Chemotactic movement of myofibroblasts is recognized as a common means for their sequestration to the site of tissue injury. Following myocardial infarction (MI), recruitment of cardiac myofibroblasts to the infarct scar is a critical step in wound healing. Contractile myofibroblasts express embryonic smooth muscle myosin, α-smooth muscle actin, as well as collagens I and III. We examined the effects of cardiotrophin-1 (CT-1) in the induction of primary rat ventricular myofibroblast motility. Changes in membrane potential (Em) and Ca2+ entry were studied to reveal the mechanisms for induction of myofibroblast migration. CT-1-induced cardiac myofibroblast cell migration, which was attenuated through the inhibition of JAK2 (25 μM AG490), and myosin light chain kinase (20 μM ML-7). Inhibition of K+ channels (1 mM tetraethylammonium or 100 μM 4-aminopyridine) and nonselective cation channels by 10 μM gadolinium (Gd3+) significantly reduced migration in the presence of CT-1. CT-1 treatment caused a significant increase in myosin light chain phosphorylation, which could be inhibited by incubation in Ca2+-free conditions or by application of AG490, ML-7, and W7 (100 μM; calmodulin inhibitor). Monitoring myofibroblast membrane potential with potentiometric fluorescent DiBAC4(3) dye revealed a biphasic response to CT-1 consisting of an initial depolarization followed by hyperpolarization. Increased intracellular Ca2+, as assessed by fluo 3, occurred immediately after membrane depolarization and attenuated at the time of maximal hyperpolarization. CT-1 exerts chemotactic effects via multiple parallel signaling modalities in ventricular myofibroblasts, including changes in membrane potential, alterations in intracellular calcium, and activation of a number of intracellular signaling pathways. Further study is warranted to determine the precise role of K+ currents in this process.
- myocardial infarction
- wound healing
- membrane potential
- cardiac fibroblast
- cardiac fibrosis
recruitment of cardiac myofibroblasts and subsequent cellular repopulation of the infarct scar are integral components of wound healing following myocardial infarction (MI; Refs. 3, 8, 14). Conventional chemotactic recognition and migration of myofibroblasts are recognized as essential processes for their sequestration to the site(s) of tissue injury (8). Until recently, the infarct scar in mammalian heart was believed to function as inert tissue, the main role of which was to prevent tissue rupture through restoration of structural integrity and tensile strength to the infarct zone (55). Myofibroblasts contribute significantly to scar formation and can be identified in the infarct site soon after influx of inflammatory cells (1, 58). Myofibroblasts within the MI scar are differentiated, specialized contractile fibroblasts that exhibit smooth muscle cell (SMC)-like features characterized by the expression of embryonic smooth muscle myosin (SMemb; Ref. 13), α-smooth muscle actin, and the ability to secrete multiple collagen types including I and III (1).
Increased cell density within the scar region, regardless of cell type, is necessary for improved left ventricular function (29, 30, 47, 57). While various cytokines exert chemotactic properties, the specific mechanism(s) governing chemotaxis of myofibroblasts in wound healing, e.g., in repopulation of the infarct scar in cardiac tissue, remains incompletely understood. However, it is known that in vivo wound healing is associated with the expression of multiple cytokines at the site of damage (15).
Cardiotrophin-1 (CT-1) is a member of the IL-6 superfamily of cytokines and is now known to be a cardioprotective agent (18, 32, 42). CT-1 is expressed in the myocardial infarct scar and has been implicated in post-MI wound healing through induction of proliferation of myofibroblasts as well as sarcomeric hypertrophy in myocytes. Recently, it has been shown that CT-1 stimulates protein synthesis, while inhibiting collagen synthesis, in isolated myofibroblasts (15, 38). However, the role of CT-1 in cardiac myofibroblast migration has not been well studied.
In mammalian myocardium, in vivo wound healing is mediated by de novo cell migration to the site of injury (15). Although there are some differences among different cell types, the migratory phenotype is characterized by the following: 1) front vs. rear cellular asymmetry, 2) development of membrane extensions or pseudopodia, 3) enhanced ability to attach to selected substrates, 4) active contractile force generation, and 5) detachment of the rear end of each cell (28). These coordinated and sequential processes occur in response to activation of ligand-dependent signaling pathways and require interactions with the extracellular matrix (ECM) through focal adhesions. Two types of force must be generated for cell movement to occur: an active extension resulting in the formation of membrane protrusions (pseudopodia, lamellipodia, and filopodia) and contractile force that moves the cell body over the substratum (28, 36). The active extension process is thought to occur via actin polymerization and cytoskeletal reorganization that is independent of myosin. In contrast, the force generation is accomplished by “signaling-dependent” activation of myosin motors (28, 36).
In smooth muscle, the activity of myosin motors is modulated by the phosphorylation of regulatory myosin light chains (rMLC), as a result of targeting myosin light chain kinase (MLCK; Ref. 1) and phosphatase (41) activities. The phosphorylation level of rMLC regulates activation of the myosin ATPase by actin, which results in cross-bridge cycling and contractile activity. Although there is relatively new evidence suggesting that MLCK can be activated by Src kinases (2, 17), the main physiologic regulator of MLCK is elevated intracellular calcium ([Ca2+]i). Ca2+ binds to calmodulin, which in turn binds to, and selectively activates, MLCK (41).
In most mammalian cells, [Ca2+]i is an important second messenger. [Ca2+]i regulates a large number of physiologic functions including cellular motility (44, 45, 62). [Ca2+]i levels are controlled by influx of Ca2+ and its mobilization from intracellular stores, mainly from the sarcoplasmic and endoplasmic reticulum (37). In a number of cell types that do not express voltage-dependent Ca2+ channels, Ca2+ influx is regulated indirectly. Initially, K+ channel activation results in membrane hyperpolarization, which increases the driving force (electrochemical gradient) for Ca2+ and in this way enhances Ca2+ influx (44, 45, 62). Often, this Ca2+ influx is mediated by TRP channels (canonical transient receptor potential-1), some of which are depletion-activated channels in nonexcitable cells (45). On the other hand, we (43) have recently demonstrated the robust expression of L-type Ca channels in primary cardiac myofibroblasts, and these channels are known to activate in response to membrane depolarization.
While ligand-based Ca2+ signaling may be important in fibroblast-mediated wound repair, the mechanisms supporting transient changes in [Ca2+]i are unknown. At present, little information is available to explain the specific mechanisms governing myofibroblast migration; however, it is known that CT-1 levels are rapidly increased within the ischemic and infarcted myocardium (14). We have carried out an in vitro study of the role of CT-1 in the induction of migration in cardiac myofibroblasts. The present experimental results extend previous work on the effects of CT-1 (14–16). Our findings characterize this cytokine as a chemoattractant for adult rat ventricular myofibroblasts by elucidating a number of important steps in this chemotaxis: 1) Chemotaxis is mediated through a Janus kinase 2 (JAK2), tyrphostin AG490-sensitive mechanism. 2) Application of CT-1 induces phosphorylation of myosin light chain 2 (MLC2), which can be blunted through sequestration of Ca2+. 3) CT-1 results in a rapid depolarization followed by prolonged hyperpolarization, and enhanced [Ca2+]i is observed during the later phase. 4) Gadolinium (Gd3+) treatment of CT-1-stimulated cells markedly reduces migration. Our results support the hypothesis that CT-1 mediates and enhances chemotaxis in ventricular myofibroblasts by augmenting [Ca2+]i levels, which promote MLCK-mediated actin-myosin crossbridge cycling.
MATERIALS AND METHODS
Isolation and characterization of cardiac myofibroblasts.
The investigation conforms with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85–23, revised 1996) as administered by the Canadian Council for Animal Care and was approved by the University of Manitoba Animal Care Committee. Ventricular fibroblasts were isolated from the hearts of 150- to 200-g adult male Sprague-Dawley rats as previously described (20). Briefly, hearts were excised and Langendorff perfused at 37°C with Joklik's medium containing 0.1% collagenase (Worthington Biochemical, Lakewood, NJ) for 20–25 min. Collagenase was then neutralized by addition of an equal volume of DMEM/F-12 medium containing 10% FBS, and liberated cells were collected by centrifugation at 500 g for 5 min. Cells were resuspended in fresh DMEM/F-12 containing 10% FBS and plated on 75-cm2 culture flasks at 37°C and equilibrated with 5% CO2 for 3 h. Nonadherent cells (myocytes) were removed by changing the culture media, and adherent cells (mainly fibroblasts) were incubated in DMEM/F-12 containing 10% FBS, 100 U/ml penicillin, 100 μg/ml streptomycin, and 100 μM ascorbate. We (15, 50) and others (34) have previously described the differentiation process of myofibroblasts from adult primary fibroblasts when plated at low density as described above. The myofibroblasts used for these experiments were collected after the first passage (P1). The purity of this cell population was ∼95%, based on routine phenotyping methods (as previously described) (39). First passage (P1) cells were observed to express α smooth-muscle actin and the embryonic form of smooth muscle myosin heavy chain (SMemb) as they are known to do in vivo in infarcted rat heart (12, 50).
At the start of each experiment, cells were rendered quiescent by incubation in serum-free medium for 24 h and then stimulated with CT-1 (10 ng/ml = 0.37 nM unless otherwise stated) for specified times before being lysed in RIPA buffer (150 mM NaCl, 1% Triton X-100, 0.5% deoxycholate, 0.1% SDS, and 50 mM Tris) using protease inhibitor cocktail (Sigma-Aldrich, Oakville, Ontario, Canada), 10 mM NaF, 1 mM Na3VO4, and 1 mM EGTA.
Western blot analysis.
Protein concentrations of cell lysates were determined by the BCA method (53). Proteins were separated by 8–12% SDS-PAGE and transferred to PVDF membrane (Roche, Indianapolis, IN) for 1 h at 300 mA for proteins with low molecular masses or for 2 h at 500 mA for proteins >100 kDa. Membranes were blocked with 5% nonfat skim milk in TBS containing 0.2% Tween 20 (TBS-T) or in 3% BSA/TBS-T for phosphorylated antigens. Proteins were visualized using ECL Plus (Amersham) after being probed with primary and secondary antibodies. Band intensity was quantified using a CCD camera configured as an imaging densitometer (GS670; Bio-Rad Laboratories, Mississauga, Ontario, Canada).
Migration of myofibroblasts.
Migration of myofibroblasts was measured in two different ways: Boyden chamber (Neuro Probe Gaithersburg, MD) assay (48) and the Transwell assay (Costar, Cambridge MA). For the Boyden chamber assay, chemoattractants were diluted in DMEM/F-12 and loaded into the lower wells at specified concentrations. A polycarbonate membrane (5-μm pore) insert was placed over the wells, and 55-μl cell suspension in DMEM/F-12 supplemented with 100 μM ascorbate was loaded into each well (1,000 cells/mm2). The chamber was incubated overnight in 5% CO2 and 100% humidity at 37°C. Myofibroblasts that had migrated through the membrane and become adherent to the lower surface were fixed with methanol and stained with Diff-Quick (Dade Behring, Düdingen, Switzerland). Cell migration was determined by counting the number of cells per high power field.
Chemotaxis of rat cardiac myofibroblasts was also determined with the Transwell system (Costar). This procedure was carried out as previously described with minor alterations (23, 26). In the lower chambers of sixwell Transwell plates, CT-1 was diluted to 10 ng/ml (0.37 nM) in serum free DMEM/F-12. P1 myofibroblasts were passaged, counted (2 × 105 cells/well), and plated directly onto inserts that were superimposed on, and separated from, the lower chambers by a polycarbonate membrane (with 8-μm pores). A stock of 5 mM GdCl3 (Sigma) in PBS was diluted at specified concentrations in serum-free media and loaded into the inserts containing cells. Plates were placed in an incubator containing 5% CO2-95% O2 and 100% humidity at 37°C. After a 24-h incubation period, cells had migrated through the membrane toward the chemoattractant and adhered to the underlying membrane on the bottom of the lower well. Media were carefully aspirated, and both inserts and wells were washed once with 1.5 ml PBS, followed by addition of 1.5 ml trypsin to the lower well. The plates were incubated for 5 min, gently agitated to detach adherent cells and then neutralized with 1.5 ml DMEM/F-12 10% FBS. The insert was removed, and a filtered PBS dilution of the cell suspension was counted with a Model ZM Coulter cell counter (Beckmen Coulter, Fullerton, CA). Measurements for each of the experimental groups were performed in duplicate.
Change in membrane potential and [Ca2+]i.
Myofibroblast membrane potential was monitored using the voltage-sensitive dye DiBAC4(3) (Molecular Probes, Eugene, OR). DiBAC4(3) is a potentiometric bisoxonol dye that partitions into the cell membrane as a function of membrane potential. It exhibits increased fluorescence upon depolarization and decreased fluorescence in response to hyperpolarization (10). To make these measurements, first passage myofibroblasts were plated onto glass coverslips, allowed to attach, and then incubated in 2 ml of control solution containing the following: 1 μM DiBAC4(3), 140 mM NaCl, 5 mM KCl, 10 mM HEPES, 2.0 mM CaCl2, 1.4 mM MgCl2, and 10 mM glucose for 20 min. Stained cells were visualized individually at ×400 on an inverted microscope (Nikon Canada). DiBAC4(3) was excited at a 470-nm wavelength, and emitted light was collected at 525 nm using a photometer (Photon Technology International, Lawrenceville, NJ). Fluorescence intensity was normalized to the baseline. All data are expressed as percent change, relative to baseline. According to the specifications of the dye, a 1% change in fluorescence intensity is equivalent to a 1-mV change in membrane potential (64). To confirm that these fluorescence intensity changes in response to changes in membrane potential were as expected, the cells were superfused with Tyrode's solution containing 1 μM DiBAC4(3) and either 1.5 or 15 mM KCl. In each such test experiment, DiBAC4(3) fluorescence decreased or increased significantly in concert with KCl-induced hyper- and depolarization of the cells, respectively (4). In the CT-1 studies, CT-1 was diluted in control solution containing 1 μM DiBAC4(3). After a stable baseline fluorescence signal was obtained, a 2-μl CT-1 stock solution was added to the recording chamber and fluorescence was monitored for an additional 10 min.
[Ca2+]i was measured using the dye fluo 3-AM, which was dissolved in DMSO at a concentration of 5 mM. Cardiac myofibroblasts were loaded with fluo 3 AM [1 μM in HEPES-buffered saline (HBS)] for 30 min at room temperature. Fluo 3-AM was then removed and replaced with fresh HBS. These experiments were done at room temperature and typically lasted 30 min. Fluo 3 was excited at 485 nm, and its fluorescence intensity was recorded at 526 nm using a photometer (Photon Technologies International, Lawrenceville, NJ). The raw data were normalized to the baseline fluorescence intensity. These data are also expressed as percent change relative to baseline.
Cell culture reagents were purchased from GIBCO (Burlington, ON, Canada) unless otherwise specified. Recombinant human CT-1 and mouse monoclonal CT-1 antibody were purchased from R&D Systems (Minneapolis, MN). Polyclonal antibodies against MLC2 (pS20) were from Abcam (Cambridge, UK), and goat anti-rabbit HRP-linked secondary antibody was purchased from Cell Signaling (New England Biolabs, Mississauga, Ontario, Canada). The antibody used to detect actin was from Santa Cruz Biotechnology (Santa Cruz, CA). Biotinylated secondary antibodies and streptavidin FITC were from Amersham-Pharmacia (Baie d'Urfe, Quebec, Canada). Inhibitors of selected targets in cell signaling pathways (e.g., ML-7, W7, and AG490) were purchased from Calbiochem (San Diego, CA). The Ca2+ buffer BAPTA was purchased from Sigma-Aldrich. GdCl3 and other laboratory grade reagents were also purchased from Sigma-Aldrich. HBS (pH 7.0) solution (25 ml of 200 mM HEPES, 11.8 ml of 100 mM NaOH, and H20 to 100 ml) was made in the laboratory and sterilized by passage through a 0.22-μM filter. Fluo 3-AM and DiBAC4(3) were purchased from Molecular Probes B-438 (Burlington, Ontario, Canada).
Data are expressed as means ± SE. Means between control and test conditions were compared using one-way ANOVA followed by either Dunnett's or Bonferonni's post hoc analysis. Significant differences are indicated by P ≤ 0.05 (n = sample size).
CT-1 induces cardiac myofibroblast migration.
The results from this study confirm and significantly extend our earlier work (16) that demonstrated that CT-1 can function as a chemokine for primary ventricular fibroblasts (Fig. 1A). We previously observed a dose-dependent migratory response to CT-1 in fibroblasts (16). Results in this study confirm this in both fibroblasts (P0) and include new findings in myofibroblast (P1) cultures (data not shown) that provide insights into the signaling mechanisms involved in CT-1-induced migration. A primary objective was to determine how the chemotactic response to CT-1 is transduced in cardiac myofibroblasts. Accordingly, we administered AG490 (25 μM), an inhibitor of JAK 2, a key component of a known intracellular signaling pathway of CT-1 signaling. AG490 blocked CT-1-induced cardiac fibroblast migration (Fig. 1B).
We also sought to evaluate whether CT-1-induced migration is dependent on MLCK activity. CT-1-induced migration was inhibited by coincubation with ML-7 (24 h at 20 μM), an inhibitor of MLCK (Fig. 1C). MLCK would be expected to regulate myosin motors through phosphorylation of regulatory light chains, which are required for effective cell movement. We therefore examined the effect of CT-1 on the phosphorylation status of MLC. CT-1-induced phosphorylation of MLC at serine 20 in a time-dependent manner (Fig. 2). To clarify the mechanism of CT-1-induced MLC phosphorylation, we sequentially examined components of the MLCK signaling cascade. MLC phosphorylation could be prevented by incubation in Ca2+ free conditions or by buffering of [Ca2+]i using BAPTA-AM (30 min at 5 μM). Phosphorylation could also be attenuated by coincubation with ML-7 (MLCK inhibitor for 30 min at 20 μM) and W7 (calmodulin inhibitor for 30 min at 100 μM; Fig. 3).
A third component of the study was to determine the mechanism of CT-1-induced MLC phosphorylation. In nonexcitable cells, Ca2+ flux is coupled to membrane potential through nonselective cation channels. Hyperpolarization, through activation of K+ channels, may conceivably influence Ca2+ influx into the cell, and we hypothesized that inhibition of K+ channels would impact on cell migration. Indeed, myofibroblast migration was significantly reduced following application of 4-aminopyridine (4-AP; 100 μM) or tetraethylammonium (TEA; 1 mM) for 24 h (Fig. 4). To establish the link between CT-1 application and altered membrane potential, we employed DiBAC4(3) potentiometric dye. Following application of the dye, fluorescence intensity provides an approximation of cell membrane potential (Em). Measurement of DiBAC4(3) fluorescence intensity yielded a consistent pattern of CT-1-induced responses (Fig. 5). After application of CT-1, the myofibroblast first depolarized for a short period of time, followed by significant, more prolonged hyperpolarization (Fig. 5). This confirmed that application of CT-1 induces first a depolarization, followed by a prolonged hyperpolarization phase wherein the calcium influx was delayed after depolarization and was attenuated at the same time that hyperpolarization occurred. We hypothesize that the elevation of Ca2+ in the cytoplasm is required for CT-1-induced activation of MLCK with subsequent MLC phosphorylation and chemotaxis in primary cardiac myofibroblasts. As we have previously demonstrated that L-type calcium channels are expressed in primary myofibroblasts, we suggest that these channels activate after CT-1 depolarization.
To confirm the change in [Ca2+]i implied by our experiments to this point, we employed the calcimetric dye fluo 3. As assessed by fluo 3 fluorescence intensity, CT-1 produced an increase in [Ca2+]i (Fig. 6). This was temporally associated with the CT-1-induced depolarization seen in the DiBAC4(3) experiments. Two mechanisms that can explain enhanced [Ca2+]i is through the activation of transient receptor potential/nonselective cation channels, as occurs in lysophophatidylcholine-induced monocyte chemotaxis (51) or via activation of L-type Ca2+ channels, as Gd3+ is nonselective. The TRP/nonselective cationic channel and L-type Ca2+ channel inhibitor Gd3+ at concentrations of 10 and 20 μM attenuated the effects of CT-1 on myofibroblast migration (Fig. 7). This indicates that cation channels participate in CT-1-induced cell movement (43).
CT-1 is expressed in the post-MI heart (15, 16) and has been shown to exert cardioprotective effects in response to myocardial injury (18, 27, 32). We (16) previously reported that CT-1 can function as a chemokine for cardiac fibroblasts. The present study extends these findings by demonstrating a critical role for MLCK. CT-1 treatment of myofibroblasts was associated with increased phosphorylation of MLC at a specific location (MLC2-phospho-S20). This phosphorylation event is normally associated with increased actin-myosin cross-bridge cycling and cellular contractility as a result of enhanced myosin heavy chain ATPase activity (21). In the current study, we demonstrated that enhanced MLC phosphorylation was associated with enhanced cell motility. In addition, we showed that Ca2+ plays a critical role in CT-1-induced activation of MLCK. Indeed, CT-1 treatment resulted in a transient elevation of [Ca2+]i as reflected by fluo 3 fluorescence. It is likely that the increase in MLC phosphorylation results from CT-1-mediated change in [Ca2+]i (60) as MLC phosphorylation could be attenuated by incubation in Ca2+-free conditions or chelation of intracellular Ca2+. We observed that MLC phosphorylation was markedly inhibited in the presence of BAPTA-AM, possibly through a Ca2+-calmodulin-dependent mechanism. This finding supports the suggestion that myofibroblast MLC may be heavily phosphorylated in basal conditions. As ML-7 did not inhibit phosphorylation to the extent as did calcium chelation, our results also support the suggestion that kinases other than MLCK may function to phosphorylate MLC in cardiac myofibroblasts. In this regard, Ding et al. (8a) have demonstrated that a Ca2+-calmodulin-activated kinase that phosphorylates MLC is present in mouse heart, and despite the wealth of information about skeletal and smooth muscles, the identity of kinases in the heart that are important for phosphorylation of MLC is not well characterized. Nonetheless, our new results provide evidence for the role of Ca2+ as a second messenger in CT-1 signaling, with respect to MLCK activation and cell migration.
Myofibroblast migration is an important component of cardiac wound healing in post-MI hearts (8). Our findings support the hypothesis that CT-1 may facilitate the induction of myocardial wound healing and reveal a number of important steps in this complex signaling system. Inhibition of JAK2 with AG490 (35) attenuated CT-1-induced chemotactic effects. Previously published work (14) has provided strong evidence for the involvement of the JAK/STAT pathway in CT-1 signaling. JAK2 is known to activate multiple downstream signaling pathways, including the MAPK pathway, Src, phosphatidylinositol 3-kinase, and STATs (46). We also provide new evidence that CT-1-induced JAK activation induces MLCK activity (MLC phosphorylation) and that [Ca2+]i acts as a second messenger in this process.
A rise in [Ca2+]i is a major trigger for cell contraction and an important stimulus for cell migration. Removal of extracellular Ca2+ and/or the depletion of Ca2+ stores can significantly reduce cell migration in response to autocrine/paracrine factors and chemoattractants (62). Others (24) have reported inhibition of migration in nonexcitable intestinal epithelial cells following removal of extracellular Ca2+. In both excitable and nonexcitable cells, membrane potential (Em) is a major determinant of the driving force for Ca2+ influx (62). Through their role in regulating Em, voltage-gated K+ channels have previously been shown to be involved in the regulation of intestinal epithelial cell migration after wounding (44, 45). We provide new evidence that supports a somewhat similar functional relationship among changes in Em, [Ca2+]i, and migration of cardiac myofibroblasts in response to CT-1 stimulation. Our data suggest that CT-1 induces alterations in membrane potential, which culminates in a rise in [Ca2+]i with subsequent activation of MLCK and cell migration. However, the exact mechanism of Ca2+ influx remains to be elucidated.
Very little is known about the physiology (or electrophysiology) of cardiac myofibroblasts that reside in the healed infarct scar in vivo. Although scar formation is completed at 3 wk post-MI (11), it is known that myofibroblasts are the dominant cell type in the infarct scar (40, 54, 56) and remain active even 8 wk post-MI (6, 24, 40). Indeed, collagen content (sourced exclusively from cardiac myofibroblasts) in the infarct scar may increase over months (11, 25). These myofibroblasts may persist in infarct scar in post-MI patients for many years, even decades, after the initial injury (6, 7, 63). The scarification of the infarct is required to maintain ventricular integrity. Thus limited fibrosis in the healing infarct scar may help to preserve ventricular function, as the new scar tissue selectively resists circumferential deformation (22). Myofibroblasts produce isometric tension within granulation tissue in vivo and in cultures (52, 59). Tension is exerted at the level of focal adhesions, which connect cells to matrix (9). Myofibroblast contraction confers matrix distortion in wound healing (59); the tensile force opposes retractile forces and promotes scar contraction in the post-MI heart (19, 31). The ionic currents that regulate episodic contraction of these cells are not well described. Previous work by members of our group and that of others has demonstrated that cardiac myofibroblasts are electrically active in vitro (4, 49) and that cardiac myofibroblasts may electrically couple to cardiomyocytes, and share depolarization patterns (33). Further, Zhang et al. (65) have demonstrated that cardiac fibroblasts upregulate connexin 43 in post-MI heart in direct contrast to neighboring cardiomyocytes. This expression allows for maintenance of electrical and metabolic coupling during a time characterized by compromised intercellular communication in cardiomyocytes from damaged hearts. While we have demonstrated that K+ currents may regulate resting membrane potential, proliferation, and contractile responses in ventricular myofibroblasts (4), little if any published data are available to describe myofibroblast responses in vivo to compounds that may disrupt K+ currents such as 4-AP or TEA. Further, there is a paucity of in vivo studies to address the effects of W7, ML-7, or gadolinium on fibroblasts in the literature. Nonetheless, the undertaking of this work in the future is warranted in lieu of our current findings and the recent discovery of electrical connectivity displayed by cardiac fibroblasts and myofibroblasts.
We (4) have previously documented expression of several K+-channels in cardiac myofibroblasts, including Kv1.6, Kir2.1, and Kir6.1. Others (61) have shown expression of Kv 1.2, Kv 1.4, Kv 1.5, and Kv2.1. Consistent with this expression, these cells were found to be responsive to alteration of extracellular K+ concentration with respect to Em as well as proliferation and collagen gel contraction (4). In the current study, we demonstrate that application of a signaling mediator, CT-1, causes changes in Em accompanied by a rise in [Ca2+]i with subsequent phosphorylation of MLC and induction of cell migration. Nonselective potassium channel inhibitors 4-AP and TEA were effective in attenuating the cellular migratory response to CT-1, as was the nonselective cationic channel inhibitor Gd3+. Our current data highlighting changes in membrane depolarization and hyperpolarization and the rise in [Ca2+]i allow us to comment on the coordination of these events. The elevation of [Ca2+]i begins at ∼400 s after CT-1 is added, and is complete after ∼1,100 s, whereas membrane potential at 400 s is nearing the end of depolarization. Further, at 1,100 s, membrane potential is at the peak of hyperpolarization. Our interpretation of these events is that transmembrane Ca2+ influx occurs after initial depolarization and that the extended hyperpolarization phase is associated with the completion of Ca2+ influx after CT-1 stimulation. Thus these findings support the hypothesis that CT-1 treatment activates a K+-channel with augmented Ca2+ influx after depolarization followed by subsequent membrane hyperpolarization/abrogation of as well as net activation of MLCK. As we have documented robust expression of L-type calcium channel expression in cardiac myofibroblasts (43), we suggest that the depolarization of treated cells is associated with activation of these channels. CT-1 application results in a biphasic pattern of changes in membrane potential in which the second or hyperpolarizing response is associated with a diminution of [Ca2+]i. Although the rise in [Ca2+]i is modest, it would appear to be sufficient to induce a level of phosphorylation of MLC that is nearly equivalent to that induced by the calcium ionophore ionomycin (data not shown). The mechanisms of CT-1-induced Em changes are likely multifactorial, with participation of both TRP and K+ channels. The involvement of nonselective cation channels is implicated as a mechanism for Ca2+ influx since an inhibition of CT-1-induced cell migration was observed following administration of Gd3+. Attempts to record CT-1-induced membrane currents by perforated patch-clamp technique in rat ventricular myofibroblasts failed to yield consistent results in contrast to the data consistently recorded with DiBAC4(3) potentiometric dye. Further work is required to elucidate the precise mechanism of CT-1-induced Ca2+ entry in cardiac myofibroblasts, including the potential role of ion channel phosphorylation, secondary to CT-1 signaling. Nevertheless, we have demonstrated the central role of [Ca2+]i and MLCK in CT-1-induced cell migration and propose a signaling cascade with these as the central players (Fig. 8).
In summary, results from this study substantiate the hypothesis that CT-1 is responsible for activation of multiple parallel signaling pathways in cardiac myofibroblasts, including alterations in Em and [Ca2+]i. Our findings provide further insights into the signaling mechanisms responsible for the effects of a novel agonist, CT-1, on a novel cell type, the cardiac myofibroblast.
This current project was funded with an operating grant from the Canadian Institutes of Health Research (I. M. C. Dixon). D. H. Freed received postdoctoral fellow salary support from the IMPACT/Canadian Institutes for Health Research program. A. L. Dangerfield and J. E. Raizman were recipients of St. Boniface Hospital and Research Foundation Studentships. I. M. C. Dixon was named for the Heart and Stroke Foundation of Manitoba R. E. Beamish Memorial Award (2007–2008) from which funding was applied to complete this research. Additional personnel support was provided by funds from the Canadian Institutes for Health Research and by the St. Boniface General Hospital and Research Foundation.
All of the authors named in this article have no conflict of interest to declare, including shareholding in or receipt of a grant or consultancy fee from a company whose product is featured in the submitted manuscript or that manufactures a competing product.
We thank Dr. Wayne Giles (Univ. of Calgary) and Jared Davies for helpful discussions and comprehensive critique of the manuscript and kind advice in completion of this work.
- Copyright © 2011 the American Physiological Society