The mode of protein antigen administration determines preferential presentation of peptide-class II complexes by lymph node dendritic or B cells

iim$$$6053

International Immunology, Vol. 9, No. 1, pp. 9–15

© 1997 Oxford University Press

The mode of protein antigen administration determines preferential presentation of peptide–class II complexes by lymph node dendritic or B cells Jean-Charles Gue´ry1, Francesco Ria2, Francesca Galbiati, Simona Smiroldo and Luciano Adorini Roche Milano Ricerche, Via Olgettina, 58, 20132 Milano, Italy 1Present 2Present

address: INSERM U28, Hoˆpital Purpan, 31059 Toulouse, France address: Istituto di Patologia Generale, Universita` Cattolica del Sacro Cuore, 00168 Roma, Italy

Keywords: antigen presentation, lysozyme Abstract We have compared the capacity of dendritic cells (DC) and B cells to present peptide–class II complexes following administration of protein in adjuvant or in soluble form. Three different antigen-presenting cell (APC) populations were separated from draining lymph node cells from mice immunized s.c. with hen egg-white lysozyme (HEL) in adjuvant or with adjuvant only followed by soluble HEL: DC (N418F, class IIF, B220–, low buoyant density), large B cells (B220F, low buoyant density) and small B cells (B220F, high buoyant density). HEL peptide–class II complexes displayed by these APC were evaluated by their capacity to activate HEL-specific T hybridoma cells. Following immunization with HEL in adjuvant, DC are the only lymph node APC population expressing detectable HEL peptide–class II complexes. Conversely, after i.v. administration of soluble HEL in mice previously injected with adjuvant only, lymph node B cells are much more efficient than DC in presenting peptide–class II complexes to T cells. Therefore, different modes of protein antigen administration lead to selective expression of antigenic complexes by different APC populations. These data correlate with the observation that, unlike B cells, DC recruited in lymph nodes of mice injected with adjuvant only present in vitro processed protein antigen much less efficiently than synthetic peptides, probably as a consequence of their maturation in vivo. Introduction The activation of CD41 T cells is initiated by the recognition of peptide–class II MHC complexes on the surface of antigenpresenting cells (APC) like macrophages, dendritic cells (DC) and B cells (1,2). Among the different APC populations capable of presenting peptide–class II MHC complexes, DC and B cells have been studied extensively, but their relative role in the presentation of protein antigen and CD41 T cell priming in vivo is still controversial. Using mice lacking B cells, presentation to CD41 T cells of antigenic complexes derived from processing of protein antigen administered in adjuvant was found in some studies to require B cells (3), whereas in others B cells were found not critical and antigenic complexes were presumably presented by DC (4). Similarly, administration of soluble protein to normal mice was found to

lead to selective expression of antigenic complexes either by antigen-specific B cells (3) or by DC (5). DC represent a system of highly specialized APC present, in different stages of maturation, in the circulation as well as in lymphoid and non-lymphoid organs (6,7). Immature DC, such as Langerhans cells in the skin, are found in nonlymphoid tissues, where they exert a sentinel function. After antigen uptake, they migrate through the afferent lymph to Tdependent areas of lymphoid organs where priming of naive T cells may occur (6,8). Based on in vitro data, it has been hypothesized that during this process they mature into potent APC by increasing their immunostimulatory properties while losing their antigen capturing capacity (9,10). Thus, mature DC should have in vivo an impaired capacity to endocytose

Correspondence to: L. Adorini Transmitting editor: G. Doria

Received 26 February 1996, accepted 17 September 1996

10 Relative role of DC and B cells in antigen presentation in vivo protein antigens but will present very efficiently peptides derived from proteins which have been previously endocytosed at the site of inflammation. To address this point, we have compared the relative capacity of DC and B cells recruited in lymph nodes during the inflammatory response induced by adjuvant administration to present protein antigen administered in different forms. We have previously shown that immune lymph node cells (LNC) from mice immunized with hen egg-white lysozyme (HEL) in adjuvant display HEL peptide–MHC class II complexes able to stimulate, in the absence of any further antigen addition, specific T hybridoma cells (11,12). In the present paper, HEL-specific T hybridoma cells were used to read-out expression by DC and B cells of antigenic complexes derived from processing of native HEL, either given s.c. in adjuvant or in soluble form i.v. Phenotypic analysis of the stimulatory APC in immune lymph nodes confirms our previous study demonstrating that following s.c. administration of HEL in adjuvant DC are the only APC expressing detectable HEL peptide–class II complexes (13). Conversely, when HEL is administered in soluble form i.v. to mice previously injected with adjuvant only, lymph node B cells are much more efficient than DC in the presentation of HEL peptides. These results demonstrate that protein antigen injected in soluble form is presented best by B cells, whereas the same protein is presented only by lymph node DC when administered in adjuvant. Therefore, different protocols of protein antigen administration lead to expression of peptide– class II complexes by different APC. These results also suggest maturation of DC in vivo. Methods

Mice, antigens and immunizations Female BALB/c mice (2–3 months old; Charles River, Calco, Italy) were used. HEL, recrystallized three times, and ovalbumin (OVA) grade were obtained from Sigma (St Louis, MO). HEL peptide sequence 105–120 (MNWVAWRNRCKGTDV) was purchased from Neosystem (Strasbourg, France). Mice were immunized s.c. into the hind footpads with the indicated amount of HEL dissolved in PBS and emulsified vol/vol in incomplete Freund’s adjuvant (IFA) (Difco, Detroit, MI). In addition, mice were injected with 1 nmole OVA-IFA in to the hind footpads and 5 days later received 100 nmols soluble HEL i.v. 3 h before sacrifice.

Enrichment of APC populations Immune LNC were depleted of T cells by cytotoxic elimination with HO-13-4 anti-Thy-1.2 mAb (TIB 99) followed by rabbit complement (Low-Tox M; Cedarlane, London, Canada). Low and high buoyant density APC were prepared from T celldepleted LNC by centrifugation over a discontinuous Percoll (Pharmacia LKB, Uppsala, Sweden) gradient containing 55– 60% and 70% layers. Cells at the medium/55–60% and 60/ 70% interface were collected separately and referred to as low- and high-density APC respectively. B cells in the lowdensity population were further depleted by magnetic cell sorting using the MiniMACS magnetic separation system following the manufacturer’s instructions (Miltenyi Biotec,

Bergisch Gladbach, Germany). Cells were incubated with B220-coated microbeads, and then separated into B2201 and B220– fractions on MiniMACS separation columns.

Flow cytometry Cells were double stained by incubating them with optimal concentrations of FITC–14.4.4S (anti-I-E) mAb, biotin-conjugated N418 (anti-CD11c) or 6B2 (anti-B220) mAb for 30 min at 4°C in PBS containing 5% FCS, 0.1% sodium azide and 1% normal rat serum to inhibit binding to FcR. 14.4.4S and 6B2 mAb were purchased from PharMingen (San Diego, CA). N418 is an hamster mAb specific for mouse splenic DC recognizing the p150/90 leukocyte integrin, likely the mouse CD11c molecule (14,15), and was obtained from the ATCC (HB-224; Rockville, MD). Biotinylated mAb were revealed using phycoerythrin–streptavidin (Southern Biotechnology Associates, Birmingham, AL). Analysis was performed on a Becton Dickinson FACScan flow cytometer (Becton Dickinson, Mountain View, CA). Data were collected on 5000–10,000 viable cells as determined by forward light scatter intensity and propidium iodide exclusion, and analyzed using Lysys II software.

Assay for antigen-presenting activity The antigen-presenting activity of lymph node APC was assessed as previously described (11) using the T cell hybridomas 1H11.3 (I-Ed, HEL108–116) (16) and 2G12.1 (I-Ad, β2M26–39) (17). Briefly, mice were immunized into the hind footpads with the indicated amount of HEL emulsified in IFA. Three to ten days after immunization, the draining popliteal lymph nodes were removed and APC enriched as detailed above. In addition, mice were injected with 1 nmole OVA-IFA into the hind footpads and 5 days later received 100 nmoles soluble HEL i.v. 3 h before sacrifice. Lymph node DC, large and small B cells were cultured in duplicate or triplicate at the indicated cell doses with appropriate T cell hybridomas (53104 cells/well) in 96-well culture plates (Costar). Culture medium was RPMI 1640 (Gibco, Basel, Switzerland) supplemented with 2 mM L-glutamine, 50 µM 2-mercaptoethanol, 50 µg/ml gentamicin (Sigma) and 10% FCS (Gibco). After 24 h of culture, 50 µl aliquots of supernatants were transferred to microculture wells containing 104 CTLL cells and, after an additional 24 h incubation, the presence of T cell growth factors, mainly IL-2, was assessed by [3H]thymidine incorporation during the last 5 h of culture. In some experiments, IL-2 concentration was determined using a two-sites sandwich ELISA with paired mAb (JES6–1A12, JES6–5H4) purchased from PharMingen as described elsewhere (13). IL-2 was quantified from two to three titration points using standard curves generated by purified recombinant mouse IL-2 and results expressed as cytokine concentration in pg/ml. The detection limit of the assay was 10 pg/ml. Results

Enrichment of DC and B cells By a combination of Percoll gradient centrifugation and magnetic cell sorting, we separated lymph node APC into three populations: small B cells, large B cells and DC-enriched

Relative role of DC and B cells in antigen presentation in vivo 11

Fig. 1. Enrichment of lymph node DC and B cells. BALB/c mice were immunized with 10 nmol HEL in IFA. Five days later, LNC from five mice were pooled, depleted of T cells, and separated into low- and high-density populations by centrifugation on Percoll. The high-density population contains only B cells and is referred to as small B. Low-density APC were further separated by magnetic cell sorting on a MiniMACS column using B220-conjugated microbeads. Positively and negatively selected populations were referred to as large B and DC-enriched populations respectively. APC were analyzed for cell surface expression of the indicated molecules (closed histograms) as described in Methods. Control stainings (open histograms) represent cells stained without primary mAb.

cells (Fig. 1). These populations were characterized by their expression of MHC class II, B220 and N418 molecules. The high-density population contains .95% B2201, class II1 N418– cells and is referred to as small B cells. Large B cells, obtained from the low-density population by positive selection using B220 mAb-conjugated magnetic particles, are .95% B2201, class II1, N4181. The negatively selected fraction, depleted of B2201 cells, is enriched in class II1, N4181 cells (50–70%) and is referred to as DC-enriched population.

The mode of protein antigen administration determines preferential presentation of peptide–class II complexes by DC or B cells We have recently shown that following administration of HEL in adjuvant antigenic complexes are only detectable on lymph node DC, whereas B cells are devoid of antigen-presenting activity (13). In the present paper, we demonstrate that DC isolated 3 and 6 days after injection of HEL in adjuvant present very efficiently the HEL epitope 108–116 bound to I-Ed molecules whereas its presentation decreases considerably by day 10 after priming. At all time points tested B cells, either large or small, fail to present or present very inefficiently HEL108–116/Ed complexes to T hybridoma cells (Fig. 2) . To determine whether the mode of protein antigen administration

determines preferential presentation of peptide–class II complexes by DC or B cells, we have analysed the capacity of resident lymph node APC to present antigenic complexes following i.v. administration of HEL in mice previously injected with adjuvant. Injection of adjuvant recruits APC in the draining lymph nodes. Five days after administration of soluble HEL in the hind footpads popliteal lymph nodes harvested from nine mice yielded a total of 35 3 106 cells which, after T cell depletion, were reduced to 1.6 3 106. Due to the very low number of cells recovered we could not proceed further with cell enrichment. Cytofluorimetric analysis revealed that this cell popluation contained 4% N4181 cells, which represents a yield of 7 3 103 cells/mouse. The average number of N4181 cells obtained from lymph node cells after injection of adjuvant is 200 3 103 mouse, corresponding to a 30-fold increase (data not shown). BALB/c mice were therefore immunized with an irrelevant antigen (OVA) in IFA to induce cell recruitment in draining lymph nodes. Five days later, soluble HEL (100 nmoles/ mouse) was injected i.v. and popliteal lymph nodes collected 3 h later. Amount and timing of soluble antigen injection were determined in pilot experiments (data not shown) and were found to be in agreement with previous results (3,5,18,19). APC populations were enriched as in Fig. 1 and assessed in

12 Relative role of DC and B cells in antigen presentation in vivo

Fig. 2. Time course of peptide–class II complex expression on lymph node DC and B cells following administration of protein antigen in adjuvant. BALB/c mice were immunized into the hind footpads with 10 nmoles/mouse HEL in IFA 3, 6 and 10 days before sacrifice. LNC from 5 (10 and 6 days) or 10 (3 days) mice were pooled and DC and B cell populations enriched as in Fig. 1. HEL108–116-specific I-Ed-restricted 1H11.3 T hybridoma cells were cultured (5 3 104 cells/well) with graded numbers of enriched APC populations. After 24 h, antigen specific IL-2 production was determined using a two-site sandwich ELISA.

vitro for their antigen-presenting activity. Results in Fig. 3 demonstrate that HEL108–116/Ed complexes derived from processing of HEL administered subcutaneously in IFA are only detectable on the DC-enriched population (Fig. 3A). Conversely, small and large B cells are much more efficient than DC to activate T cell hybridomas when HEL is administered in soluble form i.v. (Fig. 3B). To control for the intrinsic capacity of DC and B cells to activate class II-restricted T cells, we have analyzed presentation of naturally processed self-β2M peptides corresponding to the sequences 26–39 by I-Ad molecules (17). All three APC populations constitutively express endogenously synthesized self-β2M peptide–Ad complexes with the following hierarchy in presenting capacity: DC-enriched . large B cells . small B cells. The intrinsic capacity of these APC populations to present endogenously processed self-β2M peptides is comparable in both groups (Fig. 3C and D), whereas the capacity to present exogenous HEL is clearly different. These data demonstrate that the mode of protein antigen administration determines preferential presentation of peptide–class II complexes by lymph node DC or B cells. Lymph node DC present in vitro to class II-restricted T cells protein antigen much less efficiently than synthetic peptides The previous results could be explained by differences in the accessibility of circulating proteins to these APC populations, or by their different potential in antigen uptake and processing in situ. To address this point we have compared the capacity of B cells and DC freshly isolated from immune lymph nodes to present HEL or HEL peptide 105–120 in vitro (Figs 4 and 5). No qualitative differences are observed for B cells, either small or large, in their capacity to present processed protein or synthetic peptide. Conversely, DC are very efficient in presenting the HEL peptide 105–120 but have an impaired capacity to present this epitope after processing of the native HEL protein. This is demonstrated either by titration of APC

using a fixed concentration of antigen (Fig. 4), or by titration of antigen using a fixed concentration of APC (Fig. 5). In the latter case, 104 DC and 3 3 105 small B cells/well were used to obtain a comparable T cell activation. Small B cells were chosen because they are more frequent than large B cells and more active in antigen presentation after administration of soluble HEL (Fig. 3B). The results indicate that, unlike B cells, DC recruited in draining lymph nodes by s.c. administration of adjuvant have a reduced ability to endocytose and process protein antigens, while they remain extremely efficient in presenting antigenic peptides, likely as a consequence of their maturation in vivo. These data also show that for the presentation of antigenic peptides DC are 10- to 30-fold more potent APC as compared with large and small lymph node B cells, respectively (Figs 4B and 5B). Discussion Data in the present paper demonstrate that different modes of protein antigen administration lead to selective expression of antigenic complexes by different APC (Fig. 6). Following immunization with HEL in adjuvant, DC are the only lymph node APC population expressing detectable HEL peptide– class II complexes, whereas B cells are much more efficient than DC in presenting antigenic complexes derived from processing of circulating soluble HEL. To measure the capacity of resident lymph node APC (DC and B cells) to present peptides derived from processing of a circulating protein antigen, antigen has been injected shortly before lymph node harvesting to minimize loading of APC before their entry into the lymph node. Under these conditions, lymph node B cells are far superior to DC in the presentation of HEL peptide– class II complexes to T hybridoma cells. These data correlate with the impaired capacity of lymph node DC to present in vitro native protein antigen, as compared to synthetic peptide, while no qualitative differences in the presentation

Relative role of DC and B cells in antigen presentation in vivo 13

Fig. 3. In vivo formation of antigenic complexes on lymph node DC and B cells following administration of protein antigen in adjuvant or in soluble form. BALB/c mice were immunized into the hind footpads with 10 nmol per mouse HEL in IFA (A and C) or 1 nmol per mouse OVA in IFA (B and D). OVA-IFA primed mice were injected with soluble HEL (100 nmol per mouse) 3 h before removal of popliteal lymph nodes. LNC from five mice per group were pooled and APC populations enriched as in Fig. 1. The HEL-specific I-Ed-restricted 1H11.3 and the self B2M-specific I-Ad-restricted 2G12.1 T cell hybridomas (upper and lower panels respectively) were cultured (53104 cells/well) with graded numbers of enriched APC populations. After 24 h, antigen-specific IL-2 production was determined by adding 50 µl aliquots of culture supernatant to 104 CTLL cells for an additional 24 h. [3H]Thymidine (1 µCi/well) was added during the last 5 h of culture. Data are presented as mean thymidine incorporation (c.p.m.) from duplicate cultures. Background proliferation of CTLL was usually ,1000 c.p.m. Results are from one representative experiment out of five performed.

Fig. 4. B cells present in vitro protein antigen more efficiently than lymph node DC. DC and B cells were obtained as in Fig. 2 from LNC of BALB/c mice primed with IFA alone 5 days earlier. The indicated numbers of APC enriched in DC (squares), large (triangles) or small (circles) B cells were cultured with the HEL108–116-specific, I-Ed-restricted T cell hybridoma 1H11.3 in the presence of 1 µM HEL (A) or 0.5 µM HEL peptide 105–120 (B). After 24 h of culture, IL-2 concentration was determined using a two-site sandwich ELISA. Results are from one representative experiment out of three performed.

14 Relative role of DC and B cells in antigen presentation in vivo

Fig. 5. Protein antigen is presented in vitro more efficiently by small B cells as compared with immune lymph node DC. DC and small B cells were obtained as described in Fig. 1 from LNC of five mice primed with IFA alone 5 days earlier. 104 DC or 3 3 105 small B cells/well were cultured with the HEL108–116-specific I-Ed-restricted T cell hybridoma 1H11.3 in the presence of the indicated concentrations of HEL (A) or HEL105–120 (B). After 24 h of culture, IL-2 production was determined using a two-site sandwich ELISA.

Fig. 6. The mode of protein antigen administration determines selective presentation by lymph node DC or B cells. Subcutaneous immunization with adjuvant induces recruitment of DC and B cells in the draining lymph node. After administration of HEL in adjuvant DC are the only lymph node APC presenting HEL peptide–class II antigenic complexes. Conversely, after administration of soluble HEL i.v. B cells have a much more potent antigen-presenting activity than DC. These results indicate that DC but not B cells are the initiating APC in immune lymph nodes following administration of protein antigen in adjuvant. The high efficiency of DC in the presentation of antigenic peptides derived from proteins present in inflammatory sites coupled to the decreased capacity of DC recruited in immune lymph nodes to present soluble proteins is consistent with maturation of DC in vivo.

of protein and peptide are observed using B cells, in agreement with previous results (20). Even if we cannot formally exclude a possible inability of soluble antigen to access lymph node DC in vivo, the in vitro data are consistent with the interpretation that DC migrated to the lymph node have a reduced capacity to endocytose and present in situ soluble circulating protein antigen. Maturation of DC in vivo could be

due to the local secretion of proinflammatory cytokines such as tumor necrosis factor-α, which has been shown to downregulate the antigen-capturing and -processing capacity of human DC in vitro (21). Since differences in antigen-presenting activity may also reflect differences in antigen uptake and processing capacity among the three APC populations tested, we analyzed in vitro the hierarchy in antigen-presenting activity to 1H11.3 T cells using synthetic peptide. As compared to large and small B cells, 10- to 30-fold less DC were needed to induce similar levels of IL-2 production by the T hybridoma cells. The number of antigenic complexes on DC required to activate T cells is much lower than for B cells (22). Thus, the more efficient presentation of soluble antigen administered i.v. by B cells rather than DC is likely to reflect a much higher expression of antigenic complexes on B cells as compared to DC, indicating that the circulating protein has been preferentially uptaken and processed in vivo by lymph node B cells. Interestingly, both resting and activated B cells, identified as small and large B cells respectively, appear to present soluble HEL administered i.v. almost equally well. The different role of DC and B cells in the presentation of exogenous proteins in vivo exemplifies the necessity, by professional APC such as DC, to present selectively to lymph node T cells antigens available in inflammatory sites but not circulating soluble proteins. As it has been previously suggested (7), the inefficiency of mature DC to present soluble proteins, including circulating self antigens, to lymph node T cells may also provide an efficient mechanism to avoid selfantigen presentation by DC, thereby preventing the priming of potentially autoreactive T cells. On the other hand, once circulating antigen becomes available, lymph node B cells could further contribute to the clonal expansion of the antigenspecific T cells previously activated by DC (4,23,24), although this issue is still controversial (3). In this context, it is interesting to note that soluble antigen administration usually results in T cell tolerance, mediated by antigen presentation by B cells, rather than DC (25). The fact that B cells are more efficient in presenting soluble

Relative role of DC and B cells in antigen presentation in vivo 15 protein antigen as compared to DC seems to contradict previous results, where following protein administration i.v. splenic DC were the most important source of APC bearing antigenic epitopes (3,5). This could be explained by the use, to monitor the formation of antigenic complexes, of antigenspecific T cell clones (5) or naive CD41 T cell populations (3), which are probably more dependent on co-stimulation than T cell hybridomas. Th1 clones proliferate optimally in response to splenic adherent cells rather than B cells (26). Similarly, it has been shown that DC are the most potent stimulators, as compared to B cells, to activate naive CD41 T cells (27). Therefore, due to the APC requirement of such a read-out system, the antigenic complexes formed on B cells might have been underestimated. In addition, these experiments analyzed steady-state splenic APC isolated 2 h after i.v. injection of the protein antigen (5). It is therefore possible that the protein was preferentially taken up by resident immature DC, which frequency might be higher in a non-inflammatory situation like normal spleen, as compared to immune lymph nodes. In conclusion, lymph node B cells, which are not able to present exogenous protein antigen administered s.c. in adjuvant, are nevertheless the most efficient lymph node APC in presenting circulating protein. Conversely, DC, probably as a consequence of their maturation process in vivo, are inefficient in presenting soluble protein antigen after migration to lymph nodes, but they present very efficiently antigenic peptides derived from processing of protein antigen endocytosed at the site of inflammation.

Acknowledgements J.-C. G. is supported by the Biotechnology Contract BIO2-CT942005 from the European Community.

Abbreviations APC DC HEL IFA LNC β2M OVA

antigen-presenting cell dendritic cell hen egg-white lysozyme incomplete Freund’s adjuvant lymph node cells β2-microglobulin ovalbumin

References 1 Schwartz, R. H. 1985. T-lymphocyte recognition of antigen in association with gene products of the major histocompatibility complex. Annu. Rev. Immunol. 3:237. 2 Mo¨ller, G. 1988. Antigen processing. Immunol. Rev. 106:1. 3 Constant, S., Schweitzer, N., West, J., Ranney, P. and Bottomly, K. 1995. B lymphocytes can be competent antigen-presenting cells for priming CD41 T cells to protein antigens in vivo. J. Immunol. 155:3734. 4 Epstein, M. M., Di Rosa, F., Jankovic, D., Sher, A. and Matzinger, P. 1995. Successful T cell priming in B cell-deficient mice. J. Exp. Med. 182:915. 5 Crowley, M. T., Inaba, K. and Steinman, R. M. 1990. Dendritic cells are the principal cells in mouse spleen bearing immunogenic fragments of foreign proteins. J. Exp. Med. 172:383. 6 Romani, N. and Schuler, G. 1992. The immunological properties of epidermal Langerhans cells as a part of the dendritic cell system. Semin. Immunopathol. 13:265.

7 Steinman, R. M. 1991. The dendritic cell system and its role in immunogenicity. Annu. Rev. Immunol. 9:271. 8 Macatonia, S. E., Knight, S. C., Edwards, A. J., Griffiths, S. and Fryer, P. 1987. Localization of antigen on lymph node dendritic cells after exposure to the contact sensitizer fluorescein isothiocyanate. J. Exp. Med. 166:1654. 9 Schuler, G. and Steinman, R. M. 1985. Murine epidermal Langerhans cells mature into potent immunostimulatory dendritic cells in vitro. J. Exp. Med. 161:526. 10 Romani, N., Koide, S., Crowley, M., Witmer-Pack, M., Livingstone, A. M., Fathman, C. G., Inaba, K. and Steinman, R. M. 1989. Presentation of exogenous protein antigens by dendritic cells to T cell clones: intact protein is presented best by immature, epidermal Langerhans cells. J. Exp. Med. 169:1169. 11 Gue´ry, J.-C., Sette, A., Leighton, J., Dragomir, A. and Adorini, L. 1992. Selective immunosuppression by administration of MHC class II-binding peptides. I. Evidence for in vivo MHC blockade preventing T cell activation. J. Exp. Med. 175:1345. 12 Gue´ry, J.-C., Neagu, M., Rodriguez-Tarduchy, G. and Adorini, L. 1993. Selective immunosuppression by administration of major histocompatibility complex class II-binding peptides. II. Preventive inhibition of primary and secondary antibody responses. J. Exp. Med. 177:1461. 13 Gue´ry, J. C., Ria, F. and Adorini, L. 1996. Dendritic cells but not B cells present antigenic complexes to class II-restricted T cells following administration of protein in adjuvant. J. Exp. Med. 183:751. 14 Metlay, J. P., Witmer-Pack, M. D., Agger, R., Crowley, M. T., Lawless, D. and Steinman, R. M. 1990. The distinct leukocyte integrins of mouse spleen dendritic cells as identified with new hamster monoclonal antibodies. J. Exp. Med. 171:1753. 15 Crowley, M. T., Inaba, K., Witmer-Pack, M. D., Gezelter, S. and Steinman, R. M. 1990. Use of the fluorescence activated cell sorter to enrich dendritic cells from mouse spleen. J. Immunol. Methods. 133:55. 16 Adorini, L., Sette, A., Buus, S., Grey, H. M., Darsley, M., Lehmann, P. V., Doria, G., Nagy, Z. A. and Appella, E. 1988. Interaction of an immunodominant epitope with Ia molecules in T-cell activation. Proc. Natl Acad. Sci. USA. 85:5181. 17 Gue´ry, J.-C., Sette, A., Appella, E. and Adorini, L. 1995. Constitutive presentation of dominant epitopes from endogenous naturally processed self β2 microglobulin to class II-restricted T cells leads to self tolerance. J. Immunol. 154:545. 18 Kyewski, B. A., Fathman, C. G. and Rouse, R. V. 1986. Intrathymic presentation of circulating non-MHC antigens by medullary dendritic cells. An antigen-dependent microenvironment for T cell differentiation. J. Exp. Med. 163:231. 19 Mu¨ller, K. P., Schumacher, J. and Kyewski, B. 1993. Half-life of antigen/major histocompatibility complex class II complexes in vivo: intra- and interorgan variations. Eur. J. Immunol. 23:3203. 20 Constant, S., Sant’Angelo, D., Pasqualini, T., Taylor, T., Levin, D., Flavell, R. and Bottomly, K. 1995. Peptide and protein antigen require distinct antigen-presenting cell subsets for the priming of CD41 T cells. J. Immunol. 154:4915. 21 Sallusto, F. and Lanzavecchia, A. 1994. Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage colony-stimulating factor plus interleukin 4 and downregulated by tumor necrosis factor α. J. Exp. Med. 179:1109. 22 Bhardwaj, N., Young, J. W., Nisanian, A. J., Baggers, J. and Steinman, R. M. 1993. Small amounts of superantigen, when presented on dendritic cells, are sufficient to initiate T cell responses. J. Exp. Med. 178:633. 23 Fuchs, E. F. and Matzinger, P. 1992. B cells turn off virgin but not memory T cells. Science. 258:1156. 24 Ronchese, F. and Hausmann, B. 1993. B lymphocytes in vivo fail to prime naive T cells but can stimulate antigen-experienced T lymphocytes. J. Exp. Med. 177:679. 25 Eynon, E. E. and Parker, D. C. 1992. Small B cells as antigenpresenting cells in the induction of tolerance to soluble protein antigens. J. Exp. Med. 175:131. 26 Gajewski, T. F., Pinnas, M., Wong, T. and Fitch, F. W. 1991. Murine Th1 and Th2 clones proliferate optimally in response to distinct antigen-presenting cell populations. J. Immunol. 146:1750. 27 Croft, M., Duncan, D. D. and Swain, S. L. 1992. Response of naive antigen-specific CD41 T cells in vitro: characteristics and antigen-presenting cell requirements. J. Exp. Med. 176:1431.