Selective reactivation of Epstein-Barr virus-specific cytotoxic T cells by stimulation in vitro with allogeneic virus-transformed HLA-homozygous typing cells

Selective Reactivation of Epstein-Barr Virus-specific Cytotoxic T Cells by Stimulation In Vitro with Allogeneic Virus-transformed HLA-Homozygous Typing Cells Martin Rowe, Alan B. Rickinson, Stephen R. Beer, M. Anthony Epstein, and Ben A. Bradley

A B S T R A C T : Epstein-Barr (EB) virus-specific cytotoxic T-cell preparations, produced by stimulation in vitro of peripheral blood lymphocytes with the autologous virus-transformed cell line. are HLAA and B antigen-restricted and. with some donors, show preferential restriction through one or two of the four relevant antigens of the donor's HLA type. It has now been demonstrated that such EB virus-specific cytotoxic T cells may also be reactivated by stimulatio, n with allogeneic virus-transformed cells provided that there is no mismatch of the HLA-A and B antigens between the responder and stimulator cell donors. In particular, virus-transformed cell lines from HLAhomozygous donors HLA-A and B antigen-matched to one of the haplotypes of an HLA-heterozygous responder were shown to reacth;ate selectively only those EB t,irus-specific cytotoxic T cells restricted through the HLA-A and B antigens present on the allogeneic stimulating cells. In addition to confirming the polyclonal nature of the HLA-restricted EB virus-specific cytotoxic Tcell response, this new experimental procedure has allowed the production, and subsequent expansion as cell lines dependent upon T-cell growth factor, of those effector cells restricted through the "nonpreferred" HLA antigens that are poorly represented in the response induced by stimulation with autologous rirus-transformed cells. ABBREVIATIONS EB virus LCL LYDMA

Epstein-Barr virus EB virus-transformed Blymphoblastoid cell line lymphocyte-detected membrane antigen

NK PHA TCGF U M cells

natural killer phytohemagglutinin T-cell growth factor unfractionated mononuclear cells

INTRODUCTION It is n o w w e l l established that E B * v i r u s - s p e c i f i c c y t o t o x i c T - c e l l p r e c u r s o r s , which a r e p r e s e n t e x c l u s i v e l y in p r e v i o u s l y i n f e c t e d ( s e r o p o s i t i v e ) i n d i v i d u a l s , can b e r e a c t i v a t e d in c u l t u r e s o f b l o o d m o n o n u c l e a r cells e x p e r i m e n t a l l y i n f e c t e d w i t h

From the Departmentof Pathology, Unit~ersityof Bristol MedicalSchool(M.R., A.B.R., S.R.B., M.A.E. J and the UK Transplant Serrice. SW Regional Transfusion Centre. Southmead Hospital (B.A.B.~. BrL~t(d, England. Address correspondenceto Dr. Martin Roue, Department of Pathology', Unicersi(~ of Bristol Medical School. University Walk, Bristol BS8 lTD. England. ReceivedAugust 5, 1982,"acceptedOctober20, 1982. Human Immunology 6, 151-165 (1983) ~' Elsevier Science Publishing Co.. Inc., 1983 52 Vanderbilt Ave., New York, NY 10017

151 ()198-8859/8~,/0¢(}151 -15S~.00

152

M. Rowe et al. the virus [1]. The effector cells thus obtained are specific for an EB virus-induced LYDMA* and kill EB virus-transformed target cells in an HLA-A and B antigenrestricted manner [ 1-3]. The inference from these studies is that effector T-cell preparations from HLA-heterozygous donors contain a number of component reactivities (clones) each recognizing LYDMA in association with one HLA-A or B antigen, a view which has now been supported by recent results from clonal analysis [4]. Thus, the preferential restriction which certain effector preparations show for just one or two of the four possible HLA-A and B determinants presumably reflects a predominance of these particular clones within the population which is reactivated in vitro. If this is indeed the case, then it should be possible to induce selective reactivation of particular EB virus-specific cytotoxic T-cell clones from the responsive population (perhaps even those clones not normally predominant) by challenging with cells presenting LYDMA in association with some but not all of the relevant HLA antigens. Three recent developments have made it feasible to test this prediction experimentally. First, it became clear that EB virus-specific cytotoxicity, exactly as first recognized in virus-infected cultures, could also be reactivated by stimulation in vitro of blood mononuclear cells from immune donors with X-irradiated cells of the autologous LCL* [5] provided that a responder to stimulator ratio (40:1) and culture time (10 days) were used which minimized any accompanying reactivation of NK*-like cytotoxicity [ 5-7 ]. Second, such virus-specific effector preparations could be expanded into continuously growing T-cell lines using TCGF*, thus providing large numbers of effector cells for detailed analysis of their HLArestricted function [8]. Third, EB virus-transformed LCL became available from a range of donors who were homozygous for HLA-A and B antigen expression, thus offering a source of stimulating cells where LYDMA was expressed in association with one HLA-A and one HLA-B antigen only. The present paper desoribes experiments in which responder T-lymphocytes from EB virus-immune donors were stimulated in vitro with allogeneic LCL cells bearing only those HLA-A and B antigens contained within the responder phenotype, and the subsequent cytotoxic T-cell response was analyzed for EB viral antigen specificity and for HLA-antigen restriction.

MATERIALS AND METHODS

Blood donors and preparation of mononuclear cells. The donors used as a source of responder cells in these experiments were healthy adults whose serological status with respect to EB virus had been determined by immunofluorescence testing for antibodies both to the EB viral capsid antigen and to the EB virus-associated nuclear antigen [9], and whose capacity to mount an EB virus-specific cytotoxic T-cell response to autologous virus-infected B cells had been monitored in the in vitro regression assay [ 10,11]. (UM*) leukocytes were prepared from the blood of these donors by isopycnic centrifugation on Ficoll/Hypaque [ 10]. The donors used as a source of stimulating LCL cells in these experiments were selected on the basis of their particular HLA types and included some donors who were homozygous for HLA-A and B antigens. UM cell preparations from many of these homozygous donors were kindly made available through Professor J. van Rood, Academic Hospital, Leiden, The Netherlands.

Preparation of stimulator cell lines. EB virus-transformed LCL were established in vitro from selected donors by experimental EB virus infection of UM cells which had first been depleted o f T cells by E rosetting [3].

HLA Typing. HLA-A,B,C, and DR typing was performed on both the UM cells and the derived LCL cells from all the donors used in these experiments (Table

HLA-restricted EB Virus-specific Cytotoxicity TABLE 1

153

Sources of responder and stimulator cells HLA-type of LCL stimulator cells

Donor of responder UM cells

Donor of stimulator LCL cells

A

B

C

DR

CG

Auto (CG) JS AVL LGYS PHS

1,3 1,3 1,1 1,1 3,3

7,8 7,8 8,8 8,8

w7 w6,w7 --" --

2,3 3,7 3,3 3,3

SG

Auto (SG) MO BVR CJO H53 BW

3,11 3,11 11,11 11,11 3,3 3,3

w35,w40 w35,w40 w35,w35 w35,w35 w35,w35 w35,w35

w2,w4 w3,w4 w4,w4 w4,w4 w4,w4 --

6 2,4 1,1 1,1 1,1 1

SW

Auto (SW) VYF CJO

11,w24 w24,w24 11,11

7,w35 7,7 w35,w35

w4,w7

2,4

--

2,2

w4,w4

1,1

77

-

-

2,2

"Dash i--) indicates that C-locus antigens could not unequiw)cally be identified with the sera available.

1). Typing of the UM cells was performed in the usual way using a standardized N I H complement-mediated cytotoxicity assay, and typing of the LCL was performed as described fully elsewhere [12].

Cellculture. All cells were cultured in RPMI 1640 medium containing antibiotics (100 IU/ml penicillin, 100 fzg/ml streptomycin) and 10% fetal calf serum. Cultures were maintained at 37°C in humidified CO2 incubators and fed twice weekly by replacing half of the supernatant with fresh medium. Some effector T-cell lines were maintained in medium supplemented with TCGF which was prepared by a modification of the method of Inouye et al. [8,13].

Stimulation and preparation of effector T cells in vitro. Freshly prepared responder UM cells were cultured at 2 x 106 cells per 2-ml well in multiwell tissue culture plates (Linbro; Flow Laboratories, Irvine, Scotland) together with X-irradiated stimulator cells (5 x 104 cells/well) from either the autologous or from an allogeneic LCL. Effector T-cells were isolated from the cocultures after 10 days by rosetting with sheep erythrocytes treated with 2-aminoethylisothiouronium hydrobromide as already described [3]. In some experiments, these T-cell preparations were expanded to give continuously growing TCGF-dependent cell lines in the following way. First, the T cells were recultured at 5 x 105 cells per 2-ml well with fresh X-irradiated stimulator cells (1.25 x 10 ~ cells/well) for 4 days. Thereafter, the cells were harvested twice weekly and recultured in medium supplemented with TCGF, together with the relevant X-irradiated stimulator cells. Effector cells from TCGF-dependent cell lines were always cultured overnight in TCGF-free medium before being assayed for cytotoxicity [8].

Target cellsfor cytotoxicity assay. EB virus-transformed LCL from selected donors were established as described above for the stimulator cell lines. EB virus genomenegative cell lines used as control targets in these experiments were HSB2 from an acute lymphoblastic leukemia of T-cell origin [14], K562 from a chronic myeloid leukemia [15], and EB4 from an EB virus genome-negative "Burkittlike lymphoma" of B-cell origin [16]. Mitogen-stimulated lymphoblasts were

154

M. Rowe et al. prepared by culturing freshly-prepared UM cells at 2 × 10c' cells per 2-ml Linbro well in the presence of PHA* (1:100 dilution of stock Gibco solution [Flow Laboratories, Irvine, Scotland]). The PHA blasts were harvested after 3 days and recultured in mitogen-free medium for a further 18 hr before use.

Chromium-release assay for T-cell cytotoxicity. Cellular cytotoxicity was assayed by incubating effector cells with 104 51Cr-labeled target cells for 5 hr at 37°C as described fully elsewhere [3]. The effector-to-target cell ratios (E/T) used in these experiments were between 1:1 and 10:1 as indicated in the text. In the cold-target competition studies, unlabeled competitor cells were mixed at various ratios with the constant number of labeled target cells before addition of the effector cells. In the antibody blocking experiments, concentrated immunoglobulin preparations of monoclonal antibodies specific for various target cell surface antigens were included in the assay at concentrations known to give saturation of binding of the antibody to the target cells [ 17,18]. The antibodies used were W6/32 [19] and BB7.7 [20] specific for HLA-A,B,C common framework determinants; TDR 31.1 [21] specific for an HLA-DR common framework determinant; MHM6 [18] specific for an EB virus-induced B-cell surface antigen; and AC2 [ 18] specific for a lymphoblastoid cell surface antigen. (The immunoglobulin preparations of W6/32, BB7.7, and TDR31.1 were kindly provided by Drs. M.J. Crumpton and T.A. de Kretser of the I.C.R.F., Lincoln's Inn Fields, London.) Experimental Procedure In an initial set of experiments, responder UM cells from seropositive donors were cocultured at a responder-to-stimulator ratio of 40:1 with X-irradiated cells of the autologous LCL or, in parallel cultures, with X-irradiated cells of an allogeneic LCL from a donor whose HLA-A and B type exactly matched that of the responder. T-cell preparations from these two types of coculture were tested in parallel for cytotoxicity against a range of EB virus genome-positive and genome-negative target cell lines of known HLA type. In the main body of experiments, responder UM cells from seropositive donors were cocultured with allogeneic X-irradiated LCL cells established from HLAhomozygous donors whose HLA-A and -B antigen type was contained within the responder HLA type. Effector T cells reactivated in this type of coculture were expanded as cell lines dependent on TCGF and were tested for cytotoxicity against a range of EB virus genome-positive and genome-negative target cells of known HLA type, including mitogen-stimulated lymphoblasts from the donors of both the responder and stimulator cells. The specificity of these effector cells was further analyzed in cold-target competition studies and in blocking studies using monoclonal antibodies to HLA and other target cell surface antigens. RESULTS The experiments employed three seropositive donors of responder cells (CG, SG, SW) and, for each of these donors, a panel of LCL stimulators including cells from the autologous line and cells from allogeneic lines bearing relevant HLAA and B antigen matches. The various responder/stimulator combinations tested are shown in Table 1, together with the corresponding HLA antigen types. In each case, HLA antigen typing of freshly prepared UM cells and of the derived LCL gave concordant results. The feasibility of using allogeneic LCL stimulators as a means of reactivating EB virus-specific cytotoxicity in vitro was established in the initial set of exper-

HLA-restricted EB Virus-specific Cytotoxicity

155

iments using combinations of responder and stimulator cells from unrelated HLAheterozygous donors who had been selected for complete HLA-A and B antigen identity. Thus, as shown in Figure 1, the pattern o f T-cell cytotoxicity generated in 10-day cocultures o f responder UM cells from donor CG with LCL stimulator cells from donor JS (CG --> JS cocultures) was very similar to that generated from these same responder cells by autologous LCL stimulation (CG --> CG cocuhures). In each case the cytotoxic reactivity was preferentially directed against the autologous LCL with no accompanying lysis of the EB virus genome-negative cell lines, HSB2 and K562, included as sensitive indicators of nonspecific NKlike cytotoxicity [5-7]. Moreover, lysis of allogeneic LCL targets was in each case restricted to those having a degree of HLA-A and B antigen-matching with the effector cells. Indeed, it was interesting to note that an unusually strong crossrecognition o f the RT target cell line through shared HLA-A3,B7 was observed with both effector cell populations. In addition to the data in Figure 1, the marked similarity between the cytotoxic T-cell responses induced by autologous LCL stimulation and by HLA-A and B antigen identical allogeneic LCL stimulation

FIGURE 1 Percentage specific lysis of Cr-labeled cell lines (at E/T ratio of 10:1) by effector cells isolated after 10 days coculture of responder UM cells from donor CG with (a) autologous stimulator LCL cells (CG --> CG) or (b) allogeneic stimulator cells from the HLA-A and -B antigen identical donor JS (CG --> JS). The HLA-A and B antigens shared with the autologous (CG) cells are given for each target, and lysis of the stimulating cell line is indicated by the solid bar for each effector population. (o)

50"

Effector c e i l s from (CG-~CG) co-culture

30-

20 °

0o

(b)

30.

20'

10'

0 Shored r HLA-A ontlgensL HLA-B

Effector c e l l s from (CG-.~-JS) co-culture

!1 |

li

CG

JS

AVL

LGYS

PHS

RT

1,3

1,3

1

1

3

3

7,8

7,8

8

8

7

7

EBV-DosltIve target LCL cells

BVR

FC

I I ! I I

m m HSB2

K562

1 I

I I I

I

F~V-negot lye target cells

156

M. Rowe et al. was confirmed in a repeat experiment involving responder CG and stimulator JS, and was also observed in two further experiments involving responder SG and stimulator MO (data not shown). The main body of experiments involved nine different combinations of responder cells with stimulator LCL cells derived from relevant HLA-homozygous donors (see Table 1). It was generally observed throughout this work that the cytotoxic T-cell response obtained after 10 days in such cocultures was less potent than that generated by autologous LCL stimulation, and therefore the specificity of the response was best examined following expansion of the effector cells in medium supplemented with TCGF. To illustrate this point, Figure 2 presents the results obtained from parallel in vitro activations involving responder donor SW (HLA-A 11, Aw24, B7, Bw35) and stimulator LCL cells from the homozygous donors CJO (HLA-A11, Bw35) and VYF (HLA-Aw24, B7). Little cytotoxicity was detectable after 10 days coculture, but, thereafter, expansion in medium supplemented with TCGF yielded populations showing potent cytotoxicity which in each case was preferentially active against the stimulating LCL and against the autologous LCL without undue contamination from any obviously unrelated reactivities. The finer specificity of the cytotoxic T-cell response obtained from these particular types of coculture became apparent from more extensive assays of the kind illustrated in Figure 3. Thus, effector cells derived from SW --> CJO cocultures (Fig. 3a) lysed the stimulating line CJO and the autologous line SW, but not the mitogen-stimulated lymphoblasts from the stimulating donor or the NKsensitive EB virus genome-negative cell lines. Moreover, lysis of the allogeneic LCL targets was restricted to those sharing HLA-All and/or Bw35 with the effector and stimulator cells, and did not include those targets which were HLAmatched with the effector cells only through HLA-Aw24 or B7. In contrast, the T-cell populations derived from the SW --> VYF cocultures (Fig. 3b) lysed the autologous LCL and the HLA-Aw24,B7-matched stimulating LCL, but not the target lines matched with the effector cells through HLA-A11 or Bw35. This population appeared to be restricted predominantly through HLA-B7 since target cells matched only through HLA-Aw24 (AN and TH, Fig. 3b; AM, results not shown) were not killed. Two further examples of cytotoxic T-cell lines established from cocultures of responder UM cells with allogeneic LCL from HLA-homozygous donors are shown in Figure 4. Here the stimulation of donor CG cells with the HLA-A 1,B8 cell line LGYS (Fig. 4a) induced an EB virus-specific response which appeared to be restricted predominantly through HLA-B8 (although only one suitable HLA°A1 matched target line, EF, was available for testing). Again, EB virus genome-negative target lines, including the HLA-A 1,B8-matched lymphoma line EB4, were not killed to a significant extent. In parallel cocultures, stimulation of these same responder cells with the HLA-A3,B7-homozygous cell line PHS (Fig. 4b) induced an equally potent virus-specific cytotoxic response which this time appeared to be restricted through the HLA-A3 and B7 antigens. The results illustrated in Figures 3 and 4 are representative of those obtained in the large majority of experiments where responder cells were cocultured with the relevant HLA-homozygous LCL stimulators. There were exceptions however, notably those cocultures involving responder SG and the stimulating lines H53 or BW, where a complex cytotoxic response was observed with strong lysis of the stimulating line and of certain other allogeneic LCL, apparently irrespective of their HLA type, in the absence of any reactivity against the autologous LCL. The specificity of these unusual responses was not investigated further. The remaining experiments were designed to characterize more fully the ha-

157

==

% o~

u~

~r

8

"~

e'~ ~

0

~

~

~---

~

~

....~.=O

~

o

.~

,~

e-"'~

1..~ ~.~

~4

SiSal 3lJ|3a~s aso~ua3Jad

~ .o_~

"S o

s

8 ~ 4 : l::> 3

i

g

o

g

g

g sis^|

9

3IJI3ads ~601ua3]~l

~

o

~-~j r~

158

~

N 0a~

g~

II

~

0

~'~

-~-~ o ~ ~

!

o

!

~'

o "ar,'~

"1

I

. . o ~

.o..~ ~ ~ &

~ - ~ ~ ~-.~ ~

$

d d

,.'e,, , R [, d._~.-cl

u~

8

~

8

>,

.~ ~>~

J

.,~

o

~

i.u

~ ~.~

[ • ,, >~,~,:~ \ , : ( <

~1~ ~

Ok,-,

: ,:x q.(

~80~ •

|

!

|

|

!

v

!

I

N~3o ~ ~

o ~ O~ ..--, '-a ~,~,

.

HLA-restricted EB Virus-specific Cytotoxicity

159

(a)

Effectors from (C@-,-,~LGYS) co-cu]ture

(b)

Effectors from (CGI~PHS) co-culture

60

40

20

E

60"

40

20

0

Sl]ared [ HLA-A ontiqens LHLA-B

CG

JS

AVL

LGYS

EF

1,3

1,3

i

i

l

7,8

7,8

8

8

CM

8

PHS

RT

H53

3

3

3

7

7

I

VYF

HSB2

1

1

7

K562

8 I

EEV-positL.'e taTgO[ LCL e i l ~

EB4

,

I

EBV negotJve target cells

FIGURE 4 Percentage specific lysis of Cr-labeled cells (at E/T ratio of 10:1) by TCGFexpanded T cells isolated from cocultures of responder UM cells from donor CG (HLA type A1,3;B7,8) with LCL stimulator cells from the HLA-homozygous donors (a) LGYS (HLA-A1;B8) or (b) PHS (HLA-A3;B7). The same notations as in Figures 1 and 3 are used.

ture of the cytotoxicity which the HLA-homozygous LCL stimulation usually induced. Figure 5 shows representative results from cold-target competition experiments in which effector T-cells derived from SW --> CJO cocultures were assayed against the autologous target line SW and against the H L A - A ! 1,Bw35matched stimulating line CJO in the presence of various unlabeled competitor cells. Lysis of the autologous line (Fig. 5a) was inhibited by unlabeled cells of the same kind and even more so by unlabeled CJO-LCL cells; whereas, VYFLCL competitors (matched with the effectors through HLA-Aw24,B7) and EB virus genome-negative HSB2 competitors caused little inhibition. Significantly, when the stimulating cell line CJO was used as the labeled target (Fig. 5b), its lysis was influenced by this same panel of competitors in exactly the same way, strongly suggesting that the two cytotoxic activities (against the autologous LCL and against the stimulator LCL) were mediated by the same cells and not by separate subpopulations of cells within the effector preparation.

160 u. N

~

''"

-~ ~ . ~,'b

°~ ,

~

,

~

,

~

,

~

,

.E

Slla3 ~a6JD~ q3q-OC3 ~0 sisal 31Jl3ads a~o~u~3Jad

..

~

~

~ "-~

~

~"

o ~ _O

~

~ ~ 0"-~

,

~

.

~

,

~

,

Slle0 ~e~J0~ 13~-M$ ~0 SlS~l 31~10~KI$ ~0~ua3Jad

HLA-restricted EB Virus-specific Cytotoxicity

161

In the final set o f experiments, cytotoxic T-cells induced by HLA-homozygous LCL stimulation displayed the same sensitivity to monoclonal antibody blocking as has been described for EB virus-specific, HLA-restricted effector cells generated by autologous stimulation procedures [17]. Thus, as shown in Figure 6, again using cytotoxic cells derived from SW ---> CJO cocultures, lysis of both the autologous LCL and of the stimulating LCL was strongly blocked in the presence of saturating concentrations of monoclonal antibodies binding framework determinants on all HLA-A,B,C antigen molecules (W6/32 and BB7.7). On the other hand, target cell lysis was unaffected by equivalent concentrations of control monoclonal antibodies binding to the target cell surface through a framework determinant on all HLA-DRw molecules (TDR31.1) or through an EB virusassociated cell surface antigen (MHM6) or through a lymphoblastoid cell surface antigen (AC2). This same result was obtained in further experiments with three other effector T-cell populations produced by HLA-homozygous LCL stimulation.

DISCUSSION From the initial set of experiments it became clear that an EB virus-specific Tcell response, exactly as induced by autologous LCL stimulation, could be reactivated by challenging responder UM cells from a seropositive donor with allogeneic LCL cells from unrelated but HLA-A and B antigen-identical donors (Fig. 1), always providing that the correct in vitro stimulation procedure was followed in order to minimize any accompanying NK-like response. The two relevant combinations (CG ~ JS and SG -~ MO) which were tested in this way gave clear evidence of an EB virus-specific response. In separate experiments one of the reciprocal combinations US --~ CG) was tested since the responder donor JS is seronegative and has no detectable EB virus-specific T-cell memory [ 1,5,1 l]; this served as a useful control and no virus-specific response was obtained (unpublished data). It is interesting to note from these initial experiments that the EB virus-specific cytotoxic response was obtained in HLA-A and B antigen-matched combinations irrespective o f whether matching occurred at the H L A - D R locus (CG --~ JS, single HLA-DR3 match; SG --~ MO, H L A - D R mismatch). There is considerable evidence in both man and mice that whereas the class I histocompatibility products (HLA-A,B,C in man) restrict the cytotoxic effector function [22,23], the class II histocompatibility products (HLA-DR antigens in man) restrict the so-called helper T-cells [22,24,25]. However, the results obtained in this study do not necessarily imply that H L A - D R antigen-restricted helper T-cells have no role to play in the response to autologous stimulation; indeed, such cells may well be important (I. Hart, A.B. Rickinson, and M. Rowe; in preparation), but their contribution might have been met in these unusual circumstances by another population of T cells recognizing foreign HLA-DR antigens on the stimulator cell surface. O t h e r studies in this laboratory have shown that this method of stimulation with LCL cells, first devised as a means of generating virus-specific responses to autologous stimulation [11], will also induce powerful allospecific cytotoxic responses to any foreign HLA-A or B antigens (C. Rooney, unpublished). It was therefore apparent that attempts to reactivate selectively particular clones of El3 virus cytotoxic T cells must of necessity use suitably matched HLA-homozygous LCL as a source of stimulator cells in order to minimize interference from allospecific cytotoxic responses; even so, the possibility of generating such allospecific responses was constantly borne in mind during interpretation of the results. The cytotoxic response to HLA-homozygous LCL cells was initially rather

162

9WHW ["

~ ," ..4

L'LS~

~I9M

2~

IoJ;uoo

o

R

o

~ 0 ~

~ ~

o

'~

slIs3 &a6JD; q3l-0r3 ~0

~ o

stsA! 3!j!oads a6o~uao~ad

~EM 0 O0

U ~',

ZDV ~ . ~

~ :~

9WHW

o °.~ =~ Z'L~ ~19~

S [0J~u03 !

!

o

sIIa3 ~a6~D~ qOq-MS #0 S!SAI 3!J!OadS aSD~UaOJad

HLA-restricted EB Virus-specific Cytotoxicity

163

weak in a number of combinations, but increased in potency when subsequently restimulated in the presence of TCGF (Fig. 2). This is reminiscent of the type of response that seropositive donors with low numbers of EB virus-specific memory T cells mount to autologous LCL stimulation, and in such cases expansion of the effector cell population with TCGF is likewise necessary in order to critically examine the cytotoxic response [8]. Extensive tests were carried out with the TCGF-dependent T-cell lines established from cocultures of HLA-heterozygous responder UM cells and HLAhomozygous LCL stimulator cells in order to verify the EB virus-specific nature of their cytotoxicity. Thus, as exemplified in Figures 3 and 4, lyric activity was apparent not just against the stimulator LCL but also against both the autologous LCL and against those allogeneic targets sharing HLA-A and B antigens with the stimulator cells. In contrast, there was no significant killing of EB virus genomenegative cell lines, such as HSB2, K562, and EB4, known to be extremely sensitive to the nonspecific activated NK-like reactivities which can be induced by other methods of autologous or allogeneic LCL stimulation in vitro [5-7]. Moreover, there was no detectable reactivity against allodeterminants on the stimulator cells since mitogen-stimulated UM cells from the same homozygous donors were consistently unrecognized in these assays. Lysis of the stimulator LCL was inhibited by monoclonal antibodies to HLA-A,B,C framework determinants (but not by equivalent concentrations of monoclonal antibodies to other cell-surface molecules, including HLA-DR antigens) exactly as was lysis of the autologous LCL (Fig. 6). Cold target competition experiments indicated that these two targets were being recognized and killed by the same effector cells (Fig. 5). These results, taken overall, strongly support the view that the cytotoxic response to HLA-homozygous LCL stimulation is dominated by EB virus-specific components and is, in general, not seriously contaminated with allospecific or other reactivities. Occasionally, with certain combinations of responder and stimulator cells, a non-virus-specific response did occur and this was immediately apparent from the unusual pattern of results obtained, in particular by the lack of any reactivity against the autologous LCL. The data presented here clearly demonstrate the potential of this new procedure for the selective reactivation of particular components of the EB virusspecific memory T-cell pool in peripheral blood. To take a single example, the response of donor SW to autologous LCL stimulation is routinely dominated by HLA-A11- and HLA-Bw35-restricted components [5,8]; now, using HLA-homozygous LCL cells as the stimulus, the HLA-B7-restricted (but, to date, not the Aw24-restricted) precursor pool can be reactivated in the absence of these other activities (Fig. 3b). Such results provide a clear verification of the view that individual virus-specific cytotoxic T-cell precursors in peripheral blood are restricted through a single HLA-determinant and that the overall virus-specific cytotoxic response is, therefore, a composite of these particular reactivities [4]. These results also bear on another, as yet unresolved, question concerning the number of restricting determinants displayed by any one HLA-A or B antigen [26]. It is perhaps significant that, even in the absence of demonstrable alloreactivity, the cytotoxic preparations induced by HLA-homozygous LCL stimulation lysed the stimulator LCL (and certain other allogeneic LCL with relevant HLA matches) more than the autologous line. Considering the specific example in Figure 3b, if the serologically defined HLA-B7 antigen were in fact a heterogenous group of related molecules with a variety of restriction sites, not all of which were common to all members of the group [26], then in vitro stimulation of SW responder cells with VYF LCL cells might selectively induce only that fraction of the donor's range of HLA-B7-restricted reactivities that efficiently

164

M. Rowe et al. recognize the viral-induced antigen L Y D M A as presented in the context of HLAB7 on VYF LCL cells. This in turn might explain the efficiency with which the reactivated population will subsequently lyse the stimulating LCL. The experimental approaches described in the present work, allied with clonal analysis of the effector populations thus produced [4], should provide valuable cellular probes for the further investigation of HLA antigen polymorphism as it is reflected in the functional restriction of cytotoxic T cells.

ACKNOWLEDGMENTS

This work was supported by the Cancer Research Campaign, London, out of funds donated by the Bradbury Investment Company of Hong Kong. The authors are particularly indebted to V. Stinchcombe for performing the HLA typing tests and to Mrs. E. Gough, Miss A. Green, Miss J. Sweet, and Miss P. Tuck for invaluable technical help.

REFERENCES 1. Rickinson AB, Moss DJ, Wallace LE, Rowe M, Misko IS, Epstein MA, Pope JH: Long-term T-cell-mediated immunity to Epstein-Barr virus. Cancer Res 41:4216, 1981. 2. Misko IS, Moss DJ, Pope JH: HLA antigen-related restriction of T lymphocyte cytotoxicity to Epstein Barr virus. Proc Natl Acad Sci (Wash) 77:4247, 1980. 3. Moss DJ, Wallace LE, Rickinson AB, Epstein MA: Cytotoxic T cell recognition of Epstein Barr virus-infected B cells. I specificity and HLA restriction of effector cells reactivated in vitro. Eur J Immunol 11:686, 1981. 4. Wallace LE, Rickinson AB, Rowe M, Epstein MA: Epstein-Barr virus-specific cytotoxic T-cell clones restricted through a single HLA antigen. Nature 297:413, 1982. 5. Wallace LE, Rickinson AB, Rowe M, Moss DJ, Allen DJ, Epstein MA: Stimulation of human lymphocytes with irradiated cells of the autologous Epstein-Barr virustransformed cell line. I Virus-specific and non-specific components. Cell Immunol 67:129, 1982. 6. Svedmyr EA, Deinhardt F, Klein G: Sensitivity of different target cells to the killing action of peripheral lymphocytes stimulated by autologous lymphoblastoid cell lines. IntJ Cancer 13:891, 1974. 7. Jondal M, Targan S: In vitro induction of cytotoxic effector cells with spontaneous killer cell specificity. J Exp Med 147:1621, 1978. 8. Wallace LE, Rowe M, Gaston JSH, Rickinson AB, Epstein MA: Cytotoxic recognition of Epstein-Barr virus-infected B cells. III Establishment of HLA-restricted cytotoxic T cell lines using 1L-2. Eur J Immunol (in press). 9. Crawford DH, Rickinson AB, Finerty S and Epstein MA: Epstein-Barr (EB) virus genome-containing, EB nuclear antigen-negative B-lymphocyte populations in blood in acute infectious mononucleosis. J Gen Virol 38:449, 1978. 10. Moss DJ, Rickinson AB, Pope JH: Long-term T cell-mediated immunity to EpsteinBarr virus in man. I complete regression of virus-induced transformation in cultures of seropositive donor leukocytes. Int J Cancer 22:662, 1978. 11. Rickinson AB, Moss DJ, Allen DJ, Wallace LE, Rowe M, Epstein MA: Reactivation of Epstein-Barr virus-specific cytotoxic T cells by in vitro stimulation with the autologous lymphoblastoid cell line. Int J Cancer 27:593, 198l.

HLA-restricted EB Virus-specific Cytotoxicity

165

12. Stinchcombe V, Jones T, Bradley BA: Applications for a reliable technique for tissue typing transformed lymphoblastoid cell lines (submitted). 13. Inouye H, Hank JA, Alter BJ, Bach F: TCGF production for cloning and growth of functional human T lymphocytes. Scand J Immunol 12:149, 1980. 14. Krishan A, Raychaudhuri R: Chromosome studies of cell lines and tumors derived from a single specimen of human leukemic blood by cell culture and heterotransplanration. Cancer Res 30:2012, 1970. 15. Lozzio CB, Lozzio BB: Human chronic myelogenous leukemia cell line with positive Philadelphia chromosome. Blood 45:321, 1975. 16. Epstein MA, Barr YM, Achong BG: Preliminary observations on new lymphoblast strains (EB4, EB5) from Burkitt tumours in a British and Ugandan patient. Br J Cancer 20:475, 1966. 17. Wallace LE, Moss DM, Rickinson AB, McMichael AJ, Epstein MA: Cytotoxic T cell recognition of Epstein-Barr virus-infected B cells. II. Blocking studies with monoclonal antibodies to HLA determinants. Eur J Immunol 11:694, 1981. 18. Rowe M, Hildreth JEK, Rickinson AB, Epstein MA: Monoclonal antibodies to Epstein-Barr virus-induced, transformation-associated cell surface antigens: Binding patterns and effect upon virus-specific T-cell cytotoxicity. Int J Cancer 29:373, 1982. 19. Barnstable CJ, Bodmer WF, Brown G, Galfr~ G, Milstein C, Williams AF, Ziegler A: Production of monoclonal antibodies to group A erythrocytes, HLA and other human cell surface antigens--New tools for genetic analysis. Cell 14:9, 1978. 20. Brodsky FM, Parham P, Barnstable CJ, Crumpton MJ, Bodmer WF: Monoclonal antibodies for analysis of the HLA system. Immunol Rev 47:3, 1979. 21. De Kretser TA, Crumpton MJ, Bodmer JG, Bodmer WF: Demonstration of two distinct light chains in HLA-DR-associated antigens by two dimensional gel electrophoresis. EurJ Immunol 12:214, 1982. 22. Klein J: An attempt at an interpretation of the mouse H2 complex. Contemp Topics Immunobiol 5:297, 1976. 23. Zinkernagel RM, Doherty PC: MHC-restricted cytotoxic T-cells: Studies on the biological role of polymorphic major transplantation antigens determining T-cell restriction-specificity, function, and responsiveness. Adv Immunol 27:51, 1979. 24. Katz DH, Graves M, Dorf ME, Dimuzio H, Benacerref B: Cell interactions between histocompatible T and B lymphocytes. J Exp Med 141:263, 1975. 25. Bergholtz BO, Thorsky E: HLA-D restriction of the macrophage-dependent response of human T lymphocytes to PPD in vitro: Inhibition by anti HLA-DR antisera. Stand J Immunol 8:63, 1978. 26. Biddison WE, Shearer GM, Shaw S: Influenza virus-specific cytotoxic T cells are restricted by multiple HLA-A3-related self antigens: Evidence for recognition of distinct self structures in conjunction with different foreign antigens. J Immunol 127:2231, 1981.