• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of jphysiolThe Journal of Physiology SiteMembershipSubmissionJ Physiol
J Physiol. Jun 1, 2009; 587(Pt 11): 2599–2612.
Published online Apr 29, 2009. doi:  10.1113/jphysiol.2008.165258
PMCID: PMC2714024

Activation by Ca2+/calmodulin of an exogenous myosin light chain kinase in mouse arteries


Activation of myosin light chain kinase (MLCK) and other kinases was studied in the arteries of transgenic mice that express an optical fluorescence resonance energy transfer (FRET) MLCK activity biosensor. Binding of Ca2+/calmodulin (Ca2+/CaM) induces an increase in MLCK activity and a change in FRET. After exposure to high external [K+], intracellular [Ca2+] (fura-2 ratio or fluo-4 fluorescence) and MLCK activity both increased rapidly to an initial peak and then declined, rapidly at first and then very slowly. After an initial peak (‘phasic’) force was constant or increased slowly (termed ‘tonic’ force). Inhibition of rho-kinase (Y-27632) decreased tonic force more than phasic, but had little effect on [Ca2+] and MLCK activation. Inhibition of PKCα and PKCβ with Gö6976 had no effect. KN-93, an inhibitor of CaMK II, markedly reduced force, MLCK FRET and [Ca2+]. Applied during tonic force, forskolin caused a rapid decrease in MLCK FRET ratio and force, but no change in Ca2+, suggesting a cAMP mediated decrease in affinity of MLCK for Ca2+/CaM. However, receptor (β-adrenergic) activated increases in cAMP during KCl were ineffective in causing relaxation, changes in [Ca2+], or MLCK FRET. At the same tonic force, MLCK FRET ratio activated by α1-adrenoceptors was ~60% of that activated by KCl. In conclusion, MLCK activity of arterial smooth muscle during KCl-induced contraction is determined primarily by Ca2+/CaM. Rho-kinase is activated, by unknown mechanisms, and increases ‘Ca2+ sensitivity’ significantly. Forskolin mediated increases in cAMP, but not receptor mediated increases in cAMP cause a rapid decrease in the affinity of MLCK for Ca2+/CaM.

Contraction of smooth muscle is believed to be primarily dependent on phosphorylation of myosin regulatory light chains (RLCs), as determined by the relative activities of Ca2+/calmodulin-dependent myosin light chain kinase (MLCK) and myosin light chain phosphatase (MLCP). Here, we will use the term, ‘activation’ (of MLCK) to refer to the increase in kinase activity of MLCK that occurs upon binding of Ca2+/CaM, and the term, ‘regulation’ to refer to changes in the affinity of MLCK for Ca2+/CaM that might result from phosphorylation of MLCK. Phosphorylation of MLCK at serine 1760 (‘regulatory site A’) by Ca2+/CaM-dependent protein kinase II (CaMK II) or protein kinase A (PKA) is known to regulate (decrease) the affinity of MLCK for Ca2+/CaM. MLCK can also be phosphorylated by protein kinase C (PKC), p-21 activated protein kinase (PAK) and mitogen activated protein kinase/extracellular regulated kinase (MAPK/ERK) (Kamm & Stull, 2001). Activity of MLCP is regulated most importantly by rho-kinase and by 17 kDa protein kinase C-potentiated phosphatase inhibitor (CPI-17).

The extent and time course of activation of these kinases in smooth muscle cells of intact arteries during contraction is essentially unknown, but can be expected to be complex. For example, the rise in intracellular [Ca2+] that typically activates MLCK in response to G protein coupled receptor (GPCR) activation or to membrane depolarization may inevitably activate other Ca2+-dependent proteins, such as CaMK II and conventional (i.e. Ca2+ activated) forms of PKC, potentially leading to both activation and regulation of MLCK, and to regulation of MLCP. Furthermore, Ca2+ is thought also to exert an inhibitory effect on MLCP through induction of rho-kinase activation (Yoshioka et al. 2007).

Here, we use a genetically encoded (exogenous) FRET-based MLCK biosensor molecule (Isotani et al. 2004) in an attempt to provide information on how the endogenous MLCK might be activated and regulated, during contraction of intact arteries. At the most basic level, changes in FRET within the MLCK biosensor molecule result from conformational changes caused by the binding or release of Ca2+/CaM, in the biosensor molecules only. The extent to which the same processes might be occurring in the endogenous MLCK molecules in the smooth muscle studied here is not actually known. Differences might occur, for example, as a result of differences in spatial localization of the two molecules. It is known, however, that the FRET ratio reported by the MLCK biosensor (in HEK 293 cells) is quantitatively linked to the concentration of Ca2+ and to the biosensor's ability to phosphorylate myosin regulatory light chains (Geguchadze et al. 2004); the half-maximal response for FRET was obtained at a pCa of 6.2, while the pCa for half-maximal phosphorylation of myosin (by the biosensor) was 6.4. If the same situation obtains in the smooth muscle cells of the arteries studied here, the biosensor FRET ratio we observe should reflect the endogenous MLCK activity, as activated by the endogenous Ca2+/CaM. The MLCK biosensor might also be expected to provide information on possible regulation of MLCK activity during contraction. Regulation of MLCK activity (defined here as a change in the affinity of MLCK for Ca2+/CaM caused by phosphorylation of MLCK) should be revealed by changes in biosensor FRET ratio, when no changes in [Ca2+] occur (assuming that no changes in the availability of CaM occur either).

Finally, we activated contraction using both elevated external [K+] (KCl) and the α1-adrenoceptor agonist phenylephrine (PE). KCl has the advantage that it achieves a spatially uniform elevation of [Ca2+] within small arteries and thereby simplifies the analysis of spatially averaged fluorescence signals. Activation of GPCR is more physiological, but presents a more complex situation in terms of possible heterogeneity of [Ca2+] (Zang et al. 2001) and activation of many other signalling cascades.


Animals, arteries and solutions

All experiments were approved by the Institutional Animal Care and Use Committee of the University of Maryland, School of Medicine. Inbred Charles River, wild-type (WT) and transgenic (TG) mice were maintained on 12: 12 h light–dark schedule at 22–25°C and 45–65% humidity and fed ad libitum on a standard rodent diet and tap water. Adult mice (28–35 g, 12–18 weeks), were killed by inhalation of CO2.

As described previously, (Isotani et al. 2004), the TG mice express a MLCK biosensor that monitors the binding of Ca2+/CaM through changes in FRET between cyan fluorescent protein (CFP) and yellow fluorescent protein (YFP). The mesenteric arcade was dissected from the abdominal cavity, rinsed free of blood, and placed in a temperature-controlled dissection chamber (5°C) containing a solution of the following composition (in mmol l−1): 3.0 Mops, 145.0 NaCl, 5.0 KCl, 2.5 CaCl2, 1.0 MgSO4, 1.0 KH2PO4, 0.02 EDTA, 2.0 sodium pyruvate, and 5.0 glucose (pH 7.4). Segments, 2–3 mm in length, of third-order mesenteric arteries were dissected free. If calcium indicators were to be used, the selected artery was then further exposed to dissection solution containing either fura-2 AM (5.0 μm) or fluo-4 AM (10.0 μm) and loading was allowed to proceed for 1 h or 3 h, respectively, at room temperature. After mounting for force and fluorescence recording (below) superfusion was begun with the standard experimental solution containing (in mmol l−1) 112.0 NaCl, 25.7 NaHCO3, 4.9 KCl, 2.0 CaCl2, 1.2 MgSO4, 1.2 KHPO4, 11.5 glucose, and 10.0 Hepes (pH 7.4, equilibrated with gas of 5% O2, 5% CO2, 90% N2). Solutions containing elevated KCl were made by replacing the NaCl with KCl on an equimolar basis. PE was used in concentrations ranging from 0.1 μm to 10.0 μm. The pH and partial pressure of O2 in the bath, within 0.5 mm of the artery, were 7.4 and 114 mmHg, respectively, measured with micro pH and O2 sensing microelectrodes (Microelectrodes, Inc., Londonderry, NH, USA).

Arteries were studied at 32°C, as at 37°C, fura-2 and fluo-4 are transported rapidly out of the cytoplasm of healthy smooth muscle cells. This provides a compromise between physiological conditions and the experimental necessity of retaining enough Ca2+ indicators to make measurements. At 32°C, arteries develop myogenic tone, which is an important physiological parameter of vessel viability (Zacharia et al. 2007).

When necessary, sections of aorta were also removed for biochemical measurements of MLCK content or MLCK phosphorylation by immunoblotting. Aortic sections were opened longitudinally, and immersed in the same solutions as used for the mesenteric arteries, as required.

Heart rate, indirect systolic, diastolic and mean blood pressures were recorded by tail cuff plethysmography using a blood pressure analysis system SC1000 (Hatteras Instruments Inc., Cary, NC, USA). The system utilizes a pressure transducer close to the tail-cuff on the SC1000 specimen platform. The platform temperature is precisely maintained during measurements using a low voltage silicone rubber heater. This is important to promote adequate blood flow in the tail for pulse detection. Pulse detection is made with a photodiode detector in combination with high intensity light emitting diode. Conscious mice were conditioned to five inflation/deflation cycles by a trained operator. Subsequently, the average values for blood pressure and heart rate in each animal is obtained from 10 sequential cuff inflation/deflation cycles.

All data are expressed as means ±s.e.m. Statistical analysis was performed on raw data using SigmaStat (Systat Software Inc., San Jose, CA, USA). Statistical analysis of paired data was performed with Student's t-test. In all cases P < 0.05 was taken as the level of significance.

Rho-kinase was inhibited by Y-27632 [(R)-(–)-trans-N-(4-pyridyl)-4-(1-aminoethyl)-cyclohexanecarboxamide]) (Calbiochem) and PKC was inhibited by Gö6976 (Invitrogen). CaMK II was inhibited by KN-93 (Calbiochem). Isoproterenol, forskolin and PE were obtained from Sigma, H-89 and rp-cAMP were purchased from Calbiochem.

Force measurements

Arteries were transferred to a recording chamber, where they were mounted on a confocal wire myograph (Danish Myotech Technology, Denmark). The length of the wires on the wire myograph jaws, and hence the length of the artery that generates force, was always 1.76 mm. Thus, the forces reported here from different arteries are all strictly comparable. The resting tension–internal circumference (IC) relation was determined for each vessel (Mulvany & Halpern, 1977). IC was set to 0.9 × IC100, where IC100 is the internal circumference of the vessel under an effective transmural pressure of 100 mmHg (13.3 kPa).

Biochemical measurements

Sections of aorta were used in order to obtain sufficient protein for Western blots. After aortas were subjected to the particular experimental procedure, they were lysed in 50 mm Tris-HCl, pH 7.4, 1% Nonidet P-40 (NP-40), 0.1% sodium dodecyl sulfate (SDS), 150 mm NaCl, 0.5% sodium deoxycholate, 1 mm EDTA, 1 mm phenylmethylsulfonyl fluoride (PMSF), 1 mm Na3VO4, 1 mm NaF, and proteinase inhibitors. Lysates were centrifuged at 7000 g at 4°C for 15 min and supernatant was collected as total protein. Protein concentration was determined with BCA protein assay kit (Pierce Biotechnology, Inc., Rockford, IL, USA). Protein was separated on NuPAGE 4–12% Bis-Tris Gel (Invitrogen) and transferred onto Trans-Blot Nitrocellulose membrane (Bio-Rad). Polyclonal antibodies against phospho-MLCK (Ser 1760, Upstate Biotechnology, Inc., Lake Placid, NY, USA) were used following the manufacturers’ protocols. Horseradish peroxidase-conjugated goat anti-rabbit and goat anti-mouse IgG were used as secondary antibodies (Sigma). Reactions were developed with SuperSignal West Pico Chemiluminescent Substrate (Pierce) and exposure to autoradiographic film. Signalling was quantified from scanned films using ImageJ software (Scion).

Fluorescence measurements

The equipment provided (1) confocal images of Ca2+ activated fluorescence (fluo-4) in individual smooth muscle cells of WT arteries, (2) confocal images of MLCK biosensor fluorescence in individual smooth muscle cells of TG arteries, and (3) simultaneous measurements of spatially averaged Ca2+ activated fluorescence (fura-2) and MLCK biosensor fluorescence from TG arteries mounted in a widefield microscope. A Nipkow spinning disk confocal microscope equipped with excitation illumination at 488 nm or 442 nm and an intensified CCD camera (Stanford Photonics, Palo Alto, CA, USA) was used to obtain images of fluo-4 or MLCK biosensor fluorescence. For the MLCK FRET measurements the confocal microscope was also equipped with an image-splitter designed for FRET measurements of CFP/YFP type FRET indicator molecules (Dual-View, Roper Scientific, Tucson, AZ, USA). Emission filters were for CFP, 480/40 nm and for YFP, 535/35 nm (centre wavelength, bandwidth at full width at half-maximum, FWHM).

The system used for the spatially averaged (nearly) simultaneous measurements of Ca2+ indicator fluorescence (fura-2) and MLCK biosensor fluorescence has been described in detail recently (Wier et al. 2008). Briefly, because of the spectral properties of fura-2 and the MLCK biosensor, it is possible to excite fura-2 and CFP of the MLCK biosensor selectively (at slightly different times, with different excitation filters) and therefore to collect fluorescence emission selectively from each. Since the emission spectrum of fura-2 overlaps that of both CFP and YFP, emission of fura-2 can be collected with the same two emission filters that are used for the MLCK biosensor.

Spatially resolved information on [Ca2+] and MLCK activation in individual cells was obtained, in different arteries, with fluo-4 and confocal microscopy. Information on [Ca2+] and MLCK activation in the same artery was obtained with the widefield epi-fluorescence microscope. This is because our confocal microscope does not provide excitation light for both fura-2 and CFP, as would be required for simultaneous, confocal measurements of [Ca2+] and MLCK activation in the same artery. The confocal microscope does provide excitation light for both fluo-4 and CFP of the MLCK biosensor, but fluo-4 and CFP cannot be used together because of extensive spectral overlap of excitation of fluo-4 and CFP (typically 488 nm and 442, respectively).


In confocal images of CFP fluorescence, the MLCK biosensor appeared to be expressed in all the individual smooth muscle cells in the walls of TG arteries (Fig. 1A), although expression levels, as judged by the intensity of CFP fluorescence, varied amongst different cells. Western blots (Fig. 1B) revealed that the MLCK biosensor was expressed in aortas at a level ~30% of the endogenous MLCK, similar to that reported earlier for bladder smooth muscle cells in the same animals (Isotani et al. 2004). Because the expression levels are apparently not the same in all cells, this figure represents an average level of expression. The presence of the exogenous MLCK raises the possibility that the function of the TG arteries could be affected, as, for example, might happen if the MLCK biosensor bound significant amounts of Ca2+/CaM, reducing the availability of Ca2+/CaM for endogenous MLCK and other molecules that bind Ca2+/CaM. No differences were detected, however, in the blood pressure (Fig. 1C) or heart rate (Fig. 1D) of TG mice compared to WT. Similarly, no differences were detected in the contractile responses of TG and WT arteries to elevated [KCl] or PE (Fig. 1E). The fact that TG animals and arteries were the same as WT animals and arteries for the characteristics examined might reflect the relatively low expression levels of the exogenous MLCK.

Figure 1
Visualization of MLCK biosensor in intact arteries and physiological properties of arteries containing the MLCK biosensor

[Ca2+] and MLCK biosensor fluorescence in individual cells

Although we expected KCl to induce spatially uniform changes in [Ca2+] in individual smooth muscle cells (thus justifying the use of spatially averaged measurements of fura-2 and MLCK biosensor in the same artery), we confirmed this important fact by confocal imaging of fluo-4 fluorescence in individual cells of WT arteries (Fig. 2A) during KCl exposure and confocal imaging of MLCK biosensor fluorescence in individual cells of TG arteries (Fig. 2D). Fluo-4, instead of fura-2, was used to check the uniformity of [Ca2+] changes in the artery because fura-2, cannot be used in the confocal microscope. Because fluo-4 is not a ratiometric Ca2+ indicator, the measurements are subject to possible errors due to artery motion. The maximum translation of the cells in the field of view during the (nominally) isometric contraction was less than 5%. Indeed, the changes in fluo-4 fluorescence (Fig. 2B) and MLCK FRET ratio (Fig. 2E) in representative individual smooth muscle cells during KCl exposure are similar to the spatially averaged fluorescence signals (Figs. 2C, F). The magnitude and basal levels of the fluorescence signals do vary substantially amongst different cells, for unknown reasons, but the time courses of both fluo-4 and MLCK FRET ratio changes are all remarkably similar in the different cells. Thus the spatially averaged fluorescent signal can be used as an approximation to events within an (average) cell.

Figure 2
Confocal images of MLCK biosensor fluorescence and fluo-4 fluorescence in areas of interest (AOI) within individual smooth muscle cells of TG and WT arteries, respectively, during exposure to elevated KCl

Time courses of MLCK FRET ratio and [Ca2+]

Average MLCK FRET ratio and Ca2+ indicator signals during KCl exposure were typically very similar in time course, declining rapidly at first after the initial peak, and then declining very slowly throughout the remainder of the exposure to KCl (Fig. 3A). In contrast to this, however, force typically rose slowly again after the initial peak and then was maintained constant, during a substantial period of time (Fig. 3B). We have shown previously (Wier et al. 2008) that the time courses of MLCK FRET ratio and Ca2+ (fura-2 fluorescence) are not affected by artery motion or fluorescence spectral overlap.

Figure 3
[Ca2+] and MLCK FRET ratio follow similar time courses during exposure to KCl, but force follows a different time course

Ca2+ sensitivity of contraction

The differences in time courses between MLCK FRET ratio and force could be regarded as prima facie evidence of changes in ‘Ca2+ sensitivity of contraction’. Indeed, pharmacological inhibition of rho-kinase with Y-27632 markedly inhibited the tonic component of KCl induced force, but had relatively much less effect on phasic and tonic fura-2 fluorescence or phasic and tonic MLCK FRET ratio (Fig. 4A and B).

Figure 4
Kinases involved in Ca2+ sensitivity of smooth muscle contraction in small mesenteric arteries

In some smooth muscles, Ca2+-dependent forms of PKC modulate Ca2+ sensitivity through a CPI-17 mediated inhibition of MLCP (Azam et al. 2007). Application of Gö6976 (1.0 μm, 10 min), which selectively inhibits PKCα and PKCβ (Gschwendt et al. 1996), had no effect, however, on the force, fluo-4 fluorescence or MLCK FRET ratio (Fig. 4C).

It has been noted (Pfitzer, 2001) that when intracellular [Ca2+] rises in smooth muscle, CaMK II may be activated, phosphorylate MLCK at Site A, and provide negative feedback on the activation of MLCK. This concept is based on earlier work showing that the MLCK activity (a measure of the affinity of MLCK for Ca2+/CaM) decreased during sustained elevation of intracellular [Ca2+] in uterine smooth muscle and that this phenomenon was blocked by inhibitors of CaMK II (Word et al. 1994). Earlier work had shown that Site A of MLCK was phosphorylated in a Ca2+-dependent manner (Tansey et al. 1994). The similarity in time course of Ca2+ indicator signals and MLCK FRET ratio during KCl exposure (Fig. 3) argues against major changes in affinity of MLCK for Ca2+/CaM during KCl exposure. If, however, CaMK II were significantly phosphorylating MLCK and decreasing the affinity of MLCK for Ca2+/CaM, then one would expect that application of a CaMK II inhibitor would increase MLCK FRET ratio and force, provided that Ca2+ remained the same. The CaMK II inhibitor KN-93 (10.0 μm), applied during KCl exposure, decreased Ca2+ indicator signals, MLCK FRET ratio and force all to a similar extent (Fig. 4D). The mechanisms by which KN-93 lowered intracellular [Ca2+] are not known, but may involve an effect on voltage-gated Ca2+ channels (Gao et al. 2006).

Activation of certain GPCRs is known to increase Ca2+ sensitivity of contraction, by activating rho-kinase and PKC signal transduction pathways. To investigate this with the system used here, we compared the MLCK FRET ratio signals activated by the α1-adrenoceptor (AR) agonist PE with those activated by exposure to KCl (Fig. 5). In each artery (n= 6), PE was used at a concentration that gave the same tonic force as that produced by our standard KCl protocol in that artery. During the tonic phase of force, the accompanying MLCK FRET ratio elicited by PE was 62% of that elicited by KCl, providing clear evidence of increased Ca2+ sensitivity during α1-AR activation, compared to KCl. Possible spatial inhomogeneities of the MLCK FRET ratio signal, due to possible agonist-induced inhomogeneities of [Ca2+], would not have been detected in these experiments, as discussed later.

Figure 5
Increased Ca2+ sensitivity of force after α1-AR activation as compared to KCl exposure in mesenteric arteries


It has long been known that ‘pharmacological’ elevation of cAMP results in PKA-dependent phosphorylation of MLCK (de Lanerolle et al. 1984) in biochemical measurements. The prevailing concept is that phosphorylation of MLCK by PKA reduces the affinity of MLCK for Ca2+/CaM, when Ca2+/CaM is not bound to MLCK. Here we have used KCl activation as a means to control [Ca2+] during elevation of cAMP and activation of PKA after application of forskolin to intact arteries. We have shown previously that forskolin does not change intracellular [Ca2+] under these circumstances (Wier et al. 2008). First, we examined phosphorylation of Ser 1760 (i.e. the regulatory site on MLCK) using a specific phospho-antibody (Fig. 6A). Important for the present study was the result that both the exogenous MLCK and the endogenous MLCK were both phosphorylated in the presence of forskolin, and this phosphorylation was inhibited by H-89 (Fig. 6A). Thus, changes in affinity of MLCK as a result of phosphorylation of Ser 1760 can be expected to result in changes in FRET. Furthermore, Ser 1760 was phosphorylated by exposure to forskolin also in the presence of KCl and these increases were blocked by the competitive antagonist of cAMP binding to PKA, rp-cAMP (100.0 μm, n= 3). Forskolin at 1.0 μm was applied for 2 min, after KCl had been elevated for 2 min. Arteries were pretreated, or not, with a selective PKA inhibitor (100.0 μm rp-cAMP for 10 min).

Figure 6
Forskolin causes increased phosphorylation of MLCK on Ser 1760, both exogenous biosensor and endogenous, and this effect is blocked by inhibition of PKA

Forskolin was applied in two fundamentally different ways to arteries during physiological experiments. In the first series of experiments (Fig. 7A), forskolin was applied when intracellular [Ca2+] was not simultaneously elevated by exposure to elevated KCl. Previous biochemical experiments have indicated that phosphorylation of MLCK by PKA results in a change of affinity of MLCK for Ca2+/CaM only when Ca2+/CaM is not bound to MLCK (Adelstein et al. 1982; Nishikawa et al. 1984). In this set of experiments, we applied forskolin (10.0 μm) prior to exposure to KCl (Fig. 7A). The test was preceded by a control exposure to elevated K+, in order to provide a control recording of force, fura-2 ratio and MLCK FRET ratio. When the same concentration of external K+ was applied in the presence of forskolin, the increase in MLCK FRET ratio was reduced, and force development was extremely small. In the presence of forskolin and elevated KCl, MLCK FRET ratio was approximately 50% of what it had been in elevated KCl in the absence of forskolin. A similar type of experiment (Fig. 7B) was also performed as follows: (1) an initial response to elevated KCl was obtained, (2) forskolin was applied coincidentally with removal of KCl, (3) KCl was re-applied to elevate [Ca2+] in the presence of forskolin, and, finally, (4) forskolin was then removed, in the continued presence of elevated KCl. The most important results are obtained at steps (3) and (4): presumed elevation of Ca2+ by KCl caused an increase in MLCK FRET ratio to about half what it was in the presence of elevated KCl alone, and removal of forskolin in the presence of elevated KCl restored the MLCK FRET ratio completely. These results are consistent with the concept that c-AMP-dependent protein kinase (PKA) can phosphorylate MLCK when the binding site on MLCK for Ca2+/CaM is not occupied, and decrease the affinity of that site for Ca2+/CaM, when Ca2+/CaM rises subsequently.

Figure 7
Forskolin causes relaxation and decreases MLCK FRET ratio, but does not change [Ca2+]

Although it has been reported from biochemistry experiments that phosphorylation of MLCK by PKA when Ca2+/CaM is bound does not change affinity of MLCK for Ca2+/CaM, the issue had not been investigated in intact arteries. To elevate cAMP at a time when Ca2+/CaM should be bound to MLCK, and thus reveal in real time possible changes in affinity of MLCK for Ca2+/CaM immediately, we applied forskolin (0.1 to 100.0 μm) during KCl induced contractions at a time when the MLCK FRET ratio and force had both reached approximately steady, elevated levels (Fig. 8B). At a concentration of 10.0 μm (n= 6), forskolin applied at this time induced a rapid initial relaxation, followed by a much slower phase of relaxation that proceeded until the artery had relaxed completely (Fig. 8Aa). When normalized appropriately, the rates of change of MLCK FRET ratio (dFRET/dt) and force (dForce/dt) (Fig. 8Ab) were remarkably similar in time course, with the changes in MLCK FRET ratio just preceding those in force. The subsequent much slower relaxation is seen in the dForce/dt as a subsequent slowly declining phase of dForce/dt. A similar slow change in dFRET/dt did not occur. These effects of forskolin occurred over the range of 0.3 μm to 100.0 μm (Fig. 8B). Both the maximum initial rapid relaxation (open circles) and the MLCK FRET ratio change were limited to a decrease of approximately 50% over that range of concentration of forskolin, while the slow relaxation (filled triangles) proceeded to 100% (i.e. complete relaxation).

Figure 8
Concentration dependence of forskolin's effect to cause a rapid decrease in force and MLCK FRET ratio during KCl exposure

Forskolin produces an increase in [cAMP] by a direct activation of adenylyl cyclase. Physiologically, increases in [cAMP] in smooth muscle are produced after activation of β-adrenergic receptors (e.g. β2-AR). Therefore, it was important to determine whether activation of β-AR has the same effects as forskolin. During the tonic phase of force produced by KCl or by PE, we applied the β-AR agonist isoproterenol (10.0 μm) (Fig. 9A). Isoproterenol caused a strong reduction in PE-induced force, [Ca2+] (fura-2), and MLCK FRET ratio, but not in the KCl-induced force, Ca2+, or MLCK FRET ratio (Fig. 9B). In contrast, isoproterenol did relax arteries activated by α1-AR, but this was accompanied by reductions in [Ca2+] and MLCK FRET ratio. The fact that isoproterenol was unable to produce significant relaxation in arteries activated by KCl suggests that the mechanism of relaxation induced by β-AR activation involves primarily Ca2+. Thus, these data do not present evidence for β-AR-mediated changes in the affinity of MLCK for Ca2+/CaM.

Figure 9
β-AR activation by isoproterenol caused a large reduction in the force, [Ca2+] and MLCK FRET ratio caused by exposure to PE but not to KCl


On the time scale used here, the time courses of [Ca2+] (fura-2 ratio) and activation of MLCK (MLCK FRET ratio) appear approximately similar after elevation of [Ca2+] by KCl (e.g. Fig. 3A and Fig. 2 of Wier et al. 2008). The kinetics of the formation of Ca2+/CaM and of the binding of Ca2+/CaM to MLCK within the intracellular environment are not yet known in detail. However, a recent model predicts that half-times of activation of MLCK after elevation of [Ca2+] should be less than 1 s when [Ca2+] is above about 0.2 μm (Fajmut et al. 2005). Thus, on the time scale used here, significant differences in the time courses of fura-2 ratio and MLCK FRET ratio probably represent phenomena other than kinetics of Ca2+/CaM binding to MLCK.

The time course of force was typically quite different from that of fura-2 ratio and MLCK FRET ratio, with force increasing (slightly) over very long periods of time when MLCK FRET ratio was decreasing. In the example of Fig. 3B, a slow, continuous decline in MLCK FRET ratio begins about 10 s after the peak in FRET ratio and continues for another 120 s. Force, on the other hand, begins a slow increase at about 40 s after KCl and continues until the end of the record, at 300 s. Such slow changes are unlikely to represent the kinetics of force generation after activation of MLCK (much faster).

Ca2+ sensitivity of contraction

The small difference in the normalized fura-2 ratio and the MLCK FRET ratio that develops about 10 s after KCl (Fig. 3A) could indicate a slightly decreased affinity of MLCK for Ca2+/CaM, possibly mediated by Ca2+/CaM-dependent kinase II (CaMK II). Because the differences in time courses of fura-2 ratio and MLCK FRET ratios during KCl that we observed are small, however, our data are not consistent with a scheme in which CaMK II has a large effect in ‘desensitization of contraction’ to Ca2+, as seems to occur in tracheal smooth muscle (Stull et al. 1990). Also, if significant CaMK II-mediated desensitization of contraction were occurring, one might have expected inhibition of CaMK II to cause an increase in MLCK activation and force, not a decrease (Fig. 4D). However, because [Ca2+] also decreased, net MLCK activation might have decreased, despite an increased affinity for Ca2+/CaM. Evaluation of this possibility would require a quantitative analysis beyond the scope of the present work. Nevertheless, force, Ca2+, and MLCK FRET ratio all decreased to approximately the same extent (Fig. 4D). Thus, the present data do not support the idea that CaMK II causes significant desensitization of contraction when [Ca2+] increases. Previous biochemical studies on a similar (i.e. vascular) smooth muscle (swine carotid artery) have reached similar conclusions to ours (Van Riper et al. 1995; Rokolya & Singer, 2000). In fact, it has been suggested that, rather than desensitize contraction to Ca2+, CaMK II activates contraction, through a mitogen activated protein kinase (MAPK)-dependent activation of MLCK (Kim et al. 2000). Our data could be consistent with such a mechanism, although the slow increase in force that would be attributable to this mechanism would seem to be small. Furthermore, our data show that this slow increase in force, without concomitant changes in MLCK activation may be due to rho-kinase mediated modulation of MLCP, as we discuss next.

Pharmacological inhibition of rho-kinase decreased strongly the tonic or slowly increasing component of KCl induced force generation in these arteries, without significant effects on [Ca2+] or MLCK activation (Fig. 4A, B). Because the initial peak of KCl-induced force was less affected, we suggest that rho-kinase is being activated during the KCl exposure. The effects on the initial peak force might be due to inhibition of a constitutive activity of rho-kinase. Activation of rho-kinase after GPCR activation, and the resulting ‘Ca2+ sensitization’ through inhibition of the targeting subunit, MYPT1, of MLCP are well characterized (Somlyo & Somlyo, 2003). It is now being recognized, however, that depolarization (KCl) is also a stimulus for ‘Ca2+ sensitization’ (Ratz et al. 2005; Dimopoulos et al. 2007). Ca2+-dependent activation of rho and rho-kinase has been reported in rabbit aortic smooth muscle cells (Sakurada et al. 2003) and depolarization induced contraction of rat caudal arteries also involves rho-kinase (Mita et al. 2002). Rho-kinase also activates CPI-17, the PKC (conventional and novel, c/n) activated 17 kDa inhibitor of MLCP. The importance of CPI-17 in mediating Ca2+ sensitization varies markedly with the type of smooth muscle, and its importance in rat mesenteric arteries is not yet clear. However, the inhibitor of conventional (i.e. Ca2+ activated) PKCs, Gö6976, had no effect (Fig. 4C), suggesting that in this tissue, CPI-17 is not prominently activated, either by Ca2+-dependent PKCs nor by rho-kinase, during membrane depolarization induced contractions.

The Ca2+ sensitivity of contractions activated by α1-AR appeared markedly higher than that of contractions activated by KCl (Fig. 5), as MLCK FRET ratios were markedly lower with α1-AR activation than with KCl. Indeed the concept that GPCR agonists activate increases in ‘Ca2+ sensitivity is well established (Somlyo & Somlyo, 2003). However, the interpretation of our observations is complicated by the fact that Ca2+ signalling activated by α1-AR can be heterogeneous, consisting under some circumstances of asynchronous propagating Ca2+ waves in different cells (Zang et al. 2001; Zacharia et al. 2007). Corresponding heterogeneity of the MLCK FRET ratio signals, if it had occurred, would not have been detectable in the spatially averaged fluorescence measurements made here. Averaging spatially inhomogeneous MLCK FRET fluorescence changes would be expected to result in a lower FRET signal than the peak values attained in individual cells.


Understanding the possible role of PKA in physiological regulation of MLCK has proven extremely difficult; PKA is involved in numerous cellular processes that affect intracellular [Ca2+] (and thus affect activation), such as entry of Ca2+ through voltage-dependent Ca2+ channels, availability of K+ channels, Ca2+ uptake and release in the SR, and Ca2+ extrusion from the cell. Furthermore, activation of PKA by cAMP will be accompanied by other actions of cAMP, such as the direct activation of nucleotide-gated ion channels, and activation of Epacs (guanine nucleotide exchange factor directly activated by cAMP) (Schmidt et al. 2007). Application of forskolin, thus increasing cAMP, during GPCR induced contraction certainly decreases [Ca2+] by these mechanisms and this decrease in [Ca2+] is a major contributor to relaxation. In contrast, we have shown previously (Wier et al. 2008), as have others (Porter et al. 2006), that application of forskolin during KCl exposure does not change the [Ca2+]. Since [Ca2+] does not change after application of forskolin, but MLCK FRET ratio decreases by up to 50%, the most straightforward interpretation of that result is that ‘pharmacological’ elevation of cAMP, through activation of adenylate cyclase (AC), decreases the affinity of the MLCK biosensor for Ca2+/CaM. Exposure of arteries to forskolin also causes increased phosphorylation of Ser 1760 on MLCK, and the effect is blocked by a competitive antagonist (rp-cAMP) of the cAMP-dependent activation of PKA.

The fact that MLCK FRET ratio could not be decreased by more than about 50% by forskolin might indicate that Ser 1760 phosphorylated MLCK retains an appreciable affinity for Ca2+/CaM. The rapid reduction in MLCK FRET ratio was accompanied closely in time by a rapid reduction in force, also to about 50% of the ‘control’ level, prior to application of forskolin. Other processes, apparently, activated a very much slower relaxation, which could proceed to complete relaxation, at the highest concentrations of forskolin used. Since intracellular [Ca2+] does not change under these circumstances, we suggest that these processes are most likely to be the well-known stimulatory effect of cAMP on MLCP (Somlyo & Somlyo, 2003).

We attempted to determine whether PKA mediated phosphorylation of MLCK is an important mechanism of force relaxation in response to physiological (e.g. GPCR mediated), as opposed to pharmacological, increases in cAMP. This appears not to be the case in tracheal smooth muscle (Miller et al. 1983). In the arteries studied here, presumed full activation of β-AR by high concentration of isoproterenol caused only very minor relaxation and change in [Ca2+] and MLCK FRET ratio, when [Ca2+] was elevated by exposure to KCl. In contrast, isoproterenol caused large decreases in [Ca2+], MLCK FRET ratio and force in arteries contracted by activation of adrenoceptors (α1-AR). Taken together, these results indicate that the major (physiological) mechanism of β-receptor/cAMP mediated relaxation in these arteries is a decrease in [Ca2+], not regulation of MLCK or even of MLCP. If cAMP/PKA-dependent stimulation of MLCP occurred, it was clearly insufficient to cause any relaxation, when [Ca2+] was constantly elevated. Finally, the results indicate that there is something very different about the production of cAMP by direct activation of adenylyl cyclase and by receptor mediated production of cAMP. We speculate that the concentrations of cAMP achieved by forskolin may be higher, and/or the spatial localization of the cAMP is different. Nevertheless, this remains a difficult issue to resolve because, as opposed to the case of KCl exposure, [Ca2+] decreases during physiological stimuli that increase [cAMP] and cause relaxation. When both [Ca2+] and MLCK FRET ratio decrease, any change in affinity of MLCK for Ca2+/CaM will be difficult to quantify, given that MLCK FRET ratio should decrease also as a result of less available Ca2+/CaM.


After depolarization, the activation of MLCK, as judged by the FRET ratio, appears to follow the [Ca2+] closely. There is little evidence of regulation of MLCK by other kinases. On the other hand, the time course of force differs from that of [Ca2+] and MLCK activation. While the kinetics of the force generating reactions may contribute to this difference, rho-kinase also contributes, probably through the well-known modulation of MLCP activity. The mechanism(s) of activation of rho-kinase are not known, but could be Ca2+ dependent. Ca2+ activated forms of PKC seem not to be involved. The affinity of MLCK for Ca2+/CaM can be regulated by ‘pharmacological’ increases in cAMP/PKA, even when Ca2+/CaM is already bound, but not by physiological increases after β-adrenoceptor activation. Analysis of the MLCK FRET ratio supports the concept that Ca2+ sensitivity of contraction activated by α1-AR is markedly higher than that during contraction activated by depolarization, even though Ca2+ sensitivity also increases after depolarization.


The MLCK Biosensor mice and constructive comments on the work were generously provided by Professor James T. Stull, University of Texas Southwestern Medical Center at Dallas, TX, USA. Thanks are extended to Dr Li Ping He for technical assistance. Financial support was provided by the National Heart, Lung and Blood Institute (NHLBI) (USA) HL 073094 and HL 078870.



Ca2+/CaM-dependent protein kinase II
cyan fluorescent protein
17 kDa protein kinase C-potentiated phosphatase inhibitor
fluorescence resonance energy transfer
G protein coupled receptor
myosin light chain kinase
myosin light chain phosphatase
p-21 activated protein kinase
protein kinase A
protein kinase C
myosin regulatory light chain
yellow fluorescent protein

Author contributions

HR and JZ performed the functional experiments on the isolated arteries, analyzed these data and contributed to writing the manuscript. ML performed and analyzed the biochemical experiments. HR measured physiological parameters of intact mice. WGW participated in experiments and was primarily responsible for experimental design, analysis of data, and writing of the manuscript. All authors read and approved the manuscript.


  • Adelstein RS, Sellers JR, Conti MA, Pato MD, de Lanerolle P. Regulation of smooth muscle contractile proteins by calmodulin and cyclic AMP. Fed Proc. 1982;41:2873–2878. [PubMed]
  • Azam MA, Yoshioka K, Ohkura S, Takuwa N, Sugimoto N, Sato K, Takuwa Y. Ca2+-independent, inhibitory effects of cyclic adenosine 5¢-monophosphate on Ca2+ regulation of phosphoinositide 3-kinase C2α, Rho, and myosin phosphatase in vascular smooth muscle. J Pharmacol Exp Ther. 2007;320:907–916. [PubMed]
  • de Lanerolle P, Nishikawa M, Yost DA, Adelstein RS. Increased phosphorylation of myosin light chain kinase after an increase in cyclic AMP in intact smooth muscle. Science. 1984;223:1415–1417. [PubMed]
  • Dimopoulos GJ, Semba S, Kitazawa K, Eto M, Kitazawa T. Ca2+-dependent rapid Ca2+ sensitization of contraction in arterial smooth muscle. Circ Res. 2007;100:121–129. [PMC free article] [PubMed]
  • Fajmut A, Jagodic M, Brumen M. Mathematical modeling of the myosin light chain kinase activation. J Chem Inf Model. 2005;45:1605–1609. [PubMed]
  • Gao L, Blair LA, Marshall J. CaMKII-independent effects of KN93 and its inactive analog KN92: reversible inhibition of L-type calcium channels. Biochem Biophys Res Commun. 2006;345:1606–1610. [PubMed]
  • Geguchadze R, Zhi G, Lau KS, Isotani E, Persechini A, Kamm KE, Stull JT. Quantitative measurements of Ca2+/calmodulin binding and activation of myosin light chain kinase in cells. FEBS Lett. 2004;557:121–124. [PubMed]
  • Gschwendt M, Dieterich S, Rennecke J, Kittstein W, Mueller HJ, Johannes FJ. Inhibition of protein kinase Cμ by various inhibitors. Differentiation from protein kinase C isoenzymes. FEBS Lett. 1996;392:77–80. [PubMed]
  • Isotani E, Zhi G, Lau KS, Huang J, Mizuno Y, Persechini A, Geguchadze R, Kamm KE, Stull JT. Real-time evaluation of myosin light chain kinase activation in smooth muscle tissues from a transgenic calmodulin-biosensor mouse. Proc Natl Acad Sci U S A. 2004;101:6279–6284. [PMC free article] [PubMed]
  • Kamm KE, Stull JT. Dedicated myosin light chain kinases with diverse cellular functions. J Biol Chem. 2001;276:4527–4530. [PubMed]
  • Kim I, Je HD, Gallant C, Zhan Q, Riper DV, Badwey JA, Singer HA, Morgan KG. Ca2+-calmodulin-dependent protein kinase II-dependent activation of contractility in ferret aorta. J Physiol. 2000;526:367–374. [PMC free article] [PubMed]
  • Miller JR, Silver PJ, Stull JT. The role of myosin light chain kinase phosphorylation in β-adrenergic relaxation of tracheal smooth muscle. Mol Pharmacol. 1983;24:235–242. [PubMed]
  • Mita M, Yanagihara H, Hishinuma S, Saito M, Walsh MP. Membrane depolarization-induced contraction of rat caudal arterial smooth muscle involves Rho-associated kinase. Biochem J. 2002;364:431–440. [PMC free article] [PubMed]
  • Mulvany MJ, Halpern W. Contractile properties of small arterial resistance vessels in spontaneously hypertensive and normotensive rats. Circ Res. 1977;41:19–26. [PubMed]
  • Nishikawa M, de Lanerolle P, Lincoln TM, Adelstein RS. Phosphorylation of mammalian myosin light chain kinases by the catalytic subunit of cyclic AMP-dependent protein kinase and by cyclic GMP-dependent protein kinase. J Biol Chem. 1984;259:8429–8436. [PubMed]
  • Pfitzer G. Invited review: regulation of myosin phosphorylation in smooth muscle. J Appl Physiol. 2001;91:497–503. [PubMed]
  • Porter M, Evans MC, Miner AS, Berg KM, Ward KR, Ratz PH. Convergence of Ca2+-desensitizing mechanisms activated by forskolin and phenylephrine pretreatment, but not 8-bromo-cGMP. Am J Physiol Cell Physiol. 2006;290:C1552–1559. [PubMed]
  • Ratz PH, Berg KM, Urban NH, Miner AS. Regulation of smooth muscle calcium sensitivity: KCl as a calcium- sensitizing stimulus. Am J Physiol Cell Physiol. 2005;288:C769–783. [PubMed]
  • Rokolya A, Singer HA. Inhibition of CaM kinase II activation and force maintenance by KN-93 in arterial smooth muscle. Am J Physiol Cell Physiol. 2000;278:C537–545. [PubMed]
  • Sakurada S, Takuwa N, Sugimoto N, Wang Y, Seto M, Sasaki Y, Takuwa Y. Ca2+-dependent activation of Rho and Rho kinase in membrane depolarization-induced and receptor stimulation-induced vascular smooth muscle contraction. Circ Res. 2003;93:548–556. [PubMed]
  • Schmidt M, Sand C, Jakobs KH, Michel MC, Weernink PA. Epac and the cardiovascular system. Curr Opin Pharmacol. 2007;7:193–200. [PubMed]
  • Somlyo AP, Somlyo AV. Ca2+ sensitivity of smooth muscle and nonmuscle myosin II: modulated by G proteins, kinases, and myosin phosphatase. Physiol Rev. 2003;83:1325–1358. [PubMed]
  • Stull JT, Hsu LC, Tansey MG, Kamm KE. Myosin light chain kinase phosphorylation in tracheal smooth muscle. J Biol Chem. 1990;265:16683–16690. [PubMed]
  • Tansey MG, Luby-Phelps K, Kamm KE, Stull JT. Ca2+-dependent phosphorylation of myosin light chain kinase decreases the Ca2+ sensitivity of light chain phosphorylation within smooth muscle cells. J Biol Chem. 1994;269:9912–9920. [PubMed]
  • Van Riper DA, Weaver BA, Stull JT, Rembold CM. Myosin light chain kinase phosphorylation in swine carotid artery contraction and relaxation. Am J Physiol Heart Circ Physiol. 1995;268:H2466–2475. [PubMed]
  • Wier WG, Rizzo MA, Raina H, Zacharia J. A technique for simultaneous measurement of Ca2+, FRET fluorescence and force in intact mouse small arteries. J Physiol. 2008;586:2437–2443. [PMC free article] [PubMed]
  • Word RA, Tang DC, Kamm KE. Activation properties of myosin light chain kinase during contraction/relaxation cycles of tonic and phasic smooth muscles. J Biol Chem. 1994;269:21596–21602. [PubMed]
  • Yoshioka K, Sugimoto N, Takuwa N, Takuwa Y. Essential role for class II phosphoinositide 3-kinase α-isoform in Ca2+-induced, Rho- and Rho kinase-dependent regulation of myosin phosphatase and contraction in isolated vascular smooth muscle cells. Mol Pharmacol. 2007;71:912–920. [PubMed]
  • Zacharia J, Zhang J, Wier WG. Ca2+ signaling in mouse mesenteric small arteries: myogenic tone and adrenergic vasoconstriction. Am J Physiol Heart Circ Physiol. 2007;292:H1523–1532. [PubMed]
  • Zang WJ, Balke CW, Wier WG. Graded α1-adrenoceptor activation of arteries involves recruitment of smooth muscle cells to produce ‘all or none’ Ca2+ signals. Cell Calcium. 2001;29:327–334. [PubMed]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...