Skip Navigation


International Immunology Advance Access originally published online on May 4, 2006
International Immunology 2006 18(7):1001-1015; doi:10.1093/intimm/dxl035
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
18/7/1001    most recent
dxl035v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (4)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Qin, H.-Y.
Right arrow Articles by Singh, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Qin, H.-Y.
Right arrow Articles by Singh, B.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?


© The Japanese Society for Immunology. 2006. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org

A novel mechanism of regulatory T cell-mediated down-regulation of autoimmunity

Hui-Yu Qin1, Rinee Mukherjee1, Edwin Lee-Chan1, Catherine Ewen2, R. Chris Bleackley2 and Bhagirath Singh1,3,4

1 Department of Microbiology and Immunology, University of Western Ontario, London, Ontario, N6A 5C1, Canada
2 Department of Biochemistry, University of Alberta, Edmonton, Alberta, Canada
3 Robarts Research Institute, London, Ontario, Canada
4 Institute of Infection and Immunity, Canadian Institutes of Health Research, London, Ontario, Canada

Correspondence to: B. Singh; E-mail: bsingh{at}uwo.ca


   Abstract
Top
 Abstract
Introduction
 Methods
 Results
 Discussion
 References
 
We have established a novel CD4 and CD8 double-positive CD25+ T regulatory (Treg) clone, MT-5B, from lymph nodes of type 1 diabetes prone non-obese diabetic (NOD) mice immunized with CFA. CFA has previously been shown to prevent the onset of diabetes by inducing Treg cells. In vitro, clone MT-5B was anergic to a panel of antigen stimulations and exerted an immunosuppressive effect in antigen-non-specific and cell contact-independent manners. In vivo, clone MT-5B blocked the adoptive transfer of diabetes. Proteomics and immunoadsorption studies identified the suppressive proteins secreted by clone MT-5B as granzyme B (GrB) and perforin (PFN). GrB-mediated immune suppression was PFN dependent. Removal of GrB or PFN from the culture supernatant (SN) of MT-5B cells or pre-incubation of MT-5B cells with ethyleneglycol-bis(aminoethylether)-tetraacetic acid which blocks PFN activity reduced the immunosuppressive effect in vitro. Pre-incubation of diabetogenic splenocytes from NOD mice with MT-5B SN impaired their ability to transfer disease by inducing T cell apoptosis, and removal of GrB from MT-5B SN by immunoadsorption decreased the effector function of MT-5B SN on diabetogenic splenocytes. Immunization of NOD mice with CFA increased the expression of GrB+ CD4 T cells, indicating that these cells are present in vivo. In conclusion, we describe a novel mechanism of cell contact-independent immune suppression in which Treg cells maintain immune homeostasis by secreting GrB/PFN.

Keywords: autoimmunity, CFA, diabetes, granzyme B/perforin, regulatory T cells


   Introduction
Top
 Abstract
 Introduction
Methods
 Results
 Discussion
 References
 
The discovery of CD4+ regulatory T (Treg) cells has provided a new understanding of the phenomenon of immunosuppression (1). Naturally arising and induced CD4+ Treg cells inhibit both the induction and effector function of autoreactive T cells in autoimmunity (2–5).

CD4+ Treg cells can be further differentiated into natural and adaptive Treg subsets based on their distinct ontogeny, mode of action and suppressive function (3). Each subset of Treg cells expresses CD25. The naturally occurring, thymus-derived CD4+CD25+ Treg cells have a cell contact-dependent, cytokine-independent mechanism of action (6, 7). Adaptive Treg cells generated in the periphery mediate immune responses by secreting suppressive cytokines and include IL-10-producing Tr1 and transforming growth factor (TGF)-ß-producing Th3 cells (8–10). The CD4+CD25+ Treg cells maintain immune homeostasis by controlling autoimmunity, tumor immunity, infection immunity and transplantation tolerance (11, 12). In type 1 diabetes (T1D), both CD4+ T cells (13) and CD4+CD25+ Treg cells have been widely reported as being involved in the prevention of diabetes in animal model (5, 14–16). The expressions of TGF-ß, cytotoxic T lymphocyte-associated antigen (CTLA)-4 and glucocorticoid-induced tumor necrosis factor receptor family-related gene (GITR) have been shown important to the effector function of CD4+CD25+ Treg cells (17–20), but they are not necessary for immunosuppression (21–23). It was recently discovered that the transcription factor encoded by the Foxp3 gene is exclusively expressed in CD4+CD25+ Treg cells and is vital for their development and function (24, 25).

Granzyme B (GrB) is an important member of the granzyme family (26, 27). GrB and perforin (PFN) are the effector molecules that mediate target killing by NK cells and CTLs in viral infection and anti-tumor immunity. Dysregulation of this pathway is associated with certain human diseases and genetic abnormalities in mice (28). The expression of GrA and/or GrB has recently been investigated in human and murine CD4+CD25+ Treg cells (29, 30). Similar to NK cells and CTLs, it has been shown that GrB-expressing Treg cells may function in a cell contact-dependent manner. GrB and PFN work synergistically to exert a cytotoxic effect on target cells. The mechanisms underlying the delivery of GrB to the target cells may involve transmembrane pores made by PFN (31), non-specific charge interaction (32) and/or cation-independent mannose 6-P receptor-mediated endocytosis (33). Serglycin, a proteoglycan, has been shown recently to form an apoptosis-inducing multimeric complex with granule proteins GrB and PFN through non-covalent linkage (34, 35). The precise mechanism underlying PFN-dependent delivery, intracellular trafficking and functioning of GrB is still under investigation.

In this study, MT-5B, a double-positive (DP) CD25+ Treg clone established from CFA-protected non-obese diabetic (NOD) mice was analyzed. Although this clone produces immunosuppressive cytokines, we found that these cytokines were not involved in MT-5B-mediated suppression. We identified the suppressive proteins secreted by MT-5B as GrB/PFN by using proteomic and immunoadsorption approaches. Unlike natural CD4+CD25+ Treg cells and GrB-expressing NK cells, CTLs and CD4+ Treg cells, the secretion and effector functions of GrB/PFN by MT-5B cells are cell contact independent. Pre-incubation of diabetogenic splenocytes with GrB/PFN-containing supernatant (SN) down-regulates their ability to transfer disease in a GrB-dependent manner by inducing T cell apoptosis. We report for the first time a novel mechanism involving the modulation of immune responses by DP CD25+ Treg cells through the secretion of GrB/PFN in a cell contact-independent manner.


   Methods
Top
 Abstract
 Introduction
 Methods
Results
 Discussion
 References
 
Mice
NOD/LtJ and NOD.SCID mice were bred in specific pathogen-free conditions at the Robarts Research Institute animal facilities (London, Ontario, Canada). Animals were used according to the guidelines of the institutional animal care committee at the University of Western Ontario. For adoptive transfer studies, irradiated (800 rads) NOD and NOD.SCID mice were used as recipients. Diabetic NOD mice were kept on daily subcutaneous injection of 1 U human insulin (Humulin, Eli Lilly Co., Indianapolis, USA) until the day of the experiment. C57BL/6 mice were purchased from the Jackson Laboratory (Bar Harbor, ME, USA).

Antibodies and reagents
In this study, the following mAbs were used for flow cytometry, western blot, immunoadsorption and ELISA: FITC- or PE-conjugated mAbs to CD3 (145-2C11), CD4 (GK 1.5), CD8{alpha} (53-6.7), CD11C (HL3), CD25 (PC61), CD45RB (16A), CD80 (16-10A1), CD86 (GL1), CD90-2 (30-H12), TCR{alpha}ß (H57-597), CTLA-4 (UC10-4F10-11), Fas (Jo2), FasL (MFL3); PE-conjugated and purified (NA/LETM) mAbs to IFN-{gamma} (XMg1.2), tumor necrosis factor (TNF)-{alpha} (TN3-19.12), TGF-ß (MAB1835), IL-10 (JES5-2A5), CD3 (145-2C11), isotype controls, avidin–Alkaline Phosphatase and mouse rIL-2 (BD PharMingen, Ontario, Canada); goat anti-mouse GITR, purified and biotin-conjugated goat anti-mouse GrB antibodies (R&D systems, Ontario, Canada); PE-conjugated mAbs to Foxp3 (FJK-16s) and CD8ß (CT-CD8b), purified, PE and biotin-conjugated anti-mouse PFN mAbs (clone JAW 246, eBioscience, San Diego, USA). The following reagents were also used in this study: purified GrB protein, extrafluorescence-avidin, ethyleneglycol-bis(aminoethylether)-tetraacetic acid (EGTA), propidium iodide (PI) and Concanavalin (Con A) (Sigma, Ontario, Canada); FITC–TdT-mediated dUTP nick end labeling (TUNEL, Roche, Mannheim, Germany); Mycobacterium bovis BCG (Bacille Calmette–Guérin, Sanofi-Pasteur, Toronto, Canada); mouse recombinant glutamic acid decarboxylase (GAD67) was provided by J. F. Elliott (Edmonton, Canada).

Establishment of CD4+CD8+ CD25+ Treg clones
Four-week-old NOD mice were immunized with CFA in the hind footpad as previously described (36). Ten days later, popliteal lymph nodes were harvested and single-cell suspensions were prepared in complete RPMI 1640 medium supplemented with 25 mM HEPES, 2 mM glutamine, 5 x 10–5 M 2-mercaptoethanol, 100 U ml–1 penicillin, 100 µg ml–1 streptomycin and 10% heat-inactivated FCS. Cells were seeded in 96-well flat-bottom microtitre plates in the presence of 50 ug ml–1 BCG media were changed 5 days later. Ten to 14 days after the initiation of culture, cells well grown in cultures were transferred into 24-well plates in the presence of irradiated syngeneic spleen cells (2500 rads) as antigen-presenting cells (APCs) plus rat natural IL-2 (10 U ml–1, BD Labware, MA, USA). T cell lines were cloned and sub-cloned by limiting dilution in 0.3–0.5 cells per well in the presence of APC and IL-2 as above. MT-5B is one of the clones established and used in this study. Clone MT-5B cells grow in a lightly adherent manner and can be detached by culturing at 4°C for 10 min. Cells were maintained in complete RPMI medium supplemented with 5% FCS and 50 U ml–1 of mouse rIL-2.

Preparation of SN and cell lysates
For preparation of SN, MT-5B cells were washed three times with PBS to remove the residual FCS and incubated for 24–36 h in FCS-free RPMI at a concentration of 106 cells ml–1. Control SN was from nylon wool-enriched splenic T cells of 8-week-old NOD mice (36). SN was also concentrated 40-fold by using Centricon Plus-80 (PL-10 membrane, Millipore Corporation, MA, USA). Cell lysates (LS) were prepared by repeated freeze-thaw cycles in PBS. Equal volumes of concentrated supernatants (SN.C) and LS were prepared from the same number of MT-5B cells. GrB was undetectable in control SN, and was 292, 2602 and 1684 ng ml–1 in MT-5B SN, SN.C and LS, respectively, as determined by ELISA.

Cell proliferation assays

(1) For proliferation assays, 2 x 105 MT-5B cells were incubated in complete RPMI 1640 medium in the presence of Con A (4 µg ml–1), BCG (100 µg ml–1), GAD67 (10 µg ml–1), anti-CD3 mAb (1 µg ml–1) in the presence of 4 x 105 irradiated (2500 rads) syngeneic APC, or C57BL/6 splenocytes (mixed lymphocyte reaction, MLR) for 4–5 days. Nylon wool-enriched splenic T cells of 8-week-old NOD mice were used as controls. [3H]thymidine ([3H]TdR) (0.5 µCi per well) (PerkinElmer Life Analytical Science, Boston, MA, USA) was added to the plate 16 h before cell harvesting. Uptake of [3H]TdR was measured by using Liquid Scintillation Counter (MicroBeta, PerkinElmer Wallac, Québec, Canada).
(2) For co-culture assays, MT-5B cells, MT-5B SN, anti-GrB/PFN-adsorbed, trypsin-treated SN (120 µg ml–1 SN at 37°C for 30 min), corresponding control T cells or SN were added to the NOD splenocyte cultures (2 x 105 to 4 x 105 per well) at indicated regulatory to responder cell ratios or concentrations of SN.
(3) For antibody-blocking assays, each of neutralizing mAbs to cytokines was added to the culture of Con A response or two-way MLR in the presence of MT-5B cells or MT-5B SN.
(4) For transwell experiments, 105 MT-5B cells were mixed with or separated from the splenocyte cultures (2 x 106) in 24-well plate by a cell culture insert with 1.0-µm pore size at 1.5 ml medium (Becton Dickinson Labware, NJ, USA). The cell suspension from each well was then collected and distributed equally to five wells of 96-well plate just before adding [3H]TdR.
(5) For EGTA-blocking assay, MT-5B cells were pre-incubated with different concentrations of EGTA in complete medium for different durations. After washing, 0.1 ml of MT-5B cell suspension (2 x 105 ml–1) was added to each well of 96-well plates and incubated for another 8 h. Then, MT-5B cells or SN (1:3) alone and MT-5B plus SN were separately tested by adding 24-h splenocyte Con A blasts as responder cells at 1:20 ratio of regulator:responder.

Adoptive transfer of diabetes
In adoptive transfer of diabetes experiment, 12 x 106 diabetogenic splenocytes from acutely diabetic NOD mice were transferred or co-transferred (intravenously) with 6 x 106 MT-5B cells or control NOD T cells into pre-irradiated (800 rads) NOD mice. In some experiments, diabetogenic splenocytes were pre-incubated with MT-5B SN, control SN (1:20), GrB antibody or control IgG-adsorbed MT-5B SN.C (1:100) at cell concentration of 2 x 106 cells ml–1 for 36 h before transferred into NOD.SCID recipient mouse. Urinary glucose was monitored every other day beginning at 14 days after adoptive transfer with test strips (Miles Canada Inc., Rexdale, Ontario, Canada), and verified for diabetes by examining blood glucose levels (≥300 mg dl–1) with a GlucosCan 2000 blood glucose monitor (Lifescan, Mountain View, CA, USA).

Protein extraction, gel filtration chromatography and mass spectral analysis
MT-5B SN.C (1 ml) was run on a 12% polyacrylamide preparation gel under non-reducing conditions. Proteins were extracted from excised slices with 3 ml buffer (0.05 M Tris–HCl, 0.1 mM EDTA, 1 mM dithiothreitol, 0.1 mg ml–1 BSA and 0.15 M NaCl), and precipitated by adding four volumes of acetone (–20°C) in siliconized Corex tube for 30 min. After centrifugation for 10 min at 10 000 rpm, the protein pellet was rinsed with ice-cold 80% acetone: 20% buffer and reconstituted with 1 ml PBS. For gel filtration, SN.C (0.5 ml) was loaded on 10 x 250 mm Sephacryl S-100 column (Pharmacia LKB Biotechnology AB, Uppsala, Sweden). Proteins were eluted with 0.05 M sodium phosphate buffer, 0.15 M NaCl, pH 7.0 at a flow rate of 0.5 ml per tube min–1 and monitored by spectrophotometry (Spectrospec III, LKB Biotechnology AB, Uppsala, Sweden). Each tube of eluates was tested for the ability to suppress a Con A-induced proliferative response. Selected tubes with or without suppressive activity were run on 12% polyacrylamide gels under reducing conditions. Protein bands corresponding to the suppressive activity in SN.C were excised and subjected to mass spectral analysis following in-gel-tryptic digestion at the University of Western Ontario Biological Mass Spectrometery Laboratory using MicroMass ESI-Q-TOF machine.

Analysis of cell-surface markers, intracellular molecules and apoptosis by flow cytometry
For surface staining, cells were directly or indirectly stained with mAbs at 4°C for 30 min. For intracellular staining, cells or surface-stained cells were fixed in 2% formaldehyde–PBS for 15 min at room temperature (RT). After washing with 2% BSA–PBS washing buffer and 0.5% saponin-washing buffer, cells were permeabilized and stained with PE-conjugated antibodies or biotin–avidin system in 0.5% saponin-containing buffer at RT for 30 min. For PI staining, cells were collected from the cultures, washed and stained with 50 µg ml–1 of PI in 0.1% Triton-X 100, 0.1% sodium citrate–PBS. TUNEL staining was performed following manufacturer's instructions. Briefly, surface-stained cells were fixed and permeabilized with 0.1% sodium citrate, 0.1% Triton-X 100–PBS. Cells were finally incubated with TUNEL–FITC at 37°C for 60 min. In each step, cells were washed twice with washing buffer. Samples were analyzed with a FACScan Flow Cytometer (BD, Sunnyvale, CA, USA) using CELLQuest II software. Ten thousand events were acquired per sample.

Light and confocal microscopy
MT-5B cells were settled on a glass slide and air dried. Cells were stained with Wright–Giemsa (Sigma) following manufacturer's instructions. Cells were examined using light microscopy (1000x) (Carl Zeiss Inc., Thornwood, NY, USA). For confocal analysis, cells were stained as previously described for intracellular staining prior to flow cytometric analysis. Subsequently, cells were washed twice in PBS and incubated at RT on poly-L-lysine-coated glass coverslips (Sigma–Aldrich Canada Ltd., Oakville, Canada) for10 min. Coverslips were directly mounted on glass slides using Vectashield (Vector Laboratories Inc., Burlingame, CA, USA) and analyzed using a confocal laser scanning microscope. Confocal microscopy was performed using an Axiovert 100M inverted microscope equipped with a LSM 510 laser scanning unit and a 63X/1.40 NA plan Apochromat objective (Carl Zeiss Inc.). For dual analyses, green fluorescence was detected at >515 nm after excitation at 488 nm, and red fluorescence was detected at >585 nm after excitation at 615 nm.

ELISA
A sandwich ELISA was performed to detect GrB in SN and LS (5). Briefly, ELISA plates were coated with capture antibody (0.5 to 1 µg ml–1) in NaHCO3 buffer, pH 8.6 at 4°C overnight. After saturation with 0.25% Tween–PBS buffer, 50 µl samples were added to each well. The specific binding was detected by a standard antibody–biotin and avidin–AKP system. In each step, the plates were washed with 0.05% Tween–PBS buffer and incubated at 4°C for 3 h. Reactions were developed with phosphatase substrate (p-nitrophenyl phosphate disodium, Sigma). Optical densities were measured at 405 nm by ELISA reader (Bio-Rad, Hercules, CA, USA). The detectable level of GrB was 3–6 ng ml–1.

Western blot
LS, SN.C (10 µl per lane), purified GrB (100 ng per lane), anti-GrB antibody-adsorbed SN.C or GrB adsorbed on beads were run on 12% polyacrylamide gels under reducing conditions. The gel-transferred Immobilon-P Transfer Membrane (Millipore Corporation, MA, USA) was saturated with 0.5% gelatin in Tris buffer–0.05% Tween 20 at RT for 1 h or at 4°C overnight. Membranes were sequentially incubated with biotin–antibodies to GrB or PFN and avidin–AKP (1:1500) for 1 h each at RT. Rinsing and washing in Tris-buffered saline (TBS)–0.05% Tween 20 buffer were performed after each step. Finally, membranes were incubated in nitro blue tetrazolium/5-bromo-4-chloro-3-Indolyl phosphate chromatin substrate-AKP buffer until the desired levels of color were observed. The reaction was stopped by incubation of membrane with 2 mM EDTA–PBS solution.

Immunoadsorption assay
Two to four micrograms of goat anti-mouse GrB polyclonal antibody, rat anti-mouse PFN mAb or corresponding isotype controls was bound to 30–50 µl of packed GammaBindTM G-SepharoseTM beads (Amersham Biosciences AB, Uppsala, Sweden) in reaction tube with cap (0.2 ml) by incubation in 100 µl of TBS-0.1% BSA buffer, pH 7.5 at 4°C overnight. Beads were washed with buffer and incubated with SN.C, LS (50 to 100 µl) or 0.5 µg of purified GrB and diluted to 100 or 200 µl with buffer. The mixtures were incubated at 4°C for 8 h with rotation. Following centrifugation, antibody-adsorbed SN.C were analyzed by western blotting, tested for suppression of T cell proliferation response or incubated with diabetogenic splenocytes before adoptive transfer of diabetes. The beads were extensively washed with TBS buffer and the immune complexes were eluted from the beads in reducing sample buffer, with ~20 µl of eluate applied to each lane of a 12% polyacrylamide gel. Western blot and/or coomassie blue and silver staining were used for detecting PFN and/or GrB.

Statistical analysis
For statistical analysis, Student's t-test, two-way analysis of variance (Student–Newman–Keuls) and Kaplan–Meier life survival curves were used. For all analyses, a P-value < 0.05 was considered statistically significant.


   Results
Top
 Abstract
 Introduction
 Methods
 Results
Discussion
 References
 
Phenotypic analysis of clone MT-5B
Clone MT-5B was established from the popliteal lymph nodes of 4-week-old NOD mice immunized with CFA (13, 37). The phenotype patterns of clone MT-5B are shown in Fig. 1. This clone has high-level expression of CD25 and high to intermediate levels of CD11c, CD28, CD45RB, CD80, CD86, TCR{alpha}ß and GITR on the cell surface. Intermediate levels of intracellular expression of CTLA-4 and Foxp3 were observed, while surface expression of CD3 and CD62L were low. A significant feature of clone MT-5B is the dual expression of CD4 and CD8{alpha}{alpha}. Clone MT-5B does not express Fas, FasL, CD11b, CD69 or detectable surface TGF-ß (data not shown). The expression of intracellular IFN-{gamma}, TNF-{alpha} and TGF-ß are much higher than that of IL-10 in these cells. Clone MT-5B is characterized by an activated and mature memory T cell phenotype.


Figure 1
View larger version (48K):
[in this window]
[in a new window]
  		Fig. 1 Phenotypic analysis of clone MT-5B. MT-5B cells were stained directly or indirectly with a panel of labeled antibodies and analyzed by flow cytometry as described in Methods. As shown in overlap histogram plots, cells stained with antibody and isotype control are expressed in solid and dotted lines, respectively. Dot plot shows the double staining of FITC–CD4 and PE–CD8{alpha}. Data are representative of at least three independent experiments showing similar results.

 
In vitro functional characterization of clone MT-5B
Compared with splenic T cells from normal NOD mice, MT-5B cells did not respond to a large panel of stimulations including BCG, GAD67, Con A, anti-CD3 mAb and allogeneic stimulation with C57BL/6 splenocytes (Fig. 2A). In co-culture assays, MT-5B cells non-specifically suppressed both Con A response and MLR (Fig. 2B). MT-5B cell-induced suppression had no strain restriction, as it could also suppress Con A responses and MLR derived from C57BL/6 or BALB/c mice (data not shown). Therefore, clone MT-5B is anergic, and MT-5B-induced suppression is antigen non-specific.


Figure 2
View larger version (16K):
[in this window]
[in a new window]
  		Fig. 2 MT-5B cells are anergic and exert a suppressive effect in an antigen-non-specific manner. (A) In vitro proliferative responses of MT-5B cells were compared with NOD T cells by incubation with a panel of stimuli for 4–5 days in the presence of irradiated syngeneic APC or allogeneic splenocytes from C57BL/6 (MLR). (B) The suppressive effects of MT-5B cells were assessed in Con A response and MLR cultures at the indicated ratio of regulator to responder, and compared with irradiated T cells (2500 rads) from NOD mice. The dotted line indicates the level of Con A response or MLR without modification. Results are expressed as mean counts per minute (CPM) ± SD of triplicates, and representative data from one of at least three experiments are presented.

 
MT-5B cell-induced immunosuppression is cell contact independent
It has been reported that Treg cells exert their suppressive effect in either a cell contact-dependent or a suppressive cytokine IL-10 and/or TGF-ß-mediated contact-independent manner (7–10). We found that SN from MT-5B cells also exerted suppressive effect on Con A response (Fig. 3A) and MLR (Fig. 3B). The suppressive activity of MT-5B cells was found to be cell contact independent, as MT-5B cells in cultures or in transwell chambers showed the same suppressive effect in T cell proliferative responses (Fig. 3C). The suppressive activity of MT-5B SN was shown to be trypsin sensitive, which is indicative of the proteinaceous nature of the suppressive molecules (Fig. 3D). Since clone MT-5B expressed intermediate levels of IFN-{gamma}, TNF-{alpha}, TGF-ß and low level of IL-10 (Fig. 1), we investigated the possible role of these cytokines in MT-5B-induced suppression. In vitro blocking assays were performed by adding neutralizing anti-cytokine mAbs to the culture of Con A response or MLR in the presence or absence of MT-5B cells or MT-5B SN. As shown in Fig. 3(E), we did not find significant blockade of suppression at the mAb concentration which effectively neutralizes the corresponding cytokine. Similar results were also observed when mAbs were added to the cultures in various combinations (data not shown). On the other hand, the amount of suppressive cytokines secreted in SN by MT-5B cells was much less than the concentration of purified cytokines needed to induce minimal suppression (data not shown). Therefore, clone MT-5B-induced suppression is not attributed to the cytokines secreted by these cells. Our results clearly indicate that suppressive effects induced by MT-5B cells are cell contact independent, and mediated by some unidentified proteins.


Figure 3
View larger version (25K):
[in this window]
[in a new window]
  		Fig. 3 Immunosuppression induced by clone MT-5B is cell contact-independent. SN from MT-5B cells suppressed both Con A response (A) and MLR (B) at indicated concentrations. Cell contact-independent suppression was also verified by transwell chamber in which MT-5B cells were separated from the responder cells in cultures at regulator to responder ratio of 1:20 (C). The proteinaceous nature of suppressive component in MT-5B SN was indicated by the lose of activity after pre-incubation of MT-5B SN with trypsin (120 µg ml–1) at 37°C for 30 min (D). The role of cytokines in clone MT-5B-induced suppression was evaluated in Con A response and MLR (two way) by the blocking assays with neutralizing mAbs to IFN-{gamma}, TNF-{alpha}, TGF-ß or IL-10 (20 µg ml–1) in the absence or presence of MT-5B cells (regulator to responder ratio of 1:30) or MT-5B SN (1:30) (E). The normal levels of Con A response and MLR are shown as solid bars. Isotype controls had no significant effect (data not shown). Results are expressed as mean CPM ± SD of triplicates and are representative of at least three experiments. The dotted line represents the level of Con A response or MLR without modification.

 
Identification of GrB in SN from MT-5B cell culture by proteomic analysis
Given the suppressive effects of SN from MT-5B cells, we further investigated the proteins in SN.C by proteomic analysis in two ways. Firstly, proteins in the SN.C were separated by PAGE under non-reducing conditions. Bands designated B1, B2, B3 and a control band B4 were eluted and precipitated by acetone extraction (Fig. 4A) and tested for their suppressive activities (Fig. 4B). We found that band B2 with a molecular weight (MW) of 65 kDa had maximal activity. Secondly, proteins in the SN.C were separated by Sephacryl S-100 gel filtration and various fractions were tested for suppressive activity (Fig. 4C) and run on a 12% polyacrylamide gel under reducing conditions (Fig. 4D). The profiles of bands from fraction 6 (T6) through 13 (T13) of eluates with or without suppressive activity were assessed. Bands (90, 34 and 30 kDa) shown in fractions T8, T9 and T10 that correlated with suppressive activity were further analyzed by mass spectrometric analysis. Two peptides derived from the MW 34-kDa band were 100% identical to mouse GrB (GeneStream align program). These are peptide 157–170 (YSNTLQEVE LTVQK) and peptide 234–242 (VSSFLSWIK). We further confirmed the presence of GrB in the SN of clone MT-5B by using anti-GrB antibody in western blot and ELISA. Western blot analysis showed two bands with MW 32 and 34 kDa in both SN.C and LS that are identical to the mouse rGrB bands (Fig. 4E). The expressions of intracellular and extracellular GrB in LS and SN were also demonstrated in ELISA. Esterase activity of GrB in various preparations was also confirmed by enzymatic assays as previously described (38). Since both GrB and PFN are exported from the granules and work synergistically to kill target cells, we were interested in finding out if they were co-secreted by MT-5B cells. PFN was detected in both SN.C and LS as a 65-kDa band by western blot (Fig. 4E). As shown in Fig. 4(F), the amount of GrB in SN, SN.C and LS from different preparations was quantified by ELISA. The median levels of GrB were 230, 1682 and 865 ng ml–1, respectively.


Figure 4
View larger version (26K):
[in this window]
[in a new window]
  		Fig. 4 Identification of suppressive protein in MT-5B SN by proteomic analysis and western blot. (A) MT-5B SN.C was run on a 12% polyacrylamide analytical gel under non-reducing conditions, and three bands (B1, 2 and 3) and control (B4) were subjected to further analysis. (B) Proteins were eluted and precipitated by acetone extraction from indicated bands on a preparation gel loaded with 1.0 ml SN.C, and tested for suppressive activity on splenocyte response to Con A at concentration of 1:30. (C) Eluates (0.5 ml per tube) collected from Sephacryl S-100 columns loaded with 0.5 ml SN.C were tested for suppression of splenocyte response to Con A at 1:20 dilution. The short dashed lines shown in (B) and (C) are the levels of Con A responses without modification. Results are presented as mean CPM + SD of triplicates, and are representative of at least three experiments. (D) Eight tubes of eluates (T6-T13) with or without suppressive effect were run on a 12% polyacrylamide gel under reducing conditions. (E) The presences of GrB/PFN in LS and SN.C were further demonstrated by western blot analysis. Lanes 1 and 2 were blotted with anti-PFN antibody showing 65 kD bands; lanes 3, 4 and 5 were blotted with anti-GrB antibody showing bands at MW 32 and 34 kD for SN.C and LS, and at MW 30 kD as well for purified GrB. (F) SN, SN.C and LS from different preparations were analyzed for GrB by ELISA. The quantities of GrB are expressed as nanograms per milliliter and plotted in box plot from 7 to 13 different batches of samples.

 
Morphological and intracellular GrB/PFN analysis by light and confocal microscopy and flow cytometry
MT-5B cells were stained with Wright–Giemsa for morphological analysis (Fig. 5A). Clone MT-5B is characterized by its ‘blast like’ larger size, open and clumped nuclear chromatin pattern and abundant and polarized cytoplasmic granules. The cellular membranes appeared more distinct at the site of the polarized granules, which may indicate the exocytosis site of the granules. To determine the intracellular localization of GrB and PFN, MT-5B cells were stained as described in Methods and subjected to confocal microscopy. It was observed that both GrB and PFN are localized in the cells to well-defined sub-cellular vesicles. The overlay images showed coincident staining (yellow in color) specifying co-localization of the green signal from the FITC-labeled GrB and that of the red signal from the PE-labeled PFN antibodies (Fig. 5B). The intracellular co-expressions of GrB and PFN in MT-5B cells were also demonstrated by flow cytometric analysis as shown in Fig. 5(C).


Figure 5
View larger version (39K):
[in this window]
[in a new window]
  		Fig. 5 Wright–Giemsa staining, confocal and flow cytometric analysis of clone MT-5B. (A) MT-5B cells were stained with Wright–Giemsa on glass slide and examined by a light microscope (1000x). The larger size, abundant cytoplasm, polarized granules and open/clumped chromatin are the features of MT-5B. (B) Intracellular GrB (green), PFN (red) staining and co-localization staining (yellow) were shown by confocal microscopic analysis. (C) Intracellular expressions of GrB/PFN were also shown in histogram plots and dot plots by flow cytometric analysis.

 
Both GrB and PFN are needed for the suppression induced by MT-5B SN
The roles of GrB and PFN in MT-5B-induced immune suppression were further investigated by depletion assays. GrB or PFN from SN.C and LS was adsorbed by corresponding antibody-bound beads and eluted from these beads. Eluates were denatured and run on a 12% polyacrylamide gel (Fig. 6A). Compared with IgG isotype controls, the specific adsorptions of GrB and PFN from SN.C or LS were clearly shown in bands with MW 32, 34 and 65 kDa, respectively. The amount of GrB adsorbed on beads was higher in SN.C than in LS. The amount of GrB was largely reduced in SN.C after immunoadsorption with antibody-bound beads compared with that with control goat IgG-bound beads (Fig. 6B), which resulted in the decreased suppressive activity of SN.C correspondingly (Fig. 6C). Removal of both GrB and PFN by immunoadsorption did not have an additive effect (data not shown). Removal of either GrB and/or PFN by immunoadsorption decreased the suppressive activity of SN.C and LS, which indicates that MT-5B-induced suppression is mainly mediated by the secretion of GrB/PFN, and is PFN dependent. This is also indirectly supported by the data in Fig. 4(A) and (B). The maximal activity was found in band B2 (65 kD) which indicates the composition of both PFN and GrB dimmer under non-reducing condition. Band B3 (34 kD) which only contained monomeric GrB had no suppressive activity. We also found that purified recombinant mouse GrB alone had no suppressive effect (data not shown).


Figure 6
View larger version (28K):
[in this window]
[in a new window]
  		Fig. 6 Immunoadsorption of GrB or PFN decreases the suppressive effects of MT-5B SN and LS. (A) GrB and PFN from SN.C or LS were adsorbed on specific antibody-bound beads, and the immune complex was eluted from the beads in reducing sample buffer and applied to a 12% polyacrylamide gel. Coomassie blue and silver staining were used for detecting GrB and PFN, respectively. Specific antibody-adsorbed GrB and PFN on beads showed 34, 32 and 65 kD bands, respectively. Bands near 46-kD marker represent heavy chain of immunoadsorption antibody and control IgG. (B) GrB adsorbed on antibody-bound beads and free GrB in SN.C (equal to 10 µl of SN.C) were analyzed by western blot. (C) The suppressive activities of SN.C and LS after immunoadsorption by specific antibody-bound beads were also assayed in Con A response at SN.C concentration of 1:50 to 1:100. Goat IgG (gIgG) and rat IgG2a (rIgG2a) were used as isotype controls for immunoadsorption. Results are from three experiments, and presented as mean suppression (%) ± SEM. The suppressive effects of SN.C adsorbed by specific antibodies decreased significantly compared with corresponding isotypes controls, P = 0.0011 and 0.0102 for anti-GrB antibody-adsorbed SN.C and LS, respectively, and P = 0.0279 and 0.0096 for anti-PFN antibody-adsorbed SN.C and LS, respectively.

 
MT-5B cell-induced suppression is sensitive to EGTA pre-treatment
We have demonstrated that MT-5B cell-induced suppression is cell contact-independent and mediated by the secretion of GrB and PFN. EGTA has been shown to inhibit calcium-dependent exocytosis and release of granules and the action of PFN by blocking polymerization of, and pore formation by PFN (39, 40). To further unravel the relationship between GrB and PFN in MT-5B cell or MT-5B SN-induced suppression, MT-5B cells were pre-incubated in the presence of EGTA at different concentrations for different durations. After incubation, MT-5B cells were washed and incubated in medium for another 8 h. MT-5B cells or SN alone and MT-5B cells plus SN were analyzed separately for suppressive activity. As shown in Fig. 7, the suppressive activities of MT-5B cells and SN were significantly blocked by pre-incubation of MT-5B cells with EGTA in a dose-dependent manner after all three durations of incubation. Thus, the suppressive effects induced by MT-5B cells are sensitive to EGTA pre-treatment and cumulative due to the consistent secretion of GrB and PFN in the cultures.


Figure 7
View larger version (13K):
[in this window]
[in a new window]
  		Fig. 7 EGTA pre-incubation decreases the suppressive effect of clone MT-5B. MT-5B cells were pre-incubated with 0, 2.5, 5.0 or 10.0 mM EGTA for 4, 8 or 16 h. After washing, MT-5B cells were seeded in 96-well plates (2 x 104 per well) and incubated for another 8 h in 0.1 ml complete medium. MT-5B cells or SN (1:3) alone and MT-5B cells plus SN from the cultures were then tested separately in a 3-day assay using 24-h splenocyte Con A blasts as responder at 1:20 ratio of regulator: responder. Results are plotted in 4, 8 and 16 h incubation groups and expressed as mean suppression (%) + SEM from three experiments. Statistic analysis, P < 0.05 for medium versus EDTA at all concentrations and cultures except for 2.5 versus 5 mM EGTA in 4-h incubation.

 
GrB-dependent down-regulation of diabetogenic effector T cells by MT-5B cells through inducing apoptosis
The suppressive effect of MT-5B cells was also verified in adoptive transfer of diabetes assay. MT-5B cells significantly blocked the disease transfer when co-transferred with splenocytes from acutely diabetic NOD mice to recipient NOD mice (Fig. 8A). The GrB/PFN pathway is one of the mechanisms underlying the induction of apoptosis (32). Given the presence of both GrB and PFN in SN from MT-5B cells, SN-induced suppression of immune responses was also investigated by pre-incubating diabetogenic splenocytes with MT-5B SN (1:20) or GrB antibody-adsorbed SN.C (1:100) for 36 h before cell transfer. As shown in Fig. 8(A), incubation with MT-5B SN significantly impairs the ability of diabetogenic splenocytes to transfer diabetes. Removal of GrB from SN.C by immunoadsorption damages the ability of SN.C to down-regulate diabetogenic T cells. To assess the mechanism underlying the impairment of diabetogenic T cells, we looked for apoptotic cells after incubation with MT-5B SN using flow cytometry (Fig. 8B). Significantly higher proportions of PI-positive splenocytes and TUNEL-positive CD4+ and CD8+ T cells were found in cultures incubated with SN from MT-5B than that from NOD T cells, P = 0.003 for PI-positive cells and P = 0.045 and 0.011 for TUNEL-positive CD4+/CD8+ T cells, respectively. Results from both in vitro and in vivo studies confirm that clone MT-5B established from CFA-protected NOD mice down-regulates diabetogenic T cells by inducing apoptosis in a GrB/PFN-dependent fashion.


Figure 8
View larger version (42K):
[in this window]
[in a new window]
  		Fig. 8 MT-5B cells block adoptive transfer of diabetes by GrB-dependent induction of apoptosis. (A) For adoptive transfer of diabetes, diabetogenic splenocytes (DiaSCs) prepared from acutely diabetic NOD mice were transferred alone or together with MT-5B cells or control NOD T cells into pre-irradiated NOD mice. DiaSCs were also pre-incubated with SN from MT-5B cells (1:20) or control NOD T cells, and anti-GrB antibody or isotype-absorbed MT-5B SN.C (1:100) for 36 h before transferring to NOD.SCID mice. Results are presented as Kaplan–Meier life survival curve. The proportion of mice with normal glycemia was increased in MT-5B cell co-transferred NOD mice (P = 0.0079 for versus NOD T cell control and DiaSCs alone) and in MT-5B SN or isotype IgG-adsorbed SN.C groups (P = 0.0010 for versus control SN from NOD T cells, and P = 0.0227 for versus GrB antibody-adsorbed SN.C group). The number of mice in each group was shown in parenthesis and was from two to five experiments. (B) For apoptosis analysis, DiaSCs were stained with PI or TUNEL–CD4/CD8 after in vitro incubation with SN from MT-5B or control NOD T cells, and analyzed by flow cytometry. Histogram plots and dot plots are representative of PI and TUNEL-CD4/CD8 staining from three experiments. Mean (%) ± SEM of apoptotic cells are shown in each plot. Increased proportions of apoptotic cells were found in MT-5B SN-incubated DiaSCs. Compared with control SN, P = 0.02 for PI-positive splenocytes, and P = 0.045 and 0.011 for TUNEL-CD4 and -CD8-positive T cells, respectively.

 
CFA immunization up-regulates the expression of GrB in CD4 T cells
Up-regulation of GrB expression and secretion have been shown in T cells after viral or bacterial infection (41, 42). Since clone MT-5B was established from NOD mice immunized with CFA, a Mycobacterium tuberculosis-containing adjuvant, the change of GrB expression in CD4 T cells from freshly prepared or BCG-incubated draining lymph node cells was assessed 12 and 35 days after CFA immunization. As shown in Fig. 9, GrB expression in draining lymph node CD4 T cells was increased by in vivo injection of CFA or/and in vitro stimulation of draining lymph node cells with SN of BCG suspension. The significant increases of these cells were found in all groups of cells from CFA-treated mice except for the fresh cells of 35-day group, and in BCG-incubated cells from all groups of mice except for the incomplete Freund's adjuvant-treated mice of 12-day group. It seems that immunization with Mycobacterium preparation increases the expression of GrB in CD4 T cells both in vivo and in vitro.


Figure 9
View larger version (45K):
[in this window]
[in a new window]
  		Fig. 9 Up-regulation of GrB expression in CD4 T cells of CFA-immunized NOD mice. The change of GrB expression in draining lymph node CD4 T cells was analyzed by intracellular staining 12 or 35 days after CFA immunization. Draining lymph node cells were double stained with anti-CD4 and -GrB antibodies before (fresh) or after (BCG incubation) incubation with SN of BCG suspension (400 µg ml–1) at a concentration of 1:4 for 24 h. (A) Representative dot plots show the expression of GrB in CD4 T cells of incomplete Freund's adjuvant (IFA) or CFA-immunized NOD mice, and cells incubated with BCG. The number inserted in the upper right quadrant indicates the percentage of GrB+ CD4 T cells. (B) The increased GrB+ CD4 T cells were detected in freshly prepared and 24-h BCG-incubated lymph node cells. Results are presented as mean (%) + SEM of six to seven mice from three to four experiments. P < 0.05 to 0.0001 for CFA versus IFA in all groups of cells but the fresh cells of 35-day group; P < 0.05 to 0.02 for BCG-incubated versus fresh cells in all groups of mice but the IFA-treated mice of 12-day group.

 

   Discussion
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
References
 
In this study, we report the morphology, function and mechanism of action of a unique CD4+CD8+ CD25+ Treg cell clone designated as MT-5B. This clone was derived from the lymph node cells of NOD mice immunized with CFA. Immunization of young NOD mice with CFA protects these mice from T1D (37). MT-5B is characterized by a blast-like morphology with a distinct phenotype. Our results show that MT-5B is capable of suppressing a variety of immune responses by secreting GrB/PFN in a cell contact-independent manner. This clone shares the characteristics of both adaptive and natural CD4+CD25+ Treg cells. It expresses high levels of GrB/PFN but low level of CD3. MT-5B cells showed anergic proliferative response to a panel of immune stimulations. The effector molecule involved in the regulatory function of CD4+CD25+ Treg cells remains unclear, but based on recent studies in human cells and our present study in mouse cells, GrB/PFN pathway appears to be at least one of the suppressive mechanisms underlying the function of CD4+CD25+ Treg cells (29).

It has previously been shown that NK cells, CTLs and lymphokine-activated cells exert their cytotoxic effects by inducing target cell apoptosis through the GrB/PFN pathway (29, 30, 43). The expression of PFN, GrB or granulysin was also found in CD4+ T cells derived from subjects with chronic viral infections, tumors and autoimmune diseases (44–47) or in vitro-activated Tr1 cells (48). GrB and/or PFN-expressing DP T cells were found in human PBL stimulated with human cytomegalo virus (HCMV) and HIV-1 antigens and in peripheral blood lymphocytes (PBL) of normal cynomolgus monkeys (49, 50). DP T cells have also been found in PBL of rodents, swine, monkeys, chickens and humans both in healthy and diseased states (51, 52). CD8{alpha} is an activation marker for the subset of peripheral CD4+ T cells (53). MHC class II-deficient mice develop a CD25+ DP T cell population which controls colitogenic CD4+CD25– T cells (54). Their involvement in the adaptive immune responses against infectious pathogens has increased our understanding of the role of T cells in the antiviral immune response (55). Small but increased proportion of DP T cells have been found in peripheral lymphoid tissue of CFA-immunized NOD mice (data not shown). It is uncertain whether the DP phenotype of clone MT-5B is related to the anti-mycobacterium immunity induced after CFA immunization, but it does prevent the adoptive transfer of T1D to NOD recipients.

Cell contact-dependent killing is a unique feature of natural CD4+CD25+ Treg cells and GrB/PFN-expressing CD4+CD25+ Treg cells (3, 29). In this study, we have clearly demonstrated that clone MT-5B suppresses immune responses in a cell contact-independent manner. The suppression is not primarily mediated by the secretion of suppressive cytokines, as shown by the failure of neutralizing antibodies to block the suppression, but by the secretion of GrB/PFN. The detectable suppressive effects in both SN and LS indicate that GrB/PFN produced by clone MT-5B is stored as well as secreted via granules.

The antigen specificity of CD4+CD25+ Treg cell-mediated suppression is a matter of debate. It has been shown that the activation of CD4+CD25+ Treg cells can be driven by antigen stimulation or via their TCR cross-linking, but the CD4+CD25+ T effector cells are probably not antigen specific (7, 5). It remains unclear why the mechanism of action of Treg cells is cell contact-dependent but their effector function (suppression) is antigen non-specific. The mode of action of clone MT-5B is quite different from other CD4+CD25+ Treg cells in terms of cell–cell contact. Clone MT-5B is probably an activated and fully mature memory T cell population with effector function, and is programmed to produce and secrete GrB/PFN which induces non-specific suppression without cell–cell contact. The other CD4+CD25+ Treg cells or GrB/PFN-expressing CD4+CD25+ Treg cells require target cell contact, and SN from these cells fail to induce cytotoxicity against target cells (29). We propose that this difference may be due to the maturation state and/or activation state of various Treg cell subsets, otherwise clone MT-5B may represent a new subset of Treg cells.

It has been shown that the activation state of target cells appears to determine their susceptibility to the cytotoxic killing. NK cells kill activated but not resting T cells (56). PFN-expressing T cells kill autoreactive T cells (57). Autologous activated CD4+ and CD8+ T cells are preferentially killed over unactivated T cells (29). CD4+CD25+ Treg cells may control the T cell activation state (58). Based on our observation, GrB/PFN-secreting clone MT-5B exerts similar effect on activated T cells. It seems that the degree of suppression is dependent on the activation state of target cells as shown in Con A response and MLR assays. The mechanism underlying the differential sensitivity of activated and resting T cells to GrB/PFN-mediated suppression has likely evolved to avoid bystander suppressions. In transplantation, CD4+CD25+ Treg cells prevent graft rejection, but this does not compromise anti-virus immunity (59). Therefore, there is no global immunosuppression by Treg cells.

Our perspective is that aberrantly activated cells may cause intrinsic changes such as decrease of serpin protease inhibitor (SPI-6) which makes them sensitive to GrB/PFN-mediated killing and maintains immune homeostasis. Chronic infections, especially intracellular infections, have been shown to increase the proportion of GrA/PFN-expressing T cells which eliminate the pathogen-infected cells (44, 55). The important roles of GrB/PFN in autoimmune disease have been demonstrated in autoimmune encephalomyelitis by using PFN-knockout mice (60). Previously, we have shown that BCG down-regulates the activity of diabetogenic T cells by inducing apoptosis through both Fas/FasL and TNF-{alpha} pathways (61). However, clone MT-5B may represent another kind of Treg cells under a certain activation stage, and exert its effector function mainly by using GrB/PFN pathway.

GrB-induced cytotoxicity against target cells has been shown to be PFN-dependent (29, 43) but the precise role and acting mode of PFN in GrB cell-surface binding, internalization, intracellular trafficking and targeting needs to be further elucidated. To verify the requirement of PFN in GrB secretion and GrB-induced suppression, two experiments were performed in this study. One is removal of GrB or PFN from SN by immunoadsorption. The suppressive activity is significantly decreased but not completely removed after immunoadsorption. This may result from incomplete removal of GrB or PFN, but the involvement of other mechanisms can not be excluded. Another is the pre-incubation of MT-5B cells with EGTA, a Ca2+ chelator. The decreased suppressive effect of EGTA-pre-incubated MT-5B cells or its SN was significant. It suggests that GrB/PFN is constitutively secreted and the suppressive activity is accumulated in SN. It also indicates that decreased suppression after EGTA pre-incubation mainly results from the changes in secretion and function of PFN, and that cell contact-independent secretion of GrB by clone MT-5B is dependent on PFN for its function. The requirement of both GrB and PFN for triggering in vitro apoptosis has been confirmed by other studies (32, 62), but the opposite conclusion was also reported in PFN-knockout mice (30).

In this study, the establishment of a cell contact-independent DP CD25+ Treg clone raises an interesting issue regarding the mechanism of action of CD4+CD25+ Treg cells and their application for immune interventions. We propose that the cell contact-independent suppression by soluble GrB/PFN is a major mechanism for the highly activated and mature memory CD4+CD25+ Treg cells. GrB has been shown to be exocytosed as a complex with proteoglycans and PFN plays a key role in its effector mechanism (35). The restricted targeting of activated cell populations makes these Treg cells superior in suppressing disease pathology and maintaining immune homeostasis by anti-proliferation and pro-apoptosis while sparing normal resting cells. In NOD mice, diabetogenic autoreactive T cells become activated, matured and attack self-islet ß cells with age due in part to an intrinsic deficiency of regulatory cells. A decrease in TGF-ß1 and Foxp3-co-expressing Treg cells in pancreatic lymph node cells has recently been reported to be associated with the progress of diabetes in NOD mice (63). We have previously reported that CFA prevents the onset of diabetes in NOD mice by inducing cyclophosphamide-sensitive Treg cells (37). It has been reported that administration of cyclophosphamide sensitizes established tumors to immunotherapy by depletion of CD4+CD25+ Treg cells (64).

We also found that GrB+ CD4 T cells were up-regulated in CFA-immunized NOD mice. We speculate that CFA-induced Treg cells are probably the GrB/PFN-secreting Treg cells from which MT-5B was selected and cloned. The ability of diabetogenic splenocytes to transfer disease was down-regulated by pre-incubation with SN from MT-5B cells through the induction of apoptosis, which was GrB dependent. Removal of GrB from SN significantly decreased the ability of SN to down-regulate diabetogenic splenocytes. It seems that mycobacterium-induced protection against diabetes in NOD mice as shown previously may also use GrB/PFN pathway as one of the mechanisms to eliminate GrB/PFN-sensitive diabetogenic T cells. This further provides evidence for the mechanism underlying CFA-induced protection against T1D. Collectively, GrB/PFN secretion represents a novel mechanism underlying the control of immune responses and maintenance of immune homeostasis by CD4+CD25+ Treg cells in a cell contact-independent manner.


   Acknowledgements
 
We thank Rachel De Lima for critically reading the manuscript and Tetyana Pelipyagina for animal care. This work is supported by grants from the Canadian Institutes of Health Research.


   Abbreviations
 
APC, antigen-presenting cell
BCG, bacilli Calmette–Guérin
Con A, Concanavalin
CPM, counts per minute
DP, double positive
EGTA, ethyleneglycol-bis(aminoethylether)-tetraacetic acid
GAD67, glutamic acid decarboxylase
GITR, glucocorticoid-induced tumor necrosis factor receptor family-related gene
GrB, granzyme B
[3H]TdR, [3H]thymidine
LS, cell lysates
MLR, mixed lymphocyte reaction
MW, molecular weight
NOD, non-obese diabetic
PFN, perforin
PI, propidium iodide
RT, room temperature
SN, supernatant
SN.C, concentrated supernatant
TBS, Tris-buffered saline
TGF, transforming growth factor
TNF, tumor necrosis factor
Treg, regulatory T
T1D, type 1 diabetes
TUNEL, TdT-mediated dUTP nick end labeling

   Notes
 
Transmitting editor: C. J. Paige

Received 12 September 2005, accepted 3 April 2006.


   References
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 

  1. Sakaguchi S. (2004) Naturally arising CD4+ regulatory T cells for immunologic self tolerance and negative control of immune responses. Annu. Rev. Immunol. 22:531.[CrossRef][ISI][Medline]
  2. Suri-Payer E, Amar AZ, Thornton AM, Shevach EM. (1998) CD4+CD25+ T cells inhibit both the induction and effector function of autoreactive T cells and represent a unique lineage of immunoregulatory cells. J. Immunol. 160:1212.[Abstract/Free Full Text]
  3. Bluestone JA and Abbas AK. (2003) Natural versus adaptive regulatory T cells. Nat. Rev. Immunol. 3:253.[CrossRef][ISI][Medline]
  4. Hori S, Takahashi T, Sakaguchi S. (2003) Control of autoimmunity by naturally arising regulatory CD4+ T cells. Adv. Immunol. 81:331.[Medline]
  5. Mukherjee R, Chaturvadi P, Qin H-Y, Singh B. (2003) CD4+CD25+ regulatory T cells generated in response to insulin B: 9–23 peptide prevent adoptive transfer of diabetes by diabetogenic T cells. J. Autoimmun. 21:221.[CrossRef][ISI][Medline]
  6. Sakaguchi S, Sakaguchi N, Asano M, Itoh M, Toda M. (1995) Immunologic self-tolerance maintained by activated T cells expressing IL-2 receptor alpha-chains (CD25). Breakdown of a single mechanism of self-tolerance causes various autoimmune diseases. J. Immunol. 150:1151.
  7. Thornton AM and Shevach EM. (2000) Suppressor effector function of CD4+CD25+ immunoregulatory T cells is antigen nonspecific. J. Immunol. 164:183.[Abstract/Free Full Text]
  8. Groux H, O'Garra A, Bigle M, Rouleau M, Antonenko S, de Vries JE. (1997) A CD4+ T-cell subset inhibits antigen-specific T-cell responses and prevents colitis. Nature 389:737.[CrossRef][Medline]
  9. Levings MK and Roncarolo MG. (2000) T-regulatory 1 cells: a novel subset of CD4 T cells with immunoregulatory properties. J. Allergy Clin. Immunol. 106:109.
  10. Weiner HL. (2001) Oral tolerance: immune mechanisms and the generation of Th3-type TGF-beta-secreting regulatory cells. Microbes Infect. 3:947.[CrossRef][ISI][Medline]
  11. Sakaguchi S, Sakaguchi N, Shimizu J, Yamazaki S, Sakihama T, Itoh M. (2001) Immunologic tolerance maintained by CD25+ CD4+ regulatory T cells: their common role in controlling autoimmunity, tumor immunity, and transplantation tolerance. Immunol. Rev. 182:18.[CrossRef][ISI][Medline]
  12. Horwitz DA, Zheng SG, Gray JD, Wang JH, Ohtsuka K, Yamagiwa S. (2004) Regulatory T cells generated ex vivo as an approach for the therapy of autoimmune disease. Semin. Immunol. 16:135.[CrossRef][ISI][Medline]
  13. Chaturvedi P, Qin H-Y, Chou H, Singh B. (1997) Modulation of insulin-dependent diabetes mellitus (IDDM) in NOD mice by autoreactive T cells. Crit. Rev. Immunol. 17:519.[Medline]
  14. Han HS, Jun HS, Utsugi T, Yoon JW. (1996) A new type of CD4+ suppressor T cell completely prevents spontaneous autoimmune diabetes and recurrent diabetes in syngeneic islet-transplanted NOD mice. J. Autoimmun. 9:331.[CrossRef][Medline]
  15. Akhtar I, Gold JP, Pan L-Y, et al. (1995) CD4+ ß islet cell-reactive T cell clones that suppress autoimmune diabetes in nonobese diabetic mice. J. Exp. Med. 182:87.[Abstract/Free Full Text]
  16. Judes AE and von Herrath MG. (2004) Regulatory T-cells in type 1 diabetes. Diabetes Metab. Res. Rev. 20:446.[Medline]
  17. Nakamura K, Kitani A, Strober W. (2001) Cell contact-dependent immunosuppression by CD4(+)CD25(+) regulatory T cells is mediated by cell surface-bound transforming growth factor beta. J. Exp. Med. 194:629.[Abstract/Free Full Text]
  18. Green EA, Gorelik L, McGregor CM, Tran EH, Flavell RA. (2003) CD4+CD25+ T regulatory cells control anti-islet CD8+ T cells through TGF-beta-TGF-beta receptor interactions in type 1 diabetes. Proc. Natl Acad. Sci. USA 100:10878.[Abstract/Free Full Text]
  19. Read S, Malmstrom V, Powrie F. (2000) Cytotoxic T lymphocyte-associated antigen 4 plays an essential role in the function of CD25(+)CD4(+) regulatory cells that control intestinal inflammation. J. Exp. Med. 192:295.[Abstract/Free Full Text]
  20. McHugh RS, Whitters MJ, Piccirillo CA, Young DA, Shevach EM, Collins M. (2002) CD4(+)CD25(+) immunoregulatory T cells: gene expression analysis reveals a functional role for the glucocorticoid-induced TNF receptor. Immunity 16:311.[CrossRef][ISI][Medline]
  21. Piccirillo CA, Letterio JJ, Thornton AM, et al. (2002) CD4(+)CD25(+) regulatory T cells can mediate suppressor function in the absence of transforming growth factor beta1 production and responsiveness. J. Exp. Med. 196:237.[Abstract/Free Full Text]
  22. Levings MK, Sangregorio R, Sartirana C, et al. (2002) Human CD25+CD4+ T suppressor cell clones produce transforming growth factor beta, but not interleukin 10, and are distinct from type 1 T regulatory cells. J. Exp. Med. 196:1335.[Abstract/Free Full Text]
  23. Jonuleit H and Schmitt E. (2003) The regulatory T cell family: distinct subset and their interrelations. J. Immunol. 171:6323.[Free Full Text]
  24. Hori S, Nomura T, Sakaguchi S. (2003) Control of regulatory T cell development by the transcription factor Foxp3. Science 299:1057.[Abstract/Free Full Text]
  25. Fontenot JD, Gavin MA, Rudensky AY. (2003) Foxp3 programs the development and function of CD4+CD25+ regulatory T cells. Nat. Immunol. 4:330.[CrossRef][ISI][Medline]
  26. Lieberman J. (2003) The ABCs of granule-mediated cytotoxicity: new weapons in the arsenal. Nat. Rev. Immunol. 3:361.[CrossRef][ISI][Medline]
  27. Lord SJ, Rajotte RV, Korbutt GS, Bleackley RC. (2003) Granzyme B: a natural born killer. Immunol. Rev. 193:31.[CrossRef][ISI][Medline]
  28. Russell JH and Ley TJ. (2002) Lymphocyte-mediated cytotoxicity. Annu. Rev. Immunol. 20:323.[CrossRef][ISI][Medline]
  29. Grossman WJ, Verbsky JW, Barchet W, Colonna M, Atkinson JP, Ley TJ. (2004) Human T regulatory cells can use the perforin pathway to cause autologous target cell death. Immunity 21:589.[CrossRef][ISI][Medline]
  30. Gondek DC, Lu LF, Quezada SA, Sakaguchi S, Noelle RJ. (2005) Cutting edge: contact-mediated suppression by CD4+CD25+ regulatory cells involves a granzyme B-dependent, perforin-independent mechanism. J. Immunol. 174:1783.[Abstract/Free Full Text]
  31. Yagita H, Nakata M, Kawasaki A, Shinkai Y, Okumura K. (1992) Role of perforin in lymphocyte-mediated cytolysis. Adv. Immunol. 51:215.[Medline]
  32. Shi L, Keefe D, Durand E, Feng H, Zhang D, Lieberman J. (2005) Granzyme B binding to target cells mostly by charge and must be added at the same time as perforin to trigger apoptosis. J. Immunol. 174:5456.[Abstract/Free Full Text]
  33. Motyka B, Korbutt G, Pinkoski MJ, et al. (2000) Mannose 6-phosphate/insulin-like growth factor II receptor is a death receptor for granzyme B during cytotoxic T cell-induced apoptosis. Cell 103:491.[CrossRef][ISI][Medline]
  34. Metkar SS, Wang B, Aguilar-Santelises M, et al. (2002) Cytotoxic cell granule-mediated apoptosis: perforin delivers granzyme B-serglycin complexes into target cells without plasma membrane pore formation. Immunity 16:417.[CrossRef][Medline]
  35. Raja SM, Metkar SS, Froelich CJ. (2003) Cytotoxic granule-mediated apoptosis: unraveling the complex mechanism. Curr. Opin. Immunol. 15:528.[CrossRef][Medline]
  36. Sadelain MW, Qin H-Y, Lauzon J, Singh B. (1990) Prevention of type 1 diabetes in NOD mice by adjuvant immunotherapy. Diabetes 39:583.[Abstract]
  37. Qin HY, Sadelain MW, Hitchon C, Lauzon J, Singh B. (1993) Complete Freund's adjuvant-induced T cells prevent the development and adoptive transfer of diabetes in nonobese diabetic mice. J. Immunol. 150:2072.[Abstract]
  38. Ewen C, Kane KP, Shostak I, et al. (2003) A novel cytotoxicity assay to evaluate antigen-specific CTL responses using a colorimetric substrate for granzyme B. J. Immunol. Methods 276:89.[CrossRef][Medline]
  39. Takayama H and Sitkovsky MV. (1987) Antigen receptor-regulation in cytotoxic T lymphocytes. J. Exp. Med. 166:725.[Abstract/Free Full Text]
  40. Baetz K, Isaaz S, Griffiths M. (1995) Loss of cytotoxic T lymphocyte function in Chediak-Higashi syndrome arises from a secretary defect that prevents lytic granule exocytosis. J. Immunol. 154:6122.[Abstract]
  41. van Leeuwen EM, Remmerswal EB, Vossen MT, et al. (2004) Emergence of a CD4+CD28– granzyme B+, cytomegalovirus-specific T cell subset after recovery of primary cytomegalovirus infection. J. Immunol. 173:1834.[Abstract/Free Full Text]
  42. Lauw FN, Simpson AJ, Hack CE, et al. (2000) Soluble granzymes are released during human endotoxemia in patients with sever infection due to gram-negative bacteria. J. Infect. Dis. 182:206.[CrossRef][ISI][Medline]
  43. Murray PD, McGavern DB, Lin X, et al. (1998) Perforin-dependent neurologic injury in a viral model of multiple sclerosis. J. Neurosci. 1:7306.
  44. Appay V, Zaunders JJ, Papagno L, et al. (2002) Characterization of CD4(+) CTLs ex vivo. J. Immunol. 168:5954.[Abstract/Free Full Text]
  45. Norris PJ, Sumaroka M, Brander C, et al. (2001) Multiple effector functions mediated by human immunodeficiency virus-specific CD4(+) T-cell clones. J. Virol. 75:9771.[Abstract/Free Full Text]
  46. Sun Q, Burton RL, Lucas KG. (2002) Cytokine production and cytolytic mechanism of CD4(+) cytotoxic T lymphocytes in ex vivo expanded therapeutic Epstein-Barr virus-specific T-cell cultures. Blood 99:3302.[Abstract/Free Full Text]
  47. Yawalkar N, Schmid S, Braathen LR, Pichler WJ. (2001) Perforin and granzyme B may contribute to skin inflammation in atopic dermatitis and psoriasis. Br. J. Dermatol. 144:1133.[CrossRef][Medline]
  48. Grossman WJ, Verbsky JW, Tollefsen BL, Kemper C, Atkinson JP, Ley TJ. (2004) Differential expression of granzymes A and B in human cytotoxic lymphocyte subsets and T regulatory cells. Blood 104:906.[Free Full Text]
  49. Suni MA, Ghanekar SA, Houck DW, et al. (2001) CD4(+)CD8(dim) T lymphocytes exhibit enhanced cytokine expression, proliferation and cytotoxic activity in response to HCMV and HIV-1 antigens. Eur. J. Immunol. 31:2512.[CrossRef][Medline]
  50. Nam K, Akari H, Terao K, Shibata H, Kawamura S, Yoshikawa Y. (2000) Peripheral blood extrathymic CD4(+)CD8(+) T cells with high cytotoxic activity are from the same lineage as CD4(+)CD8(