Inhibition of Notch Signaling Enhances Chemosensitivity in B

Published OnlineFirst December 18, 2018; DOI: 10.1158/0008-5472.CAN-18-1617

Cancer Translational Science Research

Inhibition of Notch Signaling Enhances Chemosensitivity in B-cell Precursor Acute Lymphoblastic Leukemia Paul Takam Kamga1,2, Giada Dal Collo1, Martina Midolo1, Annalisa Adamo1, Pietro Delﬁno3, Angela Mercuri1,4, Simone Cesaro4, Elda Mimiola1, Massimiliano Bonifacio1, Angelo Andreini1, Marco Chilosi5, and Mauro Krampera1

Abstract

Notch3 and Notch4 support survival of primary B-cell acute death in B-ALL cells by upregulating intracellular levels of lymphoblastic leukemia (B-ALL) cells, suggesting a role for reactive oxygen species, which in turn modulated mTOR, Notch signaling in drug response. Here we used in vitro, in silico, NF-kB, and ERK expression. In NOG-mouse-based xenograft and in vivo mouse xenograft model-based approaches to define models of B-ALL, co-administration of the Notch inhibitor the role of the Notch pathway in B-ALL chemosensitivity. We GSI-XII with the chemotherapeutic agent Ara-C lowered bone observed significant Notch receptor and ligand expression in marrow leukemic burden compared with DMSO or Ara-C B-ALL primary cells and cell lines. Primary leukemia cells from alone, thus prolonging mouse survival. Overall, our results high-risk patients overexpressed Notch3, Notch4, and Jagged2 support the potential effectiveness of Notch inhibitors in while displaying a reduction in expression levels of Notch1-4 patients with B-ALL. following chemotherapy. We then analyzed in vitro cell survival of B-ALL cells treated with conventional chemothera- Significance: Inhibition of Notch signaling enhances the peutic agents alone or in combination with Notch signaling chemosensitivity of B-ALL cells, suggesting Notch inhibition as inhibitors. Gamma-secretase inhibitors (GSI) and anti- a potential therapeutic strategy to improve the outcome of Notch4 were all capable of potentiating drug-induced cell patients with B-ALL.

Introduction survival and chemoresistance of leukemic cells (2–6). Notch signaling consists of four receptors (Notch1–4) and five ligands, B-cell acute lymphoblastic leukemia (B-ALL) is the most com- including Jagged1-2, DLL-1, and DLL-3-4. Activation of Notch mon leukemia in pediatric population; nowadays, more than pathway occurs after binding of one of the five ligands to one of 90% of all patients obtain disease complete remission following the four receptors (7), leading to the formation of the active form treatment, but the prognosis for relapsed/refractory patients or of the receptor called "Notch intracellular domain" (NICD). In the adult B-ALL patients is still poor (1). The availability of efficient canonical sequence of the pathway, NICD enters into the nucleus therapeutic options for the latter subgroups still represents an to activate target genes (7, 8). Notch pathway controls cell unmet need. Patient stratification based on cytogenetics, age, proliferation, survival, and differentiation, being involved in cell white blood counts, or other factors explains only partially why fate determination in normal and pathologic tissues (8). A cross- some patients classified at diagnosis as "low-risk" undergo relapse talk between Notch signaling and the control of oxidative state has after treatment. In addition to molecular aberrations affecting been proposed as a critical mechanism in tissue development and leukemic cells, a role for developmental signaling pathways, such cancer. Notably, Notch signaling lowers the levels of reactive as Hedgehog, Wnt, and Notch, has been shown in promoting oxygen species (ROS) in cancer cells, improving cancer cell survival and chemoresistance (9). Although an oncogenic role 1Stem Cell Research Laboratory, Section of Hematology, Department of for Notch gene mutations in T-acute lymphoblastic leukemia Medicine, University of Verona, Verona, Italy. 2EA4340-BCOH: Biomarker (T-ALL) development has been demonstrated, there is no clear in Cancerology and Onco-Hematology, Universite de Versailles Saint- evidence of Notch mutations in B-ALL. Nevertheless, Notch Quentin-en-Yvelines, Versailles, France. 3Department of Biotechnology, Univer- receptor genes and the Notch target gene HES1 are frequently sity of Verona, Verona, Italy. 4Pediatric Onco-Hematology Unit, University of – 5 hypermethylated, leading to the inactivation of the Notch Hes1 Verona, Verona, Italy. Section of Pathology, Department of Diagnostics and axis (10). Enforcing through lentiviral vectors the expression of Public Health, University of Verona, Verona, Italy. either active forms of Notch receptors or Hes1 induces growth Note: Supplementary data for this article are available at Cancer Research arrest in murine and human B-ALL cells (11). These results, based Online (http://cancerres.aacrjournals.org/). on Notch artificial activation, have not been validated by loss-of- Corresponding Author: Mauro Krampera, University of Verona, Policlinico G.B. function approaches, pharmacologic interference or animal mod- Rossi - P.le L.A. Scuro, 10, 37134 Verona, Italy. Phone: 39-045-8124034; Fax: 39- els. Instead, it has been observed that Notch ligands protect B cells 045-8027488; E-mail: [email protected]; www.stemcellreslab-verona.it from apoptosis in germinal centers (12). Our previous doi: 10.1158/0008-5472.CAN-18-1617 works extended the knowledge on this protective role for Notch 2018 American Association for Cancer Research. signaling in B-cell malignancies, including B-ALL and B-cell

www.aacrjournals.org 639

Takam Kamga et al.

chronic lymphocytic leukemia (CLL; refs. 13, 14). We previously labeled with appropriate antibody, and acquired by LAS4000 observed high expression of Notch ligands and receptors in a (GE Healthcare) instrument. GAPDH was used as loading control. subset of 12 patients with B-ALL, as well as in stromal cells from All samples subjected to immunoblotting contained more than patients with B-ALL. This evidence suggested a Notch paracrine 80% of leukemia cells. The specificity of all Notch antibodies signaling between B-ALL cells and their microenvironment, whose was validated on three control lysates, as previously described abrogation through Notch inhibitors, such as gamma-secretase (5, 14, 15): HEK 293, HS27A stromal cell line, and CEM T-ALL inhibitors (GSI), resulted in B-ALL cells apoptosis (14). These cell line. previous observations suggested the involvement of Notch signaling not only in B-ALL cell survival, but also in chemosensitivity MTT viability assay (5, 13). In this study, we used in vitro, in silico, and in vivo To study the specific relative basal sensitivity of B-ALL cells to approaches to comprehensively define Notch signaling contribu- different agents, cells were seeded in 96-well plates and cultured tion to B-ALL cell response to chemotherapy. Overall, we dem- for 48 hours in presence of increasing concentrations of onstrated that Notch pathway inhibition sensitizes B-ALL cells to each modulator or drug. Then, the colorimetric 3-[4,5- drugs by increasing intracellular levels of ROS, thus promoting the dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT; reduction of B-ALL cell survival both in vitro and in vivo. These Sigma-Aldrich) assay was performed, as previously described findings represent a preclinical rationale for combination treat- (5, 15). ments including Notch inhibitors and chemotherapy. Cell proliferation and TOPRO-3 viability assay Cell proliferation was evaluated by carboxy fluorescein succi- Materials and Methods nimidyl ester (CFSE; Life Technologies) staining, as previously Chemicals and antibodies described (5, 15). Briefly, cells were washed twice with PBS and This information is detailed in Supplementary Materials and resuspended in 0.1% PBS-BSA, stained with CFSE (5 mmol/L) for Methods. 10 minutes in the dark at 37C, washed, and analyzed by flow cytometry. Relative cell proliferation was expressed as percentage Patients and samples of CFSE median fluorescence of treated cells compared with cells The characteristics of patients involved in this study are pre- treated with the specific vehicle. sented in Supplementary Table S1. This study was conducted in accordance with the declaration of Helsinki, all primary cell Measurement of ROS levels samples being collected from patients with B-ALL after written ROS levels were measured as previously described (16). Briefly, informed consent as approved by the Ethical Committee of cells were stained with 5 mmol/L cm-H2DCFDA (Life Technolo- Azienda Ospedaliera Universitaria Integrata Verona Italia (N. Prog. gies) for 30 minutes. Fluorescence of oxidized cm-H2DCFDA was 1828, May 12, 2010—"Institution of cell and tissue collection for measured by flow cytometry. biomedical research in Onco-Hematology"). Apoptosis Cell line culture and validation Apoptotic rate of B cells was assessed using FITC-Annexin B-ALL cell lines including VR-ALL, RS4;11, and SUP-B15 were V/propidium iodide (PI) staining, as previously described cultured in complete RPMI (RPMI supplemented with 10% FBS, (5, 14, 15). Briefly, cells were stained with APC-conjugated 1% L-glutamine, and 1% penicillin/streptomycin). VR-ALL cells anti-CD19 for 15 minutes, then resuspended in binding buffer were isolated, characterized, and validated in our laboratory, as (MiltenyiBiotec) and stained for 15 minutes with FITC-conjugated previously described (15), whereas RS4;11 and SUP-B15 cell lines Annexin V (MiltenyiBiotec) at 1 mg/mL concentration. Cells were were acquired from ATCC. Stability and identity of all B-ALL cell then analyzed by flow cytometry. lines were controlled during and at the end of the current study using flow cytometry of membrane marker, morphologic analysis Xenograft mouse model after Giemsa staining (15), and short tandem repeat (STR). NOD/Shi-scid/IL-2Rgnull (NOG) mice were purchased from The stability and identity of all other cell lines (including HEK Taconic and kept in pathogen-free conditions in the animal 293, HEK 293T, CEM, and Ramos) used as control in Western facility of the Interdepartmental Centre of Experimental Research immunoblot were validated only through flow cytometry of Service of the University of Verona. B-ALL cells (5 106) were membrane marker and morphologic analysis after Giemsa stain- injected into the tail vein of totally irradiated (1.2 Gy, 137Cesium ing. All cell lines were routinely tested to be Mycoplasma-free source), 8- to 12-week-old mice. At day 14 post-injection, mice using the Mycoplasma PCR Detection Kit from Sigma-Aldrich. were assigned to one of the following treatment arms: DMSO Once thawed cells were passaged every week and were not used (GSIs vehicles) or IgG isotype (0.5 mg/kg), Ara-C (100 mg/kg) þ for more than 3 months. This period corresponding to 8 to 12 DMSO or Ara-C þ IgG isotype, Ara-C þ GSI-XII mg/kg), or cell passages. Ara-C þ anti-Notch4 (0.5 mg/kg), all administered through intraperitoneal daily injection for 3 days. In case of combined Western blotting treatment, mice were first treated for 3 days with Ara-C followed Cells were lysed with appropriate amount of RIPA buffer by GSI-XII or anti-Notch4 for 3 days. Animals were sacrificed after (25 nmol/LTris pH 7.6, 150 mmol/L NaCl, 1% NP40, 1% 4 weeks from injection of cell lines, and bone marrow leukemic þ Na-deoxycholate, 0.1% SDS) supplemented with complete Pro- burden was evaluated as percentage of human CD19 cells. To tease Inhibitor (Roche) and 1 mmol/L Na3VO4. Proteins were assess the effect of N-acetylcysteine (NAC) on animal outcome, separated on a 10% polyacrylamide gel. Subsequently, proteins mice were treated for 6 days with NAC (150 mg/kg), starting from were transferred onto nitrocellulose membrane (GE Healthcare), the first day of Ara-C administration. To exclude any toxic effect of

640 Cancer Res; 79(3) February 1, 2019 Cancer Research

Notch, B-ALL Chemosensitivity

NAC, nontransplanted mice were also treated with NAC only. the Notch expression pattern in a larger cohort of patients (Sup- Animal experiments presented here were approved by the review plementary Table S1; ref. 51) and in B-ALL cell lines. We found board of the Italian Health Ministry. the expression of Notch1, Notch3, Notch4, Jagged2, DLL-3, and DLL-4, in B-ALL primary cells (Fig. 1A). Through Western blot Genomic sequencing and analysis analysis, we observed that Notch expression pattern was similar Genomic sequencing and analysis are detailed in Supplemen- both in primary cells and B-ALL cell lines (Fig. 1A and B). Lobry tary Materials and Methods. and colleagues (22) and Kanan and colleagues (23) previously reported that the presence of Notch receptors and ligands does not Notch gene methylation pattern always correlate with the activation of the pathway. We then Publicly available methylation data were downloaded from analyzed the expression of the Notch target gene HES1 in B-ALL NCBI GEO under accession number GSE49031 (17), and ana- cell lines and primary B-ALL cells from randomly chosen patients lyzed as described in Supplementary Materials and Methods. (n ¼ 21). We observed the expression of Hes1 in B-ALL cell lines as well as in 13/21 primary B-ALL cell samples (61.9%; Fig. 1C; Statistical analysis Supplementary Fig. S1A and S1B). In addition, proteins corre- Statistical analysis was performed using GraphPad Prism. sponding to other Notch target genes such as Hes5, Hey1, Deltex1, Mann–Whitney and Kruskal–Wallis were used to compare two and c-Myc, were also expressed in primary samples (Fig. 1C; groups or more than two groups, respectively. Pearson Chi-square Supplementary Fig. S1A). Treatment of cells lines with Notch analysis was used to test association between variables. inhibitors, such as SAHM, GSI-IX, and GSI-XII, reduced the band corresponding to Hes1, thus confirming that Hes1 expression in B-ALL cells arises at least in part from Notch activation (Supple- Results mentary Fig. S1C). We have previously shown that Notch3 and Mutational and epigenetic patterns of Notch genes in human Notch4 are the main active receptors in B-ALL samples (14). B-ALL B-ALL cells cell line lysates were probed with anti-Notch3 (D11B8) and The pathologic role of Notch signaling in T-ALL and CLL is -Notch4 (L5C5) antibodies, which recognize full length, and the often linked to the mutational status of the pathway. More than cleaved form of corresponding receptors (5) according to man- 50% of patients with T-ALL present an activating mutation of ufacturer's instruction and confirmed by the use of specific siRNA NOTCH1 (18, 19), whereas in CLL the occurrence of NOTCH1 against Notch3 and Notch4 (Supplementary Fig. S1D). B-ALL cell mutations in a subset of patients may explain the activation and lines expressed the active forms of Notch3 and Notch4, as con- therefore the oncogenic function of Notch signaling (4). To assess firmed by their protein level sensitivity to Notch inhibitors the presence of mutations in Notch family genes in B-ALL, we (Supplementary Fig. S1C). However, some samples lacking Hes1 performed whole genome sequencing in samples derived from six expression still showed the presence of Notch4 active form, patients. Consistently with other studies, we identified many suggesting a non-CSL coupling of Notch4 in B-ALL cells. mutations usually found in B-ALL, including those affecting NCOR1 and PAX5 genes (Supplementary Table S2; ref. 20). None Notch signaling overexpression in high-risk patient group of the activating Notch point mutations observed in T-ALL and Higher levels of Notch signaling have been described as prog- CLL was present in B-ALL samples. Considering the small number nostic factor in many hematologic malignancies, including, CLL of patients analyzed, we compared our results with those pub- and AML (24–26). We asked whether Notch expression could lished by other groups regarding genomic characterization of correlate with response to treatment or patient's overall survival. cytogenetic alterations in ALL samples (20). We found no evi- B-ALL patient risk stratification is based on many prognostic dence of Notch activating mutations in B-ALL. In one patient factors, including cytogenetics, age, and white blood cell count (Patient 14; ref. 15), we found missense variants both in NOTCH2 (WBC; ref. 27). We did not find any correlation between Notch (E38K) and JAGGED1 (P871R) genes as well as a putative muta- expression levels and cytogenetics or WBC, nor difference in tion in NOTCH1 (K1821N); this patient with multiple aberra- Notch expression between pediatric and adult patients. A research tions in Notch components was also affected by Alagille syn- group of St. Jude Children's Research Hospital classified as high- drome (15), a genetic disease characterized by multiple aberra- risk group, patients with blast cell persistence in bone marrow tions in JAGGED1 and/or NOTCH2 genes (Supplementary Table more than 42 days after treatment (28, 29). Thus, we classified cell S2); however, none of the aberrations found in this patient led to samples into two cohorts, that is low-risk group including patients the activation of the pathway (15). Aberrant DNA methylation is achieving complete remission within 42 days following induc- one of the main features of B-ALL cells and may influence the tion, and high-risk group, including patients with persistent course of the treatment (17, 21). Epigenetic status of Notch genes leukemia cells in the bone marrow after 42 days of treatment has been described elsewhere (GSE49031; ref. 17). Using bioin- (refractory; Supplementary Table S1). The analysis of Notch formatics tools, we analyzed the change in b-value of probes pathway expression in the samples collected from patients at corresponding to Notch genes in B-ALL cells, as compared to diagnosis revealed that the high-risk group presented higher levels normal B cells. We observed a variation in several probes when of Notch3, Notch4, and Jagged2 at diagnosis, as compared with leukemia cells at diagnosis were compared with normal B cells the low-risk group (Fig. 2). In addition, using Pearson Chi square (Supplementary Table S3), thus depicting a possible epigenetic analysis to seek association between Notch protein levels and regulation of Notch genes in B-ALL. drug response, we found a dependent association between Jag- ged1, Jagged2, and drug refractoriness (Supplementary Table S4). Notch components are highly expressed in B-ALL samples These observations suggested that Notch levels at diagnosis could We have described elsewhere the expression of Notch compo- play a role in chemosensitivity and therefore have a predictive nents in B-ALL samples (14). Here, we analyzed by flow cytometry value for the response to treatment.

www.aacrjournals.org Cancer Res; 79(3) February 1, 2019 641

Takam Kamga et al.

Figure 1. Notch expression and activation in B-ALL cell samples. A, Flow cytometric analysis of B-ALL cells (n ¼ 45) using fluorochrome-conjugated antibodies specificfor extracellular Notch receptors and ligands; data are represented as mean fluorescence intensity of each antibody normalized to specific fluorochrome-conjugated controls. B, Immunoblot of B-ALL cell lines (VR-ALL, RS4;11, and SUP-B15), the lymphoma cell line Ramos, and the T-ALL cell line CEM probed for Notch1-4, Jagged1-2, DLL-4, and Hes1 expression. Data are representative of six independent experiments. C, Representative expression of Notch targets (Hes1, Hes5, Hey1, and Deltex1) in primary B-ALL sample analyzed by flow cytometry using fluorochrome-conjugated antibodies Data are representative of 5 B-ALL cell samples (Supplementary Fig. S1A).

Notch expression pattern is modulated by chemotherapy treatment. Therefore, we focused the assessment of Notch To address a possible role for Notch signaling in B-ALL cell expression by focusing on the entire nonleukemic bone marrow response to chemotherapy, we investigated the changes in the cell population (CD19 cells) before and after the treatment, expression pattern of Notch components after chemotherapy. thus showing the reduction of Notch expression because of We have previously observed that Notch expression pattern in chemotherapy (Fig. 3A). Retrieving GSE49031 data, we B-ALL cells correlates with its expression in bone marrow observed that the epigenetic pattern of Notch genes changes stroma cells (14). Thus, considering the variability among from diagnosis to follow-up period, whereas patients at diag- patients, our aim was to assess whether, at any time point after nosis and relapsed patients had the same methylation pattern treatment, bone marrow cells (including stromal cells) were (Fig. 3B). This change in Notch expression according to the characterized by reduced levels of Notch proteins. Patients treatment status may reﬂect a possible role for Notch signaling achieving complete remission following chemotherapy display in the response to chemotherapy. To clarify this issue, we no or a few blast cells in bone marrow (1). Consequently, treated B-ALL cell lines for 72 hours with chemotherapeutic þ B-ALL blast cells (CD19 ) cannot be used to follow the general agents, including Cytarabine (Ara-C) and dexamethasone. The Notch downregulation in the bone marrow following the addition of increasing concentrations of each of these drugs in

642 Cancer Res; 79(3) February 1, 2019 Cancer Research

Notch, B-ALL Chemosensitivity

Figure 2. Notch expression as prognostic marker. Samples collected from patients at diagnosis were later classiﬁed according to treatment outcome, that is, patients responsive (11) to treatment and refractory patients (11). Then Notch expression levels were analyzed in each B-ALL sample through ﬂow cytometric analysis. , P < 0.05.

B-ALL cell culture medium induced cell death in a dose–response chemotherapeutic agents (Fig. 3C), suggesting that Notch4 could manner as recently described (15). At 24 hours of treatment, be irrelevant for drug response or, by contrast, it could be the main although MTT and Annexin V assays revealed no apoptosis in Notch receptor involved in chemoresistance. To assess the con- treated cells, immunoblots revealed a decrease of Notch1, Notch2, tribution of each pathway component to chemosensitivity, we and Notch3 protein levels in samples treated with Ara-C and decided to interfere with Notch molecules expression in B-ALL dexamethasone. Notch4 levels did not change regardless the cells upon drug treatment.

Figure 3. Treatment-induced changes in the expression of Notch receptors in B-ALL cells. A, Flow cytometric analysis of Notch expression in cell population from patient samples collected before and after the treatment. Data are represented as mean fluorescence intensity of each antibody normalized to specific fluorochrome- conjugated control. B, Percentage of hypomethylated probes in samples from patients collected at diagnosis, after the treatment, or from relapsed patients. Relative hypomethylation was defined as probes with low b-value as compared to normal B cells. C, Immunoblot analysis for Notch1-4 in B-ALL cell lines treated for 24 hours with Ara-C (5 mmol/L) and dexamethasone (0.5 mmol/L). Data are representative of four independent experiments. , P < 0.05.

www.aacrjournals.org Cancer Res; 79(3) February 1, 2019 643

Takam Kamga et al.

Figure 4. Notch inhibition reduces B-ALL cell proliferation and survival. A, Relative viability of primary B-ALL cell samples cultured for 4 days in presence of Notch inhibitors, that is, SAHM1 (20 mmol/L), GSI IX (15 mmol/L), and GSI-XII (10 mmol/L). Cells were collected, stained with Topro-3, and analyzed by flow cytometry; viable cells were identified as Topro-3-negative cells. Data are expressed as the mean SEM of four independent experiments involving 12 patient samples. B, Relative proliferation of B-ALL cell lines stained with CFSE and treated for 4 days with Notch inhibitors; CFSE dilution was analyzed by flow cytometry and is expressed as relative proliferation. Data are reported as mean SEM of four independent experiments performed in duplicate. C, Flow cytometric analysis of human CD19þ cells in bone marrow samples obtained from mice transplanted with the B-ALL cell line RS4;11. Starting from day 14 postengraftment, mice were treated for 3 days with either GSI-IX (10 mg/kg) or GSI-XII (10 mg/kg) or their vehicle (DMSO). The assay was performed with at least five mice in each group. , P < 0.05; , P < 0.01; , P < 0.001.

Notch inhibition affects B-ALL cell survival and proliferation daily for 3 days. Only a slight, and not significant reduction in þ We tested three categories of Notch inhibitors for their effects levels of human CD19 cells was observed in the bone marrow of on B-ALL cell survival, proliferation, and drug response, that is mice treated with GSI-IX and GSI-XII (Fig. 4C). Notch blocking antibodies, GSI-IX/GSI-XII, and a Notch transcription factor inhibitor (SAHM1). We previously demonstrated, Notch Inhibition potentiates B-ALL cell chemosensitivity using Western blot analysis of active form of Notch receptors as Considering the lack of significant reduction of leukemic cell well as RBP-Jk reporter genes, that all these inhibitors were capable burden in mouse bone marrow, we investigated the effect of the of inhibiting Notch signaling in leukemic cells (5, 13). In this association of Notch inhibitors with chemotherapeutic agents study, the concentration used for each inhibitor was sufficient to on B-ALL cell in vitro survival and in vivo clearance. GSI-IX or reduce the levels of Hes1 in B-ALL cells (Supplementary Fig. S1C). GSI-XII were all capable of sensitizing B-ALL cells to Ara-C, We observed that GSI-IX and GSI-XII significantly reduced B-ALL dexamethasone, and doxorubicin treatments (Fig. 5A), whereas primary cell viability in vitro (Fig. 4A). Although a reduction in cell SAHM1 was not (Supplementary Fig.S3).Theseeffectswere viability following SAHM1 addition was observed, the effect was mostly mediated by Notch4, as only its blockade reproduced all not statically significant (Fig. 4A). We have previously shown that the effects observed with GSI-IX and GSI-XII, thus promoting 5 mg/mL of anti-Notch3 or anti-Notch4 antibodies are sufficient to B-ALL cell apoptosis induced by all the drugs (Fig. 5A; Sup- inhibit Notch signaling in B-ALL cell samples (5, 14). Accordingly, plementary Fig. S2). We next investigated whether the synergy anti-Notch3 (5 mg/mL) or anti-Notch4 (5 mg/mL) blocking anti- between Notch inhibition and chemotherapy could also occur bodies decreased B-ALL leukemia cell survival in vitro (Supple- in the mouse xenograft model of B-ALL. To this aim, we mentary Table S5). We also observed a reduction in cell prolif- assessed in vitro the most effective combination schedule of eration of B-ALL cell lines cultured in media supplemented with drugs with Notch inhibitors. B-ALL cell lines were treated with the Notch inhibitors (Fig. 4B). We then generated a xenograft Ara-C, dexamethasone, or doxorubicin associated with one of model of B-ALL by injecting the B-ALL cell line RS4;11 into the tail the Notch inhibitors. Results clearly highlighted that the stron- vein of NOG mice; after 4 weeks, mice showed 69% to 80% of gest antileukemic effect was obtained in vitro with Ara-C plus þ human CD19 cells in the bone marrow (Fig. 4C). At week 2 GSI-XII combination (Supplementary Fig. S3). Therefore, 2 following injection, mice received intraperitoneal injection of weeks after injection of RS4;11 in NOG mice, animals were GSI-IX (10 mg/kg) or GSI-XII (10 mg/kg) or their vehicle (DMSO) treated for 3 days with Ara-C alone followed by 3 days of

644 Cancer Res; 79(3) February 1, 2019 Cancer Research

Notch, B-ALL Chemosensitivity

Figure 5. Notch inhibition enhances drug effects in vitro and in vivo. A, Primary B-ALL cells were cultured for 2 days in presence of Ara-C (5 mmol/L) or doxorubicin alone or in combination with Notch inhibitors, that is, GSI IX (15 mmol/L), GSI-XII (10 mmol/L), and anti-Notch4–blocking antibodies (5 mg/mL). Cells were then collected, stained with Annexin V/PI, and analyzed by flow cytometry. Data are expressed as the mean SEM of three to five independent experiments involving eight patient samples. B, Representative and quantitative flow cytometric analysis of human CD19þ cells in bone marrow samples obtained from mice transplanted with the B-ALL cell line RS4;11. Starting from day 14 postengraftment, mice were treated for 3 days with Ara-C, followed by 3 days of Notch inhibitors (GSI-XII 10 mg/kg, anti-Notch4 0.5 mg/kg) or their respective controls (DMSO and the control isotype IgG1 0.5 mg/kg). Data are reported as mean SEM of values obtained from at least six mice marrow. C, Mice survival as analyzed with the log-rank (Mantel–Cox) test. The assay was performed with at least five mice in each group. , P < 0.05; , P < 0.01.

treatment with Notch inhibitors (GSI-XII, anti-Notch4) or their Notch signaling modulates drug resistance by controlling respective controls (DMSO, IgG isotype). Analysis of mouse production of ROS species bone marrow at 4 weeks showed that Ara-C signiﬁcantly Many anticancer drugs trigger cell death through the induction þ reduced the levels of human CD19 cells, particularly of ROS species (30). We observed that pretreatment of B-ALL cell when Ara-C treatment was followed by GSI-XII or anti-Notch4 lines with anti-oxidants, such as NAC or b-mercaptoetanol (BME), treatment (Fig. 5B). Reduction of leukemic burden in the bone prevented cell death induced by Ara-C or doxorubicin (Supple- marrow correlated with increased mouse survival, with a medi- mentary Fig. S4A). It has been demonstrated that Notch inhibi- an survival of 42, 51, and 55 days for vehicle (DMSO), Ara-C, tion can potentiate cell death by increasing the levels of intracel- and Ara-CþGSI-XII, respectively (Fig. 5C). lular ROS (31). Accordingly, we found that NAC administration

www.aacrjournals.org Cancer Res; 79(3) February 1, 2019 645

Takam Kamga et al.

Figure 6. Notch inhibition controls redox system. A, Survival of mice transplanted with RS4;11 cell line and subdivided in ﬁve treatment groups. Treatments were administrated as described in Fig. 5B; NAC treatment started at day 14 and lasted for 6 days. Differences in survival curves were analyzed with the log-rank (Mantel– Cox) test. B, ROS levels in B-ALL cell lines treated with Notch inhibitors (SAHM1 20 mmol/L, GSI IX 15 mmol/L, or GSI-XII 10 mmol/L) in association with Ara-C (5 mmol/L). Cells were collected at each time point and stained with 5 mmol/L of cm-H2DCFDA for 30 minutes. Fluorescence of oxidized cm- H2DCFDA was measured by ﬂow cytometry. Data are representative of three independent experiments performed in duplicate. The control group used for statistics was Ara-CþDMSO. , P < 0.05; , P < 0.01; , P < 0.001.

to B-ALL xenotransplanted mice suppressed the survival advan- with Ara-C or doxorubicin only (Fig. 6B). To determine whether tage conferred by the co-administration of GSI-XII with Ara-C pro-oxidant effects of Notch inhibitors were responsible for the (Fig. 6A). NAC administration had no effect on survival of sensitization of B-ALL cells to Ara-C and doxorubicin, we incu- nontransplanted mouse, excluding any toxic effect of NAC alone bated cells with NAC or BME two hours before combined treat- (Supplementary Fig. S4B). To assess whether Notch inhibition ment with Ara-C/doxorubicin and Notch inhibitors. Cell death favored Ara-C- and doxorubicin-induced cell death by enhancing was analyzed at 48 hours using Annexin V/PI staining. Our results intracellular ROS levels, we used oxidation of c-H2DCFDA as showed that NAC or BME treatment abrogated cell death induced surrogate for ROS production. We observed that cells treated with by Ara-C or doxorubicin, used either as single agents or in Ara-C or doxorubicin in association with Notch inhibitors combination with Notch inhibitors (Fig. 7A; Supplementary showed increased levels of ROS, as compared with cells treated Fig. S4C). To conﬁrm this critical role for ROS in Notch

Figure 7. Notch inhibition-induced cell death is sensitive to the oxidative state. A, Primary B-ALL cell samples were cultured for 2 days in presence of Ara-C (5 mmol/L) alone or with Notch inhibitors, that is, SAHM1 (20 mmol/L), GSI IX (15 mmol/L), GSI-XII (10 mmol/L), and with or without NAC (2 mmol/L). Cells were collected, stained with Annexin V/PI, and analyzed by ﬂow cytometry. Data are expressed as mean SEM of four independent experiments using eight different patient samples. B, Representative Immunoblot analysis of prosurvival proteins in B-ALL primary cells treated with Ara-C (5 mmol/L) alone or with Notch inhibitors, that is, GSI IX (15 mmol/L), GSI-XII (10 mmol/L), anti-Notch4 (5 mg/mL), and with or without NAC (2 mmol/L). Experiments were performed with samples from at least four patients. , P < 0.05; , P < 0.01; , P < 0.001.

646 Cancer Res; 79(3) February 1, 2019 Cancer Research

Notch, B-ALL Chemosensitivity

inhibition-mediated cell death, we assessed through Western specific receptor blockade in terms of cell survival, proliferation, immunoblotting the levels of proteins capable to support the and drug response. Even if anti-Notch4 and anti-Notch3 prosurvival role of Notch towards cancer cells, that is mTOR, antibodies were all capable of lowering B-ALL cell viability, AKT, NF-kB, and ERK1/2 (5, 13, 32). We observed that Ara-C only anti-Notch4 reproduced the inhibitory effects of GSI-IX and treatment associated with Notch inhibition through either GSIs or GSI-XII by sensitizing B-ALL cells to apoptosis induced by Ara-C, anti-Notch4 reduced the levels of, mTOR, pNF-kB/NF-kB, and dexamethasone or doxorubicin. Consequently, Notch4 seemed to pERK1/2/ERK. Interestingly, NAC pretreatment prevented the be the main driving mechanism of Notch-mediated drug resis- reduction of expression levels of prosurvival proteins in B-ALL tance in B-ALL cells. In addition, the mild effect of the Notch cells incubated with Ara-C associated with Notch inhibitors. transcription factor SAHM1 suggests the involvement of nonca- However, considering mTOR and pNF-kB/NF-kB proteins, nonical Notch signaling in B-ALL cell response to chemotherapy. NAC-induced rescue was different according to the type of The latter hypothesis is supported by the fact that the presence of Notch inhibitor used (GSIs or anti-Notch4), whereas a common active form of Notch4 in B-ALL cells was not always associated to pattern of modulation between GSIs and anti-Notch4 was the expression of the Notch target gene HES1. Accordingly, we found for pERK1/2 and ERK1/2, thus suggesting that ERK1/2 have recently characterized a new B-ALL cell line (VR-ALL) dis- may act as the main downstream target involved in ROS- playing a non-CSL dependent Notch signaling; VR-ALL cell line is dependent Notch-mediated survival and chemosensitivity in sensitive to GSIs and anti-Notch4, but poorly to SAHM1 (15). The B-ALL cells (Fig. 7B). challenge of frontline therapy in B-ALL is to eradicate leukemia cells in bone marrow after induction. Patients with no evidence of minimal residual disease following induction show a better Discussion outcome as compared to patients with persistent leukemic clones, The pathogenetic role of Notch signaling in leukemic diseases, requiring new therapeutic strategies (40). The association such as T-ALL and CLL, has been ascribed to the presence of between clinical status and Notch expression suggested the activating mutations in genes coding for components of the Notch involvement of the Notch pathway in chemoresistance. But the pathway (33). However, Notch pathway overexpression due to validation on a larger B-ALL cohort of patients would further paracrine signals between leukemic cells and stromal cell micro- strengthen the prognostic value of Notch signaling monitoring. environment may result in enhanced leukemia cell survival, as Importantly, through in vitro approaches, we identified Ara- shown in vitro both in B-ALL (14) and AML cells (5, 22, 23) that CþGSI-XII as the most active combination of chemotherapeutic normally lack activating Notch mutations (34). The reason for this agents and Notch inhibitors. The effectiveness of this combina- abnormal expression of the Notch pathway is unclear, but Notch tion was confirmed in the xenograft model of B-ALL. We observed receptor and ligand overexpression can be a consequence of side that GSI-XII potentiates antileukemic effects of Ara-C by reducing mutations or cytogenetic abnormalities affecting other signaling the bone marrow leukemic burden and prolonging the overall systems, as it occurs in AML for b-catenin (35), Flt3 (36), and survival of mice undergoing xenotransplantion with the B-ALL PML-RAR (37). In AML, even in absence of Notch gene mutations, cell line RS4;11. Preclinical and clinical use of GSIs are hampered a significant expression of Notch pathway correlated with shorter by their high toxicity, likely due to off-target effects and the patient survival (6). Accordingly, gene sequencing of six patients inhibition of all Notch receptors, including those not involved with B-ALL revealed the lack of any activating mutation in Notch in the disease (41). In this study, we showed the central role of components; according to previous studies, we found some Notch4 in B-ALL chemosensitivity and we demonstrated that mutations in genes associated to the Notch pathway regulation, association of Ara-C to anti-Notch4 blocking could lead to an including NCOR1, NUMB, and TCF3 (38, 39). antileukemic effect equivalent to the one observed with GSIs In this study, we observed that B-ALL cells expressed high levels treatment. These findings may pave the way to the association of different components of the Notch pathway. Notch1, Notch3, of Notch inhibitors to chemotherapy in the induction regimen or Notch4, Jagged2, DLL-3, and DLL-4 resulted upregulated both in during consolidation therapy with the aim of eradicating minimal B-ALL primary cells and B-ALL cell lines and the pathway was residual disease. In fact, Notch inhibitors act either by potentiat- functional, as shown by the expression of Notch target genes ing the effect of drugs on specific pathways or by interfering with HES1, HES5, HEY1, DELTEX1, and c-MYC in B-ALL cells. The pathways that sustain resistance to anticancer drugs (8, 42). The functionality of the pathway was also revealed by the sensitivity of three chemotherapeutic agents used in our study have as common target genes to Notch inhibitors, such as SAHM1, GSI-IX, and feature, the capability of inducing the production of ROS in cancer GSI-XII (Supplementary Fig. S1C). cells, leading to apoptosis (43–45). Emerging data revealed that However, an important finding is that Notch pathway pro- Notch signaling protects cancer cells from apoptosis by main- motes B-ALL cell chemosensitivity in vitro and in vivo, potentiating taining low levels of ROS (46). In our study, we demonstrated that B-ALL cell survival, proliferation, and modulating drug resistance Notch inhibitors potentiate drug-induced apoptosis by increasing through the control of ROS species. We observed when analyzing ROS production and decreasing mTOR, NF-kB, and ERK levels. samples from patients, that expression levels of all Notch recep- Abrogation of these effects by NAC clearly showed that Notch tors decreased upon chemotherapy. Consistently, changes in inhibitors improve antileukemic activity of drugs through the expression levels of Notch1-3 in B-ALL cells were associated in induction of ROS. vitro to drug treatments. However, all the drugs tested in vitro were Taken together, our data provide preclinical data in favor of the unable to modulate Notch4 levels in B-ALL cell lines at 24 hours, potential usefulness of the association of Notch inhibitors with thus suggesting a critical role for Notch4 in B-ALL cell survival. To conventional chemotherapy to reduce the relapse rate. assess the specific contribution of Notch4 and other Notch receptors to B-ALL cell chemosensitivity, we studied the effects Disclosure of Potential Conflicts of Interest of either pan-Notch inhibitors, such as GSI-IX and GSI-XII, or No potential conflicts of interest were disclosed.

www.aacrjournals.org Cancer Res; 79(3) February 1, 2019 647

Takam Kamga et al.

Authors' Contributions Acknowledgments Conception and design: P.T. Kamga, A. Mercuri, S. Cesaro, M. Krampera This work was supported by Fondazione CARIVERONA. The study was Development of methodology: P.T. Kamga, G.D. Collo, M. Midolo, A. Adamo, conducted in Interdepartmental Laboratory of Medical Research (LURM) at M. Krampera the University of Verona, Verona, Italy. We would like to thank Dr. Jessicca Acquisition of data (provided animals, acquired and managed patients, Nordlund from the Uppsala University for her kind support and help for in silico provided facilities, etc.): P.T. Kamga, G.D. Collo, M. Midolo, S. Cesaro, analysis of gene methylation. We would like to thank Dr. Massimo Delledonne E. Mimiola, M. Bonifacio, A. Andreini for his kind support and help for genomic studies. Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): P.T. Kamga, A. Adamo, P. Delﬁno The costs of publication of this article were defrayed in part by the payment of Writing, review, and/or revision of the manuscript: P.T. Kamga, G.D. Collo, page charges. This article must therefore be hereby marked advertisement in S. Cesaro, M. Bonifacio, A. Andreini, M. Chilosi, M. Krampera accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): P.T. Kamga, M. Bonifacio, A. Andreini Received May 27, 2018; revised October 23, 2018; accepted December 14, Study supervision: P.T. Kamga, M. Krampera 2018; published ﬁrst December 18, 2018.

References 1. Cooper SL, Brown PA. Treatment of pediatric acute lymphoblastic leuke- 18. Aster JC. Deregulated NOTCH signaling in acute T-cell lymphoblastic mia. Pediatr Clin North Am 2015;62:61–73. leukemia/lymphoma: new insights, questions, and opportunities. Int J 2. Kamdje AHN, Kamga PT, Simo RT, Vecchio L, Etet PFS, Muller JM, et al. Hematol 2005;82:295–301. Developmental pathways associated with cancer metastasis: Notch, Wnt, 19. Gianfelici V. Activation of the NOTCH1 pathway in chronic lymphocytic and Hedgehog. Cancer Biol Med 2017;14:109–20. leukemia. Haematologica 2012;97:328–30. 3. Zou J, Hong Y, Tong Y, Wei J, Qin Y, Shao S, et al. Sonic hedgehog produced 20. Mullighan CG, Zhang J, Kasper LH, Lerach S, Payne-Turner D, Phillips LA, by bone marrow-derived mesenchymal stromal cells supports cell survival et al. CREBBP mutations in relapsed acute lymphoblastic leukaemia. in myelodysplastic syndrome. Stem Cells Int 2015;2015:1–13. Nature 2011;471:235–9. 4. Rosati E, Sabatini R, Rampino G, Tabilio A, Di Ianni M, Fettucciari K, et al. 21. Taylor KH, Pena-Hernandez KE, Davis JW, Arthur GL, Duff DJ, Shi H, et al. Constitutively activated Notch signaling is involved in survival and apo- Large-scale CpG methylation analysis identifies novel candidate genes and ptosis resistance of B-CLL cells. Blood 2009;113:856–65. reveals methylation hotspots in acute lymphoblastic leukemia. Cancer Res 5. Takam Kamga P, Bassi G, Cassaro A, Midolo M, Di Trapani M, Gatti A, et al. 2007;67:2617–25. Notch signalling drives bone marrow stromal cell-mediated chemoresis- 22. Lobry C, Ntziachristos P, Ndiaye-Lobry D, Oh P, Cimmino L, Zhu N, et al. tance in acute myeloid leukemia. Oncotarget 2016;7:21713–27. Notch pathway activation targets AML-initiating cell homeostasis and 6. Xu X, Zhao Y, Xu M, Dai Q, Meng W, Yang J, et al. Activation of Notch signal differentiation. J Exp Med 2013;210:301–19. pathway is associated with a poorer prognosis in acute myeloid leukemia. 23. Kannan S, Sutphin RM, Hall MG, Golfman LS, Fang W, Nolo RM, et al. Med Oncol Northwood Lond Engl 2011;28:S483–489. Notch activation inhibits AML growth and survival: a potential therapeutic 7. Gordon WR, Arnett KL, Blacklow SC. The molecular logic of Notch approach. J Exp Med 2013;210:321–37. signaling - a structural and biochemical perspective. J Cell Sci 2008;121: 24. Xu X, Zhao Y, Xu M, Dai Q, Meng W, Yang J, et al. Activation of Notch signal 3109–19. pathway is associated with a poorer prognosis in acute myeloid leukemia. 8. Ntziachristos P, Lim JS, Sage J, Aifantis I. From fly wings to targeted Med Oncol Northwood Lond Engl 2011;28:S483–489. cancer therapies: a centennial for notch signaling. Cancer Cell 2014;25: 25. Rossi D, Rasi S, Fabbri G, Spina V, Fangazio M, Forconi F, et al. Mutations of 318–34. NOTCH1 are an independent predictor of survival in chronic lymphocytic 9. Cai W-X, Liang L, Wang L, Han J-T, Zhu X-X, Han H, et al. Inhibition of leukemia. Blood 2012;119:521–9. Notch signaling leads to increased intracellular ROS by up-regulating 26. Ferrando A. NOTCH mutations as prognostic markers in T-ALL. Leukemia Nox4 expression in primary HUVECs. Cell Immunol 2014;287: 2010;24:2003–4. 129–35. 27. Schultz KR, Pullen DJ, Sather HN, Shuster JJ, Devidas M, Borowitz MJ, et al. 10. Kuang S-Q, Fang Z, Zweidler-McKay PA, Yang H, Wei Y, Gonzalez-Cer- Risk- and response-based classification of childhood B-precursor acute vantes EA, et al. Epigenetic inactivation of Notch-Hes pathway in human lymphoblastic leukemia: a combined analysis of prognostic markers from B-cell acute lymphoblastic leukemia. PloS One 2013;8:e61807. the Pediatric Oncology Group (POG) and Children's Cancer Group 11. Zweidler-McKay PA, He Y, Xu L, Rodriguez CG, Karnell FG, Carpenter AC, (CCG). Blood 2007;109:926–35. et al. Notch signaling is a potent inducer of growth arrest and apoptosis in a 28. Campana D. Role of minimal residual disease monitoring in adult and wide range of B-cell malignancies. Blood 2005;106:3898–906. pediatric acute lymphoblastic leukemia. Hematol Oncol Clin North Am 12. Yoon S-O, Zhang X, Berner P, Blom B, Choi YS. Notch ligands expressed by 2009;23:1083–98. follicular dendritic cells protect germinal center B cells from apoptosis. 29. Campana D, Pui C-H. Minimal residual disease–guided therapy in child- J Immunol 2009;183:352–8. hood acute lymphoblastic leukemia. Blood 2017;129:1913–8. 13. Kamdje AHN, Bassi G, Pacelli L, Malpeli G, Amati E, Nichele I, et al. Role of 30. Trachootham D, Alexandre J, Huang P. Targeting cancer cells by ROS- stromal cell-mediated Notch signaling in CLL resistance to chemotherapy. mediated mechanisms: a radical therapeutic approach? Nat Rev Drug Blood Cancer J 2012;2:e73. Discov 2009;8:579–91. 14. Nwabo Kamdje AH, Mosna F, Bifari F, Lisi V, Bassi G, Malpeli G, et al. 31. Yu H-C, Bai L, Yue S-Q, Wang D-S, Wang L, Han H, et al. Notch signal Notch-3 and Notch-4 signaling rescue from apoptosis human B-ALL cells protects non-parenchymal cells from ischemia/reperfusion injury in vitro in contact with human bone marrow-derived mesenchymal stromal cells. by repressing ROS. Ann Hepatol 2013;12:815–21. Blood 2011;118:380–9. 32. Okuhashi Y, Itoh M, Nara N, Tohda S. NOTCH knockdown affects the 15. Kamga PT, Collo GD, Bassi G, Midolo M, Delledonne M, Chilosi M, et al. proliferation and mTOR signaling of leukemia cells. Anticancer Res Characterization of a new B-ALL cell line with constitutional defect of the 2013;33:4293–8. Notch signaling pathway. Oncotarget 2018;9:18341–50. 33. Liu N, Zhang J, Ji C. The emerging roles of Notch signaling in leukemia and 16. Ikeda T, Nakata Y, Kimura F, Sato K, Anderson K, Motoyoshi K, et al. stem cells. Biomark Res 2013;1:23. doi: 10.1186/2050-7771-1-23. Induction of redox imbalance and apoptosis in multiple myeloma cells by 34. Palomero T, McKenna K, O-Neil J, Galinsky I, Stone R, Suzukawa K, et al. the novel triterpenoid 2-cyano-3,12-dioxoolean-1,9-dien-28-oic acid. Mol Activating mutations in NOTCH1 in acute myeloid leukemia and lineage Cancer Ther 2004;3:39–45. switch leukemias. Leukemia 2006;20:1963–6. 17. Nordlund J, B€acklin CL, Wahlberg P, Busche S, Berglund EC, Eloranta M-L, 35. Kode A, Manavalan JS, Mosialou I, Bhagat G, Rathinam CV, Luo N, et al. et al. Genome-wide signatures of differential DNA methylation in pediatric Leukaemogenesis induced by an activating b-catenin mutation in osteo- acute lymphoblastic leukemia. Genome Biol 2013;14:r105. blasts. Nature 2014;506:240–4.

648 Cancer Res; 79(3) February 1, 2019 Cancer Research

Notch, B-ALL Chemosensitivity

36. Stolzel€ F, Steudel C, Oelschl€agel U, Mohr B, Koch S, Ehninger G, et al. 42. Previs RA, Coleman RL, Harris AL, Sood AK. Molecular pathways: trans- Mechanisms of resistance against PKC412 in resistant FLT3-ITD positive lational and therapeutic implications of the Notch signaling pathway in human acute myeloid leukemia cells. Ann Hematol 2010;89:653–62. cancer. Clin Cancer Res 2015;21:955–61. 37. Grieselhuber NR, Klco JM, Verdoni AM, Lamprecht T, Sarkaria SM, 43. Iacobini M, Menichelli A, Palumbo G, Multari G, Werner B, Del Principe D. Wartman LD, et al. Notch signaling in acute promyelocytic leukemia. Involvement of oxygen radicals in cytarabine-induced apoptosis in human Leukemia 2013;27:1548–57. polymorphonuclear cells. Biochem Pharmacol 2001;61:1033–40. 38. Liu Y-F, Wang B-Y, Zhang W-N, Huang J-Y, Li B-S, Zhang M, et al. Genomic 44. Kim S-Y, Kim S-J, Kim B-J, Rah S-Y, Chung SM, Im M-J, et al. Doxorubicin- profiling of adult and pediatric B-cell Acute Lymphoblastic Leukemia. induced reactive oxygen species generation and intracellular Ca2þ increase EBioMedicine 2016;8:173–83. are reciprocally modulated in rat cardiomyocytes. Exp Mol Med 39. Okamoto R, Ogawa S, Nowak D, Kawamata N, Akagi T, Kato M, et al. 2006;38:535–45. Genomic profiling of adult acute lymphoblastic leukemia by single nucle- 45. Kraaij MD, van der Kooij SW, Reinders MEJ, Koekkoek K, Rabelink TJ, van otide polymorphism oligonucleotide microarray and comparison to pedi- Kooten C, et al. Dexamethasone increases ROS production and T cell atric acute lymphoblastic leukemia. Haematologica 2010;95:1481–8. suppressive capacity by anti-inflammatory macrophages. Mol Immunol 40. Pui C-H, Evans WE. Treatment of acute lymphoblastic leukemia. N Engl J 2011;49:549–57. Med 2006;354:166–78. 46. Qiang L, Wu T, Zhang H-W, Lu N, Hu R, Wang Y-J, et al. HIF-1a is critical for 41. Golde TE, Koo EH, Felsenstein KM, Osborne BA, Miele L. g-Secretase hypoxia-mediated maintenance of glioblastoma stem cells by activating inhibitors and modulators. Biochim Biophys Acta 2013;1828:2898–907. Notch signaling pathway. Cell Death Differ 2012;19:284–94.

www.aacrjournals.org Cancer Res; 79(3) February 1, 2019 649

Inhibition of Notch Signaling Enhances Chemosensitivity in B-cell Precursor Acute Lymphoblastic Leukemia

Paul Takam Kamga, Giada Dal Collo, Martina Midolo, et al.

Cancer Res 2019;79:639-649. Published OnlineFirst December 18, 2018.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-18-1617

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2018/12/18/0008-5472.CAN-18-1617.DC1

Cited articles This article cites 46 articles, 16 of which you can access for free at: http://cancerres.aacrjournals.org/content/79/3/639.full#ref-list-1

Citing articles This article has been cited by 3 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/79/3/639.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/79/3/639. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.