Homeostasis of the gut microbiota is often disrupted in patients with metabolic diseases. Numerous studies demonstrate that herbal medicines can modulate the composition of the gut microbiota and restore gut homeostasis in rodents with metabolic diseases (Figure 1). Subsequently, we summarized the major intestinal bacteria regulated by herbal medicines at different classification levels and discussed their association with metabolic diseases.

At the phylum level, Firmicutes and Bacteroidetes are the most abundant bacteria. An elevated ratio of Firmicutes to Bacteroidetes (F/B) in the gut microbiota has been linked to metabolic diseases such as T2DM, hyperlipidemia, obesity, and NAFLD (Ansari et al., 2016; Chen et al., 2018; Li et al., 2019a; Zhang et al., 2022a). Some herbal medicines have been shown to reduce the F/B ratio. For example, Alisma orientale (Sam.) Juz. (Alismataceae; Alismatis Rhizoma) decreased the F/B ratio and improved gut microbiota dysbiosis in hyperlipidemic rats induced through a high-fat and high-sucrose diet (Li et al., 2019a). Similarly, taking the Linggui Zhugan formula reduced the F/B ratio in diabetic mice (Liu et al., 2019). Moreover, Proteobacteria can cause inflammatory responses and lead to metabolic disorders (Shin et al., 2015). Studies showed that Gynostemma pentaphyllum (Thunb.) Makino (Cucurbitaceae; Gynostemmae Pentaphylli Herba) and Jiangzhi granules could reduce the abundance of Proteobacteria in NAFLD mice (Jia et al., 2018a; Wang et al., 2021).

At the family level, Lachnospiraceae bacteria are major producers of short-chain fatty acids (SCFAs), which are beneficial in improving metabolic diseases (Du et al., 2022). A study demonstrated that Jiangzhi granules could regulate the gut microbiota disturbance in NAFLD mice and resulted in a significant increase in the abundance of Lachnospiraceae bacteria (Wang et al., 2021). Ruminococcaceae bacteria are known to degrade mucin and ferment carbohydrates, which results in the production of beneficial acetate and propionate (Paone and Cani, 2020). Li et al. (2021b) reported that the Jieyu Qutan Huazhuo formula significantly increased the abundance of Ruminococcaceae bacteria in hyperlipidemia mice. Similarly, Zhang et al. (2019) demonstrated that Spatholobus suberectus Dunn (Leguminosae; Spatholobi Caulis) could increase the abundance of Ruminococcaceae bacteria in diet-induced obese mice. In addition, Desulfovibrionaceae bacteria are capable of producing harmful H₂S and lipopolysaccharide (LPS), which can damage intestinal epithelial cells (Christophersen et al., 2011; Liu et al., 2014). One study revealed that Momordica charantia L. (Cucurbitaceae; Momordicae charantiae fructus) significantly reduced the number of Desulfovibrionaceae bacteria in obese rats (Bai et al., 2016).

At the genus level, Akkermansia bacteria are beneficial microorganisms. They can degrade mucin and produce propionic acid, thus protecting the intestinal mucosal barrier (Cui et al., 2019). Moreover, Akkermansia bacteria are associated with a reduced risk of obesity (Zhou et al., 2020). Notably, Xu et al. (2021) demonstrated that herbal medicine, Berberis kansuensis C.K. Schneid. (Berberidaceae; Berberis Cortex), significantly increased the abundance of Akkermansia bacteria in T2D rats. Similarly, Li et al. (2022) reported that Penthorum chinense Pursh (Saxifragaceae; Penthori Chinensis Herba) could increase the abundance of Akkermansia bacteria in the feces of NAFLD mice. Blautia bacteria possess potential probiotic properties that help regulate host health and alleviate metabolic syndrome (Liu et al., 2021). It was reported that the Shenqi compound could increase the abundance of Blautia bacteria and its metabolites in model rats and thereby improve T2DM (Zhang et al., 2022b). Furthermore, Escherichia–Shigella bacteria are a class of opportunistic pathogens associated with inflammation. Xiao et al. (2020) found that Scutellaria baicalensis Georgi (Labiatae; Scutellariae Radix) and Coptis chinensis Franch. (Ranunculaceae; Coptidis Rhizoma) could significantly reduce the number of Escherichia–Shigella in the feces of diabetic rats.

At the species level, Akkermansia muciniphila, which is a well-known beneficial microbe, has been shown to be effective in improving metabolic diseases such as T2DM and NAFLD (Yan et al., 2021). One study demonstrated that A. muciniphila reversed high-fat diet (HFD)-induced metabolic endotoxemia, adipose tissue inflammation, and insulin resistance (Everard et al., 2013). Notably, some herbal medicines have been found to increase the abundance of A. muciniphila and improve metabolic diseases. For example, Yang et al. (2022a) reported that taking C. chinensis Franch. (Ranunculaceae; Coptidis Rhizoma) for 4 weeks could increase the number of A. muciniphila in the feces of hyperlipidemic mice. Moreover, Parabacteroides goldsteinii is considered a probiotic that can ameliorate obesity and its associated metabolic disorders (Chang et al., 2019). One study found that taking the mycelium of Ganoderma lucidum (Leyss. ex Fr.) Karst (Polyporaceae; Ganoderma) for 12 weeks could increase the number of P. goldsteinii and reduce body weight, inflammation, and insulin resistance in obese mice (Chang et al., 2015).

Homeostasis of the intestinal barrier is vital for human health. When the integrity of the intestinal barrier is compromised, toxins from gut microbes (e.g., LPS) can escape from the intestinal lumen into the bloodstream (Guerville et al., 2017). LPS is a glycolipid component of the outer membrane of gram-negative bacteria (Carnevale et al., 2019). LPS can activate the TLR4/MyD88/NF-κB signaling pathway to release pro-inflammatory indicators (e.g., IL-6, IL-1β, and TNF-α) and thus trigger a cascade of inflammatory processes that ultimately lead to chronic low-grade inflammation (Mehta et al., 2010; Mohammad and Thiemermann, 2021). It is well known that chronic inflammation is a pathogenic factor in metabolic diseases (Hotamisligil et al., 2006). Thus, strengthening the function of the intestinal barrier and reducing inflammation is beneficial for ameliorating metabolic diseases.

Tight junctions, a key component of the intestinal barrier, are essential for preventing the transmission of harmful molecules (Meijers et al., 2018). Studies have shown that the decreased expression of tight junction proteins, such as occludin, zonula occludens-1 (ZO-1), and claudin-1, can lead to an impaired epithelial barrier function and increased intestinal permeability (Hossain and Hirata, 2008; McGuckin et al., 2009). Some herbal medicines have been shown to protect the intestinal mucosal barrier by upregulating the expression of several tight junction proteins (Figure 2). For example, Jiangan Xiaozhi decoction, Quzhuo Tongbi decoction, Simiao decoction, Jiangzhi granules, and Hongqi Jiangzhi formula can upregulate the expression of tight junction-associated proteins occludin and ZO-1 in intestinal tissues of rodents with metabolic diseases (Liang et al., 2018; Liao et al., 2020; Han et al., 2021; Wang et al., 2021; Wen et al., 2021). Similarly, Qushi Huayu decoction was found to protect the intestinal barrier function by upregulating the mRNA expression of ZO-1, occludin, and claudin-1 and thereby inhibiting LPS gut leakage and ultimately improving non-alcoholic steatohepatitis induced by a high-fat diet in mice (Leng et al., 2020). In addition to blocking LPS passage through the intestinal barrier, some herbal medicines can directly inhibit LPS production by reducing the number of LPS-producing bacteria such as Escherichia–Shigella, Enterococcus, Desulfovibrio, Klebsiella, and Enterobacter. Berberis kansuensis C.K. Schneid. (Berberidaceae; Berberis Cortex) significantly improved T2DM, which was associated with reduced inflammation, by lowering the abundance of Escherichia–Shigella and Enterococcus and levels of LPS, TNF-α, IL-1β, and IL-6 in rats (Xu et al., 2021). Similarly, Dendrobium officinale Kimura & Migo (Orchidaceae; Dendrobii Officinalis Caulis) accelerated liver recovery in NAFLD mice by regulating the gut microbiota and gut–liver axis signaling pathways, including reducing the relative abundance of Desulfovibrio by 76.1% and the level of microbial LPS (Lei et al., 2019).

The gut microbiota can affect human health and disease by producing some bioactive metabolites such as SCFAs and bile acids (BAs). Next, we categorically summarized the major microbial metabolites regulated by herbal medicines and discussed their roles in metabolic diseases (Figure 3).

1) SCFAs

SCFAs, which are organic fatty acids composed of 1 to 6 carbon atoms, are a major category of microbial metabolites produced by the fermentation of bacteria in the cecum and colon (Koh et al., 2016). SCFAs, including acetate, butyrate, and propionate, have been shown to be beneficial in improving metabolic disorders. Butyrate and propionate can activate intestinal gluconeogenesis through the cAMP-dependent pathway, thereby reducing hepatic glucose production (De Vadder et al., 2014). Acetate inhibits liver lipid accumulation by upregulating several fatty acid oxidation-related proteins and the peroxisome proliferator-activated receptor α (PPARα) gene (Kondo et al., 2009). In addition, SCFAs can inhibit energy intake and appetite through the promotion of the production of satiety hormones. Acetate, butyrate, and propionate stimulate the release of peptide YY (PYY) and glucagon-like peptide-1 (GLP-1) from enteroendocrine L cells and promote the release of the satiety hormone leptin from adipose tissue (den Besten et al., 2015; Larraufie et al., 2018; Xiong et al., 2004). SCFAs can also ameliorate diet-induced insulin resistance, promote pancreatic insulin secretion, and improve insulin sensitivity (Canfora et al., 2015; Mandaliya and Seshadri, 2019). Furthermore, SCFAs can alleviate inflammation. For example, butyrate significantly reduced the production of TNF-α, monocyte chemoattractant protein 1 (MCP-1), and IL-6 and inhibited the activity of NF-κB (Ohira et al., 2013).

A number of herbal medicines have been shown to target the gut microbiome, increase SCFA levels, and thereby improve metabolic diseases. The administration of Xiexin decoction, Jinqi Jiangtang tablet, Gegen Qinlian decoction, Plantago asiatica L. (Plantaginaceae; Plantaginis Semen), and Cyclocarya paliurus (Batalin) Iljinsk. (Juglandaceae; Cylocaryae Paliuri Folium) can increase the abundance of several SCFA-producing bacteria and the level of SCFAs in T2DM rats or mice (Wei et al., 2018; Cao et al., 2019; Nie et al., 2019; Xu et al., 2020; Li et al., 2021c). Meanwhile, these herbal medicines reduce lipid levels, insulin resistance, and inflammation in the model animals. Similarly, Simiao decoction was reported to increase the abundance of SCFA-producing bacteria (Bifidobacterium and Faecalibaculum) and improve lipid metabolism and inflammation in NAFLD rodents (Han et al., 2021). Moreover, Quzhuo Tongbi decoction effectively alleviated gouty arthritis in mice by increasing the abundance of butyrate-producing bacteria and the levels of acetate, propionate, and butyrate (Wen et al., 2021).

2) Amino acids

The gut microbiota is also involved in the metabolism of amino acids in vivo. Studies have shown that the disturbance of microbial amino acid metabolism is related to the progression of metabolic diseases (Neis et al., 2015). Tryptophan can be metabolized into indole by some gut bacteria. Indole has a variety of beneficial effects on metabolic diseases; it can regulate the secretion of GLP-1 by intestinal endocrine L cells and thereby promote insulin secretion and improve glucose homeostasis (Pols et al., 2010; Du et al., 2022). Tyrosine can be converted into tyramine by the decarboxylase of the gut microbiota (Liu et al., 2020). Tyramine has been reported to improve glucose tolerance and glucose absorption (Morin et al., 2002; Visentin et al., 2003). In addition, arginine is converted into agmatine by the gut microbiota. Studies have confirmed the beneficial effects of arginine and its metabolite agmatine on metabolic diseases, including the promotion of insulin secretion, improving insulin resistance, and increasing cellular glucose uptake (Okazaki, et al., 2019; Du et al., 2022). Notably, some herbs have been found to regulate the metabolism of these amino acids in rodents with metabolic diseases. Sophora flavescens Ait. (Leguminosae; Sophorae Flavescentis Radix) significantly decreased the levels of fasted blood glucose, glycosylated serum protein, and glycosylated hemoglobin in T2DM rats by regulating gut bacteria and host-microbial metabolism, including increasing indole and tyramine levels (Shao et al., 2020). Similarly, the Naoxintong capsule improved hyperglycemia and hyperlipidemia in T2D rats, and its anti-diabetic mechanisms are related to the improvement of gut microbial disorders and regulation of tyrosine and tryptophan biosynthesis (Yan et al., 2020). Moreover, Shao et al. (2020) reported that S. flavescens Ait. (Leguminosae; Sophorae Flavescentis Radix) had good anti-diabetes effects on T2DM rats and might upregulate arginine and proline metabolism by reducing the abundance of Prevotella, Roseburia, and Faecalibacterium.

Branched-chain amino acids (BCAAs), such as leucine, isoleucine, and valine, have been shown to be closely associated with metabolic disease. Yu et al. (2022) found that a high intake of BCAAs was associated with a higher risk of T2DM. In addition, Zhou et al. (2019) demonstrated the pathogenic role of elevated BCAA levels in obesity-related insulin resistance. Some gut microbes, such as Streptococcus and Prevotella, are involved in BCAA biosynthesis and catabolism. One study found that berberine improved glycemic control and alleviated insulin resistance in HFD-fed mice, which was associated with altered gut microbiota in BCAA biosynthesis (Yue et al., 2019). Specifically, berberine treatment significantly reduced the relative abundance of BCAA-producing bacteria and serum BCAA levels. Meanwhile, berberine reduced the gut microbial genes involved in BCAA biosynthesis but enriched the genes involved in BCAA degradation and transport. In addition, Gao et al. (2018) evaluated the effects of the Qijian mixture on T2DM mice using metabonomics, gut microbiota, and network pharmacology. The results showed that the Qijian mixture significantly alleviated T2DM, and its anti-diabetic mechanisms were related to the regulation of gut microbiota and the reduction of several amino acids, including three BCAAs (leucine, isoleucine, and valine).

3) Bile acids

Cholesterol is converted to BAs in the liver under the action of cholesterol 7α-hydroxylase (CYP7A1) and CYP27A1. When BAs are secreted into the intestine, gut microbes can participate in their metabolism and maintain their homeostasis (Wahlstrom et al., 2016). For example, conjugated BAs, such as tauro-conjugated β-MCA (T-β-MCA) and glycoursodeoxycholic acid (GUDCA), can be converted into secondary BAs under the action of the bile salt hydrolase (BSH) of some gut bacteria, including Clostridium, Bacteroides, Lactobacillus, and Bifidobacterium (Jia et al., 2018b). BAs play an important role in glucose and lipid metabolism by acting on two receptors, namely, the farnesoid X receptor (FXR) and Takeda G protein-coupled receptor 5 (TGR5). Bile acid metabolism disorder has been shown to be closely related to the progression of metabolic diseases (Cai et al., 2022).

Several herbal medicines have been reported to regulate gut bacteria-related bile acid metabolism. The Tianhuang formula showed a lipid-lowering effect through the gut microbiota–T-β-MCA–FXR axis (Yang et al., 2022a). Specifically, it regulated gut microbes and inhibited their BSH activities, which thereby increased T-β-MCA levels and further inhibited intestinal FXR, which lead to increased bile acid synthesis and reduced lipid levels. Similarly, Lu et al. (2022) found that supplementation with Naoxintong capsule could effectively improve hyperlipidemia in HFD-fed rats, and its effect was related to the regulation of gut microbes and the decrease of BSH activity. In addition, S. baicalensis Georgi (Labiatae; Scutellariae Radix) improved hyperglycemia and hyperlipidemia in T2DM rats by regulating the interaction between gut microbiota and bile acid metabolism (Zhao et al., 2021). Specifically, the administration of S. baicalensis Georgi (Labiatae; Scutellariae Radix) significantly improved gut microbiota dysregulation (e.g., Lactobacillus and Feacalibaculum) and secondary BA metabolism disorders (e.g., GUDCA) by activating liver CYP7A1 and inhibiting FXR in the gut.

Phytochemicals in herbal medicines are generally low in bioavailability, but some metabolites that are transformed by the gut microbiota may exhibit better bioavailability than their precursors. Ellagitannins, for example, are a group of polyphenols found in pomegranates and Phyllanthus emblica L. (Euphorbiaceae; Phyllanthi Fructus) that have low bioavailability. However, their gut microbial metabolites, urolithins (urolithin A, B, C, and D), are more readily absorbed and have better bioavailability than ellagitannins (Espin et al., 2007; Espin et al., 2013). Interestingly, urolithins (e.g., urolithin A and B) have been shown to have beneficial therapeutic effects on metabolic diseases, including dyslipidemia, obesity, and diabetes (Raimundo et al., 2021; Toney et al., 2021; Chen et al., 2022). Thus, the therapeutic effect of pomegranates and P. emblica L. (Euphorbiaceae; P. fructus) on metabolic diseases may be attributed to urolithins produced by the gut microbiota rather than the polyphenols they contain. In addition, some metabolites produced by gut microbiota may have better bioactivity than their precursors. For instance, protopanaxadiol-type ginsenosides, including ginsenoside Rb1 in Panax ginseng, can be metabolized by the gut microbiota into compound K. There is increasing evidence that compound K has a good anti-diabetic effect (Jiang et al., 2014; Shao et al., 2015). In particular, several studies have shown that compound K has better antidiabetic, anti-inflammatory, and hepatoprotective activities than protopanaxadiol-type ginsenosides or ginsenoside Rb1 (Lee et al., 2005; Joh et al., 2011; Li et al., 2012). These findings help elucidate the key role of gut microbes in herbal treatments for metabolic diseases.

The toxicity or side effects of herbal medicines have aroused wide concern. The gut microbiota can convert some herbal compounds into less toxic metabolites. Aconitine is a well-known toxic ingredient found in Aconitum medicinal plants. Aconitine can be metabolized to benzoylaconine and lipoaconitine by human gut bacteria through deacetylation, demethylation, and esterification reactions. and thus reduce its toxicity (Kawata et al., 1999; Zhao et al., 2008; Zhang et al., 2015). Baicalin is the main active ingredient of S. baicalensis Georgi (Labiatae; Scutellariae Radix). Studies have shown that baicalin can be converted into baicalein by the gut microbiota, and baicalein has less toxicity on HepG2 cells than baicalin (Khana et al., 2012). Notably, baicalein has hepatoprotective, anti-dyslipidemia, anti-obesity, anti-inflammatory, and anti-diabetic activities (Fang et al., 2020; Rahimi et al., 2021). In addition to these direct transforming effects, some metabolites derived from gut bacteria also help reduce the toxicity of herbal medicines. Triptolide, which is a natural compound isolated from Tripterygium wilfordii Hook F (Celastraceae; Triptergii Radix et Rhizoma), has good anti-inflammatory and neuroprotective activities (Li et al., 2014). It also ameliorates hepatic lipogenesis, inflammation, and fibrosis in NAFLD (Huang et al., 2021). However, its clinical application is limited due to its severe hepatotoxicity. Recently, a study found that gut microbiota-derived propionate could ameliorate triptolide-induced hepatotoxicity (Huang et al., 2020). Specifically, propionate supplementation significantly reduces plasma transaminase, improves liver histology, and decreases liver and plasma malondialdehyde (MDA) levels.

To determine the therapeutic effect of herbal medicines on metabolic diseases and the key role of the gut microbiota, four experimental groups were created: a control group, a model group, a herbal medicine group, and an herbal medicine plus antibiotic group. After the experiment, common biochemical indicators, gut microbiota, and microbial metabolites in each group were detected and analyzed. Changes in the composition and diversity of the gut microbiota in each group were studied using 16S ribosomal RNA (rRNA) sequencing and/or metagenomics techniques. The 16S rRNA technique can be used to detect bacteria in samples based on polymerase chain reaction (PCR) amplification. The main challenges in using this technique are the lack of a standardized workflow and the difficulty in identifying bacteria at the species level. Metagenomics is a widely used technique that can identify microorganisms at the species and even strain level (Jovel et al., 2016). In addition, it can also perform a functional analysis of microbial communities. However, methodological biases and inter-individual differences must be considered during data interpretation. Recently, advanced techniques have been developed to study the composition and function of gut microbes. For example, metatranscriptomics can provide knowledge of the transcriptional profiles of microbial populations, which is beneficial in revealing the molecular activities of gut microbes and their regulatory mechanisms (Zhang et al., 2021). Similarly, metaproteomics is a powerful tool that can be used to study the functional activity of gut microbes by characterizing the complex composition of microbial proteins (Stamboulian et al., 2021).

Enzyme-linked immunosorbent assay (ELISA), real-time fluorescent quantitative PCR, Western blot, and immunohistochemistry methods can be used to detect biochemical markers related to metabolic diseases (e.g., fasting blood glucose, insulin, total cholesterol, triglycerides, alanine aminotransferase, and aspartate aminotransferase) in addition to intestinal barrier-related markers (e.g., occludin, ZO-1, and claudin-1) and inflammation-related markers (e.g., LPS, IL-1β, IL-6, and TNF-α). Metabolomics can be used to determine changes in microbial metabolites after drug administration (van Treuren and Dodd, 2020). Metabolomics analysis technology includes gas chromatography-mass spectrometry (GC-MS) and high-performance liquid chromatography-triple quadrupole mass spectrometry (HPLC-QqQ-MS), which can be used to efficiently perform qualitative and/or quantitative analysis of gut microbial metabolites. For example, GC-MS technology can accurately measure the levels of SCFAs produced by gut microbial fermentation, while HPLC-QqQ-MS technology can accurately determine the concentration of BAs. Additionally, mass spectrometry imaging (MSI) techniques, including matrix-assisted laser desorption/ionization MSI, desorption electrospray ionization MSI, and nanostructure-initiator MSI, are becoming attractive tools for the spatial visualization of microbial metabolites in tissues (Luan et al., 2017; Bauermeister et al., 2022). These MSI techniques help us understand the effects of herbal medicines on gut microbial metabolites in two or three dimensions. Additionally, bioinformatic methods can be used to study the correlation between the pharmacodynamic effects of herbal medicines and changes in gut microbes and their metabolites. In addition to antibiotic interventions, fecal microbiota transplantation (FMT) can also be used to identify the critical role of the gut microbiota in the herbal treatment of metabolic diseases.

After oral administration, herbal ingredients can be metabolized by gut microbes, and their metabolites are then absorbed into the circulation, producing pharmacological activity (van Duynhoven et al., 2011; Zierer et al., 2018). To determine whether the gut microbiota is involved in the metabolism of herbal ingredients, three experimental groups were created: a control group, a herbal medicine intervention group, and a herbal medicine intervention plus antibiotic group. After the experiment, animal biological samples, including feces or intestinal contents and serum, need to be collected. Feces or intestinal contents can be used to determine herbal metabolites after gut microbial transformation, while serum can be used to determine the absorption of these metabolites and whether bioavailability is improved. Due to the low levels of these metabolites in biological samples, high-sensitivity analytical instruments are needed for their detection. Ultra-high-performance liquid chromatography coupled with Orbitrap mass spectrometry (UPLC-Orbitrap-MS) and HPLC-QqQ-MS have been shown to accurately detect and identify metabolites of herbal phytochemicals in rat intestinal contents and serum samples (Du et al., 2020; Du et al., 2021). Furthermore, additional experimental validation is needed to determine whether the metabolites after gut microbial transformation have a better biological activity or lower toxicity than their precursors. In short, medium-pressure preparation liquid chromatography and high-speed countercurrent chromatography can be used for the targeted separation of specific metabolites. Then, cellular or animal experiments can be performed to compare the activity or toxicity of these metabolites with their precursors.