- Email: [email protected]
Function of dendritic cells and changes in T cell proliferation in antigen-induced nonresponsiveness
CELLULAR
IMMUNOLOGY
139,342-35 1 (1992)
Function of Dendritic Cells and Changes in T Cell Proliferation in Antigen-Induced Nonresponsiveness S. HILL, A. STACKPOOLE,*
I. KIMBER,? AND S. C. KNIGHT
Division of Immunological Medicine and *Section of Transplantation Biology, Clinical Research Centre, Watford Road, Harrow, Middlesex, HAI 3UJ, United Kingdom: and tlmmunology Group, C.T.L., I.C.I., Alderley Park, Macclesjeld, Cheshire, SKI0 4TJ United Kingdom Received May 8, 1991; accepted August 16, 1991
The ability of dendritic cells (DC) to acquire and present antigen to T cells during antigeninduced nonresponsiveness(AINR) in contact sensitivity was examined by studying cells from lymph nodes draining the sites of antigen challenge. Mice were pretreated on the right flank with either vehicle (AOO), oxazolone (Ox), or fluorescein isothiocyanate (FITC) and challenged 5, 10, or 20 days later with FITC on the left flank. At 5, 10, and 20 days, compared with animals pretreated with vehicle and challenged with FITC, those pretreated and challenged with FITC showed reduced acquisition of antigen by DC and the DC showed a reduced ability to stimulate naive T cells in vitro. Proliferation of T cells immediately on isolation (reflecting in vivo activity) was also reduced. When the time between pretreatment and challenge was extended to 40 days, the proliferative responsesand antigen acquisition returned to normal. Animals sensitized with Ox and challenged with FITC showed nonspecific inhibition of T cell proliferation at 5 days only and not at later times and antigen levels on the DC from these animals were normal. The results show that low T cell proliferation during specific AINR in contact sensitivity may be a consequence of reduced acquisition and presentation of antigen by DC. 0 1992 Academic press, 1~.
INTRODUCTION Some dendritic cells (DC)’ isolated from lymph nodes are thought to be derived from skin Langerhans’ cells which acquire antigen in the skin and then migrate via the afferent lymphatics to the draining lymph nodes (1). Skin painting mice with contact sensitizers such as oxazolone, picryl chloride, or fluorescein isothiocyanate (FITC) results in a large increase in the number of dendritic cells (DC) in draining lymph nodes (2) and the antigen in the draining lymph nodes is preferentially found on the DC. These antigen-bearing DC are capable of stimulating proliferation of syngeneic T cells from naive animals and also induce delayed hypersensitivity when injected into recipient mice (3). Antigen-induced nonresponsiveness (AINR) during contact sensitization appears as inhibition of proliferative responsesto one chemical by prior sensitization with the same or a different chemical (4, 5). Antigen-bearing DC are potent stimulators of primary T cell responses in vivo and in vitro (3), and ’ Abbreviations used: AINR, antigen-induced nonresponsiveness;FITC, fluorescein isothiocyanate; Ox, oxazolone; AOO, acetone:olive oil (vehicle); DBP, dibutylphthalate; LN, lymph nodes; LNC, lymph node cells; DC, dendritic cells. 342 OOOg-8749192 $3.00 Copyrisht 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved
DENDRITIC
CELLS IN NONRESPONSIVENESS
343
this study investigated the acquisition and presentation of antigen by DC during AINR. An earlier report indicated that there were changes in antigen acquisition by DC in AINR (6). The current study used different contact sensitizers, mouse strain, and challenge site from the earlier study. The data presented here confirm that, in AINR, acquisition of specific antigen by DC draining a challenge site may be reduced. However the nonspecific effectswere less marked than in the earlier study. Here we show that, during specific inhibition, the capacity of the DC to present antigen to stimulate T cells is also impaired and examine the relationship between this and the ongoing T cell proliferation in vivo. MATERIALS AND METHODS Animals Female CBA mice (6-9 weeks) were obtained from the specific pathogen-free unit at the Clinical Research Centre. Chemicals and Skin Painting 4-ethoxymethylene-2-phenyl oxazolone (Ox), [BDH, Poole, Dorset, UK] was dissolved in an acetone:olive oil mixture (AOO) (4: 1). Fluorescein isothiocyanate (FITC, isomer 1, Sigma, Poole, Dorset, UK) was dissolved in a 1:1 acetone:dibutylphthalate mixture (AOO:DBP). Mice were pretreated on the shaved right flank with 25 ~1 of 1%Ox, 0.8% FITC, or vehicle. At various times after pretreatment mice were challenged on the shaved left flank with 25 ~10.8% FITC or vehicle. Cell Separation One to three days after challenge, single-cell suspensionswere prepared from pooled axillary/brachial, inguinal, and popliteal draining lymph nodes of treated and control mice. Lymph nodes were pressedthrough wire mesh and washed in RPM1 1640 medium (Dutch modification, Flow Labs, Irvine, Ayrshire UK) supplemented with 100 III/ml penicillin, 100 pg/ml streptomycin, 1 X 10e5 2-mercaptoethanol, and 10% heat-inactivated fetal calf serum. A cell suspension (5-8 ml) at 5 X lo6 cells/ml was layered onto 2 ml metrizamide (Nygaard, Oslo, Norway; Analytical grade: 14.5 g added to 100 ml medium) and centrifuged for 10 min at 600g. Cells at the interface were collected, washed once in medium, and counted. Dendritic cells (DC) were identified by their distinctive morphology and were >70% pure as previously reported (2). Enriched T cells (>90% purity) (7) were obtained by passageof DC-depleted lymph node cells over a nylon wool column (8). In Vitro Culture DC from treated and untreated control mice were cultured in 20 ~1hanging drops (2000 DC/well) in inverted Terasaki plates containing 6-100 X lo3 enriched T cells. After 3 days the cultures were pulsed for 2 hr with 1 ~1 [3H]thymidine (2 Ci/mmol; Amersham International, Amersham, Bucks, UK) to give 1 pg thymidine/ml and harvested by blotting onto filter discs. The acid-insoluble material (5% trichloroacetic acid) was counted using a liquid scintillation counter (9).
344
HILL
ET AL.
FluorescenceAnalysis Cells were suspendedin 0.5 ml phosphate-bufferedsaline containing EDTA (1 mM), sodium azide (0.02%) and 2% FCS and were analyzed using a fluorescence-activated cell sorter (FACStar Plus, Becton-Dickinson, Mountain View, CA). Five thousand DC were analyzed for each preparation. The cells were gated on the control DC population. DC were considered to be fluorescent when the mean fluorescence was above that of the control population. Levels of fluorescence and numbers of fluorescent DC are given as mean + standard error. Statistics Results were analyzed using analysis of variance and the Student t test. RESULTS Changes in Cell Numbers Mice skin painted with contact sensitizers such as FITC show a rapid increase in the number of DC in draining lymph nodes (2,3). In this system, mice were pretreated with either vehicle (AOO), oxazolone (Ox), or FITC on the shaved right flank and then challenged with FITC on the left flank at 5, 10, 20, or 40 days afterward and lymph nodes (LN) taken 3 days after challenge. Representative experiments for each time point are shown. After pretreatment with either AOO, Ox, or FITC followed by challenge with FITC there were large increasesin the number of lymph node cells (LNC) and DC isolated from draining LN (Figs. IA and IB) but there was no significant change in the ratio of LNC:DC or between the different pretreatments (AOO, Ox, or FITC; P > 0.05). If mice were pretreated and challenged with vehicle (AOO/DBP or DBP/DBP) the number of cells obtained was either equal to, or only slightly above, control levels. Acquisition of Antigen by Dendritic Cells DC isolated from draining LN 3 days after challenge were analyzed for fluorescence using flow cytometry. When the time between pretreatment and challenge was 5 days, animals pretreated with A00 before challenge with FITC had 29% (k2) fluorescent DC (Fig. 2A). Animals pretreated with Ox had 35% (k2) fluorescent DC whilst those pretreated with FITC had fewer fluorescent DC (2 1% + 1). The differences between the groups pretreated with either AOO, Ox, or FITC and then challenged with FITC were significant (P < 0.01). This was true for all comparisons between the different treatments. These figures are based on the mean values of 12 experiments with four animals in each group. The mean fluorescence of DC from these groups followed the same pattern, (AOO/FITC: 403 f 32, Ox/FITC: 471 f 29, FITC/FITC: 213 -t 19). The differences between the mean fluorescence values of the above groups were also significant between the different treatments (P < 0.001). When the time between pretreatment and challenge was extended to 10 days (Fig. 2B), DC from animals treated with Ox/FITC had numbers of fluorescent DC similar to those from AOO/FITC-treated mice (35% +- 3 compared to 38% + 7). The mean fluorescence was also similar (458 f 147 compared to 428 f 58). Mice treated with FITC/FITC had similar numbers of fluorescent DC compared with the groups above
DENDRITIC
50
CELLS IN NONRESF’ONSIVENESS
345
B
1
d b
40-
;
30-
x
20-
2 10 01
5 days
10 days
20 days
40 days
FIG. 1. Changes in the number of lymph node cells (LNC) (A) and dendritic cells (DC) (B) isolated from draining lymph nodes (challenge site). Animals were pretreated on the right flank with either vehicle (AOO), oxazolone, or FITC and at various times later challenged on the left flank with FWC. Draining lymph nodes were taken 3 days later. Lymph nodes were taken from four animals in each experiment. Each experiment was repeated at least 3 times (12 times at 5 days and 3 times at 10, 20, and 40 days) and the results of one representative experiment for each time point are shown. Time between pretreatment and challenge: 5, 10,20, q , OxlFITC; q , FITClFITC or 40 days. 0, control; q , AOO/FITC;
and the mean fluorescence was also lower (3 1% f 1 and 232 k 52, respectively). A similar pattern of results also occurred when the time between pretreatment and challenge was extended to 20 days (Fig. 2C). Similar results were obtained in at least three experiments at each time point (10 and 20 days). The differencesbetween the numbers in groups at each time point were not significant at the 5% level (P > 0.05). At 40 days between pretreatment and challenge there was no significant difference between the groups in either the percentage of fluorescent DC or the levels of antigen they expressed(Fig. 2D). T Cell Proliferation The changes in T cell proliferation caused by AINR were also examined. On separation, enriched T cells were cultured in vitro for 2 hr with [3H]thymidine and then harvested. Reduced T cell proliferation was found in animals treated with FITC/FITC (specific suppression) in all experiments where the time between pretreatment and challenge was up to 20 days (Figs. 3A-3C). Nonspecific suppression (i.e., pretreatment with Ox and challenge with FITC) was found in half the experiments at 5 days and intermittently at later times. By 40 days levels of T cell proliferation were the same in all treated groups of mice in all three experiments (Fig. 3D).
HILL ET AL.
346
150
-I
100 50
$ c' \ :' .,
./ 0 l?4!!a+ 0 12
"..A+, 3
4
FL (40) FIG. 2. Fluorescence on dendritic cells from draining lymph nodes. Mice were pretreated with either vehicle (AOO), oxazolone (Ox), or FITC. At various times later the animals were challenged on the left flank with FITC and the draining lymph nodes taken 3 days afterward. Time between pretreatment and challenge: A, 5 days; B, 10 days; C, 20 days; D, 40 days. -, control; -, AOO/RTC; ----, Ox/FITC, . . . . . , FITC/FITC.
The functional capacity of DC from treated and control mice was assessedby culturing the DC with naive T cells in 20 ~1 hanging drop cultures. In all experiments up to 20 days there was reduced stimulation of T cells by DC from animals treated with FITC/FITC (Figs. 4A-4C) and this may be due to lower levels of antigen on the DC. Nonspecific inhibition of proliferation was seen in some experiments at 5 days (Fig. 4A), but was not significant at later times (Figs. 4B-4D). At 40 days the ability of DC treated with FITC/FITC to stimulate normal T cells was significantly inhibited in 1 of 3 experiments. The time between pretreatment and challenge was altered to compare 5 and 3 days. At both time points specific inhibition was the most marked effect. Nonspecific suppression of T cell proliferation (Figs. 5A and 5B) was found intermittently at 3 days along with inhibition of DC stimulation of naive T cells (Figs. 6A and 6B). Reducing Time after Challenge before Assay The time after challenge was reduced to 1 or 2 days to compare with 3 days to see if the nonspecific suppression of T cell proliferation and DC stimulation of naive T cells were consistent phenomena. It was found that immediate T cell proliferation above background levels could only be detected at 3 days after challenge (data not shown). Stimulation of naive T cells by DC produced unexpected results. At 1 day after challenge the mean fluorescence on DC from animals treated with FITC/FITC was lower than that of DC treated with AOO/FITC or Ox/FITC as seen using cells 3
DENDRITIC CELLS IN NONRESPONSIVENESS
347
,B
50- A 40 30 20 10 -y 9 0 5
0 / 4-
:&g -D
6.25 12.5 25
50
100
6.25 12.5 25
50
100
T cells/well (x 1Om3)
FIG. 3 T. cells were isolated from mice pretreated with either vehicle (AOO), oxazolone (Ox), or FITC and challenged at various times afterward with FITC. Immediately after separation 6.25-100 X lo3 T cells were cultured with [‘Hlthymidine for 2 hr and then harvested. Results from one experiment for each time point are shown. Time between pretreatment and challenge: A, 5 days; B, 10 days; C, 20 days; D, 40 days. 0, control; 0, AOO/FITC; n , Ox/FITC, A, FITC/FITC.
days after challenge. There were equal percentagesof fluorescent DC in each group. However, the DC from FITC/FITC were significantly more stimulatory than the DC from the other groups (P < 0.01). As the time after challenge increased both the percentage of fluorescent DC and the mean fluorescence on DC from AOO/FITC and Ox/FITC increased peaking on Day 2 and then declining. However, the percentage of fluorescent DC decreased in the FITC/FITC group and the mean fluorescence remained similar to that of Day 1. These data agree with the study of Macatonia et al. (1987) (7). DISCUSSION Up to 20 days after pretreatment with FITC, subsequent challenge with the same antigen resulted in lower acquisition of antigen by DC in the draining lymph node. These cells also caused less stimulation of syngeneic T cells and this correlated with the presence of lower T cell activity occurring in the lymph node T cells in viva Previous work by Kimber et al. (5, 6, 10) has shown that AINR in contact sensitivity is characterized by inhibition of proliferative responsesand by lower levels of fluo-
HILL ET AL.
348
6.25 12.5 25
50
100
6.25 12.5 25
50
100
T cells/well (x 10m3) FIG. 4. Stimulation of syngeneic T cells by dendritic ceils exposed to antigen in viva. Animals were pretreated with either vehicle (AOO), oxazolone (Ox), or FITC and at various times later challenged with FITC. Draining lymph nodes were taken 3 days later. Dendritic cells were then cultured with naive T cells. Uptake of [3H]thymidine after 3 days in culture is shown. Time between pretreatment and challenge: A, 5 days; B, 10 days; C, 20 days; D, 40 days. 0, control T cells only; 0, Normal DC, W,AOO/RTC; A, Ox/ FITC; +, FITC/FITC.
rescenceon DC. This study confirms the idea that DC acquire lower levels of antigen following pretreatment and challenge with the same chemical. This appears to be a robust phenomenon since this study used a different sensitizing antigen, challenge site, and mouse strain from Kimber et al. (5, 6, 10). This study now shows evidence that the changed T cell function is a consequence of the changes in acquisition and presentation of antigen by the DC. At 5 days DC isolated from animals treated with FITC/FTTC did not stimulate naive T cells in vitro to the levels seen when T cells were cultured with DC from animals treated with AOO/FTTC. There was also specific inhibition of T cell proliferation in vitro on separation (which reflectsT cell proliferation in vivo) thus showing the link between the stimulatory capacity of the DC and the activity of the T cells. Previous studies by Macatonia et al. (3) have shown the relationship between the levels of antigen on DC and the stimulation of naive T cells in vitro. The low levels of antigen on DC from mice treated with FITQFITC suggest that this may be the causeof the low-level T cell proliferation. However, an exception to this was found when DC were isolated 1 day after challenge. These cells were more
DENDRITIC CELLS IN NONRESPONSIVENESS 60
1
A
349
lB
6.25 12.5 25
50
100
6.25 12.5 25
50
100
T cells/well (x 10-S) FIG. 5 T. cells were isolated from mice pretreated with either vehicle (AOO), oxazolone (Ox), or FITC and challenged with FITC, 5 or 3 days later. Immediately after separation 6.25-100 X 10’ T cells were cultured with [3H]thymidine for 2 hr and then harvested. Results from one experiment at each time point are shown. Time between pretreatment and challenge: A, 5 days; B, 3 days. 0, control; 0, AOO/FITC; n , Ox/FITC; A, FITC/FITC.
stimulatory with lower levels of antigen than DC isolated from mice pretreated with A00 or Ox and then challenged with FITC. The idea that DC dictate responsivenessdirectly in T cells in some antigen-specific manner is not a tenable hypothesis since antigenic specificity clearly resides in T cell heterogeneity. Preliminary evidence from experiments in nude, athymic CBA suggest that the block in acquisition of antigen and reduced stimulation of T cells in vitro does not occur. The effects shown here may therefore be dependent on the presence of T cells during sensitization.
30, A
6.i5 li.5
25
50
100
6.25 12.5 25
50
100
T cells/well (x 10-Z) FIG. 6. Stimulation of syngeneic T cells by dendritic cells exposed to antigen in viva. Mice were pretreated with either vehicle (AOO), oxazolone (Ox), or FITC and challenged 5 or 3 days later with FITC. Draining lymph nodes were taken 3 days after challenge. Uptake of [3H]thymidine after 3 days in culture is shown. Time between pretreatment and challenge: A, 5 days; B, 3 days. 0, control T cells only; 0, Normal DC; n , AOO/FITC; A, Ox/FITC; +, FITQFITC.
350
HILL ET AL.
After sensitization with either picryl chloride or oxazolone, IgM anti-hapten antibody was detected within 4 days (11). Macatonia (12) has shown that sensitization with FITC induced antibody production and this antibody could be detected up to 20 days after sensitization. The presence of antibody may explain the low levels of antigen on DC isolated at 5, 10, and 20 days from animals treated with FITC/FITC. Manta et al. (13) have reported that the addition of antibody when antigen is administered reduced the ability of accessorycells to present the antigen to antigen-specific T cell hybridoma clones. The presence of specific antibody on challenge may lead to the formation of antigen-antibody complexes. These complexes are thought to be removed either via macrophages by phagocytosis or by follicular dendritic cells (FDC) via the subcapsular sinus of the draining lymph nodes. The complexes are then transported to the germinal centers leading to B cell stimulation (14). These alternative pathways for the removal of antigen may account for the low levels of antigen seen on the lymphoid DC examined in this study. The nonspecific inhibition of proliferation in vivo and in vitro appearsto be a more complicated phenomenon. The reduced acquisition of antigen by DC seemsan unlikely explanation for the intermittent nonspecific inhibition of T cell proliferation since this occurred where the antigen levels on the autologous DC were at normal levels. Antibody to Ox cross-reacting with FITC is unlikely since Ox and FITC antigenically unrelated as they do not cross-stimulate (unpublished observation). There is some evidence to suggest that when doubly haptenated antigen-presenting cells are used to sensitize guinea pigs, antigenic competition results due to competition for the antigen-binding sites on these cells (15). However, animals sensitized 5 days previously do not have detectable antigen on DC isolated from contralateral lymph nodes (16). Since DC do not appear to migrate from lymph nodes (17) it seemsunlikely that competition for binding sites in this situation can explain the nonspecific inhibition of T cell proliferation in vivo and in vitro. There are reports that T cell clones can be inactivated by high levels of antigen (18, 19) even though these cells express IL-2 receptors and produce IL-2 and this may be involved in both the specific and nonspecific inhibition of T cell proliferation. An alternative explanation for the low T cell proliferation in vivo is T cell anergy. Rammensee et al. (20) have shown that exposure of adult Mls-lb mice to cells from Mls- 1a mice resulted in specific unresponsiveness when cells from Mls- 1b were stimulated by cells from Mls- 1a mice in a mixed lymphocyte reaction and this was thought to be due to clonal anergy. Experiments by Webb et al. (21) confirmed the findings of Rammensee et al. (20) and further showed that the specific nonresponsiveness waned after severalweeksand that there was also deletion of specific T cell populations. From the data presented here an analogy may be drawn with the results of Webb et al. (21) in that specific nonresponsiveness wanes after several weeks; however, it is not known if there is deletion of T cell populations in this system. The nonresponsivenessseen here may be a result of diminished ability of DC to interact with T cells leading to reduced proliferation. However, where the block on the interaction between these cells lies is unclear. The reduced capacity of DC and T cells to interact has also been seen in some diseasesas a result of infection with viruses or bacteria; both HIV infection and inflammatory arthritis show reduced DC and T cell function (22-24). The nonresponsiveness seen in this model of AINR using contact sensitizers may provide a method for examining the changes in DC and T cell interactions during infection.
DENDRITIC
CELLS IN NONRESPONSIVENESS
351
ACKNOWLEDGMENT S. Hill is supported by I.C.I.
REFERENCES 1. Silberberg-Sinakin,I., Thorbecke, G. J., Baer, R. L., Rosenthal, S. A., and Berezowsky,V., Cell. Immunol. 83, 137, 1976. 2. Knight, S. C., Krecjci, J., Malkovsky, M., Colizzi, V., Gautam, A., Asherson, G. L., Cell. Zmmunol. 94, 427, 1985. 3. Macatonia, S. E., Edwards, A. J., and Knight, S. C., Immunology 59, 509, 1986. 4. Nakano, Y., Immunology 33, 167, 1977. 5. Kimber, I., Pierce, B. B., Mitchell, J. A., and Kinnaird, A., Znt. Arch. Allergy Appl. Zmmunol. 84, 256, 1987. 6. Kimber, I., Hill, S., Mitchell, J. A., Peters, S. W., and Knight, S. C., Immunology 71, 271, 1990. 7. Macatonia, S. E., Knight, S. C., Edwards, A. J., Griffiths, S., and Fryer, P., J. Exp. Med. 166, 1654, 1987. 8. Julius, M. H., Simpson, E., and Herzenberg, L. A., Eur. J. Zmmunol. 3, 645, 1973. 9. O’Brien, J., Knight, S. C., Quick, N. A., Moore, E. H., and Platt, A. S., J. Zmmunol. Methods 27, 219, 1979. 10. Kimber, I., Shepherd, C. J., Mitchell, J. A., Turk, J. L., and Baker, D., Immunology 66, 577, 1989. 11. Asherson, G., Colizzi, V., and Watkins, M. C., Immunology 48, 561, 1983. 12. Macatonia, S. E., “Dendritic Cells as Initiators of Delayed-Type Hypersensitivity to Contact Sensitizers.” PhD. CNAA, 1987. 13. Manta, F., Fenoglio, D., Kunkl, A., Cambiaggi, C., Sasso,M., and Celada, F., J. Zmmunol. 140, 2893, 1988. 14. Szakal, A. K., Kosco, M. H., and Tew, J. G., Annu. Rev. Zmmunol. 7, 91, 1989. 15. Polak, L., Znt. Arch. Allergy Appl. Zmmunol. 76, 275, 1985. 16. Hill, S., Kimber, I., Edwards, A. J., and Knight, S. C., Immunology 71, 277, 1990. 17. Fossum, S., Stand. J. Zmmunol. 27, 97, 1988. 18. Suzuki, G., Kawase, Y., Koyasu, S., Yahara, I., Kobayashi, Y., and Schwartz, R., J. Zmmunol. 140, 1359, 1988. 19. Ceredig, R., and Corradin, G., Eur. J. Zmmunol. 16, 30, 1986. 20. Rammensee, H.-G., Kroschewski, R., and Frangoulis, B., Nature 339, 541, 1989. 2 1. Webb, S., Morris, C., and Sprent, J., Cell 63, 1249, 1990. 22. Macatonia, S. E., Patterson, S., and Knight, S. C., Immunology 67, 285, 1989. 23. Macatonia, S. E., Lau, R., Patterson, S., Pinching, A. J., and Knight, S. C., Immunology 71, 38, 1990. 24. Stagg,A. J., Harding, B., Hughes, R. A., Keat, A., and Knight, S. C., Clin. Exp. Zmmunol. 84,66, 1991.
IMMUNOLOGY
139,342-35 1 (1992)
Function of Dendritic Cells and Changes in T Cell Proliferation in Antigen-Induced Nonresponsiveness S. HILL, A. STACKPOOLE,*
I. KIMBER,? AND S. C. KNIGHT
Division of Immunological Medicine and *Section of Transplantation Biology, Clinical Research Centre, Watford Road, Harrow, Middlesex, HAI 3UJ, United Kingdom: and tlmmunology Group, C.T.L., I.C.I., Alderley Park, Macclesjeld, Cheshire, SKI0 4TJ United Kingdom Received May 8, 1991; accepted August 16, 1991
The ability of dendritic cells (DC) to acquire and present antigen to T cells during antigeninduced nonresponsiveness(AINR) in contact sensitivity was examined by studying cells from lymph nodes draining the sites of antigen challenge. Mice were pretreated on the right flank with either vehicle (AOO), oxazolone (Ox), or fluorescein isothiocyanate (FITC) and challenged 5, 10, or 20 days later with FITC on the left flank. At 5, 10, and 20 days, compared with animals pretreated with vehicle and challenged with FITC, those pretreated and challenged with FITC showed reduced acquisition of antigen by DC and the DC showed a reduced ability to stimulate naive T cells in vitro. Proliferation of T cells immediately on isolation (reflecting in vivo activity) was also reduced. When the time between pretreatment and challenge was extended to 40 days, the proliferative responsesand antigen acquisition returned to normal. Animals sensitized with Ox and challenged with FITC showed nonspecific inhibition of T cell proliferation at 5 days only and not at later times and antigen levels on the DC from these animals were normal. The results show that low T cell proliferation during specific AINR in contact sensitivity may be a consequence of reduced acquisition and presentation of antigen by DC. 0 1992 Academic press, 1~.
INTRODUCTION Some dendritic cells (DC)’ isolated from lymph nodes are thought to be derived from skin Langerhans’ cells which acquire antigen in the skin and then migrate via the afferent lymphatics to the draining lymph nodes (1). Skin painting mice with contact sensitizers such as oxazolone, picryl chloride, or fluorescein isothiocyanate (FITC) results in a large increase in the number of dendritic cells (DC) in draining lymph nodes (2) and the antigen in the draining lymph nodes is preferentially found on the DC. These antigen-bearing DC are capable of stimulating proliferation of syngeneic T cells from naive animals and also induce delayed hypersensitivity when injected into recipient mice (3). Antigen-induced nonresponsiveness (AINR) during contact sensitization appears as inhibition of proliferative responsesto one chemical by prior sensitization with the same or a different chemical (4, 5). Antigen-bearing DC are potent stimulators of primary T cell responses in vivo and in vitro (3), and ’ Abbreviations used: AINR, antigen-induced nonresponsiveness;FITC, fluorescein isothiocyanate; Ox, oxazolone; AOO, acetone:olive oil (vehicle); DBP, dibutylphthalate; LN, lymph nodes; LNC, lymph node cells; DC, dendritic cells. 342 OOOg-8749192 $3.00 Copyrisht 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved
DENDRITIC
CELLS IN NONRESPONSIVENESS
343
this study investigated the acquisition and presentation of antigen by DC during AINR. An earlier report indicated that there were changes in antigen acquisition by DC in AINR (6). The current study used different contact sensitizers, mouse strain, and challenge site from the earlier study. The data presented here confirm that, in AINR, acquisition of specific antigen by DC draining a challenge site may be reduced. However the nonspecific effectswere less marked than in the earlier study. Here we show that, during specific inhibition, the capacity of the DC to present antigen to stimulate T cells is also impaired and examine the relationship between this and the ongoing T cell proliferation in vivo. MATERIALS AND METHODS Animals Female CBA mice (6-9 weeks) were obtained from the specific pathogen-free unit at the Clinical Research Centre. Chemicals and Skin Painting 4-ethoxymethylene-2-phenyl oxazolone (Ox), [BDH, Poole, Dorset, UK] was dissolved in an acetone:olive oil mixture (AOO) (4: 1). Fluorescein isothiocyanate (FITC, isomer 1, Sigma, Poole, Dorset, UK) was dissolved in a 1:1 acetone:dibutylphthalate mixture (AOO:DBP). Mice were pretreated on the shaved right flank with 25 ~1 of 1%Ox, 0.8% FITC, or vehicle. At various times after pretreatment mice were challenged on the shaved left flank with 25 ~10.8% FITC or vehicle. Cell Separation One to three days after challenge, single-cell suspensionswere prepared from pooled axillary/brachial, inguinal, and popliteal draining lymph nodes of treated and control mice. Lymph nodes were pressedthrough wire mesh and washed in RPM1 1640 medium (Dutch modification, Flow Labs, Irvine, Ayrshire UK) supplemented with 100 III/ml penicillin, 100 pg/ml streptomycin, 1 X 10e5 2-mercaptoethanol, and 10% heat-inactivated fetal calf serum. A cell suspension (5-8 ml) at 5 X lo6 cells/ml was layered onto 2 ml metrizamide (Nygaard, Oslo, Norway; Analytical grade: 14.5 g added to 100 ml medium) and centrifuged for 10 min at 600g. Cells at the interface were collected, washed once in medium, and counted. Dendritic cells (DC) were identified by their distinctive morphology and were >70% pure as previously reported (2). Enriched T cells (>90% purity) (7) were obtained by passageof DC-depleted lymph node cells over a nylon wool column (8). In Vitro Culture DC from treated and untreated control mice were cultured in 20 ~1hanging drops (2000 DC/well) in inverted Terasaki plates containing 6-100 X lo3 enriched T cells. After 3 days the cultures were pulsed for 2 hr with 1 ~1 [3H]thymidine (2 Ci/mmol; Amersham International, Amersham, Bucks, UK) to give 1 pg thymidine/ml and harvested by blotting onto filter discs. The acid-insoluble material (5% trichloroacetic acid) was counted using a liquid scintillation counter (9).
344
HILL
ET AL.
FluorescenceAnalysis Cells were suspendedin 0.5 ml phosphate-bufferedsaline containing EDTA (1 mM), sodium azide (0.02%) and 2% FCS and were analyzed using a fluorescence-activated cell sorter (FACStar Plus, Becton-Dickinson, Mountain View, CA). Five thousand DC were analyzed for each preparation. The cells were gated on the control DC population. DC were considered to be fluorescent when the mean fluorescence was above that of the control population. Levels of fluorescence and numbers of fluorescent DC are given as mean + standard error. Statistics Results were analyzed using analysis of variance and the Student t test. RESULTS Changes in Cell Numbers Mice skin painted with contact sensitizers such as FITC show a rapid increase in the number of DC in draining lymph nodes (2,3). In this system, mice were pretreated with either vehicle (AOO), oxazolone (Ox), or FITC on the shaved right flank and then challenged with FITC on the left flank at 5, 10, 20, or 40 days afterward and lymph nodes (LN) taken 3 days after challenge. Representative experiments for each time point are shown. After pretreatment with either AOO, Ox, or FITC followed by challenge with FITC there were large increasesin the number of lymph node cells (LNC) and DC isolated from draining LN (Figs. IA and IB) but there was no significant change in the ratio of LNC:DC or between the different pretreatments (AOO, Ox, or FITC; P > 0.05). If mice were pretreated and challenged with vehicle (AOO/DBP or DBP/DBP) the number of cells obtained was either equal to, or only slightly above, control levels. Acquisition of Antigen by Dendritic Cells DC isolated from draining LN 3 days after challenge were analyzed for fluorescence using flow cytometry. When the time between pretreatment and challenge was 5 days, animals pretreated with A00 before challenge with FITC had 29% (k2) fluorescent DC (Fig. 2A). Animals pretreated with Ox had 35% (k2) fluorescent DC whilst those pretreated with FITC had fewer fluorescent DC (2 1% + 1). The differences between the groups pretreated with either AOO, Ox, or FITC and then challenged with FITC were significant (P < 0.01). This was true for all comparisons between the different treatments. These figures are based on the mean values of 12 experiments with four animals in each group. The mean fluorescence of DC from these groups followed the same pattern, (AOO/FITC: 403 f 32, Ox/FITC: 471 f 29, FITC/FITC: 213 -t 19). The differences between the mean fluorescence values of the above groups were also significant between the different treatments (P < 0.001). When the time between pretreatment and challenge was extended to 10 days (Fig. 2B), DC from animals treated with Ox/FITC had numbers of fluorescent DC similar to those from AOO/FITC-treated mice (35% +- 3 compared to 38% + 7). The mean fluorescence was also similar (458 f 147 compared to 428 f 58). Mice treated with FITC/FITC had similar numbers of fluorescent DC compared with the groups above
DENDRITIC
50
CELLS IN NONRESF’ONSIVENESS
345
B
1
d b
40-
;
30-
x
20-
2 10 01
5 days
10 days
20 days
40 days
FIG. 1. Changes in the number of lymph node cells (LNC) (A) and dendritic cells (DC) (B) isolated from draining lymph nodes (challenge site). Animals were pretreated on the right flank with either vehicle (AOO), oxazolone, or FITC and at various times later challenged on the left flank with FWC. Draining lymph nodes were taken 3 days later. Lymph nodes were taken from four animals in each experiment. Each experiment was repeated at least 3 times (12 times at 5 days and 3 times at 10, 20, and 40 days) and the results of one representative experiment for each time point are shown. Time between pretreatment and challenge: 5, 10,20, q , OxlFITC; q , FITClFITC or 40 days. 0, control; q , AOO/FITC;
and the mean fluorescence was also lower (3 1% f 1 and 232 k 52, respectively). A similar pattern of results also occurred when the time between pretreatment and challenge was extended to 20 days (Fig. 2C). Similar results were obtained in at least three experiments at each time point (10 and 20 days). The differencesbetween the numbers in groups at each time point were not significant at the 5% level (P > 0.05). At 40 days between pretreatment and challenge there was no significant difference between the groups in either the percentage of fluorescent DC or the levels of antigen they expressed(Fig. 2D). T Cell Proliferation The changes in T cell proliferation caused by AINR were also examined. On separation, enriched T cells were cultured in vitro for 2 hr with [3H]thymidine and then harvested. Reduced T cell proliferation was found in animals treated with FITC/FITC (specific suppression) in all experiments where the time between pretreatment and challenge was up to 20 days (Figs. 3A-3C). Nonspecific suppression (i.e., pretreatment with Ox and challenge with FITC) was found in half the experiments at 5 days and intermittently at later times. By 40 days levels of T cell proliferation were the same in all treated groups of mice in all three experiments (Fig. 3D).
HILL ET AL.
346
150
-I
100 50
$ c' \ :' .,
./ 0 l?4!!a+ 0 12
"..A+, 3
4
FL (40) FIG. 2. Fluorescence on dendritic cells from draining lymph nodes. Mice were pretreated with either vehicle (AOO), oxazolone (Ox), or FITC. At various times later the animals were challenged on the left flank with FITC and the draining lymph nodes taken 3 days afterward. Time between pretreatment and challenge: A, 5 days; B, 10 days; C, 20 days; D, 40 days. -, control; -, AOO/RTC; ----, Ox/FITC, . . . . . , FITC/FITC.
The functional capacity of DC from treated and control mice was assessedby culturing the DC with naive T cells in 20 ~1 hanging drop cultures. In all experiments up to 20 days there was reduced stimulation of T cells by DC from animals treated with FITC/FITC (Figs. 4A-4C) and this may be due to lower levels of antigen on the DC. Nonspecific inhibition of proliferation was seen in some experiments at 5 days (Fig. 4A), but was not significant at later times (Figs. 4B-4D). At 40 days the ability of DC treated with FITC/FITC to stimulate normal T cells was significantly inhibited in 1 of 3 experiments. The time between pretreatment and challenge was altered to compare 5 and 3 days. At both time points specific inhibition was the most marked effect. Nonspecific suppression of T cell proliferation (Figs. 5A and 5B) was found intermittently at 3 days along with inhibition of DC stimulation of naive T cells (Figs. 6A and 6B). Reducing Time after Challenge before Assay The time after challenge was reduced to 1 or 2 days to compare with 3 days to see if the nonspecific suppression of T cell proliferation and DC stimulation of naive T cells were consistent phenomena. It was found that immediate T cell proliferation above background levels could only be detected at 3 days after challenge (data not shown). Stimulation of naive T cells by DC produced unexpected results. At 1 day after challenge the mean fluorescence on DC from animals treated with FITC/FITC was lower than that of DC treated with AOO/FITC or Ox/FITC as seen using cells 3
DENDRITIC CELLS IN NONRESPONSIVENESS
347
,B
50- A 40 30 20 10 -y 9 0 5
0 / 4-
:&g -D
6.25 12.5 25
50
100
6.25 12.5 25
50
100
T cells/well (x 1Om3)
FIG. 3 T. cells were isolated from mice pretreated with either vehicle (AOO), oxazolone (Ox), or FITC and challenged at various times afterward with FITC. Immediately after separation 6.25-100 X lo3 T cells were cultured with [‘Hlthymidine for 2 hr and then harvested. Results from one experiment for each time point are shown. Time between pretreatment and challenge: A, 5 days; B, 10 days; C, 20 days; D, 40 days. 0, control; 0, AOO/FITC; n , Ox/FITC, A, FITC/FITC.
days after challenge. There were equal percentagesof fluorescent DC in each group. However, the DC from FITC/FITC were significantly more stimulatory than the DC from the other groups (P < 0.01). As the time after challenge increased both the percentage of fluorescent DC and the mean fluorescence on DC from AOO/FITC and Ox/FITC increased peaking on Day 2 and then declining. However, the percentage of fluorescent DC decreased in the FITC/FITC group and the mean fluorescence remained similar to that of Day 1. These data agree with the study of Macatonia et al. (1987) (7). DISCUSSION Up to 20 days after pretreatment with FITC, subsequent challenge with the same antigen resulted in lower acquisition of antigen by DC in the draining lymph node. These cells also caused less stimulation of syngeneic T cells and this correlated with the presence of lower T cell activity occurring in the lymph node T cells in viva Previous work by Kimber et al. (5, 6, 10) has shown that AINR in contact sensitivity is characterized by inhibition of proliferative responsesand by lower levels of fluo-
HILL ET AL.
348
6.25 12.5 25
50
100
6.25 12.5 25
50
100
T cells/well (x 10m3) FIG. 4. Stimulation of syngeneic T cells by dendritic ceils exposed to antigen in viva. Animals were pretreated with either vehicle (AOO), oxazolone (Ox), or FITC and at various times later challenged with FITC. Draining lymph nodes were taken 3 days later. Dendritic cells were then cultured with naive T cells. Uptake of [3H]thymidine after 3 days in culture is shown. Time between pretreatment and challenge: A, 5 days; B, 10 days; C, 20 days; D, 40 days. 0, control T cells only; 0, Normal DC, W,AOO/RTC; A, Ox/ FITC; +, FITC/FITC.
rescenceon DC. This study confirms the idea that DC acquire lower levels of antigen following pretreatment and challenge with the same chemical. This appears to be a robust phenomenon since this study used a different sensitizing antigen, challenge site, and mouse strain from Kimber et al. (5, 6, 10). This study now shows evidence that the changed T cell function is a consequence of the changes in acquisition and presentation of antigen by the DC. At 5 days DC isolated from animals treated with FITC/FTTC did not stimulate naive T cells in vitro to the levels seen when T cells were cultured with DC from animals treated with AOO/FTTC. There was also specific inhibition of T cell proliferation in vitro on separation (which reflectsT cell proliferation in vivo) thus showing the link between the stimulatory capacity of the DC and the activity of the T cells. Previous studies by Macatonia et al. (3) have shown the relationship between the levels of antigen on DC and the stimulation of naive T cells in vitro. The low levels of antigen on DC from mice treated with FITQFITC suggest that this may be the causeof the low-level T cell proliferation. However, an exception to this was found when DC were isolated 1 day after challenge. These cells were more
DENDRITIC CELLS IN NONRESPONSIVENESS 60
1
A
349
lB
6.25 12.5 25
50
100
6.25 12.5 25
50
100
T cells/well (x 10-S) FIG. 5 T. cells were isolated from mice pretreated with either vehicle (AOO), oxazolone (Ox), or FITC and challenged with FITC, 5 or 3 days later. Immediately after separation 6.25-100 X 10’ T cells were cultured with [3H]thymidine for 2 hr and then harvested. Results from one experiment at each time point are shown. Time between pretreatment and challenge: A, 5 days; B, 3 days. 0, control; 0, AOO/FITC; n , Ox/FITC; A, FITC/FITC.
stimulatory with lower levels of antigen than DC isolated from mice pretreated with A00 or Ox and then challenged with FITC. The idea that DC dictate responsivenessdirectly in T cells in some antigen-specific manner is not a tenable hypothesis since antigenic specificity clearly resides in T cell heterogeneity. Preliminary evidence from experiments in nude, athymic CBA suggest that the block in acquisition of antigen and reduced stimulation of T cells in vitro does not occur. The effects shown here may therefore be dependent on the presence of T cells during sensitization.
30, A
6.i5 li.5
25
50
100
6.25 12.5 25
50
100
T cells/well (x 10-Z) FIG. 6. Stimulation of syngeneic T cells by dendritic cells exposed to antigen in viva. Mice were pretreated with either vehicle (AOO), oxazolone (Ox), or FITC and challenged 5 or 3 days later with FITC. Draining lymph nodes were taken 3 days after challenge. Uptake of [3H]thymidine after 3 days in culture is shown. Time between pretreatment and challenge: A, 5 days; B, 3 days. 0, control T cells only; 0, Normal DC; n , AOO/FITC; A, Ox/FITC; +, FITQFITC.
350
HILL ET AL.
After sensitization with either picryl chloride or oxazolone, IgM anti-hapten antibody was detected within 4 days (11). Macatonia (12) has shown that sensitization with FITC induced antibody production and this antibody could be detected up to 20 days after sensitization. The presence of antibody may explain the low levels of antigen on DC isolated at 5, 10, and 20 days from animals treated with FITC/FITC. Manta et al. (13) have reported that the addition of antibody when antigen is administered reduced the ability of accessorycells to present the antigen to antigen-specific T cell hybridoma clones. The presence of specific antibody on challenge may lead to the formation of antigen-antibody complexes. These complexes are thought to be removed either via macrophages by phagocytosis or by follicular dendritic cells (FDC) via the subcapsular sinus of the draining lymph nodes. The complexes are then transported to the germinal centers leading to B cell stimulation (14). These alternative pathways for the removal of antigen may account for the low levels of antigen seen on the lymphoid DC examined in this study. The nonspecific inhibition of proliferation in vivo and in vitro appearsto be a more complicated phenomenon. The reduced acquisition of antigen by DC seemsan unlikely explanation for the intermittent nonspecific inhibition of T cell proliferation since this occurred where the antigen levels on the autologous DC were at normal levels. Antibody to Ox cross-reacting with FITC is unlikely since Ox and FITC antigenically unrelated as they do not cross-stimulate (unpublished observation). There is some evidence to suggest that when doubly haptenated antigen-presenting cells are used to sensitize guinea pigs, antigenic competition results due to competition for the antigen-binding sites on these cells (15). However, animals sensitized 5 days previously do not have detectable antigen on DC isolated from contralateral lymph nodes (16). Since DC do not appear to migrate from lymph nodes (17) it seemsunlikely that competition for binding sites in this situation can explain the nonspecific inhibition of T cell proliferation in vivo and in vitro. There are reports that T cell clones can be inactivated by high levels of antigen (18, 19) even though these cells express IL-2 receptors and produce IL-2 and this may be involved in both the specific and nonspecific inhibition of T cell proliferation. An alternative explanation for the low T cell proliferation in vivo is T cell anergy. Rammensee et al. (20) have shown that exposure of adult Mls-lb mice to cells from Mls- 1a mice resulted in specific unresponsiveness when cells from Mls- 1b were stimulated by cells from Mls- 1a mice in a mixed lymphocyte reaction and this was thought to be due to clonal anergy. Experiments by Webb et al. (21) confirmed the findings of Rammensee et al. (20) and further showed that the specific nonresponsiveness waned after severalweeksand that there was also deletion of specific T cell populations. From the data presented here an analogy may be drawn with the results of Webb et al. (21) in that specific nonresponsiveness wanes after several weeks; however, it is not known if there is deletion of T cell populations in this system. The nonresponsivenessseen here may be a result of diminished ability of DC to interact with T cells leading to reduced proliferation. However, where the block on the interaction between these cells lies is unclear. The reduced capacity of DC and T cells to interact has also been seen in some diseasesas a result of infection with viruses or bacteria; both HIV infection and inflammatory arthritis show reduced DC and T cell function (22-24). The nonresponsiveness seen in this model of AINR using contact sensitizers may provide a method for examining the changes in DC and T cell interactions during infection.
DENDRITIC
CELLS IN NONRESPONSIVENESS
351
ACKNOWLEDGMENT S. Hill is supported by I.C.I.
REFERENCES 1. Silberberg-Sinakin,I., Thorbecke, G. J., Baer, R. L., Rosenthal, S. A., and Berezowsky,V., Cell. Immunol. 83, 137, 1976. 2. Knight, S. C., Krecjci, J., Malkovsky, M., Colizzi, V., Gautam, A., Asherson, G. L., Cell. Zmmunol. 94, 427, 1985. 3. Macatonia, S. E., Edwards, A. J., and Knight, S. C., Immunology 59, 509, 1986. 4. Nakano, Y., Immunology 33, 167, 1977. 5. Kimber, I., Pierce, B. B., Mitchell, J. A., and Kinnaird, A., Znt. Arch. Allergy Appl. Zmmunol. 84, 256, 1987. 6. Kimber, I., Hill, S., Mitchell, J. A., Peters, S. W., and Knight, S. C., Immunology 71, 271, 1990. 7. Macatonia, S. E., Knight, S. C., Edwards, A. J., Griffiths, S., and Fryer, P., J. Exp. Med. 166, 1654, 1987. 8. Julius, M. H., Simpson, E., and Herzenberg, L. A., Eur. J. Zmmunol. 3, 645, 1973. 9. O’Brien, J., Knight, S. C., Quick, N. A., Moore, E. H., and Platt, A. S., J. Zmmunol. Methods 27, 219, 1979. 10. Kimber, I., Shepherd, C. J., Mitchell, J. A., Turk, J. L., and Baker, D., Immunology 66, 577, 1989. 11. Asherson, G., Colizzi, V., and Watkins, M. C., Immunology 48, 561, 1983. 12. Macatonia, S. E., “Dendritic Cells as Initiators of Delayed-Type Hypersensitivity to Contact Sensitizers.” PhD. CNAA, 1987. 13. Manta, F., Fenoglio, D., Kunkl, A., Cambiaggi, C., Sasso,M., and Celada, F., J. Zmmunol. 140, 2893, 1988. 14. Szakal, A. K., Kosco, M. H., and Tew, J. G., Annu. Rev. Zmmunol. 7, 91, 1989. 15. Polak, L., Znt. Arch. Allergy Appl. Zmmunol. 76, 275, 1985. 16. Hill, S., Kimber, I., Edwards, A. J., and Knight, S. C., Immunology 71, 277, 1990. 17. Fossum, S., Stand. J. Zmmunol. 27, 97, 1988. 18. Suzuki, G., Kawase, Y., Koyasu, S., Yahara, I., Kobayashi, Y., and Schwartz, R., J. Zmmunol. 140, 1359, 1988. 19. Ceredig, R., and Corradin, G., Eur. J. Zmmunol. 16, 30, 1986. 20. Rammensee, H.-G., Kroschewski, R., and Frangoulis, B., Nature 339, 541, 1989. 2 1. Webb, S., Morris, C., and Sprent, J., Cell 63, 1249, 1990. 22. Macatonia, S. E., Patterson, S., and Knight, S. C., Immunology 67, 285, 1989. 23. Macatonia, S. E., Lau, R., Patterson, S., Pinching, A. J., and Knight, S. C., Immunology 71, 38, 1990. 24. Stagg,A. J., Harding, B., Hughes, R. A., Keat, A., and Knight, S. C., Clin. Exp. Zmmunol. 84,66, 1991.