Monitoring of CD4⁺CD25^highIL-7Rα^high activated T Cells in Kidney Transplant Recipients

Abstract

Summary

Background and objectives

In humans, circulating CD4⁺CD25^high T cells contain mainly regulatory T cells (Treg; FoxP3⁺IL-7Rα^low), but a small subset is represented by activated effector T cells (Tact; FoxP3⁻IL-7Rα^high). The balance between Tact and Treg may be important after transplantation. The aim of this study was first to analyze and correlate CD4⁺CD25^high Tact and Treg with the clinical status of kidney transplant recipients and second to study prospectively the effect of two immunosuppressive regimens on Tact/Treg during the first year after transplantation.

Design, setting, participants, & measurements

CD4⁺CD25^high Tact and Treg were analyzed by flow cytometry, either retrospectively in 90 patients greater than 1 year after kidney transplantation (cross-sectional analysis) or prospectively in 35 patients receiving two immunosuppressive regimens after kidney transplantation (prospective analysis).

Results

A higher proportion of Tact and a lower proportion of Treg were found in the majority of kidney recipients. In chronic humoral rejection, a strikingly higher proportion of Tact was present. A subgroup of stable recipients receiving calcineurin inhibitor–free immunosuppression (mycophenolate mofetil, azathioprine, or sirolimus) had Tact values that were similar to healthy individuals. In the prospective analysis, the proportion of Tact significantly increased in both immunosuppression groups during the first year after transplantation.

Conclusions

These data highlight distinct patterns in the proportion of circulating Tact depending on the clinical status of kidney recipients. Moreover, the prospective analysis demonstrated an increase in the proportion of Tact, regardless of the immunosuppressive regimen. The measurement of Tact, in addition to Treg, may become a useful immune monitoring tool after kidney transplantation.

Introduction

Regulatory T cells (Treg) have been described as specialized T lymphocytes that are able to suppress immune responses to self- and nonself-antigens. Multiple populations of Treg have been described (1), including so-called “natural Treg,” a specific subset of T cells coexpressing CD4 and high levels of the IL-2 receptor α chain (CD25) (2). Subsequently, Treg were also shown to play an important role in the development and maintenance of transplantation tolerance in experimental models (3). In humans, circulating Treg have been shown to inhibit anti-donor immune responses in clinically stable transplant recipients (4–6). In addition to CD4 and CD25, Treg are characterized by the constitutive expression of L-selectin (CD62L) (7), cytotoxic T lymphocyte-associated antigen-4 (8), and glucocorticoid-induced TNF receptor (9), as well as by the intracellular expression of the transcription factor forkhead box P3 (FoxP3) (10,11).

In 2006, the expression of the IL-7 receptor α chain (IL-7Rα or CD127) was reported to inversely correlate with FoxP3 expression and Treg suppressive capacity in healthy individuals (12,13), thus allowing for the distinction between two CD4⁺CD25^high T cell populations, namely Treg (IL-7Rα^lowFoxP3⁺) and activated effector T cells (Tact; IL-7Rα^highFoxP3⁻). Subsequently, we observed that circulating Tact were expanded in liver and kidney transplant recipients (14,15), were able to secrete pro-inflammatory cytokines, and constituted approximately 50% of the CD4 T cells infiltrating rejecting renal allografts, suggesting that this population may play a role in the rejection process (14). Moreover, HCV infection after liver transplantation appears to negatively modulate the proportion of circulating Tact (15). Interestingly, a recent report indicates that Treg can display an upregulated IL-7Rα expression and an inadequate suppressive capacity under specific conditions, e.g., in heart transplant recipients experiencing acute rejection (AR) (16).

The aim of this study was to precisely delineate the phenotype of circulating CD4⁺CD25^high T cell populations (Tact and Treg) in kidney transplant recipients (KTx) with various clinical conditions (first part: cross-sectional analysis), as well as the pattern of these T cell populations during the first year after kidney transplantation and the possible effect on Tact of two distinct immunosuppressive regimens (IS) (second part: prospective analysis).

Materials and Methods

Patients

From January 2005 to January 2009, a total of 125 KTx were enrolled at the Centre Hospitalier Universitaire Vaudois (CHUV), Lausanne, Switzerland, and at the Hôpitaux Universitaires de Genève (HUG), Geneva, Switzerland (Lausanne-Geneva Transplant Network). All of the patients gave informed written consent before participating in the study, which was approved by both local institutional review boards.

Cross-sectional Analysis.

Ninety patients were studied at least 12 months after kidney transplantation and were divided into four groups, according to their clinical status and IS (Table 1): (1) stable patients with standard IS (n = 54): defined as patients having a stable graft function (stable serum creatinine over the last 6 months with values less than 150 μmol/L, 24-hour proteinuria inferior to 0.5 g/d, no circulating donor-specific antibodies as detected by the Luminex assay [17,18]); they were receiving a standard double- or triple-drug maintenance IS, which included a calcineurin inhibitor (CNI) and mycophenolate mofetil (MMF) with or without steroids; (2) stable CNI-free patients (n = 23): defined as patients having a stable graft function (see above) and receiving low-dose CNI-free maintenance IS (prednisone alone, or MMF ± prednisone, or azathioprine ± prednisone, or no IS at all in one patient); in this group, the IS had been “tailored” toward minimization because of medical reasons (e.g., nephrotoxicity of CNIs, neuro-psychologic disorders); (3) patients with chronic humoral rejection (CHR) (n = 7): defined as patients having progressive allograft dysfunction (rise of creatinine and/or proteinuria >0.5 g/d) over recent months, evidence of circulating donor-specific antibodies and a biopsy-proven diagnosis of CHR (on the basis of the Banff '07 criteria (19)), with detection of C4d in peritubular capillaries; and (4) sirolimus-treated CNI-free stable patients (n = 6): defined as patients having a stable graft function (see above) and receiving CNI-free sirolimus-based IS. Seventy-three healthy volunteers were enrolled as controls.

Table 1.

Cross-sectional analysis: baseline and clinical characteristics of kidney transplant recipients (n = 90)

	All Kidney Transplant Recipients (n = 90)	Stable Standard IS (n = 54)	CHR (n = 7)	Stable CNI-free IS (n = 23)	Stable Sirolimus (n = 6)
General characteristics
recipient gender (males/females)	59/31	38/16	2/5	15/8	4/2
mean recipient age at transplant (years)	44.2 (3 to 69)	47.3 (12 to 69)	28.6 (3 to 43)	41.5 (19 to 63)	45.7 (20 to 63)
donor organ source
deceased	63 (70%)	32 (59%)	5 (71%)	21 (91%)	5 (83%)
living, related	17 (19%)	14 (26%)	2 (29%)	0	1 (17%)
living, unrelated	10 (11%)	8 (15%)	0	2 (9%)	0
retransplants	6 (7%)	0	0	5 (21%)	1 (17%)
Immunologic characteristics
HLA mismatches between recipient and donor
HLA-A	1.21	1.19	1.24	1.40	na
HLA-B	1.36	1.37	1.41	1.32	na
HLA-DR	1.30	1.27	1.39	1.34	na
Clinical characteristics
serum creatinine at the time of the study (μmol/L) (mean ± SD)	137 ± 62	136 ± 51	207 ± 139	112 ± 27	158 ± 30
immunosuppression at the time of the study
CNI (CsA or TAC)	61 (68%)	54 (100%)	7 (100%)	0	0
sirolimus	6 (8%)	0	0	0	6 (100%)
MMF	61 (68%)	44 (81%)	3 (43%)	12 (52%)	4 (67%)
azathioprine	14 (16%)	4 (7%)	0	9 (39%)	1 (17%)
prednisone	46 (51%)	20 (37%)	4 (57%)	19 (83%)	3 (50%)

CHR, chronic humoral rejection; IS, immunosuppressive regimen; CNI, calcineurin inhibitor; CsA, cyclosporin A; TAC, tacrolimus; MMF, mycophenolate mofetil; na, not available.

Prospective Analysis.

Thirty-five patients were enrolled, all receiving a first kidney transplant (Table 2). Blood samples were taken before transplantation and at 3, 6, and 12 months after kidney transplantation. Two clinical protocols were used: (1) THYMO group (n = 19): this group received four doses of rabbit anti-thymocyte globulin (Thymoglobulin®, Genzyme Transplant, Cambridge, MA) from days 0 to 3 (1.5 mg/kg per day, each day), with intravenous methylprednisolone boluses (from 500 mg on day 0 to 125 mg on day 3) administered before thymoglobulin, followed by a steroid-free maintenance IS (tacrolimus and mycophenolic acid) and (2) BSX group (n = 16): this group received two 20-mg doses of basiliximab (Simulect, Novartis Pharmaceuticals, Basel, Switzerland), on days 0 and 4, with a maintenance IS consisting of tacrolimus, MMF, and prednisone.

Table 2.

Prospective analysis: baseline and clinical characteristics of kidney transplant recipients (n = 35)

	Kidney Transplant Recipients
	THYMO Group (n = 19)	BSX Group (n = 16)
General characteristics
recipient gender (males/females)	13/6	9/7
mean recipient age at transplant (years)	42.6 (19 to 81)	49.7 (27 to 78)
donor organ source
cadaveric	0	3 (19%)
living, related	13 (68%)	6 (38%)
living, unrelated	6 (32%)	7 (44%)
Immunologic characteristics
HLA mismatches between recipient and donor
HLA-A	0.84	1.29
HLA-B	1.00	1.36
HLA-DR	0.89	1.15
Clinical characteristics
serum creatinine at the time of the study (μmol/L) (mean ± SD)
before transplant	654.1 ± 297.6	544.9 ± 205.7
at 6 months after transplant	124.6 ± 34.3	130.6 ± 40.1
acute rejection episode within 6 months after transplant	2 (10.5%)	2 (12.5%)
patient survival at 6 months after transplant	19 (100%)	16 (100%)
graft survival at 6 months after transplant	19 (100%)	16 (100%)

Flow Cytometry

Monoclonal antibodies used for surface staining included CD4-PerCPCy5.5 (Becton Dickinson, Franklin, NJ), CD25-APC (BD Pharmingen, San Diego, CA), CD45RO-FITC (BD Pharmingen), CD45RO-ECD (Immunotech, Beckman-Coulter, Marseille, France), and IL-7Rα-PE (Immunotech). 5 × 10⁵ fresh or cryopreserved peripheral blood mononuclear cells (PBMC) were incubated with appropriately titered monoclonal antibodies for 30 minutes at 4°C, washed, and fixed in 300 μl of CellFix (Becton Dickinson) until FACS analysis. For intracellular staining, PBMC were fixed and permeabilized with fixation/permeabilization buffers (eBioscience, San Diego, CA) according to the manufacturer's instructions after staining of cell surface markers, and stained with FoxP3-FITC (eBioscience). Flow cytometric analyses were performed on a FACSCalibur or LSRII (Becton Dickinson). At least 1 × 10⁵ events were recorded in the lymphocyte gate on the basis of forward scatter/sideward scatter parameters.

Gating Strategy

As described before (14,15), PBMC were gated with forward scatter and sideward scatter (Figure 1A). From this gate, cells with a CD4⁺CD25^high phenotype were gated (usually, there is a slightly less CD4 expression in this population) (20) (Figure 1B). On the basis of their expression of FoxP3, IL-7Rα, and CD45RO, Tact were defined as CD45RO⁺IL-7Rα^high cells (Figure 1C), and Treg were defined as FoxP3⁺IL-7Rα^low cells (Figure 1D), within CD4⁺CD25^high cells.

Figure 1.

Gating strategy for CD4⁺CD25^high cells, activated effector T cells (Tact), and regulatory T cells (Treg). (A) Mononuclear cells were gated with forward scatter (FSC) and sideward scatter (SSC). (B) From the gate in panel A, cells with a CD4⁺CD25^high phenotype were gated (usually there is slightly less CD4 expression in this population). (C) On the basis of its expression of CD45RO and IL-7Rα, the Tact subpopulation was defined as CD45RO⁺IL-7Rα^high cells from the gate in panel B. (D) On the basis of its expression of FoxP3 and IL-7Rα, the Treg subpopulation was defined as FoxP3⁺IL-7Rα^low cells from the gates in panels B and C. The level for considering positivity for FoxP3 was established on the total CD4⁺ cells.

Statistical Analyses

Statistical significance was calculated by the two-tailed t test. P < 0.05 was considered significant.

Results

Cross-sectional Analysis

Clinical Characteristics.

The demographics of the 90 KTx are summarized in Table 1.

Proportion of CD4⁺ and CD4⁺CD25^high T Cells.

The percentage of total CD4⁺ T cells among lymphocytes was found to be significantly lower in the 90 KTx (mean ± SEM, 42.80% ± 1.35%) as compared with the 73 healthy controls (49.30% ± 1.71%; P < 0.005). Then we analyzed the overall CD4⁺CD25^high T cells as well as the CD4⁺CD25^high T cell subpopulations (Treg and Tact) in 84 KTx (the results of the six stable KTx on sirolimus are presented below) and in the 73 controls. The overall percentage of CD25^high cells among CD4⁺ T cells was significantly reduced in the 84 KTx (1.06% ± 0.08%) as compared with controls (1.62% ± 0.16%; P < 0.005) (Figure 2). When analyzing the various study groups, the percentage of CD4⁺CD25^high T cells between the groups was not statistically different; they were found to represent 0.99% ± 0.10% of CD4⁺ T cells in stable KTx on standard IS (n = 54), 1.19% ± 0.14% in stable patients on CNI-free IS (n = 23), and 0.82% ± 0.15% in patients with CHR (n = 7).

Figure 2.

Percentage of CD4⁺CD25^high T cells in kidney transplant recipients (mean percentage ± SEM). The percentage of CD25^high cells in CD4⁺ T cells in all KTx was significantly lower as compared with healthy controls (1.06% ± 0.08% versus 1.62% ± 0.16%; P < 0.005).

Distribution of CD4⁺CD25^high T Cell Subpopulations.

In controls, as expected, the vast majority of CD4⁺CD25^high T cells were Treg, defined as FoxP3⁺IL-7Rα^low cells in CD4⁺CD25^high T cells (74.05% ± 2.01%). By contrast, the proportion of Treg was found to be significantly lower in the 84 KTx (49.23% ± 2.06%; P < 0.001) (Figure 3A); no significant statistical differences were found between stable patients on standard IS (50.62% ± 2.13%), stable patients on CNI-free IS (45.12% ± 9.33%), and patients with CHR (50.50% ± 4.15%) (Figure 3B).

Figure 3.

Percentage of regulatory T cells (Treg; CD4⁺ CD25^highFoxP3⁺IL-7Rα^low cells) in kidney transplant recipients (mean percentage ± SEM). (A) The percentage of FoxP3⁺IL-7Rα^low cells in CD4⁺CD25^high T cells (Treg) in all KTx was significantly lower as compared with healthy controls (49.23% ± 2.06% versus 74.05% ± 2.01%; P < 0.001). (B) The percentage of FoxP3⁺IL-7Rα^low cells in CD4⁺CD25^high T cells (Treg) in stable KTx with standard calcineurin inhibitor (CNI)–based immunosuppressive regimen (IS) (50.62% ± 2.13%; P < 0.001), in stable KTx with CNI-free IS (45.12% ± 9.33%; P < 0.001), and in KTx with CHR (50.50% ± 4.15%; P < 0.001) was significantly and similarly lower as compared with healthy controls (74.05% ± 2.01%).

On the other hand, the proportion of Tact, defined as CD45RO⁺IL-7Rα^high cells in CD4⁺CD25^high T cells, was found to be significantly higher in the 84 KTx (14.30% ± 0.92%) as compared with controls (5.97% ± 0.36%; P < 0.001) (Figure 4A). A strikingly higher percentage was found in KTx with CHR (25.32% ± 4.66%; P < 0.001). Of note, stable KTx on standard CNI-based IS had also significantly higher values (15.81% ± 0.94%; P < 0.001). By contrast, the group of stable KTx on CNI-free IS displayed no significant expansion of the Tact population (7.40% ± 0.75%; P = 0.08) (Figure 4B). Of note, when comparing Tact values in the subgroup of 12 KTx with chronic renal dysfunction (serum creatinine level more than 160 μmol/L), but with no evidence of humoral rejection, to the seven KTx with CHR, the results indicated that KTx with CHR had a significantly higher proportion of Tact (25.32% ± 4.66% versus 14.61% ± 1.72%; P = 0.01) (data not shown).

Figure 4.

Percentage of activated effector T cells (Tact; CD4⁺CD25^highCD45RO⁺IL-7Rα^high cells) in kidney transplant recipients (mean percentage ± SEM). (A) The percentage of CD45RO⁺IL-7Rα^high cells in CD4⁺CD25^high T cells (Tact) in all KTx was significantly higher as compared with healthy controls (14.30% ± 0.92% versus 5.97% ± 0.36%; P < 0.001). (B) The percentage of CD45RO⁺IL-7Rα^high cells in CD4⁺CD25^high T cells (Tact) in stable KTx with a standard calcineurin inhibitor (CNI)–based immunosuppressive regimen (IS) (15.81% ± 0.94%; P < 0.001) and in KTx with chronic humoral rejection (CHR; 25.32% ± 4.66%; P < 0.001) was significantly higher as compared with healthy controls (5.97% ± 0.36%); by contrast, it was comparable to healthy controls in stable KTx with CNI-free IS (7.40% ± 0.75%; P = 0.08).

Sirolimus-treated CNI-free Stable KTx.

A group of six sirolimus-treated CNI-free stable KTx was also studied. The overall percentage of CD4⁺CD25^high T cells was not statistically different in sirolimus-treated CNI-free stable KTx as compared with controls (1.72% ± 0.46% versus 1.62% ± 0.16%; P = 0.41) or with stable patients on CNI-free IS (1.19% ± 0.14%; P = 0.18), but it was statistically higher than in stable KTx on standard CNI-based IS (0.99% ± 0.10%; P < 0.05) (Figure 5A), as reported recently (21–23). The proportion of Treg was statistically lower as compared with controls (62.48% ± 4.88% versus 74.05% ± 2.01%; P < 0.05), although to a lesser extent than observed in stable CNI-treated KTx (50.62% ± 2.13%; P < 0.05) and in stable patients on CNI-free IS (45.12% ± 9.33%; P < 0.05) (Figure 5B). Interestingly, the proportion of Tact was found to be lower in sirolimus-treated CNI-free stable KTx (3.41% ± 0.74%) as compared with controls (5.97% ± 0.36%; P = 0.05) and with stable patients on CNI-free IS (7.40% ± 0.75%; P < 0.05), and much lower than in stable KTx on standard IS (15.81% ± 0.94%; P < 0.001) (Figure 5C). In support of these findings, we had the opportunity to measure the Tact population in a stable KTx before and after switching his IS from a CNI (tacrolimus) to sirolimus. Under CNI-based IS, Tact represented 20.45% of CD4⁺CD25^high T cells, but it decreased strikingly to 2.34% 9 months after the switch to a CNI-free IS with sirolimus (data no shown).

Figure 5.

Distribution of CD4⁺CD25^high T cells in sirolimus-treated CNI-free stable kidney transplant recipients (mean percentage ± SEM). (A) The percentage of CD25^high cells in CD4⁺ T cells in sirolimus-treated calcineurin inhibitor (CNI)–free stable KTx (1.72% ± 0.46%) was statistically comparable to healthy controls (1.62% ± 0.16%; P = 0.41) and to stable patients on CNI-free immunosuppressive regimen (IS) (1.19% ± 0.14%; P = 0.18), but it was statistically higher than in stable KTx on standard CNI-based IS (0.99% ± 0.10; P < 0.05). (B) The percentage of FoxP3⁺IL-7Rα^low cells in CD4⁺CD25^high T cells (Treg) in sirolimus-treated CNI-free stable KTx was significantly lower as compared with healthy controls (62.48% ± 4.88% versus 74.05% ± 2.01%; P < 0.05) but to a lesser extent than observed in stable CNI-treated KTx (50.62% ± 2.13%;P < 0.05) and in stable patients on CNI-free IS (45.12% ± 9.33%; P < 0.05). (C) The percentage of CD45RO⁺IL-7Rα^high cells in CD4⁺CD25^high T cells (Tact) in sirolimus-treated CNI-free stable KTx was significantly lower as compared with healthy controls (3.41% ± 0.74% versus 5.97% ± 0.36%; P = 0.05) and with stable patients on CNI-free IS (7.40% ± 0.75%; P < 0.05) and much lower than in stable KTx on standard IS (15.81% ± 0.94%; P < 0.001).

Absolute Counts of Treg and Tact.

We also assessed the mean absolute cell counts of Treg and Tact. As compared with healthy individuals (7.54 ± 1.09 cells/μl of blood), the absolute Treg cell counts were significantly lower in stable KTx on CNI-based IS (3.29 ± 0.50 cells/μl; P < 0.001), and they tended to be lower in KTx with CHR (2.92 ± 0.69 cells/μl; P = 0.07), as well as in sirolimus-treated CNI-free stable KTx (2.75 ± 0.71 cells/μl; P = 0.06), but not in stable CNI-free KTx (5.56 ± 0.42 cells/μl; P = 0.28) (data not shown).

We also assessed the mean absolute cell counts of Tact. Tact absolute cell counts in stable KTx on standard CNI-based IS (1.26 ± 0.25 cells/μl; P = 0.09), as well as in KTx with CHR (1.62 ± 0.40 cells/μl; P < 0.01), were higher as compared with healthy individuals (0.79 ± 0.13 cells/μl). Conversely, in stable KTx on CNI-free IS (0.55 ± 0.11 cells/μl; P = 0.11), as well as in sirolimus-treated KTx (0.33 ± 0.16 cells/μl; P = 0.08), Tact absolute cell counts were comparable to the values found in healthy individuals (data not shown).

Prospective Analysis

Clinical Characteristics.

The demographics of the 35 patients are summarized in Table 2. AR episodes occurred in two patients out of 19 (10.5%) in the THYMO group (both cell-mediated) and in two patients out of 16 (12.5%) in the BSX group (one humoral and one cell-mediated). All of these AR episodes occurred within the first 3 months after transplantation and were treated successfully with a short course of high-dose corticosteroids; serum creatinine levels at 6 months after transplantation were not statistically different between recipients with or without AR.

Treg Subpopulation.

Before transplantation, there was no statistical difference in the proportion of the Treg population between the THYMO group (74.29% ± 4.26%) and the BSX group (73.48% ± 3.42%; P = 0.24) and as compared with controls (74.05% ± 2.01%; P = 0.27 and P = 0.21, respectively). Three to 6 months after transplantation, the proportion of Treg cells decreased significantly in both study groups to a comparable extent. At 6 months, Treg represented 43.01% ± 3.55% of CD4⁺CD25^high T cells in the THYMO group and 38.45% ± 1.33% in the BSX group (P = 0.36). At 12 months, the proportion of Treg in the THYMO group had returned to “normal” values (65.36% ± 1.43%), i.e., not significantly different from that found in controls (P = 0.48), whereas it remained decreased in the BSX group (45.58%) (Figure 6). Similar results were obtained when considering the absolute count of Treg at 3, 6, and 12 months after transplantation, i.e., there were no significant differences between the two groups (data not shown).

Figure 6.

Evolution of the percentage of regulatory T cells (Treg; CD4⁺CD25^highFoxP3⁺IL-7Rα^low cells) in kidney transplant recipients (mean percentage ± SEM). The percentage of FoxP3⁺IL-7Rα^low cells in CD4⁺CD25^high T cells (Treg) decreased significantly during the first year after renal transplantation, in both THYMO and BSX study groups, as compared with pretransplant values.

Tact Subpopulation.

Before transplantation, the proportion of Tact was comparable in the THYMO group (5.94% ± 0.79%) and the BSX group (7.66% ± 0.91%; P = 0.10) and as compared with controls (5.97% ± 0.36%; P = 0.44 and P = 0.09, respectively). As soon as 3 months after transplantation, the population of Tact was significantly expanded in both groups. In the THYMO group, Tact represented 15.27% ± 1.51% of CD4⁺CD25^high T cells, whereas they represented 19.39% ± 1.70% of CD4⁺CD25^high T cells in the BSX group (P < 0.001 as compared with pretransplant values, but NS between the groups). Tact remained expanded between 3 and 12 months after transplantation, to a comparable extent in both groups (Figure 7). Of note, the proportion of Tact found at 12 months in the prospective analysis was in the same range as that found in stable KTx on standard IS in the cross-sectional part of the study (Figure 3B). Considering mean absolute cell counts, Tact were found to be lower after transplantation in the THYMO group, remaining significantly below pretransplant values during the 12 months of observation. In the BSX group, Tact remained stable at 3 months after transplantation, as compared with pretransplant values. However, Tact absolute cell counts were comparable to those found in the THYMO group at 6 and 12 months (data not shown).

Figure 7.

Evolution of the percentage of activated effector T cells (Tact; CD4⁺CD25^highCD45RO⁺IL-7Rα^high cells) in kidney transplant recipients (mean percentage ± SEM). The percentage of CD45RO⁺IL-7Rα^high cells in CD4⁺CD25^high T cells (Tact) increased significantly during the first year after renal transplantation, in both THYMO and BSX study groups, as compared with pretransplant values.

We could also assess the proportion of Tact and Treg in one patient of the THYMO group who developed a cell-mediated AR episode at day 40 after transplantation. At that time, Tact were strikingly increased (60.18%), as compared with pretransplant value (5.82%); conversely, Treg were found to be very low (21.74%), as compared with pretransplant value (84.77%). This AR episode was successfully reversed by corticosteroid boluses, and the proportion of Tact returned to 15.26% at 3 months and 17.28% at 12 months, i.e., in the same range as KTx on standard IS. Nevertheless, the value of Tact observed at the time of AR was the highest value recorded in this study (data not shown). Finally, there was no correlation between the Treg and Tact proportions and serum creatinine levels at 3, 6, or 12 months after transplantation.

Discussion

The aim of this study was to analyze and correlate the proportions of CD4⁺CD25^high subpopulations (namely Tact and Treg) with the clinical status of KTx greater than 1 year after transplantation (cross-sectional analysis) and to study the effect of two distinct IS regimens on Tact and Treg during the first year after transplantation (prospective analysis).

In the first analysis, we found that the overall percentages of CD4⁺CD25^high T cells as well as of Treg were significantly lower in all KTx to a comparable extent with the exception of KTx receiving sirolimus-based CNI-free IS. This lower proportion is probably due to the direct inhibitory effect of IS drugs, such as CNI, on the activation of T cells, and only sirolimus seemed to relatively preserve Treg, as described previously by others (21–26). By contrast, a significantly higher proportion of Tact was found in KTx, except those in the CNI-free IS group and those on sirolimus-based CNI-free therapy. The higher proportion of Tact was correlated with the clinical status, i.e., the highest percentage of Tact was found in patients with CHR, whereas KTx on CNI-free IS displayed a percentage comparable to the one of controls. In stable patients receiving standard CNI-based IS, the percentage of the Tact population was intermediate. Overall, these results suggest that the level of activation of T cells is different depending on the degree of allograft acceptance. We may also conclude that in stable KTx on standard IS, despite an apparent “clinical stability,” a significant proportion of circulating CD4⁺CD25^high T cells is composed of Tact, which may potentially play a detrimental role in the long-term outcome of allografts. Moreover, the group of KTx on CNI-free IS had Tact values comparable to controls. This group could represent what others have previously reported as “operational” or “near-tolerant” transplant recipients (27), because they were on minimal or no IS without evidence of graft dysfunction.

Our results on sirolimus-treated CNI-free stable patients are in accordance with experimental data, indicating that sirolimus relatively spares Treg (as compared with other IS medications) while at the same time preventing the expansion of effector T cells (22–26). Recent studies suggest that mammalian target of rapamycin inhibitors such as sirolimus, by contrast to CNIs, do not interfere with the suppressive capacity of Treg (28) and can induce de novo expression of FoxP3 in alloantigen-specific T cells (29) and an increase in the number of functional Treg in KTx (30), indicating that mammalian target of rapamycin inhibitors may favor Treg survival and function in the context of transplantation. It is therefore interesting to note that KTx receiving sirolimus not only had an absence of expansion of Tact but also had lower values of Tact than controls.

In the prospective part of our study, we found that in both THYMO and BSX groups the proportion of Tact increased rapidly after transplantation (as soon as 3 months), whereas the proportion of Treg decreased over the first 12 months. This reduction of Treg after basiliximab or thymoglobulin induction in clinical kidney transplantation has been previously described; however, the Tact population was not specifically analyzed (31–33). Our results indicate that despite standard induction and maintenance IS, there was a rapid expansion of Tact after transplantation, which was persistent over time, but with no apparent clinical relevance (i.e., no AR was observed in the majority of patients). Interestingly, Tact were highly increased during an AR episode in one patient and returned to usual values thereafter. This observation suggests that monitoring Tact after kidney transplantation could be of interest in assessing the alloimmune reaction of the recipient vis-à-vis the graft in the early post-transplant period.

Overall, our study further confirms that CD4⁺CD25^high T cells in humans are heterogeneous in terms of phenotype and function. Although Treg remain predominant, Tact are also represented within CD4⁺CD25^high T cells, and this subset may be relevant after transplantation. Both in the cross-sectional and the prospective parts of our study, Tact were found to be expanded in the circulation of most KTx, with the highest values observed in patients with CHR. However, most patients without apparent allograft dysfunction had significantly higher values than controls. Of note, the finding of nearly normal values in two groups of patients (sirolimus-treated CNI-free patients and patients on CNI-free IS) indicates that the population of Tact is not always increased after transplantation.

In conclusion, these results have demonstrated the expansion of the Tact population after kidney transplantation, with the highest proportion in patients with CHR. The data presented here provide the scientific basis for implementing the monitoring of Tact in large prospective clinical studies. Prospective monitoring of Tact (e.g., in addition to de novo anti-HLA antibodies) may be useful in the diagnosis of chronic rejection as well as during the so-called IS minimization strategies.

Disclosures

None.