Int. J.
Cancer: 107, 416 – 424 (2003)
© 2003 Wiley-Liss, Inc.
HUMAN PAPILLOMAVIRUS TYPE 16 AND 18 L1 PROTEIN PEPTIDE BINDING
TO VERO AND HELA CELLS INHIBITS THEIR VLPS BINDING
Ricardo VERA-BRAVO, Marisol OCAMPO, Mauricio URQUIZA, Javier E. GARCı́A, Luis E. RODRı́GUEZ, Alvaro PUENTES, Ramses LÓPEZ,
Hernando CURTIDOR, Jorge E. SUÁREZ, Elizabeth TORRES, Fanny GUZMÁN, Diana Dı́AZ, Jimena CORTES, Marı́a M. BRAVO,
Alba L. CÓMBITA, Oscar OROZCO and Manuel E. PATARROYO*
Fundación Instituto de Inmunologia de Colombia, Instituto Nacional de Cancerologı́a and Universidad Nacional de Colombia,
Bogotá, Colombia
Human papillomaviruses (HPVs) are the cause of epithelial
lesions, HPV type 16 and type 18 being associated with the
development of anogenital cancer. The L1 Major Capsid Pro-
tein (L1) represents about 90% of total HPV protein and is
involved in virus-host cell interaction, but little is known
about this binding process. L1 sequences from HPV types 16
and 18 were synthesized in 56 20-mer peptides, covering the
entire protein, HPLC-purified, 125I-radiolabeled and tested in
VERO and HeLa cell-binding assays to identify those peptides
with high specific binding activity. Peptides 18283 (residues
54 –77) and 18294 (274 –308) from HPV16 L1, as well as 18312
(59 –78) and 18322 (259 –278) from HPV18 L1, presented high
specific target cell binding activity. Peptide 18283 and 18294
affinity constants were 300 and 600 nM, respectively. Enzyme
cell treatment before binding assay indicated that VERO and
HeLa cell peptide receptor is a surface-exposed protein.
There was a 60% reduction in peptide 18283 binding to hep-
arin lyase-treated cells. Cross-linking assays showed that
these proteins molecular weights were around 69 and 54
kDa. Peptides 18283 and 18294 specifically inhibited HPV-16
VLP binding to HeLa cells. According to the L1- and VLP-
reported structure, both peptides are close on the VLP-
surface, belonging to the outer surface broad pockets sug-
gested as being potential receptor sites. Furthermore, it has
been reported that a conserved motif from peptide 18294 is
the target for neutralizing antibodies. These results suggest
that such binding sequences are used by the virus as cell-
binding regions.
© 2003 Wiley-Liss, Inc.
Key words: L1 major capsid protein; peptide; human papillomavirus;
specific binding; VLPs
Cervical cancer is the fifth most common type of cancer world-
wide (around 500,000 new cases diagnosed each year) and the
second major cause of cancer related to women’s deaths.1,2 It has
been estimated that over 95% of cervical cancer is associated with
HPV infection,3 since many cervical carcinomas and pre-invasive
lesions contain human papillomavirus DNA sequences.4
Papillomaviruses are classified as being low-risk or high-risk,
depending on their preferential association with benign or malig-
nant lesions. HPV-16 DNA is found in about 50 – 60% of human
anogenital carcinomas, HPV 18 in about 20% and other HPV
genotypes in a further 15%. There are as yet undefined papillo-
mavirus sequences in about 15% of anogenital carcinomas.5,6
Papillomaviruses (PVs) are nonenveloped particles, character-
ized by having an icosahedral capsid of about 55 nm diameter. The
capsids are composed of 2 virally encoded proteins (the L1 major
and the L2 minor capsid proteins) that migrate on SDS-PAGE at
about 55 and 75 kDa, respectively. The papillomavirus L1 major
capsid protein (L1) is the most abundant virus-encoded protein,
being highly conserved amongst different HPV types, having
amino-acid homology as high as 80%. L1 presents the ability to
self-assemble in virus-like particles (VLPs).7,8 Different PVs seem
to use the same receptor to enter their host cells.7,9,10 Trypsin
treatment of cells markedly reduced their ability to bind to VLPs,
suggesting that a surface cell protein was involved in VLP bind-
ing.11,12
Red blood cell invasion by Plasmodium falciparum merozo-
ites,13–15 reticulocytes by Plasmodium vivax merozoites14 and he-
Publication of the International Union Against Cancer
patic cells by Hepatitis C virus16 has shown that specific receptor-
ligand interactions mediate microbe binding to host cells. Using a
similar methodology, synthetic 125I-radio-labeled peptides were
thus used in receptor-ligand type binding assays to better pin/point
VERO and HeLa cell binding activity from high-risk 16 and 18
type L1 amino acid sequences. VERO and HeLa cell high activity
binding peptides (HABPs) showed saturable binding, cell recep-
tors being sensitive to enzyme treatment and the majority of bound
peptide being surface-exposed. HABPs specifically bound to 69
and 54 kDa HeLa cell surface membrane protein and inhibited
HPV-16 VLP binding to both cells. Furthermore, several HABP
residues formed part of the VLP surface, becoming the target for
neutralizing antibodies.
METHODS
Peptide synthesis
Twenty-seven sequential 20-mer peptides, corresponding to
HPV type 16 L1,17 and 29 sequential 20-mer peptides, correspond-
ing to HPV 18 L1,18 were synthesized by solid-phase multiple
peptide system.19,20 MBHA resin (0.7 meq/g), t-Boc amino-acid
and low-high cleavage techniques were used.21 Peptides were
analyzed by MALDI-TOF mass spectrometry and reverse phase-
high performance liquid chromatography (RP-HPLC). These syn-
thesized peptides are shown in 1-letter code in Figures 1a,b. An
extra Tyr residue was added to a peptide’s C-terminus for radio-
labeling purposes when such peptide did not contain Tyr.
Radio-labeling
125
I-labeling was performed according to previously described
techniques14,15,22–24 in which chloramine T (2.25 mg/mL) and 3.2
␮L Na125I (100 mCi/mL) were added to 5 ␮L peptide solution (1
␮g/␮L). Fifteen microliters of sodium bisulphite (2.75 mg/mL)
and 50 ␮L NaI (0.16 M) were added after 5 min reaction at 18°C.
The radio-labeled peptide was then separated from reaction sub-
products on a Sephadex G-10 column (Pharmacia) (80⫻5.0 mm).
Abbreviations: HABPs, high activity binding peptides; HPLC, high
performance liquid chromatography; HPV, human papillomavirus; Kd,
affinity constant; L1, the L1 major capsid protein; PVs, papillomaviruses;
VLPs, virus-like particles
Grant sponsor: President of the Republic of Colombia’s Office; Grant
sponsor: Colombian Ministry of Public Health
*Correspondence to: Fundación Instituto de Inmunologia de Colombia,
Carrera 50, No 26-00, Bogotá, Colombia. Fax ⫹57-1-4815269.
E-mail: mepatarr@mail.com
Received 28 May 2003; Revised 6 June 2003; Accepted 26 June 2003
DOI 10.1002/ijc.11433
 HPV16-18 L1 PEPTIDES BIND TO EUKARYOTIC CELLS
Cells and cell culture
VERO and HeLa cells were grown as 105 cell mL⫺1 density
monolayers in RPMI 1640 medium (Gibco, Grand Island, NY)
supplemented with 10% fetal calf serum (FCS) (Hyclone Labora-
tories, Inc., Logan, UT) at 37°C, 5% CO2 and detached by using
0.2% trypsin-0.02% EDTA. Binding cells were first washed twice
with FCS-free RPMI 1640 medium and then twice with phosphate-
buffered saline (PBS) before analysis and then held in suspension
for 90 min at room temperature during the binding assay.
Binding assay
Two and a half million VERO or HeLa cells were incubated
with increasing quantities of 125I-labeled peptide (between 3 and
50 pmol) in a 100 ␮L total volume for 90 min at room temperature
in the presence or absence of 20 ␮M unlabeled peptide to deter-
mine binding specificity (the binding assays were done for each
one of the peptides from L1 HPV16 and HPV18). After incuba-
tion, unbound peptide was removed from cells by sedimentation
through a mixture of dibutylphthalate-dioctylphthalate cushion
(d⫽1.015 g mL⫺1) and spun at 9,800⫻g for 3 min. The cell-bound
peptide was measured in an automatic gamma counter (4/200 plus
ICN Biomedicals, Inc., Costa Mesa, CA). The assays were carried
out in triplicate under identical conditions; the mean results of the
triplicate assays are reported and shown graphically.
Binding constant determination
Target cell saturation binding assays were carried out for each
one of the HABPs. According to previously described meth-
ods,14,15,22–24 2.5 ⫻ 106 HeLa cells were incubated with increasing
concentrations of between 2 to 150 pmol 125I-labeled peptide in a
100 ␮L total volume for 90 min at room temperature in the
presence or absence of 20 ␮M unlabeled peptide to determine
binding specificity as mentioned before. Each assay was performed
in triplicate.
Cross-linking assays
The high binding activity peptides were cross-linked to cells to
identify the HeLa cell binding receptor molecule. Around 1 ⫻ 106
cells were subjected to a conventional HABP binding assay
(HABPs 18283 and 18294). Then cells were washed with PBS
following incubation and subjected to cross-linking with 25 ␮M
BS3 Bis (sulfosuccinimidyl suberate) PIERCE, for 20 min at 4°C.
The reaction was stopped with 40 nM Tris-HCl (pH 7.4) and
washed again with PBS. Then cells were treated with lysis buffer
(1% Triton X-100, 10 mM iodoacetamide, 5% SDS, 100 mM
EDTA and 10 mM PMSF). The obtained proteins were solubilized
in Laemmli buffer and separated in SDS-PAGE. The gel was dried
and exposed on BioRad Imaging Screen K (BioRad Molecular
Imager FX; BioRad Quantity One Quantitation Software, ).
Enzyme treatment
The enzymes used were trypsin (Sigma Chemical Co., St. Louis,
MO) and chymotrypsin (Sigma Chemical Co.) from bovine pan-
creas, pronase (Sigma Chemical Co.) from Streptomyces griseus
and neuraminidase (ICN Biomedicals) from Clostridium perfrin-
gens.
Before binding assay. One part of a VERO cell suspension
was treated with trypsin and the other with chymotrypsin to
final 0.2 mg/ml concentration;9,11,12 or neuraminidase treated
using 100 mg/ml at 37°C for 30 min in sodium acetate buffer,
pH 5.1. HeLa cells were treated in the same way. The binding
assay was carried out after extensive cell washing with freshly
prepared isotonic PBS/(0.1 mM PMSF). Untreated cells were
used as control.
After binding assay. Briefly, HeLa cells were incubated with
125
I-labeled peptide for 90 min at 18°C in PBS (pH 7.1) and
unbound peptide was removed.9,11,12 Cells were then suspended in
the buffer and treated with the following enzymes: trypsin, chy-
motrypsin, pronase and neuraminidase, using the above assay
conditions. Cells were subsequently sedimented and washed with
417
PBS; cell-bound peptide was determined by measurement in a
gamma counter. Untreated cells were used as control.
Heparin lyases treatment
HeLa cells were Heparinase I (Sigma Chemical Co.)-treated in
digestion buffer (20 mM Tris-HCl, 50 mM NaCl, 4 mM CaCl2 and
0.01% BSA, pH 7.5), at a final 1 or 3 unit/mL concentration at
37°C for 1 hr. The same treatment was repeated with Heparinase
II (Sigma Chemical Co.). Untreated cells were used as control.
These cells were used in conventional HABP (18283 and 18294)
binding assays. The assay was carried out in triplicate under
identical conditions; the mean results of the triplicate assays are
reported in Figure 3.
Analogue peptide competition binding assay
Comparative analysis between HPV16 L1 and HPV18 L1
HABPs showed that they shared amino acid sequences. A multiple
sequence alignment for HPV16 L1 with other HPV16 strains and
types was carried out for this region (data not shown). Three
analogue peptides were synthesized to determine whether the
identified conserved motif from a high specific-binding 18283
peptide sequence was involved in binding to HeLa cells. The first
analogue peptide (21477) did not contain the second carboxyl
conserved motif. The second (21479) was a small 12-mer analogue
peptide containing only the 2 conserved motifs. The third (23485)
did not contain the first amino conserved motif (see Fig. 4a).
Briefly, cells were incubated with either native unlabeled peptide
or unlabeled analogue peptide (37 ␮M) in the presence of native
125
I-labeled peptide for competition binding assays. The assay was
done, in triplicate, as previously mentioned.
Papillomavirus VLP purification
HPV16 VLPs were produced in insect cells and purified by
ultra-centrifugation in CsCl gradients, according to a previously
described procedure.29,30 Recombinant baculoviruses encoding
HPV16 L1 protein were used to infect SF9 cells at a multiplicity
of infection (MOI) of 20. Cells were harvested 4 days post-
infection and cytoplasmic and nuclear fractions were separated by
Nonidet P-40 treatment (0.5%), followed by centrifuging
(10,000g, 15 min). The nuclear pellet was suspended in PBS and
sonicated 3 times for 15 sec at bursts of 60% maximum power
(Vibra-cell Sonicator). Nuclear lysates were loaded onto CsCl
gradients and centrifuged to equilibrium in a Beckman SW28 rotor
(4°C, 21 hr, 130,400g). CsCl gradient fractions were collected and
densities determined by refractometry. Fractions having a density
of around 1.272 were pooled in 1⫻ PBS and ultra-centrifuged
(4°C, 1 hr, 130,400g). VLP assembly was verified by electron
microscopy. VLP preparations (see Fig. 5a) were applied to 1.5%
negatively stained uranyl acetate 400 mesh carbon-coated grids
and then examined at a nominal magnification of 39,000 with a
Philips CM10 electron microscope.
HABPs block specific VLP-binding to HeLa cells
Intact VLPs were 125I-labeled by the Chloramine T method and
purified by size exclusion chromatography on Sephadex G-10.
These 125I-labeled VLPs, at 23 ␮g/ml concentration, were incu-
bated with 2.0 ⫻ 106 HeLa cells that had been pre-incubated with
nonlabeled VLPs or HABPs for 30 min at 4°C. The 125I-labeled
VLP binding was performed in the presence or absence of several
concentrations (1.2–120 ␮g/ml) of non125I-labeled VLPs or
HABPs. The radioactivity associated with HeLa cells was then
measured following 90 min incubation.
HABP location in HPV16 L1 structure
HABPs were localized in the previously determined HPV L1
3-dimensional structure by X-ray crystalography,31 using Molec-
ular Simulations, Inc. (San Diego, CA) Insight II software.
418 VERA-BRAVO ET AL.

FIGURE 1.
 HPV16-18 L1 PEPTIDES BIND TO EUKARYOTIC CELLS
RESULTS
High specific binding peptides
A previously reported highly specific and sensitive receptor-
ligand saturation assay13–15 was adapted to VERO and HeLa cells
for HPV binding process studies. Two curves were obtained in this
binding assay: the first denoted Total Binding (cells plus 125I-
labeled peptide) and the second Nonspecific Binding (cells, 125I-
labeled peptide and unlabeled peptide excess). The Specific Bind-
ing resulted from subtracting Nonspecific Binding from Total
Binding. The binding activity was defined as being the relationship
of specific binding over added peptide in the 30 – 400 nM peptide
concentration range. Those peptides exhibiting binding activity
greater than or equal to 2% of the added 125I-labeled peptide (or
0.02 pmol bound, per pmol added) were considered to be High
Activity Binding Peptides (HABPs), according to previously es-
tablished criteria.13–15
Peptides exhibited 3 types of behavior in the initial screening
binding assay: a) peptides strongly and specifically interacting with
target cells or HABPs, b) peptides that did not bind to target cells and
c) peptides binding nonspecifically to target cells. Figure 1 shows L1
HPV16 (Fig. 1a) and L1 HPV18 (Fig. 1b) synthetic peptide sequences
and binding activity to VERO and HeLa cells, respectively. The black
bars represent the binding activity. High Activity Binding Peptides
(HABPs) to VERO and HeLa cells recognized more than 50,000
specific binding sites per cell at low 125I-labeled peptide concentra-
tions. Two VERO and HeLa cell HABPs were identified for each
protein; 18283 (81PNNNKILVPKVSGLQYRVFR100) and 18294
(301LYIKGSGSTANLASSNYFPT320) for HPV16 L1, and peptides
18312 (121QDIPKVSAYQYRVFRVQLPD140) and 18322 (321FWN-
RAGTMGGTVPQSLYIKG340) for HPV18 L1. HABP HeLa cell
binding activity was higher than VERO cell binding activity. Peptides
18283 and 18312 were located in the N-terminal region, whereas
peptides 18294 and 18322 were located in the L1 central region; these
2 binding regions were separated by a cysteine-rich region.
Analogue jumbled-peptide binding activity (the same amino
acid composition as HABP-18283 but with a random sequence)
was extremely low (0.014% of specific binding activity), being
less than 7% for that of the native HABP-18283 (data not shown),
showing that binding activity was due to its specific sequence.
Affinity constants
Saturation assays, Hill and Scatchard analysis14,15,22–24 was car-
ried out only with HPV16 18283 and 18294 HABPs (since they
shared a sequence fragment with the other 2 HPV18 18312 and
18322 HABPs) (Fig. 2a). Affinity constants (Kd) were calculated
for each HABP from HPV16 L1: HABP-18283, Kd ⫽ 300; HABP-
18294, Kd⫽ 600. Hill coefficients were also calculated (18283
peptide, nh ⫽ 1.8; 18294 peptide, nh ⫽ 3); these values suggested
positive ligand-receptor cooperativity. The numbers of binding
sites per cell calculated for each peptide were 1,500,000 and
2,500,000 for HABPs 18283 and 18294, respectively.
Cross-linking assays
125
I-radio-labeled HABPs were bound to HeLa cells and then
cross-linked with BS3. Radio-labeled peptide 18283 bound to 69
kDa and 57 kDa bands in SDS-PAGE analysis (Fig. 2b). A
lessening of the radio-labeled band intensity in non specific bind-
ing was also observed, providing extra evidence of binding spec-
FIGURE 1 – HPV16-L1 (a) and HPV18-L1 (b) peptide synthesis and
determination of its binding activity. In our study, synthetic 27 and 28
sequential 20-mer peptides, belonging to HPV16-L1- and HPV18-L1-
sequences (Seedorf et al.17 and Cole et al.,18 respectively) were tested
in VERO and HeLa cell binding assays. The black bars represent the
binding activity of each peptide, determined as being the specific
binding/total added peptide ratio. High Activity Binding Peptides
(HABPs) were those peptides that exhibited binding activity higher
than or equal to 2% of total added peptide. Amino acids were num-
bered according to Chen’s article.31
419
ificity. Peptide 18294 bound to a 54 kDa band. The right lane in
Figure 2b presents Coomassie stained HeLa cell proteins electro-
phoretic pattern.
Enzyme cell treatment
HeLa and VERO cells were treated with different enzymes to
identify the chemical nature of those cell-receptors involved in
peptide binding and to determine whether the bound peptide was
surface exposed by using 2 different protocols, i.e., before- and
after-binding assays, respectively. Peptide binding activity was
compared between enzyme treated and nontreated cells (Fig. 3).
Cells treated with enzymes before binding assay. Figure 3a
shows that neuraminidase treatment, cleaving terminal sialic acids,
did not affect 18283 and 18294 HABP binding activity to HeLa
and VERO cells. HABP 18283 and 18294 binding activity signif-
icantly decreased for trypsin- and chymotrypsin-treated cells; tryp-
sin treatment had the greatest effect on binding activity (left hand
and middle column represent HeLa and VERO binding activity
respectively).
Cells treated with enzymes after binding assay. Forty to sixty
percent of bound 125I-labeled peptide was released after cells were
treated with trypsin, chymotrypsin, using the conditions described
above (Fig. 3a). The data obtained showed that neuraminidase
treatment did not remove the cell-bound 125I-labeled peptide (right
hand column Fig. 3a represents HeLa* binding activity).
Heparin lyases treatment
Previous reports have shown the importance of surface anionic
polysaccharides (heparine and heparan sulfate) in host cells for
VLP binding, where binding is decreased by treating the cells with
heparin lyases.25–27 HeLa cells were treated with heparin lyases
(heparinase I and heparinase II), which are cell-surface heparin and
heparan sulfate specific.28 Cells treated with each one of the
heparinases were submitted to conventional binding assay. Cells
treated with buffer which did not contain enzyme were used as
control (⫽ 100% specific binding).
Figure 3b shows how peptide 18283 binding was reduced by
40% when HeLa cells were treated with heparinase I (Hepar I);
binding was reduced by 60% with heparinase II (hepar II). Treat-
ment with 1 or 3 units/mL presented no significant differences.
However, peptide 18294 did not present significant changes when
cells were treated with these enzymes. On the contrary, an increase
in enzyme concentration seemed to significantly improve peptide
binding to cells (Fig. 3b).
Competition assay with analogue peptides
Comparing HPV16 L1 and HPV18 L1 HABP sequences
showed that these HABPs shared amino acid sequences. It was
shown in HPV16 L1 multiple sequence alignment (with both other
HPV16 variants and other HPV types) that the 18283 HABP
sequence was completely conserved amongst several variants of
the same HPV16 type, whereas there was only one substitution at
residue 308 (Ser-Pro) for HABP-18294 responsible for 95% con-
servation. Four- and 6-amino acid conserved motifs were found for
the HABP-18283 sequence (NKILV PKVS GL QYRVFR
IHLPD, in bold) when other high-risk HPV types (18, 31, 33, 39,
45, 51, 58 and 61) were analyzed. The HABP-18294 sequence
(GENVPDD LYIKG S-GSTANLAS) presented only 1 conserved
5-residue motif in the same type of analysis.
Three analogue peptides were synthesized to define whether the
conserved motifs (PKVS and QYRVFR) from HABP-18283 were
involved in binding to HeLa cells: the 21477 analogue peptide, which
did not contain the second conserved motif, the 21479 12-mer ana-
logue peptide, containing only 2 conserved motifs, and the 23485
analogue peptide, presenting the second conserved motif (Fig. 4a).
Competition binding assays were carried out between 125I-labeled
HABP-18283 and unlabeled 18283 or each one of the analogue
peptides, using a final 11 ␮M concentration. Peptides that did not
contain one of the conserved motifs markedly reduced the ability to
420 VERA-BRAVO ET AL.
inhibit 125I-labeled HABP-18283 binding to HeLa and VERO cells.
The small 21479 analogue peptide interacted with cells as strong as
18283 unlabeled peptide, since it displaced 125I-labeled HABP-18283
from its HeLa cell binding sites (Fig. 4b).
HPV-16 VLP binding to HeLa cells in the presence of HABPs
125
I-labeled HPV-16 VLPs bound to HeLa cells in the absence
or presence of HABPs or control peptides. Nonlabeled VLPs and
peptides 18283 (PNNNKILVPKVSGLQYRVFR), 18294 (LY-
IKGSGSTANLASSNYFPT) and 23489 (LYIKGS) dose-depen-
dently inhibited 125I-labeled VLP binding. Peptide 18283 inhibited
VLP binding just like nonlabeled VLPs did; however, peptides
18294 and 23489 needed higher peptide concentrations to obtain
similar VLP inhibition. Peptide 21479 (PKVSGLQYRVFR) did
not inhibit VLP binding at any concentration (Fig. 5b). Peptides
18283, 18294 and 23489 inhibited VLP binding to HeLa cells by
more than 80% at 120 ␮/ml peptide concentration, in a similar manner
to the control (binding inhibition by VLPs). A shorter binding se-
quence (21479) did not inhibit VLP-binding at this concentration.
DISCUSSION
The first step in the process of papillomavirus infection is the
interaction between viral capsid and host cell surface membrane.
FIGURE 2 – (a) Saturation binding curves for
HABP 18283 and 18294 binding to HeLa cells.
In the inserted Hill plot, the abscissa is Log F and
the ordinate is Log [B/(BM⫺B)] where Bm is the
maximum bound peptide; B is the bound peptide
and F is the free peptide. (b) Covalent crosslink-
ing with BS3 of 125I-labeled peptide 18283 and
for 18294 bound to HeLa cell membrane proteins
and separated by 12% SDS-PAGE.
Papillomavirus capsids are composed of the L1 major capsid
protein (L1) and L2 minor capsid protein. Highly conserved L1 is
the main protein involved in host cell recognition. The capsid has
360 L1 molecules, assembled into 72 capsomers, forming a virion
shell having icosahedral symmetry.9,12,32
It is well known that VLPs bind to different cells lines derived
from different tissues and species.7,9 ␣6 integrin has been identi-
fied as being a component of that cell receptor responsible for VLP
binding to epithelial cells;11 however the supporting experimental
evidence is not strong. Recent studies have shown heparin sulfate
role in VLP interaction with HaCaT cells.25–27 It is probable that a
primary interaction occurs with surface polysaccharides in the host
cell, followed by a specific recognition interaction with a second-
ary receptor.26
Since no reports were found concerning VLPs-L1, virus capsid
regions or amino acid sequences specifically involved in host cell
recognition, our aim was to identify L1-regions involved in virus
invasion interacting with host cells with high affinity and speci-
ficity. Peptides having binding activity to HeLa and VERO cells
equal to or higher than 2% were chosen as being HABPs. Such
binding activity was considered significant because such criterion
allowed peptides that recognize more than 50,000 specific binding
sites per cell at low 125I-labeled peptide concentration to be iden-
 HPV16-18 L1 PEPTIDES BIND TO EUKARYOTIC CELLS 421

FIGURE 3 – HABP-binding activity to enzyme-treated and nontreated HeLa and VERO cells. Each bar represents the binding activity of
125
I-labeled peptides18283 and 18294 to enzyme treated HeLa and VERO cells. (a) The binding assay was performed before [shown by the left
(HeLa cells) and middle (VERO cells columns) and after (right column showing only HeLa cells*). The cells were treated with trypsin,
chymotrypsin or neuraminidase. (b) Altering polysaccharide chains on HeLa cell-surface for Heparinase I or II at different enzyme concentra-
tions affected HABP binding to these cells. Untreated HeLa cells were used as control. Each assays was done in triplicate.

FIGURE 4 – Competition binding assays were

performed between 18283 and its analogues.
Analogue peptides 21477, 21479 and 23485,
located close to the HABP-18283 sequence in
the L1-region, were synthesized (a). Cells pre-
viously washed with PBS were subjected to a
conventional binding assay with either 18283
unlabeled peptide or analogue peptides (11
␮M) in the presence of 125I-labeled 18283 pep-
tide. The bars show HeLa cell binding inhibi-
tion percentage for each analogue peptide (b).
The inhibition presented by the 18283 peptide
was considered as being 100%.

tified; this number of HeLa cell binding sites was relevant for
HPV-HeLa cell interaction since it has been reported that HeLa
cells bind up to a maximum of 20,000 VLPs.9
Two HABPs were identified for each one of the 2 proteins
analyzed: peptides 18283 and 18294 for L1 from HPV type 16 and
peptides 18312 and 18322 for L1 from HPV type 18. HABPs
bound to HeLa cells with higher binding activity than to VERO
cells, probably due to differences in affinity and/or capacity. It is
in agreement with the fact that HeLa cells are derived from the
natural host cells for HPV type 16 and 18. It is worth mentioning
that these sequences were highly conserved, not only in different
strains, but also in different types. Their binding activity was
422 VERA-BRAVO ET AL.
dependent on peptide sequence and not amino-acid composition,
since their “jumbled“ peptides did not exhibit binding activity.
The number of binding sites per cell for each of the high-binding
peptides was between 1,500,000 and 2,500,000. These values were
higher than those observed for VLPs with 1 ⫻ 104 to 2 ⫻ 104
receptors per cell;10,12 however, each VLP contained around 360
L1 molecules, theoretically giving between 3,600,000 and
7,200,000 binding sites per cell for L1. This amount lay in the
same HABP binding site range. HABP dissociation constants (Kd)
were less than 700 nM, suggesting that these peptides possessed
high affinity for cell membrane receptors. The VLPs showed Kds
of about 100 pM.12 under saturation conditions, which was higher
than the HABPs. This could be due to VLPs presenting more than
1 binding site.
Neuraminidase treatment of VERO and HeLa cells before bind-
ing assay did not diminish HABP binding activity, indicating that
HABP cell-receptors are not sialic acid dependent. This agrees
with that reported by Volpers concerning VLP binding to eukary-
otic cells.12
Trypsin or chymotrypsin treatment of cells before binding assay
reduced 18283 and 18294 peptide (HPV type 16 L1) binding
activity, suggesting that HABP cell receptors are cell surface
exposed proteins. These results agree with those obtained by other
authors where binding to VLPs from several HPV types were very
susceptible to target cell trypsin-treatment.9,11,12 Trypsin and chy-
motrypsin HeLa cell treatment after binding assay removed at least
65% of bound peptide, showing that the majority of bound peptide
was cell surface exposed and not endocytozed or phagocytozed.
Moreover, the HABP-receptor on HeLa and VERO cells seemed to
be similar, according to these enzymes’ susceptibility.
Peptides 18283 and 18294 shared motifs from the peptide’s full
sequence with 18312 and 18322, respectively. These motifs are
FIGURE 5 – VLP-binding to HeLa cells in the presence of
HABPs. (a) Microphotography of VLPs produced as previously
reported and used in HeLa cell competition binding assays. (b)
VLP binding to HeLa cells was carried out at different HABP-
concentrations (1.2–120 ␮/ml); VLP specific binding to HeLa cells
was taken as being 1 (Relative VLP-binding); VLP binding inhi-
bition is seen as being relative VLP-binding inhibition. Each point
represents the mean of 3 experiments; the bars represent standard
deviations.
present in other HPV types, suggesting that motifs are conserved.
The LYIKG motif seems to be very important for HeLa and
VERO cell binding, since peptides 18293 and 18323 (from
HPV-16 L1 and HPV-18 L1, respectively), which did not contain
this LYIKG motif, did not exhibit binding activity. This highly
conserved motif is only present in HABPs 18322 and 18294.
Peptide 23489 (containing this motif) was also able to inhibit VLP
binding to HeLa cells (Fig. 6b,c). The conserved motifs for HABP-
18283 (PKVS GL QYRVFR in bold) were very important for
HeLa cell binding, since peptides with only 1 conserved motif
exhibited binding activity lower than peptides containing both
motifs. Peptides 18284 and 18311 (sharing sequences with
HABPS 182383 and 18312 but not containing these motifs) did not
exhibit high binding activity. Moreover, peptide 21479 (which had
the 2 motifs) did exhibit high binding activity but was unable to
inhibit VLP binding to HeLa cells.
It has been reported that both heparin and heparan sulphate play
a role VLP interaction with heparin on keratinocyte surface.25–27
Treating HeLa cells with heparinase I and II prior to binding assay
provided evidence that HABP 18283 could be interacting with
heparin or heparan sulphate, since binding significantly decreased
(60%) respecting control. However, this did not exclude the pos-
sibility of these enzymes having an indirect effect on HABP
binding activity to these cells. Peptide 18283 specific binding
decreased after heparinase II-treatment, but 40% of specific bind-
ing remained, suggesting interaction with other cell receptors.
Several viruses have been shown to bind initially to heparan
sulfate and then interact with a second receptor, leading to cell
invasion.33–36 Similar results have been obtained by Giroglou et
al.26), showing the participation of a second receptor molecule for
virion uptake, when they studied HPV pseudovirion binding to
COS-7 cells. HABP 18294 binding activity was not affected by
 HPV16-18 L1 PEPTIDES BIND TO EUKARYOTIC CELLS
FIGURE 6 – Positions of the 18283 and 18294 HABPs on the
L1-Structure: HABP 18283 is shown in yellow and 18294 in
magenta. The main loops are shown and named according to the
L1-Structure reported by Chen et al.33 Both HABPs are located on
the VLP-surface and form part of the potential receptor pockets.
enzyme treatment of HeLa cells, independently of the heparin
lyase concentration used (1 and 3 unit/mL), suggesting that this
peptide does not interact with heparan sulfate negative charges.
Experimental evidence regarding HABP sequences involved in
virus attachment to host cells could be deduced from HABP ability
to inhibit VLP binding to HeLa cells. However, peptide 21479
(which is a shorter HABP) did not inhibit VLP binding to HeLa
cells, showing that HABP 18283 N-terminal residues are very
important for efficient inhibition of VLP binding to HeLa cells.
Interestingly, the very small 23489 peptide (LYIKGS) was able to
inhibit VLP binding to HeLa cells. The results shown here suggest
that the 5-residue motif (LYIKG) present in HABPs 18294 and
18322 is absolutely necessary for binding to HeLa and VERO
cells. Synthetic peptides (containing the LYIKG sequence), re-
combinant L1 protein and VLP-elicited antibodies recognized the
LYIKG amino acid sequence.32 These antibodies also recognized
the intact virus from several HPV types, suggesting that the se-
quence was exposed on viral capsid.32,37– 40 Furthermore, this
sequence is specifically recognized by sera from patients suffering
from cervical intraepithelial neoplasia.41
HABPs 18283 and 18294 are on the VLP surface, forming part
of the C-beta strand and the B-C and F-G loops, according to the
3D structure reported by Chen et al.31 for the L1-HPV 16 protein
(Fig. 6). They showed that the H-I loop is inserted between the
anticlockwise neighbor protein F-G and E-F loops in the assem-
bled L1 pentamer,31 confirming that our HABPs were exposed on
its surface, being located close together in VLP structure. The BC
and FG loops formed part of the 5 outer surface broad pockets that
have been suggested as being potential HVP receptor sites. The
HPV-16 FG loop inserted into L1-HPV-11 generates hybrid-VLPs
able to elicit neutralizing antibodies against both HPV-16 and
REFERENCES
1. Pisani P, Parkin DM, Ferlay J. Estimates of the worldwide mortality
from eighteen major cancers in 1985: implications for prevention and
projections of future burden. Int J Cancer 1993;55:891–903.
2. Pisani P, Parkin DM, Bray F, Ferlay J. Estimates of the worldwide
mortality from 25 cancers in 1990. Int J Cancer. 1999;83:18 –29.
3. Walboomers JMM, Husman D, Van Der Brule AM, Snijders AJC,
Meijer CJLM. In: Human papillomavirus and cervical cancer: biology
and immunology, Oxford: Oxford University Press, 1994. 41–71.
4. Bosch FX, Lorinzc A, Munoz N, Meijer CJ, Shah KV. The causal
relation between human papillomavirus and cervical cancer. J Clin
Pathol 2002, 55:244 – 65
5. Bosch, F. X, Manos M, Muñoz N, Sherman M, Jansen AM, Petro J,
Schifman MH, Moreno V, Kurman R, Shah, KV. Prevalence of
human papillomavirus in cervical cancer: a worldwide perspective.
J Natl Cancer Inst 1995;87:796 – 802.
6. von Knebel Doebertz, M. Papillomavirus in human disease: Part 1.
423
HPV-11.42), supporting the role of the 18294 sequence in eliciting
a protective immune response.
So far the majority of the neutralizing antibodies have been
reported as being conformational rather than linear; however, 3
different linear B-cell epitopes have been recognized in HABP
18294. The first epitope (residue 264 to 278) is identified in native
protein.39 The second one (residue 269 to 284) is recognized by
monoclonal antibodies capable of identifying HPV-16 particles in
biopsy.32 The third (261 to 280) is recognized by monoclonal
antibodies recognizing intact VLPs.37 Residue Ser 282 located in
HABP 18294 is important for binding neutralizing monoclonal
antibody E70.43,44
Peptide specific binding was confirmed by the following facts.
Binding could be inhibited by adding non125I-labeled, concentra-
tion-dependent peptide. These HABPs showed high affinity, satu-
rable binding. Peptide binding activity was decreased by treating
cells with enzymes before binding assay. These 125I-labeled pep-
tides recognized a HeLa membrane protein by SDS-PAGE auto-
radiography. Finally, these HABPs were able to inhibit VLP HeLa
binding. Our results suggest that the HeLa and VERO cell binding
sequences reported here could be part of the binding sites used by
HPV for invading target cells.
ACKNOWLEDGEMENTS
Gloria Barrera’s contribution with electron micrography (COR-
POICA) and Pierre Coursaget’s generous donation of cells in-
fected with HPV16-L1 (Université de Tours) is greatly appreci-
ated. Jason Garry’s help in reviewing this article is greatly
appreciated. This article is in memory of Dr. Oscar Orozco.
Pathogenesis and epidemiology of human papillomavirus infections.
Med Intel 1992;1:415–23.
7. Roden, RBS, Kirnbauer R, Jenson AB, Lowy DR, Schiller JT. Inter-
action of papillomavirus with the cell surface. J Virol
1994;68:7260 – 6.
8. Rose, R. C, Bonnez W, Reichman R. C, Garcea, R. L. Expression of
human papillomavirus type 11 protein in insect cells: in vivo and in
vitro assembly of virus-like particles. J Virol 1993;67:1936 – 44.
9. Müller M, Gissman L, Cristiano RJ, Sun X, Frazer IH, Jenson AB,
Alonso A, Zentgraf H, Zhou J. Papillomavirus capsid binding and
uptake by cells from different tissues and species. J Virol 1995;69:
948 –54.
10. Qi YM, Peng SW, Hengst K, Evander M, Park DS, Zhou J, Frazer IH.
Epithelial cells display separate receptors for papillomavirus VLPs
and for soluble L1 capsid protein. Virology 1996;216:35– 46.
11. Evander M, Frazer IH, Payne E, Qi YM, Hengst K, McMillan NAJ.
424 VERA-BRAVO ET AL.
Identification of the ␣6 integrin as a candidate receptor for papillo-
maviruses. J Virol 1997;71:2449 –56.
12. Volpers C, Unckell F, Schirmacher P, Streeck, RE, Sapp M. Binding
and internalization of human papillomavirus type 33 virus-like parti-
cles by eukaryotic cells. J Virol 1995;69:3258 – 64.
13. Vera R, Marı́n V, Garcı́a J, Urquiza M, Torres E, Trujillo Mı̀, Rosas
J, Patarroyo ME. Amino terminal peptides of the ring infected eryth-
rocyte surface antigen of Plasmodium falciparum bind specifically to
erythrocytes. Vaccine 2000;18:1289 –93.
14. Ocampo M, Vera R, Rodriguez LE, Curtidor H, Urquiza M, Suarez J,
Garcı̀a J, Puentes A, Lòpez R, Trujillo M, Torres, E, Patarroyo ME.
Plasmodium vivax Duffy binding protein peptides specifically bind to
reticulocytes. Peptides 2002;23:13–22.
15. Rodrı́guez LE, Urquiza M, Ocampo M, Curtidor H, Suarez J, Garcı̀a
J, Vera R, Puentes A, Lòpez R, Pinto M, Rivera Z, Patarroyo ME.
Plasmodium vivax MSP-1 peptides have high specific binding activity
to human reticulocytes. Vaccine 2002;20:1331–39.
16. Garcı́a JE, Puentes A, Suarez J, Lòpez R, Vera R, Rodrı̀guez LE,
Ocampo M, Curtidor H, Guzmán F, Urquiza M, Patarroyo ME.
Hepatitis C virus (HCV) E1 and E2 protein regions that specifically
bind to HepG2 cells. J Hepatol 2002;36:254 – 62.
17. Seedorf, K, Oltersdorf T, Krämmer G, Röwekamp W. Identification of
early proteins of the human papilloma viruses type 16 (HPV16) and
type 18 (HPV18) in cervical cancinoma cells. EMBO J 1987;6:139 –
44.
18. Cole ST, Danos O. Nucleotide sequence and comparative analysis of
the human papillomavirus type 18 genome: phylogeny of papilloma-
viruses and repeated structure of the E6 and E7 gene products. J Mol
Biol 193:599 – 608, 1988.
19. Houghten RA. General method for the rapid solid phase synthesis of
large numbers of peptides: specifity of antigen antibody interaction at
the level of individual amino acids. Proc Natl Acad Sci 1985;82:
5131–5.
20. Merrifield, R. B. Solid phase peptide synthesis. 1. The synthesis of
tetrapeptide. J Am Chem Soc 1963;85:2149 –54.
21. Tam TP, Heath WF, Merrified RB. SN2 deprotection of synthetic
peptides with a low concentration of HF in dimethyl sulfide: Evidence
and application in peptide synthesis. J Am Chem Soc 1983;105:
6442–5.
22. Hulme EC. Receptor-ligand interactions: a practical approach. New
York: Oxford University Press, 1993.
23. Weiland GA, Molinoff PB. Quantitative analysis of drug-receptor
interactions. 1. Determination of kinetic and equilibrium properties.
Life Sci 1981;29:311–3.
24. Yamamura HI, Enna SJ, Kuhar MJ. Neurotransmitter receptor bind-
ing. New York: Raven Press, 1978.
25. Joyce JG, Tung JS, Przysiecki CT, Cook JC, Lehman ED, Sands JA,
Jansen KU, Keller PM. The L1 major capsid protein of human
papillomavirus type 11 recombinant virus-like particles interacts with
heparin and cell-surface glycosaminoglycans on human keratinocytes.
J Biol Chem 1999;274:5810 –22.
26. Giroglou T, Florin L, Schafer F, Streeck RE, Sapp M. Human papil-
lomavirus infection requires cell surface heparan sulfate. J Virol
2001;75:1565–70.
27. Combita AL, Touze A, Bousarghin L, Sizaret PY, Munoz N, Cour-
saget P. Gene transfer using human papillomavirus pseudovirions
varies according to virus genotype and requires cell surface heparan
sulfate. FEMS Microbiol Lett 2001;204:183– 8.
28. Linker A, Hovingh P. Isolation and characterization of oligosaccha-
rides obtained from heparin by the action of heparinase. Biochemistry
1972;11:563– 8.
29. Touze A, El Mehdaoui S, Sisarte P, Mougin C, Muñoz N, Coursaget
P. The L1 mayor capside protein of human papilomavirus type 16
variants affects yield of virus like particles produced in an insect cell
expresion system. J Clinical Microbiol 1988;36:2046 –51.
30. Zhou J, Sun XY, Stenzel DJ, Frazer IH. Expression of vaccinia
recombinant HPV16 L1 and L2 ORF proteins in epithelial cells is
sufficient for assembly of HPV virion-like particles. Virology 1991;
185:251–7.
31. Chen XS, Garcea RL, Goldberg I, Casini G, Harrison SC. Structure of
small virus-like particles assembled from the L1 protein of human
papillomavirus 16. Mol Cell 2000;5:557– 67.
32. Cason J, Patel D, Naylor J, Lunney D, Shepherd PS, Best JM,
McCance DJ. Identification of immunogenic regions of the major coat
protein of human papillomavirus type 16 that contain type-restricted
epitopes. J Gen Virol 1989;70:2973– 87.
33. Mondor IS. Ugolini QJ. Sattentau. Human immunodeficiency virus
type 1 attachment to HeLa CD4 cells is CD4 independent and gp120
dependent and requires cell surface heparans. J Virol 1998;72:3623–
34.
34. Chen Y, Maguire T, Hileman RE, Fromm JR, Esko JD, Linhardt RJ,
Marks RM. Dengue virus infectivity depends on envelope protein
binding to target cell heparan sulfate. Nat Med 1997;3:866 –71.
35. Shukla D, Liu J, Blaiklock P, Shworak NW, Bai X, Esko JD, Cohen
GH, Eisenberg RJ, Rosenberg RD, Spear PG. A novel role for
3-O-sulfated heparan sulfate in herpes simplex virus 1 entry. Cell
1999;99:13–22.
36. Summerford C, Samulski RJ. Membrane-associated heparan sulfate
proteoglycan is a receptor for adeno-associated virus type 2 virions.
J Virol 1998;72:1438 – 45.
37. Christensen ND, Dillner J, Eklund C, Cartes JJ, Wipf GC, Reed CA,
Cladel NM, Galloway, DA. Surface conformational and linear
epitopes on HPV-16 and HPV-18 L1 virus-like particles as defined by
monoclonal antibodies. Virology 1996;223:174 – 84.
38. Eklund C, Dillner J. A two-site enzyme immunoassay for quantitation
of human papillomavirus type 16 particles. J Virol Methods 1995;53:
11–23.
39. Kulski JK, Sadleir JW, Kelsall SR, Cicchini MS, Shellam G, Peng
SW, Qi YM, Galloway DA, Zhou J, Frazer IH. Type specific and
genotype cross reactive B-epitopes of the L1 protein of HPV16
defined by a panel of monoclonal antibodies. Virology 1998;243:275–
82.
40. Zhou J, Sun XY, Davies H, Crawford L, Park D, Frazer IH. Definition
of linear antigenic regions of the HPV16 L1 capsid protein using
synthetic virion-like particles. Virology 1992;189:592–9.
41. Cason J, Kambo PK, Best JM, McCance DJ. Detection of antibodies
to a linear epitope on the major coat protein (L1) of human papillo-
mavirus type-16 (HPV-16) in sera from patients with cervical intra-
epithelial neoplasia and children. Int J Cancer 1992;50:349 –55.
42. Christensen ND, Cladel NM, Reed CA, Budgeon LR, Embers MA,
Skulsky DM, McClements WL, Ledmerer SW, Jensen KU. Hybrid
papillomavirus L1 molecules assembled into virus like particles that
reconstitute conformational epitopes and induce neutralizing antibod-
ies to distinct HPV types. Virology 2001;291:324 –34.
43. Roden RS, Armstrong A, Haderer P, Christensen ND, Hubbert NL,
Lowy DR, Schiller JT, Kirnbauer R. Characterization of a human
papillomavirus type 16 variant dependent neutralizing epitope. J Virol
1997;71:6247–52.
44. White WI, Wilson S. Palmer-Hill FJ, Woods RM, Ghim S, Hewitt
LA, Goldman DM, Burke SJ, Jenson AB, Koenig S, Suzich JJ. Char-
acterization of a major neutralizing epitope on human papillomavirus
type 16 L1. J Virol 1999;73:4882–9.