433
Addresses
Johns Hopkins University, Bloomberg School of Public Health,
615 North Wolfe Street, Baltimore, Maryland 21205, USA
e-mail: dgriffin@welchlink.welch.jhu.edu
Current Opinion in Microbiology 2001, 4:433–434.
1369-5274/00/$ – see front matter.
© 2001 Elsevier Science Ltd. All rights reserved.
Abbreviations
ASFV African swine fever virus
HS heparan sulfate
SA sialic acid
VZV varicella-zoster virus
Viruses were the first pathogens to have complete sequence
information available. At least one representative and, often,
many representatives of each virus family have been
sequenced. The genomes have revealed and continue to
reveal much about the predicted types of viral proteins
encoded, about virus evolution, conservation of genetic
information, efficient use of small coding capacities, and about
the significant consequences of apparently minor coding and
noncoding changes for virulence. However, as more and more
viruses have been sequenced, it has quickly become apparent
that, even for those with the smallest genomes, obtaining
the sequence is only a very first step toward understanding
virus-induced disease. The ‘postgenomic era’ of virology has
provided fascinating glimpses into the complexities of
virus–host cell and virus–host interactions.
One important lesson that is illustrated by studies of a
number of viruses, including rotaviruses, varicella-zoster virus
(VZV) and African swine fever virus (ASFV), is that investi-
gation of the functions of many viral proteins cannot rely
solely on studies of virus infection of tissue culture cells.
Receptors that are used in vitro are often not the ones used
in vivo. Genes that encode proteins not essential for repli-
cation in cultured cells are often necessary for replication in
host tissue, for spread or for causing disease in the host. Some
viruses, such as the arthropod-borne viruses illustrated in this
Host–microbe interactions: viruses section of Current Opinion
in Microbiology by the flaviviruses and ASFV, have more than
one host that must be successfully infected. Thus, studies
of virus–host interactions have increasingly moved to studies
of animal model systems or infections of natural hosts to
obtain information relevant to disease pathogenesis.
Virus–receptor interactions and virus entry
Strains of viruses that are studied in the laboratory are
selected for efficient replication in the tissue culture cells
in which they are propagated. The first step in this process
is the binding of the virus to the cell surface, often regarded
as the interaction between the virus and its cellular
receptor. However, this is often an electrostatic interaction
influenced by the presence of negatively charged molecules
on the cell surface and clusters of positively charged amino
acids on the virion surface. Glycosylation of cell surface
proteins and lipids, with the addition of glycosaminoglycans
and terminal sialic acid residues, contributes much of the
negative charge. Many viruses bind the glycosaminoglycan
heparan sulfate (HS), and this molecule has been identified
as the ‘receptor’ for viruses with very different in vivo cell
and tissue tropisms (e.g. many herpes viruses, Sindbis
virus, dengue virus, foot-and-mouth-disease virus, adeno-
associated virus, respiratory syncytial virus, papillomavirus
and vaccinia virus). It is now clear that for many of these
viruses, HS binding represents a tissue culture adaptation
with selection for mutations that increase the numbers of
positively charged amino acids on the virion surface. These
amino acid changes often decrease virulence in vivo. For
instance, strains of Sindbis virus that bind HS well
replicate optimally in tissue culture, but subcutaneous
infection of mice with the same strains leads to lower level
viremia and more rapid clearance from the circulation than
is observed for strains that do not bind HS and replicate
less well in vitro [1]. Thus, a decreased ability to bind HS
leads to more efficient viral production in vivo. As Ciarlet
and Estes (pp 435–441) point out, a similar situation may
exist for sialic acid (SA) binding by rotaviruses. In this case,
one of the viral surface proteins (VP8*) binds SA in strains
that grow easily in tissue culture, but interaction of a second
viral protein (VP5*) with the cell surface is necessary for
infection, and interaction with SA is not necessary for
rotavirus-induced diarrhea in experimental animals.
The need to interact with more than one cell surface
molecule, or the use of a co-receptor, is increasingly recognized
to be necessary for the entry of many viruses. The presence
or absence of a necessary coreceptor can help to explain
host range, cell and tissue tropism and varied efficiency of
infection. In addition, viruses such as VZV (Arvin,
pp 442–449) may require the participation of more than
one viral surface protein for efficient attachment and entry.
These molecules may also induce cell-signaling events
that are necessary for virus infection, as postulated for
some rotaviruses.
How viruses traverse the cell membrane to deliver their
genomes and initiate infection is an area of active study.
Host—microbe interactions: viruses
Interactions between viruses, cells and the host immune
response that underlie the pathogenesis of viral diseases
Editorial overview
Diane E Griffin