Logo of geneticsGeneticsCurrent IssueInformation for AuthorsEditorial BoardSubscribeSubmit a Manuscript
Genetics. 2006 Apr; 172(4): 2239–2252.
PMCID: PMC1456394

Identification of Guanylyl Cyclases That Function in Thermosensory Neurons of Caenorhabditis elegans


The nematode Caenorhabditis elegans senses temperature primarily via the AFD thermosensory neurons in the head. The response to temperature can be observed as a behavior called thermotaxis on thermal gradients. It has been shown that a cyclic nucleotide-gated ion channel (CNG channel) plays a critical role in thermosensation in AFD. To further identify the thermosensory mechanisms in AFD, we attempted to identify components that function upstream of the CNG channel by a reverse genetic approach. Genetic and behavioral analyses showed that three members of a subfamily of gcy genes (gcy-8, gcy-18, and gcy-23) encoding guanylyl cyclases were essential for thermotaxis in C. elegans. Promoters of each gene drove reporter gene expression exclusively in the AFD neurons and, moreover, tagged proteins were localized to the sensory endings of AFD. Single mutants of each gcy gene showed almost normal thermotaxis. However, animals carrying double and triple mutations in these genes showed defective thermotaxis behavior. The abnormal phenotype of the gcy triple mutants was rescued by expression of any one of the three GCY proteins in the AFD neurons. These results suggest that three guanylyl cyclases function redundantly in the AFD neurons to mediate thermosensation by C. elegans.

TEMPERATURE is an extremely important environmental cue for all animals. Since most biochemical and physiological processes are very sensitive to temperature, it is critical that animals respond rapidly to changes in both external and internal temperature. Cold-blooded animals must seek the appropriate temperature to maintain active motility, whereas warm-blooded animals must maintain temperature homeostasis. Temperature may also be associated with feeding resources. Parasites recognize their hosts and snakes such as pit vipers appear to detect their prey by sensing body temperature (Campbell et al. 2002). These physiological and behavioral responses require a precise ability to sense ambient temperature. However, the molecular mechanisms for thermosensation remain poorly understood.

In mammals, temperature appears to be sensed by transient receptor potential (TRP) channels, each of which is activated by different temperature ranges (Patapoutian et al. 2003; Tominaga and Caterina 2004). The discovery of thermosensitive TRP channels in vertebrates and Drosophila was a milestone for elucidation of the molecular mechanisms required for thermosensation. However, it remains possible that other proteins are also involved in thermosensation. TREK-1, a two-pore domain K+ channel in mammals, has been shown to open gradually and reversibly in response to heat (Maingret et al. 2000), whereas Na+/K+ ATPase, ATP-gated cation channel P2X3 receptor, and Na+ channels belonging to members of the degenerin/epithelial sodium channel family also have been reported to be involved in thermosensation (Pierau et al. 1974; Souslova et al. 2000; Askwith et al. 2001).

In the nonparasitic nematode Caenorhabditis elegans, the response to temperature may be observed as a behavior called thermotaxis (TTX; Hedgecock and Russell 1975). After cultivation at a uniform temperature with sufficient food, animals preferentially migrate to the cultivation temperature and move isothermally when placed on a thermal gradient without food. C. elegans sense their environmental temperature mainly via a pair of sensory neurons (AFD), which were identified by a series of laser ablation experiments (Mori and Ohshima 1995). A small number of molecules implicated in thermosensory signal transduction in AFD have been identified by analyses of mutants defective in thermotaxis. Of these, a calcium/calmodulin-dependent protein phosphatase, calcineurin, encoded by the tax-6 gene, has been shown to negatively regulate the neuronal activity of AFD (Kuhara et al. 2002), whereas a calcium/calmodulin-dependent protein kinase I encoded by the cmk-1 gene may modulate thermosensory responses (Satterlee et al. 2004). A cyclic nucleotide-gated ion channel (CNG channel) encoded by the tax-2 and tax-4 genes has been shown to be essential for thermotaxis (Coburn and Bargmann 1996; Komatsu et al. 1996). This CNG channel is expressed in AFD and shows a higher affinity to cGMP than to cAMP, suggesting that a guanylyl cyclase producing cGMP and a phosphodiesterase degrading cGMP are also likely to be involved in thermosensory signal transduction (Komatsu et al. 1996, 1999).

In this study, we use a reverse genetic approach to identify three gcy genes encoding guanylyl cyclases and show that they play a role in thermosensation in C. elegans. These gcy genes, gcy-8, gcy-18, and gcy-23, form a subfamily and are expressed exclusively in the AFD neurons. Moreover, GFP-tagged GCY proteins are localized to the sensory endings of AFD, implicating these proteins in primary signal transduction. Single mutants of each of these gcy genes show normal thermotaxis, whereas double-mutant combinations for any pair of these gcy genes show a constitutitve cryophilic phenotype. gcy-23 gcy-8 gcy-18 triple mutants show a cryophilic or athermotactic phenotype (no response to temperature) similar to the phenotype of AFD-killed wild-type animals. The gcy triple-mutant strain shows normal chemotaxis to odorants sensed by AWA and AWC olfactory neurons and to NaCl sensed mainly by ASE gustatory neurons. The thermotaxis mutant phenotype of the gcy triple-mutant strain was rescued by expression of any one of the three GCY proteins, suggesting that the GCY-8, -18, and -23 guanylyl cyclases function redundantly in C. elegans thermosensation.



C. elegans was cultured using standard procedures (Brenner 1974). The following strains were used in this work: wild-type C. elegans variety Bristol (N2), gcy-8 (oy44) IV, gcy-18 (nj38) IV, gcy-23 (nj37) IV, and tax-4 (p678) III. Various multiple-mutant strains were constructed using standard genetic methods and verified by PCR.

Isolation of deletion mutants:

For isolation of gcy-18 (nj38) and gcy-23 (nj37), a frozen deletion mutant library was constructed and ∼2,400,000 genomes were screened by nested PCR. To induce deletion mutations, synchronized L4 larvae or young adults were treated with trimethylpsoralen (TMP) and UV light as described by Yandell et al. (1994) or Gengyo-Ando and Mitani (2000). Briefly, after incubation of 60 or 0.5 μg/ml TMP for 1 hr, the worms were irradiated with ∼340 μW/cm2 of UV light for 90 sec (∼30,000 μJ) using a hand illuminator or ∼1000 μW/cm2 of UV light for ∼5–10 sec (∼5000–10,000 μJ) using a fluorescent microscope. Mutagenized worms were incubated on seeded NGM plates in the dark at 20° for 24 hr. F1 eggs were collected and allowed to hatch for 24 hr on unseeded NGM plates. About 500 hatched larvae (∼1000 genomes) were transferred onto a seeded 6-cm NGM plate and allowed to grow until bacterial lawns were consumed. One set of the library was composed of 96 plates (12 tubes in a row and 8 tubes in a column), and the standard size of the library consisted of 12 sets (∼1,2000,000 genomes). After one generation, F1 adults and F2 larvae mixtures were collected with M9 buffer. Half of each plate was frozen for the deletion library and stored at −80°. The remaining half was treated for 8 hr at 60° with SDS worm lysis buffer, containing 50 mm Tris–HCl pH 8.0, 100 mm NaCl, 10 mm EDTA, 1% SDS, and 200 μg/ml of proteinase K. A fraction of worm lysates from each of a dozen tubes in a row were pooled, purified, and used in a first PCR screen for deletion mutants. The unpurified original tubes were stored at −20°.

Oligonucleotide primers used to detect deletions were designed using AcePrimer v1.0 (http://elegans.bcgsc.bc.ca/gko/aceprimer.shtml). PCR product sizes were ∼3 kb. External-round PCRs were 20 μl in volume and contained 1 μl pooled DNA, 0.2 μm of each external-round primer, 50 μm of each dNTP, 2 mm MgCl2, 1× standard PCR buffer, and 0.025 units/μl Taq polymerase (GIBCO, Gaithersburg, MD). Internal-round PCRs were 20 μl in volume and contained 1 μl external-round PCR reactant, 0.2 μm of each external-round primer, 200 μm of each dNTP, 1.5 mm MgCl2, 1× standard PCR buffer, and 0.025 units/μl Taq polymerase. External-round cycle conditions were 94° for 1 min, 35 cycles of 94° for 40 sec; 60° for 30 sec; 72° for 1 min; and 72° for 7 min. Internal-round cycle conditions were 94° for 1 min, 35 cycles of 94° for 40 sec; 60° for 30 sec; 72° for 2 min; and 72° for 7 min. For detection of the gcy-18 deletion, primers ina122 ZK896.8-Ael and ina123 ZK896.8-Aer were used in external-round PCR and primers ina124 ZK896.8-Ail and ina125 ZK896.8-Air were used in internal-round PCR. For detection of the gcy-23 deletion, primers ina118 T26C12.4-Ael and ina119 T26C12.4-Aer were used in external-round PCR and primers ina120 T26C12.4-Ail and ina121 T26C12.4-Air were used in internal-round PCR. Loading buffer was added to each reaction, and 10 μl of mixture was loaded into each lane of 1% agarose gels containing 10 μg/ml ethidium bromide. When positive signals were detected, worm lysates of the original 12 tubes in a row were purified and screened by nested PCR to identify a single tube containing the deletion mutant candidate. The identified tube was thawed and distributed into 96 NGM plates. DNA was extracted from each plate and screened by nested PCR. The screening cycle was repeated until a deletion mutant was isolated.

For isolation of gcy-8 (oy44), ∼1,000,000 genomes mutagenized with EMS were screened by nested PCR. The PCRs were 50 μl volume and contained 10 mm Tris–HCl, pH 8.3, 50 mm KCl, 1.5 mm MgCl2, 0.1% gelatin, 200 μm of each dNTPs, 1.2 μm of each forward and reverse primers, and 1 μl Taq polymerase. Cycle conditions were 30 cycles of 94° for 30 sec, 64° for 30 sec, and 72° for 1 min. Primers gcy8KF1 and gcy8KR1 were used in external-round PCR and primers gcy8KF2 and gcy8KR2 were used in internal-round PCR.

The isolated deletion mutants were outcrossed 10 times prior to use in behavioral assays.


For screening deletion mutants from libraries, the following primers were used:

  • For gcy-8,
  • For gcy-18,
  • ina122 ZK896.8-Ael: 5′-ATCGCTACAAACACCGGAAG;
  • ina123 ZK896.8-Aer: 5′-CCGATGAATAAAGCTTCGGA;
  • ina124 ZK896.8-Ail: 5′-TCGAAGAAATCATGCACAGG;
  • ina125 ZK896.8-Air: 5′-CGTAGGCTGGTAGGAGTTGG;
  • For gcy-23,
  • ina118 T26C12.4-Ael: 5′-AGACTCCATTCTCGAGGCAA;
  • ina119 T26C12.4-Aer: 5′-CGGTTTATGCGGCATTCTAT;
  • ina120 T26C12.4-Ail: 5′-ATAGGCAATAACGAGGTGCG;
  • ina121 T26C12.4-Air: 5′-AACTGGATTCTGGCCGTCTA.

Behavioral assays:

Thermotaxis assays using a radial temperature gradient were performed essentially as described previously (Mori and Ohshima 1995). Thermotaxis of individual animals on a radial temperature gradient was evaluated using four phenotypic categories: animals that moved to the cold region (the center of the plate) were classified as “17,” animals that moved to the 20° region were classified as “20,” animals that moved to the warm region (the periphery) were classified as “25,” and animals that moved to both cold and warm regions were classified as “17/25.” Chemotaxis to volatile odorants was assayed according to previous reports (Bargmann et al. 1993). Chemotaxis to NaCl was assayed according to Komatsu et al. (1996). Animals that moved to the concentration peak of NaCl on the chemotaxis assay plate were evaluated as “+” and animals that were insensitive to the concentration gradient were evaluated as “−.”

NGM plates containing 8-bromo-cGMP were prepared as described by Birnby et al. (2000). Briefly, plates (3.5 cm) were filled with 2 ml of NGM agar with 8-bromo-cGMP (Sigma, St. Louis) added to a given concentration. A stock of Escherichia coli OP50 in L broth was spread on the NGM plates, in which individual adult animals were placed, allowed to lay eggs at 20° for up to 12 hr, and then removed. These plates were placed at 20° for 3 days and the grown animals were used for TTX assay.

TTX plot analysis:

Two values for the TTX plot were calculated as: (20 − 17/25) = (fraction of 20) − (fraction of 17/25) and (25 − 17) = (fraction of 25) − (fraction of 17), where fraction of 20 = number of animals classified as “20”/total number of animals in one trial, and so on. Statistical significance was tested by the pairwise test for multiple comparisons using Holm's method on the TTX plot values.

Molecular biology:

Standard methods of general and C. elegans molecular biology were used (Hope 1999; Sambrook and Russell 2001). DNA constructs were made using pBluescript SK+ (Stratagene, La Jolla, CA). GFP and DsRed genes for genomic and promoter fusion constructs were amplified by PCR using pEGFP-N1 and pDsRed2 (CLONTECH, Mountain View, CA) as templates, respectively. High-fidelity PCR with Pfu DNA polymerase (Stratagene) was used for constructions and DNA sequences were confirmed.

The constructs of full-length genomic gcy∷GFP gene fusion, pINA88 for gcy-8, pINA89 for gcy-18, and pINA95 for gcy-23, contain promoter regions, ORFs, and 3′-flanking regions of each gcy gene, respectively (see Figure 2). The stop codon of each gcy gene was replaced by GFP coding sequences.

Figure 2.
Gene structures of gcy-8, gcy-18, and gcy-23 genes and schematics of promoter fusions or full-length genomic fusions to reporter genes. Exons are boxed and numbered. Solid bar indicates a deleted region.

The gcy promoter∷GFP gene fusion constructs, pINA110, pINA111, and pINA112, were generated by PCR using pINA88, pINA89, and pINA95 as templates, respectively. The gcy-8 promoter DsRed gene fusion construct, pINA176, was generated by replacement of GFP coding sequences in pINA110 by PCR-amplified DsRed sequences.

Structures of gcy-8, gcy-18, and gcy-23 genes were determined by sequencing of cDNAs of each gcy genes. The yk354a3 clone contained full-length gcy-8 cDNA. Clones yk41f4 for gcy-18 and yk1064a02 for gcy-23 contained cDNAs lacking their 5′-ends. The 5′-end fragments of gcy-18 and gcy-23 cDNA were generated by BD SMART RACE cDNA amplification kit (CLONTECH) using C. elegans mRNA purified with a PolyATract mRNA isolation system (Promega, Madison, WI).

Transgenic animals:

Germ-line transformation was performed by co-injecting test DNA and pKDK66 (ges-1∷GFP) as an injection marker into the gonad of gcy-23 gcy-8 gcy-18 or N2 animals (Mello et al. 1991). Transgenic animals carrying genomic gcy∷GFP genes were generated by co-injection of each construct at a concentration of 10 ng/μl, ges-1∷GFP at a concentration of 50 ng/μl, and pBluescript SK+ at a concentration of 40 ng/μl. Transgenic animals carrying gcy promoter∷GFP or gcy-8p∷DsRed were generated by co-injection of each construct at a concentration of 50 ng/μl and ges-1∷GFP at a concentration of 50 ng/μl. The gcy-8p∷DsRed transgene was transferred by crossing to transgenic animals carrying gcy promoter∷GFP. Transgenic animals were recognized with intestinal GFP fluorescence due to ges-1∷GFP transformation marker expression. Multiple independent transgenic lines were established and examined for each transformation experiment. Cells expressing GFP and intracellular localization of chimera GCY/GFP proteins were identified and analyzed using a light microscope AxioPlan2 (Carl Zeiss, Thornwood, NY).


The gcy-8, gcy-18, and gcy-23 guanylyl cyclase genes constitute a subfamily of guanylyl cyclase genes in C. elegans:

In C. elegans, guanylyl cyclase genes (gcy genes) form a large gene family (Yu et al. 1997). At least 26 gcy genes encoding receptor-type guanylyl cyclases and 7 gcy genes encoding soluble guanylyl cyclases are predicted from genome sequence analysis (Chen et al. 2005). Phylogenetic analysis showed that the gcy genes encoding receptor-type guanylyl cyclases are divided into several subfamilies on the basis of amino acid sequences (Figure 1A). We focused on a subfamily that consists of the gcy-8, gcy-18, and gcy-23 genes, because a previous report showed that the gcy-8 gene was expressed specifically in AFD (Yu et al. 1997). The three guanylyl cyclases encoded by these gcy genes showed high overall sequence homology (>55% identity and 73% similarity) at the amino acid sequence level. Receptor-type guanylyl cyclases contain three domains: an extracellular domain, a transmembrane region, and an intracellular region containing a kinase homology domain, hinge region, and cyclase domain (Figure 1B). GCY-8, -18, and -23 showed little homology to human guanylyl cyclases in the extracellular domain (Figure 1C). At the N terminus and just before the transmembrane region in the extracellular domain, however, these guanylyl cyclases have seven conserved cysteine residues, which have been reported to form intrachain disulfide bridges in mammalian guanylyl cyclases. In contrast, the intracellular domain, especially the cyclase domain, showed high homology among all guanylyl cyclases (Figure 1C). Three serine and threonine residues that have been reported to be reversibly phosphorylated in mammalian guanylyl cyclases were also conserved in the kinase homology domains of the GCY-8, GCY-18, and GCY-23 guanylyl cyclases.

Figure 1.
gcy-8, gcy-18, and gcy-23 form a subfamily of gcy genes and encode receptor-type guanylyl cyclases in C. elegans. (A) Phylogenetic tree of C. elegans gcy genes encoding receptor-type guanylyl cyclases. Sequences of C. elegans guanylyl cyclases (gene name ...

gcy-8, -18, and -23 genes are expressed specifically in the AFD thermosensory neurons:

To determine whether these gcy genes are expressed in AFD, the expression of GFP reporter genes under the control of gcy-8, gcy-18, and gcy-23 promoters was examined in wild-type animals (Figure 2). Expression in the AFD neurons was confirmed by colocalization with DsRed expression driven by the gcy-8 promoter (Figure 3, A, D, and G), as reported previously (Yu et al. 1997). Expression of both gcy-18 and gcy-23 genes was identical to that of the gcy-8 gene (Figure 3, B and H, for gcy-18; Figure 3, C and I, for gcy-23), and no expression was observed in other cells. This result indicates that all three gcy genes are expressed specifically in AFD.

Figure 3.
Expression analysis of gcy-8, gcy-18, and gcy-23 genes. (A–C) GFP fluorescence images of gcy-8p∷GFP (A), gcy-18p∷GFP (B), and gcy-23p∷GFP (C) transgenes. (D–F) DsRed fluorescence images of gcy-8p∷DsRed transgene ...

To determine the intracellular localization of the gcy gene products, full-length genomic gcy gene sequences fused to the GFP gene were injected into wild-type animals (Figure 2). GFP-tagged GCY-8, 18, and 23 were localized exclusively to the sensory endings of AFD (Figure 3, J–L). The TAX-4 CNG channel has also been shown to be localized to the AFD sensory endings, suggesting that these GCY proteins may function with the TAX-4 CNG channel to mediate thermosensation.

Strains doubly mutant for gcy genes show defects in thermotaxis:

To examine whether these gcy genes function in thermotaxis, deletion alleles were isolated (Figure 2). The isolated gcy deletion alleles are likely to be functional null alleles (see Figure 2). Thermotaxis behaviors of the gcy mutants were analyzed after growth at different cultivation temperatures. After cultivation at 15°, 20°, or 25°, the majority of wild-type animals migrated to their cultivation temperature on a radial thermal gradient. Animals singly mutant for one of these gcy genes also migrated to their cultivation temperature and showed nearly normal thermotaxis behavior (Figure 4, A–C, and Table 1). However, animals doubly mutant for any pair of these gcy genes showed different degrees of defects in thermotaxis behavior (Figure 4, D–F, and Table 2). When animals were cultivated at 15°, wild-type animals and gcy double mutants migrated to 15°, but the fraction of animals that showed athermotactic phenotype (classified as 17/25) was increased in any double gcy mutant combination (Figure 4D). When animals were cultivated at 20°, ∼50% of gcy-23 gcy-8 double mutants migrated to 20° but the remaining 50% migrated to colder temperatures. However, gcy-8 gcy-18 and gcy-23 gcy-18 double mutants cultivated at 20° migrated to colder temperatures and showed a severe cryophilic phenotype (Figure 4E). When animals were cultivated at 25°, gcy-8 gcy-18 and gcy-23 gcy-18 double mutants also showed a cryophilic phenotype, but the fraction of animals that showed athermotactic phenotype was increased in any double gcy mutant combination (Figure 4F). About half of gcy-23 gcy-8 double-mutant animals that contain a wild-type gcy-18 gene could migrate to the cultivation temperature when they were cultivated at 20° or 25°, suggesting that gcy-18 functions as the primary guanylyl cyclase in these temperature ranges.

Figure 4.
Thermotaxis phenotypes of gcy single and double mutants cultivated at three different temperatures. (A–C) Thermotaxis phenotypes of wild-type animals and gcy single mutants cultivated at 15° (A), 20° (B), or 25° (C). (D–F) ...
Thermotaxis in gcy single mutants
Thermotaxis in gcy double mutants

Animals triply mutant for the gcy genes show severe defects in thermotaxis but respond normally to odorants and NaCl:

gcy-23 gcy-8 gcy-18 triple mutants showed the severest cryophilic or athermotactic phenotype at any cultivation temperature (Figure 5A and Table 3). This phenotype of gcy triple-mutant animals is very similar to the phenotype of AFD-killed wild-type animals (Table 3), indicating that activity of AFD is almost abolished in the gcy triple mutants. However, significant differences were observed between the phenotype of gcy triple mutants and AFD(−) wild-type animals. The cryophilic behaviors of gcy triple mutants were completely suppressed by loss-of-function mutations in the tax-4 gene. The thermotaxis behaviors of tax-4 ; gcy-23 gcy-8 gcy-18 quadruple mutants showed no significant difference from those of tax-4 mutants (Figure 5B and Table 4). This result is consistent with the hypothesis that these guanylyl cyclases function upstream of the TAX-4 CNG channel.

Figure 5.
Behavioral phenotypes of gcy triple mutants and tax-4; gcy-23 gcy-8 gcy-18 quadruple mutants. (A) Thermotaxis phenotypes of wild-type animals and gcy triple mutants cultivated at different temperatures. Wild-type animals (n = 90) and gcy triple ...
Thermotaxis in gcy triple mutants
Thermotaxis in quadruple mutants

gcy triple mutants showed nearly normal chemotaxis toward odorants sensed by the AWA and AWC olfactory neurons (Figure 5C), as well as toward NaCl, which is sensed primarily by the ASE gustatory neuron (Figure 5D), suggesting that these gcy genes function specifically in thermotaxis.

TTX plot analysis of gcy mutants:

As shown above, various degrees of defects in thermotaxis behaviors were observed in gcy double and triple mutants but not in gcy single mutants (Figures 4 and 5 and Tables 133).). To analyze the thermotaxis behavioral defects in detail, we introduced a novel analysis—the TTX plot—for comprehensive statistical analysis (see materials and methods).

When animals were cultivated at 15°, two gcy double-mutant strains (gcy-8 gcy-18 and gcy-23 gcy-18) and the gcy-23 gcy-8 gcy-18 triple-mutant strain showed significant behavioral differences compared with wild-type animals in the TTX plot value (20 − 17/25), whereas none of the gcy single mutants and gcy-23 gcy-8 double-mutant combinations showed a significant difference (Figure 6A Table 5 and Table 6). No significant differences were observed among gcy single-, double-, and triple-mutant strains (Table 6).

Figure 6.
TTX plot of wild-type animals and gcy mutants cultivated at three different temperatures. Wild-type animals and gcy mutants cultivated at 15° (A), 20° (B), or 25° (C) were assayed and the TTX plot values were calculated as described ...
TTX plot in gcy mutants
Statistical analysis among wild-type animals and gcy mutants in the TTX plot values (20 − 17/15) and (25 − 17)

When animals were cultivated at 20°, gcy single mutants showed no significant differences compared with wild-type animals in the TTX plot (Figure 6B, Table 5, and Table 6), whereas gcy double and triple mutants showed a significant difference. gcy-23 gcy-8 gcy-18 triple mutants also showed a significant difference when compared with all strains with the exception of the gcy-8 gcy-18 double mutants.

Wild-type animals and gcy mutants were clearly divided into two groups on the TTX plot when they were cultivated at 25° (Figure 6C). One group contained wild-type animals, gcy single mutants, and gcy-23 gcy-8 double mutants, whereas another group contained the gcy-8 gcy-18 and gcy-23 gcy -18 double mutants and gcy-23 gcy-8 gcy-18 triple mutants. No significant differences were observed within each group (Table 5 and Table 6). These results of the TTX plot analysis also suggest the central role of gcy-18 in thermosensation in animals cultivated at 20° or 25°.

Functional redundancy of the three gcy genes in thermotaxis:

The defects in thermotaxis of gcy triple mutants were rescued by introduction of gcy genomic fragments (Figure 7). Transgenic gcy triple mutants carrying the gcy-8 genomic fragment showed a phenotype similar to that of gcy-18 gcy-23 double mutants. Interestingly, a relatively large fraction of transgenic gcy triple mutants carrying a gcy-23 genomic fragment showed normal thermotaxis when compared with gcy-8 gcy-18 double mutants. Overexpression of gcy genes seemed to have a little effect on thermotaxis in wild-type animals. Wild-type animals overexpressing the gcy-8 gene showed slight defects in thermotaxis, while wild-type animals overexpressing the gcy-18 or gcy-23 gene showed almost normal thermotaxis.

Figure 7.
Rescue experiments using full-length gcy∷GFP fusions. Animals were cultivated at 20°. Transgenic animals carrying full-length gcy∷GFP fusions, Ex[gcy-8∷GFP], Ex[gcy-18∷GFP], or Ex[gcy-23∷GFP], also carry ...

Although overexpression of gcy genes seems to have little effect on thermotaxis in wild-type animals, we assayed the effect of the membrane permeant cGMP analog 8-bromo-cGMP on thermotaxis to examine whether an appropriate concentration of intracellular cGMP is required for normal thermotaxis behavior (Figure 8). While ∼60% of wild-type animals migrated to their cultivation temperature when they were cultivated without 8-bromo-cGMP, wild-type animals cultivated with 8-bromo-cGMP showed slight defects: ∼35–60% of animals showed thermophilic (25) or athermotactic (17/25) phenotypes, possibly resulting from the high concentration of intracellular cGMP in AFD. This result and the phenotype of gcy triple mutants are consistent with the possibility that a low concentration of cGMP in AFD results in a cryophilic phenotype, whereas a high concentration of cGMP results in a thermophilic phenotype. The phenotype of the gcy triple mutant was not affected or rescued by the application of 8-bromo-cGMP at any concentration, suggesting that the appropriate regulation of intracellular cGMP concentration is essential for normal thermotaxis.

Figure 8.
Effect of application of 8-bromo-cGMP on thermotaxis behaviors of wild-type animals and gcy triple mutants. Animals were cultivated at 20°. For each condition, 19 or 20 animals were assayed individually. Phenotypic categories are described in ...


In C. elegans, cGMP plays a critical role in sensory signal transduction. For example, tax-2 and tax-4 genes encoding a cGMP-gated channel are required for various behaviors such as thermosensation, olfaction, and gustation (Coburn and Bargmann 1996; Komatsu et al. 1996). odr-1 and daf-11 genes encoding receptor-type guanylyl cyclases are also essential for olfaction in AWC (Vowels and Thomas 1994; Birnby et al. 2000; L'Etoile and Bargmann 2000). Recent studies have shown that a soluble guanylyl cyclase encoded by gcy-35 functions as an oxygen sensor in some sensory neurons (Gray et al. 2004; Cheung et al. 2005). In this study, we show that three gcy genes encoding receptor-type guanylyl cyclases are essential for thermotaxis in C. elegans.

Three guanylyl cyclases essential for thermotaxis in C. elegans:

Receptor-type guanylyl cyclases have been studied extensively as receptors of peptide hormones (Wedel and Garbers 2001). In mammals, ligand-induced activity of guanylyl cyclases is regulated by phosphorylation and dephosphorylation of serine and threonine residues in the kinase homology domain (Potter and Garbers 1992, 1994; Potter and Hunter 1998). Phosphorylation of these residues is essential for activation of the guanylyl cyclase; dephosphorylation results in desensitization and a reduction in ligand-induced guanylyl cyclase activity. A subset of these serine and threonine residues was also conserved in the GCY-8, 18, and 23 guanylyl cyclases in C. elegans. Recently, TAX-6 calcineurin has been shown to be involved in thermosensation in AFD of C. elegans and the tax-6 mutant showed a severe thermophilic (heat-seeking) phenotype resulting from hyperactivation of AFD (Kuhara et al. 2002). Activity of these three guanylyl cyclases might be regulated through dephosphorylation by TAX-6 calcineurin in C. elegans.

Why are three guanylyl cyclases required for thermotaxis? One possibility is that thermosensation requires a high level of cGMP concentration to maintain activity of the thermosensory neurons since temperature is a continuous stimulus unlike other chemical stimuli such as odorants. For continuous signal transduction, three guanylyl cyclases may be required to maintain high levels of intracellular cGMP concentration in AFD, as in mammalian retinal cells, which require the functions of two guanylyl cyclases (Shyjan et al. 1992; Lowe et al. 1995). Alternatively, since C. elegans can sense temperature changes with high resolution (∼0.05°) (Hedgecock and Russell 1975; Ryu and Samuel 2002), rapid synthesis of cGMP may be required to detect slight temperature changes. Application of the membrane permeant cGMP analog 8-bromo-cGMP (0.5, 1, or 5 mm) affected thermotaxis behaviors in wild-type animals, but did not rescue the abnormal phenotype of gcy triple mutants (Figure 8), suggesting that thermosensory signal transduction requires the accurate regulation of intracellular cGMP concentrations via the production or the degradation of cGMP. It is also possible that each guanylyl cyclase is involved in sensation of different ranges of temperature. In this case, the temperature ranges sensed by each guanylyl cyclase would overlap since gcy single mutants show normal thermotaxis when grown at a given cultivation temperature. The different phenotypes of gcy double mutants, in which only one guanylyl cyclase remains functional, may reflect temperature specificities of the guanylyl cyclases. The relatively weak defects of gcy-23 gcy-8 mutants suggest that the GCY-18 guanylyl cyclase may be required for responses to a wider range of temperature than either the GCY-8 or the GCY-23 guanylyl cyclase. Considering that phenotypes of gcy-8 gcy-18 and gcy-23 gcy-18 mutants cultivated at 25° were similar to those of gcy triple mutants cultivated at 25°, GCY-8 and GCY-23 guanylyl cyclases might have a minor contribution to thermotaxis in animals cultivated at 25°. Alternatively, each guanylyl cyclase might be involved in the learning of different temperatures. However, temperature-shift experiments (from 17° to 25° or from 25° to 17°) showed that gcy single mutants are able to learn a new temperature with a time course similar to that of wild-type animals (data not shown). Although results from the TTX assay, the TTX plot analysis, and the time-course assay strongly suggest that these three guanylyl cyclases function redundantly under normal conditions, it is possible that these guanylyl cyclases may have different activities or kinetics. A gcy-8 genomic fragment rescued the phenotype of gcy triple mutants to a level similar to that of gcy-23 gcy-18 double mutants in which only GCY-8 functions, but not to the level of wild-type animals. The relatively weak ability of the gcy-8 genomic fragment to rescue the phenotype of the gcy triple mutants and the slight defects observed upon gcy-8 overexpression in wild-type animals may suggest a modulatory role of GCY-8 in thermosensation. Receptor-type guanylyl cyclases appear to form homodimers or homotetramers in mammals. Combinatorial oligomerization between guanylyl cyclases with different features may contribute to sensitive regulation of signal transduction in thermosensation of C. elegans.

How many guanylyl cyclases are required in one sensory neuron? At least two guanylyl cyclases, ODR-1 and DAF-11, function in signal transduction in the AWC olfactory neurons (Vowels and Thomas 1994; Birnby et al. 2000; L'Etoile and Bargmann 2000). Expression of a few guanylyl cyclases was also reported in the ASE gustatory neurons (Yu et al. 1997). Expression of a fourth gcy gene, gcy-12, was observed in the AFD thermosensory neurons (H. Inada, H. Komatsu, M. Kosaki and I. Mori, unpublished results). However, expression of the gcy-12 gene was not restricted to the AFD neurons, but was observed in many sensory neurons, including the AWC olfactory neurons and the ASE gustatory neurons. Furthermore, the gcy-12 single mutants and the gcy-12; gcy-8 double mutants showed normal thermotaxis, suggesting that the GCY-12 guanylyl cyclase does not play a major role in AFD-mediated thermosensation. The presence of multiple gcy genes predicted by the C. elegans genome may reflect functional redundancy required for robustness of signal transduction pathways or, alternatively, may have specialized but as yet unidentified functions in individual neuron types.

Molecular mechanisms of signal transduction in the AFD thermosensory neuron:

This study and previous studies on thermosensation in C. elegans lead to a possible model of molecular mechanisms for thermosensory signal transduction shown in Figure 9. Temperature received by a thermoreceptor is transduced to changes in intracellular cGMP concentration via three guanylyl cyclases, GCY-8, GCY-18, and GCY-23, and transmitted to the TAX-4 CNG channel, resulting in a change of membrane potential of AFD.

Figure 9.
A molecular model thermosensory signal transduction in AFD. Temperature sensed by a thermoreceptor leads to changes in intracellular cGMP concentration via the function of three guanylyl cyclases, GCY-8, GCY-18, and GCY-23. cGMP regulates activity of ...

What kind of molecule can function as a thermoreceptor? G-protein-coupled seven-transmembrane thermoreceptors (7TM receptor) might exist upstream of these guanylyl cyclases and the TAX-4 CNG channel (Komatsu et al. 1996; Roayaie et al. 1998; L'Etoile and Bargmann 2000). Recently, it was reported that a few genes encoding 7TM receptors are expressed in AFD (Colosimo et al. 2004). These 7TM receptors might function as thermoreceptors in AFD. However, it remains possible that the three guanylyl cyclases described in this study function as thermoreceptors. Biochemical analysis using a heterologous expression system has shown that activity of a C. elegans guanylyl cyclase, GCY-12, was clearly dependent on temperature (Yu et al. 1997). Biochemical characterization of GCY-8, GCY-18, and GCY-23 guanylyl cyclases may clarify whether these guanylyl cyclases function directly as thermoreceptors. Analysis of thermosensory signal transduction in C. elegans may reveal thermosensory mechanisms other than thermosensitive TRP channels.


We thank S. Mitani and Gengyo-Ando for technical advice on the preparation of the deletion mutant library by TMP/UV methods; C. Hunter, Y. Shi, and C. Li for construction of a nonfrozen library and gcy-8 mutant screening; Y. Kohara for yk cDNA clones; the Caenorhabditis Genetic Center for cosmid clones; the C. elegans Sequence Consortium for updated C. elegans genome information; and members of the Mori laboratory for stimulating discussion. This work was supported by a research grant from Ministry of Education, Culture, Sports, Science and Technology (to I.M.) and by grants from the National Institutes of Health (PO1 NS44232 to P.S.; T32NS07292 and F32 GM20048 to J.S). I.M. is a Distinguished Scholar of Nagoya University.


Sequence data from this article have been deposited with the DDBJ/EMBL/GenBank Data Libraries under accession nos. AB201388, AB201389, and AB201390.


  • Askwith, C. C., C. J. Benson, M. J. Welsh and P. M. Snyder, 2001. DEG/ENaC ion channels involved in sensory transduction are modulated by cold temperature. Proc. Natl. Acad. Sci. USA 98: 6459–6463. [PMC free article] [PubMed]
  • Bargmann, C. I., E. Hartwieg and H. R. Horvitz, 1993. Odorant-selective genes and neurons mediate olfaction in C. elegans. Cell 74: 515–527. [PubMed]
  • Birnby, D. A., E. M. Link, J. J. Vowels, H. Tian, P. L. Colacurcio et al., 2000. A transmembrane guanylyl cyclase (DAF-11) and Hsp90 (DAF-21) regulate a common set of chemosensory behaviors in Caenorhabditis elegans. Genetics 155: 85–104. [PMC free article] [PubMed]
  • Brenner, S., 1974. The genetics of Caenorhabditis elegans. Genetics 77: 71–94. [PMC free article] [PubMed]
  • Campbell, A. L., R. R. Naik, L. Sowards and M. O. Stone, 2002. Biological infrared imaging and sensing. Micron 33: 211–225. [PubMed]
  • Chen, N., T. W. Harris, I. Antoshechkin, C. Bastiani, T. Bieri et al., 2005. WormBase: a comprehensive data resource for Caenorhabditis biology and genomics. Nucleic Acids Res. 33: D383–D389. [PMC free article] [PubMed]
  • Cheung, B. H., M. Cohen, C. Rogers, O. Albayram and M. de Bono, 2005. Experience-dependent modulation of C. elegans behavior by ambient oxygen. Curr. Biol. 15: 905–917. [PubMed]
  • Coburn, C. M., and C. I. Bargmann, 1996. A putative cyclic nucleotide-gated channel is required for sensory development and function in C. elegans. Neuron 17: 695–706. [PubMed]
  • Colosimo, M. E., A. Brown, S. Mukhopadhyay, C. Gabel, A. E. Lanjuin et al., 2004. Identification of thermosensory and olfactory neuron-specific genes via expression profiling of single neuron types. Curr. Biol. 14: 2245–2251. [PubMed]
  • Gengyo-Ando, K., and S. Mitani, 2000. Characterization of mutations induced by ethyl methanesulfonate, UV, and trimethylpsoralen in the nematode Caenorhabditis elegans. Biochem. Biophys. Res. Commun. 269: 64–69. [PubMed]
  • Gray, J. M., D. S. Karow, H. Lu, A. J. Chang, J. S. Chang et al., 2004. Oxygen sensation and social feeding mediated by a C. elegans guanylate cyclase homologue. Nature 430: 317–322. [PubMed]
  • Hedgecock, E. M., and R. L. Russell, 1975. Normal and mutant thermotaxis in the nematode Caenorhabditis elegans. Proc. Natl. Acad. Sci. USA 72: 4061–4065. [PMC free article] [PubMed]
  • Hope, I., 1999. C. elegans: A Practical Approach. Oxford University Press, New York.
  • Komatsu, H., I. Mori, J. S. Rhee, N. Akaike and Y. Ohshima, 1996. Mutations in a cyclic nucleotide-gated channel lead to abnormal thermosensation and chemosensation in C. elegans. Neuron 17: 707–718. [PubMed]
  • Komatsu, H., Y. H. Jin, N. L'Etoile, I. Mori, C. I. Bargmann et al., 1999. Functional reconstitution of a heteromeric cyclic nucleotide-gated channel of Caenorhabditis elegans in cultured cells. Brain Res. 821: 160–168. [PubMed]
  • Kuhara, A., H. Inada, I. Katsura and I. Mori, 2002. Negative regulation and gain control of sensory neurons by the C. elegans calcineurin TAX-6. Neuron 33: 751–763. [PubMed]
  • L'Etoile, N. D., and C. I. Bargmann, 2000. Olfaction and odor discrimination are mediated by the C. elegans guanylyl cyclase ODR-1. Neuron 25: 575–586. [PubMed]
  • Lowe, D. G., A. M. Dizhoor, K. Liu, Q. Gu, M. Spencer et al., 1995. Cloning and expression of a second photoreceptor-specific membrane retina guanylyl cyclase (RetGC), RetGC-2. Proc. Natl. Acad. Sci. USA 92: 5535–5539. [PMC free article] [PubMed]
  • Maingret, F., I. Lauritzen, A. J. Patel, C. Heurteaux, R. Reyes et al., 2000. TREK-1 is a heat-activated background K(+) channel. EMBO J. 19: 2483–2491. [PMC free article] [PubMed]
  • Mello, C. C., J. M. Kramer, D. Stinchcomb and V. Ambros, 1991. Efficient gene transfer in C. elegans: extrachromosomal maintenance and integration of transforming sequences. EMBO J. 10: 3959–3970. [PMC free article] [PubMed]
  • Mori, I., and Y. Ohshima, 1995. Neural regulation of thermotaxis in Caenorhabditis elegans. Nature 376: 344–348. [PubMed]
  • Patapoutian, A., A. M. Peier, G. M. Story and V. Viswanath, 2003. ThermoTRP channels and beyond: mechanisms of temperature sensation. Nat. Rev. Neurosci. 4: 529–539. [PubMed]
  • Pierau, F. K., P. Torrey and D. O. Carpenter, 1974. Mammalian cold receptor afferents: role of an electrogenic sodium pump in sensory transduction. Brain Res. 73: 156–160. [PubMed]
  • Potter, L. R., and D. L. Garbers, 1992. Dephosphorylation of the guanylyl cyclase-A receptor causes desensitization. J. Biol. Chem. 267: 14531–14534. [PubMed]
  • Potter, L. R., and D. L. Garbers, 1994. Protein kinase C-dependent desensitization of the atrial natriuretic peptide receptor is mediated by dephosphorylation. J. Biol. Chem. 269: 14636–14642. [PubMed]
  • Potter, L. R., and T. Hunter, 1998. Phosphorylation of the kinase homology domain is essential for activation of the A-type natriuretic peptide receptor. Mol. Cell. Biol. 18: 2164–2172. [PMC free article] [PubMed]
  • Roayaie, K., J. G. Crump, A. Sagasti and C. I. Bargmann, 1998. The G alpha protein ODR-3 mediates olfactory and nociceptive function and controls cilium morphogenesis in C. elegans olfactory neurons. Neuron 20: 55–67. [PubMed]
  • Ryu, W. S., and A. D. Samuel, 2002. Thermotaxis in Caenorhabditis elegans analyzed by measuring responses to defined thermal stimuli. J. Neurosci. 22: 5727–5733. [PubMed]
  • Sambrook, J., and D. W. Russell, 2001. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
  • Satterlee, J. S., W. S. Ryu and P. Sengupta, 2004. The CMK-1 CaMKI and the TAX-4 cyclic nucleotide-gated channel regulate thermosensory neuron gene expression and function in C. elegans. Curr. Biol. 14: 62–68. [PubMed]
  • Shyjan, A. W., F. J. de Sauvage, N. A. Gillett, D. V. Goeddel and D. G. Lowe, 1992. Molecular cloning of a retina-specific membrane guanylyl cyclase. Neuron 9: 727–737. [PubMed]
  • Souslova, V., P. Cesare, Y. Ding, A. N. Akopian, L. Stanfa et al., 2000. Warm-coding deficits and aberrant inflammatory pain in mice lacking P2X3 receptors. Nature 407: 1015–1017. [PubMed]
  • Tominaga, M., and M. J. Caterina, 2004. Thermosensation and pain. J. Neurobiol. 61: 3–12. [PubMed]
  • Vowels, J. J., and J. H. Thomas, 1994. Multiple chemosensory defects in daf-11 and daf-21 mutants of Caenorhabditis elegans. Genetics 138: 303–316. [PMC free article] [PubMed]
  • Wedel, B., and D. Garbers, 2001. The guanylyl cyclase family at Y2K. Annu. Rev. Physiol. 63: 215–233. [PubMed]
  • Yandell, M. D., L. G. Edgar and W. B. Wood, 1994. Trimethylpsoralen induces small deletion mutations in Caenorhabditis elegans. Proc. Natl. Acad. Sci. USA 91: 1381–1385. [PMC free article] [PubMed]
  • Yu, S., L. Avery, E. Baude and D. L. Garbers, 1997. Guanylyl cyclase expression in specific sensory neurons: a new family of chemosensory receptors. Proc. Natl. Acad. Sci. USA 94: 3384–3387. [PMC free article] [PubMed]

Articles from Genetics are provided here courtesy of Genetics Society of America
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Compound
    PubChem Compound links
  • Gene
    Gene links
  • Gene (nucleotide)
    Gene (nucleotide)
    Records in Gene identified from shared sequence links
  • GEO Profiles
    GEO Profiles
    Related GEO records
  • HomoloGene
    HomoloGene links
  • MedGen
    Related information in MedGen
  • Nucleotide
    Published Nucleotide sequences
  • Pathways + GO
    Pathways + GO
    Pathways, annotations and biological systems (BioSystems) that cite the current article.
  • Protein
    Published protein sequences
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem Substance links
  • Taxonomy
    Related taxonomy entry
  • Taxonomy Tree
    Taxonomy Tree

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...