Heterozygous germline mutations in the gene encoding the bone morphogenetic protein type II receptor cause familial pulmonary arterial hypertension (PAH). We previously demonstrated that the substitution of cysteine residues in the ligand-binding domain of this receptor prevents receptor trafficking to the cell membrane. Here we demonstrate the potential for chemical chaperones to rescue cell-surface expression of mutant BMPR-II and restore function. HeLa cells were transiently transfected with BMPR-II wild type or mutant (C118W) receptor constructs. Immunolocalization studies confirmed the retention of the cysteine mutant receptor mainly in the endoplasmic reticulum. Co-immunoprecipitation studies of Myc-tagged BMPR-II confirmed that the cysteine-substituted ligand-binding domain mutation, C118W, is able to associate with BMP type I receptors. Furthermore, following treatment with a panel of chemical chaperones (thapsigargin, glycerol or sodium 4-phenylbutyrate), we demonstrated a marked increase in cell-surface expression of mutant C118W BMPR-II by FACS analysis and confocal microscopy. These agents also enhanced the trafficking of wild-type BMPR-II, though to a lesser extent. Increased cell-surface expression of mutant C118W BMPR-II was associated with enhanced Smad1/5 phosphorylation in response to BMPs. These findings demonstrate the potential for rescue of mutant BMPR-II function from the endoplasmic reticulum. For the C118W mutation in the ligand-binding domain of BMPR-II, cell-surface rescue leads to at least partial restoration of BMP signalling. We conclude that enhancement of cell-surface trafficking of mutant and wild-type BMPR-II may have therapeutic potential in familial PAH.
Idiopathic pulmonary arterial hypertension (PAH) is a rare but devastating condition affecting 2–3 people per million each year (1). The condition is familial in 6–10% of cases. Without treatment, life expectancy is <3 years from diagnosis. PAH results from extensive remodelling of the pulmonary vasculature, caused by proliferation and migration of fibroblasts, endothelial and smooth muscle cells. Increased muscularization of small arteries and fibrosis of the intima leads to obliteration of small pulmonary arteries (2,3). The resulting increase in pulmonary vascular resistance ultimately leads to right heart failure.
Familial PAH is caused in 80% of cases by heterozygous germline mutations in the bone morphogenetic protein type II receptor (4,5), which is a receptor member of the transforming growth factor-β superfamily. Mutations in BMPR-II contribute to altered growth responses to BMPs, the nature of which are also cell-type specific (6). A number of disease-causing mutations in BMPR-II have been described and of these ∼30% are missense mutations in highly conserved amino acids in functionally important domains. Interestingly, the substitution of cysteine residues within the ligand-binding domain or kinase domain leads to reduced trafficking of the mutant receptor to the cell surface (7,8). This suggests that missense cysteine substitutions in the ligand-binding and kinase domains of BMPR-II may not only affect receptor activation but also impact protein processing and trafficking. Previous work from our laboratory has demonstrated that the cysteine-substituted ligand-binding domain mutant of BMPR-II, C118W, fails to reach the cell surface and is therefore unable to bind ligand, although it does posses an active kinase domain (7,9).
Signalling by wild-type BMPR-II involves heterodimerization with the transmembrane serine/threonine kinases type I BMPR-IA and -1B receptors at the cell membrane. BMP receptors show different affinities for BMP ligands. The type I receptors have high affinity for the majority of ligands (BMP-2, BMP-4, BMP-7 and GDF-5 and GDF-6), whereas BMPR-II binds with high-affinity receptor to BMP6 (10,11). On ligand binding, the constitutively active BMPR-II phosphorylates the type I receptor. Activated type I receptors phosphorylate the cytoplasmic signalling proteins known as receptor-mediated Smads (R-Smads) 1,5 and 8, which complex with the common partner Smad, Smad4, and translocate to the nucleus, where they activate downstream target genes (12). In the case of an intracellular pool of mutant receptor, signalling via Smads is compromised, as mutant receptor does not have access to extracellular ligand. The possibility exists therefore that mutant C118W BMPR-II could be trafficked to the cell membrane, where it may partially restore signalling.
The approach to improve cellular trafficking of mislocalized proteins has been explored in a variety of systems including studies on the vasopressin receptor (13–16) and the cystic fibrosis transmembrane regulator (CFTR). In the latter example, the trafficking of mutant CFTR protein to the cell membrane has been induced with the functional rescue of chloride channel activity in vitro (17–19). In addition, randomized control trials using the chemical chaperone, sodium 4-phenylbutyrate, in patients with cystic fibrosis have also shown promise (20,21). Historically, the treatment of familial PAH has been limited to agents that prevent or reverse vasoconstriction, such as calcium channel antagonists and prostacyclin. New therapies include endothelin receptor antagonists and phosphodiesterase inhibitors (22). Improved trafficking and the restoration of the signalling of mutant BMPR-II would provide a novel experimental approach to the treatment of familial PAH. In addition, since other forms of PAH are associated with the reduced expression of BMPR-II (3), targeting BMPR-II trafficking may have broader applications than the familial forms of the disease.
To provide proof-of-concept for the functional rescue of C118W BMPR-II by chemical chaperones, we employed transient and stable overexpression of wild-type and mutant BMPR-II in cell lines. Immunolocalization studies demonstrated that the mutant receptor was predominantly localized to the endoplasmic reticulum. Co-immunoprecipitation confirmed the ability of C118W BMPR-II to associate with the type IB receptor. Following treatment with a panel of chemical chaperones (sodium 4-phenyl butyrate [4-PBA], glycerol and thapsigargin), we demonstrate that C118W BMPR-II was trafficked to the cell surface through conventional protein trafficking pathways. The increase in C118W BMPR-II cell membrane expression led to the restoration of Smad signalling. These findings provide a basis for the functional rescue of mistrafficked BMPR-II mutants in familial PAH.
C118W BMPR-II mutant has reduced signalling capacity due to retention in the endoplasmic reticulum
We first confirmed that C118W BMPR-II and wild-type BMPR-II demonstrated distinct patterns of subcellular localization in transfected HeLa cells. HeLa cells transfected with GFP-tagged wild-type BMPR-II showed predominantly cell-surface localization (Fig. 1A). The level of overexpression of wild type and C118W BMPR-II was similar in these studies (data not shown). In contrast, GFP-tagged C118W BMPR-II demonstrated localization mainly in a perinuclear distribution. Co-localization studies for the ER marker, KDEL, confirmed that C118W BMPR-II was retained in the ER (Fig. 1Ba). To determine whether massive overexpression played a factor in the intracellular retention of C118W BMPR-II in the transient system, we generated immortalized MRC5(SV40T) lung fibroblasts stably expressing C118W-GFP and determined the relative expression of BMPR-II in the stable and transiently transfected cells (Fig. 1Bb). Co-localization of C118W BMPR-II with the ER also occurred in the stable cell line, but not with the wild-type receptor, indicating that massive overexpression was not the cause of the mislocalization of mutant BMPR-II (Fig. 1Bc). Staining for markers of cis- and trans-Golgi showed no localization of mutant protein in these compartments (Fig. 1Bd). Following transient transfection of HeLa cells with C118W, we confirmed that BMP4-stimulated Smad1/5 phosphorylation was reduced compared with cells transfected with wild-type BMPR-II (Fig. 1C). FACS analysis of Myc-tagged receptors following the permeabilization of transfected HeLa cells revealed that the majority of the mutant receptors was located intracellularly (Fig.1Da and b). Of the transfected population, C118W BMPR-II-containing cells showed a significantly reduced cell-membrane expression, 31.9 ± 6.7%, compared with wild-type BMPR-II, 61.6 ± 3.5% (mean ± SEM, n = 10, *P < 0.05) (Fig.1Dc).
C118W BMPR-II exerts a dominant-negative effect on BMP type I receptor trafficking
We have previously reported that C118W has a dominant-negative effect on BMP4 activation of a BMP-responsive reporter gene construct (7). To identify a potential mechanism for this, we performed co-transfection studies in HeLa cells with GFP-tagged C118W BMPR-II and Myc-tagged wild-type BMPR-1A or BMPR-IB. Immunofluorescence staining for Myc demonstrated co-localization and retention of mutant C118W-GFP with BMPR-II and BMP type I receptors within the cytoplasm (Fig. 1Eb). In contrast, wild-type BMPR-II co-localized with BMP type I receptors at the cell membrane (Fig.1Ea). Interestingly, overexpression of mutant C118W BMPR-II did not appear to cause intracellular retention of wild-type BMPR-II (Fig. 1A).
C118W BMPR-II associates normally with BMP type I receptors
We have previously shown that C118W BMPR-II associates normally with the BMPR-IA receptor by co-immunoprecipitation (9). To confirm this observation for the type BMPR-IB receptor, we transiently co-transfected HeLa cells with BMPR-II-Myc and BMPR-IB-HA. Following immunoprecipitation for the Myc tag and western blot analysis for the HA tag, a band corresponding to the appropriate size of the type I receptor was demonstrated (Fig. 2). Conversely, immunoprecipitation using an anti-HA antibody, followed by probing of the western blot using an anti-Myc antibody, demonstrated a band corresponding to BMPR-II (Fig. 2). Immunoprecipitation of the type I receptor was unimpaired whether mutant BMPR-II (C118W and C483R) or wild-type BMPR-II was transfected. After stripping, HA-IP/Myc-blots were re-probed for HA, and Myc-IP/HA-blots were re-probed for Myc to confirm equal loading of the immunoprepitated target protein.
Chemical chaperones increase the cell-surface expression of C118W BMPR-II
We employed a range of agents known to influence protein trafficking. HeLa cells transfected with Myc-tagged, wild-type or C118W BMP-RII were exposed to thapsigargin, glycerol or 4-PBA. Cell-surface Myc was detected with a secondary FITC conjugated antibody and imaged by confocal microscopy. All three agents were shown to markedly improve the cell-surface localization of C118W (Figs. 3A–C). These agents also appeared to increase the localization of wild-type BMPR-II, though the magnitude of the change was less than that observed with C118W BMPR-II. To confirm that the block in C118W transport was predominantly pre-Golgi, we repeated studies in the presence of Brefeldin A, which inhibits anterograde protein transport from the ER. These studies confirmed that Brefeldin A could prevent the enhancement of BMPR-II cell-surface localization induced by 4-PBA (Fig.3C). The trafficking effect of 4-PBA was also apparent in the C118W and wild-type stable cell lines with altered localization of mutant GFP compared to wild-type.
To quantify the magnitude of the trafficking induced by chemical chaperones, we employed FACS analysis of cell-surface Myc-tagged wild-type and C118W BMPR-II in transfected HeLa cells. All agents led to a significant increase in the cell-surface expression of Myc-C118W BMPR-II. Again, a lesser, but significant effect was observed on the trafficking of wild-type BMPR-II (Fig. 4C) (*P < 0.001, n = 6, multiple comparison ANOVA). Plasmid-containing cells were defined as cells taking up the respective Myc-tagged plasmid DNA following transfection and was determined by cell permeabilization and total cell staining for the Myc tag.
Enhanced C118W BMPR-II trafficking partially restores Smad signalling
A further important consideration is whether the rescue of retained C118W to the cell surface results in improved signalling via the BMP/Smad pathway. Following transient transfection of HeLa cells with C118W, we assessed the activation of Smad1/5 by BMPs before and after treatment with 4-PBA or glycerol. Western blotting for phospho-Smad1/5 demonstrated a significant increase in the phospho-Smad1/5 signal in response to BMP4 or BMP6 following 4-PBA treatment (Fig. 5A and B) or in response to BMP6 following pre-treatment with glycerol (Fig. 5C). An increase was also observed following glycerol pretreatment and subsequent BMP4 stimulation (data not shown). In addition, we addressed whether 4-PBA or glycerol could increase Smad1/5 signalling in cells transfected with wild-type BMPR-II. The increase in phospho-Smad1/5 wild-type BMPR-II following 4-PBA pre-treatment was less consistent; a significant increase in BMP4- but not in BMP6-induced phospho-Smad1/5 occurred, whereas glycerol pre-treatment appeared to have no effect on BMP4 or BMP6-induced phospho-Smad1/5 (data not shown). Untransfected HeLa cells alone showed no increases in BMP4 or 6-induced phospho-Smad 1/5 responses with either 4-PBA or glycerol pre-treatment (data not shown).
To provide further evidence that the increase in phospho-Smad1/5 signal achieved following treatment with trafficking agents was due to transfected mutant receptor but not to endogenous BMPR-II system, a comparative study was conducted with C118W and the kinase-deficient mutant, C483R. The cysteine kinase domain mutant, C483R, is also retained intracellularly, as shown by immunofluorescence (Fig. 6A) and following treatment with 4-PBA was also shown to be trafficked to the cell membrane via FACS analysis (Fig.6B). Importantly, in vitro kinase studies have previously shown that C483R is unable to phosphorylate the type I receptor and therefore incapable of signalling through Smads 1/5/8 (9). Figure 6Ca demonstrates the dominant-negative effect of the C483R mutant on phospho-Smad 1/5 signalling compared with HeLa cells transfected with wild-type BMPR-II. In addition, the transfection of the C483R mutant reduced BMP4-stimulated phospho-Smad 1/5 levels compared with vector control, indicating that the phospho-Smad 1/5 signal in C483R-transfected HeLa cells is due to endogenous/background levels of BMPR-II (Fig. 6Cb). Figure 6Cc shows that HeLa cells transfected with C483R BMPR-II followed by treatment with 4-PBA do not show an elevation in phospho-Smad1/5 levels in response to BMP4/6 compared with BMP4/6 stimulation alone. Taken together these findings demonstrate that following 4-PBA treatment, the increase in phospho-Smad 1/5 signalling after BMP4/6 stimulation is not due to the enhanced trafficking of the endogenous wild-type receptor but indicates that when C118W BMPR-II is expressed at the cell surface it is capable of responding to ligand.
The aim of this study was to provide proof-of-concept that the enhancement of mutant BMPR-II trafficking might provide a novel therapeutic approach in familial PAH. Previous studies from our group, and others, have confirmed that a class of cysteine-substituted ligand-binding domain mutants are retained inside the cell (7,8). The intracellular retention of this mutant is thought to be responsible for the reduced signalling via Smad1/5 following transfection into cell lines. However, the ability of these mutants to signal and to associate with type I receptors has not previously been explored in detail.
Consistent with previous findings, detailed immunolocalization studies confirmed a distinct pattern of localization between wild-type and the mutant C118W BMPR-II. We confirmed the intracellular retention of C118W BMPR-II in both transient and stably transfected cell lines. Through these immunolocalization studies, we further confirmed that the mutant receptor was localized to the endoplasmic reticulum, rather than Golgi. Previous studies (7,9) have shown that the C118W mutant exerts a dominant-negative effect on BMP signalling. Our results provide a basis for this observation. Thus, C118W mutant BMPR-II receptor caused co-retention of the type IA and IB receptors within the ER. Since BMP type I receptors are an integral part of the BMPR-II signalling axis, their ER retention would likely have a significant impact on BMP signalling over and above the loss of BMPR-II. Our results also raise the possibility that the type I and type II receptors may associate in the ER and be trafficked together, though they exist at the cell surface as BMPR-II homodimers and as heterodimers with type I receptors (23). We further showed that C118W mutant BMPR-II did not appear to retain co-expressed wild-type BMPR-II in the ER.
To provide justification for attempting to rescue cell-surface expression of C118W mutant BMPR-II, we first sought evidence that the C118W mutant possesses the characteristics of a functional BMP receptor. We confirmed, using co-immunoprecipitation studies, that mutant C118W BMPR-II was able to associate normally with the BMP-IB receptor. This is in keeping with our previous finding that C118W is capable of associating normally with the BMPR-IA receptor and phosphorylating the type I receptor (9).
Having provided further evidence for the functional integrity of the C118W BMPR-II mutant, we explored the potential for enhancing trafficking of this ER-retained protein to the cell surface. We employed a panel of agents that have previously been shown to improve protein trafficking via a number of mechanisms. We observed an increase in cell-surface localization of C118W BMPR-II and, to a lesser extent, wild-type BMPR-II following the exposure of transiently transfected HeLa cells to 4-PBA, glycerol and thapsigargin. This was inhibited by Brefeldin A, indicating conventional protein transport through ER and Golgi compartments in the cell. The increase in C118W BMPR-II receptor trafficking was associated with an increase in phospho-Smad1/5 levels following agonist stimulation. We were concerned that the increase in BMP-stimulated phospho-Smad1/5 activity could be explained by improved trafficking of endogenous wild-type BMPR-II. However, this seems unlikely because although trafficking agents improved the cell-surface localization of transfected wild-type BMPR-II, they had small and inconsistent effects on signalling, probably due to the relatively small increase in wild-type BMPR-II cell membrane expression compared with C118W BMPR-II. Moreover, we performed comparative studies with another ER-retained cysteine mutation of BMPR-II, C483R, which is deficient in kinase activity (9). The trafficking of this mutant could also be encouraged by incubation with trafficking agents but with no resulting increase in BMP-stimulated Smad1/5 phosphorylation. Taken together, these studies indicate that the improved signalling associated with the rescue of C118W BMPR-II is due to the presence of this receptor at the cell surface. However, it remains unclear whether the cysteine substitution present in the extracellular ligand-binding domain of the C118W mutant has implications for ligand binding/affinity. An increase in the molecular weight following chemical chaperone treatment, due in part to N-linked glycosylation (data not shown), also indicated that C118W BMPR-II could have been post-translationally modified following chemical chaperone treatment, suggesting the passage of mutant protein through the ER and Golgi to the cell membrane.
The retention of mutant proteins within the ER is a common feature in diseases such as cystic fibrosis and diabetes insipidus (24–27). Accumulation of mutant protein in the ER may be a result of abnormal protein processing and recognition, altered degradation pathways, an ER stress response or changes in the ER exit pathway. The ER is intrinsically responsive to calcium concentrations, and the depletion of calcium stores in the ER by the SERCA calcium pump inhibitor thapsigargin has been shown to be effective in trafficking delta F508 CFTR and vasopressin receptors to the cell membrane (16,19) although conflicting evidence exits (28–31). A limited and maintained ER calcium level has been shown to inhibit chaperone/protein interaction and allow the correction of abnormal protein trafficking (17). In addition to calcium concentrations, chaperone activity can also be modified by agents such as 4-PBA, which reduces the expression of hsc70 and increases HSP70, thus allowing trafficking of deltaF508 CFTR to the cell membrane (18,20,32,33). 4-PBA has also been implicated in suppressing ER stress (34). Other ‘chemical chaperones’ such as glycerol and organic solutes are thought to non-selectively stabilize mutant proteins and facilitate their folding (35,36).
These results provide ‘proof-of-concept’ that functional mutant C118W BMPR-II can be rescued to the cell membrane (Fig. 7). It remains to be investigated how much corrected mutant BMPR-II must reach the cell membrane to induce a clinically relevant effect. In conclusion, these data highlight a novel finding on the trafficking of a specific missense mutation of BMPR-II implicated in familial PAH. Elucidating the intracellular-signalling pathways that control the passage of mutant BMPR-II through the cell following chemical and pharmacological manipulation may lead to mutation-specific therapies for familial PAH patients.
MATERIALS AND METHODS
Cell culture conditions and reagents
HeLa cells (ATCC) were grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1× antibiotic–antimycotic at 37°C with 5% CO2, all supplied by Gibco. 4-PBA was purchased from Calbiochem. Thapsigargin and glycerol were purchased from Sigma. PGNase F was purchased from New England Biolabs.
The preparation of the wild-type and mutant pEGFP-N1-BMPR-II and plasmids has been described previously (7). The 3′-haemagglutinin (HA)-tagged pcDNA3.0-BMP-R1A (BMPRIA-HA) and –BMP-R1B (BMPRIB-HA) were generous gifts from Professor K. Miyazono (Tokyo, Japan). A 5′-Myc-tagged BMP-RII wild-type construct (Myc tag=EQKLISEEDL) was prepared by removing the BMPR-II short-form-coding sequence from the Myc-BMPR-IISF plasmid and inserting the full BMPRII-coding sequence. Mutant Myc-tagged BMPR-II plasmids were then constructed by site-directed mutagenesis using the Stratagen QuikChange protocol. Sequences of the receptor and tag sequences of the wild-type and mutant constructs were confirmed using an ABI377 sequencer with the Applied Biosystems Big Dye terminator kit DNA sequencing.
Plasmids used in transfection studies were prepared using Qiagen Plasmid EndoFree Maxi Kit, according to the manufacturer’s instructions. Transient transfections were performed on cells using Lipofectamine 2000 (LF-2000) transfection reagent (Invitrogen), according to the manufacturer’s protocol. Cells were incubated for at least 24 h before media were exchanged and transgene expression was assayed at least 48 h post-transfection. Typically, 20–50% transfection efficiency of the transgene was achieved and equivalent transfection efficiency confirmed via western blotting for BMPR-II or the Myc tag.
Generation of stable cell lines
Immortalized MRC5(SV40T) lung fibroblasts were transfected in Optimem I with pEGFP-N1 containing the full coding sequence for human wild-type BMPR-II or C118W BMPR-II using lipofectamine. Forty-eight hours post-transfection, the medium was exchanged to geneticin selection medium comprising EMEM supplemented with 10%FBS, 1 mml-glutamine, 100 units/ml penicillin, 100 µg/ml streptomycin, 0.25 μg/ml amphotericin B and 400 ug/ml geneticin. After 1 week, cells were plated at ∼0.3 cells/well in 96-well plates and clones expanded from single cells and maintained in geneticin selection medium. At early passage, we confirmed GFP fluorescence using flow cytometry, and the expression of GFP-tagged BMPR-II proteins by western blotting using antibodies against BMPR-II (BD Biosciences Pharmingen, Cowley, Oxfordshire, UK) and GFP (Roche). The relative expression of BMPR-II was also determined with quantitative PCR as described previously (11). We further confirmed the identities of the GFP-BMPR-II bands on western blots in clones that had been transfected with DharmaFECT1™ reagent alone, or DharmaFECT1™ complexed with a specific BMPR-II siRNA (BMPR-II siGenome™ Smartpool®) or a control siRNA (On-TARGETplus siControl Non-targeting Pool DharmaFECT1™), all from Perbio Science UK Ltd (Chester, Cheshire, UK). The presence of the plasmid-derived wild-type or C118W mutation in the clones used was confirmed by PCR using specific primers to amplify the plasmid-derived BMPR-II sequence followed by sequencing of the PCR product.
Expression and immunoprecipitation of Myc- and Ha-tagged receptors
HeLa cells were plated onto 100 mm culture dishes at 6 × 106 cells per plate and grown overnight to reach 90% confluence in DMEM/10% FBS, which was then changed immediately prior to transfection, as described previously. After 48 h, cells were washed with PBS and scraped into lysis buffer [20 mm Tris–HCl, pH 7.5, 150 mm NaCl, 50 mm NaF, 20 mm β-glycerophosphate, 2.5% (v/v) IGEPAL-CA630, 1 mm EDTA, 1 mm PMSF, 1 mm Na3VO4, 1 mm DTT, 10 µg/ml antipain, 10 µg/ml leupeptin, 10 µg/ml pepstatin, 10 µg/ml aprotinin] and the lysates centrifuged at 14 000g for 20 min at 4°C. For immunoprecipitation, the supernatant was incubated with either mouse anti-Myc (1:50 dilution, clone 9E10, Santa Cruz) or mouse anti-HA (1:100 dilution, Clone HA-7, Sigma) on a rotary mixer for 2 h at 4°C. Antibody:protein complexes were then isolated by incubation with Protein-G-Sepharose beads on a rotary mixer at 4°C overnight. The beads were washed three times in lysis buffer, resuspended in 1× loading buffer [62.5 mm Tris–HCL, pH 6.8, 10% (v/v) glycerol, 2% (w/v) SDS, 5% (v/v) β-mercaptoethanol, 0.003% (w/v) bromophenol blue] and boiled for 5 min followed by centrifugation at 2500g for 2 min at 4°C. Proteins were then separated on 10% SDS–PAGE followed by transfer to a nitrocellulose membrane. For Myc detection, membranes were blocked with 3% (w/v) BSA/2% (w/v) non-milk fat/0.1% Tween-20 for 1 h at room temperature. After washing, membranes were incubated with mouse anti-Myc (1:100, Santa Cruz) in 1.5% (w/v) BSA/1% (w/v) non-milk fat/0.1% Tween-20 overnight at 4°C. Blots were then washed and incubated with an anti-mouse horseradish peroxidase-conjugated secondary antibody (1:2500, Dako) in 1.5% (w/v) BSA/1% (w/v) non-milk fat/0.1% Tween-20 for 1 h at room temperature. Bands were visualized by chemiluminescence (Amersham, UK). For HA detection, membranes were blocked with 5% (w/v) non-milk fat/0.1% Tween-20 for 1 h at room temperature. After washing, membranes were incubated with mouse anti-HA (1:1000, Sigma) in 2% (w/v) BSA/0.1% Tween-20 overnight at 4°C. The secondary antibody and chemiluminescence stages were identical to the Myc protocol defined earlier. To determine loading, blots were stripped and re-probed for either HA or Myc. Co-immunoprecipitation studies were repeated in three independent experiments.
Subcellular localization of BMPR-II
HeLa cells were seeded onto glass chamberslides (Nunc) at 5 × 104 cells per well. Cells were then transiently transfected, as described previously, with pcDNA 3.0-BMPR-II 5′-Myc-tagged, Pegfp-N1-BMPR-II, plasmids, either mutant or wild-type BMPR-II and pcDNA3.0-BMP-R1A/B HA-tagged plasmids. Transfectants were grown for 48 h in DMEM/10% FBS and fixed in ice-cold methanol (10 min). All primary antibodies were added for 45 min at room temperature. Mouse anti-HA (Sigma) and mouse anti-KDEL antibodies (Santa Cruz) were used at a concentration of 1:250; mouse anti-GM180 (cis-Golgi) or p230 (trans-Golgi) (both from BD Biosciences) at a concentration of 1:500 and rabbit anti-mouse-TRITC (DAKO) at 1:500 or AlexaFluor 660 (Invitrogen) at 1:600. Chamberslides were washed three times with PBS and mounted in glycerol/PBS solution and propidium iodide or DAPI (both purchased from Vectashield).
For trafficking experiments, transfectants were treated with thapsigargin (1 µm), 4-PBA (5 mm) or 5% glycerol for a further 24 h. In the case of trafficking inhibitor studies, BFA (0.1 µg/ml) was added for 1 h prior to the addition of either glycerol or 4-PBA. To pinpoint only cell membrane localization of tagged plasmids, cells were stained prior to fixation. Briefly, slides were washed in PBS, blocked for 10 mins (in PBS/10%FBS) and a primary mouse antibody to Myc (1:100, PBS/0.1%BSA) added for 45 min at room temperature (9E10, Santa Cruz). Following washing (0.1% BSA/PBS for 3 × 5 min), cells were fixed in ice-cold methanol (10 min) and subjected to staining with a secondary rabbit anti-mouse–FITC antibody (DAKO), 1:500 dilution, for 1 h at room temperature, with rocking and protection from light. Coverslips were washed three times with PBS and mounted in glycerol/PBS solution and propidium iodide or DAPI (Vectashield). Cells were viewed and photographed using an ultraviolet confocal microscope (TCS Leica) and images captured using ImagePro Plus 4.1 software. The data shown in immunofluorescence studies are representative of at least three independent experiments.
FACS analysis of Myc-tagged receptors
HeLa cells were seeded into six-well plates (Costar) at a density of 1 × 105 cells per well and transiently transfected with 5′-Myc-tagged wild-type and C118W BMPR-II plasmids, as described earlier. Transfectants were grown for 48 h in DMEM/10% FBS and treated with thapsigargin (1 µm), 4-PBA (5 mm) or 5% glycerol for a further 24 h. Cells were then washed with PBS and 1.5 ml of EDTA (0.5 mm, PBS pH 7.4) added and placed in an incubator for 10 min at 37°C. Following cell liberation, ice-cold PBS was added and the cell suspension placed into FACS tubes and centrifuged at 600g for 10 min. To determine the percentage transfection efficiency of wild-type or C118W BMPR-II plasmids, an experimental condition was introduced whereby cells were permeabilized by adding FACS Perm solution (BD Biosciences) for 10 min at room temperature; cells were then washed in PBS prior to staining. All other conditions were not permeabilized in order to detect only cell membrane expression of transfected plasmid. Mouse monoclonal anti-Myc (Santa Cruz) antibody was added at a 1:100 dilution for 1 h on ice. Cells were washed twice with PBS and a goat anti-mouse secondary FITC-conjugated antibody (DAKO) added and the tubes incubated on ice, in the dark for 1 h. Cells were then washed twice with ice-cold PBS and resuspended in PBS before analysis on a FACS (BD Biosciences). FL1:FL2 ratios allowed definition of R1, which was, defined as a cell-membrane-positive area, through the use of a mouse IgG control (negative staining) versus an anti-Myc antibody. Cell Quest software was used for the analysis and gating of cell populations. All cell-membrane Myc-positive cells were expressed as a percentage of the transfected cell population. FACS analysis experiments were repeated six times.
HeLa cells were plated onto 10 cm2 petri dishes, grown to confluence and then transfected with pcDNA3.0, wild-type or mutant constructs, as described previously. Following transfection, cells were placed in fresh DMEM/10%FBS for 48 h and treated with 4-PBA (5 mm) or 5–10% glycerol for a further 24 h. Cells were then placed in serum-free medium overnight and then stimulated with BMP-4 or -6 (R&D Systems) at 100 ng/ml for 1 h. Protein was harvested by washing cells in cold PBS and snap-freezing in an ethanol/dry-ice bath. Cells were scraped into 200 µl of lysis buffer (125 mm Tris–HCl, 2% SDS, 10% glycerol) containing protease inhibitors, antipain, leupeptin, pepstatin (all at 1 µg/ml), PMSF (1 mm), sodium vanadate (2 mm) and sodium fluoride (2 mm). Nitrocellulose membranes were incubated with various antibodies: a rabbit monoclonal anti-phospho-Smad 1/5 (1:1000, Cell Signalling Technology, Inc., MA, USA), a mouse monoclonal anti-Myc 9E10 (1:1000, Santa Cruz), a mouse monoclonal anti-BMPR-II (1:500, BD Biosciences) in 5% BSA/0.1%Tween/TBS overnight at 4°C. Blots were then washed and incubated with an anti-rabbit, anti-mouse (1:2000, DAKO), anti-goat horseradish peroxidase-conjugated secondary antibody, respectively (in 5% milk powder/0.1%Tween/TBS), for 1 h at room temperature. Bands were visualized by chemiluminescence (Amersham, UK). Blots were then stripped and re-probed using an antibody to total pSmad 1/5 (Cell Signalling Technology, Inc.), or probed for β-actin using a mouse monoclonal antibody to β-actin (Sigma), followed by an anti-rabbit or anti-mouse HRP (DAKO) to assess the loading of samples. All immunoblotting experiments were repeated at least three times.
This study was supported by a grant from the British Heart Foundation (Programme Grant RG/03/005 to N.W.M. and R.C.T.) and from the EU Framework 6 award for PULMOTENSION.
Conflict of Interest statement. None declared.