whose compound eyes only were blinded (open circles) or whose compound eyes and ocelli were both blinded (filled triangles). The panorama consisted of a black half and a white half generating two vertical edges. The flies with only ocelli show a marked preference to walk towards the edges. Average values from 10 flies. Bars denote standard deviation of the mean. (b) Frequency distribution of the orientation of freely walking flies with blinded compound eyes. The pattern consists of a white square (22.5" side length) and opposite to it a white rectangle on a black background as indicated in the top of the figure. The flies prefer the white square. From Wehrhahn (1984).

4.5

LIGHT INTENSITY PERCEPTION FOR THE CONTROL OF DIURNAL ACTIVITY

Initiation and cessation of diurnal activities of insects often depend on the light intensity levels (Schricker, 1965; Dreisig, 1980). Under natural environmental conditions in which the light-dark cycle consists of dawn and dusk ramps, it would be advantageous for insects to be sensitive to minute changes in illumination. Thus, the ocelli are suited to control the diurnal activity whose absolute sensitivity is at least several times as high as that of compound eyes (Section 3 . 2 ) . Indeed, signals from ocelli are utilized to determine the threshold light intensity for the diurnal activities in bees (Schricker, 1965; Gould, 1975) and moths (Eaton et al., 1983; Sprint and Eaton, 1987). Foraging flight activity in bees is chiefly governed by two factors: the weather and the light intensity (Schricker, 1965). The first and last of the daily flights is dependent upon the intensity level. If the ocelli are occluded, foraging bees behave normally in most respects (but see Renner and Heinzeller, 1979). Occlusion of the ocelli, however, does interfere with the timing of the first and last foraging flights. Bees with one, two or three

B

-

1001

*NOCELL

-e- CONTRROI

--r TIME

TIME

'w O0I

0 TIME

TIME

FIG. 7 Percentage of male moths, Trichopfusiu ni, flying during the sunset period in the phase-advance (PA) experiments. (A) Day before PA. (B) First day of PA. (C) Second day of PA. (D) First day of return to original time of sunset. Anocell, anocellate moths; L. light intensity. The range of light intensity was between 100 and D a y lO>Day 15 > Day 2 virgins > Day 39 > Day 2 >Day 4 = Day 5 (Pratt et al., 1990; Stay et al., 1991a; Woodhead et al., 1989). The sensitivity of CA of various ages similarly shows a broad range of effective dose for 50% inhibition of JH biosynthesis (ED,,)), from -0.1 nM for Day 6 mated females to >1 p M for Day 4 and 5 mated females. This latter change occurred over a 2-day interval (Stay et al., 1991a) and over only 1 day (Pratt et al., 1999) which is clearly rapid and remarkable; if this response is receptor-mediated, a very rapid turnover, upregulation/downregulation or unmasking of receptors should be expected.

288

BARBARA STAY et at.

Developmental changes in responsiveness of larval CA to allatostatins show a pattern distinct from that of adult female CA (Stay et al., 1991a). CA from both penultimate and final instar larvae early in the stadium also show very high sensitivity to allatostatin I (-80% inhibition at 10nM) but this sensitivity declines as the stadium progresses. In the penultimate instar the sensitivity increases toward the end of the stadium, whereas in the final stadium the glands have lost sensitivity to allatostatins at the same time that J H biosynthesis has declined. This loss of sensitivity in final instars suggests that allatostatins are not primary regulators at this time. Larval CA show a similar degree of sensitivity to allatostatins 11, I11 and IV (Stay et al., 1991a). However, it is clear that the importance of allatostatic inhibition of JH biosynthesis differs profoundly at different developmental stages. Our understanding of the changing sensitivity of corpora allata to the allatostatins remains rudimentary and will require full assessment of all of the known allatostatins for any given species.

4.2.2 Activity of C A as a function of sensitivity to allatostatins in D. punctata The developmental changes in sensitivity of CA to allatostatins have been proposed to be inversely related to the rates of JH biosynthesis during the first reproductive cycle, with the most biosynthetically active CA (Day 4-5 adult mated females) showing the lowest degree of inhibition by allatostatins (Pratt et al., 1990, 1991a; Stay et al., 1991a). Although this generalization may be applicable to females during the first and second vitellogenic cycles and in the first half of the penultimate stadium. it clearly is not applicable to either pregnant females or to Iate penultimate or final instars (Stay et a/., 1991a), in which responsiveness decreases as rates of JH biosynthesis decline. At these times, JH biosynthesis remains at low or undetectable leveis for extended periods, suggesting that alternative tonic mechanisms may be operative. This interval, particularly in last instar animals, represents a time of developmental reorganization and transition in the CA, and the glands appear to be incapable of producing even small amounts of JH, and cannot be rescued by treatment with the penultimate precursor farnesoic acid (Yagi et a/., 1991). It is known as well that J H biosynthesis can be inhibited in pregnant females by a humoral route, so long as the brain is present (Rankin and Stay, 1985). Although immunoreactive allatostatin has been found in the haemolymph of last instar animals particularly during the allatostatin-unresponsive period, as well as during pregnancy (Yu et a [ . , 1993) and bioactive allatostatins were demonstrated after HPLC separation in haemolymph of adult females (Woodhead et al., 1993). haemolymph allatostatins would not appear to be directly functional in the inhibition of JH biosynthesis. Rather, they may act to facilitate the inhibition of JH

ALLATOSTATINS: IDENTIFICATION, PRIMARY STRUCTURES, AND FUNCTIONS

289

biosynthesis, ensuring that JH biosynthesis remains at low levels and minimizing carbon flow early in the biosynthetic pathway. 4.2.3 Changes in adult CA of P. americana Interspecies effects of allatostatins have received limited study. For example, Woodhead er al. (1989) tested the effects of D. punctufa allatostatin I (dipstatin 7) on JH production by CA of immature female P. americana and observed a 92% inhibition at 10nM. Further studies by Weaver (1991) determined the developmental changes in the sensitivity of P. umericanu CA to dipstatin 7. During times of increasing JH biosynthesis (including the onset of vitellogenesis in both basal and penultimate oocytes), sensitivity to dipstatin 7 (at 10 pM) increases dramatically and reaches a maximum (inhibition of 97%) at the time corresponding to the onset of vitellogenesis in basal oocytes. The timing is thus quite distinct from that reported in female D. punctata, in which maximal sensitivity occurs at the conclusion of vitellogenesis (Pratt ef a f . , 1990; Stay et a f . , 1991a). Similarly, in D. puncfuta, sensitivity to dipstatin 7 remains stable or declines in early adult life, reaching a minimum at the peak of vitellogenesis, whereas in P. americana sensitivity increases until the onset of vitellogenesis. Weaver (1991) conjectures that this may be a function of the different modes of reproduction of the two cockroach species (viviparity vs oviparity) (see also Tobe, 1980). However until all allatostatins of the respective species have been assessed for activity at physiological doses, it is premature to attribute such differences to different reproductive modes. The homologous allatostatin I tridecapeptides in D. punctata (dipstatin 7) and P. americana (peastatin 7) differ in two residues in the address region (see Sections 2 and 9). The effect of the peptide on the same species is two to three orders of magnitude greater than its effect on the reciprocal species, as shown in Table 4 (Weaver, 1991; Weaver et al., 1994; Stay and Woodhead, 1993). On the basis of changes in activity as a result of substitutions in the address region for dipstatins 2, 5 and 7 (see Section 3.2) tested on D. punctata CA, the difference in response of P. americana CA to their own and D. punctata allatostatin is not surprising. Although potency is reduced, the efficacy of dipstatin 7 in P. americana CA is high, confirming the importance of the message sequence in related species. 4.3

DUALITY OF RESPONSES TO ALLATOSTATINS

The responsiveness of glands of specific ages to allatostatins appears to be a complex physiological phenomenon, as for example in the apparent duality of responses (pM-nM range) to allatostatin V (Pratt el a f . , 1990, 1991b). Such duality manifests itself in flattening or plateauing of dose-response

BARBARA STAY et al.

290

TABLE 4

Interspecific effects of allatostatin I on inhibition of JH synthesis Allatostatin I (M)"

Corpora allata tested

P. americana (peastatin 7)

D. punctata (dipstatin 7)

Diploptera punctatu ( 2 day virgin)

1.2 x 1OPh

2.0 x lo-"'

Periplanetu umericana (3 day virgin)

6.2 x 10-""

6.9 x 10-'"

"Concentration required to give SO% inhibition of JH synthesis "Weaver et al. (1904). 'Stay and Woodhead (1993). W e a v e r (1991).

curves, particularly at low concentrations of peptide, and can be resolved through statistical analysis (see Loftfield and Eigner, 1969). It is likely that such duality represents interaction of a single ligand with multiple allatostatin receptors (see Section 5.2), although we cannot exclude the possibility that modification or degradation of the peptide by peptidases associated with the corpora allata occurred during the course of the assays, resulting in multiple ligands in the incubation medium. The likelihood that interaction with multiple allatostatin receptors provides for the additivity of the dual responses requires further study since it has been shown that treatment of corpora allata with more than one type of allatostatin simultaneously gives similar flattened dose-response curves from CA of specific age (S. S. Tobe, unpublished) but is nonetheless consistent with this hypothesis. Such a possibility was considered feasible by Pratt et al. (1990) and was invoked to explain the flattened dose responses realized with brain extracts. This duality in response appears to be both stage- and allatostatin-specific, since coadministration of allatostatins I and V at 10 nM to CA of Day 2 mated females did not result in additive inhibition (Pratt et al., 1991a). Such so-called 'cross-reactivity' (Pratt et al., 1990) with other allatostatin receptors at high concentrations of mixtures of allatostatins demands the presence of multiple allatostatin receptors; the resolution of this important question will require the isolation/cloning of the allatostatin receptors. 4.4

RESPONSIVENESS T O ANALOGUES OF ALLATOSTATINS

Studies on responsiveness of CA to various analogues and truncated allatostatins have in general been performed only on CA from females of a limited range of age (Days 2, 6 and 10 mated females: Pratt et al., 1991a,b; Day 2 virgin females: Stay et al., 1991b). Because there are large changes in

ALLATOSTATINS: IDENTIFICATION, PRIMARY STRUCTURES, AND FUNCTIONS

291

responsiveness to the allatostatins themselves, it is likely that similar differences exist with respect to the truncated and modified allatostatins. If such changes in responsiveness are the result, in part, of changes in receptor number or subtype, it would appear prudent to examine the effects of the analogues on CA of different developmental and physiological states.

4.5

POSSIBLE FACTORS CONTRIBUTING TO CHANGES IN RESPONSIVENESS

The dramatic developmental changes in sensitivity of CA and responsiveness to allatostatins I and V in vitro (Pratt et al., 1990, 1991a,b) could be attributed to several factors, including: (1) changes in receptor subtype; (2) changes in number and turnover of receptors; upregulation and downregulation of receptors; ( 3 ) changes in ‘cross reactivity’ of receptors to allatostatins; (4) changes in ability of CA to degrade allatostatins in vitro; ( 5 ) changes in the basal lamina surrounding the CA and ability of allatostatins to penetrate it either in vivo or in vitro; (6) changes in the quantity of allatostatins within the CA and their release during incubation in vitro. Although all of these factors no doubt influence responsiveness of CA, it is likely that the primary regulator is modulation of receptor quantity and subtype.

4.6

NEURAL A N D H U M O R A L PATHWAYS FOR ALLATOSTATIN ACTION

The dynamic range of response of CA to allatostatin treatment generally ranges between 0% and 80% inhibition of J H biosynthesis (Pratt et al., 1990; Stay et al., lYYla,b). However, Pratt et al. (1990) show an inhibition of 96% at p M concentrations for CA from Day 6 mated females. Although precise titre values for individual allatostatins in the haemolymph or within the CA are not available, the immunoreactivity data of Yu et al. (1993) suggest that in the haemolymph, the concentrations of allatostatins are not likely to be greater than in the subnanomolar to low nanomolar range. This estimate of allatostatin concentration is conservative in part because the antibody to allatostatin I used to measure it has limited cross-reactivity to other allatostatins. Nevertheless, such concentrations of individual allatostatins provide inhibitions of 50-70% only at the most sensitive stages (i.e. Day 6 mated females) (Pratt et al., 1990, 1991a; Stay et a l . , 1991a) and considerably less inhibition of CA from females of other, less sensitive stages. Interestingly, the concentration of allatostatin I immunoreactive material in haemolymph of Day 6 mated females is at its lowest point at this age (-0.1 nM) and this concentration provides 4 0 % inhibition, even in CA from Day 6 females. Release of allatostatins within and from the CA is likely to create a localized elevated concentration of the peptides. It is known that at least some allatostatins are released from CA during incubation in v i m (Yu et

BARBARA STAY et al.

292

a / . , 1993) and this probably is a manifestation of release of these peptides

within the glands and its ‘leakage’ into the surrounding medium. The allatostatin concentration within the extracellular spaces of a pair of CA can be estimated by assuming that the quantity of allatostatins within the CA is -200 fmol per pair (based on allatostatin I immunoreactivity; Yu et al., 1993); that only about 5 1 0 % of that is released following nerve depolarization with high potassium (Yu et a / . , 1993); that the volume of the extracellular space in a pair of fresh CA is 2 n l (based on measurement of fixed CA (Johnson et al., 1993) and the fact that fixed CA volume, -20111 per pair, is half that of fresh (Szibbo and Tobe, 1981)). Although the actual concentration of the peptides in the extracellular spaces cannot be determined with accuracy, localized concentration might be as high as 10pM, sufficient to cause marked inhibition of JH biosynthesis in sensitive glands. Although some allatostatins are released into and from the CA, it cannot be assumed that all allatostatins from the prohormone are cleaved and released. Processing of the precursor may result in selective packaging of the products or in selective signalling for release from appropriate terminals, be they neurohaemal or within the CA. The acidic domains of the prohormone (Donly et a/., 1993) may be important to the processing of the precursor, to the protection of the peptides from proteolytic degradation and to the targeting of release sites. At present, it can only be stated with certainty that the first five isolated allatostatins (Woodhead et a/., 1989; Pratt et a/. , 1991a) are present within the CA (Stay et al., 1991b; Stay and Woodhead, 1993). 4.7

ALLATOSTATINS AND REGULATION OF J H TITRE

It remains to be determined if allatostatins are the principal regulators of JH biosynthesis and hence of JH titre (Tobe et al., 1985). In D. punctata at physiological haemolymph concentrations in vivo, allatostatins probably cannot inhibit JH biosynthesis to a level sufficient to reduce JH titre significantly at all developmental stages. Even a 50% inhibition in JH biosynthesis, a level unlikely to be achieved at physiological concentrations of allatostatins, would provide only a small reduction in J H titre, based on data relating JH biosynthesis to JH titre (Tobe et al., 1985) (in part because the slope of the line is 0). Indeed, injection of micromolar quantities of allatostatins into haemolymph of adult female D. punctata only reduced rates of JH biosynthesis marginally and oocyte growth slightly but significantly (Woodhead et a/. , 1993); similarly in P. americana, injection of allatostatin reduced J H titre only at some stages (Weaver et a/., 1994) (see Section 7.3). Associated with this question of whether JH biosynthesis in vivo is normally restrained by allatostatins is whether inhibition of JH biosynthesis in vitro truly represents any situation in vivo, although experimental

ALLATOSTATINS: IDENTIFICATION, PRIMARY STRUCTURES, AND FUNCTIONS

293

evidence suggests that it does (see Section 11). Clearly, the degree of inhibition of J H biosynthesis necessary to effect a significant change in JH titre requires further study and should prove an important area of research of practical significance. 5 Receptors for Diploptera punctata allatostatins

The occurrence of multiple allatostatin peptides has prompted the search for receptors of these important peptides. To date, only allatostatin-binding proteins (putative receptor) for D. punctata allatostatin I have been tentatively identified (Cusson et al., 1991, 1992) and further characterization of this molecular species has proven difficult, in part because of the lack of a reliable binding assay. The search for allatostatin receptors has also been complicated by the small amounts of tissues available (the maximal volume of a typical D. punctata CA is -20 nl and its weight is