HLA class II induction in human islet cells by interferon-gamma plus tumour necrosis factor or lymphotoxin

TNF impairs in vivo and in vitro natural killer (NK) susceptibility of B16 melanoma cells

Tumour necrosis factor alpha (TNF) is a multipotent cytokine which affects many biological properties of both normal and neoplastic cells. Here we show that treatment with TNF reduces B16-A melanoma cell susceptibility to normal and in vivo- and in vitro-activated NK cell-mediated killing. This resistance is associated with an enhancement of B16-A metastatic potential in normal syngeneic mice, but not in anti-asialo GM1-treated animals, further supporting the NK dependence of TNF-induced enhancement of metastatic ability. A significant increase of MHC class I expression on B16-A murine melanoma cells is observed after TNF treatment. In all these effects TNF interacts positively with interferon gamma (IFN gamma). Taken together, these results indicate that TNF treatment negatively affects the susceptibility of B16-A murine melanoma to NK effectors in vivo and in vitro. This decreased susceptibility may be related, at least in part, to enhanced expression of MHC class I antigens on tumour cells.

Class II MHC antigens in normal human skeletal muscle

Class II MHC antigen expression has been investigated in muscle tissue and cultured cells from normal human skeletal muscle by light and electron immunocytochemistry. In muscle tissue, these antigens were detected in satellite cells, interstitial cells, and blood vessels. In cultures, muscle cells were stained with a pan-reactive anti-HLA class II antibody and with isotypes specific for DP, DQ, and DR. The staining was present on mononucleated cells and persisted on myotubes; it was stronger for DR and DQ isotypes than for DP. At the subcellular level, staining was located not only at the cell surface, but also next to the endoplasmic reticulum and in the cytosol. Thus, myosatellite cells and aneurally cultured cells from human normal skeletal muscle express class II MHC antigens. Moreover, the myotube staining and the presence of gold particles inside the cells suggested synthesis of these antigens after myoblast fusion.

Modulation of adhesion of lymphocytes to murine brain endothelial cells in vitro: relation to class II major histocompatibility complex expression

Adhesion of lymphocytes to mouse brain endothelial cells was studied after treatment of the endothelium with 1000 U/ml gamma interferon (IFN-gamma) for 1 h to 2 days. Adhesion was not significantly different from controls after 1 h but at 4 h and thereafter, adhesion increased in a time-related manner. IFN-gamma also increased the expression of class II major histocompatibility complex (MHC) and murine intercellular adhesion molecule-1 (ICAM-1) molecules on the endothelial cells. The level of expression of class II MHC molecules was related to the length of exposure to IFN-gamma. MAb blocking studies suggested that class II molecules were responsible for the IFN-gamma-induced increase in lymphocyte-endothelial cell adhesion. Transfection of a murine lung endothelial cell line with cDNA for the class II MHC molecule also produced a significant increase in lymphocyte-endothelial cell adhesion, suggesting that the class II MHC molecule may have a role in adhesion which is distinct from antigen presentation.

Expression of MHC class II antigens (HLA-DR, -DP, and -DQ) on human gastric epithelium

Class II antigen expression on gastric epithelium was investigated using an immunoperoxidase method in relation to the degree of inflammatory cell infiltration in the lamina propria. Sixty-six biopsy specimens from 43 patients with chronic gastritis were examined. The frequency of HLA-DR expression in specimens with cell infiltration was 94%, while that in specimens without cell infiltration was 24%. There was significant difference in the frequency of HLA-DR expression between the two groups (P less than 0.01). HLA-DR was most intensely expressed in the glandular neck portion. The frequency and extent of class II antigen expression on gastric epithelium with cell infiltration were in the following order: DR greater than DP greater than DQ. The extent of DR and DP, but not DQ expression generally paralleled the degree of cell infiltration. Intestinal metaplasia was found in 13 specimens. In the area of intestinal metaplasia, epithelial class II staining was absent except for one specimen. These results suggest that the respective genes of three class II antigens are regulated by different mechanisms and that an immunological mechanism plays a role in the pathogenesis of gastritis.

Cytokines as immunological adjuvants

Various cytokines have been shown to be effective immunological adjuvants in a variety of model systems, enhancing protection induced by viral, bacterial and parasitic vaccines, and increasing parameters of immunity in tumour immunization models and in clinical trials. While in most cases cytokine adjuvanticity is not as powerful as that shown by the best experimental adjuvants, such as saponin and Freund's, it can rival that of the adjuvants presently allowed for human use and there are many possible routes to improvement. The use of cytokines may allow for a choice of which immune parameters are enhanced in order to further enhance protective effects and decrease the negative effects of vaccines.

Mumps virus infects beta cells in human fetal islet cell cultures upregulating the expression of HLA class I molecules

The ability of mumps virus to infect pancreatic Beta cells and cause alterations in their HLA expression was evaluated in cultured human fetal islet cell clusters. Mumps virus could be isolated during the whole culture period (6-8 days) and 60% of cells, including Beta cells, contained viral nucleocapsid protein at the end of the culturing. A minor decrease in insulin secretion was observed in some of the infected cultures. The infection was invariably associated with an increase in the expression of HLA class I molecules. This enhancement was mediated by soluble factors secreted by infected cells. The infection could not induce the expression of HLA-DR molecules. However, external interferon-gamma was able to cause a clear rise in DR-expression which was observed only on non-Beta-cells. Rubella and coxsackie B4 viruses were also able to enhance the expression of class I molecules while herpes simplex virus type 2 was not. The results suggest that certain viruses are able to infect Beta cells and cause alterations in their immunological appearance. Increased HLA class I expression in infected islets may exaggerate the autoimmune process in pre-diabetic individuals by increasing the activity of autoreactive cytotoxic T cells.

Tumor necrosis factor alpha in murine systemic lupus erythematosus disease models: implications for genetic predisposition and immune regulation

Recombinant tumor necrosis factor alpha (TNF-alpha) administration significantly delayed the development of lupuslike nephritis in the New Zealand black x New Zealand white (NZB x NZW)F1 and to a lesser extent in the MRL-lpr/lpr model systems. TNF-alpha treatment was effective when treatment was initiated at 2, 3, or 4 months of age but was ineffective if initiated as late as 6.5 months of age. Treatment of (NZB x NZW)F1 mice for 3 months was more effective than treatment continued for 6 months. Anti-TNF-alpha antibodies did not develop in these mice. Flow microfluorometry analysis showed no major effects on B, T, or monocyte cell population in cells from the peritoneum, spleen, lymph node, and thymus. A decrease in class II Ia expression on macrophages in the peritoneum of TNF-alpha-treated mice was noticed. A correlation between the level of TNF-alpha inducibility in vitro and the effect of TNF-alpha administration in vivo could be shown. Although a limited polymorphism could be shown by restriction fragment length polymorphism, using an amplified (AC)n microsatellite located in the 5' regulatory region of TNF-alpha, a much more extensive interallelic polymorphism was found. The AC microsatellite allele found in NZW mice was unique and different from other lupus strains and nonautoimmune strains. These results have possible implications to the pathogenesis of systemic lupus erythematosus.

Extensive genetic polymorphism in the human tumor necrosis factor region and relation to extended HLA haplotypes

We have identified three polymorphic microsatellites (which we call TNFa, TNFb, and TNFc) within a 12-kilobase region of the human major histocompatibility complex (MHC) that includes the tumor necrosis factor (TNF) locus. TNFc is located within the first intron of the TNF-beta gene and has only 2 alleles. TNFa and TNFb are 3.5 kilobases upstream (telomeric) of the TNF-beta gene and have at least 13 and 7 alleles, respectively. TNFa, -b, and -c alleles are in linkage disequilibrium with alleles at other loci within the MHC, including class I, class II, and class III. TNFa, -b, and -c alleles are also associated with extended HLA haplotypes. These TNF polymorphisms will allow a thorough genetic analysis of the involvement of TNF in MHC-linked pathologies.

Overexpression of class I major histocompatibility complex accompanies insulitis in the non-obese diabetic mouse and is prevented by anti-interferon-gamma antibody

Overexpression of class I major histocompatibility complex (MHC) proteins on pancreatic islet cells is a characteristic of autoimmune Type 1 (insulin-dependent) diabetes mellitus in humans and in animal models. Studies of post-mortem pancreases from humans with Type 1 diabetes suggest that overexpression of class I MHC proteins may precede mononuclear cell infiltration of the islets (insulitis). Pancreatic histology from the earliest stages of human Type 1 diabetes is rarely available. We have used the non-obese diabetic mouse, given cyclophosphamide to accelerate Beta-cell destruction, to investigate the temporal relationship between the overexpression of class I MHC protein and mRNA and other pathological changes associated with Beta-cell destruction. Prior to cyclophosphamide, immunoperoxidase staining showed that expression of class I MHC proteins was greater on islet cells and infiltrating inflammatory cells of the non-obese diabetic mouse than on islet cells of other mouse strains, whereas staining on exocrine cells was similar. On day three after cyclophosphamide administration, when insulitis had regressed, islet class I MHC protein expression had diminished. A dramatic increase in class I MHC protein expression occurred between days seven and nine, concomitant with reinfiltration of the islets by mononuclear cells; overexpression was seen both on islet cells and on surrounding exocrine cells, but only in the presence of mononuclear cell infiltration. By day 21, class I MHC protein overexpression was again confined to the islets, the exocrine pancreas being free of infiltration. Class I mRNA also increased dramatically by day eight but had virtually returned to normal by day 12.2+ effected by cytokines secreted by activated immuno-inflammatory cells. Class I MHC overexpression should enhance targeting of cytotoxic T cells to Beta cells bearing autoantigen.

Insulitis in type 1 (insulin-dependent) diabetes mellitus in man--macrophages, lymphocytes, and interferon-gamma containing cells

This study sought to determine, firstly, the relative frequency of lymphocytes and macrophages and, secondly, the percentage of lymphocytes containing interferon-gamma in inflamed islets (insulitis) of patients with type 1 (insulin-dependent) diabetes. Autopsy pancreases of 12 patients who had died of recent-onset type 1 diabetes and one pre-diabetic patient who had died of cardiomyopathy were examined immunohistochemically. In the 87 islets that were studied, the lymphocyte macrophage ratio was 9.7:1 and approximately 40 per cent of the lymphocytes contained interferon-gamma. Interferon-gamma release in the insulitis process may be involved in the pathogenesis of type 1 diabetes.

Interferon stimulates the expression of 2',5'-oligoadenylate synthetase and MHC class I antigens in insulin-producing cells

The pathogenesis of type 1 diabetes involves autoimmune processes directed against the pancreatic beta-cells. The etiology is not known, but circumstantial evidence suggests a connection between virus infection and development of the disease. Therefore, because the interferon-(IFN) dependent 2',5'-oligoadenylate (2-5A) synthetase system constitutes an important part of the nonspecific immune defense against viral infections, the activity of the enzyme was examined in islets of Langerhans, RIN cells, and GH3 cells. First, the 2-5A synthetase was expressed constitutively in all cell types and, second, all cells were sensitive to stimulation with IFN-alpha. The 2-5A synthetase activity induced by 1,000 U/ml of IFN-alpha increased by 400% in pancreatic islets and by more than 1000% in GH3 and RIN cells. However, the IFN-alpha concentration needed to induce half-maximal 2-5A synthetase activity was nearly the same in the three cell types (i.e., ranging from 59 to 66 U/ml IFN-alpha). The 2-5A synthetase present in islets and RIN cells was highly sensitive to poly (I:C). In pancreatic islets and RIN cells, the 2-5A synthetase enzyme generated dimers and trimers of 2',5'-oligoadenylates. Furthermore, exposure of RIN cells to IFN-alpha showed an increase in MHC class I expression already at 5 U/ml and maximal expression at about 200 U/ml IFN-alpha. The examined endocrine cells express the 2-5A synthetase enzyme as well as MHC class I antigen constitutively, but also by stimulation with IFN in vitro.(ABSTRACT TRUNCATED AT 250 WORDS)

Effect of cyclosporine A on the regulation of Ia antigen keratinocytes expression

Since many skin diseases characterized by positive Ia keratinocytes show improvement with cyclosporine therapy, the purpose of this study was to determine whether cyclosporine A (CyA) alters the expression of Ia keratinocytes. Nude mice were injected with normal mouse serum (NMS) to induce keratinocyte expression of the Ia antigen. The injected mice were then divided into four groups: one was treated with oral CyA; the second was treated topically with CyA twice a day; the third was treated topically with olive oil; and the fourth was injected with nude mouse serum. The third and fourth groups served as Ia positive and Ia negative controls, respectively. The mice were treated during the first 10 days after the injections. On Day 10, epidermal sheets were analyzed for Ia expression. Analysis was made by an indirect immunoperoxidase staining method using monoclonal antibodies specific for Ia determinants. Quantitation of the number of Langerhans cells was analyzed on epidermal sheets using immunodiagnostic reagents, anti-MHC-Ia, and surface ectoenzyme, ATPase. A significant reduction of Ia-positive keratinocytes was noted in the oral CyA group vs topical and olive oil groups (64.9 +/- 29.9% vs 20.1 +/- 18.7%, respectively, P less than 0.01). In a second set of experiments mice were injected with NMS, but treatment was started only on Day 10 after injections, for 10 days. The results showed that CyA failed to down-regulate Ia expression. Topical and systemic CyA did not modify Langerhans cell population. The present study showed that systemic administration of CyA significantly reduced Ia induction by keratinocytes of nude mice that were injected with NMS.

Gamma-interferon promotes proliferation of adult human astrocytes in vitro and reactive gliosis in the adult mouse brain in vivo

Reactive gliosis is a characteristic response of astrocytes to inflammation and trauma of the central nervous system. To investigate whether soluble factors (cytokines) from inflammatory mononuclear cells that accumulate at lesion sites can provide the cellular signals to initiate gliosis and to identify such cytokines, we have tested and found that supernatants derived from subsets of activated human T lymphocytes (CD8+ or CD4+) are potent mitogens for cultured human adult astrocytes. This effect is blocked by a neutralizing antibody to gamma-interferon (IFN). Recombinant IFN alone can induce proliferation of human adult astrocytes in vitro and increase the extent of trauma-initiated gliosis in the adult mouse brain. The astrocyte proliferation-inducing activity of supernatants of glial cultures treated with IFN can be completely blocked with IFN-neutralizing antibody, suggesting that the proliferative effect does not require intermediary cytokines or cells. These results implicate IFN as an important mediator of the gliosis observed in pathologic conditions of the adult central nervous system associated with infiltrating lymphocytes.

Enhanced expression of HLA-class II molecules on activated human T lymphocytes following treatment with tumor necrosis factor alpha

Many factors induce or enhance expression of major histocompatibility complex class I and class II molecules on various cell types. Human T lymphocytes are class II negative in the resting state but show expression of class II molecules following activation. We analyzed the modulating capacity of the lymphokines recombinant interferon gamma (rIFN-gamma), interleukin-4 (IL-4), and recombinant tumor necrosis factor alpha (rTNF-alpha) on class II expression in subsets of alloactivated human T lymphocytes. The activated CD4+ T cells expressed all three class II isotypes (DR, DQ, and DP), whereas the cytotoxic CD8+ T-cell lines expressed DR and DP molecules but failed to bind DQ-specific monoclonal antibodies significantly. Treatment with rIFN-gamma and IL-4 had no effect on class II expression on any of the T-cell lines or clones, whereas rTNF-alpha enhanced class II expression in both subsets. rTNF-alpha could modulate expression of all three class II isotypes but, in principle, it appears only to affect ongoing class II synthesis as de novo synthesis of class II molecules with a resultant change in the class II phenotype from DR+ DQ- DP+ to DR+ DQ+ DP+ in the CD8+ T lymphocytes was not observed. No synergic effects of rINF-gamma and rTNF-alpha were observed; this results from the fact that activated T cells express few, if any, receptors of rIFN-gamma.

The immune-hypothalamo-pituitary-adrenal axis and autoimmunity

Immunoendocrinology is a rapidly expanding field, uncovering numerous bilateral interactions between the immune system and neuroendocrine circuits. Various hormones and neurotransmitters appear to modulate cells of the immune system and likewise cytokines control the function of neuroendocrine system. In the present paper, we discuss some lines of evidence indicating that an immunoendocrine feedback loop, which we term 'immune-hypothalamo-pituitary-adrenal system' is an integral part of the regulation of self tolerance. The finding that pathology of this immunoendocrine feedback loop is related to development of autoimmunity may lead to new prophylactic and therapeutic strategies.

The production of immunoreactive alpha- and gamma-interferon by circulating mononuclear cells in type 1 diabetes

Using a specific immunoradiometric assay method the in vitro alpha interferon response to polyinosinic:polycytidylic acid (poly-I:C) and the gamma interferon response to concanavalin A were measured in peripheral blood mononuclear cells from 11 healthy matched pairs of Type 1 diabetic patients and normal subjects. The alpha-interferon response to poly-I:C was significantly higher in the diabetic group (median 3.7 (range less than 1-25.7) u ml-1) than in the normal group (1.1 (less than 1-15.4) u ml-1, p less than 0.01). The mean gamma-interferon response to concanavalin A was 64.3 +/- 46.9 (+/- SD) u ml-1 in the diabetic patients and 49.4 +/- 18.5 u ml-1 in the normal group (NS). The higher alpha-interferon response to poly-I:C in the diabetic patients was not related to blood glucose concentration, HbA1, age of onset of diabetes, duration of diabetes, or islet cell antibody positivity, and may therefore indicate intrinsic hyper-responsiveness of circulating mononuclear cells in Type 1 diabetes.

Transfection with SV40 gene of human pancreatic endocrine cells

At present, only islet cell lines of animal origin have been successfully generated (e.g. RIN, HIT). A fully differentiated human beta cell line would be advantageous for diabetes research. We now report the generation of a human endocrine pancreatic cell line obtained by transfection using a plasmid containing the early region of SV40 viral DNA. Viral integration and transcription was assessed by Southern and Northern blotting. This cell line has been growing continuously for more than 2 years and maintains several of the characteristics of the parental cells from which they were generated. The presence of Neuron Specific Enolase, Protein Gene Product 9.5, cytokeratin, microvilli, cytoplasmic electrodense granules and the secretion of insulin, glucagon and somatostatin supports the neuroendocrine origin of this cell line. However, hormone production progressively decreased and finally stopped at passage 8. Flow cytometric analysis showed that HLA expression in this cell line is readily induced by IFN-gamma and modulated by TNF-alpha. The establishment of this human endocrine cell line indicates the feasibility of immortalizing human islets by transfection with viral oncogenes. To obtain a fully differentiated cell line it may be necessary to use other DNA constructs which immortalize the cells without fully transforming their phenotype.

Immunoregulators in the nervous system

The nervous system, through the production of neuroregulators (neurotransmitters, neuromodulators and neuropeptides) can regulate specific immune system functions, while the immune system, through the production of immunoregulators (immunomodulators and immunopeptides) can regulate specific nervous system functions. This indicates a reciprocal communication between the nervous and immune systems. The presence of immunoregulators in the brain and cerebrospinal fluid is the result of local synthesis--by intrinsic and blood-derived macrophages, activated T-lymphocytes that cross the blood-brain barrier, endothelial cells of the cerebrovasculature, microglia, astrocytes, and neuronal components--and/or uptake from the peripheral blood through the blood-brain barrier (in specific cases) and circumventricular organs. Acute and chronic pathological processes (infection, inflammation, immunological reactions, malignancy, necrosis) stimulate the synthesis and release of immunoregulators in various cell systems. These immunoregulators have pivotal roles in the coordination of the host defense mechanisms and repair, and induce a series of immunological, endocrinological, metabolical and neurological responses. This review summarizes studies concerning immunoregulators--such as interleukins, tumor necrosis factor, interferons, transforming growth factors, thymic peptides, tuftsin, platelet activating factor, neuro-immunoregulators--in the nervous system. It also describes the monitoring of immunoregulators by the central nervous system (CNS) as part of the regulatory factors that induce neurological manifestations (e.g., fever, somnolence, appetite suppression, neuroendocrine alterations) frequently accompanying acute and chronic pathological processes.

De novo HLA class II and enhanced HLA class I molecule expression in SV40 transfected human thyroid epithelial cells

Human thyroid follicular cells normally synthesize and express HLA Class I molecules but in glands affected by autoimmune or neoplastic diseases they express Class II molecules. We have investigated the effect of SV40 virus transformation on the expression of MHC molecules in human thyrocytes. Primary cultures of human thyroid from two different glands were transfected with the plasmid pX-8, containing the 'early' region of SV40 virus, and two continuous lines of thyrocytes were obtained. The cell lines maintained features of parental thyroid epithelial cells showing the characteristic cytokeratin filament network, microvillar protrusion and tight junctions. In addition, the SV40-transfected cells responded to graded doses of TSH with increased c-AMP production. Thyrocytes from both cell lines hyperexpressed Class I molecules and a significant proportion of them also acquired constitutive Class II expression, as determined by indirect immunofluorescence (IFL), flow cytometry and Northern blotting hybridization using a DR beta probe. These cells were found to be normally up-regulated by interferon (IFN)-gamma. Indirect IFL and flow cytometry analysis were used to detect and quantify the expression of HLA-DR, DP and DQ subregions. A co-ordinated expression (DR greater than DP much greater than DQ), reminiscent of the inappropriate HLA expression found in thyroid autoimmune disease in vivo and of the in vitro regulation in normal thyrocytes, was observed. Clones derived from these cell lines differed in their level of constitutive Class II expression and in their sensitivity to Class II induction by IFN-gamma. In conclusion, these thyroid cell lines could provide a useful tool for further investigation of HLA gene regulation in thyroid cells and for elucidating on the mechanism involved in the inappropriate HLA expression described in autoimmune and neoplastic diseases of the thyroid.

Influence of cytokines and immunosuppressive drugs on major histocompatibility complex class I/II expression by human cardiac myocytes in vitro

Human cardiac myocytes do not express detectable levels of major histocompatibility complex (MHC) class II antigens and express low levels, if any, of MHC class I antigens. During rejection episodes, cardiac biopsies show massive increases of MHC antigens, which are thought to be induced by cytokines released by donor-sensitized recipient mononuclear cells. In efforts to determine the nature of the cytokines that induce MHC expression on cardiac myocytes, human fetal cardiac myocyte cultures were established. Interferon-alpha (IFN-alpha), interferon-gamma (IFN-gamma), interleukin (IL)-1, IL-2, IL-3, IL-4, and tumor necrosis factor (TNF)-alpha were added to these cultures and dose/kinetics of MHC class I/II induction quantitated. Data show that IFN-gamma induces both MHC class I and II expression, and all the other cytokines (except IL-2) induce only MHC class I but not class II. Cytokines used in combination showed that IFN-alpha with TNF-alpha was the only combination that induced MHC class II expression. Addition of immunosuppressive drugs such as cytoxan, azathioprine, cyclosporine-A, and FK-506, even when added at the initiation of the cultures, did not appreciably affect the ability of the appropriate cytokines to induce MHC expression by the myocytes in vitro.

Stereospecific alignment of the X and Y elements is required for major histocompatibility complex class II DRA promoter function

The regulatory mechanisms controlling expression of the major histocompatibility complex (MHC) class II genes involve several cis-acting DNA elements, including the X and Y boxes. These two elements are conserved within all murine and human class II genes and are required for accurate and efficient transcription from MHC class II promoters. Interestingly, the distance between the X and Y elements is also evolutionarily conserved at 18 to 20 bp. To investigate the function of the invariant spacing in the human MHC class II gene, HLA-DRA, we constructed a series of spacing mutants which alters the distance between the X and Y elements by integral and half-integral turns of the DNA helix. Transient transfection of the spacing constructs into Raji cells revealed that inserting integral turns of the DNA helix (+20 and +10 bp) did not reduce promoter activity, while inserting or deleting half-integral turns of the DNA helix (+15, +5, and -5 bp) drastically reduced promoter activity. The loss of promoter function in these half-integral turn constructs was due neither to the inability of the X and Y elements to bind proteins nor to improper binding of the X- and Y-box-binding proteins. These data indicate that the X and Y elements must be aligned on the same side of the DNA helix to ensure normal function. This requirement for stereospecific alignment strongly suggests that the X- and Y-box-binding proteins either interact directly or are components of a larger transcription complex which assembles on one face of the DNA double helix.

Antierythrocyte autoantibody formation after therapy with interleukin-2 and gamma-interferon

The cardiovascular, renal, pulmonary, and dermatologic toxicities of interleukin-2 (IL-2) and gamma-interferon (IFN) are well described. However, autoimmune toxicities have only recently been noticed. The authors report the development of warm autoantibodies against erythrocytes in a patient receiving IL-2 (3.75 x 10(6) cetus units/m2 intravenous bolus three times per week) and gamma-IFN (0.1 mg/m2 subcutaneously three times per week) for metastatic renal cell carcinoma. Other potential causes of autoantibody formation, such as drugs, infection, and collagen vascular disease, were excluded. Both gamma-IFN and IL-2 have the potential to trigger or exacerbate autoimmunity due to either aberrant expression of restricted antigens or inhibition of normal cellular immune suppressor mechanisms.

Human X-box-binding protein 1 is required for the transcription of a subset of human class II major histocompatibility genes and forms a heterodimer with c-fos

A complementary DNA encoding a member of the leucine-zipper class of proteins (human X-box-binding protein, hXBP-1) that binds to the 3' end of the conserved X box (X2) of the HLA-DRA major histocompatibility complex gene was recently described. Further gel-retardation analysis has demonstrated that hXBP-1 also binds to HLA-DPB X2 but not to other X2 sequences. Transient transfection of a mammalian expression vector with the hXBP-1 cDNA inserted in the antisense orientation represses the surface expression of HLA-DR and HLA-DP in Raji cells. Cotransfection of the antisense hXBP-1 vector with a HLA-DRA/chloramphenicol acetyltransferase (but not a HLA-DQB/chloramphenicol acetyltransferase) reporter plasmid decreases chloramphenicol acetyltransferase activity in Raji cells and in gamma-interferon-treated HeLa cells relative to cells cotransfected with a control antisense vector. Moreover, hXBP-1 is shown to form a stable heterodimer with the product of the c-fos protooncogene. These data suggest that the hXBP-1 c-fos heterodimer is critical for the transcription of a subset of the human class II major histocompatibility complex genes and that the regulatory mechanisms for the different class II genes are distinct.

Stereospecific alignment of the X and Y elements is required for major histocompatibility complex class II DRA promoter function

The regulatory mechanisms controlling expression of the major histocompatibility complex (MHC) class II genes involve several cis-acting DNA elements, including the X and Y boxes. These two elements are conserved within all murine and human class II genes and are required for accurate and efficient transcription from MHC class II promoters. Interestingly, the distance between the X and Y elements is also evolutionarily conserved at 18 to 20 bp. To investigate the function of the invariant spacing in the human MHC class II gene, HLA-DRA, we constructed a series of spacing mutants which alters the distance between the X and Y elements by integral and half-integral turns of the DNA helix. Transient transfection of the spacing constructs into Raji cells revealed that inserting integral turns of the DNA helix (+20 and +10 bp) did not reduce promoter activity, while inserting or deleting half-integral turns of the DNA helix (+15, +5, and -5 bp) drastically reduced promoter activity. The loss of promoter function in these half-integral turn constructs was due neither to the inability of the X and Y elements to bind proteins nor to improper binding of the X- and Y-box-binding proteins. These data indicate that the X and Y elements must be aligned on the same side of the DNA helix to ensure normal function. This requirement for stereospecific alignment strongly suggests that the X- and Y-box-binding proteins either interact directly or are components of a larger transcription complex which assembles on one face of the DNA double helix.

Ablation of "tolerance" and induction of diabetes by virus infection in viral antigen transgenic mice

To address the mechanisms of tolerance to extrathymic proteins, we have generated transgenic mice expressing the lymphocytic choriomeningitis viral (LCMV) glycoprotein (GP) in the beta islet cells of the pancreas. The fate of LCMV GP-specific T cells was followed by breeding the GP transgenic mice with T cell receptor transgenic mice, specific for LCMV and H-2Db. These studies suggest that "peripheral tolerance" of self-reactive T cells does not involve clonal deletion, clonal anergy, or a decrease in the density of T cell receptors or accessory molecules. Instead, this model indicates that self-reactive cytotoxic T cells may remain functionally unresponsive, owing to a lack of appropriate T cell activation. Infection of transgenic mice with LCMV readily abolishes peripheral unresponsiveness to the self LCMV GP antigen, resulting in a CD8+ T cell-mediated diabetes. These data suggest that similar mechanisms may operate in several so-called "T cell-mediated autoimmune diseases."

Reduction in insulitis following administration of IFN-gamma and TNF-alpha in the NOD mouse

In insulin dependent diabetes mellitis (IDDM) beta cell destruction is associated with infiltration of the pancreatic islets by T lymphocytes and macrophages. Cytokine products from the infiltrating immunocytes not only have powerful immunoregulatory actions but also are capable of impairing islet cell functions and have thus been postulated to assume a central role in mediating anti-beta cell immunity and beta cell destruction. In an effort to explore further the role of cytokines in the pathogenesis of IDDM, we examined clinical, metabolic and pathological features of NOD/Wehi mice injected intraperitoneally with multiple doses of IFN-gamma and/or TNF-alpha. Blood glucose profiles were not significantly altered by injection of cytokines alone or in combination. Except for a hypoglycaemic rebound in mice injected with TNF-alpha, arginine stimulation tests revealed no disturbances in islet secretory function in cytokine injected mice. Compared with vehicle and cytokines alone, injection of IFN-gamma + TNF-alpha was associated with a variety of clinical and pathological changes including abdominal distention, piloerection, ascites, oedema, thymic atrophy, splenic enlargement and pancreatic distention. Histological examination of the pancreas in these mice revealed moderate to severe pancreatitis which included focal haemorrhagic necrosis, oedema and polymorphonuclear and mononuclear cell infiltration. The islets in these mice appeared normal morphologically and when stained for insulin. The injection of IFN-gamma + TNF-alpha, and to a lesser extent TNF-alpha alone, was associated with a significant reduction in the severity of insulitis. Examination of pancreatic MHC-class I and class II molecule expression revealed in mice given IFN-gamma + TNF-alpha, as compared with controls, significant and uniform induction of both these molecules on ductal and acinar cells; low level MHC-class II expression was also detectable on beta cells in these mice. MHC-class I molecules which were expressed at high levels by beta cells in control mice did not appear to change following administration of the cytokines alone or in combination. We conclude that despite their immunostimulatory actions in vitro and in other models in vivo, systemic administration of the cytokines IFN-gamma and/or TNF-alpha to NOD/Wehi mice does not activate or enhance, and may actually suppress, anti-beta cell immunity in this model.

Cytotoxic effect of IFN-gamma plus TNF-alpha on human islet cells

We have previously reported that the combination of IFN-gamma plus TNF-alpha is able to induce the de novo expression of HLA class II on human beta cells. In the present study, we have investigated the effect of these cytokines, alone or in combination, on the function and viability of human islet cells in vitro. Three hour insulin release was markedly reduced in human islet monolayer cultures after 4 days' exposure to 1000 U/ml of the combination TNF-alpha plus IFN-gamma (36.7 +/- 7.7, % of the control +/- SEM) or to TNF-alpha alone (49.5 +/- 7% of the control) while IFN-gamma had little effect. On direct inspection cell damage was clearly detected only in the cultures treated with TNF-alpha plus IFN-gamma in which staining by indirect immunofluorescence (IFL) for insulin revealed that the number of beta cells was also significantly reduced, thus suggesting a real cytotoxic effect of this cytokine combination. This effect was not beta cell specific since glucagon release and the number of alpha cells were also reduced in the cultures exposed to IFN-gamma plus TNF-alpha. 51Cr release experiments supported the cytoxicity of these cytokines to normal islet cells. There was a time course relationship between class II induction (2 days) and the cytotoxic effect of IFN-gamma plus TNF-alpha (4 days) on the same islet cells. In conclusion, these results indicate that the combination of IFN-gamma and TNF-alpha exerts a cytotoxic effect on human islet cells in vitro.

Immune reactivity to fractionated Leishmania aethiopica antigens during active human infection

Fractionated antigen preparations of Leishmania aethiopica parasites were used to stimulate the peripheral blood lymphocytes of patients with active cutaneous leishmaniasis. In assays measuring lymphocyte proliferation, 9 of 10 patients with similar clinical presentations of infection responded in a similar pattern to the fractionated antigens. Marked proliferation was observed in response to antigen fractions with molecular masses of 43 to 36, 33 to 27, and less than 22 kDa. The induction of relatively high levels of gamma interferon (IFN-gamma) and tumor necrosis factor alpha (TNF-alpha) was also observed in responses to these same three antigen fractions. In contrast, the proliferative, IFN-gamma, and TNF-alpha responses of patient lymphocytes to antigens with a molecular mass greater than 60 kDa were uniformly low. The results of this study suggest that the antigens of Leishmania parasites, which are recognized by T cells in patients with active cutaneous leishmaniasis, may be partitioned in the lower-molecular-mass antigenic determinants associated with whole-parasite preparations. The observed association between antigen-induced proliferation and IFN-gamma and TNF-alpha production may be indicative of potential disease-limiting immune effector activities which have developed during infection.

Regulatory factor-X binding to mutant HLA-DRA promoter sequences

The class II genes of the major histocompatibility complex (MHC) encode a family of cell surface glycoproteins that present processed antigen to the T cell receptor. Class II genes are regulated coordinately, responding to both immunologic and developmental signals. Conserved sequence elements 5' to class II genes have been shown to be important in transcriptional control. One of these sequences, the X box, is a specific target for the binding of the factor RF-X. In the hereditary HLA class II deficiency, a form of primary immunodeficiency, a regulatory defect in expression of class II genes is associated with a defect in the binding of RF-X. To determine the basepairs that are important for this binding interaction, a series of single basepair substitutions spanning the X box motif of the DRA gene was constructed and tested for binding of RF-X by gel electrophoresis mobility shift assays (EMSAs). Several, but not all, of the mutants severely affected binding of RF-X. In addition, the binding of both the natural and the recombinant form of RF-X was affected with the same specificity. A comparison of X box basepair positions important for RF-X binding to DRA with sequences conserved between X boxes of other class II alpha chain genes suggests that high affinity RF-X binding is important for a high level of expression and may explain differences in the levels of class II expression of different class II alpha chains.

An isotype-specific trans-acting factor is defective in a mutant B cell line that expresses HLA-DQ, but not -DR or -DP

The B lymphoblastoid cell line clone 13 (a subclone of the mutant cell line P3JHR-1) has been found to express high levels of HLA-DQ; by contrast, HLA-DR and -DP antigens are not expressed and cannot be induced by interferon gamma. Northern blot analysis using gene-specific probes indicated that the lack of surface expression of the DR and DP antigens is due to a marked decrease in the levels of steady-state RNA for both the alpha and beta chains. Southern blots demonstrated that none of the transcriptionally repressed genes are grossly deleted. Preparations of interspecific transient heterokaryons between clone 13 and the class II antigen-positive murine B cell lymphoma, A20, resulted in reactivation of the DRA gene and surface expression of both the DR and DP molecules. The efficiency of the DRA promoter relative to the DQB promoter is markedly and specifically diminished in clone 13 (P3JHR-1) as compared with the parental cell line, Jijoye, as assayed both by transient expression of appropriate chloramphenicol acetyltransferase gene (CAT) constructs and by in vitro transcription analysis. These data clearly demonstrate the existence of an isotype-specific trans-acting factor, and provide direct evidence that the highly homologous class II genes have distinct regulatory mechanisms.

Autoimmunity in periodontal disease

Periodontal disease in characterized by the loss of the normal supporting tissues of the teeth and a humoral and cellular immune response to bacterial antigen of dental plaque which accumulates at the dento-gingival junction. This review considers the evidence for the existence of an autoimmune component of the host immune response, the possible origin of such a response and the way in which such a host response may contribute to the changes observed in the periodontium in the disease.

Expression of class I and II human leukocyte antigens by thyrocytes and lymphocytic infiltration on human thyroid tumors. An immunofluorescence study

Surgical thyroid sections from 30 papillary carcinomas (PC), six medullary carcinomas (MC), three anaplastic carcinomas (AC), two follicular carcinomas (FC), and 16 adenomas (AD) were examined with an indirect immunofluorescence technique employing different monoclonal antibodies to evaluate the expression of human leukocyte antigen (HLA)-A, B, C (Class I) and DR, DP, DQ (Class II) by thyrocytes, together with the phenotype and distribution of inflammatory cells. Ten PC and four FC were also investigated for the presence of intercellular adhesion molecule-1 (ICAM-1). In situ deposits of immunocomplexes and circulating thyroid autoantibodies were also evaluated. An increased expression of Class I antigens was found in all PC and FC, in 33% of MC and AC, and in 31% of AD. An anomalous expression of Class II antigens was observed in 70% of PC, in 50% of FC, in 33% of AC, in 19% of AD, and in none of the MC. Expression of DP or DQ was revealed only in a portion of the DR-positive glands. A reduction of microsomal autoantigen expression was found. No ICAM-1-positive thyrocytes were detected. A moderate T-lymphocytic infiltrate was noticed only in PC, where it was correlated with DR and DP and/or DQ coexpression. B-cells and natural killer cells were virtually absent. The authors speculate that the weak Class II antigens expression, together with the partial or complete loss in microsomal autoantigen and the absence of ICAM-1 by thyrocytes, may account for the limited engagement of immunocompetent cells observed in thyroid tumors.

Regulation of MHC molecules on MBP positive oligodendrocytes in mice by IFN-gamma and TNF-alpha

The expression of class I and class II histocompatibility antigens by myelin basic protein (MBP)-positive oligodendrocytes, in response to exogenous cytokines, has been investigated in vitro. It has been found that interferon-gamma (IFN-gamma), although capable of class I induction, does not induce class II on oligodendrocytes. Furthermore, tumour necrosis factor-alpha (TNF-alpha), which was shown to induce class I MHC on other neural cells, failed to induce class I on oligodendrocytes. A combination of IFN-gamma and TNF-alpha also failed to facilitate the expression of class II antigens on oligodendrocytes, nor did it amplify the expression of class I seen with IFN-gamma alone. Thus it appears that MBP+ murine oligodendrocytes are refractory to class II induction, and express class I in response to IFN-gamma but not TNF-alpha. The differential regulation and class of MHC expression may have implications in terms of the initiation and targeting of immune responses directed toward the oligodendrocyte.

Essential role for interferon-gamma and interleukin-6 in autoimmune insulin-dependent diabetes in NOD/Wehi mice

Experimental studies in vitro suggest that cytokines are important mediators in the pathogenesis of autoimmune insulin-dependent diabetes mellitus (IDDM). However, there is little evidence for the role of cytokines in vivo, either in humans or in the spontaneous animal models of IDDM such as the NOD mouse or BB rat. To address this question, we used the model of cyclophosphamide (CYP)-induced autoimmune diabetes in the NOD/Wehi mouse to examine for (a) the production of IFN-gamma and IL-6 from isolated islets, and (b) the effect of anti IFN-gamma or anti IL-6 monoclonal antibodies on the development of diabetes. After cyclophosphamide, the majority of these mice develop of mononuclear cell infiltrate (insulitis) which by 10-14 d is associated with beta cell destruction. IFN-gamma activity at low levels (2.7 +/- 0.3 U/ml) could be detected only in culture supernatants from islets isolated at day 7 post-cyclophosphamide. In contrast, IL-6 activity progressively increased from 457 +/- 44 U/ml at day 0 to 6,020 +/- 777 U/ml at day 10. Culture of islets with anti-CD3 monoclonal antibody resulted in a significant increase in IFN-gamma activity from 41 +/- 7 U/ml at day 0 to 812 +/- 156 U/ml at day 10. Mice given either anti-IFN-gamma or anti-IL-6 antibody had a significantly reduced (P less than 0.001) incidence of diabetes and especially with IFN-gamma, decreased severity of insulitis. We conclude that IFN-gamma and IL-6 have essential roles in the pathogenesis of pancreatic islet beta cell destruction in this model.

Circulating tumor necrosis factor, interleukin-1 and interleukin-6 concentrations in chronic alcoholic patients

Although altered cytokine homeostasis has been implicated in the pathogenesis of alcoholic liver disease, the relationship between cytokines and metabolic consequences of alcoholic liver disease is unknown. We, therefore, sought to correlate circulating concentrations of tumor necrosis factor-alpha, interleukin-1 and interleukin-6 to clinical and biochemical parameters of liver disease in chronic alcoholic patients. We used an enzyme-linked immunosorbent assay to measure plasma tumor necrosis factor and interleukin-1 and a bioassay to measure serum interleukin-6 in three groups of alcoholic men as follows: (a) actively drinking alcoholic men without evidence of chronic liver disease, (b) nondrinking alcoholic men with stable cirrhosis and (c) patients with acute alcoholic hepatitis. Mean cytokine concentrations were elevated in cirrhotic patients and alcoholic hepatitis patients compared with controls and alcoholic patients without liver disease. Tumor necrosis factor-alpha and interleukin-1 alpha concentrations remained elevated for up to 6 mo after diagnosis of alcoholic hepatitis, whereas interleukin-6 normalized in parallel with clinical recovery. Concentrations of all three cytokines were correlated with biochemical parameters of liver injury and hepatic protein synthesis plus serum immunoglobulin concentrations. We could not demonstrate a relationship between cytokine concentrations and peripheral endotoxemia. Percentages of peripheral blood monocytes that reacted with monoclonal antibodies to CD25 (interleukin-2 receptor) and human lymphocyte antigen-DR were similar for alcoholic patients and controls. These data suggest that tumor necrosis factor-alpha and interleukin-1 alpha are related to some of the metabolic consequences of both acute and chronic alcohol-induced liver disease, whereas interleukin-6 is related to abnormalities seen in acute liver injury.

Cytokine assays: role in evaluation of the pathogenesis of autoimmunity

Cytokines are protein mediators involved in inflammation, the immune response, cell growth, repair and fibrosis. All of these processes are ongoing in active autoimmune diseases such as rheumatoid arthritis (RA), and so it would be expected that many cytokines would be actively produced in RA joints or Graves' disease (GD) thyroid glands. The cDNA cloning of cytokines has permitted the generation of pure recombinant molecules, and of newer more sensitive assays, and spurred the rapid development of knowledge in this field. Here we review the molecular strategies devised to study the possible role of cytokines in the pathogenesis of RA and GD, and describe some of the initial results. After 'cataloguing' the relative abundance of various cytokines, we sought to discover which cytokines are of major importance in pathogenesis. For that purpose we used neutralizing anti-cytokine antibodies and found that TNF alpha is one of the major signals regulating the production of IL-1 in the RA but not in the osteoarthritic (OA) joint. In order to further understand the dynamics of the cytokine network, the localization of the cytokine-producing cells by immunostaining and in situ hybridization has also been performed. The latter techniques are particularly valuable for attempting to establish the role of the target cell, such as thyroid epithelium, in the pathogenesis of disease. Cytokines act on cells via binding to high-affinity receptors. The last two years has been the cDNA cloning of many molecules encoding cytokine receptor chains, and it is now possible to begin to evaluate the other half of the cytokine pathway. Taken together, there are now exciting opportunities for the molecular dissection of the cytokine events occurring in auto-immune tissues.

Expression of intercellular adhesion molecule-1 (ICAM-1) on thyroid epithelial cells in Hashimoto's thyroiditis but not in Graves' disease or papillary thyroid cancer

In order to study the possible role of antigen-independent adhesion systems in thyroid autoimmunity, we evaluated by indirect immunofluorescence the expression of lymphocyte functional antigen-1 (LFA-1) and its ligand ICAM-1 on mononuclear cell infiltrates (when present) and thyroid follicular cells of four patients with Hashimoto's thyroiditis, 30 with Graves' disease, five with papillary cancer, two with follicular adenoma, and two normal thyroid specimens. The expression of MHC class I and class II antigens was also evaluated. Most mononuclear infiltrates were LFA-1 positive, as expected. A positivity for ICAM-1 on follicular cells was observed in three out of four Hashimoto's thyroiditis specimens; such a phenomenon was totally absent in Graves' disease or any other pathological condition, or in normal tissue. MHC class II expression on thyrocytes was observed in all the patients with Hashimoto's thyroiditis, in 27 out of 30 with Graves' disease and in three out of five papillary cancer specimens.

Lymphotoxin produced by human B- and T-cell lines appears in two distinct forms

Production and release of lymphotoxin (LT) was studied by metabolic labeling of human B- and T-cell lines with 14C-leucine and 35S-methionine. LT was immunoprecipitated with antiserum to LT and separated by sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis (PAGE) followed by fluorography. Two molecular weight forms of LT with different rates of release were found both in cell supernatants and cell extracts. Monensin, a sodium ionophore, inhibited the release of LT. LT still appeared in two molecular weight forms after deglycosylation with N-glycanase. Treatment of cells with swainsonine followed by digestion of released LT with endoglycosidase H (endo H) demonstrated that the oligosaccharides were of the complex type. Subcellular fractionation of cells on Percoll density gradients demonstrated that intracellular LT is located to intermediate density fractions. No LT was found in the high density fractions corresponding to lysosomes. Phorbol 12-myristate 13-acetate induced production of tumor necrosis factor (TNF) in the B-lymphoblastoid cell line RPMI-1788. In conclusion, we have demonstrated the presence of two distinct molecular weight forms of LT, which contain N-linked oligosaccharides of the complex type.

Soluble HLA-DR molecules in patients with HLA class II versus class I associated disorders

HLA genes have been identified as key genetic factors contributing to many chronic diseases characterized by autoimmune features. The role of HLA encoded molecules in the pathogenesis of these diseases is unresolved. We have now analysed soluble HLA-DR molecules circulating in the serum of patients with different autoimmune diseases and have defined parameters controlling serum levels. Patients with HLA-DR associated diseases were characterized by elevated serum concentrations of HLA-DR molecules and were clearly distinct from patients with HLA-B27 associated disorders. We did not find evidence for a correlation between disease activity, laboratory abnormalities and elevated serum concentrations of soluble HLA-DR molecules. Studies in normal individuals indicated that soluble HLA-DR molecules are at least partially regulated by the HLA haplotype. Highest serum concentrations were found in individuals carrying the HLA-DR3 or HLA-DR4 haplotype raising the possibility that the phenomenon of HLA-disease association reflects differences in the genetic control of soluble HLA-DR molecules. Interferon-gamma treatment caused an increase in serum concentrations of soluble HLA-DR molecules, whereas a decrease of circulating HLA-DR molecules was associated with an immunosuppressive with cyclosporine A. These data suggest that the patient's immunoresponsiveness represents a second important mechanism controlling circulating HLA-DR molecules.

Enhanced percentage of Leu M3+DR+ and Leu M3+CD25+ cells in newly diagnosed IDDM patients

The percentage of Leu M3+DR+ and of Leu M3+CD25+ cells was determined by means of immunofluorescence analysis in a group of patients with insulin dependent diabetes mellitus (IDDM). Our results show that an increased percentage of these cells may occur in the early stage of the disease. These data provide evidence for a "phenotypical" activation of Leu M3+ cells at the onset of the disease and warrant future studies to evaluate the potential role of these cells in the pathogenesis of IDDM.

Cytokine regulation of HLA on thyroid epithelial cells

The regulation of class I and class II HLA expression in human thyroid follicular cells was studied in vitro. Tumour necrosis factor-alpha (TNF-alpha) enhanced the expression of class I antigen on thyrocytes, but these cytokines had little effect on the expression of class II antigen. Interleukin-1 beta (IL-1 beta) and interleukin-6 (IL-6) did not affect class I and class II antigen expression. The combination of interferon-gamma (IFN-gamma) with TNF-alpha or IL-1 beta enhanced the induction of class I and class II antigens, compared with the effect of IFN-gamma alone. Neither class I nor class II expression was induced by IL-6 alone or in combination with IFN-gamma. These findings suggest that TNF-alpha and IL-1 beta may have an important role in inappropriate expression of HLA antigens on thyrocytes in thyroid gland.

The effect of HLA and insulin-dependent diabetes mellitus on the secretion levels of tumour necrosis factors alpha and beta and gamma interferon

Tumour necrosis factors alpha and beta (TNF-alpha and TNF-beta) and gamma interferon (IFN-gamma) were measured by ELISA in the supernatants of phytohaemagglutinin (PHA)-activated peripheral blood mononuclear cells (PBMNC) from 98 individuals (60 controls and 38 patients with insulin-dependent diabetes mellitus [IDDM]). The PBMNC were incubated with varying concentrations of PHA (0, 1, 5, and 10 micrograms/ml) for 72 h. In our population study we observed a correlation between the levels of secretion of TNF-alpha and IFN-gamma but not TNF-beta. The complete data set was analysed by non-parametric tests, and no associations with HLA phenotypes existed. Reduced levels of TNF-beta, but not TNF-alpha or IFN-gamma, secretion were found in IDDM patients stimulated with 1 and 5 micrograms/ml of PHA (P = 0.001 and 0.02 respectively). None of the lymphokine secretion levels at any PHA concentration correlated with particular HLA phenotypes. Analysis of the natural log-transformed data indicated that only for the TNF-beta levels (at 5 micrograms/ml PHA) could subjects be divided into high and low secretors, which also did not correlate with a particular HLA-B or -DR antigen.