Abstract
Rationale: Heritable pulmonary arterial hypertension (HPAH) is primarily caused by mutations of the bone morphogenetic protein (BMP) type-II receptor (BMPR2). Recent identification of mutations in the downstream mediator Smad-8 (gene, SMAD9) was surprising, because loss of Smad-8 function in canonical BMP signaling is largely compensated by Smad-1 and -5. We therefore hypothesized that noncanonical pathways may play an important role in PAH.
Objectives: To determine whether HPAH mutations disrupt noncanonical Smad-mediated microRNA (miR) processing.
Methods: Expression of miR-21, miR-27a, and miR-100 was studied in pulmonary artery endothelial (PAEC) and pulmonary artery smooth muscle cells (PASMC) from explant lungs of patients with PAH.
Measurements and Main Results: SMAD9 mutation completely abrogated miR induction, whereas canonical signaling was only reduced by one-third. miR-21 levels actually decreased, suggesting that residual canonical signaling uses up or degrades existing miR-21. BMPR2 mutations also led to loss of miR induction in two of three cases. HPAH cells proliferated faster than other PAH or controls. miR-21 and miR-27a each showed antiproliferative effects in PAEC and PASMC, and PAEC growth rate after BMP treatment correlated strongly with miR-21 fold-change. Overexpression of SMAD9 corrected miR processing and reversed the hyperproliferative phenotype.
Conclusions: HPAH-associated mutations engender a primary defect in noncanonical miR processing, whereas canonical BMP signaling is partially maintained. Smad-8 is essential for this miR pathway and its loss was not complemented by Smad-1 and -5; this may represent the first nonredundant role for Smad-8. Induction of miR-21 and miR-27a may be a critical component of BMP-induced growth suppression, loss of which likely contributes to vascular cell proliferation in HPAH.
SMAD9 mutations have been associated with heritable pulmonary arterial hypertension (PAH) but they do not significantly alter canonical bone morphogenetic protein (BMP) signaling, so the mechanism whereby they contribute to the pathogenesis of PAH is unclear.
We show that SMAD9 is essential for BMP-mediated noncanonical microRNA processing. Heterozygous SMAD9 or BMPR2 mutations had a greater impact on microRNA processing than canonical BMP signaling. Gene replacement corrected dysregulated microRNA maturation and restored BMP-mediated growth suppression. These results define a new nonredundant role for Smad-8 and emphasize the importance of noncanonical BMP pathways in the pathogenesis of PAH.
Pulmonary arterial hypertension (PAH) is a sustained elevation of pulmonary artery pressure resulting from vascular remodeling and vessel narrowing that progressively obliterates the precapillary pulmonary arteries. It may be idiopathic (IPAH) or associated with an underlying condition, such as collagen vascular disease or congenital heart defect (APAH). After a recent reclassification, individuals with an identifiable genetic mutation or known family history are grouped together as heritable PAH (HPAH) (1). The primary genetic predisposition is a heterozygous mutation of the bone morphogenetic protein (BMP) receptor type-II (BMPR-II; gene symbol, BMPR2) (2–5). Recently, mutations affecting Smad-8 (gene symbol, SMAD9), a downstream mediator of BMP signaling, were also reported (6). Smad-8 is closely related to two other mediators, Smad-1 and -5, which are collectively known as the receptor Smads (R-Smads). In canonical BMP signaling, R-Smads are phosphorylated by the activated receptor complex, whereupon they associate with Smad-4 and translocate to the nucleus. Given this functional overlap between the R-Smads in canonical signaling, it was somewhat surprising that heterozygous Smad-8 mutations could cause PAH. We therefore sought to identify nonredundant roles for Smad-8 in the pulmonary vasculature, with an emphasis on noncanonical pathways.
MicroRNAs (miRs) are evolutionarily conserved small noncoding RNA molecules that negatively regulate gene expression. The primary miR transcript is processed into a 60–80 nucleotide precursor (pre-miR) molecule by the p68-Drosha microprocessor complex. This pre-miR is then exported from the nucleus and converted to the mature 18–24 nucleotide miR by DICER. Recently, it was shown that BMP signaling directly controls processing of a subset of miRs through a noncanonical role of R-Smads (7). BMP stimulation promotes Smad-1 and -5 to recruit the primary miR transcript into a p68-Drosha complex. Levels of pre- and mature-miR increase two- to threefold, whereas the primary miR transcript does not change, confirming this is a post-transcriptional mechanism of regulation. This noncanonical pathway is independent of Smad-4. miRs regulated by the R-Smads contain the consensus binding sequence CAGAC and include miR-21, miR-100, miR-199a-5p, and miR-27a (7, 8).
Although this pathway was first defined in pulmonary artery smooth muscle cells (PASMC) (7), several important questions pertinent to PAH remain unanswered. Pulmonary artery endothelial cells (PAEC) have not been studied, the role of Smad-8 was not examined, and previous studies were only performed with short-interfering RNA (siRNA) knockdown of Smad-1 and -5, so the effects of naturally occurring PAH mutations are unknown. Here we test whether Smad-8 is important in miR processing and analyze PAEC and PASMC from patients with PAH to determine whether HPAH mutations result in dysregulated miR biogenesis. Some of the results of this study have been previously reported in the form of an abstract (9, 10).
PAEC and PASMC from explant lungs of patients with PAH were isolated as previously described (11–13). Patients comprised four HPAH, one IPAH, and one APAH (Table 1). Control cells were isolated from failed donor lungs or purchased commercially (Lonza, Allendale, NJ). Cells were maintained in EGM-2 (PAEC) or SmGM-2 (PASMC) medium (Lonza) and used at passages four to nine. Analysis for mutations in BMPR2, SMAD9, and somatic chromosome abnormalities was performed as described (5, 14, 15). Three BMPR2 mutations and one SMAD9 were identified (Table 1). Studies were approved by the Cleveland Clinic Institutional Review Board and all subjects gave written informed consent.
| Subject | WHO Class | Age | Sex | Germline Mutation | Somatic Mutation (15) |
| HPAH-1 | 1.2 | 50 | F | BMPR2 deletion exons 4–5 | None |
| HPAH-2 | 1.2 | 41 | M | BMPR2 deletion exons 1–8 | Monosomy 13, 20% of PAEC |
| HPAH-3 | 1.2 | 26 | F | SMAD9 R294X | None |
| HPAH-4 | 1.2 | 33 | F | BMPR2 R321X | None |
| IPAH | 1.1 | 49 | M | None | None |
| APAH | 1.4 | 45 | F | None | Monosomy X, 46% of PAEC |
PAEC were plated at 8.8 × 104 cells per well and PASMC at 4.2 × 104 cells per well in 12-well plates. siRNA knockdown was initiated 24 hours after seeding. Transfections were performed using DharmaFECT-1 with gene-specific siRNA pools (Dharmacon, Lafayette, CO) at a final concentration of 100 nM. siGENOME nontargeting siRNA Pool-2 (Dharmacon) was used as a negative control. After a further 24 hours, cells were treated with 3-nM BMP4 (PASMC) or BMP9 (PAEC). This choice of ligands reflects known tissue-specific responsiveness; BMP9 is the only ligand that strongly induces BMP signaling in PAEC, whereas PASMC respond weakly to BMP9 and strongly to BMP2 and BMP4 (16). Each experimental condition was performed in triplicate in independent wells. Total RNA was extracted using miRNeasy mini kit (Qiagen, Valencia, CA), reverse transcribed, and analyzed by real-time quantitative polymerase chain reaction. The effects of overexpression or inhibition of miRs on cell proliferation was studied using 2,3-Bis(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carbox-anilide assay and live cell counts. Changes in endogenous miR-21 levels with BMP stimulation were followed by transfecting PAEC with γ-32P-labeled miR-21. Correction of the SMAD9 mutation in PAEC was performed by introducing an expression construct using nucleofection. Detailed methods for these experiments are provided in the online supplement.
Statistical significance of pairwise comparisons was determined by two-tailed t test, or with Mann-Whitney rank-sum test if the data were not normally distributed. Multiple-group data were tested using one-way analysis of variance (ANOVA) followed by pairwise multiple comparison procedures (Holm-Sidak method). All tests were performed using the SigmaStat v3.5 statistical package (Systat Software Inc., Chicago, IL).
To validate siRNA knockdown at the protein level, 50 μg of protein was resolved on a 10% sodium dodecyl sulfate polyacrylamide gel under denaturing conditions, semidry transferred to polyvinylidene fluoride membrane (Biorad, Hercules, CA) and immunoblotted with anti-BMPR2 (#B19720–050, 1:200; BD Biosciences, San Jose, CA,) or anti-SMAD4 (#9515, 1:500; Cell Signaling, Danvers, MA). An antiactin antibody (#sc-1615, 1:2000; Santa Cruz Biotechnology, Santa Cruz, CA) was used as a loading control. RNA-immunoprecipitation (IP) and protein co-IP experiments were performed as detailed in the online supplement.
Before studying the role of SMAD9, we first tested whether miR processing was likely to be significantly altered in HPAH by determining the effect of loss of BMPR-II function. siRNA knockdown of BMPR2 or SMAD4 in PAEC and PASMC each achieved at least an 80% reduction in mRNA and protein levels compared with a control nontargeting siRNA pool (see Figure E1 in the online supplement). BMP stimulation of control cells induced a twofold to three fold increase in pre- and mature-miR-21 levels, which was completely abrogated by BMPR2 knockdown (Figure 1). miR-27a and miR-100 show a similar pattern (see Figure E2). In contrast, siRNA knockdown of SMAD4 had no effect on these miRs, whereas induction of ID1 and miR-181b, downstream targets known to be regulated by canonical BMP signaling (7, 17), was blocked by SMAD4 knockdown (Figure 1; see Figure E2). These results demonstrate that BMPR-II is essential for Smad-mediated miR processing and confirm that siRNA knockdown of SMAD4 effectively distinguishes between the canonical and noncanonical pathways.
Figure 1. Expression of miR-21 and its precursors in pulmonary artery endothelial cells (PAEC) (A) and pulmonary artery smooth muscle cells (PASMC) (B) after short-interfering RNA knockdown (KD) of BMPR2 or SMAD4. Bone morphogenetic protein (BMP) stimulation leads to an increase in pre- and mature–miR-21 in control and SMAD4-KD but not BMPR2-KD cells; ** P < 0.001. Basal level of mature miR-21 is significantly lower in BMPR2-KD PAEC than control cells; § P < 0.001. miR = microRNA.
[More] [Minimize]We then sought to determine whether Smad-8 promotes miR processing in the same manner as Smad-1 and -5 and if so, whether SMAD9 mutations would dysregulate this process. RNA-IP experiments confirmed that all three miRs bind to Smad-8 in a BMP-dependent manner (Figure 2) and protein co-IP confirmed association of Smad-8 with p68 (see Figure E3). Because previous studies only examined dual-knockdown of SMAD1 and SMAD5 in PASMC (7), we also studied the effects of knocking down these genes individually in both PAEC and PASMC. siRNA pools targeting individual SMADs were confirmed to be gene-specific (see Figure E4). Functional overlap between the R-Smads engenders some redundancy in the canonical BMP signaling pathway. Consistent with this, siRNA knockdown of individual R-Smads led to attenuation but not complete inhibition of canonical BMP signaling (Figure 3, ID1). Notably, SMAD9 knockdown had the least impact. In marked contrast, miR-21 processing was completely blocked by knockdown of any individual R-Smad, including SMAD9, in both cell types (Figure 3). Similar results were obtained for miR-100, -27a, and -199a-5p (see Figure E5). These results demonstrate that Smad-8 is essential for the noncanonical BMP-mediated miR processing pathway. Furthermore, unlike the canonical signaling pathway, loss of any individual R-Smad is not compensated by the remaining two, leading to a dominant–negative effect on miR biogenesis.
Figure 2. RNA-immunoprecipitation (IP) of Flag-tagged Smad-8 in pulmonary artery endothelial cells. (A) Quantitative polymerase chain reaction demonstrates little or no primary microRNA (miR) bound to Smad-8 in vehicle-treated cells and significant induction of binding with bone morphogenetic protein (BMP)-9 treatment. Pre–miR-16 is not regulated by BMPs (see Figure E15) and showed no detectable binding to Smad-8. Control IPs with nonspecific IgG were negative for all miRs. No reverse transcriptase (RT) denotes a negative control with no RT, confirming there is no DNA contamination. (B) Western blot analysis with anti-Flag antibody confirms specificity of the pulldown and equal efficiency in vehicle and BMP9-treated samples.
[More] [Minimize]
Figure 3. Short-interfering RNA knockdown (KD) of any individual R-Smad (SMAD1, SMAD5, or SMAD9) leads to a paradoxical decrease in miR-21 levels after bone morphogenetic protein (BMP) stimulation; ¶P ≤ 0.002. Similar effects are seen in pulmonary artery endothelial cells (PAEC) (A) and pulmonary artery smooth muscle cells (PASMC) (B). In contrast, expression of ID1, a downstream target of canonical BMP signaling, is still induced in both cell types; **P < 0.001. Basal level of mature miR-21 is significantly lower in Smad-KD PAEC than control cells; §P ≤; 0.006. miR = microRNA.
[More] [Minimize]Knockdown of individual R-Smads not only blocked miR induction but the levels of pre– and mature–miR-21 actually decreased by up to 70% compared with controls cells (Figure 3). This effect was seen in PAEC and PASMC and was BMP-dependent. A timecourse in SMAD1 knockdown PAEC showed that pre– and mature–miR-21 levels gradually declined for 16 hours post-BMP stimulation, then plateaued at around 40% of pretreatment levels until 20 hours and subsequently began to rise again, although they had still not returned to pretreatment levels at the 48-hour time point (Figure 4). Primary miR-21 levels remained relatively constant.
Figure 4. Timecourse of miR-21 expression in control (open symbols) and SMAD1-knockdown (KD) (solid symbols) pulmonary artery endothelial cells after bone morphogenetic protein-9 treatment at zero hours. Data are normalized to untreated cells at time zero. Primary miR-21 levels in both control and SMAD1-KD cells were essentially constant around 1; control cells are omitted for clarity. miR = microRNA.
[More] [Minimize]These results suggest that blocking noncanonical miR-21 induction unmasks a decrease in existing miR-21 driven by canonical BMP-signaling. To test this hypothesis, we first performed siRNA knockdown of p68 to block noncanonical miR processing independently of the Smads. BMP stimulation of p68-knockdown PAEC led to a threefold reduction in pre–miR-21 and a fivefold decrease in the mature miR, whereas inhibition of the canonical and noncanonical pathways with SMAD1/4 double knockdown gave no overall change (see Figure E6). We then transfected control and siRNA-treated PAEC with radiolabeled mature miR-21. In this way we could isolate what happens to the miR-21 already present at the time of stimulation, because any additional miR-21 newly induced by BMP9 would not be labeled. BMP stimulation of control PAEC led to a 50% decrease in radiolabeled miR-21 after 24 hours, whereas no change was seen in SMAD4-knockdown cells (Figure 5). This further supports our hypothesis that canonical signaling by Smad-4 leads to a decrease in miR-21 levels. Cells with siRNA knockdown of SMAD1 or SMAD9 showed decreases comparable with control cells (Figure 5), consistent with our previous evidence that they retain some canonical BMP signaling (Figure 3). Treatment with α-amanitin, puromycin, or cyclohexamide blocked the decrease, suggesting that de novo RNA transcription and protein synthesis is required. This effect may be specific to miR-21, because there was no decrease in miR-27a (Figure 5).
Figure 5. Autoradiograph and quantitation of exogenous 32P-labeled miR-21 levels 24 hours after transfection, normalized to 5S ribosomal RNA to control for the amount of total RNA loaded. Densitometric ratios were then normalized to the untreated control in each experiment. Bone morphogenetic protein (BMP) stimulation leads to a decrease in miR-21 levels, which is blocked by SMAD4 short-interfering RNA knockdown (KD) but not SMAD1 or SMAD9, suggesting it is mediated by the canonical pathway. There is no decrease in cells treated with inhibitors of transcription (α-amanitin) or translation (puromycin or cyclohexamide). miR-27a levels are unchanged with BMP treatment. miR = microRNA.
[More] [Minimize]Most PAH-associated BMPR2 mutations lead to haploinsufficiency and approximately 50% of normal gene activity (4). In contrast, siRNA knockdown reduces mRNA levels by 80–90% (see Figure E1). Thus, it was important to determine the relevance of our results in siRNA-treated cells to the in vivo human disease process. We analyzed miR processing in cells isolated from explant lungs of four patients with HPAH with heterozygous germline mutations of BMPR2 or SMAD9 (Table 1). Three of these cases (hereafter referred to as Group-A) showed no significant induction of miR-21, miR-27a, or miR-100 in PAEC (Figure 6). Indeed, consistent with our siRNA data, miR-21 levels in these cases actually decreased when stimulated with BMP9. The fourth case (HPAH-4) showed significant induction of all three miRs, as did PAEC from two patients without mutations (Figure 6). In contrast, ID1 expression was significantly up-regulated in Group-A cases, although the fold-change was smaller than in controls and other patients (Figure 6). Likewise, miR-181b was significantly induced in two of the three Group-A patients (see Figure E7). It was also notable that basal levels of miR-21 were significantly lower in all PAH cases, with the largest reduction in Group-A cells (see Figure E8). Similar results were obtained in PAH PASMC, with the interesting exception that induction of miR-199a-5p was completely absent in R321X (see Figure E9).
Figure 6. Fold-change of microRNAs (miR) and ID1 in patient pulmonary artery endothelial cells (PAEC) when treated with bone morphogenetic protein (BMP)-9. Three of four patients with germline mutations show a significant decrease in miR-21 relative to basal levels (§ P < 0.01) and nonsignificant changes in miR-27a and miR-100, whereas ID1 is significantly increased (** P < 0.001) albeit with a smaller fold-change relative to control cells. The remaining patients showed significant induction of all miRs and ID1, with fold-changes that did not differ from control cells. APAH = associated pulmonary arterial hypertension; IPAH = idiopathic pulmonary arterial hypertension.
[More] [Minimize]To investigate the physiologic effects of altered miR regulation in the pulmonary artery, we performed a growth curve analysis of control PAEC transfected with either pre–miR-21 or a miR-21 inhibitor molecule and counted live cells at each time point. Overexpression of pre–miR-21 significantly reduced cell growth and was partially dose dependent (see Figure E10). Conversely, inhibition of endogenous miR-21 with an anti-miR significantly accelerated cell growth, again with some evidence of a dose response. One-way ANOVA of cell counts at 72 hours was statistically significant (P < 0.001), with highly significant pairwise differences between pre-miR or anti-miR versus control. Cotransfection of equimolar amounts of pre- and anti-miR was not significantly different from untransfected cells, demonstrating this is a specific effect of miR-21 that can be competitively reversed. Negative control pre- and anti-miR constructs also did not differ from control cells (data not shown).
We then analyzed the growth parameters of patient PAEC and PASMC using 2,3-Bis(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carbox-anilide assays. Cells without mutations proliferated at the same rate as controls, whereas the three Group-A cases with abnormal miR regulation proliferated significantly faster (see Figure E11A). Cells from case HPAH-4 (BMPR2-R321X) were intermediate between these two groups. Because R321X is predicted to cause nonsense-mediated RNA decay, leading to haploinsufficiency, we hypothesized that perhaps the intermediate phenotype reflected higher expression of the wild-type allele. Quantitative polymerase chain reaction analysis confirmed that BMPR2 expression was 75% of control, despite complete nonsense-mediated RNA decay of the mutant allele (see Figure E12).
BMP stimulation led to a decrease in PAEC and PASMC proliferation, as previously reported (18). However, under these conditions Group-A cells still grew faster than unstimulated control cells (see Figure E11B). Indeed, the proliferation rate was highly correlated with the BMP-induced miR-21 fold-change (R2 = 0.99; P < 0.001) (Figure 7). Overexpression of miR-21 induced further growth suppression and restored a proliferation rate comparable with unstimulated control cells (see Figure E12). This was not caused by increased apoptosis (data not shown). miR-27a showed a similar effect but with less suppression of control and nonmutant PAH cells (see Figure E12). Overexpression of miR-100 was slightly proproliferative but the magnitude of the effect was very small (data not shown).
Figure 7. The growth rate of patient and control pulmonary artery endothelial cells (PAEC) stimulated with bone morphogenetic protein (BMP)-9 strongly correlates with the change in their miR-21 levels. miR = microRNA.
[More] [Minimize]Although overexpression of miR-21 or miR-27a could normalize proliferation rates of Group-A PAEC, it does not restore BMP-responsiveness and the cells still proliferate faster than BMP-treated controls. We therefore sought to restore SMAD9 levels and determine whether this would correct both the miR processing defect and cell proliferation. Overexpression of wild-type SMAD9 in control PAEC had no major effects; canonical signaling and miR processing were unchanged or slightly increased and cell proliferation over a 72-hour period post-transfection was not significantly different than the mock-transfected or green fluorescent protein-transfected cells (Figure 8). Introduction of this construct into the SMAD9-mutant PAEC not only normalized their baseline proliferation rate to that of control cells but also completely restored the additional growth suppressive effects of BMP9 stimulation (Figure 8). MiR fold-change was also normalized (Figure 8), but interestingly, the basal level of miR-21 remained significantly lower than control cells (see Figure E14).
Figure 8. (A) Overexpression of SMAD9 normalizes the bone morphogenetic protein (BMP)–induced fold-change of all three microRNAs (miRs) in pulmonary artery endothelial cells (PAEC) with the SMAD9-R294X mutation; * P < 0.05 in analysis of variance post hoc pairwise test relative to fold-change in mock-treated control cells; ns = not significant. (B) Baseline proliferation rate and growth suppression in response to BMP9 are also completely corrected (post hoc P values vs. control at 72 h: R294X + GFP, P < 0.001; R294X + SMAD9, P = 0.138; R294X + SMAD9 + BMP9 vs. control + BMP9, P = 0.39). Mock-treated cells and control-GFP are omitted for clarity. GFP = green fluorescent protein.
[More] [Minimize]Alterations in BMP signaling clearly play an important role in the pathogenesis of PAH; all known HPAH mutations affect genes in the pathway (2–5, 19–21) and there is evidence for down-regulation of BMPR-II in the lungs of other types of human PAH (22) and hypoxia and monocrotaline animal models (23–25). It was therefore surprising when a SMAD9 mutation was identified (6), because loss of Smad-8 function is largely compensated by the overlapping functions of Smad-1 and -5 (26). Thus, the SMAD9 knockout mouse is viable, whereas SMAD1 and SMAD5 knockouts are both embryonic lethal (27–29). The SMAD9 knockout does show evidence of spontaneous pulmonary vascular remodeling, supporting a role in PAH (30). We therefore hypothesized that the functions of BMP signaling most critical to PAH may be those where Smad-8 function is nonredundant. Furthermore, mutations in SMAD4, the co-Smad that mediates canonical BMP signaling, predispose to juvenile polyposis and colorectal cancer (31) but are not known to cause PAH, suggesting that noncanonical pathways may be especially important in the pathogenesis of PAH. The recent identification of a miR processing pathway mediated by Smad-1 and -5 and independent of Smad-4 (7) was an attractive candidate fulfilling these criteria, but the importance of Smad-8 has not been investigated. Here we tested whether Smad-8 is required for miR processing and analyzed the effects of naturally occurring heterozygous BMPR2 and SMAD9 mutations in explant cells from the lungs of patients with PAH.
Our data clearly show that Smad-8 is essential for miR processing, with complete loss of induction in SMAD9 siRNA knockdown cells. Remarkably, the same effect was seen in a patient with only a heterozygous mutation and this defect could be corrected by overexpressing wild-type SMAD9. Canonical BMP signaling remained relatively intact, with less than a 50% reduction in fold-change of downstream target genes. Because Smad-1 and -5 had previously only been investigated in a combined siRNA knockdown, we also tested specific knockdown of the genes individually and found that Smad-1, -5, and -8 are required for miR induction. Given this apparently dominant-negative effect of losing a single R-Smad, we hypothesized that the miR processing pathway requires formation of a complex comprising all three R-Smads to bind the p68/Drosha complex. Canonical BMP signaling is a precedent for this, where two phosphorylated R-Smads form a heterotrimeric complex with Smad-4. However, our efforts to prove this through coimmunoprecipitation of endogenous Smad-p68 complexes have been hampered by the lack of a specific Smad-8 antibody. These data also suggest that interaction of the R-Smads with Smad-4 may be favored kinetically over p68-Drosha, and so noncanonical miR processing is competitively disadvantaged by SMAD9 mutation. Further studies are required to fully characterize the nature of the complex molecular interactions involved in these two pathways.
Another remarkable finding was that dysregulation of miR processing, either by siRNA knockdown or germline mutation, was associated with a decrease in miR-21 levels on ligand stimulation. This led us to hypothesize that induction of miR-21 by the noncanonical pathway is balanced by a decrease mediated by canonical BMP signaling. In normal PAEC the induction outweighs the decrease, whereas mutations that block miR processing but leave canonical signaling partially intact lead to an overall reduction in miR-21 levels. Transfection of radiolabeled miR-21 into control and various siRNA-treated PAEC supported this hypothesis. It is presently unclear whether the decrease represents active degradation of miR-21 or if it is used up by a target mRNA induced by canonical BMP signaling. Physiologically, however, it may be critically important, because miR-21 was clearly antiproliferative in PAEC and SMC and their growth rate after BMP stimulation correlated strongly with the change in their miR-21 levels.
Baseline proliferation rates of cells from all four HPAH cases were significantly higher than controls or mutation-negative patients. Notably, the SMAD9 mutation case clustered tightly with the two BMPR2 mutations that dysregulated miR processing, despite their differential effects on canonical signaling. The fourth case was intriguing, with an intermediate phenotype correlating with elevated wild-type BMPR2 expression. Up-regulation of the wild-type allele has been proposed as a mechanism for reduced penetrance in HPAH (32), but in this case it was not protective. One of the limitations of working with explant cells is they derive from late-stage disease, have accumulated many changes, and may be influenced by previous drug treatments. Yet, overexpression of SMAD9 in mutant PAEC normalized their baseline proliferation rate, corrected miR processing, and restored BMP-mediated growth suppression to levels comparable with mock-transfected control cells. This suggests that despite the many other changes in these cells from late-stage disease lungs, their hyperproliferation is almost completely accounted for by the primary mutation and can be corrected by gene replacement.
Cells from patients with PAH also had lower basal levels of miR-21 than controls, consistent with recent data on whole-lung sections (33). Interestingly, this was not corrected by overexpression of SMAD9, even though the BMP-induced fold-change was normalized, suggesting that the mechanism may be independent of the BMP pathway. Further support for a pathogenic role for miR-21 comes from the rat monocrotaline model of pulmonary hypertension where it is also down-regulated (33). Indeed, given that monocrotaline leads to reduced BMPR2 expression and Smad signaling (24, 25, 34), the molecular basis for miR-21 down-regulation in this model may be analogous to our data in human HPAH. Together, these data raise the possibility that miR replacement might have therapeutic potential in PAH. However, considerable caution is needed because in many other tissues, miR-21 is widely considered to be oncogenic, with evidence that overexpression leads to increased proliferation and decreased apoptosis (35, 36). In the lung, overexpression of miR-21 in myofibroblasts has been implicated in the pathogenesis of idiopathic pulmonary fibrosis (37). Thus, any therapeutic strategies require very specific targeting to the pulmonary vasculature to avoid potentially serious adverse consequences. Additionally, our data show that multiple miRs are dysregulated in HPAH, further complicating miR-based therapeutic approaches. In contrast, we show that overexpressing wild-type SMAD9 in mutant cells effectively restored normal processing for multiple different miRs. Overexpression in control cells did not significantly alter BMP signaling or cell proliferation, suggesting that “overcorrection” is not detrimental. This suggests that approaches to correct HPAH mutations by increasing the availability of functional protein and correcting receptor stoichiometry (38, 39) might achieve better physiologic results by simultaneously correcting multiple different canonical and noncanonical functions of BMP signaling with fewer adverse consequences.
In conclusion, our data show that heterozygous mutations of BMPR2 or SMAD9 are sufficient to abrogate noncanonical BMP-mediated miR processing, in contrast with canonical BMP signaling, where the response is only attenuated. This difference is particularly marked for SMAD9 mutations, defining possibly the first nonredundant role for Smad-8. Cell proliferation was strongly correlated with miR-21 levels, suggesting that aberrant miR processing may play an important role in the pathogenesis of PAH.
The authors thank the patients and their families who consented to use of their tissue in this study, Michelle Koo and Jacqueline Sharp for subject recruitment and tissue acquisition, The Cleveland Clinic Pathobiology Tissue Sample and Cell Culture Core for samples and cells, and Craig Homer and Christine McDonald for assistance with the RNA-IP. Tissue samples provided by Marlene Rabinovitch, Stanford University, under the Pulmonary Hypertension Breakthrough Initiative (PHBI). Funding for the PHBI is provided by the Cardiovascular Medical Research and Education Fund.
| 1. | Simonneau G, Robbins IM, Beghetti M, Channick RN, Delcroix M, Denton CP, Elliott CG, Gaine SP, Gladwin MT, Jing ZC, et al.. Updated clinical classification of pulmonary hypertension. J Am Coll Cardiol 2009;54:S43–S54. |
| 2. | Lane KB, Machado RD, Pauciulo M, Thomson JR, Phillips JA, Loyd JE, Nichols WC, Trembath RC; The International PPH Consortium. Heterozygous germline mutations in BMPR2, encoding a TGF-beta receptor, cause familial primary pulmonary hypertension. Nat Genet 2000;26:81–84. |
| 3. | Deng Z, Morse JH, Slager SL, Cuervo N, Moore KJ, Venetos G, Kalachikov S, Cayanis E, Fischer SG, Barst RJ, et al.. Familial primary pulmonary hypertension (gene PPH1) is caused by mutations in the bone morphogenetic protein receptor-II gene. Am J Hum Genet 2000;67:737–744. |
| 4. | Machado RD, Aldred MA, James V, Harrison RE, Patel B, Schwalbe EC, Gruenig E, Janssen B, Koehler R, Seeger W, et al.. Mutations of the TGF-beta type II receptor BMPR2 in pulmonary arterial hypertension. Hum Mutat 2006;27:121–132. |
| 5. | Aldred MA, Vijayakrishnan J, James V, Soubrier F, Gomez-Sanchez MA, Martensson G, Galie N, Manes A, Corris P, Simonneau G, et al.. BMPR2 gene rearrangements account for a significant proportion of mutations in familial and idiopathic pulmonary arterial hypertension. Hum Mutat 2006;27:212–213. |
| 6. | Shintani M, Yagi H, Nakayama T, Saji T, Matsuoka R. A new nonsense mutation of SMAD8 associated with pulmonary arterial hypertension. J Med Genet 2009;46:331–337. |
| 7. | Davis BN, Hilyard AC, Lagna G, Hata A. Smad proteins control drosha-mediated microRNA maturation. Nature 2008;454:56–61. |
| 8. | Davis BN, Hilyard AC, Nguyen PH, Lagna G, Hata A. Smad proteins bind a conserved RNA sequence to promote microRNA maturation by drosha. Mol Cell 2010;39:373–384. |
| 9. | Drake KM, Comhair SA, Erzurum SC, Aldred MA. MicroRNA processing is dysregulated by mutations associated with pulmonary arterial hypertension. Proc Am Thorac Soc 2011;8:205. |
| 10. | Drake KM, Comhair SAA, Erzurum SC, Aldred MA. Dysregulated microRNA processing in pulmonary arterial hypertension contributes to endothelial and smooth muscle cell hyperproliferation. Am J Respir Crit Care Med 2011;183:A2523. |
| 11. | Masri FA, Xu W, Comhair SA, Asosingh K, Koo M, Vasanji A, Drazba J, Anand-Apte B, Erzurum SC. Hyperproliferative apoptosis-resistant endothelial cells in idiopathic pulmonary arterial hypertension. Am J Physiol Lung Cell Mol Physiol 2007;293:L548–L554. |
| 12. | Xu W, Koeck T, Lara AR, Neumann D, DiFilippo FP, Koo M, Janocha AJ, Masri FA, Arroliga AC, Jennings C, et al.. Alterations of cellular bioenergetics in pulmonary artery endothelial cells. Proc Natl Acad Sci USA 2007;104:1342–1347. |
| 13. | Aytekin M, Comhair SA, de la Motte C, Bandyopadhyay SK, Farver CF, Hascall VC, Erzurum SC, Dweik RA. High levels of hyaluronan in idiopathic pulmonary arterial hypertension. Am J Physiol Lung Cell Mol Physiol 2008;295:L789–L799. |
| 14. | Asosingh K, Aldred MA, Vasanji A, Drazba J, Sharp J, Farver C, Comhair SA, Xu W, Licina L, Huang L, et al.. Circulating angiogenic precursors in idiopathic pulmonary arterial hypertension. Am J Pathol 2008;172:615–627. |
| 15. | Aldred MA, Comhair SA, Varella-Garcia M, Asosingh K, Xu W, Noon GP, Thistlethwaite PA, Tuder RM, Erzurum SC, Geraci MW, et al.. Somatic chromosome abnormalities in the lungs of patients with pulmonary arterial hypertension. Am J Respir Crit Care Med 2010;182:1153–1160. |
| 16. | Upton PD, Davies RJ, Trembath RC, Morrell NW. Bone morphogenetic protein (BMP) and activin type II receptors balance BMP9 signals mediated by activin receptor-like kinase-1 in human pulmonary artery endothelial cells. J Biol Chem 2009;284:15794–15804. |
| 17. | Wang B, Hsu SH, Majumder S, Kutay H, Huang W, Jacob ST, Ghoshal K. TGFbeta-mediated upregulation of hepatic miR-181b promotes hepatocarcinogenesis by targeting TIMP3. Oncogene 2010;29:1787–1797. |
| 18. | Yang X, Long L, Southwood M, Rudarakanchana N, Upton PD, Jeffery TK, Atkinson C, Chen H, Trembath RC, Morrell NW. Dysfunctional Smad signaling contributes to abnormal smooth muscle cell proliferation in familial pulmonary arterial hypertension. Circ Res 2005;96:1053–1063. |
| 19. | Trembath RC, Thomson JR, Machado RD, Morgan NV, Atkinson C, Winship I, Simonneau G, Galie N, Loyd JE, Humbert M, et al.. Clinical and molecular genetic features of pulmonary hypertension in patients with hereditary hemorrhagic telangiectasia. N Engl J Med 2001;345:325–334. |
| 20. | Harrison RE, Berger R, Haworth SG, Tulloh R, Mache CJ, Morrell NW, Aldred MA, Trembath RC. Transforming growth factor-beta receptor mutations and pulmonary arterial hypertension in childhood. Circulation 2005;111:435–441. |
| 21. | Mache CJ, Gamillscheg A, Popper HH, Haworth SG. Early-life pulmonary arterial hypertension with subsequent development of diffuse pulmonary arteriovenous malformations in hereditary haemorrhagic telangiectasia type 1. Thorax 2008;63:85–86. |
| 22. | Atkinson C, Stewart S, Upton PD, Machado R, Thomson JR, Trembath RC, Morrell NW. Primary pulmonary hypertension is associated with reduced pulmonary vascular expression of type II bone morphogenetic protein receptor. Circulation 2002;105:1672–1678. |
| 23. | Takahashi H, Goto N, Kojima Y, Tsuda Y, Morio Y, Muramatsu M, Fukuchi Y. Downregulation of type II bone morphogenetic protein receptor in hypoxic pulmonary hypertension. Am J Physiol Lung Cell Mol Physiol 2006;290:L450–L458. |
| 24. | Morty RE, Nejman B, Kwapiszewska G, Hecker M, Zakrzewicz A, Kouri FM, Peters DM, Dumitrascu R, Seeger W, Knaus P, et al.. Dysregulated bone morphogenetic protein signaling in monocrotaline-induced pulmonary arterial hypertension. Arterioscler Thromb Vasc Biol 2007;27:1072–1078. |
| 25. | Ramos MF, Lame MW, Segall HJ, Wilson DW. Smad signaling in the rat model of monocrotaline pulmonary hypertension. Toxicol Pathol 2008;36:311–320. |
| 26. | Orvis GD, Jamin SP, Kwan KM, Mishina Y, Kaartinen VM, Huang S, Roberts AB, Umans L, Huylebroeck D, Zwijsen A, et al.. Functional redundancy of TGF-beta family type I receptors and receptor-Smads in mediating anti-mullerian hormone-induced mullerian duct regression in the mouse. Biol Reprod 2008;78:994–1001. |
| 27. | Lechleider RJ, Ryan JL, Garrett L, Eng C, Deng C, Wynshaw-Boris A, Roberts AB. Targeted mutagenesis of Smad1 reveals an essential role in chorioallantoic fusion. Dev Biol 2001;240:157–167. |
| 28. | Tremblay KD, Dunn NR, Robertson EJ. Mouse embryos lacking Smad1 signals display defects in extra-embryonic tissues and germ cell formation. Development 2001;128:3609–3621. |
| 29. | Chang H, Huylebroeck D, Verschueren K, Guo Q, Matzuk MM, Zwijsen A. Smad5 knockout mice die at mid-gestation due to multiple embryonic and extraembryonic defects. Development 1999;126:1631–1642. |
| 30. | Huang Z, Wang D, Ihida-Stansbury K, Jones PL, Martin JF. Defective pulmonary vascular remodeling in Smad8 mutant mice. Hum Mol Genet 2009;18:2791–2801. |
| 31. | Howe JR, Roth S, Ringold JC, Summers RW, Jarvinen HJ, Sistonen P, Tomlinson IP, Houlston RS, Bevan S, Mitros FA, et al.. Mutations in the SMAD4/DPC4 gene in juvenile polyposis. Science 1998;280:1086–1088. |
| 32. | Hamid R, Cogan JD, Hedges LK, Austin E, Phillips JA, Newman JH, Loyd JE. Penetrance of pulmonary arterial hypertension is modulated by the expression of normal BMPR2 allele. Hum Mutat 2009;30:649–654. |
| 33. | Caruso P, MacLean MR, Khanin R, McClure J, Soon E, Southgate M, MacDonald RA, Greig JA, Robertson KE, Masson R, et al.. Dynamic changes in lung microRNA profiles during the development of pulmonary hypertension due to chronic hypoxia and monocrotaline. Arterioscler Thromb Vasc Biol 2010;30:716–723. |
| 34. | Long L, Crosby A, Yang X, Southwood M, Upton PD, Kim DK, Morrell NW. Altered bone morphogenetic protein and transforming growth factor-beta signaling in rat models of pulmonary hypertension: potential for activin receptor-like kinase-5 inhibition in prevention and progression of disease. Circulation 2009;119:566–576. |
| 35. | Krichevsky AM, Gabriely G. miR-21: a small multi-faceted MA. J Cell Mol Med 2009;13:39–53. |
| 36. | Selcuklu SD, Donoghue MT, Spillane C. miR-21 as a key regulator of oncogenic processes. Biochem Soc Trans 2009;37:918–925. |
| 37. | Liu G, Friggeri A, Yang Y, Milosevic J, Ding Q, Thannickal VJ, Kaminski N, Abraham E. miR-21 mediates fibrogenic activation of pulmonary fibroblasts and lung fibrosis. J Exp Med 2010;207:1589–1597. |
| 38. | Nasim MT, Ghouri A, Patel B, James V, Rudarakanchana N, Morrell NW, Trembath RC. Stoichiometric imbalance in the receptor complex contributes to dysfunctional BMPR-II mediated signalling in pulmonary arterial hypertension. Hum Mol Genet 2008;17:1683–1694. |
| 39. | Sui X, Li D, Qiu H, Gaussin V, Depre C. Activation of the bone morphogenetic protein receptor by h11kinase/HSP22 promotes cardiac cell growth and survival. Circ Res 2009;104:887–895. |
Supported in part by AHA National Scientist Development Grant 0835146N;
Author contributions: K.M.D. and M.A.A. designed the study, analyzed the data, and wrote the manuscript. K.M.D., D.Z., and P.H. performed the experiments. L.M., L.W., S.A.C., and S.C.E. isolated and characterized the cells. All authors have contributed to the manuscript and critically evaluated the final version.
This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org
Originally Published in Press as DOI: 10.1164/rccm.201106-1130OC on September 15, 2011
Author disclosures
