Polimorfismos e haplótipos em NOTCH1 estão associados ao RDW

Abstract

Anisocytosis or Red Cell Distribution Width (RDW) increases with age and is predictive of cardiovascular disease, cancer, infections, aging, morbidity and mortality andothers. Given the implications and clinical potential of this marker, it is important to identify genes and genetic polymorphisms associated with variability in differentpopulations. In GWAS Catalog, SNPs in some genes, including NOTCH1, are associated with RDW. NOTCH1 acts in Notch pathway, crucial for embryogenesis, proliferation and apoptosis, metabolism, development and for functioning of the nervous, cardiovascular and endocrine systems. The objective of this work is to validate association between RDW and rs3124592 of NOTCH1 gene, described in theGWAS Catalog, and to identify SNPs associated with RDW in the Brazilian population.Participants in the “Health, Wellbeing and Aging” (SABE) cohort of older adults were stratified into cases and controls according to RDW values. The cases were defined by the value of coefficient of variation of RDW ≥15%. We filter SNPs located at start and end positions of NOTCH1 and 50Kb on both sides. We exclude In/Dels and unannotated SNPs. We selected SNPs with minor allelic frequency (MAF) greater thanor equal to 0.01 and HardyWeinberg equilibrium (p>0.05) and r²≥0.8. A total of 1097 participants were included, with a median age of 71.69 years, of which 703 are women(64.08%). Of these, 134 (12.22%) individuals have RDW ≥ 15%. From a total of 5599SNPs, 226 Tag SNPs were analyzed. SNP rs3124592, from the GWAS Catalog, wasnot associated with RDW in this population. After adjusting for age, gender and ancestry and considering the Bonferroni correction, the analysis showed an association between high RDW and SNP rs9411206 in the log-additive model, with thefrequency of GG genotype being higher in cases (57.5%) than controls (40.7%) (p = 0.0001477). The variants rs2229971 (p= 0.0001673), rs9411207 (p= 0.0002309) and rs11574891 (p= 0.0002440) are associated with RDW in recessive model. After adjusting for age, gender, ancestry, DBP, Hb1Ac, LDL, hsCRP, hemoglobin, and stroke, the same SNPs remained associated, except for rs9411206. Associations were observed mainly in the recessive model: rs2229971 (p= 0.00006304); rs9411207 (p=0.00007347) and rs11574891 (p=0.00002759). After this adjustment, rs9411206 was marginally associated in dominant (p= 0.0004790930) and additive (p= 0.000461617) models. Of these SNPs, rs9411206 is the only one that has eQTLs and sQTLs identified in “in silico” analysis, with results mainly in blood. rs2229971 is associated with mRNA abundance. Regulome indicates the potential of these SNPs to affect gene expression levels and rs11574891 is the most representative of this analysis. In HaploReg, all variants are associated with enhancers in lung, right ventricle, brain and spleen, as well as epigenetic modifications of histone protein H3K4me1 in brain and pancreas. In conclusion, rs3124592 was not validated in this population, however, new variants associated with RDW were identified: rs2229971, rs9411207, rs11574891 and rs9411206. Thus, our data suggest that these SNPs are associated with elevated levels of RDW and can be evaluated as biomarkers of chronic diseases and aging, corroborating recent evidence in cell cultures, animal models and clinical studies.