Owing to the potential of cancer cure by hepatectomy, the surgical limits of liver resection were pushed forward in recent years. Major liver surgery frequently requires the occlusion of hepatic inflow (Pringle maneuver) to diminish intraoperative blood loss. Reinducing blood flow after tissue ischemia leads to ischemia reperfusion injury (IRI), which may substantially compromise patient outcome. Especially as hepatotoxic preoperative chemotherapy is increasingly used to improve oncological outcome, the reduction of further harm to the liver is of major importance to allow adequate postoperative liver regeneration. Furthermore, IRI is a major concern in liver transplantation, as the donor organ shortage has prompted the inclusion of “marginal” organs, which are more likely to suffer from increased IRI. Several potential treatment strategies have been developed to reduce IRI. Ischemic preconditioning (IP), which is defined as the repetitive occlusion of blood flow to the target organ for short time periods prior to surgery, has been shown to be effective in reducing IRI. While one could assume that the effects of IP are caused by local tissue adaptation, remote ischemic preconditioning (RIPC) induced by repeated brief periods of limb ischemia distinct from the target organ, is also able to induce protection. While several findings have been implicated to account for the beneficial effects of RIPC in the target organ, the exact mechanisms remain poorly understood.
In this issue of Hepatology, Oberkofler et al. address the role of platelet-derived serotonin in the process of RIPC. They were able to demonstrate that the reduction of IRI by RIPC was associated with increased platelet activation and the protective effect of RIPC was lost in platelet-depleted mice. They further focused on serotonin as a potential effector of platelets in protecting against IRI, as it has been implicated in tissue repair after ischemia. Platelets, which are incapable of synthesizing serotonin themselves, absorb serotonin efficiently from the plasma pool and store it in their dense granules. Upon platelet activation dense granules are released first, followed by α-granules. Indeed, the authors found that after RIPC, serotonin was released by platelets and serotonin-deficient mice did not benefit from RIPC prior to IRI.
The authors further aimed to identify the downstream effector(s) of serotonin in RIPC. In accordance with the previously reported effects of serotonin on vascular endothelial growth factor (VEGF), they were able to demonstrate that serotonin induced expression of VEGF in cultured endothelial cells (ECs). Importantly, they were also able to show that inhibition of VEGF abolished the effects of RIPC. Of note, as platelet α-granules contain high levels of VEGF, the increase in circulating VEGF after RIPC is presumably derived from both ECs and platelets. Indeed, Oberkofler et al. used P-selectin to document platelet activation after RIPC, which is in fact a protein stored in α-granules such as VEGF, suggesting a relevant contribution of α-granule release after RIPC.
Using gene expression analysis the authors were further able to document that two specific target gene products mediated RIPC effects. In particular, matrix metalloproteinase-8 (Mmp-8) and interleukin-10 (IL-10) were found to be up-regulated in the target organ and the specific inhibition of both molecules eliminated the protective effect of RIPC. Importantly, Oberkofler et al. confirmed that the RIPC induced platelet-serotonin-VEGF-IL-10/Mmp-8 axis is also crucial to protect against renal IRI and demonstrated specific induction of protective gene expression in several organs (liver, kidney, lung, heart, intestine) after RIPC, rendering their observation from an organ-specific phenomenon to a presumably “general principal.” Indeed, the authors also observed a reduced expression of inflammatory genes in RIPC-treated mice in an experimental model of pharmacologically induced pancreatitis as well as the mitigation of acetaminophen-induced liver injury, suggesting a general tissue protective effect of RIPC against different stressors.
Comparably, Wang et al. reported on a similar model of RIPC in liver IRI. They proposed that high mobility group-box 1 (HMG-B1) was one of the major effectors of the protective effects of RIPC. HMG-B1 is a nuclear protein involved in transcriptional regulation that is secreted by necrotic cells as well as immune and endothelial cells. After RIPC, circulating HMG-B1 and tumor necrosis factor alpha (TNF-α) levels rose significantly. The subsequent induction of IRI was associated with a dampened TNF-α expression presumably by way of an IRAK3 (also IRAK-M)-mediated pathway resulting in the concomitant reduction of reperfusion injury.
So how can we integrate the observations by Wang et al. and the work of Oberkofler et al.? As an initial event, ischemia at the remote limb seems to induce platelet activation. This results in the release of dense granule-stored serotonin as well as α-granule proteins such as VEGF. As serotonin has been shown to induce HMG-B1 release from ECs, serotonin might also stimulate HMG-B1 secretion. Additionally, HMG-B1 might be released directly from necrotic cells at the ischemic limb site. Serotonin then further stimulates VEGF release from ECs. Moreover, HMG-B1 is known to induce secretion of VEGF. Accordingly, HMG-B1 might be contributing to the increase of circulating VEGF after RIPC. Ultimately, VEGF induces protective gene expression (IL-10 and Mmp-8) in the target organ. Importantly, both HMG-B1 and IL-10 are known to reduce TNF-α expression, suggesting another central mechanism common to both pathways. However, further research is required to elucidate the presumably close association of these pathways. A schematic presentation of a possible mechanism of RIPC in tissue protection is illustrated in Fig. 1.
RIPC has a broad spectrum of potential clinical applications. The clinical benefit of RIPC has previously been demonstrated for patients undergoing cardiac intervention or surgery. Specifically concerning liver surgery, liver injury, and liver transplantation, IP as well as RIPC might also yield major advances. However, while encouraging results for IP in hepatic resection have been reported,[2, 14] a general benefit of IP for the liver setting has not been documented to date. In particular, two large meta-analyses were unable to detect a beneficial effect of IP before liver surgery and in the procurement of liver grafts.[15, 16] While the experimental setting is obviously different from the clinics, a better understanding of the pathophysiologic mechanisms of RIPC offers the opportunity to elucidate the reasons for this discrepancy. Accordingly, the data of Oberkofler et al. contribute to an evolving understanding regarding the underlying mechanisms of RIPC and introduce platelets as a novel crucial effector of systemic tissue protection.
Finally, it should be stressed that clinical confirmation of these findings in daily practice should be top-ranked in upcoming investigations.
Patrick Starlinger, M.D.
Thomas Gruenberger, M.D., Ph.D.
Department of Surgery Medical University of Vienna General Hospital Vienna, Austria