Ligation of Fc gamma receptor IIB inhibits antibody dependent enhancement of dengue virus infection

Prévia do material em texto

Ligation of Fc gamma receptor IIB inhibits
antibody-dependent enhancement of
dengue virus infection
Kuan Rong Chana, Summer Li-Xin Zhangb, Hwee Cheng Tanc, Ying Kai Chanb, Angelia Chowc, Angeline Pei Chiew Limb,
Subhash G. Vasudevanc, Brendon J. Hansonb, and Eng Eong Ooib,c,1
aNational University of Singapore Graduate School, National University of Singapore, Singapore 117456; bDefence Science Organization National
Laboratories, Singapore 117510; and cDuke-National University of Singapore Graduate Medical School, Singapore 169854
Edited* by Diane E. Griffin, The Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, and approved June 8, 2011 (received for review
April 25, 2011)
The interaction of antibodies, dengue virus (DENV), and monocytes
can result in either immunity or enhanced virus infection. These
opposing outcomes of dengue antibodies have hampered dengue
vaccine development. Recent studies have shown that antibodies
neutralize DENV by either preventing virus attachment to cellular
receptors or inhibiting viral fusion intracellularly. However, whether
the antibody blocks attachment or fusion, the resulting immune
complexes are expected to be phagocytosed by Fc gamma receptor
(FcγR)-bearing cells and cleared from circulation. This suggests that
only antibodies that are able to block fusion intracellularly would be
able to neutralize DENV upon FcγR-mediated uptake by monocytes
whereas other antibodies would have resulted in enhancement of
DENV replication. Using convalescent sera from dengue patients, we
observed that neutralization of the homologous serotypes occurred
despite FcγR-mediated uptake. However, FcγR-mediated uptake
appeared to be inhibitedwhen neutralized heterologous DENV sero-
typeswereused instead.Wedemonstrate that this inhibitionoccurred
through the formation of viral aggregates by antibodies in a concen-
tration-dependent manner. Aggregation of viruses enabled antibod-
ies to cross-link the inhibitory FcγRIIB,which is expressed at low levels
but which inhibits FcγR-mediated phagocytosis and hence prevents
antibody-dependent enhancement of DENV infection in monocytes.
Dengue is the most common mosquito-borne viral diseaseglobally. The lack of an effective preventive measure, espe-
cially a licensed vaccine, has resulted in the global spread of this
virus (1, 2). Although neutralizing antibodies can confer lifelong
immunity against reinfection by one of the four dengue virus
(DENV) serotypes, subneutralizing antibody levels or cross-
reactive antibodies appear to enhance the risk of severe dengue in
subsequent infections (3–6). DENV bound with subneutralizing
concentrations of antibody has been shown to result in increased
virus uptake and replication in Fc gamma receptor (FcγR)-
bearing cells such as monocytes/macrophages (4, 7). Thus, de-
fining the determinants for virus neutralization will be important
for the design of an effective dengue vaccine that protects against
all four DENV serotypes while minimizing the risk of antibody-
dependent enhancement of DENV infection.
Neutralization of flavivirus infection is a multiple-hit phe-
nomenon. Recent stoichiometric studies have shown that both
antibody affinity and epitope accessibility are important deter-
minants for virus neutralization (8–10). Antibodies neutralize
DENV by either preventing virus attachment to cellular receptors
(11) or inhibiting viral fusion intracellularly (12). However,
whether the antibody blocks attachment or fusion, the resulting
immune complex is expected to be cleared from the circulation by
professional phagocytes, especially the FcγR-bearing cells. This
suggests that only antibodies that are able to block fusion intra-
cellularly would be able to neutralizeDENVupon FcγR-mediated
uptake by monocytes. We thus set out to examine the early events
of this interaction between the DENV immune complex and
monocytic cells. However, instead, we serendipitously identified a
mechanism that inhibits dengue virus infection where antibodies
aggregate viruses in a concentration-dependent manner, which in
turn allows for cross-linking of Fc gamma receptor IIB (FcγRIIB)
that inhibits uptake of the DENV immune complex.
Results
Convalescent Sera Neutralize Homologous DENV Serotypes at Levels
That Mediate Uptake of Immune Complexes but Neutralize Heterolo-
gous DENV Serotypes at Levels That Inhibit Uptake. To address the
interactions involved in antibody-mediated neutralization in mon-
ocytes, we obtained early convalescent sera from patients with
primary DENV infection. Confirmation of the primary infection
status, along with the identification of the DENV serotype with
which these patients have been infected, was carried out in the
corresponding acute serum sample. The results are shown in Table
S1. Using the plaque reduction neutralization test (PRNT) on
BHK cells, we observed that cross-reactive antibodies were pres-
ent in these early convalescent sera (Table S2), which is consistent
with previous findings (13, 14). These sera were also able to neu-
tralize the four DENV serotypes, albeit at varying titers, when the
FcγR-bearing THP-1 cells were used instead of BHK cells (Fig.
S1). However, using DiD (1, 1′-dioctadecyl-3, 3, 3′,3′-tetrame-
thylindodicarbocyanine, 4-chlorobenzenesulfonate salt)-labeled
DENV (15, 16), we observed distinct differences in the early
events when DENVwas reacted with the highest dilution of serum
that resulted in complete virus neutralization (hereafter referred
to as the DENV immune complex) to THP-1 (Fig. 1A). In a serum
sample that fully neutralized both DENV-1 and -2, uptake of
neutralized immune complexes was observed to be significantly
higher, as measured by flow cytometry, when DENV-2 (homolo-
gous) instead of DENV-1 (heterologous) was used (Fig. 1B).
Immunofluorescence showed that the neutralized DENV-2 im-
mune complexes were colocalized to LAMP-1 compartments (Fig.
1C). In contrast, no subcellular trafficking of the neutralized
DENV-1 immune complexes was observed (Fig. 1C). These
observations were reproduced in a panel of sera from DENV-2
primary patients where uptake of theDENV immune complex was
observed only when DENV-2 but not the other serotypes were
used (Fig. 1D–F). This observation was also not limited toDENV-
2. Uptake of the DENV immune complex was observed only when
Author contributions: K.R.C., S.G.V., and E.E.O. designed research; K.R.C., S.L.-X.Z., H.C.T.,
Y.K.C., and A.C. performed research; A.P.C.L., B.J.H., and E.E.O. contributed new reagents/
analytic tools; K.R.C., S.L.-X.Z., H.C.T., Y.K.C., and E.E.O. analyzed data; and K.R.C., S.G.V.,
B.J.H., and E.E.O. wrote the paper.
The authors declare no conflict of interest.
*This Direct Submission article had a prearranged editor.
Freely available online through the PNAS open access option.
1To whom correspondence should be addressed. E-mail: engeong.ooi@duke-nus.edu.sg.
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
1073/pnas.1106568108/-/DCSupplemental.
www.pnas.org/cgi/doi/10.1073/pnas.1106568108 PNAS | July 26, 2011 | vol. 108 | no. 30 | 12479–12484
M
IC
RO
BI
O
LO
G
Y
convalescent serum samples from patients with primary DENV-1,
DENV-3, or DENV-4 were reacted with the homologous but not
the heterologous serotypes (Fig. S2). These findings indicate that
neutralization of homologous DENV serotypes can occur at levels
that mediate uptake of immune complexes, but neutralization of
heterologous DENV serotypes occurs only at levels where FcγR-
mediated uptake is inhibited.
Antibody Concentration Effects on FcγR-Mediated Uptake of Immune
Complexes. The observation that neutralized viruses were taken up
and trafficked to the late endosome/lysosome compartment is
consistent with the known function of monocytes in removing
immune complexes from circulation. However, the inhibition of
theuptake of immune complexes is intriguing. Neutralization of
the heterologous DENV serotypes appeared to occur at lower
Fig. 1. Convalescent primary DENV-2 human sera neutralize homologous serotypes at levels permissible for internalization but neutralize heterologous
serotypes at levels that inhibit uptake. (A) Summary of method used to investigate early events of neutralized DENV immune complexes in THP-1 cells. (B)
Percentage of internalized DiD-labeled DENV virus (DiD+ cells) in THP-1 cells at 30 min post infection when in complex with convalescent serum (3136) at the
respective neutralizing titers, analyzed by flow cytometry. Data are normalized against cells infected with only virus (without antibodies) to account for
differences in uptake for different DENV serotypes. (C) Fate of neutralized immune complexes when 3136 is in complex with DENV-2 or DENV-1. LAMP-1 is
green, DiD-labeled DENV is blue, and h3H5 is red. (Scale bar, 7.5 μm.) (D–F) Same analysis as depicted in B and C, but with three other convalescent primary
DENV-2 sera (6583, 3111, 3598). Neutralization titers < 10 were not considered for this analysis. Data are represented as mean ± SD. **P < 0.01.
Fig. 2. Increasing antibody concentration inhibits im-
mune complex internalization by THP-1 cells. (A) Per-
centage of internalized DiD-labeled DENV-2 virus (DiD+
cells) in THP-1 cells at 30 min post uptake when com-
plexed with different neutralizing dilutions of convales-
cent DENV-2 serum, analyzed by flow cytometry. Dashed
line indicates DiD+ cells in the presence of DiD-labeled
DENV-2 only. (B) Subcellular localization of DENV-2,
DENV-2 in complex with 1:8 serum, and DENV-2 in com-
plex with undiluted serum in THP-1 cells. LAMP-1 is la-
beled green, DiD-labeled DENV-2 blue and h3H5 red. (C)
Same as A, but with various neutralization concentrations
of h3H5. (D) Same as B, but with DENV-2 only, DENV-2 in
complex with 1.56 μg/mL h3H5, or DENV-2 in complex
with 400 μg/mL h3H5. (E) Subcellular localization of
Alexa594-labeled DENV-2 in complex with 400 μg/mL
h3H5. LAMP-1 is green, Alexa594-labeled DENV-2 is red,
and h3H5 is blue. Data are represented as mean ± SD.
(Scale bar, 7.5 μm.)
12480 | www.pnas.org/cgi/doi/10.1073/pnas.1106568108 Chan et al.
dilutions of the convalescent sera than that needed for the homol-
ogous serotype (Fig. 1). This observation suggests that inhibition
of FcγR-mediated uptake is affected by antibody concentration.
However, early convalescent sera also contain IgM antibodies
that could complex DENV without interacting with FcγRs. To
address this potential interaction, we titrated a serum from a vol-
unteer who was infected with DENV-2 over 30 y ago, reacted
this to DENV-2, and determined its fate in THP-1. The results
showed that lower serum dilution (Fig. S3) resulted in a similar
reduction of DiD-labeled DENV-2 uptake whereas increasing
serum dilutions resulted in FcγR-mediated uptake of the immune
complex without DENV replication (Fig. 2 A and B).
To reduce variability resulting from the use of different sera,
we investigated the mechanism for the observed concentration-
dependent effects using the monoclonal antibody (mAb) 3H5,
which is specific against DENV-2 (17, 18). Because FcγR en-
gagement is required for FcγR-mediated phagocytosis of immune
complexes (19), we generated a mouse–human chimeric antibody
of 3H5 (h3H5) consisting of mouse VH and VL sequences and
human γ1 and κ constant sequences (20). These antibodies were
indistinguishable from the parent 3H5mAb in their ability to bind
to DENV-2 (Fig. S4A). Complete DENV-2 neutralization was
observed from 1.56 to 400 μg/mL of h3H5. Subneutralizing con-
centrations of h3H5 enhanced viral infection to a greater extent
after humanization into IgG1 but not IgG4 (Fig. S4B), indicating
that specific interactions with human FcγR were attained (21).
Using h3H5, we observed that increasing levels of antibody con-
centration along the range that fully neutralized DENV-2 resul-
ted in the reduction of DiD+ cells (Fig. 2C). Likewise, as observed
with convalescent sera, the immune complexes were trafficked to
LAMP-1 compartments at 30 min post infection although in-
creasing the h3H5 concentration to 400 μg/mL resulted in in-
hibition of uptake of the immune complexes (Fig. 2D). However,
because the DiD signal is quenched before fusion with cellular
membranes (16), it is not possible to visualize DENV outside of
the cell. To overcome this limitation, we labeled DENV with
Alexa Fluor (22) and demonstrated the presence of antibody-
bound viruses clustered outside THP-1 cells (Fig. 2E). Similar
findings were also made when primary monocytes were used in-
stead of THP-1 cells (Fig. S5). Overall, these results indicate that
immune complexes formed with neutralizing antibody can be
rapidly internalized via the FcγR but this process is inhibited with
increasing levels of antibody.
Size of DENV Immune Complex Is Dependent on the Concentration of
Antibody. One possible explanation for the antibody concentra-
tion-dependent inhibition of the immune complex uptake is that
excess antibodies competed with the immune complexes for
limited FcγR. To test this possibility, DENV complexed with
h3H5 at 400 μg/mL was compared with that at 1.56 μg/mL but
with an addition of isotype control antibodies to give a total an-
tibody concentration of 400 μg/mL The addition of isotype con-
trol antibodies did not inhibit internalization of DENV-2 (Fig. 3
A and B), indicating that inhibition of FcγR internalization can-
not be explained by competition for limited receptors by free
antibodies. Instead, increasing antibody concentration could
have cross-linked DENV, resulting in the formation of viral
aggregates. To test this hypothesis, we used a sucrose density
gradient consisting of layers extending from 60% sucrose to 10%
sucrose in 10% increments to separate immune complexes of
different sizes (23). Equal volume fractions were removed from
the bottom of each tube for quantitative PCR analysis to detect
for DENV RNA; immune complexes in the earlier fractions
would thus have greater density (Fig. 3C). The size of aggregates
in fractions that showed peak viral RNA copy numbers was de-
termined using dynamic light scattering (Fig. 3D). The average
diameter of DENV-2 in the experiment was 51.9 nm, which is
consistent with previous observations that showed that DENV is
∼50 nm in diameter (24). Reacting 33.3 μg/mL of the Fab frag-
ments of 3H5 Fab or 3 μg/mL of h3H5 with DENV-2 did not
result in significantly different particle sizes. However, with 100
μg/mL of h3H5, peak DENV RNA copies shifted to fraction 9
(Fig. 3C), and this corresponded to an increased immune complex
size of 148.2 nm in diameter (Fig. 3D). A similar increase in im-
mune complex size was also observed with DENV-2 convalescent
serum from Fig. 2A. When undiluted serum was used, the peak
DENV RNA copies shifted significantly to the smaller fractions
(Fig. 3E), which corresponded with an increased immune complex
size of 182.3 nm in diameter (Fig. 3F). Taken together, these
findings support the notion that larger aggregates were formed
with increasing antibody concentrations.
Fig. 3. Inhibition of immune-complex internalization is not due to FcγR
competition but due to increased immune-complex size. (A) Percentage of
DiD+ cells at 30 min post uptake when in complex with 400 μg/mL h3H5 or
1.56 μg/mL h3H5 with addition of 398.44 μg/mL of human isotype control. (B)
Subcellular localization of DENV-2 complexed with 1.56 μg/mL h3H5 and with
the addition of 398.44 μg/mL human IgG1 isotype control. LAMP-1 is green,
DiD-labeled DENV-2 is blue, and human antibodies are red. (Scale bar, 7.5
μm.) (C) Proportion of total viral RNA extracted from the various sucrose
fractions. Proportion of viral RNA in each fraction was determined by dividing
the viral RNA copy number in that fraction by the viral RNA copynumber in
the entire gradient. Shown are the moving averages of free virus and 33.3-μg/
mL Fab fragments of h3H5 or virus in complex with 3 μg/mL h3H5 or with 100
μg/mL h3H5 using qPCR. (D) Diameter of the immune complexes measured
using dynamic light scattering in the respective sucrose fractions with peak
viral RNA copy number. (E and F) Similar to C and D, but with 1:10 or un-
diluted serum. Data are represented as mean ± SD. **P < 0.01.
Chan et al. PNAS | July 26, 2011 | vol. 108 | no. 30 | 12481
M
IC
RO
BI
O
LO
G
Y
Aggregation of DENV Enables Antibodies to Cross-Link the Inhibitory
FcγRIIB. Given the increase in size when higher concentrations of
antibodies were reacted with DENV, it is possible that the re-
sultant larger viral aggregates enable antibodies to cross-link
inhibitory FcγR that is expressed at lower levels on the cell
membrane. One such candidate is FcγRIIB, which is known to
exert an inhibitory effect on FcγR-mediated phagocytosis (25, 26).
Because cross-linking of FcγRIIB has been previously shown to
result in SHP-1 phosphorylation that down-regulates phagocytosis
(27), we tested for SHP-1 phosphorylation when DENV-2 was
complexed with 1.56 or 100 μg/mL of h3H5. Interestingly, in-
creased phosphorylation was observed when DENV-2 was com-
plexed with 100 μg/mL h3H5 (Fig. 4A). To confirm that FcγRIIB is
functionally involved in the inhibition of immune complex uptake,
we knocked down its expression in THP-1 using siRNA. Re-
duction of FcγRIIB but not FcγRI or FcγRIIA (Fig. 4B) resulted
in significantly increased uptake of viral aggregates even with high
h3H5 concentration (Fig. 4C). DENV-2 remained neutralized
despite higher levels of uptake in the FcγRIIB knockdown cells
(Fig. S6). Conversely, overexpression of FcγRIIB in THP-1 (Fig.
4D) resulted in reduced uptake of DENV immune complexes
across all neutralizing antibody concentrations (Fig. 4E) and lower
levels of infection when subneutralizing concentrations of h3H5
were used (Fig. 4F), compared with the mock-transfected cells.
Taken collectively, our data indicate that FcγRIIB is important in
the inhibition of the uptake of larger viral aggregates.
Discussion
How the neutralizing antibody–virus complex interacts with
monocytes is poorly understood, even though it represents an
important piece of the puzzle of how dengue pathogenesis and
immunity interact. Previous investigations into the fate of the
neutralizing antibody–virus complexes have made use of kidney
cell lines, such as LLC-MK2, Vero, and BHK-1 cells (28). These
studies have shown that antibodies neutralize either by blocking
viral fusion following uptake or by viral attachment (29). These
cells, however, neither express FcγR naturally nor are primary
targets of dengue virus in human infections. Monocytes, on the
Fig. 4. FcγRIIB is involved in the inhibition of immune-complex internalization of larger viral aggregates. (A) THP-1 cells exposed to media (mock), DENV-2 in
complex with 1.56 μg/mL h3H5, and DENV-2 complexed with 100 μg/mL h3H5 30 min post infection. Cell lysates were immunoblotted with anti-SHP-1 and
anti-phospho-SHP-1 (p-SHP-1). (B) THP-1 cells transfected with a control siRNA or siRNA against FcγRIIB. Cell lysates were immunoblotted with anti-FcγRIIB,
anti-FcγRI, anti-FcγRIIA, and LAMP-1 antibodies. LAMP-1 served as a loading control. (C) Percentage of internalized DiD-labeled DENV-2 in THP-1 cells
transfected with control siRNA or siRNA against FcγRIIB when in complex with various h3H5 concentrations. Dashed line indicates DiD+ cells in the presence of
DiD-labeled DENV-2 only, without antibodies. (D) THP-1 cells subjected to mock transfection or transfected with FcγRIIB DNA. Cell lysates were immuno-
blotted with anti-FcγRIIB, anti-FcγRI, anti-FcγRIIA, and LAMP-1 antibodies (loading control). (E) Same as C, but using THP-1 cells with mock transfection or with
transfected FcγRIIB DNA. (F) Plaque titers at 72 h post infection on cells with mock transfection or with overexpression of FcγRIIB. Dashed line indicates plaque
titers in the presence of DENV-2 only, without h3H5. No significant differences between control and treated cells were observed with virus-only infection.
Data are represented as mean ± SD.
12482 | www.pnas.org/cgi/doi/10.1073/pnas.1106568108 Chan et al.
other hand, play a central role in DENV replication (30) as well
as in the removal of the antibody–virus complex in vivo (31–33).
We have explored the initial events following the introduction
of the DENV immune complex to THP-1 monocytic cells using
early convalescent sera. On average, these convalescent sera
were obtained 22 d post illness onset. Antibodies at this early
convalescent period have been shown to have a broad cross-
neutralization activity against the DENV serotypes that only be-
came more serotype-specific with time (13, 14). Using such an
approach, we observed two distinct patterns of virus neutraliza-
tion. At the highest antibody titers that produced 100% DENV
neutralization, homologous DENV serotypes were internalized
by THP-1 cells and trafficked to the late endosome/lysosome
whereas heterologous DENV serotypes were neutralized only at
titers where FcγR-mediated uptake was inhibited.
Although the uptake of neutralized DENV was expected, the
inhibition of FcγR-mediated uptake of DENV or other viral
immune complexes has not been previously described as a means
of viral neutralization. Our data indicate that, in addition to
blocking attachment or fusion, antibodies can also neutralize by
forming large viral aggregates, which in turn cross-links FcγRIIB
to inhibit phagocytosis. This was observed with both h3H5 and
human polyclonal antibodies in a convalescent serum sample.
The aggregated viruses allow the resulting immune complex to be
sufficiently large to cross-link FcγRIIB.
Unlike the receptors FcγRI and FcγRIIA, which contain the
immunoreceptor tyrosine-based activation motif that activates
phagocytosis, FcγRIIB is an inhibitory receptor with the immu-
noreceptor tyrosine-based inhibitory motif, which, when engaged,
can activate phosphatases such as SHP-1 to down-regulate
phagocytosis (34). FcγRIIB’s role in the immune response to in-
fection has not been studied extensively. Most work has involved
Streptococcus pneumoniae, a Gram-positive bacterium that is
a major cause of pneumonia. In these studies, FcγRIIB-deficient
mice have been shown to have improved FcγR-mediated phago-
cytosis of the antibody-bound bacterium (35). Likewise, FcγRIIB
deficiency is associated with an increased resistance to Staphylo-
coccus aureus (36) and Mycobacterium tuberculosis (37) infection
in mice. The only study that has examined the role of FcγRIIB
in response to viral infection showed that FcγRIIB-mediated in-
hibition of phagocytosis by dendritic cells was associated with re-
duced antibody and T-cell responses to human papilloma virus-like
particles (38). The involvement of FcγRIIB in virus neutralization
is thus unique and shows that ligation of FcγRIIB impairs phago-
cytosis and antibody-dependent enhancement in monocytes. Our
observations may also explain previously reported observations of
peak antibody titers following acute influenza or dengue infections
being associated with transient reduction in the phagocytic activity
of macrophages in experimental mouse models (39, 40).
That cross-linking of FcγRIIB can inhibit antibody-dependent
enhancement of DENV infection would suggest that, as long as
an antibody can bind to an epitope in a manner that allows for the
formation of concentration-dependent viral aggregates, DENV
can be neutralized in vivo. If immunity were to be mediated by
this mechanism alone, then antibodies would protect against all
four DENV serotypes. However, epidemiological observations
indicate that lifelong immunity is specific to the serotype with
which a person has been infected. This suggests insteadthat, for
long-lasting immunity, antibodies must be able to prevent in-
fection even at levels that do not form viral aggregates that cross-
link FcγRIIB, a scenario expected as antibody levels wane fol-
lowing acute infection. In contrast, inhibition of theh uptake of
heterologous serotypes by the early convalescent sera from
patients with primary DENV infection was observed only at low
serum dilutions. This process may be the reason why a person can
be transiently immune to the remaining three serotypes of DENV
following an acute primary DENV infection (41, 42). Further
studies, however, would be needed to establish this notion.
Our demonstration that antibodies aggregate viruses also
raises several unexplored questions. In addition to interacting
with FcγRs directly, immune complexes can also bind to com-
plement components and interact with complement receptor 1
(CR1) on red blood cells in vivo (43–45). The determinants of the
outcome of the simultaneous interaction between DENV im-
mune complexes and the CR1 receptor along with the various
types of FcγR on macrophages in the liver and spleen are un-
known. Even if phagocytosis was not inhibited in this interaction,
it would be interesting to determine if aggregation of DENV
could sufficiently reduce the stoichiometric requirement for virus
neutralization by reducing the epitopes available on each virion to
interact with cellular receptors. In addition, we observed that only
a subset of monocytes can uptake DENV, which is consistent with
previous reports indicating that only a minor population of
monocytes can uptake and replicate DENV (46, 47). These
observations suggest that additional factors may influence the
uptake of the DENV immune complex.
In conclusion, aggregation of DENV by antibodies that results
in the engagement of FcγRIIB is a mechanism that inhibits
DENV infection of monocytes.
Materials and Methods
Antibodies and Human Sera. 3H5 chimeric human/mouse IgG1 and IgG4
antibodies were constructed as previously described (20). Human sera were
obtained from early dengue infection and control (EDEN) study as previously
described (48).
Cells. BHK-21, C6/36, THP-1, and Vero cells were purchased from the American
Type Culture Collection (ATCC) and cultured according to ATCC recom-
mendation. Primary monocytes were isolated from the principal investigator
and cultured as described in SI Materials and Methods.
Virus Stock. DENV-1 (07K3640DK1) and DENV-3 (05K863DK1) are clinical
isolates obtained from the EDEN study (36). DENV-2 (ST) is a clinical isolate
from the Singapore General Hospital, and DENV-4 (H241) was obtained from
ATCC. Viruses were propagated in the Vero cell line and harvested 96 h post
infection and purified through 30% sucrose. Virus pellets resuspended in
5 mM Hepes, 150 mM NaCl, and 0.1 mM EDTA (HNE) buffer were stored
at −80 °C until use. Infectious titer was determined by plaque assay.
Virus Infection in THP-1 Cells. Serial twofold dilutions of h3H5 or human sera
were incubated for 1 h at 37 °C before being added to THP-1 cells at
a multiplicity of infection (moi) of 10. At 72 h after infection, the culture was
clarified by centrifugation, and the infectious titer of dengue virus in the
culture supernatant was quantified with plaque assay. The antibody dilution
required to mediate full virus neutralization was then determined.
Fluorescent Labeling of Virus. The method for DiD labeling and Alexa594
labeling of DENV was as previously described (22) and as detailed in SI
Materials and Methods.
Virus Internalization in THP-1.Neutralizing concentrations ofh3H5/human sera
were incubatedwithDiD-labeled DENV or AF584-DENV for 1 h at 37 °C before
adding to THP-1 cells (moi 10). Cells were then subjected to 20 min of syn-
chronization on ice, followed by 30 min of infection at 37 °C, and then fixed
with paraformaldehyde. Flow cytometry was used to determine the per-
centage of cells with internalized virus complexes, and confocal immuno-
fluorescence was used to determine localization of antibody–virus complexes
in the cell. Detailed description is provided in SI Materials and Methods.
Sucrose Gradient Analysis of DENV Immune Complex Sizes. Sucrose gradient
was formed by careful layering of 10–60 sucrose solutions (in HNE buffer) in
10% increments, starting with the densest at the bottom, in 13.5-mL Ultra-
Clear tubes (Beckman Coulter). The gradient was allowed to linearize over-
night at 4 °C. Equal amounts of purified DENV were incubated with human-
ized antibodies at 3 or 100 μg/mL h3H5 IgG or h3H5 Fab (h3H5 enzymatically
digested to give only one arm of the Fab fragment) for 1 h at 37 °C. The
samples were then carefully layered on top of the linearized 10–60% sucrose
gradient and centrifuged overnight at 25,000 × g at 4 °C in a SW41Ti rotor for
17 h. Each gradient was then harvested in 0.25-mL fractions from the bottom
of the tube. Subsequently, viral RNA was extracted from each fraction and
Chan et al. PNAS | July 26, 2011 | vol. 108 | no. 30 | 12483
M
IC
RO
BI
O
LO
G
Y
quantified by real-time PCR. The proportion of virus present in each fraction
was then tabulated by dividing it to the total viral RNA loaded in these frac-
tions. Each fraction was then plotted as an average with two neighboring
fractions to obtain smoother curves.
Dynamic Light Scattering of Peak Sucrose Fractions. The sucrose fraction that
showed the highest virus genome copy number was identified as the peak
fraction and subjected to dynamic light scattering to determine the DENV
immune complex diameter. Twenty microliters of each fraction was loaded
into a Quartz cuvettete for analysis by Zetasizer Nano S machine (Malvern) at
37 °C using 50% sucrose in HNE buffer as the dispersant. Data were then
analyzed using Zetasizer Nano software version 6.01. The diameter reports
generated are an average of more than 10 readings.
SHP-1 Phosphorylation in THP-1 Cells. DENV-2 was incubated with 1.56 or 100
μg/mL h3H5 and added to THP-1 cells for 30 min at 37 °C. For mock infection,
only RPMI media was added. Cells were then extracted and lysed before
performing Western blot as described in SI Materials and Methods.
siRNA Transfection into THP-1 Cells. Human FcγRIIB siRNA (Qiagen) and All-
Stars negative control siRNA (Qiagen) duplexes (50 nM) were incubated
with DharmaFect2 (Dharmacon) in serum-free media for 20 min and then
added to cells at a density of 2 × 105 cells/mL After 6 h incubation, cells were
replaced with RPMI growth media for 2 d to allow recovery. This was fol-
lowed by a second round of siRNA transfection. Knockdown efficiency was
determined by Western blot as described in SI Materials and Methods.
FcγRIIB Transfection of THP-1. FcγRIIB cDNA was purchased from Origene and
transfected using Lipofectamine LTX and Plus reagent (Invitrogen) in ac-
cordance with the manufacturer’s instructions. Cells were subjected to two
rounds of transfection, and transfection efficiency was determined by
Western blot as described in SI Materials and Methods.
Statistical Analysis. Two-tailed unpaired Student’s t testwas used to determine
if the difference in themeanobservedwas statistically significant (P< 0.05). All
calculations were done using GraphPad Prism v5.0 (GraphPad Software Inc.).
ACKNOWLEDGMENTS. We thank Soman Abraham, Duane Gubler, and
October Sessions for their constructive comments; Gayathri Manokaran for
her technical assistance; and the anonymous reviewer who suggested the
mechanism of inhibition of immune complex uptake by monocytes for us to
pursue. We also thank our colleagues in the early dengue infection and
control (EDEN) study from whom the serum samples were obtained. This
work was funded by the start-up funds from Duke-NUS, as well as by the
National Medical Research Council, Singapore (NMRC/TCR/005/2008 and
NMRC/CSA/025/2010).
1. Kyle JL, Harris E (2008) Global spread and persistenceof dengue. Annu Rev Microbiol
62:71–92.
2. Wilder-Smith A, Ooi EE, Vasudevan SG, Gubler DJ (2010) Update on dengue: Epide-
miology, virus evolution, antiviral drugs, and vaccine development. Curr Infect Dis
Rep 12:157–164.
3. Simmons CP, et al. (2007) Maternal antibody and viral factors in the pathogenesis of
dengue virus in infants. J Infect Dis 196:416–424.
4. Halstead SB, O’Rourke EJ (1977) Dengue viruses and mononuclear phagocytes. I. In-
fection enhancement by non-neutralizing antibody. J Exp Med 146:201–217.
5. Halstead SB (1988) Pathogenesis of dengue: Challenges to molecular biology. Science
239:476–481.
6. Kliks SC, Nimmanitya S, Nisalak A, Burke DS (1988) Evidence that maternal dengue
antibodies are important in the development of dengue hemorrhagic fever in infants.
Am J Trop Med Hyg 38:411–419.
7. Halstead SB (2007) Dengue. Lancet 370:1644–1652.
8. Pierson TC, et al. (2007) The stoichiometry of antibody-mediated neutralization and
enhancement of West Nile virus infection. Cell Host Microbe 1:135–145.
9. Pierson TC, Diamond MS (2008) Molecular mechanisms of antibody-mediated neu-
tralisation of flavivirus infection. Expert Rev Mol Med 10:e12.
10. Pierson TC, Fremont DH, Kuhn RJ, Diamond MS (2008) Structural insights into the
mechanisms of antibody-mediated neutralization of flavivirus infection: Implications
for vaccine development. Cell Host Microbe 4:229–238.
11. Crill WD, Roehrig JT (2001) Monoclonal antibodies that bind to domain III of dengue
virus E glycoprotein are the most efficient blockers of virus adsorption to Vero cells.
J Virol 75:7769–7773.
12. Gollins SW, Porterfield JS (1986) A new mechanism for the neutralization of envel-
oped viruses by antiviral antibody. Nature 321:244–246.
13. Beltramello M, et al. (2010) The human immune response to Dengue virus is domi-
nated by highly cross-reactive antibodies endowed with neutralizing and enhancing
activity. Cell Host Microbe 8:271–283.
14. Dejnirattisai W, et al. (2010) Cross-reacting antibodies enhance dengue virus infection
in humans. Science 328:745–748.
15. van der Schaar HM, et al. (2008) Dissecting the cell entry pathway of dengue virus by
single-particle tracking in living cells. PLoS Pathog 4:e1000244.
16. van der Schaar HM, et al. (2007) Characterization of the early events in dengue virus
cell entry by biochemical assays and single-virus tracking. J Virol 81:12019–12028.
17. Sukupolvi-Petty S, et al. (2007) Type- and subcomplex-specific neutralizing antibodies
against domain III of dengue virus type 2 envelope protein recognize adjacent epit-
opes. J Virol 81:12816–12826.
18. Gentry MK, Henchal EA, McCown JM, Brandt WE, Dalrymple JM (1982) Identification
of distinct antigenic determinants on dengue-2 virus using monoclonal antibodies.
Am J Trop Med Hyg 31:548–555.
19. Goncalvez AP, Engle RE, St. Claire M, Purcell RH, Lai CJ (2007) Monoclonal antibody-
mediated enhancement of dengue virus infection in vitro and in vivo and strategies
for prevention. Proc Natl Acad Sci USA 104:9422–9427.
20. Hanson BJ, et al. (2006) Passive immunoprophylaxis and therapy with humanized
monoclonal antibody specific for influenza A H5 hemagglutinin in mice. Respir Res 7:126.
21. Bruhns P, et al. (2009) Specificity and affinity of human Fcgamma receptors and their
polymorphic variants for human IgG subclasses. Blood 113:3716–3725.
22. Zhang SL, Tan HC, Hanson BJ, Ooi EE (2010) A simple method for Alexa Fluor dye
labelling of dengue virus. J Virol Methods 167:172–177.
23. Rødahl E, Iversen OJ, Dalen AB (1984) Preparative isolation of immune complexes
from serum by sucrose gradient ultracentrifugation. Scand J Immunol 20:21–26.
24. Smith TJ, Brandt WE, Swanson JL, McCown JM, Buescher EL (1970) Physical and bi-
ological properties of dengue-2 virus and associated antigens. J Virol 5:524–532.
25. Tridandapani S, et al. (2002) Regulated expression and inhibitory function of
Fcgamma RIIb in human monocytic cells. J Biol Chem 277:5082–5089.
26. Hunter S, et al. (1998) Inhibition of Fcgamma receptor-mediated phagocytosis by
a nonphagocytic Fcgamma receptor. Blood 91:1762–1768.
27. Huang ZY, Hunter S, Kim MK, Indik ZK, Schreiber AD (2003) The effect of phospha-
tases SHP-1 and SHIP-1 on signaling by the ITIM- and ITAM-containing Fcgamma re-
ceptors FcgammaRIIB and FcgammaRIIA. J Leukoc Biol 73:823–829.
28. Roehrig JT, Hombach J, Barrett AD (2008) Guidelines for plaque-reduction neutrali-
zation testing of human antibodies to dengue viruses. Viral Immunol 21:123–132.
29. van der Schaar HM, Wilschut JC, Smit JM (2009) Role of antibodies in controlling
dengue virus infection. Immunobiology 214:613–629.
30. Kou Z, et al. (2008) Monocytes, but not T or B cells, are the principal target cells for
dengue virus (DV) infection among human peripheral blood mononuclear cells. J Med
Virol 80:134–146.
31. Sun SF, et al. (2008) Stronger in vitro phagocytosis by monocytes-macrophages is
indicative of greater pathogen clearance and antibody levels in vivo. Poult Sci 87:
1725–1733.
32. Huber VC, Lynch JM, Bucher DJ, Le J, Metzger DW (2001) Fc receptor-mediated
phagocytosis makes a significant contribution to clearance of influenza virus in-
fections. J Immunol 166:7381–7388.
33. Van Strijp JA, et al. (1990) Degradation of herpes simplex virions by human poly-
morphonuclear leukocytes and monocytes. J Gen Virol 71:1205–1209.
34. Smith KG, Clatworthy MR (2010) FcgammaRIIB in autoimmunity and infection: Evo-
lutionary and therapeutic implications. Nat Rev Immunol 10:328–343.
35. Clatworthy MR, Smith KG (2004) FcgammaRIIb balances efficient pathogen clearance
and the cytokine-mediated consequences of sepsis. J Exp Med 199:717–723.
36. Gjertsson I, Kleinau S, Tarkowski A (2002) The impact of Fcgamma receptors on
Staphylococcus aureus infection. Microb Pathog 33:145–152.
37. Maglione PJ, Xu J, Casadevall A, Chan J (2008) Fc gamma receptors regulate immune
activation and susceptibility during Mycobacterium tuberculosis infection. J Immunol
180:3329–3338.
38. Da Silva DM, Fausch SC, Verbeek JS, Kast WM (2007) Uptake of human papillomavirus
virus-like particles by dendritic cells is mediated by Fcgamma receptors and contrib-
utes to acquisition of T cell immunity. J Immunol 178:7587–7597.
39. Astry CL, Jakab GJ (1984) Influenza virus-induced immune complexes suppress alve-
olar macrophage phagocytosis. J Virol 50:287–292.
40. Chaturvedi UC, Nagar R, Mathur A (1983) Effect of dengue virus infection on Fc-
receptor functions of mouse macrophages. J Gen Virol 64:2399–2407.
41. Sabin AB (1952) Research on dengue during World War II. Am J Trop Med Hyg 1:30–50.
42. Imrie A, et al. (2007) Antibody to dengue 1 detected more than 60 years after in-
fection. Viral Immunol 20:672–675.
43. Fearon DT (1979) Regulation of the amplification C3 convertase of human comple-
ment by an inhibitory protein isolated from human erythrocyte membrane. Proc Natl
Acad Sci USA 76:5867–5871.
44. Frank MM, Hamburger MI, Lawley TJ, Kimberly RP, Plotz PH (1979) Defective re-
ticuloendothelial system Fc-receptor function in systemic lupus erythematosus. N Engl
J Med 300:518–523.
45. Cornacoff JB, et al. (1983) Primate erythrocyte-immune complex-clearing mechanism.
J Clin Invest 71:236–247.
46. Tassaneetrithep B, et al. (2003) DC-SIGN (CD209) mediates dengue virus infection of
human dendritic cells. J Exp Med 197:823–829.
47. Kou Z, et al. (2011) Human antibodies against dengue enhance dengue viral in-
fectivity without suppressing type I interferon secretion in primary human mono-
cytes. Virology 410:240–247.
48. Low JG, et al. (2006) Early dengue infection and outcome study (EDEN): Study design
and preliminary findings. Ann Acad Med Singapore 35:783–789.
12484 | www.pnas.org/cgi/doi/10.1073/pnas.1106568108 Chan et al.