dendritic cell protocols

Humana Press Humana Press M E T H O D S I N M O L E C U L A R M E D I C I N E TM Dendritic Cell Protocols Edited by Stephen P. Robinson, MD, PhD Andrew J. Stagg, PhD Dendritic Cell Protocols Edited by Stephen P. Robinson, MD, PhD Andrew J. Stagg, PhD DC from Mouse Lymph Nodes 3 1 3 From: Methods in Molecular Medicine, vol. 64: Dendritic Cell Protocols Edited by: S. P. Robinson and A. J. Stagg © 2001 Humana Press Inc., Totowa, NJ Isolation of Dendritic Cells from Mouse Lymph Nodes Dmitry Gabrilovich 1. Introduction Lymph nodes are the primary sites of T-cell stimulation by dendritic cells (DC). After contact with antigens, DCs migrate to draining lymph nodes from the skin and other tissues (1–3). Investigation of the morphology and function of lymph node DCs may provide important information about the role of these cells in normal and pathological conditions. Therefore, lymph nodes are popu- lar sites for the isolation of dendritic cells. Dendritic cells isolated from lymph nodes represent “interdigitating” DCs that are localized in T-dependent regions of lymph nodes. DCs represent about 1% of the total population of lymph node cells. Therefore, in order to perform almost any functional tests, the DC fraction should be enriched. The most practical way to enrich the DC fraction is to use a density gradient. Several gradients—metrizamide (4), Nycodenz (5), and Percoll (6)—have successfully been used for enrichment of DCs obtained from different sources. When isolating DC from lymph nodes, density gradient sepa- ration produces a population of DC with a purity of 40–50%. Most contaminat- ing cells are lymphocytes with a small fraction (usually less than 5%) of macrophages. The choice of lymph nodes is dependent on the purpose of the experiment. The most commonly used lymph nodes are axillary, inguinal, and popliteal. DCs can be further enriched using monoclonal antibodies and flow cytometric cell sorting, magnetic beads separation, panning, or cytotoxic elimi- nation with complement. All these methods are based on the negative selection of DCs using anti-T, anti-B, and anti-macrophage antibodies. Since the first step of isolation involves gradient centrifugation, granulocyte contamination is negligible and further purification steps do not require use of anti-granulo- cyte antibodies. 4 Gabrilovich 2. Materials 1. Sterile dissecting forceps and scissors for lymph node extraction. 2. Sterile 6-well plates, 50-mL conical tubes, and 15-mL conical tubes (Falcon, Becton Dickinson, Franklin Lakes, NJ). Sterile 70-µm cell strainers (Falcon), 5 mL syringes, and 5 mL and 10 mL pipets. 3. Fetal calf serum (FCS) (HyClone, Logan, UT). Culture media DMEM and RPMI 1640 (Gibco-BRL, Grand Island, NY) supplemented with antibiotics. We com- monly use a combination of penicillin, streptomycin, and Fungizone (antibiotic– antimycotic, Gibco-BRL). DMEM can be used without serum. RPMI 1640 should be supplemented with 10% FCS (RPMI-FCS). 4. Metrizamide gradient. Dissolve 7.25 g metrizamide with 45 mL RPMI in a 50-mL tube. It usually takes 15–20 min. Sterilize the gradient by passing through a 0.45 µm filter. Add 5 mL of FCS, mix and prepare 2.5 mL aliquots. Store the gradient at –30 ° C. We use metrizamide produced by Nygaard, Norway. Metrizamide is also produced by Sigma, and we have had satisfactory results with Sigma’s metrizamide. 5. Hemocytometer, and microscope with 400× magnification. 6. PE conjugated anti-B7-2 (CD86) antibody, FITC conjugated anti-CD11c (N418) antibody, and PE and FITC conjugated mouse IgG2a and IgG2b as isotype con- trol (Pharmingen). 3. Methods 3.1. Isolation of DC by Density Gradient Centrifugation 1. Place a cell strainer into one well of a 6-well plate. Fill the well with DMEM. Prepare as many wells as necessary. 2. Sacrifice mice using one of the methods approved by the appropriate institutional review board. Extract the lymph nodes from at least three mice and put the lymph nodes together on the cell strainer submerged in DMEM. Make sure that the lymph nodes are covered with the medium. 3. Remove the plunger from a syringe and use it to press the lymph nodes through the mesh of the cell strainer. Make sure that lymph nodes are completely shat- tered on the mesh. Finally, wash the strainer with 3–4 mL of DMEM. 4. Discard cell strainers and collect cells from the well into a 15 mL conical tube. Wash cells once with DMEM by centrifuging for 5 min at 300g. Resuspend cells in 10 mL of RPMI-FCS. 5. Thaw the metrizamide gradient and transfer 2.5 mL into a sterile 15 mL tube. Overlay cells slowly onto the gradient. It is important not to disturb the gradient. Spin the gradients for 10 min at 500g at room temperature with no break on the centrifuge. 6. Collect cells from the interface using a 5 mL pipet and wash them once with RPMI-FCS. Resuspend cells in 1 mL of RPMI-FCS. 7. Count cells on the hemocytometer at magnification 400×. DCs may be clearly iden- tified by their distinct morphology. This DC enriched fraction can now be used in DC from Mouse Lymph Nodes 5 morphological or functional studies (see Notes 1 and 2). To confirm the purity of the samples, cells can be analyzed by flow cytometry (see Subheading 3.3.). 3.2. Purification of DC by Panning If a higher purity of DCs is desired, DCs can be further enriched by panning (see Note 3). 1. Prepare a 6-well plate for panning by coating separate wells with either anti- mouse immunoglobulin or anti-rat immunoglobulin. We usually use goat anti- mouse immunoglobulin and goat anti-rat immunoglobulin from Sigma, St. Louis, MO. Add 3 mL of each antibody at a concentration of 1 mg/mL in phosphate- buffered saline (PBS) to separate wells. 2. After at least 60-min incubation at room temperature, remove the antibody solu- tion and wash the wells four to five times with PBS (see Note 4). 3. Wash the DC fraction derived by density gradient centrifugation (see Subhead- ing 3.1.) once in PBS and resuspend in 100 µL of hybridoma supernatants of anti-CD4 antibody (L3T4, TIB-207, ATCC, Rockville, MD), 100 µL of hybri- doma supernatants of anti-CD8 antibody (Lyt-2.2, TIB-210, ATCC), and 20 µL of anti-F4/80 antibody (Serotec, Raleigh, NC) (see Note 5). 4. After a 25-min incubation on ice, wash cells twice and resuspend in 3 mL of PBS containing 0.1% mouse serum. 5. Transfer the cell suspension to a well coated with anti-rat immunoglobulin. Incu- bate cells on the plates for 60–90 min at 4 ° C. 6. Harvest the nonadherent DC enriched fraction using a 5 mL pipete. Gently wash the well with ice cold PBS to remove any partially adherent cells and add these to the nonadherent fraction (see Note 5). 7. Concentrate the harvested cells by centrifuging at 300g. Resuspend the cell pel- let in 3 mL of cold PBS with 0.1% serum and transfer into a well coated with anti-mouse immunoglobulin. Incubate for another 60-90 min and then collect cells as described previously. Cells can now be resuspended in RPMI-FCS and used for further study (see Note 5). 3.3. Analysis of DC Purity by Flow Cytometry We routinely use double labeling with anti-CD11c (N418) antibody and anti- B7-2 (CD86) antibody to identify DC as CD11c + CD86 + cells on the flow cytometer. 1. Transfer 100 µL of the purified DC cell suspension into two tubes for flow cytometry labeled “test” and “control.” 2. Wash cells once with PBS and resuspend in 100 µL of PBS. 3. Add 5 µL of anti-B7-2 and anti-CD11c antibodies into the “test” tube and 5 µL of isotype control antibodies into the “control” tube. 4. Incubate on ice for 25 min, then wash twice with 2 mL of cold PBS and analyze on the flow cytometer (see Note 6). 6 Gabrilovich 4. Notes 1. Generally, one can expect to isolate around 5–10×10 3 DCs from one lymph node. For functional tests, we usually collect lymph nodes from three or four mice, which pro- vides a sufficient number of cells (around 2×10 5 ) for several functional tests. 2. It is recommended that 10 –5 M of `-mercaptoethanol is added to the culture medium while growing mouse cells. However, `-mercaptoethanol should be added after completion of all isolation procedures, since even in low concentra- tion it may affect the binding of antibodies to the cells. 3. Owing to the relatively low number of cells, lymph nodes are not the best source of highly purified DCs. However, if the experimental goal makes it necessary to use highly purified lymph node DCs, we suggest using the panning technique as opposed to complement-dependent cytotoxicity. The latter method usually results in the nonspecific loss of some DC. If the investigator has access to a flow cell sorter, this method, as well as the magnetic bead separation technique, may pro- vide good alternatives. For all sorting procedures, the author recommends using at least six mice per sample. 4. Panning plates may be stored overnight at 4 ° C. Collected antibody solution can be used again several times. 5. Anti-Thy-1.2 antibody produced by hybridoma supernatant (ATTC TIB 107) can be used instead of anti-CD4 and anti-CD8 antibodies. All these antibodies are rat Ig2 b . In the first panning step, T cells and macrophages are eliminated using anti- rat monoclonal antibodies. In the second step, B cells are removed using anti- mouse immunoglobulin antibody. 6. It is important to perform all procedures at 4 ° C, because DCs readily adhere to plastic and some cells may be lost if the incubation is performed at room tem- perature. The purity of DCs can be verified using anti-CD11c and B7-2 antibod- ies. The final purity of the DC fraction is usually above 95% following panning. One can expect to obtain at least 10 5 highly purified DCs from five or six mice. 7. A typical example of double labeling of lymph node DCs with CD11c and B7-2 antibodies is shown Fig. 1. Cells were isolated and labeled with antibodies as described in “Methods.” Analysis was performed using FACSCalibur flow cytometer (Becton Dickinson, Mountain View, CA). It is important to note that there is no single marker that would allow for detection of 100% of lymph node DCs. N418 (CD11c) may bind to some macrophages, whereas B7-2 binds prima- rily to only mature DCs. The investigator may choose to use a combination of other DC markers depending on the goal of the study. References 1. Steinman, R. M. (1991) The dendritic cell system and its role in immunogenicity. Annu. Rev. Immunol. 9, 271–296. 2. Knight, S. C., and Stagg, A. J. (1993) Antigen-presenting cell types. Curr. Opin. Immunol. 5, 374–382. 3. Hart, D. N. J. (1997) Dendritic cells: Unique leukocyte populations which control the primary immune response. Blood 90, 3245–3287. DC from Mouse Lymph Nodes 7 4. Knight, S. C., Farrant, J., Bryant, A., et al. (1986) Non-adherent, low density cells from human peripheral bloodcontain dendritic cells and monocytes, both with veiled morphology. Immunology 57, 595–598. 5. McLellan, A. D., Starling,G. C., and Hart, D. N. C. (1995) Isolation of human blood dendritic cells by Nycodenz discontinuous gradient centrifugation. J. Immunol. Methods 184, 81–85. 6. Young, J. W. and Steinman, R. M. (1988) Accessory cell requirements for the MLR and polyclonal mitogens, as studied with a new technique for enriching blood dendritic cells. Cell Immunol 111, 167–171. Fig. 1. Flow cytometry demonstrates the purity of DC isolated from lymph nodes following density gradient centrifugation. DC from Mouse Spleen 9 2 9 From: Methods in Molecular Medicine, vol. 64: Dendritic Cell Protocols Edited by: S. P. Robinson and A. J. Stagg © 2001 Humana Press Inc., Totowa, NJ Isolation of Mouse Spleen Dendritic Cells Andrew J. Stagg, Fiona Burke, Suzanne Hill, and Stella C. Knight 1. Introduction It is now over 20 years since dendritic cells (DC) were first identified in and isolated from the spleens of mice (1,2) and they continue to be a much-studied population. Only a small proportion of spleen cells are DC, but the large size of the organ means that useful numbers of DC can still be purified. In recent years the ability to grow cells with the phenotypic and functional properties of DC from bone marrow progenitors has opened new avenues of research. However, the relationship of cells grown in this way to DC populations in vivo is unknown and the need remains to study DC present in tissues. Spleen DC are heterogeneous with differences in phenotype, function, and microanatomical location (3,4). At least two major subsets are recognized, and these can be discriminated on the basis of the presence or absence of a cell- surface __ homodimer of the CD8 molecule. The freshly isolated CD8__ + population is DEC-205 + , CD24 + , CD11b - , 33D1 - , CD4 - , whereas the CD8__ - subset is DEC205 - , CD24 - , CD11b + , 33D1 + , CD4 - . Both subsets express CD11c, and this marker appears to be expressed selectively on DC and in the mouse can be used as a pan-DC marker. The CD8__ + population predomi- nately localizes in the T-cell areas of the white pulp and corresponds to inter- digitating cells. In the steady state, the CD8__ - population is probably localized predominately in the marginal zone, between the red and white pulp, but mobi- lizes into the T-cell areas in response to lipopolysaccharide (LPS) administra- tion (5). This marginal zone DC population has a higher phagocytic activity and turnover rate than the interdigitaing cells (6). The CD8__ + and CD8__ - populations may be cells of lymphoid and myeloid lineages, respectively. They 10 Stagg et al. can both activate resting T cells, but may stimulate different types of responses. The CD8__ + population has been reported to drive preferentially Th1 responses, whereas presentation of antigen by the CD8__ - may favor Th2 responses (7). CD8__ + DC can also kill activated T cells via Fas-mediated apoptosis (8). The division of spleen DC into “lymphoid” and “myeloid” popu- lations is probably an oversimplification, and recent evidence suggests further heterogeneity with the description of a third, CD4 + , spleen DC subset (9). There are many published protocols for isolating mouse spleen cells and in choosing among these methods two factors should be borne in mind. First, different methods may favor the recovery of particular DC subsets at the expense of others. This can be a problem if the intention is to recover a repre- sentative sample of the total spleen DC, but it can also be turned to the investigator’s advantage in the purification of particular subsets. Second, DC may be altered phenotypically or functionally by the isolation process itself. This modulation occurs in methods in which DC are cultured for prolonged periods, because in vitro culture is sufficient to induce DC maturation. Changes in properties of DC may also occur during positive selection with monoclonal antibodies or digestion of tissue with proteolytic enzymes. For instance, colla- genase preparations are likely to contain significant concentrations of endot- oxin that may affect DC. In this chapter we describe a basic method for the enrichment of mouse spleen DC that involves overnight culture and separation on hypertonic metrizamide gradients and provide a suggested protocol for the further purifi- cation of these cells. We also describe an alternative method for spleen DC that avoids the need for culture and discuss how the choice of method for initial preparation of a spleen cell suspension can be used to influence the recovery of particular DC subsets. 2. Materials 1. Specific pathogen free mice. The commonly used strains in our laboratory are BALB/c, CBA, and C3H. We have used mice of either sex, and they are usually aged 6–12 wk. 2. Dissecting board or paper tissues. 3. 70% ethanol. 4. Sterile surgical instruments (forceps and scissors). 5. Complete medium: Dutch modification of RPMI-1640 (Sigma; cat. no.R-7638) supplemented with 10% heat-inactivated fetal calf serum (FCS), 2 mM L-gluta- mine,100U/mL penicillin/streptomycin, and 5×10 –5 M 2-mercaptoethanol (2-ME) (see Note 1). 6. HEPES-buffered RPMI-1640 (Sigma, cat. no. R-5886) 7. Metal cell strainers. 8. 60 mm Petri dishes (Nunc or Sterilin). DC from Mouse Spleen 11 9. 2 mL and 1 mL syringes (Terumo). 10. 10 mL conical-bottomed tubes (Sterilin; cat. no. 144AS). 11. Disposable Pasteur pipets (Alpha Labs; cat. no. LW4005) or sterilized glass equivalents. 12. Small filters for sterilization (GelmanSciences; cat. no. 6224184 [0.45mm] or cat. no. 6224192 [0.22 µm]). 13. Collagenase digestion mix: 1mg/mL collagenase D (Roche Molecular Products; cat. no. 1088 866), 20 µg/mL DNase I (Roche Molecular Products; cat. no. 1284 982), 2% FCS in HEPES-buffered RPMI-1640 (see Notes 2 and 3). To prepare collagenase stock: a. Dissolve 500 mg of collagenase D in 50 mL serum-free HEPES buffered RPMI-1640 (10 mg/mL). b. Filter sterilize (0.45 µm). c. Store in aliquots at –20°C. d. Avoid repeated freezing and thawing. e. Thaw aliquots as required and keep on ice until used. To prepare DNase I stock: a. Dissolve 100 mg in 10 mL dH 2 0 (10 mg/mL). b. Filter sterilize. c. Store in aliquots at –20°C. d. Avoid repeated freezing and thawing. To make 10 mL of digestion mix combine: a. 1 mL collagenase D stock. b. 20 µL DNase I stock. c. 0.2 mL FCS d. 8.8ml HEPES buffered RPMI-1640 Keep on ice until use. 14. 26G × 1/2 in. needles. 15. Disposable scalpels or scalpel blades. 16. T25 tissue culture flasks (Falcon; cat. no.353014) (see Note 4). 17. Cell scrapers (Falcon; cat. no. 3085). 18. Analytical grade metrizamide (Nycomed; cat. no.22.20.10) (see Note 5). 19. Sterile 5 mL (75 mm × 12 mm) push cap round bottomed tubes (Sarstedt; cat. no. 55.476.013). 20. MiniMACS buffer: PBS containing 5% bovine serum albumin (BSA) and 5 mM EDTA. Filter sterilize. Handle carefully to avoid frothing (see Note 6). 21. Heat-inactivated normal mouse serum. 22. Monoclonal antibodies and immunomagnetic microbeads (see Note 7). These include: a. “Fc-Block” (PharMingen; cat. no. 01241A/D) (see Note 8). b. Microbeads coated with anti-CD11c (N418) (Miltenyi; cat. no.520-01). c. Anti-CD11c-FITC (clone HL3) (PharMingen; cat. no.09704A/D). For some applications the same antibody labeled with an alternative fluorochrome (e.g., phycoerythrin) may also be required. 12 Stagg et al. d. Anti-CD45R-FITC (B220) (PharMingen; cat. no.01124A/D). e. Microbeads coated with anti-FITC (Miltenyi; cat. no.487-01). 23. MiniMACS magnet and holder or the varioMACS system (Miltenyi) (see Note 9) 24. MiniMACS columns (Type MS + /RS + for miniMACS or varioMACS and/or Type LS + /VS + for varioMACS, Miltenyi) (see Note 9). 3. Methods 3.1. Preparation of Single Cell Suspensions from Spleens This section describes removal of the mouse spleen and presents three dif- ferent methods for producing a single-cell suspension from the organ. The way in which the choice of methods influences the recovery of DC is discussed. 3.1.1. Removal of the Spleen 1. Kill the mouse by cervical dislocation. 2. Lay mouse on dissecting board, “left side” uppermost. 3. Surface-sterilize the skin using 70% ethanol or a proprietary compound. 4. Using one set of sterile surgical instruments (forceps and scissors), cut through the skin just below the ribcage and visualize the spleen. 5. Using a second, smaller, set of instruments, remove the spleen, trimming away any fatty tissue. 6. Place spleen into complete medium at room temperature (see Note 10). Spleens from multiple animals can be pooled. 3.1.2. Preparation of a Single-Cell Suspension using a Metal Sieve This has been our routine method for many years. It avoids the use of pro- teolytic enzymes, gives good recovery of DC numbers, and, in conjunction with overnight culture and metrizamide separation, yields a mixture of CD8__ + and CD8__ - DC (see Fig. 1). 1. Strain spleens by pouring through a sterile metal cell-strainer. Discard medium. 2. Place the strainer containing spleens into a 60 mm Petri dish and add a few milli- liters of fresh medium. 3. Using the barrel from a 2 mL syringe, press the spleens through the strainer. Continue until only a little fibrous tissue remains in the strainer. 4. Remove the strainer and place in upturned lid of the Petri dish. 5. Reinsert plunger into syringe barrel and use to transfer spleen cell suspension to a 10 mL conical tube. (A larger tube or replicate tubes will be required for mul- tiple spleens.) 6. Using fresh medium and a Pasteur pipet rinse the Petri dish and the cell strainer to ensure that all cells have been recovered. Pool with the rest of the spleen cell suspension. 7. Top-up tube with complete medium to appropriate volume (see below). [...]... (1989) The cell surface of mouse dendritic cells: FACS analyses of dendritic cells from different tissues including thymus Cell Immunol 118, 108–125 5 De Smedt, T., Pajak B., Muraille, E., et al (1996) Regulation of dendritic cell numbers and maturation by lipopolysaccharide in vivo J Exp Med 184, 1413–1424 6 Leenen, P J M., Radosevic, K., Voerman, J S A., et al (1998) Heterogeneity of mouse spleen dendritic. .. immature T-lineage cells and thymic B cells The HDCF represents 15–20% of the VLDF and contains 70–80% of N418+ DCs, the remaining N418- cells being essentially immature T-lineage cells Excluding DCs, this cell fraction is therefore devoid of other antigen presenting cells, such as macrophages or B cells and could be used as an enrichedthymic DC fraction for certain purposes Finally, N418+ cells represent... absence of plasma cells The cells do not phagocytose opsonized sheep red blood cells indicating that they are not macrophages It should also be noted that only a small proportion of the cells shows the classic dendritic morphology immediately after isolation However, after overnight culture of these cells, 95–99% of the cells will express high levels of surface MHC class II, and the dendritic processes... be encountered are : Cell loss: Since the frequency of L-DC is very low and the cell separation procedure is rather complex, poor cell yields potentially are a major problem During every step of the cell separation procedures, significant cell loss is likely to occur This can be caused by multiple factors including incorrect cell loading and washing, cell adhesion or aggregation, cell trapping in the... dendritic cells: in vivo phagocytic activity, expression of macrophage markers, and subpopulation turnover J Immunol 160, 2166–2173 7 Maldonado, L., De Smedt, T., and Michel, P., (1999) CD8alpha+ and CD8alpha– subclasses of dendritic cells direct the development of distinct T helper cells in vivo J Exp Med 189, 587–592 8 Suss, G and Shortman, K (1996) A subclass of dendritic cells kills CD4 T cells via... subpopulation of dendritic cells J Exp.Med 176, 47–58 2 Ardavin, C., Waanders, G., Ferrero, I., Anjuere, F., Acha-Orbea, H., and MacDonald, H R (1996) Expression and presentation of endogenous mouse mammary tumor virus superantigens by thymic and splenic dendritic cells and B cells J Immunol 157, 2789–2794 3 Ardavin,C., Wu, L., Li, C.-L., and Shortman, K (1993) Thymic dendritic cells and T cells develop simultaneously... 761–763 4 Ardavin, C (1997) Thymic dendritic cells Immunol Today 18, 350-361 5 Wu, L., Vremec, D., Ardavin, C., Winkel, K., Suss, G., Georgiou, H., Maraskovsky, E., Cook, W., and Shortman, K (1995) Mouse thymus dendritic cells: kinetics of development and changes in surface markers during maturation Eur J Immunol 25, 418–425 Rat Spleen and Lymph DC 29 4 Isolation of Dendritic Cells from Rat Intestinal Lymph... Hochrein, H., Wu, L., and Shortman, K (2000) CD4 and CD8 expression by dendritic cell subtypes in mouse thymus and spleen J Immunol 164, 2978–2986 Thymic DC isolation 23 3 Isolation of Mouse Thymic Dendritic Cells Fabienne Anjuère and Carlos Ardavín 1 Introduction The method described in this chapter for the isolation of mouse thymic dendritic cells (DC) is an optimization of our previously published methods... cytometry) of the cell fractions obtained: after enzymatic digestion and centrifugation in Optiprep 1.055 (VLDF); after depletion with magnetic beads (HDCF); after magnetic cell sorting of N418+ cells (purified DCs) The VLDF represents 0.2-0.4% of total thymocytes and contains around 20% N418+ cells Among N418+ cells of the VLDF around 80% are DCs and 20% are Mac-1+ F4/80+ thymic macrophages N418- cells in... contents through a cell strainer as described previously 12 Pool cells with those already on ice Rinse dish in digestion mix or medium and again pool with the other cells 13 Spin down (350g, 5 min), discard supernatant and gently resuspend cell pellet in 10 mL complete medium 14 Spin down again and resuspend in complete medium Adjust to required volume (see below) 3.1.4 Preparation of a Single -Cell Suspension . D S I N M O L E C U L A R M E D I C I N E TM Dendritic Cell Protocols Edited by Stephen P. Robinson, MD, PhD Andrew J. Stagg, PhD Dendritic Cell Protocols Edited by Stephen P. Robinson, MD, PhD Andrew. M. D., and Steinman, R. M. (1989) The cell surface of mouse dendritic cells: FACS analyses of dendritic cells from dif- ferent tissues including thymus. Cell. Immunol. 118, 108–125. 5. De Smedt,. CD8alpha– subclasses of dendritic cells direct the development of distinct T helper cells in vivo. J. Exp. Med. 189, 587–592. 8. Suss, G. and Shortman, K. (1996) A subclass of dendritic cells kills CD4 T cells via