Dendritic Cells in Xenotransplantation: Shaping the Cellular Immune Response Toward Tolerance

ABSTRACT

The molecular barriers that cause acute xenograft rejection have been identified and addressed by generating genetically modified (GM) animals, knocked out for specific xenoantigens (xenoAgs), and expressing regulatory molecules for both complement and coagulation pathways among others. The focus of xenotransplantation research now lies in delayed xenograft rejection. Dendritic cells (DC) are a specific subpopulation of professional antigen‐presenting cells (APC) that play a crucial role in the context of organ transplantation. DCs, originating from both the xenograft and the recipient, have the capacity to present xenoAgs to the recipient's immune system via their respective major histocompatibility complex (MHC) molecules leading to rejection. These processes are known as direct and indirect presentation, respectively. However, under certain microenvironmental conditions, DC develops into anti‐inflammatory regulatory cells that can induce immunological tolerance. The purpose of this review is to summarize current knowledge on the general characteristics and functions of DC from species relevant to xenotransplantation, specifically humans, non‐human primates (NHP), and pigs. It will also cover the process of xenoAg presentation, different methods for generating DC with regulatory properties in vitro, and finally, discuss the current strategies for using regulatory DC to improve xenograft acceptance by inducing tolerance.

1. Introduction

Xenotransplantation could potentially address the current shortage of organ donors through the transplantation of cells or organs from other species, such as pigs, into humans. The preclinical model of pig‐to‐non‐human primate (NHP) xenotransplantation is undoubtedly valuable [1, 2]. The survival times in such preclinical NHP models have significantly improved over the past years thanks to the availability of transgenic and knock‐out pigs, improved organ preservation techniques, and immunosuppressive protocols [3, 4, 5]. This has led to renewed clinical attempts at xenotransplantation, with a genetically modified (GM) pig heart and pig kidney being transplanted into living human recipients in January 2022 and March 2024, respectively [6, 7, 8]. However, despite more than 25 years of research and significant progress, pig organs are still rejected by the immune system of NHP and humans [4, 9]. Transplanted cells and tissues contain xenoantigens (xenoAgs) that trigger the recipient's immune system, leading to the destruction of the graft through humoral and cellular mechanisms. Antigen‐presenting cells (APC) play a major role in the activation of the adaptive immune system against xenografts resulting in the generation of specific T‐ and B‐lymphocytes. While cytotoxic T‐cells have the ability to infiltrate and damage the xenograft, B‐cells and plasma cells produce antibodies against xenoAgs [9, 10, 11]. As professional APC, activated DC upon uptake, process, and present xenoAgs to naïve T‐lymphocytes, priming antigen‐specific immune responses. DC are also involved in B‐cell activation during cellular graft rejection and under inflammatory conditions [12, 13, 14]. However, in certain situations, DC can regulate immunity, leading to tolerance [15, 16]. The term “regulatory DC” was coined to describe this phenomenon in vitro, while “tolerogenic DC” (_tolDC) is used when their regulatory properties are proven in vivo. For the sake of simplicity, we will use the term _tolDC interchangeably.

The promotion of xenograft acceptance at the cellular level involves inducing tolerance mediated by immune cells with the specific characteristic of shaping the immune response from inflammatory or reactive to tolerant. One example is provided by regulatory T‐cells (T_reg) which inhibit the maturation of naïve T‐cells into effector T‐cells [10, 17]. Remarkably, DC can be manipulated to promote graft acceptance in allotransplantation [18] and tolerance in autoimmune diseases [19]. In addition, it has been demonstrated, at least in vitro, that baboon _tolDC can induce a particular functional T_reg specific to pig xenoAgs [20]. The purpose of this review is to summarize the knowledge on DC, highlight the differences and similarities of DC among species, and discuss current progress in strategies using DC to promote long‐term graft survival in allo‐ and xenotransplantation.

2. Dendritic Cells

DC are a part of the innate immune system whose main function is to present antigens (Ags) to T‐cells using major histocompatibility complex Class‐I or Class‐II molecules (MHC‐I, ‐II). DC induce either immunity or tolerance, both centrally and peripherally [16]. DC also scan for pathogens or damaged tissue through pattern recognition receptors, and respond by secreting a wide range of cytokines and chemokines [21, 22]. As APC, DC capture and present Ags loaded in the MHC‐I and MHC‐II to the T‐cell receptor (TCR) and provide co‐stimulatory molecules and cytokines to naïve T‐cells, which are necessary to induce specific effector lymphocytes, either CD8 or CD4 T‐cells [21, 23], leading to their activation and proliferation [24, 25]. In turn, CD8 T‐cells with Ag specificity, also known as cytotoxic T‐lymphocytes (CTLs), destroy the targets; while CD4 helper T‐lymphocytes aid in CD8 and B‐cell maturation [26]. The Ag capture by DC through the engaging of Fc‐gamma, Fc‐epsilon, complement, and C‐type lectin receptors induce the maturation of DC [27, 28], changing their phenotype and functionality, and migration to secondary lymphoid organs where they ultimately present the Ag. During maturation, the expression of MHC‐II, co‐stimulation molecules, and secretion of pro‐inflammatory cytokines increases, while the capacity for Ag capture wanes. Other stimuli can induce the maturation of DC such as pro‐inflammatory cytokines, for example, tumor necrosis factor (TNF), granulocyte‐macrophage colony‐stimulating factor (GM‐CSF), interleukin (IL) 1β, interferons (IFNs), prostaglandin E2, and pathogen‐associated molecular patterns (PAMPs) [23, 29]. DC can activate both natural killer (NK) cells and CTLs through direct contact and various cytokines [30], providing a link between innate and adaptive immunity [31].

2.1. Role of Dendritic Cells in Xenotransplantation

In transplantation, the main source of foreign Ags is the MHC molecules from the donor, while Ags resulting from species divergence also play a role in chronic xenorejection. The direct presentation of Ag occurs when the donor's DC in the graft presents their MHC molecules to the recipient's T‐cells. This phenomenon can cause acute cellular rejection that occurs rapidly after transplantation. However, this type of presentation decreases over time as the donor's DC are eliminated. Indirect presentation happens when the recipient's DC capture donor Ags, process and present them in their MHC to the recipient's T‐cells. This presentation is responsible for delayed and chronic graft rejection [1, 32, 33, 34, 35]. Finally, in semi‐direct presentation, the donor's MHC remains intact and merge into the membrane of the recipient's APC, either through cell‐cell contact or fusion of MHC‐containing microvesicles with DC' membranes. Thus, the recipient's T‐cells recognize the donor's Ag‐loaded MHC through interaction with the recipient's DC. This is a combination of the direct and indirect presentations described above [35, 36, 37]. Taken together, in semi‐direct presentation, both the recipient's APC and the donor's MHC molecules are required.

In xenotransplantation [38], the direct presentation was demonstrated by Yamada et al. who showed that purified human T‐cells proliferate in response to porcine peripheral blood mononuclear cells (pPBMCs) in mixed lymphocyte reaction (MLR) assays. However, they proliferated to a lesser extent when pPBMCs were depleted of APC [14]. The authors showed that human T‐cell anti‐pig response was similar in specificity and strength to an alloresponse; involving both pathways of recognition, direct and indirect, in the human anti‐swine lymphocyte antigen Class II (SLA‐DR, porcine MHC‐II) responses where, the TCR repertoire, accessory molecule interactions, and cytokine production were required [14]. Moreover, Dorling et al. demonstrated the role of SLA‐DR in presenting of xenoAg in MLR assays by blocking it with anti‐SLA‐DR antibodies [39]. The indirect pathway of pig xenoAg presentation also occurs in NHP in vitro and in vivo [40]. Although semi‐direct presentation could theoretically occur, indirect evidence suggests it may occur in the context of xenotransplantation as extracellular vesicles derived from resting pECs carrying SLA Class I can directly induce CD8 T‐cell proliferation in vitro [41]. Notably, activated (pECs) also present Ags to human CD4 and CD8 T‐cells, as they express both SLA Class I and Class II, but less efficiently than DC [42, 43, 44]. It should be remembered that T‐cell‐mediated graft rejection is even more severe in xenotransplantation than in allotransplantation [39, 45]. Figure 1 shows a simplified scheme for the presentation of xenoAg by DC. Nevertheless, further investigation is required into the similarities and discrepancies between direct, indirect, and semi‐direct presentation pathways in xenotransplantation and their counterparts in human allotransplantation. Other unresolved issues include the manner in which these pathways contribute to tolerance in the context of xenotransplantation, both in solid organ and cellular settings.

FIGURE 1.

Antigen presentation by DC is a crucial process in xenotransplantation. (A) This drawing describes the hypothetical antigen presentation by DC in the scenario of pig‐to‐human solid organ xenotransplantation based on current knowledge obtained in vitro and pig‐to‐NHP studies. The scheme shows the different types of presentation in the lymph node of the human recipient. Here, pig DC present porcine peptides (seen as red circles) directly to human CD4 and CD8 cells on the left, while human DC present them indirectly on the right. In the semi‐direct presentation, human DC use pig‐derived SLA (center). The MHC molecules of the recipient (HLA) and donor (SLA) are depicted in brown and blue, respectively. Arrows represent cellular interactions, and those not yet proven in the xenotransplantation field are indicated by question marks. The diagram shows vesicles of cellular membranes carrying SLA molecules of the donor. (B) The graphic explanation illustrates the components of the scheme. This information is adapted from Sagoo et al. [37].

3. Dendritic Cells: A Comparison of Species Involved in Xenotransplantation

DC are a heterogeneous group of APC that can be classified based on their cytokine secretion, location, origin, and function [46]. The different types of DC include conventional DC (cDC), plasmacytoid DC (pDC), inflammatory/monocyte‐derived DC (iDC/moDC), follicular DC, and Langerhans cells. In this review, we have excluded follicular DC due to their low overall numbers and exclusive location in the follicular areas of lymph nodes [47]. Langerhans cells, which are specifically found in the skin [16], are not relevant for xenotransplantation unless skin grafts are involved.

The commitment and development of DC overlap with myeloid cells such as monocytes and macrophages, making it difficult to draw a clear distinction despite the use of transcription factors and surface markers [48]. This is particularly challenging when studying species other than humans or mice, as the lack of validated reagents or cross‐reactive antibodies further complicates the classification of DC. However, the most valuable information is derived from data collected by gene expression studies, which enables the comparison of DC among different species to determine whether they share similar characteristics [49, 50]. Information that later needs confirmation at the protein and function level. Tables S1–S5 provide a comprehensive comparison of the expression of surface markers, transcription factors, chemokine receptors, Pattern recognition receptors (PRR)s, cytokines, and chemokines found in human, NHP, and pig DC.

Surface markers are useful in defining each type of DC. However, they pose a significant challenge in the context of xenotransplantation, due to the structural and functional interactions between receptor/ligand orthologs expressed in APC and immune effector cells. These interactions can trigger or not cellular signaling, leading to functional outcomes, aspects that remain to be clarified or studied in each of the interspecies pairs. For instance, the consequences of incompatibility between recipient CD172a^high cDC2 in humans or NHP with passenger leukocytes expressing porcine CD47 remain to be elucidated due to the demonstrated incompatibilities between human CD172a and porcine CD47 [51].

Generally, cDC are of myeloid origin and are found in both lymphoid and non‐lymphoid tissues. In humans, NHP, and pigs, there are two types of cDC, namely cDC1 and cDC2. Both types of cDC have the capacity to respond to PAMPs and damage‐associated molecular patterns, but they do so differently based on their receptor expression and cytokine production patterns. Human, NHP, and pig cDC1 share the expression of markers CADM1^high, CD172a (SIRPα)^low, CD11c^int/high, and CD26^high. Additionally, humans and NHPs express the transcription factor IRF8^high and have low levels of IRF4 [52], while BATF3 transcription factor is significantly present at the transcriptional level in pigs [49]. Regarding cytokines, humans typically exhibit high levels of type‐III IFN‐λ, while TNF is common in both humans and pigs [48, 49, 53]. Conversely, in humans, cDC2 are characterized by being CADM1^low, CD172a^high, CD11c^high, and IRF4^high [52]. However, some surface markers are not shared among species or have not yet been identified.

Human and NHP pDC are characterized by being CD11c^low, IFR8^high, and IRF4^int [52]. CD45 is present on NHP pDC [52], while humans and pigs have CD45RA on their surface [48, 54]. In pigs, the CD172a marker is used for the identification of pDC. Notably, pDC produce large amounts of type‐I IFN (α and β) during viral infection [55]. Pig pDC have only TCF4, NRP1, and RUNX2 transcription factors at the messenger level [49]. (Table S3).

There is limited information available on iDC/moDC migration to lymphoid or non‐lymphoid organs during infections or inflammation in our species of interest. This is due to the difficulty in characterizing iDC/moDC based on their surface markers, the substantial heterogeneity among them, and the challenge of distinguishing them from macrophages and monocytes. This fact was elegantly reviewed by Guilliams [52]. In humans, iDC/moDC share the expression of HLA‐DR, CD11c, BDCA1, CD1a, FcɛRI, CD206, CD172a, CD14, and CD11b [56]. However, there is limited knowledge about iDC/moDC in NHP and pigs.

4. Regulatory/Tolerogenic Dendritic Cells Are Used in the Field of Cell Therapy

DC play a crucial role in inducing tolerance. Immature DC are responsible for the negative selection of self‐reactive T‐cells in both the thymus and inflammation sites when APC capture self‐Ags from dead cells [57, 58]. This occurs because immature DC present Ags to T‐cells without the necessary co‐stimulation and cytokines, leading to anergy and the generation of T_reg [21]. A particular type of T_reg that plays a role in peripheral tolerance are the Tr1 cells. These cells are of interest in the context of transplantation due to their ability to be generated by immature monocyte‐DC that secrete IL10 within the allo‐setting. A distinguishing feature of Tr1 cells is their deviation from the conventional T_reg phenotype, as evidenced by the absence of FoxP3 and high levels of CD25 expression. Nonetheless, these Tr1 cells exert their regulatory function by mechanisms that involve both IL10 and transforming growth factor‐beta (TGFβ), thereby facilitating the suppression of T‐cell responses [59].

In addition, there are DC with regulatory or “tolerogenic” properties, known as _tolDC. These cells are resistant to maturation, express relatively low numbers of MHC molecules, and have fewer co‐stimulatory molecules than other types of DC [31]. As a result, _tolDC induce T‐cell apoptosis or T‐cell anergy by inhibiting their activation [60] through both cell contact‐dependent and ‐independent mechanisms [16]. On their surface, _tolDC express several inhibitory molecules, such as programmed death‐ligand 1 and 2 (PD‐L1, PD‐L2), non‐classic MHC‐I known as HLA‐G, Ig‐like transcript‐2, ‐3, and ‐4 (ILT‐2, ‐3, ‐4), TNF‐related apoptosis‐inducing ligand (TRAIL), galectin‐1, or Fas ligand (Fas‐L/CD178). Tolerogenic DC also secrete anti‐inflammatory cytokines such as IL10 and transforming growth factor‐beta TGFβ, as well as enzymes like indoleamine 2,3‐dioxygenase (IDO), heme oxygenase‐1, inducible Ca²⁺‐insensitive nitric oxide synthase, and arginase‐1 [16, 22, 61, 62]. Tolerogenic DC induce the differentiation of naïve CD4 T‐cells into T_reg by receptor‐ligand interactions between CTLA‐4 on CD4 T‐cells and CD80/86 on DC [20, 21, 63, 64].

Cellular therapy using _tolDC represents an alternative to immunosuppressive drugs (ISDs) for controlling T‐cell activation and proliferation in transplantation and autoimmunity [16]. Tolerogenic DC with regulatory properties have already been applied for autoimmune diseases [65]. Each of these studies demonstrated a positive impact, particularly by augmenting the quantity of regulatory lymphocytes [66]. Recent reviews on the clinical use of DC in human organ allotransplantation are also available [15, 67, 68], supporting the use of _tolDC to modulate the recipient's immune response in xenotransplantation.

4.1. Approaches to Obtain Tolerogenic Dendritic Cells In Vitro

There are various protocols available for generating human DC with tolerogenic or regulatory properties. These protocols involve the use of pharmacological agents, cytokines, growth factors, or hormones to render DC resistant to maturation or to induce a regulatory phenotype (refer to Table 1). However, notably, a given agent may not necessarily generate _tolDC with the same phenotype across different species. For example, IL10 appears to have a similar effect on human and pig DC, but the expression level of TNF remains unchanged in pig _tolDC [60]. Additionally, GM porcine DC expressing human TRAIL has been reported to attenuate human T‐cells [69].

TABLE 1.

Features of in vitro differentiated tolerogenic dendritic cells for various applications.

Type of agent	Agent	Surface molecules expression a	Main cytokines secreted a	Context	References
Cytokines	IL10	↑ HLA‐G, ILT4 ↓ CD40, CD86	↓ IL12 ↑ IL10, TGFβ		[23, 70, 71]
	TGFβ	= MHC‐II ↓ CD40, CD80, CD86	↓ IL12, TNF ↑ TGFβ		[22, 23, 72, 73]
	IL3	= CD40 = CD83, CD86 ↓ CD1a	↓ IL6, IL12, TNF ↑ IL10	In vitro, allogeneic stimulation	[74]
Growth factors	VEGF	↓ MHC‐II, CD8	↓ IL12	Cancer	[75]
	HGF	↑ PD‐L1 ↑ GILZ	↑ IL27	In vitro	[76]
Metabolic/vitamins	1‐Alpha,25‐dihydroxy vitamin D3	↓ MHC‐II ↓ CD40, CD80, CD86	↓ IL12 ↑ IL10, IL35	Allotransplantation	[77, 78, 79, 80]
	Retinoic acid	↓ CD83, CD86 ↓ MHC‐II, ↑ ILT3	—	In vitro	[81]
Pharmacological	Cyclosporine A			In vitro/MLR	[82]
	Tacrolimus			In vitro/MLR	[83]
	Mycophenolate mofetil			Hypersensitivity responses/MLR	[84]
	Rapamycin	↓ MHC‐II ↓ CD40, CD80, CD83, CD86	↑ IL12 ↓ IL10	In vitro	[85, 86, 87]
	Dexamethasone	↑ MHC‐II ↓ CD40, CD83, CD86	↓ IL12, IL18, TNF ↓ IL10	Allotransplantation/ In vitro	[88, 89, 90, 91]
Hormone/lipid compounds	VIP			In vivo	[92, 93]
	PG‐E2				[34]
Genetic manipulation	CTLA‐4‐Ig expression				[69]

Abbreviations: GILZ, glucocorticoid‐induced leucine zipper; HGF, hepatocyte growth factor; PG‐E2, prostaglandin E2; VEFG, vascular endothelial growth factor; VIP, vasoactive intestinal peptide.

^a Differences compared to cDC.

From the perspective of xenotransplantation, the use of recipient‐derived _tolDC offers the advantage of compatibility, meaning that they will not be rejected and may potentially last longer than donor‐derived DC. Therefore, we aim to compare human DC generated in vitro from monocytes to those of other species, including those used in pre‐clinical animal models involving NHP. In addition, porcine donor‐derived DC are of interest as they may serve as passenger cells from the transplanted organ, which will be relevant for the direct pathway of Ag presentation.

Tolerogenic DC can be generated in vitro from various cell sources, including DC, monocytes, CD34^pos cells, stem cells, and bone marrow. In humans, _tolDC are primarily derived from monocytes [16]. The cytokines and growth factors required for _tolDC generation differ depending on the cell source. For human monocyte‐derived _tolDC, successful combinations include GM‐CSF, TNF, and IL4, or Flt3L, GM‐CSF, and stem‐cell factor, among others [94] (refer to Table 2 for details).

TABLE 2.

Protocols for in vitro differentiation of tolerogenic dendritic cells for species of interest.

	Source	Basic differentiating cocktail for DC	Main phenotype of mature tolDC a	References
Humans	Monocyte CD34+ precursor	GM‐CSF + IL4 + TNF GM‐CSF + Flt3L GM‐CSF + IL3/4/6, Flt3‐L, and SCF GM‐CSF + IL3	CD14neg, CD1a, DC‐SIGNhigh, b MHC‐II low/neg, PD‐L1, CD11chigh, CD40, CD80low, CD86 low (For cytokine production, Table S3) Resistance to maturation by LPS Inhibition T‐cell proliferation Inhibition IFNγ secretion by T‐cells Induction of Treg	[30, 74, 88, 94, 95]
NHP	Monocyte CD14+ monocytes PB stem CD34+ cells	GM‐CSF +/˗ IL4 GM‐CSF + IL4 or GM‐CSF + IL3 GM‐CSF + IL4 + IL10 + TGF‐β1 GM‐CSF + IL4 + IL10 + Vit D3 GM‐CSF + IL4 +/˗ TNF, TPO, Flt3‐L, and SCF	Elevated Ag uptake ability and migration CD14pos, c CD11chigh, CD1alow/neg c DC‐SINGlow/neg, c CD83low c Variable response to LPS Generation of xenoreactive Treg Resistant to maturation. Fail to support MLR in vitro. In vivo, tolDC + CTLA4‐Ig modulates allogeneic T‐cell responses (donor and third party) CD14pos, CD11cpos	[20, 95, 96]
Pigs	Monocyte	GM‐CSF + IL4 + IL10 + IL10 + IL3 GM‐CSF + IL13	d CD14 low/neg MHC‐IIlow, = antigen capture, ↑ IL10, no IL12 or IL6 High uptake Ag, upregulation CD80/CD86, strong naïve T‐cell stimulation	[60, 97, 98, 99]

Abbreviations: Ag, Antigen; BM, bone marrow; CTLA‐4, cytotoxic T‐lymphocyte associated protein 4; E2; Flt3‐L, Fms‐related tyrosine kinase 3 ligand; GILZ, glucocorticoid‐induced leucine zipper; GM‐CSF, Granulocyte‐macrophage colony‐stimulating factor; HGF, hepatocyte growth factor; IFNγ, interferon gamma; IL, interleukin; LPS, Lipopolysaccharide; MHC‐II, major histocompatibility complex Class II; MLR, Mixed lymphocyte reaction; PB, peripheral blood; PG‐E2, prostaglandin SCF, stem cell factor; TNF, tumor necrosis factor; TPO, thyroperoxidase; VEFG, vascular endothelial growth factor; VIP, vasoactive intestinal peptide; Vit D3, Vitamin D3.

^a The main phenotype compares the differences to cDC.

^b Depends on the agent used to generate _tolDC, generally low.

^c Highly expressed in certain primates DC.

^d Variability given by the family.

4.2. Characteristics of In Vitro–Generated DC and _tolDC

4.2.1. Humans

A comparative study by Naranjo‐Gómez et al. showed that human monocyte‐derived DC produced by the addition of rapamycin (Rapa‐DC), dexamethasone (Dexa‐DC), or vitamin D3 (VitD3‐DC) did not share the same characteristics. Specifically, VitD3‐DC exhibited a lower yield and viability compared to other types of DC. Additionally, Rapa‐DC presented a phenotype comparable to mature DC, while VitD3‐DC and Dexa‐DC shared an immature DC phenotype. Regarding cytokine production, only Dexa‐DC and, to a lesser extent, VitD3‐DC produced IL10 but not IL12p70. Dexa‐DC produced six times more IL10 than a mature DC. Most importantly, the three types of _tolDC were resistant to maturation when exposed to lipopolysaccharides (LPS) and inhibited proliferation and IFNγ secretion of T‐cells. However, only Rapa‐DC induced the differentiation of CD4 T‐cells into T_reg [88]. In another comparative study, Madelon et al. analyzed the characteristics of human Rapa‐DC and IL10‐DC derived from monocytes under basal conditions and stimulated with LPS. Once again, the expression levels of co‐stimulatory molecules CD83 and CD86 in _tolDC were lower compared to cDC, while there was a higher expression of PD‐L1. In Rapa‐DC, the level of co‐stimulatory molecules was equivalent to that of cDC, and the level of co‐inhibitory PD‐L1 was lower. In contrast, both _tolDC and cDC expressed similar amounts of MHC and ILT‐2, ILT‐4, HLA‐E, and HLA‐G [30]. However, monocyte‐derived DC differentiated by IL3 resulted in _tolDC that showed lower secretion of IL12, IL6, and TNF, but similar expression of co‐stimulatory molecules. T‐cells activated by these IL3‐derived _tolDC secreted more IL5 and IL4 and less IFNγ than T‐cells activated by cDC, indicating a transition from a Th1 to a Th2 phenotype [74].

4.2.2. Non‐Human Primates (NHP)

Here, our focus will be on the Old‐World NHP DC as they are more closely related to human DC and the animal size is “closer” to humans. In NHP, DC can be differentiated from monocytes, peripheral blood stem cells, or bone marrow cells. Jesudason et al. performed a comprehensive review of DC from various NHP used in functional studies. They compared the DC's precursors, maturation, expression of MHC and co‐stimulatory molecules, ability to induce T‐cell proliferation, MLR, cytokine production profile, and phagocytosis [95].

The morphologies of NHP DC generated in vitro are equivalent to those of humans. However, there are differences in surface molecular markers. For instance, CD14 expression persists in NHP after DC maturation, unlike in humans. CD11c can be used as a primary marker for DC because it is strongly expressed in some, but not all primate species, while CD1a is a marker for human DC but is poorly expressed in NHP. Despite a strong homology between NHP and humans, DC‐SIGN is only very rarely expressed on NHP but is strongly expressed on human myeloid DC. However, African green monkeys and Chinese macaques express this marker on DC derived from monocytes. CD83 is another maturation marker found on most primate DC in the Old‐World primate group. Finally, there are differences and similarities between the DC maturation profiles of different species. Some species’ monocytes‐derived DC react to LPS stimuli, while others do not [95].

4.2.3. Swine

In pigs, cells with DC characteristics were generated in vitro by culturing PBMC in the presence of IL4 and GM‐CSF. These DC are SLA Class‐II positive and CD14 negative to low, which is a co‐receptor for LPS, and CD1a low [60]. Further culturing with IL10 or IL3 drives them to a phenotype similar to less mature DC, with fewer SLA Class‐II molecules and better Ag capture capacity, which is characteristic of _tolDC [97]. By differentiating porcine DC in the presence of IL10, these DC produce IL10, with little or no IL12 and IL6. This results in a lower ability to stimulate T‐cell proliferation compared to cDC [60].

5. Potential of Tolerogenic DC in Xenotransplantation

Starting with what is known in clinical allotransplantation, ISDs, including corticosteroids, calcineurin inhibitors, basiliximab, rapamycin, and mycophenolate mofetil, are effective in preventing acute graft rejection. However, ISD therapy does not prevent chronic organ dysfunction [16, 100]. An alternative to unspecific general immunosuppression and its side effects is the use of regulatory cells such as _tolDC. Cellular therapy has the potential to inhibit alloresponses without affecting responses to other pathogenic Ags. It can have a long‐lasting effect [16, 101], as demonstrated by the prolongation of skin, heart, and islet of Langerhans allograft survival using _tolDC in several models [72, 102, 103, 104]. For instance, the use of Rapa‐DC in heart transplantation has indefinitely prolonged survival in mice [104]. The survival of islet grafts was promoted by TGFβ‐conditioned DC [72]. Autologous _tolDC were found to prolong skin allograft survival in mice by inducing regulatory CD8 T‐cells [105]. In an NHP model of kidney allotransplantation, donor‐derived _tolDC were injected into the recipient before transplantation and significantly prolonged graft survival. The study found that _tolDC induced apoptosis in CD95^pos memory T‐cells and reduced the stimulation of allogeneic T‐lymphocytes [106].

At the International Workshop on Clinical Immune Tolerance, the University of Pittsburgh presented the latest findings, indicating that approximately one‐third of patients who underwent liver allotransplantation were successfully weaned off immunosuppression in a clinical trial involving _tolDC derived from living donors [107]. The ONE study consortium investigated various cellular approaches, such as T_reg, DC, and monocytes, in human organ transplantation [108]. The University of Nantes, as a member of this consortium, demonstrated positive pre‐clinical results in kidney allotransplantation using autologous _tolDC. The tolerogenic cellular therapy was safe, mycophenolate was discontinued in five out of eight patients, while two remained on monotherapy. Therapy with _tolDC was associated with changes in the blood, as evidenced by a decline in the activation status of CD8 T‐cells and transient expression FoxP3 [107]. Nevertheless, graft rejection occurred at similar rates between the study groups. Notably, patients experienced fewer episodes of infections, suggesting that the approach could improve patients’ immunosuppression side effects [109]. Currently, other clinical trials using _tolDC are underway [110]. However, the therapeutic potential of _tolDC in xenotransplantation has been explored in only a limited number of studies. To achieve graft survival, it is crucial to prevent direct, indirect, and semi‐direct presentation of xenoAgs. Both donor and recipient cells can be manipulated prior to transplantation to prevent xenorejection [60].

5.1. In Vitro Studies in Xenotransplantation

In vitro, porcine DC cultured with IL10 and IL3 showed reduced xenogeneic T‐cell response in human PBMC [60]. Modifying donor‐type DC has the potential risk of promoting acute graft rejection. This was demonstrated by Smyth et al., who showed that modified donor‐type DC was given to mice recipients before heart or skin allotransplantation induced rejection. Donor‐type DC have a short lifespan in vivo, resulting in the recipient's APC capturing and presenting the alloAg [111]. However, the activation of human‐specific T‐cells can be prevented by using donor‐type DC that expresses human TRAIL (huTRAIL), which induces apoptosis in proliferating and activated human T‐cells via the death receptor [69]. This finding suggests that transgenic pigs carrying huTRAIL may be useful for xenotransplantation. Porcine huTRAIL‐DC may migrate into the recipient's secondary lymphoid organs and inhibit the activation of human‐specific T‐cells during direct presentation. Similarly, another in vitro study showed that co‐culturing GM pig fibroblasts expressing human Fas‐ligand (FasL) with human _tolDC, which were generated by culturing with dexamethasone, induced apoptosis of human PBMCs, especially activated CD4 T‐cells [112].

Regarding the use of recipients’ DC, an in vitro study demonstrated that unmodified DC, derived from human monocytes, induced specific xenogeneic responses of autologous CTLs in the presence of pECs as a source of xenoAgs. Specifically, there was an increase in IFNγ production and xenogeneic cytotoxicity. In contrast, IL10‐generated _tolDC, showed reduced IFNγ production and xenogeneic cytotoxicity by CTL and NK cells [30]. As noted, _tolDC have the capacity to induce T_reg. In another study by Li et al. baboon peripheral blood monocytes were differentiated into _tolDC using GM‐CSF, IL4, TGFβ1, and IL10. The _tolDC expressed MHC‐I and ‐II, negative regulators of T‐ and B‐cells (PD‐1, B7‐H1, and B7‐DC), less CD83, fewer co‐stimulation molecules (CD80, CD86, and CD40), and produced anti‐inflammatory cytokines. These _tolDC subsequently induced T_reg [20], which notably inhibited the activity of xenogeneic T‐effector cells specific for pig Ags, including proliferation, IFNγ production, and CD154 (CD40L) expression. Also, targeting of CD86 by siRNA was sufficient for DC to inhibit lymphocyte proliferation in vitro; production of TGFβ, IL10, and IDO was increased while IFNγ and IL2 decreased in these CD86^neg DC [113]. In summary, in vitro studies indicate that _tolDC may be a promising approach for tolerance induction in xenotransplantation.

5.2. In Vivo Studies in Xenotransplantation

Small animal xenotransplantation models have been used in most studies of _tolDC. Research has shown increased survival and function in a rat‐to‐mouse islet transplantation model due to the induction of CD4⁺CD25⁺ T_reg when diabetic mice were treated with syngeneic DC lacking CD86 before transplantation of rat islet xenografts [113]. Recently, the potential of autologous bone marrow‐derived murine DC to protect rat‐to‐mouse islet xenografts was analyzed. Tolerogenic DC were generated using GM‐CSF and IL10. In comparison to control DC, these _tolDC showed reduced levels of MHC‐II and costimulatory molecules (CD40, CD86, CD205), and lower production of pro‐inflammatory cytokines (IL‐12p70, TNF, IL6), and higher production of IL10. The survival of rat islets xenograft in diabetic mice co‐transplanted with autologous murine _tolDC was significantly prolonged in the absence of immunosuppressive treatment which may be due to the local induction of immune regulatory cells, CD8⁺ T‐cells, and myeloid cells with suppressor‐associated phenotype [114].

A different approach was explored to investigate the potential of CTLA‐4 in blocking co‐stimulation mediated by CD86/CD80 interaction with CD28. Additionally, CTLA‐4 induces the synthesis of IDO in DC, which in turn leads to the apoptosis of effector T‐cells [115]. Tian et al. modified the donor porcine monocyte‐derived immature DC to express pig CTLA‐4 fused to human IgG4‐Fc on their surface. These engineered DC were injected into insulin‐dependent diabetic mice before pig islet xenotransplantation. These immature DC expressing pCTLA4‐IgG4 were able to differentiate mouse CD4 T‐cells into T_reg and to prolong xenograft survival [115].

In the field of other APC not involving DC, Miyagawa's group has described the use of a heterogeneous population of progenitor and immature human myeloid cells, known as myeloid‐derived suppressor cells (MDSCs), to achieve long‐term graft survival and tolerance in a murine bone marrow allotransplantation model, as well as in pig‐to‐human xenogeneic responses in vitro. MDSCs were found to suppress the cytotoxicity of human CTLs against pECs through phagocytosis of CTLs and mechanisms dependent on IDO [116]. Additionally, MDSCs were found to regulate macrophage‐associated xenogeneic cytotoxicity in a contact‐dependent manner [117].

6. Conclusions and Perspectives

Cell therapy has been used in the clinic for many years and has been proven effective in numerous domains. DC are recognized as highly efficient APC that elicit adaptive immune responses. However, DC are also capable of delivering suppressive signals to T‐cells, which may contribute to long‐term xenograft acceptance and survival by promoting T_reg. Tolerogenic DC have been successfully utilized in allotransplantation models to prevent chronic graft rejection [107]. However, the protocols to be used for _tolDC therapy in the clinical setting of xenotransplantation still face several challenges. The initial consideration is the origin of _tolDC, whether from the donor (pig) or the recipient (NHP or patient). Additionally, uncertainty persist regarding whether DC derived from CD34 hematopoietic precursors or monocyte‐derived DC yield superior results, exhibit enhanced expansion, and possess superior regulatory properties. Furthermore, it is imperative to ascertain the most efficacious formula of pharmacological agents, cytokines, growth factors, or hormones in the induction of DC with regulatory characteristics. A secondary issue pertains to the protocol, namely the optimal delivery method for _tolDC, whether cellular infusion, with the xenograft, or both. This encompasses the determination of the optimal number of _tolDC and the optimal timing of administration, that is, before, during, or after transplantation. Moreover, it is imperative to ascertain whether _tolDC require ex vivo pulsing with xenoantigens or if they are capable of capturing them in vivo. The third aspect relates to the fate of _tolDC once administrated to the recipient, since the effectiveness of the therapy relies on the survival of the _tolDC and xenograft. Tolerogenic DC may experience quick destruction before they can exert their tolerogenic properties, or they may acquire an inflammatory phenotype once in the recipient, thus accelerating rejection of the xenograft. However, preclinical xenotransplantation models have demonstrated that _tolDC may induce T_reg, leading to peripheral tolerance. A number of unanswered questions regarding the potential clinical benefits in comparison to potential adverse effects remain: Will be the effect long‐lasting or transient? Will the patient require fewer or similar ISDs compared to allotransplanted patients, or none at all? Will the patient's quality of life be improved? (Figure 2). Consequently, NHP serves as an indispensable preclinical model for evaluating such strategies, with mounting evidence suggesting that _tolDC infusion enhances kidney allograft survival in NHP. There is a wide range of strategies to delay or prevent the xenograft rejection including the use of mixed chimerism [9]. The use of cell therapy by the administration of recipients’ _tolDC in xenotransplantation models represents an alternative approach and already has demonstrated encouraging outcomes [114]. This approach, combined with less aggressive immunosuppressive protocols, could contribute to the clinical feasibility of xenotransplantation and potentially address the issue of donor shortage transplantation medicine.

FIGURE 2.

Challenges in the utilization of tolerogenic DC in xenotransplantation. (A) Source of _tolDC: donor (pig)‐ or the recipient (NHP or patient)‐derived; CD34 hematopoietic precursors or monocyte‐derived DC yield _tolDC render improved expansion and regulatory properties; "tolerogenic agent” e.g. interleukin 10 (IL10), rapamycin, or Vitamin D. (B) Protocol to deliver _tolDC: in all cases, the number of _tolDC needed. In the case of cellular infusion, the optimal timing for the infusion and application scheme are necessary. If _tolDC are to be applied with the graft, the method of administration and whether complementation with infusion are required. It should be considered whether _tolDC will be pulsed with xenoantigens or allowed to capture them in vivo. The effectiveness of the therapy relies on the survival of the _tolDC and xenograft. (C) Fate of _tolDC once in the recipient. It is possible the conversion of _tolDC into an inflammatory phenotype once in the recipient, accelerating xenorejection. As well, _tolDC may experience a quick death or destruction before they can exert their tolerogenic properties. However, preclinical xenotransplantation models have demonstrated that _tolDC may induce regulatory T‐cells (T_reg), leading to peripheral tolerance. (D) Clinical benefits. Will be the cell therapy long‐lasting or temporal? Will the patient require fewer or similar immunosuppressive drugs (ISDs) compared to allotransplanted patients, or none? Has the patient's quality of life improved?

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting Information