MSI2-HOXA9 expression leads to cancer cell growth advantage and differentiation arrest

(A) Schematic of the MSI2-HOXA9 fusion protein. The fusion retains both MSI2 RNA binding domains and the HOXA9 DNA binding domain. (B) Experimental design for in vitro colony assay or in vivo transplantation of BCR-ABL/Control or BCR-ABL/MSI2-HOXA9 expressing KLS cells. Transplanted mice were subsequently assessed for chimerism, survival, and cancer cell differentiation. (C) Primary colony assay of BCR-ABL/Control or BCR-ABL/MSI2-HOXA9 expressing KLS cells. **P=0.007 (n=3 technical replicates). (D) Secondary colony assay of BCR-ABL/Control or BCR-ABL/MSI2-HOXA9 expressing KLS cells. **P=0.009 (n=3 technical replicates). (E) Chimerism of BCR-ABL/Control or BCR-ABL/MSI2-HOXA9 transplanted cells at 13 days. ***P=0.0005 (n=5 mice per group). (F-I) Lin-BCR-ABL/Control, BCR-ABL/NUP98-HOXA9, or BCR-ABL/MSI2-HOXA9 expressing cells were sorted from primary transplants and used to generate cytospins that were then stained with Giemsa and May-Grunwald stains (n=4 mice per group). (F) Quantification of blast cells. **P=0.001, ***P=0.0001 (n=4 mice per group). (G) Quantification of immature granulocytes. *P=0.01, **P=0.004. (H) Quantification of mature granulocytes. **P=0.004, ***P=0.0005. (I) Quantification of differentiated macrophages and monocytes. **P=0.002. (J) Survival of mice transplanted with BCR-ABL/Control or BCR-ABL/MSI2-HOXA9 expressing KLS cells. **P=0.002. Two-tailed unpaired Student’s t-tests were used to determine statistical significance.

BCR-ABL/MSI2-HOXA9 cell growth and differentiation is dependent on MSI2 RRM1

(A) Schematic of MSI2-HOXA9 fusion protein and mutant versions generated to test domain dependencies. ΔRRM1 and ΔRRM2 are deletions of the MSI2 RNA binding domains 1 and 2 respectively, while ΔHOXA9 introduces a stop codon at the break point to eliminate the HOXA9 DNA binding domain. (B) Primary colony formation of KLS cells expressing BCR-ABL/Control or BCR-ABL + variations of the MSI2-HOXA9 fusion protein. *=significance from BCR-ABL/Control, ***P=0.0002, ****P<0.0001; $=significance from BCR-ABL/MSI2-HOXA9, $P=0.01, $$$P=0.0001 for ΔRRM1, $$$P=0.0002 for ΔRRM2 (n=3 technical replicates). (C) Chimerism of BCR-ABL/Control, BCR-ABL/ΔRRM1, or BCR-ABL/ΔHOXA9 cells 14 days post-transplantation. ***P=0.0009 significance from BCR-ABL/Control, $$$P=0.0004 significance from BCR-ABL/ΔRRM1 (n=6 mice for BCR-ABL/Control, n=7 for BCR-ABL/ΔRRM1, n=4 for BCR-ABL/ΔHOXA9). (D) Primary colony formation of KLS cells expressing BCR-ABL/Control or BCR-ABL + MSI2-HOXA9 fusion protein or full length wild-type MSI2. *=significance from BCR-ABL/Control, *P=0.02, ****P<0.0001; $=significance from BCR-ABL/MSI2-HOXA9, $$$P=0.0002 (n=3 technical replicates). (E) Secondary colony formation of KLS cells expressing BCR-ABL/Control or BCR-ABL + MSI2-HOXA9 fusion protein or full length wild-type MSI2. *=significance from BCR-ABL/Control, ****P<0.0001; $=significance from BCR-ABL/MSI2-HOXA9, $$$P=0.0002 (n=3 technical replicates). (F-I) Lin-BCR-ABL/Control or BCR-ABL + variations of the MSI2-HOXA9 fusion protein were sorted from primary transplants and used to generate cytospins that were then stained with Giemsa and May-Grunwald stains. *=significance from BCR-ABL/Control, $=significance from BCR-ABL/MSI2-HOXA9 (n=4 mice per group). (F) Quantification of blast cells. *P=0.01, **P=0.001, ***P=0.0008, $$$P=0.0004. (G) Quantification of immature granulocytes. *P=0.01, **P=0.004, $$P=0.005. (H) Quantification of mature granulocytes. *P=0.02, **P=0.004, $$P=0.006, $$$P=0.0005. (I) Quantification of differentiated macrophages and monocytes. *P=0.01, **P=0.002. (J) Survival of mice transplanted with BCR-ABL/Control, BCR-ABL/ΔRRM1, or BCR-ABL/ΔHOXA9 expressing KLS cells. *P=0.01 significance from BCR-ABL/Control, $$P=0.007 significance from BCR-ABL/ΔRRM1. Two-tailed unpaired Student’s t-tests were used to determine statistical significance.

MSI2-HOXA9 regulates gene expression including mitochondrial, developmental, and oncogenic genes

(A) Schematic of RNAseq workflow. (B) Network map of differential genes from RNAseq of Lin-BCR-ABL/Control vs. Lin-BCR-ABL/MSI2-HOXA9. (C) Heatmap of developmental genes from RNAseq of Lin-BCR-ABL/Control vs. Lin-BCR-ABL/MSI2-HOXA9. (D) Heatmap of oncogenic genes from RNAseq of Lin-BCR-ABL/Control vs. Lin-BCR-ABL/MSI2-HOXA9. (E) Heatmap of mitochondrial genes from RNAseq of Lin-BCR-ABL/Control vs. Lin-BCR-ABL/MSI2-HOXA9. (F) Gene Set Enrichment Analysis plot of mitochondrial transcription program generated from RNAseq of Lin-BCR-ABL/Control vs. Lin-BCR-ABL/MSI2-HOXA9. (G) Validation of selected mitochondrial genes upregulated in Lin-BCR-ABL/MSI2-HOXA9 relative to Lin-BCR-ABL/Control. ****P<0.0001. Two-tailed unpaired Student’s t-tests were used to determine statistical significance.

Elevated mitochondria gene expression leads to increased cellular respiration and energetics

(A) Schematic of experimental workflow to assess Lin-BCR-ABL/Control vs. Lin-BCR-ABL/MSI2-HOXA9 cancer cell energetics using the Seahorse XFp Cell Mito Stress Test Kit. (B) Oxygen consumption rate (OCR) of Lin-BCR-ABL/Control and Lin-BCR-ABL/MSI2-HOXA9 to measure mitochondrial respiration (n=3 technical replicates for each cell type). (C) Basal respiration rate of Lin-BCR-ABL/Control and Lin-BCR-ABL/MSI2-HOXA9 determined from the OCR trace. **P=0.003. (D) ATP production rate of Lin-BCR-ABL/Control and Lin-BCR-ABL/MSI2-HOXA9 determined from the OCR trace. **P=0.002. (E) Maximum respiration rate of Lin-BCR-ABL/Control and Lin-BCR-ABL/MSI2-HOXA9 determined from the OCR trace after the first FCCP injection. *P=0.01. (F) Extracellular acidification rate (ECAR) of Lin-BCR-ABL/Control and Lin-BCR-ABL/MSI2-HOXA9 to measure glycolysis (n=3 technical replicates for each cell type). (G) “Basal” glycolysis of Lin-BCR-ABL/Control and Lin-BCR-ABL/MSI2-HOXA9 determined from the ECAR trace, ****P<0.0001. (H) Glycolytic reserve of Lin-BCR-ABL/Control and Lin-BCR-ABL/MSI2-HOXA9 determined from the ECAR trace, **P=0.006. (I) “Maximum” glycolysis of Lin-BCR-ABL/Control and Lin-BCR-ABL/MSI2-HOXA9 determined from the ECAR trace, ***P=0.0002. (J) Energetic landscape of Lin-BCR-ABL/Control and Lin-BCR-ABL/MSI2-HOXA9 generated by plotting the basal OCR vs. basal ECAR for each cell type. (K) Mitochondrial state determined from OCR data. *P=0.01. Two-tailed unpaired Student’s t-tests were used to determine statistical significance.

MSI2-HOXA9 localizes to the nucleus and increases Polrmt expression

(A) qRT-PCR of genes directly involved in regulation of mitochondrial gene expression. *P=0.02 for Tfb2m, *P=0.01 for Nrf1, ****P<0.0001 for Polrmt. (B) Representative immunofluorescent images of K562 cells expressing Flag-tagged MSI2 or Flag-tagged MSI2-HOXA9 (scale bar=10μm). (DAPI= blue, FLAG-MSI2 or FLAG-MSI2-HOXA9= green, mitotracker= red). (C) Overlap of Flag-tagged MSI2 or MSI2-HOXA9 with mitochondria. ****P<0.0001 (n=11 for MSI2, n=13 for MSI2-HOXA9). (D) Overlap of Flag-tagged MSI2 or MSI2-HOXA9 with the nucleus. ****P<0.0001 (n=11 for MSI2, n=13 for MSI2-HOXA9). (E) Proposed mechanism by which the MSI2-HOXA9 fusion protein promotes an undifferentiated aggressive cancer cell. Two-tailed unpaired Student’s t-tests were used to determine statistical significance.

Analysis of BCR-ABL/MSI2-HOXA9 leukemia

(A-C) Representative brightfield image of (A) BCR-ABL/Control, (B) BCR-ABL/NUP98-HOXA9, (C) BCR-ABL/MSI2-HOXA9 cytospins of lin-cancer stained with Giemsa and May-Grunwald stains. (D) Quantification of BCR-ABL/MSI2-HOXA9 and BCR-ABL Lin-content within the cancer population by FACS. ****P<0.0001, n=4 for each group. (E) Secondary colony formation of KLS cells expressing BCR-ABL/Control or BCR-ABL + variations of the MSI2-HOXA9 fusion protein. *=significance from BCR-ABL/Control, *P=0.04, **P=0.007 ***P=0.0004; $=significance from BCR-ABL/MSI2-HOXA9, $P=0.03 for ΔRRM1, $P=0.01 for ΔRRM2, $P=0.02 ΔHOXA9 (n=3 technical replicates). (F-H) Representative brightfield image of (F) BCR-ABL/ΔRRM1, (G) BCR-ABL/ΔRRM2 (H) BCR-ABL/ΔHOXA9 cytospins of lin-cancer stained with Giemsa and May-Grunwald stains.

Principal component analysis and mitochondrial parameters associated with BCR-ABL/MSI2-HOXA9 leukemia

(A) Principal component analysis of RNAseq data of Lin-BCR-ABL/Control and Lin-BCR-ABL/MSI2-HOXA9. (B) Coupling efficiency determined from OCR of Lin-BCR-ABL/Control and Lin-BCR-ABL/MSI2-HOXA9 (n=3 technical replicates). (C) Non-mitochondrial O2 consumption determined from OCR of Lin-BCR-ABL/Control and Lin-BCR-ABL/MSI2-HOXA9 (n=3 technical replicates). (D) Maximum respiration determined from OCR of Lin-BCR-ABL/Control and Lin-BCR-ABL/MSI2-HOXA9 following the second FCCP injection, an oversaturating concentration (n=3 technical replicates). (E) Proton leak determined from OCR of Lin-BCR-ABL/Control and Lin-BCR-ABL/MSI2-HOXA9. *P=0.01 (n=3 technical replicates). (F) Glucose uptake of Lin-BCR-ABL/Control and Lin-BCR-ABL/MSI2-HOXA9 measured by the YSI bioanalyzer at day 1 (D1) and day 3 (D3) post-plating (n=2 technical replicates). (G) Lactate production of Lin-BCR-ABL/Control and Lin-BCR-ABL/MSI2-HOXA9 measured by the YSI bioanalyzer at day 1 (D1) and day 3 (D3) post-plating (n=2 technical replicates). (H) Ratio of lactate production to glucose consumption of Lin-BCR-ABL/Control and Lin-BCR-ABL/MSI2-HOXA9 at day 1 (D1), day 2 (D2), and day 3 (D3) post-plating (n=2 technical replicates).

Examples of flow cytometry gating strategies

(A) Gating strategy used for KLS sorting from mouse bone marrow. KLS cells were used as starting material for all mouse leukemia models in this paper. (B) Gating strategy used for sorting lin-bcCML from primary transplants. This strategy was used to generate the cell source for histology, transplants, and RNAseq. (C) Gating strategy used for isolating FLAG-tagged transduced K562 cells. After 48hrs in the presence of BCR-ABL + FLAG-tagged wild-type MSI2 or FLAG-tagged MSI2/HOXA9, cells were sorted to isolate doubly infected cells. This strategy was used for the localization experiments (Fig. 5B-D).