Eukaryote cells do not have cell walls; they have only lipid bilayer cell membranes (=PLASMALEMMA ) with associated embedded proteins. They also have a very highly developed INTERNAL MEMBRANE COMPLEX, consisting of endoplasmic reticulum, Golgi apparatus, nuclear membrane, and various specialised vesicles such as LYSOSOMES, which are involved in intracellular digestion. Eukaryote cells have an intricate system of vesicle transport centred on the Golgi apparatus; this involves export of protein(s) and vesicles from the ER to the Golgi; production of EXPORT vesicles containing proteins from this for fusion with the cell membrane (=EXOCYTOSIS); production of LYSOSOMES to fuse with ENDOSOMES for digestion of material internalised by RECEPTOR-MEDIATED or non-specific ENDOCYTOSIS (PHAGOCYTOSIS for particulates; PINOCYTOSIS for liquid). Other CYTOSKELETON-DIRECTED vesicle trafficking involves targetting of vesicles back to the Golgi and to the NUCLEAR MEMBRANE.
The initially-formed ENDOCYTOTIC vesicles resulting from receptor-mediated endocytosis are coated with CLATHRIN, a structural protein which promotes curvature of the membrane at COATED PITS - which become vesicles as the protein continues to assemble and impresses the membrane into a sphere. Clathrin is recycled back to the cell membrane after endocytosis, either because of interaction of the vesicle with a lysosome, or due to another process.
 |
 |
Viruses take advantage of these
processes in a number of ways, after they have attached to the cell surface via
binding to a
specific receptor. The simplest is DIRECT MEMBRANE FUSION, where the virion membrane fuses with the cell membrane, and the virion
nucleoprotein complex is delivered into the cell cytoplasm directly. This
is generally a pH-independent
process, and requires only that the
membrane be fluid (ie: temperature in the physiological range), and
generally that some divalent cations be present.
The entry process for HIV is shown in the graphic (from the Univ Leicester material, copyright Dr AJ Cann).
Click here for an non-animated view of the process if your browser does not support GIF animation.
Here the virion attachment protein - gp120 - (see also here) attaches initially to the CD4 protein on a helper T-cell. The gp120 undergoes conformational change due to binding, and binds the accessory receptor - CCR-5, a chemokine receptor in this case, although there are others - as well. gp41 - a cleavage product of a gp160 precursor, and a part of the "spike protein" of the viral membrane - is then able to bind into the cell membrane, via a hydrophobic domain. A condensation of the gp41 structure - formation of a "6 helix bundle" - causes close juxtaposition of cell and viral membranes, which promotes membrane fusion and nucleoprotein entry into the cell.
The entire HIV life cycle is shown here: graphics kindly provided by Russell Kightley.
 |
 |
 |
|
HIV virion: outer view |
HIV virion structure: |
HIV cell entry and replication |
Another depiction of the infection process is given on the Roche site.
PARAMYXOVIRUSES are also able to enter cells in this fashion; in fact, an F or fusion (glyco)protein purified from enveloped virions of Sendai virus is often used to artificially fuse cells in in vitro experiments.
HERPESVIRUSES also are able to enter cells by direct cell membrane fusion, and to cause cell fusion (multinucleate giant cells or syncytia)
Another means of cell entry via membrane interactions is typified by influenza virus: this attaches to cell membranes via its HAEMAGGLUTININ (HA) protein - a trimeric attachment protein - which binds NEURAMINIC ACID residues on cellular glycoproteins. This and the neuraminidase (NA) protein - which exists as as a tetramer, present to about 20% of total envelope protein - contact the MATRIX (M1) protein which constitutes the internal shell. This in turn interacts with the viral nucleoprotein (vRNP) complex.
 |
 |
 |
| Diagram showing components of the influenza virion | Cutaway model of the influenza virion | Haemagglutinin trimer |
The virion binds reversibly to the cell surface via one or a few attachment proteins. Receptor consolidation - recruitment of more receptors, resulting in more binding - results in curvature of the membrane around the particle, and eventual formation of an endocytotic vesicle by a "zipper" process.
Click here for a view of the process - from Dr Sean Heaphy, Univ Leicester.
A cell surface protease cleavage divides the HA into HA1 and HA2: this causes a conformational change which exposes a hydrophobic "fusion domain" on HA2. A further pH-induced structural change (with lysosome fusion, pH drops), allows insertion of part of HA2 into the vesicle membrane. Another conformational change caused by membrane interaction results in "apposition" of membranes and fusion, allowing escape from the vesicle of the viral nucleoprotein, into the host cell cytoplasm. This is generally known is pH-dependent entry. Another version of the process, depicting the whole influenza virus life cycle, is shown below left (picture derived from original by Russell Kightley).
The M2 membrane protein - present at a low copy number of <100 in the influenza virus membrane - is a homotetramer which, at low pH, is the smallest ion channel protein yet discovered, making it a VIROPORIN. This is activated in the low-pH environment of an endosome, and allows the passage of protons (H+) into the virion: this lowers the pH, causes conformational changes in M1 which disrupt interactions between it and the vRNP, which allows its release into the cytoplasm free of the encapsulating M1 protein matrix. The vRNP then moves to the nucleus - unusual for an RNA virus - by means of NUCLEAR LOCALISATION SIGNALS in the NP contacting the cell transport machinery (see also here).
All enveloped viruses appear to share the fusion mode of entry, whether they fuse with the cell membrane directly or with the membrane around an internalised vesicle. This is mediated by three identified classes of envelope glycoproteins, which nevertheless share a very similar mechanism for inducing fusion.
Class I fusion proteins are found in retroviruses (HIV, leukaemia viruses), myxoviruses (influenza) coronaviruses (SARSCoV) and paramyxoviruses (mumps, measles), among others. The "spikes" are composed of three identical protein subunits, largely alpha-helical in structure, assembled as trimers, with subunits generated from a precursor that is cleaved into two pieces. The carboxy-terminus of one piece is anchored to the viral membrane; the new amino terminus has a characteristic stretch of 20 hydrophobic amino acids: this is the fusion peptide. Class I proteins all have a trimeric helical coiled-coil rod adjacent to the fusion peptide: this may act as a template for the refolding of protein segments during fusion, when the "six helix bundle" forms.
Interfering with this process is the basis of a highly successful new class of HIV therapeutics: entry inhibitors. See here for a movie on how one particular product - T20 or Fuzeon, from Trimeris - works. Another view of HIV chemotherapy in general can be seen here.
See here for an excellent short movie on how HIV enters its host cells.
Class II fusion proteins have distinctly different structural features: they predominantly have a β-sheet-type structure and are not cleaved during biosynthesis; the "fusion peptide" portion that inserts into the target membrane is thought to be an internal hydrophobic fusion loop. They are found in dengue, tick-borne encephalitis, yellow fever and other flaviviruses (see also here), and Semliki Forest virus and other alphaviruses, among others.
The proteins have three principal domains:
domain I begins at the amino terminus,
domain II contains the internal fusion loop, and
domain III is at the carboxy terminus.
The dimeric protein binds to one or a few cellular receptors; receptor binding consolidates and the virus is internalised. The acidic pH inside endosomes causes domain II to swing upward, allowing monomers to rearrange laterally. The fusion loop inserts into the host-cell membrane, enabling trimer formation of the viral glycoprotein. Domain III shifts and rotates to create contacts, bending the membrane. The formation of further contacts leads to unrestricted hemifusion and the final, most stable form of the protein. As with Class I proteins, several viral envelope spikes are involved in the formation of a single fusion pore.
See here for a PowerPoint presentation depiction of the process for Class I and Class II proteins (UCT MCB3024S only).
A graphic depiction of the process of membrane fusion is shown below.
Reprinted by permission from Macmillan Publishers Ltd:
Jardetzky TS and Lamb RA,
A
Class Act, Nature 427: 307-308, copyright 2004
Class III fusion proteins also form trimers of hairpins as a fusion structure by combining two structural elements. These proteins are characteristic of rhabdoviruses like rabies and vesicular stomatitis viruses, and herpesvirus gB attachment glycoproteins. Similar to Class I proteins, the post-fusion trimer has a central α-helical trimeric core; however, the fusion domains have two fusion loops at the tip of an elongated β-sheet, similar to Class II fusion proteins.
The following graphic clearly shows the
three different forms:
copied with permission from Weissenhorn W, Hinz A, Gaudin Y. Virus membrane fusion. FEBS Lett. 2007 May 22;581(11):2150-5
 |
|
Ribbon diagrams of representative structures of class I, II and III fusion proteins in their proposed post-fusion conformations. The positions of both membrane anchors at the tip of the elongated structures are indicated by black (fusion peptide, fp) and red (transmembrane, TM) arrows. (A) HIV-1 gp41 core structure; (B) Flavivirus fusion protein E and (C) VSV glycoprotein G. |
Most non-enveloped viruses, such as the complex dsDNA adenoviruses (Adenoviridae) and the simpler ssRNA picornaviruses (Picornaviridae), also enter cells via vesicles. The former appear to enter via an endocytotic vesicle, then to interact with the vesicle membrane - probably because of structural alterations due to the pH shift - so as to expel a partially-uncoated viral core structure into the cytoplasm. This is then targetted to the nucleus, where the DNA appears to enter via a nuclear pore.
Picornaviruses do something similar: that is, the capsid becomes rearranged as a result of pH-and/or receptor binding induced structural transitions, whereby the VP4 internal protein is externalised and the virion surface becomes more lipophilic, and interacts with a vesicle membrane to form a pore so as to allow exit of the RNA into the cytoplasm. A more complex process may also occur which allows RNA to enter the cell from externally-bound virions. Recent developments have allowed researchers to track picornavirus RNA and capsid separately following cell entry.
These mechanisms are summarised in the diagram below left: click here or on the box below to link to the source page.
Note that "naked" particle entry is probably an artifact of very high multiplicity of infection used when investigating the phenomenon of cell entry in tissue-cultured cells.
Reoviridae - isometric dsRNA non-enveloped viruses with up to 12 genome components - enter by endocytosis, are partially uncoated in lysosomes, and then core particles enter the cytoplasm by an unknown process, perhaps similar to what occurs with adenoviruses. Core particles then function as virus transcription "factories"
SEE ALSO the animated graphics of viral entry for individual families of viruses from the University of Calgary site mirror
A basic requirement in the infection of Eukarya would be that
DNA genomes end up in the nucleus - except for Pox-, Phyco- and some
Baculoviridae - and that RNA genomes end up in the
cytoplasm - except for myxoviruses - as these
are the sites where the respective viruses may be expected to replicate their
genomes. DNA is its own nuclear targetting signal in that naked DNA is
seen to be moved into cell nuclei if introduced into cells; however, most DNA
viruses which require a nuclear replication stage seem to have some specific
means of targetting partially-disassembled virions to the cell
nucleus.
Nucleoproteins with nuclear
localisation signals (NLS) are
very often involved.
Adenoviruses - large naked isometric virions with linear dsDNA of 30-38kbp - appear to be transported by means of the hexon protein of the partially degraded naked capsid that is released into the cytoplasm, interacting with microtubules. This reaches a nuclear pore and allows escape of viral DNA plus certain viral polypeptides into the nucleus.
A relatively recent finding (in 2001)
was that adenovirus 2 (Ad2)
docks to the nuclear pore after microtubule transport via the
nuclear-pore complex (NPC) receptor CAN/Nup214. The virus then
disassembles slowly and transfers its genetic material through the
nuclear pore while remaining attached to the NPC. The process does not
occur, however, without the linker histone H1 protein. H1 probably binds a surface-exposed acidic
amino acid cluster on hexon proteins, thereby probably destabilising the
capsid and triggering viral disassembly. In normal cells, H1 that leaks
out of the nucleus is retrieved from the cytoplasm by the proteins importin
 and importin 7: the disassembly of
Ad2 requires the same two importins, leading to
speculation that hexon-bound H1 is recognised by importin
and importin 7, and the complex is brought
partially into the
nucleus, thus triggering a gradual disassembly of the virus particle
at the NPC. This allows the DNA that exits particles to directly enter the nucleus.
(2001) Virus knocking at nucleus door. Nature Medicine 7, 1284 |
|
 |
The photographs show
two human lung carcinoma cells, one of which was injected with
anti-histone H1 antibodies. The injected cell can be seen in
the photograph on the left, stained with an injection marker
present in the antibody solution (purple). The photograph on
the right shows the two cells stained with antibodies to the
NPC (green) and Ad2 DNA (red). In the uninjected cell, Ad2 DNA
can be seen bound to NPC on the nuclear membrane but also
inside the nucleus; in the cell injected with anti-H1
antibodies all Ad2 DNA remains bound to the NPC. Reproduced with permission from the Nature Publishing Group; Rightslink License Number 2038130651955 |
Parvoviruses - small naked isometric viruses with linear ~5 kb ssDNA genomes - enter host cells by receptor-mediated endocytosis, escape from endosomal vesicles to the cytoplasm, and then replicate their DNA in the nucleus. Canine parvovirus virions at least use the motor protein dynein to be transported along microtubules to the nucleus.
Herpesviruses generally enter by
fusion with the cell membrane - by a process that involves
several envelope
glycoproteins acting in concert - and the core particles migrate to nuclear pores,
and release DNA there.
Poxviruses - which have "intracellular" single-enveloped and "extracellular" double-enveloped forms - may enter by direct cell fusion (pH-independent) or lysosomal vesicle fusion (pH-dependent and inhibitable by lysomotropic agents), with the latter perhaps predominating. Once core virions are in the cytoplasm, they uncoat further to expose a nucleoprotein complex which is first transcriptionally and later, replicationally, active.
|
Animal
Cell |
Copyright Ed Rybicki, November 1997, April 1998, June 1998,
March 1999, February 2001; September 2003; April 2008
(Unless otherwise stated)
