Induction of Chronic Lymphocytic Leukemia (CLL) Apoptosis by Nitric Oxide Synthase (NOS) Inhibitors: Drug Efficacy Correlates with Lipid Solubility and NOS1 Dissociation Constant.

Abstract

The viability of CLL cells may be dependent on the autocrine production of nitric oxide because nitric oxide synthase (NOS) inhibitors induce CLL cell apoptosis and CLL cells express inducible NOS (NOS2). Our previous study indicated that the non-specific NOS inhibitor NMMA induced CLL cell apoptosis but only at high concentrations (> 1 mM) (Levesque et al., Leukemia 17:442, 2003). Therefore, we performed the current study to identify NOS inhibitors that induce CLL cell apoptosis at lower concentrations and to understand factors that promote NOS inhibitor-induced CLL cell toxicity. We isolated and enriched CLL cells from the blood of CLL patients and cultured the CLL cells in media containing various concentrations of 21 different NOS inhibitors. We determined CLL cell viability following culture with each NOS inhibitor. We found that NOS inhibitors with specificity for neuronal NOS (NOS1) induced CLL cell death at concentrations lower than non-specific NOS inhibitors and lower than inducible NOS (NOS2) specific inhibitors. There was a weak correlation (r2 = 0.29, p = 0.1608) of the NOS1 (but not NOS2) half-maximal inhibitory concentration (IC50) of each NOS inhibitor for purified recombinant NOS and its ability to induce CLL cell death. We confirmed the specificity of the NOS inhibitors by inhibition of purified recombinant NOS1 and NOS2 enzyme activity, and we confirmed that NOS1 specific inhibitors induced CLL cell death by apoptosis. Because there was only a weak correlation of the NOS1 IC50 with NOS inhibitor induced CLL cell death, we considered whether other factors such as the Kd and hydrophobicity of each compound correlated with CLL cell death. We found that there was a direct correlation between the NOS1 (but not NOS2) dissociation constant (Kd) of NOS inhibitors and CLL cell death (r2 = 0.77, p = 0.0041) and a direct correlation of the partitioning coefficient (a measure of hydrophobicity) of each NOS inhibitor and its ability to induce CLL cell death (r2 = 0.68, p < 0.0001). Therefore, NOS inhibitors that bound tightly to NOS1 and were hydrophobic induced CLL cell death at lower concentrations. There was variable expression of CLL cell NOS1 mRNA (6 of 28 samples positive) and we were unable to demonstrate CLL cell expression of NOS1 protein by immunoblotting. This suggests that if NOS1 is present in CLL cells, it exists at very low levels. Taken together, we believe that low level NO production promotes CLL cell viability and that inhibition of CLL NOS induces CLL cell apoptosis. Importantly, our studies provide direction for the rational design and selection of NOS inhibitors that may be useful as CLL therapeutics.