Actin polymerization as part of the normal smooth muscle response to various stimuli has been reported. The actin dynamics are believed to be necessary for cytoskeletal remodeling in smooth muscle in its adaptation to external stress and strain and for maintenance of optimal contractility. We have shown in our previous studies in airway smooth muscle that myosins polymerized in response to contractile activation as well as to adaptation at longer cell lengths. We postulated that the same response could be elicited from actins under the same conditions. In the present study, actin filament formation was quantified electron microscopically in cell cross sections. Nanometer resolution allowed us to examine regional distribution of filaments in a cell cross section. Airway smooth muscle bundles were fixed in relaxed and activated states at two lengths; muscle preparations were also fixed after a period of oscillatory strain, a condition known to cause depolymerization of myosin filaments. The results indicate that contractile activation and increased cell length nonsynergistically enhanced actin polymerization; the extent of actin polymerization was substantially less than that of myosin polymerization. Oscillatory strain increased thin filament formation. Although thin filament density was found higher in cytoplasmic areas near dense bodies, contractile activation did not preferentially enhance actin polymerization in these areas. It is concluded that actin thin filaments are dynamic structures whose length and number are regulated by the cell in response to changes in extracellular environment and that polymerization and depolymerization of thin filaments occur uniformly across the whole cell cross section.
- length adaptation
- myosin filament
- actin cytoskeleton
although it has been recognized for some time that actin polymerization and depolymerization associated with the contraction-relaxation cycle in smooth muscle is important for normal function of the tissue (1, 22), details of the actin dynamics and its relation to the basic mechanism of contraction have just begun to emerge. It appears that one of the pathways mediating the activation-associated actin reorganization is through activation of the small GTPase Rho (12, 20, 21, 35) and its subsequent regulation of profilin (11, 30). Another pathway that mediates the actin reorganization is thought to be via activation of p38 mitogen-activated protein kinase and its subsequent phosphorylation of the 27-kDa heat shock protein (9, 12, 19, 36). Prevention of actin polymerization has been shown to severely reduce force production in many types of smooth muscle (7, 28, 32). In airway smooth muscle, it has been shown that there is a substantial pool of globular actin molecules in the relaxed state and that polymerization of these molecules occurs during contractile activation (23). In cultured airway smooth muscle cells, it has been shown that oscillatory strain increased actin filament formation (29); this, however, has not been demonstrated in intact cells. The actin filament formation associated with contractile activation is accompanied by an increase in the stiffness of the cytoskeleton, even after the contribution to the stiffness by actomyosin crossbridges is excluded (2). These studies suggest that in the relaxed state actin filaments are likely partially dissolved and may not be firmly attached to the dense bodies and plaques or the focal adhesion plaques, and the cytoskeleton may be malleable as a result. This interpretation is consistent with earlier findings from functional studies that showed plastic behavior in these cells (10, 26). The ability of the muscle to plastically reorganize its contractile apparatus and therefore optimize its contractile function has been attributed to malleability in both the cytoskeleton (6) and the contractile filament arrays embedded in the cytoskeleton (17). Although there is strong functional evidence supporting plastic deformation of airway smooth muscle cells subjected to large strains (both static and oscillatory), ultrastructural details associated with the remodeling of cytoskeleton and contractile apparatus are still vague, especially the details regarding actin filament formation and distribution within the muscle cell. Electron microscopy is one of the few means that allow us to visualize individual actin filaments in cell cross sections. In the present study, densities of actin filaments were quantified in electron micrographs of transverse sections of intact airway smooth muscle fixed under a variety of experimental conditions. This has allowed us to gain insights into the role of actin polymerization in the cell's adaptation to external forces.
MATERIALS AND METHODS
Porcine trachealis muscle was used for the experiments. The tracheas were obtained from a local abattoir. After removal from the animals, the tracheas were placed in physiological saline solution (PSS) at 4°C. A rectangular piece of smooth muscle tissue close to its in situ length, free of connective tissues, was cut from the trachea. The piece of tissue was then cut into multiple strips along the longitudinal axes of the cell bundles; all the strips had the same initial length as a result. The muscle preparations were ∼11 × 1 × 0.4 mm in dimension. The strips of muscle were affixed to aluminum foil clips at both ends and mounted in a muscle bath. One end of the strip was joined to a stationary hook and the other end to a length-force transducer with a signal-to-noise ratio >50 and a compliance of ∼1 μm/mN (QJin Design, Winnipeg, MB, Canada). The apparatus and tissue preparation have been described previously (33, 34). The muscle bath contained PSS with pH 7.4 at 37°C and bubbled with a gas mixture (5% CO2-95% O2). The PSS has a composition (in mM) of 118 NaCl, 4.5 KCl, 1.2 NaH2PO4, 22.5 NaHCO3, 2 MgSO4, and 2 CaCl2 and 2 g/l dextrose.
Before a trachealis preparation was chemically fixed for electronmicroscopic examination, it was equilibrated at a preset length for ∼1 h. During the equilibration period, the muscle was stimulated (with electric field stimulation, 60 Hz) periodically to produce 12-s tetani at 5-min intervals. The preparation was considered equilibrated when it developed a stable maximal isometric tetanic force. There were five experimental conditions under which the preparations were fixed (after equilibration): 1) relaxed and unstrained (at the in situ length), 2) activated at the in situ length, 3) relaxed and strained (at 1.5 times the in situ length), and 4) activated and strained (at 1.5 times the in situ length), and 5) relaxed at the in situ length and postoscillation (the length-oscillation frequency was 0.5 Hz; the peak-to-peak amplitude of strain was 60% of the in situ length; the oscillation was centered around the in situ length, which meant that the cells were subjected to a 30% stretch beyond their in situ length; and the duration of oscillation was 5 min). For the preparations fixed under an activated state, the final stimulation was produced by addition to the muscle bath 0.1 mM of acetylcholine; the muscle was fixed at the plateau of isometric contraction 120 s after the addition of acetylcholine. The average stress was found to be 101.6 ± 11.6 (SE) kPa. Acetylcholine (instead of electric field stimulation) was used in the final contraction to ensure that activation of the muscle was maintained during fixation.
Tracheas from three animals were used for the electron microscopy analysis. Five strips of muscle per trachea were dissected and equilibrated at the preset lengths according to the five experiment conditions described above. Muscle preparations were fixed for electron microscopy by using a conventional protocol described previously (18, 27). Briefly, muscle preparations were fixed with the primary fixing solution (see below for details) for 15 min while they were still attached to the experimental apparatus where isometric force was continuously monitored. The tissue was then removed from the apparatus and cut into small cubes and immersed in the primary fixing solution for an additional 2 h at 4°C. The primary fixing solution contained 2% glutaraldehyde, 2% paraformaldehyde, and 2% tannic acid in 0.1 M sodium cacodylate buffer. In the process of secondary fixation, the cubes were put in 1% OsO4 in 0.1 M sodium cacodylate buffer for 2 h. The tissue was then stained with 1% uranyl acetate, dehydrated with increasing concentrations of ethanol, and embedded in resin (TAAB 812 mix). The blocks were sectioned with a diamond knife to obtain sections of ∼90 nm of thickness. The sections (on copper grids) were further stained with 1% uranyl acetate and Reynold's lead citrate. Images of longitudinal and cross-sectional areas of smooth muscle cells were obtained by using a Phillips 300 electron microscope.
Sampling and analysis were carried out “blind.” The codes indicating experimental conditions were revealed only after the analysis of each group was finished. A total of 25 electron micrographs of muscle cell cross sections per trachea were analyzed (5 per preparation). A specialized image-analysis software (Image Pro-Plus 3.0) was used to help in the manual counting of the thin filaments by marking and keeping track of the number of filaments counted (tag-point counting). The computer program randomly placed 10 circles with an area of 0.1734 μm2 on each micrograph. Circles that fell on the areas occupied by nucleus, organelles, plasma membrane, or extracellular space were excluded from the study; only the circles that fell on cytoplasmic areas with mostly thick and thin filaments, and occasionally dense bodies, were counted. The area occupied by dense bodies was not included in the calculation of filament density. Of the 10 circles generated by the computer, ∼3–4 of them met the criteria. The thin filaments were counted by the tag-point quantification method in each of the nonexcluded circles placed in the cytoplasmic areas on the cells. According to standard morphological methods, the filaments that fell on the edge of the circles were counted as half. The total numbers of thin filaments counted were divided by the area of the circle (minus area of dense bodies, if there were any) to obtain the thin filament density. For average thick filament density, because the number of thick filaments per cross section is relatively small, the total number of thick filaments within a cell cross section was obtained and divided by the cytoplasmic area of that cell cross section.
The analysis and comparison between the groups were performed by one-way or two-way ANOVA. The n value was the number of animals used. Data from each animal (5–6 micrographs per animal) were averaged first before the means from different animals were averaged. Statistically significant difference corresponds to a P value of <0.05.
Actin thin filament density was quantified by counting the total number of thin filaments within the selected circles, as described in materials and methods, and dividing the number by the circle area. Figure 1A shows an example: 10 randomly selected areas (circles) were placed by the computer onto an electron micrograph of a cell cross section. Figure 1B is an example of an electron micrograph cross section of a cell fixed in the activated state. Only the circles that contained mostly myosin and actin filaments and free of large organelles and/or extracellular space were selected for counting; one example is shown in Fig. 1C where an area in Fig. 1A is enlarged. A myosin thick filament has an irregular cross-sectional profile, with an average “diameter” of 12–20 nm. An actin thin filament has an approximately circular profile with a diameter of ∼6–7 nm. On average, 3–4 circles (out of the 10 randomly generated by the computer) per micrograph met the selection criteria and were analyzed.
Figure 2 shows longitudinal sections of trachealis cells fixed in the relaxed and contracted states. In the activated state, the filaments were aligned with the longitudinal axes of the cells that in turn were aligned with the axis of force transmission at the time the tissue was fixed. In the relaxed state, although most filaments lay parallel to the common longitudinal axis of the cell bundle, some filaments lay with small angles to the longitudinal axis. Excluding the regions occupied by organelles, there was no evidence for regional aggregation and large degree of uneven distribution of filaments in the activated cells compared with the relaxed cells.
Cross-sectional densities of the thin filaments (such as those shown in Fig. 1) under various experiment conditions are shown in Fig. 3. Contractile activation resulted in a significant increase in the thin filament density; in cells activated at in situ length, the density increased by 19.5 ± 3.9% (SE), whereas in cells activated at 1.5 times the in situ length, the density increase was 30.1 ± 10.1%. In cells fixed in the relaxed state at the in situ length after a period of length oscillation, there was a small but significant increase in the thin filament density (9.1 ± 1.8%); this happened despite a decrease [18.4 ± 2.6% (SE)] in the thick filament density. Cross-sectional densities of the thick filaments under various conditions are shown in Fig. 4. The average density was assessed by counting all thick filaments in a cell cross section and dividing the number by the cytoplasmic area of that cell cross section. The density in the central region of a cell cross section was assessed by placing a circle (such as the ones in Fig. 1) in the central region of a cell cross section, counting the number of thick filaments within the circle, and dividing the number by the circle area. The peripheral density was assessed by randomly placing a circle in the peripheral region of a cell cross section, counting the number of thick filaments in the circle, and dividing the number by the circle area. For each condition, five micrographs from each animal were used; each averaged density value was therefore obtained from analysis of 15 micrographs. Both contractile activation and adaptation of the muscle at the longer length caused a significant increase in the thick filament density. The distribution of the thick filaments in a cell cross section was relatively even in the relaxed state; in the activated state the distribution was less even with small clusters formed in various regions in a cell, but there was no systematic aggregation of thick filaments in the central or peripheral regions (Fig. 4).
To quantitatively describe regional distribution of thin filaments in cell cross sections, we examined the density of thin filaments near dense bodies, near the peripheral regions, and near the central region of the cell cross sections. These density values were then compared with each other and to the average density measured from randomly selected areas. Figure 5 shows an example of an area that contained a dense body (that occupied ∼15% of the area within the circle) selected for examination for thin filament density around the dense bodies. The central and peripheral thin filament densities were assessed in the same way as described for thick filament density assessments (described above for Fig. 4). The filament density was examined under four conditions: relaxed vs. contracted at 1.0 times and 1.5 times the in situ length. For the comparisons of densities near the dense bodies, six micrographs from each group were used, and in each micrograph thin filaments in three circles were counted. The percentages of areas occupied by the dense bodies within the circles for the relaxed and contracted groups were 16.1 ± 0.1 and 14.8 ± 0.1%, respectively. They were not statistically different (P = 0.3). For the comparisons of central and peripheral thin filament densities, six micrographs from each group were used, and in each micrograph thin filaments within one circle placed in the central region of the cell cross section and one circle placed in the peripheral region were counted. Figure 6 summarizes the comparisons. Thin filaments were more concentrated near the dense bodies; they increased by 11.6 ± 2.8 (SE) and 11.5 ± 3.9% in the relaxed and activated cells, respectively. On activation, the density near a dense body increased by 19.2 ± 1.9% (SE), nearly identical to the increase in density in the randomly selected area during activation (19.5 ± 3.9%). Two-way ANOVA revealed that the elevated thin filament densities around dense bodies and after contractile activation were significant (P < 0.05), and there was no statistical interaction between dense body area-specific and contractile state-associated thin filament densities. The distribution of thin filaments in a cell cross section appeared to be even, and there was no systematic aggregation of the filaments in the central or peripheral areas after the cells had been activated or stretched (Fig. 6).
The dynamics of cytoskeletal structure of smooth muscle has received considerable attention recently and has been recognized as one of the key mechanisms that enable smooth muscle cells to adapt to large changes in length or to migrate (2, 4–5, 7, 20, 23–25, 28, 31). Findings of this study have revealed some previously unknown aspects of actin polymerization in airway smooth muscle and their possible roles in the muscle's adaptation to external strain and generation of active force.
Before we discuss the results in detail, it is important to point out limitations of our measurements. There was a small change in the angle of alignment of contractile filaments with the cell's longitudinal axis during contractile activation (as the filaments became more parallel to the axis of force transmission, Fig. 2); this could affect the filament density count in cell cross sections. However, the fact that the angles were in general relatively small is not likely to change the conclusion of this study (although it may quantitatively alter the density values). The filament density measured in a cell cross section at different cell lengths is directly proportional to the mass of the filaments in the cell, if the cell volume is constant. Because the filament mass is determined by the average length and number of the filaments, the filament density (D) is a product of filament length (L) and number (N); i.e., D = L × N. Unfortunately, we cannot differentiate a density change due to a change in filament length from that due to a change in the number of filaments. An increase in filament density can only be interpreted as a result of three possible changes: 1) an increase in the number of filaments, 2) an increase in filament length (of existing filaments), or 3) a mix of the above two possibilities.
Activation-dependent thin filament density and its regional variation.
Actin polymerization during contractile activation has been examined by many investigators using a variety of methods; pharmacological inhibition of the transition from globular (G)- to filamentous (F)-actin has been extensively used in the studies of both intact and cultured smooth muscle cells (1, 7, 20, 22, 25, 28); immunostaining of both G- and F-actins has also been used (4, 8, 14–15); biochemical quantification is another method used to assess the composition of G- and F-actins (23, 30). The above-mentioned studies examined the “global” changes in the contents of G- and F-actins in muscle cells; the methods used in these studies lacked the resolution to pinpoint where within the cytoplasm the polymerization or depolymerization of actin occurs in response to specific stimuli. The results presented in the present study revealed for the first time the difference in actin filament densities in the subcellular domains. The thin filament density near the cytoplasmic dense bodies was found to be slightly higher than the density in the cytoplasmic areas without dense bodies (Fig. 6). This small but significant increase in density suggests that there was a tendency for thin filaments to bundle together near dense bodies; this is perhaps related to the appearance of “myofibrils” within individual smooth muscle cells found in longitudinal sections where thin filaments attach to dense bodies in a horsetail fashion (3). The densities in the central and peripheral areas of cell cross sections were, however, not altered by contractile activation or stretching of the cells (Fig. 6).
Although a significant increase in thin filament density was found to accompany contractile activation (Fig. 3), there was no out-of-proportion regional increase (Fig. 6). The present finding is consistent with the interpretation that the actin polymerization is more or less even globally, not localized to any particular subcellular domains in a cell cross section. Polymerization of myosin thick filaments due to contractile activation, on the other hand, is rather inhomogeneous (Fig. 1B), as also reported by our laboratory previously (13, 16, 17). The clustered aggregation of thick filaments in activated cells, however, does not show systematic concentration of thick filaments in the central region of the cell or vice versa (Fig. 4).
It is still not clear why polymerization of actin is needed in smooth muscle during activation. There appears to be abundant thin filaments packed within the cytosol, even in the relaxed state. It has been postulated that lengthening of thin filaments might be needed for focal adhesion (9). The nonlocalized thin filament lengthening and formation observed in the present study suggest that the polymerization may have a function of facilitating thin-thick filament interaction and therefore force generation. However, it would appear that the relatively sparse thick filaments would be the limiting factor in this regard and that the additional thin filament formation that accompanies contractile activation would not have a major effect. More studies are needed to clarify this issue.
The present finding of 20–30% increase in thin filament density is greater than that estimated in other studies (23). We have no clear explanation for the discrepancy. There are a number of factors that could contribute to the variation. Smooth muscle preparations are known to have resting tone, and the tone varies from preparation to preparation. If the resting tone is high, the measured increase in thin filament polymerization due to contractile activation may be underestimated because of the high starting level of polymerized thin filaments. Another source of variation is the alignment of thin filaments with the longitudinal axis of the cell. The filament alignment in the relaxed state may not be as perfect as that in the activated state; this could lead to overestimation of thin filament density increase (due to contractile activation) because a nonperpendicular thin filament found in a relaxed cell cross section may not be recognized as such.
Length-dependent thin filament density.
The results presented in Fig. 3 indicate that the thin filament density is higher in muscle cells adapted at a longer length. In the relaxed state, a 50% increase in length resulted in 10.6 ± 1.9% (SE) increase in thin filament density. In the contracted state, the density increase was even greater (20.1 ± 5.8%). It appears that there is synergy between contractile activation and increased strain in augmenting actin polymerization. Statistical analysis (2-way ANOVA), however, revealed that, although the significantly increased (P < 0.05) thin filament formation can be attributed separately to contractile activation and increased cell length, there is no statistically significant interaction between activation and cell length in determining the extent of actin polymerization. Our laboratory has previously shown in the same trachealis preparation that the cell volume is conserved at different cell lengths (17). If the thin filaments (which are short compared with the cell length) are evenly distributed within the cell volume and if there is no polymerization or depolymerization of the filaments, the density of thin filaments at randomly selected cross sections should be the same at different cell lengths. An increase in density, therefore, indicates an increase in the transition of G- to F-actin. Polymerization of myosin filaments was also favored at longer lengths (Fig. 4) as our laboratory has previously found (17). In our previous study, our laboratory also found that both muscle power and shortening velocity increased with adapted length while isometric force was not changed (17). A simple explanation of the finding is that there were additional contractile units (thick filaments) added in series to the contractile apparatus of a cell adapted to a longer length. Because in smooth muscle there are many more thin filaments compared with thick filaments (see Fig. 1), one could argue that there may be enough thin filaments for the additional contractile units needed at longer cell lengths (without additional formation of thin filaments). The present finding indicates that this may not be the case. It appears that, despite a large excess of thin filaments (compared with thick filaments), actin polymerization and depolymerization are still required as part of the process of cell adaptation to different lengths. As discussed above, actin polymerization may be involved in anchoring the cytoskeleton to focal adhesion sites (9), although it is not clear why such polymerization is required.
Thin filament lability vs. thick filament lability.
Lability of myosin filament in smooth muscle has been recognized ever since the ultrastructure of the muscle was examined under electron microscope [see a review by Bagby (3)]. Although it is still controversial as to the extent of lability, it is generally accepted that, compared with those in striated muscle, thick filaments in smooth muscle are structurally less stable. Lability of actin filaments is less well documented. It has been recognized only recently that polymerization and depolymerization associated with the contraction-relaxation cycle in smooth muscle may facilitate plastic adaptation of the muscle to externally applied stress and strain (10, 23). The present results indicate that there was a 63–82% increase in thick filament density due to contractile activation, depending on the adapted length of the muscle cells (Fig. 4). The increase in thin filament density was ∼20–30% under the same conditions (Fig. 3). It appears, therefore, that the formation of thick filaments is facilitated by contractile activation to a greater extent.
Another difference between thin and thick filaments in terms of their structural stability is shown in the filaments' responses to oscillatory strain. By application of periodic (0.5 Hz) 30% stretches to a relaxed trachealis preparation for 5 min, the thick filament density was found to decrease by 18.4 ± 2.6% (SE), whereas the thin filament density was found to increase by 9.1 ± 1.8 (Fig. 3). Mechanical agitation that caused thick filaments to fall apart apparently had an opposite effect on the thin filaments. In cultured airway smooth muscle cells subjected to long-term oscillatory strain, it has been reported that there was a large increase in the amount of F-actin and that the increase was associated with RhoA activation (29). It is not known whether the increase in thin filament density observed in the present study is related to RhoA activation. It is clear, though, that short-term oscillatory strain did not cause depolymerization of thin filaments, as it did with thick filaments.
The increase in thick filament density due to contractile activation found in the present study was substantially less than that found in one of our previous studies (13). The reason for the discrepancy is not entirely clear. One factor that could explain the difference is the low isometric stress [101.6 ±11.6 (SE) kPa] produced by the muscle strips used in the present study compared with those used in the previous study (161.4 ± 11.3 kPa); the different degree of activation may be one reason for the discrepancy. The oscillatory strain-induced reduction in isometric force found in this study was also significantly less than that found in our laboratory's previous study (18). Again, the reason for the discrepancy is not clear. It is possible that the muscle preparations used in the present study possessed more stray compliance, which reduced the effectiveness of mechanical disruptions on the contractile filaments imposed by the length oscillation. Considering the large variation in results from different studies, it is important that comparisons are made only among muscle preparations dissected from the same trachea and fixed with the same batch of chemicals, as it was done in the present study.
1) There was regional variation in the density of thin filaments within the cytoplasm of trachealis cells, with the filaments slightly more concentrated near dense bodies. 2) Contractile activation was associated with an increase in the density of thin filaments; the density increase was uniform across the whole cell cross section. 3) Adaptation of trachealis cells to longer lengths resulted in a greater extent of thin filament formation. 4) Application of a brief period of oscillatory strain to the muscle caused a slight increase in thin filament density. 5) Thin filaments were less labile than the thick filaments.
This work was supported by Canadian Institutes of Health Research (CIHR) Operating Grant MOP 13271 (to C. Y. Seow). C. Y. Seow is a CIHR/British Columbia Lung Association Investigator.
We especially thank Pitt Meadows Meats Limited (Pitt Meadows, BC) for the supply of fresh porcine tracheas in kind support for this research project. In particular, we thank Cathy Pollock, Inspector in Charge, Canada Food Inspection Agency (Establishment #362), for help in obtaining the tracheas.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society