Acute liver injury following acetaminophen administration does not activate atrophic pathways in the mouse diaphragm

N-acetyl-para-amino phenol (APAP, usually named paracetamol), which is commonly used for its analgesic and antipyretic properties may lead to hepatotoxicity and acute liver damage in case of overdoses. Released cytokines and oxidative stress following acute liver damage may affect other organs’ function notably the diaphragm, which is particularly sensitive to oxidative stress and circulating cytokines. We addressed this issue in a mouse model of acute liver injury induced by administration of APAP. C57BL/6J mice (each n = 8) were treated with N-acetyl-para-amino phenol (APAP) to induce acute drug caused liver injury and sacrificed 12 or 24 h afterwards. An untreated group served as controls. Key markers of inflammation, proteolysis, autophagy and oxidative stress were measured in diaphragm samples. In APAP treated animals, liver damage was proven by the enhanced serum levels of alanine aminotransferase and aspartate aminotransferase. In the diaphragm, besides a significant increase in IL 6 and lipid peroxidation, no changes were observed in key markers of the proteolytic, and autophagy signaling pathways, other inflammatory markers and fiber dimensions. The first 24 h of acute liver damage did not impair diaphragm atrophic pathways although it slightly enhanced IL-6 and lipid peroxidation. Whether longer exposure might affect the diaphragm needs to be addressed in future experiments.

www.nature.com/scientificreports/ We therefore hypothesized that acute liver injury may affect the diaphragm and induce atrophic signaling via protease-based proteolysis and autophagy. This issue was addressed in a mouse model of acute liver injury induced by administration of APAP.

Methods
The study was approved by the appropriate governmental institution (Landesamt für Natur-, Umwelt-und Verbraucherschutz, LANUV, Germany, reference number: AZ 84-02.04.2014.A165) which granted the ethical approval for this study. The study was conducted in accordance with the principles for care and use of animals based on the Helsinki Declaration and the ARRIVE criteria.
Induction of liver injury and animal model. C57BL/6J mice were housed under specific pathogen free conditions (Janvier labs, Le Genedt-Saint-Isle, France) and were divided into 3 groups (each n = 8): (1) acute liver injury sacrified at 12 h (APAP12), (2) acute liver injury sacrified at 24 h (APAP24), (3) an untreated control group (CTRL). Acute liver injury was induced as described in detail before 10,11 . Briefly, after a fasting period of 12 h, the animals received 250 mg/kg of APAP (Actavis Deutschland GmBh& CO. Kg, Langenfeld, Germany) by a single i.v. injection into a tail vein. After sacrifice, diaphragm was removed quickly and further processed (see below). Alanine aminotransferase (ALT) and aspartate aminotransferase activity (UV test at 37 °C) was measured in serum (Roche modular preanalytics system; Roche Diagnostics International AG, Rotkreuz, Switzerland).
Diaphragm histology. Diaphragm strips were longitudinally embedded in formalin and cut in 10 µm thin slides. Staining was performed using hematoxylin and eosin as routine laboratory staining. Around 50 fibers of 3 animals of each group were used for assessment of cross-sectional areas using Image J software (Image J, NIH, USA).
Protein extraction for electrophoresis and western blotting. Diaphragm tissue samples were homogenized 1:10 (weight volume −1 ) in 5 mmol Tris·HCl and 5 mmol EDTA buffer (pH at 7.5) containing a protease inhibitor cocktail (Roche, Basel, Swiss) and centrifuged at 1500×g for 10 min at 4 °C. After collection of the resulting supernatant, diaphragmatic protein content was assessed by the method of Bradford 12 , (Bio-Rad Laboratories Inc., Hercules, CA, USA). Proteins were detected with Chemiluminescent Peroxidase Substrate (Sigma-Aldrich, Overijse, Belgium), imaged with the Proxima 2850T imaging system (Isogen Life Technologies, De Meern, Netherlands) and analyzed using the TotalLab 1D software (Isogen Life Technologies, De Meern, Netherlands).
Autophagy and Pi3K-Akt pathway. Autophagosome formation was assessed by measuring the conversion of LC3B-I to LC3B-II via immunoblotting. Proteins were separated on a 12% polyacrylamide gel and transferred onto a polyvinylidene fluoride (PVDF) membrane. Blots were incubated overnight at 4 °C with the primary antibody (# LC3A/B 4108S, Cell Signaling/Bioké, Leiden, The Netherlands) and subsequently with the appropriate secondary antibody (# P0217, Agilent, Heverlee, Belgium) for 1 h at room temperature. Data were expressed as LC3B-II/LC3B-I ratio.
Upstream signaling via Phosphatidyl-inositol-Kinase-3 (Pi3K) and Protein kinase B (AKT) was also measured by detecting the phosphorylated and non-phosphorylated form of Pi3K (Pi3K; #4249, Cell Signaling, Danvers USA) and of Protein kinase B (AKT; #4691, Cell Signaling, Danvers USA). Data were expressed as ratio between phosphorylated to total (non-phosphorylated) protein.

Calpain-1 and caspase-3 activities.
In vivo calpain-1 and caspase-3 activities were indirectly assessed, by measuring cleavage of αII-spectrin, a specific substrate for both calpain-1 and caspase-3. After separation of the proteins on a 10% polyacrylamide gel and transfer to PVDF membrane, the blot was incubated with the primary antibody against αII-spectrin (α-Fodrin FG6090, Enzo Life Science, Brussels, Belgium) and the suitable secondary antibody (P0260, Agilent, Heverlee, Belgium). Activities of calpain and caspase-3 were measured as the ratio between the densitometric values of their breakdown products to intact αII-spectrin 13 . The cleavage product of intact αII-spectrin by calpain gives a band at 150 kDa and a band at 120 kDa when cleaved by caspase-3 while intact αII-spectrin is detected at 260 kDa 14  Oxidative stress: assessment of lipid peroxidation. Diaphragmatic 4-hydroxynonenal (4-HNE) was used as a marker of lipid peroxidation. After electrophoresis (7.5% polyacrylamide gel and blotting) polyclonal anti-4-HNE antibody (# MAB3249, R&D Systems, Abingdon, UK) was used as primary antibody and a polyclonal rabbit anti-mouse (# P0260, Agilent, Heverlee, Belgium) as secondary antibody. Data were expressed as a ratio between densitometric values of 4-HNE and Vinculin 15 . . For messenger RNA (mRNA) analysis, inflammatory and proteolytic marker expression was normalized with ribosomal protein S7. RNA expression was analyzed by the ΔΔ C T method. The following primers were used (see Table 1).

Statistical analysis. Population distribution was assessed with the Kolmogorov-Smirnov test. If this test
showed normal distribution of data, comparisons between groups for each dependent variable were made by ANOVA, if normal distribution was not present, by a Kruskall Wallis test. If the group effect was significant, a Dunn's multiple comparisons test was used for pairwise comparisons between all groups. Data are shown as means ± standard deviation (SD). All statistical tests are two-tailed, significance was established at p < 0.05 (GraphPad Prism 6.0, La Jolla, CA, USA).

Liver injury characteristics. Serum levels of ALT increased significantly and similarly 12 and 24 h after
APAP administration compared to control (ctrl vs APAP 12 p = 0.01; ctrl vs APAP p < 0.001) while serum levels of AST increased significantly but only 24 h after APAP administration (ctrl vs APAP 24 p < 0.001) (see Fig. 1).
Histologic assessment of diaphragm atrophy. No  Diaphragm inflammatory response. Levels of the NfκB subunits p50 and p65 were not altered in the different groups (see Fig. S6 in Supplement, as well as the corresponding blots in Supplemental File 1, Fig. S8 and Supplemental File 2, Fig. S4). The mRNA expression of IL-10, TNF-α, IL1-β did not change significantly while mRNA levels of IL-6 were significantly elevated 24 h after APAP exposure vs ctrl (p = 0.02) (see Fig. 4).  TGT AGC CAA TGT C   TNF-a rev  GAG TAG ATG AGG TAC AGC CC   IL 6 for  GGT TCA ATC AGG AGA CCT GC   IL 6 rev  GCT TTG TCT GGA TTC TTT CCC   IL-10 for  GCA TCC ACT TCC CAA CCA G   IL-10

Discussion
This study showed that in mice, during the acute phase of liver injury, no activation of the proteolytic or autophagic pathways and no atrophy occurred in the diaphragm while lipid peroxidation was elevated and IL-6 was the only inflammatory marker to be upregulated inside the diaphragm. The effects on the diaphragm are discussed in detail below. The APAP model to induce acute liver failure is a well-established model. In previous studies, we have proven that induction of necrosis in hepatocytes after administration of APAP leads to consistent results within groups. Analysis of immune cells and their modulation showed that the model induces "two hits" during acute liver injury generated by APAP 10 . Hepatocyte necroses are induced directly by the toxic metabolite NAPQI. Otherwise, severity of the liver injury and dimension of necrosis are affected by a massive immune cell infiltration into the liver caused by the initial damage 3 . Until know it is still ambiguous which part of the immune response and which immune cells respectively aggravate or attenuate liver injury at the different stages of the disease 16 .
We detected significant changes of liver enzymes' serum levels and necrotic areas inside liver tissue, pointing to a profound damage of hepatocytes. Central clinical signs of the acute liver failure are prolonged prothrombin time/INR, declined mental function, vasodilation and a systemic inflammatory response syndrome due to the high immunological impact of the liver 4 . Summarily, acute liver failure results in multi-organ failure (Acute liver failure) 17 , although depending on etiology and duration course of illness is variable. In APAP and ischemia induced ALF the acute phase will be measured in hours and we expect systemic effects during the experimental time chosen in our model. Inflammatory response to acute liver injury has been described in humans and several cytokines in drug induced liver injury were elevated in serum 18   www.nature.com/scientificreports/ demonstrated severe muscle atrophy in the diaphragm after IL-6 administration into the circulation. However, in diaphragm the influence of inflammation on its function has been under discussion: Blocking of toll-like receptor-4 in mice in a disuse model could decrease the amount of autophagic activation and IL-6 production and ameliorate the reduction of myosin heavy chain protein 22 . Possibly, the animals had an increase in the work of breathing due to metabolic disturbances after liver injury, which might explain this increase in IL-6 in a quasiphysiologic response to overload. www.nature.com/scientificreports/ Summarily, we could not detect a profound secondary response inside the diaphragm in the following 12 h or 24 h after liver injury induction except for enhanced levels of IL-6 that were not associated with changes in NFκB subunits or other inflammatory cytokines.
Oxidative stress is a key factor in the pathogenesis of ICU-acquired diaphragmatic weakness 5 with mitochondria and to a smaller extent oxidases functioning as major sources of ROS [23][24][25] . In diaphragm, the increase of ROS interacts with the activity of main proteolytic pathways caspase 3 and calpain-1 as well as the autophagic system, so that blocking ROS results in an amelioration of atrophy via autophagic and proteolytic pathways 23,24,26,27 . In our model, we could detect a significant transient increase in the formation of lipid peroxidation (4 HNE) after 12 h of APAP induced liver injury. This increase abandoned in the 24 h APAP group, which could indicate a quick removal of oxygenated structures, as we have detected in a former experiment 28 . In our actual study, we did not distinguish further which source (mitochondria and/or oxidases) was responsible for the increase in oxidation. Importantly, we did not found a link between this increase in lipid peroxiadation and atrophy inducing pathways, although in disuse or sepsis this has been proven 26,27 . Possibly the amount of oxidative stress did not reach the level to induce proteolytic Ca-dependent proteases or autophagy.
In several diseases, proteolysis is activated inside the diaphragm and modulated by different pathways: Calcium dependent protease-based proteolysis via caspase 3 and calpain 1 have been described during sepsis, disuse (VIDD) and in chronic lung disease [29][30][31] . Caspase 3 acts a key protease, in conjunction with calpain 1, to break up the sarcomeric structures and allow downstream proteasome to further degrade the proteins 32,33 . Neither mRNA levels nor protein expression of key calcium-dependent proteases were enhanced in the diaphragm despite the presence of clear liver injury. Even more, the downstream ligases MURF and MAFbx remained unaltered in the interventional groups, although these ligases are involved in the further degradation of muscle protein after break-up by caspases 31 and are usually activated during muscle atrophy, inflammation or oxidative stress 30 . In summary, proteolytic activation in the diaphragm by acute liver injury does not seem to be present in our model.
Autophagy is an important proteolytic pathway targeting cellular components associated with muscular atrophy via autophagosomes, which form double layer structures of which LC3 is a core molecule 27,34 . Upstream, Protein kinase B (AKT) and PI3K regulate via the Forkhead-Box-O1 transcription factor the formation of autophagosome building proteins, like LC3B.
The diaphragm is susceptible to slightest disturbances in the ratio of protein synthesis and proteolysis, so that even 12 h of disuse affect this balance 35 . Transformation from LCB3BI into LC3B-II indicates the activation of autophagosomes 30 and Hussain and colleagues described this pathway as one major diaphragmatic pathway in disuse 36 which is also active in sepsis and linked to oxidative stress 30 . In addition, the autophagy pathway can modulate protein synthesis through the mammalian target of rapamycin (mTOR) and serves therefore as a switch between anabolic and catabolic states 37 .
During liver injury as in our model, autophagy assessed by the ratio of LC3BII/LC3BI and by the upstream molecules was not activated in the diaphragm.
This study has, however, some limitations that should be taken into account. According to the goal of the study, 12 and 24 h were specifically chosen as time point measures after liver injury in order to address the early phase effects of liver injury on the diaphragm. We, however, acknowledge that later time point measures may be relevant to address the secondary complications reported to arise days after acute liver injury in humans. Further studies are needed to address the influence of acetaminophen induced liver injury on diaphragm dysfunction at later time points and whether known effects on lung function (e.g. airway hyper-reactivity, asthma, etc.) may add to these changes 38 . Those were in this study beyond the scope.

Conclusion
In contrast to our hypothesis, acute liver injury induced by APAP in mouse did not induce atrophy or activate proteolysis via calcium dependent proteases or via ubiquitin proteasome pathway, neither it did stimulate autophagy. However, it resulted in enhanced diaphragm lipid peroxidation and upregulation of IL-6 which, on long-term, might be detrimental for the diaphragm. Future research might focus on long term models of liver injury to see whether the diaphragm would be affected.