Clinical and Experimental Dermatology 1992; 17: 240-245,
Flow cytometry analysis of adhesion molecules on human Langerhans cells J.VEDEL,*t Ph.VINCENDEAU,t J.-H.BEZIANf AND A.TAIEB*
* Service de Dermatologie,
Hopital des Enfants et ^Laboratoire d'Immunologie, Universite de Bordeaux II, Bordeaux, France Accepted for publication \Q September 1991
Summary The Langerhans cell (LC) migrates between the epidermis and the regional lymph nodes to present antigens. This migration pattern requires the expression of a changing repertoire of cell-surface molecules. In this work, we have investigated the expression ofthe adhesion molecules CD 11/CD 18 and CD 58 on LCs. Human epidermal cell suspensions were enriched in LCs (mean enrichment 75'^o) using a two-step technique including a Ficoll-Hypaque gradient followed by Fc receptor panning with IgG-coated sheep erythrocytes. The number ot cells obtained per experiment was 750 000 (extremes 280000-1800000), and the following antibodies were tested on fresh suspensions and/or after 48 hours in culture: BB3 (antithyroglobulin negative control IgG2a), OKT6 (anti CDla, Ortho), ami HLA-DR (BectonDickinson), MHiM 24 (anti CD 11a, leukocyte typing workshop n"3), MOl and 44 (anti CD lib, leukocyte typing workshop n"3), anti CD lie (Immunotech), 60-3 and MHM 23 (anti CD 18, leukocyte typing workshop n"2), TS2/911 (anti CD 58, leukocyte typing workshop n"3). We found that amongst CD 11 subunits, only CD 1 lc was expressed in fresh suspensions, but was weaker than CD 18, and disappeared with culture. CD 58 was not detected in fresh suspensions but appeared after 2 days of culture, confirming earlier work. Thus the LC exhibits cell surface characteristics similar to tissue macrophages (CD 18 and CD Uc) prior to culture. The expression of CD 58 after culture is in accordance with the interaction of LC with CD2 bearing T-lymphocytes during antigen presentation in peripheral lymph-nodes.
After homing to the epidermis, Langerhans ceils (LCs) originating from the bone marrow' must interact in a variety of milieus to function efficiently as antigenpresenting cells by trafficking mainly between the epidermis and regional lymph-nodes.^ Adhesion molecules have Correspondence: Professor A,Taieb, Unite de Laboratoire d'Immunologie, Bat IB, Carreire Nord, Universite de Bordeaux II, 146 rue Leo Saignat, F-33076 Bordeaux Cedex, France.
240
been shown to be ofthe utmost importance in mediating further steps in cell activation.' We first realized that adhesion was probably an important step in keratinocyteI .C interactions when adding purified I ,Cs suspensions to keratinocyte cultures and noted that LCs attached preferentially to keratinocytes as compared to blood mononuclear cells. Flow cytometry showed that the LFA-1 ft subunit (CD 18) and LFA-3 (CD 58) were expressed by LCs after 48 h in culture with or without stimulation with lipopolysaccharide (LPS)."* Several investigators using various methods of purification or staining of LCs have since published comparable findings,' " However, possibly due to different techniques, the demonstration of adhesion molecules on human LCs has not been uniform. We would like to report our results using enrichment of LCs with Fc receptor panning and analysis of surface molecules by flow cytometry and compare our data to the published literature.
Materials and methods LC enrichment
Skin specimens were obtained from adult women undergoing breast reduction. F-pidermal cell suspensions were obtained after overnight trypsinization, using 0 25"^ trypsin (Sigma) in buffer (130 mM Nad, 30 mM HEPES, 10 mM glucose, 3 mM KCl, 1 mM Na2HPO41\U\ pH adjusted to 7 4 with NaOH). Cells were filtered with .i 37 /(m nylon gauze, washed in cold buffer and resiispcnded in RPMI 1640 (Gibco), and counted with trypan blue. Density gradient centrifugation (1 077 Hisio]>;jquc Sigma) allowed pre-enrichment (6-S'\, LC) and hii:hl\ viable (>95'!o) suspensions were obtained ior pannini:. The following steps were adapted from the 1-c paniiinu: technique on sheep erythrocytes described by Rasancn ci al.^- Briefly, poly-i.-lysine treated 100 mm IVlri dishes were covered with a solution of washed sheep er\ilirncytes [Biomerieux, Lyon, I'Vance) (I",, in Dulbecco's phosphate-buffered saline fPIiSI ICiibcoj) 4,^ nm ai room temperature, rinsed three times with I'MS, aiui incu
LEUCOCYTE ADHESION MOLECULES AND LANGERHANS CELLS 45 mn at 37"C with an IgG rabbit-anti-sheep erythrocyte antiserum (Cappel, West Chester, Pennsylvannia, USA) 1 : 500 in PBS. After three washes, a maximum of 20 x 10" epidermal cell suspensions in cold RPMI 1640 were added to each dish and centrifuged 200 g for 5 min and left at 4 X for 2 h for panning. After aspiration of nonadherent cells, LCs were detached from sheep red blood eells monolayers with an erythroeyte lysing medium, washed twice in RPMI, and counted.
241
Results Cell separation and purity of LC suspensions
In a series of eight experiments included in this study, a mean of 3 3x10** epidermal cells were obtained per experiment (range 1-8-7-5 x 10") with a 75± 15% trypan blue exclusion. After panning, an average 7 5 x 10' cells were obtained per experiment (range 2-8 x 10'-l-8x 10'') with a 90 ± 5% trypan blue exclusion. Purity was checked using fluorescence microscopy (counts of a total of 500 cells using phase contrast and fluorescence in the same Monoclonal antibodies andflowcytometry field with a 20 X objective) and flow cytometry using an Table 1 shows the antibodies used in this study at a final anti CDla FITC antibody (OKT6 Ortho). As shown by dilution of I : 20. For two step immunofluorescence, a flow cytometry in Figure 1 in a representative experiGAM-FITC F (ab')2 conjugate (NEN Research Pro- ment, about 75% cells were CDla positive using both ducts) was used at 1 : 100, For flow cytometric analysis, methods. 0 5-1x10' stained cells were resuspended in 100 /il 200 400 1000 Dulbecco's PBS with 0 01% sodium azide and analysed 600 800 r [I 1—I—1—1—I—I [ I—I—r—1—I—I— with a FACScan (Becton-Dickinson) equipped with a Hewlett Packard HP 30 computer.
Table L Monoclonal antibodies used in the study
Antibody BB3t
0KT6-FITC a-HLA-DRJ BU 15 (lOM lie) MHM24II MOlq 441! 60'3II MHM23«rm// of 19, Marlin SD, Springer TA, Purified intercellular adhesion molInvestigative Dermatology 1989; 93: (lO 69. ecule-1 (lCAM-1) is a ligand for lymphocyte function-associated (i, De Panfilis G, Manara GC, Ferrari C, Torresani C^. Adhesion antigen-1 (LI"A-1) Cell 1988; 51: 813-819, molecules on the plasma membrane of epidermal cells, H, Tbe intercellular adhesion molecule 1 is constitutively present on the 20, Bierer BF, Barbosa J, Herrmann S, Burakofi SJ, Interaction of CD2 with its ligand, LFA3, in human T cell proliferation. Journa/ cell surface of human resting Langerbans cells. Journal of of Immunology 1988; 140: 3358-3363. Investigative Dermatology 1990; 94: 317 321. 21, Pattaroyo M, Makgoba MW. Leukocyte adhesion to cells: 7, Romani N, Lenz A, GIa.s.sel H, Stossel H, Stanzl U, Majdic O, molecular basis, physiological relevance, and abnormalities. Fritscb P, Scbuler G. Cultured buman I .angerhans cells resemble Scandinavian Journal of Immunology 1989; 30: 129-164. tymphoid dcntritic cells in pbenolypc and function. Journal of 22, Witmer-Pack M, Oliver W, Valinsky J, Schuler G, Steinman R, investigative Dermatology 1989; 93: 600 6(19. Granulocyte/macrophage colony-stimulating factor is essential S, Teunissen M, WormmeesterJ, Krieg SR, Peters PJ, Vogeis I.MC, for the viability and fonction of cultured murine epidermal Kapsenberg MI., lJos JD, Human epidermal Langerhans cells. Langerhans cells. Journal of Experimental Medicine 1987- 166' L ndergo profound morpbologic and phenotypical changes during 1484-1498, in vitro culture. Journal of Investigative Dermatology, ]99{); 94: 23, I leuflcr C:, Koch F, Schuler G. Granulocyte/macrophage colony 166-173. siiniLilating factor and interleukin I mediate the maturation of 9, Staquet MJ, Dezutter-Dambuyant C, Schmitt D, Tbivolet J, murine epidermal Langerhans cells into potent immuno stimulaAdhesion mediating molecules of human epidermal Langerhans tory dendritic cells. Journal of Experimental Medicine 1988; 167cdh. Journal of Invesiigalive Dermatology 1989; 92: S22. 700- 705, 10, Weber-Matthiescn K, Sterry \V, Organization of the monocyte24, Kampgen F, Heufler C, Koch F, Scbuler G, High numbers of macrophage system of normal human skin. Journal of inve.Uigative GM-CSF receptors arc expressed on epidermal Langerhans cells Dermatology 1990; 93: 83 -89. and lymphoid dendritic cells, and mediate tbe induction/increase 11, Simon J(^, Ouz PD, Iigelaar Re, Sontheimer RD, Bergstresscr of accessory cell function by GM-CSI', Archives ofDermatolomcal PR. AdbesionmoIeculesCDlla, CD18, andlCAM-I on human Research 1990, 281: 546 (Abstract). epidermal Langerhans cells serve a functional role in tbe activation of alloreactive T ce\h. Journal of Investigative Dermatology 1991; 96: 148-151,
Flow cytometry analysis of adhesion molecules on human Langerhans cells J.VEDEL,*t Ph.VINCENDEAU,t J.-H.BEZIANf AND A.TAIEB*
* Service de Dermatologie,
Hopital des Enfants et ^Laboratoire d'Immunologie, Universite de Bordeaux II, Bordeaux, France Accepted for publication \Q September 1991
Summary The Langerhans cell (LC) migrates between the epidermis and the regional lymph nodes to present antigens. This migration pattern requires the expression of a changing repertoire of cell-surface molecules. In this work, we have investigated the expression ofthe adhesion molecules CD 11/CD 18 and CD 58 on LCs. Human epidermal cell suspensions were enriched in LCs (mean enrichment 75'^o) using a two-step technique including a Ficoll-Hypaque gradient followed by Fc receptor panning with IgG-coated sheep erythrocytes. The number ot cells obtained per experiment was 750 000 (extremes 280000-1800000), and the following antibodies were tested on fresh suspensions and/or after 48 hours in culture: BB3 (antithyroglobulin negative control IgG2a), OKT6 (anti CDla, Ortho), ami HLA-DR (BectonDickinson), MHiM 24 (anti CD 11a, leukocyte typing workshop n"3), MOl and 44 (anti CD lib, leukocyte typing workshop n"3), anti CD lie (Immunotech), 60-3 and MHM 23 (anti CD 18, leukocyte typing workshop n"2), TS2/911 (anti CD 58, leukocyte typing workshop n"3). We found that amongst CD 11 subunits, only CD 1 lc was expressed in fresh suspensions, but was weaker than CD 18, and disappeared with culture. CD 58 was not detected in fresh suspensions but appeared after 2 days of culture, confirming earlier work. Thus the LC exhibits cell surface characteristics similar to tissue macrophages (CD 18 and CD Uc) prior to culture. The expression of CD 58 after culture is in accordance with the interaction of LC with CD2 bearing T-lymphocytes during antigen presentation in peripheral lymph-nodes.
After homing to the epidermis, Langerhans ceils (LCs) originating from the bone marrow' must interact in a variety of milieus to function efficiently as antigenpresenting cells by trafficking mainly between the epidermis and regional lymph-nodes.^ Adhesion molecules have Correspondence: Professor A,Taieb, Unite de Laboratoire d'Immunologie, Bat IB, Carreire Nord, Universite de Bordeaux II, 146 rue Leo Saignat, F-33076 Bordeaux Cedex, France.
240
been shown to be ofthe utmost importance in mediating further steps in cell activation.' We first realized that adhesion was probably an important step in keratinocyteI .C interactions when adding purified I ,Cs suspensions to keratinocyte cultures and noted that LCs attached preferentially to keratinocytes as compared to blood mononuclear cells. Flow cytometry showed that the LFA-1 ft subunit (CD 18) and LFA-3 (CD 58) were expressed by LCs after 48 h in culture with or without stimulation with lipopolysaccharide (LPS)."* Several investigators using various methods of purification or staining of LCs have since published comparable findings,' " However, possibly due to different techniques, the demonstration of adhesion molecules on human LCs has not been uniform. We would like to report our results using enrichment of LCs with Fc receptor panning and analysis of surface molecules by flow cytometry and compare our data to the published literature.
Materials and methods LC enrichment
Skin specimens were obtained from adult women undergoing breast reduction. F-pidermal cell suspensions were obtained after overnight trypsinization, using 0 25"^ trypsin (Sigma) in buffer (130 mM Nad, 30 mM HEPES, 10 mM glucose, 3 mM KCl, 1 mM Na2HPO41\U\ pH adjusted to 7 4 with NaOH). Cells were filtered with .i 37 /(m nylon gauze, washed in cold buffer and resiispcnded in RPMI 1640 (Gibco), and counted with trypan blue. Density gradient centrifugation (1 077 Hisio]>;jquc Sigma) allowed pre-enrichment (6-S'\, LC) and hii:hl\ viable (>95'!o) suspensions were obtained ior pannini:. The following steps were adapted from the 1-c paniiinu: technique on sheep erythrocytes described by Rasancn ci al.^- Briefly, poly-i.-lysine treated 100 mm IVlri dishes were covered with a solution of washed sheep er\ilirncytes [Biomerieux, Lyon, I'Vance) (I",, in Dulbecco's phosphate-buffered saline fPIiSI ICiibcoj) 4,^ nm ai room temperature, rinsed three times with I'MS, aiui incu
LEUCOCYTE ADHESION MOLECULES AND LANGERHANS CELLS 45 mn at 37"C with an IgG rabbit-anti-sheep erythrocyte antiserum (Cappel, West Chester, Pennsylvannia, USA) 1 : 500 in PBS. After three washes, a maximum of 20 x 10" epidermal cell suspensions in cold RPMI 1640 were added to each dish and centrifuged 200 g for 5 min and left at 4 X for 2 h for panning. After aspiration of nonadherent cells, LCs were detached from sheep red blood eells monolayers with an erythroeyte lysing medium, washed twice in RPMI, and counted.
241
Results Cell separation and purity of LC suspensions
In a series of eight experiments included in this study, a mean of 3 3x10** epidermal cells were obtained per experiment (range 1-8-7-5 x 10") with a 75± 15% trypan blue exclusion. After panning, an average 7 5 x 10' cells were obtained per experiment (range 2-8 x 10'-l-8x 10'') with a 90 ± 5% trypan blue exclusion. Purity was checked using fluorescence microscopy (counts of a total of 500 cells using phase contrast and fluorescence in the same Monoclonal antibodies andflowcytometry field with a 20 X objective) and flow cytometry using an Table 1 shows the antibodies used in this study at a final anti CDla FITC antibody (OKT6 Ortho). As shown by dilution of I : 20. For two step immunofluorescence, a flow cytometry in Figure 1 in a representative experiGAM-FITC F (ab')2 conjugate (NEN Research Pro- ment, about 75% cells were CDla positive using both ducts) was used at 1 : 100, For flow cytometric analysis, methods. 0 5-1x10' stained cells were resuspended in 100 /il 200 400 1000 Dulbecco's PBS with 0 01% sodium azide and analysed 600 800 r [I 1—I—1—1—I—I [ I—I—r—1—I—I— with a FACScan (Becton-Dickinson) equipped with a Hewlett Packard HP 30 computer.
Table L Monoclonal antibodies used in the study
Antibody BB3t
0KT6-FITC a-HLA-DRJ BU 15 (lOM lie) MHM24II MOlq 441! 60'3II MHM23«rm// of 19, Marlin SD, Springer TA, Purified intercellular adhesion molInvestigative Dermatology 1989; 93: (lO 69. ecule-1 (lCAM-1) is a ligand for lymphocyte function-associated (i, De Panfilis G, Manara GC, Ferrari C, Torresani C^. Adhesion antigen-1 (LI"A-1) Cell 1988; 51: 813-819, molecules on the plasma membrane of epidermal cells, H, Tbe intercellular adhesion molecule 1 is constitutively present on the 20, Bierer BF, Barbosa J, Herrmann S, Burakofi SJ, Interaction of CD2 with its ligand, LFA3, in human T cell proliferation. Journa/ cell surface of human resting Langerbans cells. Journal of of Immunology 1988; 140: 3358-3363. Investigative Dermatology 1990; 94: 317 321. 21, Pattaroyo M, Makgoba MW. Leukocyte adhesion to cells: 7, Romani N, Lenz A, GIa.s.sel H, Stossel H, Stanzl U, Majdic O, molecular basis, physiological relevance, and abnormalities. Fritscb P, Scbuler G. Cultured buman I .angerhans cells resemble Scandinavian Journal of Immunology 1989; 30: 129-164. tymphoid dcntritic cells in pbenolypc and function. Journal of 22, Witmer-Pack M, Oliver W, Valinsky J, Schuler G, Steinman R, investigative Dermatology 1989; 93: 600 6(19. Granulocyte/macrophage colony-stimulating factor is essential S, Teunissen M, WormmeesterJ, Krieg SR, Peters PJ, Vogeis I.MC, for the viability and fonction of cultured murine epidermal Kapsenberg MI., lJos JD, Human epidermal Langerhans cells. Langerhans cells. Journal of Experimental Medicine 1987- 166' L ndergo profound morpbologic and phenotypical changes during 1484-1498, in vitro culture. Journal of Investigative Dermatology, ]99{); 94: 23, I leuflcr C:, Koch F, Schuler G. Granulocyte/macrophage colony 166-173. siiniLilating factor and interleukin I mediate the maturation of 9, Staquet MJ, Dezutter-Dambuyant C, Schmitt D, Tbivolet J, murine epidermal Langerhans cells into potent immuno stimulaAdhesion mediating molecules of human epidermal Langerhans tory dendritic cells. Journal of Experimental Medicine 1988; 167cdh. Journal of Invesiigalive Dermatology 1989; 92: S22. 700- 705, 10, Weber-Matthiescn K, Sterry \V, Organization of the monocyte24, Kampgen F, Heufler C, Koch F, Scbuler G, High numbers of macrophage system of normal human skin. Journal of inve.Uigative GM-CSF receptors arc expressed on epidermal Langerhans cells Dermatology 1990; 93: 83 -89. and lymphoid dendritic cells, and mediate tbe induction/increase 11, Simon J(^, Ouz PD, Iigelaar Re, Sontheimer RD, Bergstresscr of accessory cell function by GM-CSI', Archives ofDermatolomcal PR. AdbesionmoIeculesCDlla, CD18, andlCAM-I on human Research 1990, 281: 546 (Abstract). epidermal Langerhans cells serve a functional role in tbe activation of alloreactive T ce\h. Journal of Investigative Dermatology 1991; 96: 148-151,
