Use of NQO1 status as a selective biomarker for oesophageal squamous cell carcinomas with greater sensitivity to 17-AAG.

BACKGROUND: Oesophageal squamous cell carcinoma (OSCC) is a major health burden in Sub-Saharan Africa, and novel chemotherapies are urgently required to combat this disease. The heat shock protein 90 (HSP90) inhibitor 17-N-allylamino-17-demethoxygeldanamycin (17-AAG) has previously been proposed as a possible candidate drug. NADPH quinone oxidoreductase 1 (NQO1) is known to increase the potency of 17-AAG, therefore we investigated the effects of 17-AAG in OSCC cell lines in the context of their NQO1 status. METHODS: We used MTT assays to compare the sensitivity of a panel of OSCC cell lines to 17-AAG. Western blotting, and RT-PCR were used to investigate NQO1 protein and mRNA levels, while an RFLP approach was used to investigate the NQO1 C609T SNP. RESULTS: Expression of NQO1 markedly increased sensitivity to 17-AAG in the OSCC cell lines, while normal fibroblasts, which expressed HSP90 at much lower levels, were more resistant to 17-AAG. In isolation, neither the C609T SNP, nor NQO1 mRNA levels was an accurate predictor of NQO1 protein levels. CONCLUSIONS: Since NQO1 greatly enhances the anti-cancer effects of 17-AAG, this could be used as a selective marker for patients that would benefit most from 17-AAG chemotherapy at low doses. Testing for the presence of the C609T SNP in both alleles could be used as a screen to exclude potentially poor responders to 17-AAG treatment at low dosages.

Differential nuclear localisation and promoter occupancy play a role in glucocorticoid receptor ligand-specific transcriptional responses.

The glucocorticoid receptor (GR) is a ligand-activated transcription factor for which a number of endogenous and synthetic ligands exist. A key question in steroid receptor biology is how different ligands elicit different maximal transcriptional responses via the same receptor and on the same promoter. This question was addressed quantitatively for the GR, using a panel of agonists, partial agonists and antagonists, on the endogenous GILZ gene in two different human cell lines. It was found that the extent of GR nuclear localization correlated with the efficacy for GILZ transactivation by the GR in U2OS cells. However, in A549 cells there was no significant correlation, with all ligands resulting in similar levels of GR nuclear localization, despite different levels of transcriptional activation of the GILZ gene. Chromatin immunoprecipitation analysis on the other hand, revealed ligand-specific differences in GILZ promoter occupancy in the A549 cells, which correlated with the transcriptional efficacy of the subset of ligands investigated. This suggests that ligand-specific differences in promoter occupancy by activated GR play a major role in discrimination between agonist, partial agonist and antagonist responses on the endogenous GILZ gene in A549 cells, while differences in nuclear localisation of liganded GR play a role in determining the transcriptional outcome in U2OS cells. These cell line-specific differences were not dependent on the amount of GR present, since transient overexpression of GR in U2OS did not alter the relative ligand-selective nuclear localisation. Our results show that there is a relationship between ligand-specific transactivation efficacy, extent of nuclear translocation and recruitment of GR to the promoter. However, the relative contribution of nuclear translocation and GR promoter recruitment to ligand-specific transactivation efficacy is cell-specific.