| Sign In to gain access to subscriptions and/or personal tools. |
Smad Signaling in the Rat Model of Monocrotaline Pulmonary Hypertension
1 Departments of Pathology, Immunology, and Microbiology and Correspondence: Address correspondence to: Dennis Wilson, DVM, PhD, DACVP, School of Veterinary Medicine, Pathology, Immunology, Microbiology, One Shields Avenue, 1044 Haring Hall, Davis, CA 95616-8617; e-mail: dwwilson{at}ucdavis.edu.
Mutations in the bone morphogenetic protein receptor type II (BMPrII) gene have been implicated in the development of familial pulmonary artery hypertension (PAH). The function of BMP signal transduction within the pulmonary vasculature and the role BMPrII mutations have in the development of PAH are incompletely understood. We used the monocrotaline (MCT) model of PAH to examine alterations in Smad signal transduction pathways in vivo. Lungs harvested from Sprague-Dawley rats treated with a single 60-mg/kg intraperitoneal (IP) injection of MCT were compared to saline-treated controls 2 weeks following treatment. Smad 4 was localized by immunohistochemistry to endothelial nuclei of the intra-acinar vessels undergoing remodeling. Smad 4, common to both BMP and transforming growth factor β (TGFβ) signaling, and BMP-specific Smad 1 were significantly decreased in western blot from whole lungs of treated animals, while no change was found for TGFβ-specific Smad 2. MCT-treated rats also had increased expression of phosphorylated Smad 1 (P-Smad 1) but not phosphorylated Smad 2 (P-Smad 2). There was a decrease in the expression of the full BMPrII protein but not its short form variant in MCT-treated rat lungs. The type I receptor Alk1 had increased expression. Collectively, our data indicate that vascular remodeling in the MCT model is associated with alterations in BMP receptors and persistent endothelial Smad 1 signaling.
Key Words: BMP receptors • Smad signal transduction • pulmonary hypertension • monocrotaline Abbreviations: BM, bone morphogenetic protein • BMPrII, bone morphogenetic protein type II receptor • BSA TBS, bovine serum albumin Tris buffered saline • DAPI, 4’-6-diamidino-2-phenylindole • DMEM, Dulbecco’s modified eagle medium • DMF, N,N-dimethyl formamide • ECL, enhanced chemiluminescence • FBS, fetal bovine serum • HPAECs, human pulmonary artery endothelial cells • HRP-DAB, horseradish peroxidase-diaminobenzidine • IFC, immunofluorescent chemistry • IHC, immunohistochemistry • IP, intraperitoneal • IPAH, idiopathic pulmonary arterial hypertension • MCT, monocrotaline • MCTP, monocrotaline pyrrole • PAH, pulmonary artery hypertension • PASMCs, pulmonary artery smooth muscle cells • PH, pulmonary hypertension • PPH, primary pulmonary hypertension • PVDF, polyvinylidene fluoride • P-Smad 1, P Smad 2, phosphorylated Smad 1 or Smad 2 • RIPA, radioimmunoprecipitation assay • SDS, sodium dodecyl sulfate • TGFβ, transforming growth factor β
Pulmonary artery hypertension (PAH) is characterized by remodeling of the pulmonary arteries with endothelial proliferation, smooth muscle hyperplasia and hypertrophy, and expansion of the adventitial matrix. An increase in pulmonary arterial pressure, right ventricular hypertrophy, and eventual cor pulmonale are the manifestations of a complex pathophysiology that composes this disease. The heterogeneity of the involved cell populations suggests that vascular remodeling is a consequence of the dysregulation of multiple converging signaling pathways that normally establish and maintain vascular integrity and function. Transforming growth factor β (TGFβ) signal transduction has been a focus of research because a number of mutations and altered expression of receptors, ligands, or secondary messenger systems have been described in a variety of vascular diseases. TGFβ receptors in particular have been attributed a role in the pathogenesis of either primary or secondary forms of PAH. Mutations in the Alk1 TGFβ type I receptor and its associated protein Endoglin occur in the vascular disease hemorrhagic telangiectasia, which, in the lung, results in secondary PAH (Lebrin et al., 2005). Mutations in another receptor in the TGFβ family, the bone morphogenetic protein type II receptor (BMPrII), have been described in familial and idiopathic PAH (Deng et al., 2000; Machado et al., 2001; Thomson et al., 2000). The TGFβ family of receptors appears to have a significant role in directing vascular remodeling in PAH. The balance between agonists, recruitment of diverse receptors, and activation of second messengers directs differing growth stimulatory or inhibitory responses in proximal and distal smooth muscle and endothelium. For example, TGFβ1 stimulates proliferation in cultured vascular smooth muscle while inducing apoptosis in endothelial cells (Pollman et al., 1999). In the pulmonary vasculature, the effect of bone morphogenetic protein 4 (BMP4) appears to be location specific. BMP4 inhibited proliferation of pulmonary artery smooth muscle cells (PASMCs) isolated from proximal pulmonary arteries but stimulated proliferation of PASMCs from peripheral arteries and conferred protection from apoptosis (Yang et al., 2005). Our laboratory recently demonstrated that in the rat, BMPrII expression is highest in pulmonary endothelium with lesser expression in vascular smooth muscle (Ramos et al., 2006). Disruption in the TGFβ/BMP balance is emerging as a key concept in the understanding of both familial and idiopathic PAH. A recent comprehensive analysis of human BMPrII mutations strongly suggests that haploinsufficiency is sufficient to predispose to PAH and that a critical threshold of signaling loss is associated with disease onset. A key consequence of mutation appears to be failure of posttranslational processing and lack of protein expression at the cell surface (Machado et al., 2006). In support of this, patients with BMPrII mutations express significantly decreased amount of BMPrII protein (Atkinson et al., 2002). This loss is manifested in a decrease in P-Smad1 in affected patients (Yang et al., 2005). Signaling in the TGFβ family is initiated through cytokine binding and subsequent heteromeric receptor recruitment and dimerization. Agonist specific type II receptors phosphorylate type I receptors, a family of proteins identified as Alk1–7, which in turn activate the secondary messenger system known as Smads. Smads 2 and 3 are phosphorylated by typical TGFβ receptors, while Smads 1, 5, and 8 are phosphorylated by BMP receptors (Mehra et al., 2000; Miyazono, 1999; Nakao, et al. 1997; Wrana and Attisano, 2000). Phosphorylated Smads complex with the common Smad 4, which then translocates to DNA binding sites in the nucleus. The transcriptional response depends on cell type, distribution, and recruitment of specific receptors; ligand dose and duration; and additional transcription cofactors. Endothelial cells within the pulmonary vasculature express both the TFGβ type I receptors Alk1 (pro-angiogenic) and Alk5 (anti-angiogenic; Goumans et al., 2003; Lebrin et al., 2005). Alk1, unlike other TGFβ receptors, signals via Smad 1, similar to BMP receptors (Chen and Massague, 1999). Differences in cell type and microenvironment explain how a mutation in a broadly expressed gene such as BMPrII is manifested within a specific location (i.e., the arterial vasculature of the lung). Monocrotaline (MCT) is a pyrrolizidine alkaloid that induces cell injury by forming protein and DNA adducts inducing cell-cycle arrest and cell death (Lame et al., 2000; Lame et al., 2005; Mukhopadhyay et al., 2005; Thomas et al., 1996). In the Sprague-Dawley rat, MCT induces endothelial injury leading to vascular remodeling and pulmonary hypertension (PH) within 2 weeks that, in many respects, resemble human PAH (Ghodsi and Will, 1981; Schultze and Roth, 1998; Wilson et al., 1992). The MCT-induced model of PAH remains a mainstay of experimental evaluations of novel therapeutic interventions. The increasing knowledge of dysfunction in TGFβ family signaling in the human disease makes it possible to design more targeted therapies. While many physiologic and morphologic similarities between the MCT model and human disease have been described, knowledge of the location and nature of TGFβ family signaling processes in remodeling arterioles of MCT-treated rats is lacking. We have previously demonstrated Smad 4 nuclear accumulation and Smad 1 phosphorylation in monocrotaline pyrrole (MCTP)-treated human pulmonary arterial endothelial cells (HPAEC; Ramos et al., 2007). We hypothesized that Smad signaling pathways are also activated in the pulmonary vasculature of MCT-treated rats. The purpose of the present investigation was to characterize the signaling processes taking place in the remodeling phase of the MCT-treated rat model of PAH and compare these changes with emerging knowledge of signaling dysfunction in affected humans.
Rabbit antibody to BMPrII was generated in our laboratory (Ramos et al., 2006) according to previously described methods (Gullick et al., 1985; Rosenzweig et al., 1995). Briefly, a peptide corresponding to amino acid sequence 185–202 of human BMPrII was coupled to keyhole limpet hemocyanin. Male rabbits were immunized according to standard methods and the presence of antibody verified by an Elisa assay. Rabbit serum was purified in 2 successive steps on keyhole limpet hemocyanin agarose (Alpha Diagnostics International, Inc., San Antonio, Texas) and thiophilic gel (Pierce Biotechnology, Rockford, Illinois) using manufactures’ suggested protocol. Final protein concentration was 7.4 µg/µl. Dulbecco’s modified eagle medium (DMEM) was obtained from Gibco BRL (Rockville, Maryland). Antibodies to Smad 4 used in western blot analysis came from Cell Signaling Technology (Beverly, Massachusetts). Anti-Smad 4 used for immunohistochemistry (IHC; cat. #7966) was obtained from Santa Cruz Biotechnology (Santa Cruz, California). Antibodies to P-Smad 1 and P-Smad 2 used in western blot were obtained from Upstate (Lake Placid, New York) and Chemicon International (Temecula, California), respectively. TGFβ1 and BMP2 antibodies were obtained from Santa Cruz Biotechnology, antibody to BMP4 was obtained from Chemicon International, and Cav-1 was obtained from BD Transduction Laboratories (San Diego, California). Enhanced chemiluminescence (ECL) detection kit and horse-radish peroxidase (HRP) conjugated antibodies (antirabbit and antimouse) used in western analysis were acquired from Amersham Biosciences (United Kingdom). Mouse IgG isotype and HRP-DAB immunohistochemistry kit were from R&D Systems (Minneapolis, Minnesota). Alexa 555 fluorescent-tagged secondary antibodies and SlowFade with 4’-6-diamidino-2-phenylindole (DAPI) were obtained from Molecular Probes (Eugene, Oregon). Additional supplies include Superblock (Pierce Biotechnology) and polyvinylidene fluoride (PVDF) membranes (Bio-Rad Laboratories; Hercules, California). MCT (Trans World Chemicals, Rockville, Maryland) was converted to the hydrochloride salt and dissolved in saline at a concentration of 30 mg/ml.
MCT Protocol Male Spraque-Dawley rats weighing approximately 200 g were randomly separated into 2 groups. Rats were administered 60 mg/kg MCT IP, while control rats were administered an equal volume of saline. Animals were weighed daily for 2 weeks and monitored for respiratory distress and anorexia. Lung tissue harvested from 3 rats per group was used for western blotting, while an additional 3 rats per group were used for enzymatic and fluorescent IHC.
Immunohistochemistry
Western Blot
IHC Evaluation of Smad 4 in Rat Lung Figure 1 demonstrates typical lesions of MCT pulmonary hypertension that occur in rats 2 weeks posttreatment. The intra-acinar vessel walls are markedly expanded by endothelial cell and smooth muscle hypertrophy and hyperplasia and matrix deposition. Vessels are surrounded by inflammatory infiltrates and edema. Smooth muscle hyperplasia is also present in alveolar septae.
Nuclear localization of Smad 4 was more prominent within endothelial cells of the pulmonary intra-acinar vessels of MCT-treated rats (Figure 2A, 2B) compared to saline-treated control rats (Figure 2C, 2B) in which staining is predominantly cytoplasmic. Nuclear accumulation most often correlated with hypertrophied endothelial cells associated with remodeled but patent vessels (Figure 3A to 3C). However, Smad 4 was predominantly in the cytoplasm of endothelial cells of occluded vessels with advanced lesions (Figure 3E, 3F). Nuclear accumulation in vascular smooth muscle was not apparent. Differences were not apparent within larger arterioles of treated animals. While Smad 4 positive nuclei were not found in counts of endothelial cells from bronchus-associated pulmonary arteries, Smad 4-stained endothelial cell nuclei were present in 45 ± 6% of intra-acinar arteries of MCT-treated rats compared with 15 ± 5% of intra-acinar arteries of in control lungs.
Western Blot Analysis Two major bands were identified for the common Smad 4, 70–75 and 40 kDa in rat lung (Figure 4A). These bands are consistent with Smad 4 splice variants described by others (Pierreux et al., 2000). A significant decrease was found in both bands (p = .0006 and p = .03, top and bottom, respectively) in treated animals verses saline controls.
Because Smad 4 is activated through either BMP or TGFβ signaling, we used quantitative western blot analysis to examine the activation of these signaling pathways in rat lung 2 weeks following treatment with MCT (Figure 4A). Smads 1 and 2, the primary receptor activated Smads in the signal cascade, are phosphorylated by ligand activated receptors. Smad 1 is phos-phorylated by either BMP receptors or [Alk]1, while Smad 2 is phosphorylated by the TGFβ receptor Alk5. We found expression of Smad 1 significantly decreased (p = .0002) in whole lungs from MCT rats versus that of saline-treated control animals, while P-Smad 1 expression significantly increased (p = .01). There was no change in expression of either Smad 2 (p = .34) or P-Smad 2 (p = .49; significance shown in Figure 4B). These results suggest preferential activation of the Smad 1 cascade. Because Smad 1 can be phosphorylated either through BMP or TGFβ receptors, we examined the protein expression of BMPrII and Alk1, a proangiogenic, TGFβ type 1 receptor that also signals through Smad 1 (Figure 5A). We observed 2 prominent bands with anti-BMPrII on western blot of whole lung, 75 and 55 kDa, representing variants of BMPrII as we have previously reported (Ramos et al., 2006). In treated animals, expression of the variant represented at 75 kDa was significantly decreased (p = .02), while expression of the variant represented at 55 kDa was not significantly changed from saline treated controls (p = .99). Conversely, Alk1 was significantly increased (p = .01) in treated animals (Figure 5A; significance shown in Figure 5B).
We also examined Cav-1 expression by western blot. We have previously associated BMPrII with caveolae in human pulmonary arterial endothelial cells (Ramos et al., 2006), and pulmonary hypertension has been described in Cav-1 knockout mice (Razani, Engelman, et al., 2001). In a prior in vitro study, we found decreased expression of caveolin and fewer ultrastructurally evident caveolae in MCTP-treated pulmonary arterial endothelial cells (Ramos et al., 2007). In this study, we found that a high level of Cav-1 expression in control animals was visually diminished in western blots from lungs of MCT-treated rats. While statistical comparison of densitometric analysis showed this to be a significant decrease (p = .005), this expression was above the linear range, making the extent of this decrease uncertain (Figure 5A; significance shown in Figure 5B).
Previous studies by us and others have characterized the progression of MCT-induced pulmonary vascular disease from initial endothelial injury to early proliferative responses and later remodeling with endothelial and smooth muscle hypertrophy, matrix deposition, and vascular occlusion beginning at 2 weeks posttreatment. In the present study, we evaluated TGFβ family signaling in the remodeling phase of the MCT model to compare this process with alterations demonstrated in humans with BMPrII mutations. Our findings with Smad 4 translocation show significant activation of TGFβ family signaling, specifically in the remodeling segments of peripheral resistance arteries. Furthermore, the majority of Smad 4 signaling was evident in endothelium rather than vascular smooth muscle, demonstrating that endothelial activation persists in the remodeling phase and suggesting a prominent role for endothelial regulation of smooth muscle hypertrophy and matrix deposition. Western blot analysis demonstrated increased amounts of P-Smad 1 but not P-Smad 2, suggesting that translocation of the common Smad 4 is a result of activation of either the BMP pathway or TGFβ signaling through Alk1. These findings differ from what is known about remodeling arterioles in humans with both idiopathic PAH and BMPrII mutations (summarized in Table 1). Normal and affected individuals demonstrate P-Smad 1 staining in nuclei of both intimal and medial vascular cells (Yang et al., 2005). The latter study also demonstrates decreased P-Smad 1 nuclear staining in individuals with PAH but no overall difference in Smad 1 protein. The implication is that affected humans have decreased BMP signaling in both endothelium and smooth muscle despite an intact second messenger system, while remodeling in our study of MCT-induced disease shows augmented BMP-like signaling limited to endothelium. What remains unclear from human studies is how generalized the P-Smad 1 findings are within the variety of BMPrII mutations, as lesions in the kinase region affect signaling more than those in other domains (Machado et al., 2006).
Most signal-transduction studies have focused on alterations of gene or protein expression occurring within the pulmonary vascular smooth muscle in PAH. BMP receptors are, however, abundantly expressed within the endothelium (Ramos et al., 2006), which also undergoes phenotypic alterations with this disease. In vivo studies that critically evaluate signaling pathways are hindered by the heterogenicity of cell populations, which makes it difficult to correlate dynamic changes in signaling proteins to a particular cell type. In this study, we used enzymatic and fluorescent-based IHC to localize Smad 4 and found nuclear accumulation predominantly in pulmonary endothelium of small arterioles and intra-acinar vessels with evidence of active remodeling. Significantly, these vessels correlate with the location of classic lesions of MCT PAH (Wilson et al., 1992). However, limited nuclear accumulation of Smad 4 occurred in small vessels with advanced lesions—those that appeared occluded histologically. Such vessels appear by angiography to be nonperfused and are the target of therapeutic interventions aimed at stimulating angiogenesis (Zhao et al., 2006). Nuclear translocation was also not evident in vascular smooth muscle or within bronchiolar epithelium and smooth muscle. Evidence of Smad 4 nuclear translocation within the endothelial cells of the pulmonary vasculature alone is not indicative of bioactivation. Studies have shown that nuclear accumulation of phosphorylated receptor-activated Smads is needed for activation of gene transcription (Inman et al., 2002; Xiao et al., 2001). In this study, western blots of whole lung showed that MCT treatment resulted in decreased expression of Smad 1 in correlation with an increase in P-Smad 1, while no change in either Smad 2 or P-Smad 2 was evident. MCT treatment also resulted in significant decreases in expression of Smad 4. These results suggest activation and/or perturbations within Smad 1 pathways of MCT-treated rats. Our finding of decreased expression of Smad 4 also fits with a general hypothesis of feedback down-regulation of TGFβ family signaling in the actively remodeling lung. We also examined protein expression of the receptors Alk1 and BMPrII. Mutations in the genes coding for these receptors have been attributed a role in the development of secondary and familial PAH, respectively (Deng et al., 2000; Lane et al., 2000; Machado et al., 2001; van den Driesche et al., 2003). Our study of remodeling in MCT-treated rats found decreased expression of the 75 kDa but not the 55 kDa isoform of BMPrII. In a previous study (Ramos et al., 2006), we compared multiple commercially available BMPrII antibodies with one developed in our laboratory and found all recognized these 2 isoforms in proteins derived from human pulmonary artery endothelial cells. While the functional differences in these isoforms have not been determined, we speculate that the 55 kDa protein may represent the BMPrII "short form" characterized by others as a splice variant lacking the cytoplasmic tail but retaining Smad activation capabilities (He et al., 2002; Ishikawa et al., 1995). In this instance, remodeling in the MCT model does mimic familial human PAH where marked decreases in BMPrII protein are present in many affected individuals, and sequence analysis shows that 71% of mutations occur in regions that would predict premature termination of the transcript and loss through failure to complete posttranslational processing (Machado et al., 2006). The selective down-regulation of the larger protein is also intriguing given that the function of the BMPrII cytoplasmic tail remains uncertain but appears to interact with actin and micro-tubule regulatory proteins (Foletta et al., 2003; Machado et al., 2003). We found up-regulation of Alk-1 in remodeling lungs of MCT-treated rats. Expression of Alk1 is predominantly limited to endothelial cells, and in contrast to other TGFβ receptors, Alk1 signaling, in association with the co-receptor endoglin, has pro-angiogenic effects by inducing endothelial cell proliferation and migration by signaling through Smad 1/5/8 pathways similar to BMP receptors (Chen and Massague, 1999; Goumans et al., 2003; Goumans et al., 2002; Lebrin et al., 2005; Oh et al., 2000). It is now appreciated that the dose-dependent biphasic effects of TGFβ on the endothelium with vastly different outcomes is at least partially attributable to the differential activation of TGFβ receptors. These effects are mediated via different receptors and signaling pathways, with low-dose TGFβ ligands preferentially stimulating the proliferative Alk1/Smad 1 pathway, while high doses activate the inhibitory Alk5/Smad 2 pathway (Goumans et al., 2002; Oh et al., 2000; Pepper et al., 1993). These alternative pathways for endothelium are summarized in Figure 6. TGFβ isoforms have previously been reported to be increased in MCTP-induced injury (Arcot et al., 1993). The up-regulation of the Alk1 receptor thus provides an alternative explanation for our Smad 1 second messenger findings. Finally, we cannot exclude the possibility that Smad 1 is phosphorylated though a non–receptor-based mechanism in this model. Although we attempted to examine expression of BMP and TGFβ ligands by western blot, this assay was not sensitive enough to detect these cytokines in whole-lung samples (data not shown).
We found a modest and subjective down-regulation of cave-olin-1 in this study. Caveolae are specialized, invaginated membrane domains that have predominantly inhibitory regulatory effects on signaling molecules (Mineo et al., 1999; Minshall et al., 2003; Rodriguez-Pascual et al., 2003). Reduced Cav-1 expression is associated with uncontrolled cellular growth (Sotgia et al., 2006; Williams and Lisanti, 2005). A number of signaling molecules have been located to caveolae, including TGFβ receptors (Razani, Zhang, et al., 2001), and more recently, BMPrII (Ramos et al., 2006). Cav-1 is of particular interest because patients with PAH have been shown to have decreased Cav-1 expression (Geraci et al., 2001) and Cav-1 knockout mice develop proliferative vascular lesions reminiscent of PAH (Razani, Engelman, et al., 2001) and dilated cardiomyopathy and PH (Zhao et al., 2002). Studies suggest that Cav-1 inhibits signaling through specific binding motifs until ligand activated (Schlegel and Lisanti, 2001). Such interactions of Cav-1 with TGFβ type 1 receptors have been established (Derynck and Zhang, 2003; Razani, Zhang, et al., 2001); however, further studies are necessary to establish whether a similar relationship exists with BMP receptors. The loss of this inhibitory molecule may have a contributing role to the pathogenesis of MCT toxicity. The mechanisms by which MCT-induced Smad 4 nuclear translocation within endothelial cells of the pulmonary vasculature and the subsequent alterations in P-Smad 1 expression were not addressed in this study. We have previously shown in vitro that MCTP induces Smad 4 nuclear translocation in HPAEC (Ramos et al., 2007). The half-life of MCTP in rat serum is measured in seconds, however, suggesting that altered signaling 2 weeks posttreatment depends on multiple cellular interactions and presumably represents a physiologic injury-repair response. MCT does, however, induce a persistent cyto and karyomegaly in a variety of lung cells that we postulate is a consequence of polyploidy resulting from covalent interactions of reactive MCT metabolites with macromolecules (Lame et al., 2000; Lame et al., 2005; Thomas et al., 1996). This raises the alternative hypothesis that persistent signaling abnormalities could arise from ongoing interactions of MCT-derived metabolites with proteins involved in cellular transcription or posttranslational processing. In summary, our results suggest a role for TGFβ family signaling in the pulmonary endothelium of remodeling arterioles in the MCT model of PAH. We postulate that the altered signaling in MCT endothelial injury in vivo is the consequence of activation and phosphorylation of Smad 1 through BMP and/or TGFβ-Alk1 signaling pathways. While MCT-treated rats have a decrease in BMPrII expression similar to that in affected humans, they appear to have augmented Smad 1-mediated signaling in contrast to the decrease reported in the human disease.
Present address for Margaret F. Ramos: Allergan Inc., Irvine, California, 92612.
This version was published on February
1, 2008 Toxicologic Pathology, Vol. 36, No. 2,
311-320 (2008)
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
Abstract
Introduction
