C/EBPδ protects from radiation-induced intestinal injury and sepsis by suppression of inflammatory and nitrosative stress

Ionizing radiation (IR)-induced intestinal damage is characterized by a loss of intestinal crypt cells, intestinal barrier disruption and translocation of intestinal microflora resulting in sepsis-mediated lethality. We have shown that mice lacking C/EBPδ display IR-induced intestinal and hematopoietic injury and lethality. The purpose of this study was to investigate whether increased IR-induced inflammatory, oxidative and nitrosative stress promote intestinal injury and sepsis-mediated lethality in Cebpd−/− mice. We found that irradiated Cebpd−/− mice show decreased villous height, crypt depth, crypt to villi ratio and expression of the proliferation marker, proliferating cell nuclear antigen, indicative of intestinal injury. Cebpd−/− mice show increased expression of the pro-inflammatory cytokines (Il-6, Tnf-α) and chemokines (Cxcl1, Mcp-1, Mif-1α) and Nos2 in the intestinal tissues compared to Cebpd+/+ mice after exposure to TBI. Cebpd−/− mice show decreased GSH/GSSG ratio, increased S-nitrosoglutathione and 3-nitrotyrosine in the intestine indicative of basal oxidative and nitrosative stress, which was exacerbated by IR. Irradiated Cebpd-deficient mice showed upregulation of Claudin-2 that correlated with increased intestinal permeability, presence of plasma endotoxin and bacterial translocation to the liver. Overall these results uncover a novel role for C/EBPδ in protection against IR-induced intestinal injury by suppressing inflammation and nitrosative stress and underlying sepsis-induced lethality.

Cxcl1 {Chemokine (C-X-C motif) ligand 1} also known as KC is another chemokine expressed by macrophages, neutrophils and epithelial cells and has neutrophil chemoattractant activity 35 . There were no significant differences in the intestinal expression of Cxcl1 between both the genotypes in unirradiated mice. At day 7 post-irradiation. Cebpd −/− mice showed a 7-fold increase, while Cebpd +/+ mice showed a 2-fold increase in the expression of Cxcl1 in the intestine tissues (Fig. 4B).
Macrophage migration inhibitory factor (Mif1-α), a pro-inflammatory cytokine which is upregulated by IR and oxidative stress 36 . MIF also has a chemokine-like function and promotes the directed migration and recruitment of leukocytes into infectious and inflammatory sites 37,38 . Unirradiated Cebpd −/− mice showed slightly higher expression of Mif-1α, however this difference was not significant when compared with unirradiated Cebpd +/+ mice. Interestingly at day 7 post-irradiation, Cebpd +/+ mice showed a 2-fold induction, while Cebpd −/− mice showed a robust upregulation of Mif-1α by almost 10-fold (Fig. 4C).
These results indicate that the elevated expression of chemokines may promote the increased recruitment of neutrophils and macrophages, in the intestine tissues of Cebpd −/− mice compared to that of Cebpd +/+ mice after exposure to IR.

Cebpd-deficiency promotes increased nitrosative stress in the intestine prior to and post-irradiation.
iNOS (inducible nitric oxide synthase) or Nos2 is involved in immune response and is known to be significantly induced by exposure of cells/tissues to IR [39][40][41][42] . Cebpd −/− mice showed robust induction of Nos2 expression in intestine by 10-fold, which is significant compared to respective Cebpd +/+ mice at 1 h post-irradiation. Both Cebpd +/+ and Cebpd −/− showed elevated expression of Nos2, however it reduced about 25-and 22-fold at 4 h and to 2.4-fold and 1.6-fold at 1 day post-irradiation respectively. At day 3.5 post-irradiation, the Nos2 transcript was induced by 2.6-fold Cebpd −/− mice compared to 3-fold in Cebpd +/+ mice and was not significant (Fig. 5A).
This post-irradiation induction of Nos2 results in increased nitric oxide levels, which react with the oxygen free radicals produced in the cells as a consequence of radiation-induced oxidative stress to form peroxynitrite, which causes nitrosylation of the cellular proteins 43,44 . The intestine tissues of Cebpd +/+ and Cebpd −/− mice were compared for the expression levels of 3-nitrotyrosine (3-NT) by HPLC. In the sham group, the intestinal tissue extracts of Cebpd −/− mice showed a 1.7-fold increase compared to Cebpd +/+ mice. Exposure to IR further showed significant increase in the 3-NT expression in the intestinal tissues of Cebpd −/− mice compared to Cebpd +/+ mice and with respect to the sham controls (Fig. 5B).
Cebpd-deficiency results in basal and IR-induced oxidative stress. Glutathione (GSH) is the global antioxidant which plays a critical role in maintaining the redox state of cells and detoxification of IR-induced oxidative stress 45,46 . We therefore examined the expression of GSH and its oxidized dimer glutathione disulfide (GSSG) in intestine tissues. The levels of GSH and GSSG were decreased by IR and were not significantly different between both genotypes ( Supplementary Fig. 1). A decrease in the ratio of GSH/GSSG is considered a measure of oxidative stress. Interestingly, we found that in the sham group, Cebpd −/− mice displayed a significant decrease in the GSH/ GSSG compared to Cebpd +/+ mice (Fig. 6A). At days 1 and 3.5 post-irradiation, although there was a decrease in the GSH/GSSG ratio in both the genotypes, however these differences were not significant. In addition, GSH also acts a scavenger of nitric oxide to form S-nitrosoglutathione (GSNO) 47,48 . We examined in both Cebpd +/+ and Cebpd −/− intestinal tissues and found that the GSNO levels were significantly elevated at basal levels as well as post-irradiation in Cebpd −/− mice compared to Cebpd +/+ mice (Fig. 6B). in sham unirradiated controls and at day 1 post-irradiation. However at day 3.5 post-irradiation, Cldn2 mRNA and protein levels were upregulated by 2-fold in the intestine tissues of Cebpd −/− mice compared to that of Cebpd +/+ mice (Fig. 7A,C). Some of the other genes such as Cldn4, Cldn11, Ocldn and Zo-1 did not any significant differences between both the genotypes either in the sham or at day 1 and 3.5 post-irradiation ( Supplementary  Fig. 2). Further immunofluorescence staining of intestine sections revealed that there were no significant differences in the expression of Claudin-2 in the sham or at day 1 post-irradiation in both the genotypes. The Claudin-2 expression was localized on the basal surface of the epithelial cells in the sham group and at day 1 post-irradiation in both Cebpd −/− and Cebpd +/+ mice. At day 3.5 post-irradiation, the localization of Claudin-2 was found in the tight junctions between the epithelial cells as well on the luminal surface in Cebpd −/− mice. In contrast the Cebpd +/+ mice showed the Claudin-2 expression on the basal surface of the epithelial cells (Fig. 7B).

Cebpd-deficient mice show increased iR-induced in vivo intestinal permeability, increased endotoxemia and bacterial translocation.
As a functional consequence of Claudin-2 upregulation, next we determined whether irradiated Cebpd −/− mice showed alterations in intestinal permeability. Cebpd −/− mice showed a 2-fold increase in FITC-dextran levels in plasma compared to Cebpd +/+ mice at day 3.5 post-irradiation indicative of increased intestinal permeability (Fig. 8A). We further confirmed whether increased intestinal permeability led to a significant increase in the lipopolysaccharide-binding protein (LBP) in Cebpd −/− mice after exposure to IR. There were no differences between both genotypes in the sham controls. However, at days 3.5 and 7 post-irradiation, compared to Cebpd +/+ mice, there was a 5-fold and a 3.7-fold increase in the plasma levels of LBP in Cebpd −/− mice, indicative of bacteria in systemic circulation (Fig. 8B). These results correlated with a  www.nature.com/scientificreports www.nature.com/scientificreports/ 2-fold increase in the amplification of 16S rRNA in liver tissues at day 3.5 post-irradiation in Cebpd −/− mice compared to that of Cebpd +/+ mice, indicating translocation of bacteria from the intestine (Fig. 8C). All these results demonstrate that the increased leakiness of the gut in the Cebpd −/− mice at day 3.5 post-irradiation leads to the onset of sepsis-like sequelae.

Discussion
Exposure to the whole or substantial parts of the body to IR often result in life-threatening injuries, primarily to self-renewing tissues such as the hematopoietic and GI 1,3,6,7 . The main cause of lethality after exposure to IR is due to the intestinal bacteria that penetrate the defective mucosal barrier and are an important source of bacteremia. An increase in mucosal permeability occurs through a combination of disruption of epithelial tight junctions and insufficient replacement of the villus epithelium, due to cell death of intestinal progenitor cells in the crypts. Enhanced intestinal permeability, leading to bacterial and lipopolysaccharide translocation are characteristics of IR-induced multiple organ dysfunction syndrome 16,49 .
Previously, we reported that the increased lethality to IR exposure observed in Cebpd −/− mice occurred due to increased thrombocytopenia, neutropenia and loss of intestinal crypts 32 . These very same processes are implicated as key hallmarks of sepsis 16,49 . In this study we show that Cebpd −/− mice show increased damage to the intestinal villi and crypts at day 7 post 8.5 Gy. Further the loss of intestinal crypts at a dose of 10 Gy suggests a direct role for C/EBPδ in protection of intestinal crypt epithelial cells.
The IR-induced inflammatory response is initiated by the production of ROS/RNS that promote the induction of apoptosis and clonogenic cell death, activation of the transcription of several pro-inflammatory cytokines, chemokines, and growth factors in the microvascular and mucosal compartments, by the recruited immune cells and by enterocytes and residing cells, depending on the severity of tissue trauma 10,15,50,51 . The production of cytokines such as Il-6 and Tnf-α is time-dependent usually peaking between 4-24 h post-irradiation with subsequent decrease to basal levels with 24 h to few days 50 . It is now realized that sepsis is associated with an "inflammatory storm", which results in multi-organ damage/failure [16][17][18] . Mcp-1 is one of the key chemokines that regulates migration and infiltration of monocytes/macrophages 34 , while Cxcl1 is implicated in recruiting neutrophils that are frequently the first immune cells to enter an inflamed or infected tissue 35 . Mif-1α is an integral component of host inflammatory responses and is known to be induced by IR and is positively associated with sepsis [36][37][38] . In the present study, we found that intestines of Cebpd −/− mice showed rapid upregulation of the pro-inflammatory cytokines, Il-6 and Tnf-α at early timepoints post-irradiation. In contrast, the expression of chemokines such as Mcp-1, Cxcl1 and Mif1-α were upregulated by days 3.5 and 7 post-irradiation compared to Cebpd +/+ mice.
Inducible nitric oxide synthase (iNOS) is expressed by infiltrating as well as resident activated macrophages in inflamed gastrointestinal tissue and is also stimulated by pro-inflammatory cytokines like Il-6 and Tnf-α as well www.nature.com/scientificreports www.nature.com/scientificreports/ as by IR 42 . In response to IR exposure, the increased nitric oxide reacts with superoxide formed by the mitochondria to form peroxynitrite (ONOO − ) which is a proxidant. Peroxynitrite reacts with the tyrosine residues in the cellular proteins and forms 3-NT 43,44 . Increased 3-NT causes increased radiation-induced intestinal toxicity and blocking or reducing 3-NT protects against radiation injury. In the present study, Cebpd −/− mice show elevated levels of 3-NT at basal as well as post-irradiation compared to Cebpd +/+ mice.
GSH is the first line of defense for oxidative stress 45 . Decreased levels of GSH indicate decrease capacity to remove free radicals. The ratio of GSH/GSSG is a measure of redox state of the cell or tissue 46 . Interestingly, the expression of GSH/GSSG in the intestine tissues of Cebpd −/− mice were significantly lower than that of Cebpd +/+ mice indicative of increased oxidative stress in the sham controls.
The elevated levels of NO produced are scavenged by the cellular antioxidant, GSH, resulting in the formation of GSNO 47,48 . GSNO is the nitrosylated form of reduced glutathione (GSH), responsible for its antioxidant cytoprotective action 47 . Although GSNO is shown to have a protective function in maintaining the epithelial barrier function 48 , we found that Cebpd −/− mice expressed significantly elevated GSNO levels in the sham as well irradiated group. At very high concentrations, GSNO is converted to GSSG and NO and the released NO can react with superoxide present in the cell and generate increased peroxynitrite as described previously 43 . This could be a plausible mechanism for the increased 3-NT levels, as we have previously shown evidence for basal as well as IR-induced oxidative stress and mitochondrial dysfunction via reduced expression of GSH levels in Cebpd-deficient MEFs 33 . The increased mRNA levels of iNOS and increased expression of 3-NT and GSNO are indicative of overall increased nitrosative stress in the intestines of Cebpd −/− mice compared to that of Cebpd +/+ mice.
The epithelial tight junctions form a barrier to the entry of allergens, toxins and pathogens across the epithelium into the interstitial tissue. The tight junction proteins, adherens junction and desmosomes seal the intercellular junctions of intestinal epithelial cells 52,53 . While the tight junctions function as a barrier from noxious molecules and a pore for the permeation of ions, solutes and water, the adherence junctions and desmosomes provide a strong adhesive bond between cells and in intercellular communication. One of the mechanisms via which inflammation promotes intestinal permeability is via the downregulation of tight junction proteins such as Occludin, junctional adhesion molecules (JAM), ZO-1, Claudins etc. [54][55][56][57][58][59][60] . In addition, ROS and RNS can promote the disruption of tight junctions and promote intestinal permeability 56,57 . Studies have shown that radiation injury led to downregulation of the tight junction proteins in the intestinal mucosa 54,55 . The upregulation of Claudin2 by IL-6 as well as TNF-α are known to cause an increase in intestinal permeability 60,61 . Similarly in this study, we found significantly elevated expression of Il-6 and Tnf-α, which correlated with the upregulation of Cldn2 in Cebpd −/− mice at day 3.5 post-irradiation. Further studies in other inflammatory intestinal disorders have reported altered localization of Claudin 2 similar to that observed in irradiated Cebpd −/− mice 62,63 . These results correlated with the increased in vivo intestinal permeability and elevated plasma LBP levels as well as bacterial translocation observed in irradiated Cebpd −/− mice. Cebpd −/− mice show increased inflammatory, oxidative and nitrosative stress that leads to increased intestinal permeability and bacterial translocation at day 3.5 post-irradiation, thus confirming that the increased mortality to radiation occurs due to underlying sepsis.
The present study uncovers a novel role for C/EBPδ in protection from IR-induced gut injury and underlying sepsis-mediated lethality by downregulating the IR-induced oxidative/nitrosative stress and inflammatory responses (Fig. 9). Further studies are warranted to elucidate whether blocking the IR-induced inflammation and oxidative/nitrosative stress can alleviate the lethality of Cebpd-deficient mice to IR. Overall these results may have relevance in terms of human exposure to IR either in accidental exposure or in the clinical setting. Recently we have described that C/EBPδ is essential to mediate the radioprotective functions of the potent radioprotector gamma-tocotrienol (GT3) 64 . We found that GT3 induces the C/EBPδ expression in the intestine and helps protect the Cebpd-WT mice, but was unable to impart protection to Cebpd-KO mice from radiation induced injury to the intestinal and hematopoietic systems. Therefore it can be speculated that agents that induce C/EBPδ expression Animals. Cebpd-heterozygous breeder mice were backcrossed for more than 20 generations to the C57BL/6 strain background. Genotyping was done as described previously 65 . In all the studies, 10-12 weeks old subjects derived from heterozygous mating pairs and litter mate controls were used whenever possible. The animals were housed in the Division of Laboratory Medicine (DLAM, University of Arkansas for Medical Sciences, Little Rock, AR) under standardized conditions with controlled temperature and humidity and a 12 h day, 12 h night light cycle. Blood, intestine and liver tissues were harvested following isoflurane inhalation to minimize suffering and the animals were euthanized by cervical dislocation. irradiation of mice. Irradiation was administered in a Mark I irradiator (J. L. Shepherd). Dose uniformity was assessed by an independent company (Ashland Specialty Ingredients) with radiographic film and alanine tablets. Alanine tables were analyzed by the National Institute of Standards and Technology (Gaithersburg, MD, USA) and demonstrated a dose rate of 1.14 Gy/min at 21 cm from the source. For each experiment the dose rate was corrected for decay.

Assessment of villus height and crypt depth. Intestinal tissue sections stained with hematoxylin
and eosin (H&E) were used to measure villus height and crypt depth using a computer-assisted image analysis platform (Imagescope ver 12.2.2.5, Leica Biosystems, e, MD, USA). Intestinal tissues fixed in Methyl-Carnoy's fixed and embedded in paraffin were cut into 2-4 μm sections with a microtome. The slides with tissue sections were de-waxed by placing in an incubator overnight set at 60 °C, cooled down to room temperature, dipped into hematoxylin solution for 30 s, washed with deionized water, stained with 1% eosin solution, dehydrated with two changes in 95% and 100% alcohol for 30 s each, washed with xylene, and finally mounted with low viscosity Permount ™ mounting media (Thermo Fisher Scientific, Grand Island, NY, USA).
Mucosal villus height was measured from the tip to the base of each villus, and crypt depth was measured from the crypt base to the top opening. Images captured at 4x magnification were analyzed for crypt to villus measurements. 20 villi and crypts were respectively analyzed for villous height and crypt depth measurements on images at 20x magnification. www.nature.com/scientificreports www.nature.com/scientificreports/ paraffin in cross-sectional orientation. For staining, sections of 5-µm thickness were cut, dewaxed, rehydrated in PBS (10 mM sodium phosphate, pH 7.4; 140 mM NaCl).
Immunofluorescence staining for Claudin-2. The intestine tissues were fixed with methanol carnoy's fixative (methanol: chloroform: acetic acid, 6:3:1) for 24 h, dehydrated, were embedded in paraffin in cross-sectional orientation. Sections were stained with rabbit anti-Claudin-2 (ab53032) (Abcam) at 1:50 dilution in blocking buffer (0.5% BSA, 0.05% Tween-20, PBS). Each section was probed and goat anti-rabbit IgG-AlexaFluor 594 (Invitrogen) at 1:400 dilution for 1 h at 37 °C and then rinsed three times with 0.05% Tween-20 in PBS. After staining, sections were counterstained with 4′,6-diamidino-2-phenylindol to visualize cell nuclei and were mounted under cover slips with Prolong Antifade kit (Invitrogen). Images for Claudin-2 and DAPI staining were acquired with an Olympus IX-51 inverted microscope (Olympus America) with a 10× objective, equipped with Hamamatsu ORCA-ER monochrome camera (Hamamatsu Photonics K.K.). For claudin-2 staining, a total of 5 fields of view per tissue section (10× objective), and the mean intensity was measured in each compartment. The average mean intensity of the field of view per tissue section was considered as a datapoint for statistical analysis. The data are presented as mean fluorescence intensity per field in n = 4 animals per timepoint. Slidebook 4.2 software (Intelligent Imaging Innovations, Inc.) was used for image capturing and analysis.
Real-time pcR. Total  HpLc assays. High-performance liquid chromatography (HPLC) was used to quantify the reduced as well as oxidized glutathione (GSH, GSSG), S-nitrosoglutathione (GSNO) and 3-nitrotyrosine (3-NT). Approximately 20 mg of intestine tissue were homogenized in ice-cold phosphate-buffered saline. To precipitate proteins, 10% metaphosphoric acid was added to the homogenate and incubated for 30 min on ice. The samples were then centrifuged at 18,000 g at 4 °C for 15 min, and 20 µl of the resulting supernatants were injected into the HPLC column for metabolite quantification, while the pellet was used for protein analysis using BCA protein assay. The details for HPLC elution and electrochemical detection of free unbound GSH, GSSG, GSNO and 3-NT in proteins (hydrolyzed by 6 N HCl treatment) have been described previously described 66 . In vivo intestinal permeability assay. Intestinal permeability was measured in 20 mice (5 per group) as described by Biju et al. 67 . Briefly, 4 days after exposure to 0 and 8.5 Gy irradiation, the mice were anesthetized with isoflurane inhalation, a midline laparotomy was performed, and the renal artery and vein were ligated bilaterally. A 10 cm small intestinal segment, located 5 cm distal to the ligament of Treitz was isolated and tied off. One hundred microliters of 4-kDa fluorescein isothiocyanate conjugated dextran (FITC-dextran 25 mg/ml in phosphate-buffered saline) was injected into the isolated intestine using a 30 Gauge needle and the abdominal incision was closed. After 90 min, blood was collected from the retro-orbital sinus and plasma was separated by centrifuging at 4 °C, 8000 rpm for 10 min. The concentration of FITC-dextran was determined with a fluorescence spectrophotometer SpectraMax M2 e (Molecular Devices, CA, USA) at an excitation wavelength of 480 nm and an emission wavelength of 520 nm. Standard curves were prepared from dilutions of FITC-dextran in PBS to calculate FITC-dextran concentration in the plasma samples. eLiSA assay. Plasma LBP level was measured using the mouse LBP ELISA kit (HK205) from Hycult Biotech, Uden, Netherlands. Briefly, blood was collected in EDTA coated tubes, centrifuged at 2000 rpm at 4 °C for 15 min and plasma samples were snap frozen and stored at −80 °C. One hundred μl of standard and plasma samples were loaded onto pre-coated 96-well plates, incubated for 2 h at room temperature. Biotinylated tracer antibody was incubated for 1 h at room temperature, developed against a streptavidin-peroxidase conjugate, and absorbance was measured at 450 nm. The concentration of LBP was determined against the standard curve. Values are expressed as nanogram LBP protein per ml plasma.
Bacterial translocation. Bacterial translocation was determined as bacterial load in liver tissue and was quantified by real time PCR using the 16S rRNA gene consensus sequence. The total load of bacteria in the liver was determined using primer sequences to amplify the highly conserved sequence for a broad species consensus as reported elsewhere 67 . Livers were removed aseptically and homogenized immediately. Bacterial translocation was quantified by real-time PCR. Briefly, DNA was isolated from sterile livers harvested at baseline and at day 3.5 post-exposure to 8.5 Gy using a DNA purification kit (Promega, Madison, WI). Real-time PCR was performed