Vasoactive intestinal polypeptide mediates circadian rhythms in mammalian olfactory bulb and olfaction

Abstract

Accumulating evidence suggests that the olfactory bulbs (OBs) function as an independent circadian system regulating daily rhythms in olfactory performance. However, the cells and signals in the olfactory system that generate and coordinate these circadian rhythms are unknown. Using real-time imaging of gene expression, we found that the isolated olfactory epithelium and OB, but not the piriform cortex, express similar, sustained circadian rhythms in PERIOD2 (PER2). In vivo, PER2 expression in the OB of mice is circadian, approximately doubling with a peak around subjective dusk. Furthermore, mice exhibit circadian rhythms in odor detection performance with a peak at approximately subjective dusk. We also found that circadian rhythms in gene expression and odor detection performance require vasoactive intestinal polypeptide (VIP) or its receptor VPAC2R. VIP is expressed, in a circadian manner, in interneurons in the external plexiform and periglomerular layers, whereas VPAC2R is expressed in mitral and external tufted cells in the OB. Together, these results indicate that VIP signaling modulates the output from the OB to maintain circadian rhythms in the mammalian olfactory system.

Keywords: VIP; circadian; clock; olfaction; olfactory discrimination; rhythms.

Figure 1.

Representative recordings of PER2::LUC expression in cultured tissues from the mouse olfactory system reveal intrinsic circadian rhythms in the isolated OE and OB but not in the PC.

Figure 2.

Circadian rhythms in PER2 expression from the OB in vivo. A, A drawing of the in vivo bioluminescence imaging setup. At 4 h intervals over 56 h, Vip^+/−;PER2::LUC mice were briefly anesthetized and transferred to a light-tight imaging box in which bioluminescence from the bilateral OBs was imaged through a cranial window in DD. B, A drawing of a mouse positioned in the imaging box with a cranial window (left) and a pseudocolored image of its bioluminescent signal from 10 min exposures (right). C, Pseudocolored images of bioluminescence recorded every 4 h from the OB of mouse 1 after 2 weeks in DD and bioluminescence levels show a daily rhythm peaking in the late subjective night (CT 20; top). PER2 expression was circadian in all seven mice with a twofold to fivefold increase from day to night. D, Bioluminescence from each mouse was normalized (minimum is 0 and maximum is 1) and then averaged in 2 h bins across mice maintained in DD for 2 weeks. PER2 levels significantly differed from peak-to-trough (*p < 0.05 for CT 17 and CT 1 on the first imaging day and CT 17 on the second day vs CT 5 and CT 9 on the first day and CT 5 on the second day; *p < 0.05 for CT 21 on the second day vs CT 5 and CT 9 on the first day; **p < 0.005 for CT 21 on the first day vs CT 13, CT 5, and CT 9 on the first day and CT 5, CT 9, and CT 13 on the second day; n = 7 mice). CT 24 denotes the daily offset of locomotion (subjective dawn), and CT 12 denotes the daily onset of locomotion (subjective dusk). Gray and black bars denote subjective light and dark times, respectively. Counts are the mean number of photons integrated over 10 min.

Figure 3.

Both VIP and VPAC2R are expressed in the mouse OB. Representative confocal images of OB sections stained with antibodies against VPAC2R (magenta) and OMP (green; A), and Vip;tdtomato (magenta) and OMP (green; C). VPAC2R staining was intense in the ET and MT cells (magenta, A, B, F; green, E). VIP staining or Vip;tdtomato expression was present mainly in the PGL and EPL (magenta, C–E). OB sections of Gad65–EGFP mice stained with antibodies against VPAC2R (magenta) or VIP (magenta) revealed that a large fraction of VIP-positive neurons was GABAergic (D), whereas none of VPAC2R-positive neurons were GABAergic (B). Protocadherin-21 (Pcadh21) labeling for MT/ET cells (F) shows that VAPC2R staining was specific to a subset of these neurons. Arrows in D and F indicate examples of cell bodies with both VIP staining and Gad65–EGFP expression and examples of MT/ET cells expressing VPAC2R, respectively. Representative neurons located mainly in the EPL immunolabeled for VIP (G, arrows) at four CTs. The quantification of the integrated intensity of immunolabeled somata in the OB shows a circadian rhythm with a peak in VIP expression at CT 18. *p < 0.05, CT 18 versus CT6. Mean ± SEM. GL, Glomerular layer.

Figure 4.

Circadian rhythmicity in the OB requires VIP. A, After 2 weeks in DD, PER2::LUC bioluminescence from the OB (pseudocolored for mouse 1) was arrhythmic in six of seven Vip^−/− mice. B, Bioluminescence was normalized in individual mice and then averaged across mice. Mice were maintained under DD for 2 weeks so they were on different CTs during imaging sessions. Each time point was pooled within a 2 h window. Mean bioluminescence did not significantly differ across CT (n = 7 mice). Gray and black bars denote subjective day (CT 0–CT 12) and night (CT 12–CT 24), respectively. Counts are the mean number of photons integrated over 10 min.

Figure 5.

Both Vip^+/− and Vip^−/− mice exhibited a circadian rhythm in PER2::LUC bioluminescence in vivo in a light/dark cycle. Gray bars denote the times of lights on and black bars the times of darkness. Bioluminescence was normalized (minimum is 0 and maximum is 1) in individual mice and then averaged across mice of the same genotype. ZT 0 and ZT 24 denote 7:00 A.M., and ZT 12 denotes 7:00 P.M. in a light/dark cycle. Mean ± SEM.

Figure 6.

Circadian rhythms in odor detection were blunted in Vip^−/− and Vipr2^−/− compared with WT mice after 48 h in DD. WT mice showed a peak during the subjective night in performance at all five concentrations of vanilla tested (A). Representative mice in each group showing odor detection performance at 1:10¹ vanilla dilution (B). The gray and black bars indicate subjective day and night, respectively. Mean ± SEM. n is number of mice per group.

Figure 7.

Proposed circuit for circadian modulation of olfactory sensitivity. Within the OB, circadian variations in VIP release from a subset of interneurons regulate excitability of MT and ET cells by acting on VPAC2R expressed on their dendrites and cell bodies. EPC, external plexiform layer cell; GC, granular cell; GL, glomerular layer; GRL, granular layer; ML, mitral cell layer; ORN, olfactory receptor neuron; PG, periglomerular cell; SA, short axon cell. Arrowhead with −, Inhibitory; arrowhead with +, excitatory.