- Email: [email protected]
Multiple suppressive effects of transforming growth factor β1 on the immune response in chickens
CELLULAR
IMMUNOLOGY
129,468-477
(1990)
Multiple Suppressive Effects of Transforming Growth Factor PI on the Immune Response in Chickens’
*Station de Pathologie Aviaire, Institut National de la Recherche, Agronomique, 37380 Monnaie, France; TDepartment ofPathology and Kaplan Cancer Center, NYU Medical Center, New York, New York 10016 Received February 12, 1990; accepted April 17, 1990 The immunosuppressive effect of human recombinant TGF-Bl on chicken immune responses in vitro was evaluated. TGF-/31 at l-10 rig/ml reduced T cell proliferation in response to concanavalin A by 50-80% and B cell proliferation in response to LPS by >90%. In contrast, when added to immune spleen cells, it reduced the secondary PFC response to sheep erythrocytes by <50%, particularly when added at the same time as antigen on Day 2 of incubation. When TGF-j31 was added during a 2-day incubation to nylon wool-nonadherent immune or normal spleen cells, it caused the maintenance and/or appearance of suppressor cells. These suppressor cells, in coculture with immune spleen cells, inhibited the secondary PFC response in vitro without any further exposure to TGF-@ 1. The phenotype of the cells giving rise to suppressor cells under the influence of TGF-j3 1 was CT8+, TCR~+((U, o), CT4-, TCR l-(-r, 6) cells. The results suggest that, in addition to direct suppressive effects on the proliferation of B cells and of some T cells, TGF-@I may suppress immune responses by maintaining or by promoting the development of suppressor T cells. o 1990 Academic press, tnc.
INTRODUCTION Transforming growth factor pl (TGF-/31)’ is a 25kDa disulfide-linked dimeric polypeptide secreted by many cell types and stored in blood platelets ( 1,2). TGF-/3 1 has been purified and its gene cloned (3,4). TGF-P 1 appears to have regulatory effects on growth and differentiation of a wide variety of cells. In general, its effects on the immune response are immunosuppressive. It inhibits B (5-9) as well as T cell (lo12) functions both with human and murine cells. The fact that TGFP-like factors are produced by tumor cells of varying origin ( 13- 17) imparts an additional importance to their immunosuppressive effect, as it has been shown that tumor specific immune and suppressor T cells can profoundly influence tumor growth in rodents (18) and chickens ( 19). In view of the evolutionary highly conserved structure of TGF-fl 1, we considered it possible that recombinant TGF-Pl of human origin might affect ’ Supported by PHS Grant NIH AG-04980 awarded by the National Institute on Aging, DHHS. 2 To whom correspondence should be addressed. 3 Abbreviations used: TGF-/3 1, transforming growth factor /31; SRBC, sheep erythrocytes; A,, agammaglobulinemic; PFC, plaque-forming cell; Con A, concanavalin A, LPS, lipopolysaccharide; PBS, phosphate-buffered saline; FCS, fetal calf serum.
0008-8749190 $3.00 Copyright Q 1990 by Academic F’res, Inc. All rights of reproduction in any form reserved.
SUPPRESSION
OF
IMMUNE
RESPONSE
BY
TGF-@I
469
chicken cells. On testing, this assumption was proven correct, i.e., both B and T cell proliferation in vitro were inhibited by TGF-B 1. The secondary antibody response to sheep erythrocytes (SRBC) with spleen cells from immunized chickens in vitro has the peculiar property of being higher when antigen is added on Day 2 rather than at the initiation of the culture (20). In previous studies this was shown to be due to the presence of antigen-induced suppressor cells in the spleen from SRBC-immune chicken that are apparently activated by the Day 0 reexposure to SRBC, but that become functionally undetectable when left in culture at 37°C for 2 days without antigen or with an unrelated antigen (20). Since in the absence of antigen these innate suppressor cells disappear, a secondary response induced on Day 2 rather than on Day 0 of culture provides a good assay system for added suppressor cells and/or factors with suppressive activity. In the present study the effect of TGF-81 on secondary antibody responses to SRBC in vitro was examined. It will be shown that addition of TGF-pl during the first 2 days of culture, before the addition of antigen, has a greater suppressive effect than addition on Day 2, because TGF-/31 causes innate CT8 suppressor cells to persist or be induced in the assay cultures. These suppressor cells then down regulate antibody formation even in the absence of additional TGF-P 1. MATERIALS
AND METHODS
Animals. Chickens of strain SC (B2/B2) were purchased as eggs from Hy-Line International
(Dallas Center, IA).
Reagents and antibodies. Anti-CT&
-CT4, and -TCR 1 murine monoclonal antibodies, that define the chicken homologs of mammalian CD8, CD4, and TCRy& respectively, were generously supplied by Dr. M. D. Cooper (Division of Developmental and Clinical Immunology, Department of Pediatrics, University of Alabama, Birmingham), and prepared as previously described (2 1,22). The anti-TCR2 monoclonal antibody, recognizing the chicken homolog of TCRaP (23), was a gift from Dr. J. Cihak (Institute for Animal Physiology, University of Munich, Munich). Recombinant TGF-@ I was obtained through the courtesy of Dr. M. A. Palladino (Genentech, San Francisco, CA). Concanavalin A (Con A) was purchased from Sigma Chemical Co. (St. Louis, MO) and lipopolysaccharide W (LPS) from Difco Laboratories (Detroit, MI). SRBC immunization. Chickens older than 4 months of age were injected intravenously with 0.25 ml of a 5% SRBC suspension. Their spleens were removed for culture 4 days after antigen injection. Spleen cell preparation. Spleen cells were prepared by gently teasing in PBS (0.165 A4, pH 7.2) and washed two times. Cells used in coculture experiments were passed through a nylon-wool column in order to deplete adherent and B cells (24). For indirect panning (25), such nylon wool-nonadherent cells were incubated for 30 min on ice with murine anti-CT& -CT4, -TCRl , or -TCR2 in 0.165 M PBS (2% FCS) at an antibody concentration previously determined. These antibody-coated cells were then washed and incubated in goat anti-mouse immunoglobulin IgG (GAM)-coated dishes, for an additional hour at 0°C and at 3 X 10’ cells per dish ( loo-mm diameter). The nonadherent cells were collected, incubated with or without TGF-@ 1 for 2 days, and used in the coculture experiments. Cell culture for antibody production (26). SRBC-primed spleen cells were cultured in petri dishes in 1 ml of modified minimum essential medium (27) with 5% agamma-
470
QUERE AND THORBECKE
globulinemic (4) chicken serum (10’ responder cells/ml), kept in a 10% CO2 + 7% O2 atmosphere and fed daily with nutrient cocktail (26). The responder cells were precultured 2 days before addition of antigen (50 ~1 of 2% SRBC). In cell-mixing experiments, nylon wool-nonadherent spleen cells from the same immune chicken used as donor of responder cells or from a normal noninjected chicken, were first precultured alone, at lo7 cells/dish, with or without TGF-Pl ( 1 rig/ml) for 2 days. They were then washed and mixed with the 2day-old responder cell cultures at an appropriate concentration. Antigen (50 ~1 SRBC 2%) was added at the same time. Plaque-forming cell (PFC) assay. PFC were detected by a slide modification of the Jerne plaque assay (28) on Day 4 of culture. During the last 45 min of culture prior to assay, 0.1 ml fresh normal chicken serum and 0.1 ml guinea pig C were added per milliliter of culture in order to lyse antibody-coated SRBC remaining in the culture which might lead to artifactual plaque formation (29). For PFC development, slides were flooded with a 10% solution of guinea pig complement (absorbed with SRBC) containing an anti-chicken immunoglobulin rabbit serum ( 1: 1000) and incubated for 1 hr at 37°C. The anti-SRBC response was expressed as PFC per dish. The mean values for a minimum of three dishes are recorded in the tables. PFC obtained in the absence of antigen were always determined and subtracted from the recorded values. Mitogen-inducedproliferative assay. The culture medium used for the proliferative assays was an equal part mixture of RPM1 1640 (GIBCO, Grand Island, NY) and EHAA (Biofluids, Inc., Rockville, MD), supplemented with 5% A, chicken serum, 1% Nutridoma-SR (Boehringer-Mannheim Biochemical, Indianapolis, IN), 5 X 1Op5 M2-mercaptoethanol (Eastman Co., Rochester, NY), penicillin and streptomycin. Spleen cells (5 X lo5 cells/well), cultured in flat-bottomed Linbro microtest plates (No. 76-003-05) for 72 hr, were incubated with Con A (2.5 pg/ml) or LPS (10-2050 pg/ml) in a final volume of 0.2 ml. [3H]thymidine (sp act, 2.0 Ci/mmol; Amersham Corp., Arlington Heights, IL) was added 6 hr before harvest at a dose of 1 PCi in 0.02 ml/well. The wells were harvested onto filter paper discs and counted in a liquid scintillation counter (Packard Instrument Co., Inc., Downers Grove, IL). RESULTS E&et of TGF-pl on PFC responses. The assay system takes advantage of the higher PFC response obtained when SRBC-immune spleen cells are cultured for 2 days before SRBC are added. In the following experiments SRBC are always added on Day 2 of culture and PFC are examined on Day 4. The effect of TGF-8 1 added on Day 0 was compared with addition on Day 2 (at the same time as antigen) (Table 1). Maximal inhibition was obtained with 10 rig/ml added on Day 0 (mean of 74.3 f 9.4% suppression). The same concentration added on Day 2 had much less effect in three out of four experiments (mean of 30.3 + 16.7% suppression). In experiments three and four this difference was statistically significant (Table 1). It was curious that 10 rig/ml gave greater inhibition than 1 rig/ml when added on Day 0, but not when added on Day 2. The greater effect of the higher concentration than of the lower concentration could possibly be explained by a lack of persistence of active TGF-P 1 until Day 2 at the lower concentration. However, that could not explain the greater effectiveness of TGF-/3 1 added on Day 0. Perhaps an effect of TGF-8 1 on the B cells prior to antigen addition could be a factor in this effect or else a sustaining/inducing effect of TGF-8 1 on innate suppressor cells.
SUPPRESSION
OF IMMUNE TABLE
471
RESPONSE BY TGF-/5I I
Effect of TGF-@I on the Secondary PFC Response of Chicken Spleen Cells to SRBC TGF-8 I Added On day 0 0 2 2
Mean PFC response/dish f SEM”
rig/ml 0 1 IO I IO
Expt I 9998 4095 5055 5095 5841
+914 + 780*** 4 271*** ? 77 I ** f 900*
Expt 2 1381 266 84 567 406
Expt 3
+ 298 f 49*+ 2 58t t 122 + 147’
5555 3043 1176 3404 4814
Expt 4
2 334 f 547 ** k 214”t * 490* + 1013”
2416 k 330 1425 + 710 599 2 195=,*** 1645 + 647 2540 f 580’
IVOW.*Lower than control group: P < 0.05. **Lower than control group: P < 0.02. ***Lower than control group: P d 0.0 I. t Lower than control group: P < 0.00 I. ’ SRBC (50 ~1, 2%) always added on Day 2. PFC/dish (IgM + IgG) determined on Day 4 of culture. n = 3. ’ Difference between these two groups: P < 0.05. ’ Difference between these two groups: P < 0.05.
Efict of TGF-@I on T and B cell proliferation. In order to determine the target of TGF-/31 action in the previous in vitro assay, it was essential to know first whether, at the dose used, TGF-/I1 could affect chicken T or B cell responses, or both, as in mice. The effect of TGF-p 1 on T cell proliferation in response to Con A is shown in Table 2. The suppressive effect of TGF-/31 was quite consistent but not complete. The degree of inhibition was similar at 1 and 10 rig/ml and amounted to 50 and 80% inhibition in experiments 1 and 2, respectively. The effect on B cell proliferation in response to stimulation by LPS is shown in Table 3. Here TGF-81 caused a greater than 90% inhibition in every experiment. Background proliferation levels were also lower in the presence than in the absence of TGF-p 1, but this effect was much less marked than the inhibition of the B cell response to LPS. Ejtrrcl of TGF-(31 on T cell-mediated suppression. The previous results showed that TGF-Pl added in culture, is able to affect chicken lymphoid cell responses both at the T and B cell levels. In order to analyze more precisely the effect of TGF-/I1 on T
TABLE 2 Effect of TGF-j3I on T Cell Proliferation Added to cultures TGF-6 I 0 1 naJml 10 rig/ml
’ One microcurie/well
[3H]Thymidine
Con A (2.5 &ml) + + + of [ ‘Hlthymidine
incorporation’(cpm
Expt I 85,546 -c 4,02 I 2,128 + 291 40,004 + 3,002 736277 42,757 + 3,286 516+215
2 SEM) (n) Expt 2
(6) (6) (6) (3) (6) (3)
140,644 k 11,189(6) 2,929 + 399 (6) 31,848 + 2,691 (6) 1,115+359 (3) 29,109 + 2,083 (6) 769+51 (3)
added after 66 hr of culture; cultures were harvested at 72 hr.
QUERE AND THORBECKE
472
TABLE 3 Effect of TGF-Bl on B Cell Proliferation [ 3H]Thymidine incorporation ’ (cpm -e SEM) (n)
Added to cultures TGF-P 1
10 20 50 0 10 20 50 0 10 20 50 0
0
1 rig/ml
10 rig/ml
’ One microcurie/well
Expt 1
Expt 2
8,601 f 721 (4) 9,575 + 871(4) 10,764 + 943 (4) 1,769 + 167 (4) 781 k 107 (4) 965 -t 109 (4) 1,121 f 157 (4) 736+77 (3) 1,055+212(4) 1,018 + 129 (4) ND 516?215(3)
ND 902 1 + 388 (6) 9440 + 409 (6) 1468 + 127 (6) ND 835 + 34 (6) 1108 k 126 (6) 346k 59 (4) ND 945 rf: 33 (6) 1035 + 93 (6) 317235 (4)
LPS Wml)
of [3H]thymidine
added after 66 hr of culture; cultures were harvested at 72 hr.
cell activity during the secondary anti-SRBC PFC response in chicken, either whole spleen cells or nylon wool-nonadherent spleen cells from the immune chickens were cultured separately with or without TGF-/31 and then added simultaneously with antigen on Day 2, after washing to remove the TGF-P 1, to other “assay” whole spleen cell cultures. Only cells which had been precultured with TGF-j31 suppressed the response of the assay cultures and the degree of suppression obtained (mean of 50.7 & 4.7%) was similar for whole spleen cells and for nylon wool-nonadherent spleen cells precultured with TGF-/31 (Table 4). Cells precultured without TGF-81 did not
TABLE 4 Suppressor T Cells Detectable after Incubation with TGF-PI Preincubated cells added on Day 2 Cells None Whole spleen NWNAC spleen
Mean PFC response/dish f SEM (n)”
Cultured with
Expt 1
Medium TGF-fi I b Medium TGF-@lb
1381 zk 298 (3) 956 f 177 (6) 370?69 (6) 1362 + 158 (6) 562+97 (6)
Expt 3
Expt 2 5555 t 334 7207a613(6) 3926 r 463 6254 k 356 3488 -+ 354
(3) (6) (6) (6)
2416 k 330 (3) ND ND 3324 zk 161(6) 1552 + 230 (6)
0 SRBC (50 ~1, 2%) added on Day 2 to SRBC-immune spleen cell assay cultures, PFC determined on Day 4. ’ TGF-j3 1 concentration in medium from Days O-2 was 1 rig/ml. It was washed out before the cells were added to the assay cultures on Day 2. c NWNA = nylon wool nonadherent.
SUPPRESSION
5x103 8 B 1x103 1
OF IMMUNE
7f ff f$ L 107
5X108
25x10’
RESPONSE BY TGF-@I
1.25X106
473
0
FcmberOfcdlSAddSd
FIG. 1. Dose-related inhibition of the anti-SRBC PFC response after addition of T cells precultured for 2 days with TGF-/31. T cells added to culture of SRBC-immune spleen cells (lO’/dish) on Day 2 of culture were precultured for 2 days with (Cl) or without(H) TGF-fil (1 ngjml). Control response of immune spleen cells without added T cells is indicated by the solid bar.
suppress the response of “assay” cultures. The degree of suppression obtained by leaving TGF+?l in the assay cultures throughout the culture period was somewhat greater than that obtained with cells precultured with TGF-Bl only, suggesting that TGF-/3 1 also had additional suppressive effects on the immune response. Results in Fig. 1 show the suppressive effect of different numbers of nylon woolnonadherent spleen cells precultured for 2 days with or without TGF-@I . Suppression in assay cultures was not obtained, regardless of the cell numbers added, when cells were taken from cultures without TGF-/3 1, whereas a dose-related decrease of the response was obtained with cells precultured with TGF-B 1. Nature of the cells aflected by TGF-@I during preculture. It seemed important to determine the phenotype of the cells which were affected by TGF-81 in culture to express suppressor activity. The results in Table 5 show that pretreatment of nylon wool-nonadherent cells from SRBC-immune chickens to remove CT4+ cells prior to the a-day culture with TGF-@I resulted in the same suppressor activity as that obtained with unfractionated T cells. On the other hand, pretreatment to remove CT8+ cells completely abolished suppressor activity (experiments l-3). Since in all the previous experiments the nylon wool-nonadherent cells, precultured with and without TGF-Pl, had been taken from the same SRBC-immune chicken as the responder cells to which they were added on Day 2 of culture, it seemed important to determine whether TGF-/3 1 would also cause the appearance of suppressor activity in spleen cells from uninjected chickens. The results of experiment 4 (Table 5) show that a significant suppression (27.5%) was obtained with unfractionated cells from normal chickens that had been precultured with TGF-@I , but not with cells precultured with medium alone or with CT8+-depleted T cells precultured with TGF-/31 . It should be noted, however, that T cells taken from normal chickens and added without preculture to SRBC-immune responder cells on Day 2 of culture together with antigen, reduce the response to approximately the same degree as do cells precultured with TGF-P 1 (30, and results not shown). These cells evidently disappear after being cultured for 2 days, unless TGF-/31 is added to the medium.
474
QUERE AND THORBECKE TABLE 5
Identification of Suppressor Cells, Detectable aher Exposure to TGF-61 in V&V, as CT4-, CT8+ T Cells Preincubated NWNA* spleen cells Mean PFC response+ SEM (a)** Added on hY2
None Unfractionated CT4+ celldepleted CT8+ celldepleted
Cultured with
Expt I
Medium TGF-@1 Medium TGF-fl I Medium TGF-8 1
4143+483(6)' 3966?476(11)" 2574f267(12)"*d 4904k 775(6)' 2770?423(6)/ 6165 ?428(6)' 6165+694(6)d
Expt 2 5508 2 171(6)' 7093k 346(10)b 5431k 322(12)"e 6198+660(6)' 3770+293(6)a 7379 ?440(6)' 7141 f 369(6)'
Expt 3 12,084+ 1,344(3) 10,378 -c 594 (6)' 6,764+ 876 (6)' 8,447 f479 (6)h 6,476 +299 (6)h
ND
Expt 4 4522f487(3) 4539 +266(6)' 3291k384(6)k
ND ND 4665 f 386(6) 4259+484(5)
* Nylon wool-nonadherent (NWNA) spleen cells (S-10 X 106), unfractionated or CT4+ or CT8+ celldepleted as indicated and precultured for 2 days + 1 ng TGF-fil at 37’C, were washed and added to responder spleen cells on Day 2 of culture together with antigen (50 ~1,2% SRBC). Donor of these “suppressor cells” was either a normal chicken (experiment 4) or the same SRBC immune chicken used for responder spleen cells (experiments l-3). ** On Day 0 responder spleen cells were taken from chickens injected iv with 0.25 ml 5% SRBC on Day -4. They were cultured at 37°C (10’ cells/dish). Antigen + “suppressor cells” were added on Day +2 and PFC were determined on Day +4. Background PFC (-antigen) were subtracted (mean background for these experiments was 569 f 133 PFC/dish). ‘-’ Statistical significance of differences between values indicated by the same letter: "P -c 0.02; 'P = 0.002; cP< 0.01; dP< o.OOO1;eP < O.OOS;'P = O.O4;8P= 0.01 hP= 10.01; 'P = O.Ol;‘P = 0.02; kP = 0.02.
To summarize results from this series of experiments, the mean degree of suppression obtained with unfractionated nylon wool-nonadherent spleen cells from SRBCimmune chickens in seven consecutive experiments, comparing TGF-/3 1-exposed to control cells, was 32%. Comparing TGF-81 exposed CT8+-depleted cells to their own controls, the suppression was 1.3%, and comparing TGF-01 exposed CT4+depleted cells to their own controls, the suppression was 35.3%. In experiments 1 and 2 (Table 5) the CT8+-depleted nylon wool-nonadherent cells caused a significant enhancement of the antibody responses in the responder cell cultures. This was not seen with CT4+-depleted cells or with CTVdepleted cells from a normal chicken (experiment 4). This finding suggests that CT4+ cells from SRBCimmune chickens, particularly after removal of CT8+ cells and regardless of preculturing with TGF-6 1, tend to enhance the response of unfractionated immune spleen cells to SRBC. In previous studies we have also shown helper activity for antibody production in unfractionated precultured cells, particularly under conditions where a helper cell deficiency was created in the assay B cell cultures (20). The results in Table 6 show that removal of TCRl+ cells prior to incubation with TGF-P 1 did not affect the suppressor activity detected at the end of the 2-day incubation (3 1.9% suppression), while removal of TCR2+ cells abolished suppression ( 1.5% suppression). It is interesting to note that the cells remaining after TCR2+-cell depletion neither enhanced nor suppressed, suggesting that the CT4+ helper cells responsible for the enhancement in CT8+ depleted cells had also been depleted by removal of TCR2+ cells.
SUPPRESSION
OF IMMUNE
475
RESPONSE BY TGF-/31
TABLE 6 Identification of Suppressor Cells, Detectable after Exposure to TGF-@I in Vitro, as TCR2+ T Cells Preincubated NWNA * spleen cells Added on Day 2 None Unfractionated TCR 1’ cell-depleted TCR2+ cell-depleted
Mean PFC response/dish + SEM (n) **
Cultured with
Expt 1
Expt 2
Medium TGF-8 1 Medium TGF-8 1 Medium TGF-#l 1
3533 -t 594 (3) 2757 + 249 (5) 1890 * 394 (6)” ND ND 3871 + 594 (6) 3814 + 653 (6)”
4260 f 397 (3) 3545 t 527 (5) 2107 f 420(6) 3150 f 281(5)b 2144rt312(6)b
* Nylon wool-nonadherent (NWNA) spleen cells (5-10 X 106), unfractionated or TCRl+ or TCR2+ cell-depleted as indicated and precultured for 2 days + 1 ng TGF-Bl at 37”C, were washed and added to responder spleen cells on Day 2 of culture together with antigen (50 ~1,2% SRBC). ** On Day 0 responder spleen cells were taken from chickens, injected iv with 0.25 ml 5% SRBC on Day -4. They were cultured at 37°C (10’ cells/dish). Antigen + “suppressor cells” were added on Day +2 and PFC were determined on Day +4. Background PFC (-antigen) were subtracted (mean background for these experiments was 427 + 115 PFC/dish). ‘zb Statistical significance ofdifferencesbetween values indicated by the same letter: “P = 0.03; bP = 0.04.
DISCUSSION The present results suggest that I- 10 ng TGF-P 1/ml has multiple immunosuppressive effects on the immune response of chicken spleen cells in vitro. In the first place it reduces the T cell response to the lectin Con A during a 3&y incubation at 40°C by 50-80s. Secondly, it reduces B cell proliferation in response to LPS by >90% during a 3-day incubation at 40°C. Most interestingly, it has also an indirect immunosuppressive effect by sustaining/inducing a suppressive effect of CT8+ T cells on antibody production. This latter effect was indirect since the TGF-@I was washed off before the cells that had been exposed to it were added to the assay culture. This effect was mediated by CT8 TCR2+ cells and was quite similar to the suppression obtained by these cells after preculture with antigen. However, the suppressive effect of cells precultured with TGF-81 could also be obtained with normal rather than SRBCimmune spleen cells. It appeared, therefore, that antigen-specific suppressor cells were not an essential component of this TGF-Pl effect. It seems more appropriate to compare the suppressor cells present after culture with TGF-@l to those induced by con A in vitro. In the present studies the suppressor cells were found to be CT8+, TCR2+ (CT4-, TCRI-). A similar phenotype was found in other studies for the Con A-induced suppressor cells (P. Quere, M. D. Cooper, and G. J. Thorbecke, manuscript in preparation). The present results do not allow a distinction between the induction and/or maintenance by TGF-/31 of suppressor T cells during the first 2 days of spleen cell cultures. Fresh normal chicken spleen cells added at the time of antigen addition to assay cultures cause a certain degree of suppression of the antibody response. Thus, some activation of suppressor cells appears to occur in fresh normal or SRBC-immune T cells when they are mixed with spleen cells responding to antigen, but the phenotype of these cells has not been established. It is possible that TGF-@ 1 maintains
476
QUERE AND THORBECKE
the activity of these same cells in vitro, so that preculturing at 37°C does not result in a loss of suppressor activity. However, it is also possible that other CT8+ T cells are induced by TGF-p 1 to exhibit suppressor activity. Studies of the actual role of TGF-@ 1 in naturally occurring suppressor cell effects or in Con A activation of suppressor cells await the availability of neutralizing antibodies to TGF-P 1. In previous studies we showed that the loss of suppressor activity in cultures of immune chicken spleen cells in the absence of antigen, which is seen at 37°C does not occur at 40°C (3 1). It is possible that under these conditions, a factor(s) is produced which, similar to TGF-@ 1, maintains or induces suppressor cell activity. It is of interest that TGF-P 1 inhibits chicken B cell proliferation to LPS to a greater extent than T cell proliferation to Con A. This high sensitivity of B cell proliferation to the suppressive effect of TGF-p 1 is in contrast with the relative resistance of antibody production in vitro when both TGF-@ 1 and antigen are added on Day 2 of culture. It appears from these observations that differentiation of B cells is much less inhibited (if at all) than B cell proliferation. This impression is also gained from recent data on the effect of TGF-61 on murine B cells, where it is found to enhance the isotype switch to IgA and thus has a B cell differentiation-promoting effect (32). ACKNOWLEDGMENTS We are greatly indebted to Dr. M. D. Cooper (Division of Clinical Immunology, Department of Pediatrics, University of Alabama, Birmingham) for his generous gifts of monoclonal antibodies to the chicken antigens CT4, CT8, and TCR 1. We thank Dr. M. A. Palladino (Genentech Corp., South San Francisco, CA) for generously providing the human TGF-PI used in these studies.
REFERENCES 1. Spom, M. B., Roberts, L. M., Wakefield, L. M., and Assoian, R. K., Science 233,532,1986. 2. Massague, J., Cell49,437, 1987. 3. Assoian, R. K., Komoriya, A., Meyers, C. A., Miller, D. M., and Spom, M. B., J. Biol. Chem. 258, 7155,1983. 4. Derynck, R., Jarrett, J. A., Chen, F. Y., Eaton, D. H., Bell, J. R., Assoian, R. K., Roberts, A. B., Spom, M. B., and Goeddel, D. V., Nature(London) 316,701,1985. 5. Kehrl, J. H., Roberts, A. B., Wakefield, L. M., Jakowlew, S., Spom, M. B., and Fauci, A. S., J. Immunol. 137,3855, 1986. 6. Blomhoff, H. K., Smeland, E., Mustafa, A. S., Godal, T., and Oholsson, R., Eur. J. Zmmunol. 17,299, 1987. 7. Lee, G., Ellingsworth, L. R., Gillis, S., Wall, R., and Kincade, P. W., J. Exp. Med. 166, 1290, 1987. 8. Petit-Koskas, E., Genot, E., Lawrence, D., and Kolb, J. P., Eur. J. Zmmunol. 18, 111, 1988. 9. Smeland, E. B., Blomhoff, H. K., Holte, H., Ruud, E., Beiske, K., Funderud, S., Godal, T., and Ohlsson, R., Exp. Cell. Res. 171,213, 1987. 10. R&ton, H. J., Proc. Natl. Acad. Sci. USA 83,5531, 1986. 11. Kehrl, J. H., Wakefield, L. M., Roberts, A. B., Jakowlew, S., Alvarez-Mon, M., Derynck, R., Spom, M. B., and Fauci, A. S., J. Exp. Med. 163, 1037,1986. 12. Ranges, G. E., Figari, I. S., Espevik, T., and Palladino, M. A., J. Exp. Med. 166,99 1, 1987. 13. Wrann, M., Bodmer, S., Martin, R., Siepl, C., Hofer-Watbinele, R., Frei, K., Hofer, E., and Fontana, A., EMBO J. 6, 1633, 1987. 14. Delarco, J. E., and Todaro, G. J., Proc. Natl. Acad. Sci. USA 75,4001, 1978. 15. Coffey, R. J., Shipley, G. D., and Moses, H. L., Cancer Rex 46,1164, 1986. 16. Derynck, R., Goeddel, D. V., Ulhich, A., Gutterman, J. U., Williams, R. D., Bringman, T. S., and Berger, W. H., Cancer Rex 47,707, 1987. 17. Newcom, S. R., Kadin, M. E., and Ansari, A. A., Amer. J. Pathol. 131,569, 1988. 18. North, R. J., Adv. Cancer Res. 45, I, 1985. 19. Edelman, A. S., Robinson, M. E., Sanchez, P., andThorbecke, G. J., Cell. Immunol. 110,321,1987.
SUPPRESSION 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32.
OF IMMUNE
RESPONSE BY TGF-81
477
Chi, D. S., Grebenau, M. D., and Thorbecke, G. J., Eur. J. Immunol. 10,203, 1980. Chan, M. M., Chen, C. H., Ager, L. L., and Cooper, M. D., J. Immunol. 140,2133, 1980. Sowder, J. T., Chen, C. H., Ager, L. L., Chan, M. M., and Cooper, M. D., J. Exp. Med. 167,315,1988. Cihak, J., Ziegler-Heitbrock, H. W. L., Trainer, H., Schranner, I., Merkenschlager, M., and Losch, U., Eur. J. Immunol. 18,533, 1988. Julius, M. H., Simpson, E., and Helzenberg, L. A., Eur. J. Immunol. 3,645, 1973, Wysocki, L. J., and Sato, V. L., Proc. Natl. Acad. Sci. USA 752844, 1978. Mishell, R. I., and Dutton, R. W., .I. Exp. Med. 126,423, 1967. Nakamura, K., J. Exp. Med. 135,476, 1972. Jerne, N. K., Nordin, A. A., and Henry, C., In “Cell Bound Antibodies” (B. Amos, and H. Koprowski, Eds.), (p. 109). Wistar Institute Press, Philadelphia, 1963. Muchmore, A. V., Koski, I., Dooley, N., and Blaese, R. M., J. Immunol. 116, 1016, 1976. Chi, D. S., and Thorbecke, G. J., In “Avian Immunology: Basis and Practice” (A. Toivanen and Towanen, Eds.), p. 16 1. CRC Press, Boca Raton, FL, 1987. Bhogal, B. S., Chi, D. S., and Thorbecke, G. J., Proc. Sot. Exp. Biol. Med. 176,414, 1984. Lebman, D., Lee, F., and Coffman, R., FASEB J. 2, A1268, 1989.
IMMUNOLOGY
129,468-477
(1990)
Multiple Suppressive Effects of Transforming Growth Factor PI on the Immune Response in Chickens’
*Station de Pathologie Aviaire, Institut National de la Recherche, Agronomique, 37380 Monnaie, France; TDepartment ofPathology and Kaplan Cancer Center, NYU Medical Center, New York, New York 10016 Received February 12, 1990; accepted April 17, 1990 The immunosuppressive effect of human recombinant TGF-Bl on chicken immune responses in vitro was evaluated. TGF-/31 at l-10 rig/ml reduced T cell proliferation in response to concanavalin A by 50-80% and B cell proliferation in response to LPS by >90%. In contrast, when added to immune spleen cells, it reduced the secondary PFC response to sheep erythrocytes by <50%, particularly when added at the same time as antigen on Day 2 of incubation. When TGF-j31 was added during a 2-day incubation to nylon wool-nonadherent immune or normal spleen cells, it caused the maintenance and/or appearance of suppressor cells. These suppressor cells, in coculture with immune spleen cells, inhibited the secondary PFC response in vitro without any further exposure to TGF-@ 1. The phenotype of the cells giving rise to suppressor cells under the influence of TGF-j3 1 was CT8+, TCR~+((U, o), CT4-, TCR l-(-r, 6) cells. The results suggest that, in addition to direct suppressive effects on the proliferation of B cells and of some T cells, TGF-@I may suppress immune responses by maintaining or by promoting the development of suppressor T cells. o 1990 Academic press, tnc.
INTRODUCTION Transforming growth factor pl (TGF-/31)’ is a 25kDa disulfide-linked dimeric polypeptide secreted by many cell types and stored in blood platelets ( 1,2). TGF-/3 1 has been purified and its gene cloned (3,4). TGF-P 1 appears to have regulatory effects on growth and differentiation of a wide variety of cells. In general, its effects on the immune response are immunosuppressive. It inhibits B (5-9) as well as T cell (lo12) functions both with human and murine cells. The fact that TGFP-like factors are produced by tumor cells of varying origin ( 13- 17) imparts an additional importance to their immunosuppressive effect, as it has been shown that tumor specific immune and suppressor T cells can profoundly influence tumor growth in rodents (18) and chickens ( 19). In view of the evolutionary highly conserved structure of TGF-fl 1, we considered it possible that recombinant TGF-Pl of human origin might affect ’ Supported by PHS Grant NIH AG-04980 awarded by the National Institute on Aging, DHHS. 2 To whom correspondence should be addressed. 3 Abbreviations used: TGF-/3 1, transforming growth factor /31; SRBC, sheep erythrocytes; A,, agammaglobulinemic; PFC, plaque-forming cell; Con A, concanavalin A, LPS, lipopolysaccharide; PBS, phosphate-buffered saline; FCS, fetal calf serum.
0008-8749190 $3.00 Copyright Q 1990 by Academic F’res, Inc. All rights of reproduction in any form reserved.
SUPPRESSION
OF
IMMUNE
RESPONSE
BY
TGF-@I
469
chicken cells. On testing, this assumption was proven correct, i.e., both B and T cell proliferation in vitro were inhibited by TGF-B 1. The secondary antibody response to sheep erythrocytes (SRBC) with spleen cells from immunized chickens in vitro has the peculiar property of being higher when antigen is added on Day 2 rather than at the initiation of the culture (20). In previous studies this was shown to be due to the presence of antigen-induced suppressor cells in the spleen from SRBC-immune chicken that are apparently activated by the Day 0 reexposure to SRBC, but that become functionally undetectable when left in culture at 37°C for 2 days without antigen or with an unrelated antigen (20). Since in the absence of antigen these innate suppressor cells disappear, a secondary response induced on Day 2 rather than on Day 0 of culture provides a good assay system for added suppressor cells and/or factors with suppressive activity. In the present study the effect of TGF-81 on secondary antibody responses to SRBC in vitro was examined. It will be shown that addition of TGF-pl during the first 2 days of culture, before the addition of antigen, has a greater suppressive effect than addition on Day 2, because TGF-/31 causes innate CT8 suppressor cells to persist or be induced in the assay cultures. These suppressor cells then down regulate antibody formation even in the absence of additional TGF-P 1. MATERIALS
AND METHODS
Animals. Chickens of strain SC (B2/B2) were purchased as eggs from Hy-Line International
(Dallas Center, IA).
Reagents and antibodies. Anti-CT&
-CT4, and -TCR 1 murine monoclonal antibodies, that define the chicken homologs of mammalian CD8, CD4, and TCRy& respectively, were generously supplied by Dr. M. D. Cooper (Division of Developmental and Clinical Immunology, Department of Pediatrics, University of Alabama, Birmingham), and prepared as previously described (2 1,22). The anti-TCR2 monoclonal antibody, recognizing the chicken homolog of TCRaP (23), was a gift from Dr. J. Cihak (Institute for Animal Physiology, University of Munich, Munich). Recombinant TGF-@ I was obtained through the courtesy of Dr. M. A. Palladino (Genentech, San Francisco, CA). Concanavalin A (Con A) was purchased from Sigma Chemical Co. (St. Louis, MO) and lipopolysaccharide W (LPS) from Difco Laboratories (Detroit, MI). SRBC immunization. Chickens older than 4 months of age were injected intravenously with 0.25 ml of a 5% SRBC suspension. Their spleens were removed for culture 4 days after antigen injection. Spleen cell preparation. Spleen cells were prepared by gently teasing in PBS (0.165 A4, pH 7.2) and washed two times. Cells used in coculture experiments were passed through a nylon-wool column in order to deplete adherent and B cells (24). For indirect panning (25), such nylon wool-nonadherent cells were incubated for 30 min on ice with murine anti-CT& -CT4, -TCRl , or -TCR2 in 0.165 M PBS (2% FCS) at an antibody concentration previously determined. These antibody-coated cells were then washed and incubated in goat anti-mouse immunoglobulin IgG (GAM)-coated dishes, for an additional hour at 0°C and at 3 X 10’ cells per dish ( loo-mm diameter). The nonadherent cells were collected, incubated with or without TGF-@ 1 for 2 days, and used in the coculture experiments. Cell culture for antibody production (26). SRBC-primed spleen cells were cultured in petri dishes in 1 ml of modified minimum essential medium (27) with 5% agamma-
470
QUERE AND THORBECKE
globulinemic (4) chicken serum (10’ responder cells/ml), kept in a 10% CO2 + 7% O2 atmosphere and fed daily with nutrient cocktail (26). The responder cells were precultured 2 days before addition of antigen (50 ~1 of 2% SRBC). In cell-mixing experiments, nylon wool-nonadherent spleen cells from the same immune chicken used as donor of responder cells or from a normal noninjected chicken, were first precultured alone, at lo7 cells/dish, with or without TGF-Pl ( 1 rig/ml) for 2 days. They were then washed and mixed with the 2day-old responder cell cultures at an appropriate concentration. Antigen (50 ~1 SRBC 2%) was added at the same time. Plaque-forming cell (PFC) assay. PFC were detected by a slide modification of the Jerne plaque assay (28) on Day 4 of culture. During the last 45 min of culture prior to assay, 0.1 ml fresh normal chicken serum and 0.1 ml guinea pig C were added per milliliter of culture in order to lyse antibody-coated SRBC remaining in the culture which might lead to artifactual plaque formation (29). For PFC development, slides were flooded with a 10% solution of guinea pig complement (absorbed with SRBC) containing an anti-chicken immunoglobulin rabbit serum ( 1: 1000) and incubated for 1 hr at 37°C. The anti-SRBC response was expressed as PFC per dish. The mean values for a minimum of three dishes are recorded in the tables. PFC obtained in the absence of antigen were always determined and subtracted from the recorded values. Mitogen-inducedproliferative assay. The culture medium used for the proliferative assays was an equal part mixture of RPM1 1640 (GIBCO, Grand Island, NY) and EHAA (Biofluids, Inc., Rockville, MD), supplemented with 5% A, chicken serum, 1% Nutridoma-SR (Boehringer-Mannheim Biochemical, Indianapolis, IN), 5 X 1Op5 M2-mercaptoethanol (Eastman Co., Rochester, NY), penicillin and streptomycin. Spleen cells (5 X lo5 cells/well), cultured in flat-bottomed Linbro microtest plates (No. 76-003-05) for 72 hr, were incubated with Con A (2.5 pg/ml) or LPS (10-2050 pg/ml) in a final volume of 0.2 ml. [3H]thymidine (sp act, 2.0 Ci/mmol; Amersham Corp., Arlington Heights, IL) was added 6 hr before harvest at a dose of 1 PCi in 0.02 ml/well. The wells were harvested onto filter paper discs and counted in a liquid scintillation counter (Packard Instrument Co., Inc., Downers Grove, IL). RESULTS E&et of TGF-pl on PFC responses. The assay system takes advantage of the higher PFC response obtained when SRBC-immune spleen cells are cultured for 2 days before SRBC are added. In the following experiments SRBC are always added on Day 2 of culture and PFC are examined on Day 4. The effect of TGF-8 1 added on Day 0 was compared with addition on Day 2 (at the same time as antigen) (Table 1). Maximal inhibition was obtained with 10 rig/ml added on Day 0 (mean of 74.3 f 9.4% suppression). The same concentration added on Day 2 had much less effect in three out of four experiments (mean of 30.3 + 16.7% suppression). In experiments three and four this difference was statistically significant (Table 1). It was curious that 10 rig/ml gave greater inhibition than 1 rig/ml when added on Day 0, but not when added on Day 2. The greater effect of the higher concentration than of the lower concentration could possibly be explained by a lack of persistence of active TGF-P 1 until Day 2 at the lower concentration. However, that could not explain the greater effectiveness of TGF-/3 1 added on Day 0. Perhaps an effect of TGF-8 1 on the B cells prior to antigen addition could be a factor in this effect or else a sustaining/inducing effect of TGF-8 1 on innate suppressor cells.
SUPPRESSION
OF IMMUNE TABLE
471
RESPONSE BY TGF-/5I I
Effect of TGF-@I on the Secondary PFC Response of Chicken Spleen Cells to SRBC TGF-8 I Added On day 0 0 2 2
Mean PFC response/dish f SEM”
rig/ml 0 1 IO I IO
Expt I 9998 4095 5055 5095 5841
+914 + 780*** 4 271*** ? 77 I ** f 900*
Expt 2 1381 266 84 567 406
Expt 3
+ 298 f 49*+ 2 58t t 122 + 147’
5555 3043 1176 3404 4814
Expt 4
2 334 f 547 ** k 214”t * 490* + 1013”
2416 k 330 1425 + 710 599 2 195=,*** 1645 + 647 2540 f 580’
IVOW.*Lower than control group: P < 0.05. **Lower than control group: P < 0.02. ***Lower than control group: P d 0.0 I. t Lower than control group: P < 0.00 I. ’ SRBC (50 ~1, 2%) always added on Day 2. PFC/dish (IgM + IgG) determined on Day 4 of culture. n = 3. ’ Difference between these two groups: P < 0.05. ’ Difference between these two groups: P < 0.05.
Efict of TGF-@I on T and B cell proliferation. In order to determine the target of TGF-/31 action in the previous in vitro assay, it was essential to know first whether, at the dose used, TGF-/I1 could affect chicken T or B cell responses, or both, as in mice. The effect of TGF-p 1 on T cell proliferation in response to Con A is shown in Table 2. The suppressive effect of TGF-/31 was quite consistent but not complete. The degree of inhibition was similar at 1 and 10 rig/ml and amounted to 50 and 80% inhibition in experiments 1 and 2, respectively. The effect on B cell proliferation in response to stimulation by LPS is shown in Table 3. Here TGF-81 caused a greater than 90% inhibition in every experiment. Background proliferation levels were also lower in the presence than in the absence of TGF-p 1, but this effect was much less marked than the inhibition of the B cell response to LPS. Ejtrrcl of TGF-(31 on T cell-mediated suppression. The previous results showed that TGF-Pl added in culture, is able to affect chicken lymphoid cell responses both at the T and B cell levels. In order to analyze more precisely the effect of TGF-/I1 on T
TABLE 2 Effect of TGF-j3I on T Cell Proliferation Added to cultures TGF-6 I 0 1 naJml 10 rig/ml
’ One microcurie/well
[3H]Thymidine
Con A (2.5 &ml) + + + of [ ‘Hlthymidine
incorporation’(cpm
Expt I 85,546 -c 4,02 I 2,128 + 291 40,004 + 3,002 736277 42,757 + 3,286 516+215
2 SEM) (n) Expt 2
(6) (6) (6) (3) (6) (3)
140,644 k 11,189(6) 2,929 + 399 (6) 31,848 + 2,691 (6) 1,115+359 (3) 29,109 + 2,083 (6) 769+51 (3)
added after 66 hr of culture; cultures were harvested at 72 hr.
QUERE AND THORBECKE
472
TABLE 3 Effect of TGF-Bl on B Cell Proliferation [ 3H]Thymidine incorporation ’ (cpm -e SEM) (n)
Added to cultures TGF-P 1
10 20 50 0 10 20 50 0 10 20 50 0
0
1 rig/ml
10 rig/ml
’ One microcurie/well
Expt 1
Expt 2
8,601 f 721 (4) 9,575 + 871(4) 10,764 + 943 (4) 1,769 + 167 (4) 781 k 107 (4) 965 -t 109 (4) 1,121 f 157 (4) 736+77 (3) 1,055+212(4) 1,018 + 129 (4) ND 516?215(3)
ND 902 1 + 388 (6) 9440 + 409 (6) 1468 + 127 (6) ND 835 + 34 (6) 1108 k 126 (6) 346k 59 (4) ND 945 rf: 33 (6) 1035 + 93 (6) 317235 (4)
LPS Wml)
of [3H]thymidine
added after 66 hr of culture; cultures were harvested at 72 hr.
cell activity during the secondary anti-SRBC PFC response in chicken, either whole spleen cells or nylon wool-nonadherent spleen cells from the immune chickens were cultured separately with or without TGF-/31 and then added simultaneously with antigen on Day 2, after washing to remove the TGF-P 1, to other “assay” whole spleen cell cultures. Only cells which had been precultured with TGF-j31 suppressed the response of the assay cultures and the degree of suppression obtained (mean of 50.7 & 4.7%) was similar for whole spleen cells and for nylon wool-nonadherent spleen cells precultured with TGF-/31 (Table 4). Cells precultured without TGF-81 did not
TABLE 4 Suppressor T Cells Detectable after Incubation with TGF-PI Preincubated cells added on Day 2 Cells None Whole spleen NWNAC spleen
Mean PFC response/dish f SEM (n)”
Cultured with
Expt 1
Medium TGF-fi I b Medium TGF-@lb
1381 zk 298 (3) 956 f 177 (6) 370?69 (6) 1362 + 158 (6) 562+97 (6)
Expt 3
Expt 2 5555 t 334 7207a613(6) 3926 r 463 6254 k 356 3488 -+ 354
(3) (6) (6) (6)
2416 k 330 (3) ND ND 3324 zk 161(6) 1552 + 230 (6)
0 SRBC (50 ~1, 2%) added on Day 2 to SRBC-immune spleen cell assay cultures, PFC determined on Day 4. ’ TGF-j3 1 concentration in medium from Days O-2 was 1 rig/ml. It was washed out before the cells were added to the assay cultures on Day 2. c NWNA = nylon wool nonadherent.
SUPPRESSION
5x103 8 B 1x103 1
OF IMMUNE
7f ff f$ L 107
5X108
25x10’
RESPONSE BY TGF-@I
1.25X106
473
0
FcmberOfcdlSAddSd
FIG. 1. Dose-related inhibition of the anti-SRBC PFC response after addition of T cells precultured for 2 days with TGF-/31. T cells added to culture of SRBC-immune spleen cells (lO’/dish) on Day 2 of culture were precultured for 2 days with (Cl) or without(H) TGF-fil (1 ngjml). Control response of immune spleen cells without added T cells is indicated by the solid bar.
suppress the response of “assay” cultures. The degree of suppression obtained by leaving TGF+?l in the assay cultures throughout the culture period was somewhat greater than that obtained with cells precultured with TGF-Bl only, suggesting that TGF-/3 1 also had additional suppressive effects on the immune response. Results in Fig. 1 show the suppressive effect of different numbers of nylon woolnonadherent spleen cells precultured for 2 days with or without TGF-@I . Suppression in assay cultures was not obtained, regardless of the cell numbers added, when cells were taken from cultures without TGF-/3 1, whereas a dose-related decrease of the response was obtained with cells precultured with TGF-B 1. Nature of the cells aflected by TGF-@I during preculture. It seemed important to determine the phenotype of the cells which were affected by TGF-81 in culture to express suppressor activity. The results in Table 5 show that pretreatment of nylon wool-nonadherent cells from SRBC-immune chickens to remove CT4+ cells prior to the a-day culture with TGF-@I resulted in the same suppressor activity as that obtained with unfractionated T cells. On the other hand, pretreatment to remove CT8+ cells completely abolished suppressor activity (experiments l-3). Since in all the previous experiments the nylon wool-nonadherent cells, precultured with and without TGF-Pl, had been taken from the same SRBC-immune chicken as the responder cells to which they were added on Day 2 of culture, it seemed important to determine whether TGF-/3 1 would also cause the appearance of suppressor activity in spleen cells from uninjected chickens. The results of experiment 4 (Table 5) show that a significant suppression (27.5%) was obtained with unfractionated cells from normal chickens that had been precultured with TGF-@I , but not with cells precultured with medium alone or with CT8+-depleted T cells precultured with TGF-/31 . It should be noted, however, that T cells taken from normal chickens and added without preculture to SRBC-immune responder cells on Day 2 of culture together with antigen, reduce the response to approximately the same degree as do cells precultured with TGF-P 1 (30, and results not shown). These cells evidently disappear after being cultured for 2 days, unless TGF-/31 is added to the medium.
474
QUERE AND THORBECKE TABLE 5
Identification of Suppressor Cells, Detectable aher Exposure to TGF-61 in V&V, as CT4-, CT8+ T Cells Preincubated NWNA* spleen cells Mean PFC response+ SEM (a)** Added on hY2
None Unfractionated CT4+ celldepleted CT8+ celldepleted
Cultured with
Expt I
Medium TGF-@1 Medium TGF-fl I Medium TGF-8 1
4143+483(6)' 3966?476(11)" 2574f267(12)"*d 4904k 775(6)' 2770?423(6)/ 6165 ?428(6)' 6165+694(6)d
Expt 2 5508 2 171(6)' 7093k 346(10)b 5431k 322(12)"e 6198+660(6)' 3770+293(6)a 7379 ?440(6)' 7141 f 369(6)'
Expt 3 12,084+ 1,344(3) 10,378 -c 594 (6)' 6,764+ 876 (6)' 8,447 f479 (6)h 6,476 +299 (6)h
ND
Expt 4 4522f487(3) 4539 +266(6)' 3291k384(6)k
ND ND 4665 f 386(6) 4259+484(5)
* Nylon wool-nonadherent (NWNA) spleen cells (S-10 X 106), unfractionated or CT4+ or CT8+ celldepleted as indicated and precultured for 2 days + 1 ng TGF-fil at 37’C, were washed and added to responder spleen cells on Day 2 of culture together with antigen (50 ~1,2% SRBC). Donor of these “suppressor cells” was either a normal chicken (experiment 4) or the same SRBC immune chicken used for responder spleen cells (experiments l-3). ** On Day 0 responder spleen cells were taken from chickens injected iv with 0.25 ml 5% SRBC on Day -4. They were cultured at 37°C (10’ cells/dish). Antigen + “suppressor cells” were added on Day +2 and PFC were determined on Day +4. Background PFC (-antigen) were subtracted (mean background for these experiments was 569 f 133 PFC/dish). ‘-’ Statistical significance of differences between values indicated by the same letter: "P -c 0.02; 'P = 0.002; cP< 0.01; dP< o.OOO1;eP < O.OOS;'P = O.O4;8P= 0.01 hP= 10.01; 'P = O.Ol;‘P = 0.02; kP = 0.02.
To summarize results from this series of experiments, the mean degree of suppression obtained with unfractionated nylon wool-nonadherent spleen cells from SRBCimmune chickens in seven consecutive experiments, comparing TGF-/3 1-exposed to control cells, was 32%. Comparing TGF-81 exposed CT8+-depleted cells to their own controls, the suppression was 1.3%, and comparing TGF-01 exposed CT4+depleted cells to their own controls, the suppression was 35.3%. In experiments 1 and 2 (Table 5) the CT8+-depleted nylon wool-nonadherent cells caused a significant enhancement of the antibody responses in the responder cell cultures. This was not seen with CT4+-depleted cells or with CTVdepleted cells from a normal chicken (experiment 4). This finding suggests that CT4+ cells from SRBCimmune chickens, particularly after removal of CT8+ cells and regardless of preculturing with TGF-6 1, tend to enhance the response of unfractionated immune spleen cells to SRBC. In previous studies we have also shown helper activity for antibody production in unfractionated precultured cells, particularly under conditions where a helper cell deficiency was created in the assay B cell cultures (20). The results in Table 6 show that removal of TCRl+ cells prior to incubation with TGF-P 1 did not affect the suppressor activity detected at the end of the 2-day incubation (3 1.9% suppression), while removal of TCR2+ cells abolished suppression ( 1.5% suppression). It is interesting to note that the cells remaining after TCR2+-cell depletion neither enhanced nor suppressed, suggesting that the CT4+ helper cells responsible for the enhancement in CT8+ depleted cells had also been depleted by removal of TCR2+ cells.
SUPPRESSION
OF IMMUNE
475
RESPONSE BY TGF-/31
TABLE 6 Identification of Suppressor Cells, Detectable after Exposure to TGF-@I in Vitro, as TCR2+ T Cells Preincubated NWNA * spleen cells Added on Day 2 None Unfractionated TCR 1’ cell-depleted TCR2+ cell-depleted
Mean PFC response/dish + SEM (n) **
Cultured with
Expt 1
Expt 2
Medium TGF-8 1 Medium TGF-8 1 Medium TGF-#l 1
3533 -t 594 (3) 2757 + 249 (5) 1890 * 394 (6)” ND ND 3871 + 594 (6) 3814 + 653 (6)”
4260 f 397 (3) 3545 t 527 (5) 2107 f 420(6) 3150 f 281(5)b 2144rt312(6)b
* Nylon wool-nonadherent (NWNA) spleen cells (5-10 X 106), unfractionated or TCRl+ or TCR2+ cell-depleted as indicated and precultured for 2 days + 1 ng TGF-Bl at 37”C, were washed and added to responder spleen cells on Day 2 of culture together with antigen (50 ~1,2% SRBC). ** On Day 0 responder spleen cells were taken from chickens, injected iv with 0.25 ml 5% SRBC on Day -4. They were cultured at 37°C (10’ cells/dish). Antigen + “suppressor cells” were added on Day +2 and PFC were determined on Day +4. Background PFC (-antigen) were subtracted (mean background for these experiments was 427 + 115 PFC/dish). ‘zb Statistical significance ofdifferencesbetween values indicated by the same letter: “P = 0.03; bP = 0.04.
DISCUSSION The present results suggest that I- 10 ng TGF-P 1/ml has multiple immunosuppressive effects on the immune response of chicken spleen cells in vitro. In the first place it reduces the T cell response to the lectin Con A during a 3&y incubation at 40°C by 50-80s. Secondly, it reduces B cell proliferation in response to LPS by >90% during a 3-day incubation at 40°C. Most interestingly, it has also an indirect immunosuppressive effect by sustaining/inducing a suppressive effect of CT8+ T cells on antibody production. This latter effect was indirect since the TGF-@I was washed off before the cells that had been exposed to it were added to the assay culture. This effect was mediated by CT8 TCR2+ cells and was quite similar to the suppression obtained by these cells after preculture with antigen. However, the suppressive effect of cells precultured with TGF-81 could also be obtained with normal rather than SRBCimmune spleen cells. It appeared, therefore, that antigen-specific suppressor cells were not an essential component of this TGF-Pl effect. It seems more appropriate to compare the suppressor cells present after culture with TGF-@l to those induced by con A in vitro. In the present studies the suppressor cells were found to be CT8+, TCR2+ (CT4-, TCRI-). A similar phenotype was found in other studies for the Con A-induced suppressor cells (P. Quere, M. D. Cooper, and G. J. Thorbecke, manuscript in preparation). The present results do not allow a distinction between the induction and/or maintenance by TGF-/31 of suppressor T cells during the first 2 days of spleen cell cultures. Fresh normal chicken spleen cells added at the time of antigen addition to assay cultures cause a certain degree of suppression of the antibody response. Thus, some activation of suppressor cells appears to occur in fresh normal or SRBC-immune T cells when they are mixed with spleen cells responding to antigen, but the phenotype of these cells has not been established. It is possible that TGF-@ 1 maintains
476
QUERE AND THORBECKE
the activity of these same cells in vitro, so that preculturing at 37°C does not result in a loss of suppressor activity. However, it is also possible that other CT8+ T cells are induced by TGF-p 1 to exhibit suppressor activity. Studies of the actual role of TGF-@ 1 in naturally occurring suppressor cell effects or in Con A activation of suppressor cells await the availability of neutralizing antibodies to TGF-P 1. In previous studies we showed that the loss of suppressor activity in cultures of immune chicken spleen cells in the absence of antigen, which is seen at 37°C does not occur at 40°C (3 1). It is possible that under these conditions, a factor(s) is produced which, similar to TGF-@ 1, maintains or induces suppressor cell activity. It is of interest that TGF-P 1 inhibits chicken B cell proliferation to LPS to a greater extent than T cell proliferation to Con A. This high sensitivity of B cell proliferation to the suppressive effect of TGF-p 1 is in contrast with the relative resistance of antibody production in vitro when both TGF-@ 1 and antigen are added on Day 2 of culture. It appears from these observations that differentiation of B cells is much less inhibited (if at all) than B cell proliferation. This impression is also gained from recent data on the effect of TGF-61 on murine B cells, where it is found to enhance the isotype switch to IgA and thus has a B cell differentiation-promoting effect (32). ACKNOWLEDGMENTS We are greatly indebted to Dr. M. D. Cooper (Division of Clinical Immunology, Department of Pediatrics, University of Alabama, Birmingham) for his generous gifts of monoclonal antibodies to the chicken antigens CT4, CT8, and TCR 1. We thank Dr. M. A. Palladino (Genentech Corp., South San Francisco, CA) for generously providing the human TGF-PI used in these studies.
REFERENCES 1. Spom, M. B., Roberts, L. M., Wakefield, L. M., and Assoian, R. K., Science 233,532,1986. 2. Massague, J., Cell49,437, 1987. 3. Assoian, R. K., Komoriya, A., Meyers, C. A., Miller, D. M., and Spom, M. B., J. Biol. Chem. 258, 7155,1983. 4. Derynck, R., Jarrett, J. A., Chen, F. Y., Eaton, D. H., Bell, J. R., Assoian, R. K., Roberts, A. B., Spom, M. B., and Goeddel, D. V., Nature(London) 316,701,1985. 5. Kehrl, J. H., Roberts, A. B., Wakefield, L. M., Jakowlew, S., Spom, M. B., and Fauci, A. S., J. Immunol. 137,3855, 1986. 6. Blomhoff, H. K., Smeland, E., Mustafa, A. S., Godal, T., and Oholsson, R., Eur. J. Zmmunol. 17,299, 1987. 7. Lee, G., Ellingsworth, L. R., Gillis, S., Wall, R., and Kincade, P. W., J. Exp. Med. 166, 1290, 1987. 8. Petit-Koskas, E., Genot, E., Lawrence, D., and Kolb, J. P., Eur. J. Zmmunol. 18, 111, 1988. 9. Smeland, E. B., Blomhoff, H. K., Holte, H., Ruud, E., Beiske, K., Funderud, S., Godal, T., and Ohlsson, R., Exp. Cell. Res. 171,213, 1987. 10. R&ton, H. J., Proc. Natl. Acad. Sci. USA 83,5531, 1986. 11. Kehrl, J. H., Wakefield, L. M., Roberts, A. B., Jakowlew, S., Alvarez-Mon, M., Derynck, R., Spom, M. B., and Fauci, A. S., J. Exp. Med. 163, 1037,1986. 12. Ranges, G. E., Figari, I. S., Espevik, T., and Palladino, M. A., J. Exp. Med. 166,99 1, 1987. 13. Wrann, M., Bodmer, S., Martin, R., Siepl, C., Hofer-Watbinele, R., Frei, K., Hofer, E., and Fontana, A., EMBO J. 6, 1633, 1987. 14. Delarco, J. E., and Todaro, G. J., Proc. Natl. Acad. Sci. USA 75,4001, 1978. 15. Coffey, R. J., Shipley, G. D., and Moses, H. L., Cancer Rex 46,1164, 1986. 16. Derynck, R., Goeddel, D. V., Ulhich, A., Gutterman, J. U., Williams, R. D., Bringman, T. S., and Berger, W. H., Cancer Rex 47,707, 1987. 17. Newcom, S. R., Kadin, M. E., and Ansari, A. A., Amer. J. Pathol. 131,569, 1988. 18. North, R. J., Adv. Cancer Res. 45, I, 1985. 19. Edelman, A. S., Robinson, M. E., Sanchez, P., andThorbecke, G. J., Cell. Immunol. 110,321,1987.
SUPPRESSION 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32.
OF IMMUNE
RESPONSE BY TGF-81
477
Chi, D. S., Grebenau, M. D., and Thorbecke, G. J., Eur. J. Immunol. 10,203, 1980. Chan, M. M., Chen, C. H., Ager, L. L., and Cooper, M. D., J. Immunol. 140,2133, 1980. Sowder, J. T., Chen, C. H., Ager, L. L., Chan, M. M., and Cooper, M. D., J. Exp. Med. 167,315,1988. Cihak, J., Ziegler-Heitbrock, H. W. L., Trainer, H., Schranner, I., Merkenschlager, M., and Losch, U., Eur. J. Immunol. 18,533, 1988. Julius, M. H., Simpson, E., and Helzenberg, L. A., Eur. J. Immunol. 3,645, 1973, Wysocki, L. J., and Sato, V. L., Proc. Natl. Acad. Sci. USA 752844, 1978. Mishell, R. I., and Dutton, R. W., .I. Exp. Med. 126,423, 1967. Nakamura, K., J. Exp. Med. 135,476, 1972. Jerne, N. K., Nordin, A. A., and Henry, C., In “Cell Bound Antibodies” (B. Amos, and H. Koprowski, Eds.), (p. 109). Wistar Institute Press, Philadelphia, 1963. Muchmore, A. V., Koski, I., Dooley, N., and Blaese, R. M., J. Immunol. 116, 1016, 1976. Chi, D. S., and Thorbecke, G. J., In “Avian Immunology: Basis and Practice” (A. Toivanen and Towanen, Eds.), p. 16 1. CRC Press, Boca Raton, FL, 1987. Bhogal, B. S., Chi, D. S., and Thorbecke, G. J., Proc. Sot. Exp. Biol. Med. 176,414, 1984. Lebman, D., Lee, F., and Coffman, R., FASEB J. 2, A1268, 1989.