AUTHORS
Alecia Alto, BS, Tiff Ricketts, PhD2, Na Xie, PhD3, Tyler Freeland, B.S.4, Mason Harris5, Gracie Jack6, Robert L Bowman, PhD7, Andrew Paterson, PhD8, Ravi Bhatia, MD9
AFFILIATIONS
1 The University of Alabama at Birmingham, Birmingham, Alabama, United States, The University of Alabama at Birmingham, Birmingham, Alabama, United States
2 The University of Alabama at Birmingham, Birmingham, Alabama, United States
3 The University of Alabama at Birmingham, Birmingham, Alabama, United States
4 The University of Alabama at Birmingham, Birmingham, Alabama, United States
5 The University of Alabama at Birmingham, Birmingham, Alabama, United States
6 The University of Alabama at Birmingham, Birmingham, Alabama, United States
7 University of Pennsylvania, Philadelphia, Alabama, United States
8 The University of Alabama at Birmingham, Birmingham, Alabama, United States
9 The University of Alabama at Birmingham, Birmingham, Alabama, United States
ABSTRACT
Hematopoietic stem cells (HSCs) localize to mesenchymal and vascular bone marrow (BM) niches that regulate their survival, self-renewal, and differentiation. BM niches also protect quiescent chronic myeloid leukemia (CML) leukemic stem cells (LSCs) from eradication following tyrosine kinase inhibition (TKI) treatment. Growing evidence suggests that inflammation-mediated alterations in BM endothelial cells (BMECs) promote disease progression and resistance to therapy in myeloproliferative neoplasms and MDS/AML. However, the vascular niche and its role in controlling LSC persistence is poorly characterized in CML. Analysis of CML patient BM obtained at the time of TKI cessation showed that patients who relapse exhibit elevated levels of chemokines CXCL8 and CXCL1, ligands for the chemokine receptor CXCR2. Our previous studies have shown a role for CXCL1-CXCR2 signaling in CML LSC expansion and TKI resistance (Agarwal et al. Cell Reports. 2021;6(2):109386). Although BMECs also express CXCR2, the role of CXCR2 signaling in altered angiogenesis and hematopoiesis has not been well studied. Therefore, our objective is to determine how endothelial CXCR2 signaling affects BMEC function in CML and regulates LSC survival.
To study CML BM vasculature in vivo, we utilize a tetracycline-regulated SCL-tTA/BCR-ABL genetic mouse model that replicates features of chronic-phase CML. We have shown that CML induction leads to drastic remodeling of the BM microvasculature (Agarwal et al. Cell Stem Cell. 2019;24(5):769-784.e6). RNA-sequencing revealed that CML-derived CD45-Ter119-CD31+ BMECs overexpressed CXCR2 and exhibited gene signatures associated with impaired endothelial function compared to WT mouse BMECs. Furthermore, an in vivo Evans Blue vascular permeability assay demonstrated increased BM vascular leakiness in CML mice compared to WT. To assess the importance of CXCR2 signaling in regulating LSC fate, we treated mice engrafted with CML cells with either TKI or vehicle for 4 weeks, supplementing half of each treatment group with CXCR2 inhibitor (CXCR2i) or DMSO control treatment. Administration of CXCR2i in addition to TKI led to improved depletion of LSCs compared to TKI alone. To assess endothelial-specific contributions to CXCR2 signaling, we generated Cxcr2fl/fl Tek-Cre KO mice. Cre+ EC-KO versus Cre- control mice had reduced counts of peripheral blood CD45+ immune cells and BM HSCs. There was no difference between progenitor and mature immune cells of the BM or spleen. A similar proportion of CD45-Ter119-CD31+ BMECs was seen in Cre+ and Cre- mice, with a modest decrease in BMEC counts seen in Tek-Cre KO mice. We are currently conducting CML BM transplants in Cxcr2 KO and control Tek-Cre mice to evaluate changes to BM vasculature, disease progression, and response to TKI treatment.
To study BMEC CXCR2 function in vitro, we used a well characterized Akt-immortalized murine BMEC line. BMECs stably express CXCR2 by flow cytometry, and produce CXCR2 ligands CXCL2 and CXCL1 at baseline. Inhibition of CXCR2 altered BMEC morphology from a spindle to a cobblestone shape. CXCR2i also reduced average BMEC counts but did not reduce viability assessed by Annexin V staining. Cell cycle analysis using Ki-67 and DNA labeling by flow cytometry showed that a greater proportion of CXCR2i-treated BMECs were in stage G0/G1 (70%) compared to DMSO control (40%). This decrease in cell cycling was validated by immunofluorescent staining, in which inhibitor-treated BMECs expressed fewer Ki-67+ cells and half as many Ki-67hi cells compared to DMSO. Assessment of BMEC function using tube formation assays showed that CXCR2i decreased average vessel length, vessel area, and number of junctions, with an increase in mean lacunarity, a measure of empty space between vessels. Therefore, CXCR2i disrupts BMEC proliferation, cell cycling, and network complexity.
Overall, our results demonstrate that CML mouse BM vasculature displays endothelial dysfunction, evidenced by RNA sequencing and functional vascular permeability studies. We also demonstrate an important role for CXCR2 signaling in regulating BMEC proliferation and network formation. Ongoing studies are investigating the role of CXCR2 in mediating vascular remodeling and regulating leukemic burden and treatment response using endothelial-specific Cxcr2 KO mice. These results are expected to inform novel clinical strategies for improving TKI efficacy by targeting endothelial perturbations in CML.
DISCLOSURES
Alecia Alto, BS: No relevant conflicts of interest to declare.