Methotrexate regulates ICAM-1 expression in recipients of rat cardiac allografts.

The means by which methotrexate (MTX) mediates immunosuppression at low doses remains to be elucidated. MTX has been shown to inhibit the adherence of neutrophils and fibroblasts to endothelial cells in vitro. The hypothesis that MTX treatment may affect cellular adherence by downregulating cell adhesion molecule expression formed the rationale for these studies. Previous studies of rat cardiac transplant recipients in our laboratory demonstrated that low-dose MTX treatment alone significantly inhibits the expression of the leucocyte beta 2 integrin subunit, CD18. These investigations have addressed whether low-dose MTX treatment might also affect the expression of the beta-integrin counter-receptor, ICAM-1, a cell adhesion molecule which may be induced on endothelial cells during an immune response. The degree to which low-dose cyclosporine A and low-dose MTX treatment alone, and in combination, impact cell adhesion molecule expression has been studied in Brown Norway (BN) to Lewis (Lew) rat accessory cervical heart allografts. According to both Northern blot and immunohistochemical analysis, ICAM-1 expression was upregulated in graft regional lymph nodes and in the spleen of untreated cardiac allograft recipients within 6 h post-transplantation. Despite induction of VCAM-1 expression, ICAM-1 expression remained low or undetectable in cardiac allograft tissue as measured both by reverse transcriptase-polymerase chain reaction (RT-PCR) and immunohistochemical analysis. These data suggest that ICAM-1 may function in leucocyte trafficking through lymphoid organs, such as the lymph nodes and spleen, but not directly in graft leucocyte recruitment during BN to Lew rat cardiac allograft rejection. Despite prolonged allograft survival with cyclosporine A alone and combination cyclosporine A/MTX, these treatments did not result in diminished steady-state ICAM-1 mRNA levels in regional lymph nodes or spleen of cardiac allograft recipients. MTX treatment alone, however, substantially diminished ICAM-1 expression in allograft recipient lymphoid tissues. These studies demonstrate for the first time in vivo using a rat model of acute allograft rejection that MTX but not cyclosporine treatment downregulates cell adhesion molecule expression. Low-dose MTX treatment alone, however, is not sufficient to result in prolonged BN to Lew rat cardiac allograft survival. The means by which combination low-dose cyclosporine A and MTX treatment results in prolonged rat cardiac allograft survival over low-dose cyclosporine treatment alone remain(s) to be clarified.

Effects of cyclosporine A and methotrexate on CD18 expression in recipients of rat cardiac allografts.

Recent advances in the study of the molecular basis of inflammation suggest that cell-cell interactions mediated by specific adhesion molecules could be new targets for immunosuppression. Methotrexate (MTX)-treated cells in vitro have demonstrated decreased neutrophil-endothelial cell adhesion associated with increased release of adenosine from endothelial cells, while the direct role of cyclosporine A (CSA) in the regulation of cell adhesion molecule (CAM) expression is less well-defined. Since the adhesion of leucocytes to endothelial cells via CAMs is necessary for leucocyte extravasation and infiltration into graft tissue during allograft rejection, these studies have addressed the hypothesis that MTX treatment of cardiac transplant recipients may affect cellular adherence by downregulating cell adhesion molecule expression. Using a vascularized method of rat cardiac transplantation, our studies have previously demonstrated that low doses of the immunosuppressive agents CSA and MTX, when used in combination, significantly increase allograft survival. According to reverse transcriptase-polymerase chain reaction (RT-PCR) methodology to measure changes in steady-state CD18 mRNA levels, and immunohistochemistry to assess transplant CD18 protein levels in situ, both CD18 transcript and protein levels were significantly increased in untreated allografts when compared to isograft tissues on days 3 through to 7 post-transplant. Whereas, both low-dose CSA alone and low-dose MTX alone treatment resulted in similar levels of graft leucocyte infiltration, MTX-treated recipients demonstrated lower levels of CD18 expression when compared to low-dose CSA alone treatment. The results of immunohistochemical staining for T cells, where significantly fewer T cells were observed in rat cardiac allografts after low-dose MTX treatment alone compared to low-dose CSA treatment, were noteworthy. Results of these studies indicate that CD18 expression and infiltrating T cell numbers in Brown Norway (BN) to Lewis (Lew) rat cardiac allografts are significantly diminished with low-dose MTX treatment. The immunosuppressive effects of MTX, therefore, may be related to its ability to interfere with an early step during the cell-mediated immune response, namely the firm binding or 'adhesion' of leucocytes to the endothelium during transendothelial migration.

Effects of cyclosporine A and methotrexate on induction of tumour necrosis factor-alpha in rat cardiac allografts.

Experimental and clinical studies have yet to determine the extent to which methotrexate (MTX) or cyclosporine A (CSA) treatment alone affects the expression in vivo of tumour necrosis factor-alpha (TNF alpha), a cytokine produced primarily by macrophages and believed to be directly involved in the pathogenesis of cardiac allograft rejection. In light of previously published findings from this laboratory examining the effects of combination CSA/MTX treatment, these studies were designed to examine the individual effects of CSA and MTX upon TNF alpha gene expression post-transplant (post-tx) using an accessory cervical heart transplant model in the rat. These studies have focused on a highly sensitive method with which to detect changes in gene expression, reverse transcriptase-polymerase chain reaction (RT-PCR) methodology and enzyme-linked immunosorbent assay (ELISA) assessment of transplant TNF alpha protein levels. Both techniques consistently demonstrated biphasic TNF alpha expression in cardiac transplant tissue obtained from untreated allograft recipients during the first week post-tx in contrast to isograft recipients. Previously demonstrated in combination to prolong cardiac allograft survival, low-dose MTX and low-dose CSA were each evaluated alone in the course of these studies to determine their impact on TNF alpha gene expression. While TNF alpha levels were up-regulated during untreated allograft rejection, both TNF alpha RNA and protein were significantly diminished with low-dose combination CSA/MTX treatment, with CSA alone, but not significantly with MTX treatment alone. In conclusion, TNF alpha gene expression in untreated allografts is consistent with the hypothesis that TNF alpha may play a role in events leading to allograft rejection. Results of these studies indicate that TNF alpha levels are significantly regulated in vivo by CSA but not by MTX treatment. These studies further implicate a role for low-dose MTX in mediating statistically significant immunosuppressive effects in conjunction with low-dose CSA.

Diminished cytotoxic gene expression in rat cardiac transplants with low-dose cyclosporine/methotrexate combination therapy.

We have previously shown that administration of a combination low-dose cyclosporine (CsA) (1.0 mg/kg/day)/methotrexate (MTX) (450 micrograms/kg/wk) treatment significantly increases the survival of rat cardiac allografts, and may therefore potentially serve as an alternative immunosuppressive therapy designed to promote transplant survival while minimizing high-dose CsA side effects. In contrast to high-dose CsA, low-dose CsA/MTX treatment does not appear to alter IL-2 gene expression, since similar patterns of IL-2 gene transcripts were found in both low-dose CsA/MTX-treated and untreated control allografts on days 1 through 8 posttransplantation (post-tx). The mechanism(s) by which low dose CsA/MTX therapy increases the time of allograft survival remains to be elucidated. The aim of the present study was to determine the effects of low-dose CsA/MTX on the expression of the cytotoxic cytokines, TNF alpha, TNF beta, or lymphotoxin (LT), and the serine proteases HF and C11 (granzymes A and B, respectively) in rat cardiac allografts during rejection. RNA blot analysis showed significant suppression of TNF alpha, LT, HF, and C11 gene expression on days 1 through 8 post-tx in cardiac allografts from low-dose CsA/MTX-treated recipients compared with untreated allograft controls. TNF protein levels in cardiac allografts from low-dose CsA/MTX-treated recipients were also found to be significantly reduced on days 1 through 8 post-tx when compared with time-matched untreated allograft controls (P < or = 0.001). We conclude that low-dose CsA/MTX treatment, while effective in prolonging cardiac transplant survival, appears to act at the mRNA level to downregulate cytotoxic cytokine gene expression. Such trials aimed at evaluating low-dose combination therapy may afford new insight into mechanisms underlying improvement in immunosuppressive treatment.

Induction of TNF alpha and TNF beta gene expression in rat cardiac transplants during allograft rejection.

The expression of the cytotoxic cytokines tumor necrosis factor alpha and TNF beta or lymphotoxin (LT) was assessed in rat cardiac transplants during rejection. Newborn rat cardiac grafts placed in adult rat ear pinnae were retrieved on days 1 through 10 posttransplantation; the average time to rejection, assessed by the absence of detectable electrocardiographic activity, was determined to be 7 days. Total cellular RNA and tissue homogenates were prepared from cardiac transplants in order that relative levels of TNF alpha and LT mRNA and TNF protein could be determined. A biphasic pattern of TNF alpha gene expression was consistently seen in cardiac allografts. TNF alpha mRNA transcripts were detected as early as day 2 post-tx, with peak levels appearing on day 3 post-tx. Although transcript levels decreased by day 4, a significant increase appeared again on day 6 post-tx, coincident with the onset of rejection. Similar to TNF alpha gene expression, LT transcripts demonstrated a biphasic pattern of induction. LT mRNA transcripts also reached peak levels on day 3 post-tx, with a second increase in transcript levels coincident with rejection. TNF protein levels in allografts displayed a biphasic pattern, similar to that shown by the cytokine mRNAs. Peak levels of TNF protein were detected on day 3 post-tx, with a second increase again coinciding with rejection. In contrast to TNF expression found in allografts, TNF alpha and LT mRNA transcripts were not detected in isografts on days 1 through 10 post-tx. TNF protein levels in cardiac isografts were consistently at or below the standard limits of detection, and on days 3 through 7 post-tx were significantly reduced (P < or = 0.001) when compared with time-matched allografts. Increased expression of the cytotoxic cytokines TNF alpha and LT, therefore, appears to be allograft-specific and is an early event during rat cardiac allograft rejection. In conclusion, induction of TNF gene expression may be an important early indicator of transplant rejection.

Use of low-dose cyclosporine A/methotrexate to prolong rat cardiac allograft survival.

Cyclosporine A (CSA) is the standard immunosuppressive agent used in human cardiac transplantation to prevent rejection; however, adverse side effects have been reported at therapeutic doses. Therefore, a need remains for the implementation of specific therapies designed to achieve transplant success with a minimum of undesirable side effects. The aims of the present study were: (1) to evaluate the efficacy of a low-dose CSA (1.0 mg/kg/day) / methotrexate (MTX) (450 micrograms/kg/week) combination therapy in prolonging rat cardiac allograft survival, and (2) to determine the effects of low-dose CSA/MTX on interleukin-2 (IL-2) gene expression in rat cardiac allografts. The average time to rejection of newborn donor Brown Norway (BN) rat hearts transplanted into the ear pinnae of CSA/MTX-treated adult Lewis recipients, measured by the absence of electrocardiographic (ECG) activity, more than doubled from day 8 post-transplantation (post-tx) to day 18 post-tx when compared to allografts in untreated control recipients (p < or = 0.01). Northern blot analysis demonstrated that IL-2 mRNA transcripts in cardiac allografts treated with low-dose CSA/MTX were detected as early as day 1 post-tx, and at increasing levels as rejection progressed post-tx. When IL-2 gene expression in allografts from CSA/MTX-treated recipients was compared to levels in allografts from untreated recipients, no significant difference in the pattern of IL-2 induction was observed. In contrast, IL-2 mRNA transcripts were not detected post-tx in allografts from recipients treated with high-dose (15 mg/kg/day) CSA or in cardiac isografts. The presence of IL-2 gene transcripts, therefore, appears to be allograft-specific.(ABSTRACT TRUNCATED AT 250 WORDS)

HLA-DR gene expression in a proliferating human thyroid cell clone (12S).

We have used a retroviral vector carrying the adenovirus E1A oncogene and the neomycin phosphotransferase gene to establish a human thyroid-derived cell line that exhibits TSH-mediated cAMP generation as well as the differential expression of HLA class II antigens in response to recombinant gamma-interferon. Twenty-two-week gestation, histologically confirmed, human fetal thyroid was collagenase digested, cultured as a monolayer, and infected directly with 12S or 13S E1A-containing retrovirus constructs. Infected clones (n = 30) were selected in a hormone-supplemented medium containing bovine TSH (bTSH; 1 mU/ml), 10% fetal bovine serum, and 0.5 mg/ml G418 antibiotic. A rapidly growing clone (designated 12S) was chosen for detailed analysis over 18 months of continuous culture. The 12S clone was sensitive to less than 10 microU/ml bTSH when assessed by extracellular accumulation of cAMP, but TSH had no influence on 72-h incorporation of [3H]thymidine. Clone 12S responded to recombinant human gamma-interferon (1-10(4) U/ml) by induction of HLA DR alpha-chain-specific mRNA and the surface expression of HLA-DR antigen detected by fluorescein isothiocyanate-labeled monoclonal antibody to nonpolymorphic HLA-DR regions using flow cytometry. These studies indicate the potential for immortalizing human thyroid cells for use as targets of anti-TSH receptor immune responses and for long term studies of human throcyte HLA gene regulation.

Localization of HLA-DR alpha-chain messenger ribonucleic acid in normal and autoimmune human thyroid using in situ hybridization.

We studied the cellular location of HLA-DR alpha-chain synthesis within the human thyroid gland using the technique of in situ hybridization. Tritium-labeled cRNA probes for the HLA-DR alpha-chain DNA sequence revealed significant HLA-DR alpha-chain gene expression in thyroid glands from patients with Graves' hyperthyroidism and Hashimoto's thyroiditis. Hybridization signals, which were RNA strand specific, were widely distributed over thyroid follicular epithelial cells as well as over areas of lymphoid infiltration, with the highest concentration of HLA-DR alpha-chain mRNA within epithelial cells in Graves' disease follicles adjacent to areas of lymphoid infiltration. Qualitatively similar results were obtained with immunoperoxidase staining of thyroid sections for DR antigen. Thyroid tissue obtained from patients who did not have autoimmune thyroid disease contained no detectable HLA-DR antigen, but in situ hybridization revealed variable levels of HLA-DR alpha-chain mRNA in thyroid epithelial cells in such glands. Some glands had no detectable HLA-DR alpha-chain mRNA levels above background, whereas other adult glands had significant DR-specific mRNA levels. Fetal thyroid tissue, however, was negative for strand-specific HLA-DR alpha-chain transcripts as well as for HLA-DR antigen. These results indicate that thyroid epithelial cells are capable of synthesizing HLA class II antigens. The DR genes were expressed to the highest degree within the thyroid glands of patients with autoimmune thyroid disease, where expression was strongly associated with lymphoid infiltration.

Allospecific DR gene expression on the human thyroid cell.

Human thyroid epithelial cells, stimulated with recombinant gamma interferon (IF) (100 U/ml), were HLA-DR typed by means of antibody-dependent, complement-mediated cytotoxicity. The thyroid cell lysis reaction patterns were found to be similar to those obtained with autologous peripheral B cells in four of five separate experiments. Since the cytotoxic reactions with thyroid cells were sometimes incomplete with many antisera, a two-color fluorescence technique was developed to measure the specificity of the antibody-dependent, complement-mediated lysis. Using constitutively HLA-D antigen-positive Raji B lymphoblastoid cells as controls, we showed that non-polymorphic HLA-D and HLA-DR determinants could be detected on the thyroid cell surfaces by this cell lysis approach. In addition, gamma IF stimulated thyroid cells from a DR3 positive individual were lysed to a significantly greater than unstimulated autologous thyroid cells with the use of a specific monoclonal antibody to HLA-DR3, thus confirming the original HLA typing studies. These data demonstrate, for the first time, that the expression of HLA-DR antigen on the surface of human thyroid cells is allospecific and may, therefore, play an important role in the immunopathology of autoimmune thyroid disease.

Induction of rat thyroid cell MHC class II antigen by thyrotropin and gamma-interferon.

Major histocompatibility (MHC) class II antigen expression by thyroid epithelial cells has been widely implicated in the pathogenesis of autoimmune thyroid disease. We have examined rat MHC (RT1) class II antigen gene regulation in 1B-6 cells, cloned in our laboratory from the Fisher rat thyroid line FRT-L5. 1B-6 cells are TSH dependent for growth and proliferation, and are responsive to more than 5 microU/ml bovine TSH (bTSH) in terms of extracellular cAMP accumulation. Recombinant rat gamma interferon (gamma IF; 1-1000 U/ml) treatment for 5 days together with bTSH (10(3) microU/ml) was able to induce class II antigen expression in up to 90% of 1B-6 cells, as detected by a murine monoclonal antibody to rat MHC class II antigen (FITC-OX-6). Total cellular RNA was examined in Northern blot analyses using an HLA-DR alpha-chain-specific RNA probe, which shares 78% sequence homology with the rat RT1.D alpha-chain gene. A 1.4-kilobase mRNA transcript was detected in gamma IF-stimulated cells, but not in untreated cells. Dose-response studies of MHC class II gene expression, using a cytoplasmic RNA slot blot technique for alpha-chain mRNA levels and FITC-OX 6 monoclonal antibody for detection of MHC class II antigen expression, indicated 1B-6 sensitivity to gamma IF in the range of 10-100 U/ml (in the presence of 10(3) microU/ml bTSH) and to bTSH in the range of 10-100 microU/ml (in the presence of 10(2) U/ml gamma IF). These data demonstrate dual control of MHC class II antigen gene expression by gamma IF and bTSH in a differentiated rat thyroid cell. Clone 1B-6 represents a powerful means of analyzing reproducibly and in long term studies the regulation of thyroid cell MHC class II gene expression.

HLA-DP, DQ and DR gene expression in Graves' disease and normal thyroid epithelium.

HLA class II antigen expression by human thyroid epithelium may influence the pathogenesis of autoimmune thyroid disease. HLA-DP, DQ and DR alpha chain-specific RNA probes and monoclonal antibodies specific for the three HLA-D subregion products were used to analyze class II antigen expression in thyroid epithelial cells from normal thyroid tissue and thyroid tissue obtained from patients with Graves' autoimmune thyroid disease. Class II alpha chain RNA transcripts, as well as immunoreactive protein representing all three HLA class II antigens, were detected in Graves' thyroid tissue, and the levels of class II antigen expression appeared to correlate with the degree of lymphocytic infiltration within the gland. Thyroid epithelial cells from normal thyroid glands contained low or absent levels of HLA class II alpha chain mRNA and contained no detectable immunoreactive HLA class II antigen. Thyroid cell monolayer cultures prepared from both Graves' disease and normal thyroid glands exhibited coordinate induction by lectin or human recombinant gamma interferon of HLA-DP, DQ and DR gene expression. These data demonstrate the potential for HLA-DP, DQ and DR antigen expression in thyroid epithelium, and imply a dose relationship between lymphocytic infiltration and coordinate transcription and translation of HLA class II genes.

Intrathyroidal MHC class II antigen expression and thyroid autoimmunity.

The association between MHC polymorphisms and autoimmune thyroid disease has been complicated by the observation of MHC class II antigen expression by the human thyroid gland. It is possible that the MHC associations observed in animal and human population studies may have mechanistic explanations at the level of the thyroid cell. There is evidence for expression of HLA-DR allospecific antigen in both normal and abnormal human thyroid cells, with enhanced expression in patients with autoimmune thyroid disease. Such MHC class II expression appears to be mediated primarily by lymphokine secretion from intrathyroidal T lymphocytes. Thyroid cell HLA-DR antigen participates in activation and amplification of T cells and is likely to be involved in presentation of thyroid antigen to the immune system. The relationship between these immune interactions and the initial events leading to the development of autoimmune thyroid disease remains to be understood.

Lymphokine regulation of HLA-DR gene expression in human thyroid cell monolayers.

Studies were conducted to examine the regulation of HLA class II gene expression in human thyroid cells in vitro. Normal human thyroid cells cultured in the absence of lectin or gamma-interferon stimulation lacked detectable HLA-DR cell surface antigen, although low levels of DR alpha-chain-specific mRNA were present. Cyclosporine A, known to inhibit lymphokine production, inhibited basal as well as lectin-mediated increases in levels of DR alpha-chain-specific mRNA and DR surface antigen expression on normal human thyrocytes. Cyclosporine had no effect on the induction of DR antigen gene expression by recombinant gamma-interferon. These data suggested that lectin enhancement of DR antigen expression in human thyroid cells may be mediated by a lymphokine(s) produced in primary human thyroid cell monolayers. This suggestion was confirmed by studies that demonstrated the abrogation of lectin responsiveness by antibody directed against gamma-interferon. Indirect immunofluorescence studies using flow cytometric analyses identified 1.6 +/- 0.2% (mean +/- SD) of cells in primary thyroid cultures as T lymphocytes, a potential source of lymphokine production. Cells derived from thyroid follicular adenomas and carcinomas demonstrated reduced lectin-mediated increases in DR antigen expression compared to normal thyroid cells. DR expression could be enhanced in these lectin-treated cells, however, by T cell coculture. Dose-response studies demonstrated that human thyroid cells were as sensitive to gamma-interferon induction of DR antigen expression as human monocyte/macrophages. These results indicate that human thyroid cell HLA-DR antigen gene expression is sensitive to low levels of lymphokines, such as gamma-interferon; an intrathyroidal T cell population, which may serve as a source of lymphokine(s), remains associated with thyroid epithelial cells in primary thyroid cultures; and lymphokine-thyroid cell interactions may be implicated in the immunopathology of human autoimmune thyroid disease.

Thyroid cell MHC class II antigens: a perspective on the aetiology of autoimmune thyroid disease.

An association exists between certain MHC polymorphisms and autoimmune thyroid disease in animals and humans. The observation of MHC class II antigen expression by the thyroid suggests that such associations may have mechanistic explanations at the level of the thyroid cell. Such class II antigen expression, rather than being a constitutive property of thyroid epithelium, appears to be primarily mediated by lymphokine secretion from intrathyroidal T lymphocytes and a variety of agents, for example TSh and TSH receptor antibodies, may amplify such lymphokine action. Thyroid cell class II antigens participate in activation and amplification of T cells and are involved in presentation of thyroid antigen to the immune system. The relationship between these local immune interactions and the initial events leading to the development of autoimmune thyroid disease requires a more fundamental understanding of the workings of the immune system at the site of antigenic stimulation.

Tyrosinase isozyme heterogeneity in differentiating B16/C3 melanoma.

The B16/C3 murine melanoma is a pigmented tumor that is rich in the copper-containing enzyme, tyrosinase. This enzyme, which converts tyrosine to melanin precursors, is largely associated with membrane fractions of cells and exists in a number of discrete isozymic forms ranging in molecular mass from 58,000 to 150,000 daltons and pI from 3.4 to 5.2. One of these isozymes (Mr = 58,000, pI 3.4) has been purified to homogeneity. The purified enzyme catalyzes the hydroxylation of L-tyrosine to L-dihydroxyphenylalanine (L-DOPA) and the conversion of L-DOPA to dopaquinone. Ascorbic acid, tetrahydrofolate, and dopamine can serve as cofactors in the hydroxylase reaction. The Michaelis constants for the purified enzyme were 7 X 10(-4) M for L-tyrosine and 6 X 10(-4) M for L-DOPA. The Vmax for L-DOPA was much greater than the Vmax for L-tyrosine indicating that tyrosine hydroxylation is rate-limiting in melanin precursor biosynthesis. Two putative copper chelators, phenylthiourea and diethyldithiocarbamide inhibited both the tyrosine hydroxylase and L-DOPA oxidase activities of the enzyme. Phenylthiourea was a noncompetitive inhibitor while diethyldithiocarbamide was a competitive inhibitor indicating that these agents act by different mechanisms. When digested with proteases and glycosidases, higher molecular weight forms of tyrosinase co-migrated with the purified enzyme in isoelectric focusing and sodium dodecyl sulfate-polyacrylamide gel electrophoresis suggesting that the isozyme was derived from larger precursors. Thus, post-translational processing of tyrosinase may underlie isozyme diversity and this may be important in the control of melanogenesis in this tumor model.

HLA-DR alpha chain expression in human thyroid cells.

Using a cDNA probe encoding the human major histocompatibility Class II antigen HLA-DR alpha chain, we have detected a single DR alpha chain-specific transcript in total cellular RNA prepared from human thyroid tissue. The hybridizable RNA in thyroid samples was indistinguishable in size from the DR mRNA in the Raji human B lymphoblastoid cell line. Of the thyroid glands examined, samples from patients with autoimmune thyroid disease consistently demonstrated the highest DR alpha chain transcript levels, with a mean +/- SEM OF 62 +/- 13% of levels found in DR-positive Raji cells. Cytoplasmic dot blot analyses of 5-day thyroid cell cultures depleted of lymphocytes and monocytes indicated that normal thyrocytes contain readily-detectable levels of DR alpha chain-mRNA. These transcript levels varied from individual to individual, with a mean of 16 +/- 9% relative to Raji cell control values and were shown to correlate after lectin or gamma interferon stimulation with enhanced numbers of immunoreactive DR antigen-positive cells. Such findings demonstrate expression of HLA Class II antigen genes in normal unstimulated human thyroid cells and suggest that quantitative variation in thyroid Class II antigen (DR) gene expression may be a major factor in thyroid immunoregulation.