https://www.abbs.info
e-mail: [email protected]
ISSN 0582-9879 Acta Biochim et Biophysica Sinica 2004, 36(1):58-64 CN 31-1300/Q

Expression of
Hepatitis C Virus E2 Ectodomain in E. coli and Its Application in the
Detection of Anti-E2 Antibodies in Human Sera
Jing LIU1#, Xin-Xin ZHANG1, 2#, Shen-Ying ZHANG2, Min
LU2, Yu-Ying KONG1, Yuan WANG1*, and Guang-Di LI1*
( 1 State Key Laboratory of
Molecular Biology, Institute of Biochemistry and Cell Biology, Shanghai
Institutes for Biological Sciences, the Chinese Academy of Sciences, Shanghai
200031, China; 2 Department
of Infectious Diseases, Ruijin Hospital, Second Medical University of Shanghai,
Shanghai 200025, China )

Abstract The second envelope glycoprotein (E2) of hepatitis
C virus has been shown to bind human target cells and has become a major target
for the development of anti-HCV vaccines. Anti-E2 antibodies have been
suggested to be of clinical significance in diagnosis, treatment and prognosis
of hepatitis C. However, large-scale expression and purification of E2 proteins
in mammalian cells is difficult. As an alternative, E2 fragment (aa 385–730)
with a four-amino-acid mutation (aa 568–571 PCNI to RVTS) was expressed as
hexa-histidine-tagged full length protein [E2N730(m)] in E. coli and
purified to over 85% purity. Purified E2N730(m) was specifically recognized by
homologous hepatitis C patient serum in Western blot, suggesting that it
displayed E2-specific antigenicity. Rabbit antiserum raised against E2N730(m)
recognized E2 glycoproteins expressed in mammalian cells in Western blot.
Purified E2N730(m) was used to detect anti-E2 antibodies in human sera and
showed better specificity and sensitivity than previously reported C-terminally
truncated E2 fragment (aa 385–565). Association between anti-E2 antibodies in
patient sera and HCV RNA status was also demonstrated using this E. coli-derived
protein. E2N730(m) might serve as an inexpensive alternative to mammalian
cell-expressed E2 proteins in clinical and research applications.

Key words hepatitis C virus; envelope protein; E2;
expression and purification; Escherichia coli

Hepatitis C virus (HCV) is the major etiological agent of
both community-acquired and post-transfusion non-A, non-B hepatitis [1]. In
1998, it was estimated that 3% of the world population (about 170 million) was
infected with HCV [2]. Prognosis of HCV infection is poor, with approximately
85% of patients developing chronic infection, and about 20% of the chronic
cases progressing onto cirrhosis and/or hepatocellular carcinoma [3]. Lack of
effective vaccines and satisfactory treatments makes HCV a global health
threat.

HCV is an enveloped plus-strand RNA virus and has been
classified as the sole member of the hepacivirus genus of the Flaviviridae family
[4]. Of the viral structural proteins, E1 and E2 are predicted to be
glycosylated type I membrane proteins and generally believed to constitute the
protein components of virion membrane [5,6].

E2 encompasses aa 384–746 of the HCV polyprotein, with
the extremely hydrophobic aa 718–746 region as its putative transmembrane
domain (TMD) [7]. In addition to the TMD, C-terminal aa 662–717 region of E2
ectodomain (aa 384– 717) is also highly hydrophobic. E2 has been suggested to
play an important role in HCV binding and entering into target cells [8–10].
Vaccination studies in chimpanzees using E2-based glycoprotein or DNA vaccines
have shown that limited but measurable protection could be achieved [11]. Data
from natural infection cases have also associated natural resolution of
infection with certain types of anti-E2 antibodies [12,13]. Therefore, E2 has
become a major target in anti-HCV vaccine research. Humoral immune responses
against E2 also have diagnostic significance, since there have been studies
showing that testing for antibodies against E2 could improve the performance of
current HCV EIA kits, which do not incorporate any forms of envelope proteins
[14–17].

In this work, E2 fragment (aa 385–730) covering the E2
ectodomain and the upstream half of TMD was expressed as hexa-histidine-tagged
protein [E2N730(m)] in E. coli. A four-residue mutation was introduced
to enhance full-length expression. Purified E2N730(m) displayed E2-specific
antigenicity and rabbit antiserum raised against E2N730(m) was able to
recognize E2 glycoproteins expressed in mammalian cells in Western blot.
Anti-E2 antibodies in human sera could be detected in EIA with E2N730(m) and
association between the presence of anti – E2 antibodies and serum HCV RNA
status was demonstrated. E2N730(m) might serve as an inexpensive alternative to
mammalian cell-expressed E2 proteins in clinical and research applications,
whereas rabbit anti- E2N730(m) could be a useful tool in biochemical and
vaccinological studies of E2.

Materials and Methods

Plasmids and bacterial host

pUC18/CE1E2 containing C, E1 and E2 coding sequences of
HCV (subtype 1b) was provided by Professor WANG Yu of Peking University,
Beijing, China (GenBank accession No. D10934) [18]. pQE8 is a N-terminal
hexahistidine fusion expression vector from Qiagen GmbH, Hilden, Germany.
pQE8/E2C’730(m) is a previously described plasmid expressing E2 aa 567–730
fragment in pQE8 background. E2 coding sequences of pQE8/E2C’730 (m) were
mutated from PCNI to RVTS at aa 568–571, due to a spontaneous frameshift by one
nucleotide affecting 36 basepairs [19]. There was also an additional
hexahistidine tag at the C-terminal of E2 coding sequences in pQE8/E2C’730(m). E.
coli strain TG-1 was used as cloning and expression host.

Construction of expression plasmid pQE8/E2N730(m)

DNA sequences encoding E2 aa 385–700 were amplified from
pUC18/CE1E2 using the primer set: 5′– GCGTTGACGGATCCACCTACGTG-3′(upstream); 5′–GCGAAGCTTGCACGTCCACGATG -3′(downstream).Amplified fragment was cloned between the BamHI
and HindIII sites on pQE8 to create pQE8/E2N (designated pQE8/E2-316 in
reference ［20］). Coding sequences for mutated aa 567–730 were amplified
from pQE8/E2C’730 (m)[19] using the primer set:  5′-GTCGGGCCCCCGTGTAACATCG-3′(upstream); 5′-CAAGCTAGCTTGGATTCTCACC-3′(downstream), and then used to replace coding sequences
for aa 567–700 on pQE8/E2N. The obtained recombinant plasmid was designated
pQE8/ E2N730(m), which carries the coding sequences for E2 aa 385–730 with aa
568–571 mutated from PCNI to RVTS [19]. All PCR and recombinant cloning steps
were performed according to standard protocols. Sequences of plasmids used for
expression were confirmed by automatic sequencing.

Expression and purification of recombinant E2 protein

Freshly saturated recombinant TG-1 culture was inoculated
into fresh LB media at 1∶100. Two
hours after inoculation, expression was induced by adding IPTG to a final
concentration of 1 mmol/L. Cells were harvested 6 hours later by centrifugation
and stored at –20 ℃.

Solubility analysis and purification of expression
products were performed as previously described [19,20]. Briefly, harvested
bacteria were resuspended in PBS, sonicated on ice-bath, and centrifuged. The
soluble and insoluble fractions were analyzed for the presence of expression
products. Insoluble recombinant E2 proteins were extracted with 6 mol/L Gu
HCl/100 mmol/L â-ME/ PBS (pH 8.0),
centrifuged, and diluted four fold with 6 mol/L Gu HCl/PBS (pH 8.0) before
loading onto preequilibrated Ni2+-NTA agarose (Qiagen). The gel matrices were sequentially
washed with 6 mol/L Gu・HCl/20 mmol/ L â-ME/PBS (pH 6.3) and 8
mol/L urea/20 mmol/L â-ME/ PBS (pH 6.3), and
then eluted with 8 mol/L urea/20 mmol/ L â-ME/PBS (pH
4.3).

E2 proteins expressed in mammalian cells

Expression of the same E2 gene in recombinant vaccinia
virus system was done by co-infecting HeLa cells with vTT7 and vCEH-2 as
previously described [21]. Recombinant vaccinia virus vCEH-2 contained coding
sequences of HCV polyprotein aa 1–730 under the control of T7 promoter, whereas
vTT7 encoded the T7 polymerase required for expression. Briefly, HeLa cells
were coinfected with vTT7 and vCEH-2 at a multiplicity of infection of 4∶4∶1 (vTT7∶vCEH-2∶cell) and
cultured

for 48 hours. Cells were collected by scraping, washed
with 4 ℃ PBS and stored at –20 ℃.

Protein analysis

SDS-PAGE under reducing conditions and Western blot were
conducted according to standard protocols. In Western blot, first antibody was diluted
1∶100 or 1∶500 for human sera and 1∶1000 for rabbit sera, and second antibody [HRP-labeled protein A (Sigma)
or swine anti-rabbit Ig (DaKo)] was diluted 1∶1000. Blots were developed using the ECL method (PerfectBio).

Animal and human sera

One 1.5 kg female rabbit (Shanghai Laboratory Animal
Center) was immunized subcutaneously on the back with 300 mg purified recombinant E2 protein emulsified in complete Freud’s adjuvant
and boosted 4 and 8 weeks later with the same amount of antigen emulsified in incomplete
Freud’s adjuvant. One week after the last boosting, total blood was collected
through the carotid artery and serum was prepared according to standard
procedures.

Human serum S94 was collected from a Chinese patient with
chronic hepatitis C and provided by Professor Yu WANG. The HCV cDNA used in
this work was cloned from the same patient. Other human sera were collected
from Chinese hepatitis patients and healthy blood donors.

Detection of anti-HCV antibodies and HCV RNA

Anti-HCV antibodies in human sera were detected using UBI
HCV EIA 4.0 (United Biomedical Inc.) according to manufacturer’s instructions.
For the detection of anti- E2 antibodies in human sera, polystyrene microplates
(Nalge Nunc International) were coated with purified recombinant E2 at 0.15 mg/hole in 100 ml 50 mmol/L carbonate
buffer at 4 ℃ for 18 h. The
microplates were then blocked with 1% BSA/2% inactivated new-born calf serum/
PBS at 37 ℃ for 2 h.
Human sera were 1∶20 diluted
and secondary antibody (HRP-labeled goat antihuman- Ig) was 1∶150 diluted in blocking buffer. Incubation was continued
for 30 minutes at 37 ℃ followed by
thorough washing with PBST. Color was developed using TMB substrate and
developing was stopped by adding 1 mol/L HCl according to standard protocols. A450 was measured
using a microplate reader. The mean A450 of 100 healthy blood
donors’ sera multiplied by 2.1 was set as cut-off value for determining
positivity. Anti-E2 antibodies in post-immune rabbit sera were detected and
titrated with similar method using HRP-labeled swine anti-rabbit Ig as
secondary antibody. HCV RNA in human sera was detected using HCV Gene Detection
Kit from Shanghai Forward Biomedical Ltd. based on RT-PCR/DNA-EIA methodology.
All EIA tests were done in duplicates and the mean absorbance value was used.

Results and Discussion

Construction of recombinant plasmid expressing E2
ectodomain

In our efforts to express different fragments of HCV E2
protein ectodomain in E. coli, we found that aa 566–622 region of E2 had
a negative effect on expression in E. coli, resulting in low or no
production of full-length protein [19,20]. However, we identified a spontaneous
mutation affecting aa 568–571 of E2 which could counteract such a negative
effect and significantly enhance full-length expression of fragments containing
this region [19]. This mutation is a frameshift of a single nucleotide,
changing aa 568–571 of E2 from PCNI to RVTS. Sequences encoding aa 567–730 of
E2 harboring this mutation were amplified by PCR from pQE8/E2C‘730(m), which expressed
mutated aa 567–730 of E2[19]. The amplified sequences were used to replace aa
567–700 coding sequences in pQE8/E2N, which expressed aa 385–700 of E2 [20].
The resultant expression plasmid was designated pQE8/E2N730(m) and carried
coding sequences for aa 385–730 of E2 fused to hexa-histidine tags at both
termini, with aa 568–571 mutated from PCNI to RVTS. Comparison of E2 coding
sequences in aa 568–571 region between pQE8/E2N and pQE8/-E2N730(m) is shown in
Fig. 1.

Fig. 1 E2 aa 568–571 sequences in pQE8/E2N and pQE8/-
E2N730(m)

Amino acid residues affected by mutation were shown in italic. ApaI
site used for cloning is underlined.

Expression and purification of recombinant E2 fusion
protein

Expression from pQE8/E2N730(m) was induced with IPTG. A
prominent band of approximately 42 kD was observed after induction [Fig. 2(A),
lane 2], matching the predicted molecular weight of 41.8 kD of full-length
product. In contrast, several similar constructs lacking the aa 568–571
mutation only expressed barely detectable level of full-length protein (data
presented in reference [19]). Densitometric scanning showed that this band
constituted over 10% of total bacterial protein. This 42 kD protein was
designated E2N730(m). After sonication, E2N730(m) was almost exclusively found
in the insoluble fraction [Fig. 2(A), compare lane 3 and 4].

Insoluble E2N730(m) was solubilized with high
concentration of strong chaotropic agent (6 mol/L Gu•HCl) in the presence of
high concentration of reducing agent (100 mmol/L â-ME) and
purified under denaturing conditions on Ni2+-NTA agarose. Purified
E2N730(m) appeared as a fairly homogenous band of 42 kD on SDS-PAGE, with minor
amounts of lower molecular weight bands, most likely non-full-length products
as a result of premature translational termination or protease degradation
[Fig. 2 (A), lane 5]. The 42 kD full-length E2N730(m) constituted over 85% of
purified proteins and its final yield was higher than 1 mg/L initial E. coli
culture.

Fig. 2 Expression and purification of E2N730(m) fusion protein
and Western blot analysis using homologous patient serum

(A) SDS/PAGE stained with Coomassie blue (12% gel). 1, 2, whole-cell
lysates of induced TG-1(pQE8) and TG-1[pQE8/E2N730(m)] respectively; 3, soluble
fraction after sonication of induced TG-1[pQE8/E2N730(m)]; 4, insoluble
fraction after sonication of induced TG-1[pQE8/E2N730(m)]; 5, purified E2N730
(m). (B) Western blot using homologous HCV patient serum S94 as first antibody (12%
gel). 1, mock purification products from induced TG-1(pQE8); 2, purified
E2N730(m).

Antigenicity and immunogenicity analysis of E2N730(m)

Reactivity of E2N730(m) against human sera in Western
blot was analyzed. Homologous hepatitis C patient serum S94 specifically
recognized E2N730(m) [Fig. 2(B)], whereas sera from healthy blood donor or
hepatitis B patient showed no specific recognition (data not shown). This
result demonstrated that, despite the presence of fourresidue mutation, E.
coli-derived E2N730(m) still displayed HCV E2-specific antigenicity. Also,
it was clear that at least some of the anti-E2 antibodies present in infected
patients’ sera were directed towards E2 polypeptide backbone, suggesting the
possibility of using bacterially expressed E2N730(m) for the clinical detection
of anti-E2 antibodies in human sera.

A rabbit subcutaneously immunized with E2N730(m) three
times produced anti-E2 antibodies with a titer of 1∶32 000. The animal was sacrificed and the obtained sera
was designated RE2N730(m) which was used to detect E2 proteins expressed in E.
coli and mammalian cells in Western blot. RE2N730(m) not only specifically
recognized E2N730(m) [Fig. 3(A)], but also showed specific recognition of
glycosylated E2 proteins expressed in recombinant vaccinia virus system [Fig.
3(B)]. In addition to the major 42 kD full-length band, RE2N730(m) also reacted
with heterogeneous bands of higher and lower mobility rates in purified
E2N730(m), although with much lower intensity [Fig. 3(A)]. Some of these bands
were also faintly observable in purified E2N730N(m) after staining or blotting
with homologous serum S94 [Fig. 2 (A), lane 5 and Fig. 2(B), lane 2]. Since
these minor bands were not observed in pQE8 transformed E. coli cells
[Fig. 3(A), lane 1], they are most likely non-full-length and polymeric forms
of E2N730(m). In reacting with RE2N730(m), glycosylated E2 appeared as two heterogeneous
species with apparent molecular weights of approximately 50 and 70 kD,
respectively [Fig. 3(B), lane2]. The difference in mobility rate of these two
species of E2 glycoprotein most likely reflected differences in the degree and
type of glycosylation. The same recognition pattern of E2 glycoproteins was
also observed with rabbit sera raised against aa 450–565 fragment of E2
expressed in E. coli [20].

Fig. 3 Western blot detection of E2 protein expressed in
various systems using RE2N730(m)

(A)     
E2 expressed in E. coli (15% gel). 1, 2, whole-cell lysates of
induced TG-1 (pQE8) and TG-1[pQE8/E2N730(m)] respectively; 3, purified
E2N730(m). (B) E2 (aa 384–730) expressed in recombinant vaccinia virus system
(15% gel). 1, HeLa cells infected with vvT7 alone; 2, HeLa cells co-infected
with vvT7 and vCEH-2. gE2 stands for glycosylated E2 proteins.

(B)  

This result indicated that E2N730(m) was able to present
E2-specific immunogenicity in vivo and elicit antibodies directed
against epitopes shared by glycosylated and unglycosylated E2 proteins. The
fact that RE2-116R was reactive against E2 proteins expressed in various
prokaryotic and eukaryotic systems regardless of glycosylation status suggested
that it could be used to detect different types of E2 proteins in biochemical,
virological and vaccinological studies of E2.

Table 1 Detection of anti-E2 antibodies in human sera
using E2N730(m) as capture antigen

Anti-E2 antibodies in human sera were detected using E. coli-derived
E2N730(m) as coating antigen. Anti-HCV antibodies were detected using UBI HCV
EIA 4.0, which does not incorporate any E2 components.

Application of E2N730(m) in the detection of anti-E2
antibodies in human sera

Some reports have suggested that including anti-E2
antibody testing into current EIA kits could improve their performance [14–17].
Our results shown here and elsewhere [19,20] demonstrated that E. coli-derived
E2 fragments reacted specifically with infected patient sera in Western blot
and elicited antibodies in rabbits reactive against glycosylated E2 expressed
in mammalian cells, indicating E2 proteins expressed in bacteria shared
antigenic/ immunogenic epitopes with mammalian E2 glycoproteins. Previously, we
developed an anti-E2 EIA using aa 385– 565 fragment of E2 expressed in E.
coli to detect anti-E2 antibodies in human sera and obtained a positivity
rate of 40% in Chinese anti-HCV EIA-positive patients [20]. Since E2N730(m)
almost covered the whole length of E2 ectodomain and the upstream half of TMD,
and the fourresidue mutation at aa 568–571 did not affect its reactivity with
homologous patient‘s sera [Fig.
2(B)], we further tested its ability to detect anti-E2 antibodies in human sera
in EIA.

Sera from 100 healthy blood donors, 20 hepatitis B
patients, 20 non-A-to-E hepatitis patients and 158 hepatitis C patients were
reacted with E2N730(m) in EIA under optimized conditions. All hepatitis
patients presented corresponding serological as well as pathological symptoms.
The results are summarized in Table 1. Readings from 100 healthy blood donor
samples were used to calculate the cut-off value for positivity and three out
of the one hundred samples (3%) gave positive readings. In order to rule out
the false-positive reactions caused by E. coli host protein
contaminations in E2N730(m) preparation causing false-positive reactions, these
three samples were confirmed using purified E2N730(m) (Fig. 4). Two samples
reacted specifically with the 42 kD E2N730(m) band, although the intensity of
the band was lower than that obtained with hepatitis C patient sera (Fig. 4,
compare lane 2 and 4 with lane 1). The third sample showed negative reaction
with E2N730(m) (Fig. 4, lane 3). The two donors positive for anti-E2 in both
EIA and Western blot might have been exposed to HCV. Unfortunately, because of
limited sample availability and difficulty in tracing the individual donors, we
were unable to ascertain the nature of this anti-E2 response in these subjects.

None of the hepatitis B and non-A-to-E hepatitis
patients’ sera reacted with E2N730(m) (Table 1).  This result and the result with healthy blood donors
demonstrated that E2N730(m) displayed high specificity and sensitivity in EIA.

Eighty-nine samples out of 158 hepatitis C patient sera
(56%) were positive for anti-E2N730(m). This positivity rate is higher than
that the 40% we obtained with E. coli-derived aa 385–565 fragment of E2
which lacked the C-terminal half of E2N730(m) [20]. Some reports in the
literature have also used E2 proteins expressed in bacteria to detect anti-E2
antibodies in hepatitis C patient sera, either in EIA or in Western blot
[22–25]. The reported positivity rate for anti-E2 varies from study to study
between 17% and 73%. Conceivably, fragment selection, antigen formulation, assay
format, and patient selection all played a role in creating such discrepancies.
Our EIA employed an E2 antigen encompassing the largest amount of E2 amino acid
residues reported so far for E. coli expression, and also used the
largest number of patient

Fig. 4 Confirmatory Western blot analysis of anti-E2
antibodies in healthy blood donor sera

Purified E2N730(m) was run on 12% SDS/PAGE gel, transferred to nitrocellulose
membrane, and blotted against human sera as described in Materials and Methods.
1, hepatitis C patient sera positive for anti-E2N730(m) in EIA as positive
control; 2–4, three healthy blood donors sera positive for anti-E2N730 (m) in
EIA; 5, healthy blood donor sera negative for anti-E2N730(m) in EIA as negative
control. Solid and open arrows indicate positive and negative signals,
respectively.

samples reported for such an assay. It is our opinion
that the 56% positivity rate obtained with E2N730(m) largely reflected the
prevalence of glycosylation- and conformation- independent anti-E2 antibodies
in Chinese hepatitis C patient sera.

Theoretically, E2 glycoproteins purified from mammalian
cells would be ideal for the detection of anti-E2 antibodies in human sera,
because they could best mimic the immunological properties of natural E2
glycoproteins on HCV virions. However, high-level and large-scale expression/
purification of E2 in mammalian cells is difficult and very expensive. It is
worth comparing E. coli-derived E2N730(m) and mammalian cell-expressed
E2 glycoproteins in EIA using the same patient group and under the same
conditions, to see whether the difference in positivity rate would be small
enough to justify using E2N730 (m) as an inexpensive alternative to E2
glycoproteins. We are now probing such a possibility.

Using mammalian cell or insect cell-derived E2
glycoproteins, some researchers have also linked the presence of anti-E2
antibodies in chronic hepatitis C patients and interferon therapy recipients
with on-going HCV replication indicated by RT-PCR positivity[26–28]. Therefore,
anti-E2 testing has been suggested to be an useful indicator for clinical
diagnosis of HCV infection and monitoring the outcome of interferon therapy. One
hundred and forty- three out of the 158 hepatitis C patient sera were subjected
to anti-E2N730(m) and HCV RNA detection to find out whether an association
between anti-E2 positivity and HCV RNA positivity could also be established
using this E. coli-derived E2 protein. The results are summarized in
Table 2. In a total of 143 samples, 86 were HCV RNA positive, of which 64 (74%)
were also anti-E2 positive. In the 57 HCV RNA negative samples, however, only
24 (42%) were anti-E2 positive. The difference was statistically significant (c2 test, P<0.05), indicating an association between the presence of anti-E2N730(m) antibodies and HCV viremia. In other words, patients positive for anti-E2N730(m) have a higher likelihood of being actively infected by HCV. This result suggested that anti-E2 EIA based on E2N730(m) might be of clinical value for both blood donor screening and hepatitis C diagnosis/ therapy monitoring.

In summary, despite a four-residue mutation and no
glycosylation or three-dimensional conformation, E2N730(m) expressed in E.
coli displayed E2-specific antigenicity and immunogenicity shared by
mammalian E2 glycoproteins. After further evaluation and optimization,
E2N730(m) could be used for preliminary clinical detection of anti-E2
antibodies in human sera, as a substitute for expensive mam-malian cell-derived
E2 glycoproteins. In addition, E2N730 (m) would be useful for the detection of
anti-E2 immune response in post-immune animals in E2-bassed HCV vaccine
research. On the other hand, rabbit antisera against E2N730(m) (RE2-116R) could be
used to detect E2 proteins expressed in different systems, and possibly, to
detect E2 antigen in liver biopsy samples. Results presented in this report
provide a possible inexpensive alternative to the cumbersome and costly route
of expressing E2 glycoproteins in mammalian cells, and might result in the
development of new diagnostic, therapeutic and prophylactic measures against
HCV.

Table
2 Association between anti-E2 antibodies in hepatitisC patient sera and viremia

References

1 Choo QL, Kuo G, Weiner AJ, Overby LR, Bradley DW, Houghton M. Isolation of
a cDNA clone derived from a blood-borne non-A, non-B viral hepatitis genome.
Science, 1989, 244: 359–362

2 Global surveillance and control of hepatitis C. Report of a WHO
Consultation organized in collaboration with the Viral Hepatitis Prevention
board, Antwerp, Belgium. J Viral Hepat, 1999, 6: 35–47

3 Seeff LB. Natural history of hepatitis C. Am J Med, 1999, 107: 10S–15S

4 Murphy FA, Fauquet CM, Bishop DHL, Ghabrial SA, Jarvis AW, Martelli GP,
Mayo MA, Summers MD eds. Virus Taxonomy: The Sixth Report of the International
Committee on Taxonomy of Viruses. Vienna and New York: Springer-Verlag, 1995,
424–426

5 Grakoui A, Wychowski C, Lin C, Feinstone SM, Rice CM. Expression and
identification of hepatitis C virus polyprotein cleavage products. J Virol, 1993,
67: 1385–1395

6 Selby MJ, Choo QL, Berger K, Kuo G, Glazer E, Eckart M, Lee C et al.
Expression, identification and subcellular localization of the proteins encoded
by the hepatitis C viral genome. J Gen Virol, 1993, 74: 1103–1113

7 Mizushima H, Hijikata M, Asabe S, Hirota M, Kimura K, Shimotohno K. Two
hepatitis C virus glycoprotein E2 products with different C termini. J Virol,
1994, 68: 6215–6222

8 Pileri P, Uematsu Y, Campagnoli S, Galli G, Falugi F, Petracca R, Weiner
AJ et al. Binding of hepatitis C virus to CD81. Science, 1998, 282:
938–941

9 Wunschmann S, Medh JD, Klinzmann D, Schmidt WN, Stapleton JT.
Characterization of hepatitis C virus (HCV) and HCV E2 interactions with CD81
and the low-density lipoprotein receptor. J Virol, 2000, 74: 10055–10062

10 Scarselli E, Ansuini H, Cerino R, Roccasecca RM, Acali S, Filocamo G,
Traboni C et al. The human scavenger receptor class B type I is a novel
candidate receptor for the hepatitis C virus. EMBO J, 2002, 21: 5017–5025

11 Choo QL, Kuo G, Ralston R, Weiner AJ, Chien D, Van Nest G, Han J et
al. Vaccination of chimpanzees against infection by the hepatitis C virus.
Proc Natl Acad Sci USA, 1994, 91: 1294–1298

12 Kobayashi M, Tanaka E, Matsumoto A, Ichijo T, Kiyosawa K. Antibody response
to E2/NS1 hepatitis C virus protein in patients with acute hepatitis C. J
Gastroenterol Hepatol, 1997, 12: 73–76

13 Ishii K, Rosa D, Watanabe Y, Katayama T, Harada H, Wyatt C, Kiyosawa K et
al. High titers of antibodies inhibiting the binding of envelope to human
cells correlate with natural resolution of chronic hepatitis C. Hepatology,
1998, 28: 1117–1120

14 Lee DS, Lesniewski RR, Sung YC, Min WK, Park SG, Lee KH, Kim HS.
Significance of anti-E2 in the diagnosis of HCV infection in patients on maintenance
hemodialysis: Anti-E2 is frequently detected among anti-HCV antibody-negative
patients.J Am Soc Nephrol, 1996, 7: 2409–2413

15 Leon P, Lopez JA, Elola C, Quan S, Echevarria JM. Typing of hepatitis C
virus antibody with specific peptides in seropositive blood donors and
comparison with genotyping of viral RNA. Vox Sang, 1997, 72: 71–75

16 Psichogiou M, Katsoulidou A, Vaindirli E, Francis B, Lee SR, Hatzakis
A. Immunologic events during the incubation period of hepatitis C virus
infection: The role of antibodies to E2 glycoprotein. Multicentre Hemodialysis
Cohort Study on Viral Hepatitis. Transfusion, 1997, 37: 858–862

17 da-Silva-Cardoso M, Sturm D, Koerner K, Dengler T, Kerowgan M, Kubanek
B. Anti-HCV envelope prevalence in blood donors from Baden-Wurttemberg. Ann
Hematol, 1997, 74: 135–137

18 Wang Y, Okamoto H, Tsuda F, Nagayama R, Tao QM, Mishiro S. Prevalence,
genotypes, and an isolate (HC-C2) of hepatitis C virus in Chinese patients with
liver disease. J Med Virol, 1993, 40: 254–260

19 Liu J, Kong Y, Zhu L, Wang Y, Li G. High-level expression of the
C-terminal hydrophobic region of HCV E2 protein ectodomain in E. coli.
Virus Genes, 2002, 25: 5–13

20 Liu J, Zhu L, Zhang X, Lu M, Kong Y, Wang Y, Li G. Expression,
purification, immunological characterization and application of Escherichia
coli-derived hepatitis C virus E2 proteins. Biotechnol Appl Biochem, 2001,
34: 109–119

21 Li Y, Li G, Kong Y, Wang Y, Wang Y, Wen Y. Expression of structural proteins
of hepatitis C virus (HCV) in mammalian cells. Sci China Ser C, 1998, 41:47–55

22 Mita E, Hayashi N, Ueda K, Kasahara A, Fusamoto H, Takamizawa A,
Matsubara K et al. Expression of MBP-HCV NS1/E2 fusion protein in E.
coli and detection of anti-NS1/E2 antibody in type C chronic liver
disease.Biochem Biophy Res Commun, 1992, 183: 925–930

23 Fournillier-Jacob A, Lunel F, Cahour A, Cresta P, Frangeul L, Perrin M,
Girard M et al. Antibody responses to hepatitis C envelope proteins in
patients with acute or chronic hepatitis C. J Med Virol, 1996, 50: 159–167

24 Hüssy P, Faust H, Wagner J, Schmid G, Mous J, Jacobsen H. Evaluation of
hepatitis C virus envelope proteins expressed in E. coli and insect
cells for use as tools for antibody screening. J Hepatol, 1997, 26: 1179–1186

25 Dueñas-Carrera S, Viña A, Garay HE, Reyes O, Alvarez-Lajonchere L,
Guerra I, Gonzãlez LJ et al. Immunological evaluation of Escherichia
coliderived hepatitis C virus second envelope protein (E2) variants. J Pept
Res, 2001, 58: 221–228

26 Lesniewski R, Okasinski G, Carrick R, Van Sant C, Desai S, Johnson R,
Scheffel J et al. Antibody to hepatitis C virus second envelope (HCV-E2)
glycoprotein: A new marker of HCV infection closely associated with viremia. J
Med Virol, 1995, 45: 415–422

27 Fong TL, Lee SR, Briggs WK, Valinluck B, Govindarajan S, Hoffman A,
Jaczko B et al. Clinical significance of hepatitis C viral RNA status
and its correlation to antibodies to structural HCV antigens in anti-HCV
reactive patients with normal liver tests. J Med Virol, 1996, 49: 253–258

28 Cerino A, Bissolati M, Cividini A, Nicosia A, Esumi M, Hayashi N,
Mizuno K et al. Antibody responses to the hepatitis C virus E2 protein:
Relationship to viraemia and prevalence in anti-HCV seronegative subjects. J
Med Virol, 1997, 51: 1–5

Received: September 2, 2003 Accepted: October 15, 2003

This work was supported by a grant from the National High Technology
Research and Development Program of China (863 Program) (No. 2001AA215171)

# Equal
contributors

* Corresponding author: Tel, 86-21-54921103; Fax, 86-21-54921011; E-mail, [email protected]