Kateřina Štechová

Healthy first-degree relatives of patients with type 1 diabetes exhibit significant differences in basal gene expression pattern of immunocompetent cells compared to controls: expression pattern as predeterminant of autoimmune diabetes.

Expression features of genetic landscape which predispose an individual to the type 1 diabetes are poorly understood. We addressed this question by comparing gene expression profile of freshly isolated peripheral blood mononuclear cells isolated from either patients with type 1 diabetes (T1D), or their first‐degree relatives or healthy controls. Our aim was to establish whether a distinct type of ‘prodiabetogenic’ gene expression pattern in the group of relatives of patients with T1D could be identified. Whole‐genome expression profile of nine patients with T1D, their ten first‐degree relatives and ten healthy controls was analysed using the human high‐density expression microarray chip. Functional aspects of candidate genes were assessed using the MetaCore software. The highest number of differentially expressed genes (547) was found between the autoantibody‐negative healthy relatives and the healthy controls. Some of them represent genes critically involved in the regulation of innate immune responses such as TLR signalling and CCR3 signalling in eosinophiles, humoral immune reactions such as BCR pathway, costimulation and cytokine responses mediated by CD137, CD40 and CD28 signalling and IL‐1 proinflammatory pathway. Our data demonstrate that expression profile of healthy relatives of patients with T1D is clearly distinct from the pattern found in the healthy controls. That especially concerns differential activation status of genes and signalling pathways involved in proinflammatory processes and those of innate immunity and humoral reactivity. Thus, we posit that the study of the healthy relative’s gene expression pattern is instrumental for the identification of novel markers associated with the development of diabetes.

Immune regulatory T cells in siblings of children suffering from type 1 diabetes mellitus

Patients with type 1 diabetes are suffering from defects in immune regulatory cells. Their siblings may be at increased risk of type 1 diabetes especially if they are carriers of certain human leucocyte antigen (HLA) alleles. In a prospective non‐randomized study, we intended to evaluate 31 healthy siblings of paediatric patients with type 1 diabetes and explore immune regulatory populations of CD4+CD25+ T cells and natural killer (NK) T cells. Tested siblings of type 1 diabetes patients were stratified according to the HLA‐associated risk of possible diabetes development. Immune regulatory function of CD4+CD25+ T cells was tested in vitro. Significant differences in CD4+CD25+ but not in NK T cells have been identified. Siblings of type 1 diabetes patients carrying high risk HLA alleles (DQA1*05, DQB1*0201, DQB1*0302) had significantly lower number of immune regulatory CD4+CD25+ T cells than the age‐matched healthy controls or siblings carrying low‐risk HLA alleles (DQB1*0301, DQB1*0603, DQB1*0602). Regulatory function of CD4+CD25+ T cells demonstrated a dose‐escalation effect. In siblings of type 1 diabetes patients, the defect in immune regulatory CD4+CD25+ T cells exists in association with genetic HLA‐linked risk for type 1 diabetes.

DNA methylation in monozygotic quadruplets affected by type 1 diabetes

To the Editor: The concordance rate of monozygotic (MZ) twins is approximately 50% for type 1 diabetes and this, along with other evidence, supports a role played by both genetic and environmental factors in determining this disease [1, 2]. The epigenome of a cell can be defined as the group of epigenetic marks, such as DNA methylation, that influence gene expression and are responsible for cell specific differentiation [3]. Epigenetic marks are influenced by environmental agents and therefore provide a potential explanation of how environmental exposure can impact upon human health [4]. The presence of DNA methylation differences in CD14 monocytes of MZ twins discordant for type 1 diabetes has recently been reported [5]. We have further explored cell specific DNA methylation differences in diabetes in an unprecedented case of MZ quadruplets of which two individuals had diabetes, one was at a prediabetic stage and one was diabetes free. The quadruplets comprise two sisters who developed type 1 diabetes at the age of 5 (T1D-1 and T1D-2), one sister who is currently at a prediabetic stage with abnormal first-phase insulin response and glucose concentration (Pre-T1D) and one more sister with no type 1 diabetes manifestations who was used as a healthy control (HC) (electronic supplementary material [ESM] Table 1) [6]. The HLA-DQ alleles carried by all sisters are DQA1*01/01 and DQ-B1*0501/0501, genotypes that are not commonly seen among type 1 diabetic patients. CD14 monocytes and CD4 T cells were isolated from venous blood using magnetic-activated cell sorting. DNA methylation was profiled using the Illumina HumanMethylation450 BeadChip in triplicate (Illumina, San Diego, CA, USA). All analyses were performed using the R package RnBeads (http://rnbeads.bioinf.mpi-inf.mpg.de/index.php) (ESM Methods). A total of 12 CD4 and 12 CD14 DNA samples (three from each individual) were analysed. A very strong correlation between replicates of the same individual and cell type was found. Only one sample (CD4 Pre-T1D replicate 3) appeared markedly different from the other two replicates and was therefore excluded (ESM Fig. 1). After removing probes of highest impurity (n=745), a total of 484,832 methylation probes were left for analysis. The β values followed a typical bimodal distribution in all samples and individual replicates clustered first according to cell type (CD4 and CD14) and then to disease status (ESM Figs 2 and 3). The number of differentially methylated probes (DMPs) with q values<0.1 was 1,575 in CD14 cells and 1,170 in CD4 Electronic supplementary material The online version of this article (doi:10.1007/s00125-013-2972-3) contains peer-reviewed but unedited supplementary material, which is available to authorised users. G. Disanto : J. Pakpoor :R. I. Elangovan :G. C. Ebers : S. V. Ramagopalan (*) Department of Physiology, Anatomy and Genetics and Medical Research Council Functional Genomics Unit, University of Oxford, South Parks Road, Oxford OX1 3PT, UK e-mail: sreeram.ramagopalan@dpag.ox.ac.uk

Eosinophils from patients with type 1 diabetes mellitus express high level of myeloid alpha-defensins and myeloperoxidase

Type 1 diabetes (T1D) is an autoimmune disease caused by T-cell mediated destruction of pancreatic beta cells. Recently, small cationic α-defensin molecules have been implicated in the pathogenesis of certain inflammatory and autoimmune diseases. The purpose of this study was to assess the α-defensin expression in patients with T1D and elucidate the cellular source of their production. Our results show that 30% of patients exhibit increased levels of α-defensin mRNAs in their capillary blood. Quantitative RT-PCR performed on FACS-sorted granulocytes identified CD15(dull)/CD14(weak) population as the cellular source of α-defensins. Surprisingly, this granulocyte subpopulation displayed augmentation of α-defensin expression in all T1D patients tested. The determination of cell surface markers, expression of cell-specific genes and confocal microscopy identified CD15(dull)/CD14(weak) cells as eosinophils. The presence of transcriptionally active eosinophils in diabetic patients suggests that eosinophils could be a part of an intricate innate immune cellular network involved in the development of diabetes.

Influence of Maternal Hyperglycaemia on Cord Blood Mononuclear Cells in Response to Diabetes‐associated Autoantigens

Perfect maternal diabetes compensation is crucial for the outcome of the baby. However, little is known how hyperglycaemia influences the specific immune response. Furthermore, babies of type 1 diabetes (T1D) mothers have less risk of development T1D than babies with a T1D father. This study aimed to analyze the effect of maternal hyperglycaemia on newborns with focus on the response to diabetes‐associated autoantigens. Populations: (1) Newborns of T1D mothers split into groups according to maternal diabetes compensation during the 3rd trimester: perfect (n = 15) or acceptable (n = 25) compensation. (2) newborns with T1D father (n = 12) (3) newborns with a mother treated for either gestational or type 2 diabetes (n = 10) (4) control newborns (n = 25). Spontaneous as well as diabetes‐associated autoantigen‐stimulated production of 23 cytokines and chemokines were tested using protein microarray. In addition, the influence of glucose on cytokine and chemokine responsiveness was analyzed in vitro. The study groups differed in their spontaneous as well as stimulated cytokine and chemokine spectra. A prominent Th1 response (high IFN‐gamma) from autoantigen stimulation was observed especially in babies of T1D fathers (P = 0.001) and also in mothers with perfect diabetes compensation during the 3rd trimester (P = 0.016) in comparison with control newborns. By contrast, cord blood mononuclear cells cultivated in vitro in high glucose concentration decreased the diabetogenic stimulated Th1 cytokine response. Maternal ‘sweet’ as well as ‘autoimmune environment’ may both lead to lower occurrence of T1D within their offspring. Further studies will reveal the exact immunological mechanism of this observation.

Cord Blood Cytokine Profile Detection in Neonates with T1D Parents – Monitoring of Cellular Auto‐reactivity Using Protein Microarray

Type 1 diabetes (T1D) is a great medical challenge and its incidence rises rapidly. T lymphocytes and their cytokine production are supposed to play a major role in T1D development. So far, there is no potent tool to recognize the early signs of cellular auto‐reactivity which leads to β‐cell damage. The naïve immune system of the newborn (not yet influenced by external factors) can be used as an important model for T1D pathogenesis studies. Cord blood samples of 22 healthy neonates born at term to a diabetic parent (T1DR) and 15 newborns with no family history of any autoimmune disease (controls) were collected. Determination of 23 cytokines was performed before and after the stimulation with diabetogenic autoantigens using protein microarray. We observed lower basal production of all detected cytokines in the T1DR group – granulocyte/macrophage colony‐stimulating factor (GM‐CSF) (P = 0.025), growth regulated protein (GRO) (P = 0.002), GRO‐α (P = 0.027), interleukin (IL)‐1‐α (P = 0.051), IL‐3 (P = 0.008), IL‐7 (P = 0.027), IL‐8 (P = 0.042), monocyte chemoattractant proteins (MCP)‐3 (P = 0.022), monokine‐induced by IFN‐γ (MIG) (P = 0.034) and regulated upon activation normal T‐cell express sequence (RANTES) (P = 0.004). Exclusively lower post‐stimulative levels of G‐CSF (P = 0.030) and GRO‐α (P = 0.04) were observed in controls in comparison with the basal levels. A significant post‐stimulative decrease in G‐CSF (P = 0.030) and MCP‐2 (P = 0.009) levels was observed in controls in comparison with T1DR neonates. We also observed the interesting impact of the risky genotype on the protein microarray results. Protein microarray seems to be a useful tool to characterize a risk pattern of the immune response for T1D also in newborns.