| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Dermatology, University of Freiburg, Germany;
Department of Biophysical Chemistry, Biocenter Basel, Basel, Switzerland; and
Academic Transfusion Medicine Unit, Department of Medicine, University of Glasgow, United Kingdom
Correspondence: C. Termeer, M.D., Department of Dermatology, University of Freiburg, Hauptstr. 7, D- 79104 Freiburg, Germany. E-mail: Termeer{at}haut.ukl.uni-freiburg.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
production and
inhibition of CD154-induced MHC class II up-regulation. However,
co-incubation with HA-degrading enzymes induced no changes in the
CD154-mediated DC clustering and maturation.
Key Words: hyaluronic acid • MHC • TNF- • IL-1ß • CD154
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(TNF-) and interleukin (IL)-1ß
[2
, 3
]. A potent, maturational signal is
delivered by the ligation of the surface receptor CD40 by its
counterreceptor CD154, expressed on activated platelets, basophils, and
T-memory cells [1
, 4
, 5
]. CD40
belongs to the TNF receptor family [6
], and
CD40-mediated signals are especially important for the priming of
CD8-mediated immune responses against viral or tumor antigens
[2
, 4
, 6
, 7
].
CD40-mediated maturation, in contrast to stimulation by
lipopolysaccharide (LPS) or TNF-, induces prolonged DC survival and
resistance to the induction of apoptotic cell death by FAS-L (CD95) and
MHC class II [8
]. Furthermore, DC maturation by CD40
ligation results in the production of high levels of IL-12 and
polarizes CD4 T-cell responses preferentially to the activation of
interferon-
(IFN-) and IL-2 producing TH1 versus
TH2 cells [3
, 9
]. In vivo
studies have underlined the potential relevance of CD40-CD154
interaction for the treatment of T-cell-mediated diseases, because a
blockade of CD40 dramatically prolonged skin and cardiac allograft
survival in mice [10
]. The first critical step for the direction of the immune response is the DC activation at the site of inflammation, followed by their migration through the lymphatic vessels to the regional lymph nodes [11 ]. During this process, DC will come into close contact with components of the cell extracellular matrix (ECM) and will form clusters within the lymphatic vessels. During DC maturation, all of these functions are mediated by adhesion molecules, with many of them highly expressed on DC, such as intercellular adhesion molecule-1 (ICAM-1), CD11b, CD18, and CD44 [11 ]. We were especially interested in CD44, because it has been proposed to modulate the immune function of DC [12 , 13 ]. CD44 is a polymorphic surface glycoprotein, which can be modified by N- and O-linked glycosylation and alternative RNA splicing [12 , 14 , 15 ]. The major ligand for CD44 is hyaluronan (HA), a macromolecular polysaccharide of a molecular weight up to 1 Mda, which forms widespread networks in the ECM, thus providing a supporting structure for the migration of DC [16 ]. In previous work, we could demonstrate that CD44 is up-regulated on activated human monocytes and mediates an increased hyaluronate-binding capacity [17 , 18 ]. In addition, an activation-dependent induction of CD44-mediated HA binding was found on CD34+ human hematopoetic progenitor cells and T lymphocytes. Up-regulated CD44 expression was accomplished not only with HA binding but also with migratory processes and cell-cell interaction of stromal cells [19 , 20 ]. Indeed, on activated Langerhans cells, which are the DC of the epidermis, we found strong up-regulation of CD44. Blocking CD44 on these cells inhibited the emigration from the epidermis into the lymphatic vessels and could also block a contact hypersensitivity response in vivo [21 ].
Besides its function as an adhesion molecule for cell-cell and cell-ECM interactions, CD44 has been described as a costimulatory factor delivering signals independent of CD28 in CD40-activated B cells to T cells [22 ]. In addition, CD44 was shown to transduce a HA-dependent costimulatory signal to activated human peripheral blood T cells and B cells [23 , 24 ]. Moreover, the interaction of CD44 and HA was shown to induce mitogen-activated protein (MAP) kinase-dependent back-signaling in bone marrow-derived macrophage-like cells, leading to secretion of IL-11ß [25 ].
Therefore, we wished to get further insight into the processes leading to DC maturation using a model of CD154-expressing murine fibroblasts and human monocyte-derived DC [1 ]. Time-lapse video microscopy was applied to investigate the impact of CD44 and its ligand HA on the clustering and maturation of DC. We found that the standard form of CD44 (CD44s) is strongly up-regulated 3–8 h after CD40 ligation. Preincubation of DCs with anti-CD44s mAbs inhibited further maturation, revealing a function for CD44s during the homotypic cell aggregation occurring at the early steps of DC maturation.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cell lines
The murine fibroblast cell lines L-M, from C3H mice,
CD154-expressing or mock-transfected control cells, were obtained via
ATCC.
Generation of human DC
Monocyte-derived DC were prepared as described previously
[26
]. In brief, CD14+ cells were isolated from human
buffy coats by immunomagnetic selection using the MACS®
magnetic-sorting system (Milteny, Bergisch-Gladbach, Germany) with a
bead-labeled anti-CD14 mAb (Milteny). These cells were cultivated for 4
days in IMDM containing granulocyte-macrophage colony-stimulating
factor (GM-CSF) and IL-4 to generate immature DC. For CD154-mediated
stimulation, 2 x 106 DC were cultivated on monolayers
of CD154-expressing murine fibroblasts in IMDM for the indicated
periods at 8% CO2 and 37°C.
Flow cytometry
Cells were stained in phosphate-buffered saline (PBS; Gibco) at
4°C for 30 min with unconjugated primary mAb [antigen-specific or
control immunogloblin G (IgG)] followed by fluorescein isothiocyanate
(FITC)-labeled (Fab)2 fragments of goat anti-mouse IgG
(Dianova, Hamburg, Germany) and analyzed with a FACScan using the
CellQuestTM software (Becton Dickinson, Heidelberg, Germany). Propidium
iodide (5 µg/ml; Sigma, Deisenhofen, Germany) was added to exclude
dead cells by appropriate gating. A total of 104 viable
cells per sample were analyzed, and mean fluorescence intensities (MFI)
were determined using CellQuest® software (Becton Dickinson).
Corrected MFI were calculated according to the following formula:
MFI(antigen-specific mAb) - MFI(control
mAb) = cMFI(antigen-specific mAb).
Human HAS-2 synthetase reverse transcriptase-polymerase chain
reaction (RT-PCR)
mRNA was isolated from human DC, freshly isolated human
keratinocytes, or HACAT cells according to standard procedures using a
quick-prep RNA isolation kit (Pharmacia, Upsala, Sweden). cDNA
synthesis was conducted in a 100 µl reaction mix containing 2 µg
total cellular RNA, 0.4 U/µl RNAse inhibitor, 1 mM dNTP, 125 pmol
random hexanucleotid primer, and 200 U superscript RT (all Gibco). The
cDNA samples (15 µl) were added to a 50µl reaction mix containing
200 nM of each oligonucleotide primer (Genescan Europe, Freiburg,
Germany), 200 µM dNTPs (Boehringer Mannheim, Indianapolis,
IN), 1.5 mM MgCl2, and 1 µl Taq DNA polymerase
(Gibco) in 1x reaction buffer. Amplification was performed in a DNA
thermal cycler (Perkin Elmer, Foster City, CA) as follows: 1 min at
95°C, 1 min at 60°C, and 2 min at 72°C for 30 cycles, followed by
10 min at 72°C. PCR products were separated on a 1.5% agarose gel
and visualized with ethidium bromide using an imaging system (Herolab,
Wiesloch, Germany). HAS-2 primer was designed according to the
published sequence as follows [27
]: Upper:
5'-TGGGGTGGAAAAAGAGAAGTC-3'; Lower: 5'-TGAGAAAGAAAGGAAAGAATC-3'; Human
ß-actin: Upper: 5'-ACTCTTCCAGCCTTCCTTCC-3'; Lower:
5'-TGTCACCTTCACCGTTCCAG-3'.
HA-RADIO-IMMUNO ASSAY
Supernatants from DC and fibroblasts were collected at the
indicated time-points, and the HA content was determined by addition of
I125-labeled HA-binding protein (HABP-I125),
according to the manufacturer’s instructions (Pharmacia) and analyzed
using a Cobra-2® -counter (Canberra Packard, Dreieich, Germany).
Analysis of DC-cluster formation
Time-lapse video microscopy was performed using a
T.I.L.L.-Photonics digital video imaging system consisting of a
TILL-imago CCD camera, a Polychrome II monochromator (TILL Photonics,
Munich, Germany) connected to an IMT-2-inverted microscope (Olympus,
Hamburg, Germany). The cells were maintained in a heated (37°C)
incubation chamber under sterile conditions in RPMI-1640 medium (PTS
GmbH, Freiburg, Germany). DC were prelabeled with the membrane dye
PKH-2 (Sigma) according to the manufacturer’s instructions.
Proliferation assay
DC were cultivated for 96 h as described above and were
then cultivated on CD154-expressing L cells or mock-transfected control
L cells for 48 h. The L cells were removed by immunomagnetic
depletion, and purity of DC was determined by fluorescein-activated
cell sorter (FACS) analysis. DC (1x106) were cultivated
with 1 x 105-purified allogeneic T cells (CD4+/CD8+)
for 144 h in U-bottom 96-well plates. Then
[3H]-thymidine incorporation was determined for the last
18 h of the experiment. Plates were harvested with a Canberra
Packard Filter-MateTM, and incorporation of [3H]
thymidine was determined by liquid scintillation spectroscopy using a
TopCountTM device (both Canberra Packard).
Western blotting
Day 4 DC were cultivated in complete IMDM on CD154-expressing
murine fibroblasts or mock-transfected control cells for the indicated
time periods. DC were separated from the fibroblasts by immunomagnetic
depletion, and 1 x 106 DCs were lysed for 1 h at
4°C in a buffer containing 1% Triton X-100, 10 mM Tris (pH 7.5), 150
mM NaCl, 3 mM ethylenediaminetetraacetate , 50 mM iodoacetamide, 2 mM
phenylmethylsulfonyl fluoride, 1 µg/ml leupeptin, and 20 µg/ml
aprotinin. Lysates were centrifuged at 9000 g for 10 min,
and the protein concentration of the supernatants was determined by
UV-absorption at 280 nm. Each lysate (20 µg) was boiled in
nonreducing sodium dodecyl sulfate (SDS) sample buffer and separated on
an 8% SDS-polyacrylamide gel electrophoresis (PAGE) gel. Proteins were
then transferred onto a nitrocellulose membrane. Unspecific binding was
blocked by addition of 5% nonfat dried milk powder in TBST for 2 h at room temperature. The membrane was then incubated for 2 h
with 2 µg/ml panCD44 mAb BU 75 (The Binding Site). After several
washes with TBS-tween, the membranes were incubated for 1 h with a
horseradish peroxidase-conjugated goat anti-mouse mAb (Dako,
Carpinteria, CA) in TBST. The protein bands were detected using the
enhanced chemiluminescence system (Amersham, Braunschweig, Germany).
Prestained molecular weight (MW) markers (BioRad, München,
Germany) were used to determine the protein size.
Preparation of HA-FITC and HA-binding assay
High MW HA (Healon®, Pharmacia) was labeled with FITC as
described previously [28
]. In brief, 1 mg HA was
incubated with 20 µg dibutyltindilaureate, 20 µg FITC, and sodium
hydrogencarbonate (all Aldrich, Steinheim, Germany) at 95°C and pH 8
for 30 min. After cooling, the solution was dialyzed against water in a
Spectra-pore® tube with a MW cut-off of 500 until no unbound FITC was
detectable. For adhesion experiments, 1 x 106 DC were
incubated with FITC-HA or unlabeled HA (200 µg/ml) for 45 min at
4°C in the dark and washed twice with PBS. HA binding was determined
by FACS analysis of 1 x 105 cells using a FACScan
device with the Cell Quest software (Becton Dickinson).
TNF- enzyme-linked immunosorbent assay (ELISA)
Supernatants from DC were collected at the indicated
time-points, and the content of TNF- was determined according to the
manufacturer’s instructions (Becton Dickinson), measured at an
extinction of 630 nm in a MR 5000 ELISA reader, and analyzed using the
Bio-LinxTM software (both Dynatech, Chantilly, VA).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Homotypic cell clustering is a very early event occurring during DC maturation in vitro. Time-lapse video microscopy was established to analyze the cluster formation of DC on CD154-transfected or mock-transfected fibroblasts. Only CD154-expressing fibroblasts were able to quickly induce firm DC adhesion after 1–2 h followed by DC-cluster formation after 3–8 h of co-incubation (unpublished results and Fig. 1A ). The size of the DC cluster formed by co-incubation on fibroblasts was quantified after 5 h by counting the 10 largest clusters per microscopic field in a Neubauer chamber. The results confirmed that only CD154-treated DC were able to form clusters of 10 cells or more (Fig. 1B) . Further, after 18–24 h, these clusters dissolved to a single-cell suspension of DC, showing the phenotype of matured DC with high expression of surface receptors like MHC class II and B7-costimulatory molecules. These matured DC showed an enhanced capacity to stimulate naive, resting T cells in a standard mixed leukocyte reaction (MLR; Fig. 2 and [1 ]). Again, these changes are restricted to co-incubation on CD154-transfected fibroblasts (Fig. 2) .
|
|
|
Fibroblasts have been described to produce high amounts of HA [30 ]. We wished to determine the amounts of fibroblast-derived HA in our system to use it for functional studies during the CD154-induced DC maturation. This is of advantage, because it avoids the use of exogenously added, potentially contaminated HA, which is important, because DC are highly susceptible to stimulation with bacterial products, and ng amounts of LPS have been shown to induce DC activation [31 ]. To investigate this, RIA assays of cell-culture supernatants were performed, showing that HA was indeed produced in large quantities up to 150 ng/ml by the murine fibroblasts (Fig. 4A ). Further, the HA production was time-dependent but independent of the CD154 transfection of the fibroblasts (Fig. 4A) . In addition, the HA present in our culture system originates solely from the fibroblasts, because human monocyte-derived DC were unable to produce HA even if stimulated with LPS or CD154 (Fig. 4A) . This was confirmed by RT-PCR experiments on the adult form of human HA synthetases, the HA synthetase-2 (HAS-2) [27 ]. In contrast to freshly isolated human keratinocytes or HACAT cells, which strongly express mRNA of HAS-2, DC do not express mRNA of HAS-2 irrespective of their degree of maturation (Fig. 4B) .
|
content 8 h after co-incubation by ELISA as a readout
parameter for DC maturation [31
]. The TNF- secretion
by CD44 mAb-pretreated DC was substantially lowered compared with the
IgG-treated control or the enzyme-treated cells (Fig. 6
). However, we could exclude a direct effect of the mAb or enzymes
on DC, because co-incubation on mock-transfected fibroblasts did not
induce any release of TNF- (Fig. 6)
. To extend these findings to a
phenotypical level, we analyzed the MHC class II expression of the mAb-
or enzyme-pretreated CD154-matured DC by FACS (Fig. 7
). In agreement with the previous findings, adding anti-CD44s mAbs
inhibited the CD154-induced MHC class II up-regulation on DC, whereas
Hyaluronidase or Heparinase had no effect. Taken together,
these data exclude the need of HA during the CD154-induced maturation
of human DC. We could also exclude that DC clustering was depending on
the heterotypic interaction of CD44 molecules expressed on murine
fibroblasts and on human DC, because addition of the mouse anti-CD44s
mAb KM201 was not effective to inhibit MHC-class II up-regulation (Fig. 7)
.
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The CD44 expression on human monocyte-derived DC has already been
studied by Haegel-Kronenberger et al. [13
]. After
48 h of DC stimulation by TNF-, an enhanced expression of a
CD44 isoform containing the CD44 splice variants v3, v6, and v9 was
found [13
]. We could complement these findings by
showing a strong up-regulation of the CD44 standard form, but no CD44
isoforms, during the first 12 h of DC maturation. On the basis of
these findings, an attractive model points to a two-stepped
up-regulation of CD44. First, the early, rapid induction of the 85 kDa
CD44 standard form on DC occurs 3–12 h after CD40 ligation. Our
observations suggest that this form of CD44 is involved in the close
cell aggregation, a crucial step for further DC maturation. Similar
observations have been made in human long-term bone marrow cultures,
which express high levels of CD44s. In these cultures, an anti-CD44s
mAb affected myeloid colony formation and maturation
[20
]. In this regard, CD44 binding could intensify
cell-cell interactions, because it is closely linked to the
cytoskeleton by the small GTPases, Rho, Rac, and Cdc42
[33
34
35
]. Recent studies show that CD44-mediated Rho
signaling is directly involved in the cytoskeletal rearrangements
needed for the generation of dendrites in activated B cells
[34
].
How CD44 confers signals to enhance DC maturation is still unclear. A
direct enhancing effect has been described for the anti-CD44s mAb J173
on human DC [13
]. Nevertheless, it has been suggested
that CD44 is unable to deliver activating signals directly to the
nucleus but works as a bystander receptor to enhance CD3-driven signals
in T cells, for example [36
]. Besides the possibility
that CD44 is directly involved in DC maturation [13
],
the close cell contact may be needed to have effective cytokine
concentrations in the surrounding of the cells. Here, TNF-, which is
produced in high amounts once DC are activated, plays a crucial role,
because we have shown that binding of TNF- by a soluble TNF receptor
abolishes DC maturation [31
].
Another possibility that was also investigated here is that binding of the major ligand HA is responsible for changes in the structural conformation of CD44 and thereby works as an activating cross-linker. Indeed, we found a strong induction of the HA affinity coinciding with CD44 up-regulation during DC maturation. A comparable regulation of the HA-binding capacity has been demonstrated in a variety of immune cells, for example activated monocytes, T cells, and hematopoietic progenitor cells [17 18 19 , 29 ]. However, some authors describe a direct activating effect of HA on DC and T cells [13 , 23 ]. In previous studies, we showed that only low MW HA fragments of oligosaccharide size were able to activate human DC [31 ]. These HA fragments have also been shown to induce intracellular signaling cascades leading to cell activation in murine macrophages [37 ]. However, whether they exert their effects through binding to CD44 is still controversial [31 , 37 , 38 ]. In contrast, high MW HA, as produced by the murine fibroblasts in our culture model, had no influence on the CD40-induced DC maturation, because enzymatic digestion of HA had an enhancing rather than blocking effect, which can be explained by the production of low MW HA in the culture system. These findings are in line with previous results showing that exogenously added high MW HA failed to induce phenotypical or functional changes in immature human or murine DC [31 ]. Further, complete DC maturation can also be achieved by co-culture with a cross-linking anti-CD40 mAb in the absence of HA ([31 ] and unpublished results). A possible explanation for these findings could be the homotypic interaction of CD44 on DC during the first 12 h of maturation. A CD44-CD44 self-interaction has been proposed for cell-cell adhesion of rat pancreatic carcinoma cells but was restricted to CD44 isoforms containing exons v6 and v7, which we could not detect on human DC during the first 12 h of stimulation [14 ]. Nevertheless, other ligands for CD44, such as the recently characterized LYVE-1, might exist on DC, thereby enhancing homotypic cell interactions [39 ].
At a later stage of maturation after 18–24 h, DC up-regulate a CD44 isoform containing the exons v3,v6, and v9 [13 ]. Again, there are similarities to other immune cells such as primary bone marrow cultures and T cells, which have also been shown to up-regulate CD44 isoforms during activation and differentiation [20 , 29 ]. We have shown that in human Langerhans cells and monocytes, the expression of CD44 isoforms is of importance for T-cell binding in the paracortical areas of the draining lymph node [17 , 21 ]. Here, CD44 isoforms might be not only of importance during the T-cell binding but also be responsible for the proposed co-stimulatory role of CD44 [22 , 23 ]. Recent data suggest that this function of CD44 is conferred by cytoskeletal rearrangements and co-localization of MHC class II complexes on the antigen-presenting cell (APC) as well as CD3 and TCR complexes on the T-cell side [34 , 36 ]. Clearly, CD44 is only one amongst other receptors that mediate the formation of the immunological synapse between the APC and the T cell, and similar observations have been made for other receptors linked to the cytoskeleton by small ezrin radixin moesin proteins such as ICAM-1 [35 , 40 ]. However, because it is very unlikely that homotypic DC clustering occurs inside the lymph node, an in vivo correlation for our findings might be the clustering of T cells around a DC following DC activation by T-memory cells via CD40L.
Another possibility is that close cell-cell contacts after CD40 ligation occur earlier, during emigration of the DC from the site of activation. This hypothesis is further supported by in vivo studies in CD154-deficient mice [41 ]. In these mice, the induction of a contact hypersensitivity response was significantly disturbed as a result of the inhibition of early DC emigration from the skin into the regional lymph node [41 ]. Indeed, similarly to cluster formation that we describe as a crucial factor for DC maturation in vitro, activated Langerhans cells form cords within the lymphatic vessels during their emigration from the skin [21 ].
In conclusion, we have demonstrated that during the CD154-induced activation of human monocyte-derived DC, early clustering events are essential for further maturation. During the first 3–12 h, CD44s is strongly up-regulated on human DC, accompanied by an increased HA-binding capacity. However, binding to HA had no direct effect on DC maturation but might be of relevance for DC migration. In contrast, blocking of CD44 by mAbs reduced the homotypic cell aggregation and thereby inhibited further DC maturation.
|
ACKNOWLEDGEMENTS |
|---|
Received January 20, 2001; revised July 3, 2001; accepted July 9, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. de la Rosa, M. Yanez-Mo, R. Samaneigo, D. Serrano-Gomez, L. Martinez-Munoz, E. Fernandez-Ruiz, N. Longo, F. Sanchez-Madrid, A. L. Corbi, and P. Sanchez-Mateos Regulated recruitment of DC-SIGN to cell-cell contact regions during zymosan-induced human dendritic cell aggregation J. Leukoc. Biol., May 1, 2005; 77(5): 699 - 709. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Kawamura, K. Iyonaga, H. Ichiyasu, J. Nagano, M. Suga, and Y. Sasaki Differentiation, Maturation, and Survival of Dendritic Cells by Osteopontin Regulation Clin. Vaccine Immunol., January 1, 2005; 12(1): 206 - 212. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
