Comparison of Volatile Anesthetic Effects on Actin–Myosin Cross-bridge Cycling in Neonatal versus  Adult Cardiac Muscle

All experimental procedures described in this article were reviewed and approved by the Institutional Animal Care and Use Committee of the Mayo Clinic. Six male Sprague-Dawley rats (Harlan Inc., Madison, WI) each from three different litters (18 rats total) were used within 3 days of parturition and comprised the neonate group (body weight, 8–10 g at birth). These neonates were gender delineated based on the distance between urethra and anus. Eighteen adult male Sprague-Dawley rats from a different litter comprised the adult group (at least 84 days old; body weight, 250–275 g). Six animals at each age were used for the halothane group, and six of the remaining animals were used for the sevoflurane group. The final six animals were used to estimate myofibrillar density of cardiac muscle.

Animals were anesthetized using intramuscular ketamine (60 mg/kg) and xylazine (2.5 mg/kg). The heart was quickly excised, cleaned of blood, and placed in a balanced salt solution (130 mM NaCl, 3 mM KCl, 1.2 mM KH²PO⁴, 1.0 mM MgSO⁴, 1.24 mM CaCl², 10 mM HEPES, 10 mM glucose, pH 7.2) vigorously bubbled with 100% O². During continued oxygenation, the ventricles were carefully dissected out, pinned in a Sylgard-coated (Dow Corning Corp., Midland, MI) Petri dish, and cut into thin strips (1–2 mm wide).

Muscle strips were stretched and pinned on cork at an estimated optimal fiber length (L^o) and flash-frozen in melting isopentane cooled by liquid nitrogen. Cross-sections were cut at 10 μm using a cryostat (Reichert-Jung model 2000E; Cambridge Instruments Inc., Buffalo, NY; −20°C), placed on glass slides, and incubated for 2 h in a primary antibody directed against cardiac myosin (antimyosin rabbit immunoglobulin G; Sigma Immunochemicals, St. Louis, MO; 1:50 dilution in 0.1 M phosphate buffer). Control sections were incubated only in buffer. All sections were then washed with phosphate buffer and incubated further for 1 h in a rhodamine-tagged secondary antibody (donkey antirabbit IgG; Jackson Immunoresearch, West Grove, PA). Sections were visualized using confocal microscopy (model MRC500, BioRad Laboratories, Hercules, CA). A fluorescence intensity threshold was established based on the residual fluorescence of the control sections. Pixels with intensities greater than threshold were considered to represent areas of myosin staining. The cross-sectional area of individual myocytes was circumscribed using manufacturer-supplied software. The proportion of pixels above threshold was then used as an index of myofibrillar density.

For each ventricular strip, aluminum foil T clips were attached to the ends to minimize compliance. These muscle strips were then skinned for 20 min with 1 mg/ml Triton X-100 and placed between two small stainless-steel hooks in a temperature-controlled, flow-through, 120-μl-volume acrylic chamber located on the stage of an inverted microscope (Olympus IMT-2; Olympus America Inc., Melville, NY). Extreme caution was used to ensure that the hooks inserted only into the aluminum clips and thus the fixed ends of the muscle. One hook was attached to a force transducer (AE-801; Aksjeselskapet, Horten, Norway) with a resonant frequency of 5 kHz, and the other hook was attached to a servo motor (G120DT; General Scanning, Watertown, MA) with a step time of 800 μs. Sarcomere length was adjusted to 2.2 μm based on calibrated microscopy of the sarcomeric pattern. Custom-built software (based on LabView from National Instruments, Austin, TX) and a data acquisition board (AT-MIO16-L9; National Instruments) were used to control the servo system and acquire length and force signals. Data were acquired at a frequency of 10 kHz.

The Triton X-100–skinned ventricular strips were perfused with solutions of different pCa (negative logarithm of the Ca²⁺concentration) prepared in accordance with the procedure described by Fabiato and Fabiato 14with stability constants listed by Godt and Lindley. 15The solutions contained the following: 10.0 mM EGTA, 1.0 mM free Mg²⁺, 5.0 mg adenosine triphosphate (ATP), 15.0 mM creatine phosphate, 50.0 mM imidazole, 2.0 mM dithiothreitol, and 1 mg/ml creatine phosphokinase with a total ionic strength of 150 mM. Strips were initially perfused with a relaxing solution of pCa 9.0. In preliminary studies, strips were sequentially exposed to different lower pCa (higher Ca²⁺) for 30 s to determine the level for maximum activation and to assess rundown of tissue. Maximal activation of the muscle was obtained at a pCa of 4.0. Exposure to pCa beyond 5.5 for greater than 45 s resulted in significant rundown of the tissue. However, exposure for 30 s allowed attainment of a stable plateau force for a particular pCa and subsequent determination of cross-bridge cycling parameters in the presence or absence of halothane.

Sarcomere lengths during relaxation and activation were maintained according to the techniques of Brenner 16as modified by Sweeney et al.  17This procedure essentially involved a repetitive sequence of rapid slackening of muscle length such that force instantaneously dropped to zero, followed by rapid restretch of the muscle to the original length. This cycle was repeated every 5 s throughout the course of the experiment. Sarcomere length was measured using video microscopy and was verified to be consistent. The average sarcomere length was found to be 2.2 ± 0.6 μm.

At the end of each experimental protocol, the diameter and depth of the muscle bundle was measured in vitro  using a calibrated microscope. The cross-sectional area of the bundle was then estimated as the area of an ellipse with diameters equal to the width and depth.

Halothane (Wyeth-Ayerst Laboratories, Philadelphia, PA) and sevoflurane (Abbott Laboratories, Deerfield, IL) were added to the aerating gas mixture via  a calibrated on-line vaporizer. The vaporizer was set to produce aqueous concentrations of anesthetic in the pCa equivalent to 1 and 2 adult rat MAC at room temperature, concentrations that are well within the range of clinical usage. All experiments were performed at 25°C. Given the lack of previously published MAC values for neonatal rats and the need to have a basis for comparison of effects at equimolar anesthetic concentrations, the adult rat MAC values were also used in the neonates. Aqueous concentrations of halothane in the perfusion chamber (1 atm) were determined by gas chromatography from anaerobically obtained samples using an electron capture detector (Hewlett-Packard 5880A; Palo Alto, CA). 18In the solutions used to examine the effects of halothane, concentrations were 0.22 ± 0.05 mM for 1 MAC and 0.48 ± 0.07 mM for 2 MAC. Sevoflurane concentrations in the perfusion chamber were determined using a flame ionization detector, and the concentrations were found to be 0.40 ± 0.12 mM and 0.64 ± 0.16 mM for 1 and 2 MAC, respectively.

A number of theoretical models have been developed to explain cross-bridge cycling in striated muscle. Most of these have limitations but are convenient frameworks for understanding the effects of various perturbations on cross-bridge cycling. In the current study, we used the two-state model developed by Huxley as a convenient model to examine the effects of volatile anesthetics. According to this model, cross-bridges are either in a force-generating state (myosin attached to actin) or a non–force-generating state (myosin detached). 19,20Contraction then involves a cyclical attachment and detachment of myosin from actin (cross-bridge cycling) and basically involves: (1) binding of ATP to the actin–myosin complex (force generating state) resulting in dissociation of the complex, (2) hydrolysis of ATP by the myosin head into adenosine 5′-diphosphate (ADP) and phosphate, (3) reattachment of actin to the myosin–ADP–phosphate complex, and (4) liberation of ADP and phosphate accompanied by a power stroke resulting in the production of force by the single cross-bridge. Due to the cyclical nature of cross-bridge attachment and detachment and a lack of synchronicity in the actin–myosin interactions across the length and width of the muscle, the steady state force produced by the muscle is proportional to the steady state fraction of cross-bridges in the force-generating state (α^fs). Accordingly, relaxation involves a reduction in α^fs. Therefore,

where n is the maximum number of cross-bridges available in parallel per half sarcomere and ¯F is the mean force produced per cross-bridge.

Previous studies have reported that changes in steady state force closely approximate changes in steady state stiffness (i.e. , the change in force produced by oscillatory length perturbations of extremely small amplitude [< 1% of optimal muscle length L^o], normalized for the length change). 21,22Oscillations are typically imposed at ∼1 kHz in both skeletal and cardiac muscle. In this technique, the length perturbations are small and fast enough as to not actually disrupt cross-bridge binding, but the incremental force required of each cross-bridge to oppose the length perturbation is a reflection of the number of attached cross-bridges. Accordingly,

where ¯S is the mean stiffness per cross-bridge.

As described earlier, ATP is essential for cross-bridge cycling to continue. During conditions of rigor, in which ATP is unavailable for cross-bridge detachment, α^fs= 1. Therefore, the α^fsduring other conditions, such as maximal Ca²⁺activation (with ATP present), can be estimated by taking the ratio of stiffness during that condition to that during rigor, thus eliminating the unknowns n and ¯S in . To determine the effects of volatile anesthetics on α^fs, stiffness measurements could then be made during conditions of maximal Ca²⁺activation and rigor, both in the presence of anesthetic.

Using the techniques employed in the current study, it is not possible to directly determine ¯F or ¯S. However, we attempted to determine the relative change in these parameters during control versus  anesthetic conditions. The total number of cross-bridges (n) is determined by contractile protein content and is unlikely to be altered by volatile anesthetics in the short time frame of their action. Accordingly, as α^fs= 1 during rigor regardless of the presence or absence of anesthetic, comparison of isometric force or stiffness during rigor in the presence and absence of anesthetic should indicate whether volatile anesthetics affect the mean force stiffness per cross-bridge. Accordingly,

where n¯F^Cand n¯F^Aare the mean forces per cross-bridge during control and anesthetic conditions, respectively. However, because both variable are unknown, only the relative change in mean force per cross-bridge can be determined from the measurement of isometric forces as

Similarly, for stiffness,

Isometric force and stiffness were measured in separate Triton-X–skinned strips by exposure to pCa 9.0 (relaxation) or pCa 4.0 (maximum activation). In the first half of each experiment, strips were first placed in high Ca²⁺rigor (pCa 4.0), relaxed with pCa 9.0, and then activated with pCa 4.0. Control values of different cross-bridge cycling parameters were obtained in this phase. The strips were then relaxed with pCa 9.0, preexposed to a single anesthetic concentration (either halothane or sevoflurane; separate sets of studies for the two anesthetics), and taken through the same protocol as the control in the continued presence of anesthetic. Thus, each strip served as its own control for either anesthetic. Rundown of the tissue preparation was estimated by exposing a separate set of strips to two pairs of rigor and pCa 4.0 activations.

Stiffness was calculated from the changes in force produced by 1 kHz sinusoidal oscillations of extremely small amplitude (< 1% L^o). Three sets of oscillations were imposed for 100 ms (100 cycles) each, with intervening periods of 3 s. Force and length data were acquired using LabView and averaged across 5 cycles in the middle of each oscillation period, and stiffness was calculated. The average stiffness across three periods was then calculated. The effect of volatile anesthetics on ¯F and ¯S were estimated by measurements during rigor in the presence or absence of 1 or 2 MAC anesthetic, as explained earlier. Separate strips were also used for each anesthetic concentration.

In a single strip, the α^fswas measured in the presence and absence of 1 or 2 MAC halothane or sevoflurane. Separate strips were used for the two anesthetics and the two concentrations. Stiffness measurements in pCa 4.0 versus  rigor were performed as described earlier.

By extending the Huxley model, two apparent rate constants can been used to describe cyclical cross-bridge kinetics 23: The rate of cross-bridge attachment (f^app), and the rate of cross-bridge detachment (g^app). These parameters are related by the expression

If an activated muscle at optimal length (L^o) is rapidly released to a shorter length (e.g. ,∼ 80–85% L^o) and then rapidly restretched back to L^o, all cross-bridges are initially broken, and force decreases abruptly to zero. As cross-bridges reform, the rate constant for force redevelopment (k^tr) is given by:

Therefore,

is valid at all values of α^fsexcept during conditions of rigor, where α^fs= 1 but k^tr= 0 and g^app= 0 (i.e. , cross-bridges do not detach). By measuring k^trand estimating α^fsfrom stiffness, the value of f^appand g^appcan be derived.

In strips where α^fswas determined, k^trwas also determined in the presence or absence of 1 or 2 MAC halothane or sevoflurane. During pCa 4.0 activation, force was allowed to stabilize at its maximum value at a muscle length of L^o. Muscle length was then rapidly dropped by 20%, held at the new length for 100 ms, and then rapidly stretched back to L^o. Decrease of the force response to zero after slackening was verified post hoc.  A curve-fitting program was used to measure k^tr. Separate strips were used for each anesthetic and concentration.

Quality control of experimental data was maintained by excluding samples that displayed greater than 10% rundown in force over a 45-s period during maximum activation. All data are expressed as means ± SD. Cross-bridge cycling parameters were compared between age groups and a single anesthetic using unpaired Student t  tests. Comparisons across age groups and anesthetics were performed using two-way analysis of variance (ANOVA) with Bonferroni corrections for multiple comparisons. All P  values were two-sided, and P < 0.05 was considered significant.

There were no significant differences between the control groups for halothane versus  sevoflurane in any of the measured or calculated parameters. Accordingly, the results of the control group were pooled in the creation of figures and tables. However, it must be emphasized that separate statistical analyses were performed for the two anesthetics using only their corresponding controls.

In both neonate and adult cardiac muscle bundles, exposure to pCa 4.0 resulted in maximal activation and a stable force response (figs. 1 and 2). In control neonatal cardiac muscle bundles, (i.e. , those not exposed to anesthetic) maximum force per cross-sectional area was ∼45% lower compared with the adult muscle bundles (P < 0.05; n = 12;fig. 3A). However, myofibrillar density, estimated using myosin staining, was 52 ± 13% in neonates and 67 ± 12% in adults (n = 6). Therefore, even when corrected for myofibrillar density (by dividing specific force by the density fraction), maximum force per cross-sectional area was still smaller in neonates by ∼30%.

Exposure to volatile anesthetics resulted in a decrease in force, with the effect being significantly more pronounced in neonates compared with adults, for both 1 and 2 MAC of either anesthetic (representative tracings in fig. 1 [halothane]and fig. 2 [sevoflurane]). Exposure to either 1 or 2 MAC anesthetic resulted in a decrease in the maximum force per cross-sectional area in both neonates and adults (figs. 3A and 3Bfor halothane and sevoflurane, respectively). At either anesthetic concentration, the effect on maximum force was significantly greater in neonates than in adults, with the effects of halothane being significantly more pronounced compared with sevoflurane (figs. 3C and 3Dfor halothane and sevoflurane, respectively; n = 6 for all groups;P < 0.05 for all comparisons). Furthermore, at any age, the effect of 2 MAC anesthetic was greater than that of 1 MAC. Overall, both halothane and sevoflurane decreased the maximum force per cross-sectional area to a greater extent in neonates, with halothane having a greater effect compared to sevoflurane.

In both neonate and adult cardiac muscle bundles, exposure to pCa 4.0 in the absence of ATP produced a rigor force response (figs. 1 and 2). In strips that were exposed to two pairs of the activating solutions (thus serving as time controls for anesthetic exposure), the rundown in force production at pCa 4.0 was found to be ∼6 and ∼7% in neonates and adults respectively. Rundown with rigor solution was found to be ∼7 and ∼8% in neonates and adults, respectively. Exposure to either 1 or 2 MAC halothane during rigor conditions also decreased the force produced. In contrast, 1 MAC sevoflurane had no significant effect on the rigor force, whereas 2 MAC sevoflurane did decrease rigor force, but to a significantly smaller effect, compared with halothane (n = 6 for all groups;P < 0.05 for anesthetic comparisons).

Similar to maximum force, maximum stiffness corrected for cross-sectional area (i.e. , elastic modulus) was ∼50% lower in control neonatal cardiac muscle compared with adult cardiac muscle. Exposure to 1 or 2 MAC halothane decreased the stiffness in muscle strips from both ages (fig. 4A), and the effect was greater in neonates and at 2 MAC (fig. 4B; n = 6 for all groups;P < 0.05). In comparison, 1 MAC sevoflurane had no significant effect on stiffness, whereas 2 MAC sevoflurane produced a small but significant effect in both neonates and adults that was considerably smaller than that produced by halothane (figs. 4C and 4D; n = 6 for groups;P < 0.05 for anesthetic comparisons).

The ratio of F^Ato F^C(mean force per cross-bridge during anesthetic and control conditions, respectively) was significantly lower than 1.0 (i.e. , decreased) in both neonates and adults after exposure to either 1 or 2 MAC halothane (P < 0.05; n = 6 for all groups;fig. 5A); this effect was greater at 2 MAC in neonates only. At 1 MAC, the effect on this ratio was significantly greater in adults compared with neonates (P < 0.05). However, at 2 MAC, this ratio was decreased to a greater extent in neonates compared with adults. In comparison, both 1 and 2 MAC sevoflurane had no significant effect on the ratio of F^Ato F^Cin neonates and only a small effect in adults at 2 MAC (fig. 5B). In time controls, the ratio of F^Ato F^Cwas found to be 0.92 ± 0.08 in neonates and 0.95 ± 0.10 in adults (n = 6 for all groups).

The ratio of S^Hto S^C(mean stiffness per cross-bridge during anesthetic and control conditions, respectively) was also significantly decreased in both neonates and adults by exposure to halothane (P < 0.05; n = 6 for all groups;fig. 5C). This effect was actually more pronounced in adults for 1 MAC halothane and comparable between ages for 2 MAC halothane. In comparison, both 1 and 2 MAC sevoflurane had no significant effect on the ratio of S^Hto S^Cin neonates but did have a small but significant effect at 2 MAC sevoflurane in adults, although considerably less than that with 2 MAC halothane (fig. 5D). In time controls, the ratio of S^Hto S^Cwas found to be 0.94 ± 0.05 in neonates and 0.94 ± 0.10 in adults.

In control tissue, the fraction of strongly bound cross-bridges (α^fs) was significantly smaller (by ∼20%) in neonates compared with adults (table 1). Exposure to 1 MAC halothane had no significant effect on α^fsin either neonates or adults, whereas 2 MAC halothane significantly decreased α^fsin both age groups (P < 0.05) but to a greater extent in neonates. In contrast, only 2 MAC sevoflurane had significant effects on α^fsand then only in neonates (table 1). These effects on neonates were comparable to those induced by 2 MAC halothane.

The rate of force redevelopment after rapid relaxation and restretch (k^tr) was significantly lower (by ∼30%) in neonates compared with adults (fig. 6and table 1). Exposure to both 1 and 2 MAC halothane significantly decreased k^tr(P < 0.05) in both groups but to a greater extent in neonates. Exposure to sevoflurane also resulted in a significant decrease in k^tr(table 1). At 1 MAC, the effects of sevoflurane were comparable between neonates and adults and both were considerably less than the effects of 1 MAC halothane (P < 0.05). At 2 MAC, sevoflurane continued to have a smaller effect on k^trin neonates compared with halothane. However, the effect in adults was greater compared with halothane.

From the control values of k^trand α^fs, the control rates of f^appand g^appwere calculated. The f^appwas significantly smaller by ∼50% in neonates compared with adults (table 1). However, the g^appwas not significantly different between neonates and adults. Exposure to 1 MAC halothane resulted in a comparable decrease in f^appin both neonates and adults, whereas g^appwas unaffected. However, 2 MAC halothane significantly decreased f^app(P < 0.05;table 1) and increased g^app(P < 0.05) to a greater extent in neonates. In comparison with halothane, 1 MAC sevoflurane produced smaller decreases in f^appin both neonates and adults; these effects were comparable between the two ages (table 1). Exposure to 2 MAC sevoflurane produced a greater decline in f^app, but only in adults. Exposure to 1 or 2 MAC sevoflurane had no significant effect on g^appin either neonates or adults.

Clinically relevant concentrations of halothane and sevoflurane generally affect force production to a greater extent in neonatal cardiac muscle compared with the adult. The depressant effects of sevoflurane on cardiac muscle were found to be generally smaller than those induced by halothane. For both anesthetics, these effects appear to be mediated, at least in part, by a direct inhibition of cross-bridge cycling, especially the rates of cross-bridge attachment and detachment. Since MAC multiples are generally higher in neonates, extrapolation of the current results using equivalent concentrations to MAC values would only suggest even greater actual myocardial depression in this age group. The results further suggest anesthetic differences in both potency and intracellular targets.

Studies were conducted at a sarcomere length of ∼2.2 μm, corresponding to ∼25% above slack length. This length approximated the plateau of the force–length relation in both neonates and adults 24and reflected a more physiologic condition. The Ca²⁺sensitivity of myofibrillar proteins is also known to be dependent on sarcomere length. 25,26Accordingly, we attempted to both standardize and optimize the conditions for age and anesthetic comparisons.

The lower force production by the neonatal cardiac muscle is attributable, at least in part, to the lower concentration of myofilaments, as reported by Reiser et al.  27and in the current study. Another contributing factor may be age-related differences in the geometry of myocytes, which would influence the actual number of cross-bridges. However, the contribution of this factor is likely to have been systematic and constant in both control and volatile anesthetic conditions within the same muscle sample.

Neonatal and adult rodent hearts predominantly express the V¹versus  the V³myosin heavy chain isoforms, respectively, 28which differ in their capacities for force generation, shortening velocity, and actomyosin ATPase activity, 29–31with the V³isoform being slower but more energy efficient. Furthermore, there are age-related differences in isoform expression of other thick and thin filament proteins. 28Accordingly, the relative expression of different myosin heavy chain isoforms may underlie, at least in part, the age-related differences in cross-bridge cycling kinetics and force generation. These issues are of significance not only in comparing volatile anesthetic effects of different age groups but also in evaluating anesthetic effects on adult hearts in pathologic conditions in which neonatal protein isoforms are reexpressed.

According to the model employed in the current study, the lower force in neonates is attributable to any of the following conditions: (1) a smaller total maximum number of cross-bridges (n), (2) a lower mean force per cross-bridge (¯F), and (3) a smaller fraction of strongly bound cross-bridges (α^fs). The total number of cross-bridges is reflected by the myofibrillar density and is known to be lower in neonates than in adults. Using the current techniques, it was not possible to estimate ¯F. Therefore, we focused on estimation of α^fsusing sinusoidal analysis, a procedure previously employed in skeletal muscle. 21,22Using this index, we found that the α^fsin neonates is significantly smaller than that in adults, suggesting an additional mechanism for lower force production. The α^fsreflects cross-bridge attachment and detachment, represented by the corresponding rates f^appand g^app. We found f^appto be significantly slower in the neonate; this was also reflected by the slower rate of force redevelopment (k^tr). These data are consistent with the slower shortening velocity of neonatal cardiac muscle reported by other groups. 29,30However, g^appwas not significantly different between ages. According to the model of Brenner et al. , 23differences in both ATPase activity and shortening velocity should parallel differences in g^app. The lack of age-related difference in g^appobserved in our study appears to be in conflict with the higher ATPase activities 31and shortening velocities 29,30of adult cardiac muscle compared with neonate cardiac muscle. This discrepancy may be more due to the limitation of the two-state cross-bridge model, in that the model does not consider the possibility that cross-bridge detachment may actually involve two rate constants, only one of which is estimated by sinusoidal analysis, but does not necessarily reflect ATPase activity. 32Regardless, the lower ATPase activity in the neonate suggests a lower actual rate of cross-bridge detachment and thus a greater proportion of time spent in the attached state, which would lead to a greater efficiency of ATP utilization. Overall, these data suggest that the lower force generated by the neonatal muscle is more a reflection of the slower f^appand lower α^fsthan the result of cross-bridge detachment.

Halothane decreased maximum force (normalized for cross-sectional area) to a greater extent in neonates, which is generally consistent with previous reports in several species. 5–8The effect of halothane is potentially attributable to the three factors described earlier for age-related differences. However, given the time frame of anesthetic action, it is unlikely that the total maximum number of cross-bridges (n), reflecting protein content, would be affected. Because muscles were preexposed to anesthetic, the possibility exists that the total number of cross-bridges were already decreased. However, the stiffness measurements were comparable in pCa 9.0 during control and anesthetic conditions (data not shown), suggesting that n was unaffected. In contrast, halothane does appear to decrease the fraction of strongly bound cross-bridges (α^fs), and thus the number of cross-bridges in the force-generating state (n ·α^fs). These findings are a direct demonstration of anesthetic effects at the level of the cross-bridge and support the suggestions made by Murat et al.  13in a previous study in skinned rat myocardium.

The considerably smaller effect of sevoflurane on force production compared with halothane is consistent with clinical and research literature on this issue. 33,34However, compared with the clinical situation, the depressant effect of sevoflurane in vitro  appears to be more pronounced in both neonates and adults. This most likely reflects the lack of compensatory neural and humoral mechanisms in vitro . As with halothane, the effects of sevoflurane may be due to an effect on ¯F or α^fs. However, we found no effects of sevoflurane on α^fsexcept at high concentrations, and then only in the neonate. This dissimilarity emphasizes the point that the two volatile anesthetics may differ in their intracellular targets.

Halothane decreased the ratio of F^Ato F^C(mean force per cross-bridge during anesthetic and control conditions, respectively) to a greater extent in the neonate compared with the adult. An inherent assumption in this technique is that the total number of cross-bridges (n) remains unchanged during rigor over time or in the presence of anesthetics (i.e. , there is no dropout of cross-bridges). In time controls, we found this dropout to be less than 5% as reflected by changes in the respective ratios, suggesting that this assumption is not strictly true. Nonetheless, this dropout was considerably less than the decrease in ratio observed with halothane, suggesting that the technique provides at least a qualitative measure of the effects of volatile anesthetic on ¯F, supporting the suggestion by Murat et al.  of a decrease in F. 13In a subset of muscle samples, we added anesthetic after the muscle had been placed in rigor (data not shown). We found that halothane decreased force production even during rigor conditions. These data are also consistent with a halothane-induced decrease in ¯F. In this regard, it is of interest that sevoflurane had a considerably smaller, and sometimes insignificant, effect on F.

Previous studies in rabbit papillary muscle have shown that the plateau of dynamic stiffness versus  frequency of length perturbations is decreased by volatile anesthetics, 13indicating a reduction in the number of cross-bridges (n ·α^fs). This effect has been largely attributed to a decreased affinity of Troponin C. However, it appears that volatile anesthetics may not affect the Ca²⁺affinity of Troponin C (TnC). 35Several other studies have used Ca²⁺concentrations sufficiently high to saturate TnC binding sites to show that decreased force due to volatile anesthetics is attributable to effects at the level of the cross-bridge itself. 9,11,12For example, in mechanically skinned rabbit 11and ferret papillary muscles 10(in which the sarcoplasmic reticulum is intact but the Ca²⁺is chelated), volatile anesthetics have been shown to decrease the maximal Ca²⁺-activated force, albeit to different extents. Furthermore, in Triton-X–skinned rat cardiac fibers (in which both sarcolemma and SR are destroyed), Murat et al.  13found that both maximal tension and Ca²⁺were decreased by volatile anesthetics. A recent study in Brij58-skinned human cardiac fibers also found that halothane and isoflurane decrease the maximum force in dose-dependent fashion. 12 

Halothane, and sevoflurane to a lesser extent, decreased the total stiffness in neonates to a greater extent than adults. Stiffness during maximum Ca²⁺activation is mainly a representation of the series elastic components of the muscle and is thought to chiefly reside in the cross-bridges themselves. In this regard, a lower total stiffness in neonates likely reflects a smaller number of attached cross-bridges, a combined effect of the smaller α^fsand protein content. In the case of anesthetic-induced changes in stiffness, the decreased stiffness at any age likely reflects the decreased α^fs.

Halothane significantly decreased α^fsand the rate for force redevelopment (k^tr). Based on the analytical models described earlier, these effects corresponded to a predominant depressant effect of halothane on the rate of cross-bridge attachment, f^app, and an acceleration of the rate of cross-bridge detachment, g^app. These effects, especially f^app, appear to be more pronounced in the neonate, especially at higher halothane concentrations. Sevoflurane also decreased k^tr, albeit to a somewhat smaller extent compared with halothane, especially in neonates. A slower f^appand a concomitant faster g^appwould lead to the cross-bridge being in the force-generating state for a smaller period of time, and thus a smaller number of attached cross-bridges at any given time. Accordingly, for force to be maintained, the rate of cross-bridge cycling would need to increase to compensate; this would be reflected by an increase in ATPase activity. In this regard, it is also of significance that the V³myosin isoform of the neonatal myocardium, where anesthetic effects on the cross-bridge appear to be greater, has a lower ATPase activity in the first place and, therefore, the proportionate compensatory increase in ATPase activity would have to be even greater. A number of studies have examined actomyosin ATPase in the presence of volatile anesthetics, albeit at relatively high concentrations, and have found either no significant changes 36or an actual decrease 37,38in ATPase activity. Nonetheless, it is unlikely that clinically relevant concentrations of volatile anesthetics will actually allow for an increase in ATPase activity. Therefore, the effects of halothane or sevoflurane at the level of the cross-bridge are likely to significantly contribute to the observed lower force.

In addition to the potential targets outlined earlier, age-related differences in myocardial sensitivity to volatile anesthetics may also be mediated via  effects on other regulatory proteins involved in force production. For example, the phosphorylation level of Troponin I may differ between neonates and adults 27and may be differentially affected by halothane. Furthermore, differences in Troponin T isoform expression 28may also contribute to age-related differences in force production. Differential effects of halothane versus  sevoflurane may also underlie the observed differences in the extent of cardiac depression. These potential mechanisms remain to be explored.