Download An RNA-binding domain in the viral haemorrhagic septicaemia virus

Journal
of General Virology (1998), 79, 47–50. Printed in Great Britain
..........................................................................................................................................................................................................
SHORT COMMUNICATION
An RNA-binding domain in the viral haemorrhagic septicaemia
virus nucleoprotein
Thania Said, Henriette Bruley, Annie Lamoureux and Michel Bre! mont
Unite! de Virologie et Immunologie Mole! culaires, Institut National de la Recherche Agronomique, 78352 Jouy-en-Josas Cedex, France
The gene encoding the nucleoprotein (N) and PCRderived subfragments from viral haemorrhagic septicaemia virus (VHSV), a salmonid rhabdovirus,
were overexpressed in Escherichia coli BL21(DE3)
transformed by recombinant expression vector
pET-14b containing N and PCR-generated subfragment cDNAs under the control of the T7 RNA
polymerase promoter. Following induction with
IPTG, recombinant His-tagged proteins were expressed, purified by affinity metal chelation chromatography under denaturing conditions and renatured. Protein blots were hybridized with various
radiolabelled nucleic acid probes. Results obtained
using genomic or messenger virus RNA as a probe
indicated that the middle part of N was possibly an
RNA-binding domain. To confirm this observation,
two more accurate approaches were undertaken : (i)
a gel retardation assay of RNA and purified protein
complexes was done ; and (ii) RNA–protein complexes were cross-linked by UV light and analysed
on a denaturing polyacrylamide gel. All these
experiments led us to conclude that the middle part
of N is the domain which interacts with RNA, despite
the absence of homology with known consensus
amino acid sequences of other RNA-binding proteins.
Viruses of the family Rhabdoviridae, identified as bulletshaped, enveloped viruses with a single-stranded RNA
genome of negative polarity, are frequently isolated from a
wide range of host fish (Wolf, 1988). Until recently, based on
the electrophoretic pattern of the structural polypeptides, they
were classified as members of either the genus Vesiculovirus,
e.g. spring viraemia of carp virus, or the genus Lyssavirus, e.g.
viral haemorrhagic septicaemia virus (VHSV) and infectious
haematopoietic necrosis virus (IHNV). VHSV and IHNV have
been the most extensively studied, due to their economic
impact. VHSV is one of the causative agents of an acute disease
Author for correspondence : Michel Bre! mont.
Fax ­33 1 34 65 26 21. e-mail bremont!biotec.jouy.inra.fr
in salmon and trout. The genomes of VHSV and IHNV encode
six proteins in the order 3« N-M1-M2-G-Nv-L 5« (Kurath et al.,
1985 ; Kurath & Leong, 1985 ; Benmansour et al., 1994). The
virion is composed of five structural proteins : a glycoprotein
(G) which is embedded in the virus membrane ; a matrix protein
(M2) ; a polymerase-associated protein (M1) ; a minor component, the polymerase (L) ; and the nucleoprotein (N). N,
which is tightly bound to the genomic negative-strand RNA,
is 404 amino acids long and represents approximately 800
molecules per virion (Leong et al., 1983). The gene encoding N
has been sequenced for the 07-71 (French ; pathogenic) (Bernard
et al., 1990) and Makah (Washington state, USA ; asymptomatic) (Bernard et al., 1992) isolates of VHSV. Comparison of
the deduced amino acid sequences shows 94±8 % similarity and
91±1 % identity between the two VHSV strains, and 62±1 % and
37±1 %, respectively, between VHSV and IHNV. The central
portion of this protein is the most highly conserved, suggesting
that this region has functional importance.
To map the RNA-binding domain on the VHSV N, the
corresponding gene and subfragments were amplified by PCR
from a cDNA clone (Bernard et al., 1992) using specific primers.
Primers were chosen in order to generate N subfragments
dividing the N gene into three overlapping fragments : A, the
5« end (1–471 nt) ; B, the middle part (442–771 nt) ; and C, the
3« end (673–1224 nt). The resulting DNA fragments were
subcloned into the blunt-ended NdeI site of the pET-14b
expression vector (Novagen). Following Escherichia coli
BL21(DE ) transformation (Studier & Moffatt, 1986), recom$
binant clones were screened by hybridization using a radiolabelled cDNA N probe and analysed for insert orientation by
restriction enzyme digestion of mini lysates. Using the pET14b expression vector, six His residues were fused to the
amino-terminal end of the recombinant proteins, enabling
purification by metal chelation affinity chromatography. Expression of recombinant proteins was induced by the addition
of IPTG to exponential phase cultures and aliquots were
analysed on a 12 % SDS–PAGE gel and visualized after
Coomassie-blue staining. As shown in Fig. 1 (a), proteins of the
expected molecular mass were synthesized at a very high level.
The yield was estimated at between 80 and 120 mg}l induced
bacteria.
To confirm the authenticity of the expressed recombinant
proteins, aliquots of induced recombinant clones separated on
0001-5070 # 1998 SGM
EH
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 13:42:06
T. Said and others
(a)
N
(b)
A
B
C
N
(c)
A
B
C
C
B
A
N
Fig. 1. Expression of N and A, B and C proteins in E. coli BL21(DE3) transformed by the respective recombinant pET constructs
and their reactivity and purification. (a) Following IPTG induction, aliquots were boiled in buffer described by Laemmli (1970)
and loaded on a 12 % SDS–PAGE gel, which was then stained with Coomassie blue. (b) North-Western blot analysis.
Recombinant E. coli cell lysates separated on a 12 % SDS–PAGE gel were transferred onto PVDF membranes and saturated as
described in the text. The blot was probed with [α-32P]UTP radiolabelled M1 RNA. (c) Coomassie blue-stained gel analysis of
the recombinant proteins purified by metal chelation chromatography.
a 12 % SDS–PAGE gel were electrotransferred onto ProBlott
membranes (Applied Biosystems) in 3-(cyclohexylamino)-1propanesulfonic acid buffer containing 10 % methanol, saturated for 2 h at room temperature in 3 % BSA in PBS (PBSA)
and then incubated with anti-VHSV polyclonal or anti-N
monoclonal antibodies diluted in PBSA containing 0±05 %
Tween 20. After washing, the bound antibodies were detected
using alkaline phosphatase-conjugated anti-mouse or antirabbit immunoglobulin (BYOSIS). 4-Nitro blue tetrazolium
chloride and 5-bromo-4-chloro-3-indolyl phosphate (GibcoBRL) were used as substrates for the alkaline phosphatase. N
and the A, B and C proteins were recognized by the 616 antiVHSV polyclonal antiserum, confirming the authenticity of the
recombinant proteins. When blots were incubated with anti-N
monoclonal antibodies, all eight monoclonal antibodies reacted
with N and either the A or C proteins, but none reacted with
the B part of N. It should be noted that this lack of reactivity
of specific monoclonal antibodies against the middle part of a
nucleoprotein has been described previously (Chen et al., 1991)
for IHNV, another fish rhabdovirus. It could be hypothesized
that this domain in the virus particle is masked and poorly
accessible to the immune system.
The first approach developed to detect binding between
the recombinant N-derived proteins and nucleic acids was
North-Western blotting, as first described by Boyle & Holmes
(1986). Recombinant E. coli lysates were separated on a 12 %
SDS–PAGE gel, transferred onto PVDF membranes and
blocked in PBSA for 1 h at room temperature. The incubation
medium was replaced by fresh PBSA containing 250 µg}ml
yeast tRNA and incubated for 30 min. Blots were then
incubated for 1 h in PBSA containing DNA or RNA probes at
10& c.p.m.}ml, obtained, respectively, by either in vitro transcription or random priming radiolabelling of cloned M1 and
EI
M2 VHSV genes. For the in vitro transcription reaction, a
pBluescript-M2 plasmid was either linearized by BsmI and used
to produce a 150 nt genomic sense RNA probe using the T3
RNA polymerase, or linearized by BsaBI and transcribed with
the T7 RNA polymerase to produce a 250 nt messenger sense
RNA probe. In both reaction mixtures, [α-$#P]UTP was
included at a final concentration of 30 µCi. The specific activity
of the probes was 7¬10( c.p.m.}µg. After four washes in
PBSA, blots were dried and autoradiographed. A typical result
obtained using RNA probes is shown in Fig. 1 (b), where
hybridization was conducted with the M2 RNA probe (the
same result was obtained with an M1 RNA probe). This result
indicates that only the B protein seems able to interact with
RNA. When DNA probes were employed, no proteins were
detected (data not shown). It should be noted that the B protein
interacts with the RNA in either the plus or minus sense. A
longer time of exposure was required to visualize the RNA–N
interaction. This is probably due to the less efficient blotting of
N versus the B protein and thus less N molecules are available
to bind the probe.
To ascertain that binding of the B protein and RNA
observed in the North-Western blotting experiment was not
an artifact, binding was assayed by electrophoretic gel
retardation in non-denaturing polyacrylamide gels. Recombinant proteins were first purified using His-bind resin columns
(Novagen). Recombinant clones were inoculated into 3 ml LB
medium containing 50 µg}ml carbenicillin, grown to
OD 0±6, then stored overnight at 4 °C. Cells were pelleted,
'!!
then resuspended in 50 ml fresh LB medium and agitated at
37 °C to OD 0±8. Expression of recombinant proteins was
'!!
induced with 0±4 mM IPTG after 3 h incubation at 37 °C. Cells
were pelleted at 5000 g for 5 min, resuspended in 12±5 ml cold
buffer A (50 mM Tris–HCl, pH 8 ; 2 mM EDTA), centrifuged
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 13:42:06
VHSV nucleoprotein RNA-binding domain
(a)
(a)
BSA
N added
N
A
C
B
Complex
RNase T1
Free
(b)
BSA
A
B
C
A
N
N
A
B
C
BS
C
on
tro
l
(b)
Monomer
Complex
Free
Fig. 2. Gel retardation analysis of VHSV RNA. (a) RNA binding to purified
N ; 20 fmol of 150 nt 32P-radiolabelled RNA was added to increasing
amounts of N and analysed on a non-denaturing polyacrylamide gel. (b)
Same as (a) using N and A, B and C proteins (6 pmol each). BSA was
also included as a negative control. The control lane contained RNA probe
incubated with binding buffer only.
and the final pellets were stored at ®70 °C. For purification,
cell pellets were resuspended in 20 ml buffer B (5 mM
imidazole ; 500 mM NaCl ; 20 mM Tris–HCl, pH 7±9) containing 200 units benzonase (Merck). After sonication, lysates
were centrifuged and the resulting pellets were resuspended
with 5 ml buffer B containing 6 M urea. Solubilization was
achieved by gentle agitation overnight at 4 °C. Solubilized
proteins were clarified by ultracentrifugation (100 000 g for
30 min) and supernatants were loaded onto His-bind resin
columns pre-equilibrated in buffer B containing 6 M urea.
Resins were washed and recombinant proteins were eluted
with 5 ml buffer C (400 mM imidazole ; 500 mM NaCl ; 20 mM
Tris–HCl, pH 7±9). Purified proteins were refolded by gradual
removal of the 6 M urea by a step-by-step dialysis (0±5 M urea
step). After the final dialysis against PBS, purified recombinant
Fig. 3. Analysis of RNA–protein complexes following UV light cross-linking.
Mixtures of RNA probe and proteins were irradiated with UV light for
various lengths of time, digested with RNase T1 and loaded on a 12 %
SDS–PAGE gel. The gel was then stained with Coomassie blue (a) and
autoradiographed (b).
proteins were stored at ®70 °C until use. Purity was estimated
on Coomassie blue-stained SDS–PAGE gels (Fig. 1 c).
For the gel retardation assay, purified proteins were mixed
with riboprobes in binding buffer (10 mM Tris–HCl, pH 8 ;
100 mM NaCl ; 1 mM EDTA) to give a total volume of 20 µl.
The mixtures were incubated at 4 °C for 20 min, after which
5 µl loading buffer (10 mM Tris–HCl, pH 8 ; 5 % glycerol ;
1 mM EDTA ; 0±05 % xylene cyanol ; 0±05 % bromophenol
blue) was added to each. The mixtures were then separated by
electrophoresis on 5 % non-denaturing polyacrylamide gels
(60 : 1 acrylamide : bisacrylamide) in 0±5¬ TBE (45 mM Tris–
borate ; 45 mM boric acid ; 1 mM EDTA) for 5 h at 50 V and
4 °C. Gels were then dried and exposed to X-ray films.
Conditions for binding were first optimized using N and the
genomic sense M2 riboprobe (20 fmol). Using low protein
concentrations (0±6–3 pmol corresponding to 28–140 ng), the
radioactive probe migrated as the protein-free RNA, but with
6 pmol (280 ng) of added N, the probe was retarded, indicating
the formation of a complex between the purified N and the
riboprobe (Fig. 2 a). The same experiment with N and the A, B
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 13:42:06
EJ
T. Said and others
and C proteins is presented in Fig. 2 (b) and it shows that N and
the B protein interact with the RNA in a similar manner. The
same experiment using the messenger sense M2 riboprobe
gave similar results (data not shown), except that part of the
probe migrated faster in lanes N and B. This increased mobility
is probably due to the large size (250 nt) of the riboprobe used
(Lane et al., 1992). Thus, these results are in agreement with
those obtained with the North-Western blot, confirming that
the B protein, the middle part of N, is able to complex the
RNA, whereas A and C proteins are not. This result also
indicates that the observed binding for N and the B protein was
not due to the six His residues present at the amino terminus
of each recombinant protein, and therefore that the B protein
could represent an RNA-binding domain.
To exclude the possibility that the interaction observed
between RNA and N or the B protein was due to contaminant
E. coli proteins, we developed a UV cross-linking assay in
which RNA–protein complexes were analysed on a denaturing
polyacrylamide gel. Mixtures of riboprobes and proteins as
described above were UV-irradiated for 5 or 10 min using a
UV General Electric USB (1±125 mW}cm#) ; complexes were
then digested with 50 U RNase T1 for 15 min at 37 °C.
RNA–protein complexes were run on 15 % SDS–PAGE gels,
which were then stained with Coomassie blue, dried and
autoradiographed. Superimposing the stained gel (Fig. 3 a) and
resulting autoradiograph (Fig. 3 b) clearly indicated that the B
protein is cross-linked to the RNA probe. Moreover, the
additional bands visualized on the autoradiograph could
correspond, by size estimation, to dimer and trimer forms of
the B protein and aggregates which are not resolved on that
gel. A longer period of UV irradiation of the samples resulted
in the appearance of N, but it was not visible for 5 min
irradiation (data not shown).
In the past, many studies on vesicular stomatitis virus or
rabies virus, the two rhabdovirus prototypes, have focused on
the interaction between the virus RNA and the nucleoprotein.
The specificity of this interaction requires the presence of the
phosphoprotein. In the absence of this protein, the nucleoprotein is able to bind any RNA (Masters & Banerjee, 1988). In
this study, we showed that a short region on VHSV N is
implicated in RNA binding, even in the absence of the M1
phosphoprotein. A comparison of N from two fish rhabdoviruses, VHSV and IHNV, showed 37 % identity at the amino
acid level, with the strongest identity between amino acids
localized in the B fragment and more precisely in the carboxyterminal half. This observation reinforces the importance of
this conserved central region in interaction with the virus
genomic RNA ; nevertheless, a search for homology with other
known RNA-binding proteins failed to show any significant
data. Detailed analysis of the B fragment sequence did not
reveal any special features such as stretches of basic amino
acids. Nevertheless, the presence of a repeat of ELAD motifs
has been detected, as well as the presence of 12 arginine and
lysine residues, but these observations cannot be related to any
FA
previously described signature for an RNA-binding domain.
Taking account of the fact that the three recombinant proteins
A, B and C produced in E. coli overlap, we may assume that the
RNA-binding domain is restricted to 68 amino acid residues
and could represent a new RNA-binding motif.
The authors are grateful to Dr J. M. Coll (CISA, Spain) for supplying
some of the anti-N monoclonal antibodies. Drs C. Aponte and D. Poncet
(INRA, France) are fully acknowledged for their valuable suggestions for
this study.
References
Benmansour, A., Paubert, G., Bernard, J. & de Kinkelin, P. (1994). The
polymerase-associated protein (P) and the matrix protein (M2) from a
virulent and an avirulent strain of viral hemorrhagic septicemia virus
(VHSV), a fish rhabdovirus. Virology 198, 602–612.
Bernard, J., Lecocq-Xhonneux, F., Rossius, M., Thiry, M. E. & de
Kinkelin, P. (1990). Cloning and sequencing the messenger RNA of the
N gene of viral haemorrhagic septicaemia virus. Journal of General
Virology 71, 1669–1674.
Bernard, J., Bremont, M. & Winton, J. (1992). Nucleocapsid gene
sequence of a North American isolate of viral haemorrhagic septicaemia
virus, a fish rhabdovirus. Journal of General Virology 73, 1011–1014.
Boyle, J. F. & Holmes, K. V. (1986). RNA-binding proteins of bovine
rotavirus. Journal of Virology 58, 561–568.
Chen, L., Ristow, S. & Leong, J. C. (1991). Epitope mapping and
characterization of the infectious hematopoietic necrosis virus nucleoprotein. In Proceedings of the Second International Symposium on Viruses of
Lower Vertebrates, pp. 65–72. Corvallis, Oreg. : Oregon State University
Press.
Kurath, G. & Leong, J. A. C. (1985). Characterization of infectious
hematopoietic necrosis virus mRNA species reveals a non virion
rhabdovirus protein. Journal of Virology 53, 462–468.
Kurath, G., Ahern, K. G., Pearson, G. D. & Leong, J. A. C. (1985).
Molecular cloning of the six mRNA species of infectious hematopoietic
necrosis virus, a fish rhabdovirus, and gene order determination by Rloop mapping. Journal of Virology 53, 469–476.
Laemmli, U. K. (1970). Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature 227, 680–685.
Lane, D., Prentki, P. & Chandler, M. (1992). Use of gel retardation to
analyze protein–nucleic acid interactions. Microbiological Reviews 56,
509–528.
Leong, J. C., Hsu, Y. L. & Engelking, H. M. (1983). Synthesis of the
structural proteins of infectious hematopoietic necrosis virus. In Bacterial
and Viral Diseases of Fish : Molecular Studies, pp. 61–71. Edited by J. H.
Crosa. Seattle : Washington Sea Grant Program.
Masters, P. S. & Banerjee, A. K. (1988). Complex formation with
vesicular stomatitis virus phosphoprotein NS prevents binding of
nucleocapsid protein N to nonspecific RNA. Journal of Virology 62,
2658–2664.
Studier, F. W. & Moffatt, B. A. (1986). Use of T7 RNA polymerase to
direct selective high-level expression of cloned genes. Journal of Molecular
Biology 189, 113–130.
Wolf, K. (1988). Fish Viruses and Fish Viral Diseases. New York : Cornell
University Press.
Received 7 July 1997 ; Accepted 9 September 1997
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 13:42:06