- Email: [email protected]
Hepatic upregulation of fetuin-A mediates acetaminophen-induced liver injury through activation of TLR4 in mice
Biochemical Pharmacology 166 (2019) 46–55
Contents lists available at ScienceDirect
Biochemical Pharmacology journal homepage: www.elsevier.com/locate/biochempharm
Hepatic upregulation of fetuin-A mediates acetaminophen-induced liver injury through activation of TLR4 in mice
T
Kang-Yo Lee, Wonseok Lee, Seung-Hwan Jung, Jungmin Park, Hyungtai Sim, You-Jin Choi, ⁎ Young-Jun Park, Yeonseok Chung, Byung-Hoon Lee College of Pharmacy and Research Institute of Pharmaceutical Sciences, Seoul National University, Gwanakro 1, Gwanak-gu, Seoul 08826, Republic of Korea
A R T I C LE I N FO
A B S T R A C T
Keywords: Acetaminophen Fetuin-A Hepatokine Chemokine Hepatotoxicity
Acetaminophen (APAP)-induced liver injury (AILI) is initiated by the generation of a reactive metabolite and ultimately leads to hepatocyte necrosis. Necrotic cells secrete damage-associated molecular patterns that activate hepatic nonparenchymal cells and induce an inflammatory response. Fetuin-A is a hepatokine with reported involvement in low-grade inflammation in many diseases, due to acting as an endogenous ligand for TLR4. However, little is known about the role of fetuin-A in AILI. In this study, we showed that fetuin-A is involved in the aggravation of hepatotoxicity during the initial phase of AILI progression. Treatment with APAP increased the expression and serum levels of fetuin-A in mice. Fetuin-A upregulated transcription of pro-inflammatory cytokines and chemokines through activation of TLR4 and also increased monocyte infiltration into the liver, leading to necroinflammatory reactions in AILI. However, these reactions were attenuated with the silencing of fetuin-A using adenoviral shRNA. As a result, mice with silenced fetuin-A exhibited less centrilobular necrosis and liver injury compared to controls in response to APAP. In conclusion, our results suggest that fetuin-A is an important hepatokine that mediates the hepatotoxicity of APAP through production of chemokines and thus regulates the infiltration of monocytes into the liver, a critical event in the inflammatory response during the initial phase of AILI. Our results indicate that a strategy based on the antagonism of fetuin-A may be a novel therapeutic approach to the treatment of acetaminophen-induced acute liver failure.
1. Introduction Acetaminophen (N-acetyl-p-aminophenol; APAP), which is one of the most commonly used analgesic and antipyretic drugs worldwide, is available over-the-counter and by prescription either as a single agent or in combination with other medications. Although it is safe when administered at the recommended dosage, APAP overdose can cause severe hepatotoxicity in experimental animals and humans. About half of all cases of acute liver failure in the US and Western countries are attributable to APAP overdose [1,2]. APAP-induced liver injury (AILI) is divided into three stages: initiation, progression, and regeneration. CYP2E1-mediated formation of N-acetyl-p-benzoquinone imine (NAPQI), a reactive metabolite of APAP, is the initiating event of APAP hepatotoxicity [3]. NAPQI depletes the cellular glutathione pool and forms a protein adduct, especially in the mitochondria. Formation of protein adducts in the electron transport chain following APAP overdose leads to inhibition of mitochondrial respiration and formation of mitochondrial oxidative stress, resulting in mitochondrial dysfunction
and oncotic necrosis of hepatocytes [4,5]. As a result of the necrotic death of hepatocytes, inflammatory mediators, such as damage-associated molecular patterns (DAMPs), activate hepatic nonparenchymal cells, resulting in sterile inflammation and progression to liver injury. A variety of cytokines released by immune cells are involved in the aggravation of hepatotoxicity via immune-mediated mechanisms [6,7]. Although the number of resident Kupffer cells decreases rapidly due to APAP cytotoxicity [8], huge numbers of monocytes are recruited into the liver, where they differentiate into monocyte-derived macrophages (MoMF) in a CCR2-dependent manner [8–10]. During the onset of acute AILI, the infiltrating MoMFs aggravate inflammation and hepatotoxicity without impairing the repair process, as confirmed by the attenuation of liver damage with the blocking of either CCL2 or CCR2/CCR5 [10]. MoMFs are indispensable for the repair and resolution of tissue injury during the recovery phase. Infiltrated macrophages express angiogenic factors, including VEGF and ANGPTL, which assist with recovery of the sinusoid and microcirculation in the liver [11]. MoMFs influence the restoration
⁎ Corresponding author at: 29-321, College of Pharmacy and Research Institute of Pharmaceutical Sciences, Seoul National University, Gwanakro 1, Gwanak-gu, Seoul 08826, Republic of Korea. E-mail address: [email protected] (B.-H. Lee).
https://doi.org/10.1016/j.bcp.2019.05.011 Received 21 March 2019; Accepted 6 May 2019 Available online 08 May 2019 0006-2952/ © 2019 Elsevier Inc. All rights reserved.
Biochemical Pharmacology 166 (2019) 46–55
K.-Y. Lee, et al.
(iNtRON, Seoul, Korea) and complementary DNA was prepared with QuantiTect Reverse Transcription Kit (QIAGEN, Hilden, Germany). cDNA was amplified by qRT-PCR using iTaq Universal SYBR Green Supermix kit (Bio-Rad, Hercules, CA).
process through the expression of Ym1, Fizz1, Arg-1, MMP-12, and MMP-9, representative markers of M2-type polarization. Therefore, deletion of CCR2 delays the repair process in AILI [8]. Fetuin-A, also known as α2-Heremans-Schmid glycoprotein (AHSG), is a secretory protein produced primarily by the liver and also by other tissues, including the tongue, placenta, and adipose tissue [12,13]. Fetuin-A was initially identified as an inducer of insulin resistance and was also found to be strongly associated with metabolic syndrome in humans [14–16]. These results suggested a possible association of fetuin-A with low-grade inflammation, which was confirmed recently with experimental and clinical data. Fetuin-A was shown to induce inflammatory cytokine expression and suppress adiponectin production in adipocytes, monocytes, and in mice [17]. Recent research indicates that fetuin-A facilitates the development of atherosclerosis through its stimulatory effects on the inflammatory response in endothelial cells and foam cell formation in macrophages [18]. Fetuin-A acts as an endogenous ligand of toll-like receptor 4 (TLR4), promoting the lipid-induced proinflammatory response and insulin resistance [19]. Moreover, lipid-induced expression of fetuin-A from adipocytes acts as an efficient chemokine to induce macrophage migration and polarization [12]. Resident macrophages and infiltrating myeloid cells secrete a variety of chemokines and cytokines, leading to sterile inflammation in AILI. As the interaction of fetuin-A with TLR4 provokes an inflammatory response, we hypothesized that fetuin-A acts as an unrecognized endogenous DAMP-like molecule and, thus, is an important mediator of AILI. In this study, we first showed that hepatokine fetuin-A is involved in the hepatotoxicity caused by APAP overdose. We demonstrated that APAP increased the expression and secretion of fetuinA in mice. Fetuin-A upregulated the expression of chemokines in hepatocytes and bone marrow-derived macrophage (BMDM) cells, leading to recruitment of monocytes to the liver. Furthermore, fetuin-A-mediated production of cytokines and chemokines aggravated APAP-induced hepatotoxicity.
2.4. Immunoblot Tissue homogenate was lysed in buffer containing 50 mM HEPES (Sigma-Aldrich), 150 mM NaCl (Sigma-Aldrich), 5 mM EGTA (SigmaAldrich), 50 mM β-glycerophosphate (Sigma-Aldrich), 1% Triton-X 100 (Sigma-Aldrich) and protease inhibitor cocktail (Roche, Indianapolis, IN). The lysates were then centrifuged and the supernatant was separated by SDS-polyacrylamide gel electrophoresis and then transferred to PVDF membrane (Millipore, Darmstadt, Germany). Transferred proteins were blocked in 5% skim milk, probed with anti-fetuin-A (sc20872, Santa Cruz Biotechnology, Santa Cruz, CA) or anti-GAPDH antibody (#2118, Cell Signaling Technology, Beverly, MA) and developed by enhanced chemiluminescence western blotting reagent (Amersham, Piscataway, NJ). 2.5. CYP2E1 activity assay Liver tissue homogenate was prepared in cold 0.15 M KCl (USB, Cleveland, OH) and centrifuged at 9000g. Supernatant was transferred to a new tube and centrifuged at 100,000g for 1 h. Microsomal pellet was resuspended in 0.15 M KCl. 200 μg of microsomal protein was incubated in 0.1 M potassium phosphate (Sigma-Aldrich) buffer with 0.4 mM p-nitrophenol (Sigma-Aldrich) and 1 mM NADPH (Roche) for 30 min at 37 °C water bath. Reaction was terminated by addition of 50% trichloroacetic acid (Sigma-Aldrich) and reaction mixture was centrifuged at 10,000g. 20 μl of 10 N NaOH (Duksan, Ansan, Korea) was added to supernatant and generation of p-nitrocatechol was determined by measuring absorbance with spectrophotometer (VersaMax, Molecular Devices, Sunnyvale, CA) at 546 nm. CYP2E1 activity of samples was deduced from standard curve of p-nitrocatechol (SigmaAldrich).
2. Materials and methods 2.1. Animal experiments
2.6. Histological analysis and immunohistochemistry
7 to 8 weeks old (20–23 g) male SPF C57BL/6J mice from Central Lab. Animal Inc. (Seoul, Korea) were housed in air-conditioned room with 12 light/dark cycle and allowed free access to water and food. Mice were fasted overnight, injected with APAP (Sigma-Aldrich, St Louis, MO) or vehicle (PBS) intraperitoneally. For in vivo silencing experiment, mice were injected with 1010 adenoviral particle containing shRNA against Ahsg (Ad-shRNA-Ahsg) or mock (Ad-shRNAmock) (Welgen Inc, Worcester, MA) intravenously. After 3 days from adenoviral knockdown, mice received APAP or vehicle. Mice were anesthetized with Alfaxan (Jurox, Kansas City, MO) and sacrificed at indicated time point. All animal experiments were performed in accordance with the guidelines of the Institutional Animal Care and Use Committee of Seoul National University.
The liver tissues were fixed in neutral buffered formalin (SigmaAldrich) and embedded in paraffin. Paraffin blocks were cut into 5 μm sections and stained with hematoxylin and eosin (H&E). Necrotic area was quantified using ImageJ (NIH, Bethesda, ML). For immunostaining, tissue slides were incubated with primary antibody directed against fetuin-A (ab187051, Abcam, Cambridge, UK) and counterstained with hematoxylin. 2.7. Proximity ligation assay (PLA) PLA was performed with Duolink PLA kit (Sigma-Aldrich) according to manufacturer’s protocol. In brief, O.C.T (Sakura Finetek, Torrance, CA)-embedded tissue slides were fixed in ice cold methanol, blocked, and probed with anti-TLR4 (sc-293072, Santa Cruz Biotechnology) and anti-fetuin-A (sc-20872, Santa Cruz Biotechnology). PLA probe was adjusted to the samples followed by addition of ligase and polymerase. Slides were covered with mounting medium with DAPI and analyzed with TCS SP8 confocal laser scanning microscope (Leica Microsystems, Wetzlar, Germany). To obtain images throughout the entire thickness of the slides, images of 15 z-plane were acquired for all presentation.
2.2. Cell isolation and treatment Primary mouse hepatocytes were isolated from 7 to 8 weeks old male SPF C57BL/6J mice via collagenase type IV (Sigma-Aldrich) perfusion as described previously [20]. Cells were stabilized and treated as indicated. BMDM were isolated from hind leg of male C57BL/6J mice as described previously [21]. Cells were differentiated into macrophages with M-CSF (Peprotech, Rocky Hill, NJ) at concentration of 30 ng/ml and treated with or without mouse fetuin-A (Sino Biological Inc, Beijing, China).
2.8. Serum biochemistry
2.3. Quantitative real-time polymerase chain reaction
Serum levels of ALT, AST and LDH were analyzed by standard clinical chemistry assays on an Automated Chemistry Analyzer (Prestige 24I; Tokyo Boeki Medical System, Tokyo, Japan).
mRNA was extracted with the Easy-Blue Total RNA extraction kit 47
Biochemical Pharmacology 166 (2019) 46–55
K.-Y. Lee, et al.
2.9. Flow cytometry
3.3. APAP induced fetuin-A co-localization with TLR4 in the liver
Hepatic nonparenchymal cells were isolated from the whole livers of mice by the homogenization using the GentleMACS Dissociator (Miltenyi Biotec, Bergisch Gladbach, Germany). Following dissociation into a single cell suspension, cells were preincubated with FcR blocker (Biolegend, San Diego, CA, #101319), stained with fluorochromeconjugated antibodies and analyzed with LSR Foresta (BD Biosciences, San Jose, CA). Antibodies used for analysis are from Biolegend (San Diego, CA): APC/Cy7-CD45.2 (#109823), FITC-Ly6G (#127605), APCF4/80 (#123116), PerCP/Cy5.5-CD11b (101227), PE/Cy7-Ly6C (#128017).
To explore the role of fetuin-A in the inflammatory reaction during the initial phase of AILI progression, 8 mice per group were injected with adenoviral shRNA that was nonspecific or against the Ahsg fetuinA gene for 3 days prior to APAP administration. Mouse which was not silenced of fetuin-A was excluded for analysis. APAP treatment led to the upregulation of mRNA and the protein expression levels of fetuin-A, which was alleviated with the silencing of fetuin-A (Fig. 4A, B). In agreement with this finding, elevated serum fetuin-A levels caused by APAP treatment were reduced with the ablation of fetuin-A (Fig. 4C). Immunohistochemistry showed intensive staining of fetuin-A with APAP treatment. We found that the necrotic region around the central vein was significantly positive in the fetuin-A staining assay in APAPtreated mice but was attenuated by the silencing of fetuin-A (Fig. 4D). To exclude the possibility of the nonspecific binding of the secondary antibodies to the damaged region, a negative control experiment was performed without the primary antibody. As shown in Fig. 4D, nonspecific binding of the antibody did not occur. We also found that the expression and activity of CYP2E1 were not affected by the silencing of fetuin-A (Fig. 5A, B). To determine whether the silencing of fetuin-A had any effect on its binding to TLR4, we performed a proximity ligation assay (PLA) between fetuin-A and TLR4, which allows in situ detection of protein–protein interactions with high specificity and sensitivity. As a negative control, we performed the same assay without primary antibodies, which should generate no PLA signal. Compared to vehicle treatment, a strong PLA signal (red dots) was detected in the APAP group, especially around the necrotic area. In contrast, a significantly fewer PLA puncta were observed in the Ad-sh-Ahsg/APAP group. For the quantification of PLA signal, we selected necrotic area around the central vein and analyzed 5 areas (approximately 500,000 μm2) per group using ImageJ software (Fig. 5C). Co-immunoprecipitation did not detect binding due to the similar molecular weights of fetuin-A and the antibody heavy chain. These results suggest that APAP increases hepatic expression of fetuin-A and its binding to TLR4 in the liver.
2.10. Statistical analysis Data were presented as mean ± S.D. for in vivo experiments and mean ± S.E.M for in vitro experiments. Data were analyzed either by student’s t-test or one-way ANOVA analysis followed by Turkey’s multiple comparison test with GraphPad Prism 5 (GraphPad, San Diego, CA). P values < 0.05 were considered statistically significant. 3. Results 3.1. APAP increased hepatic expression and secretion of fetuin-A C57BL/6J mice were treated with APAP through intraperitoneal injection, and the induction of hepatotoxicity was confirmed based on elevated serum levels of ALT, AST, and LDH at 6, 24, and 48 h following treatment. APAP-induced hepatotoxicity was ameliorated at 48 h, which corresponds with the known time of repair and resolution for such tissue injury (Fig. 1A). Most subsequent animal experiments were performed 24 h after APAP treatment. We analyzed the effects of APAP on the hepatic expression of fetuin-A. APAP increased the expression of fetuin-A, which was maintained until 48 h. Of the two bands for fetuinA on immunoblots, the intensity of the protein with the higher molecular weight, reportedly the glycosylated and secretable form of fetuinA, increased more significantly (Fig. 1B). The attenuation of fetuin-A expression at 48 h mirrored the decrease in hepatotoxicity at the same time. Expression of Ahsg, the official gene name for fetuin-A, was also upregulated at 6 and 24 h of APAP treatment (Fig. 1C). As a result, serum levels of fetuin-A gradually increased after 24 h, peaking at 48 h (Fig. 1D). Taken together, these results show that the expression and the serum levels of fetuin-A increased with APAP treatment in mice.
3.4. Silencing fetuin-A suppresses chemokine expression and monocyte infiltration by APAP in mice. TLR4 signaling is associated with the production of chemotactic and inflammatory cytokines, as well as of inflammatory infiltrate consisting of neutrophils and macrophages. The silencing of fetuin-A slightly, albeit non-significantly, reduced the mRNA expression of all chemokines and cytokines. On the other hand, silencing of fetuin-A caused significantly downregulated expression of transcripts for the chemokines (Ccl2, Ccl3, and Ccl4) and cytokines (Tnfa and Il6) that were highly upregulated with APAP treatment (Fig. 6A). Similarly, the elevated serum level of CCL2 by APAP decreased with the silencing of fetuin-A (Fig. 6B). CCL2 is one of the main chemokines driving the infiltration of myeloid cells into APAP-inflamed tissue via the CCL2–CCR2 axis [8–10]. To investigate whether the silencing of fetuin-A attenuates the infiltration of myeloid cells, we isolated hepatic nonparenchymal cells from APAP-treated mice and analyzed them using flow cytometry with the gating strategy depicted in Fig. 7A [29]. We found that APAP injection induced massive infiltration of neutrophils into the liver based on CD45.2- and Ly6G-positive cell counts. The number of infiltrated MoMF (F4/80intermediate; CD11bhigh) also increased substantially with APAP treatment, whereas Kupffer cells (F4/80high; CD11blow) decreased, likely due to the direct cytotoxic effects of APAP. In contrast, the number of MoMF decreased significantly in mice with silenced fetuin-A genes, which might be related to the downregulation of CCL2 (Fig. 6B). Taken together, these results show that APAP upregulates the expression of fetuin-A, which, in turn, enhances the production of proinflammatory cytokines and chemokines. Elevated levels of chemokines, such as CCL2, induce monocyte infiltration. Silencing fetuin-A
3.2. Fetuin-A upregulated the transcription of chemokines Hepatic nonparenchymal cells secrete chemokines that contribute to the massive accumulation and activation of myeloid cells in APAP-induced hepatotoxicity [10,22]. To investigate whether APAP-induced fetuin-A expression has a causal relationship with AILI, we first analyzed the expression of chemokines and cytokines in the liver. In accordance with previous findings, administration of APAP to mice strongly upregulated the transcript levels of chemokines, including Ccl2, Ccl3, and Ccl4, as well as cytokines, such as Il6, and Il1a (Fig. 2A). APAP treatment also induced transcript level of Tnfa, which not statistically significant. Moreover, fetuin-A increased expression of Ccl2, Ccl3, and Ccl4 in mouse primary hepatocytes and of Ccl2, Ccl3, Tnfa, Il1a, and Il6 in BMDM (Fig. 2B, C). Knockdown of fetuin-A in mouse primary hepatocytes decreased the expression and secretion of fetuin-A into the culture medium (Fig. 3A). Silencing fetuin-A in hepatocytes led to significantly downregulated expression of Ccl2. However, the mRNA levels of Ccl3 and Ccl4 were not affected by silencing of fetuin-A probably due to the low basal expression in hepatocytes. (Fig. 3B). From these results, we can conclude that fetuin-A regulates the gene expression of chemokines and cytokines in mouse primary hepatocytes, as well as in macrophages. 48
Biochemical Pharmacology 166 (2019) 46–55
K.-Y. Lee, et al.
Fig. 1. APAP increases hepatic expression and secretion of fetuin-A in mice. Male C57/BL6 mice were injected with APAP (250 mg/kg; i.p.) and sacrificed after 6, 24, or 48 h. (A) Serum ALT, AST, and LDH levels were analyzed. (B) Western blot analysis was used to determine the abundance of fetuin-A in mouse liver homogenates. Band densities were determined using an image analysis system and expressed as percentages of the control. (C) mRNA levels were measured using qRT-PCR. Relative gene expression levels are represented as fold changes compared with the control. (D) Serum fetuin-A concentrations were analyzed using ELISA. n ≥ 4 for any time point. All data were expressed mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001 compared to the corresponding vehicle group.
4. Discussion
reduced the number of infiltrating MoMF, which increased substantially with APAP administration.
AILI is initiated with the generation of a reactive metabolite, NAPQI, and the subsequent depletion of the cellular glutathione pool, leading to the generation of reactive oxygen species, the formation of protein adducts, mitochondrial dysfunction, and, finally, cell death [23]. Hepatocyte necrosis activates Kupffer cells that secrete proinflammatory cytokines and chemokines, leading to sterile inflammation. This process is mediated by the release of intracellular constituents into the surrounding milieu, some of which belong to a group of molecules known as DAMPs. DAMPs are recognized by pattern recognition receptors, including the TLR family. Several types of DAMPs have been reported to mediate the sterile inflammation process, thereby causing a second wave of destruction and contributing to the pathogenesis of AILI [22,24]. The present study demonstrates the fundamental role of the hepatokine fetuin-A in APAP-induced hepatotoxicity. The key findings of the study include that APAP increases hepatic expression and secretion of fetuin-A, which mediates the upregulation of proinflammatory cytokines and chemokines through interaction with TLR4. Elevated levels of chemokines, such as CCL2, trigger the recruitment of monocytes, which play an important role in the aggravation of hepatotoxicity during the initial phase of AILI progression. Ablation of fetuin-A in the livers of mice significantly reduced the hepatic necrotic area following APAP administration, suggesting that fetuin-A plays a role in mediating APAP-induced hepatotoxicity. TLRs are a family of pattern recognition receptors that participate in the inflammatory process by recognizing DAMPs. Although various
3.5. Silencing Ahsg attenuated APAP-induced hepatotoxicity in mice Finally, we evaluated the role of fetuin-A in APAP-induced hepatotoxicity in mice with silenced fetuin-A expression. As shown in Fig. 8, the levels of ALT, AST, and LDH increased significantly within 24 h in the Ad-sh-mock/APAP group, but it was alleviated in the Ad-sh-Ahsg/ APAP group. After 48 h of APAP administration, which is the time needed for the repair and resolution of tissue injury, most parameters of hepatotoxicity had decreased significantly. In this case, the silencing of fetuin-A had no effect (Fig. 8A). Histological examination revealed that there was no significant difference between the Ad-sh-mock/vehicle and Ad-sh-Ahsg/vehicle groups. Severe centrilobular necrosis was observed in the Ad-sh-mock/APAP group. However, Ad-sh-Ahsg/APAP mice had significantly smaller hepatic necrotic areas at both 24 and 48 h following APAP administration (Fig. 8B). As a result, the enhanced serum content of HMGB1 with APAP treatment decreased with the silencing of fetuin-A at 24 h (Fig. 8C). In conclusion, our results suggest that fetuin-A is an important hepatokine that mediates the hepatotoxicity of APAP through the production of chemokines, causing monocytes to infiltrate the liver, a critical event in the inflammatory reaction during the initial phase of AILI progression.
49
Biochemical Pharmacology 166 (2019) 46–55
K.-Y. Lee, et al.
Fig. 2. APAP treatment and fetuin-A upregulates transcription of genes downstream of TLR4. (A) Male C57/BL6 mice were injected with APAP and sacrificed after 24 h. mRNA levels were measured using qRT-PCR. Relative gene expression levels are represented as fold changes compared with the control. All data were expressed mean ± SD (n = 4 ∼ 5). (B) Primary hepatocytes were isolated, stabilized, and incubated in the presence or absence of 30 μg/ml fetuin-A for 24 h without FBS. (C) BMDMs were isolated, differentiated, and incubated in the presence or absence of 50 μg/ml of fetuin-A for 24 h without FBS. mRNA levels were measured using qRTPCR. Relative gene expression levels are represented as fold changes compared with the control. All experiments were performed three times and all data were expressed mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001 compared to the control group.
AILI [28–30]. Indeed, treatment with antibodies against TNF-α or IL1α, which are regulated by TLR4 signal transduction, rescued mice from AILI [7]. Zhang et al. emphasized that AILI is dependent on TLR4/ Myd88 signaling and the resultant production of IL-1α [40]. In the present study, we showed that APAP induced the hepatic expression and secretion of fetuin-A, which binds to TLR4 and induces expression of chemokines and cytokines, and that such production was attenuated with the silencing of fetuin-A. Similar results were observed in carbon tetrachloride-induced hepatotoxicity (data not shown). Therefore, the hepatokine fetuin-A can be defined as a DAMP-like molecule that mediates the progression of toxicity through sterile inflammation in AILI. This finding will provide new insights for the field of clinical research related to treatment of hepatic failure caused by APAP overdose, as well as for basic research related to the toxic mechanism of APAP.
liver cells express diverse TLRs, it has been suggested that TLR4, TLR3, and TLR9 signaling is involved in AILI [25,26]. DNA and RNA fragments released during APAP-induced hepatocellular necrosis can be recognized by TLR9 and TLR3, respectively, thereby inducing cytokine production through activation of NF-κB. The pathophysiological relevance of TLR3 and TLR9 to AILI has been suggested by the protective effects observed in TLR3- or TLR9-deficient mice [25,26]. HMGB1 and histones released from dying hepatocytes function as DAMP molecules for TLR4, and they are responsible for the activation of innate immunity and tissue damage following APAP hepatotoxicity in mice [22,24,27]. Suppression of TLR4 activation through the silencing or inhibition of TLR4 ligands resulted in amelioration of the injury caused by APAP overdose. Mice were also protected from AILI by the inhibition or genetic deletion of TLR4, indicating that TLR4 activation is critical for 50
Biochemical Pharmacology 166 (2019) 46–55
K.-Y. Lee, et al.
Fig. 3. Silencing of fetuin-A downregulates transcription of Ccl2 in primary mouse hepatocytes. Primary mouse hepatocytes were isolated, stabilized, infected with adenoviral shRNA targeting Ahsg or a mock treatment for 48 h. (A) mRNA levels were measured using qRT-PCR. Relative gene expression levels are represented as fold changes compared with the control. Fetuin-A levels in culture media were analyzed using ELISA. (B) mRNA levels were measured using qRT-PCR. Relative gene expression levels are represented as fold changes compared with the control. All experiments were performed three times and all data were expressed mean ± SEM. * p < 0.05, **p < 0.01, ***p < 0.001 compared to the control group.
Fig. 4. APAP-induced fetuin-A expression is evident in the region of centrilobular necrosis. Mice were injected with Ad-sh-mock or Ad-sh-Ahsg via the tail vein, treated with APAP, and sacrificed after 24 or 48 h. (A) Western blot analysis was used to determine the abundance of fetuin-A in mouse liver homogenates. Band densities were determined using an image analysis system and expressed as percentages of the control (n = 4). (B) mRNA levels were measured using qRT-PCR. Relative gene expression levels are represented as fold changes compared with the control (n ≥ 7). (C) Serum fetuin-A concentrations were analyzed using ELISA (n = 6). (D) Immunohistochemical analysis was performed using paraffin-embedded liver sections with anti-fetuin-A antibodies. Representative micrographs are shown. Original magnification, 200 × . Scale bar, 100 μm. All data were expressed Mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001 compared to Ad-shRNA-mock group, ###p < 0.001 compared to the corresponding group.
both alleviate AILI [31]. However, opposing results have also been published, with the above results explained as off-target and preconditioning effects [32,33]. Deletion of ICAM-1 or inhibition of NADPH oxidase, the primary superoxide inducer in neutrophils, did not protect against AILI. No significant differences were observed between
The role of infiltrating leukocytes in APAP-induced hepatotoxicity has been controversial. Neutrophils are recruited to an injury site by the chemokines CXCL1, CXCL2, CCL3, and CCL4. Some researchers claim that neutrophils contribute to APAP hepatotoxicity based on results showing that pretreatment with anti-Gr1 antibody and ICAM-1 deletion 51
Biochemical Pharmacology 166 (2019) 46–55
K.-Y. Lee, et al.
Fig. 5. APAP treatment increased protein interaction between TLR4 and fetuin-A, which was attenuated with the silencing of fetuin-A. Mice were injected with Adsh-mock or Ad-sh-Ahsg via the tail vein, treated with APAP, and sacrificed after 24 h. (A) mRNA levels were measured using qRT-PCR. Relative gene expression levels are represented as fold changes compared with the control (n = 4). (B) The liver microsome fraction was prepared, and CYP2E1 activity was evaluated using the liver microsome by measuring the formation of p-nitrocatechol from p-nitrophenol per mg protein/min (n = 4). (C) Proximity ligation assays with anti-TLR4 and antifetuin-A antibodies were performed using OTC-embedded liver sections as described in the Materials and Methods section. The number of red puncta denoting PLA signals was counted using ImageJ software. Five images were randomly selected from each of the four treatments. Representative micrographs are shown. Scale bar, 20 μm. All data were expressed mean ± SD. *p < 0.05, ***p < 0.001 compared to the Ad-shRNA-mock group, #p < 0.05 compared to the corresponding group. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
wild-type and gp91phox-knockout mice in an APAP challenge [34]. Moreover, APAP hepatotoxicity was not rescued by the deletion of CD18, which is critical for the adhesion of neutrophils to hepatocytes and the generation of reactive oxygen species. These results suggest that neutrophils play a role in the regeneration process, rather than detoxification, following APAP overdose [35,36]. On the other hand, the role of MoMF in APAP hepatotoxicity depends on the time since APAP exposure. CCR2-positive monocytes and their descendants mediate the inflammatory response during the early phase of AILI [10], whereas MoMF is involved in injury repair and resolution during the late phase [37]. We observed that the recruitment of MoMFs and neutrophils was attenuated in the Ad-sh-Ahsg/APAP group compared to the Ad-shmock/APAP group. We found that the expression of CCL2 was regulated by fetuin-A in hepatocytes and macrophages. The silencing of fetuin-A decreased the transcription of CCL2, which suppressed the migration of monocytes. Although the number of MoMFs was reduced in Ad-sh-Ahsg mice compared to the Ad-sh-mock mice after APAP exposure, injury repair was not impaired, and inflamed tissue recovered faster than that of control mice. This difference is likely due to the relatively small
decrease in infiltration by MoMF in the Ad-sh-Ahsg/APAP group compared to the Ad-sh-mock/APAP group. Fetuin-A is an acute-phase protein that is synthesized and released from the liver following infection and injury. In contrast to our findings, some reports have suggested that fetuin-A also plays an anti-inflammatory role. Fetuin-A functions as an antagonist of the type II TGFβ receptor, with which it shares an 18-amino-acid homologous sequence [38]. In an animal model of endotoxemia and sepsis, circulating fetuin-A levels were reduced, which was negatively correlated with levels of proinflammatory cytokines. Disruption of fetuin-A expression rendered animals more susceptible to systemic inflammation, whereas supplementation of fetuin-A increased animal survival rates in a dosedependent manner, suggesting an anti-inflammatory function of fetuinA [39]. High levels of fetuin-A in cultured macrophages reduced lipopolysaccharide (LPS)-induced IL-1β and nitric oxide production [40]. The reason for this discrepancy remains elusive, but one possible explanation is that fetuin-A could compete with LPS for the TLR4 binding site, as both are TLR4 ligands. Given that TLR4 activation by LPS is stronger than that by fetuin-A, LPS-induced production of inflammatory 52
Biochemical Pharmacology 166 (2019) 46–55
K.-Y. Lee, et al.
Fig. 6. APAP activated the transcription of genes downstream of TLR4, which was attenuated with the silencing of fetuin-A. Mice were injected with Ad-sh-mock or Ad-sh-Ahsg via the tail vein, treated with APAP, and sacrificed after 24 h. (A) mRNA levels were measured using qRT-PCR. Relative gene expression levels are represented as fold changes compared with the control (n ≥ 4). (B) Serum CCL2 concentrations were analyzed using ELISA (n = 6). All data were expressed mean ± SD. ***p < 0.001 compared to Ad-shRNA-mock group, #p < 0.05, ##p < 0.01, ###p < 0.001 compared to the corresponding group.
Fig. 7. Silencing of fetuin-A suppressed the monocyte infiltration induced by APAP. Mice were injected with Ad-sh-mock or Ad-sh-Ahsg via the tail vein, treated with APAP, and then hepatic nonparenchymal cells were isolated from them after 24 h. (A) Gating strategies to analyze subpopulations of nonparenchymal cells using flow cytometry are shown in representative plots. Total leukocytes (CD45.2+), granulocytes (CD11b+; Ly6Ghigh), nongranulocytes (CD11b+; Ly6Glow), Kupffer cells (F4/ 80high; CD11blow), and MoMF (F4/80intermediate; CD11bhigh) were gated as shown above. (B) Nonparenchymal cells were analyzed using flow cytometry. The numbers of total leukocytes, neutrophils, Kupffer cells, and MoMFs were counted (n = 6). All data were expressed Mean ± SD. ***p < 0.001, *p < 0.05 compared to AdshRNA-mock group, ##p < 0.01, ###p < 0.001 compared to the corresponding group. 53
Biochemical Pharmacology 166 (2019) 46–55
K.-Y. Lee, et al.
Fig. 8. Silencing of fetuin-A alleviates APAP-induced hepatotoxicity. Mice were injected with Ad-sh-mock or Ad-sh-Ahsg via the tail vein, treated with APAP, and sacrificed after 24 or 48 h. (A) Serum levels of ALT, AST, and LDH were analyzed (n ≥ 7). (B) Liver histology was analyzed using H&E staining, and the percent necrotic area was quantified. Original magnification, 40×. Scale bar, 100 μm. (n = 6). (C) Serum HMGB1 concentrations were analyzed using ELISA (n = 6). All data were expressed mean ± SD. **p < 0.001, ***p < 0.001 compared to Ad-shRNA-mock group, ##p < 0.01, ###p < 0.001 compared to the corresponding group.
cytokines may be compromised by fetuin-A. Further research is needed to verify this possibility. Moreover, different experimental conditions may result in the observed discrepancy. Fetuin-A at a concentration corresponding to the normal adult serum level had no effect on the macrophage response to LPS, but a higher concentration of fetuin-A, similar to fetal levels, prevented macrophages from being stimulated by LPS [40]. In conclusion, our study demonstrates that silencing fetuin-A protects against APAP-induced liver injury. This process is mediated by decreased interaction between fetuin-A and TLR4, which attenuates the transcription of pro-inflammatory cytokines and chemokines and, thus, the recruitment of MoMF. Our results indicate that an approach based on the antagonism of fetuin-A may be a novel therapeutic strategy for the treatment of acetaminophen-induced acute liver failure.
[5]
[6]
[7]
[8]
[9]
Acknowledgements
[10]
This work was supported by a grant of the Korea Healthcare Technology R&D Project, Ministry of Health & Welfare, Republic of Korea (HI14C0133) and by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No.2017R1A2B4003179).
[11]
[12]
[13]
Declarations of interest [14]
None. [15]
References [16]
[1] D. Larrey, G.-P. Pageaux, Drug-induced acute liver failure, Eur. J. Gastroenterol. Hepatol. 17 (2) (2005) 141–143. [2] A.D. Manthripragada, E.H. Zhou, D.S. Budnitz, M.C. Lovegrove, M.E. Willy, Characterization of acetaminophen overdose-related emergency department visits and hospitalizations in the United States, Pharmacoepidemiol. Drug. Saf. 20 (2011) 819–826. [3] D.C. Dahlin, G.T. Miwa, A.Y.H. Lu, S.D. Nelson, N-acetyl-p-benzoquinone imine: a cytochrome P-450-mediated oxidation product of acetaminophen, Proc. Natl. Acad. Sci. 81 (1984) 1327–1331. [4] N. Hanawa, M. Shinohara, B. Saberi, W.A. Gaarde, D. Han, N. Kaplowitz, Role of
[17]
[18]
[19]
54
JNK translocation to mitochondria leading to inhibition of mitochondria bioenergetics in acetaminophen-induced liver injury, J. Biol. Chem. 283 (20) (2008) 13565–13577. J.A. Hinson, A.B. Reid, S.S. McCullough, L.P. James, Acetaminophen-induced hepatotoxicity: role of metabolic activation, reactive oxygen/nitrogen species, and mitochondrial permeability transition, Drug. Metab. Rev. 36 (2004) 805–822. C.G. Antoniades, A. Quaglia, L.S. Taams, R.R. Mitry, M. Hussain, R. Abeles, et al., Source and characterization of hepatic macrophages in acetaminophen-induced acute liver failure in humans, Hepatology 56 (2012) 735–746. M.E. Blazka, M.R. Elwell, S.D. Holladay, R.E. Wilson, M.I. Luster, Histopathology of acetaminophen-induced liver changes: role of interleukin 1 alpha and tumor necrosis factor alpha, Toxicol. Pathol. 24 (2) (1996) 181–189. M.P. Holt, L.L. Cheng, C. Ju, Identification and characterization of infiltrating macrophages in acetaminophen-induced liver injury, J. Leukocyte Biol. 84 (2008) 1–12. D.M. Dambach, L.M. Watson, K.R. Gray, S.K. Durham, D.L. Laskin, Role of CCR2 in macrophage migration into the liver during acetaminophen-induced hepatotoxicity in the mouse, Hepatology 35 (5) (2002) 1093–1103. J.C. Mossanen, O. Krenkel, C. Ergen, O. Govaere, A. Liepelt, T. Puengel, et al., Chemokine (C-C motif) receptor 2-positive monocytes aggravate the early phase of acetaminophen-induced acute liver injury, Hepatology 64 (5) (2016) 1667–1682. Q. You, M. Holt, H. Yin, G. Li, C.J. Hu, C. Ju, Role of hepatic resident and infiltrating macrophages in liver repair after acute injury, Biochem. Pharmacol. 82 (2013) 836–843. P. Chartterjee, S. Seak, S. Mukherjee, R. Kundu, S. Mukherjee, S. Ray, et al., Adipocyte fetuin-A contributes to macrophage migration into adipose tissue and polarization of macropahges, J. Biol. Chem. 288 (39) (2013) 28324–28330. B. Denecke, S. Graber, C. Schafer, A. Heiss, W. Jahnen-Dechent, Tissue distribution and activity testing suggest a similar but not identical function of fetuin-B and fetuin-A, Biochem. J. 376 (2003) 135–145. P. Auberger, L. Falquerho, J.O. Contreres, G. Pages, G. Le Cam, B. Rossi, et al., Characterization of a natural inhibitor of the insulin receptor tyrosine kinase: cDNA cloning, purification, and anti-mitogenic activity, Cell 58 (1989) 631–640. J.H. Ix, M.G. Shlipak, V.M. Brandenburg, S. Ali, M. Ketteler, M.A. Whooley, Association between human fetuin-A and the metabolic syndrome: data from the Heart and Soul Study, Circulation 113 (2006) 1760–1767. S.T. Mathews, G.P. Singh, M. Ranalletta, V.J. Cintron, X. Qiang, A.S. Goustin, et al., Improved insulin sensitivity and resistance to weight gain in mice null for the Ahsg gene, Diabetes 51 (2002) 2450–2458. A.M. Hennige, H. Staiger, C. Wicke, F. Machicao, A. Fritsche, H.-U. Haring, et al., Fetuin-A induces cytokine expression and suppresses adiponectin production, PLoS ONE 3 (3) (2008) e1765. C. Naito, M. Hashimoto, K. Watanabe, R. Shirai, Y. Takahashi, M. Kojima, et al., Facilitatory effects of fetuin-A on atherosclerosis, Atherosclerosis 246 (2016) 344–351. D. Pal, S. Dasgupta, R. Kundu, S. Maitr, G. Das, S. Mukhopadhyay, et al., Fetuin-A acts as an endogenous ligand of TLR4 to promote lipid-induced insulin resistance,
Biochemical Pharmacology 166 (2019) 46–55
K.-Y. Lee, et al.
[30] C. Zhang, J. Feng, Z. Zhuo, S. Yang, W. Zhang, W. Wang, et al., Macrophage-derived IL-1a promotes sterile inflammation in a mouse model of acetaminophen hepatotoxicity, Cell Mol. Immunol. 14 (2017) 1–10. [31] Z.-X. Liu, B. Gunawan, N. Kaplowitz, Neutrophil depletion protects against murine acetaminophen hepatotoxicity, Hepatology 43 (6) (2006) 1220–1230. [32] C. Cover, J. Liu, A. Farhood, E. Malle, M.P. Waalkers, M.L. Bajt, Pathophysiological role of the acute inflammatory response during acetaminophen hepatotoxicity, Toxicol. Appl. Pharmacol. 216 (2006) 98–107. [33] H. Jaeschke, J. Liu, Neutrophil depletion protects against murine acetaminophen hepatotoxicity: another perspective, Hepatology 45 (6) (2007) 1588–1589. [34] L.P. James, S.S. McCullough, T.R. Knight, H. Jaeschke, J.A. Hinson, Acetaminophen toxicity in mice lacking NADPH oxidase activity: role of peroxynitrite formation and mitochondrial oxidant stress, Free Radical. Res. 37 (12) (2003) 1289–1297. [35] C.D. Williams, M.L. Bajt, A. Farhood, H. Jaeschke, Acetaminophen-induced hepatic neutrophil accumulation and inflammatory liver injury in CD18-deficient mice, Liver Int. 30 (9) (2010) 1280–1292. [36] C.D. Williams, M.L. Bajt, M.R. Sharpe, M.R. McGill, A. Farhood, H. Jaeschke, Neutrophil activation during acetaminophen hepatotoxicity and repair in mice and humans, Toxicol. Appl. Pharmacol. 275 (2) (2014) 122–133. [37] E. Zigmond, S. Samia-Grinberg, M. Pasmanik-Chor, O. Shbolet, Z. Halpern, C. Varol, Infiltrating monocyte-derived macrophages and resident kupffer cells disply different ontogeny and functions in acute liver injury, J. Immunol. 193 (2014) 344–353. [38] M. Demetriou, C. Binkert, B. Sukhu, H.C. Tenenbaum, J.W. Dennis, Fetuin/a2-HS glycoprotein is a transforming growth factor-b type II receptor mimic and cytokine antagonist, J. Biol. Chem. 271 (22) (1996) 12755–12761. [39] W. Li, S. Zhu, J. Li, Y. Huang, R. Zhou, X. Fan, et al., A hepatic protein, fetuin-A, occupies a protective role in lethal systemic inflammation, PLoS ONE 6 (2) (2011) e16945. [40] K.M. Dziegielewska, N.A. Andersen, N.R. Saunder, Modification of macrophage response to lipopolysaccharide by fetuin, Immunol. Lett. 60 (1) (1998) 31–35.
Nat. Med. 18 (2012) 1279–1285. [20] Y.J. Choi, K.Y. Lee, S.H. Jung, H.S. Kim, G. Shim, M.G. Kim, et al., Activation of AMPK by berberine induces hepatic lipid accumulation by upregulation of fatty acid translocase CD36 in mice, Toxicol. Appl. Pharmacol. 316 (2017) 74–82. [21] J. Weischenfeldt, B. Porse, Bone marrow-derived macrophages (BMM): isolation and applications, CSH Protoc. 3 (12) (2008) 1–8. [22] B.V. Martin-Murphy, M.P. Holt, C. Ju, The role of damage associated molecular patter molecules in acetaminophen-induced liver injury in mice, Toxicol. Lett. 192 (2010) 387–394. [23] H. Jaeschke, M.R. McGill, A. Ramachandran, Oxidant stress, mitochondria, and cell death mechanisms in drug-induced liver injury: lessons learned from acetaminophen hepatotoxicity, Drug Metab. Rev. 44 (1) (2012) 88–106. [24] J. Xu, X. Zhang, M. Monestier, N.L. Esmon, C. Esmon, Extracellular histones are mediators of death through TLR2 and TLR4 in mouse fatal liver injury, J. Immunol. 202 (1) (2011) 1–6. [25] K.A. Cavassani, A.P. Moreira, D. Habiel, T. Ito, A.L. Coelho, R.M. Allen, et al., Toll like receptor 3 plays a critical role in the progression and severity of acetaminophen-induced hepatotoxicity, PLoS ONE 8 (6) (2013) e65899. [26] A.B. Imaeda, A. Watanabe, M.A. Sohail, S. Mahmood|, M. Mohamadnejad, F.S. Sutterwala, et al., Acetaminophen-induced hepatotoxicity in mice is dependent on Tlr9 and the Nalp3 inflammasome, J. Clin. Invest. 119 (2) (2009) 305–314. [27] X. Wang, R. Sun, H. Wei, Z. Tian, High-mobility group box 1 (HMGB1)-Toll-like receptor (TLR)4-interleukin (IL)-23-IL-17A axis in drug-induced damage-associated lethal hepatitis: Interaction of gammadelta T cells with macrophages, Hepatology 57 (2013) 373–384. [28] J.E. Fisher, T.J. McKenzie, J.B. Lillegard, Y. Yu, J.E. Juskewitch, G.I. Nedredal, et al., Role of Kupffer cells and toll-like receptor 4 in acetaminophen-induced acute liver failure, J. Surg. Res. 180 (2013) 147–155. [29] N. Shah, M. Montes de Oca, M. Jover-Cobos, K. Tanamoto, M. Muroi, K. Sugiyama, et al., Role of Toll-like receptor 4 in mediating multiorgan dysfunction in mice with acetaminophen induced acute liver failure, Liver Transpl. 19 (2013) 751–761.
55
Contents lists available at ScienceDirect
Biochemical Pharmacology journal homepage: www.elsevier.com/locate/biochempharm
Hepatic upregulation of fetuin-A mediates acetaminophen-induced liver injury through activation of TLR4 in mice
T
Kang-Yo Lee, Wonseok Lee, Seung-Hwan Jung, Jungmin Park, Hyungtai Sim, You-Jin Choi, ⁎ Young-Jun Park, Yeonseok Chung, Byung-Hoon Lee College of Pharmacy and Research Institute of Pharmaceutical Sciences, Seoul National University, Gwanakro 1, Gwanak-gu, Seoul 08826, Republic of Korea
A R T I C LE I N FO
A B S T R A C T
Keywords: Acetaminophen Fetuin-A Hepatokine Chemokine Hepatotoxicity
Acetaminophen (APAP)-induced liver injury (AILI) is initiated by the generation of a reactive metabolite and ultimately leads to hepatocyte necrosis. Necrotic cells secrete damage-associated molecular patterns that activate hepatic nonparenchymal cells and induce an inflammatory response. Fetuin-A is a hepatokine with reported involvement in low-grade inflammation in many diseases, due to acting as an endogenous ligand for TLR4. However, little is known about the role of fetuin-A in AILI. In this study, we showed that fetuin-A is involved in the aggravation of hepatotoxicity during the initial phase of AILI progression. Treatment with APAP increased the expression and serum levels of fetuin-A in mice. Fetuin-A upregulated transcription of pro-inflammatory cytokines and chemokines through activation of TLR4 and also increased monocyte infiltration into the liver, leading to necroinflammatory reactions in AILI. However, these reactions were attenuated with the silencing of fetuin-A using adenoviral shRNA. As a result, mice with silenced fetuin-A exhibited less centrilobular necrosis and liver injury compared to controls in response to APAP. In conclusion, our results suggest that fetuin-A is an important hepatokine that mediates the hepatotoxicity of APAP through production of chemokines and thus regulates the infiltration of monocytes into the liver, a critical event in the inflammatory response during the initial phase of AILI. Our results indicate that a strategy based on the antagonism of fetuin-A may be a novel therapeutic approach to the treatment of acetaminophen-induced acute liver failure.
1. Introduction Acetaminophen (N-acetyl-p-aminophenol; APAP), which is one of the most commonly used analgesic and antipyretic drugs worldwide, is available over-the-counter and by prescription either as a single agent or in combination with other medications. Although it is safe when administered at the recommended dosage, APAP overdose can cause severe hepatotoxicity in experimental animals and humans. About half of all cases of acute liver failure in the US and Western countries are attributable to APAP overdose [1,2]. APAP-induced liver injury (AILI) is divided into three stages: initiation, progression, and regeneration. CYP2E1-mediated formation of N-acetyl-p-benzoquinone imine (NAPQI), a reactive metabolite of APAP, is the initiating event of APAP hepatotoxicity [3]. NAPQI depletes the cellular glutathione pool and forms a protein adduct, especially in the mitochondria. Formation of protein adducts in the electron transport chain following APAP overdose leads to inhibition of mitochondrial respiration and formation of mitochondrial oxidative stress, resulting in mitochondrial dysfunction
and oncotic necrosis of hepatocytes [4,5]. As a result of the necrotic death of hepatocytes, inflammatory mediators, such as damage-associated molecular patterns (DAMPs), activate hepatic nonparenchymal cells, resulting in sterile inflammation and progression to liver injury. A variety of cytokines released by immune cells are involved in the aggravation of hepatotoxicity via immune-mediated mechanisms [6,7]. Although the number of resident Kupffer cells decreases rapidly due to APAP cytotoxicity [8], huge numbers of monocytes are recruited into the liver, where they differentiate into monocyte-derived macrophages (MoMF) in a CCR2-dependent manner [8–10]. During the onset of acute AILI, the infiltrating MoMFs aggravate inflammation and hepatotoxicity without impairing the repair process, as confirmed by the attenuation of liver damage with the blocking of either CCL2 or CCR2/CCR5 [10]. MoMFs are indispensable for the repair and resolution of tissue injury during the recovery phase. Infiltrated macrophages express angiogenic factors, including VEGF and ANGPTL, which assist with recovery of the sinusoid and microcirculation in the liver [11]. MoMFs influence the restoration
⁎ Corresponding author at: 29-321, College of Pharmacy and Research Institute of Pharmaceutical Sciences, Seoul National University, Gwanakro 1, Gwanak-gu, Seoul 08826, Republic of Korea. E-mail address: [email protected] (B.-H. Lee).
https://doi.org/10.1016/j.bcp.2019.05.011 Received 21 March 2019; Accepted 6 May 2019 Available online 08 May 2019 0006-2952/ © 2019 Elsevier Inc. All rights reserved.
Biochemical Pharmacology 166 (2019) 46–55
K.-Y. Lee, et al.
(iNtRON, Seoul, Korea) and complementary DNA was prepared with QuantiTect Reverse Transcription Kit (QIAGEN, Hilden, Germany). cDNA was amplified by qRT-PCR using iTaq Universal SYBR Green Supermix kit (Bio-Rad, Hercules, CA).
process through the expression of Ym1, Fizz1, Arg-1, MMP-12, and MMP-9, representative markers of M2-type polarization. Therefore, deletion of CCR2 delays the repair process in AILI [8]. Fetuin-A, also known as α2-Heremans-Schmid glycoprotein (AHSG), is a secretory protein produced primarily by the liver and also by other tissues, including the tongue, placenta, and adipose tissue [12,13]. Fetuin-A was initially identified as an inducer of insulin resistance and was also found to be strongly associated with metabolic syndrome in humans [14–16]. These results suggested a possible association of fetuin-A with low-grade inflammation, which was confirmed recently with experimental and clinical data. Fetuin-A was shown to induce inflammatory cytokine expression and suppress adiponectin production in adipocytes, monocytes, and in mice [17]. Recent research indicates that fetuin-A facilitates the development of atherosclerosis through its stimulatory effects on the inflammatory response in endothelial cells and foam cell formation in macrophages [18]. Fetuin-A acts as an endogenous ligand of toll-like receptor 4 (TLR4), promoting the lipid-induced proinflammatory response and insulin resistance [19]. Moreover, lipid-induced expression of fetuin-A from adipocytes acts as an efficient chemokine to induce macrophage migration and polarization [12]. Resident macrophages and infiltrating myeloid cells secrete a variety of chemokines and cytokines, leading to sterile inflammation in AILI. As the interaction of fetuin-A with TLR4 provokes an inflammatory response, we hypothesized that fetuin-A acts as an unrecognized endogenous DAMP-like molecule and, thus, is an important mediator of AILI. In this study, we first showed that hepatokine fetuin-A is involved in the hepatotoxicity caused by APAP overdose. We demonstrated that APAP increased the expression and secretion of fetuinA in mice. Fetuin-A upregulated the expression of chemokines in hepatocytes and bone marrow-derived macrophage (BMDM) cells, leading to recruitment of monocytes to the liver. Furthermore, fetuin-A-mediated production of cytokines and chemokines aggravated APAP-induced hepatotoxicity.
2.4. Immunoblot Tissue homogenate was lysed in buffer containing 50 mM HEPES (Sigma-Aldrich), 150 mM NaCl (Sigma-Aldrich), 5 mM EGTA (SigmaAldrich), 50 mM β-glycerophosphate (Sigma-Aldrich), 1% Triton-X 100 (Sigma-Aldrich) and protease inhibitor cocktail (Roche, Indianapolis, IN). The lysates were then centrifuged and the supernatant was separated by SDS-polyacrylamide gel electrophoresis and then transferred to PVDF membrane (Millipore, Darmstadt, Germany). Transferred proteins were blocked in 5% skim milk, probed with anti-fetuin-A (sc20872, Santa Cruz Biotechnology, Santa Cruz, CA) or anti-GAPDH antibody (#2118, Cell Signaling Technology, Beverly, MA) and developed by enhanced chemiluminescence western blotting reagent (Amersham, Piscataway, NJ). 2.5. CYP2E1 activity assay Liver tissue homogenate was prepared in cold 0.15 M KCl (USB, Cleveland, OH) and centrifuged at 9000g. Supernatant was transferred to a new tube and centrifuged at 100,000g for 1 h. Microsomal pellet was resuspended in 0.15 M KCl. 200 μg of microsomal protein was incubated in 0.1 M potassium phosphate (Sigma-Aldrich) buffer with 0.4 mM p-nitrophenol (Sigma-Aldrich) and 1 mM NADPH (Roche) for 30 min at 37 °C water bath. Reaction was terminated by addition of 50% trichloroacetic acid (Sigma-Aldrich) and reaction mixture was centrifuged at 10,000g. 20 μl of 10 N NaOH (Duksan, Ansan, Korea) was added to supernatant and generation of p-nitrocatechol was determined by measuring absorbance with spectrophotometer (VersaMax, Molecular Devices, Sunnyvale, CA) at 546 nm. CYP2E1 activity of samples was deduced from standard curve of p-nitrocatechol (SigmaAldrich).
2. Materials and methods 2.1. Animal experiments
2.6. Histological analysis and immunohistochemistry
7 to 8 weeks old (20–23 g) male SPF C57BL/6J mice from Central Lab. Animal Inc. (Seoul, Korea) were housed in air-conditioned room with 12 light/dark cycle and allowed free access to water and food. Mice were fasted overnight, injected with APAP (Sigma-Aldrich, St Louis, MO) or vehicle (PBS) intraperitoneally. For in vivo silencing experiment, mice were injected with 1010 adenoviral particle containing shRNA against Ahsg (Ad-shRNA-Ahsg) or mock (Ad-shRNAmock) (Welgen Inc, Worcester, MA) intravenously. After 3 days from adenoviral knockdown, mice received APAP or vehicle. Mice were anesthetized with Alfaxan (Jurox, Kansas City, MO) and sacrificed at indicated time point. All animal experiments were performed in accordance with the guidelines of the Institutional Animal Care and Use Committee of Seoul National University.
The liver tissues were fixed in neutral buffered formalin (SigmaAldrich) and embedded in paraffin. Paraffin blocks were cut into 5 μm sections and stained with hematoxylin and eosin (H&E). Necrotic area was quantified using ImageJ (NIH, Bethesda, ML). For immunostaining, tissue slides were incubated with primary antibody directed against fetuin-A (ab187051, Abcam, Cambridge, UK) and counterstained with hematoxylin. 2.7. Proximity ligation assay (PLA) PLA was performed with Duolink PLA kit (Sigma-Aldrich) according to manufacturer’s protocol. In brief, O.C.T (Sakura Finetek, Torrance, CA)-embedded tissue slides were fixed in ice cold methanol, blocked, and probed with anti-TLR4 (sc-293072, Santa Cruz Biotechnology) and anti-fetuin-A (sc-20872, Santa Cruz Biotechnology). PLA probe was adjusted to the samples followed by addition of ligase and polymerase. Slides were covered with mounting medium with DAPI and analyzed with TCS SP8 confocal laser scanning microscope (Leica Microsystems, Wetzlar, Germany). To obtain images throughout the entire thickness of the slides, images of 15 z-plane were acquired for all presentation.
2.2. Cell isolation and treatment Primary mouse hepatocytes were isolated from 7 to 8 weeks old male SPF C57BL/6J mice via collagenase type IV (Sigma-Aldrich) perfusion as described previously [20]. Cells were stabilized and treated as indicated. BMDM were isolated from hind leg of male C57BL/6J mice as described previously [21]. Cells were differentiated into macrophages with M-CSF (Peprotech, Rocky Hill, NJ) at concentration of 30 ng/ml and treated with or without mouse fetuin-A (Sino Biological Inc, Beijing, China).
2.8. Serum biochemistry
2.3. Quantitative real-time polymerase chain reaction
Serum levels of ALT, AST and LDH were analyzed by standard clinical chemistry assays on an Automated Chemistry Analyzer (Prestige 24I; Tokyo Boeki Medical System, Tokyo, Japan).
mRNA was extracted with the Easy-Blue Total RNA extraction kit 47
Biochemical Pharmacology 166 (2019) 46–55
K.-Y. Lee, et al.
2.9. Flow cytometry
3.3. APAP induced fetuin-A co-localization with TLR4 in the liver
Hepatic nonparenchymal cells were isolated from the whole livers of mice by the homogenization using the GentleMACS Dissociator (Miltenyi Biotec, Bergisch Gladbach, Germany). Following dissociation into a single cell suspension, cells were preincubated with FcR blocker (Biolegend, San Diego, CA, #101319), stained with fluorochromeconjugated antibodies and analyzed with LSR Foresta (BD Biosciences, San Jose, CA). Antibodies used for analysis are from Biolegend (San Diego, CA): APC/Cy7-CD45.2 (#109823), FITC-Ly6G (#127605), APCF4/80 (#123116), PerCP/Cy5.5-CD11b (101227), PE/Cy7-Ly6C (#128017).
To explore the role of fetuin-A in the inflammatory reaction during the initial phase of AILI progression, 8 mice per group were injected with adenoviral shRNA that was nonspecific or against the Ahsg fetuinA gene for 3 days prior to APAP administration. Mouse which was not silenced of fetuin-A was excluded for analysis. APAP treatment led to the upregulation of mRNA and the protein expression levels of fetuin-A, which was alleviated with the silencing of fetuin-A (Fig. 4A, B). In agreement with this finding, elevated serum fetuin-A levels caused by APAP treatment were reduced with the ablation of fetuin-A (Fig. 4C). Immunohistochemistry showed intensive staining of fetuin-A with APAP treatment. We found that the necrotic region around the central vein was significantly positive in the fetuin-A staining assay in APAPtreated mice but was attenuated by the silencing of fetuin-A (Fig. 4D). To exclude the possibility of the nonspecific binding of the secondary antibodies to the damaged region, a negative control experiment was performed without the primary antibody. As shown in Fig. 4D, nonspecific binding of the antibody did not occur. We also found that the expression and activity of CYP2E1 were not affected by the silencing of fetuin-A (Fig. 5A, B). To determine whether the silencing of fetuin-A had any effect on its binding to TLR4, we performed a proximity ligation assay (PLA) between fetuin-A and TLR4, which allows in situ detection of protein–protein interactions with high specificity and sensitivity. As a negative control, we performed the same assay without primary antibodies, which should generate no PLA signal. Compared to vehicle treatment, a strong PLA signal (red dots) was detected in the APAP group, especially around the necrotic area. In contrast, a significantly fewer PLA puncta were observed in the Ad-sh-Ahsg/APAP group. For the quantification of PLA signal, we selected necrotic area around the central vein and analyzed 5 areas (approximately 500,000 μm2) per group using ImageJ software (Fig. 5C). Co-immunoprecipitation did not detect binding due to the similar molecular weights of fetuin-A and the antibody heavy chain. These results suggest that APAP increases hepatic expression of fetuin-A and its binding to TLR4 in the liver.
2.10. Statistical analysis Data were presented as mean ± S.D. for in vivo experiments and mean ± S.E.M for in vitro experiments. Data were analyzed either by student’s t-test or one-way ANOVA analysis followed by Turkey’s multiple comparison test with GraphPad Prism 5 (GraphPad, San Diego, CA). P values < 0.05 were considered statistically significant. 3. Results 3.1. APAP increased hepatic expression and secretion of fetuin-A C57BL/6J mice were treated with APAP through intraperitoneal injection, and the induction of hepatotoxicity was confirmed based on elevated serum levels of ALT, AST, and LDH at 6, 24, and 48 h following treatment. APAP-induced hepatotoxicity was ameliorated at 48 h, which corresponds with the known time of repair and resolution for such tissue injury (Fig. 1A). Most subsequent animal experiments were performed 24 h after APAP treatment. We analyzed the effects of APAP on the hepatic expression of fetuin-A. APAP increased the expression of fetuin-A, which was maintained until 48 h. Of the two bands for fetuinA on immunoblots, the intensity of the protein with the higher molecular weight, reportedly the glycosylated and secretable form of fetuinA, increased more significantly (Fig. 1B). The attenuation of fetuin-A expression at 48 h mirrored the decrease in hepatotoxicity at the same time. Expression of Ahsg, the official gene name for fetuin-A, was also upregulated at 6 and 24 h of APAP treatment (Fig. 1C). As a result, serum levels of fetuin-A gradually increased after 24 h, peaking at 48 h (Fig. 1D). Taken together, these results show that the expression and the serum levels of fetuin-A increased with APAP treatment in mice.
3.4. Silencing fetuin-A suppresses chemokine expression and monocyte infiltration by APAP in mice. TLR4 signaling is associated with the production of chemotactic and inflammatory cytokines, as well as of inflammatory infiltrate consisting of neutrophils and macrophages. The silencing of fetuin-A slightly, albeit non-significantly, reduced the mRNA expression of all chemokines and cytokines. On the other hand, silencing of fetuin-A caused significantly downregulated expression of transcripts for the chemokines (Ccl2, Ccl3, and Ccl4) and cytokines (Tnfa and Il6) that were highly upregulated with APAP treatment (Fig. 6A). Similarly, the elevated serum level of CCL2 by APAP decreased with the silencing of fetuin-A (Fig. 6B). CCL2 is one of the main chemokines driving the infiltration of myeloid cells into APAP-inflamed tissue via the CCL2–CCR2 axis [8–10]. To investigate whether the silencing of fetuin-A attenuates the infiltration of myeloid cells, we isolated hepatic nonparenchymal cells from APAP-treated mice and analyzed them using flow cytometry with the gating strategy depicted in Fig. 7A [29]. We found that APAP injection induced massive infiltration of neutrophils into the liver based on CD45.2- and Ly6G-positive cell counts. The number of infiltrated MoMF (F4/80intermediate; CD11bhigh) also increased substantially with APAP treatment, whereas Kupffer cells (F4/80high; CD11blow) decreased, likely due to the direct cytotoxic effects of APAP. In contrast, the number of MoMF decreased significantly in mice with silenced fetuin-A genes, which might be related to the downregulation of CCL2 (Fig. 6B). Taken together, these results show that APAP upregulates the expression of fetuin-A, which, in turn, enhances the production of proinflammatory cytokines and chemokines. Elevated levels of chemokines, such as CCL2, induce monocyte infiltration. Silencing fetuin-A
3.2. Fetuin-A upregulated the transcription of chemokines Hepatic nonparenchymal cells secrete chemokines that contribute to the massive accumulation and activation of myeloid cells in APAP-induced hepatotoxicity [10,22]. To investigate whether APAP-induced fetuin-A expression has a causal relationship with AILI, we first analyzed the expression of chemokines and cytokines in the liver. In accordance with previous findings, administration of APAP to mice strongly upregulated the transcript levels of chemokines, including Ccl2, Ccl3, and Ccl4, as well as cytokines, such as Il6, and Il1a (Fig. 2A). APAP treatment also induced transcript level of Tnfa, which not statistically significant. Moreover, fetuin-A increased expression of Ccl2, Ccl3, and Ccl4 in mouse primary hepatocytes and of Ccl2, Ccl3, Tnfa, Il1a, and Il6 in BMDM (Fig. 2B, C). Knockdown of fetuin-A in mouse primary hepatocytes decreased the expression and secretion of fetuin-A into the culture medium (Fig. 3A). Silencing fetuin-A in hepatocytes led to significantly downregulated expression of Ccl2. However, the mRNA levels of Ccl3 and Ccl4 were not affected by silencing of fetuin-A probably due to the low basal expression in hepatocytes. (Fig. 3B). From these results, we can conclude that fetuin-A regulates the gene expression of chemokines and cytokines in mouse primary hepatocytes, as well as in macrophages. 48
Biochemical Pharmacology 166 (2019) 46–55
K.-Y. Lee, et al.
Fig. 1. APAP increases hepatic expression and secretion of fetuin-A in mice. Male C57/BL6 mice were injected with APAP (250 mg/kg; i.p.) and sacrificed after 6, 24, or 48 h. (A) Serum ALT, AST, and LDH levels were analyzed. (B) Western blot analysis was used to determine the abundance of fetuin-A in mouse liver homogenates. Band densities were determined using an image analysis system and expressed as percentages of the control. (C) mRNA levels were measured using qRT-PCR. Relative gene expression levels are represented as fold changes compared with the control. (D) Serum fetuin-A concentrations were analyzed using ELISA. n ≥ 4 for any time point. All data were expressed mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001 compared to the corresponding vehicle group.
4. Discussion
reduced the number of infiltrating MoMF, which increased substantially with APAP administration.
AILI is initiated with the generation of a reactive metabolite, NAPQI, and the subsequent depletion of the cellular glutathione pool, leading to the generation of reactive oxygen species, the formation of protein adducts, mitochondrial dysfunction, and, finally, cell death [23]. Hepatocyte necrosis activates Kupffer cells that secrete proinflammatory cytokines and chemokines, leading to sterile inflammation. This process is mediated by the release of intracellular constituents into the surrounding milieu, some of which belong to a group of molecules known as DAMPs. DAMPs are recognized by pattern recognition receptors, including the TLR family. Several types of DAMPs have been reported to mediate the sterile inflammation process, thereby causing a second wave of destruction and contributing to the pathogenesis of AILI [22,24]. The present study demonstrates the fundamental role of the hepatokine fetuin-A in APAP-induced hepatotoxicity. The key findings of the study include that APAP increases hepatic expression and secretion of fetuin-A, which mediates the upregulation of proinflammatory cytokines and chemokines through interaction with TLR4. Elevated levels of chemokines, such as CCL2, trigger the recruitment of monocytes, which play an important role in the aggravation of hepatotoxicity during the initial phase of AILI progression. Ablation of fetuin-A in the livers of mice significantly reduced the hepatic necrotic area following APAP administration, suggesting that fetuin-A plays a role in mediating APAP-induced hepatotoxicity. TLRs are a family of pattern recognition receptors that participate in the inflammatory process by recognizing DAMPs. Although various
3.5. Silencing Ahsg attenuated APAP-induced hepatotoxicity in mice Finally, we evaluated the role of fetuin-A in APAP-induced hepatotoxicity in mice with silenced fetuin-A expression. As shown in Fig. 8, the levels of ALT, AST, and LDH increased significantly within 24 h in the Ad-sh-mock/APAP group, but it was alleviated in the Ad-sh-Ahsg/ APAP group. After 48 h of APAP administration, which is the time needed for the repair and resolution of tissue injury, most parameters of hepatotoxicity had decreased significantly. In this case, the silencing of fetuin-A had no effect (Fig. 8A). Histological examination revealed that there was no significant difference between the Ad-sh-mock/vehicle and Ad-sh-Ahsg/vehicle groups. Severe centrilobular necrosis was observed in the Ad-sh-mock/APAP group. However, Ad-sh-Ahsg/APAP mice had significantly smaller hepatic necrotic areas at both 24 and 48 h following APAP administration (Fig. 8B). As a result, the enhanced serum content of HMGB1 with APAP treatment decreased with the silencing of fetuin-A at 24 h (Fig. 8C). In conclusion, our results suggest that fetuin-A is an important hepatokine that mediates the hepatotoxicity of APAP through the production of chemokines, causing monocytes to infiltrate the liver, a critical event in the inflammatory reaction during the initial phase of AILI progression.
49
Biochemical Pharmacology 166 (2019) 46–55
K.-Y. Lee, et al.
Fig. 2. APAP treatment and fetuin-A upregulates transcription of genes downstream of TLR4. (A) Male C57/BL6 mice were injected with APAP and sacrificed after 24 h. mRNA levels were measured using qRT-PCR. Relative gene expression levels are represented as fold changes compared with the control. All data were expressed mean ± SD (n = 4 ∼ 5). (B) Primary hepatocytes were isolated, stabilized, and incubated in the presence or absence of 30 μg/ml fetuin-A for 24 h without FBS. (C) BMDMs were isolated, differentiated, and incubated in the presence or absence of 50 μg/ml of fetuin-A for 24 h without FBS. mRNA levels were measured using qRTPCR. Relative gene expression levels are represented as fold changes compared with the control. All experiments were performed three times and all data were expressed mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001 compared to the control group.
AILI [28–30]. Indeed, treatment with antibodies against TNF-α or IL1α, which are regulated by TLR4 signal transduction, rescued mice from AILI [7]. Zhang et al. emphasized that AILI is dependent on TLR4/ Myd88 signaling and the resultant production of IL-1α [40]. In the present study, we showed that APAP induced the hepatic expression and secretion of fetuin-A, which binds to TLR4 and induces expression of chemokines and cytokines, and that such production was attenuated with the silencing of fetuin-A. Similar results were observed in carbon tetrachloride-induced hepatotoxicity (data not shown). Therefore, the hepatokine fetuin-A can be defined as a DAMP-like molecule that mediates the progression of toxicity through sterile inflammation in AILI. This finding will provide new insights for the field of clinical research related to treatment of hepatic failure caused by APAP overdose, as well as for basic research related to the toxic mechanism of APAP.
liver cells express diverse TLRs, it has been suggested that TLR4, TLR3, and TLR9 signaling is involved in AILI [25,26]. DNA and RNA fragments released during APAP-induced hepatocellular necrosis can be recognized by TLR9 and TLR3, respectively, thereby inducing cytokine production through activation of NF-κB. The pathophysiological relevance of TLR3 and TLR9 to AILI has been suggested by the protective effects observed in TLR3- or TLR9-deficient mice [25,26]. HMGB1 and histones released from dying hepatocytes function as DAMP molecules for TLR4, and they are responsible for the activation of innate immunity and tissue damage following APAP hepatotoxicity in mice [22,24,27]. Suppression of TLR4 activation through the silencing or inhibition of TLR4 ligands resulted in amelioration of the injury caused by APAP overdose. Mice were also protected from AILI by the inhibition or genetic deletion of TLR4, indicating that TLR4 activation is critical for 50
Biochemical Pharmacology 166 (2019) 46–55
K.-Y. Lee, et al.
Fig. 3. Silencing of fetuin-A downregulates transcription of Ccl2 in primary mouse hepatocytes. Primary mouse hepatocytes were isolated, stabilized, infected with adenoviral shRNA targeting Ahsg or a mock treatment for 48 h. (A) mRNA levels were measured using qRT-PCR. Relative gene expression levels are represented as fold changes compared with the control. Fetuin-A levels in culture media were analyzed using ELISA. (B) mRNA levels were measured using qRT-PCR. Relative gene expression levels are represented as fold changes compared with the control. All experiments were performed three times and all data were expressed mean ± SEM. * p < 0.05, **p < 0.01, ***p < 0.001 compared to the control group.
Fig. 4. APAP-induced fetuin-A expression is evident in the region of centrilobular necrosis. Mice were injected with Ad-sh-mock or Ad-sh-Ahsg via the tail vein, treated with APAP, and sacrificed after 24 or 48 h. (A) Western blot analysis was used to determine the abundance of fetuin-A in mouse liver homogenates. Band densities were determined using an image analysis system and expressed as percentages of the control (n = 4). (B) mRNA levels were measured using qRT-PCR. Relative gene expression levels are represented as fold changes compared with the control (n ≥ 7). (C) Serum fetuin-A concentrations were analyzed using ELISA (n = 6). (D) Immunohistochemical analysis was performed using paraffin-embedded liver sections with anti-fetuin-A antibodies. Representative micrographs are shown. Original magnification, 200 × . Scale bar, 100 μm. All data were expressed Mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001 compared to Ad-shRNA-mock group, ###p < 0.001 compared to the corresponding group.
both alleviate AILI [31]. However, opposing results have also been published, with the above results explained as off-target and preconditioning effects [32,33]. Deletion of ICAM-1 or inhibition of NADPH oxidase, the primary superoxide inducer in neutrophils, did not protect against AILI. No significant differences were observed between
The role of infiltrating leukocytes in APAP-induced hepatotoxicity has been controversial. Neutrophils are recruited to an injury site by the chemokines CXCL1, CXCL2, CCL3, and CCL4. Some researchers claim that neutrophils contribute to APAP hepatotoxicity based on results showing that pretreatment with anti-Gr1 antibody and ICAM-1 deletion 51
Biochemical Pharmacology 166 (2019) 46–55
K.-Y. Lee, et al.
Fig. 5. APAP treatment increased protein interaction between TLR4 and fetuin-A, which was attenuated with the silencing of fetuin-A. Mice were injected with Adsh-mock or Ad-sh-Ahsg via the tail vein, treated with APAP, and sacrificed after 24 h. (A) mRNA levels were measured using qRT-PCR. Relative gene expression levels are represented as fold changes compared with the control (n = 4). (B) The liver microsome fraction was prepared, and CYP2E1 activity was evaluated using the liver microsome by measuring the formation of p-nitrocatechol from p-nitrophenol per mg protein/min (n = 4). (C) Proximity ligation assays with anti-TLR4 and antifetuin-A antibodies were performed using OTC-embedded liver sections as described in the Materials and Methods section. The number of red puncta denoting PLA signals was counted using ImageJ software. Five images were randomly selected from each of the four treatments. Representative micrographs are shown. Scale bar, 20 μm. All data were expressed mean ± SD. *p < 0.05, ***p < 0.001 compared to the Ad-shRNA-mock group, #p < 0.05 compared to the corresponding group. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
wild-type and gp91phox-knockout mice in an APAP challenge [34]. Moreover, APAP hepatotoxicity was not rescued by the deletion of CD18, which is critical for the adhesion of neutrophils to hepatocytes and the generation of reactive oxygen species. These results suggest that neutrophils play a role in the regeneration process, rather than detoxification, following APAP overdose [35,36]. On the other hand, the role of MoMF in APAP hepatotoxicity depends on the time since APAP exposure. CCR2-positive monocytes and their descendants mediate the inflammatory response during the early phase of AILI [10], whereas MoMF is involved in injury repair and resolution during the late phase [37]. We observed that the recruitment of MoMFs and neutrophils was attenuated in the Ad-sh-Ahsg/APAP group compared to the Ad-shmock/APAP group. We found that the expression of CCL2 was regulated by fetuin-A in hepatocytes and macrophages. The silencing of fetuin-A decreased the transcription of CCL2, which suppressed the migration of monocytes. Although the number of MoMFs was reduced in Ad-sh-Ahsg mice compared to the Ad-sh-mock mice after APAP exposure, injury repair was not impaired, and inflamed tissue recovered faster than that of control mice. This difference is likely due to the relatively small
decrease in infiltration by MoMF in the Ad-sh-Ahsg/APAP group compared to the Ad-sh-mock/APAP group. Fetuin-A is an acute-phase protein that is synthesized and released from the liver following infection and injury. In contrast to our findings, some reports have suggested that fetuin-A also plays an anti-inflammatory role. Fetuin-A functions as an antagonist of the type II TGFβ receptor, with which it shares an 18-amino-acid homologous sequence [38]. In an animal model of endotoxemia and sepsis, circulating fetuin-A levels were reduced, which was negatively correlated with levels of proinflammatory cytokines. Disruption of fetuin-A expression rendered animals more susceptible to systemic inflammation, whereas supplementation of fetuin-A increased animal survival rates in a dosedependent manner, suggesting an anti-inflammatory function of fetuinA [39]. High levels of fetuin-A in cultured macrophages reduced lipopolysaccharide (LPS)-induced IL-1β and nitric oxide production [40]. The reason for this discrepancy remains elusive, but one possible explanation is that fetuin-A could compete with LPS for the TLR4 binding site, as both are TLR4 ligands. Given that TLR4 activation by LPS is stronger than that by fetuin-A, LPS-induced production of inflammatory 52
Biochemical Pharmacology 166 (2019) 46–55
K.-Y. Lee, et al.
Fig. 6. APAP activated the transcription of genes downstream of TLR4, which was attenuated with the silencing of fetuin-A. Mice were injected with Ad-sh-mock or Ad-sh-Ahsg via the tail vein, treated with APAP, and sacrificed after 24 h. (A) mRNA levels were measured using qRT-PCR. Relative gene expression levels are represented as fold changes compared with the control (n ≥ 4). (B) Serum CCL2 concentrations were analyzed using ELISA (n = 6). All data were expressed mean ± SD. ***p < 0.001 compared to Ad-shRNA-mock group, #p < 0.05, ##p < 0.01, ###p < 0.001 compared to the corresponding group.
Fig. 7. Silencing of fetuin-A suppressed the monocyte infiltration induced by APAP. Mice were injected with Ad-sh-mock or Ad-sh-Ahsg via the tail vein, treated with APAP, and then hepatic nonparenchymal cells were isolated from them after 24 h. (A) Gating strategies to analyze subpopulations of nonparenchymal cells using flow cytometry are shown in representative plots. Total leukocytes (CD45.2+), granulocytes (CD11b+; Ly6Ghigh), nongranulocytes (CD11b+; Ly6Glow), Kupffer cells (F4/ 80high; CD11blow), and MoMF (F4/80intermediate; CD11bhigh) were gated as shown above. (B) Nonparenchymal cells were analyzed using flow cytometry. The numbers of total leukocytes, neutrophils, Kupffer cells, and MoMFs were counted (n = 6). All data were expressed Mean ± SD. ***p < 0.001, *p < 0.05 compared to AdshRNA-mock group, ##p < 0.01, ###p < 0.001 compared to the corresponding group. 53
Biochemical Pharmacology 166 (2019) 46–55
K.-Y. Lee, et al.
Fig. 8. Silencing of fetuin-A alleviates APAP-induced hepatotoxicity. Mice were injected with Ad-sh-mock or Ad-sh-Ahsg via the tail vein, treated with APAP, and sacrificed after 24 or 48 h. (A) Serum levels of ALT, AST, and LDH were analyzed (n ≥ 7). (B) Liver histology was analyzed using H&E staining, and the percent necrotic area was quantified. Original magnification, 40×. Scale bar, 100 μm. (n = 6). (C) Serum HMGB1 concentrations were analyzed using ELISA (n = 6). All data were expressed mean ± SD. **p < 0.001, ***p < 0.001 compared to Ad-shRNA-mock group, ##p < 0.01, ###p < 0.001 compared to the corresponding group.
cytokines may be compromised by fetuin-A. Further research is needed to verify this possibility. Moreover, different experimental conditions may result in the observed discrepancy. Fetuin-A at a concentration corresponding to the normal adult serum level had no effect on the macrophage response to LPS, but a higher concentration of fetuin-A, similar to fetal levels, prevented macrophages from being stimulated by LPS [40]. In conclusion, our study demonstrates that silencing fetuin-A protects against APAP-induced liver injury. This process is mediated by decreased interaction between fetuin-A and TLR4, which attenuates the transcription of pro-inflammatory cytokines and chemokines and, thus, the recruitment of MoMF. Our results indicate that an approach based on the antagonism of fetuin-A may be a novel therapeutic strategy for the treatment of acetaminophen-induced acute liver failure.
[5]
[6]
[7]
[8]
[9]
Acknowledgements
[10]
This work was supported by a grant of the Korea Healthcare Technology R&D Project, Ministry of Health & Welfare, Republic of Korea (HI14C0133) and by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No.2017R1A2B4003179).
[11]
[12]
[13]
Declarations of interest [14]
None. [15]
References [16]
[1] D. Larrey, G.-P. Pageaux, Drug-induced acute liver failure, Eur. J. Gastroenterol. Hepatol. 17 (2) (2005) 141–143. [2] A.D. Manthripragada, E.H. Zhou, D.S. Budnitz, M.C. Lovegrove, M.E. Willy, Characterization of acetaminophen overdose-related emergency department visits and hospitalizations in the United States, Pharmacoepidemiol. Drug. Saf. 20 (2011) 819–826. [3] D.C. Dahlin, G.T. Miwa, A.Y.H. Lu, S.D. Nelson, N-acetyl-p-benzoquinone imine: a cytochrome P-450-mediated oxidation product of acetaminophen, Proc. Natl. Acad. Sci. 81 (1984) 1327–1331. [4] N. Hanawa, M. Shinohara, B. Saberi, W.A. Gaarde, D. Han, N. Kaplowitz, Role of
[17]
[18]
[19]
54
JNK translocation to mitochondria leading to inhibition of mitochondria bioenergetics in acetaminophen-induced liver injury, J. Biol. Chem. 283 (20) (2008) 13565–13577. J.A. Hinson, A.B. Reid, S.S. McCullough, L.P. James, Acetaminophen-induced hepatotoxicity: role of metabolic activation, reactive oxygen/nitrogen species, and mitochondrial permeability transition, Drug. Metab. Rev. 36 (2004) 805–822. C.G. Antoniades, A. Quaglia, L.S. Taams, R.R. Mitry, M. Hussain, R. Abeles, et al., Source and characterization of hepatic macrophages in acetaminophen-induced acute liver failure in humans, Hepatology 56 (2012) 735–746. M.E. Blazka, M.R. Elwell, S.D. Holladay, R.E. Wilson, M.I. Luster, Histopathology of acetaminophen-induced liver changes: role of interleukin 1 alpha and tumor necrosis factor alpha, Toxicol. Pathol. 24 (2) (1996) 181–189. M.P. Holt, L.L. Cheng, C. Ju, Identification and characterization of infiltrating macrophages in acetaminophen-induced liver injury, J. Leukocyte Biol. 84 (2008) 1–12. D.M. Dambach, L.M. Watson, K.R. Gray, S.K. Durham, D.L. Laskin, Role of CCR2 in macrophage migration into the liver during acetaminophen-induced hepatotoxicity in the mouse, Hepatology 35 (5) (2002) 1093–1103. J.C. Mossanen, O. Krenkel, C. Ergen, O. Govaere, A. Liepelt, T. Puengel, et al., Chemokine (C-C motif) receptor 2-positive monocytes aggravate the early phase of acetaminophen-induced acute liver injury, Hepatology 64 (5) (2016) 1667–1682. Q. You, M. Holt, H. Yin, G. Li, C.J. Hu, C. Ju, Role of hepatic resident and infiltrating macrophages in liver repair after acute injury, Biochem. Pharmacol. 82 (2013) 836–843. P. Chartterjee, S. Seak, S. Mukherjee, R. Kundu, S. Mukherjee, S. Ray, et al., Adipocyte fetuin-A contributes to macrophage migration into adipose tissue and polarization of macropahges, J. Biol. Chem. 288 (39) (2013) 28324–28330. B. Denecke, S. Graber, C. Schafer, A. Heiss, W. Jahnen-Dechent, Tissue distribution and activity testing suggest a similar but not identical function of fetuin-B and fetuin-A, Biochem. J. 376 (2003) 135–145. P. Auberger, L. Falquerho, J.O. Contreres, G. Pages, G. Le Cam, B. Rossi, et al., Characterization of a natural inhibitor of the insulin receptor tyrosine kinase: cDNA cloning, purification, and anti-mitogenic activity, Cell 58 (1989) 631–640. J.H. Ix, M.G. Shlipak, V.M. Brandenburg, S. Ali, M. Ketteler, M.A. Whooley, Association between human fetuin-A and the metabolic syndrome: data from the Heart and Soul Study, Circulation 113 (2006) 1760–1767. S.T. Mathews, G.P. Singh, M. Ranalletta, V.J. Cintron, X. Qiang, A.S. Goustin, et al., Improved insulin sensitivity and resistance to weight gain in mice null for the Ahsg gene, Diabetes 51 (2002) 2450–2458. A.M. Hennige, H. Staiger, C. Wicke, F. Machicao, A. Fritsche, H.-U. Haring, et al., Fetuin-A induces cytokine expression and suppresses adiponectin production, PLoS ONE 3 (3) (2008) e1765. C. Naito, M. Hashimoto, K. Watanabe, R. Shirai, Y. Takahashi, M. Kojima, et al., Facilitatory effects of fetuin-A on atherosclerosis, Atherosclerosis 246 (2016) 344–351. D. Pal, S. Dasgupta, R. Kundu, S. Maitr, G. Das, S. Mukhopadhyay, et al., Fetuin-A acts as an endogenous ligand of TLR4 to promote lipid-induced insulin resistance,
Biochemical Pharmacology 166 (2019) 46–55
K.-Y. Lee, et al.
[30] C. Zhang, J. Feng, Z. Zhuo, S. Yang, W. Zhang, W. Wang, et al., Macrophage-derived IL-1a promotes sterile inflammation in a mouse model of acetaminophen hepatotoxicity, Cell Mol. Immunol. 14 (2017) 1–10. [31] Z.-X. Liu, B. Gunawan, N. Kaplowitz, Neutrophil depletion protects against murine acetaminophen hepatotoxicity, Hepatology 43 (6) (2006) 1220–1230. [32] C. Cover, J. Liu, A. Farhood, E. Malle, M.P. Waalkers, M.L. Bajt, Pathophysiological role of the acute inflammatory response during acetaminophen hepatotoxicity, Toxicol. Appl. Pharmacol. 216 (2006) 98–107. [33] H. Jaeschke, J. Liu, Neutrophil depletion protects against murine acetaminophen hepatotoxicity: another perspective, Hepatology 45 (6) (2007) 1588–1589. [34] L.P. James, S.S. McCullough, T.R. Knight, H. Jaeschke, J.A. Hinson, Acetaminophen toxicity in mice lacking NADPH oxidase activity: role of peroxynitrite formation and mitochondrial oxidant stress, Free Radical. Res. 37 (12) (2003) 1289–1297. [35] C.D. Williams, M.L. Bajt, A. Farhood, H. Jaeschke, Acetaminophen-induced hepatic neutrophil accumulation and inflammatory liver injury in CD18-deficient mice, Liver Int. 30 (9) (2010) 1280–1292. [36] C.D. Williams, M.L. Bajt, M.R. Sharpe, M.R. McGill, A. Farhood, H. Jaeschke, Neutrophil activation during acetaminophen hepatotoxicity and repair in mice and humans, Toxicol. Appl. Pharmacol. 275 (2) (2014) 122–133. [37] E. Zigmond, S. Samia-Grinberg, M. Pasmanik-Chor, O. Shbolet, Z. Halpern, C. Varol, Infiltrating monocyte-derived macrophages and resident kupffer cells disply different ontogeny and functions in acute liver injury, J. Immunol. 193 (2014) 344–353. [38] M. Demetriou, C. Binkert, B. Sukhu, H.C. Tenenbaum, J.W. Dennis, Fetuin/a2-HS glycoprotein is a transforming growth factor-b type II receptor mimic and cytokine antagonist, J. Biol. Chem. 271 (22) (1996) 12755–12761. [39] W. Li, S. Zhu, J. Li, Y. Huang, R. Zhou, X. Fan, et al., A hepatic protein, fetuin-A, occupies a protective role in lethal systemic inflammation, PLoS ONE 6 (2) (2011) e16945. [40] K.M. Dziegielewska, N.A. Andersen, N.R. Saunder, Modification of macrophage response to lipopolysaccharide by fetuin, Immunol. Lett. 60 (1) (1998) 31–35.
Nat. Med. 18 (2012) 1279–1285. [20] Y.J. Choi, K.Y. Lee, S.H. Jung, H.S. Kim, G. Shim, M.G. Kim, et al., Activation of AMPK by berberine induces hepatic lipid accumulation by upregulation of fatty acid translocase CD36 in mice, Toxicol. Appl. Pharmacol. 316 (2017) 74–82. [21] J. Weischenfeldt, B. Porse, Bone marrow-derived macrophages (BMM): isolation and applications, CSH Protoc. 3 (12) (2008) 1–8. [22] B.V. Martin-Murphy, M.P. Holt, C. Ju, The role of damage associated molecular patter molecules in acetaminophen-induced liver injury in mice, Toxicol. Lett. 192 (2010) 387–394. [23] H. Jaeschke, M.R. McGill, A. Ramachandran, Oxidant stress, mitochondria, and cell death mechanisms in drug-induced liver injury: lessons learned from acetaminophen hepatotoxicity, Drug Metab. Rev. 44 (1) (2012) 88–106. [24] J. Xu, X. Zhang, M. Monestier, N.L. Esmon, C. Esmon, Extracellular histones are mediators of death through TLR2 and TLR4 in mouse fatal liver injury, J. Immunol. 202 (1) (2011) 1–6. [25] K.A. Cavassani, A.P. Moreira, D. Habiel, T. Ito, A.L. Coelho, R.M. Allen, et al., Toll like receptor 3 plays a critical role in the progression and severity of acetaminophen-induced hepatotoxicity, PLoS ONE 8 (6) (2013) e65899. [26] A.B. Imaeda, A. Watanabe, M.A. Sohail, S. Mahmood|, M. Mohamadnejad, F.S. Sutterwala, et al., Acetaminophen-induced hepatotoxicity in mice is dependent on Tlr9 and the Nalp3 inflammasome, J. Clin. Invest. 119 (2) (2009) 305–314. [27] X. Wang, R. Sun, H. Wei, Z. Tian, High-mobility group box 1 (HMGB1)-Toll-like receptor (TLR)4-interleukin (IL)-23-IL-17A axis in drug-induced damage-associated lethal hepatitis: Interaction of gammadelta T cells with macrophages, Hepatology 57 (2013) 373–384. [28] J.E. Fisher, T.J. McKenzie, J.B. Lillegard, Y. Yu, J.E. Juskewitch, G.I. Nedredal, et al., Role of Kupffer cells and toll-like receptor 4 in acetaminophen-induced acute liver failure, J. Surg. Res. 180 (2013) 147–155. [29] N. Shah, M. Montes de Oca, M. Jover-Cobos, K. Tanamoto, M. Muroi, K. Sugiyama, et al., Role of Toll-like receptor 4 in mediating multiorgan dysfunction in mice with acetaminophen induced acute liver failure, Liver Transpl. 19 (2013) 751–761.
55