Bcl2 Signaling Pathways Dendritic Cells by Targeting YWHAZ And

Multiple Tumor-Associated MicroRNAs Modulate the Survival and Longevity of Dendritic Cells by Targeting YWHAZ and Bcl2 Signaling Pathways This information is current as of September 26, 2021. Siping Min, Xue Liang, Miaomiao Zhang, Yuan Zhang, Shiyue Mei, Jinzhe Liu, Jingyi Liu, Xiaomin Su, Shuisong Cao, Xueqing Zhong, Yueming Li, Jiatan Sun, Qiaofei Liu, Xingran Jiang, Yongzhe Che and Rongcun Yang

J Immunol 2013; 190:2437-2446; Prepublished online 25 Downloaded from January 2013; doi: 10.4049/jimmunol.1202282 http://www.jimmunol.org/content/190/5/2437 http://www.jimmunol.org/

Supplementary http://www.jimmunol.org/content/suppl/2013/01/25/jimmunol.120228 Material 2.DC1 References This article cites 46 articles, 24 of which you can access for free at: http://www.jimmunol.org/content/190/5/2437.full#ref-list-1 by guest on September 26, 2021 Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2013 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Multiple Tumor-Associated MicroRNAs Modulate the Survival and Longevity of Dendritic Cells by Targeting YWHAZ and Bcl2 Signaling Pathways

Siping Min,*,1 Xue Liang,*,1 Miaomiao Zhang,* Yuan Zhang,* Shiyue Mei,* Jinzhe Liu,* Jingyi Liu,* Xiaomin Su,* Shuisong Cao,* Xueqing Zhong,* Yueming Li,† Jiatan Sun,† Qiaofei Liu,* Xingran Jiang,* Yongzhe Che,* and Rongcun Yang*,‡

Tumors use a wide array of immunosuppressive strategies, such as reducing the longevity and survival of dendritic cells (DCs), to diminish immune responses and limit the effect of immunotherapy. In this study, we found that tumors upregulate the expression of multiple microRNAs (miRNAs), such as miR-16-1, miR-22, miR-155, and miR-503. These tumor-associated miRNAs influenced the survival and longevity of DCs by affecting the expression of multiple molecules that are associated with Downloaded from apoptotic signaling pathways. Specifically, miR-22 targeted YWHAZ to interrupt the PI3K/Akt and MAPK signaling pathways, and miR-503 downregulated Bcl2 expression. The result of the increased expression of miR-22 and miR-503 in the tumor- associated DCs was their reduced survival and longevity. Thus, tumor-associated miRNAs can target multiple intracellular signaling molecules to cause the apoptosis of DCs in the tumor environment. Use of miR-22 and miR-503 as inhibitors may therefore represent a new strategy to improve DC-based immunotherapies against tumors. The Journal of Immunology, 2013, 190: 2437–2446. http://www.jimmunol.org/

endritic cells (DCs), including conventional DCs and derived factors (3), DC apoptosis has also been observed in can- plasmacytoid DCs, are professional APCs that are critical cers, and the process parallels the induction of immunosuppres- D for the induction of adaptive immunity and tolerance. sion. A massive reduction in DC numbers has been observed in However, mature DCs also undergo apoptosis in the different tissue lymph node drainage from tumors, but not other lymph nodes, and organs, especially the lymph node (1). DC apoptosis is an in multiple cancers (4–7). In tumor lymph nodes, tumor-derived important event that regulates the balance between tolerance and TGF-b can induce DC apoptosis (8), and regulatory T cells can immunity; many molecules participate in this process, including also directly interact with tumor Ag-bearing DCs to induce their by guest on September 26, 2021 TNF-related activation-induced cytokine (TRANCE), CD154, apoptosis (9). FASL, amyloid peptide, TRAIL, LPS, type I IFN, leptin, and The apoptosis of DCs can be regulated by extrinsic and intrin- CCR7 (reviewed in Ref. 2). Defects in DC apoptosis can trigger sic pathways (10). The ratios between the antiapoptotic Bcl2/ autoimmunity, whereas environments in which the apoptosis of Bcl-xL molecules and the proapoptotic Bax/Bak molecules DCs occurs are immunosuppressive because they promote reg- determine the lifespan of different DC subsets. A lower ratio of ulatory T cell generation and functional impairment of DCs. antiapoptotic Bcl2/Bcl-xL to proapoptotic Bax/Bak has been Several different death receptors have been identified on DCs, observed in shorter lived mDCs compared with longer lived including Fas (CD95) and the TNF and TRAIL receptors. Al- plasmacytoid DCs (11). Transfection with Bcl2 or Bcl-xL pro- though regulatory DCs have been shown to be induced by tumor- longs the survival of mouse primary mDCs in vitro, and deletion of Bcl2 accelerates DC apoptosis in vivo (11). Because AKT *Department of Immunology, Nankai University School of Medicine, Nankai Uni- downregulation correlates with Bcl2 downregulation and DC versity, Tianjin 300071, China; ‡Key Laboratory of Bioactive Materials, Ministry of death, the PI3K/AKT pathway also plays an important role in the † Education, Nankai University, Tianjin 300071, China; and Tianjin “254” Hospital, DC lifespan. Indeed, AKT deficiency leads to defective DC ac- Nankai University, Tianjin 300071, China tivation and survival (12), and inhibition of PI3K antagonizes 1S.M. and X.L. contributed equally to this work. DC survival mediated by CD40, CpG, LPS, TRANCE, TNF Received for publication August 21, 2012. Accepted for publication December 27, 2012. superfamily member 11, and PGE2 (2, 13). Cell survival has also been found to be dependent on the ERK pathway in several This work was supported by National Science Foundation of China Grants 91029736, 30771967, and 30872315, the Ministry of Science and Technology (863 Program, cellular models, whereas the activation of p38 and JNK promotes Grant 2008AA02Z129), the National Theme Program of China (863 Program, Grant apoptosis (14). Interestingly, these apoptotic pathways may be 2011AA020116), and National Key Scientific Program Grant 2011CB964902. regulated by microRNAs (miRNAs), including miR-16-1 (15), Address correspondence and reprint requests to Rongcun Yang, Department of Im- miR-155 (16), miR-21 (17), and miR-451 (18). miRNAs are munology, Nankai University School of Medicine, Nankai University, Nankai District, Weijing Road No. 94, Tianjin 300071, China. E-mail address: [email protected] noncoding small RNAs that regulate gene expression and cell The online version of this article contains supplemental material. growth and differentiation. It is thought that miRNAs have a Abbreviations used in this article: BMDC, bone marrow–derived dendritic cell; DC, central role in regulating the apoptosis of DCs. In this study, we dendritic cell; miRNA, microRNA; moDC, monocyte-derived dendritic cell; qRT- demonstrate that the tumor-mediated miRNAs miR-22 and miR- PCR, quantitative real-time PCR; siRNA, small interfering RNA; TRANCE, TNF- 503 affect the survival and longevity of DCs. Specifically, in related activation-induced cytokine; UTR, untranslated region. tumor environments, these miRNAs target the YWHAZ and Bcl2 Copyright Ó 2013 by The American Association of Immunologists, Inc. 0022-1767/13/$16.00 pathways and promote DC apoptosis. www.jimmunol.org/cgi/doi/10.4049/jimmunol.1202282 2438 MODULATION OF DC SURVIVAL BY TUMOR-ASSOCIATED miRNAs

Materials and Methods (siRNAs) and the negative control siRNA were purchased from Santa Cruz Mice Biotechnology. BLOCK-iT fluorescent oligonucleotides were purchased from Invitrogen. The 39 untranslated region (UTR) of YWHAZ and Bcl2 Four- to six-wk-old female C57BL/6 and BALB/c mice (Beijing Animal was cloned from mouse spleen cell genomic DNA via PCR. Primers Center) were maintained in a pathogen-free animal facility for at least 1 wk containing the restriction enzymes XhoI and NotI were used so that the prior to use. Experiments were performed in accordance with the institu- sequences could be cloned into the siCheck-2 luciferase reporter vector tional guidelines. Various s.c. tumor models, including Lewis lung carci- (Promega). The YWHAZ and Bcl2 39 UTR mutants were generated using noma (American Type Culture Collection) and B16 melanoma in C57BL/6 four primers according to a previous method (21). Mutations were con- mice and CT-26 colon carcinoma (American Type Culture Collection) in firmed by DNA sequencing. YWHAZ and Bcl2 were directly cloned into BALB/c mice, were used in this study. The number of tumor cells injected pcDNA3.1 from total spleen cell RNA using the pcDNA3.1 TOPO kit. s.c. for each model was determined based on the ability of the cells to form The sequences of the miRNAs were obtained from the miRbase (http:// a tumor 1.5 cm in diameter within 3–4 wk injection. microrna.sanger.ac.uk/sequences/). The 39 UTR sequences were obtained from the National Center for Biotechnology Information (http://www.ncbi. Cell lines, human monocyte-derived dendritic cells, and nlm.nih.gov/sites/entrez/). Primer 3 software was used for primer design murine bone marrow–derived dendritic cells (http://frodo.wi.mit.edu/). The primers used are listed in Supplemental Table I. The following cell lines were purchased from the American Type Culture Collection: murine monocyte/macrophage RAW264.7, murine melanoma Transfection B16, murine colon carcinoma CT-26, mouse fibroblast L, human lung carcinoma PG, human breast carcinoma MCF-7, human cervical carcinoma For the miRNA mimics, inhibitors, siRNAs, and negative control oli- HeLa, human monocyte U937, human embryonic kidney 293T, and human gonucleotides, cells were transfected with the indicated oligonucleotides lung fibroblast WI38. Ovarian carcinoma 1D8 was provided by K.F. Ruby (100 nM) using the Entranster-R system (Engreen Biosystem) according to the manufacturer’s instructions. For some of the constructs, the cells were (University of Kansas Medical Center). All cells were maintained in Downloaded from complete media, which consisted of RPMI 1640 supplemented with 10% transfected using a nucleofection approach according to the manufacturer’s FCS and 1% penicillin and streptomycin. protocol (Amaxa Biosystems, Berlin, Germany). miRNA mimics, inhib- Human monocyte-derived DCs (moDCs) were prepared from monocytes itors, siRNAs, and negative control oligonucleotides were purchased from as previously described (19). Briefly, after Ficoll-Hypaque separation, pe- RiboBio (Guangzhou, China). Silencing of the target molecules was con- ripheral blood cells were resuspended in complete media. The cells were firmed by RT-PCR after transfection for 24 h. The primers used in the incubated for 1 h at 37˚C, and the nonadherent cells were removed by gentle siRNA experiments are listed in Supplemental Table I. pipetting. The adherent cells were cultured in RPMI 1640 containing 10%

miRNA array http://www.jimmunol.org/ FCS, 1000 U/ml GM-CSF (R&D Systems), and 1000 U/ml IL-4 (R&D Systems) for 7 d. Expression levels of miRNAs in 1D8 mouse ovarian tumor– and CT-26 Murine bone marrow–derived dendritic cells (BMDCs) were prepared as colon cancer–associated BMDCs or control BMDCs were analyzed using described in our previous work (20). Briefly, bone marrow cells were col- Exiqon miRNA arrays (Vedbaek, Denmark). Total RNA was isolated using lected by removing the femurs of mice, cutting off each end, and flushing TRIzol reagent (Invitrogen Life Technologies). The concentration and out the bone marrow with RPMI 1640 using a syringe. The pooled cells purity of the RNA were determined using NanoDrop ND-1000. The miR- were harvested by centrifugation at 600 3 g for 10 min and resuspended in NAs were labeled using the miRCURY array power labeling kit (Exiqon). 2 ml ACK buffer for 5 min at room temperature to lyse RBCs. The cells miRNA array hybridization was performed, and unbound miRNA labels were then washed in complete media, and DCs were generated by culturing were washed away using the miRCURY array wash buffer kit (Exiqon). The the cells in media containing 500 U/ml GM-CSF (R&D Systems) for 7 d. arrays were scanned on an Axon 4000B scanner (Molecular Devices), and

the signal intensity was determined using GenePix Pro 6.0 software (Mo- by guest on September 26, 2021 Flow cytometric analysis lecular Devices). miRNAs with a 1.5-fold increase or decrease in expression Murine BMDCs were collected in ice-cold PBS and incubated with the were regarded as differentially expressed. following anti-mouse Abs: FITC-, PE- or allophycocyanin-conjugated Gene chip array anti-mouse CD11c (N418), B220 (RA3-6B2), CD80 (16-10A1), CD86 (GL1), CD40 (5C3), CD11b (M1/70), CD4 (L3T4), or CD8a (Ly-2) Abs. A gene chip analysis was performed on RNA from untreated BMDCs and Human moDCs were collected in ice-cold PBS and incubated with the BMDCs cocultured with 1D8 tumor cells using a transwell plate. Total following anti-human Abs: FITC-, PE- or allophycocyanin-conjugated RNA was isolated from DCs using TRIzol reagent (Invitrogen) followed anti-human CD11c (3.9), CD86 (IT2.2), CD80 (2D10.4), CD40 (HB14), or by RNA cleanup with an RNeasy Mini kit (Qiagen). Processing of the CD11b (ICRF44). All of the Abs used in the study were purchased from samples was performed following Affymetrix specifications. Fluores- BD Biosciences. The cells were stained and resuspended in PBS with 1% cence was detected using the Hewlett-Packard G2500 GeneArray scanner, paraformaldehyde and 1% FCS and kept at 4˚C prior to flow cytometric and image analysis of each gene chip was performed with the Microarray analysis (FACScan; Becton Dickson). Isotype-matched control mAbs were Suite 5.0 software from Affymetrix using the standard default settings. used as negative controls for all analyses. Transcripts that were consistently significant in at least two of the four For apoptotic analysis, the cells were stained with annexin V and pro- iterative comparisons were selected for the final candidate list. Any pidium iodide according to the manufacturer’s instructions. Samples were transcript that showed at least a 2-fold change in expression level (ratios examined by FACScan. are presented as log2) between the experimental sample and the control RNA isolation and quantitative real-time PCR sample was considered significant. Total RNA was extracted using TRIzol reagent, and reverse transcription Two-dimensional gel electrophoresis was performed with the Moloney murine leukemia virus reverse tran- Two-dimensional gel electrophoresis was performed by the TEDA School scriptase according to the manufacturer’s protocol. For gene expression of Biological Science and Biotechnology, Nankai University. Total protein analysis, quantitative real-time PCR (qRT-PCR) was performed using the concentrations from RAW264.7 cells transfected with different miRNAs or QuantiTect SYBR PCR kit with a specific set of primers according to the control oligonucleotides were determined according to the Bradford method manufacturer’s instructions (Qiagen). For mature miRNAs, qRT-PCR was (Bio-Rad Laboratories, Hercules, CA) following the manufacturer’s pro- performed using a standard TaqMan PCR kit and an Applied Biosystems tocol. Proteins on the two-dimensional gels were visualized by silver staining 7900 HT sequence detection system. Amplification of U6 small RNA was for pattern comparisons, and protein spots were selected for quantitative performed to detect mature miRNAs, and amplification of GAPDH was analysis when they showed a reduction $2-fold that of the control sample. performed on each experimental sample as an endogenous control. Fold Selected protein spots were digested, desalted with ZipTip (Millipore), and ΔΔ changes were calculated using the Ct method according to the manu- subjected to analysis by MALDI-TOF MS. The peptide mass fingerprintings facturer’s protocol (Applied Biosystems). All reactions were run in trip- obtained for each protein of interest were searched against the National licate. The primer sequences used are listed in Supplemental Table I. Center for Biotechnology Information nonredundant database using the MASCOT search engine (http://www/matrixscience.com). miRNA, small interfering RNA, gene, and 39 untranslated region expression constructs Luciferase reporters miRNA mimics, miRNA inhibitors, and control oligonucleotides were Cells were plated in a 48-well plate at a density of 4 3 104 cells/well in purchased from Dharmacon. YWHAZ and Bcl2 small interfering RNAs 250 ml culture medium 24 h prior to transfection; the cells were then The Journal of Immunology 2439 cotransfected with the indicated expression plasmids, the siCHECK-2 re- (Life Technologies) for 24 h. For in vitro apoptosis analysis, murine porter plasmid, and control plasmids (Promega) using Lipofectamine reagent BMDCs or human moDCs were harvested, transfected, and then coincu- (Invitrogen) or Nucleofector technology (Amaxa Biosystems) according bated with the murine or human tumor cells in a transwell plate for 3 d. to the manufacturer’s recommendations. Renilla and firefly luciferase Specifically, 2 3 106 DCs were placed into 24-well plates in 1 ml medium, activities were measured 24 h after transfection using a dual luciferase and 2 3 105 tumor cells were placed into the inserts on the top of each kit (Promega). well. As controls, DCs were coincubated with murine or human fibroblasts or medium alone and placed in the inserts. The cells were also cultured Western blot analysis with TGF-b, TNF-a, or under conditions of GM-CSF withdrawal. Apo- Western blot analysis was performed according to our previous protocol ptotic cells were assessed using an annexin V binding assay with pro- (21). Hybridizations with primary Abs were performed for 1 h at room pidium iodide staining and an FITC-conjugated TUNEL reaction mixture temperature in blocking buffer. The protein/Ab complexes were detected that was provided with an in situ cell death detection kit (Roche Diag- using peroxidase-conjugated secondary Abs (Boehringer Mannheim) and nostics, Indianapolis, IN). ECL (Amersham Biosciences). Rabbit anti-YWHAZ (Sigma-Aldrich), anti-Bcl2 (Cell Signaling Technology), anti-Bax (Sigma-Aldrich), anti- In vivo experiments AKT (Cell Signaling Technology), anti-FOXO3 (Sigma-Aldrich), anti– To generate the mouse tumor model, B16 cells (0.5 3 106 in 100 ml PBS) phospho-AKT (Cell Signaling Technology), and anti–phospho-FOXO3 were inoculated s.c. into the inguinal region backs of C57BL/6 mice. (Cell Signaling Technology) were used in addition to the following Abs Tumors were allowed to develop for 9 d, and mice were then randomly purchased from Bios Biosynthesis Biotechnology (Beijing, China): anti- divided into different experimental groups (six mice per group). actin, anti-p38, anti-JNK, anti-ERK, anti–phospho-p38, anti–phospho-JNK, To analyze DC survival in vivo, BMDCs were transfected with differ- anti–phospho-ERK, and an HRP-labeled goat polyclonal Ab against rabbit ent miRNAs, labeled with CFSE and injected s.c. (1 3 106 cells/mouse) IgG (H+L). into the footpads of syngeneic C57BL/6 mice bearing B16 tumors. The

In vitro experiments draining popliteal lymph nodes were harvested at various time points after Downloaded from injection. The total numbers of lymph node cells were counted, and the For in vitro preparation of tumor-associated DCs, 2 3 105 murine or human percentage of CFSE+ DCs was analyzed by flow cytometry. The proportion tumor cells were cocultured with 2 3 106 murine BMDCs or human of CFSE+ DCs in the draining lymph nodes of each mouse was calculated. moDCs in a 24-well transwell plate in RPMI 1640 medium supplemented The effect of tumor-associated miRNAs on DC-based immunotherapy with 3% FCS (Life Technologies) and 1% penicillin and streptomycin against tumors was assessed according to the previously reported protocol http://www.jimmunol.org/ by guest on September 26, 2021

FIGURE 1. Tumor cells induce DC apoptosis. (A and B) Mouse tumor cells induced the apoptosis of murine BMDCs. BMDCs were cocultured with murine melanoma B16F10 cells, colon carcinoma CT-26 cells, ovarian carcinoma 1D8 cells, RAW264.7 monocytes, or mouse L fibroblast cells (mFB) in a transwell for 72 h. B16F10 in FACS data indicated the BMDCs cocultured with B16F10. The cells were stained with annexin V and propidium iodide and a FITC-conjugated TUNEL reaction mixture that was provided with an in situ cell death detection kit (Roche Diagnostics) according to the manufacturer’s instructions. Original magnification 340. Samples were examined by FACScan and fluorescence microscopy, respectively. (C) Mouse B16 tumor cells induced the apoptosis of BMDCs in vivo. The numbers in the gates indicate the proportion of CFSE+ cells in the draining lymph nodes pooled from B16 tumor mice (six mice/group) injected with 1 3 106 CFSE-labeled BMDCs at the indicated times. The mouse B16 tumor model was generated according to the protocol described in Materials and Methods. B16, the cells from the pooled draining lymph nodes of mice bearing B16 tumors; tumor-free, the cells from the draining lymph nodes of tumor-free mice. (D and E) Human tumor cells induced the apoptosis of human moDCs. Human moDCs were cocultured with human breast carcinoma MCF-7 cells, lung carcinoma PG cells, cervical carcinoma HeLa cells, U937 monocytes, or WI38 human fibroblast cells (hFB) in a transwell for 72 h. Original magnification 340. HeLa in FACS data indicated the moDCs cocultured with HeLa. The results shown are representative of three independent experiments. *p , 0.05. Ctr., Cells from pooled draining lymph nodes from mice (six mice/group) injected with 1 3 106 unlabeled BMDCs. 2440 MODULATION OF DC SURVIVAL BY TUMOR-ASSOCIATED miRNAs

(20). Mice bearing B16 tumors were randomly divided into different study, when DCs were cocultured with different types of mouse experimental groups (six mice per group) and treated via intratumoral or human tumors, the number of apoptotic and dead DCs signif- injection with the transduced DCs. Groups treated with PBS only or icantly increased compared with DCs cultured with only medium untreated DCs were used as controls. The diameter of the perpendicular , tumor was measured with a caliper to assess tumor size. Mice were sac- or fibroblasts (p 0.05); however, the tumor cells showed dif- rificed when they exhibited signs of distress or when the total tumor vol- ferent abilities to induce apoptosis (Fig. 1A, 1B). In in vivo ex- ume was .3000 mm3. periments, B16 tumors also promoted the apoptosis of BMDCs as compared with control (mean 6 SEM, p , 0.01), consistent Statistical analysis with other reports (26) (Fig. 1C). Similarly, human moDCs were Data were presented as mean 6 SD or mean 6 SEM. Statistical com- sensitive to MCF-7 human breast carcinoma, PG lung carcinoma, parisons were made by one-way ANOVA followed by the post hoc Tukey– and HeLa cervical carcinoma cell–mediated apoptosis (Fig. 1D, Kramer multiple comparison test and a Student t test. A two-sided p value ,0.05 was considered statistically significant. 1E), consistent with other reports (27). To investigate the miRNAs that may be involved in regulating the survival and longevity of DCs in tumor microenvironments, Results we first analyzed the expression of miRNAs in tumor-associated Tumor-associated miRNAs regulate the survival and longevity BMDCs. GenoExplorer miRNA arrays (http://www.ncbi.nlm. of DCs nih.gov/geo/query/acc.cgi?acc=GSE42722, GPL16352, National Cancers such as breast cancer (22, 23), hepatocellular carcinoma Center for Biotechnology Information tracking system no. (24), and melanoma (25) are associated with DC apoptosis. In this 16691611) showed that some miRNAs, such as miR-16-1, miR- Downloaded from http://www.jimmunol.org/ by guest on September 26, 2021

FIGURE 2. Expression of miRNAs in tumor-associated murine BMDCs. (A) miRNA expression in tumor-associated BMDCs. BMDCs were generated and cocultured with the CT-26 mouse colon cancer or 1D8 mouse ovarian cancer cell lines in a 24-well transwell plate for 24 h. The expression of miRNAs in the tumor-associated BMDCs was analyzed using a GenoExplorer miRNA array chip according to the protocol described in Materials and Methods; 16, 21, 22, 142-5p, 146a, 155, and 503 are miR-16-1, miR-21, miR-22, miR-142-5p, miR-146a, miR-155, and miR-503 respectively. (B) Tumor-mediated miRNA expression was further confirmed in CT-26– and B16-associated BMDCs by qRT-PCR. BMDCs were cocultured with B16 or CT-26 cells, and miRNA expression was detected at the indicated times according to the protocol described in Materials and Methods.(C) Expression of miRNAs in the draining lymph nodes from the B16 and CT-26 mouse tumor models. The DCs were isolated from the draining lymph nodes of CT-26 and B16 or control (Ctr.) mice using Dynabeads (Invitrogen) coated with an anti-mouse CD11c Ab (MBL International, Woburn, MA) according to the manufacturers’ instructions. The CT-26 and B16 mouse tumor models were generated according to the protocol described in Materials and Methods.*p , 0.05, **p , 0.01. R.E., Relative expression. The Journal of Immunology 2441

21, miR-22, miR-142-5p, miR-146a, miR-155, and miR-503, miR-503 on modulating DC apoptosis. As shown in Fig. 3, after were upregulated when the BMDCs were cultured with mouse s.c. delivery of BMDCs transfected with miR-22 and miR-503 CT-26 colon cancer cells or 1D8 ovarian cancer cells (Fig. 2A). inhibitors, the percentage of live cells from the draining lymph These upregulated miRNAs were also detected by qRT-PCR in node of murine B16 tumor model was significantly higher than BMDCs cultured with mouse tumor B16 cells. Notably, except that obtained from the lymph nodes of mice injected with control for miR-22 and miR-142-5p, the kinetics of the expression DCs (Fig. 3B, 3C). This disproportionate percentage of live cells of these miRNAs was different from the BMDCs cultured with was sustained for at least 5 d after delivery (Fig. 3B, 3C). These CT-26 cells (Fig. 2B), implying different mechanisms that are data suggest that the miR-22 and miR-503 inhibitors prolong DC involved in the 1D8- and CT-26–mediated modulation on miRNAs. lifespan. Importantly, the upregulated miRNAs were also found in DCs We also employed apoptotic models that are mediated by TNF- isolated from the draining lymph nodes of mice bearing B16 and a (28), TGF-b (29), or GM-CSF withdrawal (30) to further inves- CT-26 tumors (Fig. 2C). Significant differences were apparent tigate the effects of the tumor-associated miRNAs on the apoptosis as compared with control (mean 6 SD, p , 0.01). Thus, our of DCs. miR-22 and miR-503 mimics promoted the apoptosis of results suggest that the expression of miR-16-1, miR-21, miR-22, DCs upon TNF-a,TGF-b, and GM-CSF withdrawal, whereas miR-142-5p, miR-146a, miR-155, miR-503, and the previously miR-155 promoted DC apoptosis only when exposed to TNF-a reported miR-17-5p and miR-20 (21) may be upregulated by tumor- (not shown). miR-16 did not have an effect on DCs in these associated factors. apoptotic models (not shown), implying that multiple signaling Although there exists a possibility for the roles of the down- pathways are involved in the tumor-associated miRNA regulation regulated miRNAs in DC apoptosis, we next only assessed effects of DC apoptosis. Downloaded from of the tumor upregulated miRNAs on the tumor-induced apoptosis in vitro by coculturing murine B16 tumor cells with miRNAs or The YWHAZ-based apoptotic signaling pathway may be their inhibitor-transfected BMDCs with the demonstrated trans- directly or indirectly modulated by multiple tumor-associated fection (not shown). The results showed that tumor-associated miR- miRNAs 22 and miR-503 and the previously reported miR-16-1 (15) and Given the potential of miRNAs to regulate a large number of

miR-155 (16) significantly increased the number of apoptotic and cellular transcripts, we decided to take a broad approach to iden- http://www.jimmunol.org/ dead DCs, whereas their inhibitors resisted tumor-induced apo- tify targets and determine the mechanisms by which miRNAs ptosis of DCs (Fig. 3A and not shown). Because others have act to regulate DC apoptosis. The expression of intracellular shown the effect of miR-16-1 and miR-155 on mediating cellular proteins in the monocyte cell line RAW264.7 transfected with apoptosis (15, 16), we next focused on the effects of miR-22 and tumor-associated miR-16-1, miR-20a, miR-22, miR-155, or miR- by guest on September 26, 2021

FIGURE 3. Tumor-associated miRNAs induce the apoptosis of DCs. (A) Tumor-associated miR-16-1, miR-22, miR-155, and miR-503 induce the apoptosis of BMDCs in vitro. Murine BMDCs were transfected with miR-16-1, miR-22, miR-155, and miR-503 mimics or their inhibitors, and apoptotic cells were detected by propidium iodide and annexin V staining after the BMDCs were cocultured with B16 cells in a transwell plate for 3 d. (B and C) miR- 22 and miR-503 induce the apoptosis of BMDCs in vivo. The mice bearing B16 tumors were injected with 1 3 106 CFSE-labeled BMDCs in the footpads, and the draining lymph nodes were analyzed by FACScan at the indicated times. The BMDCs were transfected with miR-22 or miR-503 mimics or their inhibitors according to the protocol described in Materials and Methods. miR-22M and miR-503M indicate the BMDCs that were transfected with miR-22 and miR-503 mimics, respectively; miR-22I and miR-503I indicate the BMDCs that were transfected with miR-22 and miR-503 inhibitors, respectively. M1–M6 represent the different mice in each group. The numbers in the gates in (C) indicate the proportion of CFSE+ cells in the pooled draining lymph nodes of B16 tumor mice (six mice/group) injected with 1 3 106 CFSE-labeled BMDCs. **p , 0.01. Ctr., The pooled draining lymph nodes of tumor-free mice (six mice/group) injected with 1 3 106 labeled BMDCs. 2442 MODULATION OF DC SURVIVAL BY TUMOR-ASSOCIATED miRNAs

503 was investigated using two-dimensional gels and MALDI- Table II) and miR-22 inhibitors. Both YWHAZ 39 UTR mu- TOF MS. We found that these miRNAs could indeed down- tation and miR-22 inhibitors abolished the interaction between regulate the levels of multiple proteins, such as YWHAZ, ALB, miR-22 and the 39 UTRofYWHAZ(Fig.5B).Whentheex- HMOX1, LDHA, ANXA2, CCT4, ANXA5, TCP1, CCT2, ANXA1, pression of YWHAZ in tumor-associated DCs was assessed, we EEF2, VIM, PGK1, CCT7, TKT, Cofilin, CAP1, CFL1, P4HB, found that the DCs had reduced expression of YWHAZ (Fig. TPl1, ALDOA, PGAM1, CS, GLUD1, PKM2, and ENO1 (Fig. 5C). Thus, these data demonstrate that tumor-associated miR-22 4, Supplemental Fig. 1). In addition to predictions of miRNA- can inhibit the expression of YWHAZ by binding to its 39 UTR targeted molecules, multiple potential targets of miR-16-1, miR- region. 20a, miR-22, miR-155, and miR-503 were identified (Fig. 4, Previous studies have shown that YWHAZ can interact with Supplemental Fig. 1, and not shown). For example, miR-22 multiple molecules in different signaling pathways, such as AKT markedly reduced the expression of YWHAZ (Fig. 4, Sup- (32–36). Indeed, as shown in Fig. 5D, transfection with miR-22 plemental Fig. 1). Several molecules related to YWHAZ were reduced the phosphorylation of AKT, whereas miR-22 inhibitors also downregulated by tumor-associated miRNAs, including promoted the phosphorylation of AKT. We also observed that ALDOA, GLUD1, and ENO1 by miR-16, ANXA2 by miR-20a, miR-22 inhibitors induced the phosphorylation of FOXO3 after CAP1 by miR-22, ANXA 2 by miR-20a, and ENO1 by miR- transfection for 5 min. Because these phosphorylated residues 503 (Supplemental Fig. 1). These data suggest that the tumor- subsequently sequestered FOXO3 to YWHAZ to prevent apo- associated miRNAs affected the YWHAZ-based apoptotic signaling ptosis (37), miR-22 may promote the apoptosis of DCs by re- pathway by reducing the expression of YWHAZ-associated mol- ducing the expression of YWHAZ and affecting the activity of ecules. AKT and FOXO3. Downloaded from YWHAZ has been shown to promote ERK/MAPK activation Tumor-associated miR-22 targets YWHAZ but inhibit the activation of JNK and p38 MAPK (31). Indeed, as YWHAZ plays an important role in the assembly of signaling shown in Fig. 5D, administration of miR-22 mimics remarkably complexes required for the activation of pathways downstream caused the downregulation of ERK phosphorylation and the up- of growth receptors (31). Interestingly, miR-22 not only reduced regulation of JNK and p38 phosphorylation; conversely, ERK was

the protein levels of YWHAZ but also its transcript levels activated and the phosphorylation of JNK and p38 was inhibited http://www.jimmunol.org/ (Figs. 4, 5A). Further analyses demonstrated that miR-22 tar- by miR-22 inhibitors. The difference of p38 and ERK phosphor- geted the 39 UTR region of YWHAZ. We fused the 39 UTR ylation could be found after transfection for 5 min. JNK was also sequence of mouse YWHAZ to a luciferase reporter gene and remarkable phosphorylated after miR-22 transfection as com- found significant repression of luciferase activity by miR-22, pared with control transfection. Because AKT and MAPK ac- suggesting a direct effect (Fig. 5B). We also performed experi- tivity is associated with the expression of apoptotic molecules, ments with mutated target mRNA sequences (Supplemental such as Bcl2 and Bax, we next determined the effect of miR-22 by guest on September 26, 2021

FIGURE 4. Both miR-22 and miR-155 downregulate the expression of YWHAZ. (A) The levels of the YWHAZ protein (spot 1201) in miR-22– and miR-155–transfected RAW264.7 cells. miR-22–, miR-155–, or control oligonucleotide (Ctr)–transfected RAW264.7 cells were analyzed by two-dimen- sional gel electrophoresis. 1201, protein spot number. (B and C) The selected spot 1201 is the YWHAZ protein. Spot 1201 was digested and subjected to analysis by MALDI-TOF MS. The Journal of Immunology 2443 Downloaded from http://www.jimmunol.org/

FIGURE 5. miR-22 targets YWHAZ. (A) miR-22 downregulates the expression of YWHAZ. Murine BMDCs were transfected with different concentrations of miR-22 or its inhibitor. The transcript and protein levels of YWHAZ were detected by RT-PCR, qRT-PCR, and Western blot, respectively. R.E., relative expression. (B) The luciferase activity demonstrates that miR-22 targets the 39 UTR of YWHAZ. The 39 UTR of YWHAZ by guest on September 26, 2021 (Wt.UTR) or the mutated 39 UTR (Mut.UTR) were used to construct luciferase reporter vectors that were cotransfected with miR-22 (miRNA) its inhibitor or control oligonucleotides (Oligos.ctr) into 293T cells. Luciferase activity was measured after 48 h according to the protocol describedin Materials and Methods. The data are expressed as the relative luciferase activity of the control samples that were cotransfected with an equal concentration of negative control vectors and are presented as the means 6 SD from triplicate tests. The luciferase activity in 293T cells cotransfected with YWHAZ wild-type 39 UTR and control oligonucleotides (Oligos.ctr) was arbitrarily set at 1. Error bars represent the SD. *p , 0.05, **p , 0.01. (C) Tumor-associated BMDCs had reduced expression of YWHAZ. Murine BMDCs were cocultured with murine 1D8 ovarian carcinoma cells, CT-26 colon carcinoma cells, B16 melanoma cells, and murine fibroblast cells (mFB) or with different concentrations of tumor supernatant. The transcript and protein levels of YWHAZ were detected by qRT-PCR and Western blot after 24 h; 1/2, 1/4, 1/8, and 1/16 indicate that BMDCs were cultured with 50, 25, 12.5, and 6.25% tumor supernatant, respectively. 0, without tumor supernatant. The normalized YWHAZ ex- pression for control cells was set at 1. (D) miR-22 reduced the activity of AKT, FOXO3, and ERK, but p38 and JNK activity was enhanced. BMDCs were transfected with miR-22 or its inhibitors, and phospho-proteins were detected at the indicated times by Western blot according to the protocol described in Materials and Methods.(E) miR-22 inhibitors upregulated the expression of YWHAZ and Bcl2, but Bax expression was downregulated. BMDCs were transfected with miR-22, miR-155, or their inhibitors. The protein levels of Bcl2 and Bax were detected by Western blot after transfection for 24 h. The normalized YWHAZ, Bcl2, or Bax expression for control cells was set at 1. R.E., Relative expression; RLU, relative light unit.

on Bcl2 and Bax. Transfection with miR-22 reduced the ex- The Bcl2-based apoptotic signaling pathway may be pression of Bcl2 and enhanced the expression of Bax, whereas potentially modulated by multiple tumor-associated miRNAs the miR-22 inhibitor inhibited the expression of Bax and en- We analyzed the gene expression profile of BMDCs using a gene hanced the expression of Bcl2 (Fig. 5E). Thus, the ERK/MAPK chip after exposure of the DCs to tumor cells. Approximately 2701 signaling pathway is involved in the miR-22–mediated apoptosis genes were significantly downregulated in the tumor-associated of DCs. BMDCs compared with the control BMDCs (not shown). Com- Taken together, both the AKT/FOXO3 and ERK/MAPK sig- bined with analyses using the TargetScan program, we predicted naling pathways may be involved in the miR-22–mediated apo- the potential target molecules of tumor-associated miR-16-1, miR- ptosis of BMDCs. Additionally, we found that miR-155, which 17-5p, miR-20a, miR-21, miR-22, miR-142-5p, miR-146a, miR- induced DC apoptosis (Ref. 16 and Fig. 3), also downregulated the 155, and miR-503. As shown in Supplemental Fig. 2, some of expression of YWHAZ and affected the expression of Bcl2 and these downregulated genes, such as BCL2, MTF1, DNAJC7, Bax (Fig. 5E). Thus, the AKT/FOXO3 and ERK/MAPK signaling PRPF4B, FGD4, CDK6, MAP3K3, MAP3K9, KLF4, PPP3CA, pathways may also play a role in the miR-155–mediated apoptosis FKBP1A, MAPK14, DPYSL2, TRIB2, RCN2, ITGA4, TRIB2, of BMDCs. RGS2, PLXNA2, PLCB1,andSACS, may be potentially modulated 2444 MODULATION OF DC SURVIVAL BY TUMOR-ASSOCIATED miRNAs by the tumor-associated miRNAs. These genes are directly or miR-503 could regulate the apoptosis of DCs by affecting Bcl2 indirectly related to Bcl2 (Supplemental Fig. 2), which is a mol- expression. ecule that favors cell survival by inhibiting cell death (38). Thus, Inhibitors of miR-22 and miR-503 promote the DC-based the Bcl2-based apoptotic signaling pathway may be potentially immunotherapy against tumors modulated by multiple tumor-associated miRNAs. Both Bcl2 and YWHAZ play a critical role in regulating the ap- Tumor-associated miR-503 targets Bcl2 optosis of different types of cells (30, 40–42). Because miR-22 and Next, we investigated the effect of the tumor-associated miRNAs miR-503 could affect the survival and longevity of DCs in the on the predicted target molecules. We found that miR-503 and miR- tumor environment by targeting YWHAZ and Bcl2, these miR- 16-1 could potentially target Bcl2 (Fig. 6A). Indeed, as shown in NAs might be promising targets for DC-based immunotherapies Fig. 6B and 6C, miR-503 downregulated Bcl2 at both the tran- against tumors. To test the effects of miR-22 or miR-503 on an script and protein level. miR16-1 also reduced the expression of ex vivo DC vaccine against tumors, we tested the efficacy of Bcl2 (Fig. 6B), consistent with a previous report (39). We then DCs transfected with miR-22 and miR-503 inhibitors on an estab- fused the 39 UTR sequence of mouse Bcl2 to a luciferase reporter lished tumor. As shown in Fig. 7, tumor growth was slower in gene and found a significant repression of luciferase activity by mice vaccinated with DCs transfected with miR-22 and miR-503 miR-503 compared with control vectors, suggesting a direct effect inhibitors, whereas mice vaccinated with DCs transfected with (Fig. 6D). We performed further experiments with mutated target miR-22 and miR-503 mimics exhibited faster tumor growth. mRNA sequences (Supplemental Table II) and miR-503 inhib- Significant differences in tumor growth between control and miR- itors; as expected, both the Bcl2 39 UTR mutation and miR-503 22 or miR-503 inhibitor transfected groups or between control Downloaded from inhibitors abolished the interaction between miR-503 and the 39 and miR-22 or miR-503 mimics transfected group were ap- UTR of Bcl2. When the expression of Bcl2 was assessed in tumor- parent after 12 d (p , 0.05). There were significantly differences associated DCs, these tumor-associated DCs had reduced ex- in tumor weight as compared with control (mean 6 SEM, p , pression of Bcl2 (Fig. 6E). These data suggest that in addition 0.01). Consistent with other reports (43), siRNAs targeting Bcl2 to what has been previously reported for miR16-1 (39), tumor- and YWHAZ in DCs also suppressed the DC-based immuno- 9 associated miR-503 may interact directly with the 3 UTR of Bcl2 therapy against tumors; conversely, Bcl2 and YWHAZ improved http://www.jimmunol.org/ to downregulate the expression of Bcl2. Thus, tumor-associated the DC-based immunotherapy against tumors (Fig. 7). by guest on September 26, 2021

FIGURE 6. miR-503 targets Bcl2. (A) Bcl2 is a potential target of miR-503. In addition to the downregulated genes in tumor-associated BMDCs shown in Supplemental Table II, miR-503 and miR-16-1 potentially target Bcl2 based on the prediction of TargetScan 5.1 (http://www.targetscan. org/cgi-bin/vert_50/targetscan.cgi?mirg=mmu-miR-223) PICTAR (http://pictar.bio.nyu.edu/), and MIRANDA (http://cbio.mskcc.org/cgi-bin/ mirnaviewer/mirnaviewer.pl). Cycle number 2071 indicates the 2071 genes that are downregulated in tumor-associated BMDCs. Cycle numbers 243 and 821 indicate the number of predicted target molecules by miR-16-1 and miR-503, respectively. (B and C) Both miR-503 and miR-16-1 downregulate the expression of Bcl2. BMDCs were transfected with different concentrations of miR-503 or miR-16-1. The transcript and protein levels of Bcl2 were detected by qRT-PCR and Western blot, respectively. (D) miR-503 can target the 39 UTR of Bcl2. The Bcl2 luciferase reporter vector (Wt.UTR) or mutant Bcl2 luciferase reporter vector (Mut1.UTR) was cotransfected with miR-503 mimics (miR503), miR-503 inhibitors, or control oligonucleotides (Oligos.ctr.) into 293T cells. Luciferase activity was measured after 48 h according to the protocol described in Materials and Methods. The data are expressed as the relative luciferase activity of control samples that were cotransfected with an equal concentration of negative control vectors and are presented as the means 6 SD from triplicate tests. The luciferase activity in 293T cells cotransfected with Bcl2 Wt. UTR and control oligonucleotides was arbitrarily set at 1. Error bars represent the SD. *p , 0.05, **p , 0.01. (E) The expression of Bcl2 was regulated by tumor cells. Murine BMDCs were cocultured with murine melanoma B16 cells, colon carcinoma CT-26 cells, ovarian carcinoma 1D8 cells, and murine fibroblast cells (mFB). The transcript and protein levels of Bcl2 were detected by qRT-PCR and Western blot. R.E., Relative expression. The Journal of Immunology 2445

FIGURE 7. miR-22 and miR-503 impair DC-based immunotherapy against tumors. (A and B)Tumorsgrewslowerinmice injected with DCs transfected with miR- 22 and miR-503 inhibitors. BMDCs were transfected with miR-22 inhibitors (miR- 22I), miR-503 inhibitors (miR-503I), or vectors containing Bcl2 or YWHAZ. Con- trol BMDCs were transfected using a mix- ture of mock siRNAs with empty vectors. (C and D) Tumors grew faster in mice im- munized with DCs transfected with tumor- associated miRNAs. BMDCs were trans- fected with miR-22 mimics (miR-22M), miR-503 mimics (miR-503M), Bcl2 siRNAs (Bcl2siR), or YWHAZ siRNAs (YWHAZ siR). The murine B16 tumor model was established according to the protocol de- scribed in Materials and Methods. Trans- fected BMDCs were injected into tumor Downloaded from tissueatdays3,9,and15(23 106 cells/ mouse). Tumor sizes were measured at the indicated times and statistically analyzed by ANOVA followed by the post hoc Tukey– Kramer multiple-comparison Test. Tumor weights were measured at day 28 after in- jecting tumor cells and statistically analyzed http://www.jimmunol.org/ using a Student t test.

Discussion cancers (4). However, the mechanism of tumor-mediated apo- In this study, we identified murine tumor-associated miRNAs that ptosis of DCs is unclear. Our results demonstrate that tumor cells affected the survival and longevity of DCs. First, tumor-associated induce the upregulation of multiple miRNAs, including miR-22, miR-22 reduced the survival of DCs by targeting YWHAZ, which miR-503, miR-16-1, and miR-155. A recent study by Cubillos- diminished the activity of AKT and FOXO3 to promote apoptosis. Ruiz et al. (46) reported that tumor-associated DCs had reduced by guest on September 26, 2021 Second, the targeting of miR-22 to YWHAZ inhibited ERK miR-155 levels. This may be caused from different time points activity, promoted the activation of JNK/p38, and affected the and/or different models. Indeed, miR-155 and also miR-503 are expression of Bcl2 and Bax. Third, tumor-associated miR-503 immediately upregulated, but their expression is remarkably down- directly targeted Bcl2 to promote the apoptosis of DCs. Finally, regulated with time extension upon exposure to tumors. However, tumor-associated miR-16, miR-22, miR-155, and miR-503 could these tumor-associated miRNAs, particularly miR-22 and miR- also affect the expression of multiple molecules related to apoptotic 503, reduce the survival and longevity of DCs by promoting their signaling pathways. Thus, tumor-associated miRNAs can target apoptosis. Other studies have also shown that miR-16 and miR- multiple intracellular signaling molecules to reduce the survival of 155 are involved in the apoptosis of tumor cells (15, 18) and DCs in the tumor environment. DCs (16) by targeting YWHAZ and Bcl2. Tumor cells may not only promote the survival of DCs and DC-based immunotherapies have been extensively explored for induce regulatory DCs to mediate immune tolerance (3) but also cancer treatments, but they have had limited success. A major cause the apoptosis of DCs. Several types of solid and blood problem is that DCs undergo apoptosis after injection. Thus, de- cancers, such as pancreatic, breast, prostate, hepatocellular, lung, livery of apoptosis-resistant DCs may improve the efficacy of DC head and neck carcinomas, and leukemia, are accompanied by immunotherapies. The decreased viability of the DCs could be impaired function and reduced numbers of DCs (23, 44, 45). explained by the reduced expression of the antiapoptotic proteins DCs are often depleted from the tumor site itself and also from Bcl2 and YWHAZ, which have previously been shown to control the circulation. The reduced numbers of DCs may be from the the longevity of murine DCs (30, 40–42). The administration of enhanced apoptosis. Indeed, our data showed that tumor-mediated siRNAs targeting these proapoptotic molecules in DCs increases miRNAs such as miR-22 and miR-503 promoted the apoptosis DC survival. Treatment of DCs with TRANCE, a prosurvival of DCs by targeting YWHAZ and Bcl2. Others also found that factor, results in better adjuvant properties. Our data show that tumor-derived TGF-b may induce DC apoptosis (8). This may DCs transfected with miR-22 and miR-503 inhibitors may im- be an important mechanism for tumor immune tolerance and prove antitumor response. However, the role of overexpressed escape. miRNAs on DC survival/apoptosis might only be a part of the DC Apoptotic pathways may be regulated by miRNAs, including conditioned by tumor cells. Others such as soluble and membrane- miR-16 (15), miR-155 (16), miR-21 (17), and miR-451 (18). The bound factors are likely involved, and the deleterious effects on various mediators released from tumor cells (e.g., TGF-b)flow DCs is not limited to their survival. into sentinel lymph nodes and affect the survival and longevity of DCs (8, 24, 25). DC numbers in the sentinel lymph nodes of Disclosures tumors have been found to be remarkably reduced in multiple The authors have no financial conflicts of interest. 2446 MODULATION OF DC SURVIVAL BY TUMOR-ASSOCIATED miRNAs

References 24. Um, S. H., C. Mulhall, A. Alisa, A. R. Ives, J. Karani, R. Williams, A. Bertoletti, and S. Behboudi. 2004. Alpha-fetoprotein impairs APC function and induces 1. Garg, S., A. Oran, J. Wajchman, S. Sasaki, C. H. Maris, J. A. Kapp, and J. Jacob. their apoptosis. J. Immunol. 173: 1772–1778. 2003. Genetic tagging shows increased frequency and longevity of antigen- 25. Pe´guet-Navarro, J., M. Sportouch, I. Popa, O. Berthier, D. Schmitt, and presenting, skin-derived dendritic cells in vivo. Nat. Immunol. 4: 907–912. J. Portoukalian. 2003. Gangliosides from human melanoma tumors impair 2. Kushwah, R., and J. Hu. 2010. Dendritic cell apoptosis: regulation of tolerance dendritic cell differentiation from monocytes and induce their apoptosis. J. versus immunity. J. Immunol. 185: 795–802. Immunol. 170: 3488–3494. 3. Schmidt, S. V., A. C. Nino-Castro, and J. L. Schultze. 2012. Regulatory dendritic 26. Kanto, T., P. Kalinski, O. C. Hunter, M. T. Lotze, and A. A. Amoscato. 2001. cells: there is more than just immune activation. Front Immunol 3: 274. Ceramide mediates tumor-induced dendritic cell apoptosis. J. Immunol. 167: 4. Poindexter, N. J., A. Sahin, K. K. Hunt, and E. A. Grimm. 2004. Analysis of 3773–3784. dendritic cells in tumor-free and tumor-containing sentinel lymph nodes from 27. Ishida, A., M. Ohta, M. Toda, T. Murata, T. Usui, K. Akita, M. Inoue, and patients with breast cancer. Breast Cancer Res. 6: R408–R415. H. Nakada. 2008. Mucin-induced apoptosis of monocyte-derived dendritic cells 5. Movassagh, M., A. Spatz, J. Davoust, S. Lebecque, P. Romero, M. Pittet, during maturation. Proteomics 8: 3342–3349. D. Rimoldi, D. Lie´nard, O. Gugerli, L. Ferradini, et al. 2004. Selective accu- + 28. Matsue, H., and A. Takashima. 1999. Apoptosis in dendritic cell biology. J. mulation of mature DC-Lamp dendritic cells in tumor sites is associated with Dermatol. Sci. 20: 159–171. efficient T-cell-mediated antitumor response and control of metastatic dissemi- 29. Kim, S. G., H. S. Jong, T. Y. Kim, J. W. Lee, N. K. Kim, S. H. Hong, and nation in melanoma. Cancer Res. 64: 2192–2198. Y. J. Bang. 2004. Transforming growth factor-b1 induces apoptosis through Fas 6. Kara, P. P., A. Ayhan, B. Caner, M. Gultekin, O. Ugur, M. F. Bozkurt, ligand-independent activation of the Fas death pathway in human gastric SNU- A. Usubutun, and A. Uner. 2009. Analysis of dendritic cells in sentinel lymph 620 carcinoma cells. Mol. Biol. Cell 15: 420–434. nodes of patients with endometrial and patients with cervical cancers. Int. J. 30. Chen, M., L. Huang, and J. Wang. 2007. Deficiency of Bim in dendritic cells Gynecol. Cancer 19: 1239–1243. contributes to overactivation of lymphocytes and autoimmunity. Blood 109: 7. Li, X., Y. Takahashi, K. Sakamoto, and T. Nakashima. 2009. Expression of 4360–4367. dendritic cell phenotypic antigens in cervical lymph nodes of patients with 31. Xing, H., S. Zhang, C. Weinheimer, A. Kovacs, and A. J. Muslin. 2000. 14-3-3 hypopharyngeal and laryngeal carcinoma. J. Laryngol. Otol. Suppl. 31: 5–10. proteins block apoptosis and differentially regulate MAPK cascades. EMBO J. 8. Ito, M., Y. Minamiya, H. Kawai, S. Saito, H. Saito, T. Nakagawa, K. Imai, 19: 349–358. b Downloaded from M. Hirokawa, and J. Ogawa. 2006. Tumor-derived TGF -1 induces dendritic 32. Garcia-Guzman, M., F. Dolfi, M. Russello, and K. Vuori. 1999. Cell adhesion cell apoptosis in the sentinel lymph node. J. Immunol. 176: 5637–5643. regulates the interaction between the docking protein p130Cas and the 14-3-3 9. Boissonnas, A., A. Scholer-Dahirel, V. Simon-Blancal, L. Pace, F. Valet, proteins. J. Biol. Chem. 274: 5762–5768. A. Kissenpfennig, T. Sparwasser, B. Malissen, L. Fetler, and S. Amigorena. 33. Powell, D. W., M. J. Rane, Q. Chen, S. Singh, and K. R. McLeish. 2002. + 2010. Foxp3 T cells induce perforin-dependent dendritic cell death in tumor- Identification of 14-3-3z as a protein kinase B/Akt substrate. J. Biol. Chem. 277: draining lymph nodes. Immunity 32: 266–278. 21639–21642. 10. Riedl, S. J., and G. S. Salvesen. 2007. The apoptosome: signalling platform of 34. Mils, V., V. Baldin, F. Goubin, I. Pinta, C. Papin, M. Waye, A. Eychene, and cell death. Nat. Rev. Mol. Cell Biol. 8: 405–413. B. Ducommun. 2000. Specific interaction between 14-3-3 isoforms and the

11. Chen, M., L. Huang, Z. Shabier, and J. Wang. 2007. Regulation of the lifespan in human CDC25B phosphatase. Oncogene 19: 1257–1265. http://www.jimmunol.org/ dendritic cell subsets. Mol. Immunol. 44: 2558–2565. 35. Birkenfeld, J., H. Betz, and D. Roth. 2003. Identification of cofilin and LIM- 12. Downward, J. 2004. PI 3-kinase, Akt and cell survival. Semin. Cell Dev. Biol. 15: domain-containing protein kinase 1 as novel interaction partners of 14-3-3z. 177–182. Biochem. J. 369: 45–54. 13. Chen, M., Y. H. Wang, Y. Wang, L. Huang, H. Sandoval, Y. J. Liu, and J. Wang. 36. Nellist, M., M. A. Goedbloed, C. de Winter, B. Verhaaf, A. Jankie, A. J. Reuser, 2006. Dendritic cell apoptosis in the maintenance of immune tolerance. Science A. M. van den Ouweland, P. van der Sluijs, and D. J. Halley. 2002. Identification 311: 1160–1164. and characterization of the interaction between tuberin and 14-3-3z. J. Biol. 14. Xia, Z., M. Dickens, J. Raingeaud, R. J. Davis, and M. E. Greenberg. 1995. Chem. 277: 39417–39424. Opposing effects of ERK and JNK-p38 MAP kinases on apoptosis. Science 270: 37. Dong, S., S. Kang, T. L. Gu, S. Kardar, H. Fu, S. Lonial, H. J. Khoury, F. Khuri, 1326–1331. and J. Chen. 2007. 14-3-3 integrates prosurvival signals mediated by the AKT 15. Tsang, T. Y., W. Y. Tang, J. Y. Chan, N. N. Co, C. L. Au Yeung, P. L. Yau, and MAPK pathways in ZNF198-FGFR1-transformed hematopoietic cells. S. K. Kong, K. P. Fung, and T. T. Kwok. 2011. P-glycoprotein enhances Blood 110: 360–369. radiation-induced apoptotic cell death through the regulation of miR-16 and Bcl- 38. Cory, S., and J. M. Adams. 2002. The Bcl2 family: regulators of the cellular life- by guest on September 26, 2021 2 expressions in hepatocellular carcinoma cells. Apoptosis 16: 524–535. or-death switch. Nat. Rev. Cancer 2: 647–656. 16. Lu, C., X. Huang, X. Zhang, K. Roensch, Q. Cao, K. I. Nakayama, B. R. Blazar, 39. Cimmino, A., G. A. Calin, M. Fabbri, M. V. Iorio, M. Ferracin, M. Shimizu, Y. Zeng, and X. Zhou. 2011. miR-221 and miR-155 regulate human dendritic S. E. Wojcik, R. I. Aqeilan, S. Zupo, M. Dono, et al. 2005. miR-15 and miR-16 cell development, apoptosis, and IL-12 production through targeting of p27kip1, induce apoptosis by targeting BCL2. Proc. Natl. Acad. Sci. USA 102: 13944–13949. KPC1, and SOCS-1. Blood 117: 4293–4303. 40. Hou, W. S., and L. Van Parijs. 2004. A Bcl-2-dependent molecular timer reg- 17. Carletti, M. Z., S. D. Fiedler, and L. K. Christenson. 2010. MicroRNA 21 blocks ulates the lifespan and immunogenicity of dendritic cells. Nat. Immunol. 5: 583– apoptosis in mouse periovulatory granulosa cells. Biol. Reprod. 83: 286–295. 589. 18. Bergamaschi, A., and B. S. Katzenellenbogen. 2012. Tamoxifen downregulation 41. Ouaaz, F., J. Arron, Y. Zheng, Y. Choi, and A. A. Beg. 2002. Dendritic cell de- of miR-451 increases 14-3-3z and promotes breast cancer cell survival and en- velopment and survival require distinct NF-kB subunits. Immunity 16: 257–270. docrine resistance. Oncogene 31: 39–47. 42. Gautier, E. L., T. Huby, F. Saint-Charles, B. Ouzilleau, M. J. Chapman, and 19. Rongcun, Y., F. Salazar-Onfray, J. Charo, K. J. Malmberg, K. Evrin, H. Maes, P. Lesnik. 2008. Enhanced dendritic cell survival attenuates lipopolysaccharide- K. Kono, C. Hising, M. Petersson, O. Larsson, et al. 1999. Identification of new induced immunosuppression and increases resistance to lethal endotoxic shock. HER2/neu-derived peptide epitopes that can elicit specific CTL against autolo- J. Immunol. 180: 6941–6946. gous and allogeneic carcinomas and melanomas. J. Immunol. 163: 1037–1044. 43. Nopora, A., and T. Brocker. 2002. Bcl-2 controls dendritic cell longevity in vivo. 20. Zhang, Z., Q. Liu, Y. Che, X. Yuan, L. Dai, B. Zeng, G. Jiao, Y. Zhang, X. Wu, J. Immunol. 169: 3006–3014. Y. Yu, et al. 2010. Antigen presentation by dendritic cells in tumors is disrupted 44. Yanagimoto, H., S. Takai, S. Satoi, H. Toyokawa, K. Takahashi, N. Terakawa, by altered metabolism that involves pyruvate kinase M2 and its interaction with A. H. Kwon, and Y. Kamiyama. 2005. Impaired function of circulating dendritic SOCS3. Cancer Res. 70: 89–98. cells in patients with pancreatic cancer. Clin. Immunol. 114: 52–60. 21. Zhang, M., Q. Liu, S. Mi, X. Liang, Z. Zhang, X. Su, J. Liu, Y. Chen, M. Wang, 45. Mohty, M., D. Jarrossay, M. Lafage-Pochitaloff, C. Zandotti, F. Brie`re, X. N. de Y. Zhang, et al. 2011. Both miR-17-5p and miR-20a alleviate suppressive po- Lamballeri, D. Isnardon, D. Sainty, D. Olive, and B. Gaugler. 2001. Circulating tential of myeloid-derived suppressor cells by modulating STAT3 expression. J. blood dendritic cells from myeloid leukemia patients display quantitative and Immunol. 186: 4716–4724. cytogenetic abnormalities as well as functional impairment. Blood 98: 3750– 22. Pinzon-Charry, A., T. Maxwell, M. A. McGuckin, C. Schmidt, C. Furnival, and 3756. J. A. Lo´pez. 2006. Spontaneous apoptosis of blood dendritic cells in patients 46. Cubillos-Ruiz, J. R., J. R. Baird, A. J. Tesone, M. R. Rutkowski, U. K. Scarlett, with breast cancer. Breast Cancer Res. 8: R5. A. L. Camposeco-Jacobs, J. Anadon-Arnillas, N. M. Harwood, M. Korc, 23. Satthaporn, S., A. Robins, W. Vassanasiri, M. El-Sheemy, J. A. Jibril, D. Clark, S. N. Fiering, et al. 2012. Reprogramming tumor-associated dendritic cells D. Valerio, and O. Eremin. 2004. Dendritic cells are dysfunctional in patients in vivo using miRNA mimetics triggers protective immunity against ovarian with operable breast cancer. Cancer Immunol. Immunother. 53: 510–518. cancer. Cancer Res. 72: 1683–1693. Supplementary Figure legends

FIGURE S1. A potential YWHAZ-based regulatory network of apoptosis by tumor associated miRNAs. (A) Tumor associated miR-16, miR-20a, miR-22, miR-155 and miR-503 downregulated the levels of multiple proteins such as YWHAZ, Glud1,

Eno1, Aldoa, Cap1 and Anaxa2. Murine RAW264.7 cells were transfected by tumor associated miR-22, miR-155, miR-16, miR-503 and miR-20a or control oligos (Ctr.

Tr). The lytic proteins were analyses by 2-D gel with MALDI-TOF MS after transfection for 24 hrs according to the protocol described in the methods.

(B) The YWHAZ-based regulation network of apoptosis by miRNAs in tumor associated DCs. Based on the downregulated molecules in miRNAs-transfected

RAW264.7, we constructed a potential YWHAZ-based regulation network of apoptosis using the IPA software (www.ipasoftware.com). Multiple molecules in this regulatory network could be potentially targeted by tumor associated miRNAs. The targeting molecules of microRNAs were predicted by TargetScan 5.1

(http://www.targetscan.org/cgi-bin/vert_50/targetscan.cgi?mirg=mmu-miR-223),

PicTar (http://pictar.bio.nyu.edu/) and miRanda (http://cbio.mskcc.org/cgi-bin/ mirnaviewer/mirnaviewer.pl). The genes with blue color are downregulated in

RAW264.7 transfected by miR-22, miR-155, miR-16, miR-503 or miR-20a. Some of these were potential targets of tumor associated miRNAs.

FIGURE S2. A potential Bcl2-based regulatory network of apoptosis by tumor

1 associated miRNAs. (A) The potential targeting molecules by tumor associated miRNAs. The numbers in blue cycles indicate the downregulated genes in 1D8 associated DCs. The numbers in red cycle indicate the potential targets by miRNA.

Targeting molecules by tumor associated miRNAs were predicted by TargetScan 5.1

(http://www.targetscan.org/cgi-bin/vert_50/targetscan.cgi?mirg=mmu-miR-223),

PicTar (http://pictar.bio.nyu.edu/) and miRanda (http://cbio.mskcc.org/cgi-bin/ mirnaviewer/mirnaviewer.pl). Numbers indicated by arrow is the potential targeting molecules by miRNAs in the downregulated genes of tumor associated BMDCs. (B)

A potential Bcl2-based regulatory network of apoptosis by miRNAs in tumor associated DCs. Based on the downregulated genes with blue color in murine ovarian carcinoma 1D8 associated BMDCs and potential targeting molecules by miRNAs, we constructed a Bcl2-based apoptotic signal network by tumor associated miRNAs using the IPA software (www.ipasoftware.com).

2

Table S1. Oligos used in this study.

Table S2. 3’UTRs and mutated 3’UTRs of YWHAZ and Bcl2 targeted by miR-22 and miR-503. 3’UTRs of YWHAZ and Bcl2 were cloned and the mutated 3’UTR of

YWHAZ and Bcl2 were prepared according the protocol in the materials and methods.

3 Table S1. Oligos used in this study.

᧤A᧥ Oligos used for miRNA detection.

Oligo name stem loop RT primer qPT-PCR Reverse-primer mmu-miR-16-1 GTCGTATCCAGTGCAGGGTCCGAGGT CGGCTAGCAGCACGTAA ATTCGCACTGGATACGACCGCCAA ATA mmu-miR-21 GTCGTATCCAGTGCAGGGTCCGAGGT CGGCTAGCTTATCAGAC ATTCGCACTGGATACGACTCAACA TGA mmu-miR-22 GTCGTATCCAGTGCAGGGTCCGAGGT ATTCGCACTGGATACGACACAGTT GCAAGCTGCCAGTTGAA mmu-miR-142- GTCGTATCCAGTGCAGGGTCCGAGGT CGGCCATAAAGTAGAAA 5p ATTCGCACTGGATACGACAGTAGT GC mmu-miR-146a GTCGTATCCAGTGCAGGGTCCGAGGT GCCGTGAGAACTGAATT ATTCGCACTGGATACGACAACCCA CC mmu-miR-155 GTCGTATCCAGTGCAGGGTCCGAGGT CGGCTTAATGCTAATTGT ATTCGCACTGGATACGACACCCCT GA mmu-miR-503 GTCGTATCCAGTGCAGGGTCCGAGGT ATTAGCAGCGGGAACAG ATTCGCACTGGATACGACCTGCAG T

᧤B᧥ Oligos used for cloning murine YWHAZ and Bcl2.

Oligo name Sequence (5’>3’) Description YWHAZ s ATGGATAAAAATGAGCTGGTGCAG For cloning YWHAZ into YWHAZ as ATTTTCCCCTCCTTCTCCTGCTTCTG pCDNA3.1TOPO plasmid or for RT-PCR

For cloning Bcl2 into Bcl2 s ATGGCGCAAGCCGGGAGAACAGG pcDNA3.1TOPO plasmid or Bcl2 as CTTGTGGCCCAGGTATGCACCCAG RT-PCR

᧤C᧥ Oligos used for cloning 3UTR regions of murine YWHAZ and Bcl2.

Oligo name Sequence (5’>3’) Description YWHAZ 3’UTR CCGCTCGAGAACAGGGAAATAACTTG For cloning YWHAZ s for miR-22 3’UTR and mutated YWHAZ 3’UTR ATTTGCGGCCGCCTTCCTAAGTTAGAAA YWHAZ 3’UTR to as for miR-22 siCHECK-2 vector in XhoI 3’UTR mutant s TTATTCCCTCTTGCGTCGTAATGGGCT and NotI enzyme sites 3’UTR mutant as TAAAAGCCCATTACGACGCAAGAGGGA

1 BCL-2 3’UTR CCGCTCGAGGCAAATGGTCGAATCAG 3’UTR and mutated BCL-2 s for miR-503 CTA 3’UTR to siCHECK-2 BCL-2 3’UTR ATTTGCGGCCGCAGGGAAGTCTCAGT vector in XhoI and NotI as for miR-503 GGGT enzyme sites

3’UTR mutant s GCAGTCTGGGACGTAGAAGCTTTGAAG 3’UTR mutant as CAAAGCTTCTACGTCCCAGACTGCAG GC

(D) Oligos used for qRT-PCR analyses. Oligo name Sequence (5’>3’) Description YWHAZ s AGAAGACGGAAGGTGCTGAGAA qRT-PCR for mouse YWHAZ YWHAZ as CGGCCAAGTAACGGTAGTAGTCA detection BCL-2 s CGTCGCTACCGTCGTGACT qRT-PCR for mouse BCL-2 BCL-2 as CCCATCCCTGAAGAGTTCCT detection GAPDH s TCAACGGCACAGTCAAGG qRT-PCR for mouse GAPDH GAPDH as TACTCAGCACCGGCCTCA detection

2 Supplementary Table S2. 3’UTR and mutated 3’UTR of murine miR-22 and miR-503.

Predicted consequential pairing of target region and miRNA Position 745-752 of YWHAZ 3' UTR 5' ...UCCUUAUUCCCUCUUGGCAGCUA... ||| ||||||| mmu-miR-22 3' UGUCAAGAAGUUGA--CCGUCGAA ||| | | Position 745-752 of YWHAZ 3' UTR 5' ...UCCUUAUUCCCUCUUGCGTCGUA... mutation Position 439-445 of Bcl-2 3' UTR 5' ...GAAGCCUGCAGUCUGGCUGCUAG... ||| |||||| Mmu-miR-503 3' GACGUCAUGACAAGGG--CGACGAU ||| | | Position 439-445 of Bcl-2 3' UTR 5' ...GAAGCCUGCAGUCUGGGACGUAG... mutation