Ca2+-independent phosphorylation of myosin in rat caudal artery and chicken gizzard myofilaments

Abstract

1. Smooth muscle contraction is activated primarily by the Ca2+-calmodulin (CaM)-dependent phosphorylation of the 20 kDa light chains (LC20) of myosin. Activation can also occur in some instances without a change in intracellular free [Ca2+] or indeed in a Ca2+-independent manner. These signalling pathways often involve inhibition of myosin light chain phosphatase and unmasking of basal kinase activity leading to LC20 phosphorylation and contraction. 2. We have used demembranated rat caudal arterial smooth muscle strips and isolated chicken gizzard myofilaments in conjunction with the phosphatase inhibitor microcystin-LR to investigate the mechanism of Ca2+-independent phosphorylation of LC20 and contraction. 3. Treatment of Triton X-100-demembranated rat caudal arterial smooth muscle strips with microcystin at pCa 9 triggered a concentration-dependent contraction that was slower than that induced by pCa 4.5 or 6 but reached comparable steady-state levels of tension. 4. This Ca2+-independent, microcystin-induced contraction correlated with phosphorylation of LC20 at serine-19 and threonine-18. 5. Whereas Ca2+-dependent LC20 phosphorylation and contraction were inhibited by a synthetic peptide (AV25) based on the autoinhibitory domain of myosin light chain kinase (MLCK), Ca2+-independent, microcystin-induced LC20 phosphorylation and contraction were resistant to AV25. 6. Ca2+-independent LC20 kinase activity was also detected in chicken gizzard smooth muscle myofilaments and catalysed phosphorylation of endogenous myosin LC20 at serine-19 and/or threonine-18. This is in contrast to MLCK which phosphorylates threonine-18 only after prior phosphorylation of serine-19. 7. Gizzard Ca2+-independent LC20 kinase could be separated from MLCK by differential extraction from myofilaments and by CaM affinity chromatography. Its activity was resistant to AV25. 8. We conclude that inhibition of smooth muscle myosin light chain phosphatase (MLCP) unmasks the activity of a Ca2+-independent LC20 kinase associated with the myofilaments and distinct from MLCK. This kinase, therefore, probably plays a role in Ca2+ sensitization and Ca2+-independent contraction of smooth muscle in response to stimuli that act via Ca2+-independent pathways, leading to inhibition of MLCP.

Figure 18. Separation of MLCK and Ca²⁺-independent LC₂₀ kinase by CaM affinity chromatography

A, the 2 M NaCl extract of chicken gizzard myofilaments (E2) was subjected to CaM affinity chromatography following partial purification by Q-Sepharose anion-exchange chromatography as described in Methods. Proteins bound in the presence of Ca²⁺ were eluted from the affinity column by application of EGTA-containing buffer (indicated by the arrow). Fractions were assayed for their ability to phosphorylate myosin in the absence (○) and presence of Ca²⁺ (•). B, Western blot with anti-MLCK of 10 μl of the column load (lane 1), the pooled flow-through fractions (lane 2) and the pooled EGTA eluate (lane 3). C, autoradiograph of reaction mixtures assayed in the absence of Ca²⁺ (see A) and subjected to SDS-PAGE. B, blank (no column fraction); L, column load; fractions indicated by number. The sizes of M_r marker proteins are shown on the left in kDa.

Figure 1. Ca²⁺-dependent and -independent contraction of demembranated rat caudal arterial smooth muscle

Triton X-100-demembranated rat caudal arterial smooth muscle strips were stimulated to contract at time zero by increasing the free [Ca²⁺] to 1 μM (pCa 6) (○) or 30 μM (pCa 4.5) (•) (A) or by addition of 10 μM microcystin at pCa 9 (•) (B). Tension was recorded until steady-state levels were attained. B, inset: the contractile response at pCa 9 to 0.01 (□), 0.1 (▪), 1 (○) and 10 μM microcystin (•).

Figure 2. Ca²⁺-independent LC₂₀ phosphorylation in demembranated rat caudal arterial smooth muscle

Triton X-100-demembranated rat caudal arterial smooth muscle strips were incubated for 60 min with [γ-³²P]ATP at pCa 9 without (lanes 2 and 4) or with (lanes 3 and 5) 1 μM microcystin. Tissue strips were processed for SDS-PAGE and autoradiography as described in Methods. Lanes 1-3: Coomassie Blue-stained gels of M_r marker proteins, control muscle strip and muscle strip treated with microcystin, respectively. Lanes 4 and 5: autoradiogram showing phosphoproteins in control and microcystin-treated muscle strips, respectively. The sizes of M_r markers are shown on the left in kDa. The position of LC₂₀ is indicated at the right. Results are representative of five similar experiments. From densitometric scanning of the actin bands in lanes 2 and 3, similar amounts of tissue protein were loaded: (optical density × mm²) values for the actin band were 34.7 and 38.1 in lanes 2 and 3, respectively.

Figure 3. Identification of the 20 kDa phosphoprotein as LC₂₀

Triton X-100-demembranated rat caudal arterial smooth muscle strips were incubated for 60 min with [γ-³²P]ATP at pCa 9 (A, lane 1), pCa 6 (A, lane 2) or pCa 9 + microcystin (A, lane 3 and B, lanes 1 and 2) and proteins were separated by SDS-PAGE. A, phosphoproteins were detected by autoradiography. B, Western blots of a tissue sample identical to that in lane 3 (A) using antibody pLC1 that specifically recognizes LC₂₀ phosphorylated at serine-19 (lane 1) or antibody pLC2 that specifically recognizes LC₂₀ phosphorylated at both serine-19 and threonine-18 (lane 2).

Figure 4. Time course of LC₂₀ phosphorylation in response to microcystin treatment at pCa 9

Triton X-100-demembranated rat caudal arterial smooth muscle strips were incubated for 0-40 min with [γ-³²P]ATP in pCa 9 solution containing 1 μM microcystin. LC₂₀ phosphorylation was analysed by SDS-PAGE and autoradiography. Incubation times were: 0 (lane 1), 2 min (lane 2), 5 min (lane 3), 10 min (lane 4), 20 min (lane 5) and 40 min (lane 6). The sizes of M_r markers are shown on the left in kDa.

Figure 5. Ca²⁺-independent LC₂₀ phosphorylation in response to microcystin treatment occurs at serine-19 and threonine-18

Triton X-100-demembranated rat caudal arterial smooth muscle strips were incubated for 60 min at pCa 9 (lane 1) or pCa 9 with 10 μM microcystin (lanes 2-4). Unphosphorylated (LC₂₀) and phosphorylated forms of LC₂₀ (P₁-LC₂₀ and P₂-LC₂₀) were separated by urea/glycerol gel electrophoresis and identified by Western blotting using a general antibody to LC₂₀ (lanes 1 and 2) or antibodies specific for LC₂₀ phosphorylated exclusively at serine-19 (pLC1; lane 3) or both serine-19 and threonine-18 (pLC2; lane 4). Samples of 30 μl were applied per lane.

Figure 6. Myosin light chain kinase is inactive in the absence of Ca²⁺

Purified MLCK was incubated in the absence (○) and presence of Ca²⁺ (•) at a kinase concentration of 0.25 μg ml⁻¹ (A) or 10 μg ml⁻¹ (B), using LC₂₀ (0.2 mg ml⁻¹) as substrate. Samples (20 μl) of reaction mixtures were spotted onto P81 phosphocellulose paper at the indicated times for quantification of LC₂₀ phosphorylation.

Figure 7. Autophosphorylation of MLCK correlates with LC₂₀ phosphorylation in the absence of Ca²⁺

A, correlation between MLCK autophosphorylation (○) and Ca²⁺-independent phosphorylation of isolated LC₂₀ at 15 μg ml⁻¹ (•). B, correlation between MLCK autophosphorylation (○) and Ca²⁺-independent phosphorylation of intact myosin at 0.15 mg ml⁻¹ (•). Assays were carried out at 10 μg ml⁻¹ MLCK in the absence of Ca²⁺. Insets: autoradiographs of reaction mixtures in the presence of Ca²⁺ (lanes 1) or absence of Ca²⁺ (lanes 2) after 120 min. A, isolated LC₂₀ substrate; B, intact myosin substrate. Note that autophosphorylation of MLCK occurs only in the absence of Ca²⁺ (compare lanes 1 and 2 in each panel).

Figure 8. Inhibition of MLCK by synthetic peptides based on the autoinhibitory domain

MLCK (10 μg ml⁻¹) was incubated with the indicated concentrations of SM-1 (○) or AV25 (•) in the presence (A) or absence (B) of Ca²⁺. At 5 min (A) or 2 h (B,) samples (20 μl) of reaction mixtures were spotted onto P81 phosphocellulose paper for quantification of substrate (0.1 mM MLC_11-23) phosphorylation. Activities are expressed as a percentage of the activity in the absence of peptide inhibitors (65.2 mol P_i min⁻¹ (mol MLCK)⁻¹ in the presence of Ca²⁺ and 0.21 mol P_i min⁻¹ (mol MLCK)⁻¹ in the absence of Ca²⁺).

Figure 9. Lack of effect of SM-1 or AV25 on MLCK autophosphorylation

MLCK (10 μg ml⁻¹) was incubated for 2 h under phosphorylating conditions with LC₂₀ (0.1 mg ml⁻¹) and various concentrations of SM-1 or AV25 in the absence of Ca²⁺. Samples (30 μl) of reaction mixtures were subjected to SDS-PAGE and autoradiography. A, Coomassie Blue-stained gel; B, autoradiograph. Lane 1: M_r marker proteins; lanes 2-6: 0, 0.1, 1, 10 and 100 μM SM-1, respectively; lanes 7-10: 0.1, 1, 10 and 100 μM AV25, respectively. Scintillation counting of MLCK bands indicated no effect of SM-1 or AV25 on autophosphorylation. Scintillation counting of LC₂₀ bands indicated 42 % inhibition of phosphorylation at 100 μM SM-1 and 49 % inhibition at 100 μM AV25.

Figure 10. Effect of AV25 on intact and constitutively active MLCK

A, the activity of intact MLCK (5 μg ml⁻¹; 47 nM) was assayed in the presence of Ca²⁺ without (○) or with 75 μM AV25 (•). B, the activity of constitutively active (trypsin-digested) MLCK (47 nM) was assayed in the absence (□, ▪) and presence of Ca²⁺ (○, •) without (○, □) or with 75 μM AV25 (•, ▪). LC₂₀ (0.1 mg ml⁻¹) was used as substrate.

Figure 11. Preparation of chicken gizzard myofilaments

A, chicken gizzard myofilaments were prepared as described in Methods. Samples (3 μl each) of fractions obtained during the preparation were analysed by SDS-PAGE and Coomassie Blue staining. Lane 1, chicken gizzard homogenate; lane 2, supernatant after first centrifugation; lane 3, first wash of pellet; lane 4, second wash of pellet; lane 5, third wash of pellet; lane 6, myofilaments. The sizes of M_r markers are indicated on the left in kDa. The major proteins in the myofilament preparation are identified at the right. B, Western blot with anti-MLCK. Lane 1, chicken gizzard homogenate; lane 2, supernatant after first centrifugation; lane 3, first wash of pellet; lane 4, myofilaments. The sizes of prestained M_r markers are shown on the left in kDa.

Figure 12. Identification of Ca²⁺-independent LC₂₀ kinase activity in chicken gizzard myofilaments

Myosin phosphorylation in chicken gizzard myofilaments was quantified following SDS-PAGE and autoradiography of reaction mixtures in the presence of Ca²⁺ (A) or EGTA (B). The stoichiometry of LC₂₀ phosphorylation was determined by cutting the radiolabelled LC₂₀ band out of the gel and scintillation counting. LC₂₀ phosphorylated to exactly 1.0 mol P_i (mol LC₂₀)⁻¹ was used as a reference to calculate the stoichiometry of LC₂₀ phosphorylation. Reactions were carried out in the absence (○) or presence of 10 μM microcystin (•). C, SDS-PAGE and autoradiography of the reaction mixtures containing microcystin and either Ca²⁺ (2.5 min) (lanes 1 and 3) or EGTA (120 min) (lanes 2 and 4). Lanes 1 and 2, Coomassie Blue-stained gels; lanes 3 and 4, autoradiographs. D, urea/glycerol gel electrophoresis and Western blotting with the general LC₂₀ antibody (a), pLC1 (b) or pLC2 (c). Lane 1, untreated myofilaments; lanes 2 and 3, incubated with Ca²⁺ and microcystin for 5 and 120 min, respectively; lanes 4 and 5, incubated with EGTA and microcystin for 30 and 120 min, respectively.

Figure 13. Ca²⁺-dependent and -independent phosphorylation of isolated LC₂₀ by MLCK and of myofilament LC₂₀ by endogenous kinases

Time courses of LC₂₀ phosphorylation were analysed by urea/glycerol gel electrophoresis and autoradiography: A, phosphorylation of isolated LC₂₀ by MLCK in the presence of Ca²⁺; B, phosphorylation of isolated LC₂₀ by MLCK in the absence of Ca²⁺; C, phosphorylation of LC₂₀ in myofilaments by endogenous MLCK in the presence of Ca²⁺; D, phosphorylation of LC₂₀ in myofilaments by endogenous Ca²⁺-independent LC₂₀ kinase in the absence of Ca²⁺. Lane 1, 5 min; lane 2, 30 min; lane 3, 60 min; lane 4, 120 min. Exposure times were varied to give similar signal intensities.

Figure 14. Phosphopeptide mapping of isolated LC₂₀ phosphorylated by MLCK and of myofilament P₁-LC₂₀ phosphorylated by endogenous kinases

P₁-LC₂₀, separated by urea/glycerol gel electrophoresis, was digested with TPCK-trypsin in the gel slice and peptides were separated by two-dimensional thin-layer electrophoresis/chromatography. Phosphopeptides were identified by autoradiography. A, isolated LC₂₀ phosphorylated by MLCK in the presence of Ca²⁺ (see P₁-LC₂₀ in Fig. 13A). B, isolated LC₂₀ phosphorylated by PKC; C, isolated LC₂₀ phosphorylated by MLCK in the absence of Ca²⁺ (see P₁-LC₂₀ in Fig. 13B). D, a mixture of A and C.E, myofilament LC₂₀ phosphorylated in the presence of Ca²⁺ (see P₁-LC₂₀ in Fig. 13C). F, a mixture of A and E.G, myofilament LC₂₀ phosphorylated in the absence of Ca²⁺ (see P₁-LC₂₀ in Fig. 13D). H, a mixture of A and G. Exposure times were longer than in the experiment of Fig. 13 to obtain clear signals.

Figure 15. Phosphoamino acid analysis of LC₂₀ phosphopeptides

The phosphopeptides from Fig. 14 were hydrolysed to their constituent amino acids and phosphoserine (PS), phosphothreonine (PT) and phosphotyrosine (PY) were separated by two-dimensional electrophoresis. Added unlabelled phosphoamino acid standards were detected by ninhydrin staining (left panels) and radiolabelled phosphoamino acids derived from phosphorylated LC₂₀ were detected by autoradiography (right panels). A, isolated LC₂₀ phosphorylated by MLCK in the presence of Ca²⁺ (see P₁-LC₂₀ in Fig. 13A);B, isolated LC₂₀ phosphorylated by MLCK in the absence of Ca²⁺ (see P₁-LC₂₀ in Fig. 13B); C, myofilament LC₂₀ phosphorylated in the presence of Ca²⁺ (see P₁-LC₂₀ in Fig. 13C); D, myofilament LC₂₀ phosphorylated in the absence of Ca²⁺ (see P₁-LC₂₀ in Fig. 13D).

Figure 16. Effect of kinase inhibitors on Ca²⁺-independent LC₂₀ kinase activity in myofilaments

Myofilament Ca²⁺-independent LC₂₀ kinase activity was assayed in the absence of Ca²⁺ for 120 min at the indicated concentrations of staurosporine (•), AV25 (○) or [Ala9]autocamtide 2, either alone (□) or with 50 μM AV25 (▵).

Figure 17. Extraction of MLCK and Ca²⁺-independent LC₂₀ kinase from myofilaments under different conditions

A, Coomassie Blue-stained SDS-PAGE of chicken gizzard myofilaments (lane 1), the 30 mM Mg²⁺ extract (E1; lane 2), the 2 M NaCl extract (E2; lane 3) and the final pellet after the two extractions (lane 4). Loading level, 3 μl of protein sample per lane. Prestained M_r markers are indicated on the left. B, Western blot with anti-MLCK. Lanes 1-4 as in A. Prestained M_r markers are again indicated on the left.