Maximum lift production and the thermal sensitivity of lift production increase dramatically during adult maturation of Libellula pulchella dragonflies. Here, we report that the mechanistic basis for this transition appears to involve a developmental change in protein expression, which alters the Ca2+-sensitivity of muscle activation and twitch contraction kinetics. The alternatively spliced Ca2+ regulatory protein troponin T (TnT) undergoes an isoform shift during adult maturation. Skinned (demembranated) fibers of mature flight muscle are up to 13 times more sensitive to activation by Ca2+ than skinned fibers from teneral (newly emerged adult) flight muscle, and their Ca2+-sensitivity is more strongly affected by temperature. Intact muscle from mature individuals has a shorter time to peak tension and longer time to half-relaxation during twitch contractions, which is consistent with a greater Ca2+-sensitivity of mature muscle. Because it becomes activated more quickly and relaxes more slowly, mature flight muscle is able to generate, with each twitch, more force per unit area than teneral muscle; this difference in force becomes greater at high temperatures. There do not appear to be any age-related differences in actomyosin crossbridge properties, since teneral and mature flight muscles do not differ in shortening velocity, tetanic tension or instantaneous power output during isotonic contraction. Thus, variation in TnT expression appears to affect the temperature-dependent Ca2+-sensitivity of muscle activation, which in turn affects the kinetics and force production of the twitch contractions used by dragonflies during flight. This cascade of effects suggests that maturational changes in the expression of TnT isoforms may be a key determinant of overall muscle and organismal performance.

Significant behavioral and thermoregulatory changes occur during adult maturation of Libellula pulchella dragonflies. Newly emerged adults (tenerals) are relatively sedentary, spending less than 2 % of their time in flight, whereas these dragonflies are much more active at maturity, spending an average of 30 % of certain periods of the day in flight, much of which is strenuous (Marden et al. 1996). Thoracic temperature (Tth) during flight also varies with age, as tenerals have Tth values ranging from 29 to 40 °C, whereas matures have Tth values ranging from 38 to 45 °C (Marden et al. 1996). The thermal physiology of flight performance undergoes a parallel change, as matures have a mean optimal Tth (the temperature at which maximum lift is produced) of 43.6 °C compared with the teneral mean optimal Tth of 34.6 °C (Marden, 1995a). At high Tth, matures generate an average of 50 % more lift per kilogram of flight muscle and have greater wingbeat frequencies than tenerals.

The thermal shift and the marked increase in flight performance evident during maturation of L. pulchella dragonflies could have a wide variety of mechanistic bases. We hypothesized that the mechanism might involve muscle contraction, and we therefore set out to examine two fundamentally different aspects of muscle contractile physiology: (1) crossbridge kinetics, as shown by shortening velocity and tetanic tension of intact muscle, and (2) contractile regulation by thin filament proteins, as shown by the Ca2+-sensitivity of skinned muscle fibers. The results presented here suggest that a change in isoform expression of the Ca2+ regulatory protein troponin T causes a substantial increase in Ca2+-sensitivity of muscle activation which, in turn, affects the timing and force generation of twitch contractions, which ultimately determine whole-organism performance.

Dragonflies and their ages

Libellula pulchella Drury dragonflies (Odonata, Libellulidae) were collected at field sites in Centre County, Pennsylvania, USA. Relative age, broadly classified as teneral (newly emerged), adolescent (approximately midway between teneral and mature in body mass) or mature, was determined according to objective criteria described previously (Marden, 1995a). Note that flight muscle mass approximately doubles between emergence and maturity (Marden, 1995a) as a result of hypertrophy of muscle cells (Marden, 1989).

Whole-muscle contractile performance

Shortening velocity, tetanic tension and twitch contraction kinetics were measured from in situ preparations of individual mechanically isolated L. pulchella flight muscles. The head, wings and legs were removed (the abdomen was left in place to provide ventilation via its continued rhythmic contractions), after which the ventral thorax was mounted with epoxy resin to the bottom of a temperature-controlled chamber in such a way that the dorso-ventral flight muscles were oriented vertically. The first dorso-ventral muscle of the mesothorax, which drives the downstroke of the forewing leading edge, was isolated through a small incision in the dorsal thoracic cuticle, and its apodeme was attached with cyanoacrylate glue to an insect pin suspended from the lever arm of a Cambridge Technology 300B lever system. This muscle was held at a length approximately 2 % longer than its resting length, as determined by comparison with the undisturbed contralateral muscle. A 0.25 ms square-wave stimulus (a single stimulus for twitches; a 200 Hz train for tetani) from a Grass S4B stimulator was applied with fine-gauge electrodes inserted in the lateral thorax; the intensity of the stimulus was set at 125 % of supramaximal voltage. Muscle temperature was monitored with a fine-gauge copper–constantan thermocouple inserted into the thorax and connected to a Physitemp Bat-12 thermocouple thermometer. Water was pumped from a temperature-controlled bath through the water jacket of the test chamber in order to maintain muscle temperature at desired levels. A MacLab/8 (AD Instruments) was used to convert force and position signals from the lever system into 12-bit digital data and send them to a Macintosh Quadra 700, where they were monitored in real time with software emulating an oscilloscope (Scope; AD Instruments) at a sampling frequency of 4 kHz.

Shortening velocity and tetanic tension were measured for one muscle preparation from each of 10 tenerals and 12 matures. Muscles were tested at 1–6 temperatures (alternating between either an ascending or a descending order of temperatures at 3–5 °C intervals from a starting point at room temperature). Experiments were terminated whenever tetanic tension degraded by more than 25 % from levels measured at the start of the experiment. This standard of preparation integrity is more lenient than that typically used for studies of insect muscle (Stevenson and Josephson, 1990; Marden, 1995b) because our preparations showed a poor capacity to maintain their original tension production. However, shortening velocity was not affected by tetanic tension independently of temperature (P=0.27). We therefore assumed that force degradation was a result of decreasing neural or synaptic function, which resulted in submaximal recruitment of motor units, whereas crossbridge cycling of the active motor units was apparently not affected adversely. At the conclusion of experiments, cuticle and tissue surrounding the isolated muscle were dissected away, and the resting length of the muscle was measured to the nearest 0.1 mm using digital calipers. Muscle mass was measured to the nearest 0.1 mg using a Mettler balance. Cross-sectional area was estimated as the ratio of muscle mass to length.

At each experimental temperature, shortening velocity was measured from the first 1 ms of length change following the establishment of a stable tension during isotonic contraction (see Marden, 1995b, for details). Maximum shortening velocity (Vmax) was determined from a curve fitted to velocity as a function of relative tension using a hyperbolic–linear equation (Marsh and Bennett, 1986; performed iteratively using an Igor software routine supplied by R. L. Marsh). Estimates of shortening velocity (m s−1) and tension (N) at the tension yielding maximum power output (as determined by the hyperbolic–linear equation) were multiplied together and divided by fresh muscle mass (kg) in order to derive maximum instantaneous power output (W kg−1).

Measurements of twitch contraction kinetics were performed using the same arrangement as described above with a completely separate set of preparations wherein muscles were held at constant length and excited with single rather than multiple stimuli. These muscles were cooled to a temperature of approximately 16 °C, and the temperature was then raised slowly (approximately 0.5 °C min−1) until there was no longer a response to stimulation. Twitches were stimulated at least every 30 s and recorded at approximately 0.5 °C intervals. Twitch data were collected from five tenerals, three adolescents and seven matures.

For each twitch, the time to peak tension, the time to half-relaxation and the maximal force were determined. Temperatures were truncated to whole degrees, and data were pooled by age group within each temperature.

Skinned fiber experiments

Flight muscles from L. pulchella dragonflies were dissected and demembranated on ice for 1 h in an isotonic relaxing solution (20 mmol l−1 imidazole, 3 mmol l−1 MgATP, 5 mmol l−1 creatine phosphate and 5 mmol l−1 EGTA at pH 7, pCa 8 and 180 mmol l−1 total ionic strength; ionic strength adjusted using potassium methane sulfonate; Andrews et al. 1991) containing 0.5 % Triton X-100. The skinned fibers were used immediately or stored at −20 °C in 1:1 glycerol:relaxing solution for up to 3 weeks.

Solutions of similar ionic composition (pH 7 at 25 °C) without detergent were prepared at pCa 8, 7, 6.5, 6, 5.75, 5.5, 5.25, 5, 4.5 and 4, using the computer program BES_BATH (Godt and Lindley, 1982). As temperature was raised or lowered from 25 °C during experiments, the pH of these solutions varied according to the pKa of the imidazole buffer, thereby closely approximating natural temperature-induced changes in pH of the intracellular milieu (Somero, 1986). The pCa of these solutions also varied as temperature was raised or lowered from 25 °C, but this effect was small, changing only by a few hundredths of a unit at the most extreme temperatures.

A bundle of approximately 4–7 fibers was held at each end with a clamp made from a 23 gauge syringe needle and a 0.29 mm insect pin. The syringe needle was blunt-cut to a length of 1.5 cm, and the barrel was ground half-way through along 5 mm of its length, leaving a semi-circular trough at one end of the needle. One end of the muscle was laid in the trough and clamped by a length of pin which slid into the syringe needle beyond the trough and was glued in place with cyanoacrylate glue. This clamp is a modification of one described by Moss (1979). One clamp was mounted to a fixed point within a temperature-controlled chamber filled with relaxing solution. The other clamp, also within the chamber, was attached to the lever arm of a Cambridge Technology 300B lever system (which simultaneously measures both length and tension). The MacLab/8 converted force and position signals from the lever system into 12-bit digital form and sent them to a Macintosh Quadra 700, where they were monitored in real time with software emulating a chart recorder (Chart; AD Instruments) at a sampling frequency of 20 Hz.

A length–tension curve was generated for the fiber bundle in activating solution (pCa 5.5; this Ca2+ concentration was chosen because the muscle did not recover well from prolonged immersion in solutions of higher Ca2+ concentrations), and the muscle was fixed at a length just shorter than the length at which the tension started to decrease from its maximum. The length–tension curve for an individual fiber bundle was highly reproducible, provided the bundle was not stretched excessively beyond its point of maximum tension. This method fixes sarcomere length at maximum crossbridge overlap (Moss, 1979). Samples of teneral and mature fibers were fixed with glutaraldehyde while mounted at the point of maximal tension; when sectioned longitudinally and examined with a transmission electron microscope, they showed nearly uniform sarcomere lengths (2.2–2.4 μm).

As the length of the fiber bundle was held constant, its isometric tension was measured at various Ca2+ concentrations and temperatures. Data were first collected at 20 °C, and the fiber bundle was then subjected to temperatures of 15, 25, 30, 35 and 40 °C. Some fibers did not survive the entire range of temperatures (they no longer generated measurable tension changes between relaxation and full activation, or they tore). For some dragonflies, we tested more than one fiber in order to cover nearly all of the experimental temperatures. An average of 5.5 pCa50 estimates (described below) per individual dragonfly were obtained (S.D.=1.3; N=6 tenerals and N=7 matures). All activating solutions were used at each temperature; they were added in order of increasing Ca2+ concentration (a slightly different data set was generated if the solutions were added in reverse order).

The chamber and fiber were rinsed three times with each new activating solution prior to its final addition. Tension was allowed to equilibrate at each temperature and Ca2+ concentration prior to data collection.

For each skinned fiber preparation, relative tension at each point on a pCa series (the full complement of activating solutions run at a given temperature) was determined as the ratio of tension to the range of tension over that same pCa series. The pCa50 (the Ca2+ concentration at which the fiber bundle was half-maximally activated) was determined using an Igor software routine that performed an iterative fit of the Hill equation (Hill, 1910; Godt and Lindley, 1982; Andrews et al. 1991) to measurements of percentage maximal tension versus Ca2+ concentration.

SDS–PAGE gels and western blots

Flight muscle was dissected from the thorax of freshly killed L. pulchella dragonflies and demembranated on ice for 1 h in the same detergent solution as described above for preparing skinned fibers. This treatment removed most of the water-soluble and membrane-bound proteins, leaving a fraction enriched in contractile proteins. Approximately 5 mg of washed flight muscle was placed in 200 μl of SDS reducing buffer (62.5 mmol l−1 Tris, pH 6.8, 10 % glycerol, 2 % SDS, 0.5 % 2-β-mercaptoethanol and 0.0025 % Bromophenol Blue) and frozen until used.

Polyacrylamide gels (16 cm×16 cm; 0.75 mm thick) were cast and run using BioRad PROTEAN II xi electrophoresis equipment. The running gel contained 12 % polyacrylamide, 0.1 % SDS and 375 mmol l−1 Tris, pH 8.8, and was polymerized by the addition of 0.05 % ammonium persulfate and 0.03 % TEMED (N,N,N′,N′-Tetramethylethylenediamine) (Laemmli, 1970). The stacking gel contained 4 % polyacrylamide and 125 mmol l−1 Tris, pH 6.8. Frozen samples in reducing buffer were heated to 95 °C for 5 min and centrifuged at 14 000 revs min−1 for 10 min before use. A buffer volume containing soluble protein from 0.25 mg of flight muscle (approximately 10 μl) was loaded onto the gel. (Gels run with ground portions of the muscle in reducing buffer contained the same types and proportions of proteins but produced greater background staining.) The molecular mass standards used were Perfect Protein markers from Novagen, recombinant proteins of constant amino acid composition at specified molecular masses. Gels were run at 17 °C in a buffer of 0.1 % SDS, 0.025 mol l−1 Tris and 0.2 mol l−1 glycine. A constant (40 mA) current was applied using a BioRad model 200/2.0 power supply. The protocol and reagents supplied with the BioRad silver stain plus kit were used to silver stain the gels.

Proteins from gels were transferred to nitrocellulose using standard blotting methods (Towbin et al. 1979) with the BioRad trans-blot cell and the model 200/2.0 power supply. The blotting buffer contained 10 mmol l−1 Tris, pH 8.0, 150 mmol l−1 NaCl and 0.05 % Tween 20. Monoclonal antibodies to troponin T (MAC 145; Bullard et al. 1988), troponin C (MAC 352), troponin H and troponin I (MAC 143, which binds to both H and I; all antibodies provided by B. Bullard) were applied to the nitrocellulose after nonspecific blocking of the membrane using 3 % bovine serum albumin (BSA) and conjugated with a goat anti-rat alkaline phosphatase secondary antibody (Sigma). The nitrocellulose was then stained using the BioRad immuno-blot alkaline phosphatase assay kit.

Duplicate sets of lanes were run on each half of a single gel and blotted onto nitrocellulose. The nitrocellulose was then cut and one set of lanes was developed for the antibody, while the other set of lanes was stained with the BioRad colloidal gold total protein stain kit. This procedure provided confirmation of the precise location of the troponin bands on the gel.

Whole-muscle experiments

The contractile characteristics of L. pulchella whole-muscle preparations were similar to those observed in previous studies of synchronous insect flight muscle. Tetanic tension averaged 10–12 N cm−2, which is very similar to the range of values reported for katydid metathoracic flight muscles (Josephson, 1984) and flight muscles of geometrid moths (Marden, 1995b). If an adjustment is made for the 46 % of cross-sectional area in dragonfly muscle that is occupied by mitochondria (Marden, 1989), these tetanic tensions compare favorably with values obtained from vertebrate striated muscle (15–30 N cm−2; Close, 1972). Maximal shortening velocities of 8–10 muscle lengths s−1 (L s−1) are consistent with previous warm-temperature data from locusts and katydids (5–16 L s−1; Josephson, 1984) and moths (8–10 L s−1; Marden, 1995b). Thus, our dragonfly whole-muscle preparations appeared to be functionally competent, despite their somewhat accelerated rates of performance decay.

Maximum shortening velocity, tetanic tension and instantaneous power output during isotonic contraction did not vary as a function of age or age×temperature interaction (Fig. 1; Table 1). In contrast, twitch contraction kinetics varied significantly with age. Fig. 2 shows plots of time to peak tension (TTP; the interval between stimulus onset and peak tension) and time to half-relaxation (THR; the interval between peak tension and half-maximal tension during the relaxation phase) versus temperature for teneral and mature muscle. Mature muscle showed lower TTP and higher THR values at all but the very highest experimental temperatures, where teneral muscle was rapidly degenerating. A linear transformation of these data into logTTP and logTHR versus log(temperature) was made prior to statistical analysis. The linear range of the transformed data extended from 18 to 38 °C; data outside that range were excluded.

Table 1.

Analysis of variance table for the effects of age (teneral or mature) and temperature on maximum shortening velocity, tetanic tension and instantaneous power output adjusted for random variation between individual dragonflies

Analysis of variance table for the effects of age (teneral or mature) and temperature on maximum shortening velocity, tetanic tension and instantaneous power output adjusted for random variation between individual dragonflies
Analysis of variance table for the effects of age (teneral or mature) and temperature on maximum shortening velocity, tetanic tension and instantaneous power output adjusted for random variation between individual dragonflies
Fig. 1.

Upper plot: a representative series of traces showing muscle tension and length during a quick switch from isometric to isotonic contraction. Shortening velocity was measured immediately after establishment of a stable isotonic tension (for details, see Marden, 1995b). These traces were recorded at 28 °C from a teneral muscle that was 0.52 cm in length and weighed 7.4 mg. Lower plots: maximum shortening velocity (Vmax), tetanic tension and instantaneous power output during isotonic contraction for muscles from 12 tenerals and 10 matures. Individual preparations were not tested at all temperatures; plotted means (± 1 S.D.) are based on sample sizes ranging from 2 to 11.

Fig. 1.

Upper plot: a representative series of traces showing muscle tension and length during a quick switch from isometric to isotonic contraction. Shortening velocity was measured immediately after establishment of a stable isotonic tension (for details, see Marden, 1995b). These traces were recorded at 28 °C from a teneral muscle that was 0.52 cm in length and weighed 7.4 mg. Lower plots: maximum shortening velocity (Vmax), tetanic tension and instantaneous power output during isotonic contraction for muscles from 12 tenerals and 10 matures. Individual preparations were not tested at all temperatures; plotted means (± 1 S.D.) are based on sample sizes ranging from 2 to 11.

Fig. 2.

Twitch contraction kinetics of Libellula pulchella flight muscle, as characterized by the time to peak tension and time to half-relaxation. Graphs show means ± 1 S.D. for five tenerals and seven matures. Adolescent mean values were intermediate between teneral and mature mean values, but were omitted from the graph for clarity. Lower plots show raw twitch traces of L. pulchella flight muscle at low, mid-range and high thoracic temperatures. Each trace in a plot is from a different dragonfly.

Fig. 2.

Twitch contraction kinetics of Libellula pulchella flight muscle, as characterized by the time to peak tension and time to half-relaxation. Graphs show means ± 1 S.D. for five tenerals and seven matures. Adolescent mean values were intermediate between teneral and mature mean values, but were omitted from the graph for clarity. Lower plots show raw twitch traces of L. pulchella flight muscle at low, mid-range and high thoracic temperatures. Each trace in a plot is from a different dragonfly.

A split-plot analysis of variance (ANOVA), controlling for differences among individuals that were tested repeatedly over variable temperatures, was performed to determine whether logTTP is affected by age and temperature. A similar split-plot ANOVA (Table 2) was performed with the response variable logTHR. All of the effects tested were highly significant (P<0.01), making both age and temperature statistically important factors in determining twitch contraction kinetics of whole flight muscle.

Table 2.

Analysis of variance table for the effects of age (teneral, adolescent or mature) and temperature on logTTP and logTHR of flight muscle adjusted for random variation between individual dragonflies

Analysis of variance table for the effects of age (teneral, adolescent or mature) and temperature on logTTP and logTHR of flight muscle adjusted for random variation between individual dragonflies
Analysis of variance table for the effects of age (teneral, adolescent or mature) and temperature on logTTP and logTHR of flight muscle adjusted for random variation between individual dragonflies

Twitch force also varied with age and temperature. Fig. 2 shows representative traces of twitches from teneral and mature muscles at 18, 30 and 38 °C. Although twitch performance was highly variable (presumably due to variation in motor unit recruitment), general trends are evident. As dragonflies age, they are able to generate more twitch force per unit cross-sectional area of flight muscle. Likewise, warmer temperatures are associated with higher twitch forces. A split-plot analysis of variance (Table 3) was performed to determine whether age and temperature significantly affected twitch tension. Data generated above 42 °C were excluded from the statistical analysis because muscle preparations behaved erratically at high temperatures. Age, temperature and an age×temperature interaction (P<0.004 in each case) significantly affected twitch tension. Mature muscle is capable of generating approximately 75 % more twitch force than teneral muscle (least-squares age means ± 1 S.D. were 3.66±0.40, 2.75±0.45 and 2.04±0.40, respectively, for mature, adolescent and teneral muscle), with the difference increasing at high temperature, where twitch forces (Fig. 2) in the best mature muscle preparations approached tetanic tension (approximately 10–12 N cm−2; see Fig. 1).

Table 3.

Analysis of variance table for the effects of age (teneral, adolescent or mature) and temperature on force generated per unit cross-sectional area of flight muscle (N cm−2) adjusted for random variation between individual dragonflies

Analysis of variance table for the effects of age (teneral, adolescent or mature) and temperature on force generated per unit cross-sectional area of flight muscle (N cm−2) adjusted for random variation between individual dragonflies
Analysis of variance table for the effects of age (teneral, adolescent or mature) and temperature on force generated per unit cross-sectional area of flight muscle (N cm−2) adjusted for random variation between individual dragonflies

Skinned fiber experiments

A sigmoid curve of percentage maximal tension versus Ca2+ concentration (μmol l−1) was generated for each fiber at each experimental temperature. A representative series of curves from teneral and mature fibers at 35 °C is shown in Fig. 3B. At all experimental temperatures, fibers generated virtually no tension in the least concentrated Ca2+ solution (0.01 μmol l−1) and generated maximal tension at higher Ca2+ concentrations. As the temperature increased, fibers became activated at lower Ca2+ concentrations.

Fig. 3.

(A) Photograph of a skinned fiber bundle mounted between two clamps. (B) Pooled tension data from six mature and six teneral fiber bundles at 35 °C. (C) Thermal sensitivity of muscle activation, as revealed by a plot of pCa50 (negative logarithm of the Ca2+ concentration producing half-maximal tension) versus temperature for six tenerals and seven matures. Sample sizes vary among means as not all fibers were tested at all temperatures. In particular, only one mature fiber remained stable long enough to complete a pCa series at 40 °C. For both plots, data are means ± 1 S.D.

Fig. 3.

(A) Photograph of a skinned fiber bundle mounted between two clamps. (B) Pooled tension data from six mature and six teneral fiber bundles at 35 °C. (C) Thermal sensitivity of muscle activation, as revealed by a plot of pCa50 (negative logarithm of the Ca2+ concentration producing half-maximal tension) versus temperature for six tenerals and seven matures. Sample sizes vary among means as not all fibers were tested at all temperatures. In particular, only one mature fiber remained stable long enough to complete a pCa series at 40 °C. For both plots, data are means ± 1 S.D.

A split-plot analysis of variance, which controls for repeated measures from individual preparations (Table 4), was performed to test whether there was a difference in the Ca2+-sensitivity of skinned fibers from teneral and mature individuals, whether temperature affected Ca2+-sensitivity and whether the temperature-dependence of Ca2+-sensitivity (i.e. the slope of the regression line for pCa50versus temperature) differed for teneral and mature individuals. Both temperature and the temperature-dependence of Ca2+-sensitivity (expressed as age×temperature interaction) were statistically significant (P<0.0001) effects. Age did not have a significant effect independent of other factors in the model because of the large age×temperature interaction term.

Table 4.

Analysis of variance table for the effects of age (teneral or mature) and temperature on the pCa50 of skinned fibers adjusted for random variation between individual dragonflies

Analysis of variance table for the effects of age (teneral or mature) and temperature on the pCa50 of skinned fibers adjusted for random variation between individual dragonflies
Analysis of variance table for the effects of age (teneral or mature) and temperature on the pCa50 of skinned fibers adjusted for random variation between individual dragonflies

Slopes of plots of pCa50versus temperature, which quantify this temperature-dependence of Ca2+-sensitivity, were determined for each age group in the following manner. Linear regressions of pCa50 on temperature were performed for skinned fiber data from each individual, and the resultant linear equations were determined. The slopes of these equations were subjected to an analysis of variance with age as a factor (P=0.0006). This model reduced the data from each individual to a single independent data point. The least-square means of these slopes (±1 S.E.M.) were 0.057±006 and 0.020±005 pCa50 units °C−1 for mature and teneral muscle, respectively. The temperature-sensitivity of Ca2+ activation is therefore nearly three times greater in mature flight muscle.

There is a range of temperatures within which pCa50 values for teneral and mature skinned fibers are not significantly different from each other. The upper end of this range includes 15 °C, the lowest temperature tested. By 20 °C, the pCa50 values are significantly different (Fig. 3C; P<0.0062 for the split-plot ANOVA of skinned fiber data collected at 20 °C). The least-square mean pCa50 values generated by this model were 5.94 and 5.47 for mature and teneral muscle, respectively. This 0.47 unit difference in pCa50 corresponds to a threefold difference in Ca2+-sensitivity. By 35 °C, this difference in Ca2+-sensitivity between teneral and mature muscle rises to nearly 13-fold (Fig. 3C; mean pCa50 values of 6.74 and 5.63 for mature and teneral muscle, respectively, from the same split-plot ANOVA model).

SDS–PAGE gels and western blots

The maturational change in Ca2+-sensitivity of skinned fibers stimulated us to examine possible maturational changes in troponin expression. Flight muscle contractile proteins of teneral and mature L. pulchella dragonflies separated on one-dimensional SDS–PAGE gels (Fig. 4) showed a shift from a 45.7 kDa protein in teneral muscle to a protein of similar color and staining intensity at 44.6 kDa in mature muscle. Tenerals have a faint band at 44.6 kDa; adolescents (not shown) have approximately equal amounts of both bands. Western blots using monoclonal antibody identify these bands as troponin T (TnT; Fig. 4). Antibody staining of troponins H, I and C (98.2, 29.6 and 20.6 kDa respectively) showed no age-related differences on our one-dimensional gels (Fig. 4).

Fig. 4.

This silver-stained gel of flight muscle shows a shift in isoform expression of the Ca2+ regulatory protein troponin T between tenerals (T, lanes 1 and 2) and matures (M, lanes 3 and 4), which is confirmed by a western blot (teneral in lane 5; mature in lane 6). The bands corresponding to troponins H, I and C do not change in apparent molecular mass during adult maturation (teneral in lanes 7 and 9; mature in lanes 8 and 10). Molecular mass (in kDa) is shown on the left.

Fig. 4.

This silver-stained gel of flight muscle shows a shift in isoform expression of the Ca2+ regulatory protein troponin T between tenerals (T, lanes 1 and 2) and matures (M, lanes 3 and 4), which is confirmed by a western blot (teneral in lane 5; mature in lane 6). The bands corresponding to troponins H, I and C do not change in apparent molecular mass during adult maturation (teneral in lanes 7 and 9; mature in lanes 8 and 10). Molecular mass (in kDa) is shown on the left.

The absence of age-related differences in maximum shortening velocity, tetanic tension and instantaneous power output of L. pulchella flight muscle (Fig. 1) demonstrates that there are no significant age-related changes in crossbridge cycling kinetics. This result suggests that the maturational differences in muscle and organismal performance in this species (Marden, 1995a) do not involve changes in the way actin and myosin interact when the muscle is fully activated. However, dragonfly flight muscle is rarely fully activated; rather, it cycles rapidly between activated and relaxed states during each downstroke or upstroke of the wingbeat, usually stimulated by a single neural impulse (J. H. Marden, in preparation). Thus, Ca2+ is in a constant state of flux within these working muscles, and changes in excitation–contraction coupling could greatly affect muscle and organismal performance. Indeed, in L. pulchella dragonflies, a Ca2+ regulatory protein (TnT) undergoes a shift in isoform expression during adult maturation (Fig. 4), which apparently causes a dramatic (three-to 13-fold, depending on temperature) change in Ca2+-sensitivity of muscle activation between newly emerged and mature adults (Fig. 3C). Herein, as we detail below, lies the beginning of a mechanistic understanding of the dramatically increased flight performance and optimal body temperature of L. pulchella mature adults.

How do changes in Ca2+-sensitivity affect the function of a non-tetanizing muscle? This fundamental question is just beginning to be addressed by experimental research (Sweeney et al. 1993; Rome et al. 1996; Watkins et al. 1996); however, we can surmise the following. The mechanics of twitch contractions depend upon the time course of intracellular changes in Ca2+ concentration and how the muscle responds to those changes. Resting muscle has a very low Ca2+ concentration (10−8 mol l−1), which changes rapidly when a single neural impulse initiates Ca2+ release from the sarcoplasmic reticulum, thereby creating a wave of Ca2+ that spreads rapidly through the myoplasm, reaching peak concentrations of approximately 10−4 mol l−1. Resequestration of Ca2+ by soluble Ca2+-binding proteins and by active transport pumps in the membrane of the sarcoplasmic reticulum causes a rapid return of intracellular free Ca2+ concentration to resting levels. The rise and fall of intracellular Ca2+ concentration during a twitch occurs over a time scale of a few milliseconds (Tsugorka et al. 1995; Rome et al. 1996), during which the Ca2+ binding sites on troponin C do not become saturated (Cannell and Allen, 1984). Because transitions in intracellular Ca2+ concentration during a twitch are not instantaneous and do not saturate troponin C binding sites, changes in Ca2+-sensitivity of muscle activation and deactivation will necessarily bring about changes in the time course of twitch tension production. Interestingly, these effects should be more pronounced at high temperature (Moore et al. 1990; Sweeney et al. 1993). Thus, if we make the simplifying assumption that Ca2+ release and uptake kinetics in L. pulchella are age-invariant, mature muscle should show more rapid force generation in response to a Ca2+ transient, more delayed relaxation and greater twitch tension because the muscle is activated for longer (readers can visualize these effects by tracing the changes in percentage maximal tension of differently aged fibers as Ca2+ concentration rises and falls along the horizontal axis of the plot in Fig. 3B). These predictions accord with the observed maturational changes in twitch contraction kinetics of individual flight muscles (Fig. 2). Mature muscles reach peak tension more rapidly, relax more slowly and generate greater tension (P<0.001 for all of these comparisons). The exceedingly high Ca2+-sensitivity of mature muscle at temperatures above 35 °C (Fig. 3C) results in twitch tension reaching up to 90 % of tetanic tension (compare Figs 1 and 2), i.e. nearly the full tension-production capacity of the muscle.

The combination of a higher rate of force production during the rising phase of a twitch and a greater peak force (Fig. 2) should greatly enhance wing velocity and thus aerodynamic force production for mature dragonflies. We are presently initiating work-loop experiments (Josephson, 1985) and detailed measurements of wing kinematics in order to gain a fuller understanding of the relationship between variation in muscle activational biochemistry and mechanical performance.

The change in isoform expression of TnT provides a strong candidate for the molecular-level cause of age-related variation in Ca2+-sensitivity of L. pulchella flight muscle. Numerous studies of vertebrate striated muscle have found that developmental and pathological shifts in TnT isoform expression are correlated with changes in Ca2+-sensitivity of muscle activation (Schachat et al. 1987; Greaser et al. 1988; Nassar et al. 1991; Reiser et al. 1992; Gulati et al. 1994; Akella et al. 1995; Mesnard et al. 1995). Studies in which myofibrils were reconstituted with specific TnT isoforms (Tobacman and Lee, 1987; Wu et al. 1995) provide experimental proof of the role of TnT in determining Ca2+ sensitivity.

Troponin T is the most variable of all the troponins (and of all molecules outside the immune system; Andreadis et al. 1987), with at least 64 possible isoforms in mammals formed by alternative splicing of a gene that exists as one copy per haploid genome (Breitbart et al. 1985; Andreadis et al. 1987). Insects are also known to produce multiple splice variants of TnT from a single gene (Fyrberg et al. 1990); however, the developmental timing and tissue specificity have not yet been determined. The results of the present study suggest that TnT splicing may be both widespread in insects and important in terms of muscle and organismal function.

We are presently investigating the genetic composition and patterns of TnT isoform splicing in L. pulchella. Clones from a cDNA library made from RNA isolated from a mixture of teneral and mature L. pulchella flight muscle reveal a cassette of variably spliced, acidic exons near the amino terminal of the protein (M. Wolf and J. H. Marden, in preparation). This hypervariable region of TnT (Briggs et al. 1987) has been found to affect the cooperativity of muscle activation (Schaertl et al. 1995). In vertebrate striated muscle, mature isoforms of TnT have more of the alternative exons spliced out of the pre-mRNA (Jin et al. 1992) via a mechanism that predates the evolutionary divergence of vertebrates and invertebrates (Nadal-Ginard et al. 1991). Our observations of a decrease in molecular mass of TnT during adult maturation in dragonflies (Fig. 4) and a correlated change in Ca2+-sensitivity accord with this pattern. Thus, our present working hypothesis is that alternative, acidic exons near the amino terminus affect the charge structure of TnT, thereby altering molecular cooperativity among neighboring troponin units, which in turn alters the Ca2+-sensitivity of muscle activation, muscle twitch contraction kinetics and, ultimately, whole-organism performance.

Although we have demonstrated that the maturational differences in twitch contraction kinetics are consistent with changing Ca2+-sensitivity, it is not certain that increased Ca2+-sensitivity is solely responsible for these results. There could be other factors contributing to the age-related change in TTP and THR values, foremost of which is a change in the rapidity of Ca2+ release and re-sequestration by the sarcoplasmic reticulum. Age-related differences in Ca2+-sensitivity of skinned fibers are independent of these factors, however, since skinned fibers have no functional sarcoplasmic reticulum. As a result, the observed changes in whole-muscle twitch contraction kinetics must be due at least in part to the effects of the maturational changes in Ca2+-sensitivity (Fig. 3). Although we have not yet performed a detailed study, cross sections of L. pulchella flight muscle do not show a marked maturational change in the relative amount of SR or in the distance between the SR and the myofilaments.

The results of this study suggest a chain of functionally interrelated maturational changes, beginning with mRNA splice variation through subcellular, whole-muscle and organismal levels of function. Although we have demonstrated possible avenues of functional linkage between these levels, we have not yet experimentally verified the hypothesized cause/effect relationships. Establishing such linkages is an important quest for, although many previous studies have shown similar phenomena at the subcellular level (i.e. in skinned fibers), there has been little effort to trace the consequences of isoform variation in Ca2+ regulatory proteins to higher levels of tissue and organismal performance. Indeed, a recent review article (Malhotra, 1994) regarding the role of thin filament proteins in pathological states laments the absence of direct links between molecular differences in Ca2+ regulatory proteins and their physiological and pathological function. By developing an organismal model for assessing the multi-level functional effects of variable TnT expression, we may be able to contribute in important ways to achieving an integrated understanding of muscle biochemistry and function.

We thank B. Bullard for providing monoclonal antibodies, D. Maughan and J. Vigoreaux for early tips on electrophoresis, R. Cyr for teaching us western blotting, M. Wolf for access to preliminary sequences of L. pulchella troponin cDNAs, and R. Ordway, J. Day, and two anonymous reviewers for providing helpful criticisms of the manuscript. This research was supported by NSF grants IBN-9317969 and IBN-9600840 to J.H.M.

Akella
,
A. B.
,
Ding
,
X.-L.
,
Cheng
,
R.
and
Gulati
,
J.
(
1995
).
Diminished calcium sensitivity of skinned cardiac muscle contractility coincident with troponin T-band shifts in the diabetic rat
.
Circulation Res.
76
,
600
606
.
Andreadis
,
A.
,
Gallego
,
M.E.
and
Nadal-Ginard
,
B.
(
1987
).
Generation of protein isoform diversity by alternative splicing: mechanistic and biological implications
.
A. Rev. Cell Biol
.
3
,
207
242
.
Andrews
,
M. A.
,
Maughan
,
D. W.
,
Nosek
,
T. M.
and
Godt
,
R. E.
(
1991
).
Ion-specific and general ionic effects on contraction of skinned fast-twitch skeletal muscle from the rabbit
.
J. gen. Physiol.
98
,
1105
1125
.
Breitbart
,
R. E.
,
Nguyen
,
H. T.
,
Medford
,
R .M.
,
Destree
,
A. T.
,
Mahdavi
,
V.
and
Nadal-Ginard
,
B.
(
1985
).
Intricate combinatorial patterns of exon splicing generate multiple regulated troponin T isoforms from a single gene
.
Cell
41
,
67
82
.
Briggs
,
M. M.
,
Lin
,
J. J.-C.
and
Schachat
,
F. H.
(
1987
).
The extent of amino-terminal heterogeneity in rabbit fast skeletal muscle troponin T
.
J. Muscle Res. Cell Motil.
8
,
1
12
.
Bullard
,
B.
,
Leonard
,
K.
,
Larkins
,
A.
,
Butcher
,
G.
,
Karlik
,
C.
and
Fyrberg
,
E.
(
1988
).
Troponin of asynchronous flight muscle
.
J. molec. Biol.
204
,
621
637
.
Cannell
,
M. B.
and
Allen
,
D G.
(
1984
).
Model of calcium movements during activation in the sarcomere of frog skeletal muscle
.
Biophys. J.
45
,
913
925
.
Close
,
R. I.
(
1972
).
Dynamic properties of mammalian skeletal muscle
.
Physiol. Rev
.
52
,
129
197
.
Fyrberg
,
E.
,
Fyrberg
,
C. C.
,
Beall
,
C.
and
Saville
,
D. L.
(
1990
).
Drosophila melanogaster troponin-T mutations engender three distinct syndromes of myofibrillar abnormalities
.
J. molec. Biol.
216
,
657
675
.
Godt
,
R. E.
and
Lindley
,
B. D.
(
1982
).
Influence of temperature upon contractile activation and isometric force production in mechanically skinned muscle fibers of the frog
.
J. gen. Physiol.
80
,
279
297
.
Greaser
,
M. L.
,
Moss
,
R. L.
and
Reiser
,
P. J.
(
1988
).
Variation in contractile properties of rabbit single muscle fibres in relation to troponin T isoforms and myosin light chains
.
J. Physiol., Lond.
406
,
85
98
.
Gulati
,
J.
,
Akella
,
A.B.
,
Nikolic
,
S. D.
,
Starc
,
V.
and
Siri
,
F.
(
1994
).
Shifts in contractile regulatory protein subunits troponin T and troponin I in cardiac hypertrophy
.
Biochem. biophys. Res. Commun.
202
,
384
390
.
Hill
,
A. V.
(
1910
).
The possible effects of the aggregation of the molecules of haemoglobin on its dissociation curve
.
J. Physiol., Lond.
40
,
4P
7P
.
Jin
,
J.-P.
,
Huang
,
Q. Q.
,
Yeh
,
H. I.
and
Lin
,
J. J.
(
1992
).
Complete nucleotide sequence and structural organization of rat cardiac troponin T gene: a single gene generates embryonic and adult isoforms via developmentally regulated alternative splicing
.
J. molec. Biol
.
227
,
269
276
.
Josephson
,
R. K.
(
1984
).
Contraction dynamics of flight and stridulatory muscles of tettigoniid insects
.
J. exp. Biol
.
108
,
77
96
.
Josephson
,
R. K.
(
1985
).
Mechanical power output from striated muscle during cyclic contraction
.
J. exp. Biol
.
114
,
493
512
.
Laemmli
,
U. K.
(
1970
).
Cleavage of structural proteins during the assembly of the head of bacteriophage T4
.
Nature
227
,
680
685
.
Malhotra
,
A.
(
1994
).
Role of regulatory proteins (troponin–tropomyosin) in pathologic states
.
Molec. cell. Biochem.
135
,
43
50
.
Marden
,
J. H.
(
1989
).
Bodybuilding dragonflies: costs and benefits of maximizing flight muscle
.
Physiol. Zool.
62
,
505
521
.
Marden
,
J. H.
(
1995a
).
Large-scale changes in thermal sensitivity of flight performance during adult maturation in a dragonfly
.
J. exp. Biol.
198
,
2087
2094
.
Marden
,
J. H.
(
1995b
).
Evolutionary adaptation of contractile performance in muscle of ectothermic winter-flying moths
.
J. exp. Biol.
198
,
2095
2102
.
Marden
,
J. H.
,
Kramer
,
M. G.
and
Frisch
,
J.
(
1996
).
Age-related variation in body temperature, thermoregulation and activity in a thermally polymorphic dragonfly
.
J. exp. Biol.
199
,
529
535
.
Marsh
,
R. L.
and
Bennett
,
A. F.
(
1986
).
Thermal dependence of contractile properties of skeletal muscle from the lizard Sceloporus occidentalis with comments on methods for fitting and comparing force–velocity curves
.
J. exp. Biol.
126
,
63
77
.
Mesnard
,
L.
,
Logeart
,
D.
,
Taviaux
,
S.
,
Diriong
,
S.
,
Mercadier
,
J.-J.
and
Samson
,
F.
(
1995
).
Human cardiac troponin T: cloning and expression of new isoforms in the normal and failing heart
.
Circulation Res.
76
,
687
692
.
Moore
,
R. L.
,
Palmer
,
B. M.
,
Williams
,
S. L.
,
Tanabe
,
H.
,
Grange
,
R. W.
and
Houston
,
M. E.
(
1990
).
Effect of temperature on myosin phophorylation in mouse skeletal muscle
.
Am. J. Physiol
.
259
,
C432
C471
.
Moss
,
R. L.
(
1979
).
Sarcomere length–tension relations of frog skinned muscle fibres during calcium activation at short lengths
.
J. Physiol., Lond.
292
,
177
192
.
Nadal-Ginard
,
B.
,
Smith
,
C. W.
,
Patton
,
J. G.
and
Breitbart
,
R. E.
(
1991
).
Alternative splicing is an efficient mechanism for the generation of protein diversity: contractile protein genes as a model system
.
Adv. Enzyme Regul.
31
,
261
286
.
Nassar
,
R.
,
Malouf
,
N. N.
,
Kelly
,
M. B.
,
Oakeley
,
A. E.
and
Anderson
,
P. A. W.
(
1991
).
Force-pCa relation and troponin T isoforms of rabbit myocardium
.
Circulation Res.
69
,
1470
1475
.
Reiser
,
P. J.
,
Greaser
,
M. L.
and
Moss
,
R. L.
(
1992
).
Developmental changes in troponin T isoform expression and tension production in chicken single skeletal muscle fibres
.
J. Physiol., Lond.
449
,
573
588
.
Rome
,
L. C.
,
Syme
,
D. A.
,
Hollingworth
,
S.
,
Lindstedt
,
S. L.
and
Baylor
,
S. M.
(
1996
).
The whistle and the rattle: the design of sound producing muscles
.
Proc. natn. Acad. Sci. U.S.A.
92
,
8095
8100
.
Schachat
,
F. H.
,
Diamond
,
M. S.
and
Brandt
,
P. W.
(
1987
).
Effect of different troponin T–tropomyosin combinations on thin filament activation
.
J. molec. Biol.
198
,
551
555
.
Schaertl
,
S.
,
Lehrer
,
S. S.
and
Geeves
,
M. A.
(
1995
).
Separation and characterization of the two functional regions of troponin involved in muscle thin filament regulation
.
Biochemistry
34
,
15890
15894
.
Somero
,
G. N.
(
1986
).
Protons, osmolytes and fitness of internal milieu for protein function
.
Am. J. Physiol.
251
,
R197
R213
.
Stevenson
,
R. D.
and
Josephson
,
R. K.
(
1990
).
Effects of operating frequency and temperature on mechanical power output from moth flight muscle
.
J. exp. Biol.
149
,
61
78
.
Sweeney
,
H. L.
,
Bowman
,
B. F.
and
Stull
,
J. T.
(
1993
).
Myosin light chain phophorylation in vertebrate striated muscle: regulation and function
.
Am. J. Physiol.
264
,
C1085
C1095
.
Tobacman
,
L. S.
and
Lee
,
R.
(
1987
).
Isolation and functional comparison of bovine cardiac troponin T isoforms
.
J. biol. Chem.
262
,
4059
4064
.
Towbin
,
H.
,
Straehlin
,
T.
and
Gordon
,
J.
(
1979
).
Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications
.
Proc. natn. Acad. Sci. U.S.A.
76
,
4350
4354
.
Tsugorka
,
A.
,
Rios
,
E.
and
Blatter
,
L. A.
(
1995
).
Imaging elementary events of calcium release in skeletal muscle cells
.
Science
269
,
1723
1726
.
Watkins
,
H.
,
Seidman
,
C. E.
,
Seidman
,
J. G.
,
Feng
,
H. S.
and
Sweeney
,
H. L.
(
1996
).
Expression and functional assessment of a truncated cardiac troponin T that causes hypertrophic cardiomyopathy
.
J. Clin. Invest.
98
,
2456
2461
.
Wu
,
Q.-L.
,
Jha
,
P. K.
,
Du
,
Y.
,
Laevis
,
P. C.
and
Sarkar
,
S.
(
1995
).
Overproduction and rapid purification of human fast skeletal beta troponin T using Escherichia coli expression vectors: functional differences between the alpha and beta isoforms
.
Gene
155
,
225
230
.