Tolerance, Immunosuppression, and Immune Modulation: Impacts on Lung Allograft Survival

Abstract

Purpose of Review

Immune responses following lung transplantation continue to result in high rates of allograft failure and rejection, and current immunosuppression does not address the unique immunologic properties of the lung. Here, we review recent studies on lung allograft tolerance and alloimmunity and discuss implications for immunosuppression.

Recent Findings

Processes governing tolerance and alloimmunity in lung allografts differ from other solid organs. Recent studies have suggested that allorecognition is regulated at the level of the lung graft. Furthermore, certain cell populations essential for lung allograft tolerance may facilitate rejection in other organs. Induction of lung allograft tolerance is associated with the formation of tertiary lymphoid organs, which are enriched in regulatory T cells and play an important role in preventing rejection.

Summary

Recent discoveries regarding alloactivation and the regulation of tolerance following lung transplantation have introduced exciting potential avenues for the development of lung-specific immunosuppression.

Introduction

While the field of pulmonary transplantation has seen a dramatic improvement in outcomes since the first lung transplant in 1963, the clinical practice remains substantially limited by chronic rejection, which continues to affect approximately 50% of recipients within 5 years after transplantation [1,2*]. Although we have deepened our understanding of the pathways which mediate rejection and tolerance, new mechanisms of lung transplant alloimmunity have been discovered in recent years and much remains to be understood before these can be harnessed clinically to improve outcomes. Current immunosuppressive regimens for lung transplant recipients continue to fall short, utilizing agents that have largely been adopted from experiences with other solid organs and do not account for the unique immunologic features of lungs. Our research has suggested that current immunosuppression may in fact have deleterious effects on protective cell populations, such as regulatory and central memory T cells which mediate graft tolerance. These findings may, in part, explain why lung transplant recipients remain plagued by high rates of allograft failure and survival rates trail behind those of other solid organa [2*]. While modern practices have resulted in improved short-term outcomes, long-term survival rates have remained relatively constant throughout the last several decades, with 5- and 10-year survival rates of 62% and 43%, respectively [3]. For these reasons, an urgent need exists to gain better mechanistic understanding of events governing rejection and tolerance of lung grafts to improve recipient outcomes.

Recognition of donor alloantigens by the recipient immune system occurs through three routes, known as the direct, indirect, and semi-direct allorecognition pathways. Direct allorecognition involves recipient T cell recognition of donor major histocompatibility complex (MHC) molecules on donor antigen-presenting cells (APCs). Indirect allorecognition involves recipient T cell recognition of processed donor antigens presented on recipient APCs in the context of recipient MHC molecules. Semi-direct allorecognition involves recipient T cell recognition of donor MHC molecules on recipient APCs. Our research has demonstrated that allorecognition pathways, linked to both rejection and tolerance, become activated within the lung and are regulated at the graft level [4–6]. We have observed that bronchus-associated lymphoid tissue (BALT) is induced in tolerant mouse lung allografts and enriched in regulatory CD4⁺ Foxp3⁺ T cells (Tregs) [7]. In contrast to other transplanted organs, where the generation of alloimmune responses is dependent on cell trafficking to secondary lymphoid organs such as the lymph nodes or spleen, lung allografts provide a suitable environment for the interaction of immune cells [8,9]. These findings highlight the notion that alloimmunity, tolerance, and immune modulation in lung transplant recipients is substantially different from other solid organ transplants and unique approaches aimed at decreasing rejection are called for.

Here, we discuss insights gained from recent studies on lung transplant alloimmunity and tolerance. We also address how these findings apply to current immunosuppression and novel therapeutics that may serve to improve outcomes in the future.

Alloimmunity

Acute cellular rejection is primarily driven by T cell activation after recognition of donor alloantigens. Such T cell activation appears to be facilitated by innate immune pathways triggered by early graft injury [10,11**,12]. Ischemia-reperfusion injury (IRI), a frequent complication after lung transplantation, is considered to be the principal mechanism leading to primary graft dysfunction (PGD). PGD occurs in up to 80% of recipients and is associated with chronic rejection and late mortality [3,13]. Following reperfusion of ischemic lung grafts, various forms of inflammatory cell death trigger the dissolution of plasma membranes, releasing intra-cellular contents which propagate graft injury [14,15*]. Neutrophil recruitment to lung grafts is orchestrated by donor non-classical and recipient classical monocytes through secretion of CXCL2 and IL-1β, respectively [16,17]. Neutrophil accumulation within the lung allograft leads to upregulation of pro-inflammatory mediators, stimulating alloantigen-specific T cell expansion and inhibition of tolerance induction [12,18]. This innate immune activation after IRI provides a mechanistic link between PGD and adaptive alloimmunity, suggesting that early events within the lung allograft set the stage for allorecognition and rejection.

We have previously shown that bacterial infections after lung transplantation can trigger rejection as well. Following infection with Pseudomonas aeruginosa, we demonstrated that graft-infiltrating neutrophils mediate T cell activation via upregulation of B7 [18]. Similarly, others have shown that bacterial and viral infections are associated with chronic rejection in the clinics [19,20]. A recent study by our group showed that the presence of a synthetic bacterial lipopeptide in donor airways was sufficient to prevent tolerance induction [11**]. This response to bacterial lipopeptide was mediated by recipient-derived monocytes and induced expansion of CD8⁺ T cells within the lung graft.

Antibody-mediated rejection (AMR) is an increasingly recognized form of allograft rejection that can lead to early fatal complications or trigger chronic rejection [21]. AMR is characterized by allograft dysfunction, abnormal histopathology, and circulating donor-specific antibodies (DSA). In vitro studies have demonstrated that exposure of human airways to anti-HLA antibodies induces fibroblast proliferation and epithelial apoptosis [22]. We demonstrated that tolerant mouse lung allografts depleted of Tregs that reside within the BALT develop high levels of DSA and experience AMR [23]. Collectively, these findings lend support to the notion that alloimmunity following lung transplantation is regulated within the allograft itself.

Tolerance

Similar to evidence discussed above suggesting that deleterious alloimmune responses occur within the lung graft, tolerance also appears to be mediated locally. We previously demonstrated that central memory CD8⁺ T cells play an important role during the induction of tolerance following lung transplantation, which is in stark contrast to studies showing that such cells facilitate rejection in other solid organ grafts [6]. We have shown that IFN-γ production by CD8⁺ central memory T cells induces nitric oxide (NO) in the graft which is critical for downregulation of alloimmune responses. Expanding on this finding that NO is essential for tolerance, we subsequently identified recipient-derived eosinophils as NO producers during tolerance induction. NO production is dependent on inflammatory mediators such as IFN-γ and TNF-α, and resultant inhibition of T cell receptor (TCR) signaling is dependent on interactions between eosinophils and T cells mediated by PD-L1/PD-1 [24]. Importantly, intra-tracheal administration of eosinophil chemotactic chemokines decreased the degree of acute rejection [25]. We have also shown that CD8⁺ central memory T cells promote accumulation of Tregs within lung grafts thereby setting the stage for the establishment of immune regulation [6].

We have observed that lung transplant tolerance is associated with the development of BALT, a tertiary lymphoid organ enriched in Tregs [7]. In contrast to our observations in lung allografts, the formation of intragraft tertiary lymphoid organs has been associated with acute or chronic rejection in other transplanted tissues including hearts [26,27]. To this end, utilizing mouse models of re-transplantation, we have demonstrated that local immune responses occurring within pulmonary, but not cardiac grafts, regulate the maintenance of tolerance [7]. BALT in tolerant lung allografts is characterized by T cells, B cells, APCs, lymphatics, and high endothelial venules [7,28**]. We have shown that the induction of the accumulation of B cells and the formation of high endothelial venules within BALT depends on IL-22, which is predominantly produced by graft-infiltrating γδ T cells and type 3 innate lymphoid cells. Despite the absence of IL-22 expression in recipients, Treg aggregates formed in the graft where they played an important role in maintaining tolerance [29].

Tregs can suppress anti-inflammatory responses through soluble mediators or through cellular contact. There is evidence that Tregs are protective after lung transplantation in humans. For example, reduction in airway Tregs has been found to correlate with rejection in lung transplant patients [30]. Additionally, a recent study showed that increased levels of circulating Tregs were associated with improved graft survival [31]. We have also shown that selective depletion of graft-resident Tregs from tolerant lung allografts triggers AMR. Using intravital microscopy, Tregs are observed closely interacting with graft-infiltrating B cells and CD4⁺ T cells within tolerant lung grafts. When graft-resident Tregs are depleted, B cell activation ensues and DSA levels rise thereby triggering AMR [23]. These findings indicate that BALT-resident Tregs inhibit humoral immune responses following lung transplantation and provide a framework for novel therapies aimed at preventing AMR.

Tregs are capable of egressing from tolerant lung grafts via lymphatics and conferring systemic tolerance to donor antigens [28**]. We discovered that mice with tolerant lung grafts will subsequently accept a donor-matched heart allograft without immunosuppression. Acceptance of heart allografts depends on the continued presence of the lung graft, as a transplant pneumonectomy will result in rejection of the heart allograft. We found that lung allograft-derived Tregs migrate to heart grafts and disruption of lymphatic drainage from tolerant lung grafts prevents cardiac graft acceptance. Lymphatic drainage also appears to play a role in downregulating alloimmune responses through Treg-independent mechanisms, such as clearing damage-associated molecular patterns [32]. To this end, treatment with VEGF-C, which promotes lymphangiogenesis, dampens acute rejection and facilitates clearance of damage-associated molecular patterns from the lung graft. Given that lymphatic disruption is inherent to the lung transplant procedure, facilitating lymphangiogenesis after transplantation may be a strategy for blunting alloimmune responses both locally and in the periphery.

Immunosuppression and Immune Modulation

These findings provide an impetus to critically evaluate current immunosuppressive strategies used after lung transplantation, as many of them inhibit pathways essential for lung allograft tolerance. Examples include peri-operative global T cell depletion therapies, which eliminate Tregs and central memory CD8⁺ T cells, as well as calcineurin inhibitors, which suppress Treg activation and decrease NO production [33,34]. Other agents have mixed effects on tolerogenic pathways, such as mycophenolate mofetil, which increases peripheral Tregs in comparison to calcineurin inhibitors, but also inhibits early Th1 responses [35,36]. Conversely, glucocorticoids may alter the balance of T cell subsets to favor Tregs [37]. Therefore, the development of targeted, lung-specific immunosuppression is warranted to prevent inadvertent depletion of tolerogenic cell populations.

There are several unique considerations regarding the delivery of immunosuppression in lung transplant recipients. Unlike most other organs, lungs possess an easily accessible route for local drug delivery via the airway, which we have demonstrated as a successful means of administering biologic mediators for altering lung allograft rejection in mice [24]. Similarly, airway administration of viruses expressing anti-inflammatory cytokines has been shown to reduce rates of allograft rejection in mice [38]. A rationale for inhalational immunosuppression is supported by our demonstration that activation of alloreactive CD8⁺ T cells can occur within the airways [39]. There is also tremendous potential for therapeutics which can be applied directly to donor lungs prior to transplantation via ex vivo lung perfusion (EVLP). For example, administration of cyclosporin A via EVLP in rats has been associated with improved lung graft preservation, and others have similarly demonstrated promising results after priming lungs with adenosine A_2A receptor agonists [40*,41].

Several potential therapeutic avenues exist for developing novel immunosuppression targeted for lung allografts. Enhancing tolerogenic cell populations within the lung graft is one such strategy. For example, infusion of Tregs is being prominently explored in pre-clinical studies. Due to the significantly higher cell number required and the theoretical risk of non-specific immunosuppression with polyclonal Tregs, antigen-specific Tregs may be a preferred option [42]. Various methods for antigen-specific Treg production are being investigated, including CD40L-activated B cells and synthetic APCs fabricated from biodegradable polymers [43,44]. Tregs expressing chimeric antigen-specific receptors (CAR-Tregs) have shown promise in preventing rejection in models of skin transplantation, and other strategies may allow preferential localization of CAR-Tregs in inflamed lungs, which holds potential as a method for targeted drug delivery to avoid global immunosuppression [45,46]. Extracorporeal photopheresis is another strategy employed clinically with some success in the treatment of AMR and chronic rejection after lung transplantation [47,48]. While the precise mechanisms remain to be elucidated, extracorporeal photopheresis downregulates immune responses, at least in part, through enriching Tregs [49]. Eosinophils present another potential cell target to promote a tolerogenic microenvironment, and we have previously demonstrated that intratracheal eotaxin and IL-5 increases eosinophil concentrations within lung grafts in mice [25]. Taken together, these data support targeting specific cell populations and tolerogenic feedback loops to downregulate lung graft rejection while avoiding global immunosuppression.

Suppression of alloimmune responses via inhibition of T or B cell activation is another strategy for future immunosuppressive therapy. In an orthotopic lung transplant model in mice, expansion of recipient natural killer cells has been shown to eliminate donor dendritic cells, reducing T cell alloactivation [50]. As mentioned previously, we have shown that neutrophil-mediated T cell activation is inhibited by B7 blockade, representing another potential target for immunotherapy [18]. Additionally, using our mouse transplant model, we showed that neutralization of CXCL13 and inhibition of ICOS-ICOS-L / CD40-CD40L, pathways regulating activation of B cells by T follicular helper cells, can ameliorate graft rejection [23]. Such data warrant investigation in preclinical trials to determine if they can be extrapolated to human recipients.

Conclusion

As outlined above, the failure of lung transplants is largely mediated by the recipient immune response, and numerous inflammatory signals which activate either the innate or adaptive immune system have been recognized as risk factors for allograft rejection. Research over the last decade suggests that immune responses to lung grafts differ from those of other organs. We have recently discovered that alloimmune pathways determining both rejection and tolerance are regulated within the lung graft through pathways unique to lung transplantation. Thus, prevention of immune-mediated failure of pulmonary allografts requires the development of targeted therapies tailored to the lung’s unique immunologic properties.

Key points:

• Alloimmune responses after lung transplantation are regulated at the level of the graft.

• Acute rejection of lung allografts may be triggered by activation of innate immune pathways in response to early graft injury, such as IRI or infection.

• Tregs, central memory CD8⁺ T cells, and eosinophils are important modulators of lung allograft tolerance.

• Current immunosuppression regimens do not take into account the unique immunologic features of the lungs, and many immunosuppressive agents have deleterious effects on protective immune cell populations.

• Enhancement of tolerogenic cell activity and suppression of alloimmune responses within lung grafts provide potential avenues for development of novel lung-specific immunosuppression.

Abbreviations

ACR

Acute cellular rejection

AMR

Antibody-mediated rejection

APC

Antigen-presenting cell

BALT

Bronchus-associated lymphoid tissue

DSA

Donor-specific antibodies

EVLP

Ex vivo lung perfusion

HLA

Human leukocyte antigen

IRI

Ischemia-reperfusion injury

MHC

Major histocompatibility complex

NO

Nitric oxide

PGD

Primary graft dysfunction

TCR

T cell receptor

Treg

Regulatory T cell