Gastrin-releasing peptide promotes suprachiasmatic nuclei cellular rhythmicity in the absence of vasoactive intestinal polypeptide-VPAC2 receptor signaling

Abstract

Vasoactive intestinal polypeptide (VIP) and gastrin-releasing peptide (GRP) acting via the VPAC2 receptor and BB2 receptors, respectively, are key signaling pathways in the suprachiasmatic nuclei (SCN) circadian clock. Transgenic mice lacking the VPAC2 receptor (Vipr2(-/-)) display a continuum of disrupted behavioral rhythms with only a minority capable of sustaining predictable cycles of rest and activity. However, electrical or molecular oscillations have not yet been detected in SCN cells from adult Vipr2(-/-) mice. Using a novel electrophysiological recording technique, we found that in brain slices from wild-type and behaviorally rhythmic Vipr2(-/-) mice, the majority of SCN neurons we detected displayed circadian firing patterns with estimated periods similar to the animals' behavior. In contrast, in slices from behaviorally arrhythmic Vipr2(-/-) mice, only a small minority of the observed SCN cells oscillated. Remarkably, exogenous GRP promoted SCN cellular rhythms in Vipr2(-/-) mouse slices, whereas blockade of BB2 receptors suppressed neuronal oscillations. In wild-type mice, perturbation of GRP-BB2 signaling had few effects on SCN cellular rhythms, except when VPAC2 receptors were blocked pharmacologically. These findings establish that residual electrical oscillations persist in the SCN of Vipr2(-/-) mice and reveal a potential new role for GRP-BB2 signaling within the circadian clock.

Figure 1.

Coherent behavioral rhythms in Vipr2^-/- mice kept in constant darkness predict the presence of SCN firing rate rhythms. WT mice showed strong rhythms in wheel running (a) and multiunit and single-unit electrical activity (b, c) with estimated periods close to 24 h (j). In slices from all behaviorally arrhythmic Vipr2^-/- mice (d), MUA (e) and most single units are also arrhythmic (f, j). However, some Vipr2^-/- mice expressed short-period rhythms in behavior (g), and in SCN slices from these mice, MUA and the discharge of most single units displayed similar short-period rhythmicity (h, i, j). The peak, but not mean, firing rates of SCN neurons from both rhythmic and arrhythmic Vipr2^-/- mice were significantly lower than those from WT animals (k; Tukey's test). ^*p < 0.05 versus WT. MUA and single-unit data are plotted as mean firing rate (in hertz) every minute. Data in k correspond to average single-unit values ± SEM.

Figure 2.

GRP promotes cellular rhythmicity in the Vipr2^-/- SCN. The majority of slices prepared from Vipr2^-/- mice housed under LD (6 of 9) lack rhythms in MUA (a), and most single units detected within these slices are also arrhythmic (b, i). Chronic application of GRP to Vipr2^-/- SCN slices promoted MUA rhythms (c) and robust, short-period rhythmicity in most single units (d, I; χ² test). When applied in the presence of the BB₂ receptor antagonist DPDMB, GRP failed to promote rhythms in SCN multiunit (e) or single-unit (f, i) activity. Another excitatory agent, NMDA, did not promote MUA (g) or single-unit electrical rhythms (h, i) in SCN slices from Vipr2^-/- mice. Ten to 20 min after application of GRP or NMDA, the single-unit firing rate (mean ± SEM) was significantly increased compared with 10 min before drug application, whereas firing rates were significantly decreased in the same time period after GRP application in the presence of DPDMB (j; paired t tests). Whereas none of the drug treatments significantly altered peak or mean daily firing rates (k; Tukey's test; p > 0.05), GRP treatment increased the amplitude of oscillating cells compared with control Vipr2^-/- SCN neurons (k; t test). ^*p < 0.05. Shaded areas in a-h correspond to periods of darkness extrapolated from the previous LD cycle. The bars in c-h indicate timing and duration of drug application. Error bars in j and k correspond to ±SEM.

Figure 3.

GRP-BB₂ receptor signaling partially accounts for SCN electrical activity rhythms in Vipr2^-/- mice. Long-term perfusion of brain slices from behaviorally rhythmic Vipr2^-/- mice (a, d) with the BB₂ receptor antagonist DPDMB prevented MUA rhythms in four to six slices (b) and rendered the majority of single units arrhythmic (c). However, some antagonist-treated slices showed clear circadian rhythms in multiunit and single-unit discharge (e, f). Conversely, in the presence of GRP, robust MUA rhythms were observed in three of four SCN slices (h) from behaviorally arrhythmic Vipr2^-/- mice (g, j) and the majority of single units detected (i). GRP failed to induce MUA rhythms in one slice (k) but caused short-period oscillations (outside the circadian scale) in single-unit discharge (l). The bars in b-l indicate timing and duration of drug application.

Figure 4.

GRP-BB₂ receptor signaling is not essential for circadian rhythms in WT SCN electrical activity. SCN slices prepared from WT mice housed under LD displayed robust rhythms in MUA (a) and single-unit discharge (b, g). Chronic application of the BB₂ receptor antagonist DPDMB (100 nm) to WT slices did not prevent multiunit or single-unit firing rate rhythms (c, d, g). The majority (3 of 4) of WT SCN slices also sustained MUA rhythms when continually perfused with 50 nm GRP (e), whereas the number (No.) of single units showing circadian firing rate rhythms (f) was reduced (g). Neither peak/mean daily firing rates nor the amplitude of rhythmic cells was significantly altered by GRP or DPDMB (h; Tukey's test; p > 0.05). Shaded areas in a-f correspond to periods of darkness extrapolated from the previous LD cycle. The bars in c-f indicate timing and duration of drug application. Error bars in h correspond to ±SEM.

Figure 5.

GRP promotes cellular rhythmicity in the WT SCN in the absence of VPAC₂ receptor signaling. Long-term perfusion of WT slices with the VPAC₂ receptor antagonist PG 99-465 abolishes circadian rhythms in MUA (a) and most single units (b, top, e), although some neurons expressed low-amplitude rhythms (b, bottom). Chronic application of GRP promoted MUA rhythms in VPAC₂ receptor antagonist-treated WT SCN slices (c) and significantly increased the number of rhythmic neurons detected (d, e; χ² test). Single units oscillated with higher amplitude in the presence of GRP compared with PG 99-465-only-treated slices (f; t test). ^*p < 0.05; ^**p < 0.01. The shaded areas in a-d correspond to periods of darkness extrapolated from the previous LD cycle. The bars indicate timing and duration of drug application. Error bars in f correspond to ±SEM.