Emerging role of exosomes in allorecognition and allograft rejection

Abstract

Purpose of review

This article reviews recent literature on the nature of extracellular vesicles released by allogeneic transplants and examine their role in T-cell alloimmunity involved in rejection and tolerance of these grafts.

Recent findings

Donor cells release extracellular vesicles, including exosomes, after transplantation of allogeneic organs and tissues. Consequently, recipient APCs take up these exosomes and present donor MHC antigens on their surface (allo-MHC cross-dressing) thus, activating some alloreactive T cells via a mechanism called semi-direct pathway of allorecognition. In addition, one study shows that exosomes carrying noninherited maternal antigens are associated with maternal microchimerism and tolerance in offspring. Finally, a few studies describe potential utilization of exosomes as modulators of alloimmunity and biomarkers of rejection in allotransplantation.

Summary

Extracellular vesicles, including exosomes, released by allografts contribute to recognition of donor antigens by T cells after allotransplantation. This occurs through cross-dressing of recipient APCs with donor MHC antigens and subsequent activation of T cells, a process called semi-direct alloreactivity. The relevance of this phenomenon in rejection and tolerance of allografts and the potential utilization of exosomes as biomarkers in transplantation are discussed.

INTRODUCTION

Allograft rejection is initiated in the recipient’s secondary lymphoid organs by T lymphocytes recognizing donor major histocompatibility (MHC) antigens (allorecognition). During this phase, activation of alloreactive T cells occurs via two distinct pathways: the direct pathway in which T cells recognize intact allogeneic MHC proteins on donor MHC class II⁺ leukocytes (passenger leukocytes) and the indirect pathway whereby T cells recognize donor peptides processed and presented by recipient antigen-presenting cells (APCs) [1,2,3▪]. Direct allorecognition by T cells triggers a massive polyclonal inflammatory response, which leads to acute rejection of allografts [4]. In contrast, indirect alloreactivity involves a limited set of T-cell clones specific to a few dominant determinants on donor MHC [5,6]. The role of indirect alloresponses in acute allograft rejection is still unclear. On the other hand, there is strong circumstantial evidence supporting the view that indirectly activated alloreactive T cells play a key role in alloantibody production and chronic graft rejection, a condition characterized by graft tissue fibrosis and blood vessel obstruction [7–10].

Two recent studies demonstrate that, in addition to passenger leukocytes, allografts release extracellular vesicles, which migrate out of the graft and carry donor MHC molecules to the recipient’s lymphoid organs [11▪▪,12▪▪]. These donor-derived vesicles, including exosomes, are regularly taken up by recipient APCs, which then present on their surface donor MHC molecules, a phenomenon that we refer to as allo-MHC cross-dressing [11▪▪]. Previous studies by Lechler’s laboratory had shown that recipient cells displaying allogeneic MHC molecules could activate allospecific T cells. Therefore, it is plausible that this type of alloantigen recognition called semi-direct pathway could be involved in allograft rejection [13▪,14]. Alternatively, a few reports suggest that certain extracellular vesicles can promote allograft tolerance [15,16▪▪,17,18]. This article reviews recent studies evaluating the role of extracellular vesicles released after allotransplantation and their involvement in immunity associated with allograft rejection or tolerance.

EXTRACELLULAR VESICLES IN REJECTION OF SKIN ALLOGRAFTS

Conventional, nonprimarily vascularized, skin allografts induce potent direct and indirect alloresponses by CD4⁺ and CD8⁺ T cells. The traditional view is that direct priming of T cells in draining lymph nodes occurs via presentation of allogeneic MHC antigens by passenger leukocytes having migrated out of the skin graft (dendritic cells) [19–21]. However, the presence of donor dendritic cells in lymph nodes draining skin allografts has never been formally proven. Recently, our laboratory showed that, after skin grafting in mice, donor MHC molecules are actually transported to the host’s lymphoid organs by extracellular vesicles [3▪]. Recipient dendritic cells and B cells then acquire these vesicles and present intact donor MHC class I and II molecules to allospecific T cells. Furthermore, we showed that injection of naïve mice with allogeneic extracellular vesicles (purified exosomes) or with recipient cells cross-dressed in vitro with donor exosomes was sufficient on its own to elicit a direct inflammatory alloresponse in vivo and sensitize mice to donor antigens [11▪▪]. This corroborates former studies by R Lechler’s laboratory showing the ability of allo-MHC cross-dressed cells to elicit allospecific T-cell responses in vitro (mixed lymphocyte reaction) and in vivo in mice [11▪▪]. In a subsequent study, Smyth et al. [22▪▪] showed continuous acquisition of MHC–peptide complexes by recipient cells and its role in the generation of antigraft CD8⁺ T-cell immunity in skin-grafted mice. Altogether, these studies suggest that the presentation of donor MHC by recipient APCs rather than by donor passenger leukocytes represents the driving force behind direct alloreactivity and acute rejection in skin transplantation. However, it is important to note that this conclusion may not apply to primarily vascularized skin allografts, whose passenger leukocytes can migrate out of the graft through blood vessels right after their placement and activate T cells in the recipient’s spleen, as observed with heart transplants [4]. Finally, it is still unknown whether donor exosomes are involved in the initiation of indirect T-cell alloresponse, which is an essential component of skin allograft rejection. Further studies designed to inhibit the production or secretion of extracellular vesicles by donor cells or their uptake by recipient cells will be required to evaluate the actual role of exosomes and donor antigen cross-dressing in skin allograft rejection.

EXTRACELLULAR VESICLES IN REJECTION OF VASCULARIZED SOLID ORGAN TRANSPLANTS

Cardiac allografts are connected to the host’s blood vessels through surgical anastomosis during their placement. This facilitates immediate trafficking of donor passenger leukocytes out of the graft, presumably through reverse transendothelial migration. Indeed, seminal studies by Larsen [23] have documented the presence of donor dendritic cells in the spleen of mice within a day after transplantation of an allogeneic heart. Recently, we readdressed this question using imaging flow cytometry and confirmed the presence of donor MHC class II⁺ cells in the spleen of heart-transplanted mice. However, less than 100 donor cells per million spleen cells were detected at day 1 posttransplantation and their number declined rapidly thereafter [11▪▪]. On the other hand, one day after transplantation, we found 30 000 recipient spleen cells cross-dressed with donor MHC molecules. Strikingly, by day 7, nearly 100 000 cross-dressed cells were detected [11▪▪]. Such high frequencies of allo-MHC cross-dressed cells in lymphoid organs of heart-transplanted mice were also observed in a concurrent study in Morelli’s laboratory [12▪▪]. In both Marino et al.’s [11▪▪] and Liu et al.’s [12▪▪] reports, transfer of allogeneic MHC proteins was shown to involve extracellular vesicles displaying characteristic features of exosomes. This corroborates former studies by the Wong’s group showing extensive transfer of MHC molecules in vivo after cardiac transplantation [24,25]. Importantly, Morelli’s studies showed that recipient cDCs having acquired exosomes became activated and stimulated alloreactive T cells. Furthermore, depletion of recipient cDCs after cardiac transplantation markedly decreased presentation of donor MHC antigens to alloreactive T cells and delayed graft rejection in mice [12▪▪]. In further support of this view, Harper et al. [26] showed the key role of acquired alloantigens in CD8⁺ T-cell responses during acute rejection of heart allografts. These studies support the view that the same recipient APC can present allopeptides indirectly to CD4⁺ T cells and donor MHC class I directly to CD8⁺ T cells. Such a ‘three cell model’ would explain how CD4⁺ T cells help the differentiation of cytotoxic CD8⁺ T cells in a MHC class I-only disparate transplant setting (depicted in Fig. 1). Taken together, these studies support the view that release of exosomes by heart transplants and subsequent presentation of donor MHC molecules by recipient APCs in the recipient’s spleen triggers T-cell alloresponses involved in acute rejection of these allografts.

FIGURE 1.

MHC cross-dressing can promote cooperation between alloreactive CD4⁺ and CD8⁺ T cells. Panels (a) and (b) depicts two models describing how indirectly activated CD4⁺ T cells can provide help to CD8⁺ T cells after placement of a MHC class I-mismatched allograft. Panel (a) shows a classical four-cell cluster whereby CD4⁺ and CD8⁺ T cells recognize donor MHC indirectly (donor MHC peptide) and directly (intact donor MHC), respectively on distinct APCs. Panel (b) shows how donor MHC class I cross-dressing can promote CD4⁺⁻CD8⁺ T-cell cooperation by having self-MHC class II and donor peptide (indirect presentation) and donor MHC class I (direct presentation) presented on the same antigen-presenting cell (three-cell cluster).

EXTRACELLULAR VESICLES IN BONE MARROW TRANSPLANTATION

Transplantation of allogeneic bone marrow (BMT) or hematopoietic stem cells along with recipient’s conditioning is a treatment of choice for hematological malignancy. In addition, induction of allogeneic hematopoietic chimerism has been used in animal models and patients to mediate tolerance of allografts [27,28]. However, graft-versus-host disease (GvHD) caused by donor T cells reacting to recipient MHC and minor histocompatibility antigens is a major complication associated with this therapy. Apparently, both direct and indirect recognition of recipient MHC and minor histocompatibility antigens by donor T cells are involved in GvHD [29–31]. An elegant study by Markey et al. [32] demonstrated that many donor cells become cross-dressed with recipient MHC class I and II molecules following BMT. Interestingly, the presence of recipient MHC proteins on donor dendritic cells (cross-dressing) was transient in the setting of myeloablative conditioning but durable after nonmyeloablative conditioning, in which recipient hematopoietic cells remain in high numbers [32]. This suggested that only recipient professional APCs can transfer their MHC antigens to donor dendritic cells. Next, using TCR transgenic mice (TEa) whose T cells recognize a single MHC–peptide complex (A^b MHC class II bound to Eα 52–68 peptide), they observed that cross-dressed cells could not activate TEa T cells through semi-direct pathway. In turn, such cross-dressed cells could enhance indirect alloresponse by T cells recognizing the same MHC– peptide complex, presumably by increasing the level of interaction between APCs and CD4⁺ T cells [32]. These results suggest that MHC cross-dressing of donor cells with recipient MHC antigens occurs regularly after BMT. Further studies will be needed to determine the contribution of exosomes in this process and the role of semi-direct alloresponses in GvHD.

EXTRACELLULAR VESICLES AS BIOMARKERS IN TRANSPLANTATION

Reliable and early diagnosis of rejection is critical in clinical transplantation. Indeed, current methods often detect deleterious immunity whenever it is too late to revert its course and/or the organ transplant is irreversibly damaged. Most important, there is a need for noninvasive methods to detect chronic allograft rejection, a process characterized by graft tissue fibrosis and vasculopathy affecting a large proportion of organ transplants within 1–5 years after their placement [33,34]. Finally, current histological examination of small sample biopsies does not necessarily reflect the entire allograft’s disease status.

It is well documented that the presence of selected extracellular vesicles, including exosomes, in bodily fluids reflects histological and physiological features of organs and tissues. For instance, the presence of exosomes carrying defined proteins and RNA are found in urine and blood of patients with various kidney diseases [35–37]. Similar findings have been reported in patients with different types of cancer and autoimmune diseases [38–41]. This suggests that, in transplanted patients, the release of certain exosomes in the blood or urine might correlate with the type and stage of rejection of their allografts. In support of this view, a recent article from Mohanakumar’s laboratory showed that donor-derived exosomes carrying lung tissue-specific antigens (collagen V) are regularly found in the serum of lung-transplanted patients with bronchiolitis obliterans syndrome (BOS), which is characteristic of chronic rejection [42▪▪]. Interestingly, detection of graft-derived exosomes preceded clinical diagnosis suggesting that it could serve as a method to predict the onset of chronic rejection and adjust patients’ treatment accordingly. In another study, Tower et al. [43▪▪] reported that the number of plasma C4d⁺ endothelial microvesicles correlated with antibody-mediated rejection and its severity in patients, recipient of a renal allograft. Furthermore, Sigdel et al. [44] and Alvarez et al. [45] reported altered protein contents in urinary exosomes from patients undergoing acute rejection of kidney allografts. Finally, a recent article from Naji and co-workers showed the presence of donor-derived exosomes displaying the endocrine markers, insulin, glucagon and somatostatin, in blood samples of recipients after islet transplantation [46▪▪]. Therefore, monitoring exosomes and their cargoes in blood or urine of patients might represent a promising noninvasive method to evaluate the status of allografts and the nature and severity of their rejection.

EXOSOMES AND ALLOGRAFT TOLERANCE

Immune tolerance of allografts is often associated with alloantigen presentation in a tolerogenic fashion, that is, by selected APCs in a context favoring protective rather than destructive immunity. This suggests that, under certain circumstances, exosomes carrying donor antigens could be involved in tolerance rather than rejection of allografts.

In support of this view, a recent study by Burlingham’s laboratory shows the key role of donor exosomes in the amplification of maternal microchimerism and regulation of immunity to noninherited maternal antigens in adults [16▪▪]. In addition, some studies show that allogeneic exosomes given in a defined fashion and Treg-derived exosomes can be used to promote tolerance to allografts [15,17,18,47▪▪].

CONCLUSION

It is now clear that allogeneic cells release extracellular vesicles such as exosomes, after organ and tissue transplantation. Subsequently, recipient professional APCs (dendritic cells and B cells) take up these vesicles and present donor MHC antigens on their surface (MHC cross-dressing). These allo-MHC cross-dressed cells can activate recipient alloreactive in vitro and in vivo (semi-direct pathway). Also, donor MHC and minor antigens carried by donor exosomes might represent a major source of allodeterminants for processing and indirect presentation by recipient APCs. On the other hand, no or very few donor passenger leukocytes migrate to recipient lymphoid organs after transplantation and they are eliminated rapidly by the host’s immune system. This suggests that donor exosomes rather than passenger leukocytes are the main source of antigens for allorecognition by T cells after transplantation. However, this still remains to be demonstrated as well as the contribution of exosomes and antigen cross-dressing in rejection and tolerance of allografts.

KEY POINTS.

• Extracellular vesicles, including exosomes, carrying donor MHC antigens are released by allografts after transplantation.

• Graft-derived exosomes are captured by recipient antigen-presenting cells, which then present intact donor MHC molecules on their surface, a phenomenon referred to as allo-MHC cross-dressing.

• Recipient T cells are activated in vitro and in vivo following recognition of donor MHC molecules presented on recipient APCs, a process called semidirect alloreactivity.

• Selected exosomes can be used to prolong allograft survival.

• Graft-derived exosomes can serve as biomarkers in acute and chronic rejection or just chronic rejection of solid organ allografts.