Pressure-dependent contribution of Rho kinase-mediated calcium sensitization in serotonin-evoked vasoconstriction of rat cerebral arteries

Abstract

Our understanding of the cellular signalling mechanisms contributing to agonist-induced constriction is almost exclusively based on the study of conduit arteries. Resistance arteries/arterioles have received less attention as standard biochemical approaches lack the necessary sensitivity to permit quantification of phosphoprotein levels in these small vessels. Here, we have employed a novel, highly sensitive Western blotting method to assess: (1) the contribution of Ca²⁺ sensitization mediated by phosphorylation of myosin light chain phosphatase targeting subunit 1 (MYPT1) and the 17 kDa PKC-potentiated protein phosphatase 1 inhibitor protein (CPI-17) to serotonin (5-HT)-induced constriction of rat middle cerebral arteries, and (2) whether there is any interplay between pressure-induced myogenic and agonist-induced mechanisms of vasoconstriction. Arterial diameter and levels of MYPT1 (T697 and T855), CPI-17 and 20 kDa myosin light chain subunit (LC₂₀) phosphorylation were determined following treatment with 5-HT (1 μmol l⁻¹) at 10 or 60 mmHg in the absence and presence of H1152 or GF109203X to suppress the activity of Rho-associated kinase (ROK) and protein kinase C (PKC), respectively. Although H1152 and GF109203X suppressed 5-HT-induced constriction and reduced phospho-LC₂₀ content at 10 mmHg, we failed to detect any increase in MYPT1 or CPI-17 phosphorylation. In contrast, an increase in MYPT1-T697 and MYPT1-T855 phosphorylation, but not phospho-CPI-17 content, was apparent at 60 mmHg following exposure to 5-HT, and the phosphorylation of both MYPT1 sites was sensitive to H1152 inhibition of ROK. The involvement of MYPT1 phosphorylation in the response to 5-HT at 60 mmHg was not dependent on force generation per se, as inhibition of cross-bridge cycling with blebbistatin (10 μmol l⁻¹) did not affect phosphoprotein content. Taken together, the data indicate that Ca²⁺ sensitization owing to ROK-mediated phosphorylation of MYPT1 contributes to 5-HT-evoked vasoconstriction only in the presence of pressure-induced myogenic activation. These findings provide novel evidence of an interplay between myogenic- and agonist-induced vasoconstriction in cerebral resistance arteries.

Abbreviations 

CPI-17

17 kDa PKC-potentiated protein phosphatase 1 inhibitor protein

5-HT

serotonin

MLCK

myosin light chain kinase

MLCP

myosin light chain phosphatase

LC₂₀

20 kDa myosin regulatory light chain subunits

MYPT1

myosin light chain phosphatase targeting subunit 1

PdBu

phorbol 12,13-dibutyrate

PKC

protein kinase C

RMCA

rat middle cerebral artery

ROK

Rho-associated kinase

VSMC

vascular smooth muscle cell

VGCC

voltage-gated Ca²⁺ channel

Introduction

Brain function and integrity are reliant on an appropriate level of blood flow within the cerebral circulation. Precise control of blood flow is achieved through the interplay of several physiological factors that regulate the contractile state of cerebrovascular smooth muscle cells and, thereby, arterial diameter. These factors include the inherent autoregulatory, myogenic response of the smooth muscle cells to changes in intra-luminal pressure, and extrinsic modulators, such as vasoactive molecules released from the endothelium, parenchyma, cells within the vessel lumen and nerve varicosities within adventitial connective tissue surrounding the vessels (Iadecola & Nedergaard, 2007; Kulik et al. 2008).

The myogenic response plays a major role in blood flow control in the cerebral vasculature, maintaining flow constant during fluctuations in blood pressure (Faraci & Heistad, 1990; Davis & Hill, 1999). The modulatory actions of extrinsic factors are superimposed on this myogenic control to adjust flow to meet changing local requirements. 5-HT released from an extensive network of nerve varicosities surrounding cerebral vessels is an important extrinsic factor that contributes to the control of cerebral arterial diameter and modulation of regional blood flow (Cohen et al. 1996; Iadecola & Nedergaard, 2007). Moreover, abnormal 5-HT-induced vasoconstriction may contribute to the development of vasospasm following haemorrhagic stroke (Hansen-Schwartz et al. 2008). At present, however, we have a limited understanding of the cellular mechanism(s) by which 5-HT elicits cerebral vasoconstriction, particularly in the presence of physiological intra-luminal pressure.

A large body of evidence, based primarily on the study of conduit arteries, indicates that vasoconstrictor agonists elicit smooth muscle contraction principally by two distinct cellular mechanisms. Receptor occupancy can cause: (1) membrane potential (E_m) depolarization leading to Ca²⁺ influx through T- and L-type voltage-gated Ca²⁺ channels (VGCCs) and/or (2) Ca²⁺ release in the form of asynchronous, transient Ca²⁺ waves that elevate cytosolic free Ca²⁺ concentration ([Ca²⁺]_i) (e.g. Shaw et al. 2004). Force generation is subsequently initiated via Ca²⁺–calmodulin-dependent activation of myosin light chain kinase (MLCK) that phosphorylates myosin regulatory light chain subunits (LC₂₀) leading to actomyosin ATPase activation, cross-bridge cycling and force generation (Zou et al. 2000).

Alternatively, vasoconstrictor agonists are also known to increase force generation in the absence of changes in [Ca²⁺]_i through the inhibition of myosin light chain phosphatase (MLCP) (Somlyo & Somlyo, 2003). The concept that LC₂₀ phosphorylation and force generation are a function of the balance between the activities of MLCK and MLCP is fundamental to our understanding of smooth muscle contraction. Cellular signalling pathways activated by agonists that involve Rho-associated kinase (ROK) and protein kinase C (PKC) decrease MLCP activity, shifting the MLCK–MLCP balance to favour MLCK-dependent LC₂₀ phosphorylation and resulting in a leftward shift in the contractile force–[Ca²⁺]_i relationship (Somlyo & Somlyo, 2003, 2004; Swärd et al. 2003; Hirano, 2007). This mechanism is referred to as myofilament Ca²⁺ sensitization. Inhibition of MLCP activity has been shown to occur following phosphorylation of the MLCP targeting subunit 1 (MYPT1) by ROK at several sites, with MYPT1-T697 and MYPT1-T855 being the major sites in the rat. Phosphorylation at MYPT1-T697 inhibits MLCP activity (Feng et al. 1999) and phosphorylation at MYPT1-T855 interferes with the binding of MYPT1 to myosin (Velasco et al. 2002) and inhibits MLCP activity (Murányi et al. 2005). PKC-mediated phosphorylation of the 17 kDa PKC-potentiated protein phosphatase 1 inhibitor protein (CPI-17) increases the intrinsic inhibitory effect of CPI-17 on MLCP by 1000-fold (Hayashi et al. 2001; Somlyo & Somlyo, 2003; Swärd et al. 2003; Dimopoulos et al. 2007). Contractile force generation owing to MLCK activation and MLCP inhibition may be further modulated by mechanisms affecting cross-bridge cycling (i.e. thin filament regulation) (Jin et al. 2000; Kaneko et al. 2000; Morgan & Gangopadhyay, 2001) and/or dynamic cytoskeleton reorganization (Cipolla et al. 2002).

The mechanisms by which 5-HT evokes cerebral vasoconstriction have not been examined in detail. Early studies suggested a role for depolarization-induced Ca²⁺ entry in cerebral vascular smooth muscle cells (VSMCs) leading to contraction; e.g. treating cat or rabbit basilar arterial rings with 0.01–3 μmol l⁻¹ 5-HT evoked a concentration-dependent depolarization of up to 30 mV (Harder & Waters, 1983; Garland, 1987). More recent studies have considered the role of Ca²⁺ sensitization, but a clear picture has not emerged. Sandoval et al. (2007) concluded that Ca²⁺ sensitization mediated by a ROK-dependent mechanism contributes to 5-HT-induced contraction of ovine carotid arterial rings. However, Watanabe et al. (2005) failed to detect an increase in MYPT1 phosphorylation in rabbit basilar arterial rings, although the contraction induced by 5-HT was abolished by the ROK inhibitor, fasudil. Elevated phosphorylation of MYPT1-S854 and MYPT1-T853 in canine and rabbit basilar arteries (the latter equivalent to MYPT1-T855 in rat) was detected in models of experimental subarachnoid haemorrhage, but there was no change in phospho-MYPT1 content associated with 5-HT-evoked contraction under control conditions (Sato et al. 2000; Watanabe et al. 2005). The reasons for these varied observations are not known.

The absence of physiological intra-luminal pressure is a potential limitation to understanding the contribution of Ca²⁺ sensitization in 5-HT-induced constriction in previous studies of the cerebral vasculature. It was previously demonstrated that the myogenic response is altered in the presence of vasoconstrictors, such that agonist treatment was found to induce myogenic behaviour (Harder, 1988; Faber & Meininger, 1990; Meininger & Faber, 1991; VanBavel & Mulvany, 1994; Anschütz & Schubert, 2005). The reverse also occurs, in that the sensitivity to vasoconstrictors can be affected by the presence of myogenic tone (Harder, 1988; Lombard et al. 1990; Meininger & Faber, 1991; VanBavel & Mulvany, 1994). The molecular basis of this interplay is unknown, but may be due to changes in the length–tension relationship or an alteration in ionic or biochemical signalling mechanisms in the VSMCs (Meininger & Faber, 1991). Determining the nature of the biochemical mechanisms contributing to agonist-induced force generation using pressurized resistance vessels is difficult owing to their small size. Standard Western blotting methods lack the sensitivity to detect the trace quantities of phosphoprotein present in the 0.5–1 mm segments of resistance arteries commonly employed in pressure myography experiments. For this reason larger conduit arteries have been the preparation of choice for analysis of the mechanisms contributing to agonist-induced vasoconstriction. However, the applicability of the findings of studies of conduit vessels to pressurized resistance arteries is unknown.

We recently employed a novel three-step Western blot method of sufficient sensitivity to permit accurate quantification of MYPT1, CPI-17 and LC₂₀ phosphorylation in pressurized segments of rat middle cerebral arteries (RMCAs). This approach was employed to identify the contribution of ROK-mediated phosphorylation of MYPT1-T855 and Ca²⁺ sensitization in the myogenic response to steps in intra-luminal pressure from 10 to 60 or 100 mmHg that are within the physiological intra-luminal pressure range (Johnson et al. 2009). Thus, analysis of the biochemical mechanisms contributing to agonist-induced constriction of resistance arteries in the presence of physiological intra-luminal pressure is now possible. The aims of the present study were two-fold: (1) to assess the contribution of Ca²⁺ sensitization mediated by MYPT1 and CPI-17 phosphorylation to 5-HT-induced constriction of RMCAs, and (2) to determine whether there is any interplay between myogenic- and agonist-induced mechanisms of constriction via a comparison of the role of Ca²⁺ sensitization to 5-HT-evoked constriction at intra-luminal pressures of 10 and 60 mmHg.

Methods

Ethical approval

All animal experiments were conducted according to a protocol reviewed by the Animal Care Committee of the Faculty of Medicine, University of Calgary and conforming to the standards of the Canadian Council on Animal Care and The Journal of Physiology's ethical policies and regulations as outlined in Drummond (2009). A total of 170 rats were used in the completion of this study.

Intact cerebral arterial pressure myography

Male Sprague–Dawley rats (250–275 g; Charles River, Montréal, Quebec, Canada) were killed by halothane inhalation followed by exsanguination. The brain was removed and transferred to ice-cold Krebs solution of the following composition (in mmol l⁻¹): NaCl 120, NaHCO₃ 25, KCl 4.8, NaH₂PO₄ 1.2, MgSO₄ 1.2, glucose 11, CaCl₂ 1.8 (pH 7.4 when aerated with 95% air–5% CO₂). Rat middle cerebral arteries (RMCAs) from both hemispheres were removed, dissected free of the surrounding connective tissue, cut into segments of 2–3 mm in length and employed in pressure myography, as previously described (Chen et al. 2006; Johnson et al. 2009). Briefly, the arterial segments were cannulated and mounted in a myograph chamber connected to a pressure myograph (Living Systems, Burlington, VT, USA) and external arterial diameter measured by edge detection (IonOptix, Milton, MA, USA). The endothelial layer was disrupted by passing a stream of fine air bubbles through the vessel lumen and confirmed by the loss of vasodilatation to 10 μm bradykinin. Arteries were equilibrated in warm Krebs solution (37 ± 0.5°C) in the absence of intra-luminal pressure for 20–30 min at the beginning of each experiment.

Intra-luminal pressure protocol

Intra-luminal pressure was initially raised to 10 mmHg for 10–20 min after an initial equilibration period at 0 mmHg. Constrictions evoked by 5-HT and phosphoprotein quantifications were performed using vessels at 10 and 60 mmHg. Vessels studied at 10 mmHg were subjected to a further 10 min equilibration at 10 mmHg before the application of 5-HT and subsequent freezing to permit quantification of phospho-LC₂₀, phospho-MYPT1 and phospho-CPI-17 content when a stable level of constriction was achieved (at 3 to 5 min). These vessels were not subjected to conditioning pressure steps to avoid any possibility of sustained changes in phosphoprotein content owing to activation of a myogenic response. Vessel health was confirmed by the presence of a robust constriction to 5-HT. Vessels to be assessed at 60 mmHg were stepped to 60 mmHg for ∼15 min to permit stable development of myogenic constriction prior to the application of two 5 min steps from 10 to 80 mmHg. These steps to 80 mmHg ensured that all vessels had a similar level of myogenic response. Arteries that exhibited leaks or a lack of myogenic constriction during these test steps were discarded. The proportion of vessels showing tone was above 85%. Pressure was then set to 60 mmHg and the vessels were treated with 5-HT when a stable level of constriction was achieved. Some vessels at 10 and/or 60 mmHg were exposed to inhibitors of PKC (GF109203X), ROK (H1152) or cross-bridge cycling (blebbistatin) after 5-HT treatment (blebbistatin was also employed as a pretreatment prior to 5-HT in some experiments). For comparative purposes, two additional sets of arteries were pressurized to 140 mmHg from 10 mmHg to evaluate the changes in LC₂₀ and MYPT1 phosphorylation accompanying this extreme increase in intra-luminal pressure.

5-HT concentration–response protocol

Cumulative 5-HT concentration–response relations were determined for arteries at 10 and 60 mmHg. 5-HT was applied over the range of 0.001–10 μmol l⁻¹ in half-log increments; each 5-HT concentration was applied at a stable level of constriction. The effect of 5-HT on arterial diameter was measured as a percentage of maximum constriction for each tissue and fitted by a sigmoidal concentration–response relation using GraphPad Prism 4 (GraphPad Software, San Diego, CA, USA) to determine pEC₅₀ values.

Experimental protocols for analysis of phosphoprotein content

Six sets of vessels were exposed to 5-HT ± inhibitors at 10 or 60 mmHg followed by rapid freezing and quantification of phospho-MYPT1, phospho-CPI-17 and phospho-LC₂₀ content. Sister vessel segments were always employed in different treatment groups to minimize vessel-to-vessel variability in phosphoprotein content. Flash-freezing was accomplished in an ice-cold mixture of 10% trichloroacetic acid and 10 mmol l⁻¹ dithiothreitol (DTT) in acetone. Segments were washed in acetone containing 10 mmol l⁻¹ of DTT, lyophilized overnight, and stored at −80°C. Prior to protein extraction, the cannulated ends were removed from each lyophilized vessel segment to avoid contamination from tissue not exposed to the test pressure.

Set 1: vessels were maintained at 10 mmHg and exposed to control saline or 5-HT at 1, 3, or 10 μmol l⁻¹ until a stable change in diameter was achieved and then frozen.

Set 2: vessels were maintained at 10 mmHg prior to the application of 1 μmol l⁻¹ 5-HT to evoke a stable vasoconstriction followed by a washout period of 30–45 min. Vessels were then exposed to (1) control saline, (2) the ROK inhibitor H1152 (0.5 μmol l⁻¹), or (3) the PKC inhibitor GF109203X (3 μmol l⁻¹) for 15 min prior to a second application of 1 μmol l⁻¹ 5-HT and freezing when a stable constriction was achieved (at 3 to 5 min).

Set 3: vessels were maintained at 10 mmHg prior to treatment with control saline or 0.5 μmol l⁻¹ H1152 for 25 min and then frozen.

Set 4: vessels were maintained at 60 mmHg and treated with control saline or 1 μmol l⁻¹ 5-HT for 3–5 min to permit a stable level of constriction to be achieved prior to freezing.

Set 5: vessels were maintained at 60 mmHg and 1 μmol l⁻¹ 5-HT was applied until a stable constriction was achieved. Half of the segments were frozen at peak constriction whereas their sister segments were treated with 0.5 μmol l⁻¹ H1152 until a stable level of vasodilatation was achieved and then frozen.

Set 6: vessels were treated in a manner identical to Set 5 except that blebbistatin (10 μmol l⁻¹) was applied after addition of 5-HT to inhibit cross-bridge cycling. In some experiments, blebbistatin was added before stepping the pressure up to 60 mmHg.

In addition, some vessels were exposed to 1 μmol l⁻¹ phorbol 12,13-dibutyrate (PdBu) to directly activate PKC and serve as a positive control for quantification of CPI-17 phosphorylation. The concentration of PdBu used produced an increase in LC₂₀ phosphorylation similar to that obtained with 1 μmol l⁻¹ 5-HT in vessel segments pressurized to 10 mmHg (unpublished observations). In sets 4–6, the difference in constriction between treatments was assessed by comparison of arterial diameter immediately prior to freezing. Diameter rather than active constriction was employed in these instances as we could not determine the extent of active myogenic constriction from measurements of the passive diameter in Ca²⁺-free saline owing to the freezing procedure.

Protein extraction

Forty-five microlitres of sample buffer (4% SDS, 100 mmol l⁻¹ DTT, 10% glycerol, 0.01% bromophenol blue, 60 mmol l⁻¹ Tris-HCl, pH 6.8) was added to two to three lyophilized vessel segments for a single n value. Samples were heated at 95°C and rotated overnight at 4°C prior to electrophoresis. Although a single vessel segment was sufficient to detect phosphoprotein, two to three segments per sample were used to ensure robust detection and quantification of phosphoprotein content particularly when each sample was used to quantify multiple phosphoproteins.

Western blotting

A three-step Western blotting protocol was used to quantify phospho-MYPT1 (at T697 and T855), phospho-CPI-17 and phospho-LC₂₀ content, as previously described in detail (Takeya et al. 2008; Johnson et al. 2009). Phospho-MYPT1-T697 and -MYPT1-T855 levels were determined by normalization to the level of Coomassie blue-stained actin in each sample. This approach was shown to be more accurate than stripping and reprobing the blots with pan-MYPT1 antibody for loading normalization (Johnson et al. 2009). Phospho-CPI-17 and phospho-LC₂₀ were separated from unphosphorylated protein by phosphate-affinity tag sodium dodecyl sulfate polyacrylamide gel electrophoresis (Phos-Tag SDS-PAGE) and quantified as a percentage of total CPI-17 and LC₂₀, respectively, as previously described (Takeya et al. 2008; Johnson et al. 2009).

Materials

All chemicals were purchased from Sigma (Oakville, ON, Canada) unless otherwise indicated. H1152 was obtained from Calbiochem (San Diego, CA, USA) and GF109203X from Biomol International (Plymouth Meeting, PA, USA). Tween 20, Coomassie Brilliant Blue-R250, TEMED, PVDF, and nitrocellulose membranes were from Bio-Rad Laboratories (Mississauga, ON, Canada). Rabbit polyclonal antibodies specific for MYPT1 phosphorylated at T697 (anti-MYPT1-T697) or T855 (anti-MYPT1-T855) and anti-CPI-17 were from Upstate USA (Charlottesville, VA, USA). Polyclonal rabbit anti-LC₂₀ was from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Biotin-conjugated goat anti-rabbit secondary antibody was from Chemicon International (Bellerica, MA, USA) and horseradish peroxidase (HRP)-conjugated streptavidin was from Pierce Biotechnology (Rockford, CT, USA). Phos-tag™ acrylamide was from NARD Institute Ltd. (Japan).

Statistical analysis

Values are presented as means ±s.e.m., and the n values indicate the number of experiments for a given treatment (for the biochemical experiments, vessel segments from at least 3 rats contributed to each sample (n value) analysed for phosphoprotein content). Statistical difference was determined using unpaired Student's t test, repeated measures ANOVA followed by Bonferoni's post hoc test, or ANOVA followed by Dunnett's multiple comparisons test. A P value of < 0.05 was considered to be statistically significant.

Results

5-HT-evoked vasoconstriction at 10 and 60 mmHg

Vasoconstriction of RMCAs in response to 5-HT treatment was first assessed by determination of the cumulative concentration–response relation in half-log units at 10 and 60 mmHg (Fig. 1A and B). Intra-luminal pressures of 10 and 60 mmHg were selected to mimic the basal state in the absence of myogenic force generation and a state of physiological pressurization for RMCAs of 55–65 mmHg, respectively. The latter value was based on a mean arterial pressure of ∼110 mmHg in the rat (Yoshino et al. 2009) and a reduction of 40–50% from aortic pressure at the level of the cerebral arteries (Heistad, 2001). To minimize control and test sample variability, we employed pairs of vessel segments from adjacent regions of the same RMCA for all comparisons. 5-HT produced a concentration-dependent constriction in the range of 0.001 to 10 μmol l⁻¹, with maximal constriction observed at ∼3 μmol l⁻¹ at 10 and 60 mmHg. The peak constriction produced by near maximal 5-HT concentration (1 μmol l⁻¹) was on average greater at 60 compared to 10 mmHg (Fig. 1C). The EC₅₀ value at 60 mmHg (pEC₅₀= 7.1 ± 0.1) was slightly right-shifted compared to that at 10 mmHg (pEC₅₀= 7.5 ± 0.1, P < 0.05) (Fig. 1D).

Figure 1. Concentration–response relationship for 5-HT in RMCAs at 10 and 60 mmHg A and B, representative recordings of arterial diameter showing the effect of cumulative exposure to 5-HT between 0.001 and 10 μmol l⁻¹ in half-log concentrations at 10 and 60 mmHg. C, mean level ±s.e.m. of vasoconstriction due to 1 μmol l⁻¹ 5-HT at 10 (n= 14) and 60 mmHg (n= 22). Statistical significance was determined by Student's unpaired t test; *P < 0.05. D, mean ±s.e.m. concentration–response relations for 5-HT at 10 and 60 mmHg. The pEC₅₀ was shifted to the right at 60 mmHg (pEC₅₀ values of 7.1 ± 0.1 and 7.5 ± 0.1 mol l⁻¹ at 60 and 10 mmHg, respectively (n= 6 for each)).

Suppression of 5-HT-evoked vasoconstriction at 10 mmHg by ROK inhibition

The effect of the ROK inhibitor H1152 (0.5 μmol l⁻¹) on 5-HT-induced constriction of RMCAs maintained at 10 mmHg was examined. This concentration of H1152 was previously shown to produce almost maximal inhibition of myogenic constriction in RMCAs (Johnson et al. 2009). In time control vessels, a second exposure to 1 μmol l⁻¹ 5-HT evoked a similar level of constriction as the initial treatment (Fig. 2). However, treatment with H1152 almost abolished the second 5-HT-mediated constriction (Fig. 2).

Figure 2. Effect of ROK inhibition on 5-HT-induced constriction of RMCAs at 10 mmHg A, representative recordings of arterial diameter showing the response of RMCAs to two applications of 5-HT, with the second treatment applied in the absence or presence of H1152 to inhibit ROK activity. B, mean level ±s.e.m. of constriction induced by two applications of 5-HT (1st exposure, open bar), with the second application in the absence (Control; n= 18) or presence of H1152 (n= 20). Statistical significance was determined by repeated measures ANOVA followed by Bonferroni's post hoc test; *P < 0.05.

Increased phosphorylation of LC₂₀ but not MYPT1 in response to 5-HT at 10 mmHg

Phospho-LC₂₀ and phospho-MYPT1 content at 10 mmHg in the absence and presence of 5-HT were quantified by three-step Western blotting using anti-LC₂₀ and phospho-specific MYPT1-T697 and MYPT1-T855 antibodies. In the case of LC₂₀, separation of unphosphorylated and phosphorylated protein was achieved using Phos-Tag SDS-PAGE (Takeya et al. 2008; Johnson et al. 2009). 5-HT was initially employed at 1 μmol l⁻¹ (Fig. 3A) to evoke a near-maximal constriction (Fig. 1), and presumably, robust changes in phosphoprotein content and minimal measurement error. Basal phosphorylation of MYPT1-T697 and MYPT1-T855 was detected in untreated tissues at 10 mmHg (Fig. 3C). Although a substantial increase in phospho-LC₂₀ content was observed in response to 5-HT at 10 mmHg (Fig. 3B), no change in the level of phospho-MYPT1-T697 or -MYPT1-T855 was detected (Fig. 3C). To eliminate any doubt related to the sensitivity of detection, 5-HT concentrations of 3 and 10 μmol l⁻¹ were also employed (Fig. 3A). Phospho-LC₂₀ content increased from ∼25 to ∼50% of total LC₂₀ at 3 and 10 μmol l⁻¹, consistent with the change observed for 1 μmol l⁻¹ and the nearly identical level of constriction observed at the three concentrations (Fig. 3A and B). Increasing the concentration of 5-HT to 3 or 10 μmol l⁻¹ did not alter the phospho-MYPT1 content (Fig. 3C).

Figure 3. Effect of 5-HT on LC₂₀ and MYPT1 phosphorylation in RMCAs at 10 mmHg A, representative recordings of 1, 3, or 10 μmol l⁻¹ 5-HT-induced constriction at 10 mmHg. B, representative 3-step Western blot of LC₂₀ (left panel) and mean levels ±s.e.m. of phospho-LC₂₀ as a percentage of total LC₂₀ (right panel; n= 4) in the absence and presence of 5-HT at 1, 3, or 10 μmol l⁻¹. Unphosphorylated (0-P) and mono-phosphorylated (1-P) LC₂₀ were separated by Phos-Tag SDS-PAGE. C, representative 3-step Western blots of phospho-MYPT1-T697 and phospho-MYPT1-T855 and corresponding actin levels in each lane (left panel), and mean levels ±s.e.m. of phosphorylated MYPT1-T697 and MYPT1-T855 normalized to actin (right panel; n= 3) in the absence and presence of 5-HT at 1, 3, or 10 μmol l⁻¹. Statistical significance was determined by ANOVA followed by Dunnett's post hoc test; *P < 0.05.

Effect of ROK and PKC inhibition on LC₂₀ and MYPT1 phosphorylation during 5-HT treatment at 10 mmHg

The effect of H1152 (0.5 μmol l⁻¹) and GF109203X (3 μmol l⁻¹) on LC₂₀ and MYPT1 phosphorylation evoked by 1 μmol l⁻¹ 5-HT at 10 mmHg were examined. Both inhibitors caused a partial reduction in the level of 5-HT-induced LC₂₀ phosphorylation (Fig. 4A), consistent with an inhibition of constriction (Fig. 2A for H1152 and see Fig. 8A for GF109203X). MYPT1-T855 phosphorylation was reduced by >50% in tissues treated with H1152 and 5-HT, but the inhibitor did not affect MYPT1-T697 phosphorylation (Fig. 4B). The PKC inhibitor GF109203X had no effect on MYPT1 phosphorylation at T697 or T855 in the presence of 5-HT (Fig. 4B). Since 5-HT did not increase the level of phospho-MYPT1-T855 at 10 mmHg (Fig. 3C), and we previously noted that H1152 suppressed the basal level of phosphorylation at this site (Johnson et al. 2009), RMCAs were also exposed to H1152 at 10 mmHg in the absence of 5-HT according to the pressure protocol employed in this study (i.e. no pre-steps to 80 mmHg as were employed for all tissues in Johnson et al. 2009). Figure 4C shows that phosphorylation of MYPT1-T855 at 10 mmHg was significantly reduced by H1152 compared to untreated vessels, but the basal level of phospho-MYPT1-T697 was not affected.

Figure 4. Effect of ROK and PKC inhibitors on 5-HT-induced LC₂₀ and MYPT1 phosphorylation in RMCAs at 10 mmHg A, representative 3-step Western blot of LC₂₀ (left panel) and mean levels ±s.e.m. of phospho-LC₂₀ as a percentage of total LC₂₀ (right panel) in the presence of 1 μmol l⁻¹ 5-HT alone (n= 5) or 5-HT plus H1152 (0.5 μmol l⁻¹; n= 5) or GF109203X (3 μmol l⁻¹; n= 5). Unphosphorylated (0-P) and mono-phosphorylated (1-P) LC₂₀ were separated by Phos-Tag SDS-PAGE. B, representative 3-step Western blots of phospho-MYPT1-T697 and -MYPT1-T855 and corresponding actin levels in each lane (upper panels) and mean levels ±s.e.m. of phospho-MYPT1-T697 and -MYPT1-T855 normalized to actin (lower panels; n= 3 for each) in the presence of 1 μmol l⁻¹ 5-HT alone or 5-HT plus H1152 or GF109203X. C, representative 3-step Western blots of phospho-MYPT1-T697 and -MYPT1-T855 and corresponding actin levels in each lane (upper panels) and mean levels ±s.e.m. of phospho-MYPT1-T697 and -MYPT1-T855 normalized to actin (lower panels; n= 3 for each) at 10 mmHg in the absence (Con) or presence of H1152 (0.5 μmol l⁻¹). Statistical significance was determined by ANOVA followed by Dunnett's post hoc test for panels A and B and by Student's unpaired t test in panel C; *P < 0.05.

Figure 8. Effect of PKC inhibition on 5-HT-induced constriction of RMCAs A, representative recording of arterial diameter (left panel) and mean level ±s.e.m. of constriction (right panel) of RMCAs to two applications of 5-HT with the second treatment applied in the presence of GF109203X to inhibit PKC (5-HT 1st exposure, open bar; n= 18; 5-HT 2nd exposure in presence of GF109203X, filled bar, n= 18). B, representative 3-step Western blot of CPI-17 (left panel) and mean levels ±s.e.m. of phospho-CPI-17 as a percentage of total CPI-17 (right panel; n= 3) in the absence (C) and presence of PdBu (1 μmol l⁻¹) or 5-HT at 1, 3, or 10 μmol l⁻¹ at an intra-luminal pressure of 10 mmHg. Unphosphorylated (CPI-17) and phosphorylated (p-CPI-17) CPI-17 were separated by Phos-Tag SDS-PAGE. C, representative 3-step Western blot of CPI-17 (left panel) and mean levels ±s.e.m. of phospho-CPI-17 as a percentage of total CPI-17 (right panel; n= 3) in the absence (C) and presence of 5-HT (1 μmol l⁻¹) at 60 mmHg. Separation of non-phosphorylated and phosphorylated CPI-17 as in B. Statistical significance was determined by repeated measures ANOVA followed by Bonferroni's post hoc test; *P < 0.05.

Increased LC₂₀ and MYPT1 phosphorylation in response to 5-HT at 60 mmHg

RMCAs responded to an increase in intra-luminal pressure from 10 to 60 mmHg by constricting to maintain arterial diameter at or below the level at 10 mmHg owing to myogenic force generation (Fig. 5A). Exposure to 5-HT (1 μmol l⁻¹) following stabilization of arterial diameter induced a significant, further constriction that was associated with a further increase in phospho-LC₂₀ content compared to untreated segments at 60 mmHg (Fig. 5A and B). Contrary to the results obtained for 5-HT at 10 mmHg, agonist treatment at 60 mmHg caused a significant increase in the level of phosphorylation of MYPT1-T697 and MYPT1-T855 (Fig. 5C). The increase in phospho-MYPT1-T697 content was not anticipated, as we had previously failed to detect any change in MYPT1-T697 phosphorylation following a pressure step from 10 to 60 or 100 mmHg (Johnson et al. 2009). To determine whether the phosphorylation at MYPT1-T697 was specifically due to a mechanism activated by agonist application, we also assessed MYPT1 phosphorylation after stepping to an extreme pressure of 140 mmHg in the absence of agonist. This pressure lies at the transition to forced dilatation wherein myogenic force generation is no longer capable of holding arterial diameter constant and pressure steps to values greater than 140 mmHg are accompanied by dilatation (Osol et al. 2002). Figure 5D shows that a step increase in pressure to 140 mmHg was also associated with a significant increase in the level of phosphorylation at MYPT1-T697 and MYPT1-T855. This increase in MYPT1-T697 and MYPT1-T855 phosphorylation was accompanied by an increase in LC₂₀ phosphorylation (data not shown).

Figure 5. Effect of 5-HT on RMCA diameter and phosphorylation of LC₂₀ and MYPT1 at 60 mmHg A, representative recordings of RMCA diameter owing to a pressure step from 10 to 60 mmHg (left panel), the same pressure step followed by exposure to 1 μmol l⁻¹ 5-HT (middle) and the mean ±s.e.m. (n= 9) change in diameter due to pressure (60 mmHg) and pressure plus 5-HT (5-HT). B, representative 3-step Western blots of LC₂₀ and mean levels ±s.e.m. (n= 3 each) of phospho-LC₂₀ as a percentage of the total LC₂₀ at 60 mmHg in the absence (Con) and presence (5-HT) of 1 μmol l⁻¹ 5-HT. C, representative 3-step Western blots of phospho-MYPT1-T697 and phospho-MYPT1-T855 (and corresponding actin levels in each lane) (upper panels) and mean levels ±s.e.m. (n= 3 each) of phospho-MYPT1-T697 and -MYPT1-T855 normalized to actin (lower panels) at 60 mmHg in the absence (Con) and presence of 1 μmol l⁻¹ 5-HT (5-HT). D, representative 3-step Western blots of phospho-MYPT1-T697 and phospho-MYPT1-T855 and the corresponding actin (upper panels) and mean levels ±s.e.m. (n= 4 each) of phospho-MYPT1-T697 and phospho-MYPT1-T855 normalized to actin (lower panels) at 10 and 140 mmHg. Statistical significance was determined by Student's unpaired t test; *P < 0.05.

Suppression of 5-HT-induced increase in LC₂₀ and MYPT1 phosphorylation at 60 mmHg by ROK inhibition

In order to evaluate the dependence of 5-HT-induced constriction and increase in phosphoprotein content at 60 mmHg on ROK activity, sister vessel segments were pressurized to 60 mmHg and treated with 1 μmol l⁻¹ 5-HT alone or 5-HT followed by H1152 (0.5 μmol l⁻¹) after stable constrictions to pressure and 5-HT were observed (Fig. 6). ROK inhibition with H1152 suppressed the constriction due to pressure and 5-HT (Fig. 6A) and reduced the levels of LC₂₀, MYPT1-T697 and MYPT1-T855 phosphorylation (Fig. 6B). That the increase in MYPT-T697 phosphorylation evoked by 5-HT at 60 mmHg was affected by H1152 is significant, as the ROK inhibitor was previously found to be without effect on the level of phosphorylation at this site at pressures of 10, 60 and 100 mmHg (Figs 4 and 6, and Johnson et al. 2009), as well as at 10 mmHg following 5-HT treatment in this study (Fig. 4).

Figure 6. Effect of ROK inhibition on 5-HT-induced constriction and LC₂₀ and MYPT1 phosphorylation in RMCAs at 60 mmHg A, representative recordings of RMCA diameter owing to a pressure step from 10 to 60 mmHg followed by 5-HT (left) and then H1152 (middle) and the mean ±s.e.m. (n= 6) change in diameter due to pressure plus 5-HT without (5-HT) or with subsequent treatment with H1152 (H1152). B, representative 3-step Western blots of LC₂₀, phospho-MYPT1-T697 and phospho-MYPT1-T855 (and corresponding actin levels in each lane) (upper panels) and mean levels ±s.e.m. (n= 3 each) of phospho-LC₂₀ as a percentage of total LC₂₀ and phospho-MYPT1-T697 and -MYPT1-T855 normalized to actin (lower panels) at 60 mmHg plus 5-HT without (5-HT) or with subsequent treatment with H1152 (H1152). Statistical significance was determined by Student's unpaired t test; *P < 0.05.

Lack of effect of inhibition of cross-bridge cycling on 5-HT-evoked changes in MYPT1 phosphorylation

To examine whether myogenic force generation per se was responsible for the change in MYPT1 phosphorylation evoked by 5-HT at 60 mmHg, we employed blebbistatin at 10 μmol l⁻¹ (IC₅₀ 0.5–5 μmol l⁻¹) to inhibit actomyosin ATPase activity and suppress cross-bridge cycling (Eddinger et al. 2007). Blebbistatin treatment significantly reduced the constriction in the presence of 5-HT and 60 mmHg (Fig. 7A), but the extent of dilatation was less than that observed with H1152 (see Fig. 6A). Significantly, we did not detect any change in the level of MYPT1-T697, MYPT1-T855 or LC₂₀ phosphorylation in the presence of blebbistatin (Fig. 7B). Similar results were obtained when vessels were pretreated with 10 μmol l⁻¹ blebbistatin prior to stepping to 60 mmHg and treatment with 5-HT (data not shown). In the latter case, blebbistatin completely blocked the myogenic response and reduced the magnitude of constriction induced by 5-HT (i.e. 19.8 ± 5.1 μm versus∼50 μm in control tissues in Fig. 5).

Figure 7. Effect of the inhibition of cross-bridge cycling with blebbistatin on RMCA diameter and phosphorylation of LC₂₀ and MYPT1 following treatment with 5-HT at 60 mmHg A, representative recordings of RMCA diameter owing to a pressure step from 10 to 60 mmHg followed by 1 μmol l⁻¹ 5-HT without (left) or with subsequent exposure to 10 μmol l⁻¹ blebbistatin (middle) and the mean ±s.e.m. (n= 6) change in diameter due to pressure plus 5-HT without (5-HT) and with subsequent exposure to blebbistatin (Blebb) (right). B, representative 3-step Western blots of LC₂₀, phospho-MYPT1-T697 and phospho-MYPT1-T855 (and corresponding actin levels in each lane) (upper panels) and mean levels ±s.e.m. (n= 3 each) of phospho-LC₂₀ as a percentage of total LC₂₀ and phospho-MYPT1-T697 and -MYPT1-T855 normalized to actin (lower panels) at 60 mmHg plus 5-HT without (5-HT) and with subsequent exposure to blebbistatin (Blebb). Statistical significance was determined by Student's unpaired t test; *P < 0.05.

Lack of role of CPI-17 phosphorylation despite block of vasoconstriction to 5-HT by PKC inhibition

The contribution of a PKC-mediated increase in CPI-17 phosphorylation to 5-HT-evoked constriction at 10 mmHg was examined using an identical approach to that described above for MYPT1. The pan-PKC isoform inhibitor GF109203X at 3 μmol l⁻¹ (Dimopoulos et al. 2007; Maruko et al. 2009; Johnson et al. 2009) was first shown to suppress 5-HT-induced constriction at 10 mmHg (Fig. 8A). However, we were unable to detect any basal CPI-17 phosphorylation, or an increase in phosphorylation at 3–5 min following treatment with 5-HT (Fig. 8B). Moreover, increasing the concentration of 5-HT to 3 and 10 μmol l⁻¹ also failed to induce a change in phospho-CPI-17 content (Fig. 8B).

Discussion

This study examined the contribution of ROK- and PKC-mediated mechanisms of agonist-induced force generation by cerebral resistance arteries superimposed on a physiologically relevant background of pressure-induced myogenic vasoconstriction. Specifically, the role of MYPT1- and CPI-17-dependent Ca²⁺ sensitization in 5-HT-evoked constriction of RMCAs at 10 and 60 mmHg was determined. Pressure myography and a three-step immunoblotting method were used to measure arterial diameter and LC₂₀, MYPT1 and CPI-17 phosphorylation, respectively, in vessel segments treated with 1 μmol l⁻¹ 5-HT in the absence and presence of pharmacological inhibition of ROK, PKC or cross-bridge cycling. The important novel findings of this study are that: (1) the contribution of ROK-mediated MYPT1 phosphorylation to 5-HT-induced constriction was pressure dependent and only observed in the presence of myogenic tone; and (2) the site of ROK-mediated MYPT1 phosphorylation varied with the magnitude of the vasoconstrictor stimulus.

No change in phospho-CPI-17 content was detected in RMCAs treated with 5-HT in this study. The vessel samples employed for phospho-CPI-17 quantification were obtained at 3–5 min following exposure to 5-HT. Although agonist-induced CPI-17 phosphorylation is rapid and peaks within the first 30 s in rabbit femoral artery, it remains elevated for at least 5 min (Dimopoulos et al. 2007) and would have been detected using the protocol employed here if the kinetics in the femoral and cerebral arteries are similar. Also, the method was considered to be of sufficient sensitivity to detect alterations in CPI-17 phosphorylation, as direct activation of PKC with 1 μmol l⁻¹ PDBu was found to increase phospho-CPI-17 content by ∼35%. These data highlight the importance of employing a direct, biochemical approach to assess the involvement of PKC-mediated mechanisms of Ca²⁺ sensitization in agonist-induced constriction of pressurized resistance arteries. Specifically, our findings indicate that Ca²⁺ sensitization due to CPI-17 phosphorylation does not contribute to 5-HT-evoked constriction at 10 and 60 mmHg despite the fact that the responses were almost completely suppressed by PKC inhibition with GF109203X. Previous studies on resistance arteries have frequently claimed a role for PKC-induced Ca²⁺ sensitization on the basis of a similar inhibition of agonist-induced constriction without supporting biochemical evidence of an alteration in CPI-17 phosphorylation (cremaster arteries, Hill et al. 1990; Liu et al. 1994; gracilis arteries, Massett et al. 2002; mesenteric arteries, Hill et al. 1996; Wesselman et al. 2001; cerebral arteries, Osol et al. 1991; Gokina & Osol, 1998; Gokina et al. 1999; Lagaud et al. 2002; but see Akopov et al. 1998; Sandoval et al. 2007). Significantly, the PKC inhibitors used in some of these studies were shown to be of inferior selectivity, e.g. staurosporine and H-7. However, a PKC-dependent contribution to increased LC₂₀ phosphorylation cannot be ruled out as it was shown to occur in permeabilized vessels treated with a PKC activator (Hill et al. 1996). GF109203X, on the other hand, is 200-fold more selective for PKC compared to PKA and at least 5000-fold more selective for PKC compared to tyrosine kinases (Toullec et al. 1991, see Davies et al. 2000 for detailed information of selectivity compared to other kinases). It is possible that the inhibition of the 5-HT-evoked contraction of RMCAs observed here occurred via effects on: (1) non-selective cation and/or voltage-gated Ca²⁺ channel activity leading to a change in [Ca²⁺]_i (e.g. McCarron et al. 1997; Korzick et al. 2004; Cobine et al. 2007; Earley et al. 2007) that accounts for ∼15% decline in LC₂₀ phosphorylation; (2) thin filament regulation (Jin et al. 2000; Morgan & Gangopadhyay, 2001); and/or (3) dynamic reorganization of the cytoskeleton (Cipolla & Osol, 1998; Gerthoffer, 2005; Gunst & Zhang, 2008).

Our findings indicate that the contribution of ROK-mediated Ca²⁺ sensitization to agonist-induced constriction can be influenced by the presence of myogenic tone. Previous studies provide convincing evidence that 5-HT evokes depolarization of cerebral VSMC membrane potential leading to an increase in [Ca²⁺]_i and force generation, and a reduction in arterial diameter (Harder & Waters, 1983; Worley et al. 1991). Here, 5-HT treatment at 10 and 60 mmHg was associated with an increase in LC₂₀ phosphorylation, but an increase in phospho-MYPT1 content was only detected at 60 mmHg. The absence of biochemical evidence of increased MYPT1 phosphorylation due to 5-HT at 10 mmHg suggests that the increase in phospho-LC₂₀ resulted from MLCK activation owing to depolarization-mediated increase in Ca²⁺ influx, and possibly increased Ca²⁺ release from internal stores (Salomone et al. 2009), rather than an increase in ROK-mediated inhibition of MLCP activity. In contrast, 5-HT treatment at 60 mmHg caused a ROK-mediated phosphorylation of MYPT1-T697 and MYPT1-T855. Therefore, in the presence of pre-existing myogenic tone at 60 mmHg owing to depolarization, elevated [Ca²⁺]_i (Knot & Nelson, 1998), and MYPT1-T855 phosphorylation (Johnson et al. 2009), an alternative mechanism was recruited to permit the necessary increase in force generation required for 5-HT-induced constriction; specifically, a ROK-mediated Ca²⁺ sensitization mechanism leading to further suppression of MLCP activity and a greater shift in the MLCK-MLCP balance to favouring increased LC₂₀ phosphorylation. The pressure dependence of the contribution of ROK-mediated Ca²⁺ sensitization to 5-HT-induced contraction of RMCAs provides biochemical evidence of an interplay between myogenic and agonist-induced mechanisms of vasoconstriction. Moreover, the absence of Ca²⁺ sensitization at 10 mmHg provides a possible explanation for previous instances in which no change in MYPT1 phosphorylation was detected in cerebral arterial rings exposed to 5-HT in the absence of intra-luminal pressure (e.g. Watanabe et al. 2005).

The present findings make an important contribution to our understanding of MYPT1 phosphorylation in smooth muscle contraction. Currently, no consensus has been reached regarding the involvement of one or both major sites of MYPT1 phosphorylation in Ca²⁺ sensitization in smooth muscle. Agonist-induced phosphorylation at MYPT1-T855 is invariably detected in intact tissues when it is assessed (e.g. Kitazawa et al. 2003; Stevenson et al. 2004; Wilson et al. 2005), but coincident phosphorylation of MYPT1-T697 is not always observed (both sites, e.g. Neppl et al. 2009; versus only MYTPT1-T855, e.g. Freitas et al. 2009; Tsai & Jiang, 2006), despite the fact that synthetic MYPT1-T697 phosphopeptide is a ∼30-fold more potent inhibitor of PP1C activity in vitro compared to phosphorylated MYPT1-T855 peptide (Khromov et al. 2009). An explanation for the varied presence of MYPT1-T697 phosphorylation has not been forthcoming, prompting speculation that MYPT1-T697 may not be a ROK phosphorylation site in vivo, but perhaps phosphorylated by another kinase, such as zipper-interacting kinase or integrin-linked kinase (MacDonald et al. 2001; Murányi et al. 2002). This may be the case for basal phosphorylation of MYPT1-T697 in RMCAs that is not affected by H1152 treatment (Fig. 4; and Johnson et al. 2009). However, our data show that the increase in phospho-MYPT1-T697 content associated with 5-HT treatment at 60 mmHg was suppressed by H1152 and, therefore, mediated by ROK.

In this study, variable phosphorylation of the two major sites on MYPT1 was detected under different conditions of contractile stimulation. Specifically, the data indicate that ROK-mediated phosphorylation of the two main sites on MYPT1 occurred in a preferential manner, first at MYPT1-T855 and subsequently at MYPT1-T855 and MYPT1-T697, possibly providing a greater range of increase in the level of Ca²⁺ sensitization and force generation than can be attained by phosphorylation at a single site. We previously showed that myogenic contractions to intra-luminal pressure steps from 10 to 60 or 100 mmHg were accompanied by phosphorylation of MYPT1-T855, but not MYPT1-T697 (Johnson et al. 2009). Here, we found that 5-HT treatment at 60 mmHg elevated the phospho-MYPT1-T855 content beyond that caused by pressure alone, and also increased MYPT1-T697 phosphorylation. The elevated phosphorylation of MYPT1-T697 was not due to a signalling pathway activated exclusively by 5-HT, because increasing intra-luminal pressure to 140 mmHg in the absence of agonist was also associated with enhanced phosphorylation at both sites. Taken together, these data on RMCAs indicate that preferential phosphorylation of MYPT1-T855 followed by MYPT1-T697 is evoked during myogenic contractions of increasing magnitude due to intra-luminal pressure and by the combination of agonist-induced constriction superimposed on a background of myogenic tone. That MYPT1 exhibits phosphorylation at MYPT1-T855 and then at MYPT1-T855 and MYPT1-T697 under conditions of increasing force generation in situ provides a potential explanation for the varied detection of MYPT1-T855 and MYPT1-T697 phosphorylation in previous studies of intact smooth muscle.

How ROK preferentially phosphorylates MYPT1-T855 before MYPT1-T697 is an intriguing question that remains to be answered. It is unlikely that the presence of myogenic tone at 60 mmHg induces a change in the conformation of MYPT1 and exposure of MYPT1-T697 for phosphorylation by ROK. This view is supported by the lack of effect of blebbistatin inhibition of cross-bridge cycling and force generation on MYPT1 phosphorylation in the presence of 5-HT at 60 mmHg, i.e. increased MYPT1-T855 and MYPT1-T697 phosphorylation was still detected. Yet, it is possible that activated ROK has a lower affinity for the MYPT1-T697 substrate so that detectable increases in phosphorylation levels, comparable to those at MYPT1-T855, require a stronger stimulus to achieve a higher level of ROK activation. While in vitro experiments imply that there is no difference in the efficiency of phosphorylation of MYPT1-T697 and MYPT1-T855 by ROK (Murányi et al. 2005), this may not be the case in situ. Alternatively, there may be multiple pools of ROK and/or MYPT1 that are differentially recruited by vasoconstrictor stimuli of varied intensity, as previously suggested by Neppl et al. (2009). An additional possible explanation could be based on a recent observation that the phosphorylated MYPT1-T697 residue and the surrounding domain is at least 30-fold more potent as a MLCP autoinhibitor compared to the phosphorylated MYPT1-T855 residue (Khromov et al. 2009), which could potentially mean that at a given stimulus strength the same MLCP inhibition level can be obtained with a much lower level of MYPT1-T697 phosphorylation that would be undetected with weaker stimuli. Further work is required to distinguish between these possibilities.

The mechanisms that determine the level of force generation and constriction of resistance arteries in response to agonists and pressure are not known with certainty. The roles of MLCK activation and MLCP inhibition in agonist-induced contraction of conduit arteries are well documented, and our findings now extend this knowledge to the level of the resistance vasculature under physiological conditions of intra-luminal pressure. However, the effects of ROK and PKC inhibition on contractility in this study cannot be fully explained by the modulation of MLCK and MLCP activity. The absence of 5-HT-induced, ROK-mediated phosphorylation of MYPT1 at 10 mmHg is contrary to the expected result based on the strong inhibition of the constriction by H1152. However, this apparent contradiction is resolved, at least in part, because H1152 reduces basal MYPT1-T855 phosphorylation (Fig. 4, and Johnson et al. 2009). This would be expected to increase MLCP activity, enhance LC₂₀ dephosphorylation and inhibit force generation. The observed reduction in LC₂₀ phosphorylation in the presence of H1152 and 5-HT at 10 mmHg is consistent with this interpretation. However, we also found that: (1) constrictions in the presence of 5-HT at 10 and 60 mmHg were almost completely blocked by PKC or ROK inhibition, respectively, but phospho-LC₂₀ content only decreased by 10–15%; i.e. to a level equivalent to that at 60 mmHg in control vessels with considerable myogenic tone (Fig. 5, and Johnson et al. 2009). (2) The increases in phospho-LC₂₀ evoked by 5-HT at 10 and 60 mmHg were ∼25 and ∼10%, respectively, yet the amplitude of the constriction at 60 mmHg was significantly greater. These results can be explained in part because the experiments were conducted under isobaric conditions; thus, small changes in LC₂₀ phosphorylation may be sufficient to alter the balance between constriction owing to smooth muscle contraction and dilatation due to intra-luminal pressure. Also, it may be expected that the RMCAs are at a higher point on the force versus LC₂₀ phosphorylation curve at 60 compared to 10 mmHg, such that smaller changes in phospho-LC₂₀ elicit greater alterations in force generation. On the other hand, mechanisms that are unrelated to the regulation of MLCK and MLCP activity may also be involved. Possible candidate mechanisms include thin filament regulation owing to phosphorylation of calponin and caldesmon (Jin et al. 2000; Kaneko et al. 2000; Morgan & Gangopadhyay, 2001), and dynamic reorganization of the actin cytoskeleton due to alterations in G- and F-actin turnover (Cipolla & Osol, 1998; Cipolla et al. 2002; Gokina & Osol, 2002; Gerthoffer, 2005). For example, the reported increase in F-actin with pressure elevation (Cipolla & Osol, 1998; Cipolla et al. 2002; Gokina & Osol, 2002) could provide for enhanced transmission of force to the membrane and an increased level of 5-HT-evoked constriction for a smaller change in LC₂₀ phosphorylation at 60 mmHg versus 10 mmHg. The contribution of these mechanisms to force generation in the presence of 5-HT at 60 mmHg remains to be determined.

In summary, this study provides novel evidence of an interplay between agonist- and pressure-induced vasoconstriction in RMCAs that has important implications for understanding the complex integration of intrinsic and extrinsic mechanisms for control of vascular smooth muscle contraction and regulation of resistance arterial diameter. Taken together, our findings indicate that Ca²⁺ sensitization due to a sequential ROK-mediated phosphorylation of MYPT1-T855 and MYPT1-T697, but not PKC-mediated phosphorylation of CPI-17, contributes to cerebral vasoconstriction evoked by pressure and pressure plus 5-HT. It remains to be determined whether similar mechanisms are responsible for the actions of other contractile agonists on cerebral arteries and/or are present in resistance arteries of other vascular beds.

Appendix

Author contributions

All authors contributed to conception and design of the experiments, analysis and interpretation of the results, drafting of the manuscript, and have approved the final version for publication. Experiments for this study were conducted in the Cole and the Walsh laboratories in the Smooth Muscle Research Group of the Faculty of Medicine at the University of Calgary, Calgary, Alberta, Canada.

Acknowledgements

This work was supported by a grant from the Canadian Institutes of Health Research (CIHR) (MOP-10568). A.F.E. was supported by the CIHR and the Alberta Heritage Foundation for Medical Research (AHFMR) and holds the position of Lecturer of Pharmacology at the Faculty of Pharmacy at Alexandria University. R.P.J. was supported by the AHFMR and Natural Sciences and Engineering Research Council of Canada. W.C.C. is the Andrew Family Professor in Cardiovascular Research and M.P.W. is an AHFMR Scientist and recipient of a Canada Research Chair (Tier 1) in Vascular Smooth Muscle Research.