By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Please be advised that we experienced an unexpected issue that occurred on Saturday and Sunday January 20th and 21st that caused the site to be down for an extended period of time and affected the ability of users to access content on Wiley Online Library. This issue has now been fully resolved. We apologize for any inconvenience this may have caused and are working to ensure that we can alert you immediately of any unplanned periods of downtime or disruption in the future.
Chronic human lung allograft rejection is manifested by bronchiolitis obliterans syndrome (BOS). BOS has a multifactorial etiology. Previous studies have indicated that both cellular and humoral alloimmunity play a significant role in the pathogenesis of BOS. Recently, autoimmunity has also been demonstrated to contribute to lung allograft rejection in animal models. However, the significance of autoimmunity in BOS remains unknown. In this report, we investigated the role of naturally occurring CD4+CD25+ regulatory T cells (T-regs) in modulating cellular autoimmunity to collagen type V (col-V), a ‘sequestered’ yet immunogenic self-protein present in the lung tissue, following lung transplantation (LT). We demonstrated that col-V reactive CD4+ T cells could be detected in the peripheral blood of lung transplant recipients. There was a predominance of IL-10 producing T cells (TIL-10) reactive to col-V with significantly lower levels of IFN-γ and IL-2 producing T cells (Th1 cells). The col-V specific TIL-10 cells suppressed the proliferation and expansion of col-V specific Th1 cells by IL-10-dependent and contact-independent pathways. The TIL-10 cells were distinct but their development was dependent on the presence of T-regs. Furthermore, during chronic lung allograft rejection there was a significant decline of TIL-10 cells with concomitant expansion of col-V-specific IFN-γproducing Th1 cells.
Chronic allograft rejection has a multifactorial origin. Bronchiolitis obliterans syndrome (BOS) represents chronic rejection in human lung allografts (1) and is the predominant cause that limits the long-term success of lung transplantation (LT) (2). BOS develops in over 75% of human lung allograft recipients at 5 years and about 90–100% after 9 years post-transplantation (3). Obliterative bronchiolitis (OB), the histological correlate of BOS (4), is characterized by excessive fibro-proliferation and abnormal tissue remodeling. The hallmark pathological lesion of OB is collagen deposition (5), which results from excessive fibroblastic activity. Although the pathogenesis of BOS is not entirely understood, our previous studies have indicated a significant role of alloimmunity (3,6,7). We have also demonstrated that non-HLA anti-airway epithelial cell (AEC) antibodies (Abs) can develop in LT patients (3). These anti-AEC Abs induce intracellular Ca++ influx, tyrosine phosphorylation and up-regulation of transforming growth factor beta and heparin-binding epidermal growth factor in the AECs that can stimulate fibroblastic proliferation and lead to collagen deposition (3). Emerging evidence indicates that chronic allograft rejection might be an ‘injury response’ which leads to abnormal tissue remodeling, and both alloimmune-dependent and -independent events contribute to its pathogenesis (8).
Alloimmunity has been shown to influence acute and chronic allograft rejection. However, recent data have indicated that autoimmunity could also contribute to the pathogenesis of allograft rejection (9–11). Nonpolymorphic proteins like myosin (10), vimentin (12) and heat shock protein (9) have been identified as potential autoimmune targets for the development of allograft rejection. Compeling data from Wilkes' lab have also indicated that autoimmunity to collagen type V (col-V), a ‘sequestered’ yet immunogenic self-protein present in the lung tissue, may play a significant role in the development of chronic rat lung allograft rejection (13,14). OB is known to occur even in nontransplant settings, such as autoimmune connective tissue (collagen vascular) disorders (15–17) and due to a variety of inhalational injuries (16) that lead to lung parenchymal inflammation. Lung allografts are uniquely susceptible to injuries from a variety of exogenous agents due to their direct communication with the external environment. Post-transplant gastroesophageal reflux (18,19), chronic viral infections (20) and repeated acute rejection episodes (2) have been strongly associated with the development of BOS. These risk factors create an inflammatory milieu within the lung allografts that is favorable for the development of autoimmunity (21).
Recent data have shown that naturally occurring regulatory CD4+25+ T-regs can modulate both organ-specific autoimmunity in normal subjects and allo-specific immune responses in transplant recipients (22). However, the mechanisms of T-reg mediated suppression both in vitro and in vivo are unclear. In addition, the role of T-regs in maintaining peripheral tolerance to self-proteins after human allo-transplantation is not well defined. In this report, we demonstrated that col-V reactive T cells are present in LT patients and that T-regs can indirectly modulate the expansion of col-V autoreactive Th1 cells by inducing IL-10 producing suppressor T cells (TIL-10).
Materials and Methods
Patients undergoing LT at the Washington University Medical Center/Barnes-Jewish Hospital were prospectively enrolled in the study after obtaining informed consent in accordance with a protocol approved by the Institutional Review Board. At each follow-up visit, the peripheral blood mononuclear cells (PBMCs) were isolated from heparinized blood by Ficoll–Hypaque density gradient centrifugation (Pharmacia, Sweden), and stored at –135°C until further use. The patients did not have acute allograft rejection within at least 3 months of analysis. BOS was diagnosed based on the percentage decline in forced expiratory volume in 1 s (FEV1) compared to baseline (1) and graded as follows: 1 = 80–66% of baseline value, 2 = 65–51% of baseline value and 3 = 50% or less of baseline value. Other causes of decreased lung function such as infection and bronchial anastomotic stricture were ruled out. BOS positive (+ve) patients selected for our experiments were grade II & III.
The post-transplant follow-up of the BOS+ve patients until the onset of BOS was divided into four equal quadrants and one sample obtained at the midpoint of each interval was analyzed. For every BOS+ve patient a closely matched BOS−ve patient was selected. The clinico-demographic variables used to match BOS+ve and BOS−ve patients included age, sex, primary lung pathology, year of transplantation and type of transplant (unilateral or bilateral). The BOS−ve samples analyzed belonged to a comparable posttransplant interval as their BOS+ve counterparts. For example, if a patient developed BOS at 32 months after transplantation, this interval was divided into quadrants, namely 0–8, 8–16, 16–24 and 24–32 months. Samples were selected from mid-quadrant at 4 (sample 1), 12 (sample 2), 20 (sample 3) and 28 (sample 4) months, respectively. In addition, a post-BOS sample obtained at 36 months (Sample 5) was analyzed. For the matched BOS−ve control, samples analyzed were also obtained from visits at 4, 12, 20, 28 and 36 months post-transplant. The objective behind using this strategy was to analyze samples in the BOS+ve and BOS−ve groups at comparable intervals post-transplantation as that can influence the state of immune system. The clinico-demographic profile of all patients is illustrated in Table 1.
Table 1. Clinico-demographic profile of patients evaluated for collagen-V immunity
48.5 ± 13.9
48.9 ± 12.2
50.9 ± 11.9
2.69 ± 1.5
2.75 ± 1.3
2.67 ± 1.5
0.86 ± 0.8
0.87 ± 0.7
0.85 ± 0.8
Total acute rejection episodes
0.79 ± 0.96
0.92 ± 0.18
0.66 ± 0.17
272.0 ± 14.3
253.1 ± 15.5
297.0 ± 11.8
299.9 ± 13.4
289.1 ± 13.8
322.8 ± 13.5
Type of transplant
Graft survival (months)
37.2 ± 3.6
24.4 ± 5.6
49.6 ± 5.9
Mean study follow-up (months)
30.1 ± 5.6
29.6 ± 6.8
30.9 ± 6.6
In vitro expansion of col-V specific T cells
To expand col-V specific CD4+ T cells, we generated cell lines from PBMCs of 20 LT patients. PBMCs were stimulated with irradiated (3000 rads) autologous PBMCs (1:1 ratio) and human col-V (20 μg/mL, BD Biosciences, San Jose, CA, USA) in RPMI-1640 medium (Gibco BRL, Grand Island, NY, USA) supplemented with 15% heat-inactivated endotoxin-free fetal bovine serum (Hyclone, Logan, UT, USA), L-glutamine (2 mmol/L), nonessential amino acids (100 μmol/L), HEPES (25 mmol/L), sodium pyruvate (1 mmol/L), penicillin (100 U/mL), streptomycin (100 μg/mL), recombinant IL-2 (20 U/mL; Chiron, Emeryville, CA, USA) and T-stim without phytohemagglutinin (10%, BD Biosciences). The cultures were supplemented with fresh IL-2 (20 U/mL) every 4 days. After the sixth stimulation, the cells were rested in the endotoxin-free hyclone media for 8 days. Following this, CD4+ T cells were purified using negative selection with a depletional cocktail for non-CD4 cells (Dynal Biotech, Carlsbad, CA, USA).
ELISPOT assays for IL-10, IFN-γ, IL-2, IL-4 and IL-5 were performed as previously described (23). Briefly, multiscreen 96-well filtration plates (Millipore, Bedford, MA, USA) were coated overnight at 4°C with 5.0 μg/mL of capture human cytokine-specific mAb (BD Biosciences) in 0.05 M carbonate–bicarbonate buffer (pH 9.6). The plates were blocked with 1% BSA for 2 h and washed (3×) with phosphate buffered saline (PBS). Subsequently, 3 × 105 cells were cultured in triplicate in the presence of col-V (20 μg/mL) and feeder autologous PBMCs (APCs) (1:1 ratio). After 48 h, the plates were washed with PBS (3×) and 0.05% PBS-Tween-20 (3×). Then, 2.0 μg/mL of a biotinylated human cytokine-specific mAb (BD Biosciences) in PBS/BSA/Tween-20 was added to the wells. After an overnight incubation at 4°C, the plates were washed (3×) and horseradish peroxidase-labeled streptavidin (BD Biosciences), diluted 1:2000 in PBS/BSA/Tween-20, was added to the wells. After 2 h, the assay was developed by 3-amino-9-ethylcarbazole substrate reagent (BD Biosciences) for 5–10 min. The plates were washed with tap water to stop the reaction and air-dried. The spots were analyzed in an ImmunoSpot Series I analyzer (Cellular Technology, Cleveland, OH, USA). Any spots obtained by culturing T-cell lines with APCs alone without col-V were subtracted from the number of spots in the experimental cultures. The results were expressed as number of spots per million cells (spm) ± standard error.
Cells were seeded in triplicate cultures in flat 96-well plates (Falcon, BD Labware, Franklin Lakes, NJ, USA) at a concentration of 2×105 cells/well in the presence of col-V (20 μg/mL) and APCs (1:1). After 4 days, the cultures were pulsed with [3H]-thymidine (1 μCi/well) for 18 h following which [3H]-thymidine incorporation into DNA was determined by means of liquid scintillation counting. The results expressed as counts per minute (cpm) after subtracting the counts obtained from the culturing cells with autologous APCs without col-V.
Transwell experiments were performed in 24-well plates (Costar, Cambridge, MA, USA) with 0.4 μm membrane supports. In parallel, cells from BOS+ve and BOS−ve patients were also co-cultured in presence of neutralizing anti-IL10 mAbs (50 μg/mL; BD Pharmingen, San Diego, CA, USA) or anti-TGFβ (20 μg/mL, R&D Systems [Minneapolis, MN, USA], as per supplier's protocol). For all experiments, the ratio of cells from BOS+ve and BOS−ve on either side of the membrane was 1:1. At the end of co-culture, CD4+ T cells were negatively selected and col-V specific response measured using ELISPOT assays. In addition, col-V cell lines from BOS−ve patients were co-cultured with CD3/CD28 bead (Dynal Biotech) stimulated CD25− T cells (1:1 ratio) either in the presence or absence of neutralizing anti-IL10 mAbs (50 μg/mL, BD Pharmingen).
Activation-fixation of T-regs
Activation-fixation of natural T-regs was performed as described by Jonuleit et al. (24). CD25high natural T-regs were positively separated from the PBMCs using Flow-cytometry assisted cell sorting (FACS; Beckman Coulter, Miami, FL, USA) after staining with Phycoerythrin (PE)-conjugated mouse human anti-CD25 mAbs (BD Pharmingen). The CD25high T-regs isolated had a CD4+ purity of greater than 96%. Subsequently, the T-regs were cultured with anti-CD3/CD28 beads (Dynal Biotech). Following overnight incubation, the T-regs were fixed in 10% paraformaldehyde for 10 min. The feeder PBMCs depleted of all CD25+ T cells using FACS were irradiated (3000 rads), mixed with the isolated T-regs and added alongwith col-V to the cell cultures (1:1 ratio).
IL-10 producing col-V reactive T cells are present in the peripheral blood of LT patients
We first tested if col-V reactive T cells were present in the peripheral blood of LT patients and normal subjects. Toward this, the PBMCs from LT recipients obtained during the first year post-transplant and normal subjects were stimulated with col-V in the presence of irradiated autologous PBMCs as APCs to develop T-cell lines. CD4+ T cells were purified by negative selection following the last stimulation. The cells obtained were more than 80% CD4+ (data not shown). Col-V specific response was measured by ELISPOT assays in the presence of autologous APCs and col-V. Col-V T-cell lines could be developed in 17 out of 20 LT patients tested. The col-V specific CD4+ T cells in LT patients were predominantly IL-10 producing (226 ± 42.8 spm) with low IFN-γ (50.2 ± 14.3 spm), IL-2 (27.4 ±7.5 spm), IL-5 (3.2 ±0.8 spm) and no IL-4 production (Figure 1). This represents a Tr1-type immunity (25,26). In contrast, col-V T-cell lines could only be developed from three of nine (33%) normal individuals (‘Fisher's exact’ p = 0.010). Besides, there was a significantly lower frequency of col-V reactive CD4+ T cells in these cell lines: IL-10 (61.0 ± 9.5 spm, ‘two-tailed t-test’ p < 0.001), IFN-γ (19.0 ± 5.0 spm,), IL-2 (9.0 ± 3.0 spm), IL-5 (8.5 ± 0.5 spm) and no IL-4. In the absence of APCs and col-V TIL-10 cells did not constitutively produce any IL-10: (0 spm). Moreover, there was low cross-reactivity with an unrelated protein ovalbumin (12.2 ± 8.5 spm) or collagen type II (22.2 ± 8.5 spm). These results suggested that there was a higher prevalence of T-cell clones specific to col-V in the peripheral blood of LT recipients than normal subjects.
Development of BOS was associated with a decline in col-V reactive TIL-10
We next serially analyzed the col-V specific response in LT recipients. Eight BOS+ve patients were randomly selected. The interval between LT and BOS development was divided into four equal intervals. One PBMC sample from each interval was analyzed for col-V specific reactivity as described above. For each BOS+ve patient one BOS−ve patient matched for age, sex, primary lung pathology, type of immunosuppression and transplant (unilateral vs. bilateral) was selected. One sample obtained from the follow-up visit of the BOS−ve patient corresponding to the BOS+ve counterpart was analyzed in the same fashion. The aim was to compare the col-V immunity between BOS+ve and individually matched BOS−ve patients serially at comparable intervals following transplantation. There was no significant difference in the ischemia time, number of acute rejection episodes or the level of donor HLA-mismatch (Table 1). The results in each of the four intervals were then averaged separately for the BOS+ve and BOS−ve patients (Figure 2).
LT recipients were found to have a higher frequency of TIL-10 cells (226 ± 42.8 spm) with low IFN-γ (50.2 ± 14.3 spm) early during the post-transplant follow-up. However, five of the eight patients that developed BOS showed a significant decline in the frequency of these TIL-10 cells at the time of BOS development (23.7 ± 3.7 spm; >nine-fold decrease, ‘paired two-tailed t-test’ p = 0.001). This correlated with an increase in the frequency of IFN-γ producing T cells (TIFN-γ, Th1-predominance) that peaked during BOS (368.3 ± 15.6 spm; >six-fold increase, ‘paired two-tailed t-test’ p = 0.011). In contrast, all patients that remained BOS–ve maintained a consistently higher level of TIL-10 cells (mean 199.5 ± 10.9 spm) and low TIFN-γ (Th1) cells (mean 47.6 ± 24.7 spm). We and others have previously shown that development of BOS correlates with an elevation of Th1-cytokines, particularly IFN-γ, within the lung allograft (27–30). These experiments suggested that there was a derangement of regulatory mechanisms in LT patients that developed BOS that may be responsible for the reversal of immune predominance from Tr1 to Th1.
IL-10 producing CD4+ T cells suppressed col-V specific IFN-γ producing Th1 cells in LT patients
We next determined whether the TIL-10 cells were responsible for preventing the expansion of IFN-γ producing Th1 cells in patients who remained BOS–ve. To investigate this, PBMCs from BOS+ve and BOS−ve patients were first stimulated three times with col-V to first expand the col-V reactive TIFN-γ or TIL-10 cells, respectively. Following this, co-culture was performed during another three rounds of stimulation in a transwell plate. The transwell membrane prevents direct cellular contact between cells on either side but allows free diffusion of cytokines. Following the last stimulation, the col-V specific response was analyzed as before (Figure 3). After co-culture, we found a significant reduction both in the frequency of IFN-γ producing T cells (117.0 ± 16.20 spm; > three-fold decrease, ‘paired two-tailed t-test’ p = 0.03) and total cellular proliferation (22 ± 3 × 103 cpm; > 2.5-fold decrease, ‘paired two-tailed t-test’ p = 0.03) in BOS+ve cell lines compared to the initial levels: IFN-γ (368.3 ± 25.6 spm) and cellular proliferation (55 ± 7 × 103 cpm). Moreover, the inhibition of Th1-immunity in the BOS+ve cells observed after co-culture with BOS−ve was significantly reversed in the presence of neutralizing anti-IL10 mAbs: IFN-γ (262.3 ± 29.1 spm; >two-fold increase, ‘paired two-tailed t-test’ p = 0.021) and cellular proliferation (44 ± 6 × 103 cpm; >two-fold increase, ‘paired two-tailed t-test’ p = 0.019). In contrast, neutralizing TGF-β did not result in any significant reversal of inhibition. There was no statistically significant difference in the col-V response of BOS−ve cells after co-culture (data not shown). No suppression during the co-culture was observed if the BOS−ve cell lines were devoid of APCs and col-V. Moreover, co-culture of BOS+ve cell lines with those from clinically matched but HLA-disparate BOS+ve patients did not result in any suppression (Figure 3). These data demonstrated that the TIL-10 cells could limit Th1-autoreactivity in BOS−ve patients and that their decline at the time of BOS development may lead to Th1-predominant immunity to col-V. Further, TIL-10 activated in presence of col-V and autologous APCs suppressed polyclonal activation of conventional CD25− T cells (Figure 3C) that could be reversed by neutralizing IL-10 indicating that activation of TIL-10 cells can result in bystander suppression, possibly including alloimmunity.
Development of TIL-10 was dependent on natural T-regs
We next investigated the origin of these TIL-10 cells. It was possible that they represented a regulatory T-cell subset generated in response to col-V release within the lung allografts. T-regs have been shown to induce IL-10 production in naïve T cells following in vitro mitogenic stimulation (24,31). To investigate the role of T-regs in the development of IL-10 predominant col-V immunity, we stimulated the PBMCs from LT recipients in the absence of T-regs. CD25+ T cells were depleted from the original PBMC sample used to generate the cell line and from the APCs used for stimulation using FACS. If the development of these TIL-10 cells was dependent on T-regs, a loss of IL-10 production would be expected. In parallel experiments, CD25+ T cells isolated from the PBMCs were reconstituted back into the cultures. Interestingly, the cell lines generated in this manner showed not only markedly low IL-10 production (39.3 ± 12.8 spm; >seven-fold decrease, ‘paired two-tailed t-test’ p = 0.007) but also a concomitant increase in IFN-γ producing T cells (250.6 ± 52.7 spm; >four-fold increase, ‘paired two-tailed t-test’ p = 0.028) compared to when CD25+ T cells were present: IL-10 (291.0 ± 24.2 spm) and IFN-γ (57.3 ± 24.8 spm) (Figure 4). Reconstitution of CD25+ T cells back into the cell cultures restored the suppressive cytokine profile: IL-10 (221.0 ± 25.7 spm) and IFN-γ (54.7 ± 16.4 spm). These results indicated that the development of TIL-10 cells was dependent on T-regs and in the absence of IL-10 mediated suppression there was an expansion of TIFN-γ Th1 cells.
TIL-10 were distinct from natural T-regs
Since T-regs are generally anergic in vitro (32), we hypothesized that the predominant TIL-10 cells were distinct from natural T-regs. However, to eliminate the possibility of T-regs differentiating into col-V specific TIL-10 cells, we stimulated the PBMCs with col-V in the presence of activated but fixed T-regs. The original PBMCs and autologous APCs were depleted of CD25+ T cells using FACS. T-regs were collected from the CD25+ fraction by gating on CD25high T cells as shown in Figure 5 and had a CD4+ purity of greater than 96%. The T-regs were then cultured in the presence of anti-CD3/CD28 beads overnight following which they were fixed in paraformaldehyde and washed thoroughly. The CD25+ T-cell depleted APCs were irradiated (3000 rads), mixed with the activated–fixed T-regs and used as feeders for the CD25+ T-cell depleted PBMCs at 1:1 ratio in presence of col-V. This procedure eliminated the possibility of T-regs differentiating into TIL-10, yet retaining their suppressive functions which are dependent on signaling through surface molecules upregulated following activation (31,33). Cell lines developed in the presence of activated-fixed T-regs revealed a near complete restoration of TIL-10 cells (221.0 ± 25.10 spm; ‘paired two-tailed t-test’ p = 0.001) in contrast to those generated after T-reg depletion (39.0 ± 12.8 spm) (Figure 6). These results demonstrated that the predominant TIL-10 cells were distinct but required T-regs for development.
BOS and its histological correlate OB are generally accepted to have a multifactorial etiology. Previous studies from our laboratory along with reports from other investigators support the notion that BOS results from multiple injuries sustained by the lung allografts (34). Recent studies performed using animal models of solid organ transplantation have indicated that autoimmunity to nonpolymorphic self-proteins can contribute to the development of allograft rejection (9,10,35). Also, using a rat lung model of single LT, it has been demonstrated that autoimmunity to col-V may play a significant role in the development of lung allograft rejection (13,14).
Col-V is a minor collagen in the lung (36). During inflammation and lung tissue repair, the col-V levels in the lungs increase significantly (37,38). Col-V molecule has several unique properties. It maintains the –NH2 and –COOH terminal ends, making it more immunogenic (39). However, it is incorporated into collagen type-I and III (37,40), hence ‘sequestered’, within the tissues under normal circumstances. Matrix metalloproteinases (MMPs)-2,9 are capable of cleaving collagen molecules (41). Further, the activity of MMP-2 and MMP-9 has been shown to get upregulated after LT mainly during allograft rejection (42,43). In fact, col-V has been detected in the bronchoalveolar lavage fluid in human and rat lung allografts (13,44). Col-V reactive T cells were shown to be present in rat lung allografts undergoing rejection. Moreover, col-V specific T-cell lines derived from rat lung allografts with OB induced rejection of isografts when adoptively transferred into isograft recipients without significantly affecting the native lungs (13).
Lung allografts undergo continuous injury–repair cycles due to multiple insults sustained during and after transplantation. Such an inflammatory milieu is conducive for the expansion of self-reactive lymphocytes due to: i) release of previously ‘cryptic’ determinants and neo-antigens, ii) lowering of T-cell activation threshold and priming of autoreactive T cells with ‘low-affinity’ T-cell receptors (TCRs), which were previously below their activation potential (21), and iii) diversification of epitope specificity of alloreactive T cells to new allo- or autoantigens through epitope spreading (20,45). These events may have lead to the expansion of col-V reactive T cells in LT patients (Figure 1). It has been demonstrated that T cells maybe attracted to the primary target antigenic site (pulmonary tissue in our study) after activation, leaving a low frequency in the peripheral blood (46). This may explain why the detection of col-V specific T cells required prior expansion. However, initial experiments performed using a trans vivo murine model have already shown that PBMCs obtained from LT patients at the time of BOS development can directly mount a strong delayed-type hypersensitivity response to col-V injected into mouse foot-pads (47).
BOS development has been correlated with a Th1 immunity within the lung allografts (27,28). We have previously shown that the mRNA levels of Th1 cytokine IFN-γ in lung allografts progressively increase and peak at the time of BOS development (30). In the present experiments, BOS development was associated with an increase of col-V reactive Th1 cells with a concomitant loss of TIL-10 cells (Figure 2). These results are in agreement with the earlier reports and demonstrate that the BOS is associated with Th1-predominance to both allo- and self-peptides. Further studies are required to characterize the contribution of col-V autoimmunity in the pathogenesis of BOS.
Natural T-regs modulate both autoimmunity and transplantation tolerance (48). Although T-regs are anergic in vitro, this anergy can be overcome by TCR stimulation and high doses of IL-2 (32). In our study depletion of T-regs caused marked decline in IL-10 production indicating that T-regs were pivotal for the IL-10-mediated suppression of Th1 cells during co-culture with BOS+ve cells. However, reconstitution of activated-fixed T-regs significantly restored the IL-10 production. Hence the predominant IL-10 production was from T cells that were distinct but required the presence of T-regs for development. Moreover, since the T-regs were fixed, the development of TIL-10 cells required cell contact-dependent mechanisms with T-regs. However, it is unclear if T-regs induced IL-10 production in the CD25− T cells by acting indirectly through APCs or directly on the T cells. The suppressive properties of T-regs, at least in the in vitro systems, are strictly dependent on direct cellular contact (48). However, using an autoimmune-colitis murine model, it was shown that administration of anti-IL10 receptor mAb abrogates the colitis-preventing action of these T-regs (49). Moreover, if the suppressive properties of T-regs are dependent on direct cellular contact, it is difficult to explain how a subset of T cells that comprises only about 5–6% of total PBMC population can effectively mediate inhibition of other cells. Using in vitro model of mitogenic stimulation Jonuleit et al. (24) and Dieckmann et al. (31) have recently shown that T-regs might induce anergy and IL-10 production in naive T cells converting them into suppressor cells, a phenomenon termed ‘infectious tolerance’. Our results are in agreement with these reports and further demonstrate that ‘infectious tolerance’ could be an integral regulatory circuit that maintains peripheral tolerance to self-antigens in human transplant recipients. Although the predominant suppressive cytokine in the transwell experiments was found to be IL-10 alone, the role of membrane bound TGF-β in contact-dependent suppression of effector T cells cannot be ruled out.
It is now clear that human CD4+CD25+ T cells, although enriched in suppressive properties, represent a heterogenous group. Ongoing investigation to identify unique markers to more specifically define the suppressive subset in this population has so far been unsuccessful. In this regard, we have recently found that CD4+CD25+ T-regs from BOS are less efficient in suppressing conventional T cells to both allogeneic and mitogenic stimulation (Bharat A, Mohanakumar T; unpublished observations). Interestingly, the frequency of CDR45O+(memory/mature) CD4+CD25+ T cells is significantly lower in BOS+ve patients (50). Hence, we further hypothesize that the decline in the col-V specific TIL-10 during BOS is related to the loss of functional T-regs. This indirectly provides a potential for the expansion of Th1-autoreactivity. BOS is believed to affect the majority of LT patients surviving beyond 8–10 years post-transplantation. Contemporary immunosuppression has shown little improvement on its incidence and prognosis. Since the function and development of T-reg is dependent on IL-2 (48), calcineurin inhibitors like cylosporin A and tacrolimus might be detrimental to T-regs. In fact, cyclosporin A has been shown to induce autoimmunity (51–53). Hence, one plausible explanation for the decline of T-regs in transplant recipients could be related to the routine administration of immunosuppressive drugs. Therefore, identification and use of T-reg sparing immunosuppressive agents may better promote long-term allograft survival. Such studies are underway and have shown promising preliminary results with mycophenolate mofetil and rapamycin (54,55).
The data from these experiments indicate that LT patients have col-V reactive T cells that can be detected in the peripheral blood. The predominant col-V specific T cells produce IL-10 that suppresses the autoreactive Th1 cells, independent of direct cellular contact. Furthermore, the TIL-10 are capable of mediating bystander suppression, possibly including alloimmunity. T-regs are pivotal for the induction of these ‘suppressor’ TIL10 cells. During BOS, loss of T-regs leads to the failure of this regulatory circuit resulting in the expansion of Th1 cells. In summary, our studies demonstrate a natural T-reg-mediated regulatory pathway that could be responsible for preventing autoimmunity in human allograft recipients. The pathogenesis of chronic human lung allograft rejection is still unclear. Data provided in this report demonstrate that a derangement of this T-reg-mediated regulation could contribute to its development.
This work is supported by NIH HL56643 (TM). RF is recipient of ISHLT Research Fellowship Award.