Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Mar 1.
Published in final edited form as: Obesity (Silver Spring). 2020 Feb 6;28(3):676–682. doi: 10.1002/oby.22741

Low Serum Insulin-like Growth Factor-II Levels Correlate with High Body Mass Index in American Indian Adults

Yunhua L Muller 1, Robert L Hanson 1, Darin Mahkee 1, Paolo Piaggi 1, Sayuko Kobes 1, Wen-Chi Hsueh 1, William C Knowler 1, Clifton Bogardus 1, Leslie J Baier 1
PMCID: PMC7192225  NIHMSID: NIHMS1547056  PMID: 32030914

Abstract

Objective:

The insulin-like growth factor-II (IGF-II) regulates metabolism and growth. In humans both positive and negative relationships have been reported between serum IGF-II levels and obesity. We assess the relationship between serum IGF-II levels and Body Mass Index (BMI); and determine whether IGF-II levels predict weight gain.

Methods:

Serum samples were available on 911 American Indians with a recorded BMI. IGF-II was measured using Enzyme-Linked ImmunoSorbent Assay.

Results:

Serum IGF-II levels were negatively correlated with BMI (r=−0.17, P=4.4×10−7, adjusted for age, sex, storage time). The strongest correlation was in subjects ≥30 years (r=−0.28, P=3.4×10−8, N=349), a modest correlation was in subjects 20–29 years (r=−0.15, P=7.6×10−3, N=322), and subjects 15–19 years had no correlation (r=0.05, P=0.48, N=240). IGF-II levels did not predict weight gain. However, among individuals who had genotypes for 64 established obesity variants (age≥20 years, N=671), a genetic risk score for high BMI associated with lower IGF-II (β=−0.08 SD IGF-II per SD genetic risk score, P=0.025).

Conclusions:

There is a negative relationship between IGF-II levels and BMI, where the correlation is stronger at older ages. The association between genetic risk for BMI and IGF-II levels suggests that this correlation may be due to an effect of obesity on IGF-II.

Keywords: serum Insulin-like Growth Factor-II, Body Mass Index, American Indian, single nucleotide polymorphism

Introduction

The hormone insulin-like growth factor-II (IGF-II) plays an important role in the regulation of metabolism and growth. IGF-II is a ligand for both type I and type II IGF receptors as well as the insulin receptor, but its affinity is the highest for the type II IGF receptor. IGF-II expression is mainly regulated by growth hormone and nutrition (1).

Several lines of evidence in both animal and human studies have shown a relationship between IGF-II and body weight or BMI (2–3). In mouse models, overexpression of IGF-II led to a significant reduction in fat mass and lipid content (4) while reduction of IGF-II, via deletion of a control region intergenic to the Igf2 and H19 genes, associated with increased adiposity (5). In a human study of males, an ApaI polymorphism in the 3’UTR of the IGF2 gene (rs680; GG to AA) was associated with both body weight and circulating IGF-II levels, where AA carriers had a lower mean body weight, but higher serum IGF-II concentrations compared to GG carriers (2). However, human population-based studies assessing the cross-sectional association between circulating IGF-II levels and obesity are few and inconsistent. While one study reported that obese people have lower IGF-II levels compared with leaner people (6), another clinical study reported higher IGF-II concentrations in people with obesity (7).

The Pima Indians of Arizona are a relatively isolated population with a high prevalence of obesity and type 2 diabetes (8). Longitudinal data have been collected in a population-based sample of Pima Indians which allows for assessment of both a cross-sectional analysis of the relationship between serum IGF-II levels and BMI as well as a prospective analysis of serum IGF-II levels as a predictor of future weight gain. Genome-wide genotypic data are also available.

Methods

Subjects

Subjects had participated in a longitudinal study of health among the Gila River Indian Community in Arizona, where most residents are of Pima Indian heritage. Among them, ~49% were full-heritage Pima Indians (defined as 8/8 Pima including Tohono O’odham); the rest were, on average, 6/8 American Indian (4/8 Pima and 2/8 of other tribes). Residents (age≥5 years) were invited to attend outpatient biennial exams between the years 1965–2007. Each exam included measures of height and weight for calculation of BMI and as well as a 75-g oral glucose tolerance test to determine diabetes according to the criteria of American Diabetes Association (9). Serum IGF-II concentrations were measured in stored fasting blood samples. Only samples from individuals ≥15 years of age, who were examined between 1990 and 1995 (when samples were routinely stored at −80° C) and had at least one follow-up examination were included in the current analysis. All participants were determined to be non-diabetic at the time of the blood draw (N=911; age [mean±SD]=29±12 years, BMI=32.7±7.7 kg/m2, male=39%).

Measurement of circulating IGF-II

Fasting serum IGF-II was measured using the method of Enzyme-Linked ImmunoSorbent Assay (ELISA) with the commercially available kit “IFG-II-ELISA” (Mediagnost, Reutlingen, Germany) according to manufacturer’s protocol. The antibody to IGF-II had high specificity; cross-reaction with IGF-I was not detected (up to 1000ng/ml). This technique allows separation of IGF-II from the binding proteins by acidification, thus the measured IFG-II is free in solution.

Genotyping

The 911 subjects with serum IGF-II measurements were a subset of a population-based study from the Gila River Indian Community (N = 6789) with data on BMI and diabetes status as well as genotypic data derived from a custom Axiom array (Affymetrix, Santa Clara, CA) for a genome-wide association study (GWAS). Data from this array which passed all QCs constituted 515,692 SNPs and captured 92% of all common variation (minor allele frequency ≥0.05; r2≥0.85 in 300-kb windows) across the entire Pima Indian genome (10). Common variation spanning the IGF2 locus (100 kb up- and down-stream of the gene, Chr11:2050980–2269939, hg19) was captured by 42 tag SNPs on this array. Genotypic data were also available for 64 “established” BMI-associated SNPs which were identified in prior meta-analyses of GWAS for BMI from predominately large population s of European ancestry (11).

Statistical analysis

The correlation of serum IGF-II with BMI in 911 subjects was analyzed by linear regression including age, sex and storage time as covariates. The association with weight change was assessed by analysis of weight at the examination subsequent to the one for which IGF-II was measured with sex, storage time, follow-up time, baseline age, weight and height as covariates. The association of genotypes with maximum BMI observed in the longitudinal study was analyzed in 6,789 subjects by a linear mixed model fitted with a variance components covariance structure to account for genetic relatedness among individuals. The genetic relatedness matrix was estimated as the proportion of the genome shared identical by descent (IBD) between each pair of individuals who had been genotyped (a total of 29,648,850 pairs) (12). Genomic segments shared IBD were identified with the fastIBD function of Beagle package (13) using 482,616 autosomal markers with minor allele frequency>0.05. Mixed models were fitted using the SOLAR package (14). The natural logarithm of BMI was taken as the dependent variable. Results were adjusted for age, sex, birth year and the first five genetic principal components (15) derived from the genome-wide array. Associations between genotypes and serum IGF-II levels were analyzed similarly in a linear mixed model with adjustment for age, sex, storage time and the first five genetic principal components. Heritability (the proportion of phenotypic variance potentially due to genetic factors) was calculated from the variance components of the mixed model. To correct for multiple statistical comparisons, we used the Moskvina-Schmidt method, which is based on the number of effectively independent tests accounting for the linkage disequilibrium among SNPs (16).

To assess the contribution of obesity susceptibility variants to circulating IGF-II levels, we analyzed association with a genetic risk score derived from 64 established BMI variants. For each individual, the score was calculated as the sum of the number of risk alleles at each locus weighted by the published effect size for BMI in European population, divided by the sum of the weights (11). If the genetic associations of these established variants represent causal effects, then the genetic risk score can be treated as an instrumental variable that can be used to assess the potential causal effect of an independent variable (BMI) on a dependent variable (IGF-II levels) (17). Thus, a test of whether the BMI genetic risk score associates with IGF-II levels can provide an indication of whether the association between BMI and IGF-II levels is causal, provided that the standard assumptions of “Mendelian randomization” (lack of residual population stratification, absence of pleiotropy) apply (18).

Results

Circulating IGF-II levels strongly correlate with BMI

Among the 911 Pima subjects whose serum samples had been collected between the years 1990–1995 and stored at −80°C, a mean (±SD) serum IGF-II concentration of 443.3±148.9 ng/ml (values ranged: 66.2–1005.5 ng/ml) was detected. However, a significant negative correlation was observed between serum IGF-II levels and sample storage time (r=−0.23, P=3.3×10−12); therefore, all subsequent analyses were adjusted for serum storage time. Cross-sectional analyses showed that mean fasting serum IGF-II levels did not differ by sex (438.3±157.1 ng/ml in 351 males and 446.5±143.7 ng/ml in 560 females; P=0.29, adjusted for serum storage time, Figure 1A). In contrast, serum IGF-II levels had a strong negative correlation with age, where IGF-II levels were lower in older subjects (r=−0.37, P=1.6×10−30, adjusted for sex and serum storage time, Figure 1B). After adjusting for age, sex and storage time, lower IGF-II levels correlated with higher BMI (r=−0.17, adjusted P=4.4×10−7, Figure 2A). In women the correlation between IGF-II levels and BMI was −0.19 (P=7.1×10−6), while in men it was −0.14 (P=7.5×10−3). There was no statistically significant difference in the correlation between the sexes (P=0.51 for difference). There was a statistically significant interaction with age, such that the inverse correlation between BMI and IGF-II levels was stronger at older ages (P=0.0023). When stratified by age, no correlation between IGF-II levels and BMI was observed in subjects 15–19 years (r=0.05, P=0.48, N=240, Figure 2B), a modest correlation was observed in subjects 20–29 years (r=−0.15, P=7.6×10−3, N=322, Figure 2C) and the strongest correlation was observed in subjects ≥30 years (r=−0.28, P=3.4×10−8, N=349, Figure 2D). Among these 911 subjects, only 26 had a DXA measurement to estimate percent body fat within one month of the IGF-II measurement. However, if we consider all 223 subjects who had both measurements without requiring them to be proximal in time, there was a similar negative relationship between IGF-II levels and percentage body fat (r=−0.16, P=0.02).

Figure 1:

Figure 1:

Open in a new tab

Correlation of serum IGF-II levels with sex (Panel A; Data are presented as unadjusted mean±SD. P-value was adjusted for age and storage time) or age (Panel B; Data are presented as unadjusted values. P-value was adjusted for sex and storage time).

Figure 2:

Figure 2:

Open in a new tab

Correlation of serum IGF-II levels with BMI in 911 Pima Indians (Panel A) and stratification by age (Panel B-D). Data are presented as unadjusted values. P values were adjusted for sex, storage time and the first five genetic principal components.

Circulating IGF-II levels do not predict future weight gain

The 911 subjects with measured IGF-II levels had been longitudinally studied and therefore had follow-up measurements for weight (mean follow-up time to next examination=4.1±3.0 years). However, a prospective analysis showed that circulating IGF-II levels did not predict the future weight (Beta=0.0011 kg/SD IGF-II; P=0.66, adjusted for age, sex, storage time, follow-up time, height and baseline weight). When analyses were restricted to individuals ≥20 years of age, there was also no association with weight change (Beta=0.0009 kg/SD, P=0.75).

Genetic variation in/near the IGF2 locus does not statistically significantly associate with BMI or circulating IGF-II levels

Genotypes of 42 SNPs which tag common variation across the IGF2 locus (100kb up- and down-stream from the gene) were analyzed for association with BMI in a population-based sample of 6789 Pima Indians, and association with serum IGF-II levels in the subset of individuals (N=911) with this measure (Table 1). Two tag SNPs had nominal associations (P<0.05) with maximum adult BMI, and 8 tag SNPs had nominal associations with serum IGF-II levels (P<0.05). However, none of these associations were significant after a correction for multiple comparisons (requires P value<0.0012).

Table1.

Association of 42 common tag SNPs in/near IGF-II with maximum BMI and serum IGF-II level in American Indians.

SNP Position (chr.11, hg19) Allele 1/2 Allele 1 Frequency Number of SNPs tagged Maximum BMI (n=6789) American Indian Serum IGF-II level (n=911) Full-heritage Pima
Beta (Loge) P Beta P
rs61867028 2050980 T/C 0.15 singleton 0.004 0.566 13.22 0.1493
rs77530445 2062693 T/C 0.35 singleton −0.008 0.079 −0.37 0.9571
rs11042391 2063598 A/G 0.13 8 0.006 0.376 −2.04 0.8327
rs7479065 2068717 A/G 0.26 singleton 0.001 0.810 −0.07 0.992
rs148127200 2069207 A/G 0.93 singleton 0.009 0.299 −34.92 0.0054
rs73398037 2070333 T/C 0.58 5 −0.004 0.386 −0.61 0.9275
rs74764171 2072834 T/C 0.17 3 0.006 0.311 2.28 0.7853
rs10734645 2073378 T/C 0.74 2 −0.004 0.440 3.43 0.6431
rs4930030 2087220 T/C 0.19 16 0.006 0.285 −4.72 0.5837
rs74048171 2093603 A/C 0.12 30 0.010 0.137 0.74 0.9424
chr11:2118957 2118957 T/G 0.07 singleton −0.006 0.494 −1.11 0.9283
rs146587171 2123728 A/G 0.33 4 −0.005 0.294 19.21 0.0056
rs145203788 2127329 T/C 0.22 singleton 0.003 0.528 −6.41 0.3985
rs4341514 2141603 A/G 0.09 56 −0.003 0.689 −34.54 0.0088
rs12276321 2143313 T/G 0.06 3 −0.001 0.965 −31.58 0.0208
rs734351 2156213 A/G 0.86 2 0.002 0.757 26.97 0.0051
rs386580953 2156930 A/G 0.14 3 −0.003 0.675 −0.56 0.9509
rs3213216 2158179 T/C 0.44 singleton 0.006 0.175 −7.39 0.2424
rs149483638 2161530 T/C 0.15 13 0.004 0.458 −0.67 0.9421
rs3741204 2169908 T/C 0.74 6 0.005 0.299 3.55 0.6281
rs10743144 2170773 T/C 0.29 1 −0.003 0.502 13.20 0.0641
rs6357 2188238 T/C 0.10 singleton 0.005 0.495 −20.15 0.0821
rs6356 2190951 T/C 0.62 1 0.000 0.991 −4.71 0.5057
rs10743152 2195981 T/C 0.46 17 0.002 0.684 −1.36 0.8315
rs11042978 2198418 T/G 0.40 19 −0.005 0.257 −1.26 0.8546
rs11564707 2206387 C/G 0.58 9 0.000 0.917 3.91 0.5527
rs61871274 2207877 T/C 0.28 2 −0.001 0.851 12.12 0.091
rs12418303 2209413 A/G 0.32 6 0.007 0.142 −11.78 0.0936
rs7121197 2213258 T/C 0.89 33 −0.003 0.667 11.54 0.2952
rs10840523 2213452 T/C 0.60 2 0.000 0.952 −3.89 0.5755
rs7124729 2218817 T/G 0.20 1 0.006 0.330 −18.60 0.0294
rs11043069 2220652 T/C 0.31 13 −0.002 0.707 14.30 0.0378
rs117940747 2222748 A/C 0.22 singleton 0.004 0.438 −11.94 0.1294
rs7480143 2225023 T/C 0.06 7 −0.012 0.189 −20.32 0.1459
rs140996354 2232137 T/C 0.16 3 0.005 0.356 5.33 0.5442
rs7121039 2232911 C/G 0.58 53 0.000 0.924 −12.32 0.0671
rs11606404 2241215 A/G 0.89 7 0.005 0.459 10.67 0.3322
rs10765826 2241321 A/G 0.54 8 −0.006 0.188 7.77 0.2281
rs6578246 2245345 A/G 0.56 12 0.011 0.017 −12.00 0.0623
rs4506651 2259087 A/G 0.12 10 0.004 0.598 −12.57 0.2705
rs12278109 2268695 T/C 0.05 5 −0.025 0.010 −8.94 0.5552
rs181911080 2269939 A/G 0.06 singleton 0.002 0.792 35.88 0.0062
Open in a new tab

SNPs were 100kb up- and down-stream from the IGF-II gene (Chr11:2050980–2269939, hg19). For the BMI analysis, BMI was loge-transformed to approximate a normal distribution, and the betas and P values are adjusted for age, sex, birth year and the first five genetic principal components. For serum IGF-II analyses, beta and P values are adjusted for age, sex, serum storage time, the first five genetic principal components. Results for both traits are derived from a linear mixed model that accounts for genetic relationships among individuals.

Analyses of Genetic Risk Score for BMI on IGF-II Levels

All participants in whom IGF-II levels were measured also had genotypic data available on 64 SNPs which had been previously reported to be associated with BMI in a predominately European meta-analysis for BMI. A genetic risk score for BMI was calculated from these 64 SNPs. This genetic risk score was associated with BMI in the 911 participants, but did not associate with IGF-II levels (Figure 3 A and B). However, if the analysis was restricted to those individuals ≥20 years of age (N=671), the association between genetic risk score and BMI remained significant (β=0.026 log[kg/m2] per SD, P=0.0032, Figure 3C) and there was a significant association with IGF-II levels such that those participants with greater numbers of obesity susceptibility alleles had lower IGF-II levels (β=−0.081 SD of IGF-II per SD of genetic risk score, P=0.025, Figure 3D).

Figure 3:

Figure 3:

Open in a new tab

Analysis of a genetic risk score (GRS) for BMI, derived from 64 established obesity variants in Europeans, in our sample of American Indians. Categories of BMI risk score are plotted at their mean value. Panel A: Association of the GRS with BMI in 911 American Indians. Beta is the logarithm of BMI per SD of genetic risk score. Numbers of individuals in each GRS category: 40–48, n=98; 48–50, n=106; 50–52, n=123; 52–56, n=234; 56–58.5 n=153; 58.5–61, n=107; 61–71, n=88. Panel B. Association of the GRS with serum IGF-II levels in 911 American Indians. Beta is given in SD of IGF-II per SD of BMI risk score. Panel C. Association of the GRS with BMI in American Indians ≥20 years of age (n=671). Numbers of individuals in each GRS category are: 40–48, n=79; 48–50, n=79; 50–52, n=96; 52–56; n=165; 56–58.5, n=112; 58.5–61, n=77; 61–71, n=62. Panel D. Association of the GRS with serum IGF-II levels in American Indians ≥20 years of age (n=671). All data are presented as unadjusted values. Betas and p-values are adjusted for age, sex, storage time and the first five genetic principal components.

Discussion

In a cross-sectional analysis of Pima Indians, circulating IGF-II levels associated with age, where IGF-II declined with increasing age. This observation is consistent with a prior report that circulating IGF-II concentrations in humans peak at puberty and then decline throughout adulthood; whereas systemic levels of IGF-II in rodents decline soon after birth (19). Independent of the age-related decrease in IGF-II, relatively low circulating IGF-II levels correlated with higher BMIs in Pima Indians. There was also evidence to indicate that this correlation of IGF-II with BMI interacts with age, where this negative relationship was significant in older ages (>20 years), but not in younger ages (15–20 years).

Population studies assessing cross-sectional associations between circulating IGF-II concentrations and BMI or obesity are few and contradictory. Some studies have reported higher IGF-II concentrations in people with obesity as compared with lean controls (7). In contrast, a cross-sectional study showed that subjects with obesity had statistically significantly lower mean IGF-II levels compared with lean individuals (6) and another study similarly reported that low circulating IGF-II concentrations predict weight gain and obesity (20). Yet another study found no association between IGF-II levels and BMI, and weak negative associations with measures of waist-to-hip ratio (21). In comparison to these prior reports, our study had the largest number of subjects (N=911 with measures of serum IGF-II) and we observed a negative correlation between circulating IGF-II levels and BMI. Of note, due to our larger sample size, we could also analyze this relationship among different age groups and observed a significant age interaction, with the inverse correlation being stronger at older ages, although the underlying physiology of this interaction remains unclear.

In men, a polymorphism (rs680) in the 3’UTR IGF-II was reported to associate with both circulating IGF-II levels and body weight (2). However, rs680 did not associate with IGF-II and BMI in Pima Indians. Our study identified a different tag SNP (rs6578246) where the allele (G) that nominally associated with lower serum IGF-II levels also associated higher BMI, but both associations were too weak (adjusted P= 0.062 and 0.017, respectively) to survive adjustment for multiple testing. It is noteworthy that no cis-eQTL has been reported in public databases for IGF-II in blood.

Various studies have reported diverse mechanisms which support that IGF-II levels affect BMI. For example, transgenic mouse studies have shown that IGF-II levels affect fat metabolism. Mice overexpressing IGF-II have reduced fat mass and a reduction of lipid content in adipose tissue (4). Oxidation of dietary lipids is increased in IGF-II overexpressed mice, indicating that IGF-II may influence the metabolic utilization of ingested lipids (4). It has also been reported that relatively higher IGF-I and IGF-II levels may increase muscle mass and/or the respiratory quotient (ratio of fat to carbohydrate oxidation) which would impact total and macronutrient energy balance (22, 23). Alternatively, it has also been suggested that IGF-II levels impact body weight via a central-acting role of IGF-II on regulation of feeding behavior. In both humans and rodents, the ligands IGF-I, IGF-II, and insulin as well as their receptors are expressed in hypothalamic regions implicated in regulation of food intake (24). Like insulin, studies have shown that intracerebroventricular injections of IGF-II induce hypophagia and weight loss in rodents (25–26). IGF-II attenuates the release of neuropeptide Y, a potent orexigenic peptide, from the hypothalamic paraventricular nucleus in vitro (27). These central IGF-II actions may be mediated through the insulin receptor isoform A, which is the only insulin receptor isoform expressed in central nervous tissue and has high affinity for IGF-II (28).

In our cross-sectional study, we found a negative relationship between IGF-II and BMI. However, it is not clear whether lower circulating IGF-II levels led to an increase in BMI, or whether higher BMI/adiposity led to a reduction in circulating IGF-II level. In our analyses, baseline IGF-II levels were not predictive of future weight change. Moreover, in participants ≥ 20 years of age, an obesity genetic risk score was significantly associated with lower IGF-II levels, consistent with the hypothesis that genetically driven high BMI is associated with low IGF-II levels. Therefore, our data provided some support for the notion that high BMI leads to low IGF-II levels in older American Indians, but more complex relationships involving genetic pleiotropy cannot be excluded. Larger studies with longitudinal data may help resolve these issues.

In summary, our study of 911 subjects showed that low levels of circulating IGF-II associated with higher BMI, where the correlation was stronger at older ages. However, future studies in humans are required to demonstrate whether low IGF-II levels affect BMI, as had been shown in rodent models, or whether increased adiposity affects circulating IGF-II levels.

Answer the Study Important Questions.

What is already known about this subject?

  1. The hormone insulin-like growth factor-II (IGF-II) plays an important role in the regulation of metabolism and growth.

  2. In mice, overexpression of IGF-II leads to a reduction in body weight and fat mass; however, in humans both positive and negative relationships have been reported between serum IGF-II levels and obesity.

What does your study add?

  1. In our human study, serum IGF-II levels were negatively correlated with BMI, where the correlation was stronger at older ages. In comparison to the prior reports, our study in Pima Indians had the largest number of subjects (N=911 with measures of serum IGF-II).

  2. Among individuals who had genotypes for 64 established obesity variants and were ≥20 years (N= 671), a genetic risk score for high BMI associated with lower IGF-II levels. This observed correlation between serum IGF-II and BMI may be due to metabolic consequences of obesity on IGF-II levels.

How might your results change the direction of research or the focus of clinical practice?

Our study may provide new insight into the prevention and treatment of obesity.

Acknowledgements

We thank the clinical staff of the Phoenix Epidemiology and Clinical Research Branch for conducting the study. We also thank all study participants. This study utilized the high-performance computational capabilities of the Biowulf Linux cluster at the National Institutes of Health.

Funding: This work was supported by the intramural research programme of the National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health.

Footnotes

Disclosure Statement: The authors have nothing to disclose.

References

  • 1.Jones JI, Clemmons DR. Insulin-like growth factors and their binding proteins: biological actions. Endocr Rev. 1995;16:3–34. [DOI] [PubMed] [Google Scholar]
  • 2.O’Dell SD, Miller GJ, Cooper JA, et al. Apal polymorphism in insulin-like growth factor II (IGF2) gene and weight in middle-aged males. Int J Obes Relat Metab Disord. 1997;21:822–825. [DOI] [PubMed] [Google Scholar]
  • 3.Gaunt TR, Cooper JA, Miller GJ, Day IN, O’Dell SD. Positive associations between single nucleotide polymorphisms in the IGF2 gene region and body mass index in adult males. Hum Mol Genet. 2001;10:1491–1501. [DOI] [PubMed] [Google Scholar]
  • 4.Da Costa TH, Williamson DH, Ward A, et al. High plasma insulin-like growth factor-II and low lipid content in transgenic mice: measurements of lipid metabolism. J Endocrinol. 1994;143:433–439. [DOI] [PubMed] [Google Scholar]
  • 5.Jones BK, Levorse J, Tilghman SM. Deletion of a nuclease-sensitive region between the Igf2 and H19 genes leads to Igf2 misregulation and increased adiposity. Hum Mol Genet. 2001;10:807–814. [DOI] [PubMed] [Google Scholar]
  • 6.Chang S, Wu X, Yu H, Spitz MR. Plasma concentrations of insulin-like growth factors among healthy adult men and postmenopausal women:associations with body composition, lifestyle, and reproductive factors. Cancer Epidemiol Biomarkers Prev. 2002;11:758–766. [PubMed] [Google Scholar]
  • 7.Frystyk J, Vestbo E, Skjaerbaek C, Mogensen CE, Orskov H. Free insulin-like growth factors in human obesity. Metabolism. 1995;44:37–44. [DOI] [PubMed] [Google Scholar]
  • 8.Knowler WC, Pettitt DJ, Saad MF, Bennett PH. Diabetes mellitus in the Pima Indians: incidence, risk factors and pathogenesis. Diabetes Metab Rev. 1990;6:1–27. [DOI] [PubMed] [Google Scholar]
  • 9.The Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 1997;20:1183–1197. [DOI] [PubMed] [Google Scholar]
  • 10.Muller YL, Piaggi P, Chen P, et al. Assessing Variation across Eight Established East Asian Loci for Type 2 Diabetes in American Indians: Suggestive Evidence for New Sex-specific Diabetes Signals in GLIS3 and ZFAND3. Diabetes Metab Res Rev. 2017;33:doi: 10.1002/dmrr.2869. [DOI] [PubMed] [Google Scholar]
  • 11.Muller YL, Hanson RL, Piaggi P, et al. Assessing the Role of 98 Established Loci for BMI in American Indians. Obesity 2019;27(5):845–854. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Hsueh WC, Nair AK, Kobes S, et al. Identity-by-Descent Mapping Identifies Major Locus for Serum Triglycerides in Amerindians Largely Explained by an APOC3 Founder Mutation. Circ Cardiovasc Genet. 2017;10:doi: 10.1161/CIRCGENETICS.117.001809. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Browning BL, Browning SR. A fast, powerful method for detecting identity by descent. Am J Hum Genet. 2011;88:173–182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Almasy L, Blangero J. Multipoint quantitative-trait linkage analysis in general pedigrees. Am J Hum Genet. 1998;62:1198–1211. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Piaggi P, Masindova I, Muller YL, et al. A Genome-Wide Association Study Using a Custom Genotyping Array Identifies Variants in GPR158 Associated with Reduced Energy Expenditure in American Indians. Diabetes 2017;66(8):2284–2295. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Moskvina V, Schmidt KM. On multiple-testing correction in genome-wide association studies. Genet Epidemiol. 2008;32(6):567–73. [DOI] [PubMed] [Google Scholar]
  • 17.Burgess S, Thompson SG. Use of allele scores as instrumental variables in Mendelian randomization. Int J Epidemiol. 2013;42:1134–1144. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.VanderWeele TJ, Tchetgen EJT, Cornelius M, Kraft P. Methodological challenges in Mendelian randomization. Epidemiology. 2014;25:427–435. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Holly JM: The IGF-II enigma. Growth Horm IGF Res. 1998;8:183–184. [DOI] [PubMed] [Google Scholar]
  • 20.Sandhu Manjinder S., Gibson J. Martin, et al. Low Circulating IGF-II Concentrations Predict Weight Gain and Obesity in Humans. Diabetes 2003;52(6):1403–1408. [DOI] [PubMed] [Google Scholar]
  • 21.Cruickshank JK, Heald AH, Anderson S, et al. Epidemiology of the insulin-like growth factor system in three ethnic groups. Am J Epidemiol. 2001;154:504–513. [DOI] [PubMed] [Google Scholar]
  • 22.Sun G, Ukkola O, Rankinen T, Joanisse DR, Bouchard C. Skeletal muscle characteristics predict body fat gain in response to overfeeding in never-obese young men. Metabolism 2002;51:451–456. [DOI] [PubMed] [Google Scholar]
  • 23.Musaro A, McCullagh K, Paul A, et al. Localized Igf-1 transgene expression sustains hypertrophy and regeneration in senescent skeletal muscle. Nat Genet. 2001;27:195–200. [DOI] [PubMed] [Google Scholar]
  • 24.Baskin DG, Wilcox BJ, Figlewicz DP, Dorsa DM. Insulin and insulin-like growth factors in the CNS. Trends Neurosci. 1988;11:107–111. [DOI] [PubMed] [Google Scholar]
  • 25.Harel Z, Tannenbaum GS. Centrally administered insulin-like growth factor II fails to alter pulsatile growth hormone secretion or food intake. Neuroendocrinology 1992;56:161–168. [DOI] [PubMed] [Google Scholar]
  • 26.Lauterio TJ, Marson L, Daughaday WH, Baile CA. Evidence for the role of insulin-like growth factor II (IGF-II) in the control of food intake. Physiol Behav. 1987;40:755–758. [DOI] [PubMed] [Google Scholar]
  • 27.Sahu A, Dube MG, Phelps CP, Sninsky CA, Kalra PS, Kalra SP. Insulin and insulin-like growth factor II suppress neuropeptide Y release from the nerve terminals in the paraventricular nucleus: a putative hypothalamic site for energy homeostasis. Endocrinology. 1995;136:5718–5724. [DOI] [PubMed] [Google Scholar]
  • 28.Moller DE, Yokota A, Caro JF, Flier JS. Tissue-specific expression of two alternatively spliced insulin receptor mRNAs in man. Mol Endocrinol. 1989;3:1263–1269. [DOI] [PubMed] [Google Scholar]

RESOURCES