- Email: [email protected]
Fasting protects liver from ischemic injury through Sirt1-mediated downregulation of circulating HMGB1 in mice
Accepted Manuscript Fasting protects liver from ischemic injury through Sirt1-mediated downregulation of circulating Hmgb1 in mice Andreas Rickenbacher, Jaehwi Jang, Perparim Limani, Udo Ungethüm, Kuno Lehmann, Christian E. Oberkofler, Achim Weber, Rolf Graf, Bostjan Humar, Pierre-Alain Clavien PII: DOI: Reference:
S0168-8278(14)00263-3 http://dx.doi.org/10.1016/j.jhep.2014.04.010 JHEPAT 5112
To appear in:
Journal of Hepatology
Received Date: Revised Date: Accepted Date:
20 August 2013 4 April 2014 10 April 2014
Please cite this article as: Rickenbacher, A., Jang, J., Limani, P., Ungethüm, U., Lehmann, K., Oberkofler, C.E., Weber, A., Graf, R., Humar, B., Clavien, P-A., Fasting protects liver from ischemic injury through Sirt1-mediated downregulation of circulating Hmgb1 in mice, Journal of Hepatology (2014), doi: http://dx.doi.org/10.1016/j.jhep. 2014.04.010
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1 JHEPAT-D-13-01488R2
Fasting protects liver from ischemic injury through Sirt1-mediated downregulation of circulating Hmgb1 in mice
Andreas Rickenbacher1, Jaehwi Jang1†, Perparim Limani1†, Udo Ungethüm1, Kuno Lehmann1, Christian E. Oberkofler1, Achim Weber2, Rolf Graf1, Bostjan Humar1*, Pierre-Alain Clavien1 *
1
Swiss Hepato-Pancreatico-Biliary Center, Department of Surgery and 2Institute of Pathology, University Hospital Zürich, CH-8091 Zürich, Switzerland *BH and PAC share senior authorship †equal contribution
Correspondence: Pierre-Alain Clavien, MD, PhD Department of Surgery University Hospital of Zurich Raemistrasse 100, 8091 Zurich, Switzerland Tel: +41 44 255 33 00 Fax: +41 44 255 44 49 e-mail: [email protected] Word count: 4470 words, 4 figures, 0 tables Abbreviations: IR: ischemia reperfusion, ROS: reactive oxygen species, DAMP: damage-associated molecular pattern, Hmgb1: high mobility group box 1, KC: Kupffer cell, IRI: ischemia reperfusion injury, ATP: adenosine triphosphate, AST: aspartate aminotransferase, ALT: alanine aminotransferase, H&E: hematoxylin and eosin, PBS: phosphate buffered saline, TUNEL: TdT-mediated dUTP-biotin nick end labeling, TNFα: tumor necrosis factor α, IL1β: interleukine 1β, IL-6: interleukine 6, MCP1: monocyte chemotactic protein-1,
Conflict of interest: The authors have nothing to disclose. Financial support: This work was supported by an Academic Research in Surgery grant to AR, a Forschungsschwerpunkt grant by the University of Zürich to PAC, BH and RG, and a Swiss National Foundation grant (#320030_132985) to PAC.
2 Abstract (word count: 250 words)
Background&Aims: Fasting and calorie restriction are associated with a prolonged life span and an increased resistance to stress. The protective effects of fasting have been exploited for the mitigation of ischemic organ injury, yet the underlying mechanisms remain incompletely understood. Here, we investigated whether fasting protects liver against ischemia reperfusion (IR) through energy-preserving or anti-inflammatory mechanisms. Methods: Fasted C57BL6 mice were subjected to partial hepatic IR. Injury was assessed by liver enzymes and histology. Raw 264-7 macrophage-like cells were investigated in vitro. Sirt1 and Hmgb1 were inhibited using Ex527 and neutralizing antibodies, respectively. Results: Fasting for one, but not two or three days, protected from hepatic IR injury. None of the investigated energy parameters correlated with the protective effects. Instead, inflammatory responses were dampened in one-day-fasted mice and in starved macrophages. Fasting alone led to a reduction in circulating Hmgb1 associated with cytoplasmic Hmgb1 translocation, aggregate formation and autophagy. Inhibition of autophagy re-elevated circulating Hmgb1 and abolished protection in fasted mice, as did supplementation with Hmgb1. In vitro, Sirt1 inhibition prevented Hmgb1 translocation, leading to elevated Hmgb1 in the supernatant. In vivo, Sirt1 inhibition abrogated the fasting-induced protection, but had no effect in the presence of neutralizing Hmgb1 antibody. Conclusions: Fasting for one day protects from hepatic IR injury via Sirt1-dependent downregulation of circulating Hmgb1. The reduction in serum Hmgb1 appears to be mediated by its engagement in the autophagic response. These findings integrate Sirt1, Hmgb1 and autophagy into a common framework that underlies the anti-inflammatory properties of shortterm fasting. KEYWORDS: innate immunity; anti-inflammatory; liver ischemia; Kupffer cell, autophagy, Hmgb1, sirtuin, fasting.
3 Introduction
Calorie restriction has profound effects on organisms, the most remarkable of which is the impact on longevity. The extension of life span through a limited calorie intake has been documented in a variety of species ranging from yeast to primates [1-3]. The longevity response is believed to be the result of an evolutionary adaption to deal with periods of food scarcity. In particular, the ability to extend life span in times of shortage may have been selected by the need to delay reproduction until conditions are sufficiently prosperous [4]. As for longevity, calorie restriction can impart beneficial effects on various physiological processes deteriorating with age. The underlying mechanisms are thought to involve changes in energy production and utilization, in the handling of oxidative stress, insulin sensitivity, inflammatory responses and alterations in the communication between cells and organs [4, 5]. Many of these processes are regulated by signaling molecules able to sense the caloric state. Sirtuins are famous examples of such molecules, among which Sirt1 is the most conserved and best characterized. Sirt1 acts as a deacetylase dependent on NAD+, which tends to accumulate during a restricted access to calories [6]. Ischemia reperfusion (IR) is a major contributor to postoperative complications arising from surgery that requires the transient interruption of blood flow to organs. In liver exposed to IR during transplantation or major resection, tissue injury is thought to result from an excessive activation of innate immunity following reperfusion [7]. Upon re-oxygenation, a burst of reactive oxygen species (ROS) activates Kupffer cells (KCs, liver resident macrophages), which then attract neutrophils to infiltrate the liver and damage the parenchyma. The release of DAMPs (damage-associated molecular patterns), particularly Hmgb1, from immune cells and injured liver cells further activates KCs, inducing a selfperpetuating cycle that may cause widespread cell death in the liver [8]. Hmgb1 neutralization during liver IR induces marked protection, emphasizing the central role of innate immunity in
4 the acquisition of injury [9, 10]. Apart from anti-inflammatory measures, other strategies can likewise protect from ischemia reperfusion injury (IRI); ATP supplementation during IR limits hepatic damage, suggesting injury can be prevented by the provision of sufficient energy [11]. Of note, the sensitivity of organs towards IR rises with age [12, 13], and calorie restriction is able to mitigate ischemic organ injury [14-17]. Considering its success in experimental models, preoperative calorie restriction may improve the outcome of postoperative complications associated with IR. However, calorie restriction must be applied for a period of months, limiting its applicability. Apart from issues with compliance, longerterm calorie restriction is not always beneficial, as it comes along with a significant weight loss, a potential proneness to infection, has been associated with mood disorders, physical fragility, and an impaired quality of life [4, 5]. On the other hand, fasting (i.e. starvation) for a few days may confer benefits similar to those of longer-term calorie restriction, although the underlying mechanisms appear to be different and less well understood [5]. Importantly, fasting has not been associated with the undesired, chronic effects of calorie restriction [5]. Given it is also effective in protecting from organ IRI [18], short-term fasting may be a feasible strategy to decrease postoperative complications following surgery. To support its clinical translation, the mechanisms behind the fasting-induced protection against IR need to be elucidated. The aim of this study was to investigate how short-term fasting affords protection from IRI. Using a model of hepatic IR, we examined whether fasting may be protective because of its effects on energy metabolism or inflammation, two processes central to the acquisition of ischemic organ injury.
5 Materials and Methods
Animals Male wild-type mice (C57BL6; Harlan, Horst NL) were used for all experiments. Animals were fed a standard laboratory diet and kept under constant environmental conditions. Fasting was performed by moving mice to clean cages with access to water only. Fasting was started at 9am and operations were performed between 8am and 12am. All experimental procedures were approved by the Swiss animal welfare authorities and performed in accordance with the institutional animal care guidelines.
Partial hepatic inflow occlusion A 70% segmental ischemia model was used as previously described [19]. Hepatic inflow to the median and left lobes was occluded by application of micro-vascular clamps (Aesculap, San Francisco, CA) under isoflurane/O2 inhalation anesthesia. During ischemia, the abdomen was closed by a single running suture and animals allowed to wake up. Ischemia was performed for 60 minutes and terminated by removing the clamp.
6 Results
Short-term, but not long-term fasting is protective To evaluate whether fasting protects from IRI, C57BL/6, mice were fasted for one, two or three days prior to the ischemic insult. Following partial ischemia for 60 min, tissue was evaluated 6h post reperfusion. Fasting for one day (1d-fasting) - but not for two or three days - prior to surgery significantly decreased serum AST/ALT levels, the necrotic area and the number of TUNEL-positive hepatocytes (Fig. 1A.). These findings indicate that 1d-fasting protects from IRI, whereas this effect is lost upon longer fasting.
Fasting-mediated protection does not correlate with alterations in energy resources Ischemic injury is associated with high energy demands [11, 12], while short-term fasting may liberate energy resources [20]. We thus examined energy-related parameters in our model of hepatic IRI, including body weight, liver weight, hepatic glycogen stores, serum glucose, and serum and hepatic triglycerides. None of these was associated with the protective effects of fasting (Fig. S1.). Hepatic ATP content was decreased after 1d-fasting (Fig. S1.), although such a reduction would be expected to promote, rather than protect from IRI [11, 12]. Furthermore, in vivo inhibition of ghrelin, which is known to be upregulated by fasting and to protect against IR [21, 22], had no impact on IRI following 1d-fasting (Fig. S1.). Finally, we did not observe protection-associated changes in two key regulators of cellular energy resources, Pparg and Ampk (Fig. S1.). Taken together, we failed to show a correlation between the investigated energy parameters and fasting-induced protection.
7 Fasting-mediated protection correlates with mitigation of the inflammatory response Inflammatory processes are central to the acquisition of IRI [7, 23]. We measured in fasted mice pro-inflammatory cytokine expression following IR. Tnfα, Il1b, Il6 and Ccl2 were significantly reduced in 1d-fasted mice relative to controls (Fig. 1B.). No significant change was noted in mice fasted for two or three days (Fig. 1B.), revealing an association between the reduced inflammatory response and the fasting-mediated protection from IRI. In agreement, both the expression of the neutrophil marker Cd14 and the number of Mpopositive neutrophils were reduced in liver after one but not two or three days of fasting (Fig. 1B.).
Fasting dampens the responsiveness of inflammatory macrophages Following reperfusion, neutrophils are thought to be attracted to the liver through proinflammatory signaling mainly from Kupffer cells (KCs). Given that inactivation of KCs can protect from hepatic IRI [24], we investigated whether KCs are involved in the effects of fasting. First, we found fasting had no impact on the number of hepatic KC (Fig. 1B., F4/80). We next tested whether fasting affects the responsiveness of inflammatory cells in vitro. Fasting was simulated by incubation of the macrophage-like cell line Raw264-7 with descending serum concentrations over night. Cell viability was not affected (data not shown), however, Tnfα-secretion and intracellular ROS production decreased with the serum concentration in both unstimulated and LPS-challenged cells (Fig. 2A.). The reduced inflammatory response and neutrophil recruitment observed in the fasted liver after IR (Fig. 1B.) are consistent with a dampened activity of resident KCs, suggesting parenchymal damage is limited by fasting due to a suppressed activation of innate immunity through KCs.
8 Fasting protects against IR by decreasing circulating Hmgb1 levels A key signal for the activation of KCs following IR is provided by the DAMP Hmgb1 [25, 26]. Serum Hmgb1 levels were significantly reduced by 1d-fasting before (Fig. 2B.) and after IR (Fig. S2.), and correlated with the protective effects of fasting. To assess whether Hmgb1 reduction is causally related to fasting-mediated protection, recombinant Hmgb1 was injected i.v. into mice prior to IR. Hmgb1 supplementation had no impact on serum AST/ALT levels in unfasted mice subjected to IR. In contrast, Hmgb1 injection completely abolished the reduction in serum AST/ALT levels of fasted mice (Fig. 2B.). Moreover, Hmgb1 addition to fasted macrophage cultures re-installed an inflammatory phenotype (Fig. S2.). Therefore, 1dfasting seems to mitigate IRI because it attenuates KC activity through the lowering of Hmgb1 levels into circulation. In unfasted control animals, Hmgb1 was predominantly localized to the nucleus (Fig. 2C.). 1d-fasting induced cytoplasmic translocations of the protein, with particularly strong staining in sinusoid-associated KCs (Fig. 2C.). Cytoplasmic localization is associated with either Hmgb1 secretion or with the autophagic response to starvation [9] [27]. Since Hmgb1 release (i.e. circulating levels) was reduced after 1d-fasting, we examined an involvement of autophagy. Compared to unfasted controls, liver of animals fasted for one or two days had upregulated LC3 band II (Fig. 2D.) and beclin1 (Fig. S4.), indicating increased autophagy. Reduced Hmgb1 release during autophagy has been associated with the formation of Hmgb1 aggregates that cannot be secreted but instead are degraded [28]. Whilst intracellular Hmgb1 levels did not markedly change during the first two days of fasting, Hmgb1 aggregates accumulated particularly after 1d-fasting (Fig. S4.), correlating with reduced serum Hmgb1. Therefore, enhanced intracellular Hmgb1 aggregation after 1d-fasting may explain the reduced Hmgb1 release following one, but not two or three days of fasting.
9 We conclude that cytoplasmic Hmgb1 localization following 1d-fasting is associated with the authophagic response rather than the extracellular release of the molecule.
Fasting-induced protection is mediated by Sirt1 A prominent molecule that regulates responses to starvation - including the induction of the autophagic response - is Sirt1 [29]. Although Sirt1 expression was unaltered, 1d-fasting upregulated hepatic Sirt1 activity (Fig. 3A.). To assess whether Sirt1 activity contributes to fasting-mediated protection, we treated mice with the specific Sirt1 antagonist Ex527 [30]. Ex527 given before and during fasting severely abolished protection from IRI, but had no impact on AST/ALT levels in unfasted controls (Fig. 3A.). These results suggest fasting mediates most of its protective effects through the rise in Sirt1 activity.
Fasting-associated changes in Hmgb1 are regulated through Sirt1 Immunofluorescence demonstrated strong expression of Sirt1 in sinusoidal KCs (Fig. 3B.). We therefore investigated whether macrophage Sirt1 activity is required for the fasting-induced alterations in Hmgb1. Control and serum-starved Raw264-7 cells were treated with Ex527, and Hmgb1 was examined by immunofluorescence. In control cells, Hmgb1 was mainly localized to the nucleus. 1d-serum starvation promoted Hmgb1 cytoplasmic localization, an effect that was prevented by Ex527 pretreatment (Fig. 3C. & Fig. S3.). To test whether Sirt1 activity impacts on Hmgb1 release, control cells were treated with the Sirt1 agonist resveratrol. Upon treatment, the Hmgb1 concentration in the supernatant decreased by about half. Vice versa, Sirt1 inhibition in starved cells elevated Hmgb1 supernatant levels to those of untreated controls (Fig. 3C.). Therefore, cytoplasmic localization of Hmgb1 in our in vitro model was associated with an inhibition of extracellular Hmgb1 release. Consistent with a participation of Hmgb1 in autophagic processes, we observed co-localization of Hmbg1 with Beclin1 in punctae of starved cells [27](Fig. 3C.).
10 To assess whether fasting-induced protection through Sirt1 is dependent on Hmgb1 also in vivo, we pretreated fasted mice with Ex527 and/or neutralizing Hmgb1 antibodies before exposure to IR. Anti-Hmgb1 treatment alone had no significant impact on AST levels in fasted animals. However, Ex527 increased IRI in fasted mice, an effect that was prevented by concomitant αHmgb1 treatment (Fig. 3D.). The autophagy inhibitor SP600125 likewise increased IRI in fasted mice (Fig. 3D), along with an increase in serum Hmgb1 both before and after IR (Fig. S4.). These findings suggest a central role for autophagy in the fastingmediated Hmgb1 reduction and protection. Fasting hence mediates its protective effects via an Hmgb1-dependent process that is regulated through Sirt1 and involves autophagy.
11 Discussion In this study, we found that fasting confers protection against hepatic IR by dampening the response of innate immune cells. Key to the attenuation of innate immunity is the reduction in circulating Hmgb1 levels through fasting, a process dependent on fastingassociated Sirt1 activity. Our results underscore a hitherto unrecognized link between fasting, Sirt1, and Hmgb1, extending previous evidence for a role of fasting in innate immunity [3133]. Unable to find a meaningful association between energy metabolism and protection from IRI following fasting, we focused on innate immunity, chiefly responsible for the tissue injury observed in organs exposed to IR [34]. Indeed, both the expression of all investigated proinflammatory cytokines and the number of neutrophils having infiltrated the liver correlated with the protective effects of fasting. Although the number of F4/80-positive KCs, the main producers of pro-inflammatory signals in the liver, did not change upon fasting, macrophage activity in vitro was suppressed by serum starvation, as reflected in a withering Tnfα/ROS production and a reduced responsiveness to LPS. In support of a suppressed KC activity following fasting in vivo was the observation of reduced levels of circulating Hmgb1, a ligand of TLR4 [35], a potent activator of KCs [36], and an early mediator of inflammation in hepatic IR [10]. When recombinant Hmgb1 was injected into mice, the protective effects of fasting were overridden, providing evidence for a key role of reduced Hmgb1 levels in the fasting-induced protection. Likewise, Hmgb1 addition to starved macrophages led to a regain of the inflammatory phenotype. Overall, these findings suggest fasting lowers circulating Hmgb1 levels, leading to reduced activation of KCs upon inflammatory conditions such as IR, and consequently less neutrophil infiltration and tissue damage. Extracellular secretion of Hmgb1 requires its displacement from the nucleus into the cytosol. Unexpectedly, we observed an increased cytoplasmic Hmgb1 expression in liver following fasting, and hence investigated whether Hmgb1 may not be released from the
12 cytoplasm due to its engagement in autophagy, a classic response to fasting [27, 37]. Autophagy was indicated by increased levels of LC3 fragment II [38], beclin1 and Jnk1 in fasted relative to unfasted liver. Importantly, the levels of autophagy-associated intracellular Hmgb1 aggregates [28] inversely correlated with serum Hmgb1 levels. Given that protein aggregates are usually insoluble [39] and hence unreleasable, the accumulation of Hmgb aggregates provides an explanation for the reduction in serum Hmgb1 specifically after 1dfasting. Indeed, blocking of autophagy (SP600125) in 1d-fasted mice caused a re-elevation of serum Hmgb1 before and after IR, along with increased IRI. These findings suggest a central role for autophagy in fasting-induced protection and the associated alterations in serum Hmgb1. Unlike in vitro [27], cytoplasmic Hmgb1 may not be a promoter of autophagy in vivo [40]. Indeed, the association of Hmgb1 with Beclin1 may serve only for its autophagic degradation. This is supported by our finding of Hmgb1 protein aggregates [28], and further strengthens the view that the autophagic engagement of Hmgb1 counteracts its extracellular release. Although KCs are main mediators of IRI, hepatocytes also demonstrated cytoplasmic Hmgb1 translocation following fasting, suggesting their autophagic response may contribute to the abated Hmgb1 concentration in the serum. Indeed, recent findings indicate hepatocytes have an active, proinflammatory role during IR, with hepatocyte TLR4 being required for active Hmgb1 release and maximal IR injury [41]. Given that TLR4 signals also through Jnk1, SP600125 will not only inhibit fasting-induced autophagy, but also active Hmgb1 release following reperfusion [41]. As such, the loss of protection from IRI following SP600125 pretreatment seems to be mostly due to inhibition of autophagy, because reduced Hmgb1 release from SP600125-treated hepatocytes after IR should be protective. Indeed, hepatocyte autophagy may counteract cell death after IR and hence directly protect from IRI [13]. In turn, protection of hepatocytes from death would reduce passive DAMP release and thereby counteract the inflammatory cascades during IR. Therefore, it is likely that both
13 reduced serum Hmgb1 pre IR and autophagic protection from cell death post IR contribute to the anti-inflammatory effects of fasting. Of note, autophagic LC3 fragment II and beclin1 were elevated also in liver fasted for two days, yet protection from IRI was lost again at this time point. This seeming discrepancy may be explained by the re-elevation in serum Hmgb1 and the reduction in intracellular Hmgb1 aggregates after 2d- relative to 1d-fasting (considering that protection is lost upon concomitant Hmgb1 injection into 1d-fasted mice). Furthermore, human liver displays an accumulation of autophagosomes along with signs of increased cell permeability and elevations in liver enzymes following prolonged starvation [42]. Analogously, we observed an increase in AST, ALT and the cell death marker LDH in mouse liver fasted for two/three days compared to 1d-fasted or unfasted mice (Suppl. Fig. S5). Conceivably, increased liver injury before IR will counteract fasting-induced protection, providing an additional reason for the lost protection upon 2d-fasting. Moreover, leaky cell membranes will favor the passive release of Hmgb1, likely adding to the re-elevation in serum Hmgb1 after 2d-fasting. However, Michell et al [18] have found 2d/3d-fasting of mice even more protective against IR than 1d-fasting. We currently cannot explain their different findings, but reasons may relate to a different experimental set up, the use of different litter (that might be eaten by starved animals), or a different level of fecal ingestion. In our hands, only fasting for one day was consistently protective. Searching for molecules that may regulate the fasting-mediated alterations in Hmgb1, we investigated Sirt1 because of its role in fasting-induced autophagy [29]. To antagonize Sirt1 activity, we treated mice with Ex527, a highly selective inhibitor of Sirt1 (with an affinity for Sirt1 ≥200x stronger than for other deacetylases) [43]. Ex527 abolished most of the protection afforded by fasting; moreover, Ex527 prevented the nuclear-to-cytoplasmic translocation of Hmgb1 in starved macrophage cultures and increased its secretion into the supernatant, an effect that was reversed upon Sirt1 agonism [44]. These findings indicate that
14 Sirt1 activity acts upstream of Hmgb1 to regulate its cellular localization and release. Indeed, antibodies neutralizing serum Hmgb1 re-installed the fasting-induced protection in animals with inhibited Sirt1. Altogether, our results suggest a model whereby activation of Sirt1 by short fasting favors cytoplasmic localization of Hmgb1 to absorb the DAMP into autophagic processes, including aggregate formation. This Hmgb1 engagement hinders its release into circulation and hence dampens the activation of inflammatory cells, eventually providing protection from inflammatory insults such as hepatic IR (Fig. 4.). How Sirt1 may regulate Hmgb1 localization/release is currently unclear. Given that Hmgb1 nuclear-to-cytoplasmic shuttling is affected by its acetylation status [45], we examined whether Sirt1 might directly deacetylate Hmgb1. Immuno-precipitation assays and in vitro deacetylation assays did not reveal a change in Hmgb1 acetylation upon Sirt1 activation (Suppl. Fig. S3.). We thus favor an indirect interaction between Sirt1 and Hmgb1. One possibility how Sirt1 might promote Hmgb1 cytoplasmic translocation may rely on the ability of Sirt1 to deacetylate and inhibit p53 [46]. p53 in turn has been shown to bind Hmgb1 and inhibit its autophagic translocation into the cytosol [47]. Thus, Sirt1 might induce cytoplasmic Hmgb1 by inhibiting p53. Akin to fasting, Sirt1 has also been implicated in the protection of organs [48-50] from IRI. The protective properties of Sirt1 in cardiac IR, for example, have further been related to the induction of autophagy [51], which has independently been shown to minimize hepatic IRI [13]. The sensitivity of liver towards IRI is known to increase with age, an effect that has been attributed to the age-associated loss of Atg4B and the concomitant deficiency in autophagy [13]. Interestingly, autophagy has been found to provide protection via Hmgb1; non-toxic doses of cisplatin mitigate hepatic IRI by inducing autophagy and hindering Hmgb1 extracellular release [52]. Short-term fasting represents an attractive strategy to improve the outcome of surgery that requires the transient interruption of blood flow. Considering the universal function of
15 Hmgb1 in the activation of innate immunity, short-term fasting may be beneficial in many other settings, such as in viral or autoimmune disease. For the clinical translation of fasting, however, the optimal time windows and the resulting decrease in circulating Hmgb1 will need prior determination in relevant patient populations. The relative drop in Hmgb1 levels might serve as a suitable marker to estimate the efficacy of preconditioning strategies in general. In summary, we show that (i) short-term fasting for one day reduces circulating Hmgb1 levels, dampening the activity of liver-resident macrophages; (ii) fasting-associated Sirt1 activity is required to inhibit Hmgb1 extracellular release; (iii) the engagement of cytoplasmic Hmgb1 in the autophagic response and the formation of protein aggregates appears to inhibit the secretion of the molecule into circulation; and (iv) these fasting-induced alterations counteract the mobilization of innate immunity, thereby conferring protection from inflammatory insults. Altogether, our observations on Hmgb1 provide a link that places Sirt1 and autophagy into a common framework underlying the anti-inflammatory properties of short-term fasting.
16 Acknowledgements
We thank Martha Bain for excellent technical assistance.
17 References
[1] Imai S, Armstrong CM, Kaeberlein M, Guarente L. Transcriptional silencing and longevity protein Sir2 is an NAD-dependent histone deacetylase. Nature 2000;403:795-800. [2] Tissenbaum HA, Guarente L. Increased dosage of a sir-2 gene extends lifespan in Caenorhabditis elegans. Nature 2001;410:227-230. [3] Colman RJ, Anderson RM, Johnson SC, Kastman EK, Kosmatka KJ, Beasley TM, et al. Caloric restriction delays disease onset and mortality in rhesus monkeys. Science 2009;325:201-204. [4] Libert S, Guarente L. Metabolic and neuropsychiatric effects of calorie restriction and sirtuins. Annu Rev Physiol 2013;75:669-684. [5] Lee C, Longo VD. Fasting vs dietary restriction in cellular protection and cancer treatment: from model organisms to patients. Oncogene 2011;30:3305-3316. [6] Canto C, Auwerx J. Caloric restriction, SIRT1 and longevity. Trends Endocrinol Metab 2009;20:325-331. [7] Zhai Y, Busuttil RW, Kupiec-Weglinski JW. Liver ischemia and reperfusion injury: new insights into mechanisms of innate-adaptive immune-mediated tissue inflammation. Am J Transplant 2011;11:1563-1569. [8] Mosher B, Dean R, Harkema J, Remick D, Palma J, Crockett E. Inhibition of Kupffer cells reduced CXC chemokine production and liver injury. J Surg Res 2001;99:201-210. [9] Ulloa L, Messmer D. High-mobility group box 1 (HMGB1) protein: friend and foe. Cytokine Growth Factor Rev 2006;17:189-201. [10] Tsung A, Sahai R, Tanaka H, Nakao A, Fink MP, Lotze MT, et al. The nuclear factor HMGB1 mediates hepatic injury after murine liver ischemia-reperfusion. J Exp Med 2005;201:1135-1143. [11] Selzner N, Selzner M, Jochum W, Clavien PA. Ischemic preconditioning protects the steatotic mouse liver against reperfusion injury: an ATP dependent mechanism. J Hepatol 2003;39:55-61. [12] Selzner M, Selzner N, Jochum W, Graf R, Clavien PA. Increased ischemic injury in old mouse liver: an ATP-dependent mechanism. Liver Transpl 2007;13:382-390. [13] Wang JH, Ahn IS, Fischer TD, Byeon JI, Dunn WA, Jr., Behrns KE, et al. Autophagy suppresses age-dependent ischemia and reperfusion injury in livers of mice. Gastroenterology 2011;141:2188-2199. [14] Yu ZF, Mattson MP. Dietary restriction and 2-deoxyglucose administration reduce focal ischemic brain damage and improve behavioral outcome: evidence for a preconditioning mechanism. J Neurosci Res 1999;57:830-839. [15] Chandrasekar B, Nelson JF, Colston JT, Freeman GL. Calorie restriction attenuates inflammatory responses to myocardial ischemia-reperfusion injury. Am J Physiol Heart Circ Physiol 2001;280:H2094-2102. [16] Kawai SI, Vora S, Das S, Gachie E, Becker B, Neufeld AH. Modeling of risk factors for the degeneration of retinal ganglion cells after ischemia/reperfusion in rats: effects of age, caloric restriction, diabetes, pigmentation, and glaucoma. Faseb J 2001;15:1285-1287. [17] Ahmet I, Wan R, Mattson MP, Lakatta EG, Talan M. Cardioprotection by intermittent fasting in rats. Circulation 2005;112:3115-3121. [18] Mitchell JR, Verweij M, Brand K, van de Ven M, Goemaere N, van den Engel S, et al. Short-term dietary restriction and fasting precondition against ischemia reperfusion injury in mice. Aging Cell 2010;9:40-53. [19] Jang JH, Moritz W, Graf R, Clavien PA. Preconditioning with death ligands FasL and TNF-alpha protects the cirrhotic mouse liver against ischaemic injury. Gut 2008;57:492-499.
18 [20] Sokolovic M, Sokolovic A, Wehkamp D, Ver Loren van Themaat E, de Waart DR, Gilhuijs-Pederson LA, et al. The transcriptomic signature of fasting murine liver. BMC Genomics 2008;9:528. [21] Tschop M, Smiley DL, Heiman ML. Ghrelin induces adiposity in rodents. Nature 2000;407:908-913. [22] Takeda R, Nishimatsu H, Suzuki E, Satonaka H, Nagata D, Oba S, et al. Ghrelin improves renal function in mice with ischemic acute renal failure. J Am Soc Nephrol 2006;17:113-121. [23] Abu-Amara M, Yang SY, Tapuria N, Fuller B, Davidson B, Seifalian A. Liver ischemia/reperfusion injury: processes in inflammatory networks--a review. Liver Transpl 2010;16:1016-1032. [24] Suzuki S, Toledo-Pereyra LH, Rodriguez F, Lopez F. Role of Kupffer cells in neutrophil activation and infiltration following total hepatic ischemia and reperfusion. Circ Shock 1994;42:204-209. [25] Yang H, Tracey KJ. Targeting HMGB1 in inflammation. Biochim Biophys Acta 2010;1799:149-156. [26] Castiglioni A, Canti V, Rovere-Querini P, Manfredi AA. High-mobility group box 1 (HMGB1) as a master regulator of innate immunity. Cell Tissue Res 2011;343:189-199. [27] Tang D, Kang R, Livesey KM, Cheh CW, Farkas A, Loughran P, et al. Endogenous HMGB1 regulates autophagy. J Cell Biol 2010;190:881-892. [28] Zhu S, Li W, Li J, Jundoria A, Sama AE, Wang H. It Is Not Just Folklore: The Aqueous Extract of Mung Bean Coat Is Protective against Sepsis. Evid Based Complement Alternat Med 2012;2012:498467. [29] Lee IH, Cao L, Mostoslavsky R, Lombard DB, Liu J, Bruns NE, et al. A role for the NAD-dependent deacetylase Sirt1 in the regulation of autophagy. Proc Natl Acad Sci U S A 2008;105:3374-3379. [30] Peck B, Chen CY, Ho KK, Di Fruscia P, Myatt SS, Coombes RC, et al. SIRT inhibitors induce cell death and p53 acetylation through targeting both SIRT1 and SIRT2. Mol Cancer Ther 2010;9:844-855. [31] Sankary HN, Chong A, Foster P, Brown E, Shen J, Kimura R, et al. Inactivation of Kupffer cells after prolonged donor fasting improves viability of transplanted hepatic allografts. Hepatology 1995;22:1236-1242. [32] Ritz BW, Aktan I, Nogusa S, Gardner EM. Energy restriction impairs natural killer cell function and increases the severity of influenza infection in young adult male C57BL/6 mice. J Nutr 2008;138:2269-2275. [33] Shen J, Ren H, Tomiyama-Miyaji C, Watanabe M, Kainuma E, Inoue M, et al. Resistance and augmentation of innate immunity in mice exposed to starvation. Cell Immunol 2009;259:66-73. [34] van Golen RF, van Gulik TM, Heger M. The sterile immune response during hepatic ischemia/reperfusion. Cytokine Growth Factor Rev 2012;23:69-84. [35] Yu M, Wang H, Ding A, Golenbock DT, Latz E, Czura CJ, et al. HMGB1 signals through toll-like receptor (TLR) 4 and TLR2. Shock 2006;26:174-179. [36] Frink M, Hsieh YC, Thobe BM, Choudhry MA, Schwacha MG, Bland KI, et al. TLR4 regulates Kupffer cell chemokine production, systemic inflammation and lung neutrophil infiltration following trauma-hemorrhage. Mol Immunol 2007;44:2625-2630. [37] Cantino D, Mosso R, Baccino FM. Changes induced by fasting and cycloheximide in the vacuolar apparatus of rat hepatocytes. A morphometric investigation. Boll Soc Ital Biol Sper 1979;55:1884-1889. [38] Mizushima N, Yoshimori T. How to interpret LC3 immunoblotting. Autophagy 2007;3:542-545.
19 [39] Powell SR, Divald A. The ubiquitin-proteasome system in myocardial ischaemia and preconditioning. Cardiovasc Res 2010;85:303-311. [40] Huebener P, Gwak GY, Pradere JP, Quinzii CM, Friedman R, Lin CS, et al. Highmobility group box 1 is dispensable for autophagy, mitochondrial quality control, and organ function in vivo. Cell Metab 2014;19:539-547. [41] Nace GW, Huang H, Klune JR, Eid RE, Rosborough BR, Korff S, et al. Cellularspecific role of toll-like receptor 4 in hepatic ischemia-reperfusion injury in mice. Hepatology 2013;58:374-387. [42] Rautou PE, Cazals-Hatem D, Moreau R, Francoz C, Feldmann G, Lebrec D, et al. Acute liver cell damage in patients with anorexia nervosa: a possible role of starvationinduced hepatocyte autophagy. Gastroenterology 2008;135:840-848. [43] Zhao X, Allison D, Condon B, Zhang F, Gheyi T, Zhang A, et al. The 2.5 A crystal structure of the SIRT1 catalytic domain bound to nicotinamide adenine dinucleotide (NAD+) and an indole (EX527 analogue) reveals a novel mechanism of histone deacetylase inhibition. J Med Chem 2013;56:963-969. [44] Timmers S, Konings E, Bilet L, Houtkooper RH, van de Weijer T, Goossens GH, et al. Calorie restriction-like effects of 30 days of resveratrol supplementation on energy metabolism and metabolic profile in obese humans. Cell Metab 2011;14:612-622. [45] Bonaldi T, Talamo F, Scaffidi P, Ferrera D, Porto A, Bachi A, et al. Monocytic cells hyperacetylate chromatin protein HMGB1 to redirect it towards secretion. Embo J 2003;22:5551-5560. [46] Lee JT, Gu W. SIRT1: Regulator of p53 Deacetylation. Genes Cancer 2013;4:112117. [47] Livesey KM, Kang R, Vernon P, Buchser W, Loughran P, Watkins SC, et al. p53/HMGB1 complexes regulate autophagy and apoptosis. Cancer Res 2012;72:1996-2005. [48] Nadtochiy SM, Redman E, Rahman I, Brookes PS. Lysine deacetylation in ischaemic preconditioning: the role of SIRT1. Cardiovasc Res 2011;89:643-649. [49] Hsu CP, Zhai P, Yamamoto T, Maejima Y, Matsushima S, Hariharan N, et al. Silent information regulator 1 protects the heart from ischemia/reperfusion. Circulation 2010;122:2170-2182. [50] Fan H, Yang HC, You L, Wang YY, He WJ, Hao CM. The histone deacetylase, SIRT1, contributes to the resistance of young mice to ischemia/reperfusion-induced acute kidney injury. Kidney Int 2013;83:404-413. [51] Nadtochiy SM, Yao H, McBurney MW, Gu W, Guarente L, Rahman I, et al. SIRT1mediated acute cardioprotection. Am J Physiol Heart Circ Physiol 2011;301:H1506-1512. [52] Cardinal J, Pan P, Dhupar R, Ross M, Nakao A, Lotze M, et al. Cisplatin prevents high mobility group box 1 release and is protective in a murine model of hepatic ischemia/reperfusion injury. Hepatology 2009;50:565-574.
20 Figure Legends
Figure 1. One-day fasting protects mouse liver from ischemia reperfusion injury and correlates with a repressed inflammatory response. (A) Levels of liver injury markers (AST/ALT) and parenchymal cell death (histological necrosis, TUNEL-positive cells, see representative examples at bottom) following fasting are shown. (B) Inflammatory gene expression (Tnfa, Il6, Il1b, Cd14, Ccl2) and histological quantification of Mpo and F4/80 assessed at 6h post reperfusion. Expression data was normalized to unfasted controls. *P<0.05 with n=8/group.
Figure 2. Serum starvation dampens macrophage activity and lowers circulating Hmgb1 levels. (A) Secreted Tnfa and intracellular ROS levels of Raw264-7 cultures following overnight exposure to descending concentrations of serum (upper panels) or at 10% (control) and 0.1% (fasting) serum challenged with ascending LPS doses (lower panels). (B) Serum levels of Hmgb1 in fasted mice and AST levels at 6h post reperfusion in unfasted controls and 1d-fasted mice injected with NaCl or Hmgb1 prior to ischemia. (C) Immunofluorescence (top) and immunohistochemistry (bottom) showing hepatic Hmgb1 expression in unfasted controls and 1d-fasted mice. Kupffer cells are marked with red/black arrowheads. Note the strong Hmgb1 expression in KCs following fasting. (D) Hepatic LC3 immunoblots and quantification of hepatic LC3 II fragments in fasted mice (loading control β-tubulin). *P<0.05 with n=6/group.
Figure 3. IRI protection through fasting is dependent on Hmgb1 regulation through Sirt1. (A) Sirt1 mRNA/protein expression, NAD+-dependent histone deacetylase activity, and AST levels (6h post reperfusion) following Sirt1 inhibition are shown for control and fasted mice. (B) Sirt1 immunochemical localization (Sirt1 green, nuclei blue) in fasted liver. (C)
21 Raw264-7 cells were cultured with 10% serum, 0.1% serum, or 0.1% serum plus Ex527 overnight and stained for Hmgb1. Phalloidin staining (F-aktin, blue) was used to visualize cell borders. Graphs to the left show supernatant Hmgb1 concentration in response to resveratrol (10% FCS) or Ex527 (0.1% FCS). Right bottom panel: Immunofluorescence demonstrating colocalization of Hmgb1 (green) with Beclin1 (red) in starved cells (blue nuclei). (D) Serum AST levels at 6h post reperfusion in mice fasted for one day prior to ischemia. Right panel: treatment with neutralizing Hmgb1 antibody reduced AST only in mice pretreated with Ex527. Left panel: inhibition of autophagy with SP600125 abolished the protection afforded by fasting. *P<0.05 with n=6/group.
Figure 4. Model for the protective effects fasting has on ischemia reperfusion injury. DAMPs and ROS generated during IR activate KCs, which in response produce ROS and actively secrete Hmgb1 to potentiate their activation. Neutrophils, attracted by KCs, damage liver tissue, further increasing circulating Hmgb1 and propagating KC activation. Fasting activates Sirt1, thereby engaging Hmgb1 into the autophagic response and lowering its extracellular release. The reduced levels of circulating Hmgb1 dampen the selfpropagation of KCs and hence protect from IRI. Prolonged fasting, however, may lead to increased membrane leakiness due to autophagic cell death, again increasing the levels of circulating Hmgb1 and neutralizing the protective effects of the early autophagic response.
22
23
24
25
S0168-8278(14)00263-3 http://dx.doi.org/10.1016/j.jhep.2014.04.010 JHEPAT 5112
To appear in:
Journal of Hepatology
Received Date: Revised Date: Accepted Date:
20 August 2013 4 April 2014 10 April 2014
Please cite this article as: Rickenbacher, A., Jang, J., Limani, P., Ungethüm, U., Lehmann, K., Oberkofler, C.E., Weber, A., Graf, R., Humar, B., Clavien, P-A., Fasting protects liver from ischemic injury through Sirt1-mediated downregulation of circulating Hmgb1 in mice, Journal of Hepatology (2014), doi: http://dx.doi.org/10.1016/j.jhep. 2014.04.010
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1 JHEPAT-D-13-01488R2
Fasting protects liver from ischemic injury through Sirt1-mediated downregulation of circulating Hmgb1 in mice
Andreas Rickenbacher1, Jaehwi Jang1†, Perparim Limani1†, Udo Ungethüm1, Kuno Lehmann1, Christian E. Oberkofler1, Achim Weber2, Rolf Graf1, Bostjan Humar1*, Pierre-Alain Clavien1 *
1
Swiss Hepato-Pancreatico-Biliary Center, Department of Surgery and 2Institute of Pathology, University Hospital Zürich, CH-8091 Zürich, Switzerland *BH and PAC share senior authorship †equal contribution
Correspondence: Pierre-Alain Clavien, MD, PhD Department of Surgery University Hospital of Zurich Raemistrasse 100, 8091 Zurich, Switzerland Tel: +41 44 255 33 00 Fax: +41 44 255 44 49 e-mail: [email protected] Word count: 4470 words, 4 figures, 0 tables Abbreviations: IR: ischemia reperfusion, ROS: reactive oxygen species, DAMP: damage-associated molecular pattern, Hmgb1: high mobility group box 1, KC: Kupffer cell, IRI: ischemia reperfusion injury, ATP: adenosine triphosphate, AST: aspartate aminotransferase, ALT: alanine aminotransferase, H&E: hematoxylin and eosin, PBS: phosphate buffered saline, TUNEL: TdT-mediated dUTP-biotin nick end labeling, TNFα: tumor necrosis factor α, IL1β: interleukine 1β, IL-6: interleukine 6, MCP1: monocyte chemotactic protein-1,
Conflict of interest: The authors have nothing to disclose. Financial support: This work was supported by an Academic Research in Surgery grant to AR, a Forschungsschwerpunkt grant by the University of Zürich to PAC, BH and RG, and a Swiss National Foundation grant (#320030_132985) to PAC.
2 Abstract (word count: 250 words)
Background&Aims: Fasting and calorie restriction are associated with a prolonged life span and an increased resistance to stress. The protective effects of fasting have been exploited for the mitigation of ischemic organ injury, yet the underlying mechanisms remain incompletely understood. Here, we investigated whether fasting protects liver against ischemia reperfusion (IR) through energy-preserving or anti-inflammatory mechanisms. Methods: Fasted C57BL6 mice were subjected to partial hepatic IR. Injury was assessed by liver enzymes and histology. Raw 264-7 macrophage-like cells were investigated in vitro. Sirt1 and Hmgb1 were inhibited using Ex527 and neutralizing antibodies, respectively. Results: Fasting for one, but not two or three days, protected from hepatic IR injury. None of the investigated energy parameters correlated with the protective effects. Instead, inflammatory responses were dampened in one-day-fasted mice and in starved macrophages. Fasting alone led to a reduction in circulating Hmgb1 associated with cytoplasmic Hmgb1 translocation, aggregate formation and autophagy. Inhibition of autophagy re-elevated circulating Hmgb1 and abolished protection in fasted mice, as did supplementation with Hmgb1. In vitro, Sirt1 inhibition prevented Hmgb1 translocation, leading to elevated Hmgb1 in the supernatant. In vivo, Sirt1 inhibition abrogated the fasting-induced protection, but had no effect in the presence of neutralizing Hmgb1 antibody. Conclusions: Fasting for one day protects from hepatic IR injury via Sirt1-dependent downregulation of circulating Hmgb1. The reduction in serum Hmgb1 appears to be mediated by its engagement in the autophagic response. These findings integrate Sirt1, Hmgb1 and autophagy into a common framework that underlies the anti-inflammatory properties of shortterm fasting. KEYWORDS: innate immunity; anti-inflammatory; liver ischemia; Kupffer cell, autophagy, Hmgb1, sirtuin, fasting.
3 Introduction
Calorie restriction has profound effects on organisms, the most remarkable of which is the impact on longevity. The extension of life span through a limited calorie intake has been documented in a variety of species ranging from yeast to primates [1-3]. The longevity response is believed to be the result of an evolutionary adaption to deal with periods of food scarcity. In particular, the ability to extend life span in times of shortage may have been selected by the need to delay reproduction until conditions are sufficiently prosperous [4]. As for longevity, calorie restriction can impart beneficial effects on various physiological processes deteriorating with age. The underlying mechanisms are thought to involve changes in energy production and utilization, in the handling of oxidative stress, insulin sensitivity, inflammatory responses and alterations in the communication between cells and organs [4, 5]. Many of these processes are regulated by signaling molecules able to sense the caloric state. Sirtuins are famous examples of such molecules, among which Sirt1 is the most conserved and best characterized. Sirt1 acts as a deacetylase dependent on NAD+, which tends to accumulate during a restricted access to calories [6]. Ischemia reperfusion (IR) is a major contributor to postoperative complications arising from surgery that requires the transient interruption of blood flow to organs. In liver exposed to IR during transplantation or major resection, tissue injury is thought to result from an excessive activation of innate immunity following reperfusion [7]. Upon re-oxygenation, a burst of reactive oxygen species (ROS) activates Kupffer cells (KCs, liver resident macrophages), which then attract neutrophils to infiltrate the liver and damage the parenchyma. The release of DAMPs (damage-associated molecular patterns), particularly Hmgb1, from immune cells and injured liver cells further activates KCs, inducing a selfperpetuating cycle that may cause widespread cell death in the liver [8]. Hmgb1 neutralization during liver IR induces marked protection, emphasizing the central role of innate immunity in
4 the acquisition of injury [9, 10]. Apart from anti-inflammatory measures, other strategies can likewise protect from ischemia reperfusion injury (IRI); ATP supplementation during IR limits hepatic damage, suggesting injury can be prevented by the provision of sufficient energy [11]. Of note, the sensitivity of organs towards IR rises with age [12, 13], and calorie restriction is able to mitigate ischemic organ injury [14-17]. Considering its success in experimental models, preoperative calorie restriction may improve the outcome of postoperative complications associated with IR. However, calorie restriction must be applied for a period of months, limiting its applicability. Apart from issues with compliance, longerterm calorie restriction is not always beneficial, as it comes along with a significant weight loss, a potential proneness to infection, has been associated with mood disorders, physical fragility, and an impaired quality of life [4, 5]. On the other hand, fasting (i.e. starvation) for a few days may confer benefits similar to those of longer-term calorie restriction, although the underlying mechanisms appear to be different and less well understood [5]. Importantly, fasting has not been associated with the undesired, chronic effects of calorie restriction [5]. Given it is also effective in protecting from organ IRI [18], short-term fasting may be a feasible strategy to decrease postoperative complications following surgery. To support its clinical translation, the mechanisms behind the fasting-induced protection against IR need to be elucidated. The aim of this study was to investigate how short-term fasting affords protection from IRI. Using a model of hepatic IR, we examined whether fasting may be protective because of its effects on energy metabolism or inflammation, two processes central to the acquisition of ischemic organ injury.
5 Materials and Methods
Animals Male wild-type mice (C57BL6; Harlan, Horst NL) were used for all experiments. Animals were fed a standard laboratory diet and kept under constant environmental conditions. Fasting was performed by moving mice to clean cages with access to water only. Fasting was started at 9am and operations were performed between 8am and 12am. All experimental procedures were approved by the Swiss animal welfare authorities and performed in accordance with the institutional animal care guidelines.
Partial hepatic inflow occlusion A 70% segmental ischemia model was used as previously described [19]. Hepatic inflow to the median and left lobes was occluded by application of micro-vascular clamps (Aesculap, San Francisco, CA) under isoflurane/O2 inhalation anesthesia. During ischemia, the abdomen was closed by a single running suture and animals allowed to wake up. Ischemia was performed for 60 minutes and terminated by removing the clamp.
6 Results
Short-term, but not long-term fasting is protective To evaluate whether fasting protects from IRI, C57BL/6, mice were fasted for one, two or three days prior to the ischemic insult. Following partial ischemia for 60 min, tissue was evaluated 6h post reperfusion. Fasting for one day (1d-fasting) - but not for two or three days - prior to surgery significantly decreased serum AST/ALT levels, the necrotic area and the number of TUNEL-positive hepatocytes (Fig. 1A.). These findings indicate that 1d-fasting protects from IRI, whereas this effect is lost upon longer fasting.
Fasting-mediated protection does not correlate with alterations in energy resources Ischemic injury is associated with high energy demands [11, 12], while short-term fasting may liberate energy resources [20]. We thus examined energy-related parameters in our model of hepatic IRI, including body weight, liver weight, hepatic glycogen stores, serum glucose, and serum and hepatic triglycerides. None of these was associated with the protective effects of fasting (Fig. S1.). Hepatic ATP content was decreased after 1d-fasting (Fig. S1.), although such a reduction would be expected to promote, rather than protect from IRI [11, 12]. Furthermore, in vivo inhibition of ghrelin, which is known to be upregulated by fasting and to protect against IR [21, 22], had no impact on IRI following 1d-fasting (Fig. S1.). Finally, we did not observe protection-associated changes in two key regulators of cellular energy resources, Pparg and Ampk (Fig. S1.). Taken together, we failed to show a correlation between the investigated energy parameters and fasting-induced protection.
7 Fasting-mediated protection correlates with mitigation of the inflammatory response Inflammatory processes are central to the acquisition of IRI [7, 23]. We measured in fasted mice pro-inflammatory cytokine expression following IR. Tnfα, Il1b, Il6 and Ccl2 were significantly reduced in 1d-fasted mice relative to controls (Fig. 1B.). No significant change was noted in mice fasted for two or three days (Fig. 1B.), revealing an association between the reduced inflammatory response and the fasting-mediated protection from IRI. In agreement, both the expression of the neutrophil marker Cd14 and the number of Mpopositive neutrophils were reduced in liver after one but not two or three days of fasting (Fig. 1B.).
Fasting dampens the responsiveness of inflammatory macrophages Following reperfusion, neutrophils are thought to be attracted to the liver through proinflammatory signaling mainly from Kupffer cells (KCs). Given that inactivation of KCs can protect from hepatic IRI [24], we investigated whether KCs are involved in the effects of fasting. First, we found fasting had no impact on the number of hepatic KC (Fig. 1B., F4/80). We next tested whether fasting affects the responsiveness of inflammatory cells in vitro. Fasting was simulated by incubation of the macrophage-like cell line Raw264-7 with descending serum concentrations over night. Cell viability was not affected (data not shown), however, Tnfα-secretion and intracellular ROS production decreased with the serum concentration in both unstimulated and LPS-challenged cells (Fig. 2A.). The reduced inflammatory response and neutrophil recruitment observed in the fasted liver after IR (Fig. 1B.) are consistent with a dampened activity of resident KCs, suggesting parenchymal damage is limited by fasting due to a suppressed activation of innate immunity through KCs.
8 Fasting protects against IR by decreasing circulating Hmgb1 levels A key signal for the activation of KCs following IR is provided by the DAMP Hmgb1 [25, 26]. Serum Hmgb1 levels were significantly reduced by 1d-fasting before (Fig. 2B.) and after IR (Fig. S2.), and correlated with the protective effects of fasting. To assess whether Hmgb1 reduction is causally related to fasting-mediated protection, recombinant Hmgb1 was injected i.v. into mice prior to IR. Hmgb1 supplementation had no impact on serum AST/ALT levels in unfasted mice subjected to IR. In contrast, Hmgb1 injection completely abolished the reduction in serum AST/ALT levels of fasted mice (Fig. 2B.). Moreover, Hmgb1 addition to fasted macrophage cultures re-installed an inflammatory phenotype (Fig. S2.). Therefore, 1dfasting seems to mitigate IRI because it attenuates KC activity through the lowering of Hmgb1 levels into circulation. In unfasted control animals, Hmgb1 was predominantly localized to the nucleus (Fig. 2C.). 1d-fasting induced cytoplasmic translocations of the protein, with particularly strong staining in sinusoid-associated KCs (Fig. 2C.). Cytoplasmic localization is associated with either Hmgb1 secretion or with the autophagic response to starvation [9] [27]. Since Hmgb1 release (i.e. circulating levels) was reduced after 1d-fasting, we examined an involvement of autophagy. Compared to unfasted controls, liver of animals fasted for one or two days had upregulated LC3 band II (Fig. 2D.) and beclin1 (Fig. S4.), indicating increased autophagy. Reduced Hmgb1 release during autophagy has been associated with the formation of Hmgb1 aggregates that cannot be secreted but instead are degraded [28]. Whilst intracellular Hmgb1 levels did not markedly change during the first two days of fasting, Hmgb1 aggregates accumulated particularly after 1d-fasting (Fig. S4.), correlating with reduced serum Hmgb1. Therefore, enhanced intracellular Hmgb1 aggregation after 1d-fasting may explain the reduced Hmgb1 release following one, but not two or three days of fasting.
9 We conclude that cytoplasmic Hmgb1 localization following 1d-fasting is associated with the authophagic response rather than the extracellular release of the molecule.
Fasting-induced protection is mediated by Sirt1 A prominent molecule that regulates responses to starvation - including the induction of the autophagic response - is Sirt1 [29]. Although Sirt1 expression was unaltered, 1d-fasting upregulated hepatic Sirt1 activity (Fig. 3A.). To assess whether Sirt1 activity contributes to fasting-mediated protection, we treated mice with the specific Sirt1 antagonist Ex527 [30]. Ex527 given before and during fasting severely abolished protection from IRI, but had no impact on AST/ALT levels in unfasted controls (Fig. 3A.). These results suggest fasting mediates most of its protective effects through the rise in Sirt1 activity.
Fasting-associated changes in Hmgb1 are regulated through Sirt1 Immunofluorescence demonstrated strong expression of Sirt1 in sinusoidal KCs (Fig. 3B.). We therefore investigated whether macrophage Sirt1 activity is required for the fasting-induced alterations in Hmgb1. Control and serum-starved Raw264-7 cells were treated with Ex527, and Hmgb1 was examined by immunofluorescence. In control cells, Hmgb1 was mainly localized to the nucleus. 1d-serum starvation promoted Hmgb1 cytoplasmic localization, an effect that was prevented by Ex527 pretreatment (Fig. 3C. & Fig. S3.). To test whether Sirt1 activity impacts on Hmgb1 release, control cells were treated with the Sirt1 agonist resveratrol. Upon treatment, the Hmgb1 concentration in the supernatant decreased by about half. Vice versa, Sirt1 inhibition in starved cells elevated Hmgb1 supernatant levels to those of untreated controls (Fig. 3C.). Therefore, cytoplasmic localization of Hmgb1 in our in vitro model was associated with an inhibition of extracellular Hmgb1 release. Consistent with a participation of Hmgb1 in autophagic processes, we observed co-localization of Hmbg1 with Beclin1 in punctae of starved cells [27](Fig. 3C.).
10 To assess whether fasting-induced protection through Sirt1 is dependent on Hmgb1 also in vivo, we pretreated fasted mice with Ex527 and/or neutralizing Hmgb1 antibodies before exposure to IR. Anti-Hmgb1 treatment alone had no significant impact on AST levels in fasted animals. However, Ex527 increased IRI in fasted mice, an effect that was prevented by concomitant αHmgb1 treatment (Fig. 3D.). The autophagy inhibitor SP600125 likewise increased IRI in fasted mice (Fig. 3D), along with an increase in serum Hmgb1 both before and after IR (Fig. S4.). These findings suggest a central role for autophagy in the fastingmediated Hmgb1 reduction and protection. Fasting hence mediates its protective effects via an Hmgb1-dependent process that is regulated through Sirt1 and involves autophagy.
11 Discussion In this study, we found that fasting confers protection against hepatic IR by dampening the response of innate immune cells. Key to the attenuation of innate immunity is the reduction in circulating Hmgb1 levels through fasting, a process dependent on fastingassociated Sirt1 activity. Our results underscore a hitherto unrecognized link between fasting, Sirt1, and Hmgb1, extending previous evidence for a role of fasting in innate immunity [3133]. Unable to find a meaningful association between energy metabolism and protection from IRI following fasting, we focused on innate immunity, chiefly responsible for the tissue injury observed in organs exposed to IR [34]. Indeed, both the expression of all investigated proinflammatory cytokines and the number of neutrophils having infiltrated the liver correlated with the protective effects of fasting. Although the number of F4/80-positive KCs, the main producers of pro-inflammatory signals in the liver, did not change upon fasting, macrophage activity in vitro was suppressed by serum starvation, as reflected in a withering Tnfα/ROS production and a reduced responsiveness to LPS. In support of a suppressed KC activity following fasting in vivo was the observation of reduced levels of circulating Hmgb1, a ligand of TLR4 [35], a potent activator of KCs [36], and an early mediator of inflammation in hepatic IR [10]. When recombinant Hmgb1 was injected into mice, the protective effects of fasting were overridden, providing evidence for a key role of reduced Hmgb1 levels in the fasting-induced protection. Likewise, Hmgb1 addition to starved macrophages led to a regain of the inflammatory phenotype. Overall, these findings suggest fasting lowers circulating Hmgb1 levels, leading to reduced activation of KCs upon inflammatory conditions such as IR, and consequently less neutrophil infiltration and tissue damage. Extracellular secretion of Hmgb1 requires its displacement from the nucleus into the cytosol. Unexpectedly, we observed an increased cytoplasmic Hmgb1 expression in liver following fasting, and hence investigated whether Hmgb1 may not be released from the
12 cytoplasm due to its engagement in autophagy, a classic response to fasting [27, 37]. Autophagy was indicated by increased levels of LC3 fragment II [38], beclin1 and Jnk1 in fasted relative to unfasted liver. Importantly, the levels of autophagy-associated intracellular Hmgb1 aggregates [28] inversely correlated with serum Hmgb1 levels. Given that protein aggregates are usually insoluble [39] and hence unreleasable, the accumulation of Hmgb aggregates provides an explanation for the reduction in serum Hmgb1 specifically after 1dfasting. Indeed, blocking of autophagy (SP600125) in 1d-fasted mice caused a re-elevation of serum Hmgb1 before and after IR, along with increased IRI. These findings suggest a central role for autophagy in fasting-induced protection and the associated alterations in serum Hmgb1. Unlike in vitro [27], cytoplasmic Hmgb1 may not be a promoter of autophagy in vivo [40]. Indeed, the association of Hmgb1 with Beclin1 may serve only for its autophagic degradation. This is supported by our finding of Hmgb1 protein aggregates [28], and further strengthens the view that the autophagic engagement of Hmgb1 counteracts its extracellular release. Although KCs are main mediators of IRI, hepatocytes also demonstrated cytoplasmic Hmgb1 translocation following fasting, suggesting their autophagic response may contribute to the abated Hmgb1 concentration in the serum. Indeed, recent findings indicate hepatocytes have an active, proinflammatory role during IR, with hepatocyte TLR4 being required for active Hmgb1 release and maximal IR injury [41]. Given that TLR4 signals also through Jnk1, SP600125 will not only inhibit fasting-induced autophagy, but also active Hmgb1 release following reperfusion [41]. As such, the loss of protection from IRI following SP600125 pretreatment seems to be mostly due to inhibition of autophagy, because reduced Hmgb1 release from SP600125-treated hepatocytes after IR should be protective. Indeed, hepatocyte autophagy may counteract cell death after IR and hence directly protect from IRI [13]. In turn, protection of hepatocytes from death would reduce passive DAMP release and thereby counteract the inflammatory cascades during IR. Therefore, it is likely that both
13 reduced serum Hmgb1 pre IR and autophagic protection from cell death post IR contribute to the anti-inflammatory effects of fasting. Of note, autophagic LC3 fragment II and beclin1 were elevated also in liver fasted for two days, yet protection from IRI was lost again at this time point. This seeming discrepancy may be explained by the re-elevation in serum Hmgb1 and the reduction in intracellular Hmgb1 aggregates after 2d- relative to 1d-fasting (considering that protection is lost upon concomitant Hmgb1 injection into 1d-fasted mice). Furthermore, human liver displays an accumulation of autophagosomes along with signs of increased cell permeability and elevations in liver enzymes following prolonged starvation [42]. Analogously, we observed an increase in AST, ALT and the cell death marker LDH in mouse liver fasted for two/three days compared to 1d-fasted or unfasted mice (Suppl. Fig. S5). Conceivably, increased liver injury before IR will counteract fasting-induced protection, providing an additional reason for the lost protection upon 2d-fasting. Moreover, leaky cell membranes will favor the passive release of Hmgb1, likely adding to the re-elevation in serum Hmgb1 after 2d-fasting. However, Michell et al [18] have found 2d/3d-fasting of mice even more protective against IR than 1d-fasting. We currently cannot explain their different findings, but reasons may relate to a different experimental set up, the use of different litter (that might be eaten by starved animals), or a different level of fecal ingestion. In our hands, only fasting for one day was consistently protective. Searching for molecules that may regulate the fasting-mediated alterations in Hmgb1, we investigated Sirt1 because of its role in fasting-induced autophagy [29]. To antagonize Sirt1 activity, we treated mice with Ex527, a highly selective inhibitor of Sirt1 (with an affinity for Sirt1 ≥200x stronger than for other deacetylases) [43]. Ex527 abolished most of the protection afforded by fasting; moreover, Ex527 prevented the nuclear-to-cytoplasmic translocation of Hmgb1 in starved macrophage cultures and increased its secretion into the supernatant, an effect that was reversed upon Sirt1 agonism [44]. These findings indicate that
14 Sirt1 activity acts upstream of Hmgb1 to regulate its cellular localization and release. Indeed, antibodies neutralizing serum Hmgb1 re-installed the fasting-induced protection in animals with inhibited Sirt1. Altogether, our results suggest a model whereby activation of Sirt1 by short fasting favors cytoplasmic localization of Hmgb1 to absorb the DAMP into autophagic processes, including aggregate formation. This Hmgb1 engagement hinders its release into circulation and hence dampens the activation of inflammatory cells, eventually providing protection from inflammatory insults such as hepatic IR (Fig. 4.). How Sirt1 may regulate Hmgb1 localization/release is currently unclear. Given that Hmgb1 nuclear-to-cytoplasmic shuttling is affected by its acetylation status [45], we examined whether Sirt1 might directly deacetylate Hmgb1. Immuno-precipitation assays and in vitro deacetylation assays did not reveal a change in Hmgb1 acetylation upon Sirt1 activation (Suppl. Fig. S3.). We thus favor an indirect interaction between Sirt1 and Hmgb1. One possibility how Sirt1 might promote Hmgb1 cytoplasmic translocation may rely on the ability of Sirt1 to deacetylate and inhibit p53 [46]. p53 in turn has been shown to bind Hmgb1 and inhibit its autophagic translocation into the cytosol [47]. Thus, Sirt1 might induce cytoplasmic Hmgb1 by inhibiting p53. Akin to fasting, Sirt1 has also been implicated in the protection of organs [48-50] from IRI. The protective properties of Sirt1 in cardiac IR, for example, have further been related to the induction of autophagy [51], which has independently been shown to minimize hepatic IRI [13]. The sensitivity of liver towards IRI is known to increase with age, an effect that has been attributed to the age-associated loss of Atg4B and the concomitant deficiency in autophagy [13]. Interestingly, autophagy has been found to provide protection via Hmgb1; non-toxic doses of cisplatin mitigate hepatic IRI by inducing autophagy and hindering Hmgb1 extracellular release [52]. Short-term fasting represents an attractive strategy to improve the outcome of surgery that requires the transient interruption of blood flow. Considering the universal function of
15 Hmgb1 in the activation of innate immunity, short-term fasting may be beneficial in many other settings, such as in viral or autoimmune disease. For the clinical translation of fasting, however, the optimal time windows and the resulting decrease in circulating Hmgb1 will need prior determination in relevant patient populations. The relative drop in Hmgb1 levels might serve as a suitable marker to estimate the efficacy of preconditioning strategies in general. In summary, we show that (i) short-term fasting for one day reduces circulating Hmgb1 levels, dampening the activity of liver-resident macrophages; (ii) fasting-associated Sirt1 activity is required to inhibit Hmgb1 extracellular release; (iii) the engagement of cytoplasmic Hmgb1 in the autophagic response and the formation of protein aggregates appears to inhibit the secretion of the molecule into circulation; and (iv) these fasting-induced alterations counteract the mobilization of innate immunity, thereby conferring protection from inflammatory insults. Altogether, our observations on Hmgb1 provide a link that places Sirt1 and autophagy into a common framework underlying the anti-inflammatory properties of short-term fasting.
16 Acknowledgements
We thank Martha Bain for excellent technical assistance.
17 References
[1] Imai S, Armstrong CM, Kaeberlein M, Guarente L. Transcriptional silencing and longevity protein Sir2 is an NAD-dependent histone deacetylase. Nature 2000;403:795-800. [2] Tissenbaum HA, Guarente L. Increased dosage of a sir-2 gene extends lifespan in Caenorhabditis elegans. Nature 2001;410:227-230. [3] Colman RJ, Anderson RM, Johnson SC, Kastman EK, Kosmatka KJ, Beasley TM, et al. Caloric restriction delays disease onset and mortality in rhesus monkeys. Science 2009;325:201-204. [4] Libert S, Guarente L. Metabolic and neuropsychiatric effects of calorie restriction and sirtuins. Annu Rev Physiol 2013;75:669-684. [5] Lee C, Longo VD. Fasting vs dietary restriction in cellular protection and cancer treatment: from model organisms to patients. Oncogene 2011;30:3305-3316. [6] Canto C, Auwerx J. Caloric restriction, SIRT1 and longevity. Trends Endocrinol Metab 2009;20:325-331. [7] Zhai Y, Busuttil RW, Kupiec-Weglinski JW. Liver ischemia and reperfusion injury: new insights into mechanisms of innate-adaptive immune-mediated tissue inflammation. Am J Transplant 2011;11:1563-1569. [8] Mosher B, Dean R, Harkema J, Remick D, Palma J, Crockett E. Inhibition of Kupffer cells reduced CXC chemokine production and liver injury. J Surg Res 2001;99:201-210. [9] Ulloa L, Messmer D. High-mobility group box 1 (HMGB1) protein: friend and foe. Cytokine Growth Factor Rev 2006;17:189-201. [10] Tsung A, Sahai R, Tanaka H, Nakao A, Fink MP, Lotze MT, et al. The nuclear factor HMGB1 mediates hepatic injury after murine liver ischemia-reperfusion. J Exp Med 2005;201:1135-1143. [11] Selzner N, Selzner M, Jochum W, Clavien PA. Ischemic preconditioning protects the steatotic mouse liver against reperfusion injury: an ATP dependent mechanism. J Hepatol 2003;39:55-61. [12] Selzner M, Selzner N, Jochum W, Graf R, Clavien PA. Increased ischemic injury in old mouse liver: an ATP-dependent mechanism. Liver Transpl 2007;13:382-390. [13] Wang JH, Ahn IS, Fischer TD, Byeon JI, Dunn WA, Jr., Behrns KE, et al. Autophagy suppresses age-dependent ischemia and reperfusion injury in livers of mice. Gastroenterology 2011;141:2188-2199. [14] Yu ZF, Mattson MP. Dietary restriction and 2-deoxyglucose administration reduce focal ischemic brain damage and improve behavioral outcome: evidence for a preconditioning mechanism. J Neurosci Res 1999;57:830-839. [15] Chandrasekar B, Nelson JF, Colston JT, Freeman GL. Calorie restriction attenuates inflammatory responses to myocardial ischemia-reperfusion injury. Am J Physiol Heart Circ Physiol 2001;280:H2094-2102. [16] Kawai SI, Vora S, Das S, Gachie E, Becker B, Neufeld AH. Modeling of risk factors for the degeneration of retinal ganglion cells after ischemia/reperfusion in rats: effects of age, caloric restriction, diabetes, pigmentation, and glaucoma. Faseb J 2001;15:1285-1287. [17] Ahmet I, Wan R, Mattson MP, Lakatta EG, Talan M. Cardioprotection by intermittent fasting in rats. Circulation 2005;112:3115-3121. [18] Mitchell JR, Verweij M, Brand K, van de Ven M, Goemaere N, van den Engel S, et al. Short-term dietary restriction and fasting precondition against ischemia reperfusion injury in mice. Aging Cell 2010;9:40-53. [19] Jang JH, Moritz W, Graf R, Clavien PA. Preconditioning with death ligands FasL and TNF-alpha protects the cirrhotic mouse liver against ischaemic injury. Gut 2008;57:492-499.
18 [20] Sokolovic M, Sokolovic A, Wehkamp D, Ver Loren van Themaat E, de Waart DR, Gilhuijs-Pederson LA, et al. The transcriptomic signature of fasting murine liver. BMC Genomics 2008;9:528. [21] Tschop M, Smiley DL, Heiman ML. Ghrelin induces adiposity in rodents. Nature 2000;407:908-913. [22] Takeda R, Nishimatsu H, Suzuki E, Satonaka H, Nagata D, Oba S, et al. Ghrelin improves renal function in mice with ischemic acute renal failure. J Am Soc Nephrol 2006;17:113-121. [23] Abu-Amara M, Yang SY, Tapuria N, Fuller B, Davidson B, Seifalian A. Liver ischemia/reperfusion injury: processes in inflammatory networks--a review. Liver Transpl 2010;16:1016-1032. [24] Suzuki S, Toledo-Pereyra LH, Rodriguez F, Lopez F. Role of Kupffer cells in neutrophil activation and infiltration following total hepatic ischemia and reperfusion. Circ Shock 1994;42:204-209. [25] Yang H, Tracey KJ. Targeting HMGB1 in inflammation. Biochim Biophys Acta 2010;1799:149-156. [26] Castiglioni A, Canti V, Rovere-Querini P, Manfredi AA. High-mobility group box 1 (HMGB1) as a master regulator of innate immunity. Cell Tissue Res 2011;343:189-199. [27] Tang D, Kang R, Livesey KM, Cheh CW, Farkas A, Loughran P, et al. Endogenous HMGB1 regulates autophagy. J Cell Biol 2010;190:881-892. [28] Zhu S, Li W, Li J, Jundoria A, Sama AE, Wang H. It Is Not Just Folklore: The Aqueous Extract of Mung Bean Coat Is Protective against Sepsis. Evid Based Complement Alternat Med 2012;2012:498467. [29] Lee IH, Cao L, Mostoslavsky R, Lombard DB, Liu J, Bruns NE, et al. A role for the NAD-dependent deacetylase Sirt1 in the regulation of autophagy. Proc Natl Acad Sci U S A 2008;105:3374-3379. [30] Peck B, Chen CY, Ho KK, Di Fruscia P, Myatt SS, Coombes RC, et al. SIRT inhibitors induce cell death and p53 acetylation through targeting both SIRT1 and SIRT2. Mol Cancer Ther 2010;9:844-855. [31] Sankary HN, Chong A, Foster P, Brown E, Shen J, Kimura R, et al. Inactivation of Kupffer cells after prolonged donor fasting improves viability of transplanted hepatic allografts. Hepatology 1995;22:1236-1242. [32] Ritz BW, Aktan I, Nogusa S, Gardner EM. Energy restriction impairs natural killer cell function and increases the severity of influenza infection in young adult male C57BL/6 mice. J Nutr 2008;138:2269-2275. [33] Shen J, Ren H, Tomiyama-Miyaji C, Watanabe M, Kainuma E, Inoue M, et al. Resistance and augmentation of innate immunity in mice exposed to starvation. Cell Immunol 2009;259:66-73. [34] van Golen RF, van Gulik TM, Heger M. The sterile immune response during hepatic ischemia/reperfusion. Cytokine Growth Factor Rev 2012;23:69-84. [35] Yu M, Wang H, Ding A, Golenbock DT, Latz E, Czura CJ, et al. HMGB1 signals through toll-like receptor (TLR) 4 and TLR2. Shock 2006;26:174-179. [36] Frink M, Hsieh YC, Thobe BM, Choudhry MA, Schwacha MG, Bland KI, et al. TLR4 regulates Kupffer cell chemokine production, systemic inflammation and lung neutrophil infiltration following trauma-hemorrhage. Mol Immunol 2007;44:2625-2630. [37] Cantino D, Mosso R, Baccino FM. Changes induced by fasting and cycloheximide in the vacuolar apparatus of rat hepatocytes. A morphometric investigation. Boll Soc Ital Biol Sper 1979;55:1884-1889. [38] Mizushima N, Yoshimori T. How to interpret LC3 immunoblotting. Autophagy 2007;3:542-545.
19 [39] Powell SR, Divald A. The ubiquitin-proteasome system in myocardial ischaemia and preconditioning. Cardiovasc Res 2010;85:303-311. [40] Huebener P, Gwak GY, Pradere JP, Quinzii CM, Friedman R, Lin CS, et al. Highmobility group box 1 is dispensable for autophagy, mitochondrial quality control, and organ function in vivo. Cell Metab 2014;19:539-547. [41] Nace GW, Huang H, Klune JR, Eid RE, Rosborough BR, Korff S, et al. Cellularspecific role of toll-like receptor 4 in hepatic ischemia-reperfusion injury in mice. Hepatology 2013;58:374-387. [42] Rautou PE, Cazals-Hatem D, Moreau R, Francoz C, Feldmann G, Lebrec D, et al. Acute liver cell damage in patients with anorexia nervosa: a possible role of starvationinduced hepatocyte autophagy. Gastroenterology 2008;135:840-848. [43] Zhao X, Allison D, Condon B, Zhang F, Gheyi T, Zhang A, et al. The 2.5 A crystal structure of the SIRT1 catalytic domain bound to nicotinamide adenine dinucleotide (NAD+) and an indole (EX527 analogue) reveals a novel mechanism of histone deacetylase inhibition. J Med Chem 2013;56:963-969. [44] Timmers S, Konings E, Bilet L, Houtkooper RH, van de Weijer T, Goossens GH, et al. Calorie restriction-like effects of 30 days of resveratrol supplementation on energy metabolism and metabolic profile in obese humans. Cell Metab 2011;14:612-622. [45] Bonaldi T, Talamo F, Scaffidi P, Ferrera D, Porto A, Bachi A, et al. Monocytic cells hyperacetylate chromatin protein HMGB1 to redirect it towards secretion. Embo J 2003;22:5551-5560. [46] Lee JT, Gu W. SIRT1: Regulator of p53 Deacetylation. Genes Cancer 2013;4:112117. [47] Livesey KM, Kang R, Vernon P, Buchser W, Loughran P, Watkins SC, et al. p53/HMGB1 complexes regulate autophagy and apoptosis. Cancer Res 2012;72:1996-2005. [48] Nadtochiy SM, Redman E, Rahman I, Brookes PS. Lysine deacetylation in ischaemic preconditioning: the role of SIRT1. Cardiovasc Res 2011;89:643-649. [49] Hsu CP, Zhai P, Yamamoto T, Maejima Y, Matsushima S, Hariharan N, et al. Silent information regulator 1 protects the heart from ischemia/reperfusion. Circulation 2010;122:2170-2182. [50] Fan H, Yang HC, You L, Wang YY, He WJ, Hao CM. The histone deacetylase, SIRT1, contributes to the resistance of young mice to ischemia/reperfusion-induced acute kidney injury. Kidney Int 2013;83:404-413. [51] Nadtochiy SM, Yao H, McBurney MW, Gu W, Guarente L, Rahman I, et al. SIRT1mediated acute cardioprotection. Am J Physiol Heart Circ Physiol 2011;301:H1506-1512. [52] Cardinal J, Pan P, Dhupar R, Ross M, Nakao A, Lotze M, et al. Cisplatin prevents high mobility group box 1 release and is protective in a murine model of hepatic ischemia/reperfusion injury. Hepatology 2009;50:565-574.
20 Figure Legends
Figure 1. One-day fasting protects mouse liver from ischemia reperfusion injury and correlates with a repressed inflammatory response. (A) Levels of liver injury markers (AST/ALT) and parenchymal cell death (histological necrosis, TUNEL-positive cells, see representative examples at bottom) following fasting are shown. (B) Inflammatory gene expression (Tnfa, Il6, Il1b, Cd14, Ccl2) and histological quantification of Mpo and F4/80 assessed at 6h post reperfusion. Expression data was normalized to unfasted controls. *P<0.05 with n=8/group.
Figure 2. Serum starvation dampens macrophage activity and lowers circulating Hmgb1 levels. (A) Secreted Tnfa and intracellular ROS levels of Raw264-7 cultures following overnight exposure to descending concentrations of serum (upper panels) or at 10% (control) and 0.1% (fasting) serum challenged with ascending LPS doses (lower panels). (B) Serum levels of Hmgb1 in fasted mice and AST levels at 6h post reperfusion in unfasted controls and 1d-fasted mice injected with NaCl or Hmgb1 prior to ischemia. (C) Immunofluorescence (top) and immunohistochemistry (bottom) showing hepatic Hmgb1 expression in unfasted controls and 1d-fasted mice. Kupffer cells are marked with red/black arrowheads. Note the strong Hmgb1 expression in KCs following fasting. (D) Hepatic LC3 immunoblots and quantification of hepatic LC3 II fragments in fasted mice (loading control β-tubulin). *P<0.05 with n=6/group.
Figure 3. IRI protection through fasting is dependent on Hmgb1 regulation through Sirt1. (A) Sirt1 mRNA/protein expression, NAD+-dependent histone deacetylase activity, and AST levels (6h post reperfusion) following Sirt1 inhibition are shown for control and fasted mice. (B) Sirt1 immunochemical localization (Sirt1 green, nuclei blue) in fasted liver. (C)
21 Raw264-7 cells were cultured with 10% serum, 0.1% serum, or 0.1% serum plus Ex527 overnight and stained for Hmgb1. Phalloidin staining (F-aktin, blue) was used to visualize cell borders. Graphs to the left show supernatant Hmgb1 concentration in response to resveratrol (10% FCS) or Ex527 (0.1% FCS). Right bottom panel: Immunofluorescence demonstrating colocalization of Hmgb1 (green) with Beclin1 (red) in starved cells (blue nuclei). (D) Serum AST levels at 6h post reperfusion in mice fasted for one day prior to ischemia. Right panel: treatment with neutralizing Hmgb1 antibody reduced AST only in mice pretreated with Ex527. Left panel: inhibition of autophagy with SP600125 abolished the protection afforded by fasting. *P<0.05 with n=6/group.
Figure 4. Model for the protective effects fasting has on ischemia reperfusion injury. DAMPs and ROS generated during IR activate KCs, which in response produce ROS and actively secrete Hmgb1 to potentiate their activation. Neutrophils, attracted by KCs, damage liver tissue, further increasing circulating Hmgb1 and propagating KC activation. Fasting activates Sirt1, thereby engaging Hmgb1 into the autophagic response and lowering its extracellular release. The reduced levels of circulating Hmgb1 dampen the selfpropagation of KCs and hence protect from IRI. Prolonged fasting, however, may lead to increased membrane leakiness due to autophagic cell death, again increasing the levels of circulating Hmgb1 and neutralizing the protective effects of the early autophagic response.
22
23
24
25