Aims/hypothesis Sphingolipids play important roles in beta cell physiology, by regulating proinsulin folding and insulin secre- tion and in controlling apoptosis, as studied in animal models and cell cultures. Here we investigate whether sphingolipid metabolism may contribute to the pathogenesis of human type 1 diabetes and whether increasing the levels of the sphingolipid sulfatide would prevent models of diabetes in NOD mice. Methods We examined the amount and distribution of sulfatide in human pancreatic islets by immunohistochemistry, immuno- fluorescence and electron microscopy. Transcriptional analysis was used to evaluate expression of sphingolipid-related genes in isolated human islets. Genome-wide association studies (GWAS) and a T cell proliferation assay were used to identify type 1 diabetes related polymorphisms and test how these affect cellular islet autoimmunity. Finally, we treated NOD mice with fenofibrate, a known activator of sulfatide biosynthesis, to evaluate the effect on experimental autoimmune diabetes development. Results We found reduced amounts of sulfatide, 23% of the levels in control participants, in pancreatic islets of individuals with newly diagnosed type 1 diabetes, which were associated with reduced expression of enzymes involved in sphingolipid metabolism. Next, we discovered eight gene polymorphisms (ORMDL3, SPHK2, B4GALNT1, SLC1A5, GALC, PPARD, PPARG and B4GALT1) involved in sphingolipid metabolism that contribute to the genetic predisposition to type 1 diabetes. These gene polymorphisms correlated with the degree of cellular islet autoimmunity in a cohort of individuals with type 1 diabetes. Finally, using fenofibrate, which activates sulfatide biosynthesis, we completely prevented diabetes in NOD mice and even reversed the disease in half of otherwise diabetic animals. Conclusions/interpretation These results indicate that islet sphingolipid metabolism is abnormal in type 1 diabetes and suggest that modulation may represent a novel therapeutic approach. Data availability The RNA expression data is available online at https://www.dropbox.com/s/93mk5tzl5fdyo6b/Abnormal% 20islet%20sphingolipid%20metabolism%20in%20type%201%20diabetes%2C%20RNA%20expression.xlsx?dl=0. A list of SNPs identified is available at https://www.dropbox.com/s/yfojma9xanpp2ju/Abnormal%20islet%20sphingolipid% 20metabolism%20in%20type%201%20diabetes%20SNP.xlsx?dl=0. . . . . . . . . Keywords Fenofibrate Genepolymorphisms GWAS Islet autoimmunity NOD mice Prevention Sphingolipid Sulfatide Tcells Type 1 diabetes Knut Dahl-Jørgensen and Karsten Buschard contributed equally as senior Abbreviations authors CPM Counts per min Knut Dahl-Jørgensen is Principal Investigator of the DiViD Study DiViD Diabetes virus detection eQTL Expression quantitative trait loci Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00125-018-4614-2) contains peer-reviewed but GRS Genetic risk score unedited supplementary material, which is available to authorised users. IA-2 Islet antigen-2 INS-DRiP Insulin-defective ribosomal product * Karsten Buschard PBMC Peripheral blood mononuclear cell email@example.com PPI Preproinsulin SI Stimulation index Extended author information available on the last page of the article Diabetologia (2018) 61:1650–1661 1651 Introduction pancreas biopsies from newly diagnosed individuals with type 1 diabetes, we found evidence suggesting that sphingolipid Type 1 diabetes is characterised as an autoimmune disease in metabolism plays a role in type 1 diabetes pathology. Based which autoreactive T cells infiltrate the pancreatic islets and on this we tested whether increasing pancreatic sulfatide levels destroy the insulin producing beta cells . However, the po- could prevent and reverse experimental autoimmune diabetes in tential importance of beta cell dysfunction, rather than complete NOD mice. beta cell loss, in the pathogenesis of type 1 diabetes has recently been emphasised by the demonstration that a majority of indi- viduals with diabetes retain a significant proportion of insulin- Methods positive islets at disease onset [2–4]. In support of this, islets isolated from pancreatic biopsies taken from individuals with Human tissue Pancreatic tissue was collected in the Diabetes type 1 diabetes partly regained their ability to secrete insulin in Virus Detection (DiViD) studies as described previously . response to glucose when cultured in a non-diabetogenic envi- In short, individuals with diabetes between 25 and 35 years of ronment in vitro , while the majority of individuals with type age had a surgical minimal pancreatic tail resection obtained by 1 diabetes regained insulin production immediately upon treat- laparoscopy 3 to 9 weeks after the onset of type 1 diabetes. The ment with autologous system cell therapy . Potential key DiViD study was approved by The Norwegian Government’s players in this beta cell dysfunction are sphingolipids, a diverse Regional Ethics Committee (reference 2009/1907). UK tissue group of lipids found in all cellular membranes, having diverse samples from the Exeter Archival Diabetes Biobank had all been roles as both structural components and signalling molecules collected prior to the study and were made available with ethical [7–9]. Sphingolipids are known to have important roles in beta approval from the West of Scotland Research Ethics Service cell biology [10, 11] and have been linked to the development (reference 15/WS/0258). The tissue from donors with type 2 of diabetes-associated pathologies . One especially impor- diabetes and non-diabetic control donors, used for RNA analy- tant sphingolipid is sulfatide (3-O-sulfogalactosylceramide) ses, was acquired from the network of Pancreatic Organ Donors which acts as an insulin chaperone, preserves insulin crystals (nPOD; with approval by the University of Tennessee Health and regulates insulin secretion by influencing the gating of Science Center (UTHSC) local Institutional Review Board [ref- ATP-sensitive potassium channels [13, 14]. However, most erence 10-00848-XM]). The pancreatic tissue from participants studies regarding the role of sphingolipid metabolism in beta without diabetes used for sulfatide staining was acquired from cells have been conducted in animal models and cell cultures. Rigshospitalet, Copenhagen, Denmark, as completely Here, we tested the hypothesis that sphingolipid metabolism is anonymised (unknown age and sex) healthy tissue removed contributing to the pathogenesis of human type 1 diabetes from from pancreas samples after resection for surgical treatment of several perspectives. Using blood and tissue samples from pancreatic cancer and was used in accordance with the rules by 1652 Diabetologia (2018) 61:1650–1661 Region Hovedstaden Committee on Health Research Ethics. RNA analyses Frozen tissue (optimal cutting temperature Human islets for electron microscopy were obtained as described [OCT]) sections were obtained from the nPOD and previously . Peripheral blood samples were collected from DiViD tissue collections. Tissue slides were fixed and laser- individuals with type 1 diabetes, aged between 1 and 39 years capture of islets conducted as previously described . All after informed consent was obtained as approved by the Medical islets in two to five sections of tissue from each donor were Ethical Committee of Leiden University Medical Center (refer- captured and pooled and RNA extracted using the Arcturus ence CME05/68C). For all participant information see electronic PicoPure RNA Isolation Kit (Applied Biosystems, Grand supplementary material (ESM) Table 1. Island, NY, USA). Quality and quantity of RNA was deter- mined on a Bioanalyzer 2100 instrument (Agilent Immunohistochemistry Immunohistochemistry on Technologies, Santa Clara, CA, USA). Samples with suffi- neighbouring pancreatic sections from the DiViD study and cient quantity and quality of RNAwere then subjected to gene control participants was performed using anti-sulfatide anti- expression analysis using Affymetrix expression arrays body Sulph I (a gift from P. Fredman, Gothenburg University, (GeneChip Human Gene 2.0 ST) and scaled normalised gene Sweden ; diluted 1:150) or guinea pig anti-insulin (Dako, expression values produced as previously described . The Ely, UK; diluted 1:700). Visualisation was performed using normalised expression data for 70 genes of interest were then ultraView Universal DAB Detection Kit (Roche, Basel, subjected to analysis as described below. Switzerland). The light microscope BX51 (Olympus America, Melville, NY, USA) was used to analyse the stained GWAS analyses The 70 genes examined at the RNA level were specimens. For the UK sections, immunohistochemistry was also evaluated using GWAS to look for a genetic association visualised using Dako REAL EnVision Detection System, with type 1 diabetes. Immunochip SNPs for type 1 diabetes Peroxidase/DAB+ (Dako) and light microscope Nikon 50i were retrieved from Onengut-Gumuscu et al 2015 . A cut- Eclipse (Nikon, Kingston-upon-Thames, UK). The relative off p value <0.02 was used to retrieve nominally significant sulfatide level in pancreatic islets was compared with control SNPs. SNPs within (±100 kb flanking regions) of the examined participants without diabetes. Staining in all control partici- genes were identified for further analysis. We used pants was set to one (100%) and the staining intensity in Encyclopedia of DNA Elements (ENCODE) regulatory fea- minimum 30 islets from each donor was evaluated. tures (ChIP-Seq peaks, DNase I hypersensitivity peaks, DNase I footprints) from University of California Santa Cruz Immunofluorescence staining Pancreatic sections were stained (UCSC) genome browser (http://genome.ucsc.edu/)and with an anti-sulfatide antibody (diluted 1:150) and a secondary RegulomeDB  to identify potential regulatory SNPs likely Alexa Fluor 488 antibody (Life Technologies, Paisley, UK; to affect the expression of the associated gene. We also integrated data from multiple expression quantitative trait loci diluted 1:400). Pancreatic sections were co-stained with an anti-glucagon antibody raised in rabbit (Abcam, Cambridge, (eQTL) studies [24–26] to identify SNPs associated with chang- UK; diluted 1:4000) and with guinea pig anti-insulin (Dako; es in expression (cis-eQTLs) of their associated gene. The cis- diluted 1:700) plus relevant secondary antibodies labelled with eQTL effects were calculated using linear regression models in Alexa Fluor 647 and Alexa Fluor 568 (Life Technologies; (di- the selected tissues (whole blood, fibroblast and lymphoblastoid luted 1:400). Images were captured under fluorescence illumi- cell lines) using a cis window of ±1 MB around the transcription nation using a Leica AF6000 microscope (Leica, Milton start site at significance level of p <0.05 [24–26]. SNPs map- Keynes, UK). Leica application suite X software (Lecia) was ping to HLA regions were excluded from the analysis. used to remove background staining and crop images. Blood donors Peripheral blood was collected from 71 indi- Electron microscopy Isolated pancreatic human islets were viduals with type 1 diabetes. Peripheral blood mononucle- incubated overnight at 4°C with Sulph I (diluted 1:1000) ar cells (PBMCs) were isolated by Ficoll density gradient and washed in 1% PBS-BSA. Next the islets were incubated centrifugation and resuspended in Iscove’s Modified overnight at 4°C with 1 nm gold labelled goat anti-mouse IgG Dulbecco’s Media (IMDM) (Life Technologies, Paisley, (Aurion, Wageningen, the Netherlands; diluted as 1:300). The UK.) containing 10% heat-inactivated human serum (HS; islets were postfixed in 2% glutaraldehyde for 2 h, before Sanquin, the Netherlands). PBMCs were subsequently test- silver enhancement using AURION R-GENT SE-EM ed for the presence of autoreactive T cells using a T cell (Aurion). The islets were then washed in distilled water before proliferation assay. osmication in 1% OsO4 diluted in 0.1 mol/l cacodylate buffer. After washing in 0.1 mol/l cacodylate buffer, the specimens SNP genotyping and genetic risk score DNA was isolated were dehydrated in alcohol and embedded in Epon Resin 812 from PBMCs of individuals with type 1 diabetes using the before ultra-sections were examined in a Philips 208 electron DNeasy Blood & Tissue Kit (Qiagen Benelux, Venlo, the microscope (Philips, Eindhoven, the Netherlands). Netherlands). DNA concentration was determined by Diabetologia (2018) 61:1650–1661 1653 NanoDrop and samples were concentrated at 50 ng/μl. SNP (Abbott, Chicago, IL, USA) glucose monitoring. Diabetes di- genotyping was performed on the Infinium ImmunoArray-24 agnosis was based on two blood glucose measurements >12 v2 BeadChip Kit (Illumina, Eindhoven, the Netherlands) ac- mmol/l with an interval of 2 days, all measurements were cording to the manufacturer’s protocol. made between 09:00 and 13:00 hours. The date of the first To test the cumulative effect of identified SNPs on islet blood glucose measurement >12 mmol/l was used as diabetes autoimmunity, we computed a genetic risk score (GRS) in onset date. Mice were killed by cervical dislocation at onset of all individuals with type 1 diabetes. GRS is the sum of the diabetes or at the age of 217 days. Distribution of animals into number of risk alleles (0, 1 or 2) multiplied by the natural log groups and diabetes monitoring was not performed blinded. In of the OR for each SNP, divided by the total number of SNPs. the reversal studies, fenofibrate treatment was immediately SNPs were also individually analysed. The SNPs selected for commenced at the onset of diabetes and continued for 3 weeks. the study were based on the SNP with the lowest p value and No inclusion or exclusion criteria were used. the highest OR was examined for each of the genes identified in the GWAS analysis. Insulitis and sulfatide scoring of NOD mice Insulitis score was calculated from six mice in each group, at 13 weeks old. T cell proliferation assay A T cell proliferation assay was per- Pancreases were removed, fixed in 10% neutral buffered for- formed on PBMCs freshly isolated from individuals with type malin overnight, embedded in paraffin and sectioned in 5 μm 1 diabetes to investigate autoimmunity towards GAD65, sections that were subsequently stained in haematoxylin and preproinsulin (PPI), islet antigen-2 (IA-2) and insulin- eosin. The sections were evaluated randomly and blinded defective ribosomal product (INS-DRiP). Human recombi- using an Olympus BX53 microscope (Olympus America). nant proteins GAD65, PPI, IA-2 and INS-DRiP were pro- Twenty-five islets from each mouse were scored according duced as previously described [27, 28]. PBMCs were seeded to the following scale: 0, no infiltration; 1, intact islets but (150,000/well) in round-bottomed 96-well microculture plates with few mononuclear cells surrounding the islets; 2, peri- (Greiner, Nürtingen, Germany) and cultured for 5 days in insulitis; 3, islet infiltration below 50% and 4, islet infiltration IMDM containing 10% HS at 37°C in 5% CO , in a humid- above 50%. Neighbouring slides were stained for sulfatide ified atmosphere. Cells were cultured in triplicates in medium and scanned using NanoZoomer-XR (Hamamatsu, alone, with 10 μg/ml recombinant GAD65, PPI, IA-2 or INS- Hamamatsu City, Japan). For each mouse one slide was DRiP or with recombinant IL-2 (35 units/mL; Genzyme, scored and the staining intensity was evaluated using a scale Cambridge, MA, USA) as positive control. After 16 h of cul- from 0 to 4 with 0 denoting no sulfatide and 4 denoting inten- ture, 50 μl RPMI medium 1640 (Dutch modification; Gibco, sity as seen in neurons. The scoring was performed blinded. Thermo Fisher Scientific, Waltham, MA, USA) containing Statistics The statistical analysis was performed using 18500 Bq [ H]thymidine (DuPont, Boston, MA, USA) was added per well. After the cells were harvested on filters with GraphPad Prism version 6.01 (GraphPad, La Jolla, CA, an automated harvester, proliferation was determined by the USA) and data is shown as mean ± SEM unless otherwise measurement of 3H-thymidine incorporation in an automatic noted. The cumulative diabetes incidence was assessed using liquid scintillation counter. All results are calculated as mean logrank Mantel–Cox. Correlation between insulitis and counts per min (CPM) in the presence of antigen and com- sulfatide was performed with a linear regression. For compar- pared with medium alone. Stimulation index (SI) = mean isons between groups a two-tailed unpaired Student’s t test or CPM /mean CPM .AnSI ≥3 is considered a one-way ANOVA with Tukey’s multiple comparisons test. ANTIGEN MEDIUM positive. In three participants, INS-DRiP was not measured. The percentage of participants with a positive T cell response was evaluated using a χ test and two-proportions Z test. Data Animals and diabetes monitoring Female NOD mice were natural log-transformed before analysis if not normally (Taconic Biosciences, Hudson, NY, USA) were kept in a spe- distributed. A p value of less than 0.05 was considered signif- cific pathogen-free (SPF) animal facility (temperature 22°C, icant. *p <0.05; **p < 0.01; ***p <0.001; ****p < 0.0001. 12 h light cycle, air change 16 times per h and humidity 55 ± 10%). Animal experiments were approved by the Animal Experiments Inspectorate, Ministry of Food, Agriculture and Results Fisheries of Denmark (reference 2012-15-2934-00086) and experiments performed according to international guidelines Sulfatide is reduced in human pancreatic islets at the onset of for the care and use of laboratory animals. The mice had free type 1 diabetes To study sulfatide levels in human islets at the access to drinking water and standard Altromin 1320 diet onset of type 1 diabetes, a sulfatide specific antibody was (Altromin, Lage, Germany) with or without 0.01% fenofibrate employed to compare islet immunostaining. Pancreas biopsies (Sigma, St Louis, MO, USA). The mice were inspected week- from individuals with newly diagnosed type 1 diabetes includ- ly for diabetes from an age of 84 days using FreeStyle Lite ed in the DiViD study, had reduced sulfatide staining as 1654 Diabetologia (2018) 61:1650–1661 ab Insulin Sulfatide Control Control T1D T1D Glucagon cd Insulin Alpha cell Sulfatide Merged Beta cell Fig. 1 Sulfatide is present in beta cells and is lost from insulin-positive sulfatide of three healthy control participants and three newly diagnosed islets at the onset of type 1 diabetes. (a) Immunohistochemical staining of individuals with type 1 diabetes from the UK. Scale bar, 50 μm. (c) insulin and sulfatide from a control participant without diabetes and a Immunofluorescent staining of a pancreas without diabetes showing that patient with new-onset type 1 diabetes from DiViD. The pictures represent sulfatide is expressed in beta cells. Scale bar, 30 μm. (d) Electron micros- astandardislet as foundinall sixDiViD cases. Immunohistochemistry copy on an isolated pancreatic human islet stained for sulfatide. Sulfatide is shows pronounced insulin staining and no sulfatide in individuals with type localised to insulin granules in beta cells. T1D, type 1 diabetes 1 diabetes. Scale bar, 50 μm. (b) Immunohistochemical staining of compared with control participants without diabetes (Fig. 1a). control group. This loss of sulfatide was observed in all six Of the islets in the six DiViD individuals with diabetes, 63% individuals studied and was confirmed in a separate cohort of were sulfatide positive and had a relative sulfatide staining individuals with type 1 diabetes from the Exeter Archival intensity of 23% SEM ±6% (p < 0.0001) compared with the Diabetes Biobank (Fig. 1b). SPTLC2 ORMDL2 SPTSSA CERS2 CGT CERK UGCG -50 -100 * **** **** ** ** ** * ARSK B3GALT4 B3GALT5 B4GALT1 SLC1A4 SLC7A10 ** -50 -100 ** * * ** * Fig. 2 Altered expression of enzymes involved in sphingolipid metabo- shown. Solid line at zero represents control average. Control (n=18); lism at the onset of type 1 diabetes. RNA was isolated from laser-dissect- white squares, type 1 diabetes (n=5); black triangles, type 2 diabetes ed islets and analysed by microarray. The individuals with type 1 diabetes (n=8). *p<0.05; **p<0.01; ****p<0.0001. One-way ANOVA with are from the DiViD study. The mean difference in per cent ± SEM is Tukey’smultiplecomparisons test Per cent change Per cent change Diabetologia (2018) 61:1650–1661 1655 Analysis of multiple islets from four pancreases revealed an altered distribution of complex glycosphingolipids in type that sulfatide was found only in beta cells, but absent in alpha 1 diabetes, with increased expression of the cells (Fig. 1c), which was confirmed by electron microscopy galactosyltransferase, B3GALT5 (which encodes β-1,3- (Fig. 1d). Sulfatide was occasionally detectable in islet cells galactosyltransferase 5) (35%, p = 0.002), whereas B3GALT4 negatively for both insulin and glucagon (ESM Fig. 1). (encoding β-1,3-galactosyltransferase 4) and B4GALT1 (β-1,4- galactosyltransferase 1) were downregulated (by 39%, p =0.01 Reduced expression of enzymes involved in sphingolipid me- and 34%, p = 0.02, respectively). Finally, we observed a re- tabolism in human pancreatic islets at the onset of type 1 duced expression of two amino acid transporters; SLC1A4 diabetes A microarray analysis was performed to examine (encoding solute carrier family 1 member 4) (33%, p =0.009) the expression of enzymes involved in sphingolipid metabolism and SLC7A10 (solute carrier family 7 member 10) (32%, p = in human islets. RNAwas isolated from the islets of individuals 0.01) which facilitate the uptake of the sphingolipid precursor with new onset type 1 diabetes (DiViD), individuals with type 2 serine into cells . None of the examined genes had altered diabetes and control participants without diabetes. RNA levels expression in type 2 diabetes compared with control of 70 genes involved in sphingolipid metabolism were evaluat- participants. ed and 13 genes were found to have significantly altered ex- pression at the onset of type 1 diabetes compared with control SNPs in promoter regions of enzymes involved in participants (Fig. 2). SPTLC2 (which encodes serine sphingolipid metabolism associate with the development of palmitoyltransferase long chain base subunit 2), a subunit of type 1 diabetes Next, we tested whether SNPs in genes in- serine palmitoyltransferase (SPT) which catalyses the first step volved in sphingolipid metabolism were associated with ge- in sphingolipid synthesis , was reduced (by 31%, p =0.04). netic predisposition to type 1 diabetes. GWAS have identified Similarly, the expression of the SPT inhibitor ORMDL2 around 50 loci that influence the risk of developing type 1 (encoding ORMDL sphingolipid biosynthesis regulator 2) diabetes but the disease-promoting genes at these loci often  and activator SPTSSA (encoding serine remain unknown . We examined regions 100 kb upstream palmitoyltransferase small subunit A)  were also reduced and downstream of the transcriptional start site of 70 genes 53%, p < 0.0001; 56%, p < 0.0001, respectively. Expression of involved in sphingolipid metabolism (Table 1). A p value enzymes involved in the generation and modification of cer- <0.02 was selected to identify all SNPs associated with type amide were also reduced, including ceramide synthase 2 1 diabetes, as previously described . SNPs mapping to the (CERS2) which was decreased (26%, p = 0.004). Expression HLA regions were excluded from the analysis. RegulomeDB, of ceramide galactosyltransferase (CGT, also known as which ranks SNPs based on the likelihood of the SNP UGT8), ceramide kinase (CERK)and ceramide influencing gene expression, was used with a cut-off of ≤3 glucosyltransferase (UGCG) were similarly diminished; by to prioritise SNPs with a likely regulatory activity . We 30%, p =0.005; 39%, p = 0.003 and 37%, p = 0.03, respective- identified eight genes with SNPs associated with an increased ly. We also found a reduced expression of lysosomal risk of type 1 diabetes and a RegulomeDB score ≤3(Table 1) arylsulfatase K (ARSK;41%, p = 0.004). Our results indicate and OR up to 1.47. Five genes coding for enzymes involved in Table 1 Genes related to sphingolipid metabolism are in type 1 diabetes-associated genetic regions Gene Total type 1 diabetes SNPs OR p value cis-eQTL p value SNPs (p<0.02) (RegulomeDB ≤3) (RegulomeDB ≤3) (RegulomeDB ≤3) (tissue/cell-line) −8 −11 ORMDL3 155 36 1.20 (rs75290103) 1.20×10 (rs12150079) 2.6×10 (whole blood) −10 SPHK2 82 18 1.13 (rs281388) 5.28×10 (rs33988101) 0.032 (whole blood); −4 8.4×10 (cells: transformed fibroblasts) −5 B4GALNT1 54 14 1.47 (rs41292013) 9.43×10 (rs775251) – −8 −7 SLC1A5 50 14 1.16 (rs10412340) 4.72×10 (rs402072) 1.4×10 (cells: transformed fibroblasts) −5 GALC 42 3 1.06 (rs17798191) 0.01 (rs10139328) 2.3×10 (cells: EBV-transformed lymphocytes); 0.033 (whole blood) −3 −3 PPARD 27 4 1.13 (rs7744392) 8.74×10 (rs7744392) 2.7×10 (cells: transformed fibroblasts) PPARG 4 1 1.17 (rs77040839) 0.018 (rs77040839) – −4 −5 B4GALT1 1 1 1.12 (rs7019909) 5.57×10 (rs7019909) 7.7×10 (cells: transformed fibroblasts); 0.008 (cells: EBV-transformed lymphocytes) Type 1 diabetes-associated SNPs (p < 0.02) were identified in eight genes (±100 kb) involved in sphingolipid metabolism. Genes located in proximity to HLA regions have been excluded. Genes are ranked according to the total number of type 1 diabetes-associated SNPs. The RegulomeDB score, which ranks SNPs based on the likelihood of the SNP influencing gene transcription (the lower the more likely), was used with a cut-off ≤3 to prioritise SNPs with a likely regulatory function. In genes with more than one SNP, the highest OR/lowest p value is mentioned. OR is based on the minor allele for B4GALNT1, PPARG and SLC1A5 and the major allele for B4GALT1, GALC, ORMDL3, PPARD and SPHK2. The last column reports the cis-eQTL p value in disease relevant tissues/cell lines for the SNP with the lowest p value 1656 Diabetologia (2018) 61:1650–1661 sphingolipid biosynthesis (ORMDL3 [ORMDL sphingolipid Increased genetic risk defined by sphingolipid-related SNPs is biosynthesis regulator 3], SPHK2 [sphingosine kinase 2], associated with reduced proliferation of islet-specific T cells in B4GALNT1 [β-1,4-N-acetyl-galactosaminyltransferase 1], individuals with type 1 diabetes We wanted to evaluate GALC [galactosylceramidase] and B4GALT1), two transcrip- whether the identified SNPs could affect islet autoimmunity. tion factors (peroxisome proliferator-activated receptor The most promising SNPs (lowest p value and highest OR for (PPARs) D and G), which regulate the expression of enzymes each gene, as shown in Table 1) were selected and a GRS in sphingolipid metabolism  and the amino acid transport- based on the number of risk alleles and OR per SNP was er SLC1A5 (solute carrier family 1 member 5), which is in- computed. The GRS was correlated to proliferation of T cells volved in the uptake of sphingolipid precursor L-serine in response to islet autoantigens GAD65, PPI, IA-2 and INS- (Table 1). Of these ORMDL3 has been previously described DRiP  as measured by the SI. A cohort of 71 individuals . Next, we used cis-eQTLs to evaluate the predicted effect with type 1 diabetes were divided between three risk groups: of the SNPs on the expression levels of their associated gene. low genetic risk (GRS = 0.11–0.14, n = 20); intermediate We integrated data from pre-calculated cis-eQTLs . For (GRS = 0.14–0.16, n = 37) and high (GRS > 0.16, n =14). six genes (ORMDL3, GALC, SPHK2, SLC1A5, PPARD and When comparing the GRS with the proliferation against all B4GALT1) the SNPs with the strongest association with type islet autoantigens (SI =SI +SI +SI +SI SUM GAD65 PPI IA-2 INS- 1 diabetes (lowest p value) also acted as cis-eQTLs, suggest- ) we surprisingly found that islet-specific T cells from DRiP ing that these SNPs regulate the expression of their associated intermediate- and high-risk patients proliferated less than genes (Table 1). those with a low-risk (p = 0.047 and p =0.017, respectively; a b * 6 ** ** 2 0 -1 -2 0 -3 High High High High Low Intermediate High Low Intermediate Low Intermediate Low Intermediate Low Intermediate PPI INS-DRiP IA-2 GAD65 cd e 6 6 * * 4 4 0 0 -2 -2 0 -4 -4 Non-carriers Carriers Non-carriers Carriers PPI INS-DRiP IA-2 GAD65 Fig. 3 Sphingolipid-related SNPs associate with cellular islet autoimmu- autoantigen. Dashed line indicates SI=3. (c) Percentage of patients within nity in individuals with type 1 diabetes. Proliferation of Tcells specific for each risk group with positive T cell proliferation responses (SI ≥3) plotted GAD65, PPI, IA-2 and INS-DRiP in PBMCs freshly isolated from indi- per islet-autoantigen. Light grey bars, low GRS; medium grey bars, in- viduals with type 1 diabetes. Patients were divided into risk groups based termediate GRS; dark grey bars, high GRS. (d, e) Proliferation of PPI- on sphingolipid-related genetic risk: low (GRS=0.11–0.14, n=20); inter- specific T cells in heterozygous and homozygous carriers of the (d) mediate (GRS=0.14–0.16, n=37) and high (GRS >0.16, n=14). Data were rs12150079 or (e) rs33988101 risk allele vs non-carriers of the respective normalised by natural log-transformation. SI ≥3 is considered positive. risk allele. *p<0.05; **p<0.01. One-way ANOVA with Tukey’s multiple Tukey boxplots are shown. (a) Cumulative proliferation of islet-specific T comparisons, χ two-proportions Z test and two-tailed unpaired Student’s cells (SI =SI +SI +SI +SI ) in patient-risk groups. t test SUM GAD65 PPI IA-2 INS-DRiP (b) Proliferation of T cells in patient-risk groups plotted per islet- log (SI ) e SUM Patients with positive SI (%) log (SI) log (SI) log (SI) e Diabetologia (2018) 61:1650–1661 1657 Fig. 3a). Focusing on individual islet autoantigens, we found treated with fenofibrate, which is known to increase sulfatide that PPI-specific T cells proliferated less in intermediate- and levels in several organs . NOD mice, which usually de- high-risk patients compared with those at low-risk (p =0.007 velop insulitis at age 4 weeks , were treated with and p = 0.002, respectively) and IA-2-specific T cells prolifer- fenofibrate from an age of 3 weeks till 35 weeks. ated less in high-risk patients compared with those at low-risk Development of diabetes was prevented in the fenofibrate (p = 0.018), with a non-significant difference in the interme- treated mice (0/15 (0%) vs 11/15 (73%) in the control group diate group (p = 0.31); there were also non-significant differ- (p < 0.0001; Fig. 4a). Fenofibrate also reduced the degree of ences in proliferation for INS-DRiP among both high- and insulitis (p = 0.0006; Fig. 4b,c) and increased the expression intermediate risk groups compared with the low-risk group of sulfatide in islets (p =0.007; Fig. 4d). There was an inverse (p = 0.159 and p = 0.069, respectively), but not for GAD65 correlation between sulfatide and insulitis score (p =0.0004, (Fig. 3b). Looking at the percentage of patients in each risk r = 0.72; Fig. 4e) in the experimental animals. Fenofibrate group with positive proliferation responses against individual treatment initiated after onset of diabetes reversed diabetes islet autoantigens (SI ≥3), we found that fewer patients in the in 46% (6/13) NOD mice after 3 weeks of treatment. high-risk group responded to PPI compared with the low-risk group (p = 0.003; Fig. 3c). Moreover, even though there were no significant differences in absolute INS-DRiP-specific Tcell Discussion proliferation, the percentage of patients with a positive INS- DRiP response was significantly lower in the intermediate- The present study provides human data demonstrating that risk group than in the low-risk group (p = 0.034). sphingolipid metabolism contributes to genetic disease predis- Furthermore, we found that heterozygous and homozygous position and autoimmunity and that the onset of type 1 diabe- carriers of the risk alleles rs12150079 (ORMDL3)and tes is associated with altered sphingolipid metabolism and rs33988101 (SPHK2) had lower levels of T cell autoimmunity reduced expression of sulfatide in pancreatic islets (Fig. 5). (p = 0.042 and p = 0.017, respectively; Fig. 3d,e). Sulfatide is known to participate in the regulation of first- phase insulin secretion  and it is possible that the observed Increasing sulfatide levels in mice pancreatic islets is associ- loss of pancreatic sulfatide may contribute to the loss of first- ated with prevention of autoimmune diabetes in NOD mice phase insulin secretion seen during the development of type 1 Based on these findings, we considered pharmacological up- diabetes . We notice a finer granulation in the islets of regulation of pancreatic sulfatide levels as a possible therapeu- individuals with type 1 diabetes; whether this might, in part, tic approach in type 1 diabetes. NOD mice were therefore influence the reduced sulfatide-staining that is seen in insulin Fig. 4 Fenofibrate prevents a b diabetes in NOD mice. NOD Insulitis score 4 mice were treated with fenofibrate Insulitis score 3 80 80 or control from an age of 3 weeks. Insulitis score 2 (a) Diabetes incidence in the Insulitis score 1 60 60 experimental groups (n=15); Insulitis score 0 40 40 dotted line, fenofibrate; solid line, control. (b) Percentage 20 20 distribution of insulitis. Insulitis score (n=6 per group) at age 13 0 0 0 50 100 150 200 weeks on a scale from 0 (no insulitis) to 4 (above 50% Age (days) infiltration). (c) Average insulitis score. (d) Sulfatide score (n=6 per group). (e) Correlation between cd e insulitis score and sulfatide with 4 4 4 r = 0.72 *** linear regression. Show is mean ± p = 0.0004 ** SEM. **p<0.01; ***p<0.001. 3 3 3 Logrank Mantel–Cox, two-tailed unpaired Student’s t test and 2 2 2 linear regression 1 1 1 0 0 0 Sulfatide 1 2 3 Control Control Fenofibrate Fenofibrate Control Fenofibrate Insulitis Diabetes free (%) Sulfatide Per cent of islets Insulitis 1658 Diabetologia (2018) 61:1650–1661 granules, is unknown . Reduced levels of sulfatide might of disease . Altered levels of ceramide and sphingosine-1- be explained by reduced enzyme expression as suggested by phosphate could also affect islet function through regulation the transcriptome analysis, which showed reduced expression of the sphingolipid rheostat . of several enzymes involved in sphingolipid metabolism in To support these findings, GWAS data were interrogated islets from individuals with newly diagnosed type 1 diabetes and SNPs in the promoter regions of eight genes influencing (Fig. 5a). The reduced expression of CERS2 is particularly sphingolipid levels were identified (Fig. 5b). Among these, interesting since this points towards altered hydrocarbon chain the OR of 1.47 calculated for B4GALNT1 ranks it among the lengths of beta cell sphingolipids in type 1 diabetes, with highest risk genes implicated in the predisposition to type 1 lower amounts of long chains . Long chain (C24) sulfatide diabetes (ESM Fig. 3). All SNPs identified here correlated is protective against diabetes development in NOD mice . with predisposition to type 1 diabetes but not type 2 diabetes. Furthermore, a low-grade enteroviral infection was found in It should be noted that the SNPs identified could be inhibitory the beta cells of all DiViD participants  and this could be or activating and so the overall effect of these SNPs on linked to lower expression of C24 sulfatide which stimulates sphingolipid composition is difficult to predict. the natural killer (NK) T cells that normally eliminate diabe- The most promising SNPs were found to be associated togenic viruses [41, 42]. We did not find any changes in the with lower rates of T cell proliferation when these cells expression of arylsulfatase A, which degrades sulfatide in the were presented with beta cell autoantigens (Fig. 5c). This lysosome , suggesting that the reduced amount of effect was linked with autoimmunity to PPI and to a less- sulfatide did not result from enhanced rates of lipid degrada- er degree IA-2 and INS-DRiP, but not GAD65. A possible tion. The changed expression of B3GALT5, B3GALT4 and explanation for this seemingly paradoxical finding is that B4GALT1 suggest that the development of type 1 diabetes is sulfatide is involved in PPI folding  and likely the associated with changes in the composition of islet formation of INS-DRiP. T cells recognise folded PPI and glycosphingolipids with increases in the neolacto/lacto series so impaired folding due to less sulfatide would lead to a and decreased levels of gangliosides (ESM Fig. 2). An altered lower immune response against PPI. A lack of sulfatide amount of gangliosides could play a role in type 1 diabetes on the other hand would not affect autoimmunity to aetiology as ganglioside autoantibodies are found at the onset GAD65. Insulitis Control T1D c ad SNP Enzymes Fenofibrate A G G A Fig. 5 Sphingolipid metabolism is connected with type 1 diabetes. sphingolipid metabolism increases the risk for developing type 1 diabetes. Overview of how sphingolipid metabolism is related to the development (c) These genetic polymorphisms are associated with lower rates of T cell of type 1 diabetes. (a) There is a reduced amount of sulfatide in islets of proliferation when presented to beta cell autoantigens. (d)Fenofibrate individuals with newly diagnosed type 1 diabetes (images taken from Fig. stimulates sulfatide production in islets of NOD mice. (e)Thisis associ- 1b). This is related to altered expression of enzymes involved in the ated with complete protection against diabetes and a lower degree of biosynthesis of sphingolipids in islets. (b) Genetic polymorphisms in insulitis in NOD mice; insulitis image adapted from  (original cour- the promoter region of eight genes encoding enzymes involved in tesy of A. van Halteren) with permission of Springer Nature Diabetologia (2018) 61:1650–1661 1659 Open Access This article is distributed under the terms of the Creative Fenofibrate has been in use for decades to reduce LDL- Commons Attribution 4.0 International License (http:// cholesterol, triacylglycerol and cholesterol levels and has creativecommons.org/licenses/by/4.0/), which permits unrestricted use, shown beneficial effects on the prevention of diabetic neurop- distribution, and reproduction in any medium, provided you give appro- athy and retinopathy . Here we demonstrate that NOD priate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. mice were protected from insulitis and diabetes by early ex- posure to fenofibrate and that this correlated with an increase in islet sulfatide levels (Fig. 5d,e). The positive effect of References fenofibrate, however, cannot be solely credited to the in- creased amount of sulfatide as fenofibrate is also likely to 1. van Belle TL, Coppieters KT, von Herrath MG (2011) Type 1 affect other aspects of lipid biology. Previous studies have diabetes: etiology, immunology, and therapeutic strategies. otherwise shown that the ceramide synthase inhibitor Physiol Rev 91:79–118 FTY720  prevented diabetes development in NOD mice 2. Leete P, Willcox A, Krogvold L et al (2016) Differential insulitic , highlighting the diverse roles of different sphingolipids profiles determine the extent of beta-cell destruction and the age at onset of type 1 diabetes. Diabetes 65:1362–1369 in diabetes pathology. 3. Krogvold L, Wiberg A, Edwin B et al (2016) Insulitis and charac- In conclusion, we provide human evidence of an altered terisation of infiltrating T cells in surgical pancreatic tail resections islet sphingolipid metabolism in type 1 diabetes. Increasing from patients at onset of type 1 diabetes. Diabetologia 59:492–501 sulfatide levels prevents diabetes in NOD mice suggesting that 4. Coppieters KT, Dotta F, Amirian N et al (2012) Demonstration of upregulation of sulfatide biosynthesis may represent a prom- islet-autoreactive CD8 T cells in insulitic lesions from recent onset and long-term type 1 diabetes patients. J Exp Med 209:51–60 ising therapeutic route in type 1 diabetes. 5. Krogvold L, Skog O, Sundstrom G et al (2015) Function of isolated pancreatic islets from patients at onset of type 1 diabetes: insulin Data availability The RNA expression data is available online at https:// secretion can be restored after some days in a nondiabetogenic www.dropbox.com/s/93mk5tzl5fdyo6b/Abnormal%20islet% environment in vitro: results from the DiViD study. Diabetes 64: 20sphingolipid%20metabolism%20in%20type%201%20diabetes%2C% 2506–2512 20RNA%20expression.xlsx?dl=0 6. Malmegrim KC, de Azevedo JT, Arruda LC et al (2017) A list of SNPs identified is available at https://www.dropbox.com/s/ Immunological balance is associated with clinical outcome after yfojma9xanpp2ju/Abnormal%20islet%20sphingolipid%20metabolism% autologous hematopoietic stem cell transplantation in type 1 diabe- 20in%20type%201%20diabetes%20SNP.xlsx?dl=0 tes. Front Immunol 8:167 7. Maceyka M, Spiegel S (2014) Sphingolipid metabolites in inflam- Funding The DiViD study was funded by the South-Eastern Norway matory disease. Nature 510:58–67 Regional Health Authority (grant to KD-J), the Novo Nordisk 8. Chen Y, Liu Y, Sullards MC, Merrill AH Jr (2010) An introduction Foundation (grant to KD-J), and through the PEVNET (Persistent Virus to sphingolipid metabolism and analysis by new technologies. Infection in Diabetes Network) Study Group funded by the European NeuroMolecular Med 12:306–319 Union’s Seventh Framework Programme (FP7/2007-2013) under grant 9. Yamaji T, Hanada K (2015) Sphingolipid metabolism and agreement number 261441 PEVNET. Additional grant support from National Institutes of Health, UC4 DK104155, the JDRF (47-2013- interorganellar transport: localization of sphingolipid enzymes 520), Dutch Diabetes Research Foundation, and Stichting DON and lipid transfer proteins. Traffic 16:101–122 (DFN2013.40.1693), and Kirsten and Freddy Johansens Fond. 10. Boslem E, Meikle PJ, Biden TJ (2012) Roles of ceramide and sphingolipids in pancreatic beta-cell function and dysfunction. Islets 4:177–187 Duality of interest The authors declare that there is no duality of interest 11. Veret J, Bellini L, Giussani P, Ng C, Magnan C, Le Stunff H (2014) associated with this manuscript. Roles of sphingolipid metabolism in pancreatic beta cell dysfunc- tion induced by lipotoxicity. J Clin Med 3:646–662 Contribution statement KB and LJH conceived and planned the study. 12. Ng ML, Wadham C, Sukocheva OA (2017) The role of KD-J, LK, and KFH conceived and planned the DiViD study together. sphingolipid signalling in diabetesassociated pathologies LK, KFH, and KD-J provided the DiViD tissue. LJH and JPH planned (Review). Int J Mol Med 39:243–252 and performed the immunohistochemistry on the DiViD tissue. LJH and 13. Buschard K, Blomqvist M, Mansson JE, Fredman P, Juhl K, MAR planned, performed and analysed the immunofluorescence study. Gromada J (2006) C16:0 sulfatide inhibits insulin secretion in rat MAR performed the immunohistochemistry on tissue from the UK. KB beta-cells by reducing the sensitivity of KATP channels to ATP and NGM planned and supervised the study on tissue from the UK. KB inhibition. Diabetes 55:2826–2834 planned and performed the electron microscopy. ICG and CEM planned, 14. Buschard K, Bracey AW, McElroy DL et al (2016) Sulfatide pre- performed and analysed the microarray analysis. LK, KFH, and KD-J serves insulin crystals not by being integrated in the lattice but by helped plan and supervised the microarray study. LJH and KB selected stabilizing their surface. J Diabetes Res 2016:6179635 the genes to study and helped with the analysis of the microarray study. SK and FP planned, performed and analysed the GWAS study together 15. Krogvold L, Edwin B, Buanes T et al (2014) Pancreatic biopsy by with LJH and KB. LJH and KB designed, performed experiments and minimal tail resection in live adult patients at the onset of type 1 analysed data in the NOD mice study. JPH performed the insulitis and diabetes: experiences from the DiViD study. Diabetologia 57:841– sulfatide scoring of the NOD mice. LAC, BPCK and BOR designed and 843 executed the immunological studies. LJH wrote the manuscript with input 16. Osterbye T, Funda DP, Fundova P, Mansson JE, Tlaskalova- from all authors. All authors approved the final manuscript. KB is the Hogenova H, Buschard K (2010) A subset of human pancreatic guarantor of this study. beta cells express functional CD14 receptors: a signaling pathway 1660 Diabetologia (2018) 61:1650–1661 for beta cell-related glycolipids, sulfatide and beta- 34. Mirza AH, Kaur S, Brorsson CA, Pociot F (2014) Effects of GWAS-associated genetic variants on lncRNAs within IBD and galactosylceramide. Diabetes Metab Res Rev 26:656–667 17. Fredman P, Mattsson L, Andersson K et al (1988) Characterization T1D candidate loci. PLoS One 9:e105723 of the binding epitope of a monoclonal antibody to sulphatide. 35. Baranowski M, Gorski J (2011) Heart sphingolipids in health and Biochem J 251:17–22 disease. Adv Exp Med Biol 721:41–56 18. Campbell-Thompson M, Wasserfall C, Kaddis J et al (2012) 36. Barrett JC, Clayton DG, Concannon P et al (2009) Genome-wide Network for Pancreatic Organ Donors with Diabetes (nPOD): de- association study and meta-analysis find that over 40 loci affect risk veloping a tissue biobank for type 1 diabetes. Diabetes Metab Res of type 1 diabetes. Nat Genet 41:703–707 Rev 28:608–617 37. Nakajima T, Kamijo Y, Yuzhe H et al (2013) Peroxisome 19. Richardson SJ, Rodriguez-Calvo T, Gerling IC et al (2016) Islet cell proliferator-activated receptor alpha mediates enhancement of gene hyperexpression of HLA class I antigens: a defining feature in type expression of cerebroside sulfotransferase in several murine or- 1 diabetes. Diabetologia 59:2448–2458 gans. Glycoconj J 30:553–560 20. Wu J, Kakoola DN, Lenchik NI, Desiderio DM, Marshall DR, 38. Crevecoeur I, Gudmundsdottir V, Vig S et al (2017) Early differ- Gerling IC (2012) Molecular phenotyping of immune cells from ences in islets from prediabetic NOD mice: combined microarray young NOD mice reveals abnormal metabolic pathways in the early and proteomic analysis. Diabetologia 60:475–489 induction phase of autoimmune diabetes. PLoS One 7:e46941 39. Sosenko JM, Skyler JS, Beam CA et al (2013) Acceleration of the 21. Onengut-Gumuscu S, Chen WM, Burren O et al (2015) Fine map- loss of the first-phase insulin response during the progression to ping of type 1 diabetes susceptibility loci and evidence for type 1 diabetes in diabetes prevention trial-type 1 participants. colocalization of causal variants with lymphoid gene enhancers. Diabetes 62:4179–4183 Nat Genet 47:381–386 40. Laviad EL, Albee L, Pankova-Kholmyansky I et al (2008) 22. Rosenbloom KR, Sloan CA, Malladi VS et al (2013) ENCODE Characterization of ceramide synthase 2: tissue distribution, sub- data in the UCSC Genome Browser: year 5 update. Nucleic Acids strate specificity, and inhibition by sphingosine 1-phosphate. J Biol Res 41:D56–D63 Chem 283:5677–5684 23. Boyle AP, Hong EL, Hariharan M et al (2012) Annotation of func- 41. Subramanian L, Blumenfeld H, Tohn R et al (2012) NKT cells tional variation in personal genomes using RegulomeDB. Genome stimulated by long fatty acyl chain sulfatides significantly reduce Res 22:1790–1797 the incidence of type 1 diabetes in nonobese diabetic mice 24. Consortium G (2015) Human genomics. The Genotype-Tissue [corrected]. PLoS One 7:e37771 Expression (GTEx) pilot analysis: multitissue gene regulation in 42. Exley MA, Bigley NJ, Cheng O et al (2001) CD1d-reactive T cell humans. Science 348:648–660 activation leads to amelioration of disease caused by diabetogenic 25. Stranger BE, Montgomery SB, Dimas AS et al (2012) Patterns of encephalomyocarditis virus. J Leukoc Biol 69:713–718 cis regulatory variation in diverse human populations. PLoS Genet 43. Doerr J, Bockenhoff A, Ewald B et al (2015) Arylsulfatase A over- 8:e1002639 expressing human iPSC-derived neural cells reduce CNS sulfatide 26. Westra HJ, Peters MJ, Esko T et al (2013) Systematic identification storage in a mouse model of metachromatic leukodystrophy. Mol of trans eQTLs as putative drivers of known disease associations. Ther 23:1519–1531 Nat Genet 45:1238–1243 44. Dotta F, Falorni A, Tiberti C et al (1997) Autoantibodies to the 27. Franken KL, Hiemstra HS, van Meijgaarden KE et al (2000) GM2-1 islet ganglioside and to GAD-65 at type 1 diabetes onset. Purification of his-tagged proteins by immobilized chelate affinity JAutoimmun 10:585–588 chromatography: the benefits from the use of organic solvent. 45. Jessup CF, Bonder CS, Pitson SM, Coates PT (2011) The Protein Expr Purif 18:95–99 sphingolipid rheostat: a potential target for improving pancreatic 28. Kracht MJ, van Lummel M, Nikolic T et al (2017) Autoimmunity islet survival and function. Endocr Metab Immune Disord Drug against a defective ribosomal insulin gene product in type 1 diabe- Targets 11:262–272 tes. Nat Med 23:501–507 46. Osterbye T, Jorgensen KH, Fredman P et al (2001) Sulfatide pro- 29. Hanada K (2003) Serine palmitoyltransferase, a key enzyme of motes the folding of proinsulin, preserves insulin crystals, and me- sphingolipid metabolism. Biochim Biophys Acta 1632:16–30 diates its monomerization. Glycobiology 11:473–479 30. Siow D, Sunkara M, Dunn TM, Morris AJ, Wattenberg B (2015) 47. Wright AD, Dodson PM (2011) Medical management of diabetic ORMDL/serine palmitoyltransferase stoichiometry determines ef- retinopathy: fenofibrate and ACCORD Eye studies. Eye (Lond) 25: fects of ORMDL3 expression on sphingolipid biosynthesis. J Lipid 843–849 Res 56:898–908 48. Berdyshev EV, Gorshkova I, Skobeleva A et al (2009) FTY720 31. Han G, Gupta SD, Gable K et al (2009) Identification of small inhibits ceramide synthases and up-regulates dihydrosphingosine subunits of mammalian serine palmitoyltransferase that confer dis- 1-phosphate formation in human lung endothelial cells. J Biol tinct acyl-CoA substrate specificities. Proc Natl Acad Sci U S A Chem 284:5467–5477 106:8186–8191 49. Yang Z, Chen M, Fialkow LB et al (2003) The immune modulator 32. El-Hattab AW (2016) Serine biosynthesis and transport defects. FYT720 prevents autoimmune diabetes in nonobese diabetic mice. Mol Genet Metab 118:153–159 Clin Immunol 107:30–35 33. Floyel T, Kaur S, Pociot F (2015) Genes affecting beta-cell function 50. Roep BO (2003) The role of T-cells in the pathogenesis of type 1 in type 1 diabetes. Curr Diab Rep 15:97 diabetes: from cause to cure. Diabetologia 46:305–321 Diabetologia (2018) 61:1650–1661 1661 Affiliations 1 2,3 4 5 6,7 8 Laurits J. Holm & Lars Krogvold & Jane P. Hasselby & Simranjeet Kaur & Laura A. Claessens & Mark A. Russell & 9 3,10 8 7 6,11 Clayton E. Mathews & Kristian F. Hanssen & Noel G. Morgan & Bobby P. C. Koeleman & Bart O. Roep & 12 5 2,13 1 Ivan C. Gerling & Flemming Pociot & Knut Dahl-Jørgensen & Karsten Buschard 1 8 The Bartholin Institute, Department of Pathology, Rigshospitalet, Institute of Biomedical and Clinical Sciences, University of Exeter Copenhagen Biocenter, Ole Maaløes Vej 5, 2200 Copenhagen Medical School, Exeter, UK N, Denmark Department of Pathology, University of Florida, Gainesville, FL, Division of Paediatric and Adolescent Medicine, Oslo University USA Hospital, Oslo, Norway Department of Endocrinology, Oslo University Hospital, Faculty of Odontology, University of Oslo, Oslo, Norway Oslo, Norway 4 11 Department of Pathology, Rigshospitalet, Copenhagen, Denmark Department of Diabetes Immunology, Diabetes & Metabolism Research Institute, Beckman Research Institute at the City of Hope, Steno Diabetes Center Copenhagen, Gentofte, Denmark Duarte, CA, USA Department of Immunohaematology & Blood Transfusion, Leiden Department of Medicine, University of Tennessee, Memphis, TN, University Medical Center, Leiden, the Netherlands USA Department of Medical Genetics, University Medical Center, Faculty of Medicine, University of Oslo, Oslo, Norway Utrecht, the Netherlands
Diabetologia – Springer Journals
Published: Apr 18, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera