The current study is the first to demonstrate the role of cholinergic anti-inflammatory pathway in hepatic IR injury. The findings showed that administration of nicotinic ACh agonists, DMPP and nicotine, significantly attenuated the release of cytokines/chemokine, and diminished the early phase of hepatic injury caused by liver IR in mice. In contrast, the late phase of liver injury was not protected, although cytokine/chemokine production was significantly decreased. These results highlight the differential mechanisms of reperfusion injury during early and late phases of IR.
The present data indicates an association between decreased cytokine production and hepatocellular injury during cholinergic agonist treatment. However, this association appears only during the early phase of liver IR. This finding supports previously published studies [5
]. In an in vitro
study, ACh significantly attenuated the release of pro-inflammatory cytokines (i.e., IL-1β, IL-6, and TNF) in LPS-stimulated human macrophage cultures [5
]. In addition, direct electrical stimulation of the peripheral vagus nerve in vivo
during lethal endotoxemia in rats inhibited TNF synthesis in liver, decreased plasma TNF levels, and prevented the development of shock [5
]. TNF is the proximal cytokine generated during an inflammatory response that stimulates various cells including Kupffer cells and hepatocytes to produce various CXC chemokines such as MIP-2 and KC [16
]. TNF has also been implicated in the production of other inflammatory mediators such as eicosanoids, nitric oxide and ROS, which can further exacerbates inflammation and tissue injury [17
]. In hepatic IR, TNF is involved in the production of chemokines [12
]. Previous studies by Colletti et al.
] and Lentsch et al.
] have shown that neutralization of TNF-α or MIP-2 by administration of antibodies significantly attenuated hepatocellular injury induced by hepatic IR. Further, our laboratory has previously shown that Kupffer cells were the major source of the cytokines and chemokines in liver IR, and inhibition of Kupffer cell activity significantly attenuated the liver IR injury [11
]. Acetylcholine, the principal neurotransmitter of the vagus nerve, binds to nicotinic and muscarinic receptors on macrophages, and the receptor-ligand interaction inhibits the synthesis of TNF at a post-transcriptional level [5
]. The nicotinic AChR α-7 subunit is essential for the inhibition of cytokine synthesis by the cholinergic anti-inflammatory pathway [18
]. Therefore, the published literature and our data suggest that DMPP and nicotine interaction with AChR on Kupffer cells may possibly attenuate the synthesis of the pro-inflammatory cytokines as well as the development of liver injury during the early phase of liver IR in mice.
In contrast, there was no protective effect in the late phase of IR injury during cholinergic agonist treatment, as seen in the early IR injury phase. However, cytokine production was significantly attenuated in both early and late phases of IR injury. This finding does not support the findings reported by Wang et al.
, who demonstrated that treatment of mice with nicotine provided lasting protection and not merely delayed the onset of death using a sepsis model [19
]. It is interesting to note that in their study nicotine treatment did not have a significant effect on IL-6, which appears to be a marker in the prognosis in sepsis in patients and animals [20
]. In addition, it was shown that protection against sepsis was attributed to the high mobility group box-1 (HMGB-1) protein, which was significantly diminished upon nicotine treatment. HMGB-1 is a late mediator of lethal systemic inflammation in sepsis [23
], however, a recent study has suggested HMGB-1 as an early mediator of inflammation and organ damage in hepatic IR injury [24
]. In vitro
, necrotic cells as well as purified HMGB-1 trigger the production of pro-inflammatory cytokine TNF-α [25
]. HMGB-1 is a monocyte-derived protein. It is also present in the nucleus of hepatocytes in the normal liver, which is significantly up-regulated after IR [24
]. Whether the cholinergic agonists had any effect on HMGB-1 production in the study presented here remains unclear. The discrepancy between the effect of nicotine in providing long term protection against sepsis and not against liver IR injury may be attributed to the differences in the underlying pathophysiologic mechanisms of the two disease conditions.
Previous studies have shown a direct association between CXC chemokines, neutrophil recruitment and liver injury. Specifically, it has been shown that neutralization of CXC chemokines significantly attenuated neutrophil infiltration and liver injury in rat and mouse models of warm hepatic IR [12
]. In support of these findings, the current study showed that inhibition of cytokine/chemokine production correlated with liver injury (plasma ALT level) during the early phase of liver IR injury. However, our study indicated that during the late phase of hepatic IR, cholinergic agonists significantly inhibited CXC chemokine production (i.e., MIP-2), which did not correlate with neutrophil infiltration and the liver IR injury. One explanation may be that the cholinergic agonist treatment did not sufficiently abolish CXC chemokine production and that the generated MIP-2 was able to trigger neutrophil infiltration and liver injury. Another explanation is that during the late phase of liver IR injury, cytokine/chemokine role may be of limited relevance for neutrophil infiltration and liver injury, suggesting participation of other more potent inflammatory factors. In favor of the latter hypothesis, Dorman et al.
showed that neutrophil infiltration and liver injury occurred independent of CXC chemokine production in response to apoptotic cell injury in mice subjected to endotoximia [26
]. The wild type as well as CXCR2 -/- mice showed similar neutrophil infiltration and liver injury. This hypothesis is supported by the data from the present study, which showed a significant increase in MIP-2 production and liver injury (as indicated by ALT and histopathology) by 6 h of reperfusion with moderate neutrophil infiltration into the ischemic liver lobe. At the peak of neutrophil infiltration (i.e. 24 h of reperfusion), MIP-2 levels were significantly attenuated, suggesting that neutrophils had infiltrated the liver parenchyma in response to the necrotic hepatocytes. Additionally, the notion that the neutrophil response is mediated by necrotic hepatocytes is further supported by the lack of neutrophil infiltration into the non-ischemic lobe of the IR mice where there is no hepatocellular damage. Of course, one should keep in mind a potential role of other factors such as reduced neutrophil deformability [27
], sinusoidal cell swelling [28
], and vasoconstriction of the sinusoids [29
], which can facilitate sinusoidal neutrophil trapping. Other studies have underscored the role of complement factors as a mediator of IR injury [30
]. A component of the complement system, C5a, can prime and activate Kupffer cells as well as neutrophils to generate ROS. Further, the complement system through formation of the membrane attack complex can potentially mediate IR injury. Evidence has been presented that the complement system is crucially involved in the pathogenesis of renal IR injury by modulation of neutrophil-independent, as well as neutrophil-dependent pathways [32
]. The contribution of infiltrated neutrophils to hepatic IR injury and modulation of their functions by the cholinergic agonists in our study remains unclear at present and further studies are necessary to understand their potential role.
Hepatic IR injury is a complex process, and other inflammatory factors in addition to the cytokines are implicated in the pathogenesis. For example, studies have suggested a central role for ROS in the pathophysiology of liver IR injury [33
]. The ROS generated during IR can induce apoptosis [35
]. Additionally, the generated ROS can serve as a second messenger molecule in mediating TNF-α signaling and IL-8 secretion [37
]. Further, studies have shown that hepatotoxic injury mediated by oxidative stress can be prevented by administration of anti-oxidants [40
]. Whether cholinergic agonist treatment has any effect on ROS generation is of interest. It has been shown that in in vitro
experiments, nicotine treatment inhibited the production of oxygen radicals by human neutrophils and macrophages and IL-1 by macrophages [41
]. In contrast, a study by Jay et al.,
has shown that nicotine potentiates superoxide production by human neutrophils [42
]. Iho et al.
showed that nicotine induces human neutrophils to produce IL-8 (CXC chemokine) through the production of ROS and subsequent activation of NF-kB, with no effect on monocytes [43
]. The discrepancy between these studies may be explained with respect to the cell type or their activation/differentiation status as well as the nicotine concentrations applied. In support of this notion, a study by Gillespie et al.,
has shown that high concentrations of nicotine attenuated formyl peptide-induced chemotaxis, while lower doses potentiated the response [44
]. It is of interest to examine the effect of the cholinergic agonists on ROS production and the biochemical signaling in hepatic IR. Whether the cholinergic agonists protected the early phase of hepatic IR injury by attenuating a Kupffer cell-induced oxidant stress remains unclear, and further studies are necessary to examine its potential role.
The loss of protective effect of the cholinergic agonists during the late phase of liver IR remains unclear. One possible explanation may involve cholinergic receptor desensitization and/or recycling. Another possibility may be a lack of sufficient concentration of the agonists, due to their clearance from the body, and therefore, resulting in a diminished protective effect. This hypothesis was addressed by the further addition of doses of either agonist. There was no significant difference in liver IR injury after 6 and 24 h of reperfusion between the mice that received a single injection of the agonists prior to the onset of ischemia and those that received additional doses (2–4 injections) during the reperfusion. Further, the delay in hepatic injury may be due to the effect of cholinergic agonists on the hepatic microcirculation. The cholinergic receptors are present in hepatic microcirculation and have shown a role in vasodilatory function [45
]. Other studies have shown the presence of non-neuronal nicotinic AChR on endothelial cells that were upregulated upon proliferation in response to hypoxia and ischemia [47
]. Whether the cholinergic receptor agonists act on endothelial cell functions in hepatic IR remains to be elucidated.
In summary, the study presented indicates that AChR agonists provide a means to regulate cytokine production and to delay the hepatocellular injury caused by reperfusion of the ischemic liver. Modulation of the cholinergic anti-inflammatory pathway may have important therapeutic implications during the early phase of hepatic IR injury. Further studies are necessary to determine the exact role of the cholinergic system in liver IR injury during the early and late phases of reperfusion.