INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

SMOOTH MUSCLE CONTRACTION controls the tone of many involuntary muscles, such as those lining arteries and the walls of the stomach. On the basis of their contraction profiles in response to K⁺ depolarization (43), smooth muscles are divided into two general contractile phenotypes: tonic and phasic. Tonic smooth muscles demonstrate slow rates of force activation and relaxation, lower maximum speeds of shortening (V_max) as well as force maintenance, whereas phasic smooth muscles display rapid rates of force activation and relaxation and have a high V_max but poor force maintenance (reviewed in Ref. 15). In both smooth muscle types, phosphorylation of the 20-kDa regulatory myosin light chains (MLC₂₀) by Ca²⁺-calmodulin-activated myosin light chain kinase (MLCK) promotes actomyosin cycling to produce force. Although previous studies have suggested that isoform expression of myosin and actin filament-associated proteins are molecular determinants of the contractile properties of smooth muscle (46), the molecular determinants of the unique contractile properties of phasic versus tonic smooth muscle phenotypes remain unclear.

Proteins controlling the activation of the myosin ATPase activity, namely MLCK and myosin light chain phosphatase (MLCP), have been suggested as candidates that define contractile properties in phasic and tonic smooth muscles (12, 23). These proteins control MLC₂₀ phosphorylation and therefore the level of activation. However, the unique rates of force activation in phasic and tonic smooth muscles are not simply explained by differences in Ca²⁺ transients or rates of MLC₂₀ phosphorylation because phasic tissues maintain higher rates of force activation when skinned or when the MLC₂₀ is thiophosphorylated to dissociate force activation from MLCK and MLCP kinetics (22). Nonetheless, maximum shortening velocity (V_max) in smooth muscle is related to the level of activation as defined by MLC₂₀ phosphorylation (1, 10). Furthermore, it has been proposed that MLCP may play a role in the regulation of relaxation because protein kinase C activation and subsequent MLCP inhibition decreases the rate of MLC₂₀ dephosphorylation and force relaxation (25, 34), parameters thought to reflect a decrease in MLCP catalytic activity. Moreover, the MLCP holoenzyme may exist as isoforms, which differ in the expression of two unique myosin binding subunits (M130/M133; Ref. 41). Therefore, in addition to contractile protein isoform diversity, the control of actomyosin kinetics in smooth muscle may also rely on factors modulating MLC₂₀ phosphorylation.

The smooth muscle myosin heavy chain (MHC) is a single gene with two pairs of alternatively spliced exons. Two isoforms, named SM1 and SM2, differ by expression of a 43- or 9-amino acid COOH-terminus, respectively (5). Their contribution to contractility remains controversial (2, 27, 36). In addition, alternative splicing of an exon near the NH₂-terminal 25- to 50-kDa junction of the myosin head accounts for a 7-amino acid insert in visceral SM1 and SM2 MHCs (28). MHC isoform diversity at the NH₂-terminus has been hypothesized as a determinant of the tonic and phasic contractile properties in smooth muscle (28, 30, 46). Additionally, the 17-kDa essential myosin light chains (MLC₁₇) associated with the neck region of MHCs also show isoform diversity. Two splice variants of MLC₁₇ have been found in smooth muscle, and they differ by inclusion (MLC_17b) or exclusion (MLC_17a) of a 39-bp exon (38). Studies have shown a correlation between increased MLC_17b expression and decreased actin activated Mg²⁺-ATPase activity of myosin (18) as well as decreased V_max in intact tissues (31).

We assumed that the major factors controlling the contractile properties of developing gizzard and aortic smooth muscle would be the myosin isoforms (MHC and MLC₁₇) or proteins controlling myosin activation through MLC₂₀ phosphorylation (MLCK and MLCP). Because MHC and MLC₁₇ isoforms switch during development of the embryonic chicken aorta and gizzard (7), we hypothesized that changes in protein isoform expression throughout development would be reflected by changes in the contractile phenotype. Therefore, we investigated MHC, MLC₂₀, MLC₁₇, MLCK, MLCP, and thin filament-associated proteins to determine the markers for phasic and tonic contractile properties. In the aorta and gizzard, the tonic phenotype was the default during early embryonic development. Approximately 1 wk before hatching, the phasic phenotype developed in the gizzard. We demonstrate that the MLC₁₇ and MLCP isoforms are molecular markers for the phasic contractile phenotype. Expression of MLC_17a defined the rapid rate of force activation, whereas expression of a gizzard-specific MLCP isoform was a determinant for the rapid rate of relaxation, both hallmarks of phasic smooth muscles.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Identification of regulatory protein isoforms in the embryonic chicken. Fertilized chicken eggs were incubated at 37°C in humidified air and collected at embryonic days (ED) 10, 12, 14, 16, 18, and 20. The gizzard and thoracic aortas from five or more embryos were dissected in Ca²⁺-free physiological saline solution (in mM: 140 NaCl, 4.7 KCl, 0.3 Na₂HPO₄, 2 MOPS, 0.5 EGTA, 0.2 EDTA, 1.2 MgCl₂, 5.6 glucose, final pH adjusted to 7.4 with 1 N KOH). The tissues were immediately homogenized in SDS-PAGE buffer, heated to 80°C for 5 min, and clarified by centrifugation at 14,000 g in a microcentrifuge. The samples were resolved by either 12% SDS-PAGE with an acrylamide-to-bisacrylamide ratio of 29:1 (calponin, MLC₂₀) or 14% SDS-PAGE with an acrylamide-to-bisacrylamide ratio of 180:1 (MLCK, myosin binding subunit of MLCP). To ensure comparable sample loading, the samples were normalized according to the G-actin band in SDS-PAGE. Once sample loadings were comparable, the SDS-PAGE-resolved proteins were transferred to 0.22-µm nitrocellulose membrane in a transfer buffer (100 mM glycine, 10 mM Tris, pH 8.6, 20% methanol) and blocked at room temperature for 3 h using Tris-buffered saline (TBS; 150 mM NaCl, 50 mM Tris · HCl, pH 7.4) containing 1% BSA. Blocked nitrocellulose membranes were incubated overnight at 4°C in TBS-0.1% BSA with monoclonal antibodies (MAbs) against calponin (CP1; 26), MLC₂₀ (My-21; Sigma Immunochemicals), MLCK (LKH3 and LKH18; Ref. 6), or MLCP myosin binding subunit (Covance). For MLCK identification, a purified smooth muscle MLCK protein sample was used as a control. The membranes were washed three times for 10 min in TBS containing either 0.05% Tween 20 or 0.5% Triton X-100-0.05% SDS and then rinsed with TBS. The membranes were subsequently incubated with TBS-0.1% BSA containing alkaline phosphatase-conjugated goat anti-mouse IgG or IgM second antibody for 50 min at room temperature. After membranes were washed after second antibody incubation, the color reaction was developed in alkaline phosphatase substrate buffer (in mM: 110 NaCl, 5 MgCl₂, 100 Tris, final pH 9.5) containing 0.04% (wt/vol) nitro blue tetrazolium and 0.02% (wt/vol) 5-bromo-4-chloro-3-indolylphosphate.

Force of contraction of developing chicken gizzard and aorta. Embryonic chicken gizzard and aorta strips were excised from embryos at various stages of development, skinned, and allowed to contract in high Ca²⁺ (0.1 mM free Ca²⁺) and relax in low Ca²⁺ (1 nM free Ca²⁺) solutions. The solutions were mixed according to a computer program that calculated the amount of stock solutions required for a given set of free ion concentrations (3). The binding constants for the ionic species present were corrected for temperature and ionic strength. The strips were clipped between two aluminum foil T clips (11) and skinned at room temperature for 15 (gizzard) or 30 min (aorta) in a pCa9 solution [in mM: 5 Na₂ATP, 5 EGTA, 25 potassium methane sulfonate, pH 7, 6.9 MgCl₂ (1 mM free Mg²⁺), 25 creatine phosphate, 2 glutathione, and 400 µM -escin, final pH to 7 with 1 N KOH]. The times used for skinning provided optimal Ca²⁺ activated force per cross section for the preparations; longer times compromised the force response. After skinning, the tissues were transferred to pCa9 relaxing solution [in mM: 5.4 Na₂ATP, 5 EGTA, 71 potassium methane sulfonate, 7.1 MgCl₂ (1 free Mg²⁺), 25 creatine phosphate, 25 N,N-bis(2-hydroxyethyl)-2-aminoethanesulfonic acid (BES), and 12 µM CaCl₂, final pH to 7 with 1 N KOH] and mounted on the stage of a computer-controlled mechanics workstation. One end of the tissue strip was hooked to a length driver (Physik Instrumente, Waldbron, Germany) and the other to a force transducer (Akers AE 801 Sensonor, Horten, Norway). The strips were stretched to 1.3 times their resting length (or to the length for maximal force development) while in relaxing solution. Contractions were initiated by transferring the strips to a well containing pCa4 activating solution [in mM: 60 potassium methane sulfonate, pH 7, 5 EGTA, 5.3 CaCl₂, 6.98 MgCl₂ (1 free Mg²⁺), 5.56 Na₂ATP, 25 creatine phosphate, 25 BES, final pH to 7 with 1 N KOH]. Previous studies (7) have demonstrated that tissue from the developing aorta and gizzard is predominantly smooth muscle cells. Force generated was normalized to force per cross-sectional area (mN/cm²) by measuring the dimensions of the tissues and assuming that the tissues were rectangular. Tissue strips were of comparable sizes on given days of development and were typically 250 and 1,000 µm long, 80-150 µm wide, and 80-150 µm thick. Results are reported as the average ± SE. All experiments were done at room temperature.

V_max in developing chicken gizzard and aorta. The unloaded speed of shortening of the skinned tissues was determined from the force versus velocity relationship (19). A computer program based on Smith and Barsotti (42) was written in LabView (National Instruments, Dallas, TX) to control length for constant loads. Force and length signals were sampled and controlled using an analog-to-digital-digital-to-analog board (National Instruments PCI-MIO-16XE-10, 100 × 10³ samples/s, 16 bit). After activation of the gizzard and aorta strips, the load on the tissues was decreased to any set percentage of the isometric force. The tissues were maintained at a fraction of the initial isometric load (1-90%) by a program monitoring tension (200-300 samples/s) and rapidly adjusting (shortening) length to provide constant load. A force clamp from an initial tone to any desired new load was achieved within 10 (300 samples/s) to 30 ms (200 samples/s). The shortening profiles of the tissues were saved, and the speed of shortening at various loads was calculated from the tangent to the length versus time trace at 150 ms after the step change in load. The unloaded speed of shortening was calculated by fitting the force versus velocity relationship to the Hill equation (19) or taken as the speed of shortening at 1% of initial isometric load (42). The speeds of shortening were normalized to muscle lengths per second (ML/s). For each tissue, at least three separate measurements were taken at each load tested; n represents the number of tissues tested. However, the V_max for smooth muscles before ED 12 could not be reliably determined because of the low forces of Ca²⁺-activated contractions. Results are presented as the average ± SE.

Analysis of MLC₂₀ phosphorylation in developing chicken gizzard and aorta. To determine the extent of phosphorylation of MLC₂₀, a glycerol-urea-PAGE procedure (17) was adapted to resolve the phosphorylated and unphosphorylated forms of MLC₂₀. Strips of chicken gizzard or aorta were skinned as before, washed, and resuspended in the relaxing buffer. To measure the rate of phosphorylation of MLC₂₀, tissue strips were incubated for 0, 30, 60, 120, and 300 s in pCa4 activating solution. Activation of the strips was terminated by adding excess 15% (vol/vol) trichloroacetic acid in acetone. The tissues were frozen in liquid nitrogen for 30 min. After denaturation, the tissue strips were washed twice with acetone, dried using a Speed-Vac evaporator and homogenized in PAGE sample buffer containing 6 M urea, 20 mM glycine, 22 mM Tris · HCl, pH 8.6, and 1 mM EDTA. Glycerol (5% vol/vol) and bromophenol blue (0.1% wt/vol) were added to the samples before loading. The samples were resolved by a 19:1 (acrylamide-bisacrylamide) 10% polyacrylamide gel containing 40% (vol/vol) glycerol and 4 M urea and polymerized in a gel buffer containing 20 mM glycine and 22 mM Tris · HCl, pH 8.6. The gel was prerun in the electrophoresis buffer (20 mM glycine, 22 mM Tris · HCl, pH 8.6, 1 mM thioglycolic acid) for 1 h at 200 V before loading of the samples. The samples were resolved overnight at 200 V or roughly 2 h after the bromophenol blue dye front had exited the gel. Resolved proteins were transferred to nitrocellulose membrane, and the phosphorylated and unphosphorylated MLC₂₀ were identified by immunoblotting as described earlier. The percentage of MLC₂₀ phosphorylated was calculated by determining the ratio of the phosphorylated MLC₂₀ band versus the total MLC₂₀ (phosphorylated + unphosphorylated forms) identified by the mAb.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Rapid rate of force activation in the developing gizzard. Representative contraction profiles for embryonic aorta and gizzard at various stages demonstrate the development of the contractile phenotypes of these tissues (Fig. 1). Both smooth muscle tissues show a slow rate of force activation early during embryonic development (ED 8-12). The faster contractile properties of the gizzard strips become apparent by ED 16, with clearly increased rates of force activation and relaxation compared with the aorta (Table 1 and see Table 6). The rates of activation between ED 12 gizzard and aorta were not different (P > 0.05). Although the rates of force activation were not significantly different between ED 16 and 20 gizzard, they were always significantly faster than either ED 16 or 20 aorta (P < 0.05). The rates of force activation increased with development in the gizzard and were significantly faster at ED 20 versus ED 12 (P < 0.05). Thiophosphorylation of MLC₂₀ significantly increased the rates of force activation in both ED 20 aorta (0.018 ± 0.003 s¹) and gizzard (0.060 ± 0.009 s¹) versus Ca²⁺ activation (P < 0.05). However, the rate of force activation for ED 20 gizzard remained threefold faster than for ED 20 aorta with or without thiophosphorylation.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Contraction profiles of developing embryonic chick aorta and gizzard strips. Representative tracings of skinned embryonic days (ED) 8, 12, 16, and 20 chicken aorta and gizzard strips after Ca2+ activation. Early in development, both the aorta and gizzard show tonic contractile properties marked by a slow rate of force activation. The rapid rate of force activation in the gizzard emerges by ED 16, implying that the determinants of tonic and phasic contractile properties are established. A similar trend is seen for the rates of force relaxation, as ED 16 gizzard shows faster rates of relaxation compared with the aorta (Table 5).

Orientation of cells in the developing aorta and gizzard. To determine whether changes in the orientation of the cells were occurring with development, ED 12, 16, and 20 aorta and gizzard sections were stained with hematoxylin and eosin (Fig. 2). With development in both tissues, an increase in cell density was apparent. In the aorta, there was no change in the orientation of the cells with development. In the gizzard, however, the orientation of cells changed by ~20° between ED 12 and 16, with cells changing from parallel to the axis of the strip at ED 12 to 20° off the axis at ED 16. No subsequent change in cell orientation was observed from ED 16 to 20 gizzard.

View larger version (141K): [in this window] [in a new window]   Fig. 2.   The orientation of cells in embryonic aorta and gizzard tissues. Tissue samples from ED 12, 16, and 20 aorta and gizzard were sectioned and stained with hematoxylin and eosin to determine the orientation of the cells. During development of the aorta, no change in the orientation of the cells was observed. During development of the gizzard, a change of ~20° in the orientation of the cells is evident from ED 12 to 16. No change in orientation was seen between ED 16 and 20 gizzard. Bars denote 10 µm.

Protein isoform expression in the developing aorta and gizzard. The expression of multiple regulatory protein isoforms is a common theme in developing muscle. Representative smooth muscle-specific markers were chosen, and their isoform expression throughout development of the embryonic chicken gizzard and aorta was monitored (Fig. 3). Calponin, an actin filament-associated marker of smooth muscle, was present throughout development of the gizzard and aorta as a single isoform. Using MAb MY-21 (Sigma), we identified a single isoform of MLC₂₀ expressed throughout development of the aorta and gizzard. Comparison of the expression of the M130/M133 isoforms of the myosin binding subunit of MLCP (41) indicated that only M133 was expressed throughout development of the aorta. On the other hand, a developmental isoform switch of M130 for M133 was observed in the developing gizzard, with M130 expression beginning at ED 20. This is consistent with upregulation of M130 expression to become the dominant myosin binding subunit isoform expressed in adult chicken gizzard (41). The smooth muscle MLCK isoform was the abundant MLCK isoform expressed in both the gizzard and aorta during embryonic development. Because the Western blots only monitored isoform expression, we could not make any conclusions with respect to changes in protein content with development.

View larger version (24K): [in this window] [in a new window]   Fig. 3.   The expression of regulatory protein isoforms during development. With the use of specific monoclonal antibodies, the expression pattern of the myosin binding subunit of myosin light chain phosphatase (MLCP), myosin light chain kinase (MLCK), h1-calponin, and 20-kDa myosin light chain (MLC20) were determined in the developing gizzard and aorta. All proteins showed expression of a single protein isoform throughout development except for MLCP. The gizzard showed expression of the M130 myosin binding subunit beginning between ED 18 and 20, with M130 being the dominant isoform expressed at ED 20.

Force of contraction of the developing gizzard and aorta. We measured the force per cross-sectional area in developing aorta and gizzard to determine whether a correlation could be made with myosin isoform expression. Although the force per cross-sectional area increased with development for both embryonic aorta and gizzard (Table 2), ED 20 gizzard was significantly higher (P < 0.05) than ED 20 aorta, with or without thiophosphorylation of the myosin regulatory light chain. The increases in force per cross section may be achieved by differences in actin/myosin content or tissue content of smooth muscle.

V_max in embryonic chicken gizzard and aorta. To monitor changes in V_max with development of the smooth muscle contractile phenotypes, the V_max for embryonic aorta and gizzard strips was determined at three stages of development. The V_max for the developing aorta was higher than for the developing gizzard at all stages (Table 3). This was surprising given that adult phasic tissues have faster speeds of shortening than tonic tissues (15). Thiophosphorylation of MLC₂₀ with ATP--S resulted in no significant change of V_max for ED 20 aorta but a doubling in the gizzard (0.038 ± 0.004 vs. 0.084 ± 0.018 ML/s). Additional experiments with Ca²⁺-activated skinned gizzard strips from 1-wk-old chicks indicated that V_max increased after hatching (0.115 ± 0.013 ML/s, n = 3). Therefore, V_max continues to increase significantly after birth, given that adult chicken gizzard strips have a V_max of ~0.3 ML/s (1). Force-velocity curves of thiophosphorylated fibers showed that the shortening velocity as a percentage of V_max (Fig. 4B) achieved by the gizzard was higher at most loads compared with the aorta. The extent of MLC₂₀ thiophosphorylation in the preparations was comparable (Table 4).

View larger version (19K): [in this window] [in a new window]   Fig. 4.   The force versus velocity relationship in thiophosphorylated ED 20 aorta and gizzard. After thiophosphorylation of skinned ED 20 aorta and gizzard strips, the force versus velocity relationship was determined by decreasing the load on the strips to various levels and measuring the speed of shortening. For a given tissue, either the absolute speeds of shortening [muscle length (ML)/s] at various loads were reported (A) or speeds of shortening were normalized by dividing by the maximum speed of shortening (Vmax) of the tissue and expressing the speed as a fraction of the Vmax (B). A: the force versus velocity relationship from single ED 20 aorta and gizzard fibers. Data points are the average ± SE for 3 measurements at each load. B: pooled data for ED 20 aorta (n = 8) and ED 20 gizzard (n = 11) strips. The pooled data were fit to the Hill equation. Note that the gizzard displays a flatter relationship, indicating that a higher absolute Vmax is achieved at loads 40% of maximum (A) and that a greater percentage of maximum speed is achieved for isotonic contractions above zero load (B).

Time course of MLC₂₀ phosphorylation and dephosphorylation. To assess whether phosphorylation-dephosphorylation of MLC₂₀ played a significant role in the rates of force activation and relaxation, we determined the time to reach half-maximal phosphorylation and dephosphorylation in skinned ED 16 and 20 aorta and gizzard. We chose ED 16 and 20 because phasic and tonic contractile phenotypes appeared to be well established at these days of development. The levels of MLC₂₀ phosphorylation in the gizzard and aorta after 0 or 300 s of activation were not significantly different (Fig. 5). Both ED 20 aorta and gizzard showed similar levels of maximum MLC₂₀ phosphorylation (aorta, 47.3 ± 4.0%, n = 7; gizzard, 44.2 ± 3.3%, n = 7), which were not different from ED 16 aorta (42.6 ± 5.0%, n = 8) or gizzard (42.7 ± 4.6%, n = 7), indicating that the maximum level of activation does not change between ED 16 and 20. Likewise, for dephosphorylation experiments, we found no difference between the level of MLC₂₀ phosphorylation in the gizzard and aorta after 0 or 300 s of dephosphorylation (Fig. 6).

View larger version (15K): [in this window] [in a new window]   Fig. 5.   The phosphorylation of MLC20 in ED 20 aorta and gizzard. With the use of skinned ED 20 aorta and gizzard strips, the time course of MLC20 phosphorylation was followed by separation of the phosphorylated and unphosphorylated MLC20 by glycerol-urea-PAGE. There was no difference in the time course of phosphorylation or maximal level of MLC20 phosphorylation in the tonic and phasic tissues.

View larger version (13K): [in this window] [in a new window]   Fig. 6.   Dephosphorylation of MLC20 in ED 20 gizzard is faster than ED 20 aorta. After activation of the skinned tissues for 5 min, the time course of dephosphorylation of MLC20 was followed by incubating the tissue strips in a Ca2+-free rigor solution. Dephosphorylation of MLC20 was significantly faster in ED 20 gizzard compared with ED 20 aorta (P < 0.05). The solid line shows a fit to the data by a single exponential with a rate of 0.051 ± 0.016 s1 in the gizzard and 0.021 ± 0.003 s1 in the aorta.

1

1

1

1

Rate of force relaxation in embryonic chicken smooth muscles. We measured the rates of force relaxation for skinned embryonic gizzard and aorta strips at ED 12, 16, and 20 (Table 6) to determine whether relaxation could be correlated with differences in the rate of dephosphorylation of MLC₂₀. For ED 12, 16, and 20 aorta, rates of relaxation were 0.001 ± 0.001 (n = 3), 0.004 ± 0.001 (n = 5), and 0.005 ± 0.002 s¹ (n = 7), respectively. These were not different from the rate of relaxation of ED 12 gizzard (0.003 ± 0.002 s¹, n = 3). However, the rates of relaxation for ED 16 gizzard (0.019 ± 0.002 s¹, n = 8) and ED 20 gizzard (0.030 ± 0.004 s¹, n = 8) were significantly faster compared with the aorta (P < 0.05). Furthermore, the rate of relaxation in ED 20 gizzard was significantly faster (P < 0.05) than in ED 16 gizzard.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Regulatory protein isoform heterogeneity during development. Through development, regulatory proteins may be expressed as multiple isoforms. During chicken development, MHC and MLC₁₇ expression both show default tonic phenotypes at early embryonic stages, with expression of MLC_17b and noninsert containing MHC (7). However, in the gizzard, polymerase chain reaction and Western blot analyses have shown that expression of the 7-amino acid insert containing MHCs dominates expression in SM1 and SM2 MHC isoforms from ED 10 (>70%) to adult (~100%; 7, 28), whereas the aorta did not express the 7-amino acid insert in SM1 and SM2 myosins. Furthermore, MLC_17a expression is upregulated with development to hatching in the gizzard (>50% at ED 12, >70% at ED 16, and ~100% for adult; 7, 28), whereas MLC_17b is predominant throughout aorta development (>60% from ED 12 to adult; 7, 28). Therefore, in the gizzard, MHC and MLC₁₇ isoform maturation correlates with development of the phasic phenotype.

Kinetics of MLC₂₀ phosphorylation in embryonic gizzard and aorta. Although MLC₂₀ phosphorylation controls smooth muscle force production, the increased rate of force activation in the gizzard is not well explained by MLCK activity. Phosphorylation of MLC₂₀ in both aorta and gizzard preceded maximal force activation (see Figs. 1 and 5). In addition, both tissues expressed identical MLCK isoforms (Fig. 3) and reached the same maximum level of MLC₂₀ phosphorylation (Fig. 5). However, between ED 20 gizzard and aorta, we detected a significant difference in the time course of MLC₂₀ dephosphorylation (Fig. 6). If MLCP activity is constitutive in the presence and absence of Ca²⁺, this implies that steady-state MLCK activity was higher in ED 20 gizzard than in the aorta. This difference was not resolved on the time scale of our in vitro phosphorylation experiments. However, the difference between MLCK activity for the aorta and gizzard can be determined assuming that the steady-state level of MLC₂₀ phosphorylation is defined by MLCK and MLCP activities as described by MLCK/(MLCK + MLCP). Because steady-state MLC₂₀ phosphorylation levels were not significantly different between ED 20 aorta and gizzard (~45%) and assuming that the rates of dephosphorylation are accurate, the MLCK activity in the aorta would calculate to 41% of that in the gizzard. This implies that the rates of phosphorylation were underestimated. Although thiophosphorylation of MLC₂₀ in ED 20 aorta and gizzard increased the rates of force activation compared with phosphorylation of MLC₂₀, an approximately threefold difference remained (Table 1) in the rates of force activation in these tissues. Therefore, MLC₂₀ phosphorylation was not a limiting factor in the slow rate of force activation in the aorta, indicating that with phasic muscle the rapid rate of force activation is not explained by increased MLCK activity alone. Thus although the steady-state MLCK activity may be higher in the gizzard, its contribution to the differences in the rates of Ca²⁺-activated force appears to be small. Furthermore, differences in the rates of force activation between aorta and gizzard are unlikely to be due to diffusional effects. Activation of thiophosphorylated gizzard fibers (wherein ATP diffusion is limiting) showed that diffusion can sustain rates <0.060 s¹. Therefore, the differences in the rates of force activation in Ca²⁺-activated gizzard and aorta fibers (wherein rates were <0.02 s¹) would not be due to diffusion. In fact, thiophosphorylated phasic and tonic tissues activated by photolysis of caged ATP still show significant differences in the rates of force activation (21). With flash photolysis, diffusion is not a factor and differences in the rates of force activation point to unique kinetics of the actomyosin interaction in these tissues. In addition, the differences in the rates of force activation for Ca²⁺-activated smooth muscle strips was not due to differences in calmodulin content (or loss) after skinning, because thiophosphorylation achieved near complete levels of MLC₂₀ phosphorylation in both ED 20 aorta and gizzard (Table 4) while preserving the differences in the rates of force activation (Table 1). We also considered the possibility that changes in maximum force could be due to variations in the orientation of sarcomere equivalents. However, because force is related to the cosine of the angle between the axis of tissue preparation and the orientation of the cells, large changes in angle (>60°) would be required to affect maximum force by twofold. The data in Fig. 2 demonstrate that changes in cell orientation were not an issue with the developing aorta. In the gizzard, the ~20° change in orientation between ED 12 and 16 would result in the underestimation of the measured force at ED 16 and 20 by ~6%. This suggests that the increase in force with development is due to more smooth muscle cells per cross section; this is supported by data that the force per stiffness is not significantly different with development (data not shown). However, the orientation of cells is a factor only for the force measured, not the rate of force activation. The data indicate that the primary determinant for the rapid rate of force activation in the gizzard is at the level of the actin and myosin filaments. Because the MLC₂₀ phosphorylation measurements (Fig. 5) were done on skinned preparations activated directly with Ca²⁺, we excluded possible modulation of MLCK activity in the aorta and gizzard through second messenger pathways (44), but not through Ca²⁺-activated kinases. However, thiophosphorylated fibers maintained differences in the rates of force activation, downplaying the importance of Ca²⁺-activated kinases (other than MLCK), unless a kinase was specifically coupled to the contractile apparatus in only gizzard or only aortic smooth muscle. Therefore, differences in the rates of force activation between gizzard and aortic smooth muscle remain in the absence of tissue-specific agonist-dependent pathways that differentially modulate MLCK/MLCP activity through phosphorylation (9).

Rates of MLCP dephosphorylation in developing gizzard and aorta. There was a significant difference in the rates of MLC₂₀ dephosphorylation between ED 20 gizzard and ED 16 gizzard or ED 16 and 20 aorta (Table 5; Fig. 6). The increased rate of MLC₂₀ dephosphorylation in ED 20 gizzard correlated with expression of a unique isoform of the myosin binding subunit of MLCP (M130; 16, 41) that was not expressed in the aorta (Fig. 3). There was no difference in the rates of dephosphorylation in ED 16 aorta and gizzard versus ED 20 aorta (Table 4), wherein all smooth muscle tissues expressed M133. Therefore, our results suggest that the MLCP holoenzyme containing the M130 myosin binding subunit correlates with the increased rate of MLC₂₀ dephosphorylation in ED 20 gizzard, either through increased activity of this MLCP holoenzyme or an increase in the MLCP-to-myosin ratio. Coupled with data characterizing the rates of force relaxation, two conclusions may be drawn. First, we demonstrated that the rate of relaxation in the gizzard is faster than in the aorta at both ED 16 and 20. At ED 16, both tissues express the MLCP holoenzyme with the M133 subunit and show similar time courses for MLC₂₀ dephosphorylation. Therefore, the differences in the rates of force relaxation between ED 16 aorta and gizzard (Table 6) are best attributed to changes in the kinetics of the myosin isoforms in the cross-bridge cycle, likely due to inclusion (gizzard) or exclusion (aorta) of the 7-amino acid insert in the MHC or different MLC₁₇ isoform expression in the two tissues. With laser trap experiments, it has been demonstrated that heavy meromyosin composed of the noninsert (aortic) form of MHC has an attachment time on actin roughly twice as long as the insert (gizzard) form (30). This would agree with our results that relaxation in ED 16 gizzard is faster than in ED 16 aorta, although the tissues show similar rates of MLC₂₀ dephosphorylation. Nonetheless, the increase in the rate of relaxation of ED 20 versus 16 gizzard indicates that expression of MLCP holoenzymes containing the M130 myosin binding subunit correlates with increases in the rate of MLC₂₀ dephosphorylation and thus the rate of force relaxation, whether by increased MLCP isoform activity or increased MLCP-to-myosin ratio.

Unloaded speed of shortening in developing chicken gizzard and aorta. Previous reports have demonstrated that increased MLC_17a content may account for faster speeds of shortening in smooth muscle tissues (31). Increasing the proportion of MLC_17a expression by exchange of MLC₁₇ isoforms in trifluoperazine-treated smooth muscle was also shown to increase the shortening velocity (35), although contraction in these tissues was independent of Ca²⁺ or MLC₂₀ phosphorylation. On the other hand, in in vitro motility assays, exchange of MLC₁₇ isoforms had no effect on the speed of actin filament movement by myosins (28, 40). Although the embryonic gizzard expressed a higher proportion of MLC_17a than the aorta (7), in our study, the gizzard had slower speeds of shortening under all conditions tested (Table 3). Therefore, MLC_17a content was not a factor in defining a faster V_max in these tissues. It remains a possibility that changes in content or isoform of calponin and caldesmon (37) may contribute to the mechanical parameters measured in these experiments.

Role of MLC₁₇ isoforms in determining tonic and phasic contractile properties. It is well known that the 20-kDa regulatory myosin light chains play a defining role in controlling actomyosin interaction in smooth muscle. However, the function of the 17-kDa essential myosin light chain has not been as well characterized. The essential myosin light chains line the neck region of the MHC dimers, possibly acting as a scaffold for this extended helix (39). In this respect, the interaction of different MLC₁₇ isoforms with the neck region of MHC dimers may result in changes in the stiffness of this lever arm. The rapid rate of force activation in the gizzard due to MLC_17a expression may be explained if this isoform provided added stiffness to the myosin lever arm.