Beta-cell gene expression and functional characterisation of the human insulinoma cell line CM.

Animal insulinoma cell lines are widely used to study physiological and pathophysiological mechanisms involved in glucose metabolism and to establish in vitro models for studies on beta-cells. In contrast, human insulinoma cell lines are rarely used because of difficulties in obtaining and culturing them for long periods. The aim of our study was to investigate, under different experimental conditions, the capacity of the human insulinoma cell line CM to retain beta-cell function, particularly the expression of constitutive beta-cell genes (insulin, the glucose transporters GLUT1 and GLUT2, glucokinase), intracellular and secreted insulin, beta-cell granules, and cAMP content. Results showed that CM cells from an early-passage express specific beta-cell genes in response to glucose stimulation, in particular the insulin and GLUT genes. Such capacity is lost at later passages when cells are cultured at standard glucose concentrations. However, if cultured at lower glucose concentration (0.8 mM) for a longer time, CM cells re-acquire the capacity to respond to glucose stimulation, as shown by the increased expression of beta-cell genes (insulin, GLUT2, glucokinase). Nonetheless, insulin secretion could not be restored under such experimental conditions despite the presence of intracellular insulin, although cAMP response to a potent activator of adenylate cyclase, forskolin, was present indicating a viable system. In conclusion, these data show that the human insulinoma cell line CM, at both early-passage and late-passage, posseses a functional glucose-signalling pathway and insulin mRNA expression similar to normal beta-cells, representing, therefore, a good model for studies concerning the signalling and expression of beta-cells. Furthermore, we have previously shown that it is also a good model for immunological studies. In this respect it is important to note that the CM cell line is one of the very few existing human beta-cell lines in long-term culture.

Establishment of T cell lines to bovine beta-casein and beta-casein-derived epitopes in patients with type 1 diabetes.

Enhanced cellular immune response to bovine beta-casein has been reported in patients with type 1 diabetes. In this study we aimed to establish beta-casein-specific T cell lines from newly diagnosed type 1 diabetic patients and to characterise these cell lines in terms of phenotype and epitope specificity. Furthermore, since sequence homologies exist between beta-casein and putative beta-cell autoantigens, reactivity to the latter was also investigated. T cell lines were generated from the peripheral blood of nine recent onset type 1 diabetic patients with different HLA-DQ and -DR genotypes, after stimulation with antigen pulsed autologous irradiated antigen presenting cells (APCs) and recombinant human interleukin-2 (rhIL-2). T cell line reactivity was evaluated in response to bovine beta-casein, to 18 overlapping peptides encompassing the whole sequence of beta-casein and to beta-cell antigens, including the human insulinoma cell line, CM, and a peptide from the beta-cell glucose transporter, GLUT-2. T cell lines specific to beta-casein could not be isolated from HLA-matched and -unmatched control subjects. beta-Casein T cell lines reacted to different sequences of the protein, however a higher frequency of T cell reactivity was observed towards the C-terminal portion (peptides B05-14, and B05-17 in 5/9 and 4/9 T cell lines respectively). Furthermore, we found that 1 out of 9 beta-casein-specific T cell lines reacted also to the homologous peptide from GLUT-2, and that 3 out of 4 of tested cell lines reacted also to extracts of the human insulinoma cell line, CM. We conclude that T cell lines specific to bovine beta-casein can be isolated from the peripheral blood of patients with type 1 diabetes; these cell lines react with multiple and different sequences of the protein particularly towards the C-terminal portion. In addition, reactivity of beta-casein T cell lines to human insulinoma extracts and GLUT-2 peptide was detected, suggesting that the potential cross-reactivity with beta-cell antigens deserves further investigation.

Beta-cell markers and autoantigen expression by a human insulinoma cell line: similarities to native beta cells.

In the present study we have evaluated the expression of different beta-cell markers, islet molecules and auto-antigens relevant in diabetes autoimmunity by a human insulinoma cell line (CM) in order to define its similarities with native beta cells and to discover whether it could be considered as a model for studies on immunological aspects of Type 1 diabetes. First, the positivity of the CM cell line for known markers of neuroendocrine derivation was determined by means of immunocytochemical analysis using different anti-islet monoclonal antibodies including A2B5 and 3G5 reacting with islet gangliosides, and HISL19 binding to an islet glycoprotein. Secondly, the expression and characteristics of glutamic acid decarboxylase (GAD) and of GM2-1 ganglioside, both known to be islet autoantigens in diabetes autoimmunity and expressed by human native beta cells, were investigated in the CM cell line. The pattern of ganglioside expression in comparison to that of native beta cells was also evaluated. Thirdly, the binding of diabetic sera to CM cells reacting with islet cytoplasmic antigens (ICA) was studied by immunohistochemistry. The results of this study showed that beta cell markers identified by anti-islet monoclonal antibodies A2B5, 3G5 and HISL-19 are expressed by CM cells; similarly, islet molecules such as GAD and GM2-1 ganglioside are present and possess similar characteristics to those found in native beta cells; the pattern of expression of other gangliosides by CM cells is also identical to human pancreatic islets; beta cell autoantigen(s) reacting with antibodies present in islet cell antibodies (ICA) positive diabetic sera identified by ICA binding are also detectable in this insulinoma cell line. We conclude that CM cells show close similarities to native beta cells with respect to the expression of neuro-endocrine markers, relevant beta cell autoantigens in Type 1 diabetes (GAD, GM2-1, ICA antigen), and other gangliosides. Therefore, this insulinoma cell line may be considered as an ideal model for studies aimed at investigating autoimmune phenomena occurring in Type 1 diabetes.

T cell reactivity to human insulinoma cell line (CM) antigens in patients with type 1 diabetes.

Autoimmune (type 1) diabetes mellitus results from a progressive destruction of insulin secreting beta cells operated by T lymphocytes in pancreatic islets. Circulating autoreactive T cells to specific beta cell antigens are detected in patients with type 1 diabetes. To date, several beta cell autoantigens have been identified in this disease (GAD, IA-2, 38kD secretory protein, insulin, ICA69 etc.), however, it is possible that also other unidentified self molecules contribute to trigger beta cell autoimmunity. In this study we used the human insulinoma cell line CM as source of beta cell antigens to detect reactive T lymphocytes in patients with type 1 diabetes mellitus. This cell line has been previously shown to express a number of recognized beta cell antigens. Since the expression of several beta cell antigens is affected by glucose stimulation we tested two preparations of CM cells cultured under different conditions containing low (0.8 mM) and high glucose concentration (11 mM). T cell proliferation was measured using cells from 32 patients with type 1 diabetes (19 of recent onset and 13 at 3 to 22 months from diagnosis) and 27 age-matched control subjects. A significant increase in T cell proliferation to CM cells grown in high glucose conditions (11 mM) (p < 0.05) was found in type 1 diabetic patients compared to controls. No significant differences were observed when using CM cells cultured at the low glucose concentration. Furthermore, the response to both extracts of CM cells was independent of disease duration (p = 0.6 for both CM cells cultured at 0.8 and 11 mM glucose). These data indicate that T cell reactivity to homogenates of CM cells is detectable in patients with type 1 diabetes and suggest that this human insulinoma cell line is an interesting potential source of beta cell material for immunological studies of autoimmune diabetes.

A radiopharmaceutical for imaging areas of lymphocytic infiltration: 123I-interleukin-2. Labelling procedure and animal studies.

The labelling of interleukin-2 (IL-2) with 123I and its in vivo application for imaging chronic pathological lymphocytic infiltrations are described. The lactoperoxidase/glucoseoxidase technique was the labelling method of choice leading to immunoreactive IL-2 with high specific activity. Labelled IL-2 was injected in diabetes-prone non-obese diabetic (NOD) mice with pancreatic lymphocytic infiltration. As control animals, Balb/c mice were used. As specificity control, monoclonal antibodies AMT13 and UCHT1, bovine serum albumin and alpha-lactalbumin were radioiodinated and injected in mice. Eighteen NOD mice and four control Balb/c mice were used for gamma camera imaging experiments. Fifty-four NOD and 20 Balb/c mice were used for time course single organ counting and autoradiography. Gamma camera images showed that radioactivity accumulated in the pancreatic region from the 10th minute onwards in NOD mice injected with 123I-IL-2 but not in Balb/c mice, or in NOD mice injected with control radiopharmaceuticals. These findings were confirmed by counting the radioactivity present in single organs. Autoradiography of NOD pancreas, after injection of labelled IL-2, showed that radioactivity was specifically associated with infiltrating lymphocytes. In conclusion, this technique is highly specific and easy to perform and we suggest its application in humans for in vivo detection of areas of lymphocytic infiltration.

Biochemical markers of type 1 diabetes: clinical use.

In type 1 diabetes, a number of specific and non-specific antigens have been identified. The major autoantigens involved in the destructive process of beta-cells leading to the development of type 1 diabetes are proinsulin/insulin, glutamic acid decarboxylase (GAD) and the transmembrane protein tyrosine phosphatase (IA-2). These are the only autoantigens that partially satisfy the criteria by which an autoantigen or cross-reactive nonself antigen could be evaluated for a pathogenic role in autoimmune diseases. Antigens by definition induce antibody production and in type 1 diabetes, such (auto) antibodies are accepted as biochemical markers for the disease. Here we describe the main features and usefulness of these markers for disease prediction.

Tumor necrosis factor alpha (TNFalpha) and its soluble receptor p75 (sTNF-R p75) in familial combined hyperlipidemia (FCHL).

BACKGROUND AND AIM: Familial combined hyperlipidemia (FCHL) is a genetic disorder of lipid metabolism associated with insulin resistance and abnormalities in fatty acid metabolism whose underlying mechanisms are largely unknown. Perturbations in the TNFalpha/TNF-R pathway may play a role in these abnormalities. METHODS AND RESULTS: We determined plasma levels of TNFalpha and sTNF-R p75 in 85 FCHL patients (TC 245+/-45 mg/dl; TG 260+/-148 mg/dl; apoB 148+/-37 mg/dl) and in 29 age- and sex-matched normolipemic relatives (NL) (TC 187+/-22.8 mg/dl; TG 115+/-37 mg/dl; apoB 106+/-16 mg/dl). Thirty-four normolipemic subjects (TC 180+/-34 mg/dl; TG 107+/-42 mg/dl; apoB 95+/-22 mg/dl) were also included as unrelated controls (NC). Plasma free fatty acids (NEFA) were also measured and insulin sensitivity was evaluated by HOMA. Levels of sTNF-R p75 were significantly reduced in FCHL compared to NL (2.30+/-0.55 ng/ml vs. 2.64+/-0.88 ng/ml, p<0.05) but not compared to NC (2.35+/-0.68 ng/ml). HOMA values were comparable in all groups and did not show any relation with plasma levels of sTNF-R p75. Logistic analysis demonstrated that a low concentration of sTNF-R p75 was an independent predictor of the affected status within FCHL families, but this role was no longer evident when FCHL patients were compared to NC. In FCHL, age (p<0.001) was positively, and TG (p=0.029) and HDL-C (p=0.025) were negatively correlated with plasma concentrations of sTNF-R p75. In the other groups, age (in NL) and non-HDL-C (in NC) were significantly correlated with sTNF-R p75. CONCLUSIONS: Although our data do not support a causative role of TNFalpha/TNF-R alterations in FCHL, they confirm that variation in TNF-R shedding may influence lipid phenotypic expression in FCHL families.

Cell-mediated immune response to beta casein in recent-onset insulin-dependent diabetes: implications for disease pathogenesis.

BACKGROUND: The cows' milk hypothesis for the cause of insulin-dependent diabetes (IDDM) is based on the concept that early consumption of cows' milk may expose the immune system to a foreign protein possessing immunological cross-reactivity with an antigen present on pancreatic beta-cells. METHODS: We measured in-vitro peripheral lymphocyte response to beta casein, a protein in cows' milk, in 47 patients with recent-onset IDDM, in 36 healthy people and, to test disease specificity, in 10 patients with autoimmune thyroid disease. Other antigens tested for were bovine serum albumin, purified protein derivative, human serum albumin, and phytohaemagglutinin. RESULTS: Specific proliferation of T lymphocytes with bovine beta casein was seen in patients with IDDM (mean [SD] age 18.7 [9]) with a significant difference in mean stimulation index (SI) versus healthy people (p < 0.00001) or patients with autoimmune thyroid disease (p < 0.002). 24 of 47 (51.1%) patients with IDDM versus 0/10 patients with thyroid disease and 1/36 (2.7%) healthy people had a positive response to beta casein defined as a SI above the mean value +2 SD of healthy people (p < 0.00001). No significant differences were observed between the groups of subjects with respect to other antigens tested. INTERPRETATION: The association between IDDM and early consumption of cows' milk may be explained by the generation of a specific immune response to beta casein. Exposure to cows' milk triggers a cellular and humoral anti-beta casein immune response which may cross-react with a beta-cell antigen. It is of interest that sequence homologies exist between beta casein and several beta-cell molecules.

Glutathione protects a human insulinoma cell line from tumor necrosis factor-alpha-mediated cytotoxicity.

It is postulated that glutathione acting as a free oxygen radical scavenger may protect beta-cells from cytokine-mediated cytotoxicity in insulin-dependent diabetes. In this study the effect of glutathione in preventing the cytotoxic damage mediated by tumor necrosis factor-a in vitro towards a human beta-cell line (CM insulinoma) was investigated. CM cells were exposed in vitro to tumor necrosis factor-alpha, tumor necrosis factor-alpha plus glutathione or glutathione alone at different concentrations. The resulting cytotoxicity was measured using a colorimetric assay. Glutathione significantly reduced the cytotoxicity mediated by tumor necrosis factor-alpha in a dose-dependent fashion (P < 0.001). These results suggest a protective effect of glutathione on beta-cell cytotoxicity induced by tumor necrosis factor-alpha and encourage the use of glutathione in trials aimed at reducing the beta-cell damage occurring in insulin-dependent diabetes.

Cytokine profile and insulin antibody IgG subclasses in patients with recent onset type 1 diabetes treated with oral insulin.

AIMS/HYPOTHESIS: Tolerance to orally administered antigens may be generated through the induction of T helper cell type 2 and 3 (Th2/Th3) regulatory cells. We previously reported that treatment of recent onset type 1 diabetes with oral insulin had no effect on residual beta cell function. The aim of this study was to evaluate whether this treatment produces a deviation in the immune response, with polarisation of the cytokine pattern and the induction of a Th2-like antibody response. METHODS: Mononuclear cells were collected from a total of 20 patients with type 1 diabetes before and after 12 months of treatment with oral insulin (n=11) or placebo (n=9). Following stimulation of the cells with insulin or phytohaemagglutinin, levels of Th2 and Th3 cytokines (including TGF-beta, IFN-gamma, IL-4 and IL-5) in the culture supernatants were assessed by ELISA. In addition, levels of total and specific insulin antibody IgG subclasses were measured by radioimmunoassay in serum samples drawn from 33 patients with type 1 diabetes before and after 3, 6 and 12 months of therapy with oral insulin (n=18) or placebo (n=15). RESULTS: After 12 months of treatment, the release of TGF-beta was significantly higher in patients who received oral insulin compared with those who received placebo (p=0.025 and p=0.006 for lymphocytes challenged with insulin and phytohaemagglutinin respectively). The two groups had similar levels of IL-4 and IL-5 both at baseline and after 12 months of treatment. The release of IFN-gamma was markedly reduced in patients treated with oral insulin compared with those who received placebo at the 12-month follow-up. Circulating levels of IgG1 and IgG3 subclasses directed against insulin were significantly lower in the oral insulin group than in the placebo group after 12 months of treatment (p=0.05 for IgG1 and p=0.014 for IgG3). CONCLUSIONS/INTERPRETATION: The increased TGF-beta release observed in patients treated with oral insulin suggests that a regulatory response can be induced in vivo by this treatment. The lower levels of insulin antibody IgG1 and IgG3 subclasses present in patients exposed to oral insulin are consistent with a Th2 deviation of the immune response. The failure of oral insulin treatment to provide any measurable clinical benefit may be due to the timing of treatment initiation.

Bovine beta-casein antibodies in breast- and bottle-fed infants: their relevance in Type 1 diabetes.

BACKGROUND: Bovine beta-casein is a cow's milk protein that targets both humoral and cellular immune responses in patients with Type 1 diabetes and, to a lesser degree, also in normal subjects. In this study we aimed to determine whether the avoidance of cow's milk consumption early in life could prevent the development of antibody response to bovine beta-casein despite the mother being exposed on a daily basis to cow's milk consumption. MATERIALS AND METHODS: We measured the antibody response to bovine beta-casein using an ELISA method in 28 healthy infants under 4 months of age, of whom 16 were exclusively breast-fed and 12 were bottle-fed with cow's milk. In addition, beta-casein antibodies were measured in 37 prepubertal children with Type 1 diabetes and in 31 healthy children who were exposed to cow's milk or dairy products to see whether differences in antibody titers exist in this young age group. Antibodies binding to beta-casein were also evaluated by immunoblotting analysis. RESULTS: Elevated levels of beta-casein antibodies were found in bottle-fed infants compared to breast-fed infants (p<0.0001). Antibody levels to bovine beta-casein were also significantly higher in children with Type 1 diabetes compared to age-matched controls (p=0.03). By western blot analysis we confirmed specific binding to bovine beta-casein in bottle-fed infants, in children with Type 1 diabetes and in controls exposed to cow's milk, but not in infants who were exclusively breast-fed. CONCLUSIONS: The results of this study indicate that breastfeeding within the first 4 months of life prevents the generation of antibody response to bovine beta-casein despite the mothers' consumption of cow's milk during the breastfeeding period. These findings may have relevance for disease prevention.

Intradermal skin test with diabetes specific antigens in patients with type 1 diabetes.

Cell mediated immune response in vitro to a number of antigens has been reported in patients with Type 1 diabetes. The aim of the present study was to develop an in vivo intradermal (delayed type hypersensitivity) skin test using antigens known to be recognized by lymphocytes of patients with Type 1 diabetes and to compare, where possible, the in vivo response to the in vitro T cell proliferation to the same antigens. The skin test was performed in the following group of patients: 55 with recent onset Type 1 diabetes; 16 patients with Type 1 diabetes of longer duration; 10 patients with autoimmune thyroid disease and 20 patients with Latent Autoimmune Diabetes in Adults (LADA). Type 1 diabetes specific antigens for the skin test included glutamic acid decarboxilase (GAD65), insulin and beta casein, whereas diabetes non specific antigens included tetanus toxoid, diphteria, proteus, tubercolin, streptococcus, and glycerol as control. A multitest device consisting of heads delivering intradermally 10 microl of solution containing the antigens was applied to the forearms; the specific antigens were injected in one forearm whereas the non specific antigens were injected in the other forearm. Reading of the reaction, which was considered positive in the presence of a nodule of 2 mm diameter was performed 48 h after the multitest application. The in vitro T cell response to diabetes specific antigens used in the multitest was studied using conventional proliferation assays in patients with recent onset Type 1 diabetes and in age matched normal subjects. Only recent onset Type 1 diabetes patients showed an in vivo positive response to GAD65, such response being detectable in 10 patients (18%). Two patients reacted also to beta casein and insulin, all other patient groups resulted negative but 2 patients with longer duration of Type 1 diabetes. There was no apparent link between the in vivo skin test and in vitro T cell proliferation to GAD65. We conclude that in vivo cell mediated immune reaction to GAD65, insulin and beta casein can be visualized in a minority of patients with recent onset Type 1 diabetes. Further studies are required to determine specificity and whether altering the dose can improve the sensitivity of the test.

Antibodies to bovine beta-casein in diabetes and other autoimmune diseases.

Cow's milk is thought to be an environmental trigger for autoimmune response in Type 1 diabetes. In the present study, our aim was to investigate the antibody response to bovine beta-casein in different immune- and non-immune-mediated diseases and to establish whether such an antibody response is specific to Type 1 diabetes. We measured antibodies to bovine beta-casein using an enzyme-linked immunosorbent assay in a total of 519 sera from subjects as follows: 71 patients with Type 1 diabetes, 33 patients with coeliac disease, 100 patients with latent autoimmune diabetes in adults (LADA), 50 patients with autoimmune thyroid disease (ATD), 50 patients with Type 2 diabetes, 24 patients with multiple sclerosis (MS), and 3 different groups of controls (n = 191). Significantly increased levels of antibodies to beta-casein were found in patients with Type 1 diabetes, coeliac disease and in LADA compared to age-matched controls (p = 0.01, p = 0.02 and p = 0.01, respectively). No differences were observed in beta-casein antibody titres between patients with other disease conditions (MS, and ATD) and age-matched controls. The highest antibody response to beta-casein in Type 1 diabetic patients and in patients with coeliac disease could reflect the gut mucosal immune disorders common to Type 1 diabetes and coeliac disease. Furthermore, the elevated beta-casein antibody levels found in LADA patients suggest that the antibody response to this protein may be relevant in autoimmune diabetes.