Odors detected by mice deficient in cyclic nucleotide-gated channel subunit A2 stimulate the main olfactory system

Abstract

It is believed that odor transduction in the mammalian main olfactory system only involves the cAMP-signaling pathway. Here, we report on odor responsiveness in mice with a disrupted cyclic nucleotide-gated (CNG) channel subunit A2. Several odorants, including putative pheromones, can be detected and discriminated by these mice behaviorally. These odors elicit responses in the olfactory epithelium, main olfactory bulb, and olfactory (piriform) cortex of CNGA2 knock-out mice. In addition, responses to odors detected by CNGA2 knock-out mice are relatively insensitive to inhibitors of the cAMP pathway. These results provide strong evidence that cAMP-independent pathways in the main olfactory system of mammals participate in detecting a subset of odors.

Figure 3.

Odor-evoked EOG in adult CNGA2 knock-out and wild-type (WT) control mice and pharmacology in wild-type mice. A, Representative traces of EOG responses in wild-type and CNGA2 knock-out mice. Top, Responses to 100 μm odorants and 1 μm forskolin in wild-type mice (1 μm is not a saturating concentration of forskolin). Wild-type mice responded to all odorants tested with relatively large changes in field potentials. Bottom, CNGA2 knock-out mice selectively responded to 100 μm 2-heptanone and DMP but failed to respond or responded only weakly to other odorants at 500 μm. Application of Ringer's solution did not produce changes in field potential (control traces). Responses in wild-type and knock-out animals are presented in different scales. B, Average absolute value of the peak EOG responses to different concentrations of odor in CNGA2 knock-out mice (right panel) compared with the responses of wild-type mice at 100 μm odor concentration (left panel). The number of mice tested is indicated under each bar. Those values that were significantly different from the EOG elicited by forskolin in CNGA2 knock-outs (i.e., significantly different from no response) are identified with an asterisk. C, Concentration dependence for percentage inhibition of EOG responses by the adenylyl cyclase inhibitor SQ22536 in wild-type mice. Mean ± SEM (n = 5). Dose-response for forskolin was fit by a Hill equation with EC₅₀ of 72 μm, a Hill coefficient of 0.5, and a maximum percentage inhibition of 100%. Dose-response curves for all other compounds were fit with a Hill equation using the same EC₅₀ and Hill coefficient but with different maximum percentage inhibition: geraniol (77%), lilial (86%), 2-heptanone (43%), and DMP (39%). Solid lines, Curves for the best fit of the Hill equation to the data for the different odorants; black squares, forskolin; red circles, lilial; green triangles, geraniol; upside-down dark blue triangles, 2-heptanone; light blue diamonds, DMP. D, The CNG channel blocker l-cis-diltiazem (50 μm) and the adenylyl cyclase inhibitors SQ22536 (300 μm) and MDL-12330A (50 μm) partially reduced odorant (100 μm)- and forskolin (1 μm)-evoked responses in wild-type mice. Black, Forskolin; red, lilial; green, geraniol; dark blue, 2-heptanone; light blue, DMP. Note the differential effect on responses to 2-heptanone and DMP compared with forskolin, geraniol, and lilial. Number of experiments per condition is four for l-cis-diltiazem, 10 for SQ22536, and five to seven for MDL-12330A. For each drug, the percentage inhibition of the responses to forskolin, lilial, and geraniol was significantly larger than the percentage inhibition of the responses to 2-heptanone and DMP. Application of the PLC inhibitor U73122 (10 μm) (D, bars on right) produced further significant suppression of responses to 2-heptanone and DMP in the presence of MDL-12330A (50 μm). The preparation was preincubated with the inhibitors for 15 min. Inhibition by the drugs was partially reversible (the percentage inhibition decreased by 50% after a 10 min wash with Ringer's). All responses used for comparison in a single animal were obtained strictly from the same recording position with one recording electrode. Black asterisks, DMP and 2-heptanone are significantly different from forskolin, geraniol, and lilial; red asterisks, suppression by U73122 + MDL-12330A on DMP and 2-heptanone was significantly different from the suppression by MDL-12330A alone. Ba²⁺ (5 mm) abolished the EOG of all odors tested (data not shown).

Figure 1.

Detection and discrimination tests and innervation in the olfactory bulb in 3-MI-treated mice. A–C, Results of odor detection test for CNGA2 knock-out mice and controls (S + stimulus, odor; S–stimulus, air). The percentage of correct responses is plotted as a function of block number (each blockh as 20 trials); 50% represents chance response (horizontal line). Filled circles, Wild-type mice; open circles, CNGA2 knock-out mice. Stimuli were a 1/40 dilution of air equilibrated with 1% geraniol, 1% lilial, and 1% ethyl acetate. Mean ± SEM (n = 5–7 mice each). D, Odor discrimination test. Mice were water deprived and trained to detect ethyl acetate (1/40 dilution of 1%) as the S + stimulus and air as the S–stimulus. In the third block, the S–stimulus was abruptly changed to 1% propyl acetate (PA). Although the knock-out mice took longer in making the discrimination (ANOVA; p < 0.05 for phenotype), both CNGA2 knock-outs and controls were able to detect the difference between the two odors (mean ± SEM; n = 5). E, F, HRP reactivity in olfactory bulb of control (E) and 3-MI-treated (F) mice. 3-MI treatment induced denervation of all glomeruli in the main olfactory bulb but did not affect the afferent innervation of the accessory olfactory bulb (arrow).

Figure 2.

Measurement of differences in trigeminal sensitivity between wild-type and CNGA2 knock-out mice. Trigeminal stimuli elicit a depression of the rate of respiration attributable to a reflex response. Respiration was monitored by recording the changes in voltage in a thermocouple located at the naris of an anesthetized mouse (see Materials and Methods). A and B show the voltage from the thermocouple as a function of time in the absence (A) and presence (B) of air equilibrated with (1/1) × 10% ethyl acetate. Calibration: (in A) 50 μV, 1 sec. An increase in temperature produces a decrease in voltage. C and D show the change in respiratory frequency as a function of time before (open squares) and during (open circles) application of (1/1) × 10% (C) or (1/1) × 0.1% (D) ethyl acetate. E and F show the absolute value of the rate of change in respiratory frequency as a function of the concentration of ethyl acetate for wild-type (E) and CNGA2 knock-out (F) mice. The EC₅₀ values were 2.5 and 1.1% ethyl acetate for wild-type and CNGA2 knock-out mice, respectively. Mean ± SEM (n = 5 for wild-type; n = 3 for CNGA2 knock-out).

Figure 4.

Odor-evoked Fos activity in the MOB of wild-type and CNGA2 knock-out mice. Light micrographs of Fos DAB reaction from representative sagittal sections. A–D are sections from mice exposed to a mixture of 2-heptanone and DMP [(1/40) × 0.5% each]. Micrographs are representative of three independent experiments for each condition. A, Sagittal section of bulb ipsilateral to naris occlusion in a wild-type mouse. Arrowheads indicate glomeruli that were active in C but not D. Arrows indicate glomeruli activated in both C and D. B, Section ipsilateral to naris occlusion in a CNGA2 knock-out mouse. C and D are sections of the bulb ipsilateral to naris that was not occluded. C, Wild-type. D, CNGA2 knock-out. E, CNGA2 knock-out exposed to fresh air (representative of 3 independent experiments). F, CNGA2 knock-out exposed to (1/40) × 1% lilial (representative of 2 independent experiments). G, H, CNGA2 knock-out exposed to (1/40) × 1% ethyl acetate (representative of 2 independent experiments). G, Left bulb. H, Right bulb.

Figure 5.

The mixture of (1/40) × 0.5% DMP and 2-heptanone activated both GFP-positive and PDE2-positive glomeruli in CNGA2-GFP knock-out mice. A and B are black and white micrographs showing Fos immunoreactivity (black nuclei surrounding glomeruli; arrows point to glomeruli) and GFP labeling denoting axons from ORNs that would express CNGA2 in wild-type mice (white). Odor-activated glomeruli are located symmetrically in the left (A) and right (B) bulbs (arrows). GFP fluorescence images were overlaid with transmission light microscopy images. Numerous distinct glomeruli are labeled by GFP, and some are surrounded by Fos-positive juxtaglomerular cells. C, A high-magnification confocal image showing a GFP-positive glomerulus (in green) surrounded by many Fos-labeled nuclei, indicating the glomerulus was activated by the odorant exposure. GFP (green channel) was overlayed on a Nomarski image. Fos reaction (DAB) appears as dark areas. D, A confocal micrograph showing two activated glomeruli surrounded by Fos-labeled nuclei (in blue). One of these glomeruli was GFP positive. The other was GFP negative but was positive for PDE2 Ab (in red), indicating this was a necklace glomerulus activated by DMP and 2-heptanone. The image was obtained from a medial section of the bulb. Scale bars: A, B, 100 μm; C, D, 20 μm.

Figure 6.

Tyrosine hydroxylase immonoreactivity. A, B, Low-magnification fluorescence micrographs. A, Wild-type. B, CNGA2-GFP knock-out. CNGA2 knock-out GFP mice showed general reduction in TH reactivity of regular glomeruli (B) as compared with wild-type mice (A). The TH immunoreactivity surrounding necklace glomeruli remained strong (B, arrows). High-magnification images show relatively uniform TH reactivity surrounding glomeruli of control mouse (C) and heterogeneous reactivity in the knock-out mouse (D). C and D are higher magnification florescence micrographs. D was taken in the area shown in the box in B. Ipsilateral naris occlusion reduced TH activity (F, red) as compared with nonoccluded side (E) in necklace glomeruli (positive for PDE2; green) from the same CNGA2 knock-out mouse. These two pictures were taken under the same excitation intensity and exposure conditions. Scale bars: A, B, 100 μm; C–F, 20 μm. Data are representative of three independent experiments.

Figure 7.

Fos immunoreactivity in AON (L = 0.21–0.26 mm; bregma = 1.5–2.2 mm) (A and B and lower magnification in C and D) and piriform cortex (Pir; L = 1.36–1.39 mm; bregma = 1.6–1.9 mm) (E, F). A, C, and E are micrographs taken in a CNGA2 knock-out mouse exposed to fresh air, whereas B, D, and F are micrographs from the same areas in a CNGA2 knock-out mouse exposed to (1/40) × 0.5% DMP and 2-heptanone. These micrographs are taken in 30 μm brain slices cut sagittally. Scale bars: A, B, 20 μm; C–F, 50 μm. Results are representative of three independent experiments.