Murine splenic B cells express corticotropin-releasing hormone receptor 2 that affect their viability during a stress response

Chronic stress is now recognized as a risk factor for disease development and/or exacerbation. It has been shown to affect negatively the immune system and notably the humoral immune response. Corticotropin-releasing hormone (CRH) is known to play a crucial role in stress response. CRH receptors are expressed on different immune cells such as granulocytes, monocytes and T cells. However, up to now, no CRH receptor has been described on B cells which are key players of the humoral immune response. In order to highlight new pathways by which stress may impact immunity, we investigated the role of CRH in B cells. Here we show that splenic B cells express the CRH receptor 2 (CRHR2), but not CRHR1. This receptor is functional since CRH treatment of B cells activates different signaling pathways (e.g. p38) and decreases B cell viability. Finally, we show that immunization of mice with two types of antigens induces a more intense CRHR staining in secondary lymphoid organs where B cells are known to respond to the antigen. Altogether our results demonstrate, for the first time, that CRH is able to modulate directly B cell activity through the presence of CRHR2.

B cells are key players of humoral immunity through their ability to produce antibodies and enhance antibody affinity via somatic hypermutation 28 . This latter phenomenon contributes to a better protection of the organism. Depending on the nature of the antigen (T cell-dependent or T cell-independent), B cells require or not cooperation with T cells to mount their response. As T cells express CRHR, CRH can affect this cell type and consequently B cell responses in the case of T cell-dependent antigens (indirect action). However, it is also of crucial interest to determine if B cells can be directly affected by CRH. Some studies have tried to address this question but conflicting results were reported. Using human blood mononuclear cells, Leu and Singh showed that CRH inhibits antibody production while Smith et al., using murine splenocytes, showed that CRH enhances antigen-specific antibody production 29,30 . CRH microinjection into the lateral ventricle of rats was shown to slow down antibody induction in response to a T cell-dependent antigen both after primary or secondary immunization 31 . When CRH was injected intraperitoneally or subcutaneously, this diminution did not occur. Finally, it was shown that immunization of CRH transgenic (CRH-Tg) mice decreased humoral immune response and germinal center formation 32,33 . Taken together, these studies strongly suggest that CRH is able to modulate B cell activity. However, the presence of CRH receptors on these cells has never been reported. Thus, it is not known if CRH can acts directly on B cells.
As a first step to determine whether CRH action on B cells could be direct, we searched for CRHR on purified splenic murine B cells and discovered that these cells express CRHR2 but not CRHR1. These CRH receptors are functional as the addition of CRH to splenic B cells activates p38β, p38δ MAP kinases, GSK3β phosphorylation and the PKA/CREB signaling pathway. Secondly, we were interested in the CRH effect on splenocyte viability. In the presence of the hormone, splenic B cell viability was decreased while the one of T cell remained unchanged. Finally, we showed that immunizations led to a more specific CRHR labeling in secondary lymphoid organs correlating with B cell areas labeling. Taken together, these data show that CRH can act directly on B cells and affect their physiology.

Murine splenic B cells express CRHR2.
The presence of CRH receptors has been reported on murine splenic T cells and macrophages but not on splenic B cells. Thus, we investigated the presence of CRH receptors on B cells. As two types of CRH receptors have been described, CRHR1 and CRHR2, we wondered which one could be expressed by murine splenic B cells. To address this question, B cells were purified by negative selection from murine splenocytes and purity, determined by flow cytometry, was comprised between 92 and 97%. Then, RNA was extracted from these splenic B cells and RT-PCRs were performed with two pairs of primers defined in specific parts of CRHR1 and CRHR2 transcripts. As shown in Fig. 1a, no amplification was obtained with CRHR1 primers even after 40 cycles of amplification whereas amplicons were obtained with both pairs of CRHR2 oligonucleotides. Sequencing of these PCR products confirmed that amplicons were CRHR2 cDNA fragments demonstrating that murine B cells express only these receptor transcripts.
Then, we investigated whether CRHR2 protein expression level could be modulated by CRH. To address this question, splenocytes were cultured for 48 h with CRH 1 or 100 nM. As for RT-PCR experiments, B cells were purified by negative selection and the expression of CRHR2 was assessed by western blotting. Figure 1b shows that B cells express CRHR2 independently of the CRH treatment. Interestingly, two bands were observed and might correspond to different CRHR2 isoforms. To confirm CRHR2 expression by murine splenic B cells, splenocytes were cultured with 1 or 100 nM of CRH for 48 h and analyzed by immunofluorescence. As shown in Fig. 1c, CRHR2 was found to be expressed on B cells (CD19 + ). Altogether these results firmly demonstrate that splenic B cells express CRH receptors and more specifically CRHR2.

CRH activates different signaling pathways in murine splenic B cells. To determine if CRHR2
are functional, splenic B cells were purified and incubated with CRH 100 nM for different durations (15,30 or 60 min). Proteins were then extracted and applied onto a proteome phospho-MAPK array to screen signaling transduction pathways that can be activated by this hormone. No activation was noted after 15 min of incubation with CRH (Fig. 2). After 30 min, an increase of the phosphorylation levels of mTOR (≈ 20%), GSK3β and p38β (≈45%), and p38δ (≈100%) was observed in B cells treated with CRH compared to non CRH-treated B cells. Moreover, the phosphorylation level of CREB was increased by approximately 45% after 60 min of CRH incubation. These results indicate that different signaling pathways could be activated by CRH and that CRHR2 are functional in B cells.
CRH affects murine splenic B cell viability. As it was shown that CRH can induce peripheral blood lymphocyte apoptosis 34 , we next studied the effect of CRH on B cell viability. Splenocytes were cultured with or without CRH 1 or 100 nM for 48 h. Then, T and B cells were labeled with anti-CD3 and anti-CD19 antibodies, respectively, before (T0) and after 48 h of culture (T48). Flow cytometry analyses showed that the percentage of B cells decreased after 48 h of culture with or without CRH while the percentage of the T cells increased (≈ 60% of B cells and 30% of T cells at T0 with a T/B cells ratio of 0.48; ≈50% of B cells and 45% of T cells at T48 with a T/B cells ratio ranging from 0.88 to 0.96) (Fig. 3a and c). Analysis of Annexin-V (Anx-V) staining in both T and B cell populations showed that B cells were more sensitive to apoptosis than T cells (≈10% of B cells and 3% of T cells were Anx-V positive at T0 while 66% of B cells and 35% of T cells were Anx-V positive at T48 without CRH), (Fig. 3b and d). Furthermore, the treatment of splenocytes with CRH significantly increased the rate of Anx-V stained cells only in the B cell population (+6.5% and +5% with CRH 1 nM and 100 nM respectively). These last results demonstrated that, in vitro, CRH decreased B cell survival.
Scientific REPORtS | (2018) 8:143 | DOI:10.1038/s41598-017-18401-y CRHR expression is increased in the spleen after intraperitoneal immunization. As reduced humoral immune response and germinal center formation were noted after immunization of CRH-Tg mice, we performed in vivo experiments to further understand the function of CRH receptors on splenic B cells. Mice were immunized with two T cell-dependent antigens, BSA (bovine serum albumin) and NP-KLH (4-hydroxy-3-nitrophenylacetyl hapten conjugated to keyhole limpet hemocyanin), or with a T cell-independent antigen, LPS (lipopolysaccharide). Then, immunofluorescence staining was used to assess the expression of CRHR within the spleen (Fig. 4). In non-immunized mice, splenic CRHR labeling showed no precise localization. After immunization with BSA, CRHR staining was increased into B cell areas corresponding to follicles where B cells are known to respond to T cell-dependent antigens and lead to germinal center formation. This result did not depend on antigen composition because immunization with another T cell-dependent antigen, NP-KLH, led to the same CRHR staining localization (white arrows). After immunization with LPS, a more specific CRHR labeling was observed around B cell areas, corresponding to marginal zones (red arrows). In these areas, B cells are known to be activated with T cell-independent antigens. Taken together, these results suggest that whatever the antigen used for immune system activation, the splenic CRHR expression is more specifically localized in B cell areas.

Discussion
Over the past few years, it has become clear that CRH can exert direct effects on immune tissues, such as the spleen, through the presence of CRH receptors in these tissues. Nevertheless, direct CRH effects on leukocytes, notably on B cells, are not well known yet. Thus, the purpose of our study was to better understand the impact of CRH on B cells.
We first demonstrated by western blotting and immunofluorescence that splenic murine B cells express CRHR and, by sequencing of RT-PCR products, that they express only CRHR2 but not CRHR1. This result is very interesting because it shows for the first time that CRH can act directly on B cells to modulate their physiology during a stress response. Western blotting experiments performed with the anti-CRHR antibody revealed two bands. In rodents, different CRHR2 mRNAs are generated by alternative splicing leading to two isoforms: CRHR2α and CRH2β 35 . This alternative splicing leads to the deletion of exons 1 and 2 in CRHR2α mRNA and to the deletion of exon 3 in CRHR2β mRNA 36 . The predictive size of these two isoforms is 47 kDa for CRHR2α and 50 kDa for CRHR2β (UniProt; http://www.uniprot.org/uniprot/Q60748). Thus, the two bands observed in our western blotting experiments likely correspond to CRHR2α for the lower band and to CRHR2β for the higher band. This indicates that murine splenic B cells express both isoforms.
CRH is known for having a greater affinity for CRHR1 than for CRHR2 (Kd of 3 nM and 10-40 nM, respectively) 37 . However, different studies have demonstrated that CRH can act via its type 2 receptor on neurons or macrophages [38][39][40] . Once activated, rodent CRHR can activate different phospho-MAP kinases pathways like CREB (cyclic AMP response element-binding), p38 and GSK3β [41][42][43] . Consequently, we screened different signaling pathways that could be activated by CRH in murine B cells.
Based on the literature, cells were incubated with CRH 100 nM, to mimic high stress levels, because in vivo studies estimated that CRH level could rise above 100-200 nM in the hypothalamus and hippocampus of animals placed under stressing conditions 44 . Our results suggest that CRHR2 binding induces the activation of different MAP kinases pathways such as p38 (more precisely p38β and p38δ), GSK3β and CREB. Then, we further investigated the consequences of the presence of CRHR on B cell survival. Indeed, it has been shown that CRHR2 activation induces apoptosis in a murine macrophage cell line (RAW264.7) 43 . Our data show that both CRH 1 (to mimic mild stress levels) and 100 nM lead to an increase of murine splenic T/B cells ratio which is due to a decrease of B but not T cell viability. These results are consistent with in vitro studies showing that 1 nM of CRH could affect PC-12 rat adrenal cell line apoptosis and murine oocyte maturation 45,46 . Interestingly, the diminution of B cell survival correlates with proteome phospho-MAPK array results, showing the activation of different MAP kinases pathways such as p38, GSK3β and CREB known to be involved in cell death. Indeed, the increase of the phosphorylation levels of such protein kinases has been associated with apoptosis induction in various cell types such as p38δ in keratinocytes 47 or GSK3β in neurons 48 and hepatocytes 49 . In the same manner, in animal models of ischemia-reperfusion injury or neonatal hypoxia-ischemia, GSK3 inhibition protects the heart and the neurons respectively via the diminution of inflammation and apoptosis 50,51 . How CRH specifically impacts B cell survival remains unclear. However, based on our data and on the literature, one could speculate that such early signaling events induced by CRH might lead to the reduced B cell survival. Studies performed on CRH-Tg mice revealed the same alteration of T/B cells ratio in the spleen, associated with a B cell maturation blockage and a decrease of germinal center formation after immunization 32,33 . Nevertheless, in vivo studies did not dissociate CRH and glucocorticoid action. As CRH overexpression leads to glucocorticoid overproduction and because B cells express glucocorticoid receptors 53 , it is difficult to know if such results are due to a direct impact of CRH on lymphocytes. CRH could potentiate glucocorticoid action and vice-versa. Indeed, in our study, we observed an increase of glucocorticoid receptor transcription levels in murine splenocytes after CRH treatment during 48 h (Supplementary Fig. S2). However, during our in vitro studies, there was no glucocorticoid stimulation. Thus, the impact on B cell viability is the direct consequence of CRH action. Finally, we focused on the in vivo role of CRH in splenic B cell maturation and selection. Immunofluorescence staining of spleen sections from mice immunized with different antigens was performed. In non-immunized mice, immunostaining did not show any CRHR specific localization. This result is not surprising because other splenic cells such as macrophages, dendritic cells and T cells are known to express CRHR 54,55 . After immunization with T cell-dependent antigens (BSA or NP-KLH), CRHR staining was more localized in B cell follicles where germinal centers can develop after B cell activation. Germinal center formation, necessary for B cell maturation and selection, takes place in response to T cell-dependent antigens. After immunization with a T cell-independent antigen (LPS), CRHR staining seemed to be situated in marginal zones located around B cell follicles. These observations suggest that, in addition of being a stress hormone able to act directly on B cells, CRH could also play a role in B cell selection and maturation during immunization/infection. This hypothesis could explain why CRH-Tg mice exhibit a diminution of germinal center formation in the spleen, associated with a decrease of immunoglobulin class switching and antibody affinity maturation 32,33 ; this last observation having also been done in stressed animals 56 .

Conclusion and Perspectives
In conclusion, we show for the first time the expression of CRHR2 on splenic murine B cells thereby rendering them more sensitive to death induced by CRH produced during stress responses. Moreover, the increase of CRHR expression in or around B cell follicles after immunization raises the question of the function of CRH in B cell maturation and selection during immune responses. In the future, it would be interesting to determine if peripheral blood B cells also express CRHR2. Indeed, when a tissue is injured, inflammation occurs and permits not only the entry of circulating leukocytes in the stressed tissue but also the increase of HPA axis activity. As this activation leads to an increase of CRH production, it will be important to investigate if this overproduction modulates the physiology of infiltrated leukocytes, especially in the brain where CRH concentration is high. Indeed, neuro-inflammation studies have shown that blood B cells infiltrating the brain could cause deleterious or protective effects depending on the B cell subset [57][58][59] .

Methods
Mice. C57Bl/6J male mice aged 14-18 weeks (Charles River Laboratories, Saint Germain Nuelles, France) were housed in vented animal cabinets (Noroit, Bouaye, France) under controlled temperature (22 °C) and 12 h light-dark cycle with free access to food and water. Animals were treated in accordance with the national legislation. Experimental protocols were approved by the local ethics committee (Comité d'Ethique Lorrain en Matière d'Expérimentation Animale, permit number: CELMEA-2012-008).
Cell culture and B cell isolation. Mice were anaesthetized with isoflurane (AbbVie, Rungis, France) and euthanized by cervical dislocation. Each spleen was collected and dissociated with 70 μm nylon cell EASYstrainer (Greiner bio-one, Dutscher, Brumath, France). Then, splenic red blood cells were lyzed into 2 mL of 1X RBC lyzis buffer (eBioscience, Affymetrix, Rennes, France) for 2 min at 37 °C and the reaction was stopped by adding 10 mL of RPMI 1640 medium. After centrifugation at 300 g during 5 min, the pellet was suspended in RPMI 1640 medium enriched with 10% heat inactivated foetal calf serum (Sigma-Aldrich, L'Isles d' Abeau Chesnes, France), 100 U/mL penicillin, 100 μg/mL streptomycin, 10 mM HEPES, 2 mM L-glutamine, 1 mM sodium pyruvate and 1X non-essential amino-acids (all purchased from Sigma-Aldrich). After cell count, splenocytes were cultured at a density of 1 × 10 6 cells/mL at 37 °C under 5% CO 2 for 48 h with or without CRH 1 or 100 nM (to mimic mild and high stress levels, respectively), (Bachem, Interchim, Montluçon, France). After incubation with CRH, splenic murine B cells were purified using the MagCellect TM Mouse B cell isolation kit according to manufacturer's instructions (Stemcell Technologies, Grenoble, France) to perform western blotting experiments. B cell purity was checked by flow cytometry after co-staining with anti CD19-PC7 and anti CD3-APC antibodies. The degree of purity was comprised between 92 and 97%. For RNA extraction followed by RT-PCR and for protein array experiments, B cells were directly purified from mice spleen with the MagCellect TM Mouse B cell isolation kit.

RT-PCR.
Total RNA was extracted from purified splenic B cells using Trizol reagent (Invitrogen, Thermo Purification and sequencing of amplicons. CRHR1 and CRHR2 PCR products were separated on an agarose gel stained with ethidium bromide. Amplicons were excised and purified using the lysis NucleoSpin Gel and PCR clean-up kit (Macherey-Nagel, Hoerdt, France). Then, purified PCR products were sent for sequencing to GATC Biotech (Mulhouse, France).
Protein preparation. After CRH stimulation, purified splenic B cells were suspended in NP-40 cell lysis buffer (150 mM sodium chloride, 50 mM Tris pH 8.0, 1% NP-40) and protein extraction was performed in the presence of a complete protease and phosphatase inhibitor mixture (Roche Molecular Biochemicals, Mannheim, Germany). Protein concentrations were determined using a Coomassie protein assay (Bio-Rad, Ivry sur Seine, France).
Scientific REPORtS | (2018) 8:143 | DOI:10.1038/s41598-017-18401-y Protein Array. Purified B cells from a pool of three spleens were incubated with or without 100 nM of CRH during 15, 30 or 60 min. Proteins were then extracted and CRH-activated signaling pathways were analyzed using the "proteome profiler human phospho-MAPK array" kit according to manufacturer's instructions (R&D Systems, Lille, France). 80% of the antibodies present on this array can detect murine proteins according to the manufacturer. Signals were visualized by chemiluminescence (FX7, Vilbert-Lourmat) and analyzed by densitometry (ImageJ ® ).
Immunization of mice and immunostaining of spleen sections. Mice were immunized intraperitoneally with 10 µg of lipopolysaccharide (LPS, Sigma-Aldrich), 100 µg of BSA or 50 µg of 4-hydroxy-3-nitrophenylacetyl hapten conjugated to keyhole limpet hemocyanin (NP-KLH, LGC BioSearch Technologies, Steinach, Germany). All antigens were mixed with an equal volume of Freund's adjuvant (final volume of 100 µL; Sigma-Aldrich). Ten days later, animals were euthanized. After fixation by whole body perfusion, spleens were incubated for 24 h in 4% paraformaldehyde followed by 12 h in PBS containing 30% of sucrose. Frozen spleens were cut using a cryostat to obtain 14 μm slice sections that were fixed on polylysine glass slides (Thermo Scientific Fischer). Sections were permeabilized, blocked as described above and stained overnight at 4 °C with anti-CD19 (1/100, eBioscience) and anti-CRHR antibodies (1/50, BIOSS USA, Interchim) diluted in PBS-BSA 0.5% containing 0.1% of Triton X-100 and 5% of goat serum. After washing, secondary antibodies were applied as described above except that incubation lasted 2 h. Hoechst staining was also performed as described above. Signals were detected using an epifluorescence microscope Eclipse 80i (Objective 20x; Nikon). Statistical analysis. Homogeneity of variances and normality of distribution were controlled with Levene and Kolmogorov-Smirnov tests, respectively. The level of statistical significance was set at p ≤ 0.05. Intergroup differences were estimated by analysis of variances with one-way ANOVA tests followed by pairwise comparisons with Fisher's PLSD tests. All statistical tests were performed using the Statview ® software (SAS institute Inc, Cary, USA).