Evidence for the important role of inflammation in xenotransplantation.

There is increasing evidence of a sustained state of systemic inflammation after pig-to-nonhuman primate (NHP) xenotransplantation (that has been termed systemic inflammation in xenograft recipients [SIXR]). Increases in inflammatory markers, e.g., C-reactive protein, histones, serum amyloid A, D-dimer, cytokines, chemokines, and a decrease in free triiodothyronine, have been demonstrated in the recipient NHPs. The complex interactions between inflammation, coagulation, and the immune response are well-recognized, but the role of inflammation in xenograft recipients is not fully understood. The evidence suggests that inflammation can promote the activation of coagulation and the adaptive immune response, but the exact mechanisms remain uncertain. If prolonged xenograft survival is to be achieved, anti-inflammatory strategies (e.g., the administration of anti-inflammatory agents, and/or the generation of genetically-engineered organ-source pigs that are protected from the effect of inflammation) may be necessary to prevent, control, or negate the effect of the systemic inflammation that develops in xenograft recipients. This may allow for a reduction in the intensity of exogenous immunosuppressive therapy. If immunological tolerance to a xenograft is to be obtained, then control of inflammation may be essential.


Introduction
Organ transplantation is one of the medical success stories of the past 70 years, but there remain insufficient organs from deceased human donors to treat all of the patients who might benefit. For example, in the USA at present there are approximately 120,000 patients awaiting an organ of one sort or another, and yet this year only approximately 10,000 deceased human donors will become available, providing an average of three or four organs per donor [1].
The lack of human organs could be obviated if a suitable animal source of organs were available. For a number of logistic and other reasons, the pig has been identified as a potential source of organs for clinical transplantation [2]. The field of xenotransplantation (cross-species transplantation) has therefore been extensively investigated during the past 35 years [3]. Although organs from wild-type (i.e., genetically-unmodified) pigs transplanted into humans or nonhuman primates (NHPs) are rejected within minutes [4], our ability to genetically-engineer the pig to protect its organs from the primate immune response has resulted in life-supporting kidney or heart graft survival in NHPs extending to many months or even more than a year [5][6][7][8][9]. One of the barriers that has had to be overcome, but continues to be problematic, is the inflammatory response to the presence of a pig organ.
Inflammation is part of the complex biological response of body tissues to harmful stimuli, and is observed in various diseases, e.g., inflammatory disease [10], infection [11], atherosclerosis [12]. The release of appropriate pro-inflammatory cytokines and chemokines is necessary for protective immunity, but production of these factors in excess can result in various pathological states [13]. An inflammatory response follows ischemia-reperfusion injury after organ transplantation [14]. This may play an important role in initiating the allo-immune response [15], and in the development of allograft vasculopathy [16].
We here review the evidence of a prolonged systemic inflammatory response to a xenograft, and consider what steps can be taken to prevent or reduce it. We have primarily drawn on our own observations, but have supplemented these by a review of the literature.
Evidence for a sustained inflammatory response in xenograft recipients (SIXR) ( Table 1) C-reactive protein (C-RP) is an acute phase protein synthesized largely by hepatocytes in response to proinflammatory cytokines, in particular interleukin-6 (IL-6) [31]. C-RP provides the first line of defense to an invasive pathogen, and can promote activation of complement, bacterial capsular swelling, and phagocytosis [32]. It is a marker of early infection, and provides an easy objective parameter [33]. Moreover, C-RP mRNA expression increases in the presence of acute rejection of a renal allograft [34]. C-RP can contribute both to host defense against infection and enhancement of inflammatory tissue damage.
After pig-to-baboon organ transplantation, C-RP is increased for several months, suggesting a persisting inflammatory state [13,19,26] (Fig. 1a), and is deposited in the transplanted pig kidney [18] (Fig. 1b). Whether this is secondary to initial antibody binding remains uncertain.
After pig-to-baboon organ xenotransplantation, significant increases in SAA have been observed during antibody-mediated rejection (Fig. 2) or when a consumptive coagulopathy or infection is developing [26,27]. Amyloid A is deposited in the transplanted pig kidney [28]. Although the current method of measuring SAA is not fully quantitative, it is a simple and rapid indicator of the inflammatory state, allowing early investigation, e.g., for rejection, infection, or other complications.
The direct prothrombotic activity of histone-DNA complexes increases inflammatory cytokine formation, and fosters thrombotic responses by activating TLRs 2, 4, and 9 [48]. Moreover, inflammatory cytokines downregulate thrombomodulin, induce tissue factor, and upregulate plasminogen activator inhibitor [48]. Histones can also cause direct platelet activation [53,58]. Their levels increase in xenograft recipients when there is evidence for inflammation and coagulation dysfunction [26]. In the absence of IL-6-receptor blockade (with tocilizumab), the mean serum histone level after pig organ transplantation rises significantly [26] (Fig. 3a). A decrease in the number of neutrophils might reduce extracellular histone release [59,60]. In in vitro studies, histone-induced porcine EC apoptosis/death was significantly reduced by an inhibitor of nuclear factor kappa B (NF-κB), parthenolide (Fig. 3b) [26]. EC apoptosis is observed in many inflammatory and immune disorders [61].
D-dimer is a protein product of cross-linked fibrin degradation. An elevated blood concentration of Ddimer is observed in intravascular coagulation and thrombotic disease [62]. D-dimer may promote the inflammatory cascade by activating neutrophils and Fig. 1 a C-RP in baboons with pig artery patch (n = 9) or organ (n = 17) grafts. Levels of C-RP in baboons before (day 0) and after pig organ or artery patch transplantation. (Black line = without tocilizumab therapy; Red line = with tocilizumab therapy.) The mean level of C-RP in the tocilizumab-treated baboons remained < 0.5 mg/dL from day 4, which (on days 7, 14, 28, and 60) was significantly lower than in baboons not receiving tocilizumab (day 7, 0.2 vs 1.6 mg/dL, P < 0.001; day 14, 0.3 vs 1.8 mg/dL, P < 0.001; day 28, 0.3 vs 1.6 mg/dL, P < 0.001; and day 60, 0.3 vs 2.0 mg/dL, P < 0.01, respectively). (**P < 0.01; ***P < 0.001). Tocilizumab treatment therefore prevented an increase in C-RP after xenotransplantation. (The rise in C-RP on day 136 in one of the baboons in the tocilizumab-treatment group was associated with the onset of systemic infection.) (Reprinted with permission from ref. [26]). b C-RP deposition in pig kidneys transplanted into baboons, an indicator of the inflammatory response to the graft. (Left panel) At 30 min after reperfusion of an α1,3-galactosyltransferase gene-knockout (GTKO) pig kidney graft, no C-RP deposition was detected. In two different kidneys at the time of euthanasia (middle and right panels), C-RP deposition was detected in the glomeruli (arrow heads, right panel) and tubules (arrows, middle and right panels). Our data suggest that both the xenograft and the recipient contribute to C-RP production. (We detected minimal C-RP in NHPs undergoing heart allotransplantation [not shown]). (Reproduced with permission from ref. [18]) monocytes, inducing secretion of inflammatory cytokines (e.g., IL-6) [62][63][64][65].
In the absence of immunosuppressive therapy, increases in certain cytokine levels are seen after xenotransplantation, but not when immunosuppressive therapy is administered [18] (Fig. 5a,b).
Inflammation plays a key role in platelet activation and aggregation [71], which in turn plays an important role in the dysregulation of coagulation seen after xenotransplantation [72]. Extracellular histones bind to TLRs, particularly to TLR2 and TLR4, on platelets, which results in platelet aggregation [51,53]. In humans, the cytokine, IL-17, can promote platelet activation and aggregation through the ERK2 and P53 signaling pathways [73,74], although the exact mechanism remains unclear [75]. Recipient platelets might also be activated by binding directly to pig ECs [76]. Human platelets can upregulate tissue factor expression after contact with pAECs in the absence of human serum or antibodies, which can lead to coagulation through thrombin production [77].
In recipient baboons undergoing pig heart, kidney, liver, and artery patch xenotransplants, fT3 falls rapidly, and takes several days to return to pre-transplant levels [26] (Fig. 6). A negative correlation between serum IL-6 and TNF-α with thyroid hormone concentrations has been reported [80]. A persisting low level is almost certainly associated with an inflammatory response to a xenograft [26].

Evidence for the relationship between inflammation and coagulation in xenograft recipients
Until recently, a major barrier to successful pig organ transplantation in NHPs was dysregulation of coagulation resulting from excessive thrombin generation [90][91][92][93]. The activation of thrombin receptors amplifies production of the chemokine, CCL18, and the pulmonary activation-regulated chemokine by mature dendritic cells [94]. Thrombin can upregulate ICAM-1 mRNA and induce ICAM-1 expression on monocytes in vitro [95], and by activating NF-κB [96].
It is well-known that inflammation contributes to activation of coagulation dysfunction [17,18,70,97,98]. Tissue factor is not only a promoter of thrombin, but also a marker of inflammation [99,100]. TNF-α [101], IL-6 [102], and C-RP [103] increase tissue factor expression on innate immune cells, which in turn promotes the activation of coagulation [100,103]. There is an amplification circuit between coagulation and inflammation which results in activation of inflammatory mediators as well as procoagulant factors [20]. Therefore, therapeutic prevention of inflammation may be a major factor in minimizing coagulation dysregulation after pig organ xenotransplantation.
An important observation made recently indicates that, when pig vascular ECs expressing only natural pig thrombomodulin (which also has an anti-inflammatory effect) are activated by TNF-α, the expression of thrombomodulin is significantly downregulated (Fig. 7a) [98]. This suggests that, when a pig organ is exposed to inflammation (which is universal after a pig organ transplant into a NHP), thrombotic microangiopathy is likely to develop. The absence of the anti-inflammatory effect of human thrombomodulin may result in the early development of consumptive coagulopathy [6]. In contrast, transgenic expression of human thrombomodulin is not downregulated, thus maintaining both its anticoagulant and anti-inflammatory effects (Fig. 7b) [98].

Evidence for the relationship between inflammation and the immune response in xenograft recipients
The significant increase in certain cytokines/chemokines after xenotransplantation likely results from innate immune cell activity, and may well be a causative factor in xenograft injury [17,18]. Inflammation and the innate immune response augment the adaptive immune  [26]). b In vitro histone-induced porcine endothelial cell apoptosis/death is influenced by NF-κB inhibition. The NF-κB inhibitor, parthenolide (at 2 and 8 μM), significantly reduced histone (160 μg/ mL)-induced cell apoptosis/death (mean percentage apoptosis/death of 91.4% vs 54%, respectively; both P < 0.05). There was no significant difference in the protective effect of parthenolide at concentrations of 2 and 8 μM (mean percentage apoptosis/death of 54% at both concentrations). (Reprinted with permission from ref. [26]) response [70,98]. Systemic upregulation of inflammatory markers is related to inefficient blockade of the T cell-dependent adaptive immune response [105].
In an in vitro study, there was a significant increase in the human peripheral blood mononuclear cell (PBMC) proliferative response when pAECs were activated by pig IFN-γ, supporting the concept that inflammation augments the immune response to a xenograft [106] (Fig. 8a). The induction of T cell tolerance after transplantation is inhibited by inflammation [25]. By affecting the immune response, cytokine and chemokine secretions influence the outcome of allotransplantation [107,108]. Increased IL-7, IL-8, and IFN-γ-induced protein 10, chemokine ligand 9, and chemokine ligands 2 and 5 are associated with early allograft dysfunction [109][110][111].
Inflammation, coagulation, and the immune response have a complex inter-relationship [23,50]. For example, thrombin activates the human cellular response to pig cells in vitro, and induces a T cell proliferative response to the same extent as IFN-γ activation (Fig. 8b) [97].

Potential strategies to prevent inflammation in xenotransplantation recipients
Several strategies aimed at preventing or reducing excessive inflammation after xenotransplantation have been tested, some of which are clinically-approved.
Drug therapy (Table 2) Corticosteroids Corticosteroids activate several genes, including inhibitors of NF-κB, which has an anti-inflammatory effect [120]. After their administration to pig heart xenograft recipients, the levels of IL-6, IL-8, and MCP-1 were reduced [13]. However, D-dimer remained increased, irrespective of corticosteroids and/or anti-inflammatory therapy, suggesting that an inflammatory response persisted [13].
Tocilizumab has several other beneficial effects on the immune response to a graft. It reduces the number of memory B cells [130,131]) and plasma cells [132], but increases regulatory B cells [133], and the ratio of regulatory T cells [134]. It also reduces monocytes and myeloid dendritic cells [135]. Recipients of kidney allografts treated with tocilizumab suffer less antibody-mediated rejection [136], and have reduced donor-specific antibody levels [137].
However, recent evidence indicates that tocilizumab, although binding to primate IL-6 receptors, does not bind to IL-6 receptors on the pig graft [70], and therefore may have no protective effect on the graft. The IL-6 inhibitor, siltuximab, has a therapeutic effect in Castleman disease and certain inflammatory diseases by neutralizing IL-6 production [138]. IL-6 neutralization with siltuximab resulted in sustained C-RP suppression in Castleman disease [112], but it is not completely effective in xenotransplantation [Zhang G, et al., manuscript in preparation].

Anti-histone antibodies
Extracellular histones and TLR pathways are major targets for treating a variety of inflammatory conditions. Anti-histone therapy has the potential to prevent histone-induced inflammation in xenotransplantation [26]. The administration of an anti-histone antibody (e.g., anti-histone H4 monoclonal antibody) inhibits cytokine production and has a protective effect on various inflammatory injuries [45,56,[139][140][141][142][143][144][145][146][147]. The protective effects of rTBM against histone toxicity are mediated through both activated protein C-dependent and -independent ways [148]. Anti-histone antibodies have not yet been tested in in vivo models of xenotransplantation.

TNF-α inhibitors
EC activation is reduced by a TNF-α inhibitor [113]. A TNF-receptor fusion protein (TNF-RFP) has reduced inflammation in an in vivo xenoperfusion model, although the mechanism of its function is poorly understood [113].

NF-κB inhibitors
NF-κB plays a crucial role in enhancing the cellular responses to inflammation. Thrombin not only activates NF-κB, but also upregulates NF-κB-dependent genes [87]. As extracellular histones induce expression of tissue factor on ECs potentially through the NF-κB pathway, this amplifies thrombin generation [149]. The NF-κB inhibitor, parthenolide, reduced porcine EC apoptosis/death in vitro [26] (Fig. 3b). Parthenolide has also been reported to reduce endotoxic shock and prevent inflammation in immune glomerulonephritis [150]. It is used as prophylactic treatment for migraine, and has been reported to have a beneficial effect in clinical trials [151].

Platelet inhibitors
Aspirin is widely used as a preventative against vascular disease, and is associated with a reduction in myocardial infarction and stroke [168]. In addition, there is evidence that aspirin down-regulates some proinflammatory cytokines (e.g., IL-6) [116] and proinflammatory signaling pathways, including NF-κB [169][170][171].

Triiodothyronine (T3)
It remains uncertain whether, in the presence of a pig xenograft, the administration of T 3 can suppress the inflammatory state [79], but T3 treatment reduces inflammatory cytokines (e.g., TNF-α, IL-6), improving glycemic control in diabetic rats [119]. Nevertheless, as there is a fall in fT3 in all baboons following pig organ transplantation [30], we have found it beneficial to administer T3 to increase fT3 levels. Fig. 8 a IFN-γ-activation increases the proliferative response of human peripheral blood mononuclear cells (PBMCs) to wild-type (WT) and GTKO pig aortic endothelial cells (pAECs). When non-activated, the proliferative response to WT pAECs was greater than to GTKO pAECs (P < 0.05). There was an increase in the PBMC response when the pAECs were activated by IFN-γ, the response to WT pAECs again being significantly greater than to GTKO pAECs (P < 0.01) . The study illustrates how inflammation can increase the immune response to a xenograft. (CPM = counts per minute; SI = stimulation index). (Reproduced with permission from ref. [106]). b Thrombin activates T cell proliferation. The degree of activation of GTKO pig PBMCs by thrombin was comparable to that resulting from stimulation of the cells by porcine interferon-gamma (pIFN-γ). Thrombin-stimulated activation of the human cellular response was reduced by the addition of hirudin, confirming that thrombin was the stimulatory factor. (Reproduced with permission from ref. [97]) Genetic modification of the organ-source pig (Table 3) Expression of hemeoxygenase-1 (HO-1) HO-1 is known to have an anti-inflammatory effect and reduces cell apoptosis [14,[172][173][174][175][176][177][178]. It is an anti-oxidant enzyme, which is regulated by the erythroid 2-related factor 2 (Nrf2) pathway [194]. The activation of HO-1 can prevent TNF-α-induced inflammatory and oxidative damage by up-regulating the Nrf2/HO-1 signaling pathway [195]. hHO-1 expression on porcine cells prevents TNFα-and cycloheximide-mediated apoptosis ( Fig. 9) [173][174][175][176], and results in the downregulation of adhesion molecules, e.g., E-selectin, ICAM-1, and VCAM-1 [175]. Organs expressing hHO-1 were shown to be critical for prolonged survival of mouse cardiac xenografts in rats [173,177], and expression of hHO-1 in pig islets prolonged their survival in mice, and decreased immune cell infiltration and islet cell apoptosis [178].

Conclusions
Systemic inflammation may be playing a crucial role in pig organ xenotransplantation through activating the coagulation cascade and immune response. The administration of anti-inflammatory agents or the genetic modification of the organ-source pig by the introduction of human inflammation-regulatory transgenes may be beneficial to prevent or control inflammation. Control of inflammation is likely to allow a reduction in the intensity of exogenous immunosuppressive therapy. If immunological tolerance to a xenograft is to be obtained, then control of inflammation may be essential. TNF-α-induced apoptosis was reduced by transgenic expression of hHO-1. Human hemeoxygenase-1 (hHO-1) transgenic pig aortic endothelial cells (pAECs) were protected against TNF-α-mediated apoptosis, measured by a caspase 3/7 assay. pAECs from hHO-1 transgenic pigs were better protected against TNF-α-mediated apoptosis compared to WT pAECs. (Modified from ref. [175])