Journal of Hepatology 36 (2002) 254–262

www.elsevier.com/locate/jhep

Hepatitis C virus (HCV) E1 and E2 protein regions that

specifically bind to HepG2 cells
Javier Eduardo Garcia*, Alvaro Puentes, Jorge Suárez, Ramses López, Ricardo Vera,
Luis Eduardo Rodrı́guez, Marisol Ocampo, Hernando Curtidor, Fanny Guzmán,
Mauricio Urquiza, Manuel Elkin Patarroyo
Fundación Instituto de Inmunologia de Colombia, Universidad Nacional de Colombia, Avda. Calle 26 No. 51-60, Bogota, Colombia

Background/Aims: Identify hepatitis C virus (HCV) sequences in E1 and E2 protein binding to HepG2.
Methods: Synthetic 20-mer long, ten-residue overlapped peptides, from E1 and E2 proteins, were tested in HepG2 or
Raji cell-binding assays. Affinity constants, binding site number per cell and Hill coefficients were determined by
saturation assay for high activity binding peptides (HABPs). Receptors for HepG2 cell were determined by cross-
linking and sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis.
Results: Twelve HABPs were found in HCV genotype 1a, allowing six hepatocyte-binding sequences (HBSs) to be
defined: two peptide-binding regions in E1 HABPs 4913 (YQVRNSTGLYHVTNDCPNSS) and 4918 (MTPTVAT-
RDGKLPATQLRRHY). Four hepatocyte-binding regions were defined in E2: region-I, peptide 4931 (ETHVTGG-
SAGHTVSGFVSLLY); region-II, 4937–4939 (HHKFNSSGCPERLASCRPLTDFDQGWGPISYANGSGPDQR); reg-
ion-III, 4943–4945 (PVYCFTPSPVVVGTTDRSGAPTYSWGENDTDVFVLNNTR) and region-IV, 4949–4952 (CGA-
PPCVIGGAGNNTLHCPTDCFRKHPDATYSRCGSGPWITPRCLVDYPY). The underlined sequences are most rele-
vant in the binding process. HABPs 4913 and 4938 also bind to CD81 positive Raji cells. Region-II 4938 HABPs bind to
50 and 60 kDa HepG2 cell membrane surface proteins.
Conclusions: Six HVRs to the HepG2 were identified. Some HABPs have been previously found to be antigenic and
immunogenic. HABPs, 4918 (from E1), 4938, 4949, 4950, 4951 and 4952 (from E2) have not been previously recognised.
These HABPs could be relevant to HCV invasion of hepatocytes.
q 2002 Published by Elsevier Science B.V. on behalf of the European Association for the Study of the Liver.
Keywords: Hepatitis C virus; Envelope proteins; Hypervariable region-1; HepG2 binding

1. Introduction HCV mainly infects hepatocytes but can invade human

The hepatitis C virus (HCV) is considered to be the major
non-A and non-B hepatitis parental transmission agent in
the world; it has been associated with chronic liver disease,
development of cirrhosis and liver cancer. The HCV
genome encodes a 3011 amino acid polyprotein [1–3].
This polyprotein is processed in four structural proteins
[4]: (i) the core protein (22 kDa); (ii) the E1 and E2 envel-
ope proteins (33 and 70 kDa, respectively) [5–6] and (iii) P7
(7 kDa) [7].
Received 29 December 2000; received in revised form 13 August 2001;
accepted 22 October 2001
* Corresponding author. Tel.: 157-1-2207700, ext. 494; fax: 157-1-280-
3999.
E-mail address: javgar22@hotmail.com (J.E. Garcia).
B-cells. HCV binding to hepatic cells seems to be mediated
by E1/E2 virus proteins; these envelope recombinant
proteins can bind to hepatocarcinoma cell lines as well as
human B-cells [8,9]. Recombinant E2 protein binding to
human B-cells seems to be mediated by the CD81 extracel-
lular domain [11,10]; there are reports that some hepatoma
cell lines can bind to HCV envelope in the absence of CD81
[9,12]. E1/E2 vaccinated chimpanzees were protected from
homologous challenge with HCV [13]. It has been
suggested that the hypervariable region-1 (HVR-1) could
be considered to be a HCV-vaccine candidate since antibo-
dies against this region (defining linear B-cell epitopes)
were able to prevent HCV binding to target cells [14–17].
Vaccination with a HVR-1 peptide is also effective [18–19].
However, the HVR-1 shows sequence variability which

0168-8278/02/$20.00 q 2002 Published by Elsevier Science B.V. on behalf of the European Association for the Study of the Liver.
PII: S01 68- 8278(01)0026 2-8
 J.E. Garcia et al. / Journal of Hepatology 36 (2002) 254–262
might be due to immune response to HCV-infection [18,19].
However, the existence of immunogenic regions different
from HVR-1 has been suggested because B-epitopes have
been located in other regions [20,21].
Taking this data into account, there was great interest in
looking for E1/E2 protein hepatocyte-binding sequences
(HBSs). It was found that 12, 20-mer long, ten-overlapping
amino acid peptides specifically bind to HepG2 cells in
these proteins. These peptides are located in conserved,
variable and hypervariable genotype 1a sequences. The
hepatocyte-binding sequence role is discussed.
2. Materials and methods
2.1. Cell culture
HepG2 cells [22,23] were kept in RPMI-1640 (supplemented with 10%
foetal bovine serum (FBS), penicillin 100 IU/ml, streptomycin 100 mg/ml,
amphotericin B 0.25 mg/ml, vitamins ICN) and non-essential amino acids
(Gibco). The cells were grown as monolayers in flasks coated with 2%
collagen at 378C in 5% CO2. The cells were harvested by phosphate
buffered saline–ethylene diamine tetraacetic acid (PBS–EDTA) addition
and centrifugation. The cells were washed with PBS and counted in
Newbawer chamber.
2.2. Peptides
The peptides were synthesised by solid phase methodology [24], purified
by high pressure liquid chromatography (HPLC) and characterised by
matrix assisted laser desorption /ionisation-time of flight (Maldi-Tof)
mass spectrometry. The peptides were 20 amino acid long, overlapping
in ten residues; these peptides cover the entire E1 and E2 HCV protein
sequence. For those peptides with no Tyr-residue in their sequences, one
Tyr-residue was added at the N-terminal to allow 125I-radio-labelling
[25,26].
2.3. Radio-labelling
Purified peptides (2 nmol) were radio-labelled with 5 ml Na 125I
(100 mCi/ml, ICN) and 0.3 mmol chloramine-T in a 25 ml volume final
for 15 min [25]. The reaction was stopped with 0.3 mmol sodium metabi-
sulphite [27]. The radio-labelled peptides were passed through a Sephadex
G-10 column; specific activity was between 40 and 50 mCi/nmol.
2.4. Binding assays
The binding assay was standardised using the hepatitis B 27-mer 4980
(PLGFFPDHQLDPAFGANSNNPDWDFNP) peptide which has been
reported as specifically binding to HepG2 cells [28]. HepG2 cells
(1.5 £ 10 6 cells/ml) were incubated with different radio-labelled peptide
concentrations (0, 15, 30, 50 and 90 nM), in the presence or absence of
unlabelled peptide (40 mM) for 2 h at 48C [29]. An aliquot of this reaction
mixture was passed through a 60:40 dioctylftalate–dibutylftalate cushion
(1.015 g/ml), spinning at 13,200 rpm for 1.5 min. Cell-associated radioac-
tivity was quantified. The binding assays were performed in triplicate (Figs.
1 and 2).
2.5. Saturation assays
HepG2 cells (1 £ 10 6 cells/ml) were incubated with radio-labelled high
activity binding peptides (HABPs) at concentrations between 10 and
1500 nM, in the presence of absence of unlabelled peptide (80 mM). The
255
curves obtained were analysed by Scatchard analysis and the affinity
constants were determined by the Hill equation [25].
2.6. Cross-competition assays
The cross-competition assay was performed between E2 HABPs and the
other HABPs. Two 125I-labelled HABPs (0–90 nM) were incubated with
1.5 £ 10 6 cells/ml, for 2 h at 48C, in the presence of the same or other
unlabelled HABPs (0–80 mM). Cells were passed through a 60:40 dioctylf-
talate–dibutylftalate cushion (see above); cell bound peptide was quanti-
fied.
2.7. Analogue competition binding assay
HABP binding to HepG2 cells was inhibited by the presence of glycine-
scanning analogue peptides at different concentrations [7]. 125I-labelled
HABPs (300–400 nM concentrations) were incubated with 2 £ 10 6
HepG2 cells in the presence or absence of 4 and 40 mM unlabelled original
or analogue peptide in 80 ml final volume. The assay was incubated for
90 min at 48C; the cell bound 125I-labelled peptide was quantified.
2.8. Obtaining HepG2 cell membranes
Cells (2.5 £ 10 6) were suspended in 6 ml buffer A (0.25 M sucrose,
5 mM Tris–HCl pH 8.0, 1 mM MgCl2), passed ten times through 0.27
needle syringe and centrifuged at 250 £ g for 5 min at 48C; the supernatant
was skimmed off and stored at 48C. The pellet was suspended in buffer A
and the process repeated twice more. The supernatant was centrifuged at
1500 £ g for 10 min, 48C, pooled and centrifuged for 30 min at 13,200 rpm,
48C. The pellet suspended in buffer B (1.42 M sucrose, 5 mM Tris–HCl pH
8.0, 1 mM MgCl2) was poured into the bottom of the tube; then 2 ml buffer
B and 1 ml buffer A were added on top. The sucrose gradient was centri-
fuged at 80,000 £ g, 60 min at 48C. The ring observed in the interface was
suspended in PBS and centrifuged for 60 min at 13,200 rpm. Plasmatic
membranes were collected from the pellet, which was suspended in 1 ml
PBS and stored at 2208C [30].
2.9. Binding assay and sodium dodecyl sulfate-
polyacrylamide gel electrophoresis (SDS-PAGE)
HepG2 cell membranes, 50 ml, were incubated with 200 ml radio-
labelled peptide (400 nM) for 2 h at 48C. The suspension was centrifuged
at 13,200 rpm, 5 min; the pellet was PBS washed. Membrane bound peptide
was cross-linked with 20 mM Bis (sulphosuccinimidyl suberate-Bs3,
Pierce), for 4 h at 48C and the reaction was stopped with 0.1 M Tris–
HCl, pH 6.8. The pelleted membranes were PBS washed, centrifuged and
suspended in Laemmli buffer, heated at 908C, 2 min. Membrane proteins
were separated in 12% SDS-PAGE, exposed for 24 h at 2708C and auto-
radiographically developed on Kodak film [31].
3. Results
In binding assays to HepG2, the radio-labelled peptides
(0–90 nM) were incubated with HepG2 cells in the presence
(non-specific binding) or absence (total binding) of the same
non-radio-labelled peptide. Total binding minus non-speci-
fic binding gave the specific binding [29].
Specific binding activity to HepG2 cells is calculated
from specific binding curve slope; peptides showing specific
binding activity equal to or higher than control peptide were
named HABPs to HepG2 cells. Three types of HepG2-bind-
ing behaviour were found for these peptides (Fig. 1): A and
B show HABP-4913 behaviour (YQVRNSTGLYHVTN-
256 J.E. Garcia et al. / Journal of Hepatology 36 (2002) 254–262

Fig. 1. Peptide-binding behaviour to HepG2 cells. Total binding (–– £ –– £ ––) and non-specific binding to HepG2 cells (––D––D––) for peptides 4913,
4947 and 4964 at different 125I-labelled peptide concentrations is shown. HepG2 cells incubated with 125I-peptide in the presence of 400 excess
unlabelled peptide (A,C,E). The right-hand panels (B,D,F) show specific binding, resulting from subtracting total binding from non-specific binding.
Peptides 4913, 4947 and 4964 show high binding, non-specific binding and non-binding activity to HepG2 cells, respectively.

DCPNSS). C and D show the 4947 non-specific binding
peptide; this binds to HepG2 cells but there is non-inhibition
with the same non-radio-labelled peptide (D). E and F
peptides did not show binding to HepG2 cells.
Two HABPs were found in E1 protein: 4913 (Fig. 2)
located in the N-terminal region and peptide 4918
(MTPTVATRDGKLPATQLRRHY) located in the central
region. Neighbouring peptides (4914, 4917 and 4919) did
not show binding activity, indicating that residues in peptide
4913 and the entire peptide 4918 sequence are essential for
hepatocyte-binding activity.
The E2 protein showed ten HABPs. The 4931
(ETHVTGGSAGHTVSGFVSLLY) is located in the HVR;
the other nine HABPs are clustered in three regions. This
allowed us to define three HBSs.
HBS-II contains HABPs 4937–4939 ( 444HHKFNSSGC-
PERLASCRPLTDFDQGWGPISYANGSGPDQR 484). The
critical binding peptide is 4938 (underlined) in this region,
since the N-terminal sequence from 4937 (included in 4936)
and the C-terminal portion of 4939 (present in 4940) do not
bind to HepG2.
The underlined sequence is critical for binding in HBS-
III, formed by HABPs 4943–4944 ( 504PVYCFTPSP-
VVVGTTDRSGAPTYSWGEND 533), since peptide 4943
N-terminus (included in peptide 4942) and peptide 4944
C-terminus (included in peptide 4945) show no or little
binding activity.
In HBS-IV, formed by HABPs 4949–4952 ( 564CGAPPC-
VIGGAGNNTLHCPTDCFRKHPDATYSRCGSGPWITP-
RCLVDYPY 613), a similar analysis to that shown above
allows us to conclude that the underlined binding sequence
is longer, comprising 30 amino acids, involving HABPs
4950 and 4951.
According to Gene Data Bank [32–40], HABPs 4943
(PVYCFTPSPVVVGTTDRSGA) and 4952 (YSRCGSGP-
WITPRCLVDYPY) have genotype 1a conserved sequ-
ences. HBS-I is the most variable and HBS-III the most
conserved.
 J.E. Garcia et al. / Journal of Hepatology 36 (2002) 254–262 257

HABPs 4931, 4938, 4943, 4950 and 4952 affinity
constants were in the 350–600 nM range and there were
6 £ 10 5 and 3 £ 10 6 binding sites per HepG2 cell (Table 1).
The HABP 4931 had the highest affinity and highest binding
capacity; HABP 4938 had the lowest affinity constant, show-
ing high binding capacity (Fig. 3).
The cross-competition assays showed that 125I-labelled
4944 or 125I-labelled 4951 peptide binding to HepG2 cells
was inhibited by 4913, 4938, 4944 and 4951 non-labelled
peptides. The Plasmodium falciparum- (Pf )-MSP-1 1585
peptide (high affinity binding to erythrocytes [25]) did not
inhibit either 4944 or 4951 binding to HepG2 cells (Fig. 4).
HepG2 cell binding critical residues were defined by
competition binding assays. Critical residues were those
which, upon replacement with glycine, were not able to
inhibit original radio-labelled peptide binding by at least
50% (4 and 40 mM unlabelled peptide: 40 mM is shown in
Fig. 5). The critical residues (in bold and underlined) were
as follows: 4944 (VVVTTDRSAPTYSWEND), 4949 (CA-
PPCVIANNTLHCPT), 4950 (ANNTLHCPTDCFRKHP-
DAT), 4951 (DCFRKHPDATYSRCSPWI) and 4952 (Y-
SRCSPWITPRCLVDYPY); they were 50% hydrophobic
and 50% hydrophilic.
The HABPs bind to hepatocyte membrane surface mole-
cules having similar apparent molecular weight (43 and
60 kDa, Fig. 6), suggesting that the same molecule could
be the receptor for these HABPs.
Four HABPs were tested in binding assay to Raji B-
lymphoblastoid cells, which are CD81 positive. HABP
4931 did not bind to these cells.

4. Discussion

Recognition and interaction between virus and host cell is

mediated by HCV-envelope glycoprotein [8,10–12].
Neutralising antibody targets (preventing HCV-infection
in animal model) are the envelope proteins [41,42]; further-
more, these envelope protein recombinant fragments, or
peptide fragments, show protection in chimpanzees chal-
lenged with homologous HCV [13,14,41].
HepG2 cell HABP affinity constants were in the nanomo-
lar range, indicating that these sequences’ interaction with
HepG2 cells is sufficiently strong to be considered important
during hepatocyte–virus invasion. Furthermore, the number
of binding sites per hepatocyte suggests that they are suffi-
cient to be implicated in the hepatocyte–virus invasion. The
Hill coefficients (suggesting positive co-operation) support
the above hypothesis.
HABP binding cell selectivity is supported by the fact

Table 1
Saturation assay a

Peptide Sequence Kd (nM) Sites binding

per cell

4931 ETHVTGGSAGHTVSGFVSLL 350 3,000,000

4952 YSRCGSGPWITPRCLVDYPY 400 700,000
4950 AGNNTLHCPTDCFRKHPDAT 500 600,000
Fig. 2. Peptide specific binding activity to HepG2 cells. The sequence of 4938 ERLASCRPLTDFDQGWGPI 600 1,200,000
synthesised peptides (20 amino acids long and ten overlapping residues) 4943 PVYCFTPSPVVVGTTDRSGA 500 600,000
for E1 (A) and E2 (B) proteins: protein and peptide numbers are a
shown. The bars represent the binding activity. The 4980 peptide HepG2 cells were incubated with different radio-labelled peptide
(reported by Neurath et al.) is used as control. The residues within concentrations in the presence or absence of non-labelled peptide; incu-
the grey box are variable, comparing the different sequences belonging bated and non-bound peptide was eliminated (see Section 2). These
to genotype 1a (Gene Bank, Ref. [32–40]). The residues in the black box peptides’ binding is saturable and specific, having affinity constants of
indicate E2 protein HVR. The peptides with greater or equal peptide between 350 and 500 nM and the maximum binding sites between
control were considered to be HABPs. 6 £ 10 5 and 3 £ 10 6.
258 J.E. Garcia et al. / Journal of Hepatology 36 (2002) 254–262
Fig. 3. Saturation assays. The saturation curves result from graphic
concentration of free 125I-peptide vs. specifically bound 125I-peptide.
Peptide 4931 (A), 4938 (B), 4943 (C), 4950 (D) and 4952 (E) curves
can be observed. The affinity constants, as well as the maximum
number of sites per cell, are obtained from these curves. The 4931
peptide has affinity and number of sites per cell higher than peptides
4938, 4943, 4950 and 4952. The binding peptide is saturable and the
affinity constants are nanomolar.
that these peptides did not bind to erythrocytes (data not
shown); furthermore, some of these HABPs did not bind
to RDBL (Raji cells) (Table 2), but they bound to other
non-completely characterised hepatoma cells (HUH-1 like).
It has been suggested that HCV receptor is the CD81
molecule [24]. However, E2 protein binds to non-expressing
CD81 HepG2 cells [9,12]. There was thus great interest in
determining those envelope protein regions which specifi-
cally bind to human HepG2 cells (non-expressing CD81).
The 4913-HABP (YQVRNSTGLYHVTNDCPNSS),
located in E1 protein N-terminal portion, is semi-conserved
for genotype 1a (Fig. 2A). An immunogenic peptide
YEVRNVSGIYHVTNDCSNSS has been reported which
is highly homologous to HABP 4913 (underlined
sequences) [43]. HABP 4931 (E2 protein and underlined)
belongs to HVR-I (ETHVTGGSAGHTVSGFVSLLAP-
GAKQN) which is probably expressed on the outside and
is sterically free on the HCV envelope [15]. It has been
suggested that E1 and E2 N-terminal regions (where 4913
and 4931-HABPs are located) are exposed on the protein
surface [15]. HVR-1 antibodies are implicated in viral
neutralisation [15,16,44], as well as homologous protection,
in chimpanzees and rabbits [13,14,41].
Fig. 4. Cross- competition assays. (A) 125I-peptide 4944 HepG2 cell
binding inhibited by the presence of non-radio-labelled peptides
4913, 4938 and 4951 in concentrations between 0 and 80 mM. (B) Bind-
ing of radio-labelled peptide 4951 to HepG2 is inhibited by the presence
of peptides 4913, 4938 and 4944. It can be observed that peptide 4913
(belonging to protein E1) inhibits binding to a greater extent than non-
radio-labelled peptides 4944 (A) and 4951 (B). Peptides 4944 and 4951
are mutually inhibiting. These results indicate that these peptides possi-
bly bind to the same receptor. P. falciparum MSP-1 protein 1585
peptide (high affinity binding to erythrocytes) does not inhibit binding
to peptides 4944 and 4951.
 J.E. Garcia et al. / Journal of Hepatology 36 (2002) 254–262
Fig. 5. Analogue competition assay. The height of the black bars is the
percentage of the original radio-labelled peptide binding inhibition by
the original unlabelled peptide in 40 mM (*) or unlabelled analogous
peptides. Original peptide specific binding is considered to be 100%.
The letter at the bottom represents the amino acid which was replaced
by glycine in the analogue peptide. The critical residues at 4 and 40 mM
is underlined.
The ETHVTGGSAG HABP-4931 sequence seems to be
very important for HepG2 binding, since peptide 4932 binds
with much lower affinity. It has been reported that anti-
HVR-I specific antibodies at the N-terminus (middle region
or C-terminal portion of this region) are capable of prevent-
ing HCV-infection just where HABP 4931 is located [15].
In fact, it has been reported that HVR-1 N-terminal region is
effective in preventing HCV in vitro binding to human fibro-
blast cells. The literature states that C-terminal portion
HVR-1 antibodies, raised in rabbits, block HCV absorption
to cells [19], suggesting that HABP 4931 is involved in
binding and invasion processes. HVR-1 is the major anti-
genic region recognised by anti-HCV antibodies, containing
linear B-cell epitopes [18,19]; HVR-1 has also been
involved in recurrent and persistent HCV-infection,
259
Fig. 6. Binding assay to HepG2 cell membranes. Binding of peptides
4938, 4931, 4943 and 4951 to human hepatocytes. HepG2 cells were
incubated with 125I-labelled peptide 4938 (100 nM) in the absence (lane
1) or presence (lane 2) of unlabelled peptide 4938 (1 mM). 125I-labelled
peptide (100 nM) 4931 (lane 3), 4943 (lane 4) and 4951 (lane 5). After
the binding, cells were washed, BS3-cross-linked, membranes were
obtained, samples were electrophoresed and 125I-labelled bands were
visualised by autoradiography. The 4938 peptide binds to 60 and
42 kDa bands. The 4931, 4943 and 4951 peptides only bind to the 42-
kDa band. Specific activities for peptides 4931, 4938, 4943 and 4951
were 20,000, 59,700 75,000 and 164,000 cpm/nmol, respectively.
suggesting that it possesses specific epitopes for each
virus isolate [18,19].
Peptides 4944 and 4951 binding to HepG2 cells is inhib-
ited by 4913 and 4938 peptides, indicating that the binding
site for these peptides is the same or closely related (Fig. 4).
This is also supported by the fact that peptides 4931, 4938
and 4951 present a similar molecular weight radio-labelled
band in autoradiography. Inhibition of 4944 and 4951 bind-
ing to HepG2 cells caused by peptide 4913 also suggests a
possible association between E1 and E2 proteins in binding
to target cell. Rosa et al. [8] state that recombinant E1/E2
protein can bind to human cells. Competition assays using
Table 2
Binding assay to Raji cells a
Peptide Sequences Binding
HepG2 Raji
4913 YQVRNSTGLYHVTNDCPNSS 2.9 ^ 0.3 5.0 ^ 0.3
4931 ETHVTGGSAGHTVSGFVSLLY 2.0 ^ 0.2 0.1 ^ 0.05
4938 ERLASCRPLTDFDQGWGPISY 2.8 ^ 0.3 4.0 ^ 0.1
4943 PVYCFTPSPVVVGTTDRSGA 5.3 ^ 0.5 0.5 ^ 0.02
a
The binding assays to Raji cells were performed similar to HepG2 cell
binding assays. The 1 £ 10 6 cells were incubated with radio-labelled
peptide in the presence or absence of non-labelled peptide. Non-bound
peptide was eliminated and radio-labelled peptide was quantified. The
4913 and 4938 peptides bind more to Raji cells as compared to HepG2
cells and peptides 4931 and 4943 bind only to HepG2 cells.
260 J.E. Garcia et al. / Journal of Hepatology 36 (2002) 254–262
recombinant proteins should be carried out. Unfortunately,
it has not been possible to perform these assays as yet, due to
difficulties in obtaining recombinant E1 and E2.
Our results indicate that E1- and E2-binding regions have
the same receptor on these cells’ surface. This could indicate
that the virus has redundancy in the ligands used to recognise
the hepatic cells, in turn, implying different pathways for
hepatocyte recognition. The 4913 peptide has a similar
sequence to the linear epitope reported [43], suggesting
that this region is immunogenic. This would be an advantage
allowing the virus to bypass the immune system response.
Competition assays with glycine analogues show that
each one of the peptides 4944 (VVVTTDRSAPTYS-
WEND), 4949 (CAPPCVIANNTLHCPT) and 4950 (AN-
NTLHCPTDCFRKHPDAT) has three critical residues (Fig.
5); threonine is involved in all these. The peptides 4944,
4951 and 4952 present some analogues inhibiting binding
at 40 mM, but not at 4 mM (these modified residues are not
critical). The IANNT sequence is critical for binding to
HepG2 cells in peptides 4949 and 4950. Peptides 4951
(DCFRKHPDATYSRCSPWI) and 4952 (YSRCSPWIT-
PRCLVDYPY) present the YSR-binding motive. IANNT
and YSR sequences seem to be implicated in regions III and
IV binding to HepG2 cells. These binding motives could
directly interact with hepatocyte-binding sites and modifi-
cations in these sequences could be responsible for changing
the three-dimensional (3D) peptide structure.
Peptide-1 [21] (TTDRSGAPTYSWGANDTDVFV), rep-
orted as a subtype 1a specific linear epitope, shares the
underlined sequence with HABPs 4943 (PVYCFTPSP-
VVVGTTDRSGA) and 4944 (VVVGTTDRSGAPTYSW-
GEND). Our binding assays suggest that some residues
(located in 4943 N-terminal portion) could be important in
keeping the peptide binding site’s 3D structure, since their
binding activity is higher than HABP 4944. A B-cell epitope
has also been reported (TWGENETDVLLLNNTRPPQ)
[20], sharing the underlined sequence with HABP 4944
and peptide 4945 (APTYSWGENDTDVFVLNNTR). This
data suggests that 4944 and 4945 sequences are antigenic,
since they were recognised by patients’ sera, and immuno-
genic, since they were able to induce monoclonal antibody
production.
HCV ability to invade CD81 positive B-lymphocytes has
been reported. Different HABPs were tested in binding
assays to the CD81 positive B-lymphoblastoid Raji cell,
showing that HABPs 4913 and 4938 were capable of bind-
ing to Raji cells. However, other reports have shown specific
HVR-I N-terminal region antibodies (384–391), implicating
this E2 region interaction with CD81 [12]. Such results
suggest that HABP 4913 (YQVRNSTGLYHVTNDCPNSS)
and 4938 (ERLASCRPLTDFDQGWGPISY) binding sites
could be similar to those of HepG2 cells and Raji cells, but
not necessarily that the receptor-protein be the same. They
also show that the binding sites for HABPs 4931 and 4943
are not present on Raji cells.
CD81 molecule has been reported as being the E2 recep-
tor. The E2 sequences are located within regions that do not
bind to HepG2 cells. These sequences (PDQRPYCWHYPP
and PPLGNWFG) are contained in peptides 4940
(YANGSGPDQRPYCWHYPPKP) and 4946 (TDVFVL-
NNTRPPLGNWFGCTY), which do not bind to HepG2
cells. This data is in agreement with recent reports showing
that HepG2 cells do not express CD81 molecule [9,12].
There was a 43 kDa double band and another 60 kDa
band in the autoradiography. It seems that these HABPs
recognise similar HepG2 receptor cells and also interact
with more than one receptor. However, the number of bind-
ing sites to HepG2 cells is different, being around 2 £ and
5 £ higher (HABP 4931 has five times more binding sites
than HABP 4943). A hypothesis is that this difference is due
to HABPs recognising different receptor aggregation states.
Our results strongly suggest that these peptides’ binding
sites could be similar. However, the 4938-HABP bind to
two bands (43 and 60 kDa) and the 4931, 4943 and 4951
HABPs bind only to 43 kDa. CD81 molecular weight is
25 kDa [9,10]; but, HABPs bound to these bands have
different molecular weights, suggesting that receptor mole-
cules are different to those described previously.
Peptide 4938 binding two different bands in the autora-
diography could be due to this peptide possessing two bind-
ing motives for two different proteins (42 and 62 kDa). One
motif is the same for peptides 4931, 4943 and 4951; the
other motif may be for binding 4938 to the 42 kDa band,
forming a complex able to strongly bind to 18 kDa
membrane protein, justifying the 60 kDa band.
The results shown here indicate the presence of three
hepatocyte-binding regions in E2, independent of the
presence of the CD81 molecule. These regions, different
from HVR-1, are: region-II (444–484), region-III (505–
543) and region-IV (564–613). These results are in agree-
ment with Rosa et al. [8], suggesting the existence of other
hepatocyte-binding regions within the E2 protein. It has also
been demonstrated that HVR-1 specific antibodies are not
sufficient to block HCV binding to hepatocytes [8].
Some of the peptides identified here are located in regions
reported as being immunogenic. Since the HABP identifica-
tion and recognition test is highly specific and the binding is
saturable, this data is important for developing strategies for
blocking host cell–HCV interaction.
Results thus indicate that HABPs specifically recognise
receptors on HepG2 cell surface, such recognition being in
the nanomolar range, suggesting high specificity. They also
indicate the existence of a common receptor (43 kDa, deter-
mined by SDS-PAGE) for different HABPs. However,
further studies with recombinant proteins must be done to
elucidate this phenomenon.
Acknowledgements
This research project was supported by the Presidency of
the Republic of Colombia and Ministry of Public Health.
 J.E. Garcia et al. / Journal of Hepatology 36 (2002) 254–262
The collaboration of our chemistry and immunology
sections is greatly appreciated, as is that of the whole staff
at the Institute of Immunology and the excellent support
provided by Diana Tovar and Liliana Rodriguez.
Publisher’s note
Due to no author corrections any misprints are the respon-
sibility of the authors.
References
[1] Choo QL, Kuo G, Weiner AJ, Overby LR, Bradley OW, Houghton M.
Isolation of a cDNA clone derived from a blood born Non-A, Non-B
viral hepatitis genome. Science 1989;244:359–361.
[2] Choo QL, Richman K, Han H, Berger K, Lee C, Dong C, et al.
Genetic organization and diversity of the hepatitis C virus. Proc
Natl Acad Sci USA 1991;88:2451–2455.
[3] Kato N, Hijikata M, Ootsuyama Y, Nakagawa M, Ohkoshi Sh, Sugi-
mura T, et al. Molecular cloning of the human hepatitis C virus
genome from Japanese patients with non-A, non-B hepatitis. Proc
Natl Acad Sci USA 1990;87:9524–9528.
[4] Santolini E, Migliaccio G, Monica N. Biosynthesis and bichemical
properties of the hepatitis C virus core protein. J Virol 1994;68:3631–
3641.
[5] Grakoui A, Wychowski C, Lin Ch, Feinstone SM, Rice Ch. Expres-
sion and identification of hepatitis C virus polyprotein cleavage
products. J Virol 1993;67:1385–1395.
[6] Hijikata M, Kato N, Ootsuyama Y, Nakagawa M, Shimotohno K.
Gene mapping of the putative structural region of the hepatitis C
virus genome by in vitro processing analysis. Proc Natl Acad Sci
USA 1991;88:5547–5551.
[7] Lin CH, Linden B, Pragai B, McCourt D, Rice CH. Processing in the
hepatitis C virus E2-NS2 region: identification of p7 and two distinct
E2-specific products with different C termini. J Virol 1994;68:5063–
5073.
[8] Rosa D, Campagnoli S, Moretto C, Guenzi E, Counsens L, Chin M, et
al. A quantitative test to estimate neutralizing antibodies to the hepa-
titis C virus: cytofluorimetric assessment of envelope glycoprotein 2
binding to target cells. Proc Natl Acad Sci USA 1996;93:1759–1763.
[9] Petracca R, Falugi F, Galli G, Norais N, Rosa D, Campagnoli S, et al.
Structure–function analysis of hepatitis C virus envelope-CD81 bind-
ing. J Virol 2000;74:4824–4830.
[10] Pileri P, Uematsu Y, Campagnoli S, Galli G, Falagi F, Petracca R, et
al. Binding of hepatitis C virus to CD81. Science 1998;282:938–941.
[11] Higginbottom A, Quinn E, Kuo Ch, Flint M, Wilson L, Bianchi E, et
al. Identification of amino acid residues in CD81 critical for interac-
tion with hepatitis C virus envelope glycoprotein E2. J Virol
2000;74:3642–3649.
[12] Flint M, Maidens C, Loomis-price L, Shotton Ch, Dubuisson J, Monk
P, et al. Characterization of hepatitis C virus E2 glycoprotein inter-
action with a putative cellular receptor, CD81. J Virol 1999;73:6235–
6244.
[13] Choo QL, Kuo G, Ralston R, Weiner A, Chien D, Nest V, et al.
Vaccination of chimpanzees against infection by the hepatitis C
virus. Proc Natl Acad Sci USA 1994;91:1294–1298.
[14] Esumi A, Rikihisa T, Nishimura S, Goto T, Mizumo K, Zhou YH, et
al. Experimental vaccine activities of recombinant E1 and E2 glyco-
proteins and hypervariable region 1 peptides of hepatitis C virus in
chimpanzees. Arch Virol 1999;144:973–980.
[15] Zhou Y, Shimizu Y, Esumi M. Monoclonal antibodies to the hyper-
261
variable region a1 of hepatitis C virus capture virus and inhibit virus
adsorption to susceptible cells in vitro. Virology 2000;269:276–283.
[16] Zibert A, Dudziak P, Schreier E, Roggendorf M. Characterization of
antibody response to hepatitis C virus protein E2 and significance of
hypervariable region ! -specific antibodies in viral neutralization.
Arch Virol 1997;142:523–534.
[17] Zibert A, Schreier E, Roggendorf M. Antibodies in human sera speci-
fic to hypervariable region 1 of hepatitis C virus can block viral
attachment. Virology 1995;208:653–661.
[18] Kato N, Ootsuyama Y, Hitomi S, Ohkoshi Sh, Nakasawa T, Hijikata
M, et al. Genetic drift in hypervariable region 1 of the viral genome in
persistent hepatitis C virus infection. J Virol 1994;68:4776–4784.
[19] Sekiya H, Kato N, Ootsuyama Y, Nakasawa T, Yamauchi K, Shimo-
tohno K. Genetic alterations of the putative envelope proteins encod-
ing region of the hepatitis C virus in the progression to relapsed phase
from acute hepatitis: humoral immune response to hypervariable
region 1. Int J Cancer, 57:664–70.
[20] Lee JW, Kim KM, Jung SH, Lee JK, Choi EC, Sung Ych, et al.
Identification of a domain containing B-Cell epitopes in hepatitis C
virus E2 glycoprotein by using mouse monoclonal antibodies. J Virol
1999;73:11–18.
[21] Mink M, Benichou S, Madaule P, Tiollais P, Prince A, et al. Char-
acterization and mapping of a B-cell immunogenic domain in hepa-
titis C virus E2 glycoprotein using a yeast peptide library. Virology
1994;200:246–255.
[22] Aden D, Fogel A, Plotkin S, Damjanov I, Knowles B. Controlled
synthesis of HBsAg in a differentiated human liver carcinoma-derived
cell line. Nature 1979;282:615–616.
[23] Bchini R, Capel F, Dauguet Ch, Dubanchet S, Petit AM. In vitro
infection of human hepatoma (HepG2) cell with hepatitis B virus. J
Virol 1990;64:3025–3032.
[24] Merrifield RB. Solid phase peptide synthesis I. The synthesis of a
tetrapeptide. J Am Chem Soc 1963;85:2149–2154.
[25] Urquiza M, Rodriguez L, Suarez J, Guzman F, Ocampo M, Curtidor
H, et al. Identification of peptides from the Plasmodium falciparum
MSP-1 protein able to bind to red blood cells. Parasite Immunol
1996;18:515–526.
[26] Vera R, Marin V, Garcia J, Urquiza M, Torres E, Trujillo M, et al.
Amino terminal peptides of the ring infected erythrocyte surface anti-
gen of Plasmodium falciparum bind specifically to erythrocytes.
Vaccine 2000;18:1289–1293.
[27] Yamamura HY, Enna SJ, Kuhar M. Neurotransmitter receptor bind-
ing. New York, NY: Raven Press, 1978.
[28] Neurath R, Kent S, Strick N, Parker K. Identification and chemical
synthesis of a host cell receptor binding site on hepatitis B virus. Cell
1986;46:429–436.
[29] Rodriguez LE, Urquiza M, Ocampo M, Suarez J, Curtidor H, Guzman
F, et al. Plasmodium falciparum EBA-175 kDa protein peptide which
bind to human red blood cells. Parasitology 2000;120:225–235.
[30] Hubbard AL, Wall DA, Ma A. Isolation of rat heaptocyte plasma
membranes. J Cell Biol 1983;96:217–229.
[31] Puentes A, Garcia J, Vera R, Lopez R, Urquiza M, Vanegas M, et al.
Serine repeat antigens peptides bind specifically to red blood cells.
Parasitol Int 2000;49:105–117.
[32] Bukh J, Purcell R, Miller R. Sequence analysis of the core gene of 14
hepatitis C virus genotypes. Proc Natl Acad Sci USA 1994;91:8239–
8243.
[33] Yanagi M, Purcell R, Emerson S, Bukh J. Transcripts from a single
full-length cDNA clone of hepatitis C virus are infectious when
directly transfected into the liver of chimpanzee. Proc Natl Acad
Sci USA 1997;94:8738–8743.
[34] Lin HJ, Seeff LB, Hollinger FB. Occurrence of identical hypervari-
able region I sequences of hepatitis C virus in transfusion recipient
and their respective blood donors. Gen Bank Data base No Accession
AF206447, 1999.
[35] Yanagi M, Purcell RH, Emerson SU, Bukh J. Hepatitis C virus: an
infectious molecular clone of a second major genotype (2a) and lack
262 J.E. Garcia et al. / Journal of Hepatology 36 (2002) 254–262
of viability of intertypic 1a and 2a chimeras. J Virol 1999;262:250–
263.
[36] Wang YM, Ray SC, Laeyendecker O, Ticehurst JR, Thomas DL.
Assessment of hepatitis C virus sequence complexity by electrophore-
tic mobilities of both single- and double-stranded DNAs. J Clin
Microbiol 1998;36:2982–2989.
[37] Greene WK, Cheong MK, Yap KW. Prevalence of hepatitis C virus
sequence variants in South-East Asia. J Gen Virol 1995;76:211–
215.
[38] Kolykhalov AA, Agapov EV, Blight KJ, Mihalik K, Feinstone SM,
Rice CM. Transmission of hepatitis C by intrahepatic inoculation with
transcribed RNA. J Sci 1997;277:570–574.
[39] Munro J, Briggs JD, McCruden EA. Detection of a cluster of hepatitis
C infections in a renal transplant unit by analysis of sequence varia-
tion of the NS% a gene. J Infect Dis 1996;174:177–180.
[40] Bukh J, Purcell RH, Miller RH. At least 12 genotypes of hepatitis C
virus predicted by sequence analysis of the putative E1 gene of isolates
collected worldwide. Proc Natl Acad Sci USA 1993;90:8234–8238.
[41] Farci P, Shimoda A, Wong D, Cabezon T, Gioannis D, Strazzera A, et
al. Prevention of hepatitis C virus infection in chimpanzees by hyper-
immune serum against the hypervariable region 1 of the envelope 2
protein. Proc Natl Acad Sci USA 1996;93:15394–15399.
[42] Shimizu Y, Hijikata M, Iwamoto A, Harvey A, Purcell R, Yoshikura H.
Neutralizing antibodies against hepatitis C virus and the emergence of
neutralization escape mutant viruses. J Virol 1994;68:1494–1500.
[43] Ishida C, Matsumoto K, Fukada K, Matsushita K, Shiraki H, Maeda
Y. Detection of antibodies to hepatitis C virus (HCV) structural
proteins in anti-HCV-positive sera by an enzyme-linked immunosor-
bent assay using synthetic peptides as antigens. J Clin Microbiol
1993;31:936–940.
[44] Shimizu YK, Igarashi H, Kiyohara T, Cabezon T, Farsi P, Purcell RH.
A hyperimmune serum against a synthetic peptide corresponding to
the hypervariable region 1 of hepatitis C virus can prevent viral infec-
tion in cell cultures. Virology 1996;223:409–412.