A Novel Resolvin-Based Strategy for Limiting Acetaminophen Hepatotoxicity

Acetaminophen (APAP) and APAP-containing products are the most commonly used antipyretic-analgesic medications worldwide. Although APAP is safe when taken at therapeutic doses in the majority of patients, overdoses of APAP can lead to significant morbidity and mortality. In fact, APAP-induced hepatotoxicity accounts for 50% of acute liver failure cases, and is the leading reason for liver transplantation for acute liver failure in the United States.1, 2, 3 In addition, the risk of developing APAP hepatotoxicity is further increased in the large cohort of patients with preexisting chronic liver disease.4 The pathogenesis of APAP-induced hepatotoxicity begins with its metabolism by perivenular hepatocytes, leading to the generation of reactive metabolites such as N-acetyl-p-benzoquinone-imine (NAPQI) that directly trigger oxidative stress, mitochondrial damage, and hepatocellular injury.5, 6, 7 Growing evidence suggests that as this injury propagates throughout the hepatic lobule, an exuberant host inflammatory response is activated, resulting in hepatic neutrophil infiltration and significant collateral damage.8, 9, 10, 11, 12, 13 Without timely treatment, many patients develop fulminant hepatic failure and multiorgan dysfunction. N-acetyl cysteine (NAC) is the only FDA-approved pharmacologic therapy for APAP hepatotoxicity.1, 14 However, it suffers from a limited rescue window (the time between APAP ingestion and initiation of therapy), as it targets only the initial reactive metabolite-driven injury that occurs at the earliest stage of disease pathogenesis.14, 15 Unfortunately, many patients are asymptomatic during this stage and do not present for treatment. As such, there is a need to better understand the complete pathogenesis of APAP hepatotoxicity with a particular emphasis on the later stages, in order to develop novel therapies that extend the rescue window to beyond that of NAC. Immune cells play a dynamic role in activating, maintaining, and resolving inflammation at the site of tissue injury. Kupffer cells, natural killer cells, and neutrophils have all been implicated in APAP hepatotoxicity by releasing various inflammatory mediators including cytokines, chemokines, and reactive oxygen species.8, 16, 17, 18 Although neutrophil recruitment into the liver and peripheral activation has been demonstrated in APAP hepatotoxicity, neutrophil contribution to the progression and severity of injury is controversial. Whereas some data show that neutrophils establish a host inflammatory response that amplifies overall liver injury, other data suggest that their activation may be a critical event for injury resolution following APAP overdose.8, 19, 20, 21 Recently endogenous lipid mediators derived from omega-3 polyunsaturated fatty acids have been shown to control important events during inflammation, such as neutrophil migration, adhesion, activation, and clearance.21, 22, 23 In particular, docosahexaenoic acid-derived lipid mediators, known as resolvins, regulate critical cellular events in the resolution of inflammation.21, 22, 23 Resolvins are synthesized by neutrophils during the resolution phase of inflammation, and serve to block the secretion of interleukin-1 beta and tumor necrosis factor-α, as well as stimulate nitric oxide production, thereby reducing neutrophil adhesion to the endothelium and inhibiting neutrophil infiltration into the tissue.24, 25 Resolvins have also been shown to potently and specifically inhibit neutrophil chemotaxis by directly acting on circulating neutrophils.26 As such, resolvins have proven protective in murine models of inflammatory bowel disease, colitis, sepsis, asthmatic airway inflammation, conjunctivitis, myocardial ischemia-reperfusion injury, and burn injury.22, 27, 28, 29, 30, 31 In this study, we tested whether inhibition of neutrophil recruitment attenuates APAP-induced hepatoxicity. Specifically, we hypothesized that administration of a resolvin compound, 7S, 16R, 17S-trihydroxy-4Z, 8E, 10Z, 12E, 14E, 19Z-docosahexaenoic acid (RvD2),22 after APAP overdose would lessen liver injury by reducing the secondary, neutrophil-predominant sterile inflammatory phase seen with APAP hepatic injury. Human microvascular endothelial cells (HMVECs) (Lonza, United States) were cultured as previously described31 using MCDB 131 media (Caisson Laboratories, North Logan, UT). Experiments were performed using standard phenol red free Dulbecco’s modified Eagle's medium (Thermo Fisher, Waltham, MA, GIBCO) supplemented with 10% heat-inactivated fetal bovine serum (FBS) and l-glutamine (200 mM). Polymorphonuclear granulocytes (PMNs) were isolated from human peripheral blood by gradient separation using Lympholyte-poly (Cedarlane Labs, Burlington, NC) according to the manufacturer’s protocol. To assay adhesion between PMNs and HMVECs in vitro, PMNs were co-cultured for 1 h with HMVECs in 12-well plates followed by three washing steps. Experimental pretreatment of HMVECs included NAPQI at 250 μM for 4 h. Experimental pretreatment of PMNs included RvD2 at 1 μM for 1 h. For image analysis, HMVEC nuclei were labeled with Hoechst 33342, PMNs with Calcein-AM stain (Life Technologies, Grand Island, NY). Images were acquired using an inverted microscope equipped with an incubation chamber (Axiovert, Zeiss, Oberkochen, Germany) and analyzed using ImageJ 1.43 u. To account for partial detachment of NAPQI-injured HMVECs, the number of adherent PMNs was normalized to the density of HMVECs. Washing steps effectively eliminated nonspecific PMN adherence to HMVEC-free areas. The temporal pattern of APAP-induced hepatic neutrophil infiltration was then evaluated using hepatic MPO activity as a surrogate for neutrophil activation. We observed an increase in neutrophil activation at 3 h following APAP administration. However, we observed a much sharper rise in hepatic neutrophil infiltration 12 h following APAP administration (Figure 1b, c). Mice pretreated with anti-GR1 antibody before APAP-displayed attenuated hepatocellular injury, based on ALT values, compared to vehicle-treated mice at all time points throughout the study. These results are consistent with previously published data.8, 36 Notably, the difference in liver injury between anti-GR1-treated and vehicle-treated mice was most significant at later time points (12 and 24 h) after APAP administration (Figure 2a), suggesting that neutrophils play a more prominent role in ongoing liver injury and inflammation as opposed to the initial injury. Furthermore, mice treated with anti-GR1 before a lethal dose of APAP exhibited no mortality when followed for 10 days after APAP administration, in contrast to vehicle-treated mice receiving APAP, which experienced significant mortality (Figure 2b). Multiple inflammatory mediators activate neutrophils and enhance their interaction with endothelial cells, thereby promoting neutrophil infiltration into active sites of tissue injury.40,41 Previous work has demonstrated the ability of RvD2 to inhibit neutrophil migration to sites of injury.22 We showed that co-culturing of human neutrophils and endothelial cells in the absence of NAPQI leads to minimal neutrophil adhesion. Conversely, the addition of NAPQI strongly increases neutrophil adhesion to endothelial cells. Notably, NAPQI-induced neutrophil adhesion to endothelial cells is significantly attenuated by RvD2 treatment (Figure 5b). In this study we demonstrate that hepatic neutrophil infiltration occurs secondary to APAP-induced liver injury, and that neutrophil depletion attenuates liver injury. Furthermore, we show that administration of RvD2 after APAP overdose attenuates liver injury and extends the therapeutic rescue window as compared to the gold standard of treatment, NAC. Mechanistic in vitro analysis highlights RvD2’s ability to inhibit neutrophil attachment to endothelial cells in the presence of NAPQI, the reactive metabolite of APAP, suggesting RvD2 may prevent neutrophil migration into the liver following APAP injury. Although inflammation is host protective against a variety of insults and pathogens, excessive inflammation can be damaging. For example, sepsis results from an overly exuberant and persistent inflammatory response to a pathogen, resulting in multisystem organ dysfunction. Furthermore, the inability to dampen inflammation may contribute to the development of certain chronic inflammatory diseases, such as inflammatory bowel disease and rheumatoid arthritis. Accordingly, the ability to modulate the resolution phase of the inflammatory response may prove critical in the treatment of these disorders. Endogenous lipid mediators, termed resolvins, have been shown to actively contribute to the inflammation resolution phase.23 In fact, exogenous administration of resolvins has decreased the severity of inflammation in multiple preclinical models of disease, including sepsis and inflammatory bowel disease. These encouraging preclinical animal data have led to the development of clinical programs investigating the use of pro-resolving mediators for human disease, in particular ocular and neurodegenerative diseases.42 Although the mechanism for the observed protection is not fully understood, it has been demonstrated that resolvins regulate trafficking of immune cells to active sites of inflammation, downregulate immune cell production of pro-inflammatory cytokines, and preserve the vascular network surrounding sites of inflammation.25, 26, 31 It is increasingly appreciated that APAP-induced hepatotoxicity results not only from reactive metabolite-driven injury but also as a result of an exuberant host inflammatory response. Supporting this theory are the data that show APAP-related hepatocellular injury occurs long after hepatotoxic metabolites such as NAPQI have been removed from systemic circulation, suggesting an inflammation-driven secondary component to APAP hepatotoxicity.43 With these data in mind, we tested the hypothesis that exogenously administered RvD2 would dampen hepatic inflammation and injury induced by APAP. Indeed, RvD2 given as late as 12 h after APAP attenuated liver injury. Our data suggest that this protective effect is driven, at least in part, by inhibition of neutrophil migration into the liver. It is well-recognized that although NAC offers benefit to patients with APAP overdose, its efficacy is largely based on the timing of administration. NAC offers best protection against hepatotoxicity if given within 10 h of APAP ingestion.37, 38, 39, 40 Unfortunately, only a minority of patients present in this narrow time frame. Accordingly, additional therapies that offer benefit at later time points following APAP ingestion are critically needed. An intriguing finding in our study was the ability of RvD2 to extend the therapeutic rescue window following APAP overdose. In particular, RvD2 was able to rescue from APAP-induced liver injury when given up to eight times later than NAC. We believe this extended protection stems from the fact that RvD2 targets the resolution phase of the inflammatory pathways induced by APAP, whereas NAC targets the earliest phases of injury in the pathogenesis of APAP-induced liver injury. This exciting finding deserves further investigation as a potential treatment strategy, either as an alternative to or in combination with NAC, for the treatment of APAP overdose in patients. The contribution of neutrophils to the pathogenesis of APAP-induced hepatotoxicity has been the subject of debate. Consistent with our data, previous reports have documented attenuation of APAP liver injury in neutrophil-depleted mice, suggesting an essential role for neutrophils in the injury process.8,36 In addition, recent data show neutrophils contribute to the sterile inflammatory response seen in APAP-induced liver injury in a TLR-9-dependent manner.44 However, other reports suggest that neutrophils are not critical to the inflammatory response induced by APAP, and, in fact, may contribute to liver regeneration and recovery following the APAP insult.19, 45, 46, 47, 48 Although our study does not provide definitive evidence for the role of neutrophils in APAP liver injury, it provides experimental support for their importance. Despite these exciting findings, certain limitations need to be noted. As this was a proof-of-concept study, we did not test various doses of RvD2 to ascertain the lowest dose needed to obtain protection. It is unclear if the supraphysiological dose used in this study would be tolerated clinically; therefore, the use of RvD2 as a therapeutic target in humans with APAP overdose remains uncertain. Furthermore, it must be noted that in this study we did not investigate the effects of RvD2 administration systemically and on immune cells other than neutrophils, such as macrophages and natural killer cells. Further work is needed to comprehensively define the effect of resolvins on the multiple inflammatory pathways involved in the APAP liver injury. In summary, we highlight the ability of RvD2 to protect against APAP-induced liver injury, and compared to NAC therapy, significantly extend the therapeutic rescue window by eightfold following APAP overdose. Although the mechanism for RvD2-induced hepatoprotection is likely multifactorial, inhibition of neutrophil migration and activation plays an important role. Further study into the potential use of RvD2 in the treatment of APAP-induced hepatotoxicity is thus warranted. Guarantor of the article: Martin L. Yarmush, MD, PhD. Specific author contributions: Suraj J. Patel and Jay Luther designed the concept, performed experiments, and wrote manuscript. Stefan Bohr, Arvin Iracheta-Vellve, Matthew Li, and Kevin R. King performed experiments. Raymond T. Chung and Martin L. Yarmush provided final approval. Financial support: Shriner Hospital for Children Postdoctoral Fellowship Award to Suraj J. Patel; AASLD Clinical and Translational Research Grant to Jay Luther. Potential competing interests: None.