cGMP dynamics that underlies thermosensation in temperature-sensing neuron regulates thermotaxis behavior in C. elegans

Abstract

Living organisms including bacteria, plants and animals sense ambient temperature so that they can avoid noxious temperature or adapt to new environmental temperature. A nematode C. elegans can sense innocuous temperature, and navigate themselves towards memorize past cultivation temperature (Tc) of their preference. For this thermotaxis, AFD thermosensory neuron is pivotal, which stereotypically responds to warming by increasing intracellular Ca2+ level in a manner dependent on the remembered past Tc. We aimed to reveal how AFD encodes the information of temperature into neural activities. cGMP synthesis in AFD is crucial for thermosensation in AFD and thermotaxis behavior. Here we characterized the dynamic change of cGMP level in AFD by imaging animals expressing a fluorescence resonance energy transfer (FRET)-based cGMP probe specifically in AFD and found that cGMP dynamically responded to both warming and cooling in a manner dependent on past Tc. Moreover, we characterized mutant animals that lack guanylyl cyclases (GCYs) or phosphodiesterases (PDEs), which synthesize and hydrolyze cGMP, respectively, and uncovered how GCYs and PDEs contribute to cGMP and Ca2+ dynamics in AFD and to thermotaxis behavior.

Fig 1. cGMP dynamics in AFD.

A and B. Wild type animals expressing cGi-500 cGMP indicator specifically in AFD thermosensory neurons (IK3110) were cultivated at 17°C (A) or 23°C (B). Blue and yellow fluorescence was monitored during warming from 14°C to 23°C and subsequent cooling to 14°C, which were repeated twice as indicated (orange line). Warming and cooling was at the rate of 1°C/6 sec. Individual (gray) and average fluorescence ratio (CFP/YFP) change at AFD sensory ending (blue) and soma (dark blue) is shown. The temperature program was repeated twice since the increment of the fluorescence ratio was more remarkable in response to the second warming especially in 23°C-cultivate animals, probably due to the fluorescence ratio was once decreased by the first round of cooling. C. gcy-18 gcy-8 gcy-23 triple mutant animals expressing cGi-500 cGMP indicator in AFD (IK3360) were cultivated at 23°C and subjected to imaging analysis. R₀ is average of R (CFP/YFP) from t = 1 to t = 30.

Fig 2. cGMP onsets from lower temperature in gcy double mutants.

A. Wild type animals and animals in which indicated gcy gene(s) is mutated were cultivated at 17°C or 23°C and then placed on a thermal gradient. The number of animals in each section of the thermal gradient was scored, and the proportion of animals in each section was plotted on histograms. n = 3 to 6 as indicated by open circles in boxplots. The error bars in histograms represent the standard error of mean (SEM). The thermotaxis indices were plotted on boxplots. The indices of strains marked with distinct alphabets differ significantly (p < 0.05) according to the Tukey-Kramer test. B. Wild type and indicated gcy double mutant animals that express cGi-500 cGMP indicator in AFD were cultivated at 23°C and subjected to imaging analysis with temperature stimuli indicated (orange line). Warming and cooling was at the rate of 1°C/20 sec. Individual (gray) and average (blue) fluorescence ratio (CFP/YFP) change at AFD sensory ending was plotted against time. A temperature program with slower change rate than in Fig 1 was used to compare the onset temperature between different strains. Since the increment of the fluorescence ratio was more remarkable in response to the warming after cooling, the temperature program starting from cooling was used to shorten measurement time and therefore to prevent the probe from bleaching. C and D. Temperature at which cGMP level started increasing in response to warming (C) and decreasing in response to the 2^nd cooling (D) was extracted using a MATLAB command ‘findchangepts’ as detailed in S1 Fig and plotted. *** indicates p < 0.001 (Dunnett test against wild type animals).

Fig 3. pde mutants are defective for thermotaxis behavior.

Wild type and pde mutant animals indicated were cultivated at 17°C or 23°C and then subjected to thermotaxis assay. n = 4 to 8 as indicated by open circles in boxplots. The error bars in histograms represent the standard error of mean (SEM). The thermotaxis indices of strains marked with distinct alphabets differ significantly (p < 0.05) according to the Tukey-Kramer test.

Fig 4. pde-5 acts in AFD to regulate cGMP dynamics and thermotaxis.

A. Wild type and pde-5 animals and pde-5 animals that express PDE-5 specifically in AFD were cultivated at 17°C or 23°C and then subjected to thermotaxis assay. n = 4. The error bars in histograms represent the standard error of mean (SEM). The thermotaxis indices of strains marked with distinct alphabets differ significantly (p < 0.05) according to the Tukey-Kramer test. B. Wild type and pde-5 animals and pde-5 mutant animals expressing PDE-5 in AFD that express cGi-500 cGMP indicator in AFD were cultivated at 17°C or 23°C. Animals were then subjected to imaging analysis with temperature stimuli indicated (orange line). Warming and cooling was at the rate of 1°C/6 sec. Individual (gray) and average (blue) fluorescence ratio (CFP/YFP) change at AFD sensory ending is shown. Dataset of wild type cultivated at 17°C are identical to those in Fig 1A. C and D. Wild type and pde-5 animals and pde-5 mutant animals expressing PDE-5 in AFD that express GCaMP3 Ca²⁺ indicator and tagRFP in AFD were cultivated at 17°C or 23°C. Animals were then subjected to imaging analysis with temperature stimuli indicated (orange line). Individual (gray) and average (pea green) fluorescence ratio (GCaMP/RFP) change at AFD sensory ending (C) and soma (D) is shown. E. pde-5(nj49); njEx1414[gcy-8p::pde-5::GFP, ges-1p::NLStagRFP] was subjected to microscopic analysis with Zeiss LSM880 confocal microscope.

Fig 5. pde-1 and pde-5 synergize in AFD.

A. Wild type, pde-1, pde-5 and pde-5 pde-1 double mutant animals and pde-5 pde-1 animals that express pde-1 or pde-5 in AFD were cultivated at 23°C and then subjected to thermotaxis assay. n = 4. The error bars in histograms represent the standard error of mean (SEM). The thermotaxis indices of strains marked with distinct alphabets differ significantly (p < 0.05) according to the Tukey-Kramer test. B. pde-5 pde-1 mutant animals that express cGi-500 cGMP indicator in AFD were cultivated at 23°C and subjected to imaging analysis. C-D. Wild type, pde-1, pde-5 and pde-5 pde-1 double mutant animals that express GCaMP3 Ca²⁺ indicator and tagRFP in AFD were cultivated at 23°C and subjected to imaging analysis. Warming and cooling was at the rate of 1°C/20 sec. Individual (gray) and average (pea green or green) fluorescence ratio (GCaMP/RFP) change at AFD sensory ending (C) and soma (D) is shown. Data of wild type and pde-1 animals are identical to those in S5C and S5D Fig.

Fig 6. Contribution of GCYs and PDEs on cGMP and Ca²⁺ dynamics in AFD cultivated at different temperature.

In animals cultivated at 17°C, any of GCY-23, GCY-8 and GCY-18 can probably contribute to the cGMP increment in response to warming since all of three gcy double mutants show Ca²⁺ response (S2 and S3 Figs). cGMP production by GCY-18 alone might not be sufficient since gcy-23 gcy-8 shows slightly defective thermotaxis (Fig 2A). Ca²⁺ level in soma is actively decreased via the three GCYs (S3 Fig) and SLO-2 potassium channel [40]. In animals cultivated at 23°C, activity of GCY-23 and GCY-8 is suppressed below the threshold temperature by coexistence of both, possibly forming an inactive dimer. Threshold for GCY-18 seems to be adjustable by an unknow AFD-specific mechanism (See ’Discussion‘ section). Transcription of gcy-18 is increased under higher cultivation temperature [40, 41]. Transcription of gcy-8 might be regulated by GCY-18 and TAX-4 (S4A Fig). Importantly, PDE-5 and PDE-1 collaborate to suppress Ca²⁺ level under threshold temperature, which seems to be essential for thermotaxis.