Acute myeloid leukemia (AML) is the most common and deadly leukemia in adults. Although most AML patients respond to first-line chemotherapy, relapse is common. Thus, the five-year overall survive rate remains under 50%. The high relapse rate of patients who achieve “complete remission” is largely due to the difficulties of eradicating minimal residual AML cells, using current therapeutic strategies. Thus, additional therapies to target and eradicate residual AML are greatly needed.

In the Chapter 1, we report the development of a novel nanoparticle-based AML vaccination immunotherapy for the treatment of AML. Our vaccine platform utilizes AML cell membrane-coated nanoparticles (AMCNPs), in which immune stimulatory adjuvant-loaded nanoparticles are coated with leukemic cell membrane material. The AMCNP vaccine stimulates leukemia-specific immune responses by co-delivering membrane-associated antigens along with adjuvants to antigen-presenting cells. The AMCNP vaccines were efficiently acquired by antigen-presenting cells in vitro and in vivo and stimulate antigen cross-presentation. Vaccination with AMCNPs significantly enhanced antigen-specific T cell expansion and effector function. Furthermore, in an AML post-remission vaccination model, AMCNP vaccination significantly enhanced survival in comparison to control vaccination. Our AMCNP vaccination strategy is multi-antigenic, fully personalized, and obviates the need for neoantigen identification. Thus, AMCNPs are a promising platform that can be further developed as an AML vaccination immunotherapy.

In chapter 2, we examined the post-transcriptional regulation of the AML1-ETO oncofusion gene. The AML1-ETO oncofusion gene is created by the t(8;21)(q22;q22) chromosomal translocation, which characterizes a common AML subtype [t(8;21) AML]. Post-transcriptional regulation of gene expression is often mediated through interactions between trans-factors and cis-elements within transcript 3'-untranslated regions (UTR). AML1-ETO uses the 3’UTR of the ETO gene, which is not normally expressed in hematopoietic cells. Therefore, the mechanisms regulating AML1-ETO expression via the 3’UTR are attractive therapeutic targets. Here, we identify a negative regulatory element within the AML1-ETO 3’UTR and demonstrate that the let-7b microRNA directly represses AML1-ETO through this site, inhibits the proliferation of t(8;21) AML cell lines, rescues expression of AML1-ETO target genes, and promotes differentiation. We conclude that AML1-ETO is posttranscriptional regulated by let-7b, which contributes to the leukemic phenotype of t(8;21) AML.

The t(8;21)(q22;q22) generates the oncofusion protein RUNX1-ETO (AML1-ETO, RUNX1-RUNX1T1, AML1-MTG8) and is one of the most common chromosomal abnormalities associated with acute myeloid leukemia (AML). Loss of a sex chromosome (LOS) is detected in 32-59% of t(8;21) AML patients, which suggests haploinsufficiency of key tumor suppressors on the sex chromosomes may be involved in t(8;21) leukemogenesis. Though LOS is frequently observed in t(8;21) patients, there have been no mechanistic studies examining its role in the pathogenesis of t(8;21) AML. The gene encoding the granulocyte-macrophage colony-stimulating factor (GM-CSF) receptor, CSF2RA, is located on the sex chromosomes. With LOS, t(8;21) cells suffer haploinsufficiency of CSF2RA, resulting in reduced GM-CSF signaling. It was recently discovered that GM-CSF signaling, which is compromised upon LOS, inhibits t(8;21) leukemogenesis. Therefore, this dissertation explores the mechanisms by which GM-CSF signaling prevents t(8;21) leukemogenesis and aims to identify mediating mechanisms that are therapeutically targetable. Gene expression profiling of GM-CSF-treated RUNX1-ETO cells revealed that GM-CSF signaling elicits a unique gene expression profile in RUNX1-ETO cells. This gene expression profile correlates with attenuated MYC-associated gene signatures and human myelopoiesis, which together reduce the self-renewal potential of RUNX1-ETO cells and inhibit leukemogenesis. To identify GM-CSF-induced genes mediating the inhibition of RUNX1-ETO leukemogenesis, a functional screen was conducted to identify several genes, including Mxi1, which reduce the self-renewal potential of RUNX1-ETO cells. MXI1 is a well-established inhibitor of MYC; therefore, the effects of MXI1 and MYC inhibition were examined in t(8;21) cells. It was found that MYC inhibition promotes apoptosis and reduces the proliferation and self-renewal potential of t(8;21) cells, demonstrating that MYC is critical for maintaining leukemic potential and may serve as an effective target in t(8;21) AML treatment. Furthermore, it was discovered that in addition to being downregulated upon LOS-driven haploinsufficiency, CSF2RA is also downregulated in t(8;21) cell lines through its 3’ UTR. This reveals an additional mechanism present in t(8;21) cells to reduce GM-CSF signaling and promote leukemogenesis. Together, these studies highlight the importance of GM-CSF signaling in attenuating MYC in t(8;21) cells and elucidate the mechanisms by which t(8;21) cells downregulate CSF2RA to reduce GM-CSF signaling.

RASSF2 belongs to the Ras-association domain family (RASSFs) and was previously identified as a transcriptional target of RUNX1-ETO associated with the t(8;21) acute myeloid leukemia. RASSF2 was reported to interact with Hippo kinase MST1/MST2 and RAS family GTPase KRAS but its association with other GTPases remained unexplored. In this study, we carefully selected 19 small GTPases guided by their specific expression in bone marrow and association with proteins proximal to RASSF2. As a result, we discovered six novel small GTPases that interact with RASSF2, namely Rab32, Rac2, RIT1, RhoH, Rab8a, and RhoA. Structural modeling predictions suggest the possibility of direct bindings between these small GTPases and RASSF2. Notably, the expression levels of Rac2 and Rab32 correlate with the survival of AML patients. This study advances our understanding of RASSF2's biological role as a Ras effector-like protein and provides valuable insights into RASSF2’s potential contributions to hematopoiesis and the development of AML.

Acute Myloid Leukemia (AML) is one of the most common leukemia in adults with poor five years survival even with modern chemotherapy treatment. New strategies need to be developed in order to enhance the survival of AML patients. Interferon-α (IFN-α), a type I IFN, is a known as an anti-tumor agent, which used to clinical treat AML because of its anti-proliferative effect and immune response. However, the clinical trials show divergent results because of the instability and delivery difficulty. USP18 is a negative regulator of Type I IFN signaling pathway. Knockout Usp18 can enhance and prolong the Type I IFN pathway.

Here, we use AE9a-included leukemia mouse models that reflect the genetics and pathology of human acute myeloid leukemia to study the effect of Usp18 loss in leukemia development and to discover possible new therapeutic targets for leukemia. Results show that deletion of Usp18 delayed leukemia propagation in both the AE9a-induced and MLL-AF9-induced leukemia models. Usp18-deletion-mediated inhibition of myeloid leukemia propagation is IFN response dependent. More important, Usp18 deletion can delay the myeloid leukemia remission after the chemotherapy. These data indicate that pharmacological inhibition of Usp18 may be benefit in AML treatment with interferon prescribed.

Post-transcriptional regulation by RNA-binding proteins (RBPs) is an important layer of gene regulation implicated in both healthy hematopoiesis and hematologic malignancy. Among post-transcriptional mechanisms, alternative polyadenylation (APA) regulates gene expression and function, mediating normal cellular differentiation and malignant transformation across cellular systems. In hematology, APA plays a critical role in lymphocyte maturation and dysregulation contributes to multiple myeloma and lymphocytic leukemia. Despite its documented importance in immune cells, it is unknown whether APA plays a critical role in myeloid malignancy or in healthy hematopoietic stem cell (HSC) maintenance. Furthermore, RBPs that regulate APA in hematologic systems have not been identified.

Here, we first addressed the prevalence and global function of APA in myeloid malignancy. We compared poly(A) site usage in acute myeloid leukemia (AML) blasts to usage in healthy hematopoietic stem and progenitor cells (HSPCs), uncovering global patterns and individual leukemia-promoting genes altered in malignancy. By targeting the RBP and APA regulator FIP1L1, we reversed the global trends in patients and observed cellular differentiation across diverse AML subtypes by disrupting leukemogenic signaling networks. In t(8;21) AML, we validated APA regulation of AML1-ETO, showing for the first time that expression of a prominent oncofusion is sensitive to this mode of post-transcriptional regulation. Altogether, our work defines a critical role for APA in AML and illuminates a new pathway that may be exploited for differentiation therapy in patients.

We also studied the role of APA in healthy HSC pool maintenance by focusing on APA regulation of the critical hematopoietic transcription factor RUNX1. Polyadenylation upstream of the most distal 3'UTR produces a C-terminally truncated protein that antagonizes the pro-differentiation function of full-length RUNX1 in HSCs. We modeled this relevant APA event using a dual fluorescent minigene reporter and used this reporter in a CRISPR screen targeting RBPs. We identified HNRNPA1 and KHDRBS1 as regulators of RUNX1 APA, assigning a new role to these RBPs in HSC fate. Overall, our work highlights the intersection between post-transcriptional regulation and transcription factor function in healthy hematopoiesis.

Myelodysplastic Syndromes (MDS) are hematopoietic disorders characterized by myeloid dysplasia, inefficient hematopoiesis, and the propensity to transform into acute myeloid leukemia (AML). Recent advances in sequencing technologies uncovered various types of mutations associated with the disease. In particular, mutations of the splicing factor SRSF2 were significantly associated with mutations of RUNX1, a transcription factor with an established role in hematological disorders. These findings suggest a critical role of these factors in hematopoiesis and disease pathogenesis through their cooperative effects in regulating gene expression and alternative splicing. First, we characterized the roles of SRSF2 mutations in blood cell development and the splicing alternations caused by these mutations. Second, we uncovered that, besides their canonical roles in splicing, SRSF2 mutations also trigger excessive R-loop formation, leading to replication stress and DNA damage response. Third, we discovered the convergent effect of SRSF2 mutations and RUNX1 mutations in facilitating the development of MDS phenotypes, altering global splicing and gene expression programs, which elucidated the pathways critical for disease development. These findings laid a critical foundation to understand the contribution of individual mutations to MDS and the synergy effects among them and provided insights to develop novel therapeutic strategies in the future.

Tumor suppressor RASSF2 (Ras association domain family member 2) is a key regulator of the mammalian Hippo pathway, which controls the cell cycle by balancing cell proliferation and apoptosis. RASSF2 expression is often suppressed in human cancers at both the mRNA and protein levels, however little is known about its regulation. In the present study, we found that RASSF2 protein is rapidly degraded by ubiquitin-mediated proteasomal degradation. Interestingly, RASSF2 protein with a deletion of the SARAH domain (Sav, RASSF2, Hpo) is more stable than the wildtype RASSF2, suggesting that the SARAH domain may be a target for negative regulation. Therefore, we hypothesized that one or more ubiquitin E3 ligase(s) specifically target the SARAH domain of RASSF2 to promote its degradation. We employed a shRNA-mediated approach to knockdown E3 ligases that are potentially specific to the SARAH domain of RASSF2 and monitored the changes in RASSF2 protein stability. As a result, we have identified E3 ligase ITCH as a negative regulator of RASSF2 protein stability. Knockdown of ITCH in the Kasumi-1 cell line which normally harbors low RASSF2 expression, successfully increased RASSF2 protein levels. We further show that ITCH overexpression negatively modulates the stability of both RASSF2 and an additional Hippo pathway member, MST1, in a manner dependent on the E3-ligase enzymatic activity. Our results suggest that small molecule inhibitors targeting ITCH enzyme activity may provide a novel approach for cancer therapy through the ability to restore expression of RASSF2 and other Hippo signaling pathway proteins to normal levels in cancer cells.

RASSF2 is a powerful pro-apoptotic K-Ras effector that is that is inactivated in many tumors via promoter methylation and has been shown to function as a tumor suppressor in lung, colorectal, and breast cell lines. RASSF2 belongs to the Ras-association domain family (RASSF) of proteins, which are able to engage in homo/hetero-dimerization and interact with common binding partners. In the context of acute myelogenous leukemia (AML), RASSF2 is exclusively downregulated in t(8;21) AML, suggesting that its repression may be essential for t(8;21) leukemia development.

In order to further characterize RASSF2’s tumor suppressive role in a leukemic context, we performed a CRISPR/Cas9-mediated knockout of RASSF2 in two non-t(8;21) AML cell lines: HL-60 and U937, generating single cell isolated clonal lines that are RASSF2 wild type, heterozygous, and knockout. Among the clonal cell lines, we assayed for changes in proliferation, apoptosis, cell cycle, and differentiation. We observed that heterozygous knockdown of RASSF2 in the U937 cells resulted in significantly higher (p<.05) proliferation when compared to wild type U937 clonal cells. However, mono- and bi-allelic knockout of RASSF2 did not result in significant differences in differentiation, apoptosis, or cell cycle arrest. In fact, clonal lines of the same genotype were observed to be characteristically variable. Taken together, we demonstrate that RASSF2 knockout is not critical in acute myelogenous leukemia since it does not manifest in a significant functional phenotype that overcame clonal characteristic variation. This suggests that loss of RASSF2 expression leads to the rescue of its function. We also prove that the process of single cell clonal isolation selects for clonal lines that are inherently variable in differentiation, which demonstrates that functional characterizations involving single cell clonal isolation must be performed with a larger pool of samples to account for the characteristic variability of clonal cell lines.

Chromosomal abnormalities are frequent events in human cancer and are especially prevalent in hematological malignancies. Here, through the characterization of chromosomal abnormalities del(20q) and t(8;21) we reveal unexpected non-canonical functions of the mammalian Hippo kinase in hematologic cancer.

Somatic heterozygous deletions on chromosome 20q are detected in several hematopoietic malignancies, including myelodysplastic syndrome (MDS), classical myeloproliferative neoplasm (MPN), MDS/MPN overlap disorders, and acute leukemias. To date, identification of genes in the del(20q) common deleted region that contribute to disease development have remained elusive. Through assessment of patient gene expression we have identified STK4 (encoding Hippo kinase MST1) as a 20q gene that is downregulated below haploinsufficient amounts in MDS and MPN. Functional modeling of hematopoietic-specific gene inactivation in mice revealed Hippo kinase loss to cause phenotypes that closely resemble those observed in patients harboring del(20q), and to cooperate with JAK2-V617F mutation in promoting adverse MPN disease progression to myelofibrosis. Mechanistic studies revealed that myelofibrotic transformation in mice was associated with cooperative effects of JAK2-V617F and Hippo kinase inactivation on IRAK1-dependent innate immune-associated proinflammatory cytokine production, including IL-1β and IL-6. In summary, we find Hippo kinase MST1 (STK4) as having a central role in the biology of del(20q)-associated hematologic malignancies.

Through biochemical and functional studies based on the characterization of t(8;21) in acute myeloid leukemia, we also define a novel Hippo-RASSF2 signaling pathway that regulates basal Rac GTPase activity in hematological cancer and can be exploited for therapy in patients. This non-canonical signaling mechanism is independent of Hippo kinase activity, and is instead mediated by a SARAH domain-dependent interaction. Using proximity-based biotin labeling, we associate RASSF2 with Rho GTPase-related complexes and identify a direct interaction with the critical Rac guanine nucleotide exchange factor (GEF), DOCK2. RASSF2 promotes DOCK2 GEF activity in vitro; and RASSF2 knockdown is sufficient to functionally abolish GTP-bound Rac and promote growth-arrest in leukemia cells. Importantly, RASSF2 expression is broadly correlated with leukemia cell sensitivity to small-molecule inhibition of DOCK2 GEF activity, revealing novel mechanistic insight and providing a functional biomarker for sensitivity to perturbation of this pathway in hematological cancer.