http://orcid.org/0000-0002-0994-0184
Figures
Abstract
Background
Based on the association of common GLIS3 variants with various forms of diabetes and the biological role of GLIS3 in beta-cells, we sequenced GLIS3 in non-diabetic and diabetic Danes to investigate the effect of rare missense variants on glucose metabolism.
Methods
We sequenced 53 patients with maturity-onset diabetes of the young (MODY), 5,726 non-diabetic participants, 2,930 patients with newly diagnosed type 2 diabetes and 206 patients with glutamic acid decarboxylase antibody (GADA) -positive diabetes.
Results
In total we identified 86 rare (minor allele frequency < 0.1%) missense variants. None was considered causal for the presence of MODY. Among patients with type 2 diabetes, we observed a higher prevalence of rare GLIS3 missense variants (2.5%) compared to non-diabetic individuals (1.8%) (odds ratio of 1.37 (interquartile range:1.01–1.88, p = 0.04)). A significantly increased HbA1c was found among patients with type 2 diabetes and with GADA-positive diabetes carrying rare GLIS3 variants compared to non-carriers of rare GLIS3 variants with diabetes (p = 0.02 and p = 0.004, respectively). One variant (p.I28V) was found to have a minor allele frequency of only 0.03% among patients with type 2 diabetes compared to 0.2% among non-diabetic individuals suggesting a protective function (odds ratio of 0.20 (interquartile range: 0.005–1.4, p = 0.1)), an effect which was supported by publically available data. This variant was also associated with a lower level of fasting plasma glucose among non-diabetic individuals (p = 0.046).
Citation: Sun J, Have CT, Hollensted M, Grarup N, Linneberg A, Pedersen O, et al. (2019) Sequencing reveals protective and pathogenic effects on development of diabetes of rare GLIS3 variants. PLoS ONE 14(8): e0220805. https://doi.org/10.1371/journal.pone.0220805
Editor: Hui-Qi Qu, Children's hospital of philadelphia, UNITED STATES
Received: May 13, 2019; Accepted: July 23, 2019; Published: August 15, 2019
Copyright: © 2019 Sun et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: This project received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement No 667191 (authors awarded: APG, NG and TH). In addition, the study was supported from the Danish Diabetes Academy, the Danish Medical Research Council, the European Foundation for the Study of Diabetes (EFSD), the Novo Nordisk Foundation (Immunometabolism, NNF15CC0018486) and the Danish Diabetes Association. The DD2 cohort was funded by the Danish Agency for Science (grant no. 09-067009 and 09-075724) (authors awarded: JSN and JR). The Novo Nordisk Foundation Center for Basic Metabolic Research is an independent research center at the University of Copenhagen, partially funded by an unrestricted donation from the Novo Nordisk Foundation (www.metabol.ku.dk). The BGI provided support in the form of salaries for author [JS and KK], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section. The iCarbonX provided support in the form of salaries for author [WJ], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. No other funders had any role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.
Competing interests: The authors declare no competing interest. JS and KK are has received support from BGI and WJ from iCarbonX, however, these affiliations do not alter our adherence to PLOS ONE policies on sharing data and materials.
Abbreviations: BMI, body mass index; CI, confidence interval; GADA, glutamic acid decarboxylase antibody; GDM, gestational diabetes mellitus; hsCRP, high-sensitivity CRP; IFG, impaired fasting glycaemia; IGT, impaired glucose tolerance; IQR, Interquartile range; MAF, minor allele frequency; MODY, maturity-onset diabetes of the young; OHA, oral hyperglycaemic agent; OR, Odds ratio; T2D, type 2 diabetes
Introduction
GLIS3 is encoding a member of the Krüppel-like zinc finger protein subfamily GLI-similar 3 which is a transcription factor playing a critical role both as a repressor and activator of transcription [1]. Human GLIS3 exist in two isoforms: Isoform A which is the longest including a total of 930 amino acids, and isoform B which is 155 amino acids shorter at the N-terminus [2]. Isoform A is highly expressed in the pancreas, kidney, and thyroid, whereas the smaller isoform B is strongly expressed in the heart, liver, kidney, and skeletal muscle [3]. The transcriptional regulation by GLIS3 is mediated through an interaction between GLIS-binding sites (GLISBS) in the regulatory region of target genes and zinc finger domains in GLIS3 [1, 4].
Globally Glis3 deficient mice have a decreased beta-cell mass and develop neonatal diabetes in addition to hypothyroidism, and cystic kidney disease [5]. In addition, Glis3 missense variants identified in Goto-Kakizaki rats have been found to increase basal production of insulin but reduce the glucose stimulated insulin secretion in INS1-cells [6].
Also in humans, mutations in GLIS3 are a rare cause of neonatal diabetes as well as congenital hypothyroidism and in some cases in combination with polycystic kidney disease, glaucoma, and hepatic fibrosis in a recessive manner [3, 7]. It has been found for the neonatal diabetes genes GCK and PDX1, that homozygous mutations are a cause of neonatal diabetes and heterozygote mutations results in the monogenic form of diabetes called maturity onset diabetes of the young (MODY) [8, 9]. This may also be the case for GLIS3.
Type 1 diabetes (T1D) and T2D are two genetically distinct diseases, yet GLIS3 is one of the few genes in which common variants have been found to associate with both forms of diabetes [10, 11]. In addition, common GLIS3 variants associate with development of gestational diabetes [12], with increased level of fasting glucose [10] and with altered insulin clearance [13]. Thus, GLIS3 appear to be implicated in various diabetes forms.
This pathogenic effect of GLIS3 variants may be related to GLIS3 being instrumental for an optimal transcriptional activation of Ins2 in conjunction with the transcriptions factors Pdx1, NeuroD1, and MafA [14]. This has been further verified by the observation that mutations in the GLISBS of the insulin promoter are responsible for the development of neonatal diabetes in some patients [14]. However, GLIS3 has also been shown in INS1-cells to stimulate the transcription of genes of importance for apoptosis such as Atg7 and Atg4a [6] as well as the pro-apoptotic BH3-only protein Bim [15].
In humans common and low frequency GLIS3 variants (MAF > 0.1%) have previously been investigated in relation to diabetes [10]. Thus, we sequenced the coding region of GLIS3 using next-generation sequencing (NGS) in order to investigate the effect of rare GLIS3 missense variants in a well characterised Danish population. The sequencing was completed in 53 MODY patients with an unknown genetic etiology (MODYX), 5,726 non-diabetic individuals, 2,930 patients newly diagnosed with T2D and 206 patients with GADA-positive diabetes with the aim to investigate 1) if GLIS3 missense variants is involved in MODYX, 2) if rare GLIS3 missense variants increase the risk of T2D development and 3) if rare GLIS3 missense variants affect measures of glucose metabolism.
Materials and methods
Population
Individuals in the present study include: 1) 53 MODYX probands recruited from the outpatient clinic at Steno Diabetes Center, Copenhagen, Denmark, 2) 5,726 non-diabetic individuals recruited from the Danish population-based Inter99 study [16, 17] including 1,157 prediabetic individuals (having impaired fasting glucose (IFG) and/or impaired glucose tolerance (IGT)) and 4,569 glucose tolerant individuals (NGT) classified according the WHO 1999 criteria using a 2-hour oral glucose tolerance test (OGTT) [18]; 3) 2,930 patients with newly-diagnosed T2D recruited through the DD2-cohort [19]; and 4) 206 patients with GADA-positive diabetes recruited from the DD2-cohort (n = 85) and at Hospital Lillebaelt, Denmark (n = 121). A clinical description of included individuals can be found in Table 1.
Patients with newly diagnosed T2D were GADA-negative and had a fasting serum C-peptide concentration > 150 pmol/l within 1.5 years from diabetes diagnosis if C-peptide was available. GADA-positive patients were selected based on a GADA>30 IU/ml.
Prior to participation, written informed consents were obtained from all participants. The study design was in accordance with the ethical scientific principles of the Helsinki Declaration II and approved by The Scientific Ethics Committee of the Capital Region of Denmark (Inter99: KA-98155) and by the Danish National Ethical Committee on Health Research (DD2: S-20100082).
Anthropometrics measurements
Body weight, height, waist, and hip circumference were measured with light indoor clothes and without shoes. Body mass index (BMI) was defined as body mass divided by body height in units of kg/m2.
Metabolic measurements
Inter99 participants and participants recruited at Hospital Lillebaelt, Denmark.
The blood samples were drawn after a 12h fast for measures of glycated hemoglobin (HbA1c), plasma glucose, serum insulin, serum C-peptide, total cholesterol and serum triglyceride [16, 20]. In addition, individuals without known diabetes underwent a standard 75g OGTT with samples drawn at 30, and 120 minutes measured for plasma glucose, serum insulin and C-peptide. Serum insulin levels (excluding des-31,32 and intact proinsulin) were measured using the AutoDELFIA insulin kit (Perkin-Elmer, Wallac, Turku, Finland). Plasma glucose was analysed using a glucose oxidase method (Granutest; Merck, Darmstadt, Germany)[17]. Concentrations of serum triglycerides and total cholesterol were analysed using enzymatic colorimetric methods (GPO-PAP and CHOD-PAP, Roche Molecular Biochemicals, Germany). HbA1c was measured using ion-exchange high performance liquid chromatography (normal reference range: 4.1–6.4%) [21].
Participants from the DD2-cohort.
Measures of BMI, and routine laboratory measurements such as fasting plasma glucose, fasting serum C-peptide and GADA were extracted from the Danish Diabetes Database for Adults [22].
Estimates of glucose metabolism
Oral glucose-stimulated insulin response was measured as the insulinogenic index. Homeostatic model assessment of insulin resistance (HOMA-IR) index measured as previous reported [23].
Targeted resequencing
A customized oligonucleotide probe was designed including all coding regions of GLIS3 as previously described [24]. Genomic DNA was extracted from human leucocyte nuclei. Target regions were captured and subsequently underwent library construction. The captured DNA libraries were sequenced using the Illumina HiSeq2000 as paired-end 90 bp reads (following the manufacturer’s standard cluster generation and sequencing protocols). GLIS3 was covered with a minimum depth of over 20X and a mean depth of the target region of 166X. Qualified reads were aligned to the reference of human genome (UCSC hg19) using the Burrows-Wheeler Aligner tool (http://bio-bwa.sourceforge.net), and single-nucleotide polymorphisms and indels were identified using the Genome Analysis Toolkit (https://www.broadinstitute.org/gatk/). The variants were annotated using Annovar [25] with variants annotated according to transcript NM_001042413.
The linkage disequilibrium (LD) structure between presently identified and previously investigated variants was calculated using LDlink [26].
Statistical analysis
The statistical differences in carrier-frequency between cases and controls were calculated using chi-squared, a kernel-based adaptive cluster (KBAC) test as well as an analysis which collapses and combines rare variants (CMC) [27]. The CMC analysis included adjustment for principal component 1–4 in order to adjust for population stratification.
However, due to the low number of carriers of each group, the statistical difference between cases and control for variant p.I28V was calculated using a fisher’s exact test. Quantitative trait analyses were performed using linear regression using additive genetic models with adjustment for age and sex. Traits were all q-transformed prior to analysis. Statistical analyses were performed using RStudio software version R-3.4.1 (version 3.2.3; R Foundation for Statistical Computing, Boston, MA, USA) except KBAC and CMC which was performed using rvtests [28]. Burden-test for the association between variants in GLIS3 and glucose tolerance status did not undergo correction for multiple testing as this was a test of our primary hypothesis and a p-value < 0.05 was considered statistically significant. In contrast, quantitative analyses were corrected for multiple testing. Thus, for quantitative traits the adjusted level of significance is p > 0.006 (0.05/9).
Estimation of functionality
The pathogenicity of missense variants was evaluated using the Combined Annotation Dependent Depletion (CADD) score where a PHRED-scaled CADD score above 10 predicts pathogenicity in the top 10 percentile of all variants and a score above 20 predicts the top 1 percentile [29].
Results
The coding regions of GLIS3 were sequenced in 53 MODYX patients, 2,930 patients with T2D, 5,724 non-diabetic individuals and 206 patients with GADA-positive diabetes. A total 88 missense variants were identified of which three were common variants (MAF >1%), six were low frequency variants (MAF between 1% and 0.1%) and the vast majority (n = 79) were rare (MAF<0.1%) (S1 Table).
Three variants were found among the patients having MODYX. These included the common variants p.G313A and p.D512E and the rare variant p.V916L. However, the rare variant p.V916L was also found among three patients with T2D and five glucose tolerant individuals (Table 1), thus, none of these variants can be considered causal for the development of MODY.
Among the remaining patients with T2D and participants without diabetes, we found three common variants (p.G313A, p.P456Q, p.D512E) and six low frequency variants (p.P282A, p.S298Y, p.P364S, p.Q397H, p.H400R and p.E515D) which have previously been investigated in a large GWAS study combining GWAS data from close to 900,000 individuals [30]. Thus, we did not further investigate the effect of neither common nor low frequency variants but focussed on the previously un-investigated effect of rare variants in GLIS3.
A total of 79 rare variants were found among 100 non-diabetic individuals including three individuals carrying two variants and among 70 patients with T2D of whom two individuals are carrying two variants. Thus, the prevalence of rare GLIS3 variants was 2.4% among patients with T2D compared to 1.7% among non-diabetic individuals revealing a slight enrichment of rare GLIS3 variants in patients having T2D with odds ratio (OR) of 1.37 (IQR: 1.01–1.85, p = 0.04, Table 2).
The effect of GLIS3 mutations on glucose metabolism was further explored and in patients with T2D, levels of HbA1c were nominally elevated to 7.20% (SD:1.69) among carriers of rare GLIS3 variants compared to 6.80% (SD:1.23) among non-carriers, p = 0.02 (Table 3). As variants in GLIS3 also have been found to associate with T1D, we investigated the effect of rare GLIS3 variants in among patients classified as T2D, yet being GADA-positive (n = 206). Three carriers of GLIS3 variants were found with two carriers of the p.P96L and one carrier of p.P703S (S1 Table). Among these, a significantly higher level of HbA1c (11.37%, SD: 2.99) compared to non-carriers (7.62%, SD: 1.40), p = 0.004 was also found (Table 3).
In contrast to the overall prevalence of rare variants, one of the identified rare variants (p.I28V) was far more frequent among non-diabetic individuals compared to patients with T2D with ten carriers among non-diabetic individuals and only one carrier with T2D. These non-diabetic carriers had a slightly lower level of fasting plasma glucose compared to non-diabetic individuals not carrying this variant (p = 0.047, S2 Table). The prevalence of this variant was further investigated in a larger publically available dataset where we observed an excess of five carriers among non-diabetic individuals (n = 9,335) in contrast to only one carriers among patients with T2D (n = 9,121) [31] which in a meta-analysis combined with our data disclose a combined OR of 0.20 (IQR: 0.046–0.90), p = 0.04. The publically available online data also identifies a significantly lower level of fasting glucose among carriers of this variant (beta = -0.21, p = 0.02) [31], an analysis which is performed in more than 75,000 individuals.
Discussion
We investigated the effect of rare missense variants within GLIS3 on risk of T2D and on measures of glucose homeostasis. Since amino acids 1–156 are only present in isoform A, 17 of the identified variants are only affecting this longer transcript. However, the longer isoform A is expressed in the pancreas, thus, we assume that all of the identified variants may have potential effect on glucose metabolism.
A slightly elevated prevalence of rare GLIS3 missense variants was observed among patients with T2D compared to non-diabetic individuals. Supporting this observation, levels of HbA1c were also elevated among patients with T2D carrying rare GLIS3 missense variants compared to patients not carrying such variants.
Despite the known function of GLIS3 in beta-cell function, we did not observe any effect of GLIS3 mutations neither on circulating levels of C-peptide nor levels of insulin. Thus, we were unable to determine the cause of elevated HbA1c measuring the long-term glycaemic control. Our observations are well in line with GWAS, finding that common GLIS3 variants are associated with T2D and elevated levels of fasting glucose, however, no other measures of glucose homeostasis have been found to be affected by common GLIS3 variants.
It is puzzling that GLIS3 variants, do not appear to affect insulin secretion in human studies as GLIS3 has been found to be: 1) directly implemented in the transcription of INS [14], 2) been found to be involved in pancreas maturation [7], 3) been found to be involved in beta-cell apoptosis [32] and 4) the physiological consequence of GLIS3 mutations results in diabetes. Yet, direct effects of the mutations on the protein and on beta-cell function are still undisclosed.
One of the identified variants (p.I28V) was less prevalent among patients with T2D and associated with reduced levels of fasting plasma glucose both in our data and in online available data. This suggests a protective effect of this variant on the development of hyperglycaemia. This potential protective variant is only present in Europeans (MAF = 0.1%) and to a smaller extend in the Hispanic population (MAF = 0.02%), indicating that this is mutation is a de novo mutations arisen in Europeans. In a Japanese study, a GLIS3 variant (p.A908V) protecting against T1D was found among approximately 3,000 Japanese patients with T1D and control individuals [33], indicating that variants in GLIS3 may not only have a deleterious effects on diabetes.
Thus, there appear to be large differences in the pathogenicity of variants within GLIS3. Within the current population, we do not encounter variants having a highly detriment effect leading to the development of monogenic diabetes. The combined effect of rare missense variant only results in a modest increased risk of T2D and only a slight increase in level of HbA1c. In contrast, a possible protective variant was encountered. The effect of the identified GLIS3 mutations must be further investigated in functionality studies of each encountered variant.
Due to the central role of GLIS3 in glucose metabolism, GLIS3 may be considered a potential treatment target and two studies have found compounds able to significantly diminish the apoptotic effect of GLIS3. These compounds may be developed further into anti-diabetic treatments in the future [15, 32].
In conclusion, we find that rare missense variants in GLIS3 may increase risk of diabetes and elevate levels of plasma glucose. However, we also find a rare missense variant which likely protects against the development of diabetes. Thus, variants in GLIS3 may explain a fraction of heterogeneity in susceptibility to T2D.
Supporting information
S1 Table. Identified GLIS3 missense variants among non-diabetic individuals, patients with T2D and patients with GADA-positive diabetes.
https://doi.org/10.1371/journal.pone.0220805.s001
(DOCX)
S2 Table. The effect of the p.I28V variants among non-diabetic participants.
https://doi.org/10.1371/journal.pone.0220805.s002
(DOCX)
Acknowledgments
The authors would like to thank the Exome Aggregation Consortium and the groups that provided exome variant data for comparison. A full list of contributing groups can be found at http://exac.broadinstitute.org/about. From Novo Nordisk Foundation Center for Basic Metabolic Research, University of Copenhagen, Denmark, we wish to thank A. Forman, T. H. Lorentzen and G. J. Klavsen for laboratory assistance, P. Sandbeck for data management, and T. F. Toldsted and Camilla Verdich for grant management.
References
- 1. Kim YS, Nakanishi G, Lewandoski M, Jetten AM. GLIS3, a novel member of the GLIS subfamily of Kruppel-like zinc finger proteins with repressor and activation functions. Nucleic Acids Res. 2003;31(19):5513–25. pmid:14500813; PubMed Central PMCID: PMC206473.
- 2. Wen X, Yang Y. Emerging roles of GLIS3 in neonatal diabetes, type 1 and type 2 diabetes. Journal of molecular endocrinology. 2017;58(2):R73–R85. pmid:27899417.
- 3. Senee V, Chelala C, Duchatelet S, Feng D, Blanc H, Cossec JC, et al. Mutations in GLIS3 are responsible for a rare syndrome with neonatal diabetes mellitus and congenital hypothyroidism. Nature genetics. 2006;38(6):682–7. pmid:16715098.
- 4. Jetten AM. GLIS1-3 transcription factors: critical roles in the regulation of multiple physiological processes and diseases. Cellular and molecular life sciences: CMLS. 2018. pmid:29779043.
- 5. Watanabe N, Hiramatsu K, Miyamoto R, Yasuda K, Suzuki N, Oshima N, et al. A murine model of neonatal diabetes mellitus in Glis3-deficient mice. FEBS letters. 2009;583(12):2108–13. pmid:19481545
- 6. Calderari S, Ria M, Gerard C, Nogueira TC, Villate O, Collins SC, et al. Molecular genetics of the transcription factor GLIS3 identifies its dual function in beta cells and neurons. Genomics. 2018;110(2):98–111. pmid:28911974.
- 7. Kang HS, Kim YS, ZeRuth G, Beak JY, Gerrish K, Kilic G, et al. Transcription factor Glis3, a novel critical player in the regulation of pancreatic beta-cell development and insulin gene expression. Molecular and cellular biology. 2009;29(24):6366–79. pmid:19805515; PubMed Central PMCID: PMC2786880.
- 8. Osbak KK, Colclough K, Saint-Martin C, Beer NL, Bellanne-Chantelot C, Ellard S, et al. Update on mutations in glucokinase (GCK), which cause maturity-onset diabetes of the young, permanent neonatal diabetes, and hyperinsulinemic hypoglycemia. Hum Mutat. 2009;30(11):1512–26. pmid:19790256.
- 9. Nicolino M, Claiborn KC, Senee V, Boland A, Stoffers DA, Julier C. A novel hypomorphic PDX1 mutation responsible for permanent neonatal diabetes with subclinical exocrine deficiency. Diabetes. 2010;59(3):733–40. pmid:20009086; PubMed Central PMCID: PMC2828654.
- 10. Dupuis J, Langenberg C, Prokopenko I, Saxena R, Soranzo N, Jackson AU, et al. New genetic loci implicated in fasting glucose homeostasis and their impact on type 2 diabetes risk. Nature genetics. 2010;42(2):105–16. pmid:20081858; PubMed Central PMCID: PMC3018764.
- 11. Barrett JC, Clayton DG, Concannon P, Akolkar B, Cooper JD, Erlich HA, et al. Genome-wide association study and meta-analysis find that over 40 loci affect risk of type 1 diabetes. Nature genetics. 2009;41(6):703–7. pmid:19430480; PubMed Central PMCID: PMC2889014.
- 12. Ding M, Chavarro J, Olsen S, Lin Y, Ley SH, Bao W, et al. Genetic variants of gestational diabetes mellitus: a study of 112 SNPs among 8722 women in two independent populations. Diabetologia. 2018;61(8):1758–68. pmid:29947923.
- 13. Goodarzi MO, Guo X, Cui J, Jones MR, Haritunians T, Xiang AH, et al. Systematic evaluation of validated type 2 diabetes and glycaemic trait loci for association with insulin clearance. Diabetologia. 2013;56(6):1282–90. pmid:23494448; PubMed Central PMCID: PMC3651757.
- 14. ZeRuth GT, Takeda Y, Jetten AM. The Kruppel-like protein Gli-similar 3 (Glis3) functions as a key regulator of insulin transcription. Molecular endocrinology. 2013;27(10):1692–705. pmid:23927931; PubMed Central PMCID: PMC3787130.
- 15. Nogueira TC, Paula FM, Villate O, Colli ML, Moura RF, Cunha DA, et al. GLIS3, a susceptibility gene for type 1 and type 2 diabetes, modulates pancreatic beta cell apoptosis via regulation of a splice variant of the BH3-only protein Bim. PLoS genetics. 2013;9(5):e1003532. pmid:23737756; PubMed Central PMCID: PMC3667755.
- 16. Jorgensen T, Borch-Johnsen K, Thomsen TF, Ibsen H, Glumer C, Pisinger C. A randomized non-pharmacological intervention study for prevention of ischaemic heart disease: baseline results Inter99. European journal of cardiovascular prevention and rehabilitation: official journal of the European Society of Cardiology, Working Groups on Epidemiology & Prevention and Cardiac Rehabilitation and Exercise Physiology. 2003;10(5):377–86. pmid:14663300.
- 17. Glumer C, Jorgensen T, Borch-Johnsen K, Inter s. Prevalences of diabetes and impaired glucose regulation in a Danish population: the Inter99 study. Diabetes care. 2003;26(8):2335–40. pmid:12882858.
- 18.
Organization WH. World Health Organization Diagnosis and Classification of Diabetes Mellitus: Report of a WHO Consultation, in Part 1. World Health Organization, Geneva. 1999.
- 19. Sorensen HT, Friborg S, Rungby J, Christensen JS, Vaag A, Beck-Nielsen H. The Danish national type 2 diabetes cohort—the DD2 study. Clinical epidemiology. 2012;4(Suppl 1):1–5. pmid:23152706; PubMed Central PMCID: PMC3496314.
- 20. Rathmann W, Kowall B, Schulze MB. Development of a type 2 diabetes risk model from a panel of serum biomarkers from the Inter99 cohort: response to Kolberg et Al. Diabetes care. 2010;33(2):e28; author reply e9. pmid:20103554.
- 21. Lau C, Vistisen D, Toft U, Tetens I, Glumer C, Pedersen O, et al. The effects of adding group-based lifestyle counselling to individual counselling on changes in plasma glucose levels in a randomized controlled trial: the Inter99 study. Diabetes Metab. 2011;37(6):546–52. pmid:21900030.
- 22. Thomsen RW, Nielsen JS, Ulrichsen SP, Pedersen L, Hansen AM, Nilsson T. The Danish Centre for Strategic Research in Type 2 Diabetes (DD2) study: Collection of baseline data from the first 580 patients. Clinical epidemiology. 2012;4:43–8. pmid:23071401; PubMed Central PMCID: PMC3470453.
- 23. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985;28(7):412–9. pmid:3899825.
- 24. Gao R, Liu Y, Gjesing AP, Hollensted M, Wan X, He S, et al. Evaluation of a target region capture sequencing platform using monogenic diabetes as a study-model. BMC Genet. 2014;15:13. pmid:24476040; PubMed Central PMCID: PMC3943834.
- 25. Wang K, Li MY, Hakonarson H. ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. Nucleic Acids Res. 2010;38(16). doi: ARTN e164 PubMed PMID: WOS:000281720500004. pmid:20601685
- 26. Machiela MJ, Chanock SJ. LDlink: a web-based application for exploring population-specific haplotype structure and linking correlated alleles of possible functional variants. Bioinformatics. 2015;31(21):3555–7. pmid:26139635; PubMed Central PMCID: PMC4626747.
- 27. Liu DJ, Leal SM. A novel adaptive method for the analysis of next-generation sequencing data to detect complex trait associations with rare variants due to gene main effects and interactions. PLoS genetics. 2010;6(10):e1001156. pmid:20976247; PubMed Central PMCID: PMC2954824.
- 28. Zhan X, Hu Y, Li B, Abecasis GR, Liu DJ. RVTESTS: an efficient and comprehensive tool for rare variant association analysis using sequence data. Bioinformatics. 2016;32(9):1423–6. pmid:27153000; PubMed Central PMCID: PMC4848408.
- 29. Kircher M, Witten DM, Jain P, O'Roak BJ, Cooper GM, Shendure J. A general framework for estimating the relative pathogenicity of human genetic variants. Nature genetics. 2014;46(3):310–5. pmid:24487276; PubMed Central PMCID: PMC3992975.
- 30. Mahajan A, Taliun D, Thurner M, Robertson NR, Torres JM, Rayner NW, et al. Fine-mapping type 2 diabetes loci to single-variant resolution using high-density imputation and islet-specific epigenome maps. Nature genetics. 2018;50(11):1505–13. pmid:30297969; PubMed Central PMCID: PMC6287706.
- 31.
Portal TDK. Type 2 Diabetes Knowledge Portal 2019. Available from: http://www.type2diabetesgenetics.org/.
- 32. Amin S, Cook B, Zhou T, Ghazizadeh Z, Lis R, Zhang T, et al. Discovery of a drug candidate for GLIS3-associated diabetes. Nature communications. 2018;9(1):2681. pmid:29992946; PubMed Central PMCID: PMC6041295.
- 33. Awata T, Yamashita H, Kurihara S, Morita-Ohkubo T, Miyashita Y, Katayama S, et al. A low-frequency GLIS3 variant associated with resistance to Japanese type 1 diabetes. Biochemical and biophysical research communications. 2013;437(4):521–5. pmid:23856252.