These authors contributed equally to this work.
Autoimmune, Cholestatic and Biliary Disease
Expansion and de novo generation of potentially therapeutic regulatory T cells in patients with autoimmune hepatitis†
Article first published online: 7 JAN 2008
Copyright © 2008 American Association for the Study of Liver Diseases
Volume 47, Issue 2, pages 581–591, February 2008
How to Cite
Longhi, M. S., Meda, F., Wang, P., Samyn, M., Mieli-Vergani, G., Vergani, D. and Ma, Y. (2008), Expansion and de novo generation of potentially therapeutic regulatory T cells in patients with autoimmune hepatitis. Hepatology, 47: 581–591. doi: 10.1002/hep.22071
Potential conflict of interest: Nothing to report.
- Issue published online: 26 JAN 2008
- Article first published online: 7 JAN 2008
- Accepted manuscript online: 7 JAN 2008 12:00AM EST
- Manuscript Accepted: 28 SEP 2007
- Manuscript Received: 22 AUG 2007
- Alex Mowat PhD studentship from King's Medical Research Trust, UK
- Sparks for Kids, UK
- Children's Liver Disease Foundation, Birmingham, UK
CD4+CD25+ regulatory T cells (T-regs) are central to the maintenance of immune tolerance and represent an immune intervention candidate in autoimmune hepatitis (AIH), a condition characterized by impaired T-reg number and function. We investigated whether T-regs can be expanded from the existing CD4+CD25+ T cell pool and generated de novo from CD4+CD25− T cells in AIH patients and healthy controls. Purified CD4+CD25+ and CD4+CD25− T cells from 24 patients with type 1 AIH and 22 healthy controls were cultured for up to 5 weeks with anti-CD3/anti-CD28 T cell expander and high-dose interleukin-2 (IL-2). Cell phenotypes, suppressor ability, forkhead winged/helix transcription factor box P3 (FOXP3) gene, and protein expression were assessed weekly by cytofluorimetry, proliferation assay, real-time polymerase chain reaction (PCR), and immunoblot. During culture, the number of CD4+CD25+ T cells derived from the existing T-reg pool (expanded T-regs) and generated de novo from CD4+CD25− T cells (newly generated T-regs) increased constantly up to week 4 in both healthy controls and, to a lesser extent, in AIH patients. Expanded T-regs retained conventional T-reg phenotype and, compared with baseline, demonstrated more vigorous suppressive function and increased FOXP3 gene and protein expression. Newly generated T-regs not only acquired T-reg phenotype but underwent greater growth and were more resistant to apoptosis than expanded T-regs. Their suppressive function augmented throughout culture, reaching a peak at week 4, preceded by a peak FOXP3 gene and protein expression at week 2. Suppressor function and FOXP3 expression of both expanded and newly generated T-regs were higher in normal controls than in AIH patients. Conclusion: Functionally enhanced T-regs can be expanded and generated de novo in patients with AIH. This finding may assist in reconstituting impaired immune regulation and restoring peripheral tolerance through T-reg infusion in this condition. (HEPATOLOGY 2008;47:581–591.)
The mechanisms underlying the breakdown of self-tolerance leading to autoimmune disease have not been fully elucidated, though impaired immune regulation is likely to be involved. Among the several cell populations with known regulatory activities,1 CD4 T cells constitutively expressing the interleukin 2 (IL-2) receptor α chain (CD25) have emerged as key players in immunotolerance maintenance.2, 3 A number of surface markers characterize these regulatory T cells (T-regs), including the cytotoxic T lymphocyte–associated-antigen 4, the glucocorticoid inducible tumor necrosis factor receptor, and the forkhead winged/helix transcription factor box P3 (FOXP3). The latter is considered the most reliable marker of T-regs, because it is closely associated with suppressor function.4–6
Numerical and functional T-reg impairment has been documented in experimental murine models of autoimmune disorders and in human autoimmune conditions, where these cells fail to control autoreactive CD4 and CD8 T cell effector function.7–11 Because of their ability to control autoaggression, T-regs represent an attractive candidate of immune intervention aimed at restoring immunotolerance in autoimmune disorders. Their potential therapeutic use is limited by their small number in the circulation and by their low ability to proliferate, although the latter can be overcome both in mouse and humans by exposing CD4+CD25+ T cells to high-dose IL-2 and to a combined solid-phase CD3/CD28 stimulus.8, 12, 13 Studies in healthy subjects have shown that expanded T-regs maintain their phenotypic profile and their ability to suppress in vitro proliferation and cytokine production.12, 13
The CD4+CD25− T cell population also comprises a small proportion of FOXP3-positive lymphocytes capable, after exposure to saturating concentrations of IL-2, of acquiring the CD25 marker and regulatory properties.14
Patients with autoimmune hepatitis (AIH), an inflammatory liver disorder characterized by hypergammaglobulinemia, circulating autoantibodies, and histological evidence of interface hepatitis,15, 16 have numerically and functionally defective T-regs.17–19 This defect relates to the stage of liver disease, being more evident at disease presentation than during treatment induced remission, where a partial restoration is observed.17, 18
The aims of the current study were, first, to investigate whether in patients with AIH CD4+CD25+ T-regs can be expanded from the existing CD4+CD25+ T cell pool and generated de novo from the CD4+CD25− T cell pool; and, second, to characterize phenotypically and functionally both expanded and newly generated T-regs.
Patients and Methods
Patients and Controls
Twenty-four patients with antinuclear antibody or smooth muscle antibody positive AIH15 were studied (12 female; median age at the time of study, 14.7 years; range, 8–26 years). Three of them had bile duct changes characteristic of sclerosing cholangitis on retrograde cholangiography and were diagnosed as having overlap syndrome.20 All were studied during remission on immunosuppressive treatment, because we have previously demonstrated that T-regs collected at diagnosis are numerically impaired.17, 18 Median aspartate aminotransferase, bilirubin, and immunoglobulin G levels were 29 IU/L (range, 19–55; normal value < 50 IU/L], 9 μmol/L (range, 4–22; normal value < 20 μmol/L), and 12.6 g/L (range, 7.1–23.1; normal value, 6.5–17 g/L). Median antinuclear antibody and smooth muscle antibody titers were 1/40 (range, 1/20–1/160) and 1/20 (range 1/20–1/640). All patients were receiving prednisolone (2.5–5 mg/daily), in association with azathioprine (1–2 mg/kg/day) in 17. Twenty-two healthy subjects (median age, 27 years; range, 22–35; 12 female) served as normal controls, the disparity between patient and control age deriving from ethical constraints in obtaining blood from healthy children. Written consent was obtained for each subject. The study was approved by the Ethics Committee of King's College Hospital, London, UK.
Cell Separation and Purification
Peripheral blood mononuclear cells were prepared from 10 to 20 mL peripheral blood mixed with preservative-free heparin (10 U/mL), diluted 1/1 with Roswell Park Memorial Institute 1640 (Invitrogen Life Technologies, Paisley, UK), and separated using Ficoll Hypaque (Amersham Pharmacia Biotech Ltd., Little Chalfont, UK). Mononuclear cell viability, determined by Trypan blue exclusion, always exceeded 98%.
CD4+CD25+ T cells were isolated from peripheral blood mononuclear cells by CD4 negative selection, using a cocktail of antibodies to CD14, CD56, CD19, CD8, CD235a, and CD45RA and depletion beads (Dynal Biotech, Oslo, Norway) coated with an Fc-specific human immunoglobulin G4 antibody, followed by CD25-positive selection using immunomagnetic beads coated with anti-human CD25 antibodies (Dynal Biotech).19 Purified CD4+CD25+ T cells localized in the CD4+CD25high cell gated area by cell sorting (model BD Vantage SE/DiVa, Immunocytochemistry Systems, San José, CA) and exerted comparable inhibitory activity.19 CD4+CD25− T cells were obtained from the negative fraction of the 2-step selection described. As previously described,19 the purity exceeded 95% for CD4+CD25+ T-regs and 90% (range, 92%–97%) for CD4+CD25− T cells.
T-reg and CD4+CD25− T Cell Expansion and Phenotype Characterization
Freshly isolated T-regs and CD4+CD25− T cells were expanded according to Hoffmann et al.,12 with some modifications: 5 × 105 cells/well were cultured for up to 5 weeks in Roswell Park Memorial Institute 1640, supplemented with 2 mM L-glutamine, 25 mM [4-(2-hydroxyethyl)-1-piperazine ethanesulfonic acid], 100 U/L penicillin, 0.1 mg/mL streptomycin, 2.5 μg/mL amphotericin B, and 10% inactivated fetal bovine serum (FBS) at 37°C and 5% CO2, in the presence of anti-CD3/anti-CD28 T cell expander at a ratio of 4 beads/cell and recombinant IL-2 (rIL-2, Eurocetus Amsterdam, Netherlands), added at 300 U/mL for the first week and later at 100 U/mL. T cell viability and expansion were assessed weekly by Trypan blue exclusion. When sufficient numbers of peripheral blood mononuclear cells were available, T-regs generated from CD4+CD25− T cells after 4 weeks of expansion were purified as described above and their phenotype and function were assessed.
T-reg phenotype was assessed by flow cytometry, using fluorescein isothiocyanate (FITC)-conjugated anti-CD4, phycoerythrin-conjugated anti-CD25, CD45RO, CD62L, and CD127 monoclonal antibodies (BD Biosciences, San José, CA). After 35 minutes of incubation in the dark at 4°C, cells were washed, resuspended in phosphate-buffered saline (PBS)/1% FBS, acquired on a FACSCalibur (BD Immunocytometry Systems) and analyzed using CellQuest software.
Apoptosis rate was assessed by staining cells with FITC-conjugated Annexin V (BD Biosciences), an early marker of apoptosis, and with propidium iodide to exclude necrotic cells. We washed 1 × 105 cells once with PBS/1% FBS and resuspended them in 1× binding buffer containing 0.1 M [4-(2-hydroxyethyl)-1-piperazine ethanesulfonic acid] /NaOH, 1.4 M NaCl, and 25 mM CaCl2. After addition of 5 μL FITC-Annexin V and propidium iodide, cells were incubated for 15 minutes in the dark and resuspended in 400 μL 1× binding buffer. Samples were acquired within 1 hour as described previously.
Freshly isolated CD4+CD25+ and CD4+CD25− T cells, as well as expanded and newly generated T-regs, were added before and after 48-hour resting in T cell expander and r-IL-2–free medium, to autologous CD4+CD25− T cells seeded at 3 × 105/well in 96-well plates at a ratio of 1/8.17 Experiments were performed in duplicate. After a 5-day culture, cells were pulsed with 0.25 μCi/well of [3H]thymidine and harvested after 18 hours. Incorporated thymidine was measured by β-counter (Canberra Packard Ltd., Pangbourne, UK). Percentage inhibition was calculated using the formula: 1 − [counts per minute (cpm) in the presence of T-reg/cpm in the absence of T-reg] × 100. The ability of expanded and newly generated T-regs to inhibit CD4+CD25− T cell proliferation as well as their FOXP3 gene and protein expression (see below) were assessed weekly for up to 4 weeks.
Quantification of FOXP3 Gene Expression
Cells (1 × 105 to 5 × 105) were lysed with TRizol reagent (Invitrogen Life Technologies) at a concentration of 0.1 ml/1 × 106 cells and total RNA was extracted. Messenger RNA (mRNA) was reverse transcribed using Oligo-(dT)12-18 primer (Invitrogen Life Technologies) and Omniscript Reverse Transcriptase (Qiagen Inc., Chatsworth, CA) and amplified by polymerase chain reaction (PCR) with gene-specific primers, Super TAQ polymerase (HT Biotechnology Ltd., Cambridge, UK), and TaqStart antibody (Clontech Laboratories UK Ltd., Basingstoke, UK). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as control gene. Primer sequences were as described.18 The complementary DNA (cDNA) amplification was performed out in 25 μL for 35 cycles of denaturation at 94°C for 30 seconds, annealing at 61°C for 30 seconds, and extension at 72°C for 30 seconds in an automated thermal cycler (Multiblock Satellite PCR System, Thermo Electron Corp., Basingstoke, UK).
FOXP3 transcripts were quantified by real-time PCR using gene-specific probes and TaqMan Master Mix (Applied Biosystems, Warrington, UK). PCR amplification conditions were: 50°C for 2 minutes, 95°C for 10 minutes, 40 cycles of 95°C for 15 seconds, and 60°C for 1 minute. Samples were run in triplicate using a real-time PCR thermocycler (ABI Prism 7000 Sequence Detection System, Applied Biosystems, Foster City, CA), and results were analyzed by matched software. Relative expression of FOXP3 gene was determined by normalizing to GAPDH expression according to the manufacturer's instructions.
Intracellular FOXP3 Staining
Cells (3 × 105) were washed and resuspended in PBS/1% FBS and stained with FITC-conjugated anti-CD25 and Cy-conjugated anti-CD4 monoclonal antibodies (BD Pharmingen). After 35 minutes incubation at 4°C, cells were washed, fixed, and permeabilized with Cytofix/Cytoperm, then counterstained with phycoerythrin-conjugated anti-human FOXP3 antibodies (clone PCH101) (eBioscience) used at 20 μL per 1 × 106 cells according to the manufacturer's instructions and based on preliminary experiments in which 10, 20, and 40 μL per 1 × 106 cells were tested. After 30 minutes incubation at 4°C, cells were washed and analyzed by flow cytometry. Cells were gated on CD4 lymphocytes. A minimum of 10,000 events were acquired per sample.
Cells (2 × 106 to 5 × 106) were lysed in ice-cold radio immunoprecipitation assay (RIPA) buffer containing 1% NP-40, 0.5% sodium deoxycolate, 0.1% sodium dodecyl sulfate (SDS) (Sigma-Aldrich Ltd., Gillingham, UK), and freshly added phenylmethyl sulfanyl fluoride (10 mg/mL isopropanolol; Sigma-Aldrich Ltd.), aprotinin (30 μL/mL; Sigma-Aldrich Ltd.), and sodium orthovanadate (18.39 mg/mL; Sigma-Aldrich Ltd.). After 40 minutes' incubation on ice and centrifugation at 15,000g at 4°C for 20 minutes, supernatants (total cell lysates) were collected and total protein concentration assessed by Bradford assay (Bio-Rad Laboratories, Hercules, CA). After protein denaturation with SDS, cell lysates were boiled for 5 minutes and separated on a 10% SDS-polyacrylamide gel electrophoresis. Eight micrograms protein was used per lane. Gels were run for 190 minutes at 126 V; proteins were then transferred onto 0.45 μm polyvinylidene fluoride membranes (Immobilon-P, Millipore, Watford, UK), subsequently incubated at room temperature in blocking buffer containing 5% skimmed milk, PBS, and 0.1% Tween 20. After 1 hour, rat anti-human FOXP3 monoclonal antibody (clone PCH101, eBioscience, San Diego, CA) was applied at 1 μg/mL. After overnight incubation at 4°C and washing, membranes were incubated for 1 hour at room temperature with horseradish peroxidase–labeled rabbit anti-rat antibody (1/2000). Bands were visualized with a nonradioactive enhanced chemiluminescent system (ECL Plus System, Amersham Pharmacia Biotech) on X-OMAT films (Kodak) and developed. For immunoblot normalization, the same membranes were stripped using a buffer containing 5% SDS, 500 mM Tris, and β-mercaptoethanol, reprobed with mouse anti-human GAPDH monoclonal antibody (Abcam, Cambridge, UK) at 1 μg/mL and subsequently with an horseradish peroxidase–labeled goat anti-mouse polyclonal antibody at 1/5000 dilution. Because of the small number of CD4+CD25+ T cells at baseline and up to week 3 of culture, immunoblot analysis was only performed at the end of week 4, whereas it was performed at baseline and throughout the whole culture period for up to 4 weeks for CD4+CD25− cells. FOXP3 and GAPDH band density was determined using Scion Image Processing Program (Release Beta 4.0.2).
Paired and unpaired Student t tests were used for comparing normally distributed data; Wilcoxon's rank sum test or Mann-Whitney test were used for non-normally distributed data. Correlations were determined by Pearson's or Spearman's correlation coefficient. A P value less than 0.05 was considered significant. Results are expressed as mean ± SEM. T-reg expansion is expressed as median fold increase and range.
T-reg Expansion from CD4+CD25+ T Cells
After exposure to high-dose IL-2 and anti-CD3/anti-CD28 T cell expander, the number of CD4+CD25+ T cells derived from the existing T-reg pool increased progressively up to the end of week 4 in normal controls and, to a lesser extent, in patients with AIH (Fig. 1). CD4+CD25+ T cells from 14 of the 22 (64%) normal controls and from 9 of the 24 (38%, P = 0.07) patients investigated continued proliferating up to the end of week 4 (Fig. 1); CD4+CD25+ T cells from 2 of the 22 (9%) normal controls and from 2 of the 24 (8%) patients continued proliferating up to the end of week 5, the number of fold-increase being 165 and 141 in the former and 10 and 25 in the latter. In both normal subjects and patients, the percentage of CD25+ cells among expanded T-regs remained higher than 95% throughout the culture period. These cells were CD25high according to their location within the upper right quadrant in flow cytometry.19 Expanded T-regs maintained their original phenotype, being CD45RO and CD62L positive and CD127 negative (Fig. 2A, B).
Apoptosis in Expanded T-regs.
In both normal controls and patients, the percentage of Annexin V positive cells increased from 49 ± 8.4 and 77.3 ± 13.1 (P = 0.1) at baseline to 75.1 ± 7.9 (P = 0.05) and 84.9 ±3.3 [P = nonsignificant (NS)] in expanded T-regs at the end of week 4.
T-reg Generation from CD4+CD25− T Cells.
T-regs were newly generated from CD4+CD25− T cells using the same protocol described for T-reg expansion. The number of CD4+CD25+ T cells derived from CD4+CD25− T cells increased progressively up to the end of week 4, decreasing thereafter (Fig. 1). This pattern was observed in both normal controls and, to a lesser extent, in patients with AIH. Newly generated T-regs from 18 of the 22 (82%) normal controls and from 13 of the 24 (54%, P = 0.045) patients continued growing up to the end of week 4; newly generated T-regs from 9 of the 22 (45%) normal controls and from 6 of the 24 (25%) patients continued growing up to the end of week 5. In normal controls, the mean percentage of newly generated T-regs was 80.1 ± 10.7 at the end of week 2, 83.4 ± 4.2 at the end of week 3, 85.2 ± 7.3 at the end of week 4, and 79.4 ± 6 at the end of week 5; in patients with AIH, it was 51.3 ± 15.7 at the end of week 2, increasing to 76.7 ± 6.4 (P = 0.016), 85.3 ± 3.9 (P = 0.003) and 73.5 ± 5.6 (P = 0.049) at the end of weeks 3, 4, and 5, respectively. In both normal subjects and patients with AIH, newly generated T-regs were CD45RO and CD62L positive and CD127 negative (Fig. 2C, D).
Apoptosis in Newly Generated T-regs
The baseline percentage of Annexin V positive cells in CD4+CD25− T cells was 38.5 ± 9.3 in normal controls and 26.1 ± 10.2 in patients (P = 0.14), increasing respectively to 60.3 ± 4.2 (P = 0.025) and 60.1 ± 1.7 (P = 0.0009) in newly generated T-regs at the end of week 4. Subsequent experiments were performed only during the first 4-week culture period, owing to the high rate of apoptosis at the end of week 4 in both expanded and newly generated T-regs.
Suppressor Function of Expanded and of Newly Generated T-regs
The ability of expanded and newly generated T-regs to suppress freshly isolated CD4+CD25− T cell proliferation was assessed in 9 normal controls and 12 patients with AIH (Fig. 3A, D). In normal controls, addition of CD4+CD25+ T-regs decreased mean CD4+CD25− T cell proliferation by 23% (P = 0.048) at baseline, 29.4% (P = 0.001) at the end of week 2, 60% (P < 0.001) at the end of week 3, and 83.7% (P < 0.001) at the end of week 4. In AIH patients, T-regs decreased the mean CD4+CD25− proliferation by 13% (P = 0.08) at baseline, 14% (P = 0.054) at the end of week 2, 17% (P = 0.099) at the end of week 3, and 31.2% (P = 0.047) at the end of week 4.
The ability of freshly isolated CD4+CD25− T cells, and T-regs newly generated from them, to suppress CD4+CD25− T cell proliferation was tested in the same controls and patients (Fig. 3B, E). In normal controls, CD4+CD25− T cell addition did not alter the proliferation of CD4+CD25− T cells at baseline; newly generated CD4+CD25+ T cells did not reduce CD4+CD25− proliferation at the end of week 2, but they decreased it by 17.6% (P = 0.13) at the end of week 3 and by 37.1% (P = 0.028) at the end of week 4. In patients, freshly isolated CD4+CD25− T cells did not alter the proliferation of CD4+CD25− T cells. Newly generated CD4+CD25+ T cells did not reduce CD4+CD25− proliferation at the end of week 2, but reduced it by 5.4% at the end of week 3 and by 22.3% (P = 0.041) at the end of week 4. At the end of week 4, after their further purification through CD25 immunobead-positive selection, newly generated T-regs reduced CD4+CD25− proliferation by 41.5% (P = 0.03) in normal controls and by 29.5% (P = 0.047) in patients (Fig. 3C, F).
Because no inhibition of CD4+CD25− T cell proliferation was observed over the 4-week culture period when expanded and newly generated T-regs were added before resting them for 48 hours in IL-2–free and T cell expander–free medium, the assays described in the following sections were performed after cell resting.
FOXP3 Gene and Protein Expression in Freshly Isolated CD4+CD25+ T Cells and Expanded T-regs
FOXP3 gene expression was assessed in 9 controls and 12 patients with AIH. FOXP3 protein expression was measured as fluorescence intensity in cytofluorimetry in 16 controls and 12 AIH patients and as band density in immunoblot in 12 controls and 14 patients. In normal controls, the levels of FOXP3 gene expression and FOXP3 protein expression (immunoblot) were 6.2 ± 1.7 and 37.01 ± 3.16, respectively, before stimulation, remained unchanged at the end of week 2, and increased thereafter (Fig. 4A, B). In AIH patients, the level of FOXP3 gene expression in T-regs before stimulation was lower compared with normal controls (1.7 ± 0.6 versus 6.2 ± 1.7; P = 0.017) and increased progressively throughout the 4-week culture (Fig. 4C). FOXP3 protein expression was measured by cytofluorimetry only from the second week onward because of the small number of T-regs at baseline, and was 18.6 ± 1.6 at the end of week 2, remained unchanged at the end of week 3, and increased significantly at the end of week 4 (Fig. 4D). The levels of FOXP3 gene and protein expression in expanded T-regs were higher in normal controls than in patients at all time points. At the end of week 4 for gene expression and at the end of weeks 2 and 4 for fluorescence intensity, however, the difference reached conventional levels of statistical significance (P = 0.007, P = 0.027, and P = 0.041, respectively). At the end of week 4, the level of FOXP3 band density in total cell lysates was higher in normal controls than in patients with AIH (P = 0.027) (Fig. 4E). The numbers of expanded T-regs obtained from both patients and controls were not sufficient to perform all assays in each subject, precluding the possibility of establishing a correlation between levels of gene and protein expression.
FOXP3 Gene and Protein Expression in Freshly Isolated CD4+CD25− T Cells and Newly Generated T-regs
Levels of FOXP3 gene and protein expression were measured in CD4+CD25− T cells and newly generated T-regs obtained from the same healthy controls and AIH patients, described in the previous section.
In freshly isolated CD4+CD25− T cells, the level of FOXP3 gene expression was higher in normal controls (2.7 ± 0.9) than in patients (0.7 ± 0.3; P = 0.05), whereas no difference between the 2 groups was noted at the protein level tested by both cytofluorimetry and immunoblot. Newly generated T-regs from normal controls had a higher level of gene and protein expression at the end of week 2, followed by a decrease thereafter, remaining, however, higher than before stimulation (Fig. 5A–C). Similarly, in newly generated T-regs obtained from patients, there was an increase in the levels of FOXP3 gene and protein expression at the end of week 2, followed by a decrease thereafter, with levels remaining higher than before stimulation (Fig. 5D–F). Figure 6 shows the density of FOXP3 protein throughout the 4-week culture in representative normal (Fig. 6A) and AIH (Fig. 6B) cases.
Levels of gene and protein tended to be higher in normal controls than in patients at all time points (P = 0.13 at the end of week 3 and P = 0.05 at the end of week 4 for gene expression; P = 0.078 at the end of week 2 and P = 0.13 at the end of week 4 for protein expression measured by cytofluorimetry; P = 0.031 at the end of week 3 for protein expression measured by immunoblot).
Compared with the end of week 4, the level of FOXP3 gene and protein expression in newly generated purified T-regs increased in both normal controls and patients (Fig. 7A-C).
In contrast to expanded T-regs, numbers of newly generated T-regs from both patients and controls were sufficient to perform the 3 assays in most subjects. In normal controls, but not in patients, there was a positive correlation between the level of FOXP3 gene expression and that of protein expression tested by both cytofluorimetry (R = 0.8; P = 0.005) and immunoblot (R = 0.46, P = 0.043), whereas a positive correlation between the levels of FOXP3 protein expression tested by cytofluorimetry and immunoblot was found in both normal controls and patients (R = 0.67; P = 0.043 and R = 0.75; P = 0.087, respectively).
This study shows that T-regs can be expanded from circulating CD4+CD25+ regulatory T cells and generated de novo from non-regulatory CD4+CD25− T cells in healthy subjects and in patients with autoimmune hepatitis. The finding is of particular relevance in AIH, in which expansion of T-regs is a prerequisite for attempting to reconstitute impaired immune regulation and restore peripheral tolerance through T-reg infusion. The study also documents sequentially, over a period of 4 weeks, the relationship between T-reg inhibitory function and the expression of the key regulatory molecule FOXP3 at gene and protein level.
Though T-regs are known to be poor proliferators, we have succeeded in expanding them in both normal controls and, to a lesser extent, in patients with AIH using a strategy that engages the T cell receptor via CD3 and the key costimulatory molecule CD28 in the presence of high doses of IL-2, a cytokine critical for T-reg survival and growth. This approach for expanding T-regs was chosen because, compared with other strategies, such as the use of the immunosuppressant rapamycin,21–23 it allows not only the achievement of higher percentages of CD25high cells but also potential generation of T-regs from CD4+CD25− T cells, a population whose expansion is selectively blocked by rapamycin.
The protocol used in this study has allowed us to obtain expanded CD4+CD25+ T-regs with a suppressor function more vigorous than previously reported,12, 13 the expanded T-regs exerting an 80% inhibition even at the low ratio of 1/8. Moreover, with sequential experiments we have shown a progressive increase in suppressor function over the 4-week period of culture. The knowledge of the T-reg phenotypical and functional changes occurring during their expansion provides information as to when the cells exert the strongest suppressor activity and therefore can be used for immunotherapeutic purposes. During the 4-week culture period, we have also been able to monitor changes in the expression at gene and protein level of the key regulatory molecule FOXP3. In T-regs expanded from CD4+CD25+ T cells, derepression of the FOXP3 gene and an increase in suppressor function occurred sequentially after the engagement of the T cell receptor, an event necessary to T-reg development.24 In mice, a key link between T cell receptor engagement and T-reg expansion is a transmembrane adaptor protein, called a linker for activation of T cells,25 mice bearing a linker for activation of T cells knock-in mutation having defective CD4+CD25+FOXP3+ cells and developing severe autoimmune disease.25 Future studies should elucidate the role of linkers for activation of T cells in human autoimmune disease.
For potential therapeutic purposes, our ultimate aim is to maximize the pool of functional T-regs in patients. Because in both patients and controls the yield of expanded T-regs is low and their apoptotic rate high, limiting the number of cells suitable for infusion, we have explored the possibility of generating functional T-regs from CD4+CD25− T cells, a heterogeneous cell subpopulation mainly composed of effector cells but also including a subset of cells with regulatory potential. Using the same approach used for the expansion of conventional T-regs, we succeeded in generating CD4+CD25+ T cells with the phenotypic markers of T-regs, being CD25high, CD45RO and CD62L positive, both in healthy subjects and in patients with AIH. At variance with previous reports,12, 13 the CD4+CD25+ T cells we obtained were also able to exert suppressor activity: the reason for this difference with the published experience may depend on our use of lower doses of IL-2, chosen to favor T-reg suppressor activity both in vitro and in vivo.26, 27 A further reason could be the fact that we rested the cells in IL-2–free and T cell expander–free medium for 2 days before use. This approach was adopted based on previous published evidence showing that, after removal of IL-2 and anti-CD28 antibody, CD4+CD25+ T cells enhance their suppressor activity.8 Newly generated T-regs started exerting suppressor function only after the peak of FOXP3 expression, suggesting that FOXP3 expression is a prerequisite for the generation and development of T-regs.
At variance with T-regs expanded from CD4+CD25+ T-regs, newly generated T-regs have greater ability to expand and are more resistant to apoptosis, features compensating for their less powerful suppressive function. The fact that newly generated T-regs, despite their suppressor phenotype, have lower suppressor activity than those expanded from CD4+CD25+ T cells may derive from contamination with Th17 cells, a lineage sharing the same CD4 progenitor with CD4+CD25+ T-regs but exerting effector functions and restraining T-reg development.28, 29 This possibility is supported by the observation that, after further purification of newly generated CD4+CD25high T cells at the end of week 4, their suppressor activity increased.
Clinical application of adoptive transfer of autologous expanded or newly generated T-regs requires further investigation of their tolerability and half-life in humans. In mice, function of adoptively transferred expanded T-regs can still be detected 23 days after injection.30
T-reg suppressor function and FOXP3 expression during culture require a longer time to reach their peak in expanded T-regs from patients than from healthy subjects, suggesting different kinetics. Moreover, the fact that a positive correlation between the levels of FOXP3 gene and protein expression was seen in normal controls but not in patients may indicate a defect in the mechanisms regulating FOXP3 gene transcription or translation in AIH. A number of FOXP3 mutations have been described in children affected by the immune-dysregulation polyendocrinopathy enteropathy X-linked syndrome.31 Different mutations lead to different degrees of FOXP3 protein expression, ranging from undetectable to normal levels, and are accompanied by different degrees of T-reg impairment and clinical manifestations.31 Whether the lack of correlation between gene and protein expression in patients with AIH is caused by mutations within the forkhead domain of FOXP3 gene warrants further investigations.
In summary, we have provided evidence that T-regs can be expanded and generated de novo in patients with AIH. These findings may represent the basis for developing new effective forms of immune intervention, aiming at reestablishing immunotolerance. This approach should be investigated not only in conditions in which impaired suppressor function is the result of a reduced T-reg number, such as primary biliary cirrhosis,32 but also in those with a qualitative T-reg defect, such as rheumatoid arthritis10 and insulin-dependent diabetes.33
- 7Organ-specific autoimmune diseases induced in mice by elimination of T cell subset. I. Evidence for the active participation of T cells in natural self-tolerance; deficit of a T cell subset as a possible cause of autoimmune disease. J Exp Med 1985; 161: 72–87., , , .