The RUNX1-ETO target gene RASSF2 suppresses t(8;21) AML development and regulates Rac GTPase signaling

Large-scale chromosomal translocations are frequent oncogenic drivers in acute myeloid leukemia (AML). These translocations often occur in critical transcriptional/epigenetic regulators and contribute to malignant cell growth through alteration of normal gene expression. Despite this knowledge, the specific gene expression alterations that contribute to the development of leukemia remain incompletely understood. Here, through characterization of transcriptional regulation by the RUNX1-ETO fusion protein, we have identified Ras-association domain family member 2 (RASSF2) as a critical gene that is aberrantly transcriptionally repressed in t(8;21)-associated AML. Re-expression of RASSF2 specifically inhibits t(8;21) AML development in multiple models. Through biochemical and functional studies, we demonstrate RASSF2-mediated functions to be dependent on interaction with Hippo kinases, MST1 and MST2, but independent of canonical Hippo pathway signaling. Using proximity-based biotin labeling we define the RASSF2-proximal proteome in leukemia cells and reveal association with Rac GTPase-related proteins, including an interaction with the guanine nucleotide exchange factor, DOCK2. Importantly, RASSF2 knockdown impairs Rac GTPase activation, and RASSF2 expression is broadly correlated with Rac-mediated signal transduction in AML patients. Together, these data reveal a previously unappreciated mechanistic link between RASSF2, Hippo kinases, and Rac activity with potentially broad functional consequences in leukemia.

each plate. 24 hours following media change, IMDM media containing retroviral particles was collected, passed through a 0.45 μm syringe filter, pooled, and supplemented with recombinant cytokines mIL-3 (10 ng/mL), hIL-6 (10 ng/mL), and mSCF (20 ng/mL), and polybrene (4 μg/mL). For retroviral transduction, HSC-enriched bone-marrow mononuclear cells harvested from mice were resuspended in this supplemented retroviral supernatant at densities of 2-3 x 10 6 cells/mL and centrifuged (2,000x g) in 6-well plates at 32°C for 3 hours (Allegra X-12R centrifuge, Beckman Coulter); followed by overnight culture at 37°C. Two consecutive retroviral transductions were performed in this manner on subsequent days. Transduction efficiency was measured (GFP+ frequency) by flow cytometry the morning following the second centrifugation (immediately prior to transplantation into recipient mice). Low retroviral titers were utilized, which were typically measured at ~1.5-2% GFP+ following two rounds of transduction.
For lentiviral transduction of shRNA constructs: Transfections of HEK293T cells were conducted by combining 5 μg of psPAX2, 2.5 μg of pMD2.G, and 3 μg of respective pLKO.1-based shRNA vector, and 44 μl of polyethylenimine (PEI) in 1 mL of Opti-MEM reduced serum medium (Gibco, #31985-070). Approximately 16 hours post-transfection, media was aspirated, cells were washed once in PBS, and 10 mL fresh RPMI supplemented with 10% fetal bovine serum was added to each plate. 24 hours following media change, RPMI media containing lentiviral particles was collected, passed through a 0.45 μm syringe filter, pooled, and supplemented with polybrene (final working concentration: 4 μg/mL). For lentiviral transduction, THP-1 cells were resuspended in this supplemented retroviral supernatant at densities of ~ 0.5 x 10 6 cells/mL and centrifuged (2,000x g) in 6-well plates at 32°C for 3 hours (Allegra X-12R centrifuge, Beckman Coulter); followed by overnight culture at 37°C. Two consecutive transductions were performed in this manner on subsequent days. Cell lysis and RNA isolation were performed using Trizol reagent (ThermoFisher Scientific, #15596026) according to manufacturer's instructions. cDNA was prepared from 0.5 -1 μg RNA using qScript cDNA Supermix (Quanta, #95048) according to manufacturer's protocol. Quantitative PCR was performed using KAPA SYBR Fast 2X Master Mix (KAPA Biosystems, #KK4618) according to manufacturer's protocol, in 20 μl reactions, each performed in technical duplicates. qPCR reactions were performed using a BioRad CFX Connect instrument. Data analysis was performed using a standard delta-delta Ct method relative to the geometric-mean of two reference genes, GAPDH and POLR2A.

Apoptosis Measurement
Apoptotic cell death was monitored by flow cytometry using the APC Annexin V apoptosis kit with 7-AAD (BioLegend, San Diego, CA) as described by manufacturer's protocol.

PAK1-based Pull-down Assay for Active Rac-GTP Measurement
Briefly, lysates from approximately 6 x 10 6 cells were prepared as described above but with 1X GTPase assay/lysis buffer (25 mM HEPES, pH 7.5, 150 mM NaCl, 1% NP-40, 10 mM MgCl 2 , 1 mM EDTA, 2% Glycerol). After clearing, 5% of cell lysates were separated and denatured in 2X loading buffer for use as input protein measurements. For pull-down, remaining cell lysates were combined with 36 μL PAK1 PBD agarose beads (#STA-411) and incubated at 4°C with rotating either for 3 hours (human AML cell lines) or overnight (HEK293T cells). Following this, beads were washed 3x with ice-cold 1X Assay/Lysis buffer via centrifugation at 11,000 xg. Washed beads were then resuspended in 50 μl 2x loading buffer and denatured prior to loading. Immunoblotting was performed using the Li-Cor Odyssey infrared imaging instrument. Post-acquisition image analysis and cropping was performed using Li-Cor Image Studio Lite (V 5.2.5) software. For measuring Rac-GTP in cells, pan-Rac antibody (see Antibody section), rather than anti-Rac1 monoclonal antibody provided by kit was used. Positive and negative assay controls using 100X GTPγS and 100X GDP were performed in parallel using excess cell lysates.

RNA-seq Analysis
Post-QC, mRNA from eukaryotic organisms is enriched using oligo(dT) beads. First, the mRNA is fragmented randomly by adding fragmentation buffer, then the cDNA is synthesized by using mRNA template and random hexamers primer, after which a custom second-strand synthesis buffer (Illumina) , dNTPs, RNase H and DNA polymerase I are added to initiate the second-strand synthesis. Second, after a series of terminal repair, ligation and sequencing adaptor ligation, the double-stranded cDNA library is completed through size selection and PCR enrichment. The qualified libraries are fed into HiSeq/MiSeq sequencers after pooling according to its effective concentration and expected data volume. Differential expression analysis was conducted by established methods using HISAT2 for human genome alignment (hg38), followed by featureCounts and DESeq2 for differential gene expression analysis.

Gene Ontology Analysis
Gene ontology analysis was performed using Metascape (metascape.org). Gene list of RASSF2specific protein hits was uploaded and compared against full human genome list as background.

Flow Cytometry and Cell Sorting
Flow cytometric analysis was conducted using a BD FACSCanto instrument equipped with standard lasers (488 nm, 640 nm) and filters. Data collection was performed using BD FACSDiva software.
Compensation was set up using appropriate single-stained controls and positive-staining gates were established using appropriate FMO controls. Post-acquisition data analysis was performed using