SLC30A8 Nonsynonymous Variant Is Associated With Recovery Following Exercise and Skeletal Muscle Size and Strength
- Courtney Sprouse1,2,
- Heather Gordish-Dressman1,
- E. Funda Orkunoglu-Suer1,
- Jason S. Lipof3,
- Stephanie Moeckel-Cole4,
- Ronak R. Patel1,2,
- Kasra Adham1,2,
- Justin S. Larkin1,2,
- Monica J. Hubal1,
- Amy K. Kearns4,
- Priscilla M. Clarkson4,
- Paul D. Thompson5,
- Theodore J. Angelopoulos6,
- Paul M. Gordon7,
- Niall M. Moyna8,
- Linda S. Pescatello9,
- Paul S. Visich10,
- Robert F. Zoeller11,
- Eric P. Hoffman1,
- Laura L. Tosi1,2,3⇑ and
- Joseph M. Devaney1⇑
- 1Department of Integrative Systems Biology, Research Center for Genetic Medicine, Children’s National Medical Center, Washington, DC
- 2Department of Orthopedic Surgery and Sports Medicine, Children’s National Medical Center, Washington, DC
- 3George Washington University School of Medicine, Washington, DC
- 4Department of Kinesiology, University of Massachusetts, Amherst, MA
- 5Division of Cardiology, Henry Low Heart Center, Hartford Hospital, Hartford, CT
- 6Center for Lifestyle Medicine and Department of Health Professions, University of Central Florida, Orlando, FL
- 7Laboratory for Physical Activity and Exercise Intervention Research, University of Michigan, Ann Arbor, MI
- 8Department of Sport Science and Health, Dublin City University, Dublin, Ireland
- 9School of Allied Health, University of Connecticut, Storrs, CT
- 10Human Performance Laboratory, Central Michigan University, Mount Pleasant, MI
- 11Department of Exercise Science and Health Promotion, Florida Atlantic University, Boca Raton, FL
- Corresponding authors: Laura L. Tosi, , and Joseph M. Devaney, .
L.L.T. and J.M.D. contributed equally to this work.
Genome-wide association studies have identified thousands of variants that are associated with numerous phenotypes. One such variant, rs13266634, a nonsynonymous single nucleotide polymorphism in the solute carrier family 30 (zinc transporter) member eight gene, is associated with a 53% increase in the risk of developing type 2 diabetes (T2D). We hypothesized that individuals with the protective allele against T2D would show a positive response to short-term and long-term resistance exercise. Two cohorts of young adults—the Eccentric Muscle Damage (EMD; n = 156) cohort and the Functional Single Nucleotide Polymorphisms Associated with Muscle Size and Strength Study (FAMuSS; n = 874)—were tested for association of the rs13266634 variant with measures of skeletal muscle response to resistance exercise. Our results were sexually dimorphic in both cohorts. Men in the EMD study with two copies of the protective allele showed less post-exercise bout strength loss, less soreness, and lower creatine kinase values. In addition, men in the FAMuSS, homozygous for the protective allele, showed higher pre-exercise strength and larger arm skeletal muscle volume, but did not show a significant difference in skeletal muscle hypertrophy or strength with resistance training.
Genome-wide association studies (GWASs) are extensively used as a tool to discover common single nucleotide polymorphisms (SNPs) associated with complex disease phenotypes such as type 2 diabetes (T2D), dyslipidemia, and obesity. One SNP discovered utilizing a GWAS is located in the solute carrier family 30 (zinc transporter), member eight gene (SLC30A8), on chromosome 8q24.11. This nonsynonymous variant (rs13266634) is a C-to-T variant (arginine to tryptophan at position 325 [R325W]) that is associated with the risk of developing T2D (1–4). This SNP is associated with several prediabetic phenotypes, including reduced insulin secretion in response to orally and intravenously administered glucose (5), lower fasting insulin levels (6), reduced disposition index (7), and decreased proinsulin-to-insulin conversion (6). The SLC30A8 rs13266634 polymorphism is among the most confirmed genetic markers of T2D in Europeans and East Asians (8).
The SLC30A8 gene encodes a protein, Zinc Efflux Transporter 8 protein, that is expressed primarily in the pancreatic α-cells and β-cells (9). This transmembrane protein transports zinc from the cytoplasm into insulin secretory vesicles, where insulin is stored as a solid hexamer bound with two Zn2+ ions within the pancreatic islet β-cells (10). Zn2+ ions tightly regulate insulin production, and are essential for the storage and crystallization of insulin (11).
Many studies have investigated the effects of endurance exercise on T2D because of the improvement in insulin sensitivity, the reduction of glucose levels, and improved glycemic control (12–14). However, the effects of resistance exercise on T2D risk have not been widely studied. van Djik et al. (15) recently published a study on the effects of endurance versus resistance exercise on glucose homeostasis that demonstrated equal reductions in blood glucose concentrations in participants, regardless of exercise type. However, this study measured only the effects of an acute exercise intervention, and this does not always reflect long-term exercise outcomes.
We hypothesized that a copy of the risk allele (C allele) for the SLC30A8 variant will affect the response of an individual to a resistance training intervention. In addition, we wanted to compare and quantitate the genotype effect on an acute exercise intervention (single bout) versus a long-term 12-week resistance training intervention. Therefore, we included the following two different exercise cohorts: 1) Eccentric Muscle Damage (EMD) cohort allowed us to evaluate the effects of acute resistance exercise to exhaustion (16,17); and 2) the Functional Single Nucleotide Polymorphisms Associated with Muscle Size and Strength Study (FAMuSS) provided evaluation of a chronic resistance exercise program, a longitudinal 12-week resistance exercise training cohort (18,19). In this article, we demonstrated that two copies of the protective allele (TT) against the development of T2D are associated with a positive response to an acute resistance training exercise bout and show an association with pre-exercise measures of skeletal muscle size and strength. However, two copies of the protective allele (TT) did not improve skeletal muscle hypertrophy or strength after a 12-week resistance training program.
Research Design and Methods
This study was approved by the University of Massachusetts Institutional Review Board and contained a total of 203 participants; however, we were able to extract DNA from only 156 individuals (78 men and 78 women between the ages of 18 and 40 years; Table 1). The individuals in the study were primarily of Caucasian descent (74% Caucasian: 115 Caucasians, 4 African Americans, 6 Hispanics, 20 Asians, and 11 subjects who self-classified their race as “other”). The participants performed two sets of 25 maximal eccentric contractions of the elbow flexor muscles on a modified preacher curl bench. Serum samples were collected at baseline (pre-exercise), and at 4, 7, and 10 days after exercise for the analysis of biomarkers of muscle damage (creatine kinase [CK] and myoglobin [Mb]) (Supplementary Fig. 1). Participants were evaluated at 1 and 12 h, and at 3, 4, 7, and 10 days after the exercise bout for strength and muscle soreness using a visual analog scale (a 100-mm line on which participants place a mark between 0 [no soreness] and 100 mm [unbearable pain]) (Supplementary Fig. 1).
FAMuSS is a study of healthy Caucasian adults 18–40 years of age (Table 2), and this study was approved by the Children’s National Medical Center Institutional Review Board. All participants undertook 12 weeks of progressive resistance training using their nondominant arm only (18,19). Arm strength and skeletal muscle volume were measured before and after the training using maximal voluntary isometric contraction (MVC) and one-repetition maximum (1-RM) (Supplementary Fig. 1) (18,19). Upper arm skeletal muscle volume was quantified using magnetic resonance imaging (19).
Genotyping was performed with a TaqMan assay (assay identification #C_357888_10) using standard thermal cycling conditions with genotypes determined by a 7900HT system (Life Technologies).
For both cohorts, we decided to use a recessive model (TT vs. CT/CC) for the genotype/phenotype relationships. This statistical model fits with that of Sladek et al. (2), who showed that a copy of the C allele was associated with T2D; therefore, two copies of the T allele were protective. We hypothesized that a copy of the risk allele (C allele) for the rs13266634 variant might affect the response to a resistance exercise intervention. In addition, we wanted to quantitate the genotype effect on an acute bout of resistance training and a 12-week resistance exercise intervention.
The phenotypes tested for an association with SLC30A8 genotype were as follows: baseline isometric strength (MVC); relative (compared with baseline) MVC 1 h after the eccentric exercise; relative MVC at 12 h, and 3, 4, 7, and 10 days after exercise. Soreness after the exercise bout was analyzed for an association with genotype at time points 12 h, and 3, 4, 7, and 10 days after the exercise bout. Serum measurements of CK and Mb were analyzed pre-exercise (baseline) before the exercise bout and at 4 and 7 days after the exercise bout. In addition, CK was analyzed at 10 days after the exercise bout; however, the Mb assay for 10 days after the exercise bout failed, and the data are not included in the analyses. We have added a figure to show the various time points and the data that were collected (Supplementary Fig. 1).
Normality of each quantitative trait was tested using the Shapiro-Wilks normality test. Mean quantitative muscle measurements were compared with SNP genotypes using ANCOVA methods or quantile regression for those phenotypes not normally distributed. The ANCOVAs used Sidak post hoc tests to control for multiple comparisons. Analyses were conducted in sex-specific cohorts while covarying for body mass, age, and treatment.
The phenotypes measured in the FAMuSS that were tested for associations with SLC30A8 genotype were BMI, measures of strength (1-RM and MVC; before and after training), and volumetric measurements of skeletal muscle and fat of the upper arm (before and after training) Supplementary Fig. 1 shows the exercise protocol and the data points collected. Mean quantitative muscle measurements were compared with the SNP genotype in a sex-specific manner covarying with body mass and age using ANCOVA methods, or quantile regression for phenotypes not normally distributed. The percentage variation attributable to the SNP was determined with a likelihood ratio, and recessive models were used (TT vs. CT/CC).
The results are reported as raw/adjusted means ± SEM. All analyses were performed using Stata version 8.2 (StataCorp, College Station, TX) with a significance level at P < 0.05, but, given that we performed multiple statistical tests, those tests under P < 0.01 should be viewed with caution and require validation in future studies.
The allele frequency in the EMD study was 57% for CC, 39% for CT, and 4% for TT (Hardy-Weinberg equilibrium, P = 0.35). We found an association for a copy of the risk allele for T2D (CC/CT), and increased strength loss, increased soreness, increased CK levels, and increased Mb levels in males (Table 3). Males with two copies of the protective allele experienced less muscle soreness and strength loss, and had decreased levels of CK and Mb, after the muscle damage session (Table 3). The Mb data for day 10 were the same as the baseline Mb value and is not included in Table 3. The outcome measures demonstrated no association with this variant in females in the EMD study (see Supplementary Table 1). Finally, the SLC30A8 variant was not associated with BMI in males or females.
This SNP did not violate the Hardy-Weinberg equilibrium (P = 0.971). Males with a copy of the risk allele for T2D (CC/CT) had lower pre-exercise strength and lower pre-exercise upper skeletal muscle volume (Table 4). Interestingly, the SNP was not associated with gain in strength after an exercise bout, change in muscle volume, or change in fat volume in males. The outcome measures demonstrated no association with this variant in females (see Supplementary Table 2). In addition, this nonsynonymous variant in the SLC30A8 gene was not associated with BMI in males or females.
The goal of this study was to investigate whether a nonsynonymous SNP, discovered by GWAS, and related to the development of T2D (1–8), was linked with parameters of skeletal muscle size, strength, or response to a resistance exercise intervention. We discovered that a copy of the risk allele was associated with skeletal muscle strength loss, skeletal muscle soreness, and CK and Mb levels after a short bout of resistance exercise and with lower pre-exercise skeletal muscle size and strength in males.
The SLC30A8 gene encodes a zinc transporter expressed solely in the secretory vesicles of β-cells and is implicated in the final stages of insulin biosynthesis involving cocrystallization with zinc (20). The communication between skeletal muscle and β-cells is thought to contribute to healthy skeletal muscle function and mass (21). For example, interleukin (IL-6) is a prototype myokine and is identified as a muscle contraction–induced factor (22). Potentially, IL-6 affects β-cell function, but more detailed studies are needed to clarify the direct impact of this myokine on pancreatic β-cells in health and disease, especially because the α-cell has been identified as a major IL-6 target in islets (23). In addition, multiple in vitro studies have shown that multiple cytokines and chemokines can affect the function, survival, and proliferation of β-cells (21). Therefore, a variant in the SLC30A8 gene could play a role in skeletal muscle size and strength as a signal molecule in the communication pathway between skeletal muscle and β-cells. There is a link between secretory molecules from skeletal muscle (myokines) and β-cell function (21), and the SLC30A8 gene may play a role in this communication.
Interestingly, this variant showed sexually dimorphic results: only males had an association in both cohorts, which is consistent with the previous EMD and FAMuSS cohorts (17,19), and additional articles showing the genetic influence on fitness phenotypic traits (24,25). This suggests that the sex-specific genetic architecture influencing phenotypes needs to be incorporated into the design of individualized exercise interventions to optimize results.
There are some limitations to this study; it was a retrospective study, and therefore we were unable to include phenotypes, such as circulating insulin levels. The SLC30A8 gene is primarily expressed in β-cells; however, a relationship exists between β-cells and skeletal muscle that needs to be explored (21).
Understanding this missense variant in the SLC30A8 gene has many implications for the future of diabetes research, including examination of the effect of myokines on β-cells. In addition, our findings suggest that a personalized approach to exercise prescription may be warranted based upon genotype and sex. The idea that resistance exercise may be painful to men possessing the common allele may change these individuals to an endurance-based exercise program that would be beneficial in preventing the development of T2D. It would be interesting to investigate whether there are males with common SLC30A8 alleles that reap more benefit from resistance exercise compared with endurance exercise.
Duality of Interest. No potential conflicts of interest relevant to this article were reported.
Author Contributions. C.S. generated data and wrote the manuscript. H.G.-D. performed all statistical analyses on data acquired. J.S.Li. researched data and reviewed and edited the manuscript. S.M.-C. organized and analyzed data. R.R.P., K.A., and J.S.La. performed genotyping. E.F.O.-S., M.J.H., A.K.K., P.D.T., T.J.A., P.M.G., N.M.M., L.S.P., P.S.V., R.F.Z., and E.P.H. were part of the FAMuSS and edited the manuscript. P.M.C. performed, organized, and recorded all measurements on cohorts. L.L.T. and J.M.D. wrote, reviewed, and edited the manuscript. J.M.D. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
This article contains Supplementary Data online at http://diabetes.diabetesjournals.org/lookup/suppl/doi:10.2337/db13-1150/-/DC1.
- Received July 24, 2013.
- Accepted September 30, 2013.
- © 2014 by the American Diabetes Association.
Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. See http://creativecommons.org/licenses/by-nc-nd/3.0/ for details.