Metabolomics of oncogene-specific metabolic reprogramming during breast cancer

Abstract

Background: The complex yet interrelated connections between cancer metabolism and oncogenic driver genes are relatively unexplored but have the potential to identify novel biomarkers and drug targets with prognostic and therapeutic value. The goal of this study was to identify global metabolic profiles of breast tumors isolated from multiple transgenic mouse models and to identify unique metabolic signatures driven by these oncogenes.

Methods: Using mass spectrometry (GC-MS, LC-MS/MS, and capillary zone electrophoresis (CZE)-MS platforms), we quantified and compared the levels of 374 metabolites in breast tissue from normal and transgenic mouse breast cancer models overexpressing a panel of oncogenes (PyMT, PyMT-DB, Wnt1, Neu, and C3-TAg). We also compared the mouse metabolomics data to published human metabolomics data already linked to clinical data.

Results: Through analysis of our metabolomics data, we identified metabolic differences between normal and tumor breast tissues as well as metabolic differences unique to each initiating oncogene. We also quantified the metabolic profiles of the mammary fat pad versus mammary epithelium by CZE-MS/MS. However, the differences between the tissues did not account for the majority of the metabolic differences between the normal mammary gland and breast tumor tissues. Therefore, the differences between the cohorts were unlikely due to cellular heterogeneity. Of the mouse models used in this study, C3-TAg was the only cohort with a tumor metabolic signature composed of ten metabolites that had significant prognostic value in breast cancer patients. Gene expression analysis identified candidate genes that may contribute to the metabolic reprogramming.

Conclusions: This study identifies oncogene-induced metabolic reprogramming within mouse breast tumors and compares the results to that of human breast tumors, providing a unique look at the relationship between and clinical value of oncogene initiation and metabolism during breast cancer.

Keywords: Breast cancer; Metabolic reprogramming; Metabolomics; Oncogene; Transgenic mouse model.

Fig. 1

Overview of metabolomics data. a Experimental design. Breast tumors were collected from five transgenic mouse models and normal mammary tissue from littermates. Samples then were analyzed by GC-MS and LC-MS/MS to acquire metabolomics data, which was then used to acquire oncogene-specific metabolic profiles. b Principal component analysis (PCA) of tumor metabolites. Each group of tumor samples (multi-colored) are separated from normal mammary tissue samples (gray) and from other tumor models. c Unsupervised cluster analysis of 47 samples and 374 metabolites. Samples are separated by column, and metabolites are separated by row. All normal tissue was within cluster 1 (blue); tumors are nearly all in cluster 2 (red), generally with high metabolite levels. d Super pathways of universally enriched metabolites in all tumor groups compared to normal tissue. The y-axis indicates the metabolic super pathways of metabolites, and the x-axis indicates the percentage of metabolites in each super pathway. The bars indicate metabolites that were upregulated more than twofold (blue) or fivefold (red) in all five tumor groups compared to normal tissue. The number of metabolites belonging to each major pathway was normalized against the total number of metabolites detected in the same pathway. See also Additional files 1, 2, and 3.

Fig. 2

Metabolites supporting rapid growth are increased in tumors. Box and whisker plots of key metabolites of energy pathways and major catabolism/anabolism pathways in normal mammary tissue and tumors of transgenic mice. y-axis: scaled intensity of metabolites by mass spectrometry. Cross: mean value; center line: median; box lines: upper/lower quartile values; extended line: upper/lower extreme values; circles: outliers. a Glycolysis and TCA cycle intermediate levels are higher in all tumor groups compared to control, consistent with higher energy flux in rapid-growing tumors. b Amino acid metabolites are upregulated in all tumors compared to control. c Lipid and nucleotide pathway metabolites have higher levels in tumors, consistent with the need for more building blocks in rapid-growing tumors. *P < 0.05, Welch’s t test

Fig. 3

Metabolic differences between the epithelium enriched and the stroma enriched breast tissue. To determine if the metabolic differences we saw between normal mammary tissue and tumors are indeed due to cancer progression, rather than the difference in cell type heterogeneity, we compared the mammary gland metabolome of stroma enriched tissue (adipocyte rich) and epithelium enriched tissue during normal mammary gland development. a Overview of experimental design. Stroma enriched and epithelium enriched tissue were collected at 3 weeks, at the time when the mammary epithelium invades only a part of the mammary fat pad. b Of the 1365 different features detected, only 48 metabolites had significantly different levels. c Major pathway distribution of adipocyte enriched and epithelium enriched metabolites. d Cloud plot of all metabolites with significantly different levels in the two tissue groups. Each circle indicates one metabolite, with the color indicating the tissue detected in (green: epithelial cells, red: stroma). The size of the circles indicates the fold of change (compared to background noise), and the color of the circles indicate P values (darker color indicating lower P value). The most significantly changed metabolites (with the lowest P values) in both tissues were presented on the right side of the plot

Fig. 4

Metabolomic differences associated with Wnt1-initiated tumors. Metabolomics revealed significantly different eicosanoid and cysteine-methionine metabolites in Wnt1 tumors compared to other transgenic model tumors. a Eicosanoid metabolism. Left: the eicosanoid pathway, with colors indicating increased metabolite levels compared to at least three other tumor groups (red), decreased levels (green), no change (blue), or not detected (black). Right: Graphs of the quantification of eicosanoid precursors (AA) and some eicosanoids. Middle line of box plots indicates median of sample group. *P < 0.05, Welch’s t test. b Quantification of cysteine-methionine metabolism. CDO1 (black) converts cysteine to hypotaurine. *P < 0.05, Welch’s t test. Statistical comparisons were made between Wnt1 and every other sample group. Abbreviations: COX cyclooxygenase, LOX lysyl oxidase DHGLA dihomo-γ-linolenic acid, HETE hydroxyeicosatetraenoic acid, SAM S-adenosyl methionine, SAH S-adenosylhomocysteine, THF tetrahydrofolate, Cdo1 cysteine dioxygenase

Fig. 5

Metabolic comparison of PyMT vs. PyMT-DB tumors. PyMT-DB has increased glycolytic flux and more glycogen breakdown, as well as reduced inflammation, according to the metabolomics results. a Super pathway distributions of significantly up/downregulated metabolites in PyMT-DB tumors compared to PyMT tumors. Increases in energy, carbohydrate, amino acid, and lipid metabolism are apparent in PyMT-DB tumors, as is the decrease in peptide metabolites. b Starch/glycogen metabolism is significantly increased in PyMT-DB tumors compared to PyMT tumors. c Glutathione metabolism is altered in PyMT-DB compared to PyMT, with higher glutathione levels and lower γ-glutamyl cycle intermediates. d Eicosanoid levels are lower in PyMT-DB vs. PyMT. *P < 0.05, Welch’s t test. Statistical comparisons were made between PyMT-DB and every other sample group

Fig. 6

Metabolic profile of Her2/neu tumors. Her2/neu tumors have increased lipid metabolism compared to other tumors. a Super pathway distributions of significantly up/downregulated metabolites in Her2/neu tumors compared to tumors of other transgenic mouse models and normal mammary tissue. Elevated levels of lipid metabolites are observed in Her2/neu tumors compared to other tumors. b Inositol metabolites are upregulated in Her2/neu tumors compared to other tumors. *P < 0.05, Welch’s t test. Statistical comparisons were made between Her2/neu and every other sample group

Fig. 7

Metabolic profile of C3-TAg tumors. C3-TAg has lower metabolite levels in all pathways compared to other tumor groups, but still maintains higher levels of metabolites than normal mammary tissue. a Super pathway distributions of significantly up/downregulated metabolites in C3-TAg tumors compared to tumors of other transgenic mouse models and normal mammary tissue. b Decreased turnover of glutathione via the γ-glutamyl cycle in C3-TAg tumors compared to other tumors. *P < 0.05, Welch’s t test. Statistical comparisons were made between C3-TAg and every other sample group