mcu-1 is required in the AWCON sensory neuron at the adult stage for aversive odor learning.

(A) Each C. elegans chemosensory neuron pairs detect distinct attractive odorants. (B) Schematic for chemotaxis assay and chemotaxis index (CI) calculation. (C) Chemotaxis index for naïve N2 and mcu-1(tm6026) worms. (D) Schematic for aversive odor learning. (E) Chemotaxis index in worms after 60 min odor conditioning. (F,G) Aversive odor learning in neuron(rab-3p)- and AWC(ceh36p)-specific rescue strains for benzaldehyde(BZ) and butanone(BU). (H) Chemotaxis index of transgenic strains expressing MCU-1 under the heat-shock-inducible promoter (hsp-16.2p). Worms were given heat shock at different larval stages and tested as day 1 adults. P values were determined using Student’s t-test (C, E, comparison between WT and mcu-1 in H) or one-way ANOVA with Dunnett’s multiple comparisons test (F, G, comparison between larval-stage-heatshock in H). Asterisks indicate p value (ns, not significant; ****p<0.0001).

Mutants of mcu-1 are specifically defective for 60 min odor learning.

(A) Aversive odor learning was conducted with varied conditioning times. (B,C) Chemotaxis index after different conditioning times (ný6 trials per group). Statistics were (D) Wild type worms treated with pharmacological blockers of MCU, ruthenium red(RR) and RU360, display the same learning defect to butanone as mcu-1 mutants. (E) nlp-1 mutant is also defective for aversive learning after 60 min conditioning, but not after 90 min. P values were determined using Student’s t-test (B,C,E) or one-way ANOVA with Dunnett’s multiple comparisons test (D). Asterisks indicate p value (ns, not significant; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001).

MCU activation or butanone exposure causes NLP-1 secretion from AWC neurons.

(A) Diagram of NLP-1 secretion from AWC and uptake by coelomocytes. (B) NLP-1::Venus fluorescence measured in coelomocytes with and without 10 min treatment with the MCU activator kaempferol. (C) NLP-3::Venus fluorescence in coelomocytes with and without 10 min kaempferol treatment. (D) NLP-1::Venus fluorescence in coelomocytes after butanone conditioning. P values were determined using Student’s t-test (B,C) or one-way ANOVA with Dunnett’s multiple comparisons test (D). Asterisks indicate p value (ns, not significant; ***p<0.001; ****p<0.0001).

Odor conditioning causes mtROS production in AWC.

(A) roGFP localized to AWC mitochondria was used to determine the redox ratio after exposure to 10 min H2O2 or 60 min butanone. (B) NLP-1 secretion is prevented by treatment with the antioxidants NAC and mitoTEMPO during odor conditioning. (C) Antioxidant treatment inhibits 60 min aversive odor learning, but not 90 min learning. (D,E) 10 min treatment with H2O2 is sufficient for NLP-1 secretion in both wild type and mcu-1 mutant. (F) Expression of mitochondrially targeted miniSOG in the AWC neuron is sufficient for NLP-1 release under ambient light. (G) Change in chemotaxis index after 10 min odor conditioning. Activation of miniSOG by light causes fast odor learning. P values were determined using Student’s t-test (C,D,E,F) or one-way ANOVA with Dunnett’s multiple comparisons test (A,B,G). Asterisks indicate p value (ns, not significant; *p<0.05; **p<0.01; ***p<0.001).

Coincident mtROS production and odor exposure is sufficient for odor learning.

(A) Kaempferol treatment during 10 min odor conditioning results in strong learning. (B) Calcium response to odor addition and removal in the AWC neuron (n=12). (C) Effect of RU360 treatment at different time points during the 60 min butanone conditioning on odor learning. (D) NLP-1 secretion in worms treated with RU360 during the last 15 min of 60 min butanone conditioning. (E) Model of MCU activation in response to prolonged odor stimulus, which results in NLP-1 secretion and odor learning. P values were determined using one-way ANOVA with Dunnett’s multiple comparisons test. Asterisks indicate p value (ns, not significant; **p<0.01; ***p<0.001).

AWC morphology in wild type and mcu-1 mutant worms.

Strong blue light is required for mito-miniSOG to cause mitochondrial loss and fragmentation.

GFP labeling of mitochondria in the AWC neuron show that exposure to ambient light is not sufficient to cause cell ablation (blue light -). Irradiation with intense blue light for 30 min resulted in severe loss and fragmentation of mitochondria when observed immediately after the treatment (blue light +). Scale bars indicate 20 μm.