Corticotropin-Releasing Hormone Suppresses Synapse Formation in the Hippocampus of Male Rats via Inhibition of CXCL5 Secretion by Glia

Corticotropin-Releasing Hormone Suppresses Synapse Formation in the Hippocampus of Male Rats via... Abstract Corticotropin-releasing hormone (CRH) is believed to play a critical role in stress-induced synaptic formation and modification. In the current study, we explored the mechanisms underlying CRH modulation of synaptic formation in the hippocampus by using various models in vitro. In cultured hippocampal slices, CRH treatment decreased synapsin I and postsynaptic density protein 95 (PSD95) levels via CRH receptor type 1 (CRHR1). In isolated hippocampal neurons, however, it increased synapsin I–labeled presynaptic terminals and PSD95-labeled postsynaptic terminals via CRHR1. Interestingly, the inhibitory effect of CRH on synapsin I–labeled and PSD95-labeled terminals occurred in the model of neuron–glia cocultures. These effects were prevented by CRHR1 antagonist. Moreover, treatment of the neurons with the media of CRH-treated glia led to a decrease in synaptic terminal formation. The media collected from CRH-treated glial cells with CRHR1 knockdown did not show an inhibitory effect on synaptic terminals in hippocampal neurons. Unbiased cytokine array coupled with confirmatory enzyme-linked immunosorbent assay revealed that CRH suppressed C-X-C motif chemokine 5 (CXCL5) production in glia via CRHR1. Administration of CXCL5 reversed the inhibitory effects of CRH-treated glia culture media on synaptic formation. Our data suggest that CRH suppresses synapse formation through inhibition of CXCL5 secretion from glia in the hippocampus. Our study indicates that glia–neuron intercommunication is one of the mechanisms responsible for neuronal circuit remodeling during stress. Corticotropin-releasing hormone (CRH), a 41-amino-acid neuropeptide, acts as a principal component of the hypothalamic–pituitary–adrenal axis (1). In addition to hypothalamus, where CRH was originally identified, CRH is widely expressed in the central nervous system, including the neocortex, basal ganglia, amygdala, and hippocampus (2, 3). Emerging evidence indicates that CRH is the key mediator in stress-related brain disorders and impaired learning and memory. These effects result primarily from neuroplasticity including neurogenesis, synaptic alternation, and formation (4–6). The effects of CRH on synapse formation and integrity vary between studies, in particular with variety in different brain regions. For instance, CRH increases spine and synapse formation in the cerebellum (7), whereas it mediates stress-induced rapid loss of apical dendritic spines in CA1 and CA3 pyramidal cells of the hippocampus (8, 9). In corticotropin-releasing hormone receptor type 1 (CRHR1) knockout mice, the absence of CRHR1 prevents the detrimental effects of chronic stress on dendritic arborization of CA3 neurons and spatial memory (5, 10). Moreover, the effects of CRH on synaptogenesis in cultured neurons conflict with those in organotypic culture of brain slices. CRH increases dendrite outgrowth in cultured hippocampal neurons (11); in contrast, it suppresses dendrite growth in cultured slices of the hippocampus (12). Thus, the mechanisms by which CRH modulates synapse formation remain largely unknown. An increasing body of evidence has strongly suggested that glial cells play critical roles in the formation, stabilization, and elimination of synapses in both the peripheral and central nervous systems (13–15). Such effects are mediated by soluble signals secreted by glia. A number of secreted factors of glia, such as thrombospondins, cholesterol with apolipoprotein E, glypican 4 and 6, transforming growth factor β, brain-derived neurotrophic factor (BDNF), and tumor necrosis factor α, have recently been identified to regulate various aspects of synapse formation (16–21). Interestingly, some studies have demonstrated that CRH can regulate secretion of a number of factors, such as tumor necrosis factor α and BDNF, in glial cells (19, 20). We hypothesized that glia might be involved in CRH modulating synapse formation and modification. To test this hypothesis, we set up a series of experiments by using various models in vitro. Given that synapsin I links synaptic vesicles to cytoskeletal elements within the presynaptic terminal (22–24) and postsynaptic density protein 95 (PSD95) orients perpendicular to the postsynaptic membrane (25, 26), we used synapsin I and PSD95 as presynaptic and postsynaptic markers, respectively. First, we compared the effects of CRH on presynaptic and postsynaptic related proteins, such as synapsin I and PSD95, in organotypic culture of hippocampal slices and synapsin I- and PSD95-labeled terminals formation in cultured hippocampal neurons. Second, we explored whether glia contributed to CRH modulation of synapse formation in the model of neuron–glia cocultures and neuron cultures with glia-conditioned media (GCM), and subsequently we identified the key mediators produced by glia. Finally, we defined the signaling pathways responsible for CRH regulating cytokine release in glia. Materials and Methods Organotypic hippocampal slice culture All animal protocols were approved by the Ethical Committee of Experimental Animals of Second Military Medical University, China. Protocols were designed to minimize the number of animals used and their suffering. Sprague-Dawley rats were obtained from Sino-British SIPPR/BK Laboratory Animal Ltd, Shanghai, China. Hippocampal slice cultures were prepared from male rats on postnatal (P) days P2 and P3. In each independent culture, six to eight male newborn rats from same litter were used. The cultures were followed by the method described by Bender et al. (27, 28). Briefly, after decapitation rat brains were removed and placed in ice-cold oxygenated low-sodium artificial cerebral spinal fluid (containing 248 mM sucrose, 4 mM KCl, 1.25 mM NaH2PO4, 26.2 mM NaHCO3, 1 mM CaCl2, 5 mM MgCl2, and 10 mM glucose) and then carefully placed on the platform of a tissue chopper and sliced perpendicular to their longitudinal axes with a vibrating microtome (NVSLM1; World Precision Instruments Inc., Sarasota, FL), with 400 μm thickness of each slice. Slices were transferred to Millicell CM membrane inserts (Millipore, Bedford, MA) in six-well culture plates. Each well contained 1.2 mL of prewarmed Dulbecco’s modified Eagle medium (DMEM) (Invitrogen Corp., Carlsbad, CA) containing 20% horse serum (Invitrogen), 10.5 mM glucose, 12.5 mM HEPES, and 55 mM NaHCO3 (pH 7.3 to 7.4). The slices were incubated in a humidified, 5% CO2 atmosphere at 37°C overnight, and followed by treatments with increasing concentrations (10−12 to 10−8 M) of CRH (Sigma-Aldrich, St. Louis, MO) in the absence or presence of CRHR1 antagonist antalarmin (Sigma-Aldrich) and CRHR2 antagonist astressin2B (Sigma-Aldrich) for 48 hours. The dosages of CRH were selected according to the literature (29) and our previous studies (11). The media and slices were collected for subsequent analysis. CRH and astressin2B were dissolved in phosphate-buffered saline (PBS) and stored in 0.1-mM stock at −80°C, and antalarmin was dissolved in dimethyl sulfoxide to achieve a stock of 10 mM, then diluted by culture medium to achieve the final concentration of dimethyl sulfoxide <0.01%. Hippocampal neuron culture Primary hippocampal neurons were cultured as described previously (30). Briefly, hippocampi were dissected from P1 male Sprague-Dawley rats (six to eight male newborn rats from same litter were used in each independent culture) in ice-cold dissection solution containing 136 mM NaCl, 5.4 mM KCl, 0.2 mM Na2HPO4, 2 mM KH2PO4, 16.7 mM glucose, 20.8 mM sucrose, 10 mM HEPES, and 0.0012% phenol red, pH 7.4, and then incubated with 0.125% trypsin (Invitrogen) at 37°C for 15 minutes. Cell suspensions were obtained by mechanical dissociation in DMEM supplemented with 10% heat-inactivated fetal bovine serum (FBS) (Invitrogen) and 10% horse serum. Cells were plated at a density of 1 × 105 cells/cm2 on poly-l-lysine (Sigma-Aldrich) coated coverslips and six-well plates. Cultures were then maintained in 5% CO2 at 37°C in DMEM containing 10% heat-inactivated FBS and 10% horse serum overnight. Cytosine arabinoside (1 μM) was added into cultures 24 hours after plating to block the proliferation of glial cells, and the culture media were changed to serum-free B27/neurobasal medium (Invitrogen). Half of the medium was replaced with fresh medium every 3 days. After 7-day culture, cells were treated with various concentrations of CRH in the absence or presence of antalarmin and astressin2B for 48 hours. In some cases, cells were incubated with GCM or GCM plus recombinant C-X-C motif chemokine 5 (CXCL5) (rLIX, R&D Systems, Minneapolis, MN), antirat CXCL5 antibody (R&D Systems), or CXCR2 antagonist SB265610 (Sigma-Aldrich) for 48 hours. Culture of glial cells Cultures of mixed glia were established according to the method described previously (19, 31). Briefly, hippocampi were prepared from P2 to P3 male Sprague-Dawley rats (four to six male newborn rats from same litter were used in each independent culture) in ice-cold dissection solution and were dissociated by trypsin (0.125%) at 37°C for 15 minutes. After mechanical dissociation in DMEM supplemented with 15% heat-inactivated FBS, single-cell suspensions were obtained and then plated in 75-cm2 culture flasks and maintained at 37°C in a 5% CO2 humidified incubator. Half of the medium was replaced by fresh medium every 3 days for a total of 12 to 14 days. Mixed glial cells in flasks were trypsinized, counted, spun down, and resuspended in culture media and replaced onto poly-l-lysine–coated coverslips, 12-well plates, and transwell inserts. After 4-day culture, the cultures were administrated with CRH, antalarmin, and astressin2B for the indicated time. In some cases, cells were treated with the aforementioned reagents for 48 hours, and then media were harvested and spun down to remove cell debris. These media served as GCM to be used for treatment of isolated neurons. To purify astrocytes from mixed glial cells, cells were allowed to grow for 14 to 21 days. After reaching confluence, cells were shaken at 250 rpm for 18 to 20 hours in DMEM containing 10% FBS on a gyratory shaker at 37°C to remove neurons and other cell types. The cells were then plated onto poly-l-lysine–coated 12-well and 6-well plates and maintained in DMEM containing 10% FBS at 37°C and 5% CO2. Purified astrocytes were treated with various concentrations of CRH in combination with antalarmin, astressin2B, H89, Gö6976, U73122, chelerythrine, forskolin, SQ22536, 8-Br-cAMP, and pertussis toxin (PTX) for the indicated time. All the aforementioned reagents were purchased from Sigma-Aldrich. The purity of astrocytes was assessed by imunocytochemistry with anti-GFAP antibody. It was shown that >95% of such cells GFAP positive (Supplemental Fig. 1). Glia–neuron coculture Glia–neuron coculture was performed in a six-well transwell chamber system with a 4-μm pore size (Corning, Corning, NY), which permitted cell contact–independent communication via diffusible soluble factors only. Freshly isolated mixed glial cells were plated in transwell inserts at 5 × 104 cells per insert. The inserts then were placed on top of each well of the six-well plates, where neurons were cultured and maintained throughout the culture time until subsequent experiments were performed. Immunofluorescence analysis Cultured neurons, glia, and astrocytes were fixed in 4% paraformaldehyde at room temperature. The cultured cells were washed with PBS and incubated with 10% bovine serum albumin for 2 hours and then were incubated with primary antibodies antisynapsin I (AB1543P, 1:200; Millipore, Billerica, MA), anti-PSD95 (ab2723, 1:200; Abcam, Cambridge, UK), anti-CRHR1 (BS2590, 1:100; Bioworld Technology Inc., St. Louis Park, MN), anti-CRHR1 (TA313693; 1:200, OriGene Technologies Inc., MD), and anti-GFAP (sc6170, 1:200; Santa Cruz Biotechnology Inc., Santa Cruz, CA), in PBS containing 2% bovine serum albumin overnight at 4°C. Subsequently, the specimens were incubated with anti-rabbit IgG conjugated to Alexa Fluor 488 (A21206, 1:400; Thermo Fisher Scientific, Waltham, MA) or anti-goat IgG Alexa Fluor 488 (A-11055, 1:400; Thermo Fisher Scientific), anti-rabbit IgG Alexa Fluor 594 (R37119, 1:400; Thermo Fisher Scientific), or anti-mouse IgG Alexa Fluor 594 (R37115, 1:400; Thermo Fisher Scientific). The cell nuclei were visualized by applying the DNA-specific dye Bisbenzimide Hoechst 33342 (23491-52-3, 1:5000; Sigma-Aldrich). Staining images were visualized with a Leica confocal microscope (Lecia Microsystems Inc., Buffalo, NY). Quantification of synapsin I clusters, PSD95 clusters, and synapsin I–PSD95 clusters of cultured neurons were performed in Image J software. These clusters were calculated in individual neurons with easily distinguishable processes, as described in a previous study (32). The intensity of synapsin I, PSD95, and synapsin I–PSD95 overlapping positive clusters per region of interest (ROI) was taken along the same neuronal branch. At least five visual fields were acquired in each independent culture. The mean values of the positive clusters or costained contacts in each culture were determined. Western blot analysis Cells were scraped off the plates in the presence of lysis buffer, which consists of 60 mM Tris-HCl, 2% sodium dodecyl sulfate (SDS), 10% sucrose, 2 mM phenylmethylsulfonyl fluoride (Merck, Darmstadt, Germany), 1 mM sodium orthovanadate (Sigma-Aldrich), and 10 g/mL aprotinin (Bayer, Leverkusen, Germany). Hippocampal tissues were homogenized in the presence of the aforementioned lysis buffer. The lysates were quickly sonicated, boiled for 5 minutes at 95°C, and centrifuged at 12,000g for 5 minutes at 4°C. Then, the supernatants were collected and the protein concentration in supernatant was assayed with a modified Bradford assay. The samples were diluted in sample buffer [250 mM Tris-HCl (pH 6.8), containing 4% SDS, 10% glycerol, 2% β-mercaptoethanol, and 0.002% bromophenol blue] and boiled for another 5 minutes. Aliquots of the samples were separated by 10% SDS polyacrylamide gel electrophoresis and subsequently transferred to nitrocellulose membranes by electroblotting. The membrane was blocked in 5% skim milk powder in 0.1% Tris-buffered saline/Tween 20 at room temperature for 2 hours, and then was incubated with antibodies against synapsin I, PSD95, CRHR1 (Bioworld Technology Inc.), phospholipase C (PLC)-β3 (sc133231; Santa Cruz), and phosphorylated PLC-β3 (Ser 537) (2481; Cell Signaling Technology, Boston, MA) at a dilution of 1:1000 overnight at 4°C. After three washes with 0.1% Tris-buffered saline/Tween 20, the membranes were incubated with a secondary horseradish peroxidase conjugated antibody (Santa Cruz) for 1 hour at room temperature. Immunoreactive proteins were detected with and enhanced chemiluminescence Western blot detection system (Santa Cruz) and visualized with a Sygene Bio Image system (Synoptics Ltd., UK). To control sampling errors, the ratio of band intensities to β-actin (A5441; Sigma-Aldrich) was obtained to quantify the relative protein expression level. Cell viability assay To examine whether the reagents have detrimental effects on neurons, viability was tested. The cells were incubated with 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide (Sigma-Aldrich) at 0.5 mg/mL for 2 hours after treatment with various reagents. Then, the cells were lysed with dimethyl sulfoxide in Sorensen glycine buffer and the formazan crystals solubilized. Absorbance was read at 570 nm with a spectrophotometric microplate reader (Bio-Rad Laboratories, Inc., Hercules, CA). Cytokine array and enzyme-linked immunosorbent assay Glial cells were treated with CRH (10−8 M) in the presence or absence of antalarmin (10−7 M) for 48 hours, and the media were collected. The media of three independent cultures were pooled together for following cytokine array. The output of total of 34 cytokines in media was evaluated with a rat cytokine antibody array G series (Ray Biotech Inc., Norcross, GA) according to the manufacturer’s instructions. The concentration of CXCL5 (i.e., LIX) in media of cell cultures and brain slices was determined with a RayBio Rat LIX enzyme-linked immunosorbent assay (ELISA) kit (Ray Biotech Inc.) according to the manufacturer’s instructions. The LIX kit shows no cross-reactivity with other cytokines including CINC-2, CINC-3, CNTF, fractalkine, IL-1α, IL-1β, IL-4, IL-6, IL-10, GM-CSF, leptin, CCL-2, β-NGF, TIMP-1, VEGF, TNF-α, INF-γ, and MIP-3α in rats. The sensitivity of the LIX assay was <15 pg/mL. Intra-assay variation is <10%, and interassay variation is <12%. Small interfering RNA transfection The sequence-specific small interfering RNA (siRNA) targeting rat CRHR1 was designed and synthesized by GenePharma Corporation (Shanghai, China). The siRNA targeting CRHR1 (sense 5′-CCGCCUACAAUUACUUCCATT-3′; antisense 5′-UGGAAGUAAUUGUAGGCGGTT-3′) was used for knockdown of the CRHR1 gene in mixed glia in vitro. The following siRNA (sense 5′-UUCUCCGAACGUGUCACGUTT-3′; antisense 5′-ACGUGACACGUUCGGAGAATT-3′), without homology to any known rat messenger RNA sequences in the NCBI RefSeq database, was used as a negative control. The cultured glial cells were transfected with siRNA-CRHR1 and negative control siRNA with siPORT™ NeoFX™ transfection agent (Ambion, Austin, TX). Determination of activated Gi protein The level of active GTP-bound Gi protein was determined with a commercial Gi activation assay kit (NewEast Biosciences, Malvern, PA). Cultured astrocytes were treated with increasing concentrations of CRH (10−12 to 10−8 M) in the absence and presence of CRHR1 antagonist antalarmin (10−7 M) for 5 minutes and then scraped off the plate in the presence of lysis buffer supplied with the kit. The cell lysate was centrifuged for 10 seconds at 12,000g, and then all supernatants were immunoprecipitated with antiactive Gi monoclonal antibody, and protein A/G agarose beads were added to cell lysates. After incubation at 4°C for 1 hour, the beads were pelleted and resuspended three times (10 minutes each) in lysis buffer. Then the samples were centrifuged for 10 seconds at 12,000g, and the supernatant containing active GTP-bound Gi protein was collected. Samples were analyzed by Western blot with anti-Gi monoclonal antibody. To control sampling errors, the total Gi protein was determined by Western blot analysis. cAMP assay Astrocytes were treated with increasing concentrations of various reagents for 10 minutes, then scraped off the plate in the presence of 50 mM sodium acetate (pH 4.75). In some cases, the cells were transfected with CRHR1 siRNA and then treated with CRH (10−8 M) for 10 minutes. Lysates were boiled at 95°C for 10 minutes and then quickly sonicated in an ice bath. The supernatants were collected by centrifugation and used for cAMP assay according to the protocol of a commercial 125I RIA kit (Huaying Biotechnology Research Institute, Peking, China). The sensitivity of cAMP assay was <0.8 pmol/mL. Intra-assay and interassay variations are <8.9%. The kit shows no cross-reactivity with other small molecules in rats. IP3 assay Astrocytes were treated with increasing concentrations of various reagents for 10 minutes, then scraped off the plate in the presence of 50 mM sodium acetate (pH 4.75). In some cases, the cells were transfected with CRHR1 siRNA and then treated with CRH (10−8 M) for 10 minutes. After the cells were frozen three times, the supernatants were collected by centrifugation (3000g for 20 minutes). IP3 content in supernatants was determined with the IP3 ELISA kits (R&D Systems, Minneapolis, MN). The sensitivity of the IP3 assay was <2 pg/mL. There is no cross-reactivity with other small molecules in this kit. The interassay and intra-assay variations are <10%. To avoid the errors that would occur in different assays, IP3 contents were determined by bulk assay of the cells with various treatments. Statistical analysis The n value represents the numbers of independent cultures. All data are presented as mean ± standard error of the mean (SEM). After confirmation of normal distribution, one-way analysis of variance (ANOVA) followed by least significant difference (LSD)-t test or factorial design ANOVA was used to appropriately assess the difference in all variables. SPSS software version 16.0 was used. P < 0.05 was considered statistically significant. Results CRH suppresses presynaptic and postsynaptic related protein levels in hippocampal slices Treatment of cultured hippocampal slices with CRH (10−12 to 10−8 M) for 48 hours significantly suppressed synapsin I and PSD95 expression in a dose-dependent manner. The significant effect occurred at a concentration of 10−11 M in the synapsin I level. The highest effect was achieved at 10−8 M, with 58.3% ± 3.5% reduction of the synapsin I level and 52.9% ± 5.1% reduction of the PSD95 level [Fig. 1(a) and 1(b), synapsin I, F(5, 24) = 34.29, P < 0.001, CRH 10−8 M vs vehicle, LSD-t(24) = −10.12, P < 0.001; PSD95, F(5, 24) = 40.22, P < 0.001, CRH 10−11 M vs vehicle, LSD-t(24) = −11.39, P < 0.001]. The specific CRHR1 antagonist antalarmin reversed CRH inhibition of synapsin I and PSD95 protein expression [Fig. 1(c), synapsin I, F(5, 24) = 49.68, P < 0.001; CRH plus antalarmin: 92.9% ± 5.3% vs CRH: 65.8% ± 3.6%, LSD-t(24) = 2.21, P = 0.037; Fig. 1(d), PSD95, F(5, 24) = 60.17, P < 0.001; CRH plus antalarmin: 95.5% ± 4.3% vs CRH: 55.8% ± 3.4%, LSD-t (24) = 2.45, P = 0.022]. CRHR2 antagonist astressin2B did not affect the effect of CRH. Neither antalarmin nor astressin2B alone has an impact on synapsin I or PSD95 expression. Figure 1. View largeDownload slide CRH suppressed presynaptic and postsynaptic protein expression in hippocampal slices. Hippocampal slices were treated with increasing concentrations of CRH in the presence or absence of antalarmin and astressin2B for 48 hours. The protein levels of synapsin I and PSD95 were determined by Western blot analysis. (a and b) CRH (10−12–10−8 M) dose-dependently suppressed (a) synapsin I (n = 5 cultures) and (b) PSD95 (n = 5 cultures) protein level in cultured hippocampal slices. (c and d) Effects of CRH receptor antagonists on the expression of (c) synapsin I (n = 5 cultures) and (d) PSD95 (n = 5 cultures) protein levels induced by CRH in cultured hippocampal slices. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM. Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; #P < 0.05 vs CRH. Anta, antalarmin; As2B, astressin2B. Figure 1. View largeDownload slide CRH suppressed presynaptic and postsynaptic protein expression in hippocampal slices. Hippocampal slices were treated with increasing concentrations of CRH in the presence or absence of antalarmin and astressin2B for 48 hours. The protein levels of synapsin I and PSD95 were determined by Western blot analysis. (a and b) CRH (10−12–10−8 M) dose-dependently suppressed (a) synapsin I (n = 5 cultures) and (b) PSD95 (n = 5 cultures) protein level in cultured hippocampal slices. (c and d) Effects of CRH receptor antagonists on the expression of (c) synapsin I (n = 5 cultures) and (d) PSD95 (n = 5 cultures) protein levels induced by CRH in cultured hippocampal slices. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM. Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; #P < 0.05 vs CRH. Anta, antalarmin; As2B, astressin2B. CRH promotes synapse formation in isolated hippocampal neuron cultures but suppresses synapse formation in glia–neuron cocultures In a model of hippocampal neuron culture, CRH treatment (10−8 M) for 48 hours significantly increased synapsin I–labeled terminals and PSD95-labeled signals along the neuronal branches compared with vehicle control. Accordingly, the numbers of synapsin I–PSD95 costained synapses were upregulated by CRH treatment compared with vehicle control [Fig. 2(a) and 2(b), synapsin I: F(5, 24) = 80.12, P < 0.001; CRH: 9.1 ± 1.0 vs vehicle: 4.8 ± 0.2, LSD-t(24) = 11.37, P < 0.001; PSD95: F(5, 24) = 72.64, P < 0.001; CRH: 9.8 ± 0.8 vs vehicle: 4.8 ± 0.3, LSD-t(24) = 12.46, P < 0.001; costained synapses: F(5, 24) = 122.14, P < 0.001, CRH: 8.9 ± 0.9 vs vehicle: 4.8 ± 0.2, LSD-t(24) = 14.89, P < 0.001]. These effects were prevented by the CRHR1 specific antagonist antalarmin (10−7 M) but not by the CRHR2 antagonist astressin2B (10−7 M) [synapsin I: CRH plus antalarmin: 4.0 ± 0.5 vs CRH: 9.1 ± 1.0, LSD-t(24) = −14.10, P < 0.001; PSD95: CRH plus antalarmin: 4.3 ± 0.7 vs CRH: 9.8 ± 0.8, LSD-t(24) = −13.69, P < 0.001; costained synapses: CRH plus antalarmin: 4.0 ± 1.1 vs CRH: 8.9 ± 0.9, LSD-t (24) = −17.32, P < 0.001]. Figure 2. View largeDownload slide CRH stimulated synapse formation in cultured hippocampal neurons. (a and b) Hippocampal neurons were treated with CRH in the presence or absence of antalarmin and astressin2B for 48 hours. The cells were then used for immunofluorescence analysis of synapse terminals. Presynaptic terminals were labeled by synapsin I (green). PSD95 (red) was used as a marker of postsynaptic terminals. (a) The representative images show the effect of CRH on the synapse formation. Scale bar represents 10 μm. (b) The histogram summarizes the effects of CRH and CRH receptor antagonists on the numbers of synapsin I and PSD95 clusters (ROI = 10 μm). Data are presented as mean ± SEM (n = 5 cultures). (c and d) Hippocampal neurons were treated with increasing concentrations of CRH for 48 hours. (c) Synapsin I and (d) PSD95 expression protein levels of hippocampal neurons were determined by Western blot. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM (n = 6 cultures). (e and f) Hippocampal neurons were treated with CRH in the presence or absence of antalarmin and astressin2B for 48 hours. The neurons were then harvested for determination of (e) synapsin I and (f) PSD95 by Western blot. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM (n = 6 cultures). Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; ##P < 0.01 vs CRH. Anta, antalarmin; As2B, astressin2B. Figure 2. View largeDownload slide CRH stimulated synapse formation in cultured hippocampal neurons. (a and b) Hippocampal neurons were treated with CRH in the presence or absence of antalarmin and astressin2B for 48 hours. The cells were then used for immunofluorescence analysis of synapse terminals. Presynaptic terminals were labeled by synapsin I (green). PSD95 (red) was used as a marker of postsynaptic terminals. (a) The representative images show the effect of CRH on the synapse formation. Scale bar represents 10 μm. (b) The histogram summarizes the effects of CRH and CRH receptor antagonists on the numbers of synapsin I and PSD95 clusters (ROI = 10 μm). Data are presented as mean ± SEM (n = 5 cultures). (c and d) Hippocampal neurons were treated with increasing concentrations of CRH for 48 hours. (c) Synapsin I and (d) PSD95 expression protein levels of hippocampal neurons were determined by Western blot. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM (n = 6 cultures). (e and f) Hippocampal neurons were treated with CRH in the presence or absence of antalarmin and astressin2B for 48 hours. The neurons were then harvested for determination of (e) synapsin I and (f) PSD95 by Western blot. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM (n = 6 cultures). Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; ##P < 0.01 vs CRH. Anta, antalarmin; As2B, astressin2B. Western blot analysis showed that CRH (10−12 to 10−8 M) treatment dose-dependently increased the protein level of synapsin I and PSD95, suggesting that CRH increased synapse numbers mainly through stimulation of synaptic protein expression. Moreover, it was found that the maximal effect occurred at the concentration of 10−8 M, with a 157.8% ± 4.1% increase of synapsin I and a 182.1% ± 4.3% increase of PSD95, respectively [Fig. 2(c), 2(d), synapsin I: F(5, 30) = 43.29, P < 0.001, CRH 10−8 M vs vehicle, LSD-t(30) = 11.91, P < 0.001; PSD95: F(5, 30) = 39.56, CRH 10−8 M vs vehicle, LSD-t(30) = 11.43, P < 0.001]. Consistently, only the CRHR1 antagonist reversed the stimulatory effects of CRH on synapsin I and PSD95 expression [Fig. 2(e) and 2(f), synapsin I: F(5, 30) = 125.31, P < 0.001; CRH plus antalarmin: 106.0% ± 4.5% vs CRH: 179.7% ± 18.5%, LSD-t(30) = −13.47, P < 0.001; PSD95: F(5, 30) = 47.26, P < 0.001; CRH plus antalarmin: 106.8% ± 20.3% vs CRH: 179.5% ± 23.4%, LSD-t(30) = −8.66, P < 0.001]. Because the major difference between isolated neurons and hippocampal slices is the cell component (i.e., glia), we set up glia–neuron cocultures to mimic the cell component of brain slices. In this model, we found that CRH (10−8 M) treatment for 48 hours led to an inhibition of synapse formation. The numbers of synapsin I–labeled terminals, PSD95-labeled terminals, and colocalized complexes around the neuronal processes were significantly suppressed by CRH treatment compared with vehicle control [Fig. 3(a) and 3(b), synapsin I: F(5, 18) = 13.47, P < 0.001; CRH: 2.9 ± 0.3 vs vehicle: 5.2 ± 0.4, LSD-t(18) = −4.95, P < 0.001; PSD95: F(5, 18) = 47.63, P < 0.001; CRH: 2.9 ± 0.6 vs vehicle: 5.3 ± 0.6, LSD-t(18) = −8.64, P < 0.001; costained synapses: F(5, 18) = 32.27, P < 0.001; CRH: 2.7 ± 0.7 vs vehicle: 5.0 ± 0.7, LSD-t(18) = −7.13, P < 0.001]. The inhibitory effect of CRH could be abolished by the specific CRHR1 antagonist antalarmin coapplication [CRH plus antalarmin vs CRH, synapsin I: P = 0.002, LSD-t(18) = 3.62; PSD95: LSD-t(18) = 9.08, P < 0.001; costained synapses: LSD-t(18) = 6.13, P < 0.001]. Western blot analysis showed that CRH (10−12 to 10−8 M) inhibited synapsin I and PSD95 expression in a dose-dependent manner. The maximal effects were achieved at the concentration of 10−8 M. The synapsin I level was reduced by 57.2% ± 2.3%, and PSD95 expression was decreased by 53.2% ± 3.6% [Fig. 3(c) and 3(d), synapsin I: F(5, 24) = 79.96, P < 0.001; CRH 10−8 M vs vehicle, LSD-t(24) = −16.47, P < 0.001; PSD95: F(5, 24) = 57.74, P < 0.001; CRH 10−8 M vs vehicle, LSD-t(24) = −14.52, P < 0.001]. These effects of CRH were blocked by antalarmin but not astressin2B [Fig. 3(e) and 3(f), synapsin I: F(5, 24) = 65.24; P < 0.001; CRH plus antalarmin: 92.3% ± 4.3% vs CRH: 54.9% ± 1.4%, LSD-t(24) = 10.24, P < 0.001; PSD95: F(5, 24) = 65.58, P < 0.001; CRH plus antalarmin: 95.1% ± 4.1% vs CRH: 56.2% ± 2.9%, LSD-t(24) = 2.45, P = 0.022]. Cell viability assessment showed that the neuron viability remained unaffected after CRH incubation (Supplemental Fig. 2). Figure 3. View largeDownload slide Glia contribute to CRH inhibition of synapse formation in the hippocampus. (a and b) The neuron–glia cocultures were treated with various reagents as indicated for 48 hours. Cells were fixed for immunofluorescence analysis of synapse terminals as described in Materials and Methods. Presynaptic terminals were labeled by synapsin I (green). Postsynaptic terminals were marked by PSD95 (red). (a) Representative fluorescence images show the effects of CRH and CRH receptor antagonists on the synapse complex formation in cocultures. Scale bar represents 10 μm. (b) The histogram cumulated the effects of CRH and CRH receptor antagonists on the numbers of synapsin I and PSD95 clusters (ROI = 10 μm). Data are presented as mean ± SEM (n = 4 cultures). (c–f) The neuron–glia cocultures were treated with increasing concentrations of (c and d) CRH or CRH (10−8 M) in the absence or presence of (e and f) antalarmin and astressin2B for 48 hours. The neurons were harvested for determination of (c and e) synapsin I and (d and f) PSD95 levels by Western blot. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM (n = 5 cultures). Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; #P < 0.05; ##P < 0.01 vs CRH. Anta, antalarmin; As2B, astressin2B. Figure 3. View largeDownload slide Glia contribute to CRH inhibition of synapse formation in the hippocampus. (a and b) The neuron–glia cocultures were treated with various reagents as indicated for 48 hours. Cells were fixed for immunofluorescence analysis of synapse terminals as described in Materials and Methods. Presynaptic terminals were labeled by synapsin I (green). Postsynaptic terminals were marked by PSD95 (red). (a) Representative fluorescence images show the effects of CRH and CRH receptor antagonists on the synapse complex formation in cocultures. Scale bar represents 10 μm. (b) The histogram cumulated the effects of CRH and CRH receptor antagonists on the numbers of synapsin I and PSD95 clusters (ROI = 10 μm). Data are presented as mean ± SEM (n = 4 cultures). (c–f) The neuron–glia cocultures were treated with increasing concentrations of (c and d) CRH or CRH (10−8 M) in the absence or presence of (e and f) antalarmin and astressin2B for 48 hours. The neurons were harvested for determination of (c and e) synapsin I and (d and f) PSD95 levels by Western blot. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM (n = 5 cultures). Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; #P < 0.05; ##P < 0.01 vs CRH. Anta, antalarmin; As2B, astressin2B. GCM of CRH-treated glia exhibits an inhibitory effect on synapse formation We then explored whether soluble and diffusible factors secreted by glial cells upon CRH treatment would be responsible for synapse formation. It was shown that the numbers of synapsin I–labeled clusters and PSD95-positive clusters were significantly decreased when the hippocampal neurons were incubated with the GCM of CRH (10−8 M)-treated glia [Fig. 4(a) and 4(b), synapsin I: F(5, 18) = 52.95, P < 0.001, GCM of the glia treated with CRH vs vehicle GCM: LSD-t(18) = −11.39, P < 0.001; PSD95: F(5, 18) = 47.82, P < 0.001, GCM of the glia treated with CRH vs vehicle GCM: LSD-t(18) = −8.64, P < 0.001; costained synapses: F(5, 18) = 24.29, P < 0.001, GCM of the glia treated with CRH vs vehicle GCM: LSD-t(18) = −7.02, P < 0.001]. The GCM collected from the glial cells with CRH in combination with antalarmin treatment did not influence the numbers of synapsin I–positive and PSD95-positive terminals in hippocampal neurons. In contrast, the GCM obtained from the glial cells with CRH plus astressin2B treatment showed inhibition of the formation of synapsin I–positive and PSD95-positive terminals [GCM of glia treated with CRH plus astressin2B vs vehicle GCM, synapsin I: LSD-t(18) = −10.28, P < 0.001; PSD95: LSD-t(18) = −10.59, P < 0.001; costained synapses: LSD-t(18) = −6.65, P < 0.001]. Figure 4. View largeDownload slide CRH inhibits synapse terminal formation through the soluble factors secreted by glial cells. (a and b) Hippocampal neurons were incubated with GCM collected from glial cells with CRH treatment or CRH plus CRHR antagonist treatment for 48 hours. The neurons were fixed for immunofluorescence analysis of synapsin I and PSD95 clusters. Presynaptic terminals were labeled by synapsin I (green). PSD95 (red) was used as a marker of postsynaptic terminals. (a) Representative fluorescence images showed presynaptic and postsynaptic terminals in hippocampal neurons incubated with the GCM collected from CRH-treated glial cells. Scale bar represents 10 μm. (b) The histogram summarizes the effects of different GCMs on the numbers of synapsin I and PSD95 clusters (ROI = 10 μm). Data are presented as mean ± SEM (n = 4 cultures). (c–f) Hippocampal neurons were incubated with GCM collected from glial cells that were treated with increasing concentrations of CRH or CRH plus CRHR antagonists. The protein levels of (c and e) synapsin I and (d and f) PSD95 in hippocampal neurons incubated with different GCMs were determined by Western blot. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM (n = 4 cultures). Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control GCM; ##P < 0.01 vs GCM collected from glia treated with CRH (10−8 M). Anta, antalarmin; As2B, astressin2B. Figure 4. View largeDownload slide CRH inhibits synapse terminal formation through the soluble factors secreted by glial cells. (a and b) Hippocampal neurons were incubated with GCM collected from glial cells with CRH treatment or CRH plus CRHR antagonist treatment for 48 hours. The neurons were fixed for immunofluorescence analysis of synapsin I and PSD95 clusters. Presynaptic terminals were labeled by synapsin I (green). PSD95 (red) was used as a marker of postsynaptic terminals. (a) Representative fluorescence images showed presynaptic and postsynaptic terminals in hippocampal neurons incubated with the GCM collected from CRH-treated glial cells. Scale bar represents 10 μm. (b) The histogram summarizes the effects of different GCMs on the numbers of synapsin I and PSD95 clusters (ROI = 10 μm). Data are presented as mean ± SEM (n = 4 cultures). (c–f) Hippocampal neurons were incubated with GCM collected from glial cells that were treated with increasing concentrations of CRH or CRH plus CRHR antagonists. The protein levels of (c and e) synapsin I and (d and f) PSD95 in hippocampal neurons incubated with different GCMs were determined by Western blot. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM (n = 4 cultures). Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control GCM; ##P < 0.01 vs GCM collected from glia treated with CRH (10−8 M). Anta, antalarmin; As2B, astressin2B. Western blot analysis showed that synapsin I and PSD95 protein levels in hippocampal neurons were decreased by the treatment of GCM collected from glia treated with various concentrations of CRH [Fig. 4(c) and 4(d), synapsin I: F(5, 18) = 40.25, P < 0.001; GCM of the glia treated with CRH vs vehicle GCM: LSD-t(18) = −10.64, P < 0.001; PSD95: F(5, 18) = 36.56, P < 0.001; GCM of the glia treated with CRH vs vehicle GCM: LSD-t(18) = −10.63, P < 0.001]. The GCM of the glia treated with CRH in combination with antalarmin did not exhibit any inhibitory effect on synapsin I and PSD95 expression. The GCM of the glia treated with CRH plus astressin2B showed suppression on both synapsin I and PSD95 expression level [Fig. 4(e) and 4(f), synapsin I, F(5, 18) = 30.33, P < 0.001; GCM of glia treated with CRH plus astressin2B vs vehicle GCM, LSD-t(18) = −7.88, P < 0.001; PSD95: F(5, 18) = 65.56, P < 0.001; GCM of glia treated with CRH plus astressin2B vs vehicle GCM, LSD-t(18) = −11.81, P < 0.001]. Sequence-specific siRNA targeting CRHR1 was used to further confirm the role of CRHR1 in glial cells in CRH-induced suppression of synapse formation. CRHR1 siRNA transfection led to 63.3% ± 4.8% decrease in CRHR1 expression in glial cells (Supplemental Fig. 3). The GCM was collected from glial cells of CRHR1 knockdown with CRH (10−8 M) treatment of 24 hours. Treatment of hippocampal neurons with the above GCM did not affect the number of synapsin I–labeled and PSD95-labeled terminals. However, the GCM, which was harvested from CRH-treated glial cells transfected with control siRNA, significantly decreased the numbers of synapsin I–labeled and PSD95-labeled terminals [Fig. 5(a)–5(d), synapsin I: interaction F(1, 12) = 14.89, P = 0.002; PSD95: interaction F(1, 12) = 16.16, P = 0.002; costained synapses: interaction F(1, 12) = 14.66, P = 0.002]. Figure 5. View largeDownload slide CRHR1 in glia mediates CRH modulation of synapse formation in the hippocampus. (a–d) The roles of CRHR1 in glia in CRH suppression of synapsin I–labeled and PSD95-labeled terminals in hippocampal neurons. Hippocampal neurons were incubated with the GCM of CRH-treated glia with CRHR1 siRNA or control siRNA transfection for 48 hours. Immunofluorescence analysis was then performed as described in Materials and Methods. Presynaptic terminals were labeled by synapsin I (green). Postsynaptic terminals were labeled by PSD95 (red). (a) Representative fluorescence images show synapse complexes in hippocampal neurons. Scale bar represents 10 μm. (b–d) Profile plot cartograms summarize the effects of CRH-treated GCM on the estimation of marginal means of (b) synapsin I–positive clusters, (c) PSD95-positive clusters, and (d) costained synapses (ROI = 10 μm, n = 4 cultures). (e and f) The effects of the GCM of CRH-treated glial cells with CRHR1 siRNA transfection on synapsin I and PSD95 expression in hippocampal neurons. Hippocampal neurons were incubated with the GCM collected from CRH-treated glial cells with CRHR1 siRNA or control siRNA transfection for 48 hours. The levels of (e) synapsin I and (f) PSD95 in hippocampal neurons were determined by Western blot. Representative protein bands are presented at the top of the profile plot cartograms, showing the estimation of marginal means of (e) synapsin I and (f) PSD95 protein levels (n = 4 cultures). Statistical analysis was performed with factorial design ANOVA. Figure 5. View largeDownload slide CRHR1 in glia mediates CRH modulation of synapse formation in the hippocampus. (a–d) The roles of CRHR1 in glia in CRH suppression of synapsin I–labeled and PSD95-labeled terminals in hippocampal neurons. Hippocampal neurons were incubated with the GCM of CRH-treated glia with CRHR1 siRNA or control siRNA transfection for 48 hours. Immunofluorescence analysis was then performed as described in Materials and Methods. Presynaptic terminals were labeled by synapsin I (green). Postsynaptic terminals were labeled by PSD95 (red). (a) Representative fluorescence images show synapse complexes in hippocampal neurons. Scale bar represents 10 μm. (b–d) Profile plot cartograms summarize the effects of CRH-treated GCM on the estimation of marginal means of (b) synapsin I–positive clusters, (c) PSD95-positive clusters, and (d) costained synapses (ROI = 10 μm, n = 4 cultures). (e and f) The effects of the GCM of CRH-treated glial cells with CRHR1 siRNA transfection on synapsin I and PSD95 expression in hippocampal neurons. Hippocampal neurons were incubated with the GCM collected from CRH-treated glial cells with CRHR1 siRNA or control siRNA transfection for 48 hours. The levels of (e) synapsin I and (f) PSD95 in hippocampal neurons were determined by Western blot. Representative protein bands are presented at the top of the profile plot cartograms, showing the estimation of marginal means of (e) synapsin I and (f) PSD95 protein levels (n = 4 cultures). Statistical analysis was performed with factorial design ANOVA. Western blot analysis also showed that GCM collected from CRH-treated glial cells with CRHR1 knockdown did not affect levels of synapsin I and PSD95 in hippocampal neurons. In contrast, the GCM collected from CRH-treated glial cells with control siRNA transfection inhibited the protein levels of synapsin I and PSD95 [Fig. 5(e) and 5(f), synapsin I: interaction F(1, 12) = 7.74, P = 0.017; PSD95: interaction F(1, 12) = 56.70, P < 0.001]. CXCL5 produced by glial cells contributes to CRH suppression of synapse formation We then identified the secretary factors regulated by CRHR1 in glial cells. Cytokine antibody array showed that CRH (10−8 M) treatment affected the output of various cytokines, including IL-6, agrin, CCL-2 (MCP-1), and CXCL5. Among them, CRH-induced inhibition of CXCL5 was blocked by CRHR1 antagonist antalarmin [Fig. 6(a) and 6(b), Supplemental Table 1]. Figure 6. View largeDownload slide CRH suppresses CXCL5 (LIX) production in glial cells and hippocampal slices via CRHR1. (a and b) Cytokine output in glial cells by CRH/CRHR1 signaling. Glial cells were treated with CRH (10−8 M) in the presence and absence of CRHR1 antagonist antalarmin (10−7 M) for 48 hours. The media were collected for cytokine antibody array as described in Materials and Methods. (a) The raw data of the cytokine antibody array. (b) The histogram of summarizing the cytokine antibody array. (c and d) The role of CRH/CRHR1 signaling in CXCL5 secretion in glia. (c) Glial cells were treated with increasing concentrations of CRH or combined CRH and CRHR antagonists for 48 hours. The media were collected for determination of CXCL5 by ELISA (n = 4 cultures). (d) Estimation of marginal means of CXCL5 level in CRH-treated GCM collected from glial cells with CRHR1 knockdown. Glial cells were treated with CRH (10−8 M) in glial cells transfected with CRHR1 siRNA or vehicle siRNA. The media were collected for determination of CXCL5 by ELISA (n = 4 cultures). (e) The effect of CRH on CXCL5 content in hippocampal slices. Hippocampal slices were treated with CRH or in a combination of CRHR antagonists. The slices were collected for determination of CXCL5 content by ELISA (n = 4 cultures). Data are presented as mean ± SEM. The data for Fig. 6(d) were analyzed by factorial design ANOVA. The data for other figures were analyzed by one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; ##P < 0.01 vs CRH. Anta, antalarmin; As2B, astressin2B. Figure 6. View largeDownload slide CRH suppresses CXCL5 (LIX) production in glial cells and hippocampal slices via CRHR1. (a and b) Cytokine output in glial cells by CRH/CRHR1 signaling. Glial cells were treated with CRH (10−8 M) in the presence and absence of CRHR1 antagonist antalarmin (10−7 M) for 48 hours. The media were collected for cytokine antibody array as described in Materials and Methods. (a) The raw data of the cytokine antibody array. (b) The histogram of summarizing the cytokine antibody array. (c and d) The role of CRH/CRHR1 signaling in CXCL5 secretion in glia. (c) Glial cells were treated with increasing concentrations of CRH or combined CRH and CRHR antagonists for 48 hours. The media were collected for determination of CXCL5 by ELISA (n = 4 cultures). (d) Estimation of marginal means of CXCL5 level in CRH-treated GCM collected from glial cells with CRHR1 knockdown. Glial cells were treated with CRH (10−8 M) in glial cells transfected with CRHR1 siRNA or vehicle siRNA. The media were collected for determination of CXCL5 by ELISA (n = 4 cultures). (e) The effect of CRH on CXCL5 content in hippocampal slices. Hippocampal slices were treated with CRH or in a combination of CRHR antagonists. The slices were collected for determination of CXCL5 content by ELISA (n = 4 cultures). Data are presented as mean ± SEM. The data for Fig. 6(d) were analyzed by factorial design ANOVA. The data for other figures were analyzed by one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; ##P < 0.01 vs CRH. Anta, antalarmin; As2B, astressin2B. To confirm the effect of CRH on CXCL5 output, we measured CXCL5 levels in the culture media of the neurons treated with increasing concentrations of CRH. It was showed that CRH dose-dependently decreased CXCL5 output. This effect was prevented by CRHR1 antagonist antalarmin [Fig. 6(c); F(7, 24) = 20.20, P < 0.001; CRH plus antalarmin vs CRH: LSD-t(24) = 7.11, P < 0.001] and CRHR1 siRNA [Fig. 6(d), interaction F(1, 12) = 20.14, P < 0.001]. We also found that CRH treatment caused a decrease in CXCL5 content in the cultured hippocampal slice, which was blocked by antalarmin [Fig. 6(e), F(7, 24) = 25.30, P < 0.001; CRH vs vehicle, LSD-t(24) = −8.38, P < 0.001; CRH plus antalarmin vs CRH, LSD-t(24) = 8.69, P < 0.001]. We subsequently examined whether CXCL5 mediates CRH modulation of synapse formation in cultured hippocampal neurons. At first, we found that administering CXCR2 antagonist SB265610 to the neurons incubated with vehicle GCM reduced synapsin I–labeled and PSD95-labeled terminals and clusters [Fig. 7(a) and 7(b), synapsin I, F(5, 18) = 16.88, P < 0.001; SB265610 plus vehicle GCM vs vehicle GCM: LSD-t(18) = −5.96, P < 0.001; PSD95: F(5, 18) = 17.18, P < 0.001, SB265610 plus vehicle GCM vs vehicle GCM: LSD-t(18) = −5.76, P < 0.001; costained synapses: F(5, 18) = 19.07, P < 0.001, SB265610 plus vehicle GCM vs vehicle GCM: LSD-t(18) = −6.35, P < 0.001]. Administration of anti-CXCL5 (anti-LIX) antibody exhibited similar effects. It was shown that anti-CXCL5 antibody treatment decreased synapsin I–positive and PSD95-positive terminals and clusters in the neurons incubated with vehicle GCM [synapsin I, LSD-t(18) = −4.01, P = 0.001; PSD95: LSD-t(18) = −4.65, P < 0.001; costained synapses: LSD-t(18) = −4.56, P < 0.001]. However, in the neurons incubated with CRH-treated GCM, additional application of purified rLIX reversed the inhibitory effects of CRH-treated GCM on synapsin I–positive and PSD95-positive terminals [synapsin I: LSD-t(18) = 5.61, P < 0.001; PSD95: LSD-t(18) = 5.72, P < 0.001; costained synapses: LSD-t(18) = 5.75, P < 0.001]. These effects could be blocked by administration of CXCR2 antagonist SB265610 [synapsin I: LSD-t(18) = −5.90, P < 0.001; PSD95: LSD-t(18) = −5.93, P < 0.001; costained synapses: LSD-t(18) = −6.28, P < 0.001]. Figure 7. View largeDownload slide Reduced CXCL5 production in glial cells contributes to CRH inhibition of synapse formation in hippocampal neurons. (a and b) Cultured hippocampal neurons were incubated with GCM containing CXCR2 antagonist SB265610 and CXCL5 (LIX) antibody or with CRH-treated GCM containing SB265610 and LIX for 48 hours. Immunofluorescence analysis of synapsin I–labeled and PSD95-labeled terminals was performed as described in Materials and Methods. Presynaptic terminals were labeled by synapsin I (green). Postsynaptic terminals were labeled by PSD95 (red). (a) Representative fluorescence images of synapse formation in hippocampal neurons incubated with various GCMs as indicated. Scale bar represents 10 μm. (b) Cumulative data of the effects of various GCMs on the numbers of synapsin I, PSD95, and synapsin I and PSD95 costaining terminals. Data are presented as mean ± SEM (n = 4). (c–f) The effects of various GCMs on (c and e) synapsin I and (d and f) PSD95 expression in hippocampal neurons. Hippocampal neurons were incubated with the GCM containing CXCR2 antagonist SB265610 and LIX antibody or with CRH-treated GCM containing SB265610 and LIX for 48 hours. The levels of synapsin I and PSD95 were determined by Western blot. Representative protein bands are presented at the top of the histogram (n = 5 cultures). Data are presented as mean ± SEM. Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control GCM; ##P < 0.01 vs CRH-treated GCM; &&P < 0.01 vs CRH-treated GCM plus LIX. Figure 7. View largeDownload slide Reduced CXCL5 production in glial cells contributes to CRH inhibition of synapse formation in hippocampal neurons. (a and b) Cultured hippocampal neurons were incubated with GCM containing CXCR2 antagonist SB265610 and CXCL5 (LIX) antibody or with CRH-treated GCM containing SB265610 and LIX for 48 hours. Immunofluorescence analysis of synapsin I–labeled and PSD95-labeled terminals was performed as described in Materials and Methods. Presynaptic terminals were labeled by synapsin I (green). Postsynaptic terminals were labeled by PSD95 (red). (a) Representative fluorescence images of synapse formation in hippocampal neurons incubated with various GCMs as indicated. Scale bar represents 10 μm. (b) Cumulative data of the effects of various GCMs on the numbers of synapsin I, PSD95, and synapsin I and PSD95 costaining terminals. Data are presented as mean ± SEM (n = 4). (c–f) The effects of various GCMs on (c and e) synapsin I and (d and f) PSD95 expression in hippocampal neurons. Hippocampal neurons were incubated with the GCM containing CXCR2 antagonist SB265610 and LIX antibody or with CRH-treated GCM containing SB265610 and LIX for 48 hours. The levels of synapsin I and PSD95 were determined by Western blot. Representative protein bands are presented at the top of the histogram (n = 5 cultures). Data are presented as mean ± SEM. Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control GCM; ##P < 0.01 vs CRH-treated GCM; &&P < 0.01 vs CRH-treated GCM plus LIX. Western blot analysis confirmed that CXCR2 antagonist significantly inhibited synapsin I and PSD95 levels in neurons incubated with vehicle GCM [Fig. 7(c) and 7(d), synapsin I:, F(4, 20) = 15.77, P < 0.001; SB265610 plus vehicle GCM vs vehicle GCM: LSD-t(20) = −2.44, P = 0.022; PSD95: F(4, 20) = 14.10, P < 0.001; SB265610 plus vehicle GCM vs vehicle GCM: LSD-t(20) = −2.28, P = 0.034]. GCM containing anti-LIX antibody exhibited similar inhibitory effects on synapsin I and PSD95 expression [Fig. 7(e) and 7(f), synapsin I: F(5, 24) = 79.97, P < 0.001; anti-LIX plus vehicle GCM vs vehicle GCM: LSD-t(24) = −11.63, P < 0.001; PSD95: F(5, 24) = 134.56, P < 0.001; anti-LIX plus vehicle GCM vs vehicle GCM: LSD-t(24) = −14.61, P < 0.001]. Administration of rLIX blocked the inhibitory effect of CRH-treated GCM on synapsin I and PSD95 expression [Fig. 7(c) and 7(d), synapsin I: LSD-t(20) = 5.31, P < 0.001; PSD95: LSD-t(20) = 3.48, P = 0.002]. This effect was reversed by CXCR2 antagonist SB265610 [Fig. 7(c) and 7(d), synapsin I: LSD-t(20) = −5.79, P < 0.001; PSD95: LSD-t(20) = −5.76, P < 0.001]. CRH inhibits CXCL5 production in astrocytes through multiple signaling pathways Astrocytes, microglia, and oligodendrocytes are main types of glial cells in rodents. We showed that astrocytes are the predominant cell type in glia isolated from the hippocampus (Supplemental Fig. 1). We therefore investigated the signaling pathways responsible for CRH suppressing CXCL5 secretion in cultured astrocytes. CRH (10−12 to 10−8 M) decreased cAMP production and activated Gi protein in a dose-dependent manner [Fig. 8(a), F(5, 18) = 88.54, P < 0.001; Fig. 8(b), F(5, 18) = 30.55, P < 0.001]. These effects were prevented by either CRHR1 antagonist [CRH plus antalarmin vs CRH: Gi protein level, LSD-t(18) = −7.31, P < 0.001; cAMP production, LSD-t(18) = 9.29, P < 0.001] or CRHR1 siRNA [Fig. 8(c), interaction F(1, 12) = 11.21, P = 0.006]. PTX (1.5 μg/mL), a blocker of Gi protein, blocked the inhibitory effect of CRH on CXCL5 release [Fig. 8(d), F(3, 12) = 25.69, P < 0.001; CRH vs vehicle: LSD-t(12) = −8.13, P < 0.001; CRH plus PTX vs CRH: LSD-t(12) = 6.87, P < 0.001]. Application of AC inhibitor SQ22536 mimicked CRH suppression of CXCL5 secretion, whereas AC activator forskolin or cAMP analog 8-Br-cAMP reversed CRH inhibition of CXCL5 secretion [Fig. 8(e), F(6, 21) = 27.45, P < 0.001; CRH vs vehicle, LSD-t(21) = −10.36, P < 0.001; SQ22536 vs control: LSD-t(21) = −9.40, P < 0.001; CRH plus forskolin vs CRH, LSD-t(21) = 6.72, P < 0.001; CRH plus 8-Br-cAMP vs CRH, LSD-t(21) = 7.61, P < 0.001]. Figure 8. View largeDownload slide CRH inhibits CXCL5 output in astrocytes through multiple signaling pathways. (a) The effects of CRH on Gi protein activation in astrocytes. Astrocytes were treated with increasing concentrations of CRH in the presence or absence of CRHR1 antagonists for 5 minutes. The cells were collected for determination of GTP-bound Gi protein as described in Materials and Methods. Representative protein bands of GTP-bound Gi protein are on the top of histograms (n = 4 cultures). (b and c) The effects of CRH/CRHR1 signaling on cAMP production in astrocytes. (b) The concentration of cAMP in astrocytes treated with increasing concentrations of CRH in the presence or absence of CRHR1 antagonist for 10 minutes (n = 4 cultures). (c) Profile plot cartogram shows the estimation of marginal means of cAMP content in astrocytes treated with CRH in the presence of CRHR1 siRNA (n = 4 cultures). The content of cAMP in cells was determined by RIA. (d and e) The effects of PTX, SQ22536, forskolin, and 8-Br-cAMP on CRH inhibition of CXCL5 secretion in astrocytes. Astrocytes were treated with CRH (10−8 M) in the presence or absence of (d) PTX (n = 4 cultures), (e) SQ22536 (n = 4 cultures), (e) forskolin, and (e) 8-Br-cAMP for 48 hours. The media were collected for determination of CXCL5 content. (f and g) The effects of CRH on PLC-β3 activation and IP3 production in astrocytes. Astrocytes were treated with CRH (10−8 M) in the presence or absence of antalarmin for 10 minutes. The levels of (f) phosphorylated PLC-β3 (n = 4 cultures) and (g) IP3 (n = 4 cultures) were determined by Western blot and ELISA, respectively. Representative protein bands of pPLC-β3 are presented at the top of the histogram. (h) The effect of CRHR1 siRNA on CRH modulation of IP3 production. Astrocytes were transfected with CRHR1 siRNA or control siRNA and then treated with CRH (10−8 M) for 10 minutes. The concentration of IP3 was determined by ELISA. Profile plot cartogram shows the estimation of marginal means of IP3 production (n = 4 cultures). (i) The effects of PLC inhibitor U73122, PKC nonselective inhibitor chelerythrine, and PKCα/β inhibitor Gö6976 on CRH inhibition of CXCL5 secretion in astrocytes. Astrocytes were treated with CRH (10−8 M) in the presence or absence of U73122, chelerythrine, and Gö6976 for 48 hours. The media were collected for determination of CXCL5 content (n = 4 cultures). Data are presented as mean ± SEM. The data for Fig. 8(c) and 8(h) were statistically analyzed by factorial design ANOVA, whereas the data for other figures were analyzed by one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; #P < 0.05; ##P < 0.01 vs CRH. Anta, antalarmin. Figure 8. View largeDownload slide CRH inhibits CXCL5 output in astrocytes through multiple signaling pathways. (a) The effects of CRH on Gi protein activation in astrocytes. Astrocytes were treated with increasing concentrations of CRH in the presence or absence of CRHR1 antagonists for 5 minutes. The cells were collected for determination of GTP-bound Gi protein as described in Materials and Methods. Representative protein bands of GTP-bound Gi protein are on the top of histograms (n = 4 cultures). (b and c) The effects of CRH/CRHR1 signaling on cAMP production in astrocytes. (b) The concentration of cAMP in astrocytes treated with increasing concentrations of CRH in the presence or absence of CRHR1 antagonist for 10 minutes (n = 4 cultures). (c) Profile plot cartogram shows the estimation of marginal means of cAMP content in astrocytes treated with CRH in the presence of CRHR1 siRNA (n = 4 cultures). The content of cAMP in cells was determined by RIA. (d and e) The effects of PTX, SQ22536, forskolin, and 8-Br-cAMP on CRH inhibition of CXCL5 secretion in astrocytes. Astrocytes were treated with CRH (10−8 M) in the presence or absence of (d) PTX (n = 4 cultures), (e) SQ22536 (n = 4 cultures), (e) forskolin, and (e) 8-Br-cAMP for 48 hours. The media were collected for determination of CXCL5 content. (f and g) The effects of CRH on PLC-β3 activation and IP3 production in astrocytes. Astrocytes were treated with CRH (10−8 M) in the presence or absence of antalarmin for 10 minutes. The levels of (f) phosphorylated PLC-β3 (n = 4 cultures) and (g) IP3 (n = 4 cultures) were determined by Western blot and ELISA, respectively. Representative protein bands of pPLC-β3 are presented at the top of the histogram. (h) The effect of CRHR1 siRNA on CRH modulation of IP3 production. Astrocytes were transfected with CRHR1 siRNA or control siRNA and then treated with CRH (10−8 M) for 10 minutes. The concentration of IP3 was determined by ELISA. Profile plot cartogram shows the estimation of marginal means of IP3 production (n = 4 cultures). (i) The effects of PLC inhibitor U73122, PKC nonselective inhibitor chelerythrine, and PKCα/β inhibitor Gö6976 on CRH inhibition of CXCL5 secretion in astrocytes. Astrocytes were treated with CRH (10−8 M) in the presence or absence of U73122, chelerythrine, and Gö6976 for 48 hours. The media were collected for determination of CXCL5 content (n = 4 cultures). Data are presented as mean ± SEM. The data for Fig. 8(c) and 8(h) were statistically analyzed by factorial design ANOVA, whereas the data for other figures were analyzed by one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; #P < 0.05; ##P < 0.01 vs CRH. Anta, antalarmin. CRH treatment also dose-dependently increased the level of phosphorylated PLC-β3, the active PLC-β3, which was prevented by CRHR1 antagonist [Fig. 8(f), F(5, 18) = 179.82, P < 0.001; CRH vs vehicle, LSD-t(18) = 25.24, P < 0.001; CRH plus antalarmin vs CRH, LSD-t(18) = −20.26, P < 0.001]. CRH administration also increased IP3 production. CRHR1 antagonist and CRHR1 siRNA could block the aforementioned effect [Fig. 8(g), F(5, 18) = 20.90, P < 0.001; CRH vs vehicle, LSD-t(18) = 8.97, P < 0.001; CRH plus antalarmin vs CRH, LSD-t(18) = −7.77, P < 0.001; Fig. 8(h), interaction F(1, 12) = 40.95, P < 0.001]. Administration of PLC inhibitor U73122, protein kinase C (PKC) nonselective inhibitor chelerythrine, or PKCα/β inhibitor Gö6976 completely abolished the CRH-induced suppression of CXCL5 secretion [Fig. 8(i), F(7, 24) = 36.20, P < 0.001; CRH vs vehicle, LSD-t(24) = −11.37, P < 0.001; CRH plus U73122 vs CRH, LSD-t(24) = 8.88, P < 0.001; CRH plus chelerythrine vs CRH, LSD-t(24) = 6.14, P < 0.001; CRH plus Gö6976 vs CRH, LSD-t(24) = 10.64, P < 0.001]. Discussion This study demonstrated that CRH acts on glia to inhibit CXCL5 secretion, subsequently leading to suppression of synapse formation in the hippocampus. These data may indicate that glia–neuron intercommunication is one of the mechanisms responsible for neuronal circuits remodeling during stress. The hippocampus, one of the most plastic regions in brain, is highly sensitive to stress (33–35). Many studies have demonstrated that stress induces structural changes of synapses in the hippocampus, with concomitant changes in cognitive function and affective behavior (4–6, 36, 37). As mentioned, CRH is one of the key factors involved in stress-induced synaptic abnormalities, particularly synapse-bearing dendritic spines in the hippocampus (7–10, 38). CRH is synthesized mainly by pyramidal cell layer interneurons and released during stress (39, 40). Blockage of CRHR1 prevented stress-induced spine loss and dendritic atrophy induced by stress in the hippocampus (39, 41). Liao et al. (42) recently reported that CRHR1 antagonist attenuated downregulation of synapse-related proteins induced by stress in the hippocampus. In addition, an in vitro study by Chen et al. (43) showed that CRH suppresses dendritic spine density in hippocampal slices. Consistently, the current study showed that CRH suppressed synaptic formation in the hippocampus via CRHR1. The molecular mechanisms by which CRH modulates synapse formation in the hippocampus remain largely unknown. Emerging evidence has indicated that glial cells, such as astrocytes, play critical roles in neuronal differentiation and circuitry formation in the last decade (44, 45). In the hippocampus, about 64% of synapses are contacted by perisynaptic astrocytes at the synaptic cleft (46, 47). The intricate arborization and ramifications of astrocytes allow them to tightly enwrap the synaptic terminal to modulate synaptic processes and plasticity (46, 48). Previous studies have implied that glia might play a role in CRH modulation of synapse formation because it has been shown that CRH displays differential effects on synaptic formation in the model of cultured neurons and brain slices (11, 12). Interestingly, the current study also demonstrated the opposite effects of CRH on synapse formation in isolated hippocampal neurons and organotypic hippocampal slices. Moreover, we have shown that similar effects of CRH on synaptic-related protein expression and synaptic formation occurred in hippocampal slices and glia–neuron cocultures. Knockdown of CRHR1 in glial cells prevented CRH suppression of synapse formation in the model of glia–neuron cocultures. These data provided the strong evidence that glial cells played a key role in CRH suppression of synaptic formation in the hippocampus. The glial cells regulate synapse remodeling and plasticity through contact molecules and soluble secreted factors. A number of soluble factors secreted by glia have been identified to be involved in synapse formation, such as thrombospondins, transforming growth factor -β, and BDNF (16, 19, 20). In the current study, we demonstrated that a chemokine, CXCL5 (LIX), played a critical role in maintenance of synaptic formation in the hippocampus, as evidenced by blockage of CXCL5 leading to decreased synapse formation in glia–neuron cocultures. A number of studies have demonstrated that glia can synthesize and secrete various cytokines, such as IL-1, IL-6, and IL-10, and chemokines including CCL2, CCL3, CCL4, CXCL12, and CXCL14 (49–57). These cytokines and chemokines are involved in neurogenesis, neuronal survival, synaptic transmission, and synaptic plasticity. For instance, IL-10 promotes synapse formation in cultured cortical neurons (52), whereas chemokines CCL2, CCL3, and CXCL14 have been shown to modulate synaptic transmission (53, 54, 57). In the current study, we confirmed that CRHR1 was expressed in hippocampal glia in rats. Furthermore, we demonstrated that CRHR1 activation led to inhibition of AC/cAMP signaling and activation of PLC/PKC signaling in astrocytes, and these signaling pathways mediated CRH inhibition of CXCL5 secretion. CRH receptors, particularly CRHR1, can couple to multiple G proteins with subsequent activation of multiple signaling pathways in various tissues (58). Our previous studies have shown that CRHR1 can couple to Gs and Gq proteins in hippocampal neurons of rats (59, 60); in contrast, it couples to Gi, Gs, and Gq in human myometrium (61). Interestingly, Takuma et al. (62) reported that CRH increases Ca2+ influx via PLC/PKC pathway in cultured astrocytes of rats. Here we demonstrated that CRHR1 coupled with Gi and Gq proteins and subsequently led to inhibition of AC/cAMP signaling and activation of PLC/PKC signaling in astrocytes, respectively. Although no direct evidence indicates that CRHR1 leads to activation of multiple signaling pathways in glia in vivo, Blank et al. (63) showed that CRH can induce diverse signaling pathways via multiple G proteins in cultured hippocampal slices. However, our present data are not consistent with the study by Chen et al. (64), who showed that CRHR1 activation leads to accumulation of cAMP in astrocytes of cortex in rats. It is very hard to uncover the reason for the discrepant data between two groups. However, the glial cells collected from different brain regions (cortex in Chen et al.’s vs hippocampus in ours) might account for it. Nevertheless, we have demonstrated CRHR1 activation of a Gi/AC/cAMP signaling pathway by the following evidence: CRHR1 activation leads to activation of Gi protein and a decrease in cAMP production, and blockage of Gi activation prevents CRHR1 inhibition of CXCL5 secretion. As mentioned, CRH can be secreted from interneurons in the hippocampus during stress (40, 41), which contributes to spine loss during stress (9). We therefore propose that, during stress, increased CRH secretion inhibits CXCL5 secretion by glia via CRHR1, leading to suppression of synapse formation in hippocampal neurons (Fig. 9). Our study may indicate that glia–neuron communication is critical for remodeling of synapse circuits in the brain during stress. Figure 9. View largeDownload slide Scheme illustrating CRH regulation of synapse formation in the hippocampus. CXCL5 secreted by glial cells plays a critical role in synaptic terminal formation of pyramidal neurons in the hippocampus. CRH can be released from interneurons by various stimuli (such as stress) in the hippocampus. CRH then acts on CRHR1 in glia to suppress CXCL5 secretion from glia, resulting in a decrease in synapses in pyramidal neurons. AC, adenylyl cyclase; PKA, protein kinase A. Figure 9. View largeDownload slide Scheme illustrating CRH regulation of synapse formation in the hippocampus. CXCL5 secreted by glial cells plays a critical role in synaptic terminal formation of pyramidal neurons in the hippocampus. CRH can be released from interneurons by various stimuli (such as stress) in the hippocampus. CRH then acts on CRHR1 in glia to suppress CXCL5 secretion from glia, resulting in a decrease in synapses in pyramidal neurons. AC, adenylyl cyclase; PKA, protein kinase A. Appendix. Antibody Table Peptide/Protein Target  Antigen Sequence (if Known)  Name of Antibody  Manufacturer, Catalog No.  Species Raised in; Monoclonal or Polyclonal  Dilution Used  RRID  Synapsin I    Anti-synapsin I antibody  Millipore, AB1543P  Rabbit; polyclonal  IF 1:200, WB 1:1000  AB_90757  Postsynaptic density 95 kDa    Anti-PSD95 antibody  Abcam, ab2723  Mouse; monoclonal  IF 1:200, WB 1:1000  AB_303248  Corticotropin-releasing factor receptor 1    CRF-R1 polyclonal antibody  Bioworld, BS2590  Rabbit; polyclonal  WB 1:1000  AB_1663553  Glial fibrillary acid protein    GFAP antibody  Santa Cruz Biotechnology Inc, sc-6170  Goat; polyclonal  IF 1:200  AB_641021  PLC-β3    Mouse anti-PLC β3 monoclonal antibody  Santa Cruz Biotechnology Inc, sc133231  Mouse; monoclonal  WB 1:1000  AB_2299534  Phosphorylated PLC-β3 (Ser 537)    Phospho-PLC β3 (Ser537) antibody  Cell Signaling Technology, 2481  Rabbit; polyclonal  WB 1:1000  AB_2163265  β-Actin    Mouse anti-β-actin monoclonal antibody  Sigma-Aldrich, A5441  Mouse; monoclonal  WB 1:10000  AB_476744  Rabbit IgG    Donkey anti-rabbit IgG secondary antibody, Alexa Fluor 594 conjugate  Thermo Fisher, R37119  Donkey; polyclonal  IF 1:400  AB_2556547  Goat IgG    Donkey anti-goat IgG (H+L) secondary antibody, Alexa Fluor 488 conjugate  Thermo Fisher, A-11055  Donkey; polyclonal  IF 1:400  AB_2534102  Mouse IgG    Donkey anti-mouse IgG secondary antibody, Alexa Fluor 594 conjugate  Thermo Fisher, R37115  Donkey; polyclonal  IF 1:400  AB_2556543  Rabbit IgG    Donkey anti-rabbit IgG (H+L) secondary antibody, Alexa Fluor 488 conjugate  Thermo Fisher, A-21206  Donkey; polyclonal  IF 1:400  AB_2535792  Rabbit IgG    Donkey anti-rabbit IgG-HRP  Santa Cruz Biotechnology Inc, sc-2313  Donkey; polyclonal  WB 1:1000  AB_641181  Mouse IgG    Donkey anti-mouse IgG-HRP  Santa Cruz Biotechnology Inc, sc-2314  Donkey; polyclonal  WB 1:1000  AB_641170  Peptide/Protein Target  Antigen Sequence (if Known)  Name of Antibody  Manufacturer, Catalog No.  Species Raised in; Monoclonal or Polyclonal  Dilution Used  RRID  Synapsin I    Anti-synapsin I antibody  Millipore, AB1543P  Rabbit; polyclonal  IF 1:200, WB 1:1000  AB_90757  Postsynaptic density 95 kDa    Anti-PSD95 antibody  Abcam, ab2723  Mouse; monoclonal  IF 1:200, WB 1:1000  AB_303248  Corticotropin-releasing factor receptor 1    CRF-R1 polyclonal antibody  Bioworld, BS2590  Rabbit; polyclonal  WB 1:1000  AB_1663553  Glial fibrillary acid protein    GFAP antibody  Santa Cruz Biotechnology Inc, sc-6170  Goat; polyclonal  IF 1:200  AB_641021  PLC-β3    Mouse anti-PLC β3 monoclonal antibody  Santa Cruz Biotechnology Inc, sc133231  Mouse; monoclonal  WB 1:1000  AB_2299534  Phosphorylated PLC-β3 (Ser 537)    Phospho-PLC β3 (Ser537) antibody  Cell Signaling Technology, 2481  Rabbit; polyclonal  WB 1:1000  AB_2163265  β-Actin    Mouse anti-β-actin monoclonal antibody  Sigma-Aldrich, A5441  Mouse; monoclonal  WB 1:10000  AB_476744  Rabbit IgG    Donkey anti-rabbit IgG secondary antibody, Alexa Fluor 594 conjugate  Thermo Fisher, R37119  Donkey; polyclonal  IF 1:400  AB_2556547  Goat IgG    Donkey anti-goat IgG (H+L) secondary antibody, Alexa Fluor 488 conjugate  Thermo Fisher, A-11055  Donkey; polyclonal  IF 1:400  AB_2534102  Mouse IgG    Donkey anti-mouse IgG secondary antibody, Alexa Fluor 594 conjugate  Thermo Fisher, R37115  Donkey; polyclonal  IF 1:400  AB_2556543  Rabbit IgG    Donkey anti-rabbit IgG (H+L) secondary antibody, Alexa Fluor 488 conjugate  Thermo Fisher, A-21206  Donkey; polyclonal  IF 1:400  AB_2535792  Rabbit IgG    Donkey anti-rabbit IgG-HRP  Santa Cruz Biotechnology Inc, sc-2313  Donkey; polyclonal  WB 1:1000  AB_641181  Mouse IgG    Donkey anti-mouse IgG-HRP  Santa Cruz Biotechnology Inc, sc-2314  Donkey; polyclonal  WB 1:1000  AB_641170  Abbreviations: HRP, horseradish peroxidase; IF, immunofluorescence; RRID, Research Resource Identifier; WB, Western blot. View Large Abbreviations: ANOVA analysis of variance BDNF brain-derived neurotrophic factor CRH corticotropin-releasing hormone CRHR1 corticotropin-releasing hormone receptor type 1 CXCL5 C-X-C motif chemokine 5 DMEM Dulbecco’s modified Eagle medium ELISA enzyme-linked immunosorbent assay FBS fetal bovine serum GCM glia-conditioned media LIX C-X-C motif chemokine 5 LSD least significant difference P postnatal day PBS phosphate-buffered saline PKC protein kinase C PLC phospholipase C PSD95 postsynaptic density protein 95 PTX pertussis toxin rLIX recombinant C-X-C motif chemokine 5 ROI region of interest SDS sodium dodecyl sulfate SEM standard error of the mean siRNA small interfering RNA. Acknowledgments The authors thank Dr. Chen Wu from the Department of Statistics, Second Military Medical University, Shanghai, China, for help with statistical analysis. Financial Support: This work was supported by the Major State Basic Research Program of China (2013CB967404) and the National Natural Science Foundation of China (31100840). Disclosure Summary: The authors have nothing to disclose. References 1. Owens MJ, Nemeroff CB. Physiology and pharmacology of corticotropin-releasing factor. Pharmacol Rev . 1991; 43( 4): 425– 473. Google Scholar PubMed  2. Cummings S, Elde R, Ells J, Lindall A. Corticotropin-releasing factor immunoreactivity is widely distributed within the central nervous system of the rat: an immunohistochemical study. J Neurosci . 1983; 3( 7): 1355– 1368. Google Scholar PubMed  3. Herman JP, Prewitt CM, Cullinan WE. Neuronal circuit regulation of the hypothalamo-pituitary-adrenocortical stress axis. Crit Rev Neurobiol . 1996; 10( 3–4): 371– 394. Google Scholar CrossRef Search ADS PubMed  4. Chen Y, Rex CS, Rice CJ, Dubé CM, Gall CM, Lynch G, Baram TZ. Correlated memory defects and hippocampal dendritic spine loss after acute stress involve corticotropin-releasing hormone signaling. Proc Natl Acad Sci USA . 2010; 107( 29): 13123– 13128. Google Scholar CrossRef Search ADS PubMed  5. Wang XD, Su YA, Wagner KV, Avrabos C, Scharf SH, Hartmann J, Wolf M, Liebl C, Kühne C, Wurst W, Holsboer F, Eder M, Deussing JM, Müller MB, Schmidt MV. Nectin-3 links CRHR1 signaling to stress-induced memory deficits and spine loss. Nat Neurosci . 2013; 16( 6): 706– 713. Google Scholar CrossRef Search ADS PubMed  6. Wang XD, Rammes G, Kraev I, Wolf M, Liebl C, Scharf SH, Rice CJ, Wurst W, Holsboer F, Deussing JM, Baram TZ, Stewart MG, Müller MB, Schmidt MV. Forebrain CRF1 modulates early-life stress-programmed cognitive deficits. J Neurosci . 2011; 31( 38): 13625– 13634. Google Scholar CrossRef Search ADS PubMed  7. Gounko NV, Swinny JD, Kalicharan D, Jafari S, Corteen N, Seifi M, Bakels R, van der Want JJ. Corticotropin-releasing factor and urocortin regulate spine and synapse formation: structural basis for stress-induced neuronal remodeling and pathology. Mol Psychiatry . 2013; 18( 1): 86– 92. Google Scholar CrossRef Search ADS PubMed  8. Chen Y, Fenoglio KA, Dubé CM, Grigoriadis DE, Baram TZ. Cellular and molecular mechanisms of hippocampal activation by acute stress are age-dependent. Mol Psychiatry . 2006; 11( 11): 992– 1002. Google Scholar CrossRef Search ADS PubMed  9. Chen Y, Dubé CM, Rice CJ, Baram TZ. Rapid loss of dendritic spines after stress involves derangement of spine dynamics by corticotropin-releasing hormone. J Neurosci . 2008; 28( 11): 2903– 2911. Google Scholar CrossRef Search ADS PubMed  10. Wang XD, Chen Y, Wolf M, Wagner KV, Liebl C, Scharf SH, Harbich D, Mayer B, Wurst W, Holsboer F, Deussing JM, Baram TZ, Müller MB, Schmidt MV. Forebrain CRHR1 deficiency attenuates chronic stress-induced cognitive deficits and dendritic remodeling. Neurobiol Dis . 2011; 42( 3): 300– 310. Google Scholar CrossRef Search ADS PubMed  11. Sheng H, Xu Y, Chen Y, Zhang Y, Xu X, He C, Ni X. CRH-R1 and CRH-R2 differentially modulate dendritic outgrowth of hippocampal neurons. Endocrine . 2012; 41( 3): 458– 464. Google Scholar CrossRef Search ADS PubMed  12. Chen Y, Bender RA, Brunson KL, Pomper JK, Grigoriadis DE, Wurst W, Baram TZ. Modulation of dendritic differentiation by corticotropin-releasing factor in the developing hippocampus. Proc Natl Acad Sci USA . 2004; 101( 44): 15782– 15787. Google Scholar CrossRef Search ADS PubMed  13. Fields RD, Stevens-Graham B. New insights into neuron–glia communication. Science . 2002; 298( 5593): 556– 562. Google Scholar CrossRef Search ADS PubMed  14. Allen NJ, Barres BA. Neuroscience: Glia—more than just brain glue. Nature . 2009; 457( 7230): 675– 677. Google Scholar CrossRef Search ADS PubMed  15. Stogsdill JA, Eroglu C. The interplay between neurons and glia in synapse development and plasticity. Curr Opin Neurobiol . 2017; 42: 1– 8. Google Scholar CrossRef Search ADS PubMed  16. Christopherson KS, Ullian EM, Stokes CC, Mullowney CE, Hell JW, Agah A, Lawler J, Mosher DF, Bornstein P, Barres BA. Thrombospondins are astrocyte-secreted proteins that promote CNS synaptogenesis. Cell . 2005; 120( 3): 421– 433. Google Scholar CrossRef Search ADS PubMed  17. Mauch DH, Nägler K, Schumacher S, Göritz C, Müller EC, Otto A, Pfrieger FW. CNS synaptogenesis promoted by glia-derived cholesterol. Science . 2001; 294( 5545): 1354– 1357. Google Scholar CrossRef Search ADS PubMed  18. Allen NJ, Bennett ML, Foo LC, Wang GX, Chakraborty C, Smith SJ, Barres BA. Astrocyte glypicans 4 and 6 promote formation of excitatory synapses via GluA1 AMPA receptors. Nature . 2012; 486( 7403): 410– 414. Google Scholar PubMed  19. Diniz LP, Almeida JC, Tortelli V, Vargas Lopes C, Setti-Perdigão P, Stipursky J, Kahn SA, Romão LF, de Miranda J, Alves-Leon SV, de Souza JM, Castro NG, Panizzutti R, Gomes FC. Astrocyte-induced synaptogenesis is mediated by transforming growth factor β signaling through modulation of D-serine levels in cerebral cortex neurons. J Biol Chem . 2012; 287( 49): 41432– 41445. Google Scholar CrossRef Search ADS PubMed  20. Fan Y, Chen J, Ye J, Yan H, Cai Y. Proteinase-activated receptor 2 modulates corticotropin releasing hormone-induced brain-derived neurotrophic factor release from microglial cells. Cell Biol Int . 2014; 38( 1): 92– 96. Google Scholar CrossRef Search ADS PubMed  21. Wang W, Ji P, Dow KE. Corticotropin-releasing hormone induces proliferation and TNF-alpha release in cultured rat microglia via MAP kinase signalling pathways. J Neurochem . 2003; 84( 1): 189– 195. Google Scholar CrossRef Search ADS PubMed  22. Burton SD, Johnson JW, Zeringue HC, Meriney SD. Distinct roles of neuroligin-1 and SynCAM1 in synapse formation and function in primary hippocampal neuronal cultures. Neuroscience . 2012; 215: 1– 16. Google Scholar CrossRef Search ADS PubMed  23. Devader C, Khayachi A, Veyssière J, Moha Ou Maati H, Roulot M, Moreno S, Borsotto M, Martin S, Heurteaux C, Mazella J. In vitro and in vivo regulation of synaptogenesis by the novel antidepressant spadin. Br J Pharmacol . 2015; 172( 10): 2604– 2617. Google Scholar CrossRef Search ADS PubMed  24. Cuesto G, Jordán-Álvarez S, Enriquez-Barreto L, Ferrús A, Morales M, Acebes Á. GSK3β inhibition promotes synaptogenesis in Drosophila and mammalian neurons. PLoS One . 2015; 10( 3): e0118475. Google Scholar CrossRef Search ADS PubMed  25. Zeng M, Shang Y, Araki Y, Guo T, Huganir RL, Zhang M. Phase transition in postsynaptic densities underlies formation of synaptic complexes and synaptic plasticity. Cell . 2016; 166( 5): 1163– 1175.e12. Google Scholar CrossRef Search ADS PubMed  26. Jeyifous O, Lin EI, Chen X, Antinone SE, Mastro R, Drisdel R, Reese TS, Green WN. Palmitoylation regulates glutamate receptor distributions in postsynaptic densities through control of PSD95 conformation and orientation. Proc Natl Acad Sci USA . 2016; 113( 52): E8482– E8491. Google Scholar CrossRef Search ADS PubMed  27. Peña F. Organotypic cultures as tool to test long-term effects of chemicals on the nervous system. Curr Med Chem . 2010; 17( 10): 987– 1001. Google Scholar CrossRef Search ADS PubMed  28. Bender R, Heimrich B, Meyer M, Frotscher M. Hippocampal mossy fiber sprouting is not impaired in brain-derived neurotrophic factor–deficient mice. Exp Brain Res . 1998; 120( 3): 399– 402. Google Scholar CrossRef Search ADS PubMed  29. Elliott-Hunt CR, Kazlauskaite J, Wilde GJ, Grammatopoulos DK, Hillhouse EW. Potential signalling pathways underlying corticotrophin-releasing hormone–mediated neuroprotection from excitotoxicity in rat hippocampus. J Neurochem . 2002; 80( 3): 416– 425. Google Scholar CrossRef Search ADS PubMed  30. Nelson TE, Gruol DL. The chemokine CXCL10 modulates excitatory activity and intracellular calcium signaling in cultured hippocampal neurons. J Neuroimmunol . 2004; 156( 1-2): 74– 87. Google Scholar CrossRef Search ADS PubMed  31. Duan S, Anderson CM, Keung EC, Chen Y, Chen Y, Swanson RA. P2X7 receptor–mediated release of excitatory amino acids from astrocytes. J Neurosci . 2003; 23( 4): 1320– 1328. Google Scholar PubMed  32. Zheng Y, Zhang YM, Ni X. Urocortin 2 but not urocortin 3 promotes the synaptic formation in hippocampal neurons via induction of NGF production by astrocytes. Endocrinology . 2016; 157( 3): 1200– 1210. Google Scholar CrossRef Search ADS PubMed  33. Nemeroff CB, Bremner JD, Foa EB, Mayberg HS, North CS, Stein MB. Posttraumatic stress disorder: a state-of-the-science review. J Psychiatr Res . 2006; 40( 1): 1– 21. Google Scholar CrossRef Search ADS PubMed  34. Naninck EF, Hoeijmakers L, Kakava-Georgiadou N, Meesters A, Lazic SE, Lucassen PJ, Korosi A. Chronic early life stress alters developmental and adult neurogenesis and impairs cognitive function in mice. Hippocampus . 2015; 25( 3): 309– 328. Google Scholar CrossRef Search ADS PubMed  35. Barzegar M, Sajjadi FS, Talaei SA, Hamidi G, Salami M. Prenatal exposure to noise stress: anxiety, impaired spatial memory, and deteriorated hippocampal plasticity in postnatal life. Hippocampus . 2015; 25( 2): 187– 196. Google Scholar CrossRef Search ADS PubMed  36. Xia B, Chen C, Zhang H, Xue W, Tang J, Tao W, Wu R, Ren L, Wang W, Chen G. Chronic stress prior to pregnancy potentiated long-lasting postpartum depressive-like behavior, regulated by Akt-mTOR signaling in the hippocampus. Sci Rep . 2016; 6( 1): 35042. Google Scholar CrossRef Search ADS PubMed  37. Karisetty BC, Maitra S, Wahul AB, Musalamadugu A, Khandelwal N, Guntupalli S, Garikapati R, Jhansyrani T, Kumar A, Chakravarty S. Differential effect of chronic stress on mouse hippocampal memory and affective behavior: Role of major ovarian hormones. Behav Brain Res . 2017; 318: 36– 44. Google Scholar CrossRef Search ADS PubMed  38. Andres AL, Regev L, Phi L, Seese RR, Chen Y, Gall CM, Baram TZ. NMDA receptor activation and calpain contribute to disruption of dendritic spines by the stress neuropeptide CRH. J Neurosci . 2013; 33( 43): 16945– 16960. Google Scholar CrossRef Search ADS PubMed  39. Chen Y, Bender RA, Frotscher M, Baram TZ. Novel and transient populations of corticotropin-releasing hormone–expressing neurons in developing hippocampus suggest unique functional roles: a quantitative spatiotemporal analysis. J Neurosci . 2001; 21( 18): 7171– 7181. Google Scholar PubMed  40. Chen Y, Brunson KL, Adelmann G, Bender RA, Frotscher M, Baram TZ. Hippocampal corticotropin releasing hormone: pre- and postsynaptic location and release by stress. Neuroscience . 2004; 126( 3): 533– 540. Google Scholar CrossRef Search ADS PubMed  41. Ivy AS, Rex CS, Chen Y, Dubé C, Maras PM, Grigoriadis DE, Gall CM, Lynch G, Baram TZ. Hippocampal dysfunction and cognitive impairments provoked by chronic early-life stress involve excessive activation of CRH receptors. J Neurosci . 2010; 30( 39): 13005– 13015. Google Scholar CrossRef Search ADS PubMed  42. Liao XM, Yang XD, Jia J, Li JT, Xie XM, Su YA, Schmidt MV, Si TM, Wang XD. Blockade of corticotropin-releasing hormone receptor 1 attenuates early-life stress-induced synaptic abnormalities in the neonatal hippocampus. Hippocampus . 2014; 24( 5): 528– 540. Google Scholar CrossRef Search ADS PubMed  43. Chen Y, Kramár EA, Chen LY, Babayan AH, Andres AL, Gall CM, Lynch G, Baram TZ. Impairment of synaptic plasticity by the stress mediator CRH involves selective destruction of thin dendritic spines via RhoA signaling. Mol Psychiatry . 2013; 18( 4): 485– 496. Google Scholar CrossRef Search ADS PubMed  44. Allen NJ. Role of glia in developmental synapse formation. Curr Opin Neurobiol . 2013; 23( 6): 1027– 1033. Google Scholar CrossRef Search ADS PubMed  45. Todd KJ, Serrano A, Lacaille JC, Robitaille R. Glial cells in synaptic plasticity. J Physiol Paris . 2006; 99( 2–3): 75– 83. Google Scholar CrossRef Search ADS PubMed  46. Ota Y, Zanetti AT, Hallock RM. The role of astrocytes in the regulation of synaptic plasticity and memory formation. Neural Plast . 2013; 2013: 185463. Google Scholar CrossRef Search ADS PubMed  47. Witcher MR, Kirov SA, Harris KM. Plasticity of perisynaptic astroglia during synaptogenesis in the mature rat hippocampus. Glia . 2007; 55( 1): 13– 23. Google Scholar CrossRef Search ADS PubMed  48. Matyash V, Kettenmann H. Heterogeneity in astrocyte morphology and physiology. Brain Res Brain Res Rev . 2010; 63( 1–2): 2– 10. Google Scholar CrossRef Search ADS   49. Schneider H, Pitossi F, Balschun D, Wagner A, del Rey A, Besedovsky HO. A neuromodulatory role of interleukin-1beta in the hippocampus. Proc Natl Acad Sci USA . 1998; 95( 13): 7778– 7783. Google Scholar CrossRef Search ADS PubMed  50. Ross FM, Allan SM, Rothwell NJ, Verkhratsky A. A dual role for interleukin-1 in LTP in mouse hippocampal slices. J Neuroimmunol . 2003; 144( 1–2): 61– 67. Google Scholar CrossRef Search ADS PubMed  51. Balschun D, Wetzel W, Del Rey A, Pitossi F, Schneider H, Zuschratter W, Besedovsky HO. Interleukin-6: a cytokine to forget. FASEB J . 2004; 18( 14): 1788– 1790. Google Scholar PubMed  52. Chen H, Lin W, Zhang Y, Lin L, Chen J, Zeng Y, Zheng M, Zhuang Z, Du H, Chen R, Liu N. IL-10 promotes neurite outgrowth and synapse formation in cultured cortical neurons after the oxygen–glucose deprivation via JAK1/STAT3 pathway. Sci Rep . 2016; 6( 1): 30459. Google Scholar CrossRef Search ADS PubMed  53. Bray JG, Reyes KC, Roberts AJ, Ransohoff RM, Gruol DL. Synaptic plasticity in the hippocampus shows resistance to acute ethanol exposure in transgenic mice with astrocyte-targeted enhanced CCL2 expression. Neuropharmacology . 2013; 67: 115– 125. Google Scholar CrossRef Search ADS PubMed  54. Marciniak E, Faivre E, Dutar P, Alves Pires C, Demeyer D, Caillierez R, Laloux C, Buée L, Blum D, Humez S. The chemokine MIP-1α/CCL3 impairs mouse hippocampal synaptic transmission, plasticity and memory. Sci Rep . 2015; 5( 1): 15862. Google Scholar CrossRef Search ADS PubMed  55. Lee YB, Nagai A, Kim SU. Cytokines, chemokines, and cytokine receptors in human microglia. J Neurosci Res . 2002; 69( 1): 94– 103. Google Scholar CrossRef Search ADS PubMed  56. Guyon A. CXCL12 chemokine and its receptors as major players in the interactions between immune and nervous systems. Front Cell Neurosci . 2014; 8( 65): 65. Google Scholar PubMed  57. Banisadr G, Bhattacharyya BJ, Belmadani A, Izen SC, Ren D, Tran PB, Miller RJ. The chemokine BRAK/CXCL14 regulates synaptic transmission in the adult mouse dentate gyrus stem cell niche. J Neurochem . 2011; 119( 6): 1173– 1182. Google Scholar CrossRef Search ADS PubMed  58. Hillhouse EW, Grammatopoulos DK. The molecular mechanisms underlying the regulation of the biological activity of corticotropin-releasing hormone receptors: implications for physiology and pathophysiology. Endocr Rev . 2006; 27( 3): 260– 286. Google Scholar CrossRef Search ADS PubMed  59. Sheng H, Zhang Y, Sun J, Gao L, Ma B, Lu J, Ni X. Corticotropin-releasing hormone (CRH) depresses n-methyl-D-aspartate receptor–mediated current in cultured rat hippocampal neurons via CRH receptor type 1. Endocrinology . 2008; 149( 3): 1389– 1398. Google Scholar CrossRef Search ADS PubMed  60. Sheng H, Xu Y, Chen Y, Zhang Y, Ni X. Corticotropin-releasing hormone stimulates mitotic kinesin–like protein 1 expression via a PLC/PKC-dependent signaling pathway in hippocampal neurons. Mol Cell Endocrinol . 2012; 362( 1–2): 157– 164. Google Scholar CrossRef Search ADS PubMed  61. You X, Gao L, Liu J, Xu C, Liu C, Li Y, Hui N, Gu H, Ni X. CRH activation of different signaling pathways results in differential calcium signaling in human pregnant myometrium before and during labor. J Clin Endocrinol Metab . 2012; 97( 10): E1851– E1861. Google Scholar CrossRef Search ADS PubMed  62. Takuma K, Matsuda T, Yoshikawa T, Kitanaka J, Gotoh M, Hayata K, Baba A. Corticotropin-releasing factor stimulates Ca2+ influx in cultured rat astrocytes. Biochem Biophys Res Commun . 1994; 199( 3): 1103– 1107. Google Scholar CrossRef Search ADS PubMed  63. Blank T, Nijholt I, Grammatopoulos DK, Randeva HS, Hillhouse EW, Spiess J. Corticotropin-releasing factor receptors couple to multiple G-proteins to activate diverse intracellular signaling pathways in mouse hippocampus: role in neuronal excitability and associative learning. J Neurosci . 2003; 23( 2): 700– 707. Google Scholar PubMed  64. Chen SJ, Yang JF, Kong FP, Ren JL, Hao K, Li M, Yuan Y, Chen XC, Yu RS, Li JF, Leng G, Chen XQ, Du JZ. Overactivation of corticotropin-releasing factor receptor type 1 and aquaporin-4 by hypoxia induces cerebral edema. Proc Natl Acad Sci USA . 2014; 111( 36): 13199– 13204. Google Scholar CrossRef Search ADS PubMed  Copyright © 2018 Endocrine Society http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Endocrinology Oxford University Press

Corticotropin-Releasing Hormone Suppresses Synapse Formation in the Hippocampus of Male Rats via Inhibition of CXCL5 Secretion by Glia

Loading next page...
 
/lp/ou_press/corticotropin-releasing-hormone-suppresses-synapse-formation-in-the-A57tfYOCn6
Publisher
Oxford University Press
Copyright
Copyright © 2018 Endocrine Society
ISSN
0013-7227
eISSN
1945-7170
D.O.I.
10.1210/en.2017-00336
Publisher site
See Article on Publisher Site

Abstract

Abstract Corticotropin-releasing hormone (CRH) is believed to play a critical role in stress-induced synaptic formation and modification. In the current study, we explored the mechanisms underlying CRH modulation of synaptic formation in the hippocampus by using various models in vitro. In cultured hippocampal slices, CRH treatment decreased synapsin I and postsynaptic density protein 95 (PSD95) levels via CRH receptor type 1 (CRHR1). In isolated hippocampal neurons, however, it increased synapsin I–labeled presynaptic terminals and PSD95-labeled postsynaptic terminals via CRHR1. Interestingly, the inhibitory effect of CRH on synapsin I–labeled and PSD95-labeled terminals occurred in the model of neuron–glia cocultures. These effects were prevented by CRHR1 antagonist. Moreover, treatment of the neurons with the media of CRH-treated glia led to a decrease in synaptic terminal formation. The media collected from CRH-treated glial cells with CRHR1 knockdown did not show an inhibitory effect on synaptic terminals in hippocampal neurons. Unbiased cytokine array coupled with confirmatory enzyme-linked immunosorbent assay revealed that CRH suppressed C-X-C motif chemokine 5 (CXCL5) production in glia via CRHR1. Administration of CXCL5 reversed the inhibitory effects of CRH-treated glia culture media on synaptic formation. Our data suggest that CRH suppresses synapse formation through inhibition of CXCL5 secretion from glia in the hippocampus. Our study indicates that glia–neuron intercommunication is one of the mechanisms responsible for neuronal circuit remodeling during stress. Corticotropin-releasing hormone (CRH), a 41-amino-acid neuropeptide, acts as a principal component of the hypothalamic–pituitary–adrenal axis (1). In addition to hypothalamus, where CRH was originally identified, CRH is widely expressed in the central nervous system, including the neocortex, basal ganglia, amygdala, and hippocampus (2, 3). Emerging evidence indicates that CRH is the key mediator in stress-related brain disorders and impaired learning and memory. These effects result primarily from neuroplasticity including neurogenesis, synaptic alternation, and formation (4–6). The effects of CRH on synapse formation and integrity vary between studies, in particular with variety in different brain regions. For instance, CRH increases spine and synapse formation in the cerebellum (7), whereas it mediates stress-induced rapid loss of apical dendritic spines in CA1 and CA3 pyramidal cells of the hippocampus (8, 9). In corticotropin-releasing hormone receptor type 1 (CRHR1) knockout mice, the absence of CRHR1 prevents the detrimental effects of chronic stress on dendritic arborization of CA3 neurons and spatial memory (5, 10). Moreover, the effects of CRH on synaptogenesis in cultured neurons conflict with those in organotypic culture of brain slices. CRH increases dendrite outgrowth in cultured hippocampal neurons (11); in contrast, it suppresses dendrite growth in cultured slices of the hippocampus (12). Thus, the mechanisms by which CRH modulates synapse formation remain largely unknown. An increasing body of evidence has strongly suggested that glial cells play critical roles in the formation, stabilization, and elimination of synapses in both the peripheral and central nervous systems (13–15). Such effects are mediated by soluble signals secreted by glia. A number of secreted factors of glia, such as thrombospondins, cholesterol with apolipoprotein E, glypican 4 and 6, transforming growth factor β, brain-derived neurotrophic factor (BDNF), and tumor necrosis factor α, have recently been identified to regulate various aspects of synapse formation (16–21). Interestingly, some studies have demonstrated that CRH can regulate secretion of a number of factors, such as tumor necrosis factor α and BDNF, in glial cells (19, 20). We hypothesized that glia might be involved in CRH modulating synapse formation and modification. To test this hypothesis, we set up a series of experiments by using various models in vitro. Given that synapsin I links synaptic vesicles to cytoskeletal elements within the presynaptic terminal (22–24) and postsynaptic density protein 95 (PSD95) orients perpendicular to the postsynaptic membrane (25, 26), we used synapsin I and PSD95 as presynaptic and postsynaptic markers, respectively. First, we compared the effects of CRH on presynaptic and postsynaptic related proteins, such as synapsin I and PSD95, in organotypic culture of hippocampal slices and synapsin I- and PSD95-labeled terminals formation in cultured hippocampal neurons. Second, we explored whether glia contributed to CRH modulation of synapse formation in the model of neuron–glia cocultures and neuron cultures with glia-conditioned media (GCM), and subsequently we identified the key mediators produced by glia. Finally, we defined the signaling pathways responsible for CRH regulating cytokine release in glia. Materials and Methods Organotypic hippocampal slice culture All animal protocols were approved by the Ethical Committee of Experimental Animals of Second Military Medical University, China. Protocols were designed to minimize the number of animals used and their suffering. Sprague-Dawley rats were obtained from Sino-British SIPPR/BK Laboratory Animal Ltd, Shanghai, China. Hippocampal slice cultures were prepared from male rats on postnatal (P) days P2 and P3. In each independent culture, six to eight male newborn rats from same litter were used. The cultures were followed by the method described by Bender et al. (27, 28). Briefly, after decapitation rat brains were removed and placed in ice-cold oxygenated low-sodium artificial cerebral spinal fluid (containing 248 mM sucrose, 4 mM KCl, 1.25 mM NaH2PO4, 26.2 mM NaHCO3, 1 mM CaCl2, 5 mM MgCl2, and 10 mM glucose) and then carefully placed on the platform of a tissue chopper and sliced perpendicular to their longitudinal axes with a vibrating microtome (NVSLM1; World Precision Instruments Inc., Sarasota, FL), with 400 μm thickness of each slice. Slices were transferred to Millicell CM membrane inserts (Millipore, Bedford, MA) in six-well culture plates. Each well contained 1.2 mL of prewarmed Dulbecco’s modified Eagle medium (DMEM) (Invitrogen Corp., Carlsbad, CA) containing 20% horse serum (Invitrogen), 10.5 mM glucose, 12.5 mM HEPES, and 55 mM NaHCO3 (pH 7.3 to 7.4). The slices were incubated in a humidified, 5% CO2 atmosphere at 37°C overnight, and followed by treatments with increasing concentrations (10−12 to 10−8 M) of CRH (Sigma-Aldrich, St. Louis, MO) in the absence or presence of CRHR1 antagonist antalarmin (Sigma-Aldrich) and CRHR2 antagonist astressin2B (Sigma-Aldrich) for 48 hours. The dosages of CRH were selected according to the literature (29) and our previous studies (11). The media and slices were collected for subsequent analysis. CRH and astressin2B were dissolved in phosphate-buffered saline (PBS) and stored in 0.1-mM stock at −80°C, and antalarmin was dissolved in dimethyl sulfoxide to achieve a stock of 10 mM, then diluted by culture medium to achieve the final concentration of dimethyl sulfoxide <0.01%. Hippocampal neuron culture Primary hippocampal neurons were cultured as described previously (30). Briefly, hippocampi were dissected from P1 male Sprague-Dawley rats (six to eight male newborn rats from same litter were used in each independent culture) in ice-cold dissection solution containing 136 mM NaCl, 5.4 mM KCl, 0.2 mM Na2HPO4, 2 mM KH2PO4, 16.7 mM glucose, 20.8 mM sucrose, 10 mM HEPES, and 0.0012% phenol red, pH 7.4, and then incubated with 0.125% trypsin (Invitrogen) at 37°C for 15 minutes. Cell suspensions were obtained by mechanical dissociation in DMEM supplemented with 10% heat-inactivated fetal bovine serum (FBS) (Invitrogen) and 10% horse serum. Cells were plated at a density of 1 × 105 cells/cm2 on poly-l-lysine (Sigma-Aldrich) coated coverslips and six-well plates. Cultures were then maintained in 5% CO2 at 37°C in DMEM containing 10% heat-inactivated FBS and 10% horse serum overnight. Cytosine arabinoside (1 μM) was added into cultures 24 hours after plating to block the proliferation of glial cells, and the culture media were changed to serum-free B27/neurobasal medium (Invitrogen). Half of the medium was replaced with fresh medium every 3 days. After 7-day culture, cells were treated with various concentrations of CRH in the absence or presence of antalarmin and astressin2B for 48 hours. In some cases, cells were incubated with GCM or GCM plus recombinant C-X-C motif chemokine 5 (CXCL5) (rLIX, R&D Systems, Minneapolis, MN), antirat CXCL5 antibody (R&D Systems), or CXCR2 antagonist SB265610 (Sigma-Aldrich) for 48 hours. Culture of glial cells Cultures of mixed glia were established according to the method described previously (19, 31). Briefly, hippocampi were prepared from P2 to P3 male Sprague-Dawley rats (four to six male newborn rats from same litter were used in each independent culture) in ice-cold dissection solution and were dissociated by trypsin (0.125%) at 37°C for 15 minutes. After mechanical dissociation in DMEM supplemented with 15% heat-inactivated FBS, single-cell suspensions were obtained and then plated in 75-cm2 culture flasks and maintained at 37°C in a 5% CO2 humidified incubator. Half of the medium was replaced by fresh medium every 3 days for a total of 12 to 14 days. Mixed glial cells in flasks were trypsinized, counted, spun down, and resuspended in culture media and replaced onto poly-l-lysine–coated coverslips, 12-well plates, and transwell inserts. After 4-day culture, the cultures were administrated with CRH, antalarmin, and astressin2B for the indicated time. In some cases, cells were treated with the aforementioned reagents for 48 hours, and then media were harvested and spun down to remove cell debris. These media served as GCM to be used for treatment of isolated neurons. To purify astrocytes from mixed glial cells, cells were allowed to grow for 14 to 21 days. After reaching confluence, cells were shaken at 250 rpm for 18 to 20 hours in DMEM containing 10% FBS on a gyratory shaker at 37°C to remove neurons and other cell types. The cells were then plated onto poly-l-lysine–coated 12-well and 6-well plates and maintained in DMEM containing 10% FBS at 37°C and 5% CO2. Purified astrocytes were treated with various concentrations of CRH in combination with antalarmin, astressin2B, H89, Gö6976, U73122, chelerythrine, forskolin, SQ22536, 8-Br-cAMP, and pertussis toxin (PTX) for the indicated time. All the aforementioned reagents were purchased from Sigma-Aldrich. The purity of astrocytes was assessed by imunocytochemistry with anti-GFAP antibody. It was shown that >95% of such cells GFAP positive (Supplemental Fig. 1). Glia–neuron coculture Glia–neuron coculture was performed in a six-well transwell chamber system with a 4-μm pore size (Corning, Corning, NY), which permitted cell contact–independent communication via diffusible soluble factors only. Freshly isolated mixed glial cells were plated in transwell inserts at 5 × 104 cells per insert. The inserts then were placed on top of each well of the six-well plates, where neurons were cultured and maintained throughout the culture time until subsequent experiments were performed. Immunofluorescence analysis Cultured neurons, glia, and astrocytes were fixed in 4% paraformaldehyde at room temperature. The cultured cells were washed with PBS and incubated with 10% bovine serum albumin for 2 hours and then were incubated with primary antibodies antisynapsin I (AB1543P, 1:200; Millipore, Billerica, MA), anti-PSD95 (ab2723, 1:200; Abcam, Cambridge, UK), anti-CRHR1 (BS2590, 1:100; Bioworld Technology Inc., St. Louis Park, MN), anti-CRHR1 (TA313693; 1:200, OriGene Technologies Inc., MD), and anti-GFAP (sc6170, 1:200; Santa Cruz Biotechnology Inc., Santa Cruz, CA), in PBS containing 2% bovine serum albumin overnight at 4°C. Subsequently, the specimens were incubated with anti-rabbit IgG conjugated to Alexa Fluor 488 (A21206, 1:400; Thermo Fisher Scientific, Waltham, MA) or anti-goat IgG Alexa Fluor 488 (A-11055, 1:400; Thermo Fisher Scientific), anti-rabbit IgG Alexa Fluor 594 (R37119, 1:400; Thermo Fisher Scientific), or anti-mouse IgG Alexa Fluor 594 (R37115, 1:400; Thermo Fisher Scientific). The cell nuclei were visualized by applying the DNA-specific dye Bisbenzimide Hoechst 33342 (23491-52-3, 1:5000; Sigma-Aldrich). Staining images were visualized with a Leica confocal microscope (Lecia Microsystems Inc., Buffalo, NY). Quantification of synapsin I clusters, PSD95 clusters, and synapsin I–PSD95 clusters of cultured neurons were performed in Image J software. These clusters were calculated in individual neurons with easily distinguishable processes, as described in a previous study (32). The intensity of synapsin I, PSD95, and synapsin I–PSD95 overlapping positive clusters per region of interest (ROI) was taken along the same neuronal branch. At least five visual fields were acquired in each independent culture. The mean values of the positive clusters or costained contacts in each culture were determined. Western blot analysis Cells were scraped off the plates in the presence of lysis buffer, which consists of 60 mM Tris-HCl, 2% sodium dodecyl sulfate (SDS), 10% sucrose, 2 mM phenylmethylsulfonyl fluoride (Merck, Darmstadt, Germany), 1 mM sodium orthovanadate (Sigma-Aldrich), and 10 g/mL aprotinin (Bayer, Leverkusen, Germany). Hippocampal tissues were homogenized in the presence of the aforementioned lysis buffer. The lysates were quickly sonicated, boiled for 5 minutes at 95°C, and centrifuged at 12,000g for 5 minutes at 4°C. Then, the supernatants were collected and the protein concentration in supernatant was assayed with a modified Bradford assay. The samples were diluted in sample buffer [250 mM Tris-HCl (pH 6.8), containing 4% SDS, 10% glycerol, 2% β-mercaptoethanol, and 0.002% bromophenol blue] and boiled for another 5 minutes. Aliquots of the samples were separated by 10% SDS polyacrylamide gel electrophoresis and subsequently transferred to nitrocellulose membranes by electroblotting. The membrane was blocked in 5% skim milk powder in 0.1% Tris-buffered saline/Tween 20 at room temperature for 2 hours, and then was incubated with antibodies against synapsin I, PSD95, CRHR1 (Bioworld Technology Inc.), phospholipase C (PLC)-β3 (sc133231; Santa Cruz), and phosphorylated PLC-β3 (Ser 537) (2481; Cell Signaling Technology, Boston, MA) at a dilution of 1:1000 overnight at 4°C. After three washes with 0.1% Tris-buffered saline/Tween 20, the membranes were incubated with a secondary horseradish peroxidase conjugated antibody (Santa Cruz) for 1 hour at room temperature. Immunoreactive proteins were detected with and enhanced chemiluminescence Western blot detection system (Santa Cruz) and visualized with a Sygene Bio Image system (Synoptics Ltd., UK). To control sampling errors, the ratio of band intensities to β-actin (A5441; Sigma-Aldrich) was obtained to quantify the relative protein expression level. Cell viability assay To examine whether the reagents have detrimental effects on neurons, viability was tested. The cells were incubated with 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide (Sigma-Aldrich) at 0.5 mg/mL for 2 hours after treatment with various reagents. Then, the cells were lysed with dimethyl sulfoxide in Sorensen glycine buffer and the formazan crystals solubilized. Absorbance was read at 570 nm with a spectrophotometric microplate reader (Bio-Rad Laboratories, Inc., Hercules, CA). Cytokine array and enzyme-linked immunosorbent assay Glial cells were treated with CRH (10−8 M) in the presence or absence of antalarmin (10−7 M) for 48 hours, and the media were collected. The media of three independent cultures were pooled together for following cytokine array. The output of total of 34 cytokines in media was evaluated with a rat cytokine antibody array G series (Ray Biotech Inc., Norcross, GA) according to the manufacturer’s instructions. The concentration of CXCL5 (i.e., LIX) in media of cell cultures and brain slices was determined with a RayBio Rat LIX enzyme-linked immunosorbent assay (ELISA) kit (Ray Biotech Inc.) according to the manufacturer’s instructions. The LIX kit shows no cross-reactivity with other cytokines including CINC-2, CINC-3, CNTF, fractalkine, IL-1α, IL-1β, IL-4, IL-6, IL-10, GM-CSF, leptin, CCL-2, β-NGF, TIMP-1, VEGF, TNF-α, INF-γ, and MIP-3α in rats. The sensitivity of the LIX assay was <15 pg/mL. Intra-assay variation is <10%, and interassay variation is <12%. Small interfering RNA transfection The sequence-specific small interfering RNA (siRNA) targeting rat CRHR1 was designed and synthesized by GenePharma Corporation (Shanghai, China). The siRNA targeting CRHR1 (sense 5′-CCGCCUACAAUUACUUCCATT-3′; antisense 5′-UGGAAGUAAUUGUAGGCGGTT-3′) was used for knockdown of the CRHR1 gene in mixed glia in vitro. The following siRNA (sense 5′-UUCUCCGAACGUGUCACGUTT-3′; antisense 5′-ACGUGACACGUUCGGAGAATT-3′), without homology to any known rat messenger RNA sequences in the NCBI RefSeq database, was used as a negative control. The cultured glial cells were transfected with siRNA-CRHR1 and negative control siRNA with siPORT™ NeoFX™ transfection agent (Ambion, Austin, TX). Determination of activated Gi protein The level of active GTP-bound Gi protein was determined with a commercial Gi activation assay kit (NewEast Biosciences, Malvern, PA). Cultured astrocytes were treated with increasing concentrations of CRH (10−12 to 10−8 M) in the absence and presence of CRHR1 antagonist antalarmin (10−7 M) for 5 minutes and then scraped off the plate in the presence of lysis buffer supplied with the kit. The cell lysate was centrifuged for 10 seconds at 12,000g, and then all supernatants were immunoprecipitated with antiactive Gi monoclonal antibody, and protein A/G agarose beads were added to cell lysates. After incubation at 4°C for 1 hour, the beads were pelleted and resuspended three times (10 minutes each) in lysis buffer. Then the samples were centrifuged for 10 seconds at 12,000g, and the supernatant containing active GTP-bound Gi protein was collected. Samples were analyzed by Western blot with anti-Gi monoclonal antibody. To control sampling errors, the total Gi protein was determined by Western blot analysis. cAMP assay Astrocytes were treated with increasing concentrations of various reagents for 10 minutes, then scraped off the plate in the presence of 50 mM sodium acetate (pH 4.75). In some cases, the cells were transfected with CRHR1 siRNA and then treated with CRH (10−8 M) for 10 minutes. Lysates were boiled at 95°C for 10 minutes and then quickly sonicated in an ice bath. The supernatants were collected by centrifugation and used for cAMP assay according to the protocol of a commercial 125I RIA kit (Huaying Biotechnology Research Institute, Peking, China). The sensitivity of cAMP assay was <0.8 pmol/mL. Intra-assay and interassay variations are <8.9%. The kit shows no cross-reactivity with other small molecules in rats. IP3 assay Astrocytes were treated with increasing concentrations of various reagents for 10 minutes, then scraped off the plate in the presence of 50 mM sodium acetate (pH 4.75). In some cases, the cells were transfected with CRHR1 siRNA and then treated with CRH (10−8 M) for 10 minutes. After the cells were frozen three times, the supernatants were collected by centrifugation (3000g for 20 minutes). IP3 content in supernatants was determined with the IP3 ELISA kits (R&D Systems, Minneapolis, MN). The sensitivity of the IP3 assay was <2 pg/mL. There is no cross-reactivity with other small molecules in this kit. The interassay and intra-assay variations are <10%. To avoid the errors that would occur in different assays, IP3 contents were determined by bulk assay of the cells with various treatments. Statistical analysis The n value represents the numbers of independent cultures. All data are presented as mean ± standard error of the mean (SEM). After confirmation of normal distribution, one-way analysis of variance (ANOVA) followed by least significant difference (LSD)-t test or factorial design ANOVA was used to appropriately assess the difference in all variables. SPSS software version 16.0 was used. P < 0.05 was considered statistically significant. Results CRH suppresses presynaptic and postsynaptic related protein levels in hippocampal slices Treatment of cultured hippocampal slices with CRH (10−12 to 10−8 M) for 48 hours significantly suppressed synapsin I and PSD95 expression in a dose-dependent manner. The significant effect occurred at a concentration of 10−11 M in the synapsin I level. The highest effect was achieved at 10−8 M, with 58.3% ± 3.5% reduction of the synapsin I level and 52.9% ± 5.1% reduction of the PSD95 level [Fig. 1(a) and 1(b), synapsin I, F(5, 24) = 34.29, P < 0.001, CRH 10−8 M vs vehicle, LSD-t(24) = −10.12, P < 0.001; PSD95, F(5, 24) = 40.22, P < 0.001, CRH 10−11 M vs vehicle, LSD-t(24) = −11.39, P < 0.001]. The specific CRHR1 antagonist antalarmin reversed CRH inhibition of synapsin I and PSD95 protein expression [Fig. 1(c), synapsin I, F(5, 24) = 49.68, P < 0.001; CRH plus antalarmin: 92.9% ± 5.3% vs CRH: 65.8% ± 3.6%, LSD-t(24) = 2.21, P = 0.037; Fig. 1(d), PSD95, F(5, 24) = 60.17, P < 0.001; CRH plus antalarmin: 95.5% ± 4.3% vs CRH: 55.8% ± 3.4%, LSD-t (24) = 2.45, P = 0.022]. CRHR2 antagonist astressin2B did not affect the effect of CRH. Neither antalarmin nor astressin2B alone has an impact on synapsin I or PSD95 expression. Figure 1. View largeDownload slide CRH suppressed presynaptic and postsynaptic protein expression in hippocampal slices. Hippocampal slices were treated with increasing concentrations of CRH in the presence or absence of antalarmin and astressin2B for 48 hours. The protein levels of synapsin I and PSD95 were determined by Western blot analysis. (a and b) CRH (10−12–10−8 M) dose-dependently suppressed (a) synapsin I (n = 5 cultures) and (b) PSD95 (n = 5 cultures) protein level in cultured hippocampal slices. (c and d) Effects of CRH receptor antagonists on the expression of (c) synapsin I (n = 5 cultures) and (d) PSD95 (n = 5 cultures) protein levels induced by CRH in cultured hippocampal slices. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM. Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; #P < 0.05 vs CRH. Anta, antalarmin; As2B, astressin2B. Figure 1. View largeDownload slide CRH suppressed presynaptic and postsynaptic protein expression in hippocampal slices. Hippocampal slices were treated with increasing concentrations of CRH in the presence or absence of antalarmin and astressin2B for 48 hours. The protein levels of synapsin I and PSD95 were determined by Western blot analysis. (a and b) CRH (10−12–10−8 M) dose-dependently suppressed (a) synapsin I (n = 5 cultures) and (b) PSD95 (n = 5 cultures) protein level in cultured hippocampal slices. (c and d) Effects of CRH receptor antagonists on the expression of (c) synapsin I (n = 5 cultures) and (d) PSD95 (n = 5 cultures) protein levels induced by CRH in cultured hippocampal slices. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM. Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; #P < 0.05 vs CRH. Anta, antalarmin; As2B, astressin2B. CRH promotes synapse formation in isolated hippocampal neuron cultures but suppresses synapse formation in glia–neuron cocultures In a model of hippocampal neuron culture, CRH treatment (10−8 M) for 48 hours significantly increased synapsin I–labeled terminals and PSD95-labeled signals along the neuronal branches compared with vehicle control. Accordingly, the numbers of synapsin I–PSD95 costained synapses were upregulated by CRH treatment compared with vehicle control [Fig. 2(a) and 2(b), synapsin I: F(5, 24) = 80.12, P < 0.001; CRH: 9.1 ± 1.0 vs vehicle: 4.8 ± 0.2, LSD-t(24) = 11.37, P < 0.001; PSD95: F(5, 24) = 72.64, P < 0.001; CRH: 9.8 ± 0.8 vs vehicle: 4.8 ± 0.3, LSD-t(24) = 12.46, P < 0.001; costained synapses: F(5, 24) = 122.14, P < 0.001, CRH: 8.9 ± 0.9 vs vehicle: 4.8 ± 0.2, LSD-t(24) = 14.89, P < 0.001]. These effects were prevented by the CRHR1 specific antagonist antalarmin (10−7 M) but not by the CRHR2 antagonist astressin2B (10−7 M) [synapsin I: CRH plus antalarmin: 4.0 ± 0.5 vs CRH: 9.1 ± 1.0, LSD-t(24) = −14.10, P < 0.001; PSD95: CRH plus antalarmin: 4.3 ± 0.7 vs CRH: 9.8 ± 0.8, LSD-t(24) = −13.69, P < 0.001; costained synapses: CRH plus antalarmin: 4.0 ± 1.1 vs CRH: 8.9 ± 0.9, LSD-t (24) = −17.32, P < 0.001]. Figure 2. View largeDownload slide CRH stimulated synapse formation in cultured hippocampal neurons. (a and b) Hippocampal neurons were treated with CRH in the presence or absence of antalarmin and astressin2B for 48 hours. The cells were then used for immunofluorescence analysis of synapse terminals. Presynaptic terminals were labeled by synapsin I (green). PSD95 (red) was used as a marker of postsynaptic terminals. (a) The representative images show the effect of CRH on the synapse formation. Scale bar represents 10 μm. (b) The histogram summarizes the effects of CRH and CRH receptor antagonists on the numbers of synapsin I and PSD95 clusters (ROI = 10 μm). Data are presented as mean ± SEM (n = 5 cultures). (c and d) Hippocampal neurons were treated with increasing concentrations of CRH for 48 hours. (c) Synapsin I and (d) PSD95 expression protein levels of hippocampal neurons were determined by Western blot. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM (n = 6 cultures). (e and f) Hippocampal neurons were treated with CRH in the presence or absence of antalarmin and astressin2B for 48 hours. The neurons were then harvested for determination of (e) synapsin I and (f) PSD95 by Western blot. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM (n = 6 cultures). Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; ##P < 0.01 vs CRH. Anta, antalarmin; As2B, astressin2B. Figure 2. View largeDownload slide CRH stimulated synapse formation in cultured hippocampal neurons. (a and b) Hippocampal neurons were treated with CRH in the presence or absence of antalarmin and astressin2B for 48 hours. The cells were then used for immunofluorescence analysis of synapse terminals. Presynaptic terminals were labeled by synapsin I (green). PSD95 (red) was used as a marker of postsynaptic terminals. (a) The representative images show the effect of CRH on the synapse formation. Scale bar represents 10 μm. (b) The histogram summarizes the effects of CRH and CRH receptor antagonists on the numbers of synapsin I and PSD95 clusters (ROI = 10 μm). Data are presented as mean ± SEM (n = 5 cultures). (c and d) Hippocampal neurons were treated with increasing concentrations of CRH for 48 hours. (c) Synapsin I and (d) PSD95 expression protein levels of hippocampal neurons were determined by Western blot. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM (n = 6 cultures). (e and f) Hippocampal neurons were treated with CRH in the presence or absence of antalarmin and astressin2B for 48 hours. The neurons were then harvested for determination of (e) synapsin I and (f) PSD95 by Western blot. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM (n = 6 cultures). Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; ##P < 0.01 vs CRH. Anta, antalarmin; As2B, astressin2B. Western blot analysis showed that CRH (10−12 to 10−8 M) treatment dose-dependently increased the protein level of synapsin I and PSD95, suggesting that CRH increased synapse numbers mainly through stimulation of synaptic protein expression. Moreover, it was found that the maximal effect occurred at the concentration of 10−8 M, with a 157.8% ± 4.1% increase of synapsin I and a 182.1% ± 4.3% increase of PSD95, respectively [Fig. 2(c), 2(d), synapsin I: F(5, 30) = 43.29, P < 0.001, CRH 10−8 M vs vehicle, LSD-t(30) = 11.91, P < 0.001; PSD95: F(5, 30) = 39.56, CRH 10−8 M vs vehicle, LSD-t(30) = 11.43, P < 0.001]. Consistently, only the CRHR1 antagonist reversed the stimulatory effects of CRH on synapsin I and PSD95 expression [Fig. 2(e) and 2(f), synapsin I: F(5, 30) = 125.31, P < 0.001; CRH plus antalarmin: 106.0% ± 4.5% vs CRH: 179.7% ± 18.5%, LSD-t(30) = −13.47, P < 0.001; PSD95: F(5, 30) = 47.26, P < 0.001; CRH plus antalarmin: 106.8% ± 20.3% vs CRH: 179.5% ± 23.4%, LSD-t(30) = −8.66, P < 0.001]. Because the major difference between isolated neurons and hippocampal slices is the cell component (i.e., glia), we set up glia–neuron cocultures to mimic the cell component of brain slices. In this model, we found that CRH (10−8 M) treatment for 48 hours led to an inhibition of synapse formation. The numbers of synapsin I–labeled terminals, PSD95-labeled terminals, and colocalized complexes around the neuronal processes were significantly suppressed by CRH treatment compared with vehicle control [Fig. 3(a) and 3(b), synapsin I: F(5, 18) = 13.47, P < 0.001; CRH: 2.9 ± 0.3 vs vehicle: 5.2 ± 0.4, LSD-t(18) = −4.95, P < 0.001; PSD95: F(5, 18) = 47.63, P < 0.001; CRH: 2.9 ± 0.6 vs vehicle: 5.3 ± 0.6, LSD-t(18) = −8.64, P < 0.001; costained synapses: F(5, 18) = 32.27, P < 0.001; CRH: 2.7 ± 0.7 vs vehicle: 5.0 ± 0.7, LSD-t(18) = −7.13, P < 0.001]. The inhibitory effect of CRH could be abolished by the specific CRHR1 antagonist antalarmin coapplication [CRH plus antalarmin vs CRH, synapsin I: P = 0.002, LSD-t(18) = 3.62; PSD95: LSD-t(18) = 9.08, P < 0.001; costained synapses: LSD-t(18) = 6.13, P < 0.001]. Western blot analysis showed that CRH (10−12 to 10−8 M) inhibited synapsin I and PSD95 expression in a dose-dependent manner. The maximal effects were achieved at the concentration of 10−8 M. The synapsin I level was reduced by 57.2% ± 2.3%, and PSD95 expression was decreased by 53.2% ± 3.6% [Fig. 3(c) and 3(d), synapsin I: F(5, 24) = 79.96, P < 0.001; CRH 10−8 M vs vehicle, LSD-t(24) = −16.47, P < 0.001; PSD95: F(5, 24) = 57.74, P < 0.001; CRH 10−8 M vs vehicle, LSD-t(24) = −14.52, P < 0.001]. These effects of CRH were blocked by antalarmin but not astressin2B [Fig. 3(e) and 3(f), synapsin I: F(5, 24) = 65.24; P < 0.001; CRH plus antalarmin: 92.3% ± 4.3% vs CRH: 54.9% ± 1.4%, LSD-t(24) = 10.24, P < 0.001; PSD95: F(5, 24) = 65.58, P < 0.001; CRH plus antalarmin: 95.1% ± 4.1% vs CRH: 56.2% ± 2.9%, LSD-t(24) = 2.45, P = 0.022]. Cell viability assessment showed that the neuron viability remained unaffected after CRH incubation (Supplemental Fig. 2). Figure 3. View largeDownload slide Glia contribute to CRH inhibition of synapse formation in the hippocampus. (a and b) The neuron–glia cocultures were treated with various reagents as indicated for 48 hours. Cells were fixed for immunofluorescence analysis of synapse terminals as described in Materials and Methods. Presynaptic terminals were labeled by synapsin I (green). Postsynaptic terminals were marked by PSD95 (red). (a) Representative fluorescence images show the effects of CRH and CRH receptor antagonists on the synapse complex formation in cocultures. Scale bar represents 10 μm. (b) The histogram cumulated the effects of CRH and CRH receptor antagonists on the numbers of synapsin I and PSD95 clusters (ROI = 10 μm). Data are presented as mean ± SEM (n = 4 cultures). (c–f) The neuron–glia cocultures were treated with increasing concentrations of (c and d) CRH or CRH (10−8 M) in the absence or presence of (e and f) antalarmin and astressin2B for 48 hours. The neurons were harvested for determination of (c and e) synapsin I and (d and f) PSD95 levels by Western blot. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM (n = 5 cultures). Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; #P < 0.05; ##P < 0.01 vs CRH. Anta, antalarmin; As2B, astressin2B. Figure 3. View largeDownload slide Glia contribute to CRH inhibition of synapse formation in the hippocampus. (a and b) The neuron–glia cocultures were treated with various reagents as indicated for 48 hours. Cells were fixed for immunofluorescence analysis of synapse terminals as described in Materials and Methods. Presynaptic terminals were labeled by synapsin I (green). Postsynaptic terminals were marked by PSD95 (red). (a) Representative fluorescence images show the effects of CRH and CRH receptor antagonists on the synapse complex formation in cocultures. Scale bar represents 10 μm. (b) The histogram cumulated the effects of CRH and CRH receptor antagonists on the numbers of synapsin I and PSD95 clusters (ROI = 10 μm). Data are presented as mean ± SEM (n = 4 cultures). (c–f) The neuron–glia cocultures were treated with increasing concentrations of (c and d) CRH or CRH (10−8 M) in the absence or presence of (e and f) antalarmin and astressin2B for 48 hours. The neurons were harvested for determination of (c and e) synapsin I and (d and f) PSD95 levels by Western blot. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM (n = 5 cultures). Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; #P < 0.05; ##P < 0.01 vs CRH. Anta, antalarmin; As2B, astressin2B. GCM of CRH-treated glia exhibits an inhibitory effect on synapse formation We then explored whether soluble and diffusible factors secreted by glial cells upon CRH treatment would be responsible for synapse formation. It was shown that the numbers of synapsin I–labeled clusters and PSD95-positive clusters were significantly decreased when the hippocampal neurons were incubated with the GCM of CRH (10−8 M)-treated glia [Fig. 4(a) and 4(b), synapsin I: F(5, 18) = 52.95, P < 0.001, GCM of the glia treated with CRH vs vehicle GCM: LSD-t(18) = −11.39, P < 0.001; PSD95: F(5, 18) = 47.82, P < 0.001, GCM of the glia treated with CRH vs vehicle GCM: LSD-t(18) = −8.64, P < 0.001; costained synapses: F(5, 18) = 24.29, P < 0.001, GCM of the glia treated with CRH vs vehicle GCM: LSD-t(18) = −7.02, P < 0.001]. The GCM collected from the glial cells with CRH in combination with antalarmin treatment did not influence the numbers of synapsin I–positive and PSD95-positive terminals in hippocampal neurons. In contrast, the GCM obtained from the glial cells with CRH plus astressin2B treatment showed inhibition of the formation of synapsin I–positive and PSD95-positive terminals [GCM of glia treated with CRH plus astressin2B vs vehicle GCM, synapsin I: LSD-t(18) = −10.28, P < 0.001; PSD95: LSD-t(18) = −10.59, P < 0.001; costained synapses: LSD-t(18) = −6.65, P < 0.001]. Figure 4. View largeDownload slide CRH inhibits synapse terminal formation through the soluble factors secreted by glial cells. (a and b) Hippocampal neurons were incubated with GCM collected from glial cells with CRH treatment or CRH plus CRHR antagonist treatment for 48 hours. The neurons were fixed for immunofluorescence analysis of synapsin I and PSD95 clusters. Presynaptic terminals were labeled by synapsin I (green). PSD95 (red) was used as a marker of postsynaptic terminals. (a) Representative fluorescence images showed presynaptic and postsynaptic terminals in hippocampal neurons incubated with the GCM collected from CRH-treated glial cells. Scale bar represents 10 μm. (b) The histogram summarizes the effects of different GCMs on the numbers of synapsin I and PSD95 clusters (ROI = 10 μm). Data are presented as mean ± SEM (n = 4 cultures). (c–f) Hippocampal neurons were incubated with GCM collected from glial cells that were treated with increasing concentrations of CRH or CRH plus CRHR antagonists. The protein levels of (c and e) synapsin I and (d and f) PSD95 in hippocampal neurons incubated with different GCMs were determined by Western blot. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM (n = 4 cultures). Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control GCM; ##P < 0.01 vs GCM collected from glia treated with CRH (10−8 M). Anta, antalarmin; As2B, astressin2B. Figure 4. View largeDownload slide CRH inhibits synapse terminal formation through the soluble factors secreted by glial cells. (a and b) Hippocampal neurons were incubated with GCM collected from glial cells with CRH treatment or CRH plus CRHR antagonist treatment for 48 hours. The neurons were fixed for immunofluorescence analysis of synapsin I and PSD95 clusters. Presynaptic terminals were labeled by synapsin I (green). PSD95 (red) was used as a marker of postsynaptic terminals. (a) Representative fluorescence images showed presynaptic and postsynaptic terminals in hippocampal neurons incubated with the GCM collected from CRH-treated glial cells. Scale bar represents 10 μm. (b) The histogram summarizes the effects of different GCMs on the numbers of synapsin I and PSD95 clusters (ROI = 10 μm). Data are presented as mean ± SEM (n = 4 cultures). (c–f) Hippocampal neurons were incubated with GCM collected from glial cells that were treated with increasing concentrations of CRH or CRH plus CRHR antagonists. The protein levels of (c and e) synapsin I and (d and f) PSD95 in hippocampal neurons incubated with different GCMs were determined by Western blot. Representative protein bands are presented at the top of the histogram. Data are presented as mean ± SEM (n = 4 cultures). Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control GCM; ##P < 0.01 vs GCM collected from glia treated with CRH (10−8 M). Anta, antalarmin; As2B, astressin2B. Western blot analysis showed that synapsin I and PSD95 protein levels in hippocampal neurons were decreased by the treatment of GCM collected from glia treated with various concentrations of CRH [Fig. 4(c) and 4(d), synapsin I: F(5, 18) = 40.25, P < 0.001; GCM of the glia treated with CRH vs vehicle GCM: LSD-t(18) = −10.64, P < 0.001; PSD95: F(5, 18) = 36.56, P < 0.001; GCM of the glia treated with CRH vs vehicle GCM: LSD-t(18) = −10.63, P < 0.001]. The GCM of the glia treated with CRH in combination with antalarmin did not exhibit any inhibitory effect on synapsin I and PSD95 expression. The GCM of the glia treated with CRH plus astressin2B showed suppression on both synapsin I and PSD95 expression level [Fig. 4(e) and 4(f), synapsin I, F(5, 18) = 30.33, P < 0.001; GCM of glia treated with CRH plus astressin2B vs vehicle GCM, LSD-t(18) = −7.88, P < 0.001; PSD95: F(5, 18) = 65.56, P < 0.001; GCM of glia treated with CRH plus astressin2B vs vehicle GCM, LSD-t(18) = −11.81, P < 0.001]. Sequence-specific siRNA targeting CRHR1 was used to further confirm the role of CRHR1 in glial cells in CRH-induced suppression of synapse formation. CRHR1 siRNA transfection led to 63.3% ± 4.8% decrease in CRHR1 expression in glial cells (Supplemental Fig. 3). The GCM was collected from glial cells of CRHR1 knockdown with CRH (10−8 M) treatment of 24 hours. Treatment of hippocampal neurons with the above GCM did not affect the number of synapsin I–labeled and PSD95-labeled terminals. However, the GCM, which was harvested from CRH-treated glial cells transfected with control siRNA, significantly decreased the numbers of synapsin I–labeled and PSD95-labeled terminals [Fig. 5(a)–5(d), synapsin I: interaction F(1, 12) = 14.89, P = 0.002; PSD95: interaction F(1, 12) = 16.16, P = 0.002; costained synapses: interaction F(1, 12) = 14.66, P = 0.002]. Figure 5. View largeDownload slide CRHR1 in glia mediates CRH modulation of synapse formation in the hippocampus. (a–d) The roles of CRHR1 in glia in CRH suppression of synapsin I–labeled and PSD95-labeled terminals in hippocampal neurons. Hippocampal neurons were incubated with the GCM of CRH-treated glia with CRHR1 siRNA or control siRNA transfection for 48 hours. Immunofluorescence analysis was then performed as described in Materials and Methods. Presynaptic terminals were labeled by synapsin I (green). Postsynaptic terminals were labeled by PSD95 (red). (a) Representative fluorescence images show synapse complexes in hippocampal neurons. Scale bar represents 10 μm. (b–d) Profile plot cartograms summarize the effects of CRH-treated GCM on the estimation of marginal means of (b) synapsin I–positive clusters, (c) PSD95-positive clusters, and (d) costained synapses (ROI = 10 μm, n = 4 cultures). (e and f) The effects of the GCM of CRH-treated glial cells with CRHR1 siRNA transfection on synapsin I and PSD95 expression in hippocampal neurons. Hippocampal neurons were incubated with the GCM collected from CRH-treated glial cells with CRHR1 siRNA or control siRNA transfection for 48 hours. The levels of (e) synapsin I and (f) PSD95 in hippocampal neurons were determined by Western blot. Representative protein bands are presented at the top of the profile plot cartograms, showing the estimation of marginal means of (e) synapsin I and (f) PSD95 protein levels (n = 4 cultures). Statistical analysis was performed with factorial design ANOVA. Figure 5. View largeDownload slide CRHR1 in glia mediates CRH modulation of synapse formation in the hippocampus. (a–d) The roles of CRHR1 in glia in CRH suppression of synapsin I–labeled and PSD95-labeled terminals in hippocampal neurons. Hippocampal neurons were incubated with the GCM of CRH-treated glia with CRHR1 siRNA or control siRNA transfection for 48 hours. Immunofluorescence analysis was then performed as described in Materials and Methods. Presynaptic terminals were labeled by synapsin I (green). Postsynaptic terminals were labeled by PSD95 (red). (a) Representative fluorescence images show synapse complexes in hippocampal neurons. Scale bar represents 10 μm. (b–d) Profile plot cartograms summarize the effects of CRH-treated GCM on the estimation of marginal means of (b) synapsin I–positive clusters, (c) PSD95-positive clusters, and (d) costained synapses (ROI = 10 μm, n = 4 cultures). (e and f) The effects of the GCM of CRH-treated glial cells with CRHR1 siRNA transfection on synapsin I and PSD95 expression in hippocampal neurons. Hippocampal neurons were incubated with the GCM collected from CRH-treated glial cells with CRHR1 siRNA or control siRNA transfection for 48 hours. The levels of (e) synapsin I and (f) PSD95 in hippocampal neurons were determined by Western blot. Representative protein bands are presented at the top of the profile plot cartograms, showing the estimation of marginal means of (e) synapsin I and (f) PSD95 protein levels (n = 4 cultures). Statistical analysis was performed with factorial design ANOVA. Western blot analysis also showed that GCM collected from CRH-treated glial cells with CRHR1 knockdown did not affect levels of synapsin I and PSD95 in hippocampal neurons. In contrast, the GCM collected from CRH-treated glial cells with control siRNA transfection inhibited the protein levels of synapsin I and PSD95 [Fig. 5(e) and 5(f), synapsin I: interaction F(1, 12) = 7.74, P = 0.017; PSD95: interaction F(1, 12) = 56.70, P < 0.001]. CXCL5 produced by glial cells contributes to CRH suppression of synapse formation We then identified the secretary factors regulated by CRHR1 in glial cells. Cytokine antibody array showed that CRH (10−8 M) treatment affected the output of various cytokines, including IL-6, agrin, CCL-2 (MCP-1), and CXCL5. Among them, CRH-induced inhibition of CXCL5 was blocked by CRHR1 antagonist antalarmin [Fig. 6(a) and 6(b), Supplemental Table 1]. Figure 6. View largeDownload slide CRH suppresses CXCL5 (LIX) production in glial cells and hippocampal slices via CRHR1. (a and b) Cytokine output in glial cells by CRH/CRHR1 signaling. Glial cells were treated with CRH (10−8 M) in the presence and absence of CRHR1 antagonist antalarmin (10−7 M) for 48 hours. The media were collected for cytokine antibody array as described in Materials and Methods. (a) The raw data of the cytokine antibody array. (b) The histogram of summarizing the cytokine antibody array. (c and d) The role of CRH/CRHR1 signaling in CXCL5 secretion in glia. (c) Glial cells were treated with increasing concentrations of CRH or combined CRH and CRHR antagonists for 48 hours. The media were collected for determination of CXCL5 by ELISA (n = 4 cultures). (d) Estimation of marginal means of CXCL5 level in CRH-treated GCM collected from glial cells with CRHR1 knockdown. Glial cells were treated with CRH (10−8 M) in glial cells transfected with CRHR1 siRNA or vehicle siRNA. The media were collected for determination of CXCL5 by ELISA (n = 4 cultures). (e) The effect of CRH on CXCL5 content in hippocampal slices. Hippocampal slices were treated with CRH or in a combination of CRHR antagonists. The slices were collected for determination of CXCL5 content by ELISA (n = 4 cultures). Data are presented as mean ± SEM. The data for Fig. 6(d) were analyzed by factorial design ANOVA. The data for other figures were analyzed by one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; ##P < 0.01 vs CRH. Anta, antalarmin; As2B, astressin2B. Figure 6. View largeDownload slide CRH suppresses CXCL5 (LIX) production in glial cells and hippocampal slices via CRHR1. (a and b) Cytokine output in glial cells by CRH/CRHR1 signaling. Glial cells were treated with CRH (10−8 M) in the presence and absence of CRHR1 antagonist antalarmin (10−7 M) for 48 hours. The media were collected for cytokine antibody array as described in Materials and Methods. (a) The raw data of the cytokine antibody array. (b) The histogram of summarizing the cytokine antibody array. (c and d) The role of CRH/CRHR1 signaling in CXCL5 secretion in glia. (c) Glial cells were treated with increasing concentrations of CRH or combined CRH and CRHR antagonists for 48 hours. The media were collected for determination of CXCL5 by ELISA (n = 4 cultures). (d) Estimation of marginal means of CXCL5 level in CRH-treated GCM collected from glial cells with CRHR1 knockdown. Glial cells were treated with CRH (10−8 M) in glial cells transfected with CRHR1 siRNA or vehicle siRNA. The media were collected for determination of CXCL5 by ELISA (n = 4 cultures). (e) The effect of CRH on CXCL5 content in hippocampal slices. Hippocampal slices were treated with CRH or in a combination of CRHR antagonists. The slices were collected for determination of CXCL5 content by ELISA (n = 4 cultures). Data are presented as mean ± SEM. The data for Fig. 6(d) were analyzed by factorial design ANOVA. The data for other figures were analyzed by one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; ##P < 0.01 vs CRH. Anta, antalarmin; As2B, astressin2B. To confirm the effect of CRH on CXCL5 output, we measured CXCL5 levels in the culture media of the neurons treated with increasing concentrations of CRH. It was showed that CRH dose-dependently decreased CXCL5 output. This effect was prevented by CRHR1 antagonist antalarmin [Fig. 6(c); F(7, 24) = 20.20, P < 0.001; CRH plus antalarmin vs CRH: LSD-t(24) = 7.11, P < 0.001] and CRHR1 siRNA [Fig. 6(d), interaction F(1, 12) = 20.14, P < 0.001]. We also found that CRH treatment caused a decrease in CXCL5 content in the cultured hippocampal slice, which was blocked by antalarmin [Fig. 6(e), F(7, 24) = 25.30, P < 0.001; CRH vs vehicle, LSD-t(24) = −8.38, P < 0.001; CRH plus antalarmin vs CRH, LSD-t(24) = 8.69, P < 0.001]. We subsequently examined whether CXCL5 mediates CRH modulation of synapse formation in cultured hippocampal neurons. At first, we found that administering CXCR2 antagonist SB265610 to the neurons incubated with vehicle GCM reduced synapsin I–labeled and PSD95-labeled terminals and clusters [Fig. 7(a) and 7(b), synapsin I, F(5, 18) = 16.88, P < 0.001; SB265610 plus vehicle GCM vs vehicle GCM: LSD-t(18) = −5.96, P < 0.001; PSD95: F(5, 18) = 17.18, P < 0.001, SB265610 plus vehicle GCM vs vehicle GCM: LSD-t(18) = −5.76, P < 0.001; costained synapses: F(5, 18) = 19.07, P < 0.001, SB265610 plus vehicle GCM vs vehicle GCM: LSD-t(18) = −6.35, P < 0.001]. Administration of anti-CXCL5 (anti-LIX) antibody exhibited similar effects. It was shown that anti-CXCL5 antibody treatment decreased synapsin I–positive and PSD95-positive terminals and clusters in the neurons incubated with vehicle GCM [synapsin I, LSD-t(18) = −4.01, P = 0.001; PSD95: LSD-t(18) = −4.65, P < 0.001; costained synapses: LSD-t(18) = −4.56, P < 0.001]. However, in the neurons incubated with CRH-treated GCM, additional application of purified rLIX reversed the inhibitory effects of CRH-treated GCM on synapsin I–positive and PSD95-positive terminals [synapsin I: LSD-t(18) = 5.61, P < 0.001; PSD95: LSD-t(18) = 5.72, P < 0.001; costained synapses: LSD-t(18) = 5.75, P < 0.001]. These effects could be blocked by administration of CXCR2 antagonist SB265610 [synapsin I: LSD-t(18) = −5.90, P < 0.001; PSD95: LSD-t(18) = −5.93, P < 0.001; costained synapses: LSD-t(18) = −6.28, P < 0.001]. Figure 7. View largeDownload slide Reduced CXCL5 production in glial cells contributes to CRH inhibition of synapse formation in hippocampal neurons. (a and b) Cultured hippocampal neurons were incubated with GCM containing CXCR2 antagonist SB265610 and CXCL5 (LIX) antibody or with CRH-treated GCM containing SB265610 and LIX for 48 hours. Immunofluorescence analysis of synapsin I–labeled and PSD95-labeled terminals was performed as described in Materials and Methods. Presynaptic terminals were labeled by synapsin I (green). Postsynaptic terminals were labeled by PSD95 (red). (a) Representative fluorescence images of synapse formation in hippocampal neurons incubated with various GCMs as indicated. Scale bar represents 10 μm. (b) Cumulative data of the effects of various GCMs on the numbers of synapsin I, PSD95, and synapsin I and PSD95 costaining terminals. Data are presented as mean ± SEM (n = 4). (c–f) The effects of various GCMs on (c and e) synapsin I and (d and f) PSD95 expression in hippocampal neurons. Hippocampal neurons were incubated with the GCM containing CXCR2 antagonist SB265610 and LIX antibody or with CRH-treated GCM containing SB265610 and LIX for 48 hours. The levels of synapsin I and PSD95 were determined by Western blot. Representative protein bands are presented at the top of the histogram (n = 5 cultures). Data are presented as mean ± SEM. Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control GCM; ##P < 0.01 vs CRH-treated GCM; &&P < 0.01 vs CRH-treated GCM plus LIX. Figure 7. View largeDownload slide Reduced CXCL5 production in glial cells contributes to CRH inhibition of synapse formation in hippocampal neurons. (a and b) Cultured hippocampal neurons were incubated with GCM containing CXCR2 antagonist SB265610 and CXCL5 (LIX) antibody or with CRH-treated GCM containing SB265610 and LIX for 48 hours. Immunofluorescence analysis of synapsin I–labeled and PSD95-labeled terminals was performed as described in Materials and Methods. Presynaptic terminals were labeled by synapsin I (green). Postsynaptic terminals were labeled by PSD95 (red). (a) Representative fluorescence images of synapse formation in hippocampal neurons incubated with various GCMs as indicated. Scale bar represents 10 μm. (b) Cumulative data of the effects of various GCMs on the numbers of synapsin I, PSD95, and synapsin I and PSD95 costaining terminals. Data are presented as mean ± SEM (n = 4). (c–f) The effects of various GCMs on (c and e) synapsin I and (d and f) PSD95 expression in hippocampal neurons. Hippocampal neurons were incubated with the GCM containing CXCR2 antagonist SB265610 and LIX antibody or with CRH-treated GCM containing SB265610 and LIX for 48 hours. The levels of synapsin I and PSD95 were determined by Western blot. Representative protein bands are presented at the top of the histogram (n = 5 cultures). Data are presented as mean ± SEM. Statistical analysis was performed with one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control GCM; ##P < 0.01 vs CRH-treated GCM; &&P < 0.01 vs CRH-treated GCM plus LIX. Western blot analysis confirmed that CXCR2 antagonist significantly inhibited synapsin I and PSD95 levels in neurons incubated with vehicle GCM [Fig. 7(c) and 7(d), synapsin I:, F(4, 20) = 15.77, P < 0.001; SB265610 plus vehicle GCM vs vehicle GCM: LSD-t(20) = −2.44, P = 0.022; PSD95: F(4, 20) = 14.10, P < 0.001; SB265610 plus vehicle GCM vs vehicle GCM: LSD-t(20) = −2.28, P = 0.034]. GCM containing anti-LIX antibody exhibited similar inhibitory effects on synapsin I and PSD95 expression [Fig. 7(e) and 7(f), synapsin I: F(5, 24) = 79.97, P < 0.001; anti-LIX plus vehicle GCM vs vehicle GCM: LSD-t(24) = −11.63, P < 0.001; PSD95: F(5, 24) = 134.56, P < 0.001; anti-LIX plus vehicle GCM vs vehicle GCM: LSD-t(24) = −14.61, P < 0.001]. Administration of rLIX blocked the inhibitory effect of CRH-treated GCM on synapsin I and PSD95 expression [Fig. 7(c) and 7(d), synapsin I: LSD-t(20) = 5.31, P < 0.001; PSD95: LSD-t(20) = 3.48, P = 0.002]. This effect was reversed by CXCR2 antagonist SB265610 [Fig. 7(c) and 7(d), synapsin I: LSD-t(20) = −5.79, P < 0.001; PSD95: LSD-t(20) = −5.76, P < 0.001]. CRH inhibits CXCL5 production in astrocytes through multiple signaling pathways Astrocytes, microglia, and oligodendrocytes are main types of glial cells in rodents. We showed that astrocytes are the predominant cell type in glia isolated from the hippocampus (Supplemental Fig. 1). We therefore investigated the signaling pathways responsible for CRH suppressing CXCL5 secretion in cultured astrocytes. CRH (10−12 to 10−8 M) decreased cAMP production and activated Gi protein in a dose-dependent manner [Fig. 8(a), F(5, 18) = 88.54, P < 0.001; Fig. 8(b), F(5, 18) = 30.55, P < 0.001]. These effects were prevented by either CRHR1 antagonist [CRH plus antalarmin vs CRH: Gi protein level, LSD-t(18) = −7.31, P < 0.001; cAMP production, LSD-t(18) = 9.29, P < 0.001] or CRHR1 siRNA [Fig. 8(c), interaction F(1, 12) = 11.21, P = 0.006]. PTX (1.5 μg/mL), a blocker of Gi protein, blocked the inhibitory effect of CRH on CXCL5 release [Fig. 8(d), F(3, 12) = 25.69, P < 0.001; CRH vs vehicle: LSD-t(12) = −8.13, P < 0.001; CRH plus PTX vs CRH: LSD-t(12) = 6.87, P < 0.001]. Application of AC inhibitor SQ22536 mimicked CRH suppression of CXCL5 secretion, whereas AC activator forskolin or cAMP analog 8-Br-cAMP reversed CRH inhibition of CXCL5 secretion [Fig. 8(e), F(6, 21) = 27.45, P < 0.001; CRH vs vehicle, LSD-t(21) = −10.36, P < 0.001; SQ22536 vs control: LSD-t(21) = −9.40, P < 0.001; CRH plus forskolin vs CRH, LSD-t(21) = 6.72, P < 0.001; CRH plus 8-Br-cAMP vs CRH, LSD-t(21) = 7.61, P < 0.001]. Figure 8. View largeDownload slide CRH inhibits CXCL5 output in astrocytes through multiple signaling pathways. (a) The effects of CRH on Gi protein activation in astrocytes. Astrocytes were treated with increasing concentrations of CRH in the presence or absence of CRHR1 antagonists for 5 minutes. The cells were collected for determination of GTP-bound Gi protein as described in Materials and Methods. Representative protein bands of GTP-bound Gi protein are on the top of histograms (n = 4 cultures). (b and c) The effects of CRH/CRHR1 signaling on cAMP production in astrocytes. (b) The concentration of cAMP in astrocytes treated with increasing concentrations of CRH in the presence or absence of CRHR1 antagonist for 10 minutes (n = 4 cultures). (c) Profile plot cartogram shows the estimation of marginal means of cAMP content in astrocytes treated with CRH in the presence of CRHR1 siRNA (n = 4 cultures). The content of cAMP in cells was determined by RIA. (d and e) The effects of PTX, SQ22536, forskolin, and 8-Br-cAMP on CRH inhibition of CXCL5 secretion in astrocytes. Astrocytes were treated with CRH (10−8 M) in the presence or absence of (d) PTX (n = 4 cultures), (e) SQ22536 (n = 4 cultures), (e) forskolin, and (e) 8-Br-cAMP for 48 hours. The media were collected for determination of CXCL5 content. (f and g) The effects of CRH on PLC-β3 activation and IP3 production in astrocytes. Astrocytes were treated with CRH (10−8 M) in the presence or absence of antalarmin for 10 minutes. The levels of (f) phosphorylated PLC-β3 (n = 4 cultures) and (g) IP3 (n = 4 cultures) were determined by Western blot and ELISA, respectively. Representative protein bands of pPLC-β3 are presented at the top of the histogram. (h) The effect of CRHR1 siRNA on CRH modulation of IP3 production. Astrocytes were transfected with CRHR1 siRNA or control siRNA and then treated with CRH (10−8 M) for 10 minutes. The concentration of IP3 was determined by ELISA. Profile plot cartogram shows the estimation of marginal means of IP3 production (n = 4 cultures). (i) The effects of PLC inhibitor U73122, PKC nonselective inhibitor chelerythrine, and PKCα/β inhibitor Gö6976 on CRH inhibition of CXCL5 secretion in astrocytes. Astrocytes were treated with CRH (10−8 M) in the presence or absence of U73122, chelerythrine, and Gö6976 for 48 hours. The media were collected for determination of CXCL5 content (n = 4 cultures). Data are presented as mean ± SEM. The data for Fig. 8(c) and 8(h) were statistically analyzed by factorial design ANOVA, whereas the data for other figures were analyzed by one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; #P < 0.05; ##P < 0.01 vs CRH. Anta, antalarmin. Figure 8. View largeDownload slide CRH inhibits CXCL5 output in astrocytes through multiple signaling pathways. (a) The effects of CRH on Gi protein activation in astrocytes. Astrocytes were treated with increasing concentrations of CRH in the presence or absence of CRHR1 antagonists for 5 minutes. The cells were collected for determination of GTP-bound Gi protein as described in Materials and Methods. Representative protein bands of GTP-bound Gi protein are on the top of histograms (n = 4 cultures). (b and c) The effects of CRH/CRHR1 signaling on cAMP production in astrocytes. (b) The concentration of cAMP in astrocytes treated with increasing concentrations of CRH in the presence or absence of CRHR1 antagonist for 10 minutes (n = 4 cultures). (c) Profile plot cartogram shows the estimation of marginal means of cAMP content in astrocytes treated with CRH in the presence of CRHR1 siRNA (n = 4 cultures). The content of cAMP in cells was determined by RIA. (d and e) The effects of PTX, SQ22536, forskolin, and 8-Br-cAMP on CRH inhibition of CXCL5 secretion in astrocytes. Astrocytes were treated with CRH (10−8 M) in the presence or absence of (d) PTX (n = 4 cultures), (e) SQ22536 (n = 4 cultures), (e) forskolin, and (e) 8-Br-cAMP for 48 hours. The media were collected for determination of CXCL5 content. (f and g) The effects of CRH on PLC-β3 activation and IP3 production in astrocytes. Astrocytes were treated with CRH (10−8 M) in the presence or absence of antalarmin for 10 minutes. The levels of (f) phosphorylated PLC-β3 (n = 4 cultures) and (g) IP3 (n = 4 cultures) were determined by Western blot and ELISA, respectively. Representative protein bands of pPLC-β3 are presented at the top of the histogram. (h) The effect of CRHR1 siRNA on CRH modulation of IP3 production. Astrocytes were transfected with CRHR1 siRNA or control siRNA and then treated with CRH (10−8 M) for 10 minutes. The concentration of IP3 was determined by ELISA. Profile plot cartogram shows the estimation of marginal means of IP3 production (n = 4 cultures). (i) The effects of PLC inhibitor U73122, PKC nonselective inhibitor chelerythrine, and PKCα/β inhibitor Gö6976 on CRH inhibition of CXCL5 secretion in astrocytes. Astrocytes were treated with CRH (10−8 M) in the presence or absence of U73122, chelerythrine, and Gö6976 for 48 hours. The media were collected for determination of CXCL5 content (n = 4 cultures). Data are presented as mean ± SEM. The data for Fig. 8(c) and 8(h) were statistically analyzed by factorial design ANOVA, whereas the data for other figures were analyzed by one-way ANOVA followed by LSD-t test. *P < 0.05; **P < 0.01 vs control; #P < 0.05; ##P < 0.01 vs CRH. Anta, antalarmin. CRH treatment also dose-dependently increased the level of phosphorylated PLC-β3, the active PLC-β3, which was prevented by CRHR1 antagonist [Fig. 8(f), F(5, 18) = 179.82, P < 0.001; CRH vs vehicle, LSD-t(18) = 25.24, P < 0.001; CRH plus antalarmin vs CRH, LSD-t(18) = −20.26, P < 0.001]. CRH administration also increased IP3 production. CRHR1 antagonist and CRHR1 siRNA could block the aforementioned effect [Fig. 8(g), F(5, 18) = 20.90, P < 0.001; CRH vs vehicle, LSD-t(18) = 8.97, P < 0.001; CRH plus antalarmin vs CRH, LSD-t(18) = −7.77, P < 0.001; Fig. 8(h), interaction F(1, 12) = 40.95, P < 0.001]. Administration of PLC inhibitor U73122, protein kinase C (PKC) nonselective inhibitor chelerythrine, or PKCα/β inhibitor Gö6976 completely abolished the CRH-induced suppression of CXCL5 secretion [Fig. 8(i), F(7, 24) = 36.20, P < 0.001; CRH vs vehicle, LSD-t(24) = −11.37, P < 0.001; CRH plus U73122 vs CRH, LSD-t(24) = 8.88, P < 0.001; CRH plus chelerythrine vs CRH, LSD-t(24) = 6.14, P < 0.001; CRH plus Gö6976 vs CRH, LSD-t(24) = 10.64, P < 0.001]. Discussion This study demonstrated that CRH acts on glia to inhibit CXCL5 secretion, subsequently leading to suppression of synapse formation in the hippocampus. These data may indicate that glia–neuron intercommunication is one of the mechanisms responsible for neuronal circuits remodeling during stress. The hippocampus, one of the most plastic regions in brain, is highly sensitive to stress (33–35). Many studies have demonstrated that stress induces structural changes of synapses in the hippocampus, with concomitant changes in cognitive function and affective behavior (4–6, 36, 37). As mentioned, CRH is one of the key factors involved in stress-induced synaptic abnormalities, particularly synapse-bearing dendritic spines in the hippocampus (7–10, 38). CRH is synthesized mainly by pyramidal cell layer interneurons and released during stress (39, 40). Blockage of CRHR1 prevented stress-induced spine loss and dendritic atrophy induced by stress in the hippocampus (39, 41). Liao et al. (42) recently reported that CRHR1 antagonist attenuated downregulation of synapse-related proteins induced by stress in the hippocampus. In addition, an in vitro study by Chen et al. (43) showed that CRH suppresses dendritic spine density in hippocampal slices. Consistently, the current study showed that CRH suppressed synaptic formation in the hippocampus via CRHR1. The molecular mechanisms by which CRH modulates synapse formation in the hippocampus remain largely unknown. Emerging evidence has indicated that glial cells, such as astrocytes, play critical roles in neuronal differentiation and circuitry formation in the last decade (44, 45). In the hippocampus, about 64% of synapses are contacted by perisynaptic astrocytes at the synaptic cleft (46, 47). The intricate arborization and ramifications of astrocytes allow them to tightly enwrap the synaptic terminal to modulate synaptic processes and plasticity (46, 48). Previous studies have implied that glia might play a role in CRH modulation of synapse formation because it has been shown that CRH displays differential effects on synaptic formation in the model of cultured neurons and brain slices (11, 12). Interestingly, the current study also demonstrated the opposite effects of CRH on synapse formation in isolated hippocampal neurons and organotypic hippocampal slices. Moreover, we have shown that similar effects of CRH on synaptic-related protein expression and synaptic formation occurred in hippocampal slices and glia–neuron cocultures. Knockdown of CRHR1 in glial cells prevented CRH suppression of synapse formation in the model of glia–neuron cocultures. These data provided the strong evidence that glial cells played a key role in CRH suppression of synaptic formation in the hippocampus. The glial cells regulate synapse remodeling and plasticity through contact molecules and soluble secreted factors. A number of soluble factors secreted by glia have been identified to be involved in synapse formation, such as thrombospondins, transforming growth factor -β, and BDNF (16, 19, 20). In the current study, we demonstrated that a chemokine, CXCL5 (LIX), played a critical role in maintenance of synaptic formation in the hippocampus, as evidenced by blockage of CXCL5 leading to decreased synapse formation in glia–neuron cocultures. A number of studies have demonstrated that glia can synthesize and secrete various cytokines, such as IL-1, IL-6, and IL-10, and chemokines including CCL2, CCL3, CCL4, CXCL12, and CXCL14 (49–57). These cytokines and chemokines are involved in neurogenesis, neuronal survival, synaptic transmission, and synaptic plasticity. For instance, IL-10 promotes synapse formation in cultured cortical neurons (52), whereas chemokines CCL2, CCL3, and CXCL14 have been shown to modulate synaptic transmission (53, 54, 57). In the current study, we confirmed that CRHR1 was expressed in hippocampal glia in rats. Furthermore, we demonstrated that CRHR1 activation led to inhibition of AC/cAMP signaling and activation of PLC/PKC signaling in astrocytes, and these signaling pathways mediated CRH inhibition of CXCL5 secretion. CRH receptors, particularly CRHR1, can couple to multiple G proteins with subsequent activation of multiple signaling pathways in various tissues (58). Our previous studies have shown that CRHR1 can couple to Gs and Gq proteins in hippocampal neurons of rats (59, 60); in contrast, it couples to Gi, Gs, and Gq in human myometrium (61). Interestingly, Takuma et al. (62) reported that CRH increases Ca2+ influx via PLC/PKC pathway in cultured astrocytes of rats. Here we demonstrated that CRHR1 coupled with Gi and Gq proteins and subsequently led to inhibition of AC/cAMP signaling and activation of PLC/PKC signaling in astrocytes, respectively. Although no direct evidence indicates that CRHR1 leads to activation of multiple signaling pathways in glia in vivo, Blank et al. (63) showed that CRH can induce diverse signaling pathways via multiple G proteins in cultured hippocampal slices. However, our present data are not consistent with the study by Chen et al. (64), who showed that CRHR1 activation leads to accumulation of cAMP in astrocytes of cortex in rats. It is very hard to uncover the reason for the discrepant data between two groups. However, the glial cells collected from different brain regions (cortex in Chen et al.’s vs hippocampus in ours) might account for it. Nevertheless, we have demonstrated CRHR1 activation of a Gi/AC/cAMP signaling pathway by the following evidence: CRHR1 activation leads to activation of Gi protein and a decrease in cAMP production, and blockage of Gi activation prevents CRHR1 inhibition of CXCL5 secretion. As mentioned, CRH can be secreted from interneurons in the hippocampus during stress (40, 41), which contributes to spine loss during stress (9). We therefore propose that, during stress, increased CRH secretion inhibits CXCL5 secretion by glia via CRHR1, leading to suppression of synapse formation in hippocampal neurons (Fig. 9). Our study may indicate that glia–neuron communication is critical for remodeling of synapse circuits in the brain during stress. Figure 9. View largeDownload slide Scheme illustrating CRH regulation of synapse formation in the hippocampus. CXCL5 secreted by glial cells plays a critical role in synaptic terminal formation of pyramidal neurons in the hippocampus. CRH can be released from interneurons by various stimuli (such as stress) in the hippocampus. CRH then acts on CRHR1 in glia to suppress CXCL5 secretion from glia, resulting in a decrease in synapses in pyramidal neurons. AC, adenylyl cyclase; PKA, protein kinase A. Figure 9. View largeDownload slide Scheme illustrating CRH regulation of synapse formation in the hippocampus. CXCL5 secreted by glial cells plays a critical role in synaptic terminal formation of pyramidal neurons in the hippocampus. CRH can be released from interneurons by various stimuli (such as stress) in the hippocampus. CRH then acts on CRHR1 in glia to suppress CXCL5 secretion from glia, resulting in a decrease in synapses in pyramidal neurons. AC, adenylyl cyclase; PKA, protein kinase A. Appendix. Antibody Table Peptide/Protein Target  Antigen Sequence (if Known)  Name of Antibody  Manufacturer, Catalog No.  Species Raised in; Monoclonal or Polyclonal  Dilution Used  RRID  Synapsin I    Anti-synapsin I antibody  Millipore, AB1543P  Rabbit; polyclonal  IF 1:200, WB 1:1000  AB_90757  Postsynaptic density 95 kDa    Anti-PSD95 antibody  Abcam, ab2723  Mouse; monoclonal  IF 1:200, WB 1:1000  AB_303248  Corticotropin-releasing factor receptor 1    CRF-R1 polyclonal antibody  Bioworld, BS2590  Rabbit; polyclonal  WB 1:1000  AB_1663553  Glial fibrillary acid protein    GFAP antibody  Santa Cruz Biotechnology Inc, sc-6170  Goat; polyclonal  IF 1:200  AB_641021  PLC-β3    Mouse anti-PLC β3 monoclonal antibody  Santa Cruz Biotechnology Inc, sc133231  Mouse; monoclonal  WB 1:1000  AB_2299534  Phosphorylated PLC-β3 (Ser 537)    Phospho-PLC β3 (Ser537) antibody  Cell Signaling Technology, 2481  Rabbit; polyclonal  WB 1:1000  AB_2163265  β-Actin    Mouse anti-β-actin monoclonal antibody  Sigma-Aldrich, A5441  Mouse; monoclonal  WB 1:10000  AB_476744  Rabbit IgG    Donkey anti-rabbit IgG secondary antibody, Alexa Fluor 594 conjugate  Thermo Fisher, R37119  Donkey; polyclonal  IF 1:400  AB_2556547  Goat IgG    Donkey anti-goat IgG (H+L) secondary antibody, Alexa Fluor 488 conjugate  Thermo Fisher, A-11055  Donkey; polyclonal  IF 1:400  AB_2534102  Mouse IgG    Donkey anti-mouse IgG secondary antibody, Alexa Fluor 594 conjugate  Thermo Fisher, R37115  Donkey; polyclonal  IF 1:400  AB_2556543  Rabbit IgG    Donkey anti-rabbit IgG (H+L) secondary antibody, Alexa Fluor 488 conjugate  Thermo Fisher, A-21206  Donkey; polyclonal  IF 1:400  AB_2535792  Rabbit IgG    Donkey anti-rabbit IgG-HRP  Santa Cruz Biotechnology Inc, sc-2313  Donkey; polyclonal  WB 1:1000  AB_641181  Mouse IgG    Donkey anti-mouse IgG-HRP  Santa Cruz Biotechnology Inc, sc-2314  Donkey; polyclonal  WB 1:1000  AB_641170  Peptide/Protein Target  Antigen Sequence (if Known)  Name of Antibody  Manufacturer, Catalog No.  Species Raised in; Monoclonal or Polyclonal  Dilution Used  RRID  Synapsin I    Anti-synapsin I antibody  Millipore, AB1543P  Rabbit; polyclonal  IF 1:200, WB 1:1000  AB_90757  Postsynaptic density 95 kDa    Anti-PSD95 antibody  Abcam, ab2723  Mouse; monoclonal  IF 1:200, WB 1:1000  AB_303248  Corticotropin-releasing factor receptor 1    CRF-R1 polyclonal antibody  Bioworld, BS2590  Rabbit; polyclonal  WB 1:1000  AB_1663553  Glial fibrillary acid protein    GFAP antibody  Santa Cruz Biotechnology Inc, sc-6170  Goat; polyclonal  IF 1:200  AB_641021  PLC-β3    Mouse anti-PLC β3 monoclonal antibody  Santa Cruz Biotechnology Inc, sc133231  Mouse; monoclonal  WB 1:1000  AB_2299534  Phosphorylated PLC-β3 (Ser 537)    Phospho-PLC β3 (Ser537) antibody  Cell Signaling Technology, 2481  Rabbit; polyclonal  WB 1:1000  AB_2163265  β-Actin    Mouse anti-β-actin monoclonal antibody  Sigma-Aldrich, A5441  Mouse; monoclonal  WB 1:10000  AB_476744  Rabbit IgG    Donkey anti-rabbit IgG secondary antibody, Alexa Fluor 594 conjugate  Thermo Fisher, R37119  Donkey; polyclonal  IF 1:400  AB_2556547  Goat IgG    Donkey anti-goat IgG (H+L) secondary antibody, Alexa Fluor 488 conjugate  Thermo Fisher, A-11055  Donkey; polyclonal  IF 1:400  AB_2534102  Mouse IgG    Donkey anti-mouse IgG secondary antibody, Alexa Fluor 594 conjugate  Thermo Fisher, R37115  Donkey; polyclonal  IF 1:400  AB_2556543  Rabbit IgG    Donkey anti-rabbit IgG (H+L) secondary antibody, Alexa Fluor 488 conjugate  Thermo Fisher, A-21206  Donkey; polyclonal  IF 1:400  AB_2535792  Rabbit IgG    Donkey anti-rabbit IgG-HRP  Santa Cruz Biotechnology Inc, sc-2313  Donkey; polyclonal  WB 1:1000  AB_641181  Mouse IgG    Donkey anti-mouse IgG-HRP  Santa Cruz Biotechnology Inc, sc-2314  Donkey; polyclonal  WB 1:1000  AB_641170  Abbreviations: HRP, horseradish peroxidase; IF, immunofluorescence; RRID, Research Resource Identifier; WB, Western blot. View Large Abbreviations: ANOVA analysis of variance BDNF brain-derived neurotrophic factor CRH corticotropin-releasing hormone CRHR1 corticotropin-releasing hormone receptor type 1 CXCL5 C-X-C motif chemokine 5 DMEM Dulbecco’s modified Eagle medium ELISA enzyme-linked immunosorbent assay FBS fetal bovine serum GCM glia-conditioned media LIX C-X-C motif chemokine 5 LSD least significant difference P postnatal day PBS phosphate-buffered saline PKC protein kinase C PLC phospholipase C PSD95 postsynaptic density protein 95 PTX pertussis toxin rLIX recombinant C-X-C motif chemokine 5 ROI region of interest SDS sodium dodecyl sulfate SEM standard error of the mean siRNA small interfering RNA. Acknowledgments The authors thank Dr. Chen Wu from the Department of Statistics, Second Military Medical University, Shanghai, China, for help with statistical analysis. Financial Support: This work was supported by the Major State Basic Research Program of China (2013CB967404) and the National Natural Science Foundation of China (31100840). Disclosure Summary: The authors have nothing to disclose. References 1. Owens MJ, Nemeroff CB. Physiology and pharmacology of corticotropin-releasing factor. Pharmacol Rev . 1991; 43( 4): 425– 473. Google Scholar PubMed  2. Cummings S, Elde R, Ells J, Lindall A. Corticotropin-releasing factor immunoreactivity is widely distributed within the central nervous system of the rat: an immunohistochemical study. J Neurosci . 1983; 3( 7): 1355– 1368. Google Scholar PubMed  3. Herman JP, Prewitt CM, Cullinan WE. Neuronal circuit regulation of the hypothalamo-pituitary-adrenocortical stress axis. Crit Rev Neurobiol . 1996; 10( 3–4): 371– 394. Google Scholar CrossRef Search ADS PubMed  4. Chen Y, Rex CS, Rice CJ, Dubé CM, Gall CM, Lynch G, Baram TZ. Correlated memory defects and hippocampal dendritic spine loss after acute stress involve corticotropin-releasing hormone signaling. Proc Natl Acad Sci USA . 2010; 107( 29): 13123– 13128. Google Scholar CrossRef Search ADS PubMed  5. Wang XD, Su YA, Wagner KV, Avrabos C, Scharf SH, Hartmann J, Wolf M, Liebl C, Kühne C, Wurst W, Holsboer F, Eder M, Deussing JM, Müller MB, Schmidt MV. Nectin-3 links CRHR1 signaling to stress-induced memory deficits and spine loss. Nat Neurosci . 2013; 16( 6): 706– 713. Google Scholar CrossRef Search ADS PubMed  6. Wang XD, Rammes G, Kraev I, Wolf M, Liebl C, Scharf SH, Rice CJ, Wurst W, Holsboer F, Deussing JM, Baram TZ, Stewart MG, Müller MB, Schmidt MV. Forebrain CRF1 modulates early-life stress-programmed cognitive deficits. J Neurosci . 2011; 31( 38): 13625– 13634. Google Scholar CrossRef Search ADS PubMed  7. Gounko NV, Swinny JD, Kalicharan D, Jafari S, Corteen N, Seifi M, Bakels R, van der Want JJ. Corticotropin-releasing factor and urocortin regulate spine and synapse formation: structural basis for stress-induced neuronal remodeling and pathology. Mol Psychiatry . 2013; 18( 1): 86– 92. Google Scholar CrossRef Search ADS PubMed  8. Chen Y, Fenoglio KA, Dubé CM, Grigoriadis DE, Baram TZ. Cellular and molecular mechanisms of hippocampal activation by acute stress are age-dependent. Mol Psychiatry . 2006; 11( 11): 992– 1002. Google Scholar CrossRef Search ADS PubMed  9. Chen Y, Dubé CM, Rice CJ, Baram TZ. Rapid loss of dendritic spines after stress involves derangement of spine dynamics by corticotropin-releasing hormone. J Neurosci . 2008; 28( 11): 2903– 2911. Google Scholar CrossRef Search ADS PubMed  10. Wang XD, Chen Y, Wolf M, Wagner KV, Liebl C, Scharf SH, Harbich D, Mayer B, Wurst W, Holsboer F, Deussing JM, Baram TZ, Müller MB, Schmidt MV. Forebrain CRHR1 deficiency attenuates chronic stress-induced cognitive deficits and dendritic remodeling. Neurobiol Dis . 2011; 42( 3): 300– 310. Google Scholar CrossRef Search ADS PubMed  11. Sheng H, Xu Y, Chen Y, Zhang Y, Xu X, He C, Ni X. CRH-R1 and CRH-R2 differentially modulate dendritic outgrowth of hippocampal neurons. Endocrine . 2012; 41( 3): 458– 464. Google Scholar CrossRef Search ADS PubMed  12. Chen Y, Bender RA, Brunson KL, Pomper JK, Grigoriadis DE, Wurst W, Baram TZ. Modulation of dendritic differentiation by corticotropin-releasing factor in the developing hippocampus. Proc Natl Acad Sci USA . 2004; 101( 44): 15782– 15787. Google Scholar CrossRef Search ADS PubMed  13. Fields RD, Stevens-Graham B. New insights into neuron–glia communication. Science . 2002; 298( 5593): 556– 562. Google Scholar CrossRef Search ADS PubMed  14. Allen NJ, Barres BA. Neuroscience: Glia—more than just brain glue. Nature . 2009; 457( 7230): 675– 677. Google Scholar CrossRef Search ADS PubMed  15. Stogsdill JA, Eroglu C. The interplay between neurons and glia in synapse development and plasticity. Curr Opin Neurobiol . 2017; 42: 1– 8. Google Scholar CrossRef Search ADS PubMed  16. Christopherson KS, Ullian EM, Stokes CC, Mullowney CE, Hell JW, Agah A, Lawler J, Mosher DF, Bornstein P, Barres BA. Thrombospondins are astrocyte-secreted proteins that promote CNS synaptogenesis. Cell . 2005; 120( 3): 421– 433. Google Scholar CrossRef Search ADS PubMed  17. Mauch DH, Nägler K, Schumacher S, Göritz C, Müller EC, Otto A, Pfrieger FW. CNS synaptogenesis promoted by glia-derived cholesterol. Science . 2001; 294( 5545): 1354– 1357. Google Scholar CrossRef Search ADS PubMed  18. Allen NJ, Bennett ML, Foo LC, Wang GX, Chakraborty C, Smith SJ, Barres BA. Astrocyte glypicans 4 and 6 promote formation of excitatory synapses via GluA1 AMPA receptors. Nature . 2012; 486( 7403): 410– 414. Google Scholar PubMed  19. Diniz LP, Almeida JC, Tortelli V, Vargas Lopes C, Setti-Perdigão P, Stipursky J, Kahn SA, Romão LF, de Miranda J, Alves-Leon SV, de Souza JM, Castro NG, Panizzutti R, Gomes FC. Astrocyte-induced synaptogenesis is mediated by transforming growth factor β signaling through modulation of D-serine levels in cerebral cortex neurons. J Biol Chem . 2012; 287( 49): 41432– 41445. Google Scholar CrossRef Search ADS PubMed  20. Fan Y, Chen J, Ye J, Yan H, Cai Y. Proteinase-activated receptor 2 modulates corticotropin releasing hormone-induced brain-derived neurotrophic factor release from microglial cells. Cell Biol Int . 2014; 38( 1): 92– 96. Google Scholar CrossRef Search ADS PubMed  21. Wang W, Ji P, Dow KE. Corticotropin-releasing hormone induces proliferation and TNF-alpha release in cultured rat microglia via MAP kinase signalling pathways. J Neurochem . 2003; 84( 1): 189– 195. Google Scholar CrossRef Search ADS PubMed  22. Burton SD, Johnson JW, Zeringue HC, Meriney SD. Distinct roles of neuroligin-1 and SynCAM1 in synapse formation and function in primary hippocampal neuronal cultures. Neuroscience . 2012; 215: 1– 16. Google Scholar CrossRef Search ADS PubMed  23. Devader C, Khayachi A, Veyssière J, Moha Ou Maati H, Roulot M, Moreno S, Borsotto M, Martin S, Heurteaux C, Mazella J. In vitro and in vivo regulation of synaptogenesis by the novel antidepressant spadin. Br J Pharmacol . 2015; 172( 10): 2604– 2617. Google Scholar CrossRef Search ADS PubMed  24. Cuesto G, Jordán-Álvarez S, Enriquez-Barreto L, Ferrús A, Morales M, Acebes Á. GSK3β inhibition promotes synaptogenesis in Drosophila and mammalian neurons. PLoS One . 2015; 10( 3): e0118475. Google Scholar CrossRef Search ADS PubMed  25. Zeng M, Shang Y, Araki Y, Guo T, Huganir RL, Zhang M. Phase transition in postsynaptic densities underlies formation of synaptic complexes and synaptic plasticity. Cell . 2016; 166( 5): 1163– 1175.e12. Google Scholar CrossRef Search ADS PubMed  26. Jeyifous O, Lin EI, Chen X, Antinone SE, Mastro R, Drisdel R, Reese TS, Green WN. Palmitoylation regulates glutamate receptor distributions in postsynaptic densities through control of PSD95 conformation and orientation. Proc Natl Acad Sci USA . 2016; 113( 52): E8482– E8491. Google Scholar CrossRef Search ADS PubMed  27. Peña F. Organotypic cultures as tool to test long-term effects of chemicals on the nervous system. Curr Med Chem . 2010; 17( 10): 987– 1001. Google Scholar CrossRef Search ADS PubMed  28. Bender R, Heimrich B, Meyer M, Frotscher M. Hippocampal mossy fiber sprouting is not impaired in brain-derived neurotrophic factor–deficient mice. Exp Brain Res . 1998; 120( 3): 399– 402. Google Scholar CrossRef Search ADS PubMed  29. Elliott-Hunt CR, Kazlauskaite J, Wilde GJ, Grammatopoulos DK, Hillhouse EW. Potential signalling pathways underlying corticotrophin-releasing hormone–mediated neuroprotection from excitotoxicity in rat hippocampus. J Neurochem . 2002; 80( 3): 416– 425. Google Scholar CrossRef Search ADS PubMed  30. Nelson TE, Gruol DL. The chemokine CXCL10 modulates excitatory activity and intracellular calcium signaling in cultured hippocampal neurons. J Neuroimmunol . 2004; 156( 1-2): 74– 87. Google Scholar CrossRef Search ADS PubMed  31. Duan S, Anderson CM, Keung EC, Chen Y, Chen Y, Swanson RA. P2X7 receptor–mediated release of excitatory amino acids from astrocytes. J Neurosci . 2003; 23( 4): 1320– 1328. Google Scholar PubMed  32. Zheng Y, Zhang YM, Ni X. Urocortin 2 but not urocortin 3 promotes the synaptic formation in hippocampal neurons via induction of NGF production by astrocytes. Endocrinology . 2016; 157( 3): 1200– 1210. Google Scholar CrossRef Search ADS PubMed  33. Nemeroff CB, Bremner JD, Foa EB, Mayberg HS, North CS, Stein MB. Posttraumatic stress disorder: a state-of-the-science review. J Psychiatr Res . 2006; 40( 1): 1– 21. Google Scholar CrossRef Search ADS PubMed  34. Naninck EF, Hoeijmakers L, Kakava-Georgiadou N, Meesters A, Lazic SE, Lucassen PJ, Korosi A. Chronic early life stress alters developmental and adult neurogenesis and impairs cognitive function in mice. Hippocampus . 2015; 25( 3): 309– 328. Google Scholar CrossRef Search ADS PubMed  35. Barzegar M, Sajjadi FS, Talaei SA, Hamidi G, Salami M. Prenatal exposure to noise stress: anxiety, impaired spatial memory, and deteriorated hippocampal plasticity in postnatal life. Hippocampus . 2015; 25( 2): 187– 196. Google Scholar CrossRef Search ADS PubMed  36. Xia B, Chen C, Zhang H, Xue W, Tang J, Tao W, Wu R, Ren L, Wang W, Chen G. Chronic stress prior to pregnancy potentiated long-lasting postpartum depressive-like behavior, regulated by Akt-mTOR signaling in the hippocampus. Sci Rep . 2016; 6( 1): 35042. Google Scholar CrossRef Search ADS PubMed  37. Karisetty BC, Maitra S, Wahul AB, Musalamadugu A, Khandelwal N, Guntupalli S, Garikapati R, Jhansyrani T, Kumar A, Chakravarty S. Differential effect of chronic stress on mouse hippocampal memory and affective behavior: Role of major ovarian hormones. Behav Brain Res . 2017; 318: 36– 44. Google Scholar CrossRef Search ADS PubMed  38. Andres AL, Regev L, Phi L, Seese RR, Chen Y, Gall CM, Baram TZ. NMDA receptor activation and calpain contribute to disruption of dendritic spines by the stress neuropeptide CRH. J Neurosci . 2013; 33( 43): 16945– 16960. Google Scholar CrossRef Search ADS PubMed  39. Chen Y, Bender RA, Frotscher M, Baram TZ. Novel and transient populations of corticotropin-releasing hormone–expressing neurons in developing hippocampus suggest unique functional roles: a quantitative spatiotemporal analysis. J Neurosci . 2001; 21( 18): 7171– 7181. Google Scholar PubMed  40. Chen Y, Brunson KL, Adelmann G, Bender RA, Frotscher M, Baram TZ. Hippocampal corticotropin releasing hormone: pre- and postsynaptic location and release by stress. Neuroscience . 2004; 126( 3): 533– 540. Google Scholar CrossRef Search ADS PubMed  41. Ivy AS, Rex CS, Chen Y, Dubé C, Maras PM, Grigoriadis DE, Gall CM, Lynch G, Baram TZ. Hippocampal dysfunction and cognitive impairments provoked by chronic early-life stress involve excessive activation of CRH receptors. J Neurosci . 2010; 30( 39): 13005– 13015. Google Scholar CrossRef Search ADS PubMed  42. Liao XM, Yang XD, Jia J, Li JT, Xie XM, Su YA, Schmidt MV, Si TM, Wang XD. Blockade of corticotropin-releasing hormone receptor 1 attenuates early-life stress-induced synaptic abnormalities in the neonatal hippocampus. Hippocampus . 2014; 24( 5): 528– 540. Google Scholar CrossRef Search ADS PubMed  43. Chen Y, Kramár EA, Chen LY, Babayan AH, Andres AL, Gall CM, Lynch G, Baram TZ. Impairment of synaptic plasticity by the stress mediator CRH involves selective destruction of thin dendritic spines via RhoA signaling. Mol Psychiatry . 2013; 18( 4): 485– 496. Google Scholar CrossRef Search ADS PubMed  44. Allen NJ. Role of glia in developmental synapse formation. Curr Opin Neurobiol . 2013; 23( 6): 1027– 1033. Google Scholar CrossRef Search ADS PubMed  45. Todd KJ, Serrano A, Lacaille JC, Robitaille R. Glial cells in synaptic plasticity. J Physiol Paris . 2006; 99( 2–3): 75– 83. Google Scholar CrossRef Search ADS PubMed  46. Ota Y, Zanetti AT, Hallock RM. The role of astrocytes in the regulation of synaptic plasticity and memory formation. Neural Plast . 2013; 2013: 185463. Google Scholar CrossRef Search ADS PubMed  47. Witcher MR, Kirov SA, Harris KM. Plasticity of perisynaptic astroglia during synaptogenesis in the mature rat hippocampus. Glia . 2007; 55( 1): 13– 23. Google Scholar CrossRef Search ADS PubMed  48. Matyash V, Kettenmann H. Heterogeneity in astrocyte morphology and physiology. Brain Res Brain Res Rev . 2010; 63( 1–2): 2– 10. Google Scholar CrossRef Search ADS   49. Schneider H, Pitossi F, Balschun D, Wagner A, del Rey A, Besedovsky HO. A neuromodulatory role of interleukin-1beta in the hippocampus. Proc Natl Acad Sci USA . 1998; 95( 13): 7778– 7783. Google Scholar CrossRef Search ADS PubMed  50. Ross FM, Allan SM, Rothwell NJ, Verkhratsky A. A dual role for interleukin-1 in LTP in mouse hippocampal slices. J Neuroimmunol . 2003; 144( 1–2): 61– 67. Google Scholar CrossRef Search ADS PubMed  51. Balschun D, Wetzel W, Del Rey A, Pitossi F, Schneider H, Zuschratter W, Besedovsky HO. Interleukin-6: a cytokine to forget. FASEB J . 2004; 18( 14): 1788– 1790. Google Scholar PubMed  52. Chen H, Lin W, Zhang Y, Lin L, Chen J, Zeng Y, Zheng M, Zhuang Z, Du H, Chen R, Liu N. IL-10 promotes neurite outgrowth and synapse formation in cultured cortical neurons after the oxygen–glucose deprivation via JAK1/STAT3 pathway. Sci Rep . 2016; 6( 1): 30459. Google Scholar CrossRef Search ADS PubMed  53. Bray JG, Reyes KC, Roberts AJ, Ransohoff RM, Gruol DL. Synaptic plasticity in the hippocampus shows resistance to acute ethanol exposure in transgenic mice with astrocyte-targeted enhanced CCL2 expression. Neuropharmacology . 2013; 67: 115– 125. Google Scholar CrossRef Search ADS PubMed  54. Marciniak E, Faivre E, Dutar P, Alves Pires C, Demeyer D, Caillierez R, Laloux C, Buée L, Blum D, Humez S. The chemokine MIP-1α/CCL3 impairs mouse hippocampal synaptic transmission, plasticity and memory. Sci Rep . 2015; 5( 1): 15862. Google Scholar CrossRef Search ADS PubMed  55. Lee YB, Nagai A, Kim SU. Cytokines, chemokines, and cytokine receptors in human microglia. J Neurosci Res . 2002; 69( 1): 94– 103. Google Scholar CrossRef Search ADS PubMed  56. Guyon A. CXCL12 chemokine and its receptors as major players in the interactions between immune and nervous systems. Front Cell Neurosci . 2014; 8( 65): 65. Google Scholar PubMed  57. Banisadr G, Bhattacharyya BJ, Belmadani A, Izen SC, Ren D, Tran PB, Miller RJ. The chemokine BRAK/CXCL14 regulates synaptic transmission in the adult mouse dentate gyrus stem cell niche. J Neurochem . 2011; 119( 6): 1173– 1182. Google Scholar CrossRef Search ADS PubMed  58. Hillhouse EW, Grammatopoulos DK. The molecular mechanisms underlying the regulation of the biological activity of corticotropin-releasing hormone receptors: implications for physiology and pathophysiology. Endocr Rev . 2006; 27( 3): 260– 286. Google Scholar CrossRef Search ADS PubMed  59. Sheng H, Zhang Y, Sun J, Gao L, Ma B, Lu J, Ni X. Corticotropin-releasing hormone (CRH) depresses n-methyl-D-aspartate receptor–mediated current in cultured rat hippocampal neurons via CRH receptor type 1. Endocrinology . 2008; 149( 3): 1389– 1398. Google Scholar CrossRef Search ADS PubMed  60. Sheng H, Xu Y, Chen Y, Zhang Y, Ni X. Corticotropin-releasing hormone stimulates mitotic kinesin–like protein 1 expression via a PLC/PKC-dependent signaling pathway in hippocampal neurons. Mol Cell Endocrinol . 2012; 362( 1–2): 157– 164. Google Scholar CrossRef Search ADS PubMed  61. You X, Gao L, Liu J, Xu C, Liu C, Li Y, Hui N, Gu H, Ni X. CRH activation of different signaling pathways results in differential calcium signaling in human pregnant myometrium before and during labor. J Clin Endocrinol Metab . 2012; 97( 10): E1851– E1861. Google Scholar CrossRef Search ADS PubMed  62. Takuma K, Matsuda T, Yoshikawa T, Kitanaka J, Gotoh M, Hayata K, Baba A. Corticotropin-releasing factor stimulates Ca2+ influx in cultured rat astrocytes. Biochem Biophys Res Commun . 1994; 199( 3): 1103– 1107. Google Scholar CrossRef Search ADS PubMed  63. Blank T, Nijholt I, Grammatopoulos DK, Randeva HS, Hillhouse EW, Spiess J. Corticotropin-releasing factor receptors couple to multiple G-proteins to activate diverse intracellular signaling pathways in mouse hippocampus: role in neuronal excitability and associative learning. J Neurosci . 2003; 23( 2): 700– 707. Google Scholar PubMed  64. Chen SJ, Yang JF, Kong FP, Ren JL, Hao K, Li M, Yuan Y, Chen XC, Yu RS, Li JF, Leng G, Chen XQ, Du JZ. Overactivation of corticotropin-releasing factor receptor type 1 and aquaporin-4 by hypoxia induces cerebral edema. Proc Natl Acad Sci USA . 2014; 111( 36): 13199– 13204. Google Scholar CrossRef Search ADS PubMed  Copyright © 2018 Endocrine Society

Journal

EndocrinologyOxford University Press

Published: Feb 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off