SPRR1A is a key downstream effector of MiR-150 during both … – Nature.com

Sprr1a knockdown in miR-150 KO mice largely corrects cardiac dysfunction mediated by miR-150 deletion

Sprr1a is a direct target of miR-150 in vitro, miR-150 acts as a gatekeeper of CM survival in part by inhibiting proapoptotic Sprr1a [13], and their correlative cardiac actions are shown [12, 13]; but an in vivo functional relationship between miR-150 and Sprr1a in the heart has not been established. To directly investigate their in vivo functional interaction in the heart, we generated a novel miR-150 KO;Sprr1ahypo/hypo mouse line by breeding miR-150 KO mice with Sprr1ahypo/hypo mice. We first conducted permanent ligation of the left anterior descending (LAD) artery in mice to induce MI. Consistent with a previous report [12], we observe that miR-150 KO mice exhibit normal cardiac function at baseline (Supplementary Table 1 and Fig. 1) but respond differently to MI. Cardiac function is significantly compromised in miR-150-null mice following MI. First, MI significantly worsens the cardiac function of miR-150 KO mice at 3 days as indicated by a decreased ejection fraction (EF), fractional shortening (FS), diastolic left ventricular anterior wall thickness (LVAW), and systolic left ventricular posterior wall thickness (LVPW) as well as an increase in end-systolic volume (ESV) and systolic left ventricular internal diameter (LVID) compared to those of WT controls (Supplementary Table 2 and Fig. 1). MiR-150 KO mice also display impaired cardiac function at 4 weeks post-MI, shown by a significant decrease in EF, FS, diastolic LVPW, and systolic LVPW as well as a significant increase in end-diastolic volume (EDV), ESV, diastolic LVID, and systolic LVID (Supplementary Table 3 and Fig. 1). MI also causes augmented cardiac dysfunction in miR-150 KO mice at 8 weeks as evidenced by a significant decrease in EF, FS, diastolic LVAW, diastolic LVPW, and systolic LVPW as well as a significant increase in EDV, ESV, diastolic LVID, and systolic LVID (Supplementary Table 4 and Fig. 1). In contrast, WT controls show less functional impairment at 4 weeks (Supplementary Table 3 and Fig. 1) and 8 weeks following MI (Supplementary Table 4 and Fig. 1).

We next show that miR-150 KO;Sprr1ahypo/hypo mouse hearts are functionally normal at baseline (Supplementary Table 1 and Fig. 1). However, a significant improvement in cardiac function at 3 days after MI is observed in miR-150 KO;Sprr1ahypo/hypo mice compared to miR-150 KO mice, indicated by an increase in cardiac output (CO), EF, FS, and diastolic LVAW as well as a decrease in EDV, ESV, diastolic LVID, and systolic LVID (Supplementary Table 2 and Fig. 1). MiR-150 KO;Sprr1ahypo/hypo mice also display enhanced cardiac function at 4 weeks post-MI as evidenced by a significant increase in EF, FS, diastolic LVAW, systolic LVAW, diastolic LVPW, and systolic LVPW as well as a significant decrease in EDV, ESV, diastolic LVID, and systolic LVID (Supplementary Table 3 and Fig. 1) compared to those of miR-150 KO mice. Last, we show improved cardiac function in miR-150 KO;Sprr1ahypo/hypo mice at 8 weeks post-MI compared to miR-150 KO mice as shown by a significant increase in CO, EF, FS, heart rate (HR), diastolic LVAW, systolic LVAW, and systolic LVPW as well as a significant decrease in EDV, ESV, diastolic LVID, and systolic LVID (Supplementary Table 4 and Fig. 1). Our morphometric data also show that miR-150 KO;Sprr1ahypo/hypo mice have a significant decrease in the ratio of heart weight/body weight (HW/BW) and the ratio of left ventricle weight/body weight (LVW/BW) at 8 weeks after MI compared to miR-150 KO controls (Supplementary Table 4). Notably, we do not observe any difference in post-MI mortality between groups (Supplementary Tables 1, 3, and 4: see n for animal numbers per each group at week 0, week 4, and week 8 after MI).

We previously reported that miR-150 KO mice display excessive maladaptive post-MI remodeling, such as cardiac damage, inflammation, and apoptosis [12]. To determine whether repression of Sprr1a mediates the major functions of miR-150 in vivo, we employed miR-150 KO;Sprr1ahypo/hypo mice and assessed post-MI remodeling compared to that of miR-150 KO controls. We find that miR-150 KO;Sprr1ahypo/hypo hearts exhibit a decrease in the loss of normal architecture and cellular integrity (Fig. 2A) as well as decreased mRNA levels of fetal Nppa (Fig. 2B) after 8 weeks of MI compared to miR-150 KO hearts. We next examined whether an improved cardiac inflammatory cell (CI) response contributes to the decreased disorganized structure in miR-150 KO;Sprr1ahypo/hypo hearts post-MI. Notably, inflammatory Il-6, Tnf-, and Ptprc are also downregulated in miR-150 KO;Sprr1ahypo/hypo hearts (Fig. 2C, D and Supplementary Fig. 1) compared to miR-150 KO hearts post-MI. Finally, we find that miR-150 KO;Sprr1ahypo/hypo hearts contain significantly lower numbers of cleaved caspase-3-positive cells (Fig. 3A, B), indicating decreased apoptosis in miR-150 KO;Sprr1ahypo/hypo hearts. Our data further show that miR-150 KO;Sprr1ahypo/hypo hearts have decreased mRNA levels of apoptotic P53, Bak1, and Bax (Fig. 3CE) compared to levels in miR-150 KO hearts. Altogether, our data suggest that sustained Sprr1a downregulation ameliorates adverse post-MI remodeling caused by miR-150 deletion and that miR-150 is a functionally important upstream negative regulator of Sprr1a in the heart.

A Representative hematoxylin and eosin (H&E) staining of heart sections of the peri-ischemic border area from the 6 experimental groups at 8 weeks post-MI shows a decrease in the loss of normal architecture and cellular integrity as well as in disorganized structure in miR-150 KO;Sprr1ahypo/hypo hearts compared to miR-150 KO controls. Scale bars: 100m. B qRT-PCR analysis of Nppa expression representing cardiac damage in ischemic areas from WT, miR-150 KO, and miR-150 KO;Sprr1ahypo/hypo mouse hearts at 8 weeks post-MI. qRT-PCR analysis of inflammatory Il6 (C) and Tnf-a (D) expression in ischemic areas from WT, miR-150 KO, and miR-150 KO;Sprr1ahypo/hypo mouse hearts at 8 weeks post-MI. N=56 per group. qRT-PCR data (BD) are shown as the fold induction of gene expression normalized to Gapdh. Two-way ANOVA with Tukeys multiple comparison test. *P<0.05 or **P<0.01 vs. sham for each genotype; #P<0.05, ##P<0.01, or ###P<0.001 vs. WT or miR-150 KO. Data are presented as the meanSEM.

Representative cleaved caspase-3 staining images in heart sections of the peri-ischemic border area in WT, miR-150 KO, and miR-150 KO;Sprr1ahypo/hypo hearts at 8 weeks post-MI (A) and quantification of apoptosis in six 40X fields (B). Scale bars: 100 m. qRT-PCR analysis of proapoptotic p53 (C), Bak1 (D), or Bax (E) expression in the ischemic areas from WT, miR-150 KO, and miR-150 KO;Sprr1ahypo/hypo mouse hearts at 8 weeks post-MI. Data are shown as the fold induction of gene expression normalized to Gapdh. N=6 per group. Two-way ANOVA with Tukeys multiple comparison test. *P<0.05 or ***P<0.001 vs. sham for each genotype; #P<0.05, ##P<0.01, or ###P<0.001 vs. WT or miR-150 KO. Data are presented as the meanSEM.

To further determine the response of miR-150 KO;Sprr1ahypo/hypo mice to MI, we assessed the degree of fibrosis using Massons trichrome staining and picrosirius red staining of the hearts at 8 weeks post-MI. We find larger regions of fibrosis in miR-150 KO hearts than in WT MI controls, as reported previously [12]. We next observe reduced fibrosis post-MI in miR-150 KO;Sprr1ahypo/hypo hearts compared to miR-150 KO hearts (Figs. 4, 5A, B, and Supplementary Fig. 2). MiR-150 KO MI hearts also exhibit increased expression of fibrotic Col5a1, Col6a1, Col1a1, Col3a1, and Ctgf (Figs. 5C, D, and 6AC) compared to expression in WT controls, but miR-150 KO;Sprr1ahypo/hypo MI hearts exhibit decreased expression of these profibrotic genes (Figs. 5C, D, and 6AC) compared to miR-150 KO controls. Next, our in vivo protein analysis reveals significantly elevated levels of VIMENTIN and -SMA in miR-150 KO MI mouse hearts compared to WT controls and significantly decreased levels of VIMENTIN and -SMA in miR-150 KO;Sprr1ahypo/hypo hearts at 8 weeks post-MI compared to miR-150 KO controls (Fig. 6D, E, and Supplementary Fig. 3); this is consistent with the mRNA data for the profibrotic genes (Figs. 5C, D, and 6AC). Collectively, these results demonstrate for the first time that genetic knockdown of Sprr1a significantly attenuates adverse postinfarct remodeling mediated by miR-150 deletion.

Representative Massons trichrome staining (A, B) in heart sections of the peri-ischemic border area in the 6 experimental groups at 8 weeks post-MI and fibrosis quantification (C) in whole left ventricles (LVs). Fibrosis histology images from whole heart longitudinal sections (A: Scale bars: 1mm) and zoomed in images of the peri-ischemic border area (B: Scale bars: 100m). N=6 per group. Two-way ANOVA with Tukeys multiple comparison test. ***P<0.001 vs. sham for each genotype; #P<0.05 or ##P<0.01 vs. WT or miR-150 KO. Data are presented as the meanSEM.

Representative picrosirius red staining (A) from heart sections in the 6 experimental groups at 8 weeks post-MI and fibrosis quantification (B) in whole left ventricles (LVs). Fibrosis histology images from whole heart longitudinal sections (A: Scale bars: 1mm) are shown. N=6 per group. Two-way ANOVA with Tukeys multiple comparison test. ***P<0.001 vs. sham for each genotype; #P<0.05 or ###P<0.001 vs. WT or miR-150 KO. Data are presented as the meanSEM. qRT-PCR analysis of profibrotic Col5a1 (C) or Col6a1 (D) expression in ischemic areas from WT, miR-150 KO, and miR-150 KO;Sprr1ahypo/hypo mouse hearts at 8 weeks post-MI. Data are shown as the fold change of gene expression normalized to Gapdh. N=6 per group. Two-way ANOVA with Tukeys multiple comparison test. **P<0.01 or ***P<0.001 vs. sham for each genotype; #P<0.05, ##P<0.01, or ###P<0.001 vs. WT or miR-150 KO. Data are presented as the meanSEM.

qRT-PCR analysis of profibrotic Col1a1 (A), Col3a1 (B), or Ctgf (C) expression in ischemic areas from WT, miR-150 KO, and miR-150 KO;Sprr1ahypo/hypo mouse hearts at 8 weeks post-MI. Data are shown as the fold induction of gene expression normalized to Gapdh. N=46 per group. Two-way ANOVA with Tukeys multiple comparison test. *P<0.05 or ***P<0.001 vs. sham for each genotype; #P<0.05, ##P<0.01, or ###P<0.001 vs. WT or miR-150 KO. Data are presented as the meanSEM. D, E VIMENTIN protein levels were measured in ischemic areas from WT, miR-150 KO, and miR-150 KO;Sprr1ahypo/hypo mouse hearts at 8 weeks post-MI. N=56 per group. Two-way ANOVA with Tukeys multiple comparison test. *P<0.05 or **P<0.01 vs. sham for each genotype; #P<0.05 vs. WT or miR-150 KO. Data are presented as the meanSEM.

Because of the cardiac upregulation of miR-150 by Carv [11] concurrent with the downregulation of Sprr1a [13], and the downregulation of miR-150 in CFs isolated from TAC mice [15] concurrent with the upregulation of Sprr1a in CFs during MI [13], we next studied primary adult human CFs (HCFs) to test whether miR-150 and SPRR1A are inversely regulated in HCFs treated with Carv as well as HCFs subjected to H/R conditions. Indeed, SPRR1A is downregulated in HCFs subjected to H/R conditions after Carv treatment (Supplementary Fig. 4) concurrent with the upregulation of miR-150 [28]. We also observe that SPRR1A is increased in HCFs after H/R (Supplementary Fig. 4), consistent with our in vivo results in post-MI hearts and isolated CFs from ischemic myocardium [13]. Notably, we previously reported that miR-150 is downregulated in HCFs after H/R [28]. Together with other previous reports on miR-150 downregulation in H/R and MI [12] as well as I/R [29, 30], our results indicate that Sprr1a is a critical functional target of miR-150 in CFs.

Because Sprr1a expression is upregulated in CFs isolated from ischemic mouse hearts [13] concurrent with the downregulation of miR-150 in CFs isolated from TAC mice [15], and miR-150 negatively regulates mouse CF activation in vitro [15], we first confirmed whether a direct target of miR-150, SPRR1A is repressed by miR-150 in HCFs. Our loss-of-function studies indeed show that SPRR1A is increased after miR-150 inhibition in HCFs (Fig. 7A, B). We next investigated whether SPRR1A regulates HCF activation. We first observe that SPRR1A knockdown in HCFs decreases the expression of profibrotic ACTA2 and CTGF (Fig. 7C and Supplementary Fig. 5), and miR-150 knockdown increases the expression of ACTA2, CTGF, and POSTN (Supplementary Fig. 6).

HCFs were transfected with antimiR control or antimiR-150 (A, B) and with control scramble siRNA (si-control) or SPRR1A siRNA (si-SPRR1A) (C). qRT-PCR analyses for miR-150 (A) or SPRR1A (B, C) were then performed to check their expression after the indicated transfection. Data were normalized to U6 SNRNA (A) or GAPDH (B, C) and are expressed relative to controls. N=6 per group. Unpaired 2-tailed t-test. RNA interference with SPRR1A protects HCFs from the increased proliferation mediated by antimiR-150. HCFs were transfected as indicated and subjected to normoxia or hypoxia/reoxygenation (H/R). Bromodeoxyuridine (BrdU) assays were then performed under both normoxic (D, F) and H/R (E, F) conditions. The percentage of proliferating nuclei (green) was calculated by normalizing to the total nuclei (blue). N=6 per group. One-way ANOVA with Tukeys multiple comparison test. *P<0.05 or **P<0.01 vs. control: either si-control or antimiR control. #P<0.05 vs. anti-miR-150. Data are presented as the meanSEM.

To further assess the effects of SPRR1A knockdown, we examined HCF proliferation using bromodeoxyuridine assay. We find that compared to controls, SPRR1A knockdown decreased HCF proliferation (Fig. 7DF) under both normoxic and H/R conditions. This is consistent with our gene expression data, showing that HCFs with SPRR1A knockdown have decreased mRNA levels of S-phase marker PCNA, mitosis (M) marker AURKB, and G2/M-phase marker CCNB1 compared with controls (Supplementary Fig. 7). Moreover, our wound migration studies reveal that compared to controls, SPRR1A knockdown decreased HCF migration (Fig. 8AC) under both normoxic and H/R conditions. This is consistent with our gene expression data, showing that SPRR1A knockdown in HCFs subjected to H/R decreases mRNA levels of cell migration markers, CTHRC1 and TNC compared with controls (Supplementary Fig. 8). SPRR1A knockdown in HCFs also suppresses mRNA levels of CF differentiation markers, COL4A1, COL8A1, and SRF (Supplementary Fig. 9), as well as the protein levels of profibrotic -SMA and FIBRONECTION (Supplementary Fig. 10). Because TGF-1/SMAD signaling pathway plays a key role in CF activation, we next investigated the role of SPRR1A in the regulation of TGF-1 and SMADs. We observe that SPRR1A knockdown in HCFs subjected to H/R decreases mRNA levels of TGFB1, SMAD2, and SMAD3 compared with controls (Supplementary Fig. 11). This is consistent with our in vivo data, showing that Sprr1a knockdown in mice decreases Smad3 expression as well as mRNA and protein levels of TGF-1 compared with controls (Supplementary Figs. 12, 13). Our data thus suggest that SPRR1A is sufficient to increase HCF activation in part by activating TGF-1/SMAD signaling pathway.

AC HCFs were transfected and subjected to normoxia or hypoxia/reoxygenation (H/R) as indicated in Fig. 7DF. Scratch migration assays were then performed. RNA interference with SPRR1A protects HCFs from the increased migration mediated by antimiR-150. N=6 per group. Two-way repeated-measures ANOVA with Bonferroni post hoc test. One-way ANOVA with Tukeys multiple comparison test. *P<0.05, **P<0.01, or ***P<0.001 vs. control: either si-control or anti-miR control. #P<0.05 or ##P<0.01 vs. anti-miR-150. Data are presented as the meanSEM.

Finally, to establish the functional relationship between miR-150 and SPRR1A in HCF activation, we applied an antimiR/siRNA-based rescue strategy to validate the functional relevance of the direct miR-150 target SPRR1A. MiR-150 knockdown increases HCF proliferation (Fig. 7DF and Supplementary Fig. 7) and migration (Fig. 8AC and Supplementary Fig. 8), which are attenuated by siRNA against SPRR1A (Figs. 7DF, 8AC, Supplementary Figs. 7, 8). We also show that miR-150 knockdown increases the expression of profibrotic TGFB1, SMAD2, SMAD3, COL1A1, COL3A1, COL4A1, COL8A1, and SRF under normoxic and/or H/R conditions, which are attenuated by SPRR1A knockdown (Supplementary Figs. 9, 11, 14). Taken together, our data indicate that profibrotic SPRR1A is a key direct and functional target of miR-150 in HCFs and whole mouse hearts.

Read this article:
SPRR1A is a key downstream effector of MiR-150 during both ... - Nature.com