COX-1/EP1R Inhibition Ameliorate Decits in Adult Neurogenesis and Social Interaction via Alleviating NLRP3/ IL-1β Activation During Hypobaric Hypoxia

Low oxygen environments like hypobaric hypoxia (HH) are common nodes to various diseases: characterized by neuroinammation, which is detrimental to the structural and functional aspects of hippocampal circuitry. Various hypoxic conditions also lead to elevation of NLRP3 mediated neuroinammation that may contribute to cognitive decits. Components of neurogenic niches like microglia and astrocyte are largely affected by neuroinammation; however, a systematic investigation of the impact of NLRP3 mediated neuroinammation on components of neurogenic niche during hypoxia (HH) remains elusive. dependent adult neurogenesis during HH condition. Pharmacological inhibition of COX-1 and EP1R during 7 days of HH exposure reverts glial activation in DG Activation of COX-1 as well EP1R promoted neuro-inammation as well as microglia mediated neurotoxicity, we also reported COX-1 dominance in microglia in HH condition. EP1 gene ablation or selective antagonist shown to blunt microglial mediated inammatory response. We planned our next experiment to answer the question about how COX-1/EP1R antagonism during HH can modulate morphology of microglia and astrocytes as they are prominent contributor to neurogenic niche.

being clear specially during HH. Recent research has focused on inhibiting the activation of in ammasomes and thus reducing the maturation of proin ammatory cytokines such as interleukin (IL)-1β and IL-18. NSAIDS like Fenamate that inhibit COX-1 and COX-2 are reported to be speci c NLRP3 blocker that can cross BBB and FDA approved [14]. Recently COX inhibitors are reported to inhibit NLRP3 activation in various modalities like pancreatitis, ischemia [15,16,17].
Despite light shed on de cits of hippocampal-dependent tasks during HH exposure [18,19,20,21] the effect of HH exposure on social interaction in rats is still an enigma. Effect of HH exposure on neuroin ammation has been discussed earlier with primitive evidence but how glial cells activated and leads to perturbation of neurogenic niche is still unknown. The role of NLRP3 mediated in ammation in adult neurogenesis, social and anxiety like behavior during HH exposure is still very much vivid. The effect of COX-1 speci c inhibition on adult neurogenesis and its active PGE2 downstream receptor is far from being clear specially during HH. The ndings led to the proposal that microglial COX-1 via EP1 receptor induces NLRP3 mediated neuroin ammation, activation of glial cells, and perturbation of neurogenic niche in DG, culminates into defective social behaviour followed by elevation in anxiety during HH.

Experimental animals
Adult male Sprague Dawley rats of weight 230-280 grams were used for each experiment in this study. Optimal housing conditions (22 ± 2 °C, 54-60% humidity, 12 h light-dark cycle) were provided.
Experiments were performed in accordance with the guidelines of the Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA) of the Indian Government and were approved by our Institutional Animal Ethics Committee (IAEC) IAEC/DIPAS/2015-22. Food pellets (Lipton India Ltd., India) and tap water were provided ad libitum. Animals were habituated well with the experimenter before behavioural study and sacri ce.Adult rats were randomly assigned into one of the following 7 groups: cage control (CC), 1 day of HH(1HH),3 days of HH(3HH),7 days of HH (7HH), 7HH with COX-1 inhibitor (7HH + ValS), and 7HH with EP1 R antagonist (7HH + SC19220). Speci c groups and number of animals used in each experiment are mentioned in results and gure legends respectively.

High altitude hypoxia chamber
The animals were stimulated with high altitude pressure using a specially designed animal decompression chamber (Seven Star Model).The ideal conditions like atmospheric pressure 282 mmHg, PO2 59 mm Hg, 28-300 C, and relative humidity of 55-60% were maintained [7,22,23]. Rats were habituated at an altitude of 4572 m (15,000 ft) for 24 h prior to continuous exposure to HH at an altitude of 7620 m (25,000 ft) for 1, 3, or 7 days. A 12hr day-night cycle was adequately maintained. Food and water were provided ad libitum to rats. The schematic representation of experimental design provided in Supplementary g. 1.
The salt inhibitor was rst dissolved in DMSO and then diluted in 0.9% NaCl to obtain the nal concentration that was 20%. The selection of valeryl salicyclate dose was inconsistent with the previous usage in rats. [21,24].

Assessment of social interaction behavior
Social recognition and novelty preference were evaluated using a three-chambered apparatus with minor modi cations that appropriately t the size of Sprague Dawley rats [31,32,33,34]. This apparatus was made of Plexiglas dimensions: 50 cm x 50 cm x 40 cm and divided by walls, with an open middle section that allowed free access to each chamber. We habituate animals for 10 min daily with the whole instrument till rat explored it all (max. for 3 days). The movement of the experimental animals was recorded using SMART 3.0, PANLAB, Harvard Apparatus, USA. During trial 1, the experimental rat was allowed to explore between the chamber containing the wired cage (Empty) and another identical chamber containing stranger rat 1(Social). During trial 2, that was for another 10 min. The experimental rat was allowed the explore the chambers containing stranger rat 1(familiar) and stranger rat 2 (novel), and time spent in both the chamber was quanti ed. Social tendencies of test rat quanti ed using this test the principle behind the parameters a) the time spent with social stimuli versus empty enclosure and b) preference for novel versus already explored familiar stimuli. Contact behaviour was considered as the number of direct (active) contacts between the experimental rat and the wired enclosure housing or not housing the Stranger 1 or stranger 2, for each chamber individually. Direct contact between the experimental rat and the containment wired cup, or stretching of the body of the experimental rat in an area of 3-5 cm around the cup is counted as direct contact. After habituation rats were exposed to 7HH exposure and then tested again for same parameters.

Elevated plus maze
Elevated plus maze (EPM) is a widely used test for measuring anxiety-related behaviour in rats [35,36]. EPM is a plus-shaped instrument made up of four elevated arms 10cm wide and 50cm long, perpendicular to each other, the two opposite closed-arm are equipped with 30 cm high walls. An overhead IR camera, coupled to a tracking system, ANY-Maze software (Stoelting, USA), was used to track the rat in the EPM. During this behavioral test, the rat was placed at the center of the apparatus to explore evenly lit all the arms freely maximum for 5min. The time spent in the open arm by the rat was calculated; to determine the anxiety level in the rat.

Immunohistochemistry
Animals from all the groups were anesthetized deeply with urethane injection (1.2 g/kg body weight), xed using ice-cold 4% paraformaldehyde (PFA) [18,19]. IHC experiments were performed in 30μm, freeoating, coronal sections for every antibody. BrdU staining required pre-treatment with the solution (2 M HCl + Triton X-100 for 45 min at RT) for DNA denaturation and washing with PBS (pH 8.5) afterward.
Denaturation followed by incubation with a rat anti-BrdU primary antibody details provided in Table 1.
Coronal brain section of the 30μm section of the dorsal hippocampus from the bregma -3.14 to -4.30 for all the staining protocol. Each section was washed with PBST (PBS with 0.1% Tween-20. Heating for 10 min was done for epitope retrieval in sodium citrate solution at (pH 6.0). Blocking was for 2 h in blocking buffer comprising 0.03% Triton X-100, 10% goat serum in PBS followed by incubation in an appropriate primary antibody for 40 h at 4 0 C followed by Fluorophore-conjugated with secondary antibodies (Alexa Fluor 488 and Alexa Fluor 633, Invitrogen, USA) for 2h at RT.

Sholl analysis
Microglial morphology was analysed using a previously published protocol [21,37]. Microglial cells in the DG region were selected (6-8 cells/section; n = 6) on 30μm hippocampal tissue sections pre xed with 4% PFA. Concentric circles were drawn using the ImageJ concentric circles plugin. Circles were centered on the soma by manually determining the radii, which were then increased by 2μm with every circle. Sholl analysis was performed using the ImageJ Sholl analysis plugin. These data were quantitated using the following parameters: Nm = process maximum (the maximum number of intersections for the cell); Cr = critical value (the distance from the soma where the Nm occurred); maximum branch length (μm, the maximum radius at which a branch intersection occurred); and Np = number of primary branches (the number of branches originating from the microglial soma). The rami cation index (Nm/ Np) was calculated to quantify cell branching density. The cell soma area was also calculated using ImageJ.
Skeleton analysis was performed on images stained with DAB, as described previously. The following sequence was followed: Grey Scale-Subtract Background-Binary-Skeletonize-Analyse Skeleton. The ImageJ Skeletonize plugin was used to quantitate microglial cell process endpoints and calculate the summed microglial process length.

Microscopy
The dorsal hippocampus was taken for BrdU or DCX positive immune cells counting in 30μm sections as described in the protocols before [ 26,27,38] using Olympus BX51TF microscope, Japan. The three subpopulations of DCX positive immune-reactive cells based on their maturity were counted individually. The mean intensity values of both left and right hippocampus were considered for analysis.

RT PCR
Rat hippocampal tissue was collected and weighed (30-50mg). Homogenized in 1000 ml TRIZOL reagent (Sigma Aldrich, USA) at RT for 15min. Chloroform in the quantity of 250ml per 1000 ml TRIZOL reagent was added in the hippocampal lysate. Samples were vortex vigorously for the 30s and kept at RT followed by centrifuge at 12,000 rpm at 400 C. for 5min. Centrifugation lead separation of the solution in three phases. After the separation of the aqueous phase, RNA was precipitated using isopropyl alcohol. A gel-like RNA pallet was visible on the bottom or side of the wall and stored at -80°C. Samples were made sure to have a 260/280 ratio of more than 1.8. The purity of total RNA regarding size, distribution, and integrity was evaluated by the denaturation using denaturing agarose gel electrophoresis. cDNA was prepared by using the RT2 rst-strand cDNA synthesis kit (Qiagen, USA). It was then stored at -15 to -30°C freezer for further use.
The RT2 Pro ler for rat neurogenesis and neuroin ammation (PARN-404, PARN-011-Qiagen, USA) performed according to the provider's protocol. The data was analyzed using Gene Globe Data Analysis Centre facilitated software-based tool available at Qiagen website (www.qiagen.com/shop/genes-andpathways/data-analysis-center-overview-page). The spreadsheet displayed Delta Ct values was uploaded and normalized with the values of housekeeping genes (at least 2 out 5 were selected) . The values of fold change were used for comparison between groups.

Western immunoblotting
A standardized protocol for Western blotting was performed [19,21] on hippocampal tissue. The membrane was immune blotted with primary antibodies in Table 1. The blots were probed using HRP hosphor secondary antibody and developed with enhanced chemiluminescent (ECL) kit (Abcam, USA) in Molecular imager Chemidoc imaging system (Bio-Rad). Densitometry analysis to quantitate the band intensity of the blot was performed using ImageJ software (NIH).

High performance liquid chromatography (HPLC)
The Serotonin (5HT) levels in hippocampal tissue were measured using ion-pairing reversed phase HPLC-ECD. Tissue samples were sonicated in the ice-cold 0.05 M (500 ml/100 mg tissue) perchloric acid buffer and were centrifuged for 5 min at 10,000 rpm. Supernatants were ltered with syringe lter (0.45 mm, Millipore, USA) and stored at −80 °C for further use. To estimate 5HT levels by HPLC-ECD, 50 μl of the aqueous lysate was injected into a C18 reverse phase column (Waters, USA) from an auto sampler (Water 717, USA) using an HPLC pump (Waters 515, USA). The column pressure was maintained at 1800 psi. The mobile phase comprised of vacuum-degassed 8.65 mM heptane sulphonic acid, 0.27 mM Ethylene Diamine Tetra-acetic Acid (EDTA), 13% acetonitrile, 0.4% triethylamine and 0.2% phosphoric acid with ow rate at 1 ml/min in isocratic ow mode. The run time was set for 20 min, and peaks were identi ed in the sample by electrochemical detector. Compounds were identi ed based on the retention time matching with standards.

Statistical analysis
Statistical analysis was performed using Prism software (GraphPad, San Diego, CA, USA). Data were analyzed using ANOVAs to test differences of means between experimental groups mentioned with the results. All statistical tests were considered statistically reliable at a p-value of <0.05.

Results
Chronic HH exposure caused reproducible change in social interaction as well as anxiety-like behaviors HH exposure exhibits detrimental effects on body weight at 3HH (p<0.01), 7HH (p<0.05) as well on food intake at 3HH (p<0.05) and 7HH (p<0..01) as compare to control Supplementary Fig. 2.
Previous studies have shown vulnerability in hippocampal-dependent tasks, so we sought to examine whether hippocampal based social memory is affected. The behavioral schematic setup was representing two trials, each consisting of 10 min (Fig. 1a). Representative track-plots are provided for both trial 1 and trial 2 from control and 7 days HH exposed rats showings differential movement in the chamber ( Fig. 1 b). Statistical analysis showed that time spent in the chamber containing social stimuli were markedly lower in 7HH exposed (F 1,18 = 7.764, p<0.01) animals. In contrast, this time was signi cantly higher with empty enclosure (p<0.05) as compared to control rats (two-way ANOVA, Bonferroni's multiple comparison test, Fig. 1 c, n=10). However, contact behaviour that represents the direct exploration of two stimuli, was also altered in 7HH exposed rats as there was a decrease in contact behavior with social stimuli (F 1,18 = 7.709, p<0.05) and it increases when rats explored with an empty enclosure. Quanti cation of Trial 2 parameters revealed that both stressed (7HH) and control animals spent almost similar time in the chamber with the novel stimuli, but 7HH stressed animals spend more time with the familiar animal compared to control (F 1,18 = 6.225, p<0.05, Fig. 1 d). However, the direct contact behaviour of test rats from 7HH group was lower (F 1,18 = 12.51, p<0.05) with novel stimuli and higher with familiar (p<0.05).
During many psychiatric disease states, anxiety and social de cits are co-expressed as the baso-lateral amygdala and the ventral hippocampus are reported to share conventional circuitry and robust reciprocal connections. In line with this information we next evaluated anxiety-like behaviour using EPM.
Representative track plot shown for reference as lesser track movement of rats recorded in open arm ( Fig.1 e). Exposure to 7HH reduced the time spent in open arm signi cantly (one-way ANOVA, F 3, 36 = 11.35, p<0.01, Fig.1 f, Tukey's post-hoc test, n=10) as compared to CC but no signi cant reduction observed at 1HH and 3HH exposure.
Together, these experiments suggest 7 days of HH exposure diminished social interaction behaviour in rats by reducing the discrimination between social stimuli as the hippocampus being one of the affected brain regions. Along with social interaction, anxiety-like behaviour was also up-surged after 7HH exposure.

HH exposure diminishes expression of BDNF, Serotonin and pCREB in DG
Given that BDNF and Serotonin has been demonstrated to regulate mood associated behavior in rat, we investigated the effect of HH exposure on their expression. We found that the expression levels of BDNF decreased signi cantly in temporal manner straight from 1HH (one-way ANOVA, F 3, 20 = 9.108, p<0.01, Tukey's post-hoc test, Fig. 2 a, b, n=3), 3HH (p<0.01) and 7HH (p<0.01) exposure in DG indicated by representative IHC images as well. We also found decrease in mRNA levels of BDNF as well Overall, these results suggest that HH exposure upsets the microenvironment of the DG by altering the levels of BDNF, Serotonin and pCREB in day dependent manner that might associated with de cits in social and elevation of anxiety like behavior.
Representative immuno uorescence images are shown in Fig. 4a, represents a marked reduction in BrdU positive (+ve) cells after time-dependent HH exposure, which is signi cantly reduced at 7 days of exposure (one-way ANOVA, F 3, 20 = 6.928, p<0.01, Tukey's post-hoc test, Fig. 4b, n=6). Endogenous neuronal proliferation marker Ki-67, showed marked reduction at 7 HH exposure (F 3, 20 = 4.033, p<0.05, We inferred this results as 7HH exposure procreate de cits in generation of NPCs, immature neurons as well maturation of neurons whereas no change was observed in mature population of neurons and in gliogenesis. The observed de cits in neurogenesis are supported by mRNA expression of genes that contribute to different functional class associated with neurogenesis.

HH evokes NLRP3 In ammasome mediated neuroin ammation and microglial activation
To assess the expression of in ammatory cytokine and their receptors we used RT 2 pro ler for in ammatory cytokines and receptors. Quantitative PCR array analysis revealed modulation of mRNA expression. mRNA expression of pro-in ammatory genes like CXCL1, CXCL12, CXCR2, IFNG, VEGFA, CSF3, IL1B, CCR1, CCR8, CCR5, IL10RA, LTB which related to decrease in neurogenesis and microglial activation were shown to be upregulated. On the other hand mRNA expression of genes like IL1A, IL1RN, BMP2, CD40LG that are anti-in ammatory and positively regulate neurogenesis are decreased after HH exposure. We previously reported that temporal exposure of HH activates microglia and astrocytes in the DG followed by up-regulation of pro-in ammatory cytokines right from 1 day of exposure. Fold value that was more than 2 was considered for statistically signi cance (Fig 5, n=3).
Induction of pro-in ammatory cytokine lead us to further investigate the role of other neuro-in ammation associated mechanisms that may contribute to such pathophysiology we explored the expression of NLRP3/NFkB pathway. Here, we observed in ation in the phosphorylation in NFkB at its subunit 536 in hippocampus after 3HH (one-way ANOVA, F 3, 12 = 7.436, p<0.05, Tuckey's post-hoc, test n=3, Fig 6 a, b, c) and 7HH (p<0.05) exposure. We found a striking upsurge in the expression of levels of IL1β immediately after 1day (one-way ANOVA, F 3, 16 = 16.85, p<0.01, n=3, Fig 6 a, b, c), 3day (p<0.01) and continued till 7day (p<0.001) in the hippocampus. We also found increase in the mRNA levels of IL1β shown in the RT-PCR panel ( Fig 5). In consistence with that we also observed signi cant increase in the NLRP3 expression in the DG after 1HH (one-way ANOVA, F 3, 20 = 12.15, p<0.05, n=3, Fig 6 a, b, c), 3HH (p<0.001) and 7HH (p<0.001) exposure compare to control. We next evaluate the levels of downstream molecule that is caspase-1 and it also showed up regulation of the expression after 3HH (one-way ANOVA, We earlier observed induction in the PGE2 conc. And associated neuroin ammation in the hippocampus and plasma, but its active downstream receptor was not explored during HH. We next checked the expression of all PGE2 receptor expression in hippocampus. Representative immunoblot of EP1 receptor and its quantitative analysis revealed its marked increase at 3HH (one-way ANOVA, F 3, 12 = 4.693, p<0.05, Fig. 7 a) p<0.05) and at 7HH exposure (p<0.05). Expression of another prostanoid receptor EP2 was (oneway ANOVA, F 3, 12 = 3.887, p<0.05 , Fig 7 b) up-regulated after 7 day of HH exposure as indicated by representative immunoblot. Whereas, no signi cant difference was found in the expression of EP3, and EP4 receptors in the hippocampal tissue lysate after time-dependent exposure to HH represented with immunoblots and quantitative analysis (Fig 7 c and d).
These data indicate that HH differentially regulate expression of PGE2 G-protein coupled EP receptors. Only EP1 and EP2 receptors were reproducibly up-regulated after 3 and 7 days of HH exposure respectively, whereas expression EP3 and EP4 receptors were unaltered in the hippocampus after HH exposure at any day. PGE2 receptor EP1 is evident in the activated microglia as well as in astrocyte during HH exposure.
Induction in the EP1 expression in hippocampus lead us to examine its expression in the cell types in DG. Microglia and astrocyte activation was also prominent in the DG during HH exposure which opens the possibility of EP1R expression in these cell types. So we next performed a co-labelling experiment and found a signi cant increase in the CD-68+ EP1R+ (yellow) cells in the DG after 3HH exposure (Unpaired ttest, F 5, 5 = 1.202, p<0.01, Fig. 8 a, b) in the DG. Similarly, we found an interesting observation where EP1R+ GFAP+ (yellow) cells were signi cantly more in DG after 3 days of HH (Unpaired t-test, F 5,5 =2.806, p<0.001, Fig. 8 c, d) exposure as compare to control.
Taken, together we can say EP1R expression in the DG contributing to the activation of glial cells and in uencing neurogenic niche during HH exposure.
COX-1 dependent PGE2 response via downstream EP1 receptor is critical for the induction of social interaction de cits during HH exposure.
There are COX-1 and EP1 de ciency reported to abolished social avoidance and impulsive behavior under acute social and environmental stress. We found signi cant increase in protein expression of EP1R in hippocampus after 7HH exposure. The Tukey's post-hoc test revealed the considerable increase in EP1R expression after 7HH and was markedly decreased after treatment with valeryl salicyclate (ValS), COX-1 inhibitor (F 3, 12 = 8.444, p<0.05, Fig. 9 a) and SC19220, EP1R antagonist (p<0.01). We previously reported increase in COX-1 expression during 1HH, 3HH and 7HH of exposure. Signi cant positive correlation in the expression of COX-1 and EP1 was observed (Pearson correlation, r = 0.6722, p<0.01, Fig. 9 b). This striking result lead us to further carried out social interaction test in order to understand how pharmacological inhibition of COX-1 and EP1 receptor during HH affect social behaviour. Exposure of rats to HH for 7days (p<0.001) signi cantly reduced the time spent with social stimuli. Treatment withValS (Two-way ANOVA, F 3, 40 = 8.778, p<0.05) and SC19220 (p<0.01, Fig. 9 c) during 7HH increased time spent with social stimuli. It is clearly indicated in (Fig. 9 d) thatdirect contact of test rat with social stimuli was signi cantly elevated after treatment with ValS (F 3, 40 = 4.679, p<0.05) and SC19220 (p<0.05), which was reduced in 7HH (p<0.05) exposed rats. During Trial 2, 7HH exposure decrease time spent with novel animal and both treatments showed increased time spent with novel animal although this difference was not signi cant.7HH group animals spent signi cantly higher time with familiar rat (Two-way ANOVA, F 3, 40 = 6.664, p<0.05, Fig.9 e, n=6), whereas treatment with ValS (p<0.05) and SC19220 (p<0.05) signi cantly reduced the time spent in the chamber containing a familiar rat. Bonferroni's post hoc test clearly depict that (Fig. 9 f) contact behavior of test rat with novel stimuli was signi cantly reduced (F 3, 40 = 6.498, p<0.05) after 7HH exposure and treatment with ValS and SC19220 increased it (not signi cantly). Contact behavior with familiar was increased in rats exposed to HH for 7 days, which was reduced after treatment with both inhibitors, although not signi cant. Time spent in the open arm of EPM was signi cantly lower in 7HH (One-way ANOVA, F 3, 20 = 7.508, p<0.01) exposed group and increased after administration of ValS (p<0.05) and SC19220 (p<0.05, Fig. 9 g) during 7HH exposure.
Interpretation can be drawn from these results that COX-1/EP1 axis is evident during HH exposure right from 1 day. COX-1 inhibition reduced the expression of EP1 R indicate it to be the active downstream receptor and their strong correlation strengthen this. We could also found that 7HH induced alteration in social behavior in rats was reproducibly mitigated after COX-1 inhibitor (ValS), and EP1 antagonist (SC19220) probably by reducing anxiety like behavior.
Administration of COX-1 inhibitor and EP1R antagonist differentially rescued HH induced alteration in expression of BDNF, Serotonin and pCREB.
Bene cial effect of ValS and SC19220 on social interaction as well on anxiety like behaviour and alteration in levels of BDNF, Serotonin and pCREB during HH lead to this experiment. We tested the effects of ValS and SC19220 on neurotrophic factor like BDNF and we found signi cant up-regulation in the expression of BDNF after COX-1 inhibition (F 3, 12 = 8.613, p<0.05, Fig. 10 a) during 7HH exposure as compare to only 7HH group. EP1R antagonism also increased BDNF expression as compare to 7HH but the elevation was not signi cant. On the other hand Serotonin (F 3, 20 = 2.243, Fig. 10  Taken together, results showed that the inhibition of COX-1 and antagonist of EP1R rescued the perturbation of hippocampal microenvironment at some extent by elevating the expression BDNF and pCREB in DG but levels of serotonin in the hippocampus remained unchanged.

Inhibition of COX-1 as well EP1R pathway during HH stress boosts neuronal cell proliferation in DG
To test the hypothesis that the ameliorative behavioural effects of COX-1 inhibitor (ValS), and EP1 antagonist (SC19220) could be associated with adult hippocampal neurogenesis, we further tested their e cacy on HH induced de cits of neuronal proliferation in DG. Representative photomicrographs of progenitor cell marker BrdU and Ki-67 as well as DCX labelled neurons among groups are displayed (Fig.   11 a and b). Quantitative analysis of the number of BrdU labeled cells represents that 7HH exposure (Oneway ANOVA, F 3, 20 = 8.876, p<0.01, Fig. 11 c) severely reduced their number in DG, treatment with ValS (p<0.05) and SC19220 (p<0.01) elevated their number. Similarly, Ki-67 labeled cells were also decreased after 7HH exposure (p<0.05), but successfully increased after treatment with ValS (F 3, 20 = 4.327, p<0.05) and SC19220 (p<0.05) during 7HH exposure indicating increase in proliferative cells in DG upon treatment. DCX labeled neurons at the proliferative stage were signi cantly reduced at 7HH exposure (p<0.001), increased after treatment with ValS (F 3, 20 = 10.50, p<0.05) and SC19220 (p<0.01, Fig. 11 d) signi cantly. Similarly, DCX labeled cells at the post-mitotic stage were also dramatically reduced (F 3, 20 = 14.54, p<0.001) upon 7HH exposure, this reduction was mitigated after treatment with ValS (not signi cant) as well with SC19220 (p<0.01) indicating rescue of immature neuron at proliferative and post-mitotic stage but not at intermediate stage.
Representative immunomicrograph indicating higher number of co-labelled cells (BrdU+DCX+) (NeuN+DCX+) in DG after treatment with ValS and SC-19220 (Fig. 11 e) re ecting amelioration of maturation de cits caused during 7HH stress. No difference in (BrdU+NeuN+) as well (BrdU+ GFAP+). Although, BrdU+SOX2+ cells were decreased after 7HH exposure (p<0.01) and signi cantly rescued after injection of ValS (One-way ANOVA, F 3, 20 = 9.676, p<0.05) and SC19220 (p<0.01, Fig. 11  Pharmacological inhibition of COX-1 and EP1R during 7 days of HH exposure reverts glial activation in DG Activation of COX-1 as well EP1R promoted neuro-in ammation as well as microglia mediated neurotoxicity, we also reported COX-1 dominance in microglia in HH condition. EP1 gene ablation or selective antagonist shown to blunt microglial mediated in ammatory response. We planned our next experiment to answer the question about how COX-1/EP1R antagonism during HH can modulate morphology of microglia and astrocytes as they are prominent contributor to neurogenic niche. Representative immune-micrographs of Iba-1 and GFAP staining in DG among different groups (Fig. 12 a  and  dramatically altered after 7HH exposure. Pharmacological intervention with ValS and SC19220 during 7HH exposure also rescued parameters like Shoenen rami cation index (p<0.05, p<0.05), microglial soma area (p<0.05, p<0.01), and number of branch endpoints (p<0.05, p<0.001, Fig. 12 c) respectively. This indicates that ValS and SC19220 blunted microglial activation during 7HH.
We also checked the astrocyte intensity via GFAP staining in DG where exposure to HH for 7days (F 3, 20 = 6.612, p<0.01) signi cantly elevated the mean intensity of GFAP in DG, this was signi cantly mitigated after treatment with ValS (p<0.05) and SC19220 (p<0.05). Similarly, Number of resting astrocytes in DG was also reduced after 7HH exposure (F 3, 20 = 7.368, p<0.01) but ValS and SC19220 was not able to increase them signi cantly. The number of activated astrocyte were also increased considerably after 7HH exposure (F 3, 20 = 6.612, p<0.01) in DG, administration of ValS (p<0.01) and SC19220 (p<0.01, Fig.   12 d) reverted their activation signi cantly.
Taken together, we can conclude that ValS and SC-19220 effectively rescue activation of microglia and astrocyte followed by amelioration of neuroin ammatory response. This may lead to the conclusion that COX-1 inhibition as well EP1R antagonist protecting the perturbed neurogenic niche via blunting the activation of glial cells as both COX-1 and EP1 R expression was evident in both cells specially microglia.
HH induced NLRP3/NFkB mediated neuroin ammation was reverted by speci c COX-1 and EP1R inhibition HH exposure upsurge the expression of NLRP3/NFkB pathway and pro-in ammatory cytokines in the DG and COX-1, EP1R established to increase during in ammatory stimuli, so we next investigated the ability of COX-1 inhibitor and EP1R antagonist to blunt neuroin ammation during HH condition. We found elevation in the phosphorylation of NFkB at subunit 536 at 7HH (F 3, 12 = 6.021,p<0.01, n=3) exposure which was signi cantly reduced after administration of ValS (p<0.05) and SC19220 (p<0.05, Fig. 13 a).
These results indicate that HH induced neuroin ammation was mediated by NLRP3/NFkB pathways which was mitigated after treatment with COX-1 inhibitor and EP1R antagonist. Reduction of proin ammatory cytokine was visible when COX-1 and EP1R elevation blunted after treatment with both pharmacological agents.

Discussion
In this study we explored social interaction behavior and downstream molecular mechanisms under chronic HH exposure. Social memory is, in part, stored in the hippocampus and reported to affect during environmental stress like heat, sleep deprivation, and neonatal hypoxia [39,40,41,42]. HH induced de cits in social interaction behavior at 7HH exposure which is consistent with the growing understanding that external stress may also cause such de cits. Although neuroimaging studies suggest that multiple brain regions are impacted by HH exposure, the hippocampal formation is frequently identi ed as a site of injury during HH [43,44]. In rats, some early cytotoxic lesion and studies concluded that the hippocampus is dispensable for recognizing a familiar conspeci c [45,46]. HH induced alteration in hippocampus might be the reason behind alteration in social interaction behavior in rats. Another interesting observation where we found anxiety-like behavior in rats as indicated by tract-plots during social behavior, more movement in the chamber, but lesser contact behavior with the novel subject.
Neurodegeneration in both the amygdala and hippocampus after HH exposure have well established [18,47] which make rats vulnerable to anxiety. During many psychiatric disease states, anxiety and social de cits are co-expressed [48,49,50]. Our observation that HH exposure for 7days severely affects the social interaction behavior and induce anxiety like behavior in rats provided a unique opportunity to explore underlying molecular mechanism in HH. We tried to address this question by examining effects of HH on microenvironment of hippocampus particularly DG. BDNF and Serotonin both regulate development and plasticity of neuronal circuits that are involved in mood dependent behavior like social interaction and anxiety [ 51,52,53]. We found dramatic decrease in their expression upon exposure to HH that might be causing de cits in the above mentioned behavior. Phospho CREB directly regulate BDNF signalling and that also found to be decreased during HH [54]. As we found disturbed hippocampal microenvironment and very less information available regarding the effect of HH on stages of adult neurogenesis in rats. We next evaluated and found decrease in neuronal progenitor cells upon 7HH exposure in DG. However, immature neurons showed marked reduction from day 1 of the exposure. After 7HH exposure, reduction or delay in the maturation of new-born neurons of the DG region observed. Overall, HH exposure at 25000ft elicits detrimental effect on progenitor, proliferative, immature neurons in DG. In other hypoxia-related studies like intermittent hypoxia witnessed contradictory contributions towards the generation of new-born neurons in DG, whereas ischemia majorly suppresses adult neurogenesis [55,56,57]. Whereas, there is a study where number of Ki-67 and DCX+ hosphor cells elevated upon high altitude exposure in rats at 3450m via elevation in VEGF signaling [58]. Similarly, another report where hypobaric hypoxia exposure at 11000m for 3hr elevated the expression of NeuroD1 in the hippocampus [59]. The discrepancy can be explained as altitude in our condition is 7260m and for 1, 3 and 7days, which is harsher for neuronal generation in DG and VEGF elevation reported to be in pathological range [19]. Like other stresses, HH exposure also increases glucocorticoids and corticosterone in the bloodstream and brain that might be one of the reasons behind the impairment in adult neurogenesis [60,61]. Decrease in hippocampal BDNF, Serotonin and pCREB directly or indirectly causes de cits in adult neurogenesis in subgranular zone (SGZ) in rodents we observed similar string of event during HH exposure [ 62,63,64].
There are reports where proliferation blocker induced decline in BrdU and BrdU/NeuN labelled cells in DG abolished the effect of social interaction [65,66] suggests an association of adult neurogenesis and social behavior in rats, which was further validated in our study. Decrease in hippocampal BDNF, Serotonin and pCREB directly or indirectly causes de cits in adult neurogenesis might be the causal factors behind the alteration in social memory and induction in anxiety like behavior after HH exposure [62,64,67,68]. Increase in adult neurogenesis in DG improves social interaction behavior with reduced anxiety [67,69]. Besides, alteration in neuronal progenitor proliferation and dendritic development of newborn neuron resulted in elevated anxiety like behavior [70]. Decrease in adult neurogenesis similar strings of events observed in our study.
Neuro-in ammation speci cally via NLRP3 leads activation of glial and causes de cits in adult neurogenesis both observation were evident in our stress [68,71]. NLRP3 and NFkB signalling pathways were shown to activate upon HH exposure [22] supports our observation where we showed their up regulation in DG. Reduction in BDNF dependent neurogenesis was simultaneous with induction in NLRP3 dependent in ammation similarly shown earlier [72]. Dramatic increase in levels of pro-in ammatory cytokines followed by microglial activation made it imperative to target in ammatory pathways in an attempt to rescue brain neurogenic capability. COX pathway is one of the major pathways for the contribution to the process of in ammation evident by its presence in microglia and neuron both [73]. In this study, an interesting result observed where COX-1 inhibitor also decreased the expression of the EP1 receptor at 7HH exposure. We found strong correlation in the expression of COX-1 and EP1R, so we thought COX-1 might be using PGE2 receptor EP1 for its downstream effects. There are several supporting literature regarding the role of COX-1 mediated PGE2 synthesis in the induction of social avoidance. Moreover, EP1 is critical for the anxiogenic effect of repeated social defeat [74]. Both COX-1 and EP1 antagonist reduced the anxiety-like behavior similarly with the study where EP1 de cient mice showed decreased anxiety like behavior.
Fenamate (NSAIDS) which block COX-1 also were found to effective in supressing NLRP3 induced neuroin ammation and associated memory de cits in Alzheimer's disease in rodent models [14]. ValS also reduced the expression of NLRP3 during HH. During LPS induced neuroin ammation, COX-1 gene deletion reported to regulate hippocampal neurogenesis positively [12] during HH COX-1 inhibition rescued neurogenesis de cits at some extent. There are ndings, which also reported that PGE2-EP1 signaling suppresses midbrain dopamine neurons and regulates impulsive behaviors under acute stress [75,76] . Interestingly, We found that the COX-1 speci c inhibitor (ValS) and EP1 (SC19220) antagonist rescued social interaction de cits at 7HH exposure. We earlier reported COX-1 dominance in glial cells specially microglia and in this study we found dominance of EP1 R in glial cells. COX-1 and EP1 induction in activated microglia is reported to upsurge Ca 2+ levels thus disrupts Ca 2+ homeostasis leads to neurodegeneration and de cits in integration and maturation of newborn neurons in DG [77,78] .
Therefore, it is plausible that antagonism of COX-1 / EP1 could mitigate disturbance in Ca 2+ levels and rescued number of NPCs in the DG after injection of ValS and SC19220 to the rats during 7HH.There are not many studies that can elaborate on the effect of COX-1 on DCX labeled neuron directly, but COX-2 being an in ammatory marker has shown to be both negatively and positively regulating DCX expression in DG [79,80]. Importantly, depletion of intermittent progenitor cells due to LPS induced microglial activation rescued after COX-1 and COX-2 or only after COX-2 inhibitor [81,82]. To best of our knowledge, it is the rst study that has directly shown the effect of COX-1 inhibitor on DCX labeled neurons in DG. We earlier reported COX-1 neurotoxicity to be microglial dependent, hence COX-1 inhibition reverted microglia activation and rescued adult neurogenesis de cits. Similarly SC19220 also decreases microglia and astrocyte activation in DG, supported by an observation where EP -/mice showed ~70% reduction in hippocampal microgliosis [83]. NLRP3/ NFkB mediated neuro-in ammation was also decreased after COX-1/EP1R blockade results in decreased microglial activation and associated neuroin ammation.
These observations are inconsistent with the ndings where alleviation in expression of NLRP3/NFkB reduced microglial activation and proin ammatory cytokines [84,85]. Both ValS an SC19220 rescued microglial and astrocyte activation as their (COX-1/EP1) predominance in those cells thus ameliorated neurogenic niches perturbations. Consistent with our observation, there is a fascinating study where, EP1-/-mice suppressed microglial activation and rescued BrdU positive cells [86] . Our observation that EP1 antagonism effectively decreased pro-in ammatory cytokines in DG strengthen this premise further.

Conclusion
The present study demonstrates the following signi cant phenomena: The exposure to HH alters social behavior along with increase in anxiety like behavior in rats. HH disturb the microenvironment by altering the expression of BDNF, Serotonin and pCREB followed decreased proliferation of neuronal progenitor as well maturation of new born neurons in DG. HH exhibits upsurge in the NLRP3/ NFkB signalling pathway in the hippocampus particularly in DG accompanied by activation of glial cells and in ammatory milieu culminate in neurogenesis de cits. Along with COX-1, PGE2 receptor EP1R is up-regulated in the hippocampus speci cally in the glial cells in DG and their inhibition during 7HH exposure rescued alteration in the levels of BDNF, pCREB. Both pharmacological approach mitigated adult neurogenesis de cits, decreased activation of glia cells, NLRP3, NFkB signaling molecules and associated neuroin ammation, ultimately rescued social interaction de cits along with reduced anxiety like behavior.
COX-1/ EP1 speci c inhibition proved more bene cial in HH induced pathology, provides therapeutic alternatives to non-selective nonsteroidal anti-in ammatory drugs (NSAIDs), which has side effects in prolonged use. Various modalities that can enhance neurogenesis can be suggested to people before the ascent to high altitude, hence minimize alteration in associated behavior, like mood, social interaction, and cognition.