Vascular Disrupting Combretastatins Impair Bone Marrow Endothelial Cells by Depolymerizing the Microtubule Cytoskeleton.

The vascular disrupting combretastatins CA1P (OXi4503) and CA4P (Zybrestat) are known to target solid tumor neovasculature by altering endothelial cell morphology and occluding cancer neovessels. In prior work, we and others have demonstrated that combretastatins target acute myeloid leukemia (AML) cells by generating reactive oxygen species within the malignant myeloid cells. We are currently conducting a phase 1 clinical trial of OXi4503 in patients with AML (NCT01085656). However, the vascular targeting mechanisms by which combretastatins regress AML are ill defined. We hypothesized that combretastatins also target bone marrow endothelial cells (BMECs).

BMECs were isolated and cultured from bone marrow of healthy subjects and patients with AML. The cells were treated with various doses of the combretastatins CA1 and CA4 for 24 and 48 hours. Cell viability was measured by XTT assay. BMEC function in response to combretastatin treatments was quantified using a modified in vitro scratch assay and 2D capillary formation using Matrigel. Analysis of microtubule cytoskeleton was performed by staining treated BMECs with anti-β-tubulin antibodies and fluorescence microscopy. To ascertain the effects of combretastatin treatment on the ability of BMECs to support and protect AML cells, co-culture experiments were performed.

The combretastatins CA1 and CA4 did not decrease BMEC viability or metabolic activity. However, scratch assay showed impaired BMEC migration, with only 15.6% and 7.5% of the wound covered by BMECs after CA1 and CA4 respectively, compared to untreated control (80%) (p=0.0173). Matrigel assay also showed impaired BMEC function after combretastatin treatment, with complete lack of capillary tube formation compared to untreated BMECs which generated an average of 27.3 capillary tubes per 15 BMECs (p<0.001). Immunohistochemistry showed a distinct degradation of microtubule cytoskeleton in BMECs treated with CA1 and CA4. Finally, when treated with a vascular targeting combretastatin, AML cells co-cultured with BMECs showed a lower survival rate than AML cells alone (64.1% vs. 75.3%, p = 0.0267).

Combretastatins directly impair BMEC function and eliminate BMEC protection of AML cells. The dual targeting of combretastatins on BMECs and AML cells leads to enhanced AML regression. These data highlight the role of the leukemia microenvironment as a protective reservoir of disease and provide methods of tracking biologic activity of combretastatins in early phase clinical studies.

Cogle:OXiGENE: Research Funding; Leukemia & Lymphoma Society: Membership on an entity's Board of Directors or advisory committees, Research Funding.