Magnus Hillman

IgG4-subclass of glutamic acid decarboxylase antibody is more frequent in latent autoimmune diabetes in adults than in type 1 diabetes

Aims/hypothesisGlutamic acid decarboxylase autoantibodies (GADA) are the most frequent beta-cell-specific autoantibodies in type 1 diabetes and in latent autoimmune diabetes in adults (LADA). The autoimmune attack on pancreatic islet cells is associated with a T helper 1 cell (Th1) response, mainly represented by IgG1-subclass in humans. It has been proposed that the presence of IgG4 may be associated with a Th2 response. The aim of our study was to compare the GADA IgG-subclass distribution between adult patients with type 1 diabetes and LADA.MethodsPatients with type 1 diabetes (n=45) and patients with LADA (n=60) were included. Radioimmunoprecipitation assay with IgG-subclass specific Sepharose (IgG1, IgG2, IgG3 and IgG4) was used to precipitate the antibody/antigen-complex.ResultsWe only detected IgG4-subclass of GADA in subjects with LADA (26.7%; p<0.001). IgG1 was the most common GADA-subclass in both groups, however IgG1 as the solely expressed subclass was more common among type 1 diabetic patients (77.8%; p<0.05). The rank order of the frequencies of IgG-subclasses in type 1 diabetes was IgG1>IgG3>IgG2>IgG4 and in LADA patients IgG1>IgG4>IgG2>IgG3.Conclusions/interpretationThe difference in GADA IgG-subclasses could indicate a different immune response, possibly an altered balance between Th1 and Th2 cytokine profile in pancreatic islets. This difference could contribute to the slower rate of beta cell destruction in LADA patients, as reflected by a higher C-peptide level at clinical onset.

Reduced levels of active GLP-1 in patients with cystic fibrosis with and without diabetes mellitus.

Glucagon like peptide 1 (GLP-1) is an incretin hormone released as a bioactive peptide from intestinal L-cells in response to eating. It acts on target cells and exerts several functions as stimulating insulin and inhibiting glucagon. It is quickly deactivated by the serine protease dipeptidyl peptidase IV (DPP-IV) as an important regulatory mechanism. GLP-1 analogues are used as antidiabetic drugs in patients with type 2 diabetes. We served patients with cystic fibrosis (CF, n=29), cystic fibrosis related diabetes (CFRD, n=19) and healthy controls (n=18) a standardized breakfast (23 g protein, 25 g fat and 76 g carbohydrates) after an overnight fasting. Blood samples were collected before meal as well as 15, 30, 45 and 60 min after the meal in tubes prefilled with a DPP-IV inhibitor. The aim of the study was to compare levels of GLP-1 in patients with CF, CFRD and in healthy controls. We found that active GLP-1 was significantly decreased in patients with CF and CFRD compared to in healthy controls (p<0.01). However, levels in patients with CFRD tended to be lower but were not significantly lower than in patients with CF without diabetes (p=0.06). Total GLP-1 did not differ between the groups, which points to that the inactive form of GLP-1 is more pronounced in CF patients. The endogenous insulin production (measured by C-peptide) was significantly lower in patients with CFRD as expected. However, levels in non-diabetic CF patients did not differ from the controls. We suggest that the decreased levels of GLP-1 could affect the progression toward CFRD and that more studies need to be performed in order to evaluate a possible treatment with GLP-1 analogues in CF-patients.

Polymorphisms of TNF microsatellite marker a and HLA-DR-DQ in diabetes mellitus-a study in 609 Swedish subjects.

We explored the importance of the genetic markers microsatellite TNFa, HLA-DR3-DQ2, and DR4-DQ8 in diabetes mellitus. The studied groups comprised autoimmune type 1 (n = 63), nonautoimmune type 1 (n = 35), latent autoimmune diabetes in adults (LADA; n = 54), and nonautoimmune type 2 (n = 340) and these patients were compared to 117 healthy controls. HLA genotyping was done with polymerase chain reaction and sequence-specific oligonucleotides. TNFa microsatellites were determined with polymerase chain reaction and fragment size determination. Univariate analysis of these genetic risk factors demonstrated that homozygosity for TNFa2/2 was a significant risk factor for autoimmune type 1 diabetes (odds ratio (OR) = 5.82; 95% confidence interval (95%CI) 1.97-17.2), for autoimmune negative type 1 diabetes (OR = 4.63; 95%CI 1.32-16.2), and for LADA (OR = 3.90; 95%CI 1.21-12.5). Moreover, heterozygosity for HLA-DR3-DQ2/DR4-DQ8 was an important risk factor for autoimmune type 1 diabetes (OR = 16.4; 95%CI 3.60-75) as was DR4-DQ8/x (OR = 2.52; 95%CI 1.27-4.98). Heterozygosity for HLA-DR3-DQ2/DR4-DQ8 was a risk factor also for LADA (OR = 10.0; 95%CI 2.05-48.9). Neither HLA-DR3-DQ2 nor DR4-DQ8 were risk factors for nonautoimmune type 1 or type 2 diabetes. We concluded that heterozygosity for DR3-DQ2/DR4-DQ8 and to some extent homozygosity for TNFa2/2 were risk factors for autoimmune diabetes irrespective of the clinical classification.

Prevalence of zinc transporter 8 antibodies in gestational diabetes mellitus.

Diabet. Med. 29, e436–e439 (2012)

Is type 1 diabetes a food-induced disease?

The incidence of type 1 diabetes among children has almost doubled during the last decades in Sweden. Type 1 diabetes is considered as an autoimmune disease with unknown aetiology. Here we propose that the immune reaction may be initiated by food-derived mechanisms. The incidence of diabetes parallels an increased consumption of pasta, white bread, meat, cheese, low-fat milk, exotic fruits, soda, and snacks. Simultaneously, the consumption of potatoes, butter, high-fat milk, and domestic fruit has decreased. Three categories of food related reaction mechanisms are discussed against the following items (1) proteins from wheat, meat, and milk, (2) fat from processed food, and (3) exotic fruits. The current food consumption is suggested to initiate a pro-inflammatory reaction in the intestine and thereby reduce the intestinal barrier function. This may influence tolerance development and thus pave the way for an autoimmune attack on pancreatic beta cells.

Plasma levels of relaxin-2 are higher and correlated to C-peptide levels in early gestational diabetes mellitus

Relaxin-2 is an important gestational hormone in shaping the endometrium in early pregnancy and its secretion peaks during the first trimester [1]. The relaxin family of peptides are structurally found to be similar to insulin and insulinlike growth factors with A-chains and B-chains making up the bioactive molecule [2, 3] and a C-peptide connecting the pre-molecule [4]. A study investigating relaxin levels in women with type 2 diabetes found a correlation between relaxin and insulin sensitivity. Contrarily, a negative correlation between relaxin and beta cell function has been suggested [5]. Another study reported lower levels of relaxin-2 but not of relaxin-3 in patients with recent onset type 2 diabetes. This data combined would suggest that relaxin-2 could be involved in the pathophysiology of type 2 diabetes and beneficial in increasing insulin sensitivity in patients with insulin resistance [6]. Pregnancy is a normal state of insulin resistance which ensures continuous glucose transport to the fetus. Women are however not normally hyperglycemic during pregnancy. If hyperglycemic, the women are diagnosed with gestational diabetes mellitus [7]. There are currently no reports on levels of relaxin-2 in gestational diabetes mellitus. The aim of this study was to investigate plasma levels of relaxin-2 in patients with gestational diabetes mellitus, in comparison to pregnant controls without diabetes.

Determination of glutamic acid decarboxylase antibodies (GADA) IgG subclasses - comparison of three immunoprecipitation assays (IPAs)

IgG subclasses of glutamic acid decarboxylase (GAD65) antibodies (GADA) may reflect the immunological state in the pancreas of GADA‐positive patients with autoimmune diabetes. The use of biotin‐conjugated antibodies and streptavidin Sepharose are used commonly in immunoprecipitation assays (IPA) based on 125I‐ or 35S‐labelled antigens to capture IgG subclasses directed against IA‐2 or GAD65. We have compared three different immunoprecipitation assays for the determination of GADA IgG subclasses. Two of the assays were based on the biotin and streptavidin systems provided in a solid (immobilized) or liquid (mobilized) phase binding environment. The third assay was based on N‐hydroxysuccinimide (immobilized) interaction with primary amines (i.e. lysine residues) on the antibody. We found the liquid phase binding assay (LPBA) to be the most stable assay, with a comparatively low coefficient of variation and background.

The significance of biomarkers such as FABP4 in gestational diabetes mellitus and the importance of rigorous peer review

performed in the statistical software Graph Pad Prism, but the ROC curve analysis is apparently made using another statistical software clearly indicating that this section has not been carefully evaluated. All data in the study are presented as mean ± standard deviation with comparisons performed with the Students’ T test although soluble biomarkers are rarely normally distributed and any parametric tests performed for non-normally distributed data will be skewed. A clear presentation of the sample handling and preparation is essential if the reader is to draw any conclusion from the results of a study. This is, however, poorly described in the paper where the authors do not describe the collection process, the time and temperature regarding the sample storage nor time from sample collection to analysis. This could affect the stability and level of cytokines and possibly other biomarkers [2]. It is also unclear whether FABP4 has been analysed in serum or plasma as the title states plasma, while the results state serum. It is important to clearly state in which type of sample the analysis has been made in order to be able to confirm the results of the study. Further there are clear discrepancies between Table 1 and the first paragraph on page 894 regarding LGA and Apgar score. This is of particular importance since Table 1 states statistically significant differences for these variables. The authors mention that serum levels of FABP4 have previously been observed in a small study made on patients with GDM [3], while other and more comprehensive studies also have investigated circulating FABP4 levels in GDM without being discussed [4, 5]. Relating and discussing the findings in this study to the other previous studies would give the reader a better understanding of the merit and addition made by the present study regarding FABP4 in GDM. Dear Editor-in-chief We appreciate that biomarkers in gestational diabetes mellitus (GDM) recently have gained increased interest. Diagnostic criteria for GDM are still debated. It is important to find reliable diagnostic tools to complement blood glucose measurement in early GDM since diabetes during pregnancy is associated with adverse outcome for both mother and child. In their case–control study, Dr Ning et al. [1] suggested plasma fatty acid-binding protein 4 (FABP4) to be elevated in women with GDM. By performing a receiver operating characteristics (ROC) curve analysis, they have identified FABP4 as a potential novel predictor of GDM. Even though the reported results are credible, there are several issues that need to be addressed which should have been acknowledged already during the peer review process. In order to properly evaluate the scientific conclusions from a study, the scientific and statistical methods should be clearly presented to the reader. Sometimes the reviewers seem to underestimate the value of correct statistical methods, and in this case, the section describing the statistical analysis seems to have been overlooked in the review process. First of all, serum calprotectin concentrations are mentioned despite the fact that they have not been analysed in this study. Secondly, the authors state that ANOVA has been performed for multiple comparisons despite the fact that ANOVA has not been used in the study. Thirdly, the authors claim that all statistical analyses have been

Clinical use of C-peptide and β-cell specific autoantibodies during gestational diabetes mellitus

Gestational diabetes mellitus (GDM) confers a risk for developing type 2 diabetes later in life, but the risk of developing type 1 diabetes is also increased. In this study we have evaluated the clinical use of C‐peptide and β‐cell specific autoantibodies during pregnancy with GDM as predictors for later development of diabetes.

Depression in type 1 diabetes was associated with high levels of circulating galectin-3

Objective Neuroinflammatory responses are implicated in depression. The aim was to explore whether depression in patients with type 1 diabetes (T1D) was associated with high circulating galectin-3, controlling for metabolic variables, s-creatinine, life style factors, medication and cardiovascular complications. Design Cross-sectional. Methods Participants were T1D patients (n = 283, 56% men, age 18–59 years, diabetes duration ≥1 year). Depression was assessed by Hospital Anxiety and Depression Scale-depression subscale. Blood samples, anthropometrics and blood pressure were collected, and supplemented with data from medical records and the Swedish National Diabetes Registry. Galectin-3 ≥2.562 µg/l, corresponding to the 85th percentile, was defined as high galectin-3. Results Median (quartile1, quartile3) galectin-3 (µg/l) was 1.3 (0.8, 2.9) for the 30 depressed patients, and 0.9 (0.5, 1.6) for the 253 non-depressed, P = 0.009. Depression was associated with high galectin-3 in all the 283 patients (adjusted odds ratio (AOR) 3.5), in the 161 men (AOR 3.4), and in the 122 women (AOR 3.9). HbA1c, s-lipids, s-creatinine, blood pressure, obesity, smoking, physical inactivity, cardiovascular complications and drugs (antihypertensive, lipid lowering, oral antidiabetic drugs and antidepressants) were not associated with high galectin-3. Conclusions This is the first study to show an association between depression and galectin-3. Depression was the only explored parameter associated with high circulating galectin-3 levels in 283 T1D patients. High galectin-3 levels might contribute to the increased risk for Alzheimer’s disease, cardiovascular and all-cause mortality observed in persons with depression. Potentially, in the future, treatment targeting galactin-3 might improve the prognosis for patients with high galectin-3 levels.