Activation of immunoglobulin allotype-specific T cells by B cells

CELLULAR

IMMUNOLOGY

96,71-82 (1985)

Activation of lmmunoglobulin

Allotype-Specific

T Cells by B Cells

KATSUYUKIYUIANDAKIHIKOYANO Laboratory

oflmmunology, Department OfInfectious Diseases and Parasitology, School ofMedicine, 3-l-l Asahi, Matsumoto 390, Japan

Shinshu University

Received June 11, 1985; accepted July 8, 1985

To investigate the role of Ia and immunoglobulin (Ig) molecules of B cells in alloantigen-specific and nominal antigen-specific T-cell activations, the ability of B cells to stimulate Ig allotypespecificT cells was examined. T 15-primed B 1O.BRT cells respondedto MOPC 3 15 (IgA myeloma protein derived from BALB/c) as well as T15 but not to MOPC3lc (IgG, myeloma protein). These T cells were stimulated by papain-digested Fc fragment of T15. Thus, TIS-primed BlO.BR T cells were shown to be specific for Ig allotype of T 15, that is, Igh-2a. T 15-specificB I O.BR T cells were selectedby IO-day cultures with T15 in vitro. They responded to BALB.K spleen cells without addition of soluble T I5 antigen to the assayculture. Stimulator cells in this mixed lymphocyte reaction (MLR)-like responsebetween T 15-specificB lO.BR T cells and BALB.K spleen cells were Thy-l-, Iaf cells and these responseswere blocked by anti-Iak antibodies. Furthermore, Sephadex G-lo-passed BALB.K B cells stimulated the proliferation of Tl S-specific BlO.BR T cells, while they failed to stimulate allogeneic BALB/c spleen cells. The stimulating ability of B cells in this MLR-like responseof T15-specificBIO.BR T cellswas shown to be genetically restricted, namely, both H-2 and non-H-2 genesare involved in the manifestation of the stimulating ability. This system will provide a useful model for studying the role of B-cell surface Ig and Ia molecules in the activation of antigen-specific T cells and alloreactive T cells. o 1985 Academic press, IN.

INTRODUCTION Ia+ cells such as macrophages, dendritic cells, Langerhans cells and a variety of other cell types are capable of presenting foreign antigens to antigen-specific Ia-restricted T cells ( l-5). These cells that bear Ia molecules and that activate T cells in the presence of antigen are known as antigen-presenting cells (APC). Recently, it has been demonstrated that B cells are also capable of directly presenting antigen to T cells for antigen-specific, major histocompatibility complex (MHC)-restricted T-cell responses. Many B-cell tumors (6,7) and lipopolysaccharide-stimulated B lymphoblasts (8) were shown to possessAPC activity, while Ashwell et al. reported that even small resting B cells act as APC to antigen-specific MHC-restricted T cells (9). Furthermore, the role of B-cell surface Ig in T-cell activation has been demonstrated. Chesnut and Grey showed that macrophage-depleted murine spleen cells could stimulate rabbit IgGprimed T cells to proliferate in vitro in the presence of rabbit anti-mouse IgG but not rabbit IgG lacking anti-mouse Ig activity (10). They also demonstrated the relative efficiency of antigen presentation by B cells when the antigen was bound to membrane Ig (acting as antibodies) compared to antigen taken up nonspecifically (11). It has been reported that hapten-specific B lymphocytes are highly efficient at presenting 71 0008-8749185$3.00 Copyright 0 1985 by Academic Press, Inc. All rights of reproduction in any form reserved.

72

YUI AND YANO

hapten-modified antigen to T cells, and that hapten-binding surface Ig molecules are critically involved in this effective form of T-B interaction (12). It is well known that Ig molecules themselves also function as antigens. T Cells that recognize Ig isotype ( 13-16), allotype ( 17- 19), and idiotype (20-23) have been described. Several investigators suggestedthat the mechanisms of immune response to Ig molecules are the sameas those of conventional antigens. Liebermann et al. showed that the capacity to produce antibodies to Ig allotype is under the control of immune response(Ir) geneslinked to MHC (24). Schwartz and Paul (25) established that MHClinked Ir genescontrol T-cell responsivenessto T 15 myeloma protein. Furthermore, evidence showing T-cell recognition of the Igh- lb product in association with selfMHC-encoded determinants came from recent experiments with Ig allotype-specific cytotoxic T cells (Tc) ( 18, 19) or T-cell proliferative responses(26). With regard to alloreactive T-cell activation, the ability of Ia antigen on B cells in allogeneic MLR has also been examined. The relative lack of stimulation by B lymphocytes was first noted by Talmage and Hemmingsen (27) and Greineder and Rosenthal (28). A recent study using B-cell tumor lines revealed that B-cell tumor lines stimulated alloreactive T cells only in the presence of interleukin 1 (IL-l) (7). Thus, the mechanism of the functional difference of Ia antigens on B cells and other Ia-positive cells such as macrophages, dendritic cells, etc., in activation of alloreactive and antigen-reactive T lymphocytes in an MHC-restricted fashion remains to be clarified. In this study, we examined the ability of resting B cells to activate Ig allotypespecific T cells. The results showed that Tl5 allotype-specific BlO.BR T cells and T-cell lines were directly stimulated by B cells derived from BALB.K mice which have the same H-2k as responder T cells and the same Ig allotype as the myeloma protein, T 15, used as an antigen. MATERIALS

AND METHODS

Animals. Strains BlO.BR/SgSn Sic (BlO.BR) and BALB/c Cr Sic (BALB/c) were purchased from Shizuoka Laboratory Animal Center (Hamamatsu, Japan). A.TH, A.TL, and BALB.K mice were the progeny of breeding pairs kindly provided by Dr. Toshiyuki Hamaoka (Institute for Cancer Research,Osaka University Medical School, Osaka, Japan). Mice of both sexesbetween 8 and 24 weeks of age were used. Myeloma cells and purification procedures for myeloma proteins. The plasmacytoma TEPC 15 was kindly provided by Dr. Toshiyuki Hamaoka, MOPC3 1c by Dr. Masai-u Taniguchi (Laboratories for Immunology, School of Medicine, Chiba University, Chiba, Japan), and ascites of MOPC3 15 myeloma by Dr. Kenji Okuda (Department of Microbiology, Yokohama City University, Yokohama, Japan). The IgA myeloma protein, Tl5, was obtained from ascitic tumor and purified by affinity chromatography on phosphorylcholine (PC) columns according to the procedure of Chesebro and Metzger (29) with slight modifications. Briefly, the asciteswas applied to a loo-ml PC column. After the column was washed with 1 liter of phosphate-buffered saline (PBS, 0.005 A$ pH 7.8), the specifically bound protein was eluted with 200 ml of lop3 A4 phosphorylcholine chloride (Sigma Chemical Co., St. Louis, MO.). The eluted protein was concentrated and dialyzed exhaustively against PBS. After dialysis, the material was centrifuged (10,000 rpm, 60 min) and sterilized by passing it through Millipore membrane (pore size: 0.22 pm) and was stored at 4°C until use. Analysis by SDSpolyacrylamide gel electrophoresis, high-performance liquid chromatography (Phar-

T-CELL

ACTIVATION

BY B CELLS

73

macia Fine Chemicals), and the Ouchterlony method using goat anti-mouse Ig reagents revealed no contaminant protein (data not shown). Papain digestion of T 15 was done according to the procedure of Fahey and Askonas (30) with slight modification. Briefly, 20 mg of purified T15 was incubated with crystalline papain and cysteine at 37°C for 1 hr and dialyzed against phosphate buffer (PB, 0.005 M, pH 7.8) overnight. The digested fragments of T15 were separated by DEAE-cellulose chromatography using a gradient elution from 0.005 to 0.5 M PB. Two main peaks were observed by measuring the optical density at 280 nm and each fraction was pooled and concentrated. Ouchterlony double diffusion technique using goat anti-mouse Ig a-chain antibody (Cat. No. B-106, Meloy laboratories, Inc., Va.) and rabbit anti-mouse Fab fragment antibody (a kind gift of Dr. Masai-u Taniguchi) revealed that each fraction was Fc and Fab. Other antigens. Ovalbumin (OVA) was purchased from Sigma Chemical Co. Conjugation of PC with human y-globulin (HGG: Vega Biochemicals, Ariz.) was done by reacting 60 mg of HGG with 22 cul’( of p-diazonium choline in 6 ml of borate buffer (0.07 M sodium borate buffer with 0.08 A4 NaCl, pH 9.0). These conjugates contained 3.1 mol PC/mol protein and are designated PCj.,-HGG. Antibodies. Anti-Thy 1.2 monoclonal antibody (MAb) (Cat. No. NEI-005) was purchased from New England Nuclear (Boston, Mass.). Conventional A.TH anti-A.TL antibody was prepared from A.TH mice immunized weekly by intraperitoneal injections of 1 X 10’ live A.TL spleen cells. The monoclonal antibodies, TlH12 (anti-IAk, IgGZb)and TFk4 (anti-I-Ek, IgGi), were established by the hybridization of P3U 1 and X-63-Ag8-6.5.3. cells, respectively, with spleen cells from A.TH mice immunized with spleen cells from A.TL mice. The hybridization procedure for anti-Ia antibody production has been described elsewhere (31-33). The monoclonal antibodies were obtained from ascites of (BALB/cxA.TH)F, mice which were injected with hybrid cells intraperitoneally. P3Ul cell line and X63-Ag8-6.5.3 cell line were the kind gifts of Dr. Masaru Taniguchi. Immunization. The procedure for immunization of mice was described previously (3, 34). In brief, the antigen was emulsified in complete Freund’s adjuvant containing 1 mg/ml of killed Mycobacterium tuberculosis (H37Ra) organisms (Difco Laboratories, Mich.). Mice were immunized in the hind footpads and at the base of the tail with 10 pg of OVA and 100 pg of myeloma protein in a total volume of 0.1 ml of emulsion. Preparation of peritoneal exudate T-lymphocyte-enriched cells (PETLES) and lymph node lymphocytes (LNL). The original and slightly modified procedures for preparing

PETLES and LNL were described in detail elsewhere (3, 34, 35). Briefly, 2-3 weeks after immunization, thioglycollate-induced peritoneal exudate cells were harvested and passed over nylon wool columns. LNL were prepared from popliteal, inguinal, and paraaortal lymph nodes of immunized mice by passingthem through nylon wool column. Preparation of spleen cells, B cells, and Langerhans cells. Spleen cells as the MLR stimulator cells were prepared as already described (32). SephadexG- 1O-passedB cells were prepared from spleen cells of unprimed mice. Spleen cells were treated with antiThy-l plus complement (C). After the treatment, the cells were applied to Sephadex G- 10 column and the first 0.2% of the cells eluted were collected. These cells contained 89% surface Ig-positive cells detected by immunofluorescent staining. Stimulator cells were irradiated (1500 rad) with MBR-1505R (Hitachi Medico Co., Tokyo, Japan). Langerhans cells were prepared as described previously (36). Briefly, the skin was

74

YUI AND YANO

pulled off the entire tail and treated with 0.5% trypsin solution at 37°C for 1 hr. The loosened dermis was removed and covered with 0.025% DNAase, and epidermal cells were harvested. Treatmentof spleencellswith antibodiesand C. Stimulator spleen cells were treated with anti-Thy 1.2 MAb, and TFk4 (anti-I-Ek MAb), at 4°C for 20 min, washed, and then treated with 1 ml of rabbit C (diluted 1:15) at 37°C for 30 min. After treatment, the cells were washed twice with Hanks’ medium and resuspendedin Peck-Click medium supplemented with heat inactivated 10%fetal calf serum (FCS; GIBCO, Grand Island, N.Y.). In vitro primary cell cultures.LNL (2-4 X 105)were cultured with varying dosesof myeloma proteins or 1 X 104- 1 X 1O6stimulator cells in 200 ~1of Peck-Click medium supplemented with 10% FCS per well. Cultures were maintained at 37°C in a humidified atmosphere of 2% CO2 and 98% air for 5 days. Sixteen to twenty-four hours before harvesting 0.5 &i of [3H]thymidine ([3H]TdR: sp act 27 Ci/mmol, Radiochemical Center, Amersham, England) was added. [‘H]TdR incorporation was measured in a liquid scintillation counter (Packard Co., Downers Grove, Ind.). All determinations were done in triplicate or duplicate and the data were expressedeither as mean counts per minute f standard error of the mean or as the difference between antigen-stimulated and control responses(Acpm) as already described (3, 32-34). In vitro secondarycell cultures and TlS-specific T-cell lines. The procedure for in vitro secondary cell cultures has been described previously (32). Briefly, 200 ~1 of a 10% FCS supplemented Peck-Click medium containing antigen ( 100 pg/ml OVA or 250 pg/ml T15) and l-2 X lo5 PETLES were placed in each well of a sterile, U-bottom microculture plate (l-63302, Nunclon; Denmark) and incubated at 37°C in a humidified atmosphere of 2% COZ, 98% air (primary culture). Ten days later, responding cells, at a density of 1 X lo4 cells in 200 ~1of fresh medium, were restimulated with the antigen in the presence of 1 X lo5 irradiated spleen cells (secondary culture). After a 2.5-day period in culture, proliferative responseswere assessedby a 16- to 24-hr exposure to 0.5 &I of [3H]TdR. The T15-specific B lO.BR T-cell lines (BRT-1 and -2) were derived from PETLES of B lO.BR mice immunized with T 15. The T-cell lines were maintained as long-term continuous cultures by restimulation at IO-day intervals with irradiated BlO.BR spleen cells and antigen T 15 ( 100 pg/ml). Viable T 15-specific T cells were recovered 10 days after the last T 15 stimulation and used to assaythe T 15- and BALB.K-specific proliferative responses. RESULTS

Specificity of Proliferative Responsesof TlS-Primed BlO.BR LNL BlO.BR strain has been shown to be a high responder to IgA allotype or Igh-2a in antibody formation and T-cell proliferative responses(24, 25). Thus, we immunized BlO.BR mice with T15 and also OVA as a control antigen. One week after immunization, proliferative responsesof nylon wool column-passed LNL to increasing concentrations of various myeloma proteins were analyzed by 13H]TdR incorporation. As shown in Fig. la, [3H]TdR incorporation to new DNA of T15-primed BlO.BR LNL increased depending on the concentration of T 15. These T 15-primed B IO.BR LNL were highly responsiveto another IgA myeloma protein of Igh-2a allotype (MOPC 3 15, anti-DNP myeloma). By contrast, no responsivenesstoward MOPC3lc, IgGi

T-CELL ACTIVATION

75

BY B CELLS b cp=

(b)

(a) 15,ooc

10,000

5,ooc

I

10

30

100

Antigen

300 dose

1,000 (pghl)

3

10

30

100

Antigen dose

300 (ug/mll

FIG. 1. T I Sprimed B IO.BR T ceils are specific for allotypic determinant(s) on T 15. 2 X 10’ T 1S-primed B lO.BR LNL were cultured with various concentrations of T 15 (I& l ), MOPC 3 I5 (I& n ), MOPC 3 Ic (I&, q !),and PC-HGC (0) in (a), and Fc fragment of T15 (A) and Fab fragment of T15 (A) in (b).

myeloma protein of BALB/c origin, was observed. The responseswere not directed to PC which was used for purification of T15 myeloma protein, becausethe BlO.BR LNL did not respond to PC-HGG at all. Fine analysis of the antigen specificity of T 1j-primed B 1O.BR LNL was done by using papain-digested T 15 fragments, that is, Fab and Fc. As shown in Fig. lb, proliferative responses were observed when the T 15-primed LNL were stimulated by the Fc fragment of T 15. The T15-specific BlO.BR LNL responseswere mediated by Thy-l+, Lyt- 1+2-, Iacells with the stimulation of Ia+ syngeneic APC in the presence of T15 antigen (data not shown). Tl5-Specific BIO.BR T Cells Respond to BALB.K Spleen Cells To purify T15-specific BlO.BR T cells, T15-primed BlO.BR LNL were stimulated and selected in vitro by lo-days of culture with T15. As shown in Table 1, after 10 days of culture with the antigen, T 15-specific B lO.BR T cells responded to T 15 in the presenceof syngeneic APC (Acpm 38,107), and strongly responded to BALB.K spleen cells even in the absence of soluble T15 antigen (Acpm 43,238). As a control, OVAspecific BlO.BR T cells were prepared in a manner similar to T15 selection. OVAspecific BlO.BR T cells did not respond to BALB.K spleen cells alone (Acpm 732), but were stimulated with OVA presented by BALB.K spleen cells (Acpm 30,176). Thus, T15-specific BlO.BR T cells, primed in vivo and selected in vitro by 10 days of culture with the stimulation of T15, responded to BALB.K spleen cells without the antigen, T15, in secondary responses.In addition, the maximum responsesof T15specific B lO.BR T cells against either T 15 plus syngeneic APC or BALB.K spleen cells

76

WI

AND

YANO

TABLE Proliferative

Response of T15- or OVA-Specific

1 BlO.BR T Cells to BALB.K Spleen Cells Responseof OVA-specific T cells

Responseof T I5-specific T cells

Stimulator cells None BIO.BR BALB.K Acpm for MLR

Medium (cpm)

T15 (cpm)

Acpm for T15

Medium (cpm)

OVA (wm)

Acpm for OVA

1,223 + 226 1,794 f 575 45,032 f 59 I

1,286 Z!I 241 39,901 f 3,938 56,087 3~1,391

63 38,107 11,055

2800 f 100 1224k 30 1956 IT 175

1,655 f 365 23,563 + 1,524 32,132 + 1,080


43,238

732

Note. BlO.BR T cells (1.5 x 104)positively selectedin vitro with T15 or OVA were added to culture containing 2 X IO5 irradiated (1500 tad) nonimmune BlO.BR or BALB.K spleen cells with or without antigen (1000 &ml T15 or 100 @fml OVA). The data are expressedas mean counts per minute + SD of duplicate cultures or the difference of the mean (Acpm).

were observed on Day 3 as a secondary T-cell response type but not as a primary MLR response against non-H-2 gene products of BALB.K spleen cells. The addition of T 15 antigen together with BALB.K spleen cells in the culture of T 15-specificB 1O.BR T cells augmented the response slightly. Role of la Molecules on BALB.K Stimulator Cells in Activation of Tl54’pecific BlO.BR T Cells The analysis of stimulator cell population in the MLR response of T 1Sspecific BlO.BR T cells against BALB.K spleen cells showed that BALB.K stimulator cells bore Ia molecules but not Thy-l molecules. As shown in Table 2, the stimulating ability of BALB.K spleen cells was eliminated by the pretreatment of stimulator cells with anti-Ia MAb + C, but not with anti-Thy-l MAb + C. The participation of Ia molecules in the responseswas investigated by adding anti-Ia antibodies to such cultures in the absence of C. Table 3 presents the data demonstrating that MLR responsesof TlS-specific BlO.BR T cells to BALB.K Ia+ cells were completely blocked by the A.TH anti-A.TL antibody at 1 X 10e2dilution. Furthermore, inhibition of the MLR responseswas obtained with either anti-I-Ak or anti-I-Ek MAb as shown in Table 4. Proliferative responses of T15-specific BlO.BR T cells to BALB.K spleen cells were TABLE 2 MLR Response of Tl S-Selected BlO.BR T Cells Directed to Iaf BALB.K Spleen Cells Stimulator cells of MLR Treatment of stimulator cells

BlO.BR (wm)

C alone Anti-I-Ek (TFk4) + C Anti-Thy 1 + C

5795 + 936 1189 + 318 5180 + 312

BALB.K (cm) 15,831 + 2553 2,539 + 330 13,592 + 1138

Acpm 10,063 1,350 8,412

Note. BlO.BR or BALB.K stimulator cells were treated with TFk4 or anti-Thy 1 + C. After the treatment with monoclonal antibodies, live cells were counted and the cell number was adjusted. T 15-selected B lO.BR T ceils (2 X IO’) were cultured with 2 X 1OSirradiated ( 1500 rad) stimulator cells.

T-CELL ACTIVATION

77

BY B CELLS

TABLE 3 Blocking Effect of A.TH Anti-A.TL Antibody on MLR Responseof TlS-Specific B IO.BR T Cells against BALB.K Spleen Cells MLR responseof TlS-specific BIO.BR T cells” Dose of antibody @)

BlO.BR (cpm)

0 3 1 0.3 0.1 0.03 0.01

2601 f 308 1661 rt: 230 758 f 95 1798 + 175 1551 f 204 2745 f 853 2764 f 813

BALB.K (wm)

(Awn)

14,51I + 2487 719 + 145 1,250 k 352 5,404 f 1110 10,024 + 908 10,724 + 73 13,041 ?I 954

%blockingb

11,910 10 492 3,606 8,473 7,969 10,277

100 96 70 29 33 14

’ 2 x lo4 T15 specific BIO.BR T cells were added to cultures containing 2 X IO5irradiated BlO.BR or BALB.K spleen cells and various concentrations of A.TH anti-A.TL antibody. b Percentage blocking was calculated according to the formula. { 1 - [experimental response (Acpm)/ control response(Acpm)]} X 100.

inhibited by the MAb specific for I-Ak (T 1H 12) and also by the MAb specific for I-Ek (TFk4). Fifty seven and 45% inhibitions of the control response were obtained by TlH12 and TFk4, respectively. Thus, Ia molecules of stimulator cells (I-Ak and I-Ek molecules of BALB.K spleen cells) function to elicit MLR responsesof T 15-specific BlO.BR T cells. Sephadex G-lo-Passed BALB.K B Cells Stimulate Tl5-Specijc BlO.BR T Cells Having established that BALB.K Ia+ cells stimulate TlS-specific BlO.BR T cells, we undertook a study to determine whether the B-cell fraction of BALB.K spleen cells was capable of stimulating T 1Sspecific B lO.BR T cells. In these experiments, titrated TABLE 4 Effect of Monoclonal Anti-Ia Antibodies on MLR Responseof T I5-Specific BlO.BR T Cells against BALB.K Spleen Cells Responseof TlS-specific BlO.BR T cells Stimulatora cells

mAbb

wm

Acpm

%inhibition

BlO.BR BALB.K

-

2950 f 333 9443 ? 982

6492

-

BlO.BR BALB.K

T 1H 12 (anti-l-A’)

2031 + 153 4854 + 661

2823

57

BlO.BR BALB.K

TFk4 (anti-I-Ek)

1421 f 337 4984 Z!Y387

3563

45

n After TI 5 selection in vitro 1.5 X IO4B IO.BR T cells were added to culture containing 2 X lo5 irradiated (1500 rad) nonimmune BlO.BR or BALB.K spleen cells. b Each monoclonal antibody was added in culture at a final concentration of 1%.

78

YUI AND YANO

numbers of whole spleen cells or Sephadex G- 1O-passedB cells were compared for their abilities to stimulate the MLR responses.As shown in Fig. 2, Sephadex G-lOpassedBALB.K B cells were much more efficient than whole spleen cells in stimulating T 15specific BlO.BR T cells. However, the same Sephadex G-IO-passed BALB.K B-cell fraction was found to be a poor stimulator in BALB/c anti-BALB.K MLR responses,while unfractioned whole BALB.K spleen cells elicited anti-BALB.K MLR responses. In order to analyze whether Ia+ cells which do not bear Ig stimulate T 15-specific B 1O.BR T cells, we prepared Langerhans cells as stimulator cell populations. As shown in Table 5, BALB.K Langerhans cells were not stimulatory to T 15-specific B lO.BR T cells. Thus, the B-cell fraction of BALB.K spleen cells stimulate T 15-specific B lO.BR T cells in the absenceof an exogenous source of T15 antigen. Involvement of H-2 and Non-H-2 Genesfor Eliciting A4LR Responsesof TlXi’peciJic BIO.BR T Cell Lines against BALB.K Ia’ Cells Previous studies in the guinea pig and mice have demonstrated the requirement for genetic identity at the MHC for APC to present antigen effectively to primed T lymphocytes. To test the genetic restriction between TlS-specific BlO.BR T cells and stimulator Ia+ cells, H-2 congenic mice of BALB background were utilized. T15specific BlO.BR T cell lines, BRT-1 and BRT-2, were established as described under Materials and Methods. These cell lines were cultured with BlO.BR (H-29, BALB.K (H-27 or BALB/c (H-29 spleen cells as shown in Table 6. The sameantigen specificity of BRT- 1 and -2 cell lines was observed, that is, they responded to both T 15 in the presenceof syngeneic B lO.BR APC and BALB.K spleen cells without exogenous T 15. BALB/c spleen cells (bearing the same non-H-2 genesas BALB.K), however, did not A.

T15-Specific

Bl0.P.R

LNL

8.

CPl

MLa/c

spleen

CdlS

20,oot

.

--

I

,

1

I

1

3

10

30

(x10-3

100

cells/rail) Stimulator

1

I

I

I

3

10

30

(x10-3 cell

dose

100

call~/well) stimulator

cell

doao

2. Stimulating ability of Sephadex G-IO-passedBALB.K B cells for TlS-specific BIO.BR T cells and BALB/c spleen cells. TI5-specific BlO.BR LNL and BALB/c spleen cells were cultured with various concentrations of Sephadex G-lo-passed BALB.K B cells (O), whole BALB.K spleen cells (O), and syngeneic cells (A). FIG.

T-CELL ACTIVATION

79

BY B CELLS

TABLE 5 Stimulating Ability of BALB.K Spleen Cells or Langerhans Cells against Tl S-Specific BIO.BR T Cells Source of stimulator cell population Langerhans cells

Spleen cells Strain of stimulator cells BIO.BR BALB.K

(w-4 8,345 f 1635 16,765 + 3005

(Acpm)

(cpm)

@m-d

8,420

5600 f 2020 6425 + 1015

825

Note. 2 X lo5 TIS-primed BlO.BR PETLES were cultured with I X lo5 irradiated spleen cells or 2 X IO4 irradiated Langerhans cells for 5 days in microtiter plates. Acpm for T 15-specificproliferation (50 &ml of T15 in the culture) in the presence of BIO.BR spleen cells or BlO.BR Langerhans cells were 11,905 or 10,960, respectively.

stimulate T 15specific B lO.BR T-cell lines. Theseresults show that H-2 identity between T 15-specificB lO.BR T cells and stimulator Ia+ spleencells is required for manifestation of the MLR. It is obvious that non-H-2 genesare also involved in the MLR, because TlS-specific T-cell lines themselves derived from BlO.BR strain (H-27. Thus the MLR-stimulating ability of BALBK spleen cells to T15-specific BlO.BR T-cell lines was shown to be under the control of at least two genes, one located within the H-2 complex and the other located in non-H-2, probably Igh allotype genes. DISCUSSION It is known from recent reports that B cells can function as APC to MHC-restricted antigen-specific T cells (7-10, 37). On the other hand, the poor stimulating ability of purified B-cell subpopulations or B-cell tumor lines has been described. Ahmann et al. (38) reported that purified B cells (Sephadex G- 10 passedand treated with rabbit anti-mouse brain serum and C) were poor stimulators of an MLR. In this report, Sephadex G-lo-passed BALBK B cells were also poor stimulators to alloreactive BALB/c T cells (Fig. 2). Halper et al. (39) also described that leukemic B cells, despite their strong expression of Ia antigens, were extremely poor stimulating cells in the MLR. Furthermore, Minami et al. (40) showed that macrophage-enriched cells were capable of inducing an MLR against K, D, I-A, and I-E differences, whereas B cells induced little or no stimulation. We also observedthat Epstein-Barr virus-transformed human B-cell lines can elicit allogeneic MLR responsesonly with the existence of responder accessorycells (Yamamoto and Yano, J. Zmmunol., in press).Furthermore, Glimcher et al. (7) presented the data demonstrating that a seriesof Ia antigen-positive BALB/c B-cell lymphomas can stimulate an allogeneic MLR only in the presence of an exogeneoussource of IL- 1. One of the possibleinterpretations of the poor stimulating ability of B cells in allogeneic MLR is that B cells secrete small amounts of IL-l in comparison with macrophages. In contrast, Gottlieb et al. (41) reported that, in the human system, surface IgM-bearing B cells from both peripheral blood and tonsil were excellent stimulator cells in unidirectioned MLR responses.They purified IgMbearing B cells by the rosetting method by anti-IgM-coated bovine red blood cells. It

80

YLJI AND

YANO

may be possible to assume that the purified B cells were activated by the purification procedures and that such activated B cells stimulated MLR responding cells. The Sephadex G-lo-purified B cells were shown to be good stimulators to Tl5specific BlO.BR T cells but not to BALB/c alloreactive spleen cells. Both alloreactive BALB/c T cells and T 15-specificB lO.BR T cells may recognize Iak antigens of BALBK mice. Why did Iak-positive BALB.K B cells not stimulate BALB/c T cells, while they stimulated TIS-specific BlO.BR T cells? One might postulate following three explanations. (i) Ia antigens expressedby accessorycells (macrophages, dendritic cells, etc.) and B cells are intrinsically different. Cullen et al. (42) reported that the peptides obtained from accessorycell and B cell Ia molecules are identical but the (Ychains of B-cell Ia molecules are more extensively sialylated than those of accessorycells. Alloreactive BALB/c T cells may recognize Ia epitope of accessorycells but not such sialylated Ia molecules of B cells, while T 15-specific B lO.BR T cells recognize Ia epitope on B cells together with Ig allotype antigen. (ii) An additional signal (like IL-l) is required to activate alloreactive T cells in addition to the recognition of alloantigen. And accessorycells deliver this signal, but B cells are unable to do so. T15-specific BlO.BR T cells, however, do not require this signal to proliferate. Glimcher et al. (7) reported an interesting difference between the antigen-specific and allospecific responses.The addition of an exogeneous source of IL-l was necessary to obtain an allogeneic MLR, but antigen-specific proliferative responseswere elicited without it. (iii) Recently, it has been revealed that various Ia negative cells, such as endothelial cells (43), leukemic cells (44), and L cells (45), function as accessorycells in Con A responses.An undefined molecule (other than Ia antigens) existing on accessorycells but not on B cells may be involved in the activation of Con A-reactive and -alloreactive T cells. Such a molecule is, however, not necessaryto activate T 15-specific B 1O.BR T cells. The studies describedhere show that T 15-specificBlO.BR T cells respond to BALB.K B cells without the addition of soluble T 15 antigen, In contrast, Ig- Ia+ cell fraction of BALBK mice, for example Langerhans cells, did not stimulate the proliferative responseof T 15-specific B 1O.BR T cells. Since T15-specific B lO.BR T cells recognize allotypic determinants on IgA myeloma protein, T 15, BALB.K B cells which stimulate T15-specific B lO.BR T cells may be h&bearing B cells. Preliminary results in our hands indicate that the treatment of BALB.K spleen cells with anti-murine Ig LYchain + C (but not with anti-murine Ig p chain + C) abrogated the stimulating ability to T15-specific BlO.BR T cells. These results strongly suggestthat T15-specific BlO.BR T cells recognize Ig allotypic determinants of IgA on BALB.K B cells. Snodgrassand co-workers ( 19,20) demonstrated Igh- 1b-specificTc which were specifically lytic for target myeloma cells expressing Igh- 1ballotype of IgG2,. They showed that the lytic activity of the Tc was restricted to syngeneic H-2d target cells in primary cultures. Cloned Tc also displayed a preference for Igh- 1b target of H-2“ haplotype. Bikoff and Bona (26) characterized the T-cell proliferative responseto IgG2, of the b allotype. They showed that allotypic determinants on the molecule were recognized in association with self-class II molecules. As shown in Tables 3, 4, and 6, MLR responsesof T15-specific BlO.BR T cells against BALB.K B cells were also mediated by stimulator Ia molecules. Furthermore, T-cell recognition of other cell surface mol-

T-CELL ACTIVATION

81

BY B CELLS

TABLE 6 Requirement of N-2 Identity between Responder T Cells and Stimulator Cells for Activation of T I5-Specific B lO.BR T-Cell Lines Proliferative responseof T I5specific BlO.BR T-cell line

Stimulator cells T-cell line BRT- I

BRT-2

Strain

I-region

Igh-2

BlO.BR BALB.K BALB/c

( Acpm)

(cpm)

2,008 f 239 18,260 + 2318 3,192 f 1099

BlO.BR BALB.K A.TL BALB/c

1,149 f

84

7,678 +

266

16,252 1,184 6,529

1,940 ?z 31

791

2,007 f

858

627

Note. TIS-specific BlO.BR T-cell lines BRT-I and -2 (1.5 X 104)were cultured with 2 X lo5 irradiated (1500 rad) BlO.BR, BALB.K, A.TL, or BALB/c spleen cells.

ecules, such as class I alloantigens (46,47) or allogeneic Mls determinants (48) in the MLR, has also been shown to occur in association with classII molecules on stimulator cells. Similarly, the response of T15-specific BlO.BR T cells against BALB.K B cells was stimulator class II restricted. As already discussed,B cells present antigen to T cells in spite of a poor stimulating ability of alloreactive T cells in MLR. In this study, BALB.K B cells directly stimulated T15-specific B lO.BR T cells like MLR responses.Although the real mechanisms of T 15-specific B lO.BR T-cell activation by BALB.K B cells still remain to be answered, one of the possible interpretations is that BALB.K B cells, to be precise Ig a-chainpositive B cells, present their self-component of IgA with their own Ia molecules to T 15-specific B lO.BR T cells. The proliferative responsesof Ig allotype-specific T cells against Ig on B-cell surface may provide a useful model for studying the role of B-cell surface Ig and Ia molecules in the activation of antigen-specific T cells and alloreactive T cells. ACKNOWLEDGMENT The authors thank Ms. Rieko Okae for her excellent secretarial help.

REFERENCES 1. Rosenthal, A. S., and Schevach, E. M., J. Exp. Med. 138, 1194, 1973. 2. Steinman, R. M., and Witmer, M. D., Proc. Natl. Acad. Sci. USA 75, 1978. 3. Yano, A., Schwartz, R. H., and Paul, W. E., J. Exp. Med. 146,828, 1977. 4. Stingl, G., Katz, S. I., Clement, L., Green, I., and Shevach, E. M., J. Immunol. 121, 2005, 1978. 5. Schwartz, R. H., Yano, A., and Paul, W. E., Immunol. Rev. 40, 153, 1978. 6. McKean, D. J., Infante, A. J., Nilson, A., Kimoto, M., Fathman, C. G., Walker, E., and Warner, N., J. Exp. Med. 154, 1419, 1981.

7. Glimcher, L. H., Kim, K.-J., Green, I., and Paul, W. E., J. Exp. Med. 155,445, 1982. 8. Chesnut, R. W., Colon, S. M., and Grey, H. M., J. Immunol. 128, 1764, 1982. 9. Ashwell, J. D., Defmnco, A. L., Paul, W. E., and Schwartz, R. H., J. Exp. Med. 159, 881, 1984. 10. Chesnut, R. W., andGrey, H. M., J. Immunol. 126, 1075, 1981. 11. Kakiuchi, T., Chesnut, R. W., and Grey, H. M., J. Immunol. 131, 109, 1983.

82

YUI AND YANO

12. Rock, K. L., Benacerraf, B., and Abbas, A. K., J. Exp. Med. 160, 1102, 1984. 13. Tada, T., Okumura, K., and Taniguchi, M., J. Zmmunol. 111, 952, 1973. 14. Chiorazzi, N., Fox, D. A., and Katz, D. H., J. Zmmunol. 118,48, 1977. 15. Suemura, M., Kishimoto, T., Hirai, Y., and Yamamura, Y., J. Zmmunol. 119, 149, 1977. 16. Rosenberg, Y. J., and Chiller, J. M., J. Exp. Med. 150, 5 17, 1979. 17. Herzenberg, L. A., Okumura, K., Cantor, H., Sato, V. L., Shen, F.-W., Boyse, E. A., and Herzenberg, L. A., J. Exp. Med. 144,330, 1976. 18. Snodgrass,R., Wilson, D. B., and Bosma, M. J., J. Exp. Med. 154,480, 1981. 19. Snodgrass,R., Bosma, M. J., and Wilson, D. B., J. Exp. Med. 154, 491, 1981. 20. Julius, M. H., Augustin, A. A., and Cozenza, H., Nature (London) 265,251, 1977. 21. Gleason, K., and Kdhler H., J. Exp. Med. 156, 539, 1982. 22. Jorgensen, T., Bogen, B., and Hannestad, K., J. Exp. Med. 158,2183, 1983. 23. McNamara, M., and Kohler H., J. Exp. Med. 159, 623, 1984. 24. Lieberman, R., Paul, W. E., Humphrey, W. Jr., and Stimpfling, J. H., J. Exp. Med. 136, 1231, 1972. 25. Schwartz, R. H., and Paul, W. E., J. Exp. Med. 143, 529, 1976. 26. Bikoff, E., and Bona, C. A., J. Zmmunol. 131, 103, 1983. 27. Talmage, D. W., and Hemmingsen, H., J. Allergy C/in. Zmmunol. 55,442, 1975. 28. Greineder, D., and Rosenthal, A. S., J. Zmmunol. 114, 1514, 1975. 29. Chesebro, B., and Metzger, H., Biochemistry. 11, 766, 1972. 30. Fahey, J. L., and Askonas, B. A., J. Exp. Med. 115,623, 1962. 3 1. Kohler, B., and Milstein, C., Nature (London) 256, 495, 1975. 32. Yoshizawa, K., and Yano, A., J. Zmmunol. 132,2820, 1984. 33. Aosai, F., Yui, K., Shigematsu, H., and Yano, A., Microbial. Zmmunol. 28, 1223, 1984. 34. Schwartz, R. H., Jackson, L., and Paul. W, E., J. Zmmunol. 115, 1330, 1975. 35. Corradin, G., Ethnger, H. M., and Chiller, J. M., J. Zmmunol. 119, 1048, 1977. 36. Steinmuller, D., and Wunderlich, J. R., Cell. Zmmunol. 24, 146, 1976. 37. Glimcher, L. H., Hamano, T., Asofsky, R., Heber-Katz, E., Hedrick, S., Schwartz, R. H., and Paul, W. E., Nature (London) 298, 283, 1982. 38. Ahmann, G. B., Nadler, P. I., Bimkrant, A., and Hodes, R. J., J. Zmmunol. 123, 903, 1979. 39. Halper, J. P., Fu, S. M., Gottlieb, A. B., Winchester, R. J., and Kunkel, H. G., J. Clin. Invest. 64, 1141, 1979. 40. Minami, M., Shreffler, D. C., and Cowing, C., J. Zmmunol. 124, 1314, 1980. 41. Gottlieb, A. B., Fu, S. M., Yu, T. Y. D., Wang, C. Y., Halper, J. P., and Kunkel, H. G., J. Zmmunol. 123, 1497, 1979. 42. Cullen, S. E., Kindle, C. S., Shreffler, D. C., and Cowing, C., J. Zmmunol. 127, 1478, 1981. 43. Roska, A. K., Johnson, A. R., and Lipsky, P. E., J. Zmmunol. 132, 136, 1984. 44. Wakasugi, H., Hare], A., Dokhelar, M. C., Fradelizi, D., and Tursz, T., J. Zmmunol. 132, 2939, 1984. 45. Bekotf, M., Kakiuchi, T., and Grey, H. M., J. Zmmunol. 134, 1337, 1985. 46. Minami, M., and Schreffler, D. C., J. Zmmunol. 126, 1774, 1981. 47. Rock, K. L., Barnes, M. C., Germain, R. N., and Benacerraf, B., J. Zmmunol. 130,457, 1983. 48. Cress, R. E., Wesley, M. N., and Hodes, R. J., J. Zmmunol. 127, 1763, 1981.