| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,**
* Department of Molecular Histopathology, and
|| Immunology, Department of Pathology, University of Cambridge, United Kingdom; Departments of
Microbiology and
# Medicine, University of Pennsylvania, Philadelphia;
The Wistar Institute, Philadelphia, Pennsylvania; and
Dept. of Microbiology, Immunology & Molecular Genetics, UCLA School of Medicine, and
** UCLA AIDS Institute, Los Angeles, California
Correspondence: Benhur Lee, M.D., Dept. of Microbiology, Immunology & Molecular Genetics, UCLA School of Medicine, and UCLA AIDS Institute, 3825 Molecular Sciences Building, 609 Charles E. Young Drive East, Los Angeles, California 90095-1489. E-mail: benhurL{at}microbio.ucla.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: lectin • HIV • BDCA-2 • Th1/Th2 • plasmacytoid
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In addition to serving as a universal attachment factor for the primate lentiviruses and mediating DC-T-cell interactions via ICAM-3 binding, DC-SIGN may also play a role in DC trafficking. Recently, DC-SIGN was shown to bind ICAM-2 with very high affinity [5 ]. It was suggested that this high-affinity binding would allow DC-SIGN-expressing DCs to roll along the surface of endothelium, prior to emigration from vessels [5 ]. Although there has been preliminary demonstration of DC-SIGN expression in some lymphoid tissues and skin [3 ], the putative role that DC-SIGN plays in DC trafficking [5 ], T-cell costimulation [2 ], and the biology of primary HIV infection [3 , 4 ] warrants a more detailed examination of the expression, regulation, and control of DC-SIGN expression in vivo. Because the antigenic environment and immune status of the fetus are different from the adult, a comparison between DC-SIGN expression in adult and fetal tissue through its many gestational stages may shed further light on the biological function of DC-SIGN. Also, the realization that DCs are comprised of functionally distinct subsets in the blood and tissues [6 , 7 ] presents an opportunity to study if DC-SIGN expression cosegregates with any of these DC subsets. These results would provide insights into the role of DC-SIGN in the immunology of antigen presentation.
Recently, a closely related homolog to DC-SIGN, DC-SIGNR, was cloned [8 ]. We and others [8 , 9 ] demonstrated that it has similar ligand-binding properties to DC-SIGN. Initial data using DC-SIGNR-specific antiserum indicate that DC-SIGNR expression is much more restricted than DC-SIGN and that DC-SIGN and DC-SIGNR expression are not coincident [8 , 9 ]. Of note, the high conservation in the extracellular domains of these two molecules suggests that previous analysis of DC-SIGN expression can be complicated by cross-reactivity with DC-SIGNR. Indeed, one of the monoclonal antibodies (mAb) previously used to detect DC-SIGN expression in tissues has been shown to cross-react with DC-SIGNR [9 ]. Thus, for focused studies on DC-SIGN, we generated a specific antiserum to the unique carboxy-terminus of DC-SIGN, which can be used in immunohistochemical and fluorescein-activated cell sorter (FACS)-based studies.
Here, we show that DC-SIGN expression is relatively restricted to cells of the DC/macrophage lineage in human adult and fetal tissues. Peripheral blood mononuclear cells and monocyte-derived macrophages (MDM) were consistently negative for DC-SIGN-surface expression under a variety of standard culture conditions. The vast majority of peripheral blood DC (PBDC) precursors was also negative for DC-SIGN expression, except for a subpopulation of BDCA-2+ PBDC, previously suggested to represent a class of plasmacytoid DC precursors (pDC2) [10 ]. These results were supported by finding large numbers of DC-SIGN+/BDCA2+/CD123+ DCs in allergic nasal polyps, previously shown to be infiltrated by DCs with DC2 phenotypes [11 ]. Consistent with this observation, we also found that the T-helper cell type 2 (Th2) cytokine, interleukin (IL)-13, can induce expression of cell-surface DC-SIGN on MDM. These data provide insights into the biological function of DC-SIGN and suggest that DC-SIGN expression may have specific roles in certain immunological responses.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-Cter) were directly conjugated to fluorescein
isothiocyanate (FITC) or Alexa-488 according to the manufacturer’s
directions (Molecular Probes, Eugene, OR). As a negative control for
cell-surface staining, polyclonal antibodies were made against the
N-terminal (intracellular) portion of DC-SIGN (-Nter; amino acids
1–15), affinity-purified and conjugated to FITC or Alexa-488 in a
similar fashion. Absorbance calculations (A280/A494) indicated that
-Cter and -Nter antibodies were conjugated equivalently (
5
fluorochrome molecules per antibody molecule).
Immunoprecipitation
HEK 293 T cells were transfected with DC-SIGN, DC-SIGNR, or
pcDNA3 by the standard calcium-phosphate method. After overnight
expression, cells were labeled in six-well plates for 4 h with 300
µCi total [35S]cysteine and
[35S]methionine (NEN Life Science Products, Boston, MA)
in cysteine/methionine-free Dulbecco’s modified Eagle’s medium (DMEM)
supplemented with 10% dialyzed fetal calf serum (FCS; NEN Life Science
Products). After radiolabeling, cells were washed once with
phosphate-buffered saline (PBS), lysed in 600 µl standard RIPA buffer
(per well) with 1x protease inhibitor cocktail (CompleteTM, Roche
Molecular Biochemicals, Indianapolis, IN), and immunoprecipitated with
8 µl -Cter sera and protein A/G beads overnight (Santa Cruz
Biotechnologies, Santa Cruz, CA). Beads were pelleted and washed three
times with RIPA buffer supplemented with 500 mM NaCl on the last wash.
Sodium dodecyl sulfate (SDS) sample buffer was added to the beads,
heated at 55°C for 1 h, and analyzed by SDS-polyacrylamide gel
electrophoresis (PAGE). The gel was exposed to film for 48 h at
-80°C.
Selecting and obtaining tissue and tissue processing
All tissues were obtained with Local Research Ethics Committee
approval from the Department of Histopathology, Addenbrooke’s
Hospital, Cambridge, UK. Anonymized, histologically normal tissue was
provided from the following adult organs: lymph node, spleen, tonsil,
brain, skin, vulva, vagina, cervix, endometrium, ovary, testis, buccal
mucosa, larynx, bronchus, lung, oesophagus, stomach, duodenum, colon,
and kidney. A similar range of postmortem tissues and also the thymus
and placenta were obtained from fetuses of the following gestations: 12
weeks, 18 weeks, 23 weeks, 36 weeks, and 39 weeks. Adult nasal polyp
tissue was also obtained. Tissue was snap-frozen and kept at -80°C
before being processed to cryosections or was fixed in 10%
neutral-buffered formalin, followed by paraffin-wax embedding and
sectioning.
Single immunostaining of paraffin sections
Sections were immunostained with rabbit anti-DC-SIGN polyclonal
serum, with preimmune serum on serial sections as a negative control.
Further serial sections were immunostained with one of the following
mouse mAb (mmAb): anti-CD1a (Novocastra, Newcastle upon Tyne, U.K.),
anti-CD14 (Novocastra), anti-CD56 (Novocastra), anti-CD68 (Dako,
Glostrup, Denmark), anti-CD79a (Dako), anti-S100 (Dako), and
anti-HLA-DP/DQ/DR (HLA-II; Dako).
Rehydrated, 5 µm paraffin sections were pressure-cooked for 3 min in sodium citrate buffer (Dako) prior to staining. The sections were preblocked in 0.5% hydrogen peroxide in Tris-buffered saline (TBS) for 30 min and then in TBS/1% bovine serum albumin (BSA)/10% normal goat serum for 2 h. Following incubation overnight in primary antibody in TBS/1% BSA, sections were washed with TBS and incubated for 2 h in biotinylated goat anti-rabbit or biotinylated goat anti-mouse antibody (Dako). The SABC kit (Dako) was used to form an avidin-biotin-horseradish peroxidase (HRP) complex or an avidin-biotin-alkaline phosphatase (ALP) complex. Slides were developed with diaminobenzidine (brown) or diaminobenzidine/nickel (black) for HRP (Vector Laboratories, Burlingame, CA) or with Sigma Fast Red (Sigma Chemical Co., Dorset, U.K.) for ALP and counterstained with haematoxylin (Dako). Slides were dehydrated to xylene and mounted in DePX (BDH Laboratory Supplies, Poole, U.K.) or mounted aqueously in Faramount aqueous-mounting medium (Dako). Positive control tissues for the mmAb were lymph node (CD14, CD68, CD79a, HLA-II,), endometrium (CD56), peripheral nerve (S100), and skin (CD1a). For each tissue, a negative control was included using preimmune rabbit serum or omitting the primary antibody.
Staining frozen sections
Cryosections (10 µm) were fixed for 5 min in 4%
paraformaldehyde and then immunostained with anti-CD3 (Dako), anti-CD50
(anti-ICAM-3; Dako), anti-CD83 (PharMingen, San Diego, CA), anti-CD86
(PharMingen), anti-cmrf-44 (courtesy of D. Hart, Mater Institute for
Medical Research, Brisbane, Australia) [12
], and rabbit
anti-DC-SIGN polyclonal serum. Procedures and controls were as
described above. The positive control tissue for all mAb used on frozen
sections was lymph node.
Double-immunostaining sections
Double immunostaining was performed on formalin fixed and frozen
tissue, using rabbit anti-DC-SIGN serum in combination with each of the
mmAb described above. For each antibody pair, serial tissue sections
were also stained with each antibody singly and, as a negative control,
stained with preimmune rabbit serum and no mouse primary antibody.
Procedures were as above until the primary antibody stage, where both
primary antibodies were added together overnight. Following washing in
TBS, both secondary antibodies were added together. These were Envision
HRP-tagged goat anti-mouse antibody (Dako) and biotinylated goat
anti-rabbit antibody. Following washing, slides were incubated in the
streptavidin-biotin-ALP complex (Dako) and then developed first with
Sigma Fast Red (Sigma Chemical Co., Dorset, U.K.) for ALP, followed by
diaminobenzidine/nickel (black) for HRP (Vector Laboratories),
counterstained with haematoxylin (Dako), and mounted aqueously in
Faramount aqueous-mounting medium (Dako).
Preparation of monocyte-derived macrophages (MDM) and DCs
MDM were cultured from adherent monocytes as previously
described [13
]. MDM were differentiated in RPMI 1640
with human serum alone (5% pooled AB serum) plus macrophage
colony-stimulating factor (M-CSF; 50 units/ml) or granulocyte M-CSF
(GM-CSF; 10 ng/ml). On day 7, MDM were treated with lipopolysaccharides
(LPS; 10 µg/ml; Sigma Chemical Co., St. Louis, MO) or IL-13 (20
ng/ml) for 24 h before FACS analysis or processing for reverse
transcriptase-polymerase chain reaction (RT-PCR). Monocyte-derived DCs
(MDDCs) were generated as described previously [14
].
Briefly, the monocyte-enriched fraction from discontinuous Percoll
gradient centrifugation of peripheral blood was magnetically depleted
of B (CD19), T (CD2), and natural killer (NK; CD16, CD56) cells (Dynal,
Lake Success, NY) before culturing the purified monocytes (>95% CD14)
in AIM V serum-free media (Life Technologies, Rockville, MD),
supplemented with GM-CSF (50 ng/ml) and IL-4 (R&D Systems, Minneapolis,
MN; 100 ng/ml).
FACS analysis
293 T cells were cotransfected with a DC-SIGN expression vector
and a green fluorescent protein (GFP) reporter gene as a transfection
marker using standard calcium-phosphate transfection protocols.
Forty-eight hours post-transfection, cells were harvested by mechanical
dislodgment in 1 x PBS, and 5 x 105 cells were
each stained with 24 µg/ml affinity-purified -Nter or -Cter in
FACS block buffer [1xPBS, 2.5% FCS, 0.5% BSA, 0.02% sodium azide,
100 µg/ml human immunoglobulin G (IgG)]. In parallel, cells were
first permeabilized using the IntraPrepTM kit (BD Pharmingen, San Jose,
CA) before staining with an equivalent amount of -Nter or -Cter.
The primary rabbit antibodies were detected by biotinylated goat
anti-rabbit antibodies (Dako), followed by allophycocyanin-conjugated
(APC) streptavidin (Caltag, Burlingame, CA). Specificity of DC-SIGN
staining was determined by comparing the DC-SIGN fluorescence profile
on gated GFPhigh versus GFPneg cells in the
same sample. FACS analysis was performed on a FACSCalibur machine
(Becton Dickinson, San Jose, CA) using Cell QuestTM software (Becton
Dickinson) for data analysis.
For analysis of PBDC precursors, 250 µL whole blood was lysed selectively in ammonium chloride (PharMlyse, PharMingen), and the remaining leukocytes were washed twice with FACS buffer (1xPBS, 2.5% FCS, 0.5% BSA, 0.02% sodium azide) and blocked for 15 min on ice in 100 µl FACS buffer with 10% normal rabbit serum and FcR block (Miltenyi Biotec, Auburn, CA). The indicated antibodies were added directly to the blocking buffer, and cells were stained for 20 min before washing twice with FACS buffer and fixed in 2% paraformaldehyde. All directly conjugated antibodies used were as described previously [15 ], except for the Alexa-488-conjugated rabbit polyclonals against DC-SIGN described above and the phycoerythrin-conjugated BDCA-2 and BDCA-3 antibodies (Miltenyi Biotec, Auburn, CA) [10 ].
DC-SIGN RT-PCR
MDM were isolated as described above. MDM at days 0, 3, and 7
were harvested. LPS treatment was performed by incubating the cells
with 1 µg/ml LPS for 6 h. RNA was then isolated using
TRI-Reagent from Molecular Research Center (Cincinnati, Ohio).
RNA (1 µg) was reverse-transcribed using the RETROscript kit (Ambion,
Austin, TX). One-tenth of the resultant cDNA products were used for the
multiplex PCR reactions. A multiplex PCR reaction was set up according
to the manufacturer’s instructions with the following modifications:
An 18S RNA primer pair (5 µM; 18 S competimer/18 S primer in the
ratio of 7/3) was added as an internal control to normalize the DC-SIGN
mRNA expression level along with 1 µl 32P-dCTP. Forward
and reverse primers for DC-SIGN were GCCACCCCTGTCCCTGGGAATG (DCSIGNFr3)
and TAAAGGTCGAAGGATGGAGAGAAG (3UTRRv), respectively. The PCR reaction
was run for 26 cycles with a magnesium concentration of 1.5 mM, and PCR
products were separated on a 6% polyacrylamide gel. The dried gel was
exposed using a Phospho-Imager (Molecular Dynamics, Sunnyvale, CA).
Images were captured with the Storage Phospho Screen (Molecular
Dynamics), and the ImageQuant software (Molecular Dynamics) was used to
calculate the 32P counts of the DC-SIGN and 18S RNA
signals. The 18S RNA counts of all samples were divided by that of the
untreated day 0 samples to obtain a value representative of the
relative sample loading between time points (i.e., lanes). Normalized
DC-SIGN counts were then obtained by dividing DC-SIGN counts in each
lane by the value representative of the relative sample loading.
Triple-immunofluorescent staining for confocal microscopy
Triple immunofluorescence was performed on cryostat
sections fixed in 4% paraformaldehyde. Sections were blocked with 10%
goat serum/10% swine serum/1% BSA in TBS. Sections were incubated for
1 h in CD123 (Santa Cruz Biotechnology), -Cter or CD11c (Dako),
and -Cter, followed by washing in TBS and incubation for 1 h in
secondary antibodies, which were Alexa-633-conjugated goat anti-mouse
(Molecular Probes) and FITC-conjugated swine anti-rabbit (Dako).
Sections were then incubated for 1 h in directly conjugated mmAb
diluted in 1% BSA/TBS. These were phycoerythrin-conjugated BDCA-2 and
phycoerythrin-conjugated BDCA-3 (Miltenyi Biotec, Bisley, U.K.).
Finally, slides were rinsed in TBS and mounted in fluorescent-mounting
medium (Dako). As a positive control, lymph-node staining was performed
in parallel with the staining of nasal polyps. As controls, single
immunostaining with each antibody was performed in turn, as was
staining with preimmune rabbit serum and with secondary antibodies in
the absence of primary antibodies.
Confocal microscopy
Images were obtained by using a confocal laser-scanning
microscope TCS 4D (Leica Lasertechnik, Heidelberg, Germany). All triple
immunostaining was photographed using sequential scanning techniques.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-Cter) and N-terminal
(intracellular) 15 amino acids (-Nter) of DC-SIGN. Specificity of
the antisera was first demonstrated by immunoprecipitation of
35S-labeled cell extracts from cells transfected with
DC-SIGN, the closely related homolog DC-SIGNR, or the empty vector
pcDNA3. Figure 1 A
shows that a specific band of approximately 45 kDa was
immunoprecipitated only from the DC-SIGN-transfected cells.
|
-Nter and -Cter antibodies were directly
conjugated to Alexa-488 and used in subsequent FACS-based experiments.
MDDCs were stained with 24 µg/ml Alexa-488-conjugated -Cter or
-Nter antibodies. Figure 1B
shows that -Cter detected the
high level of DC-SIGN expression on these cells that has been described
previously [2
]. -Nter antibodies gave no appreciable
background staining. The positive staining by -Cter antibodies was
abrogated completely by use of an excess of unconjugated -Cter serum
but not by an excess of -Nter serum, providing further confirmation
of the epitope specificity of the antibodies used. To confirm that the
-Nter antibody used was fully active against DC-SIGN, we stained
DC-SIGN-transfected cells and showed that -Nter stained positively
only when cells were permeabilized (Fig. 1C)
. As expected, -Cter
antibodies stained permeabilized and unpermeabilized DC-SIGN
transfectants.
Finally, to confirm the specificity of staining of tissue
sections by -Cter, we demonstrated that staining of DCs in duodenum,
cervix, and lymph node and staining of macrophages in the alveoli and
decidua could be blocked completely by using an excess of the peptide
immunogen (Fig. 2
; see below).
|
-Cter rabbit serum to study the tissue distribution
of DC-SIGN in histologically normal tissue from human adult and fetus.
In the adult, large numbers of DC-SIGN+ cells were present in
lymphoid organs such as tonsil, spleen, and lymph nodes (Fig. 2e and 2f)
. DC-SIGN expression was restricted entirely to cells with a
dendritic morphology in T-cell areas and within lymphoid sinuses.
DC-SIGN expression on cells with a dendritic morphology was also seen
in the majority of tissues, particularly at mucosal surfaces, such as
duodenum, colon, gallbladder, bronchus, cervix, endometrium, and
bladder (Fig. 2a
and 2c
; unpublished results). The dendritic
morphology of these DC-SIGN+ cells is clearly illustrated by the
enlarged view of the section of cervix (Fig. 2k)
. It is interesting
that the DC-SIGN-expressing cells observed in lung (Fig. 2g)
had a
nondendritic, macrophage-type oval morphology and an interstitial
distribution within the alveoli (alveolar air spaces are marked with
arrows), as is accepted for alveolar macrophages. Only alveolar
macrophages are found within alveoli (airspaces) within the lungs, and
lung DCs are located within the connective tissue that makes up
alveolar septa [16
, 17
]. Double
immunostaining demonstrated these cells to be CD68+ CD3- CD56- CD79-
(see below). DC-SIGN expression was also observed on decidual
macrophages (Fig. 2i)
. We have demonstrated previously that DC-SIGN+
decidual macrophages express high levels of CD14 and have a macrophage
phenotype [18
]. Small numbers of DC-SIGN+ cells were present in the dermis or lamina propria of mucosal areas covered with squamous epithelium, including skin, buccal mucosa, oesophagus, larynx, vagina, and vulva, although epidermal Langerhans cells were negative for DC-SIGN, as described previously [3 , 19 ]. Within mucosae covered by columnar epithelium, such as the duodenum, large numbers of DC-SIGN+ cells were present in the lamina propria and in associated, specialized areas of lymphoid tissue (Fig. 2a) . In keeping with the accepted absence of lymphoid tissue from normal stomach wall [20 ], DC-SIGN+ cells were absent from the gastric mucosa. No expression of DC-SIGN could be found within testis, ovary, or brain (unpublished results).
In fetal tissues, a similar distribution of DC-SIGN expression (with the exception of lung) was observed to that in the adult at all gestations examined. DC-SIGN+ cells with dendritic morphology were present in lymph nodes, thymus, spleen, large and small intestine, and bladder and were associated with islands of hematopoietic tissue in liver and more sparsely in fibrous connective tissue (Fig. 3a 3b 3c 3d ). Again, no expression was seen within ovary or testis. An interesting observation was the differing tissue distribution in lung. In early gestations at the pseudo-glandular and canalicular stages where alveoli are not yet formed [21 ], small numbers of DC-SIGN+ DCs were present, scattered between the bronchioles (unpublished results). However, once alveoli developed, the alveolar-macrophage population, even in term fetuses, showed no appreciable DC-SIGN expression (Fig. 3e) , although DC-SIGN expression was seen on the macrophages of adult lung (Fig. 2g) . This absence of DC-SIGN expression in fetal lung was not because of the lack of alveolar macrophages, because CD68 staining reveals numerous alveolar macrophages in the fetal lung (Fig. 3f) .
|
|
DC-SIGN expression on MDM
Because DC-SIGN expression was detected in tissue macrophages
resident in the lung and decidua (Fig. 2a
and 2i
; ref.
[18
]), we wanted to determine if DC-SIGN was also
expressed in cultured MDM. Surprisingly, FACS analysis indicated that
MDM were uniformly negative for cell-surface DC-SIGN regardless of
culture conditions (normal human serum alone or with the addition of
M-CSF or GM-CSF) or LPS stimulation (Fig. 5 B
; unpublished results). However, RT-PCR and phosphoimage analysis
(see Materials and Methods) indicated that DC-SIGN mRNA increased by
approximately fivefold by culture day 7 (Fig. 5A)
and was responsive to
down-regulation by LPS. Because we could detect DC-SIGN expression in
specialized tissue macrophages (e.g., placenta) but not in cultured
MDM, DC-SIGN expression may be regulated by distinct microenvironmental
cues. Indeed, there have been previous studies describing that IL-13, a
Th2 cytokine, is expressed by syncytiotrophoblast and cytotrophoblast
cells in the placenta [23
, 24
]. This fact,
coupled with our observation that large numbers of DC-SIGN-positive
cells were present in allergic nasal polyps (see Fig. 8
), an accepted
Th2-mediated disorder [11
], prompted us to examine the
effects of a Th2 cytokine (IL-13) on the expression of DC-SIGN in
7-day-old MDM. It is interesting that although 24-h exposure to GM-CSF
did not alter surface DC-SIGN expression, exposure of MDM to IL-13 for
the same period (Fig. 5B)
resulted in a distinct increase in surface
DC-SIGN expression (<2% DC-SIGN+ cells to 30% DC-SIGN+ cells).
|
|
The electronic noise generated when using such a large cocktail of antibodies, coupled with our desire to investigate subsets of these blood DCs, led us to use the newly described markers for plasmacytoid (BDCA-2) and myeloid (BDCA-3) PBDC subsets [10 ] without having to resort to the multiple antibody cocktails traditionally used to identify DC subsets in blood [27 , 28 ]. We found that a reproducibly small percentage (4–14%) of plasmacytoid PBDC (BDCA-2-positive) precursors had high levels of DC-SIGN expression, and BDCA-3-positive myeloid PBDC precursors were uniformly negative for DC-SIGN (compare Fig. 6A and B). Although only a small fraction of BDCA-2-positive cells are DC-SIGN-positive, all seven donors showed a similar pattern of DC-SIGN+ cells only in the BDCA-2-positive gate (Fig. 6C ; P=0.04). Expression of DC-SIGN on the BDCA-2-positive gate was confirmed further by specificity controls (Fig. 7 ). It is interesting that the DC-SIGN-positive subset of BDCA-2-positive PBDC precursors also exhibited low levels of CD86 expression in some donors (Fig. 7) .
|
|
(CD123) on their cell surface
[29
, 30
]. When matured in vitro with IL-3
and CD40L, these cells acquire the ability to induce the secretion of
allergy-promoting Th2 cytokines from naïve T cells
[31
]. Recently, Jahnsen et al.
[11
] demonstrated that large numbers of DC2 were
found infiltrating the nasal mucosa in experimentally induced allergic
rhinitis and suggested that DC2 may be involved in triggering airway
allergy. Therefore, to substantiate our findings that DC-SIGN was
expressed on a DC2 subset in situ, we looked for evidence of DC-SIGN
expression on DCs infiltrating allergic nasal polyps. Figure 8 A
shows that a large number of DC-SIGN-expressing cells with dendritic
morphology were seen infiltrating nasal polyp tissue. We confirmed that
these cells were indeed DC2s by triple-labeled confocal microscopy.
Figure 8B shows that DC-SIGN+ cells in the nasal polyps were also
BDCA-2+- and CD123+-positive. Note that these were frozen sections, and
thus, the morphology of the cells was not as well-preserved as that
seen in Figure 8A
. BDCA-3 and CD11c, markers for DC1s, did not show
appreciable staining in allergic nasal polyp tissue (Fig. 8C)
, further
underscoring the specificity of the staining seen in Figure 8B
.
Lymph-node tissue was included as a positive control for all antibodies
(unpublished results). Thus, DC-SIGN expression on the DC2 subset was
confirmed in the peripheral blood and at an effector site with direct
antigen exposure (nasal mucosa). |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
DC-SIGN was expressed on cells with dendritic morphology at all mucosal surfaces and lymphoid tissues investigated in the adult and fetus. With the exception of lung tissue (see below), an identical distribution was seen in the fetus, suggesting that no environmental or antigenic stimuli is required for DC-SIGN expression on DCs. In fetal thymic tissue, DC-SIGN was expressed on large numbers of cells with dendritic morphology, within the cortex and medulla. Because DCs in the thymus participate in the process of negative selection of self-reactive T cells and thus have a distinct role from peripheral DCs [32 ], it is possible that the function of DC-SIGN in the thymus and periphery may differ. Recent studies have also indicated that the human thymus contains at least two DC populations and plasmacytoid DC precursors [33 ] that are functionally and anatomically distinct. Whether DC-SIGN cosegregates with these DC subsets and how it contributes to their biological function remain matters for future studies.
In an attempt to understand the origin of such DC-SIGN+ cells in
tissues, we investigated whether cells expressing DC-SIGN could be
demonstrated in peripheral blood. DC-SIGN expression on cells in blood
may also have significant implications for the dissemination of HIV
around the body. Although we found that
lineage-negative/HLA-DR-positive DC precursors were generally
DC-SIGN-negative in blood, the use of mAb specific for precursor DC1
(BDCA-3) and DC2 (BDCA-2) subsets in the peripheral blood
[10
] allowed us to identify a subset of DC2s expressing
high levels of DC-SIGN. Additional flow-cytometric analysis indicated
that these BDCA-2+/DC-SIGN+ cells were also HLA-DRlow
(unpublished results), similar to thymic plasmacytoid
(IL-3R-positive) DC precursors, which have also been described to
have an HLA-DRlow phenotype [33
].
We substantiated our evidence for DC-SIGN expression on DC2s by our
demonstration of a high level of DC-SIGN expression on cells with
dendritic morphology in allergic nasal polyp tissue, where large
numbers of IL-3R (CD123)high DC2s have been demonstrated
previously [11
]. We confirm that DC-SIGN-positive cells
in these allergic nasal polyps also coexpress CD123 and BDCA-2, markers
characteristic of DC2s. To our knowledge, this is the first study
presenting in situ evidence that DC-SIGN is colocalized with markers of
a particular DC subset. Although the vast majority of DC-SIGN+ cells in
this tissue is also BDCA-2+ and CD123+, this is not true in lymph-node
tissue, where DC-SIGN+ cells can be found that are not positive for
BDCA-2 or CD123 (unpublished results). Thus, although our findings that
the Th2 cytokine, IL-13, can induce expression of DC-SIGN on MDM
suggest that DC-SIGN may be involved in the cascade of Th2-mediated
immunity, this may not be a generalizable phenomena. Future comparative
analysis of DC-SIGN expression in tissues from classical Th1- versus
Th2-mediated diseases (e.g., Crohn’s disease vs. ulcerative colitis)
will allow for a better determination of the role of DC-SIGN in the
regulation of specific immune responses.
Our demonstration of DC-SIGN expression on decidual macrophages (Fig. 2i ; ref. [29 ]) and here on alveolar macrophages may have important implications for HIV pathogenesis. Besides the transinfection function of DC-SIGN, we have shown recently that DC-SIGN mediates high-efficiency infection of permissive CD4+ coreceptor+ cells when expressed in cis [34 ]. Macrophages are known to express CD4 in addition to a plethora of the major (CCR5 and CXCR4) and alternate coreceptors (CCR2 and CCR3) for HIV infection. Thus, DC-SIGN expression on these cells may make establishment of a viral reservoir more efficient in times of increased selection pressures, such as HAART treatment (low viral load), or when coreceptor antagonist treatment becomes a clinical reality.
Although we were unable to demonstrate surface DC-SIGN expression in macrophages under a range of standard conditions, IL-13 treatment of macrophages led to a significant increase in surface DC-SIGN expression. It is interesting that high levels of IL-13 have been shown previously to be associated with the placenta [23 , 24 ], and we have found numerous DC-SIGN-positive macrophages in the decidua (Fig. 2 ; ref. [29 ]). Therefore, we propose that the tissue microenvironment in vivo may regulate the very restricted expression of DC-SIGN by macrophages. The observation that fetal alveolar macrophages do not show DC-SIGN expression, even at term, and adult alveolar macrophages do, also argues that exposure to environmental stimuli is important in DC-SIGN macrophage expression. The expression of an erstwhile-DC-specific marker on alveolar macrophages has precedence in that DEC-205, a mannose receptor-like molecule highly expressed on mouse DCs [35 36 37 ], is also expressed on a subset of murine alveolar macrophages [38 ].
In summary, by using a highly specific antibody for DC-SIGN that does not cross-react with DC-SIGNR, we have demonstrated here the expression of DC-SIGN on subpopulations of DC precursors and specialized tissue macrophages. We have also presented functional evidence that DC-SIGN expression on macrophages may be induced by the Th2 cytokine, IL-13. Our observations of DC-SIGN expression on infiltrating DCs in allergic nasal polyps and its expression on candidate plasmacytoid DC precursors (DC2) in blood warrant future investigation into the role DC-SIGN may play in the trafficking and function of DC subsets during Th2 immune responses. The dichotomy between DC-SIGN expression on cellular subsets in situ and on in vitro-cultured cells reinforces the notion that in vitro culture conditions may not always recapitulate the microenvironment in tissues. Thus, the biological significance of DC-SIGN expression in the immunology of DC-T-cell interaction and the biology of HIV-1 infection remain to be determined in an in vivo setting.
|
ACKNOWLEDGEMENTS |
|---|
Received October 1, 2001; revised October 23, 2001; accepted October 31, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Related Articles
This article has been cited by other articles:
|
|
 |
|
  G. Cannon, Y. Yi, H. Ni, E. Stoddard, D. A. Scales, D. I. Van Ryk, I. Chaiken, D. Malamud, and D. Weissman HIV Envelope Binding by Macrophage-Expressed gp340 Promotes HIV-1 Infection J. Immunol., August 1, 2008; 181(3): 2065 - 2070. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Yamakawa, C. Pennelegion, S. Willcocks, A. Stalker, N. MacHugh, D. Burt, T. J. Coffey, and D. Werling Identification and functional characterization of a bovine orthologue to DC-SIGN J. Leukoc. Biol., June 1, 2008; 83(6): 1396 - 1403. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. J. Tacken, B. Joosten, A. Reddy, D. Wu, A. Eek, P. Laverman, A. Kretz-Rommel, G. J. Adema, R. Torensma, and C. G. Figdor No Advantage of Cell-Penetrating Peptides over Receptor-Specific Antibodies in Targeting Antigen to Human Dendritic Cells for Cross-Presentation J. Immunol., June 1, 2008; 180(11): 7687 - 7696. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. Higgins, J. Sanchez-Campillo, A. G. Rosas-Taraco, J. R. Higgins, E. J. Lee, I. M. Orme, and M. Gonzalez-Juarrero Relative Levels of M-CSF and GM-CSF Influence the Specific Generation of Macrophage Populations during Infection with Mycobacterium tuberculosis J. Immunol., April 1, 2008; 180(7): 4892 - 4900. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Serrano-Gomez, E. Sierra-Filardi, R. T. Martinez-Nunez, E. Caparros, R. Delgado, M. A. Munoz-Fernandez, M. A. Abad, J. Jimenez-Barbero, M. Leal, and A. L. Corbi Structural Requirements for Multimerization of the Pathogen Receptor Dendritic Cell-specific ICAM3-grabbing Non-integrin (CD209) on the Cell Surface J. Biol. Chem., February 15, 2008; 283(7): 3889 - 3903. [Abstract]< |
