Early growth response 2 regulates mouse cooperative behavior

Social cooperation is essential to animals’ physical health and psychological state. However, the underlying molecular and neurobiological mechanisms remain poorly understood. Here, we established a novel model for systematically evaluating the cooperative behavior of mice for the task of water reward. Using this paradigm, we characterized cooperative decits in isolated juvenile mice and the corrective effect of resocialization. Mechanistically, we found that the transcription factor early growth response 2 (Egr2)-dependent myelin maturation in the medial prefrontal cortex is necessary for the development of mice’s cooperative behavior. Additionally, corticosterone levels in the serum and medial prefrontal cortex were elevated in the isolated mice, potentially contributing to Egr2 expression reduction. This work suggests targeting Egr2 could be targeted for preventing and treating social isolation-related neuropsychiatric disorders in the future. mutual assistance skills. show juvenile SI multiple behavioral rescuable the pathological reveals that the mPFC, Mechanistically, the in by SI one two and three n = 9, 9, and 11 mice, respectively; SI: n = 9, 9, and 12 mice, respectively. e, f, Lineal tting showing negative correlation of serum corticosteroid levels with Egr2 protein (e) or mRNA (f) levels in the mPFC of SI mice. g-i, Representative Western blot images (g) and qPCR analysis (i) showing reduced Egr2 expression after corticosteroid treatment of primary oligodendrocytes; h, shows the quantied analysis of g; EGR2 expression is normalized to GAPDH in g, i. j, Schematic graph showing Egr2 regulate mice’s cooperative behavior through modulating oligodendrocyte maturation.

RNA sequencing (RNA-seq) identi ed Egr2 as a critical regulator for mice's cooperative behavior To explore the molecular mechanism underlying SI and resocialization, we performed RNA-seq on mPFC samples from the ~7-week-old GH and SI mice. We identi ed 1265 differentially expressed genes (DEGs) that were enriched in 4 pathways: cognitive behavior, myelin development, synaptic development, and ion channels (Extended Data Fig. 3a-b). After validating the mostly affected DEGs with real-time PCR (qPCR) and western blot, we further checked whether their expression could be restored by resocialization (Extended Data Fig. 3c-l). Finally, we got 3 candidate genes showing consistent reduction (in SI mice) and restoration (in GH-SI mice): early growth response 1 (Egr1), early growth response 2 (Egr2), and activityregulated cytoskeleton-associated protein (Arc) (Extended Data Fig. 3m-n). Intriguingly, the promoter regions of both Arc and Egr1 are predicted to be potential binding sites for Egr2. We performed a dualluciferase reporter experiment in 293 cells and validated Egr2 could transcriptionally upregulate Arc and Egr1 expression (Extended Data Fig. 4). Therefore, these results suggested Egr2, the known critical transcription factor for PNS myelination and also for hindbrain development 17,18 , as a possible master mediator for the effect of SI and resocialization.
Next, we generated an AAV9 virus expressing a shRNA targeting Egr2 (AAV-Egr2-RNAi). The 3-week-old weanling mice received virus injection into the bilateral mPFC and then group-housed or raised individually for 3 additional weeks (Fig. 3a, Extended Data Fig. 5a, c, e-f). AAV-Egr2-RNAi mice did not show any abnormalities in the cooperation-training period (Fig. 3b). The mice's performance on spatial working memory, fear learning and memory, locomotion, exploration, and anxiety-like behaviors were also unchanged (Extended Data Fig. 6). These results indicated Egr2 de ciency did not dampen mice's learning or emotional performance. However, in the subsequent cooperation-testing period, GH-Egr2-RNAi mice performed signi cantly worse than the control GH-GFP mice (Fig. 3c). We continued to explore if enhancing Egr2 could reverse the detrimental effect of SI. Like the AAV-Egr2-RNAi mice, overexpressing Egr2 in SI mice (SI-AAV-Egr2) did not alter mice's performance in the cooperation-training period (Extended Data Fig. 7a-b, 5b, d, g-h). In the subsequent testing period, however, the performance of SI-AAV-Egr2 mice were dramatically improved compared to SI-GFP mice, and to a level close to the control GH-GFP group (Extended Data Fig. 7c). Egr2 overexpression did not change mice's performance on spatial working memory, locomotion, and exploration behavior, but the mice displayed improved performance on fear learning and memory, and reduced anxiety-like behavior (Extended Data Fig. 8).
Therefore, both RNAi and overexpression results identi ed a critical role of Egr2 in regulating mice's cooperative behavior.
Egr2 was required for mPFC myelination and the middle stage maturation of oligodendrocyte Next, we found knocking down Egr2 led to mPFC myelin impairment, as re ected by elevated g-ratio, reduced heterochromatin proportion, and reduced MBP expression, by EM, immunohistology, and western blot analysis (Fig. 3d-j). However, we did not identify any signi cant synaptic changes in GH-Egr2-RNAi mice, despite Egr2 is indeed expressed in mPFC neurons (Extended Data Fig. 9). In contrast to the RNAi results, overexpressing Egr2 in SI mice rescued both myelin and synaptic defect (Extended Data Fig. 7dm). This synaptic improvement was possibly a dosage effect of Egr2: overexpression of Egr2 could improve synaptic function through upregulation of Arc, although synaptic development normally does not require Egr2. Indeed, we performed chromatin immunoprecipitation (ChIP) coupled with qPCR (ChIP-qPCR) experiment and identi ed a strong binding of Egr2 with the 1700-1850bp region upstream of Arc's transcription initiation site (Extended Data Fig. 10).
Egr2 is an established master regulator for PNS myelination, where it promotes the late-stage Schwann cell maturation 19 . To see if Egr2 behaves similarly in the CNS oligodendrocytes, we knocked down Egr2 in the differentiating primary oligodendrocyte progenitor cells (OPCs). We found lack of Egr2 did not alter the expression of the early-stage marker platelet-derived growth factor alpha receptor (PDGFRα), but signi cantly reduced the middle stage marker ectonucleotide pyrophosphatase/phosphodiesterase 6 (Enpp6) and the late-stage marker MBP ( Fig. 3k-m, Extended Data Fig. 11). Together, these results suggest Egr2 is required for mPFC myelination through promoting the middle-stage maturation of oligodendrocytes, but dispensable for neuronal synaptic development.

Elevated corticosteroid production in SI mice may inhibit Egr2 expression
We continued to explore how mPFC Egr2 is regulated by the social treatments. As a kind of stress, early SI could stimulate the production of corticosteroid in the adrenal cortex in juvenile mice 20 . Corticosteroid treatment, on the other hand, is reported to inhibit oligodendrocyte proliferation, in sharp contrast to Egr2's promoting role 21 . Is it the elevated corticosteroid level suppresses Egr2 expression in SI mice? On mouse brain slices, immuno uorescence validated the existence of corticosteroid receptor (GR) and phosphorylated GR (pGR) in O4-expressing mPFC oligodendrocytes (Fig. 4a-c). For the mice receiving 3week SI treatment, from the second week we evidenced a signi cant increase of the serum corticosteroid level (Fig. 4d). Additionally, we found a clear reverse correlation between a mouse's serum corticosteroid level and its mPFC Egr2 protein level ( Fig. 4e-f). Treating primary oligodendrocytes with corticosteroid reduced Egr2 mRNA and protein expression ( Fig. 4g-i). Therefore, our data suggest the elevated corticosteroid level in SI mice could lead to reduced Egr2 expression.
Taken together, as illustrated in Fig. 4j, we designed a new paradigm that could reliably and quantitatively evaluate mice's cooperative behavior. With this system, we found early SI led to cooperation defect that could be reversed by immediately followed resocialization; the myelination state and synaptic integrity in the mPFC are two main targets of SI and resocialization. We further identi ed the transcriptional regulator Egr2 conveys a major part of the effect of SI and resocialization on social cooperation. Egr2 works through promoting oligodendrocyte maturation, suggesting drugs targeting Egr2 or mPFC myelination could be explored for future preventing and treating of isolation-related neuropsychiatric disorders.

Discussion
Rodents are prosocial animals and known to cooperate with one another. A few studies investigated the ability of rats to work together to obtain water or food 9,11 . However, these behavior paradigms are relatively simple, and to some extent, not enough to re ect the characteristics of mutual adaptation and improvement in cooperative behavior. In this study, we investigated the cooperative behavior of mice in a long-term and multi-dimensional way following training them to learn to use a new device for drinking water. The results suggested that the behavior paradigm can well re ect the cooperative willingness (codrinking latency) and e ciency (number and duration of co-drinking behavior) of mice. Through the testing period, their cooperative ability is continuously enhanced with the mutual bene t of getting more water. This is clear evidence that mice have remarkable social intelligence and mutual assistance skills.
The current results show juvenile SI mice have multiple behavioral abnormalities that are rescuable by resocialization. Correspondingly, the pathological evidence reveals that resocialization greatly attenuates myelin sheath and synaptic damages in the mPFC, which is consistent with previous reports 16,22 . Moreover, we demonstrated these isolated mice have impaired cooperative ability, which can be recovered with reintroduction into a social environment. Mechanistically, our results suggest the increased serum corticosteroid level results in the downregulation of Egr2 expression in the mPFC of the juvenile SI mice. Local overexpression of Egr2 can markedly mitigate the pathological defect in the mPFC and rescue the cooperative abnormalities in SI mice. Resocialization has corrective effects on multiple behaviors for SI mice, suggesting that besides the mPFC, it might also attenuate pathological damages of other brain regions. However, reducing Egr2 expression in the mPFC affects social cooperation speci cally, without changing spatial working memory, fear learning and memory, locomotion, exploration, and anxiety-like behaviors. This may be attributed to the fact that mPFC is mainly responsible for controlling social behavior rather than emotion or working memory.
Egr2 has been shown to control Schwann cell myelination by modulating the expression of genes encoding myelin proteins and enzymes required for the synthesis of myelin lipids [23][24][25] . It also controls Schwann cell proliferation, differentiation, and death in vitro 19,26,27 . In the brain, Egr2 expression is induced under conditions like electroconvulsive shock, focal cerebral ischemia, opiate withdrawal, and limbic seizures [28][29][30] . Egr2 is also shown to play a key role in the stabilization of dentate gyrus long-term potentiation in adult mice 31 . In the present study, we have identi ed a crucial role of Egr2 in maintaining mPFC myelin and synapse integrity, which will advance our understanding of the neurobiological and molecular bases for mouse social cooperative behavior. However, Egr2 expresses in various brain regions including the neocortex, hippocampus, amygdala, olfactory bulb, striatum, cerebellum, diencephalic, and brainstem structures 28,32 . Whether there is a brain region-speci city role for Egr2 in brain functions needs to be clari ed. Indeed, a recent study reported that the neuropeptide Tac2 coordinates complex behavioral effects after chronic SI tress in a distributed manner 33 . Whether Egr2 regulates different behavioral effects of SI in different brain regions in a similar way is worth exploring.

Animals and experimental treatment
Weanling male C57BL/6J mice (3-week-old) were randomly divided into GH and SI group. Five mice (for GH group) or a single mouse (for SI group) were housed in a plexiglass cage (31 cm × 22 cm ×15 cm) for 3 weeks as the social treatments (Fig. 1a); for resocialization, two SI mice were then put in a cage containing three GH mice for 4 weeks (Fig. 2a). For virus injection, 3-week-old weanling mice were injected with indicated viruses (AAV-Egr2-eGFP/AAV-Egr2-RNAi-eGFP or AAV-Control-eGFP) in the mPFC and then underwent the social treatments (Fig. 3a, Extended Data Fig. 7a). Mice were maintained on a 12h light/dark cycle with ambient temperature (18-22°C) and humidity (30-50%), and having free access to food and water. All experiments were conducted in accordance with international standards on animal welfare and the guidelines of the Institute for Laboratory Animal Research of Nanjing Medical University.

Behavioral test
Mouse activity in the following behavioral test was collected using a computer-connected digital video camera (TopScan, CleverSys, Inc., Reston, VA). Room lights, except for a dim light, were switched off during each experimental session for mouse comfort. Before each test, the apparatuses were cleaned with 70% alcohol to eliminate possible remaining olfactory cues. All tests were performed by two independent experimenters, who were each blind to the treatment schedule.

Cooperative test
The cooperative ability of mice was evaluated by a cooperative drinking device developed by our laboratory (Fig. 1b-c, Supplemental video 1). The device was plexiglass cage (60 cm × 40 cm × 35 cm) equipped with two water valves from which drop shape water could be delivered by turning on two serial photoelectric switches. During the rst phase of training, named the training period, one photoelectric switch was defaulted ON; a mouse was placed into the chamber, and, in a 5-min period, it can explore and turn on the other switch to get water-rewards. Training was conducted over 7 consecutive days, with 2 trials per day. The latency for the rst successful drinking, as well as the times and time of drinking during each trial were collected. During the second phase, named the testing period, both switches were defaulted OFF. Two trained mice were placed in the chamber and allowed to move freely for 10 min. Only if they turn on both switches at the same time, they can get the water-reward together. The co-drinking latency, times and time of co-drinking were recorded. The test was carried out once a day for ve days. The mice were returned to their cages after each trail. To motivate mice to drink from the device, water supply was suspended for six hours before each trial. To increase their incentive for getting drinking water during the training or testing period, we made water unavailable from 6 hours before each training or testing; this short-term water deprivation had no signi cant effect on mice's exploratory activities, anxiety-like behavior, spatial working memory, or weight (Extended Data Fig. 1).

Open-eld test
The open eld test was used to evaluate the locomotion and anxiety-like behaviors 34 . The open eld consisted of a square blue box (60 cm × 60 cm × 25 cm), with an outlined center area (30 cm × 30 cm). Each mouse was placed in the center of box and allowed to move freely for 5 min within the box. The time spent in the center area, times of entering the center area and total distance traveled were calculated during the test.

Elevated plus maze test
The anxiety-like behavior was also evaluated by the elevated plus maze that was composed of four arms (50 cm ×10 cm) connected by a central square (10 cm × 10 cm) and elevated 100 cm above the oor 15 . Two opposite arms were open, while the remaining were closed with 40 cm high walls. The mice were placed into the close arm and allowed to freely explore the maze for 5 min. The percentage of time spent and the frequency of entries into the open arm were calculated.

Y-maze test
The spatial working memory of mice was examined using a Y-shaped maze 15 . The maze consisted of three arms (8× 30× 15 cm), with an angle of 120 degrees between each arm. The procedure included 5 min-training stage and 5 min-testing stage, with an interval of 2 h. During the rst stage, one arm named novel arm (NA) was blocked by a ba e, allowing the mice to move freely in the other two arms for 5 min. During the second stage, the NA was opened and mice could freely explore throughout all 3 arms for 5 min. The percentage of time traveled in the NA arm, number of entries into the NA arm were analyzed.

Fear conditioning test
The mouse was placed in the conditioning chamber for 3 min as a habituation period followed by one tone-foot-shock (tone, 30 s, 70 dB, 1 kHz; foot-shock, 2 s, 0.8 mA) 35 . The stimulation was given every two min within six min. The mouse was placed in the chamber again with tone-shock pairing (tone, 30 s, 70 dB, 1kHz) for 3 min after 24 h. Freezing behavior, de ned as the absence of all visible movement of the body except the movement necessitated by respiration, was scored.

RNA sequencing
Total RNA from the mPFC was extracted with RNAiso Plus (Takara, #9109) according to the manufacturer's instructions. The RNA quality and quantity were measured using NanoDrop 2000 (Thermo Scienti c). RNA library was constructed using the BGISEQ-500 platform according to the manufacturer's instruction; after that, adapters were ligated to each end of the RNAs and subsequently reverse transcribed to create single-stranded cDNA and sequenced on the BGISEQ-500 platform with a read length of 50 bps. Stringent criteria were set to determine signi cantly dysregulated genes: adjusted p value below 0.05 and log2FC of greater than 1. The mPFC samples from SI and GH mice were analyzed as two independent datasets. Heatmaps representing z-scores were generated using the seaborn package in python.

Stereotaxic injection
Deeply anesthetized mice were xed in a stereotaxic instrument with the skull surface exposed. A total of 1.5 µl of AAV-Egr2, AAV-Egr2-RNAi (1 × 10 13 µg) or control virus (AAV-GFP) were infused bilaterally into the mPFC at the following coordinates: +1.5 mm anterior/posterior, -0.75 mm dorsal/ventral and ±1.0 mm medial/lateral relative to Bregma use 33-gauge syringe needles (Hamilton) 36 . The infusion rate was 0.2 µl/min, and the cannula was left in the place for 5 min following completion of the infusion. Mice were allowed to recover for 2 days before isolated housing. Behavioral experiments were performed 21 days after the injection. Following the behavioral tests, mice were perfused and brain sections were examined by electron microscopy, qPCR, histology, or Western blot analysis.

Cell lines
Neuro-2a cell was a gift from Dr. Gang Hu (Nanjing Medical University) and was used for the ChIP, luciferase reporter assay and western blot analysis. For transfection, the cells were plated 24 hours before and then transfected with appropriate constructs using Lipofectamine 2000 (Invitrogen, #52887) according to the manufacturer's instruction. Culture medium was replaced 5-6 hours after transfection; transfected cells were cultured for 24 additional hours for q-PCR or 48 h for Western blotting.

Luciferase reporter assay
The luciferase reporter assay for Arc/Egr1 promoter activity was performed according to the manufacturer's instructions (Promega, E1910). Brie y, 0.05 μg luciferase reporter plasmid, 0.2 μg EGR2 expression plasmid and 1.25 ng Renilla were co-transfected into the 80-90% con uence Neuro-2a cells; 48 hours after the transfection, cells were harvested and resuspended in 50 μl passive lysis buffer, and then placed on a micro oscillator and split for 15 min. The supernatant was used to measure luciferase activity; the normalized values (Renilla/ re y activity) were used for analysis. Experiments were performed in triplicate.

ChIP and ChIP qPCR data analysis
We used a Magna ChIP Kit (Millipore, #17-10085) for the chromatin analysis. Brie y, Neuro-2a cells were cross-linked with 1% formaldehyde; the cells were collected by scratching, and then lysed and resuspended in nuclear buffer. 24 cycles of sonication (10 s for each, followed by a 20 s interval) were applied to break the chromatin into fragments between 200 and 1000 bp. The samples were incubated with anti-Egr2 antibody (Santa Cruz, #sc-293195) or anti-RNA polymerase antibody (Millipore, #17-10085) and magnetic beads overnight at 4°C. Normal IgG was used as a negative control. The immunoprecipitants were separated by magnetic rack and washed. The DNA fragments were released by incubation with proteinase K at 62°C for 3 h with continuous shaking, and isolated by ltration. qPCR was performed in a 20-μl reaction volume using the SYBR Green PCR master mix (Takara, #RR420B). For electrophoresis analysis, PCR was performed with a NEB Next High-Fidelity 2X PCR Master Mix (New England Biolabs) with product resolved by agarose electrophoresis, followed by gel imaging. ChIP-qPCR (qChIP) primers were designed in proximity to Arc binding motif sequences. For each gene of interest, separate primer pairs were designed. The PCR primers are listed in (Supplementary Table 1). qChIP results were presented as percent to input.

Corticosterone Assay
Orbit blood was collected from mice into ice-cooled centrifugal tubes and centrifuged (1700 ×g, 10 min, 4°C) to separate serum and stored below 80°C until the assay. Blood sampling took place between 09:00 and 12:00 h 37 . Corticosterone was quanti ed by the enzyme-linked immunosorbent assay (Parameter Corticosterone Assay R&D Systems, #KGE009) according to assay instructions. All samples were run in duplicate, counterbalanced across plates, and were within the standard curve. Final values were determined by averaging the results of duplicated samples.

Electron microscopy
Cardiac of the anesthetized mice was poured into 0.9% saline by perfusion pump for 2 min, followed by 2.5% glutaraldehyde, 2% paraformaldehyde in 0.1 M PB. The brain was rapidly removed and post-xed (4% paraformaldehyde and 2.5% glutaraldehyde in 0.1 M phosphate buffer containing 0.5% NaCl) for at least one week at 4°C. Following washing with distilled water, the sections were stained with 0.5% uranyl acetate in 70% ethanol for 1 h, dehydrated in a serial dilution of ethanol, and cleared in propylene oxide, embedded in Epon, and incubated at 60°C for 24 h. The tissues in Epon blocks were then trimmed and reoriented so that ultrathin (70 nm) cross sections were cut using the ultramicrotome (Leica EM UC7).

Western blotting
For Western blot analyses, samples were lysed in the RIPA buffer (Beyotime, #P0013B) containing protease inhibitors (Beyotime, #st506) and phosphatase inhibitors (Roche, #04906837001). Cell lysates were cleared of cellular debris by centrifugation (12000 rpm for 15 min) and equal amounts of protein were loaded in the SDS-PAGE on 8-12% gels and transferred to PVDF membranes. After blocking for 1 h in 5% nonfat milk in TBST, the membranes were incubated at 4℃ overnight with one of the following primary antibodies: anti-Egr1 (santa cruz, #sc-189, 1:1000 dilution), anti-Egr2 (santa cruz, #sc-293195, 1:1000 dilution), anti-Arc (santa cruz, #sc-17839, 1:1000 dilution), anti-Lef1 (Proteintech, #14972-1-AP, 1:500 dilution), anti-Nab2 (Proteintech, #19601-1-AP, 1:1000 dilution), anti-NTs (Immunoway, #YT5535,  Note a successful training is re ected by the declining (or rising) to latter attened curves. d, The testing setup for the cooperative water-drinking experiment. e, Graphs showing the cooperation performance in the testing setup for GH and SI mice. Note the separation of the two curves in each graph, as assessed by the elevated drinking latency (F(1, 80)    The cooperation defect and mPFC impairment of SI mice were rescued by resocialization. a, A timeline diagram showing mice receiving weanling, social isolation, resocialization, cooperative water-drinking test, and other behavioral tests. b, c, Graphs showing the time before the rst drinking (drinking latency), the times of drinking (drinking number), and the total time spent on drinking (drinking time) each day in the training (b) and testing (c) periods. Note curves of GH-SI mice are closer to GH ones in c. as assessed