Intranasal Mycobacterium vaccae administration prevents stress-induced aggravation of dextran sulfate sodium (DSS) colitis

Brain, Behavior, and Immunity xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Brain, Behavior, and Immunity journal homepage: www.elsevier.com/locate/ybrbi

Intranasal Mycobacterium vaccae administration prevents stress-induced aggravation of dextran sulfate sodium (DSS) colitis

T

Mattia Amorosoa, Elena Kemptera, Tasnim Eleslamboulya, Christopher A. Lowryb,c,d,e, ⁎ Dominik Langgartnera,1, Stefan O. Rebera, ,1 a

Laboratory for Molecular Psychosomatics, Department of Psychosomatic Medicine and Psychotherapy, University Ulm, Ulm, Germany Department of Integrative Physiology, Center for Neuroscience, and Center for Microbial Exploration, University of Colorado Boulder, Boulder, CO 80309, USA c Department of Physical Medicine & Rehabilitation and Center for Neuroscience, University of Colorado Anschutz Medical Campus, Aurora, CO 80045, USA d Veterans Health Administration, Rocky Mountain Mental Illness Research Education and Clinical Center (MIRECC), The Rocky Mountain Regional Medical Center (RMRMC), Aurora, CO 80045, USA e Military and Veteran Microbiome: Consortium for Research and Education (MVM-CoRE), Aurora, CO 80045, USA b

ARTICLE INFO

ABSTRACT

Keywords: Anxiety Chronic subordinate colony housing Colitis CSC Immunoregulation Inflammation Mycobacterium vaccae Old friends Intranasal administration

An increasing body of evidence indicates that immunodysregulation and subsequent chronic low-grade inflammation can promote the development of stress-related somatic and psychiatric pathologies, including inflammatory bowel disease (IBD) and posttraumatic stress disorder (PTSD). Thus, immunoregulatory approaches counterbalancing basal and/or stress-induced immune activation should have stress-protective potential. In support of this hypothesis, we recently demonstrated that repeated s.c. preimmunization with a heat-killed preparation of the immunoregulatory bacterium Mycobacterium vaccae (M. vaccae; National Collection of Type Culture (NCTC) 11659), protects mice against stress-induced general anxiety, spontaneous colitis, and aggravation of dextran sulfate sodium (DSS)-induced colitis in the chronic subordinate colony housing (CSC) paradigm, a validated model for PTSD in male mice. In the current study, we repeatedly administered M. vaccae via the non-invasive intranasal (i.n.; 0.1 mg/mouse/administration) route, prior to or during CSC exposure or single housed control (SHC) conditions, and assessed the effects on general and social anxiety, and on parameters related to the severity of DSS-induced colitis. While administration of M. vaccae prior to the onset of CSC exposure only had minor stress-protective effects, administration of M. vaccae during CSC completely prevented CSC-induced aggravation of DSS colitis. As CSC in the current experimental setting did not reliably increase general anxiety-related behavior, potential stress-protective effects of M.vaccae are difficult to interpret. Taken together, these data broaden the framework for developing bioimmunoregulatory approaches, based on the administration of microorganisms with anti-inflammatory and immunoregulatory properties, for the prevention of stress-related disorders.

1. Introduction 1.1. Cytokine theory of stress-associated mental disorders Chronic psychosocial stress is a burden of modern society and has long been recognized as a risk factor for numerous somatic and affective disorders like inflammatory bowel disease (IBD) (Bitton et al., 2003; Levenstein et al., 2000) and posttraumatic stress disorder (PTSD) (Reber et al., 2016a). While the underlying mechanisms are not yet fully understood, a considerable number of studies indicates that an

increased immune activation represents a core feature for many stressrelated disorders and might even be involved in disease pathogenesis (Gao et al., 2018; Lowry et al., 2016; Miller and Raison, 2016; Rohleder, 2014). In support of the latter and considering that stressassociated mental disorders are more prevalent in urban versus rural areas (Peen et al., 2010), healthy participants raised in an urban environment in the absence of pets relative to those raised in a rural environment in the presence of farm animals showed an exaggerated immune response to acute psychosocial stress (Böbel et al., 2018). Moreover, subjects suffering from PTSD often are characterized by a

Corresponding author at: Laboratory for Molecular Psychosomatics, Department of Psychosomatic Medicine and Psychotherapy, University Ulm, Albert-EinsteinAllee 23, 89081 Ulm, Germany. E-mail address: [email protected] (S.O. Reber). 1 These authors contributed equally. ⁎

https://doi.org/10.1016/j.bbi.2019.05.005 Received 26 November 2018; Received in revised form 2 April 2019; Accepted 2 May 2019 0889-1591/ © 2019 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).

Please cite this article as: Mattia Amoroso, et al., Brain, Behavior, and Immunity, https://doi.org/10.1016/j.bbi.2019.05.005

Brain, Behavior, and Immunity xxx (xxxx) xxx–xxx

M. Amoroso, et al.

failure in adequate immunoregulation (Sommershof et al., 2009) and a plethora of inflammatory comorbidities (O'Donovan et al., 2015). Furthermore, the extent of stress-induced immune activation has been shown to be predictive for developing PTSD in humans (Eraly et al., 2014; Pervanidou et al., 2007), as well as development of anxiety-like defensive behavioral responses (Hodes et al., 2014), and impairment of fear extinction in mice (Young et al., 2018).

with immunoregulatory capacities. For example, mycobacteria are abundant in municipal showerheads, where they may be aerosolized (Gebert et al., 2018) and are a normal component of the healthy human microbiome of the oral cavity (buccal mucosa and dental plaque) and upper respiratory tract (nostrils and oropharynx) and, therefore, are considered part of the microbiome of the upper airways (Macovei et al., 2015). Support for i.n. administered M. vaccae to have stress-protective effects is provided by data showing that both i.n. (Hopfenspirger and Agrawal, 2002) and intragastric (i.g.) administration (Hunt et al., 2005) of M. vaccae, i.e. two mucosal routes of administration, were protective against allergic airway inflammation in a mouse model of asthma. Moreover, microfold cells (M cells) and dendritic cells (DCs), which are hypothesized to mediate M. vaccae effects in the intestinal mucosa (Chieppa et al., 2006; Coombes and Powrie, 2008; Fujimura, 1986; Rescigno et al., 2001), are present and cluster together in the respiratory mucosa (Kim et al., 2011) and have been identified as gateways for inhaled antigens to affect immune signaling (Kim et al., 2011; Nair et al., 2016; Park et al., 2003). Intranasal administration of different compounds has been further shown to induce immunoregulatory mechanisms in the lungs, thereby exerting protective effects in different animal models of inflammatory diseases (Gupta et al., 2012; Minne et al., 2007; Perdomo et al., 2016). Thus, in the present study, we tested whether repeated i.n. administration of M. vaccae prior to and during CSC exposure ameliorates/ prevents stress-induced anxiety and aggravation of DSS-induced colitis.

1.2. Mycobacterium vaccae (s.c.) has immunoregulatory and stressprotective effects Consistent with the aforementioned role of immune activation in the development of mood disorders, we have recently shown that immunoregulation induced by repeated subcutaneous (s.c.) immunizations with a heat-killed preparation of Mycobacterium vaccae (M. vaccae; National Collection of Type Cultures (NCTC) 11659), an abundant soil saprophyte with immunoregulatory properties, was able to prevent stress-induced anxiety, spontaneous colitis, and aggravation of dextran sulfate sodium (DSS)-induced colitis, in a murine model of PTSD (Reber et al., 2016b). As shown previously using a mouse model of allergic airway inflammation (Zuany-Amorim et al., 2002), s.c. administered M. vaccae mediated its immunoregulatory properties and, thus, stressprotective effects through induction of regulatory T cells (Tregs) and interleukin (IL)-10 secretion (Reber et al., 2016b). Moreover, M. vaccae immunization in a Treg-independent manner reduced subordinate, flight, and avoiding behavioral responses to a dominant aggressor, indicating a shift towards more proactive stress coping (Reber et al., 2016b). Independent of the beneficial effects of M. vaccae described above, M. vaccae immunization did not prevent the stress-induced development of a gut microbial milieu with inflammatory potential, but prevented the resulting immune activation normally associated with tissue damage, colitis, and development of a chronic anxiety/fear-like state. Moreover, immunization with M. vaccae did not affect stress-induced changes in the hypothalamic-pituitary-adrenal (HPA) axis (i.e. increased adrenal weight and adrenal ex vivo adrenocorticotropic hormone (ACTH) insensitivity), suggesting that, although HPA axis changes may contribute to stress-induced exaggeration of inflammation, they do not mediate the protective effects of M. vaccae immunization.

2. Methods 2.1. Animals Male C57BL/6N mice (Charles River, Sulzfeld, Germany) weighing 17–19 g (Experiment 1) or 19–21 g (Experiment 2) were used as experimental mice and male CD-1 mice (Charles River, Sulzfeld, Germany) weighing 30–35 g were used as dominant aggressor mice. All mice were kept in standard polycarbonate mouse cages (16 cm width × 22 cm length × 14 cm height) housed in a specific-pathogenfree (SPF) animal facility under standard laboratory conditions (12-h light–dark cycle, lights on at 06.00 h, 22 °C, 60% humidity) and had free access to tap water and standard mouse diet. All experimental protocols were approved by the Committee on Animal Health and Care of the local government and performed according to international guidelines on the ethical use of animals. All efforts were made to minimize the number of animals used and their suffering. Females were not used in the current study as the CSC paradigm is based on territorial aggression and the establishment of social hierarchies, which is not typically seen in female mice. The research described here was conducted in compliance with the ARRIVE Guidelines for Reporting Animal Research (Kilkenny et al., 2010).

1.3. The chronic subordinate colony housing (CSC) paradigm as mouse model of PTSD To induce this PTSD-like phenotype, the chronic subordinate colony housing (CSC) paradigm, which is based on the chronic subordination of four male CSC mice towards a dominant resident male conspecific, was used (Füchsl et al., 2013; Füchsl et al., 2014; Langgartner et al., 2015). Briefly, compared with single-housed controls (SHC), CSC mice avoid trauma-related external reminders, indicated by a lack of social preference towards unfamiliar male mice, and develop a long-lasting increase in general and social anxiety-related behavior and alcohol consumption/preference, hyperactivity, spontaneous colitis, and an aggravated DSS-induced colitis. CSC exposure is further associated with basal hypocorticism, increased dexamethasone suppression of ACTH, and increased HPA axis reactivity toward novel stressors. Importantly, CSC mice show a reduced number of Tregs (Schmidt et al., 2010), likely contributing to the overall increased inflammatory state.

2.2. Experimental procedures Mice in Experiment 1 (SHC, Veh, n = 8; SHC, M. vaccae, n = 8; CSC Veh, n = 8; CSC M. vaccae, n = 8) were housed in groups of four on the day of arrival (day −21), and received repeated i.n. administrations of either Veh (BBS: borate-buffered saline) or M. vaccae (for details see Section 2.3) on days −21, −14 and −7 prior to stress exposure (Fig. 1). On day −7, mice in Experiment 1 were single-housed. Mice in Experiment 2 (SHC, Veh, n = 8; SHC, M. vaccae, n = 8; CSC, Veh, n = 8; CSC, M. vaccae, n = 8) were single-housed on the day of their arrival (day −7). On day 1, mice in Experiments 1 and 2 were either exposed to the CSC paradigm or kept single-housed as controls (SHC), while only mice in Experiment 2 received repeated i.n. administrations of either Veh or M. vaccae on days 2, 8 and 15 during CSC exposure (Fig. 1). Behavioral tests were performed in both experiments to assess general and social anxiety-related behaviors. Specifically, the elevated plus-maze test (EPM) was performed on day 19, the open-field/novel

1.4. Intranasal administration as a non-invasive route to promote stressresilience Although M. vaccae has proven to be stress-protective when applied via the s.c. route, it remains unknown if M. vaccae also promotes stressresilience when administered via the non-invasive intranasal (i.n.) route. In contrast to s.c. administration, i.n. administration more closely mimics the natural route of exposure to environmental microorganisms 2

Brain, Behavior, and Immunity xxx (xxxx) xxx–xxx

M. Amoroso, et al.

Fig. 1. Experimental timeline. Schematic illustration of the experimental timelines of Experiment 1 and Experiment 2. Male C57BL/6N mice were exposed to the chronic subordinate colony housing (CSC) paradigm or kept as single-housed controls (SHC). Mycobacterium vaccae (M. vaccae) or Vehicle (Veh) was either applied prior to (Experiment 1, on day -21, -14 and -7) or during (Experiment 2, on day 2, 8 and 15) CSC exposure, followed by a week of dextran sulphate sodium (DSS) treatment before euthanasia. Final group sizes following exclusion of outliers identified by statistical analysis (for details see. Section 2.12) are as follows: Experiment 1 (SHC Veh: n = 6, CSC Veh: n = 7–8, SHC M. vaccae: n = 7–8, CSC M. vaccae: n = 7); Experiment 2 (SHC Veh: n = 7–8, CSC Veh: n = 7, SHC M. vaccae: n = 8, CSC M. vaccae: n = 8). Abbreviations: EPM = elevated plus-maze test; OF/NO = open-field/novel object test; SPAT = social preference/avoidance test; i.n. = intranasal.

object test (OF/NO) was performed on day 20, and the social preference/avoidance test (SPAT) was performed on day 21. After the SPAT, all mice were housed singly and received 1% DSS (MP Biomedicals, Santa Ana, California, USA) in their drinking water for one week. On day 28 mice in both experiments were euthanized between 07.00 and 10.00 a.m. by decapitation following brief CO2 inhalation for assessment of ex vivo cytokine secretion from isolated mesenteric lymph node cells (mesLNC), colon length, and histological damage score. For assessment of changes in bodyweight, mice were weighed on days 1, 5, 8, 12, 15, 19, 21, 24, 25, 26, 27 and 28. Two mice in the VEH-SHC group were excluded from analysis due to severe splenomegaly indicating unexpected systemic immune activation and one CSC-M. vaccae mouse died during CSC, all in Experiment 1.

2.5. Elevated plus-maze (EPM) To assess treatment (M. vaccae and/or CSC) effects on anxiety-related behavior, the EPM test was performed on day 19 of the CSC paradigm as previously described (Reber et al., 2007). Mice were tested for 5 min between 07.00 a.m. and 10.00 a.m. The test took place in a dedicated behavioral box inside the animal room where all animals were housed for the whole duration of the experiment. Briefly, the plusmaze consists of two open (length: 30 cm; width: 6 cm; 140 lx) and two closed (length: 30 cm; width: 6 cm; height: 17 cm, 85 lx) arms radiating from a central platform (6 cm × 6 cm) to form a plus-shaped figure elevated 130 cm above the floor. The open arm edges were 0.3 cm in height to avoid falling. Each mouse was placed in one of the closed arms for 30 s to habituate before the test started. During this time, the entrance to the other arms of the maze was blocked. The maze was thoroughly cleaned with water after each test. The number of entries into the closed arms, as well as the relative time spent in the open arms (percentage of time spent in the open arms relative to the amount of time spent in all arms) was assessed. All parameters were analyzed using EthoVision XT (v9.0.724; Noldus Information Technology, Wageningen, The Netherlands).

2.3. Intranasal application of M. vaccae All mice received i.n. administration of a whole heat-killed preparation of M. vaccae [supplied as 10 mg/ml stock solution; strain NCTC 11659, batch ENG#1, provided by BioElpida (Lyon, France)] or Veh (BBS: borate-buffered saline). Mice in Experiment 1 received i.n. administrations on days -21, -14 and -7 prior to CSC, whereas mice in Experiment 2 received i.n. administrations on days 2, 8 and 15 during CSC. M. vaccae was diluted in BBS at a final concentration of 8.3 mg/ml and each mouse received 12 µl of M. vaccae or Veh by administering 3 µl to each nostril in two consecutive rounds of administrations. The final dose of 0.1 mg M. vaccae administered i.n. per mouse is equivalent to the dose we administered s.c. per mouse previously (Reber et al., 2016b). A 10 µl pipette tip was used, according to a procedure described previously (Lukas and Neumann, 2012).

2.6. Open-field/novel object test (OF/NO) To assess treatment (M. vaccae and/or CSC) effects on anxietyrelated behavior, the OF/NO test was conducted on day 20 between 07:00 a.m. and 10:00 a.m. of the CSC paradigm as previously described (Foertsch et al., 2017; Langgartner et al., 2015). The test took place in a dedicated behavioral box inside the animal room where all animals were housed for the whole duration of the experiment. Briefly, the test arena (length: 45 cm; width: 27 cm; height: 27 cm; 350 lx) was subdivided into an inner (9 cm × 27 cm), and an outer zone. Within each trial, the mouse was placed into the inner zone and allowed to explore the arena for 5 min. After 5 min of open-field exploration, a plastic round object (diameter: 3.5 cm; height: 1.5 cm) was placed in the middle of the inner zone. The mouse was then allowed to explore the arena containing the unfamiliar object for 5 min. In the OF/NO test, the distance moved by the animals was analyzed. Furthermore, in the NO test, the time spent exploring the object was analyzed. The test arena was cleaned thoroughly with water after each test. All parameters were analyzed using EthoVision XT (v11.5.1022; Noldus Information Technology, Wageningen, The Netherlands).

2.4. Chronic subordinate colony housing (CSC) procedure The CSC paradigm was performed as described previously (Langgartner et al., 2015; Reber et al., 2007; Reber et al., 2008). Briefly, on day 1 mice were weighed and assigned to either the SHC or the CSC group, according to their body weight. SHC mice remained undisturbed for the duration of the experiment, except for changing the bedding once a week and weighing (see Section 2.2). To induce a PTSD-like phenotype, four CSC mice were housed together with a dominant male CD-1 aggressor mouse for 20 consecutive days. To avoid habituation, CSC mice were transferred to the cages of new CD1 aggressor mice on days 8 and 15. Before the beginning of the CSC procedure, all potential aggressor mice were tested to exclude the most aggressive individuals. Mice that started to injure their opponents by harmful bites were not used for the CSC procedure. 3

Brain, Behavior, and Immunity xxx (xxxx) xxx–xxx

M. Amoroso, et al.

2.7. Social preference/avoidance test (SPAT)

stimulated mesLNC was measured using a commercially available ELISA kit according to the manufacturer’s protocol (IFN-γ, cat. no. DY485-05, DuoSet ELISA, R&D Systems Europe, Ltd., Abingdon, UK; IL10, cat. no. DY417-05 DuoSet ELISA, R&D Systems Europe).

To assess treatment (M. vaccae and/or CSC) effects on general and social anxiety, the SPAT was conducted on day 21 between 07.00 a.m. and 10.00 a.m. as described previously (Slattery et al., 2012). The test took place in a dedicated behavioral box inside the animal room where all animals were housed for the whole duration of the experiment. Briefly, the experimental mouse was placed in the SPAT box (length: 45 cm; width: 27 cm; height: 27 cm; 20 lx) for 30 s to habituate to the unfamiliar environment before a small empty wire mesh cage (length: 8.5 cm; width: 7.5 cm; height: 6.5 cm) was introduced for 150 s. Afterwards, the empty cage was exchanged with an identical cage containing an unfamiliar male conspecific for another 150 s. The half of the SPAT box containing the cage was subdivided into the direct contact zone (a 2.25-cm broad zone around the wire mesh cage) and the remaining cage zone. The box was cleaned thoroughly with water before every test. Total distance moved, time spent in the cage zone and time in, as well as entries into the direct contact zone, were recorded using EthoVision XT (v11.5.1022; Noldus Information Technology, Wageningen, The Netherlands).

2.11. Determination of colon length and histological damage score of the colon To assess CSC effects on intestinal inflammation, the colon was removed. After its length was measured, it was mechanically cleaned of feces. To prepare histological sections, 1 cm of the distal third of the colon was cut longitudinally and fixed in 4% formalin overnight. The next day, the fixed tissue was embedded in paraffin and cut longitudinally. Histological slides were obtained from the paraffin blocks, stained with hematoxylin-eosin and evaluated by histological scoring performed by an investigator blinded to treatment. Histological damage score was assessed as previously described (Reber et al., 2008): epithelium score (0: normal morphology; 1: loss of goblet cells; 2: loss of goblet cells in large areas; 3: loss of crypts; 4: loss of crypts in large areas) and infiltration score (0: no infiltration; 1: infiltrate around crypt bases; 2: infiltrate reaching to lamina muscularis mucosae; 3: extensive infiltration reaching the lamina muscularis mucosae and thickening of the mucosa with abundant edema; 4: infiltration of the lamina submucosa). The total histological score of each mouse represents the sum of the epithelium and infiltration scores and ranges from 0 to 8.

2.8. DSS treatment Immediately after the SPAT, on day 21, all experimental mice were housed singly and received 1% DSS (50 kDa; cat. no. 16011036, MP Biomedicals) through their drinking water for 7 consecutive days (days 21–28). The DSS bottles were weighed at the beginning and at the end of the treatment and the amount of DSS consumed was calculated. As a rule, any mouse that had a DSS consumption exceeding the mean of its treatment group by 75% was excluded from subsequent analysis. Given that drinking behavior and, thus, overall DSS consumption varied between the groups all physiological and colitis-related data were corrected by the amount of DSS [g] consumed.

2.12. Statistics For statistical comparisons, the software package IBM SPSS statistics (version 25.0) was used. For graphical illustration, the software package SigmaPlot (version 13.0) was used. Kolmogorov-Smirnov test using Lilliefors‘ significance was employed to test normal distribution of all acquired data sets. Outliers in normally distributed data sets were identified using Grubbs’ test (Grubbs, 1969) and excluded from further analysis. Normally distributed data sets were subsequently analyzed using two-way ANOVA (two factors, two or more independent samples) followed by post hoc analysis using Bonferroni pairwise comparison in case a significant main effect was found; in case of repeated measures, a linear mixed model approach was employed. Non-normally distributed data sets were analyzed using non-parametric statistics, i.e. MannWhitney U test (MWU; one factor, two independent samples) and Wilcoxon test (one factor, two dependent samples). Normally distributed data are presented as bar plots (mean + SEM). Non-normally distributed data are presented as box plots (median; mean; 10th, 25th, 75th and 90th percentile; outliers). The level of significance was set at p ≤ 0.05. Tendencies are reported in the graphs (i.e., p > 0.05 and < 0.06).

2.9. Isolation and incubation of mesenteric lymph node cells (mesLNC) To assess treatment (M. vaccae and CSC) effects on anti-CD3/ CD28–stimulated cytokine secretion from isolated lymph node cells, mesenteric lymph nodes were removed and stored in ice-cold Roswell Park Memorial Institute Medium (RPMI-1640, cat. no. R8758, SigmaAldrich, St. Louis, MO, USA) supplemented with 10% fetal bovine serum (FBS, cat. no. 10270106, Gibco®, Thermo Fisher Scientific, Waltham, MA, USA), 100 U/mL penicillin and 100 µg/mL streptomycin (cat. no. 15070063, Gibco®, Thermo Fisher Scientific), and 3 × 10−5 M β-mercaptoethanol (cat. no. M6250, Sigma-Aldrich). Lymph nodes were mechanically disrupted and filtered through a cell strainer (Corning™, 70 µm nylon, cat. no. 431751, Thermo Fisher Scientific). Afterward, cells were washed in cell-culture medium and adjusted to a concentration of 2 × 105 cells/100 µl. Twenty-four h before lymph node cell incubation, wells of a 96-well plate were pre-coated with 100 µl of anti-CD3 antibody (cat. no. MA1-10184, Thermo Fisher Scientific) diluted in PBS (final concentration: 2.5 µg/ml/well). The wells were then aspirated and washed twice with PBS before lymph node cells were transferred to the wells. 2 × 105 cells were plated per well. Following plating, 100 µl anti-CD28 antibody (cat. no. 16-0281, e-Bioscience™, Thermo Fisher Scientific) diluted in RPMI were added to the wells at a final concentration of 1 µg/ml/well. After 48 h incubation (37 °C, 5% CO2), supernatants were removed and stored at −20 °C until interferon gamma (IFN-γ) and interleukin 10 (IL-10) concentrations were measured in the supernatants using a commercially available ELISA kit.

3. Results 3.1. Effects of M. vaccae administration on general and social anxiety following CSC exposure General anxiety-related behavior of all mice was assessed employing the EPM (day 19) and the OF/NO (day 20); social anxiety-related behavior was assessed in the SPAT (day 21). 3.1.1. M. vaccae administration prior to CSC exposure Percentage of time spent on open arms (Fig. 2A) and number of entries into closed arms (Fig. 2B), indicative of general anxiety and locomotor activity during EPM testing, respectively, were neither affected by M. vaccae nor CSC. In line with this lack of both CSC and M. vaccae effects on general anxiety in the EPM, distance moved during OF exposure (Fig. 2C), distance moved during NO exploration (Fig. 2D) and time spent in the direct contact zone during NO exposure (Fig. 2E) were comparable between the groups. SHC mice in the Veh group show

2.10. Enzyme-linked immunosorbent assay (ELISA) for assessment of IFN-γ and IL-10 secretion from isolated and anti-CD3/CD28 stimulated mesLNC ex vivo IFN-γ concentration in the supernatants of anti-CD3/CD284

Brain, Behavior, and Immunity xxx (xxxx) xxx–xxx

M. Amoroso, et al.

Fig. 2. Effects of intranasal administration of Mycobacterium vaccae (M. vaccae) on chronic subordinate colony housing (CSC)-induced general and social anxiety. In Experiment 1, neither M. vaccae administration prior to CSC nor subsequent CSC exposure significantly affected general anxiety on the elevated plus-maze (EPM) or in the open-field/novel object (OF/NO) test (A–E). Furthermore, both M. vaccae- and Vehicle (Veh)-treated SHC but not CSC mice showed social preference in the SPAT (F-I), with M. vaccae treatment having mildly protective effects on social anxiety (H). In Experiment 2, neither M. vaccae administration during CSC nor CSC exposure had major effects on general anxiety in the EPM and the OF/NO (J-N). Locomotion, as measured by the number of entries in to the closed arms, was increased in CSC mice in the M. vaccae but not Veh group (K). Furthermore, both M. vaccae-and Veh-treated SHC mice as well as the Veh-treated CSC mice showed social preference in the SPAT (O-R). Experiment 1 (SHC Veh: n = 6, CSC Veh: n = 8, SHC M. vaccae: n = 7–8, CSC M. vaccae: n = 7); Experiment 2 (SHC Veh: n = 7–8, CSC Veh: n = 8, SHC M. vaccae: n = 8, CSC M. vaccae: n = 8). Parametric data are presented as mean + SEM. Non-parametric data are represented as boxplots. Solid line represents the median, dotted line represents the mean for each data set. Lower boxes indicate 25th, upper boxes indicate 75th percentile; 10th (lower error bar), and 90th percentile (upper error bar) as well as possible outliers beyond the percentiles (indicated by closed circles) are also shown. * p ≤ 0.05, ** p ≤ 0.01 versus respective SHC; $ p ≤ 0.05, $$ p ≤ 0.01, $$$ p ≤ 0.001 versus respective empty cage.

typical signs of social preference, indicated by an increased time in the cage zone (Wilcoxon; p = 0.05; Fig. 2F) and in the direct contact zone (Wilcoxon; p = 0.03; Fig. 2G), as well as an increased distance moved (Wilcoxon; p = 0.03; Fig. 2H) and entries into the direct contact zone (Wilcoxon, p = 0.05; Fig. 2I), all during social vs. empty cage exploration. Moreover, Veh-CSC vs. SHC mice spent less time in the cage zone (MWU; p = 0.04; Fig. 2F) and in the direct contact zone (MWU; p = 0.005; Fig. 2G), as well as moved less (MWU; p = 0.02; Fig. 2H) and by trend entered the direct contact zone less often (MWU; p = 0.06; Fig. 2I), indicating a CSC-induced lack of social preference. Although SHC mice in the M. vaccae-group did not spend more time in the cage zone (Fig. 2F) and in the direct contact zone (Fig. 2G), and did not enter the direct contact zone more often (Fig. 2I), all during social vs. empty cage exploration, an increased distance moved during social vs. empty cage exploration (Wilcoxon; p = 0.01; Fig. 2H) indicates social preference in this group. The latter was missing in M. vaccae-CSC mice.

general locomotor activity during EPM testing, was increased in CSC vs. SHC mice in the M. vaccae group (Fig. 2K; MWU; p = 0.04). Distance moved during OF (Fig. 2L) and NO (Fig. 2M) exploration, as well as time spent in the contact zone during NO exploration (Fig. 2N) were neither affected by M. vaccae nor CSC. Social preference in the SPAT (Fig. 2O-R) was seen in SHC mice of both, the Veh and M. vaccae -group. This was indicated by an increased time in the direct contact zone (Wilcoxon: p = 0.05; Fig. 2P) and an increased distance moved (Linear mixed model; F(Time)1,26.427 = 24.845, p < 0.001; F (CSC × time)1,26.427 = 7.450, p = 0.01; F(M. vaccae × time)1,26.427 = 6.617, p = 0.02; Bonferroni: p < 0.001; Fig. 2Q) in Veh-SHC mice, as well as an increased time in the direct contact zone (Wilcoxon: p = 0.05; Fig. 2P), an increased distance moved (Linear mixed model; F(Time)1,26.427 = 24.845, p < 0.001; F(CSC × time)1,26.427 = 7.450, p = 0.01; F(M. vaccae × time)1,26.427 = 6.617, p = 0.02; Bonferroni: p = 0.002; Fig. 2Q) and an increased number of entries into the direct contact zone (Wilcoxon; p = 0.01; Fig. 2R) in M. vaccae-SHC mice. Although Veh-CSC mice showed an increased distance moved during social vs. empty cage exploration (Linear mixed model; F(Time)1,26.427 = 24.845, p < 0.001; F(CSC × time)1,26.427 = 7.450, p = 0.01; F(M. vaccae × time)1,26.427 = 6.617, p = 0.02; Bonferroni: p = 0.003; Fig. 2Q), they

3.1.2. M. vaccae administration during CSC exposure Percentage of time spent on open arms (Fig. 2J), indicative of general anxiety during EPM testing, was neither affected by M. vaccae nor CSC, whereas the number of entries in the closed arms, indicative of 5

Brain, Behavior, and Immunity xxx (xxxx) xxx–xxx

M. Amoroso, et al.

spent less time in the direct contact zone during social cage exploration when compared to the respective SHC group (MWU; p = 0.05; Fig. 2P), indicating a CSC-induced lack of social preference. In line with these findings, M.vaccae-CSC mice during social cage exploration spent less time in (MWU; p = 0.01; Fig. 2P) and entered (MWU; p = 0.04; Fig. 2R) the direct contact zone less often, as well as moved less (Linear mixed model; F (Time)1,26.427 = 24.845, p < 0.001; F(CSC × time)1,26.427 = 7.450, p = 0.01; F(M. vaccae × time)1,26.427 = 6.617, p = 0.02; Bonferroni: p = 0.02; Fig. 2Q) compared to respective SHC mice, indicating that

M.vaccae did not ameliorate CSC-induced social deficits. Time in the cage zone during “social” versus “empty cage” exploration (Fig. 2O) was neither affected by M. vaccae nor CSC. 3.2. Repeated i.n. M. vaccae administration protects against CSC-induced aggravation of DSS-induced colitis To assess if repeated i.n. administrations of M. vaccae protect against CSC-induced aggravation of DSS-induced colitis in Experiment

(caption on next page) 6

Brain, Behavior, and Immunity xxx (xxxx) xxx–xxx

M. Amoroso, et al.

Fig. 3. Effects of intranasal administration of Mycobacterium vaccae (M. vaccae) on dextran sulphate sodium (DSS)-induced colonic inflammation. Neither M. vaccae administration nor subsequent chronic subordinate colony housing (CSC) exposure affected DSS consumption (A) or DSS-induced inflammatory shortening of the colon (F) in Experiment 1, although CSC induced a more pronounced body weight loss compared to SHC in both treatment groups considering delta body weight between d28-21 (B) as well as d28-24 (B, inlay). Moreover, although M. vaccae administration prevented CSC-induced increase in the histological damage score (E, G), it did not prevent the CSC-induced increase in interferon (IFN)-γ secretion from anti-CD3/CD28-stimulated mesenteric lymph node cells (mesLNC) ex vivo (C). In addition, neither M. vaccae nor CSC affected IL-10 secretion from anti-CD3/CD28-stimulated mesLNC ex vivo (D). M. vaccae administration during CSC exposure (Experiment 2; H-N) resulted in an increased DSS consumption (H) and could prevent the CSC-induced increase in the histological damage score (L, N). A less pronounced inflammatory shortening of the colon was also seen in chronically stressed M. vaccae vs. Vehicle (Veh)-treated mice (M). Moreover, M. vaccae administration prevented CSC-induced body weight loss (d28-21 (I) and d28-24 (I, inlay)) as well as the CSC-induced increase in IFN-γ (J) secretion from anti-CD3/ CD28-stimulated mesLNC ex vivo. Neither CSC nor M. vaccae exposure affected IL-10 secretion from anti-CD3/CD28-stimulated mesLNC ex vivo (K). Bars in the histological images of the colon for the prevention (G) and treatment (N) protocol represent 200 µm. Experiment 1 (SHC Veh: n = 6, CSC Veh: n = 7–8, SHC M. vaccae: n = 7–8, CSC M. vaccae: n = 7); Experiment 2 (SHC Veh: n = 7–8, CSC Veh: n = 7, SHC M. vaccae: n = 8, CSC M. vaccae: n = 8). Parametric data are presented as mean + SEM. Non-parametric data are represented as box-plots. Solid line represents the median, dotted line represents the mean for each data set. Lower boxes indicate 25th, upper boxes indicate 75th percentile, 10th (lower error bar), and 90th percentile (upper error bar) as well as possible outliers beyond the percentiles (indicated by closed circles) are also shown. * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001 versus respective SHC; ## p ≤ 0.01 versus respective Veh.

1 and Experiment 2, different inflammation-related parameters were obtained from SHC and CSC mice that received i.n. administrations of M. vaccae or vehicle either: 1) on days -21, -14, and -7 prior to CSC exposure (Fig. 3, A–G); or 2) on days 2, 8, and 15 during CSC (Fig. 3, H–N).

Moreover, M. vaccae administration prevented the CSC-induced increase in body weight loss during DSS treatment between days 21 and 28 (Fig. 3I, as well as the ex vivo secretion of IFN-γ from anti-CD3/ CD28-stimulated mesLNC (Fig. 3J). On the other hand, the ex vivo secretion of IL-10 from anti-CD3/28-stimulated mesLNC (Fig. 3K) was neither affected by M. vaccae nor CSC exposure.

3.2.1. M. vaccae administration prior to CSC exposure DSS consumption was neither affected by M. vaccae administration nor by CSC exposure (Fig. 3A). Independent of M. vaccae administration, CSC mice showed an increased body weight loss during DSS treatment between days 21 and 28 (Fig. 3B, MWU; Veh: p = 0.04; MWU; M. vaccae: p = 0.01) as well as between days 24 and 28 (Fig. 3B (inlay), MWU; Veh: p = 0.01; MWU; M. vaccae: p = 0.01). CSC mildly aggravated DSS-induced colitis in Veh-treated mice indicated by a trend towards an increased ex vivo IFN-γ secretion from anti-CD3/CD28-stimulated mesLNC (Fig. 3C; MWU; p = 0.06) in CSC versus respective SHC mice, and a trend towards an increased histological damage score (Fig. 3E, G; MWU; p = 0.06), while the inflammatory shortening of the colon (Fig. 3F) was unaltered. Although the trend towards an increased histological damage score was not detectable in CSC versus SHC in the M. vaccae-treated group, suggesting mild protective effects of prior-tostress i.n. M. vaccae administration, CSC mice in this group showed an increased ex vivo IFN-γ (Fig. 3C; MWU; p = 0.03) secretion from antiCD3/CD28-stimulated mesLNC when compared to respective SHC mice. Moreover, neither CSC nor M. vaccae affected ex vivo secretion of IL-10 from anti-CD3/CD28-stimulated mesenteric lymph node (Fig. 3D).

4. Discussion In the present study, we provide the first evidence supporting the stress-protective role of repeated non-invasive i.n. M. vaccae administration in a mouse model of PTSD, although the effectiveness seems to strongly depend on the timing of administration. While i.n. M. vaccae administered prior to stressor exposure had only moderate to no colitisprotective effects, i.n. M. vaccae administered during stressor exposure completely prevented CSC-induced aggravation of DSS-induced colitis. We previously showed that s.c. administration of M. vaccae protects against stress-induced aggravation of DSS-induced colitis (Reber et al., 2016b), but does not affect basal inflammatory processes. Thus, the moderate colitis-protective effect of i.n. M. vaccae administration prior to stress might be explained by the fact that the CSC administration protocol reliably aggravated DSS-induced colitis only in Experiment 2, when M. vaccae was given during CSC. Furthermore, as CSC did not reliably affect anxiety in mice pretreated with vehicle via the i.n. route, the stress-protective effects of i.n. administered M. vaccae are difficult to interpret.

3.2.2. M. vaccae administration during CSC exposure Mice treated with M. vaccae during CSC exposure consumed more DSS. In detail, the latter was indicated by significant main effects for factors M. vaccae (Fig. 3H; two-way ANOVA; F1,26 = 8.54; p = 0.01) and CSC (Fig. 3H; two-way ANOVA; F1,26 = 4.62; p = 0.04), as well as a significantly increased DSS consumption in M. vaccae CSC compared with both respective Veh (p = 0.01) and SHC (p = 0.03) mice, revealed by Bonferroni post hoc testing. Moreover, CSC aggravated DSS-induced colitis in Veh-treated mice as shown by an increased body weight loss during DSS treatment between days 21 and 28 (Fig. 3I; two-way ANOVA; F(CSC)1,27 = 10.91, p = 0.003; Bonferroni: p = 0.002) and between days 24 and 28 (Fig. 3I (inlay); two-way ANOVA; F (CSC)1,27 = 17.22, p < 0.001; Bonferroni: p < 0.001), histological damage score of the colon (Fig. 3L, N; MWU; p = 0.05) and secretion of IFN-γ (Fig. 3J; MWU; p = 0.01) from ex vivo anti-CD3/CD28-stimulated mesLNC in respective CSC vs. SHC mice. Importantly, treating mice with M. vaccae during CSC exposure completely blocked these effects, indicated by a significantly reduced body weight loss (Fig. 3I (inlay); two-way ANOVA; F(M. vaccae)1,27 = 4.11, p = 0.05; Bonferroni: p = 0.01), histological damage score (Fig. 3L, N; MWU; p = 0.01) and inflammatory shortening of the colon (Fig. 3M; two-way ANOVA; F(M. vaccae)1,27 = 4.08, p = 0.05; Bonferroni: p = 0.01) in M. vaccae-immunized CSC mice compared with respective Veh-treated mice.

4.1. Effects of i.n. M. vaccae administration on anxiety-like behavior In contrast to the protective effects of i.n. M. vaccae in terms of stress-induced aggravation of DSS-induced colitis, little can be said about the protective effects of i.n. M. vaccae administration on stressinduced general anxiety. In contrast to what we reliably detect in nontreated mice (Füchsl et al., 2014; Langgartner et al., 2017), CSC in the present study was not able to reliably increase general anxiety, neither when i.n. Veh was given prior to CSC exposure nor during CSC exposure. This is likely due to the fact that the i.n. administration procedure requires immobilization of experimental mice and, thus, is per se stressful and anxiogenic, masking possible CSC effects on emotionality. Support for the latter hypothesis comes from data showing that acute immobilization stress is able to increase anxiety-like behavior with a time-delay of about 10 days, measured as significant open arm avoidance in the EPM (Mitra et al., 2005). This effect was paralleled by an up-regulated spine density on dendrites of basolateral amygdala neurons (BLA), which is believed to be a site for storage of fearful memories and/or stressful experiences (Blair et al., 2001; LeDoux, 1993; Rogan et al., 1997; Schafe et al., 2001). As these molecular and behavioral consequences of acute immobilization could be prevented by prior administration of low-dose corticosterone (Rao et al., 2012), one possible option to prevent the anxiogenic effects of the 7

Brain, Behavior, and Immunity xxx (xxxx) xxx–xxx

M. Amoroso, et al.

administration procedure per se might be to administer Veh or M. vaccae during the dark phase, when corticosterone levels are elevated due to the diurnal rhythm. In contrast to general anxiety, CSC in Veh and M. vaccae-treated mice induced a lack of social preference employing both the prior and during application protocol. This is on the one hand indicating that the effects of CSC on social anxiety are more robust than its effects on general anxiety and on the other hand suggesting that M. vaccae specifically affects conflict anxiety as shown previously by our group (Reber et al., 2016b).

directly onto the nasal epithelium, probably resulting in only a small proportion of the material being available for presentation to DCs over a limited period of time. Of note, as ex vivo IL-10 secretion from anti CD3/CD28-stimulated mesLNC was neither affected by CSC nor M. vaccae administration, it is unlikely that the protective effects of i.n. M. vaccae administration on stress-induced aggravation of DSS colitis found in the present study are mediated by Treg cell induction in the mesenteric lymph nodes. Importantly, immunoregulatory and, thus, protective effects of i.n. administration of M. vaccae in terms of stress-induced aggravation of DSS-induced colitis are in line with previous studies focusing on mycobacteria-derived heat shock protein 65 (hsp65), demonstrating i.n. administration to be protective against atherosclerosis (Maron et al., 2002) and diabetes (Liang et al., 2010). In support for the i.n. route per se to represent an attractive way to induce immunoregulation, Klingenberg and colleagues further demonstrated that i.n. administration of a fusion protein containing apolipoprotein B-100 was effective in reducing atherosclerotic lesions (Klingenberg et al., 2010). Moreover, a reduced stroke size was found in a mouse stroke model after i.n. administration of myelin oligodendrocyte protein (Frenkel et al., 2003). As in both studies the immunoregulatory effects of i.n. administration were linked to the induction of Treg, which dampen overshooting immune responses by the secretion of anti-inflammatory IL-10 and transforming growth factor (TGF)-β, it seem that the nose-draining lymph node microenvironment (i.e. cervical lymph nodes and pulmonary lymph nodes) provides decisive signals for the DCs to induce a tolerogenic form of antigen presentation (Ruane et al., 2013; Unger et al., 2003). Therefore, it is likely that also M. vaccae, following translocation across the nasal epithelium, interacts with DCs, which in turn facilitate induction of immunoregulatory cells at the systemic level and, consequently, promote immunoregulation also at distant sites, including the colon. Similar mechanisms have been proposed for systemic immunoregulatory effects following exposure to another bacterium with immunoregulatory properties, H. pylori (Arnold et al., 2011; Arnold et al., 2012). Thus, the existing concept of a bidirectional gut-lung axis (Hong et al., 2016; Samuelson et al., 2015; Sze et al., 2014), which is based on the fact that immunological events in the lungs are able to affect gut microbiota, should perhaps be extended to a microbiome-gut-lung-nasal cavity axis. As a consequence, we hypothesize that M. vaccae administered via both the i.n. and s.c. route, will be able to induce immunoregulatory effects not only at the level of the gut, but also on a systemic level, probably preventing CSC-induced elevation of plasma cytokine levels and spleen cell activation. Although we could not provide evidence for the involvement of Tregs in the current settings, we cannot exclude the general involvement of immune cells with regulatory properties. In fact, several cell types, besides this CD4+CD25+Foxp3+ T cell subset, have been recognized in mice for their regulatory function, like for example tissue-resident memory cells (Perdomo et al., 2016), IL-10-producing DCs (Iwasaki and Kelsall, 1999), CD4+ TH2-like cells that produce IL-4 and IL-10 and antagonize the activity of TH1 effector cells (Polanski et al., 1997), CD4+CD45RBlow Tr1 cells that function through the production of IL10 (Groux et al., 1997; Zuany-Amorim et al., 2002) and CD4+ or CD8+ T cells producing TGF-β (TH3 cells) (Chen et al., 1994). Furthermore, intraepithelial CD8+ γδ T cells from the respiratory mucosa and from the small intestine have also been suggested to be involved in mucosal tolerance (Huang et al., 2004; McMenamin et al., 1991). Future studies are, thus, required to elucidate the subsets of immune cells implicated in the protective effects against gut pathology after i.n. administration of M. vaccae.

4.2. Effects of i.n. M. vaccae administration on inflammation Chronic subordinate colony housing causes spontaneous colitis (Füchsl et al., 2014; Langgartner et al., 2017; Reber et al., 2011) and aggravates DSS-induced colitis (Reber et al., 2008), with the former (Langgartner et al., 2017) but not the latter (Reber, unpublished data) being critically dependent on the presence of intestinal pathobionts (e.g., Helicobacter spp.). As experiments in the current study were conducted under SPF conditions, all experimental mice were treated with DSS (1%) in the drinking water over 7 consecutive days before euthanasia, to assess the stress-protective effects of repeated i.n. M. vaccae administration on colonic inflammation. To avoid confounding of our colitis data by CSC effects on general drinking behavior, we normalized all physiological and immunological parameters to the amount of DSS [g] consumed. In confirmation of our hypothesis and extending the stress-protective effects of M. vaccae given s.c. prior to stress (Reber et al., 2016b), M. vaccae administered repeatedly using the non-invasive i.n. route during CSC exposure completely protected CSC mice from CSC-induced aggravation of DSS-induced colitis. In detail, M. vaccae prevented the CSC-induced loss of body weight during DSS consumption, the increased histological damage score of the colon, as well as IFN-γ secretion from ex vivo-stimulated mesLNC. M. vaccaeimmunized CSC mice further showed a reduced inflammatory shortening of the colon compared with respective Veh-treated CSC mice. In contrast, and against our prediction, M. vaccae CSC mice were not statistically different from respective Veh-treated mice when M. vaccae was repeatedly administered via the i.n. route prior to CSC exposure. Only the lack of a CSC-induced increase in the histological damage score in the M. vaccae group, which was by trend detectable in the Veh group, suggested moderate stress-protective effects of repeated prior i.n. M. vaccae administration. The lack of profound colitis-protective effects of prior i.n. M. vaccae administration might be explained by the fact that CSC only in the during, but not prior, i.n. administration protocol reliably aggravated DSS-induced colitis in the Veh group. The only exceptions being: 1) the by-trend increase in the histological damage score; 2) the by-trend increase in ex vivo IFN-γ secretion from mesLNC; and 3) higher body weight loss in CSC vs. SHC in the Veh group. The inflammatory shortening of the colon, another typical colitis parameter, was not affected by CSC exposure in the Veh group among mice given i.n. injections before the onset of the CSC procedure. This hypothesis would be in line with our previous findings showing that s.c. pre-administration of M. vaccae only prevented stress-induced aggravation of DSS-induced colitis but did not affect basal inflammatory processes going on during DSS administration (Reber et al., 2016b). Another possible explanation for the fact that i.n. administration of M. vaccae only exerted stress-protective effects when given during, but not prior to, CSC (which is in contrast to the pronounced stress-protective effects of M. vaccae when administered prior to CSC via the s.c. route), might be the reduction in both the overall amount of heat-killed M. vaccae material that is available for presentation to DCs and the persistence of heat-killed M. vaccae material at the site of administration. When heat-killed M. vaccae material is injected s.c., essentially the entire dose is deposited inside the body, within the subcutaneous space, serving as a reservoir or depot of bacterial material that is phagocytosed by DCs over days or weeks. In contrast, during i.n. administration of heat-killed M. vaccae the material is deposited outside the body,

5. Summary and future perspectives In summary, this study is the first to demonstrate stress-protective effects of M. vaccae administered using the non-invasive i.n. route, especially concerning stress-induced aggravation of DSS-induced 8

Brain, Behavior, and Immunity xxx (xxxx) xxx–xxx

M. Amoroso, et al.

colitis. These findings set the stage for future studies investigating the underlying mechanisms with a focus on Treg induced in the nasal associated lymphoid tissue (NALT) and subsequently migrating into the gut. Given that social stress is difficult to predict for humans, the possibility of preventing negative stress consequences by repeated noninvasive i.n. administrations of aerosolized preparations of heat-killed M. vaccae during stressor exposure offers great potential for maintaining at least somatic health during times of chronic psychosocial stress/trauma.

Weiner, H.L., 2003. Nasal vaccination with myelin oligodendrocyte glycoprotein reduces stroke size by inducing IL-10-producing CD4+ T cells. J. Immunol. 171, 6549–6555. Füchsl, A.M., Langgartner, D., Reber, S.O., 2013. Mechanisms underlying the increased plasma ACTH levels in chronic psychosocially stressed male mice. PLoS One 8, e84161. Füchsl, A.M., Neumann, I.D., Reber, S.O., 2014. Stress resilience: a low-anxiety genotype protects male mice from the consequences of chronic psychosocial stress. Endocrinology 155, 117–126. Fujimura, Y., 1986. Functional morphology of microfold cells (M cells) in Peyer’s patches–phagocytosis and transport of BCG by M cells into rabbit Peyer’s patches. Gastroenterol. Jpn. 21, 325–335. Gao, X., Cao, Q., Cheng, Y., Zhao, D., Wang, Z., Yang, H., Wu, Q., You, L., Wang, Y., Lin, Y., Li, X., Wang, Y., Bian, J.S., Sun, D., Kong, L., Birnbaumer, L., Yang, Y., 2018. Chronic stress promotes colitis by disturbing the gut microbiota and triggering immune system response. Proc. Natl. Acad. Sci. U.S.A. 115, E2960–E2969. Gebert, M.J., Delgado-Baquerizo, M., Oliverio, A.M., Webster, T.M., Nichols, L.M., Honda, J.R., Chan, E.D., Adjemian, J., Dunn, R.R., Fierer, N., 2018. Ecological analyses of mycobacteria in showerhead biofilms and their relevance to human health. mBio 9, e01614–e01618. https://doi.org/10.1128/mBio.01614-18. Groux, H., O'Garra, A., Bigler, M., Rouleau, M., Antonenko, S., de Vries, J.E., Roncarolo, M.G., 1997. A CD4+ T-cell subset inhibits antigen-specific T-cell responses and prevents colitis. Nature 389, 737–742. Grubbs, F.E., 1969. Procedures for detecting outlying observations in samples. Technometrics 11, 1–21. Gupta, A., Ahmad, F.J., Ahmad, F., Gupta, U.D., Natarajan, M., Katoch, V., Bhaskar, S., 2012. Efficacy of Mycobacterium indicus pranii immunotherapy as an adjunct to chemotherapy for tuberculosis and underlying immune responses in the lung. PLoS One 7, e39215. Hodes, G.E., Pfau, M.L., Leboeuf, M., Golden, S.A., Christoffel, D.J., Bregman, D., Rebusi, N., Heshmati, M., Aleyasin, H., Warren, B.L., Lebonte, B., Horn, S., Lapidus, K.A., Stelzhammer, V., Wong, E.H., Bahn, S., Krishnan, V., Bolanos-Guzman, C.A., Murrough, J.W., Merad, M., Russo, S.J., 2014. Individual differences in the peripheral immune system promote resilience versus susceptibility to social stress. Proc. Natl. Acad. Sci. U.S.A. 111, 16136–16141. Hong, B.Y., Maulen, N.P., Adami, A.J., Granados, H., Balcells, M.E., Cervantes, J., 2016. Microbiome changes during tuberculosis and antituberculous therapy. Clin. Microbiol. Rev. 29, 915–926. Hopfenspirger, M.T., Agrawal, D.K., 2002. Airway hyperresponsiveness, late allergic response, and eosinophilia are reversed with mycobacterial antigens in ovalbuminpresensitized mice. J. Immunol. 168, 2516–2522. Huang, C.T., Workman, C.J., Flies, D., Pan, X., Marson, A.L., Zhou, G., Hipkiss, E.L., Ravi, S., Kowalski, J., Levitsky, H.I., Powell, J.D., Pardoll, D.M., Drake, C.G., Vignali, D.A., 2004. Role of LAG-3 in regulatory T cells. Immunity 21, 503–513. Hunt, J.R., Martinelli, R., Adams, V.C., Rook, G.A., Brunet, L.R., 2005. Intragastric administration of Mycobacterium vaccae inhibits severe pulmonary allergic inflammation in a mouse model. Clin. Exp. Allergy 35, 685–690. Iwasaki, A., Kelsall, B.L., 1999. Freshly isolated Peyer’s patch, but not spleen, dendritic cells produce interleukin 10 and induce the differentiation of T helper type 2 cells. J. Exp. Med. 190, 229–239. Kilkenny, C., Browne, W.J., Cuthill, I.C., Emerson, M., Altman, D.G., 2010. Improving bioscience research reporting: the ARRIVE guidelines for reporting animal research. PLoS Biol. 8, e1000412. Kim, D.Y., Sato, A., Fukuyama, S., Sagara, H., Nagatake, T., Kong, I.G., Goda, K., Nochi, T., Kunisawa, J., Sato, S., Yokota, Y., Lee, C.H., Kiyono, H., 2011. The airway antigen sampling system: respiratory M cells as an alternative gateway for inhaled antigens. J. Immunol. 186, 4253–4262. Klingenberg, R., Lebens, M., Hermansson, A., Fredrikson, G.N., Strodthoff, D., Rudling, M., Ketelhuth, D.F., Gerdes, N., Holmgren, J., Nilsson, J., Hansson, G.K., 2010. Intranasal immunization with an apolipoprotein B-100 fusion protein induces antigen-specific regulatory T cells and reduces atherosclerosis. Arterioscler. Thromb. Vasc. Biol. 30, 946–952. Langgartner, D., Fuchsl, A.M., Uschold-Schmidt, N., Slattery, D.A., Reber, S.O., 2015. Chronic subordinate colony housing paradigm: a mouse model to characterize the consequences of insufficient glucocorticoid signaling. Front. Psychiatry 6, 18. Langgartner, D., Peterlik, D., Foertsch, S., Fuchsl, A.M., Brokmann, P., Flor, P.J., Shen, Z., Fox, J.G., Uschold-Schmidt, N., Lowry, C.A., Reber, S.O., 2017. Individual differences in stress vulnerability: the role of gut pathobionts in stress-induced colitis. Brain Behav. Immun. 64, 23–32. LeDoux, J.E., 1993. Emotional memory systems in the brain. Behav. Brain Res. 58, 69–79. Levenstein, S., Prantera, C., Varvo, V., Scribano, M.L., Andreoli, A., Luzi, C., Arca, M., Berto, E., Milite, G., Marcheggiano, A., 2000. Stress and exacerbation in ulcerative colitis: a prospective study of patients enrolled in remission. Am. J. Gastroenterol. 95, 1213–1220. Liang, J., Aihua, Z., Yu, W., Yong, L., Jingjing, L., 2010. HSP65 serves as an immunogenic carrier for a diabetogenic peptide P277 inducing anti-inflammatory immune response in NOD mice by nasal administration. Vaccine 28, 3312–3317. Lowry, C.A., Smith, D.G., Siebler, P.H., Schmidt, D., Stamper, C.E., Hassell Jr., J.E., Yamashita, P.S., Fox, J.H., Reber, S.O., Brenner, L.A., Hoisington, A.J., Postolache, T.T., Kinney, K.A., Marciani, D., Hernandez, M., Hemmings, S.M., Malan-Muller, S., Wright, K.P., Knight, R., Raison, C.L., Rook, G.A., 2016. The microbiota, immunoregulation, and mental health: implications for public health. Curr. Environ. Health Rep. 3, 270–286. Lukas, M., Neumann, I.D., 2012. Nasal application of neuropeptide S reduces anxiety and prolongs memory in rats: social versus non-social effects. Neuropharmacology 62, 398–405.

Funding This study was funded by the Office of Naval Research Global (N00014-17-S-B001). Declaration of Competing Interest MA, EK, TE, DL and SOR have nothing to declare. CAL serves on the Scientific Advisory Board of Immodulon Therapeutics. Acknowledgements The authors thank P. Hornischer and U. Binder for their technical assistance and help in performing the experiments. Furthermore, the authors would also like to thank Dr. S. Ott, E. Merkel and S. Hummel (local animal research center) for their excellent support in terms of animal housing. Author contributions SOR, DL, CAL planned the study; MA, DL performed the experiments; MA TE, EK, did the statistical analysis; MA, DL, CAL, SOR wrote the manuscript. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.bbi.2019.05.005. References Arnold, I.C., Dehzad, N., Reuter, S., Martin, H., Becher, B., Taube, C., Muller, A., 2011. Helicobacter pylori infection prevents allergic asthma in mouse models through the induction of regulatory T cells. J. Clin. Invest. 121, 3088–3093. Arnold, I.C., Hitzler, I., Muller, A., 2012. The immunomodulatory properties of Helicobacter pylori confer protection against allergic and chronic inflammatory disorders. Front. Cell. Infect. Microbiol. 2, 10. Bitton, A., Sewitch, M.J., Peppercorn, M.A., deB Edwardes, M.D., Shah, S., Ransil, B., Locke, S.E., 2003. Psychosocial determinants of relapse in ulcerative colitis: a longitudinal study. Am. J. Gastroenterol. 98, 2203–2208. Blair, H.T., Schafe, G.E., Bauer, E.P., Rodrigues, S.M., LeDoux, J.E., 2001. Synaptic plasticity in the lateral amygdala: a cellular hypothesis of fear conditioning. Learn. Memory 8, 229–242. Böbel, T.S., Hackl, S.B., Langgartner, D., Jarczok, M.N., Rohleder, N., Rook, G.A., Lowry, C.A., Gundel, H., Waller, C., Reber, S.O., 2018. Less immune activation following social stress in rural vs. urban participants raised with regular or no animal contact, respectively. Proc. Natl. Acad. Sci. U.S.A. 115, 5259–5264. Chen, Y., Kuchroo, V.K., Inobe, J., Hafler, D.A., Weiner, H.L., 1994. Regulatory T cell clones induced by oral tolerance: suppression of autoimmune encephalomyelitis. Science 265, 1237–1240. Chieppa, M., Rescigno, M., Huang, A.Y., Germain, R.N., 2006. Dynamic imaging of dendritic cell extension into the small bowel lumen in response to epithelial cell TLR engagement. J. Exp. Med. 203, 2841–2852. Coombes, J.L., Powrie, F., 2008. Dendritic cells in intestinal immune regulation. Nat. Rev. Immunol. 8, 435–446. Eraly, S.A., Nievergelt, C.M., Maihofer, A.X., Barkauskas, D.A., Biswas, N., Agorastos, A., O'Connor, D.T., Baker, D.G., Marine Resiliency Study, T., 2014. Assessment of plasma C-reactive protein as a biomarker of posttraumatic stress disorder risk. JAMA Psychiatry 71, 423–431. Foertsch, S., Fuchsl, A.M., Faller, S.D., Holzer, H., Langgartner, D., Messmann, J., Strauss, G., Reber, S.O., 2017. Splenic glucocorticoid resistance following psychosocial stress requires physical injury. Sci. Rep. 7, 15730. Frenkel, D., Huang, Z., Maron, R., Koldzic, D.N., Hancock, W.W., Moskowitz, M.A.,

9

Brain, Behavior, and Immunity xxx (xxxx) xxx–xxx

M. Amoroso, et al. Macovei, L., McCafferty, J., Chen, T., Teles, F., Hasturk, H., Paster, B.J., Campos-Neto, A., 2015. The hidden ‘mycobacteriome’ of the human healthy oral cavity and upper respiratory tract. J. Oral Microbiol. 7, 26094. Maron, R., Sukhova, G., Faria, A.M., Hoffmann, E., Mach, F., Libby, P., Weiner, H.L., 2002. Mucosal administration of heat shock protein-65 decreases atherosclerosis and inflammation in aortic arch of low-density lipoprotein receptor-deficient mice. Circulation 106, 1708–1715. McMenamin, C., Schon-Hegrad, M., Oliver, J., Girn, B., Holt, P.G., 1991. Regulation of IgE responses to inhaled antigens: cellular mechanisms underlying allergic sensitization versus tolerance induction. Int. Arch. Allergy Appl. Immunol. 94, 78–82. Miller, A.H., Raison, C.L., 2016. The role of inflammation in depression: from evolutionary imperative to modern treatment target. Nat. Rev. Immunol. 16, 22–34. Minne, A., Jaworska, J., Gerhold, K., Ahrens, B., Avagyan, A., Vanbever, R., Matricardi, P.M., Schmidt, A.C., Hamelmann, E., 2007. Intranasal delivery of whole influenza vaccine prevents subsequent allergen-induced sensitization and airway hyper-reactivity in mice. Clin. Exp. Allergy 37, 1250–1258. Mitra, R., Jadhav, S., McEwen, B.S., Vyas, A., Chattarji, S., 2005. Stress duration modulates the spatiotemporal patterns of spine formation in the basolateral amygdala. Proc. Natl. Acad. Sci. U.S.A. 102, 9371–9376. Nair, V.R., Franco, L.H., Zacharia, V.M., Khan, H.S., Stamm, C.E., You, W., Marciano, D.K., Yagita, H., Levine, B., Shiloh, M.U., 2016. Microfold cells actively translocate Mycobacterium tuberculosis to initiate infection. Cell Rep. 16, 1253–1258. O'Donovan, A., Cohen, B.E., Seal, K.H., Bertenthal, D., Margaretten, M., Nishimi, K., Neylan, T.C., 2015. Elevated risk for autoimmune disorders in Iraq and Afghanistan veterans with posttraumatic stress disorder. Biol. Psychiatry 77, 365–374. Park, H.S., Francis, K.P., Yu, J., Cleary, P.P., 2003. Membranous cells in nasal-associated lymphoid tissue: a portal of entry for the respiratory mucosal pathogen group A streptococcus. J. Immunol. 171, 2532–2537. Peen, J., Schoevers, R.A., Beekman, A.T., Dekker, J., 2010. The current status of urbanrural differences in psychiatric disorders. Acta Psychiatr. Scand. 121, 84–93. Perdomo, C., Zedler, U., Kuhl, A.A., Lozza, L., Saikali, P., Sander, L.E., Vogelzang, A., Kaufmann, S.H., Kupz, A., 2016. Mucosal BCG vaccination induces protective lungresident memory T cell populations against tuberculosis. MBio 7https://doi.org/10. 1128/mBio. 01686-16. e01686–16. Pervanidou, P., Kolaitis, G., Charitaki, S., Margeli, A., Ferentinos, S., Bakoula, C., Lazaropoulou, C., Papassotiriou, I., Tsiantis, J., Chrousos, G.P., 2007. Elevated morning serum interleukin (IL)-6 or evening salivary cortisol concentrations predict posttraumatic stress disorder in children and adolescents six months after a motor vehicle accident. Psychoneuroendocrinology 32, 991–999. Polanski, M., Melican, N.S., Zhang, J., Weiner, H.L., 1997. Oral administration of the immunodominant B-chain of insulin reduces diabetes in a co-transfer model of diabetes in the NOD mouse and is associated with a switch from Th1 to Th2 cytokines. J. Autoimmun. 10, 339–346. Rao, R.P., Anilkumar, S., McEwen, B.S., Chattarji, S., 2012. Glucocorticoids protect against the delayed behavioral and cellular effects of acute stress on the amygdala. Biol. Psychiatry 72, 466–475. Reber, S.O., Birkeneder, L., Veenema, A.H., Obermeier, F., Falk, W., Straub, R.H., Neumann, I.D., 2007. Adrenal insufficiency and colonic inflammation after a novel chronic psycho-social stress paradigm in mice: implications and mechanisms. Endocrinology 148, 670–682. Reber, S.O., Langgartner, D., Foertsch, S., Postolache, T.T., Brenner, L.A., Guendel, H., Lowry, C.A., 2016a. Chronic subordinate colony housing paradigm: A mouse model for mechanisms of PTSD vulnerability, targeted prevention, and treatment-2016 Curt Richter Award Paper. Psychoneuroendocrinology 74, 221–230.

Reber, S.O., Obermeier, F., Straub, R.H., Veenema, A.H., Neumann, I.D., 2008. Aggravation of DSS-induced colitis after chronic subordinate colony (CSC) housing is partially mediated by adrenal mechanisms. Stress 11, 225–234. Reber, S.O., Peters, S., Slattery, D.A., Hofmann, C., Scholmerich, J., Neumann, I.D., Obermeier, F., 2011. Mucosal immunosuppression and epithelial barrier defects are key events in murine psychosocial stress-induced colitis. Brain Behav. Immun. 25, 1153–1161. Reber, S.O., Siebler, P.H., Donner, N.C., Morton, J.T., Smith, D.G., Kopelman, J.M., Lowe, K.R., Wheeler, K.J., Fox, J.H., Hassell Jr., J.E., Greenwood, B.N., Jansch, C., Lechner, A., Schmidt, D., Uschold-Schmidt, N., Fuchsl, A.M., Langgartner, D., Walker, F.R., Hale, M.W., Lopez Perez, G., Van Treuren, W., Gonzalez, A., Halweg-Edwards, A.L., Fleshner, M., Raison, C.L., Rook, G.A., Peddada, S.D., Knight, R., Lowry, C.A., 2016b. Immunization with a heat-killed preparation of the environmental bacterium Mycobacterium vaccae promotes stress resilience in mice. Proc. Natl. Acad. Sci. U.S.A. 113, E3130–E3139. Rescigno, M., Rotta, G., Valzasina, B., Ricciardi-Castagnoli, P., 2001. Dendritic cells shuttle microbes across gut epithelial monolayers. Immunobiology 204, 572–581. Rogan, M.T., Staubli, U.V., LeDoux, J.E., 1997. Fear conditioning induces associative long-term potentiation in the amygdala. Nature 390, 604–607. Rohleder, N., 2014. Stimulation of systemic low-grade inflammation by psychosocial stress. Psychosom. Med. 76, 181–189. Ruane, D., Brane, L., Reis, B.S., Cheong, C., Poles, J., Do, Y., Zhu, H., Velinzon, K., Choi, J.H., Studt, N., Mayer, L., Lavelle, E.C., Steinman, R.M., Mucida, D., Mehandru, S., 2013. Lung dendritic cells induce migration of protective T cells to the gastrointestinal tract. J. Exp. Med. 210, 1871–1888. Samuelson, D.R., Welsh, D.A., Shellito, J.E., 2015. Regulation of lung immunity and host defense by the intestinal microbiota. Front. Microbiol. 6, 1085. Schafe, G.E., Nader, K., Blair, H.T., LeDoux, J.E., 2001. Memory consolidation of Pavlovian fear conditioning: a cellular and molecular perspective. Trends Neurosci. 24, 540–546. Schmidt, D., Reber, S.O., Botteron, C., Barth, T., Peterlik, D., Uschold, N., Mannel, D.N., Lechner, A., 2010. Chronic psychosocial stress promotes systemic immune activation and the development of inflammatory Th cell responses. Brain Behav. Immun. 24, 1097–1104. Slattery, D.A., Uschold, N., Magoni, M., Bar, J., Popoli, M., Neumann, I.D., Reber, S.O., 2012. Behavioural consequences of two chronic psychosocial stress paradigms: anxiety without depression. Psychoneuroendocrinology 37, 702–714. Sommershof, A., Aichinger, H., Engler, H., Adenauer, H., Catani, C., Boneberg, E.M., Elbert, T., Groettrup, M., Kolassa, I.T., 2009. Substantial reduction of naive and regulatory T cells following traumatic stress. Brain Behav. Immun. 23, 1117–1124. Sze, M.A., Tsuruta, M., Yang, S.W., Oh, Y., Man, S.F., Hogg, J.C., Sin, D.D., 2014. Changes in the bacterial microbiota in gut, blood, and lungs following acute LPS instillation into mice lungs. PLoS One 9, e111228. Unger, W.W., Hauet-Broere, F., Jansen, W., van Berkel, L.A., Kraal, G., Samsom, J.N., 2003. Early events in peripheral regulatory T cell induction via the nasal mucosa. J. Immunol. 171, 4592–4603. Young, M.B., Howell, L.L., Hopkins, L., Moshfegh, C., Yu, Z., Clubb, L., Seidenberg, J., Park, J., Swiercz, A.P., Marvar, P.J., 2018. A peripheral immune response to remembering trauma contributes to the maintenance of fear memory in mice. Psychoneuroendocrinology 94, 143–151. Zuany-Amorim, C., Sawicka, E., Manlius, C., Le Moine, A., Brunet, L.R., Kemeny, D.M., Bowen, G., Rook, G., Walker, C., 2002. Suppression of airway eosinophilia by killed Mycobacterium vaccae-induced allergen-specific regulatory T-cells. Nat. Med. 8, 625–629.

10