A force-activated kinase in a catch smooth muscle

Abstract

Permeabilized anterior byssus retractor muscles (ABRM) from Mytilus edulis were used as a simple system to test whether there is a stretch dependent activation of a kinase as has been postulated for titin and the mini-titin twitchin. The ABRM is a smooth muscle that shows catch, a condition of high force maintenance and resistance to stretch following stimulation when the intracellular Ca(++) concentration has diminished to sub-maximum levels. In the catch state twitchin is unphosphorylated, and the muscle maintains force without myosin crossbridge cycling through what is likely a twitchin mediated tether between thick and thin filaments. In catch, a small change in length results in a large change in force. The phosphorylation state of an added peptide, a good substrate for molluscan twitchin kinase, with the sequence KKRAARATSNVFA was used as a measure of kinase activation. We find that there is about a two-fold increase in phosphorylation of the added peptide with a 10% stretch of the ABRM in catch. The increased phosphorylation is due to activation of a kinase rather than to an inhibition of a phosphatase. The extent of phosphorylation of the peptide is decreased when twitchin is phosphorylated and catch force is not present. However, there is also a large increase in peptide phosphorylation when the muscle is activated in pCa 5, and the catch state does not exist. The force-sensitive kinase activity is decreased by ML-9 and ML-7 which are inhibitors of twitchin kinase, but not by the Rho kinase inhibitor Y-27632. There is no detectable phosphorylation of myosin light chains, but the phosphorylation of twitchin increases by a small, but significant extent with stretch. It is possible that twitchin senses force output resulting in a force-sensitive twitchin kinase activity that results in autophosphorylation of twitchin on site(s) other than those responsible for relaxation of catch.

Fig. 1

Dependence of peptide phosphorylation on muscle stretch and force output. Panel a Experimental design showing typical force responses in stretch and release protocols. Permeabilized muscles were activated in pCa 5 followed by addition of blebbistatin (50 µM). The [Ca⁺²] was then decreased to pCa > 8. One muscle (red dashed line) was stretched by 5% as indicated and then released by 10%. The other muscle (blue line) was released by 5% followed by a stretch by 10%. The solutions containing ³²P-ATP and peptide are identified as 1, 2, and 3. Panel b Typical curves showing the radioactivity following chromatographic separation of the peptide in solutions 2 and 3. Results are shown for the muscle while released (dashed line) and while stretched (solid line). Panel c Relative phosphorylation of the peptide in different experimental designs. In the stretch versus release protocols, the phosphorylation of the peptide when the muscle was stretched is reported relative to when released. In the pCa 5 versus pCa 8 protocols, the phosphorylation of the peptide when the muscle was in pCa 5 is reported relative to that in pCa 8. See text for details of experimental designs. Data are shown as mean ± SEM

Fig. 2

Typical liquid chromatographic runs on muscle incubation solutions containing ³²P-labeled phosphorylated peptide. Fractions containing ³²P-Pi and ³²P-labeled phosphorylated peptide are shown. The muscle was incubated in a solution containing ³²P-labeled phosphorylated peptide under either stretched (blue) or released (black) conditions. Also shown is a chromatogram of a solution containing the ³²P-labeled phosphorylated peptide, but not used for muscle incubation (red)

Fig. 3

Effect of kinase inhibitors ML-9 (300 µM), ML-7 (20 µM), and Y-27632 (20 µM) on peptide phosphorylation response to stretch. Muscles were put into the catch state as shown in Fig. 1, Panel a, and then subjected to a 5% decrease in muscle length followed by a 30 min incubation and determination of peptide phosphorylation. The inhibitor was then added and measurement of peptide phosphorylation made at the released length and following stretch by 10%. The phosphorylation of the peptide in the presence of the inhibitor is shown relative to that at the released length without the inhibitor. Open bar released conditions; Shaded bar stretched conditions; Hatched bar difference between stretched and released conditions. Also shown are similar data with muscles not subjected to inhibitor treatment. *Released muscle with inhibitor significantly lower than no inhibitor. #Difference between stretched and released muscle with inhibitor significantly lower than no inhibitor. + Difference between stretched and released muscle with inhibitor significantly higher than no inhibitor. Data are mean ± SEM

Fig. 4

³²P incorporation into proteins as a result of stretch. Panel a shows typical Coomassie Blue stained gels and autoradiograms of proteins from muscles put into catch and then incubated in ³²P-ATP for 30 min under either stretched (Str) or released (Rel) conditions as shown in Fig. 1, Panel a. Gradient gels of 4–15% acrylamide were used. Panel b shows similar data for muscles in a solution with pCa > 8 and ³²P-ATP either kept at slack length (−Str) or stretched (+Str) by 10%. The muscles were frozen after 8 min. 5% acrylamide gels were used. The regions containing twitchin (Twit), myosin heavy chain (MHC), and paramyosin (Para) are identified. Panel c Quantification of the effect of stretch on radioactivity in twitchin in the design described in Panel b. The radioactivity in twitchin is normalized to the Coomassie Blue staining of twitchin. Data are reported relative to the mean ratio in the non-stretched muscles. Data are mean ± SEM. *P < 0.01