| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
Departments of Anesthesiology and Genetics, University Hospitals and Case Western Reserve University, Cleveland, Ohio
Address correspondence and reprint requests to Philip G. Morgan, MD, Department of Anesthesiology, 2400 Bolwell Building, University Hospitals, 11100 Euclid Ave., Cleveland, OH 44106. Address e-mail to philip.morgan{at}uhhs.com
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Implications: As in mammals, nonimmobilizers antagonize the effects of halothane on the nematode, C. elegans. The variation in slopes in the response to halothane in different strains is consistent with multiple sites of action. These results support the use of C. elegans as a model for the study of anesthetics.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Eckenhoff and Johannson (11) have discussed the potential error of requiring that concentrations of volatile anesthetics, necessary to achieve in vitro end points, match those needed for in vivo end points. They point out that, if multiple sites of anesthetic action exist, their effects on the anesthetized state of a whole animal may be additive. This would shift the resulting dose-response curve significantly to the left of a curve generated by a single site of action studied in vitro. Because the number of sites of action may vary among different species, these arguments can also make predictions about the 50% effective concentrations (EC50) of anesthetics for widely divergent species. Organisms with more complicated nervous systems might be expected to have more contributing sites of anesthetic action than more simple animals. The resulting dose-response curves of a complex animal would be shifted to the left (a lower EC50 or minimum alveolar anesthetic concentration) compared with those of an animal with a simpler nervous system.
In addition to comparing EC50 values between in vitro and in vivo systems, Eckenhoff and Johannson (11) also make predictions concerning the slopes of dose-response curves in these two very different experimental systems. If a behavioral end point results from the concerted activity of several anesthetic targets, then the dose-response curve for the whole animal will have a steeper slope than the dose-response curve of an in vitro assay of that same organism. However, such predictions are confounded by the genetic heterogeneity of populations of complex organisms. Genetic variability will tend to flatten any dose-response curves generated in a population of mammals, possibly blunting the differences in the data between in vitro and in vivo systems.
The wild type strain and most mutant strains of C. elegans are unique in that they are isogenic (all animals are genetically identical to each other). The slope of a dose-response curve of any one strain represents the added responses of an identical group of targets and not the genetic variation of the population. The differences in slopes between strains can then be assigned to specific mutations rather than to population variability.
Studies in C. elegans, by using reversible immobility as an end point, are most consistent with multiple sites of anesthetic action (4). Several genes have been identified and characterized that affect anesthetic sensitivity. Mutations in the gene, unc-1, cause a specific increase in sensitivity to diethylether, whereas mutations in the gene, gas-1, cause an increase in sensitivity to all volatile anesthetics (4,5). Mutations in unc-79 cause an increase in sensitivity to halothane but not to isoflurane (12,13). Unc-1 has been identified as coding for a close homologue of the mammalian protein stomatin (14), whereas gas-1 codes for a component of Complex I of the mitochondrial respiratory chain (15). The identification of unc-79 is not known at this time.
We first hypothesized that nonimmobilizers would fail to anesthetize C. elegans. The responses of the wild type nematode and several mutants to two nonimmobilizers and a transitional compound were characterized. We then hypothesized that the slopes of the responses to halothane of mutant strains with different sensitivities would also have different slopes, consistent with a different number of anesthetic targets. The results show decreased slopes of dose-response curves in the more sensitive strains and lend support to the arguments put forth by Eckenhoff and Johannson (11). In addition, they allow us to make a general prediction of the number of anesthetic targets leading to immobility in C. elegans.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Halothane was purchased from Anaquest, Liberty Corner, NJ. The nonimmobilizers 2,3dichlorooctofluorobutane (DCOFB) and 1,2dichlorohexafluorocyclobutane (DCHFCB) were purchased from PCR, Gainesville, FL. Toluene (reagent grade) was purchased from Fisher Scientific, Pittsburgh, PA.
Nematodes were anesthetized as previously described (4). C. elegans strains are isogenic with an extremely small forward mutation rate (16). Therefore, any population variability should be the result of environmental conditions. All strains are synchronized before testing and grown under identical conditions so that population variability is basically nonexistent. Adult hermaphrodite worms were allowed to lay eggs for 4 h on agar plates spread with E. coli as a food source. The adults were removed and the plates were incubated at 20°C until the offspring were all young adults (approximately 3.5 days) within 4 h of age of each other. Then, they were tested for anesthetic sensitivity.
Synchronized cultures of worms on agar plates were placed in a glass air-tight chamber. A liquid volume of anesthetic, calculated to give an appropriate gas concentration based on the chamber’s volume, was injected with a glass syringe into the sealed chamber via a stopcock. The chambers were returned to the 20°C incubator and removed later for scoring. The worms were observed through the chamber’s lid with a dissecting microscope, and judged to be anesthetized when they were immobile for at least 10 s. Normal worms move in a constant sinuous motion across the plate. The concentrations of anesthetic necessary for this end point have been correlated with those necessary for loss of response to a tap on the worm snout and found to be identical. A steady-state response was noted after 90 min and all chambers were scored at 2 h. All chambers were scored by two independent observers and the resulting anesthetized fraction was the average of the two scores.
Anesthetic concentrations were measured with a gas chromatograph. Dose-response curves were based on a minimum of 15 different concentrations of anesthetic with at least 50 animals per concentration. The animals were used only once and then discarded.
Nematodes were exposed to the nonimmobilizers and toluene initially, as described for other volatile anesthetics (previously mentioned). All concentrations of anesthetics and nonimmobilizers were measured by gas chromatography (GC). All GC samples were withdrawn from the gas over the nematodes in the glass chambers through a stopcock in the glass chambers. All chambers were checked before the experiments for stability of gas concentrations. As for other anesthetics, the concentrations of nonimmobilizers were determined by comparison to a known standard curve for each gas on each day. The standard curves were determined by injection of a known amount of liquid nonimmobilizer into a closed large flask of known volume under vacuum. After evaporation of the liquid, air was added to the flask to reach atmospheric pressure. The concentration of the resulting gas was calculated and serial dilutions were made to construct a concentration curve for the GC.
Because the maximum concentration obtained at atmospheric pressure of either nonimmobilizer had no effect on the worms, a constant amount of nonimmobilizer was added to varying amounts of halothane to obtain a dose-response curve. The resulting curve was compared with that for halothane alone to determine whether nonimmobilizers shifted the halothane curve.
EC50 values and slope constants with standard errors were calculated as described by Waud (17). The slope constant is derived from considerations of quantal responses and corresponds to the more commonly used Hill coefficient as a measure of the rate of increase of the dose-response curves. EC50 values and slope constants were compared by using analysis of variance as previously described (4). The resulting dose-response curves were plotted by using the plotting program, Sigmaplot (Jandel Corporation, San Rafael, CA). The nature of this plotting program caused the curves to not be smooth, but were not used for calculation of the EC50 values or errors.
The predicted EC50 values for DCOFB, DCHFCB, and toluene were determined by comparing their oil gas partition coefficients (O/Gs) at 23°C to the Meyer-Overton relationship previously reported for C. elegans (13,18). The mean product of the O/Gs times EC50 values (for immobility) for volatile anesthetics in C. elegans is 11.6. The predicted EC50 values were estimated by dividing 11.6 by the measured O/Gs at 23°C. The O/Gs were estimated by measuring the amount of compound dissolved in olive oil at 23°C when a known amount of the volatile compound was placed in olive oil and the concentration in air over the oil was measured as described by Taheri et al. (19). Each measurement was repeated in triplicate. The measured O/Gs were 74 ± 5 (SD) for DCHFCB, 42 ± 4 for DCOFB, and 3600 ± 400 for toluene.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
These compounds may have an anesthetic effect at a concentration larger than we could achieve at atmospheric pressure. Thus, we added a constant amount of DCOFB (15% ± 1%) to increasing amounts of halothane to see whether we could identify any additive anesthetic effects between the two drugs. The dose-response curves are shown in Figure 1 and the EC50 values and slope constants are shown in Tables 1 and 2, respectively. As noted before, the slopes and EC50 values for gas-1 and unc-79 were decreased compared with N2 (4). As noted previously, DCOFB failed to act alone as an anesthetic in any case. However, in N2 and unc-1, DCOFB demonstrated an anti-anesthetic effect, shifting the halothane dose-response curves to the right. In addition, in these cases, the dose-response curve for halothane plus DCOFB had a significantly decreased slope constant compared with that for halothane alone. Adding DCOFB did not alter the response of either unc-79 or gas-1 to halothane. DCHFCB, which had no anesthetic effect when used alone, gave effects similar to those seen with DCOFB. Specifically, DCHFCB (14 ± 1%) demonstrated an anti-anesthetic effect, shifting the halothane dose-response curves of N2 and unc-1 to the right. (Tables 1,2)
|
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
To identify whether these compounds have effects at concentrations larger than those we could achieve, we studied whether DCOFB or DCHFCB would add to the anesthetic effects of halothane. If so, we could calculate their effective EC50 values. However, we found that both DCOFB and DCHFCB antagonized the anesthetic effects of halothane on the wild type worm, N2. Because we did not see this effect in mutants which have decreased function of unc-79 or gas-1, we conclude that this antagonism requires the normal copies of both of these genes. DCOFB and DCHFCB could either act through the unc-79 and gas-1 wild type genes (and thus require their presence) or these drugs may affect other functions only sensitive to the larger concentrations of halothane required to immobilize N2.
We are not discussing the absolute slope of any of these strains, only their comparative differences. The absolute slope of any isogenic strain could be extremely steep because it results from quantal responses (all or none). Thus, it is possible that all animals could change their behavior despite very small changes in anesthetic concentration. It is not clear whether the absolute slopes of the dose-response curves result from multiple targets, characteristics of the receptor concentration-behavioral response relationship, or small differences between individuals, despite the strains being isogenic or other considerations. However, because all strains discussed here are isogenic, these considerations apply equally to each of them.
Fang et al. (10) reported that flurothyl (hexaflurodiethylether) and DCOFB acted via different mechanisms in producing convulsions in rats. We reported earlier that small doses of flurothyl were antagonistic to halothane in C. elegans (13). However, flurothyl is able to act as an immobilizer in N2 at an EC50 predicted by its O/G (13). Thus, we note strikingly different effects between flurothyl and both DCOFB and DCHFCB. These data are in agreement with the conclusion of Fang et al. (10) that these different types of convulsants work, at least partially, in different manners.
Toluene acts as a transitional compound in mammals (9). We find that this is also true in the wild type nematodes. The EC50 for N2 in toluene is 2–3 times larger than predicted by its lipid solubility. This is less of a difference than is seen in mice in which the observed EC50 is 5–6 times the predicted value. In addition, we found that unc-79 and gas-1 have increased sensitivities to toluene compared with N2 and unc-1. This is consistent with previous findings that unc-79 is more sensitive to anesthetics with lipid solubilities greater than that of halothane (calculated as 407 at 23°C) (13) and that gas-1 is more sensitive to all volatile anesthetics. However, comparing the responses of gas-1 and unc-79 to toluene with their responses to other anesthetics with high lipid solubilities, we would predict that these mutants should have EC50 values between 0.05 and 0.1 vol% in air. Thus toluene is a transitional compound for these two mutants. C. elegans responds to both nonimmobilizers and a transitional compound in patterns similar to those in mammals.
Immobility in C. elegans satisfies several criteria as a potential model for anesthesia in mammals. However, one must still consider the relatively larger doses necessary for immobilization in nematodes compared with minimum alveolar concentration in mammals. The larger EC50 for halothane is roughly one-half to one-third of the LD50 for this drug (13), also arguing that it is analogous to a mammalian EC50. It is possible that anesthetic targets in C. elegans are simply very different than those in mammals. However, it appears unlikely that such different targets would satisfy so many other criteria for the action of volatile anesthetics and have no relation to those affected in mammals. This appears especially true given the overall similarities between neurotransmitters in C. elegans and mammals (20). Certainly, one possibility is that anesthetic targets, although homologous, have diverged enough across the animal kingdom to see a relative resistance to volatile anesthetics in nematodes compared with mammals. It is also possible that anesthetic targets in these divergent species are very much the same, but that the overlay of multiple downstream physiologic processes modulate the final response of different organisms. In this case, it will be instructive to see what will be the effect of mutations in mice in genes for molecules already identified as potential targets for volatile anesthetics in C. elegans. Selective pressure must have altered at least some of the physiologic functions in these nematodes to be able to function in the temperature range of 10°C–25°C (roughly the range in which they live) and in an environment in which they are exposed to a wide array of chemicals. However, it is also possible that the arguments of Eckenhoff and Johansson (11) have bearing on this finding. If the number of targets is less in C. elegans than in mammals, then we may see a shift in the EC50 values compared with more complicated organisms based on that fact alone.
How many targets might there be in C. elegans? Because single mutations cause the increased sensitivities seen in unc-79 and gas-1 animals, it is probable that the responses in these animals are the result of a single type of anesthetic target. If we assume that the slope of the response of unc-79 (approximately 10) to halothane reflects a single target in C. elegans, then we may estimate the number of targets required leading to the slope seen for N2 (approximately 20). By using the unc-79 slope as a starting point, we added integral numbers of the response curves together until a dose-response curve with a slope of 20 was generated. By this method, we estimate that 3–4 such targets, each with EC50 values larger than N2, would be required to equally contribute to the end point seen in N2. This is a purely theoretical argument, and is dependent on the starting assumption that there is one active site of action for unc-79 or gas-1. However, it is interesting that the isolation of different genetic patterns of anesthetic susceptibility previously led us to postulate at least three anesthetic targets in C. elegans. Similar considerations are consistent with a model in which nonimmobilizers inhibit halothane action on 1–2 anesthetic sites. By themselves, the results in this study do not eliminate a single site of action for volatile anesthetics in C. elegans. However, with earlier genetic studies, we feel they are most consistent with multiple sites of anesthetic action.
In summary, we have found that the nonimmobilizers DCOFB and DCHFB have antagonistic effects on the response of C. elegans to halothane and that the presence of both unc-79 and gas-1 genes is necessary for this response. Such mutations offer a powerful approach to dissect out the gene products necessary for responses to such compounds. Because they do not cause immobility by themselves, DCOFB and DCHFCB function at least, in part, by a different mechanism than does the convulsant, flurothyl (which causes immobility). These data are consistent with C. elegans having at least three components that contribute to the end point of immobility when exposed to halothane.
|
Acknowledgments |
|---|
We are indebted for the technical assistance of Julie Seifker and to Laura Staufer in the preparation of the manuscript. Finally, we thank E. I Eger, II, for his suggestion to study nonimmobilizers.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. A. Humphrey, M. M. Sedensky, and P. G. Morgan Understanding anesthesia: making genetic sense of the absence of senses Hum. Mol. Genet., May 15, 2002; 11(10): 1241 - 1249. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
