Figure 1.

Odor response profiles of IR-expressing coeloconic sensilla neurons. Electrophysiological responses in the four antennal coeloconic sensilla classes (ac1–ac4) to 168 odors (mean ± SEM; n ≥ 4, mixed genders), representing the summed, solvent-corrected activities of the IR-expressing neurons they house, indicated in the diagrams at the top [the broadly expressed IR8a and IR25a coreceptors are not indicated (Abuin et al., 2011)]. We quantified summed responses because the small and overlapping spike amplitude of coeloconic neurons makes reliable spike sorting extremely difficult (Benton et al., 2009); a previous analysis was able to distinguish only two of the three neurons in ac1 and ac2 sensilla and was unable to discriminate the three OSNs in ac4 (Yao et al., 2005). Negative values indicate responses smaller than those evoked by the corresponding solvent, most likely attributable to suppression of basal activity in these neurons (Yao et al., 2005). Solvent responses are as follows: ac1, water, 19.69 ± 3.06; paraffin oil, 19.69 ± 3.25; ac2, water, 23.33 ± 4.79; paraffin oil, 28.22 ± 4.92; ac3, water, 14.39 ± 2.47; paraffin oil, 16.48 ± 2.16; ac4, water, 6.06 ± 1.91; paraffin oil, 1.56 ± 1.36. The summing of responses from neurons within a sensillum that display opposite changes in spike frequency will unavoidably lead to underestimations of evoked responses, but this is likely to affect only a very small minority of odor/neuron pairs. Odorants are color coded by functional group; chemicals containing multiple distinct functional groups were classified according to the IUPAC nomenclature of organic chemistry (Panico et al., 1993). We used a wild-type strain for all sensilla except for ac3, in which we recorded from a mutant line, OR35a^f02057, which lacks the function of this broadly tuned OR (Yao et al., 2005). Data for ac4 sensilla are in part adapted from our unpublished observations (Y.G., R.R., and R.B.).