In Vitro Derived Dendritic Cells
trans-Infect CD4 T Cells Primarily
with Surface-Bound HIV-1 Virions
Marielle Cavrois1, Jason Neidleman1, Jason F. Kreisberg1,2, Warner C. Greene1,2,3*
1 Gladstone Institute of Virology and Immunology, University of California San Francisco, San Francisco, California, United States of America, 2 Department of Medicine,
University of California San Francisco, San Francisco, California, United States of America, 3 Department of Microbiology and Immunology, University of California San
Francisco, San Francisco, California, United States of America
In the prevailing model of HIV-1 trans-infection, dendritic cells (DCs) capture and internalize intact virions and transfer
these virions to interacting T cells at the virological synapse. Here, we show that HIV-1 virions transmitted in trans from
in vitro derived DCs to T cells principally originate from the surface of DCs. Selective neutralization of surface-bound
virions abrogated trans-infection by monocyte-derived DCs and CD34-derived Langerhans cells. Under conditions
mimicking antigen recognition by the interacting T cells, most transferred virions still derived from the cell surface,
although a few were transferred from an internal compartment. Our findings suggest that attachment inhibitors could
neutralize trans-infection of T cells by DCs in vivo.
Citation: Cavrois M, Neidleman J, Kreisberg JF, Greene WC (2007) In vitro derived dendritic cells trans-infect CD4 T cells primarily with surface-bound HIV-1 virions. PLoS
Pathog 3(1): e4. doi:10.1371/journal.ppat.0030004
To ensure their survival, microbial pathogens have evolved
strategies to subvert the action of cellular components of the
host immune system, including dendritic cells (DCs). DCs
patrol peripheral mucosal sites, capturing and processing
potential pathogens into antigenic peptides for presentation
by major histocompatibility complex (MHC) class II to CD4 T
cells in lymphoid organs, initiating an immune response (for
a review, see ). HIV-1 has been proposed to usurp this
natural function of DCs to spread efficiently. HIV-1 entering
the body via the mucosa and other peripheral sites may be
transported by DCs to CD4 T cells deeper in the mucosa or in
lymphoid organs [2–4]. HIV-1 that reaches lymphoid organs
can also take advantage of the formation of DC–T-cell
conjugates to promote its replication and spread [5–7].
DCs can transmit HIV-1 to T cells via two pathways. In the
de novo pathway, DCs are actively infected with HIV, leading
to the budding and spread of new virions to neighboring CD4
T cells. In the prevailing model of the trans pathway, intact
HIV-1 virions are captured by alternative HIV-1 receptors,
which bind virions without triggering fusion, and internalized
into clustered compartments resembling late endosome/
multivesicular bodies (MVBs) [8,9]. After interacting with a
CD4 T cell, HIV-1–loaded DCs redistribute the virion-
containing vesicles to the virological synapse [8,10,11]; CD4,
CXCR4, and CCR5 receptors on T cells are recruited to this
region, facilitating trans-infection .
How HIV-1 virions survive the uptake pathway designed to
capture and cleave pathogens into peptides for antigen
presentation remains unknown. HIV-1 could divert the
intracellular trafficking of immunological synapse compo-
nents to avoid degradation and thus survive until later
transmitted to T cells. Alternatively, external virions could be
transmitted to CD4 T cells since some HIV-1 virions remain
deeply tangled in membrane protrusions and microvilli of the
plasma membrane . Using functional assays that detect
virion fusion and productive infection of CD4 T cells, we
investigated whether trans-infection is mediated through
internalized or external HIV-1 virions in monocyte-derived
DCs (MDDCs) and CD34-derived Langerhans cells (LCs).
Mature MDDCs Transmit HIV-1 to T Cells Primarily via the
trans Pathway while Immature MDDCs Preferentially
Transmit R5-Tropic HIV-1 via the De Novo Pathway
The potential effects of the state of DC maturation and
coreceptor utilization by HIV virions in the trans and the de
novo pathways in HIV-1 transmission from DCs to T cells
were evaluated. These studies were performed with immature
MDDCs or MDDCs matured with tumor necrosis factor a and
poly(I:C)  and with two laboratory-adapted viral strains,
CXCR4-tropic NL4–3 and CCR5-tropic 81A (Figure 1).
MDDCs were incubated with virions at 4 8C to promote viral
binding and were then either added to autologous activated T
cells immediately or incubated for 1 to 5 d at 37 8C before
mixing the autologous T cells. The MDDCs were then
incubated with T cells for 24 h to allow virion transfer to T
Editor: Michael H. Malim, King’s College London, United Kingdom
Received September 20, 2006; Accepted November 29, 2006; Published January
Copyright: ? 2007 Cavrois et al. This is an open-access article distributed under
the terms of the Creative Commons Attribution License, which permits unrestricted
use, distribution, and reproduction in any medium, provided the original author
and source are credited.
Abbreviations: AZT, azidothymidine; CFP, cyan fluorescent protein; DC, dendritic
cell; GALT, gut-associated lymphoid tissue; GFP, green fluorescent protein; LC,
Langerhans’ cell; MDDC, monocyte-derived dendritic cell; MHC, major histocom-
patibility complex; MVB, multivesicular body; PBL, peripheral blood lymphocyte;
sCD4, soluble CD4; SEB, staphylococcal enterotoxin B
* To whom correspondence should be addressed. E-mail: email@example.com.
PLoS Pathogens | www.plospathogens.orgJanuary 2007 | Volume 3 | Issue 1 | e40038
cells. After an additional 2 d of incubation in the presence of
azidothymidine (AZT), productive infection of T cells was
measured by immunostaining with anti-p24Gag(Figure 1A).
Transmission of 81A (R5-tropic) virions from immature
MDDCs to T cells was biphasic, as reported . The early
phase (0 to 1 d) involved the trans pathway; the later phase (1
to 5 d) involved the de novo pathway and was sensitive to the
HIV protease inhibitor amprenavir (not shown). During the
first day, 25% of R5-tropic 81A virions were transmitted by
the trans pathway; 75% were transmitted by the de novo
pathway over the ensuing 4 d (Figure 1B). In mature MDDCs,
however, approximately 93% of virions were transmitted by
the trans pathway during the first day. X4-tropic NL4–3
virions were transmitted by both immature and mature
MDDCs principally by the trans pathway. Similar results were
obtained when MDDCs were analyzed from nine different
normal donors (Figure 1C).
In vivo, DCs may not immediately interact with T cells after
virion capture. Accordingly, we investigated how a delay in T-
cell contact might affect transmission through the trans
pathway with a virion-based HIV-1 fusion assay [13,14].
MDDCs loaded with HIV-1 virions containing b-lactamase-
Vpr (BlaM-Vpr) were incubated with autologous T cells, and
fusion to CD4 T cells was monitored by the changes in
fluorescence of CCF2, a BlaM substrate loaded into the cells
(Figure 1D). When NL4–3 virions were presented immediately
after binding to mature MDDCs, up to 24% of CD4 T cells
displayed BlaM activity, indicating virion fusion. Trans-
mission was less efficient when virions were presented by
immature MDDCs. Fewer 81A than NL4–3 virions were
transmitted, likely because there were fewer CCR5- than
CXCR4-expressing cells in resting peripheral blood lympho-
cytes (PBLs). When virions were presented by MDDCs after
incubation at 37 8C for up to 120 min, transmission efficiency
decreased sharply (Figure 1E) in both immature and mature
MDDCs. This rapid decrease was not due to a relative lack of
sensitivity of the fusion assay in the context of trans-infection.
As in our previous studies of T-cell infection with free virions,
the fusion assay proved to be both sensitive and quantitative
over a broad range of viral inputs in these DC–T-cell mixing
experiments  (Figure S1). Thus, HIV-1 trans-infection
from MDDCs to autologous CD4 T cells is efficient only for a
limited time after virion capture.
Our results show that immature DCs preferentially trans-
mit R5-tropic HIV-1 by the de novo pathway, as described
[11,15–17], and X4-tropic HIV-1 by the trans pathway. Mature
DCs transmit both R5- and X4-tropic virions mainly by the
trans pathway. Since immature DCs are present at mucosal
sites of viral entry, while mature DCs reside in lymph nodes
and in the gut-associated lymphoid tissue (GALT), HIV-1 may
exploit different transmission strategies at different anatomic
sites in vivo. In healthy mucosa, immature DCs are likely to
transmit R5-tropic HIV-1 principally via the de novo path-
way, especially since the efficiency of the trans pathway
declines rapidly (Figure 1E and [11,16]); the trans pathway
might contribute to the local spread of virus from mucosal
immature DCs to macrophages and CD4 T cells. However, in
inflamed mucosal epithelium, which contains a greater
proportion of mature DCs, HIV-1 transmission might
preferentially involve the trans pathway, as in human cervical
explants . In lymph nodes and GALT, the proximity of
mature DCs to T cells would further favor the trans pathway.
Since DC–T-cell conjugates are major sites of HIV-1
production [5–7], trans-infection could be critical in the
intense viral replication that characterizes the acute and
chronic phases of untreated HIV-1 infection.
Neutralizing Surface-Bound Virions on DCs Abrogates
To identify the cellular compartment from which HIV-1 is
transmitted in trans, we selectively neutralized surface-bound
HIV-1 virions with truncated recombinant soluble CD4
(sCD4; AIDS Reagent Program ), which binds the HIV-1
envelope gp120 protein and prevents engagement of CD4 on
T cells. Cells were treated at 4 8C to protect internalized
virions from sCD4 exposure. NL4–3 virions containing BlaM-
Vpr were bound to MDDCs and allowed to internalize, and
cell-surface virions were neutralized with sCD4. In the
absence of an internalization step, surface-bound virions on
immature and mature MDDCs effectively fused to CD3þCD4þ
cells (Figure 2A; bars 1 and 3); this fusion was effectively
blocked by sCD4 (bars 2 and 4). However, when HIV-1 virions
were internalized at 37 8C for 30 min before treatment (bars 5
to 8), sCD4 still inhibited virion fusion (bars 6 and 8). To
confirm that sCD4 neutralized surface-bound virions without
impairing subsequent virion transfer, two sets of HIV-1
virions were successively bound to immature MDDCs, but
only the first was neutralized with sCD4 (Figure 2B). Similar
amounts of HIV-1 fused to T cells regardless of the presence
of previously neutralized cell-surface virions on the DCs (bars
3 and 5). Thus, sCD4 does not interfere with virological
synapse formation or the transfer of virions bound after
To further confirm the absence of virion transfer from
internal cellular compartments, we performed sequential
loading of HIV-1 expressing green fluorescent protein (GFP)
(GFP-HIV) and then HIV-1 expressing cyan fluorescent
protein (CFP) (CFP-HIV)  (Figure 2C). GFP-HIV was
bound to MDDCs and incubated at 37 8C to allow virion
internalization. Some residual GFP-HIV virions remained at
the surface. Next, CFP-HIV was bound but kept at 4 8C to
prevent virion internalization. MDDCs were then incubated
with autologous T cells for 48 h. sCD4 treatment after binding
PLoS Pathogens | www.plospathogens.orgJanuary 2007 | Volume 3 | Issue 1 | e40039
HIV-1 trans-Infection of T Cells by DCs
Dendritic cells (DCs) patrol peripheral mucosal sites, capturing and
processing potential pathogens into antigenic peptides for pre-
sentation to T cells of lymphoid organs, and thereby initiating an
immune response. HIV-1 had been proposed to use DCs as ‘‘Trojan
horses,’’ hiding inside the DCs and surviving the degradation
pathway to gain access to the lymph nodes and spread to the T
cells. Our study challenges this ‘‘Trojan horse’’ model by showing
that only HIV-1 virions bound to the surface of DCs, and not
internalized virions, are transmitted to T cells. Even when T cells
specifically recognized the antigen presented by DCs, the infection
of T cells was principally mediated by virions remaining at the
surface of the DCs. Interestingly, in this context of antigen-specific
recognition, which increases the trafficking toward the immuno-
logical synapse of DC internal vesicles, where HIV-1 virions seem to
hide, a few internal virions could infect T cells. Our findings suggest
that in vivo transmission to T cells of HIV-1 virions captured by DCs
should be more sensitive to neutralization than previously expected.
and internalization of GFP-HIV but before CFP-HIV binding
fully blocked transmission of GFP-HIV (Figure 2C, middle
panel), indicating that residual surface-bound GFP-HIV was
the source of virus for transmission to T cells (left panel).
Under these conditions, CFP-HIV was still transferred to T
cells, confirming that sCD4 treatment did not affect
subsequent transfer of virions and was not inherently
harmful. In the absence of treatment, many double-positive
GFPþCFPþT cells were observed, indicating that more than
one virion can be transmitted to each T cell (left panel).
Thus, the neutralization studies showed that only surface-
bound, but not internalized, HIV virions mediated trans-
infection from MDDCs to T cells. sCD4 likely neutralizes HIV-
1 virions by competing for viral binding to the CD4 receptor
on T cells, as intact HIV-1 virions, including gp120, remained
associated with the MDDCs after treatment (unpublished
data). Although sCD4 has been reported to allow HIV-1
fusion to cells that do not express CD4 , sCD4 did not
induce fusion to CD4–cells in our experiments, as demon-
strated by the absence of BlaM transfer to CD8 T cells or B
cells (not shown).
Next, we stripped surface HIV virions from MDDCs by
proteolytic digestion (Figure 2D). GFP-HIV was again allowed
to bind and internalize. CFP-HIV was restricted to the surface
of MDDCs and served as a control for neutralization by the
proteolytic enzymes. MDDCs loaded with GFP-HIV and CFP-
HIV were then treated with trypsin as described [10,21] or
pronase and incubated for 2 d with T cells. Trypsin treatment
did not effectively remove externally bound CFP-HIV virions
(unpublished data). However, in the presence of increasing
amounts of pronase, fewer surface-bound CFP-HIV virions
were transmitted to T cells, indicating increasingly effective
removal the surface-bound HIV-1 by this protease cocktail.
However, pronase had the same effect on the transfer of GFP-
Figure 1. Mature MDDCs Transmit HIV-1 to T Cells Primarily via the trans Pathway
(A–C) NL4–3 or 81A virions were bound to MDDCs. After washing, the cells were added to activated autologous T cells immediately or after 1 to 5 d of
culture at 37 8C to allow HIV-1 transmission to T cells for 24 h. The number of transmission events was measured by monitoring the appearance of
infected T cells detected by p24Gagintracellular immunostaining.
(A) Transmission of 81A virions from immature MDDCs to T cells over time. FACS plots represent the population of T cells (CD3þCD1a–) analyzed for
intracellular Gag and CD4 expression.
(B) Effects of maturation on HIV-1 transmission from MDDCs to T cells. Values are percentages of all transmission events over 5 d.
(C) Relative contribution of trans pathway or de novo pathway in transmission of virus from immature or mature MDDCs to T cells. Data are averaged
from MDDCs derived from nine donors. Transmission events from days 1 to 5 were added to determine the number of transmission events occurring via
the de novo pathway.
(D and E) NL4–3 or 81A virions containing BlaM-Vpr were bound to MDDCs at 4 8C. After washing, the cells were added to autologous T cells
immediately (T0) or after incubation for 10 to 120 min at 37 8C. HIV-1 transmission was measured with a virion-based fusion assay after gating on
(D) HIV-1 transmission from MDDCs to T cells at T0. FACS plots show CD3þCD4þCD1a–cells analyzed for virion fusion. To control for specificity, MDDCs
were incubated with T cells and entry inhibitors (500 nM TAK-779 or 500 nM AMD3100).
(E) Effect of time on NL4–3 and 81A transmission from MDDCs to T cells. The curve is representative of four experiments.
PLoS Pathogens | www.plospathogens.org January 2007 | Volume 3 | Issue 1 | e40040
HIV-1 trans-Infection of T Cells by DCs
HIV, whether it had been internalized or not. Thus,
successfully transmitted GFP-HIV virions appear to originate
from the cell surface rather than from an internal, pronase-
resistant compartment in MDDCs.
We considered the possibility that the low levels of
transmission in the presence of sCD4 could correspond to
transmission events from internalized HIV-1, masked under
our experimental conditions. However, changes in viral
Figure 2. HIV-1 Virions Transmitted In Trans from MDDCs to T Cells Are Sensitive to sCD4 and Pronase
(A) Immature and mature MDDCs were incubated at 4 8C with NL4–3 virions containing BlaM-Vpr, washed, and incubated for 30 min at 37 8C to allow
virion internalization or held at 4 8C. MDDCs were treated (or not) with sCD4 at 4 8C to inactivate cell-surface virions. After extensive washes, loaded
MDDCs were added to autologous PBLs to allow HIV-1 trans-infection of T cells. Fusion was measured with the virion-based fusion assay after gating on
the CD3þCD4þcells. The histogram presents a representative experiment independently performed five times with similar results.
(B) sCD4 treatment of MDDCs loaded with HIV-1 virions does not inhibit transfer of virions loaded after treatment. Two sets of HIV-1 virions containing
BlaM-Vpr were successively bound to immature MDDCs and neutralized with sCD4. After incubation with autologous CD4 T cells, fusion to CD3þCD4þ
cells was assessed.
(C) MDDCs were sequentially incubated with two reporter viruses (1 lg of p24Gageach). GFP-HIV was bound and internalized, but CFP-HIV was bound
only. Loaded MDDCs were incubated with autologous PBLs for 2 d. FACS plots show T cells analyzed for GFP and CFP expression when MDDCs were
untreated or treated with sCD4 at 4 8C between or after loading of the reporter viruses.
(D) GFP-HIV and CFP-HIV were either successively bound to MDDCs or GFP-HIV was, in addition, internalized by MDDCs during a 30-min incubation at
37 8C. MDDCs were then treated with increasing concentrations of pronase. Curves represent the number of T cells expressing CFP or GFP expressed as
a percentage of the untreated samples. This graph presents a representative experiment with mature MDDCs that was independently performed three
times with both immature and mature MDDCs.
PLoS Pathogens | www.plospathogens.orgJanuary 2007 | Volume 3 | Issue 1 | e40041
HIV-1 trans-Infection of T Cells by DCs
input, internalization time, and viral strains failed to reveal
significant transfer from internal compartments (Figure S2).
We also studied a second type of DC, LCs derived from CD34þ
cord blood cells (MatTek; http://www.mattek.com). As ob-
served in MDDCs, the transfer of HIV virions from LCs to
allogenic T cells again was mediated by virions bound at the
surface of the LCs (Figure S3). Since MDDCs and LCs derived
from CD34 progenitors are excellent surrogates of in vivo
DCs, we conclude that most virions transmitted in trans in
vivo likely originate from the cell surface.
These results in MDDCs and LCs sharply contrast with the
report that formed the basis for the prevailing model of trans-
infection . In that report, surface-bound virions were
neutralized by proteolytic digestion with trypsin. Although in
our hands trypsin did not digest surface-bound virions as
potently as pronase, the discordance of results likely lies in
the use of different reporter systems. Kwon et al.  used a
luciferase reporter that does not allow the distinction
between infection of T cells or MDDCs. Since the immature
MDDCs used in that study are highly susceptible to fusion
with the R5-tropic BaL envelopes  and efficiently
replicate CCR5-tropic HIV-1 [15,22], the luciferase activity
may have derived from infected immature MDDCs, not T
cells, in the coculture. Our flow-based assays, which permit a
clear distinction between infection of T cells and MDDCs in
coculture, reveal that HIV-1 virions transmitted in trans are
sensitive to pronase treatment.
Our results also differ from three later studies, further
supporting the notion that HIV infection of T cells by DCs
involves the transfer of internalized virions from DCs to
interacting CD4 T cells [9,10,15]. McDonald et al.  showed
the recruitment of ‘‘trypsin-resistant’’ HIV-1 virions to the
immunological synapse; however, functional assays were not
performed to confirm that the interacting CD4 T cells were
actually infected by the recruited virions. We suspect that,
while virions may be transported to the synapse, these virions
are not successfully transmitted. Wiley et al.  showed the
release of infectious virions from HIV-1–loaded MDDCs, even
after surface-bound virions were removed with trypsin.
However, the efficacy of the trypsin treatment was only
controlled in the experiments measuring the release of
p24Gagin the supernatant, not in studies measuring the
infectivity of these virions. Finally, Ganesh et al.  observed
the transfer of some virions from MDDCs to T cells in the
presence of neutralizing antibodies, a surprising result in
light of our findings with sCD4. In their study, the efficacy of
the antibody neutralization was measured with free virions
but not with MDDCs bearing only surface-bound HIV-1
virions. The amount of antibody required to neutralize free
virions might be lower than the amount needed to neutralize
surface-bound virions on MDDCs, a possibility that would
explain our divergent results. Of note, our findings are
supported by a recent study showing that, in Raji cells
expressing DC-SIGN (DC-specific ICAM-3 grabbing non-
integrin), surface-bound rather than internalized virions are
transmitted in trans to 293T cells expressing CD4 and CCR5
trans-Infection in the Context of Superantigen Stimulation
Several reports have suggested that captured HIV-1 virions
are stored in MVBs, raising the possibility that HIV-1
mediates trans-infection of T cells by highjacking a pathway
involved in the trafficking of internal vesicles to the
immunological synapse. In DCs, the transport of MHC class
II from the MVB to the immunological synapse requires a T-
cell–mediated signal . Only T cells of the appropriate
antigen specificity trigger this transport. Since antigen
recognition could mobilize the release of HIV-1 virions from
the MVB, we investigated trans-infection in the context of
stimulation with a superantigen, staphylococcal enterotoxin
B (SEB) (Figure 3). SEB activates T cells by crosslinking the
variable region of T-cell receptor b-chain and the MHC class
II molecule expressed on the DC surface .
NL4–3 virions containing BlaM-Vpr were bound at 4 8C to
SEB-pulsed MDDCs. Viral transfer to autologous purified CD4
T cells was measured after allowing virion internalization, or
not, by MDDCs. Again, sCD4 completely blocked HIV-1
transmission (Figure 3A). However, since the fusion assay
does not require T-cell activation to generate a positive signal,
trans-infection could be detected in the absence of engage-
ment of the T-cell receptor and MHC class II. To further
ensure that transfer was analyzed only when the MDDCs and T
cells were effectively engaged, we measured productive
infection of resting T cells by immunostaining for p24Gag
(Figure 3B). Only T cells stimulated by SEB-loaded MDDCs are
rendered permissive by releasing a postentry restriction block
created by APOBEC3G (apolipoprotein B mRNA-editing
enzyme, catalytic polypeptide-like 3G) . Again, the vast
majority of transmission events were neutralized by sCD4.
However, under these experimental conditions, a few virions
were transmitted from an sCD4-resistant compartment, as
evidenced by the slight increase in transfer when SEB-loaded
MDDCs were allowed to internalize HIV-1 virions at 37 8C.
These transmission events were not due to new virion
production by MDDCs since similar results were observed
when AZT was added after 24 h of coincubation of MDDCs
and T cells. When SEB-pulsed MDDCs were allowed to
internalize HIV-1 virions for a longer period, HIV-1 virion
transfer slightly increased peaking at 1 h of internalization
(Figure 3C). Subsequently, the efficiency of transfer from
internal compartments decreased, likely as a consequence of
degradation or inactivation of the internalized virions.
In conclusion, HIV-1 virions transmitted in trans from DCs
to T cells principally originate from the surface of DCs,
except during antigen recognition, when a few internalized
virions may also be transmitted to the antigen-specific T cells.
Whether these rare events contribute to the preferential
infection and elimination of HIV-specific T cells in vivo 
is not known. Nevertheless, even within this context, the vast
majority of transmitted virions are derived from the surface
of DCs. Our results do not challenge the prevailing view that
HIV-1 virions are internalized in the DCs. Indeed, we
detected large amounts of internalized HIV-1 virions by
microscopy. However, unless HIV-1–loaded DCs encounter T
cells of the appropriate specificity, virion internalization
appears to be a dead end for HIV-1 trans-infection. Since the
C-type lectin receptors involved in trans-infection are
localized in lipid rafts , surface-bound HIV-1 likely
exploits the clustering of lipid rafts at the immunological
synapse to enhance trans-infection of CD4 T cells. Because
trans-infection principally involves surface-bound virions, our
findings suggest that attachment inhibitors could be used to
limit trans-infection of T cells by DCs, in vivo.
PLoS Pathogens | www.plospathogens.org January 2007 | Volume 3 | Issue 1 | e40042
HIV-1 trans-Infection of T Cells by DCs
Materials and Methods
Two-phase transmission of HIV-1 from MDDCs to autologous T
cells. To study HIV-1 transmission from DCs to T cells, MDDCs (2 3
106) were incubated with 81A or NL4–3 virions (50 lg of p24Gag/ml)
for 1 h at 4 8C, washed four times in cold PBS, incubated for 0 to 5 d
at 37 8C, diluted 1:10, and added to autologous phytohemagglutinin-
activated PBLs (23106). Cocultures were maintained for 3 d in RPMI
with 10% FBS, 20 IU/ml IL-2, 25 ng/ml IL-4, 50 ng/ml GM-CSF, and
penicillin and streptomycin (100 lg/ml each); at 24 h, AZT (10 lM)
was added to prevent further infection. Infected T cells were
identified by intracellular immunostaining for p24Gagcombined with
antibodies against CD3, CD4, and CD1a. Productively infected T cells
represent the percentage of p24GagþCD4–cells in the CD3þCD1a–
population and correspond to infected CD4 T cells that had
effectively downregulated CD4 receptors due to expression of select
viral gene products, including nef, vpu, and env . In some
experiments, an HIV protease inhibitor, amprenavir (40 nM)
(Division of Acquired Immunodeficiency Syndrome, National In-
stitute of Allergy and Infectious Diseases; http://www.niaid.nih.gov),
was added during the binding step and maintained for the rest of the
Measuring trans-infection of autologous T cells by MDDCs. MDDCs
derived from CD14þmonocyte were induced to mature with
poly(I:C) and tumor necrosis factor a . The 81A or NL4–3
virions containing BlaM-Vpr (500 ng of p24Gag) [12–14] were
incubated with MDDCs (2 3 106) or with CD34-derived LCs for 1 h
at 4 8C, washed four times in cold PBS, and incubated at 37 8C for
the indicated time to allow virion internalization or kept at 4 8C.
Aliquots (2 3 105cells) were added to autologous resting PBLs (2 3
106), and incubated at 37 8C for 1 h. HIV-1 fusion to CD4þCD3þcells
was measured using the virion-based fusion assay combined with
immunostaining with CD1a-APC, CD4-PE Cy7, and CD3-APC Cy7
antibodies [13,14]. Cells were analyzed by flow cytometry (BD LSRII;
Becton Dickinson, http://www.bd.com) and analyzed with FlowJo
software (Treestar Software, http://www.flowjo.com).
Assessing trans-infection of autologous T cells by MDDCs by
measuring proviral expression. GFP-HIV virions were bound to
MDDCs for 1 h at 4 8C and allowed to internalize at 37 8C for 30 min;
CFP-HIV virions were only bound to MDDCs. As indicated, surface
virions were neutralized with sCD4 before or after the binding of
CFP-HIV or by pronase after the binding of GFP-HIV and CFP-HIV.
MDDCs were then incubated with autologous T cells for 48 h. Cells
Figure 3. HIV-1 Transmitted from SEB-Stimulated MDDCs to T Cells Remains Mostly Sensitive to sCD4
NL4–3 virions containing BlaM-Vpr were bound to SEB-pulsed MDDCs. The MDDCs were then incubated or not at 37 8C for the indicated time, treated
or not with sCD4 at 4 8C, and incubated with purified autologous CD4 T cells.
(A–C) Viral transfer was measured by analyzing HIV-1 fusion to CD4þCD3þT cells after 2 h of culture (A) and by measuring productive infection of the T
cells by intracellular Gag immunostaining 3 d later (B and C).
(B) Histogram depicts one representative experiment performed three times with cells from three independent donors.
(C) HIV-1 transfer to resting T cells from DCs pulsed with SEB and treated with sCD4 over time. This experiment was performed in triplicate; similar
results obtained with MDDCs from two additional donors.
PLoS Pathogens | www.plospathogens.orgJanuary 2007 | Volume 3 | Issue 1 | e40043
HIV-1 trans-Infection of T Cells by DCs
were immunostained with CD1a-APC, CD4-PE Cy7, and CD3-APC
Cy7 antibodies and analyzed by flow cytometry.
Neutralization of surface-bound virions. To neutralize surface-
bound virions, MDDCs or CD34-derived LCs loaded with HIV-1
virions were incubated for 90 min at 4 8C with 20 lg/ml sCD4 in RPMI
and 10% FBS and extensively washed with PBS before MDDCs or
CD34-derived LCs were added to T cells. To neutralize virions with
pronase, the HIV-1–loaded MDDCs were incubated for 30 min at 4 8C
with 50 to 400 lg/ml pronase (Roche, http://www.roche.com).
Measurement of trans-infection of autologous CD4 T cells by SEB-
pulsed MDDCs. MDDCs (23106) were pulsed with SEB (0.5 lg/ml) at
37 8C for 1 h. NL4–3 virions containing BlaM-Vpr (500 ng of p24Gag)
were allowed to bind at 4 8C to the MDDCs; as indicated, cells were
incubated for 30 min to 4 h at 37 8C to allow internalization. Surface-
bound virions were then neutralized or not with sCD4. The HIV-
loaded MDDCs were added to autologous purified resting CD4 T
cells, and trans-infection was measured with the fusion assay at 2 h or
by measuring productive infection after 3 d of coculture. Produc-
tively infected T cells were identified by intracellular immunostaining
for p24Gagcombined with antibodies against CD3, CD4, and CD1a.
We then measured the percentage of p24GagþCD4–cells in the
Figure S1. The Virion-Based Fusion Assay Is Sensitive and Quanti-
tative in the Context of trans-Infection
Serial dilutions of NL4–3 virions containing BlaM-Vpr were bound to
immature or mature MDDCs at 4 8C. After three washes, these HIV-1–
loaded MDDCs were incubated with autologous T cells for 2 h. Viral
transfer to T cells was determined by measuring the amount of BlaMþ
Found at doi:10.1371/journal.ppat.0030004.sg001 (50 KB PDF).
Figure S2. Changes in Virion Input, Internalization Time, and Viral
Strain Fail to Reveal HIV-1 trans-Infection of T Cells from Internal
Serial dilutions (A) or 500 ng (B and C) of virions containing BlaM-
Vpr was bound to MDDCs. Where indicated, viral strains other than
NL4–3 were used (C). Virions were then internalized (or not) at 37 8C
for 30 min (A and C) or for increasing amounts of time (B). After
sCD4 treatment at 4 8C to inactivate surface virions, the loaded DCs
were incubated with PBLs. Curves represent the transfer of HIV-1
from immature or mature MDDCs to T cells in the presence or
absence of sCD4 treatment.
Found at doi:10.1371/journal.ppat.0030004.sg002 (68 KB PDF).
Figure S3. LCs Transmit HIV-1 to T Cells from an External
(A) Phenotype of the CD34-derived LCs matured or not with 5 lg/ml
LPS and 50 ng/ml tumor necrosis factor a for 24 h.
(B and C) NL4–3 virions containing BlaM-Vpr were bound to
immature or mature LCs at 4 8C. After washing, the cells were added
to allogenic T cells immediately or after incubation at 37 8C as
indicated. HIV-1 transmission to T cells was measured with the
virion-based fusion assay after gating on the CD3þCD4þcells.
(B) Effect of time on HIV transfer from immature or mature LCs to T
cells. Histogram shows one of two experiments.
(C) LCs were treated or not at 4 8C with sCD4 before incubation with
the T cells.
Found at doi:10.1371/journal.ppat.0030004.sg003 (113 KB PDF).
We thank K. Stopak for discussions, D. Schols for AMD3100, S.
Ordway and G. Howard for editorial assistance, J. Carroll for graphic
arts, and S. Cammack and R. Givens for administrative assistance.
Author contributions. MC and JN conceived and designed the
experiments and analyzed the data. JN performed the experiments.
JFK contributed reagents/materials/analysis tools. MC wrote the
paper. WCG supervised the study.
Funding. These studies were supported by funding from the
National Institutes of Health (P01 HD40543, 1S10 RR022448, and R03
AI062263). MC was supported by Universitywide AIDS Research
Program (University of California, Office of the President, Oakland,
California, United States) (F03-GI-205). University of California San
Francisco–Gladstone Institute of Virology and Immunology Center
for AIDS Research (UCSF-GIVI CFAR) provided infrastructure
support (P30 AI27763). These studies were also made possible by
grant RR 18928–01 from the National Institutes of Health National
Center for Research Resources.
Competing interests. The authors have declared that no competing
1. Banchereau J, Steinman RM (1998) Dendritic cells and the control of
immunity. Nature 392: 245–252.
2.Spira AI, Marx PA, Patterson BK, Mahoney J, Koup RA, et al. (1996) Cellular
targets of infection and route of viral dissemination after an intravaginal
inoculation of simian immunodeficiency virus into rhesus macaques. J Exp
Med 183: 215–225.
3. Hu J, Gardner MB, Miller CJ (2000) Simian immunodeficiency virus rapidly
penetrates the cervicovaginal mucosa after intravaginal inoculation and
infects intraepithelial dendritic cells. J Virol 74: 6087–6095.
4. Hu Q, Frank I, Williams V, Santos JJ, Watts P, et al. (2004) Blockade of
attachment and fusion receptors inhibits HIV-1 infection of human
cervical tissue. J Exp Med 199: 1065–1075.
5. Cameron PU, Freudenthal PS, Barker JM, Gezelter S, Inaba K, et al. (1992)
Dendritic cells exposed to human immunodeficiency virus type-1 transmit
a vigorous cytopathic infection to CD4þ T cells. Science 257: 383–387.
6. Pope M, Betjes MG, Romani N, Hirmand H, Cameron PU, et al. (1994)
Conjugates of dendritic cells and memory T lymphocytes from skin
facilitate productive infection with HIV-1. Cell 78: 389–398.
7. Hu J, Miller CJ, O’Doherty U, Marx PA, Pope M (1999) The dendritic cell-T
cell milieu of the lymphoid tissue of the tonsil provides a locale in which
SIV can reside and propagate at chronic stages of infection. AIDS Res Hum
Retroviruses 15: 1305–1314.
8. Garcia E, Pion M, Pelchen-Matthews A, Collinson L, Arrighi JF, et al. (2005)
HIV-1 trafficking to the dendritic cell-T-cell infectious synapse uses a
pathway of tetraspanin sorting to the immunological synapse. Traffic 6:
9. Wiley RD, Gummuluru S (2006) Immature dendritic cell-derived exosomes
can mediate HIV-1 trans infection. Proc Natl Acad Sci U S A 103: 738–743.
10. McDonald D, Wu L, Bohks SM, KewalRamani VN, Unutmaz D, et al. (2003)
Recruitment of HIV and its receptors to dendritic cell-T cell junctions.
Science 300: 1295–1297.
11. Turville SG, Santos JJ, Frank I, Cameron PU, Wilkinson J, et al. (2004)
Immunodeficiency virus uptake, turnover, and two-phase transfer in
human dendritic cells. Blood 103: 2170–2179.
12. Cavrois M, Neidleman J, Kreisberg JF, Fenard D, Callebaut C, et al. (2006)
Human immunodeficiency virus fusion to dendritic cells declines as cells
mature. J Virol 80: 1992–1999.
13. Cavrois M, de Noronha C, Greene WC (2002) A sensitive and specific
enzyme-based assay detecting HIV-1 virion fusion in primary T lympho-
cytes. Nat Biotechnol 20: 1151–1154.
14. Cavrois M, Neidleman J, Bigos M, Greene WC (2004) Fluorescence
resonance energy transfer-based HIV-1 virion fusion assay. Methods Mol
Biol 263: 333–344.
15. Ganesh L, Leung K, Lore K, Levin R, Panet A, et al. (2004) Infection of
specific dendritic cells by CCR5-tropic human immunodeficiency virus
type 1 promotes cell-mediated transmission of virus resistant to broadly
neutralizing antibodies. J Virol 78: 11980–11987.
16. Nobile C, Petit C, Moris A, Skrabal K, Abastado JP, et al. (2005) Covert
human immunodeficiency virus replication in dendritic cells and in DC-
SIGN-expressing cells promotes long-term transmission to lymphocytes. J
Virol 79: 5386–5399.
17. Burleigh L, Lozach PY, Schiffer C, Staropoli I, Pezo V, et al. (2006) Infection
of dendritic cells (DCs), not DC-SIGN-mediated internalization of human
immunodeficiency virus, is required for long-term transfer of virus to T
cells. J Virol 80: 2949–2957.
18. Garlick RL, Kirschner RJ, Eckenrode FM, Tarpley WG, Tomich CS (1990)
Escherichia coli expression, purification, and biological activity of a
truncated soluble CD4. AIDS Res Hum Retroviruses 6: 465–479.
19. Levy DN, Aldrovandi GM, Kutsch O, Shaw GM (2004) Dynamics of HIV-1
recombination in its natural target cells. Proc Natl Acad Sci U S A 101:
20. Speck RF, Esser U, Penn ML, Eckstein DA, Pulliam L, et al. (1999) A trans-
receptor mechanism for infection of CD4-negative cells by human
immunodeficiency virus type 1. Curr Biol 9: 547–550.
21. Kwon DS, Gregorio G, Bitton N, Hendrickson WA, Littman DR (2002) DC-
SIGN-mediated internalization of HIV is required for trans-enhancement
of T cell infection. Immunity 16: 135–144.
22. Kawamura T, Gulden FO, Sugaya M, McNamara DT, Borris DL, et al. (2003)
R5 HIV productively infects Langerhans cells, and infection levels are
regulated by compound CCR5 polymorphisms. Proc Natl Acad Sci U S A
PLoS Pathogens | www.plospathogens.org January 2007 | Volume 3 | Issue 1 | e40044
HIV-1 trans-Infection of T Cells by DCs
23. Boes M, Cerny J, Massol R, Op den Brouw M, Kirchhausen T, et al. (2002) T- Download full-text
cell engagement of dendritic cells rapidly rearranges MHC class II
transport. Nature 418: 983–988.
24. Lavoie PM, Thibodeau J, Erard F, Sekaly RP (1999) Understanding the
mechanism of action of bacterial superantigens from a decade of research.
Immunol Rev 168: 257–269.
25. Chiu YL, Soros VB, Kreisberg JF, Stopak K, Yonemoto W, et al. (2005)
Cellular APOBEC3G restricts HIV-1 infection in resting CD4þ T cells.
Nature 435: 108–114.
26. Douek DC, Brenchley JM, Betts MR, Ambrozak DR, Hill BJ, et al. (2002) HIV
preferentially infects HIV-specific CD4þ T cells. Nature 417: 95–98.
27. Cambi A, de Lange F, van Maarseveen NM, Nijhuis M, Joosten B, et al.
(2004) Microdomains of the C-type lectin DC-SIGN are portals for virus
entry into dendritic cells. J Cell Biol 164: 145–155.
28. Mascola JR, Louder MK, Winter C, Prabhakara R, De Rosa SC, et al. (2002)
Human immunodeficiency virus type 1 neutralization measured by flow
cytometric quantitation of single-round infection of primary human T
cells. J Virol 76: 4810–4821.
PLoS Pathogens | www.plospathogens.org January 2007 | Volume 3 | Issue 1 | e40045
HIV-1 trans-Infection of T Cells by DCs