Epitope mapping and biological function analysis of antibodies produced
by immunization of mice with an inactivated Chinese isolate of severe
acute respiratory syndrome-associated coronavirus (SARS-CoV)
Te-hui W. Choua, Shixia Wanga, Pavlo V. Sakhatskyya, Innocent Mboudoudjecka,
John M. Lawrencea, Song Huanga, Scott Coleya, Baoan Yangb, Jiaming Lib,
Qingyu Zhub, Shan Lua,T
aLaboratory of Nucleic Acid Vaccines, Department of Medicine, University of Massachusetts Medical School, 364 Plantation Street,
Lazare Research Building, Worcester, MA 01605-2397, USA
bInstitute of Microbiology and Epidemiology, Chinese Academy of Military Medical Sciences, Beijing 100071, China
Received 9 December 2004; returned to author for revision 10 January 2005; accepted 25 January 2005
Inactivated severe acute respiratory syndrome-associated coronavirus (SARS-CoV) has been tested as a candidate vaccine against the re-
emergence of SARS. In order to understand the efficacy and safety of this approach, it is important to know the antibody specificities
generated with inactivated SARS-CoV. In the current study, a panel of twelve monoclonal antibodies (mAbs) was established by immunizing
Balb/c mice with the inactivated BJ01 strain of SARS-CoV isolated from the lung tissue of a SARS-infected Chinese patient. These mAbs
could recognize SARS-CoV-infected cells by immunofluorescence analysis (IFA). Seven of them were mapped to the specific segments of
recombinant spike (S) protein: six on S1 subunit (aa 12–798) and one on S2 subunit (aa 797–1192). High neutralizing titers against SARS-
CoV were detected with two mAbs (1A5 and 2C5) targeting at a subdomain of S protein (aa 310–535), consistent with the previous report
that this segment of S protein contains the major neutralizing domain. Some of these S-specific mAbs were able to recognize cleaved
products of S protein in SARS-CoV-infected Vero E6 cells. None of the remaining five mAbs could recognize either of the recombinant S, N,
M, or E antigens by ELISA. This study demonstrated that the inactivated SARS-CoV was able to preserve the immunogenicity of S protein
including its major neutralizing domain. The relative ease with which these mAbs were generated against SARS-CoV virions further supports
that subunit vaccination with S constructs may also be able to protect animals and perhaps humans. It is somewhat unexpected that no N-
specific mAbs were identified albeit anti-N IgG was easily identified in SARS-CoV-infected patients. The availability of this panel of mAbs
also provided potentially useful agents with applications in therapy, diagnosis, and basic research of SARS-CoV.
D 2005 Elsevier Inc. All rights reserved.
Keywords: SARS-CoV; Monoclonal antibody; Epitope mapping; Inactivated vaccine
Severe acute respiratory syndrome (SARS), a highly
virulent emerging infectious disease, can spread rapidly
among large human populations as demonstrated in the first
half of 2003. The etiological agent of SARS was identified as
a new human coronavirus, SARS-CoV (Ksiazek et al., 2003;
Marra et al., 2003; Rota et al., 2003). While the first SARS
epidemic was successfully contained with the collaborative
efforts organized by the World Health Organization, SARS
remains a potential threat due to the mysterious source of its
initial infection and highly transmittable nature of this virus
(Callow et al., 1990; Holmes, 2001; Kraaijeveld et al., 1980).
Its apparent presence in animal reservoirs provided the
possibility of reemergence, including in forms with increased
infectivity. Because the symptoms of SARS can be confused
clinically with many respiratory diseases caused by other
0042-6822/$ - see front matter D 2005 Elsevier Inc. All rights reserved.
T Corresponding author. Fax: +1 508 856 6751.
E-mail address: email@example.com (S. Lu).
Virology 334 (2005) 134–143
applicable at the early onset of SARS still needs to be
developed. The current medical strategies mainly relying on
non-specific anti-viral and supportive treatment are not
sufficient. While questions remain whether natural SARS
infection will ever return in the form seen in 2003, it is
important to test various candidate SARS-CoV vaccines and
to produce tools which can be used to diagnose and treat this
infection when it re-emerges.
The SARS-CoV is an enveloped positive-sense single-
stranded RNA virus of 29,700 nucleotides that has been
completely sequenced. Open reading frames (ORFs) analysis
by analogy with other known coronaviruses indicated that
four structural proteins might play important functions
associated with SARS-CoV infection, including the surface
spike protein S (1,255 aa), which has an N-terminal receptor
binding domain to mediate attachment to cellular receptors
(Lewicki and Gallagher, 2002; Wong et al., 2004) and C-
terminal heptad repeats (HRs) to promote virus entry by
fusion with cell membranes (Bosch et al., 2003; Luo and
2003). The nucleocapsid N protein (422 aa) is a highly basic
structural protein, which is usually known to bind viral RNA
1981) and to promote viral packaging and viral core
formation (He et al., 2004a; Hiscox et al., 2001). The matrix
membrane glycoprotein M (221 aa) is an integral membrane
protein involved in budding. There is a small envelope
protein E (76 aa), a critical component of the virus
responsible for virion envelope morphogenesis (Arbely et
al.,2004) which acts as ascaffoldproteinto trigger assembly.
SARS-CoV lacks the envelope-associated hemagglutinin-
esterase glycoprotein that is encoded by some coronaviruses.
dependent RNA-polymerase (1a/b) and eight other non-
structure proteins (Marra et al., 2003; Rota et al., 2003).
Since the discovery of SARS-CoVas the cause of SARS,
inactivated SARS-CoV has been proposed as one of the
prophylactic vaccination approaches against SARS. Animal
and early phase human studies have been started (Darnell et
al., 2004; He et al., 2004b; Marshall and Enserink, 2004;
Takasuka et al., 2004; Tang et al., 2004; Xiong et al., 2004).
However, the breadth and specificity of antibody responses
in animals immunized with inactivated SARS-CoV have not
been well characterized. At the same time, a number of
studies have mapped the major neutralizing domain on
SARS-CoV to the S protein between amino acids 261 and
672 (Sui et al., 2004). This epitope coincides with the
angiotensin-converting enzyme 2 (ACE2) receptor binding
site identified for S protein in Vero E6 cells infected with
SARS-CoV (Berry et al., 2004; Che et al., 2003; He et al.,
2003; Hua et al., 2004; Li et al., 2003; Sui et al., 2004; ter
Meulen et al., 2004; Traggiai et al., 2004; Wen et al., 2004;
Wong et al., 2004; Zhou et al., 2004).
In the current study, a panel of monoclonal antibodies
was produced from mice immunized with the inactivated
SARS-CoV isolated from an infected patient (Qin et al.,
2003; Tang et al., 2004) and the spectrum of antibody
responses that can be elicited with an inactivated SARS-CoV
was examined. Our results suggested that most mAbs
generated by this approach recognized quite diverse epitopes
on S protein but not against the other three structural proteins
(N, E, and M). S-specific mAbs showed different biological
activities. Because virion-associated antigens may resemble
the true topology of SARS-CoV antigens more closely than
recombinant SARS-CoV proteins, the current study pro-
duced a panel of mAbs against epitopes which may not be
otherwise recognized, thus providing useful tools for a wide
range of potential applications related to SARS research.
Screening of antigen specificity on monoclonal antibodies
produced by immunization with an inactivated SARS-CoV
Positive antibody responses against SARS-CoV were
identified in five Balb/C mice after being immunized three
times with inactivated SARS-CoV. After fusion of the
spleen cells with SP2/0 myeloma with several rounds of
screening, twelve individual hybridomas were finally
obtained showing antibody responses against SARS-CoV
as verified by IFA staining of SARS-CoV-infected Vero E6
cells (data not shown). The antigen specificity was further
mapped by ELISA against recombinant SARS-CoV struc-
tural proteins. Six mAbs (2A3, 1B4, 2B1, 1A5, 2C5, and
3A3) showed high-level reactivity against the full-length
recombinant Spike (S) protein produced by the mammalian
expression system (Fig. 1). The remaining mAbs showed
very poor reactivity against S protein (Fig. 1) and also failed
to recognize recombinant E, M, and N in the subsequent
ELISA tests (data not shown).
Mapping of mAbs epitope on specific domains of S
The S protein of SARS-CoV is a large type-I trans-
membrane glycoprotein composed of 1255 amino acids
including an N-terminal S1 domain and a C-terminal S2
domain (Marra et al., 2003; Rota et al., 2003; Spiga et al.,
2003). We used individual segments of S (Fig. 2) to further
map the epitopes for six mAbs that showed high reactivity
against the full-length S protein. The designation of S1 (aa
12–798) and S2 (aa 797–1255) domains in the current study
was based on the knowledge of S protein structure from
coronaviruses in general (Lai and Holmes, 2001) and the
alignment of SARS-CoV S protein sequence against S
proteins from other coronaviruses with known cleavage sites
between their S1 and S2 domains (Bosch et al., 2003;
Krokhin et al., 2003; Marra et al., 2003; Rota et al., 2003;
Spiga et al., 2003). However, there has been no direct
evidence in the literature to suggest S protein of SARS-CoV
was cleaved into S1 and S2. In our study, the S1 segment
T.W. Chou et al. / Virology 334 (2005) 134–143
was further divided into two sub-segments, S1.1 (aa 12–
535) and S1.2 (aa 534–798), based on the distribution of the
predicted N-glycan sites. The S1.1 segment was then further
divided into S1.1a (aa 12–311) and S1.1b (aa 310–535), and
the latter was reported to be the site for binding of SARS-
CoV S protein to its receptor ACE 2 (Li et al., 2003).
These recombinant S segment proteins were produced by
transiently transfected 293Tcells and used as antigens in the
ELISA and Western blot analyses. Three mAbs (2A3, 1B4,
and 2B1) were mapped to S1.1a, two mAbs (1A5 and 2C5)
to S1.1b, and one (3A3) to the S2 domain based on the
result of ELISA (Table 1). The specificity of these mAbs
was further confirmed by Western blot analysis recognizing
specific recombinant S protein fragments. Fig. 3 shows the
example of one mAb, 2A3, whose specificity was mapped
to the S1.1a domain by ELISA (Fig. 3A) and confirmed by
Fig. 1. Epitope mapping of mAbs to the S protein by ELISA. MAbs were added at 1:100,000 dilution to each well against either the recombinant full-length S
protein (bars in shade) expressed from transiently transfected 293Tcells or the vector DNA-transfected 293Tcells (bars in blank). b+Q and b?Q were control sera
from New Zealand rabbits immunized with either the full-length S DNAvaccine or the empty vector DNA, respectively. The OD450nmvalues less than 0.2 are
considered non-specific binding to the full-length S antigen in this ELISA assay.
Fig. 2. Designs of recombinant S segments used in this study. Schematic representation of the entire Spike protein is shown on top, including its natural leader
and its transmembrane domain close to the C terminal tail. Predicated N-glycosylation sites are marked by asterisks and the ACE2 receptor binding domain is
also noted. DNA plasmids expressing different segments of the S protein were shown in the lower part of the figure with their amino acid residue numbers
T.W. Chou et al. / Virology 334 (2005) 134–143
the Western blot of its recognition to recombinant S, S1,
S1.1 proteins, and the SARS-CoV virion associated S
antigen, but not to recombinant S1.2 or S2 protein (Fig. 3B).
Similarly, the epitopes of mAbs 2B1 and 2C5 were mapped
to S1.1a and S1.1b, respectively (Table 1). Another mAb,
3A3, which mapped to the S2 domain by ELISA (Table 1),
recognized not only the virus-associated full-length S
antigen and its oligomer but also a smaller fragment about
70 kDa which is likely to be the cleaved S2 (Fig. 4).
Interestingly, the other two S-specific mAbs (1B4 and 1A5)
as identified by the ELISA screening did not show any
recognitions to the recombinant S proteins by the Western
blot (Table 1), suggesting these mAbs may mainly recognize
Unexpectedly, another mAb 2D4 which failed to show
reactivity against the full-length S antigen by either ELISA
(Fig. 1) or Western blot recognized the recombinant S1
protein (Fig. 5A). In consistent with this finding, mAb
2D4 recognized a lower molecular band, but not the full-
length S associated with the SARS-CoV virion (Fig. 5A).
Additional ELISA analysis further mapped its epitope to
the first 311 aa of the S protein because this mAb showed
low but clearly positive reactivity against S1.1 and S1.1a
antigens (Fig. 5B).
Mapping of S-specific mAbs by ELISA and Western blot analyses
mAbsRecombinant S antigens recognized in ELISAWestern
S S1 S1.1 S1.1aS1.1b S1.2S2 Vector
ND: not done.
Fig. 3. Epitope mapping of mAb 2A3 using specific recombinant S antigens
expressed in 293T transfected cells. (A) Detection of various recombinant S
proteins as the coating antigens by ELISA. The OD450nmvalues in ELISA
less than 0.2 are considered non-specific binding to S antigens. (B)
Detection of S and its subdomains by Western blot analysis. Samples
included uninfected Vero E6 cell lysate (Vero E6), SARS-CoV-infected
Vero E6 cell lysate (SARS-CoV), and recombinant S proteins (S, S1, S1.1,
S1.2, and S2) as labeled.
Fig. 4. Detection of SARS-CoV-associated S protein with mAb 3A3 by
Western blot analysis. Samples loaded included uninfected Vero E6 lysate
and SARS-CoV-infected Vero E6 lysate. Full length S protein (S), the
dimmer form of S (di-S), and processed S protein (SV ) are indicated.
Fig. 5. Epitope mapping of mAb 2D4. (A) Western blot analysis of mAb
2D4 with the following samples: uninfected Vero E6 cells, SARS-CoV-
infected Vero E6 lysate, recombinant S proteins (S, S1 and S2) expressed
from transiently transfected 293T cells, and the control 293T cells
transfected with vector DNA. (B) ELISA with mAb 2D4 at 1:500 dilution
against different recombinant S antigens (S1.1, S1.1a, S1.1b, S1.2, and S2)
expressed from transiently transfected 293T cells and control 293T cells
transfected with vector DNA. OD450nmvalues less than 0.1 are considered
non-specific binding in this ELISA assay.
T.W. Chou et al. / Virology 334 (2005) 134–143
Neutralization of SARS-CoV infection to Vero E6 cells
Out of twelve mAbs tested in this study, two S1.1b
specific mAbs, 1A5 and 2C5, showed significant levels of
neutralizing activities against SARS-CoV (Fig. 6). Two
other S-specific mAbs, 2A3 and 1B4, which had their
epitopes on S1.1a domain, showed only borderline neutral-
izing activities. The remaining three S-specific mAbs (3A3,
2B1, and 2D4) did not demonstrate any neutralizing
activities (Fig. 6). The neutralizing titers of these mAbs as
measured by 50% inhibition of SARS-CoV infection were
summarized in Table 2 along with the original concen-
trations of S-specific IgG in these mAbs. This panel of
mAbs was further characterized with respect to their
immunoglobulin isotypes. Analyses on the constant regions
of their heavy chains showed that there were five IgG1, five
IgG2a, one IgG2b, and one IgM (Table 3). All of them used
n light chains instead of E chains. Two mAbs, 1A5 and 2C5,
that showed strong neutralizing activities against SARS-
CoV were IgG2a and IgG1 isotypes, respectively.
The current study confirmed that the inactivated SARS-
CoV was immunogenic in eliciting antibody responses
against SARS-CoV antigens. While it is less likely that
inactivated SARS-CoV can completely reflect the antigen
conformation of SARS-CoV, our data revealed that the S
protein is the dominant antigen among the virion-associated
proteins because majority of the mAbs produced in this
study were specific for S protein including two with
Fig. 6. Neutralizing activities of mAbs against the Urbani strain of SARS-CoV as measured by neutral red staining in infected Vero E6 cells. Neutralizing
activities are plotted as the percent inhibition of viral infection against a particular rabbit serum dilution based on the geometric means from triplet wells. The
50% inhibition (IC50) or 75% inhibition (IC75) levels are marked.
Summary of mAb neutralizing activities against SARS-CoV Urbani strain
by neutral red assay
mAbs Specificity Concentration (Ag/Al)Neutralization titers (IC50)*
Neutralizing titers are expressed as the highest sera dilutions that achieved
50% inhibition of SARS-CoV infection to Vero E6 cells. The values are the
geometric means from 3 triplet wells.
ND: not done.
Ig subtype of monoclonal antibodies against SARS-CoV
mAbs Heavy chainLight chain
IgG1IgG2aIgG2bIgG3 IgA IgM
T.W. Chou et al. / Virology 334 (2005) 134–143
excellent neutralizing activities against SARS-CoV infec-
tion (Fig. 6). It is interesting that none of the mAb could
recognize N, M, and E antigens even though anti-N
antibody has been reported as one of the major antibody
components in SARS-CoV-infected patient sera (Ying et al.,
2004). This finding is different from the recent in vitro study
using subunit SARS-CoV antigens in which M protein
interacts with the N protein through its C-terminal domain
to facilitate formation of nucleocapsids, and M was also
reported to be able to interact with S protein to form
pseudoparticles (Huang et al., 2004). In the same study,
DNA vaccine expressing N antigen elicited high titer anti-N
antibody responses thus ruling out the possibility that N is a
No epitopes were identified for five mAbs produced by
immunization with inactivated SARS-CoV. It is likely that
these mAbs may have recognized some of the antigen
determinants that were unique to the inactivated virion. The
possibility of these mAbs to recognize non-structure
proteins exists but is less likely since the mAbs were
initially selected based on their reactivity against inactivated
virion by ELISA. Knowledge related to the specificity of
antibody responses generated by inactivated SARS-CoV
vaccines is important for the evaluation of efficacy and
safety profile for such vaccination approach against SARS.
Viral enhancing antibody (or autoimmune responses) was
reported for inactivated vaccine approach against animal
coronaviruses (Marshall and Enserink, 2004; Scott, 1987).
Our data will be useful for the identification of the
contributing components for such autoimmune responses
if the candidate inactivated SARS-CoV vaccines eventually
demonstrate any similar adverse events in more advanced
In the current study, twelve mAbs against human SARS-
CoV were produced by immunizing the Balb/C mice with
an inactivated patient isolate of SARS-CoV. This is different
from other approaches that produced anti-S mAbs. MAbs
have been isolated from SARS-CoV-infected patients’ IgG
memory B lymphocytes and immortalized with EBV
(Traggiai et al., 2004). Mice receiving the high dose of
one such mAb, S3.1, were protected from viral challenge
(Traggiai et al., 2004). Another study showed that a mAb
specific for the nucleoprotein and five mAbs reacted with
the Spike protein out of the seventeen IgG mAbs screened
by ELISAwere able to neutralize SARS-CoVin vitro (Berry
et al., 2004). A human mAb CR3014, generated by antibody
phage display technology screening a large naive antibody
library, could reduce replication of SARS-CoV if prophy-
lactic administrated to ferret (ter Meulen et al., 2004).
Several mAbs were raised by using recombinant protein
fragments of the SARS CoV S protein (residues 249–667 or
485–625) or N proteins expressed from Escherichia coli
(Berry et al., 2004; Che et al., 2003; Wen et al., 2004).
Synthetic gene fragments from predicted S epitopes have
also been used to raise mAbs to study the S domain
structure (Hua et al., 2004). There were recombinant
antibodies against E and N protein isolated from the mouse
synthetic VH+ VLscFv phage display library with high
binding affinity to the SARS proteins E and N purified from
E. coli (Liu et al., 2004). Our study used the inactivated
vaccine approach to immunize the mice and the resulting
mAbs showed diverse specificities and various biological
Our data confirmed that the dominant neutralizing
domain on S protein is at the S1.1b region which coincides
with the ACE-2 receptor binding region of S protein as
reported previously (Sui et al., 2004). Specificities and
strength of two high titer neutralizing mAbs (1A5 and
2C5) were indistinguishable in our experiment, thus they
probably recognized the same or a closely related epitope.
The relative ease with which these mAbs were generated
further supports the idea that subunit vaccination with S1-
or S1.1b-like constructs may be sufficient to protect
animals and perhaps humans. Two of the three S1.1a-
specific mAbs (2A3 and 1B4) also showed slightly
positive neutralizing activities. However, their low neu-
tralizing titers would not support the role of S1.1a region
as another neutralizing domain. These two mAbs may
recognize a region on S protein which can interfere with
the binding of virion to the viral receptor but may not be
in a direct manner.
Seven mAbs in our panel were specific for S protein with
six showing strong binding to the full-length S protein. The
seventh mAb 2D4 could recognize a processed S antigen
from SARS-CoV with the apparent molecular weight of ~70
kDa (Fig. 5A) and its epitope was mapped to S1.1a region
(Fig. 5B). It is quite unique that 2D4 could not recognize
either the full-length recombinant S protein or the virion-
associated full-length S protein. The direct interpretation of
this finding is that the 2D4 epitope in the S1.1a region may
not be accessible on the full-length S protein given the large
size and heavy glycosylation of SARS-CoV S protein.
Additional analyses are needed to confirm whether this
epitope may become more exposed upon the cleavage of S
into smaller fragments.
As we recently demonstrated, the S protein of SARS-
CoV may go through a step-wise cleavage process in which
the highly specific polyclonal antibodies recognized a group
of low molecular weight S fragments rather than the explicit
S1 or S2 domains (Wang et al., 2005). Most of the reports
suggested that the S protein of SARS-CoV was not
normally processed into S1 and S2 subunits; however, it
could be cleaved by exogenous trypsin (Simmons et al.,
2004; Yao et al., 2004). In the current study, a cleaved S2
fragment at about 70 kDa was identified with S2-specific
mAb 3A3 by Western blot analysis. This finding is
consistent with the observation in a recent publication that
an S2 product was detected by mAb generated by purified
recombinant S2 protein (Wu et al., 2004). These data
revealed important evidence that S protein can be cleaved in
the lysate of SARS-CoV-infected Vero E6 cells, indicating
the presence of proteolytic processing of S protein. This
T.W. Chou et al. / Virology 334 (2005) 134–143
process may be cell dependent because no such cleavage
was observed with recombinant S protein in transfected
293T cells. The availability of specific cellular enzyme may
be critical for such cleavage. The discovery of mAbs 2D4
(specific for processed S1 protein) and 3A3 (specific for S2)
can aid further study on the proteocleavage of S protein.
It is known that the IgG subclasses may show differences
in effector functions like antigen recognition, complement
activation, and cell surface Fc receptor binding (Whitton
and Oldstone, 1996). For example, human antibodies
against viral capsid proteins were found to belong predom-
inantly to the IgG1 and IgG3 subclasses, whereas IgG2 is
the predominant subclass in immune responses against
polysaccharides. Therefore, in this study, the immunoglo-
bulin isotypes were also analyzed although it is known that
the classification of mouse IgG subclasses does not match
with that of human’s. Among 11 IgG mAbs produced from
this study, there are five IgG1, five IgG2a, and one IgG2b.
Two strong neutralizing mAbs 1A5 and 2C5 were IgG2a
and IgG1, respectively. Previously reported neutralizing
mAbs against SARS-CoV included IgG1 (Sui et al., 2004),
IgG2a, or IgG2b subtypes (Gubbins et al., 2005). So, there
is no evidence to suggest that antibody isotypes may play
any roles in determining the neutralizing activities against
In conclusion, we have characterized twelve mAbs
generated against inactivated SARS-CoV. Majority of these
mAbs recognized either the full-length or the processed S
protein. The wide spectrum of S protein epitopes recognized
by this panel of mAbs suggested the overall high immuno-
genicity of S protein associated with SARS-CoV particles.
The antibodies to SARS-CoV S protein included both linear
and conformational epitopes, thus these mAbs are useful
tools for specific applications in SARS research.
Materials and methods
Viruses and cells
The SARS-CoV BJ01 strain (Qin et al., 2003) was
isolated from a SARS patient in Beijing, China, by the
Institute of Microbiology and Epidemiology, Beijing, China
(Fang et al., 2003). Vero E6 cells (2 ? 106cells) were
infected with a multiplicity of infection (MOI) of 0.01 and
cultured at 37 8C with 5% CO2for 36 h. Cells were lysed by
freeze–thaw cycles followed by centrifugation at 6000 rpm
(Beckman 25R, Beckman Coulter Inc., Fullerton, CA) for
20 min. Preparation of inactivated SARS-CoV was pre-
viously reported (Tang et al., 2004). Briefly, the virus was
inactivated by h-propiolactone (1:2000 dilution) from the
commercial stock solution (1.146 g/mL, Sigma-Aldrich, St.
Louis, MO) and incubated at 4 8C for 24–72 h, then kept at
37 8C for 2 h. The inactivated virus suspension was
centrifuged at 6000 rpm (Beckman 25R, Beckman Coulter
Inc.) at 4 8C for 30 min. The supernatant was harvested and
concentrated with PEG20000 (Sigma-Aldrich, Inc.), then
concentrated by centrifugation (30,000 ? g, 20 min) using
Cetriplus YM-100 (Millipore Corp., Bedford, MA), fol-
lowed with purification by Sepharose 4FF column chroma-
tography (Tang et al., 2004).
The SARS-CoV Urbani strain used for neutralization
assay was obtained from U.S. Center for Disease Control
and Prevention (Atlanta, GA). To inactivate this virus for
ELISA and Western blot analysis, the virus stocks were first
filtered through a 0.45-Am membrane to remove cell debris.
Then the virus was inactivated with a buffer containing
0.2% SDS and 1% Triton-X 100 in Tris-buffered saline (pH
7.6) for 1 h at 4 8C. The inactivation of SARS-CoV was
confirmed by using a Standard Operational Procedure (SOP)
approved by the Institutional Biosafety Committee at the
University of Massachusetts Medical School.
Production of monoclonal antibodies (mAbs) by inactivated
Six-week-old Balb/C mice were injected with 2 Ag of
inactivated SARS-CoV BJ01 strain mixed with Complete
Freund’s Adjuvant (CFA, H37 Ra; Difco) on day 1. On day
10, the mice received 2 Ag of inactivated SARS-CoV in
incomplete Freund’s Adjuvant (IFA). Mice were boosted 3
weeks later with 0.8 Ag of inactivated SARS-CoV without
any adjuvants. Animals were euthanized 3 days later and the
mouse splenocytes were fused with SP2/0 myeloma cells.
Positive hybridomas were cloned out and supernatants were
screened via ELISA using inactivated virus as coating
antigen. The fused cell lines positive in ELISA were grown
briefly in cell culture and then injected into mouse
peritoneum. The ascites fluid which contains high titer
monoclonal antibodies against SARS-CoV was collected for
further analysis. Immunoglobulin isotyping was performed
using a commercial dipstick test (Roche) according to the
Production of recombinant S, N, M, and E proteins from
codon-optimized DNA expression vector
The codon usage of published SARS-CoV S, N, M, and
E gene sequences (Marra et al., 2003; Rota et al., 2003) was
analyzed by the MacVector software (V. 7.2, Accelrys, San
Diego, CA) against that of the Homo sapiens genome. The
less optimal codons in these genes were changed to codons
more preferred in mammalian systems to promote higher
expressions of these structure proteins. These codon-
optimized genes were chemically synthesized by Geneart
(Regensburg, Germany) and individually subcloned into the
DNA vaccine vector pSW3891 (Wang et al., 2004). The
human tissue plasminogen activator (tPA) leader sequence
has replaced the S natural leader in these constructs as
reported (Wang et al., 2005).
Each SARS-CoV DNA plasmid transformed in E. coli
(HB101 strain) was confirmed by DNA sequencing before
T.W. Chou et al. / Virology 334 (2005) 134–143
large amounts of DNA plasmids were prepared with a Mega
purification kit (Qiagen, Valencia, CA). The recombinant
proteins used in this study were produced from transiently
transfected 293T cells. DNA constructs were first trans-
fected into 293T cells using the calcium phosphate
precipitation method as reported (Wang et al., 2005).
Briefly, 2 ? 106293T cell of 50% confluence in a 60-mm
dish were transfected with 10 Ag of plasmid DNA and were
harvested 72 h later for ELISA or Western blot use.
ELISA (enzyme-linked immunosorbent assay)
High-binding 96-well flat-bottom plates (Costar) were
coated overnight at 4 8C with 100 Al of SARS-CoVantigens
at 1 Ag/ml. To enhance S antigen binding, the plates were
first incubated with 100 Al of ConA (50 Ag/ml) for 1 h at
room temperature, and washed 5 times with PBS containing
0.1% Triton X-100. The plates were then blocked with 200
Al/well of blocking buffer (5% non-fat dry milk, 4% whey,
0.5% Tween 20 in PBS at pH 7.2) for 1 h. After five
washings, 100 Al of serially diluted mAbs was added in
duplicate wells and incubated for 1 h. After another set of
washings, the plates were incubated for 1 h at room
temperature with 100 Al of anti-mouse IgG conjugated with
horseradish peroxidase (SouthernBiotech, Birmingham, AL)
diluted at 1:1000 in Whey dilution buffer (4% Whey, 0.5%
Tween 20 in PBS). After the final washing, the plates were
developed with 3,3V ,5,5V Tetramethybenzidine solution at
100 Al/well (Sigma, St. Louis, MO) for 3.5 min. The
reactions were stopped by adding 25 Al of 2 M H2SO4, and
the plates were read at OD 450 nm. The quantities of
mouse-purified mAbs were determined using standard of
IgG and IgM (Southern Biotech) in ELISA assay.
Western blot analysis
Purified SARS-CoV virions and transiently expressed S
antigens were first run by SDS–PAGE electrophoresis. The
samples were heated at 90 8C for 5 min in sample buffer (50
bromophenol blue, 10% glycerol), and equal amounts of
The gels were electroblotted to PVDF membranes (Bio-Rad)
using 80 mA for 2 h, then blocked overnight at 4 8C in
blocking buffer (0.2% I-block, 0.1% Tween 20 in 1? PBS).
Membranes were incubated with a 1:200 dilution of mAbs.
After being washed, blots were incubated with alkaline
phosphatase-conjugated goat anti-rabbit IgG at 1:5000
dilution, and signals were detected using a chemilumines-
cence Western-Light Kit (Tropix, Bedford, MA).
In vitro neutralization assays
The SARS-CoV neutralization assays using neutral red
staining were performed with triplicate testing wells in 96-
well flat bottom plates in the BL-3 laboratory. Initially, 400
TCID50of virus in 50 Al/well was incubated with 50 Al of
serially diluted rabbit sera or tissue culture medium for 1 h
at 37 8C. After incubation, 100 Al of Vero E6 cells (20,000
cells) was added to each well at MOI of 0.02. On day 5 of
infection when more than 70% of the cells formed CPE in
viral control wells, the culture medium was removed from
the testing wells and 100 Al of 10% neutral red in DMEM
medium was added to each well. After incubation for 1 h at
37 8C, the neutral red medium was removed, the plates were
washed twice with PBS (pH 7.2), and 100 Al of acid alcohol
(1% acetic acid in 50% ethanol) was added to each well.
After incubation for 30 min at room temperature, the
absorbance was read at A540. Percent of neutralization at a
given serum dilution was determined by calculating the
difference in absorption (A540) between test wells (cells,
serum sample, and virus) and virus control wells (cells and
virus), and dividing this result by the difference in
absorption between cell control wells (cells only) and virus
control wells (Montefiori et al., 1988). In our assay system,
sera were considered positive for neutralizing antibody
activities when the titers are above 50% inhibition as
compared with the virus controls.
This study was supported in part by NIH grants AI 40337
and AI 44338 (S. Lu). The project also used core facility
resources at the University of Massachusetts Medical
School supported by NIH grant 5P30DK32520 from the
NIDDKD. We would also like to thank US Center for
Diseases Control and Prevention (Atlanta, GA) for provid-
ing SARS-CoV Urbani strain for this study.
Arbely, E., Khattari, Z., Brotons, G., Akkawi, M., Salditt, T., Arkin,
I.T., 2004. A highly unusual palindromic transmembrane helical
hairpin formed by SARS coronavirus E protein. J. Mol. Biol. 341
Berry, J.D., Jones, S., Drebot, M.A., Andonov, A., Sabara, M., Yuan, X.Y.,
Weingartl, H., Fernando, L., Marszal, P., Gren, J., Nicolas, B.,
Andonova, M., Ranada, F., Gubbins, M.J., Ball, T.B., Kitching, P.,
Li, Y., Kabani, A., Plummer, F., 2004. Development and character-
isation of neutralising monoclonal antibody to the SARS-coronavirus. J.
Virol. Methods 120 (1), 87–96.
Bosch, B.J., van der Zee, R., de Haan, C.A., Rottier, P.J., 2003. The
coronavirus spike protein is a class I virus fusion protein: structural and
functional characterization of the fusion core complex. J. Virol. 77 (16),
Callow, K.A., Parry, H.F., Sergeant, M., Tyrrell, D.A., 1990. The time
course of the immune response to experimental coronavirus infection of
man. Epidemiol. Infect. 105 (2), 435–446.
Che, X.Y., Qiu, L.W., Pan, Y.X., Xu, H., Hao, W., Liao, Z.Y., Mei, Y.B.,
Zhang, L.Y., Wan, Z.Y., Yuan, G.Y., Huang, Z., 2003. Rapid and
efficient preparation of monoclonal antibodies against SARS-associated
coronavirus nucleocapsid protein by immunizing mice. Di Yi Jun Yi Da
Xue Xue Bao 23 (7), 640–642.
T.W. Chou et al. / Virology 334 (2005) 134–143
Darnell, M.E., Subbarao, K., Feinstone, S.M., Taylor, D.R., 2004.
Inactivation of the coronavirus that induces severe acute respiratory
syndrome, SARS-CoV. J. Virol. Methods 121 (1), 85–91.
Davies, H.A., Dourmashkin, R.R., Macnaughton, M.R., 1981. Ribonucleo-
protein of avian infectious bronchitis virus. J. Gen. Virol. 53 (Pt. 1),
Fang, L.Q., Zhang, P.H., Yang, B.A., Wu, X.M., Zhao, Q.M., Liu, W.,
Liu, H., Deng, Y.Q., Zhan, L., Han, W.G., Lu, F.S., Wu, J.S., Yang,
H., Zhu, Q.Y., Cao, W.C., 2003. The application of indirect immuno-
fluorescence assay in the diagnosis of severe acute respiratory
syndrome. Zhonghua Liu Xing Bing Xue Za Zhi 24 (6), 484–486.
Gubbins, M.J., Plummer, F.A., Yuan, X.Y., Johnstone, D., Drebot, M.,
Andonova, M., Andonov, A., Berry, J.D., 2005. Molecular character-
ization of a panel of murine monoclonal antibodies specific for the
SARS-coronavirus. Mol. Immunol. 42 (1), 125–136.
He, L., Ding, Y.Q., Che, X.Y., Zhang, Q.L., Huang, Z.X., Wang, H.J., Shen,
H., Li, Z.G., Cai, J.J., Zhang, J.H., Geng, J., Li, X., Zhang, W.L., Han,
H.X., Kang, W., Yang, L., Lu, Y.D., 2003. Expression of the
monoclonal antibody against nucleocapsid antigen of SARS-associated
coronavirus in autopsy tissues from SARS patients. Di Yi Jun Yi Da
Xue Xue Bao 23 (11), 1128–1130.
He, R., Dobie, F., Ballantine, M., Leeson, A., Li, Y., Bastien, N., Cutts, T.,
Andonov, A., Cao, J., Booth, T.F., Plummer, F.A., Tyler, S., Baker, L.,
Li, X., 2004a. Analysis of multimerization of the SARS coronavirus
nucleocapsid protein. Biochem. Biophys. Res. Commun. 316 (2),
He, Y., Zhou, Y., Siddiqui, P., Jiang, S., 2004b. Inactivated SARS-CoV
vaccine elicits high titers of spike protein-specific antibodies that block
receptor binding and virus entry. Biochem. Biophys. Res. Commun.
325 (2), 445–452.
Hiscox, J.A., Wurm, T., Wilson, L., Britton, P., Cavanagh, D., Brooks, G.,
2001. The coronavirus infectious bronchitis virus nucleoprotein local-
izes to the nucleolus. J. Virol. 75 (1), 506–512.
Holmes, K., 2001. Coronaviruses. 4th ed. In: Knipe, D., Howley,
P., Griffin, D., Lamb, R., Martin, M., Roizman, B., Straus, S.
(Eds.), Fields Virology, vol. 1. Lippincott Williams and Wilkins,
pp. 1187–1203. 2 vols.
Hua, R., Zhou, Y., Wang, Y., Hua, Y., Tong, G., 2004. Identification of two
antigenic epitopes on SARS-CoV spike protein. Biochem. Biophys.
Res. Commun. 319 (3), 929–935.
Huang, Y., Yang, Z.Y., Kong, W.P., Nabel, G.J., 2004. Generation of
synthetic severe acute respiratory syndrome coronavirus pseudopar-
ticles: implications for assembly and vaccine production. J. Virol. 78
Kraaijeveld, C.A., Reed, S.E., Macnaughton, M.R., 1980. Enzyme-linked
immunosorbent assay for detection of antibody in volunteers exper-
imentally infected with human coronavirus strain 229 E. J. Clin.
Microbiol. 12 (4), 493–497.
Krokhin, O., Li, Y., Andonov, A., Feldmann, H., Flick, R., Jones, S.,
Stroeher, U., Bastien, N., Dasuri, K.V., Cheng, K., Simonsen, J.N.,
Perreault, H., Wilkins, J., Ens, W., Plummer, F., Standing, K.G., 2003.
Mass spectrometric characterization of proteins from the sars virus: a
preliminary report. Mol. Cell Proteomics 2 (5), 346–356.
Ksiazek, T.G., Erdman, D., Goldsmith, C.S., Zaki, S.R., Peret, T., Emery,
S., Tong, S., Urbani, C., Comer, J.A., Lim, W., Rollin, P.E., Dowell,
S.F., Ling, A.E., Humphrey, C.D., Shieh, W.J., Guarner, J., Paddock,
C.D., Rota, P., Fields, B., DeRisi, J., Yang, J.Y., Cox, N., Hughes, J.M.,
LeDuc, J.W., Bellini, W.J., Anderson, L.J., 2003. A novel coronavirus
associated with severe acute respiratory syndrome. N. Engl. J. Med. 348
Lai, M.M., Holmes, K., 2001. Coronarviridae: the viruses and their
replication. 4 ed. In: Knipe, D., Howley, P., Griffin, D., Lamb, R.,
Martin, M., Roizman, B., Straus, S. (Eds.), Fields Virology, vol. 1.
Lippincott Williams and Wilkins, pp. 1163–1185. 2 vols.
Lewicki, D.N., Gallagher, T.M., 2002. Quaternary structure of coronavirus
spikes in complex with carcinoembryonic antigen-related cell adhesion
molecule cellular receptors. J. Biol. Chem. 277 (22), 19727–19734.
Li, W., Moore, M.J., Vasilieva, N., Sui, J., Wong, S.K., Berne, M.A.,
H., Farzan, M., 2003. Angiotensin-converting enzyme 2 is a functional
receptor for the SARS coronavirus. Nature 426 (6965), 450–454.
Liu, S., Xiao, G., Chen, Y., He, Y., Niu, J., Escalante, C.R., Xiong, H.,
Farmar, J., Debnath, A.K., Tien, P., Jiang, S., 2004. Interaction between
heptad repeat 1 and 2 regions in spike protein of SARS-associated
coronavirus: implications for virus fusogenic mechanism and identi-
fication of fusion inhibitors. Lancet 363 (9413), 938–947.
Luo, Z., Weiss, S.R., 1998. Roles in cell-to-cell fusion of two conserved
hydrophobic regions in the murine coronavirus spike protein. Virology
244 (2), 483–494.
Marra, M.A., Jones, S.J., Astell, C.R., Holt, R.A., Brooks-Wilson, A.,
Butterfield, Y.S., Khattra, J., Asano, J.K., Barber, S.A., Chan, S.Y.,
Cloutier, A., Coughlin, S.M., Freeman, D., Girn, N., Griffith, O.L.,
Leach, S.R., Mayo, M., McDonald, H., Montgomery, S.B., Pandoh,
P.K., Petrescu, A.S., Robertson, A.G., Schein, J.E., Siddiqui, A.,
Smailus, D.E., Stott, J.M., Yang, G.S., Plummer, F., Andonov, A.,
Artsob, H., Bastien, N., Bernard, K., Booth, T.F., Bowness, D., Czub,
M., Drebot, M., Fernando, L., Flick, R., Garbutt, M., Gray, M., Grolla,
A., Jones, S., Feldmann, H., Meyers, A., Kabani, A., Li, Y., Normand,
S., Stroher, U., Tipples, G.A., Tyler, S., Vogrig, R., Ward, D., Watson,
B., Brunham, R.C., Krajden, M., Petric, M., Skowronski, D.M., Upton,
C., Roper, R.L., 2003. The genome sequence of the SARS-associated
coronavirus. Science 300 (5624), 1399–1404.
Marshall, E., Enserink, M., 2004. Medicine. Caution urged on SARS
vaccines. Science 303 (5660), 944–946.
Montefiori, D.C., Robinson Jr., W.E., Schuffman, S.S., Mitchell, W.M.,
1988. Evaluation of antiviral drugs and neutralizing antibodies to
human immunodeficiency virus by a rapid and sensitive microtiter
infection assay. J. Clin. Microbiol. 26 (2), 231–235.
Qin, E., Zhu, Q.Y., Yu, M., Fan, B.C., Chang, G.H., Si, B.Y., Yang, B.A.,
Peng, W.M., Jiang, T., Liu, B.H., Deng, Y.Q., Liu, H., Zhang, Y., Wang,
C., Li, Y.Q., Gan, Y.H., Li, X.Y., Lu, F.S., Tan, G., Cao, W.C., Yang,
R.F., Wang, J., Li, W., Xu, Z.Y., Li, Y., Wu, Q.F., Lin, W., Chen, W.J.,
Tang, L., Deng, Y.F., Han, Y.J., Li, C.F., Lei, M., Li, G.Q., Li, W.J., Lu,
H., Shi, J.P., Tong, Z.Z., Zhang, F., Li, S.G., Liu, B., Liu, S.Q., Dong,
W., Wang, J., Wong, G.K.S., Yu, J., Yang, H.M., 2003. A complete
sequence and comparative analysis of a SARS-associated virus (Isolate
BJ01). Chin. Sci. Bull. 48 (10), 941–948.
Rota, P.A., Oberste, M.S., Monroe, S.S., Nix, W.A., Campagnoli, R.,
Icenogle, J.P., Penaranda, S., Bankamp, B., Maher, K., Chen, M.H.,
Tong, S., Tamin, A., Lowe, L., Frace, M., DeRisi, J.L., Chen, Q., Wang,
D., Erdman, D.D., Peret, T.C., Burns, C., Ksiazek, T.G., Rollin, P.E.,
Sanchez, A., Liffick, S., Holloway, B., Limor, J., McCaustland, K.,
Olsen-Rasmussen, M., Fouchier, R., Gunther, S., Osterhaus, A.D.,
Drosten, C., Pallansch, M.A., Anderson, L.J., Bellini, W.J., 2003.
Characterization of a novel coronavirus associated with severe acute
respiratory syndrome. Science 300 (5624), 1394–1399.
Scott, F.W., 1987. Immunization against feline coronaviruses. Adv. Exp.
Med. Biol. 218, 569–576.
Simmons, G., Reeves, J.D., Rennekamp, A.J., Amberg, S.M., Piefer, A.J.,
Bates, P., 2004. Characterization of severe acute respiratory syndrome-
associated coronavirus (SARS-CoV) spike glycoprotein-mediated viral
entry. Proc. Natl. Acad. Sci. U.S.A. 101 (12), 4240–4245.
Spiga, O., Bernini, A., Ciutti, A., Chiellini, S., Menciassi, N., Finetti, F.,
Causarono, V., Anselmi, F., Prischi, F., Niccolai, N., 2003. Molecular
modelling of S1 and S2 subunits of SARS coronavirus spike
glycoprotein. Biochem. Biophys. Res. Commun. 310 (1), 78–83.
Sui, J., Li, W., Murakami, A., Tamin, A., Matthews, L.J., Wong, S.K.,
Moore, M.J., St. Clair Tallarico, A., Olurinde, M., Choe, H., Anderson,
L.J., Bellini, W.J., Farzan, M., Marasco, W.A., 2004. Potent neutraliza-
tion of severe acute respiratory syndrome (SARS) coronavirus by a
human mAb to S1 protein that blocks receptor association. Proc. Natl.
Acad. Sci. U.S.A. 101 (8), 2536–2541.
Takasuka, N., Fujii, H., Takahashi, Y., Kasai, M., Morikawa, S., Itamura,
S., Ishii, K., Sakaguchi, M., Ohnishi, K., Ohshima, M., Hashimoto, S.,
T.W. Chou et al. / Virology 334 (2005) 134–143
Kimberly, R., Tashiro, M., Yoshikura, H., Takemori, T., Tsunetsugu- Download full-text
Yokota, Y., 2004. A subcutaneously injected UV-inactivated SARS
coronavirus vaccine elicits systemic humoral immunity in mice. Int.
Immunol. 16 (10), 1423–1430.
Tang, L., Zhu, Q., Qin, E., Yu, M., Ding, Z., Shi, H., Cheng, X., Wang, C.,
Chang, G., Fang, F., Chang, H., Li, S., Zhang, X., Chen, X., Yu, J.,
Wang, J., Chen, Z., 2004. Inactivated SARS-CoV vaccine prepared
from whole virus induces a high level of neutralizing antibodies in
BALB/c mice. DNA Cell Biol. 23 (6), 391–394.
ter Meulen, J., Bakker, A.B., van den Brink, E.N., Weverling, G.J., Martina,
B.E., Haagmans, B.L., Kuiken, T., de Kruif, J., Preiser, W., Spaan, W.,
Gelderblom, H.R., Goudsmit, J., Osterhaus, A.D., 2004. Human
monoclonal antibody as prophylaxis for SARS coronavirus infection
in ferrets. Lancet 363 (9427), 2139–2141.
Traggiai, E., Becker, S., Subbarao, K., Kolesnikova, L., Uematsu, Y.,
Gismondo, M.R., Murphy, B.R., Rappuoli, R., Lanzavecchia, A., 2004.
An efficient method to make human monoclonal antibodies from
memory B cells: potent neutralization of SARS coronavirus. Nat. Med.
10 (8), 871–875.
Wang, S., Heilman, D., Liu, F., Giehl, T., Joshi, S., Huang, X., Chou, T.H.,
Goguen, J., Lu, S., 2004. A DNA vaccine producing LcrV antigen in
oligomers is effective in protecting mice from lethal mucosal challenge
of plague. Vaccine 22 (25–26), 3348–3357.
Wang, S., Chou, T.W., Sakhatskyy, P.V., Huang, S., Lawrence, J.M., Cao,
H., Huang, X., Lu, S., 2005. Identification of two neutralizing regions
on SARS-CoV spike glycoprotein produced from mammalian expres-
sion system. J. Virol. 79 (3), 1906–1910.
Wen, K., Mei, Y.B., Qiu, L.W., Liao, Z.Y., Yuen, K.Y., Che, X.Y., 2004.
Preparation and characterization of monoclonal antibodies against S1
domain at N-terminal residues 249 to 667 of SARS-associated
coronavirus S1 protein. Di Yi Jun Yi Da Xue Xue Bao 24 (1), 1–6.
Whitton, J.L., Oldstone, B.A., 1996. Chapter 12. Immune response to
virus. 3rd ed. In: Field, B.N., Knipe, D.M., Howley, P.M. (Eds.),
Fields Virology, vol. 1. Lippincott-Raven Publisher, Philadephia, PA,
Wong, S.K., Li, W., Moore, M.J., Choe, H., Farzan, M., 2004. A 193-amino
acid fragment of the SARS coronavirus S protein efficiently binds
angiotensin-converting enzyme 2. J. Biol. Chem. 279 (5), 3197–3201.
Wu, X.D., Shang, B., Yang, R.F., Yu, H., Ma, Z.H., Shen, X., Ji, Y.Y., Lin,
Y., Wu, Y.D., Lin, G.M., Tian, L., Gan, X.Q., Yang, S., Jiang, W.H.,
Dai, E.H., Wang, X.Y., Jiang, H.L., Xie, Y.H., Zhu, X.L., Pei, G., Li, L.,
Wu, J.R., Sun, B., 2004. The spike protein of Severe Acute Respiratory
Syndrome (SARS) is cleaved in virus infected Vero-E6 cells. Cell Res.,
Xiong, S., Wang, Y.F., Zhang, M.Y., Liu, X.J., Zhang, C.H., Liu, S.S., Qian,
C.W., Li, J.X., Lu, J.H., Wan, Z.Y., Zheng, H.Y., Yan, X.G., Meng,
M.J., Fan, J.L., 2004. Immunogenicity of SARS inactivated vaccine in
BALB/c mice. Immunol. Lett. 95 (2), 139–143.
Yao, Y.X., Ren, J., Heinen, P., Zambon, M., Jones, I.M., 2004. Cleavage
and serum reactivity of the severe acute respiratory syndrome
coronavirus spike protein. J. Infect. Dis. 190 (1), 91–98.
Ying, W., Hao, Y., Zhang, Y., Peng, W., Qin, E., Cai, Y., Wei, K., Wang, J.,
Wan, P., Chen, T., Du, C., Li, D., Wan, J., Kuai, X., Li, W., Shi, R., Wei,
H., Cao, C., Yu, M., Liu, H., Dong, F., Wang, D., Zhang, X., Qian, X.,
Acute Respiratory Syndrome coronavirus. Proteomics 4 (2), 492–504.
Zhou, T., Wang, H., Luo, D., Rowe, T., Wang, Z., Hogan, R.J., Qiu, S.,
Bunzel, R.J., Huang, G., Mishra, V., Voss, T.G., Kimberly, R., Luo, M.,
2004. An exposed domain in the severe acute respiratory syndrome
coronavirus spike protein induces neutralizing antibodies. J. Virol. 78
T.W. Chou et al. / Virology 334 (2005) 134–143