Characterization of ORF2 and its encoded protein of the Helicoverpa armigera nucleopolyhedrovirus.

Abstract

The open reading frame 2 (ha2) of the Helicoverpa armigera single nucleocapsid nucleopolyhedrovirus (HaSNPV), a conserved gene in most baculoviruses from lepidopteran insects such as p78/83 of the Autographa californica MNPV, was characterized. It is 1,242 bp long and potentially encodes a 45.9 kDa. Ha2 is conserved among baculoviruses from lepidopteran insects. Ha2 transcripts were detected from 16 to 96 h post infection (hpi) of HzAM1 cells. Rabbit polyclonal antiserum against a GST-HA2 fusion protein reacted with three protein of 50, 46 and 35 kDa at 24-72 hpi of HzAM1 cells. Anti OpMNPV ORF2 (homologue of HA2) antibody reacted only with the 46 and 35 kDa proteins in HaSNPV-infected cells. These results demonstrate that Ha2 is modified at the mRNA or protein levels. Western blot analysis showed that only the 50 kDa product of HA2 is a structural component of proteins of both the budded virus (BV) and occlusion-derived virus (ODV) phenotypes. HA2-EGFP fusion protein showed that HA2 is localized primarily in the nucleus of HzAM1 infected cells. The HA2 was found to co-localize with actin by labelling of actin with Rhordamine-Phalloidin. In summary, the data indicated that HA2 is a structural protein and interacts with host cell actin.

Full text

Page 1

Virus Research 116 (2006) 129–135
Characterization of ORF2 and its encoded protein of the
Helicoverpa armigera nucleopolyhedrovirus
Yingchao Niea,b,1, Qian Wanga,c,1, Changyong Lianga,c,
Minggang Fanga, Zehua Yub, Xinwen Chena,∗
aState Key Lab of Virology, Wuhan Institute of Virology, Chinese Academy of Sciences, Wuhan 430071, PR China
bInstitution of Entomology, Central China Normal University, Wuhan 430070, PR China
cGraduate School of the Chinese Academy of Sciences, Beijing 100039, PR China
Received 9 June 2005; received in revised form 25 September 2005; accepted 25 September 2005
Available online 24 October 2005
Abstract
The open reading frame 2 (ha2) of the Helicoverpa armigera single nucleocapsid nucleopolyhedrovirus (HaSNPV), a conserved gene in most
baculoviruses from lepidopteran insects such as p78/83 of the Autographa californica MNPV, was characterized. It is 1242bp long and potentially
encodes a 45.9kDa. Ha2 is conserved among baculoviruses from lepidopteran insects. Ha2 transcripts were detected from 16 to 96h post infection
(hpi) of HzAM1 cells. Rabbit polyclonal antiserum against a GST-HA2 fusion protein reacted with three protein of 50, 46 and 35kDa at 24–72hpi
of HzAM1 cells. Anti OpMNPV ORF2 (homologue of HA2) antibody reacted only with the 46 and 35kDa proteins in HaSNPV-infected cells.
These results demonstrate that Ha2 is modified at the mRNA or protein levels. Western blot analysis showed that only the 50kDa product of HA2
is a structural component of proteins of both the budded virus (BV) and occlusion-derived virus (ODV) phenotypes. HA2-EGFP fusion protein
showed that HA2 is localized primarily in the nucleus of HzAM1 infected cells. The HA2 was found to co-localize with actin by labelling of actin
with Rhordamine-Phalloidin. In summary, the data indicated that HA2 is a structural protein and interacts with host cell actin.
© 2005 Elsevier B.V. All rights reserved.
Keywords: HaSNPV; p78/83; Ha2; WASP; Baculovirus; Structural protein
1. Introduction
The Baculoviridae is a large family of diverse arthropod
specific insect viruses with large, circular, supercoiled, double-
stranded DNA genomes, and occluded, rod-shaped virions. It
consists of two genera: Granulovirus (GV) and Nucleopolyhe-
drovirus (NPV), the latter of which are sub-divided into Group I
and Group II based on phylogenetic analysis (Chen et al., 1999;
Herniou et al., 2001). There are two phenotypes of virion gen-
erated during the infection of baculoviruses: budded virus (BV)
and occluded virus (ODV). ODV initiates the infection through
oralingestionofvirus-contaminatedfoodandBVspreadsinfec-
tion within larvae to different cells and tissues.
∗Corresponding author. Tel.: +86 27 87199106; fax: +86 27 87199106.
E-mail address: chenxw@pentium.whiov.ac.cn (X. Chen).
1They contributed equally to this paper.
Release of BV nucleocapsids from endosomes into the cyto-
plasm induces the formation of thick actin cables (Charlton and
Volkman,1993).Thesecablesaredetectedonlytransientlyfrom
1 to 4h post infection (hpi), during which time the entire viral
nucleocapsidtravelstoandentersthenucleus(Granados,1980).
During this period, viral nucleocapsids are often co-localized
with one end of an actin cable while the other end is oriented
away from the virion (Charlton and Volkman, 1993). It was
shown that nucleocapsids of the Autographa californica nucle-
opolyhedrovirus (AcMNPV) are capable of nucleating actin
polymerization in vitro and two viral capsid proteins, P39 and
P78/83, were found to bind actin directly (Lanier and Volkman,
1998). The protein P78/83, a nucleocapsid-associated phospho-
protein (Russell et al., 1997), shares the WH2 and acidic motif
with mammalian Wiskott-Aldrich syndrome protein (WASP)
family, although the overall arrangement of the WH2 and acidic
motifsinP78/83homologissomehowdifferentfromthatoftheir
counterparts in WASP-family proteins (Machesky et al., 2001).
Therefore, the conservation between P78/83 homologues and
0168-1702/$ – see front matter © 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.virusres.2005.09.007

Page 2

130
Y. Nie et al. / Virus Research 116 (2006) 129–135
WASP-familyproteinssuggeststhatbaculovirusesmightusethe
WASP-Arp2/3pathwaytoachievenucleo-cytoplasmictransport
and assembly of nucleocapsids.
There are intriguing evidences that host cytoskeletal pro-
teins are involved in viral egress and assembly. Sodeik et al.
(1997) provided compelling evidence that herpes virus simplex
1 parental nucleocapsids use microtubules for transport to the
nucleus, and Cudmore et al. (1995) demonstrated that vaccinia
virus progeny induce and use the vectoral polymerization of
actin to bud from cells. It has become apparent that numerous
viruses interact with actin at various stages throughout their life
cycle,bothdisruptingandrearrangingactincytoskeletontotheir
advantage (Cudmore et al., 1997). P78/83 is conserved in most
baculovirusesfromlepidopteraninsectssuggestingthatitmaybe
involved in some basic process of viral life cycle. Even charac-
terization of p78/83 homologies has been carried out in group
I baculoviruses, they are never characterized in group II bac-
uloviruses.
A p78/83 homology, open reading frame 2 (ha2), is found
in Helicoverpa armigera single nucleocapsid nucleopolyhe-
drovirus (HaSNPV) of the group II NPV (Chen et al., 2001).
Its function remains elusive. In this report, we tried to char-
acterize ha2 by transcriptional analysis, protein identification
and localization. RT-PCR analysis indicated that ha2 was tran-
scripted at the late stage of the infection. Three products with
the sizes of 50, 46 and 35kDa were detected in infected cells,
implied that the different post-translation modifications might
occurred. It might also be possible that the smaller proteins (46
and 35kDa) arise from proteolytic degradation of the 50kDa
protein. However only the 50kDa product was detected in both
BV and ODV. We also provided evident that HA2 co-localized
with cellular actin skeleton.
2. Materials and methods
2.1. Cells and virus
The H. zea cell line (BCIRL-HZ-AM1, HZ-AM1) and
Spodopterafrugiperda(Sf21)cellswereusedandmaintainedat
28◦C in Grace’s medium supplemented with 10% fetal bovine
serum. The HaSNPV G4, whose genome has been entirely
sequenced (Chen et al., 2001), and HaSNPV bacmid (HaHZ8)
(Wang et al., 2003) were used as wild type (wt) virus and prop-
agated in Hz-AM1 cells. An AcMNPV bacmid (bMON14272;
Invitrogen) was also used.
2.2. Construction of plasmids and recombinant virus
The entire ha2 was amplified from the HaSNPV G4 genome
with the primers: Ha2up 5?-GAA TTC ATG GTT CAA CTG
CAA AGT GTT-3?and Ha2down 5?-GGA TCC GCA ACT
TGC GAT TCA GTT GAC AT-3?. The PCR product was first
cloned into pGEM-T-Easy vector (Promega) (pT-HA2), and
then cloned into pEGFP-N1 vector with EcoRI and BamHI
to generate the Ha2-egfp fusion gene. To get the recombinant
virus, Ha2-egfp fusion gene was cloned into the vector pFast-
bac1 with EcoRI and NotI and produced the donor plasmid
pFast-Ha2-egfp. The plasmid pFast-Ha2-egfp was then used to
transformedcompetentDH10BcellscontaininghelperandHaS-
NPV bacmid (HaHZ8) or AcMNPV bacmid (Invitrogen). The
recombinant HaSNPV was used to transfect HzAM1 cells and
AcMNPV bacmid DNA isolated from DH10B transfected Sf21
cellsbyLipofectin(Invitrogen).RecombinantvirusesvHa-Ha2-
egfp, vAc-Ha2-egfp were harvested at 120h post transfection
(hpt). Recombinants were authenticated by PCR and restriction
enzymes analysis. Virus titres were determined by end point
dilution assay.
2.3. Preparation of antibody
Two primers, Ha2in (5?-GGA TCC TTG ATG GAA CAA
ATA CAG AAA GGA-3?) and Ha2down (5?-AAG CTT TTA AAC
TTG CGA TTC AGT TGA CAT-3?) were designed to amplify
the truncated Ha2 (corresponding to the C-terminal of Ha2: aa
201–413) from the pT-Ha2 and then cloned into the expres-
sion vector pGEX-KG with BamHI and HindIII. This generated
plasmid pGEX-cHa2, in which ha2 is in-frame and fused with
GST at the C terminus. Escherichia coli DH5?cells containing
pGEX-cHa2 were grown to an OD600of 0.4 and then induced
with 1mM IPTG. After 3h at 37◦C, cells were harvested and
lysed with lysozyme, sonicated, and centrifuged at 5000×g
for 10min at 4◦C. The fusion protein, present in the pellet
was separated in 12% SDS polyacrylamide gels and purified.
Antiserum were generated by immunizing rabbits with purified
protein (Sambrook et al., 1989) and tested by Western blot anal-
ysis. The antibody to OpMNPV ORF2 was kindly provided by
Dr.G.F.Rohrmann(DepartmentofMicrobiology,OregonState
University).
2.4. Transcription analysis of ha2
HzAM1 cells (106) were infected with HaSNPV-G4 BV at a
multiplicity of infection (MOI) of 5. Total RNA was isolated
at 0, 4, 8, 16, 24, 48, 72 and 96hpi by Trizol according to
the manufacture’s guidelines (GIBCO BRL). The RNA was
dissolved in 50?l water and quantified by absorbance mea-
surement at 260nm. Mock-infected HzAM1 cells were used as
controls.
RT-PCR was performed using 1?g total RNA as template
for each time point. First strand cDNA synthesis was performed
withAMVreversetranscriptase(Promega)anda15-ntoligo-dT
primer according to the manufacturer’s instructions. The cDNA
mixtureswereamplifiedbyPCRusingthegene-specificprimer,
Ha2down(5?-GGATCCGCAACTTGCGATTCAGTTGAC
AT-3?) and an internal primer, Ha2in (5?-GGA TCC TTG ATG
GAA CAA ATA CAG AAA GGA-3?). PCR products were ana-
lyzed on agarose gel.
2.5. Purification of HaSNPV BV and ODV fractions
Hz-AM1 cells were infected with BVs of HaSNPV G4 at
an MOI of 0.1. After 3 days, the cell culture supernatant was
collected and clarified at 2000×g for 10min at 4◦C. The
supernatant containing the BVs was passed through a 0.45mm

Page 3

Y. Nie et al. / Virus Research 116 (2006) 129–135
131
pore-size filter. BVs in the filtrate were pelleted through a 25%
(wt/wt) sucrose cushion in 0.1× TE (TE: 10mM Tris/HCl, pH
7.5, and 1.0mM EDTA) at 100,000×g for 90min at 4◦C, and
resuspended in 0.1× TE. ODVs were purified from HaSNPV
G4 polyhedra derived from infected H. armigera fourth-instar
larvae (Chen et al., 2001).
2.6. Western blotting analysis
HzAM1 cells were infected with HaSNPV-G4 at a MOI of 5.
Samples of total cell proteins were prepared from infected cells
harvested at 0, 4, 8, 12, 24, 36, 48, 72hpi. Protein samples were
separated on 12% SDS-PAGE and transferred onto Hybond-N
membrane (Amersham) by semi-dry electrophoresis (Ausubel
et al., 1994). HA2-specific antiserum and alkaline phosphatase-
conjugated goat anti-rabbit immunoglobulin (Sino-American,
China) were used as the primary and secondary antibodies,
respectively. The signal was detected using a BCIP/NBT kit
(Sino-American, China).
2.7. Fluorescence microscopy analysis
HzAM1 or Sf9 cells (1×105) were grown on glass cover
slips in Petri dishes and infected with vHa-Ha2-egfp or vAc-
Ha2-egfp at MOI of 10. At 24, 36, 48, 72 and 96hpi, the cells
werefixedandstainedwithHoechst(Beyotime),thenexamined
underaS2Leicalaserconfocalscanningmicroscopeforfluores-
cence. For the colocalization of HA2 and cellular actin, infected
cells were fixed at 48 or 72hpi for 5min on ice in 2% formalde-
hyde and 0.2% glutaraldehyde in PBS, then washed once with
PBS, solubilized with 0.15% Triton X-100 and stained with
Rhordamine-Phalloidin (Sigma). vHa-egfp was used as GFP
control.
2.8. Computer–assisted analysis
HA2 was analyzed using the ExPASy server (Appel et al.,
1994) for the prediction of signal peptides, transmembrane
regions and motifs. Protein comparisons with entries in the
updated GenBank/EMBL, SWISS-PROT, and PIR databases
were performed with BLASTP, FASTA and PSI-BLAST pro-
grams (Altschul et al., 1997). Multiple sequence alignments
wereperformedwithUW-GCGPileupcomputerprograms,with
gap creation and extension penalties set to 8 and 2, respectively
(Devereux et al., 1984). Alignment editing was done with Gen-
doc software.
2.9. Growth curves
For one-step growth curves, Hz-AM1 cells were infected
with viruses (vHa-egfp and vHa-Ha2-egfp) at an MOI of 0.5.
Supernatants were collected at the indicated times (0, 12, 24,
36, 48, 60, 72, 84, 96, 108, 120hpi). The titers of supernatants
were determined by end point dilution assay (EPDA) in Hz-
AM1 cells, and the final results were checked 7 days after
EPDA.
3. Results
3.1. Sequence analysis of ha2 and its homologues
Ha2 is located in the HindIII-A fragment of the HaSNPV
genome (nt 735–1976 (Chen et al., 2001). Sequence analysis
indicatedthatha2is1242bp,andencodesaproteinof413amino
acids with a predicted molecular mass of 45.9kDa. Two early
transcription initiation motifs, TATA and CAGT, were found
238 and 138nt upstream of the putative translational start site
of ha2, respectively. A late baculoviral transcription initiation
motif,ATAAG,wasfound397ntupstreamoftheputativetrans-
lationalstartsite,suggestingthatha2maybeanearlyorlategene
of HaSNPV. A T-rich sequence was found 63nt downstream of
thestopcodon,whichmightfunctionasthemRNAtranscription
terminationandpolyadenylationsignal(JinandGuarino,2000).
Searches of databases showed that HA2 is conserved among
all the lepidopteran 16 nucleopolyhedroviruses and 4 gran-
uloviruseswhosecompletesequencehavebeenreported,butnot
in baculoviruses from dipteran or hymenopteran hosts. Homol-
ogy analysis indicated that HA2 is most closely related to
HzSNPV ORF2 with 99% identity, consistent with the previ-
ous suggestion that HzSNPV and HaSNPV might be different
strains of the same virus (Chen et al., 2002). Sequence align-
ments (data not show) indicated low sequence identity among
all the HA2 homologues in NPV. However, the homologues are
positionally conserved in their respective genomes and share
conservedregionssuchasaprolinerichregion.Furtheranalysis
showedthatHA2containedcommonmotifsofWASPorWASP-
like protein such as RickA and ActA, including an acidic motif
(A), a central basic motif or cofilin homology and WH2 motif.
Alloftheseproteinshavethesamemotifsandorganization.The
motif conservation between HA2 and WASP proteins suggests
thatHA2maybeafunctionalhomologueofWASPproteinsand
may have the ability to interact with host actin and could well
be involved in virion transportation and/or assembly.
No signal peptide sequence was predicted on the putative
HA2 protein by PSORT and SignalP software, and one strong
transmembrane region from Gly218 to Leu236 was identified
by TMpred software at the ExPASy server (Appel et al., 1994).
In this 19 residues transmembrane domain, 17 are composed of
highly hydrophobic alanine, glycine, leucine, isoleucine, serine
and threonine. HA2 contained three potential N-linked glycosy-
lation consensus sequences (N-X-S/T) at position 15, 151 and
278. Two of these, at amino acid position 15 and 278 were
Fig. 1. Transcriptional analysis of ha2. Hz-AMI cells were infected with HaS-
NPVG4at5MOIandharvestedat0,8,12,16,24,36,48,and72hpostinfection.
RT-PCR was used to analyse ha2 transcription in each sample.

Page 4

132
Y. Nie et al. / Virus Research 116 (2006) 129–135
Fig. 2. Ha2 expression analysis and detection in virions. 1×106Hz-AMI cells were infected with HaSNPV G4 at MOI of 5. HA2 expression analysis with anti-HA2
antibody (A); HA2 detection in ODV and BV with anti-HA2 antibody (B).
predicted to be glycosylated by the NetNGlyc program of
ExPASyserver.Alsotherearenumerousserineortheoninephos-
phorylation sites predicted by NetPhos program.
3.2. Transcription analysis of ha2
Transcription of ha2 was examined by RT-PCR, using total
RNA isolated from HzAM1 infected with HaSNPV at different
timepoints.Thetranscriptofha2wasfirstdetectedat16hpiand
continued to be present until 96hpi (Fig. 1), indicating that ha2
is a late gene and may be transcribed from the late transcription
initiation motif ATAAG.
3.3. Immunodetection of the HA2 protein in infected cells
and in virus particles
Truncated HA2 (C terminal part, cHA2) was expressed in E.
coli as a GST-cHA2 fusion protein and purified by SDS-PAGE.
Specificantiserumwasgeneratedbyimmunizedrabbitswiththe
purified protein.
Western blot analysis of extracts of HaSNPV–infected Hz-
AM1 cells revealed three specific bands in size of 35, 46 and
50kDa (Fig. 2). The major band of 50kDa was first detected
at 24hpi and peaked at 48hpi, while the minor band of 46kDa
appeared at 24hpi and was undetectable by 72hpi. In contrast,
the other minor 35kDa was first observed at 48hpi. These data
areinagreementwiththeanalysisoftranscriptsandindicatethat
HA2 was synthesized at a late stage of infection. The size of the
50kDa protein is larger than the predicted 45.9kDa based on
its sequence, which suggests that this protein may be subject to
post-translationmodification.The35and46kDaformsmightbe
a cleavage product at the late stage of infection. When antibody
of OpMNPV ORF2 was used in the western blotting, only two
bandsinsizeof46and35kDaweredetectable;the50kDacould
not be detectable (data not shown.)
To investigate if HA2 is a structural component of HaSNPV
virions, Western blot analysis of BV and ODV proteins was
conducted. HA2 was detected in preparations of both ODVs
and BVs (Fig. 2B), indicating that the protein is a structural
component of ODVs ands BVs. However only the 50kDa band
was detected, the unmodified form of 46kDa and truncated
form of 35kDa did not appear to be incorporated into the virus
particles.
3.4. Cellular distribution of HA2 in HzAMI cells and Sf21
cells
The subcellular localization of the HA2 in permissive and
non-permissive cells was investigated using a C-terminal ha2-
egfp fusion construct. Two recombinant HaSNPVs were first
generatedbyBac-to-Bacsystems.Theha2-egfpfusiongenewas
inserted into HaHZ8 and AcMNPV bacmid and produced vHa-
Ha2-egfp and vAc-Ha2-egfp, respectively. Hz-AM1 cells were
infected with vHa-Ha2-egfp and examined for fluorescence by
Fig.3. HA2distributioninHz-AMI(B)and(C)andSf9(D)and(E).Schematic
diagram of recombinant baculovirus constructs (A). Hz-AM1 were infected by
vHa-Ha2-egfp (B) and (C) and Sf9 cells with vAc-Ha2-egfp (D) and (E) both
at MOI of 10, fixed and stained with Hoechst and observed at 24 and 72hpi.

Page 5

Y. Nie et al. / Virus Research 116 (2006) 129–135
133
Fig. 4. Co-localization of HA2 and cellular actin. Hz-AMI cells were infected with vHa-Ha2-egfp at MOI=5 without (A–C) or with 1?g/ml cytochalasin D (D–F).
(A) and (D), cells were stained with Rhordamin-Phalloidin at 48hpi; (B) and (E), HA2-egfp distribution in cells at 48hpi; (C) and (F), merge images.
confocal laser scanning microscopy at 24, 48, 72 and 96hpi.
The HA2-EGFP fusion protein was localized primarily in the
cytoplasm in the early stage (Fig. 3B) and in the nucleus from
48hpi (Fig. 3C) and had the tendency to form bright foci. How-
ever, when Sf9 cells infected with the vAc-Ha2-egfp, the fusion
protein was localized both in the cytoplasm and in the nucleus
(Fig. 3D and E). GFP alone showed homogeneous fluorescence
in the cytoplasm and nucleus (data not show). The different
localization of HA2 in Ha-AM1 and Sf21 might be due to the
different cell lines and/or the different viruses, implied that host
factor(s) and/or viral protein(s) might be involved.
3.5. Colocalization of HA2 and cellular actin
Hz-AM1 cells were infected with vHa-Ha2-egfp and vHa-
egfp at 10 MOI. Infected cells were stained by Rhordamine-
Phalloidin and then observed with laser confocal microscopy at
24,36,48,72and96hpi.HA2wasfoundtoshowco-localization
withcellactinespeciallyinlatecourseofinfection(Fig.4A–C).
Addition of cytochalasin D (1?g/ml) to infected cells resulted
in HA2 and actin aggregating in the nucleus and uncoupling the
co-localization (Fig. 4D–F). It is in concord with the previous
report that cytochalasin D treatment uncoupled P78/83 and F-
actin co-localization, indicating that HA2 bounds to the pointed
ends of actin filaments (Lanier and Volkman, 1998).
3.6. Biological assay of vHa-Ha2-egfp
Thetemporalkineticsoftheone-stepgrowthcurveofbudded
virus of vHa-Ha2-egfp was similar to that of vHa-egfp (Fig. 5).
Though the titer at 84hpi of vHa-Ha2-egfp was about 0.15 logs
Fig. 5. One-step growth curves in Hz-AM1 cells. Cells were infected with vHa-
egfp, vHa-Ha2-egfp at an MOI of 0.5. Progeny virus was assayed by the end
point dilution assay (EPDA).
higher than that of vHaSNPV-egfp. The growth kinetics of both
were quite similar. All titers increased to the highest at 84hpi.
Therefore, over-expression of HA2 did not appear to have an
effect on the virus growth and, the recombinant virus vHa-Ha2-
egfp has the same infectivity in HZ-AM1 cells.
4. Discussion
Ha2 is a conserved gene with homologues in most bac-
uloviruses from lepidopteran insects. The length and the iden-
tity of the homologies are quite variable, but they share the
same motifs of mammalian WASP proteins such as proline rich
domain, acidic motif, cofilin domain and WH2 motif. Acidic
motif (A), which was found to be Arp2/3 binding sequence and
seems to increase the efficiency of actin nucleation as well as
the parameter of motility (Machesky et al., 1999). Central basic
motiftermedcofilinwasreportedtobecrucialtothestimulation
of actin nucleation by Arp2/3 in vitro, and is also required for