High Mobility Group A1 (HMGA1) Chromatin Remodeling Protein Mediates Crosstalk Between Acute Myeloid Leukemia Blasts & the Tumor Microenvironment

Introduction: Acute myeloid leukemias (AMLs) are highly lethal hematologic malignancies that arise from diverse genetic abnormalities in hematopoietic progenitor cells. Unfortunately, most patients with AML will die of their disease due to failure of currently available cytotoxic chemotherapies. Accordingly, there is an urgent need to better understand the biologic interactions between AML blasts and signals from bone marrow stromal cells (BMSCs) within the niche that support their survival and protect them from exposure to chemotherapy. The high mobility group A1 (HMGA1) chromatin remodeling proteins are present at high levels during development and enriched in AML blasts, leukemic stem cells, diverse solid tumors, embryonic stem cells, and adult stem cells, such as hematopoietic stem cells (HSCs). HMGA1 proteins regulate gene expression by modulating chromatin structure and recruiting NF-κB and other transcription factor complexes to DNA. We discovered that HMGA1 acts as a potent oncogene in transgenic mouse and cultured cell models (Xu et. al, Cancer Research, 2004) by inducing stem cell transcriptional networks (Schuldenfrei et. al, BMC Genomics, 2011, Shah et. al, PLoS ONE, 2013). Here, we describe a novel role for HMGA1 in mediating crosstalk between AML blasts and BMSCs within the bone marrow microenvironment.

Methods & Results: To investigate HMGA1 in regulating AML-niche signaling, we used potent lentiviruses to deliver short hairpin RNA and silence HMGA1 in AML blasts or BMSCs. We found that silencing HMGA1 rapidly halts proliferation and induces apoptotic cell death in 3 different AML cell lines. To determine how HMGA1 mediates survival in AML blasts, we assessed expression of pro-survival genes in AML cell lines and primary AML blasts, including those encoding NF-E2-related factor 2 (NRF2), cMYC, and the C-X-C chemokine receptor type 4 (CXCR4). CXCR4 is the receptor for stromal cell-derived factor 1 (SDF-1 or CXCL12), a growth factor secreted by BMSCs that also serves as a chemo-attractant for AML blasts or HSCs within the bone marrow microenvironment. We found that silencing HMGA1 represses expression of NRF2, cMYC, and CXCR4. This led us to hypothesize that HMGA1 regulates crosstalk between AML blasts and the leukemic cell niche via CXCR4 and SDF-1. To test this, we silenced HMGA1 in cultured AML cells and assessed migration in the presence of SDF-1. Strikingly, migration was significantly impaired in the AML cells with HMGA1 knock-down. Next, we silenced HMGA1 in primary, patient-derived BMSCs, which were co-cultured with primary AML blasts from the same patients to mimic the bone marrow microenvironment. We found that knock-down of HMGA1 in BMSCs also results in apoptosis in primary AML blasts.

Conclusions: Together, our results demonstrate for the first time that HMGA1 mediates AML survival through cell-autonomous pathways in AML blasts and through non-cell-autonomous crosstalk from BMSCs within the bone marrow microenvironment. Studies are underway to determine if HMGA1 directly regulates expression of SDF-1 or other factors secreted by BMSCs within the hematopoietic niche. This knowledge should inform biologically rational strategies to enhance existing treatments and facilitate the design of novel therapies.