Investigation of platinum drug mode of action and resistance using yeast and human neuroblastoma cells as model systems

The platinum-based cytotoxic chemotherapeutics cisplatin, carboplatin, and oxaliplatin are widely used as anti-cancer drugs, but their efficacy is limited by the occurrence of resistance. The mechanisms of actions of these drugs as well as the mechanisms underlying resistance (formation) to platinum drugs remain incompletely understood. More knowledge is required to develop more effective therapies.

In this study, we used yeast cells and neuroblastoma cell lines to investigate the mode of action and resistance to platinum drugs. First, we screened a Saccharomyces cerevisiae transcription factor heterozygous gene deletion library to identify genes that when deleted result in enhanced sensitivity to cisplatin, carboplatin and/or oxaliplatin. In addition, data on platinum drug sensitivity in yeast, derived from screening a yeast whole-genome homozygous deletion library, was extracted from the Yeast Fitness Database (http://fitdb.stanford.edu/). There was a substantial overlap in the genes (and related pathways) that determine sensitivity to the individual platinum drugs, but also considerable differences. Notably, cisplatin and carboplatin are anticipated to be more similar in their mode of action compared to oxaliplatin, but the yeast data did not entirely support this notion. Amoung the genes involved in response to platinum drugs, BDF1 (Bromodomain-containing factor 1) was identified as a novel gene which, when deleted, resulted in increased sensitivity to all three platinum drugs. Its re-expression reversed platinum drug sensitivity, confirming its role in determining the yeast cell response to platinum drugs. Notably, BET proteins (the human equivalents of Bdf1) are increasingly recognised as potential anti-cancer drug targets. Our data suggest that they may have a role in sensitising cancer cells to platinum drugs.

Next, we investigated a unique panel of cell lines consisting of neuroblastoma cell lines UKF-NB-3 and UKF-NB-6 and their sub-lines with acquired resistance to cisplatin (UKF-NB-3rCDDP1000, UKF-NB-6rCDDP2000), oxaliplatin (UKF-NB-3rOXALI2000, UKF-NB6rOXALI4000), or carboplatin (UKF-NB3rCARBO2000, UKF-NB-6rCARBO2000). Adaption to platinum drugs was associated with changes in doubling times and cell morphology but there were no consistent patterns. This suggests that resistance mechanisms are complex and heterogeneous. The resistance phenotype was stable after cultivation of the resistant sub-lines for three months in the absence of drugs, indicating that resistance was not a consequence of the reversible enrichment of a pre-existing sub-population of cells, but due to a permanent, irreversible genomic change. This notion was supported by the determination of sensitivity profiles to cisplatin, carboplatin and oxaliplatin in the cell line panel. Both UKF-NB-3 and UKF-NB-6 parental lines exhibited sensitivity to all three platinum drugs, and the platinum drug-adapted parental cells displayed generally increased resistance, not just to the drug to which they were adapted, but to all three platinum drugs. However cisplatin- and carboplatin- resistant UKF-NB-3 cells displayed no cross-resistance to oxaliplatin, and oxaliplatin-resistant UKF-NB-3 cells displayed none to cisplatin and carboplatin. In contrast to the yeast data, this supports the notion that cisplatin and carboplatin are more similar in their mode of action than oxaliplatin.

Finally, in a proteomics study, we compared the UKF-NB-3 and UKF-NB-6 cells with their cisplatin-resistant sub-lines to study acquired cisplatin resistance, and also investigated their responses to acute cisplatin treatment. The resulting data, together with previous proteomics studies that investigated acquired resistance in cisplatin-adapted neuroblastoma cell lines, suggested that, despite overlaps, the resistance mechanisms are heterogeneic and cell line-specific. This was also the case, comparing our cell lines only, for the acute cisplatin responses.

In conclusion our data demonstrate that resistance formation to cisplatin is a complex and individual/cell line specific process. Further research will be required to enable a systems-level understanding of cisplatin resistance that can be translated into improved therapeutic approaches.