To determine the role of SM-B myosin in smooth muscle physiology, we have generated a mouse model deficient only in SM-B myosin (smb−/−). Removal of SM-B-specific exon 5B results in a switch to SM-A, but does not affect survival or cause any overt smooth muscle pathology (Babu et al. 2001). Physiological analysis reveals that loss of SM-B myosin results in a significant decrease in velocity of muscle shortening in bladder and in mesenteric vessels which express high levels of SM-B myosin (Babu et al. 2001, 2004; Karagiannis et al. 2003, 2004). These studies suggest that the 7 amino acid insert could be involved in the kinetics of smooth muscle contraction. However, there are reports which show no correlation between SM-B myosin content and mechanical performance of the smooth muscle (Haase & Morano, 1996; Siegman et al. 1997). In pregnant hypertrophied rat myometrium (Haase & Morano, 1996) and in the mouse megacolon of Hirschsprung's disease (Siegman et al. 1997) the maximal shortening velocity was increased, while the SM-B myosin isoform content was decreased. We have also observed altered expression of thin-filament proteins like h1-calponin (h1-CaP) and h-caldesmon (h-CaD) in the smb−/− mice, suggesting that these proteins could possibly be involved in the regulation of smooth muscle contractility in smb−/− mice (Babu et al. 2004).

Calponin, an actin-binding regulatory protein, has been found in smooth muscle and non-muscle tissues and is encoded by three different homologous genes (reviewed by Morgan & Gangopadhyay, 2001). Basic (h1) CaP is the major calponin found in the smooth muscle and has been extensively studied (reviewed by Takahashi & Yamamura, 2003). Several in vitro studies have suggested that h1-CaP inhibits actin-activated myosin ATPase activity (Horiuchi & Chacko, 1991; Haeberle, 1994; Winder et al. 1998) in a dose-dependent manner (Itoh et al. 1994; Jaworowski et al. 1995; Uyama et al. 1996). Ablation of h1-CaP resulted in faster unloaded shortening velocity of smooth muscle contraction (Matthew et al. 2000; Takahashi et al. 2000; Fujishige et al. 2002). The finding that h1-CaP levels are increased in the SM-B null mice (Babu et al. 2004) may suggest that h1-CaP plays a role in the decreased velocity of shortening observed in this model. However, it is unclear whether the up-regulation of h1-CaP in the SM-B null mice is a compensatory mechanism to maintain function or the cause for a decrease in contractile function. To critically assess the role of h1-CaP in SM-B null mice, we generated a double knockout mouse model lacking both h1-CaP and SM-B and studied smooth muscle contractility of non-vascular (urinary bladder) and vascular (mesenteric arteries) tissues.

Animals were killed by exposure to rising concentrations of CO₂, in accordance with the ethical treatment of animals, using a protocol approved by the Ohio State University and Case Western Reserve University, or by cervical dislocation and thoracotomy following isoflurane anaesthesia, according to a similarly approved protocol at the University of Vermont. The mice used for the following studies were F3 generations of both sexes and between 18 and 25 weeks old.

Bladder All solutions and the methods for tissue preparation are described in detail elsewhere (Karagiannis et al. 2003), and solutions included 250 units ml⁻¹ of calmodulin. Briefly, bladder tissues were cut into strips of approximately 800 μm by 200–300 μm by 20–50 μm and mounted between aluminium foil T-clips. Strips were then transferred to a mechanics workstation (Model 600, Aurora Scientific, Aurora, Canada) and mounted between a force transducer (Akers AE 801, MEMSCAP, San Jose, USA) and servomotor (Aurora Scientific, Aurora, Canada) in relaxing solution (pCa 9). Strips were stretched to just develop tension and then stretched to maximal length (L_o) (an additional 30%). Then the strips were skinned in pCa 9 with 1% Triton solution for 15–20 min, after which the smooth muscle strips were thiophosphorylated with a protocol consisting of bathing the fibre in rigor solution without Ca²⁺ (mm: 5 TES, 1.2 MgCl₂, 30 EGTA, 26 HDTA) for 10 min, followed by rigor with Ca²⁺ (mm: 50 TES, 1.2 MgCl₂, 28 Ca-EGTA, 1 EGTA, 26 HDTA) for another 10 min and then 30 min in rigor with Ca²⁺ and 10 mm ATPγS, then placed back in rigor solution. Tissue strips were then transferred to relaxing solution containing ATP (mm: 5.5 Na₂ATP, 56.5 potassium methane sulphonate, 7.2 MgCl₂, 25 creatine phosphate, 25 N,N-bis(2-hydroxyethyl)-2-aminoethanesulphonic acid (Bes), and 0.02 CaCl₂, pH 7.0) and force increased to a steady-state plateau. Once reaching steady state, the maximal shortening velocity (V_max) was determined.

To determine shortening velocity, force was rapidly lowered by shortening the preparation and using a computer program to adjust muscle length to clamp force at a new isotonic level (50%, 30%, 20%, 10%, 5%, 1%). The velocity of shortening was defined as the linear fit between the times of 80 ms and 90 ms to the length versus time trace. The resulting data of force versus velocity was then fitted to the Hill equation (V + b)(P + a) = b(P_o + a), where V is the rate of shortening, P is the clamped force, P_o is the maximal force (force at L_o), and a and b are constants. The maximal shortening velocity is extrapolated as the velocity at P = 0.

Mesenteric arteries Mice were shipped to the University of Vermont and allowed to acclimate for 1 week prior to experimentation. They were then anaesthetized as described above and killed by cervical dislocation. The abdomen was opened by longitudinal incision, and a section of the gut and its vessels, approximately 5–10 cm distal to the pylorus, was excised and transferred to a Petri dish filled with ice-cold N-(2-hydroxyethyl)piperazine-N′-2-ethanesulphonic acid (Hepes)-buffered physiological saline solution (Hepes-PSS). Using microdissection tools, tertiary branches of the mesenteric arcade (unpressurized arterial diameter approximately 100 μm) were dissected free of surrounding connective and fat tissue and transferred to the experimental chamber of a small-artery arteriograph. Vessels were then tied onto two glass cannulas and pressurized to 50 mmHg using a pressure-servo apparatus. Changes in vessel diameter were measured with a video dimension analyser (Living Systems Instrumentation) and recorded on a PC computer using specialized imaging software. A more detailed description of the small artery pressure system can be found elsewhere (D'Angelo & Osol, 1993).

Following 60 min equilibration at 37°C, the vessels were constricted with a solution containing phenylephrine (10 μm) and a high concentration of potassium (124 mm) to induce maximal constriction. Time to half-maximal constriction (T_1/2, s), slope of maximal velocity during the initial phase of constriction (normalized to diameter by expressing the data as the percentage change in diameter per second), and the extent of maximal constriction (percentage reduction in lumen diameter) were determined from experimental records. Vessels were then subjected to step increases in transmural pressure in 20 mmHg increments until loss of constriction was observed (forced dilatation, FD, at pressures above 180 mmHg). The pressure at which FD occurred in each vessel was used to calculate average FD pressure (mmHg), thereby providing an index of the maximal force production by vascular smooth muscle within the arterial wall.

To verify that h1-CaP is not expressed in the smooth muscles of dKO mice, Western blot analysis was carried out using total homogenate from bladder and mesenteric vessels of wild-type (WT), smb−/− and dKO mice. Figure 1 shows that CaP expression was increased in the smb−/− null bladder (2.79 ± 0.10 fold; P < 0.05, n = 3, Fig. 1A and C) and in the mesenteric vessels (2.33 ± 0.03 fold; P < 0.05, n = 3, Fig. 1B and C) compared to their respective WT tissues as reported earlier (Babu et al. 2004), but was completely absent in the dKO mice. The expression level of non-muscle β-actin protein determined as internal loading control was unchanged between the groups. These results additionally confirm the successful generation of dKO mice lacking both SM-B and CaP proteins.

Figure 1 Calponin expression in SM-B null and dKO mice bladder

To determine if loss of h1-CaP expression alters the expression of other CaP isoforms, mRNA levels of neutral (h2)-CaP and acidic (h3)-CaP in bladder tissues were analysed by reverse-transcription-PCR. Our results shown in Fig. 1D demonstrate that loss of h1-CaP does not affect the expression of h2- or h3-CaP isoforms.

Figure 2 Histological staining of SM-B null and dKO mice bladder

Figure 3 Expression of myosin heavy and light chain isoforms

Figure 4 Expression of smooth muscle α-actin and h-CaD in dKO bladder

Figure 5 Expression of smooth muscle α-actin and h-CaD in dKO mesenteric vessels

Smooth muscle-specific h-CaD has been shown to be down-regulated in the smb−/− mice (Babu et al. 2004). To determine if loss of h1-CaP affects CaD levels, Western blot analysis was carried out using smooth muscle-specific h-CaD antibody. Our results show that h-CaD protein levels are decreased in the smb−/− bladder (Fig. 4A and C) and mesenteric vessels (Fig. 5A and C) as reported earlier (Babu et al. 2004). Interestingly, ablation of CaP restores the h-CaD expression in the dKO bladder (Fig. 4A and C) as well as in the mesenteric vessels (Fig. 5A and C) and its level is significantly higher than the wild-type tissues.

Figure 6 Functional measurements on smb−/− and dKO bladder

Table 1 Contractile properties of mesenteric vessels

Relative to vessels from the smb−/− animals, the pressure at which forced dilatation occurred was significantly increased in the dKO mesenteric vessels, and reached the levels of WT control vessels (Table 1). Forced dilatation (FD; used in the present study) is an indirect measure of maximal force production, and is defined as the point at which transmural pressure overcomes the ability of vascular smooth muscle to produce an adequate counterforce to circumferential tension produced by the intraluminal pressure. These data showing enhanced force production by mesenteric artery smooth muscle from dKO versus smb−/− animals are in general agreement with those obtained from bladder smooth muscle and described above.

The major goal of the current study was to determine the functional significance of h1-CaP up-regulation in SM-B null mice (Babu et al. 2004). To critically asses the role of h1-CaP in the contractile function of SM-B null smooth muscles, we ablated h1-CaP in the SM-B null background by crossing h1-CaP and SM-B single knockout mice. The key findings of this study are as follows. (1) Ablation of CaP in the SM-B null mice bladder resulted in increased V_max and steady-state force generation. (2) In mesenteric vessels, the maximal force as measured by forced dilatation was increased in the dKO mice, and the higher maximal force allows the vessel to shorten faster under a load. The shortening velocity, however, was decreased in both smb−/− and dKO vessels. (3) Interestingly, these functional changes were observed with a decrease in smooth muscle α-actin levels and an increase in caldesmon levels.

We have previously shown that switching of SM-B to SM-A myosin in a SM-B null mouse model resulted in decreased velocity of shortening (Babu et al. 2001; Karagiannis et al. 2003) and increased expression of h1-CaP (Babu et al. 2004). The available data strongly suggest that h1-CaP is involved in the regulation of actin–myosin interaction and therefore the kinetics of smooth muscle contractility (Takahashi & Yamamura, 2003). Ablation of h1-CaP significantly increases the unloaded shortening velocity in phasic smooth muscle tissues such as bladder and vas deferens (Matthew et al. 2000; Takahashi et al. 2000). These findings taken together suggest that the up-regulation of h1-calponin could negatively influence the velocity of shortening in SM-B null mice. Therefore, the dKO mouse model lacking both h1-CaP and SM-B myosin provide us with an excellent model to determine the functional significance of h1-CaP up-regulation in SM-B null mice. In dKO mice, loss of both h1-CaP and SM-B myosin proteins does not affect survival or cause any overt smooth muscle pathology as observed in the individual knockout animals (Yoshikawa et al. 1998; Babu et al. 2001). Loss of h1-CaP is not compensated by other CaP isoforms (Fig. 1C) and consistent with the previous studies on h1-CaP single knockout (Yoshikawa et al. 1998). Interestingly, loss of CaP in the SM-B null background is associated with ~40% reduction in the smooth muscle α-actin levels. However, loss of h1-CaP does not affect the β- and γ-actin isoforms (Figs 1, 4 and 5). These changes are also consistent with changes associated with CaP single knockout mice (Matthew et al. 2000) and indicate the preservation of the CaP null phenotype in the dKO mice.

As previously reported, the loss of SM-B myosin does not alter the unitary force generation in bladder muscle compared to the WT bladder (Karagiannis et al. 2003, 2004). However, ablation of h1-CaP resulted in increased steady-state force generation in the SM-B null mice. Increased steady-state force generation under conditions where LC₂₀ were maximally, irreversibly phosphorylated, and in the absence of structural changes (as observed by histology) suggests that h1-CaP may play a role in modulating bladder smooth muscle contractility. However, these results differ from the previous report using h1-CaP single knockout mice, which showed no significant difference in force development in the skinned bladder muscles between WT and h1-CaP null mice (Matthew et al. 2000), and rather decreased force production in intact vas deferens and aortic smooth muscles (Takahashi et al. 2000; Fujishige et al. 2002). The unexpected increase in force in the dKO bladder could be due to the complete replacement of SM-B by the SM-A myosin isoform in the bladder and/or other compensatory changes yet to be analysed.

In agreement with previous results, the WT bladder shortened at a rate 20% faster than the smb−/− bladder (Karagiannis et al. 2003). The V_max, however, was significantly increased in the dKO bladder as compared to smb−/− bladder, under conditions in which the levels of myosin thiophosphorylation were similar. These results indicate that myosin light chain phosphorylation does not contribute for the observed velocity changes. A similar increase in V_max has been reported for h1-CaP single knockout (Matthew et al. 2000; Takahashi et al. 2000). We have previously shown that the 7 amino acid insert regulates the earlier rate-limiting step including nucleotide binding and inorganic phosphate release (Karagiannis et al. 2003) whereas CaP is proposed to regulate the later steps in the cross-bridge cycle such as dissociation of high-affinity cross-bridges from actin, by binding to actin (EL-Mezgueldi & Marston, 1996; Takahashi & Yamamura, 2003). Therefore, both h1-CaP and SM-B myosin could affect the cross-bridge cycling and V_max but at different kinetic steps. Taken together our studies suggest that up-regulation of h1-CaP in the SM-B null mice may be a compensatory alteration to maintain a reduced level of cross-bridge cycling over time in the absence of SM-B myosin. However, reduction in the smooth muscle α-actin levels and increase in h-CaD levels suggest that changes in other contractile and regulatory proteins may also contribute to the altered contractile properties observed in the dKO bladder.

In addition to bladder, the contractile properties of mesenteric vessels were determined in the dKO mice. We have previously reported that SM-B myosin is expressed heterogeneously in the mesenteric vessels and contributes to their unique mechanical properties (Babu et al. 2004). In that study, we also found that loss of SM-B myosin resulted in increased force generation under isometric conditions (using wire-mounted vessel rings) in response to phenylephrine + potassium (Babu et al. 2004). In the present study, which used a more physiological pressurized vessel approach, maximal constriction to an activation cocktail containing high potassium (124 mm) and phenylephrine (10 μm) was similar in all treatment groups. Since the vessel is studied as an intact unit, the maximal extent of constriction, measured as the reduction in lumen diameter from a fully relaxed state, may be limited by other factors, e.g. cell–matrix interactions, rather than simply maximal force generation.

The (1) increased half-time to maximal constriction, and (2) reduced normalized slope of constriction observed in mesenteric vessels of both smb−/− and dKO mouse models, both suggest decreased shortening velocity of smooth muscle. These findings are consistent with our previous studies showing decreased shortening velocity in the mesenteric vessels with SM-B deficiency (Babu et al. 2001, 2004), and indicate that changes in h1-CaP do not seem to be the underlying cause for the altered kinetics in this type of smooth muscle. Another study using rat portal vein also showed that the maximal velocity of vascular smooth muscle shortening was independent of h1-CaP expression (Facemire et al. 2000).

The significantly reduced forced dilatation pressure in vessels from the smb−/− but not dKO mice was an unexpected finding that suggests that the modulatory effect of calponin on vascular smooth muscle (VSM) contractility may be influenced by (1) the extent of VSM activation, (2) spatial events such as the degree of actin–myosin overlap (the length–tension relationship), and (3) the nature of the distending force acting upon the vascular wall. All three possibilities derive from our use of the more physiological pressurized vessel methodology in this study, since earlier work used the isometric wire-mounted approaches in which vessels are stretched to a fixed length prior to activation and therefore unable to shorten in response to activation. In the pressurized vessel, as in the body, activation results in constriction, which is a reflection of smooth muscle shortening. This may have implications for the spatial relationship between actin, myosin and calponin, and thereby alter the nature of thin filament regulation, as there were no detectable differences in myosin heavy and light chain isoform expression. Further, in pressurized vessels, the force experienced by the vascular wall is true transmural pressure, which has a radial as well as a circumferential component that may influence the distribution of intramural stress. Although it was beyond the scope of this study to differentiate the ultrastructural/spatial versus biomechanical components of the response, these elements deserve consideration in future studies. Finally, although the velocity data were consistent with previous findings in mesenteric vessels (Babu et al. 2004) and in bladder smooth muscle (this study), it should be noted that, in arteries, the importance of shortening velocity in vivo is questionable, while the production and maintenance of force is essential to the maintenance of vascular tone and, thereby, peripheral resistance, blood pressure and normal organ perfusion. This differs from the bladder, whose proper function is dependent on smooth muscle shortening for effective micturition. Thus, it is not unreasonable to expect differences in the nature of h1-CaP regulation based on physiological function. Based on the results of our studies, h1-CaP appears to be a key regulator of both bladder and vascular smooth muscle contractility, although its precise modulatory role may clearly vary with smooth muscle tissue type and function.

This work was supported by American Heart Association Grant 0365173B (to G.J.B.), National Heart, Lung, and Blood Institute Grant HL-38355-17 (to M.P.) and HL-44181 (to F.V.B.).