The Growth and Structure of Human Oral Keratinocytes in Culture

Dorthe Arenholt-Bindslev, D .D .S., Ph .D ., Arne Jepsen , D .D .S., Donald K. MacCallum, Ph .D., and John H. Lillie, D.D.S ., Ph .D . Hoya l DC ll ta l College (D A-B. AJ), Aarhus, Denmark , and Departmcnt of An atomy alld Cell Biology, University of Michiga n Medi ca l School (D KM , JI-IL), Ann Arbor, Michiga n, U.S.A .

Hum an keratinocytes derived from explants o f cheek (bu c­ca l) mu cosa g row vigoro usly in cu lture and can be sub­cultivated twice. The structure o f the ora l keratinocytes in vitro is the sa m e in prim ary cultures and subcultures . The cell s strati fy, are characterized by well-developed tono­fibrill ar-des moso m al co mplexes, and rarely exhibit signs of terminal diffe renti atio n. Unique fea tures o f the culture sys tem that favo r keratinocyte g rowth are: (1) incubati on

S ever; 1 different techniqu es have been used to es t3blish cultures of hum an ker3tin ocytes in the absence o f th e connecti ve ti ss ue elements th at co mprise either th e un­derl yin g dermi s o r lamin a propria. M ost recentl y, se ri al pro p3ga tion of hum3n ker3tinocy tes has been acco m­

plished by several labo rato ries thro ug h the usc of mito ticall y in­hibited murine 3T3 cells th at act as a "feeder" la yer 11 ,2 1. Ho w­ever, the presence of m etabolizin g, th o ugh mito ti ca ll y inhibited , feeder ce lls can be a confo undin g facto r in keratin ocy te cultures used to stud y in vitro m alignant tran sfo rm atio n, po tential cy­to tox ic agents, o r o th er similar problems. In o rd er ·to avoid the use of feeder cell s, va rio us o ther kera tin ocyte culture meth ods have 'been develo ped th at require: (1) coa tin g o f culture surfaces w ith m o lecules found in the extracellul ar m atri x (vari ous coll a­gens, fi bronectin , lam in in) [3-7], (2) va ryin g th e concentratio n of ca lcium in the culture m edium [8], (3) the usc o f conditio ned medium o r co mp lex bio log ic extrac ts 19], (4) th e addition o f va rio us mi togens and /or trace cl em ents [1 0, 111 , o r (5) a co mbi­natio n of2 o r 111 0 re of the above fea tures 11 2, 131. The aim o f this stud y was to devel o p a simple method fo r g rowin g and seri all y pro pagatin g human oral keratin ocy tes. T hi s impo rtant linin g mu­cosa has rarel y been studied in vitro [1 4, 151 even th oug h it ex hibits an exceptio nall y w ide range o fk eratin ocy te di fferentiJti on in situ and is subj ect to dramat ic changes in g rowth rate and phenotype (non keratinizi ng to kerJ tini zin g) in prem Jli gnant Jnd m J lignant conditi o ns.

Manuscript rcceivedJul y 28,1986; accepted fo r publica tion Septcmbcr 9, 1986.

T his stud y was supported by grants from the Danish Cancer Society and the Nat ional Institute of Denta l Resea rch (NIH).

Hepri nt requests to : Dorthe Arcnholt-Bindslev, D.D.S. , Hoya l Delltal College. Vennclyst Bouleva rd, O K 8000 Aarhus C, Denmark , or Donald K. MacCa llum , Ph .D., Department of Anatomy. Medical Science II, Uni versity of Michigan, Ann Arbo r. Michigan 48 109.

Abbrev iations: DMSO: dimeth yl sulfox ide MEM: minimum essentia l medium

at 34°C, (2) inclusio n o f 0.5 % dim eth yl sulfoxide in the culture medium, and (3) ini tiatin g subcultures as 5.0 Inn1

col onies containing 100, 000/20 ILl o f medium . O ne pri­m ary culture can yield 6 first-passage subcultures, which subsequently achi eve confluence in 10-1 2 da ys. Such cul_ tures are 3 usefu l source o fhum 3n keratin ocytes that strati fy but generally do no t undergo termin al differenti atio n . J [/I lles t D el'll/ntol 88:314- 3 '1 9, '1 987

MATE RIALS AND M ETH O D S

Culture Media and Chemicals Minimum esscntial mediu m (M E M) w ith Earle's salts supplem entcd w ith nonessenti al amino acids and g lu tamin c w erc fro m Gibco, Paisley, ScotlJnd (ca t. no . 072-1500). Fetal bo vin e serum was purchascd fro m Biochrol11 KG (Bcrlin , Wes t Ge rm any, ca t. no. S 0015) and used at a 10% final concentrati on . Gcntami cin sulfate (40 }.Lg/ ml ) was used con­tinuo usly as an antibi o ti c. M ycos tatin (160 units/ ml) was added to th c m edium for the first w eek o f culture o nl y. Dimeth yl sul f­ox id c (DM SO ) (Pi crcc C hcmi cal Co. , Rockfo rd , Illino is, sil y­btion grade, ca t. no. 20684) was used at a fiml concentratio n of 0.5%. C ulturcs wc re subculti vatcd usin g 0 .1 25% trypsin (1 :250) (Difco Labs., Detro it , Mi chi gan) and 5 mM E DT A in M E M bu ffered to pH 7.4 w ith 25 mM H E PES (N-2-h ydroxycthy lpi_ pcrazinc-N' -2- cth anesulfo nic acid).

Primary Cultures Bio psies, approximatel y 1 cm 2 X 1 mm , fro m the buccal mu cosa of 15 adult volunteers were rcm oved und cr local anesth esia. The ti ssuc w as cut into ex plants approx­imately 1 mm3

, soaked bri efl y in chi ck cmbryo ex tract (Difco), and then transfcrred to 20-}.LI dro ps of chi ck plasm a (Di fco) in 25- cm 2 NUNC LO N fl asks (NUNC, Ros kild c, I cnmark). T h e clo ts were allo wed to fo rm fo r 120 min , m edium was addcd , J n d the cultures were gassed with 5% CO2 in air befo re in cubatin g them at th e temperatures di scussed ill R es"lts. The medium was changed tw ice a w eek .

Subcultivation B y 20-25 da ys, primary epitheli al o utgrowths occupi ed mo re thJn 80% o f the culture substratum . This tim e period was selected as an o ptimum one fro m w hich to initiate subcultures beca use th cre were abundant mito ti call y acti ve ke­ratin ocy tes that could be casily dissociated usin g try psin-EDT A . Kcratinocytcs w ere subculti va tcd by in cubatin g them in 5 ml of trypsin-E DT A fo r J pprox imatci y 10 min on a wa rmin g pl ate set Jt 30°C after w hich they were dispersed into a sin glc cell sus­pension by gentle pi pettin g, and the actio n of trypsin sto pped by addin g the suspension to I n equal volume o f co mplete medium. Fo ll o wing centrifu ga tio n, th e cclls w ere res uspend ed in a sm all vo lum e o f culturc medium (0.5-0.75 ml/cultun;) to yie ld a COI1-

0022-202X/87 /$03.50 opyright © 1987 by The Society for In vestiga ti ve Dermatology , In c.

314

VO L. 88. N O. 3 MAI(C H 1')87

centratio n o f5 X 10(' cells/ml. Fivc 20-fLl dro ps ( \ 00,000 cells/drop) l were p laced o n the cu lture su rfa ce of a new Aask. T he Aasks we re

humidifi ed b y ca refull y addin g 500 fLl of m ediul11 away fro m the " d ro ps, g assed , and in cubated fo r 16-24 h to allow the cells in the i d ro ps to attach . T he subcultures were then rinsed with medi um , \ fed w ith fres h culture m ed iu m, and trea ted as conven tio nal cul-

tures the reafte r. A second subculture was initiated usin g iden t ica l techniqu es after th e first subcu lture reached conAucnce at ap-

, prox im a tely 10 d ays. T he g row th o f drop cu ltures wa s deter­m ined usin g th e m ea n of 8 rad ius m easurem ents/drop culture to ca lculate the (c ircu lar) area occupied by th e culture .

U ltrastructural Methods C ultures were rinsed in warm m e­d ium w itho ut serum before fix in g them for 2 h at 4°C in a 1:1 m ix ture o f 2.5% g lu ta raldeh yde in 0 .2 M sodium cacody late (p H 7 .3) containing 2 111 M CaC l~ with 2% aqueo us os mium tetrox ide. T he cultures were stained en bloc with 2% urany l ace tate, de­h ydrated in ascending concentratio ns of ethy l alcoho l, infiltrated w ith a 1:1 mi x ture of l 00% ethy l alco ho l-epoxy resin , and embed­ded in Epo n- Ara ldite 116 1.

RESULTS

T he percentage of ex plants that rem ained attached to the culture subst ratum fo r more th an 3 days ranged fro m 85-95% w ith a m ea n va lue of92% . Nin ety percent of the attached exp lan ts gave rise to apparentl y pure keratin ocy te o utg rowths. T he remain in g

Figure 1. Phase contrast micrograph of hum an ora l mucosa l kcratino­cytes in an epithelia l outgrowth fro lll a primary expiJnt. T he range of keratinocyte size and shape variability illustrated here is typica l of both primary cultu res and subcu ltures . Bar = 10 /Ln!.

F igure 2. Phase contras t micrograph of the expandin g edge ofa prim ary epithelial outgrowth . T he outermost cells exhibit areas where broad cy­topiJsmic processes extend out onto the culture substratum . Areas ex­hibiting such membrane activity alternated with apparentl y quiescent zones around the circum ference of the culture. Bar = 10 /L1T!.

CULTURED O RAL KEnATIN OCYTES 315

ex plants gave ri se to either o utg row th s of keratin ocy tes and fi­bro blasts o r o utg rowths co mposed entirely of fibro blasts . T hese data are th e res ult of 15 separate ex periments w ith each experi­m ent consistin g of approx imatel y 15 T - 25 Aasks (5 ex­plants/Aask). Subculture o bservatio ns described below arc based o n approx im ate ly 10-12 Aasks for fi rst subcultures and 10 Aas ks for second subcu ltures in each of the 15 experiments.

Ini tia l epitheli3l o utg rowth was o bserved 3-4 da ys afte r ex­plant3tion , w ith co nAuence usuall y achieved by 25 da ys. Incu­batio n 3t different temperatures did not m arkedly 3ffect the g rowth rate o f ·the explan t cul tures altho ug h cultures g ro wn at 32°C usu­all y reached co nAu cncy a few d3ys later than th ose g rown at 37°C . The phase microscopic appea rance of the kcratinocytes w ith in the epithelial o utg rowths was identica l rega rdless of the in cubatio n temperature.

Fibroblasts were rarely o bse rved in cultures g rown at 32°C and 34°C, and , w hen present, could eas il y be rem oved b y gentl e tryps iniza tion (1-2 min at roo m temperature). In co ntras t , fibro­blasts w ere a conspicuous feature of cu ltures g rown 3t 37°C, o ften formi ng mu ltil ayered co lo nies that occupied g reater th 3n 5 cm~ of the culture sur f3 ce, m ak in g these cultures unsuitab le fo r furth er experimentatio n.

T he ph3se co ntras t m o rph o logy of the keratinocy tes w ithin the epithelia l o u tg rowths was rem arkabl y ho m ogeneous (Fi g 1) ex­cept for the cells 3t the ad va ncin g edge of the o u tg rowths that frequentl y exhibited broad , Aattened cy toplas mic " f3 ns" that ex-

Figure 3. Phase contrast micrograph of vigorous mi to tic activity in a primary epithelial outgrowth. Although it is difficu lt to determine the precise location of the dividin g ce ll s within va rio lls cell layers with phase contras t, mall Y of the di viding cel ls (arro",s) appear to be located supra­basa ll y. Subcul tures exhibited identica l morphologic features as those illustratcd in Figs 1-3. Bar = 10 /L1l1 .

Figure 4. Phase contras t micrograph of a mu ltilayered area within a growing subculture similar to that indicated in Fig 5. The tonofibri llar network wi thin the cells such as the onc indica ted by the arro",s is con­spicuolls. Bill' = 10 /L1l1 .

316 AJ(EN HOLT-ll IN DSLEV ET AI.

tcndcd o ut o nto the culturc substratum (Fig 2). Arcas o f intcnse mitotic act iv ity werc observed prio r to the timc w hen cp ithelial outg rowths coa lesced to form a co nflucnt culture (Fig 3) . Mi tot ic activity was still o bservcd as th c co nflucnt cultures bcga n to mul­tilaycr. Evidencc of senescencc was o bservcd 5 weeks aftcr cx­plantation, w ith the numbcr of cnlarged and multinuclcatcd cell s in creasin g w ith time. Bccause of the in creasin g number o f se­nescent changes obse rved w ith time, cultures were judged not to be sat isfactory for experimcntation beyond 14 weeks, even though large areas of such cultures still contained cell s that were normal in appea rance.

Subcultures Subcultivation did no t alter th e m o rph ologic ap­pearance of th e oral mu cosa l kcratinocytes. Thc cell s g rew o ut fro m the drop cultures and eventuall y fo rm ed a confluent sheet composed of keratinocytes in a manner analogous to epi the li al o utg rowths from primary explant cultures. The cell s w ithin the o utg ro wths were s imilar to one ano ther in size and shapc. Fo l­lowin g conflu ence (app roximately 10 days), the cells began to multila yer. e lls w ithin such confluent, strat ifi ed cultures ex hib­ited prominent to nofi lam ent bundles and desmosol1les bo th with phase co ntrast (Fig 4) and electro n mi croscop y. C ulture fla sks conta ining first subcultures (drop cultures) are illustrated in Fig 5. Some sur face desquamation was observed microsco pi ca ll y in g rowin g cultures, howeve r this fcature became morc prominent in postco nflu ent , o lder cultures.

The relationship between incubation temperature and the g rowth of hum an ora l keratinocytes from hi gh-cell-density co lonies is illustratcd in Fig 6. In cubation at 34°C consistentl y yielded better growth than did culti vation at either 32°C o r 37°C. Additi onall y, culti vatio n at 34°C essentia ll y eliminated the problem of co ntam­in atin g fibrob lasts , a frequent problem with cultures m aintained

/ • !! ~

'I ® @fl .~

® \... ....--,,,. J ~-'I

5

Figure 5. Fixcd and sta ined drop cu ltures initiated from conAucnt pri­mary cultures 2 and 8 da ys after subcultivation arc illustratcd. The arca marked by the arrolV indicates a region similar to that from which the electron micrographs in this study were made.

T I-IE JOUJ(NAL 0 1= INVESTIGATIVE DERMATOLOG Y

AREAmm2

300

200

100

2

_ .. _-1--"-

5

32·C _ .. -_ .. --! 2 sue.

7 DAYS

Figure 6. The rciationship between incub3tion tcmpcrature and kerati_ nocyte growth in first and second subcultures is illustrated. Incubation at 34° consistently yielded better growth than in cubation at 37°C or 32°C . The d:lta reprcsent 15 cultures (3 Aasks) per group in this experiment. alth ough a larger numbcr of cultures co mparing just two parameters such as the effect of tcmperature on growth or the growth of onl y first sub_ cul tures yielded identical results. Bars = 2 SD.

at 37"C. Metapha se chro m oso me studies were perfo rm ed on col­chicine-a rres ted cells at the time that primary cultures were sub­culti vated . The studies indi ca ted that cu lturcd keratin ocy tcs re­tained a dip lo id chromoso m e number with no detectable pronounced chro m osom al aberration s. [This is a preliminary findin g based upon chro m oso me counts madc on 5 different culturcs (30 cells/culture) ove r a period of 1 yea r-data not shown.]

Signs of cellular senescence (progressive cellular hypertroph y, multinucleati on, detachm ent from the culture substratum) w ere observed afte r approximately 4 weeks in first subcultures and after 2 weeks in second subcultures. Mitotic activ ity ceased abo u t the sa me timc as senescent changes appeared in subcu ltures. As a conseq uence of the observcd senescent chan ges, first and secon d subcultures were judged to be unsatisfactory fo r experimentation beyond 4 weeks and 2 weeks, respectively. Subcultivation a third time resulted in in consistent and usually unsatisfactor y growth .

Cell Ultrastructure The stru cture of cells in cither primary cultures o r subcultures was typical of a cultured keratin ocyte, i. e., a fla ttened , elongated cell w ith conspi cuous to nofilam ent bundles, num erous and well-fo rm ed des moso mes, a sma ll perin uclear Golgi complex, occasional rosettes of g lycogen, and spa rcely distributed free (cytoplasmi c) and bound (rough endoplas mic reti culum) ri­boso mes. The d ispositio n o f the cell s varied depending upon the culture loca tio n sa mpl ed and ranged fro m a sin gle cell applied to the culture substratum at th e edge o fa drop culture to areas having 7-12 viable cell layers that were infrequentl y ca pped b y a ter­min all y differenti ated cell (discussed below). We have selected an area simil ar to that indica ted in Fig 5 and depicted in phase contrast in Fi g 4 as being representative of bo th hea lthy g rowin g and conflu ent cultures. In such cultures th e cells multil ayered (Fig 7a). The keratinocytes o n the culturc substratum exhibited two types of basa l cy topl as mi c arran ge ments. In one illstance, tonofi lal11ent bundlcs were prominent in the basa l cy top lasm and small hem i­desmosomes (ty pi ca l of kerat in ocytcs in cul ture) 11 6] were found alo ng the undersurface of the cell (Fig 71J) . The other arrange m ent of the basa l cell cyto plasm consisted of a zone occupied almost exclusively by 60-nm fi lam ents , typi cal of an ac tin subplasma­Icmmal netwo rk 11 7] (Fi g 8). Both types of basa l cytoplasmic confo rm ations were fo und throughout the cul tures and were not rest ri cted to anyone zone within a culture.

VOL. 88. NO. 3 MA)(CI-I 1987

Figure 7. (I , Hum an o ral kcratin ocytc subcultures sll ch as thc o nc illus­trated exhibit st ratifi cation. The ce ll s con tain abundant ronofi lalllcnts and are interconnected by dcsmoso lll cs. Ii, Kcratinocyte cytoplaslIli c slIrf.1ccs adjacent to the plastic culture substratulll fi'c'l ucntl y cxhibit slll ali . focal hcmidesmoso J11 cs si milar to thc onc indi c:1tc'd by the "'TO II I . Bar = I ILJ11 .

CULTU RED ORAL KERATINOCYTES 317

Suprabasal cells exhibited large, o rga ni zed tonofi lam ent bun­dles, well-devcloped desmosomes , and an occasional gap junctio n between adjacent ce ll s (Fig 9). Membran e- coa tin g g ranules of the ty pe found in no n keratinizin g epithelium 11 81 were no t observed nor were inclusions suggestive of keratohya line g ranules. Cells on the surface of multil ayered cultures usuall y did not exhibit signs of termina l differentiatio n such as envelope formation , lys is of cell o rganelles, o r tonofi lament dispersal (Fig 10). Occasional surfa ce cel ls did ex hibit foci of o rga nell ar lys is and tonofi lament di spersa l; rarel y, a surface cell would exhi bit envelope formation.

DISCUSSION

The hum an o ral cav ity ex hibits the g reatest range ofkera tin ocyte differentiatio n within a co nfined area of any place in the body. N onkeratini zed stratified sq uam ous epithelium abruptl y becomes keratini zed at various specifi c locations, and areas of partial ker­atini zation (pa rakeratiniza tion) alternate w ith areas of complete keratiniza tion. T his ca refull y regulated pattern of keratinocyte differentiatio n within specifi c geographi c regions of the o ral cav­ity can be altered by ph ysical o r chemi ca l irritati on to th e point where frank ca rcin om atous change can and docs occur with a very hi g h degree of frequen cy (40-50% of all ca nce rs in Sri Lanka and parts of India 11 9]). Because vve wished to stud y factors capable of switchin g the process of no rmal keratinocyte differ­entiation to an abno rmal one, we rcq uired a culturc system w here subcultures derived from the sa m c primary culture co uld bc ex­posed subsequentl y to separatc, different culture conditions. By selecti vely m od ifying conventional cell culture methods wc havc achieved reprod ucibly excellent g rowth and histod ifferentiation of hum an o ral Illucosal keratinocy tcs .

The onc feature of the culture system wc dcscribc that is , per­haps, less than desirab le is th e use of DMSO. It is no t clear what the action o f DMSO is in fac ili tatin g the growth of oral kerat­in ocytes. In our expe rience, howevcr, it is a nccessary addition in o rdcr to achicvc reprod ucibly consistcnt hum an ora l kerati­nocy te g ro wth and histod iffcrcn tiation in bo th prilllary cu ltures and subcul tures. Whcreas DMSO is not a natural product in mamlll alian ce ll system s, it has becn shown to faci li tatc g rowth of both mo use and rat kcrat inocy tcs 11 6,201.

Thc other uniquc and apparently bcncfi cial features of thc cu l­turc system we employ arc: (1) the use of drop cultures for ini­tiating subcultures, and (2) lower than normal incubation teIl1-peratures. When conflucnt primary cultures of human oral mucosa arc subculti vated at 1:2 di lutions, un satisfactory g rowth r suits. By es tabli shin g subcultures in the form of sm all colonies that have a hi gh cell density (drop cu ltures), excellent growth and differentiation res ult with an extremely favorable amplifi cation of keratinocyte numbers in culture (6 subcu ltures w ith 5 co lo­nieslAask). Establishing subcultures in the form of sm all colonies co mposed of equal numbers of cel ls also offers the additional advantages of se ri al g rowth measurem ents and direct microscopi c observation of cells as they move out over the cu lture substratum. Lower incubatio n temperatures do no t significantly alter kerati­nocyte growth but markcd ly reta rd fi broblas t g rowth that is o ften overwhelming at 37°C. T his pheno men on was first reported by one of us (A.J.) 12 years ago in a study of rat o ral keratinocytes 12 1] and was recentl y confirmcd 'in an in vitro study of hUlll an epidermis [22 1.

We ca n find onl y 2 studies in w hich successful subculti vation of hum an o ral keratinocytes has been reported 11 4,15]. In o nc case, 2-4 passages were reported using 3T3 feeder cells [1 5]; and in the other repo rt , I passage was reported-but wi th decreased mi tot ic activity 11 4 1. Whereas human keratinocy tes, whether de­rived from skin o r o ral mLi COSa, appear to have a relati vely sho rt li fespan in culture 123], the prog ressive decn:asc in the lifespa n of subcultures as co mpared with the su rvival of primary cultures observed in this stud y indica tes the process of subcu lti vation evell further decreases the lifespan of keratinocytes in culture. The

318 A REN H O L T-BINDSLEV ET AL

Figure 9. Wel l-formed desmoso mes and extensive to nofi lament bundles are characte ri sti c of spino us cell la yers in culture. The ill s£'1 dep icts the s tructure of" desmosome fro m thi s la ye r at hi gher power. Bnl' = l J.LITI , X 19,500; ill SCi X 120,000.

T H E JOU RNAL O F IN V EST IGATIV E DEHMATOLOGY

Figure 8. A po rtion o f basa l cell cyto_ plasm in a kcratinocytc subculture is il_ lustrated . Tono filamcnt bundles and dcs_ moso mcs arc present 3S arc representa tive exa mples of the types and amoun ts of Othe r o rgane ll es p resent in cultured kcrati no_ cy tes. Th is particul ar cell has a pro_ nou nced microfilam ellt (actin ) network ("I'roIllS) in the cytoplasm adjacent to the culture surface. Bnl' = 1 J.Lm .

Figure 10. Cells present on the surface o f hum an buccal mucosal sub­cultures rare ly exhibit terminal differentiation (cha racterized by tono­filam ent di spe rsa l and en velope fo rmation). The ap ica l plasma lIIembrane ex hi bits a va ri ab le number of microvilli . B"I' = I J.Lm , X 26,000; i,l s£,1 X

150, 000.

VO L. ~l:l. NO.3 MAltCH I'JH7

reasons fo r this decreased lifesp:lIl w ith progress ive subculti va tion are unknown, but may be related to the pheno men on of kerati­Ilocyte te rmin al differentiati on indu ced by detachm ent fro m the culture subs tratum 1241. the repeated ex pos ure to trypsin , o r the maj o r shi fts in ca lcium levels due to the use o f E DTA that co l­lecti ve ly redu ce th e poo l o f mitoti call y active ce lls .

The t iss ue eventuall y recotistituted by subculti vated bucca l ke­ra tin ocytes is similar, but no t identi ca l, to th e parent tiss ue just as cultured keratin ocy tes deri ved from keratini zed epithelium do

, no t re-form a f.1 ithful rep li ca o f the tiss ue fro m whi ch they were derived . A sig nal feature o f bucca l keratin ocy tes in culture is the relative lack o f keratinization (chara cterized by o rgane ll ar lys is, tono fi lam ent dispersa l, and envel ope fo rm atio n). N onkeratini z­in g epitheli a arc full y capable o f undergo ing terminal diffe renti a­tio n in culture 125i, and prev io us stud ies o f check epithelium in culture lu ve no ted varyin g degrees o f termin al diffe rentiati on exhibited b y the sur face cell s [1 4, 15,261. Even the deg ree o f sur­face cell termin al differenti ation in no rm al check mu cosa is va ri­abl e [1 8,27,28 1. W hether the la ck o f termina l differentiati on (o f the type o bse rved in either no n keratinized o r keratini zed epithelia) by human o ral keratin ocy tes in culture represents a bio logicall y sig nifi cant feature o f the tissue o r simpl y a cond itio n imposed by Culture co nditi o ns rcmains to be dete rmined. O ur pre vious ex­peri ence w ith rodent keratinizin g o ral epithelium that ex hibits a hig h degree o f keratinizat ion in vitro sugges ts tiut th e la ck o f te rmin al diffe rcnti ati on is no t du e so lel y to th e culture conditio ns [161·

KirslCII A lldrrsl'll , Lis Llllld, A llgrln Wc((o l'd, nlld 511'1'('11 MrKcll'c), I"'ol'idrd excel/r ill ICc/lllien! nssisl {IIIC<'.

n.EFE RE N C ES

1. Rhein wa ld J G , G reen H : Seria l culti vatio n o f strai ns of human epi­de rm al ke ratinocy tes: the fo rma tion of keratini zing co lo nies fro lll sin g le ce ll s. C ell 6:33 1-344. 1975

2. Hhein wa ld J G: Seri al culti va tio n of no rmal hum an ep ide rm al kera t­in ocytes. M ethods C ell Bio i 2 IA :229-254. 1980

3. Liu S-C, Karasek M : Iso lation and g rowth o f :ld ul t hum an epidermal ke ratinocytes in cell culture. J In ves t Derlll ato l 7 1: 157- 162. 1978

4. Pruni i' ras M , Regnie r M , Fo ngi: rc S. Woodle y D : Keratinocy tes synthes ize basal - lamin a p ro teins in culture. J In vest Dermato l 8 1 (suppl) :74s-81 s, 1983

5. G ilchres t 13A , N elllo re HE, Ma ci a ~ T: G rowth o f 111.1111:111 ke ratino­cy tes 0 11 fibro nectin cooted pbtc;. C dl l3i o lln t I~c p 4: 1009- 10 16. 1980

6. Lillie JH. M acCa ll ul1I DK. Jepsen A: The behav io r o f subculti vated stratifi ed squamo us epithelial cells on reconstituted ex tracellubr m atri ces: ini t ial inte ractions. Eur J C ell 13io l 29:50-60. 1982

7. Kleinman H. M cGarvey ML. Hassell .JI~ . Star VL. Ca nn on FB ,

CU LT UR ED O RAL KERATIN OCYT ES 319

Laur ic G W. M artin G R: Basement membranc complexes wi th b io logica l acti v ity. Biochemist ry 25:3 12-3 18. 1986

8. Henn ings H. M ichael D , C heng C, Steinert p. Ho lb roo k K, Yuspa SH : Ca lcium regubtion of g rowth and d iffe relltiation of mo use ep idcn nal ce lls in culture. C ell 19:245-254, 1980

9. Peehl D M , H am HG: G rowth and di ffe renti ati on of human kerat­inocy tes w ithout a feeder la ye r o r conditioned med ium . In Vi tro 16:5 16-525. 1980

10. T5ao M C. Wa ltha ll BJ , Ham I~G: lonal g row th o f no 1'11131 epide r­mal ke ratinocy tes in defin ed med ium . J C el ll'h ys io l '11 0:2 19-229. 1982

II . Watt FM , G reen H : In vo lucrin sy llthcs i5 is co rrel ated w ith ce ll size in hum an epiderm al cultures . J Cell Bio i 90:738-742, 198 1

12. Hawley-Nelson p . Sulli van J E. Kung M . Hennin gs H . Yuspa SH : Optimi zed conditions fo r the g rowth of hum an epiderm al cel ls in culture. J In ves t Derm ato I 75: 176- 182. 1980

13. G ilchres t 13A . C alho un JK . M aciag T: Attachmellt and g rowth of hum :ul ke ratinocytes in a se rulll- free enviro nmen t. J Cell Ph ysio l 11 2: 197-206, 1982

14. G uste rson 13 A. M onaghan 1': Keratinocyte diffe renti ation of hum an bu cca l l1Iu cosa in vitro . In ves t Cell I'atho l 2: 17 1- 179, 1979

15. T aichman L, Reill y S, Garan t I'R: In vit ro culti va tio n o fhul1lan o ra l knatinocy tes. Arch O ra l 13io l 24:335-34 1. 1979

I (>. J epsen A, M acCa llul1l DK , Lillie JH : Fin e stru cture o f subculri vated stratifIed squam o us. Exp C ell Hes 125: 141- 152. 1980

17. Sug rue SP. H ay ED: I~ esponsc of basa l ep itheli al cell sur f.1ce and cytoskeleton to so lub il ized ex tracellu la r matrix m olecules. J Cel l 13io l 9 1 :45-54, 198 1

18. Sq uire CA: Mem brane coa ting gra nules in non- kerat inized o ral ep­ithelium. J Ultras truct Hes 60:2 12-220, 1977 .

19. I'ind bo rg JJ: O ral Cancer and Precancer. B ri sto l. UK, Wright, 1980

20. YUSp:l SH , H arris CC: Altered diffe renti ation o f mo use epidermal cells trea ted w ith rctin yl ace tate in v itro. Ex p Cell Res 86:95-105, 1974

2 1. Jepsen A: An in vitro 11I 0del o f o ral keratini zin g squamo us epithe­lium . Sca nd J Dent I~es 82: 144- 146, 1974

22. JensCII PKA . Therkel sen AJ: C ult iva tion at low temperature as a measure to prevent contaminatio n w ith fibro blasts in epi thel ial cul tures of hulllan skin . J In ves t Derm ato I 77:2 10-2 12, 198 1

23. G reen H : T he keratin ocy te as d iffe rentia ted cell type. Ha rvey Lect 74:10 1- 139, 1979

24. G reen H: T erminal d ifferenti ati on of cu ltllred human ep ide rl1l al cell s. C ell 11 :405-4 16. 1977

25. Banks-Schlegel S, G reen H: In vo lu crin synthes is and tissue assem bly by keratinocytes in natural and cultured human epithelia. J C ell l3io I 90:732-737, 1981

26. Fb xl1lan EA . LutZller ML, Van Scon EJ : ell maturation and tissue o rganization in epithelial outg rowths fro m skin and o ral l1Iu cosa in vitro. J In ves t Dermato l 49:322-332. 1967

27. Frithio fL: U ltrastructural chan ges in the p las l1I a l1I embrane in hum an o ral epi thelium . J U ltrastruct Res 32: 1- 17. 1970

2H. Schroeder HE : Differentiation o f HUl1l an Ora l Stratifi ed E pi theli a. J3 ascJ . Sw itzerbnd . Karger. 19H I