CD200, a type 2 transmembrane molecule of the Ig supergene family, can induce immunosuppression in a number of biological systems, as well as promote increased graft acceptance, following binding to its receptors (CD200Rs). Skin and cardiac allograft acceptance are readily induced in transgenic mice overexpressing CD200 under control of a doxycycline-inducible promoter, both of which are associated with increased intragraft expression of mRNAs for a number of genes associated with altered T cell subset differentiation, including GATA-3, type 2 cytokines (IL-4, IL-13), GITR, and Foxp3. Interestingly, some 12–15 days after grafting, induction of transgenic CD200 expression can be stopped (by doxycycline withdrawal), without obvious significant effect on graft survival. However, neutralization of all CD200 expression (including endogenous CD200 expression) by anti-CD200 mAb caused graft loss, as did introduction of an acute inflammatory stimulus (LPS, 10 μg/mouse, delivered by i.p. injection). We conclude that even with apparently stably accepted tissue allografts, disruption of the immunoregulatory balance by an intense inflammatory stimulus can cause graft loss.
A number of approaches have been used to investigate the mechanisms responsible for induction of tolerance to tissue and/or organ allografts, including the use of microarray technology (1, 2, 3), subtraction hybridization approaches (4, 5), and model-driven studies using reagents directed at cells/molecules thought to be important in immunoregulation (6, 7, 8, 9, 10, 11). We have previously reported that improved acceptance of skin allografts in mice was associated with increased expression of a number of distinct mRNAs, one of which encoded CD200, a molecule expressed on the surface of dendritic cells DCs3 (4, 12); that using a soluble immunoadhesin (CD200Fc), in which the extracellular domain of CD200 was linked to a murine IgG2aFc region, or in mice which overexpression of a transgenic CD200 was under control of a doxycycline (Dox)-inducible promoter, inhibition of T cell allostimulation and type 1 cytokine production (IL-2, IFN-γ) occurred in vitro and in vivo (13, 14); and that CD200-mediated immunoregulation reflected signaling following an interaction with receptors, CD200Rs, on the surface of myeloid and or lymphoid cells (15, 16, 17), which could in turn favor development of “tolerogenic” DCs and/or regulatory T cells (Tregs) that were implicated in enhanced graft survival (18, 19). A large and growing body of work confirms the importance of Tregs in transplantation tolerance (20, 21) and the interrelatedness of this with DC function (22, 23, 24).
Unlike the many studies exploring mechanisms implicated in the induction of immunoregulation leading to longer graft survival, comparatively fewer reports exist addressing the factors important in maintenance of transplants following early induction of unresponsiveness. There are studies suggesting an important role for altered dynamics in cytokine production (25), particularly of IL-10 (26), for ongoing regulation by populations of Tregs (27), and again for a crucial role for the manner of Ag presentation in maintaining allograft tolerance (28, 29). Subtle changes in signaling through a number of costimulatory receptors, or regulatory ligands, have also been implicated in ongoing immunoregulation (30, 31). In the studies described below, we have made use of the Dox-inducible CD200-transgenic mice (CD200tg), as well as mAbs to CD200, to determine whether ongoing CD200-CD200R interactions are important in maintenance of tissue allograft survival in mice in which either skin or cardiac allograft transplant survival has initially been prolonged using conventional immunosuppression with low-dose rapamycin. A focused microarray analysis was incorporated into these studies to explore the correlation of changes seen in graft survival in different groups with altered expression of genes implicated in immunoregulation. Our data suggest that some (endogenous) CD200 expression is essential for maintenance of tissue allograft survival, and that this in turn is correlated with expression of several gene products previously reported to be involved in Treg induction/function.
Materials and Methods
Stock male C3H/HEJ and C57BL/6 mice were purchased from The Jackson Laboratory. Mice were housed five per cage under specific pathogen-free conditions and allowed food and water ad libitum. All mice were used at 8–12 wk of age. All animal experimentation was performed following guidelines of an accredited animal care committee (protocol no. AUP.1.5).
To replace the previously described Dox-inducible CD200 transgenic mouse line (14), a newer line was used incorporating a new transactivator, rtTA2S-M2, functioning to induce CD200 at a 10-fold lower Dox concentration than the previously used rtTA, with greater stability in eukaryotic cells and exhibiting essentially no background expression in the absence of Dox (32, 33). Homozygous rtTA2s-M2 CD200tg mice (on a C57BL/6 background) receiving Dox in their drinking water (1 μg/ml) for 7 days showed ubiquitous overexpression of CD200 (see Refs. 14 , 33 and Table I⇓) and suppressed allograft reactivity in vivo (see Figures below).
PE-labeled mAbs to mouse CD11b, CD205, F4/80, CD45, and CD3 were obtained from BD Pharmingen. FITC-conjugated rat anti-mouse CD200 was obtained from Serotec (distributed by Cedarlane Laboratories). Custom large-scale preparation of a previously described rat anti-mouse CD200 (10A5) was performed by Cedarlane Laboratories. Normal rat IgG2a was used as an isotype control for 10A5.
Cells (105) were incubated with PE- (or FITC-)coupled mAbs at 4°C for 60 min in PBS with 2% mouse serum to block Fcr binding. Cells were washed three times with PBS and analyzed in a Cytomics cytometer using Cytomics software (Beckman Coulter) as previously described (14, 33).
Preparation of cells and proliferation and cytotoxicity assays
These procedures generally follow previously established methods (13, 14, 15, 16). Single-cell spleen suspensions were prepared aseptically from pools of stock mice, and after centrifugation cells were resuspended in α-MEM supplemented with 2-ME and 10% FCS (αF10). In allogeneic mouse mixed leukocyte cultures (MLCs) used to induce proliferation and/or cytotoxicity (CTL), 1.5 × 106 C57BL/6 responder cells were stimulated with equal numbers of mitomycin C-treated (45 min at 37°C) C3H spleen stimulator cells in triplicate in αF10. For proliferation assays, 1 μCi of [3H]TdR was added to replicate wells at 72 h of culture and wells were harvested using a TopCount for analysis of [3H]TdR incorporation 14 h later. For cytotoxicity assays, cells from replicate wells were harvested and pooled at 5 days and titrated at different E:T ratios for killing (4 h at 37°C) of 51Cr-labeled 72 h Con A-activated C3H spleen target cells. In cultures used to assess the presence of regulatory cell populations, 1.5 × 106 responder splenocytes were used from a pool of three normal (nontransplanted) 8-wk-old C57BL/6 mice. CD4+ T cells isolated from transplanted mice for use as a regulatory pool were purified from spleen cell preparations by negative selection using a CD4+ T cell isolation kit according to the manufacturer’s recommendations (StemCell Technologies). Regulatory CD4+ cells (5 × 105) were added as described in the text, and CTL was assayed 5 days later.
Skin and cardiac grafts in mice
Groups of C57BL/6 control or rtTA2s-M2 CD200tg mice received 1-cm2 allogeneic (C3H) tail skin grafts on the shaved dermis as described elsewhere (34). All animals received low-dose rapamycin (0.5 mg/kg) at 48-h intervals for 12 days posttransplantation, a dose without significant effect on the function of Tregs (35) or cell subset distribution (R. M. Gorczynski, unpublished observations). Additionally, groups of mice were subdivided to receive either Dox (1 μg/ml) in the water supply or plain drinking water. Graft survival was followed daily by an investigator blinded to the different groups, and cells/tissue harvested from mice for further analysis were as described in individual studies below.
Cardiac transplants were performed as follows. Donor hearts were collected by dividing and excising the aorta and pulmonary artery. The grafts were transplanted into recipients by suturing donor aorta and donor pulmonary artery end to side to the recipient’s abdominal aorta and vena cava, respectively. Graft function was monitored every day by transabdominal palpation. Rejection was defined as a complete cessation of palpable beat (36). In some experiments LPS (Escherichia coli 0111:B4, 10 μg/mouse i.p.), purchased from Sigma-Aldrich, was used to provide an inflammatory stimulus.
Real-time PCR array analysis
Focused real-time PCR analysis was performed on whole RNA isolated from snap-frozen tissue harvested at various times from transplanted mice. A commercial kit purchased from SABiosciences was used to screen mouse genes implicated in T cell anergy and tolerance (RT2 Profiler PCR array (PAMM-074G), mouse; T cell anergy and immune tolerance PCR array: see www.sabiosciences.com/PCRArrayPlate.php). Paired samples were analyzed on a Roche LightCycler 480 with a 384-well block. All preliminary analytic screens with the commercial kit (detecting 84 pathway-specific genes) were performed in duplicate on samples from four mice per group. Those restricted genes reproducibly differentially expressed in experimental vs control groups (p < 0.05) were subsequently analyzed by real-time PCR using newly derived primer sequences (see below) specific for the relevant genes and fresh independent RNA samples from fou pairs of grafted mice (experimental vs control) (37).
Sequences were: Csf-1, forward primer, ATTGCCAAGGAGGTGTCAGAAC, reverse primer, AAAGGCAATCTGGCATGAAGTC; Fas, forward primer, AGTACATGGACAAGAACCATTATGCT, reverse primer, GGGTCAGGGTGCAGTTTGTT; Fos, forward primer, TCATCCTCCCGCTGCAGTAG, reverse primer, CGCAAAAGTCCTGTGTGTTGA; GITR, forward primer, TTTGGCTTCCGGTGTGTTG, reverse primer, GAACATGGTGAGAAATCCAAACTG; IL-12a, forward primer, CCCAAGGTCAGCGTTCCA, reverse primer, GGCAAGGGTGGCCAAAA; IL-17a, forward primer, GCTCCAGAAGGCCCTCAGA, reverse primer, CTTCCCAGATCACAGAGGGATATC; IL-31, forward primer, CGGTGCCCCAATATCGAA, reverse primer, GATGCCTGCTTTATGCTATAGTTGTT; Lep, forward primer, ATGTTCAAGCAGTGCCTATCCA, reverse primer, CCGACTGCGTGTGTGAAATG; Ptger2, forward primer, CCTGGCCATTATGACCATCAC, reverse primer, AGAGCTCGGAGGTCCCACTT; Stat6, forward primer, GGAATGGCACACCCTTTGAG, reverse primer, CCACTGGCTGCCCCAAT; Cma1, forward primer, CCGACACACTGCAGGAAGTA, reverse primer, TATCCCAGCACACAGCAGAG; Fas, forward primer, CCCCAGTACACCCTCTGAAAAA, reverse primer, CCATATGTGTCTTCCCATTCCA; Foxp3, forward primer, AGTCTGCAAGTGGCCTGGTT, reverse primer, GGGCCTTGCCTTTCTCATC; Gata3, forward primer, GCAGCCTGCTGGGAGGAT, reverse primer, TAGAGGTTGCCCCGCAGTT; Gzmb, forward primer, CACTCTTGACGCTGGGACCTA, reverse primer, TGATCTCCCCTGCCTTTGTC; Icam-1, forward primer, GTCCGCTGTGCTTTGAGAACT, reverse primer, GCAGAGGTCTCAGCTCCACACT; IL-4, forward primer, TCATCGGCATTTTGAACGAG, reverse primer, TTTGGCACATCCATCTCCG; INF-γ, forward primer, CCCATTCCTACTTCTCCCTCAA, reverse primer: GGAACGCACCTTTCTGGTTACA; Jak-1, forward primer, AATGTTCTCTATGAGGTCATGGTGACT, reverse primer, CAGTTTTTTCCGCTTCAGTTTATTT; and T-bet, forward primer, TGCGCCAGGAAGTTTCATTT, reverse primer, GGGCTGGTACTTGTGGAGAGACT.
In most experiments comparison between groups used ANOVA. For skin grafts, groups were compared using nonparametric tests (Mann-Whitney U test).
Increased skin and cardiac allograft survival in rtTA2s-M2 CD200tg vs control C57BL/6 mice
In a previous paper we reported that allogeneic skin graft survival in a heterozygous “first generation” CD200tg mice line was increased following induction of CD200 gene expression by Dox water (14). Despite this superior survival and diminished allosensitization in vivo, all grafts were rejected by about day 28 postgrafting. We explored survival of both cardiac and skin allografts from C3H mice in newly generated homozygous rtTA2s-M2 CD200tg animals, using in addition low-dose rapamycin (0.5 mg/kg at 36-h intervals for the first 12 days postgrafting). Data in Fig. 1⇓ show that long-term survival (>80 days) of both vascularized and nonvascularized grafts was achieved in transgenic mice overexpressing CD200 using this protocol (▵ in Fig. 1⇓). No increased survival was seen in nontransgenic mice or in transgenic mice not treated with Dox water (▴ in Fig. 1⇓).
When grafted mice were sacrificed at either day 20 or 80 postgrafting and spleen cells were assessed for responses to irradiated donor-specific (C3H) or third party (BALB/c) spleen cells in vitro (analyzing proliferation at 72 h or CTL induced at day 5 of culture), it was evident that long-term acceptance was associated with Ag-specific unresponsiveness to C3H stimulation (Fig. 2⇓). Moreover, this unresponsiveness was associated with the presence of CD4+ splenocytes able to inhibit specifically the immune response of fresh normal C57BL/6 splenocytes to stimulation with C3H Ag (Figure 3⇓).
Analysis of differential gene expression in transgenic mice with skin grafts
The data noted above support previous studies documenting the potential importance of overexpression of CD200 in prolongation of graft survival in mice. To explore evidence for more widespread altered gene expression in Dox-treated rtTA2s-M2 CD200tg mice receiving C3H skin grafts, we used a commercial focused RT-PCR kit as described in the Materials and Methods. Graft tissue was harvested at various times posttransplantation (four donors per time) from Dox-treated control or rtTA2s-M2 CD200tg mice receiving skin (Fig. 4⇓a) or cardiac (Fig. 4⇓b) grafts, mRNA was isolated and analyzed as per the manufacturer’s instructions. For those genes showing reproducibly altered over- or underexpression in transgenic relative to control mice at different times posttransplantation (p < 0.05 averaged over the four mice per group), confirmation of increased (or decreased) expression used real-time PCR with newly derived primer pairs, and four fresh pairs of tissue from grafted control and transgenic mice. Control groups, including samples in which no reverse transcription was performed, produced no detectable signal as expected. Data in Fig. 4⇓ show results for real-time PCR analysis of such transcripts in grafted mice at day 14 posttransplantation only.
Of particular note in terms of our understanding of the potential mechanisms involved in increased graft survival we noted altered expression of genes controlling expression of extracellular proteinases (Cma1), type 1, type 2, and Th17 cytokines (T-bet, GATA-3, IL-4, IL-12, IL-17, and IL-31), as well as regulatory T cell populations (Foxp3, GITR) and marked diminished expression of JAK-1, ICAM-1, Fos and Fas. Interestingly, differences existed between tissue harvested from cardiac vs skin-transplanted mice, which may be understood in terms of the greater ease with which graft acceptance is achieved for vascularized vs nonvascularized tissue (e.g., granzyme B and IL-12 levels were higher in skin transplanted mice, while Foxp3 levels were significantly higher in mice with cardiac grafts). The analysis of these gene expression profiles is assessed in more detail elsewhere (Yu, K., manuscript in preparation) and may infer the importance of different mechanisms for graft acceptance in skin vs cardiac allografts. Importantly, these preliminary data, taken together with the results in Fig. 3⇑, spurred a more detailed investigation into the long-term role of regulatory T cell populations in maintenance of graft survival.
Ongoing induction of CD200 transgene is not necessary for prolonged graft survival
To determine the importance of persistent overexpression of the CD200 transgene in long-term skin and cardiac allograft survival, we performed the following study. rtTA2s-M2 CD200tg mice received cardiac or skin allografts as described above. Nine mice per group were maintained on normal water (controls), while a further 30 per group received Dox water. For mice receiving cardiac grafts, at day 14 (12 of 12 with no grafts rejected) and day 30 (12 of 18 with grafts intact) 12 transgenic mice were subdivided into two groups of 6, only one of which continued to receive Dox water, while the other received only water to drink. Control studies confirmed that in the absence of Dox water, transgene expression was completely lost within 6 days (R. M. Gorczynski, unpublished data; see Ref. 33). For skin-grafted mice, 18 transgenic mice were used, and Dox-water was halted in 9 of the 18 mice (all with well-healed grafts) at day 14 post transplantation. Graft survival was monitored daily thereafter. Data for these studies (one of two experiments) are shown in Fig. 5⇓.
It is apparent from these data that regardless of the tissue transplant model used (skin or cardiac allograft), in CD200 overexpressing transgenic mice, once a finite period of time had elapsed following transplantation (14 days for both skin and cardiac grafts), ongoing induction of transgene expression was not necessary for maintenance of graft survival. Moreover, in independent studies in which we sacrificed four mice at day 26 from each of these three groups of skin-grafted rtTA2s-M2 CD200tg mice (no Dox water; Dox water continuously; Dox water for only the first 14 days) and used splenocytes from these mice for induction of CTL in vitro, or suppression of induction of CTL from control C57BL/6 splenocytes (see Fig. 3⇑), we found that even after withdrawal of Dox water, splenocytes from the rtTA2s-M2 CD200tg mice continued to contain a population of CD4+ regulatory cells with Ag-specific suppressive potential (see Fig. 6⇓). Similar data have been obtained in mice receiving cardiac grafts (R. M. Gorczynski, unpublished data).
Neutralization of expressed CD200 (either endogenous or transgenic) results in graft rejection
While the data in Figs. 5⇑ and 6⇑ indicated that ongoing overexpression of the transgenic CD200 in rtTA2s-M2 CD200tg mice was not needed for maintenance of graft survival, they did not address the potential role for endogenous CD200 in graft outcome in such mice. To determine the importance of endogenous expression of the CD200 in long-term skin graft survival in rtTA2s-M2 CD200tg mice, and the time period for which transgenic overexpression was needed to establish increased graft survival, we performed the following complementary studies. In the first, groups of rtTA2s-M2 CD200tg mice received C3H skin allografts as before. Eight of these, as control, received no Dox water (• in Fig. 7⇓). Forty rtTA2s-M2 CD200tg mice were placed on Dox water for the remainder of the study, and eight per group received i.v. infusions of anti-CD200mAb (10A5, 100 μg/mouse at 60-h intervals) until graft rejection, commencing on the day of transplantation (▵) or at days 10 (○), 15 (□) or 20 (⋄), while the final group of eight mice received isotype control normal rat IgG2a (▿). Skin graft survival for these mice (one of two such studies) is shown in Fig. 7⇓. In the second study, 24 rtTA2s-M2 CD200tg mice received C3H skin grafts and began treatment with Dox water from either the day of transplantation (○), or commencing at days 5 (▵) or 10 (⋄), with survival in this case as shown in Fig. 8⇓; again, a control group of rtTA2s-M2 CD200tg mice received no Dox water at any time (•).
There are a number of key features in these data. First, unlike the maintained graft survival seen in established grafts once Dox water was withdrawn (e.g., at day 14 in Fig. 5⇑), neutralization of all CD200 expression by anti-CD200 mAb led to loss of grafts regardless of time postengraftment (Fig. 7⇑). We conclude that data in Fig. 5⇑ should be interpreted as evidence that even endogenous expression of CD200 is sufficient to maintain grafts once established, but that high-level (CD200tg) expression is essential to induce prolonged survival. Second, when we investigated whether graft survival could be induced following delayed overexpression of the CD200 transgene (Fig. 8⇑), we found that delay to 10 days postengraftment produced no increased survival beyond controls (never receiving Dox-water) (see ⋄ in Fig. 8⇑). Even delay to 5 days post engraftment resulted in significantly inferior longer-term survival compared with mice receiving induction from the days of transplantation (compare ▵ and ○ in Fig. 8⇑).
Acute administration of endotoxin induces graft rejection in transplanted mice not receiving Dox water
Ongoing overexpression of the transgenic CD200 is not essential for prolonged skin graft survival in mice at >14 days posttransplantation (see Fig. 5⇑), although some endogenous expression is needed (Fig. 7⇑). We next asked whether the delicate balance in immunoregulation that maintained graft acceptance in transgenic mice in the absence of CD200 transgene expression could be easily overcome, leading to graft rejection, by a nonspecific inflammatory stimulus provided by endotoxin administration (10 μg og LPS per mouse, delivered i.p.). Thirty-six rtTA2s-M2 CD200tg mice received Dox water and C3H skin grafts as before. A control group of eight transgenic mice never received Dox water (• in Fig. 9⇓). At day 14, 18 mice (▵, ▴) were switched to receive plain water to drink, with 9 of these mice (▴) given LPS i.p. on day 18. The remaining 18 mice (□, ▪) were maintained on Dox water, but again 9 animals (▪) received LPS i.p. on day 18. Skin graft survival was followed in all mice, with typical data (pooled from two studies) shown in Fig. 9⇓.
It is apparent that when mice were maintained on Dox water throughout, challenge with LPS at day 18 produced no significant effect on longer term graft survival (▪). In contrast, when mice were taken off Dox water at day 14, while this had no effect on graft survival in the absence of additional perturbations (compare ▵ and □ in Fig. 9⇑, and data of Fig. 5⇑b), challenge with LPS at day 18 led to rapid graft rejection in all mice (see ▴ in Fig. 9⇑). In separate studies when we explored evidence for regulatory cell populations in these mice at day 24 posttransplant, cells taken from the transgenic mice off Dox water that had been challenged with LPS at day 18 also showed no evidence for CD4+ regulatory cells able to suppress CTL induction from a naive responder spleen cell pool in culture (see Fig. 10⇓).
An important immunoregulatory role for the molecule CD200 in control of allograft rejection, autoimmune arthritis and uveitis, tumor immunity, and allergy has been reported (13, 38, 39, 40, 41). This CD200-mediated suppression is mediated through interaction with a receptor, CD200R, on a target cell surface, and a family of CD200Rs has been described, with different members showing unique tissue distribution and function (16, 18, 42, 43, 44). The mechanisms by which CD200-CD200R interaction contributes to immunoregulation and prolongation of tissue allograft survival remain unclear, although we and others have reported significant suppression of inflammatory responses and the induction of regulatory T cells as being prominent among the likely candidates (45, 46, 47, 48, 49, 50). Other important issues that have yet to be investigated in any detail include the role for ongoing CD200 expression in conditions of chronic immunoregulation (as for instance in autoimmunity and/or allograft survival), and other changes in the gene expression profile that occur in association with altered CD200-CD200R interactions. The data in this manuscript represent early investigations into both of these issues.
A number of studies have focused on the importance of ongoing Ag exposure and/or of tolerogenic DCs in maintenance of allograft survival (22), with recent studies suggesting that indirect Ag presentation can break graft tolerance, while in tolerant animals, direct Ag presentation can suppress rejection (29). Some of the altered development of DCs is apparently a reflection of the altered cytokine milieu in the transplanted tissue that, along with CD200-CD200R interactions, can control the status of monocyte differentiation, promoting in vitro induction of Tregs and/or in vivo protection from autoimmune diseases (18). Multiple growth factors, including IL-10, TGFβ, G-CSF, and vasoactive intestinal peptide have been reported to modulate DC maturation and favor the differentiation of tolerogenic DCs (51).
Other data have focused on the importance of cytokines in the generation and maintenance of regulatory T cells (25). Thus, IL-4 prevented spontaneous apoptosis and the decline of Foxp3 mRNA, which occurred during culture of isolated Tregs, and Tregs exposed to IL-4 were found to express more CD25 and be more potent in suppressing the proliferation and IFN-γ production of naive CD4+ T cells when compared with Tregs cultured in medium alone (52). Independently, IL-10 was reported to be crucial for maintenance but not for developmental induction of peripheral T cell tolerance (26). Importantly, in the context of the transplant models studied here (and for extrapolation to clinical scenarios), rapamycin has been shown to enhance the percentage of CD4+CD25+Foxp3+ Tregs in the thymus and the periphery while keeping these cells functional, indicating that CD4+CD25+ Tregs are more resistant to rapamycin than other CD4+ T cells (35). Rapamycin is also implicated in generation of immunosuppressive Tregs with concomitant suppression of the differentiation of pathogenic Th17 cells (53).
Our data suggest an important role for overexpression of the immunoregulatory molecule CD200 in the induction of, but not necessarily the maintenance phase of, allograft tolerance, although some, presumably endogenous, CD200 expression was still crucial both for persistent graft survival and Treg function (Figs. 1⇑, 5⇑, 7⇑, and 8⇑). This is consistent with other reports investigating the role of other prominent immunoregulatory cell surface molecules, CTLA4 and programmed cell death-1 (PD-1), in allograft tolerance (27, 31). Early studies in an autoimmune disease model suggested that CTLA4 blockade abrogated the induction but not the maintenance phase of acquired thymic tolerance induced by intrathymic injection of myelin Ags (54). More recently, this issue was revisited in a model that used a combination of donor-specific transfusion, anti-CD45RB, and anti-CD154 to achieve >90-day survival of BALB/c skin allografts in C57BL/6 mice. In this model, engraftment remained completely dependent on CTLA4 signaling, even after grafts are healed, suggesting that prolonged engraftment was not simply attributable to depletion of responsive T cells but was actively maintained by regulation. Both CD4 and CD8 regulatory cells were required and could transfer donor-specific tolerance to naive recipients, although interestingly gradual graft loss eventually occurred (median survival time, 140 days), which the authors interpreted to reflect the capacity of alloreactive cells emerging from the thymus to overwhelm regulatory capacity of other (T) cell populations (55). Using a different approach involving a novel anti-CTLA4 Ab that interfered with CTLA4 signaling but did not antagonize B7 binding to CTLA4 or affect its ability to out-compete CD28 for B7 binding, another group concluded that competition for B7 may be important in induction of tolerance, but signaling through CTLA4 was more important in maintaining a tolerant phenotype (30). While blockade of the PD-1/PD-L1 pathway abrogates Treg-mediated immunoregulation, consistent with the idea that the PD-1/PD-L1 pathway is required for Treg suppression of alloreactivity (56), the mechanisms whereby PD-1 and its ligands regulate tolerance induction/maintenance may not be the same. Thus, PD-1 ligands on APCs have been reported to inhibit autoreactive T cells to induce peripheral tolerance, whereas those on parenchymal cells prevent tissue destruction by acting on effector T cells to maintain tolerance (57). The issue concerning the action of Tregs in preventing transplant rejection and maintaining a dominant tolerance state may be even more complex, since there is growing evidence that the target of Tregs may not even be effector T cells themselves. Several studies suggest that Treg cells may induce changes in the target tissue, promoting a state of “immune privilege” where protective genes such as heme oxygenase-1 (HO-1) and IDO play an important role (28). Note that CD200-CD200R immunoregulation has also been reported to involve perturbation of this latter pathway (46, 58). Importantly, we consistently observed evidence for populations of regulatory cells in all mice with long-term surviving grafts (both cardiac and skin allografts), regardless of expression of the CD200 transgene and even following challenge of mice with a nonspecific inflammatory stimulus (LPS) (Figs. 2⇑, 3⇑, 6⇑, and 10⇑). However, in mice challenged with LPS in the absence of the antiinflammatory effect afforded by CD200 overexpression, both graft survival and evidence for regulatory cell populations were lost (Fig. 10⇑), as indeed was the case when all CD200 expression, including endogenous levels, was neutralized by anti-CD200 (Fig. 7⇑).
Other groups have focused on whether a unique gene signature can be described that is characteristic of a tolerant vs actively rejected state in transplant recipients. Reduced costimulatory signaling, immune quiescence, apoptosis, and altered memory T cell responses are important in longer term acceptance in a renal transplant population compared with a gene signature reflecting DC activation in acute rejection (59, 60). An exhaustive microarray study to explore a signature for the “tolerant state” was performed in a mouse TCR transgenic model of experimental autoimmune encephalomyelitis using intranasal administration of the N-terminal peptide of myelin basic protein, in which tolerance was attributable to a population of IL-10-secreting regulatory T cells. A number of surprising results were reported, including evidence that overexpression of the Th1 determining gene T-bet was seen in Tregs, and that granzyme B, thought to be functionally relevant for CTL effectors, was also reported in Tregs (61). In a commentary on this study, Cobbold concluded that there was in fact considerable overlap in gene expression profile between multiple populations of Tregs, and even Th1 and Th2 cells, and that regulatory activity may be a common feature of all activated T cells, with the ultimate defining principle of a Treg being the lack of effector functions as a result of either partial or incomplete differentiation to either Th1 or Th2, rather than a positive function attributable to Tregs per se (62). Our preliminary data concerning the altered pattern of gene expression in tissue from skin-grafted mice (vs controls), for which concomitant graft survival and cellular immunological studies were available, have also provided some surprising findings (Fig. 4⇑a). Increased T-bet, granzyme B, leptin, and IFN-γ expression were observed; however, consistent with the idea that Treg function (and type 2 cytokines) may be implicated in prolonged survival (12), GATA-3, IL-4, GITR, and Foxp3 expression were also up-regulated. Interestingly, IFN-γ has recently been reported to induce suppressor of cytokine signaling-1 and up-regulation of Foxp3 and CD62L (63). Even more intriguingly, it was clear that the altered gene expression seen in skin-grafted CD200 transgenic mice differed in many instances from the gene expression profile in cardiac grafts in similar animals (Fig. 4⇑b), with notable declines in expression of IL-17, granzyme B, and IFN-γ in the cardiac grafts. Further studies are in progress to explore the implications of these data to the differences in immuneoregulatory milieu found following successful cardiac vs skin allotransplantation (Yu, K., manuscript in preparation).
In sum, our results add important new information concerning the role of expression of the immunoregulatory molecule CD200 in both the induction and maintenance phases of tolerance to tissue allografts, both vascularized (cardiac) and nonvascularized (skin). Our data further strengthen support for the hypothesis forwarded by ourselves and others (19, 27, 62, 64) that a key component of the immunoregulation responsible for both induction and maintenance of prolonged graft survival in these.
The authors have no financial conflicts of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
↵1 This work was supported by a grant from the Heart and Stroke Foundation.
↵2 Address correspondence and reprint requests to: Dr. Reginald M. Gorczynski, The Toronto Hospital, University Health Network, 2-805, MaRS Tower, 101 College Street, Toronto, Ontario M5G 1L7, Canada. E-mail address:
↵3 Abbreviations used in this paper: DC, dendritic cell; Dox, doxycycline; MLC, mixed leukocyte culture; PD-1, programmed cell death-1; Treg, regulatory T cell.
- Received January 21, 2009.
- Accepted May 27, 2009.
- Copyright © 2009 by The American Association of Immunologists, Inc.