Cellular interaction between subsets of T8 population for maximal suppression of antigen-specific antibody response

CELLULAR

IMMUNOLOGY

88, 75-84 (1984)

Cellular Interaction between Subsets of T8 Population for Maximal Suppression of Antigen-Specific Antibody Response’ CHIKAO MORIMOTO,~ JUDY ANN DISTASO, JOHN J. CHENEY, ELLIS L. REINHERZ, AND STUART F. SCHLOSSMAN Division of Tumor Immunology, Dana-Farber Cancer Institute, and the Department of Medicine, Harvard Medical School, Boston, Massachusetts 02115 Received March 8, 1984; accepted April 26, 1984 The characterization of human keyhole limpet hemocyanin (KLH)-specific T8 suppressor cell (TS KLH) generated in vitro with high doses of antigen by using peripheral blood lymphocytes is described. The cellular basis for the generation of specific suppressor-effecter functions was examined and it was shown that radioresistant T8 KLH cells could induce a second set of radiosensitive suppressor-effecter cells found in a freshly isolated T8 population. Moreover, the T8 KLH population could be divided into T8 KLH TQI+ and T8 KLH TQlsubsets.Both subsetswere required for maximal suppressionof the anti-DNP antibody response, since neither subset alone induced more than minimal suppression. These results demonstrated that several functionally distinct T8 subpopulations of cells exist, and it is suggestedthat further resolution of these complex immunological networks in man will be facilitated by the development of unique reagents capable of defining the heterogeneity of the cells involved in suppressor functions.

INTRODUCTION Recent studies have demonstrated that T-cell subsets are involved in a complex series of interactions that regulate the immune response (1, 2). Multiple cell-cell interactions appear to be necessary to translate an antigenic signal into both an effective immune response and its homeostatic controls. Human T cells can be divided into T4+ and T8+ subsets of cells which are functionally distinct (3-7). Moreover, communicative interactions occur between and within these two major populations of cells in the generation of specific effector functions. For example, interactions between subpopulations of T4 and T8 cells are required to induce suppression of Ig production in an antigen-specific and pokeweed mitogen (PWM) system or in an autologous mixed leukocyte reaction (AMLR) system (8-11). Similarly, differentiation of T8 cytotoxic effecters from precytotoxic T8 lymphocytes in mixed leukocyte reaction (MLR) requires induction by T4+ T cells (5) and, following antigen stimulation, both T4+Ia+ and T4+Ia- subsets are required to induce optimal Ig secretion by B cells (12). Furthermore, it has been suggestedthat ’ This work was supported in part by NIH Grants AI 12069 and CA 19589. 2 To whom correspondence should be addressed at: Division of Tumor Immunology, Dana-Farber Cancer Institute, 44 Binney Street, Boston, Mass. 02115. 75 0008-8749/84 $3.00 Coptight 0 1984 by Academic F’ress,Inc. All rights of reproduction in any form reserved.

76

MORIMOTO

ET AL.

the human immunoregulatory circuit is composed of discrete subsets of still poorly defined cells which interact to maintain homeostasis ( 11, 13). Recently, Gatenby et al. have suggested the existence of a suppressor-amplifier circuit in the AMLRstimulated Ig synthesis system (11). They have shown that AMLR-activated Leu 2 Iaf (T8+Iat) cells act to amplify or induce suppressor effects of fresh Leu 2 (T8+) cells (11). In earlier studies, we utilized a primary in vitro anti-dinitrophenol-keyhole limpet hemocyanin (anti-DNP-KLH) system which is T cell dependent and results in specific antibody of the IgM isotype (14). In this system, it was shown that antigenspecific T8 suppressor cells can be induced with high doses of antigen and that the suppression obtained is antigen specific (15). In the present study, we further examined the cellular interactions of subsetsof T8+ suppressor cells in the generation of specific suppressor-effector functions. MATERIALS

AND METHODS

Isolation of lymphoid populations. Human peripheral blood mononuclear cells were isolated from the blood of healthy volunteer donors by Ficoll-Hypaque density gradient centrifugation (Pharmacia Laboratories, Piscataway, N.J.). Unfractionated mononuclear cells were depleted of macrophagesby adherence to plastic as described (16). Subsequently, macrophage-depleted mononuclear cells were separated into E-rosette-positive (E+) and E-rosette-negative (E-) populations with 5% sheep erythrocytes (Microbiological Associates, Bethesda, Md.). The rosetted mixture was layered over Ficoll-Hypaque and the recovered E+ sediment was treated with 0.155 M NH&l to lyse erythrocytes. The T-cell population obtained was >95% E+ and >94% reactive with a monoclonal antibody, anti-T3, which defines an antigen present on all mature peripheral T lymphocytes (17). The E- population was highly enriched for B cells by complement (C)-mediated lysis with anti-monocyte antibody, anti-MO 1. This rationale was employed since it has been shown that E- populations contain many null cells which are reactive with anti-MO1 antibody (18). Reanalysis of anti-MO 1-lysed subpopulations of E- cells (B cells) demonstrated ~5% anti-MO lreactive cells and >90% of cells were reactive with the anti-B1 monoclonal antibody which defines an antigen present on all peripheral B cells (19). Monoclonal antibodies. Three monoclonal antibodies, termed anti-T4, anti-T8, and anti-TQ 1, were used in the present study. Their production and characterization are described elsewhere (1, 3-5, 20). In brief, anti-T4 was previously shown to react with 60% of peripheral T cells (3, 4), whereas anti-T8 defined 30% of T cells (5). Because T4 and T8 are of IgGz subclass and fix C, they were employed for C-mediated lysis. Anti-TQl antibody was reactive with 70-85% of the T4 cells but also reacted with 50% of T8+ cells (20). For the present studies, the anti-TQl which fixes C (IgG2) was used for the complement-mediated lysis of IUH-stimulated T8 cells (T8 KLH) and that which did not fix C (IgG,) was used for separation of T8+TQ l+ and TS+TQl - T cells by the plate technique. Antigens. DNP-102 KLH and DNP-bovine serum albumin (DNP-,SBSA) were provided by Dr. Yves Bore1 (Children’s Hospital Medical Center, Harvard Medical School, Boston, Mass.). The IUH was obtained from Pacific Bio-Marine Supply Co. (Venice, Calif.) and prepared as previously described (15). Complement-dependent lysis of lymphocytes with monoclonal antibodies. E+

CELLULAR

INTERACTION

BETWEEN

SUBSETS OF T8 CELLS

77

lymphocytes were treated with anti-T4 or anti-T8 monoclonal antibodies and rabbit complement (PelFreez, Rogers, Ark.). These antibodies of the IgG2 subclass have been described elsewhere (3-5). Briefly, aliquots (15 X 10’ cells) were incubated with 1 ml of antibody at a 1:250 dilution for 1 hr at room temperature and then 0.3 ml rabbit C was added to the mixture. The mixture was incubated for another hour in a 37°C shaking water bath and washed; residual cells were cultured overnight at 37°C. After lysis of cells with anti-T4 and C, >90% of the residual cells were T8+ cells, whereas ~5% were T4+ cells, after lysis with anti-T8 and C, >90% of the remaining cells were T4+ cells and ~5% were T8+ cells. These two populations will be referred to as T8+ and T4+ subsets, respectively. To eliminate T8 KLH+TQl+ cells, IUH-stimulated T8 cells ( 15 X 10’) were similarly treated with anti-TQl antibody (at 1:125) and rabbit complement. After lysis of cells with anti-TQl and C, ~5% were TQl+ cells. Separation of T8’TQl’ and T8+TQI- T-cell subsets. A modification of the plate technique previously described was utilized to separate T8+TQ I+ and T8+TQ 1subsets(13). Briefly, IUH-stimulated T8 cells (12 X 106)were incubated with 1 ml of anti-TQ 1 antibodies at a 1:125 dilution for 1 hr at 4°C and washed to remove excessantibody. Cells ( 12 X lo6 in 3 ml volume) were then applied to an affinitypurified goat anti-mouse Ig-coated plate (Fisher Scientific, Springfield, N.J.; No. 8-757- 12). After 70 min of incubation at 4°C nonadherent and adherent populations were collected. Less than 4% of the nonadherent populations were TQlf and >95% of the adherent population was TQI+. Cells from these adherent and nonadherent populations were respectively designated as TQl+ and TQl- populations. Culture conditions. Unseparated and separated populations of lymphocytes containing 5% adherent cells were cultured for 5 days with 0.6 pg per culture of DNPKLH in a total volume of 200 ~1 per flat-bottomed microtiter plate (Falcon Plastics, Oxnard, Calif.) at 37°C in a humid atmosphere with 5% CO2 in RPM1 1640 supplemented with 20% fetal calf serum (FCS) (Grand Island Biological Co. (GIBCO), Grand Island, N.Y.), 200 mM L-glutamine, 25 mM 4-(2-hydroxyethyl)1-piperazinethanesulfonic acid (Hepes) buffer, 0.5% sodium bicarbonate, and 1% penicillin-streptomycin. At the completion of the culture period, the cells were washed three times to remove antigen, resuspended in the same medium as described above, and incubated for 5 additional days at 37°C in a humid atmosphere with 5% COZ. Culture supernatants from five individual wells were pooled and stored at -20°C before radioimmunoassay (RIA) for anti-DNP antibody. To assessthe possible cellular interactions for generating suppressor-effector cells, varying numbers of autologous KLH-stimulated and 5 X lo4 fresh T8 cells were added to a mixture of 2 X lo5 T4 cells and 4 X lo5 B cells with DNP-KLH. In some experiments, fresh and/or KLH-stimulated T8 cells were irradiated (1250 R) by using a Gamma Cell 40 (Atomic Energy of Canada, Ltd., Commercial Products, Ottawa, Ontario) irradiation source. RIA for anti-DNP antibody. The RIA used for measurement of anti-DNP antibody in culture medium was the solid-phase RIA technique described earlier (10, 14). Briefly, DNP-BSA was absorbed onto flexible polyvinyl plates (Cooke Laboratory Products, Alexandria, Va.) at a concentration of 1 mg/ml for 2 hr at 4°C. Next, excessDNP-BSA antigen was removed by repeated rinsing of wells and any further nonspecific binding sites were blocked by incubation with 1% BSA in phosphatebuffered saline (PBS). After additional washing, 50 ~1 of culture supernatant was

78

MORIMOTO

ET AL.

added to each well, incubated for 3 hr at 4”C, and washed three times. Finally, 25 ~1 containing lo5 cpm of ‘251-labeledimmunoabsorbent purified rabbit anti-human F(ab’)2 was added to each well and was incubated overnight at 4°C. After unbound “‘1 was removed by extensive washing, the plates were cut out with scissors, and each well was counted for bound radioactivity in a gamma counter. Anti-DNP antibody production secreted into culture supematants was assayedin triplicate and was expressedas cpm + SE (Fig. 1). Medium background values (20% FCS in RPM1 1640) were usually below 200 cpm and were subtracted from experimental values. In vitro induction of suppressor cells. In vitro cell culture procedures used in this experiment have been described previously ( 15). Briefly, 4 X lo6 unfractionated Ef lymphocytes (T cells) with 5% macrophages were cultured in a l-ml volume of final medium with 100 pg/ml of IUH for 5 days in Linbro 24-well flat-bottomed plates (Flow Laboratories, Maler, Va.) at 37’C in 5% C02. At the completion of the culture period, the unseparated T cells were collected and washed three times. Subsequently, antigen-stimulated T8 cells were obtained by C-dependent lysis with anti-T4 antibody. RESULTS Eflect of KLH-Stimulated T8 Cells on Anti-DNP Antibody Response by Fresh T4B-Cell Mixtures In earlier studies, the suppressor function of KLH-stimulated T8 cells on the anti-DNP antibody response have been investigated by the addition of IUHstimulated T8 cells to fresh T-B-cell mixtures triggered by DNP-IUH. In the present study we undertook to determine whether the cellular interactions between IUH-stimulated T8 cells obtained after 5 days in culture and other subpopulations of T8 cells were required to generate a suppressor response. For this purpose, we Antl-DNP 0

2

Antibody 4

6

WM~lO‘~/Culture) 6

10

12

0 1 o^ b ; =

2 10

P p 25

0 1 2 10

FIG. 1. Effect of antigen-specific T8 cells on anti-DNP antibody response by a T4-B-cell mixture. Unseparated T cells were cultured with 100 pg of KLH in the presence of 5% macrophages for 5 days. Antigen-stimulated TS suppressor cells (T8 KLH) were then obtained after lysis of T4+ populations. Increasing numbers of T8 KLH cells were then added to a fresh autologous T-B- or T4-B-cell mixture (2 X IO’ T or T4 and 4 X 10’ B cells). Results are representative of five experiments performed. Values are expressedas mean cpm/culture + SE of triplicate samples.

CELLULAR

INTERACTION

79

BETWEEN SUBSETS OF T8 CELLS

compared the effect of KLH-stimulated T8 cells on the anti-DNP antibody response of freshly isolated autologous T4-B-cell mixtures which were depleted of isolated T8 cells and of fresh autologous T-B-cell mixtures. As shown in Fig. 1, when low numbers of T8 KLH cells were added to fresh autologous T-B-cell mixtures, marked suppression of anti-DNP antibody response was observed as previously described (15). However, when the same low numbers of T8 KLH cells were added to fresh autologous T4-B-cell mixtures lacking fresh T8 cells, no significant suppression of the anti-DNP antibody response was seen. Similar results were obtained in five consecutive experiments with cells from five different individuals. These results raised the possibility that interaction between T8 KLH cells and other populations of T8 cells may be necessaryfor the maximal suppression of the anti-DNP antibody response. Fresh T8 Cells Are Required for the Maximal Suppression of Anti-DNP Antibody Response To directly demonstrate that cellular interactions between T8 KLH cells and fresh T8 cells were required for the maximal suppression of an anti-DNP antibody response, the following experimental design was employed. Unseparated T cells were first cultured with 100 pg of KLH in the presence of 5% macrophages for 5 days and T8 KLH cells were then obtained by lysis of the T4 populations. Varying numbers of T8 KLH cells were then added to a fresh T4-B-cell mixture (2 X lo5 T4 and 4 X lo5 B cells) with or without 5 X lo4 fresh T8 cells in the presence of DNP-KLH. As shown in Table 1, when increasing numbers of T8 KLH cells were TABLE I Requirement of Fresh T8 Cells for Maximal Suppression of Anti-DNP Antibody Response T8 cells added” Anti-DNP antibody response? T8 KLH (X 10’)

T8 fresh (X104)

Expt 1

Expt 2

Expt 3

0

0

2 5 10 20

0 0 0 0

12,130 12,850 (-6)’ 12,540 (-3) 11,840 (2) 11,410 (6)

9,750 9,210 (6) 11,030 (-13) 9,380 (4) 9,520 (2)

14,910 ND 13,550 (9) 14,370 (4) 13,190 (12)

0 2 5 10 20

5 5 5 5 5

12,550 4,2 10 (66) 6,030 (52) 4,740 (62) 4,020 (68)

9,500 4,120 (57) 3,950 (58) 3,410 (64) 2,230 (77)

14,700 4,730 (68) 6,820 (54) 3,410 (77) 3,650 (75)

a Unseparated T cells were cultured with 100 pg of KLH in the presenceof 5% macrophages for 5 days. KLH-specific T8 suppressorcells (T8 KLH) were obtained after lysis ofT4+ populations. Increasing numbers of T8 KLH were then added to a fresh autologous T4-B-cell mixture (2 X 10’ T4+ and 4 X 10’ B cells) in the presence or absence of 5 X lo4 fresh T8 cells. b Results are mean cpm/culture of triplicate samples. SE was < 10%. ‘Numbers in parentheses = ‘% suppression calculated as [I - (cpm experimental culture/cpm control culture)] X 100.

80

MORIMOTO

ET AL.

added to T4-B-cell mixtures lacking fresh T8 cells, no significant suppression was seen. However, when increasing numbers of T8 KLH cells were added to T4-B-cell mixtures containing fresh T8 cells, marked suppression of anti-DNP response was observed. As shown, as few as 2 X lo3 T8 KLH cells were sufficient to provide a maximal signal to the fresh T8 effector cells, since larger numbers of T8 KLH cells had no further suppressive effect upon the response. Moreover, when small numbers of T8 IUH cells were added to T4-B-cell mixtures containing fresh T8 cells in the presence of DNP-fowl gammaglobulin (DNP-FGG) or small numbers of FGGstimulated T8 cells were added to T4-B-cell mixtures containing fresh T8 cells in the presence of DNP-KLH, no significant suppression was obtained (data not shown), confirming that this system was antigen specific as previously described (15). These results support the view that an interaction between T8 IUH cells and fresh T8 cells is necessary for maximal suppression of antigen-specific antibody response. Diferent Radiosensitivity of T8 Suppressor T Cells The experiments described above suggestthat at least two distinct subpopulations of T8 cells are required for the maximal suppression of the anti-DNP antibody response. We next examined the differential radiosensitivity of these two populations of T8 cells. For this purpose, increasing numbers of irradiated or nonirradiated T8 IUH cells (obtained-after 5 days in culture) were added to a fresh autologous T4B-cell mixture (2 X lo* T4 and 4 X lo* B cells) in the presence of irradiated or nonirradiated fresh T8 cells (5 X 104) with DNP-KLH. As shown in Table 2, maximum suppression was seen when both the T8 KLH and fresh populations of T8 cells were unirradiated. Irradiation of the fresh T8 cells abolished suppression but irradiation of the T8 KLH cells alone had little effect on suppression of the anti-DNP antibody response. These results suggestthat radioresistant T8 KLH cells induce or amplify the suppressor-effector function of radiosensitive fresh T8 cells. Lysis of T8 KLH TQl+ T Cells Diminished the Suppression of Anti-DNP Antibody Response Previous studies have demonstrated that a monoclonal antibody termed antiTQl reacts with approximately 50% of unfractionated lymphoid cells (20). This antibody is not only reactive with T4 and T8 subsets but also reactive with B cells and null cells. Furthermore, it reacts minimally or not at all with thymocytes. When the T cells were isolated and fractionated, anti-TQl reacted with 70-85% of T4+ cells and approximately 50% of the T8+ population. In vitro studies indicates that the subpopulations of T4+ T cells delineated by anti-TQ 1 were functionally distinct, and that the T4+-TQl- but not the T4+TQl’ subset provided the majority of T-cell help for B-cell Ig production. T8 KLH T cells generated in a 5-day culture system were treated with monoclonal anti-TQl antibody (IgG2 class) and C to remove the TQl+ populations. The residual T8 IUH TQl- T cells were examined for their effect on anti-DNP antibody response. As shown in Table 3, when T8 KLH T cells were treated with anti-TQl and C and added to the T4-B-cell mixture containing fresh T8 cells, a modest degree of suppression of anti-DNP antibody response was observed (Table 3C). In contrast, the addition of untreated T8 KLH or of T8 KLH cells treated by anti-

CELLULAR

INTERACTION

81

BETWEEN SUBSETS OF T8 CELLS TABLE 2

Different Radiosensitivity of T8 Suppressor T Cells T8 cells added’ Anti-DNP antibody responseb T8 KLH (X10’)

T8 fresh (X 104)

Expt 1

Expt 2

Expt 3

0

0

0 2 5 10

5 5 5 5

12,600 13,120 (-4)’ 5,630 (55) 4,810 (62) 3,960 (69)

11,050 12,510 (-13) 5,400 (51) 6,120 (45) 4,460 (60)

2R 5R 10R

5 5 5

7,540 (42) 6,320 (50) 5,720 (55)

7,240 (34) 6,770 (39) 5,680 (48)

7,980 (35) 8,250 (31) ND

0 2 5 10

5R 5a 5s 5R

12,530 (0) 11,760 (7) 10,970 (13) 12,130 (4)

11,430 (-3) 12,610 (-14) 11,560 (-5) 13,080 (-18)

11,510 (7) 11,730 (5) 12,570 (-2) ND

2R 5R 10R

5a 5s 5s

12,350 (2) 12,770 (0) 11,910 (6)

12,560 (-14) 13,150 (-:!I$) (’ 12,720 (~1~5);

11,180 (10) 10,690 (13) ND

12,330

1w33 (0) 6,120 (50) 6,550 (47) ND

’ Antigen-specific T8 suppressor cells (T8 KLH) were obtained as described in the notes to Table 1. Increasing numbers of irradiated or nonirradiated T8 KLH were then added to a fresh autologous T4B-cell mixture (2 X 10’ T4 and 4 X 10’ B) in the presence of irradiated or nonirradiated 5 X IO4 fresh T8 cells. R = 1250-R irradiation. b Results are mean cpm/culture of triplicate samples. SE was < 10%. ’ Numbers in parentheses = I suppression calculated as described in the notes to Table 1.

TQl without complement resulted in marked suppression of the anti-DNP antibody response (Table 3A, B). These results suggestthat either T8 KLH TQl+ is required to induce fresh T8 cells to suppress or interactions between both T8 KLH TQl+ and T8 KLH TQl- populations are necessaryto induce the fresh T8 population to suppress it. The Cellular Interaction between T8 KLH TQl+ and T8 KLH TQI- Populations Are Necessaryfor the Maximum Suppression of Anti-DNP Antibody Response To test these alternatives, T8 KLH T cells were treated with anti-TQl antibody (IgG, class) and fractionated into T8 IUH TQ 1+ and T8 IUH TQ 1- T cells on an anti-mouse Ig-coated plate. The isolated subpopulations of T8 IUH T cells were added to a fresh T4-B-cell mixture containing fresh T8 cells in the presence of DNP-KLH. As shown in Table 4, the addition of each subset of T8 KLH T cells alone resulted in minimal suppression of anti-DNP antibody response (Table 4B, C). In contrast, when both subsets of T8 KLH T cells were mixed in a 1:1 ratio, marked suppression of anti-DNP antibody response was obtained (Table 4D). These results clearly suggest that the interactions between T8 KLH TQl+ and T8 IUH TQl- populations may be necessary for the maximal suppression of anti-DNP antibody response.

MORIMOTO

82

ET AL.

TABLE 3 Lysis of KLH-Specific TS+TQ I + T Cells Diminishes the Suppression of Anti-DNP Antibody Response Anti-DNP antibody responseb Culture condition”

Expt 1

Control

14,040

Expt 2 12,950

A. Addition of T8 KLH 2 x 10’ 5 x 103 10 x 103

2,270 (84)’ 4,640 (67) 3,140 (78)

3,570 (67) 3,890 (70) 960 (93)

B. Addition of T8 KLH treated by anti-TQl alone 2 x 10’ 5 x 10’ 10 x 103

2,350 (83) 3,560 (82) 2,980 (79)

3,270 (75) 2,940 (77) 1,130 (91)

C. Addition of T8 KLH treated by anti-TQI + C 2 x 103 5 x 103 10 x 103

9,9 10 (29) 9,560 (32) 10,750 (23)

13,440 (-4) 9,090 (30) 7,050 (46)

’ Each culture contained 4 X 10s B cells containing 5% M&, 2 X IO’ T4 cells and 5 X 10” T8 cells in the presence of DNP-KLH in the culture. bResults are mean cpm/culture of triplicate samples. SE was < 10%. ’ Numbers in parentheses = % suppression calculated as described in the notes to Table 1.

TABLE 4 Collaborations Between T8 KLH TQl+ and TQl- Populations Are Necessary for Optimal Suppression of Anti-DNP Antibody Response Anti-DNP antibody response” Culture condition Control

Expt 1 11,320

Expt 2 9,980

Expt 3 10,410

A. Addition of T8 KLH 2 x 10’ 5 x 10’

2,410 (79)’ 3,100 (73)

2,740 (73) 3,080 (69)

4,460 (57) 2,350 (77)

B. Addition of T8 KLH TQl+ 2x 10’ 5 x 10”

8,030 (29) 6,980 (38)

6,540 (34) 6,150 (38)

16,550 (-59) 9,010 (13)

C. Addition of T8 KLH TQl2 x lo3 5 x 10’

7,870 (30) 6,610 (42)

6,940 (3 1) 6,490 (35)

12,770 (-23) 8,430 (19)

D. Addition of T8 KLH TQl+ and T8 KLH TQl- (1:l) 1 X 1O’each 2 x 10’ each

2,120 (81) 2,050 (82)

2,510 (75) 2,250 (77)

4,050 (61) 1,980 (81)

n Results are mean cpm/culture of triplicate samples. SE was < 10%. b Number in parentheses = % suppression calculated as described in the notes to Table I.

CELLULAR

INTERACTION

BETWEEN SUBSETS OF T8 CELLS

83

DISCUSSION In the present study, we have extended previous observations demonstrating that KLH-specific T8 suppressor cells could be generated in vitro (15) and examined the cellular basis for the generation of specific suppressor-effector functions. The studies reported above indicate that radioresistant T8 IUH cells can induce another set of radiosensitive suppressor-effector cells found in a freshly isolated T8 population. Furthermore, the collaboration between T8 KLH TQl+ and T8 KLH TQll populations appears to be necessary to generate maximal suppression of anti-DNP antibody response in this system. In earlier studies, we had examined the antigen-specific suppressor function of T8 KLH cells by adding these cells to a freshly isolated T-B-cell mixture in the presence of DNP-KLH. Those results indicated that these cells could readily induce the suppression of the anti-DNP antibody response (15). The results described in the present studies indicate that in the absence of fresh T8 cells, addition of T8 IUH cells to a T4-B-cell mixture results in no significant suppression of the antiDNP antibody response. However, when T8 KLH cells are added to the T4-B-cell mixture containing fresh T8 cells, marked suppression of the response is observed. Moreover, only small numbers of T8 KLH cells are required for this effect. Further evidence that more than one T8 population is involved in this suppressor cell pathway was obtained by investigating the radiosensitivity of the T8 KLH and fresh T8 population. These studies revealed that T8 KLH cells are relatively radioresistant whereas the suppressor-effector functions of fresh T8 cells are radiosensitive. T8 KLH cells are required to “educate” radiosensitive suppressor-effecter cells within the fresh T8 population. The results clearly support the notion that a T8 KLH cell is required to induce another set of suppressor-effector cells present in the freshly isolated T8 populations. In mice, several related but nonidentical suppressor-regulatory circuits consisting of a variety of T-T interactions have been described, each providing a distinct step in a circuit leading to the inhibition of an immune response (21-23). For example, in the GAT system in mice, suppressor-inducer cells were required to induce a population of second order suppressor T cells. The latter population was shown to be cyclophosphamide resistant and triggered a third set of effector cells (22). Gatenby et al. recently reported that Leu2+DR+ T cells (T8+Ia+) interacted with unactivated fresh T8+ cells to effect maximal suppression of Ig synthesis in AMLR. In earlier studies, we demonstrated that KLH-stimulated T8 cells (T8 IUH cells) were Ia+ and distinct from the unactivated population. These results suggest that during initial incubation of T cells with KLH, IUH-specific suppressor T8 cells are generated which are Ia positive. It appears that these cells induce a second set of suppressor-effector cells present in the fresh T8 cells. The precise cellular mechanism of induction of KLH-specific T8 suppressor cells is not clear. Our earlier studies using PWM and an antigen-driven system have provided evidence for the requirement of a radiosensitive T4+JRA+ cells to induce suppressoractivity within T8 populations (9, 10, 13). Whether T4+JRA+ cells are required for the generation of KLH-specific T8 cells and how antigen-plus-macrophage induces T4+JRA+ suppressor-inducer cells (in contrast to the T4+JRA- helper cells) are presently under investigation. The phenotypic characterization of these T8 KLH cells by anti-TQl antibody is also of interest. These results indicate that the generation of a specific suppressor-

84

MORIMOTO

ET AL.

effector cell from fresh T8 cells requires the collaboration between T8 KLH TQ 1+ and T8 KLH TQl- populations. These findings are similar to a recent report using Leu 2+ and Leu 8+ antibodies in an AMLR system (24). Thus, the different radiosensitivity of T8 populations and the requirement of collaboration between T8+TQl+ and T8+TQl- populations in the generation of suppressor-effector function demonstrate that several distinct functional subpopulations are present within the T8 population. At a minimum, using both mitogen- and antigenactivated systems, a T4+JRA+TQlf subset of cells is required to interact with antigen and macrophage to induce a T8 presuppressor population. This population appears to be heterogeneous and contains the T8+TQ 1+ and T8+TQl- populations. Freshly isolated T8+ populations also appear necessaryfor the induction of suppression. Obviously the resolution of this complex network of interacting cells in the human systems will require both improved isolation techniques and better reagents capable of distinguishing unique subsets of inducer and suppressor cells. REFERENCES 1. Reinherz, E. L. and Schlossman, S. F., N. Engl. J. Med. 303, 370, 1980. 2. Germain, R. N. and Benacerraf, B., Scund. J. Immunol. 13, 1, 1981. 3. Reinherz, E. L., Kung, P. C., Goldstein, G., and Schlossman, S. F., Proc. Nutl. Acud. Sci. USA 76, 4061, 1979. 4. Reinherz, E. L., Kung, P. C., Goldstein, G., and Schlossman, S. F., J. Immunol. 123, 2894, 1979. 5. Reinherz, E. L., Kung, P. C., Goldstein, G., and Schlossman, S. F., J. Immunol. 124, 1301, 1980. 6. Ledbetter, J. A., Evans, R. L., Lipinski, M., Cunningham-Rundles, C., Good, R. A., and Herzenberg, L. A., J. Exp. Med. 153, 310, 1981. 7. Evans, R. L., Wall, P. W., Platsoucas, C. D., Siegel, F. P., Fikrig, S. M., Testa, C. M., and Good, R. A., Proc. Natl. Acad. Sci. USA 78, 544, 1981. 8. Thomas, Y., Sosman, J., Irigoyen, D., Friedman, S. M., Kung, P. C., Goldstein, G., and Chess, L., J. Immunol. 125, 2402, 1980. 9. Morimoto, C., Reinherz, E. L., Borel, Y., Mantzouranis, E., Steinberg, A. D., and Schlossman, S. F., J. Clin. Invest. 67, 753, 1981. 10. Morimoto, C., D&so, J. A., Borel, Y., Schlossman, S. F., and Reinherz, E. L., J. Immunol. 128, 1645, 1982. 11. Gatenby, P. A., Kotzin, B. L., Kansas, G. S., and Engelman, E. G., J. Exp. Med. 156, 55, 1982. 12. Reinherr, E. L., Morimoto, C., Penta, A. C., and Schlossman, S. F., J. Immunol. 126, 67, 1981. 13. Morimoto, C., Reinhera, E. L., Bore& Y., and Schlossman, S. F., J. Immunol. 130, 157, 1983. 14. Morimoto, C., Reinherz, E. L., and Schlossman, S. F., J. Immunol. 127, 69, 1981. 15. Morimoto, C., Reinherz, E. L., Todd, R. F., Distaso, J. A., and Schlossman, S. F., J. Immunol. 131, 1209, 1983. 16. Morimoto, C., Todd, R. F., Distaso, J. A., and Schlossman, S. F., J. Immunol. 127, 1137, 1981. 17. Reinherz, E. L., Kung, P. C., Goldstein, G., and Schlossman, S. F., J. Immunol. 123, 13 12, 1979. 18. Todd, R. F., Nadler, L. M., and Schlossman, S. F., J. Immunol. 126, 1435, 1981. 19. Stashenko, P., Nadler, L. M., Hardy, R., and Schlossman, S. F., J. Immunol. 125, 1678, 1980. 20. Reinherz, E. L., Morimoto, C., Fitzgerald, K. A., Hussey, R. E., Daley, J. F., and Schlossman, S. F., J. Immunol. 128, 463, 1982. 21. Cantor, H., Hugenberger, J., McVay-Boudreau, L., Eardley, D. D., Kemp, J., Shen, F. W., and Gershon, R. K. J. Exp. Med. 148,871, 1978. 22. Germain, R. N., and Benacerraf, B., J. Immunol. 121, 608, 1978. 23. Tada, T., Taniguchi, M., and David, C. S., .I. Exp. Med. 144, 713, 1976. 24. Gatenby, P. A., Kansas, G. S., Xian, C. Y., Evans, R. L., and Engleman, E. G., J. Immunol. 129, 1997, 1982.