- Email: [email protected]
T helper cell lines that augment in vivo cytotoxic T-cell responses to minor alloantigens
CELLULAR
IMMUNOLOGY
$3, 302-3 12 ( 1984)
T Helper Cell Lines That Augment in Viva Cytotoxic T-Cell Responses to Minor Alloantigens NICHOLAS
R. J. GASCOIGNE’
AND PATRICK
K. LAI
Imperial Cancer Research Fund, Tumour Immunology Unit. Department of Zoology. University College London, Gower Street, London WClE 6BT, England Received August IS, 1983; accepted September 22, 1983 We have investigated the ability of long-term cultured T helper (Th) cell lines to help an in vivo cytotoxic T lymphocyte (CTL) responseto non-H-2 alloantigens (minor antigens). Th cell lines specific for various single or undefined minor antigens were selectedby regular restimulation with antigen in vitro. They were antigen specific and H-2 restricted in proliferation assaysand were found to be able to help primary CTL responsesto multiple minor antigens and secondary GIL responsesto single minor antigens. Although the Th were antigen specific they did not determine the specificity of the CTL. Th cells were both necessaryand limiting for an effective CTL response indicating that “helper-independent” CTL are not in themselves sufficient to generate a strong in vivo response. Under conditions where a CTL response was clearly H-2 restricted, Th cells were not. Thus, the Th cells appeared to be activated by reprocessedantigen rather than antigen on the surface of the injected antigenic cells even though the CTL themselves reacted directly to the injected antigen. INTRODUCTION
The requirement for T helper (Th)2 cells in the generation of cytotoxic T lymphocytes (CTL) has been recognized for some time (I), although most of the data that have accumulated over the last few years, and the concepts based upon them (2) have come from in vitro studies. These experiments have shown that Ly 1+2,3- T cells react with antigen presenting cells (APC) in an antigen-specific manner restricted by Class II H-2 molecules to produce a T cell growth factor (interleukin (IL)-2) which is neither antigen specific nor H-2 restricted (1, 3-8). IL-2 provides a proliferation signal for precursor CTL (CTLp), but is not sufficient in itself to induce CTL activity. The CTLp require direct contact with antigen (9) and stimulation by another soluble differentiation factor or factors (10, 11). The function of factors in vivo is not clear since IL2 at least is cleared from the serum very quickly ( 12, 13). There is uncertainty whether the requirement for Th cells in the CTL response is absolute, since some CTL clones themselves produce enough growth factors to act independently of extraneous factors in short-term proliferation assaysin vitro (1416). Such cells are restricted by ClassI molecules for both proliferation and cytotoxicity ’ Present address:Department of Medical Microbiology, Fairchild Building, Stanford University School of Medicine, Stanford, Calif. 94305. * Abbreviations used: Th, T helper cells; CTL, cytotoxic T lymphocytes; CTLp, CTL precursor; IL-2, interleukin-2; APC, antigen presenting cells; FCS, fetal calf serum; CS, calf serum; LU, lytic units; MEM, minimal essential medium; Con A, concanavalin A, [‘251]Udr, [‘251]iodo-2’deoxyuridine. 302 0008-8749184$3.00 Copyri&l 0 1984 by Academic Press. Inc. All rights of reproduction in any fom reserved.
T HELPER
CELL
LINES
FOR IN
VW0
CTL
303
RESPONSES
(15). Moreover, in some systems Class II-restricted Th cells are not needed for the induction of CTL activity, but growth factor producing “helper” cells restricted by Class I molecules are necessary(whether or not these factor producing cells are themselves cytotoxic has not been determined) (17). Since helper cell-independent CTL exist, and since they have been shown to be active in vivu under conditions where IG2dependent CTL are not (1S), the importance of Th cells for CTL responsesin vivo has been called into doubt. The data supporting the importance of their role come from indirect measurements of Th cell activity, generally relying on restimulation with antigen in vitro after in vivo priming (6-8, 17, 19-24). To assessthe in vivo requirement for Th cells directly we have adapted the footpad priming method for generating primary CTL (25), which we have previously used to characterize CTL responsesto multiple minor non-H-2 alloantigens (minor antigens) (26). We have transferred long-term cultured antigen-specific Th cells plus antigen into normal recipient mice where they enhance a primary CTL response to multiple minor antigens or into primed mice where they help a secondary response to single minor antigens. Evidence is presented that Th cells are both necessary and limiting in the response and that they are activated by reprocessedantigen presented by the recipient’s APC as well as by the APC in the inoculum. MATERIALS
AND METHODS
Mice. C57BL/lOScSn (BlO), BlO.D2, (BlO X B1O.DZ)F, (BDFi), BlO.BR, CBA, and BALB/c were obtained from ICRF breeding colony, Mill HilI and BALB.B and BALB.K from OLAC (1976) Ltd., Bicester, Oxon, England. Breeding pairs of C57BL/ 6 By (B6), B6.CH-25”By (synonym HW65), and B6.C-H-7bBy (synonym HW23) were supplied by Jackson Laboratories, Bar Harbor, Maine, and maintained at University College. Table 1 shows the major (H-2) and minor alloantigen genotypes of the mice used. The H-7b allele of B6.C-H-7b and the H-25” allele of B6.CH-25” were derived from BALB/c By (27). BlO and B6 are closely related strains which we assume to be the same for the purpose of these experiments. Media. T helper cell lines were cultured in Iscove’s modified Dulbecco medium
TABLE
1
Genotypes of Mice Used
BlO/B6 BlO.D2 BDF, BlO.BR B6.C-H-7’ B6.C-H-25 c (B6.C-H-25 c X BALB/c)FI BALB/c BALB.B BALB.K
H-2
H-7
H-25
b d b/d k b b b/d d b k
a a a a b a a/b b b b
b b b b b C C C C C
Derivation
of other genes
BlO/B6 BlO BlO BlO B6 B6 BB/BALB BALB BALB BALB
304
GASCOIGNE
AND
LA1
(Gibco) with 10% fetal calf serum (FCS) (G&co), 1 mM glutamine, penicillin, and streptomycin as described (P. K. Lai and T. Owens, submitted). IGZcontaining media were not added. Short-term cultures were performed as described (26, 28) in RPMI-1640 (ICRF) plus 5% FCS, 5 X low5 M 2-mercaptoethanol, and other sup plements as above (complete RPMI). Washing of cell suspensions and assay of CTL were performed in Eagle’s minimal essential medium (MEM) with or without 5% calf serum (CS) (ICRF). All cultures were performed at 37°C in a water-saturated atmosphere of 5% CO* in air. T helper cell lines. Antigen-specific Th cell lines were generated from BDF, mice against single (BDFI anti-B6.CH-25” or B6.C-H-7b) or undefined (BDFI anti-BALB.B) minor antigens in association with H-2b by priming in vivo followed by regular restimulation with irradiated antigen in vitro as described in the accompanying paper. These cell lines are able to help an antibody response in vivo and have no cytotoxic activity (P. K. Lai and T. Owens, submitted). They were used in proliferation and help experiments at least 1 week after restimulation. They are Ly1+2- T cells as judged by conventional immunofluorescence. Generation of cytotoxic T cells. We have previously described (26) an adaptation of the method of Starzinski-Powitz et al. (25) to the study of the primary CTL response to minor antigens. Briefly, mice were primed to minor antigens with 5 X lo6 or lo7 irradiated (2000 R) spleen cells per hind footpad. Cultured Th cells were mixed with the antigen prior to injection (2 x lo4 to 2 X lo5 per footpad). Controls were injected with either antigen alone or with normal BDFl lymph node cells plus antigen. Four days later the popliteal and inguinal lymph nodes were removed aseptically and single cell suspensions made. Cells were cultured at 4 X lo5 per well in 0.2 ml complete RPM1 in 96-well round bottom microtitre tissue culture plates (Titertek, Flow, Irvine, UK) without antigen or additional growth factors (unless otherwise stated). Equal numbers of wells were cultured for each group and after 3 days they were pooled, washed, and used as effecters in a cytotoxicity assay. Recovery of viable cells from culture was generally 30-50%. Cytotoxicity assay. Pooled and washed effector cells were titrated in threefold dilutions in duplicate against a fixed number of target cells. The target cells were concanavalin A (Con A Sigma) induced blast cells labelled with “Cr (sodium chromate, Amersham International) as described (26, 28). The assay was performed in MEM plus 5% CS in round-bottom 96-well plates for 3.5 hr at 37°C. Supematant (100 ~1) from each well was removed to count released 51Cr. Percentage specific lysis was determined by the formula % specific lysis = (E - S)/( T - S) X 100 where E is cpm released into experimental wells, S is spontaneous release, and T is the number of cpm released by detergent (1% Brij). Spontaneous release was always below 30% of total. Results are presented as titration of percentage specific lysis against the fraction of total culture or as lytic units (LU30) per culture. LUso is the reciprocal of the culture fraction producing 30% specific lysis. Proliferation assay. Th cells (l-4 X 104) were cultured with 5 X lo5 irradiated (3000 R) stimulator cells in 0.2 ml medium in flat-bottom 96-well plates for 48 hr. They were then pulsed with 0.5 &i 5-[‘251]iodo-2’deoxyuridine ([‘251]Udr) (Amersham
T HELPER
CELL
LINES
FOR IN VW0
CTL
305
RESPONSES
International) in 10 ~1medium for 6 hr and harvested on an automated cell harvester (Dynatech). DNA incorporation was assessedby gamma counting and the results expressedas cpm [i2’I]Udr uptake f SD. Con A supernatant. Con A supematant was produced by culturing CBA spleen cells (107/ml) in complete RPM1 containing 2 pg/ml Con A for 24 hr. Cells were removed by centrifugation and Con A was neutralized with 25 mMa-methylmannoside (Sigma). The Con A supematant was used at 5% v/v. RESULTS H-2 Restriction and Antigen Specificity of the Proliferation of T Helper Cell Lines T helper (Th) cell lines were produced by repeatedly restimulating primed female BDF, (H-2b X H-2d) spleen or lymph node cells with irradiated antigen in vitro (P. K. Lai and T. Owens, submitted). The antigen was always presented on H-2b (BALB.B) cells so that it was expected that the selected Th cells would be restricted by H-2b and not by H-2d. Table 2 demonstrates that the Th cell lines were indeed H-2b restricted in their proliferation and that they were also specific for the minor antigens to which they had been raised. The anti-H-7b line proliferated strongly only to strains carrying both H-7b and H-2b (B6.C-H-7b and BALB.B) and the anti-H-25” line only to strains carrying both H-25’ and H-2b (B6.C-H-25”, BALB.B, and [B6.C-H25” X BALB/c]F,). The anti-H-25” line did not respond to the male specific antigen H-Y (BDF, male). Neither line responded to allo-H-2 stimuli. Help for a Primary CTL Response to Multiple Minor Antigens Preliminary experiments showed that whereas the primary CTL responseof BALB background mice to minor antigens of the B 10 background (e.g.,BALB/c anti-B 1O.D2) were strong (26) the reciprocal combinations (BlO or BDFl anti-BALB.B and BlO.D2 TABLE 2 Proliferation
of T Helper Cell Lines” Cell line
Stimulator
Anti-H-7 b
Anti-H-25 c
-
281 (29) 408 (305) 2751 (1657) nd 3953 (2154) 300 ( 148) nd 227 (43) nd 529 (145)
307 (22) 471 (83) 2517 (193) 295 (23) nd 3728 (160) 3216 (116) nd 362 (22) nd
BALB/c BALB.B BALB.K B6.C-H7 b B6.C-H-25’ (B6.C-H-25’ X BALB/c)F, BIO.BR BDF, male BDF, female
’ 1 X lo4 (anti-H-f) or 4 X 10’ (anti-H-25’) Th cells were cultured for 48 hr with 5 X lo5 irradiated (3000 R) stimulator cells in 0.2 ml medium in 96-well flat bottom plates. Results are expressed as cpm [‘*‘I]Udr uptake (&SD).
306
GASCOIGNE AND LA1 TABLE 3 Help for a Primary CTL Response to Multiple Minor Antiget& LU30 per culture Th cell line transferred
Target: BALB.B
BIO
-
0.5
0.0
Anti-BALB.B Anti-H-7 * Anti-H-25 c
33.3 39.2 23.8
0.1 0.0 0.9
a BDF, mice were injected in the hind footpads with BALB.B spleen cells plus 4 X lo5 Th cells. The CTL obtained were assayed on BALB.B or BlO Con A blast target cells. LUs,, is the reciprccal of the culture fraction producing 30% specific lysis.
or BDF, anti-BALB/c) were not. The responses of BDFi mice could be greatly augmented by prepriming several weeks earlier (Czitrom and Gascoigne, unpublished observations). To discover if this low responsiveness was due to a low frequency of Th cells or alternatively of CTLp we tested the ability of the three Th cell lines to augment a CTL response to the minor antigens of BALB.B (Table 3). All three cell lines proved able to help the primary CTL response to BALB.B cells since the Th were all reactive with minor antigens expressed on the BALB.B cell surface. This suggested that the low primary response in this combination was due to a lack of Th cells rather than of CTLp. It was of interest to find out whether Th cells responding to a single minor antigen expressed on the BALB.B cell surface caused the preferential induction of CTL responding to that particular antigen. It is clear from Fig. 1 that they did not. When injected at the same time as BALB.B
0
I RECIPROCALOF CULTURE FRACTION
FIG. 1. Th cells against H-2Y do not preferentially induce CTL to H-2jc in a primary response against BALB.B cells. BDFr mice were primed with BALB.B cells with (solid symbols) or without (open symbols) anti-H-25’ Th cells. The CTL response was assayed on a variety of Con A blast target cells: BALB.B (A A, A A); B6.C-H-25’ (0 l,O 0); B6.C-H-7b (v -V) or B6 (Wn).
T HELPER CELL LINES FOR IN VW0
CTL RESPONSES
307
antigen, a&H-25” Th cells helped a response to minor antigens expressed by BALB.B but not by B6.C-H-25” or B6.C-H-7b. The CTL response was therefore directed against minor antigens other than H-25” or H-7b. In other words, there was not a detectable bias of the response toward the antigen recognized by the Th cells. Help for a Secondary CTL Response to H-25’
To check that Th cells were able to induce CTL to the antigen against which they were raised, BDF, mice were primed to B6.CH-25” spleen cells which were injected into the footpad. One month later they were boosted with the same antigen mixed with anti-H-25” Th cells. Unprimed BDF, mice were also injected with antigen with or without Th cells. Figure 2 demonstrates that the secondary CTL response was specific to H-25”. B6.C-H-25” and BALB.B targets were killed to approximately the same degree, but B6 and B6.C-H-7b only marginally. That these two targets were killed at all may indicate that a small proportion of the response was directed against shared viral antigens. Figure 2 also shows that there was no primary response to B6.G H-25” either with or without Th cells indicating that anti-H-25C killing was detectable only after prepriming with this antigen. H-2 Restriction and Dependence
of Th Cells on Host Antigen Reprocessing
in Vivo
The Th cells used in this study have been shown to be antigen specific and restricted to H-2b for proliferation (Table 2). To test the restriction of their in vivo helper activity, BDF, mice were injected with 10’ irradiated BALB.B or BALB/c spleen cells plus 3.5 X lo4 anti-H-7b Th cells or normal BDF, lymph node cells per hind footpad. An additional control group was injected with Th cells alone. The effect of adding Con A supematant at the in vitro culture stage was also tested. As Fig. 3 shows, the supposedly restricted Th cells appeared to help the primary response to BALB minor
I
.
81
21
9
3
RECIPROCAL OF CULTURE FIVLTI~N
FIG. 2. Specificity of a secondary CTL response to B6.C-H-25’. BDF, mice were primed 1 month previously with B6.C-H-25C spleen cells were challenged with B6.CH-25’ spleen cells plus a&H-25’ Th cells. The CTL response was assayed on various target cells: BALB.B (A A); B6.C-H-25C (0 0); B6 (D n); B6.C-H-7b (7 7). A A represents normal BDF, mice injected with B6.C-H-25’ spleen cells plus anti-H-25’ Th cells and assayed on B6.C-H-25C target cells. 0 0 is the same but without Th cells.
308
GASCOIGNE AND LA1 a
60
,
243
81
21
9
In
3
RECIPROCAL OF CULTURE FRACTION
3 243 81 21 9 RECIPROCAL OF CULTURE FRACTION
FIG. 3. Lack of H-2 restriction of Th cells in viva and the effect of adding Con A supematant in vitro. BDFr mice were injected with BALB.B or BALB/c spleen cells plus anti-H-7b Th cells or normal BDF, lymph node cells. Four days later each of these groups were cultured either alone or in the presence of IL 2 (5% Con A supematant) for a further 3 days before assay of CTL on BALB.B (a), BALB/c (b), or BDF, target cells. BALB.B + Th (A - - - A); BALB.B + Th + IL-2 (0 - - - 0); BALB.B + lymph node cells BALB/c + Th (m - - - w); BALB.B + lymph node cells + IL-2 (V - - - V); BALB/c + Th (A-A); + IL-2 (0 0); BALB/c + lymph node cells (0 0); BALB/c + lymph node cells + IL-2 (V V). An additional group was injected with Th cells alone (Cl). None of the groups gave significant killing on BDF, target cells.
antigens in a completely unrestricted manner, while the CTL response itself was H-2 restricted to the antigen with which the mice were primed. Th cells transferred alone did not generate any CTL response. Similar results have been obtained with anti-BALB.B Th cells. We interpret these results to mean that reprocessing of the antigen by the F1 host was sufficient to mask any effects of H-2 restriction of the Th. It was not possible to obtain meaningful data on the activity of these Th in parental (BlO or BlO.D2) mice due to reactions against the allo-H-2 carried by the F1 Th (unpublished data). Addition of T cell growth factors (in the form of Con A supematant) to the culture stage of the experiment had the effect of increasing the magnitude of the CTL response in groups both with and without Th. The Con A supematant was not in itself sufficient to raise the response without Th cells to that level obtained with Th cells. Further evidence for the importance of antigen processing in this system is provided in Fig. 4. BDFl female mice were primed in each hind footpad with 10’ BlO.D2 male spleen cells. Five weeks later they were challenged with 10’ B6.C-H-7b male or female, B 10 male, or a mixture of 10’ B6.C-H-7b female plus 10’ B 10 male spleen cells with or without anti-H-7b Th cells. Unprimed mice do not generate responses to H-Y in this system. A secondary CTL response to H-Y was generated only when the mice primed were injected with both H-7 incompatible cells and cells expressing the male antigen. These antigens, however, did not need to be expressedon the same cell, as might have been expected from previous reports of Th-CTL interactions (23, 24,29), again suggestingthat antigen processing by host APCs is an important factor. Alternatively, it is possible that in the limited environment of the lymph nodes, sufficient concentrations of growth factors were made to overcome the usual control mechanisms (12, 13). The antigen specificity of the Th cells is demonstrated by the
T HELPERCELL LINES FOR IN
VZVO
CTL RESPONSES
309
a
243 81 21 9 3 RECIPROCALOF CULTUREFRACTION
RECIPROCALOF CULTUREFRACTION
RECIPROCALOF CULTUREFRACTION
FIG. 4. Th cells are antigen specific but can help a CTL response to another antigen presented at the same time. BDF, femalemicewereprimedwith BIO.D2malespleencells.Fiveweekslatertheywere challenged with various antigens or mixtures of antigens with or without anti-H-7b Th cells. The CTL responseswere measured on BIO male (a), BlO female (b), and B6.C-H-7b female (c) target cells. The antigens were BlO male alone (0 -Cl), BlO male plus Th (m n); B6.C-H-7b male alone (0 - 0), B6.C-H-7b male plus Th (@0); B6.C-H-7b female plus Th (A A) or B6.C-H-7b female plus BlO male plus Th (A A).
finding that they did not augment the CTL response against H-Y when presented with H-Y alone, but did when presented with H-7b plus H-Y. This experiment also showsthat helper cells were limiting in a secondary responseto a single minor antigen, as was found in the primary response to multiple minor antigens, even though there was clearly no shortage of CTLp. DISCUSSION We have described a novel system for the study of help in in vivo CTL responses. Previous work has, with few exceptions (30), used a protocol of priming in vivo
310
GASCOIGNE
AND
LA1
followed by restimulation with antigen in vitro (6-8, 17, 19-24). The method employed here makes useof cultured Th cell lines, measuring their activity in vivo in the primary (26) or secondary response to minor antigens. These cell lines are H-2 restricted, Ly lf2- T cells which have been shown to function as helpers for antibody responses (P. K. Lai and T. Owens, submitted). To assayhelp for CTL they were injected with antigen into the hind footpads of mice. Four days later the draining lymph nodes were cultured without antigen or added growth factors and after another 3 days CTL activity was measured. Two main findings are reported here. We have shown first that it is a shortage of Th cells rather than CTLp that results in low primary and secondary in vivo responses to minor antigens; and second that the Th cells are activated not simply by APC in the spleen cells used to immunize, but also by reprocessed antigen on APC of the host animal. The evidence for the first point is that what was normally a very low primary (Table 3) or secondary (Figs. 2 and 4) responseto multiple or single minor antigens, respectively, was greatly augmented by the transfer of extra Th cells. It is possible that the extra Th cells were necessaryto overcome the effect of suppressor cells in the lymph nodes (31) though there is other evidence to support the view that help is the limiting factor: the injection of cells allogeneic at the Mls locus, which elicits strong IL-2 production (32), can facilitate a response (33) as can injection of partially purified IL-2 directly into the footpads (34). Our data on the secondary response (Figs. 2 and 4), however, suggest that CTLp as well as Th cells are limiting for a response to single minor antigens since the Th cells could not induce CTL in mice that were not previously primed to the relevant antigen (Fig. 2). Thus, even when the frequency of CTLp was increased by priming (< l/ 100,000 to l/3500 for H-Y (35)), Th cells were limiting for the CTL response. We have not determined for certain the time at which Th cells act, but we believe for two reasons that it is mainly prior to the in vitro phase. The first reason is that the addition of T cell growth factors (including IL-2) in vitro to cells from mice primed without Th cells does not increase CTL responsesup to the values obtained with antigen plus Th cells (Fig. 3). Although the growth factors have a small effect on groups primed both with and without Th cells, it is clear that the Th cannot be substituted for in vitro with Con A supernatant, which should contain all the factors necessaryto enable CTL differentiation (11, 12). The second is that in the strong primary CTL response of BALB/c mice against B 10.D2 minor antigens, IL2 production by the lymph node cells occurs in a burst from Day 3 to Day 4, declining very sharply by the fifth day (N. R. J. Gascoigne, unpublished data). This suggests that Th cell activity peaksjust before the stageat which the lymph nodes are cultured. We have shown that the H-2 heterozygous Th cells are H-2 restricted in their proliferation (P. K. Lai and T. Owens, submitted and Table 2) and yet when injected back into syngeneic F, mice their helper activity for an H-2 restricted CTL response was unrestricted (Fig. 3). One possible explanation for this is that the Th cells were unrestricted in their activation by antigen in vivo, however, this Seemsvery unlikely in view of the specificity shown in the proliferation assay. We suggestthat when H2d associated antigen was injected into BDF, mice it associated with host (H-2b’d) APC so that the Th cells could recognize it in association with H-2b. Such an interpretation is supported by the data shown in Fig. 4 where effective help occurred when the antigens that the Th cells and the CTL respond to were presented on different
312
GASCOIGNE
AND
LA1
7. Simon, M. M., Edwards, A. J., Hlmmerling, U., McKenzie, I. F. C., Eichmann, K., and Simpson, E., Eur. J. Immunol. 11, 246, 1981. 8. Michaelides, M., Sandtin, M., Morgan, G., McKenzie, I. F. C., Ashman, R., and Melvold, R. W., J. Exp. Med. 153,464, 1981. 9. Weinberger, O., Herrmann, S. H., Mescher, M. F., Benacerraf, B., and Burakoff, S. J., Eur. J. Immunol. 11, 405, 1981.
10. Wagner, H., Hardt, C., Rouse, B. T., RoBinghoff, M., Scheurich, P., and Phzenmaier, K., J. Exp. Med. 155, 1876, 1982.
11. Raulet, D. H., and Bevan, M. J., Nature (London) 296, 754, 1982. 12. Hardt, C. Rollinghoff, M., Plizenmaier, K., Moosmann, H., and Wagner, H., J. Exp. Med. 154, 2, 1981.
13. Miihlradt, P. F., and Gpitz, H. G., Eur. J. Immunol. 12, 983, 1982. 14. Widmer, M. B., and Bach, F. H., Nature (London) 294, 750, 1981. 15. Roopenian, D. C., Widmer, M. B., Orosz, C. G., and Bach, F. H., J. Immunol. 130, 542, 1983. 16. von Boehmer, H., and Turton, K., Eur. J. Immunol. 13,750, 1983. 17. Raulet, D. H., and Bevan, M. J., J. Exp. Med. 155, 1766, 1982. 18. Engers, H. D., Glasebrook, A. L., and Sorenson, G. D., J. Exp. Med. 156, 1280, 1982. 19. Zinkemagel, R. M., Callahan, G. N., Althage, A., Cooper, S., Streilein, J. W., and Klein, J., J. Exp. Med. 147, 897, 1978.
20. von Boehmer, H., and Haas, W., J. Exp. Med. 150, 1134, 1979. 21. Hurme, M., Hetherington, C. M., Chandler, P. R., Gordon, R. D., and Simpson, E., Immunogenetics 5, 453, 1977. 22. Hurme, M., Hetherington, C. M., Chandler, P. R., and Simpson, E., J. Exp. Med. 147, 758, 1978. 23. Keene, J., and Forman, J., J. Exp. Med. 155, 768, 1982. 24. Kanagawa, O., Louis, J. A., and Cerottini, J-C., In “Mechanisms of Lymphocyte Activation” (K. Reach
and H. Kirchner, Eds.), p. 404. Elsevier/North Holland Biomedical Press,Amsterdam, 1981. 25. Starzinski-Powitz, A., Pfizenmaier, K., Rollinghoff, M., and Wagner, H., Eur. J. Immunol. 6, 799, 1976. 26. Czitrom, A. A., and Gascoigne, N. R. J., Immunology 27. Bailey, D. W., Immunogenetics 2, 249, 1975.
SO, 121, 1983.
28. Lai, P. K., Waterfteld, J. D., Gascoigne,N. R. J., Shamrock,C. E. M., and Mitchison, N. A., Immunology 47, 371, 1982. 29. Mitchison, N. A., Transplant Proc. (in press), 1983. 30. Baum, L. L., and Pilarski, L. M., Trunsplunfution 32, 409, 1981.
31. Wagner, H., Starzinski-Pow&z, A., Pfizenmaier, K., and RoBinghoff, M., Eur. J. Zmmunol. 6, 873, 1976. 32. Rollinghoff, M., Ptizenmaier, K., Solbach, W., and Wagner, H., Be/wing Inst. Mitt. 67, 36, 1980. 33. Butler, L. D., and Battisto, J. R., J. Immunol. 122, 1578, 1979. 34. Wagner, H., Hardt, C., Heeg, K., Pfizenmaier, K., St&&r, H., and Rohinghoff, M., Immunobiofogy 161, 139, 1982.
35. Kanagawa, O., Louis, J., and Cerottini, J-C., J. Immunol. 128, 2362, 1982. 36. Matzinger, P., and Bevan, M. J., Cell Immunol. 33, 92, 1977. 37. Weinberger, O., Herrmann, S. H., Mescher, M. F., Benacerraf, B., and Burakoff, S. J., Proc. Nut. Acud. Sci. USA 78, 1796, 1981. 38. Rollinghoff, M., Pfizenmaier, K., and Wagner, H. Eur. J. Immunol. 12, 337, 1982. 39. Czitrom, A. A., Sunshine, G. H., Reme, T., Ceredig, R., Glasebrook, A. L., Kelso, A., and MacDonald, H. R., J. Immunol. 130, 546, 1983. 40. Mitchison, N. A., Transplant Proc. 2, 92, 1970. 41. Ahman, A., Bechtold, T. E., and Katz, D. H., J. Immunol. 122, 2484, 1979.
IMMUNOLOGY
$3, 302-3 12 ( 1984)
T Helper Cell Lines That Augment in Viva Cytotoxic T-Cell Responses to Minor Alloantigens NICHOLAS
R. J. GASCOIGNE’
AND PATRICK
K. LAI
Imperial Cancer Research Fund, Tumour Immunology Unit. Department of Zoology. University College London, Gower Street, London WClE 6BT, England Received August IS, 1983; accepted September 22, 1983 We have investigated the ability of long-term cultured T helper (Th) cell lines to help an in vivo cytotoxic T lymphocyte (CTL) responseto non-H-2 alloantigens (minor antigens). Th cell lines specific for various single or undefined minor antigens were selectedby regular restimulation with antigen in vitro. They were antigen specific and H-2 restricted in proliferation assaysand were found to be able to help primary CTL responsesto multiple minor antigens and secondary GIL responsesto single minor antigens. Although the Th were antigen specific they did not determine the specificity of the CTL. Th cells were both necessaryand limiting for an effective CTL response indicating that “helper-independent” CTL are not in themselves sufficient to generate a strong in vivo response. Under conditions where a CTL response was clearly H-2 restricted, Th cells were not. Thus, the Th cells appeared to be activated by reprocessedantigen rather than antigen on the surface of the injected antigenic cells even though the CTL themselves reacted directly to the injected antigen. INTRODUCTION
The requirement for T helper (Th)2 cells in the generation of cytotoxic T lymphocytes (CTL) has been recognized for some time (I), although most of the data that have accumulated over the last few years, and the concepts based upon them (2) have come from in vitro studies. These experiments have shown that Ly 1+2,3- T cells react with antigen presenting cells (APC) in an antigen-specific manner restricted by Class II H-2 molecules to produce a T cell growth factor (interleukin (IL)-2) which is neither antigen specific nor H-2 restricted (1, 3-8). IL-2 provides a proliferation signal for precursor CTL (CTLp), but is not sufficient in itself to induce CTL activity. The CTLp require direct contact with antigen (9) and stimulation by another soluble differentiation factor or factors (10, 11). The function of factors in vivo is not clear since IL2 at least is cleared from the serum very quickly ( 12, 13). There is uncertainty whether the requirement for Th cells in the CTL response is absolute, since some CTL clones themselves produce enough growth factors to act independently of extraneous factors in short-term proliferation assaysin vitro (1416). Such cells are restricted by ClassI molecules for both proliferation and cytotoxicity ’ Present address:Department of Medical Microbiology, Fairchild Building, Stanford University School of Medicine, Stanford, Calif. 94305. * Abbreviations used: Th, T helper cells; CTL, cytotoxic T lymphocytes; CTLp, CTL precursor; IL-2, interleukin-2; APC, antigen presenting cells; FCS, fetal calf serum; CS, calf serum; LU, lytic units; MEM, minimal essential medium; Con A, concanavalin A, [‘251]Udr, [‘251]iodo-2’deoxyuridine. 302 0008-8749184$3.00 Copyri&l 0 1984 by Academic Press. Inc. All rights of reproduction in any fom reserved.
T HELPER
CELL
LINES
FOR IN
VW0
CTL
303
RESPONSES
(15). Moreover, in some systems Class II-restricted Th cells are not needed for the induction of CTL activity, but growth factor producing “helper” cells restricted by Class I molecules are necessary(whether or not these factor producing cells are themselves cytotoxic has not been determined) (17). Since helper cell-independent CTL exist, and since they have been shown to be active in vivu under conditions where IG2dependent CTL are not (1S), the importance of Th cells for CTL responsesin vivo has been called into doubt. The data supporting the importance of their role come from indirect measurements of Th cell activity, generally relying on restimulation with antigen in vitro after in vivo priming (6-8, 17, 19-24). To assessthe in vivo requirement for Th cells directly we have adapted the footpad priming method for generating primary CTL (25), which we have previously used to characterize CTL responsesto multiple minor non-H-2 alloantigens (minor antigens) (26). We have transferred long-term cultured antigen-specific Th cells plus antigen into normal recipient mice where they enhance a primary CTL response to multiple minor antigens or into primed mice where they help a secondary response to single minor antigens. Evidence is presented that Th cells are both necessary and limiting in the response and that they are activated by reprocessedantigen presented by the recipient’s APC as well as by the APC in the inoculum. MATERIALS
AND METHODS
Mice. C57BL/lOScSn (BlO), BlO.D2, (BlO X B1O.DZ)F, (BDFi), BlO.BR, CBA, and BALB/c were obtained from ICRF breeding colony, Mill HilI and BALB.B and BALB.K from OLAC (1976) Ltd., Bicester, Oxon, England. Breeding pairs of C57BL/ 6 By (B6), B6.CH-25”By (synonym HW65), and B6.C-H-7bBy (synonym HW23) were supplied by Jackson Laboratories, Bar Harbor, Maine, and maintained at University College. Table 1 shows the major (H-2) and minor alloantigen genotypes of the mice used. The H-7b allele of B6.C-H-7b and the H-25” allele of B6.CH-25” were derived from BALB/c By (27). BlO and B6 are closely related strains which we assume to be the same for the purpose of these experiments. Media. T helper cell lines were cultured in Iscove’s modified Dulbecco medium
TABLE
1
Genotypes of Mice Used
BlO/B6 BlO.D2 BDF, BlO.BR B6.C-H-7’ B6.C-H-25 c (B6.C-H-25 c X BALB/c)FI BALB/c BALB.B BALB.K
H-2
H-7
H-25
b d b/d k b b b/d d b k
a a a a b a a/b b b b
b b b b b C C C C C
Derivation
of other genes
BlO/B6 BlO BlO BlO B6 B6 BB/BALB BALB BALB BALB
304
GASCOIGNE
AND
LA1
(Gibco) with 10% fetal calf serum (FCS) (G&co), 1 mM glutamine, penicillin, and streptomycin as described (P. K. Lai and T. Owens, submitted). IGZcontaining media were not added. Short-term cultures were performed as described (26, 28) in RPMI-1640 (ICRF) plus 5% FCS, 5 X low5 M 2-mercaptoethanol, and other sup plements as above (complete RPMI). Washing of cell suspensions and assay of CTL were performed in Eagle’s minimal essential medium (MEM) with or without 5% calf serum (CS) (ICRF). All cultures were performed at 37°C in a water-saturated atmosphere of 5% CO* in air. T helper cell lines. Antigen-specific Th cell lines were generated from BDF, mice against single (BDFI anti-B6.CH-25” or B6.C-H-7b) or undefined (BDFI anti-BALB.B) minor antigens in association with H-2b by priming in vivo followed by regular restimulation with irradiated antigen in vitro as described in the accompanying paper. These cell lines are able to help an antibody response in vivo and have no cytotoxic activity (P. K. Lai and T. Owens, submitted). They were used in proliferation and help experiments at least 1 week after restimulation. They are Ly1+2- T cells as judged by conventional immunofluorescence. Generation of cytotoxic T cells. We have previously described (26) an adaptation of the method of Starzinski-Powitz et al. (25) to the study of the primary CTL response to minor antigens. Briefly, mice were primed to minor antigens with 5 X lo6 or lo7 irradiated (2000 R) spleen cells per hind footpad. Cultured Th cells were mixed with the antigen prior to injection (2 x lo4 to 2 X lo5 per footpad). Controls were injected with either antigen alone or with normal BDFl lymph node cells plus antigen. Four days later the popliteal and inguinal lymph nodes were removed aseptically and single cell suspensions made. Cells were cultured at 4 X lo5 per well in 0.2 ml complete RPM1 in 96-well round bottom microtitre tissue culture plates (Titertek, Flow, Irvine, UK) without antigen or additional growth factors (unless otherwise stated). Equal numbers of wells were cultured for each group and after 3 days they were pooled, washed, and used as effecters in a cytotoxicity assay. Recovery of viable cells from culture was generally 30-50%. Cytotoxicity assay. Pooled and washed effector cells were titrated in threefold dilutions in duplicate against a fixed number of target cells. The target cells were concanavalin A (Con A Sigma) induced blast cells labelled with “Cr (sodium chromate, Amersham International) as described (26, 28). The assay was performed in MEM plus 5% CS in round-bottom 96-well plates for 3.5 hr at 37°C. Supematant (100 ~1) from each well was removed to count released 51Cr. Percentage specific lysis was determined by the formula % specific lysis = (E - S)/( T - S) X 100 where E is cpm released into experimental wells, S is spontaneous release, and T is the number of cpm released by detergent (1% Brij). Spontaneous release was always below 30% of total. Results are presented as titration of percentage specific lysis against the fraction of total culture or as lytic units (LU30) per culture. LUso is the reciprocal of the culture fraction producing 30% specific lysis. Proliferation assay. Th cells (l-4 X 104) were cultured with 5 X lo5 irradiated (3000 R) stimulator cells in 0.2 ml medium in flat-bottom 96-well plates for 48 hr. They were then pulsed with 0.5 &i 5-[‘251]iodo-2’deoxyuridine ([‘251]Udr) (Amersham
T HELPER
CELL
LINES
FOR IN VW0
CTL
305
RESPONSES
International) in 10 ~1medium for 6 hr and harvested on an automated cell harvester (Dynatech). DNA incorporation was assessedby gamma counting and the results expressedas cpm [i2’I]Udr uptake f SD. Con A supernatant. Con A supematant was produced by culturing CBA spleen cells (107/ml) in complete RPM1 containing 2 pg/ml Con A for 24 hr. Cells were removed by centrifugation and Con A was neutralized with 25 mMa-methylmannoside (Sigma). The Con A supematant was used at 5% v/v. RESULTS H-2 Restriction and Antigen Specificity of the Proliferation of T Helper Cell Lines T helper (Th) cell lines were produced by repeatedly restimulating primed female BDF, (H-2b X H-2d) spleen or lymph node cells with irradiated antigen in vitro (P. K. Lai and T. Owens, submitted). The antigen was always presented on H-2b (BALB.B) cells so that it was expected that the selected Th cells would be restricted by H-2b and not by H-2d. Table 2 demonstrates that the Th cell lines were indeed H-2b restricted in their proliferation and that they were also specific for the minor antigens to which they had been raised. The anti-H-7b line proliferated strongly only to strains carrying both H-7b and H-2b (B6.C-H-7b and BALB.B) and the anti-H-25” line only to strains carrying both H-25’ and H-2b (B6.C-H-25”, BALB.B, and [B6.C-H25” X BALB/c]F,). The anti-H-25” line did not respond to the male specific antigen H-Y (BDF, male). Neither line responded to allo-H-2 stimuli. Help for a Primary CTL Response to Multiple Minor Antigens Preliminary experiments showed that whereas the primary CTL responseof BALB background mice to minor antigens of the B 10 background (e.g.,BALB/c anti-B 1O.D2) were strong (26) the reciprocal combinations (BlO or BDFl anti-BALB.B and BlO.D2 TABLE 2 Proliferation
of T Helper Cell Lines” Cell line
Stimulator
Anti-H-7 b
Anti-H-25 c
-
281 (29) 408 (305) 2751 (1657) nd 3953 (2154) 300 ( 148) nd 227 (43) nd 529 (145)
307 (22) 471 (83) 2517 (193) 295 (23) nd 3728 (160) 3216 (116) nd 362 (22) nd
BALB/c BALB.B BALB.K B6.C-H7 b B6.C-H-25’ (B6.C-H-25’ X BALB/c)F, BIO.BR BDF, male BDF, female
’ 1 X lo4 (anti-H-f) or 4 X 10’ (anti-H-25’) Th cells were cultured for 48 hr with 5 X lo5 irradiated (3000 R) stimulator cells in 0.2 ml medium in 96-well flat bottom plates. Results are expressed as cpm [‘*‘I]Udr uptake (&SD).
306
GASCOIGNE AND LA1 TABLE 3 Help for a Primary CTL Response to Multiple Minor Antiget& LU30 per culture Th cell line transferred
Target: BALB.B
BIO
-
0.5
0.0
Anti-BALB.B Anti-H-7 * Anti-H-25 c
33.3 39.2 23.8
0.1 0.0 0.9
a BDF, mice were injected in the hind footpads with BALB.B spleen cells plus 4 X lo5 Th cells. The CTL obtained were assayed on BALB.B or BlO Con A blast target cells. LUs,, is the reciprccal of the culture fraction producing 30% specific lysis.
or BDF, anti-BALB/c) were not. The responses of BDFi mice could be greatly augmented by prepriming several weeks earlier (Czitrom and Gascoigne, unpublished observations). To discover if this low responsiveness was due to a low frequency of Th cells or alternatively of CTLp we tested the ability of the three Th cell lines to augment a CTL response to the minor antigens of BALB.B (Table 3). All three cell lines proved able to help the primary CTL response to BALB.B cells since the Th were all reactive with minor antigens expressed on the BALB.B cell surface. This suggested that the low primary response in this combination was due to a lack of Th cells rather than of CTLp. It was of interest to find out whether Th cells responding to a single minor antigen expressed on the BALB.B cell surface caused the preferential induction of CTL responding to that particular antigen. It is clear from Fig. 1 that they did not. When injected at the same time as BALB.B
0
I RECIPROCALOF CULTURE FRACTION
FIG. 1. Th cells against H-2Y do not preferentially induce CTL to H-2jc in a primary response against BALB.B cells. BDFr mice were primed with BALB.B cells with (solid symbols) or without (open symbols) anti-H-25’ Th cells. The CTL response was assayed on a variety of Con A blast target cells: BALB.B (A A, A A); B6.C-H-25’ (0 l,O 0); B6.C-H-7b (v -V) or B6 (Wn).
T HELPER CELL LINES FOR IN VW0
CTL RESPONSES
307
antigen, a&H-25” Th cells helped a response to minor antigens expressed by BALB.B but not by B6.C-H-25” or B6.C-H-7b. The CTL response was therefore directed against minor antigens other than H-25” or H-7b. In other words, there was not a detectable bias of the response toward the antigen recognized by the Th cells. Help for a Secondary CTL Response to H-25’
To check that Th cells were able to induce CTL to the antigen against which they were raised, BDF, mice were primed to B6.CH-25” spleen cells which were injected into the footpad. One month later they were boosted with the same antigen mixed with anti-H-25” Th cells. Unprimed BDF, mice were also injected with antigen with or without Th cells. Figure 2 demonstrates that the secondary CTL response was specific to H-25”. B6.C-H-25” and BALB.B targets were killed to approximately the same degree, but B6 and B6.C-H-7b only marginally. That these two targets were killed at all may indicate that a small proportion of the response was directed against shared viral antigens. Figure 2 also shows that there was no primary response to B6.G H-25” either with or without Th cells indicating that anti-H-25C killing was detectable only after prepriming with this antigen. H-2 Restriction and Dependence
of Th Cells on Host Antigen Reprocessing
in Vivo
The Th cells used in this study have been shown to be antigen specific and restricted to H-2b for proliferation (Table 2). To test the restriction of their in vivo helper activity, BDF, mice were injected with 10’ irradiated BALB.B or BALB/c spleen cells plus 3.5 X lo4 anti-H-7b Th cells or normal BDF, lymph node cells per hind footpad. An additional control group was injected with Th cells alone. The effect of adding Con A supematant at the in vitro culture stage was also tested. As Fig. 3 shows, the supposedly restricted Th cells appeared to help the primary response to BALB minor
I
.
81
21
9
3
RECIPROCAL OF CULTURE FIVLTI~N
FIG. 2. Specificity of a secondary CTL response to B6.C-H-25’. BDF, mice were primed 1 month previously with B6.C-H-25C spleen cells were challenged with B6.CH-25’ spleen cells plus a&H-25’ Th cells. The CTL response was assayed on various target cells: BALB.B (A A); B6.C-H-25C (0 0); B6 (D n); B6.C-H-7b (7 7). A A represents normal BDF, mice injected with B6.C-H-25’ spleen cells plus anti-H-25’ Th cells and assayed on B6.C-H-25C target cells. 0 0 is the same but without Th cells.
308
GASCOIGNE AND LA1 a
60
,
243
81
21
9
In
3
RECIPROCAL OF CULTURE FRACTION
3 243 81 21 9 RECIPROCAL OF CULTURE FRACTION
FIG. 3. Lack of H-2 restriction of Th cells in viva and the effect of adding Con A supematant in vitro. BDFr mice were injected with BALB.B or BALB/c spleen cells plus anti-H-7b Th cells or normal BDF, lymph node cells. Four days later each of these groups were cultured either alone or in the presence of IL 2 (5% Con A supematant) for a further 3 days before assay of CTL on BALB.B (a), BALB/c (b), or BDF, target cells. BALB.B + Th (A - - - A); BALB.B + Th + IL-2 (0 - - - 0); BALB.B + lymph node cells BALB/c + Th (m - - - w); BALB.B + lymph node cells + IL-2 (V - - - V); BALB/c + Th (A-A); + IL-2 (0 0); BALB/c + lymph node cells (0 0); BALB/c + lymph node cells + IL-2 (V V). An additional group was injected with Th cells alone (Cl). None of the groups gave significant killing on BDF, target cells.
antigens in a completely unrestricted manner, while the CTL response itself was H-2 restricted to the antigen with which the mice were primed. Th cells transferred alone did not generate any CTL response. Similar results have been obtained with anti-BALB.B Th cells. We interpret these results to mean that reprocessing of the antigen by the F1 host was sufficient to mask any effects of H-2 restriction of the Th. It was not possible to obtain meaningful data on the activity of these Th in parental (BlO or BlO.D2) mice due to reactions against the allo-H-2 carried by the F1 Th (unpublished data). Addition of T cell growth factors (in the form of Con A supematant) to the culture stage of the experiment had the effect of increasing the magnitude of the CTL response in groups both with and without Th. The Con A supematant was not in itself sufficient to raise the response without Th cells to that level obtained with Th cells. Further evidence for the importance of antigen processing in this system is provided in Fig. 4. BDFl female mice were primed in each hind footpad with 10’ BlO.D2 male spleen cells. Five weeks later they were challenged with 10’ B6.C-H-7b male or female, B 10 male, or a mixture of 10’ B6.C-H-7b female plus 10’ B 10 male spleen cells with or without anti-H-7b Th cells. Unprimed mice do not generate responses to H-Y in this system. A secondary CTL response to H-Y was generated only when the mice primed were injected with both H-7 incompatible cells and cells expressing the male antigen. These antigens, however, did not need to be expressedon the same cell, as might have been expected from previous reports of Th-CTL interactions (23, 24,29), again suggestingthat antigen processing by host APCs is an important factor. Alternatively, it is possible that in the limited environment of the lymph nodes, sufficient concentrations of growth factors were made to overcome the usual control mechanisms (12, 13). The antigen specificity of the Th cells is demonstrated by the
T HELPERCELL LINES FOR IN
VZVO
CTL RESPONSES
309
a
243 81 21 9 3 RECIPROCALOF CULTUREFRACTION
RECIPROCALOF CULTUREFRACTION
RECIPROCALOF CULTUREFRACTION
FIG. 4. Th cells are antigen specific but can help a CTL response to another antigen presented at the same time. BDF, femalemicewereprimedwith BIO.D2malespleencells.Fiveweekslatertheywere challenged with various antigens or mixtures of antigens with or without anti-H-7b Th cells. The CTL responseswere measured on BIO male (a), BlO female (b), and B6.C-H-7b female (c) target cells. The antigens were BlO male alone (0 -Cl), BlO male plus Th (m n); B6.C-H-7b male alone (0 - 0), B6.C-H-7b male plus Th (@0); B6.C-H-7b female plus Th (A A) or B6.C-H-7b female plus BlO male plus Th (A A).
finding that they did not augment the CTL response against H-Y when presented with H-Y alone, but did when presented with H-7b plus H-Y. This experiment also showsthat helper cells were limiting in a secondary responseto a single minor antigen, as was found in the primary response to multiple minor antigens, even though there was clearly no shortage of CTLp. DISCUSSION We have described a novel system for the study of help in in vivo CTL responses. Previous work has, with few exceptions (30), used a protocol of priming in vivo
310
GASCOIGNE
AND
LA1
followed by restimulation with antigen in vitro (6-8, 17, 19-24). The method employed here makes useof cultured Th cell lines, measuring their activity in vivo in the primary (26) or secondary response to minor antigens. These cell lines are H-2 restricted, Ly lf2- T cells which have been shown to function as helpers for antibody responses (P. K. Lai and T. Owens, submitted). To assayhelp for CTL they were injected with antigen into the hind footpads of mice. Four days later the draining lymph nodes were cultured without antigen or added growth factors and after another 3 days CTL activity was measured. Two main findings are reported here. We have shown first that it is a shortage of Th cells rather than CTLp that results in low primary and secondary in vivo responses to minor antigens; and second that the Th cells are activated not simply by APC in the spleen cells used to immunize, but also by reprocessed antigen on APC of the host animal. The evidence for the first point is that what was normally a very low primary (Table 3) or secondary (Figs. 2 and 4) responseto multiple or single minor antigens, respectively, was greatly augmented by the transfer of extra Th cells. It is possible that the extra Th cells were necessaryto overcome the effect of suppressor cells in the lymph nodes (31) though there is other evidence to support the view that help is the limiting factor: the injection of cells allogeneic at the Mls locus, which elicits strong IL-2 production (32), can facilitate a response (33) as can injection of partially purified IL-2 directly into the footpads (34). Our data on the secondary response (Figs. 2 and 4), however, suggest that CTLp as well as Th cells are limiting for a response to single minor antigens since the Th cells could not induce CTL in mice that were not previously primed to the relevant antigen (Fig. 2). Thus, even when the frequency of CTLp was increased by priming (< l/ 100,000 to l/3500 for H-Y (35)), Th cells were limiting for the CTL response. We have not determined for certain the time at which Th cells act, but we believe for two reasons that it is mainly prior to the in vitro phase. The first reason is that the addition of T cell growth factors (including IL-2) in vitro to cells from mice primed without Th cells does not increase CTL responsesup to the values obtained with antigen plus Th cells (Fig. 3). Although the growth factors have a small effect on groups primed both with and without Th cells, it is clear that the Th cannot be substituted for in vitro with Con A supernatant, which should contain all the factors necessaryto enable CTL differentiation (11, 12). The second is that in the strong primary CTL response of BALB/c mice against B 10.D2 minor antigens, IL2 production by the lymph node cells occurs in a burst from Day 3 to Day 4, declining very sharply by the fifth day (N. R. J. Gascoigne, unpublished data). This suggests that Th cell activity peaksjust before the stageat which the lymph nodes are cultured. We have shown that the H-2 heterozygous Th cells are H-2 restricted in their proliferation (P. K. Lai and T. Owens, submitted and Table 2) and yet when injected back into syngeneic F, mice their helper activity for an H-2 restricted CTL response was unrestricted (Fig. 3). One possible explanation for this is that the Th cells were unrestricted in their activation by antigen in vivo, however, this Seemsvery unlikely in view of the specificity shown in the proliferation assay. We suggestthat when H2d associated antigen was injected into BDF, mice it associated with host (H-2b’d) APC so that the Th cells could recognize it in association with H-2b. Such an interpretation is supported by the data shown in Fig. 4 where effective help occurred when the antigens that the Th cells and the CTL respond to were presented on different
312
GASCOIGNE
AND
LA1
7. Simon, M. M., Edwards, A. J., Hlmmerling, U., McKenzie, I. F. C., Eichmann, K., and Simpson, E., Eur. J. Immunol. 11, 246, 1981. 8. Michaelides, M., Sandtin, M., Morgan, G., McKenzie, I. F. C., Ashman, R., and Melvold, R. W., J. Exp. Med. 153,464, 1981. 9. Weinberger, O., Herrmann, S. H., Mescher, M. F., Benacerraf, B., and Burakoff, S. J., Eur. J. Immunol. 11, 405, 1981.
10. Wagner, H., Hardt, C., Rouse, B. T., RoBinghoff, M., Scheurich, P., and Phzenmaier, K., J. Exp. Med. 155, 1876, 1982.
11. Raulet, D. H., and Bevan, M. J., Nature (London) 296, 754, 1982. 12. Hardt, C. Rollinghoff, M., Plizenmaier, K., Moosmann, H., and Wagner, H., J. Exp. Med. 154, 2, 1981.
13. Miihlradt, P. F., and Gpitz, H. G., Eur. J. Immunol. 12, 983, 1982. 14. Widmer, M. B., and Bach, F. H., Nature (London) 294, 750, 1981. 15. Roopenian, D. C., Widmer, M. B., Orosz, C. G., and Bach, F. H., J. Immunol. 130, 542, 1983. 16. von Boehmer, H., and Turton, K., Eur. J. Immunol. 13,750, 1983. 17. Raulet, D. H., and Bevan, M. J., J. Exp. Med. 155, 1766, 1982. 18. Engers, H. D., Glasebrook, A. L., and Sorenson, G. D., J. Exp. Med. 156, 1280, 1982. 19. Zinkemagel, R. M., Callahan, G. N., Althage, A., Cooper, S., Streilein, J. W., and Klein, J., J. Exp. Med. 147, 897, 1978.
20. von Boehmer, H., and Haas, W., J. Exp. Med. 150, 1134, 1979. 21. Hurme, M., Hetherington, C. M., Chandler, P. R., Gordon, R. D., and Simpson, E., Immunogenetics 5, 453, 1977. 22. Hurme, M., Hetherington, C. M., Chandler, P. R., and Simpson, E., J. Exp. Med. 147, 758, 1978. 23. Keene, J., and Forman, J., J. Exp. Med. 155, 768, 1982. 24. Kanagawa, O., Louis, J. A., and Cerottini, J-C., In “Mechanisms of Lymphocyte Activation” (K. Reach
and H. Kirchner, Eds.), p. 404. Elsevier/North Holland Biomedical Press,Amsterdam, 1981. 25. Starzinski-Powitz, A., Pfizenmaier, K., Rollinghoff, M., and Wagner, H., Eur. J. Immunol. 6, 799, 1976. 26. Czitrom, A. A., and Gascoigne, N. R. J., Immunology 27. Bailey, D. W., Immunogenetics 2, 249, 1975.
SO, 121, 1983.
28. Lai, P. K., Waterfteld, J. D., Gascoigne,N. R. J., Shamrock,C. E. M., and Mitchison, N. A., Immunology 47, 371, 1982. 29. Mitchison, N. A., Transplant Proc. (in press), 1983. 30. Baum, L. L., and Pilarski, L. M., Trunsplunfution 32, 409, 1981.
31. Wagner, H., Starzinski-Pow&z, A., Pfizenmaier, K., and RoBinghoff, M., Eur. J. Zmmunol. 6, 873, 1976. 32. Rollinghoff, M., Ptizenmaier, K., Solbach, W., and Wagner, H., Be/wing Inst. Mitt. 67, 36, 1980. 33. Butler, L. D., and Battisto, J. R., J. Immunol. 122, 1578, 1979. 34. Wagner, H., Hardt, C., Heeg, K., Pfizenmaier, K., St&&r, H., and Rohinghoff, M., Immunobiofogy 161, 139, 1982.
35. Kanagawa, O., Louis, J., and Cerottini, J-C., J. Immunol. 128, 2362, 1982. 36. Matzinger, P., and Bevan, M. J., Cell Immunol. 33, 92, 1977. 37. Weinberger, O., Herrmann, S. H., Mescher, M. F., Benacerraf, B., and Burakoff, S. J., Proc. Nut. Acud. Sci. USA 78, 1796, 1981. 38. Rollinghoff, M., Pfizenmaier, K., and Wagner, H. Eur. J. Immunol. 12, 337, 1982. 39. Czitrom, A. A., Sunshine, G. H., Reme, T., Ceredig, R., Glasebrook, A. L., Kelso, A., and MacDonald, H. R., J. Immunol. 130, 546, 1983. 40. Mitchison, N. A., Transplant Proc. 2, 92, 1970. 41. Ahman, A., Bechtold, T. E., and Katz, D. H., J. Immunol. 122, 2484, 1979.