Loss of neuropeptide signals changes daily patterns and activity of mouse brain clock neurons

All experiments were performed in accordance with the UK Animals (Scientific Procedures) Act of 1986 using procedures approved by The University of Manchester. Age‐matched adult (>8 weeks of age) male mice including those with intact neuropeptide signalling (Vipr2^+/+; n = 23) as well as mice lacking VPAC₂ expression (Vipr2^−/−; n = 27) were used (see below). Animals were group housed under a 12 h light/12 h dark (LD) cycle. In LD conditions, lights‐on was defined as Zeitgeber time 0 (ZT0) and lights‐off as ZT12. Food (Bekay, B&K Universal, Hull, UK) and water were available ad libitum, the temperature was maintained at 20 ± 2°C and humidity at ~40%. Light intensity at the cage level in the breeding colony was ~45μWcm⁻². The original breeding stock of Vipr2^−/− animals was kindly provided by the late Prof. Tony Harmar, University of Edinburgh (see (Harmar et al., 2002) for details on the creation of this mouse line). Breeding stock for the mPer1::d2EGFP expressing or Per1::eGFP) mice was a gift from Prof. Douglas McMahon, Vanderbilt University (see (Kuhlman et al., 2000) for details on the creation of these mice). To generate Vipr2^−/− mice in which Per1 gene expression could be visualized at single cell level in brain slices, Vipr2^−/− mice were crossed with mPer1::d2EGFP (Per1::GFP) expressing mice. Subsequently, animals homozygous for the disrupted Vipr2^−/− transgene (Vipr2^−/− × Per1::GFP; referred to as Vipr2^−/−) or homozygous for non‐disrupted, Vipr2 gene (Vipr2^+/+ × Per1::GFP strain; referred to as Vipr2^+/+) were used.

To assess daily locomotor rhythms, Vipr2^−/− and Vipr2^+/+ animals were singly housed in running wheel‐equipped cages with food and water available ad libitum. Rotation of the wheel was recorded using the acquisition suite within Chronobiology Kit (Stanford Software Systems, Santa Cruz, CA., USA). Mice were initially maintained under a 12 h light:12 h dark light–dark (LD) cycle for 2–3 weeks and then transferred into a constant dark (DD). To assess potential genotype differences in entrainment to the LD cycle, two measures were used: 1) the portion of wheel‐running occurring during the lights‐on and lights‐off phases (as extracted using the Chronobiology Kit) and 2) the time of activity onset on the first day following transfer to DD conditions as manually assessed by an experienced observer.

Mice were taken from the vivarium either early (ZT1–4; daytime recordings) or late in the day (ZT9–11; night‐time recordings) and deeply anaesthetized with isoflurane (Abbott Laboratories, Kent, UK) prior to cervical dislocation. Following removal, brains were placed in a chilled (4°C) incubation artificial cerebrospinal fluid (aCSF) containing in mM: NaCl 95; KCl 1.8; KH₂PO₄ 1.2; MgSO₄ 7; NaHCO₃ 26; Glucose 15; Sucrose 50; CaCl₂ 0.5; phenol red 0.0014, pre‐gassed with 95% O₂ and 5% CO₂. Coronal slices (250 μm) were cut with a Campden 7000smz vibrating microtome (Campden Instruments Ltd., Leicestershire, UK). Slices were maintained in carbogen‐gassed aCSF at room temperature in a custom‐made brain slice keeper for 1–4 h prior to transfer to the recording chamber. Only slices corresponding to the intermediate level of the rostrocaudal axis of the SCN were used. Typically, one to two slices were used per mouse, with one to six neurons recorded per slice.

Slices containing the mid‐coronal SCN section were incubated for at least 1 h in pre‐gassed recording aCSF (containing in mM: NaCl 127; KCl 1.8; KH₂PO₄ 1.2; MgSO₄ 1.3; NaHCO₃ 26; Glucose 15; CaCl₂ 2.4; phenol red 0.0014) at room temperature within the recording chamber mounted on the stage of an Olympus BX51WI upright microscope (Olympus Corporation, Tokyo, Japan). The aCSF flow rate was 2–3 ml/min for all experiments. The microscope was equipped with inbuilt infrared differential contrast optics mounted on a vibration‐free air table (TMC, MA, USA). A Hitatchi C106005 CCD camera system was used for visualization of the cells on a high‐resolution black/white cathode ray monitor (Hitatchi Ltd., Tokyo, Japan). A Hamamatsu Orca R² CCD camera system (Hamamatsu Photonics, Hamamatsu City, Japan) was used in combination with a X‐Cite 120 Q fluorescence light source (Excelitas, Waltham, MA, USA) for visualizing and discriminating the Per1:EGFP+ve from Per1::EGFP‐ve neurons. Unfortunately, the Per1‐driven signal was reduced in SCN slices from Vipr2^−/− mice with few EGFP+ve cells per slice and consequently for analysis, we did not distinguish between EGFP+ve and ‐ve neurons and instead for both genotypes, we combined the data from these recordings. For daytime, recordings were made from ZT4 to ZT11.5, while for night‐time, recordings were made from ZT13 to 21. Photographs of the patch pipette and the puff pipette were taken in situ for accurate anatomical documentation of the recorded neurons within the SCN. Current‐clamp (CC) recordings were acquired as described previously (Belle et al., 2009, 2014; Timothy et al., 2018; Wegner et al., 2017). Briefly, current‐clamp (CC) recordings were done with a npi BA‐03X bridge amplifier (npi electronic GmbH, Tamm, Germany) or an Axoclamp 2A amplifier (Axon Instruments/Molecular Devices, Sunnyvale, USA) through a CED 1401 mk II A/D interface controlled by Spike2 software (Cambridge Electronic Design, Cambridge, UK). In pilot studies, no differences in the recorded parameters of SCN neurons were observed between the two amplifiers.

Voltage Clamp (VC) recordings were obtained with an Axopatch 200B amplifier (Molecular Devices) using methods adapted from (Jackson et al., 2004). Voltage‐activated ionic currents were evoked with fast depolarizing voltage steps (250 ms in 10 mV increments) applied from −80 to +40 mV from a holding potential of −70 mV. Patch pipettes (7–10 MΩ for CC recordings, 5 MΩ for VC recordings to reduce pipette capacitance) were pulled from thick‐walled borosilicate glass capillaries (GC150F‐10, Harvard Apparatus LTD, Kent, UK). Pipettes were filled with intracellular solution containing in mM: K‐Glutamate 130, KCl 10, MgCl2 2, Hepes 10, EGTA 0.5, K2ATP 2, NaGTP 0.5. All recordings were made at ~23°C. Access resistance was typically ~15 MΩ and series resistance (R_S) ~ 20 MΩ, with R_S optimally compensated in all VC recordings. Neurons were omitted from analysis if the series resistance value during the recording changed by more than 15%.

All drugs were purchased from Tocris bioscience, UK. For stock solutions of α‐amino‐3‐hydroxy‐5‐methyl‐4‐isoxazolepropionic acid (AMPA) hydrobromide and tetrodotoxin (TTX) citrate powder was dissolved in distilled water. The working dilution for these drugs was prepared on the day of the experiment in pre‐gassed aCSF using established protocols (Belle et al., 2014; Itri et al., 2004). TTX was delivered by gravity‐controlled bath application. AMPA was delivered by an eight‐channel valve‐controlled pressurized perfusion system (ALA‐VM8, BPS‐8) through a QMM Quartz MicroManifold (ALA Scientific Instruments, New York, USA).

Data from CC recordings were analysed with Spike2 software (versions 6 and 7, Cambridge Electronic Design, Cambridge, UK) and the spontaneous firing rate (SFR) and the resting membrane potential (RMP) of the cells were measured. Both parameters were determined within 2–5 min establishing the whole‐cell configuration, and prior to investigations of the current injection. The RMP of these cells was determined by phase–plot averages using a previously described algorithm (Belle et al., 2009). Cells were removed from analysis if the recorded RMP was unstable, e.g., sudden, unaccountable changes of >3 mV or alterations in input resistance >15%. Only cells that were discharging APs or silent because of hyperpolarization were included for SFR analysis. Cells in depolarized states including those exhibiting depolarized low amplitude membrane oscillations (DLAMOs) were not used in SFR measurements. To measure the membrane threshold at which neurons initiated or terminated the firing of APs, slow depolarizing and hyperpolarizing current ramps were performed. Briefly spontaneously firing cells were silenced by a ~ 5–10 sec injection of ramping hyperpolarizing current so that the cell no longer fires followed by a depolarizing current ramp (~ 15 sec) to reinitiate firing. This membrane potential at which the neuron began firing again was interpreted as the AP firing threshold.

VC I‐V curves were analysed in Clampfit10.3 (Molecular Devices). For all statistical comparisons peak inward currents at either −20 mV or −10 mV holding potential were used (Jackson et al., 2004).

Detailed statistical analyses were conducted using GraphPad Prism 10 (GraphPad Software, Boston, MA, USA), Kaleidagraph 5.0 (Synergy Software, Reading, PA, USA) and SPSS (Statistical Product and Service Solutions, IBM) as well as on‐line resources (www.vassarstats.net↗; www.estimationstats.com↗). Fisher's exact test (www.vassarstats.net↗) was employed to assess genotype and time of day differences in the proportions of SCN neurons in different states. To evaluate genotype and time of day effects on the properties of SCN neurons, the normal distribution of the data was assessed using the Kolmogorov–Smirnov test with normally distributed data subsequently analysed using two‐way analysis of variance (ANOVA) and post hoc pairwise comparison tests (Bonferroni corrected). In cases where the data did not exhibit a normal distribution, generalized linear model (GLM)‐ Type III tests followed by Bonferroni post hoc tests were used. To estimate the effect size in the ANOVA and GLM analysis, we used paired differences within the estimation statistics framework (www.estimationstats.com↗) (Ho et al., 2019). Estimation statistics report mean differences (effect size) as well as with confidence interval estimates to express uncertainty. This methodology generates a bootstrap sampling distribution for each paired mean difference, employing 5000 bootstrap samples, and then bias‐correcting and accelerating the confidence intervals.

For most experimental components in which the same neurons were challenged with different doses of drug, genotype and dose effects were evaluated using two‐way repeated measures ANOVA with post hoc Bonferroni or Tukey tests. To generate dose–response relationships, doses were log10 transformed and fitted with the Hill slope constrained to 2 (Graphpad Prism). T‐tests (Kaleidagraph) were also used to test group differences in tetrodotoxin studies. For one experiment using a VPAC₂ antagonist, we used independent t‐tests. For all statistical analyses, the threshold for significance was p < 0.05. A detailed breakdown of the statistical analysis is reported in Supplemental Table S1.

Graphs were made using Kaleidagraph or Graphpad Prism or via online (www.estimationstats.com↗). Figures were constructed using Inkscape (www.inkscape.org/↗.) and Canvas X Draw (Canvas GFX, Inc., Boston, MA, USA). In figures, data are plotted either as jitter plot, as mean ± SEM or as mean with fitted distribution and mean ± 95% confidence intervals (estimation plots).

Under 12 h–12 h LD conditions, both Vipr2^+/+ and Vipr2^−/− mice showed apparent entrainment, confining the majority (99.6 ± 0.17% and 85.5 ± 5.3%, respectively) of their daily wheel‐running activity to the lights‐off phase (Figure 1a, b; Vipr2^+/+ vs Vipr2^−/−; t = 2.67, df = 4, p = 0.055). On release into DD, clear and significant genotype differences emerged. In the first 24 h period in DD, Vipr2^−/− mice initiated vigorous sustained wheel‐running ~9–10 h in advance of the projected time of lights‐off, whereas Vipr2^+/+ animals began wheel‐running ~0.5 h prior to the projected time of lights‐off (Figure 1a,c; Vipr2^+/+ vs Vipr2^−/−; t = 34.84, df = 4, p < 0.001). These observations are consistent with other studies (Aton et al., 2005; Brown et al., 2005; Colwell et al., 2003; Hannibal et al., 2011; Harmar et al., 2002; Hughes et al., 2004; Hughes & Piggins, 2008) and illustrate that under LD conditions, VIP‐signalling deficient mice exhibit only apparent entrainment.

In SCN brain slices prepared from either genotype, we attempted to use epifluorescence to distinguish neurons in which Per1‐driven green fluorescent protein could be visualized (EGFP+ve neurons) from those in which it could not (EGFP‐ve neurons). However, in the Vipr2^−/− SCN, the Per1‐driven EGFP signal was diminished with few cells detectable per slice and consequently we combined data from EGFP‐ve and EGFP+ve neurons. To investigate how the electrical state and membrane properties of Vipr2^+/+ SCN neurons vary across the projected LD cycle and to determine if and how these properties are altered in Vipr2^−/− mice, we made whole‐cell current clamp recordings from SCN brain slices. Using previously published protocols (Belle et al., 2009, 2014; Timothy et al., 2018; Wegner et al., 2017) we classified the spontaneous activity of SCN cells into one of three states: 1) depolarized and not discharging action potentials (silent) or with DLAMOs, 2) spontaneously discharging action potentials (APs) and 3) hyperpolarized and silent (Figure 1d). Circadian clock gene expression is damped in the Vipr2^−/− SCN in vivo (Harmar et al., 2002; Hitrec et al., 2023), and consistent with previous reports (Hughes et al., 2008; Maywood et al., 2006).

All states were recorded from Vipr2^+/+ (n = 82 cells from 11 animals) and Vipr2^−/− SCN (n = 116 cells from 14 animals) neurons and in Vipr2^+/+ but not Vipr2^−/− SCN brain slices, the proportion of cells in the different states was associated with time of day (Fisher's exact test, p = 0.00004 (for Vipr2^+/+ and p > 0.05 [p = 0.382] for Vipr2^−/− slices; Figure 1e). In the Vipr2^+/+ SCN, depolarized cells (silent and DLAMO states) collectively constituted ~30% of neurons recorded during the day, while at night, they were not detected. Neurons in the hyperpolarized and silent state increased from ~2.5% during the day to ~26% at night. By contrast, in Vipr2^−/− brain slices, depolarized states (silent or expressing DLAMOs) were infrequent in the day (~3.4% of cells) and night (~8.9% of cells) phases, while those recorded in the hyperpolarized silent state represented ~33% of cells recorded during the day and ~25% of those recorded at night (Figure 1). During the day, SCN cell state differed from Vipr2^+/+ to Vipr2^−/− mice (p = 4 × 10⁻⁷), while at night, there was no genotype difference (p = 0.155). Thus, day‐night change in the spontaneous state of SCN neurons is altered in the Vipr2^−/− SCN, with Vipr2^−/− neurons expressing a distribution in neuronal states during day and night that resembles that expressed at night in the Vipr2^+/+ SCN (see Supplemental Table S1 for more detailed statistical analysis).

This temporal difference in the timing of neuronal state is also reflected in the resting membrane potential (RMP, Figure 2) and the spontaneous firing rate (SFR, Figure 3) of recorded SCN cells. For RMP, two‐way ANOVA showed a significant main effect of time of day (F_1,165 = 11.37, p < 0.001) as well as time of day x genotype interaction (F₁=13.63, p < 0.001) with RMP varying from day to night in Vipr2^+/+ (−42.8 ± 1.5 vs − 49.2 ± 1.3 mV; mean ± SEM; pairwise comparison, p = 0.017; with an effect size as measured by permutation t‐test, p < 0.0001), but not Vipr2^−/− SCN neurons (−44.3 ± 1.3 mV vs − 48.2 ± 1.3 mV; pairwise comparison, p > 0.05; permutation t‐test, p = 0.825) (Figure 2a,c). RMP varied between Vipr2^−/− and Vipr2^+/+ neurons during day (permutation t‐test, p = 0.0024) or night (permutation t‐test, p = 0.052; Figure 2e). Both time of day (Wald c², p = 0.016) and the time of day x genotype interaction (Wald c², p = 0.013) were factors for SFR which differed from day to night in Vipr2^+/+ neurons (2.14 ± 0.23vs 1.4 ± 0.28 Hz; pairwise comparison, p = 0.015; permutation t‐test, p = 0.0012), but not Vipr2^−/− neurons (1.93 ± 0.3 vs 1.88 ± 0.29 Hz; p > 0.05, permutation t‐test, p = 0.96) (Figure 3a,c). Permutation t‐tests further indicated a significant effect size for genotype during the day (p = 0.032; SFR of Vipr2^−/− neurons lower than that of Vipr2^+/+ neurons), but not the night (p = 0.34; Figures 3e) (see Supplemental Table S1 for more detailed statistical analysis). These findings are broadly consistent with a previous extracellular recording study of Vipr2^−/− mouse SCN neurons (Cutler et al., 2003) and provide evidence that at the ensemble level, Vipr2^−/− SCN neurons do not vary in membrane potential or spontaneous firing rate from day to night.

A cardinal property of circadian clocks is their ability to control if, when, and how they respond to environmental signals. As illustrated above and in other studies (Colwell et al., 2003; Hannibal et al., 2011; Harmar et al., 2002; Hitrec et al., 2023; Hughes et al., 2021; Hughes & Piggins, 2008) disruption of VIP‐VPAC₂ receptor signalling results in abnormal entrainment of mice to the LD cycle as well as impairing encoding of daylength (Lucassen et al., 2012). Further, Vipr2^−/− mice show aberrant responses to acute photic stimuli: 15 min pulses of light can immediately activate intracellular transduction pathways in the Vipr2^−/− SCN at any time of the circadian cycle, whereas activation of these pathways by light is mostly restricted to the night in the SCN of Vipr2^+/+ mice (Hughes et al., 2004). Since the main photic input to the SCN is glutamatergic, we next tested if such photic gating may arise from temporal regulation of individual SCN neurons by neurochemically mimicking the photic input through brief (3 s) pulses of the glutamatergic agonist AMPA to SCN brain slices (Ang et al., 2021; Mizoro et al., 2010). Further, we evaluated how the loss of the VPAC₂ receptor influenced the response of SCN neurons to this neurochemical proxy of the light input pathway.

We evaluated if the amplitude of responses to AMPA of current‐clamp recorded neurons in the ventral SCN (vSCN; the region of the SCN receiving substantial retinal innervation (Abrahamson & Moore, 2001; Antle & Silver, 2005)) differed from day to night as well as with the loss of VPAC₂ receptor signalling. To minimize the influence of RMP and to make recordings comparable between genotypes, all cells were manually held at a membrane potential of ~ − 40 mV and the effects of increasing concentrations of AMPA (1, 5, 10 and 25 μM) tested on Vipr2^+/+ and Vipr2^−/− vSCN neurons recorded during the day or night (Figure 4a,b).

Depolarization in membrane potential in response to AMPA treatment was recorded from both Vipr2^+/+ and Vipr2^−/− SCN neurons (Figure 4a,b). AMPA dose‐relatedly depolarized membrane potential of Vipr2^+/+ neurons during the day (n = 13 cells in slices from 3 animals) and night (n = 13 cells in slices from 3 animals), with larger responses recorded during the night [Dose: F_{1.8, 43.09} = 131.3, p < 0.001; Time of Day F_1,30 = 8.48, p = 0.008; Dose x Time of Day interaction: F_{1.795, 43.09} = 1.596]. For Vipr2^−/− mice, dose‐related depolarizing responses to AMPA were observed during day (n = 18 cells in slices from 3 animals) and night (n = 14 cells in slices from 3 animals) [Dose: F_{1.8, 53.65} = 169.36 (p < 0.001) and Dose x Time of Day interaction: F_1.8,53.65 = 4.23 (p = 0.023)]. To further examine the relationships in the effects of AMPA on SCN neuronal excitability dose–response curves (Hill slope constrained to = 2) were fitted (Figure 4c,d) and Log(EC₅₀) and Emax were calculated. The Log(EC₅₀) differed from day to night in Vipr2^−/− (t = −2.295, df = 30, p = 0.029; Figure 4e; permutation t‐test, p = 0.029; Supplemental Figure S1), but not Vipr2^+/+ mice (Figure 4e; Supplemental Figure S1). Further, at night, the Log(EC₅₀) of Vipr2^+/+ SCN neurons varied from that of Vipr2^−/− SCN neurons (t = 2.25, df = 25, p = 0.033, permutation t‐test, p = 0.029; Supplemental Figure S1). By contrast, Emax did not differ between Vipr2^+/+ and Vipr2^−/− mice nor from day to night in either genotype (Supplemental Figure S2) (see Supplemental Table S1 for more detailed statistical analysis). Collectively, these parametric and curve‐fitting analyses provide evidence that genetic perturbation of VIP‐VPAC₂ receptor signalling enhances nocturnal responsiveness to lower concentrations of a neurochemical mimic of the light input pathway to the SCN.

The above experiments indicate that as a population, neurons in the Vipr2^−/− deficient SCN do not exhibit the typical daily variations neuronal state, membrane potential, spontaneous firing rate as well as responsiveness to the light input pathway that are observed in the neurochemically intact SCN. To gain insight into how intrinsic excitability is altered by the loss of the VPAC₂ receptor, we assessed sodium currents since these contribute to the spontaneous AP firing of rodent SCN neurons (Bean, 2007; Jackson et al., 2004; Pennartz et al., 1997). We initially sought to determine if components of tetrodotoxin (TTX)‐sensitive sodium currents varied between the genotypes. To do this, we made voltage‐clamp recordings and measured how TTX (50 nM and 1 μM) influenced the amplitude of sodium currents in Vipr2^+/+ and Vipr2^−/− SCN neurons at night (n = 8 cells from slices prepared from 3 animals of each genotype; Figure 5a‐g). We used these concentrations as pilot studies of SCN neurons recorded in current‐clamp mode indicated that Vipr2^−/− neurons were more responsive than Vipr2^+/+ neurons to low (50 nM) as well as higher (1 μM) concentrations of TTX (see Figure 7). Neurons were randomly sampled in ventral SCN. Both TTX dose [F_{1.26, 16.385} = 258.186, p < 0.001] and genotype x TTX treatment interaction [F_1.26,16.385 = 6.092, p = 0.019] were factors on the amplitude of sodium currents_, while the effect of genotype was not [F_1,13 = 3.522, p = 0.083]. Under control conditions, currents of similar amplitude were evoked by the step‐wise protocol in both genotype, but treatment with the submaximal dose (50 nM) of TTX revealed a component of sodium current in Vipr2^+/+ SCN cells that was significantly attenuated in Vipr2^−/− SCN neurons (pairwise comparison, p < 0.001) (Figure 5c‐f)). Subsequently, application of 1 μM TTX completely abolished any residual sodium currents in both genotype, an observation that is consistent with previous work in SCN neurons (Kononenko et al., 2004; Pennartz et al., 1997) (Figure 5 e‐h). Since Vipr2^−/− mice develop in the absence of the VIP‐VPAC₂ receptor signalling, we next tested whether blockade of VPAC₂ receptors in the adult Vipr2^+/+ SCN also alters sensitivity to TTX (50 nM). Application of the VPAC₂ antagonist, PG 99–465, did not alter the amplitude of evoked sodium current (control: ~1300 pA vs control + PG 99–465–1064 pA; t‐test, p = 0.17; n = 13 cells), but did increase sensitivity to 50 nM TTX (TTX alone ~ − 433 pA vs PG 99–465 + TTX ~ 200 pA; t‐test, p = 0.004) (Figure 5h). Thus pharmacological blockade of VPAC₂ receptors in the adult Vipr2^+/+ SCN can recapitulate some aspects of the lack of TTX‐resistance of Vipr2^−/− SCN neurons (see Supplemental Table S1 for more detailed statistical analysis). This suggests that effects arising from developing in the absence of VIP‐VPAC₂ signalling cannot solely explain this altered sensitivity to TTX of Vipr2^−/− SCN neurons (but see also (Mazuski et al., 2020)).

The above voltage‐clamp recordings provide evidence that Vipr2^−/− neurons lack important TTX‐resistant currents and we next examined how bath‐applied TTX (50 nM and 1 μM) affected AP firing and membrane events in Vipr2^+/+ and Vipr2^−/− SCN neurons recorded in current clamp mode. With 5 min application of 50 nM TTX, AP firing in Vipr2^+/+ neurons was gradually eliminated to reveal small fluctuations in membrane potential that persisted in 50 nM TTX (6 of 6 cells tested from 3 animals), but were subsequently abolished by treatment with 1 μM TTX (Figure 6a). Such sodium current(s) in Vipr2^+/+ SCN cells may provide subthreshold depolarization drives enabling the membrane to reach threshold to fire APs. A similar time‐course of treatment with 50 nM TTX silenced AP firing of Vipr2^−/− neurons and abolished the small fluctuations of membrane potential seen under these conditions in Vipr2^+/+ neurons (Figure 6b; 5 of 8 cells from 3 animals; significant genotype difference: Fishers Exact test p = 0.031) (see Supplemental Table S1 for more detailed statistical analysis). These findings support the view that Vipr2^−/− SCN neurons lack some TTX‐resistant sodium current(s) observed in Vipr2^+/+ neurons and provide evidence that they are deficient in an intrinsic excitatory mechanism.

We next sought to determine if and how the membrane potential at which AP firing is initiated differs from day to night and between the genotypes. To assess this, we made targeted recordings from neurons randomly distributed within the SCN of both genotypes. Neurons were recorded in current clamp mode and subjected to a current ramp to determine the membrane potential at which the cells either ceased or initiated AP firing (Figure 7a). We also use phase‐plots to reveal any underlying depolarizing subthreshold events that contribute to driving AP firing (Figure 7b,c). Two‐way analysis of variance indicated a significant main effect of genotype (F_1,90 = 27; p < 0.001), and a significant genotype x time of day interaction (F_1,90 = 7.74; p = 0.007). The AP firing threshold of Vipr2^+/+ SCN neurons during the day (−47 ± 1.2 mV,

n = 17 cells) differed to that recorded at night (−51.6 ± 0.9 mV, n = 25; pairwise comparison, p = 0.003; permutation t‐test, p = 0.007; Figure 7d,f), whereas AP firing threshold measured in Vipr2^−/− cells did not vary from day (−44.6 ± 0.9 mV, n = 29) to the night (−43.7 ± 1 mV, n = 23; pairwise comparison, p > 0.05; permutation t‐test, p > 0.05; Figure 7d). At night, the AP firing threshold of Vipr2^−/− neurons occurred at a significantly more depolarized level compared to that measured in Vipr2^+/+ neurons at this time (pairwise comparison, p < 0.001; permutation t‐test, p < 0.001; Figure 7h) (see Supplemental Table S1 for more detailed statistical analysis). These analyses show that with the loss of VIP‐VPAC₂ receptor signalling, SCN neurons at night require a more depolarized membrane potential to fire APs.

Collectively, these studies provide evidence that genetically targeted impairment of VIP‐VPAC₂ receptor signalling compromises the daily control of the membrane potential and the firing of action potentials.

Next, we investigated if pharmacological blockade of VPAC₂ receptors in adult Vipr2^+/+ mice would also alter the AP threshold of SCN neurons. Previously, we showed that short‐term treatment with the VPAC₂ receptor antagonist, PG99–465, for 5–10 min prior to VIP application, acutely antagonizes VIP's presynaptic effects in the SCN (Pakhotin et al., 2006), while the longer‐term application (>3 h) results in alteration of the cell's intrinsic electrophysiological behaviour (Cutler et al., 2003; Itri & Colwell, 2003). Therefore, for 4 h prior to recording, SCN‐containing brain slices from Vipr2^+/+ mice were incubated in aCSF containing PG99–465. We found that with this longer‐term VPAC₂ receptor blockade with PG99–465, there was no day‐night difference in the AP firing threshold (−47 ± 1.12 mV, n = 12 cells from 3 animals vs − 47.6 ± 1.04 mV, n = 17 cells from 3 animals; t‐test, p > 0.05; Figure 7j) (see Supplemental Table S1 for more detailed statistical analysis). Thus, pharmacological blockade of VIP‐VPAC₂ signalling in Vipr2^+/+ cells mimicked the AP firing threshold recorded at night in Vipr2^−/− SCN neurons.

The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer-review/10.1111/ejn.16590↗.