Investigating gene expression patterns and amino acid metabolism in CML LSCs using transcriptomics and ex-vivo expansion

Roy, Kiron (2025) Investigating gene expression patterns and amino acid metabolism in CML LSCs using transcriptomics and ex-vivo expansion. PhD thesis, University of Glasgow.

Abstract

Chronic myeloid leukaemia (CML) is a myeloproliferative disease where a normal haematopoietic stem cell (HSC) undergoes chromosomal translocation, leading to the formation of the Philadelphia chromosome (Ph). The Ph contains the fusion oncogene breakpoint cluster region abelson murine Leukemia 1 (BCR::ABL1), which encodes a constitutively active tyrosine kinase. BCR::ABL1 signalling transforms a HSC into a leukaemic stem cell (LSC), which is responsible for propagating CML. Tyrosine kinase inhibitors (TKIs) have vastly improved patient outcomes in the long and short term, however BCR::ABL1 inhibition alone is insufficient to eradicate all disease causing LSCs.

Bulk transcriptome profiling is a method which has been used to study CML LSCs for over 20 years, and has revealed much regarding the phenotype and molecular features of the LSC. Recent advances such as single cell transcriptomics have again increased the granularity at which we understand LSCs, now providing the capacity to phenotypically identify BCR::ABL1+ LSCs, predict treatment responses, and identify features of treatment resistant LSCs. Metabolomics and transcriptomics have also shown that LSCs possess enhanced metabolism compared to their HSC counterparts, with metabolic adaption also being a mechanism by which LSCs are capable of surviving TKI treatment. Therefore, we aimed to conduct a metaanalysis on CML LSC transcriptomic datasets to determine if we could identify novel BCR::ABL1+ specific signals and metabolic genes across phenotypes, ranging from CD34+ to CD34+CD38-CD90+CD26+ cells. Additionally, we aimed to use recent advances in ex-vivo HSC culture and apply them to LSCs, to identify the metabolic weaknesses of murine CML LSCs, and validate these findings in-vivo.

In our meta-analysis, we show numerous uncharacterised genes differentially expressed in LSCs compared to bystander BCR::ABL1- or normal HSCs across datasets and phenotypes, and a core of commonly expressed genes that would likely represent meaningful therapeutic targets. We also observed widespread changes in all areas of metabolism, particularly OXPHOS, amino acid (AA) and fatty acid metabolism in Lin-CD34+CD38- LSCs. While we detected fewer changes within the Lin-CD34+CD38-CD90+ compartment, we were still able to detect metabolic changes, suggesting this weakness is present even in a highly primitive phenotype.

To assess the utility of ex-vivo culture methods, we conducted hibernation culture and poly-vinyl alcohol (PVA) based expansion culture on CML LSCs. We found that the low number of LSCs that could be purified from leukaemic mice made hibernation culture a challenge when considering hypothesis testing, even though it may be more physiologically relevant. Using PVA based ex-vivo expansion, we were capable of expanding LSCs from a c-kit enriched or sorted population under normoxia or hypoxia. We then used expanded LSCs to conduct an AA dropout screen, wherein LSCs were grown in media individually missing each of the AAs. Among the top hits were valine, leucine and isoleucine (the branched chain amino acids (BCAAs)), which we further characterised using a combination of expanded LSCs and cell lines. We found BCAA restriction in-vitro to be highly efficacious, with a 75% reduction in BCAA concentration having a strong anti-leukaemic effect in both cell lines and expanded cells. We also show that BCAA restriction reduced ex-vivo expanded HSC growth substantially. Lastly, we used dietary restriction of BCAAs to treat murine CML in-vivo, however, the combination of BCAA restriction and CML was poorly tolerated, and so we could not detect any anti-leukaemic effect.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Subjects: R Medicine > RC Internal medicine > RC0254 Neoplasms. Tumors. Oncology (including Cancer)
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Cancer Sciences
• Supervisor's Name: Helgason, Professor Vignir
• Date of Award: 2025
• Depositing User: Theses Team
• Unique ID: glathesis:2025-85649
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 12 Dec 2025 09:37
• Last Modified: 12 Dec 2025 09:41
• Thesis DOI: 10.5525/gla.thesis.85649
• URI: https://theses.gla.ac.uk/id/eprint/85649