By definition, genes that confer a phenotype when their copy number is changed encode factors that are limiting for physiological processes. In search of dosage-sensitive genes that influence arousal and/or its suppression by anesthetics, we previously screened for changes in sensitivity to halothane in lines that were missing one copy of a large block of DNA. Are the hits we found genetically complex or simple? We had previously shown that the effect of the 75 kb ED1 deletion could be reversed by adding back a 15 kb segment of this region. As a further test, we subdivided the 540 kb ED4065 deletion and the 389 kb ED2247 deletion. In both cases, dosage sensitivity could be recapitulated by a 50 kb deletion. In each of these regions, is a single gene responsible for the anesthesia phenotype? Our test used transgenic rescue constructs containing genes from the region that were wildtype or individually mutated. This showed that, for ED1, the critical gene is glutathione-S-transferase S1 (GstS1). In Drosophila this enzyme functions as an antioxidant, implicating redox regulation as an important contributor to anesthetic sensitivity. For ED4065, our mutant analysis implicates the orc4 gene as the locus of haploinsufficiency. This gene is well known as a key component of the Origin Recognition Complex, a multi-protein ensemble first identified for its role in loading replication origins with key proteins. However, in both flies and vertebrates, the same complex is found in post-mitotic neurons, where it appears to play a role in microtubule organization. Further study of this gene is likely to be relevant to the mission of NIMH since the human ortholog of another origin complex subunit, ORC3, has been implicated in determining the degree of positive symptoms expressed by schizophrenics. Strong effects on anesthesia sensitivity are caused by mutations in the narrow abdomen (na) gene of Drosophila. We earlier showed that the na gene encodes a putative ion channel, one that is expressed broadly in the central nervous system of all metazoans and one whose human ortholog has been implicated in susceptibility to bipolar disorder. Frustrated in our attempts to confirm a published report that the vertebrate ortholog functions as a non-selective cation channel, we have been unable to ask whether channel function is affected by anesthetics. Instead, we have pursued the functional anatomy of channel's contribution to anesthetic sensitivity. The central question is whether the kind of arousal measured in our assays reflects the global state of excitability in the central nervous system or is subserved by more restricted circuitry. To answer this question we use the GAL4/UAS system, which permits expression of the channel under the control of various promoters. We previously found that broad neural expression rescued the anesthesia phenotypes but none of 20 drivers that produced restricted expression gave reliable rescue (promising results with the tim-GAL4 driver using the postural assay now appear to be due to a non-specific sensitization to stressors). Happily, a new survey of restricted GAL4 lines turned up one (c522) that provides substantial rescue of the halothane phenotype, suggesting that arousal depends on localized circuitry. However, channel expression under the control of c522 fails to rescue the sevoflurane phenotypes of na mutants. This result not only proves that the halothane rescue is not the result of non-specific effects on arousal, but it also tells us for the first time that different parts of the arousal circuitry, each of which depends on the NA channel, are differentially affected by these two agents. We previously reported that many mutations isolated on the basis of increased sensitivity to halothane have much smaller effects on sensitivity to other clinical anesthetics. From this result we infer the existence of a component of the nervous system that is preferentially affected by this drug. The ryanodine receptor (Ryr) has been reported to be unusually sensitive to halothane in vitro but, to our knowledge, the importance of this component to the anesthetic sensitivity of whole animals has not been studied. To evaluate this issue in Drosophila, we acquired two insertion mutations that putatively depress gene function. To remove adventitious unlinked mutations in these strains, we extensively outcrossed them to our laboratory control strain. We then crossed the two lines to each other and compared the resulting trans-heterozygote to its congenic control. The mutant strain was much less sensitive to halothane;its concentration-response curve was shifted to the right by more than 60%. When a transgenic copy of Ryr genomic DNA was introduced into the heterozygote, normal halothane sensitivity was restored. Moreover, addition of this transgene to a wild-type strain increased sensitivity to halothane, demonstrating that Ryr is a limiting factor for this drug. Of particular importance is our recent discovery that, using the GAL4/UAS system, expression of a Ryr cDNA in the nervous system is sufficient to overcome the mutant phenotype. Thus, although Ryr is most well known for its role in muscle physiology, its action in neurons and/or glia is vital for anesthetic responsiveness. In addition to focused studies of general anesthesia, we apply our expertise in Drosophila neurobiology to genes of general interest via collaborative studies. 1) Together with Dr. Brian Mozer of NHLBI, David Sandstrom is analyzing the role of Drosophila neuroligin, an ortholog of a gene implicated in autism, in the structure and function of a prototypical synapse, the larval neuromuscular junction. New developments include the demonstrations that that NLG protein is localized to the synapse and that reduction in neurotransmitter release in nlg mutants is paralleled by reduction in the number of synaptic boutons. Further evidence that nlg regulates synaptic structure is: a)mutations in nlg and its trans-synaptic partner neurexin alter the amplitude distributions of spontaneous miniature excitatory currents;b) mutations that cause synaptic overgrowth are associated with a dramatic reduction of NLG protein. 2) BAG3 proteins are co-chaperones implicated in cell survival, proliferation and migration. In collaboration with Dr. Victoria Virador of NCI, David Sandstrom is studying the Drosophila homolog of BAG3, starvin (stv). In recent work that used a panel of isoform-specific antibodies, they discovered expression of the protein in the CNS and at the neuromuscular junction (NMJ). Loss of stv function results in large NMJ synaptic boutons that are clumped together, while overexpression of STV protein causes abnormally long strings of synaptic boutons. Responses to repetitive stimulation are abnormal both in stv mutants and upon stv overexpression. These phenotypes provide fresh insight into the role of this poorly understood co-chaperone in neural development as well as regulated exocytosis. 3) TDP1 is an enzyme that is known to function in the repair of damaged DNA. In collaboration with the lab of Dr. Yves Pommier at NCI, using a mutant line in which TDP1 levels are severely reduced, Dongyu Guo is studying the role of TDP1 in the Drosophila nervous system. We previously noted that when fly larvae are raised on food containing the oxidative agent bleomycin, very few are able to undergo metamorphosis to the pupal stage. In the past year we have shown that this phenotype is reversed by expression of TDP1 under the control of a nervous system-specific promoter. This important result proves that: a) the effect of the mutation is indeed due to reduction of TDP1 function and b) that the critically sensitive tissue is the nervous system.
