the Stimulatory Capacity of Β Cells to a
Major Histocompatibility Complex
Class-II-Restricted Proliferative
T-Cell Clone
Ellen G. van Lochern, Astrid Bakker, Simone Snijder,
Mijntje Aarts, Gijsbert C. de Gast, and Eis Goulmy
ABSTRACT: After BMT, donor Τ cells are acnvated which can display GvHD as well as GvL activities In order to study this GvL-specifk T-cell response in vitro, proliferative T-cell ebnes from post-BMT PBMCs were generated by Stimulation with a patient's leukemic cells One C D 4+ T-cell clone (designaced M-33) displayed
strong proliferative activity against the patient's leukemic cells but not against the patienc's EBV-LCLs The mduc-tion of proliferamduc-tion, however, appeared not to be leuke-mia specific Detailed analysis of the reactivity patterns revealed that T-cell clone M-33 recogmzes an as yet un-known nonpolymorphic determmant in the context of seif HLA-DRw52, presented by all but one type of APC T-cell clone M-33 prohferated upon Stimulation by
PB-MCs, freshly isolated Β cells, monocytes, dendntic cells, leukemic Β cells, and nonleukemic B-cell blasts, solely in vitro EBV-transformed Β cells and in vivo EBV-infected Β cells failed to induce proliferation of T-cell clone M-33 Neither surface expression of MHC or accessory molecules on the EBV cells nor suppression caused by the EBV-infected cells could explain their failure to stimulate T-cell clone M-33 We therefore hypothesize that the absence of the stimulatory capacity once ehe Β cells are virally infected could be the result of competmon for MHC class II binding of the Epstein—Barr viral peptides, thus affecting the postulaced DRw52-resmcted peptide for recognition by T-cell clone M-3 3 Human Immunology 42 137-144(1995) ABBRBVIATIONS APC BMT EBV EBV-LCL ER GvHD GvL mAb antigen-presenting cell bone marrow transplantation Epstein—Barr virus
EBV-transformed lymphoblastoid cell hne
erythrocyte rosette g'aft-versus-host disease
graft versus leukemia monoclonal antibody MESF mH MHC MLR NK PBMC PLT TNTR From the Cepartment ofImmunohematology andBloodBank (E GvL
AB,SS, ΜΛ, EG), Leiden Unwersity Hospital, Leiden, and the Department of Hematology (G C d C ) , Utrecht Untversity Hospital, Utrecht, Τ he Netherlands
Addnss repnnt requests to Dr L G van Lochern Department of lm-munohernatology and Blood Bank, Leiden Unwersity Hospital, Rijntbur-gerweg 10, 2333 AA Leiden, The Netherlands
Received(U) March 9, 1994, aeeeptedjuly 5 1994
molecules of equivalent soluble fluorochrome
rmnor histocompatibility major histocompatibility complex mixed leukocyte reaction natural killer
penpheral blood mononuclear cell pnmed lymphocyte test
INTRODUCTION
Graft-versus-host disease (GvHD) still forms a major barner to successful bone marrow transplantation (BMT) in both the HLA-identical as well as the HLA-matched Situation [1} Depletion of the T-cell population from the bone marrow inoculum leads to a reduction in the incidence and seventy of GvHD but IS associated with Hunvn Immunology 42, 137-144 (1995)
© American Society for Hiscocornpanbility and Immunogenencs, 1995
138 Ε G van Lochern et al
mcreased nsk of rejection and recurrence of the original disease Thus, mature Τ cells in the donor bone marrow inoculuin, which are vital for graft acceptance, cause GvHD and evoke graft-versus-leukemia (GvL) activity [2, 3] Α direct relationship between the GvHD effect and the GvL effect IS mainly supported by clinical data [4, 5], although Horowitz et al [2] and Buttunm et al {3] demonstrated a GvL effect mdependent of GvHD [2, 3}. Probably both supposmons are carrect as data from animal studies and a few in vitro studies in humans demonstrate the existence of effector cells exhibiting both as well as separate GvHD and GvL activities [6-9] In support of the latter notion, prehminary studies in our laboratory indeed showed the presence of three
function-aüy different types of cytotoxic Τ celi (CTL) clones
clones directed at host-specific minor histocompatibility (mH) antigens which were shared by the patient's pe-npheral blood mononuclear cells (PBMCs) and leukemic cells, clones recognizmg only host PBMCs but not host leukemic cells, and putative GvL clones directed at the patient's neoplastic cells only [10]
We aimed at both confirmation and extension of our first results and concentrated in this study on the gen-eration of GvL-directed prohferative T-cell responses Assummg that GvL-reactive donor Τ cells may exist in a patient's post-BMT circulation, we used the PBMCs of a patient (M) after HLA genotypically identical BMT as responding Τ cells and stimulated these cells in vitro with the patient's own leukemic cells Following specific restimulation and expansion, T-cell clones were ob-tained One of the latter clones (ι e , clone M-33), se-lected on the basis of lts high leukemia-specific prohf-erative activity, was subsequently analyzed in detail Comprehensive analyses of the reactivity patterns re-vealed that T-cell clone M-33 was not solely leukemia cell specific, on the contrary, lt seemed to recognize an as yet unknown nonpolymorphic determinant in the con-text of seif HLA-DRw52 which is presented by all but one type of antigen-presenting cell (APC) Here we re-port on the absence of the stimulatory capacity of Β cells following either in vivo or in vitro Epstein-Barr virus (EBV) infection.
MATERIALS AND METHODS
Patient. Α female patient Μ received a BMT from her
HLA-identical mixed leukocyte reaction (MLR)-negative sister for treatment of acute myeloid leukemia (AML-M5) The HLA type of the patient and the donor is HLA-A2,-B44,-B27,Cwl,-DR7)-DR12,-DRw521
-Dw25,-DQ2 The patient developed a mild form of acute GvHD after BMT.
Panel PBMCs of a panel of HLA-A, -B, -C, -DR, and
-DQ-typed blood donors were used as stirnulator cells in
primed lymphocyte tests (PLTs) Typing of these donors was done by serology Alloreactive prohferative T-cell clones specific for products of the ß3 locus of HLA-DR
were used to type the stimulator cells for Dw25 and Dw26 Two clones, designated as LBQ-1 and clone 12, specific for HLA-Dw25 and -Dw26, respectively, were kindly provided by Dr Α Termijtelen
Blood samples Hepannized penpheral blood was taken
from the patient when leukemic cells were present, dur-mg the time of complete rernission (pre-BMT), and from the BM donor PBMCs from the recipient, the BM do-nor, and randomly selected donors were isolated on Ficoll—Isopaque, washed, and cryopreserved in liquid m-trogen m medium containing 20% DMSO
PBMCs of some blood donors were separated into dif-ferent cellular fracnons To this end, PBMCs were sep-arated into T-cell-ennched (erythrocyte rosette positive [ER"1 ]) and T-cell-depleted (ER~) fractions by rosetting
with AET-treated sheep red blood cells (SRBCs) Τ cells were recovered from ERH fractions by lysing the ery
th-rocytes in NH4CL, yielding > 9 5 % CD3 + Τ cells The
T-cell-depleted fraction was cultured for 36 hours in tissue culture dishes By reculturing the dislodged cells twice, monocytes selectively attach to the plastic culture dishes The adherent monocytes were coüected by firm
scraping and yielded > 9 8 % C D l 4+ monocytes.
Non-adherent cells were layered onto 14 5% metnzamide to separate Β and natural killer (NK) cells from the
den-dntic cells and residual monocytes The Β and NK cell fraction generally consisted of about 60% Β cells and 40% NK cells.
Cell culture medtum The culture medium used was RPMI
1640 (Dutch modification, Gibco) supplemented with 3 mM L-glutamin (Gibco), antibiotics (100 IU/ml peni-cillin and 100 μ-g/ml streptomycin), and 10% (vol/vol) inactivated pooled human serum
EBV transformation For in vitro EBV transformation of
the selected cells, the supernatant of the EBV producer cell hne Β 95-8 was used [11]
Lymphoblastoid cell hnes from in vivo EBV-infected Β cells were kindly provided by Dr J W Gratama [12] Α spontaneous outgrowth of in vivo infected Β cells could be observed when PBMCs from patients and do-nors before and after BMT were cultured in RPMI-1640 medium containing 10% FCS and 0.5 μg cyclosponne Between 1 and 2 weeks after the Initiation of the rul-tures, fresh cord blood cells were added in order to rescue the EBV released from in vivo virus-carrying cells [12] Estabhshed lymphoblastoid cell hnes could be expanded on medium containing 10% FCS
Generation of B-cell blasts B-cell blasts were obtained
[13]-In short, Β cells were ennched from PBMCs using AET-coated SRBC and leucyl-Ieucine methyl ester El-4 Β 5 cells, kindly provided by Dr R Η Zubler stimulated these B-cell-enriched fractions for 6—10 days in the pres-ence of 1% PMA and 20% TCGF
T-cell line and clones T-cell line MAx was ininated wich
day-183 post-BMT patient's PBMCs The cells were cul-tured wich irradiated patient s leukemic cells at a ratio of 1 1 At day 6 the responder cells were specifically re-stimulated with the panent's leukemic cells in the pres-ence of 20 U/ml r-IL2 Thereafter, the T-cell line was cultured by weekly Stimulation with the patient s leu-kemic cells and 1% leuco-A, alternated by addition of fresh medium containmg 15% TCGF (Biotest) The T-cell line was subsequently cloned using a FACSorter at 1 cell/well The T-cell clones were expanded using an irradiated feedermix of PBMCs from six random blood donors and the patient's leukemic cells in the presence of 20 U/ml r-IL2
Proltferatton assay To assay the proliferative response
upon Stimulation, the T-cell line and clones were tested in PLTs Responder Τ cells, 104, were added in tnplicate
in 96-well flat-bottomed plates to either 105 PBMCs,
monocytes, Β cells, dendntic cells, or leukemic cells (3000 rad irradiated) or 0 25 Χ 105 EBV-transformed
lymphoblastoid cell line (EBV-LCL) or B-cell blasts (8000 rad irradiated) After 48 hours of culture 1 μ θ [iH}thymidine was added for the last 18—20 hours
Pro-liferation IS expressed in Stimulation indices according to the formula
SI = cpm RC X SC
(cpm RC alone + cpm SC alone)
where Sl IS the Stimulation index, RC means responder cells, and SC means stimulator cells Antibody blocking assays were performed using the following monoclonal antibodies (mAbs) B9 12 1 and IC2 for major histocom-patibility complex (MHC) class I and II, respecnvely, the DP-specific B7 21 (Btcton-Dickmson), the DQ-specific SPvL3 {14} and B8 11 2 {15] against the DR backbone, and 7 3 19 1 as a DRß3-cham-specific antibody [16}
FACScan analysis Both responder and stimulator cells were regularly aasayed for expression of MHC class I and class II molecules, accessory molecules ICAM-1, LFA-1, LFA-3, VLA-1, -2, -4, -6, B7, <ind CD28, CD4, and CD8 All rnouse mAbs used in the FACScan analysis were obtamed from Becton-Dickmson and are either di-rectly conjugated with FITC or PE or used in combina-non with fluorescein-conjugated GAM antibodies
Α semiquantitative analysis of expressed markers
could be performed by using microbeads with various defined amounts of bound molecules of equivalent solu-ble fluorochrome (MESF) In this way the mean fluores-cence could be related to the MESF and different cell types can be compared for their expression of certain molecules
RESULTS
MHC c/ass Η HLA-DRu>52-restrtcted recogmtton of T-cell clone M-33 Α prohferative T-cell clone, M-33, was
gen-erated from post-BMT PBMCs, as outlined in Materials
and Metbods This C D 4+ T-cell clone displayed a
pro-liferative activity against the patient's pre-BMT PBMCs and leukemic cells but not against the patient's pre-BMT EBV-LCLs(Fig 1A) The proliferative response of T-cell clone M-33 upon Stimulation by the patient's pre-BMT PBMCs or leukemic cells could be completely blocked by mAbs directed against HLA class 11 and HLA-DR, and by the DRß3-chain-specific antibody, but not by HLA
class-I-specific mAb (Fig 1B) To further define the MHC class II restnction allele, T-cell clone M-33 was assayed against a panel of PBMCs of 23 heakhy unrelated blood donors (Table 1) Among these donors, those who shared HLA-DRw52-a or -b (cellularly typed Dw25 or Dw26) with clone M-33 were snmulatory No Dw25/ Dw26-positive PBMCs were found that could not stim-ulate clone M-33 So we conclude that specific prohfer-ation of T-cell clone M-33 can be induced by Dw25- or Dw26 (DRw52)-positive stimulator cells in all cases tested
T-cell clone M-33 is not specific for the myelotd hneage As T-cell clone M-33 was generated and expanded by Stim-ulation with the patient s myeloid leukemic cells, we analyzed whether the specific recognition of M-33 was restricted to the myeloid lineage Five leukemias with a different morphology were selected on the presence of the defined restnction molecule HLA-DRw52 Table 2 sum-manzes the morphology and phenotype of these leuke-mias and their capacity to snmulate M-33 and the HLA-Dw2 5-specific T-cell clone LBQ-1 Besides the patient's myeloid leukemic cells, cells from another acute myeloid leukemia (Z) differentiation stage, M2, were recognized
Two acute lymphoblastoid leukemias with different phe-notypes (K and H) and a Burkitt-like lymphoma (N) were also stimulatory to T-cell clone M-33 One undif-ferentiated leukemia (L) with a low expression of HLA-DR was only stimulatory to the allospecific clone LBQ-1 but not to M-33
140 Ε G van Lochern et al
PBL pre BMT leukemic cells EBV-LCL pre BMT
Α stimulator cells
Β
B9 121 1C2 B7 21 SPvL3 B8112 73 191
leukemic cells as stimulator cells
iblockmg mAb
FIGURE 1 (A) Proliferation of T-cell clone M-33 upon Stimulation wich the patient's pre-BMT penpheral blood lym-phocytes (PBLs), leukemic cells, and EBV-LCLs Proliferation IS expressed as Stimulation index (SI) (B) Proliferation of T-cell clone M-33 upon Stimulation with patient's leukemic cells tn the pres-enceofdifferent mAbs B9 12 1 and IC2 are directed agamst HLA class I and class II, B7 21, SPvL3, and B8.ll 2 against HLA-DP, -DQ, and -DR, respectively, 7 3 19 1 is directed against the product of the ß3 locus of HLA-DR Proliferation
of M-33 is expressed as Stimulation index (SI)
T-cell clone M-33. Notwithstanchng the capacity of these Dw25/Dw26-positive EBV-LCLs to evoke a prolif-erattve response of the allospecific clones LBQ-1 and clone 12, respectively, none of these 18 EBV-LCL hnes were able to stimulate T-ceü clone M-33 On the con-trary, PBMCs of the same individuals are very potent snmulators of clone M-33
To address the question of whether only the EBV-transformed Β cells are mcapable of stimulating T-cell
clone M-33, PBMCs of one individual blood donor (W) were separated into different cellular fractions Figure 2 shows that unfractionated PBMCs, monocytes (95% CDl4 + ), and the B/NK cell fraction (contammg 60%
C D 1 9+ Β cells and 35% C D 1 6+ NK cells) were equally
stimulatory. Again, solely the EBV-transformed Β cells of the same individual (W) appeared to be unable to stimulate T-cell clone M-33. The unresponsiveness of clone M-33 upon EBV-LCL Stimulation could not be
overcome by the addition of third-party monocytes or B/NK cells, lacking the restnction determinant, to Dw25-posmve EBV-LCLs (Fig 2), nor could addition of IL-2 (data not shown) Cultunng EBV-LCLs in serum-free culture medmm or in culture medium supplemented with human serum instead of fetal calf serum did not change their stimulatory capacity (data not shown)
Absence of stimulatory capacity of EBV-LCLs is not due to tmpaired expression of MHC or adhesion molecules To
TABLE 1 Α panel of PBMCs of HLA-typed blood
donors is used in PLT assays as snmulator for T-cell clone M-33
Stimulator cells Leukemia paticnt PBL patienc PBL Duiv PBL Sieg PBL Geld PBL Grat PBL Guld PBL Heil PBL KluN PBL Kost PBL Krsu PBL Krui PBL Laga PBL Mast PBL Miku PBL Mimt PBL Nieu PBL Oost PBL Plai PBL Ruig PBL Ri|n PBL Vane PBL VhR PBL Brau PBL Witvl DR 7, wl2(5)* 7, wl2(5) 3, 7 wl3(6), 7 1 wll(5) 1, 3 wl3(6), wl2(5) wl5(2), 4 wl5(2), wl3(6) 3 wl5(2) 4, 10 wl3(6), 9 wl5(2), 10 1, 8 2, 3 1, wll(5) 1, 8 1, wl3(6) 1, wl2(5) wl3(6), w8 3, 4 3, 4 2 7 3 Dw 2Ϋ 25 25 19126 25 24 18/25 24 18/24 NT NT NT 18125 NT NT 24 25 NT W25 25 19/26 24 24 NT 25
sr
520 122 26) 517 301 1 397 1 1 2 1 1 258 1 1 1 62 1 211 330 211 2 2 1 312 Λ Prohferation is expressed as Stimulation index (SD* All stimulator cells atc serologically typed for HLA-DR and a majonty was also cellularly typed fot Dw25/Dw26
PBL, penpheral blood lymphocyte, and N T not tested
on EBV-LCLs in a same or even higher amount than on leukemic cells As the patient's Ieukemic cells were po-tent stimulators for clone M-33, the unresponsiveness of M-33 upon Stimulation with EBV-LCLs cannot be due to a difference in the expression of MHC or adhesion mol-ecules
Absence of stimulatory capactty ts related to EBV Iransfec-tion Α direct role for the EBV in the incapability of
EBV-LCL to stimulate M-33 became apparent when in vivo EBV-infected B-cell lines were assayed and com-pared with EL-4 B5-stimulated B-cell blasts of the same individual (Fig 3) From three different donors in vivo infected LCLs were tested in parallel with in vitro trans-formed EBV-LCLs Similar to the in vitro transtrans-formed EBV-LCLs, the in vivo infected Β cells are also not rec-ognized However, nonleukemic, non-EBV transformed but in vitro stimulated B-cell blasts (generated by snm-ulating Β cells wich EL-4 B5 cells) could be recognized by T-cell clone M-33 (Ftg. 3).
DISCUSSION
Α prohferative T-cell clone, M-33, was generated by frequently stimulating post-BMT PBMCs wtth the pa-tient's own myeloid leukemic cells Initially, a potential GvL activity of this T-cell clone was suggested based on lts prohferative response upon Stimulation with the pa-tient's leukemic cells and absence of Stimulation with the pacient's pre-BMT EBV-LCLs. Because the patient's leu-kemic cells were of the rnyeloid hneage, myeloid-specific
TABLE 2 Lymphomas with different morphology are used as stimulator cells for T-cell clone M-33 and
HLA-Dw25-specific clone LBQ-1 Patient
designaticn
HLA typing
C DR Morphology Phenotype
T-cetl clones (SI)" M-33 LBQ-1 Ζ
κ
Ν Η L AI A2, A2 A2, A2, 3 3 19 B8, B12 B38 B7 B8 B8, 27 Cw5, 7 DR3, w52 Cw5 DR5, 7, w52 B-ALL AML-M2 1%CD3 + , 95%CD14 + , 5%CD19' , >95%DR + 1%CD3 + , 2%CD14 + , 97%CD19+/CD34+/DRH DR5, 6, w52 Burkitc-hke 9%CD3 +, 3%CDl4 + , lyraphoma 85%CD19 + , 2%CD34 + , Cw3, 7 DR3, 7, w*>2 Pre-B-ALL Cw2, 7 DR3, 5, «52 AUL 90%CD19 + , <1%CD34 14%CD3 + , 2%CD14 + , 1%CD19 + , 7%CD34 + , 9%DR +142 E.G. van Lochern et al.
PBL W EBV-LCL W MonoW B/NK W PHA Τ cell blast W
PBL Μ EBV-LCL Μ " + 3rd party B/NK " + 3rd party mono
35 30 25 20 15 10 5 0 25 50 75 10O Stimulation index Stimulation index
FIGURE 2 Different ceLLulac subsets of donot W are tested for their stimulating capacity for clone M-33 and clone LBQ-1. EBV-LCLs of patient Μ pre-BMT are tested in PLT for their Stimulation in the presence of third-party B/NK cells and monocytes.
reactivity was postulated. The subsequent induction of proliferation by a series of HLA-DRw52-positive alloge-neic PBMCs (Table 1) were in line with the norion that T-cell clone M-33 would recognize a nonpolymorphic myeloid-lineage-specific antigen.
This hypothesis of lineage specificity of clone M-33 had to be rejected as, first, leukemias of Β lineage and, second, different celiular subsets of healthy blood donors were found to be stimulatory. With regard to the former, of the five leukemias selected on the presence of the required restriction molecule DRw52, four were rec-ognized by T-cell clone M-33 (Table 2). Of these four only one was of the myeloid lineage. One leukemia (L) with an undifferentiated morphology was not recognized by clone M-33. This might be due to the low expression TABLE 3 Surface expression of MHC and accessory
molecules on leukemic cells and EBV-LCL
Surfice molecules MHC class I MHC class II HLA-DR HLA-DP HLA-DQ ICAM-1 LFA-3 VLA-1 VLA-2 VLA-4 VLA-6 B7 Leukemic cells (MESF Χ 104) 14 5.8 2.8 2.3 1.1 0.77 0.53 0.34 0.03 1.2 0.64 1.3 EBV-LCL (MESF X \04) 52 56 Π 29 19 26 2.6 0.52 0.75 3.9 0.46 5.0 Ratio MESF EBV-LCL. leukemia 3.71 9.65 3.93 12.6 17.3 33.8 4.91 1.53 25.0 3.25 0 72 3.85
HLA molecules and a senes of accessory molecules are analyzed for cheir expression on leukemic ceils and pre-BMT EBV-LCLs of patient Μ. Α semi-quantitative analysis was performed by relacing the mean fluorescence to MESF.
of HLA-DR, although it was clearly recognized by the allospecific clone LBQ-1. The stimulatory capacity of different celiular fractions, such as monocytes, Β cells, and dendritic cells, derived from a healthy blood donor (Fig. 2), argues strongly for the recognition of a common determinant by T-cell clone M-33 in the context of seif HLA-DRw52.
Nonetheless, distinct from all stimulator cells tested, clone M-33 could not be stimulated when EBV-LCLs were used as APCs. We postulate different explanations for the absence of the stimulatory capacity of EBV-LCLs. First, the EBV-LCLs could be inadequate as APC for clone M-33. We reported earlier that EBV-LCLs could be inadequate as APC for mH-specific Th cell response induction, although we observed that EBV-LCLs func-tioned as potent secondary stimulator cells in PLT assays [17], We do not favor the possibility of the EBV-LCL being an inadequate APC, as allospecific proliferative T-cell clones, recognizing Dw25 or Dw26, responded well upon Stimulation with the same EBV-LCL (Fig. 3). Second, it has been shown that the level of class II MHC expression on APCs contributes in a quantitative fashion to the efficiency of T-cell activation [18, 19}- Α downregulated expression of MHC products could there-fore underlie the absence of stimulatory capacity. EBV-LCLs, however, exhibit high levels of MHC products and accessory molecules, as evaluated by semiquantitative FACScan analysis (Table 3)· Compared with the patient's leukemic cells, which are highly stimulatory to clone M-33, EBV-LCLs express equal to 30-fold higher amounts of accessory molecules or MHC products. It is therefore unlikely that an altered expression of these mol-ecules accounts for the lack of recognition by clone M-33.
FIGURE 3 Stimulation indices of PB-MCs vs EBV-LCLs of four donors were depicted for clone M-33 and clone
LBQ-1. From three donors (D, H, and M) in vivo infected EBV-LCLs are tested in par-allel for Stimulation.
PBL W EBV-LCL W El-4 B5 Β blast W PBLD in vitro EBV D in vivo EBV D PBL Η in vitro EBV Η in vivo EBV Η PBL Μ in vitro EBV Μ in vivo EBV Μ • M-33 • LBQ-I 40 30 10 Stimulation index 25 50 75 100 Stimulation index
to subsequent antigenic Stimulation [20). This T-cell anergy can be overcome by providing the costimulatory signal by third-party APCs. However, the unresponsive-ness of M-33 upon Stimulation with EBV-LCL as APC could not be overcome by third-party monocytes or Β cells (Fig. 2), nor by addition of exogenous IL-2 (data not shown).
We propose a role for the EBV infection in the abro-gation of the stimulatory capacity of Β cells. This idea is favored by our finding that both in vivo and in vitro infected Β cells are the only APCs that do not stimulate T-cell clone M-33 (Fig. 3). Freshly isolated Β cells, leu-kemic B-cell blasts, and EL4B5-stimulated B-cell blasts, on the contrary, are potent stimulator cells of T-cell clone M-33· It is known that viral peptides of EBV can be presented in the context of MHC class Π molecules leading to a proliferative T-ceü response {17}. We
hy-pothesi2e that viral-encoded peptides compete for bind-ing to MHC class II, with the antigen recognized by T-cell clone M-33. Furthermore, the results with this T-cell clone demonstrate that the screening procedure with leukemic ceils and EBV-LCLs can lead to a false Interpretation of Leukemia-specifk activity and has to be followed by subsequent specificity tests.
In conclusicn, we describe here a proliferative T-cell clone, M-33, which recognizes a common antigenic de-terminant in the context of seif HLA-DRw52. This as yet undefined antigenic determinant is expressed by all (« = 18) panel cells tested, and can be presented by all but one APC tested. EBV infection of Β cells, both in
vivo and in vitro, is found to abrogate the stimulatory capactty for this patticular T-cell clone.
ACKNOWLEDGMENTS
The authors thank Dr. J . W. Gracama for providing ehe in vivo infected EBV-LCLs and Dr. F. Claas and F. Koning for
critically reading of this manuscript. This work was supported by grants from the Dutch Cancer Foundation (Koningin Wdhelmina Fonds) and the J . A. Cohen Institute for Radiopa-thology and Radiation Protection (IRS).
REFERENCES
1. Bortin MM: Acute graft-versus-host disease following bone marrow transplantation in humans: prognostic fac-tors. Transplant Proc 19:2655, 1987.
2. Horowitz MM, Gale RP, Sondel PM: Graft-versus-leukemia reactions after bone marrow transplantation. Blood 75:555, 1990.
3. Butturini A, Bortin MM, Gale RP: Graft-versus-leukerrna following bone marrow transplantation. Bone Marrow Transplant 2:233, 1987.
4. Weiden PL, Flournoy N, Thomas ED, et al: Antileuke-mic effect of graft-versus-host disease in human reeipients of allogeneic-marrow grafts. Ν Engl J Med 300:1068,
1979.
5. Weiden PL, Sullivan KM, Flournoy N, Storb R, Thomas ED: Antileukemic effect of chronic graft-versus-host dis-ease: contnbution to improved survival after allogeneic marrow transplantation. Ν Engl J Med 304:1529, 1981. 6. Truitt RL, Shih C, Lefever AV, Tempelis LD, Andreani M, Bortin MM: Characterization of alloimmunization-induced Τ lymphocytes reactive against AKR leukemia in vitro and correlation with graft-vs-leukemia activity in vivo. J Immunol 131:2050, 1983.
7. SosmanJA, Oettel KR, Smith SD, Hank JA, Fisch P, Sondel PM: Specific recognition of human leukemic cells by allogeneic Τ cells. II. Evidence for HLA-D restricted determmants on leukemic cells that are crossreaenve with determinants present on unrelated nonleukemic cells. Blood 75:2005, 1990.
phenorne-144 Ε G van Lochern et al
non is GVL separable from GVHD-1 Bone Marrow
Trans-plant6 155, 1990
9 Faber LM, Luxemburg-Heijs SAP, WiUemze R, Falken-burg JHF Generation of leukemia-reacave cytotoxic Τ lymphocyte clones from the HLA-identical bone marrow donor of a patient wich leukemia J Exp Med 176 128-$,
1992
10 van Lochern EG, de Gast GC, Goulmy Ε In vitro Sepa-ration of host specific host and graft-versus-leukemia cytotoxic Τ cell activities Bone Marrow Trans-plant 10 181, 1992
11 Miller G, Lipman Μ Release of infectious Epstein—Barr virus by marmoset leukocytes Proc Natl Acad Sei USA 70 190, 1973
12 Gratama JW, Oosterveer MAP, Lepoutre JMM, van Rood JJ, Zwaan FE, Vossen JMJJ, Kapsenberg JG, Richel D, Klein G, Ernberg I Setological and molecular studies of Epstein—Barr virus infection in allogeneic mar-row graft recipients Transplantation 49 725, 1990 13 Zubler RH, Erard F, Lees RK, van Laer M, Mingan C,
Moretta L, MacDonald HR Mutant EL-4 thymoma cells polyclonally aenvate munne and human Β cells via direct cell interaction J Immunol 134 3662, 1985
14 Spits H, Borst J, Giphart MJ, Cohgan J, Terhorst C, de Vnes J HLA-DC anngens can serve as recognition ele-ments for hurnan cytotoxic Τ iymphocytes Eur J Immu-nol 14 299, 1984
15 Rebai N, Mahssen B, Dieres M, Acolla RS, Corte G,
Mawas C Distinct HLA-DR epitopes and distinct fami-hes of HLA-DR molecules defined by 15 monoclonal an-tibodies (mAb) either anti-DR or allo-anti-Ia cross-reacting with human DR molecule I Cross-inhibition studies of mAb cell surface fixation and differential bind-mg of mAb to detergent-solubilized HLA molecules lm-mobihzed to a solid phase by a first mAb Eur J Immunol
13 106, 1983
16 Koning F, Schreuder GMTh, Giphart MJ, Bruning JW Α mouse monoclonal antibody detecting a DR-related MT2-like specificity serology and biochemistry Hum Immunol 9 221, 1984
17 van Eis CACM, Bakker A, van Rood JJ, Goulmy Ε Induction of minor histocompatibility antigen specific Τ helper, but not Τ cytotoxic response is dependent on the source of antigen presenting cell Hum Immunol 28 39,
1990
18 Matis LA, Gltmcher LH, Paul WE, Schwartz RH Mag-nitude of response of histocompatibihty-restncted T-cell clones is a function of the produet of the concentrations of antigen and la molecules Proc Natl Acad Sei USA 80 6019, 1983
19 Lechler RI, Norcross MA, Germain RN Qualitative and quantitative studies of antigen presenting cell function by using I-A-expressing L cells J Immunol 135.2914,
1985
