Although ambient temperature is a critical environmental cue that must be sensed by an organism, little is known about thermosensory neuron development or thermosensory signal transduction. Temperature changes can elicit dramatic behavioral and metabolic changes. In C. elegans , the primary thermosensory neurons are the AFD neural pair, which have specialized microvillar sensory endings that are believed to mediate thermosensation. We conducted a genetic screen designed to isolate mutants with defects in AFD-specific gene expression, and identified mutants in four complementation groups. We have isolated two alleles of the previously identified thermotaxis mutant
ttx-1 . Mutants in
ttx-1 are defective in expression of all AFD specific genes examined, have morphological defects in the AFD sensory endings, and are cryophilic. In
ttx-1 mutants, the AFD neurons adopt an olfactory neuron-like fate. Misexpression of
ttx-1 in other sensory cells is sufficient to convert many other sensory cells to an AFD-like fate, including expression of AFD-specific genes and elaboration of AFD-like sensory endings. In addition, our work suggests that the AFD temperature input is important in regulating dauer recovery, perhaps via temperature regulation of insulin production. (Work on
ttx-1 was done in collaboration with Hiroyuki Sasakura, Atsushi Kuhara, and Ikue Mori).
ttx-1 encodes a homeodomain protein of the Otx/Crx family, making
ttx-1 the first identified transcription factor that regulates thermosensory neuron differentiation specifically, in any organism. Members of the Otx/Crx family have previously been shown to function in the specification of visual structures in Drosophila and vertebrates, and Crx in particular, has been shown to be required for maintenance of vertebrate photoreceptor cell fate. Interestingly, an orthologous gene pair, Chx10 and
ceh-10 , have also been shown to specify the fates of the postsynaptic partners of the photosensory (bipolar cells) and thermosensory (AIY) neurons respectively. This suggests, as proposed by Svendsen and McGhee (1995), that the photosensory and thermosensory circuits may share a common evolutionary origin. We have identified three other genes that function to regulate AFD-specific gene expression. Two of these genes,
tax-2 and
tax-4 , encode subunits of a cyclic nucleotide-gated channel which has been shown by other researchers to function in AFD, as well as other sensory cells. The fourth class of mutants have molecular lesions in the
cmk-1 gene which encodes the worm ortholog of CaMKI, a calcium/calmodulin-dependent protein kinase. There are no published accounts of a CaMKI mutant in any multicellular eukaryote, so the characterization of
cmk-1 mutants will reveal a great deal about the specific neuronal function of CaMKI.
cmk-1 mutants are thermophilic, and have reduced expression of some, but not all, AFD-specific genes.
cmk-1 is expressed in many neurons including AFD, although the function of this gene in other cells is still unknown. These results suggest that gene expression in the AFD neurons may be regulated by an activity-dependent mechanism. Why are
tax-2 ,
tax-4 and
cmk-1 required for proper expression of AFD specific genes? Do these molecules participate directly in thermosensory signaling? Are they part of a pathway allowing the thermosensory neurons to habituate to, or learn and remember, a food-associated cultivation temperature? Do defects in these molecules cause the worm to perceive the ambient temperature as higher or lower than in fact it is? We will present our progress in testing these hypotheses, with an emphasis on elucidating the pathway by which four cool genes make worms "feel the heat".