Noggin inhibits hypoxia-induced proliferation by targeting store-operated calcium entry and transient receptor potential cation channels

Abstract

Abnormally elevated bone morphogenetic protein 4 (BMP4) expression and mediated signaling play a critical role in the pathogenesis of chronic hypoxia-induced pulmonary hypertension (CHPH). In this study, we investigated the expression level and functional significance of four reported naturally occurring BMP4 antagonists, noggin, follistatin, gremlin1, and matrix gla protein (MGP), in the lung and distal pulmonary arterial smooth muscle cell (PASMC). A 21-day chronic hypoxic (10% O₂) exposure rat model was utilized, which has been previously shown to successfully establish experimental CHPH. Among the four antagonists, noggin, but not the other three, was selectively downregulated by hypoxic exposure in both the lung tissue and PASMC, in correlation with markedly elevated BMP4 expression, suggesting that the loss of noggin might account for the hypoxia-triggered BMP4 signaling transduction. Then, by using treatment of extrogenous recombinant noggin protein, we further found that noggin significantly normalized 1) BMP4-induced phosphorylation of cellular p38 and ERK1/2; 2) BMP4-induced phosphorylation of cellular JAK2 and STAT3; 3) hypoxia-induced PASMC proliferation; 4) hypoxia-induced store-operated calcium entry (SOCE), and 5) hypoxia-increased expression of transient receptor potential cation channels (TRPC1 and TRPC6) in PASMC. In combination, these data strongly indicated that the hypoxia-suppressed noggin accounts, at least partially, for hypoxia-induced excessive PASMC proliferation, while restoration of noggin may be an effective way to inhibit cell proliferation by suppressing SOCE and TRPC expression.

pulmonary hypertension (PH) is a disease characterized by a list of functional and structural changes in the pulmonary vasculature that lead to enhanced distal pulmonary arterial (PA) contraction and remodeling, eventually causing heart failure. During the disease process, remodeling of small vessels in the lung, due to abnormal proliferation and migration of vascular smooth muscle cells and endothelium cells, is well studied and accepted (10).

Bone morphogenetic proteins (BMPs) belong to a subgroup of the transforming growth factor-β (TGF-β) superfamily, which are a group of growth factors originally discovered by their ability to induce the formation of bone and cartilage. Recently, evidence strongly indicated that dysregulated BMP signaling is involved in the pathogenesis of PH (9, 11). BMP4, a member of BMP ligands, has been found selectively upregulated by chronic hypoxia in the lungs and plays an important role during the development of chronic hypoxia-induced pulmonary hypertension (CHPH) by regulating the proliferation and migration of pulmonary arterial smooth muscle cells (PASMC) (11, 13, 20). In mechanisms, BMP4-mediated signaling transduction is mainly regulated by two groups of molecules, the typical receptors and the extracellular soluble antagonists (3). BMP4 transduces signals by binding to type II serine-threonine kinase receptors (44), which then causes recruitment and phosphorylation of type I receptors, leading to activation of a number of cellular kinases (18).

The group of BMP antagonists belongs to naturally secreted endogenous proteins that can block the BMP ligand-receptor interaction to inhibit the BMP signaling transduction. Increasing evidence demonstrated that dysfunction of these proteins leads to excessive BMP activity and signaling transduction, which is present and may account for the development of numerous diseases. In detail, gremlin1 specifically binds to BMP2, 4, and 7 and inhibits their actions on the downstream signaling (14, 23). Evidence suggests that gremlin1 is upregulated in lungs isolated from 2-day alveolar hypoxia-exposed mice (6). The increased gremlin1 expression mostly localizes in the pulmonary endothelium, but not smooth muscle (2). Deletion of gremlin1 increases cell proliferation and migration responses in mouse embryonic fibroblasts (7). Noggin, follistatin, and matrix gla protein (MGP) are all demonstrated as BMP4 antagonists, which are found to coexpress with BMP4 at sites of oscillatory shear stress in the systemic vasculature (4). Noggin has long been known as a classic BMP antagonist with a high-affinity binding to BMP4 (25, 26, 49). Traditionally defined as an antagonist of activin protein, follistatin could also interact with BMPs (including BMP4, 5, 6, 7, and 15), though in a lower-affinity range (5, 12, 32, 35). MGP is an extracellular matrix component expressing high abundance in vascular smooth muscle cells. MGP has been defined as contributing to the development of vasculature by using MGP-deficient mice (45). MGP inhibits or activates BMPs (BMP2 and BMP4) in a concentration-dependent manner (46, 48). So far, many groups have discussed the effects of BMPs antagonists on interfering BMP signaling and participating in the development of different diseases. However, the full action of these members in CHPH remains largely unknown. We previously demonstrated that animals exposed to chronic hypoxia (CH, 10% O₂) for 21 days were used as an animal model for mimicking CHPH (21, 40). Thus, in this study, we aimed to figure out the changes in these antagonists upon hypoxic stresses and determine their roles in the process of hypoxia-induced proliferation in PASMC.¹

MATERIALS AND METHODS

Chronic hypoxic exposure of rats and primary culture of PASMC.

All animal experimental procedures were approved by the Institutional Animal Care and Use Committee of The Johns Hopkins University School of Medicine. The surgical procedure was performed under anesthesia with pentobarbital sodium (65 mg/kg ip), and all efforts were made to minimize animal suffering. Adult male Wistar rats (150–250 g) were purchased from Harlan (Frederick, MD). Rats were randomly divided into two groups and exposed to normoxia or chronic hypoxia (10% O₂) for 21 days as previously described (34, 40). The hypoxic chamber was continuously flushed with a mixture of room air and N₂ to maintain 10 ± 0.5% O₂ and CO₂ <0.5%. Chamber O₂ concentration was continuously monitored using a PRO-OX unit (RCI Hudson, Anaheim, CA). Rats were exposed to room air for 10 min every day for administering food and water, or change of cage. By the end of the exposure, rats from all groups were anesthetized with pentobarbital sodium (65 mg/kg ip) and euthanized for harvesting lung tissues. The blood in the vessel was gently completely washed out with PBS to eliminate the interference of the circulation. The primary culture of distal PASMC was based on an enzymatic digestion method, as previously described (39). Basically, distal (>4^th generation) intralobar distal PAs were dissected from the lungs. Adventitia was removed from the isolated PAs, and endothelium was denuded by opening the vessel longitudinally and rubbing the luminal surface with a cotton swab.

Real-time quantitative polymerase chain reaction.

Total RNA was extracted using TRIzol method for lung tissues, as previously described (21). DNA contamination in RNA preparations was removed by on-column DNase digestion using RNeasy column and RNase-free DNase (Qiagen, Valencia, CA). Reverse transcription was performed using iScript cDNA synthesis kit (Bio-Rad, Hercules, CA) with reaction mixture containing 1 μg total RNA in a 20 μl volume. cDNA were amplified by real-time quantitative polymerase chain reaction (RT-qPCR) using QuantiTect SYBR Green PCR Master Mix (Qiagen) in an iCyclerIQ detection system (Bio-Rad). The protocol consisted of initial enzyme activation at 95°C for 3 min, followed by 40 cycles at 95°C for 5 s and at 60°C for 15 s (22). Primer sequences were designed using Primer3 software and are listed in Table 1. Relative concentration of each transcript was calculated using the Pfaffl method (33). Efficiency for each gene was determined from 5-point serial dilutions of an unknown cDNA sample. The expression of BMP antagonists was normalized to cyclophilin B as internal control.

Table 1.

Primers for real time-qPCR

Gene	Accession No.	Source	Primer Sequence (left/right)	Product Size, bp
Follistatin	NM_012561.1	Rat	5′-AACCTACCGCAACGAATGTG-3′	138
			5′-TGATCCACCACACAAGTGGA-3′
Noggin	NM_012990.1	Rat	5′-CCTGGCTTTCTGGTTCATGT-3′	123
			5′-GCCGGGTAACTTTTGACGTA-3′
Gremlin1	NM_019282.3	Rat	5′-GACAAGGCTCAGCACAATGA-3′	118
			5′-ACTCAAGCACCTCCTCTCCA-3′
MGP	NM_012862.1	Rat	5′-TGAATCTCACGAAAGCATGG-3′	91
			5′-CCATCTCTGCTGAGGGGATA-3′
Cyclophilin B	NM_022536.1	Rat	5′-CAAGACCTCCTGGCTAGACG-3′	81
			5′-TTCTCCACCTTCCGTACCAC-3′

Western blotting.

Lung tissue samples were sonicated or homogenized in T-PER sample buffer (Pierce, Rockford, IL) containing protease inhibitor cocktail. Total protein concentration in the homogenates was determined by bicinchoninic acid protein assay (Pierce) using bovine serum albumin as a standard. Homogenates were denatured by adding dithiothreitol to 150 mM and heating at 95°C for 3 min. Homogenate proteins were resolved by 10% SDS-PAGE calibrated with Precision Plus protein molecular weight markers (Bio-Rad). Separated proteins were transferred to 0.45 μM polyvinylidene difluoride membranes (Bio-Rad). Membranes were blocked with 5% nonfat dry milk in Tris-buffered saline containing 0.2% Tween 20, and blotted with affinity-purified polyclonal antibodies specific for TRPC1 and TRPC6 (Sigma, St. Louis, MO), MGP (Proteintech, Chicago, IL), noggin (BD Biosciences, San Jose, CA), follistatin (Abcam, Cambridge, MA), gremlin1 (Abcam), p-p38, t-p38, p-ERK1/2, t-ERK1/2, p-JAK2, and p-STAT3 (Cell Signaling Technology, Beverly, MA) or monoclonal antibody against β-tubulin and β-actin (Sigma). The membranes were then washed for 10 min 3 times and incubated with horseradish peroxidase-conjugated goat anti-rabbit or anti-mouse IgG (Kirkegaard and Perry Laboratories, Gaithersburg, MD) for 1 h. Bound antibodies were detected using an enhanced chemiluminescence system (ECL, GE Healthcare, Piscataway, NJ).

Measurement of store-operated calcium entry.

Following previously described procedures (39), cultured PASMC were seeded on coverslips and serum starved for 24 h in smooth muscle basal medium (SMBM) containing 0.3% FBS and 1-h loading with 7.5 μM fluorescent dye fura-2 AM (Molecular Probes, Eugene, OR) before calcium measurement experiments. Intracellular Ca²⁺ concentration was calculated by the ratio of fura-2 fluorescence emitted at 510 nm after excitation at 340 nm to that after excitation at 380 nm (F₃₄₀/F₃₈₀). Store-operated calcium entry (SOCE) was evaluated by extracellular calcium restoration method, in which intracellular sarcoplasmic reticulum calcium stores within PASMC were depleted by perfusion of 10 μM cyclopiazonic acid (CPA), in the presence of 1 mM EGTA and 5 μM voltage-dependent calcium channel (VDCC) blocker nifedipine to exclude the influence of VDCC. By restoring extracellular calcium we could measure SOCE through store-operated calcium channels (SOCC).

Measurement of PASMC proliferation.

The proliferation of PASMC was measured with the Cell Proliferation Biotrak ELISA Kit (GE Healthcare) based on a BrdU incorporation assay, as previously described (41). Briefly, PASMC were seeded in 96-well plates in SMBM at a density of 5 × 10³ cells/well. After each different treatment, the cells were then labeled with BrdU for 24 h, fixed and blocked; antibody was probed, developed with TMB substrate, and stopped with sulfuric acid, following kit instructions. The optical density of wells was measured with a microplate reader (Bio-Rad) at 450 nm.

Materials and reagents.

Unless otherwise specified, all the reagents were obtained from Sigma-Aldrich. Fura-2 AM (Invitrogen) was prepared before the experiment as a 2.5 mM stock solution in 20% DMSO containing 20% pluronic F-127 (Invitrogen). Stock solutions of nifedipine and CPA were both made in DMSO at 30 mM. Human recombinant BMP4 protein was purchased from R&D and noggin recombinant protein was purchased from Invitrogen.

Statistical analysis.

All of the data are represented as means ± SE; n is number of experiments, which equals the number of animals providing PA or lung; number of dishes of cells for in vitro molecular biological experiments and intracellular calcium imaging experiments. Statistical analyses were performed using analysis of Students t-test and one-way ANOVA. Pairwise comparison of means was conducted with t-tests. For the groups containing multiple comparisons, one-way ANOVA was used for the statistical analysis. Differences were considered significant when P < 0.05.

RESULTS

Chronic hypoxic exposure (10% O₂, 21-day) selectively downregulates noggin expression in rat lung tissues.

First, we focused on the mRNA and protein expression of the four BMP antagonists noggin, follistatin, gremlin1, and MGP in the lungs isolated from both normoxic and CH-exposed rats. We designed specific primers for all of these members for real-time qPCR amplification (Table 1). By using real-time qPCR, we found that all of the four BMP antagonists were expressed in rat lung tissue and among which, noggin was selectively downregulated for 41.9% by hypoxic exposure (Fig. 1A). Then, by using Western blotting, we found that noggin protein levels were also significantly downregulated 45.7% (Fig. 1, B and C) by CH exposure, in comparison with normoxic control. However, the other three antagonists remained unaltered by hypoxic stress in the lung.

Fig. 1.

Chronic hypoxia (10% O₂, 21-day) downregulates noggin expression at both mRNA and protein levels in rat lung tissue. A: bar graph representing mRNA fold induction of noggin, follistatin, gremlin1, and matrix gla protein (MGP) in lungs isolated from 21-day chronic hypoxia-exposed rats normalized to levels from normoxic control rats. Cyclophilin B (CpB) served as housekeeping internal control. B and C: Western blots (B) and bar graph (C) showing protein expression levels of the four bone morphogenetic protein (BMP) antagonists in rat lung tissues under normoxic and hypoxic conditions. β-Actin served as housekeeping protein. S1, S2, and S3 mean the number of each individual sample in each group of treatment. The bar graph represents means ± SE; n = 6–12 in each group. *P < 0.05 vs. normoxic control.

Prolonged hypoxic exposure (4% O₂, 60-h) selectively downregulates noggin expression in rat distal PASMC.

Since previous studies strongly indicated that the excessive proliferation and migration of distal PASMC act as two major events contributing to distal PA remodeling and thickening during PH pathogenesis, we further explored if expression of these antagonists can also be regulated by hypoxia in cultured PASMC. Primary cultured rat distal PASMC were serum starved and exposed to prolonged hypoxia (4% O₂) for up to 60 h, the period of which has been proved to induce elevation of intracellular calcium concentration and cell proliferation (40). We used real-time PCR to measure mRNA transcription levels and Western blotting to measure protein levels of the four BMP antagonists in PASMC upon prolonged hypoxic exposure. The RT-qPCR data showed no obvious change in the mRNA levels of all these four antagonists under different time points of hypoxic exposure (Fig. 2, A–D). However, the protein level of noggin, but not the others, was selectively reduced at the 48- and 60-h hypoxic exposure time points, accompanied with upregulated BMP4 protein levels (Fig. 2, E and F). These results indicate the possibility that the hypoxia-suppressed noggin, acting in correlation with the hypoxia-induced BMP4, might account for the uncontrolled proliferation and elevated intracellular calcium concentration and SOCE in PASMC.

Fig. 2.

Prolonged hypoxia (4% O₂) inhibits noggin protein expression in rat distal pulmonary arterial smooth muscle cells (PASMC). A–D: bar graphs representing mRNA fold induction of noggin, follistatin, gremlin1, and MGP in rat distal PASMC that underwent hypoxic exposure (H) for 6–60 h, as relevant to CpB. NOR, normoxia. E and F: Western blots (E) and line chart (F) showing protein expression levels of the four BMP antagonists and BMP4 in cultured rat distal PASMC under different time points of hypoxic exposure. The bar graph represents means ± SE; n = 3–6 in each group. *P < 0.05 vs. normoxic control (BMP4 line); #P <0.05 vs. normoxic control (noggin line).

Treatment of extrogenous recombinant noggin inhibits BMP4-induced phosphorylation of p38 and ERK1/2 in PASMC.

Our previous study indicated that among the substantial cellular signaling pathways that can be induced by BMP4, the p38 and ERK1/2 were considered to be essential for the induction of downstream SOCE processes and the expression of SOCC components TRPC1 and TRPC6 (18). Thus, we applied treatment of exogenous recombinant noggin protein and detected if noggin can block BMP4-induced p38 and ERK1/2 activation. Our results showed that 200 ng/ml of noggin treatment totally blocked BMP4-induced phosphorylation of p38 and ERK1/2 (Fig. 3, A and B).

Fig. 3.

Noggin blocks BMP4-induced p38 and ERK1/2 phosphorylation in PASMC. A: Western blots showing protein expression levels of p-p38, t-p38, p-ERK1/2, t-ERK1/2, and β-tubulin in cultured rat distal PASMC upon BMP4 (50 ng/ml) treatment for 15 min after pretreatment with noggin (50 and 200 ng/ml) for 30 min. B: bar graphs representing the phosphorylation rates of p-p38 (left) and p-ERK1/2 (right) as normalized to t-p38 and t-ERK1/2. The bar graph represents means ± SE; n = 3 in each group. *P < 0.05 vs. vehicle control; #P < 0.05 vs. BMP4 (50 ng/ml)-treated group.

Treatment of extrogenous recombinant noggin inhibits BMP4-induced phosphorylation of JAK2 and STAT3 in PASMC.

As we know, p38 and ERK are not the only kinases to initiate and mediate the cellular pro-proliferation subsequences. Actually, a variety of previous publications have reported that the JAK2/STAT3 signaling axis can mediate smooth muscle cell proliferation upon activation from upstream stimulation (1, 24, 36–38). Therefore, we found that BMP4 treatment (50 ng/ml) can activate the JAK2/STAT3 signaling cascade in a time-dependent manner (Fig. 4, A and B), then, a pretreatment with noggin (200 ng/ml) for 30 min significantly attenuated the BMP4-induced phosphorylation of JAK2/STAT3 (Fig. 4, C and D).

Fig. 4.

Noggin attenuates BMP4-induced JAK2 and STAT3 phosphorylation in PASMC. A and B: BMP4 time dependently induces JAK2 and STAT3 phosphorylation in PASMC. Western blots show protein expression levels of p-JAK2, p-STAT3, and β-tubulin in cultured rat distal PASMC upon BMP4 (50 ng/ml) treatment for 5–120 min (A). Bar graphs represent the phosphorylation rates of p-JAK2 (left) and p-STAT3 (right) as normalized to β-tubulin (B). C and D: pretreatment of noggin (200 ng/ml) for 30 min attenuates BMP4 (50 ng/ml, 30-min)-induced phosphorylation of JAK2/STAT3 in PASMC. Western blots show protein expression levels of p-JAK2, p-STAT3, and β-tubulin in cultured rat distal PASMC upon BMP4 and noggin treatments (C). Bar graphs represent the phosphorylation rates of p-JAK2 (left) and p-STAT3 (right) as normalized to β-tubulin (D). The bar graph represents means ± SE; n = 3–5 in each group. *P < 0.05 vs. control; #P < 0.05 vs. BMP4 (50 ng/ml)-treated group.

Treatment of extrogenous recombinant noggin inhibits hypoxia-elevated proliferation of PASMC.

Since the dosage of 200 ng/ml of noggin treatment totally blocks BMP4-induced p38 and ERK1/2 activation, we further determined the effect of 200 ng/ml noggin on the proliferation in PASMC under both normoxic and prolonged hypoxic conditions. Our results showed that first, hypoxia markedly increased the proliferation rate 41.6% compared with normoxic control. Then, 200 ng/ml noggin significantly normalized hypoxia-elevated proliferation to basal levels, without altering the proliferation under normoxia (Fig. 5).

Fig. 5.

Noggin normalized hypoxia-elevated proliferation in PASMC. The bar graph represents the fold induction of the PASMC proliferation rate in response to prolonged hypoxia (4% O₂, 60-h) exposure with or without noggin (200 ng/ml) treatment, as normalized to normoxic control value. The bar graph represents means ± SE; n = 4 in each group. *P <0 .05 vs. normoxic control; #P < 0.05 vs. hypoxic control.

Noggin attenuates hypoxia-elevated SOCE in PASMC.

Furthermore, we also detected the role of noggin treatment on the SOCE and intracellular calcium regulation in PASMC, since hypoxia-enhanced SOCE is considered a key regulator to promote PASMC proliferation and PA remodeling (40). Our results indicated that prolonged hypoxia significantly enhanced SOCE from 295.9 nM to 372.7 nM, while noggin dramatically normalized hypoxia-enhanced SOCE to 275.7 nM, without markedly altering the SOCE under normoxia (Fig. 6).

Fig. 6.

Noggin attenuates hypoxia-elevated store-operated calcium entry (SOCE) in PASMC. A: line charts showing the dynamic traces of the intracellular calcium concentration ([Ca²⁺]_i) in response to intracellular calcium store depletion by cyclopiazonic acid (CPA, 10 μM), in the presence of the voltage-dependent calcium channel (VDCC) blocker nifedipine (Nif; 5 μM) in calcium-free solution (0 Ca²⁺) and restoration of extracellular calcium concentration afterward. B: bar graph indicating the Δ changes before and after extracellular calcium restoration was calculated as SOCE. The bar graph represents means ± SE; n = 3 experiments in each group. Four groups were designed as normoxic control (Nor), normoxia+noggin (Nor+Nog, 200 ng/ml, 60 h), hypoxic control (Hyp), and hypoxia+noggin (Hyp+Nog, 200 ng/ml, 60 h). *P < 0.05 vs. normoxic control; #P < 0.05 vs. hypoxic control.

Noggin normalizes hypoxia-induced upregulation of TRPC1 and TRPC6 in PASMC.

Given the fact that noggin significantly normalized hypoxia-induced cell proliferation and SOCE in PASMC, we therefore investigated the role of noggin on the expression of the main SOCC components TRPC1 and TRPC6, which was previously reported to mediate hypoxia-enhanced SOCE in PASMC. As seen in Fig. 7, noggin treatment also attenuates the upregulated TRPC1 and TRPC6 protein expression under hypoxic condition, but not under normoxia.

Fig. 7.

Noggin normalized hypoxic upregulation of transient receptor potential cation channels TRPC1 and TRPC6 protein expression in PASMC. A: Western blots showing protein expression levels of TRPC1, TRPC6, and β-tubulin in cultured rat distal PASMC upon hypoxic exposure (4% O₂, 60 h) and noggin (200 ng/ml) treatment. B: bar graphs representing the fold induction of TRPC1 (left) and TRPC6 (right) under different treatments, as normalized to normoxic control. β-Tubulin served as internal control. The bar graph represents means ± SE; n = 3 in each group. *P < 0.05 vs. normoxic control; #P < 0.05 vs. hypoxic control.

DISCUSSION

Previous evidence strongly indicated the essential contribution of increased BMP4 expression and elevated BMP4-mediated signaling transduction in CHPH pathogenesis. In this study, we investigated the expression and role of a group of four reported endogenous BMP4 antagonists in response to hypoxic stress in rat lung and distal PASMC. Our results indicated that among these four antagonists, noggin was selectively downregulated by hypoxic exposure in both lung and PASMC, in correlation with hypoxia-elevated BMP4 expression and activity. On the basis of the previous data, we hypothesized that hypoxia-suppressed noggin lacks the ability as an antagonist to inhibit the hypoxia-evoked BMP4 activity, leading to promoted BMP4 signaling transduction, which results in the transcription of numerous proliferative genes that triggers the cellular pro-proliferative subsequences. We further observed that extrogenous noggin treatment significantly normalized hypoxia-triggered proliferation, SOCE, and TRPC expression, suggesting that suppressed noggin contributes to the elevated BMP4 signaling and activity under hypoxia.

BMPs belong to the TGF-β superfamily, which are a group of factors originally found to induce ectopic bone formation when implanted subcutaneously (42). Then, considerable evidence demonstrated that BMPs also act as key multifunctional regulators in regulating cell proliferation, differentiation, and apoptosis in different tissues (27). BMPs exert their signals via binding with type II transmembrane serine/threonine kinase receptors, thus recruit with type 1 transmembrane serine/threonine kinase receptors, then activating the intracellular Smad-dependent and Smad-independent (e.g., ERK, JNK, and p38 MAP kinase pathways) signaling pathways in regulating substantial downstream gene expression (43). Basically, three type II receptors (BMPRII, ActRIIa, and ActRIIb) and three type I receptors (ALK2, ALK3, and ALK6) have already been indicated to be able to interfere with BMP ligands and transduce signaling (27, 28). Recently, a number of studies suggested that abnormal BMP signaling participates in the disease pathogenesis of PH. Many works have been done based on the finding of large chances of BMPRII mutation in idiopathic PAH and familial PAH patients (8, 17).

Besides the specific BMP receptors, there are also several extracellular modulators that can interfere with the BMPs ligand-receptor binding to block BMP signaling. Noggin, MGP, gremlin1, and follistatin belong to such members. These soluble antagonists dampen the BMP signaling either by competitively interacting with the BMP ligands and protecting them from binding with their specific receptors or by blocking the intracellular signal transduction without affecting ligand-receptor reaction (3, 15, 49). Previous publications addressed the different localization of these antagonists in the lung. Gremlin1 is identified in the endothelium, proximal airway epithelium, and alveolar epithelium (19, 29). Follistatin is widely expressed in bronchial epithelial cells, alveolar macrophages, and vascular smooth muscle cells (47). MGP is excessively and broadly expressed in the lung tissue (48). As it is well accepted that dysfunction of BMP4 signaling participates in and contributes to the excessive proliferation of PASMC in PH, we wondered whether the triggered BMP4 signaling is possibly, to some extent, due to the loss of expression and function of the antagonists.

Hence, we investigated the expression changes of these antagonists in response to hypoxic exposure in both the lung and PASMC. Our results showed that among these four antagonists, noggin was selectively downregulated by hypoxia in both the lung and PASMC. However, by applying the extrogenous noggin recombinant protein treatment, we further demonstrated that under hypoxia, the restoration of noggin effectively normalized hypoxia-induced proliferation, SOCE, and TRPC expression. It is important to compare the blockage efficiency between noggin treatment and direct BMP4 inhibition strategy (BMP4 knockdown). Actually, in one of our previous publications (20), we reported that knockdown of BMP4 by using specific siRNA effectively leads to dramatic decrease in the hypoxia-elevated TRPC-SOCE-[Ca²⁺]_i signaling axis. Given the fact that the intracellular free calcium concentration acts as a major factor to enter the nucleus, facilitiates the transcription of a number of pro-proliferative genes and promote cell proliferation (16, 30, 31), we can conclude that in PASMC, BMP4 is responsible for the proliferation and intracellular calcium homeostasis, especially the SOCE process. Moreover, in this study, we further showed that noggin exerts similar inhibitory roles, in comparison with direct BMP4 knockdown, on the TRPC-SOCE-proliferation signaling axis. Therefore, our data suggest that a decrease in the naturally endogenous expressing BMP4 antagonist noggin is potentially a main reason accounting for the elevated BMP4 signaling pathway. The hypoxia-inhibited noggin accounts for, at least partly, the hypoxia-triggered BMP4 activity and signaling transduction. Due to the lack of noggin's protective role, evoked BMP4 signaling leads to upregulated TRPC-SOCE axis, which eventually results in excessive PASMC proliferation in PH development. Specific noggin-targeted restoration might act as a potential novel strategy to modulate BMP4 expression and activity, thus mediating an attenuation of hypoxia-elevated proliferation in PASMC. Moreover, we also found that besides the p38/ERK-TRPC-SOCE-[Ca²⁺]_i signaling axis, BMP4 can also induce PASMC proliferation by inducing activation of JAK2/STAT3 signaling cascade. Our results strongly indicate that noggin can effectively attenuate BMP4-mediated proliferation by inhibiting both p38/ERK-TRPC-SOCE targeted calcium-dependent signaling axis and JAK2/STAT3 targeted calcium independent signaling axis.

In summary, this study presents initial evidence for the expression pattern of the four BMPs antagonists in response to hypoxic stress. We observed that the hypoxia-inhibited noggin level acts as a key element to result in excessive PASMC proliferation due to the uncontrolled elevated BMP4 signaling, as well as the downstream triggered p38/ERK-mediated calcium-dependent and JAK2/STAT3-mediated calcium-independent pro-proliferative signaling pathways, while strategies targeting noggin restoration might be a useful way to attenuate the hypoxia-elevated proliferation of PASMC. This study provides evidence of a novel potential target, noggin, which can potentially attenuate the hypoxia-elevated proliferation of PASMC and deserves further study and research.

GRANTS

This work was supported by National Institutes of Health (National Heart, Lung, and Blood Institute Grant R01 HL-093020), National Natural Science Foundation of China (81173112, 81470246, 81170052, 81220108001), Guangzhou Department of Education Yangcheng Scholarship (12A001S), Guangzhou Department of Natural Science (2014Y2-00167) and Guangdong Province Universities and Colleges Pearl River Scholar Funded Scheme (2014, W. Lu), and China Scholarship Council (201208440091 for K. Yang).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

K.Y., J.J., J.Z., H.J., and L.X. performed experiments; K.Y., W.L., and J.W. interpreted results of experiments; K.Y., J.J., and J.W. prepared figures; K.Y. drafted manuscript; J.J., J.Z., W.L., and J.W. analyzed data; M.Z., S.W., W.L., and J.W. edited and revised manuscript; W.L. and J.W. conception and design of research; W.L. and J.W. approved final version of manuscript.