| ||||||||||||||
|
|
|||||||||||||

From the *Department of Anesthesiology and Intensive Care, University of Muenster, Muenster, Germany; and
Mayo Clinic and Foundation, Rochester, Minnesota.
Address correspondence and reprint requests to Dirk Breukelmann, MD, University of Muenster, Klinik und Poliklinik fuer Anaesthesiologie und Operative Intensivmedizin, Albert-Schweitzer-Strasse 33, 48149 Muenster, Germany. Address e-mail to breukel{at}uni-muenster.de.
| Abstract |
|---|
|
|
|---|
METHODS: We investigated the effect of halothane, isoflurane, and sevoflurane on calcium-dependent kinetics of isolated human recombinant cardiac troponin C labeled with IAANS (HrcTnCIAANS) using stopped-flow and calcium titration techniques.
RESULTS: Calcium concentration at half-maximal fluorescence intensity (Kd) in the control group was 2.1 ± 0.1 mM. Volatile anesthetics increased calcium sensitivity in a concentration-dependent fashion sevoflurane (Kd 1.51.7 mM, P = 0.001) > halothane (Kd 1.71.9 mM, P < 0.01) > isoflurane (Kd 1.81.9 mM, P < 0.05). The rate constant of conformational changes after rapid dissociation of calcium from HrcTnCIAANS (koff(c)) was moderately prolonged at 4°C by halothane and isoflurane > sevoflurane.
CONCLUSION: These mechanisms may counteract the effects of lower calcium availability, and can be responsible for abbreviated, and possibly incomplete, relaxation of cardiac muscle fibers in the presence of volatile anesthetics.
| Introduction |
|---|
|
|
|---|
Halothane, isoflurane, and sevoflurane are the most common volatile anesthetics used in clinical anesthesia today. It has been demonstrated that these anesthetics reduce intracellular Ca2+ availability during contraction (1,2), and also Ca2+ sensitivity, defined as the maximum amount of force generated at a given intracellular Ca2+ concentration (25). Whereas the diminished intracellular Ca2+ availability has been related to a decrease in intracellular L-type channel Ca2+ influx (6) and decreased sarcoplasmic reticulum Ca2+ content (3), the exact locus of action to affect Ca2+ sensitivity is not known. Possible mechanisms include, but are not limited to, the binding characteristics of Ca2+ to troponin C (TnC),1 interactions between regulatory troponin subunits, troponintropomyosinactin interactions, and changes in actomyosin cross-bridge kinetics (7).
Human cardiac troponin consists of three subunits: troponin T is the tropomyosin-binding site, troponin I functions as the inhibitory subunit, and TnC has been identified as the Ca2+-binding regulatory subunit. Conformational changes of the troponin complex ultimately affect the probability of actomyosin binding, and thus regulate cross-bridge cycling in the cardiac muscle. Those effects are induced by Ca2+-binding and -release to and from the Ca2+-specific binding site II of the cardiac TnC molecule, whereas the remaining two high-affinity binding sites of the cardiac TnC molecule are saturated with Mg2+ in physiological conditions.
Therefore, a possible mechanism by which volatile anesthetics reduce cardiac cross-bridge-binding upstream in the regulatory chain could be via a direct effect on the Ca2+-dependent properties of cardiac TnC.
We have examined conformational changes of isolated human recombinant cardiac TnC labeled with sulfhydryl-specific fluorescent probe molecule 2-(4'-iodo-acet-amido-anilino)-naphthalene-6-sulfonic acid (HrcTnCIAANS) (8) on binding and release of Ca2+ to and from the Ca2+-specific site II during exposure to different concentrations of halothane, isoflurane, and sevoflurane. HrcTnCIAANS has been shown to change its fluorescence properties on calcium-induced conformational changes of cardiac troponin when bound to cysteine 35 and 84 of the native TnC molecule. The labeled protein undergoes a small decrease in fluorescence due to Ca2+ or Mg2+ binding to its two high-affinity sites, and a large increase in fluorescence on Ca2+ binding to the low-affinity Ca2+-specific site. HrcTnCIAANS exhibits a Ca2+-sensitive fluorescence peak at excitation 340 nm and emission 445 nm (9).
We hypothesized that volatile anesthetics reduce the affinity of HrcTnCIAANS site II Ca2+-binding and/or kinetics of subsequent conformational changes by exerting direct effects on the isolated protein.
| METHODS |
|---|
|
|
|---|
· cm (Nanopure Infinity UF Series 898, Barnstead/Thermolyne, Dubuque, IA) to minimize ion contamination. Solutions were kept at 4°C for a maximum of 5 days to avoid degradation. Since pH affects Ca2+ binding characteristics of myocardial TnC, 3-(N-morpholino) propanesulfonic acid (MOPS) pH buffer capacity was carefully chosen, so that Ca2+ binding to [ethylenebis-(oxyethylenenitrilo)]-tetraacetic acid (EGTA) would not change pH over a range of pCa 3.59.0 (data not shown).
Protein Purification and IAANS Labeling
IAANS labeling of human recombinant cardiac TnC (HrcTnCIAANS) was performed using standard methods (8). IAANS binding stoichiometry was determined to be 1.08 moles per mole protein using absorbance characteristics specific to IAANS. Since the wild type TnC possesses two cysteine residues (Cys35 and Cys84), about 50% of the potential IAANS binding sites can be considered labeled.
HrcTnCIAANS was extensively dialyzed against Ca2+-free protein buffer containing MOPS 120 mM, KCl 90 mM, pH 7.0 in purified water, aliquoted, and stored at 80°C until use.
Anesthetic-Saturated Buffers
To obtain anesthetic-saturated buffer, each anesthetic was allowed to equilibrate with titration buffer for at least 2 h at room temperature. Different volumes of anesthetic-saturated buffer were mixed with titration buffer to expose HrcTnCIAANS to increasing anesthetic concentrations.
Gas Chromatography
Gas chromatographic analysis was performed in a Hewlett Packard 5880A Gas Chromatograph, using either a Porapak Q 80/100 column (Hewlett Packard, Avondale, PA) with a Nickel-63 electron capture detector, for halothane and isoflurane, or a fused silica column (007-624 Series, Quadrex Corp. Woodbridge, CT) combined with a flame ionization detector for sevoflurane. Anesthetic concentrations were determined after extracting the volatile compound of the buffer in hexane, which was then injected into the gas chromatograph.
Equilibrium Titration
Steady-state fluorescence (excitation 330 nm, emission 445 nm) was determined in a constantly stirred, temperature-controlled quartz cuvette (pathlength 1 cm, total volume 4.5 mL) at 25°C in a SLM Aminco 8100 Spectrofluorimeter (Software Version 4.00) at a sampling rate of 1 Hz. Fifty microliters protein buffer (MOPS 120 mM, KCl 90 mM, HrcTnCIAANS 2000 µg/mL, pH 7.0) was added to a total volume of 1950 µL titration buffer (MOPS 120 mM, KCl 90 mM, EGTA 2 mM, pH 7.0) and allowed to equilibrate for stable baseline fluorescence. Different anesthetic concentrations in the solution were achieved by varying the portion of anesthetic-saturated buffer.
|
|
Equation 1 is the Hill equation to fit experimental data to sigmoid curve where Kd is the Ca2+ concentration with half-maximal fluorescence Y and n is the Hill coefficient, steepness of the sigmoid curve.
The expected changes in anesthetic concentrations due to gasliquid equilibration in the cuvette were calculated on the basis of temperature-corrected anesthetic partition coefficients (
) (10) under the assumption that the total amount of anesthetic is added into the liquid phase and remains constant in the liquidgas system [Eq. (2)]. Anesthetic concentrations were determined by gas chromatography before and at the end of the experiment.
Statistical analysis for differences between control and anesthetic groups was performed using KruskalWallis and Wilcoxon's tests, with P < 0.05 being considered significant (11).
Stopped-Flow Spectrofluorimetry
The effect of Ca2+ dissociation from the single regulatory site II on the kinetics of conformational changes of HrcTnC was assessed in a fluorescence stopped-flow analyzer (Applied Photophysics SX.18 MV, Leatherhead, Surrey, UK) at 4°C and 20°C. HrcTnCIAANS saturated with Ca2+ (HrcTnCIAANS 15 µg/mL, pCa 3.6 in K+ 140 mM, Cl 90 mM, MOPS 120 mM, pH 7.0) in the presence or absence of 1 mM Mg2+ was rapidly (0.9 ms) mixed 1:1 with EGTA 9.75 mM (same buffer). Fluorescence (excitation 330 nm, emission >395 nm) between 1 and 100 ms was recorded and fitted to a first-order exponential function to yield the rate constant of conformational change koff(c). Different anesthetic concentrations were obtained by varying the fraction of anesthetic saturated buffer (same composition) in both solutions before mixing. Glass syringes were used to minimize evaporation.
Anesthetic concentration in each syringe was determined by gas chromatographic analysis. Effects of the respective anesthetic were analyzed according to the anesthetic concentration for both Mg2+ 1 mM and Mg2+-free groups individually.
Statistical analysis was performed using Wilcoxon's ranked sum test to test for differences of anesthetic group versus control and effect of anesthetic concentration. P < 0.05 was considered significant (11).
| RESULTS |
|---|
|
|
|---|
Anesthetic concentrations in the liquid phase at the end of the calcium titration (expressed as a fraction of the concentration immediately after injection and mixing ± sd) were determined to be 0.82 ± 0.13 (halothane, n = 18), 0.69 ± 0.10 (isoflurane, n = 25), and 0.75 ± 0.11 (sevoflurane, n = 21).
|
|
Equation 2 is used to calculate expected anesthetic concentration in the liquid phase after gas:liquid equilibrium where ntotal is the total amount of substance injected into liquid phase; volliq is the volume of liquid phase; volgas is the volume of gas phase;
is the liquid:gas partition coefficient; and nliq is the amount of substance in liquid phase after equilibration.
Site II Ca2+ Affinity
Least square fitting and statistical analysis showed that Kd in the control group was 2.1 ± 0.1 mM; the Hill coefficient (n) for the cooperativity of the metalligand interaction was 2.7 ± 0.6 (n = 7). Halothane at concentrations of 0.2 ± 0.1 mM decreased Kd to 1.9 ± 0.2 mM (n = 7, P < 0.05 compared with control). Halothane 1.0 ± 0.2 mM led to a further decrease of Kd to 1.7 ± 0.1 mM (n = 9, P < 0.01 to control) and also increased the Hill coefficient n to 3.7 ± 0.4 (n = 9, P < 0.01 to control). A further excessive increase of halothane concentration to 2.5 ± 0.3 mM showed Kd at 1.7 ± 0.1 (n = 8, P < 0.001 to control). However, at that concentration the Hill coefficient decreased to 3.1 ± 0.3 (n = 8, n.s. compared with control). Isoflurane decreased Kd at concentrations of 0.1 ± 0.02 mM, 0.6 ± 0.1 mM, and 1.2 ± 0.3 mM to 1.8 ± 0.1 mM (P < 0.01, n = 8), 1.8 ± 0.1 mM (P < 0.01, n = 10), and 1.9 ± 0.2 mM (P < 0.05, n = 9), respectively. Isoflurane did not affect the Hill coefficient at any concentration. Sevoflurane decreased Kd to 1.7 ± 0.2 mM at 0.4 ± 0.1 mM (P < 0.001, n = 9) and 1.5 ± 0.1 mM at 0.6 ± 0.1 mM (P = 0.001, n = 9), without significant effect on the Hill coefficient (Figs. 1 and 2).
|
|
|
|
| DISCUSSION |
|---|
|
|
|---|
However, the exact mechanisms of the latter effect remain unclear. One of the possible mechanisms playing an important role in the process of force generation in cardiac muscle upstream of the cross-bridge level is the Ca2+ affinity of the regulatory protein complex, which is most likely to be determined by the Ca2+ binding characteristics of TnC.
In this in vitro study, halothane, isoflurane, and sevoflurane lead to a decrease of Kd (e.g., a left shift of the troponin-Ca2+ binding curve), whereas the Hill coefficient n, describing the steepness of the sigmoid curve, was increased (halothane) or remained largely unaffected. The effect was concentration-dependent, reaching a maximum impact at a concentration of 1 mM (1.7 vol %, halothane) and 0.1 mM (0.3 vol %, isoflurane) with no further decrease of Kd at higher concentrations. For sevoflurane Kd decreased in a concentration-dependent manner up to the highest concentration 0.6 mM (2.5 vol %) used.
All volatile compounds investigated tended to increase the rate-constant of Ca2+ dissociation-induced conformational change koff(c) of HrcTnCIAANS at lower temperatures, although it has to be noted that the changes observed are rather small with considerable concentrations of anesthetics used. At 20°C halothane, more than isoflurane, decreased koff(c) at lower concentrations and showed a progressive increase with higher anesthetic concentrations. Sevoflurane did not affect the rate-constant of conformational change.
On the basis of the law-of-mass action, increased Ca2+ sensitivity (Kd) and altered rate of conformational change after Ca2+ dissociation from HrcTnC (off-rate) suggest an increase in the rate of conformational change on Ca2+ binding to TnC (on-rate). This is consistent with experimental observations of anesthetic effects on cardiac muscle contractility showing decreased time to peak shortening and time to half-maximal relaxation for sevoflurane (2), little effect of isoflurane, and prolonged contraction and relaxation on exposure to halothane (1).
This result is in contrast with the findings of Blanck et al. (22) who were not able to demonstrate any effects of halothane on TnC-Ca2+ binding kinetic using intrinsic tyrosine fluorescence, ultraviolet circular dichroism, and 5,5'-dithiobis(2-nitrobenzoic acid) binding techniques. This might be explained by the different sensitivities of the experimental procedures used. One advantage of using a previously IAANS-labeled TnC protein is that the molecular structure of the protein-IAANS complex remains unaltered during the entire experimental procedure. It has been shown that the IAANS-labeled protein does not significantly differ in physiological Ca2+ binding, undergoes Ca2+-induced increases in
helix, and also forms complexes with other troponin subunits similar to unlabeled cTnC (8). There is, however, a possibility that volatile anesthetics do interact specifically with IAANS labeling sites, and thus, interfere with Ca2+-dependent conformational change-induced fluorescent signals.
Skinned muscle experiments have suggested that halothane and isoflurane decrease Ca2+ sensitivity and maximal force in human cardiac fibers (4,18). Since skinned muscle experiments focus on a disrupted, although vital, part of the contractile machinery, possible important interactions with enzymes, second-messengers, or even parts of the interacting filaments may remain undetected.
Physiologically, increased Ca2+ sensitivity counteracts negative inotropic side effects of halothane, isoflurane, and sevoflurane, which are to a great extent due to decreased intracellular Ca2+ availability (13). Sevoflurane exhibited the most pronounced increase in Ca2+ sensitivity in our study which may, in part, account for the relatively lower cardio-depressant activity compared with other volatile anesthetics. When mechanisms mediating negative inotropic effects are absent or weak, a sevoflurane-dependent increase in Ca2+ sensitivity might even explain an overall positive inotropic response to sevoflurane exposure, as described by Vivien et al. (23).
| CONCLUSION |
|---|
|
|
|---|
| Footnotes |
|---|
Accepted for publication October 16, 2006.
| REFERENCES |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|