Chronic social defeat stress induces meningeal neutrophilia via type I interferon signaling

Animal models of stress and stress-related disorders are also associated with blood neutrophilia. The mechanistic relevance of this to symptoms or behavior is unclear. We used cytometry, immunohistochemistry, whole tissue clearing, and single-cell sequencing to characterize the meningeal immune response to chronic social defeat (CSD) stress in mice. We find that chronic, but not acute, stress causes meningeal neutrophil accumulation, and CSD increases neutrophil trafficking in vascular channels emanating from skull bone marrow (BM). Transcriptional analysis suggested CSD increases type I interferon (IFN-I) signaling in meningeal neutrophils. Blocking this pathway via the IFN-I receptor (IFNAR) protected against the anhedonic and anxiogenic effects of CSD stress, potentially through reduced infiltration of IFNAR+ neutrophils into the meninges from skull BM. Our identification of IFN-I signaling as a putative mediator of meningeal neutrophil recruitment may facilitate development of new therapies for stress-related disorders.


One sentence summary:
Type I interferon sensing neutrophils accumulate in meninges of psychosocially stressed mice via skull bone marrow channels and are associated with the negative behavioral sequelae of stress; blockade of this pathway inhibits neutrophil trafficking and improves behavioral outcomes.

Results:
CSD increases meningeal and blood neutrophil abundance.
We tested the extent to which meningeal neutrophil numbers reflect peripheral blood neutrophil levels, modeling the effect of condition (HC vs. CSD) and blood neutrophils on meningeal counts (Figure 1G).We found no relationship between blood and iv -meningeal neutrophils.

CSD increases neutrophil trafficking between skull BM and meninges.
We used tissue clearing to visualize GFP + cells in the channels connecting skull BM to the meninges (Figures 2G-H).There were more GFP + neutrophils per vascular channel in CSD mice compared to HC (Figure 2I: *p < 0.05, t = 2.4, df = 7), suggesting this route may be important for neutrophil trafficking into the meninges.
scRNAseq was used to profile meningeal immune cells from behaviorally stratified HC and CSD mice (Figures 4A-B).Clustering (leidenalg; see Supplementary Methods) identified 20 cell clusters which were manually annotated based on expression of canonical lineage genes (Figures 4C, S6A-C).There was a large cluster of neutrophils, identified by high Cxcr2 expression, which did not appear to be undergoing local cell proliferation (Figures S6D-E).Consistently with our IHC and flow cytometric data, the meninges from CSD-stressed mice showed a relatively greater proportion of neutrophils compared to HC (Figure 4C, lavender bars).
Subclustering of neutrophils alone identified 6 subclusters with roughly equal proportions of each subtype in both HC and CSD meninges (Figure S7A).Based on expression of genes related to primary, secondary, and tertiary granule formation, as well as suppression of transcriptional machinery (Figure S7B), these subclusters likely correspond to states of increasing neutrophil maturation [29][30][31] .Pseudotime analysis of differentially expressed genes across the subclusters supported this hypothesis (Figure S7C).

Effects of CSD stress on neutrophil cell size and actin polymerization.
Gene set enrichment analysis (GSEA) of differentially expressed genes (DEGs) in the pooled neutrophil cluster uncovered 80 biological pathways associated with CSD stress (Figure S8). 8 of the 41 positively enriched pathways indicated CSD was associated with altered neutrophil cell size and/or actin polymerization, e.g. the gene ontology (GO) pathway "regulation of cellular component size" (Figures S9A-B).Chronically high demand for neutrophils leads to the release from BM of immature cells 32,33 , which are both volumetrically larger 34 and more rigid 35 than mature neutrophils.Given this and our GSEA results, we hypothesized there would be an increase in enlarged BM-derived neutrophils in the meninges.
Pathway analysis also implicated increases in the GO pathway "cytokine production" in stressed animals, with increased expression of Cxcr2 in CSD neutrophils (Figures S10A-B).However, we did not detect changes in either the frequency (%CXCR2 + ) or protein expression (CXCR2 MFI) in any examined tissue (Figures S10C-D).We did observe an increased percentage of neutrophils in CSD meninges, 6 skull, tibia, and blood (****p < 0.0001, F(4,78) = 44.8;post hoc: t = 7.9, ****p < 0.0001), consistent with our other findings and with stress-associated increases in BM neutrophil levels described elsewhere 15 .We also noted an increase in CSD CXCL1 plasma concentration (**p < 0.01, U = 9), apparently due to increased mRNA expression in the liver (*p < 0.05, U = 8), and decreased Cxcl12 expression in CSD tibia (*p < 0.05, t = 2.3, df = 17).Collectively, this would promote egress of neutrophils into the blood stream, i.e. through CXCL1 binding to CXCR2 and decreased CXCL12 binding to CXCR4 (Figure S10E).Finally, detection of ROS by intracellular flow cytometry provided no evidence for increased neutrophil ROS production on a per cell basis, consistent with the GO pathway "cell redox homeostasis" (Figure S11).

Enriched IFN-I signaling in meningeal neutrophils.
Our GSEA results also identified enrichment of the GO pathway "response to type I interferons" (Figures S8, S12A), which we pursued further given extensive literature supporting a role for IFN-I in promoting depressive-like behavior 5,6,36 .Across all meningeal cell clusters, leading-edge genes Interferon induced transmembrane proteins (Ifitm)-2 and Ifitm3 showed highest levels of expression in neutrophils and monocytes (Figure S12B).Ifitm2 and Ifitm3 expression was strongly increased in CSDstressed meningeal neutrophils generally (Figures 4D, S12C) and in several neutrophil subclusters (Figures 5A-B).In contrast, monocyte expression of Ifitm2 and Ifitm3 did not show stress-associated changes (Figure S12C).
IFN-I sensing meningeal neutrophils could migrate to meninges from peripheral BM tissue (via the circulation), or directly from skull BM [26][27][28] .We would predict a decrease in IFNAR + neutrophils in the source tissue as they leave to traffic to the meninges.IFNAR + neutrophils were present in blood, skull and tibial BM (Figures 5C,  S13A).Comparison of HC vs CSD tissue revealed a 1.7-fold decrease of IFNAR + neutrophils in CSD mice that was specific to skull BM (Figure 5D: *p < 0.05, t = 2.7, df = 17).This result, along with the neutrophils seen in skull-to-meninges vascular channels (Figure 2I), suggest skull BM as the source of stress-induced meningeal neutrophils.
IFNAR + neutrophils are a relatively mature pool of BM neutrophils.
We noted three distinct populations of neutrophils -IFNAR neg , IFNAR lo , and IFNAR hi -with tissue-specific distributions and expression of activation markers (Figure S13).In blood, the proportion of IFNAR lo and IFNAR hi neutrophils was approximately equal, with no detectable IFNAR neg neutrophils.Bone marrow IFNAR + neutrophils had higher side scatter complexity (SSC: skull, ****p < 0.0001, F(2,54) = 17.0; tibia, ****p < 0.0001, F(2,54) = 12.8)-a phenomenon corresponding to increased granule content and/or more complex nuclear morphology reflecting cell maturation.Conversely, the majority (>94%) of BM neutrophils were IFNAR neg , consistent with an expected reservoir of immature cells in this niche.We also examined MFIs in each of the three IFNAR neutrophil subtypes.Ly6C is expressed in response to IFN-I signaling 37 .Consistently, Ly6C MFI was highest in IFNAR hi neutrophils from all three tissues (Figure S13C We assessed whether blockade of IFN-I signaling rescues the negative behavioral sequelae associated with CSD stress in LysM gfp/+ mice by administering an IFNAR-blocking antibody, as shown in Figure 6A.As expected, in the USM test for anhedonia, more HC+control antibody (IgG) mice marked compared to the CSD+IgG group (Figure 6B: **p < 0.01, χ 2 = 9.5; post hoc *p < 0.05).Anti-IFNAR treatment rescued the phenotype, with no difference in marking frequency between HC+IgG and CSD+IFNAR groups.Anxiety data for the CSD+IFNAR blockade group was nonnormally distributed, with distinct groups of animals that marked (+) or did not mark (-) in the USM test, so we treated these animals separately in our remaining analyses.There was a significant difference in the OF task between HC+IgG and CSD+IFNAR(-), where CSD+IFNAR(-) animals were more anxious.As before, behavior was rescued in the CSD+IFNAR blockade group, with no significant difference in behavior from HC+IgG (Figure 6C: *P < 0.05, H = 9.1; post hoc: *p < 0.05, Z = 2.8).

IFNAR blockade attenuates neutrophil migration into, but not B cell egress from, meningeal tissue in CSD stressed mice.
Hand counting of meningeal neutrophils from these mice showed an expected increase in the CSD+IgG group, which was present irrespective of distance from blood vessel (Figure 6D: ***p < 0.001, F(3,78) = 7.9; post hoc **p < 0.01, t = 3.8).The CSD+IFNAR(-) group also showed an increase in meningeal neutrophils independent of distance from a blood vessel (**p < 0.01, t = 4.1).There was no significant difference between the HC+IgG group and the CSD+IFNAR(+) group; rather, there appeared to be a decrease in meningeal neutrophils >10 μm from a blood vessel in CSD+IFNAR(+) mice.
Unexpectedly, there were no effects of anti-IFNAR treatment or CSD on peripherally circulating WBC populations, including both neutrophils and B cells (Figure S14B).Further examination of our data suggested an effect of repeated injection stress on circulating WBCs in non-defeated control mice (Figure S15).Importantly, no effect of injection stress was evident in the meninges.

Reduced expression of chemorepellent factors in CSD stress may permit neutrophil entry into the leptomeninges.
Expression of class 3 semaphorins was examined in microarray data from meningeal tissue.Sema3b expression was significantly reduced in CSD mice compared to HC (Figure 7C: LFC = -0.30,*unadjusted P = 0.050).

Discussion
Our data robustly demonstrate increased neutrophils in the meninges following CSD stress, and that neutrophilia correlates with a negative behavioral response to chronic stress (Figures 3, S5).Moreover, we have replicated the blood neutrophilia described in other preclinical stress models 15,38 and in people with MDD 18 -though, interestingly, neutrophil levels in the meninges and blood were not significantly related (Figure 1E).These results are summarized in Figure 7A.Whereas acute stress led to a rapid increase in blood neutrophils (i.e., within 2 h of a single defeat encounter), the effect in meninges was quite specific-neither acute nor non-psychological stress recapitulated the phenotype (Figures 1H, S15A).This is consistent with a growing body of literature suggesting that the meningeal niche is governed by processes distinct from the peripheral immune system 21,25 .
It is not presently clear exactly how meningeal neutrophils influence behavior given no evidence for their entry into the brain parenchyma (Figure S4).However a mechanism for neutrophil entry into the subarachnoid space (SAS) via arachnoid cuff exit (ACE) points was recently described 39 ; this occurs via reduced expression of chemorepellent class 3 semaphorins by arachnoid barrier cells (ABCs; Figure 7B).Any soluble factors released from neutrophils within this space have the capacity to interact directly with the brain.We reexamined our data from a previously published gene expression microarray on meningeal tissue 21 in an attempt to address whether CSD stress influences semaphorin expression.Sema3b expression was reduced in CSD mice compared to HC (Figure 7C), suggesting permissive conditions for neutrophil entry into the SAS.Future efforts to demonstrate this pathway conclusively will be important.
We have previously shown that B cell depletion increases both IFN-I signaling and meningeal neutrophil levels in Cd19 -/-mice 21 , which are deficient in B cells 40 .Here, we replicated the depleting effect of CSD stress on meningeal B cells and demonstrated that CSD stress leads to an enriched IFN-I signature in neutrophils but not in other meningeal cell types ( Figures 4E, S12).Our unbiased identification of this pathway in multiple CSD cohorts using multiple experimental techniques is further strengthened by consistent evidence from humans, non-human primates, and rodents that IFN-I is sufficient to induce depression and depressive-like behavior 5,6,36 .
Differential gene expression (DEG) analysis of our single-cell RNA sequencing data indicated altered innate immune system activation across the neutrophil cluster (Figure 4D).For example, the most strongly upregulated transcript in CSD meningeal neutrophils was Wfdc17, which inhibits NFκB-mediated microglial activation 41 and is proinflammatory in monocytes 42 .There was also evidence for altered proinflammatory leukotriene signaling: Arachidonate 5-lipoxygenase-activating protein (Alox5ap) is an accessory protein required for leukotriene synthesis, and Translocator protein (Tspo) mediates neutrophil chemotaxis via leukotriene B4 (Ltb4) receptor signaling 43,44 .Conversely, decreased C1qa expression following stress may hamper phagocytosis and clearance of dying neutrophils by nearby macrophages, leading to sustained local inflammation 45,46 .
Expression of Ifitm2 and Ifitm3, the two leading-edge genes identified in CSD neutrophils from the "response to type I interferons" GO pathway, has been reported in a mature, hyperinflammatory neutrophil subtype observed in blood from humans and mice 30,47,48 .Characterization of IFN-I receptor (IFNAR) expressing neutrophils in different tissue compartments revealed highest Ly6C expression in IFNAR hi neutrophils, which-in BM-appeared to also be more mature (Figure S13C).Consistently, qualitative assessment of IFNAR + compared to IFNAR -skull neutrophils showed more hyper-segmentation of the nucleus-a feature of neutrophil maturation (Figure 5C).High Ly6C expression, as seen in IFNAR hi neutrophils, is associated with greater proinflammatory capacity in monocytes 49 .This suggests IFNAR + neutrophils are a distinct, relatively mature BM subset that may also be more proinflammatory.
We tested whether IFN-I-sensing (i.e., IFNAR + ) neutrophils extravasate into the meninges from the skull given: 1) our observation that CSD leads to increased neutrophil trafficking in vascular channels connecting skull BM to the meninges (Figures 2G-I), and 2) no evidence to suggest local meningeal proliferation (Figure S6D).Characterization of IFNAR + neutrophils from several tissues revealed CSDassociated depletion was specific to skull BM (Figure 5D).Presumably this reflects egress from skull BM into the meninges, consistent with recent work from multiple labs showing skull BM provides a specialized reservoir for innate immune cell recruitment to the brain in conditions of neurological damage or disease [26][27][28] , and supported by our findings in cleared tissue.To our knowledge, this is the first evidence for such a phenomenon under conditions of psychosocial stress.
IFN-I depletion in C57BL/6J mice normalized CSD-related meningeal neutrophil levels (Figure 6E).Our attempts to replicate these effects in LysM gfp/+ mice were less straightforward (see Limitations), though anhedonic behavior improved in the CSD+IFNAR treated mice (Figure 6B).Meningeal B cell levels were unaffected by anti-IFNAR treatment; specifically, CSD stress resulted in fewer B cells regardless of antibody treatment (Figure 6F).This would suggest B cell depletion precedes IFNARmediated neutrophil recruitment and is consistent with data from our time-course study in which meningeal B cell numbers decreased prior to an increase in meningeal neutrophils, i.e. by day 8 vs day 14 (Figures 1H, S2C).However, at present we also cannot exclude the possibility that B cell depletion and neutrophilia are completely independent events.
What processes drive B cell egress from, and neutrophil infiltration toward, the meninges following CSD stress?Whereas the mechanisms driving B cell depletion remain unclear, our previous work may shed light on the latter question.For example, we have shown that CSD stress leads to randomly occurring microhemorrhages in the neurovasculature 22,50 .As neutrophils are typically 'first on the scene' to sites of brain injury 51 , their active recruitment may facilitate progression, maintenance, or resolution of this stress-induced damage.Consistently, skull BM-to-meninges neutrophil trafficking has been demonstrated in hemorrhagic stroke, wherein blood brain barrier (BBB) damage is present 52 -though, notably, we find no evidence for infiltration of neutrophils or other leukocytes into the brain parenchyma of CSD-stressed mice 22,23 .
Another possibility is that chronic release of neutrophils into the bloodstream via repeated exposure to stress 15,16 drives formation of these microhemorrhages.For example, in neurological conditions, neutrophils are notorious drivers of bystander damage due to their release of toxic reactive oxygen species (ROS) [51][52][53] .Our GSEA results ("GO: detoxification", "GO: cell redox homeostasis"; Figure S8) prompted us to explore ROS production in blood and meningeal neutrophils.We saw no differences in ROS production between HC and CSD neutrophils on a per-cell basis (Figure S11), though the overall increase in ROS-producing neutrophils may exacerbate risk for bystander damage.10 Blood-derived CSD neutrophils also showed signs of being both more rigid and proinflammatory (Figure S9E).We observed increased phalloidin staining intensity in CSD blood neutrophils (Figure S9E), consistent with enhanced formation of filamentous actin (F-actin) 54 .Increased F-actin formation may cause neutrophil rigidity, thereby increasing the likelihood of neutrophil 'stalling' in narrow brain parenchymal capillaries.Indeed, we saw more GFP + neutrophils 'stuck' throughout the neurovasculature in CSD brains, and more neutrophil-like cells coprecipitating with neurovascular endothelia (NVE), despite aggressive exsanguination and tissue perfusion (Figure S4), consistent with another report 55 .Interestingly, in mouse models of Alzheimer's disease (AD), neutrophil stalling was stochastic, resulting in reduced blood flow to the brain and associated memory deficits 56 .Our observations of NVE damage were similarly random and could possibly stem from stalling-related bystander damage, e.g., through neutrophil-mediated ROS-release.
Importantly, depressive symptoms are common in neurological disorders like stroke 57 and AD 58 .This raises the possibility that altered neutrophil properties are a shared biological feature between neurological and psychiatric disorders; comparison of neutrophil phenotypes across brain disorders may provide fruitful insights into pharmacologically relevant treatment targets.In particular, we have identified IFN-I signaling as a putative mediator of meningeal neutrophil recruitment, which may worsen depressive-like symptoms (Figure 6).IFN-I signaling deleteriously impacts both neuronal injury after stroke 59 and synapse loss in AD 60 .Given the relationship between depressive-like behavior and IFN-I 5,6,36 , and epidemiological evidence that depression doubles the risk for dementia late in life 61 , future work in this area is warranted.

Limitations and strengths
While the social defeat paradigm is well-established and pharmacologically validated for the preclinical study of depression-and anxiety-like behavior 62 , a major limitation of this and similar studies is that they are done almost exclusively in adult male mice, though depression and anxiety disproportionately affect cis women 63 and transgender youth 64 .'Resident' CD-1 males commonly used in defeat paradigms to induce depressive-like behavior do not show aggression towards 'intruder' females unless manipulated to do so, either through surgical implant 65 or by daily application of male odorant to females 66 , and unfortunately these adaptations to the standard paradigm were not published until after we had begun data collection.
Another limitation of our study is that social defeat stress is obligatorily associated with some degree of fight-inflicted wounding.We mitigated this by regularly clipping the CD-1 aggressor's teeth, terminating defeat encounters before the 5 min duration of the stressor if an aggressive encounter involved a visible degree of biting, and removing CD-1 aggressors as stimulus animals if they were consistently hyperaggressive.We also shaved mice at the time of sacrifice to assess severity of wounds.We note that without this step, the degree of wounding is not necessarily obvious under the fur.Anecdotally, higher wound scores were associated with greater levels of circulating and meningeal monocytes.This may require a more nuanced approach when attempting to translate results from social defeat stress models into a clinical setting, as has been discussed elsewhere 67 .
Finally, depletion of IFN-I signaling showed pleiotropic effects that were straindependent.We cannot account for the variation seen in LysM gfp/+ CSD+IFNAR mice, though close inspection of HC data from different experiments suggests some individuals may be sensitive to the effects of repeated injection stress (Figure S15D).Our data also cannot rule out a role for IFNAR signaling in other cell types in behavior; future experiments using neutrophil-specific IFNAR knockouts will be important.
Our strengths include our whole organism approach to neutrophil dynamics.Triangulation and replication of our findings via multiple experimental approaches yielded strong evidence for chronic stress-induced meningeal neutrophilia.Moreover, data-led investigation of candidate mechanisms revealed IFN-I-mediated migration of neutrophils from skull BM to the meninges as a future target for translational study of stress-associated depression.

Methods
Animals: Strains used were either C57BL/6J male mice purchased from Jackson Labs (Bar Harbor, ME) or male LysM +/gfp offspring from C57BL/6J mice bred in our facility with LysM gfp/gfp mice, which strongly express GFP in neutrophils 68 (obtained from Dr. Dorian McGavern, NINDS).Upon arrival into the facility, purchased mice were randomly pair-housed in divided cages and given one week of acclimation to the facility.Mice bred in-house were weaned at 3 weeks of age into same-sex cages of littermates.Sex was assigned based on external anatomy.Animals were housed under a reverse light cycle (lights off 8:00 AM to 8:00 PM), with food and water ad libitum.At 8-10 weeks of age, mice were randomly assigned to an experimental group.All procedures were approved by the National Institute of Mental Health Institutional Care and Use Committee and conducted in accordance with National Institutes of Health guidelines.
Chronic social defeat (CSD): CSD stress was performed as previously reported 22 .Briefly, the 'intruder' test mouse was introduced into the home cage of an aggressive, CD-1 (Taconic; Rensselaer, New York) retired breeder.The two were separated by a perforated barrier and given 24 h to acclimate; the barrier allowed for olfactory, visual, and auditory communication, but not tactile contact.Each day for either 1, 2, 4, 8, or 14 consecutive days, depending on the experiment, the barrier was lifted, and agonistic encounters were allowed to occur for 5 m.Interactions were monitored by a trained individual to ensure the test mouse exhibited submissive behavior and conversely that the CD1 exhibited dominant behavior.Efforts to minimize physical damage were taken, i.e.CD1 mice were lightly anesthetized with isoflurane and incisors were trimmed prior to starting the social defeat paradigm, and on a weekly basis thereafter.Test mice were shaved and inspected for the presence of wounds at the end of the experiment; wounds were scored on a scale from 1-10 (1 = no injuries, 5 = combination of old and new bite marks, 10 = severe wounds).Animals with wound scores of 10 were excluded.
Behavioral phenotyping: All behavioral testing was done by both male and female experimenters during the dark phase of the light cycle, prior to a defeat encounter on that day.On the day of testing, mice were moved to a separate behavioral room with red lighting and acclimated for one hour prior to running the behavioral assays.Tests 12 were usually run on separate days, but when multiple tests were run on a single day, the animals were given an hour to recover in between tests.Behavioral tests were performed in the following order: Urine scent marking (USM): As described previously 24 .Briefly, mice were placed into a novel arena (50 x 50 x 50 cm) that contained a thick sheet of paper; one corner of this paper was 'spotted' with 50 μL of urine from multiple estrus females, and testing was performed with the lights off while the experimenter was outside the room.After testing, the sheets of paper were sprayed with ninhydrin and heated to indicate the presence of proteins, allowing for visualization of urine marking.Photos were taken of the sheets by an experimenter blind to group identity, then analyzed in ImageJ 69 .Reduced preference marking for the female scent is indicative of social anhedonia.
Open field (OF): Exploration of the novel open field arena, 50 x 50 x 50 cm in dimension, was performed under dim white lighting (~25 lux).Mice were placed in the middle of the arena; total distance moved, crosses to center, and time in center over a 10 min testing period whilst the experimenter was out of the room were later analyzed with automated tracking software (Clever Sys TopScan Suite) to eliminate potential human bias.Fewer crosses to center, reduced time in center, and reduced movement are all indicative of increased anxiety-like behavior.

Tissue collection:
For all studies, mice were euthanized 2 h after their final exposure to the defeat stressor.Tissue collection occurred between ~8am and noon, with modifications as indicated below.Venous blood was collected into EDTA tubes via puncture of the submandibular vein and kept on ice until processing.Retro-orbital injections were administered while mice were under light isoflurane anesthesia.After, mice were deeply anesthetized with isoflurane prior to perfusion with 35 mL room temperature PBS.When indicated, hindlegs were collected for tibial BM extraction.The head was decapitated, and intact skull cleaned with a scalpel to remove muscle and connective tissue.
Histology.LysM +/gfp mice received a retro-orbital intravascular injection of either DyLite 649-or DyLite 594-conjugated tomato lectin (TomL, Cat #DL-1178, Vector Labs), which was allowed to circulate for 5 min to label blood vessels before perfusion.In addition to PBS, mice were perfused with 10 mL of 4% cold paraformaldehyde.Skulls were postfixed in 4% PFA for 24 h at 4 o C before transferring to 25% sucrose solution for cryoprotection, dehydration, and preparation for imaging.
Single cell suspensions.WT mice were retro-orbitally injected with 4 μg CD45-FITC (Cat.#103108; Biolegend).Mice were then allowed to recover; after 25 min of circulation, mice were anesthetized lightly for venous blood collection.The decapitated skull was placed in cold HBSS + 0.1% BSA on ice until further processing.

Tissue clearing and analysis of skull-to-meningeal vascular channels:
We used the CUBIC tissue clearing method 70 on whole skulls from LysM +/gfp mice.Tissue was prepared as for histology except that samples were transferred to PBS instead of 13 sucrose after 24h fixation.Decalcification of bone prior to tissue clearing was achieved by incubation in decalcification solution (10% EDTA, 15% imidazole) for 5-7 days at 37°C with shaking 71 .The decalcification solution was refreshed once on day 3.Following tissue clearing, the whole skull was inverted and placed in a glass-bottom dish filled with fresh Reagent 2 70 and imaged using a Zeiss 780 confocal microscope fitted with 10x objective.2-5 images were collected at random locations for each skull and analyzed using IMARIS 9.7.For each image, both the number of vascular channels (labeled with TomL) and the number of discreet neutrophils (GFP + ) in a channel were counted.For each individual sample, the ratio of averaged neutrophils normalized to the average number of channels is presented.

Skull and tibial BM preparations:
To prepare for skull BM extraction, the dorsal calvarium was trimmed to be relatively flat, and meninges were removed under a dissecting microscope.Next, the skull was cut into small bone pieces with scissors in cold HBSS + 0.1% BSA.This entire slurry was transferred to a 70 μm cell strainer and mashed with the rubber end of a 3 mL syringe for approximately 2 min per sample.Tibia were prepared by first stripping away all tissue from the bone, then cutting the very top such that a 23g syringe needle could be inserted to flush out the BM into a tube of cold HBSS + 0.1% BSA.This was next transferred to a 70 μm cell strainer and mashed with the rubber end of a 3 mL syringe.For both kinds of BM, the resulting cell suspensions were then pelleted and prepared for flow cytometry.
Meningeal dissection and single cell suspension: Meninges samples were collected by first cutting around the lateral sutures of the skull; the dorsal skull and ventral skull were transferred to a fresh petri dish filled with cold HBSS + 0.1% BSA and kept on ice while pial and arachnoid meningeal membranes were gently picked off the entire outer surface of the brain into a second 'working dish' with Dumont #5 forceps (Cat.#RS-5058; Roboz, Gaithersburg, MD).Extra care was taken to avoid inclusion of choroid plexus from the 4 th ventricle.Once finished with the brain, skull pieces were transferred as necessary into the working dish to remove attached meninges; we avoided leaving skull pieces in the 'working dish' to minimize contamination with cells from skull BM.Upon completion of the meningeal dissection, samples were transferred to a fresh tube and cells were pelleted by centrifugation, then resuspended in 2mL of BSA-free HBSS supplemented with 2.5 mg/mL Collagenase D (Cat.#11088858001; Roche) and 12.5 μL of 0.5 mg/mL DNAseI (Cat.#L5002139; Worthington) for cell dissociation.The samples were incubated at 37 o C for 30 min, diluted with cold HBSS + 0.1% BSA, and mashed through a 70 μm cell strainer into single cell suspension.

Meningeal whole mount preparations and staining:
The dorsal skull was carefully removed to retain maximal attachment of the meningeal layers from tissue prepared for histology, and brains were returned to fresh 25% sucrose until sunk for further staining (see Supplemental Methods).Dorsal meninges were gently peeled from the skull as a single sheet, mounted ventral (brain) side up/dorsal (skull) side down onto slides and encircled with a Pap pen.
Meningeal immunohistochemistry: Meningeal whole mount samples from LysM +/gfp mice were dried, washed with PBS, blocked for 1 h in 4% normal goat serum in 0.4% Triton-PBS, and incubated in a humidity chamber overnight at room temperature with chicken anti-GFP (1:1000, Cat #13970, Abcam), diluted in 0.2% Triton-PBS with 2% normal goat serum.Approximately 18h later, the samples were washed 3 x 5m with PBS, and incubated for 2 h with Chicken IgY-Alexa Fluor 488 (1:500, Cat #150169, Abcam) in 0.4% Triton-PBS.Samples were washed 2 x 5m with PBS, given a quick rinse in deionized water, and counterstained with DAPI for 5 minutes.They were rinsed briefly again with deionized water then cover-slipped with PVA-DABCO (made in-house).
Image acquisition and analysis: Meningeal whole mounts were tile scanned using a 20x objective at 1024x1024 resolution and online stitching with a Zeiss 780 confocal microscope.Two independent investigators blind to treatment hand-counted confocal images of the meningeal whole mounts to quantify the density and location of neutrophils (characterized by high GFP expression, irregularly shaped nuclei, and a semi-round shape) within the tissue using ImageJ software 69 .There was excellent agreement in their counts (Pearson correlation, ****p < 0.0001, r = 0.937).Neutrophils were also examined in relationship to blood vessels; neutrophils > 10 μm from a blood vessel were provisionally called 'parenchymal,' whereas neutrophils ≤ 10 μm were called 'abluminal.' Neutrophils were otherwise considered intravascular.

Cell staining and flow cytometry:
To exclude dead cells, samples were stained with either Fixable Viability Dye eFluor™ 780 (Cat.#65-0865; eBioscience) for 10 min at room temperature at 1:2400 dilution, or with Zombie AQUA (Cat # 423102, Biolegend) for 15 min at room temperature at 1:100 dilution.The cells were then washed and blocked with 1 μL of normal goat serum (Cat #G9023-10ML; Sigma) and 1 μL of Fc block (Cat #101302; Biolegend) for 10m on ice.25 μL Brilliant Violet stain buffer (Cat #563794; BD Horizon) was then added, followed by an antibody master mix (Table S7).Cells were incubated on ice for 25 min, washed with PBS, and fixed in 2% PFA at room temperature for 15 min for analysis on either a BD LSR Fortessa or a Beckman Coulter CytoFLEX flow analyzer.Compensation was performed for each session using UltraComp eBeads (eBioscience 01-2222-42) conjugated to antibodies used in the sample panels.Viability dye and GFP + controls used cells instead of beads.Data were analyzed using FlowJo (BD) software with manual gating.Absolute cell counts were determined using CountBrite counting beads (ThermoFisher, catalog #C36950).See Supplemental Methods for details on imaging flow cytometry.
Meningeal scRNAseq: Meningeal scRNAseq data were acquired from 8 non-stressed HC mice and 4 stress-susceptible CSD mice as described previously 21 .In brief, live, nucleated, singlet cells (DAPI -DRAQ5 + ) were sorted on a BD FACS Aria Fusion into HBSS + 10% FBS prior to droplet encapsulation using 10x Genomics' Drop-seq platform (Chromium v2). 3 pooled samples of 4 mice each were generated (two HC pools, one CSD pool) and run on the same 10x Chromium chip; libraries were sequenced on Illumina NextSeq 550 and feature counts generated using Cellranger V2 pipeline.See Supplemental Methods for additional information.
Cytospin: Skulls were obtained and prepared for single-cell suspension.The cells were then fixed, pelleted onto thin coverslips using funnel centrifugation (Cat #10-354, Fisher HealthCare), and imaged using a Zeiss 780 confocal microscope fitted with 10x objective.
IFNAR-blocking assay: Anti-IFNAR (clone: MAR1-5A3, Cat # BE0241) and non-specific, IgG isotype control (clone: MOPC-21, Cat # BE0083) antibodies for repeated in vivo injections were purchased from BioXCell and diluted in InVivoPure pH 7.0 Dilution Buffer (Cat # IP0070) to a concentration of 5 mg/mL.Mice were i.p. injected with 1mg of antibody on day zero (d0), before the start of the defeat paradigm, and received 0.5 mg of antibody every third day thereafter, for a total of 5 injections per mouse.Defeats were done approximately 1-3 h after injections.LysM +/gfp mice were ear-tagged and randomly assigned to one of three groups: HC+IgG, CSD+IgG, or CSD+IFNAR.WT mice were ear-tagged and randomly assigned to one of four groups: HC+IgG, HC+IFNAR, CSD+IgG, or CSD+IFNAR.
Statistics: Prism 9.0.2 (GraphPad Software, LLC) was used for statistical testing and graphing of univariate analyses.Normality and equal variance were assessed, and appropriate tests were conducted thereafter.For simple, two-group comparisons, samples with non-equal variance were analyzed using a Mann-Whitney U test, whereas samples with equal variance were analyzed using a Student's t test.In almost all instances CSD-stressed mice failed to mark in the USM test; the behavioral output was thus better modeled statistically as a binary response (did or did not mark).We therefore used χ 2 tests for univariate analyses of group in the USM test; posthoc testing was performed with the chisq.posthoc.testpackage in R, using the Benjamini-Hochberg method for multiple comparisons corrections.For multiple comparisons of normally distributed data, ordinary one-or two-way ANOVAs were used for analysis.When a main effect was present, post-hoc analyses were conducted (Dunnett's or Tukey's, respectively).For multiple comparisons of non-parametric data, a Kruskal-Wallis test was used, and Dunn's test for multiple corrections was run for post-hoc analysis.Univariate comparisons are summarized as the mean ± SEM and considered statistically significant at p < 0.05.No power calculations were performed at the outset of the experiment.Please see Supplemental Methods for more information about statistical modeling.S1A-C).b) Mice were injected retro-orbitally with a fluorescently labeled CD45 antibody for exclusion of blood-exposed cells in the meninges.Simplified gating strategy showing identification of nonvascular neutrophils from meningeal tissue.Full gating strategy shown in Figures S1D-E. c) Flow cytometric analysis shows CSD stress causes an increase in meningeal neutrophils, defined as CD45iv -;CD11b + ;Ly6G + ;Ly6C int and (top) calculated as a percent of live CD45 + cells (nHC = 27, nCSD = 21), or (bottom) assessed for absolute, as opposed to relative, cell counts (nHC = 14, nCSD = 10).d) Flow cytometry analysis of blood neutrophils (CD45iv + ;CD11b + ;Ly6G + ;Ly6C int ) shows a robust increase following CSD stress in both the (top) percentage of neutrophils (nHC = 27, nCSD = 20) and (bottom) absolute counts (nHC = 8, nCSD = 4).e) No apparent relationship between blood and meningeal neutrophils; † values were square root transformed to improve normality.f) Schematic of acute vs chronic stress study; red arrows indicate time points in days at which mice were killed and tissue was harvested.g) Comparison of all cell types examined in this study showing equal and opposite directional fluctuations for neutrophils and B cells in both tissues.h) There was a main effect of the number of encounters for meningeal neutrophils, but only the CSD day-14 group showed a significant increase by post hoc analysis.In contrast, there was a significant increase in blood neutrophils overall and at each time point when compared to HC (subscript indicates days of defeat: nHC = 8, n1 = 7; n2 = 7; n4 = 9; n8 = 5; n14 = 6).Diagrams made with Biorender.HC = home cage, CSD = chronic social defeat stress.Data shown as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 2:
CSD stress leads to increased numbers of neutrophils in vascular channels connecting skull bone marrow to meninges.a) LysM gfp/+ mice were subjected to CSD stress.Before TomL intravascular (iv) injection, blood was drawn for flow cytometry (Figure SF3).Dorsal meninges and brain (Figure S4) were prepared for imaging.A separate cohort of mice was used for tissue clearing.Diagram made with Biorender.b) Flow cytometry data of blood shows both high GFP expression and SSC in neutrophils compared to other cell types (x axis is log scale, y axis is linear).c) Left: Representative image for dorsal whole-mount meningeal tissue.Scale bar = 1 mm.Right: GFP + neutrophils adjacent to a blood vessel; merge and individual channels.Scale bar = 10μm.d) Representative meningeal whole mounts from HC and CSD mice showing hand-counted meningeal neutrophils, normalized to indicated area.Blood vessels traced in Adobe Illustrator for visualization.Inset box represents approximate size of area shown in high resolution images from (c).e) Quantification of total meningeal neutrophils shows an increase with CSD stress.‡ natural log-transformed.f) Neutrophils were separated into three categories based on their location in the tissue: >10μm away from a blood vessel ("parenchymal"), ≤10μm away from a blood vessel ("abluminal"), and intravascular ("iv").CSD led to neutrophil elevations in all three subcompartments.
† log10 -transformed.g) Representative image from cleared HC skull showing vascular channels between skull bone marrow and the meninges.Scale bar = 100μm.Top: Merged image.Bottom: Individual channels for blood vessels and neutrophils.h) Representative image from a CSD mouse.i) Quantification of neutrophils per channel (normalized to number of channels) indicates increased egress from skull bone marrow following CSD stress (nHC = 3, nCSD = 6).j) No differences in the number of channels counted between groups.HC = home cage, CSD = chronic social defeat stress, TomL = tomato lectin, NPs = neutrophils, cMOs = classical monocytes, SSC = side scatter complexity, BM = bone marrow.Data shown as mean ± SEM. *p < 0.05.There were no differences between the HC+IgG group and the CSD+IFNAR group (nHC+IgG = 11, nCSD+IgG = 10, nCSD+IFNAR = 11).Hereafter we considered LysM gfp/+ mice in two separate groups -those that marked in the USM test (+, indicated with triangles) and those that didn't (-, indicated with circles).c) HC+IgG(+) and (-) mice showed a normal distribution for anxiety-like behavior and were collapsed into one group.CSD+IFNAR(+) mice showed control-like levels of exploration in the OF task for anxiety-like behavior.d) There was an overall effect of group on meningeal neutrophils.
Post hoc analysis indicated a CSD stress-induced increase that was normalized in the CSD+IFNAR(+) subgroup.† values were natural log-transformed to improve normality.e) In C57BL/6J wild type mice there was an IFNAR-mediated rescue in meningeal neutrophil accumulation following CSD stress, with expected differences in IgG control groups (nCSD+IFNAR = 7, otherwise n = 8).f) Anti-IFNAR treatment does not rescue the reduction in meningeal B cells seen with CSD stress.HC = home cage, CSD = chronic social defeat stress.Data shown as mean ± SEM. *p < 0.05, **p < 0.01.

Figure 7:
Proposed model for how chronic, but not acute, stress leads to dysregulation of the meningeal environment.a) At baseline, both skull and tibia bone marrow contain a small population of IFNAR + neutrophils (orange), further divisible into IFNAR lo and IFNAR hi populations.Relatively mature IFNAR hi cells express more Ly6C and may thus be more proinflammatory 37 .In blood, B cell numbers greatly outnumber neutrophils; in the meninges these two populations are roughly equal.Acute stress exposure leads to repetitive release of neutrophils into the blood but is not sufficient for accumulation of neutrophils in the meninges (Figures 1H, S15A).Prolonged exposure to psychosocial stress leads to a decline in meningeal B cells (Figure S2C) that precedes an increase in meningeal neutrophils (Figure 1H).Whereas CSD-associated meningeal neutrophilia appears to be mediated by IFN-I signaling, meningeal B cell depletion does not (Figures 6C,E).CSD stress causes random neurovascular damage 22 -potentially via stalled and rigid ROS-producing neutrophils in brain capillaries (Figures S4, S9E, S11)-which amplifies neutrophil recruitment from adjacent skull bone marrow (Figure 2G).b) Class 3 semaphorins (SEMA3) prevent migration of neutrophils across dural lymphatics into the leptomeninges at ACE points due to their chemorepellent properties 39 .Depletion of SEMA3 family expression leads to increased accumulation of neutrophils and other immune factors in SAS from whence they can directly interact with the brain to influence mood and behavior.c) Microarray analysis of SEMA3 family members in bulk meningeal tissue suggests decreased expression of Sema3b, which may permit neutrophil entry into the SAS (n = 7 per group).

Figure 1 :
Figure 1:Meningeal neutrophils are elevated following chronic, but not acute, social defeat stress.a) CSD mice were behaviorally phenotyped on days 10-13, and tissue harvested at day 14 (FiguresS1A-C).b) Mice were injected retro-orbitally with a fluorescently labeled CD45 antibody for exclusion of blood-exposed cells in the meninges.Simplified gating strategy showing identification of nonvascular neutrophils from meningeal tissue.Full gating strategy shown in FiguresS1D-E.c) Flow cytometric analysis shows CSD stress causes an increase in meningeal neutrophils, defined as CD45iv -;CD11b + ;Ly6G + ;Ly6C int and (top) calculated as a percent of live CD45 + cells (nHC = 27, nCSD = 21), or (bottom) assessed for absolute, as opposed to relative, cell counts (nHC = 14, nCSD = 10).d) Flow cytometry analysis of blood neutrophils (CD45iv + ;CD11b + ;Ly6G + ;Ly6C int ) shows a robust increase following CSD stress in both the (top) percentage of neutrophils (nHC = 27, nCSD = 20) and (bottom) absolute counts (nHC = 8, nCSD = 4).e) No apparent relationship between blood and meningeal neutrophils; † values were square root transformed to improve normality.f) Schematic of acute vs chronic stress study; red arrows indicate time points in days at which mice were killed and tissue was harvested.g) Comparison of all cell types examined in this study showing equal and opposite directional fluctuations for neutrophils and B cells in both tissues.h) There was a main effect of the number of encounters for meningeal neutrophils, but only the CSD day-14 group showed a significant increase by post hoc analysis.In contrast, there was a significant increase in blood neutrophils overall and at each time point when compared to HC (subscript indicates days of defeat: nHC = 8, n1 = 7; n2 = 7; n4 = 9; n8 = 5; n14 = 6).Diagrams made with Biorender.HC = home cage, CSD = chronic social defeat stress.Data shown as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 3 :
Figure3: Elevated neutrophil levels are associated with the negative behavioral sequelae of CSD stress.Dot and whiskers plots showing standardized effect sizes for neutrophil levels on behavioral outcomes (batch corrected for cohort).Left: Logistic regression models show negative relationships between blood and meningeal neutrophils with USM behavior.As neutrophil levels rise, mice are less likely to engage with the hedonic stimulus, i.e., odor from sexually mature females (nHC = 12, nCSD = 13).Right: Linear regression models indicate blood neutrophils are significantly associated with anxiety-like behavior.Specifically, as blood neutrophil levels increase, mice explored the OF arena less (iv -/iv + meningeal neutrophils: nHC = 15, nCSD = 17; blood: nHC = 10, nCSD = 11).See TablesS1-S2 for full statistics; see Figure S5 for comparison with LysM gfp/+ mice.HC = home cage, CSD = chronic social defeat stress, USM = urine scent marking, OF = open field.*p < 0.05.

Figure 4 :
Figure 4: Single cell sequencing of meningeal tissue validates CSD neutrophilia; differential gene expression (DGE) analysis reveals enrichment of pathways related to cytoskeletal processes.a) Timeline for sample collection and processing.Figure made with Biorender.b)WT mice were behaviorally phenotyped prior to single cell analysis; a subset with representative behavior in both the USM task for anhedonia and the OF task for anxiety-like behavior were selected.Top: HC animals that marked (+mark) and CSD animals that did not mark (-mark) were chosen.Bottom: larger, filled circles indicate individual animals used for single-cell analysis.c) Analysis of 10x Genomics single-cell data for meningeal tissue reveals 20 distinct immune cell clusters.Left: Visualization of recovered cells as a proportion of total cells recovered per group; lavender indicates neutrophils.(nHC = 8, nCSD = 4; see Methods).Plot shown previously21 .d) Volcano plot showing DGE between CSD and HC in the neutrophil cluster (excluding preneutrophil cluster).Indicated points represent DGE with LFC > 0.5 and FDR p < 0.001.Gene set enrichment analysis (GSEA) revealed several enriched pathways related to cell size and the cytoskeleton (see FigureS8).e) The Amnis ImageStream system was used to visualize cells stained with flow cytometry markers identifying neutrophils, phalloidin (to label actin) and Hoechst (to label nuclei).Top: Representative images acquired from individual meningeal neutrophils.Two differently sized populations emerged, as depicted (white bar = 'small' neutrophil diameter, black bar = 'large').Bottom: There were nearly 3x more enlarged neutrophils in CSD meninges compared to HC.No changes were evident in blood.For more details, see Figure S9.HC = home cage, CSD = chronic social defeat stress, BAM = border associated macrophage, PVM = perivascular macrophage, DBC = dural border cell, BF = brightfield.Data shown as mean ± SEM. *p < 0.05, ****p < 0.0001.

Figure 5 :
Figure5: Migration of IFNAR + neutrophils from skull bone marrow to the meninges may underlie the type I interferon neutrophil signature seen in CSD stressed mice.a) UMAP shows expression of Ifitm2 and Ifitm3, the leading-edge genes for enrichment of the GO: "Response to type I interferon" pathway in neutrophils, in CSD compared to HC animals.b) Dot plot showing gene expression in each neutrophil subcluster for all genes comprising this pathway.Expression is scaled to mean ± standard deviation.c) 60x magnification of skull bone marrow-derived neutrophils.Clockwise from top left: Merge, Ly6G + staining (cyan), IFNAR + staining (magenta), nuclei (DAPI, blue).Scale = 5μm.d) IFNAR + neutrophils, normalized to total neutrophils, for blood, skull, and tibia bone marrow.The population of IFNAR + skull bone marrow neutrophils was decreased in CSD stressed mice, and may represent a migration event to the meninges.See FigureS13for more detail.HC = home cage, CSD = chronic social defeat.Data shown as mean ± SEM. *p < 0.05.

Figure 6 :
Figure 6: Type I interferon receptor (IFNAR) blockade may improve CSD-stress related behavioral anhedonia and prevent meningeal neutrophil accumulation.a) Schematic showing drug-delivery schedule for IFNAR blocking antibodies.Mice were injected with either matched IgG control antibody or anti-IFNAR antibody on the days indicated with yellow arrows.Figure made with Biorender.b) USM results from LysM gfp/+ mice; as expected, CSD+IgG mice marked less frequently compared to the HC+IgG group.There were no differences between the HC+IgG group and the CSD+IFNAR group (nHC+IgG = 11, nCSD+IgG = 10, nCSD+IFNAR = 11).Hereafter we considered LysM gfp/+ mice in two separate groups -those that marked in the USM test (+, indicated with triangles) and those that didn't (-, indicated with circles).c) HC+IgG(+) and (-) mice showed a normal distribution for anxiety-like behavior and were collapsed into one group.CSD+IFNAR(+) mice showed control-like levels of exploration in the OF task for anxiety-like behavior.d) There was an overall effect of group on meningeal neutrophils.Post hoc analysis indicated a CSD stress-induced increase that was normalized in the CSD+IFNAR(+) subgroup.† values were natural log-transformed to improve normality.e) In C57BL/6J wild type mice there was an IFNAR-mediated rescue in meningeal neutrophil accumulation following CSD stress, with expected differences in IgG control groups (nCSD+IFNAR = 7, otherwise n = 8).f) Anti-IFNAR treatment does not rescue the reduction in meningeal B cells seen with CSD stress.HC = home cage, CSD = chronic social defeat stress.Data shown as mean ± SEM. *p < 0.05, **p < 0.01.