1. Neuroscience
Download icon

Chronic social isolation reduces 5-HT neuronal activity via upregulated SK3 calcium-activated potassium channels

  1. Derya Sargin
  2. David K Oliver
  3. Evelyn K Lambe Is a corresponding author
  1. University of Toronto, Canada
Short Report
Cited
3
Views
1,213
Comments
0
Cite as: eLife 2016;5:e21416 doi: 10.7554/eLife.21416

Abstract

The activity of serotonin (5-HT) neurons is critical for mood regulation. In a mouse model of chronic social isolation, a known risk factor for depressive illness, we show that 5-HT neurons in the dorsal raphe nucleus are less responsive to stimulation. Probing the responsible cellular mechanisms pinpoints a disturbance in the expression and function of small-conductance Ca2+-activated K+ (SK) channels and reveals an important role for both SK2 and SK3 channels in normal regulation of 5-HT neuronal excitability. Chronic social isolation renders 5-HT neurons insensitive to SK2 blockade, however inhibition of the upregulated SK3 channels restores normal excitability. In vivo, we demonstrate that inhibiting SK channels normalizes chronic social isolation-induced anxiety/depressive-like behaviors. Our experiments reveal a causal link for the first time between SK channel dysregulation and 5-HT neuron activity in a lifelong stress paradigm, suggesting these channels as targets for the development of novel therapies for mood disorders.

https://doi.org/10.7554/eLife.21416.001

eLife digest

Major depressive disorder is a common and debilitating disease that interferes with the afflicted person’s everyday life. While some patients do benefit from antidepressant treatments, these medications need to be taken for several weeks before they become effective. Still, a large proportion of patients do not recover fully and some do not respond at all to the existing treatments. As a result, there is a need to find new and more effective treatments for depression.

The most widely used antidepressant drugs target the chemical messenger or neurotransmitter called serotonin. The majority of nerve cells that produce serotonin are located in a region deep in the brain known as the dorsal raphe nucleus. When active, these nerve cells release serotonin; this in turn controls the cells’ own activity as well as the activity of a large number of connected nerve cells located throughout the brain. Any disruption in this system will have a widespread impact and can potentially increase the risk of disturbed moods. However, it was not exactly clear what alters the activity of serotonin-producing nerve cells in depression.

Now, Sargin et al. have identified a previously unknown mechanism that underlies changes to the activity of serotonin-producing nerve cells. Keeping mice isolated for a prolonged period elicits the symptoms of depression. Sargin et al. found that serotonin-producing nerve cells were dramatically less active in isolated mice and that a specific type of ion channel protein (the SK3 channel) was more abundant in these nerve cells. A higher amount of this channel inhibits the activity of nerve cells. Blocking these inhibitory SK3 channels (using a drug that can be obtained from bee venom) restored normal activity in the serotonin-producing cells. Moreover, this treatment alleviated the depressive symptoms of the isolated mice.

The findings of Sargin et al. suggest a new way to treat the symptoms of depression. Yet to translate them into an accessible treatment for patients, future work will be required to develop drugs that can specifically and potently target the affected channel.

https://doi.org/10.7554/eLife.21416.002

Introduction

Major depression is a prevalent and debilitating disease for which standard treatments remain ineffective. Social isolation has long been implicated as a risk factor for depression in humans (Cacioppo et al., 2002,2006; Holt-Lunstad et al., 2010) and induces anxiety- and depressive-like behaviors in rodents (Koike et al., 2009; Wallace et al., 2009; Dang et al., 2015; Shimizu et al., 2016; for review see Fone and Porkess, 2008; Lukkes et al., 2009). The most frequently prescribed medications for depression and anxiety disorders target the serotonin (5-HT) producing neurons (Blier et al., 1990; Nutt, 2005), the majority of which are located in the dorsal raphe nucleus (DRN) (Descarries et al., 1982). The activity of 5-HT neurons is highly vulnerable to stress (Lira et al., 2003; Bambico et al., 2009; Espallergues et al., 2012; Challis et al., 2013) and is critical for depressive-like, anxiogenic, and reward-associated behaviors (Liu et al., 2014; Teissier et al., 2015; Urban et al., 2016). Interestingly, social isolation in rodents has been shown to affect endogenous 5-HT release and 5-HT turnover in postsynaptic areas (Heidbreder et al., 2000; Muchimapura et al., 2002, 2003). However, it is unknown how social isolation affects the activity of DRN 5-HT neurons themselves. Identification of changes in the activity of these neurons may uncover novel therapeutic targets for depression and anxiety disorders.

Here, we show that chronic social isolation leads to a reduction in the excitability of DRN 5-HT neurons. Their firing activity to optical, electrophysiological, and neuromodulatory stimulation are all reduced after social isolation. Specifically, we have identified that the reduction in the firing activity of 5-HT neurons results from alterations in the function and expression of small-conductance Ca2+-activated K+ (SK) channels. Furthermore, inhibition of SK channels normalizes the activity of 5-HT neurons and restores the behavioral deficits observed after chronic social isolation.

Results

Social isolation reduces excitability of 5-HT neurons

To examine how chronic social isolation affects DRN 5-HT neurons, we performed whole-cell electrophysiology on acute slices from mice expressing ChR2-EYFP under Tph2 promoter (Zhao et al., 2011) (Figure 1A). Mice were single-housed after weaning for ≥7 weeks and group-housed littermates were used as controls. A striking difference was observed in 5-HT neuronal excitability to depolarizing current steps, with significantly fewer action potentials generated in 5-HT neurons from single-housed mice (Figure 1B–C). This difference did not arise from altered membrane characteristics (Figure 1—figure supplement 1A–C) nor altered GABA tone, as probed with picrotoxin and CGP52432 used to block GABAA and GABAB receptors, respectively (Figure 1—figure supplement 1D). The inhibitory current responses to the 5-HT1A receptor agonist 5-CT were comparable between the groups (Figure 1—figure supplement 2A–B). Furthermore, the social isolation-induced difference in the intrinsic excitability of 5-HT neurons persisted in the presence of 5-HT1A receptor blockade with WAY100635, indicating that this change does not result from altered serotonergic tone nor altered 5-HT1A receptors (Figure 1—figure supplement 2C–E).

Figure 1 with 4 supplements see all
Reduced firing of dorsal raphe 5-HT neurons after chronic social isolation is accompanied by an increase in afterhyperpolarization (AHP).

(A) Schematic representation of a coronal brainstem section (above) comprising dorsal raphe nucleus (adapted from Paxinos and Franklin, 2001) where 5-HT neurons are recorded. IR-DIC and EYFP fluorescence images (below) of a dorsal raphe 5-HT neuron being approached by a recording pipette. Scale bars, 10 μM. (B) Current-clamp traces of a 5-HT neuron from a group-housed (above) and a single-housed (below) mouse in response to a 500 pA depolarizing step. (C) Input-output curve showing spike frequency (Hz) of 5-HT neurons in response to a series of depolarizing current (pA) injections. The firing frequency (Hz) of 5-HT neurons from single-housed mice (n = 38 neurons) is reduced compared to 5-HT neurons from group-housed mice (n = 33 neurons) indicating reduced excitability (two-way repeated-measures ANOVA, effect of housing, F (1690) = 9.19, p=0.003; Newman-Keuls posthoc test, *p<0.05, **p<0.01) (D) Current-clamp recordings of a 5-HT neuron from a group-housed mouse (left, inset below) and a single-housed mouse (right, inset below) showing action potential firing in response to 20 Hz optogenetic stimulation by blue light. Note that the neuron from the group-housed mouse is able to respond with 20 Hz firing frequency to blue light whereas the neuron from the single-housed mouse responds with a 7 Hz firing frequency. (E) The frequency (Hz) of action potentials in response to optogenetic stimulation is reduced in 5-HT neurons from single-housed mice (n = 20 neurons) compared to neurons from group-housed mice (n = 17 neurons) (unpaired t-test, *p<0.05). (photocurrents; group-housed: −343 ± 32.3 pA, single-housed: −269 ± 29.4 pA, p=0.1) (F) Current-clamp recordings of a 5-HT neuron from a group-housed mouse (left, inset below) and a single-housed mouse (right, inset below) showing action potential firing in response to 20 Hz electrophysiological stimulation. The neuron from the group-housed mouse produced spike doublets in response to the strong electrophysiological stimulation (left inset). The neuron from the single-housed mouse responded with missing spikes (right inset). (G) The frequency (Hz) of action potentials in response to electrophysiological stimulation is reduced in 5-HT neurons from single-housed mice (n = 14 neurons) compared to neurons from group-housed mice (n = 13 neurons) (unpaired t-test, **p<0.01). (H) The first spike of the 25 pA depolarizing step of a current-clamp trace from each 5-HT neuron from group-housed (n = 33 neurons) and single-housed (n = 37 neurons) mice is averaged in order to obtain the resulting trace showing the AHP difference. The peak value of the AHP (mV) (I) and AHP area (mVxs) (J) are greater in 5-HT neurons from single-housed mice, relative to group-housed mice (unpaired t-test, **p<0.01). Data are represented as mean ± S.E.M.

https://doi.org/10.7554/eLife.21416.003

To investigate further the social isolation-induced changes in 5-HT neuron excitability, we evoked action potentials optogenetically (20 Hz blue light). Firing frequency of 5-HT neurons from single-housed mice in response to light was significantly reduced compared to those from group-housed littermates (Figure 1D–E). In contrast to optical stimulation, we were able to activate 5-HT neurons further using 20 Hz strong electrophysiological stimulation with current input >500 pA (with doublets in some neurons from group-housed mice). Again, 5-HT neurons from group-housed mice responded at a significantly higher frequency than those of single-housed mice (Figure 1F–G). Overall, chronic social isolation has led to a reduction in the excitability of 5-HT neurons rendering them less responsive to both optical and electrophysiological stimulations.

DRN 5-HT neurons receive numerous inputs (Dorocic et al., 2014; Ogawa et al., 2014; Weissbourd et al., 2014) that release excitatory neuromodulators such as orexins that are important for regulating their firing activity (Brown et al., 2001, 2002 Sakurai et al., 2005; Tsujino and Sakurai, 2009). Furthermore, the pacemaker activity of 5-HT neurons in vivo as well as their firing activity on slice is largely dependent on noradrenergic tone (Baraban et al., 1978; Liu et al., 2005). To examine whether chronic social isolation perturbs the firing activity of DRN 5-HT neurons in response to excitatory neuromodulation, we investigated the effects of orexinB and phenylephrine, which act respectively on hypocretin2 and α1-adrenergic receptors. Application of orexinB or phenylephrine on slice depolarized the majority of 5-HT neurons to threshold and elicited action potentials. Quantification of spike frequency revealed a significant suppression in the peak orexinB (Figure 1—figure supplement 3A,C) and the peak phenylephrine effect (Figure 1—figure supplement 3B,D) in 5-HT neurons of single-housed mice. These differences in neuromodulator-elicited excitability do not appear to reflect altered efficacy of the receptors nor channel effectors, since upon voltage-clamp examination both neuromodulators produced robust inward excitatory currents comparable between groups (Figure 1—figure supplement 3E–H). Taken together, 5-HT neurons from single-housed mice generated significantly reduced action potential frequency in response to stimulation of typical excitatory neuromodulator receptors, suggesting a disturbance in modulation and firing activity in response to excitatory inputs.

Social isolation increases afterhyperpolarization (AHP) and disturbs the normal molecular balance of SK channels

A key characteristic of 5-HT neurons is their long-duration action potential followed by a medium duration AHP (Beck et al., 2003; Kirby et al., 2003; Rouchet et al., 2008; Alix et al., 2014). Since AHP modulation is closely linked to the alterations in the firing activity of 5-HT neurons (Rouchet et al., 2008; Crespi, 2009), we sought to determine whether the observed reduction in their activity is due to a more pronounced AHP. We found that the peak and area of rheobase AHP were significantly larger for the single-housed mice (Figure 1H–J). The larger AHP in 5-HT neurons after social isolation renders them less excitable and less responsive to stimulation.

Since AHP is dependent on calcium (Ca2+) (Sah, 1996; Faber and Sah, 2003), we examined whether blocking voltage-gated Ca2+ channels (VGCC) would modify the firing activity of 5-HT neurons. Using the VGCC blocker CdCl2, the difference in firing frequency of 5-HT neurons from group- and single-housed mice was abolished (Figure 1—figure supplement 4A–B). 5-HT neurons from both groups responded with such a robust increase in their firing frequency that they went into a depolarization block at current steps >100 pA (data not shown). As the medium duration AHP in 5-HT neurons has been attributed to SK channels (Scuvée-Moreau et al., 2004), we next considered whether we could restore the firing activity of 5-HT neurons from single-housed mice by inhibiting these channels. Using apamin, a blocker of the SK family of channels (Blatz and Magleby, 1986; Köhler et al., 1996), the firing frequency of 5-HT neurons in response to depolarizing current steps became comparable between groups (Figure 1—figure supplement 4C–D). The intracellular Ca2+ levels relative to baseline (dF/F) in response to a set number of action potentials were similar between 5-HT neurons of group- and single-housed mice (Figure 1—figure supplement 4E–F), indicating that alterations in expression levels and/or sensitivity of SK channels to Ca2+ rather than changes in Ca2+ responses might contribute to the reduced firing activity of 5-HT neurons upon chronic social isolation.

Two subtypes of SK channels, SK2 (KCa2.2) and SK3 (KCa2.3), exist in the rodent DRN (Stocker, 2000). The discrete contribution of each subtype to 5-HT neuron excitability is not known. Both subtypes are sensitive to apamin (Adelman et al., 2012); but SK2 can be blocked selectively with a novel toxin Lei-Dab7 (Shakkottai et al., 2001; Aidi-Knani et al., 2015). With this pharmacological approach, we reveal for the first time a significant contribution of SK2 to the electrophysiological regulation of 5-HT neurons in group-housed mice (Figure 2A–B). Yet in single-housed mice, surprisingly, Lei-Dab7 did not produce a significant effect (Figure 2C–D). In contrast, blocking both SK2 and SK3 with apamin rendered the excitability of 5-HT neurons comparable in both groups with a significantly greater increase in firing frequency in the single-housed mice (firing frequency to 500 pA step; group-housed: 161.5 ± 6.9% baseline, n = 15 neurons, p<0.001, single-housed: 239.2 ± 16.5% baseline, n = 13 neurons, p<0.001) (Figure 2E–F). Western blot analysis in DRN revealed that there was a significant increase in expression of SK3 channels in socially-isolated mice, while the levels of SK2 channels remained unaltered (Figure 2G). Finally, apamin application decreased the rheobase AHPs of 5-HT neurons from both groups, making them comparable (Figure 2H–J).

The reduced excitability of 5-HT neurons upon chronic social isolation can be restored by blocking SK3 channels.

(A) Current-clamp traces of a 5-HT neuron in response to a 500 pA depolarizing step before (Baseline, above) and after application of SK2 blocker, Lei-Dab7 (100 nM) (+Lei-Dab7, below) from a group-housed mouse. (B) Input-output curve showing increased excitability upon application of Lei-Dab7 in 5-HT neurons of group-housed mice (Baseline, n = 17 neurons, +Lei-Dab7, n = 11 neurons) (two-way repeated-measures ANOVA, effect of drug, F (1260) = 5.87, p=0.023; Newman-Keuls posthoc test, *p<0.05, **p<0.01). (C) Current-clamp traces of a 5-HT neuron in response to a 500 pA depolarizing step before (Baseline, above) and after application of SK2 blocker, Lei-Dab7 (100 nM) (+Lei-Dab7, below) from a single-housed mouse. (D) The firing frequency (Hz) of 5-HT neurons from single-housed mice did not change upon Lei-Dab7 (Baseline, n = 16 neurons, +Lei-Dab7, n = 9 neurons) (two-way repeated-measures ANOVA, effect of drug, F (1230) = 2.39, p=0.13). (E) Current-clamp traces of 5-HT neurons in response to 500 pA depolarizing steps after subsequent application of apamin (200 nM) to block SK3 from a group-housed (above) and a single-housed (below) mouse. (F) Although the group difference persisted after SK2 blockade with Lei-Dab7, the subsequent application of apamin to block SK3 rendered the firing frequency (Hz) of neurons from group-housed (n = 15 neurons) and single-housed (n = 13 neurons) mice comparable (two-way repeated-measures ANOVA, effect of housing, F (1260) = 0.67, p=0.42). (G) Representative immunoblots (top) and quantification (bottom) showing protein levels of SK2 and SK3 channels in the DRN of group- and single-housed mice. SK3 channel expression is significantly increased in single-housed mice (n = 6) compared to group-housed mice (n = 6) (unpaired t-test, p=0.027) while SK2 channel expression seems unaltered (unpaired t-test, p=0.98). Gh; group-housed, Sh; single-housed. (H) Averaged superimposed spikes showing the AHP difference of 5-HT neurons before and after application of apamin in group-housed (Baseline; n = 18, +Apamin; n = 10 neurons) and single-housed (Baseline; n = 16, +Apamin; n = 8 neurons) mice. The peak value of the AHP (mV) (I) and AHP area (mVxs) (J) before and after application of apamin are shown (two-way ANOVA, effect of drug, peak AHP: F (1,48) = 60.02, p<0.001, AHP area: F (1,48) = 60.59, p<0.001, Newman-Keuls posthoc test, **p<0.01). Data are represented as mean ± S.E.M.

https://doi.org/10.7554/eLife.21416.008

Systemic apamin treatment normalizes anxiety- and depressive-like behaviors in socially isolated mice

To characterize anxiety-/depressive-like behaviors after chronic social isolation, group-housed and single-housed mice were subjected to three different behavioral tests: novelty suppressed feeding, tail suspension, and sucrose preference. To assess the therapeutic potential of inhibiting SK channels for treating depression and anxiety, we systemically administered the SK channel blocker apamin (i.p.) in a group of single-housed mice 30 min before each test.

Single-housed mice showed significantly longer latencies to start feeding in the novelty suppressed feeding test compared to the group-housed mice (Figure 3A), indicating elevated anxiety-like behavior after chronic social isolation. Blockade of SK channels by systemic apamin treatment resulted in a trend towards an attenuation of the increased latency to feed (Figure 3A). To assess appetitive motivation, we measured the amount of food consumed in the homecage following the novelty suppressed feeding test. Interestingly, single-housed mice consumed significantly more food in the homecage compared to the group-housed mice (Figure 3—figure supplement 1). The consumption of food in single-housed mice was restored to control levels following acute systemic apamin treatment (Figure 3—figure supplement 1), indicating not only that SK channels may play an important role in feeding behavior, but also that modulation of these channels may be useful in the treatment of stress/anxiety related eating disorders.

Figure 3 with 1 supplement see all
Decreased anxiety/depression-like behaviors in single-housed mice after systemic apamin.

(A) Single-housed mice showed increased latency to feed, which was only partially recovered upon apamin treatment in the novelty suppressed feeding test (one-way ANOVA, group effect, p=0.001, Newman-Keuls posthoc test, *p<0.05, **p<0.01; a priori t-test between Sh and Sh+Apamin, #p=0.09). (B) Single-housed mice showed increased immobility in the tail suspension test suggesting enhanced depressive-like behavior, normalized by apamin treatment (two-way ANOVA, main effect of group, p=0.01, Newman-Keuls posthoc test, **p<0.01). (C) Single-housed mice showed decreased sucrose preference, which was restored by apamin treatment (one-way ANOVA, main effect of group, p=0.02, Newman-Keuls posthoc test, *p<0.05). (D) Single-housed mice showed overall increased emotionality, which was decreased by apamin treatment (one-way ANOVA, main effect of group, p<0.01, Newman-Keuls posthoc test, **p<0.01). Data are represented as mean ± S.E.M. (n = 7–9 per group). Gh; group-housed, Sh; single-housed, ns; not significant.

https://doi.org/10.7554/eLife.21416.009

In the tail suspension test, chronic social isolation resulted in an increase in depressive-like behavior. Single-housed mice showed significantly increased cumulative immobility (Figure 3B). This impairment was completely reversed by acute systemic apamin in single-housed mice (Figure 3B). Social isolation also elicited anhedonia, a measure of depressive-like behavior as shown by decreased sucrose preference in single-housed mice, compared to the group-housed mice. SK channel blockade reduced the anhedonia to levels that were comparable to group-housed mice (Figure 3C).

In addition to exploring the behavioral data for each individual test we also combined the z-scores across all behavioral tests to produce an integrated emotionality score for each mouse (Guilloux et al., 2011). Integrated z-scores showed that chronic social isolation significantly increased overall behavioral emotionality while acute systemic treatment with apamin was sufficient to decrease emotionality back to control levels (Figure 3D).

Discussion

Our results demonstrate that chronic social isolation results in a reduction in the excitability of 5-HT neurons and renders them less responsive to stimulation. We demonstrate that inhibiting SK channels, critical regulators of AHP in these cells, can reverse the reduced excitability. This work provides the first direct link between a chronic social isolation paradigm and functional alterations in 5-HT neurons themselves. We also reveal that SK2 contributes to normal regulation of 5-HT neuronal excitability but ceases to modulate these neurons significantly after chronic stress. Behavioral analysis showed that chronic social isolation increases anxiety/depression-like behaviors, which can be normalized upon inhibition of SK channels by acute systemic apamin.

In humans, polymorphisms on the gene encoding the SK3 channel have been associated with neuropsychiatric disorders characterized by emotional dysregulation including schizophrenia, bipolar disorder, and anorexia nervosa (Chandy et al., 1998; Koronyo-Hamaoui et al., 2007; Grube et al., 2011). Administration of the SK channel blocker apamin in mice and rats reduced immobility in a forced swim test (Galeotti et al., 1999; van der Staay et al., 1999), a measure of depressive-like behavior in rodents (Cryan and Slattery, 2007). Consistent with these reports, SK3 null mice show enhanced hippocampal 5-HT release and reduced immobility in forced swim and tail suspension tests, indicative of an antidepressant like phenotype (Jacobsen et al., 2008). Our work demonstrates for the first time that social isolation, a major risk factor for depression and anxiety, results in reduced 5-HT neuronal activity due to upregulated SK3 channel function in the DRN. Moreover, we show that systemic treatment with apamin, an inhibitor of SK channels improves behavioral deficits induced by chronic social isolation. Consistent with studies that have implicated SK channel function in neuropsychiatric conditions, our current findings suggest that SK channel modulation is a promising therapeutic target for disorders of emotional disturbance such as depression and anxiety. Apamin however has been reported to have adverse side effects in rats (van der Staay et al., 1999) close to the beneficial dose in the current mouse study. The diverse expression of SK channels in different tissue types and the lack of tissue/subtype specific modulators or inhibitors are currently the limiting factors for therapeutic interventions targeting SK channels. Further research focused on understanding the differential regulation, modulation and function of SK channel subtypes may shed light on the development of effective disease treatment strategies. Importantly, development of more specific drugs targeted at inhibiting SK3 function may have significant implications for treatment of depressive and anxiety disorders.

Materials and methods

Experimental animals

All experiments were performed in Tg(Tph2-COP4*H134R/EYFP)5Gfng mice (RRID:IMSR_JAX:014555) that express channelrhodopsin EYFP fusion protein driven by the Tph2 promoter (Zhao et al., 2011), in accordance with animal protocols approved by the University of Toronto (20010374, 20011622, 20011733). All mice were maintained in C57Bl/6 background. At weaning (p21), littermate male mice were either housed individually (single-housed) or together in groups of 3–4 (group-housed). After a minimum of 7 weeks, the adult (> p70) mice were used for experiments. Taken together, this study required a total of 28 group-housed and 29 single-housed mice for electrophysiological and western blot experiments, and a total of 25 mice for behavioral experiments. All mice were housed under a 12:12h light/dark cycle with ad libitum access to both food and water.

Electrophysiology

Coronal brainstem slices (400 µm) comprising the dorsal raphe nucleus were obtained using a Dosaka Pro-7 Linear Slicer (SciMedia) in ice-cold oxygenated sucrose-substituted artificial cerebrospinal fluid (ACSF). The slices were then recovered for a minimum of 2 hr in ACSF solution containing: 128 mM NaCl, 10 mM D-glucose, 26 mM NaHCO3, 2 mM CaCl2, 2 mM MgSO4, 3 mM KCl, 1.25 mM NaH2PO4, pH 7.4 and saturated with 95% O2/5% CO2 at 31–33ºC. L-tryptophan (2.5 µM) was included during the recovery period to maintain 5-HT synthesis (Rood et al., 2014). Recording was performed in ACSF oxygenated with 95% O2/5% CO2 at 31–33ºC flowing at a rate of 3–4 ml/min. Patch pipettes (2–4 MΩ) contained the following composition: 120 mM potassium gluconate, 5 mM KCl, 2 mM MgCl2, 4 mM K2-ATP, 0.4 mM Na2-GTP, 10 mM Na2-phosphocreatine, and 10 mM HEPES buffer (adjusted to pH 7.3 with KOH). Neurons in DRN were visualized with a fixed-staged microscope (Olympus BX50WI) and 5-HT neurons were targeted based on the expression of EYFP. Recordings were focused on the dorsomedial subregion of dorsal raphe, where most of the EYFP positive neurons were visible. Whole-cell recordings were made in voltage-clamp or current-clamp mode with a Multiclamp 700B amplifier (Molecular Devices). Voltage-clamp recordings were performed at −75 mV. Optogenetic stimulation (20 Hz, 10 ms) was performed using LED light (473 nm, Thorlabs). For electrophysiological stimulation, action potentials were elicited by applying brief depolarizing current pulses (>500 pA, 10 ms) at 20 Hz from a membrane potential of −70 mV. All data were acquired at 20 kHz and low-pass filtered at 3 kHz using pClamp10.2 and Digidata1440 software.

Pharmacology

5-CT (100 nM, Tocris) was bath applied to probe 5-HT1A receptor currents. WAY100635 (30 nM, Sigma) was applied in ACSF to block 5-HT1A receptor responses. OrexinB (300 nM, GenScript) and R-(-) Phenylephrine hydrochloride (10 μM, Sigma) were applied in bath to measure responses of 5-HT neurons to excitatory neuromodulation. Picrotoxin (100 μM, Sigma) and CGP52432 (1 μM, Tocris) were included together in bath solution to block GABAA and GABAB receptor responses, respectively. Cadmium chloride (100 μM, Sigma) was used to block voltage gated calcium channels. Lei-Dab7 (100 nM, Tocris) was included in ACSF as a blocker of SK2 channels. Apamin (200 nM, Tocris) was used as a blocker of SK2 and SK3 channels.

Multiphoton calcium imaging

The calcium dye Fluo5F (150 μM, Invitrogen) was included in the pipette, along with Alexa Fluor-594 hydrazide (20 μM), for visualization of the neuron. Multiphoton imaging was performed using a titanium:sapphire laser (Newport) tuned to wavelength 800 nm and an Olympus Fluoview FV1000 microscope equipped with a 60X objective (numerical aperture 0.9). Images were acquired at a rate of ∼10 frames/s. Emissions were recorded in the green channel (Fluo5F signal, 495–540 nm) and the red channel (Alexa Fluor-594 signal, 570–620 nm). Analysis was performed using Fluoview software by selecting a pan-somatic area of interest and measuring the increases in green fluorescence relative to baseline fluorescence (dF/F).

Western blot

For Western blot analysis, brainstem slices were prepared in the same way as for electrophysiological experiments. DRN was then dissected and immediately frozen on dry ice. The tissue was lysed in lysis buffer (Biobasic) supplemented with protease and phosphatase inhibitors. Protein concentration was measured using Bradford assay (Sigma). For immunoblotting, 10 μg protein was loaded per lane onto 4–20% Mini-Protean precast polyacrylamide gels (Bio-Rad), separated by electrophoresis and transferred onto a nitrocellulose membrane (Bio-Rad). Membranes were then blocked with 5% non-fat milk in Tris-buffered saline and incubated overnight at 4oC with primary antibodies (rabbit anti-SK3, 1:1000, Alomone, RRID:AB_2040130; rabbit anti-SK2, 1:1000, Alomone, RRID:AB_2040126; rabbit anti-5-HT1AR, 1:1000, Millipore Bioscience Research Reagents, RRID:AB_805421; mouse anti-β-actin, 1:5000, Sigma, RRID:AB_476744). Immunoreactive bands were visualized using HRP-coupled secondary antibodies by ECL detection methods (Bio-Rad). Band intensities were analyzed using Bio-Rad Image Laboratory software and calculated relative to β-actin as the loading control.

Behavioral analysis

To assess anxiety- and depression-like behaviors, the novelty suppressed feeding, tail suspension, and sucrose preference tests were performed. Each test was separated by a minimum of 3 days. Mice were habituated to the testing room for at least 30 min. Apamin (0.3 mg/kg dissolved in 0.9% NaCl, Alomone Labs) or vehicle (0.9% NaCl) was injected intraperitoneally (i.p.) 30 min before each test (van der Staay et al., 1999; Chen et al., 2014).

Novelty suppressed feeding test

Following food deprivation for 18 hr, mice were provided with a single food pellet placed in the middle of a dimly lit plastic (45 x 45x 20 cm) box. The floor of the box was covered with bedding 2 cm deep. The latency to start feeding in a 5 min assay was measured by an observer blinded to the groups/treatment. Following the test, each mouse was returned to its homecage and the amount of food consumed within 5 min was measured.

Tail suspension test

Mice were individually suspended by their tails within a three-walled plastic box (45 x 25 x 25 cm) using adhesive tape securely attached to the tip of the tail. The test lasted for 6 min and the behavior of each mouse was videotaped. The amount of time spent immobile per minute was measured by an observer blinded to the groups and treatment and expressed as the cumulative immobility.

Sucrose preference test

Mice were habituated to drinking 2% sucrose from two bottles prior to the test. In order to assess the drinking amount, group-housed mice were housed in pairs or singles for a period of 3 days. Mice were given a choice of 2% sucrose or water in their homecage for 2 days. Water and sucrose intake were measured daily and the position of the bottles were changed after the first day. Drugs were administered on both days of the test. Data were expressed as the ratio of sucrose to water consumed and the average of both days were taken.

Z-scoring methodology was used as previously described (Guilloux et al., 2011) to assess the overall emotionality score for each mouse that ran through all the behavioral tests.

Data analysis

Electrophysiological responses were calculated as changes in membrane potential or holding current resulting from manipulations such as 473 nm light exposure or the bath application of drugs and analyzed by Clampfit (Molecular Devices) software. Spike frequencies for input-output curves were measured by counting the number of spikes in response to 500 ms depolarizing current steps. Spike frequencies for optogenetic and electrophysiological stimulations were determined by counting the number of spikes over a 2 s stimulation period. For illustrative purposes, averaged traces showing current responses and AHPs were compiled using Axograph software. Quantification of AHP potentials was performed by detecting action potentials automatically with a derivative threshold of 20 mV/ms by Axograph software. AHP area (mVxs) is measured as the area under the AHP between the midpoint of each spike (taken as time=0) and the 100 ms time point corresponding to the medium AHP (Sah, 1996; Faber and Sah, 2003). The maximum negative value of the AHP within this time window is determined as the peak AHP (mV). Statistical analysis was performed using Prism 5 software. For electrophysiology experiments, the numbers (n) that represent the number of neurons were obtained from at least 3 mice per group. For western blots, numbers (n) represent the number of mice (averages of 2 replicates per mouse). Data were analyzed using unpaired student’s t-test, two-way ANOVA or repeated-measures two-way ANOVA. Newman-Keuls post-hoc test was performed when appropriate. Data were expressed as mean ± S.E.M. and evaluated at a significance level of 0.05.

References

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
    A role for the serotonin system in the mechanism of action of antidepressant treatments: preclinical evidence
    1. P Blier
    2. C de Montigny
    3. Y Chaput
    (1990)
    The Journal of Clinical Psychiatry 51 Suppl:14–20.
  9. 9
  10. 10
    Convergent excitation of dorsal raphe serotonin neurons by multiple arousal systems (orexin/hypocretin, histamine and noradrenaline)
    1. RE Brown
    2. OA Sergeeva
    3. KS Eriksson
    4. HL Haas
    (2002)
    Journal of Neuroscience 22:8850–8859.
  11. 11
  12. 12
  13. 13
  14. 14
  15. 15
  16. 16
  17. 17
  18. 18
  19. 19
  20. 20
  21. 21
  22. 22
  23. 23
  24. 24
  25. 25
  26. 26
  27. 27
  28. 28
  29. 29
  30. 30
  31. 31
  32. 32
  33. 33
  34. 34
  35. 35
  36. 36
  37. 37
  38. 38
  39. 39
  40. 40
  41. 41
  42. 42
    The Mouse Brain in Stereotaxic Coordinates (2nd edn)
    1. G Paxinos
    2. KBJ Franklin
    (2001)
    San Diego: Elsevier Academic Press.
  43. 43
  44. 44
  45. 45
  46. 46
  47. 47
  48. 48
  49. 49
  50. 50
  51. 51
  52. 52
  53. 53
  54. 54
  55. 55
  56. 56
  57. 57

Decision letter

  1. Sacha B Nelson
    Reviewing Editor; Brandeis University, United States

In the interests of transparency, eLife includes the editorial decision letter and accompanying author responses. A lightly edited version of the letter sent to the authors after peer review is shown, indicating the most substantive concerns; minor comments are not usually included.

[Editors’ note: a previous version of this study was rejected after peer review, but the authors submitted for reconsideration. The first decision letter after peer review is shown below.]

Thank you for submitting your work entitled "Chronic social isolation reduces 5-HT neuronal activity via upregulated SK3 calcium-activated potassium channels" for consideration by eLife. Your article has been reviewed by two peer reviewers, one of whom, Sacha B Nelson, is a member of our Board of Reviewing Editor and the evaluation has been overseen by Richard Aldrich as the Senior Editor.

Our decision has been reached after consultation between the reviewers. Based on these discussions and the individual reviews below, we regret to inform you that your work will not be considered further for publication in eLife.

Although the topic of the experiments is interesting and the results presented are convincing, the reader is left unsure of the behavioral impact of the changes in excitability observed and as to whether this is indeed as suggested a "promising therapeutic target for disorders of emotional disturbance such as depression and anxiety." It remains possible that SK channels are too broad a target to be useful, and even that these changes in excitability are consequences of, but not causally related to, the effects of social deprivation. For example, one reviewer noted (during the reviewer discussion) that "a more recent paper used a somewhat similar model in mice (Neuropsychopharmacology. 2014 Dec; 39(13):2928-37. doi: 10.1038/npp.2014.162) and doesn't seem to see much on the anxiety tests; they did voltammetry measures of 5-HT release in different conditions, and rather see a trend for increased rather than decreased 5-HT release (see their table1). " While it would be feasible with significant additional work to shore up the present results with additional behavioral studies (e.g. demonstrating that social isolation produces specific behavioral effects which can be normalized by normalizing the excitability of DRN neurons), this would likely take several months to complete and it is the journal policy to reject manuscripts which cannot be accepted without work that could be completed in 1-2 months at most. In this case, however, we would be willing to consider a new manuscript on the same topic, provided it addressed the issue of behavioral relevance. But the reviewer and reviewing editor felt that in its present state the manuscript would be more appropriate for a more specialized journal.

Reviewer #1:

The authors show that social isolation results in reduced excitability of serotonergic neurons in the mouse dorsal raphe, perhaps explaining the impact of social isolation on depression related symptoms. The experiments demonstrate concisely and convincingly that the decreased excitability is due to an increased afterhyperpolarization (AHP) associated with increased expression of the SK3 subtype of small conductance Ca-activated potassium (SK) channels. Interestingly, polymorphisms of the SK3 gene are known to be risk factors for psychiatric disease.

The paper is short and to the point, and the experiments are well done and clearly illustrated.

I had only minor suggestions for clarifying the physiological results presented in the manuscript. However, after discussion with the other reviewer, I am persuaded that there is still an open question as to the behavioral relevance, and hence the general interest, of the results. Although it is clear that social isolation is a risk factor for depressive symptoms and anxiety, it is not clear whether the observed change in DRN excitability plays a causal role in this.

Reviewer #2:

The work nicely demonstrates that rearing mice in social isolation lowers the excitability of serotonin raphe neurons, by modifying the afterhyperpolarization (AHP) responses of the 5-HT neurons, a biophysical property that is controlled by SK channels; they find that SK3 channel expression is increased in the raphe of single-reared mice and that pharmacological blockade of SK3 normalizes AHP and raphe neuron excitability. The paper is well done, but without addition of behavioral data, it remains limited in its conclusions and does not address the possible behavioral correlates of SK3 changes (nor the possible upstream mechanisms that induce SK3 up regulation). Thus, although presenting an interesting novel observation in the field, the results seemed somewhat too preliminary to conclude on the potential relevance of SK3 up regulation for the effects of stress on mood disorders.

Specific remarks

1) Antagonist rescues hypoactivity of the 5-HT neurons. What about behaviour? There are no behavioural data to show that the protocol of social isolation really changes behaviour, nor how SK3 antagonism could modify them. This is difficult to predict from the published literature since SK3 overexpression was found to cause learning deficits but no anxiety phenotypes (Grube et al., 2011), while SK3-KO appear to have increased anxiety (Jacobsen et al., 2008).

2) Increased of SK3 is shown with western blots, however this shows an unequal signal among cases; thus dot plots would be preferable to histograms in Figure 2G. Further, SK3-immunolabelling (such antibodies are characterized in Jacobsen et al.2008) would be a nice addition, to show more specifically an SK3 increase in the 5-HT neurons.

3) In the Results section the authors claim that difference in excitability did not result from changes in GABA tone mentioning data not shown. It would be important to show these data since there is literature showing that stress affects excitability of GABA neurons but not 5-HT neurons (e.g. Challis et al., 2013).

4) When measuring AHP amplitudes and areas, at which frequency were the cells firing (Figure 1H)? The authors report the same current step for all cells (25pA); because neurons from single-housed are hypoexcitable we can infer that the frequency is lower. This would bias the measures. Measures of medium AHP are normally taken after a depolarizing current step has been applied, and selecting the current step showing the same number of action potentials for all cells.

5) Because of the heterogeneity of the dorsal raphe neurons, the authors need to specify more precisely the localisation of their recordings.

[Editors’ note: what now follows is the decision letter after the authors submitted for further consideration.]

Thank you for submitting your article "Chronic social isolation reduces 5-HT neuronal activity via upregulated SK3 calcium-activated potassium channels" for consideration by eLife. Your article has been reviewed by two peer reviewers: Sacha B Nelson (Reviewer #1), who is a member of our Board of Reviewing Editors, and Patricia Gaspar (Reviewer #2). The evaluation has been overseen by Richard Aldrich as the Senior Editor.

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

Summary:

The authors nicely demonstrate that rearing mice in social isolation lowers the excitability of serotonin raphe neurons, by modifying their afterhyperpolarizations (AHPs), a biophysical property that is controlled by small-conductance Ca2+-activated potassium (SK) channels. They go on to show that isolation also impacts behaviors associated with anxiety and depression in rodents and that these changes are reversed by inhibiting SK channels.

Essential revisions:

The requested revisions are minor and this should not need to go back for another round of review. The full text of the reviews is given below.

Reviewer #1:

In this revised manuscript, the authors have now added three behavioral tests to support the idea that the changes in DRN SK3 activity they see following social isolation has effects on feeding and the anxiety/depression related behavior of immobility following tail suspension. The results and some other minor fixes add confidence that the observed changes in the excitability of serotonergic neurons are more likely to have a behavioral impact. There is some concern of course that changes elsewhere (e.g. in the gut) could have affected the feeding behaviors, but this concern is difficult to address.

Reviewer #2:

The authors have addressed my previous concerns concerning the validation of the social isolation model in mice, by adding behavioral experiments that demonstrate the effectiveness of their protocol to induce anxiety-like phenotypes. Moreover they indicate a causal link between SK3 upregulation and these phenotypes by showing a partial rescue of the behavioral phenotype with Apamin. Finally methodological precisions and complementary explanations were added to the text as requested.

Minor comments:

1) In their rebuttal, the authors explain the difference of their results with the Dankovski 2014 paper as a difference in protocol. However all the other supporting papers they cite have been done in rats (and they do not cite the Dankovski paper). If this is the first time the social separation model could be valuably transferred to mice, this would be worth mentioning more explicitly for future studies in mice.

2) In this regard, information about the mouse strain/gender of mice used in the different experiments is important; I could not find this info in the new version of the manuscript.

3) Results, subsection “Social isolation reduces excitability of 5-HT neurons”, first paragraph: explain what picrotoxin and CGP52432 are doing (as in the Methods).

4) Results, subsection “Systemic apamin treatment normalizes anxiety- and depressive-like 160 behaviors in socially isolated mice”, last paragraph: to see how the z-score was calculated, I went to "Lin and Sibille 2015"; this was a waste of time as the Methods only refers to Guilloux 2011. Suggestion is to keep only the useful reference.

5) Conclusion - It is interesting that SK channels are a druggable compound for depression. Since this is not entirely new, as it had already been put forth and tested - to some extent - by Crespi in 2010, it may be useful to add a sentence as to why no one has really followed this up. Are there issues of toxicity? A need for more selective agents?

https://doi.org/10.7554/eLife.21416.011

Author response

[Editors’ note: the author responses to the first round of peer review follow.]

[…] Although the topic of the experiments is interesting and the results presented are convincing, the reader is left unsure of the behavioral impact of the changes in excitability observed and as to whether this is indeed as suggested a "promising therapeutic target for disorders of emotional disturbance such as depression and anxiety." It remains possible that SK channels are too broad a target to be useful, and even that these changes in excitability are consequences of, but not causally related to, the effects of social deprivation. For example, one reviewer noted (during the reviewer discussion) that "a more recent paper used a somewhat similar model in mice (Neuropsychopharmacology. 2014 Dec; 39(13):2928-37. doi: 10.1038/npp.2014.162) and doesn't seem to see much on the anxiety tests; they did voltammetry measures of 5-HT release in different conditions, and rather see a trend for increased rather than decreased 5-HT release (see their table1). " While it would be feasible with significant additional work to shore up the present results with additional behavioral studies (e.g. demonstrating that social isolation produces specific behavioral effects which can be normalized by normalizing the excitability of DRN neurons), this would likely take several months to complete and it is the journal policy to reject manuscripts which cannot be accepted without work that could be completed in 1-2 months at most. In this case, however, we would be willing to consider a new manuscript on the same topic, provided it addressed the issue of behavioral relevance. But the reviewer and reviewing editor felt that in its present state the manuscript would be more appropriate for a more specialized journal.

We appreciate that the editors and reviewers brought a recent paper into discussion that uses a different model of social isolation and employed voltammetry to address questions with respect to serotonin release. This gives us an opportunity to talk about our model of social isolation in a more detailed manner. We isolate mice at the age of postnatal (P) day 21 upon weaning and this social isolation continues for more than 7 weeks. Our behavioral work in the new Figure 3 demonstrates that this paradigm elicits robust anxiety- and depressive-like effects.

Indeed, it has been shown in various studies that there is a critical developmental window for isolation rearing experiments in rodents and that single-housing after this critical developmental window does not lead to the same robust behavioral effects (Einon and Morgan, 1977; Fone and Porkess, 2008).

In the above-mentioned paper by Dankoski et al. 2016, mice were socially isolated later, starting at adolescence (P35-42), for a relatively brief period of 3 weeks. In this model, single-housed mice did not show robust behavioral changes compared to the paired controls. Although single- housing mice in adolescence may certainly have stressful effects (Wallace et al., 2009), the Dankoski et al. model likely represents a milder stress than ours. Still interestingly, citalopram-induced facilitation of serotonin release was blocked in this model of social isolation suggesting that even milder stressful conditions may have an impact on serotonin release under challenge.

Our electrophysiological findings that serotonin neurons become less responsive to stimulation upon chronic social isolation beginning at P21 are consistent with previous reports showing decreased release of 5-HT in postsynaptic areas after single-housing rodents starting at this developmentally-sensitive time (Jaffe et al., 1993; Bickerdike et al., 1993; Heidbreder et al., 2000; Muchimapura et al., 2002; Muchimapura et al., 2003).

Reviewer #1:

[…] I had only minor suggestions for clarifying the physiological results presented in the manuscript. However, after discussion with the other reviewer, I am persuaded that there is still an open question as to the behavioral relevance, and hence the general interest, of the results. Although it is clear that social isolation is a risk factor for depressive symptoms and anxiety, it is not clear whether the observed change in DRN excitability plays a causal role in this.

We thank the reviewer for raising the point with respect to the behavioral relevance of our findings. We have now performed the behavioral experiments showing that this chronic social isolation paradigm elicits significant differences on anxiety- and depressive-like behaviors.

Furthermore, we demonstrate that this group difference can be rescued with SK channel blockade by apamin used at a dose shown to enhance the excitability of raphe neurons in vivo (Crespi, 2009).

Reviewer #2:

[…] Specific remarks

1) Antagonist rescues hypoactivity of the 5-HT neurons. What about behaviour? There are no behavioural data to show that the protocol of social isolation really changes behaviour, nor how SK3 antagonism could modify them. This is difficult to predict from the published literature since SK3 overexpression was found to cause learning deficits but no anxiety phenotypes (Grube et al., 2011), while SK3-KO appear to have increased anxiety (Jacobsen et al., 2008).

We appreciate the valuable feedback with respect to the requirement of behavioral experiments. We used male mice for our experiments. Jacobsen et al., 2011 reported an antidepressant-like phenotype in SK3 deficient male mice (reduced immobility in tail suspension and forced swim tests), which supports our findings in this study.

With the suggestion of the reviewer, we now conducted the necessary behavioral experiments and added them into the manuscript. A new Figure 3 and Figure 3—figure supplement 1 highlight the striking behavioral differences observed in our chronically socially isolated male mice compared to group- housed littermate controls, as well as the normalization of these measures upon acute administration of apamin to block a portion of SK channels.

Interestingly, Jacobsen et al., 2011 reported an anxiety-like effect in one of the tests (zero maze) in SK3 deficient female mice. Whether sex differences exist in our model is an interesting point and will be important to follow up in a future study.

2) Increased of SK3 is shown with western blots, however this shows an unequal signal among cases; thus dot plots would be preferable to histograms in Figure 2G. Further, SK3-immunolabelling (such antibodies are characterized in Jacobsen et al. 2008) would be a nice addition, to show more specifically an SK3 increase in the 5-HT neurons.

We now included the dot plots instead of the histograms for the western blot data in Figure 2. SK3 expression is reported to be predominantly in serotonergic neurons of dorsal raphe (Stocker and Pedarzani, 2000). Serotonergic neurons characteristically have a prominent medium duration apamin-sensitive AHPs (Scuvée-Moreau et al., 2004) that do not exist to the same extent in other neuronal subtypes such as GABAergic neurons in dorsal raphe. Moreover, systemic apamin injections were shown to increase the firing rate of 5-HT neurons in vivo and resulted in a subsequent increase in 5-HT release in postsynaptic regions (Crespi, 2009) suggesting that serotonergic neurons are specifically sensitive to apamin. SK channel blockade in vivo has also been shown to increase burst firing specifically in serotonergic neurons (Rouchet et al., 2008; Crespi, 2009). These effects of apamin persisted in the presence of GABAA and GABAB receptor blockers (Rouchet et al., 2008). Together with our work, these experiments suggest that serotonergic neurons would contribute disproportionately and be most sensitive to alterations in the expression of SK channels in this region.

3) In the Results section the authors claim that difference in excitability did not result from changes in GABA tone mentioning data not shown. It would be important to show these data since there is literature showing that stress affects excitability of GABA neurons but not 5-HT neurons (e.g. Challis et al., 2013).

We thank the reviewer for raising this important point. We have now included these data in Figure 1—figure supplement 1. Blockade of GABA-A and GABA-B receptors respectively with picrotoxin and CGP52432 did not attenuate the significant difference in the responsiveness of serotonin neurons between the single- and group-housed conditions. Note, it is essential to use picrotoxin instead of bicuculline for these experiments, since the latter has a known nonspecific effect of blocking SK channels (Khawaled et al., 1999).

4) When measuring AHP amplitudes and areas, at which frequency were the cells firing (Figure 1H)? The authors report the same current step for all cells (25pA); because neurons from single-housed are hypoexcitable we can infer that the frequency is lower. This would bias the measures. Measures of medium AHP are normally taken after a depolarizing current step has been applied, and selecting the current step showing the same number of action potentials for all cells.

To measure AHP amplitudes and areas, we selected the minimal depolarizing current step required to elicit action potentials in these cells. We calculated the AHPs from the first spike generated in response to the rheobase current. The frequency of firing at this point is similar between the groups (group-housed; 5.8 ± 0.2 Hz, single-housed; 5.4 ± 0.2 Hz, two-tailed t-test p = 0.2). Of note, we found the frequency differences between the groups in response to stronger depolarizing current steps (>250pA).

5) Because of the heterogeneity of the dorsal raphe neurons, the authors need to specify more precisely the localisation of their recordings.

We now included this in the Materials and methods section.

[Editors' note: the author responses to the re-review follow.]

[…] Essential revisions:

The requested revisions are minor and this should not need to go back for another round of review. The full text of the reviews is given below.

[…] Reviewer #2:

1) In their rebuttal, the authors explain the difference of their results with the Dankovski 2014 paper as a difference in protocol. However all the other supporting papers they cite have been done in rats (and they do not cite the Dankovski paper). If this is the first time the social separation model could be valuably transferred to mice, this would be worth mentioning more explicitly for future studies in mice.

We thank the reviewer for noticing the requirement of additional references for mouse studies on social isolation. We now included additional references in the Introduction.

2) In this regard, information about the mouse strain/gender of mice used in the different experiments is important; I could not find this info in the new version of the manuscript.

We used only male mice in our study. The gender and the background of our mice are included in the Methods section.

3) Results, subsection “Social isolation reduces excitability of 5-HT neurons”, first paragraph: explain what picrotoxin and CGP52432 are doing (as in the Methods).

We added this information to this subsection.

4) Results, subsection “Systemic apamin treatment normalizes anxiety- and depressive-like 160 behaviors in socially isolated mice”, last paragraph: to see how the z-score was calculated, I went to "Lin and Sibille 2015"; this was a waste of time as the Methods only refers to Guilloux 2011. Suggestion is to keep only the useful reference.

The Lin and Sibille 2015 reference is no longer included.

5) Conclusion - It is interesting that SK channels are a druggable compound for depression. Since this is not entirely new, as it had already been put forth and tested - to some extent - by Crespi in 2010, it may be useful to add a sentence as to why no one has really followed this up. Are there issues of toxicity? A need for more selective agents?

We thank the reviewer for raising this important point. We now included in the Discussion what additional steps/approaches are required for targeting SK channels for treatment of depression. Including these points also gave us a chance to explain the translational aspect of our study with the goal of finding better treatment strategies for depression and anxiety.

https://doi.org/10.7554/eLife.21416.012

Article and author information

Author details

  1. Derya Sargin

    Department of Physiology, University of Toronto, Toronto, Canada
    Contribution
    DS, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
    Competing interests
    The authors declare that no competing interests exist.
    ORCID icon 0000-0002-0253-5442
  2. David K Oliver

    Department of Physiology, University of Toronto, Toronto, Canada
    Contribution
    DKO, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
    Competing interests
    The authors declare that no competing interests exist.
    ORCID icon 0000-0003-1210-8409
  3. Evelyn K Lambe

    1. Department of Physiology, University of Toronto, Toronto, Canada
    2. Department of Obstetrics and Gynaecology, University of Toronto, Toronto, Canada
    3. Department of Psychiatry, University of Toronto, Toronto, Canada
    Contribution
    EKL, Conception and design, Analysis and interpretation of data, Drafting or revising the article
    For correspondence
    evelyn.lambe@utoronto.ca
    Competing interests
    The authors declare that no competing interests exist.
    ORCID icon 0000-0002-5994-6090

Funding

Natural Sciences and Engineering Research Council of Canada (Discovery Grant)

  • Evelyn K Lambe

Canada Research Chairs (Canada Research Chair in Developmental Cortical Physiology)

  • Evelyn K Lambe

Province of Ontario, Canada (Early Researcher Award)

  • Evelyn K Lambe

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

This work was funded by grants to EKL from the Canadian Natural Science and Engineering Research Council (NSERC), the Canada Research Chairs Program, and the Early Researcher Award from the Province of Ontario. We thank Ms. Lily Kang and Ms. Rhian Duke for expert technical assistance.

Ethics

Animal experimentation: This study was performed in accordance with the recommendations of the Canadian Council on Animal Care. All of the animals were handled according to approved institutional animal care and use guidelines under protocols approved by the Faculty of Medicine Animal Care and Use Committee at the University of Toronto. (20010374, 20011622, 20011733)

Reviewing Editor

  1. Sacha B Nelson, Reviewing Editor, Brandeis University, United States

Publication history

  1. Received: September 14, 2016
  2. Accepted: November 2, 2016
  3. Version of Record published: November 22, 2016 (version 1)

Copyright

© 2016, Sargin et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 1,213
    Page views
  • 254
    Downloads
  • 3
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.

Comments

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Download citations (links to download the citations from this article in formats compatible with various reference manager tools)

Open citations (links to open the citations from this article in various online reference manager services)

Further reading

    1. Computational and Systems Biology
    Daniel Scott Himmelstein et al.
    Research Article