Immunobiol., vol. 176, pp. 423-431 (1988)

Endocrine Department of the Medical Clinic, Klinikum Steglitz, Free University of Berlin, Germany

Interferon-y Protects Human Thyroid Epithelial Cells against Cell-Mediated Cytotoxicity*

ULRICH BOGNER, BARBARA SIGLE, and HORST SCHLEUSENER

Received August 13, 1987· Accepted in Revised Form October 21,1987

Abstract

Interferons have been shown to have antagonistic effects on cell-mediated cytotoxicity. In the present study, we investigated various types of immunologically mediated cytotoxic reactions with a 51 chromium release assay, using human thyroid cells as targets, which were cultured in the presence and absence of interferon-gamma (IFN-y). Antibody-dependent cell­mediated cytotoxicity (ADCC), natural killer (NK) cell-mediated cytotoxicity, and autolo­gous lymphocyte-mediated cytotoxicity (ALC) were measured utilizing thyroid tissue, sera and lymphocytes from patients with autoimmune thyroid disease as well as from normal subjects. ADCC, determined with sera from patients with Hashimoto's thyroiditis, was reduced by 70 % when the thyroid target cells were cultured in the presence of 1000 U human IFN-y/ml culture medium. Similarly, natural killer cell-mediated cytotoxicity, as well as autologous lymphocyte-mediated cytotoxicity, was significantly reduced after pretreatment of thyroid target cells with 500-1000 U IFN-y/ml culture medium. Although it is known that IFN-y renders target cells resistant to NK cell-mediated lysis, this is the first report on this effect using human epithelial cells, which may have major implications for the suggested role of IFN-y in the induction and perpetuation of autoimmune thyroid disease.

Introduction

IFN-y is mainly produced by T cells and plays an important role in the

modulation of the immune reactions after antigenic stimulation. The

biological effects include the activation of macrophages, NK, K and

cytotoxic T cells, the enhancement of HLA class I expression, as well as the

induction of the HLA class II molecules on the surface of epithelial cells

(1-7). Recently, in autoimmune thyroid disease, the aberrant expression of

HLA-DR on the surface of human thyroid cells was demonstrated and was

thought to be a crucial step in the initiation and perpetuation of the diseases

" This work was supported by the Deutsche Forschungsgemeinschaft

Abbreviations: NK = natural killer; ADCC = antibody-dependent cell-mediated cytotox­icity; ALC = autologous lymphocyte-mediated cytotoxicity.

424 . ULRICH BOGNER, BARBARA SIGLE, and HORST SCHLEUSENER

(8, 9). This hypothesis was further supported by the finding that cultured thyroid cells can be induced to express HLA-DR molecules on their surface after IFN-y treatment (10, 11).

In contrast to the augmentation of the cytotoxic effector mechanisms by IFN-y, a reduction of cytotoxicity was reported after preincubation of various types of target cells, such as lymphoblastoid cell lines, normal and tumorous mouse cells, carcinoma cells and human fibroblasts, with IFN-y (12-17). In the present study, we investigated antibody-dependent cell­mediated cytotoxicity, NK cell and autologous lymphocyte-mediated cytotoxicity against human thyroid cells, cultured in the presence and absence of IFN-y. We found a marked reduction of cell-mediated lysis, which is to our knowledge the first report on induced resistance by IFN-y against cell-mediated cytotoxicity using human epithelial cells as targets.

Materials and Methods

Human thyroid cells

Human thyroid tissue, obtained by surgery, was minced and digested by collagenase. Four x 106 cells were cultured in Iscove's medium (Seromed, Berlin, F.R.G.) in the absence and presence of various concentrations (5-1000 U/ml medium) of recombinant human IFN-y (Sigma Chemicals, St. Louis, MO, U.S.A.) in a water-saturated incubator for 3 days at 37°C in 5 % CO2 in air. The medium was supplemented with 10 % FCS, 100 U/ml penicillin, 100 ~g/ ml streptomycin (Seromed), 10 ~g/ml insulin, 10 nmol/ml hydrocortisone, 5 ~g/ml transferrin, 10 ng/ml somatostatin (Sigma), and 10 mUlml bTSH (Organon Technika, Munich, F.R.G.).

Cytotoxicity assay

Cytotoxicity assays were performed as previously described (18). Two x 10· thyroid cells were radiolabelled with 100 ~Ci Na2 SlCr04 (Amersham, Braunschweig, F.R.G.) for 1 h at 3rC in 5 % CO2 in air, washed and adjusted to 1 x 105 cells/m!. In ADCC assay, 5 x 103

radio labelled thyroid cells were preincubated with 1 00 ~l heat-inactivated serum. After 1 h, the cells were washed, and 1.25 X 105 effector lymphocytes were added, yielding an effector:target cell ratio (E:T ratio) of 25:1. After 18 h, an aliquot of the supernatant was aspirated (cpmexp) and counted in a gamma counter. Samples for the 100 % value (cpmm,,) and unspecific lysis (cpmump,,), contained medium instead of lymphocytes. The 100 % value was determined by counting a 100-~1 aliquot of resuspended target cells. The NK assay was performed with the same procedure as the ADCC assay, but without serum at E:T cell ratios ranging from 50:1 to 12.5:1. The incubation period was 4 h. In the autologous lymphocyte-mediated cytotoxicity assay, thyroid cells from patients with Graves' disease were cultured. On day 3, the peripheral lymphocytes from the corresponding patient were collected and used as effector cells in the cytotoxicity assay. The cytotoxicity assay was performed in the same manner as the NK assay. The cytotoxicity was expressed as follows:

0/ ·f· I· cpmexp-cpmump" x 100. /0 specl IC YSIS =

cpmmax - cpmunspec

Effector ceJJs

Mononuclear lymphocytes were separated from heparinized venous blood by Ficoll density centrifugation (Seromed). Lymphocytes were washed three times with PBS, suspended in

IFN-y-Induced Protection against Cytotoxicity· 425

Iscove's medium and adjusted to 2.5, 1.25, and 0.625 X 106/ml, which yielded E:T cell ratios of 50:1,25:1, and 12.5:1, respectively.

Clinical subjects

Sera from patients with Hashimoto's thyroiditis (n = 8) were tested in ADCC. The diagnosis was made from the clinical history of hypothyroidism, elevated TSH and positivity of microsomal and thyroglobulin antibodies. Thyroid tissue and lymphocytes from 4 patients with Graves' disease (n = 4) were collected. Graves' disease was diagnosed from the clinical history and biochemical proof of hyperthyroidism, diffuse uptake of the radiolabel in the scintiscan, and positivity of TSH-receptor-antibodies. Normal peripheral lymphocytes were collected from volunteers without history of thyroid diseases.

Statistical analysis

Statistics were assessed by Student's t-test.

Results

We have reported previously that ADCC activity is increased in patients with Hashimoto's thyroiditis (18, 19). In the experiments presented here, we measured ADCC in 8 sera from patients with Hashimoto's thyroiditis and found significant reduction of cytotoxicity using thyroid cells precul­tured with IFN-y for 3 days (Fig. la). The mean ± SD specific lysis was 24.1 ± 8.6 % in the absence and 8.3 ± 8.1 % in the presence of 1000 U

a % specific lysis

40

35

30

25

20

15

10

5

o o 1000

g-interferon (U/ml)

b % reduction of cytotoxicity

100

80

60

40

20

o o 1000

g-interferon (U/ml)

Figure 1. a) Percent specific lysis in ADCC assay with sera from patients with Hashimoto's thyroiditis (n = 8) using human thyroid cells as targets. Preculturing of thyroid cells with 1000 U human IFN-y/ml medium for 3 days resulted in a significant reduction of cytotoxicity; b) reduction of % specific lysis, expressed as % reduction of cytotoxicity from all sera tested in ADCC assay.

426 . ULRICH BOGNER, BARBARA SIGLE, and HORST SCHLEUSENER

a % specific lysis

30

25

20

15

10

5

o

OE:Tr.tio 50 :1 0" " 25,1 11" .. 12.5: 1

I I o 500

g-interferon (U Iml)

b % reduction of cytotoxic ity

100

80

60

40

20

o o 500

g-interferon (U/ml)

Figure 2. a) Percent specific lysis in NK cell-mediated cytotoxicity at different E:T cell ratios against human thyroid target cells shows a significant reduction after preculture of the target cells with 500 U IFN-y/ml culture medium for 3 days. Lymphocytes from normal subjects were used as effector cells; b) reduction of % specific lysis, expressed as % reduction of cytotoxicity from all values in Figure 2a. The mean reduction of cytotoxicity was 63 % when targets were pre cultured with IFN-y.

% specific lysis

25

20

15

10

5

o I o

I 5

o E:T ratio = 50: 1 "E:T ratio = 25:1

I 50 500

g-interferon (U/ml)

Figure 3. NK cell-mediated cytotoxiCity using target cells pre incubated with increasing concentrations of IFN-y/ml medium. A decrease of cytotoxicity is found with a minimal dosis of 50 U IFN-y/ml culture medium. Peripheral lymphocytes from normal subjects were used as effector cells.

IFN-y-Induced Protection against Cytotoxicity· 427

% spec ific lysis

30

25

20

15

10

~ 5

0

o 500 1000 g-interferon (U/ml)

Figure 4. Autologous lymphocyte-mediated cytotoxicity with peripheral lymphocytes from patients with Graves' disease against their own thyroid target cells, obtained by surgery, and cultured for 3 days in the presence of 0,500 and 1000 U IFN-y/ml medium. A dose-dependent reduction of cytotoxicity was found with increasing concentrations of IFN-y. The E:T cell ratio was 50:1.

IFN-y/ml culture medium, respectively (p < 0.0025). Expressing the results as % reduction of cytotoxicity, the mean lysis is reduced by 70 % (Fig. 1 b).

In NK cell-mediated cytotoxicity, a similar decrease of cytotoxicity was found after IFN-y pretreatment of the target cells at all E:T cell ratios tested (Fig. 2a). The mean±SD specific lysis was 19.4±4.5, 17.6±3 and 14.8 ± 0.6 % in the absence of IFN-y, and 7.8 ± 5.7,8.7 ± 3 and 8.1 ± 2.5 % in the presence of 500 V IFN-y/ml incubation medium at E:T cell ratios of 50:1,25:1, and 12.5:1, respectively (p < 0.0005). The mean % reduction of cytotoxicity was 63 % (Fig. 2b). In a dose-dependent experiment, the minimal concentration of IFN-y effective in the induction of resistance to NK cell-mediated lysis was found to be 50 Vlml culture medium (Fig. 3).

Autologous lymphocyte-mediated cytotoxicity measured in patients with Graves' disease showed a similar pattern in reduction of cell-mediated cytotoxicity with a dose-dependent decrease of lysis with increasing con­centrations of IFN -y added to the culture medium of the target cells (Fig. 4). The specific lysis (mean ± SD) was 21.5 ± 8.2, 15.6 ± 5.1 and 10.7 ± 4.3 % at E:T cell ratios of 50:1,25:1 and 12.5:1, respectively, in the absence of IFN-y; 11.8 ± 6.8, 8.4 ± 6.1 and 6.5 ± 6.3 in the presence of 500 V (p < 0.01), and 3.5 ± 5, 3.7 ± 5.2 and 4.1 ± 5.8 % in the presence of 1000 V IFN-y/ml culture medium, respectively (p < 0.0025).

428 . ULRICH BOGNER, BARBARA SIGLE, and HORST SCHLEUSENER

Discussion

IFN-y renders various types of target cells (i.e., lymphoblastoid cell lines, normal and tumor mouse cells, carcinoma cells and fibroblasts) resistant to cell-mediated lysis (12-17). This phenomenon is dose-dependent and influences almost all cell types, although the amount of this protective effect differs from cell line to cell line (13). Our data indicate that this phenome­non also occurs with human epithelial cells, demonstrating reduced cytotoxicity by various types of immunologically mediated cytotoxic mechanisms after pretreatment of human thyroid target cells with IFN-y.

The mechanism by which IFN-y induces an inhibitory activity on the susceptibility of target cells to cell-mediated lysis is not completely under­stood. Morphological, physical and biochemical alterations of the plasma membrane are observed, e.g., an increase in the density, glycoprotein content and intramembranous particles (20). The binding of the NK cells to their receptors was not influenced, but the described alterations of the membrane seem to mitigate the proper target-binding interaction which results in an inhibition of the lytic hit (13,15). Very recently, STORKUS et al. and HAREL-BELLAN et al. demonstrated an inverse correlation between NK-susceptibility of target cells and their HLA class I - but not class 11-expression, suggesting an interaction between class I antigens and NK target structures on the surface (21, 22). Furthermore, induction of HLA class I antigens by IFN-y and addition of monoclonal antibodies to the class I determinants did not preserve the resistance to lysis, whereas reduced NK-susceptibility could be abolished by addition of actinomycin D, cyclo­heximide or antibodies to IFN-y to the culture medium (13). This indicates that induction of resistance to cell-mediated lysis can be inhibited by substances which suppress the alteration of the plasma membrane on the level of the RNA or protein synthesis.

Another potent factor known to suppress NK cell activity is prostaglan­din E2 (23,24). Its generation could be shown in co-culture experiments of thyroid cells with blood mononuclear cells, but not by thyroid cells alone (25). Although we cannot completely rule out a prostaglandin E2 generation by contaminating monocytes in the thyroid cell culture, this should not influence the results, since the NK effector cells in the cytotoxicity assay are from a different source. A further argument against a possible influence of prostaglandin E2 in our test system is the recent finding that IFN-y does antagonize the prostaglandin E2 release from stimulated monocytes (26).

Although an IFN-y-induced reduction of NK cell-mediated cytotoxicity has been described in most studies, the results concerning ADCC are not unanimous. WELSH et al. reported about ADCC being reduced to a lesser extent than in NK cell activity, whereas TRINCHIERI et al. could not find any influence on ADCC after incubation of the target cells with IFN-y (13, 15). NK and K cells appear to reside within the same subpopulation of peripheral lymphocytes - the large granular lymphocytes - and it is

IFN-y-Induced Protection against Cytotoxicity' 429

suggested that NK and K cells are identical, mediating both types of cytotoxic events (27, 28). On the basis of this assumption, it is plausible that IFN-y influences NK and K cell-mediated cytotoxicity in a similar manner; reduction of cell-mediated lysis when the target cells were precultured with IFN-y, and augmentation of NK and ADCC activity when the lympho­cytes were incubated with IFN-y, as could be shown by others (2, 5, 7). Furthermore, there is evidence in ADCC that the alteration of the plasma membrane - i.e., possibly induced by IFN-y - hinders the spatial orienta­tion of the target-cell-bound antibodies, which is necessary for the binding of the K cell eliciting an optimal cytotoxic reaction (29).

T cell-mediated cytotoxicity is MHC restricted and requires the recogni­tion of HLA class I antigens on the target cells by cytotoxic T cells. Cytotoxic T cell-mediated cell lysis was reported to be increased after culturing target cells with IFN-y, known to be a potent stimulator of HLA class I expression (6, 13). Although we measured cytotoxicity in an autolo­gous system, we utilized mononuclear lymphocytes after Ficoll density centrifugation, which include NK, K, and cytotoxic T cells as well. The observed effect of IFN-y-induced reduction of cytotoxicity in the autolo­gous system is probably not cytotoxic T cell-dependent, but rather an additive effect of various types of effector cells residing in the mononuclear fraction.

The significance of the described in vitro effects of IFN-y-induced resistance of target cells to cell-mediated lysis is unclear with respect to the in vivo immune reactions in autoimmune thyroid disease. Although the effect is dose-dependent, as presented in this paper, the tissue levels of IFN-y in the affected thyroid gland are unknown. Further studies have to clarify the antagonistic effects of IFN-y in the modulation of the immune mechanisms in autoimmune thyroid diseases.

Acknowledgements

The authors wish to thank Mrs. SCHNEIDER for her excellent technical assistance.

References

1. BLOOM, B. R. 1980. Interferons and the immune system. Nature 284: 593. 2. BASHAM, T. Y., W. K. SMITH, and T. C. MERIGAN. 1984. Interferon enhances antibody­

dependent cellular cytotoxicity when suboptimal concentrations of antibody are used. Cell. Immunol. 88: 393.

3. FAUCI, A. S. 1987. Immunomodulators in clinical medicine. Ann. Int. Med. 106: 421. 4. HERON, 1., M. HOKLAND, and K. BERG. 1978. Enhanced expression of Brmicroglobulin

and HLA antigens on human lymphoid cells by interferon. Proc. Nat!. Acad. Sci. USA 75: 6215.

5. HERBERMAN, R. B., J. R. ORTALDO, and G. D. BONNARD. 1979. Augmentation of interferon of human natural and antibody-dependent cell-mediated cytotoxicity. Nature 277: 221.

6. BUKOWSKI, J. F., and R. M. WELSH. 1985. Interferon enhances the susceptibility of virus­infected fibroblasts to cytotoxic T cells. J. Exp. Med. 161: 257.

430 . ULRICH BOGNER, BARBARA SIGLE, and HORST SCHLEUSENER

7. SMALLEY, R. V., and E. C. BORDEN. 1986. Interferons: current status and future directions of this prototypic biological. Springer Semin. Immunopathol. 9: 73.

8. BOTTAZZO, G. F., R. PUJoL-BoRRELL, T. HANAFUSA, and M. FELDMANN. 1983. Role of aberrant HLA-DR expression and antigen presentation in induction of endocrine auto­immunity. Lancet II: 1115.

9. HANAFUSA, T., L. CHIOVATO, D. DONIACH, R. PUJOL-BoRRELL, R. C. G. RUSSELL, and G. F. BOTTAZZO. 1983. Aberrant expression of HLA-DR antigen on thyrocytes in Graves' disease: Relevance for autoimmunity. Lancet II: 1111.

10. TODD, I., R. PUJOL-BORRELL, L. J. HAMMOND, G. F. BOTTAZZO, and M. FELDMANN. 1985. Interferon-gamma induces HLA-DR expression by thyroid epithelium. Clin. Exp. Immunol. 61: 265.

11. WEETMAN, A. P., D. J. VOLKMANN, K. D. BURMAN, T. L. GERRALD, and A. S. FAUCI. 1985. The in vitro regulation of human thyrocyte HLA-DR antigen expression. J. Clin. Endocrinol. Metab. 61: 817.

12. TRINCHIERI, G. and D. SANTOLI. 1978. Anti-viral activity induced by culturing lympho­cytes with tumor-derived or virus-transformed cells: enhancement of human natural killer cell activity by interferon and antagonistic inhibition of susceptibility of target cells to lysis. J. Exp. Med. 147: 1314.

13. WELSH, R. M., K. KARRE, M. HANSSON, L. A. KUNKEL, and R. W. KIESSLING. 1981. Interferon-mediated protection of normal and tumor target cells against lysis by mouse natural killer cells. J. Immunol. 126: 219.

14. MOORE, M., W. J. WHITE, and M. R. POTTER. 1980. Modulation of target cell susceptibil­ity to human natural killer cells by interferon. Int. J. Cancer 25: 565.

15. TRINCHIERI, G., D. GRANATO, and B. PERUSSIA. 1981. Interferon-induced resistance of fibroblasts to cytolysis mediated by natural killer cells: specificity and mechanism. J. Immunol. 126: 335.

16. WALLACH, D. 1983. Interferon-induced resistance to the killing by NK cells: a preferential effect of IFN-gamma. Cell. Immunol. 75: 390.

17. SARZOTTI, M., S. BARON, S. K. TYRING, and G. R. KLIMPEL. 1986. Interferon-mediated protection of B16 melanoma cells from cytotoxicity by activated macrophages. Cell. Immunol. 100: 280.

18. BOGNER, U., H. SCHLEUSENER, and J. R. WALL. 1984. Antibody-dependent cell-mediated cytotoxicity against human thyroid cells in Hashimoto's thyroiditis but not Graves' disease. J. Clin. Endocrinol. Metab. 59: 734.

19. BOGNER, U., J. R. WALL, and H. SCHLEUSENER. 1987. Cellular and antibody-mediated cytotoxicity in autoimmune thyroid disease. Acta Endocrinol. Suppl. 281: 133.

20. CHANG, E. H., F. T. JAY, and R. M. FRIEDMAN. 1978. Physical, morphological, and biochemical alterations in the membrane of AKR mouse cells after interferon treatment. Proc. Natl. Acad. Sci. USA 75: 1859.

21. STORKUS, W. J., D. N. HOWELL, R. D. SALTER, J. R. DAWSON, and P. CRESSWELL. 1987. NK susceptibility varies inversely with target cell class I HLA antigen expression. J. Immunol. 138: 1657.

22. HAREL-BELLAN, A., A. QUILLET, C. MARCHIOL, R. DEMARS, T. TURSZ, and D. FRADELIZI. 1986. Natural killer susceptibility of human cells may be regulated by genes in the HLA region on chromosome 6. Proc. Natl. Acad. Sci. 83: 5688.

23. BRUNDA, M. J., R. B. HERBERMAN, and H. T. HOLDEN. 1980. Inhibition of murine natural killer cell activity by prostaglandins. J. Immunol. 124: 2682.

24. RAMSTEDT, U., C. N. SERHAN, K. C. NICOLAOU, S. E. WEBBER, H. WIGZELL, and B. SAMUELSSON. 1987. Lipoxin A-induced inhibition of human natural killer cell cytotox­icity: studies on stereospecificity of inhibition and mode of action. J. Immunol. 138: 266.

25. YAMAMOTO, M., E. A. HERMAN, and B. RAPOPORT. 1979. Prostaglandin generation in vitro by peripheral blood mononuclear cells in co-culture with human thyroid cells. J. BioI. Chern. 254: 4046.

26. BROWNING, J. L. and A. RIBOLINI. 1987. Interferon blocks interleukin I-induced prosta­glandin release from human peripheral monocytes. J. Immunol. 138: 2857.

IFN-y-Induced Protection against Cytotoxicity· 431

27. HERBERMAN, R. B. 1986. Natural killer cell activity and antibody-dependent cell-mediated cytotoxicity. Manual of Clinical Laboratory Immunology, 3rd ed. American Society for Microbiology, Washington D.C., 308 pp.

28. TIMONEN, T., J. R. ORTALDO, and R. HERBERMAN. 1981. Characteristics of human large granular lymphocytes and relationship to natural killer and K cells. J. Exp. Med. 153: 569.

29. CHRISTIAANSEN, J. E., S. S. BURNSIDE, and D. W. SEARS. 1987. Apparent sensitivity of human K lymphocytes to the spatial orientation and organization of target cell-bound antibodies as measured by the efficiency of antibody-dependent cellular cytotoxicity (ADCC). J. Immunol. 138: 2236.

Dr. ULRICH BOGNER, Klinikum Steglitz, Medizinische Klinik, Hindenburgdamm 30, D-1000 Berlin 45, Germany