Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2001 Jul 15;534(Pt 2):357–366. doi: 10.1111/j.1469-7793.2001.00357.x

Is myosin phosphatase regulated in vivo by inhibitor-1? Evidence from inhibitor-1 knockout mice

Andrew N Carr *, Roy L Sutliff *, Craig S Weber *, Patrick B Allen †, Paul Greengard †, Primal de Lanerolle ‡, Evangelia G Kranias *, Richard J Paul *
PMCID: PMC2278711  PMID: 11454956

Abstract

  1. The Ca2+ sensitivity of smooth muscle contractility is modulated via regulation of phosphatase activity. Protein phosphatase inhibitor-1 (I-1) is the classic type-1 phosphatase inhibitor, but its presence and role in cAMP-dependent protein kinase (PKA) modulation of smooth muscle is unclear. To address the relevance of I-1 in vivo, we investigated smooth muscle function in a mouse model lacking the I-1 protein (I-1(−/−) mice).

  2. Significant amounts of I-1 protein were detected in the wild-type (WT) mouse aorta and could be phosphorylated by PKA, as indicated by 32P-labelled aortic extracts from WT mice.

  3. Despite the significant presence of I-1 in WT aorta, phenylephrine and KCl concentration- isometric force relations in the presence or absence of the PKA pathway activator isoproterenol (isoprenaline) were unchanged compared to I-1(−/−) aorta. cGMP-dependent protein kinase (PKG) relaxation pathways were also not different. Consistent with these findings, dephosphorylation rates of the 20 kDa myosin light chains (MLC20), measured in aortic extracts, were nearly identical between WT and I-1(−/−) mice.

  4. In the portal vein, I-1 protein ablation was associated with a significant (P < 0.05) rightward shift in the EC50 of isoproterenol relaxation (EC50 = 10.4 ± 1.4 nm) compared to the WT value (EC50= 3.5 ± 0.2 nm). Contraction in response to acetylcholine as well as Ca2+ sensitivity were similar between WT and I-1(−/−) aorta.

  5. Despite the prevalence of I-1 and its activation by PKA in the aorta, I-1 does not appear to play a significant role in contractile or relaxant responses to any pharmacomechanical or electromechanical agonists used. I-1 may play a role as a fine-tuning mechanism involved in regulating portal vein responsiveness to β-adrenergic agonists.

Regulation of phosphoprotein phosphatases has been shown to be a major component in the Ca2+ sensitization of smooth muscle contraction (Somlyo & Somlyo, 2000). The ∼19 kDa phosphatase inhibitor-1 (I-1) protein is the classic regulator of type-1 phosphatase activity (Oliver & Shenolikar, 1998) and has been shown to be present in smooth muscle (Elbrecht et al. 1990). However, studies of the relevance of I-1 in regulating smooth muscle myosin phosphatase activity in vitro and, consequently, smooth muscle function have provided conflicting results (Alessi et al. 1992; Tokui et al. 1996). These data may be due to the inability of in vitro systems to recapitulate the complex regulatory mechanisms acting on the phosphatase, including its regulation by kinases and/or targeting subunits. Thus, an in vivo approach is essential to determine the physiological effects of the inhibitor-1 protein. To address this issue, we used the recently developed I-1 knockout (I-1(−/−)) mouse to delimit the role of I-1 in both tonic and phasic smooth muscle contractility.

A major mechanism involved in regulating smooth muscle contractility is the phosphorylation of serine-19 on the 20 kDa myosin light chains (MLC20) (de Lanerolle & Paul, 1991; Hartshorne et al. 1998), resulting in the activation of myosin-actin crossbridge cycling and contraction. The phosphorylation status of MLC20, and consequently contraction, is controlled by the dynamic balance between the activities of myosin light chain kinase (MLCK) and myosin light chain phosphatase (MLCP). MLCK has been well characterized (Gallagher et al. 1997) and its activation is Ca2+ dependent, involving a Ca2+-calmodulin complex. It is now clear that MLCP is also regulated and this regulation has been suggested to involve the cytosolic I-1 (Somlyo et al. 1989).

MLCP is a heterotrimer consisting of a ∼37 kDa catalytic subunit (PP1c), a ∼20 kDa subunit of unknown function, and a ∼110–130 kDa myosin phosphatase targeting subunit (MYPT1) (Shirazi et al. 1994; Hartshorne et al. 1998). Interaction of PP1c with MYPT1 confers selectivity of PP1c towards the myosin molecule (Alessi et al. 1992) and enhances PP1c activity (Shirazi et al. 1994; Ichikawa et al. 1996). As in other tissues, the targeting subunit may play a direct role in regulating PP1c activity. For instance, in striated muscle, phosphorylation of the glycogen targeting subunit, RGL, results in dissociation of PP1c, and inactivation of the phosphatase by I-1 (Hubbard & Cohen, 1989). By analogy, phosphorylation of MYPT1 by protein kinase C (PKC) may perform a similar function in regulating the interactions between MYPT1 and PP1c (Feng et al. 1999; Toth et al. 2000). Although the physiological significance of altered MLCP regulation is not known with certainty, myosin phosphatase activity probably plays a paramount role in certain pathological states (Solaro, 2000).

I-1 is an endogenous smooth muscle phosphatase inhibitor that is inactive when dephosphorylated. However, when activated by PKA (Cohen, 1989) or PKG (Hemmings et al. 1984; Tokui et al. 1996) phosphorylated I-1 specifically inhibits type-1 phosphatase activity. Regulation of myosin phosphatase activity has also been suggested as an important mechanism in the Ca2+-sensitizing properties of agents such as cGMP (Lee et al. 1997). Activation of I-1 by PKG suggests a possible role for I-1 in this effect, but this remains unconfirmed in vivo. In vitro experiments using the purified myosin phosphatase holoenzyme have suggested that the MLCP is relatively insensitive to I-1 (Alessi et al. 1992). Contrary to these results, activated (phosphorylated) I-1 has been shown to inhibit myosin phosphatase activity (Mitsui et al. 1992). This finding is supported by contractility experiments, performed in β-escin-skinned smooth muscle cells (Tokui et al. 1996), which demonstrated that phosphorylated I-1 enhanced contraction at submaximal [Ca2+], suggesting that I-1 may play a role in regulating the Ca2+ sensitivity of smooth muscle phosphatase activity in vivo. These results seemingly lead to an apparent paradox, since β-adrenergic receptor stimulation, which results in phosphorylation of I-1, is generally associated with smooth muscle relaxation. Oliver & Shenolikar (1998) suggest that I-1 phosphorylation may decrease type-1 phosphatase activity against the inhibitory phosphorylation site on MLCK (Conti & Adelstein, 1981). This would decrease myosin phosphorylation, and provides a path for I-1 phosphorylation in β-adrenergic receptor-mediated relaxation of smooth muscle. However, evidence to support such a conclusion is lacking. Thus, the role of I-1 in the regulation of smooth muscle contractility in vivo is not known with certainty.

The potential importance of I-1 in regulating smooth muscle phosphatase activity is underscored by two other mechanisms known to regulate the myosin phosphatase. These involve the phosphorylation of MLCP by Rho-kinase (Fujita et al. 1995; Kimura et al. 1996) or inhibition of MLCP by the ∼17 kDa PKC-activated protein phosphatase inhibitor CPI-17 (Li et al. 1998), each representing unique mechanisms for inhibiting myosin phosphatase activity in vivo (Somlyo & Somlyo, 2000). These studies suggest that phosphatase regulation via endogenous inhibitor proteins, such as I-1, may represent novel and potentially important participants in smooth muscle function.

To assess the role of I-1 in smooth muscle contractility, we have examined contractility in tonic (aorta) and phasic (portal vein) muscle from WT and I-1(−/−) mice, in response to a variety of electromechanical and pharmacomechanical agonists. Our data suggest that I-1 ablation does not affect contractility in aorta, but may play a subtle role in regulating β-adrenergic receptor responsiveness in the portal vein.

METHODS

Generation of protein phosphatase inhibitor-1 knockout mice

I-1(−/−) mice were generated as previously described (Allen et al. 2000). Handling of mice and experimental procedures were performed in accordance with protocols approved by the Institutional Animal Care and Use Committee at the University of Cincinnati.

Aortic vessel preparations

Vessels were prepared as previously described (Lalli et al. 1997). Briefly, 8- to 10-week-old C57Bl/6 WT and I-1(−/−) mice were killed by asphyxiation in a pre-charged CO2 chamber. Aortas were immediately dissected, rinsed in cold bicarbonate physiological saline solution (PSS), and connective tissue and fat were removed. Tissues were maintained in a PSS solution containing (mm): 118 NaCl, 4.73 KCl, 1.2 MgSO4, 0.026 EDTA, 1.2 NaH2PO4, 2.5 CaCl2, 11 glucose, and 25 NaHCO3; pH when bubbled with 95 % O2-5 % CO2 at 37 °C was 7.4. Experiments on aortic tissue were performed in both endothelium-intact and denuded tissue. Data presented are from denuded aorta only, unless otherwise indicated. In cases of denuded aortas, the endothelium was removed by gently rubbing the vessel between thumb and forefinger. Data were acquired using a Biopac MP100 data acquisition system and analysed using Acqknowledge software.

Western blot analysis

Aortas from both WT (n = 5) and I-1(−/−)(n = 7) mice were excised, washed, and homogenized in an ice-cold buffer containing (mm): 10 imidazole (pH 7.0), 300 sucrose, 1 DTT, 1 sodium metabisulfite, 100 NaF and 2 EDTA, and protease inhibitors 0.3 mm phenylmethylsulfonyl fluoride (PMSF), 5 μg ml−1 leupeptin and 10 μg ml−1 soybean trypsin inhibitor, pH 7.0. Brain samples from both groups were prepared using an identical buffer. Protein concentrations were determined by the BioRad method, comparing samples versus a BSA standard. Homogenates were subjected to SDS-PAGE, blocked with 5 % non-fat milk in Tris-buffered saline + 0.2 % Tween-20 (TBST), and incubated with a rabbit, polyclonal I-1-specific antibody (generated in-house and provided by Paul Greengard) for 1 h. Blots were then incubated for 1 h with anti-rabbit secondary antibodies. Membranes were subjected to Enhanced Chemiluminescence PLUS detection reagents (Amersham) and exposed to film for appropriate times.

Isolation of aortic extracts

Extracts from five WT and five I-1(−/−) aortas were obtained by homogenizing each with a ground glass pestle in 10 volumes of ice-cold buffer containing (mm): 40 Tris-HCl pH 7.5, 300 NaCl, 5 DTT, 5 EDTA and 0.2 PMSF, with 1 % nonidet P-40, 10 μg ml−1 RNAse, 10 μg ml−1 DNAse, 10 μg ml−1 leupeptin, 10 μg ml−1 pepstatin, and phosphatase inhibitors 50 nm deltamethrin and 2 nm okadaic acid. Extracts were spun at 50 000 g for 20 min and the supernatant saved for MLC20 dephosphorylation assays or 32P-labelling.

Radiolabelling of I-1 in aortic extracts

Extracts from WT or I-1(−/−) aorta (70 μl) were radiolabelled in a solution containing 11.5 mm MgSO4, 5 μg of the catalytic subunit of PKA purified from bovine heart (Sigma Chemical Co.) and 0.1 mmγ-labelled 32P-ATP (2200 c.p.m. pmol−1) in a final volume of 100 μl (1 h at room temperature). The reaction was stopped by the addition of 5 μl of 0.5 m EDTA. An aliquot (7 μl) was removed, boiled in SDS sample buffer, and subjected to SDS-PAGE and autoradiography.

Radioactive labelling of MLC20

Bacterially expressed rat aorta MLC20 (80 μg) was phosphorylated in a final volume of 100 μl in a solution containing 100 mm NaCl, 1 mm DTT, 10 mm MgSO4, 0.1 mmγ-labelled 32P-ATP (2200 c.p.m. pmol−1) and 20 mm Tris, pH 7.5, for 30 min at room temperature with 0.1 μg of bacterially expressed, catalytically active, truncated MLCK. The reaction was stopped by adding EDTA (final concentration, 15 mm).

MLC20 dephosphorylation rates

Dephosphorylation assays were performed by combining 45 μl of phosphorylated MLC20 with 40 μl of WT or I-1(−/−) extract and incubating at room temperature. Aliquots of the reaction mixture were removed at 0, 2.5, 5, 10, 20, 40 and 60 min, separated by SDS-PAGE and stained with Coomassie brilliant blue. Bands corresponding to MLC20 were excised and counted.

Portal vein force measurements

Portal veins were mounted isometrically in a longitudinal orientation, as previously described (Sutliff et al. 1999). Tissues were maintained in PSS as described for the aorta. One loop of the portal vein was attached to a triangle-shaped holder and the other loop was attached to a Harvard force transducer (Harvard Apparatus, Hollistom, MA, USA). Experiments were conducted after optimal tension was achieved by adjusting the length of the vessel until maximal peak-peak oscillations were observed. Data were acquired and analysed as stated for the aorta.

Statistical analysis

All values are expressed as means ± standard error of the mean (s.e.m.). Statistical comparisons between WT and I-1(−/−) mice were made by performing two-way repeated measures ANOVA (Stewart-Newman-Kuels) or Student's unpaired t test, where applicable. Concentration-response curves were generated using Logistic Fit from Origin software, and fitting parameters were averaged. Statistical significance was imparted at P < 0.05; n refers to the number of mice used in each experimental group.

RESULTS

Inhibitor-1 is present in the mouse aorta

Aorta from WT and protein phosphatase I-1(−/−) mice were indistinguishable by their gross anatomical parameters (Table 1). I-1 has been reported to be present in smooth muscle of higher animal species (Elbrecht et al. 1990) but has not been examined in the mouse aorta. Although the molecular mass of I-1 is ∼19 kDa, this protein has been shown to migrate at ∼26–29 kDa on SDS-PAGE (Allen et al. 2000), probably due to its abnormally low detergent-binding properties. Western blot analysis of WT and I-1(−/−) aorta or brain homogenates (included as positive and negative controls) revealed an immunoreactive band only in the WT samples (Fig. 1a), demonstrating the presence of I-1 in aortic smooth muscle from WT mice. To determine whether I-1 can be phosphorylated by PKA in mouse aorta, aortic extracts from WT and I-1(−/−) mice were incubated with PKA in the presence of [γ-32P]ATP. Autoradiograms of these extracts revealed a robust signal at ∼29 kDa only in WT aorta (Fig. 1B), consistent with the phosphorylation of I-1 under these conditions.

Table 1.

Dimensions of WT and I-1(−/−) aorta and portal vein

WT I-1(−/−)
Aorta
n 10 10
Weight (mg) 0.90±0.05 0.90±0.04
Length (mm) 5.34±0.17 5.23±0.12
Thickness (mm) 0.075±0.004 0.074±0.003
CSA (mm2) 0.79±0.05 5.76±0.03
Portal vein
n 10 10
Weight (mg) 0.32±0.05 0.36±0.05
Length (mm) 3.29±0.12 3.59±0.17
CSA (mm2) 0.09±0.01 5.09±0.01
Open in a new tab

weight, length and cross-sectional ares (CSA) were similar between WT and I-1(−/−) aorta or portal vein. CSA in aorta was calculated by the equation: CSA=[(2 × weigth)/(1.06 × circumference)]. CSA in portal vein was calculated by the equation: weigth/[(1.06 × length)]. Wall thickness was estimated in aortas by the equation: Thickness = weight/(1.05 × length × circumference).

Figure 1. Protein phosphatase inhibitor-1 is present and phosphorylated in mouse aorta.

Figure 1

Open in a new tab

A, homogenates for both wild-type (WT) and I-1(−/−) aorta (100 μg) were subjected to Western blot analysis. Homogenates (10 μg) from WT brain (lane 1) and I-1(−/−) brain (lane 2) were run as positive and negative controls, respectively. An immunoreactive band at ∼29 kDa in both WT aorta and brain (lanes 1 and 3) was absent in the I-1(−/−) tissues (lanes 2 and 4). B, extracts phosphorylated by PKA in the presence of [γ-32P]ATP were subjected to SDS-PAGE and autoradiography. WT extracts contained a signal at ∼29 kDa, which was absent in the I-1(−/−) extracts, consistent with the phosphorylation of I-1 under these conditions.

Effects of I-1 ablation on aorta contractility

Administration of phenylephrine (PE), an α-adrenergic agonist, results in activation of the phosphatidylinositol pathway and induces smooth muscle contraction (Gong et al. 1995). We anticipated that increased phosphatase activity against MLC20, due to ablation of I-1, would attenuate and/or desensitize PE-induced contractions. This would lead to a rightward shift in the PE concentration-force relationship. Steady-state isometric force and sensitivity to PE were indistinguishable between WT and I-1(−/−) aorta (Fig. 2). The time to 50 % maximal contraction/force (TC50) was also similar between the WT and I-1(−/−) aorta (37.3 ± 2.67 vs. 45.5 ± 5.2 s, n = 5, P > 0.05). Since I-1 is only active when phosphorylated by PKA or PKG, functional differences between WT and I-1(−/−) mice should be accentuated under conditions in which I-1 is activated. To ensure that I-1 is phosphorylated under these conditions and thus active in our assay, PE concentration-force relations were generated after pretreatment of aortas with isoproterenol (300 nm). Following isoproterenol pretreatment, the PE-force relations were shifted to the right (decreased sensitivity) in both groups (Fig. 2B). However, isoproterenol treatment did not reveal any contractile differences between WT and I-1(−/−) aorta. The TC50 values for PE following 300 nm isoproterenol were 40.5 ± 7.6 and 38.3 ± 4.9 s, for WT (n = 6) and I-1(−/−)(n = 6) aorta, respectively (P > 0.05).

Figure 2. Phenylephrine concentration-isometric force relations.

Figure 2

Open in a new tab

Receptor-mediated responses to phenylephrine (PE) were generated in WT (•) and I-1(−/−) (□) denuded aortas. A, force/area responses to phenylephrine were not different between the two groups (P > 0.05). B, percentage maximal force was unchanged and pretreatment with isoproterenol (dotted lines) resulted in a rightward shift in both WT (▿) and I-1(−/−) (▵) curves to a similar extent (P > 0.05, n = 4 for all groups). Values represent means ±s.e.m.

Alternatively, smooth muscle contraction can be elicited by high potassium solutions causing membrane depolarization and a resulting influx of extracellular Ca2+ (receptor independent). If I-1 regulates MLCP activity, KCl-induced contractions in I-1(−/−) mice should be attenuated and/or rightward shifted. However, the KCl concentration- isometric force relations were similar between WT and I-1(−/−) aorta (Fig. 3a). TC50 values in response to KCl-induced contractions were similar between WT and I-1(−/−) aorta (72.4 ± 22.7 vs. 66.5 ± 13.3 s, n = 5, P > 0.05). Isoproterenol, which activates I-1, has been reported to diminish the contractile response to KCl administration in a manner different from agonist-induced contractions. This is based on the observation that the desensitization occurs in the absence of alterations in intracellular Ca2+ concentrations or sensitivity of MLCK to Ca2+-CaM (Tang et al. 1992). KCl-induced contractions following isoproterenol pretreatment (300 nm) did result in desensitization to KCl (rightward shift), but were similar in WT and I-1(−/−) aorta, suggesting that I-1 does not play a role in this desensitization (Fig. 3B). The TC50 values were not different between the WT and I-1(−/−) aorta (94.2 ± 10.6 vs. 97.4 ± 13.4 s, n = 5, P > 0.05).

Figure 3. KCl concentration-isometric force relations.

Figure 3

Open in a new tab

Responsiveness to KCl was assessed in WT (•, n = 4) and I-1(−/−) (□, n = 4) denuded aortas, in parallel. A, KCl did not affect the absolute force per area generated in either group (P > 0.05). B, percentage maximal response was also not different between the two groups. Pretreatment with isoproterenol (dotted lines) resulted in a rightward shift of the concentration curves in a similar manner in WT (▿) and I-1(−/−) (▵) aorta (P > 0.05). Values represent means ±s.e.m.

I-1 in aortic relaxation

I-1 is a substrate for PKA (Huang & Glinsmann, 1976) and PKG (Tokui et al. 1996), both of which invoke smooth muscle relaxation. Based on these results, we examined whether I-1 might play a role in the response to vasodilators. To assess the role of I-1 in relaxation, aortas were contracted with an EC50 concentration of PE (300 nm) and relaxed with increasing concentrations of either forskolin (PKA mediated) or sodium nitroprusside (SNP; PKG mediated). If the myosin phosphatase is more active in the I-1(−/−) aortas, they should be sensitized to relaxation. Relaxation in response to forskolin was nearly identical between the WT and I-1(−/−) aortas (Fig. 4A). Interestingly, a small, but significant, decrease in the time to half-maximal relaxation (TR50) in response to forskolin was observed in the I-1(−/−) aortas (WT 173.7 ± 15.1 s vs. I-1(−/−) 103.7 ± 12.5 s, n = 9, P < 0.05). Relaxation in response to SNP was also not different between WT and I-1(−/−) aorta (Fig. 4B). Additionally, SNP was without effect on the rates of relaxation (TR50) in WT and I-1(−/−) aorta (39.0 ± 4.3 vs. 36.3 ± 2.7 s, n = 6, P > 0.05). In addition to the relaxant effects of forskolin and SNP, we further examined whether I-1 was involved in relaxation in response to ACh. WT and I-1(−/−) aortas responded in a similar manner to increasing ACh concentrations (data not shown).

Figure 4. PKA-mediated (forskolin) and PKG-mediated (SNP) concentration-relaxation relations.

Figure 4

Open in a new tab

Denuded aortas from WT (•, n = 4) and I-1(−/−) (□, n = 4) mice were contracted with a submaximal concentration of PE and subsequently relaxed with increasing concentrations of forskolin (A) or sodium nitroprusside (B0. Relaxation in response to either agonist, expressed as a percentage of PE-induced contraction, was nearly identical between the two groups (P > 0.05). Values represent means ±s.e.m.

MLC20 dephosphorylation rates in WT and I-1(−/−) aortic extracts

Although I-1 is present in the aorta (Fig. 1), no obvious effects of I-1 ablation were evident in aortic contraction or relaxation. Thus, we directly examined myosin phosphatase activity by examining rates of MLC20 dephosphorylation in aortic homogenates from WT and I-1(−/−) mice, in the presence of 1 μm forskolin. Forskolin was included in this assay in order to fully phosphorylate (activate) I-1, maximizing the inhibitory activity of I-1. The phosphorylation status of I-1 was maintained during the assay by inclusion of the phosphatase inhibitors okadaic acid (2 nm) and deltamethrin (50 nm) to inhibit type-2A and type-2B phosphatase, respectively, the two classes of phosphatases responsible for dephosphorylating (inactivating) I-1 (Oliver & Shenolikar, 1998). The time courses of MLC20 dephosphorylation, measured by the release of 32P from radiolabelled myosin light chains, were similar between the WT and I-1(−/−) extracts in the presence of forskolin (Fig. 5), suggesting that I-1 is not regulating myosin phosphatase activity.

Figure 5. MLC20 dephosphorylation rates.

Figure 5

Open in a new tab

32P-labelled MLC20 dephosphorylation rates were measured in aortic extracts from WT (•, n = 5) and I-1(−/−) (□, n = 5) mice treated with 50 μm forskolin. Type-2A (2 nm okadaic acid) and type-2B (50 nm deltamethrin) phosphatase inhibitors were included to maintain the phosphorylation status of I-1 during the assay. Values represent means ±s.e.m. of duplicate experiments.

Role of I-1 in portal vein smooth muscle

Phasic smooth muscles are distinct from tonic smooth muscles in several of their basic intrinsic properties. Thus, we investigated the effects of I-1 ablation in the spontaneously contracting portal veins from WT and I-1(−/−) mice under basal conditions and in response to a variety of pharmacological agents. Portal veins from WT and I-1(−/−) mice were indistinguishable by their gross anatomical parameters (Table 1). If I-1 is involved in the inhibition of phosphatase activity in portal vein, increased phosphatase activity would result in decreased force generation and rate of force development (+dF/dt), while the rate of force decline (−dF/dt) would be accelerated in the I-1(−/−) preparations. Although values for the amplitude of force generation were decreased in I-1(−/−) mice, the tension-time integral, ±dF/dt, or frequency in spontaneously contracting portal veins from WT and I-1(−/−) mice were similar (Table 2).

Table 2.

Parameters of portal vein contractility from WT and I-1(−/−) mice

Parameters WT I-1(−/−)
Amplitude of force (mN mm−2) 20.24±1.53 15.27±1.36*
Tension–time integral (mN s mm−2) 245.1±21.1 195.5±20.0
+dF/dt(mN s−1 mm−2) 62.21±4.41 67.37±9.09
−dF/dt(mN s−1 mm−2) −42.99±2.04 −45.89±4.44
Frequency (min−1) 2.46±0.15 2.17±0.12
Isoproterenol EC50(nm 3.5±0.2 10.4±1.4*
Open in a new tab

Baseline measurements were obtained from 10 WT and 10 I-1(−/−) isometrically mounted portal veins, in parallel. ±dF/dt values were obtained from 4 WT and 4 I-1(−/−) portal veins. Parameters are listed on the left with units of measure listed in parentheses. Baseline values are expressed as means ± s.e.m. Statistical significance (Student's unpaired test) is indicated by an asterisk.

*

P < 0.05.

To determine the effects of β-adrenergic stimulation in WT and I-1(−/−) portal veins, concentration-relaxation relations were examined in response to isoproterenol. Following isoproterenol administration, portal vein contractility, expressed as a percentage of the basal tension-time integral, was decreased in both the WT and I-1(−/−) preparations (Fig. 6). Although the maximal relaxation in response to isoproterenol was similar between WT and I-1(−/−) portal vein, the concentration required to achieve half-relaxation (EC50) was significantly higher in I-1(−/−) portal vein (Table 2), indicating a desensitized response to β-adrenergic stimulation in the I-1(−/−) mice. In addition, portal vein Ca2+ sensitivity was measured by depolarizing the portal veins and exposing them to increasing concentrations of Ca2+. If I-1 is involved in a Ca2+-sensitizing effect, through inhibition of MLCP, uninhibited myosin phosphatase in the I-1(−/−) portal vein should result in a rightward shift in the Ca2+ concentration-isometric force relations. Ca2+-elicited contractions were similar between the two groups (Fig. 7A). Pretreatment with isoproterenol (300 nm) resulted in a near-parallel Ca2+ desensitization of portal veins, as previously observed in response to cGMP (Lee et al. 1997), in both WT and I-1(−/−) portal veins. Finally, the effect of I-1 gene ablation on ACh-mediated contraction of portal vein preparations was examined. The tension-time integral for portal veins from WT and I-1(−/−) mice increased with increasing concentrations of ACh. However, no differences in ACh responsiveness were observed between the WT and I-1(−/−) portal veins (Fig. 7B.

Figure 6. Concentration-tension-time integral relations in response to isoproterenol in WT and I-1(−/−) portal vein.

Figure 6

Open in a new tab

Spontaneous mechanical activity in response to isoproterenol was measured in WT (•, n = 10) and I-1(−/−) (□, n = 10) portal veins. The concentration of isoproterenol resulting in 50 % relaxation was significantly rightward shifted (P < 0.05) in I-1(−/−) portal veins, compared to the WT. Values represent means ±s.e.m.

Figure 7. Percentage maximal force or tension-time integral from portal vein in response to increased calcium or ACh concentrations.

Figure 7

Open in a new tab

Portal vein contraction in response to calcium (A) or ACh (B) was measured in WT (•, n = 5) and I-1(−/−) (□, n = 5) mice. Portal vein contractility in WT and I-1(−/−) mice was similar at every calcium concentration (P > 0.05). Pretreatment with isoproterenol (dotted lines), to fully activate I-1, resulted in a parallel rightward shift in the calcium dose-response curve in both the WT (▿) and I-1(−/−) (▵) portal veins. Contraction in response to ACh, expressed as a percentage of basal values, increased in a comparable manner in both WT (•, n = 5) and I-1(−/−) (□, n = 5) portal vein (P > 0.05). Values represent means ±s.e.m.

DISCUSSION

The discovery that MLCP activity, and consequently MLC20 dephosphorylation, could be regulated in response to various agents has initiated a search for novel proteins involved in modulating the MLCP. In this study, we investigated the role of protein phosphatase inhibitor-1 (I-1), a known inhibitor of the type-1 phosphatase, in tonic (aorta) and phasic (portal vein) smooth muscle contraction and relaxation. I-1 has been shown to regulate myosin phosphatase activity in vitro and I-1 gene-targeted mice provide us with the opportunity to directly examine the role of I-1 in smooth muscle contractility in vivo.

Our original hypothesis was that I-1 plays an important role in the regulation of the smooth muscle phosphatase activity, specifically of the myosin-associated phosphatase, and thus is a regulator of smooth muscle contractility. This hypothesis was based on the findings that I-1 can regulate MLCP in vitro and that other inhibitors of myosin phosphatase, such as Rho-kinase and CPI-17, are important in regulating smooth muscle contractility (Somlyo & Somlyo, 2000). In addition, I-1 has been suggested to play a role in the Ca2+-sensitization observed in response to agonists such as cGMP (Somlyo et al. 1989). The major finding of this study is that, although I-1 was detected in smooth muscle (Fig. 1A) and phosphorylated by PKA (Fig. 1B), ablation of I-1 did not have any major affects on either aorta or portal vein contractility in vivo. These findings are supported by nearly identical MLC20 dephosphorylation rates in WT and I-1(−/−) aorta. Although the in vivo substrate for the phosphatase is native myosin, these studies on the dephosphorylation of MLC20 strongly suggest that the phosphatase activities in the WT and I-1(−/−) mice are equivalent. The absence of a difference between WT and I-1(−/−) tissues was not due to inactive (dephosphorylated) I-1 in the WT, since experiments performed in the presence of known activators of I-1 (isoproterenol/ forskolin) also showed no effect.

Relaxation induced by PKA or PKG, both of which activate I-1, was also unaffected by I-1 ablation. These data demonstrate that I-1 is not an essential component of smooth muscle contractility under basal conditions or following β-adrenergic stimulation in the aorta. We did observe a small, but significant, rightward-shifted (desensitized) response to isoproterenol in I-1(−/−) portal veins. This finding was unique to the portal vein and is not consistent with I-1 inhibiting type-1 phosphatase activity against myosin. At least two potential mechanisms could explain this finding. In the absence of I-1, increased phosphatase activity and the resulting dephosphorylation of the inhibitory phosphorylation site on MLCK (Oliver & Shenolikar, 1998) would increase MLCK activity, impairing relaxation. Alternatively, I-1 could regulate phosphatase activity in the sarcoplasmic reticulum, as described for striated muscle (MacDougall et al. 1991). If so, increased SR phosphatase activity would result in activation (dephosphorylation) of the inhibitor of the SR Ca2+-ATPase phospholamban, leading to slower sequestration of Ca2+ (Lalli et al. 1999). Although the physiological significance of this finding is not clear, these data suggest that I-1 may exist as a fine-tuning mechanism in response to β-catecholamines in the portal vein. In addition, we also observed a rightward shift in the Ca2+ response following addition of isoproterenol (desensitized) in WT portal vein, as previously described for cGMP (Lee et al. 1997). However, the I-1(−/−) portal vein curve was shifted in a similar manner, excluding I-1 as a component of the Ca2+-desensitizing effect.

Collectively, our data suggest that I-1 does not play a major role in either tonic or phasic smooth muscle contractility in vivo. The non-effect of I-1 ablation on MLC20 dephosphorylation rates between WT and I-1(−/−) aorta supports in vitro studies (Alessi et al. 1992), which demonstrated only a modest effect of I-1 on inhibiting the myosin phosphatase holoenzyme activity in avian smooth muscle (< 30 %; Alessi et al. 1992). The lack of effect of I-1 ablation on either aorta or portal vein contractility compels us to speculate on the relevance of this protein in smooth muscle, given its relative abundance. A role for I-1 in smooth muscle may be in regulating enzymes involved in glycogen metabolism (Cohen, 1989). However, despite numerous in vitro reports demonstrating a role for I-1 in this process, studies from skeletal muscles of I-1(−/−) mice showed no effect on the activities of glycogen synthase and glycogen phosphorylase (Scrimgeour et al. 1999), challenging I-1′s role in this process. Alternatively, there are several lines of evidence that suggest the presence of another myosin phosphatase (Pato & Adelstein, 1980; Erdodi et al. 1989; Takai et al. 1989; Nomura et al. 1992), probably a type-2A enzyme. In this case, inhibition of the myosin type-1 phosphatase by I-1 may represent only a minor fraction of the total myosin phosphatase activity. With respect to the portal vein, the rightward shift in the isoproterenol-induced relaxation in I-1(−/−) tissues suggests that I-1 may serve as a fine-tuning mechanism in regulating contractility in response to β-adrenergic agonists. The difference between the PKA effects of I-1 protein ablation in aorta and portal vein may reflect inherent differences in the myosin-targeting subunits or catalytic subunits themselves in each tissue (Hartshorne, 1998; Hartshorne et al. 1998). In the light of the significant amount of I-1 present, other functions for I-1 in vascular smooth muscle are a strong possibility, but remain elusive.

Acknowledgments

This work was supported by National Institutes of Health Grants HL54829, HL64942, (R.J.P.), HL59618 (P.d.L.), HL26057, HL52318, P40 RR12358, HL64018 (E.G.K.), HL07382 (A.N.C.) and MH40899 and DA 10044 (P.G.). We kindly acknowledge Peggy Bowman for excellent advice and technical assistance.

Andrew N. Carr and Roy L. Sutliff contributed equally to this work.

References

  1. Alessi D, MacDougall LK, Sola MM, Ikebe M, Cohen P. The control of protein phosphatase-1 by targetting subunits. The major myosin phosphatase in avian smooth muscle is a novel form of protein phosphatase-1. European Journal of Biochemistry. 1992;210:1023–1035. doi: 10.1111/j.1432-1033.1992.tb17508.x. [DOI] [PubMed] [Google Scholar]
  2. Allen PB, Hvalby O, Jensen V, Errington ML, Ramsay M, Chaudhry FA, Bliss TV, Storm-Mathisen J, Morris RG, Andersen P, Greengard P. Protein phosphatase-1 regulation in the induction of long-term potentiation: heterogeneous molecular mechanisms. Journal of Neuroscience. 2000;20:3537–3543. doi: 10.1523/JNEUROSCI.20-10-03537.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Cohen P. The structure and regulation of protein phosphatases. Annual Review of Biochemistry. 1989;58:453–508. doi: 10.1146/annurev.bi.58.070189.002321. [DOI] [PubMed] [Google Scholar]
  4. Conti MA, Adelstein RS. The relationship between calmodulin binding and phosphorylation of smooth muscle myosin kinase by the catalytic subunit of 3′:5′ cAMP-dependent protein kinase. Journal of Biological Chemistry. 1981;256:3178–3181. [PubMed] [Google Scholar]
  5. de Lanerolle P, Paul RJ. Myosin phosphorylation/ dephosphorylation and regulation of airway smooth muscle contractility. American Journal of Physiology. 1991;261:L1–14. doi: 10.1152/ajplung.1991.261.2.L1. [DOI] [PubMed] [Google Scholar]
  6. Elbrecht A, Direnzo J, Smith RG, Shenolikar S. Molecular cloning of protein phosphatase inhibitor-1 and its expression in rat and rabbit tissues. Journal of Biological Chemistry. 1990;265:13415–13418. [PubMed] [Google Scholar]
  7. Erdodi F, Rokolya A, Barany M, Barany K. Dephosphorylation of distinct sites in myosin light chain by two types of phosphatase in aortic smooth muscle. Biochimica et Biophysica Acta. 1989;1011:67–74. doi: 10.1016/0167-4889(89)90080-3. [DOI] [PubMed] [Google Scholar]
  8. Feng J, Ito M, Ichikawa K, Isaka N, Nishikawa M, Hartshorne DJ, Nakano T. Inhibitory phosphorylation site for Rho-associated kinase on smooth muscle myosin phosphatase. Journal of Biological Chemistry. 1999;274:37385–37390. doi: 10.1074/jbc.274.52.37385. [DOI] [PubMed] [Google Scholar]
  9. Fujita A, Takeuchi T, Nakajima H, Nishio H, Hata F. Involvement of heterotrimeric GTP-binding protein and rho protein, but not protein kinase C, in agonist-induced Ca2+ sensitization of skinned muscle of guinea pig vas deferens. Journal of Pharmacology and Experimental Therapeutics. 1995;274:555–561. [PubMed] [Google Scholar]
  10. Gallagher PJ, Herring BP, Stull JT. Myosin light chain kinases. Journal of Muscle Research and Cell Motility. 1997;18:1–16. doi: 10.1023/a:1018616814417. [DOI] [PubMed] [Google Scholar]
  11. Gong MC, Kinter MT, Somlyo AV, Somlyo AP. Arachidonic acid and diacylglycerol release associated with inhibition of myosin light chain dephosphorylation in rabbit smooth muscle. Journal of Physiology. 1995;486:113–122. doi: 10.1113/jphysiol.1995.sp020795. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Hartshorne DJ. Myosin phosphatase: subunits and interactions. Acta Physiologica Scandinavica. 1998;164:483–493. doi: 10.1046/j.1365-201X.1998.00447.x. [DOI] [PubMed] [Google Scholar]
  13. Hartshorne DJ, Ito M, Erdodi F. Myosin light chain phosphatase: subunit composition, interactions and regulation. Journal of Muscle Research and Cell Motility. 1998;19:325–341. doi: 10.1023/a:1005385302064. [DOI] [PubMed] [Google Scholar]
  14. Hemmings HC, Jr, Nairn AC, Greengard P. DARPP-32, a dopamine- and adenosine 3′:5′-monophosphate-regulated neuronal phosphoprotein. II. Comparison of the kinetics of phosphorylation of DARPP-32 and phosphatase inhibitor 1. Journal of Biological Chemistry. 1984;259:14491–14497. [PubMed] [Google Scholar]
  15. Huang FL, Glinsmann WH. Separation and characterization of two phosphorylase phosphatase inhibitors from rabbit skeletal muscle. European Journal of Biochemistry. 1976;70:419–426. doi: 10.1111/j.1432-1033.1976.tb11032.x. [DOI] [PubMed] [Google Scholar]
  16. Hubbard MJ, Cohen P. Regulation of protein phosphatase-1G from rabbit skeletal muscle. 2. Catalytic subunit translocation is a mechanism for reversible inhibition of activity toward glycogen-bound substrates. European Journal of Biochemistry. 1989;186:711–716. doi: 10.1111/j.1432-1033.1989.tb15264.x. [DOI] [PubMed] [Google Scholar]
  17. Ichikawa K, Hirano K, Ito M, Tanaka J, Nakano T, Hartshorne DJ. Interactions and properties of smooth muscle myosin phosphatase. Biochemistry. 1996;35:6313–6320. doi: 10.1021/bi960208q. [DOI] [PubMed] [Google Scholar]
  18. Kimura K, Ito M, Amano M, Chihara K, Fukata Y, Nakafuku M, Yamamori B, Feng J, Nakano T, Okawa K, Iwamatsu A, Kaibuchi K. Regulation of myosin phosphatase by Rho and Rho-associated kinase (Rho-kinase) Science. 1996;273:245–248. doi: 10.1126/science.273.5272.245. [DOI] [PubMed] [Google Scholar]
  19. Lalli J, Harrer JM, Luo W, Kranias EG, Paul RJ. Targeted ablation of the phospholamban gene is associated with a marked decrease in sensitivity in aortic smooth muscle. Circulation Research. 1997;80:506–513. doi: 10.1161/01.res.80.4.506. [DOI] [PubMed] [Google Scholar]
  20. Lalli MJ, Shimizu S, Sutliff RL, Kranias EG, Paul RJ. [Ca2+]i homeostasis and cyclic nucleotide relaxation in aorta of phospholamban-deficient mice. American Journal of Physiology. 1999;277:H963–670. doi: 10.1152/ajpheart.1999.277.3.H963. [DOI] [PubMed] [Google Scholar]
  21. Lee MR, Li L, Kitazawa T. Cyclic GMP causes Ca2+ desensitization in vascular smooth muscle by activating the myosin light chain phosphatase. Journal of Biological Chemistry. 1997;272:5063–5068. doi: 10.1074/jbc.272.8.5063. [DOI] [PubMed] [Google Scholar]
  22. Li L, Eto M, Lee MR, Morita F, Yazawa M, Kitazawa T. Possible involvement of the novel CPI-17 protein in protein kinase C signal transduction of rabbit arterial smooth muscle. Journal of Physiology. 1998;508:871–881. doi: 10.1111/j.1469-7793.1998.871bp.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. MacDougall LK, Jones LR, Cohen P. Identification of the major protein phosphatases in mammalian cardiac muscle which dephosphorylate phospholamban. European Journal of Biochemistry. 1991;196:725–734. doi: 10.1111/j.1432-1033.1991.tb15871.x. [DOI] [PubMed] [Google Scholar]
  24. Mitsui T, Inagaki M, Ikebe M. Purification and characterization of smooth muscle myosin-associated phosphatase from chicken gizzards. Journal of Biological Chemistry. 1992;267:16727–16735. [PubMed] [Google Scholar]
  25. Nomura M, Stull JT, Kamm KE, Mumby MC. Site-specific dephosphorylation of smooth muscle myosin light chain kinase by protein phosphatases 1 and 2A. Biochemistry. 1992;31:11915–11920. doi: 10.1021/bi00162a034. [DOI] [PubMed] [Google Scholar]
  26. Oliver CJ, Shenolikar S. Physiologic importance of protein phosphatase inhibitors. Frontiers in Biosciences. 1998;3:D961–972. doi: 10.2741/a336. [DOI] [PubMed] [Google Scholar]
  27. Pato MD, Adelstein RS. Dephosphorylation of the 20,000-dalton light chain of myosin by two different phosphatases from smooth muscle. Journal of Biological Chemistry. 1980;255:6535–6538. [PubMed] [Google Scholar]
  28. Scrimgeour AG, Allen PB, Fienberg AA, Greengard P, Lawrence JC., Jr Inhibitor-1 is not required for the activation of glycogen synthase by insulin in skeletal muscle. Journal of Biological Chemistry. 1999;274:20949–20952. doi: 10.1074/jbc.274.30.20949. [DOI] [PubMed] [Google Scholar]
  29. Shirazi A, Iizuka K, Fadden P, Mosse C, Somlyo AP, Somlyo AV, Haystead TA. Purification and characterization of the mammalian myosin light chain phosphatase holoenzyme. The differential effects of the holoenzyme and its subunits on smooth muscle. Journal of Biological Chemistry. 1994;269:31598–31606. [PubMed] [Google Scholar]
  30. Solaro RJ. Myosin light chain phosphatase: a Cinderella of cellular signaling. Circulation Research. 2000;87:173–175. doi: 10.1161/01.res.87.3.173. [DOI] [PubMed] [Google Scholar]
  31. Somlyo AP, Kitazawa T, Himpens B, Matthijs G, Horiuti K, Kobatashi S, Goldman YE, Somlyo AV. Modulation of Ca2+-sensitivity and of the time course of contraction in smooth muscle: a major role of protein phosphatases? Advances in Protein Phosphatases. 1989;5:181–195. [Google Scholar]
  32. Somlyo AP, Somlyo AV. Signal transduction by G-proteins, rho-kinase and protein phosphatase to smooth muscle and non-muscle myosin II. Journal of Physiology. 2000;522:177–185. doi: 10.1111/j.1469-7793.2000.t01-2-00177.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Sutliff RL, Weber CS, Qian J, Miller ML, Clemens TL, Paul RJ. Vasorelaxant properties of parathyroid hormone-related protein in the mouse: evidence for endothelium involvement independent of nitric oxide formation. Endocrinology. 1999;140:2077–2083. doi: 10.1210/endo.140.5.6700. [DOI] [PubMed] [Google Scholar]
  34. Takai A, Troschka M, Mieskes G, Somlyo AV. Protein phosphatase composition in the smooth muscle of guinea-pig ileum studied with okadaic acid and inhibitor 2. Biochemical Journal. 1989;262:617–623. doi: 10.1042/bj2620617. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Tang DC, Stull JT, Kubota Y, Kamm KE. Regulation of the Ca2+ dependence of smooth muscle contraction. Journal of Biological Chemistry. 1992;267:11839–11845. [PubMed] [Google Scholar]
  36. Tokui T, Brozovich F, Ando S, Ikebe M. Enhancement of smooth muscle contraction with protein phosphatase inhibitor 1: activation of inhibitor 1 by cGMP-dependent protein kinase. Biochemical and Biophysical Research Communications. 1996;220:777–783. doi: 10.1006/bbrc.1996.0480. [DOI] [PubMed] [Google Scholar]
  37. Toth A, Kiss E, Gergely P, Walsh MP, Hartshorne DJ, Erdodi F. Phosphorylation of MYPT1 by protein kinase C attenuates interaction with PP1 catalytic subunit and the 20 kDa light chain of myosin. FEBS Letters. 2000;484:113–117. doi: 10.1016/s0014-5793(00)02138-4. [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES