Platanosides, a Potential Botanical Drug Combination, Decrease Liver Injury Caused by Acetaminophen Overdose in Mice

Abstract

Oxidative stress plays an important role in acetaminophen (APAP) -induced hepatotoxicity. Platanosides (PTSs) isolated from the American sycamore tree (Platanus occidentalis) represent a potential new 4-molecule botanical-drug class of antibiotics active against drug-resistant infectious disease. Preliminary studies have suggested that PTS is safe and well tolerated, had antioxidant properties. The potential utility of PTS in decreasing APAP hepatotoxicity in mice in addition to an assessment of its potential with APAP for the control of infectious diseases along with pain and pyrexia associated with a bacterial infection were investigated. On PTS treatment in mice, serum alanine amino transferase (ALT) release, hepatic centrilobular necrosis and 4-hydroxynonenal (4-HNE) were markedly decreased. In addition, inducible nitric oxide synthase (iNOS) expression and c-Jun-N-terminal kinase (JNK) activation decreased when mice overdosed with APAP were treated with PTS. Computational studies suggested that PTS may act as a JNK-1/2 and Keap1-Nrf2 inhibitor and that the isomeric mixture could provide greater efficacy than the individual molecules. Overall, PTS represents a promising botanical drug for hepatoprotection and drug-resistant bacterial infections and is effective in protecting against APAP-related hepatotoxicity, which decreases liver necrosis and inflammation, iNOS expression, and oxidative and nitrative stress, possibly by preventing persistent JNK activation.

Graphical Abstract

Acetaminophen (APAP) is an effective analgesic and antipyretic that is utilized alone or in combination with other components of numerous prescription and over-the-counter medicines. APAP was approved for clinical use in the 1950s and currently is widely used throughout the United States and worldwide in both adults and children.^1–3 At doses greater than 4 g/day, APAP can cause hepatotoxicity, which its most common side effect and can lead to acute liver failure (ALF) in severe cases.^3–5 Since APAP is present in many over-the-counter combination drugs, accidental overdoses often occur due to the use of multiple non-prescription combined./formulations containing APAP, since many consumers do not read which of these formulations contain APAP or else are not knowledgeable of the dangers of overdosing. Moreover, APAP intoxication is sometimes caused by abuse or by suicidal attempts, due to its low cost and easy accessibility.⁶ APAP intoxication is the leading cause of drug-induced liver injury (DILI), and accounts for about half of overdose-related ALF cases in the United States, and is also the primary cause of acute liver failure worldwide.^3–6

Previous studies have shown that APAP hepatotoxicity is linked to the production of N-acetyl-p-benzoquinone imine (NAPQI) as an intermediate during APAP metabolism through reactions catalyzed by cytochrome P450s, primarily Cyp2E1. This pathway accounts for about 5-15% of drug metabolism whereas sulfation and glucuronidation account for 20–46% and 40–67% of APAP metabolism, respectively.^7–11 NAPQI is detoxified by conjugation with glutathione (GSH). In APAP overdose, production of NAPQI increases markedly, which depletes GSH.^7,8,12 As a result, NAPQI begins to form covalent bonds with biologically important molecules such as proteins, which leads to detrimental effects on cells.^8,9,13 Chronic alcohol use and poor nutrition, which induce Cyp2E1 and/or affect the detoxification process, increase the susceptibility to APAP-induced liver injury^1,3,5. Formation of reactive oxygen and nitrogen species (ROS and RNS), activation of JNK and binding of activated JNK to the mitochondrial protein Sab (SH3 domain-binding protein 5), mitochondrial dysfunction, and calpain activation, are also considered important mediators of APAP hepatotoxicity.^5,7,14–20 N-Acetylcysteine (NAC) is currently the treatment of choice for APAP-induced liver injury.^21,22 NAC is a cell membrane-permeable GSH precursor, which increases GSH formation in cells, thus enhancing the detoxification process of APAP.^21,22 In addition, a 2010 study showed that NAC can directly facilitate the reduction of NAPQI back to APAP.²³ Currently, NAC is the only FDA-approved drug as an antidote for APAP-induced hepatotoxicity. However, NAC has to be used early after APAP administration since its effectiveness decreases markedly at later periods after APAP overdose.²² Therefore, additional therapies for APAP hepatotoxicity as either a preventative agent or treatment are needed and may also help control other forms of liver injury.

In the present study, we explored the effects of the platanosides (PTSs) (Figure 1) against APAP-induced hepatotoxicity. The PTSs are isolated from the leaves of the American sycamore tree (Platanus occidentalis L.; Platanaceae), a common hardwood used as an ornamental and shade tree. In addition, about a dozen very closely related species in the family Platanaceae also contain PTSs.^24–26 The molecular structures of PTS consist of the components of kaempferol flavone rhamnose glycoside moeity with two para-coumaryl groups attached to the carbohydrate unit, and their metabolites are highly active against many drug-resistant bacteria, fungi and viruses.²⁷ PTS represent a botanical drug lead as the four isomers will undergo equilibration. The present study show PTS has an approximate 1.9:1.0:1.1:0.5 ratio (E,E-, Z,Z-, E,Z-, Z,E-PTS; Figure S17, Supporting Information). This isomeric mixture may perform better than an individual molecule based on the data shown in Table 1 and maybe an evolutionary advantage for the plant.²⁸ Botanical drugs are complex combinations of natural products that are now recognized and approved as a separate category by the FDA.²⁹ Therefore, PTS represents a potential botanical drug type of antibiotics active against drug-resistant infectious disease although this has not been approved by the U.S. FDA. The PTS flavone glycoside isomeric mixture with two para-coumaryl groups and multiple phenolic functionalities may scavenge reactive oxygen or nitrogen species.²⁵ In fact, such flavone glycosides have a long history of use as antioxidants and appear to be essentially nontoxic to mammalian cells.^25,30 Furthermore, previous studies on flavone compounds have shown good protection against APAP-induced hepatotoxicity (Table 1).^31,32 Since oxidative stress-triggered alterations of signaling pathways play an important role in the development of APAP-induced liver injury, in this study the hypothesis was tested that PTS protects against APAP hepatotoxicity and their mechanism of action was examined using a dual strategy of in vivo testing in mice and a computational approach. Their therapeutic effects were compared with that of NAC.

Figure 1.

A: Molecular ion networking cluster for PTS from P. occidentalis. B: Structure of E,E-, E,Z-, Z,E-, and Z,Z-PTS.

Table 1.

Protection of PTS and Kaempferol against APAP-induced Hepatotoxicity.

Markera	PTS	Kaempferol
Stress	TRX1 restored ALT, MPO, 4-HNE, iNOS, and 3-NT ↓	GSH, GPx, SOD, MnSOD, MDA, and catalase restored ALT, AST, γ–GT, ROS, TBARS, and 3-NT ↓
Metabolism	CYP2E1 no change	CYP2E1 ↓ UGT1A1 ↑
Inflammatory	IL-1β ↓	IL-1β, IL-6, and TNFα ↓
Apoptotic	n.t.	Bcl-2 ↑ Caspase-3 ↓
Others	JNK1/2 phosphorylation ↓ Sab-pJNK1/2 ↓	JNK and ERK phosphorylation ↓ SIRT1 ↑ PARP1 and RIP1 ↓ Acetylation of P53, NFKB, and FOXO1 ↓

^a3-NT: 3-nitrotyrosine; 4-HNE: 4-hydroxynonenal; ALT: alanine amino transferace; AST: aspartate amino transferace; CYP2E1: cytochrome P450 2E1; ERK: extracellular signal-regulated kinase; FOXO1: forkhead box protein O1; Gpx: glutathione peroxidase; GSH: glutathione; γ-GT: g-glutamyl transferase; IL-1β: interleukin 1β; IL-6: interleukin 6; iNOS: inducible nitric oxide synthase; JNK: c-Jun N-terminal kinase; MDA: malondialdehyde; MnSOD: manganese superoxide dismutase; MPO: myeloperoxidase; NF-κB: nuclear factor kappa-light-chain-enhancer of activated B cell; p53: tumor protein P53; PARP1: poly ADP-ribose polymerase-1; RIP1: receptorinteracting serine/threonine-protein kinase 1; ROS: reactive oxygen species; Sab: SH3 domain-binding protein 5; SIRT1: silent information regulator 1; SOD: superoxide dismutase; TBARS: thiobarbituric acid reactive substances; TNFα: tumour necrosis factor-α; TRX1: thioredoxin; UGT1A1: UDP glucuronosyltransferase family 1 member A1; n.t: not tested.

RESULTS AND DISCUSSION

Purification of the Platanosides (PTSs).

Molecular networking was utilized to visualize the molecules observed from the plant sample collected as familial groupings, as a result of which commonalities within the mass spectrometric (MS) fragmentation data were assessed via vector correlations and displayed as an MS/MS network. The visualization of metabolite networks by using Cytoscape enabled the direct observation of similarities as well as differences between two or more samples in which similar entities within the network were clustered together, while disparate or single entities were grouped separately.³³ The MS/MS data were collected from the injection of a methanol extract of P. occidentalis mature leaves and xylem. Vector similarity scores were calculated for every possible pair of spectra with a minimum of six matching fragment ions (i.e., peaks) with similarity determined using a modified cosine calculation that takes into account the relative intensities of the fragment ions as well as the precursor m/z differences between the paired spectra. The paired spectra are defined as correlated if both have high K value (Cohen’s kappa coefficient) and vector similarity score (represented as a cosine value) greater than the defined threshold. Cosine threshold values were defined as 0.7 whereby a cosine value of 1.0 indicates identical spectra. These data were then imported to Cytoscape. The organized landscape of the molecular networking produced by Cytoscape shows that the PTS as well as the analogues are localized in P. occidentalis mature leaves and xylem. In this case, P. occidentalis leaves were used to isolate PTS with the guidance of molecular ion networking (MoIN) for the assessment in vivo activity in mice for the possible control of APAP induced liver toxicity (Figure 1A).

PTS Decrease APAP Hepatotoxicity in vivo.

Previous studies have shown that APAP causes overt liver injury at 24 h.^18,19 Therefore, liver histology was examined in H&E-stained liver sections at 24 h after APAP or vehicle treatment. No necrosis was observed in livers from mice treated with vehicle or PTS alone (Figures 2A and B). After APAP treatment, widespread centrilobular necrosis occurred in about 46% of the liver tissue area (Figures 2C and E). PTS administered at the same time as APAP dosing (0 h, gavage) decreased necrosis to 23% (Figures 2D and E). ALT, an enzyme that localizes in hepatocytes, is released into the blood during liver injury. Serum ALT levels were 20-35 U/L in mice treated with vehicle or PTS alone (Figure S1, Supporting Information). At 24 h after APAP treatment, serum ALT increased to ~11,200 U/L, indicating severe liver injury. PTS treatment at 0 h markedly decreased serum ALT by ~75%.

Figure 2.

Simultaneous PTS treatment decreases APAP-induced liver necrosis. Mice were treated with APAP (300 mg/kg, i.p.). PTS (10 mg/kg, i.g.) or vehicle (Veh) were given immediately after APAP treatment. Livers were collected at 24 h after APAP treatment. A-D, representative images of H&E stained liver sections (n = 4/group). A, livers of mice treated with Veh; B, livers of mice treated with PTS; C, livers of mice treated with APAP and Veh; D, livers of mice treated with APAP and PTS. Arrows identify necrotic areas. Bar is 200 μm. E, Quantification of necrotic areas. Values are means ± SEM. a, p < 0.05 vs. Veh; b, p < 0.05 vs. PTS; c, p < 0.05 vs. APAP plus Veh.

APAP overdose causes severe liver injury and even ALF.^3–5 NAC, which increases GSH formation in cells thus enhancing the scavenge of NAPQI, has been used for therapy of APAP hepatotoxicity since the 1970s.^21,22 Up to the present, NAC remains the only FDA-approved antidote for clinical use in APAP overdose. NAC is effective when used in the early stage after intoxication but its efficacy markedly decreases when given at later stages.^21,22,34,35 Delayed NAC treatment after APAP intoxication is the main cause of ALF.^21,22 Moreover, standard NAC dosing may not be sufficient for patients who are exposed to very high APAP overdoses. Therefore, additional drug therapy is needed for APAP hepatotoxicity treatment. In the present study, it was shown that PTS markedly decreased APAP hepatotoxicity (decreasing necrosis, ALT release and inflammation, Figures 2, S1, and S2, Supporting Information). Importantly, it only requires a low dose (10 mg/kg) of PTS to achieve protection. Moreover, PTS is effective when administered at the same time or even after APAP administration (Figures 2 and 3, and S1–S3, Supporting Information). The current study demonstrated that PTS at 10 mg/kg can achieve a similar protection to NAC at 300 mg/kg when administered after APAP (Figures 3 and S3, Supporting Information). Therefore, the use of PTS may be a promising strategy for both prevention and therapy for APAP hepatotoxicity and could provide utility in a clinical setting for an antibiotic-antipyretic drug combination with reduced risks of liver injury over APAP alone.

Figure 3.

Delayed NAC and PTS treatments decrease APAP-induced liver necrosis. Mice were treated with APAP (300 mg/kg, i.p.). NAC (300 mg/kg, i.p.), PTS (10 mg/kg, i.p. or i.g.), or vehicle (Veh, i.p.) were given 2 h after APAP treatment. Livers were collected at 24 h after APAP treatment. A-D, representative images of H&E stained liver sections (n = 3-4/group). A, livers from mice treated with APAP and Veh (i.p.); B, livers from mice treated with APAP and NAC; C, livers from mice treated with APAP and PTS (i.p.). D, livers from mice treated with APAP and PTS (i.g.). Arrows identify necrotic areas. Bar is 200 μm. E, Quantification of necrotic areas. Values are means ± SEM. **, p < 0.01 vs. APAP plus Veh.

PTS Decreases Oxidative/Nitrative Stresses and Inflammation.

The formation of NAPQI, the toxic intermediate of APAP metabolism, is catalyzed primarily by CYP2E1.^7–9 Therefore, it was examined if PTS alters CYP2E1 expression. At 2 h after APAP treatment, the time when NAPQI formation peaks, CYP2E1 expression was not altered by either APAP or PTS treatment (Figure S4, Supporting Information). The adduct NAPQI forms with a protein sulfhydryl group is called an “APAP protein adduct” because the NAPQI is reduced in the binding process. Also examined was the formation of APAP protein adducts. APAP-protein adducts were undetectable in animals not receiving APAP. After APAP, NAPQI-protein adducts increased to 0.23 ng/mg protein. Surprisingly, PTS given at the same time as APAP dosing increased APAP-protein adducts to 0.48 ng/mg protein.

In this study, it was found that PTS markedly decreases APAP hepatotoxicity (Figures 2 and 3, and S1–S3, Supporting Information). However, protection by PTS is not due to a decrease in APAP metabolism. PTS increased the formation of NAPQI (Figure S4, Supporting Information). Therefore, PTS must affect the blocking of pathways downstream of NAPQI formation, such as ROS formation. ROS and RNS are shown to modulate many signaling processes that trigger or amplify cell death, inflammation and fibrosis.^36–39 Oxidative and nitrative stresses are also recognized as the key mediators in APAP hepatotoxicity.^14,40 Many different enzymes and organelles in cells can produce ROS and RNS in pathophysiological settings (e.g., NADPH oxidase, xanthine oxidase, cytochrome P450 and peroxisomes, and mitochondria). Currently, it is widely accepted that ROS and RNS are formed primarily in mitochondria during the process of APAP hepatotoxicity.^41,42

It was examined if PTS have antioxidant effects since oxidative stress plays an important role in APAP hepatotoxicity.¹⁴ 4-HNE is a reactive product of lipid peroxidation. Small amounts of 4-HNE protein adducts were observed in the liver of mice that received only the vehicle or PTS (Figure S5, Supporting Information). 4-HNE protein adducts increased to 280% of the control value at 24 h. With PTS treatment (0 h), 4-HNE adducts did not increase after APAP. The expression of TRX1, an antioxidative protein, increased by ~60% in the liver of mice treated with PTS (Figure S5, Supporting Information). After APAP dosing, TRX1 decreased to 47% of the control value. With PTS treatment (0 h), TRX1 remained above the control value. These data clearly demonstrated the antioxidant effects of PTS.

PTS may be able to scavenge ROS due to the presence of multiple phenolic functional groups. Moreover, an in silico binding experiment (see below) showed that PTS may interact with the Keap1-Nrf2 pathway. Nrf2 is a redox sensitive transcriptional factor that plays a central role in antioxidant response element (ARE)-mediated induction of numerous antioxidant proteins/enzymes (e.g., TRX, glutathione-S-transferase, glutamate cysteine ligase, SOD2, and HO-1).^43–45 Nrf2-deficient mice have a lower level of antioxidant proteins⁴³ whereas Nrf2 activators are protective against APAP toxicity.⁴⁶ Previous studies on compounds having similar flavone moieties as PTS showed protection against APAP-induced hepatotoxicity by increasing the Nrf2 activity.^47–49 TRX1, one of the Nrf2 target genes, was upregulated by PTS treatment (Figure S5, Supporting Information). Therefore, PTS may have dual effects: as a free radical scavenger that allows removing ROS rapidly and as an Nrf2 activator that increases the expression of antioxidant proteins/enzymes to ensure sustained antioxidative effects. Computational binding results (see below) have indicated that PTS may act as a Keap1 inhibitor disrupting the formation of the Keap1-Nrf2 complex and hence promoting the downstream activity of Nrf2. In addition, activation of Nrf2 inhibits NF-κB and inflammasome activation thus suppressing the formation of many proinflammatory cytokines such as IL-1β, IL-10 and TNFα.⁵⁰ Previous studies on flavone compounds, the core structure of PTS was able to function as a Keap 1 inhibitor.^51,52 Therefore, the Nrf2 pathway potentially may play a pivotal role in the regulation of inflammation. Indeed, PTS also markedly inhibited inflammation after APAP (Figure S2, Supporting Information). Since infiltrated leukocytes also produce ROS thus amplifying liver injury, inhibition of inflammation would breach this vicious cycle. Importantly, PTS can achieve protection when administered at the same time with APAP and even after APAP dosing.

Peroxynitrite, which is formed by the reaction of superoxide with nitric oxide, is considered a critical mediator in the progression of APAP-induced liver injury.¹⁵ It was further investigated if PTS alter the expression of iNOS, the enzyme that produces nitric oxide in many pathological situations. At 2 h after APAP when ROS formation had already increased, iNOS expression remained at similarly low levels in all groups (data not shown). However, at 24 h after APAP iNOS expression markedly increased and this effect was blunted by half with PTS treatment that was given simultaneously with APAP (Figure S6, Supporting Information). 3-NT protein adducts, an indicator of peroxynitrite formation,⁵³ existed at low levels in the livers of mice treated with vehicle or PTS (Figure S6, Supporting Information). APAP increased 3-NT protein adducts to 223% of the basal level and this effect was blocked by PTS treatment.

Peroxynitrite plays a critical role in APAP-induced mitochondrial dysfunction.^54,55 In the current study, it was demonstrated that PTS also decrease RNS after APAP overdose (Figure S6, Supporting Information). The phenolic functional group in PTS suggest that PTS may be able to scavenge peroxynitrite, but future studies are needed to determine whether PTS have direct effects on peroxynitrite. In previous studies, all three isoforms of NOS have been linked to NO production after APAP but the exact source of NO has been controversial.^55–57 nNOS is expressed primarily in nerve tissue but was also identified in hepatocytes.⁵⁸ nNOS inhibitor protected cultured hepatocytes from APAP-induced cell death and nNOS-deficiency in mice decreased APAP hepatotoxicity in vivo, suggesting that nNOS plays an important role.^56,59 Some studies have showed iNOS deficiency decreases APAP hepatotoxicity whereas others indicated that peroxynitrite formation and APAP hepatotoxicity are independent of iNOS.^56,60,61 Mice deficient in eNOS also showed a lower level of liver injury after APAP.⁶² In this study, an increased expression of iNOS was observed in vivo after APAP, which was diminished by PTS (Figure S6, Supporting Information). It is likely that the decrease of peroxynitrite by PTS after APAP is due, at least in part, to the inhibition of iNOS expression. Consistently, a computational study indicated that PTS binds reasonably well to iNOS (see below) and potentially can inhibit NO production and provide protection against RNS.

PTS Decreases JNK Activation and Binding to Sab.

Previous work showed that the binding of activated JNK to mitochondrial protein Sab plays an essential role in APAP hepatotoxicity.^16–18 At 24 h after APAP, the expression of JNK was similar in all groups (Figure S7, Supporting Information). In the liver of mice treated with vehicle or PTS, low levels of phospho-JNK were observed. In contrast, in mice treated with an overdose of APAP, both phospho-JNK1 and phospho-JNK2 increased markedly. The ratio of phospho-JNK1/2 to JNK1/2 increased ~5-fold, indicating JNK1/2 activation. PTS blocked JNK activation after APAP overdose.

The binding of phospho-JNK to Sab was detected by immunoprecipitation of Sab followed by immunoblotting of phospho-JNK1/2 and Sab, respectively. Phospho-JNK1/2 bound to Sab increased ~1.9 fold and this effect was blocked by PTS. The observation that PTS inhibited the activation of JNK in vivo are in agreement with the in silico results (see below) that PTS are inhibitors of JNK1/2 and MKK4, a kinase that activates JNK (Table 1).

JNK activation and binding to Sab are considered a critical step mediating APAP hepatotoxicity.⁴⁰ Oxidative stress in APAP overdose causes TRX1 oxidation that leads to activation of apoptosis signal-regulating kinase 1 (ASK1); moreover, oxidative stress also activates mixed-lineage kinase 3 (MLK3).^63,64 Both activated ASK1 and MLK3 can phosphorylate MKK4, which leads to subsequent activation of JNK in the cytosol.^63,64 Activated JNK translocated to the mitochondrial outer membrane, where it binds to Sab, a mitochondrial protein. JNK binding with Sab causes inhibition of the electron transport, which further increases ROS and peroxynitrite formation and exacerbates mitochondrial dysfunction, thus forming an adverse cycle.^17,60,65 Both ROS and peroxynitrite can trigger the onset of the MPT^66,67 and MPT has been shown to cause cell death in APAP hepatotoxicity.^18,68 In this study, PTS was shown to inhibit JNK activation and binding to Sab after APAP in vivo (Figure S7, Supporting Information). This effect could be due to inhibition of PTS on MKK4 and JNK1/2 as well as a consequence of PTS inhibition of ROS and RNS formation. Inhibition of JNK activation and binding with Sab breaks the adverse cycle that leads to cell death in APAP hepatotoxicity.

Computational Binding Affinity of PTS to Hepatotoxicity-regulating Proteins.

Nine proteins regulating liver inflammation and necrosis were evaluated in an in silico binding study to elucidate possible mechanisms of action of PTS against APAP-induced hepatotoxicity.^69,70 E,E-, E,Z-, Z,E-, and ZZ-PTS were docked to CYP2E1, Keap1, iNOS, JNK-1, JNK-2, MKK4, MKK7, TNFα, and COX-2. The in silico binding results were in good agreement with the in vivo results observed and the binding affinities are summarized in Table 2. PTS showed the poorest binding affinity (31.4 - 34.9 kcal/mol) to CYP2E1 when compared to CYP2E1 inhibitor⁷¹ (4-methyl pyrazole; −4.3 kcal/mol) and APAP (−4.3 kcal/mol). CYP2E1 has the smallest binding domain among the cytochrome P450 family.⁷¹ Thus, it is not surprising that PTS, a relatively larger molecule, fit poorly in the binding pocket of CYP2E1. Molecular modelling showed that the PTS structure is too large and extended out from the narrow CYP2E1 binding pocket (Figure S8, Supporting Information). This observation indicated that PTS should not inhibit the metabolism of APAP by CYP2E1 to form NAPQI, in agreement with the in vivo results.

Table 2.

PTS Binding Affinity (kcal/mol).

Compound	CYP2E1	Keap1	iNOS	JNK1	JNK2	MKK4	MKK7	TNFα	COX-2
E,E-PTS	34.9	−8.6	−4.5	−10.2	−8.4	−10.5	−10.0	−8.9	−4.7
E,Z-PTS	31.4	−9.5	−4.1	−9.0	−8.6	−8.5	−8.5	−9.3	−6.3
Z,E-PTS	33.9	−9.0	−3.2	−8.9	−9.7	−10.8	−10.3	−8.4	−4.2
Z,Z-PTS	31.6	−7.4	−8.9	−9.7	−7.9	−10.1	−10.0	−9.0	−4.9
APAP	−4.3	n.a	n.a	n.a	n.a	n.a	n.a	n.a	n.a
4-Methylpyrazole	−4.3	n.a	n.a	n.a	n.a	n.a	n.a	n.a	n.a
trans-Resveratrol	n.a	−6.9	n.a	n.a	n.a	n.a	n.a	n.a	n.a
S-Ethylisothiourea	n.a	n.a	−4.2	n.a	n.a	n.a	n.a	n.a	n.a
ATP	n.a	n.a	n.a	−8.6	−7.6	−9.5	−7.5	n.a	n.a
SP600125	n.a	n.a	n.a	−8.7	−9.1	n.a	n.a	n.a	n.a
aTNF-α inhibitor	n.a	n.a	n.a	n.a	n.a	n.a	n.a	−8.8	n.a
Meloxicam	n.a	n.a	n.a	n.a	n.a	n.a	n.a	n.a	−10.2

^a6,7-Dimethyl-3-((methyl(2([methyl((1-[3-(trifluoromethyl)phenyl)-1H-indol-3-yl)methyl)amino)ethyl)amino)-methyl)-4H-chromen-4-one; n.a: not applicable

PTS showed better binding affinity (E,Z-PTS; −9.5 kcal/mol) to Keap1 than Keap1 inhibitor⁵² (trans-resveratrol; −6.9 kcal/mol) at the Keap1-Nrf2 binding domain (Figure S9, Supporting Information). Molecular modeling showed that the o,p-dihydroxybenzene moiety, which is present in both PTS and trans-resveratrol, shared a similar docking space in the binding pocket. This suggests that the o,p-dihydroxybenzene scaffold could have Keap1 inhibition activity (Figure S9, Supporting Information). The binding of PTS to Keap1 would be expected to disrupt the formation of a Keap1-Nrf2 complex.⁷² The unbound Nrf2 protein will subsequently would trigger the expression of antioxidant defense enzymes, such as SOD, TRX1, Prx, GPx, and GR, to decrease oxidative stress.^69,72 Another interesting observation was that PTS gave a better binding affinity (Z,Z-PTS; −8.9 kcal/mol) to iNOS than an iNOS inhibitor⁷³ (S-ethylisothiourea; −4.2 kcal/mol) (Figure S10, Supporting Information). This indicated that PTS potentially could inhibit NO production and protect against RNS formation. PTS showed better binding affinity to JNK-1 (E,E-PTS; −10.2 kcal/mol) and JNK-2 (Z,E-PTS; −9.7 kcal/mol) than ATP (JNK-1: −8.6 kcal/mol; JNK-2: −7.7 kcal/mol) at the ATP-competitive binding domain (Figure S11 and S12, Supporting Information). The binding affinity displayed by PTS is comparable with a known JNK-1/2 inhibitor (SP600125).⁷⁴ In addition, PTS also showed better binding to MKK4 and MKK7 at the ATP-competitive binding domain, when compared to ATP. Z,E-PTS showed a binding affinity of −10.8 kcal/mol to MKK4, better than ATP (−9.5 kcal/mol) (Figure S13, Supporting Information). In the case of MKK7, ZE-PTS show a binding affinity of −10.3 kcal/mol, better than ATP binding affinity (−7.5 kcal/mol) (Figure S14, Supporting Information). This suggests that PTS could inhibit the phosphorylation of JNK-1/2, MKK4, and MKK7.

PTS were docked to two proinflammatory targets, TNFα and COX-2, in the binding experiment. The binding results showed PTS bound well to TNFα. PTS showed comparable binding affinity (E,Z-PTS; −9.3 kcal/mol) to TNFα when compared to a known TNFα inhibitor⁷⁵ [6,7-dimethyl-3-((methyl(2-(methyl((1-(3-(trifluoro-methyl)phenyl)-1H-indol-3-yl)methyl)amino)-ethyl)amino)-methyl)-4H-chromen-4-one; −8.8 kcal/mol]. Both PTS and TNFα inhibitors shared a common chromone moiety in their structure. Molecular modelling showed that the chromone moiety is docked at a similar space in the binding pocket, suggesting that the chromone scaffold could have TNFα inhibition activity (Figure S15, Supporting Information). In the case of COX-2, PTS showed poor binding affinity (−4.2 - −6.3 kcal/mol) in the binding experiment when compared to COX-2 inhibitor⁷⁶ (meloxicam; −10.2 kcal/mol). Molecular modeling showed that the PTS structures are too large and fits poorly in the narrow binding pocket (Figure S16, Supporting Information). In summary, the in silico binding experiments showed that isomers of PTS bind favorably and are potential inhibitors of Keap1, iNOS, JNK-1/2, MKK-4/7, and TNFα.

CONCLUSION

Taken together, this study showed that the platanosides markedly decreased APAP hepatotoxicity in mice when given simultaneously or even after APAP and therefore provides a promising strategy for prevention and therapy of APAP-induced liver injury as well as perhaps other forms of chemically induced liver injury. Furthermore, these studies revealed that an antibiotic-antipyretic drug combination involving PTS and APAP would reduce the risk of liver injury in addition to controlling Gram positive and negartive vancomycin-resistant bacteria and pyrexia and pain. The PTS do not function by decreasing the formation of NAPQI but by decreasing the oxidative/nitrative stress and the subsequent JNK activation/Sab binding pathways suggesting possible broader applications in the control of liver injury and disease. The reduced IL-1β, MPO, 4-HNE, iNOS, and 3-NT expressions under PTS treatment in APAP-overdosed conditions clearly demonstrated the ability of PTS in decreasing inflammation and oxidative/nitrative stress. The in vivo results also showed that PTS did not alter the expression of CYP2E1, the main enzyme that catalyzes the metabolism of APAP to NAPQI. Moreover, the in silico experiment showed that the PTS structure is too large to bind to the narrow CYP2E1 binding pocket, indicating that PTS is unlikely to inhibit CYP2E1 activity. Therefore, the protective effects of PTS are not due to a decrease in NAPQI formation.

The in vivo observations suggested the mechanism of action by PTS in decreasing APAP hepatotoxicity could proceed via the inhibition of JNK1/2 activation, reduction of inflammation factors and increase production of antioxidant enzymes (Figure 4). The decrease in JNK1/2 activation under PTS treatment in APAP-overdosed conditions could be due to the inhibition of the upstream MKK4 and/or JNK1/2 phosphorylation. Modeling data also suggested that the isomeric PTS combination may perform better than the individual molecules, further supporting the potential value of a botanical drug in human health. Last, the protection provided by PTS is consistent with previous studies on other flavone compounds (Table 1). However, the particular activity of PTS in the control of antibiotic-resistant bacteria affects considerable potential as a treatment combination with APAP, as APAP is used to control pyrexia and pain during a bacterial infection.

Figure 4.

Protection of PTS against APAP-induced hepatotoxicity.

The current study has clearly demonstrated that PTS block the key signaling pathways by APAP adducts that lead to cell death and inflammation after APAP overdose. However, the detailed mechanisms remain unclear. It will be a subject of interest for future study to elucidate the role of PTS in influencing APAP protein adducts (e.g., variety, quantity, binding property, and time course of formation and degradation), ability to reverse toxicity (e.g. after prolonged APAP exposure, combination treatment with NAC), other chemopreventive or therapeutic effects on other oxidative and nitrative stress, and JNK activation-mediated liver injury (e.g., alcoholic liver injury, NASH, liver injury caused by cancer chemotherapeutic drugs).

EXPERIMENTAL SECTION

General Experimental Procedures.

Sources of chemicals, antibodies and other reagents used in this study are listed in Table S1 (Supporting Information). The ¹H and ¹³C NMR spectra were obtained at 500 and 600 MHz for ¹H and 125 and 150 MHz for ¹³C, respectively, on an INOVA 500 MHz spectrometer and a BRUKER AV-III-600 MHz spectrometer. ESIMS was performed using an Agilent 1100 series LC/MS ion trap mass spectrometer. HRESIMS was performed using an Agilent 6520 Accurate-Mass Q-Tof LC/MS mass spectrometer. Reversed-phase preparative HPLC was carried out on Shimadzu instrument (pump LC-6AD, UV detector SPD-20A, 224 nm). Analytical HPLC was performed on a Shimadzu instrument (pump LC-20AT, UV detector SPD-M20A, 224 nm). Column chromatography (CC) was carried out on a CombiFlash Rf 200 chromatograph (Teledyine Isco, United States).

Plant material.

P. occidentalis was collected and identified as described in reference 27.²⁷

Extraction and Isolation.

Three and a half kg of P. occidentalis (American Sycamore) leaves were air dried, finely ground, and extracted with ethanol for 48 hours. The ethanol solvent was then removed to give a thick brown slurry. The crude extract (brown slurry) was added to a column (length: 3 feet 11 inches, diameter: 9 inches) and washed (96 L 200 proof of ethanol) for preliminary cleaning. The solvent was again removed using rotavap to give 465 g of crude ethanol extract. The crude ethanol extract then placed in a Buchner funnel and washed with dichloromethane followed by distilled water to remove the non-polar components and other water soluble residues. The washed crude ethanol extract was then dried under vacuum using a speed-vac to give 230 g of final ethanol extract. The extract was then subjected to column chromatography using C18 silica gel (40-63 mm particle size (230-400 mesh), pore diameter 60 Å, space surface area 500 m²/g) and eluted with water (100%) and gradually increasing the ethanol percentage (10, 33, 50, 67, and 100%) to give four major fractions, i.e., 11 g (from 10% ethanol), 20 g (from 33% ethanol), 81 g (from 50% ethanol), and 17 g (from 67% ethanol). The 67% ethanol fraction was further purified using column chromatography (40-63 mm particle size, pore diameter 60 Å, bulk density: 0.5 g/mL) using eluting solvent: 1) hexane:ethyl acetate (4:1, 3:1, 2:1, 1.5:1, 1:1, 0:1); 2) ethyl acetate: methanol (1:1, 0:1); and 3) 100% methanol to give nine subfractions. i.e., 67E-0 fraction 204 mg, 67E-1 fraction 137 mg, 67E-2 fraction 51 mg, 67E-3 fraction 94 mg, 67E-4 fraction 78 mg, 67E-5 fraction 2.5 g, 67E-6 fraction 107 mg, 67E-7 fraction 1.4 g, 67E-8 fraction 8.0 g. PTS is found in fraction 67E-5. Fraction 67E-5 was further purified using HPLC (kinetix C₈, 65% MeOH; 7 mL/min) to give an overall 0.02-0.05% yield and >95% purity. The final composition of PTS used in the present study is (E,E-, Z,Z-, E,Z-, and Z,E-PTS; 1.9:1.0:1.1:0.5; Figure S17, Supporting Information). Molecular structures of PTS are shown in Fig. 1.

Treatment of Mice with APAP, PTS, and NAC.

Male C57BL/6 mice (6-7 weeks old; 20-25 g; Jackson Laboratory, Bar Harbor, MA) were used in this study. Mice were fasted overnight (18 - 20 h) before APAP administration in vivo. APAP (Table S1, Supporting Information) was dissolved in normal saline (20 mg/mL) and injected intraperitoneally (300 mg/kg). Food was returned to mice 30 min after APAP treatment. PTS was dissolved in ethanol and Neobee M-5 (Stephan Lipid Nutrition, Maywood, NJ). PTS (10 mg/kg) or equal volume of vehicle were given at the time of APAP dosing (0 h, gavage) or at 2 h after APAP administration (gavage or i.p.). Previous studies showed that NAC (300 mg/kg), the current antidote for APAP, markedly decreases APAP hepatotoxicity when administered at 2 h after APAP dosing in mice; however its protective effect was lost when dosing in later time points (3-4 h).¹⁸ Therefore, some mice were treated with NAC (300 mg/kg, i.p.) at 2 h after APAP. Control animals were fasted overnight and received vehicle (saline and ethanol+Neobee M-5) injections. All animal procedures followed the NIH “Guide for the Care and Use of Laboratory Animals” and were preapproved by the Institutional Animal Care and Use Committee at the Medical University of South Carolina (Protocol number ARC#2019-00861).

Histology and Serum ALT Measurement.

At 2 h and 24 h after the initial APAP administration, mice were treated with ketamine/xylazine (90 mg/kg and 10 mg/kg, i.p.), and the abdomen of each was opened, and blood (~0.4 mL) was collected from the inferior vena cava. Livers were harvested after perfusion with 1 mL normal saline via the portal vein. Part of the liver was snap-frozen in liquid nitrogen and then kept at −80 °C until subsequent biochemical analysis and the other part was fixed in 10% neutral buffered formalin for 24 h. Liver tissue was processed and embedded in paraffin, and liver sections (5 μm) were deparaffinized and stained with H&E staining. The histology of the liver sections was examined under a light microscope. Five random images per liver section were acquired using a BD CARV microscope (BD Biosciences, Rockville, MD) and a 10x objective lens. Necrotic areas were quantified by image analysis in a blinded manner using Image-J (FIJI, Madison, WI).⁷⁷ Serum ALT was measured using an analytical kit (Table S1, Supporting Information) according to the manufacturer’s instructions (ALT analytical kit, Pointe Scientific, Uncoln Park, MI, United States).

NAPQI-Protein Adduct Assay.

NAPQI-protein adducts in liver tissue collected 2 h after APAP administration were measured by HPLC, as previously described.⁷⁸

Immunoprecipitation of Sab (SH3 Domain-binding Protein 5).

Mice liver tissue was collected at 24 h after APAP treatment. Livers were homogenized in ice-cold lysis buffer, and protein contents in the lysates were determined using a Pierce BCA protein assay kit (Table S1). Immunoprecipitation (IP) of Sab in liver lysates (500 μg protein) was performed using a Pierce Classic Immunoprecipitation Kit (Table S1) with the corresponding Sab antibody (5 μg, Table S1), conducted according to the manufacturer’s instruction (Pierce Classic Immuno-precipitation Kit, Pierce Biotec., Rockford, IL, United States). Protein contents in the immunoprecipitates were measured and then an equal amount of Sab was loaded to each lane.⁷⁹ Binding of phospho-JNK1/2 to Sab were determined by immunoblotting, as described below.

Immunoblotting.

Livers were collected at 2 or 24 h after APAP treatment. Livers were homogenized in ice-cold lysis buffer, and protein contents in the lysates were determined. Immunoblotting of proteins in liver tissue lysates or the immunoprecipitates (as described above) was performed using primary antibodies specific for the proteins of interest interleukin-1β (IL-1β), JNK, myeloperoxidase (MPO), 4-hydroxynonenal (4-HNE), inducible nitric oxide synthase (iNOS), phospho-JNK, SH3 domain–binding protein 5 (Sab), and house-keeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH)] at concentrations of 1:1,000 to 3,000 at 4 °C overnight, respectively. Horseradish peroxidase-conjugated secondary antibodies were applied, and detection was using a chemiluminescence kit and a LICOR Odyssey imaging system (LICOR Biosciences, Lincoln, NE, United States).⁷⁷ The sources of all antibodies are shown in Table S1 (Supporting Information).

Statistical Analysis.

Groups were compared using ANOVA plus Student-Newman-Keul’s post-hoc test using p <0.05 as the criterion of significance. Values are means ± SEM, and group numbers are given in the figure legends.

In silico Molecular Binding.

The structures of E,E-, E,Z-, Z,E-, and Z,Z-PTS, ATP, 4-methylpyrazole, trans-resveratrol, and SP600125 were optimized using the MM2 energy-minimized function in the Chem3D Ultra version 16.0. The crystal structure of the receptor proteins (Table S2, Supporting Information) were obtained from the Protein Data Bank.^80,81 AutoDockTools version 1.5.6 were used to prepare the receptor proteins and ligands for the molecular docking experiment. The grid box parameter used were: grid box spacing = 1.0 Å; x-dimension = y-dimension = z-dimension = 20. AutoDock Vina program was used to perform the docking and calculate the binding affinity.^82,83 The results were processed and analyzed using the BIOVIA Discovery Studio Visualizer version 17.2.0. The binding affinity is summarized in Table S2 (Supporting Information).