| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Institut National de la Santé et de la Recherche Médicale Unité 255, Centre de Recherches Biomedicales des Cordeliers, Paris, France;
Institut National de la Santé et de la Recherche Médicale Unité 28, Centre Hospitalier Universitaire Purpan, Toulouse, France;
Department of Oral Anatomy, Faculty of Dentistry, Hiroshima University, Hiroshima, Japan; and
Laboratoire d’Immunologie, Schering-Plough, Dardilly, France
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
DC are a family of professional APCs that reside in peripheral tissue in an immature state, optimal for Ag capture. In response to microbial stimuli and inflammatory cytokines, which are most likely produced by T cells and macrophages communicating with the immature DC, the DC mature and migrate to the T cell zone of secondary lymphoid organs (4, 5). Mature DC express high levels of costimulatory molecules and antigenic peptides, bound to MHC class I and II molecules, and are highly efficient T cell stimulators. Shortly after immunization, splenic CD11c+ DC migrate from the MZ bridging channels into the PALS, suggesting that immature DC reside at the bridging channels, and mature DC in the PALS (5, 6). DC have also been implicated in the transport of unprocessed Ag to B cells (7, 8). B cells would subsequently present antigenic peptides to T cells and receive T cell-encoded stimuli in form of cytokines and CD40 ligand (9, 10).
About 4 days following immunization, activated B cells migrate into the follicles to form germinal centers (GC), where they undergo affinity maturation and isotype switch (11, 12). Besides B cells, the GC reaction involves at least three additional cellular components: follicular DC (FDC), Ag-specific Th cells, and tingible body macrophages that act as scavengers for apoptotic lymphocytes. FDC are nonphagocytic and trap native Ag in form of immune complexes on Fc and complement receptors (13, 14, 15). Retention of immune complexes is necessary for a productive GC, as mutations that abolish immune complex retention can severely impair GC reaction. Studies of the mode of transport of immune complexes have suggested that Ag-transporting cells carry immune complexes on the cell surface from the subcapsular sinus to developing GC (16). Using an Fc chimeric protein, containing the cysteine-rich domain of the mouse mannose receptor (CR-Fc), it was observed that mannose receptor ligand+ (MR-L)+ cells migrate from the subcapsular sinus into the developing GC (17). In addition, using the same probe, Berney et al. (8) have shown that MR-L+ subcapular sinus macrophages migrate to the interface of the T cell zone and the follicle and that purified MR-L+ cells differentiate into T and B cell stimulatory DC. Therefore, it appears that MR-L+ subcapular sinus macrophages can differentiate into GC-homing putative Ag-transporting cells and bona fide DC. The splenic counterpart to the subcapsular sinus macrophages appears to be the MZ metallophils, as they also express MR-L (17, 18). However, in response to immunization, MZ metallophils appear not to migrate into the B cell follicle or to the T cell zone (17).
Studies on human tonsils have suggested the presence of T and B cell stimulatory DC in GC (19, 20, 21). However, the cellular and microanatomical origin of GCDC is currently unclear. In the mouse, GCDC have not yet been identified. Decysin, a novel disintegrin metalloprotease, has been cloned from a cDNA library made from ex vivo purified human GCDC. In situ hybridization revealed transcription of decysin in human GC, suggesting that decysin is a novel marker for this specialized DC subset (22). In this study, we have cloned the mouse homologue of decysin and show that in immunized mice it is expressed by CD11c+ DC, FDC, and tingible body macrophages. Strikingly, following immunization, MR-L+ MZ metallophils express decysin and appear to migrate into the developing GC in the course of the immune response. These MZ-derived decysin+ cells are absent in op/op mice that lack functional M-CSF, supporting the idea that the M-CSF-dependent MZ metallophils migrate into the follicle. MZ metallophils may be a precursor for the murine GCDC, and could carry immune complexes from the marginal sinus into the developing GC.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
BALB/c mice were purchased from Iffa Credo (L’Arbresle, France) and used at 8–12 wk of age. A total of 4 x 107 SRBC (Sanofi Diagnostics Pasteur, Paris, France) was injected i.p. Osteopetrotic op/op and control +/? mice were used at 3 wk of age and were immunized i.p. with 10 µg alum-precipitated chicken OVA. BALB/c mice were injected s.c. into hind footpads with 20 µg alum-precipitated OVA (grade V; Sigma, St. Louis, MO) and boosted 2 wk after. Popliteal lymph nodes were removed before secondary immunization (day 0) and 2 days after secondary injection. For each time point, three BALB/c mice and two op/op mice were used, and multiple sections were analyzed with identical results.
Cloning of mouse decysin and plasmid constructions
cDNA was amplified from reverse-transcribed BALB/c mouse spleen mRNA (Clontech, Palo Alto, CA) in a PCR using a primer derived from expressed sequencing tag (EST) AA608260 and an upstream primer derived from human decysin. The product was cloned into a T/A vector (pCR2.1; Invitrogen, San Diego, CA) and sequenced. A primer derived from this sequence was used to amplify the 5' untranslated region using Marathon rapid amplification of cDNA ends (Clontech). The PCR product was directly sequenced from both strands. To obtain the full-length mouse decysin-coding cDNA, the upstream primer (5'-ctcgagaccatgctgcctgggacttctcggc) and the downstream primer (5'-ctcgagttctgtgatgtggtgg) were used in a PCR reaction with DNA polymerase PfuTurbo (Stratagene, La Jolla, CA) and reverse-transcribed BALB/c spleen mRNA as template. After addition of 3'-adenosine overhangs by Taq polymerase (AmpliTaq; PerkinElmer Roche, Norwalk, CT), the PCR product was cloned into the T/A vector (pCR 2.1; Invitrogen) and sequenced.
The insert was liberated at XhoI sites, introduced on the primers, and subcloned into plasmid pIg (R&D Systems, Minneapolis, MN), which directs expression of a fusion protein with the human IgG1 Fc tail. The XhoI restriction fragment was also subcloned into pCDNA3.1 (Invitrogen) in frame with a carboxyl-terminal myc epitope and a 6x histidine (His) tag. Nucleotide sequence of all constructs was verified by double-stranded sequencing.
In vitro expression and furin convertase cleavage
Myc/6x His-tagged decysin was transcribed with T7 RNA polymerase and translated in rabbit reticulocyte nuclear lysate using the TnT-coupled reticulocyte system (Promega, Madison, WI) in the presence of [35S]methionine (Amersham Pharmacia Biotech, Piscataway, NJ). To purify decysin, lysates were added to 200 µl binding buffer (50 mM NaH2PO4, 300 mM NaCl, 10 mM imidazole) containing preequilibrated Ni2+-NTA slurry (Qiagen, Chatsworth, CA). After binding for 1 h at room temperature, beads were pelleted, washed in buffer (50 mM NaH2PO4, 300 mM NaCl, 20 mM imidazole), and resuspended in 20 µl furin convertase cleavage buffer (50 mM HEPES, pH 7.5, 1 mM CaCl2, 0.5% Nonidet P-40). Ten microliters were treated with 2 U of purified recombinant human furin convertase (Alexis Biochemicals, San Diego, CA) at 30°C for 1 and 3 h. Proteins were resolved on a SDS-PAGE and visualized by fluorography.
Expression of Fc-tagged recombinant decysin
A total of 3 x 105 Cos-7 cells was transiently transfected (lipofectamin; Life Technologies, Rockville, MD) with plasmid pIg decysin. Two days later, cells were preincubated for 15 min in methionine/cysteine-free RPMI medium (ICN, Costa Mesa, CA) in the presence of 10% FCS (Valbiotech, Paris, France) and then labeled for 2 h with 150 µCi/ml pro-mix L-35S (Amersham Pharmacia Biotech) (70% [35S]methionine, 30% [35S]cysteine). After a chase of 1 h in RPMI 1640 medium (Life Technologies) containing 10% FCS, pulse and chase growth medium were pooled, and protease inhibitors were added. Cells were washed twice in cold PBS and lysed at 4°C in RIPA buffer (10 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton, 0.5% sodium deoxycholate, 0.1% SDS, aprotinin, PMSF, leupeptin). Cleared cell lysates and growth medium were incubated with protein A-Sepharose CL-4B (Sigma) and then washed with RIPA buffer. Precipitated proteins were resolved by SDS-PAGE and visualized by fluorography.
RNA isolation and RT-PCR analysis
To isolate fresh DC, spleen cells from naive or SRBC-immunized
BALB/c mice were isolated by mechanical disruption in cold
Mg2+- and Ca2+-free PBS and
loaded on a Ficoll gradient (Roche Molecular Biochemicals, Basel,
Switzerland). Mononuclear cells were collected and allowed to settle on
plastic. After 2 h, nonadherent and loosely adherent cells were
detached by washing, incubated with biotinylated anti-B220 mAb
RA3-6B2, washed, and incubated with streptavidin microbeads (Miltenyi
Biotec, Auburn, CA). After two passages over a MiniMacs column
(Miltenyi Biotec), B cell-depleted cells were then positively enriched
for CD11c+ DC by incubating cells with
biotinylated N418 mAb, streptavidin microbeads, and passage over a
MiniMacs column (Miltenyi Biotec). RNA was isolated using
RNAPlus (Quantum, Durham, NC) and reverse
transcribed using the Ready To Go kit (Amersham Pharmacia). RNA from
untreated, LPS-, and TNF-
-stimulated DC line D1 (23)
were kindly provided by S. Amigorena (Institut Curie, Paris, France).
Mouse decysin was amplified by primers 5'-gaggaatgtaccaatctt (sense)
and 5'-tcaccaggattcggctcc (antisense), which generate a 497-bp
product. G3PDH primers were from Clontech.
Antibodies
Anti-decysin rabbit polyclonal Abs (pAb) were raised against the
prodomain peptide 69-NQTERYGKEEKYAPEV-74 and affinity purified
(CovalAb, Ouillins, France). Remaining Abs are listed in Table I
. Isotype-matched control Abs from
PharMingen were used in preliminary experiments to validate the
specificity of the staining. Decysin and FDC-M2 double-labeling
analyses were done in the presence of 10 µg/ml Fc receptor-blocking
Ab 2.4G2 (own production).
|
Acetone-fixed spleen sections were incubated with anti-decysin pAb (0.5 µg/ml) overnight at 4°C in PBS/2% goat serum, or with the following Ab for 1 h at room temperature in PBS/2% goat serum: N418 (5 µg/ml), Moma-1 (1/25 dilution), biotinylated RA3-6B2 (2.5 µg/ml), biotinylated peanut lectin (PNA; 2.5 µg/ml), FDC-M2 (1/200 dilution), FA11 (1/100 dilution). Decysin was revealed by biotinylated goat anti-rabbit Ab (Vector, Burlingame, CA) and alkaline phosphatase (AP)-conjugated avidin-biotin complex (Vector). Color development was in Fast Blue (Vector). For N418, FDC-M2, and FA11, the secondary Ab was HRP-conjugated goat anti-rat IgG (Southern Biotechnology Associates, Birmingham, AL) and developed in diaminobenzidine (DAB; DAKO, Carpenteria, CA). For Moma-1, the secondary Ab was AP-conjugated goat anti-rat IgG (Southern Biotechnology Associates), and color development was in Fast Red (Sigma). Biotinylated RA3-6B2 and PNA were revealed by HRP-conjugated streptavidin, and development was in DAB (DAKO).
In situ hybridization
The protocol is from reference (24). Mouse decysin-coding cDNA was transcribed in antisense direction by T7 RNA polymerase in presence of digoxigenin UTP. Sense probes were used as control and produced unspecific background staining (not shown). The RNA probe was separated from unincorporated label by gel filtration chromatography (Chroma Spin; Clontech) and used for hybridization without prior alkaline hydrolysis. Acetone-treated cryostat sections were fixed in PBS-buffered 4% paraformaldehyde for 10 min at room temperature. RNases were inactivated in PBS containing 1/1000 vol diethylpyrocarbonate (Sigma) for 30 min at room temperature. Sections were washed in 4x SSC and prehybridized for 4 h at 60°C in 4x SSC/50% deionized formamide/10% dextran sulfate/40 µg/ml salmon sperm DNA. The prehybridization solution was then replaced with fresh solution containing the RNA probe and covered with a hydrophobic coverslip (Grace Biolabs, Bend, OR). After overnight incubation at 60°C, sections were washed at 65°C in 2x SSC for 1 h and in 0.1x SSC for 1 h. Sections were then incubated at room temperature for 2 h with anti-digoxigenin AP-conjugated Fab (Roche Molecular Biochemicals) diluted 1/5000 in buffer A (0.1 M Tris-HCl, pH 7.5/150 mM NaCl) containing 0.5% blocking agent (Roche Molecular Biochemicals). After washing in buffer A, development was done in 50 ml buffer B (0.1 M Tris-HCl, pH 9.5, 0.1 M NaCl, 50 mM MgCl2) containing 175 µl 5-bromo-4-chloro-3-indolyl phosphate (BCIP) and 225 µl nitroblue tetrasodium (NBT; Kirkegaard & Perry, Gaithersburg, MD) for between 1 and 2 days at room temperature in the dark. The sections were either counterstained with Fast Nuclear Red (Certistain; Merck, West Point, PA) or subjected to immunohistochemistry. After reequilibration in PBS, sections were incubated with biotinylated PNA (5 µg/ml) in PBS/2% sheep serum, followed by HRP-conjugated streptavidin, and developed in DAB (DAKO). Alternatively, sections were incubated with 6 µg/ml chimeric probe CR-Fc in PBS/2% sheep serum, followed by biotinylated goat anti-human IgG (Southern Biotechnology Associates) and HRP-conjugated avidin-biotin complex (Vector). Development was with DAB (DAKO). All images were recorded by a 3CCD camera (Hamamatsu, Massy, France) and assembled using Adobe Photoshop (Adobe Systems, Mountain View, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Database searching for EST with significant homology to human
decysin identified a nucleotide sequence from mouse colon. The missing
open reading frame was cloned from mouse spleen cDNA using a primer
derived from the EST sequence and a primer of the 5' end of human
decysin. Comparison of the deduced amino acid sequence with human
decysin showed 78% similarity and 65% identity between the two
proteins (Fig. 1). Like human decysin,
the mouse sequence comprises a peptide leader sequence and a
recognition site for furin endopeptidases, which would separate the
N-terminal prodomain from the catalytic domain. The catalytic domain
contains the zinc binding site and a short disintegrin domain. The zinc
binding sites of both sequences comprise aspartate at position 361
(boxed residue), which is a unique feature of decysin among all other
mammalian disintegrin metalloproteases that comprise a histidine
residue at this position (25).
|
To verify that decysin is cleaved by furin convertase,
myc/6x His-tagged mouse decysin was synthesized in vitro in
the presence of [35S]methionine and purified by
Ni2+-chelate affinity chromatography. The labeled
protein was mock treated or incubated with recombinant human furin
convertase, and cleavage products were analyzed by SDS-PAGE. As shown
in Fig. 2A, decysin was
cleaved by furin convertase in a time-dependent manner, resulting in
two clearly visible protein bands corresponding to the 54-kDa
full-length protein and the cleaved 32-kDa catalytic domain. The 22-kDa
prodomain was further processed at an internal cleavage site (Fig. 1
)
and migrated with an apparent molecular mass of 18 kDa. It was
poorly visible, as it contains only two labeled methionine
residues.
|
B). Cell
lysates contained predominantly the unprocessed form, and cell culture
supernatants contained almost exclusively the processed catalytic
domain. This provides evidence that the decysin catalytic domain is
secreted while unprocessed decysin remains intracellular.
A rabbit antiserum was raised against a prodomain peptide and affinity
purified. Western blot of myc/6x His-tagged mouse decysin
produced in Escherichia coli showed that the
anti-peptide Ab recognizes the unprocessed protein, but not a
shorter fragment that migrated with the apparent m.w. expected of the
catalytic domain (Fig. 2C). The anti-myc Ab
recognized both forms.
Decysin is transcribed by DC of the T cell zone
We wondered whether, like its human homologue, mouse decysin is
transcribed by mature DC. The mouse DC line D1 (23) was
stimulated by LPS or TNF-, and the expression of decysin mRNA was
tested by RT-PCR. LPS and TNF- had previously been found to be
effective maturation agents for this cell line (23). As
shown in Fig. 3A, the message
was absent in immature D1 cells, weakly induced by LPS, and strongly
induced by TNF-. Next, CD11c+ DC were purified
from spleens of naive and immunized mice and tested for the presence of
decysin message. Decysin was transcribed in
CD11c+ DC isolated from naive mice, but its
message was up-regulated 1 day after immunization with SRBC (Fig. 3A). The presence of decysin mRNA in DC from naive mice was
most likely due to DC activation in the course of cell manipulation. To
further analyze the induction of the decysin message in DC as well as
to explore its expression profile in situ, we performed in situ
hybridization on spleen sections from naive mice and 1 day after
immunization with SRBC. As shown in Fig. 3B, in naive
spleen, no decysin transcripts were detected in the white pulp,
comprising the PALS, that surrounds the central arteriole, and the
follicles. Also the MZ, which delineates the white pulp, and the
bridging channel (asterisk) showed no significant signal above
background. A weak signal was visible in the red pulp close to the MZ,
which may reflect a low level of transcription by red pulp macrophages.
The expression profile drastically changed 1 day after immunization
(Fig. 3C). Now, cells in the PALS clearly transcribed
decysin, which correlated with the immigration of
CD11c+ DC from the bridging channels (asterisk)
into the PALS (Fig. 3D). In contrast, the
CD11c+ DC residing at the bridging channel
expressed little, if any, decysin mRNA. Taken together, these data
provide evidence that decysin transcription is induced by immunization,
and strongly suggest that decysin is transcribed by mature mouse
CD11c+ DC.
|
At day 1, a cell population located in the
Moma-1+ MZ also transcribed decysin (Fig. 3, C and D). To test whether these cells were MZ
metallophils, spleen sections 1 day after immunization were double
stained for decysin mRNA and MR-L, a marker for MZ metallophils
(17). As shown in Fig. 4A, almost all
decysin-transcribing cells expressed MR-L and were located at the inner
MZ. Few decysin+ MR-L+
cells were found within the follicle (arrowheads). To further assess
whether these cells migrate into the B cell follicle during the immune
response, spleen sections 2 days after immunization were stained for
decysin mRNA and MR-L (Fig. 4B). Now, cells expressing
decysin mRNA and MR-L were clearly seen in the B cell follicle
(arrowheads). To analyze whether the decysin+
cells migrate toward a developing GC, spleen sections taken from mice 2
and 4 days after immunization were stained for decysin using the
anti-decysin anti-peptide pAb and PNA. Two days after
immunization, decysin+ cells were positioned
between the MZ and a developing PNA+ GC (between
arrowheads) (Fig. 4C). Four days after immunization, the
decysin+ cells were no longer seen at the MZ, but
were distributed within the GC (arrowheads) (Fig. 4D). Taken
together, these data suggest that shortly after immunization,
MR-L+ MZ metallophils express decysin and home
into the developing GC in the course of the immune response.
|
E), these cells were absent in
op/op mice (Fig. 4F). The remaining
decysin+ cells were FDC (see below). Thus, the
decysin+ follicle-homing cells are M-CSF
dependent, supporting the notion that they are derived from MZ
metallophils.
To further substantiate these findings, we investigated whether in the
lymph node decysin is transcribed by the MR-L+
subcapsular sinus macrophages (8, 17). Lymph nodes were
taken from BALB/c mice before (Fig. 4G) or 2 days after
(Fig. 4H) secondary immunization with OVA and stained for
decysin mRNA and MR-L. Before immunization, the decysin-transcribing
cells were almost exclusively localized in the subcapsular sinus area,
and the majority expressed MR-L (Fig. 4G). Occasionally,
some decysin mRNA+ cells were seen in the
follicle (arrowheads). Two days after immunization, the subcapsular
sinus macrophages translocated from the subcapsular sinus toward the
center of the lymph node and continued to transcribe decysin (Fig. 4H, arrowheads). Staining of adjacent sections confirmed
that the MR-L+ decysin+ had
entered the B cell follicle (data not shown). These results extend the
observations made in the spleen by showing that also in the draining
lymph node, decysin is transcribed by the specialized
MR-L+ macrophages that home into the B cell
follicle in response to immunization with T cell-dependent
Ags.
Decysin is expressed by FDC and tingible body macrophages
Immunized op/op mice showed decysin expression in the
follicular center (Fig. 4, E and F). To address
the question of whether activated FDC express decysin, spleen sections
of BALB/c mice 4–6 days after immunization were analyzed for decysin
mRNA and protein expression. In situ hybridization revealed, besides
decysin transcription around the central arteriole, decysin
mRNA in distinct areas close to the periphery of the white pulp, where
GC were expected (Fig. 5A).
Immunohistochemical staining with the anti-decysin pAb and
anti-B220 showed decysin+ cells in the B cell
follicle (Fig. 5B), and further double-label analysis showed
decysin+ cells closely associated with
PNA+ GC B cells (Fig. 5C). The
decysin-expressing cells were tightly associated with GC B cells and
formed a network throughout the outer side of the GC, which was
reminiscent of FDC. Double-label analysis with the FDC-specific marker
FDC-M2 confirmed that decysin was expressed by FDC, creating a black
overlay staining in the center of the FDC-M2+
network (Fig. 5D).
|
E). To test
whether these cells were GCDC (19) or tingible body
macrophages, GC were double stained for decysin and macrosialin (CD68),
a protein expressed by tingible body macrophages, but not GCDC
(19). With exception of the FDC network, all
decysin+ cells expressed macrosialin (arrows,
Fig. 5F). Moreover, the double-labeled cells were large
cells and appeared to contain a number of apoptotic bodies, not stained
by the Abs, which was a typical feature of tingible body macrophages.
The tingible body macrophages did not express CD11c (data not
shown). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) and
showed that, like genuine metalloproteases, it is processed by furin
proteases (Fig. 2). However, decysin is distinct from the many other
mammalian disintegrin metalloproteases 1) by the replacement of the
conserved histidine residue by an aspartate in the catalytic site and
2) by its being secreted.
Human decysin is transcribed by mature DC and in tonsilar GC
(22). By the use of in situ mRNA hybridization and
immunohistochemistry with a pAb specific for the intracellular
prodomain, we have identified the cell types that express decysin
during an immune response in the mouse (Table II). In the spleen of nonimmunized mice,
decysin is not transcribed in the white pulp or the MZ. However, 1 day
after immunization, decysin is clearly transcribed in the PALS (Fig. 3). It had previously been shown that the majority of
CD11c+ DC presenting a peptide derived from hen
egg lysozyme bound to MHC-II are located in the PALS, whereas DC of the
MZ present little peptide (27). This observation suggests
that immature DC process Ag at the MZ bridging channels, migrate into
the PALS where they mature, and present antigenic peptides to T cells.
The decysin transcription pattern correlates with the migration of
mature DC from the MZ bridging channels and suggests that decysin is
expressed by mature CD11c+ DC. In concordance,
the unstimulated DC cell line D1 does not transcribe decysin, but its
transcription is up-regulated after maturation with LPS or TNF-.
Maturation of DC is induced by microbial products and by cytokines such
as IFN-
, IL-1, GM-CSF, and TNF-, which are most likely produced
by T cells and macrophages, communicating with the immature
DC.
|
and 4). In the course of the immune response, these MZ
decysin+ cells appeared to migrate into the B
cell follicle (Fig. 4). The majority of the MZ
decysin+ cells carry MR-L, recognized by the
CR-Fc probe, which is expressed by MZ metallophils (8, 17, 18). op/op mice, which lack functional M-CSF, are
devoid of decysin expression in the MZ, which is in accordance with
decysin being expressed by the M-CSF-dependent MZ metallophils (Fig. 4)
(26). However, as some MZ-derived
decysin+ cells lacked CR-Fc binding, we cannot
exclude that other cells, such as MZ macrophages, express decysin. On
the other hand, it might be that MR-L is lost in the course of the
immune response to release mannosylated Ag bound to the MR-L for uptake
by B cells or FDC. In the lymph node, decysin is expressed by
MR-L+ subcapsular sinus macrophages, which also
migrate into the follicle in response to immunization (Fig. 4)
(8, 17).
Berney et al. (8) have shown that
MR-L+ subcapsular sinus macrophages expressed low
levels of CD11c. Upon immunization, the cells migrate toward the outer
B cell follicle and associate with B and T lymphocytes. Purified and
reinjected into mice, they were able to prime T cells and induce the
production of Ag-specific IgM and IgG1 (8). This suggests
that lymph node MR-L+ subcapsular sinus
macrophages can differentiate into T and B cell stimulatory DC,
features previously described for human interdigitating DC and GCDC
(20, 28). This raises the interesting possibility that
also splenic MR-L+ decysin+
metallophils differentiate into DC in the outer follicle and GC.
However, we could not use decysin as a specific marker for these DC, as
tingible body macrophages and FDC also expressed this gene (Fig. 5),
thus masking detection of potential murine GCDC.
It has recently been observed in TNF-- and p55TNF-R-deficient mice
that the FDC-M2 mAb stained a cell population in the MZ, and that it is
capable of binding immune complexes (29). The authors
proposed the existence of a MZ-derived FDC population. It is possible
that the decysin+ MZ-derived cells may be
identical to these putative FDC precursors.
It has been suggested that cells close to the subcapsular sinus are
involved in the transport of immune complexes into the developing GC
(16). The cells are likely to be nonphagocytic, as they
carry immune complexes on the cell surface. It is probable that these
Ag-transporting cells are derived from the MR-L+
decysin+ subcapsular sinus macrophages, as they
home into the follicle in response to immunization (Fig. 4) (8, 17). MZ metallophils are poorly phagocytic and are ideally
positioned at the MZ-follicle interface to transfer antigenic material
from the marginal sinus into the follicles (3). The
obervation that MR-L+ MZ metallophils express
decysin and migrate into the follicle in response to immunization
raises the possibility that MZ metallophils may transport Ag into the
GC. Whether the cells then differentiate into DC or FDC-like cells
remains to be determined. As mice commonly contain natural Abs that
recognize hemagglutinins, the immunogen (SRBC) is likely to be
transported in the form of immune complexes bound to Fc or complement
receptors. It would be interesting to investigate the role of MZ
metallophils in Ag transport using op/op mice.
In addition to DC and MZ metallophils, we identified two further cell
types that expressed decysin in response to immunization: FDC and
tingible body macrophages (Fig. 5). As decysin is not expressed by FDC
in naive mice, and few genes are known to be expressed by murine FDC,
decysin may represent a useful marker for activated mouse FDC. Our
inability to detect mouse GCDC using decysin as a probe could be due to
their masking by the FDC network and tingible body macrophages. It may
also be that GCDC are restricted to human GC, which are much larger and
may require DC to stimulate T and B lymphocytes in GC.
The anti-decysin pAb specific for the prodomain recognizes decysin
expressed in the MZ-derived cells, FDC, and tingible body macrophages,
but not in DC of the PALS (data not shown). We have verified the
specificity of the anti-decysin pAb in mice deficient for decysin,
in which the Ab does not produce any cell staining (data not shown). As
the prodomain may be rapidly degraded after furin-dependent processing
of the zymogen, it is possible that the anti-prodomain Ab only
recognizes unprocessed decysin. This may imply that DC process decysin
more effectively and/or degrade the prodomain more rapidly than other
cells, a likely feature of DC, given their efficient endoprotease and
exoprotease machinery involved in Ag processing and presentation.
Whether decysin itself plays a role in Ag processing is not known.
Alike other members of its family (30), it is probable
that decysin forms heterodimers with other disintegrin metalloproteins.
This would enable it to exert different functions, specific for the
different cell types that express decysin. For example, the disintegrin
metalloprotease TNF--converting enzyme (ADAM 17) not only cleaves
TNF-, but also TGF-, and may be involved in the processing of
L-selectin and amyloprotein precursor (31, 32, 33).
Interestingly, disintegrin metalloproteases have been shown to cleave
complement component C3 (34), which plays a key role in GC
formation (35, 36, 37). An attractive hypothesis would be that
after secretion into the GC environment, decysin cleaves complement
component C3 and thus terminates GC reaction.
Despite the key role that GC play in shaping the humoral immune response, exemplified by the susceptibility of hyper IgM patients to infectious diseases (38), GC formation still remains far from being fully understood. Efforts aimed at defining the molecules that regulate cellular interactions in the GC reaction should eventually provide us with a molecular framework necessary to understand the complex cellular cross-talk involved in the formation of the secondary B cell response.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Chris G. F. Mueller, Institut National de la Santé et de la Recherche Médicale Unité 255, Centre de Recherches Biomedicales des Cordeliers, 15 rue de L’Ecole de Médecine, 75006 Paris, France. E-mail address: chmuller{at}infobiogen.fr
3 Abbreviations used in this paper: MZ, marginal zone; AP, alkaline phosphatase; BCIP, 5-bromo-4-chloro-3-indolyl phosphate; CR-Fc, cysteine-rich domain of the mouse mannose receptor; DAB, diaminobenzidine; DC, dendritic cell; EST, expressed sequencing tag; GC, germinal center; MR-L, mannose receptor ligand; NBT, nitroblue tetrasodium; pAb, polyclonal Ab; PALS, periarteriolar lymphoid sheath; PNA, peanut lectin.
Received for publication May 29, 2001. Accepted for publication August 24, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
from cells. Nature 385:729.[Medline]
. Nature 385:733.[Medline]
This article has been cited by other articles:
|
|
 |
|
  G. Chattopadhyay, Q. Chen, J. Colino, A. Lees, and C. M. Snapper Intact Bacteria Inhibit the Induction of Humoral Immune Responses to Bacterial-Derived and Heterologous Soluble T Cell-Dependent Antigens J. Immunol., February 15, 2009; 182(4): 2011 - 2019. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. J. McKenzie, P. R. Taylor, R. J. Stillion, A. D. Lucas, J. Harris, S. Gordon, and L. Martinez-Pomares Mannose Receptor Expression and Function Define a New Population of Murine Dendritic Cells J. Immunol., April 15, 2007; 178(8): 4975 - 4983. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Vermijlen, P. Ellis, C. Langford, A. Klein, R. Engel, K. Willimann, H. Jomaa, A. C. Hayday, and M. Eberl Distinct Cytokine-Driven Responses of Activated Blood {gamma}{delta} T Cells: Insights into Unconventional T Cell Pleiotropy J. Immunol., April 1, 2007; 178(7): 4304 - 4314. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Beattie, C. R. Engwerda, M. Wykes, and M. F. Good CD8+ T Lymphocyte-Mediated Loss of Marginal Metallophilic Macrophages following Infection with Plasmodium chabaudi chabaudi AS J. Immunol., August 15, 2006; 177(4): 2518 - 2526. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Grdic, L. Ekman, K. Schon, K. Lindgren, J. Mattsson, K.-E. Magnusson, P. Ricciardi-Castagnoli, and N. Lycke Splenic Marginal Zone Dendritic Cells Mediate the Cholera Toxin Adjuvant Effect: Dependence on the ADP-Ribosyltransferase Activity of the Holotoxin J. Immunol., October 15, 2005; 175(8): 5192 - 5202. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. R. Taylor, S. Zamze, R. J. Stillion, S. Y. C. Wong, S. Gordon, and L. Martinez-Pomares Development of a specific system for targeting protein to metallophilic macrophages PNAS, February 17, 2004; 101(7): 1963 - 1968. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
