Abstract

Pre-B acute lymphoblastic leukemia (ALL) is one of the leading causes of cancer-related death, largely due to the high risk of relapse. Relapse often originates in the bone marrow microenvironment where stroma protects ALL cells against drug treatment. Galectin-3 (Gal3) is a carbohydrate-binding protein which recognizes cell surface receptors by binding LacNAc branches of N-glycans. We recently found that the stromal cells of the tumor microenvironment are the main source of Gal3 in ALL cells. In co-culture of ALL cells with supportive stroma (OP9), stroma-secreted Gal3 is actively internalized by ALL cells. Expression of Gal3 in ALL cells leads to significant protection of these cells against drug-induced apoptosis. The application and therapeutic benefits of targeting Gal3 and the lectin-mediated communication between the tumor microenvironment and leukemic cells in ALL treatment have not been explored. Here, we tested the hypothesis that Gal3 conveys important signals between the tumor microenvironment and ALL cells and that it represents a viable therapeutic target.

We tested the preclinical activities of a first in class Gal3 inhibitor "KB1019.7" by studying its antileukemic effects on ALL cells in vitro. KB1019.7 is a novel carbohydrate-based compound with a micro Molar binding affinity for Gal3 as well as significant selectivity for Gal3 over Gal1. KB1019.7 acts by interfering with carbohydrate recognition activities of Gal3.

Treatment of different ALL cells with KB1019.7 resulted in a dose-dependent reduction in cell viability. ALL cells which were cultured on a protective stroma (OP9) layer were plated alone or together with OP9 cells prior to drug treatment. The IC50 values (concentration of the inhibitor at which 50% cell death occurred) did not vary significantly among different ALL subgroups and were within mid micromolar range (average ~470 μM). The high tolerance of the cells to KB1019.7 may be explained by the fact that this compound is an optimized monovalent carbohydrate and not a classic Galectin inhibitor which are usually in the form of peptide or carbohydrate ligands. Gal3 is involved in several cellular activities including cell growth, cell differentiation and apoptosis. We evaluated the effects of KB1019.7 on modulating the biological functions of Gal3 in US7 (Ph-negative) and TXL2 (Ph-positive) B-cell leukemic cells. To this end, analysis of ALL cells exposed to 250 and 500 μM of KB1019.7 revealed a dose-dependent increase in the number of cells with 4N DNA content after 48 hours, suggesting that Gal3 inhibition may lead to accumulation of ALL cells in G2/M phase. More importantly, KB1019.7 had adverse effect on the acquired resistance to Vincristine in ALL cells. We treated US7 cells with vehicle control, KB1019.7 (500 μM), Vincristine (2.5 nM) or a combination of KB1019.7 and Vincristine in the presence of protective stroma (OP9) cells. Vincristine was chosen as the model for standard cytotoxic therapy in pre-B ALL. US7 cells developed resistance to Vincristine after continued exposure (~12 days) as evident by recovery and increase in cell viability observed after a time-dependent decline in cell number. Interestingly, introduction of KB1019.7 in combination with Vincristine effectively inhibited resistance to Vincristine and led to efficient cell death. Furthermore, KB1019.7 treatment promoted a significant downregulation in pERK1/2 protein levels which is often upregulated upon emergence of environmentally mediated drug resistance in ALL cells.

Together our data demonstrate a role for Gal3 inhibition with KB1019.7 in inducing antileukemic responses in ALL cells including induction of cell death and cell cycle arrest. Furthermore, our results highlight the value of combining Gal3 inhibition with conventional chemotherapy as a tool to efficiently circumvent resistance to such therapies. Gal3 pro-tumor activities are documented in other hematologic malignancies and further studies to initiate and expand Gal3-targeted therapies in leukemia are highly warranted.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.