| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Servicio de Inmunología and
Hematología, Hospital General Universitario Gregorio Marañón, Madrid, Spain; and
Departamento de Inmunología, Centro de Investigaciones Biológicas, Madrid, Spain
Correspondence: Dr. Paloma Sánchez-Mateos, Hospital Gregorio Marañón, Servicio de Inmunología, C/ Dr. Esquerdo 46, 28007 Madrid, Spain. E-mail: inmunoonc{at}hispacom.net
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: chemotaxis • myeloid DCs • plasmacytoid DCs • CXCL12
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(MIP-1)], CCL4 (MIP-1ß), CCL5 [regulated on activation, normal T expressed and secreted (RANTES)], CCL7 [monocyte chemoattractant protein-3 (MCP-3)], CCL13 (MCP-4), and CCL22 (macrophage-derived chemokine), although CCL20 (MIP-3) only appears to induce migration of CD34-DCs. It is now well established that upon maturation, a striking change in the pattern of DC migration occurs. Hence, maturing DCs exhibit a rapid down-modulation of inflammatory chemokine receptor expression and function and the concomitant up-regulation of CCR7 [8
9
10
]. In this manner, the CCR7 ligands CCL19 (MIP-3ß) and CCL21 [secondary lymphoid-tissue chemokine (SLC)], which are produced in lymphoid tissues, facilitate the emigration of maturing DCs from inflamed tissues and direct the homing of DCs to the T cell areas of lymphoid organs through lymphatic vessels [10
, 11
]. However, the in vitro differentiation of DCs might be unrelated to the in vivo process [12
], and the migratory properties of in vivo-differentiated DCs and DC precursors remain largely unknown.
Emigration of DCs from blood into tissue requires adhesion to and transmigration through vascular endothelium. Like other leukocytes, a number of adhesion molecules expressed on the DC or the endothelium [e.g., selectins, ß1 (CD29) and ß2 (CD18) integrins, vascular cell adhesion molecule-1 (VCAM-1), platelet-endothelial cell adhesion molecule-1 (PECAM-1/CD31), and intercellular adhesion molecule-1 (ICAM-1)] may be involved in the transmigration of circulating DCs. In this regard, it has been shown that the transmigration of in vitro-derived mono-DC is mediated by ß2 integrins and PECAM-1 across resting endothelium and by 4ß1 (CD49e/CD29) across activated endothelium [13
, 14
].
A heterogeneous population of DCs (or DC precursors) is found circulating in blood, which is currently identified by their high expression of human leukocyte antigen (HLA)-DR and lack of specific lineage markers found on other leukocytes [15
, 16
]. The physiological role of blood DCs is unclear, although it is thought that most of them are en route from the bone marrow to peripheral tissues or from nonlymphoid tissues to the regional lymph nodes and spleen. Based on the differential expression of CD11c, two main DC subsets have been originally identified in human blood [17
]. The myeloid (CD11c+) DC population differentiates into mature DCs in response to inflammatory stimuli and expresses granulocyte macrophage-colony stimulating factor receptor and other myeloid cell markers, including CD13 and CD33 [18
]. Conversely, the plasmacytoid (CD123+) DC subset lacks myeloid cell markers, expresses high levels of the interleukin (IL)-3 receptor (CD123), and produces large amounts of type I interferon (IFN) upon exposure to viruses or bacteria [19
, 20
].
Transendothelial migration (TEM) studies of blood DCs have been limited by the paucity of these cells, as they represent less than 1% of circulating peripheral blood mononuclear cells (PBMCs), as well as by their ability to rapidly initiate a program of maturation in culture or during cell purification procedures, which may modify their pattern of chemotactic response [10 ]. To overcome these problems, we have performed transmigration assays with unfractionated PBMCs and evaluated the number and type of blood DCs by flow cytometry before and after migration. In this manner, we have systematically examined the chemotactic response of blood DCs to a wide panel of chemokines through resting or activated endothelium and the adhesion molecules involved in their transendothelial passage. Our results indicate that the two main subsets of blood DCs display distinct patterns of TEM in response to chemokines and require differential endothelial adhesion support.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 TS2/7, anti-2 Tea1/41, anti-3 VJ1/18, anti-4 HP1/2 (IgG1), anti-ß2 Lia3/2 (IgG1), anti-CD11a TS1/11, anti-CD11b Bear-1, anti-CD11c HC1/1 integrins, anti-ICAM-1 HU5/3 (IgG1), anti-ICAM-3 HP2/19, anti-PECAM-1 TP1/15 (IgG2a), anti-VCAM-1 4B9 (IgG1), and anti-HLA-A,B,C W6/32 (IgG2a) were a generous gift from Dr. Francisco Sánchez-Madrid (Hospital de la Princesa, Madrid, Spain). The anti-DC-specific ICAM-3-grabbing nonintegrin (DC-SIGN; CD209) MR-1 mAb has been described previously [21
]. The anti-ß3, anti-V, and anti-5 VC5 (IgG1) integrins and anti-CCR5 (clone 2D7), anti-CXCR1 (clone 5A12), and anti-CXCR2 (clone 6C6) mAb were purchased from PharMingen (San Diego, CA). The anti-CCR1 (clone #53504.111), anti-CCR3 (clone #61828.111), anti-CCR6 (clone #53103.111), anti-CCR7 (clone #150503), anti-CXCR3 (clone #49801.111), and anti-CXCR4 (clone #44717.111) mAb were purchased from R&D Systems (Minneapolis, MN). The anti-CCR2 mAb (clone #04) was provided by Dr. Carlos Martínez-A (Centro Nacional de Biotecnología, Madrid, Spain). Human recombinant CCL2, CCL7, CCL8, CCL13, CCL19, SLC, CXCL8, CXCL9, and CXCL10 were purchased from PeproTech (London, UK). CCL3, CCL4, CCL5, CCL20, CCL22, and CXCL12 were obtained from R&D Systems. The activity of all the chemokines used was assayed in the appropriated responsive cell type (e.g., tumor-infiltrating lymphocytes were used to control CXCL9 and CXCL10 activity; granulocytes for CXCL8).
Cells and cell cultures
PBMCs were obtained from buffy coats by centrifugation over Ficoll-Hypaque (LymphoprepTM, Nycomed Pharma AS, Oslo, Norway) and were resuspended in RPMI (Gibco-BRL Life Technologies, Paisley, Scotland), supplemented with 10% heat-inactivated fetal calf serum (FCS; Gibco-BRL Life Technologies). DCs were purified with a blood DC cell isolation kit (Miltenyi Biotec, Bergish Gladblach, Germany) to a purity 
95%. All purification steps were performed at 4°C.
In some experiments, PBMCs were cultured at 37°C in the presence of polyinosinic:polycytidylic acid [poly(I:C); 50 µg/ml; Sigma Chemical Co., St. Louis, MO]; oligodeoxynucleotides (ODN) in their phosphorothioate form AAC-30-CpG (5'-ACCGATAACGTTGCCGGTGACGGCACCACG) and nonstimulatory-GpC (5'-TGCTGCTTTTGTGCTTTTGTGCTT; 1 µM; Amersham Pharmacia Biotech, Amersham, UK), as described [22 ]; recombinant human CD40 ligand kit (CD40L 100 ng/ml; enhancer 1 µg/ml; Alexis Biochemicals, Lausen, Switzerland); and IL-3 (10 ng/ml; PeproTech). In mAb-blocking experiments, PBMCs were preincubated with human IgG at 250 µg/ml for 10 min at room temperature to block FcR binding and then with mAb at 20 µg/ml for 15 min before TEM assays.
Human endothelial cells (ECs) from umbilical vein (HUVEC) were obtained and cultured as described previously [23
]. Briefly, umbilical veins were cannulated, washed, and incubated with 0.1% collagenase P (Boehringer Mannheim GmbH, Mannheim, Germany) for 20 min at 37°C. Cells were seeded on tissue-culture flasks (Costar, Corning, NY) coated with 0.5% gelatine (Sigma Chemical Co.) and grown in 199 medium (Gibco-BRL Life Technologies) supplemented with 20% FCS, 50 IU/ml penicillin, 50 µg/ml streptomycin (ICN Biomedicals, Costa Mesa, CA), 250 µg/ml fungizone (Squibb Industria Farmacéutica, Barcelona, Spain), 50 µg/ml EC growth supplement (prepared from bovine brain), and 100 µg/ml heparin (Sigma Chemical Co.) and were used up to the third passage. For TEM assays, polycarbonate transwell membranes (COSTAR, Costar Europe, Badhoevedorp, The Netherlands) were coated with 20 µg/ml fibronectin (Sigma Chemical Co.) at 37°C for 1 h before HUVEC seeding onto membranes (105 cell/transwell). HUVEC were grown as a monolayer for 24 h in a total volume of 200 µl PANSERIN 401 serum-free medium (PAN Biotech GmbH, Aidenbach, Germany). To confirm the confluence of the HUVEC on the membrane and that ECs grew only on the upper side of the filter, sample monolayers were routinely stained with Diff-Quik (American Scientific Products, McGraw Park, IL). To stimulate EC monolayers, tumor necrosis factor (TNF-; Alexis Biochemicals) at 50 ng/ml was added for 12 h. In mAb-blocking experiments, HUVEC monolayers grown on transwells were preincubated with mAb at 20 µg/ml for 15 min before TEM assays.
TEM and chemotaxis assays
TEM and chemotaxis assays were performed using the transwell system (pore size 3 µm) as previously reported [5
] with some modifications. The polycarbonate transwell membranes were used uncoated for bare chemotaxis assays or coated with an EC monolayer for TEM assays. The endothelial monolayers were rinsed once with RPMI medium, and then, 1 x 106 PBMCs were added in 100 µl RPMI plus 10% FCS to the upper inserts. Inserts were transferred to wells (lower chambers) of 24-well plates containing 600 µl fresh RPMI. In some experiments, various chemokines were also added at different concentrations to the lower chambers. Cells were allowed to transmigrate for 2 h at 37°C or the indicated time in chemotaxis assays. After migration, cells in the lower chamber were collected. Additionally, to recover eventually migrated cells bound to the lower side of the membrane, the lower surface of the insert was rinsed with 600 µl ice-cold phosphate-buffered saline containing 10 mM EDTA. The cells that had migrated to the lower chamber were combined with those detached from the filter, and the total number of migrated cells was calculated by quantitative immunofluorescence flow cytometry (TruCOUNT analysis) in a FACScalibur flow cytometer (Becton Dickinson) as described previously [24
]. TruCOUNT tubes, each one containing a known number of beads, were used to determine the absolute counts of DCs. A minimum of 500 Lin-, HLA-DR+ DC events, and/or 40,000 beads was acquired for each analysis. Each sample was analyzed in duplicate. The absolute number of DCs in each sample was calculated as the average of the duplicate tubes, each being determined by comparing the cellular events with bead events. The number of cells before and after migration was calculated for each phenotype, and the percentage of migration was calculated from these values. In addition, we checked that the culture of PBMCs in the presence of chemokines from 1 to 4 h did not modify the number or the proportion of blood DC subsets. We validated the results of DC TEM and chemotaxis obtained with total PBMCs using purified DC subsets, which in parallel experiments, showed a comparable pattern of response to chemokines.
Flow cytometry
Cells were preincubated with human IgG for 10 min at room temperature and then incubated with the appropriate mAb for 30 min at 4°C. To identify CD11c+ and CD123+ peripheral blood DCs (PBDCs), triple staining with labeled mAb was performed with antilineage (CD3, CD14, CD16, CD19, CD20, CD34, CD56) FITC, anti-DR PerCP, and anti-CD11c PE or anti-CD123 PE. Blood DCs were defined within PBMCs as lineage-negative, HLA-DR+ cells that were CD11c+ or CD123+. Four-color cytometry was used to analyze the expression of adhesion molecules and chemokine receptors. PBMCs were incubated with corresponding mAb (1 µg/ml), followed by washing and labeling with biotin-labeled goat anti-mouse Ig (Dako, Glostrup, Denmark), followed by allophycocyanin-labeled streptavidin (PharMingen). After a new washing, free goat anti-mouse Ig-binding sites were saturated by excess concentrations of mouse IgGs before exposing the cells to anti-CD3, -CD14, -CD16, -CD19, -CD20, -CD34, and -CD56 FITC, CD11c PE or anti-CD123 PE, and anti-DR PerCP-labeled mAb. Fluorescence flow cytometry acquisition and analysis were performed on a FACScalibur cytofluorometer (Becton Dickinson) using CellQuest software.
Statistic analysis
Data are presented as the mean ± SD and compared by Student’s t-test or one-way ANOVA. Multiple groups were compared by nonparametric ANOVA (Kruskal-Wallis test). If P < 0.05 were obtained, pairs of groups were further compared with nonparametric Mann-Whitney U-test.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
We also performed bare migration assays using total PBMCs to a wide panel of chemokines. As shown in Figure 2 , CD11c+ DCs migrated with great potency to CCL2, CCL8, CCL7, and CXCL12. The absolute number of CD11c+ DCs mobilized in response to these chemokines was very high, with CXCL12 attracting more than 80% of the initial population after 1 h of migration. In addition, the CD11c+ subset moderately responded to CCL13, CCL5, and CCL3 and migrated weakly or did not migrate in response to CCL4, CCL19, CCL20, CCL21, CCL22, CXCL8, CXCL9, and CXCL10 (Fig. 2A) . In contrast, the CD123+ subset poorly responded to most of the chemokines tested, in spite of a modest response to CXCL12 (Fig. 2B) . Dose-response curves to chemokines were also performed, and the chemotactic responses of both DC subsets were analyzed. CD11c+ DCs migrated to CCL2, CCL3, CCL5, CCL7, CCL8, and CCL13 in a dose-dependent manner, whereas CD123+ DCs weakly responded to higher doses of these chemokines (Fig. 2C 2D 2E ; data not shown). Doses as high as 500 ng/ml were assayed for CCL19, CCL20, CCL21, CCL22, CXCL8, CXCL9, CXCL10, and CXCL11 without a significant increase in blood DC chemotaxis (data not shown). The dose-response curves of both DC subsets toward CXCL12 were bell-shaped, reaching a maximum at 50 ng/ml for CD11c+ cells and at 100 ng/ml for CD123+ cells, respectively (Fig. 2F) . These results indicate that in bare migration assays, CD11c+ DCs are highly motile cells, which respond with great potency to inflammatory and homeostatic chemokines, whereas CD123+ DCs only respond with very low potency to CXCL12. Therefore, the chemotactic responses of blood DCs obtained with unfractionated PBMCs were consistent with previous studies performed with purified DCs [25 ], suggesting that additional factors to DC isolation procedures may account for the impaired ability of plasmacytoid DCs to migrate.
|
-activated (Fig. 3
). After 2 h of transmigration, a significant level of spontaneous migration of CD11c+ DCs was observed through resting and TNF--activated endothelium (percentage of migrated cells: 20.6±6.3%, n=5 donors, and 29±10.8%, n=9, respectively; Fig. 3A
shows a representative donor), whereas the baseline migration of the CD123+ subset was much lower (resting endothelium 2.3±1.6% and TNF--activated endothelium 2.9±2.6%, n=5 donors; Fig. 3A
). In the case of CD11c+ DCs, migration was higher in response to CXCL12 (90.3±7.6%, n=7) and CCL5 (78.5±13.5%, n=8; Fig. 3A
). Similarly, CD123+ DCs strongly migrated across endothelium in response to CXCL12 (52±16.2%, n=7) and exhibited a significant migratory response to CCL5 (30±8.7%, n=8; Fig. 3A
). Therefore, the presence of an endothelial monolayer greatly favored the migration of CD123+ DCs. For comparative purposes, the transmigration responses of purified blood DCs to the more active chemokines CCL2, CCL5, and CXCL12 were also analyzed (Fig. 3B)
. As expected, both DC subsets strongly responded to CXCL12. CD123+ DCs responded moderately to CCL5 and less to CCL2, whereas CD11c+ DCs equally responded to CCL2 and CCL5. Therefore, CD123+ and CD11c+ DCs significantly migrate across endothelium toward CXCL12 and CCL5, but they differ in their ability to spontaneously transmigrate and in their response to the inflammatory chemokine CCL2.
|
|
|
4 (CD49d) and L (CD11a) integrins, which may mediate adhesion to endothelium. It is interesting that no expression of DC-SIGN was detected in blood DC subsets.
Next, to address the specific contribution of each adhesion molecule to the process of DC transmigration, TEM assays were performed in the presence of blocking mAb to ICAM-1, PECAM-1, and ß1 and ß2 integrins. CXCL12 was used as a migration inducer in these experiments, as it promotes maximal transmigration in both DC subsets. TEM, across resting endothelium of both DC subsets, was inhibited by 60–70% with anti-ß2 integrin mAb; however, no inhibition was observed with anti-ICAM-1 mAb (Fig. 5A
). Because in the same experiments, the anti-ICAM-1 mAb effectively inhibited the migration of monocytes and lymphocytes (Fig. 5A)
, it is possible that other ß2-integrin endothelial ligands such as ICAM-2 could play a role in the transmigration ability of blood DCs through resting endothelium. Anti-ß1-integrin mAb also inhibited migration of DCs across resting endothelium. This inhibition may correspond to blocking of leukocyte interactions with underlying extracellular matrix (ECM) secreted by HUVEC, as the endothelial preparations used in these studies did not express VCAM-1, as assessed by immunofluorescence flow cytometry, and anti-VCAM-1 mAb did not inhibit transmigration across resting endothelium (data not shown). In addition, to assess the role of endothelial cytokine-inducible molecules, TEM experiments were performed across TNF--stimulated endothelium (Fig. 5B)
. Under these conditions, transmigration of CD11c+ DC was partially inhibited by anti-PECAM-1 mAb and was weakly or not affected by blocking anti-ICAM-1, anti-ß1, or anti-ß2 mAb (Fig. 5B)
. On the contrary, CD123+ DC transmigration could be blocked by anti-ß2, anti-ß1, anti-ICAM-1, or anti-PECAM-1 mAb (Fig. 5B)
. We also used anti-ß1 and -ß2 mAb in combination, which was more effective than blocking each integrin alone (Fig. 5C)
. However, a large portion of CXCL12-induced CD11c+ DC migration across activated endothelium was independent of ß1 and ß2 integrins. Taken together, these results may indicate that PECAM-1 may support transmigration of both DC subsets through stimulated endothelium. In addition, functional ß1 and ß2 integrins support the adhesion/transmigration of CD123+ DCs across resting and stimulated endothelium and of CD11c+ DCs across resting endothelium. By contrast, under endothelial activation, CD11c+ DCs appear to be highly motile cells whose transmigration was weakly inhibited by antibodies against ß1 and ß2 integrins, compared with other leukocytes.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We performed migration assays with whole peripheral blood leukocytes, taking advantage of the combination of migration assays with cell counting of minor leukocyte populations by quantitative flow cytometry. Our results were highly reproducible, despite blood DCs representing a minor subpopulation of leukocytes, and were consistently observed among different blood donors. This methodology allowed us to study the pattern of migration of unmanipulated blood DCs in response to a number of chemokines and their ability to interact with endothelium. We show that the response to chemokines of unmanipulated blood DCs is similar to purified blood DCs and thus is compatible with bare migration assays performed with enriched DCs [25 ]. However, bare migration assays are limited and have little impact, as cell migration requires the interaction of adhesion receptors with ligands provided by ECM or intercellular adhesion molecules expressed by supporting cells. To migrate, some leukocytes such as lymphocytes are highly dependent on the interaction with intercellular adhesion molecules (such as ICAM-1 and -2, expressed by resting endothelium, or VCAM-1, expressed by activated endothelium) [26 ]. Therefore, we performed migration assays in the presence of EC monolayers, which provide adhesion molecules as well as synthesize ECM proteins, and together, support cell adhesion and transmigration. Previous work, performed in the absence of endothelial support, reported that plasmacytoid cells are drastically impaired to migrate to most chemokines [25 ]. It is interesting that plasmacytoid DCs are highly dependent on endothelial support to transmigrate. When well positioned to migrate across endothelium, this subset responds importantly to CXCL12 and CCL5, chemokines that may drive the accumulation of plasmacytoid DCs under pathological settings.
At least some blood DCs are hypothesized to be en route from bone marrow to tissues, where they immigrate to become tissue resident-DCs [1 ]. Our data indicate that CXCL12 may play a major role in the steady-state recruitment of blood DCs directly from blood to tissues. CXCL12 mRNA has been reported to be expressed in many different tissues [27 ], although it is preferentially expressed in lymph nodes, lung, liver, and bone marrow and less in small intestine, kidney, skin, brain, and skeletal muscle [28 ]. In addition, CXCL12 may also play a role in the recruitment of DCs to inflamed tissues, as CXCL12 is strongly up-regulated in the joint synovium of rheumatoid arthritis and may also contribute to skin inflammatory responses [29 30 31 ]. The bell-shaped, dose-response curves to CXCL12 suggest that myeloid DCs respond maximally to lower doses of CXCL12 than plasmacytoid DCs. Thus, it is possible that myeloid DCs migrate preferentially to tissues expressing low levels of CXCL12, whereas plasmacytoid DCs may migrate to tissues that express higher doses of CXCL12, such as lymph nodes. This is compatible with the reported presence of plasmacytoid DCs, but not myeloid DCs, in tumors expressing high levels of CXCL12 [32 ]. Previous work, based on the expression of CXCR3 and L-selectin (CD62L) by plasmacytoid DCs [19 ], had postulated that these cells migrated into the lymph node through high endothelial venules (HEV) in response to CXCL9 or CXCL10. In accordance, it has been recently described that plasmacytoid DCs are reduced in frequency in L-selectin-deficient mice [33 ]. However, our data indicate that plasmacytoid DCs transmigrate specifically to CXCL12 and CCL5, with rather weak chemotaxis toward CXCL9, CXCL10, and CXCL11, even through activated endothelium, which support CD62L-mediated leukocyte adhesion [34 ]. In this regard, it has been reported that CXCL12 is displayed broadly on HEV, and CXCL12 ectopic expression induces small infiltrates enriched in DCs [35 , 36 ]. In addition to blood and organized lymphoid tissue, plasmacytoid DCs have also been identified in nasal mucosa during allergic reactions [37 ] and in cutaneous lesions of lupus erithematosus [38 ]. We have also found that plasmacytoid DCs migrated in response to CCL5, whose expression has been associated with a wide range of inflammatory disorders and pathologic conditions [39 ]. Emigration of plasmacytoid DCs to tissues may be of pathogenic relevance in those inflammatory diseases associated with up-regulated CCL5 and/or CXCL12 expression.
Conversely, inducible chemokines, such as CCL5 and CCL2, may play an important role in the recruitment of myeloid DCs into inflamed tissues. In this regard, it has been reported that the treatment of rats with the selective antagonist of CCR1 and CCR5, Met-RANTES, importantly reduces the recruitment of DC precursors into airway epithelium [40 ]. The strong, migratory response of myeloid DCs to CCL2 and CCL5 might allow their rapid accumulation within inflamed tissues and explain why DCs are the earliest detectable leukocytes arriving at mucosal surfaces upon pathogen invasion [41 , 42 ]. It is interesting that the responsiveness of myeloid DCs to CCL2 is a major difference with in vitro-derived DCs, as the latter do not migrate in response to CCL2 (data not shown; refs. [4 , 5 ]).
Besides a role as precursors of tissue resident-DCs, at least some blood DCs may act as environmental sentinels circulating in blood to detect pathogen invasion. Therefore, blood DCs may capture pathogenic antigens that can access the bloodstream and afterwards, mature and emigrate directly from blood to lymph nodes. In this regard, we have shown the ability of myeloid DCs to respond to the lymph node chemokine CCL21 upon acute viral dsRNA challenge, which points out their ability to detect the presence of pathogens in the bloodstream. The positive effect of poly(I:C) in the response of myeloid DCs to CCR7 ligands is in agreement with the very high level of expression of its receptor Toll-like receptor (TLR)3 by CD11c+ DCs [43
]. Thus, poly(I:C) selectively stimulated CD11c+ DCs to produce IFN- and IL-12p75, whereas plasmacytoid DCs do not express TLR3 and do not respond to poly(I:C). Plasmacytoid DCs express high levels of TLR9 and CCR7; however, they were not responsive to CCL21, neither resting nor after short activation with the TLR9 ligand ODN containing unmethylated CpG motifs. CpG-DNA and other stimuli reported to induce maturation of this subset, such as CD40L plus IL-3, partially rendered plasmacytoid DC responsive to CCL21 after prolonged stimulation (24 h).
In addition to chemokines that guide the emigration of responsive subsets of leukocytes to target tissues, extravasation is supported by a multistep cascade of adhesive interactions with endothelium [26
, 44
]. Both subsets of blood DCs uniformly express molecules involved in the initial adhesion under flow, such as P-selectin glycoprotein ligand-l and its isoform cutaneous lymphocyte-associated antigen [45
]. We show that blood DCs also uniformly express 4 (CD49d) integrins, which like selectins, may support rolling [46
]. By contrast, no blood DC subset expresses the DC-specific adhesion receptor DC-SIGN, a C-type lectin that has been reported to mediate the transmigration of mono-DCs [47
]. To investigate adhesion molecules involved in TEM of both DC subsets, we induced transmigration with CXCL12 to achieve a maximal DC transmigration. An interesting observation is the fact that under CXCL12-induced TEM, blockade of ß1 integrins partly inhibited migration of most leukocytes, across resting and activated endothelium. These results agree with a previous work reporting that CXCL12 stimulates lymphocytes to use ß1 integrins in addition to ß2 integrins across activated endothelium [48
]. Chemokine-induced ß1-integrin activation [26
] can mediate adhesion to ECM proteins produced by HUVEC, and thus, in addition to very late antigen (VLA)-4, other ß1 integrins such as VLA-5 may mediate chemokine-induced TEM [49
]. Similar to previous monocyte TEM studies [50
], we did not observe differences in the number of DCs that migrated across resting or TNF--stimulated endothelium. A previous study by Brown et al. [51
] found that endothelial activation increased the attachment of blood DCs, and it was dependent on ß1 and ß2 integrins. It is possible that enhanced DC adhesion to activated endothelium may counteract the up-regulation of endothelial adhesion molecules. However, we did observe some differences in the role of adhesion receptors between resting and stimulated endothelium. The contribution of PECAM-1 to DC transmigration appears to be more important across activated endothelium. Transmigration of plasmacytoid DCs is supported by ß1 and ß2 integrins across resting and activated endothelium, but migration of myeloid DCs is only dependent on ß1 and ß2 integrins across resting endothelium. It is interesting that transmigration across activated endothelium of the highly migratory myeloid DCs was not inhibited by anti-ß1- and -ß2-integrin mAb and only partially by a combination of anti-ß1- and -ß2-integrin mAb, suggesting that an additional, unrecognized pathway is also involved in TEM of myeloid DCs to CXCL12.
In conclusion, this study points out the existence of important differences in the migratory ability between distinct subsets of blood DCs. Adding on the understanding of the multiple factors governing the traffic of DCs may offer us the opportunity to design novel, therapeutic strategies based on targeting the appropriate DC subset to selective tissues such as tumors.
|
ACKNOWLEDGEMENTS |
|---|
Received October 30, 2002; revised January 9, 2003; accepted January 22, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. de la Rosa, D. Yang, P. Tewary, A. Varadhachary, and J. J. Oppenheim Lactoferrin Acts as an Alarmin to Promote the Recruitment and Activation of APCs and Antigen-Specific Immune Responses J. Immunol., May 15, 2008; 180(10): 6868 - 6876. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Lehner, A. Stilper, P. Morhart, and W. Holter Plasticity of dendritic cell function in response to prostaglandin E2 (PGE2) and interferon-{gamma} (IFN-{gamma}) J. Leukoc. Biol., April 1, 2008; 83(4): 883 - 893. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Marchal-Somme, Y. Uzunhan, S. Marchand-Adam, M. Kambouchner, D. Valeyre, B. Crestani, and P. Soler Dendritic Cells Accumulate in Human Fibrotic Interstitial Lung Disease Am. J. Respir. Crit. Care Med., November 15, 2007; 176(10): 1007 - 1014. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Samaniego, L. Sanchez-Martin, A. Estecha, and P. Sanchez-Mateos Rho/ROCK and myosin II control the polarized distribution of endocytic clathrin structures at the uropod of moving T lymphocytes J. Cell Sci., October 15, 2007; 120(20): 3534 - 3543. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Sancho, J. M. Vieira, C. Casalou, M. Mesquita, T. Pereira, B. M. Cavaco, S. Dias, and V. Leite Expression and function of the chemokine receptor CCR7 in thyroid carcinomas. J. Endocrinol., October 1, 2006; 191(1): 229 - 238. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Sanchez-Sanchez, L. Riol-Blanco, and J. L. Rodriguez-Fernandez The Multiple Personalities of the Chemokine Receptor CCR7 in Dendritic Cells J. Immunol., May 1, 2006; 176(9): 5153 - 5159. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. A. Zabel, S. J. Allen, P. Kulig, J. A. Allen, J. Cichy, T. M. Handel, and E. C. Butcher Chemerin Activation by Serine Proteases of the Coagulation, Fibrinolytic, and Inflammatory Cascades J. Biol. Chem., October 14, 2005; 280(41): 34661 - 34666. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Li, F. Takeuchi, J.-a. Wang, C. Fuller, G. Pacheco-Rodriguez, J. Moss, and T. N. Darling MCP-1 overexpressed in tuberous sclerosis lesions acts as a paracrine factor for tumor development J. Exp. Med., September 6, 2005; 202( |
