- Email: [email protected]
Characterization of human cardiac infiltrating cells posttransplantation: II. CD4+ cloned T-cell lines with “anti-idiotype”-like reactivity
Characterization of Human Cardiac Infiltrating Cells Posttransplantation: II. CD4 + Cloned T-Cell Lines with "Anti-Idiotype"-Like Reactivity Y. C. Wang, K. Kanter, O. Lattouf, G. E. Rodey, K. W. Sell, and A. A. Ansari
A B S T R A C T : CD4 + T cells were cultured from posttransplant cardiac biopsies placed on irradiated feeder
cells of autologous cloned donor major histocompatibility complex class II-specific T-cell lines cultured and grown from previous biopsies. Fourteen of the CD4 + T-cell cultures were expanded and cloned using the same feeder cells. Two of the 14 cloned T-cell lines were examined in detail for their ability to proliferate in vitro. Clones 7E4 and 8G2 proliferated (as determined by primed lymphocyte testing) only when cocultured with a series of distinct autologous cloned T-cell lines from previous biopsies that were specificfor donor-specific HLA-DR3 and HLA-DR 4, respectively. In addition, when HLA-DR-specific T-cell lines were established using recipient peripheral blood mononuclear cells and a series of HLA-DR-expressing homozygous typing cells, clone 7E4 only responded to the series of distinct HLA-DR3-specific autologous cloned T-cell lines but not to autologous HLA-DR2 and -DR4, and clone 8G2 responded to 3 of 8 distinct autologous HLADR4-specific T-cell lines, but not HLA-DR2-specific T-cell lines. These data demonstrate that cardiac biopsies contain CD4 + T cells of recipient origin which show anti-idiotype-like reactivity against T-cell receptors specificfor donor-specific major histocompatibility complex class II molecules. ABBREVIATIONS
CTL dpt HTC IBFC IL-2 MHC
cytotoxic T lymphocyte days post transplant homozygous typing cells idiotype-bearing feeder cells interleukin 2 major histocompatibility complex
MLR NFC PBMC PLT
mixed lymphocyte reaction normal feeder cells peripheralblood mononuclear cells primed lymphocyte testing
INTRODUCTION In vitro culture of cardiac biopsies with media containing interleukin 2 (IL-2) allows for the exudation, growth, culture, and expansion of host-derived mononuclear cells infiltrating donor tissue posttransplantation [1-5]. Although these
From the Department of Pathology (Y.C.W.; G.E.R.; K.W.S.; A.A.A.) and the Department of Surgery (K.K.; O.L.), Emory University School of Medicine, Atlanta, Georgia. Address reprint requests to Dr. A. A. Ansari, Rm. 5008, Winship Cancer Center, Emory University School of Medicine, Atlanta, GA 30322. Received December 1, 1989; acceptedJanuary 15, 1990.
Human Immunology 28, 141-152 (1990) © American Society for Histocompatibility and Immunogenetics, 1990
141 0198-8859/90/$3 50
142
Y.C. Wang et al. TABLE 1 Relative frequency of isolation of mononuclear cells from sequential cardiac biopsies as a function of time posttransplantation Days posttransplant
Number of biopsies put into culture
No. yielding positive mononuclear cell cultures (%)
<120 121-360 >361 Total
768 589 541 1898
446 (58.1) 159(27.0) 83 (15.3) 688 (36.2)
recipient origin cells may represent biased populations of cells that are readily cultured from such biopsies and are selected for growth in IL-2, they still provide a unique opportunity to examine the function and phenotype of these cells, to characterize epitopes recognized by such cells, and to correlate these characteristics with clinical and histologic evidence of allograft rejection. Several studies, including ours, indicate that the infiltrates comprise predominantly CD4 + and CD8 ÷ T cells which contain clones that demonstrate donor-specific mixed lymphocyte reaction (MLR) and cytotoxic T-lymphocyte (CTL) activity [1-5]. Initially, we attempted to correlate the phenotype and function of these cells cultured from biopsies with the expression of major histocompatibility complex (MHC) class I and class II antigens by cardiac myocytes and endothelial cells of the same biopsy [6,7]. Normal cardiac myocytes do not express detectable levels of MHC class II antigens and relatively low levels of MHC class I antigens [8-10]. However, posttransplantation there is a marked increase in MHC antigen expression by both cardiac myocytes and endothelial cells [11-14]. Of interest was our finding that such increases in MHC expression preceded infiltration of mononuclear cells and the histopathologic evidence of rejection [7]. Thus, a radioimmunoassay for quantitation of MHC antigen expression was set up which may be of diagnostic value for the prediction of the onset of allograft rejection episodes [7]. During the course of these studies, we observed that success in obtaining mononuclear cell cultures from posttransplant cardiac biopsies gradually decreases with time (see Table 1) even though comparable numbers of mononuclear cells were present in the biopsy specimens. It was reasoned that our failure to obtain success in culturing mononuclear cells from biopsies late posttransplantation was perhaps due to absence of appropriate costimulator molecules [15] and/or other cellular signals necessary for the propagation of these cells. To test this possibility, aliquots of biopsies were either placed on feeder layers of irradiated recipient peripheral blood mononuclear cells (PBMC) or irradiated cells cloned from the same patient from earlier cardiac biopsies that demonstrate donor-specific MHC class II reactivity. Placement of cardiac biopsies on autologous irradiated PBMC feeder layers increased the frequency of isolation of cultures from the mononuclear cell infiltrates present in the biopsy specimen. However, of interest was our finding of a total of 14 CD4 + T-cell lines that were cultured on feeder layers of cloned T-cell lines from previous biopsies of the same patient. These 14 CD4 ÷ T-cell lines demonstrated unusual patterns of reactivity in a proliferation assay. The reactivity of these 14 CD4 + T-cell lines was examined in detail. The data obtained strongly suggest that they react in an "anti-idiotype"-like fashion. Results of these studies constitute the basis of this report.
Anti-Idiotype Reactivity of CD4 + T Cells
143
MATERIALS AND METHODS Cardiac biopsies. Transvenous endomyocardial biopsies were obtained from patients post cardiac transplantation at regularly scheduled intervals or as clinically indicated. An aliquot of the biopsies was transferred to test tubes containing sterile medium and transported to the laboratory. Other aliquots were submitted to the pathology service for routine histologic grading of rejection by a slightly modified criteria as described previously [ 16,17].
Cardiac biopsy cultures. The routine culture, growth, expansion, and subsequent cloning to derive T-cell lines from cardiac biopsies was performed as previously described with some modifications [4]. Cardiac biopsies were cut into four small pieces, and two were placed on irradiated (5000 rad, Gamma Cell 40, AEC Canada, Ltd.) PBMC from recipients [normal feeder cells (NFC)], while two were placed on irradiated (same dose) autologous cloned T-cell lines [idiotype-bearing feeder cells (IBFC)] derived from previous biopsies. We used IBFC which had been previously characterized to react specifically with donor MHC class II antigens. Each biopsy specimen was placed on 2 × 106 feeder cells in each well of a 24-well Costar (Cambridge, MA) plate. After the mononuclear cells exuding from the biopsy reached confluency, the cells were expanded in T-30 flasks. The cells were propagated in medium containing 1000 recombinant IL-2 U/ml (a gift from Hoffman-LaRoche, Nutley, NJ) and charged once every other week with irradiated feeder cells at a ratio of 10 cells/irradiated feeder cell. The cells were cloned by limiting dilution analysis as previously described, using appropriate feeder cells (NFC or IBFC).
Phenotypic analysis. The bulk cultures and the cloned T-cell lines were analyzed for cell surface markers using a battery of monoclonal reagents and the FACSSTAR (Becton Dickinson, Mountain View, CA) according to procedures established in our laboratory [4].
Functionalassays. Each cloned T-cell line was assayed for donor-specific MLR and CTL activity by the primed lymphocyte test (PLT) assay and a 51Cr-labeled target cell assay, as previously described [4]. HLA typing. HLA typing for HLA-A, -B, -C, -D, -DR, -DP, -DQ specificities of donor and recipient PBMC was performed by the HLA typing laboratory of the Emory University Hospital using standard typing protocols [18]. In addition, the HLA-DR specificity of the cloned T-cell lines was determined using a panel of HLA-DR homozygous typing cells (HTC) obtained from the International HTC Typing Workshop [19].
Medium. Medium used for cell isolation and functional assays consisted of RPM11640 supplemented with 1000 U/ml of penicillin, streptomycin, 20 mM L-glutamine, and 10% heat inactivated fetal calf serum (all from Gibco, Long Island,NY). For culturing mononuclear cells from cardiac biopsies and maintaining bulk and/
144
Y.C. Wang et al.
80
7O
-I-
60
0 O.
50
g >. 0 Z u.l 0 u.i ~ u.
40
ao
20
10
5-120
121-360
361
+
D A Y S POST T R A N S P L A N T
FIGURE 1 Frequency of the predominant phenotype of mononuclear cells (CD3+, CD4 +, CD8-, II; CD3 +, CD4-, CD8 +, l ; CD3 +, CD4 +, CD8 + , I ; and CD3 + , CD4-, CD8-, VA)cultured from human cardiac biopsies at varying time intervals posttransplantation as determined by flow microfluorometry. A total of 446, 159, and 83 mononuclear cell cultures were obtained from cardiac biopsies from patients between 5 and 120 days, between 121 and 360 days, and >361 days posttransplant, respectively.
or cloned T-cell lines, the medium was supplemented with 1000 rIL-2 U/ml (a gift from Hoffman-LaRoche, Nutley, NJ). RESULTS F r e q u e n c y of Isolation and P h e n o t y p i c and Functional Analysis of M o n o n u c l e a r Cells C u l t u r e d from H u m a n Cardiac Biopsies In support of observations by Weber et al. [20], our laboratory data similarly show that the relative success in obtaining mononuclear cell cultures from biopsies decreases as a function of time posttransplantation (Table 1). All bulk cultures obtained from early posttransplant biopsies (5-120 days) have a predominant phenotype consisting of CD3 +, C D 4 - , CD8 + T cells. In late posttransplant biopsies (post-121 days), however, there is a shift to a phenotype predominantly consisting of CD3 +, CD4 +, C D 8 - T ceils. A significant number of CD3 +, CD4 +, CD8 + (dual-positive) and a limited number of CD3 +, C D 4 - , C D 8 - (dual-negative) T-cell cultures have also been isolated (see Figure 1). The dual-positive Tcell cultures were mainly (>85%) cultured from biopsies between 121 and 360 days posttransplant, whereas the eight dual-negati,re T-cell cultures were all obtained from biopsies >361 days posttransplant. Most of the CD4 + and CD8 + T cells demonstrated donor-specific MLR and CTL activity, respectively. By contrast, all dual-positive cultures (36/688) failed to respond in either MLR or CTL assays. Dual-negative cells (8/688) all demonstrated donor M H C class I-specific cytolytic activity and expressed the 7/8 T-cell receptor.
Anti-ldiotype Reactivity of CD4 + T Cells TABLE 2
HLA typing results of cardiac transplant recipients (and donors) from which "anti-idiotype"-like reactive T-cell clones were derived
Cloned T-cell line
HLA type of the recipient
EU86-015-7E4
A30,36,B53,61 BW6,BW6 CW4 DR6,DRW 11,DRW52 DQW1,DQW3 A3,30,B57,BW4,CW4 DR6,7,DRW53 DQW1,2,DPW4
EU88-110-8G2
145
HLA type of the donor A1,2,B8,BW62,CW3 DR3,4,DRW52 DQW2,3
A2,31,B44,BW4,6 DR3,4,DRW52 DQW3
Functional Specificity of T-Cell Clones Generated from Cardiac Biopsies Placed on Irradiated Autologous Feeder Cells We reasoned that our failure to obtain mononuclear cell cultures exuding from cardiac biopsies late posttransplantation (> 121 days) was secondary to requirements for either other cytokines and/or other cellular signals by these cells and not by cardiac infiltrating cells during the early posttransplant period. We therefore cultured aliquots of biopsies (two each) on either autologous NFC or autologous T-cell clones derived from previous cardiac biopsies that were specific for donorspecific MHC class II molecules (IBFC). Aliquots of 57 cardiac biopsies obtained between 121 and 360 days posttransplant (dpt) and 46 biopsies >361 dpt were placed on NFC and IBFC. This strategy resulted in 24 cultures obtained by using NFC and 5 using IBFC from the total of 57 biopsies (121-360 dpt); and 15 cultures obtained by using NFC and 9 using IBFC from the total of 46 biopsies (>361 dpt). These preliminary data indicate an increase from 27 (see Table 1) to 42% for success in obtaining cultures from biopsies 121-360 dpt and from 15.3 to 32% for biopsies from patients >361 dpt with the use of NFC. A high frequency (32/39, 81%) of the mononuclear cell cultures (using NFC) were CD3 +, CD4 +, C D 8 - , the rest being CD3 +, CD4-, CD8 +. All 14 cultures using IBFC were CD3 +, CD4 +, CD8-. Functional studies showed that 27/32 CD4 + mononuclear cells cultured using NFC were specific for donor MHC class II antigens. Of great interest were our findings on the 14 cultures using IBFC. All T-cell cultures failed to react with donor cells but selectively proliferated only when cocultured with the original IBFC. Two CD3 +, CD4 +, CD8- cloned Tcell lines [EU86-015-7E4 (clone 7E4) and EU88-110-8G2 (clone 8G2)] were studied in more detail. Donor and recipient HLA typing results are shown in Table 2. Clone 7E4 was derived from a routine cardiac biopsy of a patient 286 days posttransplant (biopsy 17) that was placed on autologous irradiated feeder cells previously obtained from a cardiac biopsy at 132 days posttransplant. Clone 7E4 was derived by repeated culturing and final cloning by limiting dilution techniques on irradiated feeder cells derived from day-132 posttransplant cardiac biopsy. When tested against donor PBMC, Epstein-Barr virus-transformed cells or a series of cloned T-cell lines from the same patient, clone 7E4 failed to respond in primed lymphocyte testing to donor cells, but proliferated in response to 4 of
146
Y.C. Wang et al. 80000
uJ ui +1
70000
era O
60000
E •
50000
E
40OO0
.£ iP
_,~soooo , "
1
20000
1oooo
~
~. '-'
0
.
.
1
I
.
2
.
.
3
controls
-
I
I
.
.
.
1
2
3
.
.
4
.
-
I
HLA.DR3 specific stimulator cells
.
1
I
.
2
.
3
.
.
4
5
6
HLA.DR4 specific stimulator cells
7
I
FIGURE 2 Proliferativeresponse specificity of cloned T-cell line EU86-015-7E4 as determined by the PLT. Clone 7E4 T cells (2 x 105) were cocultured with 5 x 104 autologous irradiated (A) control cells (control 1 = responder cells alone, control 2 = normal PBMC from donor, and control 3 = Epstein-Barr virus-transformed donor PBMC), (B) a series of four distinct donor HLA-DR3-specificcloned T-cell lines obtained from previous cardiac biopsies (clones EU86-015-3H11,2B6,4B7, and 6C2), (C) a series of seven HLA-DR4-specific cloned T-cell lines from previous cardiac biopsies (clones EU86-015-1A12, 2C9, 2E12, 3B6, 3H1, 5H6, and 6D3).
the 11 cloned T-cell lines obtained from other previous biopsies (see Figure 2). All four cell lines that induced the proliferation of clone 7E4 when used as stimulator cells were cloned T-cell lines specific for donor HLA-DR3. Further testing of clone 7E4 against a panel of nine cloned T-cell lines prepared from recipient PBMC cells against 3 distinct HLA-DR3 expressing homozygous typing cells (HTC), and for control purposes, three cloned T-cell lines from recipient PBMC specific for three distinct HLA-DR2 and three specific for three distinct HLA-DR4 HTC, clearly demonstrated specific proliferative responses against only those specific for HLA-DR3 (see Figure 3). The clone 8G2 was obtained from cardiac biopsy of another patient by placing the biopsy over autologous irradiated feeder cells from a previous biopsy from the same patient. Again, clone 8G2 only proliferated in response to the original IBFC and to three other T-cell lines grown from previous biopsies of the same patient but not six other T-cell lines (Figure 4). All three cloned T-cell lines, when used as stimulator cells that induced positive proliferative responses, were HLADR4 specific. When a series of eight cloned T-cell lines were prepared from recipient PBMC specific for different allogeneic HI.A-DR4 expressing HTC, clone 8G2 only responded to three of the eight clones (see Figure 5). In addition, clone 8G2 failed to proliferate when cocultured with recipient PBMC cloned Tcell lines specific for HLA-DR2 or DR3 (see Figure 5) and also against DR5 or DR8 (data not shown). These data suggest that clone 8G2 selectively proliferates against the HLA-DR4 specific cloned T-cell lines. Since HLA-DR4 actually com-
Anti-Idiotype Reactivity of CD4 + T Cells
u.,I
40000
+l
35000
ui
O
E
147
30000 25000
10
>, =,,.
20000
I-"tL
15000
"
10000
:E 5ooo
0
1
2
3
I controls
- 1 II
2
3
1 2 3 - 1 2 3 II II I H~-D~s~cH~-DR3~cHLA-DR4s~c stimulator cells stimulator cells stimulator cells
FIGURE 3 Specificity of the PLT response of clone 7E4 against a panel of autologous HLA-DR-specific cloned T-cell lines prepared from recipient PBMC using a series of HLA-DR2, -DR3, and -DR4-expressing distinct HTC. The 7E4 responder clone T cells (2 × 105/culture) were cocultured with irradiated stimulator (5 × 104/culture) cells from (A) controls (control 1 = responder cells alone, control 2 = donor PBMC, and control 3 = recipient PBMC), (B) a series of three autologous cloned cell lines specific for HLADR2, three specific for HLA-DR3, and three specific for HLA-DR4, prepared against three distinct HLA-DR2, -DR3, and -DR4-expressing HTC.
prises four distinct gene products, it is not surprising that clone 8G2 proliferated in response to three of the eight HLA-DR4 specific autologous T-cell lines. Specificity was consistent since the tests were repeated at least four times and gave similar results. DISCUSSION Previous data from our laboratory showed that mononuclear cells cultured from cardiac biopsies of patients posttransplantation in media containing IL-2 demonstrate donor-specific alloreactivity [4]. If recipient-derived cells were tested for donor-specific alloreactivity after 2 - 3 weeks of culture, specificity for donor antigens was lost unless cultures were transferred to feeder cell layers of cells. We, therefore, employed the strategy of routinely culturing cardiac biopsy hostderived mononuclear cells (which demonstrate donor-specific MLR or CTL activity) on feeder cells of donor origin (either PBMC, spleen cells, or Epstein-Barr virus-transformed donor PBMC) within 3 weeks of propagation of the biopsyderived cells. This strategy led to the derivation of a large number of T-cell cultures. Donor-specific M H C class I and class II alloreactivity of cloned T-cell lines was maintained by routinely propagating the cells on feeder cells of donor origin [4]. During the course of these studies, we observed that cardiac biopsies from patients late in the posttransplant period failed to yield the initial cellular exudate of mononuclear cell cultures using the same culture conditions, even
148
Y.C. Wang et al. 50000 I,u 45000 4-1 =¢11 o
E
40000
35000
-
30000
-
25000
-
20000
-
>,= ¢-
15000
-
=E
1 0000
-
or .m
E el'-
"6 5000-| o J¢ ¢l
o
I~1~
~..~.1~'~1 . . . . . . 1 I
2
controls
3
1 I
2
,,.~.~,.~.~'~'l?-J~.
3
I I HLA-DR4 specific stimulator cells
1 I
2
3
.
4
5
HLA-DR3 specific stimulator cells
.
6 I
FIGURE 4 Proliferative response specificity of the cloned T-cell line EU88-110-8G2 as determined by PLT. Clone 8G2 T cells (2 x 105) were cocultured with 5 x 104 irradiated stimulator cells from (A) controls (control 1 - responder cells alone, control 2 = donor normal PBMC, and control 3 -- EBV-transformed donor PBMC), (B) a series of three distinct autologous cloned T-cell lines established from previous cardiac biopsies of the same patient that show specific response to donor HLA-DR4, (C) a series of six distinct autologous cloned T-cell lines established from previous cardiac biopsies of the same patient that show specific response to donor HLA-DR3.
when an aliquot of the same biopsy showed histologic evidence of mononuclear cell infiltrates• Histologically, the cellular infiltrates in late transplant biopsies were mostly focal and not diffuse as characteristic of biopsies early during the posttransplant period. We thus reasoned that mononuclear cells in the biopsies late posttransplant perhaps had other requirements such as cytokines, costimulator molecules, and/or cellular signals. Thus, an attempt was made to culture a series of cardiac biopsies directly on irradiated feeder cells of recipient origin. Two sources of recipient origin feeder cells were utilized• One consisted of irradiated PBMC of recipient origin, and the other consisted of recipient irradiated cloned T-cell lines derived from a biopsy at an earlier posttransplant period. Our results confirm previous observations that use of PBMC of recipient origin markedly increases the frequency of isolation of mononuclear cell cultures. Our major new finding is the growth of mononuclear cell infiltrates from biopsies placed directly on recipient-derived cloned T-cell lines from previous cardiac biopsies. Although the frequency of isolation of mononuclear cells placed on autologous cloned Tcell lines was low (5/57 from biopsies 121-360 dpt and 9/46 from biopsies >361 dpt), the functional reactivity of cloned T-cell lines established from such cultures provided some interesting data. Clone 7E4 appeared to react specifically against autologous T-cell clones that were specific for HLA-DR3 but not against T-cell clones derived from other biopsies of the same patient that were specific for HLADR4. To more specifically examine the specificity of clone 7E4, a series of HLADR-reactive cloned T-cell lines was prepared against DR2,3,4-positive HTC
Anti-Idiotype Reactivity of CD4 + T Cells
,,~ u~
35000
+t
30000
~:
25000
¢e
20000 -
:6
149
-
-
15000 -
.~
.I"
~. -,
10000 -
0
.
1
2
I
3
1 2 I
controls
I HLA-DR2
3
1 I
spec
stimulator cells
2
I HLA-DR3
.
.
3 I spec
stimulator cells
1
.
.
2
.
3
.
.
4
.
5
.
6
I
7
8 I
HLA-DR4
specific stimulator cells
FIGURE 5 Specificity of the PLT response of clone EU88-110-8G2 against a panel of autologous HLA-DR-specific cloned T-cell lines prepared from recipient PBMC using a series of HLA-DR2, -DR3, and -DR4-expressing distinct HTC. The 8G2 responder clone T cells (2 × 105/culture) were cocultured with irradiated stimulator cells (5 x 104/culture) from (A) control (control 1 -- responder cell control, control 2 -- donor PBMC, and control 3 = recipient PBMC), (B) a series of three autologous cloned cell lines specific for HLA-DR2, three autologous cloned cell lines specific for HLA-DR3, and eight autologous cloned cell lines specific for HLA-DR4, prepared from PBMC of the patient against three distinct HLA-DR2, 3 di'stinct HLA-DR3, and eight distinct HLA-DR4 homozygous typing cells.
utilizing recipient PBMC. Again, clone 7E4 only reacted with the cloned T-cell lines that were specific for HLA-DR3 but not for HLA-DR2 or -DR4 (see Figure 3). This pattern of reactivity strongly suggested that clone 7E4 was reactive against either the T-cell receptor (idiotype) specific for HLA-DR3 or an as yet unidentified molecule that is only expressed after reactivity with HLA-DR3-expressing cells. This anti-idiotype-like reactivity was not unique to clone 7E4, since clone 8G2 also demonstrates a pattern of reactivity which appears to be anti-idiotype-like (Figure 4). However, it is clear that clone 8G2 reacts with only a subset of HLADR4-reactive T-cell clones. A limited series of experiments with cloned T-cell lines specific for HLA-DR3 and HLA-DR4 established from the PBMC of unrelated individuals was also performed (Figure 5). Results demonstrate a certain degree of cross-reactivity in that clone 7E4 did proliferate against 3 of 12 HLADR3-reactive cloned T-cell lines from unrelated individuals, but 0 of 14 HLADR2 and 0 of 15 HLA-DR4-reactive cloned T-cell lines from unrelated individuals (data not shown). Clone 8G2 similarly reacted with 2 of 10 cloned T-cell lines specific for HLA-DR4 from unrelated individuals but 0 of 21 T-cell lines specific for HLA-DR2, -DR3, -DR5, -DR7, and -DR8 (data not shown). These preliminary data suggest that there must be some shared idiotypy for these or associated HLA class II molecules at the T-cell receptor level. Both clones 7E4 and 8G2 express the c~/j8 T-cell receptor, and it will be of great interest to examine the T-cell
150
Y.C. Wang et al. repertoire (C, D, J, and V gene usage) of not only the clones 7E4 and 8G2 but also of the stimulatory and nonstimulatory cloned T-cell lines. Such studies are currently in progress. In addition, experiments are being conducted to determine whether monoclonal antibodies against monomorphic T-ceU receptors and against MHC class II determinants inhibit this anti-idiotype-like reactivity. Such anti-idiotype T-T-cell reactivity is not without precedence (discussed in detail by Pereira et al. [21]). Thus, the experiments of Simm et al. [22,23], SuciuFoca et al. [24,25], Nagarkati et al. [26], and Kennedy et al. [27] all demonstrate the presence of T-cell receptor networks. The demonstration that (A × B) F1 hybrids become resistant to local graft-versus-host reactions induced by large numbers of parent A T cells but not by T cells from the parent B strain if the F1 was pretreated with small numbers of parent A CD4 + T cells containing A antiB (MHC) precursors is one of the earliest documented examples of the existence of a T-ceU receptor network [28-32]. This resistance was shown to be mediated by CD8 + T cells which not only suppress but also are cytotoxic for A anti-Breactive T cells but not A anti-C- or A anti-D-reactive T cells. Of interest was also their finding that cells from hosts made tolerant to A anti-B T-cell receptors also suppress and are cytotoxic for all other anti-B-reactive T cells prepared from other MHC haplotype donor animals. In our experiments, however, such antiidiotype-reactive T cells were CD4 + and not CD8+; and, although they did demonstrate cross-reactivity against cloned T-cell lines prepared from unrelated individuals which were specific for the same MHC class II determinant, this crossreactivity was not absolute. A molecular basis for the recognition of a T-cell receptor by another T-cell receptor (idiotype) is difficult to envision by traditional mechanisms. Thus, classically T-cell receptors recognize antigens in the context of self-MHC molecules. The T-T-cell interactions such as those proposed by results of our experiments first of all suggest that the reactivity is not secondary to shed T-cell receptors that are processed and presented by responder T-cell source antigen presenting cells (APC) since such classical APC were not introduced in the assay system that we employed. It is possible, however, that T cells can themselves present antigens, as recently shown [33]. More likely, it appears that such linked recognition of T-cell receptor must occur by direct interaction of responder and stimulator T-cell receptors and that such reactivity may have physiologic relevance in controlling allogeneic responses. Almost 15 years ago, Jerne suggested that the T-cell repertoire is created by diversification of progenitor T cells with self-reactivity [34,35]. The hypothesis implied the existence of antiself-MHC class II reactivity, and if T-cell diversity is dependent on the existence of such anti-self-reactivity, it is clear that such self-reactivity must be precisely regulated. Certainly, data from our experiments support this notion. Thus, not only can regulation occur by immunoglobulin idiotype anti-idiotype type reactions, by suppressor T cells, but also by T-cell anti-T-cell idiotype regulation. Within this context, it is important to recognize that the molecular and biochemical basis for such interactions is far from clear. Thus, the variable region of an antiidiotype antibody does not necessarily have to share peptide sequences with the peptide that it is mimicking, as recently discussed by Erlanger [36]. The overall significance of our findings has three imPortant aspects. First, the presence of mononuclear cell infiltrates in cardiac biopsies may not be of pathologic consequence. In fact, it may be quite contrary. Second, the occurrence of such anti-idiotype-like reactive T cells has important clinical therapeutic implications. Finally, the precise definition of the molecular basis for such T-T-cell reactivity will be of great importance for our understanding of the basic mechanisms of self- and alloreactivity.
Anti-Idiotype Reactivity of CD4 + T Cells
151
ACKNOWLEDGMENTS The authors are deeply grateful to the Cardiac Transplantation Unit of the Department of Surgery and the supporting staff of Emory University Hospital for providing us with all the clinical material used in these studies. In addition, the authors are grateful to Ms. Ann Mayne for her technical assistance and Ms. Elaine Bunch for her secretarial assistance. This work was supported by National Institutes of Health grant NIH R01-25566-02.
REFERENCES 1. Mayer TG, Fuller AA, Fuller TC, Lazarovits AI, Boyle LA, Kurnick JT: Immunol 134:288, 1985. 2. Zeevi A, FungJJ, Zerbe TR, Kaufman C, Rabin B, Griffith BP, Hardesty R, Duquesnoy RJ: Transplantation 41:620, 1986. 3. Duquesnoy RJ, Zeevi A, FungJJ, Kaufman C, Zerbe TR, Griffith B, Trento A, Kormos R, Hardesty R: Transplant Proc 19:2560, 1987. 4. Ahmed-Ansari A, Knopf WD, Murphy D, Tadros T, Leatherbury A, Goodroe J, Dempsey C, Sell KW: J Cardiovasc Pathol 2(3):193, 1988. 5. Moreau JF, Vie H, Peyrat MA, Soulillou JP: Transplant Proc 17:810, 1985. 6. Ahmed-Ansari A, Tadros T, KnopfWD, Murphy DA, Hertzler G, Feighan J, Leatherbury A, Sell, KW: Transplantation 45(5):972, 1988. 7. Sell KW, Tadros T, Wang YC, Hertzler G, Knopf WD, Murphy DA, Ahmed-Ansari A: J Heart Transplant 7(6):407, 1988. 8. Daar AS, Fuggle SV, Fabre JW, Ting AF, Morris PJ: Transplantation 38:287, 1984. 9. Daar AS, Fuggle SV, Fabre JW, Ting AF, Morris PJ: Transplantation 38:293, 1984. 10. Ahmed-Ansari A, Leatherbury A, Stone J, Demsey C, Sell KW: Fed Proc 46(3):946, 1987. 11. Rose ML, Coles MI, Griffin RJ, Pomerance A, Yacoub MH: Transplantation 41(6):776, 1986. 12. Milton AD, Fabre JW: J Exp Med 161:98, 1985. 13. Suitters A, Rose M, Khagani A, Yacoub M: Transplant Proc 19:2566, 1987. 14. Suitters A, Rose M, Higgins A, Yacoub MH: Clin Ex Immunol 69:575, 1987. 15. Mueller DL, Jenkins MK, Schwartz RH: Ann Rev Immunol 7:445, 1989. 16. Billingham ME: J Heart Transplant 1:25, 1982. 17. Knopf WD, Murphy DA, Sell KW: Emory U J Med 1:11, 1987. 18. van Rood JJ, van Leeuwen A, Ploem JS: Nature 262:795, 1976. 19. Yang SY, Milford E, H~mmerling U, Dupont B: Dupont B (ed): Immunology of HLA. New York, Springer-Verlag, 1989, pp 11-19. 20. Weber T, Kaufman C, Zeevi A, Zerbe TR, Hardesty RJ, Kormos RH, Griffith BP, Duquesnoy RJ: Transplant Proc 20(2):176, 1988. 21. Pereira P, Bandiera A, Coutinho A, Marcos MA, Toribio M, Martinez-AC: Ann Rev Immunol 7:209, 1989. 22. Simm GK, Augustin AA: Ann NY Acad Sci 418:272, 1983. 23. Simm GK, MacNeil IA, Augustin AA: Immunol Rev 90:49, 1986. 24. Suciu-Foca N, Rohowsky C, Kung P, King DW: J Exp Med 156:283, 1982. 25. Suciu-Foca N, Rohowsky C, Kung P, Levinson A, Nicholson J, Reemstma K, King DW: Transplant Proc 15:784, 1983.
152
Y.C. Wang et al 26. Nagarkati PS, Nagarkati M, Kaplan AM: J Exp Med 162:375, 1985. 27. Kennedy DW, Russo C, Kim YT, Weksler ME: J Exp Med 164:490, 1986. 28. 29. 30. 31. 32. 33. 34. 35. 36.
Woodland RT, Wilson DB: EurJ Immunol 7:136, 1977. BeUgrau DL, Wilson DB: J Exp Med 148:103, 1978. Bellgrau B, Wilson DB: J Exp Med 149:234, 1979. Kimura H, Wilson DB: Nature 308:463, 1984. Kimura H, Pickard A, Wilson DB: J Exp Med 160:652, 1984. Hewitt CRA, Feldman M: J Immunol 142(5):1429, 1989. Jerne NK: Am Instit Pasteur Immunol 125C:373, 1974. Jerne NK: Immunol Rev 79:5, 1984. Erlanger BF: Immunol Today 10(5):151, 1989.
A B S T R A C T : CD4 + T cells were cultured from posttransplant cardiac biopsies placed on irradiated feeder
cells of autologous cloned donor major histocompatibility complex class II-specific T-cell lines cultured and grown from previous biopsies. Fourteen of the CD4 + T-cell cultures were expanded and cloned using the same feeder cells. Two of the 14 cloned T-cell lines were examined in detail for their ability to proliferate in vitro. Clones 7E4 and 8G2 proliferated (as determined by primed lymphocyte testing) only when cocultured with a series of distinct autologous cloned T-cell lines from previous biopsies that were specificfor donor-specific HLA-DR3 and HLA-DR 4, respectively. In addition, when HLA-DR-specific T-cell lines were established using recipient peripheral blood mononuclear cells and a series of HLA-DR-expressing homozygous typing cells, clone 7E4 only responded to the series of distinct HLA-DR3-specific autologous cloned T-cell lines but not to autologous HLA-DR2 and -DR4, and clone 8G2 responded to 3 of 8 distinct autologous HLADR4-specific T-cell lines, but not HLA-DR2-specific T-cell lines. These data demonstrate that cardiac biopsies contain CD4 + T cells of recipient origin which show anti-idiotype-like reactivity against T-cell receptors specificfor donor-specific major histocompatibility complex class II molecules. ABBREVIATIONS
CTL dpt HTC IBFC IL-2 MHC
cytotoxic T lymphocyte days post transplant homozygous typing cells idiotype-bearing feeder cells interleukin 2 major histocompatibility complex
MLR NFC PBMC PLT
mixed lymphocyte reaction normal feeder cells peripheralblood mononuclear cells primed lymphocyte testing
INTRODUCTION In vitro culture of cardiac biopsies with media containing interleukin 2 (IL-2) allows for the exudation, growth, culture, and expansion of host-derived mononuclear cells infiltrating donor tissue posttransplantation [1-5]. Although these
From the Department of Pathology (Y.C.W.; G.E.R.; K.W.S.; A.A.A.) and the Department of Surgery (K.K.; O.L.), Emory University School of Medicine, Atlanta, Georgia. Address reprint requests to Dr. A. A. Ansari, Rm. 5008, Winship Cancer Center, Emory University School of Medicine, Atlanta, GA 30322. Received December 1, 1989; acceptedJanuary 15, 1990.
Human Immunology 28, 141-152 (1990) © American Society for Histocompatibility and Immunogenetics, 1990
141 0198-8859/90/$3 50
142
Y.C. Wang et al. TABLE 1 Relative frequency of isolation of mononuclear cells from sequential cardiac biopsies as a function of time posttransplantation Days posttransplant
Number of biopsies put into culture
No. yielding positive mononuclear cell cultures (%)
<120 121-360 >361 Total
768 589 541 1898
446 (58.1) 159(27.0) 83 (15.3) 688 (36.2)
recipient origin cells may represent biased populations of cells that are readily cultured from such biopsies and are selected for growth in IL-2, they still provide a unique opportunity to examine the function and phenotype of these cells, to characterize epitopes recognized by such cells, and to correlate these characteristics with clinical and histologic evidence of allograft rejection. Several studies, including ours, indicate that the infiltrates comprise predominantly CD4 + and CD8 ÷ T cells which contain clones that demonstrate donor-specific mixed lymphocyte reaction (MLR) and cytotoxic T-lymphocyte (CTL) activity [1-5]. Initially, we attempted to correlate the phenotype and function of these cells cultured from biopsies with the expression of major histocompatibility complex (MHC) class I and class II antigens by cardiac myocytes and endothelial cells of the same biopsy [6,7]. Normal cardiac myocytes do not express detectable levels of MHC class II antigens and relatively low levels of MHC class I antigens [8-10]. However, posttransplantation there is a marked increase in MHC antigen expression by both cardiac myocytes and endothelial cells [11-14]. Of interest was our finding that such increases in MHC expression preceded infiltration of mononuclear cells and the histopathologic evidence of rejection [7]. Thus, a radioimmunoassay for quantitation of MHC antigen expression was set up which may be of diagnostic value for the prediction of the onset of allograft rejection episodes [7]. During the course of these studies, we observed that success in obtaining mononuclear cell cultures from posttransplant cardiac biopsies gradually decreases with time (see Table 1) even though comparable numbers of mononuclear cells were present in the biopsy specimens. It was reasoned that our failure to obtain success in culturing mononuclear cells from biopsies late posttransplantation was perhaps due to absence of appropriate costimulator molecules [15] and/or other cellular signals necessary for the propagation of these cells. To test this possibility, aliquots of biopsies were either placed on feeder layers of irradiated recipient peripheral blood mononuclear cells (PBMC) or irradiated cells cloned from the same patient from earlier cardiac biopsies that demonstrate donor-specific MHC class II reactivity. Placement of cardiac biopsies on autologous irradiated PBMC feeder layers increased the frequency of isolation of cultures from the mononuclear cell infiltrates present in the biopsy specimen. However, of interest was our finding of a total of 14 CD4 + T-cell lines that were cultured on feeder layers of cloned T-cell lines from previous biopsies of the same patient. These 14 CD4 ÷ T-cell lines demonstrated unusual patterns of reactivity in a proliferation assay. The reactivity of these 14 CD4 + T-cell lines was examined in detail. The data obtained strongly suggest that they react in an "anti-idiotype"-like fashion. Results of these studies constitute the basis of this report.
Anti-Idiotype Reactivity of CD4 + T Cells
143
MATERIALS AND METHODS Cardiac biopsies. Transvenous endomyocardial biopsies were obtained from patients post cardiac transplantation at regularly scheduled intervals or as clinically indicated. An aliquot of the biopsies was transferred to test tubes containing sterile medium and transported to the laboratory. Other aliquots were submitted to the pathology service for routine histologic grading of rejection by a slightly modified criteria as described previously [ 16,17].
Cardiac biopsy cultures. The routine culture, growth, expansion, and subsequent cloning to derive T-cell lines from cardiac biopsies was performed as previously described with some modifications [4]. Cardiac biopsies were cut into four small pieces, and two were placed on irradiated (5000 rad, Gamma Cell 40, AEC Canada, Ltd.) PBMC from recipients [normal feeder cells (NFC)], while two were placed on irradiated (same dose) autologous cloned T-cell lines [idiotype-bearing feeder cells (IBFC)] derived from previous biopsies. We used IBFC which had been previously characterized to react specifically with donor MHC class II antigens. Each biopsy specimen was placed on 2 × 106 feeder cells in each well of a 24-well Costar (Cambridge, MA) plate. After the mononuclear cells exuding from the biopsy reached confluency, the cells were expanded in T-30 flasks. The cells were propagated in medium containing 1000 recombinant IL-2 U/ml (a gift from Hoffman-LaRoche, Nutley, NJ) and charged once every other week with irradiated feeder cells at a ratio of 10 cells/irradiated feeder cell. The cells were cloned by limiting dilution analysis as previously described, using appropriate feeder cells (NFC or IBFC).
Phenotypic analysis. The bulk cultures and the cloned T-cell lines were analyzed for cell surface markers using a battery of monoclonal reagents and the FACSSTAR (Becton Dickinson, Mountain View, CA) according to procedures established in our laboratory [4].
Functionalassays. Each cloned T-cell line was assayed for donor-specific MLR and CTL activity by the primed lymphocyte test (PLT) assay and a 51Cr-labeled target cell assay, as previously described [4]. HLA typing. HLA typing for HLA-A, -B, -C, -D, -DR, -DP, -DQ specificities of donor and recipient PBMC was performed by the HLA typing laboratory of the Emory University Hospital using standard typing protocols [18]. In addition, the HLA-DR specificity of the cloned T-cell lines was determined using a panel of HLA-DR homozygous typing cells (HTC) obtained from the International HTC Typing Workshop [19].
Medium. Medium used for cell isolation and functional assays consisted of RPM11640 supplemented with 1000 U/ml of penicillin, streptomycin, 20 mM L-glutamine, and 10% heat inactivated fetal calf serum (all from Gibco, Long Island,NY). For culturing mononuclear cells from cardiac biopsies and maintaining bulk and/
144
Y.C. Wang et al.
80
7O
-I-
60
0 O.
50
g >. 0 Z u.l 0 u.i ~ u.
40
ao
20
10
5-120
121-360
361
+
D A Y S POST T R A N S P L A N T
FIGURE 1 Frequency of the predominant phenotype of mononuclear cells (CD3+, CD4 +, CD8-, II; CD3 +, CD4-, CD8 +, l ; CD3 +, CD4 +, CD8 + , I ; and CD3 + , CD4-, CD8-, VA)cultured from human cardiac biopsies at varying time intervals posttransplantation as determined by flow microfluorometry. A total of 446, 159, and 83 mononuclear cell cultures were obtained from cardiac biopsies from patients between 5 and 120 days, between 121 and 360 days, and >361 days posttransplant, respectively.
or cloned T-cell lines, the medium was supplemented with 1000 rIL-2 U/ml (a gift from Hoffman-LaRoche, Nutley, NJ). RESULTS F r e q u e n c y of Isolation and P h e n o t y p i c and Functional Analysis of M o n o n u c l e a r Cells C u l t u r e d from H u m a n Cardiac Biopsies In support of observations by Weber et al. [20], our laboratory data similarly show that the relative success in obtaining mononuclear cell cultures from biopsies decreases as a function of time posttransplantation (Table 1). All bulk cultures obtained from early posttransplant biopsies (5-120 days) have a predominant phenotype consisting of CD3 +, C D 4 - , CD8 + T cells. In late posttransplant biopsies (post-121 days), however, there is a shift to a phenotype predominantly consisting of CD3 +, CD4 +, C D 8 - T ceils. A significant number of CD3 +, CD4 +, CD8 + (dual-positive) and a limited number of CD3 +, C D 4 - , C D 8 - (dual-negative) T-cell cultures have also been isolated (see Figure 1). The dual-positive Tcell cultures were mainly (>85%) cultured from biopsies between 121 and 360 days posttransplant, whereas the eight dual-negati,re T-cell cultures were all obtained from biopsies >361 days posttransplant. Most of the CD4 + and CD8 + T cells demonstrated donor-specific MLR and CTL activity, respectively. By contrast, all dual-positive cultures (36/688) failed to respond in either MLR or CTL assays. Dual-negative cells (8/688) all demonstrated donor M H C class I-specific cytolytic activity and expressed the 7/8 T-cell receptor.
Anti-ldiotype Reactivity of CD4 + T Cells TABLE 2
HLA typing results of cardiac transplant recipients (and donors) from which "anti-idiotype"-like reactive T-cell clones were derived
Cloned T-cell line
HLA type of the recipient
EU86-015-7E4
A30,36,B53,61 BW6,BW6 CW4 DR6,DRW 11,DRW52 DQW1,DQW3 A3,30,B57,BW4,CW4 DR6,7,DRW53 DQW1,2,DPW4
EU88-110-8G2
145
HLA type of the donor A1,2,B8,BW62,CW3 DR3,4,DRW52 DQW2,3
A2,31,B44,BW4,6 DR3,4,DRW52 DQW3
Functional Specificity of T-Cell Clones Generated from Cardiac Biopsies Placed on Irradiated Autologous Feeder Cells We reasoned that our failure to obtain mononuclear cell cultures exuding from cardiac biopsies late posttransplantation (> 121 days) was secondary to requirements for either other cytokines and/or other cellular signals by these cells and not by cardiac infiltrating cells during the early posttransplant period. We therefore cultured aliquots of biopsies (two each) on either autologous NFC or autologous T-cell clones derived from previous cardiac biopsies that were specific for donorspecific MHC class II molecules (IBFC). Aliquots of 57 cardiac biopsies obtained between 121 and 360 days posttransplant (dpt) and 46 biopsies >361 dpt were placed on NFC and IBFC. This strategy resulted in 24 cultures obtained by using NFC and 5 using IBFC from the total of 57 biopsies (121-360 dpt); and 15 cultures obtained by using NFC and 9 using IBFC from the total of 46 biopsies (>361 dpt). These preliminary data indicate an increase from 27 (see Table 1) to 42% for success in obtaining cultures from biopsies 121-360 dpt and from 15.3 to 32% for biopsies from patients >361 dpt with the use of NFC. A high frequency (32/39, 81%) of the mononuclear cell cultures (using NFC) were CD3 +, CD4 +, C D 8 - , the rest being CD3 +, CD4-, CD8 +. All 14 cultures using IBFC were CD3 +, CD4 +, CD8-. Functional studies showed that 27/32 CD4 + mononuclear cells cultured using NFC were specific for donor MHC class II antigens. Of great interest were our findings on the 14 cultures using IBFC. All T-cell cultures failed to react with donor cells but selectively proliferated only when cocultured with the original IBFC. Two CD3 +, CD4 +, CD8- cloned Tcell lines [EU86-015-7E4 (clone 7E4) and EU88-110-8G2 (clone 8G2)] were studied in more detail. Donor and recipient HLA typing results are shown in Table 2. Clone 7E4 was derived from a routine cardiac biopsy of a patient 286 days posttransplant (biopsy 17) that was placed on autologous irradiated feeder cells previously obtained from a cardiac biopsy at 132 days posttransplant. Clone 7E4 was derived by repeated culturing and final cloning by limiting dilution techniques on irradiated feeder cells derived from day-132 posttransplant cardiac biopsy. When tested against donor PBMC, Epstein-Barr virus-transformed cells or a series of cloned T-cell lines from the same patient, clone 7E4 failed to respond in primed lymphocyte testing to donor cells, but proliferated in response to 4 of
146
Y.C. Wang et al. 80000
uJ ui +1
70000
era O
60000
E •
50000
E
40OO0
.£ iP
_,~soooo , "
1
20000
1oooo
~
~. '-'
0
.
.
1
I
.
2
.
.
3
controls
-
I
I
.
.
.
1
2
3
.
.
4
.
-
I
HLA.DR3 specific stimulator cells
.
1
I
.
2
.
3
.
.
4
5
6
HLA.DR4 specific stimulator cells
7
I
FIGURE 2 Proliferativeresponse specificity of cloned T-cell line EU86-015-7E4 as determined by the PLT. Clone 7E4 T cells (2 x 105) were cocultured with 5 x 104 autologous irradiated (A) control cells (control 1 = responder cells alone, control 2 = normal PBMC from donor, and control 3 = Epstein-Barr virus-transformed donor PBMC), (B) a series of four distinct donor HLA-DR3-specificcloned T-cell lines obtained from previous cardiac biopsies (clones EU86-015-3H11,2B6,4B7, and 6C2), (C) a series of seven HLA-DR4-specific cloned T-cell lines from previous cardiac biopsies (clones EU86-015-1A12, 2C9, 2E12, 3B6, 3H1, 5H6, and 6D3).
the 11 cloned T-cell lines obtained from other previous biopsies (see Figure 2). All four cell lines that induced the proliferation of clone 7E4 when used as stimulator cells were cloned T-cell lines specific for donor HLA-DR3. Further testing of clone 7E4 against a panel of nine cloned T-cell lines prepared from recipient PBMC cells against 3 distinct HLA-DR3 expressing homozygous typing cells (HTC), and for control purposes, three cloned T-cell lines from recipient PBMC specific for three distinct HLA-DR2 and three specific for three distinct HLA-DR4 HTC, clearly demonstrated specific proliferative responses against only those specific for HLA-DR3 (see Figure 3). The clone 8G2 was obtained from cardiac biopsy of another patient by placing the biopsy over autologous irradiated feeder cells from a previous biopsy from the same patient. Again, clone 8G2 only proliferated in response to the original IBFC and to three other T-cell lines grown from previous biopsies of the same patient but not six other T-cell lines (Figure 4). All three cloned T-cell lines, when used as stimulator cells that induced positive proliferative responses, were HLADR4 specific. When a series of eight cloned T-cell lines were prepared from recipient PBMC specific for different allogeneic HI.A-DR4 expressing HTC, clone 8G2 only responded to three of the eight clones (see Figure 5). In addition, clone 8G2 failed to proliferate when cocultured with recipient PBMC cloned Tcell lines specific for HLA-DR2 or DR3 (see Figure 5) and also against DR5 or DR8 (data not shown). These data suggest that clone 8G2 selectively proliferates against the HLA-DR4 specific cloned T-cell lines. Since HLA-DR4 actually com-
Anti-Idiotype Reactivity of CD4 + T Cells
u.,I
40000
+l
35000
ui
O
E
147
30000 25000
10
>, =,,.
20000
I-"tL
15000
"
10000
:E 5ooo
0
1
2
3
I controls
- 1 II
2
3
1 2 3 - 1 2 3 II II I H~-D~s~cH~-DR3~cHLA-DR4s~c stimulator cells stimulator cells stimulator cells
FIGURE 3 Specificity of the PLT response of clone 7E4 against a panel of autologous HLA-DR-specific cloned T-cell lines prepared from recipient PBMC using a series of HLA-DR2, -DR3, and -DR4-expressing distinct HTC. The 7E4 responder clone T cells (2 × 105/culture) were cocultured with irradiated stimulator (5 × 104/culture) cells from (A) controls (control 1 = responder cells alone, control 2 = donor PBMC, and control 3 = recipient PBMC), (B) a series of three autologous cloned cell lines specific for HLADR2, three specific for HLA-DR3, and three specific for HLA-DR4, prepared against three distinct HLA-DR2, -DR3, and -DR4-expressing HTC.
prises four distinct gene products, it is not surprising that clone 8G2 proliferated in response to three of the eight HLA-DR4 specific autologous T-cell lines. Specificity was consistent since the tests were repeated at least four times and gave similar results. DISCUSSION Previous data from our laboratory showed that mononuclear cells cultured from cardiac biopsies of patients posttransplantation in media containing IL-2 demonstrate donor-specific alloreactivity [4]. If recipient-derived cells were tested for donor-specific alloreactivity after 2 - 3 weeks of culture, specificity for donor antigens was lost unless cultures were transferred to feeder cell layers of cells. We, therefore, employed the strategy of routinely culturing cardiac biopsy hostderived mononuclear cells (which demonstrate donor-specific MLR or CTL activity) on feeder cells of donor origin (either PBMC, spleen cells, or Epstein-Barr virus-transformed donor PBMC) within 3 weeks of propagation of the biopsyderived cells. This strategy led to the derivation of a large number of T-cell cultures. Donor-specific M H C class I and class II alloreactivity of cloned T-cell lines was maintained by routinely propagating the cells on feeder cells of donor origin [4]. During the course of these studies, we observed that cardiac biopsies from patients late in the posttransplant period failed to yield the initial cellular exudate of mononuclear cell cultures using the same culture conditions, even
148
Y.C. Wang et al. 50000 I,u 45000 4-1 =¢11 o
E
40000
35000
-
30000
-
25000
-
20000
-
>,= ¢-
15000
-
=E
1 0000
-
or .m
E el'-
"6 5000-| o J¢ ¢l
o
I~1~
~..~.1~'~1 . . . . . . 1 I
2
controls
3
1 I
2
,,.~.~,.~.~'~'l?-J~.
3
I I HLA-DR4 specific stimulator cells
1 I
2
3
.
4
5
HLA-DR3 specific stimulator cells
.
6 I
FIGURE 4 Proliferative response specificity of the cloned T-cell line EU88-110-8G2 as determined by PLT. Clone 8G2 T cells (2 x 105) were cocultured with 5 x 104 irradiated stimulator cells from (A) controls (control 1 - responder cells alone, control 2 = donor normal PBMC, and control 3 -- EBV-transformed donor PBMC), (B) a series of three distinct autologous cloned T-cell lines established from previous cardiac biopsies of the same patient that show specific response to donor HLA-DR4, (C) a series of six distinct autologous cloned T-cell lines established from previous cardiac biopsies of the same patient that show specific response to donor HLA-DR3.
when an aliquot of the same biopsy showed histologic evidence of mononuclear cell infiltrates• Histologically, the cellular infiltrates in late transplant biopsies were mostly focal and not diffuse as characteristic of biopsies early during the posttransplant period. We thus reasoned that mononuclear cells in the biopsies late posttransplant perhaps had other requirements such as cytokines, costimulator molecules, and/or cellular signals. Thus, an attempt was made to culture a series of cardiac biopsies directly on irradiated feeder cells of recipient origin. Two sources of recipient origin feeder cells were utilized• One consisted of irradiated PBMC of recipient origin, and the other consisted of recipient irradiated cloned T-cell lines derived from a biopsy at an earlier posttransplant period. Our results confirm previous observations that use of PBMC of recipient origin markedly increases the frequency of isolation of mononuclear cell cultures. Our major new finding is the growth of mononuclear cell infiltrates from biopsies placed directly on recipient-derived cloned T-cell lines from previous cardiac biopsies. Although the frequency of isolation of mononuclear cells placed on autologous cloned Tcell lines was low (5/57 from biopsies 121-360 dpt and 9/46 from biopsies >361 dpt), the functional reactivity of cloned T-cell lines established from such cultures provided some interesting data. Clone 7E4 appeared to react specifically against autologous T-cell clones that were specific for HLA-DR3 but not against T-cell clones derived from other biopsies of the same patient that were specific for HLADR4. To more specifically examine the specificity of clone 7E4, a series of HLADR-reactive cloned T-cell lines was prepared against DR2,3,4-positive HTC
Anti-Idiotype Reactivity of CD4 + T Cells
,,~ u~
35000
+t
30000
~:
25000
¢e
20000 -
:6
149
-
-
15000 -
.~
.I"
~. -,
10000 -
0
.
1
2
I
3
1 2 I
controls
I HLA-DR2
3
1 I
spec
stimulator cells
2
I HLA-DR3
.
.
3 I spec
stimulator cells
1
.
.
2
.
3
.
.
4
.
5
.
6
I
7
8 I
HLA-DR4
specific stimulator cells
FIGURE 5 Specificity of the PLT response of clone EU88-110-8G2 against a panel of autologous HLA-DR-specific cloned T-cell lines prepared from recipient PBMC using a series of HLA-DR2, -DR3, and -DR4-expressing distinct HTC. The 8G2 responder clone T cells (2 × 105/culture) were cocultured with irradiated stimulator cells (5 x 104/culture) from (A) control (control 1 -- responder cell control, control 2 -- donor PBMC, and control 3 = recipient PBMC), (B) a series of three autologous cloned cell lines specific for HLA-DR2, three autologous cloned cell lines specific for HLA-DR3, and eight autologous cloned cell lines specific for HLA-DR4, prepared from PBMC of the patient against three distinct HLA-DR2, 3 di'stinct HLA-DR3, and eight distinct HLA-DR4 homozygous typing cells.
utilizing recipient PBMC. Again, clone 7E4 only reacted with the cloned T-cell lines that were specific for HLA-DR3 but not for HLA-DR2 or -DR4 (see Figure 3). This pattern of reactivity strongly suggested that clone 7E4 was reactive against either the T-cell receptor (idiotype) specific for HLA-DR3 or an as yet unidentified molecule that is only expressed after reactivity with HLA-DR3-expressing cells. This anti-idiotype-like reactivity was not unique to clone 7E4, since clone 8G2 also demonstrates a pattern of reactivity which appears to be anti-idiotype-like (Figure 4). However, it is clear that clone 8G2 reacts with only a subset of HLADR4-reactive T-cell clones. A limited series of experiments with cloned T-cell lines specific for HLA-DR3 and HLA-DR4 established from the PBMC of unrelated individuals was also performed (Figure 5). Results demonstrate a certain degree of cross-reactivity in that clone 7E4 did proliferate against 3 of 12 HLADR3-reactive cloned T-cell lines from unrelated individuals, but 0 of 14 HLADR2 and 0 of 15 HLA-DR4-reactive cloned T-cell lines from unrelated individuals (data not shown). Clone 8G2 similarly reacted with 2 of 10 cloned T-cell lines specific for HLA-DR4 from unrelated individuals but 0 of 21 T-cell lines specific for HLA-DR2, -DR3, -DR5, -DR7, and -DR8 (data not shown). These preliminary data suggest that there must be some shared idiotypy for these or associated HLA class II molecules at the T-cell receptor level. Both clones 7E4 and 8G2 express the c~/j8 T-cell receptor, and it will be of great interest to examine the T-cell
150
Y.C. Wang et al. repertoire (C, D, J, and V gene usage) of not only the clones 7E4 and 8G2 but also of the stimulatory and nonstimulatory cloned T-cell lines. Such studies are currently in progress. In addition, experiments are being conducted to determine whether monoclonal antibodies against monomorphic T-ceU receptors and against MHC class II determinants inhibit this anti-idiotype-like reactivity. Such anti-idiotype T-T-cell reactivity is not without precedence (discussed in detail by Pereira et al. [21]). Thus, the experiments of Simm et al. [22,23], SuciuFoca et al. [24,25], Nagarkati et al. [26], and Kennedy et al. [27] all demonstrate the presence of T-cell receptor networks. The demonstration that (A × B) F1 hybrids become resistant to local graft-versus-host reactions induced by large numbers of parent A T cells but not by T cells from the parent B strain if the F1 was pretreated with small numbers of parent A CD4 + T cells containing A antiB (MHC) precursors is one of the earliest documented examples of the existence of a T-ceU receptor network [28-32]. This resistance was shown to be mediated by CD8 + T cells which not only suppress but also are cytotoxic for A anti-Breactive T cells but not A anti-C- or A anti-D-reactive T cells. Of interest was also their finding that cells from hosts made tolerant to A anti-B T-cell receptors also suppress and are cytotoxic for all other anti-B-reactive T cells prepared from other MHC haplotype donor animals. In our experiments, however, such antiidiotype-reactive T cells were CD4 + and not CD8+; and, although they did demonstrate cross-reactivity against cloned T-cell lines prepared from unrelated individuals which were specific for the same MHC class II determinant, this crossreactivity was not absolute. A molecular basis for the recognition of a T-cell receptor by another T-cell receptor (idiotype) is difficult to envision by traditional mechanisms. Thus, classically T-cell receptors recognize antigens in the context of self-MHC molecules. The T-T-cell interactions such as those proposed by results of our experiments first of all suggest that the reactivity is not secondary to shed T-cell receptors that are processed and presented by responder T-cell source antigen presenting cells (APC) since such classical APC were not introduced in the assay system that we employed. It is possible, however, that T cells can themselves present antigens, as recently shown [33]. More likely, it appears that such linked recognition of T-cell receptor must occur by direct interaction of responder and stimulator T-cell receptors and that such reactivity may have physiologic relevance in controlling allogeneic responses. Almost 15 years ago, Jerne suggested that the T-cell repertoire is created by diversification of progenitor T cells with self-reactivity [34,35]. The hypothesis implied the existence of antiself-MHC class II reactivity, and if T-cell diversity is dependent on the existence of such anti-self-reactivity, it is clear that such self-reactivity must be precisely regulated. Certainly, data from our experiments support this notion. Thus, not only can regulation occur by immunoglobulin idiotype anti-idiotype type reactions, by suppressor T cells, but also by T-cell anti-T-cell idiotype regulation. Within this context, it is important to recognize that the molecular and biochemical basis for such interactions is far from clear. Thus, the variable region of an antiidiotype antibody does not necessarily have to share peptide sequences with the peptide that it is mimicking, as recently discussed by Erlanger [36]. The overall significance of our findings has three imPortant aspects. First, the presence of mononuclear cell infiltrates in cardiac biopsies may not be of pathologic consequence. In fact, it may be quite contrary. Second, the occurrence of such anti-idiotype-like reactive T cells has important clinical therapeutic implications. Finally, the precise definition of the molecular basis for such T-T-cell reactivity will be of great importance for our understanding of the basic mechanisms of self- and alloreactivity.
Anti-Idiotype Reactivity of CD4 + T Cells
151
ACKNOWLEDGMENTS The authors are deeply grateful to the Cardiac Transplantation Unit of the Department of Surgery and the supporting staff of Emory University Hospital for providing us with all the clinical material used in these studies. In addition, the authors are grateful to Ms. Ann Mayne for her technical assistance and Ms. Elaine Bunch for her secretarial assistance. This work was supported by National Institutes of Health grant NIH R01-25566-02.
REFERENCES 1. Mayer TG, Fuller AA, Fuller TC, Lazarovits AI, Boyle LA, Kurnick JT: Immunol 134:288, 1985. 2. Zeevi A, FungJJ, Zerbe TR, Kaufman C, Rabin B, Griffith BP, Hardesty R, Duquesnoy RJ: Transplantation 41:620, 1986. 3. Duquesnoy RJ, Zeevi A, FungJJ, Kaufman C, Zerbe TR, Griffith B, Trento A, Kormos R, Hardesty R: Transplant Proc 19:2560, 1987. 4. Ahmed-Ansari A, Knopf WD, Murphy D, Tadros T, Leatherbury A, Goodroe J, Dempsey C, Sell KW: J Cardiovasc Pathol 2(3):193, 1988. 5. Moreau JF, Vie H, Peyrat MA, Soulillou JP: Transplant Proc 17:810, 1985. 6. Ahmed-Ansari A, Tadros T, KnopfWD, Murphy DA, Hertzler G, Feighan J, Leatherbury A, Sell, KW: Transplantation 45(5):972, 1988. 7. Sell KW, Tadros T, Wang YC, Hertzler G, Knopf WD, Murphy DA, Ahmed-Ansari A: J Heart Transplant 7(6):407, 1988. 8. Daar AS, Fuggle SV, Fabre JW, Ting AF, Morris PJ: Transplantation 38:287, 1984. 9. Daar AS, Fuggle SV, Fabre JW, Ting AF, Morris PJ: Transplantation 38:293, 1984. 10. Ahmed-Ansari A, Leatherbury A, Stone J, Demsey C, Sell KW: Fed Proc 46(3):946, 1987. 11. Rose ML, Coles MI, Griffin RJ, Pomerance A, Yacoub MH: Transplantation 41(6):776, 1986. 12. Milton AD, Fabre JW: J Exp Med 161:98, 1985. 13. Suitters A, Rose M, Khagani A, Yacoub M: Transplant Proc 19:2566, 1987. 14. Suitters A, Rose M, Higgins A, Yacoub MH: Clin Ex Immunol 69:575, 1987. 15. Mueller DL, Jenkins MK, Schwartz RH: Ann Rev Immunol 7:445, 1989. 16. Billingham ME: J Heart Transplant 1:25, 1982. 17. Knopf WD, Murphy DA, Sell KW: Emory U J Med 1:11, 1987. 18. van Rood JJ, van Leeuwen A, Ploem JS: Nature 262:795, 1976. 19. Yang SY, Milford E, H~mmerling U, Dupont B: Dupont B (ed): Immunology of HLA. New York, Springer-Verlag, 1989, pp 11-19. 20. Weber T, Kaufman C, Zeevi A, Zerbe TR, Hardesty RJ, Kormos RH, Griffith BP, Duquesnoy RJ: Transplant Proc 20(2):176, 1988. 21. Pereira P, Bandiera A, Coutinho A, Marcos MA, Toribio M, Martinez-AC: Ann Rev Immunol 7:209, 1989. 22. Simm GK, Augustin AA: Ann NY Acad Sci 418:272, 1983. 23. Simm GK, MacNeil IA, Augustin AA: Immunol Rev 90:49, 1986. 24. Suciu-Foca N, Rohowsky C, Kung P, King DW: J Exp Med 156:283, 1982. 25. Suciu-Foca N, Rohowsky C, Kung P, Levinson A, Nicholson J, Reemstma K, King DW: Transplant Proc 15:784, 1983.
152
Y.C. Wang et al 26. Nagarkati PS, Nagarkati M, Kaplan AM: J Exp Med 162:375, 1985. 27. Kennedy DW, Russo C, Kim YT, Weksler ME: J Exp Med 164:490, 1986. 28. 29. 30. 31. 32. 33. 34. 35. 36.
Woodland RT, Wilson DB: EurJ Immunol 7:136, 1977. BeUgrau DL, Wilson DB: J Exp Med 148:103, 1978. Bellgrau B, Wilson DB: J Exp Med 149:234, 1979. Kimura H, Wilson DB: Nature 308:463, 1984. Kimura H, Pickard A, Wilson DB: J Exp Med 160:652, 1984. Hewitt CRA, Feldman M: J Immunol 142(5):1429, 1989. Jerne NK: Am Instit Pasteur Immunol 125C:373, 1974. Jerne NK: Immunol Rev 79:5, 1984. Erlanger BF: Immunol Today 10(5):151, 1989.