Human Cell Culture Protocols

Establishment and Maintenance of Normal Human Keratinocyte Cultures Claire Linge 1. Introduction Keratmocytes are the major cellular component of the eptdermis, which is the strattfied squamous epttheha forming the outer-most layer of skin. The keratmocytes lie on a basement membrane and are organized mto drstmct cell layers which differ morphologically and biochemically These regions from the basement membrane outward are the basal, spmous, granular, and cormtied layers. Cellular proliferation takes place mamly m the basal layer. On division, keratmocytes give rise to either replacement progenitor cells and/or cells that are committed to undergo the process of terminal differentiation These latter cells leave the basal layer and gradually migrate upward, simultaneously progressing along the differentiation pathway as they go. Finally they reach the outer surface of the epidermis m the form of fully mature functional cells, the corneocytes. The function of these mature cells is the protectton of the underlying viable tissues from the external milieu. The reasonsfor studying keratmocytesare many-fold, and include mvesttgatton of the pathogenesis of keratmocyte-related diseases and also examination of the control mechanisms of proliferation and differentiation. The development of a long-term m vitro keratmocyte system, allowing precise experimental mampulatton of these cells, has been mstrumental in the rapid advances made m these fields over the last two decades. Initial attempts to grow keratmocytes were limited to the use of organ and explant cultures (1). Using these techniques, whole pieces of skm can be kept ahve in the short term, and growth IS confined to the tissue fragment or onto the plastic surroundmg the explant. However, these cultures have an extremely short life-span and also limited apphcatron, since mixed cultures of keratmocytes From

Methods m Molecular Medune Edlted by G E Jones Humana

1

Human Press

CeN Culture Inc , Totowa,

Protocols NJ

2

Lmge

and fibroblasts are obtamed. Fibroblasts present a major problem where keratmocyte culture is concerned. Even small amounts of tibroblast contamination can lead to their overgrowth of keratmocyte cultures This is because of the htgh proltferation rate of tibroblasts compared with that of keratmocytes even under opttmum conditions for keratinocyte growth. The greatest advance m the development of a long-term keratmocyte culture method came m 1975, when Rhemwald and Green reported the serial culttvatton of pure cultures of keratmocytes from a single-cell suspension of eptderma1 cells (2). Thts was achieved by growing the cells m serum-contammg medmm on a mesenchymal feeder cell layer (trradtated mouse 3T3 cells) Usmg this feeder layer of viable, yet nonproltferatmg, mesenchymal cells, reduced fibroblast contammation and growth vastly, if not completely, but enhanced the proliferation of keratmocytes The longevity of these keratmocyte cultures was further improved on the addition of a variety of mitogens discovered to be important for the health and growth of keratmocytes A list of these cytokmes and the relevant references are given m Section 2. The most vttal of these mttogens 1s eptdermal growth factor (EGF) (3) Smce the mtroductton by Rhemwald and Green of a method of long-term culture of keratmocytes, alternattve culture methods have been developed, each being designed for specific experimental requirements. The degree to which the pattern of keratmocyte dtfferenttatton m vtvo IS reproduced m vitro depends on the conditions and methods of culture Keratmocyte cultures can vary from undifferentiated monolayers (4) under low calcium condtttons (~0.06 mM) to fully differentiated stratified multtlayers achieved when grown m skm-equtvalent cultures (5-s). Skm equivalents have been fashtoned by growmg keratmocytes on the followmg 1 Collagen disks. Collagen-based, thin permeable membranes produced by ICN Flow 2 ECM gels Usually collagen-based, but can contain other extracellular matrix (ECM) constituents, such as lamnun or fibronectm 3 Dermal equivalent ECM gels that contain viable mesenchymal cells, such as dermal fibroblasts or fibroblastic cell lines 4 DED De-epidermized dermis Produced by multiple freeze-thawing of a piece of skin, after which the dead epidernns can simply be peeled off This treatment kills off all endogenous cells, leaving an uninhabited connective tissue skeleton, which can then be reseeded with cells of the experimenters choice The more closely the culture condtttons duplicate the ttssue environment (i.e., actdtc pH, collagenous substrate, presence of mesenchymal cells, au interface, etc.), the more complete the expression of eptdermal dtfferenttatton characteristtcs. The method detailed in this chapter is adapted from that of Rheinwald and Green, and allows the long-term maintenance of keratmocytes in culture sup-

Human Keratwcyte

Cultures

3

plymg a stock of healthy cells, which can be used either as they are or m any of the alternative methods mentioned for experlmentatlon 2. Materials 1 3T3 cells available from the European Collection of Animal Cell Cultures (ECACC #8803 1146) 2 3T3 medium Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal calf serum (FCS), 4 mA4 L-glutamme, 100 U/mL pemcillm, 100 pg/mL streptomycm The shelf life of this medium IS approx 4-6 wk Owing to the Instability of L-glutamme at 4”C, however, fresh L-glutamine can be added Media supplements are stored as concentrated stocks at -2O”C, and media (both basic and supplemented) are stored at 4°C Note All reagents are available from tissue-culture retailers FCS should be batch tested to obtain opt]mum serum for cell growth 3 Freshly isolated normal human skm m the form of abdominal or breast reductions or circumcisions 4 Skm transport medium DMEM supplemented with* 10% FCS, 100 U/mL pemclllin, 100 pg/mL streptomycin, 2 5 pg/mL fungizone, 50 pg/mL gentamicin Media supplements are stored as concentrated stocks at -2O”C, and media (both basic and supplemented) are stored at 4°C 5 Keratmocyte growth medium (KGM) made up of a 3 1 (v/v) mixture of Ham’s F12 and DMEM media, supplemented with 10% FCS, 4 mM L-glutamrne, 100 U/mL pemclllm, 100 pg/mL streptomycin, 0 4 pg/mL hydrocortlsone (9), 10-‘“Mcholera enterotoxm (lo), 5 pg/mL transferrm (II), 2 x lo-“M hothyromne (II), 1 8 x 10dA4 adenme (12), 5 pg/mL msulm (1 I), and 10 ng/mL EGF (9) This media should be used fresh If possible, but has a shelf life of approx 1 wk Media supplements are stored as concentrated stocks at -20°C, and media (both basic and supplemented) are stored at 4°C Stock supplements are usually made up m phosphate-buffered salme (PBS) contammg a carrier protein, such as 0 1% bovine serum albumin Certain media supplements require dlssolvmg as follows Na tn-lodothyromne dissolves mltlally m 1 part of HCl and 2 parts ethanol, ademne dissolves m NaOH, pH 9 0, msulm dissolves m 0 05M HCl, and hydrocortisone dissolves m EtOH 6 PBS* All PBS referred to m this text lacks calcium and magnesmm Ions and IS made up of the followmg 1% (w/v) NaCl, 0 025% KCl, 0 144% Na2HP04, and 0 025% KH2P0, This solution IS pH-adjusted to 7 2, autoclaved at 12 1“C ( 15 psi) for 15 mm, and stored at room temperature. 7 Trypsmizatlon solution 1 vol of trypsm stock IS added to 4 vol of EDTA stock and used lmmedlately. Trypsm stock. trypsm (Dlfco, Detroit, MI 1 250) IS made up of 0 25% (w/v) Trls-salme, pH 7 7 (0 8% NaCl, 0 0038% KCI, 0 01% Na*HPO,, 0 1% dextrose, 0 3% trlzma base). Stocks are filter sterilized and stored ahquoted at -2O’C EDTA stock EDTA 1smade up of 0 02% (w/v) EDTA m Caand Mg-free PBS, autoclaved at 121°C (15 psi) for 15 min, and stored at room temperature

4

Lmge

8 The followmg sterile equipment 1s reqmred: forceps, scalpel, lrls scissors, hypodermlc needles, medlcal gauze, tissue-culture flasks, and Petri dishes 9 Mltomycm-C stock solution dissolve m sterile H,O to a concentration of 400 pg/mL Store at 4°C m the absence of light. Solution IS stable for 34 mo 10. Dlspase medium 3T3 medmm contammg 2 mg/mL Dlspase (Boehnnger Mannhelm, Mannhelm, Germany) and filter sterlhzed Use unrnedlately

3. Methods 3.1. Routine Maintenance

of 3T3 Cell Line

These adherent cells are grown m 3T3 media at 37°C to near confluence (see Note 1) and passaged as follows 1 Remove media from flask, and wash cells with an equivalent volume of PBS 2 Add the trypsmlzatlon mixture to the flask at approx 1 5 mL/25 cm2 surface area, and mcubate at 37°C for approx 5 mm, or until all cells have rounded up 3 Add 4 vol of medium to deactivate the trypsm and EDTA, and disperse the cells with repeated ptpetmg 4 Estimate the cell number usmg a hemocytometer, and pellet the cells at approx 300g for 5 mm 5 Resuspend the cells m fresh media, and seed mto flasks or Petri dishes at approx 3 x 1O3 cells/cm* of surface area. Note Density of cells at seeding can be varied depending on when confluence IS required. 6 Cells should reach confluence m approx 3-5 d

3.2. Production

of Feeder Layers

1 Select flasks of exponentially growmg 3T3, which have no more than 50% of the flask’s surface area covered by cells, replace the media, and Incubate for a further 24 h 2 Add approx l-l 0 pg of mltomycm-C/ml of medium (see Note 2), and incubate for a further 12 h 3. Wash the flask three times with fresh medium Incubate the cells with the final wash for approx 1O-20 mm at 37°C 4 Harvest the cells lmmedlately by trypsmlzatlon m the usual manner (detalled m Sectlon 3 1 ) and seed m fresh flasks at approx 2 5 x lo4 cells/cm* m keratmocyte media (1 mL media/5 cm2 of plastic surface area) 5 Incubate at 37°C for approx 12 h to allow the cells to adhere and spread before seeding with keratmocytes

3.3. Initiation

of Keratinocyte

Cultures

1 Place skm sample directly from patient (see Note 3) as sterilely as possible mto a Universal contammg a covermg volume of skm transport media at 4°C (see Note 4) 2 Before processmg, remove skm from the transport media, submerge brlefly m alcohol three times, and shake dry m the tissue-culture hood

Human Keratinocyte Cultures

5

3. Place skm mto a shallow stertle container (a IO-cm Petri dash IS perfect for small skm samples), and usmg fine forceps and trts scissors, tram away the hypodermis, I e., the adipose and loose connective ttssue, unttl only the eptdermts and the relatively dense dermis remam (see Notes 5 and 6). 4 Flatten the skin, eptdermis down, onto the surface of the Petrt dish and usrng a sterile scalpel, cut the skm mto long 2-3-mm thin strtps 5 Place the strips mto a Umversal contammg at least a covermg amount of dtspase medmm, and mcubate etther overnight at 4°C or for 2-4 h at 37°C 6 After the mcubatton, remove the strtps ofskm from the dtspase media, dab excess media off on the instde of the lid of a IO-cm Petri dish, and place the relatively media-free strips mto the Petri dish Peel the epidermis away from the dermis wtth two sterile hypodermic needles The eptdermts IS a semiopaque thm layer, whereas the connecttve ttssue of the dermts will have absorbed fluid and ~111 appear as a thick swollen shghtly gelatmous layer. Thts should come away eastly. If secttons remain attached, then either the strtps were too thtck or further mcubation m dispase is required (Note This should not be a problem after mcubatton overmght at 4°C ) 7 Place the eptdern-us strtps only mto 5 mL of trypsm stock solutton, and shake rapidly for 1 mm. Add 15 mL of DMEM/lO% FCS to inactivate the trypsm. Remove upper eptdermal layer pteces by ptpetmg through sterile gauze mto a sterile universal 8 Pellet the smgle-cell suspension by centrifugatton at approx 300g for 5 mm, resuspend m keratmocyte growth medta, and count Seed at approx 2-5 x 1O4 viable cells/cm2 onto the preplated feeder layers

3.4. Routine

Culture of Keratinocyte

Strains

Regularly change the medium twice per week. With ttme, the 3T3 feeder cells will begm to dte and detach from the flask. Replace these with fresh feeder cells as necessary (see Note 7). The cultures should reach confluence within 1O-14 d It IS Important to passage the keratmocytes before they reach confluence, 1.e, when they cover approx 7&80% of the surface area of the flask or Petrt dish (see Note 8). To passage the keratmocytes, proceed as described for passaging of 3T3 cells m Section 3 1 , steps 14, with the exception that keratmocytes will take longer to trypsunze (10-I 5 mm) and wtll requtre vtgorous agitation of the flask to detach the rounded up cells from the surface. Once counted, seed the keratmocytes onto fresh feeder layers at a denstty of approx 5-50 x lo3 viable cells/cm 2 The denstty seeded depends on when confluence ts required Healthy secondary keratmocyte cultures should reach confluence within 7-10 d (see Note 9)

4. Notes 1 The 3T3 cell line 1s an undemanding cell lme to maintam m culture, and few problems should be encountered by a competent tissue culturtst The only thing

Lmge

2

3

4 5

6

7

8

9

to note IS that cells that have become overconfluent and begun to pile up (I e , form foci) should not be used either for the contmuatlon of stocks or for the production of feeder cells, since the cells appear to transform further and can become resistant to lrradlatlon or mltomycm-C treatment, mamtammg their proliferative ability and thereby overrunnmg keratmocyte cultures The exact concentration of mltomycm-C required to produce viable, yet nonprollferatmg 3T3 cells should be titrated, since It varies with batch Alternatively, feeder cells can be produced by irradiation with approx 6000 rads using a y-u-radiator (cobalt60) This can be performed on 3T3 cells that are either attached to the flask surface or m suspension (depending on the size of your irradiator) The exact dose of radlatlon required to produce viable, yet nonprohferatmg cells must be titrated Generally, keratmocyte cultures from younger patients (< 16 yr old) have a greater growth potential It 1s advlsed that cultures should not be mltlated from patients older than 60 yr Skin samples remain viable for up to 20 h when stored m skin transport media at 4°C The density of the hypodermls varies with the biopsy site For foreskins, the hypodermls 1s particularly loose and therefore easily dissectible, whereas skm taken from the back has an extremely dense hypodermls, which proves difficult to remove In the latter case, Just remove as much extraneous connective tissue as possible Skm 1soften contaminated with bacteria or yeast Submerging the skm sample m alcohol before processmg should kill most forms of contamination However, pockets of bacteria which have become trapped m sweat or sebaceous pores may be present Foreskins are particularly prone to blocked pores. Fortunately, once the skm IS stretched upside down across the Petri dish, the presence of blocked pores IS usually obvious The affected areas of skm should be carefully dissected out and discarded, taking particular care not to cut mto the blocked pore An adequate feeder layer density 1s extremely Important for the continued growth of keratmocytes and the reduction of fibroblast growth A good feeder layer should cover approx 70% of the surface area In order to mamtam healthy cultures of rapidly growing keratmocytes (see Fig 1), It IS Imperative that keratmocyte cultures are passaged well before full confluence 1s reached, when approx 70% of the flask’s surface area IS covered with keratmocytes If this IS not done, then the underlying proliferative keratmocytes will begin to die off or differentiate This is presumably owing to the relatively impermeable multiple layers of differentiating keratmocytes reducmg the nutrients available to the basal layer Keratmocyte stocks can be successfully stored m liquid nitrogen Only cultures of rapidly growing keratmocytes should be chosen for freezing, 1 e , select flasks where only 50% of the surface area 1s covered by keratmocyte colonies Trypsmlze, count, and pellet the cells as usual Resuspend the cells at approx l-5 x lo6 cells/ml m 90% FCS and 10% dlmethyl sulfoxlde, place mto cryotubes

Human Keratinocyte Cultures

7

Fig. 1. A 7-d-old primary culture showing a healthy keratinocyte colony (center) surrounded by dying feeder cells. Note the symmetrical appearance of the colony, the smooth rounded edges of which are typical of rapidly growing keratinocyte colonies. The phase-bright debris located at the center of the colony is commonly seen, particularly in primary cultures, and is thought to be owing to cellular aggregation of terminally differentiating cells to the proliferating cells before the latter adhere to the plastic and begin to grow. Magnification 200x.

immediately, insulate tubes (wrap in multiple tissue layers or place within polystyrene container), and freeze overnight at -80°C before placing into liquid nitrogen.

References 1. Cruickshank, C. N., Cooper, J. R., and Hooper, C. (1960) The cultivation of cells from adult epidermis. J. Invest. Dermatol. 34,33!9-342. 2. Rheinwald, J. G. and Green, H. (1975) Serial cultivation of strains of human epidermal keratinocytes: the formation of keratinizing colonies from single cells. Cell 6,33 1-344. 3. Rheinwald, J. G. and Green, H. (1977) Epidermal growth factor and the multiplication of cultured human epidermal keratinocytes. Nature 265,42 l-424. 4. Boyce, S. T. and Ham, R. G. (1983) Calcium regulated differentiation of normal epidermal keratinocytes in chemically defined clonal culture and serum-free serial culture. J. Invest. Dermatol. 81,33sAOs. 5. Bell, E., Sher, S., Hull, B., Merril, C., Rosen, S., Chamson, A., Asselineau, D., Dubertret, L., Coulomb, B., Lapiere, C., Nusgens, B., and Neveux, Y. (1983) The reconstitution of living skin. J. Invest. Derm. 81,2s-10s.

8

Linge

6 Prunetras, M , Regmer, M , and Woodley, D (1983) Methods of cultivation of keratmocytes at an au liquid mterface J Invest Dermatol 81, 28s-33s 7 Boyce, S T , Chrtsttanson, D J., and Hansborough, J F (1988) Structure of a collagen-glycosammoglycan dermal skm substitute opttmtsed for cultured human eptdermal keratmocytes J Boomed Mater Res 22, 939-957 8 Yannas, I V , Lee, E , Orgtll, D P , Skrabut, E M , and Murphy, G F (1989) Synthesesand charactertsatton of a model extracellular matrix that inducespartial regeneration of adult mammahanskm Proc Nat1 Acad SCI USA 86, 933-937 9 Rhemwald, J G (1980) Serial cultrvatton of normal human eptdermal keratmocytes Methods Cell Brol 21, 229-254 10 Green, H (1978) Cychc AMP m relation to prohferatton of the eptdermal cell a new vtew Cell 15,801-811 11 Watt, F M and Green, H (198 1) Involucrm synthesis IScorrelated with cell size m human eptdermal cultures J Cell Blol 90, 738-742 12 Wu, Y J., Parker, L. M , Binder, N E , Beckett, M A , Smard, J H , Grtffiths, C T , and Rhemwald, J G (1982) The mesotheltal keratms a new family of cytoskeletal protems identified m cultured mesothehalcells and non-keratmtsmg epttheha Cell 31, 693-703

Cultivation

of Normal Human Epidermal Melanocytes

Mei-Yu Hsu and Meenhard

Herlyn

1. Introduction An Important approach m studies of normal, diseased, and malignant cells IS their growth m culture. The lsolatlon and subsequent culture of human eplderma1 melanocytes has been attempted since 1957 (l-5), but only since 1982 have pure normal human melanocyte cultures been reproducibly established to yield cells m sufficient

quantity

for bIological,

biochemical,

and molecular

analyses (6). Selective growth of melanocytes, which comprise only 3-7% of epldermal cells in normal human skin, was achieved by suppressing the growth of keratmocytes and fibroblasts m epldermal cell suspensions with the tumor promoter 12-O-tetradecanoyl phorbol- 13-acetate (TPA) and the Intracellular cyclic adenosme 3’, 5’ monophosphate (CAMP) enhancer cholera toxin, respectively, which both also act as melanocyte growth promoters. Recent progress in basic cell-culture technology, along with an improved understanding of culture requirements, has led to an effective and standardized isolation method, and special culture media for selective growth and long-term maintenance of human melanocytes. The detailed description of this method IS aimed at encouragmg its use in basic and applied blologlcal research 2. Materials 1 Normal skm-transportmg medmm The medmm for collectmg normal skm IS composed of Hank’s balanced salt solution (HBSS without Ca2+ and Mg2+, Glbco-BRL [Grand Island, NY], #2 1250-089) supplemented with pemclllm (100 U/mL, USB [Cleveland, OH], #199B5), streptomycin (100 pg/mL, USB, #2 1B65), gentamlcm (100 pg/mL, BloWlttaker [Walkersvllle, MD], #17-5 1SZ), and funglzone (0 25 pg/mL; JRH Biosciences [Lenexa, KS], #59-604-076) After sterlhzation through a 0 2-pm filter, the skm-transporting medmm IS transferred mto sterile containers m 20-mL ahquots and stored at 4°C for up to 1 mo From

Methods In Molecular Medune Human Cell Culture Edlted by G E Jones Humana Press Inc , Totowa,

9

Prorocols

NJ

‘0

Hsu and Herlyn

2 Eptdermal tsolatton solutton Dilute 0 5 mL of 2 5% trypsm solutton (BIOWtttaker, #17-160E) with 4 5 mL of HBSS without Ca2+ and Mg2+ at pH 7 4 to yield a final trypsm concentration of 0 25% 3 Cell-dispersal solutton 1 25 U/mL dtspase (neutral protease, grade 11, Boehrmger Mannhetm [Indtanapohs, IN], #295-825) 0 1% (w/v) hyaluronldase (type 1S from bovme testis, Sigma [St Louts, MO], #H3506), and 10% heat-inactivated fetal calf serum (FCS, Stgma, #F2442) m MCDB 153 medium (Sigma, #M7403) supplemented with 2 mM CaCl, and mixed wtth Letbovttz’s L- 15 at a 4 1 (v/v) ratto 4 Bovine pmutary extract (7,8) The followmg should be prepared before extraction a Weigh out bovine pttmtary glands (Pel-Freeze Btologtcals [Rogers, AR], #57 133-2) mto 25-30-g batches Place m Z~ploc@ bags, and store at -70°C b Thirty liters of cold (4°C) 1X phosphate-buffered salme (PBS) wtthout Ca2+ and Mg2’ c One hter of cold 0 15MNaCl saline solution d One ltter of 0 2 mM EDTA solution e Prechill high-speed centrifuge rotor at 4°C f Boll 6000-8000 Dalton dialysis tubing (10 strips about 2 ft in length, Spectrum Medical Industries [Los Angeles, CA], #132655) twtce In ddH,O and once m EDTA solution (prepared m Section 2 . step 4d) Boil for 20 mm each time Leave tubing m beaker filled with 1X PBS, and store at 4°C for up to 3 d The bovine pmntary extract IS prepared as follows (steps g-p are performed m a cold room) Thaw and rinse pttuttary glands m ddH20, handling each batch of prewetghed g pituitary glands separately h Pulse-blend thawed pttmtary glands in a blender containing 2 38 mL cold salme solutton/g of pmutary gland to break up large pieces Pulses should not exceed 30 s because the temperature must remam low Transfer the resulting mixture mto a 2-L flask 1 Repeat steps g and h for all batches of pttmtary glands J Stir the pooled mixture m the 2-L flask for 90 mm k Pour the mixture mto plastic centrifuge bottles I Spm at 12,000g for 45 mm at 4°C m a prechtlled rotor to remove debris m Plpet supernatant from bottles into dialysis tubmg, and begm dtalysts agamst cold 1X PBS m a cold room n Change the buffer three times m 3 d 0 Filter sequentially through low protem-bmdmg filters of 0 45 urn (Mtlhpore [Marlborough, MA], SLHV025 LS) to eliminate any fragments and 0 2 pm (Milhpore, SLGV025 LS) to sterilize the filtrate Prepare 5-mL ahquots, and store at -70°C for up to 6 mo Once thawed, pttmtary extract should be diluted m medium tmmedtately P Determine the protein concentratton of the extract using a protem assay ktt (Pierce [Rockford, IL], BCA Protein Assay Reagent, #23225H), and titrate the optimal concentratton m medium (approx 40 pg/mL)

Epldermal Melanocytes 5 Melanocyte growth medium (MGM) The following stock solutions are requn-ed a Insulin (Sigma, #ISSOO), 5 mg/mL stock Dissolve 0 5 g of crystalhne lnsulm m 1 mL of 0 OlM HCl solution, and bring the volume up to 100 mL with ddH,O Filter-stenhze, prepare smgle-use ahquots m sterile vials, and store at -70°C for up to 6 mo Do not thaw and refreeze. Use 0 5 mL/500 mL of medium to give a final msulm concentration of 5 pg/mL b. Epldermal growth factor (EGF, Sigma, #E4127), 5 pg/mL stock Suspend 0 1 mg of EGF powder m 20 mL of HBSS without Ca*+ and Mg2+ Sterilize through a 0 2-pm filter, prepare I-mL allquots, and store at -70°C Use 0 5 mL/500 mL of medium to give a final EGF concentration of 5 ng/mL Avoid repeated freezing and thawmg c TPA (Chemicals for Cancer Research [Chanhassen, MN], #8005), 0 25 mg/mL stock Dissolve 10 mg of TPA m 40 mL of 100% ethanol, ahquot, seal with Parafilm@, and store at -20°C Use 20 pL/500 mL of medfum to give a final TPA concentration of 10 ng/mL MGM is prepared as follows MIX MCDB 153 (Sigma, #M7403) supplemented with 2 mA4 of CaCI,, with Lelbovltz’s L-15 (Glbco-BRL, #41300-070) at a 4 I ratio (v/v), and add 2% heat-inactivated FCS (Sigma, #F2442), 5 pg/mL of insulin, 5 ng/mL of EGF, 10 ng/mL of TPA, and bovine pltultary extract to yield 40 yg/mL of pituitary protein m the medium Store the MGM at 4°C for up to 8 d 6 Trypsm-versene solution Make a 5X stock by mixmg 0.5 mL of trypsm solution (2 5%, BloWlttaker, #17-160E) with 100 mL of versene solution composed of 0 1% EDTA (Fisher [Pittsburgh, PA], #02793-500) m Ca2’- and Mgz’-free PBS (pH 7 4) To prepare trypsln-versene solution, dilute 5X stock with HBSS (Ca2+-, Mg2’-free) to give a final concentration of 0 0025% trypsm and 0 02% versene 7 Cell-preservative medium Prepare 5% (v/v) dlmethyl sulfoxlde (DMSO; Sigma, #D2650) m 95% heat-inactivated FCS as needed

3. Methods 3.1. Day 1 1 Prepare the followmg m a lammar flow hood, one pair of sterile forceps, curved scissors, and surgical scalpel blade, 5 mL of epldermal lsolatlon solution (see Section 2., step 2) m a sterile centrifuge tube, 10 mL of Ca*+- and Mg2+-free HBSS m a sterile nontissue-culture Petri dish, and 10 mL of 70% ethanol m a separate sterile Petri dish 2. Soak the skm specimens m 70% ethanol for 30 s Transfer skm to another Petri dish contammg HBSS to rinse off ethanol (see Notes 1 and 2) 3 Cut skin-rmg open, and trim off fat and subcutaneous tissue with scissors (see Note 3) 4. Cut skm mto pieces (approx 2 x 3 mm2) using the surgical scalpel blade with one-motion cuts (see Note 4) 5 Transfer the pieces mto the tube containmg epldermal isolation solution Cap, Invert, and incubate the tube m the refrigerator at 4°C for 18-24 h (see Note 5)

12

Hsu and Herlyn

3.2. Day 2 1 Prepare the followmg m a lammar flow hood* one pan of sterile forceps and a surgical scalpel blade, 10 mL of Ca2+- and Mg2+-free HBSS m a sterrle, nontissueculture Petri dish, two empty sterile Petri dishes; and 5 mL of cell-dispersal solution m a 15-mL centrifuge tube 2 Pour tissue m eptdermal isolation solution mto one of the empty Petri dishes Transfer tissue pieces to the Petri dish containing HBSS Separate eptdetmts (thin translucent layer) from dermis (thick opaque layer) using the forceps Hold the dermal part of the skm piece with the forceps, and gently slide the epidermal side on the dry surface of a nontissue-culture Petri dish The epidermis should strck to the Petri dish Discard the dermis immediately (see Note 6) 3 Add a drop of cell-dispersal solution on the resulting eptdermis to avoid drying and to neutralize trypsm while tsolatmg the epidermis from the remaining skm pieces Repeat procedure m steps 2 and 3 for each piece of tissue (see Note 6) 4 Transfer the collected epidermal sheets with cell-dispersal solutton from the Petri dish to the centrifuge tube Incubate the tube at 37°C for 24 h depending on cell dtsaggregatton Vortex the tube vigorously to release single cells from epidermal sheets. Wash the resulting single-cell suspension three times with Ca2+- and Mgzf-free HBSS Centrifuge for 5 mm at 2000 rpm at room temperature. Aspirate the supernatant, which may contain remammg stratum corneum. Resuspend the pellet with MGM (see Note 7). 5. Plate the resultmg eptdermal cell suspension at approx 2 x IO5 cells/cm2 m the tissue-culture vessel Incubate at 37°C in 5% CO,/95% au for 48-72 h

3.3. After

2 Days

1 Wash culture with MGM to remove nonadherent cells Medium change should be performed twice each week If the culture is contaminated with tibroblasts, start treatment with MGM contammg 200 pg/mL of geneticm (G418, GibcoBRL, #11811) for 2-3 d Seventy percent confluent primary cultures can be obtained in 2 wk (see Note 8) 2 Subcultivation Primary cultures established from foreskins usually reach 70% confluence wtthm 12 d Cultures are treated with trypsm-versene solution (see Section 2 , step 6) at room temperature for 1 mm, harvested with Leibovitz’s L- 15 containing 10% heat-inactivated FCS, centrifuged at 2000 rpm for 3 mm, resuspended m MGM, remoculated at lo4 cells/cm2, and serially passaged Medium IS changed twice each week. 3 Cryopreservatton Melanocyte suspensions harvested by trypsm-versene and Letbovttz’s L- 15 containing 10% FCS are centrifuged at 2000 rpm for 5 mm and resuspended m cell-preservative medium (see Section 2 , step 7) contammg 5% DMSO as a cryopreservattve Cells are normally suspended at a density of 106/mL and transferred to cryotubes The tubes are then placed m a plastic sandwich box (NalgeneTM Cryo 1°C Freezing Container, Nalge [Rochester, NY], #5 lOO-OOOl), which is mnnedtately transferred to a -70°C freezer. The msulatton of the box ensures gradual coolmg of the cryotubes and results m over 80%

Epldermal Melanocytes

13

viability of the cells on thawing After overmght storage m the freezer, the cryotubes are placed in permanent storage in liquid nitrogen 4 Thawmg The melanocyte suspension is thawed by placing the cryotube m a water bath at 37°C When the cell-preservative medium is almost, but not totally defrosted, the outside of the tube is wiped with 70% ethanol The cell suspension is then withdrawn, quickly diluted m MGM at room temperature, centrifuged, and resuspended m fresh MGM Cell viabilny IS determined by Trypan Blue exclusion. The resultmg melanocytes are then seeded at a density of 104/cm2

3.4. Results 3 4.1. Minimal Growth Requirements Earlrer studies of normal melanocytes (6,9,1(I) were done using medta contaming 5-l 5% FCS, which provides a host of poorly characterizedgrowth-promotmg actrvmes. Deprivation of serum and brain tissue extracts from media has led to the delineation of four groups of chemrcally defined melanocyte mttogens. Peptide growth factors, mcludmg basic fibroblast growth factor (bFGF, Il. 12), which IS the mam growth-promoting polypeptlde m bovine hypothalamus and pmutary extracts, msulm/msulm-like growth factor-l (IGF-1, 13), EGF (1#,15), transforming growth factor-a (TGF-a, 16), endothelms (ET, I7), and hepatocyte growth factor/scatter factor (HGFISF, 18,19) Calcium, since reduction of Ca’+ concentrations m MGM from an optimal 2 0 to 0 03 mM reduces cell growth by approx 50% (20), and cation-bmdmg proteins, such as tyrosmase at 10-l ‘M, and ceruloplasmm at 0.6 U/mL (20) Enhancers of mtracellular levels of CAMP, mcludmg a-melanocyte-stimulatmg hormone (a-MSH) at 10 ng/mL (21); forskolm at 10e9M (20), folhcle-stunulatmg hormone (FSH) at 10m7M (22); and cholera toxin at 10-‘*M (6,20,23,24) Activators of protem kmase C (PKC), such as TPA (25), which IS ltpophthc and cannot be removed by simple washing, and 20-oxo-phorbol- 12,13-dibutyrate (PDBu) at 10-6M(25), which is a stmtlar derivative, but more hydrophilic Recent data suggest that the tigliane class phorbol compounds, such as 12 deoxyphorboLl3 isobutyrate (DPIB) and 12 deoxyphorboLl3 phenylacetate (DPPA), which possess dtmmtshed tumor-promotmg activity, are able to activate PKC as well as stimulate melanocyte proliferation (26)

3 4.2. Morphology Human epidermal melanocytesgrown m MGM normally exhibit abr- or tripolar morphology wtth varying degrees of pigmentation (Fig. 1). Dendrlcrty may increase at higher passage levels. 3.4.3. Expressron of Antigens Extensive

studies have been done to characterize

the anttgentc

phenotype

of

malignant melanoma cells (27) On the other hand, very few attempts have

Hsu and Herlyn

Fig. 1. Morphology of normal human epidermal melanocytes grown in medium supplemented with bFGF (pituitary extract), serum, and phorbol ester.

been made to produce monoclonal antibodies (MAbs) to normal melanocytes (15,28). Cultured melanocytes share with melanoma cells the expression of a variety of cell-surface antigens (melanoma-associated antigens), including p97 melanotransferrin, integrin p3 subunit of the vitronectin receptor, gangliosides CID3 and 9-O-acetyl GD,, chondroitin sulfate proteoglycan (15), and MelCAM/MUC 18 (29). However, these antigens are not expressed by normal melanocytes in situ (30). Table 1 summarizes the expression of antigens on melanocytes in situ and in culture. The observed divergent antigenic phenotype in culture and in situ suggests a role for epidermal microenvironmental signals in controlling the melanocytic phenotype. Indeed, accumulating evidence indicates that undifferentiated keratinocytes can control proliferation, morphology, pigmentation, and antigen expression of melanocytes in coculture (28,32-35). However, the underlying mechanisms responsible for keratinocyte-melanocyte interactions remain unclear. 3.4.4. Growth Characteristics Melanocytes from neonatal foreskin can be established with a successrate of 80% and have a maximum life-span of 60 doublings, with a doubling time of 2-6 d. Heavily pigmented cells isolated from black individuals have a shorter doubling time and tend to senesceafter 20-30 doublings. By contrast, epider-

Epldermal Melanocyfes Table 1 Expression

of Antigens

15 on Melanocytes

In Situ and in Culturea

Antigens

In Situ

CD26 gp145

++++ ++

++++ ++

TRP- I E-cadherm

++++ ++++

++++ ++++

a-Catenm

++++

++++

p-Catenin

++++

++++

-

++++

Tenascm Flbronectin Chondroltm sulfate proteogylcan p97 Melanotransferrm NGF-receptor (~75)

k +

+ ++++ ++++ ++++ ++++

9-O-acetyl

Ik

+++

f

++++

Integrm

& subunlt

CD,

GD3

HLA-DR

Mel-CAM

(MUC 18)

In culture”

-

-

-

++++

“-, Lack of expresslon, ?, O-20%, +, 2&40%, ++, 40-60%, +++, U&80%, +++-I-, 8&100% ‘Results were obtamed with melanocytes grown in MGM contammg TPA at passages 2-20, with the exceptlon of gp145, which IS expressed more strongly on melanocytes cultured m the absence of TPA (31 and unpubhshed data)

ma1 melanocytes from adult skm only grow m about 10% of casesand for no more than 10 doublmgs with a doublmg time of 7-14 d. The cells do not grow beyond 70% confluency and exhibit signs of growth arrest by contact mhlbltlon. Normal melanocytes do not proliferate anchorage independently m soft agar and are nontumorlgemc m athymlc nude mice (14). 4. Notes 1 Tissue source and collection* The sources of tissue for melanocyte cultures are human neonatal foreskins obtained from routine clrcumclslon and normal adult skin acquired from reduction mammoplasty At the time of exclslon, the skin is placed into a sterile container with 20 mL of normal skin-transporting medium (see Section 2 , step 1) supplied m advance and kept near the surgical area at 4°C Specimens are dehvered immediately to the tissue-culture laboratory or stored at 4°C Neonatal foreskins can be kept for up to 48 h, and normal adult skin, for up to 24 h However, the fresher the specimens, the higher the yield of live cells on lsolatlon 2 Sterlhzatlon of skm specimens Reduce contammatlon by a short treatment (30 s) of intact skm with 70% ethanol m a lammar flow hood After sterilization, rinse samples with HBSS

Hsu and Hedyn 3 Prehmmary tissue preparation Place tissues in a loo-mm nontissue-culture Petrt dish, and remove most of the subcutaneous fat and membranous material with curved sctssors 4 Adjustment of tissue size for trypsmtzatton To improve reagent penetration, cut the skm samples into small pieces (approx 2 x 3 mm2) rinsed m HBSS 5 Trypsmtzatton Since the first report m 1941 (24). eptdermal cell suspensions have usually been prepared usmg enzymes (most commonly, trypsm) Pieces of skm are incubated m eptdermal tsolatton solutton for up to 24 h at 4°C 6 Separatton of eptdermts from dermrs After mcubatton with trypsm, the eptderma1 tsolatton solution IS replaced by HBSS As ortgmally described (24), crude trypsm splits epidermis from the dermis along the basement membrane Since melanocytes are LocatedJust above the basement membrane, removal of this lowest layer of eptdermal cells requires some effort Each piece of skin IS held with forceps with dermal side up The eptdermal sheet 1sdetached by sbdmg the spectmen onto the dry surface of a nonttssue-culture Petri dish To prevent the eptdermts from drying and to stop trypsnnzatton, a drop of cell-dtspersal solutton containing 10% FCS can be added to the resultmg eptdermal sheet To avoid potential sources of tibroblast contammatton, dermal pieces should be discarded Immediately once they are separated from the epidermis, and the forceps used to hold the dermts should never touch the eptdermal sheets Contammated dermis IS easily recognized by tts white color m contrast to the yellowtsh-brown color of the epidermis Isolated eptdermal sheets m cell-dispersal solution are then transferred to a centrifuge tube 7 Cell dispersal techniques A single-cell suspension IS prepared by enzymattc treatment In the centrifuge tube, clumps of eptdermal tissue are dissociated by celldispersal solutton at 37’C for 2-4 h The resultmg single-cell suspension IS washed three times with Ca*‘- and Mg2+-free HBSS to remove enzymes Cells are then pelleted by centrtfugatton at 2000 rpm for 5 mm and resuspended m MGM The eptdermal cell suspension IS then seeded m single wells of 24-well plastic ttssueculture plates and incubated at 37°C m an atmosphere of 5% CO, and 9.5% au 8 Selective growth Most methods for growmg pure cultures of melanocytes from eptdermal cell suspensions depend on optimal condmons that enable melanocytes, but not keratmocytes, to attach to a substrate and proliferate These condtttons include high oxygen tension (36), high seeding density (37), and the presence of sodium citrate (38), phorbol esters (6), and 5-fluorouractl (39) The presence of phorbol esters (TPA) not only suppresses the growth of keratmocytes, but also promotes melanocyte growth However, despite the potency of TPA m sttmulatmg melanocytes, even mmtmal fibroblast contammatton will eventually result m overgrowth by these more rapidly dtvtdmg cells Ftbroblast contammatton can be eliminated by treatment of cultures with MGM containing 200 mg/mL of geneticm (G418, Gtbco-BRL, #118 11) for 2-3 d 9 Pitfalls and alternatives (see Table 2) The presence of TPA m the medium has been shown to reduce the numbers of melanosomes m human melanocytes m culture and to delay the onset of melamzatton (6) Thus, although this reagent

Table 2 Phenotype

of Neonatal

Foreskin

Melanocytes

in CultureaTb (37) Antigen expresslon

Growth

-L u

Morphology’

Pigmentation

NGFR

gp145

Passage

Passage

Passage -~ 1 5 8

Passage

Passage

Culture conditions

1

5

8

TPA No TPA

+++ +++

+++ +

+++ 0

1 5 8 DDD SFF

1

5

8

+++ +++ +++ + 0 0

1

5

8

++ ++++

+++ 0

+++

0

0 +++

+

“All cultures were mamtamed m the same base medmm 4 parts MCDB 153 with 1 part L-l 5, supplemented with msulm, EGF, bovme pltultary extract (BPE), and 2% FCS ‘)+ to ++++, Degree of growth, pigmentation, or antigen expression, 0, no growth or >14 d doublmg, no plgmentatlon, and no expresslon of antigen 5, spmdle; F, flat, polygonal, D, dendrltlc

18

Hsu and Herlyn supports long-term culture of human melanocytes, it may have limited use m studies of melanocyte differentiation When melanocytes are established m medium without TPA, they grow at doubling times of&7 d for the first 2-3 passages and senesce by passage 5 Imtlally, they assume a spindle morphology, which changes by passages 3-5 to a flat, polygonal morphology (40) The flat polygonal cells are unplgmented and proliferate slowly Concomitant with the morphological and prohferatlve changes, there IS a decrease m expression of the nerve growth factor (NGF) receptor and an increase m expression of gp 145 (Table 2) There are other alternative media for melanocyte culture TIP medium, a TPAcontammg medmm, consists of 85 nM TPA, 0 1 mM lsobutylmethyl xanthlne (IBMX), and lo-20 pg protem/mL placental extract m Ham’s F-10 medium supplemented with 10% newborn calf serum (41) TPA-free medium (42), composed of Ca2+-free Ml99 medmm supplemented with S-10% chelated FCS, 10 pg/mL Insulin, 10 ng/mL EGF, 1Op9A4trnodothyronme, 10 pg/mL transferrrn, 1 4 x lOAM hydrocortlsone, 1Od9A4cholera toxin, and 10 ng/mL bFGF (42), can also support short-term culture of melanocytes

References 1 Hu, F , Stancco, R J , Pmkus, H , and Fosnaugh, R (1957) Human melanocytes m tissue culture J Invest Dermatol 28, 15-32 2 Karasek, M and Charlton, M E (1980) Isolation and growth of normal human skm melanocytes Clm Res 28, 570A 3 Kltano, Y (1976) Stimulation by melanocyte stimulating hormone and dlbutyryl adenosme 3’, 5’-cyclic monophosphate of DNA synthesis m human melanocytes m vitro Arch Derm Res 257,47-52 Mayer, T. C (1982) The control of embryonic pigment cell proliferation m culture by cychc AMP Dev Blol 94, 509-614 Wllkms, L M and Szabo, G C (198 1) Use of mycostatm-supplemented media to establish pure eptdermal melanocyte culture (abstract) J invest Dermatol 76,332 Elsmger, M and Marko, 0 (1982) Selechve proliferation of normal human melanocytes m vitro m the presence of phorbol ester and cholera toxin Proc Nat1 Acad SCI USA 79,20 18-2022 7 Kano-Sueoka, T , Campbell, G R , and Gerber, M (1977) Growth stlmulatmg activity m bovine pmutary extract specific for a rat mammary carcinoma cell lme J Cell Physzol 93,417-424 8 Tsao, M C , Walthall, B J , and Ham, R G. (1982) Clonal growth of normal human epldermal keratmocytes m a defined medmm J Cell Physzol 110,2 19-229 9 Herlyn, M , Herlyn, D , Elder, D E , Bondi, E , LaRossa, D , Hanulton, R , Sears, H , Balaban, G , Guerry, D , Clark, W H , and Koprowskl, H (1983) Phenotyplc characteristics ofcells denved from precursors of human melanoma Cancer Res 43,5502-5508 10 Herlyn, M , Thurm, J , Balaban, G , Benmcelh, J , Herlyn, D , Elder, D E , Bondi, E , Guerry, D , Nowell, P , Clark, W. H , and Koprowskl, H (1985) Charactenstics of cultured human melanocytes isolated from different stages of tumor progression Cancer Res 45,5670-5676.

Epldermal Melanocytes II 12 13

14

15

16

19

Halaban, R , Ghosh, S , and Baird, A (1987) bFGF ISthe putative natural growth factor for human melanocytes In Vztro 23,47-52 Halaban, R , Kwon, B S , Ghosh. S., Dell1 BOVI, P , and Baird, A (1988) bFGF as an autocrme growth factor for humanmelanomas OncogeneRes 3, 177-l 86. Rodeck, U , Herlyn, M , Menssen, H D , Furlanetto, R W , and Koprowskl. H (1987) Metastatlc but not primary melanoma cell lines grow m vitro mdependently of exogenousgrowth factors. Int J Cancer 40,687Y690 Herlyn, M , Rodeck, U , Manclantl, M L , Cardlllo, F M , Lang, A, Zross, A H , Jambroslc,J , andKoprowskl, H (1987) ExpressIonofmelanoma-associated antigens in rapidly dlvldmg humanmelanocytesm culture CancerRes 47,3057-306 I Herlyn, M , Clark, W H , Rodeck, U , Manclantl, M L , Jambroslc, J , and Koprowskl, H (I 987) Biology of tumor progressionm human melanocytes Lab Invest 56,46 l-474 Plttelkow, M R and Shipley, G D (1989) Serum-free culture of normal human melanocytes growth kinetics and growth factor requirements J Cell Ph,vrlol 140,565-576

17 Imokawa, G , Yada, Y , and Mlyagishi, M (1992) Endothelms secreted from humankeratmocytes are mtrmslc mltogens for humanmelanocytes J B/o/ Chem 267,24,675-24.680

18 Halaban, R , Rubm, J S , Funasaka,Y , Cobb, M , Boulton, T , Faletto, D , Rosen, E , Chan, A , Yoko, K , and White, W (1992) Met and hepatocyte growth factor/scatter factor slgnal transduction m normal melanocytes and melanoma cells Oncogene 7,2 195-2206 19 Matsumoto, K , TaJima, H , and Nakamura, T ( 1991) Hepatocyte growth factor IS a potent stimulator of human melanocyte DNA synthesis and growth Bzochem Bzophys Res Commun 176,45-51 20 Herlyn, M , Manclantl, M L , Jambroslc, J , Bolen, J B , and Koprowskl, H (1988) Regulatory factors that determine growth and phenotype of normal human melanocytes Exp Cell Res 179, 322-33 1 21 Abdel-Malek, Z A (1988) Endocrine factors as effecters of mtegumental pigmentation Dermatol Clan 6, 175-184 22 Adash], E Y , Resmck, C E , Svoboda, M E , and Van Wyk, J J (1986) Folhclestimulating hormone enhancessomatomedmC bmdmg to cultured rat granulosa cells J Blol Chem 261,3923-3926 23 Gllchrest, B. A , Vrabel, M A., Flynn, E , and Szabo, G (1984) Selective cultivation of human melanocytes from newborn and adult epidermis J Invest Dermatol 83,37&376 24 Medawar, P B (1941) Sheetsof pure epldermal eplthellum from human skm Nature 148,783 25 Nledel, J E and Blackshear, P J (1986) Protein kmaseC, m Phosphomosrtrdes and Receptor Mechanwms (Putney, J W , Jr, ed ), LISS,New York, pp 47-88 26 Cela, A , Leong, I , and Krueger, J (199 1) Tlghane-type phorbols stimulate human melanocyte prohferatlon potentially safer agentsfor melanocyte culture J Invest Dermatol 96,987-990

20

Hsu and Hedyn

27 Herlyn, M and Koprowskl, H (1988) Melanoma antigens. nnmunologlcal and blological charactenzatlon and chmcal significance Ann Rev Immunol 6,283-308 28 Houghton, A N , Elsmger, M , Albino, A P , Canmcross. J G , and Old, L J (1982) Surface antigens of melanocytes and melanomas* markers of melanocyte dlfferentlatlon and melanoma subsets J Exp Med 156, 1755-l 766 29 Shah, I -M , Elder, D E , Hsu, M -Y , and Herlyn, M (1994) Regulation of MelCAM/MUC 18 expression on melanocytes of different stages of tumor progression by normal keratmocytes. J Am. Path01 145,837-845. 30 Elder, D E , Rodeck, U , Thurm, J , Cardlllo, F , Clark, W H , Stewart, R , and Herlyn, M (1989) Antigemc profile of tumor progression stages m human melanocytes, nevl, and melanomas Cancer Res 49,5091-5096 31 Valyl-Nagy, I and Herlyn, M (199 1) Regulation of growth and phenotype of normal human melanocytes m culture, in Melanoma 5, Series on Cancer Treatment and Research (Nathanson, L , ed ), Kluwer Academic, Boston, MA, pp 85-101 32 Scott G A and Haake, A R (1991) Keratinocytes regulate melanocyte number in human fetal and neonatal skm equivalents J Invest Dermatol 97, 776-78 1 33 DeLuca, M , D’Anna, F., Bondanza, S., Franzl, A T , and Cancedda, R. (1988) Human eplthehal cells Induce human melanocyte growth m vitro but only skm keratmocytes regulate Its proper dlfferentlatlon m the absence of dermis J Cell Bzol 107, 1919-1926 34 Valyl-Nagy, I , Hlrka, G , Jensen, P J , Shah, I -M , Juhasz, I , and Herlyn, M ( 1993) Undifferentiated keratmocytes control growth, morphology, and antigen expresslon of normal melanocytes through cell-cell contact Lab Invest 69,152-l 59 35 Herlyn, M and Shah, I -M (1994) Interactions of melanocytes and melanoma cells with the microenvironment Pzgment Cell Res 7,8 l-88. 36 Riley, P A (1975) Growth mhlbltlon m normal mammalian melanocytes in vitro Br J Dermatol 92,291-304 37 Mansur, J D , Fukuyama, K , Gellm, G A , and Epstein, W L (1978) Effects of 4-tertiary butyl catechol on tissue cultured melanocytes. J Invest Dermatol 70,275-279 38 Prumeras, M , Moreno, G , Dosso, Y , and Vmzens, G (1976) Studies on guinea pig skm cell cultures V Co-cultures of pigmented melanocytes and albmo keratmocytes, a model for the study of pigment transfer. Acta Dermatovenereol 56, l-9 39 Tsujl, T and Karasek, M (1983) A procedure for the isolation of primary cultures of melanocytes from newborn and adult human skin J Invest Dermatol 81,179,180 40 Herlyn, M , Clark, W H , Rodeck, U , Manclantl, M. L., Jambrosic, J , and Koprowskl, H (1987) Biology of tumor progression m human melanocytes Lab Invest 56,46 l-474 41 Halaban, R , Langdon, R , Birchall, N , Cuono, C , Ban-d, A , Scott, G , Moellmann, G , and McGulre, J (1988) Basic fibroblast growth factor from keratmocytes IS a natural mltogen for melanocytes J Cell Bzol 107, 16 1l-l 6 19 42 Tang, A , Eller, M S , Hara, M , Yaar, M , Hlrohashl, S , and Gllchrest, B A (1994) E-cadherm IS the maJor mediator of human melanocyte adhesion to keratmocytes m vitro J Cell SC! 107, 983-992

3 Cultivation of Keratinocytes from the Outer Root Sheath of Human Hair Follicles Alain Limat and Thomas Hunziker 1. Introduction The outer root sheath (ORS) of hair follicles is a multilayered tissue made up predommantly by undtfferentrated keratmocytes (1,2). Although the functions of the ORS cells for hair growth are not estabhshed, it is known that the ORS cells can contrtbute to the regeneration of the eprdermts, as during healing of superficial wounds where the ORS cells migrate out of the follrcle to repopulate the denuded area (3,4). Recent studies also suggest that stem cells for various eptthelral cell populations of the skin are located m the ORS tissue (5,6). Because ORS cells can be regarded as undifferentiated eptdermal keratmocytes (1,2,7,8), they represent a source of easily and repeatedly available keratinocytes, avoiding the dependency on surgery or suction blister material. Moreover, the use of ORS cells 1sespecially surted if cocultures with autologous cells (e.g., peripheral blood mononuclear cells) are performed (9). In the past, several methods for the cultivation of human ORS cells have been described, most of which were based on explanting plucked anagen hair follicles on different growth substrata, such as collagen (IO), bovine eye lens capsules(I I), or collagen gels populated wrth fibroblasts (22) We have developed a simple technique for the cultivation of ORS cells that yields substantially higher cell numbers and in a shorter time as compared to the explant techniques (13,14). Our laboratory routme enables the mitiation of primary cultures starting with ORS cells released from two hair follicles per culture dish. The mam steps of this technique comprise the plucking of scalp From

Methods fn Molecular Med,one Human Cell Culture Edtted by G E Jones Humana Press Inc , Totowa,

21

Protocols NJ

22

L/mat and Hunzlker

hau follicles, the dlssoclation of the ORS cells from the folhcle, and the plating of the chssoclated ORS cells m a growth-supportive medium on a preformed feeder layer (14,2.5) Because the culture IS started with low cell numbers, a crucial point of the protocol for the primary cultlvatlon IS the use of feeder layers made of postrmtotlc human dermal fibroblasts (14). We have found that the use of postmltotlc human dermal fibroblasts Instead of the conventional 3T3-feeder system (16) has a number of advantages, for instance, a higher reproduclbihty m the preparation of the feeder layers, which can be stored for several weeks m the CO2 incubator or cryopreserved m liquid nitrogen wlthout loss of the growth-promoting properties for eplthehal cells (14,15)

2. Materials 2.1. Tissue-Culture

Facility

Most items needed to isolate and cultivate ORS cells from plucked hair follicles belong to the standard equipment of a cell-culture laboratory 1 2 3 4 5 6 7 8 9 10

A tissue-culture cabmet (preferentially with vertical an- flow) A stereomicroscope A humidified, carbon dloxlde (5% m air) incubator set at 37°C An Inverted. phase-contrast microscope (magmficatlon 100 and 200x) A bench-top centrifuge (e g , Heraeus Sepatech, Osterode, Germany) Fme tweezers (curved and rectilinear ones) Gross forceps Fine scissors (curved and rectllmear ones) Mmlscalpels (Opthalmlc Knife 45”, Alcon Surgical, Fort Worth, TX) 35-, 60-, and loo-mm tissue-culture and bacterlologlcal dishes (e g , Falcon, Becton Dlckmson, Bedford, MA) 11 Pasteur pipets

2.2. Tissue-Culture

Reagents and Solutions

Phosphate-buffered salme (PBS) (Dulbecco) with Ca*+ and Mg*+ (e g , Seromed L 18 13, Blochrom Berlin, Germany) PBS (Dulbecco) wlthout Ca*+ and Mg*+ (e g , Seromed L 182-01)

Rmsmg medium* DMEM (e g , Seromed F 0435) buffered with 0 25 mM HEPES, pH 7 2 (e g , Seromed L 1613) and containing 10% fetal calf serum (FCS) (e g , Seromed or Glbco, Life Technologies, Galthersburg, MD) and 40 U/mL pemclllm

and 40 pg/mL streptomycm (e g , both from Seromed A 22 13) Trypsm 0 1% (w/v) and EDTA 0 02% (w/v) m PBS (Dulbecco) without Ca2+ and Mg *+ Trypsm and EDTA purchased, for example, from Seromed (L 2133 and L 2 113, respectively) 0 02% (w/v) EDTA m PBS (Dulbecco) wlthout Ca*’ and Mg*’ 0 05% (w/v) Trypsln, 0 02% (w/v) EDTA m PBS (Dulbecco) without Ca2+

and Mg*+

Cultivation of Keratmocytes

23

7 Medmm for primary cultures of ORS cells (Z3,17) First, 1000 mL DMEM (e g , Seromed F 0435) and 1000 mL Ham’s F12 (e g , Seromed F 08 13) are prepared, the pH adjusted to 7 &7 3 if necessary, and sterthzed through a 0 2-pm filter 8 For the preparatton of 100 mL of culture medium, 75 mL DMEM and 25 mL Ham’s F 12 are mixed, and the followmg supplements (for preparation, see steps 1G-1 3) added a 1 mL adenme solution (Boehrmger Mannhelm, [Mannhelm, Germany] 102 067) b. 0.1 mL insulin (Sigma [St Louts, MO] I 5500) c 0 1 mL truodothyronme (Sigma T 2877) d 0.2 mL hydrocorttsone (Sigma H 4001) This formulation can be stored at 4°C for up to 4 wk 9 Portions of 100 mL of final culture medium are prepared by adding the followmg supplements a 1 mL glutamme (Seromed K 0280) b IO 1tL EGF (Stgma E 4127) c 0 1 mL choleratoxm (Sigma C 3012) d 1 mL pemctllm/streptomycm (Seromed A 22 13) e 1 mL fungizone (Ctbco 15290-026) f. 10 mL FCS (e g , Seromed or Gtbco, see Note 4) This final medmm has to be used wtthm 10 d 10 Solutton of adenme Dtssolve 182 mg adenme (Boehrmger Mannhelm 102 067) m 100 mL btdtsttlled water and 0 7 mL HCl lN, stirred until complete dtssolutton Filter sterilize and make aliquots of 1 2 mL, which are to be stored at -20°C 11 Solution of insulin: Dtssolve 50 mg msulm (Sigma 15500) m 10 mL 0 005NHCl Filter sterilize and make ahquots of 0 2 mL, which are to be stored at -20°C 12 Solutton of hydrocorttsone Dissolve 25 mg hydrocorttsone (Sigma H 4001) rn 5 mL ethanol, and make altquots of 0 5 mL, which are to be stored at -20°C (stock solution) Dilute 0.4 mL of the stock solutton to 10 mL with DMEM (e g , Seromed F 0435), filter stertltze this diluted solutton, and make altquots of 0.3 mL, whtch are to be stored at -20°C 13 Solution of truodothyronme Dtssolve 13 6 mg trnodothyronme (Sigma T 2877) m a mmtmal volume of 0 02N sodium hydroxide Brmg the volume to 100 mL with H,O, filter stertltze, and store at -20°C To prepare the final solution, add 40 uL of the concentrated solution to 1960 uL DMEM, make ahquots of 0 I5 mL, and store them at -20°C 14. Solution of choleratoxm Dtssolve 1 mg choleratoxm (Sigma C 30 12) in 1 18 mL of btdtstilled water and filter stertltze (stock solution). Dilute 0 1 mL of the stock solution to 10 mL with DMEM (e g , Seromed F 0435) contammg 10% FCS, and store at 4°C 15 Solution of EGF Dissolve 100 ug EGF (Sigma E 4127) m 1 mL DMEM (e g , Seromed F 0435) contammg 10% FCS, filter stertltze, and make altquots of 15 uL which are stored at -20°C 16 Media for the subculttvatton of ORS cells Keratmocyte growth medium (KGM) used for the subculttvatton of ORS cells on tissue-culture plastic IS based on the

Limat and Hunzlker formulation of MCDB 153 (18) and can be purchased from Clonettcs Corporatton (San Diego, CA) or from Promocell (Heidelberg, Germany) 17 Medium for the cultivation of the tibroblasts. DMEM (e g , Seromed F 0435) buffered with 3.7 mg/mL NaHCO, (e g , Seromed L 1703) and containing 10% FCS (e g , Seromed or Gibco), 10 U/mL pemcillm, and 10 pg/mL streptomycm 18 Mttomycm C (Sigma M 0503) A stock solution 1s prepared by dtssolvmg mttomycm C in PBS (Dulbecco) with Ca2+ and Mg2+ at a concentration of 100 yg/mL, which IS sterilized by filtration through a 0 2-urn filter Ahquots of 0.9 mL of stock solution are stored at -20°C For treatment of the fibroblasts, mttomycm C IS used at a final concentratton of 8 ug/mL (e.g., 0 8 mL stock solution/ 10 mL DMEM containing 10% FCS)

3. Methods 3.1. Plucking of Hair Follicles Usually, we isolate scalp hatr follrcles from the occipital region. Using the same protocol, folhcles from other anatomrcal sites, such as beard, leg, and genital region, can be isolated. Optimal recovery of the ORS tissue during the pluckmg procedure IS achieved by observing carefully the followmg protocol. 1 The harrs to be plucked are exposed by pulling up the adjacent hair A few number of hairs (maximally 34) are gripped with gross sterile forceps as close as possible to the skm surface. The hairs are pulled out by a Jerky movement made perpendicular to the skm surface (see Note 1) 2 The folhcular material is then directly collected mto a 60-mm bactertological dish containing 5 mL rinsing medium, by cutting with tine sterile sctssors The remammg distal keratmtzed hair shaft is discarded At least one follicle has to be prepared per final milliliter of culture medium 3 The follicles m the anagen phase (I e , growing phase of the hair cycle (19), mdtcated by the visible ORS tissue; see Fig. 1A [p 261 and Note 2) are selected under a dtssectmg mtcroscope (stereomtcroscope) and transferred mto a new 60mm bacteriological dish contammg 5 mL of rmsmg medium 4. We usually remove the bulbar part as well as the distal fifth of the folhcular length (correspondmg to the mfundibular part) using mmtscalpels, which ensures that the only ltvmg cell populatton in the remammg folhcle 1s constttuted by ORS cells (see Note 3) In Fig 1A, the sites where the cuts are applied are marked

by arrowheads The mtcropreparattve mampulatlon

under the stereomtcroscope

does not necessarily need to be performed m the sterile cabinet, but can be done in a clean place 5 The prepared follicles are rinsed four times by consecuttve transfers m 60-mm bacteriological dishes containing 5 mL of rmsmg medium

3.2. Isolation

of the ORS Cells from the Follicles

1. The folhcles are deposited mto an empty 35-mm bactertologtcal

dtsh m such a

way that they are m close vtcmity, though separated from each other This guar-

Cultivation of Keratinocytes

2

3

4

5

25

antees free access of the trypsm during the subsequent dtsaggregatton step Some restdual medium IS aspirated with a Pasteur ptpet The folltcles are covered by a mmlmal volume of trypsm (0 l%)/EDTA (0.02%) solution (a droplet tf the number of follrcles IS 30 pg/mL may be mhrbttory

References 1. FaJardo, L F (1989) The complextty of endothehal cells Am J Clzn Pathol 24 l-250 2 Pearson, J. D (199 1) Endothehal cell btology. Radzology 179, 9-14

92,

Human Umb~hcal Vein Endothelium

109

3. Moncada, S , Palmer, R M J , and Htggs, E A. (1991) Nitric oxtde phystology, pathophystology, and pharmacology. Pharmacol Rev 43, 109-l 42 4 Jaffe, E A , Nachman, R L , Becker, C. G , and Mmtck, C R. (1973) Culture of human endothehal cells derived from umbthcal veins. tdenttficatlon by morphologic and tmmunologrc crtterra J Clan Invest 52, 2745-2756 5 Jaffe, E. A (1984) Culture and tdenttficatron of large vessel endotheltal cells, m Biology ofEndotheha1 Cells (Jaffe, E A , ed ), Martmus NiJhoff, Boston, pp l-1 3 6 Morgan, D M L , Clover, J , and Pearson, J D (1988) Effects of synthetic polycattons on leucme mcorporatton, lactate dehydrogenase release, and morphology of human umbthcal vein endothelial cells J Cell Scl 91, 23 l-238 7 Hoyer, L M , de 10s Santos, R P , and Hoyer, J. R (1973) Antthaemophtltc factor antigen Locahsation m endothehal cells by tmmunofluorescent mtcroscopy J Clw Invest 52,2737-2744 8 Voyta, J C , Via, D P , Butterfield, C E , and Zetter, B R (1984) Identrficatron and tsolatron of endotheltal cells based on then increased uptake of acetylated low-density ltpoprotem J Cell B1o1 99, 2034-2040. 9 Cut, Y C , Tat, P -C , Gatter, K C , Mason, D Y , and Spry, C J F (1983) A vascular endotheltal cell antigen with restricted dtstrtbutton m human foetal, adult and malignant ttssues Immunology 49, 183-l 89 10 Schlmgemann, R 0, Dmglan, G M , Emets, J J , Blok, J , Warnaar, S 0, and Rutter, D J (1985) Monoclonal antibody PAL-E specific for endothelmm Lab Invest 52, 7 l-76 11 Schlmgemann, R 0 , Rtetveld, F J R , de Waal, R M W , Bradley, N J , Skene, A I , Davies, A J A , Greaves, M F , Denekamp, J , and Rmter, D J (1990) Leukocyte antigen CD34 IS expressed by a subset of cultured endotheltal cells and on endothehal ablummal mtcroprocesses m the tumor stroma Lab Invest 62,690-696 12 Newman, P J , Berndt, M C , Gorskt, J., White, G C , Lyman, S , Paddock, C , and Muller, W A (1990) PECAM- 1 (CD3 1) cloning and relatron to adhesion molecules of the lmmunoglobuhn gene superfamrly. Sczence(Wash DC) 247,12 191222 13 Johnston, G I, Cook, R G , and McEver, R P (1989) Cloning of GMP-140, a granule membrane protein of platelets and endothelium: sequence stmrlartty to proteins involved m cell adheston and mflammatton. Cell 56, 1033-1044. 14 Bevtlaqua, M P., Stengelm, S , Gtmbrone, M A , and Seed, B (1989) Endotheha1 leukocyte adhesion molecule 1 an mductble receptor for neutrophtls related to complement regulatory proteins and lectins Sczence (Wash DC) 243, 1160-l 165 15 Osborn, L , Hesston, C , Trzard, R , Vassallo, C , LuhowskyJ, S., Cht-Rosso, G , and Lobb, R (1989) Direct cloning of vascular cell adhesion molecule 1, a cytokme induced endothehal protein that bmds to lymphocytes Cell 59, 1203-l 2 11. 16. Mactag, T , Cerundolo, J , Ilsley, S , Kelley, P R , and Forand, R (1979) An endothehal cell growth factor from bovine hypothalamus* tdenttticatton and partial charactertsatton Proc Nat1 Acad SIX USA 76, 5674-5678. 17. Maciag, T , Hoover, G A , Stevenson, M B , and Weinstein, R (1984) Factors whtch stimulate the growth of human umbtltcal vein endotheltal cells In vztro, in Bzology ofEndothelra1 Cells (Jaffe, E A , ed ), Martmus NtJhoff, Boston, pp 87-96

Human Thymic Epithelial Anne

Cell Cultures

H. M. Galy

1. Introduction The thymus is a very complex organ that regulates T-cell production Thymocytes (immature T-cells) constitute by far the largest cellular population m the organ (several billions of thymocytes m a child’s thymus), but small numbers of other hematopoietic cells are found m the mtrathymtc mtcroenvtronment, such as thymic monocytes, macrophages, rnterdigitatmg dendrmc cells, and R-cells Stromal cells of nonhematopoiettc origm comprise thymic epitheha1 cells (TEC) organized m a network throughout the organ, thymic fibioblasts of the capsule and mterlobular septae, and endothelial cells of the thymtc vasculature The proportion of these thymrc cellular components varies wtth age (2). Fat mfiltration becomes stgmficant at puberty and increases throughout adulthood. Intrathymtc T-cell maturation IS supported m part by TEC, which express cell-surface molecules interacting with counterreceptors on the maturing thymocytes (24) Importantly, TEC induce positive selection and major histocompatibihty complex restriction of T-cells (5) In vitro studies have shown that TEC produce numerous cytokmes (4,6,7), which may directly and/or mdtrectly contribute to T-cell maturation. Rectprocally, TEC functions are affected by mteracttons with T-cells (4,s). Monolayer cultures of TEC provide m vitro systems to study the biology of TEC (2-7) However, mvestigators should be aware that monolayer cultures may not be representative of a threedimenstonally structured TEC network, and the function of isolated cells may differ from that of cells m the midst of a complex envu-onment One method to obtain highly purified monolayer cultures of human TEC 1s described here Nontransformed human TEC can be propagated and passaged up to SIX times before cells become senescent. Careful “budgetmg” of the cell stocks by freezmg early passages can provide a long-lastmg supply of purified TEC strains with determined purity From

Methods Edlted

m Molecular by

G E Jones

Me&one Humana

111

Human Press

Cell Culture Inc , Totowa.

Protocols NJ

112

1.1. Principles

of the Method

The method to rsolate and grow TEC relies on cell density, cell-adhesion properttes, and modulatton of cell growth by appropriate factors (Note 1) Stroma1 cells need to be released from the trssue by mechanical or enzymatic dlsruptton to establish a TEC culture A large number of thymocytes are removed by densrty sedrmentatron prior to culturmg. Stromal cells (mostly thymtc eptthelial cells and thymtc fibroblasts) are separated from the lesser or nonadherent hematoporettc cells (mostly thymocytes, thymrc monocytes, and mterdtgttatmg dendrrttc cells) by culture. The cell-culture medium favors TEC expanston, partrcularly because of its high serum concentratron and the presence of eptdermal cell growth factor (EGF), a mitogen for TEC (9,lO). Hydrocorttsone (OHC) and cholera toxin (CT) further improve TEC prohferatlon, and OHC

promotes apoptosls of residual cortical thymocytes (Note 2) Thymlc stromal cell cultures tend to be overgrown by highly prohferatmg thymlc fibroblasts, and special treatment

of the cultures with EDTA

1s required to remove fibro-

blasts (II) (Note 3). The purity of TEC cultures can be measured by lmmunofluorescent staining for mtracellular keratm, which IS expressed exclusively m eplthehal cells, but not m fibroblasts or hematopoletlc cells (12) 2. Materials I Tissues Human thyme are obtained ethically m compliance with the appropriate regulations Two common sources of thymic specimens are fetal thyme collected from electively aborted fetuses and postnatal thyme obtained from patients (generally chtldren) undergoing cardrac surgery durmg whrch a portion of the thymus is removed Thymic specimens are aseptically placed m a tube contammg culture medium wrth antibiotics and transported to the laboratory, preferably within 24 h of collection Human tissues are biohazardous material and should be handled with precaution In particular, the mvestigator should wear a labcoat and gloves at all times, dispose of waste m approprrate containers, and manipulate the tissue m a certified biological safety cabinet (Biosafety Level 2) Hepatitis B vaccmation is recommended 2 RPM1 1640 3 Phosphate-buffered salme (PBS) without calcmm and magnesmm. 4 Dimethyl sulfoxide (DMSO) 5. Heat-macttvated (56°C for 30 mm) fetal bovme serum (FBS), preferably trrple 0 1 pm filtered to ensure that it is mycoplasma-free Once machvated, aliquot m 30-mL fractions m 50-mL comcal tubes, and store frozen at -20°C 6 Versene solution (phosphate-buffered salt solution contammg 0 02% EDTA) 7 Collagenase type IAS (Sigma, St Louis, MO). Reconstitute m RPM1 1640 at 5 mg/mL (or 2000 U/mL) Ahquot and store at -20°C. 8 Deoxyrtbonuclease (DNase) type II-S (Sigma) Reconstitute m RPM1 1640 at 1 mg/mL and store at -20°C

113

Thymic Epithellal Cell Cultures

9 Trypsm 10X or 0.5% (Glbco, Grand Island, NY) mycoplasma and parvovirus screened, frozen. Thaw the content of the bottle, and ahquot the 10X trypsm m IO-mL fractions that will be frozen again at -20°C When needed, an altquot of 10X trypsm will be thawed and diluted m Versene Thawed 1OX trypsm allquots can be kept at 4’C for 1 wk 10 Medium Prepared by mixing equal volumes of DMEM (with 4 5 g/L glucose, L-glutamme, and 10 mM HEPES) and HAM F12 (with L-glutamine and 25 mA4 HEPES) supplemented with 15% FBS, 50 U/mL and 50 pg/mL, respectively of pemcillm and streptomycm antibtotics (P/S), 1 mA4 sodium pyruvate, 4 mM additional L-glutamme, 0.4 ug/mL OHC, 12 5 ng/mL sterile, endotoxm-tested tissue-culture-grade EGF from mouse submaxtllary glands, and 10 ng/mL cholera toxin (CT) The DMEM, HAMF12, and pyruvate solutton can be mtxed and stored for months at 4°C Antibtotics, L-glutamme, FBS, OHC, EGF, and CT are stored m small altquots at-20°C and added freshly to TEC medium, which is subsequently kept no more than 3 wk at 4°C A frozen stock of OHC-EGF-CT supplement can be prepared as follows To 16 5 mL of FBS, add 0 7 mL of an OHC solution at 5 mg/mL m ethanol, add 100 pg of EGF, and add 75 uL of CT solutton prepared at 1 mg/mL m endotoxm-free bovine serum albumin (BSA) (BSA at 5 mg/mL m water) Ahquot this sterile supplement m 0 5-mL fractions, and store frozen at -20°C for several months to a year Use 0.5 mL of supplement for 230 mL of medium (I e ,200 mL of DMEM-HAMF 12 medium with 30 mL of FBS) 11 Eight-well Labtek chamber slide (acetone-resistant Labtek chambers, Nunc, Naperville, IL) 12 Acetone 13 Tween-80 (Sigma) 14 Mouse monoclonal antibody (MAb) anttkeratm KLl (Amac, Westbrook, ME) 15 Irrelevant mouse IgGl tsotype-matched negative control (MOPC 2 1, Sigma) 16 Fluorescem-conlugated polyclonal goat antimouse tmmunoglobulm (GAMFITC) (TAGO, Burlmgame, CA) 17 Coplm Jars. 18 Evans Blue 0 5% solution m PBS (Sigma) 19 Fluorescence mounting medium (Dako, Carpmterta, CA).

3. Methods Unless otherwise indicated, all procedures are performed under sterile conditions (Biosafety level 2). Volumes of medium and times of incubation are given for a medium-sized thymus specimen of approx 1 x 2 x 2 cm 3.1. Dissection

and Digestion

of Thymic Tissue

1 With forceps, extract thymus from the transport tube, and place into a 100 x 20-mm Petri dish with 5 mL of cold RPM1 1640 medium containing 50 ug/mL DNase

Galy

114

2 Carefully peel off and cut away the conJunctlve capsule with fine stamless-steel forceps and scissors Hemorrhagic areas, blood vessels, and nonthymlc tissues are also carefully excluded (Note 4) 3 Cut the loosened thymlc lobular structure m small cubes (approx 1 x 1 mm or smaller) usmg 4-ln -long curved mlcrosclssors This takes about 5 mm of repeated cutting and IS necessary to facllttate the enzymatic dlssoclatlon 4 Add RPMI-DNase if necessary 5 Plpet the resultmg cell mixture mto a SO-mL conical tube usmg RPMI-DNase to transfer cells remammg m the Petri dish 6 Fill tube up to 45 mL with RPMI-DNase 7 Decant the cell suspension at lg for 5 mm at room temperature Fragments should settle to the bottom of the tube 8 Discard thymocytes remaining m the upper 40 mL 9 Repeat the procedure at least three times or until the supernatant becomes clearer (complete ehmmatlon of thymocytes IS not possible) (Note 5) IO Resuspend the washed fragments m 20 mL of RPMI-DNase II Transfer fragments to a sterilized IOO-mL glass bottle contammg a magnetic stlrrmg bar 12 Add collagenase to the final concentration of 200 U/mL (or 500 pg/mL) 13 Place the bottle mto a 37°C water bath, and gently stir its contents for 2&30 mm This can be achieved sunply by placing the bottle into a water-containing glass beaker placed on a magnetic hot plate Temperature IS carefully equlllbrated and controlled with a thermometer 14 Plpet the released cells (in suspension or m the form of very small aggregates), and transfer to a 50-mL conical tube contammg cold RPM1 and 50% FBS kept on Ice 1.5 Add fresh collagenase and DNase to the rest of the tissue, and pursue digestion for another 20 mm 16 Repeat the procedure once At the end of the third dlgestlon, most of the tissue should be reduced to cell suspension 17 Collect cells m RPM1 + FBS and spm down gently (5 mm, 200g at 4°C) Cells m the pellet still contam thymocytes, but this cell preparation can be placed m culture at this point Further enrichment m stromal cells 1s achieved by layering the cells over an FBS cushion 18 Resuspend cells m 20 mL of cold RPM1 + DNase, and layer 2 mL of this mixture onto 3 mL of cold FBS m 15-mL comcal tubes that are kept undisturbed at 4°C for 40 mm 19 After decantatlon, carefully discard the top half fraction and collect the bottom fraction Stammg the cells for CD45 antigen expression (a marker of leukocytes that IS not expressed on nonhematoporetlc stroma [3/) shows significant enrichment m CD45- stromal cells m the bottom fraction (Galy, unpubhshed observations)

3.2. Culture and Purification

of TEC

1 Resuspend isolated cells m TEC medium and plate mto three 75-cm2 tlssue-culture flasks Incubated m a humidified atmosphere of 95% air, 5% CO, at 37°C for

Thymlc

2

3 4

5

6 7

8

9 10 11

Eprthel/al

Cell Cultures

115

3 d without dlsturbmg the flasks The resultmg culture should be relatively dense and Ideally would consist of platmg 0 5-l x lo5 stromal cells/75 cm2 After 3 d, wash nonadherent cells away by removing the medmm, and gently rmsmg the flask with 10 mL of RPM1 before adding fresh TEC medmm Examlnatlon of the flasks under a phase-contrast microscope shows a growmg stromal monolayer and probably very small adherent explants crowned by emerging TEC Ehminate the few remammg thymocytes by further washing and medium change, twice a week Examme the culture After about l-2 wk, large stromal areas are visible m the culture TEC are recogmzed by their polygonal morphology and fine lntracytoplasmlc permuclear granulations (Ftg 1A). As seen m Fig 1B, thymlc fibroblasts are spindle-shaped cells often oriented similarly to adJacent cells (Note 5) Thymlc monocytes and macrophages are generally round cells with a central nucleus and an Irregular outer membrane projectmg multiple pseudopodes Thymocytes are small lymphold cells that appear refractive since they are mostly nonadherent, but can also be bound to stromal cells Thymlc fibroblasts, thymlc monocytes, and thymocytes are less adherent to the culture flask than TEC and are therefore preferentially detached on treatment of the cultures with an EDTA solution (Versene) Treat cultures with EDTA as soon as cell growth IS well estabhshed and areas of thymlc fibroblasts develop The treatment will be repeated as often as needed to obtam a pure TEC culture To treat the cultures, remove the medmm, and gently wash the flasks twice with 5 mL of PBS and 5 mL of Versene Flush fibroblasts by forcefully proJectmg 5 mL of Versene (at room temperature) at a 45” angle on the cells with a cotton-plugged glass Pasteur plpet and a handoperated bulb Repeat this procedure 5-10 tnnes to flush the entire surface of the flask Concentrate on areas where tibroblasts were seen growing Collect and discard the detached cells Repeat the procedure with another 5 mL of fresh Versene This drastic procedure ~111 remove a lot of cells, mcludmg some TEC Therefore, the cultures ~111 be significantly depleted However, areas of highly adherent pure TEC should be spared and ~111 grow back Wash the flask gently with 5 mL of TEC medmm Discard Add 15 mL of TEC medium and place the flask back m the mcubator Observe cultures under the microscope for appearance of a confluent TEC monolayer wlthm 2 wk Repeat the EDTA washmg procedure If thymlc fibroblasts grow agam

3.3. Passage of TEC Cultures Confluent monolayers of TEC are passaged with trypsm EDTA. 1 Remove medmm from the flask, and wash cells twice with approx 10 mL of Versene gently spread over the entlre surface of culture Make sure all serumcontaming medium 1s removed, smce It ~111 mhlblt the actlon of trypsm

116

Galy

Fig. 1. (A) Phase-contrast photomicrograph (objective 10x) of a culture showing a highly pure TEC area. Note the proliferative cobblestone-shaped small TEC and more senescent large flattened TEC. (B) Phase-contrast photomicrograph (objective 10x) of a culture largely overgrown by thymic fibroblasts that are spindle-shaped cells organized unidirectionally. An area of TEC is seen on the left side of the photograph.

2. Add 10 mL of a 1X trypsin solution (freshly prepared by 1: 10 dilution of the 10X stock into Versene) onto the cell monolayer to cover the entire surface for about 30 s.

Thymlc

Epithellal

Cell Cultures

117

3 Remove trypsm, and repeat the procedure once Remove excess trypsm agam leavmg only a liqurd film over the cells. 4 Return flask to 37°C for 5-7 mm of mcubatron, or until all cells loosen from plasttc 5 Bang flask once on Its large side m order to detach all cells 6 Harvest TEC wtth 1O-20 mL of cold TEC medium, and spin cells down (7 mm at 400g at 4 “C) 7 Plate the resulting pellet at approx 0 5-l x IO6 cells/75 cm* flask 8 Carefully record the number of passages, smce TEC will lose then prohferattve potential after the Sh or 6th passage.

3.4. Freezing TEC Cultures The supply of TEC can be extended

and expandmg

by freezing

alrquots

of passaged

cells

only a fraction of the cultured cells.

1 Detach cells as Indicated m Sectton 3 3 2 Resuspend TEC at 2 x 10h cells/ml m cold TEC medmm 3 Add slowly an equal volume of cold DMSO solution (20% DMSO In TEC medium) to obtam a final cell suspension of 1 x IO6 TEC/mL m 10% DMSO solutron. 4 Dispense 1 mL of the cold cell-DMSO mixture m cold cryovtals 5 Freeze m controlled-freeze apparatus or more simply m Styrofoam racks placed mstde a Styrofoam box at -80°C overnight. When frozen, the vials are transferred and stored mdefimtely m hqutd nitrogen.

3.5. lmmunofluorescence of Keratin-Positive Cells

Measurement

The proportion of cells expressing intracytoplasmlc cytokeratm filaments can be measured by immunostammg and serves to identify TEC (Note 6). 1 Detach cells as mdtcated m Section 3 3 2 Plate 2 x 1O4 TEC m 200 pL of TEC medmm/well of an acetone-resistant etghtwell Labtek chamber slide Distribute cells m at least two wells (one for negative tsotype control, the other for keratm staining) 3. Place chamber slides in incubator at 37°C for adherence overnight 4 Remove medium. At this point, the procedure does not need to be carried out under stertle condtttons 5 Wash cells m the Labtek chamber three times with warm (37’C) PBS, which IS gently ptpeted m and out of the chamber well 6 Blot excess liquid after the last wash 7 Add acetone (approx 400 uL) m each well to fix the cells and permeabiltze the cellular membranes for 10 mm 8 Remove acetone, and discarded in waste contamers for volattle solvents 9 Au-dry chambers for 30 mm, and then store at -20°C for no more than 3 wk 10 For immunostainmg, rehydrate slide wtth approx 500 pL of PBS/well for 30 mm

Galy

118

11 Discard PBS and add 250 PL of the antlkeratm antibody or of the negative control each at the concentration of 20 pg/mL m PBS with 0 04% tween for 40-mm mcubatlon m a humldlfied chamber at room temperature 12 Discard antlbodles, and detach the upper part of the chamber 13 Wash the slide m a coplln Jar filled with PBS-tween 14 Repeat procedure twice for S-mm-long washes each time 15 Incubate slides m the dark for 40 mm with approx 250 pL/well of the secondary goat antimouse FITC reagent diluted 1 40 m PBS-tween, and carefully dlstnbuted over the entire surface of the chamber 16 Wash shdes twice as described m step 14, and a third time m Evans Blue counterstammg solution (0 01% solution m PBS-tween) Protect slides from light 17 Blot excess llquld, and mount slides with fluorescence mountmg medium 18 Examme slides under eplfluorescence mlcroscopy

4. Notes There seems to be ontogeny-related differences m the prohferatlve potential of thymlc stromal cells Fetal thyme or young postnatal specimen (less than a year old) will grow better than older thyme An improved medium formulation can be obtamed by addmg 1% of serum supplement LPSR- 1 (Sigma) to TEC medium However, this product 1sno longer available for import m the United States A bovine pituitary extract (Collaborative Research, Lexmgton, MA) IS mltogemc for TEC at 50 pg/mL Thymlc fibroblasts are thought to contrlbute slgmficantly to early T-cell development (13) Thymlc fibroblast cultures can be easily obtained after dIgestIon of the capsule and culture m TEC medium The nature of the culture can be assessed by its keratm negativity and lack of CD45 expresslon Careful mltlal dlssectlon of the thymlc capsule 1s very Important to reduce the chance of overgrowth by thymlc fibroblasts Fetal thyme can be placed m culture after dlssectlon mto very small fragments and thymocyte wash without enzymatic dlgestlon In that case, explants ~111 attach and TEC will grow out of the explants after 3-5 d, and the monolayer will extend rapidly to cover the surface of the dish by d 8 (plate one fetal thymus of 1 5 x 1 x 1 cm mto one 75-cm2 flask) Subsequently, the culture ~111 be treated as described m Section 3 2 , steps 5-l 1 to remove fibroblasts if needed It IS possible to analyze the mtracellular keratm content of cultured TEC by flow cytometry Trypsmlzed cell suspensions are washed m PBS, and the loosened pellet is fixed in 1 mL of cold methanol for 10 mm. About 10 mL of PBS are added to dilute the methanol, and cells are spun down and washed m PBS twice Washed cells are mununostamed with an antlkeratm antibody (CAM-5 2) specific for keratm 8,18 (Becton Dlckmson, Mountain View, CA), which gives better results on the fluorescence-actwated cell scanner (FACscan) than KLl The antlbody IS diluted at 10 pg/mL m PBS with 0 2% BSA Cells are then washed and incubated with the secondary antibody (GAM-FITC 1 40 m PBS-BSA) and washed twice Cells are analyzed on the FACscan, adjusting setting for large cells

Thymic Eplthellal Cell Cultures

119

References 1 Nakahama, M , Mohrt, N , Morr, S., Shmdo, G , Yokot, Y , and Machmamt, R (1990) Immunohrstochemtcal and histometrtcal studtes of the human thymus with special emphasis on age-related changes m medullary eptthehal and dendrrttc cells Vzrchows Arch B Cell Path01 58,245-25 1 2 Smger, K H , Dennmg, S M , Whtchard, L P , and Haynes, B F (1990) Thymocyte LFA- 1 and thymtc eprtheltal cell ICAM- 1 molecules mediate bmdmg of activated human thymocytes to thymtc epttheltal cells J Zmmunol 144, 293 1 3 Galy, A H M and Spits, H (1992) CD40 IS functtonally expressed on human thymic eptthehal cells J Immunol 149, 775-782 4 Galy, A H M and Spits, H (1991) IL-l, IL-4 and IFN-?/ differentially regulate cytokme production and cell surface molecule expression m cultured human thymtc eptthehal cells J Immunol 147, 3823-3830 5 Martin Foncheca, A, Schuurman, H J , and Zapata, A (1994) Role of thymic stromal cells m thymocyte educatron. a comparative analysts of different models Thymus 22,20 1-2 13 6 Galy, A H. M , de Waal Malefyt, R., Barcena, A , Mohan-Peterson, S , and Sptts, H (1993) Untransfected and SV40-transfected fetal and postnatal human thymrc stromal cells analysrs of phenotype, cytokme gene expression and cytokme production Thymus 22, 13-33 7 Galy, A H M , Spits, H , and Ham&on, J A (1993) Regulatron of M-CSF production by cultured human thymrc eptthehal cells. Lymphokzne Cytokzne Res 12, 265-270 8 Surh, C D , Ernst, B , and Sprent, J (1992) Growth of eptthehal cells m the thymic medulla 1s under the control of mature T cells J Exp A4ed 176, 6 116 16 9 Sun, L Serrero, G , Ptltch, A , and Hayasht, J (1987) EGF receptors on TEA3AI endocrine thymic eptthehal cells Blochem Bzophys Res Commun 148,603-608. 10 Hadden, J W., Galy, A , Chen, H , and Hadden, E. M (1989) A pmntary factor Induces thymtc eptthehal cell prohferatton m vitro Brazn Behav Zmmun 3, 149- 159 11 Singer, K H , Harden, E A, Robertson, A L , Lobach, D F , and Haynes, B F (1985) In vitro growth and phenotypic charactertzatton of mesodermal-dertved and epnhehal components of normal and abnormal thymus Hum immunol 13,16 l-l 76 12. Moll, R , Franke, W W , Schtller, D L., Getger, B., and Krepler R (1982) The catalog of human cytokeratms patterns of expression m normal epitheha, tumors and cultured cells Cell 31, 11 13. Anderson, G , Jenkmson, E J , Moore, N. C., and Owen, J J T (1993) MHC class II postttve eptthelmm and mesenchyme cells are both requned for T-cell development m the thymus Nature 362,70-73

Generation and Cloning of Antigen-Specific Human T-Cells Hans Yssel 1. Introduction The ability to grow antigen-spectfic human T-cell clones m vitro has been mstrumental m understanding T-cell function. A major breakthrough m T-cell culture m vitro was the discovery of the T-cell growth-mducmg properties of mterleukm-2 (IL-2), ortgmally called T-cell growth factor or TCGF, by Morgan et al (I), who for the first time were able to grow human T-lymphocytes, isolated from human bone marrow. Shortly thereafter, Gtlhs and Smith (2) reported the clomng and long-term culture of mouse cytotoxic T-cells, using TCGF. In sprte of the successof growmg human T-cells m vitro, the clonmg of these cells, however, turned out to be more difficult m contrast to mouse cell lures that could be maintained m culture m the presence of TCGF-contammg supernatant only, long-term cultures of cloned human allo-anttgen-specific T-cell lines (3,4) needed repetitive stimulation with specific allo-antigen for then growth (3) This led to the general assumption that antigen-specific IL-2dependent T-cell lines and T-cell clones would lose then antigen responsiveness when propagated with IL-2 m the absence of specttic antigen. In view of the presumed requirements for anttgemc stimulatton, a culture system was devised for the generation and expansion of stable allo-anttgenspecific cloned human T-cell lures At the begmning of each culture, T-cells are stimulated with a feeder cell mixture, conststmg of n-radiated peripheral blood mononuclear cells (PBMC), an Epstein-Barr virus-transformed lymphoblastoid cell lme (EBV-LCL), expressing the specific allo-antigen, and phytohemagglutmm (PHA) The use of this feeder cell mixture was found to improve clonmg effictencies and growth rates of allo-antigen-specific T-cell clones dramatically (5-7). Between repetitive resttmulations with feeder cells, T-cell clones were expanded m medium supplemented with TCGF and later, when it From

Methods Edlted

m Molecular by

G E Jones

Medrone Humana

121

Human Cell Culture

Protocols

Press

NJ

Inc , Totowa.

122

Yssel

became available, recombinant (r)IL-2 Interestmgly, It was found that T-cell clones, specific for various other antigens, could be cultured and propagated m this feeder cell mixture as well, m the absence of spectfic antigen Using this polyclonal sttmulatton protocol with feeder cells, tt was possible to generate successfully stable T-cell clones specific for tumor antigens (8,9), and recall antigens, such as tetanus toxoid (IO), bacterial antigens (I I, 12), and allergens (13) successfully, as well as T-cell receptor (TCR) y6+ T-cell clones (14) Therefore, m our culture system, the presence of antigen does not seem to be a prerequisite to mamtam antigen specrtictty and growth properties of anttgenspecific T-cells, and m this respect, our methodology to grow antigen-specific T-cell clones differs from methods reported by others who favor the presence of spectfic antigen (1.5,16) The mechamsm by which the feeder cell mixture exerts its growth-promotmg effects IS not clear. It IS noteworthy to mention that the addition of t-IL-2 alone only induces a transient acttvatlon of the T-cells, mdicatmg that this growth factor IS not able to promote long-term growth and proltferation by itself Thus 1s most likely because of the mabthty of IL-2 to induce a sustained expression of a high-affimty IL-2R Apparently, PHA can replace the requnement of antigen-mediated triggermg of the TCR for T-cell activation by mducmg the expression of the IL-2Ra chain (CD25) on the T-cells. PBMC, as well as EBV-LCL, are hkely to function as accessory cells, the presence of whtch IS needed for this mitogen to be effective In addition, PBMC will produce growth and costimulatory factors, followmg activation with PHA or EBV-LCL, which will further enhance the proltferatton of the T-cell clones. Each sttmulation of T-cell clones with feeder cells IS followed by a growth factor-dependent expanston phase, during which the growing T-ceils are expanded wtth exogenous IL-2 or other growth factors Smce this culture method IS effective m expandmg T-cell clones of vartous ortgms and spectfictttes, one does not have to worry about the avatlabthty of the relevant antigen-presenting cells (APC) Generally, foreign anttgens are processed by APC and presented on their cell surface as small linear protem fragments, where they are recogntzed by the TCR m assoctatton with autologous major hlstocompattbihty complex (MHC) molecules to which they are bound Whereas autologous MHC molecules function as a restrtction element in the recognition of antigen, foreign (mismatched) MHC itself can function as allo-antigen for T-cells as well. However, there are no dtfferences m the way the TCR and additional accessory molecules on the T-cells, such as CD4 and CD8, interact with MHC molecules, whether or not they carry foreign anttgen. CD4+T-cells recognize antigen m the context ofan MHC class II molecule, whereas the recognition of anttgen by CD8’ T-cells IS restricted by MHC class I molecules In the same vein, allospectfic CD4+ and CD8+ T-cells directly recogmze MHC Class II and class I molecules, respectively

Ant/gemSpeak

T-Cells

123

In the followmg sections, the generation of allo-antlgen- and (soluble) ant]gen-specific T-cell clones are described There are no differences m methodology for the generation of each type of cell. As mentioned, depending on the interaction with, or recognition of MHC class II or class I antigens, allo-antlgen-specific cells are CD4” or CDX+, respectively Antigen-specific T-cell clones can be identified based on their capacity to either kill specific target cells, to proliferate, or to secrete cytokmes, followmg antigen-specific stlmulatlon Once antigen-specific T-cell clones have been selected they can be cultured for extended periods of time, while mamtammg their antigen-specific properties, using the polyclonal stlmulatlon protocol, described m this chapter (see also Fig 1) 2. Materials

2.1. Culture Medium All cell cultures are grown m Yssel’s medium (27), which was orlgmally described as a serum-free medium and 1s based on a modification of Iscove’s Modified Dulbecco’s Medium (IMDM, contammg L-glutamme, 25 mMHEPES, but no a-thlogycerol, Glbco-BRL, Grand Island, NY, cat no 21056-015) (see Note I). To prepare Yssel’s medium, dissolve freshly m the followmg order m IMDM* 1 Bovme serum albumin (BSA) (Sigma, St Louis, MO, A2 153), at a final concentratlon of 0 25% (w/v) 2 2-Ammo ethanol (Sigma, E0135), at a final concentration of 1 8 pg/L 3 Transferrm (Ho10 form, Boehrmger, Indlanapohs, IN, cat no 13 17 423), at a final concentration of 40 mg/L An excellent source IS 30% saturated transferrm (652-202) m liquid form from Boehnnger, to be diluted according to the manufacturer’s mstructions (see Note 2)

4. Insulin (Sigma, 15500) at a final concentration of 5 mg/L (make stock solution of 1 mg/mL in 0 0 1M HCl and add to medium)

5 Lmolelc acid (Sigma, L 1376) and olelc acid (Sigma; 03879), both at a final concentration of 2 mg/L Make stock solution In ethanol (20 mg/mL) and add to medium (see Note 3) 6 Palmltlc acid (Sigma, P59 17), at a final concentration of 2 mg/L (stock solution [20 mg/mL] in ethanol can be stored at 4’C for mo) 7 Pemctllm/streptomycm (Gbco-BRL, cat no 59-60277P)

Ftltrate through 0.2~pm filter, ahquot, and store at -20 or -80°C 2.2. Cells and Reagents 1. PBMC, isolated from the peripheral blood of healthy volunteer donors by centrlfugatlon over fycoll hypaque (18), to be used as feeder cells for the culture of T-cell clones Keep cells at 4°C on ice prior to use

DAY 0 Primary stimulation

(1) Incubate lo6 PBMClml with antigen

DAY 40-45 Selection and expansion of T cell clones

DAY 5

(2) [3H]TdR assay (Assessment of antigen concentration)

DAY lo-14 Secondary stimulation

(3) Incubate

106 PBMC/ml

with

I 06 lrradlated autologous PBMC and antigen

DAY 20

DAY 3094 Transfer of T cell clones

Cloning, transfer and expansion of T cells IS carned out with Irradiated, allogeneic PBMCArradlated EBV-LCL and PHA, as described In section (3.3 ) and (3.4 )

(4) Harvest Cek wash and resuspend at 1 O5 cells/ml

limiting dilution

lo’

Fig. 1. GeneratIon and clonmg of antigen-spechic T-cells

lo2 lo3 cells/ml

Antigen-Specific

T-Cells

125

2 Pertpheral blood or tissue samples from healthy donors or patients to be used for the generation of T-cell lures 3 EBV-LCL, generated by mfection of B-cells with EBV (see Section 3 1 ) 4 EBV-producmg marmoset cell lme B98 5. (EBV-contammg supernatant 1s produced by growing B98.5 cells until medium is completely exhausted After centrifugation for 10 mm at 2000 rpm to remove residual B98.5 cells and filtration through a 45urn filter, virus-containing supernatant can be stored at -80°C for years ) 5 Antigen (see Note 4) 6 PHA (Wellcome, Beckenham, UK, HA-16 or HA-17) 7 Phosphate-buffered salme (PBS), supplemented with 1% human serum (HS) 8. Tissue-culture plates (24-well flat-bottom and 96-well round- and flat-bottom Lmbro plates, Flow Lab, McLean, VA) 9 Recombinant IL-2 (IL-4, IL-7, IL-12, IL-15) 10. Cyclosporm-A@ (Sandoz, East Hanover, NJ) 11 Phorbol ester (PMA or TPA, Calbiochem, La Jolla, CA; cat no 524400) 12 Calcium ionophore (Calbiochem, cat no 100105) 13. Anti-CD3 and CD28 monoclonal antibodies (MAbs). 14 Neutrahzmg ant+HLA-DR, DQ, and DP MAbs. 15 Glutaraldehyde grade I (Sigma, G5882) 16 Dimethyl sulfoxide (DMSO) (Sigma, D5879) 17. [3H] TdR (Amersham, Arlington Heights, IL) 18 5’Cr (Amersham)

3. Methods

3.1. EB V- Transformed

B-Cell Lines

1. Isolate PBMC by centrtfugation over tic011 hypaque. 2 Incubate 2 x 106-10’ PBMC/mL in a 15-mL centrifuge tube for 2 h at 37’C in 1 or 2 mL of EBV-contammg culture supernatant 3 Wash the cells twice with PBS/l% HS, and incubate m culture medium at a concentration of lo5 cells/mL m the presence of 1 pg/mL Cyclosporm@ m a 96-well tissue-culture plate at a final volume of 100 yL 4 Add 100 pL of fresh culture medium after 5 d 5 After 10-14 d, growing cultures of blastoid cells can be observed, which can be split if necessary and transferred to 24-well plates or tissue-culture flasks Cells should be mamtamed at concentrations between 2 x 1O5and 5 x 1O5cells/ml, and cultures should be split with fresh culture medium when they reach concentrations >106 to 2 x lo6 cell/mL

3.2. Generation

of Al/o-Antigen-Specific

T-Cells

1. Stimulate lo6 PBMC with 2 x lo5 Irradiated (5000 rad) HLA-mismatched EBVLCL m a total volume of 1 mL m a 24-well Lmbro plate (see Note 5) 2. Culture the cells for 10 d at 37°C 5% CO, without addition of medium or growth factors

Yssel

126

3 To evaluate the prohferatlve activity of the respondmg T-cells, transfer 100 pL of cells, m duplicate or triplicate, mto a 96-well plate at d 5 of culture, and pulse with 1 ,LKI (=37 kBq) [3H]TdR for 4 h Harvest cells onto a glass fiber filter, and determine amount of incorporated [3H]TdR by liquid scmtlllatlon counting 4 At d 10 of culture, collect the T-cells, wash once with medium, and restimulate lo6 T-cells with 2 x lo5 lrradlated (5000 rad) EBV-LCL, used m the first stlmulation, in a final volume of 1 mL in a 24-well tissue-culture plate 5 Clone the cells by hmltmg dilution between d 5 and 7 after restlmulatlon (see Sections 3 4 and 4 2 and Note 6)

3.3. Generation

of (Soluble)

Antigen-Specific

T-Cells

1 Incubate freshly isolated PBMC from mununlzed donors or patients at a concentration of 1O6 cells/ml m a 24-well Lmbro plate, and add soluble antigen at dlfferent concentrations (I e , 0 1, 1, and 10 pg/mL) (see Notes 7 and 8) 2 Culture for 10 d at 37”C, 5% CO2 without addition of medium or growth factors 3 Evaluate the proliferative activity of the responding T-cells by transferrmg 100 mL of cells into a 96-well plate at d 5 of culture and pulsing with 1 pC1 (=37 kBq) [3H]TdR for 4 h H arvest cells onto a glass fiber filter, and determine amount of incorporated [3H]TdR by liquid scintillation counting 4 Collect the T-cells at d 10 of culture, wash once with medium, and restimulate 1O6 T-cells with an optimal (based on the prohferatlon results of 3 2 3) concentration of antigen, m the presence of lo6 irradiated (4000 rad) autologous PBMC, m a final volume of 1 mL m a 24-well Lmbro plate (see Notes 9 and 10) 5 Clone the cells by hmltmg dilution between d 5 and 7 after restnnulatlon (see Sections 3 4 and 4 2 and Note 6)

3.4. Cloning

by Limiting

Dilution

Prepare a clonmg feeder cell mixture by adding together m a 50-mL centrifuge tube 5 x lo5 cells/ml of irradiated (4000 rad) PBMC from any healthy donor, 5 x 1O4cells/ml of an irradiated (5000 rad) EBV-LCL, and 50 ng/mL PHA Brmg the cloning feeder cell mixture at 37”C, and use munedlately Collect T-cells (obtained from Sections 3 2 or 3 3 ), wash once, and resuspend m culture medium at a concentration of lo5 cells/ml Make serial dilutions of 104, 103, 102, and 10 T-cells/mL with the feeder cell mixture as dlluent Starting from 10 cells/ml, make a series of higher dllutlons, 1 e , 5, 3, and 1 cells/ml Transfer 100 pL of the latter cell suspensions mto 96-well round-bottom plates The number of wells to be filled depends on the expected cloning frequencies (see Note 11) Add 100 pL of culture medium containing rIL-2 (20 U/mL) after 5 d of culture After ldentlficatlon of growing T-cell clones, usually after l&14 d of culture, transfer the cells m 200 pL to a 24-well Lmbro plate, add 300 pL of culture medmm, and add 0 5 mL of a 2X feeder cell mixture to expand the cells (see Sectlon 3 6 )

Antigen-Speclflc

127

T-Cells

3.5. Selection of Al/o-Antigen-Specific 3.5.1. Assay for Cytotox/c Actwty

T-Cell Lines

1 Wash 10h EBV-LCL to be used as target cells m the assay with PBS, and remove supernatant Incubate pellet with 100 ~CI 5’Cr m a water bath at 37°C for 45 mm Wash the cells three times with PBS, and resuspend m culture medium Seed 1O3 to 2 x 1O3target cells/well of a 96-well round-bottom plate m a volume of IOOpL Add 100 FL of effector T-cells m tnphcate wells at different effecter/target cell ratios Keep separate wells with target cells that have been incubated with medium alone (spontaneous 5’Cr release) or with 1% Tnton-Xl00 (maximal 5’Cr release) 7 Centrifuge the plates for 5 mm at 800 rpm 8 Incubate for 4-8 h at 37”C, 5% CO1 9 Harvest 100 pL of culture supernatant from each well, and count amount of released 5’Cr in a y-counter (Packard Instrument Corp , Downers Grove, IL) 10 Calculate the percentage of specific 5’Cr release with the formula (Release of sample - spontaneous release)/ (maxlmal release - spontaneous release) x 100%

(1)

and select responding T-cell clones 11 HLA-restnctlon elements can be determined by carrying out the cytotoxlclty assay m the presence of anti-HLA-DR, DQ, or DP MAb, and/or by usmg various HLA-typed EBV-LCL as target cells

3.5.2. Assay for Antigen-Specific

Proliferation

1 When using autologous EBV-LCL as APC, incubate 2 x lo4 T-cell with 4 x IO4 irradiated (5000 rad) EBV-LCL, m a 96-well round-bottom plate m a final volume of 200 pL m the presence or absence of antigen When autologous PBMC are used as APC, incubate lo5 T-cells with IO5 lrradlated (4000 rad) PBMC, m a 96-well flatbottom plate, in the presence or absence of antigen m a final volume of 200 pL Optimal concentration of antigen has been determined m Sectlon 3 3 , step 3 2 Incubate for 72 h at 37”C, pulse with 1 @I (=37 kBq) [3H]TdR, and incubate for another 4 h 3 Harvest cells onto a glass fiber filter, determine amount of incorporated [3H]TdR by llquld scmtlllatlon counting, and select responding T-cell clones 4. HLA-restnctlon elements can be determined by carrying out the proliferation assay m the presence of anti-HLA-DR, DQ, or DP MAb, and/or by usmg various HLA-typed EBV-LCL as APC

3.5 3. Stimulation of T-Cell Clones for Antigen-Speafic Cytokine Product/on Assay 1 Incubate lo6 T-cells/mL with 2 x lo6 autologous EBV-LCL or lo6 autologous PBMC as APC, m the presence or absence of antigen m either a 24 (final volume 1 mL), 48- (500 pL), or 96-well(200 pL) plate

128

Yssel

2 For polyclonal strmulation, use combmations of PMA (1 ng/mL) and A23 187 (500 ng/mL), PMA and soluble anti-CD3 MAb (1 ug/mL), ant]-CD28 MAb (1 pg/mL) and coated anti-CD3 MAb (to coat plates with anti-CD3 MAb, mcubate a 96-well flat-bottom plate with 10 ug/mL anti-CD3 MAb diluted m PBS overnight at 4”C, and wash twice with 100 pL of culture medium) 3 Culture the cells at 37°C 5% CO, for 48-72 h, and harvest the supernatants, which can be frozen once, prior to cytokine analysis 4 Determme cytokme production levels by cytokme-specific ELISA (see Note 12)

When EBV-LCL and peptldes are used, in Sections 3.5.2. and 3.5.3., EBVLCL can be fixed with glutaraldehyde. 1 2 3. 4 5

Premcubate EBV-LCL m culture medium with l-5 ug/mL peptide for 2-24 h Wash the cells once m PBS, and resuspend between lo6 and lo7 cells m 1 mL PBS Add 1 mL of 0 05% stock giutaraldehyde, and incubate for 30 s at room temperature Add 2 mL of PBS, and Incubate for an additional 10 mm at room temperature Wash twice with culture medium, and use m the assays, described m Sections 352 and353

3.6. Expansion

of Cloned T-Cell Lines

1 Prepare a 2X concentrated feeder cell mixture, consistmg of 2 x lo6 cells/ml irradiated (4000 rad) PBMC from any healthy donor, 2 x 10s cells/ml irradiated (5000 rad) EBV-LCL, and 100 ng/mL PHA, and keep at 4°C until use (see Notes 13-15) 2 Collect the T-cell clones, wash with medium or with PBS, supplemented with 2% HS or BSA, and brmg the cells to a concentration of 4 x 10s cells/ml 3 Transfer 0 5 mL of the T-cell clone suspension mto a 24-well plate, add 0 5 mL of the 2X concentrated feeder cell mixture, and Incubate at 37°C and 5% CO2 (final concentrations of feeder cells per well are lo6 PBMC and 10s EBV-LCL, and concentration of PHA IS 50 ng) 4 T-cell clones should be spht with fresh culture medium, supplemented with growth factors (r-IL-2 20 U/mL + rIL-4 10 ng/mL), usually between d 3 and 4 after restimulation with the feeder cells (see Notes 16-18) 5 Continue to expand the cultures by adding medium and growth factor(s) over the next 7-10 d 6 When the T-cells become smaller and round off, usually between d 10 and 14 of culture, repeat the culture cycle starting with step 1 (see also Fig 2 and Notes 19 and 20)

3.7. Cryopreservation 3.7. I. Cell Freezing

of T-Cell Clones

1 Centrifuge cells m a 15-mL centrifuge tube, and resuspendthe pellet in RPMI1640, supplementedwith 10% HS, fetal calf serum(FCS) or 1% BSA, at concentrations between lo6 and lo8 cells/ml and put on ice (see Note 2 1).

Antigen-Specific

T-Cells

129

Fig. 2. Different stages in the culture of cloned T-cell lines. (A) Culture aspect 24 h after addition of the feeder cell mixture. Mostly, PBMC and EBV-LCL can be seen. (B) Growing T-cells between 3 and 6 d after restimulation. The cells are growing in aggregates and have a blastoid appearance. (C) T-cells that have been cultured for 10-14 d in the presence of growth factors. The cells have stopped proliferating and have become small and round.

2. Make a fresh solution of 20% DMSO in RPMI-1640, supplemented with HS, FCS, or BSA, and put on ice (add DMSO to medium and not vice versa) (see Note 22). 3. Add dropwise the same volume of the DMSO solution to the cell suspension while gently shaking for 2-3 min. 4. Transfer the cells to the freezing vials, and freeze using a temperature-controlled freezing apparatus.

3.7.2. Cell Thawing 1. Thaw vial quickly in 37°C water bath, transfer the cells to 15-mL tube, and put on ice. 2. Add dropwise 2 mL of cold PBS, at first very slowly, while gently shaking, over a time period of about 2-3 min. 3. Using a 2-mL pipet, put 2 mL of HS or FCS under the cell suspension. 4. Centrifuge cells for 5 min at 1000 rpm and remove supernatant. 5. Wash cells once with culture medium.

4. Notes 4.1. Culture Medium 1. Yssel’s medium is suitable for the culture of T-cell clones in the absence of added HS or FCS. The addition of 1% human AB+ serum will, however, optimize

730

Yssel

growth condltlons The BSA preparation hsted m Sectlon 2 1 contains a slgmficant amount of endotoxm, which, however, does not affect T-cell growth If endotoxm-free condltlons are required for subsequent experiments with these cells, the preparation can be replaced with endotoxm-free BSA (Sigma A367.5) Yssel’s medium, supplemented with 1% human AB’ serum, can be purchased from Gemini Bloproducts Inc (Calabasas, CA, cat no 400- 113) 2 Both the Ho10 form and the Apo form of transferrm can be used However, mdlvldual batches may have mhlbltory, rather than growth-promoting, effects on T-cell growth and should therefore be tested prior to use Furthermore. transferrm preparations may contam human Ig m detectable amounts, which may cause background problems m antibody production assays, notably those for IgE determinations For such experiments, replace the transferrm source with Ig-free transferrm (Pierce, Rockford, IL, cat no 3 1152) 3 Stock lmolelc and olelc acid should be stored, as free acids. at -20” under mtrogen to prevent oxldatlon of the unsaturated bond

4.2. Generation, Cloning, and Selection of Antigen-Specific T-Cell Lines Commonly used antigens are tetanus toxoid (Lederle, Pearl River. NY), purified protein derlvatlve (PPD) (Staten Serum Institute, Copenhagen, Denmark), house dust mite (Dermatophagoldes spp), and grass pollen (Lokum spp) proteins (ALK, Copenhagen, Denmark) Most genes codmg for these protein have been cloned, and recombinant or synthetic peptides are available For the generation of allo-antigen-specific T-cells, stimulate PBMC with EBVLCL at concentrations of about 25%. since higher concentrations will hkely result m the generatron of T-cells with nonspecific, NK-like activity (3) It IS recommended to include some lower dilutions of T-cells (10, 102, and 1O3 cells/well) m clonmg experiments The growth ofthese cultures will give an early mdlcatlon that culture condltlons for the cloning experiment were optlmal For the generation of CD4+ allo-antigen-specific cells, use HLA class II-mlsmatched EBV-LCL, whereas for the generatlon of CD8+ allo-antigen-specific cells, HLA class I-mismatched EBV-LCL should be used The protocol, described m Section 3 3 , has been used successfully for the generation of both types of allo-antigen-specific T-cell clones (3,.5,19,20) For the generation of antigen-specific ThO cells, PBMC from donors who have been previously lmmumzed with tetanus toxoid can be used successfully Ideally, donors should have received a booster mjectlon about 1 mo, but not earlier, prior to the use of their PBMC Thl cells, reactive with PPD, can be generated using PBMC from donors who have previously been vaccinated with BCG or from those who have a positive Mantoux reaction Human Th2 cells can be generated from the peripheral blood of atoplc donors by stimulation with soluble allergens, such as house dust mite allergen Dermatophagoldes spp or grass pollen Lolzum perenne Suitable patients can be selected, based on their allergenspecific serum IgE levels and RAST scores Although there 1s not always a

Antigen-Speafx

T-Cells

131

posmve correlation between RAST score and success rate m the generation of allergen-specific Th2 clones, donors with a RAST of +++ or more should be chosen preferentially Since the frequency of allergen-specific T-cell clones m peripheral blood IS rather low (21), better results can be obtamed usmg skm blopsles from patients with atoplc dermatltls 9 Although EBV-LCL are efficient APC when peptldes are used, their capacity to present native antigens, such as tetanus toxoid or PPD, 1s often hmlted Therefore, preferentially autologous or HLA-matched PBMC should be used m prollferatlon or cytokme productlon assays with native antigen 10 EBV-LCL function better as APC at higher R/S ratios (3 1 or 5 1) However, high concentrations of EBV-LCL give high background readings m prollferatlon assays This problem IS avoided by using glutaraldehyde-fixed cells 1 1 Cloning efficiencies can be calculated most reliably from serial dllutlons For the calculation of clonmg frequencies, plot the fraction of negative wells @-axIs) against the number of seeded wells (x-axis) The cell dose contatmng one clonogemc cell IS determined by locatmg on the x-axis the dose that corresponds to 0 37 cells/culture Alternatively, clonmg efficlencles can be calculated from one smgle dllutlon with the formula -In[(number of negative wells/number of total wells)/ number of cells seeded/well] x 100% 12 Detailed described of MAbs be found

4.3. T-Cell

(2)

protocols for double-sandwich cytokme-specific ELISA have been m refs. 22 and 23 Most ELISAs can be purchased commercially A list and polyclonal antlbodles available for the detection of cytokmes can m ref 23

Culture

13 Smce culturing of T-cell clones requires large amounts of PBMC from different donors, donor serum should be tested for the presence of antlbodles agamst hepat&s B, CMV, and HIV antigen prior to processmg of the cells, whereas all cell cultures contammg human PBMC should be handled according to the blosafety guldelmes of the NIH (24) Freshly isolated, nonirradiated, PBMC can be stored m culture medium at 4°C for at least 2 d prior to use. 14 To remove the abundance of platelets present m peripheral blood after lsolatlon of the PBMC, wash at least four times with PBS, supplemented with 2% HS or BSA, by centnfugatton at 800 rpm for 10 mm m 50-mL centrifuge tube 15 After u-radiation of cells with a cesmm source, wash the cells at least once with culture medium to remove oxygen (O-) radicals that may have formed owing to the lrradlatlon 16 Usmg the condltlons described m Sectlon 3 6, It 1s generally not necessary to add any medium and growth factors to the T-cell cultures for the first 34 d, unless higher numbers of T-cell clones are present m the feeder cell mixture For optimal growth, when culturing the T-cell clones m tissue-culture plates or flasks,

132

17

18

19

20

21

22

Yssel tt 1s Important to keep growmg T-cells m log phase by allowing the cells to grow at rather high density ( 106-2 x lo6 cells/ml) and by frequently sphttmg the cultures with fresh culture medium, supplemented with growth factors As the primary growth factor for the expansion of T-cell clones, rIL-2 should be used, whereas t-IL-4 might be added when Th2 cell clones are grown IL- 12 has been found to have T-cell growth-promotmg effects However, tt strongly induces the productton of IFN-?/ m peripheral blood (25) and even m established Th2 clones (26) Although tt has yet been tmposstble to evaluate the amount of IL- 12 m the culture system described m thts chapter, the use of PBMC and EBV-LCL, which have been described to be producers of IL-12, may bias cultured human Th2 clones toward the ThO phenotype and may explain the frequently observed mtrmstc capacity of many of Th2 clones to produce IFN-y Prohferatmg ThO and Th 1 cell clones ~111usually remove dead feeder cells and cell debris wtthm the first 4 d of culture (see Ftg 2) Although the mechamsm for these scavenger properties 1s unknown, tt may be related to the cytotoxtc potential of these cells Indeed, Th2 cells, which have no lyttc acttvtty, grow less well under our culture condmons, and often much debris remains m the wells. which seems to have mhtbnory effects on the growth of the cells. It 1s recommended to collect the cells and wash them several times with medium or PBS supplemented with 2% HS or BSA by centrtfugatton at 800 rpm for 10 mm, after which they should be transferred to new wells at a concentratton of 5 x 10s cells, m culture medium contammg growth factor(s) This procedure can be camed out repeatedly Keep track of the age of the T-cell clones, and make frozen stocks of the cells, especially at an early stage The cells generated lust after clonmg are the most valuable and should be kept as a backup storage Although some long-term cultured T-cells seem to surpass the 30-50 cell dtvtstons, known as the Hayflick Itmtt, m the author’s experience, the growth of the T-cell clones usually slows down after about 15-20 sttmulations with the feeder cell mixture, and it becomes more and more difficult to expand the cells Cells contammated with mycoplasma show decreased growth and altered mncttonal capacities Check cell cultures, especially those that are frequently handled and maintained m culture for extended pertods of time, for the presence of mycoplasma at least every 14 d Cells that are contaminated with mycoplasma should be discarded nnrnedrately The success of T-cell cloning and culture IS greatly affected by the presence of mycoplasma T-cell clones can be frozen at any time and, after thawing, can be expanded tmmedtately with the feeder cell mixture Cells frozen between d 4 and 6 after resttmulation with the feeder cells, which are at that time still highly prohferattve, can be cultured for several days m medium with growth factors on thawing Cells frozen at later stages should be cultured with feeder cells prior to expansion wtth growth factors, Do not use IMDM, Yssel’s medium, or any other medium contammg HEPES when freezing T-cells, since the permeabthzatton of the cell membrane by DMSO results m the penetration of HEPES into the cells, which may lead to cell death.

Antigen-Specific

T-Cells

133

Acknowledgment DNAX Research Institute for Molecular by Schering-Plough Corporation.

and Cellular

Biology

IS supported

References 1 Morgan, D A , Ruscettt, F W , and Gallo, R (1976) Selective m vitro growth of T lymphocytes from normal human bone marrows Science 193, 1007,1008 2 Gtllis, S and Smith, K A (1977) Long term culture of tumour-specific cytotoxtc T cells. Nature 268, 154-I 56 3 Sptts, H , de Vrtes, J E , and Terhorst, C (198 1) A permanent human cytotoxtc T-cell lme with high klllmg capacity against a lymphoblastotd B cell line shows preference for HLA A, B target antigens and lacks spontaneous cytotoxic acttvtty Cell Immunol 59,435%44 1 4 Krensky, A M , Retss, C S , Mter, J W , Strommger, J L., and Burakoff, S J (1982) Long-term cytolyttc T-cell lines allospectlic for HLA-DRw6 antigen are OKT-4+. Proc Nat1 Acad Scl USA 79,2365-2369 5 Sptts, H , Yssel, H , Terhorst, C , and de Vries, J E (1982) Establishment of human T lymphocyte clones htghly cytotoxic for an EBV transformed B cell lme

m serum-free medmm. lsolatlon of clones that differ in phenotype and specliklty 6

7.

8

9. 10

11

12

13

J Jmmunol 128,95-99 Malissen, B , Kristensen, T., Goridrs, C , Madsen, M , and Mawas, C (198 1) Clones of human cytotoxtc T lymphocytes derived from an allo senstttzed individual HLA specificity and cell surface markers Stand J Immunol 14, 2 13-224 Spits, H , Borst, J , Terhorst, C , and de Vrtes, J. E (1982) The role of T cell dtfferentiatton markers m anttgen-spectlic and lectm-dependent cellular cytotoxictty mediated by T8+ and T4+ cytotoxtc T cell clones dnected at class I and class II MHC antigens J Immunol 129, 1563-1569 de Vrtes, J E and Sptts, H. (1984) Cloned human cytotoxtc T lymphocyte (CTL) lines reactive with autologous melanoma cells. I. In vitro generation, tsolation and analysts to phenotype and specificity J. Immunol. 132,5 10-5 19 Yssel, H., Spits, H , and de Vries, J. E (1984) A cloned human T cell line cytotoxtc for autologous and allogenetc B lymphoma cells J Exp Med 134,239-254 Yssel, H., Blanchard, D , Boylston, A., de Vrtes, J. E., and Sptts, H (1986) T cell clones which share T cell receptor epttopes differ m phenotype, functton and specificity Eur J Immunol 16, 1187-l 193 Yssel, H., Shanafelt, M -C , Soderberg, C., Schneider, P. V , Anzola, J , and Peltz, G. (199 1) B burgdor- activates T cells to produce a selective pattern of lymphokmes m Lyme arthrttts J Exp Med. 174,593-601 Haanen, J. B., de Waal MalefiJt, R., Res, P. C , Kraakman, E. M., Ottenhoff, T. H , de Vrtes, R R., and Spits, H (1991) Selection of a human T helper type 1-lake T cell subset by mycobacteria. J Exp. Med 174,583-592. Yssel, H., Johnson, K. E , Schneider, P V , Wtdeman, J , Terr, A R K , and de Vries, J. E. (1992) T cell acttvatton mducmg epttopes of the house dust

134

14

15

16

17

18

19

20

21

22

23

24

Yssel

mtte allergen Der p I Inductton of a restricted cytokme productton profile of Der p I-spectfrc T cell clones upon antrgen-specrfrc acttvatton J Immunol 148, 738-745 Alarcon, B , de Vrtes, J E , Pettey, C , Boylston, A , Yssel, H , Terhorst, C , and Sptts, H (1987) The T cell receptor y/CD3 complex tmpltcatton m the cytotoxtc acttvtty of a CD3+ CD4- CD8- human natural ktller clone Proc Nut1 Acad Scz USA 84,3861-3865 Sredm, B , Volkman, D , Schwartz, R. H , and Faucr, A S (1981) Anttgen-specrfic human T cell clones. development of clones requtrmg HLA-DR compattble presenting cells for sttmulatton m the presence of anttgen Proc Nat1 Acad Scr USA 78, 1858-I 862 Eckels, D S , Lamb, J R , Lake, P , Woody, J N , Johnson, A H , and Hartzman, R J (1982) Anttgen-specific human T lymphocyte clones Genetic restrtctton of Influenza vtrus-spectfic responses to HLA-D regron genes m man Hum Zmmunol 4,313-324 Yssel, H , de Vrtes, J E , Koken, M , Van BlitterswiJk, W , and Spits, H (1984) Serum-free medtum for the generation and propagatton of functtonal human cytotoxtc and helper T cell clones J Immunol Methods 72, 2 19-227 Boyum, A (1968) Separatton of leukocytes from blood and bone marrow tsolatron of mononuclear cells and granulocytes from human blood tsolatton of mononuclear cells by 1 mst centrtfugatton and of granulocytes by combmmg mst centrtfugatton and mst sedrmentatron at 1 g Stand J Ch Lab Invest, tl(Suppl. 97), 77 Sprts, H , Borst, J , Gtphart, M , Colhgan, J , Terhorst, C , and de Vrres, J E ( 1984) HLA-DC antigens can serve as recogmtron elements for human cytotoxtc lymphocytes Eur J Immunol 4,299-304 Sprts, H , Yssel, H , Thompson, A , and de Vrres, J E (1983) Human T4+ and T8+ cytotoxrc T lymphocyte clones dtrected at products of dtfferent class II maJor htstocompatrbthty complex IOCI J Zmmunol 131, 678-683 Sager, N , Feldmann, A , Schtllmg, G , Krettsch, P , and Neumann, C (1992) House dust-mtte-specific T cell m the skin of SubJects wrth atoptc dermatitis frequency and lymphokme profile m the allergen patch test J Allergy Clan Immunol 89,801-810 Abrams, J S , Roncarolo, M -G , Yssel, H., Andersson, U , Gletch, G J , and Silver, J (1992) Strategtes and practice of anti-cytokme monoclonal antibody development Immunoassay of IL-1 0 and IL-5 m clmlcal samples Immurtol Rev 127,5-24 Abrams, J. S (1995) Immunoenzymatrtc assays of mouse and human cytokmes, usmg NIP-lmked ant+cytokme anttbodtes, m Current Protocols zn Immunology (Cohgan, J E , Krmsbeek, A M., Margulies, D. H , Shevach, E M., and Strober, W , eds ), Wiley, New York, pp 6 20.1-6 20 15 Centers for Dtsease Control and Prevention and National Instttutes of Health (1993) Bzosafety in Mzcrobzologzcal and Medical Laboratorzes HHS publtcatton no CDC 93-8395, US Government Prmtmg Office, Washmgton, DC

Antigen-Specific

T-Cells

135

25 Chan, S H , Perussta, B , Gupta, J. W , Kobayashr, M , Posptsd, M , Young, H A ~ Wolf, S F , Young, D , Clark, S C , and Trmchrert, G (199 1) Inductton of Interferon y production by natural killer cell sttmulatory factor charactertzatton of the responder cells and synergy wtth other Inducers J Exp Med 173, 869-879 26 Yssel, H , Faster, S , de Vrtes, J E , and de Waal Malefyt, R (1994) lnterleukm 12 (IL- 12) transtently Induces IFN-y transcrtptron and protean synthests m human CD4+ allergen-specific helper type 2 (Th2) T cell clones 11zt Immunol 6, 109 1-l 096

12 Protection of Purified Human Hematopoietic Progenitor Cells by Interleukin-1 p or Tumor Necrosis Factor-a Robert J. Colinas 1. Introduction Hematopoiests is the process of blood cell formation In the adult human, the bone marrow (BM) is the primary hematopoietic organ. Each day, the BM produces billions of leukocytes, erythrocytes, and thrombocytes, which enter the circulation. Production of such enormous numbers of mature blood cells results from the exponential expansion and dtfferenttation of pluripotent hematopoietic stem cells through a series of mcreasmgly more differentiated hematopoietic progenitor cells (HPCs) Regulation of hematopoiesis is accomplished by the transduction of signals that follow mteractions of the HPCs and developmg blood cells with BM stromal cells, cytokmes, and the extracellular matrix. Functional assessment of the hematopoietic potential of human HPCs is accomplished usmg m vitro colony-forming cell (CFC) and long-term culture (LTC) methodologres that were inmally developed using rodent bone marrow (‘1,2) Of the techniques available, the CFC assay IS the most commonly employed and has been adapted for use with multiple sources of human HPCs, mcludmg BM, fetal liver, umbilical cord blood, and peripheral blood. Colony formation by HPCs m the CFC assay relies on the addition of exogenous cytokmes to cells suspended in semisolid medium. Since no colonies will form m the absence of added cytokmes, the investigator has control over which hematopoietic lineage IS being assessed. It has been shown that the oxygen partial pressure (PO,) m human BM is approx 2-5% (3). Thts is considerably lower than the near-ambient p02 (19.3%), which exists m the typtcal cell-culture incubator maintained at 7% From

Methods Edlted

In Molecular by

G E Jones

Medune Humana

137

Human Cell Culture

Protocols

Press

NJ

Inc , Totowa,

138

Colmas 80

Fig. 1. Effects of near-ambient (19 3%) and physlologlcal(5%) p02 on colony formatlon by human HPCs m vitro Normal human BMMNCs were plated m semlsolld medmm contammg 2 ng/mL rHuGM-CSF The cultures were incubated at 37°C m either air/7% CO2 or 5% 02, 7% COz. and 88% NZ, and colonies of 250 cells were counted after 14 d The data shown are from one representative experiment C02. Numerous studies have demonstrated that, even m the presence of exogenously added reducing agent, both CFC and LTC assays perform nearly twofold better when conducted at physiological p0, (4,5) (Colmas, unpublished) (Fig 1). Furthermore, m experiments using either mouse or human BM, it has been shown that the mhibitory effects of several oxidative stress-mducmg hematotoxic agents and near-ambient p02 on granulocyte/macrophage-colonystimulating factor (GM-CSF)-induced colony formation are additive (6). Thus, it is important to conduct studies on human HPCs at physiological p02 As stated, hematopoiesis is largely controlled by the mteractions of HPCs with cytokmes Interleukm- 1l3 (IL- 1 p) and tumor necrosts factor-a (TNFa) are pleiotroprc mflammatory cytokmes that are believed to play a central role n-r hematopoiesis (7,8). Cellular sources of IL-l p and TNFa mclude macrophages, lymphocytes, granulocytes, libroblasts, and endothehal cells Using unfractionated BM cells or purified HPCs (pHPCs), IL- l/3 synergizes m vitro with other cytokmes, such as IL-3, GM-CSF, IL-6, or c-kit hgand (KL), resultmg m both increased colony size and numbers (9,10). In contrast, when unfractionated bone marrow mononuclear cells (BMMNC) are treated with TNFa, colony formation IS inhibited (ZZ-23). However, mvesttgattons using pHPCs have shown that TNFa stimulates more primitive HPCs, while inhibit-

Human Hematopoietc ProgenrtoorCells Table 1 Studies Demonstrating Experimental system Mouse, Mouse, Mouse, Mouse, Mouse, Mouse, Mouse, Human, Human, Human, Human,

m VIVO, m vitro m vivo m vlvo m vlvo m VIVO m vwo m vwo m vitro m vitro m vitro m vitro

Protection

by IL-la,

139 IL-l&

or TNFa

Cytokme

Treatmenta

IL-l, TNFa IL- 1CL,TNF-a IL- 1, TNF-a TNF IL-la IL-la IL- 1, TNF-a IL-l IL- 1, TNF-a IL- 1, TNF-a TNF-a

y-Radiation Endotoxm Ischemla/reperfuslon SFU, MTX, VB y-Radlatlon with DX or DDP DX hyperoxla y-radiation 4HC DX

HQ

References (17-l 9) t..O,21) (22,23) (24) (25) (2628) (29.30) (31) (32933) (34) (34)

“The specific treatment protocols employed and the cell types protected In these studies are avallable m the correspondmg references Abbrevlatlons are as follows 5-FLJ, 5-fluorouracll, MTX, methotrexate, VB, vmblastme, DX, doxorublcm, DDP, cls-dlchlorodlammme platmum, 4HC, 4-hydroperoxycyclophosphamlde, HQ, hydroqumone

mg those progenitors committed to the erythrold and myelold hneages (1616) Thus, it appears that the net effect of TNFa on hematopolesls depends on the differentiation stateof the HPC and the specific lineage to which it 1scornmltted IL- 1p or TNFa treatment also has been shown to protect HPCs as well as other cell types against the effects of numerous inhibitory agents (Table 1). Experlmental evidence suggeststhat the inhibitory effects of these agents result from the production of oxldatlve stress through mechanisms that generate reactive oxygen mtermedlates (ROIs) or deplete glutathlone (GSH) Thus, it 1s likely that IL- 1p and TNFa produce a state within the HPC that makes it more resistant to oxldative stress.This state 1smost likely achieved by the mductlon of gene expression, the products of which are involved m antloxldant and/or detoxification responses (Fig. 2) Consistent with this hypothesis, IL-l p and TNFa have been shown to induce manganous superoxide dlsmutase (MnSOD) expression m both normal and tumor cell lines (35-37). In addition, TNFa induces aldehyde dehydrogenase (ALDH) expression m human HPCs ALDH 1s an enzyme that detoxifies the reactive 4-hydroperoxycyclophosphamide (4HC) metabohte aldophosphamlde (38). Moreover, TNFa also induces an umdentlfied response that protects human HPCs from phenylketophosphamlde toxicity (39). The followmg method describes an experimental protocol that has been used to determine whether IL- 1p or TNFa could protect human HPCs from the mhlbltory effects of doxorublcm (DX) at physlologlcal p02 and temperature

Cohas

140 IL-I p or TNFa

Toxicant

Binding Signal

Transduction Genetic

AIteratIons

Human

HPC

Fig. 2. A model for protection of human HPCs by IL- 1p or TNFa It IS believed that these cytokmes mduce genetrc changes in the HPC, whrch are capable of protectmg against a wide range of toxtc agents Some of these genetic changes are known (MnSOD and ALDH), and others are unknown (7)

(34). A summary of thts procedure is represented schematically

m Ftg. 3 Bnefly,

pHPCs are isolated from normal human BM (Section 3.2.), and pretreated with IL- l/3 or TNFa in the absence or presence of DX (Section 3.3.). The cells are washed (Section 3.4.) and plated in semtsohdmedium contammg colony-stimulatmg cytokmes(Section 3.5 ). After a 14-d mcubatron, colony numbers and types are delineated (Section 3.6.) IL-l l3or TNFa is considered protective if two criteria are met. First, pretreatment with either cytokme alone does not sigmficantly affect CFU-C frequency or type relative to the medium-pretreated control HPCs Second, the addition of either cytokme to DX-exposed pHPCs results m greater numbers of CFU-C than m the samplesexposed to DX alone. Although the protocol that follows is specific for normal CD34+ BM cells pretreated with combmations of IL- 1p, TNFa, and DX, It could be easily adapted to determine whether any identifiable HPC subpopulation can be protected by a given bioactive molecule against the mhibitory or toxic effects of any modifier (seeNote 8) 2. Materials 2.1. Human BM Source An approved source of normal human HPCs must be available. In our studies, aspirates from the posterior iliac crest of consenting hematologically

Human Hema topoie tic Progenitor Cells

141

HPC Protection Assay BM aspirate

---L

aliquot

for FACS

4 BMMNC

-

aliquot

for FACS

J pHPCs

-

aliquot

for FACS

J 22hr pretreatment flL-1 p or TNFa, 4DX 37”C, 5% 02,7% co2 J Plate +GM-CSF, IL-3 and KL J Culture 14 days 37OC, 5% 02,7% co2 J Stain colonies with Mabs J Count CFU-G, -M and GM Fig. 3. Schematic representation of the HPC protectton assay

normal 18-40-yr-old donors have been used in an Institutional Review Boardapproved outpatient procedure. The aspiration procedure is conducted by a licensed hematologist using a Sternal/Iliac BM aspn-atlon needle (Baxter Sclentlfic, Valencia, CA; cat no. DIN1515X). 2.2. Special Equipment 1 A laminar flow btologtcal safety cabmet (TC hood) (Baker Co , Sanford, ME) 2 A sealed glove box wtth an air lock (PlasLabs, Lansing, MI, model no. 830 ABCSPEC) The glove box chamber ts fitted wtth an HEPA outlet filter, a vacuum lme controlled from the mstde, and an inlet valve controlled from the outstde The arrlock has both externally controlled inlet and vacuum lmes The inlet valve is connected to a cylinder of compressed 5% O,, 7% C02, and 88% N2 fitted with an in-lme 0 2-pm gas filter 3 A flow cytometer, such as a FACScan (Becton Dickinson, San Jose, CA).

142

Coltnas

4 A water-jacketed CO, incubator (Forma Scientific, Marietta, OH, model 3 1.58) retrofitted with an OxyReducer TM (Remmg Biomstruments, Redfield, NY, model 3 11) (see Note 3) or an incubator designed to regulate both p0, and pC0, (Hareus Inc Equipment Group, South Plamfield, NJ) 5 A fluorescence microscope, such as a Zeiss Axioskop (Zeiss, Thornwood, NY)

2.3. CD34’ lmmunoselecfion

Materials

There are several commerctally available systems for immunoselectmg human HPCs All operate with similar effficienctes (approx 6580% CD34+, with I1 5% expressing cell-surface antigens found on mature blood cells, m the recovered populatton, and a 50% yield) and are based on monoclonal anttbodies (MAbs) to the CD34 cell-surface glycoprotem that has been shown to be an HPC marker (40). The CellPro biotmylated antiCD34 MAb-avidm columnbased system (Bothell, WA, cat no. LC34) has been found to be reasonably economtcal, relatively quick, and can be used m the glove box

2.4. Cyfokines In our studies, researchers have utilized IL-lp, TNFa, IL-3, GM-CSF, and KL. There are many commerctal sources of recombinant human (rHu) cytokmes, and significant differences between the same cytokmes obtained from different manufacturers have not been found. However, tt is much more economtcal, though somewhat slower and more laborious, to obtain the needed cytokmes from generous researchers that produce them m large quantities m then- own laboratories The Btological Response Modifiers Program of the National Cancer Institute (Frederick, MD) is another economical source for cytokmes. Cytokmes should be prepared as 100X stock solutions m PBS/O. 1% BSA, ahquoted and stored at -70°C

2.5. Disposable

Tissue-Culture

Supplies

Sterility of tissue cultureware 1s crucial, since the cells will be cultured for over 2 wk. Nearly all of the supplies used are generic and come presterthzed Vacutamer brand evacuated test tubes without anttcoagulant can be obtained from Fischer Sctentific (Pittsburgh, PA; cat. no 02-685 A) Obtain nonsterile I .9-mL polypropylene microcentrifuge tubes from Baxter Dtagnostics Inc. (McGaw, IL, cat no. C3520-1) and autoclave for 20 mm at 121°C then dry at 60°C before use.

2.6. Media, Fetal Bovine Serum, and Reagents 1 Low endotoxm Iscove’s modified Dulbecco’s medium (IMDM)

IS available from

Blowhlttaker (Walkersvdle, MD) and 1s stored at 4°C (see Note 4) 2 Defined fetal bovme serum (FBS) is purchased from Hyclone (Logan, UT), thawed, allquoted m 5- and lo-mL volumes, and stored at -20°C

Human Hematopo/et/c Progetxtor Cells

143

3 Phosphate-buffered salme (PBS) and 0 9% NaCl (saline) are available from Btowhtttaker 4 Ftcoll-Hypaque denstty gradtent medium 1s from Pharmacta (Uppsala, Sweden) and 1s mamtamed at room temperature m the dark 5 A 1000 U/mL solutton of preservative-free sodturn heparm (PF heparm) can be obtained from Lyphomed (Deerfield, IL) 6 DX can be purchased as a sterile 2 mg/mL solution from Chtron Therapeutics (Emeryvtlle, CA) and should be stored at 4’C m the dark 7 Powdered IMDM 1s available from Gtbco-BRL (Gatthersburg, MD, cat. no 12200-036) To make 2X IMDM, follow the manufacturer’s directions for preparation, except bring the final volume to 500 mL instead of 1 L Filter the medium through a 0 22-urn filter, and store at 4°C for no more than 1 mo 8 Low-melting-point (LMP) Seaplaque agarose is available from the FMC Corp (Rockland, ME, cat no 50102) Stocks of 0.9% LMP agarose are made tn 250-mL glass bottles by adding 1 8 g of agarose to 200 mL chilled pyrogen-free double-disttlled detomzed H,O, mtxmg well, and autoclavmg at 12 1“C for 25 mm After the autoclave has returned to ambient pressure, but the agarose 1s still molten, tighten the caps and mix well Store agarose at room temperature

2.7. lmmunophenotyping

Reagents

1 A phycoerythrm (PE)-conjugated MAb against the HPC antigen CD34 (HPCA-2) can be obtained from Becton Dickinson 2 The PE-conjugated monocyte-specific anti-CD14 and the fluorescein isothtocyanate (FITC)-conjugated granulocyte-specific anti-CD66b MAbs are avatlable from Becton Dtckmson and Immunotech (Westbrook, ME), respectively Since fluorochrome-conjugated MAbs are light- and heat-sensitive, these reagents should be stored at 4°C m the dark 3 Paraformaldehyde can be obtained from Fisher Sctenttfic Co Make PBS/2% parafotmaldehyde m a chemical fume hood by stnrmg 2 g of paraformaldehyde into 100 mL of 70-80°C PBS When the solution clears, cool, pass through a 0.22~pm filter, and store at 4°C 4 Prepare PBS, 0 1% (w/v) NaN, by drssolvmg 1 g NaNs m 1 L PBS NaN, ts a hazardous substance Follow the manufacturer’s directions for safe handling and disposal Make PBS, 0 1% NaN3, 0 1% BSA, pH 8 0, by adding 0 4 g BSA to 400 mL PBS, 0 1% NaN, and adjustmg the pH to 8 0 with 1ONNaOH Add 1 25 mg mouse IgG (Sigma, St Louts, MO; cat no 15381) to a 25-mL aliquot of this solution. Store at 4°C

3. Methods

3. I. Preliminary

Preparations

1 Two hours before beginning this procedure, clean the interior of the glove box and airlock with dismfectant and allow to dry 2 Stock the glove box wtth the necessary sterile supplies (1 e , ptpeters, serological ptpets, ptpet tips, a waste contamer, several 15- and 50-mL conical polypropy-

Co/has

144

lene centrifuge tubes, tube racks, sterile scissors, and the CellPro column stand) and seal the inner airlock door 3 With the HEPA filter completely open, flush the an out of the glove box at 10 psi for 10 mm with the 5% O,, 7% CO2 88% N, gas mixture, and then decrease the flow rate to 1 psi to mamtam a slight positive pressure.Partially close the HEPA filter (seeNote 5) 4 While the glove box IS bemg gassed, prepare 50 mL of PBS, 10 U/mL PF heparm, two 15-mL cushions of Ficoll-Hypaque m 50-mL conical polystyrene (clear) centrifuge tubes, 150 mL PBS, 5% FBS, 50 mL PBS, 1% BSA (made from CellPro 5% BSA), and 40 mL IMDM, 20% FBS, 4 mM L-glutamme120 pg/mL Gentamicm sulfate (complete IMDM, 20% FBS) 5 With the caps loosened slightly, pass the reagents prepared m step 5, a small box of ice, the CellPro biotmylated anti-CD34 MAb, 5% BSA, and the avidm column mto the glove box chamber by evacuating the airlock to 115 m Hg, and then flushing with 5% 02, 7% CO2 88% N2 back to ambient atmospheric pressure three times From this point on everything entering the glove box must be cycled through the airlock m this manner (see Note 6)

3.2. HPC Isolation,

Phenotyping,

and Yield

1 In order to prevent coagulation, 4-6 mL of BM are aspirated mto a syringe contaming 500 U of PF heparm 2 Following aspiration, a I20-gage hypodermic needle IS immediately attached to the syringe, au IS forced out, the aspirate is inJected into a Vacutamer evacuated test tube without anticoagulant, and the specimen is transported to the Iaboratory 3. On receiving the BM m the laboratory, disinfect the exterior of the tube and introduce it into the glove box Except for centrifugations, overnight storage, 37°C mcubations, or unless otherwise noted, all further manipulations of the BM cells are conducted m the glove box 4 Transfer the aspirate to a 50-mL tube, and dilute to 50 mL with PBS/l0 U/mL PF heparm It IS important to obtain a cell count at this pomt m order to calculate HPC yield. In general, undiluted BM aspirates with low peripheral blood contammation have nucleated cell counts of 35-100 x 106/mL (see Note 7a) 5 Remove 3 x lo6 cells for later flow cytometric phenotypmg, and place on ice 6 Carefully layer 25 mL of the diluted aspirate onto the top of each FicollHypaque cushion, seal the tubes tightly, and centrifuge at room temperature for 10 mm at 1OOOg Carefully remove the BM mononuclear cells (BMMNCs) banded at the interface, and pool them m a new 50-mL tube. Dilute the BMMNCs to 50 mL with PBS, 5% FBS, and pellet at 4008 for 10 mm at room temperature Discard the supernatant, resuspend the pellet in 15 mL PBS, 5% FBS, transfer to a 15-mL tube, and pellet at 4008 at 4°C. From this point on, conduct all centrifugations at 400g at 4°C 10. Wash the BMMNCs once more with 15 mL PBS/l% BSA, remove the supematant, and place the cells on ice

Human Hemafoporet~c Progenitor Cells Table 2 CD34+ Cell Content Donor 1

2

BM fraction BM aspirate BMMNC CD34+ BM aspirate BMMNC CD34+

745

of BM Subpopulations Total cell number, x1o-sa

Percent CD34+b

Total CD34+ XlOdC

Percent yieldd

252 185 20 413 249 13

2.1 1.0 72 2.4 05 65

53 19 14 99 13 08

100

35 27 100

13 8

OTotal nucleated cell counts m unfractlonated 4-6 mL BM asplrates and BMMNC and CD34+ cell fractions from two normal donors were determmed using a Coulter electromc particle counter hThe percentage of CD34+ cells m each fraction was determmed by flow cytometry CThe total number of CD34+ cells was calculated from the percent CD34+ cells and the total cell numberm each BM fraction dThe yield of CD34+ cells m each BM fraction represents the percentage of the total number of CD34+ cells m the unfractlonated BM aspirate

11 The BMMNC recovery should be approx 50-75% of the origmal total cell count (Table 2) Therefore, resuspendthe cells in a volume of PBS/l% BSA that ~111 give 10s200 x lo6 cells/ml 12 Obtain a cell count again, remove 3 x lo6 cells again for later flow cytometric analysis, and store on ice. 13 Add 40 pL of the CellPro biotinylated anti-CD34 MAb/mL of BMMNCs, mix well, and incubateon ice for 30 mm, mixmg againafter the first 15min While the cells are incubating with the MAb, preparethe avidin column asmstructed by the CellPro protocol. 14. Dilute the labeled BMMNCs to 15 mL with PBS/l% BSA, and pellet at 400g for 10 mm at 4X 15 Wash once with 15 mL of PBS/l% BSA, and resuspendthe BMMNC pellet m CellPro 5% BSA at lo&200 x lo6 cells/ml. 16 Follow the CellPro instructionsto apply the BMMNCs to the column, washaway the unboundcells (CD34-) into one 15-mL tube, and mechanicallyreleaseand elute the bound CD34’ (pHPC) subpopulationfrom the column mto a new 15-mL tube 17 At this point, pellet the BM aspirate and the BMMNC fractions reserved for FACS analysis and the isolated CD34+ cells, remove the supernatant, and resuspend each in 1 mL of complete IMDM, 20% FBS. Add 20 U of PF heparm to the ahquot of unfractionated BM aspirateto prevent coagulation 18 Obtain a cell count of the pHPCs, and remove 50 x lo3 for flow cytometric analysis In general, approx 0 5-l% of the BMMNCs are isolated aspHPCs. 19 Store the pHPCs m sealedtubes overnight in an me bath m the refrigerator Storage of the cells m this manner does not result m any loss of HPC viability or colony-formmg ability

146 20 21

Cohnas To block nonspeclfic MAb bmdmg, add an equal volume of PBS, 0 1% NaN,, 1% BSA, 50 pg/mL mouse IgG to the BM fiactlons reserved for flow cytometric analysis After blocking for at least 4 h at 4°C and wlthout washing the cells, stain the BM cell fractions with anti-CD34-PE or fluorochrome-matched lsotype control MAbs and analyze by flow cytometry usmg standard methods Typical pHPC frequency rn the various BM cell populations and yield are shown m Table 2

3.3. Pretreatment

of pHPCs

The followmg morning, disinfect, restock, and flush the glove box wtth 5% O,, 7% CO,, and 88% N, agam Plan the followmg treatment groups m trlphcate untreated control, 50 ng/mL IL- 1p. 2.5 ng/mL TNFa, 100 nM DX, IL-I/DX, TNF/DX, and IL- l/TNF/DX Thaw IL- l/3 and TNFa on Ice, and prepare the followmg m the TC hood 10 fl DX stock by addmg 43 5 pL 2 mg/mL DX to 14 95 mL chllled salme, 10 mL complete IMDM/S% FBS, and 21 capped, sterile 1 9-mL polypropylene mlcrocentrifuge tubes Introduce the Itemspreparedtn step3 mto the glove box, and equilibrate for 1 5 h Pellet the pHPCs, resuspend them at 16 7 x 103/mL m the complete IMDM, 5% FBS, and ahquot 0 3 mL mto each 1 9-mL tube To the indicated tubes first add the IL-lb and TNFa, and then the DX to final concentrations of 50 ng/mL, 25 ng/mL, and 100 nA4, respectively, mlxmg well after each addltlon Cap the tubes and place them m a 37°C incubator mamtamedat 5% O,, 7% CO,, and 88% N, for 22 h

3.4. Removal

of the Cyfokines

and Toxicanfs

Since the cells are kept Ice-cold, and the cytokmes and DX are lmmedlately diluted and then washed away, Sections 3 4. and 3.5 can be performed m the TC hood. 1. Remove the tubes from the mcubator, and place them on ice m the TC hood 2. Add 1 5-mL ice-cold complete IMDM, 5% FBS to each tube, recap, mix by mverslon, and pellet the cells 3 To prevent cell lossdurmg the washmgprocedure, aspn-ateall but the last 100 PL of the supernatant 4. Wash the cells twice with 1 5 mL ice-cold complete IMDM, 5% FBS/tube, and store on Ice until bemg plated m semlsohdmedium

3.5. CFU-C Assay 1, Loosen the bottle cap, and melt the stock 0 9% LMP agarosem a boilmg water bath Then place it m a 40°C water bath to cool 2. While the agaroseIScoohng, label 35-n-mTC dishes,thaw the ahquotsof GM-CSF,

IL-3, and KL on ice and, m a 50-mL tube, combme 14 mL 2X IMDM, 7 mL FBS, and I7 5 pL 40 mg/mL gentamlcm sulfate, cap It tightly, and warm It to 40°C

Human Hematopoietlc

Progenrtor Cells

147

3 When both the agarose and the medium reach 40°C remove both from the water bath, and m the TC hood, quickly add 14 mL 0 9% agarose to the medium, mix, and then add GM-CSF, IL-3, and KL to 2, 5, and 20 ng/mL, respectively 4 To each 35-mm TC dish, first add 1 5 mL of medium, and then the resuspended cells from one tube, mmimizmg bubble production (see Notes 7c and d) 5 MIX well by swtrlmg the dishes m both directions, and place the dishes on an icecold leveled metal tray to solidify Following this procedure, plating can be done conveniently m sets of 8-l 0 dishes (see Note 7e) 6. When all the samples are plated and have solidified, place the dishes m a wellhunndrfied incubator contannng 5% 0,, 7% CO*, and 88% N, for 14 d (see Note 7f) 7 At this point, count colonies (CFU-C) of >50 cells using a dissectmg microscope at 40X wtth oblique backhghtmg

3.6. In Situ Fixation and Staining of CFU-C After counting CFU-C, fix the cells by adding 2 mL of 2% paraformaldehyde m PBS to the center of each dish, and rock gently for 15 mm at room temperature Carefully aspirate the fixative without dlsturbmg the agarose, rinse once with 3 mL PBS, and then wash the dishes twtce for 30 mm by adding 2 5 mL of PBS, rocking gently at room temperature Add 1 mL PBS, 0 1% BSA, 0 1% NaN,, 50 pg/mL of mouse IgC to each dish, and incubate for 24 h at 4°C For each experimental group of three dishes, add fluorochrome- and isotypematched control MAbs to one control dish, and anti-CD66b-FITC and anti-CDl4-PE to the two remaining replicate dishes Rock the dishes for 24 h at 4°C m the dark Remove unbound MAbs by rmsmg once and then washing 4 times for 8-12 h each with 2 5 mL cold PBS, 0 1% BSA, 0.1% NaN,, pH 8 0, at 4°C m the dark Enumerate CFU-G, CFU-M, and CFU-GM usmg a fluorescent microscope at 100X magmficatlon. Relative to the isotype controls, CFU-G are CD66b’i CD 14ncs’to, CFU-M are CD 14+/CD66b-, and CFU-GM contam both CD66b+ and CD 14+ cells 4.

Notes

1 The mhibitory effects of ambient p02 exposure and oxidative stress-mducmg treatments on both mouse and human in vitro colony-forming assays have been shown to be additive (6). Therefore, m studies designed to assess the effects of potentially oxtdizmg toxicants, either alone or m the presence of protective cytokmes, it is crmcal that these studies be conducted at physiological pOz 2 Additional HPC sources may be used, such as BM harvested under general anesthesia for transplantation purposes, umbilical cord blood, peripheral blood, or fetal liver However, it is important to remember that several subtle differences may exist between HPCs derived from different sources 3 The OxyReducer consists of an O2 sensor, which is placed m the mcubatoi, linked to an electronic controller that maintains a preset p0, by flushmg the incubator

148

Cohnas with N2 A hqutd N2 (LN2) tank capable of gas takeoff at pressures 220 psi (Taylor Wharton, Theodore, AL; model XL-45) 1s an economical prtmary N, source As a backup, connect a cylinder of compressed N, gas mto the N, line between the LN, tank and the OxyReducer, with the regulator outlet pressure set 10 psi lower than that of the primary NZ source The htgher back-pressure from the primary N, source will prevent gas use from the compressed N, tank until the LN2 tank 1s empty Endotoxm contammation IS a persistent concern m experiments assessing the biological effects of cytokmes. Thus, all reagents and supplies should be certtfied pyrogen-free Rinse glassware well with pyrogen-free, double-dtsttlled deionized HzO, dry, and bake for 2 h at 180°C If any doubts about endotoxm remam, determme levels using the Ltmulus Amoebocyte Lysate Assay kit from Biowhittaker The glove box chamber cannot withstand a pressure differential greater than a few psi Watch carefully to avoid pressure damage All soluttons should be equtllbrated m the low-O2 envtronment for at least 1 5 h before being used Troubleshooting a Low total BM cell count On occasion, a BM aspirate will have a very low cell count (cl00 x 1O6total nucleated cells) In these cases, it should be determined whether there are enough CD34’ cells to perform the experiment To make this determmatton, stam an ahquot of BMMNC with HPCA-2-PE, and analyze by flow cytometry. BMMNC containing less than three times the absolute number of CD34+ cells needed should not be used. b High varrabrlrty between CFU-C counts m rephcate dishes The most hkely source for this sort of variabtllty is from the loss of cells during cytokme/ toxtcant removal Smce there are only 5000 cells/sample, tt is very difficult to detect cell loss Therefore, to prevent loss of cells, follow the memscus down as the supematant IS asptrated, and leave the last 100 uL m each tube c Semisolid medtum soltdiftes before completion of platmg. To prevent the semisolid medium used in the CFU assay from solidtfying before all the samples are plated out, set it in a small 37°C water bath mstde the TC hood d Microscopic observation of the CFU-C is distorted. Although distortion cannot be avoided around the edge, bubbles produced during the platmg procedure ~111 also cause optical distortions. Keep bubblmg to a mmtmum e All the CFU-C have developed on the bottom of the dash: Thts results from the cells settling to the bottom of the dish before the agarose has soltdified Thoroughly resuspend the contents of the dishes unmediately before setting them on the chilled metal tray f The dishes dry out durmg the 14-d culture pertod It 1svery important to mamtam high humidity m the incubator This is most easily accomplished by bubbling the N2 m the incubator through a beaker of sterile water and placing several evaporation pans m the mcubator.

Human Hematopoietlc Progenitor Cells

149

8 This protocol can be easily adapted to assess the effects of any cytokme or mhibitory treatment on any identtfiable HPC However, it is important first to determine appropriate cytokme and toxicant concentrations, as well as the kmettcs of their effects For example, the mhibitory effects of DX are manifest more slowly than the IL-1 p- or TNFa-mediated protective effects As a result, both cytokmes and toxtcants can be added to the pHPCs at the time of culture mmatton In contrast, the mhibnory effects of agents, such as 4HC and y-radiation, occur very quickly Therefore, the HPCs must be pretreated with IL- 1j3 or TNFa for at least 14 h before being exposed to either 4HC or y-radiation

Acknowledgments The research from which this protocol was derived was done m the laboratory of David A. Lawrence and n-r the Molecular Immunology Core Facthty at the Wadsworth Center of the New York State Department of Health. Funding for this research was provided by the NIH National Research Service Award ES05538 (R. J Colmas) and NIH grant ES05020 (D. A. Lawrence) from the National Institute of Environmental Health Sciences.

References 1 Bradley, T R. and Metcalf, D (1966) The growth of mouse bone marrow cells zn wtro J Exp Blol Med 44, 287-300 2 Dexter, T M , Allen, T D , and LaJtha, L G (1977) Conditions controlling the proliferation of haemopoietic stem cells zn v&-o J Cell Physlol 91,335-344 3 Cater, D B and Silver, I A (1960) Quantitative measurements of oxygen tension m normal tissues and in tumors of pattents after radiotherapy Acta Radzol 53, 233-241 4 Rtch, I N (1986) A role for the macrophage m normal hemopotesis II. Effect of varying physiologtcal oxygen tensions on the release of hemopoletic growth factors from bone-marrow-derived-macrophages zn vztro Exp Hematol 14, 746-75 1 5. Smith, S and Broxmeyer, H. E. (1986) The influence of oxygen tension on the long term growth zn vztro of haematopotettc progenitor cells from human cord blood Br J Haematol 63,29-34 6 Colmas, R J , Burkart, P T., and Lawrence, D. A (1994) In vztro effects of hydroquinone, benzoqumone, and doxorubicm on mouse and human bone marrow cells at physiological oxygen partial pressure Toxzcol Appl Pharmacol 129,95-102 7 Fibbe, W. E. and Wtllemze, R. (199 1) The role of mterleukm- 1 m hematopolesls. Acta Haematologxa 86, 148-154 8. Trmchreri, G. (1992) Effects of TNF and lymphotoxm on the hematopotettc system Immunol Ser 56,289-3 13 9 Muench, M 0 , Scheider, J. G., and Moore, M A S. (1992) Interactions among colony-sttmulatmg factors, IL- l/3, IL-6, kzt-hgand m the regulation of primative murme hematopoietic cells Exp Hematol 20,339-349

150

Colmas

10 Kobayasht, M , Imamura, M , Gotohda, Y , Maeda, S , Iwasaki, H , Sakurada, K , Kasal, M , Hapel, A J , and Mryazakl, T (1991) Synergistic effects of mterleukm1I3 and mterleukm-3 on the expansion of human hematopoietic progenitor cells m liquid cultures Blood 78, 1947-l 953 11 Khoury, E , Lemome, F M , Baillou, C , Kobari, L , Deloux, J , Gmgon, M , and NaJman, A (1992) Tumor necrosts factor alpha m human long-term bone marrow cultures distinct effects on nonadherent and adherent progemtors Exp Hematol 20,99 l-997 12 Beran, M , O’Brian, S , Gutterman, J U , and McCredie, K B (1988) Tumor necrosis factor and human hematopolesis I Kmetics and dlversuy of human bone marrow cell response to recombmant tumor necrosis factor alpha m short-term suspension cultures m vitro Hematol Path01 2,3 l-42 13 Akahane, K , HOSOI, T , Urabe, A , Kawakami, M , and Takaku, F (1987) Effects of recombinant tumor necrosis factor (rhTNF) on normal human and mouse hematoporetic progenitor cells Int J Cell Clomng 5, 16-26 14 Caux, C , Saeland, S , Favre, C , Duvert, V , Mannom, P , and Banchereau, J (1990) TNF alpha strongly potentiates IL-3 and GM-CSF induced proliferation of human CD34+ hematopoiemc progenitor cells Blood 75, 2292-2302 15 Backx, B , Broeders, L , Bot, F J., and Lowenburg, B (1991) Posmve and negative effects of tumor necrosis factor on colony growth from highly purified normal marrow progemtors Leukemza 5, 66-70 16 Caux, C , Favre, C , Saeland, S , Duvert, V , Durand, I , Mannom, P , and Banchereau, J (1991) Potentiation of early hematopotesis by tumor necrosis factor-a is followed by mhtbttion of granulopoietrc differentiation and proliferation Blood 78,635S644 17 Neta, R , Douches, S D , and Oppenheim, J J (1986) Interleukm 1 IS a radioprotector J Immunol 136,2483-2485 18 Constme, L S , Harwell, S , Keng, P , Lee, F , Rubm, P , and Siemann, D (1991) Interleukm 1 alpha stimulates hemopolesrs but not tumor cell proliferation and protects mice from lethal total body trradtation Znt J Radzatzon Oncology Bzol Phys 20,447-456 19 Zucali, J R., Moreb, J , Gtbbons, W., Alderman, J , Suresh, A., Zhang, Y , and Shelby, B (1994) Radioprotection of hematopoietlc stem cells by mterleukm- 1 Exp Hematol 22, 13@-135 20. Sheppard, B C , Fraker, D L , and Norton, J A. (1989) Prevention and treatment of endotoxm and sepsis lethalny with recombmant human tumor necrosis factor Surgery 106, 156-162 21 Vogel, S N , Kaufman, E N , Tate, M D , and Neta, R (1988) Recombmant mterleukm-la and recombmant tumor necrosis factor cx synergize m vlvo to induce early endotoxm tolerance and assoctated hematopoietic changes Infect Immun 56,2650-2657 22. Eddy, L J., Goeddel, D V , and Wong, G. H. W (1992) Tumor necrosis factor-a pretreatment 1s protective m a rat model of mycardtal tschemia-reperfusron inJury Blochem Blophys Res Commun 184, 10561059

Human

Hematopoletlc

Progenitor

Cells

151

23 Brown, J M , White, C W , Terada, L S., Grosso, M A, Shanley, P F , Mulvm, D W , BanerJee, A , Whttman, G J R , Harken, A. H , and Repme, J E (1990) lnterleukm 1 pretreatment decreases tschemta/reperfusion mJury Proc Nut1 Acad Scz USA 81,5026-5030 24 Slordal, L , Warren, D J , and Moore, M A S (1990) Protective effects of tumor necrosis factor on murme hematopotests during cycle-specific cytotoxtc chemotherapy Cancer Res 50.42 1W220 25 Evans, M J , Kovacs, C J , Gooya, J M , and Harrell, J P (1991) Interleukm-la protects against the toxtctty associated with combined radiation and drug therapy Int J Radzat Oncol El01 Phys 20,303-306 26 Eppstem, D A , Kurahara, C G , Bruno, N A , and Terrell, T G (1989) Prevention of doxorubtcm-induced hematotoxtctty m mice by mterleukm 1 Cancer Res 49,3955-3960. 27 Lynch, D. H , Rubm, A S , Miller, R E., and Wtlltams, D. E (1993) Protective effects of recombmant mterleukm- 1a m doxorubtcm-treated normal and tumorbearing mice Cancer Res 53, 1565-l 570 28 Damta, G , Komschhes, K L , Futamt, H , Back, T , Gruys, M E , Longo, D L , Keller, J R , Ruscettt, F W , and Wtltrout, R H (1992) Prevention of acute chemotherapy-induced death m mice by recombmant human mterleukm 1 protectton from hematologlcal and nonhematologlcal toxlcltles Cancer Res 52,4082-4089 29 Tsan, M F , White, J E , Santana, T A., and Lee, C Y (1990) Tracheal msufflation of tumor necrosis factor protects rats agamst oxygen toxtctty J Appl Physzol 68, 1211-1219 30 White, C W , Ghezzt, P , Dmarello, C A, Caldwell, S A, McMurtry, I F , and Repme, J. E. (1987) Recombmant tumor necrosis factor/cachectm and mterleukm 1 pretreatment decreases lung oxtdtzed glutathtone accumulatton, lung inJury, and mortality m rats exposed to hyperoxta J Clm Invest 79, 1868-1873 3 1. Moore, M A S , Muench, M 0 , Warren, D. J , and Laver, J (1990) Cytokme networks involved m the regulation of haemopotettc stem cell proltferation and dtfferenttatton, m Molecular Control of Haemopolesls (Bock, G and Marsh, J , eds ), Wiley, New York, pp 4361 32 Moreb, J , Zucah, J R., Gross, M A., and Weiner, R S. (1989) Protective effects of IL-1 on human hematopotettc progenitor cells treated In vztro with 4-hydroperoxycyclophosphamide J Immunol 142, 1937-l 942 33. Moreb, J , Zucah, J R., and Rueth, S (1990) The effects of tumor necrosis factor-a on early human hematopotettc progenitor cells treated with 4-hydroperoxycyclophosphamide. Blood 76,68 1689 34. Colinas, R J , Burkart, P T., and Lawrence, D. A (1995) The effects of mterleukmll3 and tumor necrosis factor-a on zn wtro colony formatton by human hematopoletic progemtor cells exposed to doxorubtcm or hydroqumone Exp Hematol in press 35 Eastgate, J , Moreb, J , Nick, H S , Suzuki, K , Tamgucht, N , and Zucah, J R (1993) A role for manganese superoxtde dtsmutase m radtoprotectton of hematopotettc stem cells by mterleukm- 1 Blood 81,639-646

152

Colinas

36 Wong, G H W and Goeddel, D V (1988) Inductton of manganous superoxide dtsmutase by tumor necrosis factor possible protective mechanism. hence 242, 94 l-944 37 Ktzaki, M , Sakashtta, A., Karmaker, A , Lm, C W , and Koeffler, H P (1993) Regulation of manganese superoxtde dismutase and other anttoxidant genes m normal and leukemic hematopoiettc cells and their relattonship to cytotoxictty by tumor necrosis factor Blood 82, 1142-l 150 38 Manthey, C L , Landkamer, G J , and Sladek, N E (1990) Identtficatton of the mouse aldehyde dehydrogenases important m aldophosphamide detoxification Cancer Res 50,4991-5002. 39 Moreb, J., Zucah, J R., Zhang, Y , Colvm, M 0 , and Gross, M A (1992) Role of aldehyde dehydrogenase m the protectton of hematopotettc progemtor cells from 4-hydroperoxycyclophosphamtde by interleukm 1p and tumor necrosis factor Cancer Res. 52,177&1774 40 Civm, C I , Strauss, L C , Brovall, C , Fackler, M J , Schwartz, J F , and Shaper, J H. (1984) Anttgemc analysts of hematpotests III A hematopotettc progenitor cell surface antigen defined by a monoclonal antibody raised against KG-la cells J Immunol 133, 157-165

13 Human Mononuclear in Tissue Culture

Phagocytes

Yona Keisari

1. Introduction Peripheral blood human monocytes (HuMo) are the major source for human mononuclear phagocytes. Such monocytes, when cultured, differentiate mto monocyte-derived macrophages (MoDM), and undergo various structural, blochemtcal, and functional changes. The most common method used for the separatton of mononuclear cells (MNC) from the blood IS Ficoll-Hypaque density gradtent centrtfugatron, essentially described by Boyum (1). Ftcoll-Hypaque at a density of 1.077 g/L IS used to separate the denser granulocytes and erythrocytes from the lighter lymphocytes, monocytes, and thrombocytes. The mononuclear cells stay at the top of the Ftcoll-Hypaque layer, whereas the denser cells smk to the bottom of the centrifuge tube. Peripheral blood monocytes are purified from the mononuclear fraction by adherence to plastic. Adherence can be carried out either directly onto ttssueculture plates m which they will be grown further (24- or 96-well plates), or onto trssue-culture flasks from which they will be removed and recultured m the required plates or chambers (2). When an enrrched monocyte cell suspension is reqmred, MNC harvested on Flcoll-Hypaque gradients can be further separated on Percoll gradients into lymphocytes and monocytes The method mittally descrtbed by Ulmer and Flad (3) was modified by Orlandt et al. (4), which used only one Percoll concentratron. After separation on Percoll, the monocytes that are less dense than lymphocytes stay on top of the Percoll layer, whereas the lymphocytes go through to the bottom of the tube. From

Methods in Molecular Medrone Human Cell Culture E&ted by G E Jones Humana Press Inc , Totowa,

153

frotocols NJ

154

Keisan

Long-term mcubatlon of monocytes under tissue-culture condrtlons results m the dlfferentlatlon of the cultured cells, and m the appearance of monocytederived macrophages (5) Long-term mcubatlon of the cells m culture results m a substantial loss of cells In various studies, it was found that granulocytemacrophage colony-stlmulatmg factor (GM-CSF) or IL-3 (6IO), as well as PKC activators/tumor promoters (1 I, 12) can be used to facilitate the survival of MoDM. The adherent human monocytes bmd firmly to plastic substrata, and it 1s very difficult to remove the cells for quantitative measurements There are several methods to enumerate or quantltate adherent monocytes/macrophages, of which four are described

2. Materials

5 6 7 8 9 10 11 12 13 14 15 16 17 18

Earle’s balanced salt solution (EBSS) Dulbecco’s PBS without Ca*’ and Mg*+ Flcoll-Hypaque (density 1 077 g/L) RPMI- 1640, supplemented with 100 pg/mL streptomycin 100 U/mL penlclllm, 300 pg/mL (2 mM) L-glutamine Newborn bovine serum (NBS) or pooled human AB serum (HABS), heat-mactlvated (56’C, 30 mm) Percoll (Pharmacla LKB Biotechnology AB, Uppsala, Sweden) 10X EBSS Human recombinant GM-CSF and IL-3 Phorbol retmoyl acetate (PRA) 120Tetradecanoyl13-phorbol acetate (TPA) Mezerem (MEZ) Blo-Rad Protein Assay, cat no 500-0006, Blo-Rad Laboratories (Munich, Germany) Dilute 1.5 in double-distilled H20, and filter before use Prepare fresh for each assay Hemacolor@ color reagents (Merck, Darmstadt, Germany), or Dlff-Qulk@ reagents (Harleco, Gibbstown, NJ) 0 5% SDS dissolved m double-distilled H,O Methanol Isopropanol(2-propanol) for analysis MTT (3-[4,5-dimethyl-thlazol-2-yl]-2,5-dlphenyl tetrazohum bromide) (5 mg/mL in PBS) Llmulus Amebocyte Lysate (LAL) reagent (Pyrogent, Whittaker, M.A , Bloproducts Inc , Walkersvllle, MD) (see Note 1)

3. Methods 3.1. The Isolation

of MNC (see Note 2)

Peripheral blood human monocytes can be separated from heparmized blood samples or from the buffy coats of normal blood bank donors with sodium

Mononuclear

Phagocytes

155

citrate as an anticoagulant. Blood bank buffy coats are obtained after centrlfugatlon of blood bank bags (ortgmal volume of 400 mL) and removal ofthe upper layer of the content. The separated fiactlon (25-40 mL) contains more than 90% of the leukocytes of the blood donation, 10% of the erythrocytes, and 5% of the plasma

4 5

6 7 8 9 10

Dilute buffy coats 1 4 (v/v) or heparmlzed blood samples 1 2 (v/v) with Dulbecco’s PBS without Ca2+ and Mg2+ Add 40 mL of the diluted blood to 50-mL conical polypropylene tubes Add 10 mL of Flcoll-Hypaque to the bottom of each tube To prevent mixing, insert a sterile Pasteur plpet mto the diluted blood down to the bottom of the tube, and Inject the Flcoll-Hypaque into the plpet by a lo-mL syringe (0 8 x 40 mm needle) Centrifuge the tubes at 700g for 30 mm at room temperature with brakes off Remove the cellular fraction that 1son top of the Flcoll-Hypaque layer, and transfer to a new 50-mL conical tube For blood bank buffy coats from the same donor, pool cells from four tubes mto one Wash three times with 50 mL cold Dulbecco’s PBS without Ca2’ and Mg2+ under the followmg condmons. (a) once at 350g for 10 mm, and (b) twice at 2308 for 7 mm Resuspend the cells m supplemented RPMI- 1640 containing 10% serum Place the tube for 16 h at 4’C to allow for separation of the MNC (cell pellet) from the thrombocytes (fluid suspension) Discard the fluid gently by low-rate aspiration, and resuspend the cells m 10 mL cold supplemented RPMI- 1640 + 2% serum For cell counts, dilute first 1’ 10 m supplemented RPMI-1640, and then 1 2 with 0 1% Trypan Blue solution, and count the cells with a hemocytometer The average yield 1s 540 + 130 x lo6 MNC (range 40~770)/ongmal400 mL blood

3.2. The Isolation of Mononuclear Phagocytes 3.2.1. Separation of Monocytes by Adherence m Mxrotiter Plates 1 Prepare MNC at 2-3 x 10’ cells/ml m supplemented RPMI- 1640 + 2% serum 2 Add 0 1 mL of the cell suspension to each well (2-3 x lo6 cells/well) of a 96-well tissue-culture plate 3 Incubate for 30 mm at 37’C and 7 5% CO, 4 Remove nonadherent cells by washing the wells three times with warm (37°C) EBSS For this purpose, use a 5- or lo-mL syringe (wlthout a needle) 5 Add 0 2 mL of supplemented RPMI- 1640 + 10% serum. The resulting monolayers contams 2 43 f 0 22 to 3 04 + 0.4 x lo5 cells/well

3.2.2. Separation of Monocytes by Adherence In Tissue-Culture Flasks 1 2 3 4

Resuspend MNC at 5 x 106/mL m supplemented RPMI- 1640 + 2% serum Add 5 mL of the cell susptnslon to 25-cm2 or 20 mL to 75-cm2 tissue-culture flasks Incubate for 30 mm at 37°C and 7 5% CO2 Remove nonadherent cells by washmg the flasks extensively three times with warm (37°C) EBSS

156

Keisari

5 Add 5 or 25 mL of supplemented RPMI-1640 + 10% serum to the 25- or 75-cm2 tissue-culture flasks, respecttvely. 6 Incubate for 16 h at 37°C and 7 5% CO,. 7 Shake the flasks firmly, and remove the medmm that contams the detached monocytes To remove stall adherent cells, wash the flasks with cold EBSS using a lo-mL syrmge and a 0.8 x 40 mm needle 8 Centrtfuge the cells at 350g for 10 mm, and resuspend the cells m supplemented RPMI-1640 + 10% serum 9 Culture 2-3 x lo5 cells/well/O 2 mL m 96-well plates, or 5 x lo5 cells/well/ 0 5 mL in 24-well plates

3 2.3. Separation

of Monocytes

on an Iso-Osmotic

Percoll

Gradient

1 Prepare tso-osmotic supplemented RPMI-1640 + 10% serum by admstmg the osmolahty to 285 mosM 2 Prepare tso-osmottc Percoll solution by mtxmg Percoll wtth EBSS (10X) 93 7 (v/v), and adjust to 285 mosM 3 Prepare a 46% solutton of tso-osmottc Percoll with tso-osmottc RPMI- 1640 4 Resuspend the mononuclear fraction obtained after Ftcoll-Hypaque separation m tso-osmottc RPMI- 1640 at 5-l 0 x 1O6 cells/ml 5 Add 5 mL of the cell suspension to 10-12 mL tubes 6 Add 5 mL 46% Percoll solution to each tube using a Pasteur ptpet as described m Section 3 1 for Ftcoll-Hypaque 7 Centrifuge at 600g for 30 mm at room temperature with brakes off 8 Remove Interface with a sterile plpet, and wash the cells twice with supplemented RPMI- 1640 9 Count cell number 10 The recovery 1s 11 3 f 3% (range S-18%) of the MNC fraction, and 86 f 6% of the cells are monocytes (range 77-95%)

3.3. Long-Term 3 3.1. MoDM

Cultures Cultured

of MoDM

in the Presence

of Colony-Stimulating

Factors

1 Prepare cultured monocytes m 96- or 24-well plates as described (see Note 3) 2 Add 50-200 U/mL GM-CSF or IL-3 to the monocyte cultures (see Note 4) 3 Incubate without changing the medium for at least 10 d to obtain human monocyte-derived macrophages (HuMoDM) 4 If extended incubation periods are required, add the indicated amount of CSF every 2 wk by replacing half of the volume of the culture medium

3.3.2. MoDM Cultured of PKC Actwators/Tumor

In the Presence Promoters (see Note 5)

1. Prepare cultured monocytes m 96- or 24-well plates as described (see Note 3) 2 Add TPA, MEZ, or PRA to monocyte cultures at a final concentratton of 2-5 nA4 (see Note 6) 3 Incubate, without changing the medium, for at least 10 d to obtain HuMoDM

157

Mononuclear Phagocytes

4 If extended mcubatton pertods are required, add the mdtcated amount of PKC activators every 3 wk by changing half of the volume of the culture medium

3.4. General Methods for the Quantitation of Cultured Adherent Mononuclear Phagocytes 3.4.1. Cell Count 1 Remove the culture medmm from the cultured monocytes 2 Add ice-cold PBS (0 2 mL) without Ca*+ and Mg*+ 3 Scrape gently the adherent cells, and count with a hemocytometer

3.4.2. Protein Determinahon

(see Note 8)

1. Wash the cultured cells extensively with EBSS to remove medium and serum 2 Lyse the cells wrth 200 pL of 0 1% Triton X- 100 for 30 mm at 37°C. 3 Add 20-uL samples of the cell lysates to 200 pL of Bto-Rad reagent m 96well plates 4 Incubate the samples for 15 mm, and read at 600 nm in an automated spectrophotometer Protem concentration IS determined using a standard curve of BSA (5-1000 ug/mL)

3.4.3. Hemacolor Colonmettx

Microtiter Assay (see Note 9)

1 Remove the supernatant from cell monolayers cultured m 96-well plates, and dry the cells qutckly m the an 2 Fix the monolayers with methanol (50 pL/well) for 30 s (do not rinse the wells between steps 2 and 4) 3 Add 80-100 pL/well of Hemacolor Reagent 2 for 60 s 4. Add 80-100 pL/well of Hemacolor Reagent 3 for another 60 s. 5 Rinse the plates three times with tap water. 6 Fill again with water and decolorize for 5 min 7 Remove the water, and dry the plates extensively (Followmg this step, the stamed cultures can be kept for several weeks m the dark.) 8 For stain extraction, add 0 2 mL/well of SDS (0.5%) dtssolved in double-dtstilled Hz0 for at least 90 mm 9. Measure OD at 600 or 630 nm with an automated microplate reader.

3.4.4. MTT Assay (see Note 10) 1, Remove the culture medium from cell monolayers cultured m 96-well plates. 2 Reconstitute each well with 0 2 mL supplemented RPMI- 1640 contammg 1 mg/mL MTT 3 Incubate the cultures for 2-4 h at 37°C. 4 Remove the supematants from the wells 5 Add 0.2 mL/well of lysmg reagent contammg 0 04N HCI m isopropanol 6. Mix the wells thoroughly 7 Read the plates m an automated mtcroplate spectrophotometer at 570 and 630 nm as reference.

Kelsan

158 3 4 5. Alternatrve

MTT Method

1 If the cultured cells are not tightly adherent and mtght be removed with the supernatant, tt IS recommended to remove only 0 1 mL of the culture medmm from the cells cultured m 96-well plates 2 Add to each well 0 025 mL of 5 mg/mL MTT m PBS 3 Incubate the cultures for 24 h at 37°C 4 Add 0 1 mL/well of lysmg reagent containing 0 04N HCl m tsopropanol 5 MIX the wells thoroughly 6 Read the plates m an automated microplate spectrophotometer at 570 and 630 nm as reference

4. Notes All the media and buffers used should be assayed for the presence of bacterial endotoxm by the Gel-clot technique (13) using the Lrmulus Amebocyte Lysate (LAL) reagent Reagents should be used only tf no detectable Lrpopolysaccharide (LPS) 1s found (sensttrvrty-0 064 Endotoxm U/mL) Steps l-5 of this procedure are carried out at room temperature (T,), and all the reagents should be at T, The use of cold reagents should be avoided at this stage of the separation For extended mcubatron periods (more than 4 d), rt 1s recommended not to culture cells m the wells at the periphery of the plates. These wells should be filled with sterile water to the top to reduce evaporatron of ltqutd from the cultures The optimal effect of CSF was achieved when added on the first day of culture After 6 d m culture, the cells did not respond to the addttron of CSF, and they behaved as nontreated cells Mrcroscoprc observations of MoDM obtained m the presence of CSF revealed a homogenous populatton of large spread-out cells, whereas nontreated cultures were more heterogeneous m then appearance and some small round cells were also apparent (6,7) Adherent HuMo cultured m 96-well plates showed a substantial loss (51%) of adherent cells m nontreated monocyte cultures after 2 wk of mcubatton. In comparison, HuMoDM cultures treated with vartous PKC acttvators/tumor promoters lost only O-26% of the cells after incubation for 2 wk (I 1,12). Prepare stock soluttons of PRA, TPA, and MEZ m DMSO at 10 w, and store m the dark at -20°C When diluting the reagents m culture media before adding to the cells, the final concentratton of DMSO should not exceed 0 1% In our laboratory, we mamtamed MoDM cultures for 4 mo by adding TPA (2 nk?) every 3 wk Determination of protein concentratron of cells cultured m 96-well plates IS according to the Bradford method (14) The Hemacolor colortmetrtc mtcrotrter assay (15) uses reagents generally used to stam blood cells and cells m tissue cultures The stammg ktt holds three soluttons Solution 1 contams methanol for fixation, solution 2 contains a xanthene dye (orange color), and solutron 3 1s a Thtazme solution contammg a mtxture of

Mononuclear Phagocytes

159

azure I dyes and methylene blue (blue reagent) A spectrophotometrlc analysis of a mixture of solutions 2 and 3 m SDS 0 5% revealed a peak of absorption at 5 I 7 nm caused by the xanthene dye, and a second peak at 634 nm caused by the thlazme solution Measurements of stained cells are carried out in an automated mlcroplate reader using 630- or 600-nm filters Hemacolor reagents can be substltuted by Dlff-Qulk reagents that serve a slmllar purpose 10 The MTT assay is based on the observation that tetrazolmm salts are reduced to formazan by cellular respiratory enzymes This actlvtty 1s performed only by viable cells, and thus the method may mdlcate the amount of viable cells present m culture The method was mltlally described by Mosmann (16) for MTT, but other tetrazohum reagents may also be used (I 7)

References 1 Boyum, A (1968) Isolation of mononuclear cells and granulocytes from human blood Scatld J Ch Lab Invest tl(Suppl. 97), 77-89 2 Treves, A J , Yagoda, D , Halmovltz, A , Ramu, N , Rachmllewltz, D , and Fuks, Z (1980) The lsolatlon and furlfication of human peripheral blood monocytes m cell suspension J Immunol Methods 39,7 l-80 3 Ulmer, A J and Flad, H -D (1979) Dlscontmuous density gradlent separation of human mononuclear leukocytes usmg Percoll as gradlent medmm J Immunol Methods 30, 1-l 0 4 Orlando, M , Bartolml, G , Chlrlcolo M , Mmghettl, L , Franceschl, C . and Tomas], V (1985) Prostaglandm and thromboxane blosynthesls m isolated platelet-free human monocytes I A modified procedure for the characterlzatlon of the prostaglandin spectrum produced by restmg and activated monocytes ProJtaglandlns, Leukotrzenes Med 18, 205-2 16 5 Zuckerman, S H , Ackerman, S K , and Douglas, S D (1979) Long-term human peripheral blood monocyte cultures* establishment, metabolism and morphology of primary human monocyte-macrophage cell cultures Immunology 38,401-4 11 6 Robin, G , Markovlch, S , Athamna, A , and Kelsan, Y (1991) Human recombtnant granulocyte-macrophage colony stlmulatmg factor augments the vlablhty and cytotoxlc actlvltles of human monocyte derived macrophages m long term cultures Lymphoklne and Cytokwte Res 10,257-263 Dlmn, R , Nlsslmov, N , and Kelsan, Y (1994) Effect of human recombinant granulocyte-macrophage colony stlmulatmg factor and IL-3 on the expression of surface markers of human monocyte derived macrophages m long term cultures Lymphokme and Cytokwze Res 14,237-243 Elliot, M. J , Vadas, M A , Eglmton, J. M , Park, L S , Blk To, L , Cleland, L G , Clark, S. C , and Lopez. A F (1989) Recombinant human mterleukm-3 and granuIocyte-macrophage colony-stlmulatmg factor show common bIologIca effects and bmdmg characteristics on human monocytes Blood 74, 2349-2359 Markowlcz, S and Engleman, E G (1990) Granulocyte-macrophage colonystlmulatmg factor promotes dlfferentlatlon and survival of human peripheral blood dendrmc cells m vitro J Clm Znvest 85, 955-961

160

Keisari

10 Elschen, A , Vincent, F , Bergerat, J P , LOUIS, B , FaradJl, A, Bohbot, A, and Oberlmg, F (1991) Long term cultures of human monocytes m vitro Impact of GM-CSF on survival and dlfferentlatlon J Immunol Methods 143, 209-22 1 11 Kelsari, Y , Bucana, C , Markovlch, S., and Campbell, D E (1990) The mteractlon between human peripheral blood monocytes and tumor promoters Effect on m vitro growth, dlfferentlatlon and function J Bzol Response Modzf 9,40 14 10 12 Markovlch, S , Kosashvllh, D , Raanam, E , Athamna, A , O’Bnan, C A , and Kelsari, Y (1994) Tumor promoters/protem kmase C activators augment human peripheral blood monocyte maturation m vitro &and J Immunol 39, 39-44 13 Yin, E T., Galanes, C., Kmsky, S., Bradshaw, R , Wessler, S , and Ludentz, 0 (1972) Picogram-sensltlve assay for endotoxm Gelatlon of hmulus polyuphemus blood cell lysate Induced by purified hpopolysaccharlde and lipid A from gramnegative bacteria Bzochzm Bzophys Acta 261,284-289 14 Bradford, M M (1976) A rapid and sensltlve method for the quantrtatlon of mlcrogram quantltles of protem utlltzmg the principle of protein-dye bmdmg Anal Bzochem 72,248-254 15 Kelsan, Y (1992) A colorlmetrlc microtiter assay for the quantltatlon of cytokme actlvlty on adherent tissue culture cells. J Immunol Methods 146, 155-161 16 Mosmann, T (1983) Rapid colorlmetrlc assay for cellular growth and survival Apphcatlon to prohferatlon and cytotoxlclty assays J Immunol Methods 65,55-63 17 Alley, M C , Scudlero, D A , Monks, A , Hursey, M L , Czerwmskl, M J , Fme, D L , Abbott, B J , Mayo, J G , Shoemaker, R H , and Boyd, M R (1988) Feaslblhty of drug screening with panels of human tumor cell lines using a mlcroculture tetrazohum assay Cancer Res 48, 589-60 1

Purification

of Peripheral

Blood Natural Killer Cells

Ian M. Bennett and Bite Perussia 1. Introduction The ability to perform biologrcal studies on Natural Killer (NK) cells requires effective methods for their isolation from hematopoietic cells of other lineages. NK cells are a discrete lymphocyte subset distinguishable from B- and T-lymphocytes on the basis of both physical and phenotypic characteristics that can be exploited for then purification. Techniques based on dtfferential cell buoyancy (centrifugation on discontinuous density gradients, such as Percoll [I]) have been used to enrich NK cells from mixed lymphocyte populations, but do not allow purification of these cells to homogeneity. The mononuclear cell suspensions obtained, although enriched in NK cells, also contam variable proportions of other cell types (notably monocytes and/or activated T- and B-lymphocytes) (2) and subsetsof NK cells of higher density are lost m these preparations. The most satisfactory purification techniques for NK cells, as well as for other leukocyte subsets, rely on then distinctive phenotype and make use of monoclonal antibodies (MAbs) directed to lineage-specific surface antigens (Ag). Although NK cell-specific surface markers have not been identified yet, lack of surface expression of T-cell receptor/CD3 complex and surface Ig, and expression of CD 16 (low-affinity receptor for the Fc portion of IgG, FcyRIIIA) (3) and CD56 (and N-CAM isoform) (4) serve to identify NK cells wrthm mononuclear cell populations. MAbs to both antigens are available, and cells sensitized with them can then be detected with a variety of secondary reagents to permit their identification and physical separation. Using this approach, homogeneous preparations of NK cells are isolated from mixed mononuclear cell populations following either of two schemes: direct isolation of NK cells using MAbs to surface Ag expressed on these cells (positive selection) and From

Methods Ed&d

m Molecular by G E Jones

Me&me Humana

161

Human Cell Culture

Protocols

Press

NJ

Inc , Totowa,

162

Bennett

and Peruss~a

depletion of all cells other than NK using a mixture of MAbs directed to Ag expressed on the former, but absent from the latter population (negative selection) The advantage of positive selection is the rapid and specific isolation of NK cells However, Ab bmdmg to antigens capable of signal transduction, like CD16, leads to modulation of NK cells’ btological functions (5,6), making them unusable for certain applications Negative selection techniques, Instead, yield cells m their least altered state that may be most suitable for most functional studies Choosmg between the two systems depends on the specific experimental requirements Here we describe m detail the use of a dependable and relatively mexpensive method of isolatmg NK cells (indirect antiglobulm resetting) that is suitable for both positive and negative selectton, results m good yields, and allows easy and rapid manipulation of large numbers of cells Indirect antiglobulm resetting is based on the use of erythrocytes (E) coated with anti-mouse Ig Ab as a secondary reagent to detect cells that have murme MAb bound at their surface which recognizes lineage-specific Ag, leading to the formation of rosettes The subsequent physical separation of Ag+ (rosetted) and Ag(nonrosetted) cells is obtained by simple centrifugation on density gradients Other reliable techniques exist that use secondary reagents coupled to different detection systems but, unlike mdirect antiglobulm resetting, may not be practical for all mvestigators owmg to unavailability of specialized equipment, low yields, or prohibitive costs For example, fluorochrome-labeled secondary reagents and fluorescence-activated cell sortmg (7) are used to purify highly homogeneous NK cell populations This requires availability of a flow cytometer, IS time and money consummg, and has the disadvantage of allowmg recovery of relatively low numbers of cells, methods using magnetic beads (8), although fast and efficient are extremely expensive; panning the Ab-sensitized cells on dishes coated with antimouse Ig (7) is efficient, fast, and economical but may become impractical when large numbers of cells need to be processed, complement (C)-dependent lysts (7), which is practtcal and effictent, can be performed only with C-fixing MAb and may result m nonspecific toxicity and, consequently, the need for screenmg numerous batches of sera for optimal use These methods (described accurately m the references provided) may however be used efficiently instead of indirect antiglobulin rosettmg when specific needs make them appropriate and practical The general approach to cell isolation discussed can also be applied to a variety of additional needs such as subfracttonation of NK cell subsets (e.g., CD8+ and CD8- cells) substitutmg appropriate MAb m purification steps followmg the isolation of NK cells by negative selection. Since the number of NK cells that can be obtained directly from peripheral blood is low and may not be sufficient for some studies, a protocol is also

Peripheral Blood NK Cells

163

provrded to Increase the number of NK cells m short-term cultures m vitro. These cultures can be used as a startmg population to separate numbers of NK cells larger than those that would he obtained from eqmvalent volumes of fresh peripheral blood NK cells prepared in thus way, however, have some charactertsttcs of activated NK cells (9), and It 1s advrsable that results of studtes usmg these cells be confirmed with primary restmg NK cells. 2. Materials Culture medium RPMI- 1640, supplemented with 10% heat-macttvated (45 mm, 56°C) fetal bovine serum (FBS), 2 mM glutamme, and, tf desired, anttbtottcs (0 5 UimL pemcillm, 0 5 pg/mL streptomycm) (complete medium) 1 077 g/mL Ftcoll-Na Metrizoate density gradient, such as Ftcoll-Hypaque (F/H) (Pharmacia, Uppsala, Sweden), tt is stored at 4°C m the dark Phosphate-buffered salme (PBS) 12 mM NaH2P0,, 12 mM Na,HPO,, pH 7 2, 0 15MNaCl 0 15MNaCl CrCl, Solutton 0 1% CrCl, 7H20 m 0 15MNaC1, pH 4 5 (stock solution) This must be prepared m advance and aged at least 1 mo before use The stock solution must be stored at room temperature m a glass contamer protected from exposure to light, the shelf life for this solution is at least 1 yr During the fiist week after preparing the solution, its pH needs to be checked every other day and adjusted to 4 5, if needed Repeat the same once per week for the followmg 3 wk This solution is used to couple anttmouse anttbodies to erythrocytes (E) Because CrCl, causes E agglutmatton by lmkmg membrane proteins, each new batch of CrC13 solutton must be titrated to determine the subagglutmatmg dtlutton to he used. For this, sheep E (25 yL of a 2% suspenston m 0 15M NaCl containing 0 1% bovine serum albumin [BSA]) are Incubated (1 1 [v v]) m roundbottom 96-well plates with serial 1 2 dtlutions of the CrCl, stock solution m 0.15MNaCl. The lowest dllutton not causing E agglutmation is determined after a 30-mm mcubatton at room temperature 1 mg/mL Goat antimouse Ig (GaMIg) m 0.15M NaCl, adsorbed on human Ig and affinity-purified. Aflimty-purified GaMIg 1s commerctally available, but may contam human Ig-crossreactive Ab that, if present, may bind Ig-bearmg cells (e g , B-lymphocytes or opsomzed monocytes) and lead to contammation of the NK cell preparation with these cell types Then depletion can be easily obtained by passing the preparation over a human IgG-CNBr Sepharose 4B column. Phosphates need to be removed from the preparations by dialysis against 0.15MNaCl (four changes are usually sufficient) After dtalysls, the preparation 1s filter sterilized (minimum concentratton 1 mg/mL) and stored m l-2-mL ahquots at 4°C for years without loss of titer It IS used to prepare the SE detection system MAbs reacting with leukocyte subsets. anti-T-cells. CD3, CD4, CD5, anttmonocytes CD14, CD32, CD64. anti-B-cells* CD19; anti-NK cells CD16, CD56. If needed (see Notes 1 and 2). antihuman E (anttglycophorm A) and anttPMN (CD 15) The murme B-cell hybrids producing these MAbs are all available

164

Bennett and Perussia

from the Amertcan Type Culture Collectton (ATCC); culture supernatants or ascttes can be used, altemattvely MAbs can he purchased from commerctal sources 8 Sheep erythrocytes (SE). These can be obtamed from several commercial sources and are stored m Alsever’s solution at 4°C for approx 1 mo These cells are coated with the affinity-purified GaMIg using the procedure described m Section 3 9. B-Lymphoblastotd cell lines to be used as feeder cells. RPMI-8866, Daudi, or possibly other B-cell lines

3. Methods 3.1. Preparation of the Starting Lymphocyte 3.1.1, Peripheral Blood Lymphocytes (PBL)

Populations

1 Peripheral blood mononuclear cells (PBMC) are first prepared by density gradient centrifugatton. The expected cell yield is approx 1 x lo6 cells/ml of blood from healthy donors (range 0 5-2 x 106) Place 15 mL F/H in a 50-mL comcal centrifuge tube and overlay 30 mL of blood (anticoagulated with heparm) slowly on top of this solution For opttmal recovery, care has to be taken not to disrupt the surface tension of the density gradient material 2 Centrifuge at 8OOg, 15-20°C for 20-30 mm Make sure that the centrifuge brake has been turned off to obtain a sharp PBMC band at the gradient’s interface 3 Carefully collect the mononuclear cell band at the interface using a IO-mL pipet, and transfer it to a new 50-mL tube Remove all cells m this band, trying to take as httle of the density gradient as possible. Mix the cell suspension with PBS (1 1 [VW]) to dilute any F/H carried over. 4. Centrifuge the cells at 35&4OOg, for 5 mm at room temperature Decant the cellfree supematant (which is turbid owing to the presence of platelets) and rap the tube against a solid surface to resuspend the pellet; if a large number of E are present m the PBMC band, which sometimes happens because of variability m E density, it may be necessary to use vacuum aspiration to remove the cell-free supernatant and avoid cell loss (see Notes l-3). 5 Resuspend the cells m PBS, and centrifuge at 15Og for 7 mm Repeat additional washes m the same conditions twice, and finally resuspend the cells m complete medium for countmg. The low-speed centrifugations are needed to reduce platelet contammatton of the final cell suspension. 6 PBL are prepared from the PBMC isolated above using an adherence step to remove the majority of monocytes. For this, PBMC m complete medium are plated m tissue-culture-treated Petri dishes. The number of cells and volume of the cell suspenston that will allow an even settling of cells depend on the size of the dash used. As an example, 50 x lo6 cells m 5 mL medium form an evenly distributed monolayer when placed m 100~mm* dishes, proporttonally lower numbers of cells (~5 x lo6 cells/lO-mm2 surface) are placed m smaller dishes, but in thus case, a relatively larger volume of medium may he needed to cover the plate evenly Be careful to avoid adding bubbles to the plates since they will prevent the cells from evenly contacting the bottom of the dishes.

Peripheral Blood NK Cells

165

7 Incubate the cells at 37°C for 30 mm m a 5% CO* atmosphere 8 Collect nonadherent cells without detachmg adherent monocytes: for this, add some PBS to each plate (not directly onto the cells, but onto the wall of the Petri dish), and swirl/rock the solution back and forth to resuspend nonadherent cells Transfer this supernatant to a tube. 9 Add PBS to one of the plates, and repeat the washing step Transfer the cell suspension to the next dish, and continue until all plates have been washed Repeat this step again, contmumg until no significant number of nonadherent cells can be seen under an inverted microscope (usually 4-5 washes). All cells collected m the washes are pooled with those collected m step 8 10 Centrifuge the cells, resuspend them m compete medium, and count

3.7.2 Short-Term PM-B-lymphoblastold

Cell Line Cocultures (9)

1. Grow the feeder cells m culture as needed; 2 x lo5 feeder cells are needed for each 1 x lo6 PBL that will be put mto culture The ability of B-lymphoblastoid cell lines to act m vitro as feeders to sustain preferential proliferation of NK cells from PBL depends on the quality of the feeder cells before they are added to the cultures (see Note 4) Exponentially growing, viable cells are essential for successful cultures 2 Irradiate these cells with 30 gy RPMI-8866 cells should be irradiated the day before they are needed and kept m a 37°C incubator until use. Daudi cells can be irradiated and placed into culture on the same day. 3 Immediately before use, centrifuge the cells (2OOg, 5 mm) and resuspend them m fresh complete medium (potentially inhibitory cytokmes produced by these cells during the overmght mcubation are removed m this way). 4 Mix the feeder cells with the PBL, prepared as m the previous section, at a 1 5 feeder cells*PBL ratio, and a final PBL concentration of 2 5 x 105/mL complete medium. 5. Add 2 mL of the cell suspension to each well of a 24-well tissue-culture plate, and place m an incubator (37“C, humidified 5% CO, atmosphere). Cultures can be set up m flasks, but the yield of total cells, and of NK cells in particular, ts lower 6. On d 6 of culture, aspirate approximately half of the medium from the wells, and replace it with fresh medium (see Note 5) 7 On d 10, collect the cultures (see Note 6) The proportion of NK cells present can be determined by surface phenotypmg (indirect unmunofluorescence is the simplest method) the day before. On average, a fivefold increase in total cell number is achieved m the cultures at this time (e.g., -50 x lo6 cells are recovered from cultures started from 10 x lo6 PBL). Typically, NK (CD1 6+/CD56-) cells represent -70-80 and 50% of the cell population when RPMI-8866 and Daudi cells are used as feeders, respectively (this represents a -20-fold increase m the total number of NK cells compared to the startmg PBL population). The remamder cells are CD3+ T-cells B-cells and monocyte/macrophages are not detectable at the end of the culture If active proliferation is observed before d 10, the cells can be collected earlier

166

Bennett and P erussia

3.2. indirect Antiglobulin Rosetting for Positive and Negative Selection of NK Cells 3.2.7. Preparatron of CrCl, SE Coated with GaMlg (10) 1 Wash SE three times with 0 15M NaCl (never use PBS* phosphates mhibtt the CrCl,-dependent coupling of proteins to cell membranes) Spm the cells (8OOg, 7 mm) and aspirate the cell-free supernatant (decanting the supernatant may result m loss of erythrocytes, if the cells were m a loose pellet) 2 Dtlute the CrCl, stock solution m 0 15MNaCl to the appropriate subagglutmatmg concentration previously determined for that batch (final optimal concentration IS usually 0 0 14 005%) Filter-sterilize the solution using a 0 45-pm filter 3 In a 50-mL comcal centrifuge tube, mix the followmg m the given order to prepare 50 mL of a 4% suspension of E-CrCl,-GaMIg 28 mL 0 15MNaC1, 2 mL packed SE, 2 mL GaMIg (the Ig are dissolved m NaCI, 0 15M usually at I mg/mL, depending on the batch, lower concentrations can be used), 8 mL CrC13 solution Smaller volumes can be prepared, depending on the need, modifymg the volumes of the different reagents, but mamtammg their relative proportions (see Note 8) 4 Incubate the suspension for 15 mm at room temperature with occasional mixmg 5 Add PBS to stop the reaction and centrifuge (8OOg, 5 mm) 6 Wash the E-CrCl,-GaMIg twice with PBS (800 g/7 mm), and resuspend m 50 mL complete medium The suspension IS stored at 4°C and can be used up to 1 mo Each time before use wash the cell suspension once with PBS to remove membranes of lysed E or free Ig, which may have come off the cells and, if present, can complete wrth the Intact SE for bmdmg to the MAb-sensitized cells Resuspend E at 4% m fresh complete medium

3 2.2. Lymphocyte Sens/t/zat/on w/th MAb 1 Resuspend the PBL preparations from which NK cells are to be purified (20 x 106/ mL complete medium) m an appropriately sized centrifuge tube 2 Dilute the desired MAb m PBS to the concentration previously determmed to be optimal for rosette formation with cells known to express the Ag of interest, and mtx (I -1 [vv]) with the PBL Culture supernatants, ascites, or purified Ig (or then F[ab’], fragments) are appropriate for use In general, culture supernatants work best at a 1 2-l 4 dilution, ascttes at a 1(3-l@+ dilutton, and purtfied Ig at 0 5-l ug/mL. The optimal concentration to be used, however, has to be determined experimentally for each batch of Ab preparation (see Note 7) For negative selection of NK cells from PBL, use a mixture of anti-T (CD3, CD5), anti-B (anti-HLA-DR, CD19), and anttmonocyte (CD14) MAb If NK cells are to be purified from the PBL-B-lymphoblastoid cells cocultures, a mixture of anti-T and antimonocyte MAb is sufficient, since B-cells are not detectable m the cultures For positive selection, use a mtxture of anti-NK cell MAb (CD16, CD56) in all cases 3 Incubate the cell suspension on ice for 30 mm Prechill the centrifuge and the tube carriers at this time (5-l O’C)

Peripheral Blood NK Cells

167

4

Wash the excess unbound Ab with ice-cold PBS (5 mm centnfugatlon, 4OOg, m the cold) 5 Decant the supernatant, and wash twice more as m step 4 6 Resuspend the cells m 10 mL ice-cold complete medium, and place on ice For optimal rosette formation, a maximum of 200 x lo6 cells can he placed m a 50-mL tube, up to 50 x lo6 cells are instead placed in a 15-mL round-bottom culture tube

3.23. Rosette Formation wdh E-CrCI,-GaMlg 1 Mix 2 5 mL of the 4% suspension of E-CrCl,-GaMIg, m Ice-cold fresh complete medium, with 200 x lo6 PBL presensltlzed, as m Section 3 2 2 , with the desired combmatlon of murme MAb Volumes of E suspension are proportionally modified to treat different numbers of cells 2 Centrifuge m the chllled carriers/centrifuge (4OOg, 7 mm) 3 Incubate the pelleted cells on ice for 30 mm 4 Resuspend the cells wrth a Pasteur pipet until all clumps have been dlsaggregated (rosettes do not break apart) Place a drop of the cell suspension on a slide with coverslip to check for percent rosettes (optical microscopy, count at least 200 cells) (see Note 9)

3.2 4 EnrlchmenVPuriftcatlon

of NK Cells

After resuspendmg the pellet (Section 3 2 3 , step 4), underlay F/H, carefully displacing the lymphocyte-SE mixture upward (13 or 5 mL F/H solution are underlayed m 50- and 15-mL tubes, respectively) Being careful not to Jar the tubes, centrifuge them at 8OOg for 15 mm After centnfugatlon, the cells expressing the antigens recognized by the MAb used are m the pellet (rosetted), and those not expressing them are at the interface of the gradient If MAb-negative cells are to be obtained (negative selection), carefully transfer the nonrosetted cells from the interface of the gradient to a new tube, and add 50% by volume of sterile PBS (see Notes l&12) Wash the cells twice with PBS, and resuspend them m complete culture medium To recover the Ab-positive cells (positive selection), aspirate the F/H, resuspend the pellet in a small volume of PBS, transfer the cells to a new tube (in order to avold contaminating these cells with Ab-negative cells, which may have adhered to the wall of the tube), fill it with PBS, and centrifuge After decanting the PBS, resuspend the pellet by rapping the tube against a solid surface, and wash twice more In order to achieve the highest degree of purlficatlon (>98%), It IS necessary to repeat the rosettmg step, without adding new MAb, on the cells obtained in step 4 or 5 (be sure to keep the cells at 4°C) For this, the cells collected at the mterface (negative selection, step 4) are pelleted with additional E-CrC13-GaMIg as before (steps l-3 m Section 3 2 1 ), and those collected from the pellet (positive selection, step 5) are resuspended m medmm and pelleted agam without adding more E After 30 mm of incubation on ice, the pellets are resuspended, F/H IS

168

Bennett

and Perussia

A PBL Preparation (3.1) - from - from

fresh peripheral m wtro cultures

B Indirect Antiglobulin

blood (3 1 1) (3 1 2)

Rosetting (3.2)

PBL

OOC

PBL + mAb (3 2 2)

’ I ;Q k

+ E-CrCI,-GolMIg **v , +,%

Density Gradient (3 2 4) Negatively

Selected

Cells

Posltwely

Selected

Cells

J Posxtwely Selected Lyse SE

Cells

I

1

Repeat Rosettmg (3 2 4) Negatively

Selected

Cells

Fig. 1. Schematic outline of the indirect anttglobulrn

rosettmg method of NK cell

purification underlayed, and the tubes are centrifuged as before (steps 1 and 2) NK cells are collected from the interface of the second F/H gradrent performed with cells negatively selected (the pellet 1sdiscarded) or from the pellet of the second F/H gradient performed with positively selected cells (cells at the interface are discarded). In this latter case, SE can be lysed by adding 0 5 mL H,O and ptpetmg carefully for 1 mm, after which the tube 1s quickly filled up with PBS and the cells are washed twice Following this protocol, homogeneous NK cell preparations are reproducibly obtained (>98% CD16+/CD56+/CD3- ceils, as determined by mdtrect or direct nmnunofluorescence [9] usmg a panel of MAb) The actual yield is, on average, 60% (range 50-75%) of the theoretical one expected on the basis of the proportion of NK cells m the starting population. e-8 x lo6 and 30 x lo6 NK cells can thus be obtained from 100 x lo6 PBL or cultured lymphocytes, respectively, where NK cells are -15 and 80% (see Notes 13 and 14). A scheme of the steps involved m the separation procedure 1sshown m Fig. 1

Peripheral Blood NK Ceils

769

4. Notes 4.1. Possible Problems in Preparing the Starting Lymphocyte Populations from Peripheral Blood 1 Occastonally, significant numbers of erythrocytes may contaminate the PBL preparations. These cells have an unusual buoyancy, will be carried over m each step, and will actually be enriched durmg the purtfication procedure. Depending on the apphcatton, it may be required that the E be lysed. For this, use HZ0 (see Section 3.2 4 , step 6), or 0 15MNH,Cl, O.OlM KHCOs, 0.1 r&4 EDTA, pH 7 2, buffer. To lyse E with this lysts buffer, the PBL suspension is incubated for 10 mm on ice with the buffer (3 mL buffer/each 100 x lo6 cells), and washed once with PBS Alternatively, anttglycophorin A MAb can be added to the mixture of MAb used for negative selection 2 PMN usually are not found at the interface of the F/H when using peripheral blood from healthy donors However, in rare donors and some patients, PMN wtth altered densities are present that may not pellet through the density gradient, and vat-table proportions of them may contaminate the PBL preparation They can be eliminated adding an anti-CD 15 MAb to the mtxture used for negative selection 3 Low yield of PBL This is usually owing to loss of cells following clumpmg during the isolation procedure There are three primary causes of this* a Insuffictent or too high speed washes of the PBMC in this case a large number of platelets that are subsequently activated (especially m the monocyte adherence step) and aggregate with other cell types may be carried over, b. Cell clumps are easily formed when the cells are either spun at too high a speed, or resuspended inaccurately after washing, and c Release of DNA from dying cells, which acts as a tenacious adhesive m vitro addition of 50 pg/mL DNase I, to these cells and incubation at 37°C for 2-3 mm will clear the clumps (store allquots of sterile DNase I, 5 mg/mL medium without FBS, for easy use). Contmumg the purification without elimmatmg these clumps results in major cell loss

4.2. Limited NK Cell Proliferation in the In Vitro PBL-Lymphoblastoid

Cells Cocultures

4. Individual laboratory clones of the RPM18866 and Daudi cell lines may vary in their abiltty to support NK cell proliferation from PBL, moreover, this property decreases after they are kept m culture for long periods of time (4-6 mo) Defrost a new batch of low-passage cells when the NK cell yields start declmmg (usually this happens gradually). 5. The precise feeding and harvesting schedule for a particular feeder cell or isolate may need to be optimtzed followmg the kinetics of NK cell proliferation Maximum NK cell proliferation may actually occur one or more days earlier or later than that described 6. Although proliferation of NK cells m the cocultures is observed with PBL from most healthy donors, donor variability may account for occasional unsuccessful cultures

Bennett and Peruss~a

170 4.3. Lack of Rosette Formation

7 Incorrect dllutlons of the MAb used for PBL sensltizatlon To solve the problem, the working dllutlon for each MAb preparation used has to be determmed For this, serial ddutions of the MAb are incubated with lymphocytes m round-bottom 96-well plates under the condltlons described m Sections 3.2 2 and 3 2 3 Rosette formation IS assessed microscopically on an allquot of the cells Choose the dllutlon resultmg m the maximum percentage of rosettes, correspondmg to that of cells detected with the MAb by mdu-ect lmmunofluorescence 8 Insufficient amount of GaMIg coating the SE a Each new batch of GaMIg has to be titrated to determine the correct concentration of Ig to be coupled to E This IS achieved by testing for rosette formation SE suspensions prepared with senal dilutions of the GaMIg (2-O 25 mg/mL) b Phosphates are present m one of the reagents used for CrC13 couplmg Because even trace amounts of phosphate inhibit the CrCl,-dependent protein coupling to cell membranes, tt IS essential that E are washed with saline, and that all reagents used are prepared and diluted m the same solution (avoid PBS at any time) c On rare occasions, couplmg of GaMIg to E IS unsuccessful for no apparent reason Just start again Each new batch of E-CrCl,-GaMIg has to be prepared and tested m advance 9 Low numbers of rosettes compared to the proportion of Ab+ cells If the reasons m Notes 1 and 2 can be excluded, the most likely explanation for this is that the temperature was not kept low during all steps This may result m capping and down-modulation of the antigen from the cell surface, thus preventing efficient rosette formation

4.4. Contamination of the Final NK Cell Preparation with Other PBL Subsets 10 Make sure that all procedures are performed m the cold and that rosettmg 1sperformed twice. 11 In the case of negative selection, the most likely explanation 1s a short time of centrifugatlon of the F/H gradient, increase the time to 20 mm and/or modify the centrlfugatlon speed 12 In the case of posltlve selection, Inaccurate resuspension of the pellets after rosette formation will result m trapping rosette Ab- PBL m the clumps These will sedlment m the pellet of the F/H gradlent, thus contammatmg the rosette Ab+ cells

4.5. Low Recovery

of NK Cells

13 In the case of negative selection the problem IS, for the most part, owing to inaccurate resuspension of the pellets after rosette formation before the separatron on F/H. Rosette-negative PBL trapped m the clumps reach the pellet, resulting m loss of cells 14 When the recovery of posltlvely selected cells 1s low, loss of cells probably occurred during the E lys~s step Careful resuspension of the cells during the lys~s and use of the NH&l buffer instead of H20 should solve the problem

Peripheral Blood NK Cells

171

References 1 Ttmonen, T , Ortaldo J R , and Herberman, R B (1981) Charactertsttcs of human large granular lymphocytes and relattonshtp to Natural Keller cells J. Exp A4ed 153,569-582 2 Perussta, B , Fannmg, V , and Trmchtert, G. (1985) A leukocyte subset bearmg HLA-DR anttgens IS responstble for m vttro alpha Interferon productton m response to vu-uses Nat1 Immun Cell Growth Regul 4, 12&131 3 Perussta, B , Starr, S , Abraham, S , Fanning, V , and Trmchtert, G (1983) Human Natural Killer cells analyzed by B73 1, a monoclonal antibody blockmg Fc receptor functions I Charactertzatton of the lymphocyte subset reacttve with B73 1 J Immunol 130,2133-2141 4 Lamer, L L , Chang, C , Azuma, M , Ruttenberg, J , Hemperly, J , and Phtlbps, J (199 1) Molecular and functtonal analysts of Natural Keller cell-associated Neural Cell Adhesion molecule (N-CAM/CD56) J Zmmunol 146,442 14426 5 Perussta, B , Acute, 0 , Terhorst, C , Faust, J , Lazarus, R , Fanning, V , and Trmchten, G (1983) Human Natural Killer cells analyzed by B73 1, a monoclonal antibody blocking Fc receptor functtons II Studies of the B73 1 anttbody antigen mteractton at the lymphocyte membrane J Immunol 130,2 142-2 148 6 Anegon, I , Cuturt, M C , Trmchtert, G , and Perussta, B (1988) Interactton of Fc receptor (CD16) with ltgands induces transcrtptton of mterleukm 2 receptor (CD25) and lymphokme genes and expression of then products m human Natural Keller cells J Exp Med 167,452-472 7 Coltgan, J , Krursbeek, A , Margutlees, D , Shevach, E , and Strober, W (eds ) (199 1) Current Protocols in Immunology, Wiley, New York, Sections 3 and 5 8 Naume, B , Nonstad, U., Stemgker, B , Funderud, S., Smeland, E , and Espevtc, E (1991) Immunomagnettc tsolatton ofNK and LAK Cells J Immunol 148,242%2436 9 Perussta, B , Ramom, C , Anegon, I , Cuturt, M C , Faust, J , and Trmchiert, G (1987) Preferential proltferatton of Natural Killer cells among peripheral blood mononuclear cells cocultured with B lymphoblastotd cell lures Nat1 Cell Growth Regul 6, 171-188 10 Godmg, J W (1976) The chromium chloride method of couplmg antigens to erythrocytes definmon of some important parameters J Immunol Methods 10, 6 l-66

15 Cystic Fibrosis Airway Epithelial

Cell Culture

Manuel A. Lega 1. Introduction 1.1. Cystic Fibrosis Cystic fibrosis (CF) is the most frequent (incidence around l/2500 live births) genetic cause of death among Caucasians. It is an autosomal recessive disorder compromismg the secretory epithelia. Clinically, CF is a polymorphic disease showing abnormal functionmg of the au-ways, the digestive apparatus (pancreas and intestine), the reproductive tract, and the sweat glands, leading to respiratory msufficiency, malnutrition, male sterility, and production of salty sweat. The average life-span of CF patients falls around 25-30 yr of age m the United States and Europe, and around 10 yr of age m Latin America (1,2). Respiratory mfections are the cause of death of more than 90% of CF patients. No curative treatments are as yet available for CF. Chloride transport is the primary function affected m CF epithelial cells. The Cl- transporter molecule involved m CF cells is a CAMP-dependent, apical membrane protein called cystic fibrosis transmembrane conductance regulator (CFTR). CFTR mutant cells have lost their ability to move chloride ions m response to CAMP (3,4). CFTR, a 1480 ammo acid polypeptide, is encoded by the CF gene The CF gene spans over 250 kb and contains 27 exons. At the time this chapter was written, more than 500 different mutations had been reported on CF alleles found m CF patients. The so-called AF508 mutation (an m-frame deletion of 3 bases causing the loss of phenylalamne at position 508 on the CFTR protein) is the mutation most frequently found m all populations so far tested; its relative frequency varies from 30-80% among different ethnic groups. The AF508 mutation affects the processmg of CFTR along the endoplasmic reticulum and the Golgi apparatus leading to the absence of CFTR on the plasma membrane (4-9). From

Methods m Molecular Edlted by G E Jones

Medmm Humana

173

Human Press

Cell Culture Inc , Totowa,

Protocols NJ

174

Vega

CFTR expression has been reported m secretory eptthehal cells (an-ways, intestine, pancreas, and eptdidymus). Moreover, there 1s some expression of CFTR m nonepttheltal cells like lymphocytes and ftbroblasts. Recently, expression of CFTR-mRNA m human ejaculated sperm cells (I 0, II, Vega et al , unpublished results) has been found. In the airways of CF patients, a deficient CFTR-mediated Cl- transport together with the observed decreased reabsorptron of Na+, leads to the alteration of the ionic and osmottc properties of the lummal mucus overlaymg the eprthehum Deprived of salt tons, the mucus becomes dehydrated and more VISCOUS, mterfermg with normal withdrawal and ctllra-mediated cleaning up of the eptthelmm The effect on the normal secretory functtons of the an-way eprthelmm thus leads to accumulatton of mucus, bacterial colomzatton, mflammatron, and final alteration of the bronchrolar and alveolar htstologtcal structure.

1.2. Airway Epithelial

Cells

As far as CF is concerned, detarled knowledge on anway eprthehal cells plays a pivotal role in two key ways: m studies aimed at getting mstght mto the basic phenomena involved m the molecular pathogenesis of the disease and m the development of gene transfer wrth therapeuttc purposes mto the eprthehum m VIVO. Human auways extend from the nose inward, mto trachea, bronchi, bronchioles, and alveoli The eptthelmm covering the respiratory tract mcludes a mixture of different cell types with speclahzed functrons, all laying on a basement membrane The relative dtstrtbutron of the dtfferent cell types varies along the different zones of the anways CFTR expression level IS, as well, not umform along the different cell types found m human airway eprtheha. rt 1s hrghest in submucosal gland cells, found all along an-way epithelmm. Alterations caused by mutant CFTR molecules are restricted, however, to the bronchial and bronchtolar tubes (22,13) The major cell types found m the bronchial pseudostrattfied eptthelmm are basal cells (pyramidal cells not reaching the lumen of the eptthelmm), ciliated cells, undifferentiated cells (presumably the airway eprthelmm stem cells), goblet cells, and serous cells. Branching from the anway, the eptthelmm forms submucosal glands covered by goblet and serous cells, resting on a membrane conttguous to the an-way epithelmm basement membrane. Goblet and serous cells, either from the submucosal glands or from the airway eprthelmm, produce mucus that fills the submucosal glands and forms “islands” over the pertcellular fluid lmmg the airway surface (142.5). The eptthelmm of the bronchtole IS mainly constttuted by two cell types. ciliated cells and Clara cells, formmg a columnar eptthelmm that lays on a basement membrane and 1s lured by a mucus layer (14,15)

Airway

Epdhellal

175

Cell Culture

A difference m cell type composltlon of bronchial, but not nasal, epltheha, between CF and normal patients IS observed. relatively less clhated cells and more basal, undifferentiated and secretory cells are found m CF bronchlai eplthehum than m normal bronchial eplthehum (26)

1.3. Epithelial

Cell Culture

When cultured m vitro, eplthellal cells show the characteristic eplthellal morphology. cells are near lsodlametnc, cells are compacted m well-defined colonies, with precise hmlts and do not spread out of the colonies Cells of some CF anway eplthellal cell lines are, however, fibroblast-shaped, and aggregate m colonies with no precise limits and with cells spreading out of the colonies Eplthehal cells express characteristic cytokeratms and form cell-to-cellJunctlons typical of epltheha Monolayers formed by m vitro growth are real eplthehal sheets that generate measurable potential differences on both sides of the monolayer. Cell lines are grown on plastic Colonies can be grown up from as little as 1 cell/well (1 mL of culture media/well). There 1s no difference m the mean size of the colotues grown from either 1 cell/well (1 mL) or 1O(r1000 cells/well (1 mL). No feeding IS then necessary for growing mdlvldual isolated clones Dlvlslon rate IS about 24 h Primary cells, however, are grown on collagen-coated dishes and usually on feeder layers Dlvlslon rate IS around 100 h, and cells support only three to five passages before they stop dlvldmg.

1.4. Mycoplasma

Decontamination

An-way eplthehal cells survive antlmycoplasma treatment, conslstmg of culturing m normal medium containing 1% mycoplasm removal agent (ICN Flow, Costa Mesa, CA; cat. no. 30-500-44) for 2 wk. Growth rate and morphology are not affected by the treatment.

1.5. Sensitivity

of Airway Epithelial

Cells to G-418

As expected, different an-way eplthehal cell lines show different sensitivity to the neomycin derivative G-4 18 At day 7 of culture m the presence of G-4 18, survival of cells varies from cell line to cell lme between 10 and 8&100% (50 pg G-41 8/mL) and between 0 and 50% (100 pg G-41 8/mL). An interesting observation, however, has been made m the laboratory concerning the natural resistance of an-way cell lines to G-4 18, suggesting that CF cell lines might be more sensitive to G-41 8 than normal cell lines. This dlfference allows for a selective recovery of the more resistant (normal) cell lme m the presence of G-418, from mixtures of two different (normal and CF) cell lines with differential sensitivity to G-4 18 (unpublished data)

Vega

176

1.6. Differential Sensitivity of CF and Normal Cells to Epinephrine It has recently been reported that CF airway epithehal cells, either primary or cells lines, show a higher sensmvity to epmephrme-induced toxicity than equivalent normal airway eptthelmm cells (17). Treatment with either epmephrme or forskolm rapidly kills CF cells, whereas normal cells are mainly unaffected. Killing of CFTR- cells by epinephrine is quite rapid: 16-24 h. CF cells transfected with CFTR expression vectors become epmephrme-resistant (Vega et al , unpublished results) In fact, both CF (CFTR-) and normal (CFTR+) cells are sensitive to epinephrme, but CF cells respond to lower concentrattons and shorter times of treatment. Sensttivity to epmephrme is dependent on cell density CFTR+ cells can be recovered out of mixtures of CFTR-and CFTR+ cells by epmephrme treatment The epmephrme-based selection method is rapid and easy to perform However, it demands a skilled operator as long as it depends on a relatively narrow window given by the differential sensitivity of CFTR- and CFTR+ cells, 1.7. Sensitivity

of CF Cells to Temperature

As mentioned, the AF508 mutation causes the CFTR to be retained at the Golgi apparatus and to not to reach the plasma membrane (9,18). However, it has been reported that when growing (AF508) CF cells at temperatures lower than usual (e.g., 20”(Z), the (AF508) CFTR can be detected on the plasma membrane, and there tt accounts for the partially recovered CAMP-dependent Cl- transport. Therefore, AF508/AF508 CF cells behave more like “normal” cells when cultured at 20°C.

2. Materials 1 Culture media* For cell lines (either normal or CF cells) use Dulbecco’s modified Eagle’s medium (DMEM).Ham’s F12 (1. l), 10% heat-inactivated fetal calf serum (mycoplasma free), 2 mA4 L-glutamme, 50 pg/mL pemctllm-G, and 50 PgimL streptomycm Filter sterthze either the mdivtdual components or the final mixture, and store at 4°C for no longer than 2 wk Anttbtottcs and L-glutamme are prepared m water, filter-stenhzed, ahquoted m 5 mL stocks, and stored at -20°C Fetal calf serum 1s heat-mactlvated (30 mm, 56”(Z), ahquoted m 50 mL, stored at -20°C and filter sterthzed Just before adding to the culture medmm For primary cells (either normal or CF cells), use DMEM Ham’s F12 (1 l), 10% heat-inactivated fetal calf serum (mycoplasma free), 2 mM L-glutamme, 50 &mL pemctlhn-G, 50 ug/mL streptomycm, 5 pg/mL transfemn, 10 ug/mL msuhn, 100 ng/mL hydrocorttsone, and 25 ng/mL eptdermal growth factor (EGF) Stock solutions (100-1000X) are prepared as follows: transferrm (human, non-free) m DMEM.Ham’s F12, msulm (bovine pancreas) m O.OOlM HCl m

Alrway Epithellal Cell Culture

2

3

4.

5

6 7

177

water, EGF (mouse) m water, and hydrocortisone (first dissolved m 100% ethanol to a concentration of 5 mg/mL) m HEPES-buffered Earle’s salts (HBES) + 5% fetal calf serum Stocks are filter stertltzed, ahquoted, and stored at -20°C (msulm solutton can be stored at 4°C for a couple of weeks) Coatmg wtth collagen Dissolve collagen (Human placenta, type IV, Stgma, St. Louis, MO) m 0 1% acetic acid by stnrmg for 48 h at 4°C Filter the collagen solution through 2-3 sterile gauze layers. Dilute with water (l/20) Pour the collagen solutton mto plastic dishes 1 mg/mL, 10-50 pg/cm2. Au dry for l-2 d (temperature not higher than 37’C) Store coated dishes at 4°C m and humtd atmosphere (for several months) Storage medium a Medium 1 10% dtmethylsulfoxide (DMSO), 80% heat-macttvated fetal calf serum, 10% culture medium b Medium 2 10% DMSO, 60% heat-inactivated fetal calf serum, 30% culture medium Store at -20°C for months Add the DMSO component immedtately before use Epmephrme solutton Prepare a 125mA4 stock solution of epmephrme (Sigma) m water The stock solutton (1000X) can be stored at -20°C for months and resists repeated cycles of freeze and thawmg Keep thawed epmephrme solutions on ice Prepare the working dilutions immediately before use Loading solution: 108 mA4NaCL4.7 mA4KC1, 1 mA4CaC12, 1 mMMgC12, 20 mM NaHC03, 0 8 mA4NazHPO,, 0 4 mMNaH,PO,, 10 Wglucose, and 5 mMHEPES, pH72 Immortahzatton buffer 10 mM Na2HP0,, 10 mMNaH,PO,, 1 mMMgC12, and 250 mA4 sucrose, pH 7 4 Filter sterthze and store at 4°C Detachment solution 0.5 mMEDTA, 0 1% trypsm in (Ca2+-free, Mg2+-free) PBS Prepare a 10X stock Filter-sterilize, ahquot, and store at-20°C Dilute m PBS to make a 1X working solutton that can be stored at -2O’C and resist thawing several times

3. Methods 3.1. Obtaining and Culturing (see Notes l-6)

of Primary Airway Epithelial

Cells

1 Cells can be obtained by epithehal brushmg of the airways through fiberopttc bronchoscopy; by scraping off nasal polyps following surgery, by scraping off the eptthehum from lung, bronchi, or tracheal necropsy, or biopsy specimens 2. In the two latter cases (scraping off from surgery pteces), protease treatment (2 5 mg/mL pronase, 60 mm, 37°C m DMEM medmm containing 1% fetal calf serum) 1sused to liberate cells 3 Recover cells m culture medium for primary cells 4. Centrifuge at 8OOg, 4’C, for 5 mm. 5. Recover in fresh medium, and culture in collagen-coated dishes at 37”C, under 5% CO, Feed with fresh medium every 2 d

Vega

178

3.2. Immortalization

of Airway Epithelial

Cells (see Notes 7 and 8)

1 Anway eprthehal cells can be nnmortahzed wtth the large T-antigen of SV40 vuus Grow primary epnhelral cells as usual 3 Harvest cells with detachment solution (5-10 mm, room temperature), and centrifuge for 10 mm at 4”C, 800g 4 Resuspend cells m 0.5 mL nnmortahzatron buffer at a density of 106-10’ cells/n& 5 MIX 10 ug of a SV40 large T-antigen plasmrd expression vector wtth the 0 5 mL cell suspension 6 Incubate for 10 mm on Ice 7 Perform electroporatron as indicated m Chapter 16, Sectton 3 1 2 8 Incubate for 10 mm at room temperature 9 Recover cells m culture medium (culture medium for cell lines), and culture at 37”C, 5% CO,, as usual Replace the culture medium wrth fresh medium every 3-5 d 10 After a couple of weeks, colonies of rmmortahzed cells will be clearly drstmgutshable 11 Continue culture of these rmmortahzed clones as usual for ceil lines 2

3.3. Culture of Airway Epithelial

Cells

Cells are grown at 37°C under 5% CO2 m the respective culture media described above Duplrcatton trme under the condlttons descrtbed IS around 24 h for the cell lmes and around 100 h for the prrmary cultures

3.4. Detachment

of Cells

1 Completely remove the culture medium lying on the cells by asprratlon 2 Immedtately overlay wrth l-3 mL detachment solutron for a small (30-mL) bottle or a IO-cm plate 3 Incubate at room temperature m the hood for 3-5 mm, avotdmg direct contact of the plate/bottle with the metallic parts of the hood 4 Shake gently by hand from time to time, and look by eye agamst a hght or a window until the monolayer detaches Fmnly close the bottle and agitate energrcally several ttmes to homogenize the cell suspension and disaggregate cell clumps 5 Once the cell suspensron IS homogeneous, add 4 vol of culture medium to neutralize trypsm and recover cells by centrrfugatton at 8OOg for 10 mm at 4°C

3.5. Cryopreservation 3 5 I Freezrng 1 2 3 4

of Cells (see Notes 9-12)

Down Cells

Grow cells up to no more than 70% confluence m normal culture medium Detach cells as prevrously described Centrifuge at 4°C SOOg for 10 mm Resuspend cellular pellet m Ice-cold storage medium to a density of around lo6 cells/ml Always keep the cell suspension on ice and process mllnedtately Splrt into Ice-cold cryotubes, 1 mL each, and keep on ice 5 Freeze down using any automatic procedure, and store in lrqutd nitrogen

179

Alrway Eplthelial Cell Culture 3.5 2. Thawmg Cells

1 Thawing of cells has to be performed with care Take the cryotubes out the hqutd mtrogen and tmmedtately put on Ice 2 Thaw by putting the tube m a water bath prevtously set up at 37°C 3 Immediately after thawing, put on me and process 4 Gently recover the cell suspenston and plpet, drop by drop, mto 5 mL of Ice-cold normal culture medmm 5 Centrifuge at 4”C, SOOg for 10 mm 6 Resuspend m an approprtate volume of culture medium, and culture as usual

3.6. Epinephrine

Selection

Procedure

(see Notes 13-15)

1 Grow cells up to 7&80% confluence as described 2 Replace culture medium wtth fresh medium containing the desired concentratton of epmephrme Final eptnephrme concentrattons that allow for dtstmctton between CFTR- and CFTR+ cells are from 100-300 mA4 3 Culture as usual 4 After 12-16 h, follow up CFTR- cell death under the microscope, observing every 2-3 h 5 Gently shake the bottle or the dish by hand to remove partially detached cells 6 To stop the selectton, replace medium contammg epinephrme with fresh eptnephrme-free culture medmm, and culture as usually 7 Continue culture of survlvmg cells, or analyze the survtvmg populatton for the expression of active CFTR

3.7. Sensitivity

of (AF508) CF Cells to Temperature

(see Note 16)

1. Grow cells up to 5&70% confluence at 37°C under 5% CO*, as usual. 2 Replace culture medium with fresh medium, and continue culture at 20°C under 5% C02, for 24 h. 3 Assay for activity of the CFTR on the plasma membrane For instance, using the epmephrme-resrstance assay a. Change the old medium with fresh medium containing the desired concentration of epmephrme (1 O&300 pA4 final concentratton) b Continue culture at 20°C (or at 37°C control cultures), under 5% CO*, for another 24- or 48-h perrod

3.8. CFTR Activity Assessment: (see Notes 17and 18)

Isotope Efflux Measurement

1 Grow cells m six-well plates until about 90% confluent, under the usual condtttons 2. Replace the culture medtum with 0.5 mL of loadmg solutton containmg ‘2511 (3-5 uCt/mL), and incubate at 37°C for 30 mm 3 Remove the extracellular isotope by washing three times, at room temperature, with 3 mL of loading solutton

Vega

180 4 Measure basal isotope every 30 s of 1 mL of 5 Add forskolin (10 pm to measure sttmulated 6 Measure ‘251- m every

efflux loading to the isotope ( I-mL)

for 34 mm at 37’C, by addition and replacement solutron on the cells subsequent 30-s-replacements of loading solutton efflux. isotope basal and stimulated efflux sample

4. Notes

4. I. Obtaining

and Culturing

Primary Airway Epithelial

Cells

1 The epttheltal nature of the cells can be vertfied by their abiltty to form charactertstic epttheltal sheets (Transeplthebal potenttal differences can be measured across the eplthelial sheet ) 2 Plating efficiency of these primary cells 1s approx 30% Confluence 1s reached around 7 d of culture 3. Under the described condmons, primary eplthehal cells proliferate until three to four passages (splitting 1.4) After that, prollferatton stops 4 Anway eptthebal primary cells can be grown on a feeder layer of 3T3 fibroblasts (like Todaro’s Swiss mouse 3T3,3T352, NIH 3T3, and Balb/c-3T3) For preparation of feeder cells a Grow 3T3 cells m 10% fetal calf serum m DMEM (they can be mamtamed as contmually growmg stocks for l-2 mo) unttl confluence b Prepare confluent cultures of 3T3 cells as feeders by nradtatron (30 gy) At this step, cells can be mamtamed m the Incubator m fresh medmm for several days before proceedmg c. Wash twtce with detachment solutton, and recover dtsaggregated cells d Spht l/3 and replace m 10% fetal calf serum m DMEM. e Wtthm 48 h, replace the DMEM medium by eptthelial cell culture medium containing the an-way eptthelial primary cells, and culture as Indicated 5 Since fibroblasts may be contammatmg the airway eptthehal cells, they can be selectively eliminated as follows a Culture the eptthehal cells until colonies of 50-200 cells are developed b Replace the culture medium by 0 02% EDTA in PBS, and Incubate for half a minute m the hood. c Pipet up and down the EDTA-PBS solution rather vigorously over all the area of the dash (By this procedure, flbroblasts-as well as feeder cells-but not the epitheltal cells are detached from the dish d. Replace the EDTA-PBS wtth fresh solution, and repeat the operation until no fibroblasts can be seen under the microscope e Wash attached cells with serum-free medium, and finally add fresh complete culture medium f Add new irradiated feeder cells to the dash, and contmue culture as usual. 6 Eprthehal cells can be characterized by lmmunocytochemical detectton of the cytokeratms typical of eplthehal cells, by detection of eptthellal cell-cell mteractton structures through electron mtcroscopy, or by measurement of potential differences across the epithehal monolayer.

Airway

Eplthelial

Cell Culture

4.2. Immortalization 7 Immortahzed eprthelral clones can be subsequently characterized by detectron of the large T-antigen expression by mmn.mocytochemrstry, by Northern analysis, or by thetr ability to pass over a crmcal number of passages (necessary for a cell culture to become a cell line). 8 Alternative to the use of SV40 large T-antigen alone as rmmortahzmg factor, hybrid SV40-adenovrrus 12 has been used to rmmortahze an-way eprthelral cells (19) When the cell lines are to be used for’the study of CF gene transfer through (adeno)vrrus vectors, however, SV40 large T-antigen alone 1s preferable.

4.3. Cryopreserva

tion

9 Cells stored at lower densities ( 104-lo5 cells/ml) are also vrable 10 Keep cells m DMSO-contammg medta always on ice Unless cold, the DMSO will affect the cells 11 Freezing down cells can be efficiently achieved by the followmg simple procedure* overnight at -20°C followed by ovemtght freezing at -80°C followed by final storage m llqurd nitrogen Cells can be stored m hqurd nitrogen for years 12 No higher concentrations of serum m the medium used for recovery and mitral culture of the thawed cells 1s required Cells usually recover raprdly and grow fast, so it 1sdesirable to dilute the thawed suspensron to a convement cell density m order to avoid the necessity of rapid sphttmg

4.4. Epinephrine

Selection

13 Cell densities 3’

Product on CFTR-mRNA

AGAACTGGAGCCTTCAGAGGG GTTGGCATGCTTTGATGACGC

( 10) (10)

158Obp (exon 10)

CGGATAACAAGGAGGAACGC(4) GCCTTCCGAGTCAGTTTCAG (7)

565 bp (exons 4-7)

CGGATAACAAGGAGGAACGC(4) TTCTGCACTAAATTGGTCGA ( 13)

1637bp(exons&13)

CTGCCTTCTGTGGACTTGGTT TTCTGCACTAAATTGGTCGA

(6a) (13)

1403 bp (exons 6a-13)

GGGGAATTATTTGAGAAAGC GGAAAACTGAGAACAGAATG

(9) (10)

GTTTTCCTGGATTATGCCTGGC ( 10) TTCTGCACTAAATTGGTCGA (13)

270 bp (exons 9-10)

467 bp (exons l&13)

PCR amphfkatlon 5 mm 94”C, 1 mm 94”C, 1 mm 55”C, 2 mm 72”C, 7 mm 72°C (35 cycles) 5 mm 94”C, 1 mm 94”C, 1 mm 60°C. 2 mm 72”C, 7 mm 72°C (35 cycles) 5 mm 94”C, 1 mm 94”C, 1 min 55°C. 2 mm 72”C, 7 mm 72°C (35 cycles) 5 mm 94”C, 1 mm 94”C, 1 mm 55”C, 2 mm 72”C, 7 min 72°C (35 cycles) 5 mm 94”C, 1 mm 94”C, 1 mm 55’C, 2 mm 72”C, 7 mm 72°C (35 cycles) 5 mm 94”C, 1 mm 94’C; 1 mm 55”C, 2 mm 72”C, 7 mm 72°C (35 cycles)

“The PCR product of 158 bp covers the codon deleted m the AF508 mutation The AF508-PCR product can be dlfferentlated from the normal PCR product by PAGE (see Sectlon 3 1 , step 15)

Amplify

as follows.

5 mm, 94°C; 60 cycles of: 1 mm, 94’C, 1 mm, 55”C, 2 mm,

72°C; and 7 mm, 72°C 2 Analyze on 6% polyacrylamlde gels to test for the presence of single-stranded DNA. 3 Sequence asymmetric PCR products (smgle-stranded DNA) by standard smglestranded DNA sequencing procedures

3.2.2. Sequencmg of Double-Stranded

DNA

1 Clone PCR products obtained from the first PCR reaction, after running an electrophoresls gel and extracting the DNA as described m Section 3 2 1. using the TA Clonmg kit with the pCRTM vector (Invitrogen Corporation, San Diego, CA) 2. Analyze transforming clones by standard DNA miniprep analysis or by PCR amplification directly on heated (10 mm, SO’C) bacterial cells using the same primers as for the first PCR reaction. 3 Sequence plasmid DNA by standard double-stranded DNA sequencing procedures

198

Vega

3.3. Easy Detection

of (AF508) CFTR-mRNA

Perform a PCR amphficatlon reactlon on the RT products obtamed as described, under the following condltlons a Primers 5’-AGAACTGGAGCCTTCAGAGGG-3’ and S-GTTGGCATGCTTTGATGACGC-3’ b Amphficatlon reaction 5 mm 94°C. 40 cycles of 0 5 mm, 94”C, 0 5 mm, 62”C, 1 mm, 72”C, and 7 mm, 72°C Analyze on 6% polyacrylamlde gels as described If two bands are observed on the polyacrylamlde gel, the starting material was a mixture of normal and AF508 molecules (either different alleles on the cellular endogenous genes, or a mixture of cellular endogenous genes and vector-produced exogenous alleles) If a single band 1s obtamed, only either AF508 or normal molecules were present m the cell Further analyze the sample as follows m order to elucidate the identity of the molecules present m the single band Put 5 yL of the PCR product of the former reaction m each of two separate PCR ampllficatlon tubes Add 5 pL of a DNA solution (at a concentration of DNA similar to that of the PCR amplification products) made of a (homoduplex) DNA known to be either normal/normal or AF508/AF508 (for instance, genomlc DNA or cDNA coming from a normal [homozygous on exon lo] mdlvldual or from a AF508/AF508 CF patient) Incubate for 10 mm at 94°C followed by 10 mm at 55°C Analyze on a 6% polyacrylamlde gel as described It two bands are obtained from the mixture with the DNA known to be normal/normal and a single band 1s obtained from the mixture with the DNA known to be AF508/AF508, then the unknown sample 1s AF508 Inverse results indicate that the unknown sample 1s normal/normal

3.4. DNA Genotypic

Analysis (see Note 8)

1 Isolate genomlc DNA as follows a Resuspend cells (2 x lo6 cells) m 1 mL of lys~s buffer, and put mto an Eppendorf tube b Centrifuge for 20 s at maximum speed m a Eppendorf mlcrofuge, discard supematant, and resuspend the pellet m 1 mL of lysls buffer c Repeat the procedure three times d Resuspend the pellet m PCR-2 buffer, and incubate for 60 mm at 56°C e Inactlvate protemase K by incubating for 10 mm at 95°C f Take 25 pL of lysate (containing approx 1 pg genomlc DNA), and perform PCR amphficatton as usual m a total volume of 100 pL 2 Determine first If the AF508 mutation IS present Perform a PCR amplification reaction on genomlc DNA using the primers and condltlons described m Section 3 3 for analysis of exon 10 on CTFR-mRNA

CF AIrway

Eplthelial

Cells

199

3 If mutations other than the AF508 are present, they can be detected and ldentlfied as mdlcated m Section 1 (3-5)

4. Notes

4.1. RT-PCR Analysis RNA lsolatlon and the RT reactlon have to be performed usmg sterile tubes, tips, and solutions, wearing gloves, keeping all solutions on ice, and, when possible, usmg disposable plasticware Water (and water for preparation of solutions) should be DEPC-treated RNA concentration can be precisely determined by spectrophotometry at 2601280 nm Alternatively, when no precise determinations are necessary, an estlmatlon can be made on an agarose gel by eye comparison with a sample of RNA of known concentration The hexanucleotlde mix used m random priming DNA-labeling procedures can be used for prlmmg the RT reaction Airway eplthehal cells, but not any tissues, contain an amount of CFTR-mRNA high enough to allow for its detectlon by PCR usmg volumes of RT reactlon products even lower than the 2 pL indicated m Section 3 1 The RT reactlon products can be stored at -20°C before proceeding into the PC’R amplification reaction Each of the PCR reactions described (see Table 1) gives a single band of the indicated size when airway eplthehal cells are used, although additional bands can be obtained when RNA from other tissues 1s analyzed

4.2. Sequencing

of CFTR-mRNA

7 DNA extraction from gel pieces can be performed in the same PCR amphficatlon tube where the asymmetric PCR reaction will take place In this case, consider the extraction volume of water for the calculation of the final volume of water to be added to the PCR reaction

4.3. DNA Genotypic

Analysis

8 When setting the PCR reaction on genomlc DNA isolated as described in Sectlon 3 4 , consider the composltlon of the lysls PCR-2 buffer in which the DNA 1s dlssolved for the calculation of the volumes of the reagents to be added

Acknowledgments The author thanks Lila N. Drittantl (Argentina), Jorge Gabbarml (Argentina), Claude Besmond (France), Mlchel Goossens (France), Pascale Fanen (France), and Bruno Costes (France), as well as AFLM (Assoclatlon Franqalse de Lutte contre la Mucovlscldose) (France), CONICET (National Research Council, Argentina), Hospital Int Gral. J. Penna (Argentina), and MICROGEN S.A.-Biotechnology (Argentina)

200

Vega

References 1 Cystzc Fzbrosu mutatzon data Prrvrleged commumcation prepared for members of the CF Genetic Analysts Consortmm, April 1995 2 Fanen, P , Ghanem, N , Vtdaud, M , Besmond, C , Martm, J , Costes, B , Plassa, F , and Goossens, M (1992) Molecular charactertzatron of cystic fibrosis 16 novel mutations Identified by analysis of the whole cystic fibrosts conductance transmembrane regulator (CFTR) coding regions and sphce site Junctrons Genomzcs 13,77&776 3 Ghanem, N , Fanen, P , Martin, J , Contevllle, P., Yahia-Cherif, Z , Vldaud, M , and Goossens, M (1992) Exhaustrve screenmg of exon 10 CFTR gene mutatrons and polymorphisms by denaturmg gradient gel electrophoresrs* apphcattons to genetrc counsellmg m cystrc fibrosrs A401 Cell Probes 6, 27-3 1 4 Costes, B , Grrodon, E , Ghanem, M , Chassrgnol, M , Thuong, N T , Dupret, D , and Goossens, M (1993) Psoralen-modified ohgonucleotrde prrmers Improve detection of mutatrons by denaturing gradient gel electrophorests and provide an alternative to GC-clamping Hum. A401 Genet 2, 393-397 5 Vidaut, M , Fanen, P , Martin, J , Ghanem, M., Nrcoles, S , and Goossens, M (1990) 3 point mutations m the CFTR gene in french cystic pattents. tdenttficatton by denaturmg gradient gel electrophoresrs Hum Genet 85,44f5-449

Human Tracheal Gland Cells in Primary Culture Marc D. Merten 1. Introduction For several years, tracheal gland cells have been cultured from different animal species, such as the cat (I), cow (2), and ferret (3). There are dlfferences, however, m the structure and function of the various animal airways, rendering it difficult to extrapolate to humans. In this chapter, the author describes techniques that facilitate the lsolatlon and culture of tracheal gland cells from humans. These techniques allow high reproduclbllity, optimal cell isolation, and high phenotyplc expression, rendering them appropriate for physiological, pharmacological, and biomedical applications 1.1. General

Considerations

Human bronchotracheal submucosal glands have long been recognized as the major secretory structure m the bronchotracheal tree (4). They are composed of mucous and serous cells (Fig. l), surrounded by myoeplthellal cells, and are connected to the bronchotracheal lumen by collectmg ducts (5). The tracheal submucosal tissue is richly innervated, vascularlzed, and also contains smooth muscle, neuroendocrine cells, and mastocytes. These elements are embedded in a parenchyme, which has numerous fibroblasts. Glands are considered the prmclpal source of the secretion of mucus, which 1s involved m the defense of the airway The mucus is a complex mixture composed of various macromolecules. Mucms, probably the most widely known macromolecules of bronchial secretion, originate predominantly from the mucous component of the glands and from the goblet cells m the surface epithelium The other proteins present in mucus stem from serum exudation and also from the serous component of the glands. Gland serous cells secrete antibacterial proteins, such as lactoferrm, lysozyme, and peroxldase (6) In addition, they secrete an From

Methods Edlted

m Molecular by G E Jones

Medune Humana

201

Human Press

Cell Culture Inc , Totowa,

Protocols NJ

Met-ten Broncho-tracheal

lumen

Surface

myoepithelial

eplthellum

cell

Submucosa

Fig. 1. Schematic representation of an HTG Both mucous and serous cells contnbute to the secretion of mucus, which IS then evacuated to the bronchotracheal lumen through collectmg ducts Mucus IS then transported by the clllary beat to the larynx

anttprotease-the bronchial inhibitor-and are the primary site m the lung for secretory IgA transcytosts. After serous and mucous cells have exocytosed their secretory products m the gland lumen, the myoepithehal cells appear to help m the evacuation of mucus, via collector ducts, mto the bronchotracheal tree In many bronchopathologres, there 1seither an overabundance of mucus or a failure m the pulmonary defense system. It 1stherefore valuable to culture tracheal gland cells m order to provide a cellular, molecular, and pharmacologtcal basis for the comprehenston of these diseases. The mtttal problem that arose when the author began the isolatton and culture of human tracheal gland (HTG) cells was the complexity

of the bronchotracheal

mucosa.

This,

m turn,

Human Tracheal Gland Cells

203

prompted the question of how tracheal gland cells could be detected after culture and how a functional homogeneous cell monolayer of this cell type could be obtained.

2. Materials Transport medium for the surgical specimen. RPM1 1640,O 1% LPSR-I (a serum replacement), 10 g/L glucose, 0 33 g/L sodium pyruvate, 200 pg/mL gentamycm, 5 U/mL amphoterlcm B (all provided from Sigma, St. Louis, MO) Storage at 4°C for 1 mo Detachment medium 110 mMNaC1,.5 mMKC1, 1 mMNa,HPO,, 1 mMKH,P04, 20 mMN-2-hydroxyethyl-plperazme-hr-2-ethanesulfomc acid (HEPES), 10 mg/mL of fraction V human serum albumin, 4 g/L glucose, 0 11 g/L pyruvate. and 2 mM ethylenedlammetetraacetlc acid (EDTA), pH 7 4 Store at -2O’C after filter sterlllzatlon m 25mL ahquots, until required DIgestion medium Identical to the detachment medium but without EDTA, and containing 1 mM MgC12 and 2 mM CaC12, m addition to the following enzymes 200 U/mL type IA collagenase (see Note l), 200 U/mL type I-S hyaluronldase, 0 1 mg/mL type I porcine pancreatic elastase, 200 UimL type II DNase (all from Sigma) Store at -20°C after sterile-filtration m 25-mL ahquots, until required Collagen solution The preparation of collagen that was found to be the most adaptable to HTG cell culture 1s as follows Use rat tails, which must first be frozen for at least 1 wk Then soak them m a 95% ethanol solution for 3 mm Subsequently, break the tails mto 1-cm-long segments using strong sterile pliers Dissect the Isolated tendons, and wash them m sterile water. The tendons are gathered and put into a 1 mM acetic acid solution for 24 h at 4°C (100 ml/tall) After centrlfugatlon to remove the nonsolublllzed material, the stock collagen solution can be used Store at 4°C for no more than 2 mo Culture medium (see Note 2) Dulbecco’s modified Eagle’s medium (DMEM), Ham’s F12, 50/50% (v/v) containing 1% LPSR-I (a serum replacement from Sigma), and the followmg substances made up to the mdlcated final concentrations 10 g/L glucose, 0 33 g/L pyruvate; 0 2 g/L leucme, lsoleucme, and valme, 0 I g/L glutamlc acid and cysteme; 3 @Zepmephrme, and antlblotrcs, 100 U/mL pemclllm G and 100 pg/mL streptomycm Glucose, pyruvate, and ammo acids are added to DMEM/Fl2 prior to filtration when the media are prepared from powdered mixtures LPSR-I and antlblotlcs are prepared together in ahquots that will be added Just before each medium change Epmephrme IS also prepared independently (stored at -80°C m a 1 ti HCl solution) and added Just before use

3. Methods

3.1. Isolation of HTG 3.1.1 Dissection of the Tissue 1 Spread out the human bronchial tissue (see Note 3) on a dissection board and carefully clean the mucous material present on the organ surface with sterile gauze

Met-ten

204 2. One hour dlgesbon with 200U/ml of collagenase IA, hyaluromdase, DNAse and 0,l mg/ml elastase

I. Dissetilon

4 Supernalant contammg the dlssoclatsd cells IS centrifuged

of the submucusa A

Cells are then cultur

centrifuged

at 500g for 10mm

Fig. 2. Isolatton of HTG cells After the mucosa of the bronchial tissue have been dissected, they are submttted to successtve enzymattc digestion procedures followed by EDTA treatment Cells isolated at each step are seeded onto collagen substrate, 30 + 10 lo6 cells/g of dissected fresh tissue are obtained Dampen the sample for 10 s with ethanol at 95% This considerably reduces the posstbtllty of bactertal contammatton It also leads to the destruction of the surface epnhelmm Sponge up the ethanol, and thoroughly soak the surface with complete cell-culture medium During the followmg operations, the surface must be kept contmually damp wtth this solutton. The mucosa, submucosa, and parttcularly the ttssue between the carttlagenous rings can now be dissected Cut this tissue mto small pieces (cl mm3) with scissors, and put them mto dtgesnon medium

3.7.2. Enzymatic Digestion of the Dissected Tracheal Mucosae The dissected ttssue can now be submitted to successtve lowed by 15 mm of cell detachment (Fig. 2). The protocol was adapted from descrtptions by Culp et al. (Z) and others (2), which had been perfected for the tsolatton of bovine gland cells

l-h dtgesttons, folfor each operatton by Fmkbemer et al and feline tracheal

1. Incubate for 1 h m gently agitating digestion medium at 37°C (25 mL solution/l 0 cm2 of bronchial sample)

Human Tracheal Gland Cells 2 Centrifuge the tissue for 2 mm at 5Og, and separate the pellet and the supernatant The supernatant and the pellet are now submitted to different operations. 3 Centrifuge the supernatant at 500g for 10 min Discard the new supernatant. The extstmg pellet contains isolated cells. 4 Incubate the first pellet with detachment medium for 15 mm at 37’C, and centrtfuge at 5Og for 2 min to separate the undigested tissue and the detached cells Remove this supematant, and centrifuge tt at 500g for 10 mm to retrieve the cells Submit the undigested tissue to a further 1-h digestion 5 Two pellets of centrifuged cells per digestion operation are obtained After the cells have been counted, seed at an rnmal densrty of 25,00&45,000 cells/cm* (see Note 4) Repeat this operation until all the tissue IS totally digested. The number of successive operations (from three to over eight times) varies accordmg to the age and sex of the donor (see Note 5) 6 The two first digestions yielded few cells, but a lot of cell debris The followmg digestions are much more producttve. As a result, after the full digestion procedure 1sachieved, 30 f 12 106ce11s/gof fresh dissected tissue are actually obtained This value may greatly vary dependmg on the time delay between death and collection of the tissue 3.2. Culture

of HTG Cells 3.2.1. Types of Cultures

Like many epithehal cells, HTG cells are able to grow m monolayers on plastic supports. In VIVO, however, the epitheha are laid on an extracellular matrix composed of collagen, glycoprotems, and proteoglycans This matrix partictpates m the polarity of the cells and plays an important role m then growth and differentiation. HTG cells proved to be very sensitive to the substratum on which they are seeded (7-9). The most convenient substratum was collagen Type 1. This substratum not only promoted cell differentiation, but was the only one to permit an optimal cell growth. There are several different procedures to prepare type I collagen either from calf skin or rat tail, both of which give sattsfactory results. Two types of HTG cell cultures can be realized. 3.2.1 1. MONOLAYER CULTURES In thus culture type, cells grow in two dimensions on thm collagen-ftlmcoated surfaces. A l/100 dtlution of collagen in Hz0 (stock solution) 1s dropped onto the plastic surfaces (0.2 mL/cm2) and left overnight This collagen solution must be removedJust before seeding the cells so that the collagen film does not dry out. Under these conditions, about 10 pg/cm2 of collagen are absorbed. 3.2.1.2.

THE COLLAGEN LATTICE

Bell et al. (ZO) first developed this technique for fibroblast culture, and it has recently been adapted for HTG cells. In this cell-culture type, HTG cells are

206

Merten

first cultivated m the aforementioned condittons and are then introduced mto a thick collagen gel where they grow three dimensionally A complete protocol IS descrtbed m detail elsewhere (II) This protocol was modified to simpltfy the technique, which is as follows, 1 Prepare a mixture, kept in ice, of 1 mL of complete medium containing 2% LPSR-I, 0 5 mL of a collagen type 1 stock solution (2 mg/mL in 0 1% acetic acid/l35 mM NaCl), and 0 25 mL of a 35-d NaOH/lOO mM NaCl 2 Add to this mixture 0 25 mL of complete medium containing 2% LPSR-1 and 10’ HTG cells 3 Transfer immediately onto a 35-mm diameter Petri dish, and place the dish in the 5% CO* incubator The collagen polymerizes to form a thick gel matrix in which cells are embedded 4 After the gel has polymerized, add fresh culture medium After 3 wk of culture, glandular-like branching structures are noticeable within the gel In many aspects, these structures resemble the in viva morphology (12)

3.2 2. Evolution and Set-/al Passaging After isolatton, microscopic exammation indicates the presence of cells of undetermmed phenotypes and abundant cell debris. During the first 4-6 d, culture medium must not be changed At this time, tt IS extremely important to avoid bacterial contammatton (see Note 6) Cells of different origm begin to grow (Fig. 3A). Small islets of highly prohferatmg eptthehotd cells can be dtstmgutshed (8) When cells reach about 75% confluence, tt is necessary to perform a partial trypsmizatton (Fig 3B,C) to separate HTG cells from all other contaminating cells (see Note 7) These Islets ~111 then contmue to grow and ~111 reach confluence within 5-7 d (Fig. 3D). One of the most dtfficult and delicate aspects of HTG cell culturmg 1s the determmation of the moment and the conditions of the passage. Since the cells grow m clusters, m the same flask, there are cells at various states of differenttation. The cells mstde the clusters are differentiated and actively secrete, but have a low prohferatlon rate. The cells at the periphery of the clusters proliferate rapidly and are undifferentiated. The passage consists of a trypsm treatment (0 025% trypsm and 0.02% EDTA ma PBS buffer) over a suffctent pertod of time (3-5 mm) to enable the tsolatton of only the cells growing at the periphery of the clusters. This is possible since the dtfferenttated cells are much more adherent to the substrate. The isolated cells are seeded at 25,000 cells/cm*, and the original flask can be cultured furthernew cells will spread again from the remaining clusters. The progression of HTG cells m then culture life time is accompanied by a decrease m the proliferatton rate The density of the cell at confluence, the vtabillty, and the capacity to secrete also decrease, and characteristics of apoptosis begin to appear This hmitatton of the cell life-span IS a necessary

Human Tracheal Gland Cells

207

Fig. 3. Phase-contrast micrographs of cultured HTG cells. (A) Cells 2 d after isolation. A mixture of cells of a different nature is isolated and will grow on the collagen gel (magnification: x320). (B) Among the cells, clusters of HTG cells are visible 1 wk after isolation (in the middle of the micrograph) (magnification: x 160). (C) The partial trypsinization allows the specific selection of the HTG cells, which are strongly adhesive to the substrate collagen (magnification: x160). (D) Multiple partial trypsinizations lead to an homogeneous monolayer of HTG cells (magnification: x320).

condition of the “normality.” The cell-growth parameters (population doubling time, cell density at confluence, interdivision time, the lag time, etc.) are identical during the first three passages and decrease dramatically thereafter. Culturing HTG cells for more than six passages was not successful.

3.2.3. Phenotypic Expression As for all cell types, there are two methods of control of the phenotypic expression, morphological control and functional control, the latter being

208

Met-ten

the most efficient. HTG cells possess differentiated charactertsttcs that have to be monitored to ascertain whether the cells are completely differentiated. These characteristics include the presence of cytokeratin wtthm the cells, a polarized secretion of the secretory products, which the glands secrete in vivo, and the responsiveness to secretagogs. Figure 4 shows the kinetics of appearance of differentiation characteristics during 30 d of culture. In the exponential growth phase, cells secrete the bronchial inhtbttor at a very low level, which 1s a specific secretory marker of HTG cells. They are also not responsive to secretagogs, nor are they polarized. There are three pertods m the stationary phase* Thts first IS 8 d where cells acqutre then dtfferentiated characteristtcs, followed by a plateau of 1O-l 5 d in which cells are fully dtfferentiated as demonstrated by then ability to be stimulated by many pharmacological and physiological agomsts (7,8,13). During this period, cells are also highly polarized (7), which is demonstrated by the apical secretion (>98%) of all the secretory products of the cells. All these characteristics (htgh polartty, higher constitutive secretion, and responstveness to agonists) appear in unison, and they were only observed when cells were grown on a collagen substratum and m the presence of epmephrme (7). The thud period corresponds to cell degeneration in which cells lose their differentiated features and die. As a consequence, the optimal time to use HTG cells IS after 8 d confluence at the third passage (whrch corresponds roughly to 6 wk after cell tsolation).

3.3. Possible Applications In fundamental as well as m applied research, the HTG cell culture represents an interest m the study of physiology, pathology, and pharmacology of bronchial secretions. Before any mvesttgations are performed, tt 1s important first to check the differentiated functions of HTG cells After these precautions, HTG cell cultures can be used efficiently. Three mam types of studies are usually performed on these cells: the study of their secretory products (which may be m their native form), the study of their pharmacology, and the study of pathologtcal cells. The aim of these studtes is to demonstrate the tmplication of these cells in the comprehension of human bronchial secretion.

3.3.1. Physiology and Pharmacology of HTG Cells Until recently, the physiology and pharmacology of bronchial secretion have not been well understood. HTG cells m culture appear to be one of the most mterestmg models for the study of regulatmg bronchial secretion mechanisms. Constderable studtes have been made on human organ culture, but with the complexity of the human bronchial tissue, it is difficult to assign the action of a phystological agent to a direct or an indirect effect, and also to distmguish

A

Confluency Exponential growth Dha8B

2 b 0

I : !

I

1 I

0

12

16

20

24

26

32

36

40

1 6

I 12

I 16

I 20

I 24

# 26

I 32

8 36

I 40

II

250

000 4

C Percentage of secretion In the apical compartment 0 Percentage of eecretlon In the baeolateral compartment loo

’

I 4

Days

of

culture

Fig. 4. Phenotypic expression of HTG cells. (A) Secretion of the specttic secretory serous marker bronchial Inhibitor during the growth cycle of HTG cells m culture. (B) HTG cells begin to respond to secretagogs (a cholmergtc agonist. carbachol and a P-adrenerglc agonist tsoproterenol) only after 8 d confluence and stay responsrve for 15 d (C) Polarrzatton of HTG cells also appears after 8 d confluence as demonstrated by the almost complete aprcal secretron of the bronchial mhrbltor 209

210

Met-ten

secretion dertved from the glands and from other secretory cells m the tissue (i.e., the goblet cells) Despite the great interest m the use of HTG cells m culture to understand the regulation of bronchial secretton, very few studies are avatlable to date. Cultured HTG cells were shown to be responstve to bradykmm and htstamme (14), to cholmergtc and adrenergtc agonists (7,8), and to the purmergtc agents ATP and UTP (13). One mterestmg observatton was the evtdence of a posrttve and negative control of secretton by HTG cells They respond to the phystologtcal neurotransmitters acetylcholme and norepmephrme by an Increase or a decrease m secretion, respectively (2.5). Furthermore, neuropepttde Y was shown to modulate the decrease m secretion Induced by norepmephrme (16) HTG cells appear to be particularly responsive to many agents that are present m the atrways, the action of whtch, at cellular and molecular levels, has only recently been documented These mvesttgatlons can be relatively easy to perform using the patch-clamp techmque, Ussmg chamber, or by measurmg some secretory products by ELISA measurements (bronchtal mhtbttor, lysozyme, or lactoferrme, for instance).

3 3.2 The Cell Phenotype of HTG Cells rn Culture HTG cells m culture were shown to secrete lysozyme, lactoferrm, and the bronchtal mhtbttor, which are proteins specific to the serous gland cells. However, gel-filtratton analyses of 3H-fucose and 35S-S04 labeled secretory products m the medium of the cultures m a Sepharose CL 6B column Indicated that the secreted htgh-mol-wt radioconJugates were partly proteoglycans (chondrottm and heparan sulfates), and also mucms as proved by resistance to all hyaluromdase, chondrottmase, heparmnase, keratanase and also by senstttvtty to p-eltmmatton (I 7) Furthermore, the buoyant density of these hydrolasereststant radiolabeled macromolecules, as well as the stzes of then glycamc chains are conststent with those expected from mucms, but not from proteoglycans or other glycoprotems. These results suggest that HTG cells are able to produce mucms m vitro. Tourmer et al (8) showed that cultured homogeneous HTG cells are mostly composed of serous cells. Sommerhof and Fmkbemer (18) observed, using specific antibodies directed against serous or mucous epttopes, an tmmunolabeling of all cells wtth both types of antibodies, suggesting that, at least m culture, HTG cells may carry both the serous and the mucous phenotypes According to the culture condtttons, HTG cells may express one phenotype more than the other Fmkbemer et al. (9), by varying cell-culture condmons, were able to obtain etther HTG cells of the serous or of the mucous phenotype. When HTG cells are cultured onto collagen, they dtfferenttate mto serous cells, but when they are cultured onto vttronectm, they differentiate mto mucous cells, HTG cell culture therefore permits both cell types present m the glands

Human Tracheal Gland Cells

211

to be obtained. In addition, it is possible to study the mechanism of transdifferentiatton and also the regulation mechanism of secretion of either mucm or protein from the serous cells

3 3.3. Cyst/c Fibrosis Cystic fibrosts derives from mutations m a membrane protein called cystic fibrosis transmembrane conductance regulator (CFTR), which 1s mvolved m the cychc AMP-dependent chloride transport Engelhardt et al. (19), followed by Jacquot et al (20), showed that m compartson to the other eptthehal cell types of the human bronchus, CFTR 1s located predominantly m tracheal glands This may signify that tracheal glands could be one of the prmctpal targets of cystic fibrosis. This disease is an exocrinopathology, the pulmonary syndrome of which 1s the most serious and which 1s characterized by mucus hypetsecrenon and lung mfectron It is useful to mvestigate the imphcations of CFTR m cultures of HTG cells provided from cystic fibrosis patients. The first data available showed a failure m the chloride transport by HTG cells m culture (21) Becq et al. demonstrated that cultured HTG cells express great quantttles of CFTR and the associated cychc AMP-dependent chloride channel activny (22). A constttuttve protein hypersecretton and an msensttivtty to secretagogs and phystologtcal neurotransmttters by HTG cells derived from cystic fibrosis patients (15) were observed These alteratrons m secretory mechanisms can be related to the macroscoptc syndrome, but until now could not be explained by the known function of the molecule CFTR

3.4. Conclusion HTG cells tend to lose then differentiating functions m culture, but differentiated eptthelial cells can be attained m very specific condittons These condltions were determmed by analyzmg the growth-supplement requirements, the substrate requn-ement, and the culture methods The first obJective was to establish conditions that allowed HTG cells to grow and especially differenttate m culture It was observed that a collagen substrate, an elevated concentration of glucose, and supplementation with epinephrme are important to HTG cell growth and differenttatton. As a result, a 1O-cm2 surgical resected sample, for example, may produce a culture comprismg more than 100 24-well plates. This relatively large quantity of cells 1s suffictent to realize many fundamental or applied studies regarding the action of phystologtcally or pharmacologtcally active substances on: receptors, membrane tonic channels, and mechanisms of effector secretion couplmg The study of human secretion, notably the complex trachea-bronchial mucosa, remains relatively unexplored, parttcularly at the cellular level Prtmary cultures of HTG cells seem an obvtous mterestmg experimental model.

212

Met-ten

Cell cultures prove then usefulness m certain mvestlgatlons and especially m the study of pathologies. Current culture procedures render the HTG cell culture one of the best models available for the study of the human bronchial secretion. 4. Notes 1 A notrceable trend m the techmcal factors observed m these studies IS the crucial importance of the nature of collagenase used m HTG cell isolation procedures The commercial products, called collagenase, are enriched pooled fracttons containing several collagenases, neutral proteases, and clostripam The composition of “collagenase” 1s greatly dependent on manufacturing processes, and attention must be given to the lot number The findmgs regarding the respective tested lots have been confirmed by other teams involved m HTG cell culture (F Dupun, INSERM U3 14, Reams, France, personal commumcatron, 1993) These results demonstrate the need for an efficient and constant form of collagenase, smce an absence of viable cells or any vtsible cell growth after tsolatron could be attributed to a toxic “collagenase” lot In thus case, it is necessary to change to another lot number of type IA collagenase 2 When the ability of different sera to promote HTG cell growth and differentiation was tested, a problem occurred with heterogeneity and the weak efficiency of the widely used fetal calf sera probably owing to some cytotoxrc stde effects The serum substitute Ultroser G (Biosepra, Villeneuve-la-Garenne, France) also marketed with the name LPSR-I (Sigma) yielded optimal and reproducible cell growth, and also dtfferentratron at confluence as Judged by polarity, cell secretion, and responsiveness to secretagogs The manufacturer advises the use of LPSR-I at the concentration of 2% By studymg the effects of mcreasmg concentrations of LPSR-I, rt was determined that 1% was the most efficient The DMEM/F 12 medium mixture contains 3 12 g/L glucose and 0 11 g/L sodmm pyruvate A rapid change m the pH of the culture medium was observed, which was attributed to an active metabolism From the measurements of the glucose, the pyruvate, and the ammo acids that are consumed by HTG cells, tt was concluded that the final concentration of these components had to be increased It is worth noting that cells use many components that are mvolved m the formation of acyl CoA, mdicatmg the importance of the metabolic activity of these cells Epithehal cells m culture can termmally differentiate, turning mto squamous and stratified or secretory and columnar cells There are factors favoring the drfferentiatton mto squamous cells, such as TGFPl and an elevated calcium concentration, but there are also other factors favoring differentiation mto columnar cells Epmephrme is countereffective to TGFPl in mducmg squamous drfferentianon (23) Epmephnne (at a 3-pA4 final concentratton) was able to promote cell growth and glandular cell drfferentiatron features, such as a high polarity and an optimal responsrveness to secretagogs at confluence (7). Add epmephrme at each medium change (every other day) from ahquots of stock solutton of epmephrme stored at -80°C m a 1-mM HCl solution

Human Tracheal Gland Cells

213

3 A prmctpal factor in the successful gathering of viable tracheal gland cells IS the time delay at room temperature before the collectton of explants Samples resected durmg surgery are of a higher quality than those obtamed after autopsy owmg to the delay that 1s mevltable between sampling and death The author rarely succeeded m obtauung satisfactory cultures from autopttc samplmg It 1s Important to put specimens as rapidly as possible mto a sterile transport medium at 4°C When samples are obtained m these condtttons, they can be stored at 4°C for several days after surgery without srgmficant reduction m either the vtabtlrty of cells or the number of isolated cells. 4. The uuttal seeding of 25,000 cells/cm2 was determined as the most convenient for growth A major problem of surgically resected samples 1s that they are frequently small m size, and the tsolatton of HTG cells from the obtained tissue IS often dtfficult However, HTG cells can grow even If their inmal quantmes or densrttes after tsolatton are low, but the presence of other cell types IS necessary to generate HTG cell cultures from very little bronchial tissue. In the case of msufficlent tissue, Instead of destroying the surface epttheltum by the ethanol treatment, the surface eptthelmm 1s removed with the submucosa, and all the dissected tissue has to be digested By mcreasmg the amount of viable tissue digested, 1 e., the total number of isolated cells, the capacity of HTG cells was augmented to develop despite their much lower percentage. The followmg parteal trypsmtzattons (see Note 7) ~111 be more carefully undertaken In addttton, Instead of adding fresh culture medium, a mixture of 50% fresh medtum and 50% of an older medium that has already been m contact with cells from a previous cell culture can be added (because of the presence of specific autocrme growth factors). Using these methods, the author succeeded m performmg HTG cell cultures from bronchtal tissue wrth as httle surface space as l/2 cm2 5 The age of the donor 1s a significant factor that must be constdered when collectmg specimens. Results m terms of number and vtabtllty of isolated cells were identical whatever the age of the donor However, the number of populatton doublmgs was dramattcally dependent on the age-the younger donors gtvmg very sattsfactory and useful cultures 6 Whether the samples are autoptic or surgical resections (most of which have been removed from artificrally ventilated patients), there 1s always a risk of contamtnation of the material by mtcroorgamsms, and care must be taken equally during experiments and m the stertllzatton of the material. Both the ethanol treatment of the tissue surface and the use of anttbtottcs m all media and solutrons constderably reduce the risk of contammatton. However, during the first 5 d of cultures, m addition to the habitually used penictllm G (100 U/mL) and streptomycm (100 pg/mL), add the broad-spectrum antlblotrc gentamycm (100 ug/mL), the antifungal agent amphotertcme B (5 pg/mL), the antrmycoplasma agent neomycme (50 ug/mL), and 5% of unheated rabbit serum, which 1s also used to ehmmate potential mycoplasma contammatton Despite these precautions, for each culture, a classic detection IS necessary to verify the absence of mycoplasms Other rare, but opportumsttc contammattons are the multtreststant Stuphylococ-

214

Merten

CMSaureus and Staphylococcus epldermldls Treatment using vancomyctn (50 yglmL) IS suffictent to eradicate thts contamination, but it can alter HTG cell growth and physiology if treatment ISprolonged for more than 3 or 4 d 7 A significant techmcal stumbling block m obtaining homogeneouscell culture 1s the risk of contamination by cells from different origins After isolatton, careful mtcroscoptc observatton allows different types of bronchial tissue cells to be dtstmgutshed Four types of different cells can be observed a Surface epnheltal cells are big, flattened, and mostly regrouped m clusters These cells are easily recognizable b Fibroblasts, the most abundantof contaminating cells, organize themselvesm regularparallelbundlesFtbroblastsgrow rapidly and spreadover all free surfaces c Endothehal cells, whtch originate from vesselspresent m bronchial tissue, are polyedric, Joined together, and proliferate at the samerate as the HTG cells They are dtfficult to distmguish from epttheltal cells and are pmpomted by the absenceof labeling by anttcytokeratm antibodies and by the presence of the Von Wtllebrand anttgen d Smooth muscle cells are eastly obtamed durmg ttssuedigestion owmg to both the presenceof elastasem the digestion medium and the useof collagen type I, which stimulates smooth muscle cell growth The processusedto obtain a homogeneousmonolayer composeduniquely of HTG cells dependson the partial trypsmtzatton methodology This technique is based on the fact that HTG cells are fixed firmly to the collagen substratum A 3-mm mcubation with trypsm/EDTA detachesthe fibroblasts, the myocytes, and most of the endotheltal cells. These contammatmg cells can then be washed out Eptthehalcellsseparateafter 7 or moreminutesof rncubattonwith trypsm/EDTA This short-time trypsmtzatton has to be performed before each passageand one or two times more during the exponential growth phaseof the first two subcultures. This operation not only allows the selection of epttheltal cells m the culture, but also an improved cell growth, since trypsmtzatton appearsto have a stimulatory effect (probably by an msulm-like effect) on the HTG cells It was ascertainedthat the cells obtained under these condtttons are HTG cells and not epithelial cells from any remammg surface epithelmm, since the cell-culture medium used for the HTG cells 1snot adapted for the surface epttheltal cells and prohibits their growth In the laboratory, a homogeneousmonolayer of HTG cells is obtained at the third passagewith four or five previous partial trysuuzatrons After tsolatton, the first partial trypsmizatton must be performed at about 75% of confluence of the cell population of the flasks

Ackowledgments The author is on a fellowshrp from Synthelabo and the “Fondatton pour la Recherche Medicale” awards and his works on this subject were supported by grants from the Assoctatton Francatse de Lutte contre la Mucovtscrdose The author thanks Anme Mascall for her help with the English text and Catherine Ftgarella for advice and fiunful dtscusstons

Human Tracheal Gland Cells

215

References I Gulp, D J., Penney, D P , and Mann, M G (1983) A techmque for the tsolatton of submucosal gland cells from cat trachea Am J Physzol 55, 1035-104 I 2 Fmkbemer, W E , Nadel, J A , and Basbaum, C B (1986) Estabhshment and charactertzatton of a cell lme derrved from bovine tracheal glands In vitro 22, 561-567 3 McBrtde, R K , Stone, K K , and Marin, M G. (1992) Arachtdomc acid increases cholmergtc secretory responstveness of ferret tracheal glands Am J Physzol 262, L694-L698 4 Read, L (1960) Measurement of the bronchial mucous gland layer a diagnostic yardstick m chronic bronchttts Thorax 15, 132-14 1 5 Meyrtck, B , Sturgess, J M , and Read, L (1969) A reconstructton of the duct system and secretory tubules of the human bronchial submucosal gland Thol-ax 69,729-736 6 Basbaum, C B , Jany, B , and Fmkbemer, W E (1990) The serous cell Annu Rev Physzol 52,97-l 13 7 Merten, M. D , Tourmer, J M , Meckler, Y , and Ftgarella, C ( 1993) Epmephrme promotes growth and dtfferenttatton of human tracheal gland cells m culture Am J Respw Cell Mel Blol 9, 172-178 8 Tourmer, J M , Merten, M , Meckler, Y , Hmnrasky, J , Fuchey, C , and Puchelle, E (1990) Culture and charactertzatton of human tracheal gland cells Am Rev Respzr Dzs 141, 128&1288 9 Fmkbemer, W E , Shen, B Q , Mrsny, R J , and Wtddrcombe, J H (1993) Inductron of mucous phenotype m cultures of glands from human atrways leads to loss of CFTR and chloride secretion Pedlatr Pulm 9, 187 10 Bell, E , Ivarssen, B , and Merrtl, C. (1979) Productton of a tissue like structure by contractton of collagen lattices by human fibroblasts of different prohferattve potential in vitro. Proc Nat1 Acad Scz USA 76, 1274-1278 11 Benalt, R , Tourmer, J M , Chevtllard, M., Zahm, J. M , Klosseck, J M , Hmnrasky, J , Galllard, D , Maquart, F. X., and Puchelle, E (1993) Tubule formation by human surface respiratory eptthehal cells cultured m a three-dimenstonal collagen lattice Am J Physlol 264, L183-L192 19 IL. Jacquot, J , Sptlmont, C , Burlet, H , Fuchey, C , Butsson, A. C , Tourmer, J M , Galllard, D , and Puchelle, E. (1994) Glandular like morphogenests and secretory activity of human tracheal gland cells m a three-dtmenstonal collagen gel matrix J Cell Physzol 161,407-418 13. Met-ten, M D , Breittmayer, J P., Figarella, C , and Frelm, C (1993) ATP and UTP increase secretion of the bronchial mhtbttor by human tracheal gland cells m culture Am J Physzol 265, L479-L484. 14 Yamaya, M , Fmkbemer, W E , and Wtddtcombe, J. H (1991) Ion transport by cultures of human tracheobronchtal glands Am J Physzol 261, L485-L490 15 Merten, M D and Ftgarella, C (1993) Constttuttve hypersecretton and msensttrvtty to neurotransmttters by cystic fibrosrs tracheal gland cells Am J Physzol 264, L93-L99

216

Met-ten

16 Merten, M D and Flgarella, C (1994) Neuropeptlde Y and norepmephrme cooperatwely mhtblt tracheal gland cell secretion. Am .I Phys~ol 266, L5 I3-L5 18 17 Merten, M D , Tourmer, J M , Meckler, Y , and Flgarella, C (1992) Secretory protems and glycoconjugates synthesized by human tracheal gland cells m culture Am J Respw Cell Mel Blol 7, 598405 18 Sommerhof, C P and Fmkbemer, W E (1990) Human tracheobronchlal submucosal gland cells In culture Am J Respv Cell Mel Bzol 2,41-50 19 Engelhardt, J F , Yankaskas, J R , Ernst, S A , Yang, Y P., Marmo, C R , Boucher, R C , Cohn, J A , and Wilson, J M (1992) Submucosal glands are the predominant site of CFTR expresslon m the human bronchus Nature Genet 2, 24G246 20 Jacquot, J , Puchelle, E , Hmnrasky, J , Fuchey, C , Bettmger, C , Spllmont, C , Bonnet, N , Dleterle, A , Dreyer, D , Pavlram, A , and Dalemans, W (1993) Locahzatlon of the cystic fibrosis transmembrane conductance regulator m airway secretory glands Eur Respw J 6, 169-176 21 Yamaya, M , Fmkbemer. W E , and Wlddlcombe, J. H (1991) Altered ion transport by human tracheal glands m cystic fibrosis Am .I Physzol 261, L49 l-L494 22 Becq, F., Merten, M D , Voelkel, M A , Cola, M , and Flgarella, C (1993) Characterization of CAMP dependent CFTR-chloride channels m human tracheal gland cells FEBS Lett 321(l), 73-78. 23 Masui, T , Wakefield, L M , Lechner, J F , Laveck, M A, Sporn, M B , and Hams, C (1986) Type p transforming growth factor IS the primary dlfferentlatlon-inducing serum factor for normal human bronchial eplthellal cells Proc Nat1 Acad Scz USA 83, 1438-1442

19 Human Chondrocyte of Cartilage-Specific

Cultures as Models Gene Regulation

Mary B. Goldring 1. Introduction Chondrocytes comprise the single cellular component of adult hyalme cartilage and are constdered to be termmally differentiated cells that maintain the cartilage matrix when turnover is low. The major components of the extracellular matrix synthesized by these spectalized cells Include highly crosslinked fibrrls of the final synthesized and secreted trtple helical type II collagen molecule that interact with other cartilage-specific collagens IX and XI, the large aggregating proteoglycan aggrecan, small proteoglycans, such as btglycan and decorm, and other spectfic and nonspectfic matrtx protems that are expressed at defined stages durmg development and growth (2,2). This matrix confers tensile strength and flexibility to arttcular surfaces and serves specialized functions m only a few other tissues Cultured chondrocytes have served as useful models for studymg chondrocyte differenttatton and the effects of cytokmes and growth factors that control the maintenance or suppression of differentiated cartilage phenotype (3). Freshly isolated human articular or costal chondrocytes express carttlagespecific type II collagen and conttnue to do so for several days to weeks m prrmary monolayer culture (4,5). During prolonged culture and serial subculture, these cells begin to express type I and type III collagens. This “switch” can be accelerated by plating the cells at low densities or by treating them wtth cytokines, such as tnterleukm-1 (IL-l) (.5,6). Early attempts to culture chondrocytes from various animal and human sources were frustrated by the tendency of these cells to “dedifferenttate” to a fibroblast-like phenotype m monolayer culture and then mabthty to proliferate in suspension culture where cartilage-spectfic phenotype was maintained (7-9). This loss of phenotype m From

Methods in Molecular MedIcme Human Cell Culture Edited by G E Jones Humana Press Inc , Totowa,

217

Protocols NJ

218

Goldring

monolayer culture was found to be reverstble if they were placed m suspension cultures m spinner flasks (20) or m dishes coated with a nonadherent substrates (II, Z2), tf they were kept at high denstty m mtcromass cultures (13,14), or if they were embedded m a solid support matrices, such as collagen gels (Z5), agarose (Z&18), or algmate (19,20) The basement membrane-type matrix commercially known as Matrtgel TM has also been shown to support mamtenance of chondrocyte phenotype, probably because of presence of growth and differentiation factors, such as IGF-I, that copurify with the matrix proteins (21) Serum-free defined media of varying composttions, but usually mcludmg msulm, have also been used, frequently m combmation wtth the other culture systems mentioned (22). Primary cultures of chondrocytes isolated from young animals that mamtam the cartilage-specific phenotype at least throughout primary culture are easily obtained and have been used widely to assess differentiated chondrocyte functions. The use of chondrocytes of human origm has been more problematical, because the source of the cartilage cannot be controlled, sufficient numbers of cells are not readily obtained from random operative procedures, and the phenotypic stabthty of adult human chondrocytes is lost more quickly on expansion m serial monolayer cultures than that of cells ofluvemle human (4) or young or embryonic animal origin (23,24) Viral oncogenes have been used to generate mrmortahzed chondrocytes from nonhuman sources that demonstrate high proliferative capacities and retain at least some differentiated chondrocyte properties (25-29). Also, chondrocyte lines have arisen spontaneously from fetal rat calvaria (30,31). Human chondrosarcoma cell lines express some aspects of the chondrocyte phenotype, but are tumorigemc (32,33) The lack of a reproducible source of chondrocytes of human origin has hampered progress m studies of cartilage function relevant to human disease Recently, the successful mnnortallzation of human chondrocytes was reported using SV40-contaming vectors (34). These cells have a high prohferative capactty that can be dissociated from their abihty to express chondrocyte-specific phenotype by using permisstve culture condttions. These cultures were established and characterized based on conditions and criteria previously defined for culturing normal human chondrocytes (4,5,35), as described m this chapter.

2. Materials 2.7.lso/ation

and Culture of Human Chondrocytes

1 Cell-culture reagents Dulbecco’s Modified Eagle’s medium (DMEM) contammg 10% fetal calf serum (FCS), Dulbecco’s phosphate-buffered Ca2+- and Mg2+free saline (PBS), trypsm-EDTA solution. These solutions have shelf hves as recommended by the supplier If DMEM IS prepared from powder, high-quaky distilled and deionized water (1 e., using a Mllh-Q apparatus), dedicated steril-

Human Chondrocyte Cultures

219

rzed bottles, and 0 22-pm filters should be used FCS and trypstn-EDTA are stored at -20°C but should not be refrozen after thawing for use 2 Hyaluromdase, I mg/mL m PBS. Prepare freshly and filter through a sterile 0 22-urn filter 3 Trypsm, 0 25% m Hank’s balanced salt solutron (HBSS) without Ca2+ and Mg2+ 4 Collagenase (bacterial, clostrtdropeptrdase A), 3 mg/mL m DMEM wtth 10% FCS for artrcular cartilage or serum-free for costal cartilage Prepare freshly m me-cold DMEM, and filter mrmedtately through a sterile 0 22-urn filter

2.2. Suspension

Cultures

1 Agarose-coated dishes Weigh out 1 g of high-melting-point agarose m an autociavable bottle, and add 100 mL of dH20 Autoclave with cap tightened loosely, allow to cool to -55°C and quickly prpet mto culture dashes (1 mL/3 5-cm, 3 mL/6-cm, or 9 mL/lO-cm ttssue-culture or bacterrologrcal dish) Allow the gel to set at 4’C for 30 mm, and wash the surface 2-3 trmes wrth PBS Use plates nnmedtately, or wrap trghtly with plastic or for1 wrap to prevent evaporation, and store at 4°C 2 Sohd suspenston culture m agarose gel. Autoclave 2% (w/v) low-gellmg-temperature agarose m dH20, cool to 37°C and dilute with an equal volume of 2X DMEM contammg 20% FCS etther without cells or wrth a chondrocyte suspension 3 Suspensron culture m algmate beads Low-vrscostty algmate, 1% (w/v) m 0 15M NaCl Stir to dissolve alginate m NaCl solutron and filter sterrhze Prepare stertlefiltered 0 15M NaCl and 102 mM CaCI,

2.3. Analysis

of Matrix

Protein

Synthesis

1 Staining for metachromattc matrix: 0 05% Alcran Blue 8GX and 2 5% glutaraldehyde m 0 4M MgCI, and 25 mM sodium acetate, pH 5.6 2 Btosynthetrc labeling of collagens* 10X solutton of ascorbic acid (ASC) and P-ammoproprromtrtle fumarate (P-APN), 5 mg of each drssolved m 10 mL of serum-free culture medmm Sterrle-filter and dilute at l/10 (v/v) to give the volume of 1X ASC/P-APN solution required for the mcubatton Prepare freshly, 25 uCr/mL L-[5-3H]prolme (1 mCl/mL, SA > 20 Ci/mmol), 50 ug/mL ASC, and 50 pg/mL j3-APN m serum-free culture medium Add 25 uL of [3H]prolme/mL of 1X ASC@-APN solution using a stertle ptpet tip 3 Collagen typmg. 2 mg/mL pepsin m 1M acetic acid. Dissolve 2 mg pepsin/ml dH20, then add 58 pL of glacral acetrc acrd/each mL of solutron, and cool on Ice Prepare freshly the volume required for the experiment. Reagents for Laemmlr SDS-PAGE are prepared as stock solutions 4. Proteoglycan syntheses. 8.OM guamdme-HCl, buffered with 0 OlM sodium acetate, and containing 0 02Mdrsodium EDTA, 0 20M6-ammocaprorc actd, with 5 0 rnA4 benzamtdme HCI, 10 n-J4 N-ethylmalermtde, and 0 5 mM PMSF added rmmedrately before use 5 Immunocytochemrstry 2% paraformaldehyde m O.lMcacodylate buffer, pH 7.4 Dissolve 10 g of paraformaldehyde m 150 mL dH20 m Erlenmeyer flask on hot

220

Goldrmg plate 1n fume hood (do not exceed 65°C) Add -2 mL of 1NNaOH while stIrrIng, and stir unt11 solution IS clear Let solution cool for 15 m1n Add 250 mL of 0 2M cacodylate buffer, pH 7 4, and adjust pH 1f necessary

2.4. Analysis of mRNA 1 Guanldme lys1s buffer 5M guanldlne monothlocyanate, 10 nnI4 EDTA, 50 nuI4 TnsHCI, pH 7 5 Add stock solutions of 1 mL of 0 5MEDTA, pH 8 0, and 2 5 mL of 1 OMTns-HCl, pH 7 5, to 5Mguan1dlne thlocyanate to volume of 50 mL Freshly add -72 pL of P-mercaptoethanol @-ME)/10 mL of buffer. 2. 4ML1Cl. Dissolve 84 8 g of bth1um chloride m -4OW50 mL of DEP-H20, allow to cool to room temperature before bringing up to final volume of 500 mL, and sterile-filter using disposable sterile vacuum flask with cap Store at -20°C 3 Solublllzatlon buffer 0.2% SDS, 1 0 mM EDTA, 10 mM Tns-HCl, pH 7 5 Add stock soluttons of 0 5 mL of 20% SDS, 0 1 mL of 0 5M EDTA, and 0 5 mL of 1M Tns-HCl, pH 7 5, to final volume of 50 mL 1n DEP-H,O Store at -20°C 4 Phenol reagent (phenol chlorofotmlsoamyl alcohol [25 24 11) Prepare freshly using phenol saturated with Tns-HWEDTA, pH 8 0, or obtain commercially 5 75% Ethanol. Dilute bottled absolute ethanol with DEP-treated water 1n sterile tube Store at -20°C

2.5. Analysis of Gene Regulatory Sequences by Transient Transfection Assays 1 2X HBSS* 50 mMHEPES, 1 5 mMNa,HPO,, 10 mA4KCl,280 mA4NaC1, 12 mM glucose, pH 7 05 Note Use an accurately calibrated pH meter, since the pH IS cr1t1cal for formation of a fine CaPO,/DNA preclprtate Filter-ster111ze and store at 4°C 2 2M CaCl, Filter sterilize and store at room temperature 3 15% Glycerol 1n HBSS Combine 30 mL of 50% (w/v) glycerol 1n dH,O, 50 mL of dH*O, and 20 mL of dH,O Filter-sterilize and store at 4°C 4. Prepare CaP04/DNA precipitate by combining m order. Hz0 plasmid DNA 2M CaCl, 2X HBSS

to make final volume of 1 mL 1O-25 pg 62 pL 500 uL

Gently add DNA and CaC12 to H,O without mixing. Use 1-mL p1pet attached to an automatic p1peter with tip placed m bottom of tube to add slowly the 2X HBSS buffer and gently release -5 bubbles to mix. Allow precipitate to form for 15-30 mm at room temperature

3. Methods 3.7. Isolation

and Culture of Human Chondrocytes

1. Human adult arucular cartilage 1s obtained from knee Joints or hips after surgery for Joint replacement or at autopsy, and dIssected free from underlying bone and

Human Chondrocyte Cultures

221

Table 1 Culture Vessel Area vs Chondrocyte Number for Plating Density of -2.6 x lo4 cells/cm2 Diameter 16-mm well 2 2-cm well 3 5-cm plate 6-cm plate 1O-cm plate

Required

Area, cm2

No. of cells plated

2 38 10 28 79

50,000 100,000 250,000 750,000 2X 106

any adherent connective tissue Juvenile costal cartilage 1s obtained from rtbs removed during pectus excavatum repair and dissected free from perichondrmm Place shces of cartilage m 1O-cm dishes containing PBS and use - 10 mL vol of protemases for dtgestion of each gram of tissue. Wash shces several ttmes m PBS, and incubate at 37’C m hyaluromdase for 10 mm followed by 0 25% trypsm for 30-45 mm with 2-3 washes m PBS after each enzyme treatment Add collagenase solution, chop the cartilage in small pieces usmg a scalpel and blade, and Incubate at 37°C overnight (18-24 h) for articular cartilage and up to 48 h for costal cartilage until the cartilage matrix is completely dtgested and the cells are free in suspension (see Note 1) Break up any clumps of cells by repeated aspiration of the suspension through a IO-mL ptpet or a 12-cc syringe without needle. Transfer cell suspension to a sterile 50-mL conical polypropylene tube, wash the plate with PBS to recover remammg cells, and combme m tube Centrifuge cells at -lOOOg m benchtop centrifuge for 10 mm at room temperature and wash cell pellet three times with PBS, resuspendmg cells each time and centrifuging. Resuspend the final pellet m culture medium containing 10% FCS, count with a Coulter counter or hemocytometer, and bring up to volume with culture medium to give 1 x 1O6cells/ml For monolayer culture, plate the cells at - 1 3 to 2 6 x 1O4 cells/cm2 (Table 1) m dishes already containing some culture medium, and agitate without swnlmg to distribute the cells evenly Incubate at 37’C m an atmosphere of 5% CO2 m air with medium changes every 34 d, as described (4,5) (see Note 2) Primary cultures of adult articular and Juvenile costal chondrocytes incubated m the absence and presence of IL-l are shown m Fig. 1A-D Preparation of subcultured cells (see Note 3) Remove culture medium by aspnanon with a sterile Pasteur pipet attached to a vacuum flask, and wash with PBS Add trypsm-EDTA (1 mL/lO-cm dish), and incubate at room temperature for 10 mm with periodic gentle shaking of dish and observation through microscope to assure that cells have come off the plate If significant numbers of cells remain attached, continue the mcubation for a longer time (120 mm) or higher temperature (37°C) and/or scrape the cell layer with a stertle plastic scraper, TeflonTM policeman, or syringe plunger Repeatedly aspirate and expel the cell suspen-

Goldring

Fig. 1. Phase-contrast micrographs of human adult articular and juvenile costal chondrocytes in primary culture. (A,B) Articular chondrocytes at d 13 of culture and (C,D) costal chondrocytes at d 14 were photographed after incubation for 24 h in culture medium alone (A and C) or with 5 pM IL- 1p (B and D). (E) The same costal chondrocyte preparation as shown in C and D was left in primary culture for 2.5 mo, trypsinized, and incubated for 1 wk in suspension culture. Note that most of the cells in clumps have begun to break up and form a single-cell suspension.

sion into the plate using a 5- or lo-mL pipet containing culture medium, and then transfer to a sterile conical 15- or 50-mL polypropylene tube. Perform cell counts, or determine the split ratio required, distribute equal volumes of the cell

Human Choncirocyte Cultures suspenston m dashes or wells that already contam culture medtum, and agttate the culture plate tmmedtately after each addttton to assure umform platmg denstty on the culture surface (see Note 4)

3.2. Suspension

Cultures

1 Flmd suspension culture above agarose (see Note 5) Trypsunze monolayer cultures, spm down cells, wash wtth PBS, centrifuge, and resuspend m culture medium at 1 x 10h cells/mL Plate chondrocyte suspenston m culture medtum contammg 10% FCS m dishes that have been coated with 1% agarose and culture for 2-4 wk Change the medtum weekly by carefully removmg the medmm above settled cells while ttltmg the dish, centrtfugmg the remammg suspended cells, and replacing them m the dish after resuspension m fresh culture medtum The cells first form large clumps that begin to break up after 7-10 d and eventually form smgle-cell suspenstons (Ftg. 1E) 2 Recovery and analysts of agarose suspension cultures To recover cells for direct experimental analysts, for redtstrtbutton m agarose-coated wells, or for culture m monolayer, transfer the cell suspenston to 15- or 50-mL conical tubes, wash agarose surface at least twice wtth culture medium to recover remammg cells, and spin down and resuspend cells m an appropriate volume of culture medium for replatmg or in wash or extraction buffer for subsequent expertmental analysts 3 Solid suspension culture m agarose Precoat plasttc tissue-culture dishes with cell-free 1% agarose m culture medium (0 5 mL/3 5-cm, 1 5 mL/ 6-cm, or 4 5 mL/ IO-cm dish), and allow to gel at room temperature Add the same volume of 1% agarose medium contammg chondrocytes at a density of l-4 x 10h cells/mL of gel, incubate at 37°C for 2&30 mm to allow the cells to settle, and leave at room temperature until the agarose IS gelled Add culture medium contammg 10% FCS, and incubate at 37’C with medium changes every 3-4 d 4 Analysts of solid agarose cultures. After incubations with test reagents and/or radtotsotopes m mmlmal volumes of approprtate culture mednun, the whole cultures may be stored frozen, or medium and gel can be treated separately For subsequent analysts, add appropriate guamdme extraction buffer directly to the gel (for proteoglycan or RNA extraction), or whole cultures may be adjusted to 0 5M acetic acid, treated with pepsm, and neutralized, as described m Section 3 3 , step 3 for analysts of collagens To remove agarose and debrts, the samples are centrifuged at htgh speed, e g , m Eppendorf centrifuge at top speed, 4°C 5 Suspension culture m algmate (see Note 6) Resuspend pelleted cells m sterile 0 15M NaCl contammg low-viscostty algmate gel (1%) at a density of l-4 x IO6 cells/ml, and then slowly express through a 22-gage needle m a dropwtse fashton into a 102-mA4 CaCl, solutton Leave the beads to polymerize further m the CaC12 solution for 10 mm at room temperature Decant CaCl, solutton, and wash three ttmes m 0 15MNaCl and once m culture medmm, usmg 10 volumes of each wash solutton/volume of packed beads, decantmg each wash after the beads have settled Ptpet the beads mto culture dishes or wells (7G80 beads/ml of medium), and incubate at 37°C For medium changes, carefully ptpet the culture medium from the top of the settled beads

224

Goldring

6 Analysis of algmate cultures+ Decant the growth medium, and solubillze the algmate gel by addmg 3 vol of a solution of 55 mA4 Na cm-ate m 0 194 NaCl for 10 mm at 37°C Add appropriate guamdme extraction buffer or centrifuge at 5OOg for 10 mm to recover the chondrocytes with pericellular matrix or at 20008 for 5 mm followed by trypsm-EDTA treatment to recover dispersed cell suspensions.

3.3. Analysis of Matrix Protein Synthesis 1 Stammg for metachromattc matrix Remove culture medium, and wash wtth PBS Add Alcian Blue/glutaraldehyde solution at room temperature for several hours, remove excess stain by washing with 3% acetic acid, and store cultures in 70% ethanol for subsequent exammation by hght photomicrography 2 Biosynthetic labelmg of collagens (see Note 7) Remove serum-contammg culture medium, wash with serum-free medium, and add [3H]prolme at 25 @i/mL for a further 24 h m serum-free culture medium supplemented with 50 pg/mL ascorbate and 50 pg/mL P-APN (or without P-APN to retam collagen in the pericellular matrix) Remove culture medium, and store at -20°C Wash cell layer with PBS, and solubthze by adding equal volumes of serum-free culture medium and 1M ammomum hydroxide (an ahquot may be analyzed for DNA by the diphenylamme method) 3 Collagen typmg. To analyze pepsin-reststant collagens, add pepsm/HAc solutton to equal volume of either labeled culture medium or solubillzed cell solution for 16 h at 4°C lyophibze, redissolve m 2X SDS sample buffer, and neutralize with 1-pL additions of 2M NaOH to titrate the color of the bromophenol blue m the sample buffer from yellow-green to blue (but not to vtolet). To analyze procollagens and fibronectm, add 2X SDS sample buffer contammg 0 2% P-ME to an equal volume of the culture medium Heat samples to boiling for 10 mm, and load on SDS gels (5% acrylamide running gels or 7-15% gradtent gels) that include a radiolabeled rat tail tendon collagen standard m one lane Perform delayed reduction with 0.1% P-ME on pepsmtzed samples to distmguish al(II1) from al(1 or II) collagens. Absence of the a2(I) collagen band indicates the virtual absence of type I collagen synthesis. In cultures contammg a mixture of type I and type II collagen, definmve identification of these collagens requires Western blottmg using specific antibodies and standard techniques 4 Proteoglycan synthesis (see Note 7) Add [35S]sodmm sulfate at 20 pCi/mL m culture medium contammg 25 pg/mL ASC, and Incubate at 37°C for a further 4-24 h Extract labeled medium and cell layer with equal volume of 8M guamdme hydrochloride solution containmg 20 mM EDTA and protemase mhibttors at 4°C for 48 h To quantttate 3SS-1abeledPGs, pass extracts over Sephadex G-25M m PD 10 columns, elute under dissociative conditions, and measure by scmtillation counting. Further characterization of proteoglycans may be performed by agarose/polyacrylamide composite gel electrophoresis, Western blottmg, or gelfiltration chromatography using standard published techniques. 5 Immunocytochemtstry (see Note 7) Plate cells m plastic Lab-Tech 4-chamber slides (Nunc, Inc Napervtlle, IL) at 6 x IO4 cells/chamber m culture medium

Human Chondrocyfe

Cultures

225

containing 10% FCS, and grow for 3-5 d to subconfluence. Change to culture medium containing 25 yg/mL ASC, and test reagents At the end of the mcubanon period, carefully wash the chambers three times with PBS, and fix the cells with 2% paraformaldehyde m 0 IA4 cacodylate buffer, pH 7 4, for 2 h at 4°C After two rinses with 0 IMcacodylate buffer, add monoclonal antibodies (MAbs) specific for human type II collagen, aggrecan core protein, and so forth to different chambers at concentrations recommended by the supplier Incubate separate chamber shdes with chondroitmase ABC for 30 mm at 37°C prior to addition of MAbs against chondroitm sulfates m order to expose epitopes Visualize the stammg by mcubation with a gold-conJugated secondary antibody (Auroprobe LM, Amersham, Arlington Heights, IL) followed by silver enhancement (e.g , IntenSE Kit, Amersham)

3.4. Analysis of mRNA Total RNA for Northern blots has been extracted successfully from human chondrocytes by several methods (35-3 7). We currently use a procedure modtfied from that of Cathala et al (38) (see Notes 8 and 9). 1 IO-cm Dishes Remove medium, and wash cell layer with PBS Trypsnuze cells m 0 5 mL of trypsm-EDTA (add 1 5 mL and then remove 1 mL immediately) at room temperature for I1 0 mm Suspend cells, and bring up suspension with 2 mL of DMEM/lO% FCS. Transfer to sterile polypropylene 10 x 75 mm tubes (15 mL) Wash plates with 2 mL of DMEM/FCS, and combme m tube with cell suspension Place tube on ice Remove ahquot for cell count, if required Spm down cells at 1200-1500 rpm, and wash with cold PBS Resuspend and combme pellets from 1-4 dishes (1-5 x IO6 cells) for each extraction 2 To final pellet add 500 uL of guamdme lysis buffer. Vortex Homogenize cell suspension by aspiration 5-l 0 times through tuberculm syrmge with needle Add 3 mL of 4M LiCl Vortex Store at 4°C overnight (can be left for a few days) 3 Spm at 9000 rpm m high-speed centrifuge for 90 min at 4°C Carefully remove supernatant with heat-treated Pasteur pipet (do immediately, standing at centrifuge) Respm if pellet slips. 4. To pellet add 500 uL of solubihzation buffer. Let stand at room temperature for

45 min, vortexmg every 10 mm until pellet IS dtssolved, or freeze on dry Ice (samples can be left at -80°C at this stage), and vortex while thawmg to break up pellet 5 Transfer sample to sterile 1 2-mL tubes with caps, Add 500 pL of phenol reagent (phenol.chloroform*isoamyl alcohol [25:24: 11). MIX by vortexmg and/or shakmg Leave on ice 5 mm Centrifuge at 12,000 rpm (m Eppendorf centrifuge) for 15 mm at 4°C Remove upper aqueous phase to fresh sterile 1.2-mL tube on ice, taking care not to take any of the interphase (Discard lower organic phase with interphase containing DNA and protems.)

6 To prectpttate the RNA, add l/l0 vol of 4M LICI, vortex, then add 2 5 vol of ethanol and vortex Leave overnight at-20°C Centrifuge for 30 mm at 12,000 rpm at 4°C Discard supernatant

226

Goldring

7 Wash pellet with 800 pL of 75% ethanol, and shake tube or vortex to break up pellet Centrifuge at 12,000 rpm for 15 mm at 4°C (Note Samples can be left m ethanol at -2O”C, If necessary ) 8 Speed vat final pellet, but not to dryness Dissolve pellets m 50 yL of dlethylpyrocarbonate (DEP)-treated H,O or 10 mMHEPES This takes time on ice Samples may be left overnight at 4°C and/or heated briefly at 65’C to speed up dlssolutlon 9 Read ODs at 240, 260, and 280 Use OD 260 to calculate RNA concentration The final preparations should give yields of approx 10 pg of RNA/l x IO6 cells with the appropriate A,,, A,,, ratio of approx 2 0 Store at -20°C in nonselfdefrostmg freezer or at -80°C 10 Separate RNAs (5-10 pg of total RNA/lane) on 0 8% agarose gels m the presence of 2% formaldehyde Transfer to mtrocellulose or nylon-supported mtrocellulose membranes in 10X SSC overnight 11 Label DNA plasmld probes with 32P-dCTP and 32P-dGTP by mck translation or cDNA mserts with 32P-dCTP by random primer labelmg Prehybrldlze blots m 50% formamide, 5X SSC, 5X Denhardt’s solution, and 0 3% SDS at hybrldlzatlon temperature Add 100 ng/mL DNA probe, and hybndlze for at least 16 h at 54°C for collagen probes and 42°C for noncollagen probes Stammg with ethldmm bromide and hybrldlzatlon with a glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA probe are employed to monitor umform loadmg of RNA on Northern blots A typical

3.5. Analysis by Transient

Northern

blot of human chondrocyte

mRNAs

1s shown

In Fig 2

of Gene Regulatory Sequences Transfection Assays

1 Plate chondrocytes at 1 x IO6 cells/lO-cm dish in culture medium contaming 10% FCS, and allow to settle down for 16-20 h Change the medium 2-4 h prior to the transfectlon 2 Add l&25 pg of plasmld DNA m a CaP04 copreclpltate m a dropwlse fashion throughout the dish, and agitate gently Incubate at 37°C for 4 h, and perform glycerol shock for 2 mm Wash cell layers with serum-free medium, and then allow the cells to recover overnight m culture medmm containing 10% FCS Change to medium contammg required serum concentration or serum substitute and test reagents, incubate for 3&48 h, and harvest for assay of reporter gene activity (see Note 10)

4. Notes I

Chondrocytes are quite resilient, and tolerate the prolonged mcubatlon times required for complete dlssoclatlon of the matrix and the absence of serum m the costal cartilage digestion If the digestion IS not complete by the end of the allotted time, then more collagenase solution may be added, or the suspension may be recovered and the fragments left behind for further digestion These condltlons result m suspensions that are essentially single-cell, and therefore, it 1s not necessary to resort to filtration through a nylon mesh, as has been done by others

227

Human Chondrocyte Cultures

Human

Costal

Chondrocyte

Aggrecan

mRNAs 9.5 kb

Type

Xi

collagen

7.5

kb

Type

II collagen

5.5

kb

Type

IX collagen

4.0

kb

Collagenase

2.2

kb

Stromelysln

1.9 kb

GAPDH

1.3 kb 1

2

3

4

Fig. 2. Expression of mRNAs encoding cartilage-specific matrix proteins and IL- linduced metalloproteinases by human costal chondrocytes. Costa1 chondrocytes at d 11 of primary culture were incubated without (lanes 1 and 2) or with 5 pA4 IL-ll3 (lanes 3 and 4) for 24 h prior to harvest of the cells for RNA extraction. Total RNAs (5 pg/lane) were electrophoresed on a 0.8% agarose gel in the presence of 2% formaldehyde, blotted on nylon-supported nitrocellulose membranes, and hybridized with the [32P]-labeled cDNA probes encoding large aggregating proteoglycan core protein (aggrecan), a2(XI) procollagen (type XI collagen), al(I1) procollagen (type II collagen), CL1(IX) procollagen (type IX collagen), collagenase, stromelysin, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The sizes of the mRNA transcripts in kilobases are indicated on the right. Descriptions and sources of the probes are found in ref. 34.

when shorter digestion times are used (I 7). These considerations are important for decreasing the loss of chondrocytes during their isolation from valuable human cartilage specimens. 2. After initial plating of the primary cultures, the chondrocytes require 2-3 d before they have settled down and spread out completely. Culture for -4-7 d is required before reasonable amounts of total RNA may be extracted. Juvenile costal chondrocytes continue to express chondrocyte phenotype (e.g., type II collagen mRNA) for several weeks and will form multilayer cultures. Adult articular chondrocytes are strongly contact-inhibited, and they may lose phenotype within 2-4 wk of monolayer culture. After they are subcultured, both types of chondrocytes stop expressing chondrocyte matrix proteins, but this loss of phenotype is reversible and the cells may be redifferentiated in suspension culture within or on top of a nonadherent matrix.

228

Goldrmg Smce chondrocytes are strongly adherent to tissue-culture plastic, possibly because of the strongly charged glycosaminoglycans m their matrix, a trypsnEDTA solution rather than trypsm alone should be used for full recovery of chondrocytes from tissue-culture plastic during passaging It IS preferable not to use any antibiotics in order that any contamination that arises becomes apparent rmmedlately If necessary, standard concentrations of pemclllm-streptomycm, gentamycm, and so on, that are suggested for fibroblast cultures are acceptable for use in chondrocyte cultures Primary chondrocyte cultures should be used for experlmental analyses nnmedlately before or Just after confluence 1sreached to permit optimal matrix synthesis and cellular responsiveness. If the cells are not used or subcultured, they may be left at confluence for several weeks wtth weekly medmm changes as long as the volume of the culture medmm is maintained If long-term culture results m the deposition of excessive matrix that IS not easily dlgested with trypsm-EDTA, then a single-cell suspension may be obtained by using a ddute solution of collagenase (0 25%) and trypsm (0 25%) m PBS Although the growth and mamtenance of chondrocytes m primary culture or after subculture reqmre the use of 10% FCS. the loss of phenotype that occurs under these conditions may be delayed if the cells are plated at 4-10 times higher density Smce high cell yields are not usually attamable from human cartilage sources, the reversibility of the loss of phenotype may be exploited by expanding the chondrocyte populations m monolayer cultures, redifferentlatmg the cells in flmd suspension culture, and replatmg them m monolayer munedlately before performing the experimental procedure Biosynthetic labeling and lmmunocytochemlstry procedures are readily performed on chondrocytes m a sohd suspenston system, such as algmate, agarose, or collagen gels Algmate culture may be the method of choice, smce the chondrocytes are easily recovered by depolymerlzatlon of the algmate with a calcium chelator For long-term algmate cultures, high-vlscoslty algmate may provide more stable beads Concentrations of serum as low as 0 5%, serum substitutes, or combmatlons of growth and differentiation factors or hormones have been used successfully, depending on the experlmental protocol, to permit chondrocyte phenotyplc expression Optlmrzation of extracellular matrix synthesis and depositIon* ASC IS not routmely added during growth and maintenance culture of chondrocytes, smce It 1s known to mhlblt the transcription of cartilage-specific matrix genes m long-term mcubatlons (37) It 1sa requlrement, however, for synthesis and secretlon ofcollagen and should be added at 25-50 pg/mL during I- to 3-d mcubatlons as reqmred for btosynthetlc labelmg of collagens or proteoglycans, or extracellular deposition of matrix proteins that will be extracted for lmmunoblottmg or analyzed by immunocytochemistry General recommendations for RNA extractions: Use sterile plasttc tubes and plpets and/or heat-treated glassware (not to be touched by human hands) Wear .. gloves at all times (even when touchmg outside of tubes). Never use parafilml All procedures are

Human Chondrocyte Cultures

229

done on ice (4°C) unless Indicated otherwise All solutions (except phenol solution) and dH,O are DEP-treated and autoclaved and/or sterile-filtered 9 Optimization of RNA extraction technique. Use of trypsmization of cell cultures prior to extractton and of a lithium chloride precipnatton step during extraction will reduce or ehmmate contammatton with proteoglycans and other glycoprotems The amount of total RNA loaded per well on agarose gels should not exceed 10 ug, and sufficiently long gels should be run to resolve collagen and large proteoglycan mRNAs that migrate more slowly than 28s RNA on Northern blots and tend to smear if overloaded. Northern blots on nitrocellulose or nylon-supported mtrocellulose and hybrtdtzations m the absence of dextran sulfate result m vtrtually no background m the detection of high ktlobase transcrtpts m chondrocyte RNA preparations High stringency hybridization conditions are used to prevent crossreacttvittes among some collagen probes 10. Optimtzation of transient expression assays m normal primary and subcultured chondrocytes Chondrocytes are plated at somewhat higher densities than nnmortahzed cell lines, such as NIH/3T3 cells routinely used for transient transfecttons Subcultured chondrocytes are transfected the day after passagmg In contrast, primary cultures must be left several (4-7) days, with at least one interim medium change, until the cells have settled down and begun to undergo cell divtsion Recovery overnight after the transfection by mcubation m culture medium also permits optimal expression of the reporter gene and responsiveness to test reagents Primary or subcultured costal chondrocytes incubated m an msulmcontammg serum substitute during transient expression of type II collagen gene regulatory sequences have been used successfully (35)

References 1 Heinegard, D and Oldberg, A (1989) Structure and biology of cartilage and bone matrix noncollagenous macromolecules FASEB J 3,2042-205 1 2 Mayne, R. and Brewton, R. G (1993) Extracellular matrix of cartilage. collagen, m Joint Carthage Degradation Basic and Clwucal Aspects (Woessner, .I F , Jr and Howell, D S., eds ), Marcel Dekker, New York, pp 81-108 3 Goldrmg, M. B (1993) Degradation of arttcular carttlage in culture regulatory factors, m Jomt Cartilage Degradatton: Basic and Clrnlcal Aspects (Woessner, J F., Jr. and Howell, D S., eds ), Marcel Dekker, New York, pp. 28 I-345 4. Goldrmg, M. B , Sandell, L J , Stephenson, M L., and Krane, S. M (1986) Immune interferon suppresses levels of procollagen mRNA and type II collagen synthesis m cultured human arttcular and costal chondrocytes J. Bzol Chem 261,9049-9056 5. Goldring, M B , Btrkhead, J , Sandell, L. J , Ktmura, T , and Krane, S. M (1988) Interleukm 1 suppresses expression of cartilage-specific types II and IX collagens and increases types I and III collagens m human chondrocytes. J Clrn Invest 82, 2026-2037

6 Goldrmg, M B and Krane, S M (1987) Modulatton by recombmant interleukm 1 of synthesis of types I and III collagens and associated procollagen mRNA levels m cultured human cells J Blol Chem 262, 16,72&16,729

230

Goldring

7 Holtzer, J , Abbott, J , Lash, J , and Holtzer, A (1960) The loss of phenotyplc traits by dlfferentlated cells m vitro I Dedlfferentlatlon of cartilage cells Proc Nat1 Acad Scl USA 46, 1533-1542 8 Ham, R G and Sattler, G L (1968) Clonal growth of dlfferentlated rabbit cartllage cells J Cell Phys~ol 72, 109-l 14 9 Green, W T , Jr (1971) Behavior of artlcular chondrocytes m cell culture C/zn Orthopaed Related Res 75,248-260 10 Norby, D P , Malemud, C J , and Sokoloff, L (1977) Differences m the collagen types synthesized by lapme artlcular chondrocytes m spmner and monolayer culture Arthritis Rheum 20, 709-7 16 11 Glowackl, J , Trepman, E , and Folkman, J (1983) Cell shape and phenotyplc expression m chondrocytes Proc Sot Exp. Blol Med 172, 93-98 12 Castagnola, P , Moro, G , Descalzl-Cancedda, F . and Cancedda, R (1986) Type X collagen synthesis during m vitro development of chick embryo tibia1 chondrocytes J Cell &ol 102,23 IO-23 17 13 Kuettner, K E , Pauh, B U , Gall, G , Memoll, V A , and Schenk, R K (1982) Synthesis of cartilage matrix by mammalian chondrocytes m vitro I Isolation, culture charactenstlcs, and morphology J Cell Blol 93, 743-750 14 Bassleer, C., Gysen, P , Foldart, J M , Bassleer, R., and Franchlmont, P (1986) Human chondrocytes m trldlmenslonal culture In Vitro Cell Dev Blol 22, 113-l 19 15 Gibson, G J , Schor, S L , and Grant, M E (1982) Effects of matrix macromolecules on chondrocyte gene expression synthesis of a low molecular weight collagen species by cells cultured wlthm collagen gels J Cell Bzol 93, 767-774 16 Benya, P D and Shaffer, J D (1982) Dedlfferentlated chondrocytes reexpress the differentiated collagen phenotype when cultured m agarose gels Cell 30, 2 15-224 17 A&house, A L , Beck, M , Fnffey, E , Sanford, J , Arden, K , Machado, M A , and Horton, W A (1989) Expression of the human chondrocyte phenotype m vitro. In Vitro Cell Dev Blol 25,659-668. I8 Aydelotte, M B and Kuettner, K E (1988) Differences between sub-populations of cultured bovine artlcular chondrocytes I. Morphology and cartilage matrix production Connect TLSSRes 18, 205-222 19 Guo, J , Jourdlan, G W , and MacCallum, D K (1989) Culture and growth characteristics of chondrocytes encapsulatedm algmate beads Conn TLW Res 19, 277-297 20 Hauselmann,H J., Aydelotte, M B , Schumacher,B L , Kuettner, K E , Gltehs, S H , and Thonar, E J -M A (1992) Synthesis and turnover of proteoglycans by human and bovine adult artlcular chondrocytes cultured m algmate beads Matrix 12,11&129 21 Vuklcevlc, S , Klemman, H K., Luyten, F P., Roberts, A B , Roche, N S , and Reddl, A H (1992) Identification of multiple active growth factors in basement membraneMatrlgel suggestscaution m interpretation of cellular actlvlty related to extracellular matrix components Exp Cell Res 202, l-8

Human Chondrocyte Cultures

231

22 Adolphe, M., Froger, B , Ronot, X , Corvol, M. T , and Forest, N (1984) Cell multiphcatton and type II collagen productton by rabbit articular chondrocytes cultivated m a defined medium Exp Cell Res 155, 527-536 23 Gerstenfeld, L C , Kelly, C M , Von Deck, M , and Lian, J B (1990) Comparative morphological and biochemtcal analysis of hypertrophic, non-hypertrophtc and 1,25(OH),D, treated non-hypertrophic chondrocytes Connect Tlss Res 24,2%39 24 Adams, S L., Pallante, K M , Nm, Z , Leboy, P S , Golden, E B , and Pacifici, M (1991) Rapid mduction of type X collagen gene expression m cultured chick vertebral chondrocytes Exp Cell Res 193, 19&197 25 Alema, S , Tato, F , and Boetttger, D ( 1985) Myc and src oncogenes have complementary effects on cell proliferation and expression of specific extracellular matrix components m definitive chondroblasts Mol Cell BIOI 5, 538-544 26 Gionti, E , Pontarelh, G , and Cancedda. R (1985) Avtan myelocytomatosis virus immortalizes differentiated quail chondrocytes Proc Nat1 Acad Scl USA 82, 27562760 27 Horton, W E , Jr, Cleveland, J , Rapp, U . Nemuth, G , Bolandet, M , Doege, K , Yamada, Y , and Hassell, J R (1988) An established rat cell lme expressing chondrocyte properties Exp Cell Res 178,457-468 28 Thenet, S , Benya, P D , Demtgnot, S , Feunteun, J , and Adolphe, M (I 992) SV40-mnnortaltzatton of rabbit arttcular chondrocytes alteration of dtfferentiated functions J Cell Physzol 150, 158-l 67 29 Mallem-Germ, F and Olsen, B R (1993) Expression of simian virus 40 large T (tumor) oncogene m chondrocytes induces cell proliferation without loss of the differentiated phenotype Proc Nat1 Acad Scl USA 90, 3289-3293 30 Grigoriadis, A E , Heersche, J N M , and Aubm, J E (1988) Differentiation of muscle, fat, cartilage, and bone from progenitor cells present m a bone-derived clonal cell populatton effect of dexamethasone J Cell Bzol 106, 2 139-2 I5 1 31 Bermer, S M and Goltzman, D (1993) Regulation of expression of the chondrogemc phenotype m a skeletal cell lme (CFK2) m vitro J Bone Miner Res 8, 475484 32. Block, J A , Inerot, S E , Gttehs, S , and Kimura, J H (1991) Synthesis of chondrocytic keratan sulphate-containing proteoglycans by human chondrosarcoma cells m long-term cell culture J Bone Joznt Surg Am 73, 647-658 33 Takigawa, M , Pan, H 0, Kmoshita, A., TaJima, K , and Takano, Y (1991) Establishment from a human chondrosarcoma of a new immortal cell line with high tumortgemcity m VIVO, which 1s able to form proteoglycan-rich cartilagelike nodules and to respond to msulm m vitro Int J Cancer 48, 7 17-725 34 Goldrmg, M B , Birkhead, J R , Suen, L -F , Yamin, R , Mizuno, S , Glowacki, J , Arbiser, J L , and Apperley, J F (1994) Interleukm- 1P-modulated gene expression in immortaltzed human chondrocytes J Cfln fnvest 94,2307-23 16 35 Goldrmg, M B , Fukuo, K , Btrkhead, J R , Dudek, E , and Sandell, L J (1994) Transcriptional suppression by mterleukm-1 and interferon-y of type II collagen gene expression m human chondrocytes J Cell Blochem 54, 85-99

232

Goldring

36 Chomczynskl, P and Sacchl, N (1987) Smgle step method of RNA lsolatlon by acid guamdmmm thlocyanate-phenol-chloroform extractlon Anal Btochem 162, 156-159 37 Sandell, L J and Daniel, J C (1988) Effects of ascorbic acid on collagen mRNA levels m short term chondrocyte cultures Connect Tzss Res 17, 11-22. 38 Cathala, G , Savouret, J -F , Mendez, B., West, B L , Karm, M., Martial, J A , and Baxter, J D (1983) A method for lsolatlon of Intact, translatlonally actwe rlbonuclelc acid DNA 2, 32!L335

Isolation and Culture of Bone-Forming (Osteoblasts) from Human Bone James A. Gallagher,

Cells

Roger Gundle, and Jon N. Beresford

1. Introduction

The most conspicuous function of the osteoblast is the formatton of bone. Durmg phases of active bone formation, osteoblastssynthesizebone matrix and prime it for subsequent mmeralization. Active osteoblasts are plump, cuboidal cells rich m organelles involved m the synthesis and secretion of matrix proteins. Unlike fibroblasts, they are obviously polarized, secretmg matrix onto the underlying bony substratum which consequently grows by apposition. Some osteoblasts are engulfed m matrix during bone formation and are entombed m lacunae. These cells are described as osteocytes and remam m the bone matrix in a state of low metabolic activity. At the completion of a phase of bone formation, those osteoblasts which avoided entombment m lacunae lose their prominent synthetic function and become inactive osteoblasts, otherwise known as bone-lmmg cells. In mature bone, linmg cells cover most of the bone surfaces. Osteocytes and bone-lining cells should not be considered as mactive cells since they play a major role m the regulation of bone modeling and remodeling and m calcium homeostasis (I). Much of our knowledge of the biology of bone tissue has been derived from morphological studies. The heterogeneity of cell types in bone, the highly cross-linked extracellular matrix and the mineral phase, combine to make bone a difficult tissue to study at the cellular and molecular level (2). Consequently the earliest attempts to isolate specific cell populations utilized enzymic digestion of fetal or neonatal tissue from experimental animals which is poorly mmeralized and highly cellular. Although these studies undoubtedly furthered our knowledge of bone cell biology, there are major advantages m attempting to investigate cells isolated from human bone. First, there are differences m cell From

Methods Edlied

tn Molecular by

G E Jones

Medrone Humana

233

Human Press

Cell Culture Inc , Totowa,

Protocols NJ

Gallagher,

Gundle,

and Beresford

phystology between species, and also between adults and neonates within a species Second, the ability to culture human bone cells opens up the prospect of mvesttgatmg the pathologtcal mechanisms that underlie bone diseases mcludmg age-related bone loss. The earliest report to describe the tsolatton of viable cells from adult human bone IS that of Bard et al. (3) The tsolatton procedure they employed involved extensive demmerahzatton of the tissue m a solutton of ethylenedtammetetraacettc acid (EDTA) followed by digestion with collagenase. The cells obtamed expressed low levels of alkaline phosphatase and mcorporated radtolabeled prolme mto macromolecular material, but the hydroxyprolme/prolme ratto mdtcated that collagen formed only a minor proportion of the total protem synthesized. The cultured cells remained viable for periods of up to 2 wk, but failed to proliferate leading the authors to conclude that osteocytes were the predominant cell type present (4). In a study of patients with Paget’s disease, Mills et al. succeeded m culturmg cell populattons from explants of trabecular bone (5) These populattons responded to crude extracts of the parathyrotd gland with an increase m radtolabeled thymtdme mcorporatton and a proportion of the cells expressed alkalme phosphatase actrvrty Encouraged by these reports, m the early 1980s we (6-15) and several other groups of mvesttgators (16-22) developed systems to isolate human bonederived cells (HBDCs) We undertook a systematic mvesttgatton to identtfy the phenotyptc charactertsttcs of HBDC populattons and to determine the condtttons of culture that favored their prohferatton and dtfferenttatton. Wtthm a relatively short space of time tt was shown that the cell populattons obtained reproducibly expressed an osteoblast-like phenotype and that they represented a viable alternative to the use of normal or neoplasttc cell lines of avtan or rodent ortgm HBDCs have been widely used to investigate the biology of the human osteoblast Then use has facthtated several major developments m our understanding of the hormonal regulation of human bone remodeling, mcludmg the first demonstration of an effect of cytokmes on bone-forming cells (IO), tdenttficatton of direct effects of oestrogen on osteoblasts (14,23), and recently the tdenttficatton of purmergtc receptors m bone (24,2.5). We are now moving mto a new phase of research m which HBDCs, isolated from patients with specific disorders including Paget’s disease (26), McCune Albrtght syndrome (27), and osteoporosts (28-30) are being used to investigate the cellular and molecular pathology of bone disease. Cultured chondrocytes have been used to repair cartilage defects with some success (31) Recently we have started to identify condtttons whrch promote the osteogemc potential of HBDCs m vitro Now we have the prospect that

235

Culture of Osteoblasts

human bone cell culture may become an important tool m tissue engineering, allowing the autologous transplantation of osteoblastlc populations expanded m vitro and seeded onto suitable carrier matrices. The purpose of this chapter IS threefold: 1 To describe m detail the methodology currently m use m the authors’ laboratories for the isolation and culture of HBDCs, 2 To demonstrate that, by employmg the methods described, it IS possible to obtain a cell population that 1s phenotyplcally stable and that retams the potential for osteogenic dlfferentlatlon m vitro and in VIVO, and 3 To promote the wider use and continued development of the HBDC culture system

2. Materials 2.1. Tissue-Culture

Media and Supplements

1 Phosphate-buffered salme (PBS) without calcium and magnesium, pH 7 4 (Glbco, Galthersburg, MD) 2 Dulbecco’s modlflcatlon of mmlmum essential medium (DMEM) (Glbco) supplemented

to a final

concentration

of 10% with

heat-mactlvated

fetal calf

serum(FCS), 2 mA4~-glutamme,25-50 U/mL pemclllm, 25-50 pg/mL streptomycm, and 50 yg/mL freshly prepared L-ascorbic acid 3 Serum-free DMEM (SFM) 4 FCS (see Note 1) 5 Tissue-cultureflasks (75 cm2)or Petri dishes(loo-mm diameter)(seeNote 2)

2.2. Preparation

of Explants

1 Bone rongeurs and/or bone curet from any surgical Instrument supplrer 2 Solid stainlesssteel scalpelswith Integral handles(BDH Merck)

2.3. Passaging

and Secondary

Culture

1 Trypsn-EDTA solution* 0 05% trypsin and 0 02% EDTA m Ca2+- and Mg2+free PBS, pH 7 4 (Glbco) 2 0 4% Trypan blue m 0 85% NaCl (Sigma Aldrich) 3 70 pm “Cell Strainer” (Becton Dlckmson) 4. Neubauer Hemocytometer (BDH Merck). 5 Collagenase(Sigma type VII from Clostndmm hlstolytlcum) 6. DNAse I (Sigma Aldrich).

2.4. Phenotypic

Characterization

1 Calcltrlol (Leo Pharmaceuticals) 2 Menadione (vitamin K3) (Sigma Aldrich)

3 Stamlng Kit 86-R for alkaline phosphatase (Sigma Aldrich) 4

Osteocalcm radlolmmunoassay (see Note 3)

(RIA)

(Cts UK Ltd , Hugh Wycombe

Bucks, UK)

236

Gallagher, Gundle, and Beresford

2.5. In Vitro Mineralization 1 2 3 4

L-Ascorbic acid 2-phosphate (Wako Pure Chemtcal Industries Dexamethasone (Sigma Aldrtch) Hematoxylm (BDH Merck) DPX (BDH Merck)

Ltd )

3. Methods 3.1. Bone Explant Culture System 3.1 7. Establishment of Primary Explant Cultures A scheme outlmtng

the culture technique

is shown

m Fig

1.

1 Transfer tissue, removed at surgery or biopsy, mto a sterile contamer with PBS or serum-free medium (SFM) for transport to the laboratory with mmtmal delay, preferably on the same day (see Note 4) Ideally, the bone used should be radtologrcally normal An excellent source is the upper femur of patients undergoing total hip replacement surgery for osteoarthritis. Cancellous bone IS removed from this site prior to the msertton of the femoral prosthesis and would otherwise be discarded The tissue obtained is remote from the hip Joint itself, and thus from the site of pathology, and IS free of contammatmg soft tissue (see Note 5). 2 Remove extraneous soft connecttve tissue from the outer surfaces of the bone by scrapmg with a sterile scalpel blade. Rinse the tissue m sterile PBS and transfer to a sterile Petri dish containing a small volume of PBS (5-20 mL, depending on the size of the specimen) If the bone sample is a femoral head, remove cancellous bone directly from the open end using a bone curet or a solid stainless steel blade with integral handle Disposable scalpel blades may shatter during this process With some bone samples (e.g., rib), it may be necessary to gam access to the cancellous bone by breaking through the cortex with the aid of sterile surgical bone rongeurs 3 Transfer the cancellous bone fragments to a clean Petri dish containing 2-3 mL of PBS and dtce mto pteces 3-5 mm m diameter. This can be achieved m two stages using a scalpel blade first, and then fine scissors Decant the PBS and transfer the bone chips to a 50-mL polypropylene tube containing 15-20 mL of PBS. Vortex the tube vigorously three times for 10 s and then leave to stand for 30 s to allow the bone fragments to settle Carefully decant off the supernatant containing hematopotettc ttssue and dtslodged cells, add an additional 15-20 mL of PBS, and vortex the bone fragments as before Repeat this process a mimmum of three times, or until no remaining hematopotettc marrow is visible and the bone fragments have assumed a white, ivory-like appearance 4 Culture the washed bone fragments as explants at a density of 0 2-O 6 g of tissue/ loo-mm diameter Petri dash or 75-cm2 flask (see Note 2) m 10 mL of medium at 37°C in an humidified atmosphere of 95% an-, 5% CO2 5 Leave the cultures undisturbed for 7 d after which time replace the medium wtth an equal volume of fresh medium taking care not to dtslodge the explants. Feed again at 14 d and twice weekly thereafter

237

Culture of Osteoblasts

dice bone q.0 Pm. 49 .i*.’

&. .. .

l

/

remove cancellous bone with curette

plate out explants

7-10da scell outgrowt i s appear I 4-6weeks confluent cultures

,---

remove explants, ‘^

Pl ric--.~.

.”

--~

._-,

:-

-3 __~..

__

.-.‘.-

__._ ._._ 4 .I_ ._

4 - 6 weeks replated explants give rise to confluent cultures

E2 Fig. 1. Schematicdiagram outlining the techniqueusedto isolate culturesof osteoblastsfrom bone.

With the exception of small numbers of isolated cells, which probably become detached from the bone surface during the dissection, the first evidence of cellular proliferation is observed on the surface of the explants, and this normally occurs within 5-7 d of plating. After 7-10 d, cells can be observed migrating from the explants onto the surface of the culture dish (see Fig. 2). If care is taken not to dislodge the explants when feeding, and they are left undisturbed between media changes, they rapidly become anchored to the sub-

238

Fig. 2. Photomicrograph osteoblasts.

Gallagher, Gundle, and Beresford

of explanted cancellous bone showing migration

of

stratum by the cellular outgrowths. Typical morphology of the cells is shown in Fig. 3, but cell shape varies between donors from fibroblastic to cobblestone-like. Cultures generally attain confluence 4-6 wk postplating, and typically achieve a saturation density of 29,000 * 9000 cells/cm* (mean + SD, n = 11 donors). 3.1.2. Passaging and Secondary Culture 1. Remove and discard the spent medium. 2. Gently wash the cell layers three times with 10 mL of PBS. 3. To each flask add 5 mL of freshly thawed trypsin-EDTA solution at room temperature (2O“C). Incubate for 2 min at room temperature with gentle rocking every 30 s to ensure that the entire surface area of the flask and explants is exposed to the trypsin-EDTA solution. Remove and discard all but 2 mL of the trypsinEDTA solution, and then incubate the cells for an additional 5 min at 37°C. 4. Remove the flasks from the incubator and examine under the microscope. Look for the presence of rounded, highly refractile cell bodies floating in the trypsinEDTA solution. If none, or only a few, are visible, tap the base of the flask sharply

Culture of Osteoblasts

Fig. 3. Photomicrograph

5.

6.

7.

8.

9.

showing the typical morphology

of HBDCs.

on the bench top in an effort to dislodge the cells. If this is without effect, incubate the cells for a further 5 min at 37OC. When the bulk of the cells has become detached from the culture substratum, transfer the cells to a 50-mL polypropylene tube containing 5 mL of DMEM with 10% FCS to inhibit tryptic activity. Wash the flask two to three times with 10 mL of SFM and pool the washings with the original cell isolate. To recover the cells centrifuge at 250g for 10 min at 15OC. Remove and discard the supematant, invert the tube, and allow to drain briefly. Holding the top of the tube, sharply flick the base of the tube with the first finger to dislodge and break up the pellet. Add 2 mL of SFM containing 1 pg/mL DNAse I for each dish or flask treated with trypsin-EDTA, and using a narrow bore 2-mL pipet, repeatedly aspirate and expel the medium to generate a cell suspension. Filter the cell suspension through a 70-pm “Cell Strainer” (Becton Dickinson) to remove any bone spicules or remaining cell aggregates. For convenience and ease of handling, the filters have been designed to fit into the neck of a 50-mL polypropylene tube. Wash the filter with 2-3 mL of SFM containing DNAse I, and add the filtrate to the cells. Take 20 pL of the mixed cell suspension and dilute to 80 p.L with SFM. Add 20 p.L of trypan blue solution, mix, and leave for 1 min before counting viable (round and refractile) and nonviable (blue) cells in a Neubauer Hemocytometer. Using this procedure, typically l-l .5 x IO6 cells are harvested per 75-cm2 flask of which 275% are viable. Plate the harvested cells at a cell density suitable for the intended analysis. We routinely subculture at 5 x 103-104 cells/cm* and achieve plating efficiencies measured after 24 h of 270% (see Note 6).

Gallagher,

240 ALKALINE

PHOSPHATASE

Gundle,

and Beresford

ACTIVITY

200 175 150 125 100

i/// 0.01

0.1

1

Log [CALCITRIOL]

10

100 (n&l)

SKIN

BONE

CELL TYPE Fig. 4. The expression of alkalme phosphatase by cultures of HBDCs and skmderived cells at first passage. Confluent cell monolayers were lysed m alkaline buffer (Sigma Aldrich) contammg 0 1% Trtton X-100, chilled on ice, and somcated for 10 s three times usmg an MSE Model 150 Somcator with a Micro-tip at 20% of full output power The homogenates were centrifuged to remove msoluble material and an ahquot of the supernatant used for the determmation of alkalme phosphatase activity as described previously (6) Open bar, skm fibroblasts (n = 6 donors); closed bar, bonedertved cells (n = 18 donors). The values shown are the mean + SD Inset Stimulation of HBDC alkahne phosphatase activrty by calcitnol The values shown are the mean f SD of data from three donors. Over the same dose range there was no effect on the alkaline phosphatase acttvtty of skin fibroblasts derived from the same donors (data not shown). If dishes have reached confluence but the cells are not required ately, the cells can be stored by cryopreservatton (see Note 7).

3.1 3. Phenotypic

Characteristics

tmmedt-

at Fvst Passage

Compared with cultures of skm fibroblasts obtained from the same donor, cultures of HBDCs expresshigh basal levels of the enzyme alkaline phosphatase, a widely accepted marker of early osteogenic dtfferenttatron (Fig. 4). Basal actrvrty IS mmally low, but increases with increasmg cell density (10) (see Note 8). Treatment with calcttrtol(l,25[OH],D3; the active metabolite of vita-

Culture of Osteoblasts

241 GROWTH

0

5

CHARACTERISTICS

10

TIME

15

20

25

(DAYS)

Fig. 5. Representative growth curves from HBDCs and skin-derived cells at first passage Cells were subcultured mto multiwell trays at a density of 5 x 10’ cells/cm* and cultured for the indicated time pertods in medium supplemented with 10% (v/v) heatinactivated FCS The media were changed every third day. Open symbols, skm fibroblasts, closed symbols, HBDCs Tissues were obtained from a 10 5-yr-old male donor undergoing corrective surgery. Inset: Mean + SD doubling times at first passage for HBDCs and skin-derived cells (n = 4 donors) cultured under identical conditions Closed bars, HBDCs (n = 6 donors), open bars, skin fibroblasts (n = 4 donors). min D3) increases HBDC

alkaline

phosphatase

activity,

but not that of skin

fibroblasts (see Ftg. 4 inset). The magnitude of this sttmulatory effect decreases as cell density and, hence, basal alkaline phosphatase activity increases (ZO). When plated at similar densities and cultured under identical condrtions, compared with skin fibroblasts obtained from the same donor, HBDCs proliferate less rapidly and reach lower saturation densities (Fig. 5)

In common with cells of the osteogenic lineage from all species studied, HBDC respond to PTH with an increase m the levels of mtracellular CAMP (Fig. 6) (I&22). As shown, however, a simtlar response can be observed m cultures of human skm fibroblasts (Fig. 6). This IS m agreement with a report that human dermal fibroblasts possess PTH receptors comparable to

Gallagher, Gundle, and Beresford

242 RESPONSE

TO PTH

5-

T 0 [bPTH]

T

2 (U/d)

0.08

[bPTH]

0.4

10

(Units/ml)

Fig. 6. The effect of bPTH on the productton of 3’, 5’-cychc adenosme monophosphate by HBDCs and skin-derived cells at first passage Confluent cell monolayers were incubated for 10 mm at 37°C m medmm supplemented with 1% (v/v) charcoalstripped FCS and 500 p~%4rsobutyl-I-methyl-xanthme m the absence or presence of bPTH (0.08-10 U/mL, National Institute of Biologtcal Standards and Control reagent #77/533,230 U/100 c(g protem) (see Note 13) Total CAMP (medium + cell layer) was measured by specific RIA Cultures were established from tissues obtained from the same donor descrtbed m Fig 1 Open bars, skm fibroblasts, closed bars, HBDCs Inset Mean k SD index of stimulatton (treated/control ratto) for cultures of HBDCs treated with 2 U/mL bPTH (n = 6 donors)

those present on cells derived from bone and kidney tissue (32). Over the same dose range, m addttion to increasing the level of mtracellular CAMP,

PTH mhtbtts basal- and calcrtrtol-stimulated alkalme phosphatase acttvtty (Table 1). Over a 24-h period, takmg mto account the relative abundance of prolme m collagen compared to noncollagenous protem, collagen accounts for 19 6 f 1 0% of the protem secreted mto the culture medium (mean 3~SEM, n = 16 donors) As wrth basal alkalme phosphatase acttvtty, the percentage of collagen syntheSISmcreases wtth Increasing cell density (12). Of the total amount of collagen

Culture of 0s teoblas ts

243

Table 1 The Effect of PTH on the Basaland Calcitriol-Stimulated Alkaline

Phosphatase

Activity

of HBDC

Alkalme phosphatase, mU/mg protem Experiment B” bPTH, U/mL 00

0 04 02 10 50

Experiment Aa 252 f 14 193 k 09’ 15 1 + 07‘

138 f 13‘

Basal 639 * 48 f 43 f 40 k

+ 50 nM Calcitrtol 30

3 4c 2 4c 0 6’

987 k 20’ 76 f 1 6” 48

k 3 O’/

“Experiment A Confluent cells (EIPl) were Incubated for 72 h m medmm supplemented with 5% (v/k) FCS m the absence or presence of bovme PTH (bPTH) (see Note 13) at the mdlcated concentrations Cells were obtained from the femur of a 30-yr-old male undelgomg surge1 y followmg an mdustrlal accident “Experiment B Confluent cells (EIPl) were incubated for 48 h In medium supplemented with 10% (v/v) FCS m absence or presence ofbPTH, calcttnol, or both hormones m combmatlon at the Indicated concentrations Cells were obtained from the tlbla of a 6-yr-old female undergomg corrective surgery for non-union of a previous fracture Alkalme phosphatase actlvlty m the solublhzed cell layers was measured as described previously (8) ‘Slgmficantly different (p < 0 01) when compared with the control “Slgmficantly different (p < 0 01) when compared with calcltrlol alone

synthesized,