Content uploaded by Geert Leroux-Roels
Author content
All content in this area was uploaded by Geert Leroux-Roels on Nov 11, 2015
Content may be subject to copyright.
Downloaded from www.microbiologyresearch.org by
IP: 24.222.136.41
On: Thu, 05 Nov 2015 07:06:13
Journal of General Virology (2002), 83, 1281–1289. Printed in Great Britain
...................................................................................................................................................................................................................................................................................
Hepatitis B virus surface antigen suppresses the activation of
monocytes through interaction with a serum protein and a
monocyte-specific receptor
Peter Vanlandschoot, Freya Van Houtte, Annelies Roobrouck, Ali Farhoudi and Geert Leroux-Roels
Center for Vaccinology, Department of Clinical Biology, Microbiology and Immunology, Ghent University Hospital, De Pintelaan 185,
9000 Ghent, Belgium
During hepatitis B virus (HBV) infection, high numbers of non-infectious HBV surface antigen
(HBsAg) particles are present in circulation. It is shown here that recombinant HBsAg (rHBsAg)
particles, which contain the S protein only, bind almost exclusively to monocytes. Attachment of
rHBsAg to the THP-1 pre-monocytic cell line occurs upon 1,25-dihydroxyvitamin D3-induced
differentiation. Binding to monocytes is enhanced by a heat-labile serum protein and is inhibited
by Ca2M/Mg2M, low pH and an HBsAg-specific monoclonal antibody. Furthermore, it is shown that
rHBsAg suppresses lipopolysaccharide- and IL-2-induced production of cytokines. These results
suggest the existence of a monocyte-specific receptor, the engagement of which by HBsAg
suppresses the activity of these cells.
Introduction
Worldwide, hepatitis B virus (HBV) causes more than 1
million deaths per year and about 350–400 million people are
persistently infected with this agent. Most adult individuals
will clear a primary infection, which can be asymptomatic or
can result in an acute, generally self-limited, liver inflammation
causing varying degrees of hepatocellular damage. However,
approximately 5–10% of infected subjects will not resolve the
primary infection and go on to develop a persistent infection.
Vaccination is an effective way to prevent infection, but the
giant virus reservoir in persistent carriers is a major obstacle to
rapid eradication of the virus.
Embedded in the viral membrane are three related viral
membrane proteins, L, M and S, which share 226 carboxy-
terminal amino acids. During HBV infection, non-infectious
subviral lipoprotein HBV surface antigen (HBsAg) particles are
produced in large quantities by the infected hepatocytes and
are secreted into circulation, where concentrations of 50–
300 µg\ml are attained (Ganem, 1996). Like virions, these
lipoprotein particles contain predominantly the S protein,
smaller amounts of M protein and hardly any L protein.
Recombinant expression of only the S protein in yeast,
Author for correspondence: Peter Vanlandschoot.
Fax j32 9 240 36 54. e-mail Peter.Vanlandschoot!rug.ac.be
mammalian and insect cells demonstrated that this protein has
the unique property to form these HBsAg particles, which
contain up to 30% of cell-derived lipids. The reason for the
existence of, or the possible advantage of, the production of
HBsAg remains elusive, until today. It is, however, remarkable
that, in both acutely and chronically infected persons, a cellular
and humoral anti-HBsAg response is lacking, despite the
presence of HBsAg (Milich, 1997). Because anti-envelope
antibodies are clearly detectable only in patients who recover
from acute hepatitis and not in chronically infected subjects,
these are thought to play a critical role in virus clearance.
The mechanism by which HBV establishes a persistent
infection is at present still unclear. Studies with HBV transgenic
mice led to the generally accepted idea that tolerance at the T
cell level is an important underlying mechanism for the
establishment of the persistent state, especially in neonates
(Milich, 1997; Chisari & Ferrari, 1995; Chisari, 1995). How-
ever, defects in the antigen-presenting activity of dendritic
cells, rather than functional defects in T or B cells are claimed
to be responsible for the induction of HBV persistence (Akbar
et al., 1993; Kurose et al., 1997). In vitro studies demonstrate a
reduced capacity of PBMCs from chronically infected persons
to produce cytokines upon stimulation with lipopolysaccharide
(LPS) (Muller & Zielinski, 1990, 1992), while HBsAg inhibits
the release of LPS-induced cytokines by human macrophages
(Jochum et al., 1990). Taken together, these results suggest that
0001-8347 #2002 SGM BCIB
Downloaded from www.microbiologyresearch.org by
IP: 24.222.136.41
On: Thu, 05 Nov 2015 07:06:13
P. Vanlandschoot and othersP. Vanlandschoot and others
HBV infection or virus products may interfere with the normal
function of antigen-presenting cells, like monocytes, macro-
phages and dendritic cells, which may add to the development
of HBV persistence. To examine if and how HBsAg can
influence the activity of monocytes, the physical interaction of
HBsAg with PBMCs was studied by FACS, using biotinylated
yeast-derived S particles. It is reported here that such particles
bind almost solely to monocytes and not to T cells, while some
interaction with B cells is observed. It is further shown that
recombinant HBsAg (rHBsAg) particles not only inhibit LPS-
induced secretion of IL-1βand TNFα, but also inhibit IL-2-
induced secretion of IL-8.
Methods
rHBsAg. Purified rHBsAg (subtype adw#) produced in Saccharomyces
cerevisiae [lots DVP23 (752 µg\ml in PBS) and DVP93\1(1mg\ml)] was
kindly provided by GlaxoSmithKline Biologicals. The purity of these
rHBsAg preparations, as judged by HPLC analysis as well as SDS–PAGE
with Coomassie staining, was 98%. rHBsAg is composed of well-
defined subvirus particles, which contain the non-glycosylated S protein
only. Similar preparations are used worldwide as human HBV vaccines
after adsorption onto aluminium hydroxide.
Biotinylation of rHBsAg. rHBsAg was biotinylated using an
enhanced chemiluminescent protein biotinylation module (RPN 2202,
Amersham Pharmacia). 300 µl rHBsAg was mixed with 270 µlH
#O,
30 µl0n8 M bicarbonate buffer, pH 8n6, and 15 µl biotinylation reagent.
The mixture was incubated at room temperature for 1 h, after which
24 µl 1 M Tris was added. Biotinylated rHBsAg (b-rHBsAg) was
purified by gel filtration on a Sephadex G25 column using PBS. Fractions
of 1 ml were collected and the two b-rHBsAg peak fractions, as
determined by ELISA, were pooled.
ELISA. Maxisorb 96-well plates (Nunc) were coated with rHBsAg or
b-rHBsAg in PBS. The wells were blocked with 0n1% BSA in PBS,
followed by washing three times (0n1 % Triton X-100). HBsAg-specific
monoclonal antibodies (MAbs) (1 µg\ml) or streptavidin–horseradish
peroxidase were added and plates were incubated for 1 h at room
temperature. MAbs were detected with goat anti-mouse or goat anti-
human antibodies labelled with peroxidase. After three washes, 3,3h,5,5h-
tetramethylbenzidine (Sigma) was added and, 30 min later, the reaction
was stopped with 1 N H#SO%.
Antibodies. Mouse anti-human CD3–FITC (clone SK7), CD14–
FITC (clone MP9), CD19–FITC (clone 4G7), IgG1–FITC isotype control
and streptavidin–phycoerythrin (Strep–PE) antibodies were purchased
from Becton Dickinson. Mouse anti-human CD14–FITC (clone My4) and
IgG2b–FITC isotype control antibodies were purchased from Immuno-
tech. Human anti-HBsAg clones F47B and F9H9 were a kind gift from Lia
Sillekens (Centraal Laboratorium van de Bloedbank, Amsterdam). Human
MAb anti-a was developed in the laboratory. Mouse anti-d and anti-y
were a kind gift from DiaSorin.
Cells. Human PBMCs were isolated from buffy coats using
Ficoll–Hypaque (density l1n077 g\ml, Nycomed Pharma) centrifuga-
tion. Cells were stored in liquid nitrogen. Phorbol 12-myristate 13-
acetate treatment (25 ng\ml) (PMA, Sigma) was performed for 4 h in
cRPMI (RPMI 1640 containing 10 % foetal calf serum, 2 mM -glutamine,
1 mM sodium pyruvate, 50 U\ml penicillin, 50 µg\ml streptomycin and
20 µMβ-mercaptoethanol). THP-1 cells were grown in cRPMI. To
induce differentiation, 100 nM 1,25-dihydroxyvitamin D3 (1,25-VitD3,
Calbiochem) was added for 24 or 48 h. Cultured cells were detached
mechanically or by using non-enzymatic cell dissociation buffer (Sigma),
washed twice with 2% non-heat-inactivated human AB serum (HS, Bio-
Whitaker) in Hankhs Balanced Salt Solution (Gibco BRL) (2% HS–HBSS)
and stained as described below.
Staining of cells. PBMCs were thawed and washed twice with 2 %
HS–HBSS. Approximately 10'cells were incubated with b-rHBsAg in
200 µl 2% HS–HBSS for 1 h on ice. After two washes with the same
solution, cells were incubated with Strep–PE and\or FITC-labelled
antibodies in 2% HS–HBSS for 1 h on ice. After two washes, cells were
resuspended in 1 ml 2% HS–HBSS or PBS containing propidium iodide
(PI) and analysed on a FACScan flow cytometer (Becton Dickinson). Dead
cells, which incorporated PI, were gated out of analysis. At least 5000
cells were counted per analysis. Fluorescence (530 nm for FITC and
580 nm for PE) was measured. Median fluorescence was determined in
each case. Signals were acquired in a logarithmic mode for Fl1 (FITC) and
Fl2 (PE). Threshold levels were set according to negative (Strep–PE only)
and isotypic controls.
LPS treatment of THP-1 cells and PBMCs. THP-1 cells
(5i10&) were treated for 24 h with 100 nM 1,25-VitD3. After washing,
the cells were incubated in cRPMI either with or without 10 or 50 ng\ml
LPS (Escherichia coli 0111:B4, Sigma), to which 0, 0n1, 1, 10 or 50 µg\ml
rHBsAg was added. In separate experiments, 10'PBMCs were incubated
in cRPMI either with or without 10 or 50 ng\ml LPS to which 0,
0n1, 1, 10, 25 or 50 µg\ml rHBsAg was added. Cell supernatants were
collected after 24 h and tested for the presence of IL-1βand TNFα.
IL-2 treatment of PBMCs. Approximately 10'PBMCs were
incubated in cRPMI either with or without 1000 U\ml IL-2 (Eurocetus),
after which, 0, 1, 10, 25 or 50 µg\ml rHBsAg was added. Cell super-
natants were collected after 24 h and tested for the presence of
IL-8.
Cytokine determination. The concentrations of IL-1β, TNFαand
IL-8 in the cell supernatants were determined using commercially
available kits (Bioscource) according to the manufacturerhs instructions.
Results
Effect of biotinylation on the antigenicity of rHBsAg
rHBsAg was biotinylated and purified as described. Because
the three lysine residues that can be biotinylated all lie in the
major antigenic region of the S protein, the recognition of b-
rHBsAg by four HBsAg-specific MAbs was investigated (Fig.
1). Biotinylation did not influence the binding of MAb F47B
(Stricker et al., 1985), which recognizes an epitope in the
carboxy-terminal end of the S protein (Paulij et al., 1999), a
region that is predicted to form a membrane-spanning domain
(Stirk et al., 1992). Binding of MAb anti-d was reduced strongly
after biotinylation. This is not unexpected as the lysine residue
at position 122 is the key determinant for the d serotype.
Binding of MAb F9H9 was reduced strongly and that of anti-
a was reduced only slightly. This diminished reactivity is
probably due to biotinylation at residue 141, which lies within
the a determinant.
BCIC
Downloaded from www.microbiologyresearch.org by
IP: 24.222.136.41
On: Thu, 05 Nov 2015 07:06:13
rHBsAg binds to monocytesrHBsAg binds to monocytes
Fig. 1. Biotinylation of rHBsAg interferes with the binding of MAbs specific
for the major antigenic region. ELISA plates were coated with twofold
dilutions of rHBsAg (#) or b-rHBsAg ($). After blocking with BSA,
human MAbs F47B, F9H9 and anti-a and the mouse MAb anti-d were
allowed to bind to the particles. MAbs were detected with either goat anti-
mouse or goat anti-human antibodies labelled with peroxidase.
Biotinylated rHBsAg binds to CD14MPBMCs
In preliminary binding experiments, it was observed that
attachment of b-rHBsAg was enhanced in the presence of
1–2% non-heat-inactivated HS (see below). Therefore, all
binding experiments were performed under these conditions.
PBMCs were incubated with 10 µg\ml b-rHBsAg in 2 %
HS–HBSS, followed by an incubation with MAbs specific for
monocytes, T or B cells. b-rHBsAg was detected with
Strep–PE. Using this concentration of b-rHBsAg, very strong
binding to CD14+cells and some attachment to CD19+cells
were observed. Binding of b-rHBsAg to CD3+cells did not
occur (Fig. 2). Although biotinylation clearly altered the
attachment of some MAbs, it did not prevent the interaction
with the monocytes of PBMCs. Additional experiments
showed that when using 1–2 µg\ml b-rHBsAg, only mono-
cytes stained positive (data not shown). Based on these results,
it was decided to use approximately 1–2 µg\ml b-rHBsAg in
most binding experiments.
Attachment of b-rHBsAg to monocytes is influenced by
serum components
Serum proteins have been shown to bind to HBsAg and are
thought to deliver HBsAg to a cellular membrane protein
Fig. 2. Attachment of b-rHBsAg to CD14+PBMCs. Cells were incubated
for 1 h with 10 µg/ml b-rHBsAg in 200 µl 2 % HS–HBSS, washed twice
with the same buffer and incubated with anti-CD14–FITC (clone P9),
anti-CD19–FITC or anti-CD3–FITC and Strep–PE.
(Imai et al., 1979; Neurath et al., 1992; Budkowska et al.,
1993; Gagliardi et al., 1994; Mehdi et al., 1994). Therefore the
effect of different serum concentrations on the attachment of b-
rHBsAg to CD14-expressing cells was investigated. As shown
in Fig. 3(a), attachment of b-rHBsAg to monocytes was clearly
enhanced at low serum concentrations (1–3%). At higher
concentrations (7–10%), binding was slightly inhibited com-
pared to the serum negative control. These effects were not
observed when BSA was used at similar protein concentrations
(Fig. 3a). Incubation of serum at 56 mC for 30 min inactivated
the factor responsible for the enhanced binding of b-rHBsAg to
PBMCs (Fig. 3b), thus demonstrating its thermolability. Heat
inactivation did not destroy the factor responsible for the
reduction in attachment. These results demonstrate that
different serum components were responsible for the en-
hancement and inhibition.
Non-biotinylated rHBsAg prevents binding of
b-rHBsAg to monocytes
To demonstrate the specificity of the observed interaction,
PBMCs were incubated with different amounts of non-
BCID
Downloaded from www.microbiologyresearch.org by
IP: 24.222.136.41
On: Thu, 05 Nov 2015 07:06:13
P. Vanlandschoot and othersP. Vanlandschoot and others
Fig. 3. (a) Attachment of b-rHBsAg to monocytes is enhanced by low
concentrations of serum and reduced at higher concentrations. PBMCs
were incubated for 80 min on ice with b-rHBsAg in HBSS containing
different concentrations of HS (black bars) or BSA (white bars). After two
washes, b-rHBsAg was detected with Strep–PE. Median fluorescence was
determined. Data shown represent the average of three separate
experiments. Error bars represent SD. (b) Heat-inactivation of serum
destroys the factor that enhances binding of b-rHBsAg to monocytes.
PBMCs were incubated for 80 min on ice with b-rHBsAg in HBSS without
HS (white bar), with 2% HS (black bar) or with 2% HS that was heat
inactivated for either 30 (dark grey bar) or 60 (light grey bar) min. After
two washes, b-rHBsAg was detected with Strep–PE. Median fluorescence
was determined. Data shown represent the average of three separate
experiments. Error bars represent SD.
Fig. 4. (a) Non-biotinylated rHBsAg competes with attachment of b-
rHBsAg to monocytes. PBMCs were incubated with different amounts of
rHBsAg in 200 µl 2 % HS–HBSS. After 1 h, b-rHBsAg was added and the
cells were incubated for another hour. After two washes, cells were
stained with Strep–PE. Median fluorescence was determined. The data
shown represent the average of three separate experiments. Error bars
represent SD. (b) MAb F47B reduces attachment of b-rHBsAg to
monocytes and induces attachment to B cells. b-rHBsAg was incubated
either with (white bars) or without (black bars) 5 µg/ml MAb F47B. After
1 h, PBMCs were added and the cells were incubated for 90 min. After
two washes, cells were stained with Strep–PE, washed twice and analysed.
Median fluorescence was determined. Data shown represent the average
of three separate experiments with a total of six samples per experiment.
Error bars represent SD.
biotinylated rHBsAg. After 1 h, b-rHBsAg was added to this
mixture. As shown in Fig. 4(a), unlabelled rHBsAg blocked the
binding of b-rHBsAg. Inhibition was also observed when non-
biotinylated rHBsAg from lots DVP93\1 and DVP93\2 was
used (data not shown).
BCIE
Downloaded from www.microbiologyresearch.org by
IP: 24.222.136.41
On: Thu, 05 Nov 2015 07:06:13
rHBsAg binds to monocytesrHBsAg binds to monocytes
MAb F74B inhibits attachment of b-rHBsAg to
monocytes
If the binding of b-rHBsAg to monocytes is specific,
antibodies to b-rHBsAg should be able to block the interaction.
To test this prediction, b-rHBsAg was incubated with different
concentrations of F47B, F9H9, anti-a and anti-d antibodies.
Only with MAb F47B was a dose-dependent reduction in
binding observed (data not shown). This was not unexpected,
as biotinylation interferes strongly with the binding of MAbs
F9H9 and anti-d to rHBsAg. Although recognition of b-
rHBsAg by MAb anti-a was reduced only slightly, doses of up
to 20 µg\ml anti-a did not inhibit attachment of b-rHBsAg
to PBMCs. Additionally, with MAb F47B, a dose-dependent
binding of b-rHBsAg to 20–35 % of lymphocytes was detected,
which were identified as B cells (data not shown). Maximal
inhibition of b-rHBsAg binding to the monocytes and maximal
attachment to B cells was obtained already with 5 µg\ml MAb
F47B (Fig. 4b).
Divalent cations reduce binding of b-rHBsAg to
monocytes
Divalent cations that are present at millimolar concentra-
tions in serum can often modulate the interaction between
ligands. Therefore, the effect of Ca#+and Mg#+on the binding
of b-rHBsAg to PBMCs was investigated. The addition of
increasing amounts of Ca#+and Mg#+caused reduced binding
of b-rHBsAg (Fig. 5a) ; the addition of 5 mM EDTA to the
mixture restored attachment. The addition of Ca#+and Mg#+
after binding had no effect (Fig. 5a). Reduced binding was also
observed when only Ca#+or only Mg#+was added (data not
shown). These experiments were all performed in 2 % HS in
50 mM Tris–HCl and 150 mM NaCl instead of HBSS to
prevent acidification when adding 5 mM EDTA.
Low pH reduces binding of b-rHBsAg to monocytes
Conflicting results about the influence of low pH on HBsAg
attachment to cells have been reported (Komai & Peeples,
1990; Mehdi et al., 1996). Therefore, the effect of pH on
particle binding was measured at pH 7, 6 and 5. As shown in
Fig. 5(b), b-rHBsAg binding was strongly reduced when the
pH was lowered. Pre-incubation of the cells or b-rHBsAg at the
same pH and for the same time did not have any effect (Fig. 5b).
Effect of monocyte differentiation state on the
attachment of b-rHBsAg
THP-1 cells, a pre-monocytic cell line, differentiate towards
a monocytic cell type by treatment with 1,25-VitD3. This
differentiation is easily detected by the expression of CD14.
Binding of b-rHBsAg to this pre-monocytic cell line and 1,25-
VitD3-differentiated THP-1 cells was investigated. CD14
expression was detected using two different specific MAbs,
clones P9 and My4. Both antibodies were used because they
Fig. 5. (a) Ca2+and Mg2+interfere with attachment of b-rHBsAg to
monocytes. PBMCs were incubated with b-rHBsAg in the presence of Ca2+
and Mg2+(black bars) or Ca2+and Mg2+and 5 mM EDTA (grey bars) in
2% HS in 50 mM Tris–HCl and 150 mM NaCl, pH 7n4. After two washes
with the same buffer, cells were stained with Strep–PE. Median
fluorescence was determined. Data shown represent the average of two
separate experiments. Error bars represent SD. Ca2+and Mg2+added after
attachment of b-rHBsAg (white bars) does not reverse binding. (b) Low
pH prevents attachment of b-rHBsAg to monocytes. PBMCs were
incubated with b-rHBsAg in 2% HS–HBSS or 2% HS in 10 mM citrate
buffer and 150 mM NaCl, pH 6 and 5 (black bars). After one wash step
with the same buffer and one wash step with 2 % HS–HBSS, cells were
stained with Strep–PE, washed twice and analysed. Median fluorescence
was determined. Data shown represent the average of two separate
experiments. Error bars represent SD. Pre-treatment of PBMCs (grey bars)
or b-rHBsAg (white bars) with the same buffers does not prevent binding.
BCIF
Downloaded from www.microbiologyresearch.org by
IP: 24.222.136.41
On: Thu, 05 Nov 2015 07:06:13
P. Vanlandschoot and othersP. Vanlandschoot and others
Fig. 6. (a) 1,25-VitD3-differentiated THP-1 cells express CD14 and bind
b-rHBsAg. Non-differentiated THP-1 cells (white bars) or 48 h
differentiated cells (black bars) were incubated with anti-CD14–FITC clone
P9, clone MY4 or b-rHBsAg. Data shown represent the average of two
separate experiments. Error bars represent SD. (b) PMA treatment
reduces binding of b-rHBsAg. Non-treated cells (white bars) and PMA-
treated cells (black bars) were stained with anti-CD14–FITC clone P9,
clone MY4 or b-rHBsAg. Data shown represent the average of two
separate experiments. Error bars represent SD.
recognize different forms of CD14, which may differ in
expression levels on monocytes and monocytic cell lines
(Pedron et al., 1995). Undifferentiated THP-1 cells showed no
detectable expression of CD14 and did not bind b-rHBsAg
(Fig. 6a). 1,25-VitD3-differentiated THP-1 cells expressed
CD14 molecules that were recognized by both antibodies.
These differentiated cells did bind b-rHBsAg (Fig. 6a), which
demonstrates that only monocytes in a certain maturation state
bind rHBsAg.
Fig. 7. rHBsAg reduces LPS-induced IL-1βand TNFαsecretion. 1,25-
VitD3-differentiated THP-1 cells (upper panels) or PBMCs (lower panels)
from donor 1 (black symbols) and donor 2 (white symbols) were
incubated with 0 (5,4), 10 (,)or50(=,>) ng/ml LPS and
different concentrations of rHBsAg. Cell supernatant was collected after
24 h and cytokine concentrations were determined.
Effect of PMA treatment on the attachment of
b-rHBsAg
PMA has several effects on monocytes, like changes in cell
shape, endocytosis and shedding of membrane proteins. PMA
also induces differentiation of (pre)-monocytes into a more
mature state. To monitor PMA-induced changes in monocytes,
anti-CD14 antibodies, clones P9 and My4, were used to detect
changes in CD14 levels; an established feature of CD14 is that
monocytes rapidly shed CD14 when treated with PMA.
Similar to previous reports (Pedron et al., 1995), PMA
treatment removed CD14 molecules recognized by clone P9
almost completely, while a two- to threefold reduction in
CD14 molecules recognized by clone My4 was obtained (Fig.
6b). More importantly, a 50% reduction in attachment of
b-rHBsAg to the PMA-treated monocytes was observed.
Effect of non-biotinylated rHBsAg on the function of
monocytes
Because rHBsAg interacts specifically with monocytes, the
effect of rHBsAg on the function of these cells was investigated
(Fig. 7). Monocytes were activated with LPS or IL-2. First,
THP-1 cells were incubated for 24 h with 100 nM 1,25-VitD3,
washed and incubated either with or without LPS in the
BCIG
Downloaded from www.microbiologyresearch.org by
IP: 24.222.136.41
On: Thu, 05 Nov 2015 07:06:13
rHBsAg binds to monocytesrHBsAg binds to monocytes
Table 1. rHBsAg reduces IL-2-induced IL-8 production
The IL-2-induced increase in levels of IL-8 was determined by first subtracting IL-8 concentrations of non-
stimulated cells from those of stimulated ones. These values were compared to one of the non-rHBsAg-
treated PBMC sample levels, set at 100%.
IL-2
k j IL-2-induced
PBMC donor rHBsAg µg/ml IL-8 pg/ml increase (%)
a 0 2659 4946 100
1 3006 4403 61
10 2307 4095 78
25 2049 3018 42
b 0 4993 16306 100
1 4393 9995 88
10 4750 9710 44
50 2347 6694 38
c 0 13122 25889 100
1 6631 18188 91
10 5280 13187 62
50 4019 14576 83
d 0 906 1428 100
1 941 1419 92
10 997 1296 57
25 999 1168 32
presence of different concentrations of rHBsAg. Culture
supernatant was collected after 24 h of incubation and tested
for the presence of IL-1βand TNFα. rHBsAg alone did not
induce any detectable cytokine production, while LPS induced
the secretion of both cytokines (Fig. 7). High levels of TNFα
were produced, while lower concentrations of IL-1βwere
detected. The highest cytokine levels were obtained with
50 ng\ml LPS. In the presence of rHBsAg, LPS-induced
cytokine production decreased in a dose-dependent manner.
The effect of rHBsAg on LPS-induced cytokine production
(IL-1βand TNFα) was also studied with PBMCs from different
donors. As shown in Fig. 7, PBMCs from donor 1 produced
higher concentrations of TNFαcompared to the PBMCs from
donor 2. Nevertheless, the dose-dependent reduction in TNF
concentrations in the presence of rHBsAg was more pro-
nounced with PBMCs from donor 1. Similar results were
obtained for IL-1β. Contrary to the reduced secretion of these
pro-inflammatory cytokines, levels of an anti-inflammatory
cytokine, IL-10, increased or remained unaffected (data not
shown).
The effect of rHBsAg on IL-2-induced cytokine production
was studied with PBMCs from four donors. As shown in Table
1, non-stimulated PBMCs already produced IL-8, the levels of
which increased by adding IL-2. In the presence of rHBsAg,
IL-8 production of both stimulated and non-stimulated
PBMCs decreased in a dose-dependent manner.
Discussion
Plasma-derived or recombinant HBsAg particles have been
used widely to identify cellular receptors for HBV attachment
and entry. Several receptors have been put forward but, so far,
convincing data to support these claims are lacking. Serum
proteins that bind to HBsAg were identified as well. These
are thought to bridge virus and cellular receptors. Again,
data to support these possibilities are lacking. The possible
modulation by HBsAg of cellular and immunological responses
during HBV infection has received much less attention. It has
been speculated that the large number of HBsAg particles may
induce T cell anergy and may prevent antibody-mediated
neutralization of HBV. In vitro studies have demonstrated a
reduced capacity of PBMCs from chronically infected persons
to produce cytokines upon stimulation with LPS (Muller &
Zielinski, 1990, 1992). Moreover, ten years ago it was shown
that HBsAg can suppress the production by human macro-
phages of different cytokines induced by different agents such
as LPS, vesicular stomatitis virus and granulocyte-macrophage
colony-stimulating factor. A role for endoribonuclease V as a
post-transcriptional inactivator of cytokine transcripts was
demonstrated (Jochum et al., 1990). Taken together, these
results suggest that HBsAg interacts with one or more
receptors on antigen-presenting cells, like monocytes, macro-
phages and dendritic cells. Attachment of HBsAg to mono-
BCIH
Downloaded from www.microbiologyresearch.org by
IP: 24.222.136.41
On: Thu, 05 Nov 2015 07:06:13
P. Vanlandschoot and othersP. Vanlandschoot and others
cytes was reported previously, an interaction that was shown
to occur via pre-S1 (Pontisso et al., 1991 ; Neurath et al., 1992).
Data presented here clearly demonstrate that monocytes
express a receptor, which is recognized by – S protein only –
HBsAg expressed in yeast. Moreover, using THP-1 cells and
PMA-treated monocytes, it is shown that this receptor is
present in a more mature differentiation state only. Binding to
macrophages, obtained by culturing monocytes for 7 days,
was observed as well (data not shown). A protein present in HS
enhances attachment of rHBsAg to the plasma membrane of
monocytes. This protein is rapidly inactivated by incubation at
56 mC, a procedure routinely performed with sera used for
tissue culture. Binding of rHBsAg is partially inhibited by an S
protein-specific MAb (F47B), while F47B-dependent binding
to B cells is observed. When the same experiment was
performed with heat-inactivated serum, attachment of rHBsAg
to monocytes and B cells was observed, but only in the
presence of MAb F47B (data not shown). This observation
suggests that the complete inhibition of binding to monocytes
is masked, probably by the alternative attachment of rHBsAg–
antibody complexes to Fc receptors. The binding of rHBsAg to
B cells most probably results from such interactions as well.
Binding of b-rHBsAg was decreased in the presence of
Ca#+\Mg#+and H+(low pH).
To find a possible biological function for this rHBsAg–
receptor interaction, the effect of rHBsAg on LPS- and IL-2-
induced activation of monocytes was investigated. One of the
most potent activators of monocytes is LPS, which induces the
secretion of several cytokines, such as IL-1β, IL-6, IL-12 and
TNFα. rHBsAg particles themselves did not induce any of
these cytokines, while LPS-induced secretion of IL-1βand
TNFαwas reduced in the presence of rHBsAg. Using
macrophages and plasma-purified HBsAg, identical results for
TNFαhave been reported previously. However, in contrast to
our results, human macrophages produced IL-1βin response to
HBsAg (Jochum et al., 1990). A second potent activator of
monocytes is IL-2, which, among several other activities,
increases the secretion of cytokines like IL-8, IL-6 and TNFα.
As shown previously (Bosco et al., 1997), blood monocytes
already secreted IL-8 when cultured without IL-2. This
production was downregulated by rHBsAg. More importantly,
IL-2-induced IL-8 secretion was reduced in the presence of
rHBsAg.
Viruses have long been viewed as simple genetic parasites
that use the host cellular machinery to propagate themselves.
However, it has become clear that the co-existence of these
pathogens and their hosts have shaped the immune system and
resulted in a surprising diversity of virus strategies to
manipulate different cellular and immune regulatory systems.
Viruses have targeted cellular cytokine production and cyto-
kine receptor-signalling pathways, apoptotic pathways, cell
growth and activation pathways, MHC-restricted antigen
presentation pathways and humoral immune responses (Alcami
et al., 2000; Tortorella et al., 2000). Our results suggest
strongly that monocytes express a receptor that is recognized
by HBsAg. Engagement of this receptor, through interaction
with a serum protein, suppresses the activity of monocytes.
These observations suggest that HBV produces HBsAg in
excess amounts to interfere with the normal function of
antigen-presenting cells.
The authors are indebted to GlaxoSmithKline Biologicals for the
recombinant yeast-derived HBsAg. The authors thank Lia Sillekens and
DiaSorin for the generous gift of MAbs F47B and F9H9, and MAb anti-
d, respectively.
References
Alcami, A. & Koszinowski, U. H. (2000). Viral mechanisms of immune
evasion. Immunology Today 21, 447–455.
Akbar, S. M., Onji, M., Inaba, K., Yamamura, K.-I. & Ohta, Y. (1993).
Low responsiveness of hepatitis B virus-transgenic mice in antibody
response to T-cell-dependent antigen: defect in antigen-presenting
activity of dendritic cells. Immunology 78, 468–475.
Bosco, M. C., Espinoza-Delgado, I., Rowe, T. K., Malabarba, M. G.,
Longo, D. L. & Varesio, L. (1997). Functional role for the myeloid
differentiation antigen CD14 in the activation of human monocytes by
IL-2. Journal of Immunology 159, 2922–2931.
Budkowska, A., Quan, C., Groh, F., Bedossa, P., Dubreuil, P., Bouvet,
J. P. & Pillot, J. (1993). Hepatitis B virus (HBV) binding factor in human
serum: candidate for a soluble form of hepatocyte HBV receptor. Journal
of Virology 67, 4316–4322.
Chisari, F. V. (1995). Hepatitis B virus transgenic mice : insights into the
virus and the disease. Hepatology 22, 1316–1325.
Chisari, F. V. & Ferrari, C. (1995). Hepatitis B virus immunopatho-
genesis. Annual Reviews in Immunology 13, 29–60.
Gagliardi, M. C., Nisini, R., Benvenuto, R., De Petrillo, G., Michel,
M. L. & Barnaba, V. (1994). Soluble transferrin mediates targeting of
hepatitis B envelope antigen to transferrin receptor and its presentation
by activated T cells. European Journal of Immunology 24, 1372–1376.
Ganem, D. (1996). Hepadnaviridae and their replication. In Fields Virology,
3rd edn, pp. 2703–2738. Edited by B. N. Fields, D. M. Knipe & P. M.
Howley. Philadelphia: Lippincott–Raven.
Imai, M., Yanase, Y., Nojiri, T., Miyakawa, Y. & Mayumi, M. (1979). A
receptor for polymerized human and chimpanzee albumins on hepatitis B
virus particles co-occurring with HBeAg. Gastroenterology 76, 242–247.
Jochum, C., Voth, R., Rossol, S., Meyer zum Buschenfelde, K. H., Hess,
G., Will, H., Schroder, H. C., Steffen, R. & Muller, W. E. (1990).
Immunosuppressive function of hepatitis B antigens in vitro : role of
endoribonuclease V as one potential trans inactivator for cytokines in
macrophages and human hepatoma cells. Journal of Virology 64,
1956–1963.
Komai, K. & Peeples, M. E. (1990). Physiology and function of the vero
cell receptor for the hepatitis B virus small S protein. Virology 177,
332–338.
Kurose, K., Akbar, S. M., Yamamoto, K. & Onji, M. (1997). Production
of antibody to hepatitis B surface antigen (anti-HBs) by murine hepatitis
B virus carriers: neonatal tolerance versus antigen presentation by
dendritic cells. Immunology 92, 494–500.
Mehdi, M., Kaplan, M. J., Anlar, F. Y., Yang, X., Bayer, R., Sutherland,
K. & Peeples, M. E. (1994). Hepatitis B virus surface antigen binds to
apolipoprotein H. Journal of Virology 68, 2415–2424.
BCII
Downloaded from www.microbiologyresearch.org by
IP: 24.222.136.41
On: Thu, 05 Nov 2015 07:06:13
rHBsAg binds to monocytesrHBsAg binds to monocytes
Mehdi, H., Yang, X. & Peeples, M. E. (1996). An altered form of
apolipoprotein H binds hepatitis B virus surface antigen most efficiently.
Virology 217, 58–66.
Milich, D. R. (1997). Pathobiology of acute and chronic hepatitis B virus
infection: an introduction. Journal of Viral Hepatitis 4, 25–30.
Muller, C. & Zielinski, C. C. (1990). Impaired lipopolysaccharide-
inducible tumor necrosis factor production in vitro by peripheral blood
monocytes of patients with viral hepatitis. Hepatology 12, 1118–1124.
Muller, C. & Zielinski, C. C. (1992). Interleukin-6 production by
peripheral blood monocytes in patients with chronic liver disease and
acute viral hepatitis. Journal of Hepatology 15, 372–377.
Neurath, A. R., Strick, N. & Sproul, P. (1992). Search for hepatitis B
virus cell receptors reveals binding sites for interleukin 6 on the virus
envelope protein. Journal of Experimental Medicine 175, 461–469.
Paulij, W. P., de Wit, P. L. M., Su
$nnen, C. M. G., van Roosmalen, M. H.,
Petersen-van Ettekoven, A., Cooreman, M. P. & Heijtink, R. A. (1999).
Localization of a unique hepatitis B virus epitope sheds new light on the
structure of hepatitis B virus surface antigen. Journal of General Virology
80, 2121–2126.
Pedron, T., Girard, R. & Chaby, R. (1995). Variation of LPS-binding
capacity, epitope expression, and shedding of membrane-bound CD14
during differentiation of human monocytes. Journal of Immunology 155,
1460–1471.
Pontisso, P., Morsica, G., Ruvoletto, M. G., Zambello, R., Colletta, C.,
Chemello, L. & Alberti, A. (1991). Hepatitis B virus binds to peripheral
blood mononuclear cells via the pre S1 protein. Journal of Hepatology 12,
203–206.
Stirk, H. J., Thornton, J. M. & Howard, C. R. (1992). A topological
model for hepatitis B surface antigen. Intervirology 33, 148–158.
Stricker, E. A. M., Tiebout, R. F., Lelie, P. N. & Zeijlemaker, W. P.
(1985). A human monoclonal IgG1λanti-hepatitis B surface antibody.
Production, properties, and application. Scandinavian Journal of Im-
munology 22, 337–343.
Tortorella, D., Gewurz, B. E., Furman, M. H., Schust, D. J. & Ploegh,
H. L. (2000). Viral subversion of the immune system. Annual Review
of Immunology 18, 861–926.
Received 18 January 2002; Accepted 8 February 2002
BCIJ
