 |
|
 |
|
|
|
Bluetongue virus (BTV) is an orbivirus of the Reoviridae family. It infects sheep and cattle and, if not treated, can lead to mortality rates in excess of 70% in sheep. The viral genome of BTV codes for 2 outer capsid proteins, 5 core proteins and three nonstructural proteins. The structure of the (empty) viral capsid has been determined [J.M. Grimes, J.N. Burroughs, P.Gouet, J.M. Diprose, R. Malby, S. Zientara, P.P. Mertens, D.I. Stuart (1998) Nature, 395, 470]. Our own interest is specifically in the non-structural proteins and generally in the mechanism by which the various genome segments are segregated and sorted prior to encapsidation.
NS1 forms tubules in infected cells although for different viruses the
structure of the tubules differs. For BTV the RNA segment encoding this
protein is transcribed at the highest rate such that NS1 comprises roughly
25% of total viral protein. The structure of the tubules to 40Å resolution
have been described using cryo-electronmicroscopy. The function of the
523Å diameter tubules is unknown . NS2, which has been implicated
in morphogenisis of BTV, seems to be located at the surface of viral inclusion
bodies of infected cells. The protein has been shown to have non-specific
affinity for ssRNA and in this respect is similar to the s-NS protein from
the ubiquitous orthoreovirus. It might be that NS2 is involved in the selection
and condensation of the 10 mRNA species (corresponding to each of the 10
dsRNA segments of the viral RNA). NS2 has been expressed to high levels
in baculovirus in insect cells, the protein having 357 amino acids
and a molecular weight of 41kD. It has been shown that a fragment without
the N terminal 92 amino acids does not bind ssRNA whereas C-terminal deletions
of 49 and 130aa still bind ssRNA. NS2 is a phosphoprotein, phosphorylated
at two serine residues, although phosphorylation appears unimportant for
ssRNA binding. The tenth, and smallest, segment of the bluetongue virus
genome codes for two closely related proteins NS3(229aa) and NS3A(216aa)
whose functions are again poorly understood but are related to viral release
from infected cells. The proteins contain two hydrophobic regions which
are thought to be membrane spanning and the proteins have been located
in the Golgi complex of infected cells.
| Full length NS2 appears heavily agregated, but crystals (see above) appeared after a while and were shown to be a proteolytic product, namely the N-terminal domain. This domain when expressed either in E. coli or in baculovirus infected insect cells is totally insoluble. Well diffracting crystals where obtained by cloning a TEV cleavage site between the N and C terminal domains and by controlled preoeolysis with TEV protease crsyatls diffracting to sufficiently high resolution could be obtained and the structure solved (see right) using selenomethionine substituted protein. |  |
 |
In the crystal the molcules form an infinite spiral (the surface representation is shown on the left). The dimer in the asymmetric unit has an NCS 2-fold perpendicular to the crystallographic 6 fold screw axis. The molecules form the spiral with two important interfaces (see above). The first is formed by the extension of two beta sheets (red and yellow molecules) and the second involves the terminal extension (yellow and blue molecules). The structure explains the insolubiulity of the construct as well as the reason why the protein forms viral inclusion bodies in vivo. However the C-terminal domain, which would be situated inside the spiral for the full-length protein would not, for packing reasons, allow the spiral to close. This would agree with measurements of the sedimentation coefficient for the full-length protein |
Author: Paul Tucker
Last Edited: 10th November 2000