(1) Departamento de Patología, Medicina y Cirugía Bucal, Universidad Santa María, e 
Instituto de Investigaciones Odontológicas, Universidad Central de Venezuela, Caracas, Venezuela
(2) Department of Oral Pathology, School of Clinical Dentistry, University of Sheffield, 
Claremont Crescent, Sheffield S10 2TA, United Kingdom
(3) Department of Histopathology, Queen Victoria Hospital, Holtye Road, East Grinstead, 
West Sussex RH19 3DZ, United Kingdom

Address:
Mariana Villarroel Dorrego
Facultad de Odontología, Módulo 10. Universidad Santa María
Vía Mariches. Caracas-Venezuela
Fax: 0058- 2129780403
E-mail: mvillarroeldorrego@cantv.net

Received: 4-06-2004  Accepted: 11-02-2005

Keratinocytes are the cells that form the bulk of the stratified squamous epithelium which covers the skin and mucous membranes of the upper aerodigestive and female genital tracts. Keratinocytes are able to express MHC class II molecules, thus they might have an immunological function and even act as antigen presenting cells (APC). Subsequently, the importance of the membrane-bound costimulatory proteins in immune potency became apparent. Among membrane-bound costimulatory pathways, two of the critical interactions are between CD80/CD86 and their receptor CD28, and CD40 and its ligand CD40L (CD154) (1).

CD80, a 60 kD protein, is a member of the immunologlobulin (Ig) superfamily and was the first ligand to be identified for CD28 (2). CD86, a 80 kD protein, a member of the same family, was subsequently recognized (3,4). The overall structure of CD80 is very similar to CD86 (5). Both molecules have two Ig-like extracellular domains, a transmembrane domain, and a cytoplasmic tail.

CD80 and CD86 expression is almost exclusively restricted to lymphoid tissues, primarily on "professional" APC, including Langerhans cells (6), monocytes (3) and activated B cells (7). CD80 is absent in resting APC, in contrast to CD86 which is constitutively expressed at low levels (8) and rapidly upregulated after antigen presentation (9). This leads to the notion that CD86 may initiate the immune response, whereas CD80 may serve to regulate and amplify it (10).

Signals delivered through the cytoplasmic tail of MHC class II molecules induce CD86 and CD28 upregulation, amplification and interaction, with simultaneous activation of CD40-CD40L. CD40 is a 48 kD phosphorylated glycoprotein belonging to the tumour necrosis factor receptor superfamily. In humans, CD40 is composed of a 171 amino acid extracellular domain, a 22 amino acid transmembrane domain, and a 62 amino acid cytoplasmic tail (11). CD40 is expressed by "professional" APC, as well as some non-lymphoid cells such as fibroblasts and keratinocytes (12).

Expression of CD80 and CD86 has been observed on cell lines derived from gastric, oesophageal, colorectal and hepatic carcinomas which have been stimulated with IFNγ (13,14). By contrast, these costimulatory molecules have not been observed on head and neck, including oral, squamous cell carcinomas either in vivo or in vitro (15-18).

Expression of CD40, contrary to CD80 and CD86, has also been found on numerous cell lines derived from carcinomas from different anatomical sites (19), raising the question of the role of CD40 in carcinogenesis. Presence of CD40 on head and neck carcinoma cell lines has been observed previously (15,18,20,21) however, the role of CD40 on oral keratinocytes remains unclear.

The apparent significance of CD80 and CD86 in the host response to malignant neoplasia is illustrated by findings which showed that MHC class II expression, in the absence of CD80 and CD86, limited CD4+ T cell activation to levels which were insufficient to induce tumour regression (22,23). The reason for tumour growth in the presence of lymphocytes is uncertain. It has been suggested that the immune response kills "less dangerous" cells, thus favouring the selection of a particularly malignant phenotype of tumour cells, that the immune system is suppressed by the carcinogenic stimulus, that the tumour produces suppressive factors which inhibit the immune response, and/or the tumour escapes recognition by host T lymphocytes through loss of MHC class II or costimulatory molecules (24-26).

The aim of this study was to determine the expression of MHC class II and the costimulatory molecules CD80, CD86 and CD40 in keratinocytes derived from oral squamous cell carcinomas.

Oral keratinocyte lines. Primary cultures from oral healthy mucosa and seven established cellular lines derived from oral squamous cell carcinomas were included in this study. Normal oral keratinocyte (NOK) were prepared as outgrowths from redundant normal mucosal tissue obtained with permission during minor oral surgical procedures. Cultures were established by the direct explant method (27). Biopsy material was soaked briefly in alcohol, phosphate buffer saline (PBS) and finally in keratinocyte growth medium (Dulbecco’s modified Eagle’s media with sodium piruvate and 1000mg/l glucose supplemented with nutrient mixture F-12 [HAM] with L-glutamine, 10% foetal bovine serum, epidermal growth factor, hydrocortisone, insulin, adenine, penicillin, streptomycin and amphotericin B). Epithelium was removed and cut into small pieces, approximately 1mm³ in size, and seeded in 25cm² flasks. Tissue was left to adhere and then cultured in keratinocyte growth medium. Cells were cultured at 37° C in a humidified incubator in 5% CO₂/95% air and microscopically examined daily until they were confluent. Contaminating fibroblasts were identified morphologically and removed mechanically (27). Keratinocyte phenotype was verified by cytokeratin immunostaining. H103, H157, H314, H357, H376, H400, H413 human keratinocyte cell lines derived from oral squamous cell carcinomas (OSCC) were also analysed (28). Clinical features of carcinoma lines are summarized in Table 1. The FC-7 Epstein Barr virus-transformed B cell line was included as a positive control.

Antibodies. All antibodies were monoclonal mouse anti-human and were diluted in PBS. For NOK phenotyping anti-pan cytokeratin clone MNF116 (Dako), which identifies cytokeratins 5, 6, 8 and 17, was used at 1:100 dilution. For MHC class II, CD40, CD80 and CD86, the following antibodies were used: anti- HLA class II (DP+DQ+DR) clone IQU9 (Novocastra) at a concentration of 1:200; anti-CD40 clone LOB7/6 (Serotec) at a concentration of 1:100; anti-CD80/B7-1 clone DAL-1 (Serotec) and L307.4 (BD Pharmingen) both at 1:20 dilution and finally anti- CD86/ B7-2 clone BU63 (Serotec) and clone IT2.2 (BD Pharmingen) diluted at 1:20.

Flow cytometry. When the keratinocytes were confluent, IFNγ was added to each flask and cells cultured in a humidified incubator in 5% CO₂/95% air at 37^o C for 36 hours. Cholera toxin was omitted from the culture medium to prevent competitive inhibition with IFNγ (29). Approximately 1x10⁶ cells/ml were incubated with primary antibodies diluted in PBS for one hour at room temperature with constant agitation. Negative controls included omission of the primary antibody, or incubation with a primary antibody of identical immunoglobulin (Ig) isotype. The positive control was the FC-7 cell line. After one hour the cells were incubated with the secondary goat anti-mouse R-phycoerythrin-conjugated F(ab’)₂ or rabbit anti-mouse fluorescein isothiocyanate-conjugated F(ab’)₂ Ig fragments (Dako) for 30 minutes. Cells were finally washed three times with fresh medium and analysed using a Becton Dickinson FACScan 420 flow cytometer running Cell Quest^TM## software. Background levels were calibrated before each experiment using the negative control samples. 1x10⁴ "events" were recorded for each sample. The data were collected for each sample as a histogram and dot plot, from which mean and median of fluorescence intensity and percentages of positive events were obtained using the same software.

Statistical analysis. Data were analyzed using the SPSS^® package, version 11.0 (SPSS Inc, Chicago, IL, USA). The difference in mean values between two groups was evaluated using Student’s t-test for independent samples and multiple comparisons by ANOVA (using the Bonferroni test). The statistically significant level was set at the 0.05 level (two-tailed).

RESULTS

Expression of MHC class II molecules in oral keratinocytes. Primary cultures of NOK and the H357 line did not express MHC class II constitutively. Very low percentages of positive MHC class II cells were observed on the rest of keratinocyte lines. By contrast, IFNγ significantly increased MHC class II expression in NOK and all oral keratinocyte lines (p≤ 0.01) when geometric means of fluorescence intensity and percentage of labelled cells were compared. When IFNγ-stimulated geometric means of fluorescence were compared, no statistically significant difference was found between NOK and cell lines, with the exception of H376, which showed the highest MHC class II fluorescence intensity (p<0.05). However, H376 and H103 lines showed the lowest relative percentage of labelled cells (62.5% and 55.85% respectively) (Fig. 1).

Expression of CD80 and CD86 on oral keratinocytes. All cell lines derived from OSCC showed virtually no expression of CD80 with or without IFNγ treatment, though low levels were detectable on NOK and H376 cells which increased slightly after stimulation with IFNγ. There were no statistically significant differences between these and the other lines, either when analysed for fluorescence intensity or relative percentages of positive cells (Fig. 2).

Marginally higher, though still minimal, levels of CD86 were detected on keratinocyte lines. However, a statistically significant increase (p=0.05) in CD86 expression was observed on NOK after stimulation with IFNγ (0.68% to 6.24%) (Fig. 3).

Expression of CD40 on oral keratinocytes. All cell lines constitutively expressed CD40 (Fig. 4). When geometric means of fluorescence intensity were compared, NOK expressed a similar level of CD40 as the H357 cell line. However, H357 showed a significantly higher percentage of CD40-postive cells (p=0.025). The other keratinocyte lines expressed constitutive levels of CD40 statistically significantly higher than NOK both in terms of fluorescence intensity and percentages of positive cells. IFNg caused a statistically significant increase in the fluorescence intensity CD40 expression on NOK and all keratinocyte lines except H413 (Fig. 4). However, percentages of positive cells were unchanged.

The results of these experiments are consistent with previous data, in that there was lack of constitutive MHC class II antigen in cultures of NOK (18, 30) and most keratinocyte lines (31, 32). The induction of MHC class II molecules on keratinocytes from normal oral mucosa and skin following exposure to IFNγ in vitro is well established (18,30,31,33-36) and was reproducible in this study. That the same levels of MHC class II expression were observed on stimulated NOK and stimulated keratinocyte lines is also in accordance with previous studies (18). Immunohistochemistry of squamous cell carcinomas of oral, skin, cervical and laryngeal epithelium (31,18,37) have corroborated these findings in vivo. Using an oral squamous cell carcinoma model, Matthews et al. (38) demonstrated that expression of MHC class II by oral keratinocytes was induced during the first stages of carcinogenesis.

The expression of CD40 observed in this study, and upregulation in intensity of expression after IFNγ stimulation, is also consistent with previous reports on oral (17,18) and skin keratinocytes (39) and a diverse range of carcinomas (40), and raises the question as to whether CD40 is expressed not only at a "resting" level, but also in an activated state.

Expression of CD80 has not been studied on normal oral mucosa in vivo or in vitro previously. Absence or very low expression of CD80 and CD86 on keratinocyte lines found in this study was consistent with previous studies of keratinocytes derived from head and neck carcinomas (15,17,18,20), but it is noteworthy that IFNγ induced minimal or no upregulation on either NOK or keratinocyte lines derived from OSCC. This absence may have biological significance. Although research in the early 1990’s suggested that keratinocytes induce T lymphocyte proliferation independent of costimulatory signals (41-43) later reports, such as that of Thomas et al. (25,44), associated the loss of CD80 by oral keratinocytes with the early development of OSCC, and considered it a marker for tumour progression and aggressiveness. Others support the view that establishment of CD80 and CD86 on tumour cells induces and/or increases T lymphocyte responses (25,44-47), and several workers have concluded that the absence of CD80 and CD86 on keratinocytes, oral and epidermal, is the major limitation to driving T lymphocyte activation (15,17,48). Thus the ability of some squamous cell carcinomas to evade the host’s defences may be because of the failure of keratinocytes to express CD80 and CD86.

An in vivo scenario can be proposed whereby an antigenic challenge to the oral epithelium is met by T lymphocytes which are activated by a "professional" APC, for example the local Langerhans cell population, as a result of which cytokines are released and oral keratinocytes, which are already CD40-positive, upregulate CD40 and start to express MHC class II. If CD86 and CD80 are also induced, an immune response is propagated. If keratinocytes fail to express CD86 and CD80, the malignant cells are able to evade host defences, grow and invade. Hence oral stratified squamous epithelium may act as an immunological organ not only as a physical barrier, but also though complex molecular and cellular mechanisms involving keratinocytes and leucocytes, both intra- and extra-epithelial.

BIBLIOGRAFÍA         [ Links ]

2. Yokochi T, Holly RD, Clark EA. B lymphoblast antigen (BB-1) expressed on Epstein-Barr virus-activated B cell blasts, B lymphoblastoid cell lines, and Burkitt’s lymphomas. J Immunol 1982;128:823-7.         [ Links ]

3. Azuma M, Ito D, Yagita H, Okumura K, Phillips JH, Lanier LL et al. B70 antigen is a second ligand for CTLA-4 and CD28. Nature 1993;366:76-9.         [ Links ]

4. Freeman GJ, Gribben JG, Boussiotis VA, Ng JW, Restivo VA, Jr., Lombard LA et al. Cloning of B7-2: a CTLA-4 counter-receptor that costimulates human T cell proliferation. Science 1993;262:909-11.         [ Links ]

5. Henry J, Miller MM, Pontarotti P. Structure and evolution of the extended B7 family. Immunol Today 1999;20:285-8.         [ Links ]

6. Symington FW, Brady W, Linsley PS. Expression and function of B7 on human epidermal Langerhans cells. J Immunol 1993;150:1286-95.         [ Links ]

7. Valle A, Aubry JP, Durand I, Banchereau J. IL-4 and IL-2 upregulate the expression of antigen B7, the B cell counterstructure to T cell CD28: an amplification mechanism for T-B cell interactions. Int Immunol 1991;3:229-35.         [ Links ]

8. Chambers CA. The expanding world of co-stimulation: the two-signal model revisited. Trends Immunol 2001;22:217-23.         [ Links ]

9. Schweitzer AN, Borriello F, Wong RC, Abbas AK, Sharpe AH. Role of costimulators in T cell differentiation: studies using antigen-presenting cells lacking expression of CD80 or CD86. J Immunol 1997;158:2713-22.         [ Links ]

10. Najafian N, Khoury SJ. T cell costimulatory pathways: blockade for autoimmunity. Expert Opin Biol Ther 2003;3:227-36.         [ Links ]

11. Vogel LA, Noelle RJ. CD40 and its crucial role as a member of the TNFR family. Semin Immunol 1998;10:435-42.         [ Links ]

12. Gaspari AA, Sempowski GD, Chess P, Gish J, Phipps RP. Human epidermal keratinocytes are induced to secrete interleukin-6 and co-stimulate T lymphocyte proliferation by a CD40-dependent mechanism. Eur J Immunol 1996;26:1371-7.         [ Links ]

13. Li J, Yang Y, Inoue H, Mori M, Akiyoshi T. The expression of costimulatory molecules CD80 and CD86 in human carcinoma cell lines: its regulation by interferon gamma and interleukin-10. Cancer Immunol Immunother 1996; 43:213-9.         [ Links ]

14. Tatsumi T, Takehara T, Katayama K, Mochizuki K, Yamamoto M, Kanto T et al. Expression of costimulatory molecules B7-1 (CD80) and B7-2 (CD86) on human hepatocellular carcinoma. Hepatology 1997;25:1108-14.         [ Links ]

15. Lang S, Whiteside TL, Lebeau A, Zeidler R, Mack B, Wollenberg B. Impairment of T-cell activation in head and neck cancer in situ and in vitro: strategies for an immune restoration. Arch Otolaryngol Head Neck Surg 1999;125:82-8.         [ Links ]

16. Lang S, Atarashi Y, Nishioka Y, Stanson J, Meidenbauer N, Whiteside TL. B7.1 on human carcinomas: costimulation of T cells and enhanced tumor-induced T-cell death. Cell Immunol 2000;201:132-43.         [ Links ]

17. Vonderheide RH, Butler MO, Liu JF, Battle TE, Hirano N, Gribben JG et al. CD40 activation of carcinoma cells increases expression of adhesion and major histocompatibility molecules but fails to induce either CD80/CD86 expression or T cell alloreactivity. Int J Oncol 2001;19:791-8.         [ Links ]

18. Farmer I, Freysdottir J, Dalghous AM, Fortune F. Expression of adhesion and activation molecules in human buccal epithelial cell lines and normal human buccal epithelium in situ. J Oral Pathol Med 2001;30:113-20.         [ Links ]

19. Ottaiano A, Pisano C, De Chiara A, Ascierto PA, Botti G, Barletta E et al. CD40 activation as potential tool in malignant neoplasms. Tumori 2002; 88:361-6.         [ Links ]

20. Posner MR, Cavacini LA, Upton MP, Tillman KC, Gornstein ER, Norris CM, Jr. Surface membrane-expressed CD40 is present on tumor cells from squamous cell cancer of the head and neck in vitro and in vivo and regulates cell growth in tumor cell lines. Clin Cancer Res 1999;5:2261-70.         [ Links ]

21. Loro LL, Ohlsson M, Vintermyr OK, Liavaag PG, Jonsson R, Johannessen AC. Maintained CD40 and loss of polarised CD40 ligand expression in oral squamous cell carcinoma. Anticancer Res 2001;21:113-7.         [ Links ]

22. Schwartz RH. Models of T cell anergy: is there a common molecular mechanism? J Exp Med 1996;184:1-8.         [ Links ]

23. Byrne SN, Halliday GM. High levels of Fas ligand and MHC class II in the absence of CD80 or CD86 expression and a decreased CD4+ T cell Infiltration, enables murine skin tumours to progress. Cancer Immunol Immunother 2003; 52:396-402         [ Links ]

24. Rice SQ, Crane IJ, Scully C, Prime SS. Production of a suppressor of lymphocyte proliferation by two human oral carcinoma cell lines. Scand J Immunol 1992;36:443-52.         [ Links ]

25. Thomas DW, Matthews JB, Patel V, Game SM, Prime SS. Inflammatory cell infiltrate associated with primary and transplanted tumours in an inbred model of oral carcinogenesis. J Oral Pathol Med 1995;24:23-31.         [ Links ]

26. Eck SC, Turka LA. Generation of protective immunity against an immunogenic carcinoma requires CD40/CD40L and B7/CD28 interactions but not CD4(+) T cells. Cancer Immunol Immunother 1999;48:336-41.         [ Links ]

27. Kedjarune U, Pongprerachok S, Arpornmaeklong P, Ungkusonmongkhon K. Culturing primary human gingival epithelial cells: comparison of two isolation techniques. J Craniomaxillofac Surg 2001;29:224-31.         [ Links ]

28. Prime SS, Nixon SV, Crane IJ, Stone A, Matthews JB, Maitland NJ et al. The behaviour of human oral squamous cell carcinoma in cell culture. J Pathol 1990;160:259-69.         [ Links ]

29. Friedman RM, Vogel SN. Interferons with special emphasis on the immune system. Adv Immunol 1983;34:97-140.         [ Links ]

30. Li J, Farthing PM, Thornhill MH. Cytokine regulation of major histocompatibility complex antigen expression by human oral and skin keratinocytes. Arch Oral Biol 1996;41:533-8.         [ Links ]

31. Crane IJ, Rice SQ, Luker J, de Gay L, Scully C, Prime SS. The expression of MHC antigens on cultured oral keratinocytes and relationship to malignancy. Br J Exp Pathol 1988;69:749-58.         [ Links ]

32. Mutlu S, Matthews JB, Midda M, Scully C, Prime SS. MHC antigen expression in human oral squamous carcinoma cell lines. J Pathol 1991;165:129-36.         [ Links ]

33. Basham TY, Nickoloff BJ, Merigan TC, Morhenn VB. Recombinant gamma interferon induces HLA-DR expression on cultured human keratinocytes. J Invest Dermatol 1984;83:88-90.         [ Links ]

34. Basham TY, Nickoloff BJ, Merigan TC, Morhenn VB. Recombinant gamma interferon differentially regulates class II antigen expression and biosynthesis on cultured normal human keratinocytes. J Interferon Res 1985;5:23-32.         [ Links ]

35. Volc-Platzer B, Leibl H, Luger T, Zahn G, Stingl G. Human epidermal cells synthesize HLA-DR alloantigens in vitro upon stimulation with gamma-interferon. J Invest Dermatol 1985;85:16-9.         [ Links ]

36. Savage NW, Walsh LJ, Seymour GJ. Expression of class I and class II major histocompatibility complex antigens on oral mucosal epithelium. J Oral Pathol 1987;16:153-7.         [ Links ]

37. Schwartz R, Momburg F, Moldenhauer G, Dorken B, Schirrmacher V. Induction of HLA class-II antigen expression on human carcinoma cell lines by IFN-Gamma. Int J Cancer 1985;35:245-50.         [ Links ]

38. Matthews JB, Pitigala-Arachchi A, Crane IJ, Scully C, Prime SS. The relationship between epithelial Ia expression and the inflammatory cell infiltrate during experimental oral carcinogenesis. Virchows Arch A Pathol Anat Histopathol 1988;413:521-8.         [ Links ]

39. Peguet-Navarro J, Dalbiez-Gauthier C, Moulon C, Berthier O, Reano A, Gaucherand M et al. CD40 ligation of human keratinocytes inhibits their proliferation and induces their differentiation. J Immunol 1997;158:144-52.         [ Links ]

40. Young LS, Eliopoulos AG, Gallagher NJ, Dawson CW. CD40 and epithelial cells: across the great divide. Immunol Today 1998;19:502-6.        [ Links ]

41. Mutis T, de Bueger M, Bakker A, Ottenhoff TH. HLA class II+ human keratinocytes present Mycobacterium leprae antigens to CD4+ Th1-like cells. Scand J Immunol 1993;37:43-51.         [ Links ]

42. de Bueger M, Bakker A, Goulmy E. Human keratinocytes activate primed major and minor histocompatibility antigen specific Th cells in vitro. Transpl Immunol 1993;1:52-9.         [ Links ]

43. Pawelec G, Kalbacher H, Pohla H, Boshell M, Max H, Friccius H et al. Requirements for stimulation or anergy induction in alloreactive human T cell clones. Cell Immunol 1994;158:241-52.         [ Links ]

44. Thomas GR, Chen Z, Enamorado I, Bancroft C, Van Waes C. IL-12- and IL-2-induced tumor regression in a new murine model of oral squamous-cell carcinoma is promoted by expression of the CD80 co-stimulatory molecule and interferon-gamma. Int J Cancer 2000;86:368-74.         [ Links ]

45. Yang G, Hellstrom KE, Mizuno MT, Chen L. In vitro priming of tumor-reactive cytolytic T lymphocytes by combining IL-10 with B7-CD28 costimulation. J Immunol 1995; 155(8):3897-903.         [ Links ]

46. Baskar S, Glimcher L, Nabavi N, Jones RT, Ostrand-Rosenberg S. Major histocompatibility complex class II+B7-1+ tumor cells are potent vaccines for stimulating tumor rejection in tumor-bearing mice. J Exp Med 1995;181:619-29.         [ Links ]

47. Wang YC, Zhu L, McHugh R, Graham SD, Jr., Hillyer CD, Dillehay D et al. Induction of autologous tumor-specific cytotoxic T-lymphocyte activity against a human renal carcinoma cell line by B7-1 (CD8O) costimulation. J Immunother Emphasis Tumor Immunol 1996;19:1-8.         [ Links ]

48. Grousson J, Concha M, Schmitt D, Peguet-Navarro J. Effects of CD40 ligation on human keratinocyte accessory function. Arch Dermatol Res 1998; 290:325-30.        [ Links ]