|
|
||||||||
CUTTING EDGE |


* Department of Dermatology and
Department of Cell Biology and Histology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands
| Abstract |
|---|
|
|
|---|
1-driven development of LCs in vitro. Consequently, and in contrast to DDCs, LCs weakly respond to bacterial TLR2, TLR4, and TLR5 ligands in terms of cytokine production and maturation, as well as to whole Gram-positive and Gram-negative bacteria, whereas their responsiveness to viral TLR ligands and viruses is fully active and comparable to DDCs. Unresponsiveness of LCs to bacteria may be a mechanism that contributes to tolerance to bacterial commensals that colonize the skin. | Introduction |
|---|
|
|
|---|
1 present in the epidermal microenvironment (6), and the observation that TGF-
1 knockout mice are devoid of LCs further establishes that TGF-
1 is crucial for the development of LCs (7, 8). Accordingly, TGF-
1 determines the differentiation of peripheral blood monocytes into LC-like cells (9, 10). To date, it is unknown whether LCs and DDCs have different roles in immunity and tolerance. In this study, we report that LCs, in contrast to DDCs, have a remarkable, strongly impaired expression of cell surface TLR2, TLR4, and TLR5, resulting in a selective low reactivity to bacteria. This attenuated reactivity of LCs to bacteria may be a mechanism contributing to tolerance to the bacterial commensal skin flora. | Materials and Methods |
|---|
|
|
|---|
Epidermal and dermal cell suspensions were obtained from residual skin obtained from plastic surgery as described previously (11, 12). Overnight incubation at 4°C in 0.2% dispase II (Boehringer Mannheim) enabled separation of epidermis and dermis, whereafter epidermal sheets were incubated in 0.25% trypsin solution (5 min at 37°C; Invitrogen Life Technologies), and dermal sheets in IMDM containing 0.2% collagenase D (Boehringer Mannheim), 40 U/ml DNase I (Boehringer Mannheim), and 2% FCS for 2 h at 37°C to yield single cells. After labeling with fluorescent-conjugated mAbs, (BD Biosciences), highly pure (>95%) CD1a+HLA-DR+ LCs and CD3CD14HLA-DR+CD11c+CD45+ DDCs were isolated by FACS sorting (BD Biosciences). Monocyte-derived DCs (moDCs) were generated by culture of peripheral blood monocytes in medium supplemented with IL-4 and GM-CSF (13), whereas additional supplementation with 10 ng/ml TGF-
1 (R&D Systems) resulted in monocyte-derived LCs (moLCs) (9).
Stimulation of DCs
Immature moDCs and moLCs were stimulated at a density of 3.5 x 104 cells in a 96-well culture plate with TLR agonists, UV-inactivated bacteria (range 105107), or heat-inactivated viruses (30 min, 56°C). TLR agonists used were as follows: LPS (Escherichia coli), PGN (Staphylococcus aureus), and dsRNA (poly(deoxyinosinic-deoxycytidylic acid)) from Sigma-Aldrich. PAM3CSK4, ssRNA (LyoVec), resiquimod (R848), flagellin (Salmonella typhimurium), lipoteichoic acid (S. aureus), and FSL were purchased from InvivoGen. Cytokine levels were analyzed by ELISA and the expression of maturation markers CD86 and CD83 by FACS (13). Gram-negative bacteria: E. coli B12G1 (14), Klebsiella pneumoniae (clinical isolate), Neisseria meningitidis H44/76 (14), and Gram-positive bacteria: S. aureus 42D (15), Corynebacterium xerosis (clinical isolate), and Enterococcus faecalis 19916 (15) were provided by S. A. J. Zaat (Department of Microbiology, Academic Medical Center, Amsterdam, The Netherlands). Varizella Zoster virus and CMV were provided by H. W. M. van Eijk (Department of Clinical Virology, Academic Medical Center). Influenza virus (FLU) A/PR/8/34 was a gift from G. Rimmelzwaan (Department of Virology, Erasmus Medical Center Rotterdam, Rotterdam, The Netherlands). The titers of virus were FLU A, 1.16 x 1010/ml; Varizella Zoster virus, 6.36 x 107/ml, and CMV, 1.07 x 108, determined by quantitative RT-PCR (16).
Real-time quantitative RT-PCR analysis
moLCs were purified (>98%) by FACS sorting (BD Biosciences) based on CD1a and E-cadherin expression, and moDCs were sorted based on forward side scatter. mRNA was extracted (isolation kit from Machery-Nagel) and cDNA transcripts (MBI Fermentas) were quantified by real-time quantitative PCR (iCycler iQ MultiColor Real-Time PCR Detection System; Bio-Rad) with specific primers (17) and general SYBR green (Bio-Rad) fluorescence detection. mRNA expression of each sample was normalized to GAPDH (18).
Statistics
We performed Wilcoxon or Student t tests for paired measurements with GraphPad software (GraphPad InStat). Values of p < 0.05 were considered significant.
| Results |
|---|
|
|
|---|
When analyzing highly purified ex vivo DDCs and LCs (Fig. 1A), we found that DDCs expressed TLR1 through TLR8, but surprisingly, LCs expressed TLR1, TLR3, TLR6, and TLR7, whereas the expression of TLR2, TLR4, TLR5, and TLR8 was weak or even absent (Fig. 1B). Both DDCs and LCs did not express TLR9 and TLR10. Quantitative RT-PCR confirmed the significantly lower expression of TLR2, TLR4, TLR5, and TLR8 by LCs (Fig. 1C). No significant differences in TLR1 and TLR3 expression were found, whereas TLR6 expression was significantly higher in LCs than in DDCs. However, it may be expected that TLR1 and TLR6 expression in LCs is not functional because these TLRs form heterodimers with TLR2, which is absent in LCs. These data clearly indicate that LCs lack expression of functional TLRs involved in the recognition of bacterial components.
|
MoDCs and moLCs, generally considered as equivalents of LCs and DDCs, typically exhibited the phenotype of immature DCs (CD14, CD1a+/, HLA-DR+, CD83, and CD86low), and moLCs could be discriminated by expression of E-cadherin and Langerin (data not shown). The TLR expression profiles of FACS-purified moDCs and moLCs (Fig. 2A) were similar to those of ex vivo DDCs and LCs, respectively. Compared with moDCs, moLCs expressed significantly lower levels of TLR2, TLR4, TLR5, and TLR8, nonsignificantly lower levels of TLR7, similar levels of TLR1 and TLR3, and significantly enhanced levels of TLR6 mRNA (Fig. 2B). This phenotypical similarity strongly suggests that TGF-
1 plays an important role in the differential TLR expression profile of LCs and DDCs.
|
To test whether their different TLR expression profile results in different responses to microbial compounds, moDCs and moLCs were stimulated with various specific TLR agonists at different concentrations (Fig. 3) for different periods of time (data not shown). Ligation of intracellular TLRs induced equal levels of IL-6 and IL-12p70 production in moDCs and moLCs (Fig. 3A). In these experiments, TNF-
production was significantly increased in moLCs upon TLR3 ligation. In notable contrast, the bacterial-derived agonists LPS (TLR4), flagellin (TLR5), Pam3CSK4 (TLR2/TLR1), lipoteichoic acid (TLR2; data not shown), and FSL (TLR2/TLR6; data not shown) induced a substantially lower IL-6 and TNF-
production by moLCs compared with moDCs at all concentrations (Fig. 3B). None of the extracellular TLR agonist induced detectable levels of IL-12p70 by moDCs and moLCs. These results stress that the strongly reduced expression of cell surface TLRs by LCs has the functional consequence that these cells are unable to recognize bacterial components. The only exception was PGN, which induced considerable TNF-
production in moLCs (Fig. 3B). Notably, no significant effect of TGF-
1 on the responsiveness to microbial compounds was observed when this cytokine was present during TLR triggering of moLCs and moDCs, regardless of which TLR was stimulated (data not shown). This indicates that the TGF-
1-driven loss of bacteria recognizing TLRs is accomplished during the differentiation of moLCs from their precursors, whereas the actual responsiveness of differentiated DCs to TLR ligands is not affected by presence of TGF-
1. In line with the cytokine responses, bacterial-derived agonists, with the exception of PGN, failed to induce expression of the maturation markers CD86 (Fig. 3C) and CD83 (data not shown) in moLCs, but not in moDCs, whereas both DC subsets acquired a mature phenotype after triggering with intracellular TLRs.
|
The inability to respond to bacterial TLR ligands strongly suggests that LCs may be unable to respond to whole bacteria. Therefore, we tested the responsiveness of moDCs and moLCs to whole bacteria and viruses using the production of IL-12, IL-6, and TNF-
as read out (Fig. 4, A and B). In sharp contrast to moDCs, moLCs produced no or significantly lower amounts of cytokines in response to Gram-positive and Gram-negative bacteria (Fig. 4A). No difference was found in the responsiveness of moDCs and moLCs to viruses (Fig. 4B). Furthermore, whole bacteria were unable to induce maturation in moLCs at all tested concentrations but did so in moDCs, whereas the FLU virus induced CD86 (Fig. 4C) and CD83 (data not shown) in both moLCs and moDCs. These results clearly show that moLCs have an impaired reactivity to bacteria, while their ability to react effectively to viruses remained fully operational.
|
| Discussion |
|---|
|
|
|---|
1 in cultures of the same pool of monocytes, the experiments suggest that TGF-
is a key molecule in determining the inability of LCs to recognize bacteria. LCs displayed a strongly diminished production of cytokines and up-regulation of maturation markers CD83 and CD86 in response to specific ligands for TLR2, TLR4, and TLR5, confirming the impaired expression of these TLRs. PGN formed the only exception because this ligand triggered LCs to produce TNF-
and to express CD83 and CD86. This reactivity may result from weak TLR2 expression by LCs. However, PGN also triggers other PRRs, such as NOD2 and NALP3, resulting in TNF-
production (22, 23, 24). Indeed, moLCs expressed NOD2 mRNA at a comparable level as moDCs (data not shown), indicating that moLCs can respond to PGN via other pathways. The significance of the reactivity to PGN remains to be established since LCs do not respond to whole bacteria. The differential responsiveness to single TLR agonists and whole bacteria may be explained by the fact that bacteria trigger multiple PRRs, which may amplify as well as inhibit responses to single TLR agonists (25). In contrast to the impaired bacterial recognition, we showed that LCs are fully responsive to viruses. However, it is not clear yet whether LCs are the actual inducers of antiviral immunity because two studies with in vivo models on epithelial HSV infection argue against a role for LC in direct priming of protective antiviral-specific T cells (26, 27). Our findings are in line with a study by Takeuchi et al. (28), reporting that human moLCs have low TLR2 and no TLR4 expression and respond weakly to PGN and not to LPS. However, this study did not detail on the questions to what extend LCs do respond to other TLR ligands and, more importantly, whether these cells differentially respond to whole microorganisms. Another study (29) reported that LC-like cells derived from CD34+ cord blood cells express TLR110, which is not in line with our study and the study by Takeuchi et al. (28). The validity of cord blood-derived LC-like cells as a model of LCs in this study is unclear because these cells were not purified, and their TLR profile is dissimilar to the TLR profile of authentic human LCs.
The inability of LCs to respond efficiently to bacteria, as shown in this study, implicates that the presence of bacteria will not easily provoke LC-mediated local inflammation and onset of specific immunity. This impaired reactivity and maturation of LCs may be a mechanism that contributes to immunological tolerance to commensal bacteria of the skin. In addition, uptake of bacteria by LCs in the absence of TLR activation could result in specific tolerance induction since studies in mice have demonstrated that immature or semimature DCs, that do not produce polarizing cytokines, induce Ag-specific tolerance (30, 31). However, bacteria that are successful in colonizing the more vulnerable dermal layer of the skin will readily trigger DDCs, even in the presence of TGF-
1. LCs are generally referred to as the first immunological barrier against pathogens. The impaired reactivity of LCs to bacteria suggests that this is only partially true. We propose that constitutive production of TGF-
in the epidermis creates an immunopriviliged compartment through the enforcement of a unique DC phenotype associated with selective tolerance to extracellular bacteria. This prevents unnecessary inflammation in response to harmless skin commensals while preserving responsiveness to intracellular pathogens.
| Acknowledgments |
|---|
| Disclosures |
|---|
|
|
|---|
| Footnotes |
|---|
1 This work was supported by the Netherlands Institute for Pigment Disorders. ![]()
2 E.C.d.J. and M.B.M.T. contributed equally to this work. ![]()
3 Address correspondence and reprint requests to Dr. Marcel B. M. Teunissen, Department of Dermatology, University of Amsterdam, Academic Medical Center, Meibergdreef 9, 1105AZ Amsterdam, The Netherlands. E-mail address: m.b.teunissen{at}amc.uva.nl ![]()
4 Abbreviations used in this paper: DC, dendritic cell; LC, Langerhans cell; DDC, dermal DC; PRR, pattern recognition receptor; moDC, monocyte-derived DC; moLC, monocyte-derived LC; PGN, peptidoglycan; FLU, influenza virus. ![]()
Received for publication November 6, 2006. Accepted for publication December 11, 2006.
| References |
|---|
|
|
|---|
1 regulation of dendritic cells. Microbes Infect. 1: 1283-1290. [Medline]
1 in Langerhans cell biology: the skin of transforming growth factor
1 null mice is devoid of epidermal Langerhans cells. J. Exp. Med. 184: 2417-2422.
1 and -
2(1). J. Invest. Dermatol. 117: 1574-1580. [Medline]
1, in the presence of granulocyte/macrophage colony-stimulating factor and interleukin 4, induces differentiation of human peripheral blood monocytes into dendritic Langerhans cells. J. Exp. Med. 187: 961-966.
B. J. Biol. Chem. 276: 4812-4818.
+ dendritic cells but not by Langerhans cells. Science 301: 1925-1928.
induce antigen-specific protection of mice from autoimmunity. J. Exp. Med. 195: 15-21. [Medline]This article has been cited by other articles:
![]() |
S. Chamorro, J. J. Garcia-Vallejo, W. W. J. Unger, R. J. Fernandes, S. C. M. Bruijns, S. Laban, B. O. Roep, B. A. 't Hart, and Y. van Kooyk TLR Triggering on Tolerogenic Dendritic Cells Results in TLR2 Up-Regulation and a Reduced Proinflammatory Immune Program J. Immunol., September 1, 2009; 183(5): 2984 - 2994. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. S. Duraisingham, J. Hornig, F. Gotch, and S. Patterson TLR-Stimulated CD34 Stem Cell-Derived Human Skin-Like and Monocyte-Derived Dendritic Cells Fail to Induce Th17 Polarization of Naive T Cells but Do Stimulate Th1 and Th17 Memory Responses J. Immunol., August 15, 2009; 183(4): 2242 - 2251. [Abstract] [Full Text] [PDF] |
||||
![]() |
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] |
||||
![]() |
O.V. Horst, K.A. Tompkins, S.R. Coats, P.H. Braham, R.P. Darveau, and B.A. Dale TGF-{beta}1 Inhibits TLR-mediated Odontoblast Responses to Oral Bacteria Journal of Dental Research, April 1, 2009; 88(4): 333 - 338. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. R. Mathers, B. M. Janelsins, J. P. Rubin, O. A. Tkacheva, W. J. Shufesky, S. C. Watkins, A. E. Morelli, and A. T. Larregina Differential Capability of Human Cutaneous Dendritic Cell Subsets to Initiate Th17 Responses J. Immunol., January 15, 2009; 182(2): 921 - 933. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Reske, G. Pollara, C. Krummenacher, D. R. Katz, and B. M. Chain Glycoprotein-Dependent and TLR2-Independent Innate Immune Recognition of Herpes Simplex Virus-1 by Dendritic Cells J. Immunol., June 1, 2008; 180(11): 7525 - 7536. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Peiser, J. Koeck, C. J. Kirschning, B. Wittig, and R. Wanner Human Langerhans cells selectively activated via Toll-like receptor 2 agonists acquire migratory and CD4+T cell stimulatory capacity J. Leukoc. Biol., May 1, 2008; 83(5): 1118 - 1127. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Kumar, J. Yin, J. Zhang, and F.-S. X. Yu Modulation of Corneal Epithelial Innate Immune Response to Pseudomonas Infection by Flagellin Pretreatment Invest. Ophthalmol. Vis. Sci., October 1, 2007; 48(10): 4664 - 4670. [Abstract] [Full Text] [PDF] |
||||
![]() |
C.-L. Ku, H. von Bernuth, C. Picard, S.-Y. Zhang, H.-H. Chang, K. Yang, M. Chrabieh, A. C. Issekutz, C. K. Cunningham, J. Gallin, et al. Selective predisposition to bacterial infections in IRAK-4 deficient children: IRAK-4 dependent TLRs are otherwise redundant in protective immunity J. Exp. Med., October 1, 2007; 204(10): 2407 - 2422. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |