We have identified and cloned a point mutation in the mouse homolog of the unc-79
gene, to which the allele name Lightweight
was assigned due to the resultant phenotypic traits. The Lightweight
mutation creates a premature stop codon in the middle of the unc-79 protein. We introgressed a ~1.4 Mb region surrounding the Lightweight
mutation fully onto the C57BL/6J genetic background. Lwt/Lwt
homozogotes are born at the expected Mendelian ratios and appear normal at birth, but survive a maximum of two days, similar to what has been reported for a knockout allele of this gene 
. Although Lightweight
heterozygotes exhibit a normal birthweight, soon thereafter and as adults, body weight is significantly less than their wild-type littermates. In addition, Lightweight
heterozygotes have increased lean tissue and reduced body fat in comparison to wild-type littermates. Furthermore, Lightweight
heterozygotes consume more food on a B6 background and more water on a B6D2 F1 background. Together, these results suggest a higher metabolic rate, increased protein turnover, or differences in energy usage.
By testing a control congenic strain in parallel, we have demonstrated that strain-specific polymorphisms from the introgressed D2 background do not produce or influence any of the phenotypes observed here. This possibility is often ignored 
but is an important concern due to the quantitative nature of several of the phenotypes tested. Although we have ruled out strain-specific polymorphisms, we cannot formally rule out the possibility that an additional ENU-induced mutation in our congenic strain might influence some of the phenotypes we have observed. The mutation rate of ENU depends on many factors, but hovers around 1×10−6 
; therefore, it is unlikely that more than one or two point mutations were induced within this 1.4 Mb congenic region, and furthermore, the likelihood of mutating a coding exon is very rare 
. We have sequenced the majority of coding exons in this congenic interval without finding any additional mutations other than Lightweight
. Complete resequencing of the whole 1.4 Mb interval would confirm the absence of other confounding mutations.
The alterations that we observed in Lightweight
heterozygotes in response to ethanol and anesthetics are qualitatively similar to what has been reported previously in C. elegans
; however, important differences exist. In invertebrates, unc-79
mutants are very hypersensitive to halothane 
; whereas, we observed no alterations in MAC for halothane in Lightweight
heterozygotes. However, we did observe that Lightweight
heterozgotes are resistant
to isoflurane. It is unclear at this point why these species differences exist. One possibility is that most of the unc-79
alleles tested in invertebrates were true null mutations; whereas, we tested heterozygous mutant mice. Another possibility is that a truncated protein might be produced in Lightweight
heterozygotes, which could interfere with the proper functioning of this pathway in unanticipated, even opposite, ways. It should be noted that the fact that Lightweight
heterozygotes were resistant to isoflurane and not to all anesthetic agents tested suggests that the anesthetic phenotype does not simply reflect a non-specific alteration in neuronal function.
The finding that C. elegans unc-79
, and nca-1;nca-2
double mutants were hypersensitive to acute ethanol exposure suggests that this pathway is important for determining ethanol sensitivity. As in vertebrates, UNC-79 and UNC-80 proteins are thought to regulate the NALCN-related leak channel encoded by nca-1
and nca-2 
. It was possible however that the ethanol phenotypes of unc-79
were due to additional, as yet undiscovered, roles of these genes. The fact that the nca-1;nca-2
double mutant exhibits the same ethanol hypersensitivity suggests that the normal function of this leak channel is indeed required for ethanol responses. Ethanol may directly modulate the nca-encoded leak channel or perhaps indirectly modulate the channel through an effect on unc-79
. The ethanol hypersensitivity of Lightweight
mice may also reflect defective NALCN function.
Because Lightweight animals exhibited dramatically altered responses to acute ethanol injection, we assayed free choice ethanol preference and consumption using a two bottle choice test. On both a B6 and a B6D2 F1 hybrid background, Lightweight heterozygotes displayed a higher preference for and consumption of ethanol than wild-type littermates. However, on a B6D2 F1 background, in addition to having a higher preference for and consumption of ethanol (at 6% and 10%), Lightweight animals also consume more water than wild-type littermates, and when offered water or water plus a tastant (saccharin), they tend to choose the bottle containing the tastant over water alone. Given these observations, in combination with the body composition and food consumption analysis performed in B6 Lightweight animals, we favor a conservative interpretation of the two bottle choice data: Lightweight animals have a higher preference for and consumption of ethanol; however, we cannot exclude the possibility that taste preference and/or alterations in metabolism could contribute to this phenotype. Because the unc-79 protein is widely expressed throughout the central nervous system, tissue-specific deletion using conditional mutagenesis or RNA interference may yield more specific results. Nonetheless, it is clear that Lightweight heterozygotes voluntarily consume more ethanol than wild-type littermates.
There is a consistent correlation between rodent populations that voluntarily consume more ethanol in a two bottle choice assay and increased operant oral self administration of ethanol 
. Because oral self administration of ethanol stimulates dopamine release 
, it is possible that the unc-79 gene product acts in the ventral tegmental area (VTA) to mediate or modulate NALCN function. In dopaminergic VTA neurons, neurotensin and other neuromodulators activate a nonselective cationic conductance with properties similar to the NALCN channel 
. Furthermore, a neurotensin- and substance P-activated inward sodium current is absent from cultured NALCN knockout VTA neurons 
. It is possible that ethanol acts directly upon NALCN channels in the VTA to influence firing rate. Alternatively, ethanol might alter the release of neurotensin, substance P, or other neuromodulators in the VTA to influence NALCN/unc-79 function secondarily. Testing of these hypotheses in the future may lead to insight into the mechanism of action of ethanol on the nervous system and specifically into mechanisms of reward-related behavior.
In conclusion, the conservation of ethanol and anesthesia phenotypes across species suggests that unc-79 and associated unc-80 and NALCN genes are part of a novel and relatively unexplored pathway that mediates the physiological effects of these agents. It is notable that we have observed a wide variety of phenotypes in heterozygotes. If variants of unc-79 and associated genes exist in the human population, it is possible that even heterozygotes may exhibit ethanol- and weight-related phenotypes.