Published online before print February 5, 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Institute of Molecular Biology and Bioinformatics, Charité CBF, Berlin, Germany;
Deutsches Rheuma-Forschungszentrum, Berlin, Germany; and
Institute of Microbiology, Technische Universität, München, Germany
1Correspondence: Institute of Molecular Biology and Bioinformatics, Charité, Arnimallee 22, 14195 Berlin, Germany. E-mail: matthias.peiser{at}charite.de
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
by stimulation via TLR2 than by stimulation via TLR3, -4, -5, -8, and -9. In contrast to MoLCs, dendritic cells, generated from the same donor monocytes, were activated by agonists of TLRs other than TLR2 as well. Lipopeptides triggering TLR2 induced IL-1R-associated kinase-1 phosphorylation and migration toward the chemokines CCL19 and CCL21 in epidermal LCs and CD1c+MoLCs. Up-regulation of CD86, CD83, and CCR7, TNF- and IL-6, and NF-
B activation and proliferation of CD4+T cells could be inhibited TLR2-specific blockage using antibodies prior to TLR2 activation. Application of anti-TLR1, anti-TLR6, and anti-TLR2 indicated an exclusive role of TLR2 in IL-6 induction in human LCs. Collectively, our results show that TLR2 expressed by LCs mediates inflammatory responses to lipopeptides, which implicates a central role in sensing pathogens in human skin.
Key Words: TLR2 • dendritic cell • skin • peptidoglycan • lipopeptide • chemotaxis
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In the periphery, DCs act as sentinels for detecting pathogenic viruses, bacteria, yeasts, and parasitic protozoae via pathogen recognition receptors. Aside from complement receptors and cytosolic proteins of the nucleotide-binding oligomerization domain (NOD)-like receptor family, Nod1 and Nod2, TLRs mediate sensing of foreign and dangerous molecules derived from pathogens [1 2 3 4 ]. These receptors are expressed constitutively or upon activation of various cell types, but DCs are provided with the most complete arsenal of TLRs [5 ]. Once activated by stimuli such as the TLR4 ligand LPS, DCs secrete proinflammatory cytokines and up-regulate costimulatory and maturation-associated molecules such as CD40, CD80, CD86, and CD83. Activated DCs follow a chemokine track by detection via CCR7 to regional lymph nodes [6 ]. Whereas expression and stimulation of TLRs in mouse and human DCs are investigated extensively [5 ], only sparse data exist about LCs. Here, we present data about primary LCs from human skin and demonstrate that exogenous, bacteria-derived TLR agonists elevate maturation, migration, and the capacity to induce T cell proliferation.
As in initial studies, secretion of IL-12p70 from epidermal LCs upon stimulation with peptidoglycan (PGN; ref. [7
]) was analyzed, we focused on activation of TLR2 by bacterial products and analogs. TLR2 (CD282) belongs to the family of 13 mammalian TLRs. The extracellular domain of the receptor binds pathogen-associated molecular patterns (PAMPs) through leucine-rich repeats [3
]. TLR2 signals through the MyD88-dependent Toll-mediated pathway including IL-1R-associated kinase (IRAK) phosphorylation, which culminates in activation of transcription factors, such as NF-B, which drives production of inflammatory cytokines such as IL-6, TNF-, and IL-18 [3
, 8
]. TLR2 was found to recognize a broad panel of different pathogens. TLR2 mediates detection of PGN and lipoteichoic acid of Gram-positive bacteria [9
]. For sensing of bacterial triacyl lipopeptides and the synthetic analog N-pamitoyl-S-[2,3-bis(palmitoyloxy)-(2RS)-propyl]-(R)-Cys-Ser-(Lys)4 (Pam3CSK4), TLR2 cooperates with the structurally related TLR1 [10
11
12
]. In contrast, phospholipomannan and zymosan from yeast, diacyl lipopeptides macrophage-activating lipopeptide-2 (MALP2) from Mycoplasma fermentans, and S-[2,3-bis(palmitoyloxy)-(2RS)-propyl]-(R))-Cys-Ser-(Lys)4 (Pam2CSK4) are recognized by dimers composed of TLR2/6 [3
, 13
]. To investigate if human LCs can detect TLR2 agonists, we stimulated epidermal LCs and monocyte-derived LCs (MoLCs) with PGN, Pam3CSK4, and Pam2CSK4. Activation of LCs by lipopeptides resulted in chemokine-induced cell migration, up-regulation of CD86 and CD40 in MoLCs, additionally in TNF-, and IL-6 secretion and activation of NF-B. TLR2 antibodies blocked LC-induced proliferation of allogeneic CD4+T cells. Co-blocking of TLR2/1 and TLR2/6 heterodimers did not result in further down-regulation of IL-6 secretion as compared with blocking of TLR2 only.
In conclusion, synthetic or bacterial lipopeptides and PGN activate human LCs TLR2-dependently, according to our data.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Isolation of CD1c+MoDCs/MoLCs
Human PBMCs were enriched from buffy coat preparations by Ficoll gradient, and untouched monocytes were isolated by depletion of magnetically labeled nonmonocytes (monocyte isolation kit II, Miltenyi Biotec). MoDCs were generated in RPMI 1640, supplemented with 2 mM L-glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin, and 10% FCS (all from Biochrom) from monocytes by 6-day culture with GM-CSF (100 ng/ml) and IL-4 (10 ng/ml) with complete medium/cytokine exchange at Days 2 and 4. For generation of MoLCs, IL-4 was only added at Day 0, and TGF-β1 (10 ng/ml, all from R&D Systems, Wiesbaden-Nordenstadt, Germany) was added at Days 0, 2, and 4, as described [15
]. In parallel, MoDCs and MoLCs were sorted by positive selection with CD1c, as described above for epidermal cells.
Cell stimulation by TLR agonists
The isolated LCs, MoDCs, and MoLCs at Day 6 of cell culture were washed two times and cultured at 106 cells per ml for 2 days without further cytokine addition in 24-well tissue-culture plates (Greiner, Frickenhauser, Germany). Stimulation of TLR1–9 was performed by challenge with LPS from Escherichia coli serotype R515 (Apotech, Epalinges, Switzerland); PGN from Staphylococcus aureus (Fluka, Buchs, Switzerland); Pam3CSK4, polyinosinic:polycytidylic acid [poly(I:C)], flagellin, MALP2, poly(U), and CpG oligodeoxynucleotide (ODN) 2395 (all from Apotech); or Pam2CSK4 (Invivogen, San Diego, CA, USA). Concentrations and stimulation conditions were optimized for MoLCs and were as indicated in the figure legends. All stimuli used, other than LPS, contained endotoxin below 0.1 EU/µg compound, as determined by using the Limulus amebocyte lysate assay (BioWhittaker, Walkersville, MD, USA).
Flow cytometry
Single-cell suspensions were stained by following anti-human mAb: anti-CD40 (5C3), anti-CD80 (L307.4), anti-HLA-DR (G46-6), directly conjugated to FITC, anti-CD54 (HA58), anti-CD83 (HB15e), anti-CD86 (FUN-1), anti-B7H2 (2D3), directly conjugated to PE, and anti-CD1a (HI149), directly conjugated to allophycocyanin. Anti-CCR7 (2H4) was indirectly stained by biotinylated anti-IgM (R6-60.2) and streptavidin-FITC conjugate (BD Biosciences, Heidelberg, Germany), and Langerin-PE (CD207, DCGM4) was from Coulter (Krefeld, Germany). The TLR phenotype of MoLCs and LCs was analyzed by using anti-TLR1 (GD2.F4), anti-TLR2 (TL2.1), anti-TLR4 (HTA125), and anti-TLR9 (eB72-1665, eBioscience, San Diego, CA, USA), all PE-coupled. Anti-TLR5 (85B152.5), anti-TLR6 (86B1153, Imgenex, San Diego, CA, USA), and purified IgG1 isotype control (MOPC-21, BD Biosciences) were detected by PE-labeled, rat anti-mouse IgG1 (A85-1, BD Biosciences) as recommended. Cells were analyzed on a FACSCalibur using 7-amino-actinomycin (BD Biosciences) to exclude dead cells. For measurement of intracellular TLR9, a fixation and permeabilization procedure was performed (Cytofix/Cytoperm, BD Biosciences).
Cytokine detection
Secretion of IL-6 and TNF- was determined after 48 h in the supernatant of MoDCs and MoLCs by sandwich ELISA (DuoSet, R&D Systems).
Western blot analysis
Proteins of CD1c+MoLCs were extracted by 5 min incubation in complete Lysis-M buffer (Roche) containing a protease inhibitor. Equal amounts of protein were separated by SDS-PAGE and transferred onto a PolyScreen polyvinylidene difluoride hybridization transfer membrane (Perkin Elmer, Wellesley, MA, USA). Western blot was performed detecting human TLR2, -1, and -6, using 0.2 µg/ml goat IgG anti-TLR2 antibody (AF2616, R&D Systems), 1 µg/ml rabbit IgG anti-TLR1 (sc-30000), and 1 µg/ml rabbit IgG anti-TLR6 (sc-30001, Santa Cruz Biotechnology, Santa Cruz, CA, USA). Blots were incubated with the following HRP-linked secondary antibodies: anti-goat IgG (0.2 µg/0.5 ml, sc-2033, Santa Cruz Biotechnology) and anti-rabbit IgG (1:2000, Cell Signaling Technology, Beverly, MA, USA). In indicated experiments CD1c+MoLCs were stimulated for 2 h as described, treated with cell lysis buffer (Cell Signaling Technology) on ice, and sonicated. Blot membranes were incubated with rabbit anti-human IRAK1, antiphospho-IRAK1 (Ser376), antiphospho-IRAK1 (Thr387, all 1:1000), and secondary HRP-conjugated anti-rabbit IgG (1:2000). Analyses were performed using the Phototope-HRP Western blot detection system (all Cell Signaling Technology) with Lumiglo and Peroxide reagent as recommended.
Detection of NF-B p65
Stimulated cells were sonicated on ice in cell lysis buffer (Cell Signaling Technology) containing 1 mM PMSF. Soluble proteins of the supernatant were analyzed by sandwich ELISA detecting phosphorylization of the NF-B subunit p65 at serine 536 (PathScan, Cell Signaling Technology), according to the manufacturer’s recommendations. This assay includes application of capture antibody mouse anti-p65 (8F8) and detection antibody rabbit antiphospho-p65 (93H1), recognized by HRP-linked anti-rabbit IgG.
Migration assays
MoLCs and LCs were incubated at Day 6 with TLR2 agonists for 3 h. Cells were washed two times with PBS and cultured for 48 h without further stimulation as described for migratory DCs [16
]. For fluorescence-based migration assay, the cells were stained with 5 µmol pH-insensitive tracer 5- (and 6-)carboxy-2',7'-dichlorofluorescein-diacetate (CDFDA; Molecular Probes, Eugene, OR, USA) for 10 min at 37°C and washed two times. Migration was assessed in 24-well plates with inserts containing 8 µm holes in light-tight PET membranes (490–700 nm, HTS FluoroBlok, BD Biosciences). In short, 105 cells were plated as triplicates in inserts in 50 µl RPMI/10% FCS, and chemokines CCL5, CCL19, and CCL21 (R&D Systems) were added in 350 µl medium to the bottom chamber. Plates were incubated for 1.5 h, and emission was quantified at 485/520 nm in the bottom-reading Fluostar Optima (BMG Labtech, Durham, NC, USA). Graded numbers of CDFDA-stained MoLCs were used in 96-well plates with the same well diameter as the Fluoroblok inserts to generate standard curves and calculation of numbers of migrated cells.
Blocking experiments
The following mAb were used in blocking experiments: anti-TLR1 (GD2.F4, eBioscience), anti-TLR2 (2229, Cell Signaling Technology, and T2.5, eBioscience), and anti-TLR6 (86B1153, Imgenex).
MLR
Untouched CD4+T cells were purified from PBMCs by depletion of magnetically labeled, non-CD4+T cells using CD4+T cell isolation kit II (Miltenyi Biotec), according to the manufacturer’s instructions. As proven by FACS, more than 94% were CD4-positive, and these naive CD4+T cells were stained for 10 min at 37°C with 0.5 µM CFSE (Molecular Probes). Then, CD4+T cells were washed two times with RPMI/10% FCS and plated at 5 x 104 cells/well. Stimulated, allogeneic LC or MoLCs, partly preincubated with blocking antibody T2.5, as described above, were recovered at Day 8 and added at a ratio of 1:10 as stimulators for CD4+T cells. As positive controls, proliferation was induced in CD4+T cells by 104 anti-CD3/anti-CD28-coated Dynabeads (Dynal Biotech, Oslo, Norway) per well. After 5 days of coculture, cells were stained by anti-CD4-Cy5 (TT1, a gift of Dr. Andreas Thiel, DRFZ, Berlin), and proliferation of CD4+T cells was measured by FACS.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (26K): [in a new window] |
Figure 1. Primary epidermal LCs, CD1c+MoLCs, and CD1c+MoDCs express TLR2. (A) LCs were prepared from human skin, and other DC types were generated from human blood by depletion of non-CD14 cells, culture in the presence of GM-CSF and IL-4 (MoDCs) or GM-CSF, IL-4, and TGF-β1 (MoLCs). All DC types were additionally enriched by CD1c sorting. Cells were stained at Day 6 of culture (CD1c+MoDCs, CD1c+MoLCs) or at day of preparation (LCs) with mAb against extracellular TLR1, -2, -4, -5, and -6 and intracellular TLR9. Histograms show TLR expression (white) of LCs, CD1c+MoDCs, and CD1c+MoLCs from one donor; each panel represents four donors with similar results; isotype, gray. (B) MoLCs at Day 6 were lysed, and for immunoblot analysis, the membranes were incubated with anti-TLR1, anti-TLR2, or anti-TLR6 antibody. One representative experiment of three is shown.
|
 View larger version (38K): [in a new window] |
Figure 2. TLR agonists induce phenotypical activation of Langerin-expressing CD1c+MoLCs. (A) CD1c+MoLCs were isolated by CD1c sorting and analyzed by FACS for expression of Langerin and CD1a. (B) CD1c+MoLCs were challenged by the following TLR agonists: 100 ng/ml Pam3CSK4 for TLR1/2 (Pam3), 100 µg/ml poly(I:C) for TLR3 (I:C), 1 µg/ml LPS for TLR4, 100 ng/ml flagellin (Flag) for TLR5, 100 ng/ml MALP2 for TLR2/6, 10 µg/ml poly(U) for TLR8 (PU), 10 µg/ml CpG ODN 2395 for TLR9, and without stimulus (Wo). Coexpression of activation-associated molecules HLA-DR and CD54, CD40, and CD86; CCR7 and CD83; and CD80 and B7H2 on viable cells was analyzed after 2 days by FACS. Quadrants indicate isotype controls. One representative experiment of five with different donors is shown.
|
and activate MoLCs. Neither poly(I:C) and poly(U), triggering virus-detecting TLRs, nor flagellin or CpG ODN induced significant up-regulation of IL-6 and TNF- protein secretion. Focusing on activation of TLR2, we analyzed the capacity of lipopeptides to stimulate CD1c+MoLCs. In the TLR2-, -1-, and -6-dependent signaling pathway, IRAK4 induces the phosphorylation of IRAK1 [17
]. Serine and threonine phosphorylation of IRAK1 but no degradation of IRAK1 could be detected in CD1c+MoLCs after 2 h of stimulation with Pam3CSK4 (Fig. 3C)
. In contrast, MALP2 failed to induce IRAK1 phosphorylation.
 View larger version (21K): [in a new window] |
Figure 3. TLR2 agonists induce IL-6, TNF-, and IRAK1 phosphorylation in CD1c+MoLCs. MoLCs and MoDCs were enriched by CD1c sorting and stimulated by TLR agonists at concentrations as described above. After 48 h, IL-6 and TNF- in the supernatants of MoLCs (A) and MoDCs (B) were measured by ELISA. Columns represent results of experiments with four different donors (±SD). Monocytes of single donors were split for generation of MoLCs and MoDCs. (C) CD1c+MoLCs were generated and isolated as described above and stimulated by 0.5 µg/ml LPS, Pam3CSK4, or MALP2 for 2 h. After cell lysis, Western blot analyses were performed with anti-IRAK1, antiphospho (P)-IRAK1 (Ser376), and antiphospho-IRAK1 (Thr387). Shown is the result of one experiment that represents three experiments with cells of different donors.
|
and IL-6 protein levels were enhanced along with increasing concentrations of PGN (Fig. 4 A
). Moreover, the lipopeptides Pam3CSK4, Pam2CSK4, and MALP2 induced TNF- and IL-6 in a concentration-dependent manner. Higher ligand concentrations did not result in an increase of the amounts of cytokines released (not shown). In CD1c+MoLCs, not solely LPS but the combination with costimulatory CD40 ligand (CD40L) is a prerequisite for maturation and secretion of IL-12p70 [7
]. Several TLR agonists, alone or in combination, are reported to exhibit a strong potency to up-regulate CD86 and CD40 in DCs in a dose-dependent manner [18
19
20
]. With regard to the up-regulation of maturation-associated molecules (Fig. 2)
, we investigated the concentration dependency of TLR2 agonists on up-regulation of CD86 and CD40. Triggering TLR2 on real LCs and on CD1c+MoLCs by PGN was strictly dose-dependent, resulting in strong CD86 and CD40 up-regulation (Fig. 4B)
. Pam3CSK4 and Pam2CSK4 enhanced CD86 and CD40 expression slightly on epidermal LCs, and high- versus low-dose stimulation maintained cell viabiltity (19% and 71%, respectively; not shown). CD86 and CD40 were strongly up-regulated on CD1c+MoLCs at different concentration ranges, as it was observed for secretion of IL-6 and TNF- (Fig. 4A)
. Still PGN, Pam3CSK4, and Pam2CSK4 at higher concentrations failed to amplify CD86 and CD40 on LCs and CD1c+MoLCs (not shown).
 View larger version (45K): [in a new window] |
Figure 4. TLR2 ligands dose-dependently induce TNF-/IL-6 and CD40/CD86 in LCs. (A) CD1c+MoLCs of one donor were stimulated at increasing amounts of TLR2 ligands. After 48 h, supernatants of 2.5 x 105 cells/ml were analyzed for TNF- and IL-6 by ELISA; stimuli indicated in µg/ml, error bars are ± SE. Shown is one experiment that represents four with different donors. (B) Real LCs were isolated from human epidermis, MoLCs were generated, and both populations were additionally enriched by CD1c MACS. LCs and MoLCs were stimulated for 48 h by PGN, Pam3CSK4, and Pam2CSK4. Cells were stained with anti-CD40 and anti-CD86 and analyzed by flow cytometry. Shown is the dynamic range of TLR2 agonist concentrations in µg/ml, as indicated left of the dot plots. The LC and MoLC plots represent three different donors with similar results.
|
 View larger version (11K): [in a new window] |
Figure 5. Pam2CSK4 and Pam3CSK4 induce LC migration. Epidermal LCs (A) and CD1c+MoLCs (B) were activated transiently for 3 h by 1 µg/ml Pam3CSK4 or 0.1 µg/ml Pam2CSK4 lipopeptides. After washing and an additional 48 h of cell culture in the absence of further stimuli, 105 cells were added in duplicates to fluorescence-dense filters with 8 µm pores. After 1.5 h, cell migration in response to 100 ng/ml CCL19 and CCL21, supplemented to the medium in the lower chambers, was detected by a bottom-reading fluorescence plate reader. Cell numbers, corresponding to the fluorescence intensity detected in the media of the lower chambers, are shown as mean values ± SD of four independent experiments.
|
, and IL-6 and IL-6 in CD1c+MoLCs was studied (Fig. 6B)
. In PGN-stimulated cells, both of the antibodies decreased the secretion of the cytokines. In a dose-dependent way, anti-TLR2 (T2.5) down-regulated secretion of TNF- and IL-6 in CD1c+MoLCs stimulated by Pam2CSK4 and Pam3CSK4. The polyclonal antibody (2229) did not block Pam3CSK4-induced TNF- and IL-6.
 View larger version (31K): [in a new window] |
Figure 6. Inhibition of CD86/CD40/CD83/CCR7 and of TNF-/IL-6 by TLR2 blockage. (A) Epidermal LCs and CD1c+MoLCs were preincubated for 30 min with anti-TLR2 (T2.5) at concentrations indicated. Isotype controls (Iso) were performed with IgG1 (MOPC-21) at the same conditions. After washing, the cells were stimulated with 20 µg/ml PGN for 48 h. Histograms show expression of CD86, CD40, CD83, and CCR7, as detected by flow cytometry. Gray line indicates the peak of basic expression level for surface molecules in unstimulated, unblocked cells. Shown is one experiment that represents the results of three experiments with different donors. (B) CD1c+MoLCs were treated with anti-TLR2 (mAb T2.5, gray bars; and polyclonal antibody 2229, white bars) at increasing concentrations. Cells were stimulated with 20 µg/ml PGN, 1 µg/ml Pam3CSK4, or 0.1 µg/ml Pam2CSK4 for 48 h. Isotype controls MOPC-21 and rabbit IgG did not block cytokine secretion in any case (not shown). (C) TLR2 coreceptors were blocked by anti-TLR1 and anti-TLR6 in different combinations. CD1c+MoLCs were incubated with 1 µg/ml anti-TLR2 (T2.5), 10 µg/ml anti-TLR1 (GD2.F4), and 10 µg/ml anti-TLR6 (86B1153) antibodies for 30 min. Afterwards, cells were stimulated by media as control, 20 µg/ml TLR2 agonist PGN, 1 µg/ml TLR1/2 agonist Pam3CSK4, or 0.1 µg/ml TLR2/6 agonist Pam2CSK4 for 48 h. Columns show the mean of duplicates ± SD from one donor, as detected by ELISA, and were repeated four (B) and three (C) times with similar results.
|
Ligation of TLR2 by lipopeptides activates NF-B and induces CD4+T cell proliferation
Triggering TLRs with various ligands resulted in the release of transcription factor NF-B from its inhibitor, its translocation to the nucleus, and expression of proinflammatory cytokines [17
]. As inhibition of the phosphorylation of the subunit p65 (RelA) in TLR2-blocked cells indicates effective blockage of cell activation by Pam3CSK4 and Pam2CSK4, we investigated cell lysates by a p65 ELISA. Increased serine phosphorylation of p65 was measured in CD1c+MoLCs after 30 min of stimulation with Pam3CSK4, Pam2CSK4, or PGN (Fig. 7 A
). In TNF--stimulated cells, phosphorylation was enhanced as compared with unstimulated cells. Treatment with TLR2-blocking antibody (T2.5) prevented p65 phosphorylation in TLR2-stimulated but not in TNF--stimulated cells. As TLR2-triggered CD1c+MoLCs enhanced HLA-DR and costimulatory molecules (Fig. 2)
, functional assays determining the capacity to induce T cell proliferation were performed. CD1c+MoLCs stimulated by Pam2CSK4 and Pam3CSK4 strongly stimulated proliferation of allogeneic CD4+T cells (Fig. 7B)
. In both cases, preincubation with anti-TLR2 (T2.5) abrogated T cell proliferation. TLR2-triggered epidermal LCs mediated limited proliferation to T cells, but this capacity could also be blocked by anti-TLR2. Treatment with anti-TLR2 did not affect proliferation of CD4+T cells stimulated with anti-CD3/CD28-coated beads.
 View larger version (28K): [in a new window] |
Figure 7. Ligation of TLR2 by lipopeptides results in phosphorylation of NF-B and enhanced MLR. (A) Isolated CD1c+MoLCs were incubated for 30 min with 10 µg/ml isotype MOPC-21 or anti-TLR2 (T2.5) and then stimulated by 1 µg/ml Pam3CSK4, 0.1 µg/ml Pam2CSK4, 20 µg/ml PGN, or 20 ng/ml TNF-. After 30 min, cells were lysed and analyzed for phosphorylated p65, a subunit of transcription factor NF-B, by ELISA. Means of duplicates ± SD of one representative experiment of three are shown. (B) Epidermal LCs and CD1c+MoLCs were incubated for 30 min with 10 µg/ml isotype MOPC-21 or anti-TLR2 (T2.5), and unbound antibodies were removed by washing. Cells were left untreated or stimulated by 1 µg/ml Pam3CSK4 or 0.1 µg/ml Pam2CSK4 for 48 h. Naive CD4+T cells were isolated by depletion from PBMCs and stained by CFSE. Allogeneic CD4+ T cells were cocultured with untreated CD1c+MoLCs/LCs as APCs (Wo) or with TLR2-activated APCs. For control, CD4+T cells were stimulated by anti-CD3/anti-CD28 beads without MoLCs. FACS plots show CFSE intensity of CD4+cells after 5 days. Results are representative of four independent experiments.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, and cell migration. Blocking of agonist binding to TLR2 inhibited NF-B activation and CD4+T cell proliferation. Finally, no cooperation of TLR2 with TLR1 and TLR6 in respect of IL-6 secretion was observed in co-blocked TLRs on MoLCs. In contrast to numerous investigations about expression and stimulation of TLRs on DCs [5 , 21 ], a lack of data about LCs is evident. There are only few reports about human LCs or LC-like cells [22 23 24 25 ]. By staining with reliable mAb, we, for the first time, analyzed more comprehensively TLR expression of LCs and showed the functional consequences of TLR2 binding by measurement of cell migration and T cell proliferation. FACS analyses revealed expression of TLRs involved in sensing of bacterial PAMPs. In accordance with former studies about TLR4 in human LCs, we found TLR4 protein not to be expressed in epidermal LCs [22 , 24 , 25 ] but in MoLCs and MoDCs [22 , 23 ].
In a report using real-time PCR, the primers used detected mRNA expression of TLR1, TLR2, and TLR4 and lesser amounts of mRNA for TLR6 and TLR9 in CD34-derived, LC-like cells [23 ]. In another study using real-time PCR, no mRNA for TLR4 and TLR9 but for TLR1, TLR2, and TLR6 was found in LCs [24 ]. In contrast, a recent real-time PCR study reports on weak or absent TLR2 expression on CD1a+/HLA-DR+LCs [25 ]. In our analyses, we found protein expression for TLR1, TLR2, TLR6, and TLR9 in LCs, MoLCs, and MoDCs. In human blood, monocytes were found to express the largest amount of TLR2 [26 ]. However, DCs are the central cells for TLR-mediated pathogen sensing [5 ]. Among the subtypes of human DCs, TLR2 was detected in MoDCs and in isolated myeloid DCs but not in plasmacytoid DCs [5 , 27 ]. Our observations that epidermal LCs and MoLCs express TLR2 are supported by data about TLR2 surface expression in epidermal LCs/MoLCs and mRNA expression in LCs generated from CD34+cells [22 , 23 ]. In accordance with human LCs, murine LCs purified from the epidermis by the panning method also expressed transcripts of TLR2 [28 ]. In contrast to TLR2 expression in bacterial-stimulated murine macrophages [29 ], its level in epidermal LCs was unchanged after stimulation (not shown).
Initial studies demonstrated an up-regulation of costimulatory molecules in DCs after exposition to a parasite extract and to the endogenous danger signal CD40L [30 ]. Increased CD86 was reported in TLR2-stimulated MoDCs and in LCs expanded from murine fetal skin and mobilized by the inflammatory mediator LPS [31 , 32 ]. As a result of the lack of reports about the costimulatory phenotype after triggering TLRs expressed by human LCs, studies using stimulus CD40L were compared. As presented by our study in CD1c+MoLCs, the levels of CD80, CD86, and HLA-DR were elevated by stimulation with MALP2 and Pam3CSK4, similar to data about human LCs generated from CD34+progenitor cells and epidermal LCs after a 2-day stimulation with CD40L [33 34 35 ].
Although MoLCs and LCs expressed TLR5, cellular challenge with flagellin only slightly enhanced costimulatory molecule expression (Fig. 2B) . In parallel, proteins of TLR3, TLR7, and TLR8 were expressed by epidermal LCs and CD1c+MoLCs (own unpublished data), but CD80, CD86, and CCR7 were also only slightly enhanced by poly(I:C) and poly(U) (Fig. 2B) . These TLR agonists are known for mimicking ligands derived from nonbacterial pathogens.
In addition to analyses of surface molecules on stimulated cells, we detected the proinflammatory cytokines IL-6 and TNF- in supernatants of CD1c+MoLCs that were stimulated with specific TLR ligands. Notably, the TLR2 agonists MALP2 and Pam3CSK4 triggered increased release of both cytokines. Moreover, in CD1c+MoDCs, poly(I:C), a synthetic mimetic of viral, double-stranded RNA, induced only a weak secretion of IL-6 and TNF-. poly(U), a mimetic of viral, single-stranded RNA, did not induce secretion of IL-6 and TNF-. These results point to a pronounced response against bacteria in human LCs. However, LCs do not seem to sense bacterial pathogens exclusively. A role of LCs in detection of viral pathogens was recently described for TLR3-activated LCs secreting small amounts of IL-6 and TNF- [24
]. Other inflammatory mediators such as CXCL9 and CXCL11 might also be involved in antiviral responses, as shown for poly(I:C)-stimulated MoLCs [23
].
In our analyses and in a report of others [22
], MoLCs stimulated by the same concentrations of TLR agonists obviously secreted lower amounts of cytokines than MoDCs. In agreement with our data about IL-6, in another report [28
], the levels of IL-6 and IL-12p40 released from murine LCs were about 10 times lower than in murine splenic DCs after stimulation with the TLR2 ligand S. aureus Cowan 1. It has to be taken into account that TGF-β might prevent MoLCs from reaching the same degree of cell maturation that can be achieved by MoDCs. An alternative explanation for the differences in cytokine secretion by LCs and DCs might lie in the different physiological locations in which they reside. In contrast to blood DCs that secrete high amounts of cytokines to communicate with distant effector cells [36
], epidermal LCs are described to form a LC unit, in which one LC controls 
50 neighboring cells [37
]. In contrast to differences in cytokine secretion of TLR-stimulated human MoLCs and MoDCs from one individual donor, we could not identify significant differences in expression of costimulatory molecules (not shown).
Analyses of different agonists of TLR2 regarding their potential to induce IL-6 and TNF- demonstrated that MALP2, Pam3CSK4, and Pam2CSK4 provoked significant but lower cytokine amounts as compared with PGN. In contrast to other lipopetides, PGN is reported to bind TLR2 homomers, not TLR1 or TLR6. On the other hand, additional intracellular receptors can sense PGN. PGN-recognition proteins and NOD proteins detect bacterial PGN with high affinity and are involved in inflammatory responses [38
39
40
]. However, in our blocking experiments, an anti-TLR2 antibody at a concentration of 10 µg/ml reduced the TNF- secretion to the level of unstimulated cells and strongly decreased IL-6 secretion. The antagonistic activity of the T2.5 antibody used was shown to block human and murine TLR2 in macrophages and prevented septic shock-like syndrome upon systemic bacterial challenge [41
]. As shown for lipopeptide-stimulated MoDCs [31
], TLR2 blocking diminished CD86 in PGN-stimulated CD1c+MoLCs. This would indicate that intracellular PGN-sensing proteins may not have access to extracellular PGN or may not be involved in PGN detection by LCs.
Our results give a promising outlook on immunotherapy and vaccination strategies. In mice, a mucosal vaccine based on HIV-Tat in combination with MALP2 elicited antigen-specific, IFN-
-producing cells [42
]. As synthetic and bacteria-derived lipopeptides activate LCs via TLR2, they might serve as an adjuvant in therapies of viral skin and mucosa infections or of basal cell carcinomas.
Received August 24, 2007; revised December 5, 2007; accepted December 12, 2007.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
cooperate in the generation of dendritic Langerhans cells Nature 360,258-261[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  Y. Ogawa, T. Kawamura, T. Kimura, M. Ito, A. Blauvelt, and S. Shimada Gram-positive bacteria enhance HIV-1 susceptibility in Langerhans cells, but not in dendritic cells, via Toll-like receptor activation Blood, May 21, 2009; 113(21): 5157 - 5166. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
