Metabolic responses induced by DNA damage and poly (ADP-ribose) polymerase (PARP) inhibition in MCF-7 cells

Abstract

Genomic instability is one of the hallmarks of cancer. Several chemotherapeutic drugs and radiotherapy induce DNA damage to prevent cancer cell replication. Cells in turn activate different DNA damage response (DDR) pathways to either repair the damage or induce cell death. These DDR pathways also elicit metabolic alterations which can play a significant role in the proper functioning of the cells. The understanding of these metabolic effects resulting from different types of DNA damage and repair mechanisms is currently lacking. In this study, we used NMR metabolomics to identify metabolic pathways which are altered in response to different DNA damaging agents. By comparing the metabolic responses in MCF-7 cells, we identified the activation of poly (ADP-ribose) polymerase (PARP) in methyl methanesulfonate (MMS)-induced DNA damage. PARP activation led to a significant depletion of NAD+. PARP inhibition using veliparib (ABT-888) was able to successfully restore the NAD+ levels in MMS-treated cells. In addition, double strand break induction by MMS and veliparib exhibited similar metabolic responses as zeocin, suggesting an application of metabolomics to classify the types of DNA damage responses. This prediction was validated by studying the metabolic responses elicited by radiation. Our findings indicate that cancer cell metabolic responses depend on the type of DNA damage responses and can also be used to classify the type of DNA damage.

Keywords: DNA damage; MCF-7; NMR; PARP; metabolomics.

Figure 1

Principal component analysis (PCA) and hierarchical clustering of MCF-7 cell metabolic responses to different DNA damaging agents. a) Scores plot between 1st and 2nd principal component (PC) with 95% confidence region (represented by the dashed ellipses). The corresponding variance explained in each PC is shown in the bracket. b) Heat map showing the hierarchical clustering of metabolic concentrations (top 25 ranked according to ANOVA) using Pearson’s correlation as the similarity measure and Ward’s linkage as the clustering algorithm. Abbr. CNT: control, CSP: cisplatin, MMS: methyl methanesulfonate and ZEO: zeocin. The index for abbreviations used in heat map is given in Table 1 and Supplementary table 1 (Online Resource 1).

Figure 2

Pathway topological analysis of metabolic responses relative to control in MCF-7 cells. The pathways which are found to be significantly enriched (number of hits >1, FDR <0.05 and a non-zero impact score) in different DNA damaging agents-treated samples are shown. Metabolite importance was assessed using relative betweenness centrality measure and a global test was used to identify enriched pathways in treatment relative to control (adjusted for multiple testings). Abbreviation: CSP: cisplatin, MMS: methyl methanesulfonate and ZEO: zeocin. Grey color represent that the pathways is not significantly enriched.

Figure 3

Principal component analysis (PCA) of metabolic responses to MMS in the presence or absence of PARP inhibition. a) Scores plot between 1^st and 2^nd principal components (PC) with 95% confidence region (represented by the dashed ellipse). The cells were treated with MMS in presence or absence of ABT-888. The corresponding variance explained in each PC is shown in the bracket. b) Heat map showing the hierarchical clustering of metabolic concentrations (top 25 ranked according to ANOVA) using Pearson’s correlation as the similarity measure and Ward’s linkage as the clustering algorithm. Abbr. CNT: control, MMS: methyl methanesulfonate PI: PARP inhibitor and MMS+PI: combination of MMS and PI. The index for abbreviations used in heat map is given in Table 1 and Supplementary table 1 (Online Resource 1).

Figure 4

Effect of PARP inhibition as a single agent and in combination with MMS. Loading plots for principal component 1 (a) and principal component 2 (b) are shown. a) Loadings 1 represent the relationship between the metabolites which are mainly responsible for the differences between MMS and MMS+PI and b) Loadings 2 represents the metabolites which are mainly responsible for the differences due to addition of MMS. The index of abbreviations is given in the Supplementary table 1 (Online Resource 1).