Limits...
Metabolomic characterisation of the effects of oncogenic PIK3CA transformation in a breast epithelial cell line

View Article: PubMed Central - PubMed

ABSTRACT

Somatic mutations in PIK3CA are frequently found in a number of human cancers, including breast cancer, altering cellular physiology and tumour sensitivity to chemotherapy. This renders PIK3CA an attractive molecular target for early detection and personalised therapy. Using 1H Nuclear Magnetic Resonance spectroscopy (NMR) and Gas Chromatography – Mass Spectrometery (GC-MS) together with 13C stable isotope-labelled glucose and glutamine as metabolic tracers, we probed the phenotypic changes in metabolism following a single copy knock-in of mutant PIK3CA (H1047R) in the MCF10A cell line, an important cell model for studying oncogenic transformation in breast tissues. We observed effects in several metabolic pathways, including a decrease in glycerophosphocholine level together with increases in glutaminolysis, de novo fatty acid synthesis and pyruvate entry into the tricarboxylic acid cycle. Our findings highlight altered glyceroplipid metabolism and lipogenesis, as key metabolic phenotypes of mutant PIK3CA transformation that are recapitulated in the MCF10A cellular model.

No MeSH data available.


Modelled metabolic parameters from fatty acid Isotopomer Spectral Analysis (ISA).ISA parameters were modelled based on mass isotopomer distribution (MID) of methyl palmitate. In the glutamine tracer data, the graphs represent the mean ± SEM from four independent biological replicate experiments and in the glucose tracer data the graphs represent the mean ± SEM data from three independent biological replicate experiments. Two-tailed Student’s t-test was used to determine statistical significance, and *denotes t-test p < 0.05.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5385542&req=5

f4: Modelled metabolic parameters from fatty acid Isotopomer Spectral Analysis (ISA).ISA parameters were modelled based on mass isotopomer distribution (MID) of methyl palmitate. In the glutamine tracer data, the graphs represent the mean ± SEM from four independent biological replicate experiments and in the glucose tracer data the graphs represent the mean ± SEM data from three independent biological replicate experiments. Two-tailed Student’s t-test was used to determine statistical significance, and *denotes t-test p < 0.05.

Mentions: Intracellular lipids were analysed by GC-MS using an extraction and derivatisation method that enabled both fatty acid esters and free fatty acids to be detected. The fatty acid methyl esters detected were not restricted to a specific class of lipid molecule, but are fatty acid chains from all lipid molecules, which were transesterified in the dervitisation process. These transesterified fatty acids could comprise various lipids, such as; membrane phospholipids (phosphatidylcholine) or signalling and functional lipids (phosphatidic acid or diacylglycerol). Our analysis successfully measured various lipid bound fatty acids, and we were able to examine the relative abundance of a number of methyl esters (lipid bound fatty acids) and the free fatty acid oleate (Table S2). For example, we found that both ratios of esterified linolenate to palmitate and esterified linolenate to oleate were significantly lower (pairwise t-test p < 0.05) in the PIK3CA mutant extracts. Whilst oleate and palmitate can be synthesised de novo, linolenate (C18:3) is an essential polyunsaturated fatty acid in mammalian cells and must be imported from the culture medium directly; our data indicated there was a possible shift in the PIK3CA mutant cells, away from relying on fatty acid uptake and towards de novo biosynthesis. Furthermore, the mass isotopomer data from U-13C6 glucose and U-13C5 glutamine both provided strong independent evidence that the rate of de novo biosynthesis of fatty acids was elevated in the PIK3CA mutant cells. In particular, we found increased incorporation of both glucose and glutamine derived two-carbon acetyl-CoA units into methyl palmitate (Figures S5 and S6), the most abundant fatty acid chain in mammalian cells. By modelling the mass isotopomer distribution of methyl palmitate using Isotopomer Spectral Analysis (ISA), a technique that untangles the effect of changes in the acetyl CoA pool contribution from the biosynthetic rate24, it was established that the increases in 13C tracer label incorporation were results of higher rates of de novo biosynthesis in the PIK3CA mutant cells (Fig. 4, Table S6). Upon culture with U-13C5 glutamine the PIK3CA mutant cells demonstrated higher de novo biosynthesis of palmitate (lipid bound methyl palmitate was 50% higher (p < 0.05) when compared to the wild type MCF10A parental line). Additionally, the modelled data suggest that glucose-derived citrate was preferentially used as substrate for forming lipogenic acetyl-CoA; roughly 60% of acetyl-CoA came from glucose as opposed to around 10% from glutamine (Fig. 4).


Metabolomic characterisation of the effects of oncogenic PIK3CA transformation in a breast epithelial cell line
Modelled metabolic parameters from fatty acid Isotopomer Spectral Analysis (ISA).ISA parameters were modelled based on mass isotopomer distribution (MID) of methyl palmitate. In the glutamine tracer data, the graphs represent the mean ± SEM from four independent biological replicate experiments and in the glucose tracer data the graphs represent the mean ± SEM data from three independent biological replicate experiments. Two-tailed Student’s t-test was used to determine statistical significance, and *denotes t-test p < 0.05.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5385542&req=5

f4: Modelled metabolic parameters from fatty acid Isotopomer Spectral Analysis (ISA).ISA parameters were modelled based on mass isotopomer distribution (MID) of methyl palmitate. In the glutamine tracer data, the graphs represent the mean ± SEM from four independent biological replicate experiments and in the glucose tracer data the graphs represent the mean ± SEM data from three independent biological replicate experiments. Two-tailed Student’s t-test was used to determine statistical significance, and *denotes t-test p < 0.05.
Mentions: Intracellular lipids were analysed by GC-MS using an extraction and derivatisation method that enabled both fatty acid esters and free fatty acids to be detected. The fatty acid methyl esters detected were not restricted to a specific class of lipid molecule, but are fatty acid chains from all lipid molecules, which were transesterified in the dervitisation process. These transesterified fatty acids could comprise various lipids, such as; membrane phospholipids (phosphatidylcholine) or signalling and functional lipids (phosphatidic acid or diacylglycerol). Our analysis successfully measured various lipid bound fatty acids, and we were able to examine the relative abundance of a number of methyl esters (lipid bound fatty acids) and the free fatty acid oleate (Table S2). For example, we found that both ratios of esterified linolenate to palmitate and esterified linolenate to oleate were significantly lower (pairwise t-test p < 0.05) in the PIK3CA mutant extracts. Whilst oleate and palmitate can be synthesised de novo, linolenate (C18:3) is an essential polyunsaturated fatty acid in mammalian cells and must be imported from the culture medium directly; our data indicated there was a possible shift in the PIK3CA mutant cells, away from relying on fatty acid uptake and towards de novo biosynthesis. Furthermore, the mass isotopomer data from U-13C6 glucose and U-13C5 glutamine both provided strong independent evidence that the rate of de novo biosynthesis of fatty acids was elevated in the PIK3CA mutant cells. In particular, we found increased incorporation of both glucose and glutamine derived two-carbon acetyl-CoA units into methyl palmitate (Figures S5 and S6), the most abundant fatty acid chain in mammalian cells. By modelling the mass isotopomer distribution of methyl palmitate using Isotopomer Spectral Analysis (ISA), a technique that untangles the effect of changes in the acetyl CoA pool contribution from the biosynthetic rate24, it was established that the increases in 13C tracer label incorporation were results of higher rates of de novo biosynthesis in the PIK3CA mutant cells (Fig. 4, Table S6). Upon culture with U-13C5 glutamine the PIK3CA mutant cells demonstrated higher de novo biosynthesis of palmitate (lipid bound methyl palmitate was 50% higher (p < 0.05) when compared to the wild type MCF10A parental line). Additionally, the modelled data suggest that glucose-derived citrate was preferentially used as substrate for forming lipogenic acetyl-CoA; roughly 60% of acetyl-CoA came from glucose as opposed to around 10% from glutamine (Fig. 4).

View Article: PubMed Central - PubMed

ABSTRACT

Somatic mutations in PIK3CA are frequently found in a number of human cancers, including breast cancer, altering cellular physiology and tumour sensitivity to chemotherapy. This renders PIK3CA an attractive molecular target for early detection and personalised therapy. Using 1H Nuclear Magnetic Resonance spectroscopy (NMR) and Gas Chromatography &ndash; Mass Spectrometery (GC-MS) together with 13C stable isotope-labelled glucose and glutamine as metabolic tracers, we probed the phenotypic changes in metabolism following a single copy knock-in of mutant PIK3CA (H1047R) in the MCF10A cell line, an important cell model for studying oncogenic transformation in breast tissues. We observed effects in several metabolic pathways, including a decrease in glycerophosphocholine level together with increases in glutaminolysis, de novo fatty acid synthesis and pyruvate entry into the tricarboxylic acid cycle. Our findings highlight altered glyceroplipid metabolism and lipogenesis, as key metabolic phenotypes of mutant PIK3CA transformation that are recapitulated in the MCF10A cellular model.

No MeSH data available.