Tumor-specific helper T cell activity was detected in the spleens of tumor-bearing ... suppressor cells: (a) T cells which suppress tumor-specific T helper ...
Eur. J. Immunol. 1982.12: 671-675
Sarah Howie and William H. MNcBride Department of Bacteriology, Edinburgh University Medical School, Edinburgh
Tumor-specific lymphocyte responses
671
Tumor-specific T helper activity can be abrogated by two distinct suppressor cell mechanisms* Tumor-specific helper T cell activity was detected in the spleens of tumor-bearing mice using a sensitive, simple and reproducible in vitro assay based on those previously described for soluble hapten-carrier antigens. Helper activity disappears when the tumor is large. This is coincident with the appearance of two distinct types of suppressor cells: (a) T cells which suppress tumor-specific T helper responses and (b) non-T cells which are nonspecific in their action.
1 Introduction
2 Materials and methods
One of the major problems of tumor immunology is to arrive at an understanding of the various cellular interactions between T and B lymphocytes and accessory cells which occur during all aspects of the immune response to tumors. The role of T cell subslets [including helper (Th) and suppressor (T,) cells] in the generation of specific cytotoxic cells against tumors has been extensively documented in the literature [1-4]. Far less well-characterized is the nature of the cellular interactions between Th and T, cells in the generation of the humoral (antibody-forming) immune response to tumor antigens. Part of the difficulty with this aspect of tumor immunology arises from the problems associated with the detection of specific antibody responses to weak cellular antigens whose physico-chemical nature is largely unknown and may be unique to individual tumors. A sensitive, siimple and reproducible in vitro assay system is described herein, based on those previously described for soluble haptl-nated antigens [5-71, which bypasses the need to measure anti-tumor antibodies directly. A similar assay has been described by Fujiwara et al. [8] who demonstrated that a mammary carcinoma induced tumor-specific Th lymphocytes in the spleens of tumor-bearing mice which would collaborate in vivo with hapten-primed B lymphocytes in the presence of haptenated tumor. We have extended these observations to demonstrate the development of tumor-specific Th cells involved in humoral responses at different stages of tumor growth. We have also demonstrated that help begins to disappear when tumors are of a significant size and is absent when the tumors are large. This disappearance of helper activity coincides with the appearance of suppressor cells. These suppressor cells are of two distinct types: (a) nylon wool (NW)nonadherent tumor-specific T, cells and (b) NW-adherent nonT, non-tumor-specific suppressor cells.
[I 35971
* This work was supported by the Cancer Research Campaign, Grant number SP1364. Correspondence: William H. McBride, Department of Bacteriology, Edinburgh University Medical School, Teviot Place, Edinburgh, Scotland Abbreviations: CRBC: Calf red blood cells SRBC: Sheep red blood cells HBSS: Hanks’ balanced salt solution TNP 2,4,6-Trinitrophenyl PFC: Plaque-forming cells NW: Nylon wool C: Complement T,,:T helper T,:T suppressor SC: Spleen cell(s) 0 Verlag Chemie GmbH, D-6940 Weinheim, 1982
2.1 Mice
C3WBu Kam female mice aged 14-18 weeks were maintained in the animal unit, Department of Bacteriology, University of Edinburgh.
2.2 Tumors All tumors were syngeneic for C3Hf/Bu Kam mice and were maintained by in vivo passage. Fsa R is a well described methylcholanthrene-induced tumor that is relatively strongly immunogenic, as determined by both the lung colony [9] and S.C. TDSO[lo] assays, and was used at pass 8-12. HTN has previously been called NFsa by others [ l l , 121. We do not believe it is a fibrosarcoma because in our hands over 60% of the cells from this tumor have Fc receptors even after in vitro culture. We believe it may be of macrophage origin and have accordingly changed its name. It arose spontaneously and is a poorly immunogenic tumor as determined by the ability of specifically sensitized mice to influence the take of tumor cells injected either i.v. or S.C. [ l l ] . In this study it was used in isotransplant generation 11. FsaA is a less well-described methylcholanthrene-induced fibrosarcoma that is of intermediate immunogenicity between FsaR and HTN as judged by comparison of the ability of mice preimmunized with two i.p. injections of lo6 healthy irradiated cells to prolong the latency and inhibit the growth of S.C. challenge with 4 x lo5 viable cells [13]. This tumor was used at isotransplant generation 7. Tumor cell suspensions were prepared by finely chopping solid tumors and disaggregating using Dispase (Boehringer, Mannheim, FRG) as described previously [14]. Solid tumors were induced in mice by inoculation with 4 x lo5 viable tumor cells S.C.in the right flank. Mice were primed against the tumors by i.p. inoculation with 5 x lo6 irradiated (5000 rd) tumor cells on three occasions, each 2 weeks apart. 2.3 Erythrocytes
Calf red blood cells (CRBC) in Alsever’s solution were purchased from Tissue Culture Services, Slough, GB. Sheep red blood cells (SRBC) from sheep No. 10966, were the kind gift of Mr. C. Birrells, Moredun Research Institute, Edinburgh. 2.4 Trinitrophenylation of tumor and red blood cells
Irradiated (5000 rds) tumor cells at lo6 celldm1 or washed CRBC or SRBC at 10% (v/v) were mixed with an equal vol0014-2Y80/82/0808-0671$02.50/0
672
S. Howie and W. H . McBride
ume of cacodylate buffer containing 2 mg/ml2,4,6-trinitrobenzene sulfonic acid (Sigma, Poole, GB) for 10-15 min at room temperature. At the end of this time the reaction was halted by the addition of 10 mg/ml glycylglycine. The coupled cells were then spun down and washed 3 times with Hanks’ buffered saline solution (HBSS). Trinitrophenyl (TNP)-coupled tumor cells were resuspended in HBSS containing 10% dimethyl sulfoxide and 20% fetal calf serum, the cells were aliquoted and stored in liquid nitrogen. TNP-coupled erythrocytes were stored at 4 “C in HBSS and remained stable for 3-5 days. To prime mice against TNP they were given two i.p. injections of 0.1 ml 1%TNP-CRBC 4 weeks apart.
Eur. J. Immunol. 1982.12: 671-675
excess of TNP-primed B cells, were titrated cells containing potential T cell help. For suppressor cell assays, cultures were set up as described with 3 X 105/mlTh cells and into the mixture were titrated cells containing putative T, cells. In every experiment the following controls were performed to determine the background response: (a) T cell-depleted TNPCRBC-primed SC cells alone, (b) T cell-depleted TNPCRBC-primed SC with TNP-tumor cells, (c) T cell-depleted TNP-CRBC-primed SC with TNP-CRBC, (d) other cell populations (either tumor or TNP-CRBC-primed cells, e.g. NWpassed cells) with their “priming” antigens (either TNP-CRBC or TNP-tumor), in all cases for every experiment such background responses were always less than 80 PFC/106 cells cultures.
2.5 Antisera
Rat anti-mouse Thy.-1.2 monoclonal antibody was a kind gift from Dr. H. S. Micklem, Department of Zoology, University of Edinburgh. Anti-mouse IgG enhancing serum was a pool from three rabbits inoculated three times each with 1 mg ammonium sulfate-precipitated normal mouse IgG taken from a variety of strains and given in Freund’s complete adjuvant.
2.6 Complement
Pooled normal guinea pig serum was absorbed for 1 h at 4 “C with 3 0 4 0 % by volume-packed SRBC (for plaque-forming cell assays) or packed mouse spleen and thymocytes (for antiThy-1.2 cytotoxicity). The cells were then spun down, the serum aliquoted and stored in liquid nitrogen.
2.9 Plaque-formingcell (PFC) assay
This is essentially the method described by Cunningham and Szenberg [16]. At the end of the 5-day culture period, quadruplicate cultures were harvested, pooled, washed in HBSS, resuspended in HBSS and screened for anti-TNP PFC in Cunningham chambers in the presence of TNP-SRBC, guinea pig C and rabbit anti-mouse IgG enhancing serum. Results are expressed as numbers of anti-TNP PFC (direct indirect)/ lo6 B cells, after subtraction of the number detected against SRBC. It is necessary to use pool cultures because individual cultures would make the experiments technically unmanagable. To compensate for this, each experiment was repeated in its entirety at least 3 times and in part up to 15 occasions.
+
3 Results 2.7 Spleen cell (SC) separations
T lymphocytes were removed from SC suspensions lo7celldm1 with a 1:50 dilution of anti-Thy-1.2 antiserum for 1 h at room temperature. The cells were then spun down, resuspended in a 1:20 dilution of guinea pig complement (C) and incubated for 1 h at 37 “C. At the end of this time the cells were washed, resuspended and the number of viable cells determined by trypan blue exclusion. After such treatment 40-50% of the nucleated cells were viable. Splenic T lymphocytes were purified by passing them over NW columns using the method of Julius et a1 [15]. The NWnonadherent cells recovered from such columns (10-20% of the starting number) were 85-95% Thy-1.2’ by immunofluorescence. NW-adherent cells were recovered by vigorous washing of the columns with medium; this procedure resulted in the recovery of 40-50% of the number of those cells initially applied.
3.1 Tumor-specifichelper cells in the spleens of tumor-primed mice
C3Hf/Bu Kam mice were injected with irradiated, syngeneic tumor cells of three distinct types, FsaR, FsaA and HTN. NWpassed SC from such tumor-primed mice were shown reproducibly to contain populations of helper cells capable of collaborating with hapten-primed B cells, but only if the hapten was carried on the correct priming carrier “antigen” (tumor cell). The results of one such experiment are shown in Table 1. The help obtained from animals primed with FsaR was greater than that from mice primed with FsaA, which in turn was greater than that from mice primed with HTN; it is interesting Table 1. The specificity of help in spleens of mice primed with individual syngeneic tumors Antigen used in viiro
NW-passed cells from mice primed with FsaR FsaA HTN TNP-CRBC Nothing
TNP-FsaR TNP-FsaA TNP-HTN TNP-CRBC
2590”)
2.8 Cell cultures
Quadruplicate 200-p1 cultures in Iscove’s medium + 5% fetal calf serum in microtiter plates were incubated for 5 days in a humidified C 0 2 (5%) incubator. For Th cell assays cultures consisted of 2.5 x 106/mlviable SC from TNP-CRBC-primed mice depleted of T cells as a source of TNP-primed B cells and either 105/mltrinitrophenylated tumor cells (TNP-FsaR, TNPFsaA or TNP-HTN) or 107/ml TNP-CRBC as a source of haptenated antigen. Into this mixture, which contains an
240 110 10
10
10
2220
210
10 140
110 210
0 290 40
250
36yo
0
40
a) Total (direct and indirect) PFC/106cells generated in 5-day culture of 106/ml NW-nonadherent cells and antigen (5 X 105/ml TNPtumor cells or 107/mlTNPXRBC with 2.5 X 1O6/mlTNP-primed B cells.
Eur. J. Immunol. 1982.12: 671-675
Tumor-specific lymphocyte responses
673
._1
that this order is the same as that of immunogenicity of tumors in protection assays [9-131 i.e., FsaR > FsaA > HTN.
19w
3.2 Tumor-specifichelper cells are T lymphocytes and require physical linkage between hapten and carrier
15M)
C3Hf/Bu Kam mice were inoculated with 4 x lo5 viable FsaR cells S.C. in thle right flank. This inoculum gives rise after approximately 10 days to a tumor which grows progressively and kills untreated mice in 30-35 days. Spleens from tumorbearing rnice were removed when the tumors were between 5 and 12 mm in diameter (10-20 days post-inoculation). NWpassed SC from these mice contained helper cells which could cooperate in a specific fashion with hapten-primed B cells. This help was dependent upon physical linkage between the hapten and the priming “carrier” antigen (the FsaR cell), as can be seen from the results shown in Table 2 . The source of this helper activity is associated with the Thy-1.2-bearing SC i e . , T cells (Fig. 1).
I
s
10‘
lo5
10‘
3
lo5
N O of FsoA bearer 5 C I 3a1O6 TNP -piimed B cells
3.3 Appearance and disappearanceof Th activity during tumor growth In this series of experiments, the kinetics of the appearance and disappearance of splenic T cells during the period of FsaR tumor growth was monitored (Fig. 2). It is not possible to directly compare the numbers of PFC obtained on different days over a long period of time. To standarize the responses, they are expressed as percent of the maximum help obtained on a particular day, where 100% is taken as the number of PFC generated by 2.5 x 106/ml T cell-depleted TNP-CRBCprimed SC in the presence of 107/mlTNP-CRBC and 3 x lo5/ ml NW-passed TNP-CRBC-primed SC. Each point has been repeated in a minimum of 3 and a maximum of 15 completely separate experiments. Tumor-specific Thcells were detected as early as 4 days post-inoculation with 4 X lo5 viable FsaR tumor cells and remained detectable up until 21 days after. At
Figure 1. Helper activity is contained within the Thy-1.2’ cells of tumor-bearer spleens. ( 6 0 )Help obtained with whole SC from FsaR tumor-bearing mice; (A-A) Help obtained with NW-passed Help obtained with Thy-1.2 SC (8&90% Thy-1.F) and (0----0) + C-treated SC. Background (Bgd.): no. of PFC generated by 2.5 X 106/mlTNP-CRBC-primedB SC in the presence of TNP-FsaR (ld/ml) alone; 100%: no of PFC generated by 2.5 X 106/mlTNP-CRBCprimed B cells in the presence of 107/mlTNP-CRBC and 3 x 10S/ml NW-passed TNP-CRBC-primed SC. Each point represents the mean k 1SE of spleens from 4 mice, each time point has been repeated 3-15 times.
T
100
90
80
I0
Table 2. Spleens of tumor-bearing mice contain specific helper cells
that require hapten-carrier linkage
* B o
6
Antigen used in vitro
Antigen used Anti-TNP PFC/106input cells to prime T cells Exp. 1 Eltp. 2
FsaR FsaR CRBC CRBC TNP-FsaR TNP-FsaR TNP-CRBC TNP-CRBC TNP-FsaR + CRBC TNP-FsaR + CRBC TNP-CRBC + FsaR TNP-CRBC + FsAR
FsaR TNP-CRBC FsaR TNP-CRBC FsaR TNP-CRBC FsaR TNP-CRBC FsaR TNP-CRBC FsaR TNP-CRBC
0 60
70
z
40
ND
10
loo
40
ND
590 20
ND
750
no0
50 1340
660 80 30
840 ND 90
600
ND
10
50
r“
a) Irradiated FsaR or TNP-FsaR (105/ml)with or without lo7 CRBC or TNP-CRBC were cultured with 2.5 x 106/mlTNP-primed B cells in the presence or absence of 5 x 105/mlNW-passed SC from
either FsaR tumor bearers or TNP-CRBC-primed mice.
30
20
10
4
1
14
21
28
D a y paet tumor cell innoeulatim
Figure 2. Kinetics of the appearance and disappearance of tumorspecific help during tumor growth. Numbers of PFC generated by TNP-CRBC-primed B cells in the presence of TNP-FsaR cells and NW-passed SC ( 6 0 106/ml,0---0 105/ml,X-X 104/ml)taken from mice 4, 7, 14, 21 and 28 days after S.C.inoculation with 4 x lo5 viable tumor cells.
S . Howie and W. H. McBride
674
Eur. J. Immunol. 1982.12: 671-675
28 days the tumors were very large (18-22 mm diameter), and concomitant with this, invariably there was an almost total loss of detectable T cell help. 3.4 Disappearance of T cell help with advanced tumor growth is coincident with the presence of both specific and nonspecific suppressor cells
To determine whether loss of detectable helper activity was due to the absence of Th cells or the presence of suppressor cells, SC populations from mice with tumors 18-22 mm in diameter were titrated into cultures where help would normally be found. To measure tumor-specific suppression, cells were titrated into cell mixtures where the help provided by 10to 20-day FsaR tumor-bearing, NW-passed SC and TNP-FsaR cells were used as antigen. To measure nonspecific suppression, cells were titrated into cell mixtures where the help was provided by TNP-CRBC primed, NW-passed SC and TNPCRBC were used as antigen. The results from two such experiments are presented in Table 3. It can be seen that the spleens from mice bearing very large tumors contained at least two distinct populations of suppressor cells capable of abrogating the activity of tumor-specific Th cells. One population of very potent suppressors was antigen-specific, and not retained on NW columns (presumably T, cells); the other population was antigen-nonspecific, resistant to treatment with anti-Thy-1.2 antiserum and C, and adhered to NW columns (presumably a B cell or more likely a macrophage-like cell).
4 Discussion Immune responses to tumor antigens are mediated at least in part by T lymphocytes [3, 41. We have shown that the Thcom-
ponent for antibody formation can readily and reproducibly be measured using haptenized tumor cells as antigen. The T h responses thus generated follow the same rules as soluble hapten-carrier responses i.e., the responses show Thcell specificity for individual tumors and require the hapten to be physically linked to the appropriate tumor “carrier” antigen. The method used to demonstrate tumor-specific Th cells is based upon the ideas of Mitchison [17] and is similar to limited studies performed previously using tumor [8], parasite “1, and virus [19] as antigen. The demonstration of a variety of tumor-specific immune responses in the spleen of mice with progressively growing tumors has been made previously and raises the question as to their value to the host. The Th cell responses in our system are presumably involved in the production of anti-tumor antibodies which, it has been suggested, could have a deleterious influence [20]. We do however know that in the same tumor system (FsaR) and using T cell populations prepared under identical conditions to those used in this study, protective immunity can be demonstrated in a Winn-type assay and that the kinetics of development of this protective immunity are identical to the kinetics of development of Th cells [21]. It therefore seems possible that the Th cells may be involved in the generation of the protective response. Further studies are being undertaken to elucidate this point. In this study it has been shown that T, cells develop within the spleen of mice when the tumor burden becomes great. The suppression has two components. One is tumor-specific and mediated by T cells. Tumor-specific T, cells have been described in other tumor systems and, from what is known, appear to be subject to the same regulatory circuits as T suppressors of conventional responses [22, 231, although little information is available on their site of action. The T, cells that
Table 3. Suppressor cells in spleens from mice bearing tumorsa)
Antigen used in vitro
Source of T help (NW treated SC)
Treatment of spleens from mice with tumors Expt .
TNP-CRBC
TNP-CRBC-primed
TNP-CRBC
TNP-CRBC-primed
TNP-CRBC
TNP-CRBC-primed
Whole spleen Anti-Thy 1
+ C-treated
NW-passed
TNP-CRBC-primed
N W-adherent
TNP-FsaR
FsaR-bearing
Whole
TNP-FsaR
FsaR-bearing
Anti-Thy-1 t C-treated
TNP-FsaR
FsaR-bearing
N W-passed
TNP-FsaR
FsaR-bearing
NW-adherent
TNP-CRBC
1 2 1 2 1 2 I
7+ Suppression”’ (anti-TNP PFC/lObcells) at 10” 10‘ I o4 103
Suppressor cells/ml 6Y
2
71 87 67 26 25 73 65
1 2 1 2 1 2 1 2
73 78 68 85 81 90 80 78
8 53 51
42 38 40 18 5 11 38 32
21 0 10 9 4 6 7 8
62 82 46 62 80 93 68 55
45 77 16 26 88 83 57 12
46 48 17 0 55
63 56 52 51
4
Yo
27 2
a) Tumors were 18-22 mm in diameter. Anti-TNP PFC generated in 5-day culture of 1.5 X 1O6/m1TMP-primed B cells + 3 X 105/mlhelper cells 105/mlTNP-FsaR or 107/mlTNP-CRBC in the presence of varying doses of putative suppressor cell populations. b) Maximum responses (1) Anti-Thy-1 + C spleen + TNP-CRBC + TNP-CRBC helper cells gave 1260 anti-TNP PFC/106cells in Exp. 1 and 1420 anti-TNP PFC/106 cells in Exp. 2. (2) Anti-Thy-1 + C spleen + TNP-FsaR + FsaR helper cells gave 860 anti-TNP PFC/106 cells in Exp. 1 and 1000 anti-TNPPFC/106 cells in Exp. 2.
+
Eur. J. Immunol. 1982.12: 671-675
we have demonstrated act directly to inhibit the responses by primed Th cells i.e., act at the effector phase of the immune response. Nonspecific suppression of responses by cells from tumor-bearing mice (and humans) has also been noted before [23] and attributed to the activities of B cells [24, 251 and macrophages 126, 271. These nonspecific suppressors are presumably involved in the general depressed immune responsiveness associated with high tumor burden in animals and humans. This “clean” assay system provides an easy means of distinguishing between tumor-specific and -nonspecific suppressor cell responses and studying their interactions. Suppressor cell development has been used as an argument to explain why tumors grow [22, 28, 291. However, from the time at which suppressor cells appear in our system, we suggest that at least those that develop within the spleen are unlikely to be relevant to tumor growth. This does not mean to say that the same cells in another site, for example within the tumor, might develop with different kinetics and be more relevant to control of tumor growth. This possibility is at present under investigation. The authors would also like to thank Mr. D . Walkingshawfor technical assistance.
Received .January 18, 1982; in revised form April 23, 1982.
5 References Holden,, H. T., Kirchner, H. and Herberman, B., J . Immunol 1975. 115: 327. Fujiwara, H., Hamaoka, T., Shearer, G. M., Yamomoto, H. and Terry, W., J . Immunol. 1980. 124: 863. Shimizu, K. and Fung-Win, S . , J . Immunol. 1979. 122: 1162. Yamauchi, K., Fujimoto, S . and Tada, T., J. Zmmunol. 1979.123: 1653. Chan, 1:. L. and Henry, C., J . Zmmunol. 1976. 117: 1132.
Tumor-specific lymphocyte responses
675
6 Waldmann, H., Munro, A. and Hunter, P., Eur. J . Immunol. 1973. 3: 167. 7 North, J. R. and Askonas, B. A., Eur. J. Immunol. 1976. 6: 8. 8 Fujiwara, H., Hamaoka, T., Toshima, K., Oaki, H. and Kiligawa, M., Immunology 1976. 31: 239. 9 Milas, L., Hunter, N., Mason, K. and Withers, H. R., Cancer Res. 1974. 34: 61. 10 Suit, H. D. and Kastelan, A., Cancer 1970. 26: 232. 11 Ando, K., Hunter, N. and Peters, L. J., Cancer Immunol. Immunother. 1979. 6: 151. 12 Milas, L., Hersh, E. M., Stringfellow, D. A. and Hunter, N., J . Natl. Cancer Inst. 1982. 68: 139. 13 Walkingshaw, D., Fellowship Thesis, Institute for Medical Laboratory Sciences, London 1982. 14 Moore, K. and McBride, W. H., Int. J. Cancer 1980. 26: 609. 15 Julius, M. H., Simpson, E. and Herzenberg, L. A,, Eur. J . Immunol. 1973. 3: 645. 16 Cunningham, A. J. and Szenberg, A., Immunology 1968.14: 599. 17 Mitchison, N. A., Transplant Proc. 1970. 2: 92. 18 Abbas, A. K., James, S. L. and Sher, A., J . Zmmunol. 1981. 126: 1022. 19 Pope, B. L., Hapel, A. J., Martin, W. J., Merchant, B. and Ennis, F. A,, J . Zmmunol. Methods 1981. 40: 95. 20 Hellstrom, K. E. and Hellstrom, I., Adv. Immunol. 1974. 18: 209. 21 McBride, W. H., Peters, L. J., Mason, K. A. and Barrow, G., J . Reticuloendothel. SOC. 1979. 27: 151. 22 Perry, L. L. and Greene, M. I., Fed. Proc. 1981. 40: 39. 23 Naor, D., Adv. Cancer. Res. 1979.29: 45. 24 Kilburn, D. G., Smith, J. B. and Gorczynski, R. M., Eur. J . Immunol. 1974. 4: 784. 25 Rudczynski, A. B. and Mortensen, R. F., J . Natl. Cancer Inst. 1978. 60:205. 26 Bluestone, J. A. and Lopez, C . , J . Natl. Cancer Inst. 1979. 63: 1221. 27 Kolb, J. P., Arrian, S. and Zolla-Pazner, S . , J. Immunol. 1977. 118: 702. 28 Broder, S., Muul, L. and Waldmann, T. A,, J . Natl. Cancer Inst. 1978. 61: 5. 29 Greene, M. I., Fujimoto, S. and Schon, A. H., J . Immunol. 1977. 119: 757.
