Published ahead of print on February 19, 2004, doi:10.1165/rcmb.2003-0240OC
© 2004 American Thoracic Society DOI: 10.1165/rcmb.2003-0240OC
Mechanical Strain Inhibits Airway Smooth Muscle Gene Transcription via Protein Kinase C SignalingDepartment of Physiology, and Asthma/COPD Research Centre, Department of Internal Medicine, Section of Respiratory Diseases, and Department of Oral Biology, University of Manitoba, Winnipeg; and Biology of Breathing Research Group, Manitoba Institute of Child Health, Winnipeg, Manitoba, Canada Address correspondence to: Andrew J. Halayko, Section of Respiratory Diseases, University of Manitoba, Rm RS321 Respiratory Hospital, 810 Sherbrook Street, Winnipeg, MB, R3A 1R8 Canada. E-mail: ahalayk{at}cc.umanitoba.ca
Mechanical strain affects airway myocyte phenotype, cytoskeletal architecture, proliferation, and contractile function. We hypothesized that (i) short-term mechanical strain modulates transcription of smooth muscle–specific gene promoters for SM22 and smooth muscle myosin heavy chain (smMHC); and (ii) strain-induced change is mediated by altered actin polymerization in association with activation of protein kinase C (PKC). Primary cultured canine tracheal myocytes were transiently transfected with luciferase reporter plasmids harboring a murine SM22, human smMHC, or artificial serum response factor (SRF)-specific gene promoter and then subjected to cyclic strain for 48 h. This strain protocol significantly reduced transcriptional activity of SM22 and smMHC promoters and an artificial SRF-dependent promoter by 55 ± 5.9%, 57 ± 6.4%, and 75 ± 7.9%, respectively, with concomitant reduction in F/G actin ratio by 31 ± 8%. PKC inhibitors, GF109203X or Gö6976, significantly attenuated these affects. Similar to strain, strain-independent activation of PKC inhibited SM22, smMHC, and SRF-dependent promoter activity by 61 ± 4%, 66 ± 5%, and 28 ± 15%, respectively, and reduced the F/G actin ratio by 30 ± 5%. Gel shift assay revealed that PKC activation led to decreased binding of the required transcription factor, SRF, to CArG elements in the SM22 promoter. These data suggest a previously unknown role for PKC isoforms in mechanosensitive signaling in airway myocytes that is associated with coordinated regulation of actin cytoskeletal dynamics and smooth muscle–specific gene transcription.
Abbreviations: airway smooth muscle, ASM • Dulbecco's modified Eagle's medium, DMEM • deoxyribonuclease, Dnase • electrophoretic mobility shift assay, EMSA • filamentous actin, F-actin • fetal bovine serum, FBS • globular actin, G-actin • nonessential amino acids, NEAA • o-nitrophenyl-ß-D-galactoside, ONPG • phosphate-buffered saline, PBS • protein kinase C, PKC • phorbol-12-myristate-13-acetate, PMA • smooth muscle myosin heavy chain, smMHC • serum response factor, SRF • Tris Borate EDTA, TBE
Fetal breathing movements and spontaneous bronchial contraction during lung development are essential for coordinated differentiation of mesenchymal cells and structural modeling of the lung and airways (1, 2). Further, in vitro studies have revealed that prolonged unidirectional cyclic mechanical strain applied to primary cultured airway smooth muscle (ASM) cells induces cell proliferation, marked reorientation of myocytes perpendicular to the direction of stretch, changes in the abundance of cytoskeleton-associated proteins, and increased agonist-induced contractile response (3–5). Brief mechanical perturbation of intact ASM also leads to strain-induced disruption of thick and thin filaments composed of myosin and actin, respectively (6–8). External uniaxial mechanical strain applied to cultured ASM cells modulates the direction, number, and size of stress fibers (4, 9). These observations suggest that mechanical strain imposed on ASM in vitro or by physiologic maneuvers such as breathing likely trigger intracellular mechanosensitive signaling and cytoskeletal adaptation that may be important in the control of airway myocyte gene expression and cell function. Due to its critical role in actin filament and stress fiber formation, the monomeric GTPase RhoA has been implicated as one of the important effectors of intracellular signaling induced by external mechanical force (10). Of note, signaling of RhoA via its downstream intermediate, the serine-threonine kinase, ROCK-1, is also required for inducing transcription of genes encoding smooth muscle–specific proteins such as smooth muscle myosin heavy chain (smMHC) and SM22 that are associated with the contractile apparatus (11). In ASM, activation of RhoA signaling leads to nuclear redistribution of serum response factor (SRF), a transcription factor that binds as a dimer to CArG box DNA elements (12). These DNA elements exist uniquely in pairs in the 5' promoter region of smooth muscle–specific genes, and binding of SRF is required for activation of transcription (12–14). Furthermore, cytoskeletal actin polymerization, which is regulated in part by RhoA signaling, is also an active player in the regulation of the gene expression of smooth muscle contractile phenotype markers (15). Vascular smooth muscle cells treated with latrunculin to induce truncation of filamentous actin (F-actin) reduced SM22 promoter activity 4-fold, whereas promoter activity was increased 22-fold when cells were treated with jasplakinolide to promote actin polymerization. Despite the potential for monomeric GTPases to mediate cellular responses to mechanical signals, two recent studies using cultured ASM and vascular smooth muscle cells report conflicting evidence that RhoA signaling can be activated or reduced depending on the nature of the mechanical strain applied (16, 17). These divergent findings suggest that unique integrated signaling cascades may be recruited in response to different types of external mechanical strain. For example, mechanical strain activates protein kinase C (PKC), a multigene family of at least 11 isozymes, in many cell types. Takei and coworkers (18) reported that 10% strain generated from a vacuum device stimulated a 62% increase in PKC activity coupled with translocation to the membrane in keratinocytes. Liu and colleagues (19) showed mechanical strain exerted on fetal lung cells elicited PKC translocation to the plasma membrane with a concomitant 6- to 10-fold increase in membrane enzyme activity within 5–15 min. Importantly, PKC isoforms may play a role in integrating networks of signal transduction pathways that control actin cytoskeletal dynamics and gene expression. For example, a recent study revealed that strain-independent activation of PKC causes significant depolymerization of F-actin in the A7r5 rat aortic smooth muscle cell line (20), and some PKC isoforms modulate RhoA-mediated expression of the c-fos gene in fibroblasts (21). Based on current understanding, we tested the hypotheses that short-term uniform biaxial cyclic mechanical strain may concomitantly modulate actin cytoskeletal dynamics and transcriptional activity of the SRF-dependent smooth muscle–specific gene promoters for SM22 and smMHC via both RhoA and PKC signaling pathways. Our studies revealed that cyclic strain led to a significant PKC-dependent loss of RhoA-mediated SM22 and smMHC basal promoter activity in primary cultured canine tracheal myocytes. We also showed that direct activation of PKC with phorbol ester significantly reduced both the activity of these promoters, and the ratio of filamentous-to-globular actin (F/G actin ratio). Moreover, we report that external mechanical strain also induced a significant PKC-dependent reduction of F/G actin ratio. These data reveal that external cyclic biaxial mechanical strain induces coordinated signaling of PKC and RhoA pathways in ASM leading to reduced filamentous actin and SRF-dependent gene transcription. In light of previous work by other investigators (16, 17) our data illustrate that the nature of the response to mechanical strain may be dependent, in part, on the characteristics of the mechanical stimulus.
Materials Cell culture media and LipofectAMINE were purchased from Invitrogen (Carlsbad, CA). Lysis buffer for luciferase assays was purchased from Promega (Madison, WI). Texas Red–phalloidin and Oregon Green–DNase (DNase I) were purchased from Molecular Probes (Eugene, OR). PKC inhibitors, Gö6976 and GF109203X, were purchased from Calbiochem (San Diego, CA). Phorbol-12-myristate-13-acetate (PMA) was purchased from Sigma (St. Louis, MO). Luciferase reporter plasmids containing SM22 and smMHC promoters and the MSVßgal plasmid vector were generous gifts of Dr. Julian Solway, University of Chicago (11, 14). A luciferase reporter plasmid driven by an artificial SRF-specific promoter was purchased from Stratagene (La Jolla, CA).
Cell Culture
Application of Mechanical Strain
Effects of Mechanical Strain and PKC on Transcription of SM22, smMHC, and SRF To measure the contribution of PKC activity to strain-induced effects on promoter activity, in some studies pharmacologic inhibitors of PKC, Gö6976 or GF109203X (1 µM), were included in the media of control or mechanically strained cells for 48 h after transient transfection. To determine whether strain-independent activation of PKC modulated transcriptional activity of p-SM22-luc, p-smMHC-luc, and p-SRF-luc, in one set of experiments myocytes plated on uncoated plastic 6-well culture plates as described above were treated with phorbol-12-myristate-13-acetate (PMA) (1 and 10 µM) for 1.5–2 h immediately after transient transfection. The range of phorbol ester concentration used was chosen based on published reports that disruption of filamentous actin in cultured fibroblasts with up to 10 µM PMA was completely blocked using two different inhibitors of PKC, indicating the response was predominantly mediated by PKC (20). Thereafter the cells were maintained in fresh serum-free DMEM in the absence of PMA for an additional 46 h before measuring luciferase activity. To measure luciferase activity, cell lysates were collected and assayed using the Luciferase Assay System (Promega, Madison, WI) as per manufacturer's instructions. All steps were completed at 4°C unless otherwise specified. Cells were washed with ice-cold Hanks' balanced salt solution, then scraped off of each well in 120 µl of Cell Lysis Buffer. Afterward the cell lysate was centrifuged (14,000 x g, 15 min), and luciferase activity was measured in the supernatant using a Berthold Lumat LB 9507 Luminometer. Luciferase activity was normalized to relative ß-galactosidase activity, which was measured colorimetrically (410 nm) using o-nitrophenyl-ß-D-galactoside (ONPG) as a substrate. All experiments were completed in triplicate using at least three different canine tracheal myocyte cell lines.
Visualization and Quantification of F- and G-Actin
Nuclear Extraction
Electrophoretic Mobility Shift Assay
Statistical Analysis
Mechanical Strain Reduces SM22-, smMHC-, and SRF-Dependent Promoter Activity Luciferase reporter assays revealed that cyclic biaxial strain for 48 h significantly reduced p-SM22-luc, p-smMHC-luc, and p-SRF-luc activity by 55 ± 5.9% (n = 16, P < 0.01), 57 ± 6.4% (n = 16, P < 0.01), and 75 ± 7.9% (n = 19 without strain and n = 9 with strain, P < 0.05), respectively (Figure 1). The inhibitory effect of biaxial mechanical strain on the activity of SM22 and smMHC promoters was not affected when experiments were performed using medium supplemented with 10% FBS, which is a potent activator of RhoA signaling (Table 1). No difference in ß-galactosidase activity, which was used to normalize luciferase activity for potential differences in transfection efficiency, was found between strained and unstrained cells. These data provide clear evidence for biaxial strain-induced inhibition of smooth muscle-specific gene promoters in transiently transfected canine tracheal myocytes.
PKC Activity Is Required for Strain-Induced Inhibition of SM22 and smMHC Promoter Activity As PKC is known to be activated by mechanical strain (19, 25), and strain-independent PKC activation ameliorates RhoA-dependent cell responses (20), we next investigated whether PKC activity was required for strain-induced inhibition of SM22 and smMHC promoter activity. We tested whether two inhibitors of PKC, GF109203X (1 µM), which is pan-PKC isoform specific, and Gö6976 (1 µM), which inhibits Ca2+-sensitive PKC isoforms, affected the strain-induced reduction of the SM22 promoter activity in transiently transfected airway myocytes. Strained cells treated with GF109203X or Gö6976 for the duration of the strain regime exhibited over 50% higher SM22 promoter activity compared with cells exposed to strain alone (P < 0.05, n = 4) (Figure 2A). Thus, in the absence of PKC activity, cyclic biaxial strain-mediated SM22 promoter repression is inhibited by over 70% (P < 0.05, n = 4). Importantly, in the presence of PKC inhibitors, no statistically significant loss of promoter activity occurred during 48 h cyclic biaxial mechanical strain (P > 0.05, n = 4), thus revealing that PKC activity is required for strain-dependent effects on promoter activity. Neither GF109203X nor Gö6976 reduced basal activity of the SM22 promoter in unstrained cells. Similar qualitative and quantitative observations in response to strain and PKC inhibitors were also seen for the smMHC promoter (P < 0.05, n = 4) (Figure 2B). Collectively, these data indicate that activity of PKC associated with external cyclic biaxial mechanical strain applied to cultured airway smooth muscle cells appears to be part of a signal transduction branch required for strain-induced inhibition of smooth muscle–specific transcription activity.
Strain-Independent Activation of PKC Inhibits SM22-, smMHC-, and SRF-Dependent Promoter Activity To better determine whether activation of PKC alone is sufficient to inhibit transcription from promoters for smooth muscle–specific genes, we next investigated the effects of direct activation of PKC in airway smooth muscle. We used the phorbol ester, PMA (1 and 10 µM), to activate PKC in cultured canine tracheal myocytes transiently transfected with p-SM22-luc, p-smMHC-luc, or p-SRF-luc. Treatment for 1.5–2 h resulted in a profound concentration-dependent loss of activity for all three promoters; 1 µM PMA reduced activity of p-SM22-luc and p-smMHC-luc by 29 ± 2% and 25 ± 6%, respectively (n = 3, P < 0.05). Treatment with 10 µM PMA inhibited SM22, smMHC, and SRF promoter activity by 61 ± 4% (n = 3, P < 0.05), 66 ± 5% (n = 3, P < 0.05), and 28 ± 15% (n = 5, P > 0.05), respectively (Figure 3). These data confirm that transient strain-independent activation of PKC is sufficient to inhibit transcription activity of SRF-dependent smooth muscle–specific gene promoters in cultured airway smooth muscle cells.
PKC Activation Inhibits the Association of SRF with CArG Elements in the SM22 Promoter Transcription activity of smooth muscle–specific gene promoters requires RhoA-dependent binding of the transcription factor, SRF, to at least two CArG Box elements, which are characteristic of the regulatory region of genes such as SM22 and smMHC (12, 14, 15). To further explore the mechanism by which PKC activation might affect SRF-dependent activity of the SM22 promoter, we performed gel shift assays. Oligonucleotides that contain either the relatively more 5' or the relatively more 3' CArG element from the murine SM22 promoter region were equilibrated with nuclear extracts from cultured tracheal myocytes pretreated with PMA for up to 2 h. Results of the EMSA revealed a consistent biphasic effect. As shown in Figure 4, the binding of SRF to CArG Box oligonucleotides decreased after 15 min of PMA exposure, returned to near basal levels by 30 min, then this was followed by a second phase in which a dramatic loss of SRF/CArG Box binding was seen 60 min after initial PMA exposure, and this then returned to basal binding levels 60 min later. This pattern was the same for oligonucleotides representative of either the 5' or 3' SM22 CArG Box; a typical EMSA example, using the 5' oligonucleotide sequence is shown in Figure 4.
The specificity of the association between SRF and the CArG-containing oligonucleotides was confirmed using two controls. First, inclusion of a 100-fold excess of unlabeled CArG oligonucleotide, but not the inclusion of an oligonucleotide containing an AP2 element, effectively competed for SRF binding with labeled oligonucleotide, thus leading to a loss of band intensity (Figure 4). Second, during incubation of nuclear lysates with CArG-containing oligonucleotides the addition of anti-SRF antibody, but not the addition of an antibody for AP2, resulted in the formation of a larger protein/DNA complex that appeared as a higher molecular weight band on the EMSA autoradiogram (Figure 4). These studies provide provocative new evidence that is consistent with or experiments that showed PKC activation inhibited activity of an artificial SRF-dependent luciferase reporter (Figure 3), suggesting that PKC-mediated blockade of smooth muscle–specific gene transcription occurs by a mechanism involving inhibition of the RhoA-dependent association of SRF with CArG Box elements in the regulatory region of these genes.
Strain-Dependent and -Independent Activation of PKC Inhibits Actin Polymerization
The Texas Red–phalloidin conjugate used to label filamentous actin revealed stress fibers oriented in parallel with the long axis of the myocytes (Figure 5A). Oregon Green–conjugated DNase I revealed a diffuse distribution of G-actin with the highest intensities in perinuclear regions. Application of external mechanical strain led to a loss in intensity of F-actin labeling, resulting in a significant 31 ± 8% decrease in the F/G actin ratio (P < 0.01, n = 5) (Figure 5B). Mean F/G actin ratio in the absence of strain was 1.62 ± 0.06 (n = 5), whereas after strain the ratio was reduced to 1.11 ± 0.15. Confirming a role for strain-induced PKC activation in reducing F/G actin ratio in cultured airway myocytes, the pan-PKC inhibitor GF109203X (1 µM) completely blocked strain-mediated F/G actin ratio suppression (Figure 5). Inhibition of Ca2+-sensitive PKC subfamilies using Gö6976 inhibited strain-mediated reduction of F/G actin ratio by > 50% (P < 0.05, n = 5). Similar to uniform biaxial strain-induced PKC activation, exposure of myocytes with PMA to promote strain-independent PKC activation induced a significant 30 ± 5% reduction of F/G ratio (P < 0.01) (Figure 6B) compared with untreated myocytes. Indeed, the conditions used in our studies (1 µM PMA, 1.5–2 h) led in some cases to an almost complete loss of F-actin (Figure 6A). Collectively these studies indicate that strain-dependent and -independent activation of PKC appears to generally inhibit RhoA-dependent cell responses such as SRF-mediated transcription of smooth muscle–specific genes and polymerization of filamentous actin in cultured airway smooth muscle cells.
This study was undertaken to investigate the effects of uniform biaxial cyclic mechanical strain on transcription of genes encoding proteins associated with the contractile apparatus (i.e., SM22 and smMHC) in airway smooth muscle, and to identify intracellular signal transduction pathways that mediate strain-induced changes. Previous elegant work by Smith and colleagues (3–5, 16) revealed that prolonged uniaxial cyclic mechanical strain, designed to mimic the perturbation expected with repeated deep inspiration, induced proliferation, and accumulation and rearrangement of contractile and cytoskeletal proteins, with concomitant activation of RhoA-mediated signal transduction in cultured canine airway myocytes. In contrast, using similar primary cultured canine tracheal myocytes, our studies show that external uniform biaxial cyclic mechanical strain significantly reduces RhoA-mediated cell responses including transcriptional activity of SRF-driven promoters such as those for SM22 and smMHC (Figure 1 and Table 1), and F/G actin ratio (Figure 5). Our studies further reveal that uniform biaxial strain does not induce reorientation of myocytes nor an increase in cell number. Our experiments also reveal a previously unknown modulator role for PKC of RhoA-dependent gene transcription (Figure 2) and cytoskeletal plasticity (Figure 5) during applied mechanical strain in airway myocytes. Further, for the first time, strain-independent activation of PKC was observed to be sufficient to inhibit RhoA-dependent activity of SRF-driven promoters for SM22 and smMHC (Figure 3), binding of SRF to CArG elements (Figure 4), and for inducing depolymerization of F-actin (Figure 6) in airway smooth muscle cells. Collectively these data provide new insight into mechanisms that may be critical in determining smooth muscle phenotype during lung development, and the pathogenesis of disease states such as asthma. There appears to be a striking difference in the results of our studies compared with previous reports concerning primary cultured ASM cells (3–5, 16). However, there are specific dissimilarities in the direction, duration, and magnitude of strain used in the current study, and these are likely critical determinants of the unique results that were obtained. By virtue of its design, the strain apparatus used in our studies applied uniform biaxial circumferential stretch to the flexible bottom of the collagen I–coated plates on which cultured myocytes were grown. This contrasts other studies that employed centripetal uniaxial stretch using similar culture plates (16). An important characteristic of uniaxial stretch is that strain is not applied uniformly; rather it is greatest near the periphery of each well and likely approaches a compressive force near the center. This induces cell proliferation and reorientation of myocytes perpendicular to the direction of strain, a process requiring cell motility and the active reorganization of the actin cytoskeleton under the control of small GTPases including RhoA (3, 16, 26, 27). In contrast, in our studies using 48 h of uniform biaxial strain generated by a vacuum of similar magnitude and frequency to that used by other investigators to induce uniaxial strain, we observed a significant decrease in two RhoA-dependent myocyte features (i.e., the activity of SRF-dependent promoters, and F/G actin ratio), and no change in myocyte orientation or number. In a similar fashion, a recent study revealed that tensile strain decreased GTP-loading of RhoA whereas compressive strain increased RhoA activation in cultured vascular smooth muscle cells (17). Thus it seems likely that disparate effects of strain on RhoA-dependent cell responses in our studies and from previous studies using uniaxial strain are manifest from critical differences in the modality of strain that was used. Mechanical forces on airway smooth muscle from fetal breathing movements and spontaneous contractions in the developing lung, during normal breathing, and as the result of bronchial spasm are thought to be an important determinant of myocyte growth, phenotype, and function (1, 28). Prolonged cyclic uniaxial strain leads to a significant increase in the accumulation of contractile proteins and other contractile phenotype markers in airway smooth muscle cells (5). Cyclic mechanical strain increased expression of interleukin-8 by human airway myocytes (29), and of extracellular matrix proteins by vascular smooth muscle cells (30). There is little information available concerning intracellular signaling that mediates strain-induced changes in smooth muscle cell gene expression, in particular genes for proteins considered to be phenotype markers for smooth muscle. Kumar et al. (29) observed that CEBP and AP-1 transcription factors, activated through the mitogen activated protein kinase pathway, regulate strain-induced expression of interleukin-8 in human airway myocytes. Uniaxial strain induced Egr-1 and Sp-1 transcription factors, which drive expression of platelet-derived growth factor A in vascular smooth muscle cells (27). Our experiments reveal that activity of an artificial SRF-specific promoter, and promoters for SM22 and smMHC is reduced by uniform biaxial mechanical strain via PKC-dependent signaling. We also found that strain-independent activation of PKC reduced binding of SRF to CArG box elements in SM22 and smMHC promoters. Because both SRF-binding and the induction of these genes have previously been shown to require RhoA signaling through ROCK-1 (12, 15), our new data suggest a modulator role for PKC in the expression of genes encoding smooth muscle–specific contractile apparatus-associated proteins. Of note, we observed that inhibition of smMHC and SM22 promoter activity by biaxial strain occurred when myocytes were grown in either serum-free or serum-supplemented medium (Table 1). As serum is a potent activator of RhoA, these data suggest that PKC-mediated regulation of SM22 and smMHC promoters is not affected directly by activation of RhoA. Though the precise nature of interactions between PKC and RhoA signal cascades cannot be completely elucidated from our study, the results are consistent with other reports that PKC becomes activated in response to mechanical strain in a variety of cell types (18, 19, 25), and that crosstalk occurs between some PKC isoforms and RhoA signal cascades (20, 21). Our study is also consistent with a recent report (17) that short-term biaxial tensile strain inhibits activation of RhoA in cultured vascular smooth muscle cells. In contrast, our data is in opposition to data indicating that RhoA is activated in airway myocytes exposed to uniaxial mechanical strain (16). Using mammalian expression vectors for active and inactive PKC, Soh and colleagues (21, 31) have demonstrated that specific isoforms appear to play a role in integrating networks of signal transduction pathways, including ERK, JNK, and RhoA that control gene expression through a number of cis-acting regulatory elements. In addition, the activation of PKC with phorbol ester in A7r5 vascular smooth muscle cells appears to lead to activation of p190RhoGAP through the tyrosine kinase Src, leading to diminished RhoA activity. Therefore, our data that show uniform biaxial strain induces PKC-dependent inhibition of Rho-regulated cell responses such as actin polymerization and SRF-driven promoter activity conform well to developing paradigms in which PKC modulates RhoA-dependent cell function. The promoter region of genes encoding smooth muscle–specific proteins contain at least two CArG boxes, which are essential for promoter function and are sites for binding of SRF (13). RhoA–ROCK-1 signaling and actin polymerization are required for nuclear localization of SRF, SRF binding to CArG elements, and activation of promoters for smooth muscle–specific genes (11, 12, 15). There is evidence that polymerized actin is required to repress the inhibitory effect of YY1 on SRF-driven SM22 promoter activity in vascular smooth muscle cells (32). Also, Mack and coworkers (15) have shown that loss of filamentous actin induced by latrunculin or Y-27632, a ROCK-1 inhibitor, dramatically decreases SM22 promoter activity in vascular myocytes. Recent reports reveal that acute tensile strain applied to cultured vascular smooth muscle cells or to vascular myocytes in intact veins results in a marked loss of F-actin (17, 33). In the present study, we measured F- and G-actin abundance using phalloidin and DNase I (34), and found that though basal F/G actin ratio was consistent with that published by other investigators (35, 36), the F/G actin ratio was significantly reduced by uniform biaxial strain in a PKC-dependent fashion. Collectively, these observations and our data indicate that PKC activity, during the application of biaxial mechanical strain or resulting from treatment with phorbol esters, likely exerts an inhibitory effect on smMHC and SM22 transcription, in part, through its ability to disrupt filamentous actin.
PKC is a family of phospholipid-dependent serine/threonine kinases comprising 11 isoforms in 3 subclasses (i.e., conventional, novel, and atypical). These enzymes have a motif structure consisting of common catalytic domain in all isoforms, a Ca2+-binding C1 region, present only in conventional isoforms, and a diacylglyceride-binding site in region C2 that is present in members of conventional and novel subclasses (37). In various cell types, cyclic strain stimulates PKC activity coupled with its translocation from the cytosol to the plasma membrane. Our findings revealed an equal ability of GF109203X and Gö6976 to inhibit strain-mediated repression of promoter activity and actin polymerization. Gö6976 selectively inhibits Ca2+-dependent PKC as well as PKCµ (38) and GF109203X acts as a competitive inhibitor for the ATP-binding site of PKC, showing high selectivity for PKC Cells of the lung are subject to mechanical strain throughout life, for example during early development in the form of fetal breathing movements, and thereafter as the consequence of normal breathing, and intermittent bronchospasm. The mechanical forces to which airway myocytes are exposed are complex. Due to their location and orientation in the airway wall, which is tethered to the lung parenchyma, and in combination with their ability to contract, airway myocytes are likely subjected to frequent multivectorial strain. In this regard, the 12.5% stretch imposed with biaxial mechanical strain regime employed in our studies for monolayer airway myocyte cultures may in part approximate mechanical forces that develop in vivo. Under physiologic conditions, airway smooth muscle is exposed to 4–5% stretch during tidal breathing, whereas during a deep inspiration when lung volume is doubled, airway smooth muscle is likely stretched by as much as 25% (40, 41). Our studies indicate that PKC and RhoA signaling, which are both known to be affected by external mechanical strain in a number of lung cell types, are not mutually exclusive, and cell responses to strain are likely manifest from coordinated integration of each signaling cascade. The mechanism for integration is poorly understood at present. That PKC appears to be a key modulator of the F/G actin ratio and is required for mechanical strain–mediated downregulation of smooth muscle–specific gene transcription, implies a much broader role for the plasticity of the contractile apparatus described for cells and intact tissues (7, 8). This study provides evidence that mechanical perturbation not only disrupts already formed contractile apparatus, but that this disruption may also regulate transcription of contractile apparatus–associated proteins such as smMHC and SM22 through integrated signaling likely involving PKC and RhoA. Expression of genes encoding these proteins is associated with myocyte differentiation and phenotype plasticity that are determinants of airway smooth muscle function in health and disease (10).
The authors thank Dr. Julian Solway from University of Chicago for the generous gift of luciferase reporter plasmids. They thank A. Vörös, K. Kassiri, C. T. Hillier, J. Tam, and Z. Al-Hariri for helpful suggestions. L.W. is supported by a CIHR/CLA/AstraZeneca Senior Research Fellowship. A.J.H. is a CIHR/CLA Scholar. This study was supported by operating funds from CIHR, CIHR/CLA/AstraZeneca, MICH, MHRC, MMSF, and CFI. Received in original form June 24, 2003 Received in final form January 29, 2004
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
24 h after replating). Three luciferase reporter plasmids constructed in the pGL2basic vector were used in these studies: first, p-SM22-luc, harboring the minimal murine SM22 5' promoter (bp –445 to +41) to drive luciferase expression; second, p-smMHC-luc, harboring 3.3 kb of human smMHC 5' promoter that includes the first intron to drive luciferase expression; and, third, p-SRF-luc, harboring an artificial promoter comprised of five copies of SRF binding sites (CC[A/T]6GG or "CArG box") in series, upstream from minimal TATA box. For transient transfection of cells grown on 6-well BioFlex dishes, cells in each well were incubated in 1 ml of serum-free Optimem medium containing 6 µg Lipofectamine, 0.4 µg of p-SM22-luc, p-smMHC-luc, or p-SRF-luc, and 0.4 µg of the pMSVßgal expression plasmid. The pMSVßgal plasmid was constructed in the pGL2basic vector, in which the viral MSV-LTR promoter was used to drive ß-galactosidase expression. By including pMSVßgal, luciferase activity for each promoter construct could be normalized for transfection efficiency in all experiments. Transfection was terminated after 5–6 h by replacing transfection medium with serum-free DMEM. Cells were then incubated in serum-free medium for an additional 48 h with or without applied mechanical strain as described above. In some studies, to compare the effects of strain on cells grown in serum-free and serum-supplemented medium, myocyte cultures were maintained in DMEM/10% FBS for 48 h after transfection. 
-32P] ATP. The oligonucleotides encompassed the 5' (5'-GCTGCCCATAAAAGGTTTTTG-3') or 3' (5'-CTTTCCCCAAATATGGAGCCTG-3') CArG boxes (underlined) of the mouse SM22 promoter. A quantity of 20,000 dpm (1–5 fmol) radiolabeled oligonucleotide was preincubated for 15 min with 1.5 µl binding buffer (50 mM Tris HCl pH 7.5, 20% Ficoll, 375 mM KCl, 5 mM EDTA, 5 mM DTT) and 1 µg poly (dI-dC). Molar excess (200-fold) of unlabeled competitor oligonucleotide was added. Binding reactions using 5 µg of nuclear protein extract were performed at room temperature in 15 µl for 30 min. For supershift experiments, 4 µg of antibody was also added to the incubation mixture. Supershift antibodies included anti-SRF and anti-AP2
(Santa Cruz Biotechnology, Santa Cruz, CA). DNA–protein complex formation was analyzed by electrophoresis on 5% non-denaturing polyacrylamide gels in TBE buffer (40 mM Tris Borate, 1 mM EDTA) and autoradiography. 
, and 
isozymes (
