Researchers investigate O-glycosylation in cells infected with Ebola

Ebola virus (EBOV) is a hemorrhagic virus endemic to Africa, which can spread rapidly in populated areas and has a high rate of fatality. Significant media attention was garnered during outbreaks over the last decade and has helped raise funding for research into the dangerous disease.

Researchers from the Copenhagen Center for Glycomics have been investigating O-glycosylation in cells infected with EBOV in order to better understand the disease.

The study

HEK293 cells lacking O-linked glycan elongation (C1GALT1C1 KO) and cells deficient in individual isoforms of initiating enzymes (GALNT1 KO, GALNT2 KO and GALNT3 KO) were infected with EBOV in order to address the function of these O-glycans. When intracellular viral replication was observed on days 1 and 6 post-infection, the mutant cells showed significantly less viral replication.

The amounts of viral RNA recovered on day 6 reflected the same trend, with the deletion of the individual GALNT genes showing a larger effect than the total elimination of O-glycan elongation. Viral titers in cell culture supernatants were measured on days 1 and 6 in order to evaluate the infectivity of extracellular virions, revealing comparable titers on day 1, but a significant reduction in all four knock out cell lines by day 6.

Following this, both native virus-infected cells and those transfected with GP were investigated to assess EBOV glycoprotein (GP) expression in the knock-out cell lines. Immunofluorescence staining showed that the wild-type cells had clusters of infected cells with surface expression of GP. GALN2 KO cells often exhibited small lesions, but other than this the KO cells were largely competent for GP surface expression. Full-length EBOV GP was then expressed in the engineered cell lines, co-stained with E-cadherin in order to confirm surface expression in non-permeabilized cells. This revealed recombinant GP on the surface of all five cell lines – suggesting that alterations in O-glycosylation do not affect the trafficking of the glycoproteins.

EBOV virus-like particles (VLPs) presenting the GP on rod-shaped particles generated by co-expression with VP40 protein were analyzed in order to identify the O-linked glycosylation sites on the GP. As well as this, the full length His-tagged EBOV GP was expressed in both wild-type and all KO cell lines, in order to determine the individual contribution of the different GALNT proteins. After purification, digestion and labeling, sequential lectin weak affinity chromatography (LWAC) of desiaylated tryptic digets was used to enrich core O-glycan structures and their intermediates. This approach allowed the researchers to identify 47 O-glycosties, 45 of which were unambiguous. Most of the glycosites were located in the MLD and glycan cap agreements, and 32 were in common between the VLPs and GP.

This same process allowed for the enrichment of core 2 structures, which were found at many different glycosites – often coexisting on adjacent amino acids. Several areas of GP also showed no elongation taking place at all due to the density of the glycosylation. Seven glycosites were found on a single peptide, most of which were single GalNAc structures.

Again, the glycosites and site-specific structures were in agreement on both the GP and the VLPs, suggesting the identified glycoproteins are mature. Some peptides carried only a single O-glycan, and the relative abundance of these peptides allowed the individual contributions of GalNAc-T1, T2 and T3 to specific sites to be evaluated. As expected, all three GalNAc-Ts contributed to glycosylation of the EBOV GP, but GalNaC-T1 was the most active. A few glycosites received contributed from several different GalNAc-Ts, and non-regulated sites suggesting a different isoform was preferred, or glycosylation was non-selective at these sites.

Finally, wild-type EBOV GP-VSV was titrated on the panel of glycoengineered cells, with the luciferase reporter expression upon entry quantified to address GP-mediated entry to cells lacking O-glycan elongation. All mutant cells showed diminished entry, further suggesting host O-glycosylation could help prevent the propagation and reduce the infectivity of EBOV.

Conclusions

The authors have successfully shown that mutant cells with a lack of transferase enzymes to initiate O-glycan elongation are significantly harder for EBOV to infect, and that viral propagation is slower once these cells are infected. As well as this, they have strong evidence that alterations in O-glycosylation do not affect the movement of glycoproteins. Many different GP glycosites were identified, and the pattern of glycosylation was evaluated, as well as which GalNAc-Ts contributed to the glycosylation. This research could prove valuable for drug manufacturers and other Ebola virus researchers, and could help identify treatments for the disease.

*Important notice

bioRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.

Written by

Sam Hancock

Sam completed his MSci in Genetics at the University of Nottingham in 2019, fuelled initially by an interest in genetic ageing. As part of his degree, he also investigated the role of rnh genes in originless replication in archaea.