Racial differences in microRNA and gene expression in hypertensive women

Abstract

Systemic arterial hypertension is an important cause of cardiovascular disease morbidity and mortality. African Americans are disproportionately affected by hypertension, in fact the incidence, prevalence, and severity of hypertension is highest among African American (AA) women. Previous data suggests that differential gene expression influences individual susceptibility to selected diseases and we hypothesized that this phenomena may affect health disparities in hypertension. Transcriptional profiling of peripheral blood mononuclear cells from AA or white, normotensive or hypertensive females identified thousands of mRNAs differentially-expressed by race and/or hypertension. Predominant gene expression differences were observed in AA hypertensive females compared to AA normotensives or white hypertensives. Since microRNAs play important roles in regulating gene expression, we profiled global microRNA expression and observed differentially-expressed microRNAs by race and/or hypertension. We identified novel mRNA-microRNA pairs potentially involved in hypertension-related pathways and differently-expressed, including MCL1/miR-20a-5p, APOL3/miR-4763-5p, PLD1/miR-4717-3p, and PLD1/miR-4709-3p. We validated gene expression levels via RT-qPCR and microRNA target validation was performed in primary endothelial cells. Altogether, we identified significant gene expression differences between AA and white female hypertensives and pinpointed novel mRNA-microRNA pairs differentially-expressed by hypertension and race. These differences may contribute to the known disparities in hypertension and may be potential targets for intervention.

Figure 1. Diagram of study work flow.

Figure 2. Global changes in gene expression by race and hypertension.

Total RNA was isolated from PBMCs from African American (AA) or white (W) age-matched females who were normotensive (NT) or hypertensive (HT). (A thru D) Gene expression was assessed using microarray and differences in expression by either race or presence of hypertension were analyzed by PCA (n = 6/group). (E) Venn diagram of the total number of significantly upregulated (up arrow) or downregulated (down arrow) mRNAs for each comparison. Each comparison highlights significant changes in hypertensives compared with normotensive racial controls or changes observed in AA hypertensives compared with white hypertensives (AAHT vs. WHT) or AA normotensives compared with white normotensives (AANT vs. WNT). (F) Heat map of significantly-changed, hypertension-related mRNAs identified using IPA. Red (up) and green (down) indicate relative Z-ratio for each comparison. (G,H) Venn diagrams of significantly upregulated or downregulated mRNAs in the hypertension-related gene set (G) or inflammation-related gene set (H). AANT, AA normotensive; AAHT, AA hypertensive; WNT, white normotensive; WHT, white hypertensives.

Figure 3. miRNA levels in PBMCs are differentially expressed by race and/or hypertension status.

(A) Heat map of all 36 significantly-changed miRNAs in PMBCs identified by microarray (n = 10/group). Red (up) and green (down) indicate relative fold-changes for each comparison. Red boxes indicate which individual miRNAs were PCR-validated. (B) miRNAs were validated in an expanded cohort (n = 20/group) using RT-qPCR with miRNA-specific primers, after exclusion of outliers. Histograms represent the mean ± SEM. *P < 0.05, **P < 0.01, ^#P < 0.09; Student’s t-test.

Figure 4. Hypertension and race-associated miRNA target validation.

HUVECs (left) and HAECs (right) were transiently transfected with precursor mimics of miR-20a-5p (A), miR-4763-5p (B), miR-4717-3p (C), or miR-4709-3p (D). miRNAs and their predicted mRNA targets were quantified by RT-qPCR. miRNA levels in transfected cells are compared to scrambled control. Histograms for each individual mRNA are compared to scrambled control. Representative immunoblots of target protein levels from at least 3 independent experiments. Histograms represent the mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001; Student’s T-test.

Figure 5. Analysis of miRNA binding sites within mRNA targets using reporters.

3′ UTR fragments from target mRNAs containing the miRNA binding sites were cloned downstream of a Renilla luciferase (RL) reporter (Panels A–D, left). Each plasmid also contains a Firefly luciferase (FL) reporter, which serves as a transfection control. Above each construct, the base pairs of the cloned 3′ UTRs are indicated relative to the start of each mRNA’s 3′ UTR sequence. Red bars indicate the approximate location of each miRNA binding site (see Supplemental Fig. S3 for detailed miRNA binding sites). HeLa cells were transfected with each luciferase reporter plasmid containing the partial wild-type 3′ UTRs of MLC1 (A), APOL3 (B), or PLD1 (C,D) and either a scrambled miRNA control or indicated miRNA mimic. The ratio of RL/FL activity is shown and each wild-type plasmid was normalized and compared with the scrambled control. MicroRNA seed sequences were mutated (mut) for the indicated sites (see Supplemental Fig. S3 for specific mutations) and luciferase activity was measured as above and normalized to the scrambled control. Histograms represent the mean ± S.D. of three independent experiments. **P < 0.01, ***P < 0.001; Student’s t-test.

Figure 6. mRNA levels in PBMCs are differentially expressed by race and/or hypertension status.

14 mRNAs identified in the microarray and subsequent miRNA target analysis were validated in an expanded cohort (n = 20/group) using RT-qPCR with mRNA-specific primers, after exclusion of outliers. Histograms represent the mean ± SEM. *P < 0.05, **P < 0.01, ^#P < 0.09; Student’s t-test.