Breast cancer metabolism and responsiveness to dichloroacetate: relationships with 15N and 13C natural abundance

Abstract

Abstract Background Metabolic reprogramming is a hallmark of breast cancer (BrCa), with alterations in glycolysis, glutamine metabolism, and the urea cycle contributing to tumour progression. Dichloroacetate (DCA), a pyruvate dehydrogenase kinase (PDK) inhibitor, shifts metabolism toward oxidative phosphorylation and has been proposed as a therapeutic agent. While isotope tracing is well-established, natural isotope abundance (δ¹³C, δ¹⁵N) is emerging as a biomarker of metabolic alterations in cancer. Methods We investigated the relationship between isotope composition and metabolism in BrCa using two BALB/c mouse mammary tumour models (V14 and 4T1) and assessed the effects of DCA treatment using metabolomics, lipidomics and isotopomics. Results V14 and 4T1 tumours exhibited isotopic patterns similar to human tumours, with δ¹³C enrichment and δ¹⁵N depletion relative to non-cancerous mammary tissue. V14 tumours were more δ¹⁵N-depleted than 4T1, reflecting differences in nitrogen metabolism. Multivariate analysis integrating isotopic, metabolomic, and lipidomic data revealed isotopic features as key discriminators between tumours and normal tissues. Compared to V14, 4T1 tumours were enriched in TCA intermediates, sphingolipids, and amino acids, whereas V14 tumours showed elevated glutaminolytic and nitrogenous metabolites. DCA treatment differentially affected tumour growth, with V14 tumours more sensitive than 4T1. DCA altered nitrogen metabolism, increasing the arginine-to-ornithine ratio, and modulating δ¹⁵N values in a tumour-specific manner increasing V14 and decreasing 4T1 δ¹⁵N values. DCA had little effect on δ¹³C. δ¹³C values were primarily determined by the balance between lipid and TCA cycle metabolites, rather than glycolytic flux. δ¹⁵N variation was linked to nitrogen metabolism, including urea cycle intermediates and sphingolipid composition, with a potential role for choline-related fractionation in δ¹⁵N depletion. Altered gene expression of Hacd2 and Acot12 in V14 tumours after DCA treatment was reflected in shorter fatty acid tails in phosphatidyl cholines, supporting the lipidomics data. Conclusions These findings support the hypothesis that cancer-associated metabolic reprogramming influences natural isotope abundance. Correlations between isotope shifts and metabolic signatures highlight the potential of lipid-derived δ¹⁵N as a biomarker of tumour metabolic state, with implications for noninvasive metabolic profiling in BrCa. Abstract Figure