R5- and X4-HIV-1 use differentially the endometrial epithelial cells
HEC-1A to ensure their own spread: Implication for mechanisms
of sexual transmission
Héla Saïdi⁎, Giuliana Magri, Nadine Nasreddine, Mary Réquena, Laurent Bélec
Unité INSERM U743 Equipe “Immunité et Biothérapie Muqueuse”, Centre de Recherches Biomédicales des Cordeliers, 15,
rue de l’Ecole de Médecine, 75270 Paris, Cedex 06, France
Received 13 April 2006; returned to author for revision 31 May 2006; accepted 20 July 2006
Available online 24 August 2006
The mechanism of viral transmission across the mucosal barrier is poorly understood. Using the endometrial epithelium-derived cell line HEC-
1A, we found that the cells are capable of sequestering large numbers of HIV-1 particles but are refractory to cell-free viral infection. The removal
of heparan sulfate moieties of cell-surface proteoglycans (HSPG) from the apical pole of HEC-1A accounted for at least 60% of both R5- and X4-
HIV-1 attachment, showing their important implication in viral attachment. HEC-1A cells also have the capacity to endocytose a weak proportion
of the attached virus and pass it along to underlying cells. Fucose, N-acetylglucosamine and mannosylated-residues inhibited the transcytosis of
some virus isolates, suggesting that mannose receptors can be implicated on the both R5- and X4-HIV-1 transcytosis. The inhibition of HIV
transcytosis by blocking CCR5 mAb suggests the implication of specific interaction between the viral gp120 and sulfated moiety of syndecans
during the transcytosis of mostly R5- and X4-HIV-1. At the basolateral pole of HEC-1A, HSPG sequestered X4- and not R5-HIV-1, highlighting
the important role of HEC-1A as an X4 virus reservoir. The cell-free virus particles that have transcytosed could infect activated Tcells but with a
weaker efficiency than virus that had not transcytosed. The specific stimulation of HEC-1A by R5-HIV-1 increased the release of monocytes/
chemokines-attracting chemokines (IL-8 and GR0) and proinflammatory cytokines (TNF-β and IL-1α) that enhanced the production of virus by
activated T cells. This study suggests that R5 and X4 viruses can differentially use epithelial cells to ensure their own spread.
© 2006 Elsevier Inc. All rights reserved.
Keywords: HIV; HEC-1A; Heparan sulfate proteoglycans
The genitourinary tract is the main route of natural infection
for human immunodeficiency virus type 1 (HIV-1). The
mucosae of the vaginal tracts have a covering of polarized
epithelial cells. Polarized simple epithelial cells have a plasma
membrane that is separated into clearly distinct domains by tight
junctions: the apical domain, which faces the tract lumen, and
the basolateral domain, which faces the serosal side and the
internal milieu containing HIV target cells (Dezzutti et al.,
2001; Howell et al., 1997). Epithelial cells express a variety of
adhesion molecules and proinflammatory cytokines that allow
these cells to interact with the underlying macrophages and
lymphocytes (Carreno et al., 2002; Prakash et al., 2003). The
intimate interaction between epithelial cells and immune cells
has important implications for HIV-1 transmission. The
epithelial cells have been suggested to be a reservoir for HIV-
1 (Dezzutti et al., 2001; Howell et al., 1997). Several studies
have shown that epithelial cells can be infected in vitro, but
there are no conclusive data about the productive infection of
genital epithelial cells in vivo (Dezzutti et al., 2001). Interest-
ingly, Bomsel et al. have shown that epithelial cells endocytose
HIV-1 and pass it along to underlying cells (Bomsel, 1997).
The initial cellular targets of HIV in the first hours of
infection have not yet been identified. Several in vitro models
have been developed to study the mechanisms involved in the
Virology 358 (2007) 55–68
⁎Corresponding author. Fax: +33 1 43 29 29 74.
E-mail address: email@example.com (H. Saïdi).
0042-6822/$ - see front matter © 2006 Elsevier Inc. All rights reserved.
transmucosal passage of the virus (Hocini et al., 2001; Lorin et
al., 2005). The transcytosis of HIV-1 through a tight monolayer
of epithelial cells, such as the human endometrial carcinoma
type A (HEC-1A) cells, has been chosen herein to evaluate the
initial interactions between cell-free virus and mucosal
epithelial cells, as well as the intracellular transport of HIV
from the apical to the basolateral pole of the monolayer.
Given that HEC-1A cells lack the entry receptor CD4 (Kozak
et al., 1999; McDougal et al., 1991), HIV-1 has to use
attachment receptors distinct from CD4. Several receptors have
been reported to facilitate the HIV attachment into CD4-
negative cells. Specifically, galactosyl-ceramide (Harouse et al.,
1991, 1995; Yahi et al., 1992), adhesion molecules such as
ICAM-1 and LFA-1(Bounou et al., 2004; Losier et al., 2003;
Tardif and Tremblay, 2005), C-type lectin such as DC-SIGN
and the mannose receptor (Chehimi et al., 2003; Geijtenbeek et
al., 2000; Turville et al., 2002) and proteoglycans containing
heparan or chondroitin sulfate proteoglycan chains (HSPG and
CSPG, respectively) have all been shown to promote HIV-1
attachment and/or entry into cells that lack CD4 (Mondor et al.,
1998). Primary human epithelial cells express HSPG richly both
in vitro and in vivo, and these HSPG efficiently capture HIV-1
particles on the surfaces of the cells (Shafti-Keramat et al.,
2003). Thus, HSPG represent prime candidates for HIV-1
attachment and/or entry into HEC-1A.
In the present study, we examined the contribution of HSPG
and oligosaccharides in the sequential steps of HIV-1
transcytosis through HEC-1A and the ability of HIV-1-exposed
epithelial cells to transfer virus to activated immune cells.
Efficiency of HIV-1 replication in endometrial epithelial cells
The characterization of HEC-1A has been reported else-
where but it had not been confirmed whether these cells may be
infected with HIV-1 (Berlier et al., 2005; Lorin et al., 2005;
Phillips et al., 1994). Our preliminary results indicated that
HEC-1Awere resistant to HIV-1 infection (Hocini et al., 2001).
To determine whether the epithelial cells require higher viral
input to become infected, HEC-1A were exposed to increasing
concentrations of virus. With a viral input of 5, 10 and 20 ng/ml
of R5 virus HIV-1Ba-L, which successfully infected intestinal
cells HT-29 (Fig. 1A), there was no appreciable level of p24
detected during the period of culture of HEC-1 cells (Fig. 1B).
We also tested another R5-virus (HIV-1JR-CSF), X4-viruses
(HIV-1LAIand HIV-1NDK), dual tropic virus (HIV-189.6) and a
non-replicating virus HIV-1NL43ΔENV(Fig. 1C). All viral strains
tested efficiently replicated in HT-29 and PBL, with the
exception of HIV-1NL43ΔENV. We found that the profiles of
p24 in the culture supernatants over the time did not appear to
have any relationship with the viral phenotypes. With all the
isolates tested, the levels of p24 found in HEC-1A supernatants
were significantly lower than 200 pg/ml whereas the levels of
p24 recovered in HT-29 supernatant reached 890±36 pg/ml
with HIVBa-L. To ensure that early replication was not taking
place, HEC-1A were incubated with increasing concentrations
of each virus tested above and levels of HIV DNA were
measured using an HIV DNA real-time PCR assay. Integrated
provirus was not found in HEC-1A incubated with virus.
Moreover, no cytopathic effect was observed in any culture of
cells incubated with each virus. Taken together, these results
showed that HIV-1 is unable to productively infect HEC-1A
Expression of HIV-1 receptor and coreceptors on HEC-1A cells
Since the lack of HIV-1 receptors and/or coreceptors would
account for the inability of HIV-1 to infect these cells, we
analyzed the cell surface expression of CD4, CCR5, CXCR4
and the alternative viral receptor, GalCer. Flow cytometric
analysis showed that HEC-1A lacked CD4 (data not shown) and
CCR5 but did express GalCer and CXCR4 whereas HT-29 cells
expressed GalCer, CCR5 and CXCR4 molecules (Fig. 1D).
Although GalCer and the CXCR4 coreceptor were present on
the HEC-1A cell surface, it apparently did not explain that these
endometrial cells were refractory to productive X4-HIV
Efficiency of HIV-1 capture and sequestering by HEC-1A cells
We speculated that the increasing levels of p24 in the
supernatant were the result of virus attachment to epithelial
cells. To determine the amount of virus bound to HEC-1A, real-
time PCR analysis was performed to quantify the virus released
from HEC-1A that had been exposed to virus. To serve as a
comparison, similar numbers of CD4−HeLa cells, a cervical-
carcinoma epithelial cell line that forms monolayer in culture,
were treated the same way. Indeed, independently of the
expression of CD4, HeLa cells bind HIV with the same
efficiency as HEC-1A cells (Fig. 2A). To remove any viral
particles attached on the surface of the cells, the monolayer was
treated with trypsin. As shown in the Fig. 2B, without trypsin
treatment, HEC-1A cells released 5-fold more virus than did
HeLa cells (11404±2340 HIV RNA/106of HEC-1A incubated
with 20 ng/ml of HIVBa-Lvs. 2300±134 HIV RNA/106of
HeLa incubated with 20 ng/ml of HIVBa-L). HEC-1A treated
with trypsin exhibited a similar level of virus than those
detected in HeLa cultures (Fig. 2B).
Next, we examined the efficiency of HEC-1A to bind R5 and
X4-HIV-1 strains. We did not observe a significant difference in
the efficiency of attachment between 4 and 37 °C except for
HIVNDK, suggesting that the initial attachment of HIV-1 to
HEC-1A was mostly temperature-independent (Fig. 2C).
Interestingly, HEC-1A cells bound X4-HIV strains more
efficiently than R5-HIV strains, and dual-tropic HIV-1 is
weakly bound, suggesting that the efficiency of viral attachment
We evaluated also the efficiency of HEC-1A to mediate
internalization of bound viruses. When HEC-1A cells were
incubated with virus and treated with trypsin, less than 1% of
attached particles penetrated HEC-1A at 4 °C (data not shown).
Experiments were conducted at both 4 and 37 °C, and the
56H. Saïdi et al. / Virology 358 (2007) 55–68
percentage of internalization was calculated by comparison of
the virus bound at 37 °C to virus bound at 4 °C. We showed that
20% of HIV-1LAI, HIV-1NDKand HIV-1JR-CSF, and only 10% of
HIV-1Ba-Land HIV-189.6were internalized, suggesting that 80
to 90% of viruses remained concentrated at the HEC-1A surface
(Fig. 2D). As depicted in the Fig. 2E, PBLs incubated with these
HIV particles released at different times of culture in the apical
pole of HEC-1A cells produced significant amounts of virus,
suggesting that the released viruses are infectious.
Altogether, these results showed that HEC-1A cells
concentrated and sequestered efficiently viruses via trypsin-
sensitive receptors expressed at their surface. Importantly, these
endometrial cells were able to release HIV particles that were
infectious until a period of 9 days.
Implication of heparan sulfate proteoglycans (HSPG) to
mediate virus attachment to the HEC-1A cells
We attempted to determine the cellular mechanisms of viral
attachment to HEC-1A. Given the demonstrated efficacy of
HSPG to function as HIV-1 receptors on CD4-negative primary
human epithelial cells, the role of HSPG, in viral attachment to
HEC-1A, was investigated. To date, there are two major classes
of cell surface HSPG, the syndecans and the glypicans. The
syndecans are transmembrane proteins consisting of an
extended extracellular domain that bears three heparan sulfate
chains (a long way from the plasma membrane) and one or two
chondroitin sulfate chains (close to the plasma membrane). In
contrast, glypicans are GPI-linked proteins consisting of a
globular extracellular domain, which bears two or three heparan
sulfate chains and no chondroitin sulfate chains (Bernfield et al.,
1999; Esko and Lindahl, 2001). To investigate the hypothesis
that proteoglycans are richly expressed on the surface of
epithelial cells (Shafti-Keramat et al., 2003), we evaluated the
ability of heparin to block viral attachment to HEC-1A. The
addition of heparin reduced HIV-1Ba-Land HIV-1NDKattach-
ment to HEC-1A by 41±11% and 50±8%, respectively,
compared with the untreated cells (Fig. 3A). This reduction
was dose-dependent. To further evaluate the implication of
HSPG on HIV attachment on HEC-1A cells, we checked, at
first, whether the removal of the heparan and chondroïtin sulfate
moieties via heparinase III and chondroitinase ABC treatment
diminished the expression of the Syndecan 1 (CD 138) (Fig.
3B). The treatment of HEC-1A cells by heparinase III or
chondroitinase ABC reduced attachment of all HIV-1 strains to
HEC-1A by at least 60%. Chondroitinase ABC treatment
weakly reduced the HIV-1JR-CSFattachment on HEC-1A and
significantly increased HIV-1Ba-L and HIV-1LAI attachment
HEC-1A by 53±2% and 46±16%, respectively, and tended to
facilitate HIV-1NDKattachment on HEC-1A by 20%±12% (Fig.
3C). However, treatment with both heparinase III and
chondroitinase ABC reduced attachment of all HIV-1 strains
to HEC-1A by at least 37%, suggesting that HIVattachment to
HEC-1A is mediated mainly through HSPG. Using phospho-
lipase C to detach all GPI-linked from the cell surface had no
effect, suggesting that glypicans are either absent or weakly
expressed on HEC-1A cells. Interestingly, HEC-1A cells
express high level of CXCR4 but addition of anti-CXCR4
mAbs did not limit the HIVadsorption (data not shown).
HIV-1 transcytosis through a tight monolayer of HEC-1A
Tofurther evaluate theinitial interactions between HIV-1 and
HEC-1A, the HIV-1 transcytosis assay was used. Since we have
shown that HSPG are implicated in HIV-attachment on HEC-
1A, we hypothesized that they may also mediate HIV
transcytosis. However, the removal of the heparan sulfate
moieties, using heparinase, did not interfere with transcytosis of
either R5- or X4-HIV-1 (Fig. 4A). The HIV transcytosis ranged
from 0.15% for HIV-1JR-CSFand HIV-1NDKto 0.51% for HIV-
1Lai, suggesting that, for these viruses, the efficiency of HIV
transcytosis does not depend on the gp120 coreceptor usage
(Fig. 4B). No transcytosis was observed in the case of HIV-
1NL43Δenv, suggesting that the gp120 is required for HIV
transcytosis (data not shown). In contrast to irrelevant
immunoglobulins (data not shown), polyclonal anti-gp160
antibodies inhibited more than 95% of the R5- and X4-HIV-
transcytosis. We evaluated the ability of mannan, D(+) and L(−)
fucose, N-acetylglucosamine, anti-CCR5 and anti-CD40 mAbs
to block viral transcytosis, by exposing HEC-1A to virus in the
presence of these inhibitor candidates. CD40 is strongly
expressed at the cell surface of HEC-1A cells (data not shown)
and is used as negative control. At 50 μg/ml, anti-CD40 mAbs
afforded a 23%, 13% and 27% decrease in HIV-1JR-CSF, HIV-
1NDKand HIV-1LAItranscytosis, respectively. As previously
found (Hocini et al., 2001), at 200 μg/mL, mannan limited 19%,
38% and 25% of the transcytosis of HIVJR-CSF, HIVNDKand
treated with irrelevant immunoglobulins. D(+) and L(−) fucose
and N-acetylglucosamine significantly inhibited transcytosis of
HIV-1JR-CSF, HIV-1NDK and HIV-1LAI in a dose-dependent
manner. Anti-CCR5 mAbs also inhibited significantly transcy-
tosis of HIV-1JR-CSFand HIV-1NDKin a dose-dependent manner
and limited the transcytosis of HIV-1LAI.
Assessment of the role of the basolateral pole of HEC-1A
The HEC-1A cells were grown on a polycarbonate perme-
able support, allowing access to the basolateral pole of HEC-
1A. After HIV transcytosis, the HEC-1A basolateral pole was
treated with heparinase III. The amount of HIV-associated to the
HEC-1A cells was then assessed by quantitating p24 antigen.
As illustrated in Fig. 4C, the removal of the heparin sulfate
moieties expressed at the basolateral pole of HEC-1A
significantly reduced the amount of X4-HIV-1NDKbut not the
amount of R5-HIV-1JR-CSFassociated to the HEC-1A cells.
Resultant effect of the interaction between R5- and X4-viruses
and the epithelial cells on the release of monokines in the
To check whether the HIV attachment on the apical pole of
epithelial cells could influence the secretion of chemokines at
the basolateral pole of HEC-1A. The apical pole of the HEC-1A
57H. Saïdi et al. / Virology 358 (2007) 55–68
was exposed to either R5-HIVJR-CSFor X4-HIVNDK. Under
these experimental conditions, the secretion of chemokines/
cytokines was detected in the basolateral medium by using the
“Proteomic Antibody array”. HIV-stimulated or not HEC-1A
secreted with the same efficiency the inflammatory cytokine
TNFα, the tolerizing cytokine IL-10, the differentiating factor
towards macrophages M-CSF and the monocytes/macrophage-
attracting chemokines such as MCP-2 and MCP-3 (Fig. 5). In
addition, there was no induction of the release of GM-CSF or of
IL-13, suggesting that HEC-1Awould not induce differentiation
of monocytes into dendritic cells. No detectable amount of
CXCR4 ligand SDF-1α was measured under these conditions.
Compared to non-stimulated HEC-1A, HIV-1JR-CSF-stimulated
HEC-1A cells released significantly more: (i) monocyte/
macrophage-attracting chemokines such as IL-8 (63%) and
GRO (53%); (ii) inflammatory cytokines as TNF-β (53%) and
IL-1α (92%), but they released the same amount of the CCR5-
interacting chemokines RANTES. Compared to unstimulated
Fig. 1. Kinetics of HIV-1 release by HEC-1A, HT-29 cells and by PBLs. HEC-1A (A) and HT-29 cells (B) were exposed to increasing amounts of HIVBa-L. (C) HEC-
1A, HT-29 cells and by PBLs were exposed to 20 ng of p24 of HIVBa-L, HIVJR-CSF, HIV89.6, HIVNDK, HIVLAIand HIV-1NL-43Δenv. After infection, levels of p24
antigen in culture supernatants were assessed at the indicated time points. All viral strains efficiently replicated in PBMCs, except HIV-1NL-43Δenv. Each data point
represents mean of duplicate wells, and each experiment was repeated at least 3 times. (D) HEC-1A and HT-29 cells were analyzed for GalCer, CXCR4 and CCR5 cell
surface expressions by FACS. The percentage of positive cells (upper number) and mean fluorescence intensity (MFI, lower number) defined according to the isotypic
control are shown in the upper right corner. One experiment representative of three is presented.
58H. Saïdi et al. / Virology 358 (2007) 55–68
Fig. 2. (A) Attachment of HIVBa-Lby epithelial cells. HEC-1A or HeLa cells were exposed to increasing amounts of HIVBa-L. After 3 h, the cells were washed
thoroughly. To measure the HIV-associated to epithelial cells, each type of cells was lysed with a solutionof Triton X-100 and the concentrationof p24 was determined
by ELISA. Each data point represents mean of replicate wells and the bars represent SEs. (B) Sequestration of virus by HEC-1A cells. HEC-1A or HeLa cells were
exposed to increasing amounts of HIVBa-L. After 3 h, the cells were washed thoroughly, were replaced in medium or were treated with trypsin, washed and recultured.
Culture supernatants were collected 6 days after exposure to virus, and RNA copy was measured by real-time polymerase chain reaction. Each data point represents
mean of replicate wells and the bars represent SEs. (C) Attachment of HIV-1 on HEC-1A cells. HEC-1A cells were exposed to 5 ng of HIV for 1 h at 4 or 37 °C. Cells
were washed and lysed. Amounts of attached virus were determined by p24 ELISA on cell lysates. Results are representative of those from three independent
experiments. (D) Internalization of virus by HEC-1A cells. HEC-1Awere exposed to 5 ng of HIV for 3 h at 4 °C or 37 °C. The cells were then washed and lysed, and
the p24 antigen concentrations were determined. The percentage of internalization was calculated by comparison of the virus bound at 37 °C to virus bound at 4 °C.
Results are representative of those from three independent experiments. The bars represent SEs. (E) Kinetic of infectiousness of viral particles released at the apical
pole of HEC-1A cells. HIVBa-Lparticles released were recovered at 3, 6 and 9 days from the apical pole of the HEC-1 cells. PBLs were infected with the same quantity
of virus (100 pg/106cells) for 3 h. After four washes, PBLs were cultured for 6 days. The levels of p24 antigen in culture supernatants were assessed by ELISA.
59H. Saïdi et al. / Virology 358 (2007) 55–68
HEC-1A cells, HIV-1NDK-stimulated HEC-1A cells released
less TNF-β (38%), IL-1α (100%), IL-10 (20%), RANTES
(42%), IL-8 (22%) and GRO (25%). Altogether, these results
showed that R5- but not X4-HIV enhanced the specific
release of monocyte/macrophage-attracting chemokines and
Fig. 3. Effect of treatment with heparin, heparinase or chondroitinase ABC on the viral attachment on HEC-1A cells. (A) HEC-1A cells were exposed to increasing
amountsof heparinpreviousto add5 ng of HIVBa-Lor HIVNDK. Amountsof attached virusweredetermined byp24 ELISAof cell lysates.Results are representative of
those from three independent experiments. The bars represent SEs. (B) HEC-1A cells were treated with or without heparinase III (Hase III) and chondroitinase ABC
(Case ABC) and further incubated with the isotypic control antibody or the anti-CD138-FITC at 4 °C. Cells were analyzed by flow cytometry as described in Materials
and methods. Histograms represent CD 138 staining (gray) and isotypic control staining (white). Percentage of CD 138 positive cells is depicted in upper right corner
and Mean of Fluorescence Intensity (MFI) is indicated in parenthesis. Results are representative of 3 independent experiments. (C) HEC-1A cells were pretreated with
heparinase III and/orchondroitinase ABCfor 1h, asdescribedinMaterialsandmethods.Thecellswerenotwashedbutthe viruswasaddedfor 1h. Thecellswerethen
washed and lysed, and the p24 was determined. Results are representative of those from three independent experiments. The bars represent SEs.
60 H. Saïdi et al. / Virology 358 (2007) 55–68
Fig. 4. (A) Effect of treatment with heparinase or chondroitinase ABC on viral transcytosis through HEC-1 cells. The apical pole of HEC-1A cells was pretreated with
heparinase III or chondroitinase ABC for 1 h, as described in Materials and methods. The apical pole was not washed but the virus was added for 24 h. Transcytosis
was measured by quantitating p24 antigen in the basal medium. The results are expressed as means±standard deviation of the percentages of transcytosis inhibition
obtained in four separate experiments. (B) Inhibition of transcytosis of cell-free HIV-1 through a tight HEC-1 epithelial barrier by polyclonal antibodies to gp160,
mannan, D(+) or L(−) fucose, N-acetylglucosamine (N-acetylGlcn), monoclonal antibodies anti-CCR5 and anti-CD40. Free virus (5 ng of HIV-1JRCSF, HIV-1NDKand
HIV-1Lai) was added to the apical medium overlying the monolayer of epithelial cells that contained the indicated tested molecule. Transcytosis was measured by
quantitating p24 antigen in the basal medium after 24 h incubation.The results are expressed as means±standard deviation of the percentages of transcytosis inhibition
obtained in four separate experiments. (C) Sequestration of viruses at the basolateral pole of HEC-1A. After the viral crossing, apical and basolateral pole of HEC-1
cells was washedandthe basolateralpole of HEC-1Awas treated with heparinaseIII for 1 h at 37 °C, as describedin Materialsand methods. Amountsof attached virus
on HEC-1Athat weretreated (+ HaseIII-T) or not treated (Without HaseIII-T) weredeterminedby p24 ELISA of cellslysates. Resultsare representative of thosefrom
12 independent experiments.
61 H. Saïdi et al. / Virology 358 (2007) 55–68
Transmission of HEC-1A-released viruses to PBLs
We compared the susceptibility of T cells to HIV infection
using HIV recovered from the apical pole (virus that had not
transcytosed) to that recovered from the basolateral pole
(virus that had transcytosed). The viral solutions were ultra-
centrifuged before their utilization in the subsequent assays.
PBLs were infected with comparable amount of virus (as
confirmed by p24 and viral RNA quantification) and at 72 h
post-infection we measured the HIV DNA content (Fig. 6A).
T cells infected with HIV recovered from the apical pole of
HEC-1A exhibited a 132–143-fold higher integration of
HIVNDKand HIVJR-CSFDNA than T cells infected with the
same quantity of HIV recovered from the basolateral pole of
HEC-1A cells. As compared to the virus collected at the
apical pole of HEC-1A, the virus collected at the basolateral
pole of HEC-1A resulted in a weakly infection of T cells,
whatever the viral strain.
In order to characterize the impact of the HEC-1A super-
natants (HEC-1A-CM) on HIV replication by infected T cells,
we examined the efficiency of these infected Tcells to replicate
X4-HIV-1 (Fig. 6B) and R5-HIV-1 (Fig. 6C), in the presence of
RPMI/FCS, apical or basolateral HEC-1A-CM or medium
collected at the basolateral pole of HIV-stimulated HEC-1A
In RPMI/FCS, T cells infected with HIVJR-CSFrecovered at
the apical pole of HEC-1A cells, produced 53-fold higher
HIVJR-CSFthan T cells infected with HIVJR-CSFrecovered at
the basolateral pole of HEC-1A (649±20 pg/ml vs. 12.5±
6 pg/ml of p24 concentrations, respectively). In the presence
of medium collected at the apical pole of HEC-1A cells, there
was no significant difference observed in HIVJR-CSFproduc-
tion, but the production of HIVJR-CSF was enhanced by 6-
folds in the presence of the medium collected at the
basolateral pole of HEC-1A cells and by 12-folds in the
presence of the HIV-HEC-1A-CM (12.5±6 pg/ml vs. 71±
5 pg/ml and 150±18 pg/ml of p24 concentrations, respec-
tively) (Fig. 6B).
In RPMI/FCS, T cells infected with HIVNDKrecovered at
the apical pole of HEC-1A cells, produced 25-fold more
HIVNDKthan T cells infected with HIVNDKrecovered at the
basolateral pole of HEC-1A cells (651±20 pg/ml vs. 26±
9 pg/ml of p24 concentrations, respectively). In the presence
of medium collected at the apical pole of HEC-1A cells, there
was no significant difference observed in HIVNDKproduction.
However, the production of HIVNDKwas enhanced by 4.7–
4.8-fold in the presence of medium collected at the
basolateral pole of HEC-1A cells or HIV-HEC-1A-CM (26±
9 pg/ml vs. 126±40 pg/ml of p24 concentrations, respectively)
Fig. 5. Differential chemokine and cytokine production by HIVexposed HEC-1A cells. Cells were exposed to HIV for 15 min at 37 °C. Following washes and further
24 h of culture, supernatants of HEC-1A exposed to HIVJR-CSFor HIVNDKwere collected and chemokine/cytokine productions were assessed by Chemokine
Antibody Array I, as described in Materials and methods. Mean of three experiments is presented. *p≤0.05.
62H. Saïdi et al. / Virology 358 (2007) 55–68
Of note, the HIV-HEC-1A-CM and HEC-1A-CM collected
at the basolateral pole of HEC-1A efficiently enhanced the
production of HIV by viruses recovered at the apical pole (data
Taken together, these data indicate that the medium
collected at the basolateral pole of HEC-1A cells stimulated
by R5 viruses enhanced the production of HIV by T cells,
whatever the viral strain, suggesting that this supernatant may
contain soluble factors enhancing the replication of CCR5 and
CXCR4-tropic HIV strains.
We demonstrate herein that the virus detected in the
endometrial epithelial HEC-1A culture supernatant was not
the result of productive infection. Despite HEC-1A cells
expressing the alternative receptor of HIV, the galactosyl-
ceramide and the coreceptor CXCR4, less than 20% of viruses
were effectively internalized. Our observation is consistent with
the general theory that genital epithelial cells are refractory to
cell-free HIV infection. In vivo studies in nonhuman primate
Fig. 6. (A) Differential integration of HIV by PBLs infected by viruses that had transcytosed or not. PBLs were infected with the same amount of HIV-1JR-CSFor HIV-
1NDKthat had recovered from the apical medium overlying the monolayer of epithelial cells or from the basolateral medium. PBLs were harvested for subsequent
integrated HIV-DNA quantification, as described in Materials and methods. Differential replication of HIV by PBLs infected by viruses that had transcytosed or not,
in the presence or in the absence of the basolateral medium of HEC-1A cells. Infected PBLs with HIV-1JR-CSF(B) or HIV-1NDK(C) were cultured in the presence of
PHA and rhIL-2 in RPMI/FCS medium, basolateral medium of HEC-1A (HEC-1A-CM) or basolateral medium of HEC-1A that were stimulated with HIV-1JR-CSF
(HIV-1JR-CSF-HEC-1A-CM) or HIV-1NDK(HIV-1NDK-HEC-1A-CM). Following 12 days of culture, the infected T cells were recovered and their supernatants were
monitored for viral p24 production. Error bars represent standard deviations. One experiment representative of three is presented, *p≤0.05.
63H. Saïdi et al. / Virology 358 (2007) 55–68
models have shown that atraumatic inoculation of cell-free SIV
onto the vagina led to a productive infection of associated
immune cells, but not of epithelial cells (Sodora et al., 1998).
An interesting finding in this study is that the HEC-1A
were capable of concentrating and sequestering significant
amounts of both R5- and X4-HIV-1 at their apical pole.
Because HeLa and HEC-1A cells are of similar sizes and
their membrane-associated proteins do not differ greatly (Wu
et al., 2003), it is probable that this ability is associated with
the content of specific viral-binding proteins on the cells.
Epithelial cells are known to carry large amounts of HSPG
(Shafti-Keramat et al., 2003), and the HEC-1A cell line has a
high level of heparan sulfate on its surface (Wu et al., 2003).
Conversely to chondroitinase ABC and phospholipase C
treatments, heparatinase III treatment significantly inhibited
both R5- and X4-HIV adsorption on HEC-1A, suggesting that
heparan sulfate chains of syndecans expressed far from the
plasma membrane of HEC-1A are mostly implicated on the
viral concentration at their surface. Since cell-surface heparan
sulfate has been shown to mediate HIV-1 attachment by
interacting with viral gp120 (Bobardt et al., 2003; de Parseval
et al., 2005), X4-HIV strains were bound more efficiently
than R5-HIV strains on HEC-1A cells, which is consistent
with previous studies suggesting that R5 gp120 has a lower
affinity for HSPGs than does X4 gp120 (Bobardt et al.,
2003). Specifically, we showed that X4 gp120 binds heparin
with a higher affinity than R5 gp120 (Moulard et al., 2000).
Because HSPG bind the V3 loop of gp120 (de Parseval et al.,
2005), they may also modulate gp120–coreceptor interactions
by altering electrostatic contacts between the positive charges
of the V3 loop and the coreceptor-binding domain of gp120
and the negative charges of the extracellular domain of the
coreceptor, thus preventing the fusion step necessary for viral
entry and the subsequent infection of cells. Even if there was
no release of SDF1-α detected at either the apical and
basolateral pole, we cannot exclude the possibility of the
presence of CXCR4-interacting SDF1-α at the surface of
HEC-1A. Since the removal of chondroitin sulfate chains
accounted for a greater attachment of R5 and X4 viruses, our
results suggest that, in addition to HSPG, other cell surface
molecules may be involved in the initial viral attachment to
HEC-1A cells and the identification of this (or these)
molecule(s) remains to be performed.
To further characterize the initial interactions between cell-
free virus and mucosal epithelial cells, the transcytosis of HIV-1
has been studied. The efficiency of cell-free HIV-1 transcytosis
was extremely low (less than 1%). The HIV-1 mutant lacking
envelope glycoproteins (Δenv) was not capable of transcytosis
and the addition of polyclonal anti-gp160 antibodies blocked
the HIV-1 transcytosis, irrelevant of the viral strain. This is
consistent with previous reports showing that gp120 is required
for HIV transcytosis and that HIV transcytosis does not depend
on gp120 coreceptor usage (Bomsel, 1997; Hocini et al., 2001).
Our results suggest that transcytosis could depend on amino-
acid sequences of gp160 likely not implicated in viral tropism,
but, more viral strains need to be used to support this
It is well established that monosaccharides with hydroxyl
groups equivalent to the three −OH and four −OH equatorial
groups of mannose, such as N-acetylglucosamine and fucose
bind to mannose-type CRDs (Ng et al., 1996). Indeed, mannose
receptors display Ca2+-dependent lectin activity towards
terminal mannose, fucose and N-acetylglucosamine residues
(Lennartz et al., 1987; Wileman et al., 1986). Indeed, we have
found that both fucose and N-acetylglucosamine inhibited R5-
and X4-HIV transcytosis more efficiently than mannan,
suggesting that mannose receptors expressed on epithelial
cells are implicated on HIV transcytosis. However, HEC-1A
cells do not express the macrophage mannose receptor (Nguyen
and Hildreth, 2003) nor DC-SIGN molecule (Geijtenbeek et al.,
2000) (data not shown). Since higher inhibitory activity on HIV
transcytosis was not obtained by pre-incubating these sacchar-
ides with cells (data not shown), both oligosaccharides of the
HIVenvelope and expressed at the HEC-1A cell surface may be
implicated in the HIV transcytosis. A recent study demonstrated
that the heparan sulfated motifs of syndecans, the tyrosine
sulfated motifs of CCR5 and sulfated anti-gp120 antibodies
contain autologous sulfated motifs that permit gp120 contact
(de Parseval et al., 2005). Despite of the absence of CCR5
expression at the HEC-1A cell surface, the use anti-CCR5
mAbs, that can recognize the N-terminus of CCR5 containing
sulfated motifs, blocked 52–62% of the HIVJR-CSFand HIVNDK
transcytosis. However, purified anti-CCR5 antibodies that
recognize the CCR5 peptide CSSHFPYSQYQFWKNFQTLK
corresponding to the second extracellular loop of CCR5 (II.E/C-
CCR5) (Bouhlal et al., 2005) did not inhibit the HIV
transcytosis, suggesting that sulfated motifs of syndecans
could be implicated specifically in the HIV transcytosis.
However, the observation that heparinase treatment of HEC-
1A did not limit the HIV transcytosis strongly suggests that
heparan sulfate moieties are not necessary for the HIV
transcytosis. Since we have shown that anti-CCR5 mAbs
weakly limited the HIVLAI-transcytosis, a simple explanation
for this apparent discrepancy is that some viral strains are able to
use an alternative receptor to cross through HEC-1A cells. At
the basolateral pole of HEC-1A, we found that HSPG were
capable of sequestering significant amounts of X4-HIV-1
exhibiting a higher positive charge on their gp120 than R5-
HIV-1. Since HSPG expressed at the apical pole mediated the
attachment of both R5 and X4-HIV-1, these results suggest that:
(i) R5-HIV-1 has lost their efficiency to bind to HSPG during
their transcytosis or (ii) HSPG expressed at the basolateral pole
do not bind or bind with a weaker affinity R5-HIV-1 than those
expressed at the apical pole. Even though proteases, like
cathepsin D, are present in the vaginal tract, protease inhibitors,
such as secretory leukocyte protease inhibitor, are also present
and can inactivate the locally secreted proteases (Seemuller et
al., 1986), suggesting that these sequestered virus can be
transmitted mostly by cell-to-cell contact between genital
epithelial cells and sub-mucosal permissive cells. Corroborating
this hypothesis, recent studies showed that CD4-negative cells
bind HIV-1 and efficiently transfer the virus to Tcells (Dezzutti
et al., 2001; Wu et al., 2003). However, the endurance of HIV-1
attached to host cells via HSPG remains to be determined.
64 H. Saïdi et al. / Virology 358 (2007) 55–68
After the HIV transcytosis, HEC-1A cells were capable of
efficiently transmitting virus to HIV-target cells recruited at the
basolateral pole of these epithelial cells. Importantly, we have
shown that the stimulation of HEC-1A cells by R5-HIV-1JR-CSF
but not X4-HIV-1NDK significantly enhanced the release of
monocyte/macrophage-attracting chemokines such as IL-8 and
GRO and of inflammatory cytokines as TNF-β and IL-1α.
Kottilil et al. (2006) have recently shown that NK cells
exposed to X4-subtype HIV gp120 showed a significant
decrease in the level of CC chemokines, while exposure to R5-
subtype HIV gp120 had minimal effect. This interesting study
showed that the exposure of NK cells to HIVenvelope proteins
resulted in abnormalities at the level of gene expression. We
showed herein in our epithelial cell model that, with respect to
its tropism, the virus would display such disparate intracellular
signals leading to differential release of monokines. We have
found that HIVJR-CSFstimulated-HEC-1 cells secreted more
efficiently monocytes/macrophages chemokines and cytokines
inducing the differentiation of monocytes into macrophages. It
would be interesting to define the macrophages populations
recruited by the HIV-stimulated HEC-1A cells (Stout and
Our observation is consistent with the study of Dezzutti et al.
(2001), which showed that human primary urogenital epithelial
cells could not be infected with cell-free HIV but are capable of
transmitting the virus to PBMCs in coculture. However, we
have shown that the virus recovered from the basolateral pole of
HEC-1A cells infected with weaker efficiency than those
recovered from the apical pole (the infectivity was reduced
between 25 and 50%), suggesting that, during its crossing
through HEC-1A, the transcytosed virus may (i) incorporate
host cell surface or (ii) lose HIV-1 envelope factors important in
its infectiousness. This observation is in agreement with studies
showing that: (i) sequestered virus was resistant to treatment
with trypsin and (ii) the epithelial cell-associated virus remained
infectious for up to 9 days longer than in DCs (Dezzutti et al.,
2001; Wu et al., 2003). Even if the infectiousness of viruses that
have transcytosed remained weaker than the infectiousness of
virus recovered at the apical pole of HEC-1A, the basolateral
medium, that contains some LTR-stimulating monokines (i.e.
IL-8, IL-6, TNF-α and IL-1), significantly enhances the R5- and
X4-HIV production by activated T cells. DNA microarray
analyses are required to study intra-cellular signals resulting of
the interaction between R5- or X4-HIV virus particles and
HEC-1A cells or primary epithelial cells.
Previous studies have suggested that epithelial cells may
be a reservoir for HIV and may explain why HIV-1 is
transmitted through sexual exposure (Howell et al., 1997). We
have shown herein that (i) X4 viruses attach more efficiently
than R5 viruses to the HEC-1A surface, (ii) the efficiency of
the both cell-free X4- and R5-HIV transcytosis is extremely
low, (iii) there is probably no correlation between coreceptor
usage and transcytosis, (iv) HSPG are implicated in the X4-
and not R5-HIV sequestration at the basolateral pole of HEC-
1A and (v) HEC-1A cells released at their basolateral pole
factors enhancing the X4- and R5-HIV production by
activated T cells. We have found that the specific interaction
of R5- and not X4-HIV-1 induce the release of chemokines
attracting monocytes/macrophages and proinflammatory cyto-
kines that could enhance its spread. It will be interesting to
study the efficiency of these sub-epithelial environments to
recruit HIV-sensitive cells. Given that majority of viruses
transmitted sexually are R5 viruses (Margolis and Shattock,
2006; van't Wout et al., 1994; Zhu et al., 1993), our study
highlights the important role of epithelial cells in the rapid
emergence of R5 viruses by attracting macrophages that are
well known to produce R5 and not X4 viruses. In contrast,
X4-HIV-1 and not R5-HIV-1 could be sequestered at the
basolateral pole of HEC-1A, suggesting that X4 viruses could
be transmitted by cell-to-cell contacts. Previous studies have
shown that the sequestered virus particles remained infectious
for more than 6 days. We cannot exclude the possibility that
HEC-1A could release X4 viruses continuously over the time,
resulting in the emergence of X4-HIV-1 in favorable
environment conditions. A more thorough work is required
to study the environmental conditions that induce the
emergence of R5 or X4 viruses.
Materials and methods
HeLa cells were obtained through the AIDS Research
whereas HT-29 and HEC-1A were obtained from American
Type Culture Collection (Manassas, VA). These epithelial cell
lines were routinely passaged by treatment with trypsin. GalCer
and CXCR4 expression was carefully examined at the time of
attachment, infection or transcytosis assays by FACS analysis.
Peripheral blood lymphocytes cells (PBL) from healthy donors
were isolated by Ficoll-Hypaque gradient centrifugation.
The X4-HIV-1LAIlaboratory adapted strain, the primary X4-
HIV-1NDK, R5/X4-HIV-189.6strains (gifts from Prof. F. Barré-
Sinoussi, Institut Pasteur, Paris, France), the primary R5-HIV-
1JR-CSFand R5-HIV-1Ba-Lwere grown in PBL of healthy donors
stimulated with phytohemagglutinin (PHA) (2.5 μg/ml; Pepro-
tech (Rocky Hill, NJ)) and interleukin-2 (rhIL-2) (1 μg/ml
Peprotech (Rocky Hill, NJ)). The X4-HIV-1NL-43δenv was
obtained from the Reagent Program, Division of AIDS,
Fluorescence-activated cell sorter (FACS) analyses
One million cells were incubated with antibodies (1 μg)
in 500 μl of PBS containing 0.25% bovine fetal serum. Anti-
galactosyl-ceramide (anti-GalCer) MAB342 monoclonal anti-
body was obtained from Chemicon International, Paris, France.
The anti-CD4, anti-CXCR4, anti-CCR5 and anti-CD138
monoclonal antibodies were obtained from R&D Systems
Europe (Abingdon, Oxon, UK). Note that the antibody stain-
ing was performed on adherent cells before prior to cell
65H. Saïdi et al. / Virology 358 (2007) 55–68
stripper (CellGro; Mediatech Inc.) detachment. Analyses were
performed using FACScalibur and CellQuest software (Becton
Dickinson, San Jose, CA) on at least 1000 gated events.
DNA or RNA extraction, quantification of HIV-1 DNA or of
HIV-1 RNA by real-time polymerase chain reaction (RT-PCR)
DNA and RNA were isolated from HIV-infected cells on a
silica column system according to the manufacturer's recom-
mendations (Qiagen DNA or RNA minikit, AG, Basel,
Switzerland). HIV-1 DNA was quantified by using 5′ nuclease
assay in the long terminal repeat (LTR) gene and carried out on
the LightCycler instrument (Roche Applied Science), using the
forward primer NEC152 (GCCTCAATAAAGCTTGCCTTGA)
and the reverse primer NEC131 (GGCGCCA CTGCTAGA-
GATTTT) in the presence of a dually (FAM and TAMRA)
labeled NEC LTR probe (AAGTAGTGTGTGCCCGTC-
TGTTRTKTGACT). Primers and probes were synthetized by
Eurogentec (Eurogentec SA, Seraing, Belgium). The LC-PCR
master mix contained 1× Fast-Start Taq DNA polymerase
reaction buffer (Roche Applied Science), 3 mM MgCl2, 0.3 μM
of each primer and probe. Cycling conditions were as follows:
initial denaturation/FastStart Taq DNA polymerase activation at
95 °C/10 min, 45 cycles of denaturation at 95 °C/10 s, annealing
and extension at 60 °C/30 s with a ramp of 5 °C/s.
The first PCR cycle allowing fluorescence detection
permitted quantification of HIV-1 DNA by reference to a
standard curve (tenfold dilutions of 8E5 cell DNA). All
reactions were performed in triplicate and tested in the same
assay. The level of albumin DNA copies in the cell pellet was
used as an endogenous reference to normalize the variation in
number of cells as previously described (Laurendeau et al.,
HIV-1 RNA quantification was carried out by RT-PCR using
primers (forward: 5′-GGCGCCACTGCTAGAGATTTT-3′;
reverse: (5′-GCCTCAATAAAGCTTGCCTTGA-3′) and exo-
nucelase probe (5′-FAM-AAGTAGTGTGTGCCCGTCT-
GTTRTKTGACT-TAMRA-3′) designed to amplify a fragment
in the Long Terminal Repeat gene. Reverse transcription and
amplification were achieved in a one step RT-PCR using the
LightCycler-RNA Master Hybridization Probes kit (Roche
diagnostics corporation). (A standard graph of the Cp values
the HIV-1 subtype A strain.) The normalized value of cell-
associated HIV-1 DNA and HIV-1 RNA loads corresponding to
the ratio [(HIV-1 copy number/albumin copy number)×2×106]
was finally expressed as the number of HIV-1 DNA or HIV-1
RNA copies per 106cells.
Once epithelial cells formed tight monolayers, HEC-1Awere
stimulated by R5or X4 viruses for 15 min, was washed and then
cultured. After 24 h, the medium at the basolateral pole of HIV-
stimulated HEC-1A cells (HIV-HEC-1A-CM) was collected.
The absence of virus in this basolateral medium was confirmed
by the measurement of HIV RNA by RT-PCR.
Infectivity and p24 assays
The epithelial cells (HEC-1A and HeLa cells) or PBL (105
cells) were exposed to virus (5 ng of p24) for 3 h at 37 °C. Cells
were washed 5 times to remove excess virus particles and were
re-cultured in 200 or 1000 μL. The culture supernatants were
then collected at various time points, were depleted of cellular
debris and were assayed for p24 antigen (HIV-1 p24 ELISA,
Ingen, Belgium) or number of RNA copies measurements. In
several experiments, HIV-1 viruses (100 pg p24), that had
transcytosed or not, were added to PBL (106cells) in the
presence of basolateral HEC-1A supernatant for 3 h at 37 °C.
Each sample was performed in duplicate. After 4 washes to
remove exceeding virus, cells were cultured for 3 days,
harvested and lysed for subsequent HIV-DNA quantification.
Supernatants were collected and p24 concentrations were
monitored by HIV-p24 antigen ELISA. The sensitivity of the
assay was <5 pg/mL p24.
HIV adsorption assay on HEC-1A
HEC-1A cells, seeded at confluency in 48-well plates, were
incubated with or without increasing concentrations of heparin
and then with HIV-1 (5 ng p24). Each sample was performed in
triplicate. In some experiments, to remove cell surface of
heparan and chondroitin sulfates, HEC-1 cells were pretreated
1 h at 37 °C in Hank's balanced salt solution with heparinase III
(5 U/ml from Sigma), phospholipase C (5 U from Sigma) or
chondroitinase ABC (10 U from Sigma), as described
previously (Saphire et al., 2001), and the virus was then
added. After 1 h of incubation at 37 °C or 4 °C, 4 washes
removed unattached virus and cells were lysed by adding Triton
X-100 1% for 45 min at 37 °C. The concentration of HIV p24
antigen was measured by p24 antigen capture ELISA.
To obtain a tight monolayer of polarized epithelial cells,
HEC-1Awere grown in RPMI1640 10% FCS on a 0.4 μm pore
polycarbonate permeable support (Costar, Cambridge, MA)
(Hocini et al., 2001). After 7to 9 days of culture, the tightness of
the monolayer was monitored by measuring resistance
(>300 Ω/cm2) at the apical and basolateral poles of the cells
(Ohmeter, Millicell, Millipore). HEC-1A were incubated with
increasing concentrations of fucose and N-acetylglucosamine
(Sigma Aldrich, USA), anti-CCR5 (clone 45549.111) and anti-
CD40 monoclonal antibodies (AIDS Reagent Program, Divi-
sion of AIDS, NIAID, NIH), and then with free virions (5 ng of
p24) for 1 h. Transcytosis was assessed after 24 h by measuring
p24 antigen concentration in the basal chamber. Positive control
wells contained HEC-1A cells that had been incubated with
anti-gp160 polyclonal antibodies purified by immunoaffinity
from pooled sera of HIV-1 seropositive individuals or mannan
(Sigma Aldrich, USA) before addition of virus whereas
negative control wells contained HEC-1A in RPMI alone or
incubated with irrelevant IgG antibodies. Inhibition of transcy-
tosis was expressed as percentage of p24 antigen recovered in
66 H. Saïdi et al. / Virology 358 (2007) 55–68
the basal chamber in the presence of antibodies, compared with
the amount of p24 antigen recovered in the absence of antibody.
A percentage of inhibition superior to 20% (mean of the values
obtained after pre-incubation with irrelevant IgG plus 3
standard deviations) was considered as significant. In some
experiments, to remove cell surface of heparan and chondroitin
sulfates, HEC-1 cells were pretreated 1 h at 37 °C in Hank's
balanced salt solution with heparinase III (5 U/ml from Sigma)
or chondroitinase ABC (10 U from Sigma), as described
previously (Saphire et al., 2001), and the virus was then added
for 24 h. The concentrations of p24 antigen recovered in the
basal chamber were determined by ELISA. In other experi-
ments, after the viral crossing, apical and basolateral pole of
HEC-1 cells was washed and the basolateral pole of HEC-1A
was treated with heparinase III (5 U/ml) for 1 h at 37 °C. The
basolateral pole of HEC-1 cells was then washed. To determine
the HIV-associated to cells, HEC-1 cells were lysed with Triton
X-100 and the amount of p24 antigen was measured by ELISA.
Detection of the cytokine and chemokine release by human
Polarized HEC-1A were incubated with HIV for 15 min at
37 °C. Cells were washed 4 times and cultured for 24 h. The
basolateral medium was then harvested and analyzed with the
“Human Chemokine Antibody Array Series I” (RayBiotech,
Norcross, GA). Signals were captured by a 2-minute exposure
on a Kodak X-OR film. Signal integration was performed with
NIH version 1.3, membrane negative control values were
subtracted and signal intensities were normalized against the
membrane positive controls.
A paired Student t test was performed for all tests to
determine the statistical significance of the data. p<0.05 was
considered to be the level of statistical significance.
This work was supported in part by the Agence Nationale de
Recherches sur le SIDA, and the Institut National de la Santé et
de la Recherche Médicale (INSERM), France. H.S. was
supported by the EMPRO program of the VI° PCRD. The
authors thank Professor Françoise Barre-Sinoussi, Institut
Pasteur de Paris, for providing HIV-1 strains and Dr Cédric
Carbonneil for helpful discussion and technical assistance.
Competing interests statement: The authors declare that they
have no competing financial interests.
Berlier, W., Bourlet, T., Lawrence, P., Hamzeh, H., Lambert, C., Genin, C.,
Verrier, B., Dieu-Nosjean, M.C., Pozzetto, B., Delezay, O., 2005. Selective
sequestration of X4 isolates by human genital epithelial cells: implication for
virus tropism selection process during sexual transmission of HIV. J. Med.
Virol. 77 (4), 465–474.
Bernfield, M., Gotte, M., Park, P.W., Reizes, O., Fitzgerald, M.L., Lincecum, J.,
Zako, M., 1999. Functions of cell surface heparan sulfate proteoglycans.
Annu. Rev. Biochem. 68, 729–777.
Bobardt,M.D., Saphire, A.C.,Hung,H.C., Yu, X., Van der Schueren,B., Zhang,
Z., David, G., Gallay, P.A., 2003. Syndecan captures, protects, and transmits
HIV to T lymphocytes. Immunity 18 (1), 27–39.
Bomsel, M., 1997. Transcytosis of infectious human immunodeficiency
virus across a tight human epithelial cell line barrier. Nat. Med. 3 (1),
Bouhlal,H., Latry, V., Requena, M., Aubry, S., Kaveri, S.V., Kazatchkine,M.D.,
Belec, L., Hocini, H., 2005. Natural antibodies to CCR5 from breast milk
block infection of macrophages and dendritic cells with primary R5-tropic
HIV-1. J. Immunol. 174 (11), 7202–7209.
Bounou, S., Giguere, J.F., Cantin, R., Gilbert, C., Imbeault, M., Martin, G.,
Tremblay, M.J., 2004. The importance of virus-associated host ICAM-1 in
humanimmunodeficiency virustype1 disseminationdependson the cellular
context. FASEB J. 18 (11), 1294–1296.
Carreno, M.P., Krieff, C., Irinopoulou, T., Kazatchkine, M.D., Belec, L., 2002.
Enhanced transcytosis of R5-tropic human immunodeficiency virus across
tight monolayer of polarized human endometrial cells under pro-
inflammatory conditions. Cytokine 20 (6), 289–294.
Chehimi, J., Luo, Q., Azzoni, L., Shawver, L., Ngoubilly, N., June, R., Jerandi,
G., Farabaugh, M., Montaner, L.J., 2003. HIV-1 transmission and cytokine-
induced expression of DC-SIGN in human monocyte-derived macrophages.
J. Leukocyte Biol. 74 (5), 757–763.
de Parseval, A., Bobardt, M.D., Chatterji, A., Chatterji, U., Elder, J.H., David,
G., Zolla-Pazner, S., Farzan, M., Lee, T.H., Gallay, P.A., 2005. A highly
conserved arginine in gp120 governs HIV-1 binding to both syndecans and
CCR5 via sulfated motifs. J. Biol. Chem. 280 (47), 39493–39504.
Dezzutti, C.S., Guenthner, P.C., Cummins Jr., J.E., Cabrera, T., Marshall, J.H.,
Dillberger, A., Lal, R.B., 2001. Cervical and prostate primary epithelial cells
are not productively infected but sequester human immunodeficiency virus
type 1. J. Infect. Dis. 183 (8), 1204–1213.
Esko, J.D., Lindahl, U., 2001. Molecular diversity of heparan sulfate. J. Clin.
Invest. 108 (2), 169–173.
Geijtenbeek, T.B., Kwon, D.S., Torensma, R., van Vliet, S.J., van Duijnhoven,
G.C., Middel, J., Cornelissen, I.L., Nottet, H.S., KewalRamani, V.N.,
Littman, D.R., Figdor, C.G., van Kooyk, Y., 2000. DC-SIGN, a dendritic
cell-specific HIV-1-binding protein that enhances trans-infection of T cells.
Cell 100 (5), 587–597.
Harouse, J.M., Bhat, S., Spitalnik, S.L., Laughlin, M., Stefano, K., Silberberg,
D.H., Gonzalez-Scarano, F., 1991. Inhibition of entry of HIV-1 in neural cell
lines by antibodies against galactosyl ceramide. Science 253 (5017),
Harouse, J.M., Collman, R.G., Gonzalez-Scarano, F., 1995. Human immuno-
deficiency virus type 1 infection of SK-N-MC cells: domains of gp120
involved in entry into a CD4-negative, galactosyl ceramide/3′ sulfo-
galactosyl ceramide-positive cell line. J. Virol. 69 (12), 7383–7390.
Hocini, H., Becquart, P., Bouhlal, H., Chomont, N., Ancuta, P., Kazatchkine,
M.D., Belec, L., 2001. Active and selective transcytosis of cell-free human
immunodeficiency virus through a tight polarized monolayer of human
endometrial cells. J. Virol. 75 (11), 5370–5374.
Howell,A.L., Edkins,R.D.,Rier, S.E.,Yeaman,G.R.,Stern, J.E.,Fanger, M.W.,
Wira, C.R., 1997. Human immunodeficiency virus type 1 infection of cells
and tissues from the upper and lower human female reproductive tract. J.
Virol. 71 (5), 3498–3506.
Kottilil, S., Shin, K., Jackson, J.O., Reitano, K.N., O'Shea, M.A., Yang, J.,
Hallahan, C.W., Lempicki, R., Arthos, J., Fauci, A.S., 2006. Innate immune
dysfunction in HIVinfection: effect of HIVenvelope-NK cell interactions. J.
Immunol. 176 (2), 1107–1114.
Kozak, S.L., Kuhmann, S.E., Platt, E.J., Kabat, D., 1999. Roles of CD4 and
coreceptors in binding, endocytosis, and proteolysis of gp120 envelope
glycoproteins derived from human immunodeficiency virus type 1. J. Biol.
Chem. 274 (33), 23499–23507.
Laurendeau, I., Bahuau, M., Vodovar, N., Larramendy, C., Olivi, M., Bieche, I.,
Vidaud, M., Vidaud, D., 1999. TaqMan PCR-based gene dosage assay for
predictive testing in individuals from a cancer family with INK4 locus
haploinsufficiency. Clin. Chem. 45 (7), 982–986.
Lennartz, M.R., Cole, F.S., Shepherd, V.L., Wileman, T.E., Stahl, P.D., 1987.
67 H. Saïdi et al. / Virology 358 (2007) 55–68
Isolation and characterization of a mannose-specific endocytosis receptor
from human placenta. J. Biol. Chem. 262 (21), 9942–9944.
Lorin, C., Saidi, H., Belaid, A., Zairi, A., Baleux, F., Hocini, H., Belec, L., Hani,
K., Tangy, F., 2005. The antimicrobial peptide dermaseptin S4 inhibits HIV-
1 infectivity in vitro. Virology 334 (2), 264–275.
Losier, M., Fortin, J.F., Cantin, R., Bergeron, M.G., Tremblay, M.J., 2003.
Virion-bound ICAM-1 and activated LFA-1: a combination of factors
conferring resistance to neutralization by sera from human immunodefi-
ciency virus type 1-infected individuals independently of the disease status
and phase. Clin. Immunol. 108 (2), 111–118.
Margolis, L., Shattock, R., 2006. Selective transmission of CCR5-utilizing
HIV-1: the ‘gatekeeper’ problem resolved? Nat. Rev., Microbiol. 4 (4),
McDougal, J.S., Maddon, P.J., Orloff, G., Clapham, P.R., Dalgleish, A.G.,
Jamal, S., Weiss, R.A., Axel, R.A., 1991. Role of CD4 in the penetration of
cells by HIV. Adv. Exp. Med. Biol. 300, 145–154 (discussion 155-8).
Mondor, I., Ugolini, S., Sattentau, Q.J., 1998. Human immunodeficiency virus
type 1 attachment to HeLa CD4 cells is CD4 independent and gp120
dependent and requires cell surface heparans. J. Virol. 72 (5), 3623–3634.
Moulard, M., Lortat-Jacob, H., Mondor, I., Roca, G., Wyatt, R., Sodroski, J.,
Zhao, L., Olson, W., Kwong, P.D., Sattentau, Q.J., 2000. Selective
interactions of polyanions with basic surfaces on human immunodeficiency
virus type 1 gp120. J. Virol. 74 (4), 1948–1960.
Ng, K.K., Drickamer, K., Weis, W.I., 1996. Structural analysis of monosacchar-
ide recognition by rat liver mannose-binding protein. J. Biol. Chem. 271 (2),
Nguyen, D.G., Hildreth, J.E., 2003. Involvement of macrophage mannose
receptor in the binding and transmission of HIV by macrophages. Eur. J.
Immunol. 33 (2), 483–493.
Phillips, D.M., Zacharopoulos, V.R., Tan, X., Pearce-Pratt, R., 1994.
Mechanisms of sexual transmission of HIV: does HIVinfect intact epithelia?
Trends Microbiol. 2 (11), 454–458.
Prakash, M., Patterson, S., Gotch, F., Kapembwa, M.S., 2003. Recruitment of
CD4Tlymphocytesandmacrophagesintothe cervicalepitheliumof women
after coitus. Am. J. Obstet. Gynecol. 188 (2), 376–381.
Saphire, A.C., Bobardt, M.D., Zhang, Z., David, G., Gallay, P.A., 2001.
Syndecans serve as attachment receptors for human immunodeficiency virus
type 1 on macrophages. J. Virol. 75 (19), 9187–9200.
Seemuller, U., Arnhold, M., Fritz, H., Wiedenmann, K., Machleidt, W., Heinzel,
R., Appelhans, H., Gassen, H.G., Lottspeich, F., 1986. The acid-stable
proteinase inhibitor of human mucous secretions (HUSI-I, antileukopro-
tease). Complete amino acid sequence as revealed by protein and cDNA
sequencing and structural homology to whey proteins and Red Sea turtle
proteinase inhibitor. FEBS Lett. 199 (1), 43–48.
receptors for human papillomaviruses. J. Virol. 77 (24), 13125–13135.
Sodora, D.L., Gettie, A., Miller, C.J., Marx, P.A., 1998. Vaginal transmission of
SIV: assessing infectivity and hormonal influences in macaques inoculated
with cell-free and cell-associated viral stocks. AIDS Res. Hum. Retroviruses
14 (Suppl. 1), S119–S123.
Stout, R.D., Suttles, J., 2004. Functional plasticity of macrophages: reversible
adaptation to changing microenvironments. J. Leukocyte Biol. 76,
Tardif, M.R., Tremblay, M.J., 2005. LFA-1 is a key determinant for preferential
infection of memory CD4+ T cells by human immunodeficiency virus type
1. J. Virol. 79 (21), 13714–13724.
Turville, S.G., Cameron, P.U., Handley, A., Lin, G., Pohlmann, S., Doms, R.W.,
Cunningham, A.L., 2002. Diversity of receptors binding HIV on dendritic
cell subsets. Nat. Immunol. 3 (10), 975–983.
van't Wout, A.B., Kootstra, N.A., Mulder-Kampinga, G.A., Albrecht-van Lent,
N., Scherpbier, H.J., Veenstra, J., Boer, K., Coutinho, R.A., Miedema, F.,
Schuitemaker, H., 1994. Macrophage-tropic variants initiate human
immunodeficiency virus type 1 infection after sexual, parenteral, and
vertical transmission. J. Clin. Invest. 94 (5), 2060–2067.
Wileman, T.E., Lennartz, M.R., Stahl, P.D., 1986. Identification of the
macrophage mannose receptor as a 175-kDa membrane protein. Proc.
Natl. Acad. Sci. U. S. A. 83 (8), 2501–2505.
Wu, Z., Chen, Z., Phillips, D.M., 2003. Human genital epithelial cells capture
cell-free human immunodeficiency virus type 1 and transmit the virus to
CD4+ cells: implications for mechanisms of sexual transmission. J. Infect.
Dis. 188 (10), 1473–1482.
type 1 on human colon epithelial HT29 cells. J. Virol. 66 (8), 4848–4854.
Zhu, T., Mo, H., Wang, N., Nam, D.S., Cao, Y., Koup, R.A., Ho, D.D., 1993.
Genotypic and phenotypic characterization of HIV-1 patients with primary
infection. Science 261 (5125), 1179–1181.
68H. Saïdi et al. / Virology 358 (2007) 55–68