Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Sep 29.
Published in final edited form as: Neuron. 2019 Dec 9;105(4):621–629.e4. doi: 10.1016/j.neuron.2019.11.011

Daily Oscillation of the Excitation-Inhibition Balance in Visual Cortical Circuits

Michelle Bridi 1,†, Fang-Jiao Zong 2,3,†, Xia Min 2,3, Jiaqian Qiu 2,3, Nancy Luo 1, Xue-Ting Zhang 2,3, Trinh Tran 1, Daniel Severin 1, Guanglin Wang 2, Zheng-Jiang Zhu 2,3, Kai-Wen He 2,3,*, Alfredo Kirkwood 1,*
PMCID: PMC9520672  NIHMSID: NIHMS1824787  PMID: 31831331

SUMMARY

A dynamic balance between synaptic excitation and inhibition (E/I balance) maintained within a narrow window is widely regarded to be crucial for cortical processing. In line with this idea, measures of the E/I balance are reportedly comparable across neighboring neurons, across behavioral states and developmental stages, and altered in mouse models of autism and schizophrenia. Motivated by these ideas, we examined whether synaptic inhibition changes over the 24-hour day to compensate for the well-documented sleep-dependent changes in synaptic excitation. We found that in pyramidal cells of visual and prefrontal cortices as well as in hippocampal CA1, the strength of synaptic inhibition also changes over the 24h light/dark cycle, but surprisingly, in the opposite direction of synaptic excitation. The upregulation of inhibition in visual cortex during the light phase is sleep-dependent. Notably, in visual cortex, these changes in the E/I balance were observed in feedback, but not feedforward, circuits. The observations that the E/I balance is not constant but oscillates over the course of the day, and only in specific circuits, opens new and interesting questions on the function and regulation of the E/I balance.

INTRODUCTION

Cortical processing critically depends on the balanced interplay of glutamatergic excitatory synapses to propagate neural firing, and GABAergic inhibitory synapses to limit that propagation in time and space. The prevailing view states that the balance between synaptic excitation and inhibition (E/I balance) is dynamically maintained within a permissible window to ensure proper neural function (see (Carandini and Heeger, 2012; Denève et al., 2017; Froemke, 2015; Isaacson and Scanziani, 2011; Keck et al., 2017; Rubin et al., 2017; Vogels et al., 2011) for reviews and theoretical considerations). Therefore, a central question is how the E/I balance is regulated in neural circuits.

The importance of maintaining proper E/I balance in neural processing is underscored by studies showing that direct manipulations of the E/I balance can disrupt social behavior and sensory perception (Ferguson and Gao, 2018; Shen et al., 2011; Yizhar et al., 2011). Moreover, a wealth of studies report E/I balance alterations in mouse models of various neurological disorders, including autism and schizophrenia (Antoine et al., 2019; Gkogkas et al., 2013; Han et al., 2012, 2014; Tabuchi et al., 2007), Angelman and Rett syndromes (Calfa et al., 2015; Judson et al., 2016; Rotaru et al., 2018; Wallace et al., 2012), Alzheimer’s disease (Busche and Konnerth, 2016; Busche et al., 2015) and tuberous sclerosis (Bateup et al., 2013). These changes in the E/I balance, largely due to changes in inhibition, are considered a primary contributing factor to the cognitive impairments associated with these pathologies (for review see (Anticevic and Lisman, 2017; Nelson and Valakh, 2015; Rosenberg et al., 2015).

The notion that the E/I balance is dynamically maintained within an optimal range is also supported by studies showing that the E/I ratio remains stable during neural development (Tao and Poo, 2005), between neighboring neurons (Xue et al., 2014), and transitioning within and between brain states (Zhou et al., 2014). In addition, research has identified multiple synaptic homeostatic mechanisms capable of maintaining the E/I balance within a target range, particularly during learning and cortical remodeling (D’amour and Froemke, 2015; Froemke, 2015; Nanou and Catterall, 2018).

The homeostatic mechanisms mentioned above primarily affect the strength of inhibition in response to changes in excitation. In that context, it was of great interest to us that prior studies have shown marked sleep-dependent changes in the strength and number of cortical excitatory glutamatergic synapses at different times of the day (de Vivo et al., 2017; Diering et al., 2017; Gilestro et al., 2009; Liu et al., 2010; Maret et al., 2011; Vyazovskiy et al., 2008). We asked, therefore, whether changes in excitation throughout the day are compensated by complementary changes in inhibition to maintain the E/I balance. Surprisingly, we found that in all cortical areas examined, inhibition and excitation change in opposite directions over the course of the 24h day: when the frequency of excitatory synaptic events is high, the frequency of inhibitory events is low, and vice versa. These changes indicate a large oscillation of the E/I balance over the course of the day, opening new and interesting questions as to how the E/I balance is regulated.

RESULTS

Modulation of synaptic excitation and inhibition in opposite directions across the light/dark cycle

In several cortical areas, the number of spines and frequency of miniature excitatory synaptic events changes across the day in a sleep-dependent manner (Liu et al., 2010; Maret et al., 2011). These observations prompted us to examine whether the E/I balance is maintained across the day by compensatory changes in synaptic inhibition. To that end, we recorded miniature excitatory and inhibitory postsynaptic currents (mEPSCs and mIPSCs) in layer 2/3 pyramidal cells of primary visual cortex (V1) in slices harvested at two times of the day, the end of the dark (zeitgeber time (ZT) 0) and light (ZT12) phases (Fig 1A. See methods for details). Consistent with previous findings in the frontal cortex (Liu et al., 2010), in the visual cortex the mEPSC frequency (Fig 1B), but not amplitude (Fig 1D), was significantly higher in slices harvested at ZT0 than at ZT12. Inhibitory synaptic transmission also changed during the day, but surprisingly in the opposite direction: the mIPSC frequency was higher at ZT12 than at ZT0 (Fig 1C). As observed for mEPSCs, the average mIPSC amplitude did not change (Fig 1 E), although in both cases we detected a modest shift in the cumulative distributions of individual events at ZT12: leftward in the case of mEPSCs (Fig 1D, p < 0.0001, K-S test), rightward in the case of mIPSCs (Fig 1E, p = 0.023, K-S test). In addition to synaptic changes, alterations in neuronal excitability shape circuit function. We therefore measured maximal firing rate and action potential threshold in pyramidal cells and the most common subtype of inhibitory cells, parvalbumin-positive interneurons. However, we detected no changes in either of these measures at different times of day (Supplementary Fig 1).

Figure 1.

Figure 1.

Open in a new tab

Regulation of mEPSCs and mIPSCs in opposite directions over the light/dark cycle in V1 L2/3 pyramidal cells. (A) Mice were entrained to a normal or reversed 12:12 L:D cycle for at least two weeks. Acute slices containing V1 were obtained at the end of the dark (ZT0) or light (ZT12) phase (arrowheads). (B) mEPSC frequency was higher at ZT0 than ZT12. Top: Example raw traces. Bottom left: mEPSC frequency was higher at ZT0 than ZT12 (ZT0: 6.94±0.66 Hz; ZT12: 4.15±0.45 Hz; t test). Bottom right: There was a significant leftward shift in the cumulative probability histogram of the interevent intervals in the ZT0 group (KS P < 0.0001). Sample size is indicated as (cells, mice). (C) mIPSC frequency was lower at ZT0 than ZT12. Top: example traces. Bottom: Average mIPSC frequency was significantly lower at ZT0 than ZT12, as indicated by lower average frequency (left: ZT0: 6.28±0.99 Hz; ZT12: 11.02±1.36 Hz; t test), and a rightward shift in the cumulative probability histogram (right: KS P < 0.0001). Sample size is indicated as (cells, mice). (D) mEPSC amplitude and kinetics did not differ between ZT0 and ZT12. Top: average traces of all well-isolated events. Bottom: average mEPSC amplitude was comparable (left: ZT0: 12.82±0.44; ZT12; 12.20±0.21, t test P= 0.78). The distribution of individual event amplitudes was slightly but significantly different between groups (KS P = 0.009). (E) mIPSC amplitude and kinetics did not differ between ZT0 and ZT12. Top: average traces of well-isolated events. Bottom: average mIPSC amplitude did not differ between groups (ZT0: 50.07±4.97; ZT12; 52.83±4.44; t test P=0.68). The cumulative distribution was slightly shifted leftward at ZT0 (KS P = 0.023). (D, E) Sample size as indicated in (B, C).

These opposing changes in excitatory and inhibitory synaptic transmission were not restricted to V1. Similar changes were observed in the medial prefrontal cortex (mPFC) layer 2/3 and the hippocampal CA1 (HC) (Fig 2), indicating that the modulation of excitation and inhibition in opposite directions over the course of the light/dark cycle may be a global phenomenon. Altogether, these results indicate that the ratio of the mEPSC/mIPSC frequency, often used as an indicator of the E/I balance (Han et al., 2012; Tabuchi et al., 2007), is not constant as commonly assumed, but spontaneously changes between ZT0 and ZT12.

Figure 2.

Figure 2.

Open in a new tab

Pyramidal cells from L2/3 medial prefrontal cortex (mPFC) and hippocampal area CA1 showed the same mEPSC and mIPSC changes as in V1. (A) mEPSC frequency, but not amplitude, was lower at ZT12 than ZT0. mPFC freq: ZT0 9.61±1.17 Hz; ZT12 6.43±0.27. Hippocampus freq: ZT0 1.13±0.17 Hz; ZT12 0.61±0.09. mPFC amp: ZT0 12.61±0.42 pA; ZT12 13.60±0.50; P=0.15. Hippocampus amp: ZT0 19.60±1.04 pA; ZT12 20.22±0.95; P=0.66. (B) mIPSC frequency, but not amplitude, was higher at ZT12 than ZT0 in mPFC and hippocampus. mPFC freq: ZT0 10.51±0.95 Hz; ZT12 18.84±1.51. Hippocampus freq: ZT0 9.86±0.63 Hz; ZT12 12.34±0.96. mPFC amp: ZT0 32.65±2.26 pA; ZT12 32.31±2.62; P=0.92. Hippocampus amp: ZT0 55.21±3.02 pA; ZT12 51.49±2.99; P=0.39. Sample size is indicated as (cells, mice).

Spontaneous ISPCs are modulated across the light/dark cycle

We next quantified spontaneous IPSCs (sIPSCs: in the absence of TTX and synaptic blockers, see methods) in V1, which better approximate natural conditions (Dani et al., 2005; Jurgensen and Castillo, 2015). We quantified inhibitory strength as the total charge (nC) in one second. To get a more detailed picture of how inhibitory synaptic transmission is modulated, we measured the inhibitory strength at six zeitgeber times (Fig 3A). Consistent with the mIPSCs (Fig 1C), sIPSC charge at ZT12 was over twice of that at ZT0. Notably, the sIPSC charge was comparable across the light phase (ZT4, 8, and 12), suggesting a rapid upregulation of inhibition that then stabilizes during the light phase. Downregulation during the dark phase follows a similar temporal pattern. To further investigate how rapidly upregulation occurs in the light, we performed additional recordings at ZT1 (Supplementary Fig 2A). At ZT1, sIPSC charge showed a nonsignificant increase compared to ZT0 and was significantly lower than at ZT4, suggesting that the sIPSCs gradually increase between ZT0 and ZT4.

Figure 3.

Figure 3.

Open in a new tab

sIPSC charge oscillations across the light/dark cycle are modulated by sleep and CBR signaling. (A) Top left: Acute brain slices were obtained at six different times of day (arrowheads). Bottom left: Representative traces showing spontaneous IPSCs recorded at +10 mV with no drugs in the bath. Scale bar: 50pA, 200 msec. Right: sIPSC charge was higher when the animal had been in the light (rest) phase prior to sacrifice. ZT0: 25.12±3.2 nC; ZT4: 52.38±6.0; ZT8: 49.47±4.2; ZT12: 59.65±5.3; ZT16: 40.86±2.5; ZT20: 40.09±2.4. One-way ANOVA F(5, 133)=7.554, P<0.0001, Holm-Sidak post-hoc analysis. (B) Top: Mice underwent sleep deprivation (SD) or were allowed to sleep ad libitum for the first four hours of the light cycle. Bottom left: EEG and EMG recordings confirmed the efficacy of the sleep deprivation. Sleep group: n=4; SD group: n=4 mice. Bottom right: SD prevented the increase in sIPSCs that normally occurs between ZT0 and ZT4. Sleep: 35.20±3.41 nC; SD: 22.43±1.94. (C) Slices were obtained at either ZT0 or ZT12 and pre-incubated with 10 μM SR 141716A (SR) or WIN 55,212–2 (WIN) in 0.1% DMSO for at least 1h, then sIPSCs were recorded in the presence of drug. Slices from the same animals were used as controls (in 0.1% DMSO). SR increased sIPSC charge at ZT0 but not ZT12; conversely, WIN decreased sIPSCs at ZT12 but not ZT0. SR experiment: Control ZT0: 35.16±3.5 nC: SR ZT0: 51.91±3.7; Control ZT12: 62.37±7.0; SR ZT12: 58.75±4.0; Kruskal-Wallis P=0.0007. WIN experiment: Control ZT0: 32.29±3.5 nC; WIN ZT0: 27.89±1.8; Control ZT12: 56.67±6.2; WIN ZT12: 26.14±1.8; Kruskal-Wallis P<0.0001. Dunn’s post-hoc analysis was used for multiple comparisons. For all panels, sample size is indicated as (cells, mice).

Modulation of inhibition across the light/dark cycle depends on sleep

Sleep controls the daily changes in mEPSC frequency (Liu et al., 2010). Therefore, we asked whether this is also the case for the changes in inhibitory transmission. We tested whether sleep plays a role in the upregulation of the sIPSC charge between ZT0 and ZT4, when mice spend more time asleep. To that end, mice were instrumented for polysomnography and divided into two groups. One group was sleep deprived (SD) by gentle handling for four hours (ZT0-ZT4), and the other was allowed to sleep ab libitum; SD efficacy was confirmed using the EEG and EMG recordings (Fig 3B). At ZT4, mice were sacrificed and sIPSCs were measured in V1 L2/3 pyramidal neurons. SD mice had significantly lower sIPSC charge compared to controls (Fig 3B). The magnitude of sIPSC charge in SD mice was comparable to ZT0 (Fig 3A), suggesting that sleep upregulates inhibition during the light phase.

Endocannabinoid signaling may modulate inhibition across the light/dark cycle

Endocannabinoids suppress inhibitory transmission in the cortex (Fortin et al., 2004; Trettel and Levine, 2003; Trettel et al., 2004) and exhibit time-of-day variations in the rat brain (Murillo-Rodriguez et al., 2006; Valenti et al., 2004) and human plasma (Hanlon et al., 2016). Therefore, we examined whether signaling via cannabinoid receptors (CBRs) could be a mechanism to suppress sIPSCs during the dark phase. To test this, we collected slices at either ZT0 or ZT12, then pre-incubated and recorded the slices in ACSF containing either the CB1R antagonist SR 141716A (SR), or the CBR agonist WIN 55,212–2 (WIN). We found that inhibiting CB1R with SR at ZT0 increased inhibitory transmission to levels similar to ZT12, while SR had no effect at ZT12 when sIPSC levels were already high (Fig 3C, left). The agonist WIN had the converse effect: WIN decreased inhibitory transmission at ZT12 but had no effect at ZT0 (Fig 3C, right). Since CBR activation can suppress inhibitory transmission during the dark phase, we then asked whether endocannabinoid levels are, in fact, higher at this time of day. Mass spectrometry measurements of visual cortical, hippocampal, and frontal cortical tissue revealed higher levels of the most abundant endocannabinoid, 2-arachidonoylglycerol (2-AG) (Hanlon et al., 2016), in the dark phase (Supplementary Fig 2B). Together, these results suggest that endocannabinoid signaling may exist as a mechanism to actively suppress inhibitory transmission during the dark phase.

Input-specific modulation of the E/I ratio across the light/dark cycle

Inhibitory interneurons in L2/3 participate in feedforward circuits driven by vertical ascending excitatory inputs, primarily from L4, and in feedback circuits driven by lateral inputs, primarily from L2/3. We therefore asked whether the E/I ratio changes between ZT0 and ZT12 in each of these pathways. The vertical inputs were stimulated optogenetically using a mouse expressing channelrhodopsin 2 (ChR2) specifically in L4 (Fig 4A, Supplementary Fig 3; see Methods); lateral stimulation was performed electrically (Fig 4A). In both cases, we recorded evoked EPSCs and IPSCs in the same pyramidal cell by holding the membrane at the reversal potential for GABA and AMPA receptors, respectively (Supplementary Fig 3). Since the E/I ratio varies with stimulation intensity (Morales et al., 2002), for each cell we stimulated at a range of intensities to determine the range over which the E/I ratio is stable. We used only values in this stable range for the analysis (Supplementary Fig 3E).

Figure 4.

Figure 4.

Open in a new tab

Circuit-specific modulation of the E/I ratio. (A) Synaptic currents were evoked either laterally by electrical stimulation in layer 2/3 or vertically by light-evoked release from channelrhodopsin-2 expressing layer 4 cells (see Methods). Layer 2/3 pyramidal cell responses were recorded in whole-cell voltage clamp configuration. (B) The laterally evoked E/I ratio was higher when the animal had been in the dark phase prior to sacrifice (ZT0: 0.37±0.02; ZT8: 0.26±0.02; ZT12: 0.25±0.02; ZT20: 0.35±0.02; one-way ANOVA F(3, 113)=11.11, P<0.0001, post-hoc Holm-Sidak test). Example traces are normalized to peak IPSC response magnitude; Scale bar: 50 msec. (C) The vertical E/I ratio did not change over the light/dark cycle (ZT0: 0.20±0.02; ZT12: 0.21±0.02, 2-tailed t test t(46)=0.3229, P=0.748). Example traces are normalized to peak IPSC response magnitude; Scale bar: 50 msec. (D) For a subset of cells in (B, C), we recorded responses to both the vertical and lateral stimulation in the same cell. At ZT0, the E/I ratio in the lateral pathway was significantly higher than in the vertical pathway (2-tailed paired t test t(8)=5.494). However, at ZT12, the E/I ratio did not differ between pathways (2-tailed paired t test t(9)=0.7393, P=0.4786). (E) Visual experience does not affect modulation of the E/I ratio. Mice were kept in complete darkness 24h prior to experimentation. As for animals on a normal light/dark cycle, the lateral E/I ratio was higher at ZT0 than at ZT12 (ZT0: 0.37±0.02; ZT12: 0.29±0.01; 2-tailed Mann-Whitney test U(45)=83), and the vertical E/I ratio remained unaffected (ZT0: 0.24±0.01; ZT12: 0.22±0.01; 2-tailed t test t(44)=1.253, P=0.2168). (F) Modulation of the lateral E/I ratio affects spike output. Cells were patched in cell-attached mode and the lateral pathway was stimulated to identify spike threshold. Then, the seal was broken and the AMPA receptor response at spike threshold was recorded in whole-cell configuration (scale bar: 5 msec, 200 pA). At ZT12, when the E/I ratio is lower, more AMPA receptor current was required to reach threshold (ZT0: 570.1±44.8 pA; ZT12: 817.3±56.1; 2-tailed t test t(39)=3.425). For all panels, sample size is indicated as (cells, mice).

With lateral stimulation, the E/I ratio was higher when the animal had been in the dark phase (ZT0, ZT20) than when the animal had been in the light phase (ZT8, ZT12) (Fig 4B). This finding is consistent with the mEPSC, mIPSC, and sIPSC results (Fig 1, 3). We then wished to confirm that these differences reflect the time of sacrifice, and were not affected by the E/I ratio changing in slices ex vivo over time. For each group, we examined the correlation between the time a cell was patched and the measured E/I ratio. The E/I ratio did not correlate with time ex vivo in any group (Supplementary Fig 3H).

In contrast with lateral stimulation, the E/I ratio measured with vertical stimulation was not different between ZT0 and ZT12 (Fig 4C). This differential and pathway-specific regulation of the E/I ratio was also observed when we examined a subset of experiments in which we measured both pathways in the same cell. In these cases, at ZT0 the E/I ratio was larger with lateral than vertical stimulation. At ZT12, however, the E/I ratio was comparable in both pathways (Fig 4D).

Visual deprivation in the form of dark exposure specifically modulates lateral, but not vertical, inputs (Petrus et al., 2015), similar to the changes in the E/I ratio reported here (Fig 4B–D). Therefore, we considered the possible role of visual experience over the light/dark cycle in the regulation of E/I balance. In these experiments we exploited the fact that mice entrained to a 12:12 L:D cycle maintain their activity patterns for 5–7 days when placed in the dark (Faradji-Prevautel et al., 1990). We therefore entrained mice to a normal or reversed 12:12 L:D cycle. Then, 24 hours prior to experimentation, we placed the mice in constant and complete darkness to remove all visual experience. Similar to mice with normal experience, in visually deprived mice the E/I balance was larger at ZT0 than at ZT12 with lateral stimulation, and unchanged with vertical stimulation (Fig 4E). These results indicate that in L2/3 pyramidal neurons of V1, the changes in E/I balance across the light/dark cycle are not regulated by recent visual experience.

Finally, we explored whether these differences in the lateral synaptic E/I ratio affect action potential firing, with the expectation that circuit excitability will be reduced when the E/I ratio is low. Therefore, we determined the amount of AMPA receptor current needed for each cell to reach spike threshold at ZT0 and ZT12. We held cells in cell-attached mode while electrically stimulating the lateral pathway to identify the intensity that elicited spikes in approximately half of the trials (spike threshold) (Fig 4F, bottom). Then, the seal was broken and the AMPA receptor current at spike threshold was recorded in whole-cell configuration. More AMPA receptor current was required to reach spike threshold at ZT12, when the E/I ratio is lower than at ZT0 (Fig 4F). These findings are consistent with our finding that neurons do not compensate for changes in the synaptic E/I ratio with changes in excitability (Supplementary Fig 1) and demonstrate that E/I fluctuations in the lateral circuit over the course of the day have a meaningful impact on spiking output.

DISCUSSION

Maintaining the E/I balance within a permissive window is considered crucial to ensure proper neural processing. Here we report that in cortex the E/I balance is not constant, but it changes markedly over the light/dark cycle. Over the course of the day, synaptic measures of inhibitory and excitatory strength change in opposite directions, and in a sleep-dependent manner. Moreover, these changes are not uniform across cortical circuits. The E/I ratio of lateral inputs to layer 2/3 pyramidal cells change, impacting neuronal spiking, but the ascending inputs do not. These observations add complexity to our understanding of the E/I balance and pose intriguing questions.

The observation that the E/I ratio changes over the course of the day might seem in conflict with previous studies showing that in cortical circuits the E/I ratio remains constant during different behavioral states (Tao and Poo, 2005; Xue et al., 2014; Zhou et al., 2014), and with the idea that experience-dependent cortical remodeling mechanisms exist to return the E/I ratio to a target set-point (Froemke, 2015). However, these ideas are not necessarily contradictory. These observations can be reconciled if the set point for the E/I ratio slowly changes across the 24h day. In this scenario, the E/I ratio within a short time window will be dynamically maintained at that target set point by fast-acting correction mechanisms. Thus, at a particular time of day, the E/I value in a given circuit will be comparable across individuals and across neural states. This explanation calls for plastic mechanisms operating at two distinct time scales: A fast mechanism operating within seconds to minutes to correct deviations from the target set point, and a slower one operating over hours to modify the set point.

The mechanisms underlying the daily oscillation of the synaptic E/I balance remain to be fully understood and open questions on multiple levels. At the level of global arousal state, sleep promotes a decrease in excitation (de Vivo et al., 2017; Liu et al., 2010; Maret et al., 2011) and an increase in inhibition (Fig 3), but what drives the complementary process during the dark phase is less clear. At an elementary synaptic level, an intriguing aspect of the changes in E/I ratio is that they manifest as changes in the frequency, not in the amplitude, of miniature excitatory and inhibitory events (Fig 1, 2). This suggests a change in the number of synapses, not their potency, in agreement with prior report of sleep-dependent changes in the turnover of excitatory spines in cortex (Maret et al., 2011). This interpretation is also consistent with our findings that neither the paired-pulse ratio (a crude estimate of the release probability) nor the AMPA/NMDA ratio (a crude estimate of silent synapses) varies over the course of the light/dark cycle (Supplementary Fig 3). Also consistent with this interpretation, molecular markers of GABAergic and glutamatergic synapses are higher during the light and dark phases, respectively, in lateral hypothalamus and cerebellum (Cirelli et al., 2004; Laperchia et al., 2017) (but see (del Cid-Pellitero et al., 2017)). Besides structural changes in synapse number, other mechanisms might also shape the E/I ratio. For example, neuromodulatory tone, which varies greatly across arousal states, may control the spontaneous fusion of vesicles. Indeed, norepinephrine, which is high during waking, can directly increase the frequency of mEPSCs in the cortex (Choy et al., 2018), and mIPSC frequency can be affected by neuromodulation as well (Cilz and Lei, 2017; Gao et al., 2017; Madison and Nicoll, 1988). Here, we find that one potential contributing mechanism for suppression of inhibition in the dark phase is endocannabinoid signaling. Pharmacological activation of the CB receptor suppresses sISPC levels at ZT12 when they are normally high, while inhibiting the CB1 receptor elevates sIPSC levels at ZT0 when they are normally low. Consistent with these observations, the endocannabinoid 2-AG is elevated during the dark phase, when it could actively suppress inhibitory transmission.

Intriguingly, evoked synaptic responses showed that changes in the E/I ratio occur specifically in the lateral, not feedforward pathway (Fig 4). This does not contradict our observations of spontaneous events because the most abundant source of synaptic inputs to layer 2/3 V1 cells is lateral connections, not vertical inputs from layer 4 (Petrus et al., 2015). Interestingly, lateral, but not vertical, inputs are susceptible to long-term modification by prolonged altered sensory experience (Petrus et al., 2015). Differences in the E/I regulation between the two pathways may also reflect complex changes in the disynaptic circuitry, e.g. changes in excitatory inputs onto inhibitory interneurons, resulting in differential recruitment of inhibition. This scenario is in line with the observation that parvalbumin-positive interneurons are differentially recruited across arousal states (Niethard et al., 2016).

Changes in excitation and inhibition over the 24h day are seemingly a common feature across brain areas, as described above, and have even been suggested in humans (Chellappa et al., 2016). Although the functional consequences of sleep-dependent increases in inhibitory synaptic transmission reported here remain to be determined, they dovetail with two major theories of sleep function. On one hand, enhanced inhibition may facilitate memory consolidation during sleep by improving spike timing precision. Precise timing of spikes in relation to hippocampal sharp-wave ripples is important for long-term potentiation during replay (Sadowski et al., 2016). On the other hand, increased inhibition promotes long-term depression (Steele et al., 2011), hence sleep may promote homeostatic weakening of excitatory synaptic transmission (Tononi and Cirelli 2014). Finally, it is worth noting that alterations in the E/I ratio are often thought to contribute to neural dysfunction. Notably, the difference in the E/I ratio between the dark and light phases reported here are comparable to, if not larger than, the E/I alterations documented in models of autism (Antoine et al., 2019; Gkogkas et al., 2013; Han et al., 2012; Tabuchi et al., 2007). This opens the question of whether daily E/I cycling is dysregulated in these mouse models.

EXPERIMENTAL PROCEDURES

The studies were conducted in brain slices prepared from young adult mice of both sexes aged 5 to 8 weeks-old, and synaptic responses were recorded and quantified with whole-cell recordings as previously described (He et al., 2015). For more details see supplemental experimental methods.

Supplementary Material

supplenetrary material

REFERENCES

  1. Anticevic A, and Lisman J (2017). How Can Global Alteration of Excitation/Inhibition Balance Lead to the Local Dysfunctions That Underlie Schizophrenia? Biol. Psychiatry 81, 818–820. [DOI] [PubMed] [Google Scholar]
  2. Antoine MW, Langberg T, Schnepel P, and Feldman DE (2019). Increased Excitation-Inhibition Ratio Stabilizes Synapse and Circuit Excitability in Four Autism Mouse Models. Neuron 101(4): 648–661. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bateup HS, Johnson CA, Denefrio CL, Saulnier JL, Kornacker K, and Sabatini BL (2013). Excitatory/inhibitory synaptic imbalance leads to hippocampal hyperexcitability in mouse models of tuberous sclerosis. Neuron 78, 510–522. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Busche MA, and Konnerth A (2016). Impairments of neural circuit function in Alzheimer’s disease. Philos. Trans. R. Soc. B Biol. Sci 371, 20150429. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Busche MA, Kekuš M, Adelsberger H, Noda T, Förstl H, Nelken I, and Konnerth A (2015). Rescue of long-range circuit dysfunction in Alzheimer’s disease models. Nat. Neurosci 18, 1623–1630. [DOI] [PubMed] [Google Scholar]
  6. Calfa G, Li W, Rutherford JM, and Pozzo-Miller L (2015). Excitation/inhibition imbalance and impaired synaptic inhibition in hippocampal area CA3 of Mecp2 knockout mice. Hippocampus 25, 159–168. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Carandini M, and Heeger DJ (2012). Normalization as a canonical neural computation. Nat. Rev. Neurosci 13, 51–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Chellappa SL, Gaggioni G, Ly JQM, Papachilleos S, Borsu C, Brzozowski A, Rosanova M, Sarasso S, Luxen A, Middleton B, et al. (2016). Circadian dynamics in measures of cortical excitation and inhibition balance. Sci. Rep 6, 1–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Choy JMC, Agahari FA, Li L, and Stricker C (2018). Noradrenaline Increases mEPSC Frequency in Pyramidal Cells in Layer II of Rat Barrel Cortex via Calcium Release From Presynaptic Stores. Front. Cell. Neurosci 12, 213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. del Cid-Pellitero E, Plavski A, Mainville L, and Jones BE (2017). Homeostatic Changes in GABA and Glutamate Receptors on Excitatory Cortical Neurons during Sleep Deprivation and Recovery. Front. Syst. Neurosci 11, 1–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cilz NI, and Lei S (2017). Histamine facilitates GABAergic transmission in the rat entorhinal cortex: Roles of H1 and H2 receptors, Na+ -permeable cation channels, and inward rectifier K+ channels. Hippocampus 27, 613–631. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cirelli C, Gutierrez CM, and Tononi G (2004). Extensive and Divergent Effects of Sleep and Wakefulness on Brain Gene Expression. Neuron 41, 35–43. [DOI] [PubMed] [Google Scholar]
  13. D’amour JA, and Froemke RC (2015). Inhibitory and excitatory spike-timing-dependent plasticity in the auditory cortex. Neuron 86, 514–528. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dani VS, Chang Q, Maffei A, Turrigiano GG, Jaenisch R, and Nelson SB (2005). Reduced cortical activity due to a shift in the balance between excitation and inhibition in a mouse model of Rett Syndrome. Proc. Natl. Acad. Sci 102, 12560–12565. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Denève S, Alemi A, and Bourdoukan R (2017). The Brain as an Efficient and Robust Adaptive Learner. Neuron 94, 969–977. [DOI] [PubMed] [Google Scholar]
  16. Diering GH, Nirujogi RS, Roth RH, Worley PF, Pandey A, and Huganir RL (2017). Homer1a drives homeostatic scaling-down of excitatory synapses during sleep. Science (80-.). 355, 511–515. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Faradji-Prevautel H, Cespuglio R, and Jouvet M (1990). Circadian rest-activity rhythms in the anophthalmic, monocular and binocular ZRDCT/An mice. Retinal and serotoninergic (raphe) influences. Brain Res 526, 207–216. [DOI] [PubMed] [Google Scholar]
  18. Ferguson BR, and Gao W-J (2018). PV Interneurons: Critical Regulators of E/I Balance for Prefrontal Cortex-Dependent Behavior and Psychiatric Disorders. Front. Neural Circuits 12, 37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Fortin DA, Trettel J, and Levine ES (2004). Brief trains of action potentials enhance pyramidal neuron excitability via endocannabinoid-mediated suppression of inhibition. J. Neurophysiol 92, 2105–2112. [DOI] [PubMed] [Google Scholar]
  20. Froemke RC (2015). Plasticity of Cortical Excitatory-Inhibitory Balance. Annu. Rev. Neurosci 38, 195–219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Gao M, Whitt JL, Huang S, Lee A, Mihalas S, Kirkwood A, and Lee HK (2017). Experience-dependent homeostasis of ‘noise’ at inhibitory synapses preserves information coding in adult visual cortex. Philos. Trans. R. Soc. B Biol. Sci 372, 11–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Gilestro GF, Tononi G, and Cirelli C (2009). Widespread changes in synaptic markers as a function of sleep and wakefulness in drosophila. Science (80-.). 324, 109–112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Gkogkas CG, Khoutorsky A, Ran I, Rampakakis E, Nevarko T, Weatherill DB, Vasuta C, Yee S, Truitt M, Dallaire P, et al. (2013). Autism-related deficits via dysregulated eIF4E-dependent translational control. Nature 493, 371–377. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Han S, Tai C, Westenbroek RE, Yu FH, Cheah CS, Potter GB, Rubenstein JL, Scheuer T, de la Iglesia HO, and Catterall WA (2012). Autistic behavior in Scn1a+/− mice and rescue by enhanced GABAergic transmission. Nature 489, 385–390. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Han S, Tai C, Jones CJ, Scheuer T, and Catterall WA (2014). Enhancement of Inhibitory Neurotransmission by GABA A Receptors Having α 2,3 -Subunits Ameliorates Behavioral Deficits in a Mouse Model of Autism. Neuron 81, 1282–1289. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Hanlon EC, Tasali E, Leproult R, Stuhr KL, Doncheck E, de Wit H, Hillard CJ, and Van Cauter E (2016). Sleep Restriction Enhances the Daily Rhythm of Circulating Levels of Endocannabinoid 2-Arachidonoylglycerol. Sleep 39, 653–664. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Isaacson JS, and Scanziani M (2011). How Inhibition Shapes Cortical Activity. Neuron 72, 231–243. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Judson MCC, Wallace MLL, Sidorov MSS, Burette ACC, Gu B, van Woerden GMM, King IFF, Han JEE, Zylka MJJ, Elgersma Y, et al. (2016). GABAergic Neuron-Specific Loss of Ube3a Causes Angelman Syndrome-Like EEG Abnormalities and Enhances Seizure Susceptibility. Neuron 90, 56–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Jurgensen S, and Castillo PE (2015). Selective Dysregulation of Hippocampal Inhibition in the Mouse Lacking Autism Candidate Gene CNTNAP2. J. Neurosci 35, 14681–14687. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Keck T, Hübener M, and Bonhoeffer T (2017). Interactions between synaptic homeostatic mechanisms: an attempt to reconcile BCM theory, synaptic scaling, and changing excitation/inhibition balance. Curr. Opin. Neurobiol 43, 87–93. [DOI] [PubMed] [Google Scholar]
  31. Laperchia C, Imperatore R, Azeez IA, Del Gallo F, Bertini G, Grassi-Zucconi G, Cristino L, and Bentivoglio M (2017). The excitatory/inhibitory input to orexin/hypocretin neuron soma undergoes day/night reorganization. Brain Struct. Funct 222, 3847–3859. [DOI] [PubMed] [Google Scholar]
  32. Liu Z-W, Faraguna U, Cirelli C, Tononi G, and Gao X-B (2010). Direct Evidence for Wake-Related Increases and Sleep-Related Decreases in Synaptic Strength in Rodent Cortex. J. Neurosci 30, 8671–8675. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Madison DV, and Nicoll RA (1988). Norepinephrine decreases synaptic inhibition in the rat hippocampus. Brain Res 442, 131–138. [DOI] [PubMed] [Google Scholar]
  34. Maret S, Faraguna U, Nelson AB, Cirelli C, and Tononi G (2011). Sleep and waking modulate spine turnover in the adolescent mouse cortex. Nat. Neurosci 14, 1418–1420. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Morales B, Choi S-Y, and Kirkwood A (2002). Dark Rearing Alters the Development of GABAergic Transmission in Visual Cortex. J. Neurosci 22, 8084–8090. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Murillo-Rodriguez E, Désarnaud F, and Prospéro-García O (2006). Diurnal variation of arachidonoylethanolamine, palmitoylethanolamide and oleoylethanolamide in the brain of the rat. Life Sci 79, 30–37. [DOI] [PubMed] [Google Scholar]
  37. Nanou E, and Catterall WA (2018). Calcium Channels, Synaptic Plasticity, and Neuropsychiatric Disease. Neuron 98, 466–481. [DOI] [PubMed] [Google Scholar]
  38. Nelson SB, and Valakh V (2015). Excitatory/Inhibitory Balance and Circuit Homeostasis in Autism Spectrum Disorders. Neuron 87, 684–698. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Niethard N, Hasegawa M, Itokazu T, Oyanedel CN, Born J, and Sato TR (2016). Sleep-Stage-Specific Regulation of Cortical Excitation and Inhibition. Curr. Biol 26, 2739–2749. [DOI] [PubMed] [Google Scholar]
  40. Petrus E, Rodriguez G, Patterson R, Connor B, Kanold PO, and Lee H-K (2015). Vision Loss Shifts the Balance of Feedforward and Intracortical Circuits in Opposite Directions in Mouse Primary Auditory and Visual Cortices. J. Neurosci 35, 8790–8801. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Rosenberg A, Patterson JS, and Angelaki DE (2015). A computational perspective on autism. Proc. Natl. Acad. Sci. U. S. A 112, 9158–9165. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Rotaru DC, van Woerden GM, Wallaard I, and Elgersma Y (2018). Adult Ube3a Gene Reinstatement Restores the Electrophysiological Deficits of Prefrontal Cortex Layer 5 Neurons in a Mouse Model of Angelman Syndrome. J. Neurosci 38, 8011–8030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Rubin R, Abbott LF, and Sompolinsky H (2017). Balanced excitation and inhibition are required for high-capacity, noise-robust neuronal selectivity. Proc. Natl. Acad. Sci 114, E9366–E9375. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Sadowski JHLP, Jones MW, and Mellor JR (2016). Sharp-Wave Ripples Orchestrate the Induction of Synaptic Plasticity during Reactivation of Place Cell Firing Patterns in the Hippocampus. Cell Rep 14, 1916–1929. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Shen W, McKeown CR, Demas JA, and Cline HT (2011). Inhibition to excitation ratio regulates visual system responses and behavior in vivo. J. Neurophysiol 106, 2285–2302. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Steele PM, Mauk MD, De Marchena J, Roberts AC, Middlebrooks PG, Valakh V, Yashiro K, Wilfley LR, and Philpot BD (2011). Inhibitory Control of LTP and LTD : Stability of Synapse Strength Inhibitory Control of LTP and LTD : Stability of Synapse Strength. 1559–1566. [Google Scholar]
  47. Tabuchi K, Blundell J, Etherton MR, Hammer RE, Liu X, Powell CM, and Südhof TC (2007). A neuroligin-3 mutation implicated in autism increases inhibitory synaptic transmission in mice. Science 318, 71–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Tao HW, and Poo MM (2005). Activity-dependent matching of excitatory and inhibitory inputs during refinement of visual receptive fields. Neuron 45, 829–836. [DOI] [PubMed] [Google Scholar]
  49. Trettel J, and Levine ES (2003). Endocannabinoids mediate rapid retrograde signaling at interneuron right-arrow pyramidal neuron synapses of the neocortex. J. Neurophysiol 89, 2334–2338. [DOI] [PubMed] [Google Scholar]
  50. Trettel J, Fortin DA, and Levine ES (2004). Endocannabinoid signalling selectively targets perisomatic inhibitory inputs to pyramidal neurones in juvenile mouse neocortex. J. Physiol 556, 95–107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Valenti M, Viganò D, Casico MG, Rubino T, Steardo L, Parolaro D, and Di Marzo V (2004). Differential diurnal variations of anandamide and 2-arachidonoyl-glycerol levels in rat brain. Cell. Mol. Life Sci 61, 945–950. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. de Vivo L, Bellesi M, Marshall W, Bushong EA, Ellisman MH, and Tononi G (2017). Ultrastructural evidence for synaptic scaling across the wake/sleep cycle. Science (80-.). 355, 507–510. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Vogels TP, Sprekeler H, Zenke F, Clopath C, and Gerstner W (2011). Inhibitory plasticity balances excitation and inhibition in sensory pathways and memory networks. Science (80-.). 334, 1569–1573. [DOI] [PubMed] [Google Scholar]
  54. Vyazovskiy VV, Cirelli C, Pfister-Genskow M, Faraguna U, and Tononi G (2008). Molecular and electrophysiological evidence for net synaptic potentiation in wake and depression in sleep. Nat. Neurosci 11, 200–208. [DOI] [PubMed] [Google Scholar]
  55. Wallace ML, Burette AC, Weinberg RJ, and Philpot BD (2012). Maternal Loss of Ube3a Produces an Excitatory/Inhibitory Imbalance through Neuron Type-Specific Synaptic Defects. Neuron 74, 793–800. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Wang Z, Ma J, Miyoshi C, Li Y, Sato M, Ogawa Y, Lou T, Ma C, Gao X, Lee C, et al. (2018). Quantitative phosphoproteomic analysis of the molecular substrates of sleep need. Nature 558, 435–439. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Xue M, Atallah BV, and Scanziani M (2014). Equalizing excitation-inhibition ratios across visual cortical neurons. Nature 511, 596–600. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Yizhar O, Fenno LE, Prigge M, Schneider F, Davidson TJ, Ogshea DJ, Sohal VS, Goshen I, Finkelstein J, Paz JT, et al. (2011). Neocortical excitation/inhibition balance in information processing and social dysfunction. Nature 477, 171–178. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Zhou M, Liang F, Xiong XR, Li L, Li H, Xiao Z, Tao HW, and Zhang LI (2014). Scaling down of balanced excitation and inhibition by active behavioral states in auditory cortex. Nat. Neurosci 17, 841–850. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supplenetrary material
NIHMS1824787-supplement-supplenetrary_material.pdf (352.2KB, pdf)

RESOURCES