Human B Cell Precursor Acute Lymphoblastic Leukemia Research Articles

A brain organoid/ALL coculture model reveals the AP-1 pathway as critically associated with CNS involvement of BCP-ALL

AbstractCentral nervous system (CNS) involvement remains a clinical hurdle in treating childhood B-cell precursor acute lymphoblastic leukemia (BCP-ALL). The disease mechanisms of CNS leukemia are primarily investigated using 2-dimensional cell culture and mouse models. Given the variations in cellular identity and architecture between the human and murine CNS, it becomes imperative to seek complementary models to study CNS leukemia. Here, we present a first-of-its-kind 3-dimensional coculture model combining human brain organoids and BCP-ALL cells. We noticed significantly higher engraftment of BCP-ALL cell lines and patient-derived xenograft (PDX) cells in cerebral organoids than non-ALL cells. To validate translatability between organoid coculture and in vivo murine models, we confirmed that targeting CNS leukemia–relevant pathways such as CD79a/Igα or C-X-C motif chemokine receptor 4–stromal cell-derived factor 1 reduced the invasion of BCP-ALL cells into organoids. RNA sequencing and functional validations of organoid-invading leukemia cells compared with the noninvaded fraction revealed significant upregulation of activator protein 1 (AP-1) transcription factor–complex members in organoid-invading cells. Moreover, we detected a significant enrichment of AP-1 pathway genes in PDX ALL cells recovered from the CNS compared with spleen blasts of mice that had received transplantation with TCF3::PBX1+ PDX cells, substantiating the role of AP-1 signaling in CNS disease. Accordingly, we found significantly higher levels of the AP-1 gene, jun proto-oncogene, in patients initially diagnosed as CNS-positive BCP-ALL compared with CNS-negative cases as well as CNS-relapse vs non–CNS-relapse cases in a cohort of 100 patients with BCP-ALL. Our results suggest CNS organoids as a novel model to investigate CNS involvement and identify the AP-1 pathway as a critical driver of CNS disease in BCP-ALL.

Acquired Venetoclax Resistance in an In Vivo Model of B-Cell Precursor Acute Lymphoblastic Leukemia Is Characterized By Altered Functions of Apoptosis Regulators

Deregulation of cell death pathways such as apoptosis is a hallmark of cancer and contributes to leukemogenesis and treatment failure in B-cell precursor acute lymphoblastic leukemia (BCP-ALL). Apoptosis is controlled by different pro- and anti-apoptotic molecules either inducing or counteracting cell death signaling. Inhibition of anti-apoptotic molecules of BCL-2 family cell death inhibitors has been developed as therapeutic strategy. Venetoclax (VEN) binds selectively to BCL-2 resulting in apoptosis induction and has shown clinical activity in hematologic malignancies, but primary and secondary resistance have also been described. In this study, we aimed to identify resistance mechanisms acquired during VEN treatment leading to therapy failure of BCL-2 inhibition. In order to mimic a clinical situation with consecutive treatment cycles, we employed a patient-derived human BCP-ALL xenograft mouse model with subsequent courses of VEN administration to ALL bearing recipients in vivo. First, we estimated VEN-sensitivities in a series of BCP-ALL PDX samples (N=30) by ex vivo titration of individual VEN-EC 50 values and identified a range of heterogeneous VEN-activities. A VEN-sensitive (low VEN-EC50) PDX-ALL sample with a pro-B ALL phenotype carrying a KMT2A:: MLL1 gene fusion was selected and transplanted onto immunodeficient NOD/SCID recipient mice. Upon leukemia manifestation (presence of 5% human ALL cells in peripheral blood), mice were either treated with VEN (100 mg/kg, two times five days) or vehicle control. Recipients were sacrificed upon onset of leukemia-related morbidity and ALL cells isolated from spleens were transplanted onto secondary recipients, which were again treated with VEN or vehicle. ALL cells isolated from secondary recipients were retransferred onto tertiary mice and again exposed to in vivo VEN-treatment providing a total of three subsequent transplantation and in vivo VEN-therapy cycles. Assessing VEN sensitivities of the ALL cells initially transplanted and isolated after VEN-treatment cycles by using an ex vivo cytochrome c release assay showed increasing VEN-resistance. Accordingly, comparison of VEN to control in vivo treated cells showed significantly decreased sensitivities over treatment cycles. Importantly, also the time periods between treatment and leukemia re-occurrence in mice indicating decreasing VEN-sensitivity became shorter with every additional in vivo therapy course. Paired RNA- seq analysis revealed that transcript levels of the pro-apoptotic molecules BAX and BIM were significantly downregulated in VEN-exposed/resistant ALL as compared to vehicle control treated cells. Basal BH3 profiling analysis showed impaired apoptotic priming (significantly lower cytochrome c release after BIM and PUMA peptide exposure) and decreased dependence on anti-apoptotic BCL-2 and MCL-1 indicating a shift from a pro-apoptotic to a pro-survival state. This shift was also reflected in an e x vivo drug treatment assay showing decreased sensitivity to the MCL-1 inhibitor S63845 and the BCL-XL inhibitor A-1331852 in ALL cells from VEN- treated mice compared to control- treated mice (S63845 EC 50 1.5 vs. 2.2 µM, A-1331852 EC 50 9.3 vs. 15.3 µM). In line with the low BAX levels in ALL cells with acquired VEN-resistance, PDX BCP-ALL samples with intrinsic VEN resistance also showed lower BAX protein expression. Taken together, acquired VEN-resistance was recapitulated in a co-clinical trial model of BCP-ALL with repeated in vivo treatment cycles showing lower drug sensitivities along with increasingly reduced in vivo anti-ALL activity of VEN. Characterization of acquired VEN-resistance revealed decreased functional dependency on anti-apoptotic proteins and downregulation of pro-apoptotic BIM and BAX, thus pointing to an imbalance of pro- and anti-apoptotic molecules, which can be potentially targeted by directed compounds bypassing resistance to specific BCL-2 inhibition.

Pax5 Heterozygosity Affects B-Cell Differentiation and Generates a Deregulated Precursor Population in the Bone Marrow

Introduction: PAX5 is a well-known master regulator for B-cell development, commitment and identity. Hence, germline or somatic deregulation of PAX5 facilitates the development of B-cell precursor acute lymphoblastic leukemia (BCP-ALL). Previously, our group has described B-cell development abnormalities in three families carrying PAX5 germline variants ( Escudero et al., Leukemia, 2022) as well as established a link between Pax5 heterozygosity and infection exposure, thereby recapitulating the incomplete penetrance of the human BCP-ALL ( Martin-Lorenzo et al., Cancer Discovery, 2015). Nevertheless, the molecular single-cell characteristics of the susceptible precursor population mediated by reduced Pax5 levels is still poorly understood. Methods: Here, we deeply characterized the susceptible precursor population in Pax5+/- mice utilizing single-cell RNA Sequencing (scRNA-Seq), multicolor flow cytometry analyses and in-vivo transplantation models. Results: Our data show, that Pax5 heterozygosity leads to an aberrant pre-BII population (B220 +CD19 +IgM -CD25 +) in the bone marrow (BM) of Pax5+/- mice (p=0.0026, Student's t-test) (Fig. 1A). This population is stable over time and characterized by higher CD25 and IL-7-Receptor expression (p<0.0001, Student's t-test). Furthermore, bulk RNA-Sequencing of the pre-BII population revealed B-cell receptor (BCR) signaling as one of the top downregulated pathways (including CD79a/b, Lyn and CD72), while DNA replication and cell cycle signaling pathways were upregulated in Pax5+/- vs. Wildtype (WT) pre-BII cells. ScRNA-Seq analyses confirmed these results (Fig. 1B) and further showed that pre-BII cells of Pax5+/- mice are skewed toward the less commonly used lambda light chain BCR-rearrangements. The observed delay in the transition to IgM-positivity was additionally validated in a murine in-vivo transplantation model. Here, after 72 hours pre-BII cells from Pax5+/- mice preferably homed to the BM, while matured transplanted WT cells were predominantly found in the spleen (p=0.0023 and p=0.0140, respectively, Student's t-test). To depict the full spectrum of malignant transformation based on the described deregulated precursor B-cell population, we additionally performed scRNA-Seq on different stages of Pax5-mediated BCP-ALL evolution. Our data show that in a first step, arrested pre-leukemic cells of Pax5+/- mice lose their B-cell identity and display high Myc levels. Sequentially, full blown BCP-ALL arises in a second step after acquisition of additional oncogenic driver mutations (including Jak1 and Jak3). Interestingly, although the analyzed BCP-ALLs of Pax5+/- mice were all based on the same predisposition, they greatly varied in their transcriptional profile and cell-cycle state, depending on their oncogenic 2 nd hit. Conclusion: In summary, we deeply characterized how reduced Pax5 transcriptional activity in the BM generates a predisposed precursor B-cell environment, which is susceptibility for malignant transformation. These findings are important for understanding the molecular mechanisms to prevent or treat a significant proportion of childhood BCP-ALLs.

Synergistic Anti-Cancer Effects of Second-Generation Proteasome Inhibitor Carfilzomib with Doxorubicin and Dexamethasone Via p53-Mediated Apoptosis in Pre-B Acute Lymphoblastic Leukemia Cells

Background: The ubiquitin-proteasome system (UPS) plays a crucial role in regulating the levels and functions of a large number of proteins in the cell, which are important for cancer cell growth and survival. The proteasome is highly activated in B-cell precursor acute lymphoblastic leukemia (BCP-ALL), which is the most common malignancy in children. The attempt to inhibit proteasome as a therapeutic strategy has been successful for some malignancies.
 Materials and Methods: In this experimental study, human BCP-ALL cell lines NALM-6 and SUP-B15 were treated with carfilzomib with and without the chemotherapeutic agents. The XTT assay evaluated the viability of the cells. Cell cycle analysis and apoptosis assay were assessed by flow cytometry. RQ-PCR and western blotting evaluated the expression of pro-/anti-apoptotic signals. A drug combination study for synergistic or additive effects of carfilzomib with doxorubicin or dexamethasone was performed.
 Results: We observed that carfilzomib alone induced G2/M cell cycle arrest and caspase-dependent apoptosis in the human BCP-ALL cells (NALM-6 and SUP-B15). Gene and protein expression analysis indicated the upregulation of pro-apoptotic as well as downregulation of the cell survival and proliferative signals (P-value<0.05). The synergy of carfilzomib with doxorubicin or dexamethasone was revealed in BCP-ALL cells.
 Conclusion: Our results indicated that proteasome inhibition induces p53-mediated apoptosis in BCP-ALL cells. Since carfilzomib has a synergistic effect with anti-leukemic agents doxorubicin and dexamethasone in BCP-ALL cells, this combined-modality approach might be befitting for patients who do not respond well to conventional chemotherapy.

A Genome-Wide CRISPR-Cas9 Screen Reveals GATOR1 Complex Is a Critical Regulator of Glucocorticoid Sensitivity in B-Cell Precursor Acute Lymphoblastic Leukemia

Glucocorticoids (GCs) are essential components of B-cell precursor acute lymphoblastic leukemia (B-ALL) therapy. Upon GC binding, GC receptor (GR) translocates to the nucleus and activates GR target genes. While GCs reportedly induce cell death via apoptosis and/or cell cycle arrest, their molecular mechanisms remain largely unknown. To identify molecules/pathways governing GC sensitivity in B-ALL cells, we performed genome-wide CRISPR-Cas9 dropout screens in the presence or absence of dexamethasone (Dex) using two human B-ALL lines (NALM-6 and MUTZ-5). To do so, we employed a baboon envelope pseudotyped lentiviral vector system, which enabled highly efficient transduction of human B cells (Sasaki et al. Blood 2022). We identified a total of 90 and 46 genes, whose single-guide RNAs (sgRNAs) were significantly enriched upon Dex treatment in NALM-6 and MUTZ-5, respectively. A total of 15 genes were identified in both lines, and among them included those previously implicated in GR resistance, such as NR3C1 and TBL1XR1. Strikingly, sgRNAs targeting DEPDC5, NPRL2 and NPRL3, each encoding components of the GATOR1 complex, a negative regulator of mTORC1, were significantly enriched in the presence of Dex in both lines. To validate these findings, we depleted each gene via lentivirus-based E2-Crimson-tagged sgRNA and assessed effects on cell growth over 11 days with or without Dex. Cells expressing sgRNA targeting either DEPDC5, NPRL2 or NPRL3, but not those transduced with empty vector, demonstrated a proliferative advantage relative to non-transduced cells in the presence of Dex. Importantly, Dex treatment significantly reduced phospho-P70S6K levels, a readout of the mTOR pathway activation, in GATOR1-intact cells, but not in NPRL2 or DEPDC5 knockout cells, suggesting that Dex suppresses mTORC1-mediated growth signals via the GATOR1 complex. We next asked whether Dex/GATOR1-mediated mTORC1 suppression is mediated by transcriptional activity of GR. To do so, we established a NALM-6 line deficient in endogenous GR in which a GR mutant lacking DNA binding capacity was exogenously expressed. Dex treatment did not reduce phospho-P70S6K levels in these cells, suggesting that the Dex/GATOR1-mediated mTORC1 suppression seen in GATOR1-intact cells primarily depends on transcriptional capacity of GR. Given that loss-of-function mutations of DEPDC5 were reported in therapy-refractory B-ALL cases and that GATOR1 complex reportedly functions in amino acid metabolisms, our findings suggest a potential link between GC sensitivity and amino acid metabolism in B-ALL cells.

Association between Dysfunction of the Nucleolar Stress Response and Multidrug Resistance in Pediatric Acute Lymphoblastic Leukemia.

Simple SummaryPediatric patients with B-cell precursor acute lymphoblastic leukemia (BCP-ALL) relapse or are refractory to chemotherapy despite the low frequency of TP53 mutations. The nucleolar stress response is one of the mechanisms that activate P53 by ribosomal protein L11 (RPL11). We hypothesized that the lack of nucleolar stress response is related to chemoresistance and relapse in some pediatric BCP-ALL cases. We revealed that clinical BCP-ALL therapeutics, such as 6-mercaptopurine, methotrexate, daunorubicin, and cytarabine, induced the nucleolar stress response, and its treatment susceptibility was dependent on the nucleolar stress response. Furthermore, we observed decreased RPL11 expression at relapse in seven children with BCP-ALL in comparison to that at onset. Our findings provide new insights into the anti-leukemia mechanism in BCP-ALL and multidrug resistance and relapse via the nucleolar stress response, suggesting that the nucleolar stress response may be a potential therapeutic strategy to predict chemosensitivity and improve chemoresistance in pediatric BCP-ALL.Approximately 20% of pediatric patients with B-cell precursor acute lymphoblastic leukemia (BCP-ALL) relapse or are refractory to chemotherapy despite the low frequency of TP53 mutations. The nucleolar stress response is a P53-activating mechanism via MDM2 inhibition by ribosomal protein L11 (RPL11). We analyzed the role of the nucleolar stress response using BCP-ALL cell lines and patient samples by drug sensitivity tests, Western blotting, and reverse transcription polymerase chain reaction. We revealed that the nucleolar stress response works properly in TP53 wild-type human BCP-ALL cell lines. Next, we found that 6-mercaptopurine, methotrexate, daunorubicin, and cytarabine had anti-leukemic effects via the nucleolar stress response within BCP-ALL treatment. Comparing the samples at onset and relapse in children with BCP-ALL, RPL11 mRNA expression decreased at relapse in seven of nine cases. Furthermore, leukemia cells with relapse acquired resistance to these four drugs and suppressed P53 and RPL11 expression. Our findings suggest that the nucleolar stress response is a novel anti-leukemia mechanism in BCP-ALL. As these four drugs are key therapeutics for BCP-ALL treatment, dysfunction of the nucleolar stress response may be related to clinical relapse or refractoriness. Nucleolar stress response may be a target to predict and improve the chemotherapy effect for pediatric BCP-ALL.

An instructive role for Interleukin-7 receptor \u03b1 in the development of human B-cell precursor leukemia

Kinase signaling fuels growth of B-cell precursor acute lymphoblastic leukemia (BCP-ALL). Yet its role in leukemia initiation is unclear and has not been shown in primary human hematopoietic cells. We previously described activating mutations in interleukin-7 receptor alpha (IL7RA) in poor-prognosis “ph-like” BCP-ALL. Here we show that expression of activated mutant IL7RA in human CD34+ hematopoietic stem and progenitor cells induces a preleukemic state in transplanted immunodeficient NOD/LtSz-scid IL2Rγnull mice, characterized by persistence of self-renewing Pro-B cells with non-productive V(D)J gene rearrangements. Preleukemic CD34+CD10highCD19+ cells evolve into BCP-ALL with spontaneously acquired Cyclin Dependent Kinase Inhibitor 2 A (CDKN2A) deletions, as commonly observed in primary human BCP-ALL. CRISPR mediated gene silencing of CDKN2A in primary human CD34+ cells transduced with activated IL7RA results in robust development of BCP-ALLs in-vivo. Thus, we demonstrate that constitutive activation of IL7RA can initiate preleukemia in primary human hematopoietic progenitors and cooperates with CDKN2A silencing in progression into BCP-ALL.

Glucocorticoid Resistance in IKZF1-Deleted BCP-ALL: It Is PTEN Again

▪Single copy deletions of IKZF1, which occur in 10-15% of all B cell precursor acute lymphoblastic leukemia (BCP-ALL) cases, are associated with a poor outcome. We previously showed that loss of IKZF1 dictates resistance to glucocorticoids (GC) in BCP-ALL cell lines, a knockout mouse model and ex-vivo analysis of primary leukemic cells. When we analyzed the initial response to prednisolone therapy, we found that pediatric patients who suffer from an IKZF1 deleted leukemia are strongly enriched in the poor responder group (14 vs 7%, p<0.05). Although we were able to show that glucocorticoid receptor (GR)-mediated transcription regulation is attenuated in IKZF1 deleted cells, in remained unclear how loss of IKZF1 affects GR signaling.In T-ALL, GC resistance is frequently induced by genomic mutations that result in hyperactivation of AKT. AKT mediated phosphorylation of the GR results in suppression of GR function and resistance to GC treatment. Although the genomic loci that are frequently mutated in T-ALL resulting in hyperactivation of AKT, such as PTEN, are unaffected in BCP-ALL, we tested whether aberrant activation of AKT may also cause resistance to GC in BCP-ALL. Indeed, both CRISPR/Cas9 induced knockout and shRNA mediated knockdown of IKZF1 resulted in activation of AKT in human BCP-ALL cell lines. This hyperactivation was also observed in splenic B cells isolated from Ikzf1+/- mice. Active AKT in turn promotes phosphorylation of the GR on Ser143. This phosphorylation inhibits GR function, at least in part, by preventing shuttling of the ligand bound protein to the nucleus.In order to understand how a single copy loss of IKZF1 activates AKT, we tested expression of upstream regulators of AKT activity and found a near complete loss of PTEN expression, the phosphatase that inhibits PI3K and AKT signaling. Indeed, shRNA mediated knockdown of PTEN in wildtype cells resulted in activation of AKT and induced resistance to GCs. Furthermore, in IKZF1+/- cells we observed an upregulation of HES1, a transcriptional factor that is known to repress expression of the PTEN gene. Of note, HES1 was previously identified as a target of IKZF1 mediated transcriptional repression in erythroid and T cells.To test whether this AKT mediated mechanism of GC resistance could be reverted through pharmacological inhibition, we tested the effect of AKT inhibitors on prednisolone induced apoptosis of human BCP-ALL cell lines (figure A). In contrast to control cells, where inhibition of AKT had little to no effect on the sensitivity to GC, in IKZF1 deleted cells AKT inhibition reverted the resistance phenotype to a level comparable to control cells. Importantly, also in patient derived xenografts we observed that AKT inhibition was able to sensitize cells to GC treatment (Figure B).In conclusion, we identified hyperactivation of AKT, as a result of aberrant HES1 expression, as the mechanism causing GC resistance in IKZF1 deleted BCP-ALL. The availability of AKT inhibitors that are under clinical evaluation, may allow the development of combination therapies that restore the response to GCs in IKZF1-deleted BCP-ALL.Figures:A) IKZF1, PTEN, AKT, pAKT(Ser473), pGR(Ser134) and GR protein expression levels of NALM6 IKZF1+/- cell lysates were analyzed by western blot. Non-targeting controls were used as control and actin was used as a loading control.B) NALM6 parental, non-targeting controls or IKZF1+/- cells were treated for 48 hours with increasing concentrations of prednisolone with or without 0.5µM AKT inhibitor MK2206 and analyzed using an MTT based viability assay. All values were normalized to untreated NALM6 cells. Error bars represent ± standard error of the mean (SEM).(C) hTERT immortalized MSCs were seeded in a 96 wells format and allowed to settle for 24 hours prior to the addition of BCP-ALL patient-derived xenografts (PDX). PDX cells were allowed to settle for 24 hours before prednisone was added in increasing concentrations with or without AKT inhibitor MK2206. After 3 days, sensitivity to treatment was measured by flowcytometry using an amine staining to test for membrane integrity as a measure of cell viability. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

PAX5-ELN oncoprotein promotes multistep B-cell acute lymphoblastic leukemia in mice

PAX5 is a well-known haploinsufficient tumor suppressor gene in human B-cell precursor acute lymphoblastic leukemia (B-ALL) and is involved in various chromosomal translocations that fuse a part of PAX5 with other partners. However, the role of PAX5 fusion proteins in B-ALL initiation and transformation is ill-known. We previously reported a new recurrent t(7;9)(q11;p13) chromosomal translocation in human B-ALL that juxtaposed PAX5 to the coding sequence of elastin (ELN). To study the function of the resulting PAX5-ELN fusion protein in B-ALL development, we generated a knockin mouse model in which the PAX5-ELN transgene is expressed specifically in B cells. PAX5-ELN-expressing mice efficiently developed B-ALL with an incidence of 80%. Leukemic transformation was associated with recurrent secondary mutations on Ptpn11, Kras, Pax5, and Jak3 genes affecting key signaling pathways required for cell proliferation. Our functional studies demonstrate that PAX5-ELN affected B-cell development in vitro and in vivo featuring an aberrant expansion of the pro-B cell compartment at the preleukemic stage. Finally, our molecular and computational approaches identified PAX5-ELN-regulated gene candidates that establish the molecular bases of the preleukemic state to drive B-ALL initiation. Hence, our study provides a new in vivo model of human B-ALL and strongly implicates PAX5 fusion proteins as potent oncoproteins in leukemia development.

High STAP1 expression in DUX4-rearranged cases is not suitable as therapeutic target in pediatric B-cell precursor acute lymphoblastic leukemia

Approximately 25% of the pediatric B-cell precursor acute lymphoblastic leukemia (BCP-ALL) cases are genetically unclassified. More thorough elucidation of the pathobiology of these genetically unclassified (‘B-other’) cases may identify novel treatment options. We analyzed gene expression profiles of 572 pediatric BCP-ALL cases, representing all major ALL subtypes. High expression of STAP1, an adaptor protein downstream of the B-cell receptor (BCR), was identified in BCR-ABL1-like and non-BCR-ABL1-like B-other cases. Limma analysis revealed an association between high expression of STAP1 and BCR signaling genes. However, STAP1 expression and pre-BCR signaling were not causally related: cytoplasmic Igμ levels were not abnormal in cases with high levels of STAP1 and stimulation of pre-BCR signaling did not induce STAP1 expression. To elucidate the role of STAP1 in BCP-ALL survival, expression was silenced in two human BCP-ALL cell lines. Knockdown of STAP1 did not reduce the proliferation rate or viability of these cells, suggesting that STAP1 is not a likely candidate for precision medicines. Moreover, high expression of STAP1 was not predictive for an unfavorable prognosis of BCR-ABL1-like and non-BCR-ABL1-like B-other cases. Remarkably, DUX4-rearrangements and intragenic ERG deletions, were enriched in cases harboring high expression of STAP1.

Erratum to: Bortezomib interferes with adhesion of B cell precursor acute lymphoblastic leukemia cells through SPARC up-regulation in human bone marrow mesenchymal stromal/stem cells.

The poor prognosis of adults with B cell precursor acute lymphoblastic leukemia (BCP-ALL) is attributed to leukemia cells that are protected by the bone marrow (BM) microenvironment. In the present study, we explored the pharmacological targeting of mesenchymal stromal/stem cells in BM (BM-MSCs) to eliminate chemoresistant BCP-ALL cells. Human BCP-ALL cells (NALM-6 cells) that adhered to human BM-MSCs (NALM-6/Ad) were highly resistant to multiple anti-cancer drugs, and exhibited pro-survival characteristics, such as an enhanced Akt/Bcl-2 pathway and increased populations in the G0 and G2/S/M cell cycle stages. Bortezomib, a proteasome inhibitor, interfered with adhesion between BM-MSCs and NALM-6 cells and up-regulated the matricellular protein SPARC (secreted protein acidic and rich in cysteine) in BM-MSCs, thereby reducing the NALM-6/Ad population. Inhibition of SPARC expression in BM-MSCs using a small interfering RNA enhanced adhesion of NALM-6 cells. Conversely, recombinant SPARC protein interfered with adhesion of NALM-6 cells. These results suggest that SPARC disrupts adhesion between BM-MSCs and NALM-6 cells. Co-treatment with bortezomib and doxorubicin prolonged the survival of BCP-ALL xenograft mice, with a significant reduction of leukemia cells in BM. Our findings demonstrate that bortezomib contributes to the elimination of BCP-ALL cells through disruption of their adhesion to BM-MSCs, and offer a novel therapeutic strategy for BCP-ALL through targeting of BM-MSCs.

Bortezomib interferes with adhesion of B cell precursor acute lymphoblastic leukemia cells through SPARC up-regulation in human bone marrow mesenchymal stromal/stem cells.

The poor prognosis of adults with B cell precursor acute lymphoblastic leukemia (BCP-ALL) is attributed to leukemia cells that are protected by the bone marrow (BM) microenvironment. In the present study, we explored the pharmacological targeting of mesenchymal stromal/stem cells in BM (BM-MSCs) to eliminate chemoresistant BCP-ALL cells. Human BCP-ALL cells (NALM-6 cells) that adhered to human BM-MSCs (NALM-6/Ad) were highly resistant to multiple anti-cancer drugs, and exhibited pro-survival characteristics, such as an enhanced Akt/Bcl-2 pathway and increased populations in the G0 and G2/S/M cell cycle stages. Bortezomib, a proteasome inhibitor, interfered with adhesion between BM-MSCs and NALM-6 cells and up-regulated the matricellular protein SPARC (secreted protein acidic and rich in cysteine) in BM-MSCs, thereby reducing the NALM-6/Ad population. Inhibition of SPARC expression in BM-MSCs using a small interfering RNA enhanced adhesion of NALM-6 cells. Conversely, recombinant SPARC protein interfered with adhesion of NALM-6 cells. These results suggest that SPARC disrupts adhesion between BM-MSCs and NALM-6 cells. Co-treatment with bortezomib and doxorubicin prolonged the survival of BCP-ALL xenograft mice, with a significant reduction of leukemia cells in BM. Our findings demonstrate that bortezomib contributes to the elimination of BCP-ALL cells through disruption of their adhesion to BM-MSCs, and offer a novel therapeutic strategy for BCP-ALL through targeting of BM-MSCs.

Unique Long Non-Coding RNA Expression Signature in ETV6/RUNX1-Driven B-Cell Precursor Acute Lymphoblastic Leukemia

Emerging evidence suggests that long non-coding RNAs (lncRNAs) are critically involved in a variety of human tumor entities and the identification of cancer-associated lncRNAs might reveal new prognostic biomarkers or novel therapeutic targets for the treatment of human cancer. In this study, we characterized the lncRNA expression signature associated with ETV6/RUNX1-positive pediatric B-cell precursor acute lymphoblastic leukemia (BCP-ALL), one of the most prevalent genetic subtypes of childhood leukemia.First, we used primary leukemia patient samples to identify an ETV6/RUNX1 specific expression signature consisting of 596 lncRNA transcripts. Besides using primary BCP-ALL patient samples, we also integrated these lncRNA expression data with RNA sequencing results from a panel of human BCP-ALL leukemia cell lines and identified a unique lncRNA expression profile of 16 lncRNAs exclusively associated with the presence of the ETV6/RUNX1 fusion protein. Notably, lnc-SARDH-1 (also known as DBH-AS1) was the only lncRNA from this list for which an oncogenic role has previously been postulated.Given that previous studies revealed putative overlap between lncRNA expression and the presence of regulatory enhancer elements, we also evaluated the distribution of H3K27ac, a histone modification associated with enhancer activity, at the genomic loci of the 16 ETV6/RUNX1 specific lncRNAs mentioned above. Notably, broad H3K27ac binding was identified for lnc-NKX2-3-1, lnc-RTN4R-1, lnc-GIP-1, lnc-LRP8-3, lnc-TCF12-2, lnc- C8orf4-1, lnc-C8orf4-2, lnc-TINAGL1-1 and lnc-LSM11-4 in ETV6/RUNX1-positive REH cells, whereas more discrete H3K27ac peaks were identified at putative promoter regions for lnc-TIMM21-5 and lnc-CPT2-7. Interestingly, coding genes adjacent to some of these lncRNAs showed unique overexpression in ETV6/RUNX1-positive BCP-ALLs, suggesting a possible cis regulatory relationship between these lncRNAs and their nearby protein coding genes. Finally, we applied shRNA-mediated silencing of endogenous ETV6/RUNX1, integrated these expression profiles with the patient and cell line data, to show that lnc-NKX2-3-1, lnc-TIMM21-5, lnc-ASTN1-1 and lnc-RTN4R-1 were the only lncRNAs that were truly regulated by the oncogenic fusion protein.Interestingly, subsequent lncRNA modulation experiments using LNA GapmeR technology revealed that lnc-TIMM21-5 and lnc-ASTN1-1 modulation has no effect on overall transcription, suggesting that these lncRNAs might act at the translational level or/and at various steps of mRNA processing. In contrast, lnc-NKX2-3-1 and lnc-RTN4R-1 perturbations resulted in severe changes in gene expression, suggesting an alternative mechanism of action for these lncRNAs that is more transcriptionally oriented. Most notably, loss of lnc-RTN4R-1expression significantly affected part of the ETV6/RUNX1-specific mRNA expression signature as exemplified by reduced levels of AK7, PTPRKand GBA3.Altogether, our study identified a panel of ETV6/RUNX1 specific lncRNAs that might be implicated in the biology of human BCP-ALL and could serve as novel biomarkers or novel therapeutic targets for the treatment of this prevalent subtype of human leukemia.TLM and PVV are shared last authors DisclosuresNo relevant conflicts of interest to declare.

Dissecting the mechanisms underling acute lymphoblastic leukemia brain metastasis

Abstract Despite improved treatments, patient with B-cell precursor acute lymphoblastic leukemia (BCP-ALL) experience relapse with devastating tumor cell dissemination into the central nervous system (CNS). As an “immune privilege site”, CNS is protected from infiltration of immune cells by specialized structure of blood-brain barrier with astrocyte endfeet that surround the blood vessels. To assess whether leukemia cells hijack mechanisms employed by inflamed immune cells to invade into the brain parenchyma, we set up leukemia mouse xenograft models with well-characterized human BCP-ALL cell lines and primary cells derived from patients. We found massive infiltration/colonization of GFP+ leukemia cells inside brain vessels and parenchyma addition to meningeal space. Immunohistochemistry staining showed that activation of astrocytes following the induction of VEGF-A in brains of leukemia burden mice. We address phenotype changes in leukemia cells, that are associated with brain metastasis by comparing the expression of chemokine receptors and cytokines in leukemia cells recovered from bone marrow and other extramedullary sites (spleen, liver) for comparison to the cells recovered from brain tissue. By investigation of extravasation steps in intact brain, we also try to figure how leukemia cells bleach the BBB and find new niche to colonize at brain parenchyma. Ongoing studies will provide new insights into the mechanisms that regulate leukemia CNS infiltration.

Tumor Suppressors BTG1 and IKZF1 Cooperate during Mouse Leukemia Development and Impact Relapse Rate in Childhood Acute Lymphoblastic Leukemia

Deletions and mutations affecting transcription factor IKZF1 are associated with increased relapse risk and poor outcome in B cell precursor acute lymphoblastic leukemia (BCP-ALL). However, additional genetic events may either enhance or negate the effects of IKZF1 on prognosis. We observed that deletions of the gene encoding the transcriptional coregulator BTG1 frequently co-occured with loss of IKZF1 function, suggesting a synergistic role for these events during leukemia development or progression. Targeted deletion of Btg1 predisposed both Btg1+/- and Btg1-/- mice to T cell malignancies, similar to what has been observed in Ikzf1 heterozygous knockout animals. Hence, while somatic single single-copy losses of either BTG1 and IKZF1 in the patient are predominantly found in BCP-ALL, targeted deletion of these genes in the mouse gives rise to T cell malignancies.To establish whether loss of BTG1 function affected the tumor suppressive role of IKZF1, the Btg1 knockout allele was crossed onto mice heterozygous for a loss-of-function Ikzf1 allele. Leukemia penetrance in these compound mice increased in a Btg1 dose-dependent manner. These leukemias were characterized by clonal TCRb rearrangement and aggressive infiltration into secondary organs, indicating synergistic roles for these tumor suppressors during mouse leukemia development.To investigate the effects of combined IKZF1/BTG1 loss in human BCP-ALL, we examined a large pediatric cohort of BCP-ALL cases, and found that the combined presence of BTG1 and IKZF1 deletions was associated with a markedly higher incidence of relapse, relative to IKZF1-deleted cases without BTG1 aberrations. Similar to BTG1 copy number losses, deletions in EBF1, PAX5, RB1 and CDKN2A/B appeared to be selectively enriched in IKZF1 deleted ALL. However, in contrast to BTG1, none of these other copy number alterations affected relapse incidence or outcome in this patient group. In conclusion, our data demonstrate synergy between the tumor suppressors BTG1 and IKZF1 during mouse leukemia development while the combined (single copy) loss of these two tumor suppressors identifies a patient group with an extremely poor outcome.Event free survival in a cohort of 514 children newly diagnosed with BCP-ALL, divided into four categories based on IKZF1 and BTG1 deletion status. [Display omitted] DisclosuresPieters:Eusa Pharma: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees.

Bortezomib Attenuates Adhesion of B Cell Precursor Acute Lymphoblastic Lleukemia Cells to Bone Marrow Mesenchymal Stromal/Stem Cells Via Regulating SPARC Expression

The prognosis of adults with B cell precursor acute lymphoblastic leukemia (BCP-ALL) is poor. Many cases in complete remission experience relapse of leukemia despite intensive chemotherapy or hematopoietic stem cell transplantation. This clinical observation suggests that minimal residual disease (MRD) still exists after these intensive therapies. Cell adhesion-mediated drug resistance (CAM-DR) is one of the mechanisms to support MRD in the bone marrow microenvironment (Clin Cancer Res 14:9, 2008). Mesenchymal stromal/stem cells (MSCs) are a cellular component of bone marrow (BM) and maintain physiological precursor B lymphopoiesis (Nature Rev Immunol 6:107, 2006). We investigated our hypothesis that survival of BCP-ALL cells is supported by their direct adhesion to BM-MSCs in BM microenvironment. First, we confirmed that BCP-ALL cells exhibited their drug resistant phenotype through adhesion to human BM-MSCs using human BCP-ALL cell line, Nalm6. We isolated human BM-MSCs from normal bone marrow samples (AllCells, Emeryville, CA). When Nalm6 cells were co-cultured with human BM-MSCs, they were prone to adhere to BM-MSCs. Nalm6 adhered to BM-MSCs (Nalm6/ad) were more resistant to the treatment with various anti-cancer drugs including doxorubicin than Nalm6 in suspension (Nalm6/su). Immunoblot analysis showed that expression level of anti-apoptotic protein Bcl-2 and phosphorylation level of pro-survival kinase Akt are higher in Nalm6/ad compared with those in Nalm6/su. In cell cycle analysis using Ki67/PI fluorescence-activated cell sorter (FACS) assay, the percentage of populations in G0 phase and S/G2/M phase was higher in Nalm6/ad compared with that in Nalm6/su, which indicated the increase of MRD and the high proliferation of leukemic cells in adhesive population, respectively. Therefore, we considered that detachment of Nalm6 from BM-MSCs is important to restore chemosensitivity of BCP-ALL cells and to eliminate these cells. Accordingly, we screened drugs that are capable of disrupting the adhesion of Nalm6 to BM-MSCs, and found that several proteasome inhibitors had a such activity. BM-MSCs were treated with bortezomib (10nM for 24h or 100nM for 1h), carfilzomib (3 nM for 24 h or 100 nM for 1 h) or oprozomib (10 nM for 24 h or 300 nM for 4 h), and then washed with PBS and co-cultured with Nalm6 cells. We confirmed that the number of BM-MSCs is not affected with these drugs irrespective of their concentration and incubation time by MTT assay. Co-cultured cells were divided into suspension cells and adherent cells. Each population was stained with monoclonal antibodies against CD19 and CD90 to detect Nalm6 and BM-MSCs, respectively, and analyzed by FACS. When BM-MSCs were pre-treated with bortezomib, carfilzomib or oprozomib, the number of Nalm6/ad was significantly decreased, and inversely, the number of Nalm6/su was increased, as compared to that of co-cultures with untreated BM-MSCs. Intriguingly, BM-MSCs treated with bortezomib showed aberrant expression of secreted protein acidic and rich in cysteine (SPARC). Immunoblot analysis showed that SPARC expression of BM-MSCs was transiently increased after bortezomib treatment. In co-cultures with BM-MSCs transfected with siRNA targeting SPARC, the number of Nalm6/ad was increased. In addition, the number of Nalm6 adhered to BM-MSCs pre-treated with recombinant human SPARC was lower than that adhered to control non-treated BM-MSCs. Therefore, SPARC showed anti-adhesive property and was one of the molecules associated with adhesion of BM-MSCs to Nalm6. Finally, we tested whether bortezomib shows therapeutic effects in vivo. Bortezomib administration supported survival of BCP-ALL xenograft model mice utilizing Nalm6, which was concomitant with the decrease in leukemia cells in the femur and in the size of spleen, as compared to those in non-treated control mice. In summary, Nalm6 cells that adhered to BM-MSCs contributed to construction of chemoresistant population similar to MRD. Here, we found that this population could be detached by treatment of BM-MSCs with bortezomib through transient increase in their SPARC expression. Our findings suggest that bortezomib or other proteasome inhibitors might be promising drugs to treat BCP-ALL through attenuating the adhesion of leukemic cells to BM-MSCs, and shed new insight into a therapeutic strategy in BCP-ALL by targeting BM-MSCs. Disclosures No relevant conflicts of interest to declare.

Functional Analysis of the Subclonal Architecture of B-Cell Precursor Acute Lymphoblastic Leukemia (BCP-ALL) at a Single Cell Level

Introduction: Transforming events in B-lineage (ALL) occur primarily at the level of committed progenitor cells, but the phenotype, frequency and hierarchical organization of leukemia-initiating cells (LICs) are controversial. Pronounced clonal heterogeneity in ALL and the lack of preclinical models impede functional analysis of LICs and dissecting the clonal architecture of B-cell precursor acute lymphoblastic leukemia (BCP-ALL).Aims and methods: To identify cellular features associated with cell fate and decisions, engraftment potential, growth kinetics and resistance to therapy of BCP-ALL LICs at the clonal level, we generated multiple single-cell-derived clones from patient derived-long-term cultures (PD-LTCs) of human BCP-ALL. This in vitro ALL-LTC model allows prolonged serum-, cytokine-, and stroma-free culture of patient-derived BCP-ALL cells that are largely stable in terms of cytogenetic and immunophenotypic features for up to 6 months [Nijmeijer B et. al, Exp Hematol. 2009;37:376; Badura S et al., PLoS One. 2013; 8:e80070] Three ALL-LTCs were selected on the basis of different patterns of surface marker expression and cytogenetics, i.e. Ph+ ALL (LTCs PH and BV; common-ALL) and BCR-ABL negative ALL (LTC CR; pre B-ALL), respectively.Results: Clonogenic growth in semisolid medium ranged from 0.25% to 8% for these ALL-LTCs (CR 8%, PH 0.45% and BV 0.25%), consistent with functional diversity within the bulk as well as the immunophenotypically defined subsets. To analyze these subpopulations at the clonal level, we isolated leukemic cell subsets by cell sorting on the basis of CD20, CD34 and CD38 expression, and then generated hundreds of clones by expanding single-cell sorted subpopulations. All sorted cell fractions from all three ALL-LTCs were capable of initiating long-term growth in vitro. In limiting dilution cultures, 1% to 5% of single sorted cells derived from these cell fractions gave rise to prolonged leukemic cell growth. To evaluate the leukemia-initiating capacity of the clonal ALL subpopulations in vivo, we examined engraftment and outgrowth kinetics of different ALL clones in a xenograft model with sublethally irradiated NSG mice. Individual clones derived from different sorted subpopulations displayed pronounced differences in engraftment potential and aggressiveness based on analyses of blood, bone marrow and spleen. This ranged from rapid appearance of ALL and death within 70-80 days (clone PH 20), to complete lack of leukemic outgrowth 155 days after transplantation (clone PH clone 14). The other four clones derived from this ALL (PH) displayed intermediate engraftment and outgrowth kinetics. Notably, leukemogenic properties were not associated with a distinct surface marker profile based in CD20, CD34 and CD38 expression. A similar heterogeneity of leukemogenic potential was determined with clonal subpopulations from the LTC BV.To assess the hierarchical relationship between different clones and the association with LIC capacity, we examined the pattern of immunoglobulin VDJ-rearrangement of different clones. The Ig rearrangement patterns confirm that the clones represent distinct clonal populations derived from the original polyclonal LTC, but we observed no hierarchical relationship with respect to the biological properties of different clones. Further functional evaluation of these subclones was performed by videomicroscopy-based single cell tracking that allows analysis of individual cells and their progeny over many cell divisions. We observed concordance between cell cycle length and cell death events as determined by single cell tracking and the engraftment potential and aggressiveness of LTCs derived single cell clones in vivo.To elucidate the molecular basis for the biologic differences between the clonal ALL populations, we compared highly and poorly leukemogenic clones by SILAC based proteomics. Only 5% of 6500 proteins measured in clones differing in their leukemia-initiating potential were differentially regulated, and these proteins could be assigned to a limited number of pathways.In conclusion, comparative evaluation of clonal lymphoblast populations generated from patient-derived ALL long-term cultures by combined in vitro analysis, single cell videomicroscopy, xenotransplantation and proteomics is a promising approach to identify specific markers and therapeutic strategies that target LIC in ALL. DisclosuresWojcik:Novartis: Research Funding, Travel, Accommodation Other; SAnofi Aventis: Consultancy, Travel, Accommodation, Travel, Accommodation Other. Lang:Novartis: Research Funding, Travel, Accommodation Other. Brüggemann:Amgen Inc.: Consultancy, Research Funding.

Btg1-Deficiency Promotes ETV6-RUNX1-Mediated Leukemic Transformation By Upregulation of BCL6

Translocation t(12;21) (p13;q22), giving rise to the ETV6-RUNX1 fusion gene, is the most common genetic abnormality in childhood B-cell precursor acute lymphoblastic leukemia (BCP-ALL). The ETV6-RUNX1 translocation arises in utero, but its expression is insufficient to induce leukemia and requires other cooperating genetic lesions for BCP-ALL to develop. Deletions affecting the transcriptional coregulator BTG1 are commonly observed in BCP-ALL (9%), but are significantly enriched in ETV6-RUNX1-positive leukemia (25%). The BTG1 protein displays no intrinsic enzymatic activity but may act by recruiting effector molecules like protein arginine methyltransferase 1 (PRMT1) to specific transcription factors. Here, we show that ETV6-RUNX1 interacts both with BTG1 and PRMT1, and this interaction is lost in c-Kit+Ter-119-Btg1-/- fetal liver (FL) derived hematopoietic progenitors (HPCs). Moreover, targeted deletion of Btg1 enhanced the proliferative capacity of ETV6-RUNX1 in FL-HPCs as measured by enhanced colony-forming and serial replating capacity (Figure 1). The combined loss of Btg1 function and ETV6-RUNX1 expression correlated with strong upregulation of the proto-oncogene Bcl6 and downregulation of BCL6 target genes, such as p19Arf and Tp53 (Figure 2). Similarly, ectopic expression of BCL6 promoted both proliferation and replating capacity of FL-derived progenitor cells in the presence of SCF, FLT3L and IL-7. This phenotype correlated with a fivefold suppression of p19Arf and a twofold suppression of Tp53 expression. Inhibition of BCL6 in a variety of human BCP-ALL cell lines by the peptide inhibitor RI-BPI resulted in decreased proliferation and induction of apoptosis as measured by Annexin-V staining. These included the ETV6-RUNX1-positive cell lines UOC-B6, AT-2 and REH, the BCR-ABL1-positive cell line SD1, as well as Nalm6. Together our results point to a novel role for BCL6 in promoting cell proliferation of primitive progenitor B cells and suggest that targeted inhibition of BCL6 may be effective in the treatment of various BCP-ALL subtypes. [Display omitted] [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

Abstract B25: A human-mouse xenograft model to evaluate therapies and study the role of TSLP-induced signals in Ph-like ALL

Abstract While the overall survival rate for children with B cell precursor acute lymphoblastic leukemia (B-ALL) is high, a subset of children with this disease are at high risk for relapse and death. Genome-wide analysis has shown that gene expression profiles in these high-risk B-ALLs is similar to that of Philadelphia chromosome–positive ALL and these are designated Ph-like ALL. Approximately half of Ph-like ALL are characterized by genetic defects resulting in overexpression of CRLF2. CRLF2, together with the IL-7Rα, forms a receptor complex that is activated by the cytokine, TSLP. The JAK-STAT5 pathway is phosphorylated downstream of this receptor complex activation. The activating JAK mutations found in some CRLF2 B-ALL have led to speculation that TSLP stimulation is not a factor in CRLF B-ALL. In preliminary studies to address this question we evaluated the effect of TSLP on a CRLF2 B-ALL cell lines with JAK defects and which have been reported to exhibit constitutive JAK-STAT5 activation. Our data show that TSLP increases STAT5 phosphorylation in these cell lines and also in primary CRLF2 B-ALL cells. Our next step was to evaluate the role of TSLP-CRLF2 interactions in vivo in the human-mouse xenograft model. However, mouse TSLP is different from most other cytokines produced in the xenograft in that it is species-specific and does not activate the human TSLP receptor complex that includes CRLF2. Thus, traditional xenograft models do not provide the TSLP-CRLF2 interactions that we believe to be a major factor in CRLF2 B-ALL. To overcome this obstacle we engineered immune-deficient NOD/SCID IL-2Rγ null (NSG) mice to express human TSLP (hTSLP+ mice) as well as control mice that lack the TSLP cytokine (hTSLP– mice). ELISA assays show serum hTSLP levels in the hTSLP+ mice that approximate the normal range in human serum. We used this hTSLP+/- xenograft model system to study the in vivo effects of TSLP on mice transplanted with a CRLF2 B-ALL. We used this hTSLP+/– xenograft model system to evaluate the in vivo effects of TSLP on survival and proliferation of transplanted CRLF2 B-ALL cells harboring a JAK defect (MUTZ5 cell line). Mice were euthanized at 5 weeks and BM was harvested. Evaluation of BM disease by flow cytometry showed that the percentage of viable human leukemia cells in hTSLP+ mice was twice that observed in hTSLP– mice. Evaluation of cell cycle progression in human CRLF2 B-ALL cells isolated from xenograft BM showed that the percentage of cycling cells in hTSLP+ mice was 2.5 fold higher than in hTSLP– mice. When primary Ph-like ALL cells were transplanted to produce hTSLP+/– xenografts, the viable pre-B ALL cells present in the BM of hTSLP+ mice showed higher expression levels of the TSLPR components (CRLF2 and IL-7Rα) than those in the hTSLP- mice. These data provide evidence that the TSLP produced in this model is active and that it impacts primary pre-B ALL cells. Preliminary data obtained from this model suggests that TSLP provides a signal that promotes in vivo survival of CRLF2 B-ALL cells and that it may play a role in selection of leukemia clones during in vivo leukemogenesis. Microarray analysis comparing gene expression in primary CRLF2 B-ALL cells isolated from hTSLP+ and hTSLP– xenograft mice identified 565 that genes are differentially regulated (&amp;gt; 2 fold up or downregulated; p&amp;lt;.05). Ingenuity Pathway Analysis is currently underway to identify the signaling pathways that are regulated by hTSLP in CRLF2 B-ALL in vivo in the hTSLP+/– xenograft model. The identification of genes downstream of TSLP-CRLF2 signaling has the potential of providing drug targets for combination therapy to effectively treat Ph-like B-ALL. The hTSLP+/– xenograft model provides an important tool for evaluating the in vivo efficacy of these and other drugs to treat CRLF2 B-ALL. Citation Format: Ruijun Su, Francis L. Olivia, Shannalee R. Martinez, Ineavely Baez, Terry Ann Milford, Terrence Bennett, Ross Fisher, Christopher L. Morris, Sinisa Dovat, Kimberly J. Payne. A human-mouse xenograft model to evaluate therapies and study the role of TSLP-induced signals in Ph-like ALL. [abstract]. In: Proceedings of the AACR Special Conference: The Translational Impact of Model Organisms in Cancer; Nov 5-8, 2013; San Diego, CA. Philadelphia (PA): AACR; Mol Cancer Res 2014;12(11 Suppl):Abstract nr B25.

Abstract B21: A human-mouse xenograft model to evaluate therapies and study the role of TSLP-induced signals in Ph-like ALL

Abstract While the overall survival rate for children with B cell precursor acute lymphoblastic leukemia (B-ALL) is high, a subset of children with this disease are at high risk for relapse and death. Genome-wide analysis has shown that gene expression profiles in these high-risk B-ALLs is similar to that of Philadelphia chromosome–positive ALL and these are designated Ph-like ALL. Approximately half of Ph-like ALL are characterized by genetic defects resulting in overexpression of CRLF2. CRLF2, together with the IL-7Rα, forms a receptor complex that is activated by the cytokine, TSLP. The JAK-STAT5 pathway is phosphorylated downstream of this receptor complex activation. The activating JAK mutations found in some CRLF2 B-ALL have led to speculation that TSLP stimulation is not a factor in CRLF B-ALL. In preliminary studies to address this question we evaluated the effect of TSLP on a CRLF2 B-ALL cell lines with JAK defects and which have been reported to exhibit constitutive JAK-STAT5 activation. Our data show that TSLP increases STAT5 phosphorylation in these cell lines and also in primary CRLF2 B-ALL cells. Our next step was to evaluate the role of TSLP-CRLF2 interactions in vivo in the human-mouse xenograft model. However, mouse TSLP is different from most other cytokines produced in the xenograft in that it is species-specific and does not activate the human TSLP receptor complex that includes CRLF2. Thus, traditional xenograft models do not provide the TSLP-CRLF2 interactions that we believe to be a major factor in CRLF2 B-ALL. To overcome this obstacle we engineered immune-deficient NOD/SCID IL-2Rγ null (NSG) mice to express human TSLP (hTSLP+ mice) as well as control mice that lack the TSLP cytokine (hTSLP– mice). ELISA assays show serum hTSLP levels in the hTSLP+ mice that approximate the normal range in human serum. We used this hTSLP+/- xenograft model system to study the in vivo effects of TSLP on mice transplanted with Ph-like B-ALL. First, we used the hTSLP+/– xenograft model system to evaluate the in vivo effects of TSLP on survival and proliferation of transplanted CRLF2 B-ALL cells harboring a JAK defect (MUTZ5 cell line). Mice were euthanized at 5 weeks and BM was harvested. Evaluation of BM disease by flow cytometry showed that the percentage of viable human leukemia cells in hTSLP+ mice was twice that observed in hTSLP– mice. Evaluation of cell cycle progression in human CRLF2 B-ALL cells isolated from xenograft BM showed that the percentage of cycling cells in hTSLP+ mice was 2.5 fold higher than in hTSLP– mice. When primary Ph-like ALL cells were transplanted to produce hTSLP+/– xenografts, the viable pre-B ALL cells present in the BM of hTSLP+ mice showed higher expression levels of the TSLPR components (CRLF2 and IL-7Rα) than those in the hTSLP- mice. These data provide evidence that the TSLP produced in this model is active and that it impacts primary pre-B ALL cells. Preliminary data obtained from this model suggests that TSLP provides a signal that promotes in vivo survival of CRLF2 B-ALL cells and that it may play a role in selection of leukemia clones during in vivo leukemogenesis. Microarray analysis comparing gene expression in primary CRLF2 B-ALL cells isolated from hTSLP+ and hTSLP– xenograft mice identified 565 that genes are differentially regulated (&amp;gt; 2 fold up or downregulated; p&amp;lt;.05). Ingenuity Pathway Analysis is currently underway to identify the signaling pathways that are regulated by hTSLP in CRLF2 B-ALL in vivo in the hTSLP+/– xenograft model. The identification of genes downstream of TSLP-CRLF2 signaling has the potential of providing drug targets for combination therapy to effectively treat Ph-like B-ALL. The hTSLP+/– xenograft model will provide an important tool for evaluating the in vivo efficacy of these and other drugs to treat CRLF2 B-ALL. Citation Format: Olivia Francis, Ruijun Su, Shannalee Martinez, Ineavely Baez, Terry-Ann Milford, Terrence Bennett, Ross Fisher, Christopher L. Morris, Sinisa Dovat, Kimberly J. Payne. A human-mouse xenograft model to evaluate therapies and study the role of TSLP-induced signals in Ph-like ALL. [abstract]. In: Proceedings of the AACR Special Conference on Pediatric Cancer at the Crossroads: Translating Discovery into Improved Outcomes; Nov 3-6, 2013; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2013;74(20 Suppl):Abstract nr B21.