Limits...
Exometabolom analysis of breast cancer cell lines: Metabolic signature.

Willmann L, Erbes T, Halbach S, Brummer T, Jäger M, Hirschfeld M, Fehm T, Neubauer H, Stickeler E, Kammerer B - Sci Rep (2015)

Bottom Line: Samples were analyzed by application of reversed phase chromatography coupled to a triple quadrupole mass spectrometer.Collectively, we determined 23 compounds from RNA metabolism, two from purine metabolism, five from polyamine/methionine cycle, one from histidine metabolism and two from nicotinate and nicotinamide metabolism.Differences in metabolite excretion resulting from cancerous metabolism can be integrated into altered processes on the cellular level.

View Article: PubMed Central - PubMed

Affiliation: Center for Biological Systems Analysis ZBSA, Albert-Ludwigs-University Freiburg, 79104 Freiburg, Germany.

ABSTRACT
Cancer cells show characteristic effects on cellular turnover and DNA/RNA modifications leading to elevated levels of excreted modified nucleosides. We investigated the molecular signature of different subtypes of breast cancer cell lines and the breast epithelial cell line MCF-10A. Prepurification of cell culture supernatants was performed by cis-diol specific affinity chromatography using boronate-derivatized polyacrylamide gel. Samples were analyzed by application of reversed phase chromatography coupled to a triple quadrupole mass spectrometer. Collectively, we determined 23 compounds from RNA metabolism, two from purine metabolism, five from polyamine/methionine cycle, one from histidine metabolism and two from nicotinate and nicotinamide metabolism. We observed major differences of metabolite excretion pattern between the breast cancer cell lines and MCF-10A, just as well as between the different breast cancer cell lines themselves. Differences in metabolite excretion resulting from cancerous metabolism can be integrated into altered processes on the cellular level. Modified nucleosides have great potential as biomarkers in due consideration of the heterogeneity of breast cancer that is reflected by the different molecular subtypes of breast cancer. Our data suggests that the metabolic signature of breast cancer cell lines might be a more subtype-specific tool to predict breast cancer, rather than a universal approach.

No MeSH data available.


Related in: MedlinePlus

Hierarchical cluster analysis of target compounds detected in breast cancer cell lines (BCCL) MDA-MB-231, -453, BT-474 and breast epithelial cell line (BECL) MCF-10A (n=3).(Rows: samples, columns: target compounds; abbreviations of target compounds are shown in Table 2).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4544000&req=5

f5: Hierarchical cluster analysis of target compounds detected in breast cancer cell lines (BCCL) MDA-MB-231, -453, BT-474 and breast epithelial cell line (BECL) MCF-10A (n=3).(Rows: samples, columns: target compounds; abbreviations of target compounds are shown in Table 2).

Mentions: For the comparison of the metabolite excretion of the three breast cancer cell lines with the breast epithelial cell line MCF-10A, we executed a semi-quantitative analysis using MRM operating mode. Therefore we related the peak area of the quantifier mass transitions of each analyte to the peak area of the internal standard isoguanosine. The quantifier MRM transitions of analyzed compounds are shown in Fig. 3. The ascertained values for the compounds in media were subtracted. The resulting values were normalized using the counted number of cells at the moment of supernatant sampling. The cell lines showed significant differences in metabolite excretion (Table 3). Compound levels were generally considered as “elevated” or “decreased”, if the values were a minimum of two standard deviations (2σ-concept) higher or lower than the reference value of MCF-10A. In our semi-quantitative analysis of cell culture supernatants, we observed major differences of metabolite excretion pattern between breast cancer cell lines and MCF-10A, just as well as between the different breast cancer cell lines themselves (Table 3). Breast cancer cell lines could be separated from the reference cell line MCF-10A by PLSDA with component 1 reaching 55.6% variance and component 2 reaching 21.5%, respectively (Fig. 4a). The prediction accuracies have been assessed by cross validation with different numbers of components (Fig. 4b). Q2 is a prediction error measure with an optimal value of 1, but Q2 can also assume negative values for bad predictive models. Although the best performance could be obtained with five components, a number of two components already delivered an adequate prediction model. The VIP score obtained by PLSDA is an important measure of each independent variable. Higher VIP scores are considered more relevant in classification. The most influential features with a VIP score of 1.2 to 1.6 and moderately influential features with a VIP score of 0.8 to 1.0 are depicted in Fig. 4c. Especially the target compounds m5U, N6-Methyl-N6-threonylcarbamoyladenosine (m6t6A), t6A, N6-Succinyloadenosine (N6-SAR), m1A, m1G, ac4C, X, Ψ, 3-Methylcytidine (m3C) and 2-Methylthio-N6-threonylcarbamoyladenosine (ms2t6A) are elevated, have a high VIP score and should consequently be considered as potential biomarkers for breast cancer. Hierarchical cluster analysis (Fig. 5) of breast cancer cell lines and the breast epithelial cell line MCF-10A revealed that BT-474, followed by MDA-MB-231, is the most similar to MCF-10A based on excretion levels of investigated compounds. The cell line MDA-MB-231 was found to be the most different to the breast epithelial cell line MCF-10A. The differential excretion of metabolites deriving from different biochemical origin will be discussed in the following paragraphs.


Exometabolom analysis of breast cancer cell lines: Metabolic signature.

Willmann L, Erbes T, Halbach S, Brummer T, Jäger M, Hirschfeld M, Fehm T, Neubauer H, Stickeler E, Kammerer B - Sci Rep (2015)

Hierarchical cluster analysis of target compounds detected in breast cancer cell lines (BCCL) MDA-MB-231, -453, BT-474 and breast epithelial cell line (BECL) MCF-10A (n=3).(Rows: samples, columns: target compounds; abbreviations of target compounds are shown in Table 2).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4544000&req=5

f5: Hierarchical cluster analysis of target compounds detected in breast cancer cell lines (BCCL) MDA-MB-231, -453, BT-474 and breast epithelial cell line (BECL) MCF-10A (n=3).(Rows: samples, columns: target compounds; abbreviations of target compounds are shown in Table 2).
Mentions: For the comparison of the metabolite excretion of the three breast cancer cell lines with the breast epithelial cell line MCF-10A, we executed a semi-quantitative analysis using MRM operating mode. Therefore we related the peak area of the quantifier mass transitions of each analyte to the peak area of the internal standard isoguanosine. The quantifier MRM transitions of analyzed compounds are shown in Fig. 3. The ascertained values for the compounds in media were subtracted. The resulting values were normalized using the counted number of cells at the moment of supernatant sampling. The cell lines showed significant differences in metabolite excretion (Table 3). Compound levels were generally considered as “elevated” or “decreased”, if the values were a minimum of two standard deviations (2σ-concept) higher or lower than the reference value of MCF-10A. In our semi-quantitative analysis of cell culture supernatants, we observed major differences of metabolite excretion pattern between breast cancer cell lines and MCF-10A, just as well as between the different breast cancer cell lines themselves (Table 3). Breast cancer cell lines could be separated from the reference cell line MCF-10A by PLSDA with component 1 reaching 55.6% variance and component 2 reaching 21.5%, respectively (Fig. 4a). The prediction accuracies have been assessed by cross validation with different numbers of components (Fig. 4b). Q2 is a prediction error measure with an optimal value of 1, but Q2 can also assume negative values for bad predictive models. Although the best performance could be obtained with five components, a number of two components already delivered an adequate prediction model. The VIP score obtained by PLSDA is an important measure of each independent variable. Higher VIP scores are considered more relevant in classification. The most influential features with a VIP score of 1.2 to 1.6 and moderately influential features with a VIP score of 0.8 to 1.0 are depicted in Fig. 4c. Especially the target compounds m5U, N6-Methyl-N6-threonylcarbamoyladenosine (m6t6A), t6A, N6-Succinyloadenosine (N6-SAR), m1A, m1G, ac4C, X, Ψ, 3-Methylcytidine (m3C) and 2-Methylthio-N6-threonylcarbamoyladenosine (ms2t6A) are elevated, have a high VIP score and should consequently be considered as potential biomarkers for breast cancer. Hierarchical cluster analysis (Fig. 5) of breast cancer cell lines and the breast epithelial cell line MCF-10A revealed that BT-474, followed by MDA-MB-231, is the most similar to MCF-10A based on excretion levels of investigated compounds. The cell line MDA-MB-231 was found to be the most different to the breast epithelial cell line MCF-10A. The differential excretion of metabolites deriving from different biochemical origin will be discussed in the following paragraphs.

Bottom Line: Samples were analyzed by application of reversed phase chromatography coupled to a triple quadrupole mass spectrometer.Collectively, we determined 23 compounds from RNA metabolism, two from purine metabolism, five from polyamine/methionine cycle, one from histidine metabolism and two from nicotinate and nicotinamide metabolism.Differences in metabolite excretion resulting from cancerous metabolism can be integrated into altered processes on the cellular level.

View Article: PubMed Central - PubMed

Affiliation: Center for Biological Systems Analysis ZBSA, Albert-Ludwigs-University Freiburg, 79104 Freiburg, Germany.

ABSTRACT
Cancer cells show characteristic effects on cellular turnover and DNA/RNA modifications leading to elevated levels of excreted modified nucleosides. We investigated the molecular signature of different subtypes of breast cancer cell lines and the breast epithelial cell line MCF-10A. Prepurification of cell culture supernatants was performed by cis-diol specific affinity chromatography using boronate-derivatized polyacrylamide gel. Samples were analyzed by application of reversed phase chromatography coupled to a triple quadrupole mass spectrometer. Collectively, we determined 23 compounds from RNA metabolism, two from purine metabolism, five from polyamine/methionine cycle, one from histidine metabolism and two from nicotinate and nicotinamide metabolism. We observed major differences of metabolite excretion pattern between the breast cancer cell lines and MCF-10A, just as well as between the different breast cancer cell lines themselves. Differences in metabolite excretion resulting from cancerous metabolism can be integrated into altered processes on the cellular level. Modified nucleosides have great potential as biomarkers in due consideration of the heterogeneity of breast cancer that is reflected by the different molecular subtypes of breast cancer. Our data suggests that the metabolic signature of breast cancer cell lines might be a more subtype-specific tool to predict breast cancer, rather than a universal approach.

No MeSH data available.


Related in: MedlinePlus