MIF Maintains the Tumorigenic Capacity of Brain Tumor–Initiating

Published OnlineFirst March 15, 2016; DOI: 10.1158/0008-5472.CAN-15-1011 Cancer Tumor and Stem Cell Biology Research

MIF Maintains the Tumorigenic Capacity of Brain Tumor–Initiating Cells by Directly Inhibiting p53 Raita Fukaya1, Shigeki Ohta2, Tomonori Yaguchi3, Yumi Matsuzaki2, Eiji Sugihara4, Hideyuki Okano2, Hideyuki Saya4,Yutaka Kawakami3,Takeshi Kawase1, Kazunari Yoshida1, and Masahiro Toda1

Abstract

Tumor-initiating cells thought to drive brain cancer are embed- different mechanisms, depending on glioma cell line and p53 ded in a complex heterogeneous histology. In this study, we status. MIF physically interacted with wild-type p53 in the nucleus isolated primary cells from 21 human brain tumor specimens to and inhibited its transcription-dependent functions. In contrast, establish cell lines with high tumorigenic potential and to identify MIF bound to mutant p53 in the cytoplasm and abrogated the molecules enabling this capability. The morphology, sphere- transcription-independent induction of apoptosis. Furthermore, forming ability upon expansion, and differentiation potential of MIF knockdown inhibited BTIC-induced tumor formation in a all cell lines were indistinguishable in vitro. However, testing for mouse xenograft model, leading to increased overall survival. tumorigenicity revealed two distinct cell types, brain tumor– Collectively, our ﬁndings suggest that MIF regulates BTIC initiating cells (BTIC) and non-BTIC. We found that macrophage function through direct, intracellular inhibition of p53, shedding migration inhibitory factor (MIF) was highly expressed in BTIC light on the molecular mechanisms underlying the tumorigenicity compared with non-BTIC. MIF bound directly to both wild-type of certain malignant brain cells. Cancer Res; 76(9); 1–11. 2016 AACR. and mutant p53 but regulated p53-dependent cell growth by

Introduction contribute significantly to the tumor mass, but not to the initiation and maintenance of the tumor. BTICs are thus considered a Glioblastoma, the most common form of primary malignant promising target for future therapies. brain tumor, is nearly always fatal (1). Although neurosurgeons Macrophage migration inhibitory factor (MIF) was originally may be able to remove the main tumor mass, this procedure identified as a proinflammatory cytokine, but was later found to typically leaves behind a small population of tumor cells that be a critical regulator of tumor progression as well (14–18). MIF retain the capacity for tumorigenesis (2). For this reason, glio- upregulation has been observed in human melanoma (19), lung blastomas almost always relapse, highlighting the need for more cancer (20), prostate cancer (21), hepatocellular carcinoma (22), precisely targeted therapeutic approaches. A number of recent and glioblastoma (23–27). MIF expression was significantly studies have demonstrated that tumors are composed of cells with higher in the tumor tissues of glioblastoma patients than in functional heterogeneity that form a hierarchy, especially in the normal brain tissues (24, 25). The precise roles of MIF in the context of tumor initiation (3, 4). Characterized by its heteroge- initiation and progression of glioma, however, remain unclear. neous tumor types (5–8), glioblastoma is now also thought to be MIF plays numerous roles through receptor-mediated signaling initiated and maintained by a subpopulation of tumor cells, pathways that are activated by cell surface receptors, such as CD74, called brain tumor stem cells or brain tumor–initiating cells CD44, and CXCRs (28–30), or by intracellular interactions (BTIC; refs. 9–13). The clinical significance of BTICs is empha- through binding to JAB1 (31). MIF has also been shown to sized by the strongly enhanced capacity of these cells to initiate function as a p53 inhibitor (16) that directly binds and antag- brain tumors, in contrast to the majority of tumor cells, which may onizes p53 function in the cells (32). The p53 protein is an important and well-characterized tumor suppressor (33). In response to various forms of cell stress, p53 1 Department of Neurosurgery, Keio University School of Medicine, regulates diverse target molecules, which may induce cell-cycle Tokyo, Japan. 2Department of Physiology, Keio University School of Medicine, Tokyo, Japan. 3Division of Cellular Signaling, Institute for arrest and apoptosis in a transcription-dependent manner (34). Advanced Medical Research, Keio University School of Medicine, p53 has also been shown to possess other biologic activities, 4 Tokyo, Japan. Division of Gene Regulation, Institute for Advanced including induction of apoptosis, in mitochondria in a transcrip- Medical Research, Keio University School of Medicine, Tokyo, Japan. tion-independent manner (35–37). Note: Supplementary data for this article are available at Cancer Research In this study, we show that MIF expression is significantly Online (http://cancerres.aacrjournals.org/). higher in BTICs isolated from human glioma tissues than in Corresponding Author: Masahiro Toda, Keio University School of Medicine, 35 non-BTICs from the same tissue source and that MIF acts as a Shinanomachi, Shinjuku, Tokyo 1608582, Japan. Phone: 813-5363-3807; Fax: direct intracellular p53 inactivator that regulates cell proliferation 813-3358-0479; E-mail: [email protected] and apoptosis in glioma cells. In addition, we highlight the link doi: 10.1158/0008-5472.CAN-15-1011 between the roles of MIF and tumor-initiating cells associated 2016 American Association for Cancer Research. with p53, a candidate for the dedifferentiation barrier.

www.aacrjournals.org OF1

Fukaya et al.

Table 1. Isolation of BTICs Differentiation Inoculation of In vivo tumor Tumor type Age (y) Sex Culture potential the tumor cells formation P53 hG002 Pilocytic astrocytoma 29 M Impossible hG016 Anaplastic astrocytoma 43 M Temporarily þ Tested hG001 Glioblastoma 46 M Temporarily Not tested Not tested hG004 Glioblastoma 64 F Temporarily þ Not tested hG006 Glioblastoma 25 F Impossible hG008 Glioblastoma 61 F Permanently þ Tested þ M hG010 Glioblastoma 50 F Temporarily þ Tested hG011 Glioblastoma (Gliosarcoma) 40 M Temporarily þ Tested hG013 Glioblastoma 39 F Temporarily þ Tested hG014 Glioblastoma 37 F Temporarily þ Tested hG015 Glioblastoma 81 F Temporarily þ Tested hG017 Glioblastoma 37 F Temporarily þ Tested hG018 Glioblastoma 85 F Temporarily þ Tested hG020 Glioblastoma 53 M Permanently þ Tested þ WT hG021 Glioblastoma 73 M Impossible hG007 Oligodendroglioma 30 M Temporarily þ Tested hG005 Anaplastic oligodendroglioma 81 F Temporarily þ Not tested hG012 Anaplastic oligodendroglioma 36 F Temporarily þ Tested hG019 Anaplastic oligodendroglioma 41 M Permanently þ Tested þ WT hG003 Medulloblastoma 34 M Temporarily þ Tested hG009 Olfactory neuroblastoma 54 M Impossible Abbreviations: M, mutant; WT, wild-type; þ, possess the potential; , none of the potential.

Materials and Methods were prepared. hG008 cells were infected with lentivirus expressing either MIF shRNA or control shRNA [multiplicity of infection Human tissues and cell culture (MOI) ¼ 1] for 24 hours and washed twice in PBS, after which the Approval to use human neurosphere cultures was obtained number of live cells was counted. A total of 10,000 cells were from the Keio University ethics committee. Tissue procurement inoculated into the brain striatum (n ¼ 5); the mice were killed procedures were performed in accordance with the Declaration of and tumor generation was evaluated by histologic examinations Helsinki and in compliance with the ethical guidelines of the six weeks after implantation. For the treatment study using Japan Society of Obstetrics and Gynecology and the Network of lentivirus, intracranial transplantation of BTICs was performed European CNS Transplantation and Restoration. Brain tumor as described above. To establish glioma xenografts for lentivirus- samples were obtained from patients undergoing surgery at Keio mediated MIF shRNA treatment, hG008 cells were injected into University Hospital (Tokyo, Japan) following approval from the the brains of NOD/SCID mice (3,000 cells/mouse, eight mice/ Institutional Review Board at Keio University School of Medicine group). Lentivirus expressing either MIF shRNA or shRNA control (Tokyo, Japan). Tumor samples were mechanically and enzymat- (MOI ¼ 1) was injected twice into the tumor sites at two and three ically dissociated into single cells. All tumor-derived cells and weeks after tumor cell implantation. The effects of MIF shRNA neural stem cells (NSC) were cultured at 37C in a humidified treatment were evaluated to compare the survival times of each incubator with 5% CO in neurosphere medium (NSM) consist- 2 group by Kaplan–Meier analysis. ing of Neurobasal Medium (Invitrogen), B27 (Invitrogen), human recombinant bFGF and EGF (PeproTech, 20 ng/mL each), Immunoprecipitation assay and heparin (10 ng/mL; Sigma). U87MG and T98G cells were Immunoprecipitation was performed using an nProtein A- provided by Dr. Y. Ohashi (Keio University) and maintained in Sepharose 4 Fast Flow Kit (GE Healthcare) following the manu- DMEM (Wako) supplemented with 10% FBS. Short tandem facturer's instructions. Nuclear and cytoplasmic extracts of tumor repeat DNA profiling of T98G and U87MG cells was performed cells (1 mg) were suspended in 1 mL RIPA buffer (Pierce) and using the Cell ID System (Promega) in August 2011. Human coupled to 50 mL nProtein A-Sepharose (GE Healthcare) for 1 astrocytes were obtained from Lonza and maintained in Astrocyte hour at 4C for precleaning. The supernatant was incubated with Basal Medium (Lonza). mouse anti-p53 (DO-1; 1:200; Santa Cruz Biotechnology) or fi Animal models nonspeci c mouse IgG (sc-2025; Santa Cruz Biotechnology) antibody for 4 hours at 4C. A total of 50 mL nProtein A-Sepharose For the brain tumor initiation assay, 6- to 8-week-old female NOD/SCID mice were purchased from Sankyo Lab. Samples of was added to the samples, which were incubated for 1 hour at 4 C. 1,000, 10,000, or 100,000 primary cells cultured from brain The immune complexes were washed twice with RIPA buffer and tumor tissues were stereotactically injected into the brain striatum once with wash buffer (50 mmol/L Tris, pH 8.0). The beads were (2.5 mm right of the midline, 3 mm deep). The mice were killed resuspended in the sample buffer (1% SDS, 100 mmol/L DTT, 50 when they exhibited neurologic symptoms or at 150 days after mmol/L Tris, pH 7.5) and boiled for 5 minutes before the super- inoculation, whichever came first. All procedures involving ani- natants were subjected to Western blot analysis. mal experiments were approved by the Animal Care and Use Committee of the Keio University School of Medicine (Tokyo, Gel shift assay Japan). To evaluate the tumor-initiating potential of BTICs by Gel shift assays were performed using a P53 EMSA Kit treatment of lentivirus expressing MIF shRNA, two groups of mice (Panomics) following the manufacturer's instructions. In each

OF2 Cancer Res; 76(9) May 1, 2016 Cancer Research

MIF Maintains the Tumorigenic Capacity of BTICs

Figure 1. Characteristics of BTICs. A, tumor spheres expressing nestin. Scale bar, 100 mm (left); 50 mm (right). B, BTIC spheres can differentiate into three cell lineages: glial, neural, and oligodendroglial cells. Scale bar, 50 mm. C, typical hematoxylin and eosin staining images of BTICs implanted in NOD/SCID mice. In mouse brain, BTIC- generated tumors were highly inﬁltrative and hemorrhagic, features characteristic of human glioblastoma. Scale bar, 2 mm (left); 100 mm (right).

sample, 10 mg of protein from nuclear extracts isolated from treatment with shRNAs. P < 0.05 was considered to represent U87MG cells was incubated with biotinylated p53 consensus significance. binding sequence oligonucleotides (AY1032P; Panomics). Sep- All other methods are available in Supplementary Material and aration of bound and free probe was achieved by electropho- Methods. resis using a 6.0% nondenaturing polyacrylamide gel, and the – protein DNA complexes were transferred to a nylon membrane Results (Hybond-N, GE Healthcare). The DNA was crosslinked to the membrane using a UV cross-linker. Protein–DNA complexes Isolation and characterization of BTICs were detected using streptavidin–horseradish peroxidase and Twenty-one brain tumors were dissociated into single cells and visualized after exposure to Hyperfilm ECL film (Amersham plated at a clonal density (38) in NSM. These brain tumors and Biosciences). Experiments were performed independently at their characteristics are shown in Table 1. Four of the tumors, least twice. including two glioblastomas, one pilocytic astrocytoma, and one olfactory neuroblastoma, could not be cultured in NSM. Most of Statistical analysis the primary cells could be cultured temporarily as glioma spheres Statistical analyses were performed using Statcel2. Student t but did not culture for more than 8 passages. Only three cultures, tests were performed to determine the statistical significance of the hG008, hG019, and hG020, could be cultured for more than 20 quantitative PCR study, cell proliferation study, double knock- passages and proliferate permanently. These three cell lines down study, reporter gene assay, and apoptosis assay. Mann– included two glioblastomas, hG008 and hG020, and one ana- Whitney tests were used to assess in vivo tumor initiation after plastic oligodendroglioma, hG019. Notably, hG008 and hG019

www.aacrjournals.org Cancer Res; 76(9) May 1, 2016 OF3

Fukaya et al.

Figure 2. MIF expression in glioma cells and tissues. A, MIF gene expression was signiﬁcantly higher in BTICs (hG008, hG019, and hG020) compared with non-BTICs (hG012, hG013, and hG014; mean SD). , P < 0.01. B, immunohistochemical analysis of MIF expression in human glioblastoma tissues. MIF was expressed heterogeneously in tumor cells (top; N, necrotic region). Strong MIF expression was observed in both the nucleus and cytoplasm, and most of the MIF- expressing tumor cells showed nuclear hypertrophy in glioblastoma tissues (bottom). Scale bar, 200 mm. C, protein expression of MIF and p53 in the cytoplasmic and nucleic fractions.

were obtained from tumors that had recurred after surgery, hG019, and hG020 were BTICs. The other primary cells, which did radiation, and chemotherapy. hG020 was isolated from a patient not generate brain tumors in vivo but showed multilineage dif- who had undergone surgery for a newly diagnosed glioblastoma. ferentiation potential and limited potential for expansion in vitro, Multiple point mutations in the DNA-binding domain of the p53 were deﬁned as non-BTICs. gene were found in hG008 (Supplementary Table S1). Wild-type p53 was expressed in hG019 and hG020. BTICs usually show BTICs expressed MIF and p53 potential for tumorigenesis, proliferation, self-renewal, and mul- To identify molecules involved in brain tumorigenesis, we tidifferentiation (39). These three expandable primary cells gen- analyzed gene expression in BTICs and non-BTICs. MIF is a erated glioma spheres and expressed nestin, a marker for NSCs cancer-related gene that has previously been shown to promote (Fig. 1A). To assess their multilineage differentiation potential, cell survival and proliferation in mouse NSCs (40). Intriguingly, the primary cells were cultured in differentiation medium con- the MIF (Fig. 2A) gene was more highly expressed in BTICs than in taining 1% FBS without EGF and FGF2. All of the analyzed non-BTICs, astrocytes, and NSCs. Next, MIF expression in human primary cells, including BTICs and non-BTIC, were able to dif- glioblastoma tissues was assessed by immunohistochemical anal- ferentiate into cells of three neural lineages: glial (GFAP), neu- ysis. Some tumor cells expressed MIF in both the cytoplasm and ronal (b-III tubulin), and oligodendroglial (O4; Fig. 1B; Table 1). the nucleus (Fig. 2B). MIF was highly expressed in BTICs and Next, to determine whether the primary cells were tumorigenic in glioblastoma tissues, although the gene expression level of CD74 vivo, the primary cells were xenografted into the striatum of NOD/ and CXCR4, known as MIF direct binding receptors, was lower in SCID mice. We attempted to generate tumors in the mouse brain cultured BTICs than in NSCs (Supplementary Fig. S1). We decided by inoculating 1,000, 10,000, or 100,000 primary cells and to focus on the intracellular functions of MIF in this study. MIF observed tumorigenesis solely from hG008, hG019, and hG020 expression was further investigated in the cytoplasm and nucleus cells (Fig. 1C; Table 1), from which we concluded that hG008, by immunoblotting (Fig. 2C). MIF protein expression was veriﬁed

OF4 Cancer Res; 76(9) May 1, 2016 Cancer Research

MIF Maintains the Tumorigenic Capacity of BTICs

in the cytoplasmic and nucleic fractions in BTICs and glioblas- U87MG cells, we did not observe an increase in the transcriptional toma cell lines (T98G, U87MG). Intriguingly, MIF expression was activity of the p21 and BAX promoters following MIF siRNA low in the cytoplasm and was not detected in the nuclei of NSCs treatment in the reporter gene assay, suggesting that mutated and astrocytes. MIF has previously been identified as a molecule p53 does not activate the transcription of target genes in T98G that inhibits p53 function (16), so we investigated p53 expres- cells (Supplementary Fig. S3A). sion. The cytoplasmic fractions of hG008, hG019, and T98G We therefore focused on the function of p53 in the cytoplasmic displayed p53 expression, whereas hG020, NSCs, and U87MG fraction of T98G cells and observed that p53 binds directly to MIF showed very low levels of p53 expression. In the nuclear fractions, in this fraction (Fig. 4D). MIF siRNA treatment led to the induc- p53 expression was high in BTICs and T98G and moderate in tion of caspase-3/7 activity (Fig. 4E) without the induction of cell- NSCs, astrocytes, and U87MG (Fig. 2C). cycle arrest in T98G cells (Supplementary Fig. S3B). We hypoth- esized that the decrease in cell proliferation induced by MIF gene MIF binds to p53 and inhibits p53 function silencing in T98G is regulated by proapoptotic events induced by We first examined the role of MIF in two glioblastoma cell lines, p53 in the cytoplasmic fraction. In fact, MIF gene silencing led to U87MG and T98G. Both cell lines expressed MIF and p53, but the accumulation of p53 in the mitochondria fraction (Fig. 4F), these exhibit different p53 gene mutation statuses: U87MG which may lead to increased apoptosis in mitochondria. More- express wild-type p53, and T98G has a homozygous mutation over, treatment with PFT-m, a known inhibitor of p53 transloca- at codon 237 (M ! I) of p53 (41). MIF gene expression tion to mitochondria, abrogated the suppression in proliferation was knocked down using MIF siRNA in U87MG, and cell prolif- induced by MIF gene silencing in T98G cells (data not shown). eration was assessed. U87MG cells treated with MIF siRNA These findings underscore the importance of MIF-dependent/ showed decreased cell proliferation compared with siRNA con- induced p53 translocation to the mitochondria, which induces trols (Fig. 3A). Next, we tested the role of p53 in this system. A p53 apoptosis in T98G cells. Collectively, these results show that MIF knockdown system was established using p53 siRNA in U87MG is able to bind to both wild-type and mutant p53, and to support and T98G cells (Supplementary Fig. S2), and we found that p53 cellular proliferation in a p53-dependent manner in both U87MG gene silencing abrogated the decrease in cell proliferation induced and T98G cells, although the underlying mechanisms differ by MIF gene silencing in U87MG cells (Fig. 3B and Supplementary between the two glioma cell lines. Fig. S2). This suggests a strong functional relationship between MIF and p53. We next explored the physical interaction between MIF is the therapeutic target for BTICs MIF and p53 in U87MG cells. Immunoprecipitation analysis was To confirm the effects of MIF on proliferation and apoptosis in performed using U87MG nuclear protein, and we confirmed that BTICs, a lentivirus expressing MIF shRNA (lenti-shMIF) was p53 binds directly to MIF in the nuclei of U87MG (Fig. 3C). To employed, and the efficacy of the lentivirus was assessed by investigate the regulation of p53 target genes under conditions in immunoblot analysis in hG008 cells, which confirmed that which MIF expression is suppressed, reporter gene assays were changes had occurred in p53 protein expression (Fig. 5A). We performed using p21-Luc and Bax-Luc reporter plasmids. The evaluated the effects of lenti-shMIF on cell proliferation in three luciferase activities of P21 and BAX were elevated by MIF siRNA BTIC lines (hG008, hG019, and hG020). Dose-dependent sup- treatment compared with controls in U87MG (Fig. 3D). Immu- pression of cell proliferation was observed in all BTICs on lenti- noblot studies also confirmed upregulation of BAX and P21 shMIF treatment (Fig. 5B). Microscopic observations revealed that protein in a dose-dependent manner by MIF siRNA treatment in lenti-shMIF infection decreased cell survival and inhibited glioma U87MG cells (Fig. 3E). Cell-cycle analysis was performed by flow sphere formation compared with the control (Fig. 5C). Suppres- cytometry and G1 cell-cycle arrest was observed in U87MG in sion of cell proliferation by MIF was also supported by p53 in response to MIF siRNA treatment (Fig. 3F). In addition, a caspase- hG008 cells (Supplementary Fig. S4). 3/7 activity assay showed a dose-dependent increase in apoptosis We further analyzed the gene expression change of p21 and following MIF siRNA treatment (Fig. 3G). Taken together, these BAX, known as transcription-dependent downstream targets of results indicate that MIF downregulation induces cell-cycle arrest p53, in two BTICs (hG008, p53 mutant-type; hG020, p53 wild- and apoptosis in U87MG cells. type), U87MG (p53 mutant-type), and T98G (p53 mutant-type) To elucidate the functional relationship between p53 and MIF, to investigate the effect of MIF knockdown. p21 and BAX genes changes in MIF-regulated p53 DNA-binding ability was assessed were upregulated on infection with lenti-shMIF in hG020 and by gel shift assay. A p21 promoter DNA probe containing a p53- U87MG compared with hG008 and T98G in quantitative PCR binding region was used in a gel shift assay using nuclear protein study (Supplementary Fig. S5A and S5B). We also confirmed that isolated from U87MG cells transfected with either MIF siRNA or MIF gene silencing led to the accumulation of p53 in the mito- control siRNA. The gel shift assay showed that p53-specific bind- chondria fraction in hG008, similar to the effect in T98G (Sup- ing was upregulated by MIF siRNA compared with the control plementary Fig. S5C). These results suggested that MIF inhibits the U87MG cells (Fig. 3H). MIF has thus been shown to bind to p53 in function of p53 in transcription-dependent and -independent the nucleus of U87MG cells, and MIF siRNA-mediated gene manner, based on the status of the p53 gene in glioma cells and silencing increased p53 DNA-binding ability, cell-cycle arrest, BTICs, respectively. Lenti-shMIF treatment of hG008 and hG020 and apoptosis. We also assessed MIF function in T98G cells in cells also led to an increase in caspase-3/7 activity in a dose- the same manner as performed in U87MG cells and observed that dependent manner (Supplementary Fig. S5D). MIF gene silencing by siRNA treatment decreased cell prolifera- We next examined the therapeutic effect of MIF targeting using tion compared with the controls in a dose-dependent fashion mouse xenograft models. First, 1 104 hG008 cells infected with (Fig. 4A and B). This reduction in cell proliferation was also p53 either lenti-shMIF or lenti-shControl were implanted into the dependent in T98 cells, as seen in U87MG cells (Fig. 4C and brains of immunodeficient NOD/SCID mice, and tumorigenicity Supplementary Fig. S2). However, in contrast to the results from was evaluated in each group at 42 days after implantation. The

www.aacrjournals.org Cancer Res; 76(9) May 1, 2016 OF5

Fukaya et al.

Figure 3. MIF binds to p53 in the nucleus and inhibits p53 function in a transcription-dependent manner in U87MG cells. A, proliferation of U87MG cells was suppressed by MIF siRNA (siMIF) treatment compared with the controls (sicont.; mean SD). , P < 0.01; , P < 0.001. B, the ability of MIF siRNA knockdown to suppress cell proliferation was abrogated by p53 gene silencing in U87MG cells at 7 days after transfection (mean SD). , P < 0.05. C, MIF binds to p53 in the nucleic fraction of U87MG cells. (Continued on the following page.)

OF6 Cancer Res; 76(9) May 1, 2016 Cancer Research

MIF Maintains the Tumorigenic Capacity of BTICs

Figure 4. MIF binds to p53 in the cytoplasmic fraction of T98G cells and inhibits the apoptotic function of p53. A, decreases in MIF protein expression by gene silencing did not affect p53 protein expression at 3 days after MIF siRNA (siMIF) transfection. B, cell proliferation was suppressed by MIF siRNA transfection in a dose- dependent manner (mean SD). , P < 0.05; , P < 0.001. C, the ability of MIF siRNA to suppress cell proliferation was abrogated by p53 gene silencing at 7 days after transfection (mean SD). , P < 0.001. D, MIF physically bound to p53 in the cytoplasmic fraction of T98G cells. E, MIF gene silencing in T98G cells led to an increase in caspase-3/7 activity compared with that of U87MG cells, 3 days after MIF siRNA transfection (mean SD). , P < 0.05; , P < 0.001. F, quantitation of p53 protein in mitochondria fraction by Western blot (WB) analysis showed increased p53 protein localization in the mitochondria of T98G cells following MIF gene silencing by infection with a lentivirus expressing MIF shRNA at 3 days after infection (mean SD). , P < 0.05. IP, immunoprecipitate. implantation of hG008 cells infected with the control virus led to (Fig. 5D). Next, we attempted to validate the therapeutic effect brain tumor formation in 4 of 5 mice, whereas no tumors were of MIF gene silencing in BTICs. First, 1 103 hG008 cells were observed in mice implanted with lenti-shMIF–infected hG008 implanted in the brains of NOD/SCID mice and then a lentiviral cells (n ¼ 5), suggesting that MIF may be an essential factor in vector expressing either MIF-shRNA or control-shRNA was admin- maintenance for brain tumors in this experimental system istered to the BTIC implantation site on days 7 and 14

(Continued.) D, the transcriptional activities of cell-cycle regulator p21 and apoptotic factor BAX were elevated by MIF gene silencing at 3 days after transfection (mean SD). , P < 0.05; , P < 0.001. E, BAX and p21 protein expression was induced by MIF gene silencing in U87MG cells at 3 days after siRNA transfection. F, G1 arrest was induced by MIF gene silencing at 3 days after siRNA transfection. G, caspase-3/7 activity was increased at 3 days after MIF siRNA transfection in a dose-dependent manner (mean SD). , P < 0.05; , P < 0.001. H, the physical binding between a p53-speciﬁc DNA probe and nuclear extracts from U87MG cells was increased by MIF gene silencing. These physical interactions were antagonized by a cold probe and an anti-p53 antibody. The nuclear extracts were prepared from U87MG cells transfected with different siRNA concentrations at 3 days after transfection. IP, immunoprecipitate; LA, luciferase; WB, Western blotting.

www.aacrjournals.org Cancer Res; 76(9) May 1, 2016 OF7

Fukaya et al.

Figure 5. MIF regulates the initiation of tumorigenesis in BTICs, and MIF gene silencing shows a therapeutic effect on BTICs in a xenograft model. A, hG008 cells infected with lenti-shMIF showed a signiﬁcant decrease in MIF protein expression compared with controls (lenti-shControl) without affecting p53 protein expression at 3 days after infection. B, BTICs infected with lenti-shMIF inhibited cell proliferation in a virus dose-dependent manner at 7 days after infection (mean SD). , P < 0.05; , P < 0.01; , P < 0.001. C, hG008 cells infected with lenti-shMIF did not form tumor spheres, whereas control cells grew while forming healthy tumor spheres. Scale bar, 200 mm. D, hG008 cells infected with lenti-shMIF did not generate brain tumors in mice, whereas 4 of 5 mice injected with hG008 cells infected with control virus generated brain tumors at 42 days after cell implantation (mean SD; P ¼ 0.01). E, the brain tumor–bearing mice implanted with hG008 cells were infected with either lenti-shMIF (n ¼ 8) or shControl (n ¼ 8) virus at two and three weeks after cell implantation. Mice treated with lenti-shMIF showed longer survival compared with control mice, as evaluated by Kaplan–Meier analysis (mean SD; P ¼ 0.04).

OF8 Cancer Res; 76(9) May 1, 2016 Cancer Research

MIF Maintains the Tumorigenic Capacity of BTICs

postimplantation. Survival periods were evaluated for each ther- knockdown both in glioma cells and BTICs, indicating that apeutic group, and the mice inoculated with lenti-shMIF lived the MIF–p53 axis was essential for glioma tumorigenesis based significantly longer than the control mice (n ¼ 8/group, P ¼ on BTICs. A more detailed analysis of MIF function in BTICs 0.04; Fig. 5E and Supplementary Fig. S6). These results suggest may thus be important, especially when considering the devel- that MIF plays a key role in the proliferation and tumorigenesis of opment of drugs targeting MIF. BTICs, both in vitro and in vivo. Histologically immature neoplasms are generally more malignant and aggressive in their development (47). Indeed, Discussion BTICs are assumed to be the origin of brain tumors and also to remain in an extremely undifferentiated status. Recently, it has High-viability brain tumor cells are known to be present in been reported that p53 acts as a barrier to somatic cell repro- the restricted ring–enhanced region surrounding the tumor gramming, a dedifferentiation process resembling tumor for- mass on gadolinium-enhanced MRI (42). Indeed, the efficien- mation (48). Our observations suggest the following hypoth- cy of isolation of BTICs from glioma tissues is known to be esis: if p53 in a normal glial cell is directly inactivated by quite low. One study showed that nine cell lines established accumulated MIF by factors such as cell stresses, including from 19 brain tumor tissues could be cultured for more than hypoxia or hypoglycemia, the cell may lose its resistance to eight passages as tumor spheres and that only six of the nine dedifferentiation and develop into an immature tumor cell cell lines were transplantable into mouse brains for examina- resemblingaBTIC,thecelloforiginanddrivingforceoftumor tion of their tumorigenic potential (43). In this study, we growth. In this hypothesis, BTICs may be derived from fully identified only three BTICs from 21 brain tumor masses, which differentiated cells, such as astrocytes, not only from NSCs in we have compared in the current study to non-BTICs that the brain. This possibility is supported by at least one recent showed no capacity for tumorigenicity. Previous BTIC studies study (49). Additional evidence also suggests that glioblastoma have distinguished primary tumor cell types based on their develops largely in subcortical white matter and rarely in expression of molecular markers, such as CD133. There are, ventricular zone, in which NSCs are abundantly present in however, no previous studies that distinguish BTICs and non- adult human brain (50). BTICs based on their functional properties. Therefore, the In conclusion, we have shown that MIF is highly expressed in identification of high MIF expression in BTICs is important BTICs and plays critical roles in the tumorigenicity of BTICs in vivo. to gain a better understanding of the mechanisms underlying Our study further suggests that MIF plays an important role in tumorigenesis. transforming benign glial cells to BTICs in the niche region, where We first investigated the functions of MIF in two glioblastoma glial cells are more likely to be exposed to cellular stresses. cell lines, U87MG and T98G, as these were available for most Moreover, MIF may represent a molecular therapeutic target and reproducible examinations, have well-known biologic properties, clinical marker for glioma therapy. including the status of the p53 gene, and also exhibit tumor- initiating potential. Disclosure of Potential Conflicts of Interest We found high levels of MIF expression in the nuclei and No potential conflicts of interests were disclosed. cytoplasm of human glioma cells, but low expression of CD74, a direct binding receptor of MIF (29). A previous study reported Authors' Contributions scant expression of CD74 in several glioma cells (44). On the Conception and design: R. Fukaya, S. Ohta, K. Yoshida, M. Toda basis of the intracellular distribution of MIF in glioma cells, we Development of methodology: R. Fukaya, S. Ohta focused on physical interactions between MIF and p53. Similar Acquisition of data (provided animals, acquired and managed patients, to the distinction observed the mutation status of p53, the provided facilities, etc.): R. Fukaya, S. Ohta, T. Yaguchi, Y. Matsuzaki intracellular distribution of p53 differed between U87MG Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): R. Fukaya, S. Ohta, T. Yaguchi, E. Sugihara and T98G cells. p53 was expressed in both the nucleus and Writing, review, and/or revision of the manuscript: R. Fukaya, S. Ohta, cytoplasm of T98G cells, but only in the nucleus of U87MG H. Okano, H. Saya, Y. Kawakami, K. Yoshida, M. Toda cells. Our findings suggest that MIF exerts its functions Administrative, technical, or material support (i.e., reporting or organizing through association with p53 mainly in the nucleic fraction data, constructing databases): R. Fukaya, S. Ohta, Y. Matsuzaki, H. Okano of U87MG cells in a transcription-dependent manner. In T98G, Study supervision: S. Ohta, Y. Kawakami, T. Kawase, K. Yoshida, M. Toda MIF may inhibit transcription-independent apoptosis by binding to cytoplasmic p53, as the mutant p53 lacks transcription- Acknowledgments activating ability (Supplementary Fig. S7). It has recently been The authors thank Dr. C. Ishioka (Tohoku University) for providing the p21- reported that p53 exerts various biologic functions, including Luc and Bax-Luc pMX-Ig vectors, Dr. Y. Kanemura for providing human NSCs, involvement in mitochondrial apoptosis in a transcription- and Prof. Douglass Sipp (Keio University) for invaluable comments for the manuscript. independent fashion (35–37). In most cells with p53 mutations, including T98G and hG008 cells, point mutations are located in the DNA-binding domain of p53,leadingtoa Grant Support complete lack or decrease in DNA-binding ability. In the This work was supported by JSPS KAKENHI grant no. 22791356 (R. Fukaya) cytoplasm of some cell types, mutant p53 has recently been and MEXT Global COE program project base no. F14 (H. Okano). The costs of publication of this article were defrayed in part by the payment of shown to induce apoptosis through a transcription- indepen- page charges. This article must therefore be hereby marked advertisement in dent pathway more effectively than wild-type p53 (45, 46), accordance with 18 U.S.C. Section 1734 solely to indicate this fact. which is consistent with our results (U87MG vs. T98G cells, hG020 vs. hG008 cells). The decrease of cell proliferation Received April 15, 2015; revised February 19, 2016; accepted March 4, 2016; following MIF gene silencing was abrogated by p53 gene published OnlineFirst March 15, 2016.

www.aacrjournals.org Cancer Res; 76(9) May 1, 2016 OF9

Fukaya et al.

References 1. Scoccianti S, Magrini SM, Ricardi U, Detti B, Buglione M, Sotti G, et al. 24. Markert JM, Fuller CM, Gillespie GY, Bubien JK, McLean LA, Hong RL, et al. Patterns of care and survival in a retrospective analysis of 1059 patients with Differential gene expression profiling in human brain tumors. Physiol glioblastoma multiforme treated between 2002 and 2007: a multicenter Genomics 2001;5:21–33. study by the Central Nervous System Study Group of Airo (italian Asso- 25. Munaut C, Boniver J, Foidart JM, Deprez M. Macrophage migration ciation of Radiation Oncology). Neurosurgery 2010;67:446–58. inhibitory factor (MIF) expression in human glioblastomas correlates with 2. Bao S, Wu Q, McLendon RE, Hao Y, Shi Q, Hjelmeland AB, et al. Glioma vascular endothelial growth factor (VEGF) expression. Neuropathol Appl stem cells promote radioresistance by preferential activation of the DNA Neurobiol 2002;28:452–60. damage response. Nature 2006;444:756–60. 26. Bacher M, Schrader J, Thompson N, Kuschela K, Gemsa D, Waeber G, et al. 3. Visvader JE, Lindeman GJ. Cancer stem cells in solid tumours: accumu- Up-regulation of macrophage migration inhibitory factor gene and protein lating evidence and unresolved questions. Nat Rev Cancer 2008;8:755–68. expression in glial tumor cells during hypoxic and hypoglycemic stress 4. Sugihara E, Saya H. Complexity of cancer stem cells. Int J Cancer 2013; indicates a critical role for angiogenesis in glioblastoma multiforme. Am J 132:1249–59. Pathol 2003;162:11–7. 5. Ricci-Vitiani L, Lombardi DG, Pilozzi E, Biffoni M, Todaro M, Peschle C, 27. Baron N, Deuster O, Noelker C, Stuer C, Strik H, Schaller C, et al. Role of et al. Identification and expansion of human colon-cancer-initiating cells. macrophage migration inhibitory factor in primary glioblastoma multi- Nature 2007;445:111–5. forme cells. J Neurosci Res 2011;89:711–7. 6. Kim CF, Jackson EL, Woolfenden AE, Lawrence S, Babar I, Vogel S, et al. 28. Bernhagen J, Krohn R, Lue H, Gregory JL, Zernecke A, Koenen RR, et al. MIF Identification of bronchioalveolar stem cells in normal lung and lung is a noncognate ligand of CXC chemokine receptors in inflammatory and cancer. Cell 2005;121:823–35. atherogenic cell recruitment. Nat Med 2007;13:587–96. 7. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF. 29.LengL,MetzCN,FangY,XuJ,DonnellyS,BaughJ,etal.MIFsignal Prospective identification of tumorigenic breast cancer cells. Proc Natl transduction initiated by binding to CD74. J Exp Med 2003;197: Acad Sci U S A 2003;100:3983–8. 1467–76. 8. Lapidot T, Sirard C, Vormoor J, Murdoch B, Hoang T, Caceres-Cortes J, et al. 30. Shi X, Leng L, Wang T, Wang W, Du X, Li J, et al. CD44 is the signaling A cell initiating human acute myeloid leukaemia after transplantation into component of the macrophage migration inhibitory factor-CD74 receptor SCID mice. Nature 1994;367:645–8. complex. Immunity 2006;25:595–606. 9. Fukaya R, Ohta S, Yamaguchi M, Fujii H, Kawakami Y, Kawase T, et al. 31. Kleemann R, Hausser A, Geiger G, Mischke R, Burger-Kentischer A, Flieger Isolation of cancer stem-like cells from a side population of a human O, et al. Intracellular action of the cytokine MIF to modulate AP-1 activity glioblastoma cell line, SK-MG-1. Cancer Lett 2010;291:150–7. and the cell cycle through Jab1. Nature 2000;408:211–6. 10. Hemmati HD, Nakano I, Lazareff JA, Masterman-Smith M, Geschwind DH, 32. Jung H, Seong HA, Ha H. Critical role of cysteine residue 81 of macrophage Bronner-Fraser M, et al. Cancerous stem cells can arise from pediatric brain migration inhibitory factor (MIF) in MIF-induced inhibition of p53 tumors. Proc Natl Acad Sci U S A 2003;100:15178–83. activity. J Biol Chem 2008;283:20383–96. 11. Galli R, Binda E, Orfanelli U, Cipelletti B, Gritti A, De Vitis S, et al. Isolation 33. Lane DP. Cancer. p53, guardian of the genome. Nature 1992;358:15–6. and characterization of tumorigenic, stem-like neural precursors from 34. Harms K, Nozell S, Chen X. The common and distinct target genes of the human glioblastoma. Cancer Res 2004;64:7011–21. p53 family transcription factors. Cell Mol Life Sci 2004;61:822–42. 12. Singh SK, Hawkins C, Clarke ID, Squire JA, Bayani J, Hide T, et al. 35. Speidel D. Transcription-independent p53 apoptosis: an alternative route Identification of human brain tumour initiating cells. Nature 2004;432: to death. Trends Cell Biol 2010;20:14–24. 396–401. 36. Green DR, Kroemer G. Cytoplasmic functions of the tumour suppressor 13. Yuan X, Curtin J, Xiong Y, Liu G, Waschsmann-Hogiu S, Farkas DL, et al. p53. Nature 2009;458:1127–30. Isolation of cancer stem cells from adult glioblastoma multiforme. Onco- 37. Moll UM, Wolff S, Speidel D, Deppert W. Transcription-independent pro- gene 2004;23:9392–400. apoptotic functions of p53. Curr Opin Cell Biol 2005;17:631–6. 14. Mitchell RA. Mechanisms and effectors of MIF-dependent promotion of 38. Gritti A, Parati EA, Cova L, Frolichsthal P, Galli R, Wanke E, et al. Multi- tumourigenesis. Cell Signal 2004;16:13–9. potential stem cells from the adult mouse brain proliferate and self-renew 15. Mitchell RA, Bucala R. Tumor growth-promoting properties of macrophage in response to basic fibroblast growth factor. J Neurosci 1996;16: migration inhibitory factor (MIF). Semin Cancer Biol 2000;10:359–66. 1091–100. 16. Hudson JD, Shoaibi MA, Maestro R, Carnero A, Hannon GJ, Beach DH. A 39. Piccirillo SG, Reynolds BA, Zanetti N, Lamorte G, Binda E, Broggi G, et al. proinflammatory cytokine inhibits p53 tumor suppressor activity. J Exp Bone morphogenetic proteins inhibit the tumorigenic potential of human Med 1999;190:1375–82. brain tumour-initiating cells. Nature 2006;444:761–5. 17. Calandra T, Roger T. Macrophage migration inhibitory factor: a regulator of 40. Ohta S, Misawa A, Fukaya R, Inoue S, Kanemura Y, Okano H, et al. innate immunity. Nat Rev Immunol 2003;3:791–800. Macrophage migration inhibitory factor (MIF) promotes cell survival and 18. Brock SE, Rendon BE, Xin D, Yaddanapudi K, Mitchell RA. MIF family proliferation of neural stem/progenitor cells. J Cell Sci 2012;125:3210–20. members cooperatively inhibit p53 expression and activity. PLoS One 41. Van Meir EG, Roemer K, Diserens AC, Kikuchi T, Rempel SA, Haas M, et al. 2014;9:e99795. Single cell monitoring of growth arrest and morphological changes 19. Shimizu T, Abe R, Nakamura H, Ohkawara A, Suzuki M, Nishihira J. High induced by transfer of wild-type p53 alleles to glioblastoma cells. Proc expression of macrophage migration inhibitory factor in human melano- Natl Acad Sci U S A 1995;92:1008–12. ma cells and its role in tumor cell growth and angiogenesis. Biochem 42. Tynninen O, Aronen HJ, Ruhala M, Paetau A, Von Boguslawski K, Salonen Biophys Res Commun 1999;264:751–8. O, et al. MRI enhancement and microvascular density in gliomas. Corre- 20. Tomiyasu M, Yoshino I, Suemitsu R, Okamoto T, Sugimachi K. Quanti- lation with tumor cell proliferation. Invest Radiol 1999;34:427–34. fication of macrophage migration inhibitory factor mRNA expression in 43. Gunther HS, Schmidt NO, Phillips HS, Kemming D, Kharbanda S, Soriano non-small cell lung cancer tissues and its clinical significance. Clin Cancer R, et al. Glioblastoma-derived stem cell-enriched cultures form distinct Res 2002;8:3755–60. subgroups according to molecular and phenotypic criteria. Oncogene 21. Meyer-Siegler KL, Iczkowski KA, Vera PL. Further evidence for increased 2008;27:2897–909. macrophage migration inhibitory factor expression in prostate cancer. 44. Zeiner PS, Preusse C, Blank AE, Zachskorn C, Baumgarten P, Caspary L, BMC Cancer 2005;5:73. et al. MIF receptor CD74 is restricted to microglia/macrophages, associated 22. Ren Y, Tsui HT, Poon RT, Ng IO, Li Z, Chen Y, et al. Macrophage migration with a M1-polarized immune milieu and prolonged patient survival in inhibitory factor: roles in regulating tumor cell migration and expression of gliomas. Brain Pathol 2015;25:491–504. angiogenic factors in hepatocellular carcinoma. Int J Cancer 2003;107: 45. Lambert JM, Gorzov P, Veprintsev DB, Soderqvist M, Segerback D, Bergman 22–9. J, et al. PRIMA-1 reactivates mutant p53 by covalent binding to the core 23. Savaskan NE, Fingerle-Rowson G, Buchfelder M, Eyupoglu IY. Brain miffed domain. Cancer Cell 2009;15:376–88. by macrophage migration inhibitory factor. Int J Cell Biol 2012;2012: 46. Kakudo Y, Shibata H, Otsuka K, Kato S, Ishioka C. Lack of correlation 139573. between p53-dependent transcriptional activity and the ability to

OF10 Cancer Res; 76(9) May 1, 2016 Cancer Research

MIF Maintains the Tumorigenic Capacity of BTICs

induce apoptosis among 179 mutant p53s. Cancer Res 2005;65: 49. Suva ML, Rheinbay E, Gillespie SM, Patel AP, Wakimoto H, Rabkin 2108–14. SD, et al. Reconstructing and reprogramming the tumor-propagating 47. Kumar V, Abbas AK, Fausto N, Aster JC. Robbins and Cotran pathologic potential of glioblastoma stem-like cells. Cell 2014;157: basis of disease, professional edition: expert consult-online. Amsterdam, 580–94. The Netherlands: Elsevier Health Sciences; 2009. 50. Louis DN, Ohgaki H, Wiestler OD, Cavenee WK. WHO classiﬁcation of 48. Tapia N, Scholer HR. p53 connects tumorigenesis and reprogramming to tumours of the central nervous system . Lyon, France: International Agency pluripotency. J Exp Med 2010;207:2045–8. for Research on Cancer; 2007.

www.aacrjournals.org Cancer Res; 76(9) May 1, 2016 OF11

MIF Maintains the Tumorigenic Capacity of Brain Tumor− Initiating Cells by Directly Inhibiting p53

Raita Fukaya, Shigeki Ohta, Tomonori Yaguchi, et al.

Cancer Res Published OnlineFirst March 15, 2016.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-15-1011

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2016/03/15/0008-5472.CAN-15-1011.DC1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/early/2016/04/20/0008-5472.CAN-15-1011. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.