Full text of DOI:10.1111/j.1348-0421.2002.tb02668.x

Microbiol. Immunol., 46(2), 119―129, 2002
Editor-Communicated Paper
Engineering of a Sagiyama Alphavirus RNA-Based
Transient Expression Vector
Yuka Yamaguchi1, and Yukio Shirako*, 2
1Graduate School of Agricultural Life Science, and 2Asian Center for Bioresources and Environmental Sciences (ANESC), Uni-
versity of Tokyo, Bunkyo-ku, Tokyo 113―8657, Japan
Communicated by Izumu Saito: Received December 14, 2001. Accepted December 18, 2001
Abstract: Sagiyama virus (SAGV), a strain of Getah virus in the genus Alphavirus in the family Togaviridae,
has a broad host range in vertebrates and invertebrates but is not pathogenic for humans. We engi-
neered the SAGV genome as an efficient transient expression vector using the full-length infectious cDNA
clone pSAG2 as the background. A green fluorescent protein (GFP) gene was used as a reporter gene and
expressed from a subgenomic mRNA. When the GFP gene was placed downstream of the intact capsid pro-
tein gene or an internally deleted capsid protein gene encoding the N-terminal 9 amino acids and C-terminal
149 amino acids, autoproteolysis occurred efficiently at the boundary site to release GFP from the N-ter-
minally-fused capsid-protease domain. GFP was also expressed efficiently without the 5'-terminal region
of the capsid protein gene, suggesting that SAGV capsid protein gene does not have a translation enhancer
sequence. To provide structural proteins for pseudovirion formation, a nonviable mutant construct,
pSAG2.3L, which contains a Gly-to-Leu substitution at the －2 position of the nsP3/4 cleavage site, was used
as a helper. GFP was expressed up to 50 µg from 1×106 BHK21 cells after inoculation of pseudovirions. The
C6/36 mosquito cell was also a suitable host for a large scale expression of GFP using pseudovirions. In addi-
tion to high-level transient expression, safeness of SAGV should give an advantage over other alphavirus
expression vectors.
Key words: Sagiyama virus, Alphavirus, Expression vector
The genus Alphavirus in the family Togaviridae (20,
31) contains more than 20 species, including important
pathogens of humans and domestic animals such as
Venezuelan, Eastern and Western equine encephalitis
viruses as well as members which do not cause nor-
mally serious illness to humans such as Sindbis virus
(SINV), the type species of the genus Alphavirus, and
Semliki Forest virus (SFV). Alphaviruses have wide
host ranges in vertebrates and invertebrates and are
transmitted in nature by mosquitos.
from a subgenomic messenger RNA which is transcribed
from a strong internal promoter on the full-length minus-
sense RNA, whereas the four nonstructural proteins,
nsP1―
4, required for RNA replication and transcription,
are encoded in the 5'-terminal region of the genomic
RNA and translated directly from it (30, 31).
Since SINV and SFV express large amounts of struc-
tural proteins from a subgenomic mRNA and since these
viruses are avirulent for humans, they have been devel-
oped as RNA-based transient expression vectors using
full-length cDNA clones from which infectious RNA can
Alphaviruses have an 11
―
12 kb, messenger-sense
RNA genome encapsidated by capsid proteins arranged
in a T＝4 structure, which is surrounded by an outer lipid
bilayer membrane containing 80 spikes, each composed
of three E1/E2 heterodimers (30). The structural proteins
are encoded in the 3'-terminal 4-kb region and expressed
be transcribed in vitro (1
foreign gene can be expressed from an infectious recom-
―
3, 7, 11, 15, 18, 25, 26, 33). A
Abbreviations: Cc, the C-terminal 149 amino acids of the cap-
sid protein; CTAB, hexadecyltrimethylammonium bromide;
GETV, Getah virus; GFP, green fluorescent protein; GST, glu-
tathione S transferase; IPTG, isopropyl-thiogalactoside; LMP,
low-melting point; PBS, phosphate buffered saline; SAGV,
Sagiyama virus; SDS-PAGE, sodium dodecyl sulfate-polyacryl-
amide gel electrophoresis; SFV, Semliki Forest virus; SINV,
Sindbis virus.
*Address correspondence to Dr. Yukio Shirako, Asian Center
for Bioresources and Environmental Sciences (ANESC), Uni-
versity of Tokyo, 1
―1
―
1 Yayoi, Bunkyo-ku, Tokyo 113
―8657,
Japan. Fax:＋ 81
―
3
―
5841 7524. E-mail: shirako@ims.u-
―
tokyo.ac.jp
119

120
Y. YAMAGUCHI AND Y. SHIRAKO
binant RNA containing the gene of interest under the
control of an additional subgenomic promoter. Another
strategy is to express a gene of interest from a self-
replicating, non-infectious replicon RNA. Structural
proteins for formation of infectious pseudovirions can be
provided by a helper RNA or by packaging cell lines
(19).
11,698 nucleotides of full-length SAGV cDNA flanked
by the T7 promoter at the 5' terminus and a 43-A stretch
at the 3' terminus in the plasmid Proteus1 derived from
pBR322 (22). (ii) In pSAG2.C:GFP, the genes encoding
the E3-E2-6K-E1 polyprotein downstream of the capsid
protein gene were replaced with an enhanced red-shift
GFP gene (720 bases encoding 27-kDa protein) which
Sagiyama virus (SAGV), a strain of Getah virus
(GETV), was isolated from mosquitoes (Culex tritae-
niorlyncus) in 1956 in a suburb of Tokyo (23, 24). The
virus was also isolated from swine and horses and anti-
bodies against SAGV and GETV have also been detect-
ed in human sera, but there have been no cases of illness
in humans caused by SAGV infection (12), indicating
that SAGV is nonpathogenic for humans.
Previously, we determined the complete nucleotide
sequence of the SAGV genome and constructed a full-
length infectious cDNA clone, pSAG2, from which
infectious RNA can be transcribed in vitro (28).
In this study, we engineered the SAGV genome as an
efficient RNA-based transient expression vector by
replacing the entire structural region or the genes encod-
ing the glycoproteins with a green fluorescent protein
(GFP) gene. In vitro transcripts of the replicon vector
RNA were electroporated into BHK21 cells. As a helper
construct, pSAG2.3L which has a lethal Gly→Leu sub-
stitution at the －2 position at the nsP3/nsP4 cleavage site
was used effectively to produce pseudovirions infec-
tious for both BHK21 cells and C6/36 mosquito cells.
was derived from pQBI25 (TaKaRa). Nucleotides 7042
―
8330 of pSAG2 were amplified by PCR using primers
TA25 and TA78 and Ex Taq DNA polymerase (TaKaRa),
and the 1.3-kb product was purified by electrophoresis in
a low-melting point (LMP) agarose gel followed by
extraction using hexadecyltrimethylammonium bromide
(CTAB) (Sigma, U.S.A.) (13). The GFP gene was
amplified from pQBI25 by PCR using primers TA79
and TA80 and the resultant 0.7-kb product was purified
by the LMP/CTAB method. The 3'-terminal region of
the 1.3-kb fragment and the 5'-terminal region of the 0.7-
kb fragment, which overlapped by 20 bases, were con-
nected by PCR using primers TA25 and TA80. The
resulting 2.0-kb product was digested with ClaI at the
position 7126 in SAG2 and with NsiI immediately down-
stream of the GFP termination codon. The 4.0-kb ClaI-
NsiI fragment in pSAG2 was replaced with the 1.9-kb
ClaI-NsiI fragment, resulting in pSAG2.C:GFP. Both the
ClaI site and the NsiI site (position 11096) are unique in
pSAG2. (iii) pSAG2.∆C:GFP has an internal deletion of
330 bases (nts 7554
―
7883) in the 5'-terminal region of
the capsid protein gene in pSAG2.C:GFP. To delete
the 330 bases from pSAG2.C:GFP, two fragments were
prepared by PCR using pSAG2.C:GFP as a template; a
0.5-kb fragment using primers TA25 and TA127 and a
1.2-kb fragment prepared using primers TA126 and
TA80, purified by the LMP/CTAB method, and con-
nected by PCR using primers TA25 and TA80. The
resultant 1.7-kb product was digested with ClaI and
NsiI and the 1.6-kb ClaI-NsiI fragment was cloned into
ClaI/NsiI-digested pSAG2 vector. (iv) pSAG2.∆C:GFP.H
was constructed in the same way as pSAG2.∆C:GFP
but using TA131 instead of TA80 to add six histidine
codons prior to the TGA termination codon for the GFP
gene. (v) In pSAG2.GFP.H, the entire structural protein
region was replaced with the GFP gene as follows; nts
Materials and Methods
Cells and culture conditions. BHK21 cell was pur-
chased from The American Type Culture Collection.
C6/36 cells were kindly provided by A. Igarashi, Institute
of Tropical Medicine at Nagasaki University, Japan.
Cells were cultured and maintained in Eagle’s mini-
mum essential medium (Type 1, Nissui Pharmaceuti-
cal, Japan) supplemented with 0.03% glutamine, 0.12%
NaHCO3, 5% fetal bovine serum (Bio Whittaker) and 1
mM non-essential amino acids (Gibco BRL). BHK21
cells were grown at 30 C or at 37 C whereas C6/36
cells were grown at 30 C.
Bacterial host and preparation of miniprep DNA.
Escherichia coli strain MC1061 was used as a host for
plasmid amplification. Miniprep DNA was prepared
using a modified boiling method (10).
Oligodeoxynucleotides. Primers used for preparing
expression vector and helper constructs are listed in
Table 1.
7042
―
7529 in pSAG2 were amplified by PCR using
primers TA25 and TA111 and the resulting 0.5-kb frag-
ment was purified by the LMP/CTAB method. The
GFP gene was amplified by PCR with TA112 and TA80
and the resulting 0.7-kb fragment was isolated. The
0.5-kb and 0.7-kb fragments which overlap by 17 bases
at nts 7513
―
7529 were connected by PCR using TA25
Replicon plasmid construction. (i) A full-length
infectious cDNA clone of SAGV, pSAG2 (Fig. 1A), has
been described previously (28). pSAG2 contains the
and TA80. The resulting 1.2-kb product was digested
with ClaI and NsiI, and the 1.1-kb ClaI-NsiI fragment
was cloned into ClaI/NsiI-digested pSAG2 vector. For all

121
SAGIYAMA ALPHAVIRUS EXPRESSION VECTOR

122
Y. YAMAGUCHI AND Y. SHIRAKO
vector constructs, the inserts between the ClaI site and
NsiI site were sequenced by an ABI model 377 auto-
mated sequencer to confirm the nucleotide sequences.
Helper plasmid construction. pSAG2.3L which has a
Gly-to-Leu substitution at the －2 position at the
nsP3/nsP4 cleavage site (Fig. 1B) resulting in nonvia-
bility, was used as a helper construct to provide the
structural proteins for packaging a vector replicon.
Primers TA89 (plus sense) and TA90 (minus sense)
were used to replace the Gly codon (GGC) at the posi-
Fig. 1. Scheme for construction of replicons and a vector. (A) Replicons. (i) pSAG2 (WT) used as the background; (ii) pSAG2.C:GFP;
(iii) pSAG2.∆C:GFP; (iv) pSAG2.∆C:GFP.H; (v) pSAG2.GFP.H. (B) pSAG2.3L helper. Nucleotide sequences and positions in
oligonucleotide primers used for constructing these clones are listed in Table 1.

123
SAGIYAMA ALPHAVIRUS EXPRESSION VECTOR
5643 to a Leu codon (CTG) by PCR as fol-
tions 5641
lows; nts 4985
and TA90, whereas nts 5632
―
ons and incubated at 30 C for 2 days. Cells were washed
once with PBS, and trypsinized to prepare a cell sus-
pension. The cells were collected by centrifugation at
1,000×g for 3 min. Pelleted cells were resuspended in
PBS and collected again by centrifugation. The pellet
was suspended and lysed in 0.5 ml of a lysis buffer (50
mM NaH2PO4, pH 8.0, 0.3 M NaCl, 10 mM imidazole, 1%
NP40) by gentle agitation. The lysed cell mixture was
centrifuged at 15,000×g for 3 min and the supernatant
was recovered. After addition of 100 µl of Ni2＋ -NTA
agarose suspension (QIAGEN), the mixture was incu-
bated at 4 C for 1 hr with continuous rocking, followed
by centrifugation at 15,000×g for 10 sec. Aliquot of the
supernatant was saved as an unbound fraction. The pel-
let was suspended in 0.5 ml of a wash buffer (50 mM
NaH2PO4, pH 8.0, 0.3 M NaCl, 20 mM imidazole) and
centrifuged at 15,000×g for 10 sec. Wash was repeat-
ed. Aliquots of the supernatants after centrifugation
were saved as the wash fractions 1 and 2. The pellet was
suspended in 0.5 ml of an elution buffer (50 mM
NaH2PO4, pH 8.0, 0.3 M NaCl, 250 mM imidazole) and
centrifuged at 15,000×g for 10 sec. The supernatant
was collected as the eluate fraction 1. The pellet was
again suspended in 0.5 ml of the elution buffer and cen-
trifuged. The supernatant was saved as the eluate fraction
2.
―
5652 were amplified by PCR using TA93
―
5970 were amplified by
PCR using TA89 and TA22. The two PCR products
were purified and connected over 21 bases using TA93
and TA22. The resulting 1.0-kb fragment was digested
with BamHI at the position 5104 and with AseI at the
position 5867 and the 0.7-kb BamHI-AseI fragment was
cloned into a BamHI/AseI-digested intermediate vector
containing the 2.4-kb BglII-ClaI fragment from pSAG2.
The 2.4-kb BglII-ClaI fragment containing the GGC→
CTG mutation at the positions 5641
―
5643 was cloned
into BglII/ClaI-digested 11.2-kb pSAG2 vector, resulting
in pSAG2.3L. The nucleotide sequence between BamHI
and ClaI sites in pSAG2.3L was confirmed using the
ABI377 automated sequencer.
In vitro transcription and electroporation. All vector
and helper constructs were linearized by digestion with
XbaI downstream of the 43-A stretch at the 3' terminus.
In vitro transcription was carried out in a 20 µl reaction
containing 0.2 µg of linearized plasmid DNA, 40 mM
Tris-HCl, pH 7.6, 6 mM MgCl2, 4 mM spermidine, 1
mM each ATP, CTP, UTP and GTP, 5 mM dithiothreitol,
0.5 mM cap analog (New England Biolabs), 2.0 units/µl
of RNase inhibitor (Promega), and 2.5 units/µl of T7
RNA polymerase (TaKaRa), at 37 C for 1 hr. After the
reaction, 1 µl of transcripts were run in a 0.8% agarose
gel, followed by staining in ethidium bromide to check
the quantity and integrity of the product. For electro-
poration, BHK21 cells were collected from T75 flasks by
trypsinization, washed once in complete BHK21 medi-
um, once in phosphate buffered saline (PBS), and final-
ly resuspended in PBS at a concentration of 1.0×107
cells/ml. Three hundred µl of cell suspension containing
3.3×106 cells was transferred into an electroporation
cuvette (0.4-cm gap, BTX) and mixed with 20 µl con-
taining approximately 20 µg of replicon transcripts.
Electroporation was carried out by giving two pulses
at 850 V and 50 µF at maximum resistance. Electropo-
rated cells were diluted in the complete medium at least
1:25 and incubated at 30 C or 37 C for 2 days.
Preparation of rabbit polyclonal antiserum against the
C-terminal 149-amino acid region of the SAGV capsid
protein. SAGV nts 7884
―
8330 encoding the C-terminal
149 amino acids of the capsid protein (Cc) were ampli-
fied by PCR using a plus-sense primer TA71, agaG-
GATCCTGCATCTTCGAGGTCAAGCT, which con-
sists of 3 extra bases (lower case) followed by BamHI site
(boldface) and SAG2 nts 7884
―
7903, and a minus-sense
primer TA72, tctCTCGAGCTACCATTCTTCGGTTC-
CTTCTG, which consists of SAG2 nts 8311
―
8330, fol-
lowed by TAG (boldface, italic) and XhoI site (bold-
face) plus 3 extra bases (lower case), using pSAG2 DNA
as the template. The resulting 0.5-kb product was digest-
ed with BamHI and XhoI and ligated into BamHI/XhoI-
digested pGEX6P1 (Amersham Pharmacia) so that Cc
was expressed as a glutathione S transferase (GST)-
fusion protein. Ligated DNA was transformed into E.
coli MC1061. Recombinant E. coli cells were grown in
LB medium and expression of the GST:Cc fusion protein
was induced by isopropyl-thiogalactoside (IPTG).
GST:Cc fusion protein was extracted from the E. coli
lysate and cleaved at the fusion boundary with PreScis-
sion Protease (Amersham Pharmacia). The Cc was
purified using glutathione Sepharose 4B according to the
manufacturer’s protocol. Two mg of purified Cc protein
was used to immunize a rabbit (Sawady Technology
Co., Japan).
Formation of pseudovirions. BHK21 cells were elec-
troporated with a mixture of replicon and helper tran-
scripts as described above. Electroporated cells were
diluted in the complete medium and incubated at 30 C for
2 days to rescue primary pseudovirions in the medium.
In order to increase titers of pseudovirions, the primary
pseudovirion-containing medium was directly used for
secondary infection to BHK21 monolayer cells in a T25
flask. After incubating at 30 C for 5 days, the medium
was harvested.
Purification of native histidine-tagged GFP. BHK21
or C6/36 monolayer cells were infected with pseudoviri-

124
SDS-PAGE and Western blot analysis. Protein sam-
Y. YAMAGUCHI AND Y. SHIRAKO
other amino acids at their N termini for cleavage effi-
ples were mixed with the same volume of 2× sample
buffer (0.1 M Tris-HCl, pH 9.0, 4% SDS, 30% sucrose,
0.1% bromophenol blue and 10% 2-mercaptoethanol),
heated for 3 min at 95 C and run on a 12.5% SDS-
PAGE gel. Proteins were stained with Coomassie blue
by the standard method. For Western blot analysis of the
capsid protein and GFP, proteins after SDS-PAGE was
transferred onto nitrocellulose membrane (S&S). Rab-
bit polyclonal antisera against GFP (Clontech) or against
the C-terminal region of the capsid protein (Cc) were
used as the primary antibodies and anti-rabbit goat IgG
conjugated with alkaline phosphatase (KPL) was used as
the secondary antibody. Heparin was added in the anti-
body dilution buffer at 5 units/ml (6). Antigen-anti-
body reaction was detected using bromo-chloro-indryl
phosphate and nitro blue tetrazolium as substrates.
ciency and stability of the protein.
The C-Terminal 149 Amino Acids Are Sufficient for
Autoproteolysis
The capsid protein of alphaviruses consists of the N-
terminal unstructured region very rich in basic amino
acids and the C-terminal, highly-structured trypsin-like
serine proteinase domain (4, 5, 9). In this study, we
deleted amino acid positions 10
―
119 in the capsid protein
so that the N-terminal 9 amino acids and C-terminal
149 amino acids are fused in a 17-kDa protein designated
as ∆C. Total cytoplasmic proteins from BHK21 cell
lysates after electroporation of in vitro transcripts from
pSAG2.∆C:GFP contained the 27-kDa GFP and the 17-
kDa ∆C but not a 44-kDa uncleaved protein (Fig. 3A,
lane 1). This result indicates that the C-terminal 149
amino acids in ∆C retained the full proteinase activity and
could cleave the fusion boundary as efficiently as the full-
length capsid protein.
Results and Discussion
Autoproteolysis Occurs at the C:GFP Boundary
BHK21 cells were transfected with in vitro transcripts
from pSAG2.C:GFP by electroporation and incubated at
30 C or at 37 C for 2 days. Transfection efficiency was
almost 100%, as judged by GFP fluorescence observed
with a fluorescence microscope (IX-70, Olympus Opti-
cal Co., Japan) (Fig. 2A). To examine whether GFP
was expressed as a part of a 57-kDa capsid-fusion prod-
uct or cleaved from the capsid protein by autoproteoly-
sis, total cytoplasmic proteins were analyzed by SDS-
PAGE and Coomassie blue staining (Fig. 2B). Two addi-
tional proteins of 27 kDa and 30 kDa were detected
from RNA transfected cells but a 57-kDa protein was not,
suggesting that cleavage occurred at the fusion boundary
and the 27-kDa GFP and the 30-kDa capsid protein
accumulated in the cytoplasm. To confirm cleavage at
the fusion boundary site, proteins separated by SDS-
PAGE were subjected to Western blot analysis using
antisera against the C-terminal region of the capsid pro-
tein (Cc) (Fig. 2C) and against GFP (Fig. 2D). With
the anti-Cc serum, only the 30-kDa protein was detected
in transfected cell lysates (Fig. 2C, lanes 2 and 4), which
migrated at the same position as the capsid protein of
SAGV virions purified by sucrose density gradient cen-
trifugation (Fig. 2C, lane 5). On the other hand, the
27-kDa protein was the only protein detected from trans-
fected cell lysates with the GFP antiserum (Fig. 2D,
lanes 2 and 4). These results indicate that cleavage at the
C:GFP boundary occurred efficiently at both 30 C and 37
C, probably while the protein was nascent during trans-
lation even when the amino acid at the ＋1 position is
methionine in GFP instead of serine at the the same
position of E3. It is of interest to examine proteins with
SAGV Does Not Need a Sequence Downstream of the
Capsid Initiation Codon for Efficient Translation of
GFP Gene
In the case of SINV and SFV, it has been reported that
a nucleotide sequence downstream of the AUG initiation
codon for the capsid protein is required to facilitate
translation (8, 29). Such a translational enhancer
sequence was not found in Venezuelan encephalitis
alphavirus (21). To examine whether SAGV has an
enhancer sequence in the 5'-terminal region of the capsid
protein gene, the structural protein region was directly
replaced with the GFP gene (Fig. 1A (v)). After elec-
troporation of BHK21 cells with in vitro transcripts
from pSAG2.GFP.H, a large amount of GFP was
expressed in the transfected cells (Fig. 3B, lane 2) at the
similar level as that from pSAG2.∆C:GFP (Fig. 3A,
lane 1). This result indicates that SAGV does not need a
sequence downstream of the capsid initiation codon for
efficient translation of GFP gene similar to the case for
Venezuelan equine encephalitis virus (21).
Pseudovirion Formation for Expression and Purifica-
tion of GFP from BHK21 Cells and C6/36 Mosquito
Cells
Because of the lack of the genes encoding the viral
structural proteins, replicon RNAs such as those from
pSAG2.C:GFP, pSAG2.∆C:GFP.H, or pSAG2.GFP can-
not be packaged into infectious viral particles. To pro-
vide structural proteins in trans for producing infec-
tious pseudovirions, we prepared a helper construct,
pSAG2.3L (Fig. 1B) in which the Gly (GGC) at the －2
position of the nsP3/P4 cleavage site was substituted

125
SAGIYAMA ALPHAVIRUS EXPRESSION VECTOR
Fig. 2. Expression of GFP in BHK21 cells using in vitro transcripts from pSAG2.C:GFP. (A) A fluorescent micrograph of BHK21 cells
transfected with in vitro transcripts from pSAG2.C:GFP and incubated at 30 C for 24 hr. (B) 12.5% SDS-PAGE gel stained with Coomassie
brilliant blue R. Three hundred µl of cell lysate was prepared from a T25 flask after incubation for 48 hr at 30 C (lanes 1 and 2) or at 37
C (lanes 3 and 4). Ten µl was loaded per well. －, not transfected (lanes 1 and 3); ＋, transfected (lanes 2 and 4). Positions of size mark-
er proteins are indicated in the left margin; 97.4-kDa phosphorylase b, 66-kDa bovine serum albumin, 45-kDa ovalbumin, 28.8-kDa car-
bonic anhydrase, 18.4-kDa β-lactoglobulin and 14.3-kDa lysozyme from the top. (C) Western blot analysis using antiserum against the
C-terminal region of the capsid protein (Cc). Lanes 1 to 4 are same as those in (B), but using 1/10 as much lysate. Lane 5, purified SAGV
virion. Virus was purified by sucrose density gradient centrifugation as described previously (27). (D) Western blot analysis using anti-
serum against GFP. Lanes 1 to 4 are same as those in (B), but using 1/10 as much lysate.

126
Y. YAMAGUCHI AND Y. SHIRAKO
on-containing medium was applied to 1 × 106 cells,
indicating that production of infectious pseudovirions was
not efficient after RNA transfection. However, when the
infected cells were incubated further at 30 C, clusters of
fluorescing cells started increase in number and size,
and by 96 hr after inoculation, all cells in the T25 flask
fluoresced (Fig. 4A, 48 hr, 72 hr and 96 hr). The medi-
um was collected 5 days after inoculation and inoculat-
ed to BHK21 monolayer cells after serial dilution.
BHK21 monolayer cells infected with 1/10-diluted medi-
um gave fluorescence of all cells, indicating the con-
centration of pseudovirions containing SAG2.C:GFP
was more than 107 particles/ml. This result was consis-
tently reproducible. The increase of replicon-containing
pseudovirions during secondary culture may be due to
change and increase in the ratio of helper RNA to repli-
con RNA because the more helper RNA should be
required for production of the higher titers of infectious
pseudovirion containing the vector replicon RNA.
Thus after the second passage, high titers of infectious
pseudovirions containing SAG2.∆C:GFP.H RNA were
obtained using SAG2.3L as the helper. GFP.H was
expressed in BHK21 cells efficiently and separated from
∆C (Fig. 4B, lane 1 and 4C, lane 1). Using Ni2＋-NTA
agarose, GFP.H was specifically eluted mostly in the
first fraction (Fig. 4B, lane 6 and 4C, lane 5). About 50
µg of GFP.H were recovered from a confluent monolayer
of about 1 × 106 cells in a T25 flask.
Fig. 3. Profiles of total cytoplasmic proteins from BHK21 cells
after separation on a 12.5% SDS-PAGE gel and staining with
Coomassie brilliant blue R. (A) Cell lysates after electroporation
of in vitro transcripts from pSAG2.∆C:GFP (lane 1). M, size
marker proteins indicated in the left margin as in Fig. 2 (B). (B)
Cell lysate after electroporation of in vitro transcripts from
pSAG2.GFP.H (lane 2) and that from untransfected cells (lane 1).
The positions of GFP, ∆C and GFP.H are indicated by arrows in
the right margins.
with Leu (CTG). Previously it has been shown that the
cleavage between nsP3 and nsP4 is essentially required
for SINV RNA replication (14, 27). When we examined
infectivity of in vitro transcripts from pSAG2.3L, trans-
fected BHK21 cells did not develop cytopathic effects
after incubation at 30 C or at 37 C even 5 days after RNA
transfection. The medium recovered after RNA trans-
fection did not contain infectious virus as judged by no
formation of any plaques and by absence of cytopathic
effects at either 30 C or 37 C using BHK21 monolayer
cells. Thus the Gly→Leu substitution at the －2 position
of the nsP3/nsP4 cleavage site was proved to be lethal in
SAGV as it was in SINV.
pSAG2.C:GFP and pSAG2.3L were separately tran-
scribed in vitro and 20 µg each of transcripts were mixed
with a BHK21 cell suspension for electroporation. After
electroporation, cells were incubated at 30 C for 2 days,
when all cells fluoresced from GFP (data not shown). To
examine formation of infectious pseudovirions, the medi-
um was collected and used to inoculate monolayer
BHK21 cells without dilution. After 24 hr incubation at
30 C, less than about 10% of the cells fluoresced without
forming clusters (Fig. 4A, 24 hr). The concentration of
pseudovirions containing SAG2.C:GFP RNA must be
less than 1×105 particles/ml since 1 ml of the pseudoviri-
We also tried C6/36 cells for a large-scale expres-
sion using pseudovirions since SAGV replicated very
efficiently in this mosquito cell line (28). As shown in
Fig. 4C, GFP.H was expressed highly and purified in
amounts similar to those obtained from BHK21 cells.
The SAGV RNA-based transient expression vector
established in this study has several advantages over
other alphavirus vectors. (1) SAGV itself is quite safe
without any cases of human illness so far. In the cases of
SINV and SFV, it has been reported that some isolates
are pathogenic for humans (16, 17, 32). (2) No transla-
tional enhancer in the capsid region has been detected for
SAGV, making it possible to express a foreign gene
without including additional residues derived from the N-
terminal region of the capsid protein. (3) In the SAGV
system, GFP is expressed and cleaved utilizing the serine
proteinase activity located in the C-terminal region of the
capsid protein. Proteins with amino acids other than
methionine at the N terminus can be expressed and
cleaved from the capsid protein, expanding the usefulness
of this vector system.
We thank Ellen Strauss of California Institute of Technology
for correcting English of this manuscript. This work was sup-
ported in part by grants from Ministry of Education, Science,

127
SAGIYAMA ALPHAVIRUS EXPRESSION VECTOR
Fig. 4. Use of SAG2.3L helper RNA for production of infectious pseudovirion for expression and purification of GFP.H in BHK21 cells
and C6/36 cells. (A) Fluorescent micrographs of BHK21 cells infected with undiluted pseudovirions harvested after incubation at 30 C
for 48 hr after co-transfection of BHK21 cells with in vitro transcripts from pSAG2.C:GFP and pSAG2.3L. BHK21 cells inoculated with
pseudovirions were incubated at 30 C and pictures were taken at 24 hr, at 48 hr, at 72 hr and at 96 hr postinoculation. (B) Protein pro-
files of BHK21 cell lysate subjected to Ni2＋-NTA agarose purification. Three hundred µl of cell suspension was prepared from a T25 flask.
Five µl was loaded on each well. Lane 1, unfractionated cell lysate; lane 2, insoluble fraction suspended in 300 µl of PBS; lane 3, super-
natant after addition of Ni2＋-NTA agarose and centrifugation; lane 4, 1st wash with 300 µl of wash buffer; lane 5, 2nd wash with 300
µl of wash buffer; lane 6, 1st eluate with 300 µl of elution buffer; lane 7, 2nd eluate with 300 µl of elution buffer. M, size marker pro-
teins indicated in the left margin as in Fig. 2 (B). (C) Protein profiles of C6/36 cell lysate subjected to Ni2＋-NTA agarose purification.
Details are as shown in (B). Lane 1, unfractionated cell lysate; lane 2, unbound to Ni2＋-NTA agarose resin; lane 3, 1st wash; lane 4, 2nd
wash; lane 5, 1st eluate; lane 6, 2nd eluate.

128
Y. YAMAGUCHI AND Y. SHIRAKO
Sports and Culture, Japan.
A., Daguisy, A.M., and Georges, A.J. 1990. An outbreak of
human Semliki Forest virus infections in Central African
Republic. Am. J. Trop. Med. Hyg. 42: 386 393.
―
References
17) Niklasson, B., Aspmark, A., LeDuc, J.W., Gargan, T.P.,
Ennis, W.A., Tesh, R.B., and Main, J.A.J. 1984. Association
of a Sindbis-like virus with Ockelbo disease in Sweden.
1) Agapov, E.V., Frolov, I., Lindenbach, B.D., Pragai, B.M.,
Schlesinger, S., and Rice, C.M. 1998. Noncytopathic Sind-
bis virus RNA vectors for heterologous gene expression.
Am. J. Trop. Med. Hyg. 33: 1212―1217.
18) Pettigrew, M.M., and O’Neill, L. 1999. Semliki Forest virus
Proc. Natl. Acad. Sci. U.S.A. 95: 12989―12994.
as an expression vector in insect cell lines. Insect Mol. Biol.
8: 409―414.
2) Berglund, P., Sjoberg, M., Garoff, H., Atkins, G.J., Sheahan,
B.J., and Liljestrom, P. 1993. Semliki forest virus expression
system: production of conditionally infectious recombinant
19) Polo, J.M., Belli, B.A., Driver, D.A., Frolov, I., Sherrill, S.,
Hariharan, M.J., Townsend, K., Perri, S., Mento, S.J., Jolly,
D.J., Chang, S.M.W., Schlesinger, S., and Dubensky, T.W.
1999. Stable alphavirus packaging cell lines for Sindbis
virus and Semliki Forest virus-derived vectors. Proc. Natl.
particles. Bio/Technology 11: 916―920.
3) Bredenbeek, P.J., Frolov, I., Rice, C.M., and Schlesinger, S.
1993. Sindbis virus expression vectors: packaging of RNA
replicons by using defective helper RNAs. J. Virol. 67:
Acad. Sci. U.S.A. 96: 4598―4603.
6439―6446.
20) Power, A.M., Brault, A.C., Shirako, Y., Strauss, E.G., Kan,
W., Strauss, J.H., and Weaver, S.C. 2001. Evolutionary rela-
tionships and systematics of the alphaviruses. J. Virol. 75:
4) Choi, H.K., Lu, G., Lee, S., Wengler, G., and Rossmann,
M.G. 1997. Structure of Semliki Forest virus core protein.
Proteins 27: 345―359.
10118―10131.
5) Choi, H.K., Tong, L., Minor, W., Dumas, P., Boege, U.,
Rossmann, M.G., and Wengler, G. 1991. Structure of Sind-
bis virus core protein reveals a chymotrypsin-like serine
proteinase and the organization of the virion. Nature 354:
21) Pushko, P., Parker, M., Ludwig, G.V., Davis, N.L., Johnston,
R.E., and Smith, J.F. 1997. Replicon-helper systems from
attenuated Venezuelan equine encephalitis virus: expres-
sion of heterologous genes in vitro and immunization against
37―43.
heterologous pathogens in vivo. Virology 239: 389―401.
6) Dietzgen, R.G., and Francki, R.I. 1987. Nonspecific binding
of immunoglobulins to coat proteins of certain plant viruses
in immunoblots and indirect ELISA. J. Virol. Methods 15:
22) Rice, C.M., Levis, R., Strauss, J.H., and Huang, H.V. 1987.
Production of infectious RNA transcripts from Sindbis virus
cDNA clones: mapping of lethal mutations, rescue of a tem-
perature-sensitive marker, and in vitro mutagenesis to gen-
159―164.
7) Frolov, I., Hoffman, T.A., Pragai, B.M., Dryga, S.A., Huang,
H.V., Schlesinger, S., and Rice, C.M. 1996. Alphavirus-
based expression vectors: strategies and applications. Proc.
Natl. Acad. Sci. U.S.A. 93: 11371―11377.
8) Frolov, I., and Schlesinger, S. 1994. Translation of Sindbis
virus mRNA: effects of sequences downstream of the initi-
erate defined mutants. J. Virol. 61: 3809―3809.
23) Scherer, W.F., Funkenbusch, M., Buescher, E.L., and Izumi,
T. 1962. Sagiyama virus, a new group A arthropod-borne
virus from Japan. I. Isolation, immunologic classification, and
ecologic observations. Am. J. Trop. Med. Hyg. 11: 255―268.
24) Scherer, W.F., Izumi, T., McCown, J., and Hardy, J.L. 1962.
Sagiyama virus. II. Some biologic, physical, chemical and
ating codon. J. Virol. 68: 8111―8117.
9) Hahn, C.S., and Strauss, J.H. 1990. Site-directed mutagenesis
of the proposed catalytic amino acids of the Sindbis virus
immunologic properties. Am. J. Trop. Med. Hyg. 11: 269
―
282.
capsid protein autoprotease. J. Virol. 64: 3069―3073.
25) Schlesinger, S. 1993. Alphaviruses-vectors for the expression
of heterologous genes. Trends Biotechnol. 11: 18 22.
10) Holmes, D.S., and Quigley, M. 1981. A rapid boiling method
―
for the preparation of bacterial plasmids. Anal. Biochem. 114:
26) Schlesinger, S., and Dubensky, T.W., Jr. 1999. Alphaviruses
vectors for gene expression and vaccines. Curr. Opin. Biote-
193―197.
11) Huang, H.V., Rice, C.M., Xiong, C., and Schlesinger, S.
1989. RNA viruses as gene expression vectors. Virus Genes
chol. 10: 434―439.
27) Shirako, Y., and Strauss, J.H. 1994. Regulation of Sindbis
virus RNA replication: uncleaved P123 and nsP4 function in
minus-strand RNA synthesis, whereas cleaved products
from P123 are required for efficient plus-strand RNA syn-
3: 85
―91.
12) Kono, Y. 1988. Getah virus disease, p. 21
―
36. In Monath, T.P.
(ed), The arboviruses: epidemiology and ecology, CRC
Press, New York.
thesis. J. Virol. 68: 1874―1885.
28) Shirako, Y., and Yamaguchi, Y. 2000. Genome structure of
Sagiyama virus and its relatedness to other alphaviruses. J.
13) Langridge, J., Langridge, P., and Bergquist, P.L. 1980.
Extraction of nucleic acids from agarose gels. Anal.
Biochem. 103: 264―271.
Gen. Virol. 81: 1353―1360.
14) Lemm, J.A., Rumenapf, T., Strauss, E.G., Strauss, J.H., and
Rice, C.M. 1994. Polypeptide requirements for assembly
of functional Sindbis virus replication complexes: a model for
the temporal regulation of minus- and plus-strand RNA syn-
29) Sjoberg, E.M., Suomalainen, M., and Garoff, H. 1994. A sig-
nificantly improved Semliki Forest virus expression system
based on translation enhancer segments from the viral capsid
gene. Biotechnology 12: 1127―1131.
thesis. EMBO J. 13: 2925―2934.
30) Strauss, J.H., and Strauss, E.G. 1994. The alphaviruses:
gene expression, replication, and evolution. Microbiol. Rev.
15) Liljestrom, P., and Garoff, H. 1991. A new generation of ani-
mal cell expression vectors based on the semliki forest virus
58: 491
―562.
replicon. Biotechnology 9: 1356
―1361.
31) Weaver, S.C., Dalgarno, L., Frey, T.K., Huang, H.V., Kinney,
16) Mathiot, C.C., Grimaud, G., Garry, P., Bouquety, J.C., Mada,

129
SAGIYAMA ALPHAVIRUS EXPRESSION VECTOR
R.M., Rice, C.M., Roehrig, J.T., Shope, R.E., and Strauss,
E.G. 2000. Family Togaviridae, p. 879 889. In van Regen-
cause of a fatal case of human encephalitis. Science 203:
1127 1129.
―
―
mortel, M.H.V., Fauquet, C.M., Bishop, D.H.L., Carstens,
E.B., Estes, M.K., Lemon, S.M., Maniloff, J., Mayo, M.A.,
McGeoch, D.J., Pringle, C.R., and Wickner, R.B. (eds),
Virus taxonomy, Academic Press, New York.
33) Xiong, C., Levis, R., Shen, P., Schlesinger, S., Rice, C.M.,
and Huang, H.V. 1989. Sindbis virus: an efficient, broad
host range vector for gene expression in animal cells expres-
sion vectors: packaging of RNA replicons by using defective
32) Willems, W.R., Kaluza, G., Boschek, C.B., Bauer, H., Hager,
H., Schetz, H.J., and Feistner, H. 1979. Semliki Forest virus:
helper RNAs. Science 243: 1188―1191.