Effects of Zhenggan Huayu decoction (正肝化瘀方) combined with entecavir on gut microbiota in patients with chronic hepatitis B fibrosis

Abstract

OBJECTIVE: To evaluate the effects of Zhenggan Huayu decoction (正肝化瘀方, ZGHY) combined with entecavir (ETV) on the gut microbiota in patients with chronic hepatitis B (CHB) fibrosis.

METHODS: A total of 59 CHB-related fibrosis patients were enrolled and treated with ZGHY combined with ETV (ZGHY + ETV) and ETV alone. Fecal samples were collected from patients at weeks 0, 12, and 24 after treatment and gut microbiota were analyzed by 16S rRNA gene sequencing.

RESULTS: Compared to the ETV group, microbiota diversity in the ZGHY + ETV group was increased after 24 weeks. Some potentially pathogenic bacteria, including Fusobacteriaceae spp., Acidaminococcus spp., and Erysipelotrichaceae spp. were reduced in the ZGHY + ETV group, while Eubacteriaceae spp., Lactococcus spp., and several other beneficial bacteria were increased.

CONCLUSION: Decreases in pathogenic bacteria and increases in probiotics were not always observed in the Traditional Chinese Medicine (TCM) group (e.g., Rothia was abundant). As an adjuvant TCM formulation for ETV, ZGHY had a positive role in the treatment of CHB patients.

1. INTRODUCTION

It has been reported that more than 250 million people worldwide are infected with the hepatitis B virus (HBV), which is responsible for some 600 000 deaths/year.¹ For chronic hepatitis B (CHB) patients, the ongoing liver damage process leads to the development of scar tissue, eventually resulting in liver fibrosis.² The final liver fibrosis stage is cirrhosis, which can be compensatory cirrhosis or decompensated cirrhosis. It is also the main cause of mortality in CHB patients. About 40% of cirrhosis cases can be attributed to HBV infections.^3,4 CHB is the main cause of cirrhosis in China.^5,6 The 5-year survival rate for patients with compensatory cirrhosis is about 55%, while for patients with decompensated cirrhosis is about 14%.⁷ CHB is treated by delaying liver fibrosis with drugs to prevent progression to decompensated cirrhosis or hepatocellular carcinoma (HCC).

Nucleoside (acid) analogues, including entecavir (ETV), adefovir dipivoxil, tenofovir, and dipivoxil are recommended for CHB treatment. Entecavir is one of the main drugs used to treat CHB-related fibrosis in China.⁸ However, the efficacy of ETV therapy in reducing HCC risk in CHB patients, which has been reported to be associated with lactic acidosis, has not yet been fully established.^9,10 Adjuvant treatment of CHB with Traditional Chinese Medicines (TCM) has been found to achieve better results.^8,11,12 Many natural pharmaceutical components are safe immunomodulators for regulating gene expressions of immune proteins.¹¹ The Chinese Pharmacopoeia has many TCM formulations for treating CHB-related fibrosis, such as Anluo Huaxian pill (安络化纤丸), ¹³ Bijia Ruangan tablet (鳖甲软肝片),¹⁴ and Fuzheng Huayu capsule (扶正化瘀胶囊).¹⁵ These TCM formulations are efficacious, and when combined with ETV, they can improve treatment outcomes of CHB-related fibrosis. ^8,16,17 Combinations of traditional and modern medicines may be effective and low-risk treatment options for CHB-related fibrosis.

Gut microbiota affects the metabolic processes of the body and is directly or indirectly involved in immune responses.¹⁸ The pathogenesis of many chronic liver diseases, including CHB,¹⁹ alcoholic liver disease (ALD),²⁰ and non-alcoholic fatty liver disease (NAFLD)²¹ are associated gut microbiota. Gut microbiota has an important role in promoting CHB development.²² A previous study found that CHB patients present with dysregulated gut microbiota and an increased abundance of potential pathogens.²³ Gut microbiota activities may be a biological factor in CHB-related fibrosis and its progression. Therefore, elucidating the association between drugs and gut microbiota may inform the pathogenesis of CHB and provide a basis for developing effective treatments.

Zhenggan Huayu decoction (ZGHY) (正肝化瘀方) is a TCM decoction composed of Huangqi (Radix Astragali Mongolici) and combined with Danshen (Radix Salviae Miltiorrhizae), and other medicines. In clinical application, this decoction reverses liver fibrosis, while its combination with ETV for CHB-related fibrosis treatment can significantly improve the degree of liver fibrosis. However, changes in gut microbiota during the treatment of CHB-related fibrosis by ZGHY combined with ETV have not yet been established. Thus, in order to investigate the effects of ZGHY and ETV combined treatments on the gut microbiota of patients with CHB-related fibrosis, we used high throughput 16S rDNA sequencing to characterize the gut microbiota of patients in the two treatment regiments.

2. MATERIALS AND METHODS

2.1. Zhenggan Huayu decoction

Zhenggan Huayu decoction (ZGHY) (正肝化瘀方) is a TCM decoction composed of Huangqi (Radix Astragali Mongolici) 30 g, Biejia (Carapax Trionycis) 30 g, Sanleng (Rhizoma Sparganii) 10 g, Ezhu (Rhizoma Curcumae Phaeocaulis) 10 g, Chuanxiong (Rhizoma Chuanxiong) 10 g, Danshen (Radix Salviae Miltiorrhizae) 30 g, Baihuasheshecao (Herba Hedyotdis) 30 g and Banzhilian (Herba Scutellariae Barbatae) 10 g.

2.2. Patinent recruitment and treatments

Patients with CHB-related fibrosis were recruited from the Shenzhen Hospital of Beijing University of Chinese Medicine. The inclusion criteria were the following: (a) diagnosed as CHB-related fibrosis by the viral hepatitis control guild of China; (b) HBeAg positive; (c) aged 16 -60 years old. Patients with non-HPV hepatitis virus infections, severe hepatitis, and decompensated liver cirrhosis were excluded. A total of 59 patients were finally recruited and assigned into two groups, i.e., the ZGHY + ETV group (n = 38) treated with ZGHY 1 potion (oral administration in two doses) and ETV 0.5 mg/d, and the ETV group (n = 21) treated with ETV 0.5 mg/d.

2.3. Sample collection, and 16S rRNA gene sequencing

Fecal samples were obtained from each participant before treatment (0 week) and at weeks 12 and 24 after treatment, and were immediately stored at -20 ℃. After collection, dry ice was used to wrap the samples, which were sent to the laboratory for DNA extraction, which was performed using the MOBIO PowerSoil ® DNA Isolation kit (QIAGEN, Hilden, Germany) as instructed by the manufacturer. Universal primers, 515F (5’-GTGYCAGCMGCCGCGGTAA-3’) and 806R (5’-GGACTACNVGGGTWTCTAAT-3’) were used to amplify the V4 region of 16S rDNA. Mixture compositions and PCR temperature were as described.²⁴ PCR products were purified by a Quant-iT PicoGreen dsDNA Assay kit (Invitrogen, Carlsbad, CA, USA). The amplicon library was combined in equimolar amounts and quantified using KAPA Library Quantification Kit (Kapa biosystems, Capetown, South Africa) according to the manufacturer’s instructions. The constructed library was sequenced on the Illumina HiSeq 6000 platform (San Diego, CA, USA) by 150 bp paired-end sequencing.

2.4. Bioinformatic analyses

The original reads sequence underwent strict quality control. Low-quality reads that were too short or contained N-bases were removed. The adapter, primers, and low-quality bases at both ends were trimmed from reads to obtain high-quality reading sequences. The Quantitative Insights into Microbial Ecology v2 (QIIME2) platform and relevant tools were used for stitching, clustering, and sequence annotation.²⁵ DADA2 algorithm was used to cluster amplicon sequence variants (ASVs) and generate the abundance characteristic table. Shannon and Simpson's indices were calculated using the 'q2-diversity' plugin. The distance between samples was calculated based on the Bray-Curtis distance, while β diversity between groups was assessed by distance-based redundancy analysis (dbRDA). The Naive Bayes classifier was trained based on pruned SILVA database ²⁶ and was used to determine taxonomic assignments for each ASV. To avoid sampling depth bias, each sample was normalized (randomly resampling 20 000 reads) when calculating ASVs abundance. PICRUSt2²⁷ with default parameters was used to predict ASV functionality. The functional diversity analysis method was consistent with species analysis.

2.5. Ethical statement

The Ethics Commission of Shenzhen Hospital of Beijing University of Chinese Medicine approved this study, and informed consent was obtained from every participant. Study protocols were in accordance with the Declaration of Helsinki.

2.6. Statistical analyses

All analyses were performed using R. The Specaccum function in the Vegan package was used for rarefaction analyses, with default parameters used for calculation. Wilcoxon signed rank-test was used to compare differences between samples; P<0.05 denoted statistical significance. Permutational multivariate analysis of variance (PERMANOVA, adonis) was used to assess the degree of variation between groups. Linear discriminant analysis Effect Size (LEfSe) analysis was used to calculate species differences at each classification level, and log10 (LDA) score>2 was considered as the species with significant differences in abundance. Spearman correlation coefficient analysis was used to evaluate correlations between OTUs and clinical indicators. The absolute value of correlation coefficient>0.6 and P<0.05 denoted significant correlations. The ggpolt2 package was used for data visualization.

3. RESULTS

3.1. Patient statistics and 16s rRNA gene sequencing

There were no significant differences in basic clinical features between the two groups [ETV (ETV group, n = 21) and ETV + ZGHY (TCM group, n = 38)] (Table 1). A total of 2184 ASVs were obtained using the QIIME2 platform, with an average of 104 ASVs per sample. Based on the SLIVA database, these ASVs were classified into bacterial taxa. About 58.7% of ASVs were classified at the genus level and 89.9% at the family level.

Table 1.

Characteristics and clinical indices of participants

	Before treatment (0 week)				After treatment (24 weeks)
Item	ETV (n = 21)	TCM (n = 38)	P value		ETV (n = 21)	TCM (n = 38)	P value
Sex (F/M)	3/17	3/35	0.405		0.176	0.086	0.405
Age	40.45±7.35	41.70±9.29	0.456		40.45±7.35	41.70±9.29	0.456
ALT	86.31±105.55	241.84±808.09	0.697		25.32±9.48	26.93±12.22	0.563
AST	80.46±108.09	68.51±52.44	0.715		27.24±4.97	31.34±15.09	0.953
GGT	63.01±48.67	92.53±138.73	0.841		36.38±24.81	40.73±31.62	0.759
ALP	98.60±36.55	106.80±58.69	0.804		80.57±32.67	89.26±51.46	0.991
TBA	13.59±13.09	24.39±38.46	0.532		11.33±17.36	13.60±28.28	0.654
TP	73.59±6.55	72.82±5.75	0.637		71.79±8.05	74.48±5.25	0.328
TBIL	22.89±13.56	26.70±23.48	0.953		19.06±6.41	18.63±6.64	0.715
DBIL	6.01±5.98	8.59±12.24	0.741		3.45±1.14	3.97±2.80	0.878
IBIL	17.44±7.68	18.10±11.97	0.663		15.61±5.44	14.23±5.08	0.268
CHE	7485.12±2208.47	7555.21±2626.55	0.805		7815.45±1555.8	8081.00±2335.77	0.596
TG	1.28±0.71	1.37±0.69	0.340		1.16±0.41	1.56±1.21	0.396
TC	4.71±1.11	4.47±0.89	0.509		4.88±1.13	4.68±0.88	0.841
HDL-C	1.39±0.34	1.22±0.37	0.283		1.43±0.27	1.28±0.22	0.087
LDL-C	3.13±0.84	3.02±0.71	0.588		3.33±0.83	3.27±0.80	1.000
UREA	4.87±1.15	4.69±0.91	0.540		29.08±107.35	4.86±0.78	0.334
CR	72.57±13.14	70.72±8.75	0.741		75.46±9.86	73.99±11.98	0.841
UA	365.34±102.41	342.92±79.17	0.760		369.56±110.51	361.81±65.52	0.769
HA	182.92±237.93	202.16±272.55	0.800		87.81±88.53	60.15±41.98	0.266
PIIINP	12.13±10.30	11.51±9.62	0.888		8.92±5.80	6.63±2.14	0.490
IV-C	100.76±88.14	104.07±143.10	0.422		56.73±32.23	45.04±27.97	0.254
LN	2604.00±10286.62	119.10±67.09	0.746		113.61±75.35	93.10±38.79	0.526
RL	122.30±10.74	123.58±12.82	0.962		121.55±9.55	122.88±8.65	0.509
DPV	10.85±1.24	10.50±0.76	0.617		10.60±1.28	10.50±0.65	0.440
TS	41.15±5.72	37.38±6.42	0.041		36.55±5.01	35.50±5.33	0.594

Notes: patients in the TCM group were treated with ZGHY 1 potion (oral administration in two doses) and ETV 0.5 mg/d, and the ETV group were treated with ETV 0.5 mg/d. All patients were treated for 24 weeks. ALT: alanine aminotransferase; AST: aspartate aminotransferase; GGT: γ-glutamyl transferase; ALP: alkaline phosphatase; TBA: total bile acid; TP: total protein; TBIL: total bilirubin; DBIL: direct bilirubin; IBIL: indirect bilirubin; CHE: cholinesterase; TG: triglyceride; TC: total cholesterol; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; CR: creatinine; UA: uric acid; HA: hyaluronic acid; PIIINP: type III procollagen peptide; IV-C: Type IV collagen; LN: laminin; RL: right lobe of liver oblique diameter; DPV: portal vein diameter; TS: spleen thickness.

3.2. TCM corrected the continuous decline of gut microbiota diversity in patients

The number of ASVs in the TCM group was higher compared to the ETV group at each time point with the same sample size (Figure 1A). However, the curve was still not saturated at nearly 40 samples, indicating the need to include more samples in future analyses.

Figure 1. Effects of ETV and TCM on the gut microbiota of patients.

A: rarefaction curve analysis of ASVs in each group of samples. The number of identified ASVs in each group was calculated based on a randomly selected specific number of samples with 100 replacements. Quartiles numbers are plotted by a line segment; B, C: differences in gut microbiota diversity between ETV and TCM groups at different treatment periods. The significance level in the Wilcoxon signed-rank test was marked as a numerical P-value; D: the dbRDA analysis based on microbiota composition showed separations between groups. The R² and P values generated by PERMANOVA analysis are shown in the lower-left quadrant of the plot; E: dbRDA analysis showed separations at different treatment periods within ETV and TCM groups, and R² as well as P values generated by PERMANOVA analysis are shown at the top of the image; F: PERMANOVA analysis showed differences between treatment periods. Patients in the TCM group were treated with ZGHY 1 potion/d (oral administration in two doses) and ETV 0.5 mg/d, and the ETV group were treated with ETV 0.5 mg/d. All patients were treated for a total of 24 weeks, fecal samples were collected and analyzed in the 12 and 24 weeks, respectively. ETV: entecavir; TCM: Traditional Chinese Medicine (FZHY+ETV group); CON: control (ETV group); ZGHY: Zhenggan Huayu decoction; ASVs: amplicon sequence variants; dbRDA: distance-based redundancy analysis; PERMANOVA: permutational multivariate analysis of variance. Significant differences in TCM group: 12 w vs 24 w, ^aP<0.05; 0+12 w vs 24 w, ^aP<0.05. n = 38.

The Shannon and Simpson indices, which were used to assess gut microbiota richness (Figure 1B, 1C), revealed that differences in gut microbiota diversity between the groups were insignificant at different treatment time points; however, compared to the ETV group, the TCM group had increased diversity. Gut microbiota diversity in patients treated with ETV continued to decline over time, but the reduction was not significant. After 24 weeks of treatment, gut microbiota diversity in patients in the TCM group showed a recovery trend, indicating that TCM treatment reversed the decline in gut microbiota diversity caused by the common treatment.

Distance-based Redundancy Analysis (dbRDA) based on Bray Curtis distance showed a certain degree of separation between the ETV and TCM groups, which was not significant (Figure 1D; R² = 0.8%, P>0.05). β diversity in the ETV group was relatively stable at different treatment stages (R² = 2.9%, P>0.05), while the TCM group showed certain temporal dynamics (R² = 0.264, P = 0.044) (Figure 1E). PERMANOVA analysis showed significant separation between the TCM group at week 12 and the other two time points (Figure 1F; 0 w + 12 w vs 24 w, P<0.05). Thus, continuous TCM treatment significantly improved the diversity of gut microbiota in patients.

3.3. TCM increased the abundance of gut probiotics in patients with hepatitis B fibrosis

In order to determine whether TCM treatment improved gut ecology, we compared gut species abundances between ETV and TCM groups. At the phylum level, although differences between groups were insignificant, a trend was observed. Compared to the ETV group, actinomycete abundances were increased while firmicutes were decreased in the TCM group at week 12 of treatment (Figure 2A). At 24 weeks, the TCM group still maintained low firmicutes abundance. Proteobacteria abundance was also increased at week 24 of TCM treatment (Figure 2B). At the genus level, differences in abundance among the top 10 genera were insignificant (Figure 2C). However, there were some trends, e.g., after 12 weeks of TCM treatment, the abundance of Bifidobacterium was increased compared to the ETV group (Figure 2D). Bifidobacterium abundance remained high after 24 weeks of TCM treatment.

Figure 2. Microbiota composition in patients’ gut.

A: the composition of phyla levels in different groups. Only the top 10 members in relative abundance are shown; B: species differences at phylum levels between groups; C: the composition of family levels in different groups. Only the top 10 members in relative abundance are shown; D: species differences at family levels between groups. Patients in the TCM group were treated with ZGHY 1 potion/d (oral administration in two doses) and ETV 0.5 mg/d, and the ETV group were treated with ETV 0.5 mg/d. All patients were treated for a total of 24 weeks, fecal samples were collected and analyzed in the 12 and 24 weeks, respectively. ETV: entecavir; ZGHY: Zhenggan Huayu decoction; TCM: Traditional Chinese Medicine (FZHY + ETV group); CON: control (ETV group).

LEfSe can help us identify biomarkers in multiple groups. Compared to the ETV group, the abundances of gut Clostrudium and Methylpentosum were higher, while those of Catneibacterium Coprobacillus, and Enterococcus were lower in the TCM group at week 0 (supplementary Figure 1A). Variations in the abundance of these bacteria may be due to factors such as an individual's genetic background and gut type. After 12 weeks of TCM treatment, abundances of Leuconostoc and Eubacteriaceae were increased, while those of Clostridium sordellii, Fusobacteriaceae, and Eggerthella were depleted (supplementary Figure 1B). After 24 weeks of TCM treatment, Eubacteriaceae, Lactococcus, Micrococcaceae, and Rothia were significantly enriched, while Acidaminococcus, Fusobacteriaceae, and Erysipelotrichaceae were depleted (supplementary Figure 1C). Most of the bacteria that were increased in the TCM group were considered to have probiotic characteristics. For instance, Eubacteriaceae and Rothia are important SCFA-producing bacteria. Moreover, TCM treatment significantly reduced the abundance of Fusobacteriaceae, a pathogen associated with colon cancer and other diseases. These results suggest that TCM may help to maintain the stability of gut microbiota and alleviate the damage to gut ecology caused by drug therapy.

3.4. Changes in gut microbiota improved patients' clinical indicators

Rothia, Paraeggerthella, Erysipelotrichaceae, Acidaminococcus and Clostridium were significantly positively correlated with multiple liver indicators, such as ALT, AST, CHE, HA, LDL-C, LN, PIIINP, UA, UREA, and TS. Paraeggerthella, Enterococcus, Acidaminococcus, and Clostridium were negatively correlated with TP, TS, UA, and CHE, respectively (Figure 3).

Figure 3. Heat maps show correlations between feature genera and clinical indicators.

Spearman correlation test showed that: * q<0.05; + q<0.01. A correlation coefficient>0.6 was considered as a positive correlation, and<-0.6 as a negative correlation. ALP: alkaline phosphatase; ALT: alanine aminotransferase; AST: aspartate aminotransferase; CHE: cholinesterase; CR: creatinine; DBIL: direct bilirubin; GGT: γ-glutamyl transferase; HA: hyaluronic acid; HDL-C: high-density lipoprotein cholesterol; IBIL: indirect bilirubin; IV-C: Type IV collagen; LDL-C: low-density lipoprotein cholesterol; LN: laminin; PIIINP: type III procollagen peptide; TBA: total bile acid; TBIL: total bilirubin; TC: total cholesterol; TG: triglyceride; TP: total protein; UA: uric acid; UREA: urea; RL: right lobe of liver oblique diameter; DPV: portal vein diameter; TS: spleen thickness.

Considering findings from species analysis, we focused on TCM-associated bacteria after 24 weeks of treatment. Bacteria that were enriched in the TCM group, such as Enterococcus and Rothia, were significantly correlated with PIIINP and TS; bacteria that were enriched in the ETV group, such as Acidaminococcus and Erysipelotrichaceae, were significantly correlated with ALT, HA, UA, CHE, and LDL-C.

3.5. Effect of TCM treatment on gut microbiota functions

In order to assess the effects of CHB-related fibrosis on the functional characteristics of gut microbiota, we used PICRUSt2 to make functional predictions for each sample. A total of 6588 Kyoto Encyclopedia of Genes and Genomes Orthologys (KOs) were annotated. More KOs were observed in the TCM group (Figure 4A), especially after 24 weeks of treatment. Consistent with species level, functional diversity of gut microbiota in CHB patients decreased first and then increased (Figure 4B). After 24 weeks of treatment, the functional diversity of bacteria in the TCM group was significantly higher than in the other two periods.

Figure 4. Effects of ETV and TCM on gut microbiota functions.

A: rarefaction curve analysis of KOs in each group of samples. The number of annotated KO genes in each group was calculated based on a randomly selected specific number of samples with 100 replacements. Quartile numbers were plotted by a line segment. B: boxplot showing the Shannon index of gut microbiota function for patients in ETV and TCM groups at different treatment periods, revealing microbiota function diversity in patients’ gut. C: fold change of gut microbiota function between patients in TCM and ETV groups. Only functional items with significant differences (P<0.05) are shown, TCM group: n = 38, ETV group, n = 21. Patients in the TCM group were treated with ZGHY 1 potion/d (oral administration in two doses) and ETV 0.5 mg/d, and the ETV group were treated with ETV 0.5 mg/d. All patients were treated for a total of 24 weeks, fecal samples were collected and analyzed in the 12 and 24 weeks, respectively. ETV: entecavir; TCM: Traditional Chinese Medicine (FZHY + ETV group); CON: control (ETV group); KO: Kyoto Encyclopedia of Genes and Genomes Orthology; ZGHY: Zhenggan Huayu decoction.

Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis revealed 379 enriched pathways. We tested the bacterial function of the two groups using the Wilcoxon sign rank test. Interestingly, all significant differences were found in the TCM group. At week 0, adenosylcobalamin biosynthesis I and GDP-Mannose biosynthesis were markedly enriched in the TCM group relative to the ETV group. At week 12, l-methionine biosynthesis III and chorismate biosynthesis II were significantly enriched in the TCM group compared to the ETV group. At week 24, Thiazole Biosynthesis II, reductive acetyl-coenzyme A Pathway, superpathway of (R, R)-butanediol biosynthesis, and L-glutamate degradation V (VIA hydroxyglutarate) were significantly enriched in TCM group (Figure 4C). Accordingly, TCM treatment affected both gut bacteria composition and gene functions. Functional gene enrichment suggests that TCM may have an active role in improving CHB-related fibrosis, which is consistent with clinical observations.

4. DISCUSSION

Previous studies have reported on gut microbiota distribution in HBV-positive patients during disease development;²³ however, the effects of drug treatment on gut microbiota have yet to be fully established. In clinical applications in China, the ZGHY decoction has been found to improve liver fibrosis; nonetheless, it is still in the empirical decoction stage due to a lack of case-control reports. In some hospitals in China, clinicians choose ZGHY combined with ETV to treat patients' CHB-related fibrosis; however, the underlying mechanism remains unclear. Thus, we evaluated the effects of ZGHY from a gut microbiota perspective. To the best of our knowledge, this is the first study that elucidated the effects of ZGHY combined with ETV on gut microbiota in patients with CHB-related fibrosis. We also assessed the dynamic changes in gut microbiota gene functions during drug treatment.

Gut microbiota dysregulation may aggravate liver fibrosis or cirrhosis in CHB patients.²² Cirrhosis has been found to decrease fecal microbiome diversity, while diversity from cirrhosis to early HCC tends to increase.²⁸ A different study reported that the gut microbiota of CHB patients changed before the onset of severe liver damage. ²³ However, none of the previous studies have reported whether medications such as ETV worsen or improve the process. Herein, we evaluated the effects of TCM treatment (ZGHY + ETV) and ETV alone on gut microbiota alpha diversity of patients by Shannon and Simpson indices. While there were no marked differences, microbiota diversity in ETV alone group decreased over time, which is not consistent with findings from mice models.²⁹ This inconsistency might be due to various factors, such as administration duration, liver fibrosis degree, and sampling position. In the TCM group, longer treatment (after 24 weeks) partially corrected the decrease in microbiota diversity. However, it is necessary to include more samples and more detailed treatment cycles in future studies to explain this pattern.

In CHB patients, Firmicutes reduction is usually accompanied by an increased abundance of Bacteroidetes.²³ The proliferation of gram-negative Bacteroidetes can lead to increased lipopolysaccharide (LPS) concentrations in the intestinal tract and promote disease progression.^30,31 We did not find significant changes in Bacteroidetes abundance in each group; however, TCM treatment may have promoted the reduction of Firmicutes, although not to significant levels. The increased Bacteroidetes/Firmicutes ratio has been shown to increase the risk of LPS exposure.³¹ At the genus level, Fusobacteriaceae in the TCM group was significantly reduced compared to the ETV group. Fusobacteriaceae belongs to Fusobacteria, which is an anaerobic, gram-negative, non-spore-producing bacteria strongly associated with inflammatory bowel diseases.³² Fusobacteriaceae damage the glycocalyx layer of the intestinal epithelium, disrupting desmosome junctions between epithelial cells and increasing intestinal permeability.^33,34 Erysipelotrichaceae, another potential pathogen, was also found to be significantly reduced in the intestinal tract of TCM group patients. Erysipelotrichaceae belongs to the firmicutes group and is abundant in colon cancer patients.³⁵ Erysipelotrichaceae bacteria are positively correlated with TNF-α,³⁶ which may promote liver inflammation. Moreover, some pathogenic bacteria, such as Paraeggerthella, were also reduced in the TCM group, while Paraeggerthella, Erysipelotrichaceae, and Acidaminococcus were positively correlated with ALT, CHE, and HA. Elevated ALT, CHE, and HA are associated with chronic hepatitis and cirrhosis.³⁷ Therefore, although TCM treatment may increase LPS exposure risk, inhibition of the above-mentioned pathogens may be more effective in improving liver fibrosis.

Among the bacteria enriched in the TCM group, there were some pathogenic bacteria species, including Enterococcus, Rothia, and Leuconostoc. Enterococcus is a gram-negative bacterium whose abundance is positively correlated with LPS, serum IL-6, IL-1, and other inflammatory cytokines.³⁸ As an ethanol-producing bacterium, Enterococcus promotes liver damage. ³⁹ We found that Enterococcus was negatively correlated with thickness spleen (TS); thus, the promoting effect of Enterococcus on liver fibrosis may not be significant. Rothia, a member of Firmicutes, is an opportunistic pathogen ⁴⁰ that also produces SCFA (Butyric acid producer) in the gut. ⁴¹ Among others, SCFAs include butyrate, acetic acid, propionic acid, and isobutyrate. They are the main energy sources of intestinal epithelial cells. SCFA has an important role in maintaining intestinal PH stability and immune regulation.⁴² We established that Rothia was proportional to PIIINP levels and TS levels, which may contribute to liver damage. Moreover, some bacteria with probiotic characteristics, such as Lactococcus and Eubacteriaceae, were also enriched in the TCM group. These bacteria are important probiotics and SCFA-producing bacteria in the gut. ^{43, 44} These results suggest that TCM treatment strategies have a positive role in ameliorating gut microbiota disorders.

In CBH patients, liver fibrosis may lead to metabolic disorders in the gut.²³ PICRUSt2 analysis showed that functional diversity in the TCM group was significantly higher than in the ETV group, which may be attributed to the correction of gut microbiota disorders by TCM. We also observed the enrichment of l-methionine biosynthesis Ⅲ and chorismate biosynthesis II4, Thiazole Biosynthesis Ⅱ, reductive acetyl-coenzyme a pathway, superpathway of (R, R)-butanediol biosynthesis, and L-glutamate degradation V (VIA hydroxyglutarate) in the gut microbiota of patients from TCM group. These pathways are involved in amino acid synthesis and degradation, energy metabolism, and other functions. These results suggest that TCM treatment may improve the functional disorders of CHB-related fibrosis.

The main limitation of the present study is that it did not assess the effects of drugs on the microbiota of healthy people. However, it is unrealistic for healthy people to take TCM without any reason. However, this can be done in animal models (SPF as well as germ-free animals), which are good in vivo models to study interactions between drugs, bacteria, and diseases. Future studies should consider using animal models to better control variables. In addition, future studies should include more samples and more time points, which is crucial for revealing the impact of TCM on gut microbiota.

In conclusion, compared to ETV alone, ZGHY combined with ETV therapy has a certain correction effect on gut microbiota dysregulation in patients with CHB-related fibrosis and can significantly improve liver fibrosis. Although TCM treatment is not completely characterized by a decrease in pathogenic bacteria and an increase of beneficial bacteria, it can still inhibit some pathogenic bacteria. Gut microbiota may be one of the targets of ZGHY drug actions, and analysis of the association between microorganisms, drugs, and CHB provides new insights for the treatment of CHB and other liver diseases.