Premium
Multi‐Trait Genetic Mapping Reveals Novel Loci Responsible for Genetic and Genetic‐by‐Diet Interactions Affecting Bone, Vitamin D, and Calcium Metabolism
Author(s) -
Fleet James C.,
ReyesFernandez Perla C.,
Replegle Rebecca A.,
Chanpaisaeng Krittikan
Publication year - 2017
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.31.1_supplement.299.3
Subject(s) - biology , quantitative trait locus , phenotype , genetics , genetic analysis , gene , genetic variation , calcium metabolism , peak bone mass , bone remodeling , genetic model , calcium , osteoporosis , endocrinology , medicine , bone density
Attaining maximal Peak Bone Mass (PBM) is a critical approach for reducing the lifetime risk of osteoporosis. While PBM is compromised by inadequate dietary calcium (Ca) intake and is influenced by genetics, it is not clear how genetics interacts with diet to influence PBM. Previously, we identified genetic loci controlling PBM and Ca metabolism in mice, as well as their response to low dietary Ca intake (i.e. a gene‐by‐diet interaction, GxD). Others have shown that multi‐trait analysis improves detection of genetic effects and we applied this approach to identify novel genetic regulators and GxD interactions that influence PBM and Ca metabolism. Mice from 51 BXD recombinant inbred lines were fed either adequate (0.5%) or low (0.25%) Ca diets from 4 to 12 wks of age (n=8/diet/line). Ca absorption (CaAbs), bone mass (BMC and BMC), serum 1,25 dihydroxyvitamin D (1,25 D), and FGF‐23 were measured in both diet groups and the response to dietary Ca restriction (RCR) was calculated for each phenotype in each line. Linear regression analysis revealed many correlations among our phenotypes. This suggests there are genetic loci that coordinately regulate multiple traits. To test this hypothesis we used principle components analysis (PCA) to group traits and we used the resulting principle components (PC) for genetic mapping using composite interval mapping. Candidate genes underlying loci were identified with PROVEAN analysis (protein coding effects) or eQTL analysis in WebQTL (mRNA level effects). On Chr 9 we found a loci controlling the PC for the positive correlation between CaAbs and 1,25 D (LOD = 13) and the PC for the negative relationship between BMD/BMC RCR and serum 1,25 D (LOD = 8.4). Within this locus is a cis eQTL controlling renal mRNA level for Ets‐1 , a transcription factor that controls vitamin D‐regulated CYP24A1 gene expression and vitamin D degradation. On Chr 4 we found two PC's that condensed BMD RCR and BMC RCR (LOD=3.8) or CaAbs on the adequate or low Ca diet (LOD = 4.0). Tceanc2 mRNA was identified as cis eQTL in this region; its bone and kidney mRNA levels correlate positively to BMD (r = 0.4) and negatively to BMC RCR (r = −39) and CaAbs (r= −0.57). Novel QTL were identified by mapping a PC for the positive relationship among CaAbs, BMD RCR, and BMC RCR: Chr11 (LOD=3.8), Chr13 (LOD=5.5), Chr15 (LOD=7.7), Chr19 (LOD=8.5). Dennd3 (Chr15) was a cis eQTL expressed in kidney; this protein influences endocytic recycling of proteins. Slc22A29 (Chr19) has a non‐synonymous coding variant that could affect cation transport in the kidney. Our study demonstrates that multi‐trait mapping can capture novel loci not identified in single trait mapping. Using this approach, we have identified several interesting candidate genes that influence multiple bone, vitamin D, and Ca metabolism phenotypes. The importance of this work is that it establishes a framework for understanding how natural genetic variation can interact with dietary Ca intake to influence the development of PBM. Support or Funding Information Supported by NIH award ES019103 to JCF.