The Bioavailability of Glycyrrhizinic Acid Was Enhanced by Probiotic Lactobacillus rhamnosus R0011 Supplementation in Liver Fibrosis Rats

Glycyrrhizinic acid (GL) is clinically applied to treat liver injury, and the bioavailability of orally administered GL is closely related to the gut microbiota. Therefore, the dysbiosis of gut flora in liver injury could significantly influence GL bioavailability. Still, less is known about the impact of probiotic supplementation on the bio-absorption process of oral medication, especially under a pathological state. Herein, probiotic L. rhamnosus R0011 (R0011) with a high viability in the harsh gastrointestinal environment was selected, and the effect of R0011 on the GL bioavailability in rats was investigated. Four groups of rats (n = 6 per group) were included: the normal group (N group), the normal group supplemented with R0011 (NLGG group), CCl4-induced chronic liver injury model (M group), and the model group supplemented with R0011 (MLGG group). Our results showed that liver injury was successfully induced in the M and MLGG groups via an intraperitoneal injection of 50% (v/v) CCl4 solution. Healthy rats supplemented with R0011 could increase the bioavailability of GL by 1.4-fold compared with the normal group by plasma pharmacokinetic analysis. Moreover, the GL bioavailability of MLGG group was significantly increased by 4.5-fold compared with the model group. R0011 directly improved gut microbial glucuronidase and downregulated the host intestinal drug transporter gene expression of multidrug resistance protein 2 (MRP2). More critically, R0011 restored the gut microbiota composition and regulated the metabolic function, significantly enhancing the microbial tryptophan metabolic pathway compared with the pathological state, which may indirectly promote the bioavailability of GL. Overall, these data may provide possible strategies by which to address the low bioavailability of traditional medicine through probiotic intervention.


Introduction
Oral administration is one of the most convenient and safest approaches for drug delivery [1], mainly applied in Chinese herbal medicine utilization [2]. The drug bioabsorption process is traditionally associated with drug physicochemical properties, including drug solubility, polarity, permeability, and stability [3]. Moreover, orally administered drugs are usually absorbed in the gut to enter the circulation system. The gut microbiota, as an "important organ", is closely correlated with the host metabolism, infection, immune system, brain function, tumors, etc. [4,5]. In addition, the gut microbiota has a chance to come into contact with oral drugs in the gut. Therefore, the gut microbiota acting as a bioreactor may play an important role in the bioabsorption of oral medicines [6].
The gut microbiota can directly transform the parent drugs into bioactive components using intestinal microbial enzymes. Zimmermann has proved that the chemical structures CCl 4 and olive oil were bought from Macklin Biochemical Technology Co. (Shanghai, China). Trypsin (1:250) and pepsin (1:10,000) were purchased from Solebo Technology Co. (Beijing, China). Alanine aminotransferase (ALT) and aspartate aminotransferase (AST) assay kits were obtained from Jiancheng Bioengineering Institute (Nanjing, China).
An total of 100 µL of six lactobacillus liquid cultures was added to 10 mL of simulated gastric juice and simulated intestinal juice. The simulated gastric juice and simulated intestinal juice were prepared according to the methods of the United States Pharmacopoeia. The mixtures of simulated gastric juice and bacterial liquid were incubated at 37 • C for 3 h, and then the viable bacteria were counted on MRS agar. The mixtures of simulated intestinal juice and bacterial liquid were incubated at 37 • C for 4 h, and then the viable bacteria were counted on MRS agar [25].
A dose-dependent (1-20 mg/mL of GL) study was designed to observe the effect of GL on the growth of L. rhamnosus R0011. The master liquor of GL was prepared at the concentration of 200 mg/mL and diluted to 0.5, 2.5, 5, 10, 15, and 20 mg/mL using ddH 2 O. After that, the sample was added to 96-well plates, with each well containing 100 µL of diluted bacterial solution (1 × 10 3 CFU/mL) and 100 µL GL solutions of different concentrations (1-20 mg/mL). For the control group, 100 µL of MRS broth medium and 100 µL diluted GL solutions of different concentrations were added. The bacterial turbidity was measured at OD 600 after 24 and 48 h of incubation at 37 • C.

Preparation of Animal Models
Male Sprague Dawley (SD) rats (180-220 g) were purchased from Qinglong Mountain experimental animal Co., Ltd. We used flat-bottomed cages with three rats per cage. These rats were fed standard laboratory food and water ad libitum, housed under controlled conditions (light/dark, every 12 h; temperature, 20-22 • C; humidity, 50 ± 10%), and acclimated for one week before the experimental operation. The body weights of rats in different groups were recorded every week. The experiment was approved by the Ethics Committee of Nanjing University of Chinese Medicine (ACU180305) on 12 March 2018. All experimental procedures were strictly performed according to the guidelines of the Experimental Animal Care and Ethics Committee of Nanjing University of Chinese Medicine.
The rats were randomly divided into four groups (n = 6 per group) as follows: the control group (N group), supplementation with L. rhamnosus R0011 in the control group (NLGG group), the CCl 4 -induced chronic liver injury model (M group), and supplementation with L. rhamnosus R0011 in the model group (MLGG group). The animal experimental protocol is shown in Figure 1D. Chronic liver injury was induced via an intraperitoneal injection of a 50% (v/v) CCl 4 solution in olive oil at a dose of 1 mL/kg [26]. In the normal group, CCl 4 was replaced with olive oil. Three intraperitoneal injections were administered in the first week, followed by twice weekly for six weeks. After the last intraperitoneal injection and a 12 h period of fasting, blood was taken from the orbital vein, and blood samples were centrifuged at 6000 rpm for 10 min to separate the plasma for analysis. The plasma levels of alanine aminotransferase (ALT) and aspartate transaminase (AST) were assessed using assay kits according to the manufacturer's directions. In the seventh week of the experiment, rats in the N and M groups were gavaged normal saline at a dose of Nutrients 2022, 14, 5278 4 of 18 1 mL, and rats in the other groups were gavaged L. rhamnosus R0011 at a dose of 1 mL (1 × 10 9 CFU/mL) once daily for 1 week. To avoid direct contact with rats' skin or urine, fresh feces of all rats from the anus were collected into a sterile tube after the last L. rhamnosus R0011 and saline treatment [27]. After collection, the feces were quickly stored at −80 • C as early as possible. ministered in the first week, followed by twice weekly for six weeks. After the last intraperitoneal injection and a 12 h period of fasting, blood was taken from the orbital vein, and blood samples were centrifuged at 6000 rpm for 10 min to separate the plasma for analysis. The plasma levels of alanine aminotransferase (ALT) and aspartate transaminase (AST) were assessed using assay kits according to the manufacturer's directions. In the seventh week of the experiment, rats in the N and M groups were gavaged normal saline at a dose of 1 mL, and rats in the other groups were gavaged L. rhamnosus R0011 at a dose of 1 mL (1 × 10 9 CFU/mL) once daily for 1 week. To avoid direct contact with rats' skin or urine, fresh feces of all rats from the anus were collected into a sterile tube after the last L. rhamnosus R0011 and saline treatment [27]. After collection, the feces were quickly stored at −80 °C as early as possible.
After pretreatment with L. rhamnosus R0011 or normal saline for 1 week, rats were gavaged a 10 mg/kg dose of GL dissolved in normal saline after fasting with free access to water for 12 h. Blood samples (each 0.5 mL) were collected from the ophthalmic artery plexus of each rat into heparin sodium tubes 0.17, 0.33, 0.5, 0.75, 1, 2, 4, 6, 8, 12, 24, 48, and 72 h after oral administration [26,28]. The plasma was gathered by centrifugation at 6000 rpm for 10 min and stored at −80 • C until further analysis. After blood from the ophthalmic artery plexus was collected, all rats were euthanized with 2% sodium pentobarbital (2 mL/kg, intraperitoneal injection), and then the tissues of the large and small intestine were removed immediately and frozen at −80 • C for qRT-PCR analysis.
Plasma concentration-time profiles were used to evaluate the peak plasma concentration and time of maximum plasma concentration (T max ) of GL or GA. Other kinetic parameters of the plasma samples were determined using The Drug and Statistics v3.0 (Drug clinical Research Center of Shanghai University of Traditional Chinese Medicine , Shanghai, China )with a non-compartmental statistical model.

Quantitative Real-Time PCR (qRT-PCR)
Total RNA was isolated from intestinal tissues using RNA-easy TM Isolation Reagent and reverse-transcribed into cDNA using the HiScript ® II Q RT SuperMix kit with Gdna Eraser according to the manufacturer's protocol. Finally, quantitative RT-PCR was executed on a CFX96 real-time PCR detection system (Bio-Rad, Hercules, CA, USA) using the ChamQ SYBR qPCR Master Mix kit. The comparative cycle threshold method quantified the foldchange of the target genes from the internal standard GAPDH. The sequences of the primers are presented in Table 1.

β-Glucuronidase Enzyme Activity Assay in the Feces and Intestinal Tissues
The extraction of crude enzymes from the feces and intestines followed the method of Kim et al. [29]. For the activity assay of fecal and intestinal β-glucuronidase, the reaction mixture (total volume of 0.5 mL), which contained 0.1 mL of 1 mmol/L p-nitrophenyl-β-Dglucuronide, 0.3 mL of 0.05 mol/L phosphate buffer (pH 7.2-pH 7.4), and 0.1 mL of the feces and intestinal supernatant, was incubated at 37 • C for 30 min. The reaction was terminated by the addition of 0.5 mL of 10 mmol/L NaOH, and the absorbance was measured at 405 nm (UV-Vis recording spectrophotometer, catalog UV-2401PC; Shimadzu). Enzyme activity was indicated as the amount required to catalyze the formation of 1.0 nmol pnitrophenol/h under standard assay conditions. Specific activity was defined as millimoles per hour per gram of wet feces.

Fecal Metabolic Profiling by UPLC-MS/MS
An total of 50 mg of each fecal sample was pretreated with 400 µL methanol: water (4:1, v/v) solution. The mixture was processed with a Wonbio-96c high-throughput tissue Nutrients 2022, 14, 5278 6 of 18 disruptor (Shanghai Wanbo Biotechnology Co., Ltd., Shanghai, China) at 50 Hz for 6 min, followed by vortexing for 30 s and ultrasound for 30 min at 5 • C. All the samples were placed at −20 • C to precipitate proteins, and then the supernatant was transferred to sample bottles for LC-MS/MS analysis after centrifugation at 13,000× g for 15 min at 4 • C.
After UPLC-TOF/MS analyses, the raw data were imported to Progenesis QI 2.3 (Waters, Nonlinear Dynamics, Miford, MA, USA) for peak detection and alignment. Partial least squares discriminant analysis (PLS-DA) was performed in "ropls". The potential biomarkers were selected using the variable importance in projection (VIP > 1) and statistically significant differences (p < 0.05) through a two-tailed t-test. Pathway enrichment analysis of the biomarkers was performed on scipy stats (Python package) (https: //docs.scipy.org/doc/scipy/, accessed on 8 August 2021) based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) library (https://www.kegg.jp/kegg/pathway.html, accessed on 7 August 2021). The results were analyzed using the Majorbio Cloud Platform (https://www.majorbio.com, accessed on 8 August 2021).

Statistical Analysis
Graphpad prism 9.0 (GraphPad Software, San Diego, CA, USA, https://www.grap hpad.com) was used for statistical analyses. All data were analyzed via Student's t-test or ANOVA. The results are expressed as the standard error of the mean (±SEM), and p < 0.05 was considered to be statistically significant. Spearman correlation analysis was used in the correlation analysis.

In Vitro Screening of Lactobacillus Based on Tolerance to the Gastrointestinal Environment
Gastrointestinal tolerance is a vital requirement for practically applying probiotics, and bacterial strains with high survival rates in the harsh gastrointestinal tract environment are essential in order to exhibit beneficial effects [32,33]. In this study, several available Lactobacillus probiotics, including Lactobacillus paracasei 2001-12, Lactobacillus salivarius JCM 1231, Lactobacillus plantarum NCIMB 8826, Lactobacillus acidophilus JCM 1132, Lactobacillus bulgaricus KLDS 1.0207, and Lactobacillus rhamnosus R0011, were firstly selected and screened for survival ability in the artificially simulated gastrointestinal environment. As shown in Figure 1A,B, all six Lactobacilli showed a certain survival ability. However, L. rhamnosus R0011 showed a better performance, with a 42.87% survival rate in the simulated gastric fluid and a 51.67% survival rate in the simulated intestinal fluid. Additionally, GL was reported to have antibacterial bioactivity; therefore, the GL resistance of L. rhanmosus R0011 was also investigated. Our results showed that L. rhanmosus R0011 possessed a solid tolerance to GL, and the highest concentration of GL reached 20 mg/mL, as shown in Figure 1C. Overall, L. rhamnosus R0011 maintained a high viability under the challenging environment, which could guarantee further application.

Effects of L. rhamnosus R0011 on Pharmacokinetics of Glycyrrhizinic Acid (GL)
To investigate the clinical effect on GL pharmacokinetics, L. rhamnosus R0011 was applied under normal and liver injury conditions in Figure 1D. CCl 4 was used to induce chronic liver fibrosis in rats. Body weight was an essential indicator of rat health status. Figure 2A showed a significant decrease in body weight in M and MLGG groups compared with the normal group since the fifth week. The liver functional biochemical analysis of ALT and AST indicated that liver injury was successfully induced in the CCl 4 treatment group ( Figure 2B). The effect of L. rhamnosus R0011 on the pharmacokinetics of orally administered GL was further investigated over a week of intragastric administration. The plasma concentration levels of GL and its metabolite GA were determined in the normal and model groups. Figure 2 and Table 2 summarize the pharmacokinetic parameters of GL and its metabolite GA in the different experimental groups. The AUC value of GL was observed to have no change after L. rhamnosus R0011 administration (AUC = 2.98 ± 1.14 µg/mL h) compared with the normal group (AUC = 2.66 ± 0.56 µg/mL h), as shown in Figure 2C. In contrast, the AUC value of GA significantly increased by 1.44 times compared with the normal group after the L. rhamnosus R0011 intervention, as shown in Figure 2D. concentration of GA was significantly increased after probiotic intervention, which suggests that L. rhamnosus R0011 could dramatically promote the bioavailability of GL, especially under the pathological state.

Effects of L. rhamnosus R0011 on Intestinal Biotransformation of Glycyrrhizinic Acid (GL)
As glucuronidase is capable of converting GL to GA through the deglycosylation process [34], we examined glucuronidase activity in intestinal tissues as well as in feces. Figure 3A shows that host intestinal metabolic enzyme activities, whether from the tissue extracts of the small intestine or large intestine, were not significantly changed after L.  Furthermore, the addition of L. rhamnosus R0011 in the model group was also investigated. A similar tendency was observed in the MLGG group. The AUC value of GL was just slightly improved compared with the model group, as shown in Figure 2E. However, remarkably increased plasma levels of GA (AUC = 181.62 ± 138.60 µg/mL h) were observed after L. rhamnosus R0011 supplementation in comparison with the model group (AUC = 39.83 ± 20.49 µg/mL h) ( Figure 2F). Our results showed that the accumulative concentration of GA was significantly increased after probiotic intervention, which suggests that L. rhamnosus R0011 could dramatically promote the bioavailability of GL, especially under the pathological state.

Effects of L. rhamnosus R0011 on Intestinal Biotransformation of Glycyrrhizinic Acid (GL)
As glucuronidase is capable of converting GL to GA through the deglycosylation process [34], we examined glucuronidase activity in intestinal tissues as well as in feces. Figure 3A shows that host intestinal metabolic enzyme activities, whether from the tissue extracts of the small intestine or large intestine, were not significantly changed after L. rhamnosus R0011 administration. In contrast, the fecal glucuronidase activities were elevated considerably after L. rhamnosus R0011 treatment both in the normal and model groups ( Figure 3B). These results indicate that supplementation with L. rhamnosus R0011 may directly participate in GL biotransformation, which promotes the bioavailability of GL. Treatment with L. rhamnosus R0011 could improve the microbial enzyme activities rather than modulating the host intestinal enzyme activities.

Effect of L. rhamnosus R0011 on the Gene Expression of Intestinal Drug Transporters
Intestinal membrane transporters play critical roles in the oral bioavailability of numerous drugs. Therefore, the small-intestinal mRNA levels of drug export pumps, including MDR1, MRP2, and BCRP, related to GA export were investigated after the probiotic intervention. Our results showed that the relative gene expression level of MRP2 was prominently elevated in the model group compared with the normal group, and the trend was reversed after supplementation with L. rhamnosus R0011 ( Figure 4A). Additionally, L. rhamnosus R0011 slightly reduced the relative gene expression of transporters BCRP and MDR1 ( Figure 4B,C). These findings show that the downregulation of small-intestinal exporter gene expression by L. rhamnosus R0011 treatment might directly contribute to the absorption of GL and GA in vivo. rhamnosus R0011 administration. In contrast, the fecal glucuronidase activities were elevated considerably after L. rhamnosus R0011 treatment both in the normal and model groups ( Figure 3B). These results indicate that supplementation with L. rhamnosus R0011 may directly participate in GL biotransformation, which promotes the bioavailability of GL. Treatment with L. rhamnosus R0011 could improve the microbial enzyme activities rather than modulating the host intestinal enzyme activities.

Effect of L. rhamnosus R0011 on the Gene Expression of Intestinal Drug Transporters
Intestinal membrane transporters play critical roles in the oral bioavailability of numerous drugs. Therefore, the small-intestinal mRNA levels of drug export pumps, including MDR1, MRP2, and BCRP, related to GA export were investigated after the probiotic intervention. Our results showed that the relative gene expression level of MRP2 was prominently elevated in the model group compared with the normal group, and the trend was reversed after supplementation with L. rhamnosus R0011 ( Figure 4A). Additionally, L. rhamnosus R0011 slightly reduced the relative gene expression of transporters BCRP and MDR1 ( Figure 4B,C). These findings show that the downregulation of small-intestinal exporter gene expression by L. rhamnosus R0011 treatment might directly contribute to the absorption of GL and GA in vivo.   The results of the data statistics are expressed as the mean ± SEM (n = 6). One-way ANOVA was used for the analysis (* p < 0.05).

Effect of L. rhamnosus R0011 on the Gene Expression of Intestinal Drug Transporters
Intestinal membrane transporters play critical roles in the oral bioavailability of numerous drugs. Therefore, the small-intestinal mRNA levels of drug export pumps, including MDR1, MRP2, and BCRP, related to GA export were investigated after the probiotic intervention. Our results showed that the relative gene expression level of MRP2 was prominently elevated in the model group compared with the normal group, and the trend was reversed after supplementation with L. rhamnosus R0011 ( Figure 4A). Additionally, L. rhamnosus R0011 slightly reduced the relative gene expression of transporters BCRP and MDR1 ( Figure 4B,C). These findings show that the downregulation of small-intestinal exporter gene expression by L. rhamnosus R0011 treatment might directly contribute to the absorption of GL and GA in vivo.

Structural Modulation of Gut Microbiota by L. rhamnosus R0011 Intervention
Orally administered probiotics could also change the gut microbiota composition. To obtain further insights into the impact of L. rhamnosus R0011 on the structure of the gut microbiota, 16S rRNA sequencing of the fecal samples was performed. A total of 1,005,069 high-quality sequences were obtained from 24 samples, with an average of 41,878 reads per sample. According to the tendency of individual rarefaction curves and Shannon-Wiener curves ( Figure 5A,B), sequencing data were reasonable and could reflect the microbial information in all samples. The α diversity, as represented by the Chao1 and Shannon indexes, is usually used for evaluating the community richness and diversity. In this study, the Chao index ( Figure 5C) and the Shannon index ( Figure 5D) showed no significant difference between any two groups. The NLGG group just slightly increased the gut microbiota richness compared with the N group. Principal coordinate analysis (PCoA) and nonmetric multidimensional scaling (NMDS) analysis were performed to evaluate the β diversity among the four groups. As shown in Figure 5E,F, the OTU-based PCoA and NMDS showed differences between microbial communities of the four groups. Nevertheless, most symbols representing the N and NLGG groups were located close to each other. The structural shift among the four groups indicated that L. rhamnosus R0011 could be more inclined to change the gut microbiota composition of the M group. Additionally, we profiled the composition of the gut microbiota at the phylum level in each group through 16S rRNA sequencing analysis. As depicted in Figure 6A, the relative abundance of Bacteroidetes and Actinobacteria at the phylum level was elevated in the model group compared with the normal group, while Firmicutes was reduced to some extent. The disorder of the gut microbiota composition was repristinated after the oral administration of L. rhamnosus R0011 under the pathological state. The genus-level distribution of the four groups is illustrated in Figure 6B, showing that the relative abundance of Clostridium in the NLGG and MLGG groups decreased after oral administration of L. rhamnosus R0011. As shown in Figure 6C, the relative abundance of Ruminococcus and Anaerostipes markedly increased in the MLGG group compared with the M group. The abundance of Lactobacillus increased slightly after the gavage of L. rhamnosus R0011 under Additionally, we profiled the composition of the gut microbiota at the phylum level in each group through 16S rRNA sequencing analysis. As depicted in Figure 6A, the relative abundance of Bacteroidetes and Actinobacteria at the phylum level was elevated in the model group compared with the normal group, while Firmicutes was reduced to some extent. The disorder of the gut microbiota composition was repristinated after the oral administration of L. rhamnosus R0011 under the pathological state. The genus-level distribution of the four groups is illustrated in Figure 6B, showing that the relative abundance of Clostridium in the NLGG and MLGG groups decreased after oral administration of L. rhamnosus R0011. As shown in Figure 6C, the relative abundance of Ruminococcus and Anaerostipes markedly increased in the MLGG group compared with the M group. The abundance of Lactobacillus increased slightly after the gavage of L. rhamnosus R0011 under pathological conditions. In addition, gut microbiota such as unclassified_f__Erysipelotrichaceae, Clostridium_sensu_stricto _1, and Coriobacteriaceae_UCG-002 were massively enriched in the model group and decreased after L. rhamnosus R0011 treatment. On the other hand, supplementation with L. rhamnosus R0011 will not hugely disturb the richness and composition of a healthy gut microbiota, which could maintain the stability of the gut microbiota in a healthy state. pathological conditions. In addition, gut microbiota such as unclassified_f__Erysipelotrichaceae, Clostridium_sensu_stricto _1, and Coriobacteriaceae_UCG-002 were massively enriched in the model group and decreased after L. rhamnosus R0011 treatment. On the other hand, supplementation with L. rhamnosus R0011 will not hugely disturb the richness and composition of a healthy gut microbiota, which could maintain the stability of the gut microbiota in a healthy state.  (C) Alterations in the relative abundance of bacterial taxa. Data are presented as the mean ± SEM (n = 6). One-way ANOVA was used to evaluate differences between groups (* p < 0.05).

Effect of L. rhamnosus R0011 on the Modification of Microbial Metabolite Pathways
The liquid chromatography-mass spectrometry (LC-MS) analytical method was applied to study the fecal metabolome, which partially reflects the functional features of the gut microbiota under probiotic treatment. We built partial least squares discriminant analysis (PLS-DA) models to evaluate the cluster tendencies among the four groups. As shown in Figure 7A,B, distinct clustering of the N, NLGG, M, and MLGG groups was observed in both positive and negative ion modes. In the negative ion mode, the values of R 2 Xcum, R 2 Ycum, and Q 2 cum were 0.316, 0.546, and 0.517. At the same time, values of 0.576, 0.988, and 0.821 were detected in the positive ion mode, indicating a good classification and prediction ability, which could subsequently be used for analyzing the differences among the four groups. Volcano plots ( Figure 7C,D) were used to screen biomarkers in both positive and negative ion modes. The variables with VIP > 1 and statistically significant differences (p < 0.05) were considered as potential biomarkers.

Effect of L. rhamnosus R0011 on the Modification of Microbial Metabolite Pathways
The liquid chromatography-mass spectrometry (LC-MS) analytical method was applied to study the fecal metabolome, which partially reflects the functional features of the gut microbiota under probiotic treatment. We built partial least squares discriminant analysis (PLS-DA) models to evaluate the cluster tendencies among the four groups. As shown in Figure 7A,B, distinct clustering of the N, NLGG, M, and MLGG groups was observed in both positive and negative ion modes. In the negative ion mode, the values of R 2 Xcum, R 2 Ycum, and Q 2 cum were 0.316, 0.546, and 0.517. At the same time, values of 0.576, 0.988, and 0.821 were detected in the positive ion mode, indicating a good classification and prediction ability, which could subsequently be used for analyzing the differences among the four groups. Volcano plots ( Figure 7C,D) were used to screen biomarkers in both positive and negative ion modes. The variables with VIP ˃ 1 and statistically significant differences (p < 0.05) were considered as potential biomarkers.  Based on the result, 14 biomarkers were further analyzed among the model group and the L. rhamnosus R0011 treatment group, as shown in Figure 7E. It was observed that the relative abundance of 12 compounds (Adenosine diphosphate ribose, Cytosine, N -Fomylkynurenine, 2-Formaminobenzoylacetate, Xanthurenic acid, 6-Hydroxykynurenic acid, Prostaglandin E2, Tetrahydrocortisone, Phenylacetylglycine, 16-Hydroxy hexadecanoic acid, Guanosine, L-Carnitine) was significantly increased with the L. rhamnosus R0011 supplementation. The identified metabolites were annotated to eight related KEGG metabolic pathways, which were displayed in a metabolome map ( Figure 7F). After one week of probiotic intervention, the most significant differences were found in tryptophan metabolism. Evidence has shown that tryptophan metabolites profoundly affect the gut microbial composition, microbial metabolism, and host immune system-intestinal microbiota interactions [35]. Numerous tryptophan metabolites, such as xanthurenic acid and kynurenic acid, have been reported to be aryl hydrocarbon receptor (AhR) ligands. AhR is abundantly present at mucosal surfaces, which are activated to improve intestinal homeostasis and may indirectly affect the bioabsorption of oral drugs [36].
To determine the functional correlation between the intestinal microbiota and their metabolites, correlation analysis was conducted by calculating Spearman's correlation coefficient. The resulting metabolic association heatmap ( Figure 8) revealed potential correlations between metabolic profiles and the gut microbiome. The correlation analysis showed that Anaerostipes had positive associations with Adenosine diphosphate ribose and N'-Formylkynurenine (p < 0.05), whereas Clostridium_sensu_stricto _1 and unclassi-fied_f__Erysipelotrichaceae were negatively associated with L-Carnitine (p < 0.05). Additionally, Ruminococcus was positively correlated with the elevation in N'-Formylkynurenine and Phenylacetylglycine (p < 0.05). Of particular note is that Lactobacillus, which increased in the MLGG group, was positively correlated with Xanthurenic acid (p < 0.05).
Based on the result, 14 biomarkers were further analyzed among the model group and the L. rhamnosus R0011 treatment group, as shown in Figure 7E. It was observed that the relative abundance of 12 compounds (Adenosine diphosphate ribose, Cytosine, N'-Fomylkynurenine, 2-Formaminobenzoylacetate, Xanthurenic acid, 6-Hydroxykynurenic acid, Prostaglandin E2, Tetrahydrocortisone, Phenylacetylglycine, 16-Hydroxy hexadecanoic acid, Guanosine, L-Carnitine) was significantly increased with the L. rhamnosus R0011 supplementation. The identified metabolites were annotated to eight related KEGG metabolic pathways, which were displayed in a metabolome map ( Figure 7F). After one week of probiotic intervention, the most significant differences were found in tryptophan metabolism. Evidence has shown that tryptophan metabolites profoundly affect the gut microbial composition, microbial metabolism, and host immune system-intestinal microbiota interactions [35]. Numerous tryptophan metabolites, such as xanthurenic acid and kynurenic acid, have been reported to be aryl hydrocarbon receptor (AhR) ligands. AhR is abundantly present at mucosal surfaces, which are activated to improve intestinal homeostasis and may indirectly affect the bioabsorption of oral drugs [36].
To determine the functional correlation between the intestinal microbiota and their metabolites, correlation analysis was conducted by calculating Spearman's correlation coefficient. The resulting metabolic association heatmap ( Figure 8) revealed potential correlations between metabolic profiles and the gut microbiome. The correlation analysis showed that Anaerostipes had positive associations with Adenosine diphosphate ribose and N'-Formylkynurenine (p < 0.05), whereas Clostridium_sensu_stricto _1 and unclassi-fied_f__Erysipelotrichaceae were negatively associated with L-Carnitine (p < 0.05). Additionally, Ruminococcus was positively correlated with the elevation in N'-Formylkynurenine and Phenylacetylglycine (p < 0.05). Of particular note is that Lactobacillus, which increased in the MLGG group, was positively correlated with Xanthurenic acid (p < 0.05).

Discussion
Growing evidence has indicated that the gut microbiota plays a crucial role in the pharmacokinetics of orally administered drugs [37]. Recent research has shown that the gut microbiome contributed more than 50% of the metabolism of oral medicines [38], and over 60% of the drug response is closely related to the individual gut microbiota [39]. The concept of pharmacomicrobiomics has attracted considerable attention [40]. Gut microbes can produce various enzymes directly involved in the biotransformation of oral drugs and thereby influence the bioavailability of these drugs. Moreover, the gut microbiota can regulate the host gene expression and modulate the host metabolic function, affecting the drug transport and bioabsorption process. Therefore, the gut microbiota has become a promising target for improving the bioavailability and efficiency of oral drugs.
Most traditional Chinese herbal medicines are clinically applied via oral administration [2,41], and their absorption is inevitably affected by the intestinal microbiota. GL is a well-known pentacyclic triterpenoid glycoside obtained from the traditional Chinese medicine licorice, which has been used as a hepatoprotective agent for chronic hepatitis [42]. GL is poorly absorbed after oral administration due to its polar characteristics, which leads to a low bioavailability and limits its clinical efficacy. In our previous study, L. murine was isolated from healthy rats. Supplementation with L. murine significantly improved the bioavailability of GL [26]. However, the specific effect of L. murine on orally administered GL in vivo has not been explored. Moreover, due to the potential pathogenicity, L. murine could not be further used in practical applications. Probiotics have been wildly accepted in our daily life and medical use due to their beneficial effect [43]. However, the effect of probiotics on GL bioavailability has not been investigated. L. rhamnosus is one of the most widely studied probiotics, and it has antiallergy [44], antioxidative [45], and gut microflora balancing effects [46]. L. rhamnosus R0011 is a food-grade probiotic with a clear genetic background, which has been proven to positively affect human clinical use. It has been suggested that L. rhamnosus can produce glycosidases and thus hydrolyze glycosidic compounds [47]. Moreover, our results showed that L. rhamnosus R0011 could tolerate the harsh intestinal environment and resist the antibacterial activity of GL, which meets the essential requirement for in vivo research.
In the intestine, GL is deglycosylated into glycyrrhetinic acid (GA) by microbial βglucuronidase. In the current study, fecal glucuronidase activity was significantly increased in the L. rhamnosus R0011-treated group. Recently, Lactobacillus has been proven to play an important role in the bioconversion of glycoconjugated phytochemicals by phosphotransferase systems (PTS) in the uptake of carbohydrates and phosphor-glucosidases for the deglycosylation process [48]. Rossi et al. reported that probiotic Lactobacillus can produce several glycosyl-hydrolases, and aglycone release can be accelerated by glucosidase [49]. The experimental results support the possibility that L. rhamnosus R0011 directly participates in the GL biotransformation process. Accumulated evidence has suggested that drug export pumps, including MDR1, MRP2, and BCRP in the small intestine, play a pivotal role in GL bioavailability. Furthermore, the intestinal drug transporter gene expression of MPR1 and MRP2 was found to be upregulated under inflammatory conditions [50]. Our results showed that supplementary L. rhamnosus R0011 could directly downregulate the relative gene expression of MRP2, which may reduce the export of GA that contributes to the improvement of the GA concentration in the blood. Strong evidence exists showing that L. rhamnosus reveals pili containing a human mucus-binding protein that promotes strong adhesion interactions with host intestinal epithelial cells [51,52]. Specifically, L. rhamnosus R0011 can effectively adhere to intestinal epithelial cells and block the adhesion of various pathogens [53,54]. In this context, we propose that orally administered L. rhamnosus R0011 could attach to the GI tract and regulate the host intestinal transporter gene expression.
Intervention with probiotics could further modulate the gut microbiota composition and regulate the metabolic function of the gut microbiota. Our results showed that the relative abundances of Ruminococcus and Anaerostipes belonging to Lachnospiraceae were significantly increased compared with the model group after the L. rhamnosus R0011 intervention. T. Akno reported that the gut microbiota Ruminococcus sp. PO1-3 can directly metabolize GL to GA by producing glucuronidase [55]. A functional metagenomic study also revealed that Lachnospiraceae contains genes encoding glycosidases [28,56,57]. Meanwhile, oral administration of L. rhamnosus R0011 can also recover the reduction in Lactobacillus caused by liver injury, which may also benefit GL metabolism. In addition, our results showed that the densities of Coriobacteriaceae, Clostridium_sensu_stricto-1, and unclassified_f__Erysipelotrichaceae, which are closely related to liver fibrosis and epithelial inflammation, were remarkably decreased after L. rhamnosus R0011 administration [58,59]. In short, we hypothesized that oral administration of probiotic L. rhamnosus R0011 could recruit more gut microbiota to participate in the GL biotransformation. On the other hand, L. rhamnosus R0011 can regulate the gut microbiota dysbiosis and restore the intestinal balance.
According to the analysis of fecal metabolites, it was discovered that the metabolite profile in the pathological state was remarkably changed after supplementation with L. rhamnosus R0011. The different metabolites were mainly involved in lipid metabolism (cutin suberine and wax biosynthesis, alpha-Linolenic acid metabolism, or arachidonic acid metabolism) and nucleotide metabolism (purine metabolism or pyrimidine metabolism). In particular, tryptophan metabolism was significantly increased after the probiotic intervention. Tryptophan metabolites such as kynurenine and xanthurenic acid have been reported to bind and activate the aryl hydrocarbon receptor (AhR), which is closely associated with intestinal barrier integrity and mucosal immune homeostasis [36]. It was reported that tryptophan metabolism was impaired under chronic inflammatory diseases [60]. Supplementation with Lactobacillus reuteri, with a high AhR ligand-production capacity, could improve gut metabolic homeostasis [61]. Additionally, a state of intact intestinal architecture supports drug absorption [62]. Based on our results, we suggest that supplementation with L. rhamnosus R0011 may repair intestinal homeostasis by modulating tryptophan metabolism, which could indirectly improve the bioavailability of GL.
As we know, GL is mainly used in chronic liver injury patients, so L. rhamnosus R0011 was applied in normal and liver injury rats to explore the effect of probiotics on GL bioavailability. Supplementation with L. rhamnosus R1100 increased the AUC value of GA by 1.4-fold compared with the normal group. Moreover, the AUC value of GA was significantly increased by 4.5-fold compared with the model group by adding L. rhamnosus R0011. The different effects of L. rhamnosus R0011 on the healthy and pathological states also attracted our attention.
Based on our experimental data, the drug transporter gene expression of MRP2 was significantly upregulated under liver injury conditions compared with the normal group. Supplementation with L. rhamnosus R0011 significantly downregulated the relative gene expression of MRP2 compared with the M group. However, L. rhamnosus R0011 did not reduce the relative gene expression of MRP2 compared with the N group, as shown in Figure 4A, mainly due to the expression of MPR2 being kept at a low level under the healthy state. We propose that the enhancement of GL bioavailability in the NLGG group was mainly attributed to the elevation in microbial glucuronidase, which was related to the supplementation with Lactobacillus and the increase in Ruminocoucccus, as shown in Figure 6C.
It has been reported that the gut microbiota structure of liver injury models was disordered, and the dynamic balance between different microbes was disrupted [14,23]. The gut microbiota disorder in liver injury can significantly influence oral drug absorption. In our study, supplementation with L. rhamnosus R1100 in liver injury rats increased microbial glucuronidase and downregulated the gene expression of the drug transporter MRP2. More critically, L. rhamnosus R1100 regulated the disorder of the gut microbiota and microbial metabolic function under the liver damage state, which indirectly promoted the bioavailability of GL. Therefore, supplementation with probiotic L. rhamnosus R1100 has a more profound effect on the bioabsorption of GL under an irrational state. However, the exact mechanism of the different effects of L. rhamnosus R1100 on healthy and pathological conditions still requires further investigation.

Conclusions
In summary, the probiotic L. rhamnosus R0011 tolerated the harsh gut environment and significantly improved the bioavailability of GL in rats, especially under a liver fibrosis state. Our research showed that L. rhamnosus R0011 intervention directly improved the microbial glucuronidase activity, as well as downregulating the host drug export gene expression. On the other hand, L. rhamnosus R0011 restored the gut microbial composition and repaired intestinal homeostasis, indirectly promoting GL bioabsorption. Overall, these data may provide possible strategies to address the low bioavailability of traditional medicines through probiotic intervention.
Author Contributions: H.L. performed experiments, analyzed the data, and drafted the manuscript. J.W., Y.F., K.Z. and Z.D. contributed significantly to perform experiments and analyze data. L.D. and J.S. revised the manuscript. T.Y. and S.J. designed the study and review of the manuscript. All authors have read and agreed to the published version of the manuscript.