Address for Correspondence Justin A. MacDonald, Department of Biochemistry & Molecular Biology, Faculty of Medicine, University of Calgary, 3280 Hospital Drive N.W., Calgary, Alberta, T2N 4Z6, Canada. Tel: +403 210 8433; fax: +403 270 2211; e-mail: firstname.lastname@example.org
Background Colonic smooth muscle contractility is altered in colitis, and several protein kinase pathways can mediate colonic smooth muscle contraction. In the present study, we investigated whether protein kinase C (PKC) pathways also play a role in colonic hypercontractility observed during TH2 colitis in BALB/c mice.
Methods Colitis was induced in BALB/c mice by provision of 5% dextran sodium sulfate (DSS) for 7 days. Changes in smooth muscle contractility were examined using dissected circular smooth muscle preparations from the distal colon. The contribution of conventional and novel PKC isozymes to the hypercontractile response was examined with pharmacological PKC inhibitors. Western blot analyses were used to examine protein expression and phosphorylation changes.
Key Results Colonic smooth muscle was associated with inflammation-induced hypercontractility and altered PKC expression. Carbachol-induced peak (phasic) and sustained (tonic) contractions were increased. Chelerythrine was the most effective PKC inhibitor of both phasic and tonic contractions. There was no general difference in the percent contribution of conventional and novel PKC isozymes toward the DSS-induced hypercontractility, but inhibition of sustained force with GF109203x was higher for inflamed muscle. The CPI-17 phosphorylation was equally suppressed in both normal and DSS conditions by Gö6976 and chelerythrine, but only for the phasic component of contraction.
Conclusions & Inferences The outcomes suggest that both conventional and novel PKC isozymes contribute to the phasic and tonic contractile components of BALB/c colonic circular smooth muscle under normal conditions, with novel PKC isozymes having a greater contribution to the tonic contraction. However, no effect of inflammation was observed on the relative contribution of PKC and CPI-17 toward the observed hypercontractility.
The coordinated regulation of contraction is a key property of gastrointestinal (GI) smooth muscle, which when functioning normally, contributes to general health and wellness, but when dysfunctional is associated with morbidity and mortality.1,2 In overt inflammatory conditions of the bowel, such as Crohn’s disease and ulcerative colitis (i.e., IBD), there have been longstanding observations of altered motility and impaired function of the GI smooth muscle.3,4 Alterations in GI motility with resultant changes in transit can contribute to the abdominal pain, intestinal cramping, and diarrhea characteristically associated with intestinal inflammation. Impairments in GI motility are a common feature of a variety of important disease manifestations of varying etiologies.5,6 However, a central mechanistic feature of GI dysmotility is an alteration in the contractile processes that occur at the level of the GI smooth muscle.
Smooth muscle contraction is a highly regulated process but is ultimately governed by the phosphorylation of the regulatory light chain (LC20) of myosin II at Ser-19 (reviewed in7–9). Increase in intracellular [Ca2+] leads to the activation of myosin light chain kinase (MLCK) which is responsible for LC20 phosphorylation and force generation. Smooth muscle myosin light chain phosphatase (MLCP) is responsible for the dephosphorylation of LC20 resulting in relaxation of smooth muscle; however, it is the balance between MLCK and MLCP activities that dictates the contractile activity of the tissue. Indeed, MLCP functions independently of Ca2+and can be regulated by G-protein-coupled signaling pathways. Inhibition of MLCP leads to an increase in both LC20 phosphorylation and contractile force development in smooth muscle without any changes in [Ca2+]. This enhancement of the contractile response to Ca2+ is commonly referred to as ‘calcium sensitization’, and the phenomenon can be mediated by phosphorylation of a protein kinase C (PKC)-potentiated phosphatase inhibitor protein-17 kDa (CPI-17).7
The PKC/CPI-17 pathway is an important regulator of intestinal smooth muscle contractility under normal conditions, and changes in PKC signaling can contribute to motility dysfunction under pro-inflammatory insults.10,11 The phosphorylation of CPI-17 at the Thr-38 site can potentiate its inhibition of MLCP,12 and this regulatory mechanism has been extensively studied in vascular smooth muscle tissues.13 Integrin-linked kinase (ILK),14 zipper-interacting protein kinase (ZIPK),15,16 Rho-associated protein kinase (ROK),17 and PKC12 have also been shown to induce this phosphorylation in vitro. However, only PKC and to a lesser extent ROK have been demonstrated to phosphorylate CPI-17 in vivo.
In a previous study,18 the administration of dextran sodium sulfate (DSS) induced a TH2 cytokine-mediated colitis in BALB/c mice. These animals exhibited increased carbachol (CCh)-induced contraction and increased activation of mitogen-activated protein kinase (MAPK) pathways (i.e., ERK and p38MAPK). In gastrointestinal smooth muscle, CCh is thought to primarily activate muscarinic M2 and M3 receptors that signal through G proteins to regulate contractile force (reviewed in9). The M3 receptors contribute to regulate cellular Ca2+ levels via the canonical excitation–contraction coupling pathway that signals by activation of phospholipase Cβ to generate inositol 1,4,5-trisphosphate (IP3) and diacylglycerol (DAG). While IP3 is the second messenger responsible for Ca2+ release from intracellular stores, DAG is responsible for PKC activation. In light of the important contributions of PKC and CPI-17 to smooth muscle contraction and previous results that suggest selective modulation of PKC and CPI-17 in GI smooth muscle by proinflammatory effectors (i.e., TH1 cytokines: IL-1β, TNF-α), we hypothesized that PKC/CPI-17 signaling may also have a significant role in the hypercontractile phenotype associated with enhanced TH2 cytokine (i.e., IL-4, IL-13) production. Our results describe PKC expression in colonic circular smooth muscle and examine the roles of both conventional and novel PKC isoforms, and the downstream target CPI-17 in colonic contractility under both normal and DSS-mediated colitis in BALB/c mice.
Materials and methods
Materials and chemicals
All chemicals were reagent grade unless otherwise indicated. Chelerythrine, GF109203x and β-escin were obtained from Sigma (St. Louis, MO, USA). The DSS (MW, 36 000–50 000) was from MP Biomedicals (Solon, OH, USA). The Gö6976 was purchased from Calbiochem (San Diego, CA, USA). Polyclonal antibodies specific for CPI-17, LC20 and MYPT1 were purchased from Upstate Biotechnology (Charlottesville, VA, USA), and antibodies to PKC isoforms were purchased from Santa Cruz Biotechnology (PKCα, PKCβI and PKCβII; Santa Cruz, CA, USA) or from Life Technologies (PKCδ and PKCε; Carlsbad, CA, USA). Phos-tag acrylamide was obtained from the NARD Institute (Amagasaki, Japan).
Induction and assessment of colitis
All animal experimentation was approved by the University of Calgary Animal Care and Use Committee and followed the guidelines established by the Canadian Council on Animal Care. Colonic inflammation was induced in female BALB/c mice (19–20 g) by administering 5% (w/v) DSS in the drinking water.18 Normal and DSS-treated animals were sacrificed under 3% isofluorane inhalation anesthesia after 7 days exposure to normal or DSS drinking water, respectively.
Expression of protein kinases in colonic circular smooth muscle
The expression profile of 76 different protein kinases (including 8 PKC isoforms) in normal colonic smooth muscle was assessed with the Kinexus KPKS 1.2 kinase screening service (Kinexus Bioinformatics, Vancouver, BC). Segments of distal colon, located adjacent to sections used for force measurements, were removed, and the circular smooth muscle layer was dissected and flash frozen in liquid N2. Tissue samples (approximately 20 mg) were homogenized in 1.0 mL of Kinexus Cell Lysis Buffer containing 20 mmol L−1 MOPS (pH 7.0), 2 mmol L−1 EGTA, 5 mmol L−1 EDTA, 30 mmol L−1 sodium fluoride, 60 mmol L−1β-glycerophosphate, 20 mmol L−1 sodium pyrophosphate, 1 mmol L−1 sodium orthovanadate, 1% Triton X-100, and 1 mmol L−1 dithiothreitol (DTT) with protease inhibitor cocktail (Roche, Indianapolis, IN, USA). The homogenates were sonicated on ice to shear nuclear DNA and then centrifuged (90 000 ×g, 30 min, 4 °C). The soluble protein fractions were collected, and the final protein concentrations adjusted to 0.75 mg mL−1 with SDS-PAGE sample buffer. Samples were delivered on dry-ice to Kinexus Bioinformatics for protein kinase screening. In addition, samples were examined by Western blotting for PKC-α, -βI, -βII; -δ and -ε isoforms. Proteins from normal and DSS-treated mice were resolved on 10% SDS-PAGE gels and then transferred to polyvinylidene fluoride (PVDF) membrane (0.2 μm pore). The membranes were blocked with 5% (w/v) non-fat dry milk and then incubated with primary antibodies for PKC isoforms according to the manufacturer’s recommendations. The blots were developed with enhanced chemiluminescence (ECL) reagent (GE Healthcare, Piscataway, NJ, USA); the PKC bands were quantitated by densitometry and normalized against β-actin as a loading control.
Force measurement of colonic circular smooth muscle
Force measurements of colonic circular smooth muscle strips were performed as previously described.18 In brief, normal and DSS-treated mice were sacrificed, and the distal colon was removed. Circular smooth muscle sheets were dissected and cut into strips (250 μm × 2 mm). Strips were mounted onto a force transducer (AE801; SensoNor Inc, Hensen, Norway) and stretched in the circular axis until the resting force reached 0.1 mN. Strips were firstly equilibrated in normal extracellular solution (NES), then the contraction in response to a 118 mmol L−1 K+ extracellular solution (KES) was used to assess muscle quality. All contractile measurements were carried out at room temperature (23 °C) with a computerized data acquisition system (PowerLab/8SP data recording unit and Chart software; ADInstruments, Colorado Springs, CO, USA). To examine the contribution of PKC to contractile force, muscle strips were subjected to sequential treatment with carbachol (CCh, 10 μmol L−1) in NES. The contractile response to the first application of CCh was used as a reference for a subsequent contraction produced when a PKC inhibitor was included. Pharmacological inhibitors of PKC (Gö6976, 10 μmol L−1; chelerythrine, 10 μmol L−1 or GF109203x, 100 nmol L−1) were applied 10 min before the second administration of CCh.
Western blot analysis of CPI-17 and MYPT1
Colonic circular smooth muscle from normal or DSS-treated mice was flash frozen by immersion in a dry-ice/acetone solution containing 10% (w/v) trichloroacetic acid (TCA) and 10 mmol L−1 dithiothreitol (DTT) and lyophilized overnight. Muscle proteins were extracted in a buffer containing 1% SDS, 30 mmol L−1 Tris-HCl, pH 6.8, 12.5% (v/v) glycerol, and (p-amidinophenyl)methanesulfonyl fluoride (APMSF) with a glass-glass, hand-held homogenizer. For CPI-17 analysis, proteins were resolved on 15% SDS-PAGE gels and then transferred to PVDF membrane in a buffer containing 10 mmol L−1 cyclohexylaminopropane sulfonic acid (CAPS), pH 11, and 10% (v/v) methanol. The membranes were blocked with 5% (w/v) non-fat dry milk and then incubated with primary antibody (1 : 1000 dilution of total-CPI-17 and β-actin) in TBST containing 1% (w/v) non-fat dry milk. After washing, blots were incubated for 1 h with horseradish peroxidase-conjugated secondary antibody (1 : 40 000 dilution for total-CPI-17 and 1 : 2500 dilution for β-actin) in 1% (w/v) non-fat dry milk in TBST. Blots for total-CPI-17 and β-actin were developed with Supersignal West Femto-ECL reagent (Pierce, Rockford, IL, USA) and ECL reagent (GE Healthcare, Piscataway, NJ, USA), respectively. The bands were quantitated by densitometry, and the ratio of total-CPI-17 to β-actin was calculated. Western blot analysis of total MYPT1 expression was carried out as previously described.18,19
Measurement of LC20 and CPI-17 phosphorylation
The phosphorylation status of CPI-17 was examined with Phos-tag SDS-PAGE gels.20 Contractile responses were halted by immersion of muscle strips in a dry-ice/acetone solution containing 10% (w/v) TCA. The muscle strips were washed with a 10 mmol L−1 DTT/acetone solution and then lyophilized overnight. Phos-tag ligand (final concentration, 30 μmol L−1) and MnCl2 (final concentration, 60 μmol L−1) were added to the separating gel (11% polyacrylamide) before polymerization. After electrophoresis, gels were soaked in transfer buffer without methanol (10 mmol L−1 CAPS, 2 mmol L−1 EDTA, pH 11). Proteins were transferred to PVDF membranes as described previously. Western blotting was carried out with a total-CPI-17 polyclonal antibody that detects both phosphorylated and unphosphorylated protein. The bands were quantitated by densitometry and relative phosphorylation levels of CPI-17 were expressed as a function of the density of the total protein. As two isoforms of CPI-17 were detected in the colonic smooth muscle of BALB/c mice, the CPI-17 phosphorylation level (%) was calculated from the following equation:
where U and L indicate the upper and lower bands within the CPI-17 doublet and P0 and P1 indicate un- and mono-phosphorylated CPI-17, respectively. The LC20 phosphorylation was analyzed with Phos-tag SDS-PAGE gels as previously described.18
All data are expressed as the mean ± SE of the mean (SEM) with independent analysis of n = 4–8 separate mice. Student’s t-test (two-tailed, P <0.05) was used to identify statistically significant differences between normal and DSS-treatment. One-way analysis of variance (anova) followed by post hoc Student–Newman–Keuls test was used to identify statistically significant differences among PKC inhibitors on contraction and CPI-17 phosphorylation; P <0.05 was considered to be significant.
A hypercontractile response to carbachol is observed in inflamed colonic smooth muscle
Consistent with previous results,18 contractile responses to the muscarinic receptor agonist carbachol (CCh) were significantly increased in colonic circular smooth muscle (Fig. 1A) isolated from DSS-treated BALB/c mice. The absolute force obtained at the peak of CCh (10−5 mol L−1)-induced contractions was twofold higher for inflamed colonic circular smooth muscle strips (Fig. 1B), rising from 7.0 ± 0.3 to 13.8 ± 0.4 mN mm−2. The increased CCh-dependent contractile force observed with DSS-treatment trended toward an elevated LC20 phosphorylation in the phasic (peak) phase of contraction; however, the change did not reach statistical significance. For the phasic (peak) contraction, the extent of LC20 monophosphorylation in normal mice was determined to be 10.4 ± 3.4% (n = 3) of total LC20, and the amount of LC20 phosphorylation measured in DSS-treated muscle was 16.3 ± 3.7% (n = 4) of total LC20 (Fig. 1C). No measureable amounts of LC20 diphosphorylation were observed. The tonic, sustained component of the CCh-induced contractions obtained from DSS-treated mice was also significantly elevated over the controls (Fig. 1B), rising from 4.3 ± 0.2 to 7.2 ± 0.6 mN mm−2. The LC20 monophosphorylation levels measured for the tonic components were 6.7 ± 1.8% and 6.8 ± 1.2% in normal and DSS-treated mice, respectively (Fig. 1C). Again, no measureable amounts of LC20 diphosphorylation were observed.
Expression of protein kinases in normal murine colonic circular smooth muscle
Initially, we examined the expression profile of protein kinases in the circular colonic smooth muscle isolated from normal BALB/c mice (Fig. 2). A variety of protein kinases known to be important in gastrointestinal smooth muscle contraction (e.g., PAK1, ROKα, PKG1, PKAc, p38MAPK, and ERK1/29,10,18,19,21) were detected. The PKC is an important regulator of intestinal smooth muscle contractility under normal conditions,9,19,21,22 and changes in PKC signaling are thought to contribute to motility dysfunction under inflammatory insult.10 Protein kinase screening (Fig. 2) for conventional PKC isozyme family members detected PKCα and PKCβΙ, but not PKCγ. The expression of PKCβII was not examined by this method due to technical limitations. For the novel PKC family members, PKCδ, PKCε, and PKCθ were all detected. Furthermore, for the atypical PKC isozymes, the expression of PKCζ was observed.
Contribution of conventional and novel PKC isozymes to CCh-induced contraction in normal colonic circular smooth muscle
As a functional role for PKCα, PKCβΙ/βΙΙ, PKCδ, and PKCε in smooth muscle contraction had been previously demonstrated,23 any or all of these PKC isozymes could contribute to the normal contractile properties of the circular, colonic smooth muscle. So, the relative contributions of the conventional (PKCα and PKCβΙ/βΙΙ) and the novel (PKCδ and PKCε) PKC isozymes were examined during CCh-induced contraction of the circular smooth muscle of normal mouse colon. At the concentration employed, Gö6976 was considered to be selective for conventional PKCα and PKCβΙ/βΙΙ isoforms without effects on PKCδ or PKCε activities.24 In contrast, chelerythrine and GF109203x were considered to be broadly acting inhibitors of both conventional and novel PKC isozymes, including PKCα, PKCβΙ/βΙΙ, PKCδ, and PKCε.25,26 The peak force developed with CCh in normal colonic tissue (Fig. 3G) was reduced by pretreatment with Gö6976 (Fig. 3A; 49.6 ± 5.5%, n = 4), chelerythrine (Fig. 3B; 82.9 ± 8.6%, n = 4), and GF109203x (Fig. 3C; 52.9 ± 5.7%, n = 4). The inhibition of peak force by chelerythrine was significantly greater than that obtained with Gö6976 and GF109203x. Similar results were observed during the sustained phase of CCh-induced contraction (Fig. 3H), with a significant force reduction upon application of Gö6976 (34.9 ± 0.9%, n = 4), chelerythrine (74.4 ± 2.3%, n = 4), and GF109203x (51.2 ± 4.7%, n = 4). As with peak force, the inhibition of sustained force by chelerythrine was significantly greater than that obtained with Gö6976 and GF109203x. Taken together, the results suggest that both conventional and novel PKC isozymes contribute to both the initial peak and the sustained phase of CCh-induced contraction in colonic circular smooth muscle.
Contribution of conventional and novel PKC isozymes to potentiated carbachol-induced contraction in inflamed colonic circular smooth muscle
The impact of colitis on the relative contribution of conventional and novel PKC isozymes to CCh-induced contraction was also examined. For DSS-treated mice, there was a significant reduction in the peak force developed with CCh administration in the presence of Gö6976 (Fig. 3D; 35.5 ± 4.3%, n = 4), chelerythrine (Fig 3E; 73.9 ± 8.7%, n = 4), and GF109203x (Fig. 3F; 55.1 ± 3.6%, n = 4). There were also significant reductions in the sustained force developed with CCh administration in the presence of Gö6976 (Fig. 3D; 46.2 ± 4.2%, n = 4), chelerythrine (Fig. 3E; 83.9 ± 9.8%, n = 4), and GF109203x (Fig. 3F; 65.7 ± 2.8%, n = 4). As found under normal conditions, it appeared that both conventional and novel PKC isozymes contributed to the initial peak of CCh-induced contraction of inflamed colonic circular smooth muscle. While the absolute force reductions were found to be higher when PKC inhibitors were applied to inflamed smooth muscle (data not shown); the reduction in contractile force when taken as a percentage of the peak force was quite similar between normal and DSS animals (Fig. 3G). In this case, the only significant difference was an increase in the ability of GF109203x to attenuate the sustained phase of CCh-mediated contraction of inflamed colonic circular smooth muscle when compared with the normal contractile responses. Again, GF109203x is a broadly acting inhibitor of both conventional and novel PKC isozymes, including PKCα, PKCβΙ/βΙΙ, PKCδ, and PKCε,25,26 and so, we were unable to specify the specific PKC target involved.
The expression of PKC-βII and -ε, but not CPI-17 or MYPT1, in circular colonic smooth muscle is affected by colitis
To study the expression of the various conventional and novel PKC isozymes that contribute to smooth muscle contractility, Western blot immunoreactivity was determined in tissue homogenates prepared from dissected colonic circular smooth muscle isolated from normal and DSS-treated mice (Fig. 4A). Specific monoclonal antibodies were used, and immunoreactive bands were obtained at approximately 80 kDa (PKCα), 80 kDa (PKCβI), 85 kDa (PKCβII), 85 kDa (PKCδ), and 95 kDa (PKCε). Modest but significant changes in the expression of PKCβII and PKCε were observed in tissue isolated from DSS-treated mice. The PKCβII protein levels were down-regulated by 15%, while PKCε protein levels were up-regulated by 17%. Conversely, the expression of PKCα, PKCβI, and PKCδ did not change in circular colonic smooth muscle during the intestinal inflammation induced by DSS exposure.
The expression level of CPI-17 is a critical determinant of the extent of smooth muscle contraction induced through G-protein-coupled receptors.13 Changes in CPI-17 expression are known to occur under inflammatory conditions, such as colitis27,28 and asthma,29 and result in abnormal smooth muscle contraction. Therefore, we examined whether the expression of CPI-17 was altered during DSS-induced colitis in BALB/c mice. Two immunoreactive bands were detected in extracts of colonic circular smooth muscle tissue from mice following Western blot analysis with CPI-17 antibody (Fig. 4B). Treatment of samples with alkaline phosphatase (data not shown) did not influence the detection of a CPI-17 doublet by Western blotting. Furthermore, a single immunoreactive band was detected when the same antibody was used to assess the CPI-17 content of rat ileal, colonic and caudal arterial smooth muscle (data not shown). Thus, the data support the expression of two isoforms of CPI-17 in the colonic circular smooth muscle; however, no significant differences in CPI-17 expression were observed in inflamed relative to normal colonic circular smooth muscle. Likewise, no significant effect of inflammation was observed on the expression level of MYPT1 (Fig. 4C).
Both novel and conventional PKC signaling pathways contribute to CPI-17 phosphorylation during carbachol-induced, peak (phasic) contraction of circular colonic smooth muscle
We next examined whether CPI-17 activation by PKC-dependent phosphorylation was associated with the increased contractile force elicited by application of CCh to inflamed colonic smooth muscle. Four bands were detected in Phos-tag SDS-PAGE gels with a CPI-17 antibody. These bands collapsed into a doublet when Mn2+ was omitted from the Phos-tag SDS-PAGE gel (Fig. 5A), indicating that the two higher molecular weight bands corresponded to the phosphorylated versions of the unphosphorylated CPI-17 doublet. The phosphorylation of CPI-17 increased following administration of CCh to colonic circular smooth muscle strips isolated from normal mice. No significant difference was observed when the upper and lower CPI-17 species were assessed individually (data not shown). For this reason, the data were calculated as overall CPI-17 phosphorylation (Fig 5B) and are reported as the percentage of total-CPI-17. The CPI-17 phosphorylation level (27.2 ± 2.0%, n = 4) at the peak of CCh-induced force development was significantly greater than that found in NES before the addition of CCh (2.0 ± 1.1%, n = 4). This increase in CPI-17 phosphorylation was significantly reduced by pretreatment with chelerythrine (a 27.0 ± 1.8% reduction, n = 4) and Gö6976 (a 51.5 ± 14.2% reduction, n = 4). The Gö6976 tended to reduce CPI-17 phosphorylation more effectively than chelerythrine, suggesting that the conventional PKC isoforms were involved in CPI-17 phosphorylation at the peak of CCh-induced contraction. Furthermore, neither chelerythrine nor Gö6976 completely inhibited CCh-induced CPI-17 phosphorylation, suggesting that additional protein kinase signaling pathways might also contribute to CPI-17 phosphorylation at the peak of CCh-induced contraction. The CPI-17 phosphorylation levels (13.4 ± 4.2%, n = 4) observed during the sustained phase of CCh-induced contraction were again significantly greater than those found in NES (2.1 ± 0.5%, n = 4) although the amount of CPI-17 phosphorylation was much lower than that observed at the peak of contraction. Significantly, neither chelerythrine nor Gö6976 had any effect on the level of CPI-17 phosphorylation during the sustained phase of contraction.
No change in CPI-17 phosphorylation is observed during carbachol-induced contraction of inflamed colonic circular smooth muscle
The phosphorylation levels of CPI-17 were also increased during CCh-induced contraction of inflamed colonic circular smooth muscle (Fig. 5A and B). However, CPI-17 phosphorylation levels determined at the peak (31.1 ± 0.1%, n = 4) and sustained (10.8 ± 0.1%, n = 4) contractions of colonic tissue isolated from DSS-treated mice were not different from those measured in normal mice. Again, the data are reported as overall CPI-17 phosphorylation as no significant differences were observed when the upper and lower CPI-17 species were assessed individually. Likewise, there were similar effects observed with the administration of PKC inhibitors during the peak contraction of colonic smooth muscle isolated from DSS-treated mice; CPI-17 phosphorylation was significantly reduced by Gö6976 and chelerythrine. Neither chelerythrine nor Gö6976 had any effect on the amount of CPI-17 phosphorylation measured during the sustained phase of contraction in DSS-treated mice. While chelerythrine tended to suppress the CCh-induced CPI-17 phosphorylation associated with sustained (tonic) contraction of smooth muscle, this inhibition was not statistically significant.
Discussion and conclusion
There have been longstanding observations of motility dysfunction and altered smooth muscle contractility in a variety of animal models of intestinal inflammation. Both increased and decreased contractility have been observed with ileitis and colitis. Smooth muscle contractility was increased in nematode (Trichinella spiralis)-induced gut inflammation30–32 and DSS-induced colitis in mice.18 On the other hand, smooth muscle contractility was decreased with intestinal inflammation induced by administration of trinitrobenzene sulfonic acid (TNBS) to rats,33,34 ethanol/acetic acid to dogs,10 and DSS to C57BL/6 mice.28 Moreover, ex vivo studies with smooth muscle strips in an organ culture system have demonstrated that incubation with IL-1β and TNF-α attenuate smooth muscle contractions in both rat27,35 and mouse ileum.34 These contrasting smooth muscle contractile responses were postulated to result from differences in cytokine profiles. Collins and colleagues36 have proposed that nematode-induced hypercontractility is mediated by TH2 cytokines (i.e., IL-4 and IL-13). Other groups have suggested that TNBS-induced hypocontractility is associated with intestinal inflammation mediated by TH1 cytokines (i.e., IL-1β, TNF-α, and IL-12).6,27,34 Thus, it appears that contractile dysfunction depends on the intestinal region, the particular animal model, and the character of the inflammatory stimulus.
We previously demonstrated that hypercontractility of colonic smooth muscle was associated with a TH2 cytokine profile in BALB/c mice following administration of DSS.18 This response was associated with alterations in Ca2+-sensitizing pathways, and uniquely, with contributions from the extracellular-regulated kinase (ERK) and p38 mitogen-activated protein kinase (p38MAPK) that may act in concert with PKC signaling.19 The PKC is an important regulator of colonic smooth muscle contractility under normal conditions,7,9,19,21 and changes in PKC signaling can contribute to motility dysfunction under inflammatory insult.10,11 The mechanism(s) responsible for contractile alterations in colonic smooth muscle following intestinal inflammation is/are not fully understood, and the main objective of this study was to examine the role of PKC-dependent mechanisms in colonic smooth muscle hypercontractility associated with DSS-induced intestinal inflammation in the BALB/c mouse. Accordingly, we identify: (i) the expression of conventional, novel, and atypical PKC isoforms in circular colonic smooth muscle, (ii) a significant involvement of novel and conventional PKC isoforms toward both the initial peak and the sustained phase of CCh-induced contraction, (iii) a noteworthy contribution of CPI-17 to peak force development following CCh-stimulation of normal or inflamed smooth muscle, and (iv) no apparent effect of inflammation on the relative contribution of PKC and CPI-17 toward the observed colonic hypercontractility.
The majority of PKC isoforms are expressed in smooth muscle cells; for example, the expression of all PKC isoforms, except for PKCη, was detected by immunoblotting of human bronchial smooth muscle cells.37 Through a protein kinase screen, we showed that conventional, novel, and atypical PKC isoforms were expressed in colonic circular smooth muscle. This pattern of expression was similar to that provided in a previous report,10 which identified α, βΙ/βΙΙ, and γ (conventional), δ and ε (novel), and ι/λ and ζ (atypical) isoforms in isolated canine colonic smooth muscle cells. We observed distinct differences in the modulation of PKC isozyme expression by inflammation in the BALB/c mouse. For example, we identified decreased expression of PKCβII and increased expression of the PKCε isoform. In previous examinations of canine colonic smooth muscle cells, hypocontractility during colitis was associated with decreased expression and impaired activation of specific PKC isoforms, notably the PKCα, PKCβ, and PKCε isoforms.10 Furthermore, total PKC expression was also reduced in inflamed colonic rat smooth muscle and was associated with reduced carbachol-induced contractions.11 One important distinction separates our examination from the results presented in,10 namely, we have examined PKC expression in dissected circular smooth muscle tissue rather than isolated cells. For this reason, it is possible that the PKC expression observed in our studies may originate from sources other than smooth muscle cells. There is scarce evidence available regarding inflammation-associated changes in expression of PKC isozymes in smooth muscle cells during pathological conditions. Interestingly, a redistribution of PKC isoform expression has been observed for arthritic osteoblasts.38 Proinflammatory TH1 cytokines, such as IL-1β and TNF-α, elicited decreased PKCζ isoform expression. In contrast, no changes in PKCε isoform expression were identified, suggesting that its modulation might not depend on proinflammatory cytokines, such as IL-1β and TNF-α. Furthermore, another study provided evidence for the activation of PKCδ by TNF-α in rat hepatoma cells.39 Taking all of these findings together, we speculated that TH1 or TH2 inflammation-induced alterations in PKC expression could be key for the development of a particular contractile phenotype in colonic smooth muscle.
The ability of different PKC isoforms to mediate Ca2+ sensitization of smooth muscle contraction has been previously pursued.23,40–42 For example, it was reported that PKCα and PKCδ activated CPI-17 in vascular smooth muscle by phosphorylation of Thr-38.40 The contraction of intestinal circular and longitudinal muscles from guinea pig was also induced by agonists, and reflected activation of PKCε and conventional PKC isoforms (i.e., PKCα, PKCβI/βII, and/or PKCγ).23 In another report, PKCε was found to mediate contraction of esophageal circular smooth muscle by activation of Ca2+-independent kinases, such as mitogen-activated protein kinases (MAPKs), zipper-interacting protein kinase (ZIPK), and integrin-linked kinase (ILK),21 which in turn could modulate myosin phosphatase activity. Furthermore, the activation of the PKCβ/CPI-17 pathway was associated with enhanced contraction of the pregnant human myometrium.42 Hence, it is likely that some PKC isoforms have more important roles in Ca2+ sensitization of intestinal smooth muscle than others. In our study, the development of hypercontractility in colonic smooth muscle following DSS administration was associated with increased expression of PKCε and decreased expression of PKCβΙΙ, and it was possible that the increase in expression of PKCε accounted for a portion of the contractile force augmentation observed with inflammation. Indeed, the involvement of PKCε in the generation of sustained contractile forces in intestinal smooth muscle has been clearly demonstrated in previous reports.21,23 As there is a role for both conventional and novel PKC isoforms in Ca2+-sensitizing processes, we applied a battery of PKC inhibitors (i.e., Gö6976, chelerythrine, and GF109203x) with selectivity for the different PKC isoforms. All three inhibitors were able to suppress both peak and sustained contractions induced by CCh in smooth muscle isolated from normal and DSS-treated mice; however, only GF109203x was more effective at attenuating the augmented contractile force induced with inflammation. Increase in peak and sustained forces elicited by DSS-treatment were not suppressed by Gö6976, a selective inhibitor of the conventional PKC isoforms (α and β).24 Although the results for GF109203x are suggestive of some increased role of novel PKC isoforms (e.g., ε and/or δ) with inflammation, we are unable at this time to ascribe the DSS-induced hypercontractile phenotype to an individual PKC isoform.
It is clear that the expression levels of CPI-17 and MYPT1 can contribute appreciably to contractile alterations associated with inflammation.34 But, it appears that hypercontractility associated with DSS-inflammation in BALB/c mice was not associated with changes in the expression of CPI-17 or myosin phosphatase (i.e., the MYPT1-PP1cδ complex). In addition, the percent of phosphorylated CPI-17 did not change upon CCh-stimulation between normal and DSS-treatment groups (Fig. 5), indicating that there was no change in the efficiency of PKC-mediated CPI-17 phosphorylation with inflammation. The PKC may have more roles in smooth muscle contraction than just regulating CPI-17 phosphorylation. One of the membrane-bound targets of PKC is the L-type calcium channel; PKCα, PKCβ, and PKCε activation promotes channel opening and greater influx of extracellular Ca2+, ultimately stimulating greater contractile force.6,10 In our study, there were distinct differences in the contractile responses to membrane depolarization elicited by application of KES solution (i.e., 118 mmol L−1 extracellular K+) when comparing the normal and DSS-treated mice (Fig. 2). Although not pursued herein, it would be useful to obtain a thorough understanding of the alterations to K+, Ca2+, and non-selective cation channels that are important initiators of excitation-contraction pathways.
Previous studies from our laboratory18,19 and others9,43,44 have established that MAPKs are pivotal mediators of contractile function in the intestine. As described herein, acute inflammation and a TH2 cytokine response were associated with hyperresponsiveness of colonic smooth muscle in BALB/c mice. Under these conditions, the expression and activation of two MAPKs, ERK and p38MAPK, were increased.18 The ERK is a downstream target of PKC, and this MAPK can in turn activate a diverse cadre of effector proteins. Caldesmon and HSP27 phosphorylation are two means, whereby MAPKs can contribute to smooth muscle contraction.9 Alternatively, MAPKs are thought to influence the activity of Ca2+-sensitizing kinases (i.e., ILK and ZIPK) that, in turn, are known to phosphorylate LC20 and MYPT1 to generate increased contractility.16 In addition, both ILK and ZIPK are able to phosphorylate CPI-17 in vitro, although PKC is still thought to be predominantly responsible for this phosphorylation in vivo.16,45 It is interesting to note the discrepancy in chelerythrine and Gö6976 effects on CCh-dependent force development and CPI-17 phosphorylation. Namely, the inhibitory effect of chelerythrine on CCh-induced contraction was significantly greater than that of Gö6976, whereas the effect of chelerythrine on CPI-17 phosphorylation was smaller. As PKC is thought to activate ILK and/or ZIPK,21 and modulate L-type Ca2+ channel activity, potentiated CCh-induced contraction may occur independent of PKC-mediated CPI-17 phosphorylation. As such, the administration of chelerythrine would contribute to contractile attenuation, without being reflected in the levels of CPI-17 phosphorylation.
In conclusion, murine colonic circular smooth muscle expresses a variety of PKC isoforms from the conventional, novel, and atypical groups, and our data suggest that PKC and CPI-17 are important mediators of colonic contractile responses. Both conventional and novel PKC isozymes contribute to both the peak and the sustained phases of CCh-induced colonic smooth muscle contraction. A significant induction of CPI-17 phosphorylation with peak (but not sustained) force development was observed following CCh-stimulation. While inflammation influenced PKC isoform expression in colonic circular smooth muscle tissue, the total CCh-induced contribution of PKC and CPI-17 to hypercontractility was not increased in inflamed smooth muscle.
Acknowledgements and Disclosures
E.I. was a recipient of a Uehara Memorial Foundation Fellowship (Japan) and Canadian Association of Gastroenterology/Canadian Institutes of Health Research/AstraZeneca fellowship. S.T. was a recipient of an Alberta Innovates- Health Solutions studentship. The research was supported by the Canadian Institutes of Health Research and a Canada Research Chair (Tier II) in Smooth Muscle Pathophysiology (J.A.M.).
EI and MC performed the experiments; EI and SRT analyzed the data and wrote the paper; JAM designed the research study, analyzed the data, and wrote the paper.