Bracteanolide A abrogates oxidative stress‐induced cellular damage and protects against hepatic ischemia and reperfusion injury in rats

Abstract Liver diseases, including viral hepatitis, liver cirrhosis, and liver cancer, mostly remain silent until the late stages and pose a continuing threat to millions of people worldwide. Liver transplantation is the most appropriate solution in the case of liver failure, but it is associated with hepatic ischemia and reperfusion (I/R) injury which severely reduces the prognosis of the patients. In order to ameliorate I/R injury, we investigated the potential of bracteanolide A, from the herb Tradescantia albiflora Kunth in protecting the liver from I/R injury. We first determined the protective effect of bracteanolide A against oxidative stress and DNA damage using HepG2 hepatocyte cell line and then assessed the levels of inflammatory cytokines and antioxidant proteins in response to hepatic insult using an animal model of hepatic I/R injury. The results showed bracteanolide A greatly enhanced cell survival and decreased reactive oxygen species (ROS) production under H2O2 induction. It also upregulated the expression of nuclear factor (erythroid‐derived 2)‐like2 (Nrf2) and its downstream cytoprotective proteins NAD(P)H quinone oxidoreductase 1 (NQO1) and heme oxygenase‐1 (HO‐1). Bracteanolide A effectively reduced the severity of liver lesions in I/R‐injured rats revealed by histological analysis and significantly decreased the levels of alanine transaminase (ALT), aspartate transaminase (AST), cyclooxygenase‐2, and inflammatory cytokines interleukin (IL)‐1β and tumor necrosis factor (TNF)‐α. Bracteanolide A preconditioning effectively protected the liver from I/R damage in the animal model, and this easily applied procedure may provide a new means to ameliorate hepatic I/R injury during liver surgeries.


| INTRODUC TI ON
Chronic liver disease affects millions of people and its complications, including cirrhosis and hepatocellular carcinoma, account for 3.5% of all deaths worldwide (Asrani et al., 2019;Moon et al., 2019).
Apart from significant advances in vaccination and treatment of viral hepatitis, incidence of metabolic liver diseases, on the other hand, is on the rise, and this is associated with increasing alcohol and/or drug abuse and compromised health conditions such as diabetes and obesity (Xiao et al., 2019;Younossi et al., 2020). The end stage of chronic liver diseases calls for liver transplantation as a way to extend the patient's life. Liver ischemia and reperfusion injury (I/R injury) is a pathophysiological condition that occurs during and after liver resection, liver transplantation, and other surgical procedures involving the liver (Cannistrà et al., 2016). I/R injury is responsible for significant morbidity and mortality resulting from liver graft failure and multiple organ dysfunctions (Cannistrà et al., 2016). The two stages of an I/R injury, ischemia and reperfusion, affect different cells and tissues. The main cells being affected during ischemia are sinusoidal endothelial cells, which result in altered microcirculation, causing hypoxia and subsequent impaired ATP synthesis. When blood and oxygen supply is re-established during reperfusion, the sequential activation of CD4 + T lymphocytes, Kupffer cells, and neutrophils releases reactive oxygen species (ROS) and inflammatory cytokines such as tumor necrosis factor-alpha (TNFα), interleukin-1 (IL-1), and platelet-activating factor (Konishi & Lentsch, 2017).
Inducible nitric oxide synthase (iNOS) is induced by inflammatory mediators and leads to the production of nitric oxide (NO), an initial form of reactive nitrogen species (RNS) which signals to nuclear factor (erythroid-derived 2)-like2 (Nrf2)/antioxidant response element (ARE) signaling, a major cellular antioxidant-detoxification mechanism in response to pathological stress (Ma, 2013;Molaei et al., 2021). NO exerts dual roles depending on dose and duration of exposure. In physiological levels, it attenuates liver I/R damage through regulating levels of ATP, cytokines, and antioxidants, but in large amounts it causes tissue damage (Guan et al., 2014).
Following reperfusion, hydrogen peroxide and superoxide radicals generated through nicotinamide adenine dinucleotide phosphate (NADPH) oxidase and the xanthine-oxidase system give rise to oxidative stress, leading to lipid peroxidation, DNA damage, and cell death (Ildefonso & Arias-Díaz, 2010). In response to oxidative stress, Nrf2 is released from ubiquitin-proteasomal degradation and translocates to the nucleus where it binds to antioxidant response element, inducing the expression of cytoprotective genes, such as heme oxygenase-1 (HO-1), superoxide dismutase 1 (SOD1), NAD(P)H quinone oxidoreductase 1 (NQO1), and catalase (Wasik et al., 2017), which counterbalance oxidative stress through glutathione (GSH) regulation (Liu et al., 2017). Overexpression of HO-1 in macrophages suppressed the production of inflammatory cytokines and preserved mouse and rat liver histology from I/R insult (Ke et al., 2010;Shen et al., 2011). Upregulation of NQO1 is found in various liver pathologies, and this may represent an adaptive stress response to halt further disease progression by detoxifying reactive species (Aleksunes et al., 2006;Cheng et al., 2015).
Both Nrf2 and HO-1 have been identified as promising therapeutic targets for the management of hepatic I/R injury Ma, 2013;Richards et al., 2010).
Intermittent inflow occlusion and ischemic preconditioning, as well as the administration of pharmaceutical agents such as antioxidants, are current strategies used to overcome hepatic ischemia and reperfusion injury and irreversible damage (Saidi & Kenari, 2014).
Given the increasing prevalence of liver illness on a global scale that ultimately requires liver transplantation or other surgical operations, which is largely susceptible to hepatic ischemia and reperfusion injury, we choose to investigate new compounds with the potential to reduce I/R damage to the liver. Bracteanolide A, whose structure is shown in Figure 1, is the major active compound of Tradescantia albiflora Kunth, a herb used to treat gout, edema, nephritis, enteritis, and diarrhea (Wang et al., 2016). However, evidence-based medical efficacy of bracteanolide A is limited. Bracteanolide A is reported to be one of the main compounds in the methanol extract of Murdannia bracteata (MB-m), which belongs to the same family as T. albiflora, and exhibits potent inhibitory effect on iNOS (Wang et al., 2007), which suggests its potential in the prevention and treatment of diseases involving an increased expression of iNOS. In addition, MB-m possesses antioxidant and free radical scavenging activities, contributing to its protective effect against CCl 4 -induced hepatic damage (Yam et al., 2010).
According to the abovementioned, bracteanolide A seems to be a suitable candidate as a hepatoprotective agent. Thus, we aim to investigate the potential of bracteanolide A to protect against liver damage using HepG2 cells and a rat model of hepatic ischemia and reperfusion injury.

| Cell culture
All materials and reagents for cell culture were obtained from Gibco, Thermo Fisher Scientific (Walham, MA, USA) unless otherwise the animal model, and this easily applied procedure may provide a new means to ameliorate hepatic I/R injury during liver surgeries.

| Bracteanolide A toxicity and cell viability
The cytoxicity of bracteanolide A alone on HepG2 cells was examined using the WST-1 assay according to manufacturer's protocol.
HepG2 cells were seeded at 2 × 10 5 per well in a 96-well culture plate. Following incubation of bracteanolide A at 0, 5, 10, 20, 50, 100, or 200 μM for 24, 48, and 72 hr, spent medium was aspirated and WST-1 reagent diluted 50x in cell culture medium was added to cell and incubated for 2.5 hr. Then, the culture medium was collected and absorbance OD 450-620 was measured. Modified methylene blue assay was used to evaluate the viability of cells exposed to oxidative stress with or without bracteanolide A treatment. The method was described in Felice et al. (2009). Briefly, HepG2 cells were seeded at 8 × 10 4 cells per well in a 24-well plate and allowed to attach for 24 hr.
For coincubation, cells were treated with vehicle or 10, 20, or 50 μM bracteanolide A together with 750 μM H 2 O 2 for 24 hr. For preincubation, cells were first incubated with bracteanolide A at the above same concentrations for 24 hr prior to the cotreatment with H 2 O 2 .
Then, media were removed and each well was rinsed with PBS. Two hundred and fifty microliters (250 μl) of methylene blue solution (1X PBS containing 1.25% glutaraldehyde and 0.6% methylene blue) was added to each well of the culture plate. After incubation at 37℃ for 60 min, the methylene blue solution was removed from the wells, and then the plates were rinsed by gentle submersion in distilled water for six times. The excess water was allowed to drain and the plate was briefly air-dried. Four hundred microliters (400 μl) of elution solution (consisted of 1:1 ratio of ethanol:PBS containing 1% acetic acid) was added to each well. After 15 min on a plate rotator at room temperature, the elution solutions were centrifuged at 12,000 × g for 3 min.
The solutions were transferred to a 96-well plate in duplicates, 175 μL into each well. Plates were read at 620 nm using a microplate reader (Thermo Fisher Scientific Instruments Co. Ltd, USA).

| Intracellular ROS assay
The generation of intracellular ROS was determined as described in Wang and Joseph (1999). Briefly, HepG2 cells were seeded at a density of 8 × 10 5 per well in a 6-well plate and incubated for 24 hr at 37℃.
After different treatments for the indicated durations, the cells were added with 100 μM dichlorodihydrofluorescein diacetate (DCFH-DA) (from Sigma-Aldrich, St. Louis, MO, USA) dissolved in medium and incubated at 37℃ for 30 min. After the excessive DCFH-DA was removed by several rinses with 1X PBS, the cells were exposed to 1 mM H 2 O 2 for 1 hr to trigger ROS production. After lysing the cells, aliquots of the cell lysates (150 μl) were transferred to a black 96-well fluorometric plate in duplicates. The fluorescence was measured using Thermo Scientific Fluoroskan Ascent fluorescence spectrometer (Thermo, USA). The excitation filter was set at 480 nm and the emission filter was set at 538 nm.

| Animals and I/R model
Eighteen healthy male Sprague Dawley rats (250-300 g) were obtained from the National Laboratory Animal Center, Taiwan, Republic of China. Animals were housed two in a cage of polycarbonate material and provided with corn cobs bedding, given a standard diet and clean water ad libitum, and maintained at 23 ± 2℃, 60 ± 5% RH with adequate ventilation on a 12-hr light/dark cycle. The experimental procedure was approved by the Institutional Animal Care and Use Committee of National Taiwan University (approval ID: NTU-103-EL-91). After one week of acclimatization, rats were randomly divided into three groups: Sham, I/R, and I/R+Bracteanolide A groups. The I/R injury rat model was generated as described in Yamauchi et al. (Yamauchi et al., 1982). Briefly, each rat was anesthetized with 1%-2% isoflurane in pure oxygen. The liver was exposed through an upper midline incision, and two pieces of fine silk thread were looped along the right and left branches of the portal vein, hepatic artery, and bile duct to block blood supply to the median and left lobes. To elicit I/R injury, ischemia of the median/left lobes was maintained for 60 min after which threads were released and blood was resupplied.
In the sham-operated group (Sham), rats were given anesthesia and subjected to laparotomy as well as exposure of the portal triad without hepatic ischemia (n = 6). In the I/R group, rats were subjected to ischemia and reperfusion as described above (n = 6). In the bracteanolide A-pretreated group (I/R+Bracteanolide A), the rats received intraperitoneal injection of bracteanolide A (2.5 mg/kg.bw), dose determined from preliminary experiment, at 30 min prior to ischemia (n = 6). Bracteanolide A was diluted to 0.625 mg/ml in 1% carboxymethylcellulose (Sigma-Aldrich). The rats in Sham and I/R groups received an equivalent volume of vehicle. After ischemia, animals were sutured, wounds swabbed with iodine solution, and given baytril (10 mg/kg BW, s.c. injection), and then returned to cages provided with lamps to avoid hypothermia. Inhalation anesthesia was chosen for ease of application and monitoring of animal physiology, as well as uninterrupted duration of sufficient anesthesia. Blood pressure was monitored by tail vein sphygmomanometer throughout the operation. Signs of distress such as lethargy or reduced food intake and grooming were not observed during recovery. The rats were sacrificed 24 hr after reperfusion by anesthesia followed by cardiac puncture; blood as well as liver tissue samples were harvested for analysis.

| Histological analysis
The left lobes of livers were harvested and trimmed to suitable sizes before being fixed in a 10% (v/v) formalin solution and then processed by standard histological routines through dehydration, clearing, and wax embedding to paraffin sections. Tissue sections (4-5 μm) were stained with hematoxylin and eosin and examined under a light microscope. Sections were scored by a pathologist from 0 to 4 for degrees of sinusoidal congestion, vacuolization of hepatocyte cytoplasm, and necrosis as described in Suzuki et al. (1991). Criteria are shown in Table 1.

| Biochemical analyses
2.6.1 | Serum ALT and AST levels Blood samples were allowed to clot at room temperature for 30 min, followed by centrifugation at 3,000 × g for 5 min. The supernatant was collected and designated as serum. Serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels were measured by SPOTCHEM SP-4410 clinical chemistry analyzer (Arkray Co, Kyoto, Japan).

| Gene expression analysis
2.7.1 | RNA extraction and reverse transcription polymerase chain reaction (RT-PCR) Total RNA in liver tissue/cell pellet was isolated using TriPure Isolation Reagent (Roche Applied Science, Basel, Switzerland).

| Quantitative PCR (q-PCR)
Quantitative PCR analyses were performed on aliquots of cDNA to detect iNOS, COX-2, HO-1, NQO1, Nrf2, and 18S (as an internal standard) gene expression using a thermal cycler (Applied Biosystems, Foster City, CA, USA). Reactions were carried out in a volume of 10 μL containing: 5 μL 2× KAPA SYBR®, 0.2 μL each of 10 μM forward and reverse primers, 2.6 μL PCR water, and 2 μL cDNA. After an initial enzyme activation for 3 min at 95℃, 40 amplification cycles were performed (3 s for 95℃ denaturation, 20 s for 60℃ annealing and extension). The primers used in this study were purchased from Sigma-Aldrich (USA) and are listed in Table 2.

| Statistical analysis
Results were expressed as mean ±standard error of the mean (SEM). Comparisons in different groups were made using one-way or two-way analysis of variance (ANOVA) where appropriate followed by post hoc tests. Statistical significance was defined as p < .05. TA B L E 1 Suzuki scores for the assessment of liver damage following I/R injury

| Bracteanolide A protects HepG2 cells against H 2 O 2 -induced oxidative stress
We first examined the cytoxicity of bracteanolide A on HepG2 cells (from the hepatocyte lineage) using WST-1 assay. The result shows that bracteanolide A does not induce cytotoxicity at all the concentrations treated from low to high doses ( Figure 2a) Figure 2c).

| Bracteanolide A induces nuclear translocation and the protein expression but not mRNA levels of Nrf2
To investigate the underlying reason for bracteanolide A's protection against H 2 O 2 -induced oxidative stress and ROS production, we looked into changes of Nrf2, a key regulator in the cellular antioxidant response. First, we measured time-course Nrf2 mRNA expression using tert-butylhydroquinone (t-BHQ), a common antioxidant food additive in fats and oils, as a positive control for inducing Nrf2. We found 17 hr treatment can induce a significant increase compared to shorter or longer durations; thus, 17 hr was used as the treatment duration for mRNA expression in subsequent experiments involving NQO1 and HO-1. Then, we treated cells with increasing concentrations of bracteanolide A from 0, 10, 20, 50, 100, and 200 μM and found out there was no significant difference among the various doses (Figure 3a), suggesting bracteanolide A did not regulate Nrf2 at the gene level, which is not surprising for a transcription factor. Later, we went on to investigate whether bracteanolide A regulates Nrf2 at the protein level. We incubated cells with 20 μM bracteanolide A for different durations to find out the most suitable time for Nrf2 induction, which was 3 hr as shown in Figure 3b. Thus, cells were then incubated for 3 hr in different concentrations of bracteanolide A. The results (Figure 3b-d) showed Nrf2 protein induction was most evident at 50 μM bracteanolide A (Figure 3c), which induced a 5.03-fold increase in Nrf2 total protein and nuclear protein accumulation, indicating Nrf2 activation. Subsequently, 50 μM bracteanolide A was the dose used to investigate protein expression of Nrf2 downstream targets NQO1 and HO-1.

| Bracteanolide A protects HepG2 cells from oxidative stress-induced DNA damage
Phosphorylation of H2AX is an early biomarker for DNA doublestrand breaks (Kuo & Yang, 2008). Apart from protecting cells from oxidative stress-induced cell death by abrogating ROS production and stimulating the levels of cytoprotective proteins,
Bracteanolide A preincubation for 16 hr, however, reduced the expression of p-H2AX, with a concomitant decrease in NQO1 and HO-1 levels.

| Bracteanolide A ameliorates I/R-induced hepatic damage
In order to further investigate the potential of bracteanolide A, we tested its effects on a rat model of hepatic ischemia and reperfusion injury, which is associated with oxidative and inflammatory insult, causing damage to the liver. We successfully reproduced distinct morphological and structural alterations to the liver by occlusion of blood supply to the median/left lobes for 1 hr followed by 24 hr of reperfusion as a model of I/R injury. Liver histological sections show normal liver parenchyma (Figure 6a i,iv,vii) and a large area of coagulative necrosis and periportal neutrophil infiltration in I/R-injured liver (Figure 6a ii,v,vii). Bracteanolide A pretreatment at 30 min before I/R greatly reduced cellular necrosis, inflammatory infiltration as well as significantly decreased the Suzuki score (grading scheme shown in Table 1) compared to I/R rats (Figure 6a iii, vi, ix and 6b). Moreover, bracteanolide A also significantly reduced the plasma levels of ALT and AST, markers of liver damage, which soared in I/R-injured rats ( Figure 7).

| Bracteanolide A restores I/R-induced alterations in the mRNA levels of inflammatory and cytoprotective enzymes
It was clear that bracteanolide A pretreatment significantly ameliorated I/R injury as shown by improvement in morphology and structure of the liver and significant reduction in ALT and AST levels.

F I G U R E 2 Effects of bracteanolide A on cell viability and reactive oxygen species production under H2O2 induction. (a) Cells
were treated with bracteanolide A at 5, 10, 20, 50, 100, and 200 μM for 24, 48, or 72 hr and then harvested to test for cytotoxicity of bracteanolide A on the cells using WST-1 assay. (b) Cells were treated with 750 μM H2O2 and bracteanolide A at 0, 10, 20, and 50 μΜ for 24 hr, with or without pretreatment of bracteanolide A at the same concentrations for 24 hr. Modified methylene assay was then performed to estimate cell viability. The relative percentage of viability to control cells (set as 100%) is shown as mean±SEM from at least two independent experiments. (c) Cells were treated with 1 mM H2O2 and bracteanolide A at the concentrations indicated for 1 hr, without or without pretreatment of bracteanolide A at the same concentrations for 24 hr. The ROS concentration was measured by fluorescence at ex 480/em 538 nm. Differences among various concentrations within each treatment duration (a) or the two incubation groups (b and c) were analyzed by two-way ANOVA followed by Tukey's multiple comparisons test. Different letters represent statistical significance at p <.05 We went on to look at the levels of inflammatory indices iNOS and COX-2 as well as Nrf2-mediated cytoprotective enzymes NQO1 and HO-1. Hepatic I/R significantly increased the mRNA levels of iNOS and COX-2, as well as those of HO-1 and NQO1. Administration of bracteanolide A significantly suppressed COX-2 and HO-1 levels, but not iNOS and NQO1 (Figure 8b,c).

| Bracteanolide A decreases serum levels of inflammatory cytokines
Serum inflammatory cytokines were significantly increased upon I/R insult, as shown in Figure 9. However, consistent with the reduction in COX-2 mRNA levels, bracteanolide A administration effectively reduced the levels of the inflammatory cytokines, even to levels comparable to Sham rats.

| D ISCUSS I ON
Plant extracts contain phytochemicals which are active in preventing various diseases and may be beneficial to health with moderate consumption. Many studies have explored the potential of plant extracts against liver injuries (Park et al., 2020;Wang et al., 2019;Zhu et al., 2019). The pathology of I/R-induced hepatic injury reveals that oxidative stress plays a crucial role in the scenario. The Keap1/Nrf2 pathway is an essential cellular defense mechanism against reactive oxygen species (Baird & Dinkova-Kostova, 2011;Kobayashi & Yamamoto, 2006), making it rational to assess Nrf2 expression levels in our study. To find a preconditioning candidate that can activate Nrf2, we first screened compounds with antioxidant potential through a luciferase-based platform (data not shown).
Bracteanolide A came out to be the most promising agent among the candidates investigated; this result was confirmed by bracteanolide A's potent protection against H 2 O 2 -induced cell death in this study (Figure 2b). Furthermore, bracteanolide A was revealed to be an effective ROS scavenger, significantly inhibiting the production of intracellular ROS with pretreatment ( Figure 2c). Usually, transcription factors were regulated at the post-translational level (Filtz et al., 2014), which is consistent with our finding that without influencing the mRNA level of Nrf2, bracteanolide A increased

F I G U R E 5 Effect of bracteanolide
A on p-H2AX protein expression (DNA damage). Cells were pretreated with 50 μM bracteanolide A for 4 or 16 hr, followed by 750 μM H2O2 for 2 hr. The expression pattern of phosphorylated H2AX (p-H2AX) along with NQO1 and HO-1 was analyzed by Western blot with p-H2AX, NQO1, and HO-1 antibodies, respectively. β-actin was used as a loading control. Representative data are shown from at least two independent experiments as well as NO production through protein kinase B (Akt), NF-κB, and iNOS pathways. The upregulation of NF-κB increases TNFα and IL-6; TNFα in turn causes the overproduction of chemokines and ROS, leading to activation of T lymphocytes and neutrophil accumulation that aggravate liver injury (Cannistrà et al., 2016).
In this study, bracteanolide A triggered a significant reduction of I/R-induced increase in COX-2 mRNA levels ( Figure 8b) and a concurrent decline in IL-1β and TNFα levels (Figure 9a.b), showing bracteanolide A effectively attenuated the propagation of I/R injury. IL-6 and IL-10, however, have been reported as protective F I G U R E 6 Effects of bracteanolide A on rat liver pathological lesions and Suzuki score. (a) Histological analysis of livers in sham rats (i, iv, vii), I/R rats (ii, v, viii), and bracteanolide A-treated I/R rats (iii, vi, ix Figure 5).

F I G U R E 9
Effects of bracteanolide A on cytokine levels in the serum of I/R-injured rats. Effects of ischemia/ reperfusion and treatment with bracteanolide A on (a) interleukin 1β (IL-1β), (b) interleukin 6 (IL-6), (c) interleukin 10 (IL-10), (d) tumor necrosis factorα (TNFα) levels in rat serum. Data are shown as mean±SEM. N = 6. Procedures are described in Materials and Methods. Different letters indicate statistical significance at p < .05 inflammatory-mediated damage is diminished and hepatic I/R injury in rats is effectively minimized, as shown by the restoration of normal structure and function of the liver. The results of this study provided evidence to bracteanolide A's protective effect on I/R-induced liver damage, and preconditioning with bracteanolide A may be a novel means to reduce the severity of I/R injury to hepatic tissues and to improve the clinical outcome of patients going through surgical operations on the liver.

ACK N OWLED G EM ENTS
This work was financially supported by grants MOST

CO N FLI C T O F I NTE R E S T
The authors declare that they do not have any conflict of interest.

E TH I C A L S TATEM ENT
All procedures regarding animal study and design were approved by the Institutional Animal Care and Use Committee of National Taiwan University (IACUC No. NTU-98-EL-109) and carried out according to national and university guidelines.

I N FO R M E D CO N S E NT
This study does not involve any human subjects.