Fig.3 Drugs with hepatotoxicity modulated by gut microbiota: acetaminophen (A), tacrine (B), triptolide (C), and geniposide (D). Red structures signify hepatotoxic drugs or metabolites; green ones signify gut-dervied metabolites that confers hepatoprotective effects against relevant drugs. A: PPD is a gut-derived metabolites that exacerbates APAP-induced hepatotoxicity by competing for GSH and so doesp-cresol which competes for sulfotransferase. DOPA, 4-HPA, and fructose are prebiotics that ameliorate APAP-induced hepatotoxicity. B: β-glucuronidase producing bacteria including Bacteroides ,Enterobacteriaceae and Lactobacillus aggravate tacrine hepatotoxicity. C: Gut-derived propionate ameliorates triptolide-induced hepatotoxicity. D: Geniposide transforms into hepatotoxic genipin under the action of gut micorbiota. Abbreviations: 4-HPA, 4-hydroxyphenylacetic acid; APAP, acetaminophe; CYPs, cytochrome 450 proteins; DOPA, 3,4-dihydroxyphenylacetic acid; GSH, glutathione; NAPQI, N-acetyl-p-benzoquinone imine; SULTs, sulfotransferases; UGTs, UDP-glucuronosyltransferases.
3.1 Acetaminophen
Acetaminophen (APAP) is a representative hepatotoxic drug accounting for most cases of DILI in the United States (Chao et al., 2018). When administered orally, it mainly undergoes two ways of metabolism: (a) sulfonated or glucuronidated by phase Ⅱ enzymes and (b) first converted into the hepatotoxic N-acetyl-p-benzoquinone imine (NAPQI) by CYPs and then conjugated with glutathione (GSH) (Chang et al., 2020c). Both ways of metabolism function as detoxification and thus any factor that influences them could have an impact on APAP-related hepatotoxicity. A human experiment demonstrated that p-cresol , microbiota-derived metabolite from tyrosine, competed for sulfation with APAP, which might induce APAP accumulation and result in hepatotoxicity (Clayton et al., 2009). This experiment introduced gut microbiota into the understanding of APAP-related hepatotoxicity for the first time. Indeed, germ-free mice are less sensitive to APAP-induced hepatotoxicity compared with conventional mice (Possamai et al., 2015) and antibiotic treatment decreases urinary sulfonated APAP (Malfatti et al., 2020), suggesting that intestinal microbiota is implicated in APAP-related hepatotoxicity and metabolism. It has long been known that APAP-induced liver injury exhibits diurnal variation, which is more severe in the night than during the daytime (Kim and Lee, 1998). Despite the hepatic GSH level and drug metabolic activity mentioned as contributing factors of this phenomenon (Kim and Lee, 1998), the gut microbiota is thought to play a pivotal role therein. Evidence suggests that gut microbiota biogeography and metabolome undergo diurnal fluctuation, which consequently regulates rhythmic liver transcriptome and detoxification pattern as proved by APAP-induced circadian hepatotoxicity (Thaiss et al., 2016). The role of intestinal microbiota in rhythmic hepatotoxicity of APAP is further confirmed by another study and it also reported that 1-phenyl-1,2-propanedione (PPD), a gut microbial metabolite exacerbated APAP-induced liver injury by depleting hepatic GSH (Gong S, Lan T, Zeng L, Luo H, Yang X, Li N, Chen X, Liu Z, Li R, Win S, Liu S, Zhou H, Schnabl B, Jiang Y, Kaplowitz N, 2018). A pharmacokinetic study demonstrated that treatment with probiotics, especiallyLactobacillus reuteri K8, diminished absorption of orally administered APAP (Kim et al., 2018), so it is speculated that pretreatment of moderate probiotics may ameliorate APAP-induced hepatotoxicity while maintaining efficacy simultaneously.
Based on the gut microbial mechanism of APAP-induced liver injury, some microbiota, diet-derived substances, and probiotics have been discovered to exert hepatoprotective effect against APAP by targeting the gut microbiota. 3,4-dihydroxyphenylacetic acid (DOPA), a microbial metabolite of quercetin, protects the liver from damage of APAP by activating nuclear factor erythroid 2-related factor 2 (Nrf-2) and enhancing phase II and antioxidant enzymes (Xue et al., 2016), so does 4-hydroxyphenylacetic acid (4-HPA), a major microbial metabolite of polyphenols (Zhao et al., 2018). As for food-derived hepatoprotective agents, a study found that supplemental fructose respited APAP-induced liver injury by altering gut microbial community, especially the genusAnaerostipes , which closely relates to hepatic levels of CYPs and GSH (Cho et al., 2017). Furthermore, probiotics like bacillus spores have been proven to protect rat from APAP-induced acute liver injury by altering proinflammatory cytokines levels (Neag et al., 2020). It seems gut microbial modulation does work in attenuating APAP-triggered liver injury, mainly by dietary modulation and probiotics intake, but it has to be verified in human trails for future application.
3.2 Tacrine
Tacrine was the first approved drug prescribed to treat Alzheimer’s disease but was withdrawn from the market for its large pharmacokinetic variation (Jarrott, 2017) and consequent unpredictable hepatotoxicity (Bethesda (MD): National Institute of Diabetes and Digestive and Kidney Diseases, 2018). Although it is no longer used in clinical, underlying mechanisms involved in tacrine-induced liver injury is still being investigated (Park et al., 2015) and it is utilized to establish liver injury model in cells and animals (Stachlewitz et al., 1997; Choi et al., 2015). Previous researches mainly focused on the hepatic metabolism of tacrine and found that tacrine was extensively metabolized by CYP2A1 and one of the hydroxylated metabolites exerted hepatotoxicity (Madden et al., 1993). Fluvoxamine is a potent inhibitor of CYP2A1 and is thought to attenuate tacrine-induced hepatotoxicity by reducing the formation of toxic metabolites but needs further validation (Rasmussen et al., 1998). A recent comprehensive study combining pharmacokinetics, metabolomics, and metagenomics revealed that rats responded differently to tacrine with strong responders exhibiting higher tacrine exposure, greater deglucuronidation capabilities, and an abundance of β-glucuronidase producing bacteria including Bacteroides ,Enterobacteriaceae and Lactobacillus (Yip et al., 2018). Further validation indicated that coadministration of oral β-glucuronidase or antibiotics with tacrine increased or decreased susceptibility to tacrine-induced hepatotoxicity, indicating that gut microbiota especially the gut microbial β-glucuronidase plays a pivotal role in tacrine-related hepatotoxicity (Yip et al., 2018). It implicates the significance of gut microbiota in interindividually varied toxicity of drugs. While current researches attempt to decipher individual medication differences by genome but with limited achievements, the future focus can be shifted to gut microbiota.
3.3 Triptolide
Hepatotoxicity is one of the major factors that hamper the application and progress of herbal medicine nowadays. Triptolide is the major bioactive component of the herbTripterygium wilfordii Hook F which exerts anti-inflammatory and anti-autoimmune functions but also confers hepatotoxicity. A study elucidated that triptolide exerted anti-inflammatory effect against ulcerative colitis with gut microbiota involved (Wu et al., 2020), indicating an interactive relationship between gut microbiota and triptolide. Molecular mechanisms of triptolide-induced hepatotoxicity have been comprehensively delineated, encompassing membrane damage, mitochondrial disruption, metabolism dysfunction, endoplasmic reticulum stress, oxidative stress, apoptosis, and autophagy (Xi et al., 2017), but the role of gut microbiota plays in it is put forward only recently. Researchers found that pretreatment of antibiotics exacerbated liver injury induced by triptolide in mice while supplemental microbiota-derived propionate proffered liver protection against triptolide (Huang et al., 2020a). It reveals that gut microbiota protects the liver from triptolide, contrary to the case in tacrine. Microbial metabolites of the prototype drugs may be an important factor herein, which indicates that microbial metabolites of triptolide confer lower hepatotoxicity while those of tacrine higher. Indeed, gut microbiota plays a unique role in triptolide metabolism since gut microbiota and liver microsomes produce different metabolites of triptolide (Peng et al., 2020). It indicates that gut microbial metabolism constitutes a significant part in drug metabolism and under some circumstances it influences and even determines toxic outcomes of drugs. Besides, triptolide treatment induced bile acids accumulation in rats and it was associated with suppression of FXR and hepatic Sirtuin (Sirt1) expression, activation of which ameliorated triptolide-induced hepatotoxicity (Yang et al., 2017).
3.4 Geniposide
Geniposide is the major bioactive component of herb Gardeniae Fructus which exhibits a wide range of pharmacological functions including neuroprotective, hypoglycemic, hepatoprotective, anti-inflammatory, analgesic, cardio-protective, antioxidant, immune-regulatory, antithrombotic, and antitumoral effects (Zhou et al., 2019b). The herb has long been used as choleretic to treat jaundice in Chinese history and modern research also reveals that geniposide attenuates APAP-induced liver injury in mice (Yang et al., 2019). Nonetheless, it has been reported that geniposide elevated serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels in rats and the conversion of geniposide to genipin made a big difference therein (Yamano et al., 1990). Genipin, the aglycon of geniposide, is produced from geniposide by intestinal bacteria but not by the liver (Aburada and Kobashi, 1994) and displayed much higher toxicity than geniposide in HepG2 cells (Kang et al., 2012), indicating the important role of gut microbiota in geniposide-induced liver injury. Antibiotic treatment of rats enhanced in vivo exposure of geniposide by suppressing metabolic activities of gut microbiota (Jin et al., 2014) and it might reduce hepatotoxicity because of a decreased level of genipin. A further study demonstrating mechanism involved in geniposide-induced liver injury attributed it to bile acids synthesis alteration (Tian et al., 2017), but the role of intestinal microbiota plays in it remains to be investigated. A recent study also proved the significance of intestinal microbiota in geniposide-related hepatotoxicity and further demonstrated that genipin dialdehyde intermediate, which was capable of conjugating with amino acids spontaneously, was the origin of geniposide-induced hepatotoxicity (Li et al., 2019).
3.5 Substances other than drugs
Besides drugs, the gut microbiota is also extensively engaged in other substances-induced hepatotoxicity which will be discussed here as well. Some of the researches are profound in mechanistic elucidations and can provide as references for mechanistic studies of drug-induced hepatotoxicity.
Aflatoxins B1 (AFB1), a kind of mycotoxin, causes liver fibrosis and hepatocellular carcinoma both in humans and livestock. AFB1 exposure caused a wide range of metabolic disturbances and particularly elevated gut microbial cometabolites including phenylacetylglycine and hippurate in rats (Zhang et al., 2011). Other metabolomics based studies also found that AFB1 disrupted several important metabolic pathways of gut microbiota, including synthesis of SCFAs, pyruvic acid-related pathways, amino acids, bile acids, and long-chain fatty acids metabolism (Zhou et al., 2018b, 2019a), suggesting that gut microbial metabolism disruption contributes to the pathogenesis of AFB1-induced diseases. Probiotics supplementation including Lactobacillus Species (Liew et al., 2019; Wacoo et al., 2020), Propionibacterium freudenreichii(El-Nezami et al., 2006), Bacillus subtilis , andCandida utilis (Chang et al., 2020b) exerted protective effects against AFB1 both in animal experiments and human trials. These studies illustrate that AFB1 perturbs gut microbiota which can be restored by probiotics and they also help degrade AFB1 and thus lower toxicity.
Some simple substances form another main category of environmental toxicants that exert hepatotoxicity. Long-term exposure of low-dose cadmium causes severe liver damage with gut microbiota implicated. It was found that cadmium induced gut dysbiosis, mainly decreasingFirmicutes and γ-proteobacteria levels, and increasing serum lipopolysaccharide (LPS) and consequently induced hepatic inflammation (Zhang et al., 2015). It also increased intestinal permeability and hence inflicted accumulation in the liver but could be partially restored by antibiotic treatment (Liu et al., 2020). Arsenic is another example here. Although it is structurally simple, it undergoes complex biotransformation and forms a series of products with different toxicity. In vitro incubation elucidated the unique metabolic role of gut microbiota and in vivo study illustrated that gut microbiota reduced arsenic load by promoting methylation and thus protected the host from hepatotoxicity of arsenic (Chi et al., 2019a). Similarly, another study also found that a stable gut microbiome was a key determinant of survival to arsenic exposure, andFaecalibacterium prausnitzii protected it (Coryell et al., 2018).
Other environmental toxicants that are found to elicit hepatotoxicity with gut microbiota engaged include polychlorinated biphenyl 126 (Chi et al., 2019b), 2,3,7,8-tetrachlorodibenzo-p -dioxin (Fader et al., 2017), titanium dioxide nanoparticles (Chen et al., 2019), etc. These studies share similarities in researching method: first, 16S rRNA sequencing analysis is utilized to demonstrate gut dysbiosis-induced by the toxicants and then omics analysis especially metabolomics is used to elucidate disturbed metabolites or metabolic pathways by focusing on gut microbial metabolites and further deduce possible mechanisms involved by analyzing results of both. It is a current mainstream in gut microbiota associated mechanistic studies particularly in researches of metabolic diseases (Zeng et al., 2020).