CD4 CD25 Foxp3 Indirect Alloreactive T cells from Renal Transplant ...

doi: 10.1111/j.1365-3083.2007.01976.x ..................................................................................................................................................................

CD4+CD25+Foxp3+ Indirect Alloreactive T cells from Renal Transplant Patients Suppress Both the Direct and Indirect Pathways of Allorecognition M. Spadafora-Ferreira*,1, C. Caldas*, K. C. Fae´*, I. Marrero*, S. M. Monteiro*, H. T. Lin-Wang*, A. Socorro-Silva*, S. G. Fonseca*, J. A. Fonseca , J. Kalil*,à,§ & V. Coelho*,à,§

Abstract *Heart Institute (InCor), Medical School, University of Saõ Paulo, Saõ Paulo, Brazil; Renal Transplantation Unit, Medical School, University of Saõ Paulo, Saõ Paulo, Brazil; àInstitute for Investigation in Immunology, Millennium Institute, Saõ Paulo, Brazil; and §Clinical Immunology and Allergy, Department of Clinical Medicine, University of Saõ Paulo, Saõ Paulo, Brazil 1 Present address: Laboratory of Immunochemistry, Butantan Institute, Saõ Paulo, Brazil.

Received 10 May 2007; Accepted in revised form 19 May 2007 Correspondence to: V. Coelho, Laborato´rio de Imunologia, InCor, HC ⁄ FMUSP, Av. Dr. Eneás de Carvalho Aguiar 44, 9o andar, Bloco 2, Saõ Paulo, SP 05403-000, Brazil. E-mail: vecoelho @usp.br

Alloreactive T cells recognize donor antigens by two routes: direct and indirect pathways of allorecognition. Although the direct pathway is reported to be dominant in allograft rejection, indirect allorecognition also plays an important role. Indirect alloreactivity is also observed in renal transplant patients irrespective of rejection. Previously we showed a predominance of interleukin (IL)-10 induced by indirect allorecognition of donor human leucocyte antigen (HLA)-DR peptides, suggesting the existence of indirect alloreactive T cells displaying regulatory activity. In the present work, our objective was to characterize these regulatory T cells. We detected indirect alloproliferation of peripheral blood mononuclear cells (PBMC) from renal transplant patients, induced by donor HLA-DR peptides, dependent on IL-4 or IL-10, suggesting regulatory activity as part of the alloreactive T-cell repertoire. PBMC-derived indirect alloreactive T-cell lines were established and produced both inflammatory and regulatory cytokines. We showed that two of these T-cell lines which were able to inhibit both direct and indirect alloproliferation of another T-cell line from the same patient presented a CD4+CD25+Foxp3+ T-cell population. These data support the idea that indirect alloreactive T cells may also have regulatory activity and may contribute to the maintenance of the human renal allograft.

Introduction Alloreactive T cells recognize donor major histocompatibility complex (MHC) molecules by two distinct but non-exclusive pathways: the direct and indirect routes [1]. In the direct pathway, the recipient’s T cells recognize alloantigens presented by donor antigen-presenting cells (APC). In the indirect pathway, donor alloantigens are processed and presented as peptides to the recipient’s T cells by self-APC, analogous to recognition of nominal antigens. Currently, a third semi-direct pathway was proposed, where recipient dendritic cells (DC) may acquire intact MHC molecules from donor cells or tissues and stimulate direct anti-donor alloimmune responses [2]. In contrast to the direct pathway of allorecognition, indirect alloreactive T cells have a low precursor frequency comparable to that observed for nominal antigens [3, 4]. Different groups reported that both direct and indirect pathways of allorecognition are involved in acute

rejection both in animal models [3, 5–7] and human [8] transplantation. Other authors observed donor indirect alloreactivity associated with chronic rejection in murine [9] and human transplantation [4, 10, 11]. On the other hand, MHC-allopeptides have been used to induce tolerance in experimental transplantation models, suggesting that the indirect pathway of allorecognition also participates in the induction of tolerance [12, 13]. In addition, we and others have observed donor indirect allorecognition in renal transplant patients, irrespective of rejection [14–16]. We also observed predominant interleukin (IL)-10 production induced by indirect alloreactivity to donor human leucocyte antigen (HLA)-DR mismatched peptides, in renal transplant patients without rejection [17]. These results indicate that activation through the indirect pathway involves production of IL-10 and possibly activation of regulatory T (Treg) cells in human transplantation, as shown in murine cardiac transplantation [18]. 2007 The Authors

352

Journal compilation 2007 Blackwell Publishing Ltd. Scandinavian Journal of Immunology 66, 352–361

M. Spadafora-Ferreira et al. Human Indirect Alloreactive CD4+CD25+Foxp3+ T cells 353 ..................................................................................................................................................................

A population of CD4+CD25+ Treg cells was described to be involved in the maintenance of immune homeostasis and peripheral tolerance to self and alloantigens in mice [19, 20]. Other Treg cells, such as CD4+CD45RBlow and CD8+CD28), were also described to contribute to transplantation tolerance in mice, and were suggested to be important in human transplantation [18, 21, 22]. The forkhead-family transcription factor Foxp3 is important for the development and suppressive effect of CD25+CD4+ Treg cells [23, 24]. The mechanisms of action for Treg cells remain poorly understood. Some Treg cells have been described to be dependent on cell–cell contact [25–27], and others on cytokines, such as IL-4 [28, 29], IL-10 or the transforming growth factor b (TGF-b) [18, 30–32]. Over the past few years, many groups characterized antigen-specific CD4+CD25+ Treg in autoimmunity [33] and transplantation [34, 35]. However, it remains unknown whether Treg cells participate in the indirect pathway of allorecognition and whether indirect alloreactive T cells express Foxp3. In this study, we investigated the occurrence of T cells with regulatory activity reactive through the indirect pathway of allorecognition in renal transplant patients.

allograft nephropathy (CAN) during the period of study, but the other four patients presented CAN later on (Table 1). The patients were followed for a mean of 33 months (11–56 months). All the patients were under triple therapy (azathioprine, prednisone and cyclosporine A) during the period of the study. Four patients received the renal transplant from a cadaveric donor, and seven from a living donor. All the patients displayed at least one donor HLA-DR mismatch, detected by high-resolution polymerase chain reaction (PCR)-SSP HLA-DRB1* typing. Peptides. Peptides (14–21 mer) from the b-chain hypervariable regions of HLA-DRB1*0101, DRB1*1501, DRB1*0301, DRB1*0701 (Table 2) were synthesized as described previously [17]. Peptide purity was performed by HPLC and mass spectrometry (Maldi-Tof; Micromass ⁄ TofSpec SE, Manchester, UK). HLA-DRB1* peptides are referred as HLA-DR peptides. Peripheral blood mononuclear cells. Peripheral blood mononuclear cells (PBMC) were isolated by FicollHypaque density gradient centrifugation of heparinized blood samples, as previously described [17]. Cells were cryopreserved in RPMI 1640 medium (Sigma, St Louis, MO, USA) with 40% fetal calf serum and 10%

Material and methods Patients. Eleven patients from the Hospital das Clıńicas, University of Saõ Paulo Medical School, with up to 4 years of renal transplantation were studied (Table 1). Blood withdraw was approved by the Hospital’s Committee of Ethics (protocol 337 ⁄ 99) and patients gave their informed consent. Renal biopsies were performed by clinical indication and classified according to Banff criteria [36]. All but three patients developed at least one acute rejection episode in the first 3 months post-transplantation (Tx). Only two developed chronic

Table 2 HLA-DRB1* synthetic peptides. Peptide sequences HLA allele DRB1*0101 ⁄ 02 DRB1*0301 DRB1*1501 ⁄ 03

[1] 6–25 6–21 1–20

[2]

[3]

[4]

[5]

21–41 22–41 22–40

42–62 41–60 41–60

63–80 63–80 61–80

81–94 81–94 81–94

The numbers in the table represent the amino acid position on the protein sequence.

Table 1 Clinical data and characteristics of renal transplant recipients. Rejection eventsb (months post-Tx)

HLA-DR typing Patient

Age (yr)

Donor typea

Recipient

Donor

1 2 3 4 5 6 7 8d 9 10 11

70 65 14 52 28 42 21 7 12 28 37

C C L L L C L C L L L

1601, 0301, 1001, 1101, 0701, 0404, 1101, 0101, 1001, 0411, 0302,

1503, 0301, 1001, 0101, 0701, 1501, 1101, 0101, 1001, 0411, 0302,

1101 1401 0801 1301 1001 1104 0804 0701 0801 0404 0801

0404 0101 1501 1406 0301 X 1503 0301 1501 0301 0102

AR

CAN

1, 2 1 1 – – 1 – 1 3 1 1, 3

– CAN CAN CAN – – CAN CAN CAN – –

(2) III (84) I (89)

I (7) II (92) I (45)

Present clinical status (months post-Tx)

Last blood sample date (months post-Tx)

Present creatinin (mg ⁄ dl)

GF (87)C GF (49) CAN Stable Stable Stable FGS GF (92) PTLDe Stable Stable

37 38 27 29 12 44 42 56 41 35 24

NA NA 2.8 1.9 1.1 0.9 2.0 NA NA 1.5 1.6

a

C, cadaveric donor; L, living donor; bAccording to Banff classification; AR, acute rejection, CAN, chronic allograft nephropathy; GF, graft failure; FGS, focal glomerulosclerosis; cPatient died with sepsis; dRetransplant; ePatient died with PTLD, post-Tx lymphoproliferative disease; yr, year; Tx, transplantation; –, no rejection; NA, not available. 2007 The Authors Journal compilation 2007 Blackwell Publishing Ltd. Scandinavian Journal of Immunology 66, 352–361

354 Human Indirect Alloreactive CD4+CD25+Foxp3+ T cells M. Spadafora-Ferreira et al. .................................................................................................................................................................. dimethylsulphoxide (Merck, Darmstadt, Germany) and stored in liquid nitrogen until use. T-cell lines. Nine alloreactive PBMC-derived T-cell lines from three patients were established by stimulation with donor cells or HLA-DR peptides. Five PBMCderived T-cell lines were stimulated either with irradiated (50 Gy) donor spleen cells (cadaveric donor) or PBMC (living donor) cells and four T-cell lines were established by stimulation with a pool of five donor mismatched HLA-DR peptides 10 lg ⁄ ml each (Table 3). All T-cell lines were cultured in DMEM medium (Sigma) supplemented with 1 mM L-glutamine (Sigma), 40 lg ⁄ ml gentamicin (HC-USP, Saõ Paulo, Brazil), 1 mM sodium pyruvate, nonessential amino acids, vitamins (Gibco, Rockville, MD, USA) and 10% inactivated normal human serum (NHS) with addition of IL-2 (40 U ⁄ ml), IL-7 (2.5 ng ⁄ ml) and IL-15 (2.5 ng ⁄ ml). We then added IL-4 (100 U ⁄ ml) or IL-10 (100 U ⁄ ml) to favour the expansion of regulatory cells as described by others [28, 37]. T-cell lines were further expanded with anti-CD3 ⁄ anti-CD28 T-cell expansion beads (Dynal Biotech ASA, Oslo, Norway). Proliferation assays. Peripheral blood mononuclear cells (2 · 105 cells ⁄ well) or T-cell lines (5 · 104 cells ⁄ well) were plated in triplicate, in 96-well round bottom plates, with donor mismatched HLA-DR peptides (10–50 lg ⁄ ml) in DMEM with 10% NHS, as described earlier [17]. After 10 days of expansion, T-cell lines were washed to remove any residual cytokine and assayed for proliferation. Autologous irradiated (50 Gy) PBMC (1 · 105 ⁄ well) were used

as feeder cells (APC) for T-cell lines. Proliferation assays with patients’ PBMC were performed in three different conditions: (i) without exogenous cytokines, (ii) with addition of IL-4 (200 U ⁄ ml) or (iii) IL-10 (100 U ⁄ ml). [3H]-thymidine (GE Healthcare, Buckinghamshire, UK) was added in the last 18 h of 7-day culture for PBMC and a 5-day period for T-cell lines. A longer period of time was used for PBMC, as these cells were not primed with peptides as T-cell lines. Cells were harvested and [3H]-thymidine incorporation was measured by liquid scintillation in a b-counter (Betaplate, Wallac-PerkinElmer, Waltham, MA, USA). Responder PBMC alone or T-cell lines plus irradiated APC without peptide were used as a negative control for PBMC and T-cell line proliferation respectively. Phytohaemagglutinin-P (PHA-P) was used as a positive control (2.5 lg ⁄ ml). Proliferation was considered positive with a stimulation index (SI) ‡2.0 [SI = mean counts per minute (cpm) of cells with antigen ⁄ cpm of cells without antigen], as previously described [17]. A patient was considered positive when proliferation was detected against at least one peptide. Regulatory T-cell function assays. The inhibitory capacity of T-cell lines over the indirect or direct [mixed lymphocyte reaction (MLR)] pathway of allorecognition was tested with or without cell contact using a transwell system (Millipore Corp., Bedford, MA, USA). Briefly, 5 · 105 responder T cells ⁄ well were distributed into a 24-well plate with irradiated (50 Gy) autologous PBMC (5 · 105 ⁄ well) in the presence of a pool of five donor HLA-DR peptides for the indirect pathway or allogeneic irradiated PBMC for the

Table 3 PBMC-derived T-cell lines from renal transplant patients – Phenotype, cytokine production and proliferation induced by donor HLADRB1* peptides. Indirect alloresponse to donor HLA-DR peptidesc Patient Rejection no. events

PBMC sample T-cell perioda linesb

2

18 mo

HLA-DR % % IL-10 CD4+ CD8+ peptide

IL-4

TGF-b IL-5

IFN-c

TNF-a

Proliferation

T2.1 T2.3 T2.5 T2.6

31 93 13