Daily rhythms in olfactory discrimination depend on clock genes but not the suprachiasmatic nucleus

Abstract

The suprachiasmatic nucleus (SCN) regulates a wide range of daily behaviors and has been described as the master circadian pacemaker. The role of daily rhythmicity in other tissues, however, is unknown. We hypothesized that circadian changes in olfactory discrimination depend on a genetic circadian oscillator outside the SCN. We developed an automated assay to monitor olfactory discrimination in individual mice throughout the day. We found olfactory sensitivity increased approximately 6-fold from a minimum during the day to a peak in the early night. This circadian rhythm was maintained in SCN-lesioned mice and mice deficient for the Npas2 gene but was lost in mice lacking Bmal1 or both Per1 and Per2 genes. We conclude that daily rhythms in olfactory sensitivity depend on the expression of canonical clock genes. Olfaction is, thus, the first circadian behavior that is not based on locomotor activity and does not require the SCN.

Figure 1

Circadian rhythms in olfactory discrimination. (A) Wild-type (wt) mice (n = 18) trained to detect vanilla diluted 400-fold showed a higher proportion of correct responses (mean ± SEM) during the early subjective night. Npas2−/− (n = 6) were also circadian, whereas Per1^mPer2^m (n = 5) and Bmal1−/− (n = 5) mice were arrhythmic. (B) Representative olfactory discrimination rhythms from one mouse (top) showed peak olfactory discrimination shortly after the daily onset of locomotor activity (bottom) over multiple days in constant darkness. The gray and black bars indicate subjective day and night, respectively.

Figure 2

Olfactory sensitivity peaks at night. Sensitivity increased by about 6-fold during the early night (CT16) in wild-type (wt) mice (n = 15). Mice were able to discriminate vanilla dilutions below 2:10⁴ at CT16 but not at CT8. Dashed lines indicate the dose required to evoke a response above threshold (75% correct) and reflect the day-night shift in sensitivity.

Figure 3

IndividualPer1^mPer2^m (left column; n= 5) and Bmal1−/− (right column; n = 5) mice showed no daily rhythms in olfactory discrimination.

Figure 4

Circadian cycling of olfactory discrimination in SCN-lesioned mice. Olfactory performance in SCNx (A–E) and control mice (F–I) was scored as circadian in all mice except mouse D. In contrast to control animals, mice tested 8 to 10 days after SCN ablation varied in the time of their daily peak in olfactory sensitivity. (J) Running-wheel activity from SCN-lesioned mouse E showing representative arrhythmic locomotor activity. * denotes time of surgery.

Figure 5

Neither decision latency (A) nor sniff duration (B) was circadian in wild-type (wt) mice (n = 15). The time spent in the odor port (sniff duration, C) and required to poke in a side port (decision latency, D) did not vary with time of day in individual SCNx mice (n = 5; dashed lines) or controls (n = 5 representative mice; continuous lines), indicating that daily rhythms in olfactory performance did not arise from changes in motivation or alertness.