Open Access

Linkage and association analysis of circulating vitamin D and parathyroid hormone identifies novel loci in Alaska Native Yup’ik people

  • Stella Aslibekyan1,
  • Laura K. Vaughan2, 4,
  • Howard W. Wiener1,
  • Bertha A. Hidalgo1,
  • Dominick J. Lemas3,
  • Diane M. O’Brien5,
  • Scarlett E. Hopkins5,
  • Kimber L. Stanhope6,
  • Peter J. Havel6,
  • Kenneth E. Thummel7,
  • Bert B. Boyer5 and
  • Hemant K. Tiwari2Email author
Genes & NutritionStudying the relationship between genetics and nutrition in the improvement of human health201611:23

https://doi.org/10.1186/s12263-016-0538-y

Received: 17 March 2016

Accepted: 18 July 2016

Published: 2 August 2016

Abstract

Background

Vitamin D deficiency is a well-documented public health issue with both genetic and environmental determinants. Populations living at far northern latitudes are vulnerable to vitamin D deficiency and its health sequelae, although consumption of traditional native dietary pattern rich in fish and marine mammals may buffer the effects of reduced sunlight exposure. To date, few studies have investigated the genetics of vitamin D metabolism in circumpolar populations or considered genediet interactions with fish and n-3 fatty acid intake.

Methods

We searched for genomic regions exhibiting linkage and association with circulating levels of vitamin D and parathyroid hormone (PTH) in 982 Yup’ik individuals from the Center for Alaska Native Health Research Study. We also investigated potential interactions between genetic variants and a biomarker of traditional dietary intake, the δ15N value.

Results

We identified several novel regions linked with circulating vitamin D and PTH as well as replicated a previous linkage finding on 2p16.2 for vitamin D. Bioinformatic analysis revealed multiple candidate genes for both PTH and vitamin D, including CUBN, MGAT3, and NFKBIA. Targeted association analysis identified NEBL as a candidate gene for vitamin D and FNDC3B for PTH. We observed significant associations between a variant in MXD1 and vitamin D only when an interaction with the δ15N value was included. Finally, we integrated pathway level information to illustrate the biological validity of the proposed candidate genes.

Conclusion

We provide evidence of linkage between several biologically plausible genomic regions and vitamin D metabolism in a circumpolar population. Additionally, these findings suggest that a traditional dietary pattern may modulate genetic effects on circulating vitamin D.

Keywords

Alaska Native Vitamin D Linkage n-3 fatty acids Parathyroid hormone

Background

Vitamin D, a steroid hormone synthesized in the skin or acquired through diet, and parathyroid hormone (PTH), secreted by parathyroid glands, work synergistically to promote absorption of dietary calcium and phosphates (reviewed by [31]). In addition, vitamin D affects cell cycle, proliferation, differentiation, and apoptosis through vitamin D receptor response elements found on several hundred genes (reviewed by [40]). As a consequence of its physiological importance, a large body of evidence links circulating vitamin D to numerous health outcomes, including skeletal health, cardiovascular disease, type 1 diabetes, type 2 diabetes, metabolic syndrome, autoimmune diseases, multiple sclerosis, and some types of cancer ([22], reviewed by [15, 16]).

The primary form of circulating vitamin D, 25-hydroxy-vitamin D (25(OH)D), is converted to the metabolically active form 1,25-dihydroxy-vitamin D (1,25(OH)2D). The conversion is regulated by PTH as part of a feedback system maintaining calcium and phosphate homeostasis (reviewed by [13, 17, 31]). Concentrations of 25(OH)D in the blood, which aggregately reflect endogenous generation through UVB exposure (D3), exogenous dietary intake (D3 from animal sources, D2 from plant sources), and supplementation, are considered the best indicator of vitamin D status (reviewed by [50]). In addition to the environmental inputs of sunlight and diet, high inter-individual variation in circulating 25(OH)D as well as family studies suggest a role for genetic determinants [1, 14]. Large-scale genome-wide association studies (GWAS) of 25(OH)D have confirmed associations with polymorphisms near cholesterol synthesis, hydroxylation, and vitamin D transport genes in individuals of European descent [1, 46]. Despite the documented importance of ancestry to vitamin D status [42, 46], genetic risk factors for other populations are less established.

Of particular salience is the study of vitamin D in Arctic populations, where exposure to sunlight is greatly reduced for at least half of the year (reviewed in [15]). However, this deficit may be corrected by traditional subsistence diets that are rich in vitamin D due to high intake of fish such as salmon or halibut and marine mammals. Furthermore, there is evidence of physiological adaptations leading to more efficient calcium absorption. For example, in Greenland Inuit people, vitamin D is produced at a lower rate than in Europeans, but the rate of PTH-mediated conversion to 1,25(OH)2D is higher [37]. Another study documented normal serum calcium despite low 25(OH)D and dietary calcium in Canadian Inuit children, postulating that evolutionary pressures have selected for vitamin D receptors that bind more strongly to the vitamin D molecule [10, 41]. For these reasons, investigating genetic determinants of vitamin D and PTH in Arctic populations, as well as gene-diet interactions with the intake of vitamin D-rich fatty fish, is likely to yield unique insights.

To that end, we performed the first genome-wide linkage analysis of both 25(OH)D and PTH in a population of Yup’ik people participating in the Center for Alaska Native Health Research (CANHR) study. We followed up our findings with targeted association and gene-diet interaction analyses and used complementary bioinformatic tools to identify putative genetic contributors to vitamin D and PTH homeostasis in this community.

Methods

The CANHR population and study sample

The CANHR studies genetic, behavioral, and nutritional risk factors for obesity and related cardiometabolic diseases among Yup’ik people in a community-based setting [29]. Recruitment of Yup’ik families was initiated in 2003 and continues in 11 Southwest Alaskan communities, where all residents are invited to participate, resulting in a convenience sample. The present study sample was comprised of 982 non-pregnant Yup’ik individuals (age range, 131–94) and were informative for genetic linkage analysis; of those, 926 passed quality control checks and had either 25(OH)D or PTH measurements or both. All participants signed informed consent documents, and the study protocols were approved by the Institutional Review Boards of the University of Alaska and the National and Alaska Area Indian Health Service Institutional Review Boards, as well as the Yukon Kuskokwim Human Studies Committee.

Laboratory measurements

Blood samples were collected from participants after an overnight fast and were immediately processed in the field, stored at −15 °C for up to a week, and then stored at −80 °C at the University of Alaska Fairbanks [5]. Total 25(OH)D was ascertained through a quantitative chemiluminescent immunoassay (ARUP Laboratories, Salt Lake City, UT). This assay accurately quantifies the sum of 25(OH)D3 and 25(OH)D2. The intra- and inter-assay coefficients of variation (CVs) were 5.5 and 12.7 %, respectively. ARUP participates in the College of American Pathologists Laboratory Accreditation Program and has CLIA (Clinical Laboratory Improvement Amendments) certification through the Centers of Medicare and Medicaid Services. PTH levels were assayed at the University of California at Davis using the DSL-8000 ACTIVE® Intact PTH IRMA Kit (Diagnostic Systems Laboratories, Inc., Webster, TX). The procedure employs a two-site immuno-radiometric assay principle [28]. The intra- and inter-assay CVs were 4.2 and 10.3 %, respectively. Traditional dietary intake was assessed using red blood cell (RBC) nitrogen stable isotope ratios (15N/14N) as previously described [34] and validated [30]. By convention, isotope ratios were expressed in relative abundance using delta values, as follows: δ15N = (15N/14Nsample − 15N/14Nstd)/15N/14Nstd × 1000 ‰, in which the standard is atmospheric nitrogen (15N/14N = 0.0036765). Analytical precision as assessed using reference materials was within 0.2 ‰.

Genotyping and quality control

Detailed descriptions of genotyping procedures, pedigree analyses, and data cleaning have previously been published [3]. Briefly, genotyping at 6090 loci was performed at the Center for Inherited Disease Research at Johns Hopkins University using the Illumina Linkage-IV panel (Illumina, San Diego, CA, USA), spanning the entire genome with an average genetic distance of 0.58 cM. Due to residual kurtosis following trait transformation, all linkage analyses involving vitamin D levels used the t-distribution option in Sequential Oligogenic Linkage Analysis Routines (SOLAR) to control type I error rate [2].

The genotypic data was subject to several quality control measures, specifically the removal of single nucleotide polymorphisms (SNPs) exhibiting inconsistencies with Mendelian segregation or Hardy-Weinberg equilibrium, prior to whole-genome linkage analysis as previously described [3]. Variants that passed quality checks were used to ascertain a cryptic population substructure, deriving “community group” as a dichotomous covariate indicating proximity to the coast.

Linkage and association analysis

We estimated the heritability and conducted a whole-genome linkage scan using the variance component approach implemented in the SOLAR program as previously described [3]. All models run in SOLAR included age (quartiles), sex (dichotomous), community group (dichotomous, inland vs. coastal), and δ15N value quartiles.

We tested all SNPs under the linkage peaks with a LOD score greater than 1.5 for association with the trait using ASSOC from the S.A.G.E. package, which models the full correlation structure of the pedigree [7, 11]. All association models included the same non-genetic covariates that were used for the linkage analysis: sex, age, community, and δ15N value quartiles. A sensitivity analysis additionally adjusted for the season of blood sample collection as a categorical variable. For each SNP, the following two models were run: one containing non-genetic covariates and the SNP genotype as an additive effect and one containing non-genetic covariates, SNP genotype as an additive effect, and the interaction between SNP effect and δ15N value quartiles. Likelihood ratio tests were used to compare the models. Corrections for multiple testing were ascertained separately for each linkage peak of interest, calculating the effective number of tested SNPs (i.e., independent SNPs not in linkage disequilibrium (LD)) using spectral decomposition of the correlation matrix [23, 32]. Bonferroni corrections for the effective number of SNPs in each linkage peak were used to determine the number of full model results that met the significance threshold (Additional file 1: Table S1). Full models were defined to include both the additive genotype term and the interactions of that term with n-3 polyunsaturated fatty acids (PUFA) intake.

Bioinformatic analysis

In order to limit the list of potential genes from regions identified by linkage analysis (LOD score >1.5), we used the SIMWALK2 [43] to phase the genotype data. This analysis indicates the positions of recombination events in the haplotypes of the analyzed individuals. We determined the haplotypes for each individual that had no recombination events and included the position of peak evidence for linkage with the trait of interest. The regions thus defined were examined for the full sample, as well as the subsets limited to both the highest and lowest quartile of the trait of interest. Regions in which less than 5 % of the examined sample experienced a recombination event were included in subsequent bioinformatic analysis. This region was very narrow for chromosome 3, so the threshold was relaxed for that region. Within the regions, potential candidate genes were identified using three complementary algorithms: Ingenuity (Qiagen, Redwood City, CA), TOPPGene Suite [6], and ENDEAVOUR [44]. For all three algorithms, training gene lists were based on the keyword “PTH” or “Vitamin D” in the HuGE navigator [49]. Genes were selected if they attained a HuGE score of at least 0.10 and were related to vitamin D and PTH only. Gene names were converted to Ensembl using BioMart Central ID converter as necessary. The settings for all three algorithms were identical to those in a previous analysis by our group [45]. Genes were considered to be putative candidates if the P values for both TOPPGene and ENDEAVOUR were ≤0.005 and if they were shown by Ingenuity to interact with the chosen training genes.

Results

Descriptive characteristics of the Yup’ik study participants are presented in Table 1. All of the P values in this table are based on simple two-sample t tests, without taking the pedigree correlation into account. RBC δ15N values and circulating PTH were both significantly higher among women, while 25(OH)D levels did not vary by gender. The correlation between serum 25(OH)D and PTH levels was estimated at −0.09. Transformations of both traits reduced residual kurtosis from 9.96 to 0.45 for PTH and from 4.00 to 1.10 for 25(OH)D. Heritability was estimated at 0.43 ± 0.07 for log-transformed PTH (P value = 1.2 × 10−10) and at 0.54 ± 0.07 for Box-Cox transformed 25(OH)D (P value = 2.3 × 10−18).
Table 1

Descriptive statistics of the study sample (n = 926)

 

Male

Female

P value

N a

419 (45)

507 (55)

0.004

Coastal community groupa

184 (50)

186 (50)

0.92

Inland community groupa

235 (42)

321 (58)

0.0003

Age, yearsb

36.56 (17.56)

38.07 (17.61)

0.17

δ15N, ‰b

8.80 (1.51)

9.13 (1.54)

0.0001

Parathyroid hormone, pg/mLc

52.7 (24.21–114.84)

65.60 (27.83–154.61)

<0.0001

25(OH)D, ng/mLc

34.57 (11.47–83.32)

33.96 (12.21–77.69)

0.56

The overall study sample consists of 926 individuals after QC; however, two were missing values for 25(OH)D and two different individuals were missing values for parathyroid hormone

a N = sample size (%), P value based on binomial distribution

bMean (SD), P value based on two-sample t test

cMean (95 % CI), P value based on two-sample t test

Results for the genome-wide linkage scan are presented in Table 2 and Figs. 1 and 2. Three regions on chromosomes 2, 10, and 22 yielded a LOD score of at least 2 for 25(OH)D. The linkage peak on chromosome 2 exhibited the strongest signal, with the LOD score exceeding 4 for 25(OH)D. Also, three regions on chromosomes 3, 14, and 17 were linked with PTH with LOD scores greater than 2.
Table 2

Linkage peaks with a LOD score >2

Chromosome bands

Peak LOD score

Start SNP

End SNP

Genomic region

25(OH)D

    

 2p16.2–2p12

4.25

rs1483869

rs741418

52,906,147–75,363,186

 10p13–10p12.1

2.52

rs873849

rs638324

14,509,807–27,730,261

 22q13.1–22q13.31

2.17

rs138383

rs929090

38,799,861–44,313,849

Parathyroid hormone

    

 3q25.32–3q26.32

2.48

rs6799097

rs6443567

158,811,068–178,427,716

 14q12–14q22.1

2.93

rs7149108

rs735265

32,962,300–52,120,264

 17p13.1–17p11.2

2.15

rs9217

rs1037037

7,363,088–18,765,021

Fig. 1

Linkage results for vitamin D. Whole-genome linkage scan for circulating vitamin D in Yup’ik people (n = 924). The X-axis shows the chromosomal location, and the Y-axis displays the LOD score

Fig. 2

Linkage results for parathyroid hormone. Whole-genome linkage scan for parathyroid hormone in Yup’ik people (n = 924). The X-axis shows the chromosomal location, and the Y-axis displays the LOD score

After adjusting for multiple testing, two variants on chromosome 2 and one variant on chromosome 10 were significantly associated with 25(OH)D at adjusted α = 0.001 and 0.002, respectively, while only one variant on chromosome 3 was significantly associated with PTH at adjusted α = 0.001. However, the two variants on chromosome 2 were in perfect LD. Significant SNPs and the associated genes are presented in Table 3. Significant interactions of variants on chromosome 2 with n-3 PUFA intake were observed for 25(OH)D, and there was evidence suggestive of similar interactions between a variant on chromosome 3 and circulating PUFA in the PTH models (Additional files 2 and 3: Figures S1 and S2). Additional adjustment for season did not appreciably change the observed genetic associations (data not shown).
Table 3

Significant associations under the linkage peaks

SNP

Chr

Gene

P (additive)a

P (full)b

χ 2 (LRT)c

P (LRT)c

25(OH)D

      

 rs10205487/rs897119d

2

MXD1

0.22

0.0006

18.15

0.0004

 rs10828183

10

NEBL

0.0005

0.004

3.18

0.37

Parathyroid hormone

      

 rs1353894

3

FNDC3B

0.03

0.0008

14.21

0.003

a P value for the model including age, sex, community group, δ15N quartiles, and additive genotype

b P value for the model including all of the above plus the interaction term between the additive genotype and δ15N quartiles

cChi-square statistic and P value for the likelihood ratio test comparing the additive model with the full model

drs10205487 and rs897119 were found to be in perfect linkage disequilibrium, with identical estimates of association signal

Results of the bioinformatic analysis are summarized in Table 4. None of the a priori identified training genes were located under the linkage peaks. All peaks except for the one on chromosome 2 contained candidate genes proposed by all three bioinformatic algorithms.
Table 4

Potential candidate genes identified by bioinformatic analysis of genes within linkage peaks for 25(OH)D and parathyroid hormone

Chromosome bands

Putative candidate genes

25(OH)D

 

 10p13–10p12.1

CUBN, RSU1

 22q13.1–22q13.31

APOBEC3F, APOBEC3G, ATF4, CACNA1I, CBX7, CBY1, DDX17, DMC1, GTPBP1, MGAT3, PDGFB, RPL3

Parathyroid hormone

 

 3q25.32–3q26.32

NLGN1

 14q12–14q22.1

CFL2, FOXA1, MYH1, NFKBIA, NKX2-8, PAX9, PSMA6, SFTA3, SNX6

 17p13.1–17p11.2

GAS7, MIPOL1, MYH2, MYH4, RCVRN

Discussion

The CANHR Yup’ik study population is ideal for investigation of the genetic determinants of circulating vitamin D and potential gene-diet interactions with n-3 PUFA. Despite low levels of sunlight exposure, mean 25(OH)D levels in our study sample were well within the optimal range currently recommended for the general population [4]. This paradox may be explained by both genetic adaptations as well as the high vitamin D content of the traditional diet of the Yup’ik people, which may buffer the effects of insufficient sunlight at higher latitudes (reviewed by [15, 17]). Indeed, a previous analysis of a CANHR subsample demonstrated that intake of locally harvested foods, namely fatty fish, fish roe, fish liver, and wild game, is correlated with higher levels of 25(OH)D [24]. Yet, another study showed that 25(OH)D mediates PTH levels in our study population [25]. Using genotype data from extended pedigrees, biochemical measurements of 25(OH)D, PTH, and the δ15N value (a marker of traditional diet rich in vitamin D), and established biological pathways, this report further illustrates the complexity of vitamin D metabolism in the Yup’ik people.

Three linkage peaks were identified for 25(OH)D and three for PTH. The peaks for the two traits did not overlap, reflecting distinct genetic contributions underlying the weakly correlated 25(OH)D and PTH levels in our study population as well as others [21, 39]. All were novel findings except for the locus on chromosome 2, which was initially reported as linked to 25(OH)D3 in Northern European families with asthma [48] and associated with bone mineral density in a large meta-analysis [8]. Few genome-wide linkage studies on 25(OH)D and none on PTH are available to use as a comparison. Interestingly, none of our observed linkage regions contained genes previously reported in genome-wide association studies (GWAS) of 25(OH)D or 1,25(OH)2D, e.g., GC on chromosome 4 [1] or CYP2R1 and DHCR7 on chromosome 11 [46]. Similarly, the previous GWAS association of loci in the PTH gene with circulating PTH was not replicated [27]. The reasons for non-replication may include our genotyping approach, which utilized a linkage-specific panel that was not designed to include all physiologically relevant loci, the uniqueness of our population, and chance. The remaining linkage peaks at 10p13–p12.1 and 22q13.1–13.31 for 25(OH)D and 3q25.32–26.32, 14.12–22.1, and 17p13.1–11.2 for PTH are novel and contain several biologically plausible loci.

Using complementary methods, bioinformatic analysis identified several strong candidate genes located under observed linkage peaks. Of those, the most notable is CUBN (cubilin), located on chromosome 10. The product of CUBN, cubilin, facilitates megalin-mediated delivery of 25(OH)D3 to kidney epithelial cells, impacting circulating levels of the hormone [33]. In both human and animals, mutations causing cubulin dysfunction are associated with abnormal vitamin D metabolism (e.g., urinary excretion of 25(OH)D3 in humans) [33]. Another biologically interesting finding from the bioinformatic analysis is MGAT3 (beta-1,4-mannosyl-glycoprotein 4-beta-N-acetylglucosaminyltransferase), located on chromosome 22. MGAT3 encodes an enzyme implicated in protein glycosylation and the immune response [20]. Preliminary data illustrate MGAT3 mRNA as a potential biomarker of response to vitamin D therapy in Alzheimer’s disease, hinting at its translational potential; however, these findings remain to be independently validated [9]. Isoforms of two other potential candidate genes, CACNA1I (calcium channel, voltage-dependent, T type, alpha 1i subunit) and PDGFB (plate-derived growth factor beta polypeptide), have been shown to be regulated by the vitamin D receptor, respectively affecting calcium homeostasis [12] and mitosis [36].

Several strong candidate genes for PTH were also identified based on their involvement in relevant physiological pathways. NFKBIA (nuclear factor kappa light polypeptide gene enhancer in B cells inhibitor, alpha) is a member of the NFkB inhibitor family that binds with the product of REL (v-rel reticuloendotheliosis viral oncogene homolog, regulated by the vitamin D receptor) to inhibit Rel/NFkB complexes. In a previous transcriptome-wide study, the expression of NFKBIA in white blood cells was shown to be decreased by vitamin D supplementation, potentially affecting downstream immune response [18]. Reflecting the pleiotropic functions of the vitamin D/PTH axis, other potential candidate genes were also implicated immune/inflammatory processes (NKX2, PSMA6, SNX6) as well as cancer (NKX2, NKX3, PAX9).

Extended pedigrees such as those present in the CANHR study provide a powerful framework for targeted association following whole-genome linkage [47]. Targeted association analysis of the observed linkage peaks identified two SNPs in MXD1 (MAX dimerization protein 1) and one in NEBL (nebulette) that were significantly associated with 25(OH)D levels and one in FNDC3B (fibronectin type III domain containing 3B) associated with PTH levels. Both MXD1 and FNDC3B are involved in the VDR/RXR activation pathway, the former being regulated by the vitamin D/VDR/RXR complex [38] and the latter suppressing the expression of RUNX2 [19], which in turn binds to and activates vitamin D/VDR/RXR [35]. On the other hand, NEBL has been implicated in calcium homeostasis [26], although a direct link to vitamin D metabolism has not been investigated.

This study presents novel evidence of linkage between several biologically plausible genomic regions and 25(OH)D/PTH in a study population of Yup’ik people with seasonally low exposure to sunlight and high intake of vitamin D-rich foods. These findings are particularly informative in the context of the vitamin D/VDR/RXR pathway, which illustrates the involvement of candidate loci in maintaining the vitamin D/PTH/calcium homeostasis. To our knowledge, this is the first whole-genome linkage study of PTH overall and the first such study of both phenotypes in a circumpolar population. In addition to replicating a previously reported linkage region for 25(OH)D and identifying novel loci for both 25(OH)D and PTH, we present novel evidence of gene-diet interactions with n-3 PUFA intake. Specifically, a variant in MXD1 was significantly associated with 25(OH)D only after accounting for its interaction with δ15N, a marker of n-3 PUFA intake and traditional lifestyle in Yup’ik people. To the best of our knowledge, no other circumpolar cohorts currently have comparable genotypic and phenotypic measurements, limiting our opportunities for external replication. Future studies in high-latitude populations with unique vitamin D dietary sources and requirements are warranted to establish validity of our findings, laying the groundwork for disease-prevention efforts in Arctic communities.

Conclusions

We provide evidence of linkage between several biologically plausible genomic regions, located on chromosomes 2, 3, 10, 14, 17, and 22, and vitamin D metabolism in a circumpolar population. Additionally, these findings suggest that the traditional dietary pattern of the Yup’ik people, characterized by high intake of n-3 PUFA, may modulate genetic effects on circulating 25(OH)D.

Abbreviations

1,25(OH)2D, 1,25-dihydroxy-vitamin D; 25(OH)D, 25-hydroxy-vitamin D; CANHR, Center for Alaska Native Health Research; CV, coefficient of variation; HWE, Hardy-Weinberg equilibrium; PTH, parathyroid hormone; PUFA, polyunsaturated fatty acids; SNP, single nucleotide polymorphism; SOLAR, Sequential Oligogenic Linkage Analysis Routines

Footnotes
1

There was only one 13 year old individual, who turned 14 within two weeks of enrollment.

 

Declarations

Acknowledgements

We thank the community field research assistants for helping with the study recruitment and data collection and all study participants and their communities for welcoming and teaching our research team so much about the Yup’ik way of life.

Funding

Genotyping services were provided by the Center for Inherited Disease Research (CIDR). CIDR is fully funded through a federal contract from the National Institutes of Health to The Johns Hopkins University, contract HHSN268200782096C. Other funding sources included the National Institutes of Health grants K01DK080188 (LKV), DK074842 (BBB), and RR016430 (BBB). The National Institutes of Health have no influence over the research process, including but not limited to the study design, the data collection/analysis/interpretation, or writing of the manuscript.

Availability of data and material

The community, through the Community Planning Group (CPG), and the research team jointly owns all data, consistent with the principles of community-based participatory research (CBPR). Results are shared at every meeting of the CPG. Datasets are shared by request and approval of the community and CPG as they become available.

Authors’ contributions

BBB and HKT designed the study. DMO, SEH, BBB, KLS, and PJH collected the samples and measured the phenotypes and covariates. HWW, BAH, and LKV analyzed the data. SA, BAH, DJL, KT, and LKV wrote the manuscript. SA, BBB, and HKT had primary responsibility for the final content. All authors read and approved the final manuscript.

Competing interests

The authors declare that they have no competing interests.

Consent for publication

Not applicable (the manuscript contains no individual person’s data).

Ethics approval and consent to participate

All participants signed informed consent documents, and the study protocols were approved by the Institutional Review Boards of the University of Alaska and the National and Alaska Area Indian Health Service Institutional Review Boards, as well as the Yukon Kuskokwim Human Studies Committee.

Financial support

This research was funded by the National Institutes of Health grants and contracts HHSN268200782096C, DK080188, DK074842, RR016430.

Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Authors’ Affiliations

(1)
Department of Epidemiology, University of Alabama at Birmingham
(2)
Department of Biostatistics, University of Alabama at Birmingham
(3)
Department of Health Outcomes and Policy, College of Medicine, University of Florida
(4)
Department of Biology, King University
(5)
Center for Alaska Native Health Research, Institute of Arctic Biology, University of Alaska Fairbanks
(6)
Departments of Molecular Biosciences and Nutrition, University of California at Davis
(7)
Department of Pharmaceutics, University of Washington

References

  1. Ahn J, et al. Genome-wide association study of circulating vitamin D levels. Hum Mol Genet. 2010;19(13):2739–45. doi:https://doi.org/10.1093/hmg/ddq155.View ArticlePubMedPubMed CentralGoogle Scholar
  2. Almasy L, Blangero J. Multipoint quantitative-trait linkage analysis in general pedigrees. Am J Hum Genet. 1998;62(5):1198–211.View ArticlePubMedPubMed CentralGoogle Scholar
  3. Aslibekyan S, et al. Evidence for novel genetic loci associated with metabolic traits in Yup’ik people. Am J Hum Biol. 2013;25(5):673–80. doi:https://doi.org/10.1002/ajhb.22429.View ArticlePubMedPubMed CentralGoogle Scholar
  4. Bischoff-Ferrari HA, et al. Estimation of optimal serum concentrations of 25-hydroxyvitamin D for multiple health outcomes. Am J Clin Nutr. 2006;84(1):18–28.PubMedGoogle Scholar
  5. Boyer BB, et al. Building a community-based participatory research center to investigate obesity and diabetes in Alaska Natives. Int J Circumpolar Health. 2005;64(3):281–90.View ArticlePubMedGoogle Scholar
  6. Chen J, et al. ToppGene Suite for gene list enrichment analysis and candidate gene prioritization. Nucleic Acids Res. 2009;37(Web Server issue):W305–11. doi:https://doi.org/10.1093/nar/gkp427.View ArticlePubMedPubMed CentralGoogle Scholar
  7. Elston RC, et al. The Elston-Stewart algorithm for continuous genotypes and environmental factors. Hum Hered. 1992;42(1):16–27.View ArticlePubMedGoogle Scholar
  8. Estrada K, et al. Genome-wide meta-analysis identifies 56 bone mineral density loci and reveals 14 loci associated with risk of fracture. Nat Genet. 2012;44(5):491–501. doi:https://doi.org/10.1038/ng.2249.View ArticlePubMedPubMed CentralGoogle Scholar
  9. Fiala M, et al. MGAT3 mRNA: a biomarker for prognosis and therapy of Alzheimer’s disease by vitamin D and curcuminoids. J Alzheimers Dis. 2011;25(1):135–44. doi:https://doi.org/10.3233/JAD-2011-101950.PubMedGoogle Scholar
  10. Frost P. Vitamin D deficiency among northern Native Peoples: a real or apparent problem? Int J Circumpolar Health. 2012;71:18001. doi:https://doi.org/10.3402/IJCH.v71i0.18001.PubMedGoogle Scholar
  11. George VT, Elston RC. Testing the association between polymorphic markers and quantitative traits in pedigrees. Genet Epidemiol. 1987;4(3):193–201. doi:https://doi.org/10.1002/gepi.1370040304.View ArticlePubMedGoogle Scholar
  12. Gezen-Ak D, et al. The effects of vitamin D receptor silencing on the expression of LVSCC-A1C and LVSCC-A1D and the release of NGF in cortical neurons. PLoS One. 2011;6(3):e17553. doi:https://doi.org/10.1371/journal.pone.0017553.View ArticlePubMedPubMed CentralGoogle Scholar
  13. Heaney RP. The vitamin D requirement in health and disease. J Steroid Biochem Mol Biol. 2005;97(1-2):13–9. doi:https://doi.org/10.1016/j.jsbmb.2005.06.020.View ArticlePubMedGoogle Scholar
  14. Hiraki LT, et al. Exploring the genetic architecture of circulating 25-hydroxyvitamin D. Genet Epidemiol. 2013;37(1):92–8. doi:https://doi.org/10.1002/gepi.21694.View ArticlePubMedGoogle Scholar
  15. Holick MF. Sunlight and vitamin D for bone health and prevention of autoimmune diseases, cancers, and cardiovascular disease. Am J Clin Nutr. 2004a; 80(6 Suppl): 1678S-88.Google Scholar
  16. Holick MF. Vitamin D: importance in the prevention of cancers, type 1 diabetes, heart disease, and osteoporosis. Am J Clin Nutr. 2004b; 79(3): 362-71Google Scholar
  17. Holick MF. Vitamin D deficiency. N Engl J Med. 2007;357(3):266–81. doi:https://doi.org/10.1056/NEJMra070553.View ArticlePubMedGoogle Scholar
  18. Hossein-Nezhad A, et al. Influence of vitamin D status and vitamin D3 supplementation on genome wide expression of white blood cells: a randomized double-blind clinical trial. PLoS One. 2013;8(3):e58725. doi:https://doi.org/10.1371/journal.pone.0058725.View ArticlePubMedPubMed CentralGoogle Scholar
  19. Kishimoto K, et al. Fad104, a positive regulator of adipogenesis, negatively regulates osteoblast differentiation. Biochem Biophys Res Commun. 2010;397(2):187–91. doi:https://doi.org/10.1016/j.bbrc.2010.05.077.View ArticlePubMedGoogle Scholar
  20. Lauc G, et al. Loci associated with N-glycosylation of human immunoglobulin G show pleiotropy with autoimmune diseases and haematological cancers. PLoS Genet. 2013;9(1):e1003225. doi:https://doi.org/10.1371/journal.pgen.1003225.View ArticlePubMedPubMed CentralGoogle Scholar
  21. Lee WP, et al. Correlations among serum calcium, vitamin D and parathyroid hormone levels in the elderly in southern Taiwan. J Nurs Res. 2002;10(1):65–72.View ArticlePubMedGoogle Scholar
  22. Levin GP, et al. Genetic variants and associations of 25-hydroxyvitamin D concentrations with major clinical outcomes. JAMA. 2012;308(18):1898–905. doi:https://doi.org/10.1001/jama.2012.17304.View ArticlePubMedPubMed CentralGoogle Scholar
  23. Li J, Ji L. Adjusting multiple testing in multilocus analyses using the eigenvalues of a correlation matrix. Heredity (Edinb). 2005;95(3):221–7. doi:https://doi.org/10.1038/sj.hdy.6800717.View ArticleGoogle Scholar
  24. Luick B, et al. Locally harvested foods support serum 25-hydroxyvitamin D sufficiency in an indigenous population of Western Alaska. Int J Circumpolar Health. 2014;73. doi: https://doi.org/10.3402/ijch.v73.22732
  25. Luick B, et al. Associations of parathyroid hormone concentration among Alaska Natives of Western Alaska. FASEB J. 2015;29:735(Abstract).Google Scholar
  26. Maiellaro-Rafferty K, et al. Altered regional cardiac wall mechanics are associated with differential cardiomyocyte calcium handling due to nebulette mutations in preclinical inherited dilated cardiomyopathy. J Mol Cell Cardiol. 2013;60:151–60. doi:https://doi.org/10.1016/j.yjmcc.2013.04.021.View ArticlePubMedPubMed CentralGoogle Scholar
  27. Melzer D, et al. A genome-wide association study identifies protein quantitative trait loci (pQTLs). PLoS Genet. 2008;4(5):e1000072. doi:https://doi.org/10.1371/journal.pgen.1000072.View ArticlePubMedPubMed CentralGoogle Scholar
  28. Miles LE, et al. Measurement of serum ferritin by a 2-site immunoradiometric assay. Anal Biochem. 1974;61(1):209–24.View ArticlePubMedGoogle Scholar
  29. Mohatt GV, et al. The Center for Alaska Native Health Research Study: a community-based participatory research study of obesity and chronic disease-related protective and risk factors. Int J Circumpolar Health. 2007;66(1):8–18.View ArticlePubMedGoogle Scholar
  30. Nash SH, et al. Stable nitrogen and carbon isotope ratios indicate traditional and market food intake in an indigenous circumpolar population. J Nutr. 2012;142(1):84–90.View ArticlePubMedGoogle Scholar
  31. Nussey S, Whitehead S. Endocrinology: an integrated approach. Oxford: BIOS Scientific Publishers, Ltd.; 2001.View ArticleGoogle Scholar
  32. Nyholt DR. A simple correction for multiple testing for single-nucleotide polymorphisms in linkage disequilibrium with each other. Am J Hum Genet. 2004;74(4):765–9. doi:https://doi.org/10.1086/383251.View ArticlePubMedPubMed CentralGoogle Scholar
  33. Nykjaer A, et al. Cubilin dysfunction causes abnormal metabolism of the steroid hormone 25(OH) vitamin D(3). Proc Natl Acad Sci U S A. 2001;98(24):13895–900. doi:https://doi.org/10.1073/pnas.241516998.View ArticlePubMedPubMed CentralGoogle Scholar
  34. O’Brien DM, et al. Red blood cell delta15N: a novel biomarker of dietary eicosapentaenoic acid and docosahexaenoic acid intake. Am J Clin Nutr. 2009;89(3):913–9. doi:https://doi.org/10.3945/ajcn.2008.27054.View ArticlePubMedPubMed CentralGoogle Scholar
  35. Paredes R, et al. The Runx2 transcription factor plays a key role in the 1alpha,25-dihydroxy vitamin D3-dependent upregulation of the rat osteocalcin (OC) gene expression in osteoblastic cells. J Steroid Biochem Mol Biol. 2004;89–90(1-5):269–71. doi:https://doi.org/10.1016/j.jsbmb.2004.03.076.View ArticlePubMedGoogle Scholar
  36. Pedigo N, et al. A 5’-distal element mediates vitamin D-inducibility of PDGF-A gene transcription. Growth Factors. 2003;21(3-4):151–60.View ArticlePubMedGoogle Scholar
  37. Rejnmark L, et al. Vitamin D insufficiency in Greenlanders on a westernized fare: ethnic differences in calcitropic hormones between Greenlanders and Danes. Calcif Tissue Int. 2004;74(3):255–63. doi:https://doi.org/10.1007/s00223-003-0110-9.View ArticlePubMedGoogle Scholar
  38. Salehi-Tabar R, et al. Vitamin D receptor as a master regulator of the c-MYC/MXD1 network. Proc Natl Acad Sci U S A. 2012;109(46):18827–32. doi:https://doi.org/10.1073/pnas.1210037109.View ArticlePubMedPubMed CentralGoogle Scholar
  39. Saliba W, et al. The relationship between serum 25(OH)D and parathyroid hormone levels. Am J Med. 2011;124(12):1165–70. doi:https://doi.org/10.1016/j.amjmed.2011.07.009.View ArticlePubMedGoogle Scholar
  40. Samuel S, Sitrin MD. Vitamin D’s role in cell proliferation and differentiation. Nutr Rev. 2008;66(10 Suppl 2):S116–24. doi:https://doi.org/10.1111/j.1753-4887.2008.00094.x.View ArticlePubMedGoogle Scholar
  41. Sellers EA, et al. Adaptation of Inuit children to a low-calcium diet. CMAJ. 2003;168(9):1141–3.PubMedPubMed CentralGoogle Scholar
  42. Signorello LB, et al. Blood vitamin D levels in relation to genetic estimation of African ancestry. Cancer Epidemiol Biomarkers Prev. 2010;19(9):2325–31. doi:https://doi.org/10.1158/1055-9965.EPI-10-0482.View ArticlePubMedPubMed CentralGoogle Scholar
  43. Sobel E, Lange K. Descent graphs in pedigree analysis: applications to haplotyping, location scores, and marker-sharing statistics. Am J Hum Genet. 1996;58(6):1323–37.PubMedPubMed CentralGoogle Scholar
  44. Tranchevent LC. ENDEAVOUR update: a web resource for gene prioritization in multiple species. Nucleic Acids Res. 2008;36(Web Server issue):W377–384. doi:https://doi.org/10.1093/nar/gkn325.View ArticlePubMedPubMed CentralGoogle Scholar
  45. Vaughan LK, et al. Linkage and association analysis of obesity traits reveals novel loci and interactions with dietary n-3 fatty acids in an Alaska Native (Yup’ik) population. Metabolism. 2015;64(6):689–97. doi:https://doi.org/10.1016/j.metabol.2015.02.008.View ArticlePubMedPubMed CentralGoogle Scholar
  46. Wang TJ, et al. Common genetic determinants of vitamin D insufficiency: a genome-wide association study. Lancet. 2010;376(9736):180–8. doi:https://doi.org/10.1016/S0140-6736(10)60588-0.View ArticlePubMedPubMed CentralGoogle Scholar
  47. Williams-Blangero S, Blangero J. Collection of pedigree data for genetic analysis in isolate populations. Hum Biol. 2006;78(1):89–101. doi:https://doi.org/10.1353/hub.2006.0023.View ArticlePubMedGoogle Scholar
  48. Wjst M, et al. A genome-wide linkage scan for 25-OH-D(3) and 1,25-(OH)2-D3 serum levels in asthma families. J Steroid Biochem Mol Biol. 2007;103(3-5):799–802. doi:https://doi.org/10.1016/j.jsbmb.2006.12.053.View ArticlePubMedGoogle Scholar
  49. Yu W, et al. A navigator for human genome epidemiology. Nat Genet. 2008;40(2):124–5. doi:https://doi.org/10.1038/ng0208-124.View ArticlePubMedGoogle Scholar
  50. Zerwekh JE. Blood biomarkers of vitamin D status. Am J Clin Nutr. 2008;87(4):1087S–91.PubMedGoogle Scholar

Copyright

© The Author(s) 2016

Advertisement