Mechanical induction of growth factor synthesis may mediate adaptive responses of smooth muscle cells (SMC) to increases in physical load. We previously demonstrated that cyclic mechanical stretch induces expression of the SMC, fibroblast, and epithelial cell mitogen heparin-binding epidermal growth factor-like growth factor (HB-EGF) in bladder SMC, an observation that suggests that this growth factor may be involved in compensatory bladder hypertrophy. In the present study we provide evidence that the activator protein-1 (AP-1) transcription factor plays a critical role in this mechanoinduction process. Rat bladder SMC were transiently transfected with a series of 5′ deletion mutants of a promoter-reporter construct containing 1.7 kb of the mouse HB-EGF promoter that was previously shown to be stretch responsive. The stretch-mediated increase in promoter activity was completely ablated with deletion of nucleotide positions −1301 to −881. Binding of AP-1, as evaluated by electrophoretic mobility shift assay, to a synthetic oligonucleotide containing an AP-1 binding site increased in response to stretch, and binding was inhibited by excess unlabeled DNA corresponding to nucleotides −993 to −973 from the HB-EGF promoter, a region that contains a previously recognized composite AP-1/Ets site. Stretch-induced promoter activity was significantly inhibited by site-directed mutagenesis of the AP-1 or Ets components of this site. Consistent with the promoter and gel-shift studies, curcumin, an inhibitor of AP-1 activation, suppressed the HB-EGF mRNA induction after stretch. Stretch also specifically increased mRNA levels for matrix metalloproteinase (MMP)-1, the promoter of which contains a functional AP-1 element, but not for MMP-2, the promoter of which does not contain an AP-1 element. The stretch response of the MMP-1 gene was also completely inhibited by curcumin. Collectively, these findings indicate that AP-1-mediated transcription plays an important role in the regulation of gene expression in bladder muscle in response to mechanical forces.
- heparin-binding epidermal growth factor
- mechanical signaling
- gene expression
- activator protein
a variety of cell types have been shown to respond to mechanical forces by activating multiple signal transduction cascades, altering their program of gene expression and increasing their rates of protein and/or DNA synthesis (9, 10, 15, 16, 18, 26, 29, 30). These physiological responses have been observed in culture systems in which cells are subjected to quantitative application of forces of various kinds, such as radial stretch and shear stress. In vitro systems have provided evidence that pathophysiological tissue remodeling seen in vivo, such as cardiac hypertrophy in response to hemodynamic overload, is the result of regulated mechanochemical signaling within specific cellular compartments (13, 31).
Bladder hypertrophy occurs as an adaptive response to physical or neuromuscular obstruction of the lower urinary tract arising in adult men primarily from age-related growth of the prostate gland and in children in association with several congenital uropathic syndromes. The bladder detrusor muscle increases in size as a compensatory response to urine outflow obstruction. Experiments in animal models have indicated that compensatory bladder growth follows a highly regulated, idiosyncratic course, which includes alterations in the mRNA levels of some growth-related genes, an early proliferative response by the uroepithelium and fibromuscular cells of the lamina propria, and hypertrophic as well as hyperplastic expansion of the bladder muscle (reviewed in Refs. 5 and 14). The molecular mechanisms that underlie these physiological changes observable in vivo and that ultimately result in bladder decompensation and functional failure, are largely unknown.
Repetitive stretch and relaxation applied to bladder smooth muscle cells (SMC) in vitro has been used to model increases in urodynamic load experienced by the bladder detrusor muscle under conditions of bladder outlet obstruction. In a recent report from our laboratory, Park et al. (22) used such an in vitro system to identify the mitogen heparin-binding epidermal growth factor-like growth factor (HB-EGF) as a stretch-responsive protein in bladder SMC. HB-EGF is an activating ligand for the ErbB1 receptor tyrosine kinase (EGF receptor) and a growth factor with a mitogenic potency for SMC that is greater than that seen with other cognate ligands for the same receptor, such as EGF (reviewed in Ref. 25). HB-EGF is synthesized by the detrusor muscle in vivo, and HB-EGF mRNA and protein expression increase predominantly in the muscle compartment during acute urinary obstruction in mice (unpublished observations; 8). Consequently, HB-EGF is a potential physiological mediator of SMC growth in the urinary tract.
A repetitive stretch-relaxation stimulus applied to bladder SMC adhered to a deformable membrane resulted in increased expression of HB-EGF mRNA and protein in bladder SMC by a mechanism that involved secretion of the peptide hormone ANG II and activation of the angiotensin receptor type 1 (AT1) (22). In these experiments, stretch-induced increases in HB-EGF mRNA levels could be completely accounted for by activation of the HB-EGF promoter, suggesting that the primary mechanism of HB-EGF induction under these conditions is transcriptional activation. In the present experiments we have analyzed the HB-EGF promoter in bladder SMC in a search forcis-acting DNA elements that mediate the stretch response.
MATERIALS AND METHODS
Rat bladder SMC were harvested by an enzymatic dispersion method, as described previously (22). Cells were cultured in medium 199 (GIBCO) supplemented with 20% fetal bovine serum (FBS; Hyclone Laboratory), penicillin (100 U/ml), and streptomycin (100 μg/ml) and were maintained in a humidified 5% CO2-95% air atmosphere at 37°C. All experiments were performed on cells betweenpassages 2 and4.
Application of cyclical stretch-relaxation.
Bladder SMC were grown on six-well silicone Elastomer-bottomed culture plates that had been coated with collagen type I (Bioflex, Flexcell, McKeesport, PA). Cells were rendered quiescent by incubation in medium 199 supplemented with 0.5% FBS for 48 h. The cells were then subjected to continuous cycles of stretch and relaxation by use of a computer-driven, vacuum-operated stretch-inducing device (Flexercell Strain Unit FX-3000) for variable duration as indicated. Each cycle consisted of 5 s of stretch and 5 s of relaxation (0.1 Hz). The vacuum induced an ∼20% maximum radial stretch at the periphery of the membrane.
Semiquantitative RT-PCR assays were performed to assess relative mRNA levels. Total RNA extraction was performed using Tri-Reagent (Molecular Research, Cincinnati, OH). RT was performed using Maloney’s murine leukemia virus reverse transcriptase with oligo(dT) as the first-strand primer. Primers were selected from the previously published rat HB-EGF, matrix metalloproteinase (MMP)-1 and -2, and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) sequences (19, 24). A 413-bp HB-EGF product was amplified using the sense primer 5′-TCC CAC TGG AAC CAC AAA CCA G-3′ (nt 157–178) and the antisense primer 5′-CCC ACG ATG ACA AGA AGA CAG AC-3′ (nt 570–548). A 494-bp MMP-1 product was amplified using the sense primer 5′-AGG TGA AAA GGC TCA GTG CTG C-3′ (nt 1119–1140) and the antisense primer 5′-GAT CCT TGG GGC TCT CAA TTT C-3′ (nt 1591–1612). A 591-bp MMP-2 product was amplified using the sense primer 5′-TTC AGA AGG TGC CCC ATG TG-3′ (nt 1502–1521) and the antisense primer 5′-TTC CCT GCG AAG AAC ACA GC-3′ (nt 2073–2092). A 571-bp GAPDH product was amplified using the sense primer 5′-TCA CCA TCT TCC AGG AGC G-3′ (nt 245–263) and the antisense primer 5′-CTG CTT CAC CAC CTT CTT GA-3′ (nt 816–797). PCR reactions were performed in a total volume of 25 μl, containing 22 μl of PCR SuperMix (GIBCO), 0.5 μl each of sense and antisense primers (20 pmol/μl), 0.1 μl of [32P]dCTP (3,000 Ci/mmol, Amersham), and 2 μl of cDNA. PCR amplification was performed for 26–28 cycles. PCR products were subjected to size separation by PAGE. All samples were normalized to GAPDH mRNA levels, and a limiting dilution method was used to make semiquantitative comparisons.
HB-EGF promoter deletion and mutation.
The full-length HB-EGF promoter-luciferase construct (pHB-EGF-luc) has been described previously (6). It contains a ∼1.7-kbMboI-Not I fragment derived from the 5′-untranslated region of a murine HB-EGF genomic clone (corresponding to sequences −1837 to −155 relative to the translation initiation site) ligated upstream of a luciferase reporter construct pGL2Basic (Promega, Madison, WI). A series of deletion mutant constructs from −1837 to −477, generated by PCR, were previously used to demonstrate the requirement for sequence upstream of position −912 to elicit HB-EGF promoter activation in response to Raf-1 activation (20). To examine the possible role of a composite AP-1/Ets binding site at nucleotide position −988 to −974 in stretch-induced activation of the HB-EGF promoter, we performed site-directed mutagenesis of the −1301pHB-EGF-luc deletion mutant from this series. The numbering convention we have adopted is based on that of Chen et al. (6); however, sequencing of the deletion mutant −1301pHB-EGF-luc revealed several discrepancies with the reported sequence, and we have amended the sequence and numbering accordingly. With use of a PCR-based approach, two fragments were generated for each transcription factor binding site mutant to be created. The upstream fragment was created using a wild-type sense primer and an appropriate antisense mutant primer. The downstream fragment was created using an antisense wild-type primer and an appropriate sense mutant primer. This resulted in the generation of two fragments that overlapped in the region of the desired mutation. The appropriate fragments were then mixed in a 1:1 molar ratio and used as a template for primer extension with use of wild-type primers wt-S and wt-AS and subsequent amplification to regenerate the full-length constructs incorporating mutations in the AP-1 or Ets component of the site or in both components. All PCR reactions were performed using the Expand High-Fidelity PCR System (Boehringer Mannheim), and products were purified using the High Pure PCR Product Purification Kit (Boehringer Mannheim). Incorporation of the desired mutations was confirmed by DNA sequencing, and products were digested with NheI and BglII restriction enzymes before ligation intoNhe I/BglII-digested −1301 pHB-EGF-luc deletion construct. Putative recombinants were confirmed by analytic digests, and large-scale preparation of transfection-quality plasmid was performed using the Qiagen Maxi-Prep Kit.
Transfection and luciferase assay.
Cells were grown to 60–70% confluence (∼1 × 105/well) on Bioflex plates. DNA-Superfect (Qiagen) mixtures (total plasmid 2 μg) were added to cells and incubated for 1 h at 37°C. Cells were washed and incubated for 24 h in the standard culture medium containing 20% FBS. Cells were then rendered quiescent by incubation in 0.5% FBS containing medium 199 for 48 h and subjected to stretch stimulation for 12 h. Cell lysates were harvested, and luciferase activity was measured immediately after the addition of luciferase substrate. A sample of cells was similarly transfected with the promoterless pGL2-basic and served as control.
Preparation of nuclear extracts.
Quiescent bladder SMC were subjected to cyclical stretch and relaxation for 2 h. Cells were collected in ice-cold PBS and centrifuged. The cell pellet was resuspended and incubated for 15 min on ice in a buffer containing 10 mM HEPES (pH 7.9), 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, 2 μg/ml aprotinin, 2 μg/ml leupeptin, 1 mM dithiothreitol (DTT), and 1 mM phenylmethylsulfonyl fluoride (PMSF). NP-40 (3% vol/vol) was added and mixed vigorously for 10 s to lyse cell membranes. After centrifugation at 25,000 g for 30 s, the resulting nuclear pellet was resuspended and incubated for 30 min on ice in the extraction buffer containing 20 mM HEPES (pH 7.9), 0.42 M NaCl, 1 mM EDTA, 1 mM EGTA, 2 μg/ml aprotinin, 2 μg/ml leupeptin, 1 mM DTT, and 1 mM PMSF. After centrifugation at 25,000g for 5 min, supernatant was collected as nuclear extracts and stored at −80°C.
Electrophoretic mobility shift assay.
Nuclear extract (2 μg) was incubated with 0.5 μg of poly(dI-dC) (Pharmacia) and 104 cpm32P-labeled DNA fragment in 20 μl of buffer containing 20 mM HEPES (pH 7.9), 50 mM KCl, 5 mM EDTA, 1 mM DTT, 3 mM MgCl2, and 5% glycerol. After incubation on ice for 20 min, the binding reaction was analyzed by electrophoresis on 4% polyacrylamide gel (19:1 bisacrylamide) in 0.25× Tris-boric acid-EDTA buffer. A double-stranded oligonucleotide, 5′-CGCTTGATGACTCAGCCGGAA-3′, containing a putative AP-1 binding element was labeled with32P by use of T4 polynucleotide kinase and used as a probe. Competition experiments were performed by preincubating nuclear extracts with a 50-fold molar excess of unlabeled competitor DNA fragment at 4°C for 10 min. An unlabeled oligonucleotide, 5′-TCT TCTTCCTGT-3′, derived from the HB-EGF promoter region (nt −993 to −973) containing the implicated AP-1 binding site (underscored), was used as a competitor. A c-Fos polyclonal rabbit antibody (Santa Cruz Biotechnology) was used in the supershift experiments to identify AP-1. Rabbit polyclonal IgG served as a control. Nuclear extracts were incubated with antibody for 10 min before the32P-labeled DNA fragments were added.
Stretch stimulates HB-EGF expression in bladder SMC.
Quiescent bladder SMC were grown to near confluence on collagen-coated silicone membranes and subjected to continuous cycles of stretch and relaxation for 0, 1, 2, 4, and 8 h. Relative HB-EGF mRNA levels were assessed using semiquantitative RT-PCR with normalization to GAPDH mRNA levels. With stretch, HB-EGF transcript levels increased rapidly, reaching the peak levels after 4 h (Fig.1). This pattern was reproducible in three independent experiments, consistent with our previously published results (22).
HB-EGF promoter activation by stretch requires a proximal AP-1 element.
We demonstrated previously that mechanical stretch stimulates HB-EGF promoter activity (22). Bladder SMC that had been transiently transfected with a DNA reporter construct containing >1.7 kb 5′-flanking region of the mouse HB-EGF gene cloned upstream to a luciferase reporter gene demonstrated a significant increase in promoter activity after stretch compared with nonstretch controls. This DNA fragment encompasses positions −1885 to −155 upstream of the translation initiation codon in exon 1 and contains possible binding sites for nuclear factor-κB (NF-κB), AP-1, Ets, E-box, Pit-1, and SP-1 transcription factors on the basis of consensus sequence analysis (6). To identifycis-acting elements within this promoter region that confer stretch responsiveness to the HB-EGF gene, we transiently transfected bladder SMC with a series of deletion mutant constructs cloned upstream of luciferase and subjected the transfected cells to cyclical stretch and relaxation for 12 h (Fig.2). Successive deletion of the promoter from nucleotides −1885 to −1301 had no discernible effect on promoter activation by mechanical stretch compared with the full-length construct. However, deletion of the sequences between nucleotides −1301 and −881 resulted in an almost complete loss of promoter activity after mechanical stretch. Consistent with this finding, further deletions through nucleotide −477 were similarly unresponsive to stretch. These results suggested that DNA sequences between positions −1302 and −881 were required for full activation of the HB-EGF promoter by mechanical stretch.
Next we investigated whether specific transcription factor binding sites within this region were required for activation of the HB-EGF promoter after stretch. A composite AP-1/Ets transcription factor binding site located at positions −988 and −974 was previously identified by DNA sequence analysis (6). The Ets component of this site has been demonstrated to be functional in NIH/3T3 cells in response to ligand-dependent activation of a conditionally activatable Raf-1 kinase fusion protein (20); however, AP-1 binding to this site has not been demonstrated. We were not able to identify another consensus transcription factor binding site in the −1301 to −881 region of the published sequence by computer analysis (Fig.3). Therefore, we introduced specific mutations into the −1301 deletion construct at the AP-1/Ets site in an attempt to alter the stretch response. We used mutations shown previously to ablate Raf-1-induced transcription at this site (20). During DNA sequencing to confirm incorporation of the desired mutations into the promoter, we identified several discrepancies with the published sequence. These discrepancies were previously noted by McCarthy et al. (20); however, these investigators did not amend the published sequence. The new sequence data are presented in Fig. 3, with the numbering amended accordingly.
Bladder SMC were transiently transfected with wild-type and mutant variants of the −1301 construct in which the AP-1 and Ets components of the targeted AP-1/Ets site were modified (Fig.4 A). Transfected cells were subjected to cyclical stretch and relaxation for 12 h (Fig. 4 B). The cells that had been transfected with the partially deleted wild-type construct demonstrated full promoter activation in response to mechanical stretch, whereas the constructs transfected with the mutant AP-1 and mutant Ets constructs, and a double mutant in which both sites were modified, showed a significant attenuation (≥70%) of stretch responsiveness (Fig. 4 B). Mutation of the AP-1 component resulted in a quantitatively significant decrease in unstimulated transcription, suggesting that AP-1 is required for significant levels of basal expression of the HB-EGF gene. These results indicate that full activation of the HB-EGF promoter by the stretch signal requires the presence of the AP-1/Ets site located at nucleotides −988 and −974.
We then determined whether an increase in binding activity of the AP-1 transcription factor could be detected at this site after mechanical stretch by performing electrophoretic mobility shift assay (EMSA). Quiescent bladder SMC were subjected to cyclical stretch and relaxation for 2 h. Compared with nonstretch controls, cells that had been stretched demonstrated a marked increase in AP-1 binding activity (Fig.5). This binding was completely inhibited when nuclear extracts were coincubated with unlabeled competitor oligonucleotides corresponding to positions −993 and −973 of the HB-EGF promoter, which include the AP-1 site at position −988 to −982 and flanking sequences. Incubation of nuclear proteins before EMSA with anti-c-Fos antibody resulted in a supershifted complex as well as diminution of AP-1 binding intensity. Preincubation of nuclear extracts with a control antibody (rabbit IgG) did not result in supershift of the complex.
Curcumin, an AP-1 inhibitor, suppresses stretch-induced HB-EGF expression.
Previous studies have demonstrated that curcumin (diferuloylmethane), a potent inhibitor of tumor promotion, inhibits AP-1-mediated transcription (11). As an additional test of the involvement of AP-1 in the stretch response of the endogenous HB-EGF promoter, we examined the effect of curcumin on HB-EGF mRNA expression in bladder SMC after mechanical stretch. Quiescent bladder SMC were subjected to cyclical stretch and relaxation for 4 h, in the presence or absence of 20 μM curcumin, and the relative levels of HB-EGF mRNA were assessed by semiquantitative RT-PCR. Stretch-induced expression of HB-EGF was almost completely attenuated in the presence of curcumin (Fig.6). This finding is consistent with the conclusion that AP-1 activation is required for transcriptional induction of the HB-EGF gene by stretch.
The above findings led us to test the hypothesis that the presence of functional AP-1 cis-elements in the promoter region could predict stretch responsiveness of other genes in this system. We therefore examined the effect of mechanical stretch on the expression of two MMP genes, one with a functional AP-1 binding site and one without an AP-1 binding site in its promoter. MMPs are frequently upregulated in vivo at sites of tissue remodeling (reviewed in Ref. 21) and metalloproteinase activation has been previously associated with acceleration of rates of soluble HB-EGF secretion (7,28). The rat interstitial collagenase (MMP-1) gene has been shown to contain a functional AP-1 element in its promoter in response to basic fibroblast growth factor stimulation (1). In contrast, the rat 72-kDa type IV collagenase (MMP-2) gene does not contain any known AP-1 binding site in its promoter (3). Quiescent bladder SMC were subjected to cyclical stretch and relaxation for 4 h. By semiquantitative RT-PCR, steady-state MMP-1 mRNA levels increased significantly (∼6-fold) after stretch, and the level of MMP-2 mRNA was not affected by stretch. Furthermore, the AP-1 inhibitor curcumin completely attenuated the stretch-induced increase in MMP-1 levels, but it had no discernible effects on MMP-2 mRNA expression (Fig. 6).
In this study we have identified the AP-1 transcription factor as an essential element in the mechanism of transcriptional activation of the HB-EGF gene in response to a repetitive mechanical stimulus. These studies and other recent reports from our group (unpublished observations; 8, 22) have identified HB-EGF as a mechanically regulated factor of potential physiological relevance to bladder physiology. HB-EGF was first identified as a stretch-responsive gene in bladder SMC (22). In this first report, induction of HB-EGF gene expression by stretch was found to be under partial control of signaling through the angiotensin receptor AT1. HB-EGF is synthesized by the bladder detrusor muscle on the basis of immunohistochemical analysis of human and mouse bladder tissue (unpublished observations; 8). In these studies the membrane-anchored precursor of mature HB-EGF (proHB-EGF) was immunolocalized to the detrusor muscle with use of two different antibodies directed against the proHB-EGF cytoplasmic tail domain, as well as antibodies to the mature growth factor. HB-EGF synthesis was also shown in the uroepithelium, and HB-EGF was identified as an autocrine uroepithelial cell growth factor (8). HB-EGF mRNA and protein expression were also observed to increase, primarily in the muscle compartment, after bladder outlet obstruction in mice (unpublished observations). The predominant cognate receptor for HB-EGF, the ErbB1 receptor tyrosine kinase, is present in the detrusor muscle and in the epithelial layer of the bladder mucosa (2; unpublished observations). These findings, taken together, suggest a role for HB-EGF in the bladder’s response to injury and/or as a potential mediator of bladder wall thickening in compensatory hypertrophy.
The system we used for the present and an earlier study (22) employs a vacuum device that deforms, in a highly controlled fashion, the substrate on which adherent cells are plated. A similar or identical system has been used by other groups to examine various responses of other cell types, including cardiac myocytes and fibroblasts, vascular SMC and endothelial cells, and glomerular mesangial cells, to mechanical stimuli. Many fewer studies have been attempted using bladder SMC, and few details of the molecular mechanism of compensatory bladder hypertrophy are known. Cardiac hypertrophy is a condition where tissue remodeling is believed to proceed as a compensatory response of the heart muscle to increases in hemodynamic load. This tissue response bears at least a superficial resemblance to the response of the bladder wall to outlet obstruction. A series of studies using cardiac myocytes have identified a variety of molecular mediators of stretch-induced intracellular signaling. Stretch of cardiac myocytes or fibroblasts by use of a system similar to that used here induces the activation of mitogen-activated protein kinase cascades and the expression of immediate-early genes as well as genes the expression of which is characteristic of fetal development of the heart (13, 18, 31). These results have been interpreted to reflect the response of the cardiac muscle during cardiac hypertrophy. Importantly, the vasoregulatory peptide ANG II, signaling through the AT1 angiotensin receptor, has been identified as a mechanically induced mediator of cardiac hypertrophy (31). These findings are consistent with our previously reported results in bladder SMC and suggest that mechanical forces regulate similar sets of genes in both cell types. We previously identified ANG II as a peptide secreted by bladder SMC in response to the repetitive stretch stimulus used in the present study (22). Blockade of angiotensin receptors inhibited induction of HB-EGF mRNA promoter activity and DNA synthesis by bladder SMC in response to stretch, suggesting that the angiotensin system is a mediator of the stretch response. This signaling system may be relevant to bladder hypertrophy in vivo, because angiotensinogen, the ANG II precursor, and AT1 are present in the bladder wall (27).
In the present study we identified AP-1 as an essential mediator of the stretch-induced stimulation of HB-EGF expression by1) deleting the region of the HB-EGF promoter containing a composite AP-1/Ets binding site, showing loss of the stretch response; 2) mutational analysis of the putative binding site, demonstrating that inactivating mutations abolished the stretch response;3) EMSA, demonstrating an increase in AP-1 binding activity with stretch, and specific ablation of this response by competition with an oligonucleotide corresponding to the putative stretch-responsive region of the HB-EGF promoter; and4) pharmacological inhibition of HB-EGF induction by stretch with curcumin, an AP-1 inhibitor. Additional evidence implicating AP-1 as a mediator of the stretch signal was provided by our demonstration that the rat MMP-1 gene, the promoter of which contains an AP-1 site demonstrated in previous studies to be activated in response to growth factor stimulation, is also responsive to stretch in this system, whereas the MMP-2 gene, which is not regulated by AP-1, does not respond to stretch or to curcumin. Like the HB-EGF gene, stretch-induced activation of MMP-1 mRNA is inhibitable by curcumin. Consistent with our present results, AP-1 has been identified as a mediator of mechanical stretch in vascular SMC, endothelial cells, and cardiac myocytes (10, 13, 30). Consequently, AP-1 appears to be an important mediator of mechanical regulation of gene expression in several cell types.
In our study, introduction of an inactivating mutation into the Ets component of the AP-1/Ets site also abolished the stretch signal, suggesting that an Ets family transcription factor cooperates with AP-1 to mediate the stretch response. Preliminary results indicate that stretch activates Ets binding activity in the bladder SMC system (unpublished observations); however, we have not been able to confirm that the responsible transcription factor at the AP-1/Ets site in the HB-EGF promoter is Ets-2, which has been shown to bind this site in NIH/3T3 cells. The promoter for tissue inhibitor of MMP-1 (TIMP-1), an endogenous inhibitor of MMP-1 and other metalloproteinases that is often coordinately regulated with MMPs during tissue remodeling conditions, contains a composite AP-1/Ets site in which AP-1 and Ets-1 act synergistically to activate TIMP-1 transcription (17). Preliminary data from our laboratory indicate that TIMP-1 is also a stretch-responsive gene in bladder SMC (unpublished observations). This finding may have physiological relevance to bladder disease, because our laboratory has previously demonstrated that upregulation of TIMP-1 was associated with bladder hypertrophy in an in vivo model of bladder outlet obstruction (23). An attractive hypothesis is that the AP-1/Ets site in the TIMP-1 promoter is a critical mediator of the stretch response and participates in the dysregulation of the proteolytic balance associated with obstruction-induced bladder fibrosis.
It remains to be determined how the stretch signal, which may be converted to a biochemical pathway by membrane-bound receptors, possibly integrins (18), is transmitted to the nucleus in bladder SMC. One possibility is the JNK pathway, a known regulator of AP-1 via phosphorylation of the Jun component of the heterodimeric AP-1 transcription factor by the SAPK/JNK mitogen-activated protein kinase. Activation of the JNK pathway by stretch has been demonstrated in cardiac myocytes (13, 16) and vascular SMC (9) and by shear stress in endothelial cells (15).
It will also be of interest to determine whether stretch-responsive genes identified in other cell systems, particularly in SMC, are similarly stretch responsive in bladder SMC. Such a result would suggest that mechanoinduction of biochemical pathways related to tissue growth are conserved across several cell types and organ systems. Consistent with this possibility, Igura et al. (12) demonstrated that HB-EGF is rapidly induced in neointimal cells in rat carotid arteries in response to balloon injury, possibly in response to a stretch signal. An alternative possibility is that mechanical signals activate cell-specific pathways and responses. This question remains to be resolved. Studies in the literature provide evidence that induction of gene expression in response to mechanical signals can vary with cell type, the strength of the mechanical stimulus, and the extracellular matrix composition of the substratum (18, 26, 29). For example, in contrast to our findings, Yang et al. (32) recently reported that platelet-derived growth factor- or tumor necrosis factor-α-induced synthesis of MMP-1 was suppressed by small mechanical strains in vascular SMC. These studies suggest that mechanoregulation of gene expression may be cell and context dependent.
In conclusion, these studies have identified HB-EGF as a mechanically regulated gene in bladder SMC and have determined that this mechanochemical regulatory mechanism requires stretch-induced activation of the transcription factor AP-1. We also provide preliminary evidence that this transcriptional mechanism may apply to a network of similarly regulated genes. One important goal will be to identify other genes in this network so as to better understand mechanical signaling of growth-regulatory pathways in the bladder wall and possibly in other hollow organs such as the heart. We propose that our studies suggest that HB-EGF may play a role in obstruction-induced increases in DNA and protein synthesis in the bladder wall in animal models of bladder outlet obstruction and in the human disease.
This study was funded by National Institutes of Health Grants RO1 DK-47556, RO1 DK-47582, and RO1 CA-77386.
Address for reprint requests and other correspondence: M. R. Freeman, Enders Research Laboratories, Rm. 1151, Children’s Hospital, 300 Longwood Ave., Boston, MA 02115 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.
- Copyright © 1999 the American Physiological Society