FTO Obesity Variant Circuitry and Adipocyte Browning in Humans Melina Claussnitzer, Ph.D., Simon N
Total Page:16
File Type:pdf, Size:1020Kb
The new england journal of medicine established in 1812 September 3, 2015 vol. 373 no. 10 FTO Obesity Variant Circuitry and Adipocyte Browning in Humans Melina Claussnitzer, Ph.D., Simon N. Dankel, Ph.D., Kyoung‑Han Kim, Ph.D., Gerald Quon, Ph.D., Wouter Meuleman, Ph.D., Christine Haugen, M.Sc., Viktoria Glunk, M.Sc., Isabel S. Sousa, M.Sc., Jacqueline L. Beaudry, Ph.D., Vijitha Puviindran, B.Sc., Nezar A. Abdennur, M.Sc., Jannel Liu, B.Sc., Per‑Arne Svensson, Ph.D., Yi‑Hsiang Hsu, Ph.D., Daniel J. Drucker, M.D., Gunnar Mellgren, M.D., Ph.D., Chi‑Chung Hui, Ph.D., Hans Hauner, M.D., and Manolis Kellis, Ph.D. abstract BACKGROUND Genomewide association studies can be used to identify disease-relevant genomic The authors’ affiliations are listed in the regions, but interpretation of the data is challenging. The FTO region harbors the Appendix. Address reprint requests to Dr. Claussnitzer at the Gerontology Division, strongest genetic association with obesity, yet the mechanistic basis of this asso- Beth Israel Deaconess Medical Center ciation remains elusive. and Hebrew SeniorLife, Harvard Medical School, 1200 Centre St., Boston, MA METHODS 02215, or at melina@ broadinstitute . org; We examined epigenomic data, allelic activity, motif conservation, regulator ex- or to Dr. Kellis at the Computer Science and Artificial Intelligence Laboratory, MIT, pression, and gene coexpression patterns, with the aim of dissecting the regula- 32 Vassar St., Cambridge, MA 02139, or tory circuitry and mechanistic basis of the association between the FTO region and at manoli@ mit . edu. obesity. We validated our predictions with the use of directed perturbations in Drs. Dankel and Kim contributed equally samples from patients and from mice and with endogenous CRISPR–Cas9 genome to this article. editing in samples from patients. This article was published on August 19, RESULTS 2015, at NEJM.org. Our data indicate that the FTO allele associated with obesity represses mitochon- N Engl J Med 2015;373:895-907. drial thermogenesis in adipocyte precursor cells in a tissue-autonomous manner. DOI: 10.1056/NEJMoa1502214 Copyright © 2015 Massachusetts Medical Society. The rs1421085 T-to-C single-nucleotide variant disrupts a conserved motif for the ARID5B repressor, which leads to derepression of a potent preadipocyte enhancer and a doubling of IRX3 and IRX5 expression during early adipocyte differentiation. This results in a cell-autonomous developmental shift from energy-dissipating beige (brite) adipocytes to energy-storing white adipocytes, with a reduction in mitochondrial thermogenesis by a factor of 5, as well as an increase in lipid stor- age. Inhibition of Irx3 in adipose tissue in mice reduced body weight and increased energy dissipation without a change in physical activity or appetite. Knockdown of IRX3 or IRX5 in primary adipocytes from participants with the risk allele restored thermogenesis, increasing it by a factor of 7, and overexpression of these genes had the opposite effect in adipocytes from nonrisk-allele carriers. Repair of the ARID5B motif by CRISPR–Cas9 editing of rs1421085 in primary adipocytes from a patient with the risk allele restored IRX3 and IRX5 repression, activated browning expression programs, and restored thermogenesis, increasing it by a factor of 7. CONCLUSIONS Our results point to a pathway for adipocyte thermogenesis regulation involving ARID5B, rs1421085, IRX3, and IRX5, which, when manipulated, had pronounced pro-obesity and anti-obesity effects. (Funded by the German Research Center for Environmental Health and others.) n engl j med 373;10 nejm.org September 3, 2015 895 The New England Journal of Medicine Downloaded from nejm.org at MIT LIBRARIES on September 14, 2015. For personal use only. No other uses without permission. Copyright © 2015 Massachusetts Medical Society. All rights reserved. The new england journal of medicine besity affects more than 500 mil- Methods lion people worldwide and contributes to type 2 diabetes, cardiovascular disorders, Participants O 1 and cancer. Obesity is the result of a positive Primary human adipose–derived progenitor cell energy balance, whereby energy intake exceeds cultures were obtained from the subcutaneous expenditure, resulting in the storage of energy, adipose tissue of 100 healthy Europeans who primarily as lipids in white adipocytes. Energy were 20 to 50 years of age and had a BMI (the balance is modulated by food consumption and weight in kilograms divided by the square of the physical activity, as well as by the dissipation of height in meters) of 20 to 24, which was in the energy as heat through constitutive thermogen- normal range. The sample included 52 partici- esis in mitochondria-rich brown adipocytes in pants (hereafter referred to as risk-allele carriers) brown fat and through inducible thermogenesis who were homozygous for the risk allele for the in beige adipocytes in white fat.2-6 Thermogene- tag variant rs9930506 (which has been reported sis is triggered by mechanisms within the cells in genomewide association studies), as well as for themselves or by the sympathetic nervous system the associated variants rs1421085 and rs1558902; (e.g., through β-adrenergic receptor agonists), in together, these variants make up part of the risk response to exercise, diet, or exposure to cold. haplotype, which in its entirety includes 89 vari- Regulators of mitochondrial thermogenesis (in- ants in introns 1 and 2 of FTO that are common cluding IRF4, PGC1α, PRDM16, and TBX15) in European populations. The sample also includ- control the expression of the gene encoding ed 48 participants who were homozygous for the uncoupling protein 1 (UCP1),4,7-9 which depolar- nonrisk allele for all three of these variants izes the inner mitochondrial membrane, causing (hereafter referred to as nonrisk-allele carriers). proton transfer and heat dissipation. Primary cell cultures were used for the preparation Body-mass index (BMI) has a strong genetic of mitochondrial and nuclear messenger RNA, component (40 to 80% heritability) involving quantitative polymerase-chain-reaction (PCR) several genes that have expression in the hypo- gene-expression analysis, assays of mitochon- thalamus and fulfill roles in appetite regula- drial function and thermogenesis, determination tion.10,11 The strongest genomewide association of lipolysis rates, small interfering RNA–mediated signal lies in introns 1 and 2 of the gene FTO12,13; knockdown, doxycycline-mediated overexpression, this region contains 89 common variants and CRISPR–Cas9 genome editing. In addition, (Fig. 1A) that are in high linkage disequilibrium whole adipose tissue and adipose-derived progeni- in Europeans (r2≥0.8)12,13 across approximately tor cells were obtained and RNA was isolated 47,000 nucleotides, which makes the identifica- from a second European cohort of nongenotyped tion of the likely causal variant challenging. The participants, including 12 severely obese patients associated region lacks protein-altering variants, (BMI, 35 to 52) who were undergoing bariatric which has prompted numerous gene-regulatory surgery and 22 healthy nonobese participants studies17-21; these studies have predicted diverse (BMI, 18 to 28) who were undergoing elective and conflicting target genes and tissues, includ- surgery. We received approval from the local ethics ing FTO itself in a whole-body knockout,17 IRX3 committees in Germany, Norway, and Sweden. in pancreas20 or brain,19 RBL2 in lymphocytes,21 All participants gave written informed consent. and RPGRIP1L in brain.18 However, the identifica- tion of a mechanistic basis for the association Study Design between the FTO locus and obesity in humans We used techniques of epigenomics, compara- has been elusive, the relevant cell types and tar- tive genomics, human genetics, genome editing, get genes remain unresolved, and the causal and directed perturbations in samples from pa- variant remains uncharacterized. In this study, tients and from mice to dissect the regulatory we sought to identify a causal variant with regu- circuitry and mechanistic basis of the FTO obe- latory roles, its upstream regulator, and its sity-associated locus. Our methods are described downstream target gene in order to provide a in detail in the Methods section in the Supple- candidate mechanistic basis for the association mentary Appendix, available with the full text of between FTO and obesity. this article at NEJM.org. 896 n engl j med 373;10 nejm.org September 3, 2015 The New England Journal of Medicine Downloaded from nejm.org at MIT LIBRARIES on September 14, 2015. For personal use only. No other uses without permission. Copyright © 2015 Massachusetts Medical Society. All rights reserved. FTO Obesity Variant and Adipocyte Browning On the basis of epigenomic annotations A across 127 human cell types, we predicted the SNVs 60 rs1421085 rs1558902 rs9930506 cell type in which the genetic variant was likely r2 1.0 80 to act, and we validated the prediction with the 50 0.8 0.6 Recombi- use of haplotype-specific enhancer assays. We 40 0.4 nation 60 0.2 rate analyzed long-range chromatin interactions in LD in CEU P value) 30 0.0 the region surrounding FTO to define potential 10 40 20 target genes, and to validate genetic targets, we (–log 20 conducted an expression quantitative-trait-locus 10 Genetic Association with BMI (eQTL) analysis in primary human adipocytes 0 0 Recombination Rate (cM/Mb) Chromosome 53.77 53.80 53.85 53.87 Mb from risk-allele carriers and nonrisk-allele carri- 16 ers. We predicted the cellular processes affected FTO gene by the obesity-associated variants on the basis of Exon 2 Exon 3 correlated expression with the target genes B across participants, and we validated their genetic Tissue or Cell Type control with the use of a trans-eQTL analysis ESC iPSC of energy-balance genes (i.e., an eQTL analysis of ES-derived energy-balance genes at large genomic distances Blood and T cell from the FTO locus) in adipocytes, as well as by HSC and B cell measuring cellular phenotypes in risk-allele car- Mesenchymal Epithelial riers and nonrisk-allele carriers.