Volatile general anesthetics produce a distinctive change in function of the nervous system, summarized as a loss of consciousness. Despite profound changes, much nervous system activity persists during anesthesia. As a complement to world-wide studies at the biochemcial and physiological level, this project uses a genetic approach to probe the action of general anesthetics. The isolation and characterization of mutants that have altered sensitivity to anesthetics could be of value in two ways. First, the mutations could identify genes that are important for neural function, especially those involved in higher-order integration. Second, some of the mutants might help to identify the direct molecular targets of the anesthetics, a subject, which has been in contention for the past 50 years. For our effort we have chosen the fruit fly, Drosophila melanogaster. In the past we showed that Drosophila responds to concentrations of a broad spectrum of volatiles similar to those effective in vertebrates. And halothane, a typical general anesthetic, is more potent at a circuit in the brain of the fly than a motor output circuit, again reminiscent of vertebrate anesthesia. Because the brain circuit used in our previous work is too complex to permit analysis at the cellular level, in our current work we have investigated two simpler circuits. We also have achieved the molecular identification of a gene that is an important determinant of anesthetic sensitivity of the fly. Effect of Shaker channels on the ERG The electroretinogram (ERG) is a mass extracellular recording taken from the surface of the eye in response to a flash of light. Primarily, it reports currents generated by voltage changes in two cell types: photoreceptors (PRs)and the large monopolar cells (LMCs) on to which they synapse In previous work we concluded that there was a single, specific alteration induced by modest concentrations of halothane. The steady photoreceptor potential and the LMC on-transient were unaffected. But, the off-transient was significantly affected, especially when the light pulse was kept short. The circuitry for the off-transient is less clearly established than that for the other components of the ERG. However, work in larger flies indicates that the off-transient relies on cholinergic stimulation of LMCs and strongly implicates a particular cell type, the amacrine, in this stimulation. We therefore think that it is likely that halothane is depressing either photoreceptor signaling to amacrines, amacrine signaling to LMCs or LMC reception of the cholinergic signal. An important new insight into the effect of halothane is our discovery that the identical effect is produced by inactivation of Shaker-encoded potassium channels. Three strong alleles of this gene have the same phenotype but two weaker alleles have little or no effect. Feeding wild-type flies an inhibitor of Shaker channels (4-aminopyridine) produces a very similar phenotype as genetic inactivation, so loss of the off-transient reflects a role for Shaker channels in the physiology of the eye rather than its development. Our finding not only raises the possibility that halothane acts via inhibition of Shaker channels, it provides a suitable experimental framework to test it. Anesthetic effects at the larval neuromuscular junction The simplest well-defined circuit in the fly is the neuromuscular junction of the 3rd larval instar. A Staff Scientist, David Sandstrom, has explored the utility of this circuit for anesthesia research. He first devised simple methods for observing larval behavior in the presence of volatiles. At concentrations comparable to those that affect adult flies, these agents caused larvae to stop moving. He also set up for two-electrode voltage clamp recording and made some fundamental observations on the way anesthetics affect the physiology of the neuromuscular junction. Although halothane provoked muscle contractures, isoflurane proved to be a workable agent. Concentrations as low as 0.2 mM produced two reliable effects. First, there is a modest increase in the frequency of spontaneous miniature excitatory junctional currents. This indicates that isoflurane somehow affects the release of vesicles from presynaptic termini of the larval motor neuron. An even more impressive effect of isoflurane on the larval motorneuron is a substantial increase in the threshold for eliciting action potentials. This phenomenon suggests that isoflurane perturbs the function of voltage-gated ion channels in fly axons. Both effects are reminiscent of those induced by anesthetics in higher organisms. The way has been opened for a cellular analysis of anesthetic effects and the way anesthesia mutants perturb them. Molecular identification of the har/na mutations The har mutations, specifically har38 and har85, were isolated in this lab by their altered behavioral response to halothane. We subsequently identified a classic mutation, na, as allelic to the har mutations. In virtually every assay of anesthesia sensitivity, these mutants have strong effects on responses to halothane. They are thus likely to be deeply involved in the anesthetic process, either as an anesthetic target or as important modulator of excitability in neurons that contain anesthetic targets. The har and na mutants show alterations in neural function even in the absence of anesthetics. The flies show a characteristic hesitant walking behavior and colleagues to whom we have sent these flies report that they have abnormal circadian rhythms and abnormal sleep/rest cycles. We have long been interested in identifying this gene at the molecular level. However, no transposon inserts have been isolated and an extensive survey of chromosomal aberrations failed to narrow the endpoints of the gene to a workable limit. With the publication of the Drosophila genome sequence, we learned that the entire region had a very low density of ORFs. We mounted an SSCP scan directed at the most promising candidates and discovered that har38 and har85 carried the identical change in one of the genes, CG1517. The change was in an intron but close to an exon/intron boundary. RT-PCR showed that the mutation depressed correct splicing and uncovered a cryptic splice site. Brute force sequencing of the na strain showed that CG1517 was again changed, suffering a 9 nt deletion in one exon. A genetic experiment showed that the alteration in sequence of CG1517 of har85 was tightly linked to the mutant phenotype. CG1517 is predicted to encode an ORF with strong sequence homology to the superfamily of sodium and calcium voltage-gated ion channels. But, it is sufficiently distinct to be placed in a sub-family that is separate from all the well-characterized channels. Other members of this sub-family are found in C. elegans and rats, and man. The rat ortholog has been reported to be widely expressed in brain but for none of these organisms is it known what processes depend on channel function. For this, the Drosophila anesthesia mutants can provide an answer.
Showing the most recent 10 out of 17 publications
