Figure 4. First-order interneurons contribute to CO₂ avoidance and attraction through experience-dependent modulation of their CO₂-evoked activity.

(A) AIY is inhibited by CO₂ in animals raised at ambient CO₂ and activated by CO₂ in animals raised at high (2.5%) CO₂. Both responses are BAG-dependent. *p<0.05, **p<0.01, ***p<0.001, Kruskal-Wallis test with Dunn’s post-test. n=8–14 animals per genotype and condition.

(B) RIA is activated by CO₂ in animals raised at ambient CO₂, and inhibited by CO₂ in animals raised at high (2.5%) CO₂. Both responses are BAG-dependent. *p<0.05, ***p<0.001, Kruskal-Wallis test with Dunn’s post-test. n=9–10 animals per genotype and condition.

(C) RIG is activated by CO₂ in animals raised at ambient CO₂, but does not respond to CO₂ in animals raised at high (2.5%) CO₂. The response is BAG-dependent. **p<0.01, ***p<0.001, Kruskal-Wallis test with Dunn’s post-test (raised at ambient CO₂) or p=0.1060, unpaired t test (raised at high CO₂). n=8–14 animals per genotype and condition.

(D) AIZ is activated by CO₂ exposure in animals raised at both ambient and high (2.5%) CO₂. Both responses show BAG-dependent and BAG-independent components. **p<0.01, ***p<0.001, one-way ANOVA with Sidak’s post-test. n=10 animals per genotype and condition.

For A-D, calcium responses were measured using the ratiometric calcium indicators yellow cameleon YC3.60 or YC2.12. Left graphs show composite calcium responses to a 20-s pulse of 15% CO₂. Solid lines indicate average calcium responses; shading represents SEM. Blue lines indicate the response of wild-type animals to CO₂; green lines indicate the response of wild-type animals to an air control; orange lines indicate the response of BAG-ablated animals to CO₂; purple lines indicate the response of BAG-ablated animals to an air control. Right graphs show maximum values (for excitatory or neutral responses) or minimum values (for inhibitory responses) of % ΔR/R0 for each animal. Lines show medians and interquartile ranges.

(E) A model for CO₂ response in C. elegans. Animals raised at ambient CO₂ avoid CO₂, while animals raised at high CO₂ (2.5% CO₂) are attracted to CO₂. CO₂ response valence is determined by the coordinated activity of three interneuron pairs postsynaptic to the CO₂-sensing BAG neurons: AIY, RIA, and RIG. In animals raised at ambient CO₂, activation of RIG and RIA combined with inhibition of AIY results in CO₂ avoidance. In animals raised at high CO₂, activation of AIY, inhibition of RIA, and silencing of RIG results in CO₂ attraction. Activation of a fourth interneuron pair, AIZ, dampens behavioral sensitivity to CO₂ regardless of valence. CO₂ response is regulated by a combinatorial code of neuropeptides: NLP-1 reduces CO₂ avoidance in animals raised at ambient CO₂, FLP-16 reduces CO₂ attraction in animals raised at high CO₂, and FLP-27 enhances CO₂ response under both conditions.

See also Figures S3 and S4.