- Email: [email protected]
Transplantation tolerance in heart transplant recipients as demonstrated by unresponsiveness in cell-mediated lympholysis
Transplantation Tolerance in Heart Transplant Recipients as Demonstrated by Unresponsiveness in Cell-Mediated Lympholysis Peter L. J. Wijngaard, Henk-Jan Schuurman, Frits H. J. Gmelig Meyling, George Jambroes, and Jan C. C. Borleffs
A B S T R A C T : Transplantation tolerance or-adaptation
to an allograft is associated with unresponsiveness to donor-specific transplantation antigens measured in in vitro cell-mediated lympholysis (CML). We here demonstrate in a longitudinal follow-up that CML nonreactivity develops in seven of ten patients following heart transplantation. The first manifestation of this nonreactivity manifested between 3 and 27 months after transplantation. CML nonreactivity correlated with time after transplantation and the percentage of activated lymphocytes in
peripheral blood. CML nonreactivity was also associated with good graft function, i.e., in condition of nonresponsiveness patients did not manifest acute rejection. The only exception was seen in one patient in whom the immunosuppressive therapy was strongly reduced. A more detailed evaluation of this patient indicated that the underlying mechanism for CML nonreactivity is clonal anergy or active suppression of the alloreactive cells. Human Immunology 34, 167-172 (1992)
ABBREVIATIONS
CML EMB HLA
cell-mediated lympholysis endomyocardial biopsy human leukocyte antigen
MNC rII-2
mononuclear cell recombinant interleukin-2
INTRODUCTION T-lymphocytes play a crucial role in (acute) allograft rejection [1]. T h e y contribute to the rejection process mainly through cytotoxicity and delayed-type hypersensitivity responses [2]. As in most T-cell-mediated responses, recognition of alloantigen is specific and immunologic m e m o r y is generated. It is well known that reactivity to the graft gradually decreases after transplantation, which is reflected by the fact that the doses
From the Heart Transplant Team University Hospital Utrecht, Departments of Clinical Immunology (P.L.J.W.; F.H.J.G.M.) Pathology (H.-J.S.), the Heart-Lung Institute (G.J.), and Internal Medicine (H.-J.S.; J.C.C.B.), University Hospital, Utrecht, The Netherlands. Address reprint requests to P. L.J. Wijngaard, Department of Climcal Immunology, University Hospital, PO Box 85.500, 3508 GA Utrecht, The Netherlands. Received November 21, 1991; acceptedApril 6, 1992. Human Immunology 34, 167-172 (1992) © American Society for Histocompadbility and Immunogenetics, 1992
of immunosuppressive agents can usually be decreased in time. This adaptation to the graft or transplantation tolerance can be explained by various mechanisms. A m o n g these, clonal deletion or clonal anergy of the alloreactive cells, and the active suppression of these cells, are the most favored ones [ 3 - 7 ] . A combination of such mechanisms is also possible. It is thought that clonal deletion occurs mainly during T-cell repertoire generation in the thymus and that clonal anergy and active suppression are the mechanisms for extrathymically induced tolerance [ 8 - 1 0 ] . T h e cell-mediated lympholysis assay (CML) enables the m e a s u r e m e n t of the recipient's cytotoxic potential toward spleen-derived mononuclear cells (MNCs) from the donor [ 11]. In clinical kidney transplantation, tolerance was reflected by nonreactivity in the CML assay 167 0198-8859/92/$5.00
168
P.L.J. Wijngaard et al.
[ 12-15]. This nonreactivity correlated with good renal function and could be abrogated by culturing the effector cells with recombinant interleukin-2 (rI1-2) before performing the CML assay [16]. We studied the phenomenon of transplantation tolerance after heart transplantation. In a longitudinal follow-up, ten heart transplant recipients were tested for their cytotoxicity toward donor and third-party, spleenderived MNCs in the CML assay. The percentages of donor-specific cytotoxicity were correlated with various cytomorphologically and phenotypically defined lymphocyte subsets. We also tested the recipients' cytotoxicity toward donor and third-party spleen-derived MNCs without the allostimulation phase. MATERIALS A N D M E T H O D S
Patients. In a longitudinal follow-up after orthotopic heart transplantation, ten patients were tested for their cytotoxic potential against donor and third-party spleen-derived MNCs. Recipient and donor were matched for ABO blood group antigens, but not for HLA histocompatibility; a negative crossmatch between patients' serum and donor leukocytes was required. In the preparative phase for transplantation, all patients received at least one blood transfusion from an unrelated donor. The standard immunosuppressive regimen consisted of cyclosporine and prednisolone (five patients), or cyclosporine, prednisolone, and azathioprine (five patients). In case of acute rejection (grade 2 or 3, diagnosed by histopathologic assessment of the endomyocardial biopsy [EMB] [17], the patients were treated with a methylprednisolone pulse for 3 days and/ or with rabbit antithymocyte globulin. In Table 1, rele-
TABLE 1 Development of CML nonreactivity (NR) in relation to clinical data Patient
CML NR ~
Rejections b
1 2 3 4 5 6 7 8 9 10
8 -24 5 12 27 --23 21
4 4 0 2 4 3 3 3 1 2
Immunosuppression ~ CsA CsA CsA CsA CsA CsA CsA CsA CsA CsA
pred pred pred pred pred, pred, pred, pred pred, pred,
aza aza aza aza aza
Months after transplantation of first manifestation of CML nonreactivity. b Number of acute rejections encountered during follow-up in CML assay. c Standard immunosuppressive therapy: CsA, cyclosporine A; pred, prednisoIon; and aza, azathioprine.
vant clinical data are summarized in relation to the development of CML nonreactivity in these patients.
Preparation ofcellsamples. Peripheral blood MNCs were isolated from heparinized venous blood samples by Ficoll-Isopaque density-gradient centrifugation directly before and at regular intervals after transplantation. Spleen-derived MNCs from the donor were isolated by mincing a small part of the donor spleen with surgical blades. From this suspension, the MNC fraction was isolated by Ficoll-Isopaque density-gradient centrifugation. By use of the MNC isolates, the recipients were monitored for phenotypically and cytomorphologically defined lymphocyte subsets with the cytoimmunologic monitoring approach, described previously [18, 19]. This technique enables one to diagnose periods of acute rejection or infection after transplantation by determining the number of cytomorphologically activated lymphocytes and large granular lymphocytes, and the percentages of CD3, CD4, CD8, and Leu-7-positive lymphocytes. The remainder of the MNC preparations were resuspended in 10% dimethylsulfoxide (BDH, Poole, England) and 40% fetal calf serum (Life Technologies, Paisley, Scotland) in minimal essential medium (Life Technologies) and stored in liquid nitrogen until tested in the CML assays. CML assay. The standard CML assay [11] was used. Briefly, peripheral blood lymphocytes from the recipient were sensitized for 6 clays in a 1 : 1 ratio with irradiated spleen-derived MNCs from their respective donors or from third-party donors, selected from the pool of transplant donors on the basis of a complete HLA-A, -B, -C, and -DR mismatch with the recipients' donor. Cytotoxicity to phytohemagglutinin-stimulated target cells was determined in a 4-hour 5~Cr-release assay in triplicate and at different effector-target ratios. The specific release was calculated using the following formula specific release _- experimental cpm - spontaneous c p m , 100% maximum cpm - spontaneous cpm Maximum release was measured by lysing the target cells in 2.5% Triton X-100; spontaneous release was measured from target cells cultured in medium only. The standard deviation of the triplicate cultures never exceeded 5%. Nonreactivity was defined as a specific release in the CML of < 15 % at an effector-target ratio of 40 : 1. In addition to cytotoxicity after in vitro allostimulation, we assayed cytotoxicity to donor or third-party spleen-derived MNCs without the phase of prior in vi-
CML Nonreactivity After Heart Transplantation
tro allosensitization. This was done in the 4-hour 51Crrelease assay by culturing the recipient's MNCs directly after thawing from liquid nitrogen with target cells at different effector-target ratios. We also measured cytotoxicity after a 6-day culture period only with rIL-2, following thawing of the cells.
Statistical analysis. Correlation analysis was done with the aid of the computer-assisted statistical program SPSS-PC+version 3.0. The percentage of donor-specific cytotoxicity was used as dependent variable; independent variables were time after transplantation; the percentages of CD3, CD4, CD8, and Leu-7-positive cells; and the percentage of cytomorphologically defined activated lymphocytes and large granular lymphocytes [ 1820]. RESULTS In this study, ten patients were tested longitudinally for their cytotoxic potential against spleen-derived MNCs from their respective donors following heart transplantation. In two patients (designated as open triangles in Fig. 1), donor-directed cytotoxicity before and after transplantation was not found. The other patients demonstrated donor-directed cytotoxicity with values varying between 22% and 79% (Fig. 1A). This cytotoxicity was specifically directed to the donor because there was minimal or no lysis of autologous and third-party target cells (Fig. 1B and C). In seven of these patients, the donor-directed cytotoxicity decreased to values of <15% within 3-27 months after transplantation (Fig. 1A). We considered this 15% value as the cutoff between reactivity and nonreactivity, based on the fact that cytotoxicity to autologous targets cells normally is below this value (Fig. 1B). Based on this cutoff value, one patient (open diamonds in Fig. 1) did not develop CML nonreactivity toward the donor, but only demonstrated a strong decrease in the CML in time from 79% before transplantation to 25% at 36 months after transplantation. We analyzed whether the observed nonreactivity is donor-specific, by allostimulating the recipient's MNCs with HLA-mismatched third-party spleen-derived MNCs and subsequent CML toward these cells. The low yields of recipients' MNCs after isolation permitted this control experiment only for four of the patients that demonstrated donor-directed nonreactivity. The data, expressed as the ratio of donor-directed cytot'oxicity over third-party-directed cytotoxicity, showed a decrease in time after transplantation (Fig. 2). This means that cytotoxicity toward third-party target cells is still present, whereas at the same time the cytotoxicity to-
169
ward donor targets' cells is decreased below values of 15%. During donor-directed nonreactivity in the CML, the cytotoxicity toward the third-party target cells ranged between 21% and 63%. The development of CML nonreactivity coincided with good functioning of the graft. Episodes of acute rejection did not occur during the period of unresponsiveness, with the exception of one patient (+ in Fig. 1A). This patient, a 37-year-old woman, was on maintenance immunosuppressive therapy comprising 300 mg cyclosporine A and 10 mg prednisolone per day, and had developed CML nonreactivity 8 months after transplantation. This remained so at least until 32 months after transplantation. At 16 months after transplantation, the cyclosporine and prednisolone doses were reduced by 25%. One month later, the endomyocardial biopsy showed grade-2 rejection, which was adequately treated with an oral pulse of 200 mg prednisolone for 3 days. The next biopsy specimen, taken 9 days later, showed no signs of rejection (grade 0). The biopsy specimen taken a subsequent week later again manifested grade-2 rejection. This was successfully treated with a pulse of 1000 mg methylprednisolone on each of 3 successive days. Thereafter, acute rejection episodes no longer occurred. The nonresponsiveness in terms of cytotoxic activity in vitro could be abrogated when the recipient's effector cells (106 MNCs/ml)were cultured with 20 U/ml rII-2 for 6 days after the initial stimulation phase. A cytotoxic potential toward the donor of 2 4 % 35% was measured after this preculture. No cytotoxicity toward autologous and third-party target cells was observed. We performed correlation analyses with regard to cytomorphologic and immunologic phenotype of blood MNCs [18, 19] and with time after transplantation using donor-directed cytotoxicity as an independent variable. A significant inverse correlation was found between the percentage of donor-directed cytotoxicity and time after transplantation (r = -0.281, p = 0.014). This is illustrated in Fig. 1. Furthermore, a significant positive correlation was observed with the percentage of lymphocytes with the cytomorphology of activated cells in blood (r = 0.367, p = 0.010) (Table 2). To test whether in vivo alloactivated lymphocytes are capable of exerting cytotoxicity to donor and control autologous and third-party target cells, we tested MNCs of four patients in the CML assay without prior allostimulation. No lysis of donor, autologous, and third-party target cells was observed. Also, when the patient's blood MNCs (106 MNCs/ml) were first cultured with 20 U/ml rII-2 for 6 days before testing the cytotoxic response, no cytotoxicity toward the same target cells was demonstrable (results not shown).
170
P . L . J . Wijngaard et al
A 100
80 >., 4-,
60
X
o o :>, {3
40
2O
0 -1
0
1 2
3
4
6
8
9 101213151718202122232427293031323641 months
after
transplantation
B IO0
80 >, 4-, X 0
60
0 0
40
20
0 -1
0
1 2
3
4
6
8
9 101213151718202122232427293031323641 months
after
transplantation
C
100
80
0 X 0
60
o >, O
40
g
Oi -1
~-~---~___a'--m. ~__~ 0
1 2
3
4
6
~ 8
~
.
,--
~
:
.-"~:- ~
'
t
-
-
.
~
9 101213151718202122232427293031323641 months
after
transplantation
I
I
I
F I G U R E l Development of nonresponsiveness in cellmediated lympholysis (CML) after heart transplantation. The recipients' mononuclear cells (MNCs) were stimulated with donor spleen-derived MNCs and tested in CML assays using the following target cell populations: (A) donor spleen-derived MNCs, (B) autologous blood MNCs, and (C) HLA-mismatched third-party spleenderived MNCs. The 15 % cytotoxicity level is indicated by the dotted line. The patient indicated by + developed an acute rejection after reduction of immunosuppression during CML nonresponsiveness 16 months after transplantation. The patient indicated as © developed donor-directed nonresponsiveness without manifesting any acute rejections in the first 3 years after transplantation.
CML Nonreactivity After Heart Transplantation
TABLE 2
Variable t > n r pd t >
171
CML nonreactivity coincided with good functioning of the graft. Rejection crises were not observed during the presence of CML nonreactivity. This was except for one patient, who, after a 25% reduction in the immunosuppressive therapy, manifested acute rejection grade 2 during the period of nonreactivity. The nonresponsiveness in this patient could be abrogated by culturing the recipient's effector cells after the sensitization phase with exogeneous rI1-2 for another 6 days. This indicates the presence of donor-specific alloreactive (precursor) cytotoxic T cells (CTLs) in the circulation of this patient. Apparently, these cells were rendered anergic or were actively suppressed, but they obviously were not completely deleted. Whether this is the case for all patients developing CML nonreactivity after heart transplantation has to be examined. For kidney transplantation, it has been demonstrated that the donor HLA-antigen inducible precursor CTL frequency gradually decreases in time [21]. At present, we are analyzing precursor CTL frequencies in a longitudinal follow-up in heart transplant recipients in relation to the development of CML nonreactivity. One patient did not manifest any acute rejection episode during the first 3 years after transplantation (Table 1). Donor-directed cytotoxicity dropped from 41% before to 9% two years after transplantation (indicated by open circles in Fig. 1). Apparently, CML nonreactivity develops without strong in vivo alloactivation as envisaged by clinical rejections as diagnosed by EMB histopathology. The percentage of donor-specific cytotoxicity significantly correlated with time after transplantation as was expected, but it also correlated with the percentage of cytomorphologically defined activated lymphocytes in peripheral blood. In previous studies, we have demon-
Correlation coefficients (r) for percentage of donor-specific lysis as independent variable Tx ~
CD3
CD4
CD8
Leu-7
ALb
LGL'
61 23 23 23 21 40 40 -0.281 0.177 0.189 0.340 0.154 0.367 0.065 0.014 0.210 0.194 0.056 0.252 0.010 0.346
Tx, time af[er [ransplantation.
AL, cytomorphologically defined activated lymphocytes. c LGL, cytomorphologically defined large granular lymphocytes. p, level of significance.
DISCUSSION Patients who have received an allograft can develop tolerance to the alloantigens presented in the graft. This in vivo adaptation enables one to decrease the dosage of immunosuppressive agents after transplantation. Transplantation tolerance may be reflected in vitro by the absence of donor-specific cytotoxicity in the CML assay [12-15]. H e r e , we demonstrate the development of CML nonreactivity in seven of eight patients after heart transplantation. These manifested CML reactivity before or in the first 3 months after transplantation, and the CML nonreactivity occurred for the first time between 3 and 27 months after transplantation (Fig. 1). The CML nonreactivity was donor specific because lymphocytes from the recipients were able to lyse thirdparty target cells that were completely HLA-A, -B, -C, and -DR mismatched with donor and recipient (Fig. 2). In two patients, the CML reactivity was not detectable and, therefore, these patients were considered negative for developing CML nonreactivity. FIGURE 2 Nonresponsiveness in cell-mediated lympholysis (CML) assays is donor specific. Recipients' mononuclear cells (MNCs) were stimulated with either donor spleenderived MNCs or with HLA-mismatched thirdparty spleen-derived MNCs and tested in the CML assay with these respective stimulator cells as target cells. The ratio of cytotoxicity to donor over that to third party is plotted against time after transplantation. Data shown is for individual patients, indicated by the same symbols as in Fig. 1.
3.00
o
2.40
~ c~
1.80
£ o 8 x~
1.20
0.60 0.00 -1
0
1
3 months
10 after
17
transplantation
27
31
41
172
strated that the percentages of the cytomorphologically activated lymphocytes are increased during periods of acute rejection [18, 19]. Apparently, this process is also reflected in the percentage of donor-directed cytotoxicity. To test whether these in vivo alloactivated lymphocytes are capable of exerting in vitro cytotoxicity, we tested M N C s in the CML assay without the prior in vitro allosensitization phase. In this condition, the recipients' lymphocytes were unable to lyse donor target cells, even when the lymphocytes were first stimulated with rI1-2 for 6 days. This absence of detectable cytotoxicity in this direct CML assay was also evident in cases with enhanced percentages of lymphocytes with the cytomorphology of activated cells, reflecting the presence of rejection. Thus, cytomorphologically defined activated lymphocytes in peripheral blood during periods of acute rejection reflect the rejection process, but may not be necessarily responsible for the cytotoxic reactions toward the graft as c o m p o n e n t in the rejection process. The mechanisms underlying transplantation tolerance are not completely resolved. In studies in mouse it has been postulated that clonal deletion, clonal anergy, and/or the active suppression of the alloreactive cells may play a role [ 3 - 7 ] . Our data in a single patient (indicated as + in Fig. 1) indicate a role for clonal anergy or active suppression for the development of CML nonreactivity after heart transplantation. Whether this is the more general mechanism has to be further analyzed.
REFERENCES 1. Mason DW: Effector mechanisms in allograft rejection. Annu Rev Immunol 4:119, 1986. 2. Mason DW, Dallman MJ, Arthur RP, Morris PJ: Mechanisms of allograft rejection: The roles of cytotoxic T-cells and delayed-type hypersensitivity. Immunol Rev 77:16, 1984. 3. Kamada N: Transplantation tolerance and immunosuppression following liver grafting in rats. Immunol Today 6:336, 1985. 4. Dallman MJ, Shiho O, Page TH, Wood KJ, Morris PJ: Peripheral tolerance to alloantigen results from altered regulation of the interleukin 2 pathway. J Exp Med 13:79, 1991. 5. Robbins PA, McMichael AJ: Immune recognition of HLA molecules downmodulates CD8 expression on cytotoxic T-lymphocytes. J Exp Med 13:221, 1991. 6. Slavin S: Total lymphoid irradiation. Immunol Today 8:88, 1987.
P.L.J. Wijngaard et al.
7. Silvers WK, Kimura H, Desquenne-Clark L, Miyamoto M: Some new perspectives on transplantation immunity and tolerance. Immunol Today 8:185, 1987. 8. Ramsdell F, Fowlkes BJ: Clonal deletion versus clonal anergy: the role of the thymus in inducing self tolerance. Science 248:1342, 1990. 9. Sprent J, Gao E-K, Webb SR: T-cell reactivity to MHC molecules: immunity versus tolerance. Science 248:135, 1990. 10. MacDonald HR, Lees RK: Programmed death of autoreactive thymocytes. Nature 343:642, 1990. 11. Goulmy E, Termijtelen A, Bradley BA, van Rood JJ: HLA restriction of non-HLA-A, -B, -C and -D cell mediated lympholysis (CML). Tissue Antigens 8:317, 1976. 12. Goulmy E, Persijn G, Blokland E, D'Amaro J, van Rood JJ: Cell-mediated lympholysis studies in renal allograft recipients. Transplantation 31:210, 1981. 13. Van Rood JJ, Claas FHJ: The influence of allogeneic cells on the human T and B cell repertoire. Science 248:1388, 1990. 14. Pfeffer PF, Thorsby E, Hirschberg H: Donor-specific decreased cell-mediated cytotoxicity in recipients of well functioning, one HLA haplotype-mismatched kidney allografts. Transplantation 35:156, 1983. 15. Pfeffer PF, Hirschberg H, Thorsby E: Donor-specific decreased primary and secondary cell-mediated immune responses in patients with well functioning grafts. Transplant Proc 13:1604, 1981. 16. Goulmy E, Blokland E, Persijn G, Paul LC, Wilmink J, van Rood JJ: HLA regulates postrenaI transplant CML nonreactivity. J Immunol 135:3082, 1985. 17. Billingham ME: Diagnosis of cardiac rejection by endomyocardial biopsy. J Heart Transplant 1:25, 1981. 18. Wijngaard PLJ, Heyn A, van der Meulen A, de Vries H: Cyto-immunologic monitoring following heart transplantation. Bibl Cardiol 43:137, 1988. 19. Wijngaard PLJ, Gimpel JA, Schuurman H-J, van der Meulen A, Gmelig Meyling FHJ, Jambroes G: Rejection monitoring after heart transplantation: cyto-immunologic monitoring on blood cells and quantitative birefringence measurements on endomyocardial biopsies. J Clin Pathol 43:137, 1990. 20. Ertel W, Reichenspurner H, Hammer C, Lersch C, Plahl M, Brendel W, Reichart B, Kemkes BM, Reble B, Gokel JM: Cyto-immunologic monitoring: a method to reduce biopsy frequency after cardiac transplantation. Transplant Proc 17:204, 1985. 21. Herzog W-R, Zanker B, Irschick E, Huber C, Franz HE, Wagner H, Kabelitz D: Selective reduction of donorspecific cytotoxic T-lymphocyte precursors in patients with a well functioning kidney allograft. Transplantation 43:384, 1987.
A B S T R A C T : Transplantation tolerance or-adaptation
to an allograft is associated with unresponsiveness to donor-specific transplantation antigens measured in in vitro cell-mediated lympholysis (CML). We here demonstrate in a longitudinal follow-up that CML nonreactivity develops in seven of ten patients following heart transplantation. The first manifestation of this nonreactivity manifested between 3 and 27 months after transplantation. CML nonreactivity correlated with time after transplantation and the percentage of activated lymphocytes in
peripheral blood. CML nonreactivity was also associated with good graft function, i.e., in condition of nonresponsiveness patients did not manifest acute rejection. The only exception was seen in one patient in whom the immunosuppressive therapy was strongly reduced. A more detailed evaluation of this patient indicated that the underlying mechanism for CML nonreactivity is clonal anergy or active suppression of the alloreactive cells. Human Immunology 34, 167-172 (1992)
ABBREVIATIONS
CML EMB HLA
cell-mediated lympholysis endomyocardial biopsy human leukocyte antigen
MNC rII-2
mononuclear cell recombinant interleukin-2
INTRODUCTION T-lymphocytes play a crucial role in (acute) allograft rejection [1]. T h e y contribute to the rejection process mainly through cytotoxicity and delayed-type hypersensitivity responses [2]. As in most T-cell-mediated responses, recognition of alloantigen is specific and immunologic m e m o r y is generated. It is well known that reactivity to the graft gradually decreases after transplantation, which is reflected by the fact that the doses
From the Heart Transplant Team University Hospital Utrecht, Departments of Clinical Immunology (P.L.J.W.; F.H.J.G.M.) Pathology (H.-J.S.), the Heart-Lung Institute (G.J.), and Internal Medicine (H.-J.S.; J.C.C.B.), University Hospital, Utrecht, The Netherlands. Address reprint requests to P. L.J. Wijngaard, Department of Climcal Immunology, University Hospital, PO Box 85.500, 3508 GA Utrecht, The Netherlands. Received November 21, 1991; acceptedApril 6, 1992. Human Immunology 34, 167-172 (1992) © American Society for Histocompadbility and Immunogenetics, 1992
of immunosuppressive agents can usually be decreased in time. This adaptation to the graft or transplantation tolerance can be explained by various mechanisms. A m o n g these, clonal deletion or clonal anergy of the alloreactive cells, and the active suppression of these cells, are the most favored ones [ 3 - 7 ] . A combination of such mechanisms is also possible. It is thought that clonal deletion occurs mainly during T-cell repertoire generation in the thymus and that clonal anergy and active suppression are the mechanisms for extrathymically induced tolerance [ 8 - 1 0 ] . T h e cell-mediated lympholysis assay (CML) enables the m e a s u r e m e n t of the recipient's cytotoxic potential toward spleen-derived mononuclear cells (MNCs) from the donor [ 11]. In clinical kidney transplantation, tolerance was reflected by nonreactivity in the CML assay 167 0198-8859/92/$5.00
168
P.L.J. Wijngaard et al.
[ 12-15]. This nonreactivity correlated with good renal function and could be abrogated by culturing the effector cells with recombinant interleukin-2 (rI1-2) before performing the CML assay [16]. We studied the phenomenon of transplantation tolerance after heart transplantation. In a longitudinal follow-up, ten heart transplant recipients were tested for their cytotoxicity toward donor and third-party, spleenderived MNCs in the CML assay. The percentages of donor-specific cytotoxicity were correlated with various cytomorphologically and phenotypically defined lymphocyte subsets. We also tested the recipients' cytotoxicity toward donor and third-party spleen-derived MNCs without the allostimulation phase. MATERIALS A N D M E T H O D S
Patients. In a longitudinal follow-up after orthotopic heart transplantation, ten patients were tested for their cytotoxic potential against donor and third-party spleen-derived MNCs. Recipient and donor were matched for ABO blood group antigens, but not for HLA histocompatibility; a negative crossmatch between patients' serum and donor leukocytes was required. In the preparative phase for transplantation, all patients received at least one blood transfusion from an unrelated donor. The standard immunosuppressive regimen consisted of cyclosporine and prednisolone (five patients), or cyclosporine, prednisolone, and azathioprine (five patients). In case of acute rejection (grade 2 or 3, diagnosed by histopathologic assessment of the endomyocardial biopsy [EMB] [17], the patients were treated with a methylprednisolone pulse for 3 days and/ or with rabbit antithymocyte globulin. In Table 1, rele-
TABLE 1 Development of CML nonreactivity (NR) in relation to clinical data Patient
CML NR ~
Rejections b
1 2 3 4 5 6 7 8 9 10
8 -24 5 12 27 --23 21
4 4 0 2 4 3 3 3 1 2
Immunosuppression ~ CsA CsA CsA CsA CsA CsA CsA CsA CsA CsA
pred pred pred pred pred, pred, pred, pred pred, pred,
aza aza aza aza aza
Months after transplantation of first manifestation of CML nonreactivity. b Number of acute rejections encountered during follow-up in CML assay. c Standard immunosuppressive therapy: CsA, cyclosporine A; pred, prednisoIon; and aza, azathioprine.
vant clinical data are summarized in relation to the development of CML nonreactivity in these patients.
Preparation ofcellsamples. Peripheral blood MNCs were isolated from heparinized venous blood samples by Ficoll-Isopaque density-gradient centrifugation directly before and at regular intervals after transplantation. Spleen-derived MNCs from the donor were isolated by mincing a small part of the donor spleen with surgical blades. From this suspension, the MNC fraction was isolated by Ficoll-Isopaque density-gradient centrifugation. By use of the MNC isolates, the recipients were monitored for phenotypically and cytomorphologically defined lymphocyte subsets with the cytoimmunologic monitoring approach, described previously [18, 19]. This technique enables one to diagnose periods of acute rejection or infection after transplantation by determining the number of cytomorphologically activated lymphocytes and large granular lymphocytes, and the percentages of CD3, CD4, CD8, and Leu-7-positive lymphocytes. The remainder of the MNC preparations were resuspended in 10% dimethylsulfoxide (BDH, Poole, England) and 40% fetal calf serum (Life Technologies, Paisley, Scotland) in minimal essential medium (Life Technologies) and stored in liquid nitrogen until tested in the CML assays. CML assay. The standard CML assay [11] was used. Briefly, peripheral blood lymphocytes from the recipient were sensitized for 6 clays in a 1 : 1 ratio with irradiated spleen-derived MNCs from their respective donors or from third-party donors, selected from the pool of transplant donors on the basis of a complete HLA-A, -B, -C, and -DR mismatch with the recipients' donor. Cytotoxicity to phytohemagglutinin-stimulated target cells was determined in a 4-hour 5~Cr-release assay in triplicate and at different effector-target ratios. The specific release was calculated using the following formula specific release _- experimental cpm - spontaneous c p m , 100% maximum cpm - spontaneous cpm Maximum release was measured by lysing the target cells in 2.5% Triton X-100; spontaneous release was measured from target cells cultured in medium only. The standard deviation of the triplicate cultures never exceeded 5%. Nonreactivity was defined as a specific release in the CML of < 15 % at an effector-target ratio of 40 : 1. In addition to cytotoxicity after in vitro allostimulation, we assayed cytotoxicity to donor or third-party spleen-derived MNCs without the phase of prior in vi-
CML Nonreactivity After Heart Transplantation
tro allosensitization. This was done in the 4-hour 51Crrelease assay by culturing the recipient's MNCs directly after thawing from liquid nitrogen with target cells at different effector-target ratios. We also measured cytotoxicity after a 6-day culture period only with rIL-2, following thawing of the cells.
Statistical analysis. Correlation analysis was done with the aid of the computer-assisted statistical program SPSS-PC+version 3.0. The percentage of donor-specific cytotoxicity was used as dependent variable; independent variables were time after transplantation; the percentages of CD3, CD4, CD8, and Leu-7-positive cells; and the percentage of cytomorphologically defined activated lymphocytes and large granular lymphocytes [ 1820]. RESULTS In this study, ten patients were tested longitudinally for their cytotoxic potential against spleen-derived MNCs from their respective donors following heart transplantation. In two patients (designated as open triangles in Fig. 1), donor-directed cytotoxicity before and after transplantation was not found. The other patients demonstrated donor-directed cytotoxicity with values varying between 22% and 79% (Fig. 1A). This cytotoxicity was specifically directed to the donor because there was minimal or no lysis of autologous and third-party target cells (Fig. 1B and C). In seven of these patients, the donor-directed cytotoxicity decreased to values of <15% within 3-27 months after transplantation (Fig. 1A). We considered this 15% value as the cutoff between reactivity and nonreactivity, based on the fact that cytotoxicity to autologous targets cells normally is below this value (Fig. 1B). Based on this cutoff value, one patient (open diamonds in Fig. 1) did not develop CML nonreactivity toward the donor, but only demonstrated a strong decrease in the CML in time from 79% before transplantation to 25% at 36 months after transplantation. We analyzed whether the observed nonreactivity is donor-specific, by allostimulating the recipient's MNCs with HLA-mismatched third-party spleen-derived MNCs and subsequent CML toward these cells. The low yields of recipients' MNCs after isolation permitted this control experiment only for four of the patients that demonstrated donor-directed nonreactivity. The data, expressed as the ratio of donor-directed cytot'oxicity over third-party-directed cytotoxicity, showed a decrease in time after transplantation (Fig. 2). This means that cytotoxicity toward third-party target cells is still present, whereas at the same time the cytotoxicity to-
169
ward donor targets' cells is decreased below values of 15%. During donor-directed nonreactivity in the CML, the cytotoxicity toward the third-party target cells ranged between 21% and 63%. The development of CML nonreactivity coincided with good functioning of the graft. Episodes of acute rejection did not occur during the period of unresponsiveness, with the exception of one patient (+ in Fig. 1A). This patient, a 37-year-old woman, was on maintenance immunosuppressive therapy comprising 300 mg cyclosporine A and 10 mg prednisolone per day, and had developed CML nonreactivity 8 months after transplantation. This remained so at least until 32 months after transplantation. At 16 months after transplantation, the cyclosporine and prednisolone doses were reduced by 25%. One month later, the endomyocardial biopsy showed grade-2 rejection, which was adequately treated with an oral pulse of 200 mg prednisolone for 3 days. The next biopsy specimen, taken 9 days later, showed no signs of rejection (grade 0). The biopsy specimen taken a subsequent week later again manifested grade-2 rejection. This was successfully treated with a pulse of 1000 mg methylprednisolone on each of 3 successive days. Thereafter, acute rejection episodes no longer occurred. The nonresponsiveness in terms of cytotoxic activity in vitro could be abrogated when the recipient's effector cells (106 MNCs/ml)were cultured with 20 U/ml rII-2 for 6 days after the initial stimulation phase. A cytotoxic potential toward the donor of 2 4 % 35% was measured after this preculture. No cytotoxicity toward autologous and third-party target cells was observed. We performed correlation analyses with regard to cytomorphologic and immunologic phenotype of blood MNCs [18, 19] and with time after transplantation using donor-directed cytotoxicity as an independent variable. A significant inverse correlation was found between the percentage of donor-directed cytotoxicity and time after transplantation (r = -0.281, p = 0.014). This is illustrated in Fig. 1. Furthermore, a significant positive correlation was observed with the percentage of lymphocytes with the cytomorphology of activated cells in blood (r = 0.367, p = 0.010) (Table 2). To test whether in vivo alloactivated lymphocytes are capable of exerting cytotoxicity to donor and control autologous and third-party target cells, we tested MNCs of four patients in the CML assay without prior allostimulation. No lysis of donor, autologous, and third-party target cells was observed. Also, when the patient's blood MNCs (106 MNCs/ml) were first cultured with 20 U/ml rII-2 for 6 days before testing the cytotoxic response, no cytotoxicity toward the same target cells was demonstrable (results not shown).
170
P . L . J . Wijngaard et al
A 100
80 >., 4-,
60
X
o o :>, {3
40
2O
0 -1
0
1 2
3
4
6
8
9 101213151718202122232427293031323641 months
after
transplantation
B IO0
80 >, 4-, X 0
60
0 0
40
20
0 -1
0
1 2
3
4
6
8
9 101213151718202122232427293031323641 months
after
transplantation
C
100
80
0 X 0
60
o >, O
40
g
Oi -1
~-~---~___a'--m. ~__~ 0
1 2
3
4
6
~ 8
~
.
,--
~
:
.-"~:- ~
'
t
-
-
.
~
9 101213151718202122232427293031323641 months
after
transplantation
I
I
I
F I G U R E l Development of nonresponsiveness in cellmediated lympholysis (CML) after heart transplantation. The recipients' mononuclear cells (MNCs) were stimulated with donor spleen-derived MNCs and tested in CML assays using the following target cell populations: (A) donor spleen-derived MNCs, (B) autologous blood MNCs, and (C) HLA-mismatched third-party spleenderived MNCs. The 15 % cytotoxicity level is indicated by the dotted line. The patient indicated by + developed an acute rejection after reduction of immunosuppression during CML nonresponsiveness 16 months after transplantation. The patient indicated as © developed donor-directed nonresponsiveness without manifesting any acute rejections in the first 3 years after transplantation.
CML Nonreactivity After Heart Transplantation
TABLE 2
Variable t > n r pd t >
171
CML nonreactivity coincided with good functioning of the graft. Rejection crises were not observed during the presence of CML nonreactivity. This was except for one patient, who, after a 25% reduction in the immunosuppressive therapy, manifested acute rejection grade 2 during the period of nonreactivity. The nonresponsiveness in this patient could be abrogated by culturing the recipient's effector cells after the sensitization phase with exogeneous rI1-2 for another 6 days. This indicates the presence of donor-specific alloreactive (precursor) cytotoxic T cells (CTLs) in the circulation of this patient. Apparently, these cells were rendered anergic or were actively suppressed, but they obviously were not completely deleted. Whether this is the case for all patients developing CML nonreactivity after heart transplantation has to be examined. For kidney transplantation, it has been demonstrated that the donor HLA-antigen inducible precursor CTL frequency gradually decreases in time [21]. At present, we are analyzing precursor CTL frequencies in a longitudinal follow-up in heart transplant recipients in relation to the development of CML nonreactivity. One patient did not manifest any acute rejection episode during the first 3 years after transplantation (Table 1). Donor-directed cytotoxicity dropped from 41% before to 9% two years after transplantation (indicated by open circles in Fig. 1). Apparently, CML nonreactivity develops without strong in vivo alloactivation as envisaged by clinical rejections as diagnosed by EMB histopathology. The percentage of donor-specific cytotoxicity significantly correlated with time after transplantation as was expected, but it also correlated with the percentage of cytomorphologically defined activated lymphocytes in peripheral blood. In previous studies, we have demon-
Correlation coefficients (r) for percentage of donor-specific lysis as independent variable Tx ~
CD3
CD4
CD8
Leu-7
ALb
LGL'
61 23 23 23 21 40 40 -0.281 0.177 0.189 0.340 0.154 0.367 0.065 0.014 0.210 0.194 0.056 0.252 0.010 0.346
Tx, time af[er [ransplantation.
AL, cytomorphologically defined activated lymphocytes. c LGL, cytomorphologically defined large granular lymphocytes. p, level of significance.
DISCUSSION Patients who have received an allograft can develop tolerance to the alloantigens presented in the graft. This in vivo adaptation enables one to decrease the dosage of immunosuppressive agents after transplantation. Transplantation tolerance may be reflected in vitro by the absence of donor-specific cytotoxicity in the CML assay [12-15]. H e r e , we demonstrate the development of CML nonreactivity in seven of eight patients after heart transplantation. These manifested CML reactivity before or in the first 3 months after transplantation, and the CML nonreactivity occurred for the first time between 3 and 27 months after transplantation (Fig. 1). The CML nonreactivity was donor specific because lymphocytes from the recipients were able to lyse thirdparty target cells that were completely HLA-A, -B, -C, and -DR mismatched with donor and recipient (Fig. 2). In two patients, the CML reactivity was not detectable and, therefore, these patients were considered negative for developing CML nonreactivity. FIGURE 2 Nonresponsiveness in cell-mediated lympholysis (CML) assays is donor specific. Recipients' mononuclear cells (MNCs) were stimulated with either donor spleenderived MNCs or with HLA-mismatched thirdparty spleen-derived MNCs and tested in the CML assay with these respective stimulator cells as target cells. The ratio of cytotoxicity to donor over that to third party is plotted against time after transplantation. Data shown is for individual patients, indicated by the same symbols as in Fig. 1.
3.00
o
2.40
~ c~
1.80
£ o 8 x~
1.20
0.60 0.00 -1
0
1
3 months
10 after
17
transplantation
27
31
41
172
strated that the percentages of the cytomorphologically activated lymphocytes are increased during periods of acute rejection [18, 19]. Apparently, this process is also reflected in the percentage of donor-directed cytotoxicity. To test whether these in vivo alloactivated lymphocytes are capable of exerting in vitro cytotoxicity, we tested M N C s in the CML assay without the prior in vitro allosensitization phase. In this condition, the recipients' lymphocytes were unable to lyse donor target cells, even when the lymphocytes were first stimulated with rI1-2 for 6 days. This absence of detectable cytotoxicity in this direct CML assay was also evident in cases with enhanced percentages of lymphocytes with the cytomorphology of activated cells, reflecting the presence of rejection. Thus, cytomorphologically defined activated lymphocytes in peripheral blood during periods of acute rejection reflect the rejection process, but may not be necessarily responsible for the cytotoxic reactions toward the graft as c o m p o n e n t in the rejection process. The mechanisms underlying transplantation tolerance are not completely resolved. In studies in mouse it has been postulated that clonal deletion, clonal anergy, and/or the active suppression of the alloreactive cells may play a role [ 3 - 7 ] . Our data in a single patient (indicated as + in Fig. 1) indicate a role for clonal anergy or active suppression for the development of CML nonreactivity after heart transplantation. Whether this is the more general mechanism has to be further analyzed.
REFERENCES 1. Mason DW: Effector mechanisms in allograft rejection. Annu Rev Immunol 4:119, 1986. 2. Mason DW, Dallman MJ, Arthur RP, Morris PJ: Mechanisms of allograft rejection: The roles of cytotoxic T-cells and delayed-type hypersensitivity. Immunol Rev 77:16, 1984. 3. Kamada N: Transplantation tolerance and immunosuppression following liver grafting in rats. Immunol Today 6:336, 1985. 4. Dallman MJ, Shiho O, Page TH, Wood KJ, Morris PJ: Peripheral tolerance to alloantigen results from altered regulation of the interleukin 2 pathway. J Exp Med 13:79, 1991. 5. Robbins PA, McMichael AJ: Immune recognition of HLA molecules downmodulates CD8 expression on cytotoxic T-lymphocytes. J Exp Med 13:221, 1991. 6. Slavin S: Total lymphoid irradiation. Immunol Today 8:88, 1987.
P.L.J. Wijngaard et al.
7. Silvers WK, Kimura H, Desquenne-Clark L, Miyamoto M: Some new perspectives on transplantation immunity and tolerance. Immunol Today 8:185, 1987. 8. Ramsdell F, Fowlkes BJ: Clonal deletion versus clonal anergy: the role of the thymus in inducing self tolerance. Science 248:1342, 1990. 9. Sprent J, Gao E-K, Webb SR: T-cell reactivity to MHC molecules: immunity versus tolerance. Science 248:135, 1990. 10. MacDonald HR, Lees RK: Programmed death of autoreactive thymocytes. Nature 343:642, 1990. 11. Goulmy E, Termijtelen A, Bradley BA, van Rood JJ: HLA restriction of non-HLA-A, -B, -C and -D cell mediated lympholysis (CML). Tissue Antigens 8:317, 1976. 12. Goulmy E, Persijn G, Blokland E, D'Amaro J, van Rood JJ: Cell-mediated lympholysis studies in renal allograft recipients. Transplantation 31:210, 1981. 13. Van Rood JJ, Claas FHJ: The influence of allogeneic cells on the human T and B cell repertoire. Science 248:1388, 1990. 14. Pfeffer PF, Thorsby E, Hirschberg H: Donor-specific decreased cell-mediated cytotoxicity in recipients of well functioning, one HLA haplotype-mismatched kidney allografts. Transplantation 35:156, 1983. 15. Pfeffer PF, Hirschberg H, Thorsby E: Donor-specific decreased primary and secondary cell-mediated immune responses in patients with well functioning grafts. Transplant Proc 13:1604, 1981. 16. Goulmy E, Blokland E, Persijn G, Paul LC, Wilmink J, van Rood JJ: HLA regulates postrenaI transplant CML nonreactivity. J Immunol 135:3082, 1985. 17. Billingham ME: Diagnosis of cardiac rejection by endomyocardial biopsy. J Heart Transplant 1:25, 1981. 18. Wijngaard PLJ, Heyn A, van der Meulen A, de Vries H: Cyto-immunologic monitoring following heart transplantation. Bibl Cardiol 43:137, 1988. 19. Wijngaard PLJ, Gimpel JA, Schuurman H-J, van der Meulen A, Gmelig Meyling FHJ, Jambroes G: Rejection monitoring after heart transplantation: cyto-immunologic monitoring on blood cells and quantitative birefringence measurements on endomyocardial biopsies. J Clin Pathol 43:137, 1990. 20. Ertel W, Reichenspurner H, Hammer C, Lersch C, Plahl M, Brendel W, Reichart B, Kemkes BM, Reble B, Gokel JM: Cyto-immunologic monitoring: a method to reduce biopsy frequency after cardiac transplantation. Transplant Proc 17:204, 1985. 21. Herzog W-R, Zanker B, Irschick E, Huber C, Franz HE, Wagner H, Kabelitz D: Selective reduction of donorspecific cytotoxic T-lymphocyte precursors in patients with a well functioning kidney allograft. Transplantation 43:384, 1987.