UCLA

Genome-wide association studies (GWAS) have been successful in identifying variants with low to moderate effects on serum lipid levels. However, determining the causal variant and underlying mechanism in an associated region has been difficult since linkage equilibrium (LD) of the associated variants often extends to large regions covering multiple genes, and GWAS provide no functional information. Using adipose gene expression data, we provide evidence of mechanism underlying a low-density lipoprotein cholesterol (LDL-C) GWAS signal in the apolipoprotein B (APOB) region. First, we determined that an LDL-C GWAS signal uncovered in a population of European ancestry (rs7575840) replicates in a Mexican study sample (1). Mexicans are an understudied population with a high prevalence of dyslipidemia; 44% of the population has high total cholesterol levels (above 200 mg/dl) (2). To further elucidate the mechanism underlying the association between rs7575840 and LDL-C, we measured lipid particle subclasses using Nuclear Magnetic Resonance (NMR), and we determined that rs7575840 is also associated with apoB-containing lipid particles, including very small very-low-density lipoprotein, intermediate lipoprotein, and LDL particles (1). We also discovered a possible mechanism underlying the increase in apoB-containing particles; rs7575840 disrupts a transcription factor binding site of the transcription factors CCAAT/enhancer binding protein alpha and CCAAT/enhancer binding protein beta, impacting gene expression of APOB and a noncoding RNA BU630349 in adipose tissue (1).

In our second study, we searched for whether novel genes and biological gene expression networks associated with triglyceride (TG) levels replicated and shared across populations (Haas et al. submitted (3)). Adipose tissue has previously been shown to be involved in regulation of serum TG levels in humans (4). High serum TG levels predispose to coronary heart disease (CHD). Searching for novel TG-associated biological networks in adipose tissue may provide novel insights into TG regulation. We measured gene expression in adipose samples from two Finnish and one Mexican study sample. In each study sample, we observed a module (biological gene expression network) that was significantly associated with serum TG levels (3). The most significant TG modules observed in each of the three study samples significantly overlapped (p<10-10) and shared 34 genes, utilizing two unique methods of measuring gene expression, microarrays and RNA sequencing (RNAseq). Thus, the results should be robust to the measurement method of gene expression. In the 34 genes shared between the three TG modules, more nonsynonymous variants (p=0.034) and overall variants (p=0.018) were observed in individuals with high TGs when compared with the individuals with low TGs (3). Seven of the 34 genes (ARHGAP30, CCR1, CXCL16, FERMT3, HCST, RNASET2, SELPG) were identified as the key hub genes of all three TG modules (p<10-7) (3). As only 11 of the 34 genes have prior evidence of involvement in CHD, type 2 diabetes, or obesity, our study provides 23 new candidates for TG regulation. Furthermore, two of the 34 genes (ARHGAP9, LST1) reside in previous TG GWAS regions, suggesting them as the regional candidates underlying the GWAS signals. This study presents a novel TG biological network shared across populations.

Our next study focused on determining whether Procadherin 15 (PCDH15) variants are involved with Familial Combined Hyperlipidemia (FCHL). Previous studies have revealed that PCDH15 resides in a region linked to FCHL (5-7), so we hypothesized that nonsynonymous variants in PCDH15 may be associated with FCHL in 92 Finnish and Dutch families. We discovered that one variant (rs10825269) associates with serum TG, apoB, and TC (Total Cholesterol) levels in these families (8). Next, we made a PCDH15 knockout model in collaboration with investigators at Case Western Reserve University and discovered that serum TG and TC levels decreased in mice with both copies of PCDH15 knocked out (8).