Understanding selectivity of metabolic labelling and click-targeting in multicellular environments as a route to tissue selective drug delivery.

Cancer cells generally exhibit higher metabolic demands relative to that of normal tissue cells. This offers great possibilities to exploit metabolic glycoengineering in combination with bio-orthogonal chemistry reactions to achieve tumour site-targeted therapeutic delivery. This work addresses the selectivity of metabolic glycan labelling in diseased (i.e., cancer) versus normal cells grown in a multicellular environment. Dibenzocylooctyne (DBCO)-bearing acetylated-d-mannosamine (Ac4ManNDBCO) was synthesised to metabolically label three different types of cell lines originating from the human lung tissues: A549 adenocarcinomic alveolar basal epithelial cells, MeT5A non-cancerous mesothelial cells, and MRC5 non-cancerous fibroblasts. These cell lines displayed different labelling sensitivity, which trended with their doubling time in the following order: A549 ≈ MeT5A > MRC5. The higher metabolic labelling efficiency inherently led to a higher extent of specific binding and accumulation of the clickable N3-conjugated gold nanoparticles (N3-AuNps, core diameter = 30 nm) in the DBCO-glycan modified A549 and MeT5A cells, but to a less prominent effect in MRC5 cells. These findings demonstrate that relative rates of cell metabolism can be exploited using metabolic labelling to recruit nanotherapeutics whilst minimising non-specific targeting of surrounding tissues.

A single residue substitution in the receptor-binding domain of H5N1 hemagglutinin is critical for packaging into pseudotyped lentiviral particles.

BACKGROUND: Serological studies for influenza infection and vaccine response often involve microneutralization and hemagglutination inhibition assays to evaluate neutralizing antibodies against human and avian influenza viruses, including H5N1. We have previously characterized lentiviral particles pseudotyped with H5-HA (H5pp) and validated an H5pp-based assay as a safe alternative for high-throughput serological studies in BSL-2 facilities. Here we show that H5-HAs from different clades do not always give rise to efficient production of H5pp and the underlying mechanisms are addressed. METHODOLOGY/FINDINGS: We have carried out mutational analysis to delineate the molecular determinants responsible for efficient packaging of HA from A/Cambodia/40808/2005 (H5Cam) and A/Anhui/1/2005 (H5Anh) into H5pp. Our results demonstrate that a single A134V mutation in the 130-loop of the receptor binding domain is sufficient to render H5Anh the ability to generate H5Anh-pp efficiently, whereas the reverse V134A mutation greatly hampers production of H5Cam-pp. Although protein expression in total cell lysates is similar for H5Anh and H5Cam, cell surface expression of H5Cam is detected at a significantly higher level than that of H5Anh. We further demonstrate by several independent lines of evidence that the behaviour of H5Anh can be explained by a stronger binding to sialic acid receptors implicating residue 134. CONCLUSIONS: We have identified a single A134V mutation as the molecular determinant in H5-HA for efficient incorporation into H5pp envelope and delineated the underlying mechanism. The reduced binding to sialic acid receptors as a result of the A134V mutation not only exerts a critical influence in pseudotyping efficiency of H5-HA, but has also an impact at the whole virus level. Because A134V substitution has been reported as a naturally occurring mutation in human host, our results may have implications for the understanding of human host adaptation of avian influenza H5N1 viruses.