Year of Award

2007

Document Type

Dissertation

Degree Type

Doctor of Philosophy (PhD)

Degree Name

Toxicology

Department or School/College

Department of Biomedical and Pharmaceutical Sciences

Committee Chair

Howard D. Beall

Commitee Members

Mark Pershouse, Fernando Cardozo-Pelaez, Christopher Esslinger, Kent D. Sugden

Keywords

lavendamycin, molecular modeling, NQO1, oxidative DNA damage, oxidative stress, structure-based design

Abstract

NAD(P)H:quinone oxidoreductase 1 (NQO1) is a two-electron reductase that catalyzes an NAD(P)H-dependent activation of many quinone-based antitumor agents. NQO1, expressed at high levels in many human solid tumors, can be used as a target for enzyme-directed bioreductive antitumor drug development. We hypothesized that lavendamycins, quinolinedione antitumor antibiotics, can be activated by NQO1 in cancer cells that overexpress NQO1 to exhibit selective toxicity toward those cells. The effects of functional group changes on the metabolism of lavendamycins by recombinant human NQO1 were studied using a spectrophotometric assay. These structure-activity relationship (SAR) studies determined key structural features that were required for lavendamycin substrate specificity. Cytotoxicity toward human colon adenocarcinoma NQO1-deficient (BE) and NQO1-rich (BE-NQ) cells was also determined using colorimetric and clonogenic assays. The best lavendamycin substrates for NQO1 were also the most selectively toxic to the BE-NQ cells compared to BE cells. To facilitate structure-based design of more optimal lavendamycin substrates and NQO1-directed lavendamycin antitumor agent development, we developed a 1H69 crystal structure-based in silico model of the NQO1 active site and performed lavendamycin-docking studies. The docking was performed using the FlexX module of SYBYL software. Lavendamycin analogues were designed as NQO1 substrates utilizing our SAR and docking data as structure-based design criteria. Docking and biological studies on the analogues were performed and were consistent suggesting the in silico model of the enzyme possessed practical predictive power. Our results also suggested practicality of the design criteria resulting in the discovery of good NQO1 substrates with selective toxicity toward BE-NQ cells. The mechanisms of NQO1-mediated selective cytotoxicity of good lavendamycin substrates in BE and BE-NQ cells were also investigated including induction of oxidative stress and apoptosis. Biomarkers of oxidative stress including formation of 8-hydroxy-2'-deoxyguanosine (8-oxo-2dG), an indicator of oxidative DNA damage, and depletion of reduced glutathione (GSH) were examined using an HPLC-based method and a colorimetric assay, respectively. Induction of apoptosis was examined using a colorimetric assay. Our results revealed that oxidative stress and subsequent apoptosis induction by a good lavendamycin substrate was NQO1 dependent and that the poor substrate for NQO1 caused neither oxidative stress nor apoptosis.

Share

COinS
 

© Copyright 2007 Mary Hassani