Potential Impact of Extracorporeal Photopheresis on Trained Immunity and Organ Transplant Acceptance

Abstract

Extracorporeal photopheresis (ECP) is a well-established, safe, and effective immunomodulatory therapy currently used in clinics to decrease T cell–mediated immunity in various disorders, including autoimmune diseases and chronic rejection in organ transplantation. Although the ECP procedure has been shown to induce apoptotic cells that are reintroduced into the patient at the end of the treatment, the precise tolerogenic mechanisms mediated by ECP are not fully understood. Previous in vitro studies have demonstrated that early apoptotic cells express annexins on their cell surface, which suppress myeloid cell activation on stimulation with bacterial lipopolysaccharide through Toll-like receptors. Mechanistically, annexins prevent the upregulation of costimulatory molecules (CD40 and CD86) and decrease the secretion of proinflammatory cytokines (tumor necrosis factor and interferon-γ) through nuclear factor kappa B signaling pathways, altogether inhibiting antigen-specific T-cell responses in vivo. In human and mouse bone marrow-derived macrophages, binding of annexin to Dectin-1, a c-type lectin receptor, promotes peripheral tolerance through the spleen tyrosine kinase signaling pathway and NADPH oxidase 2 downstream activation. In animal models, the synergistic activation of Dectin-1 and Toll-like receptor 4 by damage-associated molecular patterns in graft-infiltrating monocytes leads to the induction of trained immunity. Because trained immunity prevents long-term allograft survival in organ transplant recipients, we hypothesize pretreatment with ECP represents a potential unexplored therapeutic option to favor transplantation tolerance. Specifically, ECP may serve as a prophylactic therapy to prevent trained immunity in contexts involving the activation of the Dectin-1 pathway.

INTRODUCTION

Extracorporeal photopheresis (ECP) is an established immunomodulatory therapy that involves intravenous infusion of autologous apoptotic leucocytes. Originally developed for the management of cutaneous T-cell lymphoma, ECP is now an approved therapy in Europe to prevent heart or lung transplant rejection in adults and acute or chronic graft-versus-host disease in patients older than 3 y.¹ During the ECP process, leucocytes are collected from a patient by apheresis, which is subsequently treated with a photosensitizing agent called 8-methoxypsoralen and ultraviolet A irradiated to induce apoptosis.² On reinfusion, these apoptotic cells induce antigen-specific regulatory T cells (Treg), regulatory B cells, and regulatory natural killer cells. The mechanisms by which ECP controls alloimmunity are complex and not fully understood; however, the primary action of ECP is through macrophages (Mφ) and dendritic cells (DCs) in inflamed tissues or lymphoid tissues, which develop an anti-inflammatory phenotype when exposed to apoptotic cells.^3,4 A recent study that combines in vitro and animal models has shown that the apoptotic cells reinfused during ECP preferentially migrate to inflamed tissues, where they are engulfed by Mφ that subsequently acquire a regulatory phenotype and secrete adiponectin, a molecule exerting broad anti-inflammatory effects on the local microenvironment.⁵

Modern immunosuppressive therapies work by suppressing T- and B-cell responses to prevent the immune system from attacking the transplanted organ. These typically include calcineurin inhibitors to block T-cell activation, antimetabolites to inhibit lymphocyte proliferation, and corticosteroids to broadly suppress inflammation.⁶ They have participated in significantly improved short-term transplant outcomes; however, chronic graft rejection remains a major obstacle, with little improvement in long-term graft survival rates during the past 2 decades.⁷ Although ECP is clinically inconvenient, it has become guideline therapy for treatment-refractory heart and lung transplant rejection because it slows functional decline.^8,9 The empirical success of ECP in managing chronic transplant rejection emphasizes the need for novel immunosuppressive therapies that target innate immunity and tissue repair, especially Mφ and DC.

THE ROLE OF INNATE IMMUNITY IN GRAFT REJECTION

Growing evidence highlights the critical role of innate immune mechanisms in organ rejection, as well as the establishment and maintenance of allograft tolerance.¹⁰ Of special interest, donor- and recipient-derived Mφ play critical roles in acute graft injury, tissue repair processes, chronic rejection, and transplant acceptance.¹¹ The contribution of Mφ to transplant pathology is context dependent.^12,13 Tissue-resident Mφ and DC of donor origin respond to sterile insults, including ischemia/reperfusion injury (IRI) caused by donor death and organ retrieval, leading to an immediate inflammatory response that recruits blood monocytes and neutrophils of recipient origin into the graft. This wave of infiltrating cells triggers an early induced response associated with significant tissue damage and recruitment of recipient T cells. Alloreactive T cells further exacerbate the ongoing injury, resulting in acute cellular rejection (ACR) and transplant loss.¹⁴ Fortunately, the modern combination of glucocorticoids, calcineurin inhibitors, and antimetabolites is very effective in preventing ACR. When early ACR is adequately controlled, organs enter a phase of tissue repair, characterized by the presence of recipient-derived, tissue-reparative Mφ that remove necrotic cells, stimulate parenchymal regeneration, promote revascularization, and locally suppress T cells.⁴ Ideally, the organ would then return to its normal state of immunological and physiological homeostasis; however, in a substantial proportion of cases, this process is incomplete, leading to persistent inflammation, graft fibrosis, and, in some cases, development of late de novo donor-specific antibodies.

Work from our group and others in mouse and human in vivo studies has highlighted the importance of a third population of Mφ, which we call “regulatory macrophages” or Mreg, in establishing transplant tolerance. These graft-infiltrating cells, which originate from colony-stimulating factor 1–stimulated blood monocytes, differentiate into a distinct subset of Mφ with potent immunoregulatory properties.^15,16 They were investigated as a cell-based therapy to resolve chronic inflammation in grafted organs and reduce corticoid and immunosuppressive treatments in patients.¹⁷ A combination of in vitro experiments, animal models, and early-phase clinical studies suggests that donor-derived Mreg, when administered before kidney transplantation, may promote alloantigen-specific unresponsiveness. In vitro and preclinical studies have shown reductions in T-cell populations and increased Treg induction, whereas preliminary clinical investigations indicate potential for immunomodulatory effects in patients.^18-20. In vitro assays with human or mouse cells attribute these effects on T-cell populations, on the one hand, to the ability of Mreg to promote the expansion of interleukin (IL)-10-producing T cell immunoreceptor with Ig and ITIM domains Forkhead box P3 regulatory T cells (Treg)²¹ and, in contrast, to the direct immunosuppressive effects of Mreg on T cells by depleting them from essential metabolites such as l-arginine through inducible nitric oxide synthase expression, leading to the inhibition of T-cell proliferation and function.²² Additionally, Mreg influence antigen-presenting cells, such as DCs, by preventing their maturation and reducing the expression of costimulatory molecules (CD80/CD86), thereby limiting effector T-cell activation and fostering a tolerogenic environment.²¹ Mreg also regulate cytokine responses by suppressing proinflammatory mediators such as interferon-γ and IL-12 while enhancing IL-10 and transforming growth factor-beta secretion, further reinforcing immune regulation.²³

TRAINED IMMUNITY AND ORGAN REJECTION

Trained immunity is the phenomenon of long-term functional reprogramming of innate immune cells, especially Mφ, to respond more vigorously to rechallenge. Following an initial stimulus or injury, macrophages can become trained to respond more strongly to later exposures, independently of the adaptive immune system.^24,25 This enhanced response relies on modifications in Mφ phenotype characterized by a metabolic shift toward glycolysis and stable epigenetic modifications (eg, H3K4me3, H3K27ac) at proinflammatory gene loci, which lead to increased production of proinflammatory cytokines on restimulation (eg, tumor necrosis factor-α, IL-6) and upregulation of costimulatory molecules, such as CD40 and CD86.^24,25

In humans, trained immunity has been extensively studied after vaccination, the most documented one being the Bacillus Calmette-Guérin vaccine.²⁶ A well-described in vitro protocol for inducing trained immunity in human monocytes relies on bacterial stimulation, where monocytes are first exposed to β-glucan binding to Dectin-1, which acts as a first stimulus, followed by restimulation with lipopolysaccharide (LPS) binding to Toll-like receptor (TLR)4.^27,28 Although trained immunity has been mostly explored in the context of infectious disease, recent reviews highlight growing preclinical evidence from in vitro studies and animal models, suggesting it may also play a role in transplantation-related inflammation and graft rejection.^11,29 This is particularly relevant in kidney transplantation where experimental models have associated heightened trained immunity responses with poorer graft outcomes and an increased risk of rejection, although causal relationships in humans remain to be established.^30,31

Indeed, on encountering pathogens or necrotic cells, Mφ recognize damage-associated molecular patterns (DAMPs) and pathogen-associated molecular patterns (PAMPs) through pattern recognition receptors such as TLRs and C-type lectin receptors such as Dectin-1. In the context of organ transplantation, it has been demonstrated in mouse models that the synergistic stimulation of Dectin-1 and TLR4 by DAMPs such as vimentin and high-mobility group box 1 (HMGB1), which are released during IRI, can induce trained immunity in Mφ.^29,32 Dectin-1 binds to many ligands, resulting in different immunogenic effects, some of which are proinflammatory and/or training-inducing effects.^33,34 Among them, β-glucan is interesting as its activation pathways are widely described. One of the initial signaling events triggered by β-glucan-mediated activation of Dectin-1 is the phosphorylation of spleen tyrosine kinase (SYK) by Src family kinases.^35,36 SYK is a cytoplasmic nonreceptor tyrosine kinase essential for signaling downstream of immune receptors such as Dectin-1. On β-glucan binding, the hemITAM motif of Dectin-1 is phosphorylated by Src family kinases, which are membrane-associated tyrosine kinases that initiate early immune signaling. This phosphorylation recruits and activates SYK via its SH2 domains. Activated SYK then phosphorylates adaptor proteins such as CARD9.³⁷ This strong phosphorylation activates multiple intracellular signaling pathways, including the nuclear factor kappa B (NF-κB) and Akt pathways, leading to the upregulated transcription of proinflammatory genes.³⁸ The NF-κB pathway is a central proinflammatory signaling cascade activated downstream of pattern recognition receptors such as Dectin-1. On stimulation, NF-κB translocates to the nucleus and drives the transcription of key inflammatory mediators, including tumor necrosis factor-α, IL-6, and IL-1β, as well as chemokines and costimulatory molecules. This leads to a heightened proinflammatory response, shifting Mφ toward a more responsive, proinflammatory phenotype.³⁹ During organ transplantation, NF-κB is rapidly activated in both donor and recipient tissues due to IRI and the recognition of alloantigens by innate immune receptors. This activation triggers a cascade of inflammatory responses, including the maturation of DCs and the priming of alloreactive T cells, ultimately driving graft rejection. Sustained NF-κB signaling amplifies cytokine production and immune cell recruitment. In contrast, its inhibition can attenuate these responses and promote graft tolerance.⁴⁰

The Akt pathway, which is also central to the trained immunity mechanistic, induces an important metabolic rewiring within the cell. Innate immune memory has been shown to be supported by different metabolic pathways, such as glycolysis, oxidative phosphorylation, the tricarboxylic acid cycle, amino acid, and lipid metabolism.³⁸ The modification in the normal activity of these pathways promotes the synthesis of signaling metabolites that will affect the regulation of epigenetic signatures within the cell. For instance, the accumulation of fumarate, a metabolite of the tricarboxylic acid cycle, inhibits histone demethylases such as KDM5, leading to increased histone methylation and promoting a more accessible chromatin structure for gene transcription.³⁸ These epigenetic modifications enhance the accessibility of inflammatory gene loci, enabling a more robust transcriptional response upon reexposure to pathogens. These epigenetic modifications are long-lived and will remain after the resolution of inflammation⁴¹ (Figure 1).

FIGURE 1.

The dual role of Dectin-1 and the complex metabolic pathways it induces depending on the ligand it binds. Created in BioRender. Ochando, J. (2025) https://BioRender.com/j73h547. AKT, protein kinase B; HIF-1α, hypoxia-inducible factor 1-alpha; NF-κB, nuclear factor kappa B; NOX-2, NADPH oxidase 2; ROS, reactive oxygen species; TCA, tricarboxylic acid.

ECP AS POTENTIAL TRAINED IMMUNITY-INHIBITING THERAPY TO PROMOTE TRANSPLANTATION TOLERANCE

As trained immunity can be stimulated by components of necrotic cells, whereas ECP suppresses Mφ activation through exposure to apoptotic cells, we hypothesize that apoptotic cells might antagonize training of Mφ. The clearance of apoptotic cells by Mφ, known as efferocytosis, is not a passive process that simply avoids the release of DAMPs. In contrast, apoptotic cells reprogram Mφ through soluble factors, receptor interactions, and the production of key metabolites.⁴

TLRs are expressed by monocytes, Mφ and DC, allowing them to detect PAMPs or DAMPs. TLR activation plays a central role in Mφ training; for example, TLR4 stimulation by LPS is a strong driver of trained immunity. In solid organ transplantation, chronic inflammation is perpetuated by the release of DAMPs, such as HMGB1 protein. HMGB1 is present in injured tissues and its recognition by TLR4 triggers immune inflammation and a trained phenotype in Mφ.³² In contrast to inflammatory forms of cell death, apoptosis downregulates TLR responses in Mφ and DC.⁴² Specifically, efferocytotic Mφ prevent proinflammatory responses against LPS and zymosan (a PAMP recognized by TLR2) through secretion of anti-inflammatory mediators, including transforming growth factor-beta, platelet-activating factor, and prostaglandin E2.⁴³ Inhibition of TLR signaling by apoptotic cells does not rely on blocking ligand binding; it rather affects downstream elements in TLR signaling pathways, such as NF-κB. Interestingly, recent insights into the ECP mechanistic action suggest that ECP-induced apoptotic cells migrate preferentially to inflamed tissues, where their uptake by Mφ triggers the secretion of adiponectin, a key anti-inflammatory mediator that suppresses the NF-κB signaling pathway, inducing local immune regulation.⁵

A particularly intriguing mechanism linking ECP to trained immunity involves the interaction of Dectin-1 with apoptotic cells. In in vitro settings, Dectin-1 is shown to mediate alternative signals, depending on the ligand it binds. β-glucan binding to Dectin-1 activates proinflammatory pathways such as NF-κB or Akt, promoting a proinflammatory response and inducing trained immunity. By contrast, apoptotic cells express several members of the annexin family receptors on their surface⁴⁴ able to interact with Dectin-1. Annexin A1, A5, and A13 interact with Dectin-1 and induce tolerogenic effects.⁴⁵ These annexin receptors not only bind to a different site on Dectin-1 than β-glucan but also depend on SYK early phosphorylation for the cascade they induce. Nevertheless, the SYK tyrosine phosphorylation profile that follows their binding is slightly different and results in the activation of very different downstream pathways.^46,47 Indeed, after annexin binding, the NF-κB is not triggered. In contrast, an increased production of reactive oxygen species due to increased NADPH oxidase 2 activity will follow-up, leading to an anti-inflammatory cascade that prevents NF-κB activation. Consequently, Mφ stimulated with annexin-A1, 5, and 13 shift toward a regulatory phenotype rather than a trained state⁴⁵ (Figure 1).

CONCLUSIONS

During ECP treatment, patients are exposed to large quantities of autologous apoptotic cells, which drive recipient Mφ toward a proresolving phenotype. Beyond its well-documented anti-inflammatory effects, we hypothesize that ECP generates ligands capable of blocking the training of Mφ by tissue DAMPs. Consequently, ECP treatment may limit the establishment of trained immunity in Mφ by saturating their receptors with apoptotic cell surface markers such as annexin proteins. In contrast, ECP may also favor the emergence of a Mreg phenotype, which has been associated in some studies with immunoregulatory functions. By shaping a tolerogenic environment, ECP could contribute to the differentiation or functional reinforcement of such cells within the graft.

In conclusion, ECP is emerging as a promising cell-based therapy in solid organ transplantation. The exTra consortium is actively working to expand the indications for ECP in transplantation medicine by fostering a deeper understanding of its immunological effects.^48–50 As part of exTra, we are collaborating with research groups across Europe to harmonize in vitro and in vivo methodologies for studying ECP-induced immune modulation.^51,52