Search

In order to explore complement-dependent humoral immunity in insulin-dependent diabetic patients, their serum was incubated with Mouse isolated pancreatic islets in the presence of complement, after which the response of these islets to glucose was estimated by measuring insulin secretion. Out of 13 sera from insulin-dependent diabetic patients, 5 caused a significant inhibition of the insulin response. Morphological observation of islets suggested that this inhibition was due to a cytotoxic effect of the sera. In the absence of complement, the same sera were not able to suppress insulin release. No beta-cell inhibition or morphological damage was detectable on islets incubated with sera from 13 healthy subjects and 4 non-insulin dependent diabetic patients, in the presence or in the absence of complement. Complement-dependent humoral immunity may play a role in the beta-cell injury of insulin-dependent diabetes.

Increasingly, experimental results are underlining the role played by autoimmune mechanisms in the pathogenesis of type I insulin-dependent diabetes mellitus (IDDM). It has appeared logical to attempt to preclude the onset of IDDM by suppressing immune responses, but trials using steroids or azathioprine were unequivocal. Subsequently, a Canadian and a French group performed pilot studies to assay a new immunosuppressive drug, cyclosporine A (CyA) in human diabetes (Stiller et al., 1984; Assan et al., 1985). However, further testing is required to evaluate and confirm its potential benefits and possible risks.

Mice peritoneal macrophages obtained after i.p. injection of a medium containing thioglycollate (Institut Pasteur, Paris) have increased levels of beta glucuronidase and acid phosphatase, when compared with macrophages obtained without previous thioglycollate injection. The maximum increase is observed between 7 and 10 days after injection and reaches up to 3 to 4 times the normal activity. When using separately the various components of the medium, the same phenomenon may be observed but never to the same degree. The injection of thioglycollate also produces: a) an increase in total number of peritoneal cells, and b) an increase in the percentage of macrophages among these cells. Kinetic studies suggest that thioglycollate injection first induces an increase in the number of peritoneal macrophages and subsequently macrophage modifications resulting in lysosomal enzyme increase.

Macrophage-rich adherent peritoneal cells from mice sensitized by two i.p. injections of allogeneic cells at days -- 12 and --2, were found non-specifically cytotoxic against (a) tumoral target cells, as evaluated by the chromium release test, and (b) normal target cells, as evaluated by a functional test based on the insulin secretion response to glucose stimulation of isolated pancreatic islet cells. This non-specific cytotoxicity was identical to that of macrophages "activated" by an in vitro incubation with supernatants of MLC between allograft donor and recipient as reported elsewhere. Moreover this cytotoxicity was not abolished by anti-theta serum treatment of adherent peritoneal cells before contact with target cells. It is concluded that an in vivo macrophage activation was obtained in our experiment. Using this same model, we found that the survival time of skin grafts from male C57BL/6 to female C57BL/6 mice was shortened if the graft was performed at the time when the recipient macrophages were found non-specifically cytotoxic (two days after the second allogeneic cell injection). The role of macrophage cytotoxicity in this accelerated allograft rejection is suggested by (a) the lack of any detectable cross-histocompatibility antigens between the immunizing cells and the skin graft, and (b) the absence of any alteration of the graft survival if the second immunizing injection two days prior to grafting is omitted (in that case only lymphocytes are found specifically cytotoxic against the strain of injected cells, while the non-specific cytotoxicity of macrophages is lacking).

The non-obese diabetic (NOD) mouse is an animal model of insulin-dependent diabetes mellitus (IDDM), in which 80% of the females become diabetic after the age of 12 weeks. Using an in vitro assay we investigated the capacity of spleen lymphocytes from NOD mice to inhibit the insulin secretion of normal islet cells after stimulation by theophylline plus arginine. Spleen cells from diabetic NOD mice inhibited the insulin release of DBA/2 islet cells. Depletion experiments using monoclonal antibodies demonstrated that inhibitory cells belonged to the Lyt2 positive T lymphocyte subset. The phenomenon was not restricted by the MHC class I K region, shared by NOD and DBA/2 mice, since lymphocytes from diabetic NOD mice also inhibited the insulin secretion of normal Wistar rat islet cells. Inhibitory T cells were detected in overtly diabetic mice but also in non-diabetic females aged 5-11 weeks indicating that they are not secondary to metabolic disturbances and might contribute to their onset. Conversely they were not found in male NOD mice although some of these mice show insulitis. The presence of these inhibitory T cells might thus represent an early and sensitive marker of anti-islet cell-mediated autoimmunity.

It has been suggested that the immune system may be responsible for the destruction of insulin secreting cells in some types of diabetes. In order to test this hypothesis, we studied the consequences of immune-mediated reactions on the function of pancreatic islet cells in vitro. A model was set up in vitro where mouse pancreatic islet cells are exposed to human lymphocytes or sera + complement then stimulated for the release of insulin or glucagon. A selective inhibition of insulin secretion, but not of glucagon secretion, was observed in the presence of lymphocytes from 37 out of 40 insulin-dependent diabetic (IDD) patients and in the presence of sera (+ complement) from 22 out of 40. Lymphocytes were found inhibitory in almost all patients in both groups, with and without associated autoimmune diseases. In contrast, inhibitory sera were observed almost only in patients with associated autoimmune diseases or recent onset diabetes. The selective inhibition of insulin secretion, but not of glucagon secretion, suggests that lymphocytes or sera may be involved in a destructive process of insulin secreting cells in vivo. This cell-mediated effect depends on direct T lymphocyte cytotoxicity, rather than antibody-dependent cell cytotoxicity, as suggested by the lack of any effect of aggregated immunoglobulins on the reaction. In contrast, when C57BL/6 mice were immunized by mastocytoma cells from a DBA2 strain, their lymphocytes and sera blocked both secretions of insulin and glucagon when incubated in vitro with DBA2 islet cells. This non-selective inhibition may be due to anti-H2 immunity, rather than immunity directed against insulin secreting cells.

To explore humoral immunity in insulin-dependent diabetic (IDDM) patients, we studied insulin release from isolated mouse islets stimulated by glucose + theophylline after incubation with the sera of these patients and complement. Eleven of 21 IDDM sera suppressed the stimulated insulin release while the arginine-stimulated glucagon release remained unchanged. Morphologic evidence and the trypan-blue exclusion test suggested that the suppression of insulin release was due to a cytotoxic effect of the sera. No beta-cell inhibition of morphologic damage was detectable in the presence of sera from 30 healthy subjects, 8 non-insulin-dependent diabetic patients, and 5 nondiabetic patients with autoimmune diseases. Beta-cell inhibition by IDDM sera was not observed when complement was omitted. After serum fractionation, the cytotoxic potency of IDDM sera was located in the immunoglobulin G fraction. Using human islets, insulin release was suppressed by 3 of 6 IDDM sera. Complement-dependent cytotoxicity was found in 1 of 5 recent-onset IDDM patients and 11 of 16 IDDM patients with autoimmune phenomena. It was associated in all cases with the presence of islet cell antibodies as detected by immunofluorescence, and with the presence of circulating lymphocytes which suppressed insulin release in vitro. Complement-fixing antibodies may contribute to the selective beta-cell damage in IDDM.

Anti-islet immunity was studied in six spontaneously insulin-dependent diabetic (IDD) dogs, using mouse islets of Langerhans cells as targets, in vitro. Insulinopenia was demonstrated in all dogs by an i.v. glucose tolerance test. A significant lymphocytopenia was detected in the peripheral blood of this diabetic group. Pancreatic tissue from one of these animals was obtained shortly after death and the islets displayed a marked loss in beta cells without significant changes in the other types of islet cells. No insulitis was observed. Circulating mononuclear cells from the diabetic dogs induced an increased basal insulin (IRI) release from islet cells and a suppressed stimulated IRI release. Damage to or depth of beta cells may account for these findings. The stimulated IRI release was also suppressed when islets were incubated with the diabetic sera + complement, while the D-cell response to arginine was not altered, and the A-cell response was reduced but not abolished. A lysis of islet cells in the presence of IDD sera + complement was demonstrated by an increased release of 51Cr from labeled cells. These anomalies were observed neither when complement was heat-inactivated nor in the presence of control sera + complement. Canine IDD may be a new animal model for the study of anti-islet cellular and humoral immunities.

Monoclonal autoantibodies cytotoxic for murine islet cells have been obtained by cell fusion using spleen cells from non-immunized diabetic C57BL/KsJ (db/db) Mice. These monoclonals, of IgG2a subclass, blocked in vitro insulin secretion induced by arginine in Mouse pancreatic cells.

We previously showed that circulating lymphocytes from more than 90% of insulin-dependent diabetics, block extra insulin secretion induced by stimulatory media in mouse pancreatic cells in vitro, without altering the secretion of glucagon. The present work demonstrates that this phenomenon depends on lymphocytes having the OKT3 marker, i.e., thymodependent lymphocytes. However, OKT4+ T helper cells are not required for the above phenomenon as proved by experiments using monoclonal sera against the OKT4 marker. When diabetes is associated with other autoimmune diseases, the pancreatic lymphocyte cytotoxicity observed in vitro is inhibited by the addition of a normal lymphocyte population; this could indicate that a "suppressor" factor is lacking in these patients. Conversely, addition of normal lymphocytes does not prevent lymphocyte cytotoxicity in diabetics without associated autoimmune diseases. Such a difference confirms the present trend to make a distinction between these two categories of diabetes.

The diabetic db/db mice of the C57 BL/KsJ strain display anti-islet immunity, thymic dysfunction, and lymphopenia. In the present work, lymphocytes, T-cells, and T-cell subsets were enumerated in thymus and spleen from diabetic db/db mice and their db/ + heterozygote littermates from the 10th day to the 10th month of life. A significant lymphopenia was detected in thymus and spleen from the second month on, involving specifically the T-cell compartment, as assessed by use of a monoclonal anti-Thy1 antibody in indirect fluorescence. The study of T-cell subsets by monoclonal anti-Lyt1 and anti-Lyt2 antibodies revealed a significant increase in Lyt1+ cells and a decrease in Lyt2+ cells, with a corresponding increase of the Lyt1+/Lyt2+ ratio. These anomalies appeared early in life, and were apparently linked neither with the degree of hyperglycemia nor with weight loss or infection. The T-cell depletion in thymus was more pronounced in young male (less than 3 mo) than in young female db/db mice. These alterations may correspond to an increase in the helper/suppressor-cytotoxic ratio and could be linked with the thymic anomalies present in these mice, contributing to the development of anti-islet autoimmunity.

The anti-pancreatic immune reaction of genetically diabetic homozygote C57Bl/KsJ db/db mice was studied with an in vitro test using murine islet of Langerhans cells as target cells. C57Bl/KsJ db/db spleen lymphocytes inhibited insulin secretion by the islet cells. This inhibition was abolished when T cells were eliminated by treatment with anti-Thy 1.2 monoclonal antibody in the presence of complement. Together with this cell-mediated cytotoxicity, complement-dependent antibody (CDA) and antibody-dependent cell cytotoxicity (ADCC) were found in the sera of these mice. A longitudinal study showed that this anti-pancreatic toxicity was detectable as early as the 10th day of life and lasted throughout the entire life span of the animal. None of these anomalies was found in control heterozygote mice.

The insulin response of isolated islet cells to glucose and theophylline in vitro was studied after incubation with lymphocytes. The test was employed to explore cell-mediated immunity in diabetics. A significant inhibition of insulin response to glucose and theophylline as compared to insulin release in a "basal" medium was found after incubation with blood lymphocytes from 21 out of 23 insulin-dependent diabetics (mean secretion index 18 +/- 18 versus 118 + 8 (SEM) % in control subjects). Most of the patients studied had associated autoimmune diseases: all of these displayed inhibition of insulin release. In six cases, the diabetes had a recent onset with no associated autoimmune disease: four of them displayed the same inhibition. No inhibition was found in the 26 control subjects and in seven non-insulin-dependent diabetics (mean secretion index 134 +/- 17 versus 145 +/- 23 (SEM) % in four control subjects). Lymphocytes inhibiting islet cell response were not cytotoxic against mouse fibroblasts. Twenty-two insulin-dependent diabetics showed islet cell antibodies to human and/or mouse pancreatic islets. However, an inhibition of insulin release was found with no detectable islet cell antibodies in one case, and the converse in two cases. Lymphocyte cytotoxicity to islet cells could play a role in the natural history of insulin-dependent diabetes.

Antilymphocyte sera (ALS), either allogeneic (for example, Lewis anti-August rat serum) or xenogeneic (the classic ALS), may retain the antigen used for the immunization of the animal producing the serum in the form of antigen-antibody complexes, at least when the animal is bled within the first 2 weeks following the last immunizing injection. Evidence for this has been obtained in vivo and in vitro. In vivo experiments showed that injection of Lewis anti-August serum to a normal Lewis rat can induce a delayed active immunization against the August strain. In vitro studies proved the presence of August antigens in the anti-August sera, by precipitation of the antigen-antibody complexes with 6 M urea, removal of the antibody by filtration, and demonstration that the filtrate specifically inhibits complement-dependent cytotoxicity of anti-August antibodies. The interpretation of a number of previous works on so-called passive enhancement of grafts or on various effects of ALS may have to be reconsidered on the basis of these findings.

Anti-beta-cell-specific cell-mediated immunity was studied over a 12-mo period in 65 recently diagnosed diabetic patients randomly receiving either cyclosporin or placebo. Anti-beta-cell cellular immunity was assessed by an in vitro test based on the inhibition of insulin release from cultured rat islet cells by patients' mononuclear cells. This beta-cell-suppressive effect disappeared in cyclosporin A-treated patients within 1 mo and did not reappear during 12 mo of follow-up. Conversely, the suppressive effect persisted unchanged in placebo-treated patients during 12 mo of follow-up. These changes were predictive neither of cyclosporin A-induced remission nor of relapses. Results of the insulin-release inhibition test were not correlated to islet cell autoantibodies or HLA phenotype.

Anti-islet cell and anti-insulin antibody production was studies over a 12-mo period in 82 recently diagnosed diabetics randomly receiving either cyclosporin or placebo. Cyclosporin had only minimal effects on the production of anti-islet cell antibodies whether directed to islet cytoplasmic (immunofluorescence) or membrane (cytotoxicity assay) antigens even in patients undergoing remission. These data suggest that these antibodies do not play a major role in the pathogenesis of the disease particularly since their (irregular) presence is not predictive of the clinical response to cyclosporin. Conversely, cyclosporin completely suppressed the synthesis of antibodies elicited by exogenous insulin irrespective of the insulin doses received, and decreased the autoantibody production against thyroid antigens, indicating that cyclosporin has variable effects on antibody production against various antigens.

This paper reviews the rosette inhibition test for the evaluation of antilymphocyte serum potency and reappraises the biological significance of the assay. Evidence is also presented which indicates the presence, in antilymphocyte serum, of immune complexes.

The administration of colchicine to dysmyelinating mutant mice may serve as an in vivo pharmacological tool for the study of the mechanisms involved in the formation of the myelin sheath. The study of the acute toxicity of colchicine in these mutants demonstrated that male animals were much more sensitive than female animals. All of the mutants and their controls were also more resistant to colchicine than the Swiss strain usually used in toxicity studies.

Anti-islet immune reactions were studied in vitro in genetically diabetic homozygote C57BL/KsJ db/db mice, using murine islet cells as a target. Spleen lymphocytes inhibited insulin secretion by the islet cells. This inhibition was abolished when T-cells were eliminated by treatment with anti-Thy 1.2 monoclonal antibody in the presence of complement. Anti-islet complement-dependent antibody (CDA) and antibody-dependent cell cytotoxicity (ADCC) were also found in the sera of these mice. This anti-islet immunity was detectable as early as the tenth day of life and lasted throughout the entire life span of the animals. A significant lymphopenia was detected in thymus and spleen cell populations. None of these anomalies was found in control heterozygote mice. Thymic function was explored in the same mice by evaluating their serum thymic factor (FTS) levels using a rosette assay. The age-dependent decline of FTS levels was significantly accelerated in diabetic mice as compared with heterozygous littermates. Furthermore, FTS inhibitory immunoglobulins were detected in db/db mouse sera, which inactivated in vitro the biologic potency of synthetic FTS. Histologically, the thymus displayed an accelerated involution. It was shown by indirect immunofluorescence using anti-FTS monoclonal antibodies that the number of FTS+ cells was reduced in db/db mouse thymuses. Histologic study of the islets of Langerhans showed early signs of beta-cell hyperactivity and hypertrophy, followed by beta-cell rarefaction and profound dislocation of islet architecture. Insulitis was not detected.

Hyperglycaemia in the afternoon is often not prevented by increasing the intermediate insulin dose in C-peptide deficient Type 1 diabetic patients, particularly children, treated with two injections of mixtures of short- and intermediate-acting insulins. To provide a better understanding of this, 19 Type 1 diabetic patients were studied using the euglycaemic clamp technique, after SC injection of either soluble, isophane (NPH) or lente insulin, or mixtures of these preparations. SC injection of 0.6 U kg-1 insulin, as either intermediate-acting (isophane (NPH) or lente) insulin alone or as a mixture of 50% soluble + 50% intermediate-acting insulins, indicates that, for the same total dose of SC insulin, a shift from soluble to intermediate-acting insulin results in a major loss of activity during the first 6 h after injection, with no compensatory increase over the next 6 h. The morning hypoglycaemic activity depends on different mechanisms with isophane (NPH) or lente insulin; the effects of soluble and isophane (NHP) insulin are superimposed, while with lente insulin, the peak of soluble insulin is blunted. For the afternoon, even large doses of isophane (NPH) or lente insulin do not meet the peak of insulin requirements following lunch.