Correlation Between the Gly482Ser Polymorphism and Increased Type 2 Diabetes Susceptibility

Introduction

1 – Type 2 Diabetes

1.1 – Obesity and Type 2 Diabetes

Over the past 24 years, the number of people affected by Type 2 diabetes (T2D) has nearly quadrupled, growing from 108 million in 1980 to 422 million in 2014. Not only has the global prevalence of diabetes has nearly doubled during this time, but now even younger generations are becoming affected by the disease [1]. It has been suggested that the increase in T2D among children and adolescents can likely be attributed to higher degrees of obesity, rather than overall diabetes prevalence alone [2].

This climb in obesity is reflective of socioeconomic changes that encourage increased caloric intake and decreased energy expenditure (i.e. lack of exercise).  Poor lifestyle choices often manifest themselves in the form of hyperglycemia; typical fasting blood glucose should range from 3.9-5 mmol/l whereas those with T2D are greater than 7mmol/l. Hyperglycemia can be the result of: insulin resistance of peripheral tissues, increased gluconeogenesis by the liver and/or insufficient insulin secretion by pancreatic beta cells. Obesity is typically associated with chronic, low grade systemic inflammation which is the result of elevated levels pro-inflammatory cytokines originating from adipose and/or liver tissues [3]. Upon hyperphagia, inflammatory cytokine and circulating free fatty acid (FFA) levels rise and begin to accumulate in metabolically active tissues, including muscle and liver [4]. This results in impaired insulin signaling and mitochondrial function in these tissues which, in turn, impairs glucose uptake and largely contributes to systemic insulin resistance[4, 5]. Increased hepatic glucose production (HGP) is one consequence of increased intracellular lipids accumulating in the liver. Unwarranted HGP further impairs insulin signaling as well as lipid metabolism, driving T2D development [6].  In attempt to compensate for elevated blood glucose levels, beta cells will upregulate insulin secretion in attempt to maintain normoglycemia [7]. Unfortunately, this high level of insulin secretion is only sustainable for so long before beta cells begin to fail, allowing T2D to progress. Currently there remains a lack of consensus whether insufficient insulin secretion is the result of beta cell dysfunction, decreased beta cell mass, or both.

1.2 – The Genetic Basis of Diabetes

Metabolic syndrome is a broad spectrum of diseases that are typically characterized by several risk factors such as: hypertension, abdominal obesity, hyperglycemia, abnormal triglyceride and cholesterol levels and low high-density lipoprotein (HDL) levels. Presence of one, or more of these risk factors often gives rise to increased susceptibility disease including, but not limited to, diabetes, fatty liver disease and/or cardiovascular disease. Although environmental factors such as lifestyle and aging contribute significantly to disease susceptibility and development of the metabolic syndrome, the influence of genetics cannot be ignored. In type 2 diabetes (T2D), there are multiple lines of evidence supporting that genetics play both important roles in the pathogenesis of the disease as well as individual response to treatment. For example, those with two parents with T2D have a 70% chance of developing diabetes, with twin studies identifying >70% concordance amongst monozygotic twins [8] compared to 10-20% between dizygotic twins [9]. While these strong associations provide convincing evidence for a strong genetic basis, efforts to identify specific risk genes have proven challenging. There also remains a lack of agreement and certainty pertaining to the majority of genetic risk markers for diabetes and to what degree environment affects the epigenetic regulation of said markers  [10]. Although less than 15% of heritability has been explained by findings of Genome Wide Association Studies (GWAS) [11], these studies have significantly advanced our knowledge of genetic factors and molecular pathways influencing diabetes risk. Single nucleotide polymorphisms (SNPs) associated with increased disease risk often lie in uncharacterized genomic regions and/or have unknown/unpredictable effects on gene function, limiting usefulness of this information for prevention or therapeutic targeting. Furthermore, each SNP in isolation has arguably minor effects on risk and “genetic risk signatures” remain difficult to prove experimentally and do not seem to increase our ability to predict risk better than traditional methods [12]. To date, there are at least 80 SNPs identified shown to have significant links to diabetes risk [12, 13]. Many of these genes are associated with pancreatic β-cell biology or insulin sensitivity. Unfortunately, functional consequences of identified SNPs have yet to be elucidated, and often even when the predicted protein target is clear, the biology of many candidate proteins remains unknown. Lastly, although it is widely accepted that diabetes risk is influenced by ethnicity, most GWAS studies for diabetes involve Caucasian populations of European decent or Asian populations of Japanese or Chinese decent [12]. Recent efforts in multi-ethnic analyses have confirmed many previously identified genes and identified new candidates. However, this approach to identify common underlying genes of worldwide diabetes risk often direct us to ignore genes with significant effects only in certain populations. Furthermore, complexity introduced by geographical, cultural, environmental and even sociological (i.e. gender) influences undoubtedly have significant effects on outcomes of these large-scale epidemiological analyses. In the age of personalized medicine, understanding biological roles of gene variants, and how they are influenced by environment, could have more value over standardized lists of “risk-associated” and “protective” genes.

2 – PGC-1α: A versatile transcriptional coactivator of metabolism

2.1 – PGC-1α discovery and characterization

In 1998, an autosomal genomic scan of 363 non-diabetic Pima Indians designed to reveal genetic loci linked to pre-diabetic traits identified gene locus 4p15-q12 to be associated with increased fasting insulin [14], an established hallmark of T2D. This domain of human chromosome 4 contains approximately 44 known genes, including clock circadian regulator (CLOCK), huntingtin (HTT), Wolfram syndrome 1 (WFS1), and peroxisome proliferator-activated receptor gamma, coactivator 1 alpha (PPARGC1A). PGC-1α was originally identified by yeast two-hybrid screen in brown adipose where it was found to interact with peroxisome proliferator-activated receptor gamma (PPAR), a powerful transcription factor that is heavily involved in metabolism. In functional studies, it was classified as a powerful transcriptional coactivator of gene pathways controlling mitochondrial biogenesis and oxidative capacity [15, 16]. Following initial characterization, the importance of PGC-1α in controlling multiple aspects of mitochondrial structure and biology has become well established, leading to the protein often being referred to as a master regulator of mitochondrial function [17].

PGC-1α is a large, uniquely structured protein with activation and transcription factor binding domains largely concentrated in its N-terminal regions [18], as well as RNA arginine-serine (RS) rich domains that can interact with RNA processing factors [19]. The coactivator is expressed in many metabolically active tissues including muscle, neurons, adipose tissue, liver, and heart [20]. In addition to PPAR, PGC-1α enhances the activity of other PPARs [21, 22] and associates with transcription factors required for oxidative metabolism, mitochondrial biogenesis, reactive oxygen species (ROS) homeostasis, insulin and glucagon signalling [20]. PGC-1α is involved in glucose uptake by skeletal muscle through regulation of Glut4 expression and translocation and impacts metabolism of glucose and lipids in liver by stimulating transcription of several enzymes required for hepatic gluconeogenesis and fatty acid oxidation [23-25]. In mouse muscle, PGC-1α is required to maintain expression of mitochondrial genes, oxidative phosphorylation, and exercise capacity [26, 27], while in the beta cell and adipose, PGC-1α is dispensable for maintenance of mitochondria, but significantly impacts insulin secretion and insulin sensitivity, respectively [28-30]. These examples highlight the highly tissue-specific nature of PGC-1α and its considerable involvement in whole-body metabolism.

2.2 – The Gly482Ser polymorphism

Since the discovery of PGC-1α’s link to both mitochondrial function and diabetes risk, interest in this protein and the effects of genetic variation on metabolic health have grown. Deeper analysis of select genes centered on energy metabolism revealed that a common polymorphism in the coding region of PPARGC1A (rs8192678, 1444 G > A, Gly482Ser) is associated with acute measures of insulin secretion in Pima Indians [31].  This SNP represents a single amino acid difference in the mRNA sequence, encoding either a serine or a glycine residue at position 482 of the protein product, PGC-1α. A subsequent study investigating the link between seven PGC-1α variants and T2D in Danish Caucasians revealed that only Gly482Ser polymorphism had a significant association with T2D. Subjects harboring the serine allele had a significantly higher incidence of T2D compared to those with the glycine allele, corresponding to a 1.34 relative risk of disease [32].  Subsequent analyses did not reproduce the association of the Gly482Ser SNP with T2D in Pima Indians; however, non-diabetic Pima Indians carrying one or two serine alleles had higher insulin secretion 3- and 30-minutes following glucose infusion with equal blood glucose levels [31], suggestive of increased insulin resistance. They also had lower FFA, smaller adipocyte size, and higher rates of lipid oxidation even in the presence of clamped insulin [31], further suggesting that the serine allele reduces insulin effectiveness, possibly in a dominant manner. Consistent with effects on insulin sensitivity, obese Caucasians of Italian decent with a serine allele also had decreased insulin sensitivity (measured by HOMA-IR and increased fasting insulin) independent of age, sex, BMI, HDL-cholesterol or triglycerides, regardless of hetero- or homozygosity at the locus [33]. Interestingly, incidence of the serine-containing allele alone is not consistently elevated in subjects with diabetes; however, haplotypes containing the Gly482Ser and other PGC-1α polymorphisms (e.g. the synonymous Thr394Thr PGC-1α variant) are significantly associated with T2D [34, 35] and impaired oral glucose tolerance in offspring of type 2 diabetic subjects [36], suggesting an additive or modifying role for the Gly482Ser polymorphism in regulating insulin sensitivity.In a study of 3244 participants aged 20-59 (Netherlands), carriers of the serine allele with a BMI less than 25 kg/m² have lower non-fasting blood glucose [37], while this relationship is inversed when BMI is >25, with obese carriers of the serine allele having significantly higher fasting glucose. While these observations suggest that the Gly482Ser polymorphism has a more modifying than causative role in diabetes pathogenesis, the genotype appears to have significant implications on risk assessment, disease severity, and treatment strategy in carriers.

2.3 – The impact of ethnicity on the Gly482Ser polymorphism and diabetes

The PGC-1α protein, and the domain containing the GlySer482 SNP in particular, are highly conserved across species [38]. Interestingly, while the serine containing allele (482Ser) is arguably the “minor allele” in many human populations [38], most other vertebrates only have a serine in this position. Searching available databases, we and others have found only humans (Homo sapiens), chickens (Gallus gallus) and wild turkeys (Meleagris gallopavo), to have a glycine containing allele; however, many species of birds (ie. Columba livia, or rock pigeon) and the fruit bat (Rousettus aegyptiacus) alternatively have an arginine (R) residue. These limited data suggest the glycine-containing variant may have appeared later in evolution and became enriched in humans due to selective pressure, or it was selected against in other species. A higher than expected prevalence of Gly/Gly might suggest that this allele provides some selective advantage to certain groups. In support of this theory, the Gly482 variant is found over represented in elite endurance athletes of Russian [39], German [40], Turkish [41] and Israeli [42] decent. However, if the 482Ser variant is truly a strong risk allele for metabolic disease in humans, why does this variant remain so prevalent in humans and why is it not selected against in other species?

A plausible explanation might be that the onset of metabolic disease often occurs years after reproductive age, providing no resistance to the allele being passed onto future progeny. However, if the SNP has no influence prior to disease onset, one would expect classical mendelian ratios of SNP prevalence, which is not the case in most populations tested.  The majority of published clinical studies on this SNP were performed on subjects of European descent, where the prevalence of the Gly/Gly phenotype averages around 50%, with Ser/Ser homozygotes detected at rates of 10-15% [43]. Interestingly, the prevalence of each variant seems to vary greatly depending on geographical location [44]. Sampling data suggest that Ser482 “risk” allele prevalence can reach >80% in some Polynesian island nations of the South Pacific, while many areas within Africa, Papua New Guinea and Indonesia have frequencies less than 3% (based on data from the Human Genome Diversity Cell Line Panel) [44]. These large variant frequency differences between populations led to a theory that the 482Ser allele may be considered a “thrifty gene”, providing advantages to species that depend on fat storage capacity for survival (e.g. during periods of famine or hibernation in rodents) [44]. However, in times of relative food abundance, having the serine variant may contribute to obesity and increased metabolic disease prevalence. Recently, this hypothesis was challenged, as statistical testing did not find evidence for departure from natural evolution for this locus in a range of Polynesian, Asian, European or African populations [45]. Additional evidence against the thrifty gene hypothesis comes from the fact that relative risk of T2D associated with the 482Ser allele is not the same across populations of humans that are now exposed to relatively similar diet and lifestyles. In Caucasian populations, although significant, the risk of T2D is only modestly increased (e.g. odds ratio ranging between 1.1 – 1.8), while odds ratios for T2D risk increase greatly in 482Ser carrying subjects of Northern Indian (OR 2.04-3.19), Iranian (OR 9.0), Chinese (OR 1.64-1.85) and Tunisian decent (OR 1.17–2.98) [46-50]. Thus, it is plausible that differences in disease risk between ethnicities linked to this polymorphism are due to additive or synergistic effects with other genetic modifiers or environmental factors specific to geographical region. This is supported by evidence that haplotypes containing both the Gly482Ser of PGC-1α and the Pro12Ala of PPARy have a greater risk of diabetes [51]

2.4 – Effects of the Gly482Ser polymorphism on metabolism

Adiposity and BMI

While investigating effects of metformin and lifestyle interventions (weight loss and increased physical activity) on T2D development in patients with high fasting glucose and impaired glucose tolerance, researchers found that the Gly482Ser PGC-1α polymorphism independently associated with increased adiposity [52]. 482Ser allele carriers have elevated baseline HOMA-IR and subcutaneous adiposity, but the association is not statistically significant following adjustment for BMI [52]. The serine allele is also associated with elevated body fat mass in Korean children of normal body weight [53] and in overweight, non-diabetic Chinese adults [54]. Moreover, an increase in total body fat mass, hip circumference, BMI, and body fat ratio is observed in Ser/Ser homozygotes in a Mexican-Mestizo population [55]. Furthermore, excessive weight gain is associated with the Gly482Ser polymorphism in males with type 1 diabetes receiving intensive diabetes therapy [56]. On the contrary, no association between the SNP and obesity was found in studies of Asian Indians [57] or Portuguese children [58]. In the Pima Indian population, fasting nonesterified fatty acids (NEFA) level is lower for 482Ser carriers [31], suggesting that adipose tissue lipolysis might be impaired. Carriers of the serine allele also have reduced clearance of NEFA following an oral glucose tolerance test [59]. These elevated NEFA levels may inhibit insulin signaling and glucose disposal to increase T2D susceptibility.

Taken together, these data suggest that this polymorphism negatively impacts adipose tissue biology or function in certain populations; and specifically, the serine-containing allele confers higher risk of obesity. Thus, links to increased T2D susceptibility may simply be due to effects on adiposity. This may also explain why carriers of the 482Ser allele seem to benefit more from interventions aimed at weight loss, including caloric restriction [60], bariatric surgery [61], and acarbose treatment [51], than Gly482 allele carriers.

Insulin Secretion

Beta cell function (measured by HOMA-%B) is increased in serine allele carriers [31, 33], although it cannot be determined whether this is a primary effect of the polymorphism to improve beta cell function or increased insulin secretion secondary to insulin resistance. Pancreatic islets from non-diabetic donors carrying the 482Ser allele (Ser/Ser or Gly/Ser) have reduced capacity to secrete insulin in response to glucose compared to those homozygous with the Gly482 allele [28].

High levels of PGC-1α are detected in islets of Zucker diabetic fatty (ZDF) rats and ob/ob mice, two animal models of type 2 diabetes [62].  However, overexpression of PGC-1α in mouse islets within physiological levels in vivo does not impair β-cell function in adult mice [63] and β-cell specific knockout of both PGC-1α and PGC-1β diminishes glucose stimulated insulin secretion [29]. These data in mice are consistent with data in humans, suggesting that PGC-1α gene expression in the pancreatic islets directly correlates with insulin secretion [64]. Thus, the Gly482Ser polymorphism may have a direct impact on beta cell function; whose chronic deregulation can in itself lead to augmentation of adiposity and changes in insulin sensitivity.

Cardiac and Liver Metabolism:

In addition to T2D, the Gly482Ser polymorphism is associated with both cardiovascular disease and non-alcoholic fatty liver disease (NAFLD). Studies investigating associations between the polymorphism and cardiovascular disease reveal that Mongolian carriers of the 482Ser allele have higher incidences of hypertension [65] and hypertensive patients homozygous for the serine allele have greater left ventricle hypertrophy and lower diastolic function [66]. Moreover, there is a slightly increased risk of hypertrophic cardiomyopathy (OR 1.11 – 2.11) in carriers of the Ser482, which leads to thickening of the ventricle walls, usually resulting in hypertension [67].  There is also a higher incidence of the Ser482 allele among coronary artery disease patients of Chinese descent (OR 1.68) [68].

Fatty liver disease is a significant risk factor for diabetes, increasing chances of developing T2D by two-fold [69]. With recent interest in metabolic liver disease as a predisposing factor to diabetes, more research is focused on identifying NAFLD risk genes. The PGC-1α Gly482Ser polymorphism increases risk of NAFLD in obese Taiwanese children (OR 1.74), compared to controls homozygous for the glycine variant [70]. In obese Taiwanese adults, having at least one serine-containing allele is an independent risk factor for non-alcoholic hepatosteatosis (NASH), is associated with higher steatosis and ballooning degeneration of liver cells, and carries an additive effect of the Gly482Ser with the PNPLA3 rs738409 SNP on NASH incidence [71]. In contrast, newly diagnosed German subjects with type 1 diabetes homozygous for the Gly482 allele have lower hepatic adenosine triphosphate, suggesting impaired mitochondrial metabolism [72]. These opposing findings provide further evidence that environmental and/or lifestyle factors influence SNP impact on metabolic health and that tissue-specific effects that may be additive.

3 –  The effect of polymorphism on PGC-1α structure and function

3.1 – PGC-1α Stability

We believe that the Gly482Ser polymorphism increases T2D (and other metabolic disease) susceptibility in some populations due to impacts on protein stability. Because lower PGC-1α expression is commonly associated with poor health outcomes [73], we hypothesize that decreased protein stability in carriers of the 482Ser allele may result in impaired coactivator activity, therefore increasing T2D and metabolic syndrome risk. It has been shown in previous studies that there is a 5-fold increase in PGC-1α protein in nuclear extracts following 1 hour treatment with the proteosomal inhibitor, MG132 [74]. Interestingly, after 6 and 12 hour treatment, PGC-1α protein was found to be polyubiquitinated and potentially protected from proteolysis but likely functionally inactive [75]. This suggested to us that PGC-1α levels and/or activity could be, in part, regulated by the ubiquitin proteosome pathway.

In agreement with these findings, we have shown that the half-life of the Ser482 variant is shorter than the Gly482 variant in cultured liver cells, corresponding with reduced coactivator activity on target genes involved in ROS detoxification [76]. This was also found in INS1 beta cells [77]. It is also possible that the differences in protein stability could be attributed, at least in part, to differences in mRNA expression, as has been shown in muscle [78] and islets [28] of serine allele carriers with T2D. These studies provided the first piece of evidence suggesting that a serine at position 482 destabilizes PGC-1α mRNA and protein. These data are in concordance with studies showing that the 482Ser variant increases fat deposition and impairs PEPCK expression in liver cells exposed to palmitate [79].  Conversely, the Gly482 variant has reduced coactivator activity on PPRE and Tfam promoters Chang human hepatocytes [80]; there is also no difference in coactivation of the ACBP-1 promoter when expressed HepG2 human hepatocytes [81], suggesting cell-type specific regulation of protein stability and/or differential sensitivity or targeting of specific promoters by the variants.

3.2 – Post-translational modifications of PGC-1α

PGC-1α transcription, protein stability and activity are very precisely controlled. PGC-1α expression levels change rapidly and significantly in response to physiological stressors or increased energy demand (e.g. cold, exercise, fasting, inflammation) [82], and the protein is quickly degraded [74]. Influences of external stimuli on PGC-1α activity, mechanisms controlling mRNA transcription and protein stability are also tissue-specific [17, 83-85]. An additional level of complexity to PGC-1α regulation is added by extensive post-translational modification (PTMs). Phosphorylation, acetylation, ubiquitination and methylation all affect PGC-1α stability and activity [17]. For example, phosphorylation by Akt at the 570Ser residue enhances PGC-1α activity, whereas Clk2 phosphorylation of serine residues in the SR region (spanning residues 564-635) decreases coactivator activity [17]. Coactivator activity can also be increased by stabilizing the protein itself. For instance, p38 MAPK phosphorylates PGC-1α at Ser/Thr residues and increases the half-life of the protein by approximately 3 times (6.27h vs. 2.38h) [86]. Conversely, there are other cases where phosphorylation has the opposite effect: phosphorylation by glycogen synthase kinase 3β (GSK-3β) serves as a marker for ubiquitination and subsequent intranuclear degradation [87]. In some cases, there are phosphorylation events that generate phosphodegrons, which are defined as short, motifs that, once phosphorylated, serve as markers for protein ubiquitination [88]. Skp1-Cullin-F-box ligase (SCF) is an example of a ubiquitin ligase that has been shown to play a role in PGC-1α ubiquitination and subsequent proteosomal degradation [89]. With these data in mind we therefore propose the model that phosphorylation of the 482Ser residue leads to decreased protein stability and thus, impaired coactivator function.

4 – Phosphoproteomics 

Objectives

Many studies show that there is a correlation between the Gly482Ser polymorphism and increased type 2 diabetes susceptibility.  However, little is known about the polymorphism and studies focusing on the pathophysiology of the variant are sparse, particularly concerning its function within the β-cell. Based on previous findings suggesting that the 482Ser variant is less stable in INS-1 beta cells, we hypothesized that a phosphorylation event at the 482Ser could be responsible for PGC-1α destabilization. The objectives of this study are to 1) ascertain the 482Ser residue as a phospho-site, 2) elucidate the kinase responsible for the phosphorylation and 3) assess the effect of the Gly482Ser polymorphism on coactivator activity.

Materials & Methods

Cell culture and treatments: INS-1 β-cells were cultured in RPMI 1640 (Winsent) medium with 10% Heat Inactivated FBS (Winsent), 1% penicillin and streptomycin (Winsent), and 1x supplement (10mM HEPES, 1mM sodium pyruvate, 50 µM β- mercaptoethanol). Cells were plated in 6 well plates the night before experiments. Cells were transfected when 80-90% confluent, with Lipofectamine 2000 (2.5 µl per well) and 1 µg of DNA, with OPTIMEM reduced serum media. After 4 hours, the transfection media was replaced with normal INS-1 media. Cells were left for 24-48 hours (as specified) at 37°C with 5% CO₂. Cycloheximide chase experiments were done with 50 mg/ml of cycloheximide (Sigma-Aldrich), MG-132 (Sigma-Aldrich) used at a concentration of 10μM per ml and actinomycin D (Sigma-Aldrich) at 1 µg/ml.

Western Blot: Cells were lysed in RIPA buffer (50mM Tris, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS) with protease cocktail inhibitor (Calbiochem), and phosphatase inhibitor (Roche). Protein concentration was estimated using DC assay on BSA standards. Equal amounts of protein (20-40 µg) were loaded onto 10-12% polyacrylamide gels, and were transferred onto polyvinylidene diflouride (PVDF) membranes (GE healthcare). PGC-1α was detected using anti-PGC-1α mouse mAb (4Cl.3) antibody (1:1000, Calbiochem) diluted in 2% milk. Membranes were incubated overnight at 4°C. Secondary antibody used was anti-mouse IgG antibody (BioRad) conjugated to HRP (1:5000). Signal detection was performed using the ECL detection system.

RNA isolation and cDNA synthesis: Total RNA from INS1 cells was extracted using Trizol reagent (Invitrogen) as indicated by the manufacturer’s protocol.

For cDNA synthesis, 1 µg of RNA was incubated with 1 U/ml DNAse1 at 37°C for 15 minutes followed by 15 minutes at 65°C for DNAse1 heat inactivation. Total RNA in a total volume of 20 µl was reverse transcribed with 50 U Multiscribe reverse transcriptase (Applied Biosystems) and 20 U RNAs inhibitor (Biobasic). cDNA was synthesized at 25°C for 10 minutes, 37°C for 120 minutes and 85°C for 5 minutes. 80 µl of water (1:5 dilution) was added to each sample and was stored at -20°C. cDNA samples were further assessed with qPCR.

Quantitative Real-Time PCR: cDNA was subjected to amplifications for the gene of interest and for the endogenous control hypoxanthine-guanine phosphoribosyl transferase (HPRT). 5 µl reactions were set up in a 384 well plate, using Power SYBR green PCR Master Mix (Life Technologies). The cycling program was in two steps, a polymerase activation step for 2 minutes at 50°C and 10 minutes at 95°C, followed by 40 cycles of 15 seconds at 95°C and 1 min of 60°C using the Viia 7 system from Life Technologies.  Data was normalized to the endogenous control and relative mRNA expression was determined using the ∆∆Ct method. Graphpad Prism was used for graphing results and performing statistical analysis.

Immunoprecipitation & Mass Spectrophotometry: HEK293 cells were cultured in DMEM (Winsent) medium with 10% FBS (Winsent) and 1% penicillin and streptomycin (Winsent) in 15 cm² plates. At 70% confluency cells were infected with Flag-PGC-1α glycine and serine constructs using Adenovirus. Media was replaced after 24 hours and cells were harvested after 48 hours. Cells were pelleted under cold conditions and then loaded into a syringe so they could be prepared for cryolysis using a PM-100 grinder. 10-15 g of cell powder was added to extraction buffer and subsequently, the mixture was poured onto Flag-conjugated magnetic beads (Dynabeads by ThermoFisher Scientific). The proteins were immunoprecipitated for 30 minutes at 4°C on a rotator, and then washed 3x in fresh extraction buffer without Tween. Flag-PGC-1α proteins were digested on the beads using 500 ng of LysC for 4 hours and overnight, both at at 37°C. Phosphopeptides were enriched using ZipTips (EMD Millipore)

CRISPR mice: 2 gRNA specific for the desired PGC-1α mutation (1444 A to G) were cloned into the Cas9-encoding px330 plasmid. Embryos from B6C3F1 mice were microinjected with both plasmidstogether with the single stranded oligodioxynucleotides (ssODN) that acts as a template during homology directed repair (HDR). The embryos were implanted into pseudo-pregnant female mice. After weaning, the pups were screened using PCR F: 5’-GCTAATGGATCCTACATTTCTTTTTGTTTC -3’ and R: 5’-GCAACTTGCCTCTTAGCGC-3’ to amplify the mutated region from genomic DNA, and subsequent Age1 restriction digest. Taqman SNP Genotyping Assay (Advanced Biosystems) was also used to screen for the presence of the Gly482 allele.

Kinome Screen: 2 sets of annealed oligos, encoding: N – QAVFDDEADKT[G/S]ELRDSDFSNEQ – C, were cloned into the GST-encoding  pgex 4t1 plasmid. Single BL-21 bacterial colonies containing the GST-fusion were incubated overnight at 37°C in LB broth containing 100 ug/ml ampicillin (Amp). Next, 500 mL of LB-Amp with 0.1 mM IPTG was added to the flask to induce GST-peptide expression. After 4 hours, the cells were pelleted by centrifugation at 400 rpm and stored at -80°C. The pellets were resuspended in 1X PBS and then sonicated until the mixture was no longer viscous. The sample was centrifuged and the supernatant retained for fusion peptide purification using the GST Fusion Protein Purification Kit (Genscript) as per manufacturers instruction in the TM0185 manual. Crude, input, wash and sample aliquots were retained and assessed using coommassie staining of a 10% SDS-PAGE gel.

1. Organization., W.H., Global report on diabetes, in

. 2016, World Health Organization: France.

2. Reinehr, T., Type 2 diabetes mellitus in children and adolescents. World Journal of Diabetes, 2013. 4(6): p. 270-281.

3. Ye, J., Mechanisms of insulin resistance in obesity. Frontiers of medicine, 2013. 7(1): p. 14-24.

4. Bouzakri, K. and J.R. Zierath, MAP4K4 gene silencing in human skeletal muscle prevents tumor necrosis factor-alpha-induced insulin resistance. J Biol Chem, 2007. 282(11): p. 7783-9.

5. Daniele, G., et al., Chronic Reduction of Plasma Free Fatty Acid Improves Mitochondrial Function and Whole-Body Insulin Sensitivity in Obese and Type 2 Diabetic Individuals. Diabetes, 2014. 63(8): p. 2812-2820.

6. Rui, L., Energy Metabolism in the Liver. Compr Physiol, 2014. 4(1): p. 177-97.

7. Kahn, S.E., The importance of the beta-cell in the pathogenesis of type 2 diabetes mellitus. Am J Med, 2000. 108 Suppl 6a: p. 2s-8s.

8. Barnett, A.H., et al., Diabetes in identical twins. A study of 200 pairs. Diabetologia, 1981. 20(2): p. 87-93.

9. Newman, B., et al., Concordance for type 2 (non-insulin-dependent) diabetes mellitus in male twins. Diabetologia, 1987. 30(10): p. 763-8.

10. Raciti, G.A., et al., Understanding type 2 diabetes: from genetics to epigenetics. Acta Diabetologica, 2015. 52(5): p. 821-827.

11. Bouchard-Mercier, A., et al., Associations between polymorphisms in genes involved in fatty acid metabolism and dietary fat intakes. J Nutrigenet Nutrigenomics, 2012. 5(1): p. 1-12.

12. Wang, X., et al., Genetic markers of type 2 diabetes: Progress in genome-wide association studies and clinical application for risk prediction. J Diabetes, 2016. 8(1): p. 24-35.

13. Ali, O., Genetics of type 2 diabetes. World Journal of Diabetes, 2013. 4(4): p. 114-123.

14. Pratley, R.E., et al., An autosomal genomic scan for loci linked to prediabetic phenotypes in Pima Indians. J Clin Invest, 1998. 101(8): p. 1757-64.

15. Puigserver, P., et al., A cold-inducible coactivator of nuclear receptors linked to adaptive thermogenesis. Cell, 1998. 92(6): p. 829-39.

16. Wu, Z., et al., Mechanisms controlling mitochondrial biogenesis and respiration through the thermogenic coactivator PGC-1. Cell, 1999. 98(1): p. 115-24.

17. Fernandez-Marcos, P.J. and J. Auwerx, Regulation of PGC-1alpha, a nodal regulator of mitochondrial biogenesis. Am J Clin Nutr, 2011. 93(4): p. 884s-90.

18. Devarakonda, S., et al., Disorder-to-order transition underlies the structural basis for the assembly of a transcriptionally active PGC-1alpha/ERRgamma complex. Proc Natl Acad Sci U S A, 2011. 108(46): p. 18678-83.

19. Wallberg, A.E., et al., Coordination of p300-Mediated Chromatin Remodeling and TRAP/Mediator Function through Coactivator PGC-1α. Molecular Cell, 2003. 12(5): p. 1137-1149.

20. Finck, B.N. and D.P. Kelly, PGC-1 coactivators: inducible regulators of energy metabolism in health and disease. Journal of Clinical Investigation, 2006. 116(3): p. 615-622.

21. Kleiner, S., et al., PPAR{delta} agonism activates fatty acid oxidation via PGC-1{alpha} but does not increase mitochondrial gene expression and function. J Biol Chem, 2009. 284(28): p. 18624-33.

22. Vega, R.B., J.M. Huss, and D.P. Kelly, The coactivator PGC-1 cooperates with peroxisome proliferator-activated receptor alpha in transcriptional control of nuclear genes encoding mitochondrial fatty acid oxidation enzymes. Mol Cell Biol, 2000. 20(5): p. 1868-76.

23. Wu, H., et al., PGC-1α, glucose metabolism and type 2 diabetes mellitus. Journal of Endocrinology, 2016. 229(3): p. R99-R115.

24. Wende, A.R., et al., A Role for the Transcriptional Coactivator PGC-1α in Muscle Refueling. Journal of Biological Chemistry, 2007. 282(50): p. 36642-36651.

25. Rhee, J., et al., Partnership of PGC-1alpha and HNF4alpha in the regulation of lipoprotein metabolism. J Biol Chem, 2006. 281(21): p. 14683-90.

26. Sczelecki, S., et al., Loss of Pgc-1alpha expression in aging mouse muscle potentiates glucose intolerance and systemic inflammation. Am J Physiol Endocrinol Metab, 2014. 306(2): p. E157-67.

27. Correia, J.C., D.M. Ferreira, and J.L. Ruas, Intercellular: local and systemic actions of skeletal muscle PGC-1s. Trends Endocrinol Metab, 2015. 26(6): p. 305-14.

28. Ling, C., et al., Epigenetic regulation of PPARGC1A in human type 2 diabetic islets and effect on insulin secretion. Diabetologia, 2008. 51(4): p. 615-622.

29. Oropeza, D., et al., PGC-1 coactivators in β-cells regulate lipid metabolism and are essential for insulin secretion coupled to fatty acids. Molecular Metabolism, 2015. 4(11): p. 811-822.

30. Kleiner, S., et al., Development of insulin resistance in mice lacking PGC-1alpha in adipose tissues. Proc Natl Acad Sci U S A, 2012. 109(24): p. 9635-40.

31. Muller, Y.L., et al., A Gly482Ser missense mutation in the peroxisome proliferator-activated receptor gamma coactivator-1 is associated with altered lipid oxidation and early insulin secretion in Pima Indians. Diabetes, 2003. 52(3): p. 895-8.

32. Ek, J., et al., Mutation analysis of peroxisome proliferator-activated receptor-γ coactivator-1 (PGC-1) and relationships of identified amino acid polymorphisms to Type II diabetes mellitus. Diabetologia, 2001. 44(12): p. 2220-2226.

33. Fanelli, M., et al., The Gly482Ser missense mutation of the peroxisome proliferator-activated receptor gamma coactivator-1 alpha (PGC-1 alpha) gene associates with reduced insulin sensitivity in normal and glucose-intolerant obese subjects. Dis Markers, 2005. 21(4): p. 175-80.

34. Hara, K., et al., A genetic variation in the PGC-1 gene could confer insulin resistance and susceptibility to Type II diabetes. Diabetologia, 2002. 45(5): p. 740-3.

35. Esterbauer, H., et al., Peroxisome Proliferator-Activated Receptor-γ Coactivator-1 Gene Locus. Diabetes, 2002. 51(4): p. 1281.

36. Pihlajamaki, J., et al., Haplotypes of PPARGC1A are associated with glucose tolerance, body mass index and insulin sensitivity in offspring of patients with type 2 diabetes. Diabetologia, 2005. 48(7): p. 1331-4.

37. Povel, C.M., et al., Glucose levels and genetic variants across transcriptional pathways: interaction effects with BMI. Int J Obes (Lond), 2010. 34(5): p. 840-5.

38. Soyal, S., et al., PGC-1alpha: a potent transcriptional cofactor involved in the pathogenesis of type 2 diabetes. Diabetologia, 2006. 49(7): p. 1477-88.

39. Maciejewska, A., et al., The PPARGC1Agene Gly482Ser in Polish and Russian athletes. Journal of Sports Sciences, 2012. 30(1): p. 101-113.

40. Norbert, S., et al., Genetic Variations in PPARD and PPARGC1A Determine Mitochondrial Function and Change in Aerobic Physical Fitness and Insulin Sensitivity during Lifestyle Intervention. The Journal of Clinical Endocrinology & Metabolism, 2007. 92(5): p. 1827-1833.

41. Tural, E., et al., PPAR-α and PPARGC1A gene variants have strong effects on aerobic performance of Turkish elite endurance athletes. Molecular Biology Reports, 2014. 41(9): p. 5799-5804.

42. Eynon, N., et al., Do PPARGC1A and PPARα polymorphisms influence sprint or endurance phenotypes? Scandinavian Journal of Medicine & Science in Sports, 2010. 20(1): p. e145-e150.

43. Kunej, T.T., A Gly482Ser polymorphism of the peroxisome proliferator-activated receptor-gamma coactivator-1 (PGC-1) gene is associated with type 2 diabetes in Caucasians. Folia biologica, 2004. 50(5): p. 157-8.

44. Myles, S., et al., Testing the thrifty gene hypothesis: the Gly482Ser variant in PPARGC1A is associated with BMI in Tongans. BMC Medical Genetics, 2011. 12: p. 10-10.

45. Cadzow, M., et al., Lack of direct evidence for natural selection at the candidate thrifty gene locus, PPARGC1A. BMC Medical Genetics, 2016. 17: p. 80.

46. Bhat, A., et al., PGC-1alpha Thr394Thr and Gly482Ser variants are significantly associated with T2DM in two North Indian populations: a replicate case-control study. Hum Genet, 2007. 121(5): p. 609-14.

47. Rai, E.E., Interaction between the UCP2-866G/A, mtDNA 10398G/A and PGC1alpha p.Thr394Thr and p.Gly482Ser polymorphisms in type 2 diabetes susceptibility in North Indian population. Human Genetics. 122(5): p. 535-40.

48. Shokouhi Shabnam, S., Association between PGC-1alpha gene polymorphisms and type 2 diabetes risk: a case-control study of an Iranian population. Canadian Journal of Diabetes. 39(1): p. 65-72.

49. Sun, L., et al., The Gly482Ser variant of the PPARGC1 gene is associated with Type 2 diabetes mellitus in northern Chinese, especially men. Diabet Med, 2006. 23(10): p. 1085-92.

50. Jemaa, Z., et al., The Gly482Ser polymorphism of the peroxisome proliferator-activated receptor-gamma coactivator-1alpha (PGC-1alpha) is associated with type 2 diabetes in Tunisian population. Diabetes Metab Syndr, 2015. 9(4): p. 316-9.

51. Andrulionyte, L., et al., Common polymorphisms of the PPAR-gamma2 (Pro12Ala) and PGC-1alpha (Gly482Ser) genes are associated with the conversion from impaired glucose tolerance to type 2 diabetes in the STOP-NIDDM trial. Diabetologia, 2004. 47(12): p. 2176-84.

52. Franks, P.W., et al., Common variation at PPARGC1A/B and change in body composition and metabolic traits following preventive interventions: the Diabetes Prevention Program. Diabetologia, 2014. 57(3): p. 485-90.

53. Ha, C.D., et al., Relationship of PGC-1alpha gene polymorphism with insulin resistance syndrome in Korean children. Asia Pac J Public Health, 2015. 27(2): p. Np544-51.

54. Weng, S.-W., et al., Gly482Ser polymorphism in the peroxisome proliferator&#x2013;activated receptor <em>&#x3b3;</em> coactivator&#x2013;1<em>&#x3b1;</em> gene is associated with oxidative stress and abdominal obesity. Metabolism – Clinical and Experimental. 59(4): p. 581-586.

55. Vázquez-Del Mercado, M., et al., The 482Ser of PPARGC1A and 12Pro of PPARG2 Alleles Are Associated with Reduction of Metabolic Risk Factors Even Obesity in a Mexican-Mestizo Population. BioMed Research International, 2015. 2015: p. 285491.

56. Deeb, S.S. and J.D. Brunzell, The Role of the PGC1α Gly482Ser Polymorphism in Weight Gain due to Intensive Diabetes Therapy. PPAR Research, 2009. 2009: p. 649286.

57. Vimaleswaran, K.S., et al., Effect of polymorphisms in the PPARGC1A gene on body fat in Asian Indians. Int J Obes, 2006. 30(6): p. 884-891.

58. Albuquerque, D., et al., Association study of common polymorphisms in MSRA, TFAP2B, MC4R, NRXN3, PPARGC1A, TMEM18, SEC16B, HOXB5 and OLFM4 genes with obesity-related traits among Portuguese children. J Hum Genet, 2014. 59(6): p. 307-13.

59. Franks, P.W., et al., PPARGC1A coding variation may initiate impaired NEFA clearance during glucose challenge. Diabetologia, 2007. 50(3): p. 569-573.

60. Goyenechea, E., et al., Enhanced short-term improvement of insulin response to a low-caloric diet in obese carriers the Gly482Ser variant of the PGC-1alpha gene. Diabetes Res Clin Pract, 2008. 82(2): p. 190-6.

61. Geloneze, S.R., et al., PGC1[alpha] gene Gly482Ser polymorphism predicts improved metabolic, inflammatory and vascular outcomes following bariatric surgery. Int J Obes, 2012. 36(3): p. 363-368.

62. Yoon, J.C., et al., Suppression of β Cell Energy Metabolism and Insulin Release by PGC-1α. Developmental Cell, 2003. 5(1): p. 73-83.

63. Valtat, B., et al., Fetal PGC-1α Overexpression Programs Adult Pancreatic β-Cell Dysfunction. Diabetes, 2013. 62(4): p. 1206-1216.

64. Barroso, I., et al., Meta-analysis of the Gly482Ser variant in PPARGC1A in type 2 diabetes and related phenotypes. Diabetologia, 2006. 49(3): p. 501-5.

65. Su, Y., et al., [Association study between PPARGC1A Thr394Thr/ Gly482Ser polymorphisms and type 2 diabetes]. Yi Chuan, 2008. 30(3): p. 304-8.

66. Rojek, A.e.a., Impact of the PPARGC1A Gly482Ser polymorphism on left ventricular structural and functional abnormalities in patients with hypertension. . J Hum Hypertens,, 2014. 28(9): p. 557-563.

67. Wang, S., et al., Polymorphisms of the peroxisome proliferator-activated receptor-γ coactivator-1α gene are associated with hypertrophic cardiomyopathy and not with hypertension hypertrophy, in Clinical Chemical Laboratory Medicine. 2007. p. 962.

68. Zhang, Y.e.a., Association between PPARGC1A gene polymorphisms and coronary artery disease in a Chinese population. Clin Exp Pharmacol Physiol, 2008. 35(10): p. 1172-1177.

69. Ballestri, S., et al., Nonalcoholic fatty liver disease is associated with an almost twofold increased risk of incident type 2 diabetes and metabolic syndrome. Evidence from a systematic review and meta-analysis. J Gastroenterol Hepatol, 2016. 31(5): p. 936-44.

70. Lin, Y.C., et al., A common variant in the peroxisome proliferator-activated receptor-gamma coactivator-1alpha gene is associated with nonalcoholic fatty liver disease in obese children. Am J Clin Nutr, 2013. 97(2): p. 326-31.

71. Tai, C.M., et al., Derivation and validation of a scoring system for predicting nonalcoholic steatohepatitis in Taiwanese patients with severe obesity. Surg Obes Relat Dis, 2016.

72. Gancheva, S., et al., Variants in Genes Controlling Oxidative Metabolism Contribute to Lower Hepatic ATP Independent of Liver Fat Content in Type 1 Diabetes. Diabetes, 2016. 65(7): p. 1849.

73. Handschin, C. and B.M. Spiegelman, Peroxisome Proliferator-Activated Receptor γ Coactivator 1 Coactivators, Energy Homeostasis, and Metabolism. Endocrine Reviews, 2006. 27(7): p. 728-735.

74. Sano, M., et al., Intramolecular control of protein stability, subnuclear compartmentalization, and coactivator function of peroxisome proliferator-activated receptor gamma coactivator 1alpha. J Biol Chem, 2007. 282(35): p. 25970-80.

75. Sano, M., et al., Intramolecular Control of Protein Stability, Subnuclear Compartmentalization, and Coactivator Function of Peroxisome Proliferator-activated Receptor γ Coactivator 1α. Journal of Biological Chemistry, 2007. 282(35): p. 25970-25980.

76. Besse-Patin, A., et al., Estrogen Signals through PPARG coactivator 1 alpha to Reduce Oxidative Damage Associated with Diet-induced Fatty Liver Disease. Gastroenterology.

77. Khan, N., Exploring the regulation of PGC-1α-investigating the polymorphism of PGC-1α and the PGC-1α4 isoform, in Department of Medicine. 2015, McGill University.

78. Ling, C., et al., Multiple environmental and genetic factors influence skeletal muscle PGC-1α and PGC-1β gene expression in twins. The Journal of Clinical Investigation, 2004. 114(10): p. 1518-1526.

79. Chen, Y., et al., Gly482Ser mutation impairs the effects of peroxisome proliferator–activated receptor γ coactivator-1α on decreasing fat deposition and stimulating phosphoenolpyruvate carboxykinase expression in hepatocytes. Nutrition Research, 2013. 33(4): p. 332-339.

80. Choi, Y.-S., et al., Impaired coactivator activity of the Gly482 variant of peroxisome proliferator-activated receptor γ coactivator-1α (PGC-1α) on mitochondrial transcription factor A (Tfam) promoter. Biochemical and Biophysical Research Communications, 2006. 344(3): p. 708-712.

81. Nitz, I., et al., Analysis of PGC-1alpha variants Gly482Ser and Thr612Met concerning their PPARgamma2-coactivation function. Biochem Biophys Res Commun, 2007. 353(2): p. 481-6.

82. Melloul, D. and M. Stoffel, Regulation of Transcriptional Coactivator PGC-1{alpha}. Science of Aging Knowledge Environment, 2004. 2004(9): p. pe9.

83. Akimoto, T., et al., Exercise stimulates Pgc-1alpha transcription in skeletal muscle through activation of the p38 MAPK pathway. J Biol Chem, 2005. 280(20): p. 19587-93.

84. Akimoto, T., et al., Skeletal muscle adaptation in response to voluntary running in Ca2+/calmodulin-dependent protein kinase IV-deficient mice. Am J Physiol Cell Physiol, 2004. 287(5): p. C1311-9.

85. Gomez-Ambrosi, J., G. Fruhbeck, and J.A. Martinez, Rapid in vivo PGC-1 mRNA upregulation in brown adipose tissue of Wistar rats by a beta(3)-adrenergic agonist and lack of effect of leptin. Mol Cell Endocrinol, 2001. 176(1-2): p. 85-90.

86. Puigserver, P., et al., Cytokine stimulation of energy expenditure through p38 MAP kinase activation of PPARgamma coactivator-1. Mol Cell, 2001. 8(5): p. 971-82.

87. Anderson, R.M., et al., Dynamic regulation of PGC-1α localization and turnover implicates mitochondrial adaptation in calorie restriction and the stress response. Aging Cell, 2008. 7(1): p. 101-111.

88. Holt, L.J., Regulatory modules: Coupling protein stability to phopshoregulation during cell division. FEBS Letters, 2012. 586(17): p. 2773-2777.

89. Olson, B.L., et al., SCF(Cdc4) acts antagonistically to the PGC-1α transcriptional coactivator by targeting it for ubiquitin-mediated proteolysis. Genes & Development, 2008. 22(2): p. 252-264.