| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
review-article |
University Medicine of Berlin,4 Charité Centrum 12, Internal Medicine and Dermatology, Campus Virchow, 13353 Berlin, Germany
Department of Biological Sciences,5 University of Essex, Wivenhoe Park, Colchester C04 3SQ, United Kingdom
Department of Gynaecology and Obstetrics,6 University of Würzburg, D-97080, Würzburg, Germany
Department of Microbiology,7 Parasitology and Immunology, School of Medicine, University of Buenos Aires, C1121ABG Buenos Aires, Argentina
Department of Physiology and Immunology,8 Medical Faculty, University of Rijeka, 51000 Rijeka, Croatia
Thomas E. Starzl Transplantation Institute and Department of Surgery,9 University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania 15213
ABSTRACT
Pregnancy is a unique event in which a fetus, despite being genetically and immunologically different from the mother (a hemi-allograft), develops in the uterus. Successful pregnancy implies avoidance of rejection by the maternal immune system. Fetal and maternal immune cells come into direct contact at the decidua, which is a highly specialized mucous membrane that plays a key role in fetal tolerance. Uterine dendritic cells (DC) within the decidua have been implicated in pregnancy maintenance. DC serve as antigen-presenting cells with the unique ability to induce primary immune responses. Just as lymphocytes comprise different subsets, DC subsets have been identified that differentially control lymphocyte function. DC may also act to induce immunologic tolerance and regulation of T cell-mediated immunity. Current understanding of DC immunobiology within the context of mammalian fetal-maternal tolerance is reviewed and discussed herein.
decidua, dendritic cells, placenta, tolerance
The question as to why the placenta, despite the expression of paternally derived fetal alloantigens, is accepted by the mother remains to be fully explained, although several immune tolerance mechanisms have been postulated [1–3].
Dendritic cells (DC) represent a population of cells with relevance to mucosal sites, such as the skin, airways, gut, and decidua. This is because DC have the unique property of being able to induce both antigen-specific activation and suppression during the immune response. DC may also play a role in tolerance induction through the generation of T cells with regulatory properties, which in turn limit the proliferation of effector T cells [4, 5] or delete antigen-specific T cells [6, 7].
In the present review, we will discuss the role of DC subsets and their activation signals in the regulation of fetal-maternal tolerance. In this context, we will highlight the link between DC and the cellular and molecular network components involved in the biology of pregnancy, namely T cells, cytokines, and hormones. These endocrine and immune signals ultimately determine the maturation, differentiation, and activation status of DC. In turn, these changes are responsible for the dual capacity of DC to trigger an immune response that is associated with fetal rejection and to promote cell tolerance at the fetal-maternal interface.
Fetal allograft hypothesis. The fetal-maternal relationship is a unique immunologic phenomenon. Fetal tissues express alloantigens that are encoded by polymorphic MHC genes inherited from the father. However, in a normal pregnancy, although the mother can raise antibodies against parental gene products, typically HLA allele-specific antibodies, these do not elicit fetal rejection.
Initial studies in pregnancy invoking the classical allograft rejection theories implicated CD4 and CD8 T cells. This concept has been expanded by the notion that NK cells are also involved in rejection [8]. In addition, the role of 
T-cell recognition of trophoblasts has been documented in mice [9] and humans [10]. Moreover, NKT cells, most likely with TCR, have been implicated in fetal rejection [11], as have NK
ß T cells [12]. CD4+ T cells that recognize paternal antigens but not trophoblasts are presented by antigen-presenting cells (APC) in decidual tissues [13, 14]. These observations suggest that the properties of APC in the uterine lining are important in modulating local immunity and tolerance. No single mechanism to explain the so-called fetal-maternal immunologic paradox has been identified. As in tissue allografts, interactions between the adaptive and innate immune systems, at both the cellular and molecular levels and including cytokines, are required to regulate pregnancy to term.
The Th1 to Th2 ratio as a paradigm of fetal-maternal tolerance. The maternal balance between Th1 and Th2-Th3 cytokines contributes to the success of pregnancy [15, 16]. It is now accepted that during pregnancy, local changes in the balance between Th1/Th2-Th3 cytokines occur within the uterus and fetal-placental unit. Cytokine shifts influence trophoblast cells and uterine immune cells (macrophages, NK cells, and lymphocytes), which contribute to implantation of the embryo, development of the placenta, and survival of the fetus to term. The predominant cytokines at each stage of gestation function both to limit maternal immune rejection of the semi-allogeneic fetus, especially at the fetal-maternal interface, and to facilitate the ongoing physiological processes within the maternal reproductive tract. Furthermore, in experimental models, cytokine imbalances with inappropriate activation of macrophages and NK cells have been shown to be detrimental to fetal survival [2].
We have previously reported [17] on the decisive roles played by IL10 and IL12 in driving the acceptance or rejection of the fetus. A strong, pregnancy-protective role for the Th2 cytokine IL10 is supported by data from mouse models, which show that while purified exogenous IL10 administered during pregnancy decreases resorption rates [18], injection of anti-IL10 antibody enhances these rates. In support of these results, injection of IL12 during early gestation has been found to induce abortion and increase the levels of the proinflammatory cytokines TNF and IFNG at the fetal-maternal interface [19].
Special Role of DC at Mucosal Interfaces
DC are particularly responsive to signals received from the tissue microenvironment. In mucosal tissues, DC can acquire unique features or phenotypes that are necessary for the regulation of specific local immune functions, which include tolerance to food antigens in the gastrointestinal tract, to respiratory antigens in the airways, and protection against pathogenic organisms [20]. The phenotypes of DC are influenced by cytokines and anti-inflammatory mediators produced by epithelial cells, either under physiological conditions or in response to microbial signals [20].
Previously, we have proposed [17] that decidual DC are the "gatekeepers" of the uterine mucosal immune system, inducing tolerance under normal physiological conditions. DC are also sufficiently responsive to inflammatory stimuli to allow T-cell priming and protective immunity when necessary. It is well known that exposure to agents such as prostaglandin E2 (PGE2) [21] polarizes the maturation of myeloid DC into Th2-promoting DC, while other cytokines, such as transforming growth factor ß (TGFB), promote tolerogenic DC [22]. Both PGE2 and TGFB are present at the fetal-maternal interface in normal pregnancy and may be utilized by DC to regulate placental health (Fig. 1). However, the assumption that decidual DC have similar functions to mucosal DC needs to be further investigated.
|
Tolerogenic DC: Different Maturation States or Different Subsets?
The induction of tolerance by DC appears to contradict the immunostimulatory function of these cells. However, a role for DC in the induction of peripheral tolerance is supported by several studies [23]. Two general hypotheses have been proposed to explain how DC might maintain peripheral tolerance. The first hypothesis is based on studies performed by Dhodapkar et al. [24] and Steinman et al. [25], which support the concept that antigen presentation by immature DC induces tolerance, whereas antigen presentation by mature DC induces immunity. This paradigm has been questioned, since mature DC have also been found to induce CD4 T-cell tolerance [26, 27]. Current data suggest that steady-state DC are sufficiently mature to express intermediate levels of MHC class II and costimulatory molecules but are not yet activated to the extent that they can induce an immune response [27–29]. Immunity occurs only in the context of infection-associated signals, which induce full maturation of DC.
The second hypothesis suggests that unique DC subsets are responsible for the induction of tolerance [30]. For example, murine CD8A+ DC have been shown to act as a specialized subset of tolerogenic or regulatory DC in experimental systems of allograft rejection and in vivo studies [31, 32].
In support of the subset hypothesis, a DC subset with the ability to prime type 1 regulatory T (Treg) cells has recently been described. This subset produces IL10 upon stimulation, exhibits a stable immature phenotype (CD45RBhigh ITGAXlow MHC class IIintermediate CD86low) and plasmacytoid morphology [33, 34].
Immature DC express CD200R2, type 2 receptors for CD200, a glycoprotein with pregnancy-protective function, which is expressed on trophoblast cells and in certain areas of the deciduas on days 8–12 of pregnancy [35]. In DBA/2J-mated CBA/J female mice [18] (a murine model in which increased abortion rates can be induced by application of lipopolysaccharide, Th1 cytokines or stress exposure), blocking CD200 greatly increases fetal rejection and conversely, administering CD200Fc or CD200+ splenocytes from BALB/c mice reduces fetal rejection. Moreover, CD200 triggering of DC induces a CD4+IL2RA+ Treg population with suppressive activity, as demonstrated by increased FOXP3 expression and inhibition of mixed leukocyte reactions [36]. In line with this finding, DBA/2J-mated CBA/J females that are exposed to sound stress show elevated percentages of resident decidual DC that express moderate levels of MHC II, CD80/86, and ICAM1 on gestation Day 7.5 [17].
In humans, the number of mature CD83+ DC increases in the uterine mucosa during the late secretory phase of the menstrual cycle in comparison to other endometrial phases [37]. This is in contrast to first trimester decidua, in which the vast majority of DC express CD209 (previously known as DCSIGN), which is a marker for immature or inactivated DC [38, 39]. Interestingly, freshly isolated decidual CD209+ DC take up antigen efficiently but are unable to stimulate naïve allogeneic T cells. However, when stimulated with an inflammatory cytokine cocktail, these cells become mature IL2RA+/CD83+ DC with potent stimulatory capacity for allogeneic T cells and are similar to those DC that are sparsely distributed in early pregnancy decidua [39, 40].
Therefore, it can be hypothesized that the maturation state of DC plays a role in the etiology of spontaneous pregnancy failure. Indeed, the high abortion rates observed in the CBA/J x DBA/2J murine model are related to DC that exhibit a more mature phenotype (higher expression of MHC class II, CD80, and ICAM1), as compared to low-abortion-rate mice [41]. Deciduas from women with recurrent miscarriage at 8 wk of gestation have been shown to contain significantly more CD83+ DC than gestational age-matched normal controls [42]. The presence of mature DC correlates with higher abortion rates. It is conceivable that these activated DC can overcome the tolerance to paternal antigens and induce fetal rejection (Fig. 2).
|
Dendritic Cells Within the Decidual Cellular Environment
Cross-talk between DC and NK cells: two players in innate immunity. Recent data have highlighted the cooperation between DC and NK cells in the control/switch of innate immunity. NCAM1++FCGR3– (formerly CD56++CD16–) uterine NK (uNK) cells represent the main population of lymphoid cells in the uterine mucosa during early pregnancy [43]. However, their number decreases from midgestation onwards, and they are almost absent at term. Although the interaction between trophoblast MHC class I molecules and uNK cells is likely to provide the pivotal control for successful implantation in humans [44], it is unclear how this process is mediated. It is likely that uNK cells play a dual role in reproduction: to monitor mucosal integrity throughout the menstrual cycle, and to control trophoblast invasion during pregnancy [43]. In contrast, during mouse pregnancy, uNK cells accumulate on the mesometrial side of the placenta and play a central role in decidualization [45]. Indeed, Il15tm1Imx/tm1Imx mice, which lack uNK cells, present decidual abnormalities, which include thickening of the arterial walls with luminal narrowing and a hypocellular decidua basalis. However, uNK cells are not required for successful pregnancy, since IL15–/– mice have normal implantation and nonabortive pregnancies [46].
Little is known about the potential interactions between DC and NK cells under homeostatic resting conditions. However, it has been shown that interaction between immature DC and activated NK cells results in either DC maturation or cell death [47]. The mechanisms that determine the death or maturation outcome depend on a dynamic interplay between the ratio of DC:NK cells and the DC maturation state [48]. Gerosa et al. [49] have shown that the cross-talk between immature DC and resting NK cells leads to cell activation only in the presence of microbial stimuli. Furthermore, only under conditions of low DC:NK ratios do DC-NK cell interactions result in NK activation [50]. The cytokine pattern of DC activation after interaction with uNK cells in response to antigenic stimuli could also influence the polarization of T-cell responses. There may be physiopathologic conditions (i.e., spontaneous abortion) under which disequilibrium of the ratio between uNK cells and DC results in aberrant cell activation and perturbation of the course of pregnancy (Fig. 2). In particular, it has been reported that the vast majority of decidual immature DC are in close contact with uNK cells, whereas the small number of mature decidual DC are clustered with CD3+ T cells, but not with uNK cells. Moreover, in vitro studies have shown that decidual CD83+ DC cocultured with NCAM1+ autologous uNK cells stimulate the proliferation of NCAM1+ cells in a DC-dependent manner, increase KLRC1 inhibitory receptor, and decrease KLRC2- and KLRK1-activating receptor expression [51].
DC-derived Cytokines and Pregnancy
Various types of DC-derived molecules with the ability to polarize Th-cell responses have been identified. DC from IL12-deficient mice fail to induce Th1 responses, which suggests a critical role for IL12 family members (e.g., IL12, IL23, and IL27) in DC-induced Th1 responses [52, 53]. Consistent with this notion, murine CD8A+ DC, but not CD8A– DC, can be induced to produce large amounts of IL12 and IFNG [52, 54, 55]. These findings indicate that murine DC subsets that differ in CD8A expression can prime naïve Th cells to produce either Th1 or Th2 cytokines, respectively [52, 56]. The mechanism by which CD8A– DC induce Th2 cytokines has not been established, although IL13 [57] and IL10 [56] are suitable candidates.
The activities of cytokines, such as IL10 and TGFB, as T-cell polarizing factors are well-documented [58]. In accordance, we have noted that the intracellular level of IL10 in uterine DC is significantly higher than the level of IL12 during most of the gestation period [17]. However, there is a transient increase in the incidence of uterine CD8A+ DC in early gestation, namely on Day 5.5, concomitant with a decrease in the relative number of IL10-producing DC. Our current paradigm is that DC have remarkable plasticity in their abilities to induce Th1 or Th2 cytokines and to exhibit temporal regulation and subtle discrimination of antigen stimuli [59]. This prompted us to suggest that in early gestation, fetal antigens may be stimulating CD8A+ uterine DC to produce Th1 cytokines, thus contributing to the inflammatory milieu characteristic of the implantation period. At later stages, a regulatory IL10 cytokine pattern predominates, which mediates the communication between embryonic cells and maternal uterine cells and the maintenance of pregnancy [60–62]. Moreover, recent experiments have shown that uterine DC from mice with high abortion rates display increased expression of CD8A+ and increased IL12:IL10 ratios compared to mice with a low incidence of fetal rejection [41].
Compared with mouse DC subtypes, human DC exhibit considerable plasticity in response to cytokines and pathogens. This feature makes it difficult to assign a function to each DC lineage. In spite of this, recent findings indicate that myeloid DC produce high levels of IL12 when stimulated with TNF or CD40 ligand (CD40LG) and stimulate a Th1-skewed cytokine profile in T cells, whereas fully mature plasmacytoid DC prime T cells towards a Th2 cytokine profile [63]. Even though protection from fetal rejection has been traditionally linked to a Th2-type immune response, myeloid DC predominate in human and mouse early pregnancy decidua [17, 38, 40]. Although these findings apparently contradict the Th1/Th2 paradigm, they are in line with the new importance given to the state of activation over the lineage of resident DC. To date, only one study has examined the expression of IL12 by decidual DC, albeit in relation to peripheral blood DC in early human pregnancy. The results show that IL12-producing myeloid DC (ITGAX+) or lymphoid DC (IL3RA+) numbers are significantly reduced in the decidua compared to the peripheral blood. In addition, the level of IL12 secreted by decidual myeloid cells stimulated with lipopolysaccharide (LPS) or CD40LG was found to be significantly lower than that secreted by peripheral blood counterparts [40]. Thus, decidual DC can regulate the balance between Th1 and Th2 cytokines, resulting to the maintenance of pregnancy.
Regulatory T cells. Treg block the functions of effector CD4+ and CD8+ T cells. As a result, tolerance is achieved [64]. Naturally occurring (i.e., thymically generated) CD4+ IL2RA+ Treg appear as mediators to prevent autoimmune responses to self antigens, whereas Treg induced by antigenic stimulation in the periphery (also referred to as ‘Tr1’ cells or ‘adaptive regulatory cells’) appear to be involved in the maintenance of peripheral tolerance at mucosal surfaces [65, 66]. Mucosal tissues are faced with the constant challenge of maintaining nonresponsiveness to nonself peptides (antigens), which means that the immune system must employ potent regulatory mechanisms. Several studies have demonstrated that Treg can prevent graft rejection [67–69]. A similar situation seems to occur during pregnancy, whereby the maternal immune system tolerates paternal antigens expressed by the fetus. Recently, it has been demonstrated that normal human pregnancy is associated with an increase in the blood Treg cell population. Analysis by stage of pregnancy has shown a significant increase from the nonpregnant controls to the first trimester and from the first trimester to the second trimester, with no significant change from the second to third trimester. The proportion of CD4+ IL2RA+ T cells was significantly lower 6–8 wk following delivery than during the third trimester but remained higher than in nonpregnant controls [70]. In addition, decidual CD4+ IL2RA+ T cells represented 14% of all CD4+ T cells in the third trimester [71]. DC production of IL10 may be critical for the generation of Treg [26, 33]. The presence of decidua-resident, semimature DC subsets with intermediate expression of MHC and secretion of IL10 [17, 38, 40] raises the question as to whether the tolerogenic function of DC is achieved through the priming of Treg in the decidua. Moreover, Gorczynski et al. [72] have shown that triggering CD200R2, which is a pregnancy protective receptor that is present on immature DC, leads to the generation of CD4+IL2RA+FOXP3+ Treg that have the ability to suppress mixed lymphocyte reactions and allograft rejection.
Other regulatory T cells.
In addition to CD4+ Treg, CD8+ T cells may play a role in oral tolerance [73]. Furthermore, CD8+CD28–FOXP3+ T-suppressor cells have recently been reported to induce tolerogenic endothelial cells through the induction of inhibitory receptors and the down-regulation of costimulatory and adhesion molecules [74]. In murine pregnancy, a CD8A+ cell subpopulation that prevents abortion is present between gestation Days 4.5 and 8.5 [2, 75]. These CD8A+ cells, which may coexpress the TCR, appear to play a crucial role in the regulation of the Th1/Th2 balance [2, 76]. Our studies suggest that during stress-challenged pregnancy, there is a decrease in uterine CD8A+ cells. Since we observed a low fetal rejection index when CD8A+ cells were restored by dydrogesterone (a progesterone derivate product), it appears that these cells are highly regulated by the adequate levels of progesterone at the fetal-maternal interface [77]. Moreover, depletion of CD8A+ cells abrogated the protective effect of dydrogesterone in mice undergoing stress-challenged pregnancy. Although they have not been classified as "professional" regulatory cells, these uterine CD8A+ cells may be involved in the suppression of immune responses.
Tryptophan 2,3-dioxygenase (TDO2) and indoleamine 2,3-dioxygenase (INDO). Tryptophan catabolism is one of the many mechanisms involved in tolerance [78]. This essential amino acid is used by vertebrates in protein synthesis and is a precursor to serotonin, melatonin, kynurenine, and quinolinic acid. Moreover, some tryptophan catabolites induce apoptosis of T cells [79]. Expression of INDO is induced by IFNG in many tissues and in certain types of activated macrophages and DC, suggesting a role for INDO in the regulation of immune responses [80].
Tatsumi et al. [81] have demonstrated that TDO2 is induced in endometrial stromal cells concomitant with decidualization, which suggests its involvement in the implantation process by regulating the tryptophan level at the implantation site. On the other hand, INDO is expressed in the syncytial trophoblast and defends the conceptus against rejection by reducing the tryptophan level and suppressing T-cell activity. Several animal studies that have investigated the role of INDO in murine pregnancy have given contradictory results. Pharmacologic inhibition of INDO activity results in rejection of allogeneic fetuses by the maternal immune system [13]. However, Indotm1Alm/tm1Alm mice do not exhibit fetal rejection in allogeneic matings, which suggests that INDO is not an essential mechanism for tolerance in the context of pregnancy [82]. Moreover, Mackler et al. [83] have shown that INDO is not expressed in murine trophoblasts but it is expressed in stromal cells of the decidua basalis and metrial gland. In contrast, Baban et al. [82] have shown that INDO expression is restricted to the perinuclear regions of primary trophoblast giant cells at the middle stage of mouse gestation. Further investigations are needed to clarify the role of INDO in pregnancy maintenance.
Influence of the Endocrine System During Pregnancy
Dendritic cells and steroid hormones. An excellent body of literature exists regarding the influence of steroid hormone pathways on immune cell populations [84, 85]. Progesterone probably contributes to the natural suppression of cell-mediated immunity that accompanies pregnancy and allows the fetus to be tolerated during gestation. Progesterone acts through two isoforms of the progesterone receptor (PGR) [86]. Until recently, the consensus has been that PGR is induced by estrogen, and that many of the effects of progesterone can be attributed to the combined effects of estrogen and progesterone. Interestingly, it has been demonstrated that progesterone is essential for the induction of uterine decidualization, since this process fails to occur in mice that lack the Pgr gene [87, 88]. Indeed, Liu et al. [89] have reported that estrogen decreases the production of proinflammatory cytokines TNF, IFNG, and IL12 in mature DC. High estrogen levels at the fetal-maternal interface could influence uterine DC to reduce inflammatory cytokine production.
Recent studies have implicated progesterone in the modulation of DC differentiation, maturation, and function. Liang et al. [90] have shown that progesterone inhibits immune responses by producing more immature bone marrow (BM)-derived DC in mice. The addition of progesterone to BM-derived DC cultures results in significantly increased IL10 production and decreased TNF production. In addition, Butts et al. [91] have found that progesterone at concentrations similar to those occurring during the menstrual cycle or pregnancy has a crucial effect on rat BM-derived mature DC by modulating the ability of these cells to function as stimulators of proinflammatory responses. In vitro studies have shown that DC derived from the decidua of pregnant women increase their CD83 and HLADR expression levels when treated with progesterone at levels similar to those present during early pregnancy [92].
Vitamin D and PGE2: novel modulators?
Experimental evidence suggests that vitamin D may play a role in maternal-conceptus cross-talk. Besides its classical activity, the active form of vitamin D, 1,25-dihydroxyvitamin D3 or 1,25(OH)2D3 has well-known immunomodulatory and antiproliferative properties on human uNK cells [93]. Female fertility seems to be markedly reduced in vitamin D-deficient murine models. Both decidua and trophoblast cells express vitamin D receptor (VDR), as well as CYP27B1 (previously known as 25(OH)D 1-hydroxylase), which suggests an autocrine or paracrine role for 1,25(OH)2D3 within these tissues [94]. In particular, Zehnder et al. [95] have postulated that local activation of vitamin D influences implantation, either through the established immunomodulatory effects of 1,25(OH)2D3 or via the regulation of specific target genes associated with implantation. Indeed, 1,25(OH)2D3 regulates developmental target genes, such as the homeobox (Hox) genes. In particular, the Hoxa10 gene has been shown to be intimately associated with implantation [96].
Another immunomodulatory molecule that is present at the fetal-maternal interface is PGE2. PGE2 suppresses a number of T-cell reactions (e.g., cellular cytotoxicity and T-cell and killer-cell activity) and promotes the development of DC with low capacities to produce IL12, which supports the generation of Th2 cells. Harizi et al. [97] have demonstrated that exogenously added or endogenously released PGE2 acts on DC by inducing endogenous IL10, which in turn suppresses IL12 production and alters antigen presentation through the inhibition of MHC class II protein expression. As vitamin D and PGE2 are both present in decidua, these molecules may inhibit IL12 production by the uterine DC, thereby shifting the cytokine balance to a tolerogenic state at the fetal-maternal interface.
DC and Promotion of Transplant Tolerance
Evidence implicating DC as the main immunological stimulus for graft rejection has been provided by the renal allograft "parking" experiments of Lechler and Batchelor [98]. Subsequently, Larsen et al. [99] were the first to show that DC trafficked from heart allografts to recipient spleens, which in nonimmunosuppressed hosts is associated with rejection, concomitant with the disappearance of the donor DC. Studies conducted by Lu et al. [100] have revealed that donor-derived myeloid DC persist in and can be propagated from the BM of mice that accepted fully mismatched (liver) allografts without immunosuppressive therapy. This cannot be achieved with animals that acutely reject heart grafts from the same donor strain [100], unless both donor BM cells and immunosuppressive therapy are administered [101]. Further studies have shown that immature donor myeloid DC, which are deficient in surface costimulatory molecules, can induce alloantigen-specific T-cell hyporesponsiveness in vitro [102]. More significantly, several groups have shown that when administered before, during or even after transplantation, immature donor DC prolong MHC-mismatched allograft (including skin graft) survival. In some instances, indefinite donor-specific graft survival is achieved [103–110]. Recently, it has been shown that TGFB-treated (‘alternatively-activated’) IL10+ DC resist phenotypic and functional maturation in response to TLR4 ligation and induce long-term heart allograft survival when coadministered with a single dose of CTLA4-Ig. Graft survival is associated with Treg infiltration and the absence of vasculopathy [111]. There is also evidence that persistence of donor-derived DC may be important for the long-term maintenance of transplantation tolerance, as well as for the prevention of chronic rejection of heart allografts [112].
Collectively, these findings suggest that donor-derived DC, most likely but not exclusively at an immature/progenitor stage, can subvert recipient T-cell responses to alloantigens, through as yet ill-defined mechanisms. Although comparatively little research has been performed using syngeneic/autologous DC to subvert the indirect pathway, recent animal studies support the feasibility of this approach. Thus, Lechler's group [109] has shown that indefinite organ graft survival can be achieved in the absence of chronic rejection in rats that are receiving pharmacologically modified myeloid DC that express donor and recipient MHC antigens and only a short course of peri-operative cyclosporine (to control the direct pathway). Significantly, this effect is associated with the induction of indirect-pathway Treg. It appears that antigen-specific Treg can be induced under systemic tolerogenic regimens by plasmacytoid DC that acquire alloantigen in the graft and traffic to secondary lymphoid tissues [113].
Understanding the immunological paradox of maternal tolerance towards the fetal allograft is of great importance. The complexity of the uterine network clearly indicates that successful pregnancy does not rely entirely on immunological factors. DC appear to be key regulators of pregnancy, involving both innate and adaptive mechanisms. Decidual DC likely function to co-ordinate the spatial and temporal immunological shifts necessary for implantation and the progression of pregnancy.
FOOTNOTES
3These authors contributed equally to this work. 
1Supported by grants from Charité (to S.M.B. and P.C.A.), DFG (KFO-124/2 to U.K.), and the Croatian Ministry of Science (0376-0377 to D.R.). S.M.B. is supported by Charité (Habilitations-Stipendium), and C.A.S. and G.B. are supported by fellowships from CONICET and DAAD, respectively. S.M.B., P.C.A., and D.R. are part of the EMBIC Network of Excellence, which is cofinanced by the European Commission through the FP6 framework program "Life Science, Genomics and Biotechnology for Health."
Correspondence: 2Sandra M. Blois, Charité Centrum 12, Internal Medicine and Dermatology, BMFZ, Raum 2.0547, Augustenburger Platz 1, 13353 Berlin, Germany. FAX: 49 30 450 553997; e-mail: sandra.blois{at}charite.de or smblois{at}essex.ac.uk
Received: 5 February 2007.
First decision: 20 February 2007.
Accepted: 6 June 2007.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||
