Serp-2, a virus-derived apoptosis and inflammasome inhibitor, attenuates liver ischemia-reperfusion injury in mice

Ischemia-reperfusion injury (IRI) is a two-step process characterized by an initial transient blockade of blood flow and oxygen delivery. With IRI, there is an initial, sub-lethal damage followed by restoration of blood flow and a paradoxical acceleration of injury. Induction of liver ischemia-reperfusion injury (LIRI) is inevitable with transplantation surgery, occurring during organ resection, harvest, and graft implant, and can also occur with trauma and hemorrhagic shock [1]. IRI is a primary cause of early graft failure after transplant and can lead to a higher incidence of early acute and also long term chronic rejection. This ongoing injury to transplanted organs contributes to significant graft loss after the first year post transplant, creating the need for repeat transplantation and is one cause for the acute shortages of donor organs available for transplantation [1].

As ischemia-reperfusion injury is seen often during transplantation, attention has been centered on identifying methods to reduce IRI during and post-transplant (e.g., liver, lung, heart, and kidney transplants). For liver grafts, the post-transplantation standard of care commonly involves targeting IL-2 with either monoclonal antibodies (e.g., Basiliximab) [2], mycophenolic acid [3] or FK506 [4] (frequently as a cocktail) in an effort to prevent T cell proliferation and activity. Other approaches include inhibition of the mammalian target of rapamycin (mTOR) with drugs such as rapamycin and Everolimus [5], or generalized immune suppression with cyclosporine A or steroids [6]. While these treatments have significantly improved outcomes following liver transplantation in the past three decades, many adverse effects persist. Most notably, transplant recipients are at increased risk for malignancy [7], viremia [8, 9], bone loss [10], new-onset diabetes [11] and cardiovascular disease [12], many of which are associated with post-transplant immune suppression. Experimental approaches to reduce post-ischemic injury while also avoiding these adverse effects have included activation or inactivation of specific signaling pathways by genetic or small molecular perturbation [13‐15], and pre- or post-treatment with a variety of small molecules [16‐18]. Despite experimental advances, movement towards the clinic has been slow and there is a substantial need for steroid-sparing, immune modulatory treatments designed to prevent tissue loss after IRI.

Viruses have evolved advanced and highly potent immune-modulating strategies over millions of years of co-evolution with mammalian hosts. A key example of one such virus is Myxomavirus, a leporipoxvirus and the causative agent of a lethal infection, myxomatosis, in the European rabbit (Oryctolagus cuniculus). Interestingly, due to incompletely understood mechanisms, Myxomavirus is host-restricted to O. cuniculus and is not pathogenic in other rabbit species and in humans [19].

Myxomavirus has evolved to a highly effective pathogen in rabbits through development of potent immune modulating proteins deployed to subvert, suppress and overwhelm the host immune response. We have previously demonstrated therapeutic benefit through delivery of these immune modulators as either recombinant, purified proteins or a coding sequence DNA in Adeno-associated viral vectors (AAV) in animal model studies of disease. For example, the Myxomavirus protein M-T7 is a chemokine-GAG interaction inhibiting protein that reduces renal transplant rejection in both rats [20] and mice [21], and decreases vascular balloon injury in rabbits and rats [22]. In other work, we have demonstrated that treatment with Serp-1, a member of the serpin superfamily of proteins, as well as peptides derived from the Serp-1 reactive center loop (RCL), reduce severity and prolong survival in a lethal, herpesvirus-induced model of large vessel vasculitis [23‐25]. These and other examples demonstrate that immune modulatory proteins employed by Myxomavirus for anti-immunological evasion are attractive proteins for repurposing as new therapeutic approaches.

Serp-2 is a second Myxomavirus-derived serpin that is a critical virulence factor for Myxomavirus. Viruses deficient in Serp-2 cause robustly attenuated infections with substantial increases in virus-limiting inflammation [26]. Early molecular work on Serp-2 has demonstrated cross-class inhibitor activity for caspase-1 in the inflammasome signaling pathway, as well as caspases 8 and 10 and granzyme B in the apoptosis pathway [27‐29]. Thus, by inhibiting both inflammasome and apoptotic signaling, Serp-2 enables Myxomavirus to suppress inflammation and avoid immune clearance.

In prior work, we tested Serp-2 treatment as an immune modulatory, anti-inflammatory protein therapeutic to reduce disease pathology in mouse models. A single administration of Serp-2 treatment significantly reduced aortic aneurysm formation and plaque growth in an aortic angioplasty model in Apolipoprotein E-deficient (ApoE^−/−) mice over a period of 4 weeks [30]. In other work, Serp-2 potently reduced plaque growth and inflammation in two separate models: a rat model of iliofemoral balloon angioplasty injury, as well as aortic allograft transplant of plasminogen activator inhibitor 1-deficient (PAI-1^−/−) or ApoE^−/− aortas into Balb/C recipient mice [31]. Serp-2 lost activity in granzyme B/ApoE double knock-out aortic allograft transplants. Interestingly, in a carotid cuff injury model in ApoE^−/− mice, Serp-2 displayed systemic effects against plaque growth at the aortic root, a site distal to the acute cuff injury [31]. Thus, Serp-2 has been demonstrated as an effective and potent systemic, cross-class immune modulator against tissue injury in a variety of inflammatory in vivo models.

This short report extends prior studies with Serp-2 as a virus-derived, therapeutic immune modulator to an analysis of the potential for treatment with Serp-2 in a mouse model for LIRI. Progression of LIRI has been attributed to a variety of cellular mechanisms. Among the proposed mechanisms, perturbation of the apoptotic and inflammasome signaling cascades has demonstrated efficacy in in vivo models [14, 32‐38]. On this basis, we hypothesized that the apoptosis and inflammasome inhibitory functions of Serp-2 would reduce pathology in liver ischemia-reperfusion injury. Here, we investigated LIRI as a controlled, outcomes-focused (i.e., survival) model for testing further applicability of Serp-2 as a therapeutic protein.

Despite advances in surgical procedure techniques and post-transplant care and immunosuppression, IRI remains a primary cause for early graft loss after liver transplantation [1]. While questions remain as to the exact mechanism of graft loss caused by LIRI, many groups have shown a crucial role for apoptotic [42, 43] and also inflammasome pathway activation and signaling [14, 34, 36, 38]. Study of these pathways has led to substantially improved treatments in other sterile diseases, and thus attention has now been focused on developing similar treatments in LIRI.

Here, we investigate the potential for Serp-2, a Myxomavirus-derived serine proteinase inhibitor (serpin) with known cross-class inhibition of caspase-1 in the inflammasome signaling cascade and caspases 8 and 10 and granzyme B in the apoptosis signaling cascade, to ameliorate LIRI severity in mice. We have demonstrated here that while saline-treated mice die early after LIRI, repeated injections of Serp-2 (every alternate day) at a dose of 100 ng/g, in the absence of other immune suppression treatments, significantly prolonged survival for up to 10 days. Ten days was the endpoint of our study (Fig. 1C). Protection against ischemia-reperfusion injury was likely initiated early and not due to accelerated healing as ALT, an indicator of acute liver injury, was reduced by Serp-2 at 24 h (Fig. 1B). The improved survival and reduced damage were specific to the function of Serp-2. Similar same-dose treatments with M-T7, an unrelated Myxomavirus-derived immune modulating protein, did not provide any survival advantage, reduction in acute injury markers nor preservation of tissue viability (Figs. 1 and 2). We also note that not only do mice treated with Serp-2 survive, but their livers display significant reductions in necrosis and overall pathology when compared to saline-treated mice (Fig. 2 and Table 2). In investigating a potential physiologic mechanism for Serp-2-dependent survival advantage in this model, we note that in addition to reduced infarct scarring of the liver, areas of infarct had smaller inflammatory cell infiltrates, identified as macrophages by immunohistochemistry (Fig. 3). While we cannot specifically identify whether the macrophage infiltrates in the infarcted region are so-called “cavity” macrophages [44] or tissue-resident Kupffer cells, we note that the survival outcomes from Serp-2 treatment combined with the reduction of invasive monocyte-lineage inflammatory cells agrees with prior clinical studies [41].

Ischemia-reperfusion injury creates a complex physiological state in the liver, involving hypoxia, reactive oxygen and nitrogen species formation, multiple forms of cell death and subsequent damage associate molecular pattern (DAMP) release, all of which initiate a feed-forward cascade of damage [45]. Indeed, the mechanism of IRI in the liver remains a topic of active debate, particular with respect to the role of apoptosis, necrosis and other forms of cell death in propagating post-ischemic tissue damage [37, 42, 46, 47]. Accordingly, caspase cleavage has been reported even in the presence of a number of other small molecule and biologic treatments shown to preserve liver viability during warm IR procedures [48‐51]. Here, we found profound protection against ischemia-reperfusion injury by treatment with Serp-2, despite still observing caspase-1, − 3 and − 8 activation at 24 h follow-up (Additional file 1: Figure S1). Our data therefore agree with the principle that activation of apoptotic and inflammatory caspases is not by itself a direct nor the sole indicator of tissue viability or injury following ischemia-reperfusion, and that histopathology or functional readouts (e.g., circulating markers of injury such as ALT) are preferred for assessing the effect of protection in this model [37]. It should also be noted that Serp-2 may alter circulating levels of proteases rather than tissue levels and that the tissue isolates will represent a composite of multiple cell types that may respond heterogeneously to the Serp-2 mediated protective functions. Thus, the precise physiological mechanism and the precise cell targets for Serp-2 protection against ischemia-reperfusion injury in the liver, as described in this brief report, remains to be elucidated. Further work will focus on potential extracellular effects of Serp-2, such as inhibition of extracellular and circulating active caspases released from dying cells, a recently reported mediator of inflammation amplification [52‐54], or on potential therapeutic benefit of Serp-2-derived metabolites and peptides, such as we have described for Serp-1, a related serpin from Myxomavirus [24, 25].

Myxomavirus-derived proteins as therapeutic agents have proven highly effective in a wide variety of animal models [20, 21, 55, 56]. Of importance, the evolution of these proteins within poxvirus vectors has led these proteins to exhibit very low immunogenicity. Indeed, Serp-1 was proven safe and effective in a Phase IIa clinical trial in humans with acute unstable coronary syndrome [57]. Serp-1 dose-dependently reduced markers of heart damage with a Major Adverse Cardiac Event (MACE) score of zero and without induction of neutralizing antibody. The safety and cross-species efficacy of Myxomavirus-derived proteins, and in this study with Serp-2, highlights the potential for developing these agents for treatment of inflammatory diseases. The substantial and significant Serp-2 mediated therapeutic benefit post-LIRI as demonstrated in this study indicates the potential for Serp-2 treatment in liver transplantation. Further study is warranted for testing Serp-2 and other Myxomaviral proteins in preserving engrafted tissues.