SNPs in Lipid and Carotenoid Metabolism and Absorption Genes Impact Carotenoid Responses to a Tomato‐Soy Juice Intervention

Abstract

Several single nucleotide polymorphisms (SNPs) in genes regulating lipid and carotenoid absorption and metabolism have been associated with individual differences in plasma concentrations and absorption kinetics of beta‐carotene (BC) and lycopene (LYC). However, whether these SNPs impact plasma and prostatic carotenoid concentrations in response to a dietary tomato product intervention is not known. The primary question addressed was whether any of seven common SNPs, previously found to be related to plasma carotenoid levels, are associated with the plasma LYC response in prostate cancer patients (n=47) consuming 0, 1, or 2 cans/d of a custom formulated tomato‐soy juice (20.6 mg LYC/can) for 23±1.7 d before scheduled therapeutic prostatectomy. Secondarily, the association of these SNPs with the plasma concentrations of other tomato carotenoids, phytoene (PE), phytofluene (PF), and BC, and prostate concentrations of LYC, PE, and PF were also explored. BCMO1 (rs12934922, rs7501331, and rs6564851), SCARB1 (rs11057841), ABCA1 (rs2230808), MTTP (rs2306985), and APOB48 (rs679899) genotypes were determined using TaqMan Genotyping Assays, and plasma and prostatic carotenoid levels were measured by HPLC. Significant relationships were defined using single SNP linear models with covariates for intervention duration, body mass, baseline plasma carotenoid concentrations, and extraneous carotenoid intake; plasma carotenoid concentrations were cholesterol‐adjusted. Significant SNPs and covariates were combined into a final, predictive linear model. The genotypes of BCMO1 rs12934922 and rs6564851 significantly interacted with treatment group (p=0.021 and p=0.046, respectively) to explain 5.6% of the variance in plasma LYC response to the dietary intervention, with baseline plasma LYC and treatment group being the major predictors of the plasma LYC response (15%, p<0.001 and 55%, p<0.001, respectively), body mass being a minor predictor (2%, p=0.021), and the remainder of the variation (23%) currently unexplained. The rs12934922 genotype interacted with group (p=0.085) to contribute 4% of the variability in prostate LYC concentrations, the rs12934922 genotype alone contributed 4% (p=0.018), and rs6564851 contributed 2% (p=0.053), while treatment group contributed 1% (p=0.035) and baseline plasma LYC contributed 16% (p=0.002), with the remaining variance being unexplained (72%). The rs12934922 genotype was significantly related to the plasma PF and BC responses, the rs7501331 genotype to the plasma PE response, and the rs2306985 and rs2230808 to the plasma and prostatic PF response, respectively. These findings indicate that common SNPs in genes related to lipid and carotenoid absorption and metabolism contribute to the variability in physiologic responses to dietary carotenoids in a controlled trial, and such SNPs may be important to consider when analyzing clinical data. Furthermore, these findings suggest novel relationships of these genes with PE and PF biodistribution. Future studies on the mechanistic effects of these SNPs on tomato carotenoid transport and metabolism are warranted. Clinicaltrials.gov #NCT01009736Support or Funding InformationNIH‐R01CA112632, NIH‐UL1TR001070, NIH‐P30CA01605, The Ohio State University's Food Innovation Center and The Center for Advancement of Functional Foods Research and Entrepreneurship