Derivation and Characterization of Myeloproliferative Disorder Stem Cells from Human ES Cells

Summary of Overall Progress This grant focuses on generation of MPN stem cells from hESC or CB and correlates leukemic potential with MPN patient samples. In the first year of this grant, we have demonstrated that 1) hESC differentiate on AGM stroma to the CD34+ stage, which is associated with increased GATA-1, Flk2, GATA-2 and ADAR1 expression; 2) hESC CD34+ differentiation is enhanced in vitro and in vivo in the presence of a genetically engineered mouse stroma, which produces human stem cell factor, IL-3 and G-CSF; 3) hESC CD34+ cells can be transduced with our novel lentiviral BCR-ABL vector, which, unlike retroviral BCR-ABL, can transduce quiescent stem cells; 4) BCR-ABL expression by CP CML progenitors does not sustain engraftment but rather leukemic transformation is predicated, in part, on bcl-2 overexpression; 5) JAK2V617F expression in hES or CB stem cells is insufficient to induce leukemic transformation; 6) BCR-ABL transduced hESC CD34+ cells have significantly higher BCR-ABL transplantation potential than CP CML progenitors suggesting that they have higher survival capacity; 7) lentiviral -catenin transduction of BCR-ABL hESC CD34+ cells leads to serial transplantation indicative of LSC formation; 8) CML BC LSC persist in vivo despite potent BCR-ABL inhibition with dasatinib therapy and will likely require combined inhibitor therapy to eradicate. Currently, HEEBO arrays and phospho-flow studies are underway to detect bcl-2 family members and self-renewal protein expression in BCR-ABL and JAK2 V617F transduced hESC and CB CD34+ cells compared with MPN patient derived progenitors. This will aid in development of combined MPN stem cell inhibitor strategies in this grant.

This grant focuses on generation of myeloproliferative disorder or neoplasm (MPN) stem cells from pluripotent (hESC) or multipotent (CB) stem cells and seeks to correlate their leukemic potential with that of MPN patient sample-derived stem cells. To provide a platform for testing induction of stem cell differentiation, survival and self-renewal by BCR-ABL versus JAK2, hESC were utilized in the first year and as more patient samples and cord blood became available these were utilized. In the first year of this grant, we found that hESC undergo hematopoietic differentiation on AGM stroma to the CD34+ stage resulting in increased GATA-1, Flk2, ADAR1 and GATA-2 expression. Moreover, CD34+ differentiation was enhanced on a genetically engineered mouse stroma (SL/M2) secreting human SCF, IL-3 and G-CSF. Lentiviral BCR-ABL transduced hESC-derived CD34+ cells had higher BCR-ABL+ cellular transplantation potential than chronic phase (CP) CML progenitors, indicative of a higher survival capacity. However, they sustained self-renewal only when co-transduced with lentiviral -catenin (Rusert et al, manuscript in preparation) suggesting that blast crisis evolution requires acquisition of both enhanced survival and self-renewal potential. Similarly, lentiviral mouse mutant JAK2 expression in hESC or CB stem cells was insufficient to produce self-renewing MPN stem cells, indicating that the cellular context, nature of the genetic driver and responses to extrinsic cues from the microenvironment play seminal roles in regulating therapeutically resistant MPN stem cell properties such as aberrant survival, differentiation, self-renewal and dormancy. In the second year of this five year grant, we have focused on human cord blood (CB) stem cells compared with a large number of MPN patient samples propagated on SL/M2 stroma or in RAG2-/-c-/- mice to more adequately recapitulate the human MPN stem cell niche. Also, to more faithfully recapitulate human (rather than the previously published lentiviral mouse JAK2 vectors, Cancer Cell 2008) JAK2 driven MPNs, we cloned human wild-type JAK2 and human JAK2 V617F from MPN patient samples into lentiviral-GFP vectors (Court Recart A*, Geron I* et al, manuscript in preparation). We also incorporated full transcriptome RNA (ABI SOLiD 4.0) sequencing, PCR array and nanofluidic phosphoproteomics technology to better gauge the impact of JAK2 versus BCR-ABL on stem cell fate, survival, self-renewal and dormancy in the context of specific malignant microenvironments and the relative susceptibility of MPN stem cells in these niches to single agent molecularly targeted inhibitors.

This grant focuses on generation of myeloproliferative disorder or neoplasm (MPN) stem cells from pluripotent human embryonic stem cells (hESC) or multipotent cord blood (CB) stem cells, and seeks to correlate their leukemic potential with that of disease progression in MPN patient sample-derived stem cells. In the first and second years of this grant, we found that lentiviral BCR-ABL transduced hESC-derived CD34+ cells had higher leukemic transplantation potential than chronic phase (CP) chronic myeloid leukemia (CML) progenitors. However, they sustained self-renewal only when co-transduced with lentiviral beta-catenin suggesting that blast crisis (BC) evolution requires acquisition of both enhanced survival and self-renewal potential. Similarly, we have shown using lentiviral vectors that mouse and human mutant JAK2 were insufficient to produce self-renewing MPN stem cells. New results in Year 3 demonstrate that BCR-ABL and JAK2 activation drive differentiation of hematopoietic progenitors towards an erthyroid/myeloid lineage bias. We have used full transcriptome RNA-Sequencing (RNA-Seq) technology to evaluate the genetic and epigenetic status of BCR-ABL and JAK2-transduced normal progenitor cells as well as patient-derived MPN progenitors. This has allowed us to probe the mechanisms of aberrant differentiation and self-renewal of MPN progenitors and identify unique gene expression signatures of disease progression. We previously found that overexpression and splice isoform switching of a key RNA editing enzyme – adenosine deaminase acting on dsRNA (ADAR), and splice isoform changes in pro-survival BCL2 family members, correspond with disease progression in CML. In the current reporting period, RNA-Seq analyses revealed that ADAR1-driven activation of RNA editing contributed to malignant progenitor reprogramming, promoting aberrant differentiation and self-renewal of MPN stem cells. Knocking down ADAR1 using lentiviral shRNA vectors reduced the self-renewal potential of CML progenitors. This work has culminated in a manuscript that has now been submitted to PNAS (Jiang et al.). Recent results also show that ADAR1 is activated in progenitors from patients with JAK2-driven MPNs. Thus, ADAR1 may be an important factor that works in concert with BCR-ABL or JAK2 to facilitate disease progression in MPNs. Our results show that another self-renewal factor that may drive BCR-ABL or JAK2-mediated propagation of disease from quiescent MPN progenitors is Sonic hedgehog (Shh). We have examined the expression patterns of this pathway in MPN progenitors using qRT-PCR and RNA-Seq, and have tested a pharmacological inhibitor of this pathway in a robust stromal co-culture model of MPN progression to Acute Myeloid Leukemia (AML). In sum, we have utilized full transcriptome RNA-Seq and qRT-PCR coupled with hematopoietic progenitor assays and in vivo studies to evaluate the impact of JAK2 versus BCR-ABL on stem cell fate, survival, self-renewal and dormancy. These techniques have allowed us to investigate in more detail the role of genetic and epigenetic alterations that drive disease progression in the context of specific malignant microenvironments, and the relative susceptibility of MPN stem cells in these niches to single agent molecularly targeted inhibitors.

The main objectives of this project are generation of myeloproliferative disorder or neoplasm (MPN) stem cells from pluripotent human embryonic stem cells (hESC) or multipotent stem cells, and identification of crucial leukemia stem cell (LSC) survival and self-renewal factors that contribute to the development and progression of BCR-ABL and JAK2-driven hematopoietic disorders. A key finding of our work thus far is that in addition to activation of BCR-ABL or JAK2 oncogenes, generation of self-renewing MPN LSC requires stimulation of other pro-survival and self-renewal factors such as β-catenin, Sonic hedgehog (SHH), BCL2, and in particular the RNA editing enzyme ADAR1, which we identified as a novel regulator of LSC differentiation and self-renewal. We have now completed comprehensive gene expression analyses from next-generation RNA-sequencing studies performed on normal and leukemic human hematopoietic progenitor cells from primary cord blood samples and adult normal peripheral blood samples, along with normal cord blood transduced with BCR-ABL or JAK2 oncogenes, and primary samples from patients with BCR-ABL+ chronic phase and blast crisis chronic myeloid leukemia (CML). These studies revealed that gene expression patterns in survival and self-renewal pathways (SHH, JAK2, ADAR1) clearly distinguish normal and leukemic progenitor cells as well as MPN disease stages. These data provide a vast resource for identification of LSC-specific biomarkers with diagnostic and prognostic clinical applications, as well as providing new potential therapeutic targets to prevent disease progression. New results from RNA-sequencing studies reveal high levels of expression of inflammatory mediators in human blast crisis CML progenitors and in BCR-ABL transduced normal cord blood stem cells. Moreover, expression of the inflammation-responsive form of ADAR1 correlated with generation of an abnormally spliced GSK3β gene product that has been previously linked to LSC self-renewal. These results have now been published in the journal PNAS (Jiang et al.). Together, we have demonstrated that ADAR1 drives hematopoietic cell fate by skewing cell differentiation – a trend which occurs during normal bone marrow aging – and promotes LSC self-renewal through alternative splicing of critical survival and self-renewal factors. Notably, inhibition of ADAR1 through genetic knockdown strategies reduced self-renewal capacity of CML LSC, and may have important applications in treatment of other disorders that transform to acute leukemia. Thus, these results suggest that RNA editing (ADAR1) and splicing represent key therapeutic targets for preventing LSC self-renewal – a primary driver of leukemic progression. Whole transcriptome profiling studies coupled with qRT-PCR, hematopoietic progenitor assays and in vivo studies have shown that combined inhibition of BCR-ABL and JAK2 is another effective method to reduce LSC self-renewal in pre-clinical models. New results show that lentivirus-enforced BCR-ABL or JAK2 expression in normal cord blood stem cells drives generation of distinct splice isoforms of STAT5a. While inhibition of JAK2/STAT5a signaling or BCR-ABL tyrosine kinase activity alone did not eradicate self-renewing LSC, combined JAK2 and BCR-ABL inhibition dramatically impaired LSC survival and self-renewal in the protective bone marrow niche, and increased the lifespan of serial transplant recipients. These effects were associated with reduction in STAT5a isoform expression – which represents a novel molecular marker of response to combined BCR-ABL/JAK2 inhibition – and altered expression of cell cycle genes in human progenitor cells harvested from the bone marrow of transplanted mice. These results are the subject of a new manuscript currently under review (Court et al.). Moreover, this work has led to the development of new experimental tools that will facilitate study of LSC maintenance and cell cycle status in the context of normal versus diseased bone marrow microenvironments. In sum, studies completed thus far have uncovered a role for RNA editing and splicing alterations in leukemic progression, particularly in specific microenvironments. Using specific inhibitors targeting BCR-ABL and JAK2, along with strategies to block RNA editing and aberrant splicing activities, we have been able to establish the relative susceptibility of MPN stem cells to molecular inhibitors with activity against LSC residing in select hematopoietic niches that are difficult to treat with conventional chemotherapeutic agents.

In the final year of this project, we focused on elucidating the mechanisms of leukemia stem cell (LSC) generation in JAK2 compared with BCR-ABL1 initiated myeloproliferative neoplasms (MPN, previously called myeloproliferative disorders). To this end, we investigated the MPN stem cell propagating effects of BCR-ABL1 or JAK2 alone or in combination with activation of the human embryonic stem cell RNA editase, ADAR1. Recently, we discovered that ADAR1, which edits adenosine to inosine bases in the context of primate specific Alu sequences, leads to GSK3β missplicing and β-catenin activation in chronic phase (CP) CML progenitors leading to blast crisis (BC) transformation and LSC generation. In addition, variant isoform expression of a Wnt/β-catenin target gene, CD44, was also characteristic of LSC. In a previous report (Jiang et al., PNAS 2013), identification of ADAR1 as a malignant reprogramming factor represented the first description of RNA editing as a regulator of reprogramming. When lentivirally overexpressed, ADAR1 endows committed CP myeloid progenitors with self-renewal capacity. Further studies revealed that JAK2/STAT5a activates ADAR1 leading to deregulation of cell cycle progression and global down-regulation of microRNA expression thereby uncovering two additional key mechanisms of LSC generation in MPNs. This is consistent with our findings from gene expression profiling studies performed in the previous year, along with functional classification and network analysis using Ingenuity Pathway Analysis (IPA), showing that cell cycle-related genes were significantly altered in human progenitors from xenografted mice treated with combination JAK2 and BCR-ABL inhibitor therapy compared with single agent therapies alone. Together these data suggest that combined BCR-ABL and JAK2 inhibition impairs LSC survival and self-renewal via cell cycle modulation. ADAR1 and other stem cell regulatory pathways such as CD44 represent novel targets to detect and eradicate the self-renewing LSC. We also performed new studies that elucidate the stem cell-intrinsic genetic changes that occur during human bone marrow aging, which may contribute to BCR-ABL or JAK2-dependent functional alterations. This work has led to discovery of a novel role for embryonic stem cell genes and splice isoforms, including ADAR1 p150 and a transcript variant of CD44, in the maintenance of LSC that promote MPN progression. In addition, through the course of this research we have 1) developed novel lentiviral tools for investigating normal hematopoietic stem and progenitor (HSPC) and malignant LSC survival, differentiation, self-renewal, and cell cycle regulation, and 2) devised innovative LSC diagnostic strategies and 3) tested therapeutic strategies targeting LSC-associated RNA editing and splice isoform generation that selectively inhibit LSC self-renewal.