JAK/BCL2 inhibition acts synergistically with LSD1 inhibitors to selectively target ETP-ALL

ETP-ALL (Early T cell Progenitor Acute Lymphoblastic Leukemia) represents a high-risk subtype of T cell acute lymphocytic leukemia (T-ALL). Therapeutically, ETP-ALL patients frequently relapse after conventional chemotherapy highlighting the need for alternative therapeutic approaches. Using our ZEB2Tg ETP-ALL mouse model we previously documented the potential utility of the catalytic LSD1 inhibitor (GSK2879552) for treating mouse/human ETP-ALL. However, this approach proved to be inefficient, especially in killing human LOUCY cell ETP-ALL xenografts in vivo. Here we have revealed the novel involvement of ZEB2/LSD1 complexes in repressing the intrinsic apoptosis pathway by inhibiting the expression of several pro-apoptotic proteins such as BIM (BCL2L11) as a major driver for ETP-ALL survival. Treatment with LSD1i (particularly with the steric inhibitor SP2509) restored the expression of ZEB2/LSD1 pro-apoptotic BIM (BCL2L11) target. In combination with a JAK/STAT pathway inhibitor (JAKi, Ruxolitinib) or with a direct inhibitor of the anti-apoptotic BCL2 protein (BCL2i, ABT-199) resistance of human and mouse ETP-ALL to LSD1i was reversed. This new combination approach efficiently inhibited the growth of human and mouse ETP-ALL cells in vivo by enhancing their differentiation and triggering an apoptotic response. These results set the stage for novel combination therapies to be used in clinical trials to treat ETP-ALL patients.

unique peaks were defined by bedtools Intersect intervals (version 2.29.0). We used the plotHeatmap (Version 3.3.2.0.1) to create a heatmap for score distributions across LSD1, ZEB2 and STAT5 peak genomic regions. Association of common peaks (ZEB2, LSD1 and STAT5) with genes was performed with GREAT (http://great.stanford.edu/public/html/index.php) using the Basal plus extension setting. Check the GO biological process obtained with GREAT. The list of genes related to regulation of intrinsic apoptotic signaling pathway was illustrated by STRING (https://string-db.org).

Animal experimentation and handling.
All experiments were performed according to the regulations and guidelines of the Ethics Committee for care and use of laboratory animals at the University of Manitoba. The mouse (Mus musculus) cohorts used in these experiments were sibling littermates and maintained. Both male and female mice were included in the various analyses and ranged in age from 3 to 60 weeks.

Patient-derived xenograft models in NSG mice.
NSG (NOD.Cg-Prkdc<scid>Il2rg<tm1Wjl>SzJ) male/female mice (8-10 wk of age) were purchased from The Jackson Laboratory and maintained under sterile conditions at the University of Manitoba Animal Care Facility. Leukemic blasts from patients (characterized and provided by The Quebec Leukemia Cell Bank) were expanded in NSG mice by transplanting 0.5 × 10 6 to 2 × 10 6 cells via intravenous injection. Engraftment of human leukemic blasts in the peripheral blood was assessed by FACS staining for human CD45 and human CD7 during a period of 1-6 months.
Mice were euthanized when the engraftment of leukemic blasts was 5-30%, and BM and spleen cells were harvested. All the ETP-ALL and mature T-ALL cell patients are characterized and provided by The Quebec Leukemia Cell Bank (BCLQ) with the consent of all subjects involved.

Drug treatment of xenograft Patient-derived cells.
To ensure the high purity of blasts harvested from the BM of engrafted NSG mice, we used the EasySep™ FITC Positive Selection Kit II (StemCell Technologies, 17682) to sort the stained blasts with the anti-human CD7 FITC. For the drug treatment in vitro, from each condition, the sorted blasts were cultured on OP9 stromal cells expressing the NOTCH ligand Delta-like-1 (DL1) as previously described (Benyoucef et al. 2015). Briefly, 1x 10 5 cells from each sample were plated in α minimum essential medium with 10% fetal calf serum (StemCell Technologies,06450) and 10% serum (Sigma-Aldrich, H4522) supplemented with 50 ng/mL hSCF cytokine, 20 ng/mL Flt3-L cytokine, 20 nM insulin, 10ng/mL IL-7, and 100 U/mL penicillin and 100 mg/mL streptomycin. All human cytokines are provided by peprotech Canada. The cells are treated for 5 days with DMSO (as control condition), Ruxolitinib alone (5uM), GSK-LSD1 alone (100nM), SP-2905 alone (250nM), , or with drug combination indicted in the Figure 7 and S8. The apoptosis of blasts after treatment was assessed as mentioned previously.

Human mRNA expression in primary human samples.
We used the mRNA expression data in human patient ETP-ALL (n=12) vs mature T-ALL (n=40) samples previously published (Zhang J et al, 2012, Gutierrez A et al. 2011, and publicly available at Gene Expression Omnibus-NCBI, GSE28703.           Table S1.
Supplementary table1 excel file. It contains the common ChIP peak locations. List of genes associated with common peaks. The results of the PANTHER, GREAT analysis, STRING intrinsic apoptosis genes of the common peaks. Table S2.
The list of primers used in this study.

Figure S1
E l u t i o n 1 I n p u t s u p e r n a t a n t E l u t i o n 2 E l u t i o n 1 I n p u t s u p e r n a t a n t