|Home | About | Journals | Submit | Contact Us | Français|
Contractile airway smooth muscle (ASM) cells retain the ability for phenotype plasticity in response to multiple stimuli, which equips them with capacity to direct modeling and remodeling during development, and in disease states such as asthma. We have shown that endogenously expressed laminin is required for maturation of human ASM cells to a contractile phenotype, as occurs during ASM thickening in asthma. In this study, we profiled the expression of laminin-binding integrins α3β1, α6β1, and α7β1, and tested whether they are required for laminin-induced myocyte maturation. Immunoblotting revealed that myocyte maturation induced by prolonged serum withdrawal, which was marked by the accumulation of contractile phenotype marker protein desmin, was also associated with the accumulation of α3A, α6A, and α7B. Flow cytometry revealed that α7B expression was a distinct feature of individual myocytes that acquired a contractile phenotype. siRNA knockdown of α7, but not α3 or α6, suppressed myocyte maturation. Thus, α7B is a novel marker of the contractile phenotype, and α7 expression is essential for human ASM cell maturation, which is a laminin-dependent process. These observations provide new insight into mechanisms that likely underpin normal development and remodeling associated with airways disease.
This study reveals new biological mechanisms relevant to fibroproliferative disorders such as asthma that are characterized by the accumulation of smooth muscle, and in which smooth muscle phenotype switching occurs.
During development and the pathogenesis of fibroproliferative disease, smooth muscle cells exhibit a dynamic phenotype in response to changes in biological cues such as growth factors and the extracellular matrix (ECM). This enables myocytes to act as determinants of acute and chronic pathologic features in diseases, such as airway remodeling in asthma. From studies with in vitro cell culture systems and in vivo animal models, it is well established that plastic phenotypic behavior of differentiated smooth muscle cells is marked by reversible modulation and maturation between contractile and proliferative/synthetic phenotypic states (1). Numerous ultrastructural, biochemical, and functional differences between phenotypic states, as well as numerous gene transcriptional and protein translation mechanisms that regulate phenotype expression, have been identified (2, 3). Among the external factors that can affect phenotype expression, the ECM plays a prominent role (4).
Laminin-2 is required for commitment of mesenchymal cells to the airway smooth muscle (ASM) lineage during lung development (5). In vitro studies with myocytes obtained from adult tissues show that although ECM proteins such as fibronectin and collagen I promote a proliferative phenotype (6), the laminin family of proteoglycans can suppress modulation of ASM cells from a contractile to proliferative phenotype (4). Moreover, we recently reported that maturation of human ASM from the proliferative to the contractile phenotype is associated with increased endogenous expression of the α, β, and γ laminin chains that constitute laminin-2 (7). Notably, using competing peptides for the integrin-binding YIGSR domain in these laminin chains, we further demonstrated that ASM binding to laminin-2 is essential for maturation of contractile phenotype myocytes enriched in protein markers such as desmin and calponin (7). This is of significance to understanding the pathogenesis of bronchial asthma, which is characterized by the concomitant deposition of ECM, including the laminin α2 chain (8), and a marked increase in contractile smooth muscle abundance in association with ASM hypertrophy. These observations strongly suggest the existence of a self-regulated biological mechanism, mediated through laminin–ASM interactions, that underpins key components of airway remodeling in asthma.
Although ECM constituents such as laminin are principal biological cues regulating phenotype plasticity of smooth muscle cells, relatively little is known about the repertoire of cell surface receptors needed to mediate their effects. The integrins are a large family of transmembrane proteins that exist as noncovalent heterodimers of α- and β-subunit splice variants that form receptors with different selectivity for individual ECM constituents (9). A specific group of laminin-binding integrins, including α3β1, α6β1, and α7β1, has been identified (9). Glukhova and colleagues reported that vascular and colon smooth muscle cells exhibit concomitant changes in the spatial-temporal expression of laminin isoforms and laminin-binding integrins during development and maturation to adulthood (10). However, no studies have directly investigated the specific role of laminin-binding integrins in the maturation of differentiated smooth muscle cells to a contractile phenotype.
In the present study we characterized the repertoire of laminin-binding integrins expressed by adult human ASM cells, and tested the hypothesis that these receptors are required for maturation of myocytes to a contractile phenotype mediated by endogenously expressed laminin. With human ASM cell lines we used immunoblotting and real-time PCR to compare the expression of α3, α6, and α7 integrins and their splice variants in proliferating cultures, and cultures subjected to prolonged serum deprivation, which induces a subpopulation of human ASM cells to acquire the contractile phenotype (7, 11). Moreover, using flow cytometry and fluorescence microscopy we examined the unique repertoire of cell surface integrins expressed by human ASM cells of divergent phenotype. To test the requirement of specific integrins in the acquisition of a contractile phenotype, we employed selective siRNAs to silence expression of individual integrins and assessed the effect on expression of stringent phenotype markers. Collectively, these studies demonstrate for the first time that the laminin-binding α7 integrin subunit, most likely α7B, is exclusively needed to promote contractile phenotype acquisition in differentiated human ASM cells.
All chemicals used were of analytical grade or higher. All compounds were purchased from Sigma (St. Louis, MO) unless stated otherwise. Rabbit anti-desmin antibody (H-76, sc14026) was purchased from Santa Cruz Biotechnology Inc. (Santa Cruz, CA). Antibodies directed against integrins α3A (clone 29A3), α3B (clone 54B3), α5, α6A (clone 1A10), α6B (clone 6B4), and β1 (clone Lia1/2) were purchased from Chemicon (Temecula, CA). Antibodies directed against integrins α7A and α7B were gifts from Dr. G. Tarone, Università di Torino, Italy (12).
For all studies senescence-resistant human ASM cell lines were generated using stable MMLV retroviral transduction of the human telomerase reverse transcriptase gene (hTERT) into passage 1 or 2 primary cultures as we have described (7, 13). Primary cultured human ASM cells were prepared from macroscopically healthy segments of second- to fourth-generation main bronchus obtained after lung resection surgery from patients with a diagnosis of adenocarcinoma (7, 13). All procedures were approved by the Human Research Ethics Board (University of Manitoba). hTERT-expressing human ASM cells retain the ability to express markers of the contractile phenotype including smooth muscle myosin heavy chain (smMHC), calponin, sm-α-actin, and desmin to passage 10 and higher (13). At least four different hTERT-human ASM cell lines between passages 16 and 22 were used in all studies.
Unless otherwise indicated, human ASM cells were seeded at a density of 1 × 104 cells/cm2, and grown to confluence in Dulbecco's Modified Eagle's Medium (DMEM) containing 10% vol/vol fetal bovine serum (FBS) (14). To induce a contractile phenotype in hTERT-human ASM, confluent cultures were switched to DMEM supplemented with ITS (insulin 5 μg/ml; transferrin 5 μg/ml, selenium 5 ng/ml) for up to 7 days as we have described (3, 7, 13, 14). Serum-free media was replaced every 3 days. With this protocol, increased expression of desmin and sm-α-actin was consistently induced, indicative of phenotype maturation in both hTERT-human ASM and primary cultured human ASM cells.
For analysis of myocyte phenotype, protein lysates were prepared as we have described (7) in ice-cold lysis buffer (100 mM NaCl; 10 mM Tris-HCl, pH 7.5; 2 mM EDTA; 0.5% wt/vol deoxycholate; 1% vol/vol triton X-100; 1 mM phenylmethylsulphonylfluoride; 10 mM MgCl2; 5 μg/ml aprotinin; 100 μM sodium orthovanadate). Soluble protein content was determined using the Bio-Rad protein assay (BioRad, Hercules, CA). For measuring integrin protein abundance, samples were size fractionated by SDS-PAGE using loading buffer lacking β-mercaptoethanol (β-MC). For all other proteins β-MC was included in the gel loading buffer. Western blotting was performed as we have described using nitrocellulose membranes (7). Optimal dilutions for each antibody were pre-determined: anti-desmin (1:500), anti–sm-α-actin (1:1,000), anti-α3A (1:1,000), anti-α5 (1:5,000), anti-α6A (1:500), anti-α7B (1:1,000), and anti-β1 (1:500). After protein transfer membranes were developed using secondary horseradish peroxidase–conjugated antibody, and visualized by enhanced chemiluminescence reagents (Amersham, Oakville, ON, Canada). Membranes were re-probed with antibodies for β-actin (Sigma) to normalize for equal sample loading. Densitometry was performed using the Epson Perfection 4180 Station and TotalLab TL100 software (Nonlinear Dynamics, Durham, NC).
Total RNA was extracted using the Qiagen RNeasy Mini Kit (Qiagen, Mississauga, ON, Canada) according to the manufacturer's protocol. RNA (2 μg) was reversed transcribed using M-MLV reverse transcriptase (Promega, Madison, WI), incubated for 2 hours at 37°C followed by 5 minutes of incubation at 95°C, and diluted 1:10 with RNase-free water. Real-Time PCR was performed using the resulting cDNA with primer pairs listed in Table 1. Reaction parameters were: 40 cycles consisting of 92°C for 45 seconds (denaturation), 60°C for 45 seconds (annealing), and 72°C for 90 seconds (extension). Assays were performed in duplicate in 20-μl reactions, and the cycle threshold (CT = amplification cycle number) values for each reaction were determined using the Roche Molecular Biochemicals LightCyler 3 (version 3.5; Indianapolis, IN). Data were analyzed using the comparative CT method as previously described (7) using the Applied Biosystems (Streetsville, ON, Canada) guidelines (15). The fold change in gene expression normalized to an endogenous reference gene (18 s rRNA) and relative to the untreated control (Day 0) is given by the equation 2−ΔΔCT. The ΔCT value is determined by subtracting the average 18 s rRNA CT value from the average CT value of the corresponding target gene of interest. The calculation of ΔΔCT values involves subtraction by the ΔCT calibrator value (Day 0). For the untreated sample (Day 0) ΔΔCT = 0 and 2° equals 1. For the treated samples, evaluation of 2−ΔΔCT indicates the fold change in gene expression relative to Day 0.
Our previous studies show human ASM is a heterogeneous population in which ~ 1/6th of cells are induced to a fully contractile phenotype with prolonged serum deprivation (3, 14). Therefore, we used flow cytometry to directly assess co-expression of cell surface integrin subunits and the intracellular intermediate filament protein desmin, a stringent contractile phenotype marker, in individual myocytes from cultures maintained for 7 days in serum-free DMEM/ITS media. Cells were detached from dishes using 0.025% vol/vol trypsin/10 mM EDTA, and rapidly transferred to a large volume of PBS containing 0.2%vol/vol FBS to quench trypsin activity (16). Cell suspensions were centrifuged (800 × g, 5 min), resuspended in 10 ml PBS, filtered through 70-μm nylon mesh, and cell number was determined using a hemocytometer. Aliquots containing 106 intact cells were incubated (30 min, room temperature) with surface-staining integrin antibodies (1 μg/100 μl) that had been pre-conjugated with Alexa-Fluor (AF)-488 using Zenon Labeling Kit (3.5:1 molar ratio of Zenon labeling reagent to antibody) according to manufacturer's instructions (Molecular Probes, Invitrogen, Eugene, OR). After incubation with Zenon AF-488-anti-integrin IgG, human ASM cells were fixed in 3% paraformaldehyde/PBS (5 min, room temperature), washed three times with PBS, and then permeablized in PBS/0.1% Triton X-100 (5 min, room temperature). Cells were then centrifuged (800 × g, 5 min), washed once with 1 ml PBS, and then resuspended and incubated (30 min, room temperature) in rabbit anti-desmin IgG1 (1 μg/100 μl) which had been pre-conjugated with the R-Phycoerythrin (R-PE) using the Zenon Labeling Kit. Cells were then washed and resuspended in PBS in preparation for flow cytometry. For all experiments background and nonspecific labeling was determined using control cells incubated with isotype-matched nonimmune serum that had been conjugated with either AF-488 or R-PE using protocols that matched those used for primary antibodies.
Flow cytometry was performed using a Beckman Coulter EPICS ALTRA Flow Cytometer (Beckman Coulter Canada Inc., Mississauga, ON, Canada). All analyses used 488-nm (150 mW) laser excitation. Forward light and side scatter histograms were used to identify and gate for intact cells. AF-488 and R-PE emission was collected simultaneously using 525-nm and 575-nm bandpass filters, respectively. For each histogram 10,000-gated events were collected. Labeling for integrins or desmin was considered “positive” when fluorescence intensity was greater than that observed for all but the highest 5% of cells stained with isotype-matched antibodies. EXPO32 MultiCOMP MFA software (Version 1.2B; Beckman Coulter Canada) was used to assess each myocyte for the percent overlap in coincident positive and negative staining for cell surface integrins and desmin.
hTERT-human ASM cells were grown to confluence on 25 mm2 microscope glass slides, subjected to 7-day serum-free culture in DMEM/ITS, then were fixed, permeablized, and immunolabeled as we have previously described (3, 14). Human bronchial specimens obtained from lung resection were quick-frozen in OCT compound and then ~ 7-μm cross-sections were cut. Sections were mounted on microscope slides, then fixed and permeablized and labeled with antibodies as described for cultured cells. Primary antibodies and their respective dilutions included rabbit anti-desmin IgG (1:100) alone, and in combination with either mouse anti-smMHC IgG1 (clone hSMv; 1:100; Sigma ImmunoResearch) or rabbit-anti-integrin α7B IgG1 (1:50) (12). Fluorescein isothiocyanate–and Texas Red–conjugated secondary antibodies (1:100; Jackson ImmunoResearch, West Grove, PA) were used to detect primary antibody bound to labeled cells. Nuclei were then labeled with Hoechst 3342 (10 μg/ml), and coverslips were mounted using ProLong Antifade (Molecular Probes, Invitrogen). Thereafter fluorescence micrographs were obtained as we have described (3) using an Olympus LX70 microscope equipped with charge coupled camera controlled by UltraView Software (Olympus, Hicksville, NY).
The small interfering RNA (siRNA) Generation Kit (Gene Therapy Systems, San Diego, CA) was used to prepare siRNA from human ASM cDNA with primers that amplified integrin α3 cDNA (558 base pairs [bp]), α6 cDNA (523 bp) and α7 cDNA (547 bp, Table 2). PCR primers were designed to amplify unique regions of each specific integrin that did not share sequence homology with any other integrins. Primer sequences listed in Table 2 also included the 52-bp T7 promoter sequence linkers (5′-GCGTAATACGACTCACTATAGGGAGA-target DNA-3), which were incorporated into the DNA template PCR product to allow for in vitro transcription with the TurboScript T7 Transcription Kit (Gene Therapy Systems). Double-stranded RNA (dsRNA) was generated using the TurboScript T7 RNA Transcription Kit and then diced into 21-bp fragments using recombinant human dicer enzyme following the manufacturer's instructions (Gene Therapy Systems).
In preliminary experiments, from a concentration–response curve (0.05–1 μg/ml) we determined that 1 μg/ml of siRNA provided optimal silencing of mRNA and protein for all integrin targets (α3, α6, α7) in hTERT-human ASM cells. For subsequent studies the day before siRNA transfection, confluent human ASM cells were re-seeded into 6-well plates in DMEM/10% vol/vol FBS so that 50 to 60% confluence was achieved 16 hours later. Transfection of siRNA (1 μg/ml) was performed with Genesilencer reagent (Gene Therapy Systems) according to manufacturer's instructions, and cells were maintained in serum-free DMEM/ITS thereafter. A second transfection was performed 3 days later at the same time that culture media was refreshed. This double transfection strategy was developed based on preliminary experiments, which showed it was most effective at silencing integrin mRNA and protein expression for at least 6 days. Transfection efficiency at each stage was greater than 95%, as we have previously reported (13). For negative control studies the transfection protocol was performed in the absence of siRNA.
Data are expressed as mean ± SEM. All experiments were completed in duplicate using hTERT-human ASM generated from at least four different primary bronchial smooth muscle cell cultures. Protein expression data is expressed as fold change compared with initial levels (Day 0 in most experiments); all values are corrected for equal loading based on β-actin abundance. Comparisons were made using one-way ANOVA, with repeated measures, followed by Bonferroni's post hoc t test. A probability value of P < 0.05 was considered significant.
We have shown that maturation of hTERT-human ASM cells to a contractile phenotype during prolonged serum-free culture is marked by laminin-2 dependent accumulation of contractile phenotype marker proteins such as desmin, calponin, and smooth muscle sm α-actin (7, 13). Indeed, concomitant with myocyte maturation the cells exhibit increased expression of the α2, β1, and γ1 chains of laminin-2. Thus, we assessed whether human ASM cell maturation was also associated with changes in expression of laminin-binding integrin subunits α3, α6, and α7, and their respective A and B cytoplasmic tail splice variants.
Using immunoblotting we observed that the abundance of integrins α3A, α6A, and α7B were approximately doubled (Figure 1) in culture conditions that promoted the maturation of human ASM cells to a contractile phenotype (7). In contrast, integrin α3B (1.0 ± 0.02) and α7A (1.0 ± 0.02) abundance was unchanged during myocyte maturation. Moreover, neither the fibronectin selective integrin α5 subunit (17) nor integrin β1 were changed in abundance with serum deprivation (Figure 1). To further confirm these results we employed quantitative real-time PCR analyses, which revealed a 3- to 8-fold increase in mRNA for integrins α3A, α6A, and α7B over the course of myocyte maturation (Figure 1). These experiments reaffirmed that there was no change in expression of α3B, α7A, α5, and β1 subunits, and also demonstrated that expression of α6B, for which we had no suitable antibodies to perform immunoblotting, was similarly unchanged during myocyte maturation (not shown). Collectively these experiments reveal for the first time that increased expression of laminin-binding integrin subunits α3A, α6A, and α7B occurs in culture conditions that promote the maturation of human ASM cells to a contractile phenotype.
In additional experiments we performed PCR analyses to determine whether the profile of laminin-binding integrins expressed by contractile human ASM in prolonged serum-free culture mimicked that of myocytes from intact tissue. We designed primers that generated PCR products of differing size for each of the A and B splice variants of α3, α6, and α7 integrins (Table 3). Using mRNA isolated from human ASM cells dissected from second-generation mainstem human bronchi, we observed that α3B, α6A, and α7B isoforms were highly expressed, α3A and α6B transcript was present in low abundance, and α7A was undetectable (Figure 1).
Though our immunoblotting and PCR analyses confirm increased expression of laminin-binding integrins in conditions that promote maturation, our previous studies have revealed there is a disparate induction of the contractile phenotype marker proteins in distinct subpopulations of cultured airway myocytes under these conditions (3, 14). Thus, we used flow cytometry to assess cell surface abundance of integrins in individual myocytes after 7 days of serum-free culture. Figure 2 shows representative frequency distribution histograms of human ASM cells that were surface stained for laminin-binding integrin subunits using fluorochrome-conjugated primary antibodies. To confirm the threshold for positive labeling, for all experiments we prepared parallel negative control samples using fluorochrome-conjugated isotype-matched immunoglobulin. Distinct fractions of cells exhibiting positive fluorescence labeling for integrins α3A (46 ± 16%), α6A (42 ± 19%), α6B (27 ± 19%), and α7B (16 ± 6%) were observed (Figure 2). In contrast, though a small fraction of cells labeled for α3B or α7A was evident, these myocytes appeared as part of a shoulder from the larger population of negatively stained myocytes. As a positive control for cell surface integrin labeling, we also performed experiments to assess α5 and β1 integrin expression (Figure 2). Notably, compared with respective isotype controls, a distinct positively stained peak that included virtually all myocytes was evident for both α5 and β1. Collectively, in a pattern mimicking our previous investigation of contractile phenotype marker protein expression (3, 14), the current experiments show that disparate expression of cell surface laminin-binding α-integrin subunits is induced in subpopulations of human airway myocytes when cells are grown in conditions that promote contractile phenotype maturation.
We next used flow cytometry to investigate whether there was a direct association between induction of laminin-binding α-integrin subunits and the acquisition of a contractile phenotype in individual myocytes. For this purpose we developed double labeling protocols to assess the expression of membrane integrins and intracellular desmin, an intermediate filament protein that is a stringent marker of the contractile phenotype in smooth muscle cells (11). No direct correlation between the expression of integrin α3A and desmin was seen (Figure 3), despite our previous observation that both proteins accumulate significantly during contractile phenotype development (Figure 1). Indeed, only about half (43 ± 2%) of the myocytes that expressed α3A also expressed desmin. Similarly, there was no correlation between α6A and desmin, with co-expression in only half (43 ± 2%) of α6A-positive myocytes (Figure 3). Labeling for integrins α3B, α6B, or α7A, which were unchanged in total abundance during 7-day serum deprivation, was also not associated with an increased tendency for expression of desmin (Table 4). In contrast, labeling for integrin α7B exhibited a significant positive correlation with desmin expression in individual human ASM cells, as 66 ± 3% of α7B-positive myocytes co-expressed desmin (P < 0.05, ANOVA with repeated measures, and Bonferroni's post hoc t test, Figure 3). These data indicate that though integrins α3A, α6A, and α7B each accumulate in total abundance in human ASM cell culture during prolonged serum deprivation (Figure 1), only the α7B subunit appears to be selectively induced in the unique subpopulation of myocytes that acquires a contractile phenotype (Figure 3). This finding was consistent in each of the three independent experiments that we performed.
To further confirm the selective induction of integrin α7B in human ASM cells expressing a contractile phenotype, we performed immunofluorescence imaging of primary cultured human ASM cells and intact bronchial tissue (Figure 4). Primary human ASM cells subjected to 7-day serum deprivation were double labeled for α7B and either desmin or smooth muscle myosin heavy chain (smMHC), two stringent protein markers of the contractile phenotype that show strong coincident staining in contractile phenotype myocytes (Figure 4). Notably, as expected from our prior flow cytometry analyses, we observed abundant expression of α7B in a unique subpopulation of human ASM cells, and this appeared to be entirely coincident with desmin and smMHC labeling. Moreover, these cells exhibited a large elongate morphology that is characteristic of contractile phenotype smooth muscle cells, and expression of both smMHC and desmin was a unique feature of these cells (Figure 4). We also performed immunofluorescence imaging of intact mainstem human airway specimens, and consistently observed strong labeling for integrin α7B in the contractile smooth muscle layer surrounding the bronchi (Figure 4). Collectively these data confirm a strong correlation for abundant expression of α7B integrin in ASM cells of a contractile phenotype.
Based on our current flow cytometry and immunofluorescence studies, our previous experiments showing that endogenous laminin, likely laminin-2, is required for human ASM cell maturation (7), and that integrin α7 selectively binds laminin-2/4 (18), we next assessed the requirement for integrin α7 in human ASM cell maturation using siRNA technology. Human ASM cells were seeded at confluence and maintained in serum-free conditions for 6 days, during which time cells were transfected on both Day 0 and Day 3, with siRNA selective for integrin α7 (Table 2). We first assessed the effectiveness of siRNA-mediated silencing of integrin α7 expression during 6-day serum deprivation using PCR and immunoblotting for α7B (Figure 5). Both analyses revealed that the siRNA protocol we developed completely abrogated the accumulation of integrin α7B that typically occurs in human ASM cells, and was also evident in our control samples.
We next examined the effects of inhibiting α7 integrin accumulation with siRNA on human ASM cell maturation by measuring the accumulation of the contractile phenotype marker proteins desmin and sm-α-actin (7, 11, 14). Consistent with phenotype maturation of ASM cell myocytes that we have described previously (7, 13, 14), before serum deprivation human ASM cells expressed low levels of desmin and sm-α-actin (Figure 5), but these levels were approximately doubled after 6 days of growth in serum-deficient conditions. Notably, with inhibition of integrin α7 expression using siRNA, accumulation of both desmin and sm-α-actin was completely prevented (Figure 5), suggesting an abolition of phenotype maturation. Importantly, accumulation of desmin and sm-α-actin after 6 days in control transfection samples was the same as that seen in untreated cultures. For example, in untreated samples and control transfection samples, desmin increased 1.8 ± 0.2 and 1.9 ± 0.3 times, respectively (P > 0.05), compared with Day 0.
To confirm the validity of our findings and a possible unique functional role for α7 integrin in myocyte maturation, we next performed experiments to ensure that our siRNA protocol targeting α7 integrin was selective, and that there were no indirect effects on expression of other integrin subunits. Indeed, silencing of integrin α7 had no concomitant effects on α3A and α6A (Figures 6A and 6B), the α5 subunit (Figure 6C), or on integrin β1 (Figure 6D). These findings strongly implicate a unique requirement for α7 integrin in phenotype maturation of human ASM cells from the adult lung.
To ensure that the ability of silencing α7 integrin to prevent myocyte phenotype maturation was not due to cell toxicity, we assessed cell integrity by phase contrast imaging (Figures 6E–6H), and measured cell number after completing the siRNA protocol. There was no evidence of myocyte stress, nor were there any differences in the total number of adherent cells after the 6 days of siRNA treatment (1.0 ± 0.1 × 105 cells/well versus 1.0 ± 0.1 × 105 cells/well in control treated and α7 siRNA-treated respectively; P > 0.05).
Though our data using selective α7 integrin silencing reveal a requirement for this receptor in phenotype maturation, they do not preclude the possibility that other laminin-binding integrins may also be involved. Thus we next developed siRNA protocols for integrins α3 and α6 and measured the effects of silencing these subunits on myocyte phenotype expression in serum-free culture conditions. In a manner similar to our studies using α7 siRNA, confluent cultures of human ASM cells were maintained for 6 days in serum-free culture, during which time cells were transfected with selective siRNA for α3 or α6 integrin (Table 2) on Day 0 and Day 3. To confirm the effectiveness of α3A and α6A silencing, we measured the abundance of each protein after 6 days of serum deprivation in the presence of the respective siRNA we generated (Figure 7). Importantly, the increase in α3 or α6 integrin that we observed in control conditions (Figure 1) was abrogated by their corresponding siRNA (Figure 7).
Silencing of integrin α3 was selective, as α3 siRNA was without effect on the accumulation of integrin α6A and α7B during prolonged serum withdrawal (Figure 7, Table 5). Integrin α3 silencing did, however, promote protein accumulation of the fibronectin-binding receptor, integrin α5, during 6-day serum-free conditions, a response that was mimicked by siRNA for α6 integrin, and that was not seen in control conditions (Table 5). Nonetheless, in contrast to results we obtained with α7 siRNA (Figure 5), silencing of α3 failed to prevent myocyte phenotype maturation, as the accumulation of desmin and sm-α-actin was similar in control and α3 siRNA treated cultures (Figure 8). Collectively, these data indicate that our previous observation of a unique requirement for increased α7 integrin expression for human ASM cell phenotype maturation is not dependent on a concomitant increase in α3 integrin expression.
In the design of primers to generate integrin-selective siRNAs, we chose coding regions that had poor homology between integrins (Table 2). Despite these efforts, we found that treatment with siRNA we generated for α6 integrin blocked accumulation of both α6A and α3A integrin proteins, and thus had a double gene-silencing effect (Figure 7). Importantly, however, consistent with our results using siRNA for α3 integrin, integrin α6 siRNA treatment did not affect the accumulation of integrin α7B that occurs over 6 days of serum deprivation (Table 5). With respect to human ASM cell phenotype maturation, in marked contrast to the results we obtained with selective α7 silencing (Figure 5), treatment with α6 siRNA, and subsequent silencing of both α6 and α3, was without effect on the accumulation of desmin and sm-α-actin during 6 days of serum withdrawal (Figure 8). These data further strengthen our observations that suggest a unique requirement of laminin-binding α7 integrin in human ASM cell phenotype maturation, as concomitant suppression of α6 and α3 integrin expression failed to prevent acquisition of a contractile phenotype in vitro.
Phenotype plasticity of smooth muscle cells in differentiated tissues is a key determinant of organ remodeling associated with fibroproliferative disorders such as asthma. In this regard, primary focus has been given to modulation of myocytes to a proliferative/synthetic state. Thus, though it underpins increased smooth muscle mass that leads to altered vascular and lung function, less is known about the mechanisms promoting myocyte maturation. Intracellular markers of mature smooth muscle cells have been described; however, the current study is the first to show that increased expression of a select repertoire of integrins that bind laminin is both a hallmark of the contractile phenotype, and is required for the process of phenotype maturation. Our studies demonstrate that expression of the laminin-binding integrins, α3A, α6A, and α7B, is significantly increased in human airway myocytes under conditions that promote phenotype maturation. Notably, with prolonged serum-free culture, distinct subpopulations of human ASM cells are revealed; using dual immunofluorescence flow cytometry protocols, we demonstrate that among the laminin-binding integrins, only α7B integrin is preferentially expressed by the individual myocytes in the subgroup that acquires a contractile phenotype. Moreover, using gene-silencing strategies with integrin-selective siRNAs, we show that α7 integrin (likely α7B), but not α3 and α6, mediates laminin-dependent myocyte maturation. Our observations are significant because they reveal intrinsic ECM-associated mechanisms regulating phenotype plasticity and heterogeneity of smooth muscle cells that may be critical biological components of fibroproliferative disorders.
Laminin is essential for the commitment and differentiation of embryonic lung mesenchyme into ASM, and for its circumferential organization around bronchi (19). In primary culture of ASM cells from adult tissues, laminin can slow modulation from the contractile phenotype to a proliferative state (4). Moreover, recent studies using bovine tracheal smooth muscle strips show that laminin, by itself, does not affect contractility or proliferation but reduced the effect of the mitogen, PDGF, on these parameters (20). This would support the notion that laminin regulates both phenotype and function of intact ASM. We recently showed that endogenously expressed laminin, likely laminin-2, is also needed for differentiated ASM cells to re-acquire a contractile phenotype after modulation to a proliferative state in vitro (7). This is consistent with reports showing laminin-1 and -2 to be essential for accumulation of differentiated smooth muscle during development of the airways and gastrointestinal tract (19). Though the ECM is recognized as a regulator of smooth muscle phenotype, before our study, the functional role of ECM receptor expression in mediating phenotype maturation in differentiated myocytes has not been elucidated. Notably, changes in integrin expression is associated with atherosclerosis and post-angioplasty restenosis (21), and some studies have assessed ECM receptor expression in association with the proliferative, apoptotic, and pro-inflammatory responses of phenotypically modulated myocytes (6, 16, 22). Though there appears to be a requirement for β1 integrins in smooth muscle growth, a precise determination of the ECM selectivity of the receptor heterodimers involved has not been reported. Thus, the current study is unique both in its focus on the role of integrins in mediating ASM cell maturation to a contractile phenotype, and on receptors with an established binding selectivity for a single ECM proteoglycan, laminin (9).
The selectivity of α3β1, α6β1, and α7β1 integrin heterodimers for laminin is a determinant of α-subunit identity (9). Co-ordinated expression of laminin isoforms and laminin-binding integrin subunits occurs in smooth muscle of the developing vasculature and gastrointestinal tract (10), and is altered in response to tissue injury (21). However, the precise functional role of laminin-binding integrins in phenotype determination of differentiated smooth muscle cells is not known. Our experiments reveal that maturation of ASM cells is associated with increased total abundance of α3A-, α6A-, and α7B-subunits in the absence of any changes in expression of β1-subunit, or α5 integrin, which has affinity for fibronectin. To ensure that this profile of integrin subunits is similar to that of intact tissue, we used immunoblotting, PCR, and immunohistochemistry to assess integrin expression in human bronchial smooth muscle. Indeed, α3A, α6A, α7B, and β1 were highly expressed. In tissue and cultured myocytes we also detected α3B and α6B, but α7A was only present in cultured cells; the abundance of each was unchanged during contractile phenotype acquisition. Interestingly, although β-subunits induce downstream signaling events that regulate cell responses such as cycle progression and actin cytoskeleton assembly (23), α-subunits and their binding to ECM ligands modulates the cascade of pathways invoked by the β-subunit, and thus the α integrins are critical in directing specific cell responses (23, 24). In this context, the selective increase in expression of total α3A, α6A, and α7B that occurs during contractile phenotype acquisition is consistent with a possible functional role for these α-integrin subunits in myocyte maturation.
Heterogeneity of the cytoplasmic domains of α-subunits can result from post-translational modifications and through alternative splicing, which produces so-called A and B isoforms (9). Though the precise role of cytoplasmic domain splicing is not clear, it may underpin differences in intracellular signaling and confer unique functional roles to integrins. Indeed, as ASM responses to different laminin isoforms vary during myocyte differentiation and maturation (7, 25), changes in expression of integrin splice variants may be an associated mechanism. Our data reveal that total and cell surface integrin α7B, but not integrin α7A, was approximately doubled in abundance in ASM cell cultures induced to a contractile phenotype. This is consistent with reports indicating that integrin isoform switching is an important mechanism in skeletal muscle cell differentiation, as cytoplasmic splice variants of integrin α7 are differentially expressed during development (26). The presence of alternative splice variants of laminin-binding integrin α3 and α6 is also well known (9), and our data indicate the A isoform of each is selectively induced during myocyte maturation. In contrast with the B isoforms, the A cytoplasmic variants of α3 and α6 appear to be phosphorylated on serine residues, only weakly on tyrosine residues, and are a major target for PKC after phorbol ester exposure (27). No such differences in the phosphorylation states of integrin α7 isoforms have been documented to date; however, alternative splicing of the cytoplasmic domains for α6 and α7 is similar (26), suggesting that parallel differences in signaling profiles between α7A and α7B may exist.
Our experiments do not address functional differences between desmin- and integrin α7–expressing human ASM subpopulations. However, in previous experiments using canine tracheal smooth muscle cells, we sorted myocytes on the basis of contactile protein content, and in subsequent subculture we observed a significantly suppressed proliferative response in cells that retained elevated abundance of smMHC and sm-α-actin (1). This suggests that the α7B- and desmin-enriched human myocytes we identified may also exhibit reduced proliferative response.
A unique aspect of this study is our use of human ASM cultures subjected to prolonged serum-free culture conditions to induce a contractile phenotype in a subset of cells, as we have reported previously (3, 14). Phenotypically heterogeneous subpopulations of smooth muscle cells in culture and intact tissues have been distinguished by differences in morphology, molecular signature, and functional properties (28–34). Consistent with published reports (3, 14), our immunofluorescence analyses using antibodies for smMHC and desmin, stringent molecular markers of contractile phenotype smooth muscle cells (11, 35), confirmed that maturation occurred in only a select subpopulation of myocytes. Thus, immunoblotting and PCR assays, which rely on sampling of all cells, do not necessarily reflect the direct association between integrin expression and myocyte maturation. To assess the association of integrin expression in those myocytes that undergo maturation, we developed novel flow cytometry–based protocols in which we prepared fluorochrome-conjugated primary antibodies for simultaneous labeling of cell surface integrin subunits and intracellular desmin. By this approach, subgroups of ASM cells comprising 20 to 40% of all myocytes were revealed with labeling for desmin or the individual laminin-binding α-integrins that increased in total abundance during phenotype maturation. Notably, myocytes that expressed α7B were distinct in showing significant positive co-expression with desmin, as nearly 70% of these cells were of the contractile phenotype. Due to the sensitivity of the flow cytometry assay we developed, the exclusion cutoff used for positive integrin labeling was at the 95th percentile of control cells, thus our gating criteria were considerably more rigorous than standard single color flow cytometry where mean fluorescence intensity is compared between groups. This could explain why we did not see 100% correlation between α7B and desmin co-expression, and may have reduced the apparent co-expression of α3A or α6A with desmin in individual cells, which was less than 50%. Nonetheless, our flow cytometry experiments are significant, as they provide first time evidence for a differential association of a single laminin-binding α-integrin spice variant, α7B, with contractile phenotype acquisition of human ASM cells. Moreover, these data suggest that the pattern of laminin-binding α-integrin expression is closely related to phenotype heterogeneity of human ASM cells.
Our data from PCR, immunoblotting, immunofluorescence and flow cytometry support the existence of an intimate relationship between laminin-binding integrins and ASM cell maturation; however, they do not conclusively demonstrate a requirement for these receptors. Therefore, we developed dual-transfection siRNA protocols, which effectively abrogated the increased expression of individual target integrin proteins that would typically occur in parallel with accumulation of contractile phenotype marker proteins during prolonged serum-free culture. By this approach we demonstrated that though total α3A, α6A, and α7B integrin was increased during myocyte maturation, only the accumulation of α7B is required for acquisition of a contractile phenotype (Figure 5). This is consistent with a previous study by Yao and colleagues (36) showing that α7 expression correlates with the differentiated smooth muscle phenotype in vascular, gastrointestinal, and genitourinary systems. Moreover, as our data pertain to differentiated myocytes, they also extend recent observations from studies with an embryonic lethal α7 knock-out mouse, in which α7β1 is important for recruitment and survival of vascular smooth muscle cells during development (37). It appears that the principal effect of our α7 siRNA protocol is to prevent α7B accumulation, as our protocols had only a marginal effect in reducing α7A (not shown), which is not induced with serum withdrawal. Our siRNA strategy for α7 was also highly selective, as no change in accumulation of α3 or α6, or in the steady state expression of β1 or α5 integrin was observed. Notably laminin-2 and -4 are exclusive ligands for integrin α7 (18), and in a recent study we showed that endogenously synthesized laminin-2 is required for human ASM maturation (7). Collectively our findings indicate that ASM maturation is principally manifested through the binding of endogenously expressed laminin, likely laminin-2, to integrin α7B, which becomes markedly increased during contractile phenotype acquisition.
The intracellular pathways necessary for α7B-mediated myocyte maturation cannot be elucidated from our studies, but a number of integrin-linked pathways are also determinants of smooth muscle phenotype. A pathway including phophatidylinositol-3 kinase, mammalian target of rapamycin, p70S6 kinase, and 4E-BP1 is required for ASM cell hypertrophy and contractile protein accumulation in serum-free culture and in response to transforming growth factor β1 (3, 38), insulin, or Rho-Kinase (39). Phophatidylinositol-3 kinase signaling is implicated in integrin-mediated pathways that modulate a number of cellular processes, including differentiation and hypertrophy (40). Expression of smooth muscle specific genes, which is required for myocyte maturation, is induced via RhoA-Rho kinase signaling, and is modulated by PKC (2, 41, 42). RhoA and PKC have close ties with integrins (43), and thus they are also likely candidates for mediating α7B effects on myocyte maturation. Using HEK293 cell lines that express integrin α7B, Mielenz and colleagues (44) reported that laminin binding selectively induces phosphorylation of p125FAK, paxillin and p130CAS; these signaling effectors are associated with regulation of cytoskeletal dynamics, which is a determinant of ASM cell contraction, migration, and the transcription of smooth muscle–specific genes such as smMHC and sm-α-actin (2, 11, 42, 45). Clearly, our present observations provide a platform for future mechanistic studies to dissect the precise signaling cascades associated with α7B control of smooth muscle phenotype.
We were successful in using siRNA to prevent accumulation of α3- and α6-integrin during prolonged serum deprivation; however, these interventions were without effect on ASM cell phenotype maturation. This was the case even with siRNA for α6 integrin, which serendipitously blocked induction of both α3A and α6A in serum-free culture. These data confirm that there is no requirement for increased expression of α3 and α6 in ASM contractile phenotype acquisition; however, as siRNA did not completely deplete α3A or α6A protein, it is not certain if the small amount of protein remaining may be important in supporting the effects of α7B. This possibility seems unlikely, however, in light of our flow cytometry results that indicated fewer than half of myocytes that expressed α3A or α6A also expressed the contractile phenotype marker desmin. Interestingly, we did observe an increase in integrin α5 in myocytes transfected with α3 or α6 siRNA. Consistent with this finding, α3-null keratinocytes display increased adhesion to fibronectin (46), which selectively binds integrin α5β1 (17). However, as our flow cytometry studies showed that α5 and β1 integrins are expressed by all cells, future studies are required to examine whether changes in α5 are linked to those cells that undergo phenotype maturation.
In summary, we show that laminin-2 (7) and laminin-binding integrin expression is concomitantly induced in ASM cells under conditions that promote acquisition of a contractile phenotype. Moreover, we show that α7B integrin is unique, as it is required for myocyte maturation to occur. These data reveal new biological mechanisms relevant to fibroproliferative disorders such asthma and atherosclerosis that are characterized by the accumulation of smooth muscle, and in which smooth muscle phenotype switching occurs (47, 48). At present, though differences in ECM composition are reported in the airways of patients with asthma, it is not known if changes in integrin expression, in particular on ASM cells, are directly associated with disease progression (49). Notably, in patients with congenital muscular dystrophy due to laminin α2 chain deficiency, and in dy/dy mice that do not express laminin α2 chain, α7 integrin expression becomes severely diminished, indicating that changes in integrin abundance can be directly affected by changes in the ECM (50). Furthermore, as changes in α7 integrin expression are associated with the pathogenesis of vascular remodeling in response to injury (21), future studies are clearly necessary to assess the functional role of α7 integrin expression in other disease states, such as asthma. Though α7 integrin knockout mice have been generated (37), the mutation is embryonic lethal, thus more sophisticated transgenic mouse models that target this gene are necessary to investigate its role in disease models.
The authors thank Ms. Amanda Reinisch and Ms. Shan Li for technical assistance. They also thank Dr. Schaafsma for critical reading of this manuscript.
This work was supported by grants from the SickKids Foundation / Institute of Human Development, Child and Youth Health (#XG05–011), Canadian Institutes of Health Research (CIHR), and the Manitoba Institute of Child Health. This work was undertaken, in part, thanks to funding from the Canada Research Chairs Program and the Canada Foundation for Innovation. A.J.H. holds a Canada Research Chair in Airway Cell and Molecular Biology. T.T. is the recipient of fellowships from GlaxoSmithKline/Canadian Lung Association/CIHR and the CIHR National Training Program in Allergy and Asthma. W.T.G. was supported by a grant from the US National Institutes of Health, HL077726.
Originally Published in Press as DOI: 10.1165/rcmb.2007-0165OC on July 19, 2007
Conflict of Interest Statement: A.J.H. received an open research grant of $65,000 from Merk Frost Canada Inc. None of the other authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript.