The Differential Role of CD86 and CD80 Co-Stimulatory Molecules in the Induction and the Effector Phases of Contact Hypersensitivity

Home , CD80, CD86

The differential role of CD86 and CD80 co-stimulatory molecules in the induction and the effector phases of contact hypersensitivity

1 2 1 1 1

Satoru Nuriya - , Hideo Yagita , Ko Okumura and Miyuki Azuma Downloaded from https://academic.oup.com/intimm/article/8/6/917/677184 by guest on 23 September 2021 department of Immunology, Juntendo University School of Medicine, 2-1-1 Hongo, Bunkyc-ku, Tokyo 113, Japan 2Second Department of Oral and Maxillo-Facial Surgery, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113, Japan

Keywords: dendritic cells, Langerhans cells, regional lymph node, T cell activation

Abstract Contact hypersensitivity (CH) has served as a useful model for investigating the allergen-specific immune responses of T cells and skin-associated antigen-presenting cells. We examined the distinct role between CD80 and CD86 on hapten-lnduced CH in both an induction and an effector phase. Intraperltoneal injection of mAb against CD86, but not CD80, 2 h before sensitizatlon with epicutaneous application of dinttrofluorobenzene led to an almost complete inhibition of ear swelling, and histologically a marked reduction of edema, Inflammatory polymorphonuclear cells and lymphocyte Infiltration In the dermis. In contrast, the administration with either anti-CD80 or CD86 mAb 2 h before challenge partially inhibited CH reactions and a combination of both mAb did not improve the inhibitory effect Although Langerhans cells (LC) expressing MHC class II were observed in both the epidermis and dermis 24 h after primary sensitizatlon, CD86+ LC were observed only In the subepidermal regions and CD80-bearing cells were not detected. Dendritic cells (DC) expressing both CD80 and CD86 were preferentially observed in the T cell areas of draining lymph nodes 24 h after challenge. The administration of anti-CD8€ mAb In the induction phase prominently reduced the up-regulatlon of CD80 and CD86 on DC In the lymph nodes. The predominant role of CD86 was further supported by a marked reduction In proliferation of lymph node T cells against the sensitized hapten after the antl-CD86 treatment. These results suggest that CD86 plays a critical role in the initiation of primary immune responses In the skin, while CD80 and CD86 are not essential in the effector phase of CH reactions.

Introduction An optimal antigen-specific T cell activation requires two human peripheral blood DC and monocytes constitutively distinct signals (1,2). One originates from the ligation of the express CD86, but not CD80 (6-8). However, the expression TCR and the second signal is provided by the direct interaction of both CD80 and CD86 can be augmented by culture between co-stimulatory molecules on T cells and their ligands with various stimuli (7,9-11). In contrast to the constitutive on antigen-presenting cells (APC). It is generally accepted expression of CD86 on DC and monocytes, substantial that the signal through the binding of CD28 on T cells with expression of either CD80 or CD86 was not observed on its ligands, CD80 (B7/BB1, B7-1) and CD86 (B70, B7-2), on freshly isolated epidermal Langerhans cells (LC) or in situ in APC is a crucial co-stimulatory pathway (3-5). CD80 and tissues, but culture with media or several cytokines is capable CD86 are expressed on professional APC such as dendritic of inducing the expression of both CD80 and CD86 on LC cells (DC), monocytes and macrophages. Only subpopula- (6,12-14). tions of freshly isolated, non-cultured murine splenic DC and The skin allergic contact hypersensitivity (CH) response is

Correspondence tor. M. Azuma Transmitting editor. T. Takemori Received 13 January 1996, accepted 6 March 1996 918 Differential role of CD86 and CD80 in contact hypersensitivity a classical animal model for delayed-type hypersensitivity Assay for contact hypersensitivity (DTH), which results from cellular immune responses medi- CH was induced by sensitization with 0.5% of dinitrofluoro- ated by T 1 cells. Epidermal LC are the principal APC in CH h benzene (DNFB; Sigma, St Louis, MO) or 5% trinitrochloro- reactions. Contact with allergens results in the increased benzene (TNCB; Tokyo Kasei, Tokyo, Japan) dissolved in expression of MHC class II molecules on LC (15), which then acetone:olive oil (4:1) by painting to the shaved abdominal migrate to regional lymph nodes (16) where they induce skin of BALB/c mice on two consecutive days (days 0 and primary T cell responses. However, the actual molecular 1). Three days after the final sensitization (day 4), mice were events involved in these reactions have not yet been com- challenged on both surfaces of the ears with 10 nl of 0.2% pletely understood. It was reported that IL-1P is a critical DNFB or 1% TNCB. The ear thickness was measured using cytokine produced by LC after exposure to contact allergen a dial thickness gauge (Peacock, Ozaki, Tokyo, Japan). (17), and that IL-10 (18,19) or UV-B radiation (20) inhibits Measurement of each lobe was repeated and a total of four antigen-presenting function of LC and reduces the CH reac- measurements per mouse was performed. The ear thickness tion. In addition, recent reports suggest that the inhibition of was measured before and 24 h after challenge, since prelimin- the antigen-presenting function of LC or peripheral DC by Downloaded from https://academic.oup.com/intimm/article/8/6/917/677184 by guest on 23 September 2021 ary experiments on the time-kinetics of ear swelling demon- IL-10 or UV-B radiation results from a failure in up-regulation strated maximal responses at 24 h. The individual value of of CD86 (14,21-23). Thus, the regulation of CD80 and CD86 ear swelling was taken as the mean value of four measure- on LC seems to play an important role in antigen-specific T ments per mouse post-challenge minus the mean value before and LC interactions in the skin. To clarify the direct involvement challenge. The ear thicknesses from mice that were painted of CD80 and CD86 molecules in CH reactions, we examined with the vehicle control (acetone/olive oil) alone and then the effect of in vivo administration of anti-CD80 and/or anti- were challenged are presented as negative controls. CD86 mAb on hapten-induced CH reactions and the in situ expression of these molecules in the skin and draining lymph nodes. Antibody treatment Each group of four to six mice received i.p. injection with 250 ^g/mouse of the indicated mAb in a total volume of 0.5 ml Methods 2 h prior to every sensitization or challenge. The mice in control groups received either PBS or 250 ng/mouse of Mice irrelevant rat Ig (Sigma). The percent suppression was calculated according to the formula: % suppression = (1 - net ear Female BALB/c mice were obtained from Japan SLC swelling in treated mice/ net ear swelling in PBS-treated (Hamamatsu, Shizuoka, Japan) and used at 7-8 weeks of age. mice)xi00. Generation of anti-CD80 and -CD86 mAb CD86 co-stimulation assay Rat anti-mouse CD80 mAb (RM80, lgG2a) and CD86 mAb [PO1 (IgM), PO2 (lgG2b), PO3 (lgG2b), PO4 (lgG1) and PO5 Splenic T cells from BALB/c mice were purified by passage (lgG2a)] were obtained from a fusion of P3U1 myeloma cells through nylon-wool columns and treatment with anti-l-A and with SD rat splenocytes immunized with BCL1 cells expressing anti-CD24 mAb and rabbit complement. Purified T cells were large amounts of both CD80 and CD86 or J774 cells co-cultured with irradiated (80 Gy) CD86-P815 in the presence expressing high level of CD86 antigen alone. The specificity of anti-CD3 (2C11, 2 ng/ml) mAb. Blocking mAb were added of mAb was confirmed by specific reactivity with a murine at the start of the assay at the indicated concentrations. 3 mastocytoma P815 cell lines transfected with either murine Cultures were pulsed with [ H]thymidine (0.5 nCi/well) CD80 or CD86 cDNA, as described previously (24). RM80 (DuPont-NEN, Boston, MA) for the final 6 h of a 3 day culture reacted with CD80-P815 but not with CD86-P815, while mAb and harvested using a Micro 96 Harvester (Skatron, Lier, PO1-5 reacted with CD86-P815 but not with CD80-P815 (not Norway). Incorporated radioactivity was measured on a micro- shown). In agreement with previous results (25, 26), mAb plate p counter (Micro Beta Plus; Wallac, Turku, Finland). PO2-5 precipitated a single molecule of -65 kDa from the Percent inhibition of proliferation was calculated as: [(c.p.m. lysate of BCL1 cells under reducing conditions (data not without mAb - c.p.m. with mAb)/c.p.m. without mAb]xi00. shown). These mAb were purified from ascites by standard procedures using caprylic acid extraction and the purity of Hapten-specific T cell proliferation assay mAb was verified by SDS-PAGE analysis. Lymph nodes from normal mice and draining regional lymph nodes from the mice treated with either PBS, RM80 or PO3 mAb at the time of TNCB sensitization and unsensitized negative GL1 (anti-CD86, rat lgG2a) and 1G10 (anti-CD80, rat lgG2a) control mice were surgically removed 72 h after the final were obtained from PharMingen (San Diego, CA). For immuno- sensitization and the T cells were purified as described histochemical analysis, anti-l-A(M5/114.15.2, rat lgG2b), anti- above. Purified T cells (2xiOs/well) were co-cultured with the CD4 (GK1.5, rat lgG2b), anti-CD80 (RM80) and anti-CD86 indicated number of trinitrobenzensulfonate (TNBS)-pulsed or (PO3) were used. Hybridomas producing mAb against I-A, unpulsed normal irradiated (30 Gy) splenocytes in 96-well CD4, CD3 (2C11, hamster IgG) and CD24 (J11d, rat IgM) flat-bottom plates as previously described (19). Cultures were were obtained from the American Type Culture Collection pulsed with 1.0 (iCi/well [3H]thymidine for the last 16 h of a (Rockville, MD). 72 h incubation. Differential role of CD86 and CD80 in contact hypersensitivity 919

Histological examination 100 Tissue specimens of the ears and regional lymph nodes from the treated or untreated mice were surgically removed immediately after determination of ear swelling, embedded in OCT compound (Tissue-tek®, Miles, IN), frozen in liquid nitrogen and stored at -70°C until used. Cryostat sections of 3-4 urn thickness were fixed in absolute acetone. For immunohistochemical staining, the sections were incubated with either rat anti-mouse MHC class II (M5/114), CD4 (GK1.5), CD80 (RM80) or CD86 (PO3) mAb after blocking with 10% normal rabbit serum and then incubated with secondary biotinylated rabbit anti-rat Ig (\fector, Burlingame, CA). Localiz- ation of antigen-antibody complexes was performed by treat-

5 4 3 2 Downloaded from https://academic.oup.com/intimm/article/8/6/917/677184 by guest on 23 September 2021 ment with peroxidase-conjugated ABC complex (Vectastain 10~ 10" 10' 10" 10 ABC; Vector) and the substrate, diaminobenzidine. Between ug/ml each incubation the sections were rinsed in PBS. Finally, the Rg. 1. Inhibitory effects of anti-CD86 mAb in vitro. Purified splenic sections were rinsed in distilled water and counterstained with T cells (ixioS/well) were co-cultured with irradiated CD86-P815 methyl green. Specificity of positive staining was confirmed on pxiCrVwell) in the presence of anti-CD3 mAb. The serially diluted every section by the absence of staining with the-second- anti-CD86 mAb (GL1, PO2, PO3, PO4 and PO5) were added at the step reagent alone. Alternate sections were stained with H&E. start of the assay. Cultures were pulsed with pHJthymidine for 6 h and harvested on day 3. Percent inhibition of proliferation was calculated as- [(c p m. without mAb - c p.m. with mAb)/c p.m. without Statistical analysis mAb]xiOO The actual mean incorporation in the absence of blocking Student's f-test was performed to determine the statistical mAb co-cultured with CD86-P815 was 88594 c p m. No substantial proliferation was observed by transfectants alone or in co-cultures of significance of ear swelling responses. transfectants with T cells in the absence of anti-CD3 mAb (data not shown). Each value is the mean percent inhibition of triplicate wells. The data are representative of three experiments with similar results Results

Inhibitory effects of anti-CD86 mAb in vitro In order to evaluate the functional inhibitory effects of anti- from PBS-treated mice as a control group in the following all CD86 mAb, we performed a CD86-co-stimulation assay. experiments, since we observed no differences between mice Purified splenic T cells were co-cultured with CD86-P815 treated with PBS and irrelevant rat Ig in our preliminary transfectants in the presence of a suboptimal dose of anti- experiments. Administration of anti-CD86 mAb resulted in an CD3 mAb and the serial dilutions of each anti-CD86 mAb almost complete inhibition of the ear swelling compared with were added to the culture. As shown in Fig. 1, the highest the PBS-treated control group, while the treatment with anti- amount (50 ng/ml) of all anti-CD86 mAb, including a commer- CD80 mAb did not affect ear swelling (Fig. 2A). The in vivo cially available GL1 mAb, almost completely inhibited the inhibitory efficacy of individual anti-CD86 mAb seemed to CD86-dependent proliferation of T cells, and this inhibitory correlate with their blocking ability in CD86-co-stimulated effect occurred in a dose-dependent manner. Lower amounts T cell proliferation in vitro, since PO2 and PO4 exhibited less (0.05-0.005 |ig/ml) of mAb caused partial inhibitions, especi- inhibitory effects on ear swelling (Fig. 2B). To negate the ally in the case of PO2, PO4 and PO5, suggesting that these possibility that the inability of anti-CD80 mAb to inhibit ear three mAb have relatively less inhibitory efficacy. In previous swelling resulted from a suboptimal dose or lower efficacy of experiments using a similar assay system, we demonstrated the RM80 mAb, we compared RM80 with another anti-CD80 that the blocking ability of PO3 in CD86-dependent T cell mAb, 1G10, and obtained substantially identical results (not proliferation was comparable to that of RM80 in CD80- shown). Furthermore, the treatment with a combination of dependent proliferation (24). both RM80 and PO3 did not significantly affect the ear swelling more than the results observed with PO3 alone. Treatment with anti-CD86 mAb in the induction phase effici- The efficacy of in vivo treatment with anti-CD86 mAb ently inhibits CH reaction was also demonstrated by histological examination. Figure 3 CH is a form of DTH in which the target organ is the skin and shows representative samples of ears unsensitized (Fig. 3a) the inflammatory response is elicited by the epicutaneous or treated with either PBS (Fig. 3b), RM80 (Fig. 3c) or PO3 application of a reactive hapten. The response consists of an (Fig. 3d). Histological examination of the PBS-treated mice induction phase at the sensitization and an effector phase at showed severe swelling, edema and abundant leukocyte the hapten challenge. We first investigated the effect of anti- infiltration, in comparison to unsensitized mice. However, the CD80 (RM80) and/or anti-CD86 (PO3) mAb on the induction ears from the PO3-treated mice demonstrated an almost phase of CH reactions. Two hundred and fifty microgram of complete absence of swelling, edema and cellular infiltration. either RM80, PO3 or both mAb in 0.5 ml PBS or the same Consistent with the results of the ear swelling assay, histo- volume of control reagents was administered i.p. 2 h prior to logically, no obvious improvement was observed in the RM80- every sensitization with DNFB and the ear swelling at 24 h treated mice (Rg. 3c). These results clearly demonstrate that post-DNFB challenge was examined. We present the data the in vivo treatment with anti-CD86, but not anti-CD80 mAb, 920 Differential role of CD86 and CD80 in contact hypersensitivity

aensmzanon i reatmeni onanenqe

vehicle (-) DNFB

DNFB PBS DNFB

DNFB RM80 DNFB

DNFB PO3 DNFB

DNFB RM80+PO3 DNFB

5 10 15 20 40 60 80 100 Ear Swelling (|jm) % suppression

Fig. 2. Administration of anti-CD86 mAb in the induction phase efficiently inhibits CH reactions. mAb against CD80 (RM80) and/or CD86 (PO2,

PO3, PO4, PO5 and GL1) were injected i p 2 h before sensitization with 0.5% DNFB. On day 4, each ear was challenged with 0.2% DNFB Downloaded from https://academic.oup.com/intimm/article/8/6/917/677184 by guest on 23 September 2021 and the ear thickness was evaluated 24 h later. (A) The effects of treatment with either RM80, PO3 or both Data represent the mean changes in ear thickness ± SD. A summary of six experiments are presented. (B) The effects of treatment with various anti-CD86 mAb Percent suppression was calculated as described in methods. Each group consisted of 6-10 mice The mean ear thickness change in control PBS- treated group was 9.1 ± 2.9 urn. Statistically different from the control PBS-treated group ("p < 0 05, "p < 0 005).

efficiently inhibits the hapten-induced CH reactions elicited the number of MHC class II positive cells to the control level by epicutaneous sensitization in an induction phase. (Fig. 4g-i).

Migrate LC in the dermis express abundant CD86 24 h after Lymph node DC from the CH-elicited mice express both CD80 sensitization and CD86 We investigated further the possibility that this dominant effect It was reported that LC migrate to draining lymph nodes and of anti-CD86 mAb resulted from the preferential expression that the increase in the number of DC with antigens in the of CD86 on APC in the CH reactive site. It was previously regional lymph nodes was detected within 30 min, reached reported that phenotypic and functional change of LC in the a peak at -24 h and returned to normal levels by day 6 after epidermis occurred within 24 h after application of haptens skin painting (27). We, therefore, investigated the CD80 and (15). We, therefore, examined the expression of MHC class CD86 expression in draining regional lymph nodes 24 h after II, CD80 and CD86 in the skin 24 h after primary application challenge (Fig. 5). The location of the B cell follicles and T cell with hapten. Figure 4 shows the results of immunohisto- areas were distinguished by the comparative staining with chemical staining with anti-MHC class II (M5/114) (Fig. 4a, d anti-CD4 and anti-MHC class II mAb (not shown). It was and g), anti-CD80 (RM80) (Fig. 4b, e and h) and anti-CD86 reported that, in normal lymph nodes, DC expressing CD86 (PO3) (Fig. 4c, f and i) mAb in the ears from mice unsensitized but not CD80 were found mainly in the T cell area as control (Fig. 4a-c), 24 h after the primary sensitization with previously described (6). In the draining lymph node from 0.2% DNFB (Fig. 4d-f) and the anti-CD86 mAb treated mice the unsensitized mice, scattered CD86+ DC (Fig. 5b) were 24 h after DNFB challenge (Fig. 4g—i). Consistent with the observed mostly in the T cell-rich areas, which was stained previous report, we observed that numerous cells morpho- positive with anti-CD4 mAb. In contrast, no apparent CD80+ logically resembling DC expressed MHC class II antigen in cells were observed (Fig. 5a). No positive staining with the control sections (Fig. 4a). These cells were detected in the anti-CD80 or anti-CD86 mAb was detected on cells that basal layer of epidermis, as well as dermis. The numbers of morphologically resembled B cells in the cortical areas. The MHC class ll-positive cells were clearly increased at 24 h lymph node from the sensitized and challenged mice strongly after hapten application (Fig. 4d). Although CD80 or CD86 expressed CD80 (c) and CD86 (d), preferentially in the T cell- positive cells were hardly detected in control sections (Fig. rich area, in parallel with the MHC class II expression (not 4b and c), the cells strongly expressing CD86 were observed shown). The cell profiles at higher magnification indicated 24 h after primary sensitization (Fig. 4f). It should be noted that these positive cells were mostly DC and some macro- that most of CD86+ cells were found in the subepidermal phages but not T cells. The numbers of CD80+ and CD86+ zone, but not within the epidermis. The CD86+ cells in upper cells were moderately reduced by the anti-CD86 treatment dermis appeared to be LC from their morphology, and the (Fig. 5e and f). These results suggest that the anti-CD86 mAb CD86+ cells in the deeper layer of the dermis appeared to treatment abrogated the subsequent interaction of DC and consist of both LC and macrophages. In contrast, we could T cells in the regional lymph nodes by inhibiting the primary hardly detect any CD80 positive cells even after hapten interaction between naive T cells and the antigen-bearing LC. painting (Fig. 4e). Although an obvious difference was not Furthermore, it is likely that the failure of the anti-CD80 mAb observed in the expression of MHC class II and CD86 in the treatment to effect in the induction phase of CH responses sensitized and challenged-ears, as compared with the ears resulted from the relative lack or low expression of CD80, or after primary hapten application, a few weakly CD80 positive non-functional expression on LC at the time of sensitization. cells were detected (not shown). Interestingly, anti-CD86 However, once the antigen is presented by LC expressing treatment, but not anti-CD80 treatment (not shown), clearly CD86, both CD80 and CD86 were comparably induced by reduced the number of CD86 expressing cells and decreased the consecutive reactions in the lymph node. Differential role of CD86 and CD80 in contact hypersensitivity 921

•-£&? '•*'•*• •-=• Downloaded from https://academic.oup.com/intimm/article/8/6/917/677184 by guest on 23 September 2021

Fig. 3. Reduction of inflammatory responses by anti-CD86 mAb treatment. The ears were surgically excised from unsensitized (a), sensitized mice treated with PBS (b), RM80 (c) or P03 (d) at 24 h after DNFB challenge. Rgures show H&E-stained sections of the corresponding central areas Magnification. x25

Hapten-specific T cell proliferation is inhibited by the anti- lated in vitro with TNBS-pulsed splenocytes (Fig. 6). The CD86 mAb treatment. treatment with anti-CD86 mAb (P03) in vivo significantly To determine whether draining lymph node T cells from mice reduced the proliferative response to TNBS, whereas the treated with anti-CD86 mAb become hyporesponsive to the in vivo treatment with anti-CD80 mAb (RM80) failed to inhibit sensitized antigen, we examined the in vitro proliferative proliferative responses. Lymph node T cells from RM80- responses. Purified lymph node T cells from TNCB-sensitized treated mice consistently exhibited a higher proliferative or unsensitized mice were co-cultured with TNBS (water response to unpulsed syngeneic splenocytes than the other soluble analogue of TNCB)-pulsed or unpulsed intact syn- three groups (Fig. 6B). Further studies are required to deter- geneic splenocytes for 72 h. Unsensitized T cells failed to mine the reason for this observation. Taken together, our data proliferate in response to either TNBS-pulsed or unpulsed indicate that contact allergen-specific T cell responses can splenocytes. Lymph node T cells from TNCB-sensitized mice be inhibited by the anti-CD86 mAb treatment in the induction showed a -3-fold higher proliferative response when stimu- phase of CH reactions. I Downloaded from https://academic.oup.com/intimm/article/8/6/917/677184 by guest on 23 September 2021 i cB" o

Q. O 1

§ o I 55 I to win-

Fig. 4. Migrated dermal LC strongly express CD86 but not CD80. Immunohistochemical detection of MHC class II (a, d and g), CD80 (b, e and h), and CD86 (c, f and i) in the ear sections from mice 24 h after primary vehicle control alone (a-c) or DNFB (d-f) application and from the anti-CD86 treated mice in the induction phase 24 h after DNFB challenge (g-i) Cryostat sections were stained with either anti-MHC class II, anti-CD80 or anti-CD86 mAb as described in Methods. Magnification. X50. Differential role of CD86 and CD80 in contact hypersensitivity 923 Downloaded from https://academic.oup.com/intimm/article/8/6/917/677184 by guest on 23 September 2021

Fig. 5. Expression of CD80 and CD86 in the T cell areas of the sensitized lymph node. Draining lymph nodes from unsensitized mice (a and b) or sensitized mice treated with PBS (c and d) or with anti-CD86 mAb (e and f) were removed 24 h after challenge and were stained with anti-CD80 (a, c and e) or anti-CD86 (b, d and f) mAb. Magnification- X50.

Either treatment with anti-CD80 or CD86 mAb in the effector any additive effects by a combination of both mAb. These phase partially inhibits CH responses results suggest a partial contribution of both CD80 and CD86 We next determined whether anti-CD80 and/or anti-CD86 in the effector phase of CH reactions and the presence of treatment would modulate the CH responses in the effector some CD80/CD86-independent mechanism. phase. DNFB-sensitized mice were treated with control- reagents, RM80, PO3 or both 2 h prior to challenge and the Discussion ear swelling was measured 24 h after challenge. The data represent a summary of four experiments from 22 to 26 mice The co-stimulatory pathway mediated by CD28 has been per group (Fig. 7). Although there were marked individual implicated in various immune responses and is an important variations in the ear thickness after treatment, partial but target for immune intervention. CTLA-4-lg treatment, which statistically significant (P < 0.005) inhibitory effects were blocks the interactions of CD28 on T cells with CD80 and observed in either anti-CD80 or anti-CD86 mAb-treated group, CD86 counter-receptors on APC, has succeeded in inhibiting as compared to the PBS-treated group. We have not observed immune responses in the various experimental models in vivo 924 Differential role of CD86 and CD80 in contact hypersensitivity

5 5 5 6x10 2X10 1x10 6X105 2X105 1X105

Number of splenocytes (cells/well)

Rg. 6. In vivo treatment with anti-CD86 mAb inhibits secondary hapten-specific proliferative response of fymph node T cells in vitro. Draining Downloaded from https://academic.oup.com/intimm/article/8/6/917/677184 by guest on 23 September 2021 lymph nodes from TNCB-sensitized mice (closed symbols) treated with either PBS (circles), RM80 (triangles) or PO3 (squares) or mice painted with the vehicle control alone (open circles) were removed 72 h after the final sensitization. Purified lymph node T cells (2xiO5/well) were co-cultured with the indicated number of TNBS-pulsed (A) or unpulsed (B) irradiated (30 Gy) syngeneic splenocytes for 72 h. Incorporation of [^Jthymidine (1 nCiAvell) for the final 16 h was measured. Each value is the mean of triplicate wells SEM of the triplicates was <10% The data are representative of three experiment with similar results

Sensitizatton Treatment Challenge roles in the induction of T cell-mediated immunity against environmental, infectious and tumor-associated antigens in vehicle (-) DNFB the skin. However, resident LC in the epidermis and the LC DNFB PBS DNFB freshly isolated from normal skin are inefficient with regard to their capacity to induce antigen-specific T cell proliferation, DNFB RM80 DNFB as compared with DC derived from lymph nodes, spleen and DNFB PO3 DNFB peripheral blood (39-41). Within 24 h after antigen application, LC in the epidermis increase in size and in their expression DNFB. RM80+PO3 DNFB of MHC class II and exhibit an enhanced capability for 5 10 15 Ear Swelling (Mm) inducing T cell proliferation (15). Our immunohistochemical examination of the contact allergen-sensitized skin demon- Fig. 7. Treatment with anti-CD80 and/or anti-CD86 mAb in the effector strated a selective up-regulation of CD86 on LC in the dermis. phase partially inhibits CH reactions DNFB-sensitized mice received We could not detect substantial staining of CD80 on LC in i.p. injection with PBS, RM80, PO3 or both 2 h before DNFB challenge. the skin 24 h after the primary sensitization or after the Ear thickness was determined as described in Rg 2 These data secondary challenge. It was reported that CD80 expression represent a summary of three to four experiments with 20-26 mice on LC was induced by in vitro culture with cytokines such as per group. Statistically different from the control PBS-treated group CP < 0.005) granulocyte macrophage colony stimulating factor (GM-CSF), tumor necrosis factor-a, IL-1a, IL-ip and IL-4 (14) although the expression of CD86 was much higher than CD80 (13). Similar findings were obtained from studying cultured human peripheral blood DC and monocytes (7,42,43). (28-34). However, the mechanisms for immunosuppression and induction of tolerance have not yet been completely At present, there is no information on the differential regula- defined. In this study, we demonstrated the differential roles tion of CD80 and CD86 in LC after the in vivo antigen of CD80 and CD86 co-stimulatory molecules in two distinct stimulation. Our study demonstrates the dominant up-regula- phases of CH reactions. Our results revealed that CD86, but tion of CD86 expression after specific-antigen stimulation in not CD80, plays a potent role in the induction phase of CH situ. Similar to our findings on LC, B cells activated with a reactions, which is mediated by the interaction of antigen- combination of anti-n and IFN-yor anti-lgD-dextran preferen- bearing LC and naive T cells. Recently, several studies tially increased CD86 expression in correlation with antigen suggest the distinct co-stimulatory signal between CD80 presentation activity (44,45). Abundant expression of CD86 and CD86 or the preferential contribution of CD86-mediated on LC after sensitization seems to correlate with the functional signals to Th2 cell differentiation (35-38). However, it is a ability of skin-associated APC to induce initial T cell responses general agreement that the CH response results from cellular to allergen. Our data further indicate that the treatment with immune responses mainly mediated by Tf,1 cells from T cell anti-CD86 mAb induces a contact allergen-specific T cell transfer experiments and the Th1 cytokine blocking studies. unresponsiveness in the induction phase of CH. As the Therefore, the dominant effect of anti-CD86 mAb in this model absence of co-stimulation in antigen presentation may lead is unlikely to relate with the balance between Th1- and 1^2- to T cell anergy, blocking of the interaction of LC with T cells mediated immune responses. using the anti-CD86 mAb may induce tolerance in the reaction It is known that epidermal LC are specialized APC in of naive T cells to APC. Our results that lymph node T cells epithelial tissues derived from bone marrow and play key did not respond to secondary stimulation in vitro and in vivo Differential role of CD86 and CD80 in contact hypersensitivity 925 suggests the T cell responses were inhibited by the treatment specific T cell responses. However, CD80 and CD86 are not with anti-CD86 mAb. Rechallenging after a prolonged period essential for the effector functions involved in CH reactions. will clarify whether antigen-specific tolerance is induced. We Our present findings will be useful for elucidating the patho- cannot completely exclude the possibility that the administra- genesis of allergic CH in humans. tion of anti-CD86 mAb itself may modify the migration of LC to lymph nodes. However, we have confirmed that PO3 mAb was not associated with the depletion of target cells in our Acknowledgements previous studies (24) and the signaling capacity by anti-CD86 V\fe thank Dr L. L Lanier (DNAX research institute) for critical reading mAb was not confirmed at present. Therefore it seems unlikely of the manuscript, and Drs N. Sueyoshi, A. Furuhata and K. Yokomizo that the binding of PO3 mAb to LC affected the migration. for technical assistance on tissue staining This work was supported by grants from Ministry of education, Culture, and Science and the Further studies using monovalent mAb fragments might Ministry of Health, Japan answer this question. In addition, the failure of suppression with the anti-CD80 mAb in secondary responses may result from the low expression of CD80 on LC in the initial sensitiza- Abbreviations Downloaded from https://academic.oup.com/intimm/article/8/6/917/677184 by guest on 23 September 2021 tion with allergen. Alternatively, CD80 might deliver a regu- APC antigen-presenting cells latory signal interacting with one of its counter receptors, CH contact hypersensitivity CTLA-4 (46). Recently, negative regulatory roles of CTLA-4 DC dendritic cells were demonstrated by the studies using CTLA-4 deficient DNFB dmitrofluorobenzene mice (47,48). If it is the case that CD80 is a major functional DTH delayed-type hypersensitivity LC Langerhans cells •receptor for CTLA-4, the spontaneous higher proliferative TNBS trinitrobenzensulfonate responses of T cells derived from the anti-CD80 treated mice TNCB trinitrochlorobenzene might be explained. Administration of either anti-CD80 or CD86 mAb in the References effector phase comparably reduced CH reactions. This result can be explained from the equivalent expression of both 1 Mueller, D L , Jenkins, M K. and Schwartz, R H. 1989. Clonal CD80 and CD86 in draining lymph node DC in the sensitized expansion versus functional clonal inactivation. a co-stimulatory signalling pathway determines the outcome of T cell antigen mice. After the ear challenge, DC expressing both CD80 and receptor occupancy Annu Rev. Immunol. 7.445 CD86 may present antigens transported by LC and rapidly 2 Schwartz, Ft. H. 1990 A cell culture model for T lymphocyte clonal activate memory T cells which quickly migrate into the anergy. Science 248:1349. challenged site. Furthermore the fact that mice initial treated 3 Harding, F. A., McArthur, J G., Gross, J. A., Raulet, D. H. and with anti-CD86 mAb reduced CD80 and CD86 expression on Allison, J -P 1992. CD28 mediated signalling co-stimulates murine T cells and prevents induction of anergy in T cell clones Nature lymph node DC suggests that the interaction of CD86-bearing 356.607 LC with T cells is essential for up-regulation of CD80 and 4 Linsley.PS andLedbetter, J.A 1993 TheroleoftheCD28receptor CD86 in lymph node DC. Previous reports demonstrated that during T cell responses to antigen Annu. Rev. Immunol 11191. LC migrate from epidermis to the dermis and transport 5 Allison, J P 1994. CD28-B7 interactions in T-cell activation. Curr antigens to regional draining lymph node via lymphatic vessels Opin Immunol 6:414 6 Inaba, K., Witmer-Pack, M., Inaba, M., Hathcock, K. S., Sakuta, H., within 24 h after skin painting and initiate T cell activation Azuma, M., Yagita, H , Okumura, K , Linsley, P S., Ikehara, S , (16,27,40,49). Based on this concept, the expression of CD80 Muramatu, S., Hodes, R. J. and Steinman, R. M. 1994. The tissue and CD86 on DC may be induced by cytokines secreted in distribution of the B7-2 co-stimulator in mice, abundant expression the initial activation of naive T cells. Since GM-CSF and IFN-y on dendritic cells in situ and during maturation in vitro. J Exp produced by T cells are capable of inducing CD80 and CD86 Med. 180.1849. 7 McLellan, A. D., Starling, G C , Williams, L A., Hock, B. D and on DC (12), these factors may work as potent regulators of Hart, D. N. J 1995. Activation of human peripheral blood dendritic lymph node DC. cells induces the CD86 co-stimulatory molecule Eur. J. Immunol. Despite the treatment with both anti-CD80 and CD86 mAb, 25:2064. 8 Azuma, M , Ito, D., Yagita, H.r Okumura, K., Phillips, J. H., we failed to obtain complete inhibition of CH responses at an Lanier, L. L. and Somoza, C 1993. B70 antigen is a second ligand effector phase. The early phase of CH reactions (within the for CTLA-4 and CD28 Nature 366:76. first 24 h after challenge) manifests characteristic neutrophil 9 Young, J. W., Koukova, L,Soergel, S. A ,Clark, E. A., Steinman, R. M. and Dupont, B. 1992. The B7/BB1 antigen provides one of several infiltration and edema, which is mediated by leukocyte adhe- + sion to vascular endothelium. Predominant lymphocyte infil- co-stimulatory signals for the activation of CD4 T lymphocytes by human blood dendritic cells in vitro. J. Clin. Invest. 90:229. tration was observed at 48 and 72 h. Therefore other receptor/ 10 Larsen, C. P., Ritchie, S. C, Pearson, T. C , Linsley, P. S and Lowry, ligand pairs that permit adhesion of cells to endothelium, such R. P 1992. Functional expression of the co-stimulatory molecule, as the very late antigen molecule-4/vascular cell adhesion B7/BB1, on murine dendritic cell population. J. Exp. Med. 176:1215. molecules-1 and intercellular adhesion molecule-1 (CD54)/ 11 Caux, C, vanbervliet, B., Massacrier, C, Azuma, M., Okumura, K., leukocyte function-associated molecule-1 (CD11a/CD18) may Lanier, L. L. and Banchereau, J. 1994. B70/B7-2 is identical to CD86 and is the major functional ligand for CD28 expressed on be involved in the early phase of these reactions (50,51). human dendritic cells. J. Exp. Med. 180:1841. Taken together, we demonstrated that CD86 plays a domin- 12 Larsen,C ,Ritchie,S.C.Hendrix,R.,Linsley,P.S .Hathcock,K.S., ant role in the interaction between migratory LC and naive Hodes, R. J., Lowry, R. P. and Pearson, T. C. 1994. Regulation of immuno-stimulatory function and co-stimulatory molecule (B7-1 T cells in the draining lymph nodes after epicutaneous applica- and B7-2) expression on murine dendritic cells. J. Immunol. tion of hapten. Blockade of the CD86-CD28 co-stimulatory 152:5208. pathway in the induction phase could inhibit contact allergen- 13 Kawamura, T. and Furue, M 1995. Comparative analysis of B7-1 926 Differential role of CD86 and CD80 in contact hypersensitivity

and B7-2 expression in Langerhans cells, differential regulation by 32 Rnck, B. K , Linsley, P S and Wofsy, D. 1994 Treatment of murine T helper type 1 and T helper type 2 cytokines. Eur. J. Immunol lupus with CTLA-4lg Science 265:1225. 25:1913 33 Baliga, P., Chavin, K. D., Qin, L, Woodward, J., Lin, J., Linsley, P S 14 Chang, C -H., Furue, M and Tamaki, K 1995 B7-1 expression of and Bromberg, J S. 1994. CTLA-4lg prolongs allograft survival Langerhans cells is up-regulated by proinflammatory cytokines, while suppressing cell-mediated immunity Transplantation and is down- regulated by interferon-y or by intereukm-10 Eur J. 58 1082. Immunol. 25:394. 34 Knoerzer, D. B , Karr, R. W., Schwartz, B. D and Mengle-Gaw, L J 15 Aiba, S. and Katz, S I. 1990. Phenotypic and functional 1995. Collagen-induced arthritis in the BB rat. J. Clm Invest. 96:987 characterization of in two-activated Langerhans cells. 35 Kuchroo, V K., Das, M. P, Brown, J A., Ranger, A. M., Zamvil, S. J. Immunol 145.2791. S., Sobel, R. A., Wemer, H. L, Nabavi, N. and Glimcher, L H. 1995. 16 Krispke, M. L., Munn, C G , Jeevan, A., Tang, J. and Bucana, C. B7-1 and B7-2 co-stimulatory molecules activate differentially the 1990 Evidence that cutaneous antigen-presenting cells migrate to Th1/Th2 developmental pathways: application to autoimmune regional lymph nodes during contact sensitization. J. Immunol disease therapy. Cell 80707 145.2833. 36 Thompson, C. B. 1995. Distinct roles for the co-stimulatory ligands 17 Enk, A. H., Angeloni, V. L, Udey, M. C and Katz, S. I. 1993. An B7-1 and B7-2 in T helper cell differentiation? Cell 81.979 essential role for Langerhans cell-derived IL-1(5 in the initiation of 37 Freeman, J G., Boussiotis, V. A., Anumanthan, A., Bernstein, G. M , primary immune responses in skin J. Immunol. 150.3698 Ke, X -Y, Rennert, P. D., Gray, G. S , Gribben, J. G. and Nadler, L M. Downloaded from https://academic.oup.com/intimm/article/8/6/917/677184 by guest on 23 September 2021 18 Enk, A. H., Saioga, J., Becker, D., Mohamadzadeh, M. and Knop, J. 1995 B7-1 and B7-2do not deliver identical co-stimulatory signals, 1994 Induction of hapten-specific tolerance by interleukin-10 since B7-2 but not B7-1 preferentially co-stimulates the initial in vivo. J Exp. Med 179-1397 production of IL-4. Immunity 2:523 19 Schwarz, A , Grabbe, S., Riemann, H., Aragane, Y., Simon, M , 38 Lenschow, D J , Ho, S. C , Sattar, H., Rhee, L., Gray, G , Nabavi, N., Manon, S., Andrade, S., Luger, T. A , Zlotnik, A and Schwarz, T. Herold, K. C. and Bluestone, J. A 1995. Differential effects of anti- 1994 In vivo effects of interleukin-10 on contact hypersensitivity B7-1 and anti-B7-2 monoclonal antibody treatment on the and delayed-type hypersensitivity reactions J. Invest Dermatol development of diabetes in the nonobese diabetic mouse J Exp 103211 Med. 181.1145 20 Noonan, F P., Krispke, M. L , Pedersen, G. M and Greene, M I. 39 Schuler, G and Stemman, R. M 1985 Murine epidermal 1981 Suppression of contact hypersensitivity in mice by ultraviolet Langerhans cells mature into potent immuno-stimulatory dendritic irradiation is associated with defective antigen presentation cells in vitro J Exp. Med 161 526. Immunology 43.527 40 Inaba, K, Romani, N and Stemman, R. M. 1989. An antigen- 21 Mitra, R. S , Judge, T. A , Nestle, F O., Turka, L A. and Nickoloff, B J independent contact mechanism as an early step in T cell- 1995 Psoriatic skin-derived dendritic cell function is inhibited by proliferative responses to dendritic cells J Exp. Med 170 exogenous IL-10. J Immunol 1542668 41 Larsen, C. P., Steinman, R M., Witmer-Pack, M , Hankins, D F, 22 Young, J. W., Baggers, J. and Soergel, S. A 1993 High-dose UV- Morris, P J and Austyn, J M. 1990. Migration and maturation of B radiation alters human dendritic cell co-stimulatory activity but Langerhans cells in skin transplants and explants. J. Exp Med does not allow dendritic cells to tolerize T lymphocytes to 1721483. alloantigen in vitro. Blood81.2987 42 Fagnoni.F F,Takamizawa,M.,Dodfrey, W R.,Rivas, A.,Azuma,M , + 23 Weiss, J M.,Renkl, A. C.Denfeld, R. W., de Roche, R.,Sprtzlei, M., Okumura, K. and Engleman, E G 1995. Roleof B70/B7-2 in CD4 Schopf, E and Simon, J C. 1995 Low-dose UVB radiation perturbs T-cell immune responses induced by dendritic cells. Immunology the functional expression of B7. 1 and B7. 2 co-stimulatory 85:467. molecules on human Langerhans cells. Eur J. Immunol. 252858 43 Fujihara.M , Takahashi.T. A., Azuma, M.,Ogiso,C.,Maekawa,T L , 24 Nakajima, A., Azuma, M , Kodera, S , Nuriya, S., Terashi, A , Yagita, H., Okumura, K. and Sekiguchi, S 1996 Decreased Abe, M , Hirose, S , Shirai, T, Yagita, H. and Okumura, K. 1995. inducible expression of CD80 and CD86 in human monocytes after Preferential dependence of autoantibody production in munne ultraviolet-B irradiation- its involvement in inactivation of lupus on CD86 co-stimulatory molecule Eur. J. Immunol. 25:3060 allogenecity. Blood 87:2386. 25 Hathcock, K. S., Laszlo, G , Dickler, H B., Bradshaw, J., Lmsley, P 44 Hathcock, K S , Laszlo, G , Pucillo, C , Linsley, P. and Hodes, R J and Hodes, R. J. 1993 Identification of an alternative CTLA-4 ligand 1994. Comparative analysis of B7-1 and B7-2 co-stimulatory co-stimulatory for T cell activation. Science 262:905. ligands: Expression and function J. Exp. Med. 180.631. 26 Chen, C, Faherty, D. A., Gault, A., Connaughton, S. E , Powers, G 45 Morokata, T., Kato, T., Igarashi, O. and Nariuchi, H. 1992 D., Godfrey, D. I. and Nabavi, N. 1994. Monoclonal antibody 2D10 Mechanism of enhanced antigen presentation by B cells activated recognizes a novel T cell co-stimulatory molecule on activated with anti-n plus interferon-7: role of B7-2 in the activation of naive + murine B lymphocytes. J. Immunol 152:2105. and memory CD4 T cells. Eur. J. Immunol. 25.1992 27 Macatonia, S. E., Knight, S C, Edwards, A. J , Griffiths, S and 46 Sethna, M R.Panjs, L. V.Sharpe, A H , Abbas, A K and Freeman, Fryer, P. 1987. Localization of antigen on lymph node dendritic cells G J 1994 . A negative regulatory function of B7 revealed in B7-1 after exposure to the contact sensrtizer fluorescein isothiocyanate. transgenic mice. Immunity. 1.415. Functional and morphological studies. J. Exp. Med 166:1654. 47 Waterhouse, P., Penninger, J. M., Timms, E , Wakeham, A., 28 Linsley, P S, Wallace, P. M., Johnson, J., Gibson, M. G., Shahinian, A., Lee, K. P, Thompson, C B., Griesser, H. and Greene, J. L., Ledbetter, J A., Singh, C. and Tepper, M A 1992. Mak, T. M. 1995 Lymphoproliferative disorders with early lethality Immunosuppression in vivo by a soluble form of the CTLA-4 T cell in mice deficient in ctla-4. Science. 270.985. activation molecule. Science 257 792. 48 Tivol, E. A., Borriello, F, Schweitzer, A N., Lynch, W. P, 29 Lenschow, D. J., Zeng, Y, Thistlethwaite, J. R., Montag, A., Brandy, Bluestone, J A. and Sharps, A. H 1995. Loss of CTLA-4 leads to W.,Gibson,M.G., Linsley, P S. andBluestone, J. A 1992.Long-term massive lymphoproWeration and fatal multiorgan tissue survival of xenogeneic pancreatic islet grafts induced by CTLA-4lg destruction, revealing a critical negative regulatory role of CTLA-4. Science 257 789. Immunity 3:541 30 Turka, L. A., Linsley, P. S., Lin, H., Brady, W, Leiden, J M , Wei, R.- 49 Macatonia, S. E., Edwards, A. J. and Knight, S C 1986. Dendritic O., Gibson, M L, Zheng, X.-G., Myrdal, S., Gordon, D., Bailey, T, cells and the initiation of contact sensitivity to fluorescein Boiling, S F. and Thompson, C. B. 1992. T-cell activation by the isothiocyanate. Immunology59:509 CD28 ligand B7 is required for cardiac allograft rejection in vivo. 50 Chisholm, P. L., Williams, C A. and Lobb, R. R. 1993. Monoclonal Proc Natl. Acad. Sci USA 89:11102. antibodies to the integrin ot-4 subunit inhibit the murine contact 31 Wallace, P. M , Johnson, J. S., MacMaster, J. F., Kennedy, K. A., hypersensitivity response. Eur. J. Immunol. 23.682 Gladstone, P. and Linsley, P. S. 1994. CTLA-4lg treat- 51 Scheynius, A., Camp, R. L. and Pure, E. 1993 Reduced contact ment ameliorates the lethality of murine graft-versus-host sensitivity reactions in mice treated with monoclonal antibodies to disease across major histocompatibility complex barriers. leukocyte function-associated mo)ecule-1 and intercellular Transplantation 58.602. adhesion molecule-1. J. Immunol 150:655.