Virology tidbits: EBOV VP40 and Vesiculation: role of Phosphatidylserine

Ebola Virus (EBOV) is a filamentous negative strand ssRNA virus of about 19kb in size causing viral hemorrhagic fever in humans with a fatality rate of 60-90% transmitted by contact with infected animals, particular via the consumption of bush meat and maybe contaminated fruits as well as body fluids from patients. Although experimental drugs (e.g. ZMapp) and vaccines are available, current treatment focuses on treating the symptoms as well as preventing the spread of the disease by contact tracing and quarantining infected individuals rather than antiviral treatment. In general, outbreaks are sporadic and rare with the current outbreak in West Africa numbering an unprecedented excess of 17000 total reported cases with more than 6000 deaths, thus placing a significant burden on the healthcare system as well as on the economy in the affected countries.

EBOV in West Africa: data and geographical distribution as of Dec 05 2014

The genome of EBOV encodes for seven proteins with four proteins that are required for transcription and replication (NP; Nucleoprotein, VP35; Polymerase co-factor, VP30; Transcription activator, L; RNA dependent RNA Polymerase) and three structural proteins (GP; Glycoprotein, VP40; Membrane protein, VP24; Membrane protein), with the GP being required for viral entry and fusion with the endosome using the Niemann-Pick Type C1 cholesterol transporter as discussed in a prior posting.

Genome structure of EBOV and MARV two related but different Filoviruses

Localisation of the structural proteins within the EBOV virion

In contrast to viral entry, the process of viral assembly and viral egress is less well understood, but based on a number of studies identified the VP40 as a key protein involved in facilitating the formation of vesicular structures originating from the inner leaflet of the plasma membrane. Indeed, the expression of VP40 induces the formation of virus-like particles in the absence of other viral proteins and the expression of both GP and VP40 has been shown to produce filamentous particles which are highly immunogenic. EBOV VP40 is expressed as a 360 AA long protein with an N terminal domain required for dimerization and a C-terminal domain that binds to the inner leaflet of the plasma membrane. In addition to allow dimerization, the N terminal domain is also required to mediate binding to the cellular endosomal complex via Tsg101, and thus targeting to the plasma membrane, via a PTAP and PPxY motif, utilising the cellular ESCRT and ubiquitination machinery.

EBOV VP40 induced budding of vesicles derived from artificial membranes has been shown to be dependent on a hydrophobic patch consisting of four amino acids (L213, I293, L295, and V298) within the C-terminal domain which interacts with the inner leaflet of the (plasma) membrane, followed by oligomerization of the monomer and subsequent budding of the VLP.

Domains of VP40

Following the transfection of HEK-293T and CHOK-1 cells with a plasmid allowing the expression of wt EBOV VP40, EBOV VP40 localises to the plasma membrane with prominent extensions of the plasma membrane representing VLPs. In contrast, the expression of mutant VP40 harbouring mutations within the assembly domain (AA 60-185), in particular K127A, T129A, or N130A mutations, exhibit a predominantly cytoplasmic localisation with no filamentous extensions of the plasma membrane, indicating that the assembly domain is necessary for targeting VP40 to the plasma membrane and inducing the formation of filamentous VLPs. Furthermore, the expression of those mutant VP40 constructs also reduces viral egress as well as oligomerization of VP40 both in HEK-293T and CHOK-1 cells. Therefore the presence of hydrophobic residues within the central assembly domain is important for both dimerization and the formation of VLP whereas the hydrophobic patch within the C-terminal domain may anchor VP40 inside the inner leaflet. To my knowledge, the individual contribution of the central domain and the C-terminal domain remains to be investigated, but I can envision a model where the dimerization of VP40 is necessary for not only inducing vesiculation but also stabilising the interaction of VP40 with membrane lipids whereas the C-terminal domain is responsible for the (initial) targeting VP40 to the plasma membrane followed by conformational change allowing the formation of oligomers, a view which is supported by findings that the expression of mutant lacking the C-terminal domain does prevent the localisation of VP40 to the PM whilst leading to the accumulation of VP40 hexamers in the cytoplasm. As always however I am happy to revise my opinion in this matter.

The question remains then on how the C-terminal domain of VP40 is targeted to the PM and inserted into the inner leaflet. Whilst the latter process is straight forward -mediated by the interaction of the hydrophobic patch with membrane lipids- the targeting of VP40 to lipids is or was largely unknown.

In general, many cellular protein such as GTPases that are attached to the PM post-translationally modified by four major types of lipid modification, (1) acylation (addition of saturated fatty acids), (2) prenylation (addition of polyunsaturated isoprenoid groups), (3) esterification (addition of cholesterol), and (4) conjugation (addition of glycosylphosphatidylinositol (GPI)). In the case of cellular proteins targeted to the inner leaflet of the PM, the most common modification is acylation by the addition of palmitoyl or myristoyl groups to the protein. In the case of EBOV VP40 however no acylation signal is present, indicating that VP40 is not post-translationally modified but interacts with lipids directly, probably via the C-terminal domain. In order to identify the lipid that interacts with VP40 and characterise the budding process giant unilamellar vesicles (GUVs) -unilamellar liposomes with an average diameter up to 100 μm- with varying compositions of phospholipids were mixed with varying concentrations of purified VP40 and imaged. Intraluminal vesicles containing VP40 were only formed if the GUV contained at least 5% of Phosphatidylserine, independent of the presence of Cholesterol as well as the anionic charge since the formation of VP40 positive vesicles cannot be observed in GUVs containing PI, PI(4,5)P2, or PIP3. Interestingly, an antibody targeting Phosphatidylserine, PGN401, has been shown to bind both to EBOV infected Vero E6 cells as well as purified EBOV viral particles. Since in non-infected/non-apoptotic cells Phosphatidylserine is confined to the inner leaflet of the membrane, these results suggest that the infection of cells with EBOV induces the exposure of Phosphatidylserine to the outer membrane and thus might phagocytes to infected cells. EBOV virions containing Phosphatidylserine might also be able to bind Gas6 and thus bind Axl, a receptor tyrosine kinase expressed on target cells such as phagocytes and part of a group of receptors known as Phosphatidylserine -mediated virus entry enhancing receptors (PVEERs).

EBOV therefore might be able to infect those cells through a process termed as “apoptotic mimicry”, which might explain why a bona-fide receptor for EBOV has not been identified. If however the expression of VP40 is sufficient to induce Phosphatidylserine exposure to the cell surface and the formation of Phosphatidylserine positive VLPs - this remains to be seen.

Virology tidbits

Virology tidbits

Monday, 8 December 2014

EBOV VP40 and Vesiculation: role of Phosphatidylserine

No comments:

Post a Comment