Human CD4+ effector T lymphocytes generated upon TCR ... - Nature

Cellular & Molecular Immunology (2013) 10, 261–274 ß 2013 CSI and USTC. All rights reserved 1672-7681/13 $32.00 www.nature.com/cmi

RESEARCH ARTICLE

Human CD41 effector T lymphocytes generated upon TCR engagement with self-peptides respond defectively to IL-7 in their transition to memory cells Gabriela Gonza´lez-Pe´rez1, Norma C Segovia1, Amaranta Rivas-Carvalho1, Diana P Reyes1, Honorio Torres-Aguilar1, Sergio R Aguilar-Ruiz1, Claudine Irles2, Gloria Soldevila3 and Carmen Sańchez-Torres1 The peripheral repertoire of CD41 T lymphocytes contains autoreactive cells that remain tolerant through several mechanisms. However, nonspecific CD41 T cells can be activated in physiological conditions as in the course of an ongoing immune response, and their outcome is not yet fully understood. Here, we investigate the fate of human naive CD41 lymphocytes activated by dendritic cells (DCs) presenting endogenous self-peptides in comparison with lymphocytes involved in alloresponses. We generated memory cells (Tmem) from primary effectors activated with mature autologous DCs plus interleukin (IL)-2 (Tmauto), simulating the circumstances of an active immune response, or allogeneic DCs (Tmallo). Tmem were generated from effector cells that were rested in the absence of antigenic stimuli, with or without IL-7. Tmem were less activated than effectors (demonstrated by CD25 downregulation) particularly with IL-7, suggesting that this cytokine may favour the transition to quiescence. Tmauto and Tmallo showed an effector memory phenotype, and responded similarly to polyclonal and antigen-specific stimuli. Biochemically, IL-7-treated Tmallo were closely related to conventional memory lymphocytes based on Erk-1/2 activation, whereas Tmauto were more similar to effectors. Autologous effectors exhibited lower responses to IL-7 than allogeneic cells, which were reflected in their reduced proliferation and higher cell death. This was not related to IL-7 receptor expression but rather to signalling deficiencies, according to STAT5 activation These results suggest that ineffective responses to IL-7 could impair the transition to memory cells of naive CD41 T lymphocytes recognizing self-peptides in the setting of strong costimulation. Cellular & Molecular Immunology (2013) 10, 261–274; doi:10.1038/cmi.2012.71; published online 4 March 2013 Keywords: human CD41 T cells; IL-7; memory; self-peptides

INTRODUCTION Memory T cells promote long-lasting protective immune responses. However, since naive lymphocytes directed against self-antigens (Ags) and allo-Ags are present within the peripheral T-cell pool, memory responses can also be detrimental by mediating autoimmune diseases or graft rejection.1,2 Responses to self-peptide/major histocompatibility complexes (MHCs) by T lymphocytes also play a key role in their biology without promoting disease. Proliferation of naive T cells via interleukin (IL)-7 and T-cell receptor (TCR)/self-MHC engagement has been described for both CD41 and CD81 T

cells in lymphopenic hosts, and causes their differentiation into memory-like cells. Likewise, the maintenance of naive CD41 and CD81 T cells requires IL-7 and TCR signals from selfpeptide/MHC complexes in lymphoreplete hosts (reviewed in Ref. 3). The autologous mixed lymphocyte response (AMLR) has been recognized for many years. Cai and Hafler,4 by using autologous immature dendritic cells (DCs) as antigen presenting cells, estimated a precursor frequency of ,0.04% human CD41 T cells within the peripheral pool that proliferate in response to self-peptide/MHC complexes. However, the exact

1 Department of Molecular Biomedicine, Centro de Investigacioń y de Estudios Avanzados del I.P.N. (CINVESTAV-IPN), Mexico City, Mexico; 2Department of Physiology, Facultad de Medicina, Mexico City, Mexico and 3Department of Immunology, Instituto de Investigaciones Biome´dicas, Universidad Nacional Autońoma de Me´xico (UNAM), Mexico City, Mexico Correspondence: Dr C Sańchez-Torres, Department of Molecular Biomedicine, Centro de Investigacioń y de Estudios Avanzados del I.P.N. (CINVESTAVIPN), Av. I.P.N. 2508, C.P. 07360 Mexico City, Mexico. E-mail: [email protected] Received: 26 June 2012; revised 16 November 2012; accepted: 30 November 2012

Memory lymphocytes generated in autologous responses G Gonza´lez-Pe´rez et al

262

fate and characteristics of lymphocytes stimulated by self-Ag remain unclear. It has been suggested that during the primary AMLR, the responding lymphocytes exhibited features of immunological memory and specificity.5 Despite some studies have confirmed that AMLR led to the generation of lymphocytes with suppressive abilities,6 Zwickey and co-workers7 demonstrated in a murine model that self-Ag presented on MHC class II molecules by DCs during Listeria monocytogenes infection or after injection of anti-CD40 antibodies resulted in the activation of autoreactive T cells and disease. Other studies have corroborated that in vivo TCR engagement with selfMHC in the presence of strong adjuvants (dsRNA, type I interferons (IFN)) led to bystander T-cell activation.8,9 Thus, it is plausible that in these inflammatory settings, bystander lymphocytes might be activated by signals from mature DCs presenting self-Ag, as well as by the cytokine environment. Indeed, it has been demonstrated that the activation of human blood DCs, but not monocytes, is essential to initiate CD41 T-cell proliferation in the AMLR, and depends on the presence of costimulatory and MHC class II molecules.10 Taking into account these observations, we asked whether human memory CD41 T cells can be generated in a primary AMLR under circumstances resembling an active immune response, i.e., activated by mature DCs and cytokines known for their ability to stimulate T-cell bystander responses, such as IL-2.11 The pathways of differentiation and the cellular precursors of memory T cells are not entirely defined. However, it has been extensively documented that CD41 memory T cells can develop directly from differentiated effector lymphocytes responding to their cognate Ag.12,13 The precursors of memory T cells have selective survival within an effector cell pool that is otherwise prone to die. Cytokines that signal through the common gamma chain (cc), such as IL-7 and IL-2, have been implicated in the survival of effector CD41 T cells.14–17 In addition, cessation of stimulation has been proposed as a necessary step for effector to memory cell transition.18–20 TCR signalling intensity has been established as a key parameter for memory T-cell formation. Some studies showed that persistent or strong stimulation through the TCR can dampen memory generation.21 However, others indicated that a high threshold of TCR signalling is required by naive CD41 T lymphocytes in order to fully differentiate into effector cells that can convert to memory cells. Limited amounts of Ag, as well as non physiologically high precursor frequency of Ag-specific naive T cells competing for antigen presenting cells interactions or nutrients, results in defective memory formation.22,23 Therefore, and according to the quality and strength of differentiation signals, it is feasible that memory cells generated by stimulation with self-Ag would differ from memory lymphocytes generated upon activation with their nominal Ag. In the present study, we analysed whether human primary effector CD41 T cells generated under different conditions of activation (allo- or self-Ag) give rise to different types of memory lymphocytes, and the contribution of IL-7 to this process. Cellular & Molecular Immunology

MATERIALS AND METHODS Media and reagents Lymphocytes and monocytes were cultured in RPMI 1640 medium supplemented with 10% heat-inactivated autologous serum, 2 mM L-glutamine, 1 mM sodium pyruvate, 0.1 mM non-essential amino acids (Hyclone Laboratories, Logan, UT, USA), 1% penicillin/streptomycin, and 50 mM 2-ME (Gibco, Grand Island, NY, USA), referred to as complete medium. The following reagents were used: human recombinant (hr) granulocyte-macrophage colony-stimulating factor (1000 U/ml, kindly provided by Probiomed, Mexico City, Mexico), hrIL-4 (15 ng/ml) and prostaglandin E2 (0.1 mg/ml) (Calbiochem, San Diego, CA, USA), tumor-necrosis factor (40 ng/ml; R&D Systems, Minneapolis, MN, USA), hrIL-7 (5 ng/ml; Peprotech, Rocky Hill, NJ, USA), hrIL-2 (80 U/ml; Biovision, Mountain View, CA, USA), carboxy-fluorescein diacetate succinimidyl ester (CFSE, 10 mM, Molecular Probes, Eugene, OR, USA) and PKH26 red fluorescent cell linker (2 mM, Sigma Aldrich, St Louis, MO, USA). Cell separation and differentiation of peripheral blood monocytes Peripheral blood mononuclear cells were purified from buffy coats of healthy volunteers by Ficoll-Hypaque (Nycomed Pharma, New York, NY, USA) density–gradient centrifugation. The procedures and all the experiments of the study were in accordance to the ethical standards, and donors signed an informed consent. CD141 monocytes were separated by magnetic cell sorting, using MACS isolation kits (Miltenyi Biotec, Auburn, CA, USA). CD41 T lymphocytes were separated by negative selection using the MACS CD41 T-cell isolation kit (Miltenyi Biotec), followed by incubation with MACS antiCD45RA antibody to obtain naive (CD45RA1) and memory (CD45RA2) cells. Central (CCR71, TCM) and effector memory T cells (CCR72, TEM)24 were isolated from CD41CD45RA2 lymphocytes by FACS using an anti-CCR7 antibody (3D12; BD Biosciences, Bedford, MA, USA). CD252 Tmem were isolated by negative selection using the CD251 cell isolation kit (Miltenyi Biotec). Isolated monocytes were cultured at 106 cells/ml in complete medium supplemented with granulocyte-macrophage colonystimulating factor and IL-4 to obtain DCs. The cultures were fed with fresh medium and cytokines every 2 days. After 6 days, non-adherent DCs were harvested and replated at 53105 cells/ ml in medium containing granulocyte-macrophage colonystimulating factor and IL-4. Immature DCs were maturated with tumor-necrosis factor and prostaglandin E2 for two more days. In vitro generation of primary effectors and memory CD41 T cells To generate effector T cells, human naive CD41 T cells were stimulated with either autologous or allogeneic DCs, at a ratio of 8 : 1 for 8 days. In most of the experiments conducted with autologous DCs the culture medium was supplemented with IL-2, added once on the first day of culture. An aliquot of the


263

culture supernatants was harvested every day to quantify the amount of IL-2 by ELISA (BD Biosciences). In some assays, a blocking anti-HLA-DR monoclonal antibody (mAb) (G46-6, 25 mg/ml; BD Biosciences) was added from the beginning to the autologous cultures. After 8 days of coculture, effector T cells were purified by negative selection using MACS anti-CD45RA antibodies (Miltenyi Biotec). To generate memory cells (Tmem), purified effector T cells were maintained in the absence of stimulus for up to 10 days, with or without IL-7 added every other day along the resting period. Primary effectors were also generated from freshly isolated naive CD41 T cells for biochemical studies. Cells (13106 cells/ml) were cultured on plate-immobilized anti-CD3 (UTCH1, 5 mg/ml; BD Biosciences) and anti-CD28 (CD28.2, 2.5 mg/ml; BD Biosciences) mAbs for 3 days. Flow cytometry assays Lymphocytes were stained with fluorochrome-conjugated mAbs directed against human CD25 (M-A251), CD45RA (HI100), CD45RO (UCHL1), CD62L (DREG-56), CD127 (hIL-7RM21), all from BD Biosciences, and CCR7 (150503), from R&D Systems. Lymphocytes were also stained with biotin-conjugated annexin V, followed by allophycocyanin-labelled streptavidin (BD Biosciences) and propidium iodide (PI; SigmaAldrich). For IFN-c production assays, 105 freshly isolated naive and memory cells, and in vitro generated effectors and Tmem, were stimulated for 6 h with phorbol 12-myristate 13-acetate (PMA, 50 ng/ml) plus ionomycin (500 ng/ml) or with plateimmobilized anti-CD3 (5 mg/ml) and anti-CD28 (2.5 mg/ml) mAbs. Then, cells were stained with FITC-conjugated antiCD45RO mAb, fixed and permeabilized using the Cytofix/ Cytoperm kit (BD Biosciences), and incubated with allophycocyanin-labelled anti-IFN-c mAb (45-15; Miltenyi Biotec). Active caspase-3 was detected with the caspase-3 detection kit (RedDEVD-FMK) from Calbiochem. Data were acquired in a FACSCalibur using the CellQuest software (BD Biosciences). Proliferation assays Freshly isolated naive and memory CD41 T lymphocytes, as well as in vitro generated effectors and Tmem, were labelled with CFSE in order to analyse their proliferation in response to autologous or allogeneic DCs (naive and Tmem) or plateimmobilized anti-CD3 (1 mg/ml) and anti-CD28 (0.5 mg/ml) mAbs (naive, TEM, TCM and Tmem). Likewise, freshly isolated memory T cells and in vitro generated effectors and Tmem were labelled with CFSE and analysed during the resting period, in the presence or absence of IL-7. Suppression assays were carried out as published elsewhere.25 Briefly, allogeneic Tmem (Tmallo) or naive T cells from donor A were used as indicator responder T cells (Tind). Tind were labelled with CFSE and cultured at different ratios with PKH26-labelled autologous Tmem (Tmauto) from donor A and allogeneic DCs from donor B (10 : 1 T cell/DC ratio) for 3–4 days. Tmallo were previously generated with DCs from donor B. The CFSE dilution of Tind was analysed by flow cytometry.

Activation of lymphocytes for detection of signalling intermediates For detection of Jak3 and STAT5 phosphorylation, cells were cultured for 4 h at 37 uC in RPMI 1640 containing 0.5% foetal bovine serum (FBS), and then IL-7 (10 ng/ml) was added for additional 10, 20, 30 or 90 min. For detection of lck, ZAP-70, LAT, SLP-76 and Erk-1/2, cells were cultured in the same medium for 6 h. Subsequently, an anti-CD3 mAb was added and cells were incubated on ice for 15 min. Then, cells were washed and resuspended in cold RPMI 1640 containing 0.5% FBS, goat anti-mouse Igs (BD Biosciences) was added and cells were kept on ice for additional 15 min. After that, cells were washed and resuspended in pre-warmed (37 uC) RPMI 1640 with 0.5% FBS and incubated at 37 uC for 3 min. Following activation, lymphocytes were pelleted and lysed in cold 1% NP-40 lysis buffer with protease/phosphatase inhibitors and used for western blotting. For detection of intracellular phospho-proteins by flow cytometry, lymphocytes were cultured in RPMI-1640 containing 0.5% FBS for 5 h at 37 uC. Then, cells were stimulated with mouse anti-CD3 and anti-CD28 mAbs for 15 min at 4 uC and incubated with a goat anti-mouse Igs for additional 15 min at 4 uC. Cells were washed and resuspended in pre-warmed (37 uC) Dulbecco’s phosphate-buffered saline containing 1% FBS and incubated at 37 uC for 3 min. Cells were subsequently washed at 4 uC, fixed and permeabilized with Phosflow buffer (BD Biosciences). Then, lymphocytes were incubated with Alexa Fluor 488-conjugated anti-phospho-SLP-76 (J141688.36.58) mAb, or with purified anti-phospho-tyrosine (PY20) mAb (all from BD Pharmingen) followed by a FITClabelled goat anti-mouse Ig (Dako, Carpinteria CA, USA). Cells were analysed by flow cytometry. Western blotting Protein extracts from 13106 lymphocytes were resolved on reducing 8–15% polyacrylamide gels and transferred onto nitrocellulose sheets. Membranes were incubated with mAbs to lck (3A5), to SLP-76 (F-7), to Erk-1/2 (K-23), rabbit polyclonal antibody to ZAP-70 (LR), all from Santa Cruz Biotechnology (Santa Cruz, CA, USA), mAbs to phosphoErk-1/2 (E10), to phospho-Jak3 (D44E3), to phospho-STAT5 (14H2), to Jak3 (5H2), to STAT5 (3H7), all from Cell Signalling Technology (Danvers, MA, USA), rabbit polyclonal antibody to LAT (Upstate Biotechnology, Lake Placid, NY, USA), or mAb to b-actin (AG8, kindly supplied by Dr Manuel Hernandez, CINVESTAV, Mexico, USA), followed by HRP-conjugated secondary antibodies (Dako). Bands were detected using enhanced chemiluminescence (Amersham Biosciences, Piscataway, NJ, USA). Blots were scanned, and densitometric analysis of the autoradiograms was performed with the Image/J software (version 1.42q). Statistical analysis Data were expressed as mean6s.d. of independent experiments. The statistical significance of the data was determined by Student’s two-tailed paired t-test, assuming equal variances. Cellular & Molecular Immunology


264

RESULTS Generation of effector CD41 T cells Effector T cells were generated by stimulation of CD41CD45RA1 naive T cells with either autologous or allogeneic mature DCs (Figure 1a). Some of the autologous cultures were supplemented with IL-2 in order to simulate the context of an ongoing Ag-specific immune response, where some lymphocytes with an unrelated TCR could be activated by cytokines secreted by specific T lymphocytes. At day 8, we found around 60%–80% of dividing cells in the allogeneic response, whereas in the AMLR, the proliferation was poor (about 10%–30% of cycling cells), but increased with the addition of IL-2 (30%–60%). At the end of stimulation, we recovered about 250% of the plated cells (referred as 100%) in the allogeneic response, around 20% in the autologous response, and 100% when IL-2 was added (Figure 1b). Effector T cells (CD45RO1) were able to produce IFN-c in any condition (Figure 1c). Furthermore, IL-2 was endogenously produced in the allogeneic cultures at levels comparable or superior to that was found in the AMLR when IL-2 was added (Figure 1d). In both cases, the levels of IL-2 were highly variable depending on the donors, which might be associated with the magnitude of their responses in terms of IL-2 production and consumption. The autologous lymphocytes cultured in the absence of IL-2 produced very low levels of this cytokine (Figure 1d). The reduced recovery obtained in this condition did not allow us to follow their fate along the transition to memory cells; hence, the data subsequently provided for this response correspond to T cells stimulated in the presence of IL-2. In order to assess whether IL-2 by itself induced T-cell division in the AMLR, we blocked the TCR-MHC interaction with an anti-HLA-DR mAb throughout the stimulation with DCs (Figure 1e). The almost absent cycling cells in the presence of the mAb indicated that the proliferation of autologous T cells is initiated by the interaction of their TCR with MHC–self-peptide complexes, and IL2 would mainly amplify the response. Besides IL-2, inflammatory cytokines such as tumor-necrosis factor or IL-6 could influence the overall response of naive lymphocytes.26 The secretion of theses cytokines would result from the cross-talking between readily activated lymphocytes and DCs, and could be reduced in the AMLR compared with the allogeneic system. To investigate this issue, the autologous cultures were supplemented with supernatants of allogeneic cultures. These supernatants were collected and added every day from parallel allogeneic cultures. At the end of the 8-day culture, the recovery obtained in the AMLR in the presence of the supernatants was very similar to that obtained when IL-2 was added once at the beginning of the co-cultures (Supplementary Figure 1a). Then, IL-2 was sufficient to maintain a degree of proliferation and/or survival in the AMLR similar to the soluble factors found in the allogeneic response. Generation of memory CD41 T cells Once the effector cells were generated, they were purified and cultured in the absence of stimulation for up to 10 days Cellular & Molecular Immunology

in order to generate resting Tmem. We also analysed the influence of IL-7 in the generation of these cells. At 8–10 days, there were no proliferating cells in the cultures. Figure 2a shows the proliferation of Tmem (day 10) and of memory CD41 T cells isolated from blood and cultured for 10 days, in the presence or absence of IL-7. At that time, cells were labelled with CFSE and cultured for 2 additional days, with or without IL-7. The data demonstrated that, during the resting period, in vitro generated effector cells returned to quiescence, and they behave similarly to peripheral memory T cells (Figure 2a). Furthermore, IL-7treated Tmem showed forward (FSC) and side (SSC) light scatter profiles similar to small, resting lymphocytes. Control Tmem were more heterogeneous and, in addition to small lymphocytes, they also included enhanced numbers of bigger and more complex cells (Figure 2b). The recovery of Tmem was significantly higher with IL-7 for both Tmauto and Tmallo with respect to their controls (Figure 2c). Strikingly, IL-7 promoted higher recovery of Tmallo compared with their autologous counterparts (Figure 2c). To evaluate whether the low response of Tmauto to IL-7 was specifically related with the amount of IL-2 used during the generation of their effectors, we analysed the recovery of autologous Tmem from effectors generated with allogeneic culture supernatants (Supplementary Figure 1b). Our data indicated that these Tmem also responded poorly to IL-7 compared with the same lymphocytes activated with allogeneic DCs. These results suggest that lymphocytes receiving low strength of stimulation are impaired to fully respond to IL-7. Phenotype of memory T cells Tmallo and Tmauto showed lower percentages of CD251 cells and higher percentages of CD62L1 cells than their respective effectors (Figure 3), indicating that Tmem were less activated than effector cells. Regarding CCR7, higher numbers of Tmallo expressed CCR7 compared with their effectors, whereas the percentage of Tmauto expressing CCR7 was lower and comparable to their effectors (Figure 3). Nevertheless, most cells (,70%) were CCR72CD62L1/2, which is in accordance with a TEM phenotype.24 Tmallo showed diminished CD25 expression than Tmauto, both in the absence or presence of IL-7, which suggests that allogeneic effectors might return faster to a resting state. Additionally, Tmem generated with IL-7 showed lower CD25 expression than their controls, suggestive of a role of IL-7 in the return to quiescence. These data are in agreement with the FSC/ SSC profiles shown in Figure 2b, where higher numbers of large cells were detected in the control compared with the IL-7treated Tmem. When we evaluated the expression of CD25 in these large cells (gate R1; Supplementary Figure 2a), the results were very similar to those found in the whole cell population: control Tmem had higher percentages of CD251 cells than cells treated with IL-7, and Tmallo showed lower percentages of CD251 cells than Tmauto (Supplementary Figure 2b, left panel). However, the larger lymphocytes had increased percentages of CD251 cells in both Tmallo and Tmauto with respect to


265

a Day 8 200

250

Day 6

13

29

b

31

Auto DC

* M1

M1

M1

*

M1

102

5

101

102

104

100

55

101

103

104

100

50

101

102

102

103

300

104

60

M1

M1

103

104

100

200

100

103

Cell recovery (%)

101

0

100

0

104

0

103

150

102

104

M1

0 200

101

103

18

M1

100

102

300

3

101

0

100

0

104

250

103

200

102

200

200

101

0

0

350 100

73

101

102

103

250 *

200 150

Plated cells

100

104

50 76

100

M1

0

0

M1

101

102

103

104

100

101

102

103

104

Auto DC

M1

Allo DC

Auto DC/IL-2

0

Allo DC

0 M1

0

Auto DC+IL-2

Day 4 200

250

Day 2 3

100

101

102

103

104

100

101

102

103

104

CFSE

c

26

101

102

103

100

12

48

101 100

101 100

101 100

1.5 39

103

3

44 27

102

8

104

104 86

+ IL-2

103

5

100

IFN-c

Allogeneic DC

102

1.4

102

103

104

Autologous DC

101

102

103

104

100

102

101

103

104

CD45RO

e

CD45RA

600 400 200

104 9 26 51 13

100

0 D1

D2

D3

D5

D7

2 65

103

Auto DC/IL2

6 27

100 101 102

Auto DC

800 IL-2 (pg/mL)

103 104

1000

a-HLA-DR

Control

Allo DC

100 101 102

d

101

102

103

104

100

101

102

103

104

CFSE

Figure 1 Proliferation and cell recovery of naive CD41 T lymphocytes stimulated with autologous or allogeneic DCs. Human naive CD41 T cells were labelled with CFSE and stimulated for 8 days with autologous DCs (Auto DC), Auto DC plus IL-2 (Auto DC1IL-2) or allogeneic DCs (Allo DC). (a) Proliferation of naive CD41 T cells after 2, 4, 6 and 8 days of co-culture with Auto DC (top), Auto DC1IL-2 (middle) or Allo DC (bottom) in a representative experiment, evaluated by flow cytometry. Numbers indicate the percentage of cycling cells. (b) Percentage of T cells recovered after 8 days of coculture with DCs. Cell recoveries were determined by counting live, trypan blue-excluding cells, and calculated as the percentage with respect to the number of plated cells at the beginning of the cultures (100%). Shown is the mean6s.d. of 13 independent donors. (c) IFN-c production by DC-activated T cells at the 8th day of coculture in a representative experiment. Cells were previously stimulated with PMA and ionomycin. (d) Quantification of IL-2 in the supernatants of T-cell cultures established with Auto DC, Auto DC1IL-2 or Allo DC, evaluated at days (D) 1, 2, 3, 5 and 7 of activation The data represent the mean6s.d. of three independent donors. (e) Cocultures of Auto DC and CFSE-labelled naive T cells were established as above, in the presence of a blocking anti-HLA-DR mAb (a-HLA-DR) or an isotype control mAb (Control) for 6 days. Then IL-2 was added, and proliferation was evaluated by flow cytometry 6 days later. Numbers in the plots indicate the percentage of positive cells in each quadrant. Statistical analysis: *P,0.05. CFSE, carboxy-fluorescein diacetate succinimidyl ester; DC, dendritic cell; IFN, interferon.

Cellular & Molecular Immunology


266

4 102 103 10

4 102 103 10

1450

320

500

100 101

1

M1

100 101

2

M1

0

100 101

IL-7

Ctrl

2

M1

0

4 102 103 10

IL-7

1

M1

0

0

350

750

60

2

M1

100 101

Ctrl

M1

4 102 103 10

0

IL-7

1

Peripheral Memory

Tmauto

Tmallo Ctrl

0

a

100 101

4 102 103 10

100 101

4 102 103 10

CFSE

Tmauto

Tmallo

1023

1023

0

Ctrl 1023

1023

IL-7

0

1023 0

1023

0

0

1023

0

Effectors Auto

IL-7

1023

Ctrl

0

1023 0

SSC

Effectors Allo

1023

0

0

1023

0

b

0

1023

FSC

c

* * *

Cell recovery (%)

200

Tmallo Tmauto

150 Plated cells

100 50 0 Control

IL-7

Figure 2 IL-7 promotes the long-term recovery of in vitro generated memory T cells. Effector lymphocytes were generated by coculture with DCs (Allo DC or Auto DC1IL-2) as in Figure 1, and then CD45RO1 cells were isolated and rested for 8–10 days in the absence (Ctrl) or presence of IL-7. (a). Effector cells were rested for 10 days to obtain Tmallo and Tmauto, in the absence (Ctrl) or presence of IL-7. Freshly isolated CD41CD45RO1 memory T cells (peripheral memory) were cultured for 10 days as well, with or without IL-7. Then, cells were labelled with CFSE and cultured for 2 additional days under the same condition they were generated (presence or absence of IL-7). Proliferation was evaluated by flow cytometry. Numbers in the top of histograms indicate the percentage of dividing cells. (b) Forward and side light scatter profile of in vitro generated allogeneic (Effectors Allo) and autologous (Effectors Auto) effector T cells, as well as of 8-day Tmallo and Tmauto generated without (Ctrl) or with IL-7. (c) Longterm recovery of memory T cells. Effector T cells derived from Allo or Auto DC were rested in the absence (Ctrl) or presence of IL-7 for 10 days in order to generate Tmallo (grey) or Tmauto (black). Cell recoveries were determined as in Figure 1b, and calculated as the percentage with respect to the number of plated effectors at the beginning of the resting period (100%). Data are the mean6s.d. of 10 independent donors. Statistical analysis: *P,0.05. CFSE, carboxy-fluorescein diacetate succinimidyl ester; DC, dendritic cell.

the fraction of small lymphocytes (gate R2; Supplementary Figure 2a and b, right panel), suggesting that the large cell population may represent the remaining effectors. We also detected higher percentages of annexin V1 cells in the fraction of large lymphocytes compared with the small cells, regardless of the T-cell subpopulation evaluated (Supplementary Figure 2c). Taking into account that the cells with high FSC/SSC are overrepresented in control Tmem, the data suggest that the absence of IL-7 may impair or delay the transit of some effectors to resting cells, which subsequently favour their death. Functional features of memory T cells To determine whether there were functional differences between Tmallo and Tmauto, we analysed their IFN-c production and proliferative ability. We detected high number of IFN-c1 cells in all Tmem populations upon stimulation with Cellular & Molecular Immunology

PMA and ionomycin, with no significant differences between autologous and allogeneic lymphocytes (Figure 4a). Freshly isolated naive CD41 T lymphocytes did not produce IFN-c, while memory cells did (Figure 4a). In order to assess whether Tmem exhibited differences in the proximal TCR signalling which are bypassed by PMA/ionomycin, lymphocytes were stimulated with anti-CD3 and anti-CD28 mAbs (Figure 4b). In contrast with the above result, lower numbers of Tmem maintained in the absence of IL-7 produce IFN-c compared with lymphocytes cultured with the cytokine (only significant with allogeneic cells), which suggests a certain role of IL-7 in the maintenance of a proper TCR signalling machinery. The ability to proliferate upon TCR/CD3/CD28 cross-linking was intact in all the Tmem populations tested. It was similar to that was found in freshly isolated memory CD41 T cells (TCM and TEM lymphocytes), and significantly increased with


267 Allo

Auto

* * *

100

*

100

100

80

80

*

*

60

*

40

*

60

* *

*

40 20

20 0 Tmem Ctrl

Tmem IL-7

Tmem Ctrl

*

40

Effector

Tmem IL-7

Tmem Ctrl

Tmem IL-7

* * *

80 *

60 40

* CD62L (MFI)

*

80 *

50 CCR7 (MFI)

120

*

60

*

100

60

0 Effector

* 140

*

20

0 Effector

CD25 (MFI)

%CD62L+ T cells

* *

%CCR7+ T cells

%CD25+ T cells

* 80

40 30 20

60 40 20

10

20

0

0 Effector

Tmem Ctrl

Tmem IL-7

0 Effector

Tmem Ctrl

Tmem IL-7

Effector

Tmem Ctrl

Tmem IL-7

Figure 3 Phenotype of memory T cells generated in vitro. Human naive CD41 T cells were stimulated with either Allo DC (grey bars) or Auto DC1IL2 (black bars) to generate effector cells, as in Figure 2. Then, purified effectors were cultured in the absence (Ctrl) or presence of IL-7 for 10 days in order to generate memory T cells (Tmem). Effectors and memory T cells were stained with mAbs to CD25, CCR7 and CD62L, and analysed by flow cytometry. Data are the mean6s.d. of seven independent donors and are shown as the percentage of positive cells (top) and the MFI of each marker (bottom). Statistical analysis: *P,0.05. DC, dendritic cell; MFI, mean fluorescence intensity.

respect to naive CD41 T cells, as expected (Figure 4c). Furthermore, Tmallo and Tmauto showed a high degree of specificity for the Ag with which they were primed. Thus, Tmallo respond preferentially to allogeneic vs autologous DCs, while the opposite was observed in Tmauto (Figure 4d). Altogether, these data indicate that Tmallo and Tmauto functionally behaved like conventional memory cells. According to some studies, the AMLR leads to the generation of lymphocytes with suppressive abilities.6 Thus, we wondered whether Tmauto have suppressive capacity on naive or memory T cells. To this purpose, naive T cells or Tmallo (referred to Tind) were stimulated with allogeneic DCs, either alone or in the presence of different ratios of Tmauto generated with IL-7. We did not detect significant inhibition of Tind proliferation by Tmauto in any donor tested (Figure 4e). Likewise, Tmauto generated in the absence of IL-7 did not abrogate the proliferation of Tind (data not shown). Therefore, we conclude that Tmauto do not have not suppressive abilities, at least in the allogeneic models evaluated. TCR-mediated signalling in effector and memory CD41 T cells Biochemical differences between naive, effectors and memory CD41 T cells have been previously reported in mice27–32

and humans,33,34 such in the total protein tyrosine phosphorylation, in the expression of CD3f-associated kinase 70 (ZAP-70) and SH2-containing leukocyte molecule of 76 kDa (SLP-76), or in the activation of the mitogen-activated protein kinase Erk-1/2. Thus, we next analysed the biochemical profile of Tmem to evaluate whether they actually acquired memory-cell properties. Initially, we analysed the signalling through the TCR in naive and memory CD41 T cells obtained directly from peripheral blood, and in primary effectors derived by in vitro stimulation of naive cells with anti-CD3 and anti-CD28 mAbs. Among the molecules tested (p56lck kinase (lck), ZAP-70, linker for activation of T cells (LAT), SLP-76, Erk-1/2 and total tyrosine phosphorylation), the main biochemical difference between naive, memory and effector CD41 T cells resided in Erk-1/2 phosphorylation, which was consistently diminished in memory vs naive and effector lymphocytes in response to anti-CD3 mAbs (Figure 5a and b and Supplementary Figure 3). Thus, we took advantage of this feature to characterize Tmem. Effector T cells derived from the autologous and allogeneic stimulation and their respective Tmem cultured without IL-7 exhibited similar levels of Erk-1/2 phosphorylation (Figure 5c– e), which indicate that these Tmem share biochemical features of Cellular & Molecular Immunology


268

a

b

Tmauto

Tmallo

Tmallo

Tmauto

40

60

*

%IFN-c+ cells

%IFN-c+ cells

50 40 30 20

30

20

10 10 0

0 Naive

Memory

c

Control

*

Auto DC

Tind naive

*

80

*

40

60 40 20

20 0

0 Naive

TCM

TEM Control

IL-7

Ctrl

Tind mem

Tmauto

Tmauto

IL-7 Tmallo

100

100

80

80

60

60

40

40

20

20

* % Cycling cells

% Cycling cells

*

60

IL-7

*

*

*

80

Control

e Allo DC

* * 100

Memory

Naive

d Tmauto

Tmallo

% Cycling cells

IL-7

Ctrl IL-7 Tmauto

0

0 0,5 : 1 1 : 1 2 : 1 0,5 : 1 1 : 1 2 : 1 Tmauto:Tind ratio

Figure 4 Functional properties of in vitro generated memory T cells. Effector CD41 T lymphocytes stimulated with DCs as in Figure 2 were rested for 10 days in the absence (Control) or presence of IL-7. (a, b) IFN-c production in freshly isolated naive (CD45RA1) and memory (CD45RO1) T cells, and in Tmallo and Tmauto generated in the absence (Control) or presence of IL-7. T cells were stimulated with PMA and ionomycin (a) or with immobilized mAbs to CD3 and to CD28 (b), and then stained with an anti-IFN-c mAb. Cells were analysed by flow cytometry (c). Proliferative ability of freshly isolated naive, TCM and TEM memory cells, and of Tmallo and Tmauto generated as in (a, b). T cells were labelled with CFSE and stimulated for 4 days with immobilized mAbs to CD3 and to CD28. In a–c, data shown are the mean6s.d. of three to seven independent experiments. (d) Proliferative ability of CFSE-labelled Tmallo and Tmauto generated in the absence (Ctrl) or presence of IL-7, stimulated for 3 days with the Allo (grey bars) or Auto (black bars) DCs used for priming. The graph indicates the percentage of proliferating cells as the mean6s.d. of five donors evaluated. (e) Suppressive capacity of Tmauto. Naive CD41 (Tind naive) or Tmallo (Tind mem) from the same donor were labelled with CFSE and cultured with allogeneic DCs, in presence (grey bars) or absence (black bars) of different ratios of PKH26-labelled Tmauto generated with IL-7. At day 3 (Tind mem) or 4 (Tind naive), lymphocytes were harvested and CFSE dilution was analysed by flow cytometry. Shown is a graphical representation of the data (mean6s.d.) obtained from two to four separate donors, where the percentage of cycling Tind was set to 100%. Statistical analysis: *P,0.05. CFSE, carboxy-fluorescein diacetate succinimidyl ester; DC, dendritic cell; IFN, interferon; mAb, monoclonal antibody.

effectors even after long-term stimuli removal. IL-7-treated Tmallo exhibited low level of Erk-1/2 phosphorylation, analogous to freshly isolated memory CD41 T cells (Figure 5c and e); in contrast, Tmauto generated with IL-7 showed an extent of Erk-1/2 phosphorylation comparable to effector cells (Figure 5d and e). These results suggest that IL-7 may contribute to the acquisition of biochemical markers of memory T cells only when naive lymphocytes are activated above a certain threshold of TCR engagement. Effect of IL-7 in the proliferation and survival of memory CD41 T cells We previously observed that IL-7 led to higher recovery of Tmallo than of Tmauto (Figure 2c). Since the recovery is the sum of cell proliferation and survival, we evaluated these parameters during the resting period. First, effector cells were Cellular & Molecular Immunology

labelled with CFSE and maintained for 10 days in the presence or absence of IL-7. We detected residual proliferation in the absence of IL-7 in Tmallo and Tmauto (Figure 6a). IL-7 induced proliferation in both cell types, but the proliferation was more robust in Tmallo (Figure 6a). Moreover, IL-7 promoted higher survival rates in these cells, as demonstrated by the staining with annexin V and PI, while Tmauto did not seem to be affected at that level by the cytokine (Figure 6b). The increased susceptibility of Tmauto to cell death prompted us to determine the activation of the effector caspase-3. Although IL-7 decreased the percentage of active caspase-31 cells in both Tmallo and Tmauto with respect to their controls, the percentage of active caspase-31 cells in IL-7-treated Tmauto was significantly higher than in Tmallo (Figure 6c). Taken together, these data indicate that Tmallo are more responsive to IL-7 in comparison with Tmauto in terms of proliferation and survival.


269

a CD3

c

Memory 1°Effector _ _ + +

Naive _ +

CD3

Lck

Memory

Eff. Allo

_

_

+

Tmallo Ctrl _

+

Tmallo IL-7 _

+

+

pErk-1/2

Zap-70 Erk-1/2 LAT SLP-76

d CD3

pErk-1/2

pErk-1/2

Erk-1/2

_

+

_

- a-CD3 Naive

_

+

* * 2.5

400

512

1100

*

1

2

3

4

10 10 10 10 10

pErk/Erk ratio

512 0

1.5

a-CD3

* *

1.0

Ctrl

*

*

*

*

*

0.5 0.0

80% 0

pTyr

10 10 10 10 10

4

0

3

100 101 102 103 104

92%

93% 2

0

100 101 102 103 104

700

1100

93%

0

0

88%

100 101 102 103 104

0

Tmauto IL-7

+

*

1°Effector

Memory

95%

1

_

e

+ a-CD3

2.0

0

Tmauto Ctrl +

Erk-1/2

b

pSLP-76

Eff. Auto

Memory

b-actin

0

1

2

3

Memory Eff Auto Control 4

10 10 10 10 10

IL-7

Tmauto

Eff Allo Control

IL-7

Tmallo

Figure 5 TCR signalling in freshly isolated and in vitro generated human CD41 T lymphocytes. (a, b) Expression of lck, ZAP-70, LAT, SLP-76, bactin, phosphorylated (p) SLP-76, Erk-1/2, pErk-1/2 and total tyrosine phosphorylation (P-Tyr), in freshly isolated naive (CD45RA1) and memory (CD45RO1) CD41 T cells, and in primary (1u) effectors derived in vitro from naive cells, detected by western blot (a) or flow cytometry (b). In a, T cells (13106) were directly lysed (2) or previously stimulated with a mAb to CD3 (1) before lysis. Lysates were resolved by SDS–PAGE, blotted onto nitrocellulose membranes, and probed with the corresponding antibodies. In b, T cells were stimulated as above, but an anti-CD28 mAb was included together with the anti-CD3 mAb. Non-activated cells are shown as grey dotted lines, and activated cells as black solid lines. Numbers indicated the percentage of activated cells. Data are representative of three independent donors. (c–e) Erk-1/2 expression (Erk-1/2) and activation (pErk-1/2) in effector and memory lymphocytes generated in vitro compared to freshly isolated CD41CD45RO1 T cells (Memory). (c, d) Allogeneic (Eff Allo) and autologous (Eff Auto) effector cells generated as in Figure 2, as well as Tmallo and Tmauto generated in the absence (Ctrl) or presence of IL-7 for 10 days, were lysed in the absence (2) or presence (1) of anti-CD3 cross-linking. Protein extracts were resolved as in a and probed with the corresponding antibodies Shown is a representative experiment. (e) Densitometric analysis of pErk-1/2 and Erk-1/2 bands from the western blots of c and d. The results are the mean6s.d. of five independent donors, and are expressed as pErk-1/2/Erk-1/2 ratio.

IL-7 receptor signalling in Tmem In order to evaluate the mechanisms underlying the improved responses of allogeneic lymphocytes to IL-7, we first analysed the expression of IL-7 receptor (IL-7R). All naive lymphocytes expressed the a chain of IL-7R, which was progressively downregulated upon activation.35 Both allogeneic and autologous effectors showed low percentage of IL-7Ra1 cells (20%–30%) (Figure 7a). Notably, Tmauto and Tmallo upregulated IL-7Ra expression in the presence of IL-7 with respect to their effectors and to their counterparts cultured without IL-7 (Figure 7a). There were no substantial differences in the percentage of IL7Ra1 cells or in the amount of receptor/cell (measured by the mean fluorescence intensity) of IL-7Ra between Tmauto and

Tmallo in each condition (Figure 7a). However, upon IL-7 stimulation, we detected higher phosphorylation levels of the signal transducer and activator of transcription (STAT) 5 in Tmallo than in Tmauto (Figure 7b and c). STAT5 phosphorylation by IL-7 is mediated by members of the Janus kinase family such as Jak3.36 Thus, we next evaluated Jak3 activation by analyzing the level of tyrosine phosphorylation in the activation loop of its kinase domain with specific mAbs. The results depicted in Figure 7d indicate that the autologous and allogeneic Tmem had comparable levels of Jak3 phosphorylation upon IL-7 stimulation; hence, this kinase does not seem to be involved in the differential activation of STAT5 between Tmallo and Tmauto. Cellular & Molecular Immunology


270 Tmallo

Tmauto IL-7

91

M1

100

101

102 103

104

100

30

101

102 103

104

100

*

Ctrl IL-7

20 10 0

0

104

M1

0

102 103

81

M1

0 101

34

M1

0 100

130

26

IL-7

Ctrl 190

60

20

Ctrl

Proliferation index

a

101

102 103

Tmallo

104

Tmauto

CFSE

c

b

Eff

60

* Ctrl IL-7

50 40 30 20 10 0 Tmallo

Tmauto

Active caspase-3 (% cells)

% Anenin V-PI-T cells

*

Tmem Ctrl Tmem IL-7 *

25 20

*

*

*

15 10 5 0 Allo

Auto

Figure 6 Cell proliferation and survival of effector lymphocytes during the resting period. (a) Proliferation of effector T lymphocytes during the resting period. Effector T cells generated as in Figure 2 were purified and labelled with CFSE. Then, cells were cultured for 10 days in the absence (Ctrl) or presence of IL-7. The proliferation of Tmallo and Tmauto was evaluated by flow cytometry. A representative donor is shown in the left panel. The graph in the right panel represents the mean6s.d. of the proliferation index of four independent donors, calculated as the % cycling cells/% non-cycling cells. (b) Autologous and allogeneic lymphocytes were rested for 8 days (Tmauto and Tmallo, respectively) in the absence (Control) or presence of IL-7, then stained with annexin V and PI and analysed by flow cytometry. Shown is the percentage of viable cells (annexin V2PI2) as the mean6s.d. of four independent donors. (c) Percentage of cells expressing active caspase-3 in allogeneic (Allo) and autologous (Auto) effector and Tmem cells. The results are expressed as the average6s.d. of four different donors. Statistical analysis: *P,0.05. PI, propidium iodide.

The Tmem generated in our system displayed differences in their phenotype, essentially in CD25 expression. Thus, we considered the possibility that Tmallo and Tmauto could contain diverse proportions of T-cell subtypes, such as remaining effectors (CD251) together with more differentiated memory cells (CD252), and that the differential STAT5 phosphorylation could rely on such cell heterogeneity. To address this subject, we evaluated STAT5 activation in purified CD252 Tmallo and Tmauto. As shown in Figure 7e, IL-7 also induced higher phosphorylation of STAT5 in CD252 Tmallo than in CD252 Tmauto. Altogether, these data led us to conclude that lymphocytes activated with self-peptides responded less efficiently to IL-7 than lymphocytes receiving allogeneic stimulation as a consequence of their altered signalling through IL-7R. DISCUSSION The existence and the in vivo significance of the AMLR is a subject of controversy. Several studies indicate that this response takes place in vitro in the absence of xenogeneic/allogeneic Ags,10,37 and has been shown to bear memory and specificity.5 Moreover, its deficiency in vitro has been correlated with a variety of autoimmune diseases,38,39 suggesting a regulatory role of AMLR-activated cells.5,10,37 In this study, we have focused in the transition of these lymphocytes into memory cells particularly when they are primed in a setting resembling an active immune response, an issue that could be physiologically relevant since memory lymphocytes recognizing self-Ags Cellular & Molecular Immunology

are a powerful potential source of autoimmunity.7 We developed an in vitro model where we studied the fate of human polyclonal naive CD41 T cells stimulated with autologous mature DCs in the absence of foreign Ags and in the presence of IL-2, compared with those activated by allogeneic DCs. We choose to study alloresponses of naive T cells in order to compare them with responses to self-Ags because the TCR can recognize peptides bound to allogeneic MHC.40 Hence, our system would represent a truly comparison of low (self-peptides) and low-to-high (allo-peptides) strength of stimulation acting through the TCR. In vivo, most of the nonspecific activation of CD41 T lymphocytes has been associated with the memory T-cell pool, and is independent of TCR signalling.41–43 However, homeostatic maintenance and homeostatic expansion into lymphopenic hosts of naive CD41 T cells required TCR interaction with low-affinity self-peptides–MHC complexes and IL-7.3 Since most of the MHC molecules are occupied by self-peptides even during microbial infection,44 and these peptides partially overlap with those involved in thymic positive selection,45 it is feasible that in an ongoing immune response, a number of heterologous naive lymphocytes might be activated with these peptides presented by mature DCs and with the cytokines secreted by the responding T cells. In our system, the autologous stimulation would be mediated by self-Ags, cytokines (IL2), and probably by strong costimulation delivered by mature DCs, as it has been reported.7,10 In fact, we have proved that


271

a

Allo

Auto

* *

80

50 MFI IL-7Ra+ T cells

%IL-7Ra+ T cells

* 60 40 20

40 30 20 10

0

0 Tmem Ctrl

Effector

Tmem IL-7

Effector

b

Tmem Ctrl

c

Tmem IL-7

* *

IL-7

_

+

+

_

*

Ctrl +

-IL-7

Tmauto

Ctrl

IL-7

IL-7 _

Tmallo

_

pSTAT5/STAT5 ratio

Tmauto

Tmallo

+

pSTAT5

STAT5

+IL-7

2.5 2 *

1.5 1 0.5 0 Tmallo Ctrl

d IL-7 (min)

0

10

30

60

0

10

30

Tmallo IL-7

Tmauto IL-7

e

Tmauto

Tmallo

Tmauto Ctrl

60 IL-7

Tmauto

Tmallo

_

_

+

+

pJak3 Jak3

pSTAT5

pSTAT5 STAT5 STAT5

Figure 7 IL-7 receptor signalling in Tmem. (a) Percentage of cells expressing IL-7Ra (left) and MFI of IL-7Ra (right) in autologous (black) or allogeneic (grey) effectors and in Tmem generated in the absence (Ctrl) or presence of IL-7. Data shown are the mean6s.d. of seven independent donors. (b) STAT5 expression and activation in Tmallo and Tmauto generated in the absence (Ctrl) or presence of IL-7 for 8 days. Tmem were cultured for 1.5 h with (1) or without (2) IL-7, and then lysed. Protein extracts were blotted and probed with the corresponding antibodies. Shown is the result of a representative experiment out of five performed. (c) Densitometric analysis of five separate experiments performed as in b, expressed as the mean6s.d. of the pSTAT5/STAT5 ratio. (d) Jak3 and STAT5 expression and activation in Tmallo and Tmauto generated in the presence of IL-7 for 8 days. Tmem were cultured for 0, 10, 30 or 60 min with IL-7, and then lysed. Protein extracts were blotted onto nitrocellulose membranes and hybridized with the corresponding antibodies on the same membrane. (e) STAT5 expression and activation in CD252 Tmallo and Tmauto generated in the presence of IL-7 for 8 days. The cell lysates were obtained from Tmem cultured for 60 min with IL-7. Statistical analysis: *P,0.05. MFI, mean fluorescence intensity.

TCR stimulation in the AMLR is critical for naive T cells to proliferate in response to IL-2, which is in agreement with previous data.10,46 Then, it is plausible that the tonic activation with self-peptides together with costimulatory signals might upregulate the IL-2 receptors, enabling the T cells to respond to the cytokine. Memory T lymphocytes are resting cells that maintain several phenotypical and functional characteristics of effector cells, such as rapid expansion and production of effector

cytokines upon restimulation.47 It has been proposed that effectors return to a resting state in the absence of stimulation, and that in vitro rested effectors have attributes of memory cells in vivo.18,19 Thus, our model to generate memory cells included a resting period after DC stimulation, which last for up to 10 days. We showed that rested Tmem were non-proliferating cells with small size and complexity, analogous to conventional memory T cells. They also downregulated the activation marker CD25, produced IFN-c and proliferated at levels Cellular & Molecular Immunology


272

comparable to freshly isolated memory cells. In addition, we observed that most Tmem were CCR72CD62L1/2, in keeping with a TEM phenotype.24 Thus, Tmallo and Tmauto phenotypically and functionally behave as classical memory cells. However, we also showed that Tmauto are biochemically more similar to effector cells in terms of Erk-1/2 activation upon TCR triggering, and they also express higher levels of CD25 than Tmallo. These results suggest that the absence of activation stimuli may not be sufficient to acquire a complete memory profile in effector cells activated by autologous DCs, and other factors may be implicated. It has been documented that IL-7 protects human CD41 effector/memory T cells from death induced upon the deprivation of stimulation and cytokines.48 When the resting period was conducted without IL-7, the resulting Tmem appeared bigger and more complex than their counterparts cultured with IL-7, had residual levels of proliferation and survived poorly. We also observed that IL-7 is partially responsible, directly or indirectly, of downregulating CD25 in effector cells upon stimulation withdrawal, being more effective in Tmallo than in Tmauto. In that sense, we cannot completely rule out that the differences in CD25 expression between Tmallo and Tmauto might account for the preferential induction or maintenance of T regulatory lymphocytes in the AMLR, as it has been reported.6 However, this situation seems unlikely because Tmauto do not appear to have suppressive capacities, at least on allogeneic responses. As mentioned before, the surviving rested Tmallo and Tmauto did not seem to be incompetent to function as classical memory cells. However, one of the most relevant data of our study is that Tmallo had improved rates of proliferation and survival than Tmauto in response to IL-7. Both Tmem generated in the presence of IL-7 exhibited equivalent percentages of IL-7Ra1 cells, yet Tmallo displayed higher levels of phosphorylated STAT5 in response to IL-7. Our results indicate that Jak3, one of the major kinases involved in STAT5 phosphorylation, was activated at similar levels in Tmallo and Tmauto when they were stimulated with IL-7. These data suggest that the lower activation of STAT5 in Tmauto may result from a negative regulation of the IL-7 signalling pathway involving members of the Janus kinase family other than Jak3 (i.e. Jak1), the suppressor of cytokine signalling-1, or phosphatases such as Src homology 2-containing protein tyrosine phosphatases (SHP)-1 and -2.36 In this line, the differential phosphorylation of Erk-1/2 between IL-7-treated Tmallo and Tmauto might not only reflect their relationship with conventional memory cells, but also their signalling through IL-7R. Thus, IL-7 induced the transcription of dual-specificity phosphatase 5,49 a negative regulator of Erk1/2 activity. Moreover, Erk-1/2 could decrease STAT5 activation through phosphorylation of Grb2-associated binder family member (Gab2) and consequent SHP-2 activation.50 Therefore, the high Erk-1/2 activation in Tmauto might also be associated with defective signalling through IL-7R. The elucidation of these signalling pathways is currently under investigation. Cellular & Molecular Immunology

Altogether, the above data suggest that efficient responses to IL-7 of human CD41 effector/memory T cells depend on an adequate strength of stimulation through the TCR at priming. Several studies have proposed that the strength of stimulation during the primary response dictates memory formation and longevity.23,51 This has led to the concept of ‘T-cell fitness’,51 where weak total signal strength hampers the response of primed T cells to homeostatic cytokines such as IL-7, and therefore their persistence. Another level of interplay between IL-7 and TCR signalling is referred to the homeostatic maintenance of naive T cells. Here, the relative responsiveness to IL-7 appears to be partially regulated by the amount of TCR signalling, which relies on the affinity of the receptor and the availability of the specific selfligands.52 In our experimental approach, the avidity of TCRs for MHC–peptide complexes would be overall decreased in the AMLR compared with the allogeneic system, which could later affect the responses of the autologous cells to IL-7. Our results confirm those described by other authors in that TCR engagement with low-avidity peptides (mimicking self-peptides) in the presence of costimulation drives proliferation of naive CD41 T cells, but leads to massive cell death of cycling cells.8 Moreover, it has been suggested that it would function as a mechanism of tolerance maintenance during ongoing infections. Therefore, our study provides a putative mechanism to explain the poor lymphocyte recovery following such stimulation. Whether this mechanism operates in vivo is still unknown, but if it is the case, then it might be associated with peripheral tolerance to avoid unwanted responses from autoreactive memory lymphocytes. However, our system also revealed the possibility of a break of tolerance. Despite high numbers of autologous effectors die and do not generate memory cells, we have noticed that some Tmauto survived and responded to the challenge with self-Ag presented by mature autologous DCs, thus representing a potentially harmful pool of lymphocytes when activated in subsequent immune responses.7 Nonetheless, we postulate that this scenario is unlikely, and eventually these cells would die in the absence of a proper maintenance by IL-7. In summary, we presented here a mechanism by which naive CD41 T cells recognizing self-peptides in a setting resembling an active immune response are impaired to transit to memory cells in vitro, primarily based on their inefficient signalling through IL-7R. CONFLICT OF INTEREST The authors declare no conflict of interest.

ACKNOWLEDGEMENTS We gratefully acknowledge the generosity of the blood donors and the support of the Blood Bank staff at Centro Medico Nacional La Raza Hospital (Mexico City) in providing the blood samples. We also thank Victor H. Rosales for help with flow cytometry, Dr. Nicola´s Villegas for his aid in caspase-3 detection, Julio C. Ramı´rez for technical assistance, and Ms. Ninfa Arreola for her aid in the preparation of the manuscript.


273

This work was supported by grants to C.S.T. from CONACYT (no. 81112 and 140980).

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

Kuchroo VK, Anderson AC, Waldner H, Munder M, Bettelli E, Nicholson LB. T cell response in experimental autoimmune encephalomyelitis (EAE): role of self and cross-reactive antigens in shaping, tuning, and regulating the autopathogenic T cell repertoire. Annu Rev Immunol 2002; 20: 101–123. Tang AL, Bingaman AW, Kadavil EA, Leeser DB, Farber DL. Generation and functional capacity of polyclonal alloantigenspecific memory CD4 T cells. Am J Transplant 2006; 6: 1275–1284. Boyman O, Krieg C, Homann D, Sprent J. Homeostatic maintenance of T cells and natural killer cells. Cell Mol Life Sci 2012; 69: 1597– 1608. Cai G, Hafler DA. Multispecific responses by T cells expanded by endogenous self-peptide/MHC complexes. Eur J Immunol 2007; 37: 602–612. Weksler ME, Kozak R. Lymphocyte transformation induced by autologous cells. V. Generation of immunologic memory and specificity during the autologous mixed lymphocyte reaction. J Exp Med 1977; 146: 1833–1838. Verhasselt V, Vosters O, Beuneu C, Nicaise C, Stordeur P, Goldman M. Induction of FOXP3-expressing regulatory CD4pos T cells by human mature autologous dendritic cells. Eur J Immunol 2004; 34: 762– 772. Zwickey HL, Unternaehrer JJ, Mellman I. Presentation of selfantigens on MHC class II molecules during dendritic cell maturation. Int Immunol 2006; 18: 199–209. Choi JY, Craft J. Activation of naive CD41 T cells in vivo by a selfpeptide mimic: mechanism of tolerance maintenance and preservation of immunity. J Immunol 2004; 172: 7399–7407. Marshall HD, Prince AL, Berg LJ, Welsh RM. IFN-alphabeta and selfMHC divert CD8 T cells into a distinct differentiation pathway characterized by rapid acquisition of effector functions. J Immunol 2010; 185: 1419–1428. Scheinecker C, Machold KP, Majdic O, Ho¨cker P, Knapp W, Smolen JS. Initiation of the autologous mixed lymphocyte reaction requires the expression of costimulatory molecules B7-1 and B7-2 on human peripheral blood dendritic cells. J Immunol 1998; 161: 3966–3973. Chakir H, Lam DK, Lemay AM, Webb JR. ‘Bystander polarization’ of CD41 T cells: activation with high-dose IL-2 renders naive T cells responsive to IL-12 and/or IL-18 in the absence of TCR ligation. Eur J Immunol 2003; 33: 1788–1798. McKinstriy KK, Golech S, Lee WH, Huston G, Weng NP, Swain SL. Rapid default transition of CD4 T cell effectors to functional memory cells. J Exp Med 2007; 204: 2199–2211. Harrington LE, Janowsk KM, Oliver JR, Zajac AJ, Weaver CT. Memory CD4 T cells emerge from effector T-cell progenitors. Nature 2008; 452: 356–360. Kondrack RM, Harbertson J, Tan JT, McBreen ME, Surh CD, Bradley LM. Interleukin 7 regulates the survival and generation of memory CD4 cells. J Exp Med 2003; 198: 1797–1806. Purton JF, Martin CE, Surh CD. Enhancing T cell memory: IL-7 as an adjuvant to boost memory T-cell generation. Immunol Cell Biol 2008; 86: 385–386. Li J, Huston G, Swain SL. IL-7 promotes the transition of CD4 effectors to persistent memory cells. J Exp Med 2003; 198: 1807– 1815. Dooms H, Wolslegel K, Lin P, Abbas AK. Interleukin-2 enhances CD41 T cell memory by promoting the generation of IL-7R alphaexpressing cells. J Exp Med 2007; 204: 547–557. Harbertson J, Biederman E, Bennett KE, Kondrack RM, Bradley LM. Withdrawal of stimulation may initiate the transition of effector to memory CD4 cells. J Immunol 2002; 168: 1095–1102. Hu H, Huston G, Duso D, Lepak N, Roman E, Swain SL. CD41 T cell effectors can become memory cells with high efficiency and without further division. Nat Immunol 2001; 2: 705–710.

20 Brinkman CC, Sheasley-O’Neill SL, Ferguson AR, Engelhard VH. Activated CD8 T cells redistribute to antigen-free lymph nodes and exhibit effector and memory characteristics. J Immunol 2008; 181: 1814–1824. 21 Jelley-Gibbs DM, Brown DM, Dibble JP, Haynes L, Eaton SM, Swain SL. Unexpected prolonged presentation of influenza antigens promotes CD4 T cell memory generation. J Exp Med 2005; 202: 697–706. 22 Blair DA, Lefrançois L. Increased competition for antigen during priming negatively impacts the generation of memory CD4 T cells. Proc Natl Acad Sci USA 2007; 104: 15045–15050. 23 Williams MA, Ravkov EV, Bevan MJ. Rapid culling of the CD41 T cell repertoire in the transition from effector to memory. Immunity 2008; 28: 533–545. 24 Sallusto F, Geginat J, Lanzavecchia A. Central memory and effector memory T cell subsets: function, generation, and maintenance. Annu Rev Immunol 2004; 22: 745–763. 25 Torres-Aguilar H, Aguilar-Ruiz SR, Gonza´lez-Pe´rez G, Munguıá R, Bajanã S, Meraz-Rıós MA et al. Tolerogenic dendritic cells generated with different immunosuppressive cytokines induce antigen-specific anergy and regulatory properties in memory CD41 T cells. J. Immunol 2010; 184: 1765–1775. 26 Geginat J, Sallusto F, Lanzavecchia A. Cytokine-driven proliferation and differentiation of human naive, central memory, and effector memory CD41 T cells. J Exp Med 2001; 194: 1711–1719. 27 Ahmadzadeh M, Hussain SF, Farber DL. Effector CD4 T cells are biochemically distinct from the memory subset: evidence for longterm persistence of effectors in vivo. J Immunol 1999; 163: 3053– 3063. 28 Chandok MR, Okoye FI, Ndejembi MP, Farber DL. A biochemical signature for rapid recall of memory CD4 T cells. J Immunol 2007; 179: 3689–3698. 29 Farber DL. Differential TCR signaling and the generation of memory T cells. J Immunol 1998; 160: 535–539. 30 Farber DL, Acuto O, Bottomly K. Differential T cell receptor-mediated signaling in naive and memory CD4 T cells. Eur J Immunol 1997; 27: 2094–2101. 31 Hussain SF, Anderson CF, Farber DL. Differential SLP-76 expression and TCR-mediated signaling in effector and memory CD4 T cells. J Immunol 2002; 168: 1557–1565. 32 Watson ARO, Lee WT. Differences in signaling molecule organization between naive and memory CD41 T lymphocytes. J Immunol 2004; 173: 33–41. 33 Krishnan S, Warke VG, Nambiar MP, Wong HK, Tsokos GC, Farber DL. Generation and biochemical analysis of human effector CD4 T cells: alterations in tyrosine phosphorylation and loos of CD3f expression. Blood 2001; 97: 3851–3859. 34 Krishnan S, Warke VG, Nambiar MP, Tsokos GC, Farber DL. The FcRc subunit and Syk kinase replace the CD3f-chain and ZAP-70 kinase in the TCR signaling complex of human effector CD4 T cells. J Immunol 2003; 170: 4189–4195. 35 Lozza L, Rivino L, Guarda G, Jarrossay D, Rinaldi A, Bertoni F et al. The strength of T cell stimulation determines IL-7 responsiveness, secondary expansion, and lineage commitment of primed human CD41IL-7Rhi T cells. Eur J Immunol 2008; 38: 30–39. 36 Al-Rawi MA, Mansel RE, Jiang WG. Interleukin-7 (IL-7) and IL-7 receptor (IL-7R) signalling complex in human solid tumours. Histol Histopathol 2003; 18: 911–923. 37 Suzuki R, Suzuki S, Takahashi T, Kumagai K. Production of a cytokine with interleukin 3-like properties and cytokine-dependent proliferation in human autologous mixed lymphocyte reaction. J Exp Med 1986; 164:1682–1699. 38 Knospe S, Ko¨hlerE, Rjasanowski I, Titze K. Autologous mixed lymphocyte reaction in newly diagnosed type-1 diabetes. Exp Clin Endocrinol 1990; 95: 148–156. 39 Hirsch RL. Defective autologous mixed lymphocyte reactivity in multiple sclerosis. Clin Exp Immunol 1986; 64: 107–113. 40 Felix NJ, Donermeyer DL, Horvath S, Walters JJ, Gross ML, Suri A et al. Alloreactive T cells respond specifically to multiple distinct peptide– MHC complexes. Nat Immunol 2007; 8: 388–397. Cellular & Molecular Immunology


274 41 Eberl G, Brawand P, MacDonald HR. Selective bystander proliferation of memory CD41 and CD81 T cells upon NK T or T cell activation. J Immunol 2000; 165: 4305–4311. 42 Di Genova G, Roddick J, McNicholl F, Stevenson FK. Vaccination of human subjects expands both specific and bystander memory T cells but antibody production remains vaccine specific. Blood 2006; 107: 2806–2813. 43 Bangs SC, Baban D, Cattan HJ, Li CK, McMichael AJ, Xu XN. Human CD41 memory T cells are preferential targets for bystander activation and apoptosis. J Immunol 2009; 182: 1962–1971. 44 Germain RN, Stefanova´ I. The dynamics of T cell receptor signaling: complex orchestration and the key roles of tempo and cooperation. Annu Rev Immunol 1999; 17: 467–522. 45 Marrack P, Ignatowicz L, Kappler JW, Boymel J, Freed JH. Comparison of peptides bound to spleen and thymus class II. J Exp Med 1993; 178: 2173–2183. 46 Ge Q, Palliser D, Eisen HN, Chen J. Homeostatic T cell proliferation in a T cell-dendritic cell coculture system. Proc Natl Acad Sci USA 2002; 99: 2983–2988. 47 Rogers PR, Dubey C, Swain SL. Qualitative changes accompany memory T cell generation: faster, more effective

Cellular & Molecular Immunology

48

49

50

51

52

responses at lower doses of antigen. J Immunol 2000; 164: 2338–2346. Chetoui N, Boisvert M, Gendron S, Aoudjit F. Interleukin-7 promotes the survival of human CD41 effector/memory T cells by up-regulating Bcl-2 proteins and activating the JAK/STAT signalling pathway. Immunology 2010; 130: 418–426. Kovanen PE, Rosenwald A, Fu J, Hurt EM, Lam LT, Giltnane JM et al. Analysis of gamma c-family cytokine target genes. Identification of dual-specificity phosphatase 5 (DUSP5) as a regulator of mitogenactivated protein kinase activity in interleukin-2 signaling. J Biol Chem 2003; 278: 5205–5213. Arnaud M, Crouin C, Deon C, Loyaux D, Bertoglio J. Phosphorylation of Grb2-associated binder 2 on serine 623 by ERK MAPK regulates its association with the phosphatase SHP-2 and decreases STAT5 activation. J Immunol 2004; 173: 3962–3971. Gett AV, Sallusto F, Lanzavecchia A, Geginat J. T cell fitness determined by signal strength. Nat Immunol 2003; 4: 355– 360. Kieper WC, Burghardt JT, Surh CD. A role for TCR affinity in regulating naive T cell homeostasis. J Immunol 2004; 172: 40–44.