Sánchez-Ventura et al. Journal of Neuroinflammation (2019) 16:124 https://doi.org/10.1186/s12974-019-1511-7 RESEARCH Open Access BET protein inhibition regulates cytokine production and promotes neuroprotection after spinal cord injury Judith Sánchez-Ventura, Jesús Amo-Aparicio, Xavier Navarro and Clara Penas* Abstract Background: Spinal cord injury (SCI) usually causes a devastating lifelong disability for patients. After a traumatic lesion, disruption of the blood-spinal cord barrier induces the infiltration of macrophages into the lesion site and the activation of resident glial cells, which release cytokines and chemokines. These events result in a persistent inflammation, which has both detrimental and beneficial effects, but eventually limits functional recovery and contributes to the appearance of neuropathic pain. Bromodomain and extra-terminal domain (BET) proteins are epigenetic readers that regulate the expression of inflammatory genes by interacting with acetylated lysine residues. While BET inhibitors are a promising therapeutic strategy for cancer, little is known about their implication after SCI. Thus, the current study was aimed to investigate the anti-inflammatory role of BET inhibitors in this pathologic condition. Methods: We evaluated the effectiveness of the BET inhibitor JQ1 to modify macrophage reactivity in vitro and to modulate inflammation in a SCI mice model. We analyzed the effects of BET inhibition in pro-inflammatory and anti-inflammatory cytokine production in vitro and in vivo. We determined the effectiveness of BET inhibition in tissue sparing, inflammation, neuronal protection, and behavioral outcome after SCI. Results: We have found that the BET inhibitor JQ1 reduced the levels of pro-inflammatory mediators and increased the expression of anti-inflammatory cytokines. A prolonged treatment with JQ1 also decreased reactivity of microglia/macrophages, enhanced neuroprotection and functional recovery, and acutely reduced neuropathic pain after SCI. Conclusions: BET protein inhibition is an effective treatment to regulate cytokine production and promote neuroprotection after SCI. These novel results demonstrate for the first time that targeting BET proteins is an encouraging approach for SCI repair and a potential strategy to treat other inflammatory pathologies. Keywords: BET protein, Inflammation, Cytokine, Neuroprotection, Spinal cord injury Background are no effective treatments available [2]. Thus, it is import- Spinal cord injury (SCI) commonly causes lifelong disabil- ant to develop new therapeutic strategies to alleviate this ity that reduces the quality of life and increases morbidity harmful condition. for the affected patients [1, 2]. Direct consequences of the SCI courses with an inflammatory response that com- injury are the loss of motor, sensory and autonomic func- prises mainly microglia and blood-derived macrophages tions, producing paralysis, anesthesia, and dysautonomia. [4, 5]. Both immune cells induce and magnify the im- Besides, neuropathic pain arises in many cases, further re- mune response by secreting inflammatory molecules and ducing the quality of life of SCI patients [3]. To date, there influencing glial cells including astrocytes, oligodendro- cytes, and microglia [6]. Thus, the reaction of the im- * Correspondence: Clara.Penas@uab.cat mune system induces a feed-forward response that Institut of Neurosciences, Dept Cell Biology, Physiology and Immunology, Centro de Investigación Biomédica en Red sobre Enfermedades produces a persistent pro-inflammatory microenviron- Neurodegenerativas (CIBERNED), Universitat Autonoma de Barcelona, ment, which contributes to cell death and glial scar Barcelona, Spain © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Sánchez-Ventura et al. Journal of Neuroinflammation (2019) 16:124 Page 2 of 12 formation [2, 7]. Secreted pro-inflammatory cytokines ketamine (90 mg/kg), xylazine (10 mg/kg), and saline (such as IL-1β, IL-6, TNF-α), as well as cytotoxic factors, mixture. After laminectomy, a moderate contusion including reactive oxygen and nitrogen species, contrib- was performed to the exposed spinal cord at T11 ute to enhance inflammation by recruiting inflammatory using the Infinite Horizon Impactor device (60 kDyn, cells to clear debris at the injury area [8], but have the tissue displacement between 450–600 μm). Animals potential to cause toxicity and increase tissue damage [5, received a i.p. dose of 30 mg/kg JQ1 diluted in 5% 6, 9, 10]. In addition, the persistent release of these pro- DMSO and 5% Tween-80 in saline 2 h post-operation inflammatory modulators plays a critical role in the in- and daily onwards until specified. Vehicle-treated ani- duction and maintenance of neuropathic pain [11]. In mals received only the vehicle solution. Sham mice contrast, anti-inflammatory cytokines, such as IL-4, IL- underwent the same surgery except the contusion 10, and IL-13, help to suppress inflammation by redu- injury. cing pro-inflammatory cytokine/chemokine production and contributing to wound healing and tissue remodel- BMDM culture ing [9, 10, 12]. Bone marrow-derived macrophages (BMDMs) were ob- Bromodomain and extra-terminal domain (BET) pro- tained from mouse femurs following a specific isolating teins belong to the reader family. BET proteins are char- protocol [17]. Briefly, femurs from 8-week-old C57BL/6 acterized by the presence of two tandem bromodomains mice were cut in a sterile glass and the bone marrow (BD1 and BD2) and an extra-terminal domain (ET), cavity was flushed out with chilled PBS using 10-ml syr- which are critical for their function [13, 14]. These pro- inge with 25-G needle. Once the cells were isolated and teins bind to acetylated lysine residues in both histone centrifuged at 500×g, cells were cultured in DMEM/F-12 and non-histone proteins to recruit transcriptional com- medium containing HEPES, 1% 200 mM L-glutamine, plexes and thus regulate gene expression. As a result, 10% fetal bovine serum, and 1% penicillin/streptomycin. genes related to proliferation and inflammation are In addition, to initiate macrophage differentiation, the expressed [13–15]. BET family includes BRD2, BRD3, medium was supplemented with 10 ng/ml of macro- BRD4, and BRDT, and among them, BRD4 seem to be phage colony-stimulating factor (M-CSF; Stanley) for 9– the main responsible for pro-inflammatory gene expres- 10 days. During differentiation, the medium was replaced sion by interacting with the acetylated lysine-310 at the every 3 days to eliminate contaminating non-adherent p65 transactivation domain of NF-kB [14, 16]. Taking to- cells. After the differentiation phase, adherent BMDMs gether, the inhibition of this epigenetic phenomenon is a were seeded at a density of 0.3 × 106 cells/well in two promising target to control the inflammatory process. different 6-well plates containing the medium above Given the importance of BET proteins in inflamma- mentioned. tion, we hypothesized that these epigenetic mediators At days 9 and 10, DMSO (0.001%), JQ1 (1 μM), or LPS might be a therapeutic target after SCI. We used the (100 ng/ml) alone or a combination of JQ1 and LPS was BET inhibitor JQ1 and analyzed the effects after SCI. added to the cultures. First, cells were treated with 100 We observed that JQ1 produced a decrease of pro- ng/ml of LPS to initiate the inflammatory response dur- inflammatory and an increase of anti-inflammatory cyto- ing 1 h. Then, either DMSO or 1 μM JQ1 was added for kine expression acutely after SCI. We found that pro- 2 h. Finally, cells were collected for RNA extraction. longed treatment with JQ1 reduced inflammatory response, increased neuroprotection, and enhanced RNA isolation, RNA reverse transcription, and RT-qPCR functional outcome after SCI. Overall, the results of the For in vitro analysis, RNA was extracted using the present study indicate that BET proteins are a thera- RNeasy Micro Kit (QIAGEN) according to the manufac- peutic target after SCI. turer’s recommendations. For in vivo analysis, adult mice were perfused with sterile saline 4 h (n = 24) or 4 days Materials and methods post-SCI (n = 14). Then, a 4-mm-long segment around Spinal cord injury the SCI epicenter was removed and snap-frozen. Tissue All experimental procedures were approved by the was homogenized with QIAzol lysis reagent (QIAGEN), Universitat Autònoma de Barcelona Animal Experimentation and RNA was extracted using the RNeasy Mini Kit Ethical Committee (CEEAH 1188R3-DMAH 6131) (QIAGEN) following the manufacturer’s guidelines. and followed the European Communities Council RNA was quantified with a spectrophotometer (Nano- Directive 2010/63/EU. A total of 72 8-week-old fe- Drop Technologies) and reverse-transcribed using an male C57BL/6 mice were used for all the procedures Applied Biosystems kit (Thermo Fisher Scientific). Then, and randomly divided into three groups depending on the expression of target sequences was quantified by RT- the experimental design. General anesthesia was in- qPCR using SYBR ® Green QPCR Master Mix (Agilent duced by an intraperitoneal injection (i.p.) of a Technologies) and the corresponding primers (Table 1). Sánchez-Ventura et al. Journal of Neuroinflammation (2019) 16:124 Page 3 of 12 Table 1 List of primers used in RT-qPCR Gene F′ R′ ARG1 GTGAAGAACCCACGGTCTGT CCAGAGATGCTTCCAACTGC IL-1b CTTCAAATCTCACAGCAGCACATC CCACGGGAAAGACACAGGTAG IL-4 CATGGGAAAACTCCATGCTT TGGACTCATTCATGGTGCAG IL-6 AACCACGGCCTTCCCTACTTCA TCATTTCCACGATTTCCCAGAG IL-10 GCTCCTAGAGCTGCGGACT TCATTTCCGATAAGGCTTGG IL-13 TCCAATTGCAATGCCATCTA TGGGCTACTTCGATTTTGGT CD206 ATTGTGGAGCAGATGGAAGG ATTTGCATTGCCCAGTAAGG CCL2 ATGGGTACCGTCACAACCTC CCTGCTGCTGGTGATTCTCTT CCL5 TGCCCACGTCAAGGAGTATTTCTA TGGCGGTTCCTTCGAGTGACAA CD68 CCAATTCAGGGTGGAAGAAA ATGGGTACCGTCACAACCTC CX3CR1 CTTTGGGGGCATATTCTTCA ACGCCCAGACTAATGGTGAC GAPDH TGGCCTTCCGTGTTCCTAC GAGTTGCTGTTGAAGTCG GFAP GGAGAGGGACAACTTTGCAC CAGCCTCAGGTTGGTTTCAT INOS AATCTTGGAGCGAGTTGTGG CAGGAAGTAGGTGAGGGCTTG TNFa AGGCACTCCCCCAAAAGATG TCACCCCGAAGTTCAGTAGAC Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) indicating painful perception. Thermal algesimetry test was used as a housekeeping gene. [20] assesses the time that mice can endure heat ir- radiation. This test was performed at 28 days once the Cytokine protein expression mice could sustain their weight on the hind paws. Adult female C57/Bl6 mice were perfused with sterile saline and a 5-mm-long segment of sham or the con- Histological analyses tused spinal cord was collected at 4 and 72 h after SCI Animals used for the functional assessment were sacri- (n = 3 per group and time point) and snap-frozen. The ficed and perfused with 4% paraformaldehyde in 0.1 M spinal cords were homogenized, and protein concentra- phosphate buffer (PB) at day 28 (n = 16). An 8-mm length tion was determined using the DC Protein Assay (Bio- of the spinal cord containing the epicenter of the lesion Rad). Samples were concentrated to 4 mg/ml using was removed and cut into 20-μm-thick sagittal sections MicroCon centrifugation filters (Millipore) to ensure using a cryostat. A subset of sections was used to evaluate equal amounts of protein. Cytokine protein levels were demyelination with Luxol Fast Blue staining (LFB; Sigma- then analyzed using the Milliplex MAP Mouse Cyto- Aldrich). Briefly, after dehydration, slides were placed in a kine/Chemokine magnetic bead panel (Millipore) on a 1 mg/mL LFB solution in 95% ethanol and 0.05% acetic Luminex (Millipore) as per the manufacturers’ protocol. acid at 37 °C overnight. Then, sections were washed in 95% ethanol and distilled water before being placed into a Functional assessment 0.05% Li2CO3 solution for 3 min at room temperature. Fi- Locomotion recovery was evaluated in an open-field nally, after dehydration, sections were mounted in DPX test using the 9-point Basso Mouse Scale (BMS) at 3, mounting medium (Sigma-Aldrich). Myelin sparing of the 7, 10, 14, 21, and 28 days post-operation [18]. This three groups of animals was calculated by delineating the scale ranges from 0, indicating complete paralysis, to spared LFP-stained tissue area, and the total area, from 9, indicating normal movement of the hind limbs. images taken at the epicenter of the injury and every The analysis was carried out by two researchers 100 μm rostral and caudal to the lesion site using ImageJ which strictly followed blinding procedures to deter- software. Then, the area of residual white matter was mine the score for each animal. To evaluate neuro- expressed as the percentage with respect to the total area pathic pain, mechanical and thermal algesimetry tests of the spinal cord section. were performed. Mechanical nociception was mea- For immunohistochemical staining, another subset of sured once a week (at 7, 14, 21, and 28 days post- sections was rehydrated in PBS and blocked with 5% FBS operation) by an electronic Randall-Selitto test [19]. in PBS 0.3Tx for 1 h at room temperature. The spinal cord Briefly, an increasing mechanical force was applied sections were then incubated overnight at 4 °C with pri- with a probe on the plantar surface of the hind paws mary antibodies against glial fibrillary acidic protein until the animal produced retraction of the paw, (GFAP; for astrocytes; 1:1000; AB5804-Millipore), ionized Sánchez-Ventura et al. Journal of Neuroinflammation (2019) 16:124 Page 4 of 12 calcium-binding adapter molecule 1 (Iba1; for macro- treated with JQ1 compared to vehicle-treated mice phages/microglia; 1:500, ab5076-Abcam), and NeuN (for and for CCL2 at 4 h and CX3CR1 at 72 h post-lesion neurons; 1:75; MAB377x-Millipore, conjugated with Alexa (Additional file 1: Figure S1B) in sham-treated mice 488). Nuclei were labeled with DAPI staining (1:1000; compared to vehicle-treated animals. These results D9564-10MG-Sigma). The detection was made with ap- validate previous studies [21]. propriate secondary antibodies conjugated with Alexa In addition, our results revealed that the anti- Fluor 594 (A11058 for Goat and A21207 for Rabbit; Invi- inflammatory effect of JQ1 resides not only on reducing trogen). Finally, after several washes in PBST-Tween 0.1%, pro-inflammatory markers but also to upregulate anti- PBS and PB, a coverslip was applied in Fluoromount-G inflammatory cytokines. JQ1 increased IL-4 and IL-13 mounting medium (Southern Biotech). Microphotographs mRNA expression at 4 h post-injury and IL-10 levels at were taken at × 20, and then, immunoreactivity was ana- 72 h after SCI (Fig. 1b). To note, JQ1 increased IL-4 and lyzed by measuring the integrated density of a region of IL-13 also in sham-treated animals (Fig. 1b). To ascer- interest (ROI) after defining a threshold for background tain if the changes observed on cytokine mRNA levels correction. The ROI had an area of 0.05 mm2 and was were traduced on protein levels, we analyzed IL-4, IL-10, placed on the gray matter and on the white matter. Both IL-13, and IL-6 protein levels after SCI. IL-4 levels were markers were measured from 7 spinal cord sections (sepa- undetectable in most of the samples because the detec- rated 200 μm) of each animal. Finally, neuronal survival tion levels of the assay were too high for the biological was assessed by counting the number of neurons located levels of the protein (data not shown). However, we cor- rostrally and caudally from the epicenter in both ventral roborated the increase of IL-10 and IL-13 in sham and horns. Measurements were performed with ImageJ injured mice treated with JQ1 compared to vehicle- software. treated mice (Fig. 2). We also corroborated a decrease of IL-6 protein expression at 4 h post-injury in JQ1-treated Data analysis animals compared to vehicle-treated mice. Data are expressed as the mean ± standard error of the mean In addition, other inflammatory markers related to (SEM), and results were analyzed using GraphPad Prism 7 macrophage infiltration and microglial activation were software. For statistical analysis of cytokine/chemokine RNA studied. Considering macrophages and microglia, JQ1 expression, one-way analysis of variance (ANOVA) followed treatment after SCI greatly reduced the pro-inflammatory by Tukey’s post hoc test was performed. For cytokine protein macrophage marker INOS at 72 h post-lesion expression, unpaired t test was performed. Functional test (Additional file 1: Figure S1), although it did not affect the results were analyzed by two-way ANOVA followed anti-inflammatory macrophage markers ARG1 and by Sidak correction for multiple comparisons. Finally, CD206. Thus, apparently, BET inhibition at the dosage histological quantification results, including tissue used did not modify the balance of M1/M2 macrophage sparing and immunohistochemistry, were compared phenotype. However, a higher number of M1/M2 markers by two-way ANOVA followed by post hoc Tukey’s and fluorescence-activated cell sorting (FACS) should be multiple comparison test. All differences were consid- performed to confirm these results. Besides, astrocyte re- ered statistically significant when p < 0.05. activity determined by GFAP expression was not affected by JQ1 administration (Additional file 1: Figure S1B). Results Finally, we investigated whether the effects observed BET inhibition reduces pro-inflammatory and enhances after BET inhibition in SCI could be produced by af- anti-inflammatory cytokine production fecting macrophage reactivity. To simulate the per- After SCI, a large number of pro-inflammatory formed in vivo studies, and as a difference with markers become upregulated. To investigate the effi- previous similar studies where macrophages where cacy of BET inhibition to modulate the secondary in- pre-treated with JQ1 [15, 21], primary cultures of jury response, we monitored the expression of BMDMs received a stimulation with LPS during 1 h inflammation-related genes after SCI or in sham mice and followed or not by JQ1 treatment during 2 h. treated with the BET inhibitor JQ1 or with vehicle. Treatment with JQ1 resulted in a reduction of the As shown in Fig. 1a, at 4 h post-lesion, the expression pro-inflammatory modulators IL-6 and INOS after of the pro-inflammatory cytokines IL-6, IL-1β, and LPS stimulation (Additional file 2: Figure S2). Regard- TNF-α was significantly reduced in injured mice ing the anti-inflammatory cytokines IL-4, IL-10, and treated with JQ1 compared to vehicle-treated mice. IL-13, we found that administration of JQ1 upregu- From these cytokines, IL-1β was maintained downreg- lated their transcription. For IL-10 and IL-13, the up- ulated at 72 h post-lesion. We also detected a general regulation after JQ1 treatment was found with and decrease in chemokine levels after JQ1 treatment but without previous LPS stimulation, and for IL-4, only only significantly for CCL2 at 72 h in injured mice without LPS stimulation (Additional file 2: Figure S2). Sánchez-Ventura et al. Journal of Neuroinflammation (2019) 16:124 Page 5 of 12 Fig. 1 BET inhibitor JQ1 reduces expression of pro-inflammatory and increases expression of anti-inflammatory cytokines and chemokines acutely after SCI. Real-time PCR quantification of a pro-inflammatory cytokines and chemokines at 4 and 72 h after SCI and b anti-inflammatory cytokines at 4 and 72 h after injury, normalized to the GAPDH levels. N = 4–5 mice per group and time. *p < 0.05, **p < 0.01, ***p < 0.005, ****p < 0.001 by one-way ANOVA followed by Tukey post hoc test. Data are represented as mean ± SEM fold changes of gene expression Overall, while JQ1 suppressed the expression of key microglia/macrophages and astrocytes was analyzed at pro-inflammatory genes, it also produced an early ex- 28 days post-injury. Iba1 staining detects both quiescent pression of anti-inflammatory cytokines after SCI. The and reactive microglia and also infiltrated macrophages. changes observed in mRNA levels were concomitant The long-term JQ1 treatment showed a reduced immu- with changes on protein expression after SCI. noreactivity of Iba1 at the 200 μm rostrally to the lesion site (Fig. 3a), compared to vehicle treatment. The label- BET inhibition reduces microglia/macrophage reactivity ing for GFAP revealed that JQ1 did not affect astroglial after SCI reactivity (Fig. 3b). These results are consistent with our To further study the role of JQ1 in modulating the in- previous RT-qPCR findings at 72 h, showing no signifi- flammatory response after SCI, we evaluated the expres- cant differences in mRNA expression of GFAP after JQ1 sion of two hallmark markers of inflammation at longer administration (Additional file 1: Figure S1B). Therefore, time periods. Sham or SCI mice were treated with JQ1 these results confirm that BET inhibition reduces micro- or vehicle during 4 or 20 days, and immunoreactivity of glial reactivity without affecting astroglial reactivity. Sánchez-Ventura et al. Journal of Neuroinflammation (2019) 16:124 Page 6 of 12 Fig. 2 BET inhibitor JQ1 decreases pro-inflammatory and increases anti-inflammatory cytokine protein expression after SCI. The protein levels of the pro-inflammatory cytokine IL-6 (a) and the anti-inflammatory cytokines IL-10 and IL-13 (b) were quantified via Luminex analysis at 4 and 72 h after operation. N = 3–4 mice per group and time point. *p < 0.05, **p < 0.01, unpaired t test BET inhibition promotes tissue sparing and neuronal from the epicenter, as well as at 800 μm caudal in preservation after SCI long-term JQ1-treated animals compared to vehicle- The LFB staining results revealed that administration and to short-term-treated mice (Fig. 4e, f ). Overall, of JQ1 for 20 days led to significant white matter histopathological analysis revealed that long-term sparing in rostral and caudal areas to the injury treatment of JQ1 conferred significant neuroprotec- epicenter compared to vehicle-treated mice (Fig. 4b). tion after SCI. Besides, the total labeled myelin area was significantly higher in long-term JQ1-treated mice than in vehicle- BET inhibition enhances functional recovery after SCI treated mice (Fig. 4c). To assess whether BET inhibition had an effect on func- The number of surviving neurons in the ventral horn tional recovery after SCI, locomotion was tested in an was counted at the epicenter of the injury and at 200, open field and scored with BMS scale. As shown in 400, 600, and 800 μm rostral and caudally. While quanti- Fig. 5a, a gradual recovery of hindlimb locomotion was tative analysis of NeuN staining showed that most neu- observed over the weeks following SCI in all mice. How- rons died at the injury site and surrounding areas in the ever, only mice that received JQ1 long-term treatment three groups of mice, neuron survival was signifi- (n = 5) showed significant improvement of BMS scores cantly increased at different distances from the epi- at 14 and 21 days compared to vehicle-treated animals. center. As it is shown in Fig. 4e, f, neuronal survival Despite JQ1 administration during the first 4 days after was significantly increased at 600 and 800 μm rostral SCI (n = 6) slightly increased the BMS score, this Sánchez-Ventura et al. Journal of Neuroinflammation (2019) 16:124 Page 7 of 12 Fig. 3 JQ1 treatment reduces macrophage and microglia reactivity after SCI. a Iba1 and b GFAP immunoreactivity quantification at 800 μm rostrally and caudally from the injury epicenter. Histograms represent the mean integrated density ± SEM quantified in gray (left) and white (right) matter in the ventral zone of the spinal cord sections. Representative images of Iba1 (a) and GFAP (b) at 200 and 400 μm, respectively, rostral to the epicenter are shown. Scale bar = 100 μm. N = 4–5 mice per group. *p < 0.05 compared to vehicle-treated animals and #p < 0.05 compared to short-term-treated animals as calculated by two-way ANOVA with Tukey post hoc test increase was not significantly different from vehicle- Discussion treated mice at any time point. Although the interest in epigenetics has rapidly in- Finally, neuropathic pain was evaluated with Randall-Selitto creased over the last years, the function of epigenetic and thermal algesimetry tests (Fig. 5b). Mice treated signaling in central nervous system trauma remains with JQ1 for 20 days presented reduced mechanical largely unexplored. Thus, the overall goal of this allodynia compared to the other groups, with significant study was to investigate the therapeutic role of target- differences at 7 days. The thermal algesimetry test showed ing the epigenome, specifically BET proteins, after a non-significant increase in the withdrawal threshold to SCI. For this purpose, we used in vitro and in vivo hot stimulation in JQ1-treated mice. approaches to study the effect of the BET inhibitor Thus, BET inhibition enhances functional outcome JQ1 on inflammation, neuroprotection, functional re- and has a transient effect on neuropathic pain after SCI. covery, and ultimately on neuropathic pain. The Sánchez-Ventura et al. Journal of Neuroinflammation (2019) 16:124 Page 8 of 12 Fig. 4 BET inhibition increases tissue preservation and neuronal survival after SCI. a, b Quantification of the percentage of white matter sparing, with respect to the total area of the spinal cord, from the injury epicenter up to 1000 μm rostral and caudally. c Percentage of total spared white matter in 2-mm spinal cord samples from different mouse groups. d Representative micrographs of sections stained with LFB at 300 μm rostral to the epicenter from vehicle-treated, JQ1 short-term-treated and JQ1 long-term-treated animals. e Quantification of ventral horn neuron survival from the injury epicenter to 800 μm rostral and caudally. f Representative micrographs showing sparing ventral horn neurons from vehicle- treated, JQ1 short-term-treated and JQ1 long-term-treated animals at 600 μm rostral to the lesion site. N = 4–5 mice per group. Scale bars, 500 μm (d, f). *p < 0.05, ***p < 0.005, and ****p < 0.001 compared to vehicle-treated group and #p < 0.05 and ####p < 0.001 compared to short-term-treated group as calculated by two-way ANOVA with Tukey post hoc test. Data are represented as mean ± SEM results of the present study indicate that inhibition of present study highlight the importance of epigenetic BET proteins reduces inflammation and may repre- regulation on inflammation after SCI. sent a novel therapeutic target for SCI. We have ob- served that JQ1 decreased pro-inflammatory and Effects of BET inhibition on the inflammatory response increased anti-inflammatory cytokine expression One of the major findings of the present study is that acutely after SCI. We used JQ1 treatment at two dif- BET inhibition, besides reducing pro-inflammatory cyto- ferent time regimes and observed that the prolonged kine expression, which had been previously described treatment with JQ1 reduced the inflammatory re- [15, 21], enhances anti-inflammatory cytokine produc- sponse and increased neuroprotection. Altogether, tion. This effect was observed after SCI and in a primary these neuroimmune outcomes were associated with culture of bone marrow-derived macrophages. BET pro- an improvement of functional recovery and reduction teins are epigenetic readers, i.e., they read the acetylated in neuropathic pain. Overall, the results of the residues in histones located in specific promoters, and Sánchez-Ventura et al. Journal of Neuroinflammation (2019) 16:124 Page 9 of 12 Fig. 5 BET inhibition treatment increases functional outcome and reduces neuropathic pain after SCI. a BMS locomotion score along 28 days post-contusion. b Mechanical algesimetry results (left), expressed as percentage of baseline values, during follow-up, and thermal algesimetry results (right), expressed as absolute values in seconds, recorded at 28 days post-operation. N = 5–6 mice per group; *p < 0.05 as calculated by two-way ANOVA followed by Sidak correction for the multiple comparisons. Data are represented as mean ± SEM promote gene expression. Thus, its inhibition commonly in inducing an alternative activation to macrophages, reduces gene expression. The reduction of these which is linked to SCI recovery [1, 27, 28]. However, we inflammatory-related genes, which are mediated by NF- found that the enhancement in anti-inflammatory cyto- kB signaling, can be explained by the disruption of the kines could reduce the expression of the M1 marker normal functioning of BET proteins by JQ1 treatment iNOS but could not increase M2 markers (ARG1 and during an inflammatory response [22, 23]. However, mo- CD206), and thus, it might not affect the switch of mac- lecular mechanisms for increasing anti-inflammatory rophages into an anti-inflammatory phenotype. Accord- cytokine production after BET inhibition are elusive. ingly to literature, this result could be explained because This effect may be indirect. However, BET protein BRD4 not all of the M1 and M2 markers seem to change ex- can confer transcriptional repression by interacting with pression in a coordinated fashion after SCI [5]. Never- several repressive complexes [24–26]. Thus, it might be theless, a higher number of M1/M2 markers and possible that BET inhibition treatment produces fluorescence-activated cell sorting (FACS) of macro- chromatin release of these complexes and triggers phages and microglia harvested from the injured spinal promoter de-repression as suggested previously [25], cord would provide a better understanding of the inducing increases of anti-inflammatory cytokine pro- changes in macrophage phenotype after JQ1. Besides, it duction. However, further experiments should be per- might be possible also that the dosage or timing of the formed to decipher the molecular events undergoing treatment was not sufficient to induce a substantial in- these effects. crease of IL-4 to observe these effects. In fact, we were The importance of macrophages and microglia in SCI not able to detect IL-4 protein with the cytokine protein lies in their influence to exacerbate inflammation as well expression assay (data not shown). We cannot affirm as repair. These opposite actions are possible through that BET inhibition did not enhance IL-4 production, macrophage/microglia polarization [5]. The raised levels since the detection range of the assay was higher than of the anti-inflammatory cytokines IL-4, IL-10, and IL- the protein biological levels. However, these levels are 13 observed after JQ1 treatment are particularly interest- below those needed to produce a shift towards an M2 ing since many studies have highlighted their relevance phenotype. Sánchez-Ventura et al. Journal of Neuroinflammation (2019) 16:124 Page 10 of 12 Effects of BET inhibition on neuroprotection and SCI [29, 33], and reduce demyelination in multiple functional recovery sclerosis [34]. Besides, it is reported that BET pro- A challenge for researchers who work in the CNS teins play a role in most of neuronal types [21, 35]. trauma field is to control the cross-talk between the Therefore, further studies should be performed ana- immune and nervous system to prevent neurodegen- lyzing the role of BET inhibition in neuroprotection, eration and tissue loss. In the current study, we ob- axonal regeneration, and demyelination after a pro- served that JQ1 can promote neuroprotection, tissue longed treatment. preservation, and locomotion improvement. These beneficial effects may be attributed to the role of JQ1 Effects of BET inhibition on neuropathic pain modulating inflammation. The observed reduction in Functional assessment of neuropathic pain showed pres- pro-inflammatory cytokines and also the increase of ervation of the mechanical threshold at 7 days post- anti-inflammatory cytokines may overcome the neuro- injury. There are 2 main mechanisms thought to con- toxic environment of the acute phase of SCI, which tribute to hyperexcitability and therefore to the ontogeny eventually culminates in neuronal and oligodendrocyte of neuropathic pain (reviewed in [36]). One is the release death as well as tissue loss. In the end, these benefi- of neuromodulators such as pro-inflammatory cytokines cial outcomes might improve functional recovery. and chemokines, which enhance neuronal excitability. This conclusion is formed in view of the substantial The other one is the altered expression, trafficking, and literature related to the neurotoxic effects of TNF-α and functioning of receptors and ion channels expressed by IL-1β and the neuroprotective effects of IL-4 and IL-10 primary sensory neurons. Thus, probably BET treatment [5, 29]. Regarding pro-inflammatory cytokines, it has is only effective on reducing pro-inflammatory media- been described that blocking TNF-α and IL-1β confers tors, such as TNF-α, CCL2, and CX3CR1, which are dir- neuroprotection in SCI and stroke models [8]. For in- ectly implicated in neuropathic pain. Of interest, both stance, it has been reported that the genetic or pharma- CCR2 and CX3CR1-knockout mice exhibit reduced cological inhibition of TNF-α reduces inflammation, neuropathic pain [37]. A combination with any other injured tissue, and the number of apoptotic cells which epigenetic treatment that regulates gene expression in overall ameliorate the recovery of limb function [30]. primary sensory neurons could produce a prolonged ef- Considering anti-inflammatory cytokines, it has been fect on neuropathic pain. demonstrated that a treatment of IL-4 reduces tissue We and several authors report beneficial effects of damage and locomotion impairment after SCI [28]. BET inhibition reducing tumor growth [38–40], in- Despite our LFB staining revealed an enhanced tissue creasing brain plasticity [41], promoting neuroprotec- preservation, this fact cannot be directly linked to a pro- tion [42], and decreasing inflammation in several tective action against oligodendrocyte death. Thus, fu- pathologies [43, 44]. However, higher doses of JQ1 ture histological procedures will be performed to analyze can block memory formation [35] or produce lymph- the effects of JQ1 treatment on oligodendrocytes. How- oid and hematopoietic toxicity in mice [45] and can ever, we could observe enhanced neuronal survival, even enhance undesirable virus replication [46, 47]. In demonstrating a neuroprotective effect of JQ1 on these fact, we observed that doubling the dose of the com- cells after SCI, probably due by the decreased cytotoxic pound produced a decrease in body weight and environment produced by JQ1 treatment. This preserva- mouse death after SCI (data not shown). Therefore, tion was statistically significant in areas adjacent to the further studies carefully dissecting the dosage, analyz- injury epicenter. ing the molecular mechanisms of action, and deter- Our studies demonstrate an acute effect of BET in- mining undesirable effects of BET inhibitors should hibition by reducing pro-inflammatory and increasing be performed to find the balance between beneficial anti-inflammatory cytokine/chemokine levels after and detrimental effects of BET inhibition in each SCI. However, neuroprotection was observed only pathological condition and be able to progress to- after a prolonged treatment during 20 days and not wards clinical trials [48]. after a short treatment of 4 days, timing when most of the enhanced cytokines/chemokines return to con- Conclusions trol levels after SCI [8]. Thus, other chronic effects of This work elucidates an epigenetic mechanism which BET inhibition may also account for these mecha- contributes to the pathogenesis of inflammation and nisms. It may be possible that a sustained increase of SCI. Although BET inhibition has been previously re- anti-inflammatory cytokines could promote prolonged ported to have an anti-inflammatory effect after SCI neuroprotective/regenerative effects. In fact, IL-4, IL- [21], a beneficial outcome of BET inhibition treatment 10, and IL-13 high levels promote enhanced axonal after SCI has never been reported before. Thus, the regeneration [31, 32], promote neuroprotection after present study demonstrates for the first time that Sánchez-Ventura et al. Journal of Neuroinflammation (2019) 16:124 Page 11 of 12 BET proteins are a target for the treatment of SCI, Received: 15 January 2019 Accepted: 27 May 2019 promoting neuroprotection, behavioral recovery, and decreased neuropathic pain. Neuroinflammation is a common feature of several human neurodegenerative References diseases. Therefore, the discoveries from the present 1. Lima R, Monteiro S, Lopes JP, Barradas P, Vasconcelos NL, Gomes ED, et al. Systemic interleukin-4 administration after spinal cord injury modulates project may provide therapeutic innovation for many inflammation and promotes neuroprotection. Pharmaceuticals (Basel). 2017; other pathologies, increasing the translational impact 10(4). of the present study. 2. Silva NA, Sousa N, Reis RL, Salgado AJ. From basics to clinical: a comprehensive review on spinal cord injury. Prog Neurobiol. 2014;114:25–57. Additional files 3. van Gorp S, Kessels AG, Joosten EA, van Kleef M, Patijn J. Pain prevalence and its determinants after spinal cord injury: a systematic review. Eur J Pain. 2015;19(1):5–14. Additional file 1: JQ1 treatment modifies the M1 marker INOS but do 4. Alexander JK, Popovich PG. Neuroinflammation in spinal cord injury: not modify markers of M2 macrophage phenotype, or astrocyte reactivity. therapeutic targets for neuroprotection and regeneration. Prog Brain Res. Real-time PCR quantification of (A) M1 and M2 markers and (B) macro- 2009;175:125–37. phage and glial markers at 4 and 72 h after SCI, normalized to the GAPDH levels. N =3-4 mice/group. *p < 0.05, ***p < 0.005, as calculated 5. David S, Kroner A. Repertoire of microglial and macrophage responses after by one-way ANOVA followed by Tukey post-hoc test. Data are repre- spinal cord injury. Nat Rev Neurosci. 2011;12(7):388–99. sented as mean ± SEM fold changes of gene expression. (TIF 1771 kb) 6. Zhou X, He X, Ren Y. Function of microglia and macrophages in secondary damage after spinal cord injury. Neural Regen Res. 2014;9(20):1787–95. Additional file 2: BET inhibition affects macrophage reactivity in vitro. 7. Oyinbo CA. Secondary injury mechanisms in traumatic spinal cord Real-time PCR quantification of (A) pro-inflammatory and (B) anti- injury: a nugget of this multiply cascade. Acta Neurobiol Exp (Wars). inflammatory cytokines at 3 h after LPS (100 ng/ml) stimulation, normal- 2011;71(2):281–99. ized to the GAPDH levels. Experiments were repeated 3 independent 8. Donnelly DJ, Popovich PG. Inflammation and its role in neuroprotection, times. *p < 0.05, **p < 0.01, ***p< 0.005, as calculated by one-way ANOVA axonal regeneration and functional recovery after spinal cord injury. Exp followed by Tukey post-hoc test. Data are represented as mean ± SEM Neurol. 2008;209(2):378–88. fold changes of gene expression. (TIF 1378 kb) 9. Ren Y, Young W. Managing inflammation after spinal cord injury through manipulation of macrophage function. Neural Plast. 2013;2013:945034. 10. Hoeksema MA, de Winther MP. Epigenetic regulation of monocyte and Abbreviations macrophage function. Antioxid Redox Signal. 2016;25(14):758–74. ARG1: Arginase 1; BET: Bromodomain and extra-terminal domain; BMS: Basso 11. Walters ET. Nociceptors as chronic drivers of pain and hyperreflexia after Mouse Scale; CCL: C-C motif chemokine ligand; CD206: Cluster of spinal cord injury: an adaptive-maladaptive hyperfunctional state differentiation 2006; CXCR: C-X-C chemokine receptor; FACS: Fluorescence- hypothesis. Front Physiol. 2012;3:309. activated cell sorting; GFAP: Glial fibrillary acidic protein; Iba1: Ionized 12. Shapouri-Moghaddam A, Mohammadian S, Vazini H, Taghadosi M, Esmaeili calcium-binding adaptor molecule 1; IL: Interleukin; LBF: Luxol Fast Blue; SA, Mardani F, et al. Macrophage plasticity, polarization, and function in LPS: Lipopolysaccharide; NF-kB: Nuclear factor kappa-light-chain-enhancer of health and disease. J Cell Physiol. 2018;233(9):6425–40. activated B cells; SCI: Spinal cord injury; TNF-α: Tumor necrosis factor-α 13. Taniguchi Y. The bromodomain and extra-terminal domain (BET) family: functional anatomy of BET paralogous proteins. Int J Mol Sci. 2016;17(11). Acknowledgements 14. Jahagirdar R, Attwell S, Marusic S, Bendele A, Shenoy N, McLure KG, et al. We acknowledge Dr. James Bradner and Dana-Farber Cancer Institute (DFCI) RVX-297, a BET bromodomain inhibitor, has therapeutic effects in preclinical for kindly providing us with the compound JQ1. models of acute inflammation and autoimmune disease. Mol Pharmacol. 2017;92(6):694–706. Authors’ contributions 15. Nicodeme E, Jeffrey KL, Schaefer U, Beinke S, Dewell S, Chung CW, et al. JS performed the cell culture, qPCR, histological, and behavioral analyses. JA Suppression of inflammation by a synthetic histone mimic. Nature. 2010; performed behavioral analyses. JS, XN, and CP wrote the manuscript. CP 468(7327):1119–23. designed the study and performed the in vivo injuries. All authors read and 16. Hytti M, Tokarz P, Maatta E, Piippo N, Korhonen E, Suuronen T, et al. approved the final manuscript. Inhibition of BET bromodomains alleviates inflammation in human RPE cells. Biochem Pharmacol. 2016;110-111:71–9. Funding 17. Zhang X, Goncalves R, Mosser DM. The isolation and characterization of This work was supported by CIBERNED (CB06/05/1105) and TERCEL (RD16/ murine macrophages. Curr Protoc Immunol. 2008;Chapter 14(Unit 14):1. 0011/0014) funds from the Instituto de Salud Carlos III of Spain. CP holds a 18. Pajoohesh-Ganji A, Byrnes KR, Fatemi G, Faden AI. A combined scoring fellowship of the P-SPHERE program funded by the MSC COFUND Action. method to assess behavioral recovery after mouse spinal cord injury. Neurosci Res. 2010;67(2):117–25. Availability of data and materials 19. Randall LO, Selitto JJ. A method for measurement of analgesic activity on The datasets used and/or analyzed during the current study are available inflamed tissue. Arch Int Pharmacodyn Ther. 1957;111(4):409–19. from the corresponding author on reasonable request. 20. Hargreaves K, Dubner R, Brown F, Flores C, Joris J. A new and sensitive method for measuring thermal nociception in cutaneous hyperalgesia. Pain. 1988;32(1):77–88. Ethics approval 21. Rudman MD, Choi JS, Lee HE, Tan SK, Ayad NG, Lee JK. Bromodomain and We have the consent of the Comitè Ètic d'Experimentació Animal (CEEA) extraterminal domain-containing protein inhibition attenuates acute from the Universitat Autonoma de Barcelona approved the protocol CEAAH inflammation after spinal cord injury. Exp Neurol. 2018;309:181–92. 4053_4057 for the use of animals to repair and regenerate lesions after 22. DeMars KM, Yang C, Castro-Rivera CI, Candelario-Jalil E. Selective spinal cord injury. degradation of BET proteins with dBET1, a proteolysis-targeting chimera, potently reduces pro-inflammatory responses in Consent for publication lipopolysaccharide-activated microglia. Biochem Biophys Res Commun. Not applicable 2018;497(1):410–5. 23. Hajmirza A, Emadali A, Gauthier A, Casasnovas O, Gressin R, Callanan MB. Competing interests BET family protein BRD4: an emerging actor in NFkappaB signaling in The authors declare that they have no competing interests. inflammation and cancer. Biomedicines. 2018;6(1). Sánchez-Ventura et al. Journal of Neuroinflammation (2019) 16:124 Page 12 of 12 24. Conrad RJ, Fozouni P, Thomas S, Sy H, Zhang Q, Zhou MM, et al. The short 46. Lv B, Li J, Li M, Zhuo Y, Ren K, Li E, et al. Enhancement of adenovirus isoform of BRD4 promotes HIV-1 latency by engaging repressive SWI/SNF infection and adenoviral vector-mediated gene delivery by bromodomain chromatin-remodeling complexes. Mol Cell. 2017;67(6):1001–12 e6. inhibitor JQ1. Sci Rep. 2018;8(1):11554. 25. Shen C, Ipsaro JJ, Shi J, Milazzo JP, Wang E, Roe JS, et al. NSD3-short is an 47. Banerjee C, Archin N, Michaels D, Belkina AC, Denis GV, Bradner J, et al. BET adaptor protein that couples BRD4 to the CHD8 chromatin remodeler. Mol bromodomain inhibition as a novel strategy for reactivation of HIV-1. J Cell. 2015;60(6):847–59. Leukoc Biol. 2012;92(6):1147–54. 26. Rahman S, Sowa ME, Ottinger M, Smith JA, Shi Y, Harper JW, et al. The Brd4 48. Andrieu G, Belkina AC, Denis GV. Clinical trials for BET inhibitors run ahead extraterminal domain confers transcription activation independent of pTEFb of the science. Drug Discov Today Technol. 2016;19:45–50. by recruiting multiple proteins, including NSD3. Mol Cell Biol. 2011;31(13): 2641–52. 27. Bethea JR, Nagashima H, Acosta MC, Briceno C, Gomez F, Marcillo AE, et al. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in Systemically administered interleukin-10 reduces tumor necrosis factor- published maps and institutional affiliations. alpha production and significantly improves functional recovery following traumatic spinal cord injury in rats. J Neurotrauma. 1999;16(10):851–63. 28. Francos-Quijorna I, Amo-Aparicio J, Martinez-Muriana A, Lopez-Vales R. IL-4 drives microglia and macrophages toward a phenotype conducive for tissue repair and functional recovery after spinal cord injury. Glia. 2016;64(12):2079–92. 29. Thompson CD, Zurko JC, Hanna BF, Hellenbrand DJ, Hanna A. The therapeutic role of interleukin-10 after spinal cord injury. J Neurotrauma. 2013;30(15):1311–24. 30. Genovese T, Mazzon E, Crisafulli C, Di Paola R, Muia C, Esposito E, et al. TNF- alpha blockage in a mouse model of SCI: evidence for improved outcome. Shock. 2008;29(1):32–41. 31. Atkins S, Loescher AR, Boissonade FM, Smith KG, Occleston N, O'Kane S, et al. Interleukin-10 reduces scarring and enhances regeneration at a site of sciatic nerve repair. J Peripher Nerv Syst. 2007;12(4):269–76. 32. Vidal PM, Lemmens E, Dooley D, Hendrix S. The role of “anti-inflammatory” cytokines in axon regeneration. Cytokine Growth Factor Rev. 2013;24(1):1–12. 33. Dooley D, Lemmens E, Vangansewinkel T, Le Blon D, Hoornaert C, Ponsaerts P, et al. Cell-based delivery of interleukin-13 directs alternative activation of macrophages resulting in improved functional outcome after spinal cord injury. Stem Cell Reports. 2016;7(6):1099–115. 34. Guglielmetti C, Le Blon D, Santermans E, Salas-Perdomo A, Daans J, De Vocht N, et al. Interleukin-13 immune gene therapy prevents CNS inflammation and demyelination via alternative activation of microglia and macrophages. Glia. 2016;64(12):2181–200. 35. Korb E, Herre M, Zucker-Scharff I, Darnell RB, Allis CD. BET protein Brd4 activates transcription in neurons and BET inhibitor Jq1 blocks memory in mice. Nat Neurosci. 2015;18(10):1464–73. 36. Penas C, Navarro X. Epigenetic modifications associated to neuroinflammation and neuropathic pain after neural trauma. Front Cell Neurosci. 2018;12:158. 37. Ji RR, Xu ZZ, Gao YJ. Emerging targets in neuroinflammation-driven chronic pain. Nat Rev Drug Discov. 2014;13(7):533–48. 38. Pastori C, Kapranov P, Penas C, Peschansky V, Volmar CH, Sarkaria JN, et al. The Bromodomain protein BRD4 controls HOTAIR, a long noncoding RNA essential for glioblastoma proliferation. Proc Natl Acad Sci U S A. 2015; 112(27):8326–31. 39. Pastori C, Daniel M, Penas C, Volmar CH, Johnstone AL, Brothers SP, et al. BET bromodomain proteins are required for glioblastoma cell proliferation. Epigenetics. 2014;9(4):611–20. 40. Fu LL, Tian M, Li X, Li JJ, Huang J, Ouyang L, et al. Inhibition of BET bromodomains as a therapeutic strategy for cancer drug discovery. Oncotarget. 2015;6(8):5501–16. 41. Benito E, Ramachandran B, Schroeder H, Schmidt G, Urbanke H, Burkhardt S, et al. The BET/BRD inhibitor JQ1 improves brain plasticity in WT and APP mice. Transl Psychiatry. 2017;7(9):e1239. 42. Li J, Zhao L, Urabe G, Fu Y, Guo LW. Epigenetic intervention with a BET inhibitor ameliorates acute retinal ganglion cell death in mice. Mol Vis. 2017;23:149–59. 43. Jiang Y, Zhu L, Zhang T, Lu H, Wang C, Xue B, et al. BRD4 has dual effects on the HMGB1 and NF-kappaB signalling pathways and is a potential therapeutic target for osteoarthritis. Biochim Biophys Acta Mol Basis Dis. 2017;1863(12):3001–15. 44. Magistri M, Velmeshev D, Makhmutova M, Patel P, Sartor GC, Volmar CH, et al. The BET-bromodomain inhibitor JQ1 reduces inflammation and tau phosphorylation at Ser396 in the brain of the 3xTg model of Alzheimer’s disease. Curr Alzheimer Res. 2016;13(9):985–95. 45. Lee DU, Katavolos P, Palanisamy G, Katewa A, Sioson C, Corpuz J, et al. Nonselective inhibition of the epigenetic transcriptional regulator BET induces marked lymphoid and hematopoietic toxicity in mice. Toxicol Appl Pharmacol. 2016;300:47–54.