Ketamine rapidly reverses stress-induced impairments in GABAergic transmission in the prefrontal cortex in male rodents

Journal Pre-proof Ketamine rapidly reverses stress-induced impairments in GABAergic transmission in the prefrontal cortex in male rodents

Sriparna Ghosal, Catharine H. Duman, Rong-Jian Liu, Min Wu, Rosemarie Terwilliger, Matthew J. Girgenti, Eric Wohleb, Manoela V. Fogaca, Emily M. Teichman, Brendan Hare, Ronald S. Duman PII:

S0969-9961(19)30344-4

DOI:

https://doi.org/10.1016/j.nbd.2019.104669

Reference:

YNBDI 104669

To appear in:

Neurobiology of Disease

Received date:

24 April 2019

Revised date:

29 October 2019

Accepted date:

5 November 2019

Please cite this article as: S. Ghosal, C.H. Duman, R.-J. Liu, et al., Ketamine rapidly reverses stress-induced impairments in GABAergic transmission in the prefrontal cortex in male rodents, Neurobiology of Disease(2019), https://doi.org/10.1016/j.nbd.2019.104669

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2019 Published by Elsevier.

Journal Pre-proof Article Type: Original Article

Ketamine rapidly reverses stress-induced impairments in GABAergic transmission in the prefrontal cortex in male rodents Sriparna Ghosal, Catharine H. Duman, Rong-Jian Liu, Min Wu, Rosemarie Terwilliger, Matthew J. Girgenti, Eric Wohleb, Manoela V. Fogaca, Emily M. Teichman, Brendan Hare, and Ronald S. Duman* [email protected] Departments of Psychiatry and Neurobiology, Yale University School of Medicine, 34 Park

Corresponding author at: 34 Park Street, New Haven, CT

oo

*

f

Street, New Haven, CT 06520

pr

Abstract

e-

Dysfunction of medial prefrontal cortex (mPFC) in association with imbalance of inhibitory and excitatory neurotransmission has been implicated in depression. However, the precise

Pr

cellular mechanisms underlying this imbalance, particularly for GABAergic transmission in

al

the mPFC, and the link with the rapid acting antidepressant ketamine remains poorly understood. Here we determined the influence of chronic unpredictable stress (CUS), an validated

model

of

rn

ethologically

depression,

on

synaptic

markers

of

GABA

Jo u

neurotransmission, and the influence of a single dose of ketamine on CUS-induced synaptic deficits in mPFC of male rodents. The results demonstrate that CUS decreases GABAergic proteins and the frequency of inhibitory post synaptic currents (IPSCs) of layer V mPFC pyramidal neurons, concomitant with depression-like behaviors. In contrast, a single dose of ketamine can reverse CUS-induced deficits of GABA markers, in conjunction with reversal of CUS-induced

depressive-like

behaviors.

These

findings

provide

further

evidence

of

impairments of GABAergic synapses as key determinants of depressive behavior and highlight ketamine-induced synaptic responses that restore GABA inhibitory, as well as glutamate neurotransmission.

1

Journal Pre-proof Introduction Depression remains a leading cause of disability worldwide (Kessler et al., 2003), yet our understanding of the neurobiology of this disorder and treatment response remains incomplete. Furthermore, the time lag of current medications to produce antidepressant effects (Greenberg et al., 2015), and the existence of high rates of non-responsive patients (Fournier et al., 2010) are major limitations that require the development of faster and more effective antidepressant drugs. The NMDA receptor antagonist ketamine produces rapid and sustained

oo

f

antidepressant actions in depressed patients, even in those considered treatment resistant (Berman et al., 2000). Comparable rapid and sustained antidepressants-like effects are

pr

observed in rodent models allowing for studies to determine the molecular and cellular

e-

mechanisms underlying the actions of ketamine (Autry et al., 2011; Li et al., 2010; Maeng et

Pr

al., 2008).

Human and rodent studies demonstrate that depression and chronic stress are

al

associated with decreased volume, neuronal atrophy, and reductions of excitatory synapse

rn

density in the medial prefrontal cortex (mPFC) (Duman et al., 2016; McEwen et al., 2015;

Jo u

Savitz and Drevets, 2009). Rodent studies further demonstrate a reduction in synaptic proteins, including GluR1, PSD95, and synapsin 1 in the mPFC (Li et al., 2011). Conversely, the antidepressant actions of ketamine in rodent models are associated with reversal of the deficits caused by stress and increased synapse number and function in the mPFC (Li, et al. 2010; Li, et al. 2011). Together, these studies are consistent with the hypothesis that functional and structural deficits of mPFC excitatory synapses act as a potential locus for depression pathology and that rapid agents like ketamine act in part via reversal of these deficits. In addition to disruption of excitatory-glutamate function, there is consistent evidence that cortical inhibitory GABAergic dysfunction contributes to the pathophysiology of

2

Journal Pre-proof depression and may serve as a target for novel therapeutic interventions (Luscher et al., 2011, 2015; Duman et al., 2019; Lener et al., 2017; Mohler, 2012; Banasr et al., 2017; Ghosal et al., 2017). Clinical studies have demonstrated reduced GABA levels in the brains of depressed individuals compared to healthy subjects (Sanacora et al., 1999; Luscher et al., 2011, 2015). Postmortem studies also report deficits of one of the major GABA interneuron subtypes, somatostatin (SST) expressing cells in the brains of depressed subjects (Douillard-Guilloux et al., 2017; Seney et al., 2015; Tripp et al., 2011), consistent with preclinical studies reporting a

oo

f

role of SST interneurons in rodent chronic stress models (Fee et al., 2017; Lin and Sibille, 2015; Prevot et al., 2017). Conversely, clinical studies have reported that normalization of

pr

GABA levels is associated with remission of depressive symptoms in response to treatment

Pr

2010; Luscher et al., 2011, 2015).

e-

with typical antidepressant therapies (Sanacora et al., 2002; Hasler et al., 2005;Shen et al.,

Animal models of stress show that behavioral phenotypes relevant to depression, such

al

as anhedonia and neophobia, are observed in GABA receptor mutant mice (Luscher and

rn

Fuchs, 2015) or in mutant mice with a 50% tissue reduction in hippocampal and cortical

Jo u

GABA (Kolata et al., 2018). In addition, recent electrophysiology studies report that ketamine enhances GABA synaptic function (Ren et al., 2016), possibly via disinhibition of glutamate signaling (Widman and McMahon, 2018). However, the molecular determinants underlying these effects and whether rapid acting antidepressants can reverse GABA deficits caused by chronic stress have not been determined. Here, we investigated the influence of chronic unpredictable stress (CUS) on GABA, as well as glutamate synaptic proteins and function within the mPFC and tested the influence of a single dose of ketamine on these synaptic deficits. Based on the evidence implicating SST interneurons in stress and depression we also conducted studies on this GABA interneuron subtype.

3

Journal Pre-proof Materials and methods Animals The following lines of mice bred in house on a C57B L/6 (Jackson Laboratories; C57BL/6J; #000664) background were used (males only) at 8-10 weeks of age: Thy1-eGFP (Thy1-eGFP (MJrs/J, #007788; Jackson Laboratories, glutamic acid decarboxylase-67 (Gad1) cre recombinase (Taylor et al., 2014), and somastostatin (Sst)-tdTomato (Wohleb et al., 2016). Male Sprague-Dawley rats (Charles River Laboratories, MA, USA) were used for

oo

f

electrophysiology studies of layer V mPFC neurons. All animals were allowed ad libitum access to food and water and maintained on 12-h light/dark cycle, unless otherwise noted. All

Pr

CUS exposure and drug administration

e-

pr

procedures were performed in accordance with Yale University and the NIH guidelines.

Animals underwent 21 days of CUS as previously described (Li et al., 2011). The CUS is a chronic stress paradigm that causes robust neurobiological changes that

al

well-studied

rn

contribute to the development of depressive-like behaviors (Li et al., 2011). CUS included

Jo u

random exposure to light overnight, cage tilt, wet bedding, rat odor, white noise, food deprivation, water deprivation, light off during day, isolation, restraint, and strobe light. CUS exposure generally occurred over 2-4 hour intervals in the morning and evening. Same sequence of stressors was followed for all animals included in this study. Control animals were handled daily, but otherwise left undisturbed in their home cage. On day 21, animals received a single acute dose of ketamine (10 m g/kg, i.p.) or saline (control); 24 hr after the drug

administration animals were subjected

recordings, or tissue collection.

Behavioral testing

4

to

behavioral testing,

electrophysiological

Journal Pre-proof Forced Swim Test (FST) was conducted as previously described (Ghosal et al., 2018). Mice were placed for 10 min in a clear cylinder filled with water (24±1C, 18 cm depth). Sessions were video-recorded and scored for total immobility time. Time immobile during the 2-6 minute block is reported. Novelty Suppressed Feeding Test (NSFT) was conducted as previously described (Ghosal et al., 2018). Mice were food-deprived for 18 h and placed in a dimly lit, novel environment (24×40×14 cm, fresh bedding) with food in the center. The latency to feed was recorded.

oo

f

Then, the amount of food consumption in home cages for 10 min was measured to verify motivation to feed.

pr

Locomotor activity (LMA) was measured using the Med-PC software (Med Associates, St

e-

Albans, VT). In LMA, each animal was placed individually in clean cages (30 cm × 19 cm ×

Pr

13 cm) for 30 min, during which time the number of beam breaks was measured as previously

rn

Western Blot

al

described (Ghosal et al., 2018).

Jo u

Western blots on crude synaptoneurosome or cell homogenates of mouse whole PFC were carried out as previously reported (Li et al., 2011). Primary antibodies included rabbit antigephyrin (Abcam, ab32206), rabbit anti-GAD 67 (Cell Signaling, 5305), rabbit anti-GAD 65 (Cell Signaling, 3988), rabbit anti-vesicular glutamate transporter 1 (Vglut1) (Cell Signaling, 12331), rabbit anti-vglut2 (Cell Signaling, 14487), rabbit anti-PSD95 (Cell Signaling, 2507), guinea pig anti-vesicular GABA transporter (VGAT) (Synaptic system, 131004), and rabbitanti GAPDH (Cell Signaling, 14C10). Densitometry was used to quantify protein bands using Image J Software (NIH) and proteins were normalized to GAPDH.

Immunofluorescence and Image Analysis

5

Journal Pre-proof Immunofluorescence labeling was performed on fixed brain tissues using standard techniques. Brains were collected from mice after transcardiac perfusion with saline and 4% paraformaldehyde (PFA). Brains were post-fixed in 4% PFA for 24 hours and incubated in 30% sucrose for an additional 48 hours. Fixed brains were frozen and sectioned at 25 um using a Microm HM550 cryostat. Free-floating sections were washed, then blocked for 1 hour at room temperature followed by incubation with primary antibody. Primary antibodies: guinea pig anti-VGAT (Synaptic system-131004), guinea pig anti-VGLUT 1 (Synaptic

oo

f

system-135304), rabbit anti-VGLUT 2 (Synaptic system-135403). Washed sections were then incubated with secondary antibodies (Alexa Fluor 568-conjugated goat anti-guinea pig IgGs

pr

#A21450 or Alexa Fluor 647-conjugated goat anti-rabbit IgGs #A11011 (Life Technologies).

e-

Sections were washed, mounted on glass slides and coverslipped with Prolong Gold Anti-fade

Pr

reagent (Life Technologies).

High-resolution images were acquired using an Olympus (FV1000) laser scanning

al

confocal microscope using a 60x oil-immersion objective (N.A. 1.42) with optimized pinhole

rn

diameter. Z-stack images (16-bit depth) were collected using 0.3 um step size, 3x digital

Laser intensity,

Jo u

zoom, 2 us/pixel scan speed, 800 x 800 image size and 0.088 um x 0.088 um pixel size in xy. photomultiplier amplification and offset were optimized for maximum

dynamic range with no saturated pixels and were held constant for all image collection. Image stacks were deconvolved (AutoQuant X v.3.0.1, Media Cybernetics) and quantitative analysis was performed on thresholded images in ImageJ. Images were binarized using a threshold that was determined in an initial survey of images, to optimize the detection of moderate to bright label intensity while maximizing the visual separation of object masks. For a given label, the resulting threshold value was held constant for the analysis of the entire image set. Watershed segmentation was applied in ImageJ and particle analysis was set to include labelled structures < 2.3 for VGAT and VGLUT1, and < 3.8 um2 for the larger VGLUT2 terminals,

6

Journal Pre-proof consistent with reported sizes for these axon terminal boutons in mouse cortex (Nakamura et al., 2005). Two to four images were collected per cortical layer in each tissue section and were averaged from 1-2 tissue sections to give a value per animal for each layer.

Brain slice preparation and electrophysiology For the analyses of layer V pyramidal cells, mPFC brain slices were prepared as described previously (Li et al., 2011) with minor modifications. During recording, cells were voltage

oo

f

clamped at -65 mV to simultaneously record inhibitory and excitatory postsynaptic currents (IPSCs and EPSCs). The PSC responses were tested in brain slices 24 hours following in vivo

pr

injection of saline or ketamine (10mg/kg). Sst-tdT cells were recorded as previously described

e-

(Wohleb et al., 2016). Recordings of the ChR2-positive GAD cells were performed at least

Pr

two weeks following virus injection as detailed in the supplementary materials.

al

Virus Injection and Optical Manipulation

rn

Bilateral infusions of purified double floxed AAV2-EF1a-DIO-hChR2(H134R)-eYFP or

Jo u

floxed AAV2-EF1a-DIO-Arch-eYFP viruses were made in Gad-Cre positive and littermate wildtype mice at AP: +1.9, ML –0.2, DV –2.7 with immediate placement of optical fibers (200 µm core, 0.22NA, Doric Lenses). Optical stimulations were for 60 min (473 nm laser: 15 ms pulse width, 5 Hz, 5 mW at the tip of the fiber, 1 min on and 1 min off for 30 cycles) or (590 nm laser: 5 Hz, 1-3 mW at the tip of the fiber, continuous for 60 min).

Statistical Analysis For experiments that included two groups, the results were analyzed using two-tailed Student’s t-tests or two-tailed Mann-Whitney U test. For experiments including more than two groups, the results were analyzed by ANOVA; in the event of a significant interaction,

7

Journal Pre-proof differences between group-means were determined by Bonferroni or Fisher’s LSD post hoc tests. Significance was determined at P < 0.05. Viral expression and implant placement were verified before mice were included in the analysis.

Results Ketamine and CUS induce opposite effects on GABAergic and glutamatergic synaptic proteins.

oo

f

Previous studies have demonstrated that a single dose of ketamine produces rapid and sustained antidepressant effects in rodent models consistent with the antidepressant effects in

pr

depressed patients (see Duman et al., 2016). These behavioral effects are associated with

e-

activity dependent, rapid and sustained increases in synapse number and function in the PFC

Pr

(Duman et al., 2016). In the current study we examined the influence of CUS and a single dose of ketamine on behavior and GABA and glutamate synaptic proteins in male mice 24 hr

al

after ketamine dosing. Behavioral studies were conducted to confirm a CUS-induced

rn

depressive phenotype. CUS significantly increased immobility in the FST (Figure 1b) and the

Jo u

latency to feed in a novel environment (Figure 1c). Moreover, a single dose of ketamine significantly decreased both the immobility time in the FST and latency to feed in the NSFT in control as well as CUS mice (Figure 1b, c). There were no changes in baseline locomotor activity measured on day 25 or in home cage feeding which was measured for 24 hours following the NSFT (Figure 1d, e). To identify possible abnormalities in neural inhibition linked to these behaviors, we measured GABAergic synapse-associated proteins in the PFC; levels of glutamate markers were also examined for comparison. Levels of vesicular GABA transporter (VGAT), GABA synthetic enzymes (GAD65 and GAD67), and GABA-A receptor associated postsynaptic protein (gephyrin) were determined by western blot of PFC synaptoneurosomes. CUS

8

Journal Pre-proof exposure significantly decreased levels of GAD67, and gephyrin, and this effect was reversed by a single dose of ketamine for both GAD67 and gephyrin (Figure 1g).

A significant main

effect of ketamine was observed for all of the GABAergic markers tested (Figure 1g). Analysis of glutamate markers included the two major vesicular transporters, VGLUT1 and VGLUT2, which play an important role in the uptake of glutamate into presynaptic vesicles (Fremeau et al., 2001), and the postsynaptic protein PSD95. In the CUS exposure significantly decreased levels of VGLUT1 and this effect was Ketamine also significantly increased levels of VGLUT2

f

reversed by ketamine (Figure 1h).

oo

present study,

and PSD95 (Figure 1h). We have previously reported the effect of CUS to decrease levels of

pr

GluR1 and PSD95 (Li et al., 2011).

e-

We also used immunohistochemical labeling for the vesicular transporters to examine

Pr

the effects of CUS and ketamine and obtained results consistent with the western blot findings. These studies were conducted in Thy1-GFP mice to allow visualization of the layer

al

5 pyramidal neurons, however we could not carry out our planned analysis of appositions of

rn

labeled terminals onto GFP-expressing pyramidal cells due to the unexpected low incidence

Jo u

of GFP-expressing neuronal structures in the puncta quantification zone (the more superficial areas of the tissue sections where antibody penetration was optimal). Low power images of VGAT, VGLUT1, and VGLUT2 in mPFC illustrate the expected distribution of labeled puncta as has been reported (Chaudhry et al., 1998; Fremeau et al., 2001; Fujiyama et al., 2001) (Figure 2a,c,e). VGAT labelling in mPFC was significantly decreased by CUS (Figure 2a’, b), and this effect was reversed by a single dose of ketamine (Figure 2a’, b). Confocal images demonstrate close apposition of VGAT immune-labelled puncta with somatic and dendritic elements of GFP-positive pyramidal cells in layer V (Figure 2g), illustrating the potential for a direct influence of treatment-regulated, GABA changes on pyramidal cells in this region.

9

Journal Pre-proof CUS exposure significantly decreased the density of VGLUT1 puncta in mPFC (layers I, and II/III) and ketamine treatment normalized these effects (Figure 2c’, d). The density of VGLUT2 puncta was decreased by CUS (layer V), and ketamine normalized this effect (Figure 2e’, f). Higher magnification images from layer I of mPFC show immuno-labelled puncta for VGLUT1 and VGLUT2 in close apposition with GFP-labelled dendrites (Figure 2h, upper and lower panels respectively). These sites of apparent contact suggest the possibility that ketamine and CUS effects on glutamatergic inputs to mPFC could be directly

oo

f

transmitted to layer V pyramidal cells; further studies are needed to confirm direct apposition onto dendrites. We also examined somatosensory cortex (S1) and found similar trends for

pr

CUS to decrease and ketamine to increase immuno-labelling for the vesicular transporters

e-

(Supplementary Figure 1), indicating that these effects may not be selective for mPFC. We

Pr

also examined the vesicular choline transporter (VCHAT) and found that CUS and ketamine had no effects on VCHAT puncta in layer I, II, and III, and actually increased levels in layer

al

V of mPFC, suggesting differential effects compared to what was observed with VGAT,

Jo u

rn

VGLUT1, and VGLUT2 (Supplementary Figure 2).

Ketamine and CUS-induce opposite cells.

effects on postsynaptic currents in mPFC pyramidal

We have reported that the frequency of 5-HT- and hypocretin-induced EPSCs in layer V pyramidal neurons is decreased by CUS and reversed by ketamine (Li et al., 2011), but have not examined IPSCs under these conditions. Moreover, previous studies recording IPSCs from PFC pyramidal cells report heterogeneous results (increased or decreased IPSCs), possibly as a result of different types of stressors and length of exposure, as well as differences in the mPFC target region (Czeh et al., 2018; McKlveen et al., 2016). As in prior studies (Li et al., 2011), we used whole cell patch with voltage clamp at -65 mV and recorded

10

Journal Pre-proof IPSCs, as well as EPSCs simultaneously in the same cell. About two-thirds of the pyramidal cells recorded had a large hyperpolarizing voltage sag resulting from the presence of an Ihcurrent that is indicative of type 1 neurons, while the remaining cells lacked the Ih-current and are referred to as type 2 neurons (Supplementary Figure 3a) (Lee et al., 2014). We found that CUS exposure significantly reduced the frequency of baseline spontaneous (sIPSCs), as well as hypocretin-induced IPSCs in type 1 neurons in layer V mPFC (Figure 3a,b); there was also a trend for a reduction in 5-HT-induced IPSCs. Similar trends were observed in type 2 cells

oo

f

although the effects of CUS did not reach significance (Supplementary Figure 3b,c) due to the smaller number of cells examined. Spontaneous IPSCs were unaltered when the two cell

pr

types were combined due to increased variability in their synaptic currents (data not shown).

e-

We also found that CUS significantly decreased the frequency of baseline sEPSCs, along with

Pr

5-HT- and hypocretin-induced EPSCs in type 1 layer V cells (Figure 3a,c), confirming our previous report (Li et al., 2011). In a subsequent experiment we examined the effect of CUS ketamine administration. Ketamine significantly increased 5-HT-induced

al

with or without

We also found that CUS decreased and a single dose of ketamine

Jo u

5-HT) (Figure 3d).

rn

IPSCs although the effect of CUS on IPSCs did not reach statistical significance (P =0.06 for

increased baseline, 5-HT- and hypocretin- induced EPSCs (Figure 3e), consistent with previous work (Li et al., 2011). There was no change in the mean amplitudes regardless of the cell type, CUS exposure or ketamine treatment (Supplementary Figure 4). To determine whether these effects were specific to pyramidal cells, we also recorded from SST interneurons. In SST-tdTomato reporter mice, we found that SST+ neurons are distributed throughout the mPFC with higher numbers in layers II/III and V compared to layer I (Figure 4a) as expected (Markram et al., 2004). We found that CUS exposure significantly decreased sIPSCs on SST GABA interneurons (Figure 4b, c, d). There was no significant

11

Journal Pre-proof effect of ketamine treatment (Figure 4e, f). There was no change in the mean amplitudes regardless of the CUS exposure or ketamine treatment (Supplementary Figure 5).

Optogenetic studies of GABA interneurons in the mPFC The above results demonstrate opposing effects of CUS and ketamine on indices of GABA neurotransmission, suggesting that reduction of GABA function could contribute to the behavioral deficits caused by CUS and conversely that induction of GABA function could

oo

f

underlie, in part the actions of ketamine. To directly test this hypothesis, the influence of inhibition or stimulation of GABA interneurons on antidepressant-related behaviors was

pr

examined using an optogenetic approach. For stimulation studies, mice received viral

e-

infusions of floxed ChR2-eYFP (AAV2-EF1a-DIO-hChR2(H134R)-eYFP) into the mPFC of

Pr

Gad1-Cre+ (Figure 5a, b) or littermate controls. Current clamp analysis of Gad1-ChR2-eYFP positive cells in mPFC slices shows that laser stimulation (1-50 Hz, 473 nm) causes high

al

fidelity activation of GAD interneurons, particularly in the range of 5-10 Hz (Figure 5c, d).

rn

Based on this information we chose to use 5 Hz stimulation for the in vivo studies. For

Jo u

behavioral experiments, ChR2-eYFP virus was infused and mice were implanted with bilateral fiber optic cannula in the mPFC dorsal to the virus infusion sites. Ketamine and vehicle treatment groups were included to test the additional hypothesis that a ketamine effect depends on a glutamate burst (secondary to blockade of NMDA receptors on GABA interneurons) that could be blocked by photostimulation of Gad+ interneurons. (Duman et al., 2016; Widman and McMohan, 2018). Based on this hypothesis, we predict that photostimulation of Gad+ interneurons would block the actions of ketamine. Ten minutes after vehicle or ketamine administration, the mice received laser stimulation (1 min on, 1 min off, 1 hr) (Figure 5e); this stimulation time was chosen based on our previous studies showing that 1 hr stimulation of Camk2a expressing pyramidal neurons in the mPFC is sufficient to

12

Journal Pre-proof produce an antidepressant behavioral response (Fuchikami et al., 2015). Behavior analysis was conducted 1 to 3 days after stimulation to avoid testing during the stimulation period and to determine if the effects are sustained. Results show significant main effects of ketamine and GAD1-Cre+ genotype in the FST and NSFT (Figure 5f,g). The significant ketamine x genotype interaction and post hoc comparisons in the NSFT demonstrated that this behavioral effect of ketamine was blocked in GAD1-Cre+ mice (Figure 5g). Next, we tested the influence of laser-inhibition on behavior by infusion of Arch-YFP

oo

f

(AAV2-EF1a-DIO-Arch-eYFP) bilaterally into the mPFC of Gad1-Cre+ or control mice followed by implantation of bilateral optic fibers (Figure 5h). Mice were then subjected to

pr

laser stimulation (1 hour continuously), and behavior analyzed 24 hr later (Figure 5i-j). There

e-

was no effect of GABA interneuron inhibition under these conditions on immobility in the

Pr

FST or latency to feed in the NSFT in the Gad1-Cre+ mice (Figure 5i, j).

al

Discussion

rn

In this study, we identify specific CUS-induced deficits of GABA neurotransmission

Jo u

in the mPFC that are associated with alterations of neuronal function and show that a single dose of ketamine rapidly reverses these deficits as well as associated behavioral changes. These findings extend previous studies, describing deficits of GABA interneurons that could contribute to the pathophysiology of depression (Luscher et al., 2011, 2015), and importantly demonstrate that ketamine rapidly increases GABA function in naïve mice and can reverse CUS-induced deficits of GABA neurotransmission in mPFC. CUS resulted in depression-like behaviors that were accompanied by reductions in levels of

vesicular GABA transporter (VGAT), synthetic enzyme (GAD67), and the

postsynaptic protein (gephyrin) in the mPFC (Figure 6). Reduction of GABA neurotransmitter function by CUS was also demonstrated by decreased frequency of basal and hypocretin-

13

Journal Pre-proof induced IPSCs in mPFC layer V pyramidal neurons. The recorded cells were primarily Type 1 pyramidal neurons, but similar trends were observed in Type 2 cells. These neurons differ in their morphology, projection patterns and response to neurotransmitters, and are thought to represent functionally distinct pyramidal cell subtypes in layer 5 (Lee et al., 2014). In addition, decreased frequency of IPSCs in SST-tdT+ interneurons in the mPFC provided further evidence of GABA deficits following CUS exposure. These findings are consistent with previous reports (Banasr et al., 2017; Ghosal et al., 2017, Ren et al., 2016) and support

oo

f

the hypothesis that impairment of GABA neurotransmission represents a key determinant of chronic stress exposure and are consistent with GABA deficits reported in depression

pr

(Luscher et al., 2011, 2015).

e-

The results also demonstrate that ketamine reverses CUS-induced GABAergic deficits

Pr

in conjunction with antidepressant behavioral responses. A single dose of ketamine increased GAD67 and gephyrin in naïve animals and rapidly reversed CUS-induced deficits in GAD67,

al

VGAT, and gephyrin (Figure 6). These changes were accompanied by increased GABA

rn

function indicated by increased 5-HT-induced IPSCs in mPFC layer V pyramidal neurons

Jo u

after ketamine treatment. Alterations of both pre- (GAD67, and VGAT) and postsynaptic (gephyrin) GABAergic proteins indicate that ketamine-enhancement of 5-HT -induced IPSCs could result from increased function of pre- and/or postsynaptic elements. The fact that sPSC amplitudes were unchanged by CUS and ketamine for both GABAergic and glutamatergic synapses, suggests that changes in gephyrin expression were secondary to presynaptic changes and representative of altered synapse numbers rather than altered strength of individual synapses,

although future studies are needed to directly test this hypothesis by

recording miniature EPSCS and IPSCs. We did not observe significant effects of ketamine on SST-tdT+

interneuron function, although further studies are needed to examine the

interactions between ketamine and CUS.

14

Journal Pre-proof The mechanisms underlying the up-regulation of GABA function by ketamine are unclear. One possibility is that increased GABA function is in part a compensatory induction of inhibitory neurotransmission in response to the initial ketamine-induced burst of glutamate and increased glutamate synaptic function. However, it is also possible that ketamine has direct effects on GABA function via an unknown mechanism that is independent of the increase in glutamate function. Ketamine is also reported to rescue gephyrin and IPSC deficits in mutant mice that are heterozygous for the GABA-A γ2-subunit receptor (Ren et al., 2016).

oo

f

The results are consistent with the hypothesis that increased GABA function in the mPFC contributes to the antidepressant actions of ketamine. This could include rapid antidepressant

pr

actions that are observed within hours and/or the sustained (1-week) effects of ketamine

e-

observed in patients and in rodent models. The possibility that increased GABA function is

Pr

sufficient to produce an antidepressant response has been demonstrated in mice with heterozygous deletion of the GABA-A γ2-subunit from SST interneurons, which reduces

rn

2017).

al

inhibitory control of these cells and results in antidepressant-like responses (Fuchs et al.,

Jo u

A strength of this study is the data demonstrating that chronic stress and ketamine dynamically regulate the glutamate, as well as GABA neurotransmitter systems in the same animals. These findings are consistent with previous studies of PSD95 (Li et al., 2011), but also extend this work by demonstrating regulation of VGLUT1 and VGLUT2 in mPFC (Figure 6). Neurons expressing VGLUT1 and VGLUT2 and the punctate terminal labeling for these transporters show distinct distributions that are largely non-overlapping. In general, VGLUT1-positive terminals originate from cortical cells and VGLUT2-positive terminals from thalamic projection neurons. Punctate and fiber immunoreactivity for VGLUT1 is relatively evenly distributed through cortical layers whereas immunoreactivity for VGLUT2 occurs more heterogeneously. Our results demonstrating that both VGLUT1 and VGLUT2

15

Journal Pre-proof are decreased by CUS and increased by ketamine indicate that both cortical and thalamic afferents to PFC are altered by these treatments. The results of our electrophysiological studies are also consistent with these findings, as 5-HT and hypocretin-induced EPSCs are associated with cortical-cortical, and thalamo-cortical projections respectively and we found both to be altered by CUS and ketamine (Liu and Aghajanian, 2008). Increased glutamate synaptic function opposes the neuronal atrophy caused by chronic stress and is associated with the rapid antidepressant actions of ketamine (Li et al., 2010,

oo

f

2011). These effects of ketamine are dependent on increased neuronal activity and BDNF release (Maeng et al., 2008; Autry et al., 2011) resulting from a rapid, transient burst of

pr

glutamate (Moghaddam et al., 1997). The burst of glutamate is thought to occur via blockade

e-

of NMDA receptors on GABA interneurons, as these cells are tonic firing and are more

Pr

sensitive to the open channel blocking actions of ketamine (Homayoun and Moghaddam, 2007) (Figure 6). This is supported by evidence that ketamine administration leads to

al

decreased spontaneous firing of GABA interneurons in the PFC and a delayed increase in the

rn

firing rate of pyramidal cells (Homayoun and Moghaddam, 2007). However, there is also

Jo u

evidence that the ketamine metabolite, (2R,6R)-hydroxynorketamine produces antidepressant actions that are independent of NMDA receptor blockade and that are mediated by a mechanism similar to blockade of mGluR2 autoreceptors (Zanos et al., 2016, 2019). In either case, stimulation of Gad1+ interneurons in the mPFC during a 1 hr period after ketamine administration, when glutamate is elevated (Moghaddam et al., 1997), was able to completely block the behavioral effects of ketamine in the NSFT. These findings indicate that increased glutamate transmission, which would be blocked by optogenetic stimulation of GABA interneurons, is required for the antidepressant behavioral actions of ketamine, consistent with the disinhibition hypothesis that the initial cellular trigger underlying the actions of ketamine is a transient blockade of NMDA receptors on GABA interneurons in the mPFC (Figure 6).

16

Journal Pre-proof Although

ketamine

suppression

of GABA

neurotransmission

appears

to

be

contradictory to evidence from postmortem depressed subjects and chronic stress in rodents that GABA hypofunction contributes to depressive behavior (Ghosal et al., 2017), this is a transient blockade and glutamate burst (~1 hr) by ketamine that drives synaptic plasticity of principle neurons. Based on the results of the current study, we propose that ketamine, either via the transient glutamate burst and/or a direct effect on GABA interneurons also enhances GABA neurotransmitter plasticity and function, in part via recruitment of SST interneurons.

oo

f

The possibility that enhanced GABA neurotransmitter function could contribute to the antidepressant behavioral actions of ketamine is supported by a study showing that

pr

potentiation of SST interneuron function via deletion of GABA-A γ2-subunit produces

e-

anxiolytic and antidepressant-like responses (Fuchs et al., 2017). Together, these findings

Pr

raise the possibility that ketamine administration, possibly via disinhibition of glutamate transmission causes an adaptive response that restores GABA inhibitory function, as well as

al

glutamate neuronal actions.

rn

Optogenetic studies to test the influence of GABA interneuron stimulation on

Jo u

depression-like behaviors were inconclusive (nonequivalent results for the FST and NSFT), while the results with optogenetic inhibition were negative. Gad1+ interneurons in mPFC were either stimulated or inhibited and behavioral tests were initiated the next day to avoid the acute effects of cell manipulation; this paradigm is based on our previous report that 1 hr stimulation (1 min on, 1 min off) of mPFC pyramidal neurons is sufficient to produce sustained antidepressant actions (Fuchikami et al., 2015). The 1 hr time point was also chosen to avoid possible tissue damage caused by more long-term laser exposure, although it is possible that the 1 hr paradigm is too short, and that more sustained or repeated stimulation or inhibition paradigms are needed to induce sustained alterations of GABA function and behavior. It is also possible that targeting a specific subtype of GABA interneuron, notably

17

Journal Pre-proof the SST using Sst-Cre recombinase mice could result in significant effects, as was shown for SST specific GABA-A γ2-subunit deletion mutant mice (Fuchs et al., 2017). Another study using a DREADD approach reported that acute inhibition of SST interneurons in the mPFC caused anxiogenic responses in the elevated plus maze and NSFT, while repeated, daily inhibition of SST interneurons (3 weeks) resulted in anxiolytic effects in these tests (Soumier and Sibille, 2014). However, behavioral testing in these experiments was conducted during the time when GABA neurons were being inhibited, which would complicate the outcome

oo

f

and interpretation of these studies. We conducted preliminary DREADD studies to manipulate Gad1+ interneurons but so far the results have been inconsistent, and additional

pr

studies are needed to further investigate this question.

e-

GABA neurotransmission plays a crucial role in assembling microcircuitry during

Pr

development and in orchestrating cortical activity in adulthood, and the results of the current study indicate that CUS and ketamine impact the overall function of the mPFC. Dendrite-

al

targeting SST interneurons gate excitatory inputs from converging intracortical and thalamic

rn

signals and maintain activated and baseline pyramidal neuronal output patterns (Gentet et al.,

Jo u

2012). SST interneurons also target and provide inhibitory control over the parvalbumin neurons that send projections to the soma of principal neurons, thereby controlling spike output. Hence, reduced SST/GABA function might result in decreased inhibitory gating of excitatory inputs to the dendritic arbor of pyramidal neurons, but could also result in reduced inhibitory control of parvalbumin neurons that target glutamatergic cell bodies, resulting in increased inhibition and decreased output of principal neurons. The net consequences would be imbalance of input output activity and altered cortical control of behavioral responses. However, it is possible that CUS exposure also decreases the function of parvalbumin interneurons, with opposing effects of ketamine. In any case, our results suggest that up-regulation of GABA markers and reversal of

18

Journal Pre-proof CUS-induced GABA neurotransmitter deficits could contribute to the sustained antidepressant response to ketamine (Figure 6). Further characterization of the neurotransmitter receptors that control GABA function could lead to new insights for targeting GABA interneurons that will provide alternative and more efficacious therapeutic strategies for the treatment of depression. In addition, future studies are needed to determine if chronic stress causes sexspecific effects on GABA function (e.g., greater loss of GABA function in females compared to males) that could contribute to the increased incidence of depression in women compared

oo

f

to men. Together, these studies could identify novel therapeutic targets for the the

pr

development of more efficacious antidepressant treatments for men and women.

e-

Funding and disclosure:

Pr

R.S.D. has received consulting fees from Taisho, Johnson & Johnson, and Naurex, and grant support from Taisho, Johnson & Johnson, Naurex, Allergan, Navitor, Lundbeck, and Lilly.

al

All other authors declare no biomedical financial interests and potential conflicts of interest.

rn

None of the above-listed companies or funding agencies had any influence on the content of

Jo u

this article.

Acknowledgements:

This research was supported by NIMH Grants MH045481 and MH093897 to R.S.D and the State of CT. We would like to thank Xiao-Yuan Li for technical assistance.

References: 1. Autry, A.E., Adachi, M., Nosyreva, E., Na, E.S., Los, M.F., Cheng, P.F., Kavalali, E.T., Monteggia, L.M., 2011. NMDA receptor blockade at rest triggers rapid behavioural antidepressant responses. Nature 475, 91-U109. 19

Journal Pre-proof 2. Banasr, M., Lepack, A., Fee, C., Duric, V., Maldonado-Aviles, J., DiLeone, R., Sibille, E., Duman, R.S., Sanacora, G., 2017. Characterization of GABAergic marker expression in the chronic unpredictable stress model of depression. Chronic stress 1. 3. Berman, R.M., Cappiello, A., Anand, A., Oren, D.A., Heninger, G.R., Charney, D.S., Krystal, J.H., 2000. Antidepressant effects of ketamine in depressed patients. Biological psychiatry 47, 351-354. 4. Chaudhry, F.A., Reimer, R.J., Bellocchio, E.E., Danbolt, N.C., Osen, K.K., Edwards,

oo

f

R.H., Storm-Mathisen, J., 1998. The vesicular GABA transporter, VGAT, localizes to synaptic vesicles in sets of glycinergic as well as GABAergic neurons. J Neurosci. 18,

pr

9733-9750.

e-

5. Czeh, B., Vardya, I., Varga, Z., Febbraro, F., Csabai, D., Martis, L.S., Hojgaard, K.,

Pr

Henningsen, K., Bouzinova, E.V., Miseta, A., Jensen, K., Wiborg, O., 2018. LongTerm Stress Disrupts the Structural and Functional Integrity of GABAergic Neuronal

al

Networks in the Medial Prefrontal Cortex of Rats. Frontiers in cellular neuroscience

rn

12, 148.

Jo u

6. Douillard-Guilloux, G., Lewis, D., Seney, M.L., Sibille, E., 2017. Decrease in somatostatin-positive cell density in the amygdala of females with major depression. Depression and anxiety 34, 68-78. 7. Duman, R.S., Aghajanian, G.K., Sanacora, G., Krystal, J.H., 2016. Synaptic plasticity and depression: new insights from stress and rapid-acting antidepressants. Nature medicine 22, 238-249. 8. Duman, R.S., Sanacora, G., Krystal, J.H., 2019. Altered Connectivity in Depression: GABA and Glutamate Neurotransmitter Deficits and Reversal by Novel Treatments. Neuron 102, 75-90.

20

Journal Pre-proof 9. Fee, C., Banasr, M., Sibille, E., 2017. Somatostatin-Positive Gamma-Aminobutyric Acid Interneuron Deficits in Depression: Cortical Microcircuit and Therapeutic Perspectives. Biological psychiatry 82, 549-559. 10. Fournier, J.C., DeRubeis, R.J., Hollon, S.D., Dimidjian, S., Amsterdam, J.D., Shelton, R.C., Fawcett, J., 2010. Antidepressant drug effects and depression severity: a patientlevel meta-analysis. Jama 303, 47-53. 11. Fremeau, R.T., Jr., Troyer, M.D., Pahner, I., Nygaard, G.O., Tran, C.H., Reimer, R.J.,

oo

f

Bellocchio, E.E., Fortin, D., Storm-Mathisen, J., Edwards, R.H., 2001. The expression of vesicular glutamate transporters defines two classes of excitatory synapse. Neuron

pr

31, 247-260.

e-

12. Fuchikami, M., Thomas, A., Liu, R., Wohleb, E.S., Land, B.B., DiLeone, R.J.,

Pr

Aghajanian, G.K., Duman, R.S., 2015. Optogenetic stimulation of infralimbic PFC reproduces ketamine's rapid and sustained antidepressant actions. Proceedings of the

al

National Academy of Sciences of the United States of America 112, 8106-8111.

rn

13. Fuchs, T., Jefferson, S.J., Hooper, A., Yee, P.H., Maguire, J., Luscher, B., 2017.

Jo u

Disinhibition of somatostatin-positive GABAergic interneurons results in an anxiolytic and antidepressant-like brain state. Molecular psychiatry 22, 920-930. 14. Fujiyama, F., Furuta, T., Kaneko, T., 2001. Immunocytochemical localization of candidates for vesicular glutamate transporters in the rat cerebral cortex. The Journal of comparative neurology 435, 379-387. 15. Gentet, L.J., Kremer, Y., Taniguchi, H., Huang, Z.J., Staiger, J.F., Petersen, C.C., 2012. Unique functional properties of somatostatin-expressing GABAergic neurons in mouse barrel cortex. Nature neuroscience 15, 607-612. 16. Ghosal, S., Bang, E., Yue, W.Z., Hare, B.D., Lepack, A.E., Girgenti, M.J., Duman, R.S., 2018. Activity-Dependent Brain-Derived Neurotrophic Factor Release Is

21

Journal Pre-proof Required for the Rapid Antidepressant Actions of Scopolamine. Biological psychiatry 83, 29-37. 17. Ghosal, S., Hare, B., Duman, R.S., 2017. Prefrontal Cortex GABAergic Deficits and Circuit Dysfunction in the Pathophysiology and Treatment of Chronic Stress and Depression. Current opinion in behavioral sciences 14, 1-8. 18. Greenberg, P.E., Fournier, A.A., Sisitsky, T., Pike, C.T., Kessler, R.C., 2015. The economic burden of adults with major depressive disorder in the United States (2005

oo

f

and 2010). The Journal of clinical psychiatry 76, 155-162.

19. Hasler, G., Neumeister, A., van der Veen, J.W., Tumonis, T., Bain, E.E., Shen, J.,

pr

Drevets, W.C., Charney, D.S., 2005. Normal prefrontal gamma-aminobutyric acid

e-

levels in remitted depressed subjects determined by proton magnetic resonance

Pr

spectroscopy. Biological psychiatry 58, 969-973. 20. Homayoun, H., Moghaddam, B., 2007. NMDA receptor hypofunction produces

al

opposite effects on prefrontal cortex interneurons and pyramidal neurons. The Journal

rn

of neuroscience : the official journal of the Society for Neuroscience 27, 11496-11500.

Jo u

21. Kessler, R.C., Berglund, P., Demler, O., Jin, R., Koretz, D., Merikangas, K.R., Rush, A.J., Walters, E.E., Wang, P.S., National Comorbidity Survey, R., 2003. The epidemiology of major depressive disorder: results from the National Comorbidity Survey Replication (NCS-R). Jama 289, 3095-3105. 22. Kolata, S.M., Nakao, K., Jeevakumar, V., Farmer-Alroth, E.L., Fujita, Y., Bartley, A.F., Jiang, S.Z., Rompala, G.R., Sorge, R.E., Jimenez, D.V., Martinowich, K., Mateo, Y., Hashimoto, K., Dobrunz, L.E., Nakazawa, K., 2018. Neuropsychiatric Phenotypes Produced by GABA Reduction in Mouse Cortex and Hippocampus. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 43, 1445-1456.

22

Journal Pre-proof 23. Lee, A.T., Gee, S.M., Vogt, D., Patel, T., Rubenstein, J.L., Sohal, V.S., 2014. Pyramidal neurons in prefrontal cortex receive subtype-specific forms of excitation and inhibition. Neuron 81, 61-68. 24. Lener, M.S., Niciu, M.J., Ballard, E.D., Park, M., Park, L.T., Nugent, A.C., Zarate, C.A., Jr., 2017. Glutamate and Gamma-Aminobutyric Acid Systems in the Pathophysiology of Major Depression and Antidepressant Response to Ketamine. Biological psychiatry 81, 886-897.

oo

f

25. Li, N., Lee, B., Liu, R.J., Banasr, M., Dwyer, J.M., Iwata, M., Li, X.Y., Aghajanian, G., Duman, R.S., 2010. mTOR-dependent synapse formation underlies the rapid

pr

antidepressant effects of NMDA antagonists. Science 329, 959-964.

e-

26. Li, N., Liu, R.J., Dwyer, J.M., Banasr, M., Lee, B., Son, H., Li, X.Y., Aghajanian, G.,

Pr

Duman, R.S., 2011. Glutamate N-methyl-D-aspartate receptor antagonists rapidly reverse behavioral and synaptic deficits caused by chronic stress exposure. Biological

al

psychiatry 69, 754-761.

rn

27. Lin, L.C., Sibille, E., 2015. Somatostatin, neuronal vulnerability and behavioral

Jo u

emotionality. Molecular psychiatry 20, 377-387. 28. Liu, R.J., Aghajanian, G.K., 2008. Stress blunts serotonin- and hypocretin-evoked EPSCs in prefrontal cortex: role of corticosterone-mediated apical dendritic atrophy. Proceedings of the National Academy of Sciences of the United States of America 105, 359-364. 29. Luscher, B., Fuchs, T., 2015. GABAergic control of depression-related brain states. Advances in pharmacology 73, 97-144. 30. Luscher B, Shen Q, Sahir N (2011): The GABAergic deficit hypothesis of major depressive disorder. Mol Psychiatry 16:383–406.

23

Journal Pre-proof 31. Maeng, S., Zarate, C.A., Jr., Du, J., Schloesser, R.J., McCammon, J., Chen, G., Manji, H.K., 2008. Cellular mechanisms underlying the antidepressant effects of ketamine: role of alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid receptors. Biological psychiatry 63, 349-352. 32. Markram, H., Toledo-Rodriguez, M., Wang, Y., Gupta, A., Silberberg, G., Wu, C., 2004. Interneurons of the neocortical inhibitory system. Nature reviews. Neuroscience 5, 793-807.

oo

f

33. McEwen, B.S., Bowles, N.P., Gray, J.D., Hill, M.N., Hunter, R.G., Karatsoreos, I.N., Nasca, C., 2015. Mechanisms of stress in the brain. Nature neuroscience 18, 1353-

pr

1363.

e-

34. McKlveen, J.M., Morano, R.L., Fitzgerald, M., Zoubovsky, S., Cassella, S.N.,

Pr

Scheimann, J.R., Ghosal, S., Mahbod, P., Packard, B.A., Myers, B., Baccei, M.L., Herman, J.P., 2016. Chronic Stress Increases Prefrontal Inhibition: A Mechanism for

al

Stress-Induced Prefrontal Dysfunction. Biological psychiatry 80, 754-764.

rn

35. Moghaddam, B., Adams, B., Verma, A., Daly, D., 1997. Activation of glutamatergic

Jo u

neurotransmission by ketamine: a novel step in the pathway from NMDA receptor blockade to dopaminergic and cognitive disruptions associated with the prefrontal cortex. The Journal of neuroscience : the official journal of the Society for Neuroscience 17, 2921-2927. 36. Mohler, H., 2012. The GABA system in anxiety and depression and its therapeutic potential. Neuropharmacology 62, 42-53. 37. Nakamura, K., Hioki, H., Fujiyama, F., Kaneko, T., 2005. Postnatal changes of vesicular glutamate transporter (VGluT)1 and VGluT2 immunoreactivities and their colocalization in the mouse forebrain. Journal of Comparative Neurology 492, 263-288. 24

Journal Pre-proof 38. Prevot, T.D., Gastambide, F., Viollet, C., Henkous, N., Martel, G., Epelbaum, J., Beracochea, D., Guillou, J.L., 2017. Roles of Hippocampal Somatostatin Receptor Subtypes in Stress Response and Emotionality. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 42, 1647-1656. 39. Ren, Z., Pribiag, H., Jefferson, S.J., Shorey, M., Fuchs, T., Stellwagen, D., Luscher, B., 2016. Bidirectional Homeostatic Regulation of a Depression-Related Brain State by Gamma-Aminobutyric Acidergic Deficits and Ketamine Treatment. Biological

oo

f

psychiatry 80, 457-468.

40. Sanacora, G., Mason, G.F., Rothman, D.L., Behar, K.L., Hyder, F., Petroff, O.A.,

pr

Berman, R.M., Charney, D.S., Krystal, J.H., 1999. Reduced cortical gamma-

e-

aminobutyric acid levels in depressed patients determined by proton magnetic

Pr

resonance spectroscopy. Archives of general psychiatry 56, 1043-1047. 41. Sanacora G, Mason GF, Rothman DL, Krystal JH. Increased occipital cortex GABA

al

concentrations in depressed patients after therapy with selective serotonin reuptake

rn

inhibitors. Am J Psychiatry 2002; 159: 663–665.

Jo u

42. Savitz, J., Drevets, W.C., 2009. Bipolar and major depressive disorder: neuroimaging the developmental-degenerative divide. Neuroscience and biobehavioral reviews 33, 699-771.

43. Seney, M.L., Tripp, A., McCune, S., Lewis, D.A., Sibille, E., 2015. Laminar and cellular analyses of reduced somatostatin gene expression in the subgenual anterior cingulate cortex in major depression. Neurobiology of disease 73, 213-219. 44. Shen Q, Lal R, Luellen BA, Earnheart JC, Andrews AM, Luscher B. gammaAminobutyric acid-type A receptor deficits cause hypothalamic-pituitary-adrenal axis hyperactivity and antidepressant drug sensitivity reminiscent of melancholic forms of depression. Biol Psychiatry 2010; 68: 512–520.

25

Journal Pre-proof 45. Soumier, A., Sibille, E., 2014. Opposing effects of acute versus chronic blockade of frontal cortex somatostatin-positive inhibitory neurons on behavioral emotionality in mice. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 39, 2252-2262. 46. Taylor, S.R., Badurek, S., Dileone, R.J., Nashmi, R., Minichiello, L., Picciotto, M.R., 2014. GABAergic and glutamatergic efferents of the mouse ventral tegmental area. The Journal of comparative neurology 522, 3308-3334.

oo

f

47. Tripp, A., Kota, R.S., Lewis, D.A., Sibille, E., 2011. Reduced somatostatin in subgenual anterior cingulate cortex in major depression. Neurobiology of disease 42,

pr

116-124.

e-

48. Widman, A.J., McMahon, L.L., 2018. Disinhibition of CA1 pyramidal cells by low-

Pr

dose ketamine and other antagonists with rapid antidepressant efficacy. Proceedings of the National Academy of Sciences of the United States of America 115, E3007-

al

E3016.

rn

49. Wohleb, E.S., Wu, M., Gerhard, D.M., Taylor, S.R., Picciotto, M.R., Alreja, M.,

Jo u

Duman, R.S., 2016. GABA interneurons mediate the rapid antidepressant- like effects of scopolamine. The Journal of clinical investigation 126, 2482-2494.

26

Journal Pre-proof Figure Legends

Figure 1. Ketamine normalizes CUS-induced deficits in behavior and GABAergic proteins. (a) Male Thy1-eGFP-mice were exposed to 21 days of CUS, followed by ketamine (10mg/kg, ip) or saline (control) administration on day 21 and then screened for depression-like behaviors. (b) CUS exposure significantly increased the immobility duration in the FST and ketamine administration significantly decreased the FST immobility duration [** main effect

f

of stress (F1,22 = 4.80, P < 0.05), ## main effect of ketamine (F1,22 = 40.45, P < 0.001)] (bars

oo

indicate mean ± s.e.m.; number of animals (n) = 6-7/ group.. (c) CUS exposure increased the

pr

latency to feed in the NSFT, and ketamine administration decreased the latency to feed in the

e-

NSFT [** main effect of stress (F1,22 = 3.15, P < 0.05), ## main effect of ketamine (F1,22 = 18.93, P < 0.001)] (mean ±  s.e.m.; number of animals (n) = 6-7/ group. (d) Home cage

Pr

feeding showed no differences regardless of stress exposure or ketamine treatment (mean ± s.e.m.; number of animals (n) = 6-7/ group). (e) Basal locomotor activity was not

al

significantly different in the stress exposed or ketamine administered mice (mean ±  s.e.m.;

rn

number of animals (n) = 6-7/ group). (f) Experimental timeline of CUS procedure, ketamine

Jo u

administration, and tissue collection of whole PFC.

(g) Western blot analyses of pre- and

post- synaptic protein fold change related to GABAergic transmission revealed significant changes in the PFC synaptosomes of CUS mice and ketamine-treated mice. Representative blots are shown above the graphs. CUS exposure significantly decreased levels of GAD67 and gephyrin, and ketamine reversed this effect [GAD 67: Significant stress x ketamine interaction (F1,15 = 6.25, P <0.05), *P < 0.05 vs. control, # P < 0.05 vs. CUS

post-hoc

Bonferronis test;.significant main effect of ketamine (F1,15 = 26.79, P < 0.05); for gephyrin: Significant stress x ketamine interaction (F1,15 = 14.42, P <0.05) *P < 0.05 vs. control, #

P < 0.05 vs. CUS

post-hoc Bonferronis test; significant main effect of stress (F1,15 = 5.55,

P < 0.05) significant main effect of ketamine (F1,15 = 47.08, P < 0.05)]. 27

## Significant main

Journal Pre-proof effect of ketamine for GAD 65 (F1,15 = 13.59, P <0.05), and VGAT (F1,15 = 9.80, P <0.05). (h) Western blot analyses of pre- and post- synaptic proteins related to glutamatergic neurotransmission also revealed significant changes in the PFC of CUS and ketaminetreatment mice. Representative blots are shown above the graphs.

Levels of VGLUT1 were

significantly decreased by CUS and ketamine reversed the cus-induced decrease [Significant stress X ketamine interaction (F1,15 = 11.71, P <0.05), post-hoc Bonferroni test, *P < 0.05 vs. control, and # P < 0.05 vs. CUS]. ## Significant main effect of ketamine to increase VGLUT1

oo

f

(F1,15 = 27.05, P <0.05), VGLUT2 (F1,15 = 9.40, *P <0.05), and PSD95 (F1,15 = 14.86, *P

pr

<0.05). Data represent mean ±  s.e.m.; number of animals (n) = 5-7/ group.

e-

Figure 2. Immunofluorescence analysis of GABA and glutamate markers in mPFC: effects of

Pr

CUS and ketamine. (a) Upper panel: Low power images showing immunofluorescence labelling for VGAT in PFC in (control) Thy1-GFP-expressing mice. Locations of sampling (a’) Lower panel: Higher magnification images

al

for cortical layers I, II/II and V are shown.

(b) VGAT

rn

showing representative VGAT expression for each of the treatment groups.

Jo u

labeling in mPFC as determined by immunofluorescence, was decreased by CUS and this effect was reversed by a single ketamine administration.

[(Significant main effect of

ketamine: layer II/III (F1,19 = 11.45, P <0.05), layer V (F1,19 = 7.56, P <0.05); Significant stress X ketamine interaction: layer I (F 1,19 = 11.77, P <0.05), layer II/III (F1,19 = 11.45, P <0.05), layer V (F1,19 = 12.5, P <0.05). * P < 0.05 significantly different from control group; # P < 0.05 significantly different from CUS group, post-hoc Fischer’s LSD. Sal n = 6; ketamine n = 6; CUS n = 5; CUS+ketamine n = 6]. (c) Upper panel: Low power images showing distribution of VGLUT1 in mPFC. Locations of sampling for cortical layers I, II/II and V in PL PFC are shown. (c’) Lower panel: Higher magnification images showing representative VGLUT1 expression for each treatment group. (d) VGLUT1 expression was

28

Journal Pre-proof significantly decreased by CUS and this effect was reversed by ketamine. [Significant main effect of ketamine: layer I (F1,20 = 12.85, P <0.05), layer II/III (F1,19 = 4.30, P <0.05); Significant main effect of stress: layer I (F1,20 = 9.42, P <0.05), layer II/III (F1,19 = 5.74, P <0.05); Significant stress X ketamine interaction: layer I (F 1,20 = 21.69, P <0.05), layer II/III (F1,19 = 12.12, P <0.05). *P < 0.05, significantly different from control group; # P < 0.05, significantly different from CUS group, post-hoc Fischer’s LSD. Sal n = 6; ketamine n = 6; CUS n = 5-6; CUS+ketamine n = 6]. (e) Upper panel: Low power images showing

oo

f

distribution of immunofluorescence for VGLUT2 in mPFC. Locations of sampling for cortical layers I, II/II and V are shown. (e’) Lower panel: Higher magnification images

pr

showing representative VGLUT2 expression for each of the treatment groups. (f) VGLUT2

e-

expression as determined by immunofluorescence, was significantly decreased in layer V in

Pr

CUS mice compared with control, and ketamine treatment reversed the CUS effect. [Significant main effect of ketamine: layer I (F1,12 = 6.79, P <0.05), layer II/III (F1,12 = 5.41, P

al

<0.05); Significant stress X ketamine interaction: layer I (F 1,12 = 4.92, P <0.05), layer II/III

rn

(F1,12 = 5.81, P <0.05), layer V (F1,11 = 8.11, P <0.05). *P < 0.05 significantly different from

Jo u

control group; # P < 0.05 significantly different from CUS group, post-hoc Fisher’s LSD. Sal n = 3-4; ketamine n = 4; CUS n = 4-5; CUS+ketamine n = 3-4]. (g) VGAT immuno-positive puncta (red) can be seen in close apposition with a GFP-positive layer V pyramidal cell. Images are shown for sequential 1.2 um thick sub-stacks, each comprised of four 0.3 um Z sections. (h)

VGLUT1 (upper set) and VGLUT2 (lower set) expressing processes (red)

appear to contact GFP positive dendrites in layer I of mPFC. Scale bars: A,C,E = 200 um, A’,C’,E’ = 20 um, G = 5 um, H left panels = 2 um, right panels = 1 um.

Figure 3: CUS and ketamine exert opposite effects on spontaneous postsynaptic currents of mPFC pyramidal type 1 cells. (a) Representative traces of baseline, 5-HT and hypocretin-

29

Journal Pre-proof induced sPSCs in layer V type 1 pyramidal cell of mPFC brain slice from non-stressed control and CUS rats are shown. Two-thirds of the pyramidal cells recorded had a large hyperpolarizing voltage sag belonging to type 1 pyramidal cells. (b) CUS exposure resulted in a significant decrease in baseline and hcrt induced sIPSC frequency. [Baseline: t(34)=2.06, P=0.02, 5-HT: t(56)=1.58, P=0.06, hcrt: t(50)=12.21, P=0.02]. (c) CUS exposure resulted in a significant decrease in baseline, 5-HT and hypocretin-induced sEPSC frequency. [Baseline: t(54)=3.15, P=0.001, 5-HT: t(57)=2.73, P=0.004, hcrt: t(43)=5.02, P<0.0001]. N=30 cells/9

oo

f

rats for control, N=29 cells/10 rats for CUS. * P<0.05, ** P<0.01 compared with control. (d) A single dose of ketamine increased the frequency of 5-HT induced IPSCs [## Significant

pr

main effect of ketamine (F1,73 = 4.49, P <0.05), main effect of stress n.s. (F1,73 = 3.59, P

e-

=0.06)]. There were similar but non significant trends for baseline IPSCs [main effect of

Pr

ketamine n.s., F1,73 = 2.76, P =0.10); main effect of stress n.s., F1,73 = 2.84, P =0.10) ]. (e) CUS decreased frequencies of baseline, 5-HT and hypocretin-induced sEPSCs [** Significant

Conversley, ketamine increased frequencies of baseline, 5-HT and

rn

= 15.26, P <0.05)].

al

main effects of CUS: baseline (F1,73 = 8.68, P <0.05), 5-HT (F1,73 = 6.12, P <0.05), Hcrt (F1,72

Jo u

hypocretin-induced sEPSCs [## Significant main effect of ketamine: baseline (F1,73 = 6.11, P <0.05), 5-HT (F1,73 = 10.41, P <0.05), Hcrt (F1,72 = 5.89, P <0.05)].

Figure 4. CUS decreases IPSC frequency in layer II/III SST neurons. (a-b) SST-tdTomato mice were used for brain slice electrophysiology (triangle in b indicates patch pipette, scale bar,10 μm). (c) Sample traces of inhibitory postsynaptic currents (IPSC) and excitatory postsynaptic currents (EPSC) recorded from SST neurons of control mice and CUS mice. (d) CUS significantly reduced IPSC frequency ** p<0.01, 2-tail t-test), but not EPSC frequency. (n=32 for control IPSC, n=28 for CUS IPSC, n=30 for control EPSC, and n=25 for CUS EPSC). (e Sample traces of inhibitory postsynaptic currents (IPSC) and excitatory

30

Journal Pre-proof postsynaptic currents (EPSC) recorded from SST neurons of control mice and ketamine treated mice. (f) There were no significant effects of ketaminetreatment on IPSC or EPSC frequencies in SST neurons . (n=29 for control IPSC, n=27 for ketamine IPSC, n=32 for control EPSC, and n=25 for ketamine EPSC).

Figure 5. Antidepressant behavioral response to optogenetic regulation of mPFC GAD+ interneurons. (a) Mice received viral infusions of floxed ChR2-eYFP (AAV2-EF1a-DIO-

oo

f

hChR2(H134R)-eYFP) into the mPFC of Gad1-Cre+ or littermate controls. Representative image of ChR2-eYFP expression in mPFC after viral injection in Gad-Cre+ mice (white

pr

triangle indicates cannula placement). (b) Representative image of layer II/III GAD+ neurons

e-

(scale bar= 20 μm ). (c) Targeting of fluorescently labeled GAD+ neurons for current clamp

Pr

recordings scale bar,10 μm). (d) Example traces of GAD-ChR2 cell in the mPFC showing high fidelity, laser activation of GAD+ neurons in the mPFC upon laser stimulation at

al

different frequencies. (e) Schematic of the experimental timeline for laser stimulation and

rn

ketamine administration using Gad1-Cre+ and wild type (WT) littermates infused with floxed

Jo u

ChR2-eYFP (AAV2-EF1a-DIO-hChR2(H134R)-eYFP). Mice underwent a pre-swim and the next day were administered saline or ketamine and 10 min later received laser stimulation for 60 min. Behavior was examined 24 hrs later in the FST (f), and 72 hrs later in the NSFT (g). Ketamine administration resulted in a significant antidepressant effect in the FST and genotype showed an opposite, depressive-like effect for GAD1- Cre+ mice [## Significant main effect of ketamine (F1,20 = 10.07, P <0.05); ** Significant main effect of genotype (F1,20 = 32.20, P <0.05); Interaction n.s. ( F 1,20 = 3.61, P = 0.07). (g) Ketamine administration resulted in a significant antidepressant effect in the NSFT that was absent in GAD1- Cre+ mice; [Significant ketamine x genotype interaction (F1,20 = 12.09, P <0.05), *P < 0.05 vs. WT-saline group, # P<0.05 vs. Wt-ketamine group, post-hoc 31

Journal Pre-proof Fischer’s LSD test. There were significant main effects of ketamine (F 1,20 = 25.04, P <0.05) and genotype (F1,20 = 6.64, P <0.05). Results are the mean ± S.E.M., n = 6 per group. (h) Schematic of the experimental timeline for laser inhibition using Gad1-Cre+ and WT littermates with bilateral mPFC infusions of Arch-YFP (AAV2-EF1a-DIO-Arch-eYFP). There was no effect of laser inhibition of GABA interneurons in the mPFC on immobility

Figure

6.

Stress

and

ketamine-induced

oo

f

time in the FST ( (i) or latency to feed in the NSFT (j). Results are ±S.E.M. n=4-6/ group.

synaptic

events

in

the

mPFC.

GABA

pr

neurotransmission, driven by tonic firing GABA interneurons is required for balance and

e-

control of excitatory cortical function. Chronic stress or depression is associated with

Pr

decreases in the levels of the GABA transporter (VGAT) and synthetic enzymes GAD65 and GAD67, and functional capacity of GABA interneurons as demonstrated by decreases in the

al

frequency of basal and hypocretin-induced IPSCs in mPFC layer V pyramidal neurons. This

rn

results in alterations in inhibition/excitation balance in the mPFC. In contrast, ketamine

Jo u

treatment increases synaptic features of GABA as well as glutamate, neurotransmission. Tonic firing of GABA interneurons is driven by glutamate and stimulation of NMDA receptor, and the active, open channel state allows ketamine to enter and block channel activity. Optogenetic experiments indicate that the actions of ketamine are blocked by simultaneous stimulation of GABA interneurons. Thus it is possible that disinihibition of GABA

interneurons

resulting

in

a

glutamate

burst

that

stimulates

AMPA-induced

depolarization. This leads to activation of voltage dependent Ca2+ channels, release of BDNF and stimulation of TrkB-Akt that activates mTORC1 signaling and increased synthesis of proteins required for synapse maturation and formation. However, further time-course investigation of the circuit level changes following chronic stress and ketamine treatment will 32

Journal Pre-proof help determine the directionality of connectivity changes in the PFC downstream

Jo u

rn

al

Pr

e-

pr

oo

f

targets.

33

Journal Pre-proof Highlights 1) CUS exposure causes lasting deficits in GABA synaptic function in the medial PFC. 2) Ketamine increases GABA synaptic function and reverses the CUS-induced deficits.

Jo u

rn

al

Pr

e-

pr

oo

f

3) These GABA alterations are accompanied by changes in glutamate synaptic function.

34