Influence of anti-CD49f and anti-CD29monoclonal antibodies on mitoticactivity of epithelial cells (HaCaT) andgingival fibroblasts in vitro

Becker Ch, Buttler P, Graber HG. Influence of anti-CD49f and anti-CD29monoclonal antibodies on mitotic activity of epithelial cells (HaCaT) and gingivalfibroblasts in vitro. Eur J Oral Sci 2002; 110: 137–143. # Eur J Oral Sci, 2002

A major complication in the treatment of periodontitis marginalis is thereepithelization of periodontal defects inhibiting collagen fiber attachmentand periodontal regeneration. In this study we investigated the possibility of amolecular blockade of epithelial mitosis in vitro. Monoclonal antibodies againstthe VLA-6 laminin receptor subunit a6 interrupted interactions between epithelialcells (HaCaT cells) and their extracellular matrix and thus resulted in reductionof proliferation rates by more than 50%. The same effect was observed withantia1-antibodies. In contrast, collagen-producing and -secreting gingival fibro-blasts, which play an important role in periodontal regeneration, remainedunaffected by the applied antia6 antibodies. Correspondingly, these cells werefound to lack VLA-6 laminin receptors. Selective molecular inhibition of epithelialproliferation and apical migration by monoclonal antia6 antibody applicationmay provide an adjuvant periodontitis therapy resulting in an enhancedperiodontal regeneration.

ChristophBecker1,2, PatriciaButtler1,Hans Georg Graber1

1Department for Operative Dentistry,Periodontology and Preventive Dentistry,Medical Faculty, Rheinisch WestfalischeTechnische Hochschule (RWTH) Aachen,Germany, 2Interdisciplinary Center of ClinicalResearch in Biomaterials and Tissue–MaterialInteraction in Implants ‘BIOMAT.’, MedicalFaculty, RWTH Aachen, Germany

Dr rer. nat. Christoph Becker, Department forOperative Dentistry, Periodontology andPreventive Dentistry, Medical Faculty, RWTHAachen, Pauwelsstraße 30, D-52074 Aachen,Germany

Telefax: +49–241–8082468E-mail: cbecker@ukaachen.de

Key words: periodontal regeneration; epithelialproliferation; mitosis inhibition; monoclonalanti-integrin a6 antibody; VLA-6 lamininreceptor

Accepted for publication January 2002

Marginal periodontitis is an inflammatory disease ofthe periodontium including the adjacent soft tissue.Insufficient oral hygiene results in plaque accumulation,gingivitis and, finally, degradation of connective tissue,root cementum and alveolar bone. Conventional therapycomprises surgical cleaning and disinfection (surfacescaling or curettage). A major problem with this type ofintervention is a postoperative proliferation and migra-tion of the junctional epithelium in apical direction. Theepithelial coating of root surfaces inhibits reattachmentof supraalveolar and periodontal ligament collagen fibersas well as de novo synthesis of destroyed structures. Thusperiodontal regeneration is prevented. More elaboratetherapies try to inhibit epithelial extension using guidedtissue regeneration. However, these treatments are equallyunsatisfactory (1).

Therefore, we explored the possibility to inhibit epithe-lial proliferation on the molecular level as an adjuvanttherapy. Proliferation and apical migration of epitheliumis initiated by adhesive contacts between epithelial cellsand their extracellular matrix (ECM) (2, 3). The cohesionbetween epithelial cells and the ECM is mediated byspecific cellular adhesion receptors called integrins (4).Integrins are cell surface receptors with distinct bindingspecificity. Various integrin subtypes play an importantrole in the contacts of basal epithelial cells and the basallamina (5–7), cell migration and wound healing (8, 9), the

initiation of intercellular adhesion (10, 11), and celldifferentiation as well as proliferation (4, 12, 13).It has been shown that adhesion of epithelial cells

to different kinds of ECM components in vitro can beblocked by antibodies directed against integrin subunits(14, 15). Furthermore, studies on cultured human gingivademonstrated that migration of junctional epitheliumcould be suppressed by blocking the integrin subunits a6,a1 or both with monoclonal antibodies (16).We chose the immortalized human keratinocyte cell

line HaCaT as a model system to quantify inhibitionof mitotic activity by blocking the integrin subunitsa6 and a1 with monoclonal antibodies. HaCaT cellsare described as a well characterized and standardizedepithelial cell line (17) which is easily manageable in vitro.They were used in a number of different studies, amongother things, as model for junctional epithelium in pro-liferation studies (18). We are well aware that HaCaTcells are different from junctional epithelium in vivo.However, in some features this cell line shows similarbehavior to junctional epithelial cells, particularly withrespect to integrin expression and proliferation capacity:(i) HaCaT cells have a high basal proliferation rate (17);(ii) On their cell surface, HaCaT cells express similarintegrin subunit patterns, namely a2, a3, a5, a6, aV, a1,a4 and a5 (4, 9, 19–22); (iii) HaCaT cells synthesizeand deposit ‘classical’ basement membrane (BM) proteins

Eur J Oral Sci 2002; 110: 137–143Printed in UK. All rights reserved

Copyright# Eur J Oral Sci 2002

European Journal ofOral Sciences

ISSN 0909-8836

such as laminin-5, laminin-1, collagen IV, nidogen andsome other BM components (21, 23); (iv) HaCaT cellsinteract with extracellular wound-matrix molecules suchas fibronectin (24, 25); (v) HaCaT cells show ECMcomponent-mediated migration which can be suppressedby antibodies against integrin subunits (9, 26).

In parallel, we studied the inhibition of proliferation bythe same integrin antibodies in gingival connective tissue.Decreased proliferation of gingival and periodontal fibro-blasts is undesirable regarding their important functionof producing and secreting collagen fibers. In early stagesof periodontitis, tissues near the dentogingival junctionare affected by the destructive progression of the disease.As a representative in vitro model, we chose primarycultivated gingival fibroblasts, which form a main popu-lation of the marginal gingiva and are significantlyinvolved in the formation of the supraalveolar fiberapparatus.

Material and methods

Cells and culture conditions

Organ culture: Biopsies from human gingival sulcus tissuederived from healthy donors and non-separated pieceswere cultivated as described elsewhere (16). Integrin expres-sion in the epithelial stratum was determined immuno-histochemically from day 0 to day 30 after biopsy.

The human permanent keratinocyte cell line HaCaTwas obtained from the German Cancer Research Center(DKFZ), Heidelberg, Germany (17). Cells from frozenstocks were grown on polystyrene culture dishes (Greiner,Solingen, Germany) using DMEM High Glucose (1r)(PAA Laboratories, Linz, Austria), supplemented by10% fetal calf serum (FCS), penicillin (100 U mLx1) andstreptomycin (100 mg/mL) at 37uC and 7.5% CO2 asdescribed elsewhere (23). For proliferation assays we usedthe 3rd passages after thawing.

Primary culture of human gingival fibroblasts: Gingivalsulcus tissue (3–6 mm3) was obtained from clinically healthydonors after premolar extraction. Tissue was rinsed for 24 hin MEM (Gibco, Paisley, Scottland), supplemented bygentamycin (100 mg/mL) and fungizione (25 mg/mL), andwashed in phospate-buffered saline (PBS). The epitheliumwas mechanically separated, and connective tissue was cutinto fragments of approximately 0.5 mm3. These specimenswere attached to polystyrene dishes wetted with DMEMHigh Glucose (1r), supplemented by 10% FCS and genta-mycin (50 U mLx1). After 6 h at 37uC and 7.5% CO2,loosely attached tissue fragments were removed and theoutgrown fibroblasts were covered with medium. Mediumwas changed every 2nd day. For proliferation assays we usedcells from 5th to 7th passages.

Antibodies and immunoreagents

Monoclonal rat-anti-mouse a6: anti-CD49f (cloneGoH3; Immunotech, Marseille, France). Monoclonalmouse-anti-human a1: anti-CD29 (clone P4C10; Biomol,Hamburg, Germany). goat-anti-rat IgG FITC-conjugated(Sigma, Deisenhofen, Germany). sheep-anti-mouse IgGCy3-conjugated (Sigma). APAAP-System: rabbit-anti-rat/–mouse IgG; APAAP; New Fuchsin Substrate (Dako).Rat-IgG2a; mouse-IgG1 (Dako, Glostrup, Denmark).

Bovine serum albumine (BSA), fraction V (Boehringer,Ingelheim, Germany).

Immunohistochemistry

Cryo-sections of cultured gingiva were fixed in acetone,blocked with 2% BSA/PBS and incubated with primary anti-integrin-antibodies (GoH3: 3 mg/mL, P4C10: 0.5 mg/mL),blocking buffer only (negative control) or with rat-IgG2a(non-specific control for GoH3) or mouse-IgG1 (non-specific control for P4C10) in proportionate proteinconcentrations. Detection was carried out using theAPAAP-New Fuchsin-system, according to the manufac-turer’s instructions. Integrin localization was visualizedby light microscopy.HaCaT cells and gingival fibroblasts were fixed in acetone/

ethanol (1 : 1) and blocked with 2% BSA/PBS. Sampleswere incubated consecutively with primary anti-integrin-antibodies and fluorescence labeled secondary antibodies indefined concentrations: rat-antimouse a6 (GoH3): 3 mg/mL;mouse-antihuman a1 (P4C10): 0.5 mg/mL; goat-antirat IgGFITC: 1 : 100; sheep-antimouse IgG Cy3: 1 : 50. Negativeand non-specific controls were carried out with blockingbuffer and rat-IgG2a or mouse-IgG1, respectively, asdescribed above. Integrin localization was visualized byconfocal laser-scan microscopy: green fluorescence (FITC)at e=488 nm and red fluorescence (Cy3) at e=565 nm.

Cell proliferation

HaCaT cells and primary gingival fibroblasts(30.000 cells/mL) were seeded in 96-well polystyrene culturedishes (Becton Dickinson, Meylan, France). Cells wereincubated with either anti-integrin antibodies in supra-maximal concentrations (antia6 (GoH3): 30 mg/mL; antia1(P4C10): 5 mg/mL), medium only (negative control), 10%ethanol (positive control) or 2% BSA (non-specific control)at 37uC for 72 h. Mitotic activity was estimated using acolorimetric cell proliferation assay with bromo-2-desoxy-uridine (Boehringer, Mannheim, Germany). Viability ofcells was determined by measuring the activity of mito-chondrial dehydrogenases in WST-1 cell proliferation assay(Roche; Mannheim, Germany).

Results

An important consideration to the use of the humanepithelial cell line HaCaT as a representative model ofhuman junctional epithelium is that these cells shouldexpress the same integrin receptor subtypes (especiallyregarding a6 and a1). Using monoclonal antibodies aspreviously described (16), we tested both the presenceof integrin subunit a6 and of a1 in epithelial stratum oforgan culture from human gingival biopsies and inmonolayers of the HaCaT cell line (Figs. 1 and 2). The cellsurface proteins a6 and a1 were found to be uniformlydistributed in the epithelial stratum of cultivated humangingiva (Fig. 1, arrows). These integrins were consistentlypresent from day 0 until day 30 of primary culture ofbiopsies (not shown). Likewise, the surface of HaCaTcells was surrounded by dense bands of integrin a6(Fig. 3b) as well as integrin a1 (Fig. 3c). The expression ofboth proteins was not affected by the number of pas-sages, nor by repeated freezing and thawing procedures

138 Becker et al.

(up to 50 times), nor by the degree of confluence ofcells (not shown). Specificity of immunostaining wasensured by non-specific controls using IgG2a and IgG1,respectively.

We next incubated HaCaT cells with monoclonalantibodies against the integrins a6 and a1 and determinedcellular proliferation in a BrdU-assay. Compared tountreated cells (negative control), the proliferation of theHaCaT cells was significantly reduced after treatmentwith both antia6 and antia1 antibodies (Fig. 3): mitotic

Fig. 1. Expression pattern of integrin subunits a6 and a1 inepithelial stratum of human gingiva. Organ cultures of humangingiva were immunostained with monoclonal antibodiesagainst the intergrin subunits a6 and a1, respectively, and theAPAAP-New Fuchsin-system. Images were generated usinglight microscopy. Magnification: 250r. 1a: integrin subunits a6(arrows). 1b: integrin subunits a1 (arrows).

Fig. 2. Expression pattern of integrin subunits a6 and a1 inthe epithelial cell line HaCaT.Monolayers of HaCaT cellswere immunostained with monoclonal antibodies against theintergrin subunits a6 and a1, respectively, and fluorescencelabeled secondary antibodies. Images were generated usingconfocal laser-scan microscopy. Magnification: 400r. 2a: Cellsin phase contrast modus. 2b: FITC-stain of integrin sub-units a6 (eE=488 nm). 2c: Cy3-stain of integrin subunitsa1 (lE=565 nm).

Mitosis inhibition of epithelial cells 139

activity was inhibited by 53% and 51%, respectively.The inhibition was nonadditive in that the combinedapplication of both monoclonal antibodies did not resultin a higher degree of inhibition (47% of the negativecontrol). Samples containing bovine serum albumine(non-specific controls) did not significantly differ fromuntreated samples (negative controls), indicating thatthe measured inhibition of mitotic activity is due to thespecific action of the monoclonal antibodies, and thus tothe specific blockade of the integrins a6 and a1. Viabilityof the cells was proved using WST-1 proliferation assay.Cells incubated with antia6 or antia1-antibody, or both,did not differ from cells of the negative control.

In contrast to the results obtained with HaCaT cells,the antibodies against the integrins a6 and a1 exerteda negligible effect on the mitotic activity of the assayedfibroblasts (Fig. 4). In particular, the antibody againsta6-integrin caused no significant difference in the pro-liferation rate compared to the negative control. Withthe antia1 antibody, we observed a decrease of mitoticactivity by 12%. Nearly the same value (89% of thenegative control) was found in the samples which wereexposed to the combined antibodies, while viability ofthe cells remained unchanged.

In a further step, we determined the integrin expressionpattern in the fibroblasts. Immunostaining of the integrinsrevealed the presence of subunit a1 (Fig. 5c) but not of

subunit a6 (Fig. 5b). This pattern was consistently foundfrom day 0 (growth of the fibroblasts out of the tissuefragments) until passage number 7, the highest passageused in the proliferation assays (not shown). Specificitywas proved by non-specific controls.

Discussion

Our observations in the epithelial cell line HaCaT suggestthat integrin binding and signalling through the integrinsa6 and a1 partly regulates the mitotic activity. Inparticular, the following findings support this hypothesis:(i) blocking the cell surface receptors a6 and a1 withmonoclonal antibodies resulted in a marked reduction ofthe mitotic activity of the HaCaT cells by more than half(Fig. 3, columns 4, 5); (ii) both the antibodies againstintegrin a6 and the ones against integrin a1 producednearly the same degree of inhibition; (iii) the inhibitoryeffects of the two antibodies were not additive (Fig. 3,column 6). A cell surface protein composed of a6 and a1integrins is the VLA-6 laminin receptor (CD49f/CD29),a member of the Very Late Activation antigen familyknown to be involved in cell–cell and cell–extracellularmatrix interactions (27–29). According to the expressionexperiments in the present study, this protein is likely akey regulator of proliferation in epithelial cells.

Fig. 3. Effects of antia6 and antia1 antibodies on proliferationof the epithelial cell line HaCaT.Log-phase cells were incubatedfor 72 h with either anti-integrin antibodies and their combina-tion, medium only (negative control), 1 mg mLx1 BSA (non-specific control) or with 10% (v/v) ethanol (positive control).Subsequently, the mitotic activities of the cells were determinedusing BrdU-assay. ELISA was measured at l=450 nm.Statistical evaluation: data originate from n=4 (positive andunspecific controls) or n=6 (antibody-samples and negativecontrol) independent measurements, respectively; datatSEM;(ns): no specific difference and (**): p<0,01 compared tonegative control (one-way analysis of variance, ANOVAreferring to Dunnet).

Fig. 4. Effects of antia6- and antia1-antibodies on proliferationof primary cultivated human gingival fibroblasts. Log-phasecells were incubated for 72 h with either anti-integrin anti-bodies and their combination, medium only (negative control),1 mg mLx1 BSA (non-specific control) or with 10% (v/v)ethanol (positive control). Subsequently, the mitotic activitiesof the cells were determined using BrdU-assay. ELISA wasmeasured at l=450 nm. Statistical evaluation: data originatefrom n=4 independent measurements; datatSEM; (ns):no specific difference and (**): p<0,01 compared to negativecontrol (one-way analysis of variance, ANOVA referring toDunnet).

140 Becker et al.

In the present study we compared the integrin-dependent proliferation of epithelial and fibroblasticcells. Studies with human primary epithelia are consis-tently fraught with the fact that most epithelia cannotbe readily cultured. Accordingly, we repeatedly failedto establish primary junctional epithelial cell cultures dueto lack of sufficient donor material. In cases wherecultures could be grown, the cells invariably lost theirepithelial phenotype before sufficient numbers of cellshad been generated.We therefore resorted to HaCaT cellswhich have been repeatedly used as model cells forhuman epithelial differentiation and proliferation studies(4, 9, 18, 20, 21, 26, 30). Two findings described in thisstudy may support the use of HaCaT cells as a substituteof junctional epithelium: (i) both HaCaT cells (Fig. 2)and cultured explants of human gingival epithelium(Fig. 1) showed expression of integrin a6 as well as ofintegrin a1; (ii) in previous studies, we demonstrated anti-integrin antibody (antia6 and antia1)-induced inhibitionof epithelial migration using organ culture of healthygingiva (16). Another hint to a physiological relevantin vitro model for junctional epithelium is the high basalproliferation rate of HaCaT cells (0.03/h, as determinedin growth curves).An important aspect for assessing a possible future

therapeutic use of the examined monoclonal antibodiesis their effect on growth characteristics of other perio-dontal and gingival tissue fractions, regarding, e.g. theirimportant feature producing adherent collagenic fibers.Any inhibitory action would oppose periodontalregeneration. In the present study the applied antibodieshad no (antia6) or only negligible (antia1) effect onthe proliferation of primary cultivated human gingivalfibroblasts (Fig. 4, columns 4, 5). The absence of integrina6 indicates that the gingival fibroblasts do not expressthe VLA-6 laminin receptor, and thus mitosis cannotbe blocked by antibodies against this molecule. Theinhibitory effect on the fibroblast’s mitotic activity causedby antia1 antibodies (Fig. 4, column 5) contributes tointeractions with other a1-containing heterodimericmolecules distinct from VLA-6, for example the VLA-5fibronectin receptor (a5a1 31–33). The integrin subunita1 is ubiquitously expressed and frequently used as aplasma membrane marker in mammalian cells (34–36).However, these a1-molecules seem to have minorregulatory influence on proliferation of these cells than,on the other hand, VLA-6 receptors affect this parameterin, e.g. epithelia. Since antia6 antibodies had no effect onthe proliferation of the gingival fibroblasts, the prolifera-tion repression detected in samples incubated with bothantibodies (Fig. 4, column 6) may likely be ascribedto the exclusive action of antia1.In conclusion, our results obtained by integrin

blockades of the two cell types, HaCaT cells and gingival

Fig. 5. Expression pattern of integrin subunits a6 and a1 inprimary cultivated human gingival fibroblasts. Monolayersof human gingival fibroblasts were immunostained withmonoclonal antibodies against the intergrin subunits a6

and a1, respectively, as desribed above. Images were generatedusing confocal laser-scan microscopy. Magnification: 400r.5a: Cells in phase contrast modus. 5b: FITC-stain of integrinsubunits a6 (eE=488 nm). 5c: Cy3-stain of integrin subunits a1(lE=565 nm).

Mitosis inhibition of epithelial cells 141

fibroblasts, suggest that epithelial proliferation maybe dependent on intact VLA-6 laminin receptor bindingand signalling; a lack in the external signalling (e.g. byblocking the receptor’s subunit (s) with antibodies) aswell as a lack in the receptor’s expression (e.g. of subunita6) result in a loss of mitosis regulation via this pathway.Therefore, selective molecular inhibition of epithelialproliferation and apical migration by antia6 antibodiesshould provide a novel and effective adjuvant therapyof periodontitis.

Acknowledgments – We thank Prof Dr Willi Jahnen-Dechentfor his editorial support and the Institute of Pathology(Medical Faculty, RWTH Aachen) for their helpful collabora-tion in Confocal Laser-Scan Microscopy. This researchproject is supported by the Interdisciplinary Center of ClinicalResearch in Biomaterials and Tissue–Material Interaction inImplants ‘BIOMAT’ (BMBF project no. 01 KS 9503/9).

References

1. TROMBELLI L. Periodontal regeneration in gingival recessiondefects. Periodontol 2000, 1999; 19: 138–150.

2. GRINNELL F. Wound repair, keratinocyte activation and integrinmodulation. J Cell Sci 1992; 101: 1–5.

3. BARTOLD PM, WALSH LJ, NARAYANAN AS. Molecular and cellbiology of the gingiva. Periodontol 2000, 2000; 24: 28–55.

4. GRABER HG, CONRADS G, WILHARM J, LAMPERT F. Role ofinteractions between integrins and extracellular matrix compo-nents in healthy epithelial tissue and establishment of a longjunctional epithelium during periodontal wound healing: areview. J Periodontol 1999; 70: 1511–1522.

5. MARCHISIO PC, BONDANZA S, CREMONA O, CANCEDDA R,DE LUCA M. Polarized expression of integrin receptors (alpha 6beta 4, alpha 2 beta 1, alpha 3 beta 1, and alpha v beta 5) andtheir relationship with the cytoskeleton and basement mem-brane matrix in cultured human keratinocytes. J Cell Biol 1991;112: 761–773.

6. SONNENBERG A, CALAFAT J, JANSSEN H, DAAMS H,RAAIJ-HELMER LM, FALCIONI R, KENNEL SJ, APLIN JD,BAKER J, LOIZIDOU M. Integrin alpha 6/beta 4 complexis located in hemidesmosomes, suggesting a major role inepidermal cell-basement membrane adhesion. J Cell Biol 1991;113: 907–917.

7. HORMIA M, VIRTANEN I, QUARANTA V. Immunolocalization ofintegrin alpha 6 beta 4 in mouse junctional epithelium suggestsan anchoring function to both the internal and the externalbasal lamina. J Dent Res 1992; 71: 1503–1508.

8. DOUGLASS GD, ZHANG K, KRAMER RH. The role of integrinadhesion receptors in gingival wound healing. J Calif DentAssoc 1992; 20: 37–40.

9. SCHARFFETTER-KOCHANEK K, KLEIN CE, HEINEN G, MAUCH C,SCHAEFER T, ADELMANN-GRILL BC, GOERZ G, FUSENIG NE,KRIEG TM, PLEWIG G. Migration of a human keratinocyte cellline (HACAT) to interstitial collagen type I is mediated bythe alpha 2 beta 1-integrin receptor. J Invest Dermatol 1992;98: 3–11.

10. CARTER WG, WAYNER EA, BOUCHARD TS, KAUR P. The roleof integrins alpha 2 beta 1 and alpha 3 beta 1 in cell-cell andcell-substrate adhesion of human epidermal cells. J Cell Biol1990; 110: 1387–1404.

11. LARJAVA H, PELTONEN J, AKIYAMA SK, YAMADA SS,GRALNICK HR, UITTO J, YAMADA KM. Novel function forbeta 1 integrins in keratinocyte cell–cell interactions. J Cell Biol1990; 110: 803–815.

12. HOTCHIN NA, KOVACH NL, WATT FM. Functional down-regulation of alpha 5 beta 1 integrin in keratinocytes isreversible but commitment to terminal differentiation is not.J Cell Sci 1993; 106: 1131–1138.

13. WATT FM, KUBLER MD, HOTCHIN NA, NICHOLSON LJ,ADAMS JC. Regulation of keratinocyte terminal differentiationby integrin– extracellular matrix interactions. J Cell Sci 1993;106: 175–182.

14. HERGOTT GJ, NAGAI H, KALNINS VI. Inhibition of retinalpigment epithelial cell migration and proliferation with mono-clonal antibodies against the beta 1 integrin subunit duringwound healing in organ culture. Invest Ophthalmol Vis Sci 1993;34: 2761–2768.

15. TAKADA Y, PUZON W. Identification of a regulatory regionof integrin beta 1 subunit using activating and inhibitingantibodies. J Biol Chem 1993; 268: 17597–17601.

16. GRABER HG, WILHARM J, CONRADS G. Monoclonal antibodiesagainst integrin subunits alpha6 and beta1 inhibit migrationof gingival epithelium in organ culture. J Periodontol 1999;70: 388–393.

17. BOUKAMP P, PETRUSSEVSKA RT, BREITKREUTZ D, HORNUNG J,MARKHAM A, FUSENIG NE. Normal keratinization in aspontaneously immortalized aneuploid human keratinocytecell line. J Cell Biol 1988; 106: 761–771.

18. POLLANEN MT, SALONEN JI, GRENIER D, UITTO V-J. Epithelialcell response to challenge of bacterial lipoteichoic acids andlipopolysaccarides in vitro. J Med Microbiol 2000; 49: 245–252.

19. BOUKAMP P, FUSENIG NE. ‘Trans-differentiation’ from epi-dermal to mesenchymal/myogenic phenotype is associated witha drastic change in cell-cell and cell-matrix adhesion molecules.J Cell Biol 1993; 120: 981–993.

20. BREITKREUTZ D, STARK HJ, MIRANCEA N, TOMAKIDI P,STEINBAUER H, FUSENIG NE. Integrin and basement membranenormalization in mouse grafts of human keratinocytes –implications for epidermal homeostasis. Differentiation 1997;61: 195–209.

21. BREITKREUTZ D, SCHOOP VM,MIRANCEA N, BAUR M, STARK HJ,FUSENIG NE. Epidermal differentiation and basement mem-brane formation by HaCaT cells in surface transplants. Eur JCell Biol 1998; 75: 273–286.

22. TOMAKIDI P, MIRANCEA N, FUSENIG NE, HEROLD-MENDE C,BOSCH FX, BREITKREUTZ D. Defects of basement membraneand hemidesmosome structure correlate with malignantphenotype and stromal interactions in HaCaT-Ras xenografts.Differentiation 1999; 64: 263–275.

23. BREITKREUTZ D, STARK HJ, PLEIN P, BAUR M, FUSENIG NE.Differential modulation of epidermal keratinization in immor-talized (HaCaT) and tumorigenic human skin keratinocytes(HaCaT-ras) by retinoic acid and extracellular Ca2+.Differentiation 1993; 54: 201–217.

24. SCHOOP VM, MIRANCEA N, FUSENIG NE. Epidermal organiza-tion and differentiation of HaCaT keratinocytes in organo-typic coculture with human dermal fibroblasts. J InvestDermatol 1999; 112: 343–353.

25. KOIVISTO L, LARJAVA K, HAKKINEN L, UITTO VJ, HEINO J,LARJAVA H. Different integrins mediate cell spreading,haptotaxis and lateral migration of HaCaT keratinocyteson fibronectin. Cell Adhes Commun 1999; 7: 245–257.

26. TENAUD I, LEROY S, CHEBASSIER N, DRENO B. Zinc, copperand manganese enhanced keratinocyte migration through afunctional modulation of keratinocyte integrins. Exp Dermatol2000; 9: 407–416.

27. KUMARNM, SIGURDSON SL, SHEPPARDD, LWEBUGA-MUKASA JS.Differential modulation of integrin receptors and extra-cellular matrix laminin by transforming growth factor-beta 1in rat alveolar epithelial cells. Exp Cell Res 1995; 221: 385–394.

28. SALANOVA M, STEFANINI M, DE CURTIS I, PALOMBI F. Integrinreceptor alpha 6 beta 1 is localized at specific sites of cell- to-cellcontact in rat seminiferous epithelium. Biol Reprod 1995;52: 79–87.

29. LOTZ MM, RABINOVITZ I, MERCURIO AM. Intestinal restitution:progression of actin cytoskeleton rearrangements and integrinfunction in a model of epithelial wound healing. Am J Pathol2000; 156: 985–996.

30. HAKKINEN L, WESTMARK J, JOHANSSON N, AHO H, PELTONEN J,HEINO J, KAHARI VM. Suprabasal expression of epidermalalpha 2 beta 1 and alpha 3 beta 1 integrins in skin treated withtopical retinoic acid. Br J Dermatol 1998; 138: 29–36.

142 Becker et al.

31. HORMIA M, KONONEN M. Immunolocalization of fibronectinand vitronectin receptors in human gingival fibroblastsspreading on titanium surfaces. J Periodont Res 1994;29: 146–152.

32. IVANOVSKI S, KOMAKI M, BARTOLD PM, NARAYANAN AS.Periodontal-derived cells attach to cementumattachment proteinvia alpha 5 beta 1 integrin. J Periodont Res 1999; 34: 154–159.

33. SCRAGG MA, CANNON SJ, RANGARAJAN M, WILLIAMS DM,CURTIS MA. Targeted disruption of fibronectin–integrininteractions in human gingival fibroblasts by the RI proteaseof Porphyromonas gingivalis W50. Infect Immun 1999;67: 1837–1843.

34. FUCHS E, DOWLING J, SEGRE J, LO SH, YU QC. Integratorsof epidermal growth and differentiation: distinct functionsfor beta 1 and beta 4 integrins. Curr Opin Genet Dev 1997;7: 672–682.

35. LEE W, MCCULLOCH CA. Deregulation of collagen phagocytosisin aging human fibroblasts: effects of integrin expression andcell cycle. Exp Cell Res 1997; 237: 383–393.

36. SINCOCK PM, MAYRHOFER G, ASHMAN LK. Localization ofthe transmembrane 4 superfamily (TM4SF) member PETA-3(CD151) in normal human tissues: comparison with CD9,CD63, and alpha5beta1 integrin. J Histochem Cytochem 1997;45: 515–525.

Mitosis inhibition of epithelial cells 143