The regulation and function of the CD44 family of surface glycoproteins were investigated in human monocyte-derived dendritic cells (DCs). Variant CD44 isoform transcripts encoding exons v3, v6, and v9 are differently regulated during the differentiation of monocytes into DCs. TNF-α treatment, which induces the maturation of DCs, up-regulates the expression of all v3-, v6-, and v9-containing isoforms examined. CD44 molecules are involved in the adhesion of DCs to immobilized hyaluronate (HA), and v3- and v6-containing variants participate in this function, whereas anti-CD44v9 mAbs were unable to inhibit DC adhesion to HA. The consequences of ligand binding to CD44 were examined by culturing DCs on dishes coated with HA or various anti-CD44 mAbs. HA, the anti-pan CD44 mAb J173, and mAbs directed against v6- and v9-containing (but not v3-containing) isoforms provoked DC aggregation, phenotypic and functional maturation, and the secretion of IL-8, TNF-α, IL-1β, and granulocyte-macrophage CSF. In addition, IL-6, IL-10, and IL-12 were released by DCs stimulated with either J173 or HA, although these cytokines were not detected or were found only at low levels in the culture supernatants of DCs treated with anti-CD44v6 or anti-CD44v9 mAbs. Our study points to distinct capacities of the v3-, v6-, and v9-containing isoforms expressed by human DCs to mediate cell adhesion to HA and/or a signal inducing DC maturation and the secretion of cytokines.

Dendritic cells (DCs)4 are potent immunostimulatory leukocytes found in all tissues and organs of the body (1). Various DC populations can be distinguished according to their stage of maturation and microenvironment. In nonlymphoid organs, immature DCs are able to capture and process Ags, which they efficiently present to sensitized T lymphocytes, initiating a secondary immune response. In some circumstances, such as exposure to TNF-α, haptens, or chemical allergens, DCs can emigrate from the peripheral tissues to colonize the T cell areas of secondary lymphoid organs through afferent lymphatics or the blood (1, 2, 3). Phenotypic and functional changes known as DC maturation underlie this particular behavior (4, 5, 6). Among these changes, DCs up-regulate the expression of MHC class I and class II Ags and of accessory molecules involved in antigenic stimulation and lose their ability to efficiently process new Ags, while they increase their T cell stimulatory capacity. Consequently, in lymphoid tissues, mature DCs can sensitize resting T cells and initiate a primary immune response. Mature DCs are considered to be the most powerful “professional” APCs and represent a promising potential tool to improve the immunotherapy of cancer (7).

Among the different adhesion molecules expressed by DCs, CD44 comprises a family of polymorphic surface glycoproteins and represents the principal cell surface receptor for hyaluronate (HA), a glycosaminoglycan found in extracellular matrixes (ECM) and in association with cell surfaces (8, 9, 10). A variety of functions have been attributed to CD44 receptors, including leukocyte adhesion, differentiation, homing, activation, and extravasation to inflammatory sites (10). Recent studies have implicated CD44 and HA binding in signal transduction events (11), notably in T lymphocytes and NK cells, where CD44 participates in a tyrosine kinase-dependent signaling pathway (12, 13, 14). CD44 molecules are highly heterogeneous, and extensive glycosylation can modulate their ability to bind HA (15, 16, 17, 18). Moreover, insertion of an additional membrane-proximal region resulting from the alternative splicing of up to 10 exons (termed v1 to v10) gives rise to the so-called CD44 variants (19, 20). Whereas the common (“standard” or “hemopoietic”) form of CD44 is widely and often highly expressed, variant isoforms are more restricted. Some carrying the v6 region have been implicated in tumor metastasis and lymphocyte activation, suggesting a possible role in cell homing to the lymph nodes (19), while variants containing v3 represent targets for glycosaminoglycan modification resulting in growth factor binding and presentation (21). Exons v8 to v10 are represented in the so-called epithelial form of CD44 (CD44E) (22), and a variant containing v10 has been shown to mediate a signal inducing endothelial cell proliferation (23). Nevertheless, the functions of CD44 variant isoforms generally remain poorly understood.

Mouse Langerhans cells and DCs isolated from blood and lymph nodes strongly express CD44 (24). In humans, DCs differentiated from blood monocytes cultured in the presence of GM-CSF and IL-4 are CD44 positive (5). However, the role of the CD44 molecules expressed by DCs remains largely unknown, and the pattern of CD44 variant isoforms has not yet been determined in this cell type. Accumulating evidence suggests that regulation of CD44 on the surface of DCs may modulate their immune function or adhesive phenotype. Thus, recent work has demonstrated the involvement of CD44 in the emigration of Langerhans cells from the epidermis and their adhesion to the T cell zones of lymph nodes (25). TNF-α has been shown to increase the overall expression of CD44 in a variety of cell types, including mouse Langerhans cells and human monocyte-derived DCs (5, 26, 27). In the latter, expression of a v9-containing CD44 variant was detected following TNF-α-induced maturation (5). Therefore, we attempted in the present work to better characterize the CD44 isoforms expressed by human monocyte-derived DCs and to study their regulation during DC differentiation and maturation induced by TNF-α. To gain insight into the function(s) of these receptors, we sought to determine their implication in DC adhesion to HA and examined the effects on DC function of the binding of CD44 epitopes.

Cultures were established in RPMI 1640 medium containing Glutamax-1 (Life Technologies, Paisley, U.K.), supplemented with 10% FCS (Life Technologies), 1% nonessential amino acids (Life Technologies), 1% sodium pyruvate (Sigma, St. Louis, MO), 50 U/ml penicillin, and 50 U/ml streptomycin (Life Technologies). Hyaluronic acid (H1751) and BSA (A3350) were purchased from Sigma. Recombinant human (rh) GM-CSF was a generous gift from Schering-Plough (Levallois-Perret, France), while rhIL-4 was obtained from PeproTech (Rocky Hill, NJ), and rhTNF-α from Genzyme Corp. (Cambridge, MA).

Cells were washed twice in PBS (Life Technologies) and incubated for 30 min with the following primary mAbs: L243 (anti-HLA-DR, IgG2a, Becton Dickinson, San Jose, CA), W6/32 (anti-HLA-A, -B, and -C, IgG2a, Dako, Copenhagen, Denmark), BL6 (anti-CD1a, IgG1, Immunotech, Marseille, France), MAB89 (anti-CD40, IgG1, Immunotech), MAB104 (anti-CD80, IgG1, Immunotech), IT2.2 (anti-CD86, IgG2b, PharMingen, San Diego, CA), HB-15a (anti-CD83, IgG2b, Immunotech), J173 (anti-pan CD44, IgG1, Immunotech), 5F12 (anti-pan CD44, IgG2a, Neomarkers, Fremont, CA), 3G5 (anti-CD44v3, IgG2b, R&D Systems, Abingdon, U.K.), VFF-8 (anti-CD44v5, IgG1, Bender, Vienna, Austria), 2F10 (anti-CD44v6, IgG1, R&D Systems), VFF-9 (anti-CD44v7, IgG1, Bender), FW11-24-17-36 (anti-CD44v9, IgG1, HB-258, American Type Culture Collection, Manassas, VA), UCHT-1 (anti-CD3, IgG1, Immunotech), MΦP9 (anti-CD14, IgG2b, Becton Dickinson), 3G8 (anti-CD16, IgG1, Medarex, Annandale, NJ), J4-119 (anti-CD19, IgG1, Becton Dickinson), or BBIG-I1 (anti-CD54, IgG1, R&D Systems). Mouse IgG1, IgG2a, and IgG2b (Immunotech) were used as isotype controls. The cells were then washed twice in Dulbecco’s PBS and incubated for 30 min with the FITC-conjugated affinity-isolated F(ab′)2 fraction of a sheep anti-mouse IgG Ab (Silenus, Hawthorn, Australia). Direct staining was performed using PE-conjugated L243 (anti-HLA-DR, Becton Dickinson), HB-15 (anti-CD83, Immunotech), or IgG1 (Immunotech). All incubation and washing steps were conducted at 4°C, and the stained cells were analyzed by flow cytometry on a FACScan cytometer (Becton Dickinson) using LYSYS II software (Becton Dickinson).

Human blood monocytes were isolated by continuous flow centrifugation leukopheresis and counterflow centrifugation elutriation as previously described (28). The cells were then cultured in complete medium supplemented with 50 ng/ml rhGM-CSF and 200 U/ml rhIL-4 for 7 days. Cell density was adjusted to 106 cells/ml, and the medium was changed on day 4 of culture. Differentiation of monocytes into DCs was monitored by flow cytometric analysis of cell surface phenotypes. On day 7, DCs expressed high levels of MHC class I and class II (HLA-DR) molecules and the Ags CD1a, CD40, CD80, and CD44, but only small amounts of the CD86 Ag and neither CD83 nor CD3, CD14, CD16, or CD19, in agreement with previous observations (5, 29).

In TNF-α stimulation experiments, day 7 DCs were incubated in complete medium supplemented with 50 ng/ml rhGM-CSF, 200 U/ml rhIL-4, and 20 ng/ml rhTNF-α. After 24 to 48 h of culture, the cells were harvested, washed in PBS, and used for phenotypic or functional analyses. In CD44 stimulation experiments, day 7 DCs were cultured in 60-mm petri dishes (Falcon 1007, Becton Dickinson) coated at 4°C with 10 μg/ml of various anti-CD44 mAbs or a control IgG1 in 3.5 ml of PBS, or with 50 μg/ml HA in 5 ml of PBS. The dishes were coated overnight and rinsed twice with PBS before plating the DCs, which were previously depleted of the few remaining monocytes by treatment with magnetic beads coated with an anti-CD14 mAb (M450, Dynal, Oslo, Norway) according to the manufacturer’s instructions. Formation of cell aggregates was occasionally observed after 24-h culture. The cells were then harvested for phenotypic or functional analyses, and the supernatants were recovered for determination of cytokine levels.

All Ab and HA solutions employed for coating dishes were previously tested for their endotoxin content using a Limulus amebocyte assay (Biogenix, Biogenic, Maurin, France), and when necessary, control dishes were prepared using appropriate concentrations of Escherichia coli endotoxin (BioWittaker, Walkersville, MD) dissolved in PBS. Anti-CD44v3, -v6, and -v9 mAb solutions were devoid of detectable contaminating endotoxins. On the contrary, HA isolated from human umbilical cord was found to contain endotoxins at levels of 20 to 30 IU/ml at the concentration of 50 μg/ml HA chosen for coating dishes. Therefore, to ensure that the effect observed with HA was specific and was not due to endotoxin stimulation, control cells were plated on dishes coated overnight with a solution containing 30 IU/ml endotoxin and rinsed twice with PBS, which failed to stimulate DC maturation. This was not the case when DCs were incubated with 30 IU/ml soluble endotoxin added directly to the culture medium or on dishes coated with a highly concentrated (e.g., 900 IU/ml) endotoxin solution; in both cases DC maturation was triggered, as shown by FACS analysis of the cell surface phenotype. In subsequent experiments the low HA concentration chosen for coating dishes (50 μg/ml) could be considered to exclude an effect of contaminating endotoxins on the maturation of DCs.

Total RNA was isolated from highly purified DCs using the guanidium isothiocyanate method (30). cDNA synthesis was conducted in a 100-μl reaction mix containing 2 μg of total cellular RNA, 0.4 U/μl RNase inhibitor (Eurogentec, Seraing, Belgium), 1 mM dNTP (Boehringer Mannheim, Mannheim, Germany), 1× concentrated hexanucleotide mixture (Boehringer Mannheim), and 0.4 U/μl AMV reverse transcriptase (Finnzymes Oy, Espoo, Finland) in the reaction buffer provided by the manufacturer. After 45-min incubation at 42°C, the mixture was heated to 95°C for 5 min. The whole variable region of CD44 was amplified using oligonucleotides hybridizing with sequences present in the flanking exons 5 and 15 (Table I), which code for the common or standard region of CD44 (19). cDNA samples (5 μl) were added to a 50-μl reaction mix containing 200 nM of each oligonucleotide (Eurogentec), 0.5 mM dNTP (Boehringer Mannheim), 1.5 mM MgCl2, and 0.05 U/μl Taq DNA polymerase (Goldstar, Eurogentec) in 1× reaction buffer. Amplification was performed in a DNA thermal cycler (Hybaid, OmniGene, Teddington, U.K.) as follows: 30 s at 95°C, 15 s at 55°C, and 2 min at 72°C for 35 cycles, followed by 10 min at 72°C. PCR products were separated on a 2% agarose gel and transferred onto Hybond-N+ membranes (Amersham France, Les Ulis, France), and Southern blots were hybridized with a panel of oligonucleotide probes corresponding to the CD44 variable exons v3 to v10 (Table I). The probes were labeled with digoxigenin-dideoxyUTP (DIG oligonucleotide 3′ end labeling kit, Boehringer Mannheim). Homologous DNA fragments hybridized on Southern blots were revealed by chemiluminescence using a DIG luminescent detection kit (Boehringer Mannheim). After a series of short exposures to Hyperfilm-MP (Amersham), band intensities were quantified with the program Visiolab 2000 (Biocom, Les Ulis, France). All results were normalized relative to an actin sequence amplified from each cDNA sample. Actin PCR was conducted using the oligonucleotides given in Table I, with 25 cycles of 30 s at 95°C, 15 s at 56°C, and 60 s at 72°C. The PCR product was run on a 1.5% agarose gel, the actin band intensity was quantified by gel scanning as described above, and the result was taken as a standard indicative of the total amount of cDNA added to the amplification mixture.

Table I.

Oligonucleotides used for the amplification or detection of CD44 variant and cytokine transcripts

Sequence
CD44  
5′ standard (exon 5) 5′-CCTCCAGTGAAAGGAGCAGCACT-3′ 
3′ standard (exon 15) 5′-GGGGTGGAATGTGTCTTGGTCTC-3′ 
v3 5′-GGCTGGGAGCCAAATGAAGAAAA-3′ 
v4 5′-AAAACAGAACCAGGACTGGACCC-3′ 
v5 5′-AAACTGGAACCCAGAAGCACACC-3′ 
v6 5′-GTGGTTTGGCAACAGATGGCATG-3′ 
v7 5′-GGACAGTTCCTGGACTGATTTCT-3′ 
v8 5′-CAGGTTTGGTGGAAGATTTGGAC-3′ 
v9 5′-CTCTACATCACATGAAGGCTTGG-3′ 
v10 5′-TCAGCTAAGACTGGGTCCTTTGG-3′ 
Cytokines  
5′ IL-6 5′-ACGAATTCACAAACAAATTCGGTACA-3′ 
3′ IL-6 5′-CATCTAGATTCTTTGCCTTTTTCTGC-3′ 
5′ IL-8 5′-TTCTGCAGCTCTGTGTGAAGG-3′ 
3′ IL-8 5′-GTGGATCCTGGCTAGCAGACTAGG-3′ 
5′ IL-10 5′-CACTCATGGCTTTGTAGATGCC-3′ 
3′ IL-10 5′-AGTCTGAGAACAGCTGCACCCAC-3′ 
5′ IL-12p35 5′-CCCTGCAGTGCCGGCTCAGCATGTG-3′ 
3′ IL-12P35 5′-GCCCGAATTCTGAAAGCATGAAG-3′ 
5′ IL-12p40 5′-TCTCTGCAGAGAGTCAGAGGG-3′ 
3′ IL-12p40 5′-ACGGATCCTGATGGATCAGGTCATAAGAG-3′ 
5′ TNF-α 5′-CTAAGCTTGGGTTCCGACCCTAAGCCCCC-3′ 
3′ TNF-α 5′-GCGAATTCCCTCCTGGCCAATGGCGTGG-3′ 
5′ MIP-1α 5′-CACTCAGCTCCAGGTCACT-3′ 
3′ MIP-1α 5′-CAGGTCTCCACTGCTGCC-3′ 
5′ GM-CSF 5′-GCCTGCAGCATCTCTGCACCCGCC-3′ 
3′ GM-CSF 5′-CTTTCAAGCTTTCAAAGGTGATAATCTG-3′ 
5′ IL-1β 5′-AGGGGATCCTCTTAGCACTACCCTAAG-3′ 
3′ IL-1β 5′-AAAAGCTTGGTGATGTCTGGTCCA-3′ 
5′ IL-15 5′-CAAGTTATTTCACTTGAGTCCGGAG-3′ 
3′ IL-15 5′-TTCTAAGAGTTCATCTGATCCAAGG-3′ 
5′ actin 5′-GACTACCTCATGAAGATCCT-3′ 
3′ actin 5′-ATCCACATCTGCTGGAAGGT-3′ 
Sequence
CD44  
5′ standard (exon 5) 5′-CCTCCAGTGAAAGGAGCAGCACT-3′ 
3′ standard (exon 15) 5′-GGGGTGGAATGTGTCTTGGTCTC-3′ 
v3 5′-GGCTGGGAGCCAAATGAAGAAAA-3′ 
v4 5′-AAAACAGAACCAGGACTGGACCC-3′ 
v5 5′-AAACTGGAACCCAGAAGCACACC-3′ 
v6 5′-GTGGTTTGGCAACAGATGGCATG-3′ 
v7 5′-GGACAGTTCCTGGACTGATTTCT-3′ 
v8 5′-CAGGTTTGGTGGAAGATTTGGAC-3′ 
v9 5′-CTCTACATCACATGAAGGCTTGG-3′ 
v10 5′-TCAGCTAAGACTGGGTCCTTTGG-3′ 
Cytokines  
5′ IL-6 5′-ACGAATTCACAAACAAATTCGGTACA-3′ 
3′ IL-6 5′-CATCTAGATTCTTTGCCTTTTTCTGC-3′ 
5′ IL-8 5′-TTCTGCAGCTCTGTGTGAAGG-3′ 
3′ IL-8 5′-GTGGATCCTGGCTAGCAGACTAGG-3′ 
5′ IL-10 5′-CACTCATGGCTTTGTAGATGCC-3′ 
3′ IL-10 5′-AGTCTGAGAACAGCTGCACCCAC-3′ 
5′ IL-12p35 5′-CCCTGCAGTGCCGGCTCAGCATGTG-3′ 
3′ IL-12P35 5′-GCCCGAATTCTGAAAGCATGAAG-3′ 
5′ IL-12p40 5′-TCTCTGCAGAGAGTCAGAGGG-3′ 
3′ IL-12p40 5′-ACGGATCCTGATGGATCAGGTCATAAGAG-3′ 
5′ TNF-α 5′-CTAAGCTTGGGTTCCGACCCTAAGCCCCC-3′ 
3′ TNF-α 5′-GCGAATTCCCTCCTGGCCAATGGCGTGG-3′ 
5′ MIP-1α 5′-CACTCAGCTCCAGGTCACT-3′ 
3′ MIP-1α 5′-CAGGTCTCCACTGCTGCC-3′ 
5′ GM-CSF 5′-GCCTGCAGCATCTCTGCACCCGCC-3′ 
3′ GM-CSF 5′-CTTTCAAGCTTTCAAAGGTGATAATCTG-3′ 
5′ IL-1β 5′-AGGGGATCCTCTTAGCACTACCCTAAG-3′ 
3′ IL-1β 5′-AAAAGCTTGGTGATGTCTGGTCCA-3′ 
5′ IL-15 5′-CAAGTTATTTCACTTGAGTCCGGAG-3′ 
3′ IL-15 5′-TTCTAAGAGTTCATCTGATCCAAGG-3′ 
5′ actin 5′-GACTACCTCATGAAGATCCT-3′ 
3′ actin 5′-ATCCACATCTGCTGGAAGGT-3′ 

Oligonucleotide primers for IL-6, IL-8, IL-10, IL-12p35, IL-12p40, TNF-α, MIP-1α, GM-CSF, and IL-1β amplification were provided by Dr. Eric Tartour (Institut Curie, Paris, France), while primers for IL-15 amplification were synthesized by Eurogentec (Table I). Serial 1/10 dilutions of each cDNA sample were subjected to PCR according to the protocol described above, using amplification cycles of 1 min at 95°C, 1 min at 55°C, and 1.5 min at 72°C. The number of cycles (n = 25–35) was adjusted for each cytokine so as to remain in the linear range. PCR products were separated on a 2% agarose gel in the presence of ethidium bromide (Appligene, Illkirch, France) and quantified with the program Visiolab 2000. Again, results were normalized relative to an actin sequence amplified from each cDNA sample.

Cytokine concentrations in DC culture supernatants were determined using two-site sandwich ELISA kits (human IL-6, IL-8, IL-10, IL-12p70, GM-CSF, IL-1β, and TNF-α, Quantikine, R&D Systems) according to the manufacturer’s recommendations. Absorbances (450–540 nm) were read in an ELISA microplate reader (Molecular Devices, Palo Alto, CA).

Flat-bottom 96-well plates (Maxisorp, Nunc, Rockilde, Denmark) were coated overnight with 5 mg/ml HA (Sigma) in PBS, rinsed with PBS, and saturated with 0.5% BSA (Sigma) for 2 h at 4°C. Control wells were coated either with BSA or with an endotoxin concentration equivalent to that found in 5 mg/ml HA solutions (∼3000 IU/ml) to measure unspecific cell adhesion, which was negligible. Anti-CD44 or control mAbs (100 μl of a 40 μg/ml solution in serum-free medium) were added to a first well, 50 μl of serum-free medium was added to the following well, and serial 1/2 dilutions were performed to obtain a final volume of 50 μl of diluted mAb solution in all wells. Cells (105 in 50 μl) were added to each well and left to sediment on ice for 1 h. The wells were then filled with cold PBS to obtain a positive meniscus, the plates were wrapped in Parafilm and centrifuged in an inverted position for 5 min at 400 rpm (22 g), and medium and unbound cells were eliminated by flicking. Adherent cells were stained using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide dye reduction assay (MTT; Sigma) as described by Mossmann (31). Aliquots (200 μl) of culture medium containing 10% MTT were added to all wells, while a series of calibration wells contained graded quantities of DCs added directly to the MTT medium as standards indicative of total cell number. After 3-h incubation at 37°C, the plates were briefly centrifuged, and the supernatant medium (150 μl) was carefully removed from each well. Newly formed blue formazan crystals were dissolved by adding 100 μl of 10% Triton-100 (Sigma) in acidified isopropanol, and absorbances (540–650 nm) were read in an ELISA microplate reader.

The stimulator cells were DCs that had been cultured for 24 h in the presence of anti-CD44 or isotype control mAbs, HA, or control endotoxin as described above. After resuspension, these cells were washed in PBS and irradiated (3000 rad from a 137Cs source). Various numbers of stimulator cells (ranging from 104-625) were added to allogeneic responder cells (1.5 × 105) in the wells of round-bottom 96-well plates (Nunc) in a final volume of 200 μl of culture medium containing 10% human serum (ETS, Strasbourg, France). The responder cells were PBLs or allogeneic CD4+ naive T cells obtained from lymphocyte suspensions depleted of CD45R0+ and CD8+ cells. All cultures were set up in triplicate. Controls included DCs alone, responder cells alone, and responder cells cultured in the presence of 10 U/ml IL-2 (Chiron France, Suresnes, France) and 100 ng/ml Con A (Sigma). After 5 days of culture, 1 μCi of [3H]thymidine (Amersham) was added to each well, and thymidine incorporation was measured 18 h later using a Skatron cell harvester and a Betaplate counter (LKB Wallac, EG&G Instruments, Evry, France).

CD44 expression is known to be modulated on the surface of DCs (5, 24, 26). In particular, the appearance of a v9-containing CD44 isoform has been reported in monocyte-derived DCs following maturation induced by TNF-α (5). Given the diversity of CD44 isoforms already characterized in various cell types, we wished to know whether other variants were expressed by monocyte-derived DCs. Hence we chose to examine the profile of transcripts encoding CD44 isoforms and to compare variant CD44 mRNAs in monocyte-derived DCs to those expressed by fresh blood monocytes. To this aim, the entire variable region of CD44, spanning exons v1 to v10, was PCR amplified (Fig. 1,a), and the PCR products were hybridized with oligonucleotide probes corresponding to all exons of the variable region from v3 to v10. Since v1 is known to be nonfunctional in human cells (20), while v2 has been found only in large and rarely expressed CD44 variants (10), probes for these two exons were not tested. Some variable exons were barely detectable in either monocytes or DCs, for instance v4, v5, and v7 (data not shown), which may have been due either to the absence of such variants or to the large size of their PCR products, as long fragments are known to be less efficiently amplified by the PCR technique. In contrast, other exons were readily amplified, and their relative abundances were modified upon DC differentiation from monocytes (Fig. 1,b). The relative abundances of two v3-containing mRNAs and a v6-containing transcript did not vary significantly between monocytes and day 7 DCs (Fig. 1,b, d0–d7). The size of the v6-containing variable domain indicated that it must contain only the v6 exon and therefore correspond to the CD44v6 form known to be induced in activated lymphocytes (19). Two main v9-containing CD44 variant transcripts, giving bands of approximately 380 and 520 bp, were significantly decreased (about fourfold by densitometry scanning) in DCs compared with those in monocytes (Fig. 1,b). Hybridization with the v10 probe gave two doublets of 300 to 400 and 520 to 640 bp, and the second doublet diminished about twofold in DCs relative to monocytes (Fig. 1 b). Using probes for v8, a 520-bp band was also found to decrease in intensity following DC differentiation from monocytes (data not shown). In view of its size and hybridization pattern, this 520-bp PCR product may well correspond to the CD44E-type isoform (CD44v8–v10), while the overall results obtained using probes for v8, v9, and v10 probably reflect a decrease in the abundance of this CD44E-type transcript upon DC differentiation. It is noteworthy that although quantitative regulation was observed, no change in the general splicing pattern of the variable domain could be detected following DC differentiation from monocytes, nor was this splicing pattern modified when DCs were treated with TNF-α on day 7 of culture. However, using semiquantitative PCR analysis, mature DCs stimulated with TNF-α for 48 h showed elevated levels of v3-, v6-, v9-, and v10-containing transcripts compared with unstimulated DCs on day 7 or 9 of culture (data not shown). This led us to study more closely the regulation of CD44 isoforms on the surface of DCs during their maturation induced by TNF-α.

FIGURE 1.

Regulation of CD44 variant transcripts during DC differentiation from monocytes. a, Schematic representation of the PCR strategy for amplification of the CD44 variable region, using spanning oligonucleotides hybridizing with standard CD44 sequences. b, The variable region of CD44 was RT-PCR amplified from total RNA isolated either from fresh monocytes or from differentiated DCs on day 7 of culture. PCR products were hybridized using probes corresponding to the variable exons v3, v6, v9, and v10 as indicated, and the result shown is from one of two independent experiments.

FIGURE 1.

Regulation of CD44 variant transcripts during DC differentiation from monocytes. a, Schematic representation of the PCR strategy for amplification of the CD44 variable region, using spanning oligonucleotides hybridizing with standard CD44 sequences. b, The variable region of CD44 was RT-PCR amplified from total RNA isolated either from fresh monocytes or from differentiated DCs on day 7 of culture. PCR products were hybridized using probes corresponding to the variable exons v3, v6, v9, and v10 as indicated, and the result shown is from one of two independent experiments.

Close modal

We therefore examined whether the CD44 isoforms characterized at the transcript level were in fact expressed on the surface of DCs and regulated upon TNF-α stimulation. In view of the preceding results, the surface expression of v3-, v6-, and v9-containing isoforms and that of total CD44 were studied by immunocytochemistry and flow cytometry. After 7-day differentiation in the presence of rhGM-CSF and rhIL-4, DCs showed intense and relatively homogeneous staining with the anti-pan CD44 mAb J173. Although staining with anti-CD44v5 or anti-CD44v7 mAbs was almost negative (data not shown), the three variant epitopes, v3, v6, and v9, were indeed found to be expressed on the surface of DCs (Fig. 2), in agreement with the PCR results. However, their levels of expression were somewhat weak on unstimulated, immature DCs, with mean fluorescence intensities about 100-fold lower than that of total CD44 stained with J173. Moreover, despite a relative homogeneity in variant CD44 expression among unstimulated DC cultures (Fig. 2), a minor population of cells stained negatively for the v3, v6, and v9 epitopes. TNF-α induced the maturation of DCs, as assessed by staining for HLA-DR, CD40, CD80, and CD86, all of which were up-regulated on the cell surface following TNF-α treatment (5), and for CD83, a marker of mature DCs (29) (data not shown). After 48-h stimulation with TNF-α, total CD44 expression was slightly, but significantly, up-regulated (1.5- to 2-fold; Fig. 2, upper left panel). On the other hand, expression of the v3, v6, and v9 isoforms was more strongly induced (Fig. 2); the mean fluorescence intensities corresponding to these epitopes increased up to 5-fold in DCs stimulated by TNF-α. This would suggest that variant CD44 isoforms may play a more important role in DCs that have undergone maturation in the presence of TNF-α.

FIGURE 2.

Induction of CD44 isoform expression on the surface of DCs stimulated with TNF-α. Day 7 monocyte-derived DCs were treated with TNF-α for 48 h, and the surface expression of total CD44 (stained with the mAb J173) and of variant isoforms containing the v3, v6, and v9 epitopes was compared in stimulated and unstimulated DCs. Dotted profiles indicate controls stained with an irrelevant IgG; plain profiles show unstimulated DCs; bold profiles indicate DCs treated with TNF-α.

FIGURE 2.

Induction of CD44 isoform expression on the surface of DCs stimulated with TNF-α. Day 7 monocyte-derived DCs were treated with TNF-α for 48 h, and the surface expression of total CD44 (stained with the mAb J173) and of variant isoforms containing the v3, v6, and v9 epitopes was compared in stimulated and unstimulated DCs. Dotted profiles indicate controls stained with an irrelevant IgG; plain profiles show unstimulated DCs; bold profiles indicate DCs treated with TNF-α.

Close modal

Some members of the CD44 family have been implicated in cell adhesion to the extracellular matrix component HA (9), but the contributions of variant isoforms to this process remain a matter of debate. As a first step to investigate the function(s) of the CD44 molecules expressed by DCs, we examined their involvement in cell attachment to HA. DCs were indeed capable of adhering to a plastic surface coated with a highly concentrated HA solution (5 mg/ml) within 1 h of incubation at 4°C (Fig. 3). Although the percentage of adherent cells varied between experiments (40–80%; data not shown), possibly due to donor-specific parameters or to slight variations in experimental conditions, Ab inhibition clearly demonstrated that CD44 molecules were involved in DC adhesion to HA. Cell adhesion was measured in the presence of graded concentrations of the mAbs J173 and 5F12, which recognize common CD44 epitopes and are known to block HA binding mediated by CD44 (32, 33). At saturating concentrations, 5F12 and J173 were able to partially inhibit the adhesive interaction between DCs and HA (by ∼50% compared with a control IgG). Unexpectedly, DCs stimulated with TNF-α, which expressed higher levels of CD44, showed no significant modulation of their overall capacity to bind to immobilized HA (data not shown). Regulatory events occurring at the post-translational level may therefore further modulate HA binding regardless of the total amount of surface CD44.

FIGURE 3.

DC adhesion to immobilized HA mediated by CD44 isoforms. Adhesion of day 7 differentiated DCs to HA-coated wells was measured at 4°C in the presence of saturating concentrations (20 μg/ml) of anti-pan CD44 (J173 or 5F12) or anti-variant CD44 (v3, v6, or v9) mAbs or a control IgG. The result shown is the mean of four independent experiments testing immature day 7 DCs, expressed as the percentage of adherent cells relative to the maximal adhesion obtained in the absence of mAbs. Error bars correspond to a 10% imprecision as evaluated in our adhesion assays. Statistical analyses were performed using paired Student’s t test, and p values (∗, p ≤ 0.05; ∗∗, p ≤ 0.01) are shown for each set of conditions compared with the control conditions where DC adhesion was determined in the presence of irrelevant IgG1.

FIGURE 3.

DC adhesion to immobilized HA mediated by CD44 isoforms. Adhesion of day 7 differentiated DCs to HA-coated wells was measured at 4°C in the presence of saturating concentrations (20 μg/ml) of anti-pan CD44 (J173 or 5F12) or anti-variant CD44 (v3, v6, or v9) mAbs or a control IgG. The result shown is the mean of four independent experiments testing immature day 7 DCs, expressed as the percentage of adherent cells relative to the maximal adhesion obtained in the absence of mAbs. Error bars correspond to a 10% imprecision as evaluated in our adhesion assays. Statistical analyses were performed using paired Student’s t test, and p values (∗, p ≤ 0.05; ∗∗, p ≤ 0.01) are shown for each set of conditions compared with the control conditions where DC adhesion was determined in the presence of irrelevant IgG1.

Close modal

Interestingly, mAbs directed against CD44v3 or CD44v6 could reduce the HA adhesion of DCs to residual levels of approximately 55% and 65%, respectively, while a combination of these mAbs almost totally inhibited DC adhesion (Fig. 3). It should be noted that the blocking effect observed in the presence of both anti-CD44v3 and anti-CD44v6 mAbs was stronger than that obtained using anti-pan CD44 mAbs despite the lower level of surface expression of the variant epitopes compared with total CD44 (Fig. 2). In contrast, an anti-CD44v9 mAb was totally ineffective at inhibiting DC adhesion to HA. These results indicate that CD44 variants containing v3 and v6 play an important role in the binding of DCs to HA, whereas v9-containing variants do not seem to contribute to this function.

Since a role for surface CD44 in triggering cell aggregation has been described in lymphoid and transfected cell lines (9, 34, 35), we decided to investigate the effects on DCs of the ligation of various CD44 epitopes, including variant region domains. To this aim, day 7 DCs were incubated in the presence of immobilized Abs directed against all forms of CD44 (J173, 5F12), against v3-, v6-, or v9-containing isoforms, or in the presence of control IgG. In parallel, DCs were cultured in dishes coated with a dilute solution of HA (50 μg/ml). Large cell aggregates were observed following 24-h incubation with either the mAb J173 (Fig. 4) or immobilized HA (data not shown), while aggregation was also obtained by culture of DCs in the presence of anti-CD44v6 (Fig. 4) and anti-CD44v9 (data not shown) mAbs. The aggregates formed by stimulation with J173 were often larger than those induced by treatment with anti-CD44v6 or anti-CD44v9, which could be due to the greater number of common CD44 epitopes on the cell surface or to the existence of a small DC population negative for variant CD44 expression (Fig. 2). The mAbs 5F12 and anti-CD44v3 did not provoke any detectable cell aggregation compared with control IgG (Fig. 4). To ensure that aggregation of DCs was not induced by contaminating endotoxin in HA or mAb samples, we regularly titrated endotoxin levels in our reagents and performed control cultures on dishes coated with solutions containing equivalent concentrations of endotoxin (see Materials and Methods). Furthermore, the anti-variant CD44 mAb solutions used for coating dishes were devoid of detectable endotoxin. It is noteworthy that anti-v6 and anti-v9, but not anti-v3, Abs induced cell aggregation, even though all three variants showed comparable levels of surface expression (Fig. 2). Our experiments therefore point to distinct capacities of CD44 variant isoforms to trigger DC aggregation.

FIGURE 4.

DC aggregation mediated by CD44 isoform triggering. Day 7 DCs were cultured for 24 h in petri dishes coated with anti-CD44 mAbs, HA, control IgG, or E. coli endotoxin. DC aggregation was induced in the presence of the anti-pan CD44 mAb J173 (upper right), anti-CD44v6 (lower right), anti-CD44v9, or HA (not shown), but not in the presence of the anti-pan CD44 mAb 5F12 (upper left), anti-CD44v3 (lower left), control IgG, or E. coli endotoxin (not shown).

FIGURE 4.

DC aggregation mediated by CD44 isoform triggering. Day 7 DCs were cultured for 24 h in petri dishes coated with anti-CD44 mAbs, HA, control IgG, or E. coli endotoxin. DC aggregation was induced in the presence of the anti-pan CD44 mAb J173 (upper right), anti-CD44v6 (lower right), anti-CD44v9, or HA (not shown), but not in the presence of the anti-pan CD44 mAb 5F12 (upper left), anti-CD44v3 (lower left), control IgG, or E. coli endotoxin (not shown).

Close modal

DC maturation is accompanied by the modulation of adhesion and costimulatory molecule expression and by the appearance of CD83 on the cell surface (5, 6, 29). In view of the aggregation observed in cultures stimulated through ligation of CD44 epitopes, the surface phenotype of DCs was analyzed by flow cytometry after 24-h incubation in the presence of immobilized anti-CD44 mAbs, control IgG, HA (50 μg/ml), or control endotoxin. Results from six independent experiments are summarized in Figure 5,a. DCs from cultures that had aggregated in the presence of the anti-pan CD44 mAb J173, anti-CD44v6, or anti-CD44v9 mAbs or HA showed enhanced surface expression of MHC class II, CD40, and CD86 (B7-2) Ags (Fig. 5,a) and of CD54 and CD80 (data not shown) compared with DCs cultured on dishes treated with control IgG or endotoxin. In contrast, no significant variation in the expression of these five maturation markers was found on cells from the nonaggregated cultures stimulated with 5F12 or anti-CD44v3. Figure 5,b (left panel) presents superimposed FACS profiles of DCs treated with 5F12, J173, or anti-CD44v9 and subsequently stained for class II (DR), CD40, and CD86 expression. Induction of these Ags on the surface of DCs treated with HA is shown in Fig. 5 b (right panel) relative to the endotoxin control. Together, these results indicate that ligation of CD44 by HA or anti-CD44 Abs that induce DC aggregation leads to a concomitant up-regulation of the expression of surface markers, reflecting enhanced cell maturation and potentially implicated in T cell interactions.

FIGURE 5.

Phenotypic maturation of DCs induced by ligation of CD44 isoforms. a, Surface expression of HLA-DR, CD40, CD86, and CD83 on DCs following 24-h culture in dishes coated with anti-CD44 mAbs (anti-pan CD44 J173 or 5F12, anti-CD44v3, -v6, or -v9), control IgG, HA, or E. coli endotoxin (ET). Staining was performed either indirectly (HLA-DR, CD40, and CD86) or directly (HLA-DR and CD83) in a series of five independent experiments. b, FACS profiles of HLA-DR, CD40, CD86, and CD83 surface expression on DCs cultured in the presence of the anti-pan CD44 mAbs 5F12 (left panel, plain profiles) and J173 (left panel, dotted profiles), anti-CD44v9 (left panel, thick profiles), control E. coli endotoxin (right panel, plain profile), or HA (right panel, thick profile). Controls for the direct staining of CD83 were obtained using phycoerythrin-conjugated IgG (not shown). Controls for the indirect staining of HLA-DR, CD40, and CD86 were obtained using irrelevant IgG followed by a FITC-conjugated secondary Ab (not shown). In this last case, increased nonspecific fluorescence was observed only for DCs stimulated with J173 and was due to residual mAb bound to cell membranes (mean fluorescence intensities of controls increased from about 5 to 30). Despite this background staining, increased levels of HLA-DR, CD40, and CD86 expression were clearly distinguished on the basis of mean fluorescence intensities on DCs stimulated with J173.

FIGURE 5.

Phenotypic maturation of DCs induced by ligation of CD44 isoforms. a, Surface expression of HLA-DR, CD40, CD86, and CD83 on DCs following 24-h culture in dishes coated with anti-CD44 mAbs (anti-pan CD44 J173 or 5F12, anti-CD44v3, -v6, or -v9), control IgG, HA, or E. coli endotoxin (ET). Staining was performed either indirectly (HLA-DR, CD40, and CD86) or directly (HLA-DR and CD83) in a series of five independent experiments. b, FACS profiles of HLA-DR, CD40, CD86, and CD83 surface expression on DCs cultured in the presence of the anti-pan CD44 mAbs 5F12 (left panel, plain profiles) and J173 (left panel, dotted profiles), anti-CD44v9 (left panel, thick profiles), control E. coli endotoxin (right panel, plain profile), or HA (right panel, thick profile). Controls for the direct staining of CD83 were obtained using phycoerythrin-conjugated IgG (not shown). Controls for the indirect staining of HLA-DR, CD40, and CD86 were obtained using irrelevant IgG followed by a FITC-conjugated secondary Ab (not shown). In this last case, increased nonspecific fluorescence was observed only for DCs stimulated with J173 and was due to residual mAb bound to cell membranes (mean fluorescence intensities of controls increased from about 5 to 30). Despite this background staining, increased levels of HLA-DR, CD40, and CD86 expression were clearly distinguished on the basis of mean fluorescence intensities on DCs stimulated with J173.

Close modal

CD83 has been identified as a specific surface marker predominantly (if not exclusively) expressed on mature cells of the DC lineage (29). Its expression, i.e., low to negative in nonaggregated DC cultures, was strongly induced when the cells had been stimulated through ligation of CD44 by J173, anti-CD44v9, or HA (Fig. 5, a and b), thus providing further evidence of a mature DC phenotype. In addition, the expression of v3-, v6-, and v9-containing CD44 variants was higher on aggregated DCs incubated in the presence of HA (data not shown), similar to the up-regulation of these isoforms observed during DC maturation induced by TNF-α (Fig. 2). Finally, cultures stimulated with J173, anti-CD44v9, or HA showed a marked increase (about twofold) in the expression of the ICAM-1 adhesion molecule compared with DCs treated with the mAb 5F12 or control IgG (data not shown).

Modification of the immunostimulatory potential of DCs following their in vitro maturation has been described by several authors (4, 5, 36), and mature DCs, in particular, are more potent inducers of T cell proliferation in MLR. To evaluate the consequences of the phenotypic changes triggered through binding of CD44 epitopes, DCs cultured for 24 h in the presence of anti-CD44 mAbs or HA were used to stimulate allogeneic PBLs or naive T cells. Induction of T cell proliferation over the control level could be measured following DC stimulation by CD44 ligands inducing aggregation, namely the J173, anti-CD44v6, and anti-CD44v9 mAbs or HA (Fig. 6). Conversely, culture of DCs in the presence of 5F12 or anti-CD44v3 mAbs, which did not induce DC maturation, did not augment their stimulatory capacity compared with that of cells cultured in the presence of control IgG (Fig. 6). Thus, the up-regulation of DC maturation markers was accompanied by an increase in T cell proliferation in MLR. This result together with those presented in Figure 3 show that the binding of common or variant CD44 domains, whether these are involved in adhesion to HA (like the epitopes of the J173 and anti-CD44v6 mAbs) or do not bind HA (like the CD44v9 epitope), can trigger the maturation of DCs in vitro.

FIGURE 6.

T cell proliferation in MLR induced by DCs stimulated through CD44 isoforms. After 24-h incubation with immobilized anti-CD44 mAbs (anti-pan CD44 J173 and anti-CD44v3, -v6, or -v9), control IgG, HA, or E. coli endotoxin (ET), various numbers of DCs were used to stimulate the proliferation of 1.5 × 105 allogeneic PBL. Black bars, 10,000 DCs/well; dark gray bars, 5,000 DCs/well; light gray bars, 2,500 DCs/well; white bars, 625 DCs/well. The data shown are from one experiment representative of five and are expressed as the mean of triplicates (±SEM). Statistical analyses were performed using paired Student’s t test, and p values (∗, p ≤ 0.08; ∗ p ≤ 0.05; ∗∗∗, p ≤ 0.01) are given for anti-CD44 mAb compared with control IgG1 treatment and for HA compared with ET treatment. Similar results were obtained using allogeneic naive T cells as responder cells (not shown).

FIGURE 6.

T cell proliferation in MLR induced by DCs stimulated through CD44 isoforms. After 24-h incubation with immobilized anti-CD44 mAbs (anti-pan CD44 J173 and anti-CD44v3, -v6, or -v9), control IgG, HA, or E. coli endotoxin (ET), various numbers of DCs were used to stimulate the proliferation of 1.5 × 105 allogeneic PBL. Black bars, 10,000 DCs/well; dark gray bars, 5,000 DCs/well; light gray bars, 2,500 DCs/well; white bars, 625 DCs/well. The data shown are from one experiment representative of five and are expressed as the mean of triplicates (±SEM). Statistical analyses were performed using paired Student’s t test, and p values (∗, p ≤ 0.08; ∗ p ≤ 0.05; ∗∗∗, p ≤ 0.01) are given for anti-CD44 mAb compared with control IgG1 treatment and for HA compared with ET treatment. Similar results were obtained using allogeneic naive T cells as responder cells (not shown).

Close modal

Lastly, we addressed the question of whether the binding of surface CD44 isoforms, in particular those involved in triggering DC maturation, could lead to the production of cytokines by DCs. This property has previously been described for the CD44 molecules expressed on monocytes (37), T cells (38), NK cells (12), and macrophages (39). The pattern of cytokine production was examined in DCs cultured for 24 h in the presence of HA (50 μg/ml) or immobilized anti-CD44 mAbs or on control plates. Cytokine transcripts were amplified using a semiquantitative RT-PCR method (Fig. 7,a). mRNAs encoding GM-CSF, IL-1β, IL-6, IL-8, IL-10, IL-12p35, IL-12p40, IL-15, MIP-1α, and TNF-α were readily detected in all samples, including controls (Fig. 7,a and data not shown), thus demonstrating the presence of basal levels of cytokine transcripts in both aggregated and nonaggregated DC cultures. There were no significant variations in the levels of transcripts encoding GM-CSF, IL-1β, IL-12p35, IL-12p40, IL-15, MIP-1α, or TNF-α between different DC cultures (data not shown). In contrast, IL-8 mRNA was clearly up-regulated in DCs incubated with J173 or anti-CD44v6, but not in DCs incubated with 5F12 or a control IgG (Fig. 7,a). Treatment with the mAb J173 also increased levels of IL-6 and IL-10 mRNAs, whereas DC stimulation with anti-CD44v6 did not significantly modulate levels of these transcripts (Fig. 7 a and data not shown).

FIGURE 7.

Cytokine production by DCs stimulated through CD44 isoforms. a, Semiquantitative PCR amplification of IL-6 and IL-8 transcripts (upper panels) isolated from DCs after 24-h culture in the presence of immobilized anti-CD44 mAbs (anti-pan CD44 J173 or 5F12, or anti-CD44v6) or control IgG. Serial 1/10 dilutions were performed for each mRNA sample and use of equivalent amounts of mRNA for the amplification of cytokine mRNAs was checked by parallel amplification of a β-actin fragment (lower panel). b, Cytokine secretion was quantified in 24-h culture supernatants of DCs incubated in the presence of immobilized anti-CD44 mAbs (anti-pan CD44 J173 or 5F12, and anti-CD44v3, -v6, or -v9), control IgG, HA, or control E. coli endotoxin (ET). It should be noted that the basal levels of cytokine secretion in unstimulated DCs could vary between donors. The result shown is from one experiment, representative of eight.

FIGURE 7.

Cytokine production by DCs stimulated through CD44 isoforms. a, Semiquantitative PCR amplification of IL-6 and IL-8 transcripts (upper panels) isolated from DCs after 24-h culture in the presence of immobilized anti-CD44 mAbs (anti-pan CD44 J173 or 5F12, or anti-CD44v6) or control IgG. Serial 1/10 dilutions were performed for each mRNA sample and use of equivalent amounts of mRNA for the amplification of cytokine mRNAs was checked by parallel amplification of a β-actin fragment (lower panel). b, Cytokine secretion was quantified in 24-h culture supernatants of DCs incubated in the presence of immobilized anti-CD44 mAbs (anti-pan CD44 J173 or 5F12, and anti-CD44v3, -v6, or -v9), control IgG, HA, or control E. coli endotoxin (ET). It should be noted that the basal levels of cytokine secretion in unstimulated DCs could vary between donors. The result shown is from one experiment, representative of eight.

Close modal

The release of cytokines following CD44 stimulation was next examined by ELISA analysis of supernatants from 24-h cultures. Figure 7,b shows the results of one experiment, representative of eight. IL-6 and IL-10 production was strongly enhanced in DCs stimulated with mAb J173 compared with that in DCs stimulated with a control IgG or HA (50 μg/ml) or with control endotoxin (Fig. 7,b, upper panel). In addition, IL-8 production was up-regulated not only by incubation with J173 or HA but also by treatment with anti-CD44v6 or anti-CD44v9 (Fig. 7,b, lower panel). Whereas the up-regulation of IL-6, IL-8, and IL-10 mRNAs in DC populations paralleled the increases in cytokine levels in the culture medium, no such variations were detected in IL-1β and TNF-α mRNAs. Since the production of these latter cytokines is known to be regulated at both the transcriptional and the post-transcriptional level (40), this may account for their increased secretion by DCs in the absence of mRNA up-regulation. TNF-α and low concentrations of IL-1β and GM-CSF were likewise detected in culture supernatants of DCs stimulated by J173 or HA or by ligation of v6- or v9-containing CD44 isoforms. IL-6 and IL-10 secretion was sometimes detectable after stimulation with anti-CD44v6 or anti-CD44v9, but these responses were generally weak relative to the cytokine levels obtained using J173 or HA. However, stimulation of DCs with a combination of anti-CD44v6 and anti-CD44v9 could increase the secretion of IL-6 and IL-10, whereas addition of anti-CD44v3 to either anti-CD44v6 or anti-CD44v9 did not enhance the secretion of these cytokines (data not shown). IL-12 (p70) was secreted by DCs treated with J173 or HA, but this cytokine was never detected in cultures stimulated with anti-variant CD44 mAbs even when these were used in combination. Significant induction of cytokine secretion was not observed following treatment of DCs with 5F12 or with anti-CD44v3. Our results therefore demonstrated that DC maturation induced by the ligation of common or variant CD44 epitopes on the cell surface was accompanied by cytokine secretion (Table II).

Table II.

Modification of DC phenotype and function through ligation of CD44 epitopesa

αCD44 v3αCC44 v6αCD44 v9HAJ1735F12
Inhibition of adhesion to HA − ND 
Aggregation − − 
Phenotypic maturation − − 
Induction of CD83 ND ND − 
ICAM-1 up-regulation ND ND − 
T cell proliferation in MLR − − 
Cytokine secretion − − 
αCD44 v3αCC44 v6αCD44 v9HAJ1735F12
Inhibition of adhesion to HA − ND 
Aggregation − − 
Phenotypic maturation − − 
Induction of CD83 ND ND − 
ICAM-1 up-regulation ND ND − 
T cell proliferation in MLR − − 
Cytokine secretion − − 
a

The effects of anti-pan CD44 mAbs (J173 and 5F12), anti-variant CD44 mAbs and HA were tested as described in Figures 3–7. Results are summarized as − (no effect), + (effect), or ND (not determined).

The aim of this study was to gain insight into the regulation and function of the variety of CD44 molecules expressed by human monocyte-derived DCs. It was shown that 1) the expression of CD44 isoforms is differentially regulated during the in vitro differentiation of DCs from blood monocytes and their maturation induced by TNF-α; 2) v3- and v6-containing isoforms play an important role in DC adhesion to the ECM component HA, whereas variants carrying v9 do not appear to contribute to this function; and 3) signaling through CD44 can be achieved not only by binding of HA or of the anti-pan CD44 mAb J173 to this receptor, but also by selective stimulation of v6- or v9-containing CD44 variants. This signaling induces DC maturation, as reflected by phenotypic changes and increased T cell stimulatory capacity in MLR, cell aggregation in culture, and cytokine secretion. The effects of CD44 ligation on DC phenotype and function are summarized in Table II.

Transcripts encoding v3-, v6-, and v9-containing CD44 variants are differentially regulated during monocyte differentiation into DCs. Although levels of v3- and v6-containing mRNAs were relatively stable, two transcripts containing v9 were markedly reduced in DCs compared with fresh monocytes. One of these v9-containing transcripts was identified as CD44v8-v10, the epithelial form of CD44 (CD44E) (22). No changes were apparent in the general splicing pattern of CD44, as we failed to detect DC-specific mRNA species. Maturation of DCs in the presence of TNF-α increased the surface expression of v3-, v6-, and v9-containing isoforms compared with immature DCs, as did incubation of DCs with LPS, another inducer of DC maturation (41) (unpublished results). These findings are in agreement with other studies demonstrating the modulation of CD44 expression on DCs by TNF-α (5, 26). An increase in the CD44v9 epitope on the surface of mature DCs has previously been reported by Sallusto and Lanzavecchia (5). Interestingly, v9-containing variants are also known to be expressed by tissue macrophages at inflammation sites (42), suggesting that these variants may play a role in local immune responses.

Anti-CD44v3 and anti-CD44v6 mAbs were able to partially block the binding of DCs to immobilized HA, while a combination of these two mAbs almost totally inhibited DC adhesion, a blocking effect even stronger than that observed with either of the anti-pan CD44 mAbs, J173 and 5F12. Our data therefore suggest that v3- and v6-containing isoforms play a major role in the adhesion of DCs to immobilized HA. On the contrary, an anti-CD44v9 mAb was unable to block the binding of DCs to HA in our adhesion assays. The capacity of v9-containing CD44 isoforms to bind HA has long been a matter of debate (9, 10, 22), and more recently, the CD44E (v8–v10) and other v9-containing isoforms have been demonstrated to poorly bind immobilized HA (17). Moreover, other authors have reported an anti-CD44v9 mAb to be ineffective in blocking the adhesion of human DCs to frozen sections of paracortical lymph nodes, whereas anti-CD44v6 and anti-pan CD44 mAbs could block this binding (25). These results and those of our study support the hypothesis that DC adhesion to the lymph nodes could be mediated by the binding of v3- or v6-containing isoforms to HA present in this organ, but through a mechanism independent of CD44v9.

The engagement of CD44 molecules expressed on the surface of DCs, either by specific mAbs or by the ECM ligand HA, can induce their phenotypic and functional maturation. Thus, HLA class II molecules, CD40, ICAM-1, CD80, and CD86 were up-regulated and CD83 was induced on the surface of DCs stimulated by CD44 ligation, and these cells showed increased allogeneic T cell stimulatory capacity. Interestingly, only one of the two anti-pan CD44 mAbs, J173 but not 5F12, was able to stimulate DC maturation, whereas both these Abs could inhibit cell attachment to HA. Since other studies have demonstrated that CD44 undergoes conformational changes in the course of ligand binding (10), it is possible that the difference between the effects of these mAbs results from the blockage by 5F12 of a conformational change necessary for the signal transduction event. A number of investigations have implicated CD44 in signal transduction events leading to the activation of diverse cell types, and in the case of lymphocytes, CD44 bound by HA was found to behave as a costimulatory molecule (11, 13, 14, 33, 43, 44). Our study indicates that HA can in addition induce the maturation of DCs. Control experiments were performed to check that contamination of HA preparations with endotoxin was not responsible for this effect. Thus, culture of DCs on plastic dishes coated with a concentration of endotoxin equivalent to that found in HA solutions failed to trigger their maturation. Nevertheless, it is still possible that HA and endotoxin could bind to each other, resulting in increased endotoxin deposition on culture dishes in the presence of HA, or that the DC stimulatory activity of HA may be seen only in the presence of low levels of endotoxin. It should also be noted that another ECM component, type 1 collagen, has been reported to induce the maturation of DCs derived from mouse liver progenitors (45). Interestingly, CD44 modified by chondroitin sulfate has been shown to bind to type 1 collagen (10, 46), suggesting that DC maturation might also be triggered by interaction of CD44 with collagen.

Stimulation through CD44 has been shown to produce homotypic cell aggregation in a variety of hemopoietic cells (34, 35). Cell clustering, which is likewise a characteristic feature of our mature DC cultures, probably results from the up-regulation of several surface molecules mediating cell-cell interactions, and preliminary results suggest, for instance, that ICAM-1 (CD54) is one receptor involved in the formation of DC aggregates (H. Haegel-Kronenberger, unpublished observations). Previous studies have also implicated mAb binding to CD44 in the formation of clusters between blood DCs and T cells (47). In vivo, the binding of CD44 molecules present on the surface of DCs may favor both direct interactions and cooperation among DCs or between DCs and T cells. Thus, CD44v6 has been implicated in the onset of a primary immune response in the mouse (48). In addition, the standard form of CD44 has been shown to stimulate T cell proliferation when expressed by APC (49), suggesting that increased CD44 expression on the surface of DCs may modulate T cell stimulation, either through a direct effect on T cells or by inducing cytokine secretion in DCs.

The maturation of DCs could be induced by stimulation with an anti-CD44v6 or an anti-CD44v9 mAb, but not by treatment with an anti-CD44v3 mAb. This is the first report showing that variant isoforms of CD44 carrying the v6 or v9 region possess signaling activity. Variants containing v6 are therefore potentially involved in both cell attachment to HA and the triggering of DC maturation. Whether signaling through CD44v6 occurs in vivo after the binding of HA by DCs, particularly in the lymph nodes (19, 25), remains to be determined and may be of importance for the physiologic role of human DCs. In the case of v9-containing CD44 isoforms, while our results indicate that they can trigger a maturation signal in DCs, there is still no positive evidence that these variants bind HA. Thus, in contrast to the v6-containing variants implicated in both adhesion to HA and signal transduction, v9-containing isoforms may contribute only to this latter function. The possible existence of a specific ligand for CD44v9 other than HA that is capable of triggering a maturation signal in vivo remains a subject for further investigation. Finally, v3-containing CD44 isoforms seem to mediate DC attachment to HA without inducing signal transduction. This lack of a stimulatory effect might be related to the anti-CD44v3 mAb used, which could block adhesion but be unable to trigger a signal, in a manner similar to that of the anti-pan CD44 mAb 5F12. Alternatively, it is possible that the lack of signaling activity is an intrinsic feature of v3-containing CD44 isoforms. In summary, concerning the adhesive and signaling functions of CD44 variants, the results of this study point to distinct capacities of the v3-, v6-, and v9-containing isoforms expressed by human DCs to mediate either one or both of these properties.

Secretion of IL-8, IL-1β, TNF-α, and GM-CSF increased during the maturation of DCs in response to all CD44 stimuli, namely HA and the mAbs J173, anti-CD44v6, and anti-CD44v9. The DC maturation observed under these conditions might well result from secretion into the culture medium of TNF-α and IL-1β, acting in an autocrine fashion. It is noteworthy that monocytes have been shown to secrete IL-1β and TNF-α and that alveolar macrophages produce IL-8 in response to CD44 triggering (37, 39). This would, in fact, appear to be a common feature of myeloid cell lineages stimulated through CD44. IL-6, IL-10, and IL-12 production were also triggered following CD44 ligation by HA or J173. In contrast, when v6- or v9-containing variants were bound by mAbs, secretion of IL-6 and IL-10 was either not induced or was induced only at low levels compared with those obtained with anti-pan CD44 stimuli. Simultaneous stimulation with anti-CD44v6 and anti-CD44v9 nevertheless triggered the release of larger amounts of these cytokines. In the case of IL-12, secretion was not observed following treatment with anti-CD44v6 or CD44v9 mAbs, either alone or in combination. Rather than a specific property of variant CD44 isoforms, this may reflect the requirement for binding of a greater number of CD44 molecules (and hence of the standard or common form) on the cell surface, to trigger IL-6, IL-10, or IL-12 secretion. The pattern of cytokine release induced in our DC cultures is characteristic of mature DC lineage cells (41, 50) and therefore provides additional evidence for the mature phenotype of DCs stimulated through CD44.

Our observations suggest that the in vivo engagement of CD44 epitopes on DCs at sites of inflammation could contribute to several aspects of the immune response. Firstly, T cell responses to antigenic stimulation may be optimized notably through the action of cytokines involved in T cell activation or differentiation. Thus, IL-6 secreted from skin and lymph node DCs behaves as a T cell costimulatory signal (51). IL-10 can skew the Th1/Th2 balance to a Th2-type response (52), while IL-12 has been shown to act mainly as a stimulator of Th1-type responses (53), suggesting that the balance between the IL-10 and IL-12 produced by DCs may regulate Th responses. Secondly, the recruitment and/or differentiation of other immune cell types, including DCs or DC precursors, may be facilitated through the secretion of other cytokines. In this respect, the proinflammatory chemokine IL-8 is notably involved in the transendothelial migration of neutrophils, monocytes, and lymphocytes (54, 55). TNF-α and IL-1β participate in inflammatory responses and regulate the maturation of DCs (56, 57). Moreover, TNF-α can in combination with GM-CSF induce the differentiation of DCs from CD34+ precursors (58), while GM-CSF also plays a role in the differentiation of DCs from blood monocytes. Thus, the triggering of surface CD44 molecules on DCs may regulate not only the functional maturation of DCs but, more generally, the efficiency of the immune response.

We thank Dr. Eric Tartour for the gift of oligonucleotides, H.T.L. and Centre d’Analyse et de Recherches Experimentales de Fougères (Fougères, France) for kindly providing HA, Dr. Anita Stierlé and Centeon (Lingolsheim, France) for the titration of endotoxins, Huguette Bausinger and Dominique Fricker for technical assistance, Juliette Mulvihill for reviewing the English of the manuscript, and Gérard Vetter for photographs.

1

This work was supported by grants from Institut National de la Santé et de la Recherche Médicale (CJF 94-03) and the Agence Française du Sang (FORTS 96) and by the Etablissement de Transfusion Sanguine de Strasbourg.

2

This paper is dedicated to the memory of Professor Francis Oberling, who died on January 2, 1998.

4

Abbreviations used in this paper: DC, dendritic cell; HA, hyaluronate; ECM, extracellular matrix; GM-CSF, granulocyte-macrophage CSF; rh, recombinant human; MIP-1α, macrophage inflammatory protein-1α; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide.

1
Steinman, R. M..
1991
. The dendritic cell system and its role in immunogenicity.
Annu. Rev. Immunol.
9
:
271
2
Cumberbatch, M., I. Kimber.
1995
. Tumor-necrosis factor-α is required for accumulation of dendritic cells in draining lymph nodes and for optimal contact sensitization.
Immunology
84
:
31
3
Roake, J. A., A. S. Rao, P. J. Morris, C. P. Larsen, D. F. Hankins, J. M. Austyn.
1995
. Dendritic cell loss from nonlymphoid tissues after systemic administration of lipopolysaccharide, tumor necrosis factor, and interleukin 1.
J. Exp. Med.
181
:
2237
4
Schuler, G., R. M. Steinman.
1985
. Murine epidermal Langerhans cells mature into potent immunostimulatory dendritic cells in vitro.
J. Exp. Med.
161
:
526
5
Sallusto, F., A. Lanzavecchia.
1994
. Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage colony-stimulating factor plus interleukin 4 and downregulated by tumor necrosis factor α.
J. Exp. Med.
179
:
1109
6
Steinman, R. M., K. Inaba, G. Schuler.
1995
. Cutaneous dendritic cells: distinctive antigen-presenting cells for experimental models and disease states. H. Moll, ed.
The Immune Functions of Epidermal Langerhans Cells
1
-13. Springer Verlag, Heidelberg.
7
Girolomoni, G., P. Ricciardi-Castagnoli.
1997
. Dendritic cells hold promise for immunotherapy.
Immunol. Today
18
:
102
8
Aruffo, A., I. Stamenkovic, M. Melnick, C. B. Underhill, B. Seed.
1990
. CD44 is the principal cell surface receptor for hyaluronate.
Cell
61
:
1303
9
Underhill, C..
1992
. CD44: the hyaluronan receptor.
J. Cell Sci.
103
:
293
10
Lesley, J., R. Hyman, P. W. Kincade.
1993
. CD44 and its interaction with extracellular matrix.
Adv. Immunol.
54
:
271
11
Aruffo, A..
1996
. CD44: one ligand, two functions.
J. Clin. Invest.
98
:
2191
12
Galandrini, R., M. Piccoli, L. Frati, A. Santoni.
1996
. Tyrosine-kinase-dependent activation of human NK cell function upon triggering through CD44 receptor.
Eur. J. Immunol.
26
:
2807
13
Taher, T. E. I., L. Smit, A. W. Griffioen, E. J. M. Schilder-Tol, J. Borst, S. T. Pals.
1996
. Signaling through CD44 is mediated by tyrosine kinases: association with p56lck in T lymphocytes.
J. Biol. Chem.
271
:
2863
14
Sconocchia, G., J. A. Titus, D. M. Segal.
1997
. Signaling pathways regulating CD44-dependent cytolysis in natural killer cells.
Blood
90
:
716
15
Lesley, J., N. English, A. Perschl, J. Gregorov, R. Hyman.
1995
. Variant cell lines selected for alterations in the function of the hyaluronan receptor CD44 show differences in glycosylation.
J. Exp. Med.
182
:
431
16
Katoh, S., Z. Zheng, K. Oritani, T. Shimozato, P. W. Kincade.
1995
. Glysosylation of CD44 negatively regulates its recognition of hyaluronan.
J. Exp. Med.
182
:
419
17
Bennett, K. L., B. Modrell, B. Greenfield, A. Bartolazzi, I. Stamenkovic, R. Peach, D. G. Jackson, F. Spring, A. Aruffo. Regulation of CD44 binding to hyaluronate by glycosylation of variably spliced exons.
J. Cell Biol.
131
:
1995b
1623
18
Skelton, T. P., C. Zeng, A. Nocks, I. Stamenkovic.
1998
. Glycosylation provides both stimulatory and inhibitory effects on cell surface and soluble CD44 binding to hyaluronan.
J. Cell Biol.
140
:
431
19
Herrlich, P., M. Zöller, S. T. Pals, H. Ponta.
1993
. CD44 splice variants: metastases meet lymphocytes.
Immunol. Today
14
:
395
20
Screaton, G. R., M. V. Bell, J. I. Bell, D. J. Jackson.
1993
. The identification of a new alternative exon with highly restricted tissue expression in transcripts encoding the mouse Pgp-1 (CD44) homing receptor.
J. Biol. Chem.
268
:
12235
21
Bennett, K. L., D. G. Jackson, J. C. Simon, E. Tanczos, R. Peach, B. Modrell, I. Stamenkovic, G. Plowman, A. Aruffo. CD44 isoforms containing exon v3 are responsible for the presentation of heparin-binding growth factor.
J. Cell Biol.
128
:
1995a
687
22
Stamenkovic, I., A. Aruffo, M. Amiot, B. Seed.
1991
. The hematopoietic and epithelial forms of CD44 are distinct polypeptides with different adhesion potentials for hyaluronate-bearing cells.
EMBO J.
10
:
343
23
Lokeshwar, V. B., N. Iida, L. Y. W. Bourguignon.
1996
. The cell adhesion molecule, GP116, is a new CD44 variant (ex14/v10) involved in hyaluronic acid binding and endothelial cell proliferation.
J. Biol. Chem.
271
:
23853
24
Aiba, S., S. Nakagawa, H. Ozawa, K. Miyake, H. Yagita, H. Tagami.
1993
. Up-regulation of α4 integrin on activated Langerhans cells: analysis of adhesion molecules on Langerhans cells relating to their migration from skin to draining lymph nodes.
J. Invest. Dermatol.
100
:
143
25
Weiss, J. M., J. Sleeman, A. C. Renkel, H. Dittmar, C. C. Termeer, S. Taxix, N. Howells, M. Hofmann, G. Köhler, E. Schöpf, H. Ponta, P. Herrlich, J. Simon.
1997
. An essential role for CD44 variant isoforms in epidermal Langerhans cell and blood dendritic cell function.
J. Cell Biol.
137
:
1137
26
Osada, A., H. Nakashima, M. Furue, K. Tamaki.
1995
. Up-regulation of CD44 expression by TNF-α is neutralized by IL-10 in Langerhans cells.
J. Invest. Dermatol.
105
:
124
27
Haegel, H., M. Hofmann, C. Tölg, R. Ceredig.
1993
. Activated astrocytes and T cells express similar CD44 variants: role of CD44 in astrocyte-T cell binding.
J. Cell Biol.
122
:
1067
28
Faradji, A., A. Bohbot, M. Schmitt-Goguel, J. C. Siffert, S. Dumont, M. L. Wiesel, Y. Piemont, A. Eischen, J. P. Bergerat, J. Bartholeyns, P. Poindron, J. P. Witz, F. Oberling. Large scale isolation of human blood monocytes by continuous flow centrifugation leukopheresis and counterflow centrifugation elutriation for adoptive cellular immunotherapy in cancer patients.
J. Immunol. Methods
174
:
\n\n
297
29
Zhou, L.-J., T. F. Tedder.
1996
. CD14+ blood monocytes can differentiate into functionally mature CD83+ dendritic cells.
Proc. Natl. Acad. Sci. USA
93
:
2588
30
Chomczynski, P., N. Sacchi.
1987
. Single step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction.
Anal. Biochem.
162
:
156
31
Mossmann, T..
1983
. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays.
J. Immunol. Methods
65
:
55
32
Liao, H.-X., D. M. Lee, M. C. Levesque, B. F. Haynes.
1995
. N-terminal and central regions of the human CD44 extracellular domain participate in cell surface hyaluronate binding.
J. Immunol.
155
:
3938
33
Galandrini, R., E. Galluzzo, N. Albi, C. E. Grossi, A. Velardi.
1994
. Hyaluronate is costimulatory for human T cell effector functions and binds to CD44 on activated T cells.
J. Immunol.
153
:
21
34
Koopman, G., Y. van Kooyk, M. de Graaff, C. J. L. M. Meyer, C. G. Fidgor, S. T. Pals.
1990
. Triggering of the CD44 antigen on T lymphocytes promotes T cell adhesion through the LFA-1 pathway.
J. Immunol.
145
:
3589
35
Belitsos, P. C., J. E. K. Hildreth, J. T. August.
1990
. Homotypic cell aggregation induced by anti-CD44(Pgp-1) monoclonal antibodies and related to CD44 (Pgp-1) expression.
J. Immunol.
144
:
1661
36
Caux, C., C. Massacrier, B. Vanbervliet, B. Dubois, C. Van Kooten, I. Durand, J. Banchereau.
1994
. Activation of human dendritic cells through CD40 cross-linking.
J. Exp. Med.
180
:
1263
37
Webb, D. S. A., Y. Shimizu, G. A. van Seventer, S. Shaw, T. L. Gerrard.
1990
. LFA-1, CD44 and CD45: physiologic triggers of human monocyte tumor necrosis factor and interleukin-1 release.
Science
249
:
1295
38
Denning, S. M., P. T. Le, K. H. Singer, B. F. Haynes.
1990
. Antibodies against the CD44 p80 lymphocyte homing receptor augment human peripheral blood T cell activation.
J. Immunol.
144
:
7
39
McKee, C. M., M. B. Penno, M. Cowman, M. D. Burdick, R. M. Strieter, C. Bao, P. W. Noble.
1996
. Hyaluronan (HA) fragments induce chemokine gene expression in alveolar macrophages.
J. Clin. Invest.
98
:
2403
40
Schindler, R., B. D. Clark, C. A. Dinarello.
1990
. Dissociation between IL-1β mRNA and protein synthesis in human peripheral blood mononuclear cells.
J. Biol. Chem.
265
:
10232
41
Verhasselt, V., C. Buelens, F. Willems, D. de Groote, N. Haeffner-Cavaillon, M. Goldman.
1997
. Bacterial lipopolysaccharide stimulates the production of cytokines and the expression of costimulatory molecules by human peripheral blood dendritic cells.
J. Immunol.
158
:
2919
42
Levesque, M. C., B. F. Haynes.
1996
. In vitro culture of human peripheral blood monocytes induces hyaluronan binding and up-regulates monocyte variant CD44 isoform expression.
J. Immunol.
156
:
1557
43
Bourguignon, L. Y. W., V. B. Lokeshwar, X. Chen, W. G. L. Kerrick.
1993
. Hyaluronic acid-induced lymphocyte signal transduction and HA receptor (GP85/CD44)-cytoskeleton interaction.
J. Immunol.
151
:
6634
44
Rafi, A., M. Nagarkatti, P. S. Nagarkatti.
1997
. Hyaluronate-CD44 interactions can induce murine-B cell activation.
Blood
89
:
2901
45
Lu, L., J. Woo, A. S. Rao, Y. Li, S. C. Watkins, S. Qian, T. E. Starzl, A. J. Demetris, A. W. Thomson.
1994
. Propagation of dendritic cell progenitors from normal mouse liver using granulocyte/macrophage colony-stimulating factor and their maturational development in the presence of type-1 collagen.
J. Exp. Med.
179
:
1823
46
Faassen, A. E., J. A. Schrager, D. J. Klein, T. R. Oegema, J. R. Couchman, J. B. McCarthy.
1992
. A cell surface chondroitin sulfate proteoglycan, immunologically related to CD44, is involved in type-1 collagen-mediated melanoma cell motility and invasion.
J. Cell Biol.
116
:
521
47
Scheeren, R. A., G. Koopman, S. Van der Baan, C. J. L. M. Meijer, S. T. Pals.
1991
. Adhesion receptors involved in clustering of blood dendritic cells and T lymphocytes.
Eur. J. Immunol.
21
:
1101
48
Moll, J., A. Schmidt, H. van der Puten, R. Plug, H. Ponta, P. Herrlich, M. Zöller.
1996
. Accelerated immune response in transgenic mice expressing rat CD44v4–v7 on T cells.
J. Immunol.
156
:
2085
49
Guo, Y., Y. Wu, S. Shinde, M.-S. Sy, A. Aruffo, Y. Liu.
1996
. Identification of a costimulatory molecule rapidly induced by CD40L as CD44H.
J. Exp. Med.
184
:
955
50
Zhou, L. J., T. F. Tedder.
1995
. A distinct pattern of cytokine gene expression by human CD83+ blood dendritic cells.
Blood
86
:
3295
51
Cumberbatch, M., R. J. Dearman, I. Kimber.
1996
. Constitutive and inducible expression of interleukin-6 by Langerhans cells and lymph node dendritic cells.
Immunology
87
:
513
52
De Smet, T., M. Van Mechelen, G. De Becker, J. Urbain, O. Leo, M. Moser.
1997
. Effect of IL-10 on dendritic cell maturation and function.
Eur. J. Immunol.
27
:
1229
53
Lamont, A. G., L. Adorini.
1996
. IL-12: a key cytokine in immune regulation.
Immunol. Today
17
:
214
54
Oppenheim, J. J., C. O. C. Zachariae, N. Mukaida, K. Matsushima.
1991
. Properties of the novel proinflammatory supergene “intercrine” cytokine family.
Annu. Rev. Immunol.
9
:
617
55
Santamaria Babi, L. F., B. Moser, M. T. Perez Soler, R. Moser, P. Loetscher, B. Villiger, K. Blaser, C. Hauser.
1996
. The interleukin-8 receptor B and CXC chemokines can mediate transendothelial migration of human skin homing T cells.
Eur. J. Immunol.
26
:
2056
56
Enk, A. H., V. L. Angeloni, M. C. Udey, S. I. Katz.
1993
. An essential role for Langerhans cell-derived IL-1β in the initiation of primary immune responses in skin.
J. Immunol.
150
:
3698
57
Sallusto, F., M. Cella, C. Danieli, A. Lanzavecchia.
1995
. Dendritic cells use macropinocytosis and the mannose receptor to concentrate macromolecules in the major histocompatibility complex class II compartment: downregulation by cytokines and bacterial products.
J. Exp. Med.
182
:
389
58
Caux, C., C. Dezutter Dambuyan, D. Schmitt, J. Banchereau.
1992
. GM-CSF and TNF-α cooperate in the generation of dendritic Langerhans cells.
Nature
360
:
258