Investigating the role of the UDP-galactose transporter SLC35A2 in the regulation of HIF signalling

Abstract

Background: The hypoxia inducible factor (HIF) family of dimeric transcription factors play a vital role in the cellular response to low oxygen (hypoxia). HIF activation leads to the transcriptional upregulation of a range of genes that are involved in diverse physiological and pathophysiological processes. In malignant tumours increased HIF-α expression, and HIF activation are commonly observed, and are associated with poorer prognosis and disease progression. HIF activity is controlled by a number of different mechanisms, and the identification of molecular regulators of HIF is of intense interest. The Golgi UDP-galactose transporter SLC35A2 was identified by the Ashcroft group as a novel regulator of HIF-α, and CHO cells that had lost SLC35A2 exhibited elevated HIF-α protein in normoxia. It is known that the absence of SLC35A2 in the Golgi leads to wide-ranging glycosylation defects. The underlying mechanisms linking SLC35A2 to HIF-α protein regulation were not known. This thesis aims to further characterise the role effect of SLC35A2 loss on HIF signalling and tumour cell behaviour, in CHO cells and a human cell system, specifically by exploring the hypotheses that these cells may harbour an autophagy defect, or differences in overall UDP-sugar content, in particular O-GlcNAc. Methods and Results: I used parental and Slc35a2 mutant CHO cells, SLC35A2 knockout (KO) and wild type (WT) suspension HeLa (sHeLa) cells, and a panel of cancer cells lines, including 786O, RCC4 renal carcinoma cells which exhibit constitutive HIF activation due to loss of VHL function. Confirming previous work from the Ashcroft group, I showed that Slc35a2 mutant (M6.19) CHO cells exhibit elevated normoxic HIF-1α protein levels and evidence of a glycosylation defect. Expanding on these findings, I found that SLC35A2 loss led to elevated normoxic HIF-α (HIF-1α and HIF-2α) protein and HIFA mRNΑ, and elevated expression of HIF-α target genes (GLUT1, VEGF). Using global RNAseq analysis of SLC35A2 KO and WT sHeLa cells, I found that SLC35A2 loss was associated with highly upregulated expression of a range of genes. Further exploration of the observed glycosylation defect in Slc35a2 mutant CHO cells, showed that SLC35A2 KO sHeLa cells also exhibited altered mobility of GLUT1 protein, which I found was consistent with treatment of WT sHeLa cells with tunicamycin, a protein N-glycosylation inhibitor. UDP-sugar analysis by HPLC of Slc35a2 mutant CHO and SLC35A2 KO sHeLa cells indicated increased levels of UDP-GlcNAc compared to their WT counterparts. Furthermore, I found similar patterns of increased UDP-GlcNAc levels in SLC35A2 KO compared to WT sHeLa cells in normoxia and hypoxia, while stable reconstitution of Slc35a2 in Slc35a2 mutant (M6.19) CHO cells reduced levels of UDP-GlcNAc and rescued CMP-sialic acid levels. Consistent with my findings and the involvement of UDP-GlcNAc in glutamine metabolism, I found that Slc35a2 mutant (M6.19) CHO cells were significantly more sensitive to glutamine withdrawal compared to parental (C4.5) CHO cells. Slc35a2 mutant (M6.19) and SLC35A2 KO sHeLa cells exhibited a glycosylation defect of LAMP2A, a protein involved in HIF-1α lysosomal degradation and critical for chaperone mediated autophagy. SLC35A2 mutant (M6.19) CHO cells were significantly more sensitive to the inhibition of proliferation by the inhibitors of autophagy bafilomycin and 3-methyladenine, and showed elevated levels of the autophagy marker LC3B. Finally, from analyses of the TCGA-KIRC database I found SCL35A2 expression was higher in patients with non-mutant VHL versus mutant VHL in renal cancers. In support of these findings, I showed that basal SLC35A2 protein and mRNA levels were higher in patient-derived 786O renal carcinoma cells reconstituted with wild type VHL (786O-VHL) compared to matched 786-O empty vector control (786O-EV) cells. I also found that high SLC35A2 expression is associated with poor prognosis in patients with VHL non-mutant ccRCC. Conclusion: Taken together, my thesis identifies SLC35A2 as a regulator of HIF-α and metabolism, potentially through its role in regulating UDP-sugars, and reveals a possible novel role for SLC35A2 in lysosomal processing and autophagy. SLC35A2 expression in renal cell carcinoma is associated with VHL status, which may provide a new route for dysregulation of HIF, altered metabolism, and changes in lysosomal processing and autophagy.