Fhl-1, a New Key Protein in Pulmonary Hypertension
Background— Pulmonary hypertension (PH) is a severe disease with a poor prognosis. Different forms of PH are characterized by pronounced vascular remodeling, resulting in increased vascular resistance and subsequent right heart failure. The molecular pathways triggering the remodeling process are poorly understood. We hypothesized that underlying key factors can be identified at the onset of the disease. Thus, we screened for alterations to protein expression in lung tissue at the onset of PH in a mouse model of hypoxia-induced PH.
Methods and Results— Using 2-dimensional polyacrylamide gel electrophoresis in combination with matrix-assisted laser desorption/ionization time-of-flight analysis, we identified 36 proteins that exhibited significantly altered expression after short-term hypoxic exposure. Among these, Fhl-1, which is known to be involved in muscle development, was one of the most prominently upregulated proteins. Further analysis by immunohistochemistry, Western blot, and laser-assisted microdissection followed by quantitative polymerase chain reaction confirmed the upregulation of Fhl-1, particularly in the pulmonary vasculature. Comparable upregulation was confirmed (1) after full establishment of hypoxia-induced PH, (2) in 2 rat models of PH (monocrotaline-treated and hypoxic rats treated with the vascular endothelial growth factor receptor antagonist SU5416), and (3) in lungs from patients with idiopathic pulmonary arterial hypertension. Furthermore, we demonstrated that regulation of Fhl-1 was hypoxia-inducible transcription factor dependent. Abrogation of Fhl-1 expression in primary human pulmonary artery smooth muscle cells by small-interfering RNA suppressed, whereas Fhl-1 overexpression increased, migration and proliferation. Coimmunoprecipitation experiments identified Talin1 as a new interacting partner of Fhl-1.
Conclusions— Protein screening identified Fhl-1 as a novel protein regulated in various forms of PH, including idiopathic pulmonary arterial hypertension.
Received December 26, 2007; accepted July 10, 2008.
Pulmonary hypertension (PH) is a fatal disease of multifactorial background characterized by increased pulmonary vascular resistance and right heart afterload, concomitant with right ventricular hypertrophy that may culminate in cor pulmonale. The pathological hallmark of many forms of PH is a vascular remodeling process resulting in narrowing and obstruction of small pulmonary arteries. Underlying structural changes are caused by increased migration and proliferation of smooth muscle cells (SMCs) and fibroblasts, as well as abnormal endothelial cell proliferation (for overviews, see Humbert1 and Jeffery and Wanstall2). Although the pathological changes in PH have been well defined, the etiology of this disease remains unclear. Genetic abnormalities account for 10% to 20% of all patients with idiopathic pulmonary arterial hypertension (IPAH), 50% of which arise from mutations in the bone morphogenetic protein receptor type II gene (BMPRII).3 In addition to other factors, hypoxia is one of the causes of PH.2,4 The hypoxia-inducible transcription factor (HIF) has been suggested as a key regulator of hypoxia-induced PH.5 Consequently, mice partially deficient for HIF-1α develop attenuated PH.6 Moreover, recent evidence has shown that HIF-1α is also involved in hypoxia-independent forms of PH.7
Clinical Perspective p 1194
Various studies have identified several factors, such as endothelin, serotonin, platelet-derived growth factor, metabolic alterations, and reactive oxygen species, to be involved in the vascular remodeling processes.7–13 To date, many investigators have focused on transcriptional changes that occur in PH;14,15 however, the full spectrum of pathophysiological mechanisms that result in PH is yet to be determined. In addition, most studies have analyzed late time points in experimental models of PH or end-stage human PH when the remodeling process is already fully established; therefore, factors triggering onset of the disease may be overlooked.14,16 Against this background, we hypothesized that key triggers of PH should be detectable at the onset of the disease. To test this hypothesis, we used a proteomic screening approach to analyze lungs from a mouse model of hypoxia-induced PH after only 24 hours of chronic hypoxia by 2-dimensional polyacrylamide gel electrophoresis (2D-PAGE) in combination with mass spectroscopy. The findings from this approach were then verified in different animal models of PH as well as in human IPAH.
Lung Preparation of Mice Under Hypoxia/Normoxia
Male BALB/c AnNCrlBR mice were exposed to normobaric normoxia (Fio2 of 0.21) and to normobaric hypoxia (Fio2 of 0.10) for 1, 7, and 21 days, respectively (n=5 each). The isolation and preparation of mouse lungs have been described in detail previously.15
Protein Isolation From Lung Homogenate for 2D-PAGE
For protein isolation, part of the right lung from 24-hour hypoxic and normoxic mice (n=5 each) was homogenized by grinding under liquid nitrogen. Lysis buffer (0.7 mL; 8 mol/L urea, 2 mol/L thiourea, 4% [m/v] CHAPS, 30 mmol/L dithiothreitol, 20 mmol/L Tris) was added to the ground tissue. Samples were concentrated by acetone precipitation. Proteins were resuspended in lysis buffer and centrifuged (18 000g, 30 minutes, 4°C). Protein concentration was determined with the 2D QUANT protein determination kit (Amersham Biosciences, Freiburg, Germany). Protein (250 μg) in the supernatants was subjected to isoelectric focusing.
Isoelectric Focusing and 2D-PAGE
The 2D-PAGE was performed according to O'Farrell17 and Görg et al18 with the following modifications. Solubilized proteins in lysis buffer (8 mol/L urea, 2 mol/L thiourea, 4% CHAPS, 2% Pharmalyte buffer [v/v, pH 3 to 10], 30 mmol/L dithiothreitol, 20 mmol/L Tris) were subjected to isoelectric focusing by rehydration of immobilized pH gradient strips (length, 11 cm; linear pH range, 3 to 10 (Amersham Biosciences). The second-dimension SDS gels contained 12.5% (v/v) acrylamide. Gels were stained with Coomassie brilliant blue, and bands were quantified by scanning on an image scanner (Amersham Biosciences). For details, please refer to the online-only Data Supplement.
Computer-assisted 2D analysis of 4 independent experiments was performed with Proteomweaver software version 2.2.2 (Definiens, Munich, Germany). For details, see the online-only Data Supplement.
In-Gel Digestion and Peptide Mass Fingerprinting
Gel pieces were rehydrated in a 10 ng/μL trypsin solution (sequencing grade; Roche Diagnostics, Mannheim, Germany). Aliquots (2 μL) of the solution were applied to a thin layer of α-cyano-4-hydroxycinnamic acid on an AnchorChip target (Bruker Daltonik, Bremen, Germany). Mass fingerprints of tryptic digests were obtained by matrix-assisted laser desorption/ionization (MALDI) mass spectrometry with the use of an Ultraflex time-of-flight (TOF)/TOF mass spectrometer (Bruker Daltonics). Proteins were identified by database searching with peptide masses with the use of the program Mascot at http://www.matrixscience.com. Protein identification was completed with significance set at P<0.05 for probability-based values on Mowse Scores (≥61). For details, see the online-only Data Supplement.
Accession numbers from genes regulated under hypoxic conditions were subjected to a screen for biological processes by using the Gene Ontology page (AmiGo: www.godatabase.org/cgi-bin/amigo/go.cgi).
Western Blot Analysis
For a description of Western blot protocol, please refer to the online-only Data Supplement.
For a detailed description of immunohistochemistry protocol, please refer to the online-only Data Supplement.
Microdissection was performed with the use of the Laser Microbeam System (P.A.L.M., Bernried, Germany), as described in detail previously.15
RNA Isolation and cDNA Synthesis
RNA isolation, cDNA synthesis, reagents, and incubation steps were performed as described previously.15
Relative mRNA Quantification by Real-Time Polymerase Chain Reaction
Real-time polymerase chain reaction (PCR) was used for relative quantification of the Fhl-1, Fhl-2, and Fhl-3 mRNA. The reactions were performed in an ABI 7700 Sequence Detection System (Applied Biosystems, Foster City, Calif) (see the online-only Data Supplement).
Rat Models of PAH
In rats, PH was induced either by monocrotaline or by exposure of rats to hypoxia (10% O2) and the vascular endothelial growth factor receptor antagonist SU5416, as described previously.11,19 Lung samples were obtained either after 4 weeks of monocrotaline injection (n=7 each) or 3 weeks after hypoxic+SU5416 exposure (n=4 each).
Patient Characteristics and Measurements
Samples from human lung tissue were obtained from 5 donors and 5 IPAH patients, as described previously.20
Preparation of Primary Pulmonary Artery SMCs
Primary pulmonary artery SMCs (PASMCs) were isolated as described previously.20
For immunofluorescence analysis, please refer to the online-only Data Supplement.
Full-length mouse Fhl-1 cDNA was cloned into pCMV3A-Myc (Stratagene) and pEGFP-C1 (Clontech) vectors. Full-length human Fhl-1 was cloned into pcDNA3.1(+) (Invitrogen). For more details, please see the online-only Data Supplement.
Transfection of Primary PASMCs
Transfection with plasmids and small-interfering RNA (siRNA) was performed as described previously.20 The siRNA sequences are provided in the online-only Data Supplement.
Cell proliferation of PASMCs was assessed by [3H]thymidine incorporation as described previously.20
After transfection, PASMCs were serum-starved for 24 hours and then plated at a concentration of 7.5×103 on 8.0-μm pores (Transwell inserts; Greiner-bio-one, Frickenhausen, Germany). To quantify the migrated cells, inserts were removed after 24 hours, the inner surface was scraped, and the outer surface was fixed and stained with crystal violet solution (Sigma, Munich, Germany).
For a description of the apoptosis assay, please refer to the online-only Data Supplement.
Hypoxia Response Element
The Fhl-1 gene was screened 5000 bp downstream and upstream from coding sequence for the presence of hypoxia response elements (HRE). The consensus sequence chosen for HRE was BACGTSSK, where B can be T, G, or C; S, G or C; and K, T or G. The Fhl-1 sequence was obtained from http://www.ncbi.nlm.nih.gov/mapview/.
Hypoxia Treatment of PASMCs
Hypoxia was induced in a chamber equilibrated with a water-saturated gas mixture of 1% O2, 5% CO2, and 94% N2 at 37°C. For details, please refer to Eul et al.21 The siRNA preparation for HIF-1α and HIF-2α was performed as described previously.21
Electrophoresis Mobility Shift Assay
Electrophoresis mobility shift assay was performed with nuclear extracts (Pierce, Rockford, Ill) from human PASMCs, which had been grown on 10-cm dishes for 12 hours at 1% or 21% O2, respectively. The oligonucleotide probe corresponding to the downstream HRE consensus sequence in the human promoter (5′-CAG TGG CGG GGG CAC GTG GGC GCG CGG GGT GCG-3′) was designed and labeled with γ-[32P]ATP with the use of T4 polynucleotide kinase (Fermentas, St Leon-Rot, Germany). Samples were separated on a native 4% polyacrylamide gel. The gels were vacuum dried and subjected to autoradiography at −80°C. For details, see the online-only Data Supplement.
For the coimmunoprecipitation protocol, please see the online-only Data Supplement.
Statistical analyses were performed with the use of R (www.r-project.org). Before statistical analyses, values obtained from technical replicates were averaged. Values are presented throughout as mean±SEM. Data were tested for deviations from normality with the Shapiro-Wilk test (α=0.05). Probability values were determined as a measure for the evidence of differences in means with the t distribution. Probability values for proliferation and migration data were calculated from normalized data to account for experiment-wise differences in cell numbers. Normalization was done by (xi,treated−xi,control)/(xi,control), where xi is measured values from experiment i. Means of >2 samples were compared by 1-way ANOVA. After significant ANOVA (α=0.05), probability values of the treatment-to-control contrasts were obtained with the Dunnett multiple-to-1 distribution. Nonnormal distributed data were analyzed by the Friedman test; probability values of pairwise comparisons were determined by Wilcoxon U distribution with Bonferroni adjustment. The respective paired analyses were done for data from matched controls and treatments. All probability values determined are given on the graphs.
The authors had full access to the data and take responsibility for its integrity. All authors have read and agreed to the manuscript as written.
2D-PAGE Analysis and Subsequent Identification of Regulated Proteins in Mouse Lungs After Short-term Hypoxic Exposure
After 2D-PAGE analysis, proteins differentially regulated in hypoxic or normoxic lung tissue were divided into 2 groups for further analysis: (1) upregulated in hypoxia (≥2-fold upregulation) and (2) downregulated in hypoxia (≥2-fold downregulation). Within these groups, 95 spots were found to be upregulated, and 95 spots were downregulated. In Figure 1A, examples of differentially regulated proteins with their own identification numbers are displayed. The fold changes versus the average spot intensities (“MA plot”) are illustrated in Figure 1B. The discrete distribution of spots at low-intensity values is due to the detection limit. Most of the proteins detected exhibited regulation factors ≤2 (90%). Proteins that displayed a spot intensity >0.1 were identified by MALDI-TOF mass spectroscopy and matched to the Mascot database (Table I in the online-only Data Supplement). The scores in Table I in the online-only Data Supplement represent the matching probability of detected peptides with theoretical database peptides. One of the most upregulated proteins (factor 6.6) was Fhl-1 (four-and-a-half LIM domain-1), which belongs to the muscle development group (Figure I in the online-only Data Supplement). Western blot of homogenized lungs confirmed the results from the 2D-PAGE analysis (Figure 1C). Interestingly, no significant differences in mRNA levels from lung homogenates were observed by real-time PCR (Figure 1D). To localize Fhl-1, immunostaining was performed on mouse lung tissue. The strongest expression of Fhl-1 was observed in vascular SMCs, whereas bronchial epithelial cells exhibited weaker staining (Figure 1E, Figure II in the online-only Data Supplement). No immunoreactivity was observed in the absence of primary antibodies (Figure 1E, control). A direct comparison of Fhl-1 expression in lung sections from normoxic and hypoxic mice revealed stronger staining in vessels from hypoxic animals (Figure 1F).
Regulation of Fhl-1 in Exposure to Sustained Hypoxia
To investigate whether pronounced Fhl-1 expression also occurred during prolonged hypoxia, we performed a Western blot analysis on mouse lung extracts from mice maintained in hypoxia for 7 and 21 days. As determined previously, PH is fully developed in mice within 21 days of hypoxic exposure.11,14 A time-dependent increase in Fhl-1 expression was observed (Figure 2A). Similar to 1 day of hypoxic exposure, Fhl-1 was primarily localized in intrapulmonary vessels (Figure 2B). Additionally, neomuscularized resistance vessels exhibited positive staining. To assess whether the differences in the vascular protein regulation of Fhl-1 were also detectable on the mRNA level, laser microdissection was performed to specifically isolate pulmonary arteries from mouse lung tissue. Quantitative mRNA analysis revealed an increased Fhl-1 expression in small pulmonary arteries of mice exposed to hypoxia for 1, 7, and 21 days in comparison to normoxia (Figure 2C). Analysis of alveolar septae revealed that the hypoxia-dependent upregulation of Fhl-1 mRNA was restricted to the pulmonary vasculature (Figure 2D). Hypoxic upregulation of Fhl-1 was specific to the pulmonary system because systemic aortas showed consistent downregulation of Fhl-1 at 1, 7, and 21 days of hypoxia exposure compared with normoxia (Figure 2E). A similar pattern was observed in carotid arteries (21 days of hypoxia, 2.49±0.40; normoxia, 3.50±0.28). Under normoxia, basal levels of Fhl-1 expression were similar in the pulmonary and systemic vasculature.
Fhl-1 Expression in Rat Models of PH
To further explore the regulation of Fhl-1 in PH, we investigated 2 different rat models of PH: (1) hypoxia-induced PH in combination with the vascular endothelial growth factor receptor antagonist SU5416 and (2) monocrotaline-induced PH. Both models result in severe PH.11,19 In chronic hypoxia-exposed rats injected with SU5416, Fhl-1 mRNA expression was upregulated compared with control rats (Figure 3A). Similarly, Fhl-1 protein expression was also increased (Figure 3B and 3C). Analogous results were observed for monocrotaline-induced PH (Figure 3D, 3E, and 3F).
Regulation of Fhl-1 in Patients With IPAH
To analyze whether the upregulation of Fhl-1 observed in animal models of PH is mirrored in human IPAH, real-time PCR analysis was performed on IPAH and donor lung samples. IPAH samples exhibited higher expression of Fhl-1 than donor samples (Figure 4A). Interestingly, we did not observe any changes in the expression levels of the related genes Fhl-2 and Fhl-3 in lung homogenate samples (Figure 4B and 4C). Additionally, increased Fhl-1 protein was observed by Western blot (Figure 4D and 4E). Immunohistochemical staining revealed expression of Fhl-1 predominantly in the vasculature (Figure 4F and Figure III in the online-only Data Supplement). Notably, intense staining for Fhl-1 was observed in plexiform lesions of IPAH patients, as depicted in Figure 4F. In microdissected pulmonary vessels from IPAH lungs, a strong increase in Fhl-1 mRNA was detected compared with donors (Figure 4G). A similar upregulation of Fhl-1 mRNA and protein levels was found in PASMCs isolated from IPAH compared with donor lungs (Figure 5A, 5B, and 5C). Immunofluorescence staining of Fhl-1 in human and mouse PASMCs revealed that Fhl-1 was predominantly restricted to the cytoplasm (Figure 5D and Figure IV in the online-only Data Supplement).
Hypoxia-Dependent Regulation of Fhl-1
To explore the molecular mechanisms of the transcriptional regulation of human Fhl-1, we subjected PASMCs to hypoxia (1% O2). In comparison to normoxic conditions, Fhl-1 demonstrated pronounced expression after 24 hours of hypoxic exposure (Figure 6A). We next addressed the possible roles of HIF-1 and HIF-2 in hypoxic Fhl-1 expression. The HIF constitutes a family of basic helix-loop-helix transcription factors that function as major gene regulators at low oxygen tension.22 To analyze the individual roles of the HIF subunits in the transcriptional regulation of Fhl-1, siRNA specifically targeting HIF-1α or HIF-2α was applied. The efficiency of siRNA knockdown was assessed by real-time PCR (data not shown). Knockdown of either HIF-1α or HIF-2α significantly reduced expression of Fhl-1 under hypoxia (Figure 6B). Consequently, we screened 5 kb of human promoter for HRE upstream and downstream from the coding sequence. Several putative HIF binding sites (BACGTSSK) were found in the promoter region of the Fhl-1 gene, as indicated in Figure V in the online-only Data Supplement. One of these sequences (CACGTGGG) was in close proximity to the transcription start site (Figure V in the online-only Data Supplement). To test whether HIF proteins bind to the identified HREs in the Fhl-1 promoter, an electrophoresis mobility shift assay was performed with nuclear extracts from normoxic and hypoxic (1% O2, 12 hours) human PASMCs. Binding activity was obtained at the HIF recognition sequence with nuclear extracts of hypoxic (1% O2) but not of normoxic (21% O2) PASMCs (Figure 6C).
Fhl-1 Is Involved in Migration and Proliferation of Primary PASMCs
As migration and proliferation of PASMCs are involved in the vascular remodeling underlying PH and because Fhl-1 expression was significantly increased in IPAH, we investigated the role of Fhl-1 in these processes. For this purpose, we designed siRNA directed against Fhl-1. As depicted on Figure 7A and 7B, expression of Fhl-1 was reduced by ≈50% to 70%, as assessed by real-time PCR and Western blot analysis. Fhl-1 knockdown resulted in significantly decreased migration and proliferation of human PASMCs compared with random siRNA transfected cells (Figure 7C and 7D). Similar results were observed in mouse PASMCs (Figure VIA, VIB in the online-only Data Supplement). Moreover, silencing of Fhl-1 protein led to lower expression of cyclinD1 (Figure 7E). To provide further evidence for a role of Fhl-1 in PASMC motility and proliferation, we cloned full-length and overexpressed Fhl-1 in primary PASMCs. Overexpression was confirmed by Western blot analysis (Figure 7F). Overexpression of Fhl-1 enhanced migration and proliferation of human PASMCs compared with cells transfected with control vector (Figure 7G, 7H and Figure VII in the online-only Data Supplement). In a manner similar to the endogenous protein, overexpressed Fhl-1 was localized in the cytoplasm of primary PASMCs (Figure 7I). Interestingly, silencing and overexpressing of Fhl-1 protein in human PASMC did not have any influence on apoptosis as indicated by flow cytometry analysis (Figure VIII in the online-only Data Supplement).
Identification of Talin1 as a Novel Interaction Partner of Fhl-1
To decipher the molecular mechanisms underlying the impact of Fhl-1 on migration and proliferation, we performed coimmunoprecipitation to screen for novel interaction partners of Fhl-1. Immunoprecipitates from NIH cells overexpressing either myc or Fhl-1-myc were compared by SDS-PAGE. One unique band in the Fhl-1 extracts corresponding to Talin1 was identified by MALDI-TOF analysis. These results were confirmed in PASMCs overexpressing Fhl-1-myc followed by Western blot against Talin1 (Figure 8A). To further investigate the interaction of Fhl-1 with Talin1, we analyzed the localization of Talin1 in human PASMCs and human lung tissue. At the single-cell level, Talin1 demonstrated strong colocalization in the cytoplasm with both endogenous (Figure 8B, a and b) and overexpressed Fhl-1 protein (Figure 8B, c). Talin1 and Fhl-1 partially colocalized with focal adhesion kinase (Figure 8B, d and e). Localization of Fhl-1 to actin filaments was further validated by phalloidin staining (Figure 8B, f). In addition, both Talin1 and Fhl-1 were detected in PASMCs of human lung tissue (Figure 8C). Similarly, to the depletion of Fhl-1, knockdown of Talin1 led to decreased migration and proliferation of human PASMCs (Figure 8D and 8E). The efficiency of siRNA knockdown was assessed by real-time PCR and Western blot analysis (Figure 8F and 8G).
Exposure to chronic hypoxia results in vascular remodeling in the mouse model of PH. We hypothesized that the analysis of the very early phase of hypoxia could be used to detect key triggers of this disease and that such findings may be relevant for the pathogenesis of PH. Using a 2D-PAGE approach, we identified a variety of proteins differentially regulated when comparing lungs from mice exposed to 24-hour hypoxia with those maintained under normoxic conditions. The majority of identified proteins (48%) belonged to the group “metabolism,” which consists of proteins responsible for protein biosynthesis, metabolism of aldehydes and fatty acids, peptide cross-linking, protein folding, and NADH oxidation and response to oxidative stress. In this context, it is noteworthy that metabolic alterations and oxidative stress have been suggested to be involved in the development of hypoxia- and non–hypoxia-induced PH.7,12,23,24 Interestingly, one of the most prominently upregulated proteins identified by 2D-PAGE was Fhl-1, allocated to the group “muscle development.” This protein belongs to the family of LIM proteins, being an acronym of the first 3 identified homeodomain containing factors, Lin-11, Isl-1, and Mec-3.25–27 The LIM domain is a cysteine-rich, double zinc finger motif that mediates protein-protein interactions of transcription factors, signaling, and cytoskeletal proteins. Proteins containing LIM domains play critical roles in mediating tissue differentiation, oncogenesis, and cytoskeletal organization.28–31 In the heart, Fhl-1 is strongly expressed in the developing outflow tract and to a lesser extent in myocardium. In cardiac hypertrophic and dilated cardiomyopathy mouse models, cardiac ventricular expression of Fhl-1, but not of related proteins Fhl-2 or Fhl-3, was altered.32 Additionally, increased Fhl-1 expression has been shown in stretch-induced muscle hypertrophy.32–34 Thus, we hypothesized that Fhl-1 may also be involved in pulmonary vascular remodeling.
In our study, the enhanced expression of Fhl-1 in hypoxia-induced PH was not only present in the early time point of hypoxia exposure (1 day) but was also sustained during prolonged hypoxia (7 and 21 days), when PH is already evident and fully established.11,14 Moreover, the specific role for Fhl-1 in hypoxia-induced PH is supported by our finding that Fhl-1 is upregulated in the pulmonary but not in the systemic vasculature. This is well in agreement with the different effects of hypoxia: increasing the pulmonary but decreasing the systemic vascular resistance. Similar to the sustained increase in Fhl-1 expression observed in the hypoxic mouse model, we found elevated Fhl-1 levels in lungs from 2 different rat models of PH: monocrotaline-induced PH and PH induced by the combined exposure to chronic hypoxia and the vascular endothelial growth factor receptor antagonist SU5416. Parallel results were observed in human IPAH. These data support our hypothesis that investigations during the onset of PH in animal models may be useful to identify new candidate proteins important for the development of human PH. Interestingly, in analogy to findings from mouse models of cardiac hypertrophy and dilated cardiomyopathy,32 we did not observe any differences in the expression levels of Fhl-2 or Fhl-3 in lungs from IPAH patients. Moreover, our detailed analysis revealed that Fhl-1 was expressed in PASMCs of pulmonary arteries and arterioles, a key site of vascular remodeling in PH. We also demonstrated that hypoxia-driven expression of Fhl-1 in human PASMCs is controlled by HIF-1α as well as HIF-2α. This is all the more interesting because HIF is involved in the pathogenesis of both hypoxia- and non–hypoxia-induced forms of PH, and high levels of HIF-1α were detected in arterial lesions in IPAH.5,7,35
To prove a functional role of Fhl-1 in vascular remodeling underlying PH, we demonstrated that inhibition of Fhl-1 expression by siRNA significantly decreased PASMC migration and proliferation, whereas overexpression of Fhl-1 had the opposite effect. These findings are in accordance with data from rat skeletal myoblasts in which overexpression of Fhl-1 led to higher migration and spreading of cells, but, to the best of our knowledge, are novel for the pulmonary circulation.36 To address the role of Fhl-1 in cell cycle kinetics, we investigated cyclinD1 levels in Fhl-1 silencing experiments. In proliferating cells, G1 phase progression depends on the sustained expression of D-type cyclins.37 Moreover, integrin-mediated cell adhesion, which may also play a role in vascular remodeling in PH, can also control the accumulation of cyclinD1.38 Knockdown of Fhl-1 resulted in lower cyclinD1 protein levels and therefore inhibition of G1 progression and consequently reduced proliferation.
To further decipher the molecular mechanism underlying Fhl-1 action, we focused on interaction partners of Fhl-1. To date, only 1 interacting partner, the major myosin thick filament–associated protein MyBP-C, has been identified.39 We hypothesized that the cellular activity of Fhl-1 may arise from protein-protein interactions because (1) Fhl-1 contains 4 and a half LIM domains, motifs that are involved in both intramolecular and intermolecular interactions,40 (2) LIM domain proteins can act as scaffolds on which multiprotein complexes are formed,41 and (3) several findings have shown that Fhl-2 and Fhl-3 are involved in regulation of actin cytoskeletal dynamics. By applying coimmunoprecipitation followed by MALDI-TOF, we identified Talin1 as a novel Fhl-1 interacting partner. This interaction is further supported by the fact that Talin1 contains LIM and actin binding domains.42,43 Talin is a major cytoskeletal protein that colocalizes with activated cytoplasmic domains of β-integrins44 and links them to the intracellular actin cytoskeleton. Several binding partners for Talin have been described; these include integrins, focal adhesion kinase, and F-actin.43,45,46 Downregulation of Talin1 in HeLa cells results in a decreased rate of cell spreading.47 In undifferentiated mouse embryonic stem cells, disruption of Talin1 prevents assembly of focal adhesions and causes defects in cell spreading.48 Moreover, decreased expression or inhibition of Talin attenuates the migration of fibroblasts and lymphocytes.49,50 In our study, silencing of Talin1 in human PASMCs resulted in decreased migration and proliferation, indicating a comparable cellular role of Fhl-1 and Talin1. Fhl-1 may alter the conformation of cytoskeletal molecules like Talin and actinin and therefore could play an important role in Talin-mediated regulation of integrin signaling and cytoskeletal organization as well as in establishing or maintaining connections between Talin and actin assembly (Figure 8H). The identification of the novel Fhl-1 binding partner, Talin1, together with the findings that Talin colocalizes with Fhl-1 in human PASMCs, suggests that this interaction has a significant role in integrin-actin communication and therefore in proliferation and migration of PASMCs.
In conclusion, we demonstrated that a proteomic approach using an animal model of hypoxia-induced PH at an early time point in the development of the disease could be useful to identify new candidates that are involved in the pathogenesis of PH. Our analysis revealed Fhl-1 as a new protein involved in the proliferation and migration of human PASMCs. In addition to various animal models, we confirmed an upregulation in human IPAH. In combination with cell-specific analysis, we demonstrated that the molecular mechanism of Fhl-1 function might involve Talin1. Fhl-1 thus may be considered a new player in the vascular remodeling processes underlying the development of PH.
The authors thank U. Seay for help with PASMC isolation, M.M. Stein and K. Quanz for technical support, and Rory E. Morty for careful reading of the manuscript.
Sources of Funding
This study was funded by a start-up grant (Anschubfinanzierungsprojekt) from the Faculty of Human Medicine of the Justus-Liebig University Giessen, Giessen, Germany; Deutsche Forschungsgemeinschaft, SFB 547, project B7, Z1; Excellence Cluster Cardiopulmonary System; and EU FP6 “PULMOTENSION” (LSHM-CT-2005-018725).
Durmowicz AG, Stenmark KR. Mechanisms of structural remodeling in chronic pulmonary hypertension. Pediatr Rev. 1999; 20: e91–e102.
Bonnet S, Michelakis ED, Porter CJ, Andrade-Navarro MA, Thebaud B, Haromy A, Harry G, Moudgil R, McMurtry MS, Weir EK, Archer SL. An abnormal mitochondrial-hypoxia inducible factor-1alpha-Kv channel pathway disrupts oxygen sensing and triggers pulmonary arterial hypertension in fawn hooded rats: similarities to human pulmonary arterial hypertension. Circulation. 2006; 113: 2630–2641.
Eddahibi S, Fabre V, Boni C, Martres MP, Raffestin B, Hamon M, Adnot S. Induction of serotonin transporter by hypoxia in pulmonary vascular smooth muscle cells: relationship with the mitogenic action of serotonin. Circ Res. 1999; 84: 329–336.
Archer SL, Huang J, Henry T, Peterson D, Weir EK. A redox-based O2 sensor in rat pulmonary vasculature. Circ Res. 1993; 73: 1100–1112.
Michelakis ED, Dyck JR, McMurtry MS, Wang S, Wu XC, Moudgil R, Hashimoto K, Puttagunta L, Archer SL. Gene transfer and metabolic modulators as new therapies for pulmonary hypertension: increasing expression and activity of potassium channels in rat and human models. Adv Exp Med Biol. 2001; 502: 401–418.
Xu W, Koeck T, Lara AR, Neumann D, DiFilippo FP, Koo M, Janocha AJ, Masri FA, Arroliga AC, Jennings C, Dweik RA, Tuder RM, Stuehr DJ, Erzurum SC. Alterations of cellular bioenergetics in pulmonary artery endothelial cells. Proc Natl Acad Sci U S A. 2007; 104: 1342–1347.
O'Farrell PH. High resolution two-dimensional electrophoresis of proteins. J Biol Chem. 1975; 250: 4007–4021.
Taraseviciene-Stewart L, Kasahara Y, Alger L, Hirth P, Mc Mahon G, Waltenberger J, Voelkel NF, Tuder RM. Inhibition of the VEGF receptor 2 combined with chronic hypoxia causes cell death-dependent pulmonary endothelial cell proliferation and severe pulmonary hypertension. FASEB J. 2001; 15: 427–438.
Zakrzewicz A, Hecker M, Marsh LM, Kwapiszewska G, Nejman B, Long L, Seeger W, Schermuly RT, Morrell NW, Morty RE, Eickelberg O. Receptor for activated C-kinase 1, a novel interaction partner of type II bone morphogenetic protein receptor, regulates smooth muscle cell proliferation in pulmonary arterial hypertension. Circulation. 2007; 115: 2957–2968.
Eul B, Rose F, Krick S, Savai R, Goyal P, Klepetko W, Grimminger F, Weissmann N, Seeger W, Hanze J. Impact of HIF-1alpha and HIF-2alpha on proliferation and migration of human pulmonary artery fibroblasts in hypoxia. FASEB J. 2006; 20: 163–165.
Mittal M, Roth M, Konig P, Hofmann S, Dony E, Goyal P, Selbitz AC, Schermuly RT, Ghofrani HA, Kwapiszewska G, Kummer W, Klepetko W, Hoda MA, Fink L, Hanze J, Seeger W, Grimminger F, Schmidt HH, Weissmann N. Hypoxia-dependent regulation of nonphagocytic NADPH oxidase subunit NOX4 in the pulmonary vasculature. Circ Res. 2007; 101: 258–267.
Brown MC, Perrotta JA, Turner CE. Identification of LIM3 as the principal determinant of paxillin focal adhesion localization and characterization of a novel motif on paxillin directing vinculin and focal adhesion kinase binding. J Cell Biol. 1996; 135: 1109–1123.
Hwang DM, Dempsey AA, Wang RX, Rezvani M, Barrans JD, Dai KS, Wang HY, Ma H, Cukerman E, Liu YQ, Gu JR, Zhang JH, Tsui SK, Waye MM, Fung KP, Lee CY, Liew CC. A genome-based resource for molecular cardiovascular medicine: toward a compendium of cardiovascular genes. Circulation. 1997; 96: 4146–4203.
West J, Fagan K, Steudel W, Fouty B, Lane K, Harral J, Hoedt-Miller M, Tada Y, Ozimek J, Tuder R, Rodman DM. Pulmonary hypertension in transgenic mice expressing a dominant-negative BMPRII gene in smooth muscle. Circ Res. 2004; 94: 1109–1114.
Zhu X, Ohtsubo M, Bohmer RM, Roberts JM, Assoian RK. Adhesion-dependent cell cycle progression linked to the expression of cyclin D1, activation of cyclin E-cdk2, and phosphorylation of the retinoblastoma protein. J Cell Biol. 1996; 133: 391–403.
McGrath MJ, Cottle DL, Nguyen MA, Dyson JM, Coghill ID, Robinson PA, Holdsworth M, Cowling BS, Hardeman EC, Mitchell CA, Brown S. Four and a half LIM protein 1 binds myosin-binding protein C and regulates myosin filament formation and sarcomere assembly. J Biol Chem. 2006; 281: 7666–7683.
Feuerstein R, Wang X, Song D, Cooke NE, Liebhaber SA. The LIM/double zinc-finger motif functions as a protein dimerization domain. Proc Natl Acad Sci U S A. 1994; 91: 10655–10659.
Arber S, Caroni P. Specificity of single LIM motifs in targeting and LIM/LIM interactions in situ. Genes Dev. 1996; 10: 289–300.
Coutts AS, MacKenzie E, Griffith E, Black DM. TES is a novel focal adhesion protein with a role in cell spreading. J Cell Sci. 2003; 116: 897–906.
Albiges-Rizo C, Frachet P, Block MR. Down regulation of talin alters cell adhesion and the processing of the alpha 5 beta 1 integrin. J Cell Sci. 1995; 108(pt 10): 3317–3329.
Priddle H, Hemmings L, Monkley S, Woods A, Patel B, Sutton D, Dunn GA, Zicha D, Critchley DR. Disruption of the talin gene compromises focal adhesion assembly in undifferentiated but not differentiated embryonic stem cells. J Cell Biol. 1998; 142: 1121–1133.
Nuckolls GH, Romer LH, Burridge K. Microinjection of antibodies against talin inhibits the spreading and migration of fibroblasts. J Cell Sci. 1992; 102 (pt 4): 753–762.
Smith A, Carrasco YR, Stanley P, Kieffer N, Batista FD, Hogg N. A talin-dependent LFA-1 focal zone is formed by rapidly migrating T lymphocytes. J Cell Biol. 2005; 170: 141–151.
Pulmonary arterial hypertension is a life-threatening disease that is characterized by aberrant pulmonary vascular remodeling. Pathological alterations of the vasculature include dysregulated smooth muscle cell migration and proliferation in the pulmonary vessel media, which require alterations in cytoskeletal organization. To date, several factors involved in vascular remodeling remain to be identified and characterized. In the present study, a proteome screening approach was used to identify four-and-a-half LIM domain protein (Fhl-1) as a novel molecule involved in pulmonary arterial smooth muscle cell proliferation. This protein belongs to the family of LIM proteins that contain a cysteine-rich, double zinc finger motif that mediates protein-protein interactions of transcription factors, signaling, and cytoskeletal proteins. Proteins containing LIM domains mediate tissue differentiation, oncogenesis, and cytoskeletal organization. In hypoxia-induced pulmonary arterial hypertension, Fhl-1 expression was enhanced at both the onset of the disease and when disease was fully established. In pulmonary arteries from patients with idiopathic pulmonary arterial hypertension, strong upregulation of Fhl-1 was noted compared with healthy donor lungs. The cellular role of Fhl-1 was deciphered by identification of the cytoskeletal protein Talin1 as a novel binding partner of Fhl-1. Both molecules colocalized in pulmonary arterial smooth muscle cells in vitro and in vivo and triggered pulmonary arterial smooth muscle cell migration and proliferation, suggesting that increased levels of Fhl-1 alter cytoskeletal dynamics via interaction with Talin. Furthermore, these observations suggest Fhl-1 as a novel molecular target with therapeutic efficacy in patients with pulmonary arterial hypertension.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.107.761916/DC1.