Chemically skinned single fibers from adult rat skeletal muscles were used to test the hypothesis that, in mammalian muscle fibers, myosin heavy chain (MHC) isoform expression and Ca2+- or Sr2+-activation characteristics are only partly correlated. The fibers were first activated in Ca2+- or Sr2+-buffered solutions under near-physiological conditions, and then their MHC isoform composition was determined electrophoretically. Fibers expressing only the MHC I isoform could be appropriately identified on the basis of either the Ca2+- or Sr2+-activation characteristics or the MHC isoform composition. Fibers expressing one or a combination of fast MHC isoforms displayed no significant differences in their Ca2+- or Sr2+-activation properties; therefore, their MHC isoform composition could not be predicted from their Ca2+- or Sr2+-activation characteristics. A large proportion of fibers expressing both fast- and slow-twitch MHC isoforms displayed Ca2+- or Sr2+-activation properties that were not consistent with their MHC isoform composition; thus both fiber-typing methods were needed to fully characterize such fibers. These data show that, in rat skeletal muscles, the extent of correlation between MHC isoform expression and Ca2+- or Sr2+-activation characteristics is fiber-type dependent.
- contractile proteins
- myosin light chain isoforms
- isometric activation characteristics
- skinned fiber
- extensor digitorum longus
mammalian skeletal muscle fibers contain a multitude of myosin heavy chain molecular forms (MHC isoforms) that confer the contractile apparatus-specific characteristics with respect to myofibrillar ATPase activity and velocity of shortening (20, 28). The contractile phenotype of a muscle fiber is described, however, not only by its myofibrillar ATPase activity and maximum velocity of shortening but also by characteristics associated with the process of isometric activation by Ca2+, such as Ca2+ threshold for activation, sensitivity to Ca2+, maximum Ca2+-activated force per cross-sectional area (CaFmax/CSA), and degree of cooperativity in development of Ca2+-activated force. Moreover, skeletal muscle fibers display quite striking differences with respect to some contractile activation characteristics when Sr2+ replaces Ca2+ as the activating ion (11).
Ca2+- or Sr2+-activation properties (11,38) and MHC isoform composition (22) have often been used separately as criteria for fiber typing in biochemical and physiological studies of muscle contractility. If these two criteria were tightly correlated, then each method, used alone, should provide the experimenter with complete information on the MHC isoform composition and physiological profile of a given fiber. Alternatively, if MHC isoform expression and Ca2+- or Sr2+-activation properties were not tightly correlated, then the two methods should be used in tandem when seeking to characterize a muscle fiber in molecular and functional terms.
Because the Ca2+- or Sr2+-activation characteristics of a muscle fiber depend not only on its MHC(s) but also on several other myofibrillar proteins, such as troponin C (which, in turn, can exist in different structural and/or functional states; see discussion), MHC isoform expression may not be fully correlated with Ca2+- or Sr2+-activation characteristics. The main aim of this study was to test the hypothesis that, in mammalian muscle fibers, MHC isoform expression is only partly correlated with Ca2+- or Sr2+-activation characteristics. To test this hypothesis, it was important to examine a broad range of fiber types, which would include all four types known to occur in trunk and limb muscles of rodents (type I, IIA, IID, and IIB) and hybrid fibers expressing combinations of two or three MHC isoforms. Such fibers were obtained by random dissection from soleus (Sol), extensor digitorum longus (EDL), and diaphragm muscles of Wistar-Kyoto (WKY) rats and from Sol of spontaneously hypertensive rats (SHR). The latter muscle was used because, in a recent study (2), we found that SHR Sol muscles are enriched in IIA, I + IIA, and IIA + IID fibers and that there are no differences with respect to contractile activation characteristics between fibers of the same type from age-matched SHR and WKY animals. A minor aim of this study was to test whether, in rat skeletal muscles, there is a tight correlation between myosin light chain (MLC) and MHC isoform expression.
MATERIALS AND METHODS
Animals and Muscle Dissection
Forty-three male adult rats (WKY and SHR), 15–24 wk old, were obtained from Baker Medical Research Institute (Melbourne, Australia) and housed at La Trobe University Animal House. The rats were maintained in a temperature-controlled environment (23°C) with a 12:12-h light-dark cycle and were allowed access to rat food and water ad libitum. On the day of the experiment, rats were killed by deep halothane inhalation; this was in accordance with the Animal Ethics Permits granted by the respective committees at Victoria University of Technology and La Trobe University. The Sol, EDL, and diaphragm muscles were quickly dissected, blotted dry on filter paper, and placed under paraffin oil (Ajax Chemicals, Sydney, Australia) in a petri dish with a Sylgard 184 transparent resin base (Dow Corning). When not used for dissection of single fibers, the muscles were refrigerated at 6.0 ± 1.0°C.
Preparation of Triton X-100-Skinned, Single Muscle Fiber Segments
All EDL and diaphragm muscles used in this study were obtained from WKY rats. Sol muscles were obtained from both SHR (49%) and WKY rats. Single fibers were isolated randomly under paraffin oil with the aid of dissecting microscope, fine jewellers forceps, and iris scissors. Paraffin oil 1) facilitates the visualization of a single fiber by having a different refractive index from that of the fiber, 2) precludes fiber swelling or fiber water loss, and3) confers the fiber a quasi-circular CSA through the surface tension exerted on the fiber at the oil-fiber interface. The length and width (mean of three values along the length) of the fiber segment were measured using a video camera monitor system (Olympus), and the volume of the segment was calculated with the assumption that the fiber was cylindrical with a diameter equal to the mean value of the fiber width (2).
After measurement of fiber dimensions, the preparations were mounted under oil at slack length (average sarcomere length 2.65 ± 0.02 μm, n = 45 measured by laser diffraction) between a force transducer (Sensonor 802, Horten, Norway) and a pair of fine Barcroft forceps. Each preparation was then transferred for 10 min to a relaxing solution (solution A; see below) containing 2% Triton X-100 to disrupt cellular membranes (30).
Determination of Ca2+- or Sr2+-Activation Characteristics of Chemically Skinned, Single Muscle Fiber Segments
Details regarding the composition and preparation of the solutions used in these experiments can be found in a series of earlier papers (1, 11, 18, 19, 31). Solutions containing different concentrations of ionized Ca2+(pCa = −log[Ca2+] ≥ 4.25, where brackets indicate concentration) were obtained by mixing (in various proportions) a relaxing solution (solution A) containing 50 mM EGTA2− (pCa ≥ 9) with a maximally Ca2+-activating solution (solution C) that contained 49.4–49.8 mM calcium EGTA2− and 0.2–0.6 mM excess EGTA2− (pCa = 4.7–4.25). Similarly, solutions containing different concentrations of Sr2+ (pSr = −log[Sr2+] ≥ 3.7) were obtained by mixing solution A with a maximally Sr2+-activating solution (solution D) that contained 39.8–40.4 mM strontium-EGTA2− and 9.6–10.2 mM EGTA2− (pSr 3.5). A preactivating solution (solution B) containing 49.75 mM hexamethylenediamine-N,N,N′,N′tetracetate and 0.25 mM EGTA2− (pCa ≈ 8; pSr ≥ 8) was also used to facilitate the rapid activation of the skinned fiber preparation (18) and thus minimize deterioration in the fiber induced by prolonged activation (26). All the aforementioned solutions also contained (in mM) 117 K+, 36 Na+, 1 Mg2+ (free), 90 HEPES, 8 total ATP, 10 creatine phosphate, and 1 sodium azide. The apparent affinity constants of Ca2+ and Sr2+ to EGTA were those measured for the conditions used in this study and were 4.78 × 106 M−1 and 1.53 × 104M−1, respectively (11, 36). The pH, osmolality, and ionic strength of all solutions were 7.10 ± 0.01 at 22°C , 295 ± 5 mosmol/kgH2O, and 234 ± 2 mM, respectively. The methods for the preparation of solutions and for the determination of pCa and pSr values in each solution have been described in detail elsewhere (11, 29, 31).
The activation characteristics of individual muscle fiber segments were determined from the isometric force responses developed at 21 ± 2°C in a series of strongly Ca2+- and Sr2+-buffered solutions. Figure1 shows representative Ca2+-activated force responses generated by a fast-twitch skinned muscle fiber exposed to a set of 10 activating solutions in decreasing order of pCa values (from 6.68 to 4.27) under near-physiological conditions. Before exposure to the strongly buffered Ca2+ or Sr2+-activating solutions, the Triton X-100-treated fiber segment was briefly relaxed in solution A and then equilibrated in the preactivating solution B, which was weakly buffered for both Ca2+ or Sr2+. Subsequently, the fiber preparation was maximally activated by Ca2+ in solution C, relaxed insolution A, preactivated in solution B, and exposed to the Ca2+-activating solutions. The fiber segment was then relaxed again in solution A, preactivated insolution B, maximally activated by Sr2+ insolution D, relaxed in solution A, equilibrated in solution B, and exposed to a set of 10 Sr2+-activating solutions in decreasing order of pSr values (from 6.4 to 3.5). After relaxation in solution A and equilibration in solution B, the preparation was once more activated in the set of Ca2+-activating solutions to ensure that the preparation did not markedly deteriorate during the experiment (data not shown). On the rare occasion (2 of 243 fibers) when, at the end of such an experiment, a fiber failed to generate at least 50% of the initial CaFmax, the data were not used for analysis. Finally, the preparation was placed in 12 μl of SDS-PAGE solubilizing buffer (see SDS-PAGE Analysis of MHC and MLC Segments) to determine the type(s) of MHC isoforms expressed in that fiber segment. Only fiber segments for which a complete set of Ca2+- or Sr2+-activation characteristics and MHC profiles were obtained were used in this study.
To define parameters that described the relationship between force and Ca2+ or Sr2+ concentrations, the fraction of the CaFmax or maximum Sr2+-activated force (SrFmax) developed by a fiber segment in various activating solutions (%maximal force; Pt ) had to be calculated. This was done by dividing the steady-state isometric force attained in a solution of a particular pCa or pSr by the interpolated value of the CaFmax or SrFmax response as described in detail elsewhere (26). This strategy corrects for the small degree of deterioration in force production associated with the repeated activation of the skinned fiber preparation (see Fig.1). Pt was then plotted against the pCa and/or pSr values in those solutions, and theoretical Hill curves of the type described by the equation Equation 1were fitted through the data points using a nonlinear regression analysis protocol (GraphPad Prism software). In Eq. 1 ,ni is the associated Hill coefficient, xis pCa or pSr, and x 50 is the pCa or pSr value for which 50% of the maximum activation isometric force (pCa50, pSr50) is reached. In the present study, the curves described by Eq. 1 are called “simple” Hill curves.
For some fibers, computer-fitted simple Hill curves did not lie very close to all the data points. Thus a criterion had to be defined for deciding what was a “best-fitted” curve. If experimental errors involved in estimating the position of the data points in the force-pCa or force-pSr plots are taken into consideration, then a best-fitted curve to the data points must pass no further than 0.05 CaFmax (or SrFmax) from individual data points in the force-pCa (pSr) plots. Therefore, if the force-pCa and force-pSr plots for a particular fiber were deemed to be well-fitted by the best curve derived from the simple Hill equation (Eq. 1 ), that fiber was classified as a simple fiber.
For all 241 fibers investigated here, force-pCa data points for the two complete sets of data were within 0.05 CaFmax from simple Hill curves where all the data points fell within 0.05 SrFmax from simple Hill curves. This was also the case for the force-pSr curves for 233 fibers. However, for eight fibers, the best-fitted simple Sr2+-Hill curve passed outside the individual experimental data points by more than 5% of the maximum activated force. In this case, a combination of two Hill curves (“composite” curve) was used to fit the experimental data points (38) with the GraphPad Prism software: Equation 2where w 1 and w 2are normalized weighting factors (w 1 +w 2 = 1), n 1 andn 2 are the corresponding Hill coefficients, and pSr50/1, pSr50/2 are the corresponding pSr50 values for the two Hill curves. Fibers that were characterized by simple Hill curves for both Ca2+ and Sr2+ are referred to as simple fibers, whereas fibers that displayed simple Hill curves for Ca2+ but composite curves for Sr2+ are referred to as composite fibers. Because the force-pSr data points for all eight fibers could be well fitted by composite Hill curves derived from Eq. 2 , these fibers were classified as composite fibers. The composite fibers represented a population of functionally heterogeneous fibers that provided us with valuable information regarding the correlation between MHC isoform composition and Ca2+ or Sr2+ activation characteristics. Fibers with composite force-pSr curves have been previously reported in mammalian muscle (12, 15, 38).
Simple fibers were further subdivided into two groups based on their relative sensitivity to Ca2+ and Sr2+ expressed as pCa50 − pSr50 (Δ50): “simple slow” fibers (Sslow; 119 fibers) (Δ50 ≤ 0.5) and “simple fast” fibers (Sfast; 114 fibers) (Δ50 ≥ 1.0). The composite fibers were also further subdivided into three groups based on their relative sensitivity to Ca2+ and Sr2+: “composite slow-slow” fibers (Comps-s; 2 fibers) (pCa50 − pSr50/1, pCa50 − pSr50/2 ≤ 0.5), “composite slow-fast” fibers (Comps-f; 4 fibers) (pCa50 − pSr50/1 ≤ 0.5 and pCa50 − pSr50/2 ≥ 1) and “composite fast-fast” fibers (Compf-f; 2 fibers) (pCa50 − pSr50/1, pCa50− pSr50/2 ≥ 1).
The following activation parameters were determined for each individual fiber segment: 1) CaFmax/CSA (kN/m2), determined from the amplitude of the first force response of a mechanically skinned fiber in the maximally Ca2+-activating solution and from its estimated cross-sectional area measured in paraffin oil before exposure to aqueous solution, 2) pCa10 (activation threshold for Ca2+), 3) pSr10 (activation threshold for Sr2+), 4) pCa50(sensitivity to Ca2+), 5) pSr50 (sensitivity to Sr2+; pSr50/1and pSr50/2 for composite fibers), 6)n Ca (minimum number of cooperating Ca2+-binding sites) (31), 7)n Sr (or n Sr1 andn Sr2 for composite fibers) (minimum number of cooperating Sr2+-binding sites), and8) pCa50 − pSr50 = log([Sr2+]50/[Ca2+]50) (Δ50; relative sensitivity to Ca2+ and Sr2+).
SDS-PAGE Analysis of MHC and MLC Isoforms
After contractile experiments were performed, single fiber segments were incubated in SDS-PAGE solubilizing buffer (62.5 mM Tris, 2.3% SDS, 5% β-mercaptoethanol, 12.5% glycerol, 13.6% sucrose, 0.01% bromphenol blue, 0.1 mM phenylmethylsulfonyl fluoride, 2 μM leupeptin, and 1 μM pepstatin) for 24 h at room temperature and then boiled for 5 min. MHC and MLC isoforms were analyzed on 0.75-mm-thick slab gels using the Hoefer Mighty Small gel apparatus. MHC isoforms were separated on SDS-PAGE by using the protocol of Bortolotto et al. (2) or a slightly modified version of the protocol of Talmadge and Roy (32). Briefly, the modified version of the Talmadge and Roy protocol comprised a separating gel consisting of 7.84% acrylamide, 0.16% bis-acrylamide, 200 mM Tris (pH 8.8), 0.4% SDS, and 100 mM glycine and a stacking gel consisting of 3.9% acrylamide. Separating and stacking gels both contained 42.5% vol/vol glycerol. The gels were run at 130 V. A 6-μl aliquot containing 0.15–0.30 nl of fiber was applied to each electrophoretic well, and the gels were stained using Bio-Rad silver stain plus. An internal standard was included in each gel (Fig.2, lane 1) to ensure that the gel resolved all four MHC isoforms and enabled for correct identification of MHC bands in single muscle fibers. Details of the protocols used for MLC isoform separation by SDS-PAGE can be found in Bortolotto et al. (2). The gels for both high- and low-molecular-weight proteins were analyzed using a Molecular Dynamics personal densitometer and the volumetric quantitation method, with background correction provided by the ImageQuant software version 4.1 (Molecular Dynamics). For more details regarding densitometric analyses, see Bortolotto et al. (2).
All data are expressed as means ± SE. Unless stated otherwise, statistical comparisons were performed between groups comprising at least three sets of data using a one-way ANOVA followed by the Bonferroni test. Statistical significance was accepted atP < 0.05.
Fiber Types Classified According to MHC Composition
Trunk, limb, and respiratory muscles of adult rats have been shown to express four MHC isoforms: MHC I (slow), MHC IIb (fast), MHC IId (fast), and MHC IIa (fast) [Pette and Staron's nomenclature (23)], which can be distinguished by differences in electrophoretic mobility on SDS-polyacrylamide gels. On the basis of their MHC composition, the 241 single fiber segments examined in the present work (146 Sol fibers, 52 EDL fibers, and 43 diaphragm fibers) belonged to three major groups: 1) pure fibers, expressing only one MHC isoform (fiber types I, IIA, IID, and IIB), 2) slow/fast hybrid fibers, coexpressing one slow and one fast (fiber types I + IIA and I + IID) or one slow and two fast MHC isoforms (fiber type I + IIA + IID), and 3) fast hybrid fibers coexpressing two or three fast MHC isoforms (fiber types IIA + IID, IID + IIB, and IIA + IID + IIB) (Fig.2). No hybrid fibers coexpressing four MHC isoforms or the combinations of MHC isoforms I + IIb, IIa + IIb, I + IIa + IIb, or I + IIb + IId were found in this study. The largest proportion (73.4%) of fibers contained only one MHC isoform, 23.7% contained two MHC isoforms (2 MHC hybrids), and 2.9% contained three MHC isoforms (3 MHC hybrids).
A summary of the proportion of different fiber types detected in the fiber populations obtained from Sol, EDL, and diaphragm muscles of 43 adult SHR and WKY rats is given in Fig.3. The fiber populations dissected from Sol and diaphragm contained 89% and 69.7% pure fibers, respectively; in contrast, only 32.7% of the EDL fibers displayed a single MHC isoform. The largest proportion of fibers dissected from EDL (65.4%) coexpressed two MHC isoforms [IIa + IId (17.3%) and IId + IIb (48.1%)]. The fiber population obtained from diaphragm muscles comprised the largest proportion (13.9%) of hybrid fibers coexpressing three MHC isoforms.
Fiber Types Classified According to Ca2+- or Sr2+-Activation Characteristics
Muscle cell physiologists from several laboratories have classified mammalian fibers according to a number of criteria related to the isometric activation of myofibrillar proteins by Ca2+ or Sr2+ (6, 11, 17, 37, 38). According to two of these criteria, namely, the type of Hill curve that best fit the Sr2+-activated force data points (simple or composite) and the relative sensitivity of the contractile regulatory system to Ca2+ and Sr2+ (for details of the criteria see materials and methods), the fibers examined in this study belonged to five groups: Sslow, Sfast, Comps-s, Comps-f, and Compf-f.
Figure 4 shows representative force-pCa and force-pSr curves produced by Sslow and Sfast fibers, and Fig. 5shows the force-pSr curves for Comps-s, Comps-f, and Compf-f fibers. It is worth noting that, based on our condition that the computer-fitted curve should be no further than 5% of the maximum activated force from the data points, the rising phase of the force-pSr curve that fitted the data for the Comps-s and Compf-f fibers passed through only three and four experimental points, respectively, whereas that for the Comps-f fiber passed through seven data points (Fig. 5). Thus the probability that a fiber classified as a composite fiber was a misclassified simple fiber would be lower for Comps-f and higher for Compf-f and Comps-s fibers.
Correlation Between the MHC Isoform-Based and Ca2+- or Sr2+-Activation Property-Based Fiber-Typing Methods
The majority (97.5%) of the pure MHC type I fibers and 40% of hybrid type I + IIA fibers produced force-pSr curves of the Sslow type. All pure IIA (n = 16), IID (n = 35), and IIB (n = 6) fibers, all hybrid IIA + IID fibers (n = 25), all hybrid IIA + IID + IIB fibers (n = 2), all hybrid I + IIA + IID fibers (n = 5), 23 hybrid IID + IIB fibers (92%), and two hybrid I + IIA fibers (40%) generated force-pSr curves of the Sfast type. Finally, eight fibers of various types generated composite force-pSr curves. These included three pure type I fibers (2.5%), two hybrid type I + IID fibers (100%), two hybrid type IID + IIB fibers (8%), and one hybrid I + IIA fiber (20%).
To allow a full perspective of the correlation between the contractile activation-based and MHC isoform-based fiber-typing methods, these data are presented as proportions of different MHC-based fiber groups comprised in each class of fibers defined by Sr2+-activation characteristics (Fig.6). The majority (98.3%) of the 119 fibers that were classified as Sslow fibers contained only the slow MHC isoform and a small proportion (1.7%) coexpressed a slow and fast MHC isoform. Details of the relative proportion of the two MHC chain isoforms and contractile activation parameters in these and two other hybrid I + IIA fibers are given in Table 3. In contrast, only 50% of the 114 Sfast fibers examined contained one fast MHC isoform, whereas the other 50% coexpressed two fast MHC, three fast MHC, one slow and one fast MHC, and one slow and two fast MHC isoforms. The population of fibers displaying heterogeneous Sr2+-activation characteristics (composite fibers) was relatively small (8 fibers) and comprised 62.5% fibers that coexpressed two MHC isoforms and 37.5% fibers that contained only the slow MHC I isoform.
Maximum Specific Ca2+or Sr2+-Activated Force and Ca2+- or Sr2+-Activation Characteristics of Fibers Typed According to MHC Composition
CaFmax/CSA, CaFmax-to-SrFmax ratio (data not shown), sensitivity to Ca2+ or Sr2+ of the contractile-regulatory system (pCa50, pSr50), relative sensitivity of the contractile-regulatory system to Ca2+ and Sr2+ (pCa50− pSr50; Δ50), activation thresholds for Ca2+ and Sr2+ (pCa10, pSr10), and Hill coefficients (n Caand n Sr) were determined for all 241 fibers examined. Results, grouped according to MHC-based fiber type and muscle type (Sol, EDL, diaphragm), are presented in Tables 1-4.
CaFmax/CSA and CaFmax/SrFmax.
For each of the fiber type groups that could be subjected to statistical analysis (IIA, IID, and IIA + IID), there were no significant differences between the maximum Ca2+-activated specific forces developed by fiber populations from Sol, EDL, and diaphragm muscles (Table 1). In EDL muscles, pure or hybrid fibers containing the fast MHC isoform IIb (IIB and IIB + IID fibers) consistently developed the highest specific forces recorded in this study. These forces were significantly higher than those developed by IIA + IID fibers from EDL. The lowest value for CaFmax/CSA was encountered for the diaphragm fiber expressing MHC IIa + IId + IIb, suggesting that some of the expressed myosin may not have been functional.
For all fiber types investigated, the CaFmax-to- SrFmax ratio was close to 1 (data not shown), which suggests that under our conditions Ca2+and Sr2+ were equally effective in maximally activating the contractile-regulatory system.
Ca2+- or Sr2+-activation characteristics.
As seen in Tables1-3, for each of the three muscles examined, there were no significant differences with respect to pCa10, pCa50, pSr10, and pSr50 between fibers expressing only fast MHC isoforms.
More marked interfiber type differences related to Ca2+- or Sr2+-activation characteristics were found between fibers dissected from Sol muscles. Thus significant differences with respect to pCa10 were found between the two Sol type I fiber subgroups Sslow and Comp. Furthermore, both type I (Sslow) and type I (Comp) fibers showed a higher sensitivity to Ca2+ or Sr2+ activation than all Sol fast-twitch fiber types. For the two groups of fast-twitch fibers that allowed statistical analysis (IIA and IIA + IID), this difference was highly statistically significant. The only type I fiber obtained from diaphragm also displayed higher pCa10, pCa50, pSr10, and pSr50 values than the diaphragm fibers containing only fast MHC isoforms.
It is worth noting that type IIA fibers from Sol and EDL muscles differed significantly with respect to pCa10 and pCa50; as seen in Table 1, Sol fibers appeared to be more sensitive to Ca2+ activation by ∼0.14–0.16 log units than their EDL counterparts. Because there were no muscle-specific differences with respect to sensitivity to Sr2+ activation between fibers of the same type, the significant difference with respect to pCa50 between homologous Sol and EDL fibers was also translated into significant difference with respect to the value of Δ50.
As seen in Tables 1-3, both groups of fast-twitch fibers dissected from Sol (IIA and IIA + IID) displayed significantly highern Ca and n Sr values than pure type I fibers. In Sol and diaphragm muscles, no significant differences with respect to n Ca andn Sr were found between fast-twitch fiber types; however, in EDL, the contractile activation curves for IIA + IID fibers displayed significantly lower slopes (i.e., smaller nvalues) than the curves for pure IIA fibers (significantly so forn Ca). The IIA + IID fibers dissected from EDL and diaphragm muscles also displayed significantly lower values forn Ca and n Sr than their counterparts in the Sol muscles. It is important to note that the average values for n Ca andn Sr recorded for all fiber groups in Tables1-3, which contained more than one fiber, were very similar [index of similarity: (n Ca −n Sr)/(n Ca +n Sr) <0.1; see Ref. 38], which suggests that molecular events that follow the binding of the activating ion Ca2+ or Sr2+ to the regulatory system are very similar.
The fiber population examined also comprised 1) a group of eight hybrid and pure fibers with respect to MHC composition that produced composite force-pSr curves and 2) a group of four hybrid fibers that displayed a combination of fast and slow MHC isoforms but behaved as simple fibers in terms of Sr2+activation. Tables 3 and 4 present the Ca2+- or Sr2+-activation characteristics and MHC isoform composition of each individual fiber from these two groups.
As shown in Table 3, the two slow/fast MHC hybrids, which coexpressed MHC isoforms I and IId, behaved as Comps-f fibers. However, there appeared to be no tight quantitative correlation between the proportion of MHC isoforms (estimated densitometrically) and the proportion of the two contractile activation components (seematerials and methods). For example, diaphragm (I + IID) fiber 1 contained 1.6-fold more fast MHC isoform; however, the proportion of the fast component seen in the force-pSr curve was only about one-half that of the slow component. This numerical discrepancy could be attributed either to a genuine lack of correlation between MHC isoform composition and the contractile activation parameters under consideration or to difficulties in calculating the correct proportion of the two components from the force-pSr curves, if the two components were not operating in parallel (16). The rather large values forn Sr in several instances in Table 3 could be the result of relatively too few data points over the pSr range relevant to the individual components.
Each individual fiber from the second group of composite fibers in Table 3, namely, the type I fibers dissected from Sol muscles, appeared to contain only one MHC isoform (MHC I); however, either slow-fast characteristics (fiber 1: pCa50 − pSr50/1 = 0.28; pCa50 − pSr50/2 = 1.09) or slow-slow characteristics (fibers 2 and 3: pCa50 − pSr50/1 < 0.3; pCa50 − pSr50/2 ≤ 0.5) were displayed. The discrepancy between MHC expression and contractile activation characteristics for fiber 1 could be due to the inability of the densitometric method to detect the presence of a minor fast MHC isoform rather than to its genuine absence, whereas the discrepancies for fibers 2 and3 may be due to the data-fitting constraints used in this study (i.e., that no force-pSr data points should lie further than 0.05 SrFmax from the computer-fitted Hill curve; seematerials and methods). The two fibers from the last group in Table 3 displayed Compf-f characteristics and were also found to express a combination of MHC IId and IIb isoforms.
Table 4 shows results obtained with four fibers that coexpressed MHC isoforms I and IIa but displayed simple force-pSr characteristics. Infibers 3 and 4, the fast MHC IIa isoform content was higher than 90% and the Δ50values were greater than 1, which is a characteristic of fast fibers.Fibers 1 and 2, on the other hand, contained 80–90% MHC I and less than 20% MHC IIa and their Δ50 values were <0.5, which is characteristic of slow fibers.
In summary, the data presented in Tables 3 and 4 do not support a tight correlation between MHC composition and contractile activation characteristics as such. Nevertheless, lack of correlation between MHC composition and fast or slow contractile activation characteristics was encountered only in very rare instances, and this is illustrated in Fig. 7, in which the relative sensitivity for Ca2+ and Sr2+ (pCa50 − pSr50) was plotted for all 241 fibers studied as a function of the fast MHC fraction expressed. The data points for fibers expressing either only fast MHC isoforms (MHC II-to-MHCtotal ratio = 1.0) or only MHC I isoform (MHC II-to- MHCtotal ratio = 0.0) are clustered in relatively narrow bands that do not overlap, and the data points for the 12 fibers that expressed both fast and slow MHCs lie somewhere in between.
MLC Expression in Fibers Typed According to MHC Isoform Composition
The correlation between MLC and MHC isoform expression was examined in a subpopulation of 155 simple fibers and 5 composite fibers dissected from Sol and diaphragm muscles. As shown in Table5, there was full agreement between MLC1 and MLC2 isoform expression and MHC isoform composition in most of the fibers with simple force-pSr curves [83% (105/131) of Sol fibers; 91.7% (22/24) of diaphragm fibers] and in all composite fibers. Isoforms of MLC3 were not included in this analysis because of uncertainty regarding their identity on SDS-PAGE gels. All type I fibers displayed full correlation between MHC and MLC expression. The pool of fibers that showed disagreement between MHC and MLC isoform composition included pure IIA fast fibers and various types of hybrid fibers (I + IIA, IIA + IID, IIA + IIB + IID). Interestingly, all Sol fibers expressed the slow MLC I isoform even when only fast MHC isoforms were expressed (Table 5). Also, the diaphragm fiber that expressed a combination of all three fast MHC isoforms (fibers IIA + IID + IIB) expressed the slow isoforms of both types of MLCslow, MLC1slow and MLC2slow. The electrophoretograms of two fibers, one showing full correlation (MHC I + MLC1slow + MLC2slow; lane 1) the other only partial correlation (MHC IIa + MHC IId + MLC1slow + MLC1fast + MLC2fast; lane 2) between the isoform expression of myosin subunits, are presented in Fig. 8.
So far, studies concerned with the Ca2+- or Sr2+-activation characteristics of single muscle fibers, classified according to MHC isoform composition, have been small in number and involved only a limited range of animals, fiber type groups, and/or contractile parameters. For example, Danieli-Betto et al. (7) determined several parameters related to Ca2+ activation (but not Sr2+ activation) in type I, IIA, and IIB fibers from rat Sol, EDL and diaphragm muscles but deliberately discarded all hybrid fibers from their analyses. Because this study was undertaken before the discovery of the MHC IId/x isoform (35), one cannot therefore be certain about the true identity of the type II fibers. More recently, Cordonnier et al. (6) determined several Ca2+- or Sr2+-activation parameters of electrophoretically typed pure and hybrid fibers of rhesus monkey Sol muscles, but the fiber population examined by these authors contained only type I, IIA, and I + IIA fibers. A broader range of human MHC-based fiber types, which included slow-fast and fast-fast hybrid fibers as well as pure fibers (I, IIA, IIB, I + IIA and IIA + IIB), was investigated with respect to contractile activation characteristics by Bottinelli et al. (3), but this study was conducted at 12°C and the authors considered only parameters associated with Ca2+activation of the contractile-regulatory system.
The present work fills in this knowledge gap by considering fibers from ten MHC-based fiber-type groups (I, IIA, IID, IIB, I + IIA, I + IID, IIA + IID, IID + IIB, I + IIA + IID, and IIA + IID + IIB) and contractile parameters related to both Ca2+ and Sr2+ activation under more physiological conditions with respect to the ionic composition and temperature of activating solutions. The fibers, obtained by random dissection from fresh Sol, EDL and diaphragm muscles of adult rats, were examined systematically with respect to size, maximum Ca2+ or Sr2+ activated tension/CSA, threshold to Ca2+ or Sr2+ activation (pCa10, pSr10), sensitivity of the contractile-regulatory system to Ca2+ or Sr2+ (pCa50, pSr50), relative sensitivity of the contractile-regulatory system to Ca2+ and Sr2+(Δ50), and degree of cooperativity (n) between Ca2+ or Sr2+binding sites.
It is worth noting that the sample size was small (n < 5) in some of the fiber type groups, based on MHC isoform composition (I + IID, n = 2; IIA + IID + IIB,n = 2) or Ca2+- or Sr2+-activation characteristics (Comps-f,n = 4; Comps-s, n = 2; Compf-f, n = 2). Therefore, caution was applied when discussing results obtained with these fibers.
Fiber Type Populations in Sol, EDL and Diaphragm Muscles of SHR and WKY Rats
As expected from published data on the fiber type composition of adult rat Sol muscle (85–93% MHC I and 7–15% MHC IIa; Refs.5, 10, and 13), the majority of fibers isolated from Sol muscles of WKY and SHR contained only the slow MHC I isoform. A small proportion of Sol fibers contained MHC IIa or a combination of MHC I and IIa isoforms, but most of these fibers were obtained from Sol muscles of SHR. The diaphragm muscles of WKY rats produced a very large proportion (more than 70%) of fibers expressing only the MHC IId isoform and a sizeable proportion of fiber expressing MHC IId in combination with one or two other fast MHC isoforms. Together, these data provide strong evidence that Sol and diaphragm muscles of adult rats, regardless of the strain, are rich sources of pure type I and type IID fibers, respectively. The number of pure type I fibers dissected from the diaphragm muscle was very low (1/43). However, the number of hybrid fibers that expressed MHC I isoform (in combination with MHC II isoforms) was relatively high (21% of the total number of diaphragm fibers). This is in very good agreement with reports from other laboratories that rat diaphragm muscle contains ∼15–26% MHC I isoform (24, 25).
Random dissection of single fibers from EDL muscles of WKY rats produced neither type I fibers nor a large proportion of one particular type of fast fibers. In fact, most EDL fibers examined in this study coexpressed two (particularly IIb + IId) or three fast MHC isoforms, indicating that EDL is a remarkably heterogenous tissue with respect to fiber-type composition. A relatively large number of IID + IIB fibers (37%) was also detected by Li and Larsson (14) in a population of EDL fibers (43) randomly dissected from young adult Wistar rats, which suggests that the relatively high proportion of fast-fast hybrid fibers obtained from EDL muscles in the present study is related to the type of muscle from which the fibers were dissected rather than to the rat strain.
So far, hybrid muscles fibers have been largely regarded as fibers in transition because their proportion was found to increase in muscles undergoing transformation (23). The possibility that hybrid fibers may also represent stable “phenotypes” has been recently discussed to some length by Rivero et al. (27) and Talmadge et al. (34) who found that skeletal muscles of clinically healthy, freely moving, untrained horses contain a large proportion of hybrid fibers (>80%) (27) and that in rat Sol, hybrid fibers whose number increased dramatically after spinal cord transection persisted in high proportion (>80%) a long time (1 yr) after surgery (34). As already mentioned, a large proportion (67%) of the EDL fibers and 25.6% of the diaphragm fibers examined in the present study contained two or three MHC isoforms. If one assumes that these muscles were not undergoing a process of transformation, because they were obtained from functionally normal, adult animals that had not been subjected to intense or reduced mechanical activity, one can conclude that the hybrid fiber types described here are also stable rather than transitory cellular species.
Two of the 43 diaphragm fibers examined in this study expressed a combination of MHC isoforms (I + IId), which according to Pette et al. (21) is not compatible with the sequential fiber-type transition I ↔ IIA ↔ IID ↔ IIB. To date, no study has reported the presence of fibers displaying “atypical nonnearest neighbor combination” [phrase introduced by Pette et al. (21)] of MHC isoforms in normal, nontransforming muscles. Hybrid fibers of type I + IID have been detected, however, by others in non-weight-bearing Sol muscles from hyperthyroid rats (4) and from rats subjected either to a spaceflight program or hindlimb suspension (33).
One possible explanation for our finding is that rat diaphragm fibers undergo a continuous level of transformation even in adult animals that are not subjected to muscle transformation-inducing conditions. Alternatively, the I + IID fibers detected by us in the diaphragm of WKY rats may be stable rather than transitory entities and as such do not have to obey the rule of “nearest neighbor” MHC isoform combination predicted by Pette et al. (21) for transforming fibers. Finally, the two fibers could be I + IIA + IID fibers, misclassified as I + IID fibers, because they expressed only a nondetectable amount of MHC IIa isoform. Based on a close examination of the intensities of MHC isoform bands I and IId on the stained electrophoretograms, we believe this latter possibility to be very unlikely.
MHC Composition and Ca2+- or Sr2+-Activation Characteristics Under Near-Physiological Conditions
When the temperature and ionic composition of the activating solutions were close to physiological conditions, significant inter-fiber-type differences were observed with respect to the sensitivity of the myofibrillar compartment to Ca2+ or Sr2+ activation, degree of cooperativity in development of Ca2+- or Sr2+-activated force, and relative sensitivity to Ca2+ or Sr2+ (Δ50) between fibers expressing the slow MHC I isoform and fibers expressing one or several MHC II isoforms. For example, in SOL muscles, type I fibers displayed a higher sensitivity to Ca2+ or Sr2+ and lower n values than type IIA and type IIA + IID fibers. A higher sensitivity to Ca2+ or Sr2+ and lower n values were also observed when composite type I fibers from Sol were compared with the two composite type IID + IIB fibers from EDL. These results are in agreement with previous data reported by us (11) and others [for review see Schiaffino and Reggiani (28)] regarding the higher sensitivity to Ca2+ and Sr2+ activation in slow-twitch compared with fast-twitch fibers.
No significant differences with respect to threshold for Ca2+ or Sr2+ activation and sensitivity of the contractile-regulatory system to Ca2+ and Sr2+were observed between fibers belonging to the five fiber type groups from EDL muscles that expressed only fast MHC isoforms (IIA, IID, IIB, IIA + IID, and IIB + IID). Furthermore, with the exception of IIA + IID fibers that had a significantly lowern Ca than IIA fibers, no difference between EDL fiber groups expressing only fast MHC isoforms were found with respect to the minimum number of cooperating Ca2+- or Sr2+-binding sites (n Ca andn Sr). Similar observations were made when comparing the three fiber type groups expressing fast MHC isoforms in diaphragm muscles (IID, IIA + IID, and I + IIA + IID).
These results disagree with previous reports that IIB fibers are more sensitive to Ca2+ than IIA fibers (3, 7, 9) and that the cooperativity between Ca2+-binding sites in IIB fibers is higher than that in IIA fibers (7, 9). We believe that this discrepancy cannot be attributed to muscle-related differences because data reported by Eddinger and Moss (9) and Danieli-Betto et al. (7) were obtained with rat diaphragm and rat EDL, respectively. However, we noted marked differences in the composition and temperature of activating solutions between our study and those of Eddinger and Moss (9), Danieli-Betto et al. (7), and Bottinelli et al. (3), which could be responsible for observed differences between these studies. For example, measurements of Ca2+- or Sr2+-activation characteristics were carried out at 21–23°C in our study but at 12–15°C in previous studies (3, 7), and it is possible that the temperature dependence of the various parameters is different between fiber types.
An unexpected finding of this study was that IIA + IID and IIA fibers from Sol displayed significantly higher sensitivity to Ca2+ than IIA + IID fibers from EDL and diaphragm and IIA fibers from EDL, respectively. This is the first report of intermuscle differences with respect to Ca2+ sensitivity of isometric tension between fibers of the same fast type. Interestingly, most IIA Sol fibers (10/11) and IIA + IID Sol fibers (11/12) coexpressed MLC1slow and MLC1fast isoforms and the remaining 8% of the (IIA + IID) SOL fibers expressed the slow and fast isoforms of both MLC1 and MLC2. In contrast, the type IIA + IID fiber from diaphragm was found to contain only the fast isoforms of MLC1 and MLC2. These results suggest that the presence of slow MLC isoforms in Sol fibers of types IIA and IIA + IID may be somehow related to the higher Ca2+ sensitivity of isometric force displayed by these fibers when compared with fibers of the same type from diaphragm muscles.
The presence of slow MLC isoforms in Sol fibers of types IIA and IIA + IID as well as in the fast-fast-fast MHC isoform hybrid (IIA + IID + IIB) dissected from diaphragm indicates that MHC and MLC isoform expression are not tightly correlated in rat Sol and diaphragm muscles. This conclusion is in agreement with previous reports from other laboratories [for review see Pette et al. (21)].
A Comparison Between the MHC Isoform-Based and Ca2+- or Sr2+-Activation Properties-Based Muscle Fiber Typing Methods
Regarding the identification of the slow-twitch fibers, there was almost complete overlap between the MHC isoform-based and the Ca2+- or Sr2+-activation properties-based methods. Thus 97.5% of the pure MHC I fibers have been identified as Sslow fiber type and 98.3% of the Sslow fibers have been identified as pure MHC I fibers (Table 1 and Fig. 6). Of the pure type I fibers, 1.7% and 0.8% displayed Comps-s and Comps-f characteristics, respectively. If one regards the Comps-s fiber type as a subtype of slow-twitch fibers, then 99.2% of MHC I fibers displayed slow-twitch Ca2+- or Sr2+-activation characteristics. Conversely, only 1.7% of the type Sslow fibers were hybrid fibers, expressing both MHC isoforms I and IIa. As mentioned in results, the small discrepancy between the two methods with respect to the classification of slow-twitch fibers could be reduced even further if experimental errors in the classification of fibers by two methods are taken into account. Therefore, one could reliably use either of the two methods to identify slow-twitch fibers in the rat muscle.
All fibers that expressed only fast MHC isoforms (MHC IIa, MHC IIb, MHC IId, and combinations thereof) displayed fast Ca2+- or Sr2+-activation characteristics, 98.2% being classified as Sfast and 1.8% as Compf-f (Table 1). Fast Ca2+- or Sr2+-activation characteristics were also displayed by a small but significant fraction (6.1%) of fibers that contained slow and fast MHC isoforms (4.4% expressing MHC I + MHC IIa + MHC IId and 1.7% expressing MHC I + MHC IIa), but all these fibers expressed predominantly fast MHC isoforms. These data indicate that one could broadly predict the Ca2+- or Sr2+-activation characteristics of a fiber found to contain one, several, or a majority of fast MHC isoforms, but one could not predict the MHC composition of a simple fiber displaying fast Ca2+- or Sr2+-activation characteristics.
Further rigorous subclassification of type Sfast fibers into groups that could be closely related to the expression of specific fast MHC isoforms was not possible, although there was a clear tendency for fibers expressing MHC IIb either alone or in combination with other fast MHC isoforms to produce larger CaFmax/CSA when maximally activated (Table 2). This difference was, however, not marked or consistent enough to allow prediction of the MHC composition of a fast-twitch fiber based on its Ca2+- or Sr2+-activation characteristics. Significant differences were found between pure IIA fibers from Sol and EDL muscles with respect to pCa10, pCa50, and pCa50 − pSr50, which made it difficult to accurately predict the absolute values of Ca2+- or Sr2+-activation parameters of a fast fiber from its pattern of MHC expression.
From the group of 12 fibers that expressed both fast and slow MHC isoforms, only 25% were classified as Comps-f (3 fibers), 58.3% were classified as Sfast (7 fibers), and 16.7% were classified as Sslow (2 fibers). Conversely, 75% of the Comps-f fibers were hybrid with respect to fast and slow MHC isoform expression.
Thus, although it is recognized that there is not 100% correspondence between the MHC isoform expression and isometric Ca2+- or Sr2+-activation characteristics, the overall results indicate that in the adult rat muscles there is more than 90% overlap between the two methods of classification of fast and slow fibers. This is also reflected in Fig. 7, in which the fiber-type classification based on the pCa50 − pSr50 value alone is closely correlated with the fraction of fast MHC isoforms expressed in the fiber.
The partial rather than full correlation between the two methods demonstrates that MHC expression and isometric Ca2+- or Sr2+-activation characteristics are not causally related. The Ca2+- or Sr2+-activation characteristics likely depend on the isoform of troponin C expressed in the particular fiber (12). In skeletal mammalian fibers, troponin C exists in two isoforms (one slow and one fast) that are known to have distinct properties with respect to the relative affinity for Ca2+ and Sr2+ (8) and to the number of Ca2+-binding sites. The high level of correlation between the expression of fast and slow MHC isoforms and the isometric Ca2+- or Sr2+-activation characteristics found in the present study indicates that the expression of fast and slow troponin C isoforms and fast and slow MHC isoforms are closely correlated.
There is compelling evidence that troponin C is not the only determinant of the isometric force-pCa/pSr relationship because it is known that alteration in the structure of other myofibrillar components, including MHC, MLC, protein C, troponin T, and tropomyosin, can alter the position and the steepness of the isometric force-pCa/pSr curves (20, 38). This explains the considerable variability in the various isometric Ca2+- or Sr2+-activation characteristics within certain fibers expressing the same MHC isoform or combination of isoforms.
In conclusion, results presented in this study show that 1) in rat skeletal muscles, the extent of correlation between MHC isoform expression and Ca2+- or Sr2+-activation characteristics is fiber type dependent, 2) to obtain structural and functional information on pure or hybrid fast-twitch fibers, both MHC isoform composition-based and Ca2+- or Sr2+-activation characteristic-based fiber typing methods have to be used, 3) in Sol and diaphragm muscles of the rat, the expressions of MHC and MLC isoforms are not tightly correlated,4) nontransforming diaphragm and EDL muscles obtained from functionally normal rats contain a large proportion of hybrid fibers, and 5) rat diaphragm muscles from functionally normal adult rat contain fibers expressing an “atypical” combination of MHC isoforms.
This work was supported by the National Health and Medical Research Council (Australia) and the Australian Research Council.
Address for reprint requests and other correspondence: G. M. M. Stephenson, School of Life Sciences and Technology, Victoria Univ. of Technology, PO Box 14428, MCMC, Melbourne, Victoria 8001 Australia (E-mail:).
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 © 2000 the American Physiological Society