Activity of cortical and thalamic neurons during the slow (<1 Hz) rhythm in the mouse in vivo
- 1.3k Downloads
During NREM sleep and under certain types of anaesthesia, the mammalian brain exhibits a distinctive slow (<1 Hz) rhythm. At the cellular level, this rhythm correlates with so-called UP and DOWN membrane potential states. In the neocortex, these UP and DOWN states correspond to periods of intense network activity and widespread neuronal silence, respectively, whereas in thalamocortical (TC) neurons, UP/DOWN states take on a more stereotypical oscillatory form, with UP states commencing with a low-threshold Ca2+ potential (LTCP). Whilst these properties are now well recognised for neurons in cats and rats, whether or not they are also shared by neurons in the mouse is not fully known. To address this issue, we obtained intracellular recordings from neocortical and TC neurons during the slow (<1 Hz) rhythm in anaesthetised mice. We show that UP/DOWN states in this species are broadly similar to those observed in cats and rats, with UP states in neocortical neurons being characterised by a combination of action potential output and intense synaptic activity, whereas UP states in TC neurons always commence with an LTCP. In some neocortical and TC neurons, we observed ‘spikelets’ during UP states, supporting the possible presence of electrical coupling. Lastly, we show that, upon tonic depolarisation, UP/DOWN states in TC neurons are replaced by rhythmic high-threshold bursting at ~5 Hz, as predicted by in vitro studies. Thus, UP/DOWN state generation appears to be an elemental and conserved process in mammals that underlies the slow (<1 Hz) rhythm in several species, including humans.
KeywordsEEG Oscillations Sleep Neocortex Thalamocortical Electroencephalogram T-type calcium channel Thalamus Neocortical neurons
In anaesthetised cats, UP/DOWN states in TC neurons are reliant on intact corticothalamic projections . Because such projections are known to activate the group I metabotropic glutamate receptor (mGluR) , mGluR1a, on TC neurons, we earlier hypothesised that such activation is essential for the expression of UP/DOWN states in these cells . This was later confirmed in studies in thalamic slices where UP/DOWN states only became apparent in TC neurons following either the pharmacological or synaptic activation of mGluR1a , leading to the view that TC (and NRT) neurons behave as conditional oscillators during the slow rhythm . Interestingly, when TC neurons in this condition are either tonically depolarised in the presence of a small amount of mGluR1a activation or when mGluR1a is more intensely activated, UP/DOWN states can be replaced by a novel form of rhythmic burst firing, termed high-threshold (HT) bursting (Fig. 3a–c, upper panels) [31, 33]. Whilst there is substantial evidence from extracellular recordings from the cat lateral geniculate nucleus (LGN) that HT bursting also occurs in vivo [33, 42], this phenomenon has yet to be unequivocally observed with intracellular recordings in the intact brain.
Up to now, the majority of cellular level investigations into the neuronal basis of the slow (<1 Hz) rhythm have largely been restricted to cats (e.g. [13, 60, 61, 62]) and rats (e.g. [35, 48, 54, 56]), with a limited number of key studies performed in humans using depth recordings (e.g. [8, 38, 52]). Whilst in the main, these studies have uncovered a consistent picture with regard to the correlation between slow waves and UP/DOWN states in specific neuronal groups, the applicability of these findings to other mammals is largely assumed on the basis that sleep occurs ubiquitously across different species . Whilst this assumption may well turn out to be justified, one species where it is becoming increasingly important to definitively establish that the mechanisms of slow rhythm generation are conserved is the mouse. Given the ever-increasing application of genetic manipulation in these animals as well their growing use in basic sleep research, it is paramount to show that the basic features of UP/DOWN states in this species are equivalent to those observed in higher mammals. To this end, we have now obtained in vivo intracellular recordings of UP and DOWN states in neocortical and thalamic neurons in anaesthetised mice. In doing so, we demonstrate that neurons from both regions exhibit UP/DOWN state properties that are broadly equivalent to those previously described in other mammals . We also show that tonic depolarisation of TC neurons that are exhibiting UP/DOWN states can reveal rhythmic HT bursting at ~5 Hz, as predicted by in vitro studies [31, 33]. The importance of these findings is discussed within the context of slow (<1 Hz) rhythm-related memory processes and basic cortical and TC circuit operations.
All procedures were carried out in accordance with local ethical committee guidelines and the UK Animals (Scientific Procedure) Act, 1986. All efforts were made to minimise the suffering and number of animals used in each experiment.
Surgical procedures and in vivo electrophysiology
Adult mice (C57BL/6, P30-60, Harlan, UK) were anaesthetised with urethane (0.8 g/kg) and supplemented with ketamine and xylazine (20 and 2 mg/kg, respectively) to maintain stable anaesthesia as required (note: we did not observe any notable differences in the appearance of the slow oscillation across the age range of animals used). Body temperature was maintained at constant levels using a heating plate (Supertech Ltd., Pecs, Hungary). After exposing the skull, a small hole was drilled above the target area at stereotaxic coordinates (0.5 mm anterior of bregma; 3.5 mm lateral to midline) . The dura was then punctured and a glass electrode lowered into the cortex. Stainless steel screws (0.8 mm diameter) were placed into the skull over the cerebellum and the frontal cortex, which served as a ground and reference, respectively, and were used to monitor the EEG. Extracellular single units were recorded using glass pipettes filled with 0.5 M NaCl (impedance, 5–15 MΩ) connected to a Neurolog 104 differential amplifier (Digitimer Ltd., Welwyn Garden City, UK). Intracellular recordings were obtained using glass micropipettes filled with 1 M potassium acetate (impedance, 60–80 MΩ) connected to an Axoclamp-700B amplifier (Molecular Devices, Sunnyvale CA, USA). All extracellular and intracellular recordings were obtained from a depth corresponding to layer 5 (0.6–0.8 mm). Data were digitally acquired using a Digidata 1440A interface (Molecular Devices) and stored on a PC for offline analysis. At the end of the experiment, animals were given a lethal dose of urethane.
Average subthreshold UP states were calculated for a period of at least 1,000 ms following the detected start of an UP state and were constructed from at least 10 consecutive UP states. Average firing rate plots were computed from at least 20 individual UP states. For a given neuron, the voltage difference between UP and DOWN states was taken as the difference between the two peaks in the membrane potential histogram. For examining the frequency components of subthreshold UP states, we averaged the power spectra for at least 20 consecutive UP states. Unless otherwise stated, all quantitative data are expressed as the mean±SEM and statistical significance was assessed with Student’s t test.
Properties of the slow rhythm in mice under anaesthesia
Recordings of extracellular multiunit activity (MUA) obtained in combination with depth LFP recordings revealed that, as expected, the negative component of the LFP (i.e. UP state) was associated with neuronal firing (mean frequency, 45.6 ± 3.3 Hz; duration of epochs, 552 ± 65 ms; n = 200 epochs from 10 recordings), whilst the positive phase (i.e. DOWN state) was mostly related to neuronal silence (duration of silent epochs, 779 ± 85 ms; n = 200 epochs from 10 recordings; Fig. 4c, d). Examination of the average firing during UP states revealed that action potential output was at a maximum near the start of the UP state, decreased to a stable level from ~200 to 400 ms after UP state commencement and then gradually declined (Fig. 4c, right). The slower modulation of the slow rhythm as noted in the EEG was also apparent in MUA recordings (Fig. 4d).
Extracellular single-unit recordings of neocortical neurons during the slow rhythm reveal two main types of neuronal activity
Intracellular recordings of neocortical neurons during the slow rhythm
Intracellular recordings of thalamocortical neurons during the slow rhythm
Interestingly, when TC neurons were subjected to a small amount of steady hyperpolarising current, additional LTCPs were often observed during the DOWN state in a manner reminiscent of that previously shown for cat TC neurons, recorded both in vivo [15, 62] and in vitro [31, 33, 73], where it has been referred to as a ‘grouped’ δ oscillation (Fig. 9a, lower traces; cf Figs. 2c and 3a–c, bottom panels). Further steady hyperpolarisation of TC neurons to prevent both action potential firing and LTCP generation revealed a striking pattern of subthreshold activity consisting of barrages of synaptic activity (mean peak frequency, 31.4 ± 2.3 Hz; n = 3) that were preceded by a stereotypical depolarising event that had the appearance of a ‘subthreshold LTCP’ (indicated by arrows in Fig. 9b). Interestingly, at times, these events occurred rhythmically in groups during DOWN states in a manner that was essentially identical to the grouping of ‘full-blown’ LTCPs described above (i.e. Fig. 9a, lower traces). These events appeared to be non-synaptic in nature because their amplitude did not markedly change when cells were depolarised or hyperpolarised (not illustrated). As with neocortical RS cells, 2 out of the 4 recorded TC neurons also exhibited spikelets (time to peak, 1.2 ± 0.4 ms; τ decay, 2.2 ± 0.2 ms; n = 20 events from 2 neurons; Fig. 9d).
Lastly, when TC neurons exhibiting UP/DOWN states were subjected to sufficient steady depolarising current, we observed HT bursting at 5.3 ± 0.4 Hz (n = 3). This HT bursting could occur simultaneously with UP/DOWN states (Fig. 10a, top) or could overwhelm UP/DOWN state generation to fully define the output of these cells (Fig. 10b, top). In all senses, the HT bursting observed in vivo was indistinguishable from that previously noted in vitro (i.e. Fig. 3a–c, upper panels) [30, 31, 33, 34, 41, 42, 43].
The main findings of this study are: (1) under anaesthesia, the mouse EEG exhibits a prominent slow rhythm as has been observed in other mammalian species ; (2) this slow rhythm can be modulated by an additional slower oscillation at ~0.05–0.3 Hz; (3) the slow rhythm is accompanied by epochs of γ oscillations during UP states; (4) extracellular and intracellular recordings of neocortical neurons in layer 5 reveal two types of cells which exhibit distinct patterns of firing during UP states, namely, RS and IB neurons; (5) in RS cells, action potentials during UP states are seemingly sporadic, synaptic activity occurs in the γ frequency range and fast events resembling so-called spikelets are sometimes apparent; (6) in IB cells, UP states commence with a high-frequency burst that is followed by additional firing, synaptic activity occurs at a lower peak frequency than in RS cells and spikelets are not evident; and (7) intracellular recordings of TC neurons show that UP states in these cells always commence with an LTCP and associated burst of action potentials, which is not followed by additional firing, exhibit subthreshold events which resemble LTCPs and also display spikelets. Thus, neocortical and TC neurons in the mouse brain exhibit UP/DOWN states that possess broadly similar properties to those observed in other species .
Similarities between UP and DOWN states in different species
In the main, the results of the current study conducted in mice are in good agreement with those obtained from investigations in other mammals. Firstly, in all previous studies, the slow rhythm that is observed under anaesthesia is a highly rhythmic entity with a frequency close to, or slightly below, 1 Hz [35, 54, 56, 60, 61]. Although the mean frequency in mice is slightly greater than that noted for rats [35, 54, 56], and especially cats [60, 61], it clearly represents the same phenomenon. This is most evident from examining intracellular recordings of neocortical and TC neurons where the properties of UP and DOWN states are extremely similar to those seen previously in higher mammals [60, 61]. In particular, the sculpting of neocortical UP states by recurrent barrages of synaptic activity, the presence of spikelets during both neocortical and thalamic UP states, the commencement of UP states in TC neurons by LTCP-mediated bursts and the presence of so-called ‘grouped’ δ activity in TC neurons are also consistent findings in other species [15, 31, 33, 62, 73]. Secondly, during UP states in mice, we observed prominent γ frequency network oscillations as also observed during UP states in the ferret [12, 26], cat [63, 64] and human  neocortex. Thirdly, the observation that some UP states in the neocortex emerge from a baseline of low-frequency firing again fits well with work carried out in neocortical slices from the ferret [12, 55] and in the intact brain of cats  and humans [52, 70].
In addition to the above similarities, we also noted that the slow rhythm in mice was modulated by an additional slower oscillation at ~0.04–0.3 Hz. Whilst a cyclic modulation of brain state with a periodicity of ~10 min has previously been shown to occur under urethane anaesthesia , to our knowledge, a rhythmic modulation of slow waves on the timescale of several seconds has not been described. A further novel and striking finding in this study was that, following tonic depolarisation, UP/DOWN states in TC neurons could be replaced by rhythmic HT bursting at ~5 Hz. This is entirely consistent with findings in thalamic slices [31, 33] and shows unequivocally that HT bursting occurs in vivo (see also [33, 42]).
Potential insights into slow rhythm mechanisms
The current study provides several potential insights into the mechanisms of slow rhythm generation. Most notably, the finding that UP states in TC neurons always start with a robust LTCP-mediated burst supports our previous suggestion that this event may be a key trigger for initiating activity in the neocortex [18, 24]. Indeed, evidence from cats clearly shows that, in the absence of thalamic input, UP states in the neocortex become much less frequent and robust [66, 68]. An additional noteworthy finding in this study is the extent to which a large proportion of layer 5, presumably pyramidal, neurons also exhibit prominent bursts at the beginning of an UP state. This phenomenon is also apparent in intracellular recordings from the cat neocortex (see Fig. 2b) [60, 61] and strongly implies that these neurons may also play a key role in reinforcing UP states and maintaining rhythmicity in the neocortical network.
An often overlooked finding from earlier studies performed in cats is the presence of so-called spikelets (sometimes referred to as fast prepotentials) that are present during UP states in both neocortical  and TC neurons [59, 67]. In these studies, such spikelets were proposed to represent regenerative dendritic potentials. However, an alternative interpretation is that spikelets are action potentials from electrotonically coupled neurons, i.e. communicated via gap junctions (GJs) [6, 32, 33, 41, 43]. Whilst evidence for GJ coupling between neocortical pyramidal neurons in mature animals is not extensive (but see ), we have previously shown that spikelets in TC neurons recorded from slices of the cat LGN maintained in vitro are almost certainly due to GJ coupling with other TC neurons [6, 32, 33, 41]. If GJ coupling is indeed operational between TC neurons in the mouse during UP states in vivo, the low-pass nature of this coupling predicts that LTCP events should be efficiently propagated between cells. In fact, upon steady hyperpolarisation, subthreshold UP states in TC neurons commence with a slow event that intuitively has the appearance one would expect from an LTCP that has been electrotonically communicated from a distinct TC neuron (see arrows in Fig. 9b). Thus, as indicated previously [18, 33], the possibility exists that a certain amount of synchronisation of the slow rhythm may occur locally in the thalamus via electrotonic coupling. The possibility that such coupling is also involved in synchronising neocortical activity during UP states should also not be discounted.
A host of studies in both humans and animals now support the idea that the slow rhythm of NREM sleep is involved in memory consolidation . However, the manner in which such consolidation takes place at the cellular and network level is still not well understood. One line of thought suggests that the coordinated replay of spike sequences during UP states in NREM sleep that have occurred in prior wakefulness may form the basis for strengthening connections between relevant groups of neurons . In terms of the synaptic changes that take place, several studies suggest that sleep is associated with a net synaptic depression [3, 69], indicating that long-term depression (LTD) may be central to understanding the cellular changes that are linked to sleep-related memory processes. This is interesting because it has been demonstrated that the pairing of EPSPs with burst firing in layer 5 pyramidal neurons is an effective way to bring about LTD in these cells [4, 23]. As we have shown here, the combination of EPSPs and burst firing is commonly observed in layer 5 cells during the slow rhythm (i.e. mainly at the start of UP states; e.g. Figs. 5c, d and 7a, b), thereby potentially setting up the ideal context for the induction of LTD. Of course, whether bursting in layer 5 neocortical neurons occurs to the same extent during the slow rhythm of natural sleep as it does during anaesthesia remains to be established. However, it is tempting to speculate that, were this to be the case, it may play a key role in mediating synaptic plasticity and, ultimately, the consolidation of memory. Aside from intracortical processes, the robust burst of action potentials that consistently marks the beginning of an UP state in TC neurons may also play an important role in inducing plasticity during sleep. For example, apart from acting as a potential trigger for initiating widespread UP states in the neocortex, the significant Ca2+ entry associated with these bursts, both directly (i.e. in TC neurons) and indirectly (i.e. in neocortical neurons) may establish a window of opportunity for subsequent modifications in synaptic strength in either cell type .
In this study, we have shown that UP/DOWN states in neocortical and thalamic neurons in mice possess broadly equivalent properties to those observed in other mammals. This suggests that the cellular basis of slow rhythm generation is a conserved property of corticothalamic networks across different species. The main consequence of this is that insights garnered from mice into both the basic mechanisms of slow sleep rhythms as well as the fundamental dynamics of the neocortex are likely to be highly translatable to higher organisms, including humans.
This work was supported by the Wellcome Trust grants 71436, 78403, 91882 awarded to VC and 78311 awarded to SWH.
Conflicts of interest
The authors declare that they do not have any conflicting interests regarding the findings or interpretations of this manuscript.
This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.
- 53.Paxinos G, Franklin KBJ (2001) The mouse brain in stereotaxic coordinates. Academic, San DiegoGoogle Scholar