Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 190 (12), 1755-68

Macrophage Inflammatory Protein 3alpha Is Involved in the Constitutive Trafficking of Epidermal Langerhans Cells

Affiliations

Macrophage Inflammatory Protein 3alpha Is Involved in the Constitutive Trafficking of Epidermal Langerhans Cells

A S Charbonnier et al. J Exp Med.

Abstract

Certain types of dendritic cells (DCs) appear in inflammatory lesions of various etiologies, whereas other DCs, e.g., Langerhans cells (LCs), populate peripheral organs constitutively. Until now, the molecular mechanism behind such differential behavior has not been elucidated. Here, we show that CD1a(+) LC precursors respond selectively and specifically to the CC chemokine macrophage inflammatory protein (MIP)-3alpha. In contrast, CD14(+) precursors of DC and monocytes are not attracted by MIP-3alpha. LCs lose the migratory responsiveness to MIP-3alpha during their maturation, and non-LC DCs do not acquire MIP-3alpha sensitivity. The notion that MIP-3alpha may be responsible for selective LC recruitment into the epidermis is further supported by the following observations: (a) MIP-3alpha is expressed by keratinocytes and venular endothelial cells in clinically normal appearing human skin; (b) LCs express CC chemokine receptor (CCR)6, the sole MIP-3alpha receptor both in situ and in vitro; and (c) non-LC DCs that are not found in normal epidermis lack CCR6. The mature forms of LCs and non-LC DCs display comparable sensitivity for MIP-3beta, a CCR7 ligand, suggesting that DC subtype-specific chemokine responses are restricted to the committed precursor stage. Although LC precursors express primarily CCR6, non-LC DC precursors display a broad chemokine receptor repertoire. These findings reflect a scenario where the differential expression of chemokine receptors by two different subpopulations of DCs determines their functional behavior. One type, the LC, responds to MIP-3alpha and enters skin to screen the epidermis constitutively, whereas the other type, the "inflammatory" DC, migrates in response to a wide array of different chemokines and is involved in the amplification and modulation of the inflammatory tissue response.

Figures

Figure 1
Figure 1
Chemotactic response of LC and non-LC DC precursors generated in vitro from CD34+ HPCs. CB-derived HPCs were stimulated for 6 d with GM-CSF/TNF-α and then studied in the transwell chemotaxis assay. MIP-3α (A), MCP-1 (B), SDF-1α (C), or MIP-3β (D) was placed in individual wells of 24-well plates, transwell devices were inserted, and cell suspensions were layered into the transwell inserts. After 2 h at 37°C, the cells that transmigrated into the lower compartment were harvested, stained with anti-CD1a–FITC and anti-CD14–PE, and subjected to FACS® analysis. Results are given as the number of transmigrated cells in percentage of the input cell number of individual subpopulations (CD1a+CD14, filled circles; CD14+CD1a, open squares; CD14CD1a, open triangles). Mean ± SEM values obtained in three different experiments are shown.
Figure 3
Figure 3
Differential responses of LC and non-LC DC precursors to MIP-3α and MIP-1α. (A and B) MIP-3α induces migration of LC but not non-LC DC precursors. Day 6 CD1a+CD14 and CD14+CD1a DC precursors were flow sorted, recultured overnight in GM-CSF/TNF-α–supplemented medium, and thereafter subjected to the transwell chemotaxis assay. MIP-3α or buffer only was placed in individual wells of 24-well plates, transwell devices were inserted, and cell suspensions were layered into the transwell inserts. After 2 h at 37°C, migrated cells that detached into the lower compartment and migrated cells that remained membrane bound were counted. (A) Mean percentages (± SD; n = 2) of all migrated cells (detached plus membrane bound); (B) mean percentages (± SD; n = 2) of migrated, membrane-bound cells; CD1a+CD14, filled circles; CD14+CD1a, open squares. (C) MIP-1α mediates rapid actin polymerization in non-LC DC but not LC precursors. Cells were incubated for the indicated time periods in the presence or absence of 100 ng/ml MIP-1α. Cells were fixed with PFA, exposed to anti-CD1a–CyChrome/anti-CD14–PE, and F-actin was stained by an incubation with saponin/phalloidin–FITC. The F-actin response (y-axis) of individual cell populations is given as the ratio of the mean phalloidin-FITC fluorescence intensity in the presence of MIP-1α over the mean phalloidin-FITC fluorescence intensity in its absence. X-axis, elapsed time; CD1a+CD14 LC precursors, filled circles; CD14+CD1a non-LC DC precursors, open squares; CD1aCD14 cells, open triangles. Mean ± SD values obtained in two different experiments.
Figure 2
Figure 2
MIP-1α and MIP-3α elicit different migratory and/or adhesive responses in CD34+ HPC-derived LC and DC precursors. Results shown in A and B were obtained using the transwell chemotaxis assay and the 48-well Boyden-type chamber chemotaxis assay, respectively. For both assays, day 6 LC and/or DC precursors were harvested and tested for their migratory responses to MIP-1α (filled circles), MIP-3α (open squares), and MIP-3β (open triangles). Mean percentages (± SEM) of migrated and detached (A; n = 3) and migrated, membrane-bound cells (B; n = 2) are shown. (C) Representative vertical sections through 5-μm pore size membranes used in the transwell chemotaxis assay. The assays were performed using the indicated stimuli (MIP-1α at 100 ng/ml, MIP-3α at 1 μg/ml, and the combination of both), or buffer alone. Membrane-bound cells were fixed and labeled, and membranes were subjected to confocal laser scanning microscopy. Broken lines denote the position of the membrane.
Figure 4
Figure 4
Chemotactic response of LCs and non-LC DCs generated in vitro from CD34+ HPCs. CB-derived HPCs were stimulated for 12 or 14 d with GM-CSF/TNF-α and then studied in the transwell chemotaxis assay. MIP-3α (A), SDF-1α (B), or MIP-3β (C) was placed in individual wells of 24-well plates, transwell devices were inserted, and cell suspensions were layered into the transwell inserts. After 2 h at 37°C, the cells that transmigrated into the lower compartment were harvested and stained with anti-CD1a–FITC, anti-CD11b–PE, and biotinylated anti-E-cad, followed by SA-Cy5. Results are given as the number of transmigrated LCs and non-LC DCs in percentage of the input cell number of each DC subpopulation (LCs: E-cad+CD11bCD1a+, filled circles; non-LC DCs: E-cadCD11b+CD1a+, open squares) at day 12 (broken lines) and day 14 (solid lines). Mean ± SEM values obtained in three different experiments are shown.
Figure 5
Figure 5
RT-PCR analysis of transcripts encoding chemokine receptors in different DC types generated in vitro and purified ex vivo. cDNAs from the following cell populations were prepared and subjected to PCR: (A) in vitro–generated LCs, derived from flow-sorted CD1a+ LC precursors of 6-d-old CD34+ HPC cultures, analyzed on day 12; (B) epidermal LCs, freshly isolated from skin and purified by FACS®; (C) flow-sorted epidermal LCs cultured for 48 h in the presence of GM-CSF/TNF-α; (D) bulk progeny of GM-CSF/TNF-α–stimulated CD34+ HPCs at day 6; (E) freshly isolated CD11c+ PB-DCs. RT, reverse transcription of mRNA; M, molecular size markers. For primer sequences, see Table .
Figure 6
Figure 6
Regulation of chemokine receptor expression during LC and non-LC DC development. (A) CD1a+ LC precursors express only CCR6, whereas CD14+ non-LC DC precursors display a wide array of chemokine receptors. CD34+ HPCs were stimulated for 6 d with GM-CSF/TNF-α, and cells were exposed simultaneously to CyChrome-labeled anti-CD1a; FITC- or PE-labeled anti-CD14; and FITC-, PE-, or biotin-conjugated anti-chemokine receptor mAbs. The binding of biotinylated mAbs was revealed by SA-PE. CD1a+CD14 LC precursors and CD14+CD1a non-LC DC precursors were electronically gated and analyzed for their CCR1, CCR2, CCR5, CCR6, and CXCR4 expression by FACS®. Filled histograms, reactivities of anti-chemokine receptor mAbs; open histograms, reactivities of label-matched control mAbs. (B) Detection of CCR6-expressing cells in normal human epidermis by immunohistochemistry. CCR6 immunoreactivity is seen primarily in basal and suprabasal layers of the epidermis. Strongly immunoreactive cells (arrows) have a dendritic configuration (inset), whereas a less pronounced and granular staining pattern is observed in cells with keratinocyte-like appearance (original magnifications: ×650; inset, ×1,600). (C) Freshly isolated epidermal LCs express CCR6, but rapidly downregulate this receptor during in vitro maturation. Freshly prepared (0 h) and cultured (4 h, 20 h) LC-enriched epidermal cell suspensions were exposed to anti–B7-2–FITC, anti-CCR6–PE, and anti–HLA-DR–PerCP, and analyzed by FACS®. HLA-DR+ LCs (which homogeneously displayed CD1a; data not shown) were electronically gated and analyzed for CCR6 (filled histograms, left) and B7-2 expression (filled histograms, right); open histograms, reactivities of fluorochrome-matched isotype control mAbs.
Figure 6
Figure 6
Regulation of chemokine receptor expression during LC and non-LC DC development. (A) CD1a+ LC precursors express only CCR6, whereas CD14+ non-LC DC precursors display a wide array of chemokine receptors. CD34+ HPCs were stimulated for 6 d with GM-CSF/TNF-α, and cells were exposed simultaneously to CyChrome-labeled anti-CD1a; FITC- or PE-labeled anti-CD14; and FITC-, PE-, or biotin-conjugated anti-chemokine receptor mAbs. The binding of biotinylated mAbs was revealed by SA-PE. CD1a+CD14 LC precursors and CD14+CD1a non-LC DC precursors were electronically gated and analyzed for their CCR1, CCR2, CCR5, CCR6, and CXCR4 expression by FACS®. Filled histograms, reactivities of anti-chemokine receptor mAbs; open histograms, reactivities of label-matched control mAbs. (B) Detection of CCR6-expressing cells in normal human epidermis by immunohistochemistry. CCR6 immunoreactivity is seen primarily in basal and suprabasal layers of the epidermis. Strongly immunoreactive cells (arrows) have a dendritic configuration (inset), whereas a less pronounced and granular staining pattern is observed in cells with keratinocyte-like appearance (original magnifications: ×650; inset, ×1,600). (C) Freshly isolated epidermal LCs express CCR6, but rapidly downregulate this receptor during in vitro maturation. Freshly prepared (0 h) and cultured (4 h, 20 h) LC-enriched epidermal cell suspensions were exposed to anti–B7-2–FITC, anti-CCR6–PE, and anti–HLA-DR–PerCP, and analyzed by FACS®. HLA-DR+ LCs (which homogeneously displayed CD1a; data not shown) were electronically gated and analyzed for CCR6 (filled histograms, left) and B7-2 expression (filled histograms, right); open histograms, reactivities of fluorochrome-matched isotype control mAbs.
Figure 6
Figure 6
Regulation of chemokine receptor expression during LC and non-LC DC development. (A) CD1a+ LC precursors express only CCR6, whereas CD14+ non-LC DC precursors display a wide array of chemokine receptors. CD34+ HPCs were stimulated for 6 d with GM-CSF/TNF-α, and cells were exposed simultaneously to CyChrome-labeled anti-CD1a; FITC- or PE-labeled anti-CD14; and FITC-, PE-, or biotin-conjugated anti-chemokine receptor mAbs. The binding of biotinylated mAbs was revealed by SA-PE. CD1a+CD14 LC precursors and CD14+CD1a non-LC DC precursors were electronically gated and analyzed for their CCR1, CCR2, CCR5, CCR6, and CXCR4 expression by FACS®. Filled histograms, reactivities of anti-chemokine receptor mAbs; open histograms, reactivities of label-matched control mAbs. (B) Detection of CCR6-expressing cells in normal human epidermis by immunohistochemistry. CCR6 immunoreactivity is seen primarily in basal and suprabasal layers of the epidermis. Strongly immunoreactive cells (arrows) have a dendritic configuration (inset), whereas a less pronounced and granular staining pattern is observed in cells with keratinocyte-like appearance (original magnifications: ×650; inset, ×1,600). (C) Freshly isolated epidermal LCs express CCR6, but rapidly downregulate this receptor during in vitro maturation. Freshly prepared (0 h) and cultured (4 h, 20 h) LC-enriched epidermal cell suspensions were exposed to anti–B7-2–FITC, anti-CCR6–PE, and anti–HLA-DR–PerCP, and analyzed by FACS®. HLA-DR+ LCs (which homogeneously displayed CD1a; data not shown) were electronically gated and analyzed for CCR6 (filled histograms, left) and B7-2 expression (filled histograms, right); open histograms, reactivities of fluorochrome-matched isotype control mAbs.
Figure 7
Figure 7
Chemokine expression in normal human skin. (A) Pronounced MIP-3α immunostaining of basal and suprabasal keratinocytes. (B) Rabbit Ig control staining (Ig-Co). (C) Only basal keratinocytes display SDF-1α immunoreactivity. (D) No MIP-3β immunoreactivity is seen in the epidermis, but scattered dermal cells are stained (arrow). (E) Neither the epidermis nor the dermis reacts with anti–MCP-1 Abs. (F and G) ECs lining afferent lymphatics display MIP-3β and MIP-3α immunoreactivity (asterisks in F and G, respectively). (H) MIP-3α–positive ECs lining a postcapillary venule (original magnifications: A and B, ×1,400; C, D, and E, ×900; F and G, ×2,100; H, ×3,000). (I) RT-PCR analysis of MIP-3α and MIP-3β mRNA expression by keratinocytes (KC); freshly isolated (LC fresh) as well as ex vivo–matured epidermal LCs (LC cult.); fibroblasts (FB); DMECs; and HUVECs, the latter two with and without IFN-γ stimulation. RT, reverse transcription of mRNA; M, molecular size markers; PC, positive control for MIP-3α and MIP-3β mRNA expression (i.e., reverse-transcribed mRNA from tonsillar extracts, reference 14). PCR with β-actin–specific primers is shown (bottom; control for the cDNA content of individual samples).
Figure 7
Figure 7
Chemokine expression in normal human skin. (A) Pronounced MIP-3α immunostaining of basal and suprabasal keratinocytes. (B) Rabbit Ig control staining (Ig-Co). (C) Only basal keratinocytes display SDF-1α immunoreactivity. (D) No MIP-3β immunoreactivity is seen in the epidermis, but scattered dermal cells are stained (arrow). (E) Neither the epidermis nor the dermis reacts with anti–MCP-1 Abs. (F and G) ECs lining afferent lymphatics display MIP-3β and MIP-3α immunoreactivity (asterisks in F and G, respectively). (H) MIP-3α–positive ECs lining a postcapillary venule (original magnifications: A and B, ×1,400; C, D, and E, ×900; F and G, ×2,100; H, ×3,000). (I) RT-PCR analysis of MIP-3α and MIP-3β mRNA expression by keratinocytes (KC); freshly isolated (LC fresh) as well as ex vivo–matured epidermal LCs (LC cult.); fibroblasts (FB); DMECs; and HUVECs, the latter two with and without IFN-γ stimulation. RT, reverse transcription of mRNA; M, molecular size markers; PC, positive control for MIP-3α and MIP-3β mRNA expression (i.e., reverse-transcribed mRNA from tonsillar extracts, reference 14). PCR with β-actin–specific primers is shown (bottom; control for the cDNA content of individual samples).

Similar articles

See all similar articles

Cited by 64 PubMed Central articles

See all "Cited by" articles

References

    1. Steinman R.M. The dendritic cell system and its role in immunogenicity. Annu. Rev. Immunol. 1991;9:271–296. - PubMed
    1. Banchereau J., Steinman R.M. Dendritic cells and the control of immunity. Nature. 1998;392:245–252. - PubMed
    1. Stingl G., Maurer D., Hauser C., Wolff K. The epidermisan immunologic microenvironment. In: Freedberg I.M., Eisen A.Z., Wolff K., Austen K.F., Goldsmith L.A., Katz S.I., Fitzpatrick T.B., editors. Dermatology in General Medicine. Vol. 1. McGraw-Hill; New York, NY: 1999. pp. 343–370.
    1. Maurer D., Stingl G. Dendritic cells in the context of skin immunity. In: Lotze M.T., Thompson A.W., editors. Dendritic Cells. Academic Press; San Diego, CA: 1999. pp. 111–122.
    1. Robert C., Fuhlbrigge R.C., Kieffer J.D., Ayehunie S., Hynes R.O., Cheng G., Grabbe S., von Andrian U.H., Kupper T.S. Interaction of dendritic cells with skin endotheliumA new perspective of immunosurveillance. J. Exp. Med. 1999;189:627–636. - PMC - PubMed

Publication types

LinkOut - more resources

Feedback