With a growing increase in changes in diet, lifestyle and other factors, there has been an explosive epidemic trend of metabolic syndrome characterized by obesity, type II diabetes, non-alcoholic fatty liver disease (NAFLD), and atherosclerosis (Ley et al., 2006; Turnbaugh et al., 2006; Zhang H. et al., 2010). As one of the most serious diseases threatening public health (Haslam and James, 2005; Moustafa and Froguel, 2013; Hall et al., 2015), metabolic syndrome can also cause other diseases, such as NAFLD, which can further develop into steatohepatitis, liver fibrosis, cirrhosis, liver cancer (Osborn and Olefsky, 2012; Yoshimoto et al., 2013; Ni et al., 2021), diabetes, or hypertension and cause multiple organ damage (Thandavarayan et al., 2011; Mozaffari-Khosravi et al., 2012; Wada et al., 2012). Although various drugs are available to treat metabolic syndrome and its complications in the clinical setting, there is presently no specific treatment for this condition (Zhang et al., 2012, 2015; Chang et al., 2015). For example, although metformin and acarbose can reduce blood sugar, its gastrointestinal side effects make it impossible for many patients to tolerate the maximum dose. While thiazolidinediones (TZDs) can reduce insulin resistance, it also causes weight gain and hypoglycaemia (Srinivasan et al., 2015). Therefore, exploring the methods and mechanisms for prevention and treatment of metabolic syndrome has important theoretical and clinical application value.

Metabolic syndrome is typically accompanied by low-level, systemic chronic inflammation, leading to impaired insulin activity and the development of metabolic abnormalities (Boulangé et al., 2016; Reilly and Saltiel, 2017). It has been reported that the activation of inflammation effector molecules contributes to the desensitization of insulin signaling pathways (Ribes-Navarro et al., 2019). At the molecular level, the activation of IκB kinase complex, extracellular signal-regulated protein kinases 1 and 2 (ERK1/2), and c-Jun N-terminal kinases (JNKs) in the inflammatory tissues of obese individuals can reduce insulin receptor substrate (IRS) protein tyrosine phosphorylation levels, leading to the attenuation of the insulin signal (Tanti et al., 2012). Furthermore, the production of cytokines, such as tumor necrosis factor alpha (TNF-α), in rodent and human visceral adipose tissues regulates insulin sensitivity by altering the expression levels of genes encoding insulin receptor 1 (IRS-1), glucose transporter GLUT4, and PPAR-α (Larsen et al., 2007; Tack et al., 2012). In turn, the loss of insulin sensitivity can trigger fasting hyperglycemia and increase liver lipid synthesis and fat accumulation (Saltiel and Kahn, 2001; Delarue and Magnan, 2007). These responses enhance chronic low-grade inflammation, induce the recruitment and activation of various mature immune cells (including mast cells, macrophages, and dendritic cells) and adipocytes in metabolic tissues, and ultimately, further augment the inflammatory response (Lumeng and Saltiel, 2011; Sell et al., 2012).

Various microorganisms inhabit the intestines of vertebrates. Cooperatively, the gut microbiota play an important role in the growth, development, metabolism, and immunity of the host (Turner, 2009; Levy et al., 2017; Cani et al., 2019; Seth et al., 2019; Ruff et al., 2020). Additionally, recent studies have demonstrated that gut microbes play an important role in the occurrence and development of metabolic diseases such as obesity, insulin resistance, atherosclerosis, and NAFLD (Cani et al., 2007). Gut microbiota help break down hard-to-digest food components, such as dietary fiber (Qin et al., 2010), and can also regulate the metabolism of bile acids, lipids, and amino acids in the host by exchanging metabolites with the host and participating in signal pathways and regulating the host gene expression and energy homeostasis (Bäckhed et al., 2004; Turnbaugh et al., 2006; Velagapudi et al., 2010). However, structural imbalance of the gut microbiota due to high-fat diet (HFD) intake may damage the intestinal barrier, thereby leading to increased levels of inflammatory factors and lipopolysaccharides (LPS) in the circulatory system, which can trigger metabolic inflammation, and subsequently, induce insulin resistance, obesity, and diabetes (Cani et al., 2008). Moreover, gut microbiota imbalance, particularly a reduction in the abundance of Bifidobacterium spp. that protect the intestinal barrier function and an increase in the abundance of pathogenic bacteria such as Desulfovibrio spp. that produce endotoxin and H₂S, can compromise the barrier function of the intestine (Zhang et al., 2008). This causes high levels of intestinal toxins such as LPS to enter the circulatory system of the host and induce chronic systemic low-grade inflammation, a condition known as “metabolic endotoxemia.” Eventually, long-term chronic inflammation can lead to a series of metabolic diseases such as decreased insulin sensitivity and NAFLD (Yin et al., 2008a).

Berberine hydrochloride exhibits a modest inhibitory effect on bacteria, fungi, and viruses. Berberine is mainly used clinically to treat intestinal infections and bacterial diarrhea (Hwang et al., 2003; Villinski et al., 2003). In addition, berberine can also be used to treat various metabolic diseases such as diabetes, hyperlipidemia, inflammation, and cardiovascular disease (Li et al., 2008; Yin et al., 2008a,b; Zhang et al., 2008; Zhang C. H. et al., 2010). The anti-inflammatory effects of berberine are primarily linked to the regulation of blood lipids, inhibition of oxidative stress, reduction of cell apoptosis, regulation of cell energy metabolism and inflammatory factor expression, and increased adiponectin levels (Wu and Wang, 2008; Wu et al., 2010). Furthermore, berberine can inhibit the activation and transcription of inflammatory cytokines by suppressing the activation of nuclear factor-kappa B (NF-κB), thus displaying activity against metabolic syndrome (Zhu et al., 2018). For example, berberine can reduce the expression of TNF-α, IL-6, and leptin in 3T3-L1 adipocytes (Choi et al., 2006). In addition, berberine plays a role in reducing insulin resistance by inhibiting the phosphorylation of IKKβ (Ser181) and the translocation of NF-κB (P65) into the nucleus (Yi et al., 2008). Since the oral bioavailability of berberine is very low and most of the drug is not absorbed and excreted intact in feces, a major part of the drug can directly interact with the gut microbiota (He et al., 2009). Studies have shown that the gut microbiota-derived metabolites of berberine can inhibit the entry of NF-κB (P65) into the nucleus by downregulating the expression of TLR4 and MyD88, which in turn inhibits the TLR4-MyD88-NF-κB signaling pathway. This indicates that colitis can be alleviated by reducing the expression of related inflammatory factors such as TNF-α, NF-κB, IL-6, and inducible nitric oxide synthase (iNOS; Li et al., 2020). It has been reported that berberine intervention in hamsters with HFD-induced obesity inhibits weight gain and fat deposition; reduces blood sugar, blood lipid, and cellular lipopolysaccharide levels; improves inflammation and insulin resistance; and relieves fatty liver disease and leads to increased production of short chain fatty acids (SCFA; Guo et al., 2019). This indicates that the metabolic regulation mechanism of berberine is likely closely related to the gut microbiota status, which represents a potential target for the development of therapeutic drugs. However, whether berberine mediates its anti-inflammatory and blood sugar and lipid lowering effects solely by adjusting the structure of the gut microbiota or by first directly regulating the expression of host pro-inflammatory proteins and activating macrophages and then acting on gut microbiota, is currently unclear. The answer to this question will help clarify the anti-inflammatory mechanism of berberine and provide theoretical guidance for the subsequent development of berberine and antibiotics, or berberine and probiotics for the treatment of metabolic syndrome.

Accordingly, this study used SPF-grade C57BL/6J male mice to construct an obese mouse model and conduct a systematic study on liver tissue pathology, inflammatory factor expression, and gut microbiota structure. We screened the targets of berberine in gut microbiota and showed that the molecular mechanism of berberine activity against metabolic syndrome involves the regulation of gut microbiota structure and expression of host inflammatory factors.

The inflammatory signaling pathway, wherein the transcription factor NF-κB participates, plays a central role in inflammation. As a cell signaling factor of the NF-κB signal transduction pathway, nitric oxide (NO) can participate in multiple physiological processes such as cellular communication, cell apoptosis, and the autoimmune response (Moncada and Palmer, 2001; Tripathi et al., 2007). When NO is expressed at high levels in cells, it can induce the production of inflammatory factors such as TNF-α, IL-1β, and IL-6. In turn, these pro-inflammatory factors can enhance the activity of iNOS and promote further production of NO, resulting in a vicious circle that intensifies the inflammatory response (Aktan, 2004; Nagy et al., 2007). Cox-2 is an inducible gene that is either not expressed or expressed in very low levels in most tissues or cells under normal conditions. However, extracellular stimuli, such as endotoxins and TNF-α, can induce the expression of COX-2 to help it exert its pathophysiological effects, which, in turn, can mediate an inflammatory cascade (Lanes et al., 2000). Reg3γ is an important factor in innate immunity and plays an important role in maintaining the dynamic balance among mammalian gut microbes and protecting the body from pathogen infection. Consistent with previous research (Xin et al., 2014; Wang Y. et al., 2019), we found that HFD successfully induced mice obesity and related metabolic disorders, as indicated by increased fat deposition, insulin resistance, and hepatic steatosis. These signs of metabolic syndrome were effectively improved after berberine treatment. In this study, berberine treatment not only reduced obesity but also considerably decreased the concentration of TNF-α, iNOS, Cox-2, and Reg3γ. The two common characteristics of NAFLD are lipid accumulation in liver cells and elevated production of pro-inflammatory factors by resident or infiltrating macrophages (Stojsavljevic et al., 2014). Liver macrophages, among which Kupffer cells derived from embryos are the most common, are dynamic participants in maintaining liver homeostasis (Guilliams et al., 2016). During the occurrence of NAFLD, liver macrophages are polarized, recruited, and proliferated (Meli et al., 2014). Macrophages are polarized across a broad spectrum, from a pro-inflammatory M1-like state, to an anti-inflammatory and pro-fibrogenic M2-like state (Menon and Sudheer, 2007). During infection with intracellular pathogenic microorganisms, macrophages typically show M1 polarization and express the corresponding pro-inflammatory factors to eliminate pathogenic microorganisms. However, in the presence of extracellular pathogenic microorganisms or allergens, macrophages undergo substitution activation (M2) and express Arg1 and other anti-inflammatory factors to promote tissue repair. Therefore, the precise regulation of macrophage polarization is very important for the treatment of metabolic diseases (Lin et al., 2019). The chemokines CCL2 and TNF-α, which are secreted by inflammatory liver cells, can cause Kupffer cells to proliferate and polarize toward M1 and express IL-6, IL-1β, TNF-α, and inducible nitric oxide synthase 2 (INOS2). Macrophages and T cells are important in the study of phagocytosis, cellular immunity, and molecular immunology (Gordon, 2016; Peterson et al., 2018). Moreover, these immune cells are closely related to the occurrence of liver fat metabolism diseases (Hill et al., 2014; Gordon and Pluddemann, 2017). Here, we observed an increased percentage of macrophages and T cells in the HFD group, which was suppressed by berberine treatment. In agreement with these findings, we observed that the expression of Arg1 in HFD mice increased significantly, compared to that in NCD mice, and was efficiently reduced by berberine administration. Studies have shown that excessive calorie intake, increased fat accumulation, and fat toxicity can activate the production of effector molecules (cytokines) and cells that are mainly involved in innate immunity (Cai et al., 2005; Cani et al., 2007). These products can enhance chronic low-grade inflammation and induce the recruitment and activation of various mature immune cells (including mast cells, macrophages, and dendritic cells) and fat cells in metabolic tissues, particularly in adipose tissue, thereby further enhancing the inflammatory response (Lumeng and Saltiel, 2011; Sell et al., 2012). In this study, we found that an HFD hindered fat metabolism in mice livers and increased the inflammatory response. After berberine intervention, these phenomena were significantly improved.

As a “second genome” for modulating the health phenotype of the superorganism host, gut microbiota are closely associated with host nutrition, metabolism, and immunity (Jia et al., 2008; Jin et al., 2020). It has been reported that an HFD disrupts gut microbiota in two ways: via diminishing the levels of gut barrier-protecting probiotics such as Bifidobacteria and promoting the growth of endotoxin producers. These changes can result in high levels of lipopolysaccharide (LPS) in the host blood, thereby causing inflammation, and consequently, obesity and insulin resistance (Cani et al., 2007). Studies have reported that the antibacterial activity of berberine in the intestine is related to its anti-obesity effects. Berberine can reduce the entry of intestinal endotoxins into the blood circulation, thereby reducing the level of low-grade inflammation and preventing the development of metabolic diseases (Zhang et al., 2012). Similarly, our results showed that the LPS producers Mucispirillum and Psychrobacter (Smith et al., 2017; Herp et al., 2019) were significantly enriched in the HFD group, but not in the HFDBBR group. This result suggested that berberine might play a role in suppressing the inflammatory response by reducing the content of LPS in the blood. It has been shown that Adlercreutzia is a marker of abnormal lipid metabolism (Wei et al., 2018). In addition, a multi-omics analysis of patients with liver steatosis revealed that steatosis is significantly negatively correlated with Coprococcus (Alferink et al., 2021). Moreover, it has been reported that Ruminococcus gnavus plays an important role in the upregulation of oxidative stress-related gene pathways in the gut microbiota of patients with inflammatory bowel disease (Hall et al., 2017). Our research showed that these bacterial groups were significantly enriched in the HFD group, but not in the HFDBBR group. These findings suggest that berberine can improve lipid metabolism and inflammation by regulating the gut microbiota. Studies have shown that Parabacteroides can significantly improve obesity, insulin resistance, lipid metabolism disorders, and NAFLD symptoms in HFD-induced obese mice and in an ob/ob obesity mouse model (Wang K. et al., 2019). Furthermore, it has been reported that Sutterella in the intestine of patients with constipation-autistic spectrum disorder is reduced significantly (Dan et al., 2020). Additionally, it has been demonstrated that Bacteroides regulates liver lipid metabolism and reduces NAFLD by activating the Bacteroides-folate-liver axis pathway (Qiao et al., 2020). Akkermansia muciniphila can ameliorate the negative effect of interferon-γ on glucose tolerance, which is negatively related to obesity, diabetes, cardiovascular diseases, and low-grade inflammation (Greer et al., 2016). We found that these bacterial species were significantly enriched in the HFDBBR group, which indicated that berberine could significantly improve obesity, insulin resistance, lipid metabolism disorders, and NAFLD symptoms.

Studies have shown that SCFA-producing bacteria are beneficial to the host, as they protect the mucosa from pathogen-induced damage, supply colonocyte nutrients, and mitigate inflammation (Maslowski et al., 2009; Filippo et al., 2010). Together with preventing HFD-induced insulin resistance, berberine also significantly increases the abundance of SCFA-producing flora and SCFA levels in the rat intestines (Zhang et al., 2012). Our results showed that the SCFA producers Butyricimonas, Eubacterium, and Clostridium (Sakamoto et al., 2014; Xie et al., 2017; Mukherjee et al., 2020) were significantly enriched in the HFDBBR group, suggesting that SCFA-producing bacteria likely play an important role in the efficacy of berberine. Thus, we infer that the selective modulations of specific gut microbial phylotypes, particularly the enrichment of SCFA-producing bacteria, may participate in berberine-mediated alleviation of metabolic diseases in the host. However, some SCFA-producing bacteria, such as Roseburia, Lachnospiraceae unidentified species, Coprococcus, and Clostridiales unidentified species, were also enriched in the group of HFD. Although the results of functional prediction showed that the relative abundance of SCFA-producing genes in gut microbiota were significantly reduced in the HFD group, and BBR treatment could alleviate the reduction, whether berberine can increase the production of SCFA by regulating gut microbiota needs to be determined by measuring the content of fecal SCFA in the future.

Notably, previous studies have shown that Odoribacter splanchnicus is negatively correlated with obesity (Etxeberria et al., 2017), Roseburia is negatively correlated with the risk score of healthy high-risk gene carriers (Imhann et al., 2018), and Oscillospira is negatively correlated with body mass index and inflammation (Konikoff and Gophna, 2016). In a study demonstrating that sucralose alters the gut microbiota of mice and promotes liver inflammation, the abundance of Dehalobacterium was significantly reduced (Bain et al., 2017) and AF12 was related to the beneficial effects of fecal bacteria transplantation (Lai et al., 2018). The baseline abundance of Coprobacillus correlated with COVID-19 severity (Zuo et al., 2020). Proteus can induce inflammatory responses in vivo and in vitro and may play a key role in the pathogenesis of Crohn’s disease (Zhang et al., 2020). Trabulsiella and Robinsoniella are the most abundant species in infants delivered by cesarean section and cecal contents and mucosa, respectively (Linnenbrink et al., 2013; Kuang et al., 2016). However, we found that Odoribacter, Roseburia, Oscillospira, Dehalobacterium, and AF12 were significantly enriched in the HFD group, whereas Coprobacillus, Proteus, Trabulsiella, and Robinsoniella were significantly enriched in the HFDBBR group. The reasons for this need to be further explored. At the phylum level, significant differences in the relative abundance of Firmicutes and Bacteroidetes in gut microbiota have been observed between host groups with obesity and lean phenotypes (Turnbaugh et al., 2006). In a rat study, significantly higher abundances of the Actinobacteria and Verrucomicrobia phyla were observed in the HFD group than in the NCD group, but this trend was completely reversed by berberine co-administration (Zhang et al., 2012). Similarly in this study, we found that HFD significantly increased the relative abundance of Firmicutes and Deferribacteres in the fecal flora of mice, but significantly reduced Bacteroidetes abundance. Furthermore, the relative abundance of Firmicutes in the fecal flora of mice in the HFD group was significantly reduced after berberine intervention, but was still higher than that in the normal control group. By comparison, the relative abundance of Bacteroidetes in the fecal flora was significantly increased, but was lower than that in the normal control group. A possible explanation for these observations may be that the Bacteroides phylum is positively correlated with LDL and HDL cholesterol levels, whereas the Firmicutes phylum is negatively correlated with total cholesterol and LDL and HDL cholesterol levels (Harata et al., 2017). This indicates that berberine intervention improves lipid metabolism. Previous studies on the Shannon–Wiener diversity index have revealed that berberine significantly decreases the bacterial diversity of the gut microbiota in both NCD- and HFD-fed rats (Zhang et al., 2012, 2015). However, our results indicated that HFD significantly increased the Shannon and Simpson indexes of the fecal flora of mice, but significantly reduced the number of OTU. These findings demonstrated that the composition of the fecal microbiota of mice treated with berberine and HFD was closer to that of mice fed on a normal diet. Furthermore, our study indicated that berberine can improve the gut microbiota disorder caused by HFD to some extent.

Although high-throughput sequencing of 16S rRNA gene has been widely used to analyze microbiota composition in various environments (Wu et al., 2017; Gong et al., 2021), it is difficult to distinguish microorganisms into species or strain level, which hinders a more detailed analysis of the function of gut microbes at species or strain level. Additionally, it is important to further exploit the gut microbiota-derived metabolites of berberine through metabonomics analysis. However, how to distinguish between gut microbiota-derived and host-derived metabolites is still challenging.

Our findings showed that berberine could improve chronic inflammatory metabolic syndrome induced by HFD to some extent and effectively improved the metabolism of high-fat food in mice, which correlated with the gut microbiota structure. This study may improve our understanding of host-microbe interactions during the treatment of metabolic diseases and provide useful insights into the action mechanism of berberine.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://bigd.big.ac.cn/gsa/browse/CRA004451, CRA004451.

The animal study was reviewed and approved by the Animal Ethics Committee of the Qilu Normal University.

JJL, JLL, and HS designed the experiments. JJL, JLL, and SW performed the animal experiments. JJL and JN conducted gut microbiota analysis. JJL, CZ, JJ, GW, and HS measured biochemical data and conducted immunohistochemical analysis. JJL and HS wrote the draft of the manuscript. All authors revised and approved the final version of the manuscript.

JN is an employee of Guangdong Meilikang Bio-Science Ltd., China. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.