A Stomatin and a Degenerin Interact to Control Anesthetic Sensitivity in Caenorhabditis elegans

Corresponding author: Phil G. Morgan, Department of Anesthesiology, 2400 Bolwell Bldg., University Hospitals of Cleveland, 11100 Euclid Ave., Cleveland, OH 44106., pgm2{at}po.cwru.edu (E-mail)

Communicating editor: I. GREENWALD

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The mechanism of action of volatile anesthetics is unknown. In Caenorhabditis elegans, mutations in the gene unc-1 alter anesthetic sensitivity. The protein UNC-1 is a close homologue of the mammalian protein stomatin. Mammalian stomatin is thought to interact with an as-yet-unknown ion channel to control sodium flux. Using both reporter constructs and translational fusion constructs for UNC-1 and green fluorescent protein (GFP), we have shown that UNC-1 is expressed primarily within the nervous system. The expression pattern of UNC-1 is similar to that of UNC-8, a sodium channel homologue. We examined the interaction of multiple alleles of unc-1 and unc-8 with each other and with other genes affecting anesthetic sensitivity. The data indicate that the protein products of these genes interact, and that an UNC-1/UNC-8 complex is a possible anesthetic target. We propose that membrane-associated protein complexes may represent a general target for volatile anesthetics.

WHILE many drugs can be combined to produce general anesthesia, the volatile anesthetics are unique in that, by themselves, they can cause loss of consciousness, amnesia, analgesia, and muscle relaxation, the four hallmarks of a complete general anesthetic. These gases disrupt neuronal function across phyla in a manner that suggests that they affect conserved functions. However, despite more than 150 years of use, very little is known about the specific mechanisms of action of these drugs (MILLER et al. 1989 ; FRANKS and LIEB 1994 ). Volatile anesthetics encompass a wide variety of chemical structures. Compounds as simple as xenon or argon, or as complicated as halothane (CF3CHBrCl) or diethylether, all cause essentially the same behavior. Such diversity does not seem easily compatible with a standard lock-and-key model often applicable to other drugs.

However, the volatile anesthetics do have some characteristics in common. Since the turn of the century, it has been known that in all animals studied, the potency of a volatile anesthetic is primarily a function of its lipid solubility (MEYER 1899 ; OVERTON 1901 ). This relationship, termed the Meyer-Overton rule, states that the product of the oil/gas partition coefficient (in olive oil or octanol) and the EC₅₀ (the effective concentration at which 50% of animals are anesthetized) is approximately a constant for all the volatile anesthetics. In addition, the normalized effects of the different anesthetics are additive, i.e., adding one-half of an effective dose of one anesthetic to one-half of an effective dose of a second anesthetic acts like a full effective dose of either anesthetic alone. Because of these characteristics, it was long assumed that the volatile anesthetics acted rather nonspecifically, probably on lipid bilayers in cell membranes, and that they all functioned by an identical method. However, recent work has indicated that a nonspecific model may be too simplistic. Several studies have shown that volatile anesthetics can bind to proteins, either soluble or associated with membranes, and that they can affect the function of such proteins (FRANKS and LIEB 1994 ; ECKENHOFF 1996A , ECKENHOFF 1996B ). Hypotheses about the mechanism of action of volatile anesthetics now include both protein and lipid models. However, the unitary hypothesis of general anesthetics maintains that a universal mechanism underlies the effects of all the volatile anesthetics in all species.

Several researchers have undertaken genetic approaches to identify the site of action of volatile anesthetics. Studies in mice (BAKER et al. 1980 ; KOBLIN and DEADY 1981 ; MCCRAE et al. 1993 ), fruit flies (GAMO et al. 1981 ; KRISHNAN and NASH 1990 ), and nematodes (SEDENSKY and MENEELY 1987 ; MORGAN and SEDENSKY 1994 ; VAN SWINDEREN et al. 1999 ) have identified genetic changes that affect responses to volatile anesthetics. However, several of these genetic changes alter behavior in some volatile anesthetics differently than others. Differences have also been found in the electrophysiologic responses to various volatile anesthetics in studies of the hippocampus of rats (MACIVER and KENDIG 1991 ). The unitary hypothesis, in its simplest form, predicts that mutations at anesthetic sites of action should affect sensitivities to all volatile anesthetics in a similar fashion and that the neurophysiologic changes caused by each should be similar. Taken together, these data indicate that the unitary hypothesis may be an oversimplification and that the mechanism of action of volatile anesthetics may involve several molecular sites (KRISHNAN and NASH 1990 ; MORGAN et al. 1990 ).

The choice of anesthetic endpoints remains a concern in all these genetic studies. We have chosen immobility as the endpoint in Caenorhabditis elegans, because it correlates closely with loss of a response to a noxious stimulus, a similar behavioral endpoint as that used in humans. The ratio between the EC₅₀'s and LC₅₀'s (the lethal concentration for 50% of animals) is similar for immobility in C. elegans and surgical anesthesia in mammals. Anesthetic-induced immobility is totally reversible and obeys several long-held doctrines of anesthetic practice, including the Meyer-Overton rule, additivity of EC₅₀'s and a decrease in potency of alcohols and alkanes when the size of the carbon chain becomes too long (the cutoff effect; MORGAN and CASCORBI 1985 ; ANTON et al. 1992 ). However, immobilization of nematodes requires a higher EC₅₀ than seen in mammals for other behavioral endpoints, including surgical incision in humans. In C. elegans, others have used endpoints corresponding to subclinical behavioral changes (such as failure to move normally) in an effort to construct an endpoint with an EC₅₀ similar to that for humans (CROWDER et al. 1996 ; VAN SWINDEREN et al. 1999 ). However, behaviorally similar subclinical endpoints in humans would have correspondingly lower EC₅₀'s than the EC₅₀ for loss of consciousness. Thus, at this time, it is unclear what the corresponding mammalian endpoints might be for any anesthetic endpoint chosen in nematodes.

Using immobility as an endpoint, we have identified several genes that affect sensitivity to volatile anesthetics in C. elegans (SEDENSKY and MENEELY 1987 ; MORGAN et al. 1990 , 1994). The interactions between mutations in at least six genes (unc-79, unc-80, unc-1, unc-7, unc-9, and gas-1) suggest that they lie in a functional pathway controlling anesthetic sensitivity. The gene unc-1 occupies a central position in this pathway. Loss-of-function mutations in unc-1 suppress the altered sensitivities of upstream mutations (unc-79 and unc-80) and confer their own changes in sensitivities to volatile anesthetics. The characterization of unc-1 serves as a key to understand the action of volatile anesthetics.

The UNC-1 protein has been identified as a homologue of the mammalian protein stomatin (RAJARAM et al. 1998 ). Alterations in mammalian stomatin are associated with a congenital hemolytic anemia termed overhydrated hereditary stomatocytosis (STEWART et al. 1993 ). Erythrocytes from patients with this abnormality show a 20-fold increase in membrane leakage of the cations Na⁺ and K⁺. Thus, mammalian stomatin has been postulated to control the function of an associated ion channel in a ball-and-chain manner (STEWART et al. 1993 ). In this study, we examined the responses of unc-1 mutants to several volatile anesthetics. Different mutations in unc-1 can increase or decrease anesthetic sensitivity. We also determined the expression pattern of the UNC-1 protein using the green fluorescent protein (GFP; CHALFIE et al. 1994 ) and found that it was widely expressed in the nervous system.

A homologue of stomatin in C. elegans is coded by mec-2. MEC-2 is found in nerve cells necessary for transducing mechanical stimulation to a behavioral response (CHALFIE and AU 1989 ; HUANG et al. 1995 ). Chalfie has postulated that this protein gates a passive sodium channel that is a member of the superfamily of epithelial sodium channels (ENaCs; DRISCOLL and CHALFIE 1991 ; SHREFFLER et al. 1995 ). A subunit of a different ENaC is coded by the unc-8 gene and is expressed in motor neurons (SHREFFLER et al. 1995 ; TAVERNARAKIS et al. 1997 ). We speculated that UNC-8 might represent a subunit of the channel affected by the UNC-1 protein. Therefore, we have examined the genetic interaction of unc-8 mutants with several different alleles of unc-1, as well as the anesthetic sensitivity of multiple unc-8 mutants. We have also tested the effects of unc-8 alleles on other genes in the pathway controlling sensitivity to volatile anesthetics. The expression pattern of UNC-1 suggests that it may interact with UNC-8. unc-8 mutations can increase or decrease the sensitivity of the animal to volatile anesthetics. Genetic interactions of unc-8 with unc-1 and other genes that control behavior in anesthetics support the hypotheses that UNC-8 plays a central role in the response of C. elegans to anesthetics and that it may interact directly with UNC-1. We hypothesize that UNC-1, UNC-8, and UNC-79 are part of a protein complex that interacts with volatile anesthetics.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Nematodes:
Nematodes were anesthetized as described previously (MORGAN et al. 1990 ), and EC₅₀'s with standard errors were calculated as described by WAUD 1970 . Nematodes were cultured as described previously (BRENNER 1974 ); all experiments were performed at 20°–22°. The wild-type nematode is N2 in all experiments and was obtained from the Caenorhabditis Genetics Center. The mutant bearing the dominant unc-1 X allele, n494, was obtained from Carl Johnson. The unc-1 mutants, e580 (null), e114 (hypomorph), and hs2-5 (temperature sensitive) were obtained from the Caenorhabditis Genetics Center. The temperature-sensitive unc-1 mutant hs1 was obtained from Ralph Hecht. The only unc-79 III and unc-80 V mutants that were studied carried the null alleles ec1 and e1272, respectively. Because only one allele was used for each, we have referred to them simply as unc-79 and unc-80 in the text. The same is true for unc-7(e5)X and unc-9(e101)X. unc-8 IV mutants were isolated previously (BRENNER 1974 ; PARK and HORVITZ 1986 ; SHREFFLER et al. 1995 ) and obtained from the Caenorhabditis Genetics Center (e15, e49, n491, e15lb145, n491n1193, and n491lb82).

Park and Horvitz originally described four classes of unc-1 mutants (PARK and HORVITZ 1986 ). Members of class I are semidominant coilers, members of class II are semidominant coilers and recessive kinkers, class III mutants are hypomorphic recessive kinkers, and class IV mutants are null recessive kinkers (the kinkers do not coil and move forward more poorly than backward). A representative of each class was tested for anesthetic sensitivity. Two different classes of unc-8 alleles have been described (PARK and HORVITZ 1986 ). The first class consists of semidominant kinker/coilers (the animals back poorly, first kinking and then coiling from both the tail and head ends). The phenotype of this class of mutants is distinguishable from those of unc-1 mutants; thus, epistasis can be established. Recessive loss-of-function alleles of unc-8 have been described as almost wild type, moving with a flattened wave form (TAVERNARAKIS et al. 1997 ).

Genetics:
All genetic crosses were performed as described by BRENNER 1974 . The least straightforward crosses are described below. The presence of both genes in all double homozygotes was confirmed by noncomplementation tests or, in the case of the dominant alleles, by reisolation of the original homozygous parents from crosses with N2. Temperature shifts to determine the temperature-sensitive period for unc-1 were done as described previously (HECHT et al. 1996 ), except that ether sensitivity and suppression of unc-79 were scored.

Double mutants of unc-1, unc-7, and unc-9 with unc-8(sd):
unc-1, unc-7, and unc-9 recessive mutants are kinkers that are distinguishable from the kinker/coiler unc-8 mutants. unc-1(n494) is a semidonimant coiler. The effects of this allele also are easily distinguished from the unc-8(sd) and are discussed separately at the end of this section. unc-8(sd)/+ males were mated to unc-1, unc-7 or unc-9 hermaphrodites. In the F₁ generation, we picked hermaphrodites that were mildly uncoordinated in phenotype (consistent with the unc-8/+ genotype). These were allowed to self-fertilize. In the F₂ generation, hermaphrodites that showed the recessive kinked phenotype, i.e., Unc-1, Unc-7, or Unc-9, were isolated. These hermaphrodites were allowed to self-fertilize. In the F₃ generation, offspring that exhibited the kinker/coiler phenotype (consistent with the homozygous unc-8/unc-8 genotype) were isolated. The presence of both the recessive kinker alleles and the unc-8 allele were proven by noncomplementation with the appropriate recessive allele or by segregation of the Unc-8-specific phenotype after mating to N2 males.

In the case of unc-1(n494), 30–40 unc-8(n491)/+ males were mated to 3 n494 lon-2(X) animals on a small spot plate. NonLon mild coiler hermaphrodites were picked in F₁(n491/+; ++/n494lon-2). In the F₂ generation, we picked both NonLon Unc-8s(n491/n491; ++/n494lon-2 or n491/n491+/+) and Lon Unc-1s(n491/+;n494lon-2/n494lon-2 or +/+;n494lon-2/n494lon-2). Both F₂ parents produced strong Lon kinkers in the F₃ generation. Similar crosses were done with unc-8(e15) and unc-8(e49); in each case, about one-fourth of the offspring in the F₃ generation were Lon Unc-8. The presence of both mutant alleles was proven by noncomplemenation with the appropriate alleles or by segregation of the appropriate allele-specific phenotypes after mating to N2 males.

Double mutants of unc-1, unc-7, and unc-9 with unc-8(lf):
Heterozygous unc-8(lf) males were made by mating N2 males to unc-8 hermaphrodites and were then mated to unc-1, unc-7, or unc-9 homozygotes. Wild-type hermaphrodites were isolated in the F₁ generation and allowed to self-fertilize. Hermaphrodites with the Unc-8 phenotype were isolated in the F₂ generation and again allowed to self-fertilize. Kinkers were then isolated in the F₃ generation. The presence of each mutation was then proven by noncomplementation with the appropriate allele.

unc-79;unc-1 double mutants:
unc-79;unc-1 double mutants were constructed as described previously. Briefly, N2 males were mated with unc-79 homozygotes, and unc-79/+ males were isolated. These males were mated to unc-1 animals, and the resulting wild-type offspring were collected in the F₁ generation. Animals with the Unc-79 phenotype were isolated in the F₂ generation (presumed homozygous unc-79 animals). The offspring of these animals were screened for the appearance of the kinked Unc-1 phenotype in the F₃ generation. The presence of both the unc-79 and unc-1 alleles was confirmed by noncomplementation with the appropriate alleles.

unc-79;unc-8 double mutants:
N2 males were mated with homozygous unc-79 dpy-17 III hermaphrodites. Male offspring (genotype unc-79 dpy-17/++) were then mated to the kinker/coiler unc-8 mutant of interest (e15, e49, or n491). Each of the kinker/coiler unc-8 alleles is semidominant, but the heterozygote is easy to separate from the homozygote. The resulting hermaphrodite offspring (genotypes unc-79dpy-17/++; unc-8/+ and +/+; unc-8/+) were allowed to self-fertilize, and Unc-8 animals were moved to separate plates. Offspring of the Unc-8 animals were screened for appearance of the Dpy phenotype. Because dpy-17 and unc-79 map 1.5 cM apart, it was likely that the unc-79 allele would cosegregate with the dpy-17 allele. Several DpyUnc-8 animals were picked, one per plate, and allowed to propagate to prove that they were homozygous for the selected markers. The presence of both unc-79 and unc-8 mutant alleles was then proven by noncomplementation with the unc-79 or by segregation of unc-8 in the proper ratio.

As reported in RESULTS, not all unc-8 loss-of-function alleles cause identical phenotypes. e15lb145 caused the phenotype reported earlier (TAVERNARAKIS et al. 1997 ), which is decreased amplitude and wave length of the normal sinusoid motion of the nematode. However, another loss-of-function allele, n491n1193, resulted in an abnormal motion described as fainting. These animals move in short bursts of normal motion, but then abruptly stop (i.e., faint), wait a few seconds, and then repeat the motion. This phenotype is identical to the abnormal motion associated with alleles of unc-79 and unc-80. A third nonkinked homozygote of unc-8(e15lb82) has a phenotype similar to that of e15lb145. However, heterozygotes (e15lb82/+) are severely kinked (SHREFFLER et al. 1995 ).

In the case of the nonkinked alleles of unc-8(lf), different approaches for constructing double mutants were used. Again, unc-79 dpy-17/++ males were mated to the unc-8 homozygote of interest. In the case of e15lb145, wild-type F₁ animals were picked and allowed to self-fertilize. Both Unc-8 and Unc-79 Dpy-17 animals were picked in the F₂ generation. The F₂ Unc-8 animals gave rise to Dpy non-Unc-79 offspring. The F₂ Unc-79 Dpy-17 animals gave rise to Dpy non-Unc-79 animals in the F₃ generation. A similar approach was used for unc-8(n491 n1193), the allele of unc-8 that, like unc-79, is associated with fainting. In this case, we picked non-Dpy fainters in the F₂ generation (presumed unc-8 homozygotes) and screened for Dpys in the F₃ generation. In crosses of unc-79 dpy-17/++ with e15lb82, kinked animals were picked in the F₁. Animals with decreased amplitude of motion were picked in the F₂, and Dpys were picked in the F₃. The presence of both unc-8 and unc-79 alleles was proven in all cases by noncomplementation or, in the case of unc-8(e15lb182), by the presence of kinked animals in the F₁ generation after mating with N2.

unc-1::EGFP reporter and translational fusion constructs:
Primers containing unique sites for BamHI and XmaI were used to amplify 4.7 kb of genomic sequence upstream of the first ATG of the unc-1 coding sequence (not inclusive). This region was predicted to contain the promoter for the unc-1 gene. The primer sequences (5'-3') were as follows: U1pGFPF1, AAT GGA TCC GGC CTC TGT TAC TAA; U1pGFPR1, TTA CCC GGG TTA CCT GGA AAA CTT. The PCR product was purified using the Qiaquick PCR purification kit (QIAGEN, Chatsworth, CA) and sequentially digested with BamHI and XmaI (with an intermediate phenol/chloroform extraction and ethanol precipitation step). The resulting fragment was then cloned into the BglII and XmaI sites of EGFP-1 (CLONTECH, Palo Alto, CA).

A second GFP construct contained the unc-1 coding sequence (minus the codons for the three terminal amino acids) plus 2.4 kb of genomic sequence upstream of the first ATG of the unc-1 coding sequence fused in frame upstream of the GFP coding sequences. The primer sequences (5'-3', shown below) contained sites for the enzymes BglII and XmaI; the insert thus generated was ligated into the corresponding BglII and XmaI sites of the EGFP-1 vector. U1pGFPF3, TAT GAG CCG AAA GAA GAT CTC CCT; U1FPR3, TAA TAT TAC CCG GGT TTT CAT AAA TGC TCC. The accuracy of the cloned sequences was ascertained by restriction digestion and by Southern blotting using a genomic fragment from the unc-1 gene as the probe. The frame of the translational fusion was verified by sequencing across the junctions of the insertion.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

UNC-1 controls sensitivity to volatile anesthetics:
Wildtype C. elegans has a characteristic sinuous motion when moving on an agar plate. unc-1 alleles cause two main phenotypes: recessive kinkers and semidominant coilers that have been categorized into four classes (PARK and HORVITZ 1986 ). The null alleles e580 and e719 have severely kinked motions consisting of extra bends along their bodies as they move. Each allele results in an increased sensitivity to diethylether and a slightly decreased sensitivity to halothane. Each mutant is otherwise wild type in its response to volatile anesthetics (MORGAN et al. 1990 ; RAJARAM et al. 1998 ). e719 and e580 also suppress the hypersensitivity of unc-79 to halothane. We determined the EC₅₀'s for representatives of the four classes of unc-1 alleles in both diethylether and halothane (Table 1). All have some increase in sensitivity to diethylether, with the largest increase seen in the dominant negative allele n494 and the null allele e580. Other alleles (hs1-5, fc16, fc17, fc53, and n1598) had sensitivities representative of their classes (data not shown). When exposed to halothane, only n494 and e580 differed in sensitivity from the wild type; however, e580 is slightly resistant to halothane, while n494 is hypersensitive. In fact, n494 is hypersensitive to all volatile anesthetics in which it has been tested (data not shown). The double mutant unc-79;unc-1(n494) was also exposed to halothane. This animal was a sluggish coiler and much more sensitive (EC₅₀ 0.5%) to halothane than either single mutant (EC₅₀ 1.0% for unc-79, 0.9% for n494).Thus, unc-1 can be altered to either increase or decrease sensitivity to halothane and to either suppress or add to the halothane hypersensitivity of unc-79. The abnormal sensitivities of both e580 and n494 are rescued by microinjection of the unc-1 gene. Thus, the abnormal sensitivities of these alleles specifically result from mutations in unc-1 (RAJARAM et al. 1998 ).

UNC-1 expression:
Temperature shifts at any time during development affect both the sensitivity of unc-1 mutants to diethylether and the ability of unc-1 to suppress the abnormal sensitivities of unc-79 mutants (data not shown). This is consistent with data showing that temperature shifts at any time during development also altered the kinked phenotype conferred by the ts alleles (HECHT et al. 1996 ). Therefore, because UNC-1 must be present for the entire life of the nematode and can affect anesthetic sensitivity in a variety of ways, we postulated that it forms part of a molecular target for volatile anesthetics. Because UNC-1 is a potential target for anesthetics, we localized the expression pattern of UNC-1 to further understand its function. We wished to determine whether UNC-1 was primarily expressed in nerve, muscle, or both.

Reporter construct: A reporter construct of the unc-1 promoter and GFP showed that unc-1 is expressed broadly in the nervous system. Consistent with the temperature shift experiments, GFP expression is seen at all postembryonic stages of development (Figure 1) as well as in eggs shortly before hatching. GFP expression is seen in many cells of the nerve ring and in the motor neurons of the ventral cord. The vulval muscles, but not the body wall muscles, also express the reporter.

View larger version (64K): In this window In a new window Download PPT slide   Figure 1. (A) unc-1::GFP reporter construct expressed in an N2 adult worm. Note that the reporter is expressed widely in the nervous system, primarily in the circumpharyngeal nerve ring (small arrow) and in the ventral cord (large arrowhead). No staining is seen in body wall muscle. Note also that the vulval muscles express the construct. The worm appears twisted because the construct is coinjected with a dominant roller gene, rol-6, as a marker for successful injections. (B) The same construct as in A, but expressed in a larval worm. Again, fluorescence is widely expressed in the nervous system. The large arrowhead indicates the ventral nerve cord; the small arrow indicates neurons in the head and pharyngeal nerve ring. (C) Expression of a translational fusion construct containing the unc-1 gene fused to GFP. This construct fully rescued the Unc-1 phenotype. Note the punctate staining along nerve tracts, as noted by the arrowhead. As with the reporter construct, fluorescence was seen primarily in the nerve ring and in the ventral nerve cord, with no fluorescence seen in body wall muscle.

Translational fusion construct: We also constructed a fusion protein in which the entire GFP coding sequence was fused in frame to the unc-1 gene at the C terminus of UNC-1. This construct lacked the three terminal amino acids of UNC-1, but contained an eight-amino-acid spacer between UNC-1 and EGFP to minimize steric interference (STEWART et al. 1993 ). Because the expression of this chimeric protein is driven by the unc-1 promoter, it is expected to accurately reveal the subcellular localization of UNC-1. The UNC-1::GFP fusion construct rescued both the uncoordinated motion and altered sensitivities of unc-1(e580) mutants, indicating that the chimeric protein is functional in at least the minimum distribution necessary for UNC-1 function. UNC-1 localization is depicted in Figure 1C. Note that the protein appears extensively in the nerve ring and in a punctate pattern along nerve tracts. Body wall muscle did not express GFP, which suggests that UNC-1 affects anesthetic sensitivity primarily through the nervous system. Staining animals carrying the UNC-1::GFP fusion molecule with anti-GFP antibodies indicates that a low level of expression may exist in the body muscle of this overexpression system, but confirms that the predominant expression is in neurons (data not shown).

As noted in the Introduction, in C. elegans the MEC-2 protein is a partial homologue of UNC-1. Chalfie has suggested that MEC-2 gates a passive sodium channel of the degenerin family (HUANG et al. 1995 ). Degenerins, in turn, are members of a larger family of channels known as ENaCs. UNC-8, another ENaC-like subunit, is expressed predominantly in the ventral cord of C. elegans, in an expression pattern similar to UNC-1 (TAVERNARAKIS et al. 1997 ). Alleles of unc-8 confer phenotypes reminiscent of unc-1 alleles. Therefore, we hypothesized that UNC-1 may regulate a channel that includes UNC-8 as a subunit. We examined the genetic interactions of unc-8 mutations with several alleles of unc-1 and other genes in the pathway controlling sensitivity to volatile anesthetics.

Volatile anesthetics and unc-8:
As noted above, unc-1 alleles confer two main phenotypes: kinkers and coilers. unc-8 mutations also result in two distinct phenotypes. Semidominant Uncs are mildly uncoordinated as heterozygotes but more severe kinkers/coilers as homozygotes. Recessive loss-of-function alleles confer a subtle defect in locomotion that produces a "flat wave" track on agar plates (TAVERNARAKIS et al. 1997 ). The phenotypes associated with unc-8 alleles, n491, e15, and e49, were reminiscent of, though not identical to, the phenotypes associated with unc-1 alleles.

Each unc-8(sd) allele conferred an increased sensitivity to diethylether similar to that of unc-1 and unc-79. Like e580, unc-8(n491) and unc-8(e15) mutants were mildly resistant to halothane (Figure 2A; Table 2). The loss-of-function allele unc-8(e15lb145) conferred a small increase in sensitivity to all volatile anesthetics tested. These abnormal sensitivities cosegregate with the Unc-8 phenotypes described previously.

View larger version (15K): In this window In a new window Download PPT slide   Figure 2. Dose-response curves of several C. elegans strains in halothane using immobility as the endpoint. Procedures for determining dose-response curves, EC50's and standard errors are described in the text. (A) Comparison of two different unc-8 mutants with N2. Note that n491 is mildly resistant to halothane, while n491n1193 has an increased sensitivity. EC50's (standard errors, probability that EC50 differs from that of N2): N2, 3.2 (0.1); n491, 3.5 (0.2, P < 0.05); n491n1193; 1.0 (0.2, P < 0.001). (B) Suppression of unc-8(n491n1193) by unc-1(e580). EC50's: N2, 3.2 (0.1); n491n1193, 1.0 (0.2, P < 0.01); n491n1193;e580, 3.2 (0.2). n491n1193 and e15lb145 (see text) were isolated by Horvitz and Wolinsky, respectively, as revertants of the semidominant kinker/coiler alleles n491 and e15 in a screen to identify the loss-of-function phenotype. (C) Suppression of unc-79 by n491. Note that n491 suppresses the sensitivity to halothane of unc-79 but not the fainting phenotype. The unc-79;4nc-8(n491) double mutant is a kinker/coiler like n491 alone. EC50's: N2, 3.2 (0.1); ec1, 1.0 (0.1, P < 0.001); ecl;n491, 3.3 (0.2, P > 0.5).

Another loss-of-function allele, unc-8(n491n1193), confers a phenotype that had been previously overlooked. Specifically, n491n1193 is a fainter, in that it moves forward or backward a short distance in a normal motion and then stops abruptly. When stimulated, n491n1193 animals will continue in this pattern, never moving more than a few body lengths before "fainting." This phenotype is identical to that of unc-79 animals. unc-79 animals are hypersensitive to halothane and to diethylether, slightly resistant to enflurane, and unchanged in sensitivity to isoflurane. n491n1193 also responds to volatile anesthetics identically to unc-79 (Figure 2A; Table 2). n491n1193 fully complements two separate alleles of unc-79, ec1 and e1068, and the weaker fainter unc-80(e1272) V. Sequence data confirmed that n491n1193 carries a new mutation in unc-8 (N. TAVERNARAKIS, personal communication). In addition, unc-79 is fully rescued by fragments of the cosmid E03A3 from chromosome III, while the abnormal sensitivity of n491n1193 maps to chromosome IV (M. M. SEDENSKY and P. G. MORGAN, unpublished results). Therefore, n491n1193 is not an allele of unc-79 or unc-80.

n491n1193 and e15lb145 were placed over a deletion in the region as well as over each other. e15lb145/stDf8 had the same phenotypes as the e151b145/e151b145 homozygote. In contrast, n491n1193/stDf8 did not faint, was not strongly sensitive to halothane, and moved like e15lb145. We conclude that e15lb145 is the true null, and that n491n1193 is either a hypomorphic allele or a neomorph with complicated genetic interactions. The assignment of e15lb145 as the null is in agreement with the findings of TAVERNARAKIS et al. 1997 . In summary, different alleles of unc-8 confer phenotypes similar to those associated with the kinked alleles of unc-1, which are suppressors of unc-79, or with unc-79 itself. The phenotype of the mutant in air predicted its behavior in volatile anesthetics. Like unc-1, unc-8 can be altered to either increase or decrease sensitivity to volatile anesthetics.

Interactions of unc-8 with other genes:
Several double mutants of unc-1 and unc-8 alleles were constructed to test for epistasis with regard to their motion in air. The motions conferred by the unc-8(sd) alleles are distinct from the uncoordinated motions associated with the loss-of-function and semidominant alleles of unc-1. Therefore, it was possible to score the phenotype of the double mutants as being like one or the other parent. In general, the semidominant alleles of unc-8 were epistatic to unc-1 (Table 3). n491 was also epistatic to the other kinked suppressors of unc-79 and unc-80, unc-7 and unc-9. The exception to this rule was that the double mutant unc-8(n491);unc-1(n494) (both single mutants are coilers) was a very strong kinker with no tendency to coil. unc-8(n491);unc-1(n494) had a normal sensitivity to halothane, indicating that n491 suppresses the increased sensitivity of n494.

As noted above, unc-8(n491n1193) is identical in behavior to unc-79. Because unc-1(e580) suppresses the abnormal anesthetic sensitivity of unc-79, we tested whether kinked alleles of unc-1 would also suppress n491n1193. As shown in Figure 2B, unc-1(e580) does suppress the abnormal sensitivity of n491n1193, as do unc-7 and unc-9 (Table 4). This is in contrast to the previous data showing that unc-8(sd) alleles were generally epistatic to kinked unc-1 alleles. Suppression or epistasis is determined by the phenotype rather than by the genotype.

Like the kinked alleles of unc-1, unc-8(sd) alleles also suppress the abnormal anesthetic sensitivity of unc-79 (Figure 2C; Table 4). Because unc-8 can be a fainter or a suppressor of the altered sensitivity of fainters, we tested if the function of unc-8 was necessary for the kinked phenotype or the fainter phenotype. The unc-8(0) allele e15lb145 was a partial suppressor of the kinked motion of unc-1(e580) and unc-1(e114), a kinked hypomorphic mutant. Thus, unc-1 is partially dependent on the function of unc-8. In addition, unc-79;e151b145 was identical to e151b145 in air and in halothane, i.e., e15lb145 completely suppressed both the altered halothane sensitivity of unc-79 and the fainting phenotype in air (Table 3 and Table 4). e15lb145 also suppressed unc-80 (data not shown). We also tested whether unc-8(n491lb82), another loss-of-function allele, suppressed unc-79. n491lb82 is a kinker/coiler as a heterozygote and exhibits the loss-of-function phenotype as a homozygote. As a heterozygote, it suppresses the abnormal sensitivity of unc-79, but as a homozygote, it does not suppress the abnormal sensitivity or the fainting phenotype of unc-79 (data not shown). Double mutants of unc-79 and n491n1193 showed no additive effects.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

It is of interest to know how UNC-1 might affect anesthetic sensitivity. Of course, mutations that primarily affect muscle could alter anesthetic sensitivity when using locomotion as an endpoint. This is very unlikely for unc-1, because UNC1:GFP fusion was primarily expressed in the nervous system. The fusion construct is also interesting, because it is consistent with UNC-1 functioning in axonal membranes.

Because stomatin-like proteins are thought to interact with sodium channels, any ENaCs that are expressed in the same cells as UNC-1 are potentially important in the function of UNC-1. UNC-8 is expressed in a pattern similar to UNC-1. The findings that mutations in UNC-8 also affect anesthetic sensitivity and confer phenotypes similar to UNC-1 raise the possibility that they affect the same physiologic function. The additional finding that n491 and n494 double mutants give a novel phenotype is consistent with the hypothesis that these protein products interact directly. In addition, the interactions between unc-79, unc-80, and unc-8 indicate that UNC-79 and UNC-80 protein products may function through the UNC-8 subunit. Thus, in several ways, UNC-8 is implicated as an important member of a pathway controlling anesthetic sensitivity.

Mammalian stomatin is known to form oligomers and interact with other proteins (MAYER et al. 1998 ; SNYERS et al. 1998 ). Thus, it was of interest to determine whether UNC-1 also showed interactions with other gene products in C. elegans. UNC-1 is expressed primarily in the nervous system of C. elegans, being seen in ventral cord motorneurons that send axons to both the ventral and dorsal cords. Translational fusion constructs between GFP and UNC-1 fully rescued the Unc-1 phenotype and showed a punctate distribution of the fusion product predominantly along the major nerve tracts. We have linked the function of UNC-1 to that of a neuronal channel protein, UNC-8, an ENaC-like subunit. Multiple lines of evidence suggest that unc-1, unc-79, and unc-8 affect the same physiologic function: (1) unc-1(lf) alleles confer similar phenotypes in air to those associated with unc-8(sd) alleles; (2) unc-1(lf) is similar to unc-8(sd) in sensitivity to multiple volatile anesthetics; (3) unc-8(n491);unc-1(n494) exhibits a novel phenotype compared with either single mutant; (4) unc-1(lf) and unc-8(sd) interact similarly with unc-79 and unc-80 in sensitivity to volatile anesthetics; (5) different unc-8(lf) alleles can either confer a phenotype similar to that of unc-79 or suppress unc-79 (and unc-80); and (6) UNC-1 is spatially expressed in a pattern similar to that of UNC-8. However, UNC-1 appears to be more widely expressed in the nervous system than does UNC-8. Thus, UNC-1 probably has other functions than those involving UNC-8, and the genetic interactions are more compelling than the expression patterns alone.

These findings are particularly exciting, because the other stomatin homologue, MEC-2, is thought to directly regulate a degenerin channel (GU et al. 1996 ). It is possible that this type of interaction may be a general one. Response to volatile anesthetics may be controlled in a similar fashion by the interaction of UNC-1, UNC-8 and UNC-79. The similarity in phenotypes conferred by the unc-8 (semidominant kinker/coiler) and unc-1 (recessive kinker) alleles is consistent with this model, because both could lead to increased sodium flux through the channel. In addition, the interactions between unc-79, unc-80, unc-1, and unc-8 alleles indicate that fainting, kinking, and coiling result from different changes in the physiologic function controlled by these genes. It is unclear whether these genes exert a direct or indirect effect on anesthetic response. However, several observations make a direct effect more likely: (1) unc-79 mutations alter halothane binding in C. elegans (R. G. ECKENHOFF, personal communication); (2) unc-79 alters responses to stereoisomers of halothane differently than the racemate (SEDENSKY et al. 1994 ); (3) the null allele of unc-8 does not have as strong an effect on sensitivity as does n491n1193; and (4) suppressors of the kinked phenotype of unc-1 do not alter the ability of these mutations to suppress unc-79 (P. G. MORGAN and M. M. SEDENSKY, unpublished results).

In conclusion, we have shown that unc-1 encodes a neuronal protein capable of altering anesthetic sensitivity in an allele-specific fashion. We hypothesize that unc-1 and unc-8 (together with unc-79) interact to control ion flux through a sodium channel that is crucial in mediating the response to volatile anesthetics. In mammals, both stomatin-like proteins and members of the ENaC family have been identified in multiple tissues (GARCIA-ANOVEROS et al. 1997 ; MAYER et al. 1998 ; SEIDEL and PROHASKA 1998 ), including the central nervous system. In C. elegans, the sequencing consortium has identified >20 homologues of ENaC subunits. If several of these are sensitive to volatile anesthetics, such a family of proteins may give rise to multiple sites of action for volatile anesthetics. The potential complexity of the expression of ENaCs in mammals could easily explain the widespread effect of volatile anesthetics on diverse physiologic processes in humans. Equally important, others have identified protein complexes that influence behavior in volatile anesthetics (RAJARAM et al. 1998 ; FROEMMING et al. 1999 ; KAYSER et al. 1999 ; VAN SWINDEREN et al. 1999 ). Because such complexes are likely to have hydrophobic pockets and be susceptible to steric alterations, they represent a general type of potential target of volatile anesthetics.

	ACKNOWLEDGMENTS

We thank Janet Duerr, Helen Salz, Monica Driscoll, and Nektarios Tavernarakis for sharing unpublished data and critical discussions, as well as Carl Johnson, Eric Jorgensen, and Ralph Hecht for sharing their unc-1 alleles. We also thank Helmut Cascorbi, John Humphrey, Ernst-Bernhard Kayser, Helen Salz, and Monica Driscoll for reviewing the manuscript. This work was supported in part by National Institutes of Health grant GM45402.

Manuscript received May 7, 1999; Accepted for publication August 30, 1999.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ANTON, A. H., A. I. BERK, and C. H. NICHOLLS, 1992  The anesthetic effect of alcohols and alkanes in Caenorhabditis elegans (C.e).. Res. Comm. Chem. Pathol. Pharmacol. 78(1):69-83[Medline].

BAKER, R., C. MELCHIOR, and R. DEITRICH, 1980  The effect of halothane on mice selectively bred for differential sensitivity to alcohol. Pharmacol. Biochem. Behav. 12:691-695[Medline].

BRENNER, S., 1974  The genetics of Caenorhabditis elegans.. Genetics 77:71-94[Abstract/Free Full Text].

CHALFIE, M. and M. AU, 1989  Genetic control of differentiation of the Caenorhabditis elegans touch receptor neurons. Science 243:1027-1033[Abstract/Free Full Text].

CHALFIE, M. and E. WOLINSKY, 1990  The identification and suppression of inherited neurodegeneration in Caenorhabditis elegans.. Nature 345:410-416[Medline].

CHALFIE, M., Y. TU, G. EUSKIRCHEN, W. W. WARD, and D. C. PRASHER, 1994  Green fluorescent protein as a marker for gene expression. Science 263:802-805[Abstract/Free Full Text].

CROWDER, C. M., L. D. SHEBESTER, and T. SCHEDL, 1996  Behavioral effects of volatile anesthetics in Caenorhabditis elegans.. Anesthesiology 85:901-912[Medline].

DRISCOLL, M. and M. CHALFIE, 1991  The mec-4 gene is a member of a family of Caenorhabditis elegans genes that can mutate to induce neuronal degeneration. Nature 349:588-593[Medline].

ECKENHOFF, R. G., 1996a  An inhalational anesthetic binding domain in the nicotinic acetylcholine receptor. Proc. Natl. Acad. Sci. USA 93:2807-2810[Abstract/Free Full Text].

ECKENHOFF, R. G., 1996b  Amino acid resolution of halothane binding sites in serum albumin. J. Biol. Chem. 271:15521-15526[Abstract/Free Full Text].

FRANKS, N. P. and W. R. LIEB, 1994  Molecular and cellular mechanisms of general anaesthesia. Nature 367:607-614[Medline].

FROEMMING, G. R., D. J. DILLANE, and K. OHLENDIECK, 1999  Complex formation of skeletal muscle Ca²⁺-regulatory membrane proteins by halothane. Eur. J. Pharmacol. 365:91-102[Medline].

GAMO, S., M. OGAKI, and E. NAKASHIMA-TANAKA, 1981  Strain differences in minimum anesthetic concentrations in Drosophila melanogaster.. Anesthesiology 54:289-291[Medline].

GARCIA-ANOVEROS, J., B. DERFLER, J. NEVILLE-GOLDEN, B. T. HYMAN, and D. P. COREY, 1997  BNaC1 and BNaC2 constitute a new family of human neuronal sodium channels related to degenerins and epithelial sodium channels. Proc. Natl. Acad. Sci. USA 94:1459-1463[Abstract/Free Full Text].

GU, G., G. A. CALDWELL, and M. CHALFIE, 1996  Genetic interactions affecting touch sensitivity in Caenorhabditis elegans.. Proc. Natl. Acad. Sci. USA 93:6577-6582[Abstract/Free Full Text].

HECHT, R. M., T. V. NORMAN, and W. JONES, 1996  A novel set of uncoordinated mutants in Caenorhabditis elegans uncovered by cold-sensitive mutations. Genome 39:459-464[Medline].

HUANG, M., G. GU, E. L. FERGUSON, and M. CHALFIE, 1995  A stomatin-like protein necessary for mechanosensation in C. elegans.. Nature 378:292-295[Medline].

KAYSER, E.-B., P. G. MORGAN, and M. M. SEDENSKY, 1999  GAS-1: a mitochondrial protein controls sensitivity to volatile anesthetics in C. elegans.. Anesthesiology 90:545-554[Medline].

KOBLIN, D. D., 1990 How do inhaled anesthetics work? pp. 67–99 in Anesthesia, Ed. 4, edited by R. D. MILLER. Churchill Livingstone, New York.

KOBLIN, D. D. and J. E. DEADY, 1981  Anaesthetic requirement in mice selectively bred for differences in ethanol sensitivity. Br. J. Anaesth. 53:5-10[Abstract/Free Full Text].

KRISHNAN, K. S. and H. A. NASH, 1990  A genetic study of the anesthetic response: mutants of Drosophila melanogaster altered in sensitivity to halothane. Proc. Natl. Acad. Sci. USA 87:8632-8636[Abstract/Free Full Text].

MACIVER, M. B. and J. J. KENDIG, 1991  Anesthetic effects on resting membrane potential are voltage-dependent and agent-specific. Anesthesiology 74:83-88[Medline].

MAYER, H., U. SALZER, J. BREUSS, S. ZIEGLER, and A. MARCHLER-BAUER et al., 1998  Isolation, molecular characterization, and tissue-specific expression of a novel putative G protein-coupled receptor. Biochim. Biophys. Acta 1395:301-308[Medline].

MCCRAE, A. F., E. J. GALLAHER, P. M. WINTER, and L. L. FIRESTONE, 1993  Volatile anesthetic requirements differ in mice selectively bred for sensitivity or resistance to diazepam: implications for the site of anesthesia. Anesth. Analg. 76:1313-1317[Medline].

MEYER, H. H., 1899  Theorie der Alkoholnarkose. Arche Exp. Pathol. Pharmakol. 42:109.

MILLER, K. W., L. L. FIRESTONE, J. K. ALIFIMOFF, and P. STREICHER, 1989  Nonanesthetic alcohols dissolve in synaptic membranes without pertubing their lipids. Proc. Natl. Acad. Sci. USA 86:1084-1087[Abstract/Free Full Text].

MORGAN, P. G. and H. F. CASCORBI, 1985  Effect of anesthetics and a convulsant on normal and mutant Caenorhabditis elegans.. Anesthesiology 62:738-744[Medline].

MORGAN, P. G. and M. M. SEDENSKY, 1994  Mutations conferring new patterns of sensitivity to volatile anesthetics in C. elegans.. Anesthesiology 81:888-898[Medline].

MORGAN, P. G., M. M. SEDENSKY, and P. M. MENEELY, 1990  Multiple sites of action of volatile anesthetics in C. elegans.. Proc. Natl. Acad. Sci. USA 87:2965-2969[Abstract/Free Full Text].

OVERTON, E., 1901 Studien Über die Narkose. Verlag von Guston Fischer, Jena.

PARK, E. C. and H. R. HORVITZ, 1986  Mutations with dominant effects on the behavior and morphology of the nematode Caenorhabditis elegans.. Genetics 113:821-852[Abstract/Free Full Text].

RAJARAM, S., M. M. SEDENSKY, and P. G. MORGAN, 1998  A stomatin homologue controls sensitivity to volatile anesthetics in C. elegans.. Proc. Natl. Acad. Sci. USA 95:8761-8766[Abstract/Free Full Text].

SEDENSKY, M. M. and P. M. MENEELY, 1987  Genetic analysis of halothane sensitivity in C. elegans.. Science 236:952-954[Abstract/Free Full Text].

SEDENSKY, M. M., H. F. CASCORBI, J. MEINWALD, P. RADFORD, and P. G. MORGAN, 1994  Genetic differences affecting the potency of stereoisomers of halothane. Proc. Natl. Acad. Sci. USA 91:10054-10058[Abstract/Free Full Text].

SEIDEL, G. and R. PROHASKA, 1998  Molecular cloning of hSLP-1, a novel human brain-specific member of the band 7/MEC-2 family similar to Caenorhabditis elegans UNC-24. Gene 225:23-29[Medline].

SHREFFLER, W., T. MAGARDINO, K. SHEKDAR, and E. WOLINSKY, 1995  The unc-8 and sup-40 genes regulate ion channel function in Caenorhabditis elegans motorneurons. Genetics 139:1261-1272[Abstract].

SNYERS, L., E. UMLAUF, and R. PROHASKA, 1998  Oligomeric nature of the integral membrane protein stomatin. J. Biol. Chem. 273:17221-17226[Abstract/Free Full Text].

STEWART, G. W., A. C. ARGENT, and B. C. J. DASH, 1993  Stomatin: a putative cation transport regulator in red cell membrane. Biochim. Biophys. Acta 1225:15-25[Medline].

TAVERNARAKIS, N., W. SHREFFLER, S. WANG, and M. DRISCOLL, 1997  unc-8, a DEG/ENaC family member, encodes a subunit of a candidate mechanically gated channel that modulates C. elegans locomotion. Neuron 18:107-119[Medline].

VAN SWINDEREN, B., O. SAIFEE, L. SHEBESTER, R. ROBERSON, and M. L. NONET et al., 1999  A neomorphic syntaxin mutation blocks volatile-anesthetic action in Caenorhabditis elegans.. Proc. Natl. Acad. Sci. USA 96:2479-2484[Abstract/Free Full Text].

WAUD, D. R., 1970  On biological assays involving quantal responses. J. Pharmacol. Exp. Ther. 183:577-607[Abstract/Free Full Text].

This article has been cited by other articles:

				B. T. Carroll, G. R. Dubyak, M. M. Sedensky, and P. G. Morgan Sulfated Signal from ASJ Sensory Neurons Modulates Stomatin-dependent Coordination in Caenorhabditis elegans J. Biol. Chem., November 24, 2006; 281(47): 35989 - 35996. [Abstract] [Full Text] [PDF]

				H. Yokoyama and I. Matsui A Novel Thermostable Membrane Protease Forming an Operon with a Stomatin Homolog from the Hyperthermophilic Archaebacterium Pyrococcus horikoshii J. Biol. Chem., February 25, 2005; 280(8): 6588 - 6594. [Abstract] [Full Text] [PDF]

				P. Syntichaki and N. Tavernarakis Genetic Models of Mechanotransduction: The Nematode Caenorhabditis elegans Physiol Rev, October 1, 2004; 84(4): 1097 - 1153. [Abstract] [Full Text] [PDF]

				A. H. Hawasli, O. Saifee, C. Liu, M. L. Nonet, and C. M. Crowder Resistance to Volatile Anesthetics by Mutations Enhancing Excitatory Neurotransmitter Release in Caenorhabditis elegans Genetics, October 1, 2004; 168(2): 831 - 843. [Abstract] [Full Text] [PDF]

				M. M. Sedensky, J. M. Siefker, J. Y. Koh, D. M. Miller III, and P. G. Morgan A stomatin and a degenerin interact in lipid rafts of the nervous system of Caenorhabditis elegans Am J Physiol Cell Physiol, August 1, 2004; 287(2): C468 - C474. [Abstract] [Full Text] [PDF]

				P. Nagele, L. B. Metz, and C. M. Crowder Nitrous oxide (N2O) requires the N-methyl-D-aspartate receptor for its action in Caenorhabditis elegans PNAS, June 8, 2004; 101(23): 8791 - 8796. [Abstract] [Full Text] [PDF]

				N. L. Hiller, T. Akompong, J. S. Morrow, A. A. Holder, and K. Haldar Identification of a Stomatin Orthologue in Vacuoles Induced in Human Erythrocytes by Malaria Parasites: A ROLE FOR MICROBIAL RAFT PROTEINS IN APICOMPLEXAN VACUOLE BIOGENESIS J. Biol. Chem., November 28, 2003; 278(48): 48413 - 48421. [Abstract] [Full Text] [PDF]

				D. Rubin and F. Ismail-Beigi Distribution of Glut1 in detergent-resistant membranes (DRMs) and non-DRM domains: effect of treatment with azide Am J Physiol Cell Physiol, August 1, 2003; 285(2): C377 - C383. [Abstract] [Full Text] [PDF]

				K. Kobayakawa, R. Hayashi, K. Morita, K. Miyamichi, Y. Oka, A. Tsuboi, and H. Sakano Stomatin-Related Olfactory Protein, SRO, Specifically Expressed in the Murine Olfactory Sensory Neurons J. Neurosci., July 15, 2002; 22(14): 5931 - 5937. [Abstract] [Full Text] [PDF]

				H. A. Nash In vivo genetics of anaesthetic action Br. J. Anaesth., July 1, 2002; 89(1): 143 - 155. [Abstract] [Full Text] [PDF]

				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]

				M. M. Sedensky, J. M. Siefker, and P. G. Morgan Model Organisms: New Insights Into Ion Channel and Transporter Function. Stomatin homologues interact in Caenorhabditis elegans Am J Physiol Cell Physiol, May 1, 2001; 280(5): C1340 - C1348. [Abstract] [Full Text] [PDF]