Loss of oncogenic miR-155 in tumor cells promotes tumor growth by enhancing C/EBP-β-mediated MDSC infiltration

---

INTRODUCTION

miR-155 is a well-known oncogenic microRNA encoded by the BIC (B-Cell Insertion cluster) gene [1]. BIC transcribes a 1.5kb noncoding RNA, which is processed to generate mature miR-155[2]. As the name suggests, the BIC locus was identified as a proviral insertion site causing B-cell lymphoma [1]. Subsequently, miR-155 was identified in the transcript and shown to be responsible for the oncogenic nature of this locus [2]. Many studies have shown that miR-155 is up-regulated in multiple types of cancer including lymphoid, breast, colon, lung and pancreas [3–5]. A number of studies have been undertaken to identify the targets of miR-155 [6, 7]. There are at least 150 validated targets (from MirTar Base) of miR-155. More recently, an Ago-CLIP-Seq study using WT or miR-155-deficient T-cells revealed 250 targets [7]. Although most of these new targets remain to be validated, the existence of such many potential targets suggests that miR-155 has multiple functions that might be dependent on cellular context.

Functional studies using knockout mouse models have revealed that miR-155 is involved in B and T-cell development as well as germinal center formation [8, 9]. Subsequent studies have provided evidence to support a role for miRNA in normal immune system development and function [10, 11]. It is induced upon the activation of B or T cells and controls inflammatory response by functioning in monocytes, macrophage and dendritic cells [12]. Its role in the tumor microenvironment has also been studied. A study using EL4 and B16F10 melanoma cell lines in a allograft model reported that the absence of miR-155 in recipient mice resulted in faster tumor growth [13]. In this study, miR-155 found to modulate IFN gamma production in T cells. Another study revealed that miR-155 knockdown in myeloid cells accelerated tumor growth [14]. miR-155 is also shown to be required for tumor associated macrophage (TAM) mediated antitumor function [14, 15].

In contrast to TAM that negatively regulates tumor growth, there are a number of suppressor cells that inhibits normal antitumor immune activity. Among them, the myeloid derived suppressor cells (MDSC) are known to be regulated by miR-155 via SHIP-1 [16]. In a line with this finding, MDSCs was shown to require miR-155 to facilitate tumor growth [17]. However, the latter report showed the loss of host miR-155 overall promoted antitumor activity, which is not consistent with the results of other studies. Interestingly, another recent study revealed the loss of miR-155 in MDSCs enhanced its recruitment and function in solid tumor [18], suggesting that the role of miR-155 may vary depending on the tumor type and the model system.

To address some of the inconsistencies in previous studies and dissect the diverse role of miR-155 in breast cancer, we have examined its function in two different contexts. First, we have utilized a well-characterized Brca1^cko/cko;Trp53^cko/cko; K14-Cre mouse model because miR-155 is upregulated in BRCA1-deficient tumors [19]. We have examined the effect of miR-155 loss on tumorigenesis by germ-line inactivation of miR-155. In these mice, miR-155 is deficient in both the tumor as well as the tumor microenvironment. Second, we have used an allograft model to examine the growth of LLC1 cells with stable knockdown of miR-155 in miR-155^ko/+ or miR-155^ko/ko mice. Our results suggest a dual role for miR-155 in the tumor and the tumor microenvironment. The implications of our findings on the use of miR-155 as a therapeutic target are discussed.

RESULTS

Germ-line inactivation of miR-155 does not affect the tumor free survival of Brca1/Trp53 mutant mice

Considering the oncogenic role of miR-155 and its up-regulation in BRCA1-deficient tumors [20] we reasoned its inactivation might increase the tumor latency or reduce the tumor incidence. To examine the possibility, we crossed miR-155^ko/ko mice with well established Brca1^cko/cko;Trp53^cko/cko;K14-Cre mouse model and monitored a cohort of Brca1^cko/cko;Trp53^cko/cko;K14-Cre; miR-155^ko/ko and Brca1^cko/cko;Trp53^cko/cko;K14-Cre; miR-155^ko/+ mice for tumor incidence [21]. Surprisingly, we found the tumor free survival of Brca1^cko/cko;Trp53^cko/cko;K14-Cre mice was not significantly changed in miR-155^ko/ko and miR-155^+/ko genetic backgrounds (Figure 1A). In addition, we did not find any significant histological differences between the tumors from the two groups (Supplementary Figure 1A, Supplementary Table 1). Consistent with our previous findings, we observed a significant up-regulation of miR-155 in tumors from Brca1^cko/cko;Trp53^cko/cko;K14-Cre mice compared to EMT6 mammary cancer cells and primary mouse embryonic fibroblasts (MEF) (Figure 1B). These results suggest that germline inactivation of miR-155 does not affect the tumor free survival of Brca1^cko/cko;Trp53^cko/cko;K14-Cre mice. However, considering the known oncogenic role of miR-155 in breast cancer cell as well as its tumor suppressive effects via normal immune activation[14, 22], it is possible that the two opposing effects might compensate each other.

Increased MDSC infiltration in miR-155-deficient tumor microenvironment

Loss of miR-155 in MDSCs is reported to exert increased migration and immune suppressive potential [17]. Therefore, we first examined whether MDSCs play a role in tumorigenesis in Brca1^cko/cko;Trp53^cko/cko;K14 Cre; miR-155^ko/ko mice as well. To measure the infiltrated MDSCs in mammary tumor, we used two known MDSC markers, CD11b and Gr1. FACS analysis using these two antibodies labeled with FITC or PE showed a double positive population that is enriched in the tumor and peripheral blood from the tumor-bearing mouse (Figure 2A and 2B). Using this approach, we compared the number of infiltrated MDSCs in the spleen (n=44) and tumors (n=28) of mutant and control mice. We also examined CD4 and CD8 positive cells in these samples. In the spleen, we observed no difference in the number of MDSCs between the two genotypes, suggesting that the generation of MDSCs in the tumor bearing mice is not affected by the presence or absence of miR-155 (Figure 2C). We did not observe any significant difference in the CD4 and CD8 positive cells. In contrast, we found an increase in the MDSC population in the tumors from miR-155^ko/ko mice (Figure 2D). We also observed increased CD4 positive and decreased CD8 positive cells in the miR-155^ko/ko tumors but the differences were not significant due to high variability observed in the tumors. When the infiltration rate was calculated by dividing the portion of MDSC in tumor by the portion of MDSC in spleen, we observed a significant increase in the ratio in miR-155^ko/ko mice (Figure 2E). This was further confirmed by visualizing MDSCs in the tumors by IHC staining. As shown in Figure 2F, the degree of MDSC infiltration in the peripheral region of tumor detected by CD11b and GR1 antibodies, greatly varies between samples. However, when we scored the degree of the infiltration using an arbitrary four tier grading system (-, +, ++ and +++), we observed that the tumors from miR-155^ko/ko mice have higher degree of the MDSC infiltration (Figure 2G, Table 1). Based on these observations, we concluded that the MDSCs are more efficiently recruited to the tumor sites in the miR-155^ko/ko mice. It is possible that this contributes to the tumor growth in miR-155-deficient background (Figure 1A). These findings are consistent with other reports showing increased MDSC infiltration in the absence of miR-155 [18].

Cytokine array analysis revealed dynamic changes in cytokine/chemokine expression in the absence of miR-155

We next examined whether miR-155-deficient tumor cells are better adapted to attracting MDSCs than the tumor cells expressing miR-155. Indeed, an in vitro migration experiment using normal MDSC and conditioned medium from miR-155^ko/ko and miR-155^+/ko. Tumor cells showed that the miR-155^ko/ko conditioned medium induced much more MDSC migration than miR-155^+/ko (Figure 3A). This suggested that some unknown tumor derived factor(s) from miR-155^ko/ko tumor cells were attracting more MDSCs. To identify such factor(s), we designed a co-culture system consisting of tumor cells plus differentiated 3T3-L1 cells [23] (Figure 3B, Supplementary Figure 2). Using this system, we aimed to mimic the known interplay between breast tumor and adipose cells, which produce high levels of cytokines/chemokines, in vitro. We confirmed that indirect co-culture system was functioning properly, by measuring the expression of a number of cytokines that are known to be up-regulated [24] (Supplementary Figure 3). Using this indirect co-culture system and a PCR array, we examined the cytokine/chemokine expression changes in 3T3-L1 derived-adipocytes and miR-155^ko/ko and miR-155^+/ko tumor cells separately. We identified a number of tumor-driven cytokines/chemokines that are differentially regulated in the absence of miR-155 (Table 2). These included CXCL5, 9, 11 as well as IL-4, 6,10,13, which are up-regulated and known to be involved in the MDSC migration [25, 26]. Real time PCR analysis confirmed up-regulation of these factors (Figure 3C–3I) in co-cultured miR-155^ko/ko cells.

Over-expression of miR-155 induces cytokine changes in human breast cell lines

We next examined if miR-155-mediated cytokine/chemokine regulation also occurs in human breast cancer cells. We transfected MCF7 and MCF10A cells that have low levels of endogenous miR-155 with miR-155 mimic. The transfectants were then indirectly co-cultured with differentiated adipocytes as shown in Figure 3B. We measured RNA expression levels of cytokine/chemokines including CXCL5, 9,11 and IL-4,6,10, which are shown to be up-regulated in Figure 3C-H. As shown in Figure 4A-F, over-expression of miR-155 mimic in both cells dramatically decreased the expression of the cytokines, although the relative basal level was low for some cytokines (e.g., CXCL9 in MCF10A). The cytokine regulation by miR-155 was further confirmed by ELISA assay for IL-6 and CXCL9 by over expressing miR-155 in MCF7 and MCF10A (low endogenous miR-155) and silencing miR-155 in MDA-MB-231 and MDA-MB-436 (high endogenous miR-155). We observed significant reduction in the levels of the two cytokines by miR-155 over-expression (miRH155) whereas inhibition of the miR-155 (miRZIP155) increased cytokine production (Figures 4G and H). These results support our findings from the mouse model and demonstrate that the miR-155-mediated cytokine regulation occurs in human cells as well.

C/EBP-β is responsible for the up-regulation of cytokines/chemokines

To understand how the absence or inhibition of miR-155 induces a set of cytokines/chemokines, we examined the possibility that miR-155 could bind to the UTR of the cytokine/chemokine genes. However, analysis using TargetScan and PicTar prediction tool resulted in no miR-155 binding sites in the UTR of these genes (data not shown). Therefore, we hypothesized that a transcription factor regulating the genes encoding these cytokine/chemokines might be targeted by miR-155. We focused on CCAAT Enhancer Binding Protein-beta (C/EBP-β) because a number of reports have shown that it regulates the expression of various kinds of cytokines in immune as well as cancer cells [27, 28]. Moreover, it is known as a direct target of miR-155 in macrophage [29]. To elucidate a possible role of C/EBP-β in breast cancer cells, we first analyzed miR-155^+/ko and miR-155^ko/ko breast cancer cells to measure the C/EBP-β protein levels. We found a marked increase in C/EBP-β and its phosphoylated form in miR-155^ko/ko cells relative to miR-155^+/ko cells (Figure 5A). In addition, we observed up-regulated luciferase activity in miR-155^ko/ko cells when C/EBP-β 3’UTR reporter plasmids were introduced into three independent miR-155^ko/+ or miR-155^ko/ko cell lines (Figure 5B). The regulation of C/EBP-β by miR-155 is further confirmed by a C/EBP-β UTR reporter assay in MDA-MB436 and MDA-MB-436-miRZIP155 (miR-155 knock-down cells) or in MCF7/MCF7-miRZIP155 cells, showing miR-155 dependent UTR regulation (Figure 5C and 5D left panel). In addition, we showed that ‘miR-155 mimic significantly inhibited the luciferase activity of the CEB/P-β 3’UTR reporter but did not affect CEB/P-β 3’UTR reporter with mutated miR-155 binding sites (Mut). Collectively, these results indicate that CEB/P-β is a direct target of miR-155.

We next searched the promoter regions of the genes encoding these cytokines/chemokines for C/EBP-β binding by using ENCODE database and found ChIP signal on IL-1β, 6, 10, 13 and CXCL9 (Supplementary Figure 4A-E). We further examined IL-6 and CXCL9, as IL-6 is a well-known target whereas the CXCL9 is uncharacterized target of the C/EBP-β in breast cancer[30, 31]. By using luciferase-based reporter constructs for these two promoters we observed that these are activated when C/EBP-β is ectopically expressed (Figure 5E and 5F). We examined direct binding of C/EBP-β to IL-6 and CXCL9 promoters by performing chromatin immunoprecipitation (ChIP, Figure 5G, Supplementary Figure 4F for quantitation). We observed a clear ChIP signal in all two promoters with marked increase upon the inhibition of miR-155 (miRZIP155) in IL-6 and CXCL9 promoters. These results suggest that C/EBP-β directly transactivates these cytokine genes. Finally, we confirmed that the up-regulation of IL-6 and CXCL9 expression mediated by miR-155 inhibition is dependent on C/EBP-β. We knocked down C/EBP-β by siRNA treatment, which reduced IL-6 and CXCL9 expression induced by miR-155 inhibition in two breast cancer cell lines (Figure 5H~5K). Taken together, these results demonstrate that C/EBP-β is, at least in part, responsible for the up-regulation of cytokine/chemokines induced by miR-155 inhibition.

Opposing roles of miR-155 inactivation in cancer cells and the tumor microenvironment on tumor growth

The results described above strongly suggest a tumor suppressive role of miR-155 in the tumor microenvironment via C/EBP-β mediated cytokine production and MDSC infiltration. In contrast, previous findings have demonstrated an oncogenic role of miR-155 in tumor cells [32, 33]. Therefore, we hypothesized that these two opposing roles might compensate each other in vivo, when the miR-155 is lost in both compartments. This idea is supported by the observation that germ-line inactivation of miR-155 did not affect tumor formation in Brca1^cko/cko;Trp53^cko/cko;K14 Cre;miR-155^ko/ko mice (Figure 1A). To test this hypothesis, we generated Lewis Lung carcinoma cells (LLC1) with stable knockdown of miR-155 and injected the cells into isogenic (C57BL/6) miR-155^ko/+ or miR-155^ko/ko mice and monitored tumor growth (Figure 6A and B, Supplementary Figure 5A). The level of miR-155 in tumors was confirmed by real time PCR (Figure 6C). We observed that when injected into miR-155^ko/+ mice, the knockdown of miR-155 (indicated as miRZIP155) in LLC1 cells slowed down tumor growth compared to the control vector (Figure 6B; compare red line with blue line). More importantly, the growth defect of miRZIP155 LLC1 cells was restored when they were introduced into the miR-155^ko/ko mice (Figure 6B, compare purple line with red line). Interestingly, the growth of miRZIP155 LLC1 cells in miR-155^ko/ko mice was comparable to the growth of control LLC1 cells in miR-155^ko/+ mice (Figure 6B, compare blue and purple lines). These observations provide strong evidence to suggest a compensatory effect of miR-155 in the tumor and the tumor microenvironment, which may explain why we did not observe any significant difference in the tumor growth in Brca1^cko/cko;Trp53^cko/cko;K14 Cre mice in miR-155^ko/+ and miR-155^ko/ko backgrounds (Figure 1A).

Increased cytokine expression and MDSC infiltration is caused by the loss of miR-155 in both cancer cell and tumor microenvironment

In order to confirm our results that miR-155 dependent C/EBP-β up-regulation in cancer cells mediates cytokine up-regulation and MDSC infiltration (Figure 3, Figure 5), we analyzed the LLC1 xenograft model in more detail. First, we detected increased C/EBP-β and phospho-C/EBP-β levels in miR-155 KD LLC1 cells (Figure 6D, Supplementary Figure 5B for confirmation and 5C for quantitation). Of note, the level of C/EBP-β is increased not only by the miR-155 inhibition, but also by grafting cells into the miR-155^ko/ko mice, suggesting that there are host-driven factors that induce C/EBP-β. When we measured MDSCs from the dissected tumors by FACS, we found the inhibition of miR-155 in tumor cells increase MDSC infiltration ratio (Figure 6E Blue and Red bars, Supplementary Figure 5D and E, Supplementary Table 2). This increase is further enhanced when the cells are introduced into miR-155^ko/ko mice (compare red with purple bars in Figure 6E). These data suggest that loss of miR-155 in the tumor as well as in the tumor microenvironment contribute to the increased MDSC infiltration into the tumor. Furthermore, we found this increase of MDSC infiltration is correlated with the up-regulated level of IL-6 and CXCL9, as measured by ELISA (Figure 6F and G, Supplementary Figure 6A and B). Taken together, these results demonstrate a novel function of miR-155 in the tumor and its microenvironment, which is mediated by C/EBP-β and its regulatory function on cytokine expression that in turn results in altered MDSC infiltration into the tumor (Summarized in Figure 6H).

DISCUSSION

In the past few years several studies have been published describing the role of mir-155 in tumor progression that suggest its dual role as an oncogene as well as a tumor suppressor [32–34]. However, some of the findings are conflicting. The first study published by Huffaker et. al. showed that the loss of miR-155 in host animal results in faster xenograft growth due to increased response of T cells to interferon gamma [13]. On the other hand, Li et. al. demonstrated that miR-155 promotes expansion of functional MDSCs suggesting that loss of miR-155 might negatively affect MDSC proliferation [16]. However, another recent report by Wang et.al. revealed that miR-155 deficiency enhances the recruitment and functions of MDSCs [17], which is beneficial to the tumor. This finding is contradictory to another study reported by Chen et. al. where the loss of miR-155 enhanced antitumor T cell activity and reduced MDSC infiltration into the tumor [18].

At present, it is unclear why such conflicting results have been obtained. However, these studies have all used different mouse models, either xenograft models using different cell lines or a chemically induced, acute tumor model, which may contribute to some of the variability. In this report, we analyzed a Brca1/Trp53-based spontaneous breast cancer model with genetic loss of miR-155. Some of our results support the notion that loss of miR-155 in MDSC is beneficial for tumor growth. Our results show that loss of miR-155 in the host animal allows more MDSCs to infiltrate into the tumor (Figure 6E, compare 1^st and 3^rd, 2^nd and 4^th bar). More importantly, our findings reveal that the miR-155 deficient tumor cells also contribute to the MDSC infiltration (Figure 3A and Figure 6E; compare 1^st and 2^nd, 3^rd and 4^th bar). Therefore, it is likely that MDSC infiltration is maximized when both of the tumor and MDSCs lack miR-155.

Among the likely genes that are transcriptionally affected by the co-culture system, we focused on the cytokines and chemokines because these can potentially mediate the recruitment of the MDSCs. We hypothesized that certain tumor cell driven factor can be differentially expressed in miR-155^ko/ko cells that can aid in the recruitment of more MDSCs, which in turn will create a favorable microenvironment for tumor cell growth. Indeed, we showed that IL-6 and CXCL9 are such cytokine and chemokine. Future studies using neutralizing antibody against IL-6 and CXCL9 will reveal the clinical applicability of this finding.

Finally, because miR-155 is an important therapeutic target, our findings suggest that its systemic inhibition as an antitumor therapy can have detrimental effects due to increased MDSC infiltration. Even though we show only MDSC is significantly enriched in miR-155-deficient tumor microenvironment in the breast cancer model, we expect other immune effector cells to also differentially infiltrate in human cancer when miR-155 is inactivated by therapeutic agents, such as chemically modified AntagomiR. Finding such cells and understanding how the elegant interplay among these immune cells can affect tumor growth will be critical for developing better therapeutic strategies to target miR-155.

MATERIALS AND METHODS

Generation and genotyping of mice

Brca1^cko/cko;Trp53^cko/cko;K14-Cre mice were generated by mating Brca1^cko/cko mice with Trp53^cko/cko and K14-Cre mouse (all mouse lines were obtained from NCI-Frederick repository). The genotyping was performed as described previously[35]. The mice were further breeded with miR-155^ko/ko (Jackson Laboratory, Bar Harbor, USA) to obtain Brca1^cko/cko;Trp53^cko/cko;K14-Cre; miR-155^ko/ko and Brca1^cko/cko;Trp53^cko/cko;K14-Cre; miR-155^ko/+ mice. miR-155 mutation was genotyped as described previously[9]. Primers are listed in Supplementary Table 3. All animal studies were performed by the Guideline for the Care and Use of Laboratory Animals and approved by the NCI-Frederick Animal Care and the Laboratory of Animal Research in Asan Institute of Life Sciences.

Tumor immunohistochemical and immunofluoresence staining

For histological analysis, FFPE sections were stained with hematoxylin and eosin (H&E). For immunofluoresence staining, frozen tissue sections were co-immunostained with anti-mouse Gr1-FITC (1:1000; Sigma) and anti-CD11b-Cy3 (1:1000; Sigma) antibodies. Images were captured using a LMS510 confocal microscope (ZEISS, Germany), analyzed using Image-Pro Plus 6.0 package (Media Cybernetics, MD, USA).

FACS analysis of immune cells in spleen or tumor

To detect immune cells, tumors and spleens were processed for FACS analysis. Briefly, tumors were minced and digested in 5 ml of dissociation solution (RPMI1640 medium with Collagenase type l, 200 U/ml) for 1-2 hours at 37°C;. Spleens were crushed in RPMI1640 medium using cell strainer (100 mM, Falcon, NJ, USA). Digested tumors were also filtered and washed twice. Erythrocytes were lysed by RBC lysis buffer (Sigma-Aldrich, St. Louis, USA). 1 × 10⁷ processed cells were incubated with PE-conjugated Gr1 or CD8 mAb (BD Pharmigen, CA, USA) or FITC-conjugated CD11b or CD4 mAb in FACS buffer for 20 mins at RT. Samples were washed twice with FACS staining buffer, analyzed in the FACSCalibur (Figure 2) or Accuri Flow Cytometer (Figure 6 and Supplementary Figure 5, BD Biosciences). Data were analyzed using CellQuest Pro or CFlow software.

Derivation of mouse primary breast cancer cells and cell lines

miR-155 heterozygous (miR-155^ko/+) or knockout (miR-155^ko/ko) primary cancer cells were isolated from tumor tissues using modified MEC (mammary epithelial cell) isolation procedure described previously[19]. Briefly, fresh tumor was minced and incubated with collagenase type I and IV in DMEM with antibiotics (Hyclone) for 6 hrs. The digested tissue was passed through strainer (40 mM pore size, Falcon) and the cell suspension was washed once. The primary cells were cultured in DMEM containing 10% FBS and antibiotics. MEF, EMT6, MCF7, MDA-MB-231, MDA-MB-436, HeLa, LLC1 (ATCC, Manassas, VA, USA) and 293TN (System Biosciences, SBI, Mountain View, CA, USA) cells were maintained in DMEM containing 10% FBS and 1% antibiotics. MCF10A (ATCC) cells were cultured in RPMI1640 medium with 10% FBS, hEGF (200 ng/ml), Hydrocortisol (10 mg/ml), transferrin (10 mg/ml) and 1% penicillin/streptomycin.

Differentiation of 3T3-L1 cells to adipocytes

The 3T3-L1 preadipocyte cell line was provided by Prof. Young-Sup Song (University of Ulsan, college of Medicine) and maintained in DMEM/10% FBS. Adipocyte differentiation was induced by a mixture containing 10% FBS, 1 unit insulin (Eli Lilly), 0.5 mM 3-isobutyl-1-methylxanthine (Sigma), 2 mg/ml Dexamethasone (Sigma) and 0.5 mM Rosiglitazone (Cayman Chemical Company, Michigan, USA). Oil red O staining was performed at the end point of culture. Indirect co-culture was performed as described previously with some modifications[36]. The 3T3-L1 cells were seeded onto trans-well insert with 0.3 mm pore size (Corning Incorporated Life Sciences, MA, USA). The cells were differentiated into adipocytes as described above and then co-cultured with miR-155^ko/+ or miR-155^ko/ko cells or cancer cell lines for 24, 48, and 72 hours.

miR-155 knockdown or overexpression

For knockdown of miR-155, lentiviral constructs containing anti-miR-155 sequence (miRZIP155) was used. The lentiviral production was performed following the manufacturer’s instructions. Control or miRZIP155-containing the lentivirus particles were infected into MDA-MB-436, MDA-MB-231 and three mouse miR-155^ko/+ primary cells. After 24 hours, cells were selected by Puromycin (1.5 mg/ml; Sigma) for 7 days. For miR-155 overexpression, the MCF7 and MCF10A cells were transfected with 30 nM miR-155 mimic or scrambled miRNA by G-fectin (Genolution, Seoul, Republic of Korea)

RT2 profiler PCR array

First strand cDNA for qPCR was synthesized using the RT2 First Strand cDNA Kit (SABiosciences, MD, USA). Gene expression analyses were done using RT2 Profiler PCR Array qPCR kit with the RT2 SYBR Green qPCR Master Mix (SABiosciences), in the LightCycler 480 ll (Roche Applied Sciences, IN, USA). Expression analysis was using the manufacturer’s analysis tool and gene expression was normalized to housekeeping genes. Differential expression was measured as fold expression relative to the miR-155^ko/ko cells.

Reporter construction and luciferase assay

For reporter construction, IL-6 and CXCL9 promoter regions were amplified (primer sequences in Supplementary Table 3) and introduced into the pGL3 Enhancer vector (Promega) via SpeI and HindIII sites. MCF7 and MDA-MB-436 cells were co-transfected with pcDNA-C/EBP-β (a gift from Dr. Young-Sup Song) and reporter vectors (pGL3-IL-6 or pGL3-CXCL9 or Empty vector). After 48 hours, luciferase activity was measured using Dual-Luciferase Reporter Assay System (Promega Corp, WI, USA) and Victor Luminometer (Perkin-Elmer, MA, USA). For C/EBP-β UTR reporter assay, sequences of the 3’ UTR for mouse and human C/EBP-β gene (NM_009883 and NM_005194) were cloned (primer sequences are in Supplementary Table 3) into the pMIR-REPORT Luciferase vector (Applied Biosystems). Mutant reporter was generated by site-directed mutagenesis within the miR-155 seed match region. The miR-155 target site in the C/EBP-β 3′ UTR was mutated by deleting the 8 nt miR-155 seed match sequence (AGCAUUAA at nucleotide positions 554–561 in the C/EBP-β 3′ UTR) using the QuikChange Site-Directed Mutagenesis kit according to manufacturer's instructions (Stratagene, CA, USA). The pMiR-C/EBP-β-3’UTR (Wt or Mut) was co-transfected into MCF7 cells either with scramble miRNA or miR-155 mimic using Lipofectamine2000 (Invitrogen, CA, USA) according to the transfection procedures.

Chromatin immunoprecipitation (ChIP) assay

ChIP was performed by a protocol (http://cshprotocols.cshlp.org/content/ 2009/9/pdb prot5279. full) with minor modifications. Briefly, MDA-MB-436 cells infected with control or miRZIP155 lentiviral vectors were cross-linked with 1% formaldehyde for 10 minutes at RT. Then the cells were collected, lysed, and the chromatin was sonicated, and incubated with a C/EBP-β antibody overnight. PCR was used to detect ChIP signal with the primers listed in Supplementary Table 3.

LLC1 Xenograft experiment

To establish subcutaneous tumors in miR-155^ko/+ or miR-155^ko/ko mice, 1 × 10⁶ LLC1 cells infected with control or miRZIP155 lentiviruses were implanted subcutaneously. Tumor growth was measured by caliper every 3-4 days. Tumor volume was determined by the formula: length X width²/2.

Enzyme-linked immunosorbent assay (ELISA)

The level of IL-6 and CXCL9 in tumor or culture supernatant was assessed by ELISA. Cell extracts were prepared from snap-frozen tumor and spleen tissues. Culture supernatants were collected from MCF10A, MCF7, MDA-MB-231 and MDA-MB-436 after 48 hrs of the indirect co-culture. The IL-6 and CXCL9 concentration in the media and tissue extracts was determined by ELISA kits from R&D Systems (Minneapolis, MN, USA). An aliquot of 100 μl of conditioned medium or 25 μl of tissue extract made to 100 μl with diluent (0.1% bovine serum albumin, 0.05% Tween 20, 10 mM Tris and 150 mM sodium chloride, pH 7.3) was analyzed.

Western blot analysis

Western blot analysis was performed as previously described[19]. Briefly, cells were lysed in lysis buffer [150 mM NaCl, 1 % Triton X-100, 1% Sodium deoxycholate, 50 mM Tris-HCl (pH 7.5), 2 mM EDTA (pH 8.0), and 0.1% SDS]. 10~50 μg of protein were separated on SDS PAGE, transferred to Nitrocellulose membrane, probed with anti-C/EBP-β or anti-phospho-C/EBP-β (Thr235)(1:1000, Cell Signaling) antibodies. Membranes were stripped and re-probed with anti β-actin antibody (1:1,000; Santa Cruz Biotechnology, CA, USA) to ensure equal loading. Densitometric analysis was performed using the imageJ 1.47v system.

Quantitative RT-PCR (qRT-PCR)

The expression levels of cytokines, chemokines, and C/EBP-β were measured by SYBR Green PCR Kit, in LightCycler 480 II (Roche). The primer sequences are shown in Supplementary Table 3. Human and mouse RpL13a was used as an internal control gene. Relative quantification was calculated using the 2-(ΔΔCt) method [37]. For miR-155 quantization, miScript ll RT Kit (Qiagen, Hilden, Germany) and miScript SYBR Green PCR Kit (Qiagen) were used. Small nuclear RNA molecule U6 (RNU6) was used as an internal control.

Statistics

All data are presented as means ±standard error mean (SEM). Statistical significance between two groups was calculated using t-test. P-value less than 0.05 was considered to be significant.

CONFLICTS OF INTEREST

The authors have no conflicts of interest to declare.

ACKNOWLEDGMENTS

We thank Drs. Peter Johnson and Jonathan Keller and for helpful discussions and critical review of the manuscript. This study was supported by the Intramural Research Program, Center for Cancer Research, National Cancer Institute, U.S. National Institutes of Health (S.K.S) and grants to S.C. from the Korean Health Technology R&D Project, Ministry if Health & Welfare, Republic of Korea (Grant No. HI14C2640) and Asan Institute for Life Sciences (Grant No. 2015-571)

REFERENCES

1. Tam W, Ben-Yehuda D, Hayward WS. bic, a novel gene activated by proviral insertions in avian leukosis virus-induced lymphomas, is likely to function through its noncoding RNA. Molecular and cellular biology. 1997; 17:1490–1502.

2. Eis PS, Tam W, Sun L, Chadburn A, Li Z, Gomez MF, Lund E, Dahlberg JE. Accumulation of miR-155 and BIC RNA in human B cell lymphomas. Proceedings of the National Academy of Sciences of the United States of America. 2005; 102:3627–3632.

3. Tili E, Croce CM, Michaille JJ. miR-155: on the crosstalk between inflammation and cancer. International reviews of immunology. 2009; 28:264–284.

4. Faraoni I, Antonetti FR, Cardone J, Bonmassar E. miR-155 gene: a typical multifunctional microRNA. Biochimica et biophysica acta. 2009; 1792:497–505.

5. Jongen-Lavrencic M, Sun SM, Dijkstra MK, Valk PJ, Lowenberg B. MicroRNA expression profiling in relation to the genetic heterogeneity of acute myeloid leukemia. Blood. 2008; 111:5078–5085.

6. Meier J, Hovestadt V, Zapatka M, Pscherer A, Lichter P, Seiffert M. Genome-wide identification of translationally inhibited and degraded miR-155 targets using RNA-interacting protein-IP. RNA biology. 2013; 10:1018–1029.

7. Loeb GB, Khan AA, Canner D, Hiatt JB, Shendure J, Darnell RB, Leslie CS, Rudensky AY. Transcriptome-wide miR-155 binding map reveals widespread noncanonical microRNA targeting. Molecular cell. 2012; 48:760–770.

8. Rodriguez A, Vigorito E, Clare S, Warren MV, Couttet P, Soond DR, van Dongen S, Grocock RJ, Das PP, Miska EA, Vetrie D, Okkenhaug K, Enright AJ, Dougan G, Turner M, Bradley A. Requirement of bic/microRNA-155 for normal immune function. Science. 2007; 316:608–611.

9. Thai TH, Calado DP, Casola S, Ansel KM, Xiao C, Xue Y, Murphy A, Frendewey D, Valenzuela D, Kutok JL, Schmidt-Supprian M, Rajewsky N, Yancopoulos G, Rao A, Rajewsky K. Regulation of the germinal center response by microRNA-155. Science. 2007; 316:604–608.

10. Seddiki N, Brezar V, Ruffin N, Levy Y, Swaminathan S. Role of miR-155 in the regulation of lymphocyte immune function and disease. Immunology. 2014; 142:32–38.

11. Vigorito E, Kohlhaas S, Lu D, Leyland R. miR-155: an ancient regulator of the immune system. Immunological reviews. 2013; 253:146–157.

12. Lind EF, Ohashi PS. Mir-155, a central modulator of T-cell responses. European journal of immunology. 2014; 44:11–15.

13. Huffaker TB, Hu R, Runtsch MC, Bake E, Chen X, Zhao J, Round JL, Baltimore D, OʽConnell RM. Epistasis between microRNAs 155 and 146a during T cell-mediated antitumor immunity. Cell reports. 2012; 2:1697–1709.

14. Zonari E, Pucci F, Saini M, Mazzieri R, Politi LS, Gentner B, Naldini L. A role for miR-155 in enabling tumor-infiltrating innate immune cells to mount effective antitumor responses in mice. Blood. 2013; 122:243–252.

15. Dueck A, Eichner A, Sixt M, Meister G. A miR-155-dependent microRNA hierarchy in dendritic cell maturation and macrophage activation. FEBS letters. 2014; 588:632–640.

16. Li L, Zhang J, Diao W, Wang D, Wei Y, Zhang CY, Zen K. MicroRNA-155 and MicroRNA-21 promote the expansion of functional myeloid-derived suppressor cells. Journal of immunology. 2014; 192:1034–1043.

17. Wang J, Yu F, Jia X, Iwanowycz S, Wang Y, Huang S, Ai W, Fan D. MicroRNA-155 deficiency enhances the recruitment and functions of myeloid-derived suppressor cells in tumor microenvironment and promotes solid tumor growth. International journal of cancer. 2015; 136:E602–613.

18. Chen S, Wang L, Fan J, Ye C, Dominguez D, Zhang Y, Curiel TJ, Fang D, Kuzel TM, Zhang B. Host miR155 promotes tumor growth through a myeloid-derived suppressor cell-dependent mechanism. Cancer research. 2015; 75:519–531.

19. Chang S, Wang RH, Akagi K, Kim KA, Martin BK, Cavallone L, Kathleen Cuningham Foundation Consortium for Research into Familial Breast C, Haines DC, Basik M, Mai P, Poggi E, Isaacs C, Looi LM, Mun KS, Greene MH, Byers SW, et al. Tumor suppressor BRCA1 epigenetically controls oncogenic microRNA-155. Nature medicine. 2011; 17:1275–1282.

20. Chang S, Sharan SK. Epigenetic control of an oncogenic microRNA, miR-155, by BRCA1. Oncotarget. 2012; 3:5–6. doi: 10.18632/oncotarget.3428.

21. Rottenberg S, Nygren AO, Pajic M, van Leeuwen FW, van der Heijden I, van de Wetering K, Liu X, de Visser KE, Gilhuijs KG, van Tellingen O, Schouten JP, Jonkers J, Borst P. Selective induction of chemotherapy resistance of mammary tumors in a conditional mouse model for hereditary breast cancer. Proceedings of the National Academy of Sciences of the United States of America. 2007; 104:12117–12122.

22. Yu F, Jia X, Du F, Wang J, Wang Y, Ai W, Fan D. miR-155-deficient bone marrow promotes tumor metastasis. Molecular cancer research. 2013; 11:923–936.

23. Zebisch K, Voigt V, Wabitsch M, Brandsch M. Protocol for effective differentiation of 3T3-L1 cells to adipocytes. Analytical biochemistry. 2012; 425:88–90.

24. Tan J, Buache E, Chenard MP, Dali-Youcef N, Rio MC. Adipocyte is a non-trivial, dynamic partner of breast cancer cells. The International journal of developmental biology. 2011; 55(7-9):851–859.

25. Lechner MG, Megiel C, Russell SM, Bingham B, Arger N, Woo T, Epstein AL. Functional characterization of human Cd33+ and Cd11b+ myeloid-derived suppressor cell subsets induced from peripheral blood mononuclear cells co-cultured with a diverse set of human tumor cell lines. Journal of translational medicine. 2011; 9:90.

26. Thakur A, Schalk D, Tomaszewski E, Kondadasula SV, Yano H, Sarkar FH, Lum LG. Microenvironment generated during EGFR targeted killing of pancreatic tumor cells by ATC inhibits myeloid-derived suppressor cells through COX2 and PGE2 dependent pathway. Journal of translational medicine. 2013; 11:35.

27. Zahnow CA. CCAAT/enhancer-binding protein beta: its role in breast cancer and associations with receptor tyrosine kinases. Expert reviews in molecular medicine. 2009; 11:e12.

28. Huber R, Pietsch D, Panterodt T, Brand K. Regulation of C/EBPbeta and resulting functions in cells of the monocytic lineage. Cellular signalling. 2012; 24:1287–1296.

29. He M, Xu Z, Ding T, Kuang DM, Zheng L. MicroRNA-155 regulates inflammatory cytokine production in tumor-associated macrophages via targeting C/EBPbeta. Cellular & molecular immunology. 2009; 6:343–352.

30. Faggioli L, Costanzo C, Merola M, Bianchini E, Furia A, Carsana A, Palmieri M. Nuclear factor kappa B (NF-kappa B), nuclear factor interleukin-6 (NFIL-6 or C/EBP beta) and nuclear factor interleukin-6 beta (NFIL6-beta or C/EBP delta) are not sufficient to activate the endogenous interleukin-6 gene in the human breast carcinoma cell line MCF-7. Comparative analysis with MDA-MB-231 cells, an interleukin-6-expressing human breast carcinoma cell line. European journal of biochemistry. 1996; 239:624–631.

31. Galien R, Evans HF, Garcia T. Involvement of CCAAT/enhancer-binding protein and nuclear factor-kappa B binding sites in interleukin-6 promoter inhibition by estrogens. Molecular endocrinology. 1996; 10:713–722.

32. Mattiske S, Suetani RJ, Neilsen PM, Callen DF. The oncogenic role of miR-155 in breast cancer. Cancer epidemiology, biomarkers & prevention: a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology. 2012; 21:1236–1243.

33. Wang J, Wu J. Role of miR-155 in breast cancer. Frontiers in bioscience. 2012; 17:2350–2355.

34. Chen Z, Ma T, Huang C, Hu T, Li J. The pivotal role of microRNA-155 in the control of cancer. Journal of cellular physiology. 2014; 229:545–550.

35. Liu X, Holstege H, van der Gulden H, Treur-Mulder M, Zevenhoven J, Velds A, Kerkhoven RM, van Vliet MH, Wessels LF, Peterse JL, Berns A, Jonkers J. Somatic loss of BRCA1 and p53 in mice induces mammary tumors with features of human BRCA1-mutated basal-like breast cancer. Proceedings of the National Academy of Sciences of the United States of America. 2007; 104:12111–12116.

36. Wang L, Zhang H, Rodriguez S, Cao L, Parish J, Mumaw C, Zollman A, Kamoka MM, Mu J, Chen DZ, Srour EF, Chitteti BR, HogenEsch H, Tu X, Bellido TM, Boswell HS, et al. Notch-dependent repression of miR-155 in the bone marrow niche regulates hematopoiesis in an NF-kappaB-dependent manner. Cell stem cell. 2014; 15:51–65.

37. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001; 25:402–408.