• Kidney transplantation;
  • tolerance;
  • T cells;
  • T-cell receptor


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients, Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Most kidney transplant recipients who discontinue immunosuppression reject their graft. Nevertheless, a small number do not, suggesting that allogeneic tolerance state (referred to operational tolerance) is achievable in humans. So far, however, the rarity of such patients has limited their study. Because operational tolerance could be linked to anergy, ignorance or to an active regulatory mechanism, we analyzed the blood T-cell repertoire usage of these patients. We report on comparison of T-cell selection in drug-free operationally tolerant kidney recipients (or with minimal immunosuppression), recipients with stable graft function, chronic rejection and healthy individuals. The blood T cells of operationally tolerant patients display two major characteristics: an unexpected strongly altered T-cell receptor (TCR) Vβ usage and high TCR transcript accumulation in selected T cells. The cytokine transcriptional patterns of sorted T cells with altered TCR usage show no accumulation of cytokine transcripts (IL10, IL2, IL13, IFN-γ), suggesting a state of hyporesponsiveness in these patients. Identification of such a potential surrogate pattern of operational tolerance in transplant recipients under life-long immunosuppression may provide a new basis and rationale for exploration of tolerance state. However, these data obtained in a limited number of patients require further confirmation on larger series.


normal individual;


drug-free operationally tolerant patient;




chronic rejection;


stable graft function;


CalciNeurin inhibitor;


peripheral blood mononuclear cells;


antigen presenting cell;


complementary determining region 3-length distribution;


hypoxanthine phosphoribosyl transferase;


anti lymphocyte globulins;


polymerase chain reaction;


T-cell receptor;






Fluorescein isothiocyanate;


panel reactive antibody;


human leukocyte antigen;




hepatitis C virus;


post-transplant lymphoproliferative disorder;


anti-thymocyte globulin;


multivariate statistical analysis of variance;


Epstein Barr virus;


herpes simplex virus;






  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients, Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Kidney transplantation is the reference treatment for patients with end-stage renal disease (1). However, the rate of progressive graft loss 1 year after transplantation has been little influenced by increasingly efficient immunosuppressants (2,3), suggesting that immunologically driven chronic graft rejection might not be the primary mechanism involved in graft loss and that chronic dysfunction is also fundamental (2,4). In addition, a link between an increased incidence of skin cancer and lymphoma and long-term exposure to immunosuppression has been well documented (5). Consequently, the possibility of decreasing or withdrawing immunosuppression in kidney recipients, particularly in the long term, would be a major advance.

Anecdotal cases of kidney recipients having interrupted immunosuppression without subsequent rejection (6–9) have raised the possibility that some patients with well-functioning grafts may have become operationally tolerant. Certain graft recipients with stable function have been shown to develop circulating T cells exhibiting regulatory properties in vitro for autologous pre-graft T cells stimulated by donor leukocytes, whereas patients with chronic rejection more frequently develop effector cells against donor cell determinants (10,11). Nevertheless, there is currently no established procedure to determine who, among stable immunosuppressed recipients, may be operationally tolerant and thus could be withdrawn from chronic immunosuppression.

Because operational tolerance could be linked to clonal deletion, anergy, ignorance or to an active regulation, we analyzed T-cell selection and activation in the blood of kidney recipients who, many years after transplantation, are ‘operationally tolerant’ to their graft. We show that drug-free operationally tolerant patients have previously unobserved peripheral T-cell clonal alteration. In addition, sorted T-cell populations using selected T-cell receptors (TCR) from drug-free operationally tolerant recipients are mainly CD8+, characterized by an absence of cytokine transcript accumulation. Finally, minimally immunosuppressed kidney recipients display the same TCR alteration patterns as drug-free operationally tolerant recipients, suggesting they may be operationally tolerant too. These new observations open up the possibility that a decrease in the diversity of the CD8+ TCR repertoire in these patients may contribute to the maintenance of their operationally tolerant state. Taken together, our data also suggest that altered clonal distribution in the blood of operationally tolerant recipients may act as a marker, which could help to identify a state of operational tolerance in recipients under chronic immunosuppression with stable graft function.

Patients, Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients, Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References


The protocol was approved by the University Hospital Ethical Committee and the Committee for the Protection of Patients from Biological Risks. All patients were recruited at the Nantes Institute for Transplantation (except for one who was recruited at the St-Louis Hospital in Paris). All patients and normal individuals were informed of the protocol and gave informed consent. The University Ethical Committee recommended against performing biopsies in long-term recipients with normal stable function.

Drug-free operationally tolerant (DF-Tol) group:  DF-Tol recipients with stable kidney graft function had not taken any immunosuppressive drugs for at least 3 years before analysis (n = 5) (Table 1). These patients were the only ones identified as being drug-free and operationally tolerant among an estimated 25 000 or more kidney transplant recipients in France. Figures 1A (left panel) and 1B show the renal function (serum creatinine) of these drug-free tolerant patients, which remained remarkably stable and within the limit of 200 μM of blood creatinine and 1.5 g/24 h of proteinuria. The mean drug-free time period was 8 ± 3.3 years (range 3–12).

Table 1.  Clinical history of DF-Tol and Ster patients
 DF-Tol 1DF-Tol 2DF-Tol 3DF-Tol 4DF-Tol 5Ster 1Ster 2Ster 3Ster 4Ster 5
  1. s: Indicates typing by serology.

  2. mb: Indicates typing by molecular biology.

  3. Note: Table 1 summarizes the clinical history of DF-Tol (n = 5) and Ster (n = 5) patients. One patient (patient DF-Tol 4) received a second graft. The mean HLA-A, -B and -DR incompatibility was 3. Two patients received an induction therapy after transplantation with a 10-day course of an anti-IL2-R-α antibody (DF-Tol 1) and a 6-day therapy of ATG (DF-Tol 5). One patient (DF-Tol 3) underwent two acute rejection episodes 14 days and 7 months after transplantation, which were both treated with steroid boluses. No patient in this group had a CMV disease during their follow-up. Patient DF-Tol 4 was chronically infected by the hepatitis C virus. One patient (Ster 2) received a second graft. The mean HLA-A, -B and -DR incompatibility was 2.8 ± 1.9. The mean pre-graft anti-T PRA was 30 ± 33%. Four patients received ATG induction therapy for 14 ± 4 days. Patient Ster 5 received azathioprine and steroids, and underwent one episode of acute rejection just after transplantation, which was treated with a 10-day course of ALG. Patient Ster 4 had a CMV disease 4 years after transplantation, which was treated with gancyclovir. Patient Ster 3 suffered from serial virus infections, that is, two herpes zoster virus reactivation at 4 and 9 years, EBV at 8 years and hepatitis C infection at 12 years post-transplantation.

Patient age at time of study67657371247683296654
Patient sexMaleMaleFemaleMaleMaleMaleMaleFemaleFemaleFemale
Number of transplants1112112111
Year of transplantation1987198019931974199119971990199819831980
Induction therapy (duration)Anti RIL2 mAb (10 days)NoNoNoATG  (6 days)ATG  (11 days)ATG  (17 days)ATG  (11 days)ATG  (19 days)No No
HLA-A-B-DR inc.*3s3s3sND33mb0s5mb4s2s
Anti-T PRA prior to transplantation (%)0200ND007206020
Anti-T PRA at time of study (%)000ND0ND0000
Acute rejection episodeNoNo2NoNoNoNoNoNo1
CMV infectionNoNoNoNoNoNoNoNoYes (1987)No
Other viral infectionsNoNoNoHCVNoNoHSV (1991)Zona (1992) EBV (1996) Zona (1997) HCV (2000)NoNo
Reason for IS interruptionPTLDNon-complianceCNI toxicityNon-compliancePTLDRecurrent serious bacterial infectionsLarge bowel carcinomaPTLDVocal cord epitheliomaCutaneous epithelioma
Time period : drug-free or low dose8 years>10 years8 years12 years3 years5 years4 years6 years15 years9 years

Figure 1. Graft function in the various kidney recipient groups. (A) The left-hand panel shows the creatinemia (▪) and proteinuria levels (♦) for each drug-free patient (DF-Tol group) before and after withdrawal of immunosuppressive treatment until the time of analysis. The corresponding data for patients under low-dose corticosteroid monotherapy (Ster group) are displayed on the right-hand panel. The vertical black bars indicate the time of immunosuppression interruption or the initiation of the very low-dose immunosuppressive regimen. Patients DF-Tol 1 and 5 presented a lymphoma, 52 and 96 months following transplantation without recurrence when tested at 7 and 3 years, respectively. Patient DF-Tol 3 stopped immunosuppressive treatment due to kidney dysfunction 6 months after transplantation. He returned to chronic dialysis, his graft function progressively recovered after Cyclosporine A (CsA) interruption and his dialysis was interrupted 1 year later. Patients DF-Tol 2 and 4 stopped immunosuppressive treatment voluntarily without any medical reason. (B) The mean blood creatinemia values for ‘operationally tolerant’ patients (DF-Tol and Ster) before and after immunosuppression interruption and for patients with stable function under immunosuppression and with chronic rejection are shown. In the group of patients with chronic rejection (CR group), the values for patients under chronic dialysis are not indicated (resulting in the apparent delayed decreased in mean creatinine levels).

Download figure to PowerPoint

Minimally immunosuppressed kidney recipients:  Another group of patients received low doses (≤10 mg/day) of prednisone or prednisolone as the only immunosuppressive drug for at least 3 years before analysis (9.8 ± 4.4 years (range 4–15)). These minimally immunosuppressed patients are referred to as the Ster group (n = 5)(Table 1). Figure 1A (right panel) shows the creatinemia levels before and after the onset of monotherapy in each of these patients.

Patients with chronic rejection, patients with stable graft function under an immunosuppressive maintenance regimen and normal, healthy individuals:  Because there were several possible rationales for the most appropriate control group (based on having a stable graft function or an absence of immunosuppression), several clinical situations were studied. The stable function group (Sta) included kidney recipients with long-term stable graft function who were under a dual-therapy immunosuppressive regimen (n = 7) including a calcineurin inhibitor (CNI+; n = 4) or not (CNI; n = 3). The chronic rejection group included patients with a degradation of renal function and histological chronic rejection lesions (CR group; n = 7) associated or not with an allograft glomerulopathy. In 4 patients, immunosuppressive treatment at the time of the study consisted of a combination of tacrolimus and azathioprine or prednisone. Three CR patients were on dialysis following allograft nephrectomies and had been free of immunosuppressive drug treatment (DF-CR) for at least 1 year. Figure 1B summarizes the kinetics of graft function in these groups. The apparent late decrease in creatinemia in the chronic rejection group is due to the fact that the values of the patients who had reverted to dialysis were not taken into account. Normal individuals, 25–60 years of age, included healthy individuals (NL group, n = 15) with normal blood results and no known infectious pathology for at least 6 months prior to the study.

Blood samples

Blood was collected in EDTA tubes. Ionogram, white blood cells (WBC), formula and calcineurin inhibitor (CNI) blood levels were performed on each sample. peripheral blood mononuclear cells (PBMC) were separated on a Ficoll layer (Eurobio, Les Ulis, France) and frozen in TRIzol® reagent (Invitrogen, Life technologies, CA) for RNA extraction. Negative selection of CD4+/CD8+ T cells was performed using MACS® human cell isolation kits (Miltenyi Biotec, Germany). Purity was >90%.

TCR repertoire analysis

Diversity and T-cell selection were assessed by analysis of TCR (β chain) usage biases. In an immunologically quiescent state, the length distribution of the TCR-β chain complementary determining region 3 (CDR3) hypervariable region is gaussian. Alteration of this gaussian distribution is the hallmark of an activation phenomenon. Total mRNA was reverse-transcribed using a cDNA synthesis kit (Boehringer Mannheim, Indianapolis, IN) and complementary DNA was amplified by PCR, elongated, loaded onto a 6% acrylamide, 8-M urea gel and seperated by electrophoresis as previously described (12). Analysis of CDR3-length distribution (LD) was performed using Immunoscope® software (Institut Pasteur, Paris, France) (12–15) that provides distribution profiles of CDR3 lengths in amino acids. The CDR3 length profiles were compared with those from 15 normal individuals used as controls (15) and defining non-disturbed gaussian profiles. The average of these normal distributions, for each Vβ family separately, was then used as a control distribution for analysis of the other samples. All perturbations, including those of the controls, were assessed by comparison with the control distribution for each individual Vβ family individually. The alteration per Vβ family is thus defined as the sum of the absolute values of difference for all CDR3 lengths in that profile. No alteration gives a difference of 0%. A difference of 50% represents a very significant alteration. The average of the alterations in all Vβ families in one sample gives the global percentage of alteration for each individual. To assess the magnitude of the Vβ transcript accumulation, the level of Vβ RNA was measured by real-time quantitative PCR. The data are displayed as a tridimentional TcLandscape® (Nantes Cedex 01, France) of the entire blood T-cell repertoire for each patient analyzed. A topview of the TcLandscape allows an easier assessment of CDR3-LD alterations. Percentages of CDR3-LD alterations are represented by a color code, ranging from deep blue (−50%) to dark red (50%). The X-axis displays the 24 Vβ families in humans. Correspondence between numbers indicated on the X axis and the different Vβ families is given in Figure 2. The Y-axis gives the 13 CDR3 lengths and the Z-axis gives the amount of each Vβ concerned, normalized against levels of the housekeeping gene HPRT (Vβ/HPRT ratio) (16,17).


Figure 2. TcLandscape representation of the T-cell Vβ transcriptome in blood of transplanted patients. The five TcLandscape from drug-free tolerant kidney recipients (DF-Tol), 1 out of 15 healthy individuals (NL); 2 out of 5 minimally immunosuppressed patients under low dose of steroids (Ster), 1 out of 7 patients with stable graft function (StaCNI); 2 out of 7 patients with chronic rejection (CR).

Download figure to PowerPoint

IL2-R-α, perforinand cytokine transcript quantification

IL2-R-α and cytokine transcript measurement were performed on RNA from PBMC, CD4+/CD8+ subpopulations and Vβ families positively selected using MACS® anti-PE micro-beads and a MS+ type separation column (Miltenyi Biotec, Germany). RNA was prepared using the SMART kit (Boehringer Mannheim, Indianapolis, IN). Real-time quantitative PCR was performed for IL2-R-α, IL2, IFN-γ, IL8, IL10, TGF-β, IL13 and perforin and normalized against the HPRT transcript level (16).

T-cell Vβsurface staining

Two color labeling was performed on PBMC using FITC-conjugated anti-CD3, -CD4 and -CD8 antibodies (Abs) (Immunotech, Coulter Company, Marseille, France), and the 24 available PE-conjugated anti-Vβ Abs (Immunotech, Coulter Company, Marseille, France) or isotype-matched control Abs (IgG-FITC and IgG-PE). The staining for each Vβ family was expressed as a percentage of the total staining.

Statistical analysis

The different groups were compared on the basis of CDR3-LD and Vβ/HPRT transcript ratio values, considering their possible correlation (Manova), with a global test (F-test) and the related pairwise comparisons. After checking for normal distribution, Manova was performed with SAS© software (Cary, NC) and the ‘Mixed’ procedure. The influence of age, HLA mismatch, graft number, PRA (%), previous acute rejection, CMV, lymphoma, cancer, corticosteroids and the different immunosuppressive drugs used on the CDR3-LD and Vβ/HPRT transcript ratio values was evaluated using Manova with a global test (F-test) and the related pairwise comparisons. Comparison was performed between the different groups: drug-free patients (DF-Tol), minimally immunosuppressed patients (Ster), patients with stable graft function (Sta), patients with chronic rejection (CR) and normal individuals. Due to the low number of patients in each group, patients considered clinically as operationally tolerant (i.e. drug-free patients (DF-Tol) and minimally immunosuppressed patients (Ster)) were also pooled for CDR3-LD and Vβ/HPRT transcript ratio analyses. Statistical analyses of cytokine transcript accumulation were performed using a non-parametric Kruskal–Wallis test. The results were considered significant when p < 0.05.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients, Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Vβtranscriptome analysis and the blood T-cell repertoire of normal individuals

Our new method provides an integrating vision of in vivo T-cell clonal distribution where a green landscape indicated unselected T cells and red spots indicated strong T-cell selection. CDR3-LD and Vβ/HPRT ratios were also statistically compared (see Patients, Materials and Methods for technical details). A representative TcLandscape topology of the 15 normal individuals are displayed in Figure 2 and online Figure 1 (see for online figures). Normal individuals exhibited only slightly altered CDR3-LD with 93% of Vβ families displaying <30% CDR3-LD alteration and about 1% of Vβ families displaying more than 50% CDR3-LD alteration. Compared to the mean CDR3-LD alteration observed in the 15 healthy individuals, global CDR3-LD alteration for each normal individual (sum of alteration of the 24 Vβ families for each individual) remained within 10.9 ± 2.9%. In addition, in normal individuals, 83% of Vβ families had a Vβ/HPRT transcript ratio <5 whereas only 1% had ratios >15. The TCR patterns were found to be roughly similar in time when tested every 2 months for 8 months (M. Guillet et al., unpublished). In the following section, the patterns for each recipient group are described. The statistical differences between groups are detailed in a separate section.

The blood T cells of drug-free operationally tolerant patients display altered CDR3-LD patterns together with high Vβ/HPRT transcript ratios

Figure 1A shows the graft function (creatinemia and proteinuria) of drug-free operationally tolerant patients (DF-Tol) (left panel) before and after withdrawal of classical immunosuppressive treatment until the time of testing. Figure 1B shows the graft function of the recipient groups. A summary of the clinical history of each patient of the DF-Tol group is provided in Table 1 (left-hand part). Detailed CDR3-LD values of DF-Tol patients are displayed in Table 2 and TcLandscape topologies (Figure 2 left-hand panel). In DF-Tol recipients, 9% of Vβ families showed more than 50% CDR3-LD alteration (Table 2) whereas the global CDR3-LD alteration was 28.3 ± 11%. In addition, these recipients showed high Vβ/HPRT transcript ratios with approximately 8% more than 15 (Table 2). These ratios corresponded to a strong accumulation of Vβ transcripts since HPRT levels were unchanged in operationally tolerant patients compared to the other groups (data not shown). This phenomenon was not associated with deletion of the remaining Vβ transcripts since no decrease was observed in operationally tolerant compared to stable and chronic rejection patients, nor did it correlate to significant changes in the percentage of staining for the corresponding anti-Vβ antibodies. For example, the transcripts of the Vβ1, 4 and 11 families represented 29, 6.5 and 12%, respectively, of the whole Vβ transcriptome in one representative patient (DF-Tol 4 patient) whereas the percentages of T-cells staining for the corresponding Vβ protein were only 6.9, 0.2 and 0.5%, respectively, (Figure 3C) and did not differ from the values of normal individuals.

Table 2.  CDR3-LD alteration and Vβ/HPRT transcript ratios in normal individuals and kidney recipients
Alterations:<30%30–50%>50%Mean (%)
  1. *% of Vβ families in the alteration interval.

  2. Note: (a) the distribution of Vβ families according to the percentage of CDR3-LD alteration, global CDR3-LD alterations and (b) the Vβ/HPRT transcript ratios are given for the different clinical groups of kidney recipients tested (drug-free (DF-Tol) and those under low-dose steroids (Ster); Stable = stable with (CNI+) and without (CNI) a calcineurin inhibitor; CR = chronic rejection with (CR) or without (DF-CR) treatment) and healthy age-mached individuals.

NL93*6110.9 ± 2.9
DF-Tol5932928.3 ± 11
Ster32303842 ± 15
Sta8116320.6 ± 3.7
CR6826623.9 ± 5.5
Vβ/HPRT (%)0–55–15>15 

Figure 3. (A) CDR3-LD analysis was first performed on purified CD4+ and CD8+ T-cell populations in 4 DF-Tol patients. Figure 3A gives a representative example of a CDR3-LD analysis performed on three families sorted from patient DF-Tol 2. The alterations were mainly found to correspond to a CD8+ T-cell population (>3/4 of the Vβ families analyzed). The CD4+ T-cell population displayed a gaussian CDR3-LD pattern. (B) CDR3-LD analyses performed at 6-month intervals in 2 DF-Tol patients (patients DF-Tol 2 and 4) showed roughly stable patterns in time. (C) Two color labeling using anti-Vβ and anti-CD3 antibodies was performed on PBL from DF-Tol and Ster group patients. Vβ staining (▪) was expressed as a percentage of the total staining. Similarly, the Vβ/HPRT ratio (▪) was expressed as the percentage of Vβ RNA for the 19 corresponding Vβ families identified by available antibodies. This figure gives a representative example of patient DF-Tol 4 where several small size (low-percentage-specific Vβ positive staining) T-cell populations were characterized by a strong Vβ transcript accumulation. Vβ1, 4 and 11 represented 29, 6.5 and 12%, respectively, of the whole Vβ transcriptome whereas cells staining for the corresponding Vβ protein represented only 6.9, 0.2 and 0.5%, respectively, and did not differ from normal individual values. (D) Cytokine transcript measurement was performed on purified CD4+ and CD8+ T-cell populations and sorted Vβ families from 4 DF-Tol patients. The lack of IL2 transcript accumulation was restricted to sorted Vβ families with high Vβ/HPRT transcript ratios and CDR3-LD alterations and not to CD4+/CD8+ T-cell sub-populations.

Download figure to PowerPoint

CDR3-LD analysis was performed on purified CD4+ and CD8+ T-cell populations in 4 DF-Tol patients. Most of the alterations were found within the CD8+ T-cell population (Figure 3A) (corresponding to at least 80% of CD8+ T cells as measured by flow-cytometry) whereas the CD4+ T-cell population displayed a gaussian CDR3-LD pattern. CDR3-LD analysis performed over a 6-month interval in 2 DF-Tol patients (DF-Tol 2 and 4) revealed patterns roughly stable in time (Figure 3B).

Patients under low-dose steroid monotherapy (Ster) exhibit a TCR Vβ pattern, which resembled that observed in drug-free operationally tolerant patients

An analysis of the blood TCR repertoire in Ster patients also revealed highly altered CDR3-LD patterns (Figure 2 and online Figure 2): 38% of Vβ families showed more than 50% CDR3-LD alteration and the global CDR3-LD alteration was extremely high (42 ± 15%) (Table 2). Furthermore, these patients had high Vβ/HPRT transcript ratios with 8% of Vβ families having a Vβ/HPRT ratio >15 (Table 2). Thus, drug-free patients and those under low-dose steroid monotherapy (Ster) were both characterized by highly altered CDR3-LD patterns and strong Vβ/HPRT transcript ratios for several Vβ families.

The blood T cells of recipients with stable graft function or with chronic rejection exhibit moderate CDR3-LD alterations and low Vβ/HPRT transcript ratios

The TCR repertoire of patients with stable graft function (Sta) under calcineurin inhibitor (CNI) treatment displayed few CDR3-LD alterations and low Vβ/HPRT transcript ratios (Figure 2 and online Figure 3A) (Table 2). More than 62% of the Vβ families of StaCNI+ patients displayed less than 30% CDR3-LD alterations and their global CDR3-LD alteration was 20.6 ± 3.7%. In these patients, low Vβ/HPRT transcript ratios (<5) were observed for 84% of the Vβ families, whereas only 1% of families had a ratio >15 (Table 2). Three patients without CNI displayed the same pattern (online Figure 3A) (Table 2). Thus, patients with stable kidney graft function, with or without calcineurin inhibitor treatment, had CDR3-LD and Vβ/HPRT ratios within the range of normal individuals. Slightly more altered patterns were observed in the blood of patients with chronic rejection (CR) (Figure 2 and online Figure 3A): 54% of Vβ families showed less than 30% CDR3-LD alteration but 11% showed >50% alteration (Table 2). The global CDR3-LD alteration was 27 ± 6.3%. Whereas the CDR3-LD showed more alterations in CR patients than in normal individuals, Vβ/HPRT transcript ratios in CR patients remained low, with 90% of Vβ families having a Vβ/HPRT ratio <5 and only 1% of them having a ratio >15 (Table 2). As chronic immunosuppressive drug intake could modify the magnitude of the alterations, drug-free CR patients (DF-CR) were also analyzed (online Figure 3B). In these individuals, 59% of Vβ families showed CDR3-LD alterations below 30% whereas only 1% of families displayed alterations >50% (Table 2) and their global CDR3-LD alteration was 20.9 ± 4.8%. These patients also displayed low Vβ/HPRT transcript ratios with 98% of Vβ families having ratios <5 and no ratio >15 (Table 2).

Global statistical correlations

Whereas drug-free operationally tolerant patients and minimally immunosuppressed patients (low-dose steroid monotherapy) did not significantly differ, neither for CDR3-LD alteration nor for Vβ/HPRT transcript ratios, statistical analysis (Manova) showed that both of these groups displayed significantly stronger CDR3-LD alteration and higher Vβ/HPRT transcript ratios than any of the other groups analyzed: recipients with stable graft function (p < 0.01 and p = 0.01, respectively) or chronic rejection (p = 0.01 and p < 0.01, respectively) and normal individuals (p < 0.01 and p < 0.01, respectively). Finally, all recipient groups had significantly more altered CDR3-LD (p < 0.01) than normal individuals whereas patients with stable graft function and patients with chronic rejection did not significantly differ from each other in their CDR3-LD. Because specific clinical parameters could have influenced TCR patterns, special attention was paid to age, HLA mismatch, graft numbers, PRA %, previous acute rejection, CMV, malignancies and specific treatments (presence of steroids, type of immunosuppressive treatment). Two of 5 drug-free patients had a history of malignancies (see Table 1). These malignancies (2 post-transplant lymphoproliferative disorders (PTLD) and no other cancers) were considered to be totally cured and the corresponding patients were tested 7 (for DF-Tol 1) and 2 (for DF-Tol 5) years after diagnosis. Of importance is that these patients were not the most characteristic in terms of TCR alteration. For instance, DF-Tol 2, without a history of malignancy, had the second strongest global CDR3-LD alteration and the highest Vβ/HPRT ratio. Finally, the listed clinical parameters were also analyzed for their possible influence on CDR3-LD and Vβ/HPRT transcript ratio in drug-free and minimally immunosuppressed patients, either considered together or separately, using Manova with a global t-test (F-test) and related pairwise comparison (see Patients, Materials and Methods). None of these clinical parameters, and particularly age, lymphoma, cancer and corticosteroid therapy, was a determining factor in significantly shaping the blood T-cell alterations in each group.

Sorted T cells from families with strong CDR3-LD alterations in drug-free operationally tolerant recipients accumulate lower levels of cytokine transcripts and higher levels of IL2-R-α than those from patients with chronic rejection

Because operationally tolerant patients displayed strongly biased blood clonal distribution, we investigated the possibility that their blood T cells could have a particular transcriptional pattern for a panel of representative cytokines (IL2, IL10, IL13, TGF-β, IFN-γ and IL8), perforin and IL2-R-α. First, we tested the transcripts of unseparated PBMC from operationally tolerant patients, minimally immunosuppressed patients, patients with chronic rejection and normal individuals and found no significant differences (data not shown). However, because any differences may only concern the selected T-cell populations, we specifically analyzed purified T cells from sorted Vβ families with strong CDR3-LD alteration in DF-Tol patients, in recipients with chronic rejection (Figure 4A) and normal individuals. Whereas most of the sorted T-cell populations from chronic rejection patients accumulated high levels of cytokine transcripts, the corresponding T cells from operationally tolerant recipients did not, particularly for IL2 (p < 0.05), IL8, IL13 (p < 0.05), TGF-β, IFN-γ and perforin (Figure 4B). In this context, the pattern observed in operationally tolerant patients was much closer to that observed in normal individuals than that of chronically rejecting patients. Interestingly, sorted T cells from operationally tolerant recipients accumulated higher levels of IL2-R-α transcripts than the corresponding T cells from chronic rejection patients (p < 0.05). Figure 3D shows that this decreased transcript accumulation (particularly IL2, p < 0.05) was not observed when purified global CD4+ or CD8+ T-cell populations were tested, fitting with the patterns obtained in PBMC and restricting the difference to the T cells having engaged in clonal selection. Thus, despite a strongly selected TCR usage, sorted T cells from operationally tolerant patients had only low but detectable levels of cytokine transcripts (IL2 (p < 0.05), IL13 (p < 0.05), IFN-γ, IL8, perforin, TGF-β and IL10) (Figure 4B) and increased levels of IL2-R-α (p < 0.05) transcripts as compared to CR patients.


Figure 4. Cytokine transcript accumulation in sorted Vβ families from drug-free operationally tolerant patients and recipients with chronic rejection. (A) CDR3-LD of T cells in sorted Vβ families analyzed for cytokine transcript accumulation. (B) Cytokine transcript measurements were performed on sorted families with altered CDR3-LD from DF-Tol patients (n = 7 to 11), CR patients (n = 7) and normal individuals (n = 7–11).

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients, Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

The identification and characterization of patients who have developed operational tolerance to an allotransplant may make it possible to identify other transplant recipients who may no longer require ‘life long’ immunosuppression. A number of isolated drug-free operationally tolerant recipients have been reported with normal or decreased anti-donor T-cell responses (6–9). In human graft recipients, key tests (such as the rejection of a third-party allograft) that have been established in rodents (18,19), cannot be performed. Furthermore, no prospective studies of a possible tolerance-associated pattern can be carried out. Indeed, the 5 drug-free operationally tolerant patients described in this report are the only ones identified in France and there is currently no a priori indication that would enable the screening of a population of ‘potentially tolerant’ patients. However, it is also necessary to consider that immune tolerance has not been characterized in humans up to now and therefore no a priori theory should be accepted in terms of the possible mechanisms involved in such a phenomenon. In this article, we analyzed the blood TCR repertoire and T-cell selection and activation in such operationally tolerant patients with stable graft function many years after immunosuppression interruption. We show that these patients have a unique blood TCR pattern combining a restricted CDR3-LD and abnormal Vβ/HPRT transcript ratios. Furthermore, these peripheral oligoclonal T-cell selection expansions, which are the cellular basis of these findings, were mostly characterized by a lack of cytokine transcript accumulation, contrasting with those from chronically rejecting patients and suggesting a state of hyporesponsiveness. However, these 5 patients did not present symptoms of unusual immunoincompetence against specific human viral antigens after their treatment interruption (or while under monotherapy). In fact, the DF-Tol patients had the lowest incidence of infection (1 HCV and 1 bacterial infection) despite the post-transplant survey often spanning a decade, suggesting that they are specifically tolerant to their grafts.

Despite reports of certain T-cell selections in animal models of tolerance (20–22), including our own work (12,23), we did not expect to find such highly altered CDR3-LD in the blood, years after transplantation, given that previous studies have found TCR alterations in tolerated graft lymphocytes (20,23). To our knowledge, the blood was not studied in these reports. It is possible that the selected T cells found in the blood of operationally tolerant patients are no longer able to enter the graft after such a long time and in the absence of local inflammation. It is also possible that these selected T-cell populations have a specific distribution of cell surface molecules involved in lymphocyte homing. Given that TCR-β chain transcript accumulation is not observed before the onset of responding cell proliferation when T cells are stimulated in vitro by allogeneic APC (N. Degauque et al., unpublished), the high Vβ/HPRT ratios probably developed progressively, in parallel with T-cell expansion, rather than reflecting a short-term activation. In addition, these ratios were not related to an increased percentage of lymphocytes in DF-Tol and Ster blood as compared to CR and stable recipients or normal individuals. In fact, whereas the CD4+/CD8+ ratio remained unchanged, the number of T cells was even lower in the low-dose steroid group of patients (792 ± 551 lymphocytes/μL blood, p < 0.05). Taking this parameter into account would even further increase the already extremely high Vβ/HPRT ratios in such patients. The cause and significance of this TCR transcript accumulation in T cells with restricted CDR3-LD is still unknown but could be linked to memory T-cell clones, fitting with their predominant CD8+ phenotype. However, only 2 recipients (DF-Tol 4, Ster 3) out of 10 drug-free or minimally immunosuppressed patients demonstrated a clinical history of chronic HCV and CMV diseases and the non-compliant DF-Tol 2 patient, who had no chronic viral infection, displayed the second most altered CDR3-LD pattern and the highest Vβ/HPRT ratio. The fact that patients under low-dose steroid monotherapy exhibited a similar pattern to drug-free operationally tolerant patients and that multivariate analysis did not link this pattern to steroids, strongly suggests that the former patients are also operationally tolerant, fitting with their clinical condition. Moreover, multivariate statistical analyses showed an independence between TCR alterations and Vβ/HPRT ratios, lymphoma and cancer history. In this context, these data provide a rationale for the progressive weaning of these patients from this low steroid intake.

It is interesting to note that T cells sorted from families with strongly altered TCR from DF-Tol patients did not accumulate Th1 (IL2, IFN-γ) or Th2 (IL10, IL13) cytokine transcripts. Therefore, on this basis, operationally tolerant recipients display a profile that does not differ from normal individual contrasting with the situation of chronic rejection where the families are bearing selected T cells that seem activated memory T cells. Considering their concomitantly selected CDR3-LD pattern and high Vβ/HPRT transcript ratios, this is compatible with a state of anergy (24) rather than ignorance. The stability of these patterns in time also suggests a continuous selection pressure. In addition, altered CDR3-LD were found to be mostly made up of CD8+ T cells. CD8+ T-cell subsets have been shown to have regulatory properties in several animal models as well as in humans (12,25,26). However, there was also a significantly higher level of CD4+CD25+ T cells (p < 0.05), another T-cell population with putative regulatory properties, in the blood of DF-Tol and Ster patients compared to patients with chronic rejection (20.5 ± 7.3% vs. 10.6 ± 5.2%) (S. Louis et al., submitted for publication). Determining such regulatory activity would however require functional studies using donor cells. Given the unavailability of frozen donor cells from transplantations, the majority of which were performed more than a decade ago and for whom most of the donor typings were performed by serology, this could not be done.

The moderate clonal selection observed in the blood of CR patients, which gives a partial overlap in terms of magnitude with operationally tolerant patients, also mostly concerning CD8+ cells, may be due to the presence of circulating alloreactive cells restricted to donor determinants, as previously reported in chronic rejection (11,27). This possibility is sustained by the accumulation of cytokine transcripts in CD8+ T cell from families with altered CDR3-LD in this group. In contrast, no accumulation was observed in families with altered CDR3-LD from operationally tolerant patients. Interestingly, 1 of 7 patients of the Sta group (StaCNI+ 3) displayed a pattern that could fit with the profile observed in operationally tolerant recipients. Thus, theoretically speaking, this group is not expected to be homogeneous.

Our data also address the question of the relationship of these altered T-cell clonal patterns with the CD8+ or CD4+ clonal expansions observed in elderly individuals. Indeed, ageing has been shown to be associated with persisting blood CD8+ T-cell expansions that are hyporesponsive or anergic (28–34). Nevertheless, the strong CDR3-LD alteration observed in the drug-free and low-dose steroid groups could not be explained by the age of the patients in the multivariate analysis. It is nevertheless possible that operationally tolerant recipients are also those who developed the most prominent clonal expansions because of a unique recruitment of naive T cells into the memory pools, combined with a lack of replacement of naive T cells from the thymus (35), a process that is age-dependent. This gives grounds for the hypothesis that the mechanisms involved in this ‘naturally’ occurring tolerance are much more complex than a pure active suppression and may also be related to a state of homeostatic competition where amplified clones with ‘simplified’ CDR3-LD induce a reduction of the CD8+ repertoire and a decreased alloreactivity. Interestingly, CD8+ T cells have been suggested to play a role in chronic rejection (36). This mechanism does not therefore exclude the possible presence of regulatory cells in the CD4+ population, as suggested by the higher numbers of CD25+CD4+ cells in these patients compared to patients with chronic rejection. Our data thus open up a new field of exploration in kidney recipients with stable function. However, we are conscious that the small number of ‘tolerant’ patients require confirmation of the data. This confirmation, in a larger study, of patterns emulating those observed in operationally tolerant patients, may serve, together with other biological markers (10,37), as a possible surrogate indicator of tolerance in these patients.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients, Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

The authors thank Prof. C. Legendre (Hôpital St Louis, Paris, France) for providing samples and the clinical history of DF-Tol 4 patient, Dr. P. Richardot (DM consultant, Paris, France), Dr. J.C. Doré (MNHN, Paris, France), Dr. J. Veziers (INSERM U643, Nantes, France) and Dr. J.M. Bouler (Institut des Biomateriaux, Nantes, France) for help in statistical analyses and Dr. J. Ashton-Chess for editing the manuscript.

This work was partially supported by the ‘Fondation Progreffe’, the ‘Etablissement Français des Greffes’ and a ROTRF (Roche) grant (559688078). The study of the TCR in normal individuals was performed as a project of the Immune Tolerance Network (T00S563254), a collaborative clinical research project headquartered at the University of California, San Francisco and supported by the National Institute of Allergy and Infectious Diseases, the National Institute of Diabetes, Digestive and Kidney Disease and the Juvenile Diabetes Research Foundation.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients, Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  • 1
    Hariharan S, Johnson CP, Bresnahan BA, Taranto SE, McIntosh MJ, Stablein D. Improved graft survival after renal transplantation in the United States, 1988 to 1996. N Engl J Med 2000; 342: 605612.DOI: 10.1056/NEJM200003023420901
  • 2
    Opelz G. Influence of treatment with cyclosporine, azathioprine and steroids on chronic allograft failure. The Collaborative Transplant Study. Kidney Int Suppl 1995; 52: S89S92.
  • 3
    Soulillou JP, Giral M. Controlling the incidence of infection and malignancy by modifying immunosuppression. Transplantation 2001; 72(12 Suppl.): S89S93.DOI: 10.1097/00007890-200107150-00018
  • 4
    Giral M, Pascuariello G, Karam G et al. Acute graft pyelonephritis and long-term kidney allograft outcome. Kidney Int 2002; 61: 18801886.DOI: 10.1046/j.1523-1755.2002.00323.x
  • 5
    Dantal J, Hourmant M, Cantarovich D et al. Effect of long-term immunosuppression in kidney-graft recipients on cancer incidence: randomised comparison of two cyclosporin regimens. Lancet 1998; 351: 623628.DOI: 10.1016/S0140-6736(97)08496-1
  • 6
    Strober S, Benike C, Krishnaswamy S, Engleman EG, Grumet FC. Clinical transplantation tolerance twelve years after prospective withdrawal of immunosuppressive drugs: studies of chimerism and anti-donor reactivity. Transplantation 2000; 69: 15491554.DOI: 10.1097/00007890-200004270-00005
  • 7
    Jacobsen N, Taaning E, Ladefoged J, Kristensen JK, Pedersen FK. Tolerance to an HLA-B, DR disparate kidney allograft after bone-marrow transplantation from same donor. Lancet 1994; 343: 800.DOI: 10.1016/S0140-6736(94)91881-3
  • 8
    Burlingham WJ, Grailer AP, Fechner JH, Jr. et al. Microchimerism linked to cytotoxic T lymphocyte functional unresponsiveness (clonal anergy) in a tolerant renal transplant recipient. Transplantation 1995; 59: 11471155.
  • 9
    Zoller KM, Cho SI, Cohen JJ, Harrington JT. Cessation of immunosuppressive therapy after successful transplantation: a national survey. Kidney Int 1980; 18: 110114.
  • 10
    VanBuskirk AM, Burlingham WJ, Jankowska-Gan E et al. Human allograft acceptance is associated with immune regulation. J Clin Invest 2000; 106: 145155.
  • 11
    Ciubotariu R, Liu Z, Colovai AI et al. Persistent allopeptide reactivity and epitope spreading in chronic rejection of organ allografts. J Clin Invest 1998; 101: 398405.
  • 12
    Douillard P, Pannetier C, Josien R et al. Donor-specific blood transfusion-induced tolerance in adult rats with a dominant TCR-Vbeta rearrangement in heart allografts. J Immunol 1996; 157: 12501260.
  • 13
    Brouard S, Vanhove B, Gagne K et al. T cell repertoire alterations of vascularized xenografts. J Immunol 1999; 162: 33673377.
  • 14
    Pannetier C, Even J, Kourilsky P. T cell repertoire diversity and clonal expansions in normal and clinical samples. Immunol Today 1995; 16: 176181.DOI: 10.1016/0167-5699(95)80117-0
  • 15
    Gorochov G, Neumann AU, Kereveur A et al. Perturbation of CD4+ and CD8+ T cell repertoires during progression to AIDS and regulation of the CD4+ repertoire during antiviral therapy. Nat Med 1998; 4: 215221.DOI: 10.1038/nm0298-215
  • 16
    Sebille F, Gagne K, Guillet M et al. Direct recognition of foreign MHC determinants by naive T cells mobilizes specific Vbeta families without skewing of the complementarity-determining region 3 length distribution. J Immunol 2001; 167: 30823088.
  • 17
    Guillet M, Sebille F, Soulillou J. TCR usage in naive and committed alloreactive cells: implications for the understanding of TCR biases in transplantation. Curr Opin Immunol 2001; 13: 566571.DOI: 10.1016/S0952-7915(00)00260-0
  • 18
    Josien R, Heslan M, Brouard S, Soulillou JP, Cuturi MC. Critical requirement for graft passenger leukocytes in allograft tolerance induced by donor blood transfusion. Blood 1998; 92: 45394544.
  • 19
    Waldmann H, Cobbold S. Regulating the immune response to transplants. a role for CD4+ regulatory cells? Immunity 2001; 14: 399406.DOI: 10.1016/S1074-7613(01)00120-0
  • 20
    Zhai Y, Li J, Hammer M, Busuttil RW, Volk HD, Kupiec-Weglinski JW. Evidence of T cell clonality in the infectious tolerance pathway: implications toward identification of regulatory T cells. Transplantation 2001; 71: 17011708.DOI: 10.1097/00007890-200106270-00001
  • 21
    Baron C, McMorrow I, Sachs DH, LeGuern C. Persistence of dominant T cell clones in accepted solid organ transplants. J Immunol 2001; 167: 41544160.
  • 22
    Baron C, Sachs DH, LeGuern C. A particular TCR beta variable region used by T cells infiltrating kidney transplants. J Immunol 2001; 166: 25892596.
  • 23
    Guillet M, Brouard S, Gagne K et al. Different qualitative and quantitative regulation of Vbeta TCR transcripts during early acute allograft rejection and tolerance induction. J Immunol 2002; 168: 50885095.
  • 24
    Schwartz RH. Models of T cell anergy: is there a common molecular mechanism? J Exp Med 1996; 184: 18.DOI: 10.1084/jem.184.1.1
  • 25
    Douillard P, Vignes C, Josien R et al. Reassessment of the role of CD8+ T cells in the induction of allograft tolerance by donor-specific blood transfusion. Eur J Immunol 1999; 29: 19191924.DOI: 10.1002/(SICI)1521-4141(199906)29:06<1919::AID-IMMU1919>3.3.CO;2-#
  • 26
    Ciubotariu R, Vasilescu R, Ho E et al. Detection of T suppressor cells in patients with organ allografts. Hum Immunol 2001; 62: 1520.
  • 27
    Vella JP, Spadafora-Ferreira M, Murphy B et al. Indirect allorecognition of major histocompatibility complex allopeptides in human renal transplant recipients with chronic graft dysfunction. Transplantation 1997; 64: 795800.DOI: 10.1097/00007890-199709270-00001
  • 28
    Schwab R, Szabo P, Manavalan JS et al. Expanded CD4+ and CD8+ T cell clones in elderly humans. J Immunol 1997; 158: 44934499.
  • 29
    Ku CC, Kappler J, Marrack P. Characteristics and maintenance of CD8+ T cell clones found in old mice. Mech Ageing Dev 1997; 94: 4153.DOI: 10.1016/S0047-6374(96)01865-9
  • 30
    Ku CC, Kappler J, Marrack P. The growth of the very large CD8+ T cell clones in older mice is controlled by cytokines. J Immunol 2001; 166: 21862193.
  • 31
    Colombatti A, Doliana R, Schiappacassi M, et al. Age-related persistent clonal expansions of CD28- cells: phenotypic and molecular TCR analysis reveals both CD4+ and CD4+CD8+ cells with identical CDR3 sequences. Clin Immunol Immunopathol 1998; 89: 6170.
  • 32
    Ricalton NS, Roberton C, Norris JM, Rewers M, Hamman RF, Kotzin BL. Prevalence of CD8+ T cell expansions in relation to age in healthy individuals. J Gerontol A Biol Sci Med Sci 1998; 53: B196B203.
  • 33
    Halapi E, Jeddi-Tehrani M, Blucher A et al. Diverse T-cell receptor CDR3 length patterns in human CD4+ and CD8+ T lymphocytes from newborns and adults. Scand J Immunol 1999; 49: 149154.DOI: 10.1046/j.1365-3083.1999.00469.x
  • 34
    Ginaldi L, De Martinis M, D'Ostilio A, Marini L, Loreto MF, Quaglino D. Immunological changes in the elderly. Aging (Milano) 1999; 11: 281286.
  • 35
    Marrack P, Lo D, Brinster R et al. The effect of thymus environment on T cell development and tolerance. Cell 1988; 53: 627634.
  • 36
    Strom TB, Tilney NL, Carpenter CB, Busch GJ. Identity and cytotoxic capacity of cells infiltrating renal allografts. N Engl J Med 1975; 292: 12571263.
  • 37
    Thomson AW, Mazariegos GV, Reyes J et al. Monitoring the patient off immunosuppression. Conceptual framework for a proposed tolerance assay study in liver transplant recipients. Transplantation 2001; 72(8 Suppl.): S13S22.DOI: 10.1097/00007890-200107150-00006