The Transcription Factor IRF8 Counteracts BCR-ABL to Rescue Dendritic Cell Development in Chronic Myelogenous Leukemia

Cancer Molecular and Cellular Pathobiology Research

Tomoya Watanabe1,2, Chie Hotta1, Shin-ichi Koizumi1, Kazuho Miyashita1,3, Jun Nakabayashi1, Daisuke Kurotaki1, Go R. Sato1, Michio Yamamoto1, Masatoshi Nakazawa4, Hiroyuki Fujita3, Rika Sakai5, Shin Fujisawa5, Akira Nishiyama1, Zenro Ikezawa2, Michiko Aihara2, Yoshiaki Ishigatsubo3, and Tomohiko Tamura1

Abstract BCR-ABL tyrosine kinase inhibitors (TKI) have dramatically improved therapy for chronic myelogenous leukemia (CML). However, several problems leading to TKI resistance still impede a complete cure of this disease. IFN regulatory factor-8 (IRF8) is a transcription factor essential for the development and functions of immune cells, including dendritic cells. Irf8 / mice develop a CML-like disease and IRF8 expression is downregulated in patients with CML, suggesting that IRF8 is involved in the pathogenesis of CML. In this study, by using a murine CML model, we show that BCR-ABL strongly inhibits a generation of dendritic cells from an early stage of their differentiation in vivo, concomitant with suppression of Irf8 expression. Forced expression of IRF8 overrode BCR- ABL (both wild-type and T315I-mutated) to rescue dendritic cell development in vitro, indicating that the suppression of Irf8 causes dendritic cell deficiency. Gene expression profiling revealed that IRF8 restored the expression of a significant portion of BCR-ABL–dysregulated genes and predicted that BCR-ABL has immune- stimulatory potential. Indeed, IRF8-rescued BCR-ABL–expressing dendritic cells were capable of inducing CTLs more efficiently than control dendritic cells. Altogether, our findings suggest that IRF8 is an attractive target in next-generation therapies for CML. Cancer Res; 73(22); 6642–53. 2013 AACR.

þ Introduction chronic phase in which clonal BCR-ABL hematopoietic stem Chronic myelogenous leukemia (CML) is a myeloprolifera- cells give rise to increased numbers of their progenies, partic- tive disorder that accounts for approximately 15% of newly ularly myeloid precursors and mature cells (mainly neutro- diagnosed cases of adult leukemia (1, 2). The causal molecule of phils) in the bone marrow, blood, and extramedullary tissues. this disease is BCR-ABL, a 210 kDa oncoprotein encoded by the After 3 to 5 years, if not treated, the disease progresses into a BCR-ABL fusion gene, which is generated by a t(9;22) chromo- fatal blast crisis, characterized by accumulation of immature somal translocation (3). BCR-ABL possesses deregulated ABL precursor cells with differentiation arrest. – tyrosine kinase activity that stimulates many downstream During the last decade, therapy with BCR-ABL selective signals including the MYC, NF-kB, RAS, mitogen-activated tyrosine kinase inhibitors (TKI), such as imatinib, nilotinib, and protein kinase, phosphoinositide 3-kinase (PI3K), and STAT dasatinib, has greatly improved the prognosis of patients with pathways. These signals lead to stimulation of cell prolifera- CML (4). However, several problems still impede a complete fi tion, inhibition of apoptosis, and alteration of adhesion to cure of this disease. First, a minor but signi cant subset of stroma cells and extracellular matrix. CML begins with a patients fails to respond to or becomes resistant to TKIs mainly due to acquisition of point mutations in the ABL gene, such as T315I. Second, because leukemic stem cells are not efficiently Authors' Affiliations: Departments of 1Immunology, 2Environmental eliminated by TKIs, discontinuing TKI therapy often results in 3 Immuno-Dermatology, Internal Medicine and Clinical Immunology, and CML recurrence (4). Third, while CML is highly sensitive to 4Experimental Animal Science, Yokohama City University Graduate School of Medicine; and 5Department of Hematology, Yokohama City University tumor immunity, TKIs may suppress functions of T cells and Medical Center, Yokohama, Japan dendritic cells (5–7). Therefore, next-generation therapies for Note: Supplementary data for this article are available at Cancer Research CML are required. Online (http://cancerres.aacrjournals.org/). Dendritic cells are the most potent professional antigen- T. Watanabe and C. Hotta contributed equally to this work. presenting cells and are essential for the initiation and regu- lation of innate and adaptive immunity against pathogens and Corresponding Author: Tomohiko Tamura, Department of Immunology, Yokohama City University Graduate School of Medicine, 3-9 Fukuura, tumors (8). Dendritic cells are a heterogeneous population that Kanazawa-ku, Yokohama 236-0004, Japan. Phone: 81-45-787-2612; Fax: can be divided into classical dendritic cells (cDC; further þ þ 81-45-787-2615; E-mail: [email protected] divided into CD8a dendritic cells, CD4 dendritic cells, and doi: 10.1158/0008-5472.CAN-13-0802 double-negative dendritic cells in mice) and plasmacytoid þ 2013 American Association for Cancer Research. dendritic cells (pDC) both in humans and mice (9). CD8a

6642 Cancer Res; 73(22) November 15, 2013

þ dendritic cells (BDCA3 in humans) are a unique subset used at 8 to 12 weeks of age. All animal experiments were capable of priming CTLs by cross-presentation of dead cell carried out in accordance with the Guidelines for Proper materials and are thus critical for tumor immunity. pDCs Conduct of Animal Experiments (Science Council of Japan), produce a large amount of type I IFN, that is, IFN-a/b, upon and all protocols were approved by the Institutional Review Toll-like receptor (TLR)-7/8/9 signaling. Type I IFN elicits Boards of Yokohama City University (Yokohama, Japan; Pro- antiviral and antitumor responses. Of note, IFN-a was once tocol #F11-85). used as a major therapeutic agent for CML (10, 11). CML has long been suggested to be highly sensitive to T cell– Flow cytometry mediated tumor immunity (12). CML responds to immune- Flow cytometry was performed using FACSCanto II (BD mediated therapies, such as IFN-a, allogeneic stem cell trans- Biosciences), and data were analyzed using the FlowJo soft- plantation (alloSCT), and donor lymphocyte infusion (13). IFN-a ware (TreeStar). For antibodies and their clone names, see induces a specific T-cell response in CML (10, 11). In alloSCT, T- Supplementary Materials and Methods. cell depletion from donor bone marrow cells results in a significant increase in relapse, especially in patients with CML. Separation of hematopoietic progenitors Yet, the fact that the disease develops in patients with CML Murine lineage marker–negative (Lin ) cells were purified suggests that tumor immunity is insufficient in patients with from bone marrow cells by the magnetic-activated cell sorting CML. Because T cells per se in patients with CML rarely express (MACS) system using the Lineage Cell Depletion Kit (Miltenyi BCR-ABL, it is reasonable to suspect dendritic cells as a culprit. Biotec) and anti-interleukin (IL)-7 receptor (IL-7R/CD127) þ þ Indeed, several studies have shown that patients with CML have antibody. For isolation of Lin Sca-1 c-Kit (LSK) cells, reduced numbers of cDCs and pDCs in the chronic phase, as MACS-purified Lin cells were stained with antibodies against compared with healthy individuals (14, 15). Other studies have Sca-1 and c-Kit, and were then purified by the fluorescence- reported that dendritic cells generated in vitro from monocytes activated cell sorting (FACS) system using a FACSAria II (BD þ or CD34 cells of patients with CML are functionally defective in Biosciences). multiple aspects, such as actin organization, antigen processing, migration, maturation, and cytokine production (16, 17). In a Retroviral transduction and CML model mice murine CML model, defective homing and impaired induction The following murine stem cell virus (MSCV) retroviral of CTLs by BCR-ABL–expressing dendritic cells have been vectors were used: MIG [MSCV-internal ribosome entry site reported (18). However, the mechanism of impaired dendritic (IRES)- GFP], MIG-p210BCR-ABL (MSCV-p210BCR-ABL-IRES- cell development in CML remains unknown. GFP), MICD8 [MSCV-IRES-human truncated CD8 (hCD8t)], IFN regulatory factor-8 (IRF8) is a hematopoietic transcrip- and MICD8-IRF8 (MSCV-IRF8-IRES-hCD8t). Lin cells were tion factor that regulates the development of multiple immune precultured for 24 hours with stem cell factor (SCF) and cell types (19). IRF8 is required for differentiation of mouse thrombopoietin (TPO) for transplantation, or SCF, IL-6, and þ cDCs (particularly CD8a dendritic cells; refs. 20, 21), pDCs IL-3 for in vitro experiments, and were then transduced with þ (22), and monocytes (particularly Ly6C monocytes; ref. 23), MSCVs by spinoculation for 2 consecutive days. All cytokines while inhibiting myeloid cell proliferation and neutrophil were purchased from PeproTech. For generating CML model differentiation (24). IRF8 is also indispensable for the differ- mice, C57BL/6-Ly5.1 LSK cells transduced with MIG or MIG- entiation of CTLs (25). Thus, Irf8 / mice develop immuno- p210BCR-ABL were injected intravenously into lethally irradiated deficiency and a CML-like syndrome (26). Importantly, muta- syngenic C57BL/6-Ly5.2 recipients. tions in the human IRF8 gene are associated with dendritic cell immunodeficiency (27). Furthermore, IRF8 expression is dra- qRT-PCR, ELISA, and immunoblot analysis matically decreased in patients with CML (28, 29). IRF8 also Quantitative PCR (qPCR) with quantitative reverse tran- overcomes the mitogenic activity of BCR-ABL in differentiating scription PCR (qRT-PCR) was performed using RNAiso Plus myeloid progenitors in vitro (30). Coexpression of IRF8 in bone (Takara Bio), DNase I (Invitrogen), PrimeScript (Takara Bio), marrow progenitors can ameliorate BCR-ABL–induced mye- Thunderbird SYBR qPCR Mix (Toyobo), and an ABI PRISM loproliferative disorder in mice (31). Coexpression of IRF8 in a 7900 sequence detection system (Applied Biosystems) accord- BCR-ABL–transformed mouse pro-B cell line induces a che- ing to the manufacturers' protocols. Primer sequences are mokine-dependent antileukemia immunity in mice (31–33). described in Supplementary Materials and Methods. Data were These findings imply that there is an antagonistic relationship analyzed using the DDCT method and normalized against between IRF8 and CML pathogenesis. However, this idea has Gapdh levels. ELISA for IFN-a and IL-12p40 was performed not been tested in terms of dendritic cell biology yet. In this using commercially available kits (PBL and BioLegend, respec- study, we have investigated how BCR-ABL and IRF8 are tively). For immunoblot analysis, anti-IRF8 (C-19; Santa Cruz involved in the development and function of dendritic cells Biotechnology), c-ABL (Cell Signaling Technology), phospho- in CML using a murine model. tyrosine (4G10; Millipore), and b-actin (AC-74; Sigma-Aldrich) antibodies were used. Materials and Methods Mice Dendritic cell culture C57BL/6-Ly5.1 or -Ly5.2 congenic mice and OT-I or OT-II T Bone marrowLin cells transduced with MSCVs were washed cell receptor transgenic mice (The Jackson Laboratory) were 24 hours after the last spinoculation and cultured with human

www.aacrjournals.org Cancer Res; 73(22) November 15, 2013 6643

fms-like tyrosine kinase-3 (Flt3)-ligand (Flt3L) for 7 days to BCR-ABL inhibits an early stage of dendritic cell induce dendritic cell differentiation. TKIs and a STAT5 inhibi- differentiation tor [N0-((4-oxo-4H-chromen-3-yl)methylene)nicotinohydrazide] We nextanalyzed progenitor populations inthe bone marrow. were purchased from Santa Cruz Biotechnology. Trichostatin A Dendritic cells originate from bone marrow hematopoietic stem (TSA) and 5-azacytidine (5-Aza) were purchased from Focus cells through monocyte-dendritic cell progenitors (MDP; Lin þ þ Biomolecules and Sigma-Aldrich, respectively. Sca-1 CD127 c-Kitint/ CD115 ) and then common dendritic þ þ cell progenitors (CDP; Lin Sca-1 CD127 Flt3 c-KitintCD115 ; þ refs. 9, 35). MDPs are a CD115 population in myeloid progeni- Microarray þ tors (Lin CD127 Sca-1 c-Kit ), and overlap with granulocyte- RNAs from two independent experiments were analyzed þ monocyte progenitors (GMP; Lin CD127 Sca-1 c-Kit CD16/ using a Whole Mouse Genome 8 60 K Microarray (Agilent) þ þ according to the manufacturer's protocol. Microarray data are 32 CD34 ) and to a lesser extent common myeloid progenitors þ low þ available at the GEO/NCBI database (GSE44920). For details, (CMP; Lin CD127 Sca-1 c-Kit CD16/32 CD34 ). Within myeloid progenitors, CMPs progress toward either GMPs or see Supplementary Materials and Methods. megakaryocyte-erythroid progenitors (MEP; Lin CD127 Sca- þ 1 c-Kit CD16/32 CD34 ). T-cell response assays þ Interestingly, the bone marrow GFP population in BCR-ABL Antigen presentation capability of dendritic cells via MHC mice lacked MDPs and CDPs (Fig. 1E). The percentages of total class II (MHC-II) was evaluated using OT-II T cells essentially þ GFP myeloid progenitors were comparable between BCR-ABL as previously described (21). In vitro CTL assays were per- þ mice and MIG mice (Supplementary Fig. S1E). GFP myeloid formed using OT-I T cells essentially by the method previously progenitors in BCR-ABL mice had relatively low percentages of established (34). CMPs and GMPs, which could be partially due to the lack of MDPs, while they contained a modestly increased percentage of þ Results MEPs and harbored an aberrant CD34 CD16/32 population. BCR-ABL inhibits dendritic cell development and Irf8 Despite the sharp decrease in MDP counts, we noted that expression monocyte counts were not significantly affected in BCR-ABL To investigate how dendritic cell development is affected mice (data not shown), consistent with the fact that monocyte in CML, we first used a murine CML model in which counts are not reduced in patients with CML. The developmen- þ hematopoietic progenitors (LSK cells) were transduced with tal route of these BCR-ABL monocytes is unknown, but one bicistronic MIG-p210BCR-ABL retrovirus (carrying cDNAs possibility is that they are derived from upstream myeloid encoding BCR-ABL and GFP) in the presence of SCF and progenitors without transiting MDPs. Overall, these data suggest TPO, and then transplanted these cells into lethally irradiated that BCR-ABL affects the early stage of dendritic cell differen- mice. These mice (BCR-ABL mice), but not mice transplanted tiation, particularly the generation of MDPs. with empty MIG-transduced cells (MIG mice), exhibited We analyzed IRF8 expression in BCR-ABL and MIG mice. splenomegaly and died by 4 to 8 weeks after the transplan- qRT-PCR analysis revealed that BCR-ABL suppressed Irf8 tation (Fig. 1A and Supplementary Fig. S1A). Wright–Giemsa mRNA expression both in Lin bone marrow progenitors and staining and flow-cytometric analyses of splenic cells splenic cells (Fig. 1F). Immunostaining for IRF8 in splenic revealed that BCR-ABL mice had significant increases in the dendritic cells demonstrated that the remaining few BCR- þ þ þ percentage and number of GFP (therefore BCR-ABL ) ABL dendritic cells expressed lower levels of IRF8 than neutrophils, a hallmark of CML, as compared with MIG mice control dendritic cells (Fig. 1G and Supplementary Fig. S1F). (Fig. 1B and C). We found that both cDCs (CD11chigh cells) and pDCs IRF8 overrides BCR-ABL to rescue dendritic cell þ (CD11clowPDCA-1 cells) were dramatically diminished in differentiation BCR-ABL mice as compared with MIG mice; the percentage To further investigate the mechanisms involved in this þ and number of GFP cDCs showed 16.7- and 6.5-fold reduc- process, we examined whether BCR-ABL inhibits dendritic tions, respectively, and those of pDCs showed 18.1- and 8.1-fold cell differentiation in vitro. We transduced MIG-p210BCR-ABL reductions, respectively (Fig. 1D). More detailed analysis dem- or empty MIG retrovirus into bone marrow Lin progenitors in þ onstrated that all cDC subsets, that is, CD8a dendritic cells, the presence of SCF, IL-6, and IL-3, and the transduced cells þ CD4 dendritic cells, and CD8a CD4 (double-negative) den- were then cultured with Flt3L for 7 days to induce differen- þ dritic cells, were significantly diminished (Supplementary Fig. tiation toward cDCs (CD11chighMHC-II ) and pDCs þ þ þ S1B). In addition, we noted that the remaining GFP cDCs in (CD11c B220 CD11blow; ref. 21). In MIG-transduced control BCR-ABL mice expressed lower levels of MHC-II compared cultures, cDCs and pDCs were efficiently generated (Fig. 2A). with those in MIG mice (Supplementary Fig. S1C). In fact, these In BCR-ABL–transduced cultures, however, the development spared dendritic cells exhibited a reduced capability of pre- of both cDCs and pDCs was significantly inhibited in terms of senting ovalbumin (OVA) antigen, as demonstrated by the percentages and absolute cell counts, even though BCR-ABL– proliferation of and IFN-g production by OT-II T cells (Sup- transduced cells showed a greater increase in cell number plementary Fig. S1D). These results indicate that BCR-ABL during the Flt3L culture than MIG-transduced cells. Wright– strongly and globally inhibits the generation of dendritic cells Giemsa staining revealed that while many of the MIG- in vivo. transduced cells showed typical dendritic cell morphology,

6644 Cancer Res; 73(22) November 15, 2013 Cancer Research

A B GFP+ cells 30 ** MIG MIG BCR-ABL

) BCR-ABL 7 20 10

× ** 10

0 spleen ( + MIG BCR-ABL Absolute cell number/ cell Absolute Total GFP Figure 1. Inhibition of dendritic cell C GFP+ cells GFP+ neutrophils development and Irf8 expression MIG BCR-ABL by BCR-ABL in vivo. LSK cells 60 ** 4 ** ) cells 6

transduced with empty MIG or + 3 BCR-ABL 40 MIG-p210 were 10 × × 2 transplanted into lethally irradiated

CD11b 20 mice. Spleens and bone marrow 6.8% 45.8 1 cells in recipient mice were 0 0 analyzed 4 to 5 weeks after Ly6G ( spleen MIG MIG ABL ABL % in total GFP total in % BCR- BCR- transplantation (n ¼ 5 each except number/ cell Absolute for Fig. 1G, in which n ¼ 3 each). D GFP+ cells cDCs pDCs cDCs pDCs A, spleens (left) and splenic cell ** ** ** ** numbers (right). B, Wright–Giemsa MIG BCR-ABL 6 1.2 150 20 ) cells

þ 4 stains of FACS-sorted GFP 1.1% 0.15 + 15 4 0.8 10 100

– ﬂ × spleen cells. C E, ow-cytometric 10 analyses of splenic neutrophils (C), 2 0.4 50 5

splenic dendritic cells (D), and bone PDCA-1 Irf8 marrow MDPs and CDPs (E). F, 4.9 0.5 0 0.0 0 0 mRNA expression levels in FACS- spleen ( % in total GFP total in % MIG MIG MIG MIG ABL ABL ABL þ ABL CD11c cell number/ Absolute BCR- BCR- BCR- sorted GFP bone marrow (BM) BCR- Lin cells and splenic cells determined by qRT-PCR in E F Irf8 mRNA triplicate. G, IRF8 protein BM Lin− Spleen þ ** ** expression levels in splenic GFP GFP+ Lin− CD127− Sca-1− cells MDPs high þ 0.020 0.03 CD11c cDCs and GFP PDCA- 12 ** þ MIG BCR-ABL ) low cells 0.015

1 CD11c pDCs determined by − 11.5% 0.7 9 0.02 ﬂow cytometry. The mean

Lin 0.010 + ﬂuorescent intensity (MFI) was 6 Gapdh 0.01 (/ 0.005 calculated by subtracting the 3 c-Kit 0.000 0.00 autoﬂuorescence intensity. All Expression level 0 values in bar graphs are the mean MIG MIG ABL ABL % in GFP BCR- BCR-

P P MIG

SD. , < 0.05 and , < 0.01 ABL t CD115 BCR- (Student test). G IRF8 protein in DCs cDCs pDCs GFP+ Lin− c-Kitint Flt3+ cells CDPs ** MIG BCR-ABL 5 ** * cells 40 100

− 4 30 75 13.2 0.4 Lin 3 + 20 50 2 MFI 1 10 25 CD115 0 0 0 % inGFP MIG ABL MIG ABL MIG ABL

CD127 BCR- BCR- BCR-

BCR-ABL–transduced cells exhibited neutrophil-like, macro- We next asked whether restoration of IRF8 expression phage-like, or immature morphologies (Fig. 2B). Furthermore, could improve dendritic cell differentiation impaired by the induction of Irf8 mRNA expression during the Flt3L culture BCR-ABL. To this end, we cotransduced Lin cells with was almost completely abrogated by BCR-ABL (Fig. 2C). When MICD8-IRF8 (that expresses IRF8 and hCD8t) together with a kinase-dead K1172R mutant of BCR-ABL was transduced into MIG-p210BCR-ABL, and treated them with Flt3L. hCD8t, used cells, neither dendritic cell development nor IRF8 expression as a transduction marker, does not transmit any signals was affected, suggesting that the kinase activity of BCR-ABL is because it contains no cytoplasmic domain. Flow-cytometric þ þ required for the observed inhibitory effects (Fig. 2D). The analysis of doubly transduced (i.e., GFP hCD8t ) cells expression levels of wild-type (WT) and K1172R BCR-ABL revealed that the forced expression of IRF8 efﬁciently res- were comparable, and K1172R lacked autophosphorylation at cued differentiation of both cDCs and pDCs (Fig. 3A). The tyrosine residues. expression levels of Irf8 mRNA and IRF8 protein in

www.aacrjournals.org Cancer Res; 73(22) November 15, 2013 6645

+ A GFP+ cells cDCs D GFP cells MIG BCR-ABL 100 ** MIG BCR-ABL K1172R

72.4% 5.8 cells 80 65.8% 10.5 69.8 + 60 40 MHC-II MHC-II 20

% in GFP in % 0 MIG CD11c ABL CD11c BCR- + + GFP+ CD11c+ cells GFP CD11c cells pDCs MIG BCR-ABL 16 ** MIG BCR-ABL K1172R

cells 12 + 8 4 CD11b 9.6 0.1 8.7 CD11b 13.5 0.1

% inGFP (8.9) (0.02) (8.1) (12.3) (0.02) 0

MIG B220 B220 ABL BCR- Irf8 mRNA Total cDCs pDCs 40 5 *** * 80 * ** ** 0.20 4 60 30 ) 0.15 BCR-ABL 3 / mL) 40 20 4 2 0.10 Tyr-P Gapdh 10 20 10 BCR-ABL (/ × × 1 0.05 (

Relative level level Relative IRF8 0 0 0 0.00 b Absolute cell number Absolute -Actin MIG MIG MIG ABL ABL ABL MIG BCR- BCR- BCR- K1172R

B C BCR-ABL

+ ) GFP cells Irf8 mRNA MIG BCR-ABL 0.25 **

Gapdh 0.20 MIG 0.15 BCR-ABL 0.10 ** 0.05 ** 0.00 0357

Relative level (/ level Relative Days

Figure 2. Inhibition of dendritic cell development and Irf8 expression by BCR-ABL in vitro.BonemarrowLin cells were transduced with MIG or MIG-p210BCR-ABL and then cultured for 7 days in the presence of Flt3L to induce dendritic cell differentiation. A, flow-cytometric analysis of dendritic cells. Numbers in þ þ þ middle left indicate percentages of pDCs relative to GFP CD11c cells (without parentheses) and those relative to total GFP cells (in parentheses). For evaluating þ cell numbers, FACS-purified GFP –transduced cells were seeded at 2 105 cells/mL at the beginning of Flt3L culture. B, Wright–Giemsa stains of þ þ FACS-sorted transduced GFP cells. C, Irf8 mRNA expression levels in FACS-sorted GFP cells determined by qRT-PCR. D, flow cytometry, qRT-PCR, and immunoblot analysis in cells transduced with K1172R BCR-ABL. Tyr-P BCR-ABL indicates tyrosine-phosphorylated BCR-ABL. b-Actin expression is shown as a loading control. All values in bar graphs are the mean SD from three independent experiments. , P < 0.05 and , P < 0.01 (Student t test).

þ IRF8-transduced BCR-ABL cells were comparable with and S2B for quality control data). We first calculated the those of endogenous IRF8 in control cells (Fig. 3B and C, number of genes whose expression was significantly affected compare the first and fourth lanes). Importantly, IRF8 also by BCR-ABL [fold change (FC) > 3; false discovery rate (FDR) efficiently overrode the TKI-resistant T315I-mutant BCR- < 0.05] and those normalized by IRF8 (Fig. 4A and Supple- ABL to rescue dendritic cell differentiation (Fig. 3D). WT mentary Tables S1 and S2). The criteria of normalization and T315I BCR-ABL were expressed at comparable levels were as follows: IRF8 significantly reverses the expression in þ (Fig. 3B and C). These results indicate that the suppression BCR-ABL cells (FC > 2; FDR < 0.05), and expression of the of Irf8 is a cause, rather than a result, of dendritic cell gene becomes comparable between control dendritic cells þ deficiency by BCR-ABL in CML. and IRF8-rescued BCR-ABL dendritic cells (i.e., FC is no longer > 3andFDR< 0.05). BCR-ABL decreased the expres- IRF8 normalizes the expression of a significant portion sion of 1,433 genes and increased the expression of 1,915 of BCR-ABL–dysregulated genes genes (FC > 3). Among these genes, despite the stringent We next performed transcriptome analysis by microarray criteria, IRF8 normalized the expression of 688 genes (48%) in bone marrow Lin cells transduced with empty vectors, and 564 genes (29%), respectively. These results indicate that MIG-p210BCR-ABL, or MIG-p210BCR-ABL þ MICD8-IRF8 fol- BCR-ABL impairs the expression of many genes via suppres- lowed by 5-day Flt3L culture (see Supplementary Fig. S2A sing IRF8 expression.

6646 Cancer Res; 73(22) November 15, 2013 Cancer Research

A GFP+ hCD8t+ cells cDCs MIG + MICD8 IRF8 BCR-ABL BCR-ABL + IRF8 100 cells ** ** + 75 77.6% 85.2 9.8 52.4

% in 50 hCD8t

+ 25 MHC-II

GFP 0 BCR-ABL – – ++ CD11c IRF8 ––+ + GFP+ hCD8t+ CD11c+ cells pDCs MIG + MICD8 IRF8 BCR-ABL BCR-ABL + IRF8 cells 16 ** ** + 12 Figure 3. Dendritic cell

% in 8

differentiation impaired by BCR- hCD8t ABL can be restored by forced +

CD11b 4 expression of IRF8 in vitro. Bone 13.2 9.2 0.2 10.2 (12.2) (9.1) (0.04) (6.8) GFP 0 marrow Lin cells were BCR-ABL – +– + cotransduced with empty MIG, B220 IRF8 ––+ + MIG-p210BCR-ABL or MIG- T315IBCR-ABL, and empty MICD8 or Irf8 MICD8-IRF8 and then cultured for B mRNA C ** – – WT WT T315IT315I BCR-ABL 7 days in the presence of Flt3L. 0.25 ** ** ** A, cotransduction with IRF8 and ––+++ – IRF8 ) 0.20 BCR-ABL. Numbers in bottom left BCR-ABL 0.15 indicate percentages of pDCs þ þ þ 0.10 relative to GFP hCD8 CD11c Gapdh IRF8 cells (without parentheses) and (/ 0.05 þ þ level Relative b those relative to total GFP hCD8 0.00 -Actin cells (in parentheses). B, qRT-PCR BCR-ABL – – WTWT T315I T315I for Irf8 mRNA expression levels in IRF8 –––+++ þ þ FACS-sorted GFP hCD8t cells. C, immunoblot analysis of IRF8 and D þ þ + + BCR-ABL in GFP hCD8t cells. GFP hCD8t cells cDCs 100 D, cotransduction with IRF8 and MIG + MICD8 IRF8 T315I T315I + IRF8 cells ** ** + T315I BCR-ABL. All values in bar 77.3% 78.5 10.8 48.8 75 graphs are from three independent

% in 50 experiments. hCD8t + 25 MHC-II

GFP 0 T315I – – ++ CD11c IRF8 ––+ + GFP+ hCD8t+ CD11c+ cells pDCs MIG + MICD8 IRF8 T315I T315I + IRF8

cells 16 ** ** + 12

% in 8 hCD8t + 4 CD11b 10.2 7.8 0.1 11.5

(9.8) (7.5) (0.02) (7.2) GFP 0 T315I – +– + B220 IRF8 ––+ +

Gene set enrichment analysis (GSEA; ref. 36) showed that Supplementary Materials and Methods) was significantly expression of dendritic cell signature genes (37) was indeed induced by BCR-ABL, but suppressed by IRF8 (Supplementary significantly suppressed by BCR-ABL, and was efficiently rein- Fig. S2C). duced by coexpression of IRF8 (Fig. 4B). qRT-PCR analysis of We also paid attention to the genes dysregulated by BCR- several of these genes (Ciita, H2-Eb2, and H2-Aa) in the same set ABL but not normalized by IRF8. We used Ingenuity Pathway of transduced cells as well as cells transduced with IRF8 alone Analysis (IPA) software to predict transcription factors that on day 7 and in freshly isolated Lin cells confirmed the may act upstream of these genes. The results showed that findings (Fig. 4C, top). Expression of other genes critical for the genes induced by BCR-ABL, irrespective of IRF8 supple- þ the function of CD8a dendritic cells (e.g., cross-presentation), mentation, were significantly enriched with those inducible such as Clec9a, Xcr1, and Tlr3, was also suppressed by BCR-ABL by NF-kB, a transcription factor activated by BCR-ABL and and restored by IRF8 (Fig. 4C, bottom). Conversely, GSEA involved in many activation signals in dendritic cells showed that expression of granulocyte signature genes (see (Supplementary Fig. S2D). Furthermore, GSEA showed that

www.aacrjournals.org Cancer Res; 73(22) November 15, 2013 6647

A Genes dysregulated by BCR-ABL (FC > 3) Decreased by BCR-ABL Increased by BCR-ABL Total : 1,433 Total : 1,915

547886 465 1,351 Normalized Normalized by IRF8 by IRF8

B DC signature genes MIG + BCR-ABL MIG+ BCR-ABL MICD8 BCR-ABL BCR-ABL +IRF8 MICD8 +IRF8 0.8 0.0 P < 0.001 P < 0.001 0.3 P = 0.33 0.6 NES = 2.42 –0.2 NES = –2.52 0.2 NES = 1.07 Figure 4. Gene expression proﬁling 0.4 –0.4 0.1 of cells transduced with BCR-ABL 0.2 –0.6 0.0 –0.1 and/or IRF8. Bone marrow Lin 0.0 –0.8 cells were cotransduced with empty MIG or MIG-p210BCR-ABL Enrichment score and empty MICD8 or MICD8-IRF8, 0 50 100 150 200 0 50 100 150 200 0 50 100 150 200 þ þ and FACS-sorted GFP hCD8t Rank in ordered dataset (× 100) cells were then cultured for 5 (for microarray) or 7 days (for qRT- C Ciita H2-Eb2 H2-Aa PCR) in the presence of Flt3L. 0.05 0.005 0.06 Microarray and qRT-PCR were 0.04 0.004 performed in biologic duplicates. 0.03 0.003 0.04 A, Venn diagrams for the numbers of genes that displayed more than a 0.02 0.002 0.02 3-fold change in expression by 0.01 0.001 BCR-ABL and those normalized by 0.00 0.00 0.000 IRF8. B, GSEA for dendritic cell ) BCR-ABL + + Lin− + + Lin− + + Lin− (DC) signature genes. NES, IRF8 + + + + + + normalized enrichment score. C,

Gapdh qRT-PCR. Values are from two

(/ Clec9a Xcr1 Tlr3 independent experiments, in which 0.06 0.08 0.08 each PCR was performed in 0.06 0.06 triplicate. D, GSEA for dendritic cell mRNA expression level expression mRNA 0.04 maturation-related signature 0.04 0.04 0.02 genes. 0.02 0.02 0.00 0.00 0.00 BCR-ABL + + Lin− + + Lin− + + Lin− IRF8 + + + + + +

D DC maturation-related genes MIG + BCR-ABL MIG + BCR-ABL MICD8 BCR-ABL BCR-ABL +IRF8 MICD8 +IRF8 P < 0.001 P 0.0 0.0 0.4 = 0.007 P < 0.001 NES = –1.7 NES = 1.61 –0.2 NES = –2.42 –0.2 0.2 0.0 –0.4 –0.4 –0.2 –0.6 Enrichment score 0 50 100 150 200 0 50 100 150 200 0 50 100 150 200 Rank in ordered dataset (× 100)

þ IRF8-rescued BCR-ABL dendritic cells expressed higher BCR-ABL rather than exogenous IRF8. These results lead us levels of dendritic cell maturation-related genes (38) than to speculate that IRF8-rescued, BCR-ABL–expressing den- control dendritic cells (Fig. 4D, right). This was even more dritic cells may be in a "preactivated" state. Other predicted þ apparent in IRF8-untransduced BCR-ABL cells compared transcription factors includedGATA1andSTAT5(seeDis- þ with control dendritic cells and IRF8-rescued BCR-ABL cussion). Indeed, the expression of Gata1 itself was 15-fold dendritic cells (Fig. 4D, left and middle), suggesting that higher in BCR-ABL–transduced cells than in control den- dendritic cell maturation-related genes were upregulated by dritic cells, regardless of IRF8 supplementation.

6648 Cancer Res; 73(22) November 15, 2013 Cancer Research

A IFN-α IL-12p40 * ** ** ** ** ** ** ** ** * * * 60 45 ) 4

2 Ctrl Ctrl Ctrl CpG-A CpG-B LPS 10 3 40 30 × 2

Figure 5. Functionality of IRF8- ng/mL þ 20 15 rescued BCR-ABL dendritic cells. 1 Double transduction of Lin cells pg/mL ( 0 0 0 followed by 7-day Flt3L culture was BCR-ABL + + + + + + performed as in Fig. 3. A, cytokine þ IRF8 + + + + + + production. FACS-sorted GFP þ hCD8t cells were stimulated with MHC-II the indicated TLR ligands for 24 MHC-I ** hours and their supernatants were BC** 18 ** ** 10 ** analyzed by ELISA. B, expression ) ** ** 3 levels of antigen presentation– 8 10 12 6 ** × related molecules determined by × 80 ﬂow cytometry. C, induction of 6 4 60 CTLs. Antigen-speciﬁc MFl ( 2 cytotoxicity induced by 0 0 þ þ 40 FACS-sorted GFP hCD8t BCR-ABL + + + + cells was analyzed. All values IRF8 + + + + 20 killing ratio are from three independent CD80 CD86 CD40 OVA-specific % 0 experiments. , P < 0.05 and ** ** P t , < 0.01 (Student test). 150 ** 80 ** 5 BCR-ABL + + ) 2 60 4 IRF8 + + 10 100

× 3 40 50 2 20 MFl ( 1 0 0 0 BCR-ABL + +++ + + IRF8 + + + + + +

þ IRF8-rescued BCR-ABL dendritic cells have higher cell surface, suggesting the occurrence of posttranscriptional cytokine production and CTL induction capacities control. The expression level of another costimulatory mole- We next evaluated the functionality of the rescued dendritic cule CD40 was low and not affected by BCR-ABL. Thus, IRF8- þ cells. To examine their ability to produce IFN-a and IL-12p40, rescued BCR-ABL dendritic cells expressed higher levels of transduced Lin cells followed by 7-day Flt3L culture were MHC-I and some costimulatory molecules than regular den- stimulated with TLR9 ligands CpG-A (for pDCs) or CpG-B (for dritic cells. þ cDCs), or a TLR4 ligand lipopolysaccharide (LPS; Fig. 5A). TLRs CD8a dendritic cells have been shown to be critical for recognize not only the pathogen-associated molecular pat- eliminating tumor cells by priming CTLs via cross-presenta- þ terns (PAMP) but also damage-associated molecular patterns tion in mice (41). In addition, CD8a dendritic cells require (DAMP) that may be involved in antitumor immunity (39). type I IFN to induce CTLs that reject tumors (42, 43). These BCR-ABL almost completely abolished production of these previous findings, together with the above data on cytokine cytokines, but their expression was restored by coexpression of production and MHC-related molecules, prompted us to exam- þ IRF8. Interestingly, IRF8-rescued BCR-ABL dendritic cells ine the ability to cross-present an antigen (OVA) to induce a produced even higher levels of these cytokines than control CTL response (Fig. 5C). Transduced Lin cells followed by 7- dendritic cells, especially for IL-12p40, whose expression is day Flt3L culture were pulsed with OVA protein, and then dependent on NF-kB and IRF8 (19, 40). mixed with T cells from OT-I mice. The primed T cells were We also examined the expression of MHC-related molecules then mixed with target splenocytes loaded with OVA peptide, by flow cytometry. BCR-ABL strongly inhibited MHC-II expres- and OVA-specific CTL activity was calculated. Although con- sion, but increased MHC-I expression 4.1-fold (Fig. 5B and trol dendritic cells efficiently induced CTLs, BCR-ABL–trans- þ Supplementary Fig. S3). When IRF8 expression was supple- duced cells failed to do so. Strikingly, IRF8-rescued BCR-ABL mented, MHC-II expression was recovered, and MHC-I expres- dendritic cells regained the ability to induce CTLs, and the sion remained high. We also found that expression of the observed response was even stronger than that of control costimulatory molecules CD80 and CD86 was significantly dendritic cells. induced by BCR-ABL and remained high in IRF8-cotransduced Transduction of IRF8 without BCR-ABL did not augment the cells. We noted that the mRNA levels of MHC-I, CD80, and above functional parameters. Thus, these results suggest that CD86 did not correlate with protein expression levels on the BCR-ABL harbors an immune-stimulatory potential that is

www.aacrjournals.org Cancer Res; 73(22) November 15, 2013 6649

manifested when dendritic cell differentiation is restored by significant portion of BCR-ABL–dysregulated genes, leading to supplementation with IRF8. efficient recovery of dendritic cell differentiation. Thus, the suppression of Irf8 expression is the cause of impaired den- Imatinib partially restores Irf8 expression, but cancels dritic cell development by BCR-ABL, which we infer accounts, the enhancement of functionality in IRF8-rescued at least in part, for the inability of patients with CML to þ BCR-ABL dendritic cells establish antileukemic immunity. To investigate whether clinically available TKIs can restore Gene expression profiling indicated that the expression of a Irf8 expression as well as dendritic cell differentiation and significant numbers of NF-kB target genes was increased by þ function, imatinib, nilotinib, and dasatinib were added at BCR-ABL, and predicted that IRF8-rescued BCR-ABL den- concentrations two times higher than those required for dritic cells may be in a preactivated state. In fact, we found that 90% inhibition (IC90) of BCR-ABL kinase activity during the the rescued dendritic cells had higher cytokine production and Flt3L treatment of transduced Lin cells. These TKIs partially CTL induction capacity than normal dendritic cells, and such restored the expression of Irf8 expression in BCR-ABL–trans- enhancement was completely canceled by imatinib. These duced cells (Fig. 6A, top). Efficient inhibition of BCR-ABL results indicate that BCR-ABL harbors an immune-stimulatory kinase activity by imatinib was confirmed by immunoblot potential, which would not be elicited in the natural course of analysis for autophosphorylation of BCR-ABL (Fig. 6A, bot- CML or during the therapy with TKIs. Thus, IRF8 seems to tom). Presumably, due to the modest but significant increase in "convert" BCR-ABL from a dendritic cell suppressor to a Irf8 expression, imatinib restored cDC differentiation (Fig. 6B). dendritic cell activator. However, the recovery of pDC differentiation and MHC-II We have previously shown that IRF8 overrides the mitogenic expression was not complete (Fig. 6B and C). In addition, the activity of BCR-ABL in differentiating myeloid progenitor cells recovery of cytokine production and CTL induction was very (30). It has also been shown that IRF8 antagonizes the anti- modest (Fig. 6D). Interestingly, imatinib canceled the enhance- apoptotic effect of BCR-ABL (44). Moreover, coexpression of ment of dendritic cell functionality by BCR-ABL in IRF8- IRF8 inhibits BCR-ABL–induced myeloid hyperplasia in vivo þ rescued BCR-ABL dendritic cells; MHC-I, CD80, and CD86 (31). Neutrophil differentiation is promoted by BCR-ABL (45), expression, cytokine production, and CTL induction all but potently inhibited by IRF8 (24). Therefore, we expect the returned to unenhanced levels. Thus, BCR-ABL augments restoration of IRF8 to have two distinct effects. One is to dendritic cell functionality via its tyrosine kinase activity. We directly reduce tumor burden by inhibiting abnormally also noted that in control dendritic cells, TKIs modestly enhanced growth and survival of myeloid cells as well as inhibited Irf8 expression, and that imatinib slightly suppressed disproportionate neutrophil differentiation. The other is to the ability of control dendritic cells to of induce CTLs (Fig. 6E). induce effective anti-CML immunity by restoring dendritic cell Overall, these data suggest that dendritic cells rescued by IRF8 development and by cooperating with BCR-ABL to strengthen are more functional than those rescued by TKIs, at least under dendritic cell functionality. In this regard, IRF8 may be the key the condition used in this study. to establishing multidisciplinary therapy for CML. To elucidate the pathway through which BCR-ABL inhibits We propose that the selective restoration of IRF8 will serve Irf8 expression, we tested whether a STAT5 inhibitor, a histone as a hallmark for new CML therapies that may overcome the deacetylase inhibitor (TSA), or a DNA methytransferase inhib- problems with current TKI therapy. First, the effects of IRF8 itor (5-Aza) improved Irf8 expression and dendritic cell differ- supplementation are not affected by TKI-resistant mutations entiation. TSA and 5-Aza were tested because we speculated in BCR-ABL. Second, unlike TKIs, IRF8 therapy would generate þ that epigenetic mechanisms may be involved, given the dis- BCR-ABL dendritic cells with enhanced functionality to crepancy between the effects of the kinase-dead BCR-ABL– induce anti-CML CTL responses. Finally, once anti-CML CTLs mutant and of imatinib treatment after BCR-ABL expression are induced, it would be possible to discontinue the IRF8 on Irf8 expression. However, none of these agents restored Irf8 therapy. It should be noted, however, that the possibility of expression or dendritic cell differentiation in BCR-ABL–trans- autoimmune reactions by these hyperfunctional dendritic cells duced cells, and instead tended to suppress these events in will need to be evaluated. control cells (Supplementary Fig. S4). Combinations of TSA or To develop a method for restoration of IRF8 expression, it 5-Aza with imatinib also failed to rescue Irf8 expression and will be essential to understand the mechanism through which dendritic cell differentiation. BCR-ABL inhibits IRF8 expression and dendritic cell development. Our data seem to provide some insights into these mechanisms. Unlike Irf8 / mice, which show severe reduc- Discussion tion in CDP counts but not in MDP counts (23, 46), CML model In this study, the antagonistic relationship between BCR- mice lack both MDPs and CDPs. Thus, the dendritic cell ABL and IRF8 was investigated in the context of dendritic cell differentiation stage affected by BCR-ABL is upstream of that development in CML. Our results revealed that BCR-ABL affected by the loss of IRF8. This suggests that BCR-ABL acts on abolishes dendritic cell development at the level of generating machinery that is required for the expression of Irf8 (the MDPs. This causes severe reductions in dendritic cell counts mechanism of which has not been clarified yet) as well as the for all subsets in CML mice, consistent with previous findings generation of MDPs. in patients with CML (14, 15). BCR-ABL inhibits Irf8 expression, One possibility raised from the above context is inhibition of and supplementation with IRF8 restores the expression of a the transcription factor PU.1 by GATA1. PU.1 is expressed in

6650 Cancer Res; 73(22) November 15, 2013 Cancer Research

GFP+ hCD8t+ cells A B MIG + MICD8 8FRI BCR-ABL 8FRI+LBA-RCB

Irf8 mRNA 73.3% 72.4 9.6 46.8 ** ) * * Ctrl

** MHC-II

Gapdh 0.20 0.15 0.10 72.1 67.7 68.9 66.8 0.05

0.00 Imatinib MIG BCR- ABL Expression level Expression level (/ Ctrl CD11c Imatinib GFP+ hCD8t+ CD11c+ cells Nilotinib MIG + MICD8 IRF8 BCR-ABL BCR-ABL + IRF8 Dasatinib

Figure 6. The effects of TKIs on dendritic cell differentiation, Ctrl 9.8 7.4 0.04 10.5 function, and Irf8 expression. CD11b (9.1) (7.2) (0.01) (7.4) Lin cells were singly transduced BCR-ABL BCR-ABL + imatinib MIG as in Fig. 2 or doubly transduced as Tyr-P in Fig. 3 and cultured in the BCR-ABL presence of Flt3L for 7 days. TKIs b (imatinib at 2 mmol/L, nilotinib at -Actin Imatinib 9.4 5.1 4.9 9.6 50 nmol/L, or dasatinib at 5 nmol/L) (8.6) (4.8) (3.8) (7.7) were added during the Flt3L culture. Their differentiation status B220 and function were analyzed as C MHC-II MHC-I CD80 in Figs. 2 and 5. In A, D, and E, ** * ** ** ** ** 16 100 12 ) )

transduced cells were FACS- ) 2 3 sorted before the Flt3L culture. 12 75 3 9 10 10 10 × × × A, qRT-PCR (top) and 8 50 × 6 immunoblot analysis (bottom). 4 25 3 MFI ( MFI MFI ( MFI B, representative data of ﬂow- ( MFI cytometric analysis. C, expression 0 0 0 – BCR-ABL + +++ + + levels of antigen presentation IRF8 + + + + + + related molecules. D, cytokine CD86 ** ** CD40 production. E, induction of CTLs. 8 500 )

Imatinib was added to the Flt3L 3 6 400 Ctrl culture of dendritic cells but not 10 300 Imatinib ×

4 MFI to the coculture with T cells. All 200 values in bar graphs are from 2 100 MFI ( MFI three independent experiments. 0 0 , P < 0.05 and , P < 0.01 BCR-ABL + + + + (Student t test). IRF8 + + + + D IL-12p40 60 ** ** * 80 ** ** * Ctrl 60 Ctrl 40 CpG-B LPS Imatinib 40 Imatinib

ng/mL 20 Imatinib Imatinib +CpG-B 20 +LPS 0 0 BCR-ABL + + + + IRF8 + + + + α IFN- E ** ** ** 400 ** 80

300 Ctrl 60 * Ctrl CpG-A Imatinib 200 Imatinib 40 pg/mL Imatinib 100 20

+CpG-A ratio killing

0 OVA-specific % 0 BCR-ABL + + BCR-ABL + + IRF8 + + IRF8 + +

www.aacrjournals.org Cancer Res; 73(22) November 15, 2013 6651

similar sets of hematopoietic cell lineages as IRF8 and is Disclosure of Potential Conflicts of Interest required for the generation of MDPs (35). PU.1 activity is No potential conflicts of interest were disclosed. inhibited by GATA1, which is essential for the generation of MEPs (47). Our data showed that BCR-ABL strongly induced Authors' Contributions Conception and design: T. Tamura Gata1 expression. Moreover, GATA1 targets were indeed sig- Development of methodology: T. Watanabe, C. Hotta, T. Tamura nificantly enriched in genes induced by BCR-ABL and that the Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): T. Watanabe, C. Hotta, S.-I. Koizumi, M. Nakazawa, percentage of MEPs was increased in BCR-ABL mice. Inter- H. Fujita, R. Sakai, S. Fujisawa, Z. Ikezawa, M. Aihara, Y. Ishigatsubo estingly, GATA1 has been shown to inhibit Irf8 expression, Analysis and interpretation of data (e.g., statistical analysis, biostatistics, possibly by preventing PU.1 recruitment to the Irf8 promoter in computational analysis): S.-I. Koizumi, J. Nakabayashi, A. Nishiyama, M. Aihara, T. Tamura dendritic cells (48). Moreover, GATA1 activity can be augment- Writing, review, and/or revision of the manuscript: T. Watanabe, S.-I. ed via PI3K /AKT, a signaling pathway stimulated by BCR-ABL Koizumi, T. Tamura Administrative, technical, or material support (i.e., reporting or orga- (49). Therefore, inhibition of PU.1 through induction and nizing data, constructing databases): C. Hotta, K. Miyashita, D. Kurotaki, activation of GATA1 by BCR-ABL is an attractive possibility. G.R. Sato, M. Yamamoto, A. Nishiyama Other possibilities included the involvement of STAT5. It has Study supervision: T. Tamura been reported that GM-CSF inhibits Irf8 expression via STAT5 (50). Thus, it was tempting to speculate that BCR-ABL–acti- Acknowledgments The authors thank Drs. Warren Pear, Ricardo A. Feldman, and Huabao Xiong vated STAT5 suppresses IRF8 expression. However, our experi- for providing retroviral vectors. ments using the STAT5 inhibitor did not support this. Finally, even though we observed that TSA or 5-Aza, as single agents or Grant Support in combination with imatinib, failed to rescue Irf8 expression, it This work was supported by KAKENHI Grants-in-Aid 24390246 and 24118002 from the Japan Society for the Promotion of Science, a Special Coordination Fund is still important to consider the possibility that other histone for Promoting Science and Technology from the Ministry of Education, Culture, modifications, such as methylation, may be involved. Sports, Science and Technology (MEXT) of Japan and grants from the Yokohama In conclusion, our present study provides new insights into Foundation for Advancement of Medical Science and the Kato Memorial Bioscience Foundation (T. Tamura). CML pathogenesis and the basis of new therapy for CML in The costs of publication of this article were defrayed in part by the payment terms of dendritic cell biology. Further investigation of the of page charges. This article must therefore be hereby marked advertisement mechanisms through which BCR-ABL suppresses IRF8 expres- in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. sion will be required to develop a method to selectively restore Received March 20, 2013; revised August 15, 2013; accepted September 7, 2013; IRF8 expression. published online November 15, 2013.

References 1. Sawyers CL. Chronic myeloid leukemia. N Engl J Med 1999;340: 12. Lim SH, Coleman S. Chronic myeloid leukemia as an immunological 1330–40. target. Am J Hematol 1997;54:61–7. 2. Jabbour E, Kantarjian H. Chronic myeloid leukemia: 2012 Update on 13. Kolb HJ, Mittermuller J, Clemm C, Holler E, Ledderose G, Brehm G, diagnosis, monitoring, and management. Am J Hematol 2012;87: et al. Donor leukocyte transfusions for treatment of recurrent chronic 1037–45. myelogenous leukemia in marrow transplant patients. Blood 1990;76: 3. Wong S, Witte ON. The BCR-ABL story: bench to bedside and back. 2462–5. Annu Rev Immunol 2004;22:247–306. 14. Mohty M, Isnardon D, Vey N, Briere F, Blaise D, Olive D, et al. Low blood 4. O'Hare T, Zabriskie MS, Eiring AM, Deininger MW. Pushing the limits of dendritic cells in chronic myeloid leukaemia patients correlates with targeted therapy in chronic myeloid leukaemia. Nat Rev Cancer 2012; loss of CD34þ/CD38 primitive haematopoietic progenitors. Br J 12:513–26. Haematol 2002;119:115–8. 5. Appel S, Balabanov S, Brummendorf TH, Brossart P. Effects of 15. Boissel N, Rousselot P, Raffoux E, Cayuela JM, Maarek O, Charron D, imatinib on normal hematopoiesis and immune activation. Stem Cells et al. Defective blood dendritic cells in chronic myeloid leukemia 2005;23:1082–8. correlate with high plasmatic VEGF and are not normalized by imatinib 6. Krusch M, Salih HR. Effects of BCR-ABL inhibitors on anti-tumor mesylate. Leukemia 2004;18:1656–61. immunity. Curr Med Chem 2011;18:5174–84. 16. Eisendle K, Lang A, Eibl B, Nachbaur D, Glassl H, Fiegl M, et al. 7. Fujita H, Kitawaki T, Sato T, Maeda T, Kamihira S, Takaori-Kondo A, Phenotypic and functional deficiencies of leukaemic dendritic cells et al. The tyrosine kinase inhibitor dasatinib suppresses cytokine from patients with chronic myeloid leukaemia. Br J Haematol 2003; production by plasmacytoid dendritic cells by targeting endosomal 120:63–73. transport of CpG DNA. Eur J Immunol 2013;43:93–103. 17. Dong R, Cwynarski K, Entwistle A, Marelli-Berg F, Dazzi F, Simpson E, 8. Steinman RM, Banchereau J. Taking dendritic cells into medicine. et al. Dendritic cells from CML patients have altered actin organization, Nature 2007;449:419–26. reduced antigen processing, and impaired migration. Blood 2003; 9. Liu K, Nussenzweig MC. Origin and development of dendritic cells. 101:3560–7. Immunol Rev 2010;234:45–54. 18. Mumprecht S, Claus C, Schurch€ C, Pavelic V, Matter MS, Ochsenbein 10. Molldrem JJ, Lee PP, Wang C, Felio K, Kantarjian HM, Champlin RE, AF. Defective homing and impaired induction of cytotoxic T cells by et al. Evidence that specific T lymphocytes may participate in the BCR/ABL-expressing dendritic cells. Blood 2009;113:4681–9. elimination of chronic myelogenous leukemia. Nat Med 2000;6: 19. Tamura T, Yanai H, Savitsky D, Taniguchi T. The IRF family transcrip- 1018–23. tion factors in immunity and oncogenesis. Annu Rev Immunol 2008; 11. Burchert A, Wol€ fl S, Schmidt M, Brendel C, Denecke B, Cai D, et al. 26:535–84. Interferon-a, but not the ABL-kinase inhibitor imatinib (STI571), 20. Aliberti J, Schulz O, Pennington DJ, Tsujimura H, Reis e Sousa C, induces expression of myeloblastin and a specific T-cell response in Ozato K, et al. Essential role for ICSBP in the in vivo development of þ chronic myeloid leukemia. Blood 2003;101:259–64. murine CD8a dendritic cells. Blood 2003;101:305–10.

6652 Cancer Res; 73(22) November 15, 2013 Cancer Research

21. Tamura T, Tailor P, Yamaoka K, Kong HJ, Tsujimura H, O'Shea JJ, 36. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette et al. IFN regulatory factor-4 and -8 govern dendritic cell subset MA, et al. Gene set enrichment analysis: a knowledge-based approach development and their functional diversity. J Immunol 2005;174: for interpreting genome-wide expression profiles. Proc Natl Acad Sci U 2573–81. S A 2005;102:15545–50. 22. Schiavoni G, Mattei F, Sestili P, Borghi P, Venditti M, Morse HC III, et al. 37. Miller JC, Brown BD, Shay T, Gautier EL, Jojic V, Cohain A, et al. ICSBP is essential for the development of mouse type I interferon- Deciphering the transcriptional network of the dendritic cell lineage. producing cells and for the generation and activation of CD8a(þ) Nat Immunol 2012;13:888–99. dendritic cells. J Exp Med 2002;196:1415–25. 38. Lindstedt M, Johansson-Lindbom B, Borrebaeck CA. Global repro- 23. Kurotaki D, Osato N, Nishiyama A, Yamamoto M, Ban T, Sato H, et al. gramming of dendritic cells in response to a concerted action of Essential role of the IRF8-KLF4 transcription factor cascade in murine inflammatory mediators. Int Immunol 2002;14:1203–13. monocyte differentiation. Blood 2013;121:1839–49. 39. Green DR, Ferguson T, Zitvogel L, Kroemer G. Immunogenic and 24. Tamura T, Nagamura-Inoue T, Shmeltzer Z, Kuwata T, Ozato K. ICSBP tolerogenic cell death. Nat Rev Immunol 2009;9:353–63. directs bipotential myeloid progenitor cells to differentiate into mature 40. Murphy TL, Cleveland MG, Kulesza P, Magram J, Murphy KM. Reg- macrophages. Immunity 2000;13:155–65. ulation of interleukin 12 p40 expression through an NF-kB half-site. Mol 25. Miyagawa F, Zhang H, Terunuma A, Ozato K, Tagaya Y, Katz SI. Cell Biol 1995;15:5258–67. Interferon regulatory factor 8 integrates T-cell receptor and cyto- 41. Hildner K, Edelson BT, Purtha WE, Diamond M, Matsushita H, þ kine-signaling pathways and drives effector differentiation of CD8 T Kohyama M, et al. Batf3 deficiency reveals a critical role for CD8a cells. Proc Natl Acad Sci U S A 2012;109:12123–8. dendritic cells in cytotoxic T cell immunity. Science 2008;322: 26. Holtschke T, Lohler J, Kanno Y, Fehr T, Giese N, Rosenbauer F, et al. 1097–100. Immunodeficiency and chronic myelogenous leukemia-like syndrome 42. Fuertes MB, Kacha AK, Kline J, Woo SR, Kranz DM, Murphy KM, in mice with a targeted mutation of the ICSBP gene. Cell 1996;87: et al. Host type I IFN signals are required for antitumor CD8þ T cell 307–17. responses through CD8aþ dendritic cells. J Exp Med 2011;208: 27. Hambleton S, Salem S, Bustamante J, Bigley V, Boisson-Dupuis S, 2005–16. Azevedo J, et al. IRF8 mutations and human dendritic-cell immuno- 43. Diamond MS, Kinder M, Matsushita H, Mashayekhi M, Dunn GP, deficiency. N Engl J Med 2011;365:127–38. Archambault JM, et al. Type I interferon is selectively required by 28. Schmidt M, Nagel S, Proba J, Thiede C, Ritter M, Waring JF, et al. Lack dendritic cells for immune rejection of tumors. J Exp Med 2011;208: of interferon consensus sequence binding protein (ICSBP) transcripts 1989–2003. in human myeloid leukemias. Blood 1998;91:22–9. 44. Burchert A, Cai D, Hofbauer LC, Samuelsson MK, Slater EP, Duyster J, 29. Schmidt M, Hochhaus A, Nitsche A, Hehlmann R, Neubauer A. et al. Interferon consensus sequence binding protein (ICSBP; IRF-8) Expression of nuclear transcription factor interferon consensus antagonizes BCR/ABL and down-regulates bcl-2. Blood 2004;103: sequence binding protein in chronic myeloid leukemia correlates with 3480–9. pretreatment risk features and cytogenetic response to interferon- 45. Kobayashi S, Kimura F, Ikeda T, Osawa Y, Torikai H, Kobayashi A, et al. alpha. Blood 2001;97:3648–50. BCR-ABL promotes neutrophil differentiation in the chronic phase of 30. Tamura T, Kong HJ, Tunyaplin C, Tsujimura H, Calame K, Ozato chronic myeloid leukemia by downregulating c-Jun expression. Leu- K. ICSBP/IRF-8 inhibits mitogenic activity of p210 Bcr/Abl kemia 2009;23:1622–7. in differentiating myeloid progenitor cells. Blood 2003;102: 46. Becker AM, Michael DG, Satpathy AT, Sciammas R, Singh H, Bhat- 4547–54. tacharya D. IRF-8 extinguishes neutrophil production and promotes 31. Hao SX, Ren R. Expression of interferon consensus sequence binding dendritic cell lineage commitment in both myeloid and lymphoid protein (ICSBP) Is downregulated in Bcr-Abl-induced murine chronic mouse progenitors. Blood 2012;119:2003–12. myelogenous leukemia-like disease, and forced coexpression of 47. Zhang P, Behre G, Pan J, Iwama A, Wara-Aswapati N, Radomska ICSBP inhibits Bcr-Abl-induced myeloproliferative disorder. Mol Cell HS, et al. Negative cross-talk between hematopoietic regulators: Biol 2000;20:1149–61. GATA proteins repress PU.1. Proc Natl Acad Sci U S A 1999;96: 32. Deng M, Daley GQ. Expression of interferon consensus sequence 8705–10. binding protein induces potent immunity against BCR/ABL-induced 48. Shimokawa N, Nishiyama C, Nakano N, Maeda K, Suzuki R, Hara M, leukemia. Blood 2001;97:3491–7. et al. Suppressive effects of transcription factor GATA-1 on cell type- 33. Nardi V, Naveiras O, Azam M, Daley GQ. ICSBP-mediated immune specific gene expression in dendritic cells. Immunogenetics 2010;62: protection against BCR-ABL-induced leukemia requires the CCL6 and 421–9. CCL9 chemokines. Blood 2009;113:3813–20. 49. Zhao W, Kitidis C, Fleming MD, Lodish HF, Ghaffari S. Erythropoietin 34. Nakagawa Y, Watari E, Shimizu M, Takahashi H. One-step simple stimulates phosphorylation and activation of GATA-1 via the PI3- assay to determine antigen-specific cytotoxic activities by single-color kinase/AKT signaling pathway. Blood 2006;107:907–15. flow cytometry. Biomed Res 2011;32:159–66. 50. Esashi E, Wang YH, Perng O, Qin XF, Liu YJ, Watowich SS. The 35. Geissmann F, Manz MG, Jung S, Sieweke MH, Merad M, Ley K. signal transducer STAT5 inhibits plasmacytoid dendritic cell devel- Development of monocytes, macrophages, and dendritic cells. Sci- opment by suppressing transcription factor IRF8. Immunity 2008; ence 2010;327:656–61. 28:509–20.

www.aacrjournals.org Cancer Res; 73(22) November 15, 2013 6653

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 2013 American Association for Cancer Research. The Transcription Factor IRF8 Counteracts BCR-ABL to Rescue Dendritic Cell Development in Chronic Myelogenous Leukemia

Tomoya Watanabe, Chie Hotta, Shin-ichi Koizumi, et al.

Cancer Res 2013;73:6642-6653.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/73/22/6642

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2021/03/20/73.22.6642.DC1

Cited articles This article cites 50 articles, 25 of which you can access for free at: http://cancerres.aacrjournals.org/content/73/22/6642.full#ref-list-1

Citing articles This article has been cited by 1 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/73/22/6642.full#related-urls

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 Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/73/22/6642. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.