El Ansari, Rokaya
(2020)
Amino acid transporters in breast cancer: promising prognostic markers and emerging targets across molecular subtypes.
PhD thesis, University of Nottingham.
Abstract
Background: To maintain the necessary energy and cellular building blocks, tumour cells selectively upregulate expression of cell-surface nutrienttransporters. One of the most prominent are amino acid transporters, which catalyse the uptake of vital amino acids, particularly glutamine. Glutamine promotes cell growth by supporting bioenergetic and biosynthetic metabolism, maintaining redox balance and activating the mTORC1 signalling pathway, which supports protein translation and prevents apoptosis in cancer cells. These metabolic alterations are mediated by mutations in oncogenes and/or tumour suppressors that result in sustained cancer cell growth and proliferation. Cellular interaction with a tumour microenvironment, at the metabolic level, can also promote cancer aggressiveness and metastasis. Breast cancer (BC) is a heterogeneous disease with various subtypes that differ in terms of biology and clinical behaviour. There is growing evidence that differences in amino acid metabolism exist between BC molecular subtypes. However, the expression of amino acid transporters and the subsequent prognostic implications in BC remain elusive, especially with regard to the BC molecular subtypes. It was therefore hypothesised that amino acid transporters involved in the glutamine pathway are overexpressed in BC and they have prognostic implications that vary between BC molecular subtypes. This study aimed to determine the prognostic significance of the eminent amino acid transporters and their downstream signal in different BC molecular subtypes and to investigate the usability of the key amino acid transporters as potential targets for therapeutic interventions in BC. Furthermore, this study sought to find whether upregulation of amino acid uptake in BC participates in shaping the tumour microenvironment through attracting specific immune cell subtypes, including the subsequent impact on patient outcome.
Methodologies: The solute carriers SLC1A5, SLC7A5, SLC3A2, SLC38A2, SLC7A11 and SLC7A8 and the downstream signal,p70S6K, were each assessed at the genomic level using METABRIC data (n=1980) and Breast Cancer GeneExpression Miner dataset (n= 4904), and at the proteomic level using immunohistochemical analysis and tissue microarray (TMA) sections constructed from a large, well-characterised primary BC cohort (n=2770). Their expression was correlated with clinicopathological parameters, BC molecular subtypes, relevant biological markers and patient outcome. K-mean and PAM clustering algorithms were applied to stratify the BC cohort into accredited “SLC clusters” based on the protein expression data of the key solute carriers (SLC1A5, SLC7A5 and SLC3A2). The clinicopathological relevance of the SLC clusters was assessed, including associations with patient outcome. In addition, SLC clusters were correlated with various immune cell markers (CD3, CD8, FOXP3, CD20, CD68, PD1 and PDL1) as components of tumour microenvironment. The co-expression of the two compartments was associated with patient outcome, considering the BC molecular subtypes. In a different approach, in vitro studies using different BC cell lines were conducted to investigate the effect of transient siRNA knockdown of SLC1A5 and SLC7A5 on cell proliferation, cell invasion and amino acid uptake. The effect of silencing the two solute carriers on PDL1 expression was also assessed.
Results: All biomarkers included in this study, apart from SLC7A8, showed a strong correlation with poor clinicopathological parameters and oestrogen receptor negative (ER-) status. The results of the survival analysis differed among BC molecular subtypes. While high SLC1A5, SLC7A5 and SLC3A2 expression shared the characteristics of being predictive of patient outcome only in the ER+ highly proliferative/luminal B subtype, SLC38A2 showed similar associations in triple negative (TN) tumours, and p70S6K and SLC7A8 were limited to the ER+ low proliferative/luminal A subtype. In multivariate analysis, all aforementioned biomarkers were independent factors for predicting patient survival, excluding SLC1A5, which was significant only in the univariate analysis, and SLC7A11, which did not show any significant associations with patient outcome. A significant association was also observed with other relevant biological markers.The most prominent was the key regulator of tumour cell metabolism, MYC, and a group of glutamine metabolic enzymes, particularly those involved in the glutamine-proline regulatory axis. Clustering analysis of SLC1A5, SLC7A5, and SLC3A2 protein expression resulted in classifying BC patients into three clusters; low SLCs (SLC1A5-/SLC7A5- /SLC3A2-), high SLC1A5 (SLC1A5+/SLC7A5-/SLC3A2-), and high SLCs (SLC1A5+/SLC7A5+/SLC3A2+). Each cluster had distinct correlations to known prognostic factors and patient outcome. Multivariate analysis showed that the SLC clusters were independent risk factors for shorter BC-specific survival only in the ER+ highly proliferative tumours. The high SLCs cluster showed a significant association with different immune cell markers, primarily in the TN subtype. In addition, the co-occurrence of SLCs with immune markers had an additive effect on predicting patient outcome. Targeting SLC7A5 but not SLC1A5 significantly reduced the expression of PDL1 in the TN cell line. In this context, while silencing SLC1A5 or SLC7A5 significantly reduced amino acid uptake in the respective BC cell lines, the impact on cell proliferation and cell invasion was significant only upon silencing SLC7A5.
Conclusion: Continued refinement in understanding the biological diversity of BC and linked development of classification strategies suitable for routine clinical use are essential to achieve a personalised approach to BC management. This thesisdemonstrated that the amino acid transport system for glutamine has a crucial prognostic role in BC, which is variable among BC molecular subtypes. Solute carriers that showed the liability of being independent predictors for poor patient outcome can be potential targets for BC treatment, particularly in synergism. Furthermore, this study deduced that enhanced amino acid uptake by cancer cells can influence the composition of immune cell infiltrates, which has an additional effect on patient outcome. Co-targeting the key amino acid transporters and the pro-tumorigenic immune cells may be a novel approach for BC treatment, particularly in the TN subgroup. Further elaborated functional studies may provide new insights into the specific role played by the amino acid transporters in the aggressive BC molecular subtypes.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
![[thumbnail of Rokaya Elansari-PhD thesis (final submission).pdf]](https://eprints.nottingham.ac.uk/style/images/fileicons/application_pdf.png "Rokaya Elansari-PhD thesis (final submission).pdf")