Regulation of heme oxygenase-1 protein expression by miR-377 in combination with miR-217

Abstract

Heme oxygenase-1 (HO-1) enzyme plays a critical role in metabolizing the excess heme generated during hemolysis. Our previous studies suggested that during intravascular hemolysis the expression of HO-1 protein is not sufficient to reduce the oxidative burden of free heme in the vasculature. This led us to hypothesize that a post-translational mechanism of control exists for HO-1 expression. Micro-RNAs (miRNA) affect gene expression by post-transcriptional gene regulation of transcripts. We performed in silico analysis for the human HMOX1-3' untranslated region (3' UTR) and identified candidate miRNA binding sites. Two candidate miRNAs, miR-377 and miR-217, were cloned and co-transfected with a luciferase vector containing the human HMOX1-3'UTR region. The combination of miR-377 and miR-217 produced a 58% reduction in HMOX1-3'UTR luciferase reporter expression compared with controls. The same constructs were then used to assess how overexpression of miR-217 and miR-377 affected HO-1 levels after induction with hemin. Cells transfected with the combination of miR-377 and miR-217 exhibited no change in HMOX1 mRNA levels, but a significant reduction in HMOX1 (HO-1) protein expression and enzyme activity compared with non-transfected hemin-stimulated controls. Transfection with either miR-377 or miR-217 alone did not produce a significant decrease in HO-1 protein expression or enzyme activity. Knockdown of miR-217 and miR-377 in combination leads to up-regulation of HO-1 protein. Exposure to hemin induced a significant reduction in miR-217 expression and a trend toward decreased miR-377 expression in two different cells lines. In summary, these data suggests that the combination of miR-377 and miR-217 help regulate HO-1 protein expression in the presence of hemin.

FIGURE 1.

Predicted interactions between the HMOX1-3′ UTR with miR-217 and miR-377. The human HMOX1 gene consists of 5 exons with a 603-base pair 3′ UTR, indicated in gray. Mature hsa-miR-217 (miR-217) is predicted to interact with the HMOX1–3′ UTR from positions 462 to 468 via a 8-mer seed match interaction. Mature hsa-miR-377 (miR-377) is predicted to interact with the HMOX1–3′ UTR from positions 576 to 582 via a 8-mer seed match interaction.

FIGURE 2.

miR-217 and miR-377 interact with the HMOX1-3′ UTR. Combinations of the three 3′ UTR luciferase reporter vectors containing the HMOX1–3′ UTR or a mutation in the proposed miR-217 binding site or the proposed miR-377 binding site HMOX1–3′ UTR was transfected into HEK 293 cells along with pcDNA^TM6.2-GW/EmGFP-miR-217 (miR-217), pcDNA^TM6.2-GW/EmGFP-miR-377 (miR-377), scrambled (scrmiR) controls pcDNA^TM6.2-GW/EmGFP-scrmiR-217 (scrmiR-217), and hsa-scrmir-377 (scrmiR-377). Twenty-four hours after transfection a luciferase assay was performed. Results demonstrate that miR-377 alone, and the combination of miR-217 and miR-377 significantly (*, p < 0.05) decreased luciferase reporter activity when compared with reporter alone (Kruskal-Wallis One Way Analysis of Variance on Ranks with Student-Newman-Keuls multiple comparisons). Data are presented as mean relative percent luciferase expression, mean ± S.E., with n = 2–3 replications per condition.

FIGURE 3.

Overexpression of miR-217 and miR-377 does not change HMOX1 mRNA levels but decreases HO-1 protein expression. A, HEK 293 cells were transfected with miR-217 + miR-377 or scrmiR-217 + scrmiR-377. After 24 h the cells were treated with 10 μm hemin for 1 h, after which it was removed and the cells incubated for an additional 24 h in basal medium. RNA was then collected and expression of HMOX1 mRNA was evaluated and normalized to levels of GAPDH mRNA. Data are presented as mean ± S.E., with n = 3 replications per condition. B, representative HO-1 and GAPDH Western blots for HEK 293 cells that were treated with 10 μm hemin 24 h after transfection with either scrmiR-217 + scrmiR-377 or miR-217 + miR-377 constructs followed by an additional 24-h incubation in basal medium. C, representative HO-1 and GAPDH Western blots for primary HUVEC cells, which were treated with 20 μm hemin for 1 h, 24 h after nucleofection with miRIDIAN miRNA-217 and miR-377 mimics followed by an additional 8-h incubation in basal medium. Bar graphs below Western blots represent densitometry of HO-1 protein expression normalized to GAPDH protein expression, in which n = 3 independent trials. *, p < 0.05 compared with hemin-treated cells using a one-way ANOVA with Bonferroni t test.

FIGURE 4.

Overexpression of miR-217 + miR-377 decreases HO-1 enzyme activity. Forty-one h after transfection HEK 293 cells were treated for 1 h with 10 μm hemin, followed by a 6-h incubation in growth medium. Heme oxygenase activity for each independent sample was calculated as picomole of bilirubin formed/mg of microsomal protein/h and normalized to the non-transfected hemin-treated cells by division to eliminate intra-assay variation. Non-transfected hemin-treated cells represent 100% HO-1 enzyme activity. Data are presented as mean ± S.E., *, indicates a p < 0.05 significant difference between the sample and non-transfected hemin-treated cells using a one-way ANOVA with Bonferroni t test.

FIGURE 5.

Inhibition of miR-217 + miR-377 increases HO-1 protein expression. Western blot of primary HUVEC cell samples that were treated for 1 h with 20 μm hemin followed by an additional 24-h incubation in growth medium 36 h after nucleofection with the miRIDIAN hairpin inhibitor against miRNA-217 and miRIDIAN hairpin inhibitor against miR-377. Bar graph below Western blots represents densitometry of HO-1 protein expression normalized to GAPDH protein expression, in which n = 3 independent trials. *, p < 0.05 compared with hemin-treated cells using a one-way ANOVA with Bonferroni t test.

FIGURE 6.

Treatment with hemin affects miRNA expression levels. A, HEK 293 cells were treated with 10 μm hemin for 1 h followed by an 8-h incubation in basal medium. Treatment with hemin significantly reduced levels of miR-217 compared with control (p < 0.05). Levels of miR-377 were also reduced (p = 0.08) compared with control. B, primary HUVEC cells were treated with 10 μm hemin for 1 h followed by either an 8- or 24-h incubation in basal medium. Levels of miR-217 were significantly reduced after 24 h compared with control cells (p < 0.05). Levels of miR-377 were not significantly reduced compared with control cells. C, treatment with hemin followed by an 8-h incubation did not significantly alter levels of miRNA of four other miRNAs (miR-873, miR-520a-5p, miR-525–5p, and miR-1205) predicted to interact with the HMOX1–3′ UTR. Data are presented as mean ± S.E., with n = 3 trials per condition. *, p < 0.05 using a one-way ANOVA with Bonferroni t test post-test.

FIGURE 7.

Proposed model of miRNA regulation of HMOX1. A, at baseline, levels of HMOX1 translation are balanced by expression of miR-217 and miR-377, which interact with the 3′ UTR to attenuate protein expression. B, although heme may increase miRNA biogenesis, our experiments indicate that intracellular levels of mature miR-217 and miR-377 decrease in the presence of heme. This reduces miRNA interaction with the HMOX1–3′ UTR allowing for increased HO-1 protein expression. C, conversely, decreased HO-1 protein expression occurs when miR-217 and miR-377 are overexpressed.