I was awarded the Department of Medicine School for Cancer Studies scholarship. This allowed me the opportunity to undertake a PhD at the Cancer Research UK funded Beatson Laboratories in Glasgow.
Working under the supervision of Professor Gerry Graham, my project focused on characterising the role of a newly discovered transcription factor called Gfi1B in haematopoiesis. Initial experiments were carried out to determine the temporal and spatial expression of Gfi1B during mouse embryogenesis. These studies involved isolation of the embryonic regions that had been shown to harbour the earliest haematopoietic stem cells, via tissue dissection. mRNA was extracted from these regions and the expression of Gfi1B analysed by RT-PCR. I also determined the expression profile of Gfi1B within the adult haematopoietic compartment. This involved isolating bone marrow from femur bones and then separating cells from different lineages and stages of differentiation using magnetic beads targeted to cell surface markers. The expression data suggested that Gfi1B was expressed in early haematopietic stem cells (HSCs) and progenitor cells but was shut off upon differentiation to more mature cell types in all lineages except for the erythroid lineage.
I then used a retroviral over-expression strategy to determine a functional role for Gfi1B within the haematopoietic compartment. Enforced expression was achieved in the model cell line FDCPmix. This cell line exhibits growth factor dependence, requiring IL-3 for survival, and can be induced to differentiate down both the myeloid and erythroid lineages. My hypothesis was that enforced expression of Gfi1B would enhance erythroid differentation at the expense myeloid differentiation and as a consequence there would be fewer myeloid and more erythroid colonies in the methylcellulose assays after differentation conditions were initiated. This did not happen – Gfi1B expressing FDCPmix cells produced the same amount of myeloid and erythroid colonies as the control cells. However, there was a striking difference in the erythroid colonies that were produced. The Gfi1B+ cells produced much larger colonies which appeared blocked in the final stages of maturation.
Transcription factors work in complexes with fellow transcription factors and accessory proteins to regulate gene expression. The binding partners often offer clues as to what the downstream target genes are and this in turn can shed light on overall function. I constructed a yeast-2-hybrid library from FDCPmix and bone marrow cells to fish out binding partners of Gfi1B. These experiments were tricky as it turned out Gfi1B, despite being classed as a transcriptional repressor based on it’s homology to Gfi1, had a strong activation domain. As the yeast-2-hybrid system is based on a series of transcriptional reporter assays this proved problematic and resulted in a large number of false positives. After sifting through the results I carried out co-immunoprecipitation assays to validate binding partners discovered in the Y2H screen.