Portal de Periódicos da CAPES

T2D Risk Genes: Exome Sequencing Goes Straight to the Source

2019; Cell Press; Volume: 30; Issue: 1 Linguagem: Inglês

10.1016/j.cmet.2019.06.010

1932-7420

Laura Alonso,

Bioinformatics and Genomic Networks

Genome-wide association studies have identified hundreds of genomic variants associated with human T2D risk, but translating such findings to clinically useful information has proved challenging. A new study in Nature (Flannick et al., 2019Flannick J. Mercader J.M. Fuchsberger C. Udler M.S. Mahajan A. Wessel J. Teslovich T.M. Caulkins L. Koesterer R. Barajas-Olmos F. et al.Broad Genomics PlatformDiscovEHR CollaborationCHARGELuCampProDiGYGoT2DESPSIGMA-T2DT2D-GENESAMP-T2D-GENESExome sequencing of 20,791 cases of type 2 diabetes and 24,440 controls.Nature. 2019; 570: 71-76Crossref PubMed Scopus (151) Google Scholar) breaks this gridlock, using direct exome sequencing to identify functional coding variants, providing critical complementary gene-level information. Genome-wide association studies have identified hundreds of genomic variants associated with human T2D risk, but translating such findings to clinically useful information has proved challenging. A new study in Nature (Flannick et al., 2019Flannick J. Mercader J.M. Fuchsberger C. Udler M.S. Mahajan A. Wessel J. Teslovich T.M. Caulkins L. Koesterer R. Barajas-Olmos F. et al.Broad Genomics PlatformDiscovEHR CollaborationCHARGELuCampProDiGYGoT2DESPSIGMA-T2DT2D-GENESAMP-T2D-GENESExome sequencing of 20,791 cases of type 2 diabetes and 24,440 controls.Nature. 2019; 570: 71-76Crossref PubMed Scopus (151) Google Scholar) breaks this gridlock, using direct exome sequencing to identify functional coding variants, providing critical complementary gene-level information. The risk of type 2 diabetes (T2D) is strongly heritable. Understanding the genetic causes of T2D holds great promise for improving human health, with the potential for aspirational advances such as predicting individual disease risk, developing novel gene-directed therapies that address pathogenic biology to treat or prevent disease, and improving outcomes by selecting treatments that address an individual patient's specific biology while minimizing side effects. Enormous resources have been invested worldwide toward these goals. An exciting new study published in Nature (Flannick et al., 2019Flannick J. Mercader J.M. Fuchsberger C. Udler M.S. Mahajan A. Wessel J. Teslovich T.M. Caulkins L. Koesterer R. Barajas-Olmos F. et al.Broad Genomics PlatformDiscovEHR CollaborationCHARGELuCampProDiGYGoT2DESPSIGMA-T2DT2D-GENESAMP-T2D-GENESExome sequencing of 20,791 cases of type 2 diabetes and 24,440 controls.Nature. 2019; 570: 71-76Crossref PubMed Scopus (151) Google Scholar) moves this bar forward. To date, genetic studies have provided a rich landscape of information on T2D pathogenesis (Fuchsberger et al., 2016Fuchsberger C. Flannick J. Teslovich T.M. Mahajan A. Agarwala V. Gaulton K.J. Ma C. Fontanillas P. Moutsianas L. McCarthy D.J. et al.The genetic architecture of type 2 diabetes.Nature. 2016; 536: 41-47Crossref PubMed Scopus (704) Google Scholar) but frustratingly few clinically useful advances. A growing number of monogenic syndromes have been identified: gene mutations with high effect size such that the diabetes phenotype segregates in Mendelian fashion. However, even considering that monogenic diabetes is frequently misdiagnosed as T2D, these high-impact rare variants account for only a small fraction of adult-onset diabetes. Recognizing that T2D is a complex multigenic trait, excitement shifted toward genome-wide association studies (GWAS), with the expectation that identifying common variants with small effect size would solve the genetic basis of T2D. GWAS have been extremely successful, identifying hundreds of variants associated with T2D risk. However, most polymorphisms that show the strongest association with T2D are in noncoding regions, leaving the precise genetic mechanisms driving metabolic risk uncertain. A critical gap has been direct gene-level information at a sufficient depth to test for rare variants that may contribute heritability not localized by common variant analysis (Cirulli and Goldstein, 2010Cirulli E.T. Goldstein D.B. Uncovering the roles of rare variants in common disease through whole-genome sequencing.Nat. Rev. Genet. 2010; 11: 415-425Crossref PubMed Scopus (874) Google Scholar). The new study by Flannick et al. fills this gap as they exome sequenced a very large population of T2D cases and controls (n = 20,791 and 24,440, respectively) (Figure 1). Samples were drawn from six consortia: T2D-GENES, SIGMA T2D, Go T2D, ESP, LuCamp, and ProDiGY (Fu et al., 2013Fu W. O'Connor T.D. Jun G. Kang H.M. Abecasis G. Leal S.M. Gabriel S. Rieder M.J. Altshuler D. Shendure J. et al.NHLBI Exome Sequencing ProjectAnalysis of 6,515 exomes reveals the recent origin of most human protein-coding variants.Nature. 2013; 493: 216-220Crossref PubMed Scopus (679) Google Scholar, Fuchsberger et al., 2016Fuchsberger C. Flannick J. Teslovich T.M. Mahajan A. Agarwala V. Gaulton K.J. Ma C. Fontanillas P. Moutsianas L. McCarthy D.J. et al.The genetic architecture of type 2 diabetes.Nature. 2016; 536: 41-47Crossref PubMed Scopus (704) Google Scholar, Lohmueller et al., 2013Lohmueller K.E. Sparsø T. Li Q. Andersson E. Korneliussen T. Albrechtsen A. Banasik K. Grarup N. Hallgrimsdottir I. Kiil K. et al.Whole-exome sequencing of 2,000 Danish individuals and the role of rare coding variants in type 2 diabetes.Am. J. Hum. Genet. 2013; 93: 1072-1086Abstract Full Text Full Text PDF PubMed Scopus (106) Google Scholar, Williams et al., 2014Williams A.L. Jacobs S.B. Moreno-Macías H. Huerta-Chagoya A. Churchhouse C. Márquez-Luna C. García-Ortíz H. Gómez-Vázquez M.J. Burtt N.P. Aguilar-Salinas C.A. et al.SIGMA Type 2 Diabetes ConsortiumSequence variants in SLC16A11 are a common risk factor for type 2 diabetes in Mexico.Nature. 2014; 506: 97-101Crossref PubMed Scopus (334) Google Scholar). The resulting dataset provides an unprecedented look at T2D-associated coding variants, including 2.26 million nonsynonymous variants, of which 871,00 are insertions or deletions. The team found that 93.5% of the identified variants were rare, with a minor allele frequency <0.5%, and thus unlikely to be identifiable by GWAS. Fifteen variants achieved exome-wide significance, a high bar for rare variants (Fuchsberger et al., 2016Fuchsberger C. Flannick J. Teslovich T.M. Mahajan A. Agarwala V. Gaulton K.J. Ma C. Fontanillas P. Moutsianas L. McCarthy D.J. et al.The genetic architecture of type 2 diabetes.Nature. 2016; 536: 41-47Crossref PubMed Scopus (704) Google Scholar). Five of these were synonymous, which could impact RNA biology such as transcription rate, splicing, translation, or co-translational folding. Of the ten nonsynonymous variants, eight were previously known, one was a novel variant in a known gene (MCR4), and one was unreplicated. Gene-level analysis at exome-wide significance confirmed three genes already known to be associated with T2D risk, SLC30A8, MCR4, and PAM. Meta-analysis combining the new dataset with prior datasets revealed an additional novel gene, UBE2NL. The real power and impact of this study lie in the pool of genes that did not achieve exome-wide significance. Genes not meeting exome-wide significance are nonetheless informative, containing many genes relevant to T2D pathogenesis such as T2D drug targets, MODY genes, and genes linked to metabolic phenotypes in mice. Tremendous value comes from this new gene-level information. Identifying multiple coding variants within a gene generates an allelic series, providing actionable functional coding mutations to study biology and pathogenesis. Sophisticated bioinformatic tools can now predict mutation "deleteriousness"; i.e., an estimate of how severely a variant may disrupt function. Prediction efficacy was demonstrated by T2D drug target genes: for all four targets for which the drug increases function (i.e., GLP1R, IGF1R, PPARG, and INSR), loss-of-function alleles were associated with an increased risk of T2D. Conversely, for targets in which a drug decreases function (i.e., SLC5A2, DPP4, and ABCC8), loss-of-function alleles were associated with protection against T2D. The one exception was KCNJ11, in which variants cause either neonatal hyperinsulinism or diabetes, confounding the function prediction. Applying this directionality of variant impact to known genes is critical for drug development. As an example, it has been controversial whether activating or inhibiting SLC30A8, which encodes ZnT8, would be beneficial. In this study, loss-of-function variants of SLC30A8 reduced risk of T2D, suggesting that ZnT8 inhibition may be useful to treat T2D. Perhaps the most important result from this work is the use of coding sequence variants at GWAS loci to illuminate which locus gene influences disease risk. Since exome sequencing is a complementary technique, identifying rare variants mostly not imputable from GWAS, this is powerful new information. To pinpoint causative genes, a curated list of 94 GWAS loci (Flannick and Florez, 2016Flannick J. Florez J.C. Type 2 diabetes: genetic data sharing to advance complex disease research.Nat. Rev. Genet. 2016; 17: 535-549Crossref PubMed Scopus (98) Google Scholar) was examined. Excitingly, of 595 genes within 250 kb of any T2D GWAS index variant, 40 genes were statistically associated with T2D in the exome data. Only 14 of the 40 were the gene the locus was "named" for, reinforcing the hazard of assumptions regarding which nearby gene mediates T2D-variant impact. In one recent example, T2D-associated variants at the CDKN2A/B locus, which were assumed to regulate pancreatic islet CDKN2A expression, may instead impact a long non-coding RNA at the locus (Kong et al., 2018Kong Y. Sharma R.B. Ly S. Stamateris R.E. Jesdale W.M. Alonso L.C. CDKN2A/B T2D genome-wide association study risk SNPs impact locus gene expression and proliferation in human islets.Diabetes. 2018; 67: 872-884Crossref PubMed Scopus (36) Google Scholar). As many of the T2D risk variants are expected to impact islet biology, it will be of great interest to compare these exome sequencing predictions with emerging high sample number islet eQTL studies (Khamis et al., 2019Khamis A. Canouil M. Siddiq A. Crouch H. Falchi M. Bulow M.V. Ehehalt F. Marselli L. Distler M. Richter D. et al.Laser capture microdissection of human pancreatic islets reveals novel eQTLs associated with type 2 diabetes.Mol. Metab. 2019; 24: 98-107Crossref PubMed Scopus (18) Google Scholar, Viñuela et al., 2019Viñuela A. Varshney A. van de Bunt M. Prasad R.B. Asplund O. Bennett A. Boehnke M. Brown A. Erdos M.R. Fadista J. et al.Influence of genetic variants on gene expression in human pancreatic islets – implications for type 2 diabetes.bioRxiv. 2019; https://doi.org/10.1101/655670Crossref Google Scholar). In conclusion, this study represents a breakthrough in T2D genetics. Due to the large sample size, it provides remarkable depth of gene-level insight. The wealth of new information will support bioinformatic approaches to determine whether implicated genes converge on key targetable pathways and to prioritize genes for future studies. Importantly, the data are freely available to the public through the user-friendly AMP T2D Knowledge Portal. Many future questions remain. This work does not support the persistent hypothesis that significant missing heritability of T2D is due to unknown rare variants with high effect size; rare variants were estimated to comprise at most 25% of the risk imparted by common variants. As such, the question of missing heritability remains to be solved. Although this work significantly moves the bar on predicting which genes contribute to T2D risk, the hard work of deciphering the underlying cell biology and developing therapeutic approaches to modulate this biology to prevent or treat diabetes still lies ahead. Nonetheless, this work represents a remarkable step forward toward the goal of personalized medicine for T2D. This work was supported by NIH/NIDDK (R01DK114686 and R01DK113300; L.C.A.), the American Diabetes Association grant #1-18-IBS-233 (L.C.A.) in collaboration with the Order of the Amaranth, and the George F. and Sybil H. Fuller Foundation. The funding sources had no involvement in the analysis of the literature, the writing of this review, or the decision to submit the article for publication.