Quantile-specific heritability of monocyte chemoattractant protein-1, and relevance to rs1024611-disease interactions.

Pubmed ID: 34624603

Pubmed Central ID: PMC10124179

Journal: Cytokine

Publication Date: Jan. 1, 2022

MeSH Terms: Humans, Male, Adult, Female, Genotype, Phenotype, Chemokine CCL2, Disease

Grants: N01HC25195, HHSN268201500001I, R21 ES020700, HHSN268201500001C

Authors: Williams PT

Cite As: Williams PT. Quantile-specific heritability of monocyte chemoattractant protein-1, and relevance to rs1024611-disease interactions. Cytokine 2022 Jan;149:155722. Epub 2021 Oct 5.

Studies:

Abstract

BACKGROUND: Monocyte chemoattractant protein-1 (MCP-1) concentrations are 34% to 47% heritable. Larger -2518 G/A (rs1024611) genotypes differences are reported for: 1) MCP-1 production in stimulated vs. basal cells; and 2) MCP-1 concentrations in diseased (sepsis, brain abscess, hepatitis B virus, Alzheimer's disease, Behcet's disease, and systemic lupus erythematosus) vs. healthy patients. Those results suggest that the -2518 G/A effect size may depend on whether the phenotype is high or low relative to its distribution (quantile-dependent expressivity). METHOD: To test whether quantile-dependent expressivity applies more broadly to genetic influences on MCP-1 concentrations, quantile-specific offspring-parent (β<sub>OP</sub>) and full-sib regression slopes (β<sub>FS</sub>) were estimated by applying quantile regression to the age- and sex-adjusted serum MCP-1 concentrations of Framingham Heart Study families. Quantile-specific heritabilities were calculated as h<sup>2</sup> = 2β<sub>OP</sub>/(1 + r<sub>spouse</sub>) and h<sup>2</sup>={(1 + 8r<sub>spouse</sub>β<sub>FS</sub>)<sup>0.5</sup>-1}/(2r<sub>spouse</sub>)). RESULTS: Heritability (h<sup>2</sup> ± SE) of MCP-1 concentrations increased from 0.15 ± 0.05 at the 10th percentile of the MCP-1 distribution, 0.23 ± 0.04 at the 25th, 0.32 ± 0.05 at the 50th, 0.43 ± 0.07 at the 75th, and 0.44 ± 0.07 at the 90th percentile, or an 0.0041 ± 0.0009 increase for each one-percent increment in the MCP-1 distribution (P<sub>linear trend</sub> = 2.4 × 10<sup>-5</sup>) when estimated from β<sub>OP</sub>, and (P<sub>linear trend</sub> = 7.7 × 10<sup>-9</sup>) when estimated from β<sub>FS</sub>. Compared to the 10th percentile, β<sub>OP</sub>-estimated h<sup>2</sup> was 3-fold greater at the 90th percentile (P<sub>difference</sub> = 0.0003), and 6.9-fold greater when estimated from β<sub>FS</sub> (P<sub>difference</sub> = 3.3 × 10<sup>-6</sup>). Re-analysis of in vivo comparison of MCP-1 concentrations in controls vs. patients with MCP-1-elevating conditions, and in vitro studies of MCP-1 production in basal vs. stimulated cells, show rs1024611 genotypes differences that were consistent with quantile-dependent expressivity. CONCLUSION: The heritability of circulating MCP-1 concentrations is quantile-dependent.