Overcoming Wnt-β-catenin dependent anticancer therapy resistance in leukaemia stem cells.

Leukaemia stem cells (LSCs) underlie cancer therapy resistance but targeting these cells remains difficult. The Wnt–β-catenin and PI3K–Akt pathways cooperate to promote tumorigenesis and resistance to therapy. In a mouse model in which both pathways are activated in stem and progenitor cells, LSCs expanded under chemotherapy-induced stress. Since Akt can activate β-catenin, inhibiting this interaction might target therapy-resistant LSCs. High-throughput screening identified doxorubicin (DXR) as an inhibitor of the Akt–β-catenin interaction at low doses. Here we repurposed DXR as a targeted inhibitor rather than a broadly cytotoxic chemotherapy. Targeted DXR reduced Akt-activated β-catenin levels in chemoresistant LSCs and reduced LSC tumorigenic activity. Mechanistically, β-catenin binds multiple immune-checkpoint gene loci, and targeted DXR treatment inhibited expression of multiple immune checkpoints specifically in LSCs, including PD-L1, TIM3 and CD24. Overall, LSCs exhibit distinct properties of immune resistance that are reduced by inhibiting Akt-activated β-catenin. These findings suggest a strategy for overcoming cancer therapy resistance and immune escape. Cohorts of leukemic mice were treated with vehicle, chemotherapy, [Low]DXR, clinical-dose DXR or chemotherapy + [Low]DXR as in Figure S4. At 10 days post-treatment (day 1 = start of treatment), blast cells, LSCs, and HSPCs were sorted using stringent gates (see Figure 5A) for these populations. For each population, two biological replicates were made with 1-3 technical replicates each. Each biological replicate was a pool of sorted cells from 4-5 male and 4-5 female mice. Technical replicates were samples of sorted cells from each biological replicate. 1,000 cells per sample were sorted into 96-well plates with 7 ul lysis buffer containing RNase inhibitor (2 U/μl) from BM. First-strand cDNA synthesis and cDNA libraries were constructed using the SMARTer ultra low input RNA kit for sequencing - v3 (Clontech) following the manufacturer’s instructions. cDNA quality was determined by Agilent high sensitivity DNA kit on Agilent 2100 BioAnalyzer (Agilent Technologies). Libraries were sequenced at 50bp on the Illumina HiSeq 2500. Reads were aligned to UCSC mm10 with Tophat 2.1.1 [Kim et al 2013], using gene models from Ensembl 80. Read counts per gene were obtained with HTSeq-count 0.6.0 [Simon et al 2014]. Analysis was done in R with the EdgeR package [Robinson et al 2010] using default methods. Significantly changed genes were determined at FDR<1e-3 and fold change exceeding ±1.5. GO analysis was generated using significantly changed genes by Metascape (www.metascape.org). Kim D, Pertea G, Trapnell C, Pimentel H, Kelley R, Salzberg SL. TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. . Genome Biology 2013, 14:R36 Simon Anders, Paul Theodor Pyl, Wolfgang Huber. HTSeq — A Python framework to work with high-throughput sequencing data. Bioinformatics (2014), in print, online at doi:10.1093/bioinformatics/btu638 Robinson MD, McCarthy DJ and Smyth GK (2010). “edgeR: a Bioconductor package for differential expression analysis of digital gene expression data.” Bioinformatics, 26, pp. -1.