CCAR2 negatively regulates IL-8 production in cervical cancer cells

Cell cycle and apoptosis regulator 2 (CCAR2) is a multifaceted protein that controls diverse cellular functions; however, its function in cancer is unclear. To better understand its potential role in cancer, we examined gene expression patterns regulated by CCAR2 in cervical cancer cells. Cytokine and chemokine production by CCAR2-deficient cells increased under oxidative conditions. In particular, H2O2-treated CCAR2-depleted cells showed a significant increase in interleukin-8 (IL-8) production, indicating a negative regulation of IL-8 by CCAR2. Upregulation of IL-8 expression in CCAR2-deficient cells occurred via activation of transcription factor AP-1. The negative correlation between CCAR2 and IL-8 expression was confirmed by examining mRNA and protein levels in tissues from cervical cancer patients. Furthermore, CCAR2-regulated IL-8 expression is associated with a shorter survival of cervical cancer patients. Overall, the data suggest that CCAR2 plays a critical role in controlling both the cancer secretome and cancer progression.


INTRODUCTION
Cell cycle and apoptosis regulator 2 (CCAR2, formerly known as DBC1 [deleted in breast cancer 1]) is a multifaceted protein that regulates a diverse subset of cellular functions. CCAR2 controls transcription, mRNA splicing, DNA damage responses, circadian rhythm, inflammation, metabolism, differentiation, proliferation, survival, and apoptosis [1][2][3]. With respect to cancer growth, CCAR2 is thought to act as a tumor promoter or tumor suppressor depending on the context. The role of CCAR2 as a tumor promoter is supported by findings that it promotes proliferation [4][5][6][7][8][9][10][11][12], migration, and invasion of cancer cells [6,8,9,13] and suppresses cancer cell anoikis [14]. Depleting CCAR2 inhibits the growth of tumor xenografts derived from colon cancer [8] and laryngeal squamous cell carcinoma cells [5]. By contrast, CCAR2 functions as a tumor suppressor by stabilizing p53 through competitive inhibition of Mdm2 binding [15]. CCAR2 knockout mice spontaneously develop tumors of the liver and lung, teratomas, and lymphomas [15]. CCAR2 knockout mouse embryonic fibroblasts proliferate faster than wildtype cells [15], which is the opposite phenotype to that observed in cancer cell lines. Therefore, although the exact role of CCAR2 as a tumor promoter or tumor suppressor in humans is unclear, CCAR2 plays an intimate part in tumor development and progression.
Recently, the tumor microenvironment has emerged as a critical niche that regulates tumor progression, angiogenesis, and metastasis [16]. Beyond tumor cells, the microenvironment comprises diverse types of cells, secreted soluble factors (e.g., cytokines and growth factors), and the extracellular matrix [17]. Within this heterogeneous tumor microenvironment, tumor cells, endothelial cells, cancer-associated fibroblasts, and infiltrating inflammatory cells produce many different cytokines that control local networks [18]. Cytokines
Although the link between CCAR2 and tumor cells has been studied, the role of CCAR2 in the tumor microenvironment remains unclear. In addition, although the previous studies show that CCAR2 regulates tumor cell proliferation, migration, and invasion, the mediators of CCAR2 functions have not been identified. Here, we show that CCAR2 modulates expression and secretion of IL-8 by tumor cells under oxidative conditions. Thus, CCAR2 is a critical factor that fine-tunes the tumor microenvironment.

CCAR2 regulates the expression of cytokine and chemokine following oxidative stress
The role of CCAR2 in the oxidative tumor microenvironment is unclear. First, to explore whether CCAR2 controls responses to oxidative stress, we used a genome-wide microarray to identify transcriptional target genes coregulated by CCAR2 and hydrogen peroxide (H 2 O 2 ) ( Figure 1A). Differentially (2-fold change (FC) up or down) regulated genes corresponding to individual probes were counted and compared ( Figure  1B). Following H 2 O 2 treatment, 78 targets in CCAR2 siRNA-transfected (siCCAR2) cells showed differential expression when compared with non-transfected control cells (27 upregulated and 51 downregulated) ( Figure  1B). Among them, seven genes corresponding to eight probes showed oxidative stress-dependent transcriptional activation: IL1A, NEU4, FOS, RPRM, NR4A2, OASL, and IL8 ( Figure 1C). The 78 differentially regulated targets in H 2 O 2 -treated siCCAR2 cells were classified according to biological process using Gene Ontology (GO) analysis to identify the molecular functions most sensitive to CCAR2 levels in response to H 2 O 2 . Genes involved in inflammatory responses were significantly affected in H 2 O 2 -treated siCCAR2 cells ( Figure 1D). Next, we performed Gene Set Enrichment Analysis (GSEA) to further characterize differences between control cells and siCCAR2 cells in response to H 2 O 2 . GSEA revealed that cytokine and chemokine activity gene sets were enriched significantly among genes upregulated in H 2 O 2 -treated siCCAR2 cells (Figure 2A). This suggests that CCAR2 regulates production of cytokines and chemokines under oxidative conditions. Indeed, expression of interleukins and chemokine ligands including IL-8, IL-1, IL-24, IL-6, CC chemokines, and CXC chemokines increased significantly ( Figure 2B). In particular, expression of IL-8 showed the most significant fold change in H 2 O 2 -treated siCCAR2 cells compared with control cells ( Figure 2C). CCAR2 expression was not significantly altered by H 2 O 2 treatment ( Figure 2C). Overall, these data suggest that CCAR2 plays a key role in regulating H 2 O 2 -responsive genes.

CCAR2 regulates IL-8 production
Among cytokines and chemokines regulated by CCAR2 following oxidative stress, expression of IL-8 mRNA showed the most significant change in the microarray. Significant upregulation of IL-8 expression by H 2 O 2 -treated siCCAR2 cells was validated by RT-PCR ( Figure 3A). To confirm IL-8 secretion into the extracellular space, we measured the amount of IL-8 in culture medium. IL-8 production by H 2 O 2 -treated cells was higher than that by non-treated control cells. Production of IL-8 protein was significantly induced in siCCAR2-HeLa and siCCAR2-SiHa cells than that in siUniversaltransfected cells following H 2 O 2 treatment ( Figure 3B and 3C). This demonstrates that CCAR2 deficiency potentiates IL-8 upregulation following oxidative stress. In addition, the data suggest that CCAR2 deficiency fosters a proinflammatory tumor microenvironment.

The upregulation of IL-8 is mainly mediated by AP-1 transcription factor
To investigate the mechanism by which CCAR2 regulates IL-8 transcription, cells were transfected with luciferase reporter constructs containing the IL-8 promoter harboring binding sites for various transcription factors (AP-1, NF-IL-6, and NF-kB) ( Figure 4A). Both before and after H 2 O 2 treatment, the full length ( -143/ + 44) promoter-containing reporter generated higher luciferase activity than the shorter promoter-containing reporters in both siUniversal and siCCAR2 cells ( Figure 4B). In addition, siCCAR2 cells showed the highest luciferase activity after H 2 O 2 treatment ( Figure 4B). Furthermore, the increase in luciferase activity in siCCAR2 relative to that in siUniversal cells was significant only in cells harboring the full length reporter after H 2 O 2 treatment ( Figure 4B). www.impactjournals.com/oncotarget This implies that the AP-1 transcription factor mediates upregulation of IL-8 in CCAR2-deficient cells under oxidative stress. AP-1 is a dimeric transcription factor comprising Jun, Fos, ATF, Maf, and JDP family members [30]. Among these, we measured activation of c-Jun by measuring its phosphorylation. siCCAR2 cells showed increased phosphorylation of c-Jun in response to H 2 O 2 ( Figure 4C). This demonstrates that AP-1 activation is required for IL-8 upregulation in CCAR2-deficient cells.

The expression of CCAR2 and IL-8 is negatively correlated
The above data demonstrate that CCAR2 is one of the factors that regulate IL-8 expression in vitro. Next, CCAR2-dependent IL-8 expression was verified in cervical tumor tissues isolated from cervical cancer patients (Table  1). Immunohistochemistry (IHC) revealed that, while CCAR2 expression was confined to the nucleus, that of IL-8 was evident mainly in the cytoplasm, or focally in the nucleus. The association between CCAR2 and IL-8 showed a significant negative correlation ( Figure 5A). To confirm this negative correlation, we examined mRNA expression using data derived from 82 and 304 cervical cancer patients downloaded from Gene Expression Omnibus (GEO) ( Figure 5B) and The Cancer Genome Atlas (TCGA) ( Figure 5C), respectively. Expression of CCAR2 showed a weak (-0.40 < r <-0.20) ( Figure 5B) and very weak (-0.20 < r < 0.00) ( Figure 5C) negative relationship, respectively, with IL-8. These negative relationships were statistically significant (P < 0.05). This suggests that CCAR2 is a negative regulator of IL-8 expression.

High expression of IL-8 is associated with low overall survival of cervical cancer patients
Next, we investigated the association between expression of CCAR2 and IL-8 and overall survival. While the level of CCAR2 expression had no effect on overall survival, high expression of IL-8 is associated with shorter survival of cervical cancer patients ( Figure 6A and 6B). Thus, it suggests that CCAR2 itself does not affect the survival of cancer patients, but CCAR2-dependent IL-8 expression does. Next, we divided the patients into four groups based on the expression status of each gene: (i) low CCAR2/low IL-8, (ii) high CCAR2/low IL-8, (iii) low CCAR2/high IL-8, and (iv) high CCAR2/high IL-8. As expected, overall survival of the low CCAR2/high IL-8 group was shorter than that of the low CCAR2/low IL-8 and high CCAR2/low IL-8 groups ( Figure 6C), suggesting that low CCAR2 expression is associated with high IL-8 expression and contributes to create aggressive tumor microenvironment. However, unexpectedly, we found that the overall survival of the high CCAR2/high IL-8 group was shorter than that of the low CCAR2/high IL-8 group. The high CCAR2/high IL-8 group showed the poorest overall survival rate (5 year overall survival, 39%; P = 0.000). These data suggest that CCAR2 also plays an independent role in patients with high IL-8 expression. Overall, high IL-8 expression regulated by CCAR2 plays a critical role in aggressive progression of cervical cancer.

DISCUSSION
Here, we showed that CCAR2 negatively controls the expression of IL-8 through the regulation of AP-1 transcription factor. Considering that IL-8 is a chemokine fertilizing tumor microenvironment, CCAR2 is one of critical factors to be tightly regulated.
In addition to IL-8, CCAR2 can be used to assess a patient's prognosis and response to chemotherapy. High expression of CCAR2 correlates with low overall survival or a poor prognosis for patients with breast cancer [44,45], clear cell renal cell carcinoma [46], colon cancer [8,47], diffuse large B cell lymphoma [48], esophageal squamous cell carcinoma [13], gastric cancer [9,49,50], hepatocellular carcinoma [51], laryngeal and hypopharyngeal carcinoma [52], osteosarcoma [6], ovarian carcinoma [53], and soft tissue sarcoma [54]. While there is a negative correlation between CCAR2 expression and survival for various cancers, our study shows that there is no correlation between CCAR2 and overall survival of cervical cancer patients. In addition, CCAR2 deficiency induces sensitivity to UV irradiation [55,56], suggesting that CCAR2 confers resistance to genotoxic stress-induced chemotherapy.   Several studies show that CCAR2 affects gene expression profiles. CCAR2 deficiency downregulates expression of mitosis-regulatory genes in squamous cell carcinoma cells [5] and expression of Wnt/β-catenin target genes in colon cancer cells [8]. In addition, CCAR2 controls expression of pro-inflammatory cytokines, including IL-6 and IL-8. IL-6 levels are reduced at the early stage of adipocyte differentiation in CCAR2 shRNA-transfected cells, and in adipose tissues from CCAR2 knockout mice fed a high fat diet [57]. TALEN-mediated knockdown of CCAR2 in SW480 colon cancer cells [8] and shRNA-mediated depletion of CCAR2 from MDA-MB-231 breast cancer cells [58] result in low expression of IL-8 compared with that in wild-type cells. In contrast to previous studies, we found that CCAR2 siRNA-transfected cells show the upregulation of IL-8 in HeLa and SiHa cervical cancer cells following oxidative stress; the reason for this may be that previous studies used different cell types or stimuli. The detailed mechanism underlying the differential regulation of IL-8 by CCAR2 requires further investigation. Our study showed that AP-1 activation was required for the upregulation of IL-8 in CCAR2-deficient cells in the absence or presence of H 2 O 2 . Phosphorylation of c-Jun, an AP-1 family member, was involved in the AP-1 activation in H 2 O 2treated CCAR2-deficient cells, but not was detected in the absence of H 2 O 2 . It implies that other AP-1 family members including Fos, ATF, Maf, and JDP participate in its activation for the upregulation of IL-8. The mechanism by which CCAR2 deficiency activates AP-1 transcription factor needs to be investigated.
Until now, the mechanism underlying the role of CCAR2 in cytokine production was unclear. Here, we show that CCAR2 regulates IL-8 production under conditions of oxidative stress. Taken together, the results suggest that CCAR2 may be a potential therapeutic target for inflammatory diseases and cancer.

Cell culture and treatment
HeLa and SiHa cervical cancer cells were maintained in Dulbecco's modified Eagle's medium (Welgene Inc., Korea) and Eagle's minimum essential medium (Welgene Inc., Korea), respectively, supplemented with 10% fetal bovine serum, 100 U/ml penicillin G sodium, 100 µg/ml streptomycin sulfate, and 0.25 µg/ml amphotericin B. Cells were incubated at 37°C in a 5% CO 2 incubator and treated with hydrogen peroxide (H 2 O 2 ) for the indicated times.

Transfection of small interfering RNA (siRNA)
Universal and CCAR2 siRNAs were synthesized by ST Pharm. Co., LTD. (Korea). The siRNA duplexes were as follows: universal siRNA (siU), CCAUAAUUCUUGCCUCUUUdTdT. Transfection of 20 nM siRNA was performed using Lipofectamine RNAiMax (Invitrogen, USA). All experiments were performed 48 h after transfection.

Gene expression analysis
Total RNA was isolated from HeLa cells using an RNeasy Plus Micro kit (Qiagen, USA). Gene expression profiling was based on three independent experiments using an Illumina Human HT-12 v4 Expression BeadChip with 47,322 probes (Illumina, Inc. USA). The expression values for each gene were Log2-transformed and processed using upper quartile (UQ) normalization. Genes showing a significant difference in differential expression were selected with a cut-off value of a 2.0-fold change (FC) and a P-value < 0.05. Gene Ontology (GO) analysis was performed using The Database for Annotation, Visualization and Integrated Discovery (DAVID) 6.8 and functional annotation clustering. Gene Set Enrichment Analysis (GSEA) was performed using GSEA software (http://www.broadinstitute.org/gsea/). The data discussed in this publication have been deposited in Expression of CCAR2 and IL-8 protein in tumor tissues from cervical cancer patients was examined by IHC (n = 47). The immunostaining score is described in the Methods. The association between CCAR2 and IL-8 expression was determined using Pearson's chi square (χ 2 ) test. (B, C) Information regarding expression of CCAR2 and IL-8 mRNA in tumor tissues from cervical cancer patients was downloaded from the GEO (GSE27469) (n = 82) (B) and TCGA (n = 304) (C) databases. Pearson correlation analysis was used to quantify the relationship between CCAR2 and IL-8. r, Pearson's correlation coefficient; P, P-value. the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO) and are accessible through GEO series accession number GSE101612.
The levels of mRNA were quantified using ImageJ software.

Preparation of crude cell extracts and Western blotting
Cells were lysed on ice for 10 min using NETN lysis buffer (100 mM NaCl, 1 mM EDTA, 20 mM Tris-HCl [pH 8.0], 0.5% Nonidet P-40, 50 mM β-glycerophosphate, 10 mM NaF, and 1 mM Na 3 VO 4 ) containing a protease inhibitor cocktail (535140, Millipore, USA). After centrifugation at 100× g for 5 min, the supernatant was saved as a crude cell extract. This was boiled in Laemmli buffer and loaded onto a SDS-polyacrylamide gel [59]. Western blotting was performed according to a standard protocol. The following antibodies were used for Western blotting: c-Jun (9165), c-Jun-pS63 (2361), and β-actin (4970), all from Cell Signaling Technology (USA). Anti-CCAR2 antibody was generated as described previously [60]. mRNA expression data set was obtained from TCGA for cervical cancers. Expression of CCAR2 and IL-8 was defined as high (above median) or low (below median). Kaplan-Meier survival analysis was performed using data from 304 patients. P-values were calculated using the log-rank test.

Statistical analysis
All assays were repeated more than three times under independent conditions. Data were expressed as the mean ± standard error of mean (SEM). Individual expression values were expressed as fold increases above/below that of control cells, whose expression was set to 1 (with no variance [SEM = 0]) to reduce the effect of inter-experimental variations. Differences between three or more groups were evaluated by one-way analysis of variance (ANOVA), followed by a Tukey's honest significant difference (HSD) comparison. Statistical differences were considered significant when P < 0.05 (indicated by *, # , or + ).

Patients
Tissue samples from 47 cervical cancer patients were obtained from the Korea Gynecologic Cancer Bank (KGCB) of the Infrastructure Project for Basic Science of the Ministry of Education, Science and Technology (MEST, Korea). The use of patient specimens was authorized by the Institutional Review Board of Kyung Hee University (KHSIRB-14-044(EA)) and KGCB (KGCB-2014-3). All patients provided informed consent via Yonsei University Gangnam Severance Hospital Gene Bank.

Immunohistochemistry (IHC)
IHC was performed using a Bond Polymer Intense Detection System (Leica-Vision BioSystems, Germany) [45]. In brief, 4 μm sections of formalin-fixed, paraffinembedded tissues were deparaffinized with Bond Dewax Solution and immersed in Bond ER solution for 30 minutes at 100°C to retrieve antigen. Endogenous peroxidase was quenched by incubation with H 2 O 2 for 5 min at room temperature. The sections were incubated for 15 min at room temperature with appropriate primary antibodies using a Bond Intense R detection Kit (Vision BioSystems, Australia). Nuclei were counterstained with hematoxylin. One tissue section showing strong immunoreactivity in a pilot test was used as a positive control. Normal horse serum was substituted for the primary antibody as a negative control. All immunohistochemical slides were examined by three independent investigators (W Kim, BJ Noh, and J Lee) who were blinded to the clinical data. The percentage of stained tumor cells and its staining intensity were evaluated semi-quantitatively. To measure CCAR2 nuclear staining, H-scores were calculated using the following equation: H-score = (% cells with intensity of negative [0] × 0) + (% cells with intensity of [1+] × 1) + (% cells with intensity of [2+] × 2) + (% cells with intensity of [3+] × 3) [62]. The values ranged from 0 to 300. Expression was defined as high (above median) or low (below median). To measure IL-8 with cytoplasmic or nuclear staining, staining intensity was quantified using a score ranging from 0 to 3. Scores 3 and 2 were defined as high, and scores 1 and 0 were defined as low. Pearson's chi square (χ 2 ) test was conducted to measure the correlation between two proteins.

GEO data analysis
mRNA expression data were obtained from the NCBI's GEO database (Accession No. GSE27469). The levels of CCAR2 and IL-8 transcripts were obtained from 82 cervical cancer patients. After upper quartile normalization, the expression value was Log2transformed. Pearson's correlation coefficient (r) was calculated to determine the correlation between the two transcripts. The strength of correlation was interpreted following the guide of Evans [63]; 0.00-0.19, very weak; 0.20-0.39, weak; 0.40-0.59, moderate; 0.60-0.79, strong; 0.80-1.00, very strong. A P-value < 0.05 was considered statistically significant.

TCGA data analysis
Gene expression data from 304 cervical squamous cell carcinoma and endocervical adenocarcinoma cases were downloaded from TCGA (https://tcga-data.nci.nih. gov). The levels of CCAR2 and IL-8 transcripts were determined using Illumina HiSeq2000 RNA Sequencing Version 2 (RNA-Seq V2). Normalized RSEM (RNA-Seq by Expectation-Maximization) values were processed to Log2 values. Pearson's correlation coefficient (r) was calculated to determine the correlation between two transcripts. A P-value < 0.05 was considered statistically significant.

Association between CCAR2 and IL-8 expression and patient survival
The RSEM values were processed as described above. Patient information, including overall survival, was downloaded from the TCGA. For each sample, expression was defined as high (above median) or low (below median). The survival time of the patients was the day of death for deceased patients or the last contact day in alive patients for censoring. The association between transcript levels and patient survival was assessed using Kaplan-Meier curves, and the significance of differences was assessed using the log-rank test. A P-value < 0.05 was considered statistically significant between the population survival curves.

Author contributions
WK and JEK conceived and designed the study. WK and JP carried out experiments and analysed data. WK, BJN, JL and JEK performed data analysis. WK, JWJ and JEK wrote the manuscript. All authors were involved in writing the paper and had final approval of the submitted and published versions.