Male 8-week-old C57L/J mice were obtained from Hunan Slyke Jingda Laboratory Animal Co., Ltd (Changsha, China). The construction of an orthotopically transplanted HCC model was performed following one week of adaption [16]. Briefly, 1 $\times$ 10${}^6$ Hepal-6 cells were subcutaneously injected into the right axilla. When the tumor grew to a volume of 1–2 cm${}^3$, it was harvested and cells cultured. The fourth generation of cells were selected for the orthotopically transplanted HCC model used in this investigation. Following anesthesia (10% chloral hydrate, 0.3 mL/100 g), the livers of mice were exposed and 1 $\times$ 10${}^6$ fourth-generation Hepal-6 cells were directly seeded into the left lobe of the liver and compressed with a gelatin sponge for hemostasis. Mice in the Control group were subjected to all surgical procedures except without injection of Hepal-6 cells.

The experimental treatment groups were listed as follows: Control, Model, Quercetin, anti-PD-1, and Quercetin+anti-PD-1, and each group contained 8 mice. The corresponding intervention was administered to mice 14 days after cell implantation. Mice in the Control and Model groups were gavaged with the same dose of distilled water for 14 days. Mice in the Quercetin group were gavaged daily with quercetin (100 mg/kg$\cdot$d, Q817162, Macklin, Shanghai, China) for 14 days [17]. Mice in the anti-PD-1 group were injected with anti-PD-1 antibody (20 mg/kg, B276435, Aladdin, Shanghai, China) through the tail vein twice a week for 14 days [18]. Mice in the Quercetin+anti-PD-1 group received both treatments as outlined above simultaneously for 14 days. On day 28 after cell implantation, the mice were sacrificed, and blood, livers, and feces were collected. Livers were fixed with 4% paraformaldehyde followed by embedding in paraffin, and subsequently sectioned for staining analysis.

The liver tissue slices were first dewaxed by placing them in xylene for 20 min followed by dehydration with an ethanol gradient (75–100%). After this, the slices were stained with hematoxylin (AWI0001a, Abiowell, Changsha, China) for 1–10 min and subsequently rinsed in phosphate buffer saline (PBS). Afterward, the slices were stained with eosin (AWI0029a, Abiowell) for 1–5 min, washed with distilled water, and then dehydrated with an ethanol gradient (95–100%). Finally, the slices were placed in xylene for 10 min for tissue transparency and sealed with neutral gum (AWI0238a, Abiowell) prior to observation.

Total RNA was extracted from liver tissues using Trizol reagent (15596026, Thermo, Waltham, MA, USA). Following this, mRNA was reverse transcribed into cDNA using a reverse transcription kit (CW2569, CWBIO, Beijing, China), followed by quantitative real-time polymerase chain reaction (RT-qPCR). The primers were as follows: $\beta$-actin: F: 5′-ACATCCGTAAAGACCTCTATGCC-3′, R: 5′-TACTCCTGCTTGCTGATCCAC-3′; interleukin-6 (IL-6): F: 5′-GACTTCCATCCAGTTGCCTT-3′, R: 5′-ATGTGTAATTAAGCCTCCGACT-3′; IL-12a: F: 5′-TGAAGACATCACACGGGACCA-3′, R: 5′-CAGCTCCCTCTTGTTGTGGAA-3′; IL-1$\beta$: F: 5′-TGCCACCTTTTGACAGTGATG-3′, R: 5′-AAGGTCCACGGGAAAGACAC-3′; tumor necrosis factor-$\alpha$ (TNF-$\alpha$): F: 5′-AGCACAGAAAGCATGATCCG-3′, R: 5′-CACCCCGAAGTTCAGTAGACA-3′; arginase-1 (Arg-1): F: 5′-GGGCACACTCATGCATTCCT-3′, R: 5′-GCGTCAGCTGCGAGTACATA-3′; IL-10: F: 5′-GTTCCCCTACTGTCATCCCC-3′, R: 5′-AGGCAGACAAACAATACACCA-3′; transforming growth factor-$\beta$ (TGF-$\beta$): F: 5′-CTCCCGTGGCTTCTAGTGC-3′, R: 5′-GCCTTAGTTTGGACAGGATCTG-3′; matrix metalloproteinase-9 (MMP-9): F: 5′-GCCCTGGAACTCACACGACA-3′, R: 5′-GTAGCCCACGTCGTCCACC-3′. The relative gene expression level was calculated using the 2${}^{-\mathrm{\Delta }\mathrm{\Delta }\text{Ct}}$ method with $\beta$-actin used as the internal reference.

Liver tissues from different groups of mice were suspended in radio-immunoprecipitation assay (RIPA) lysate (AWB0136, Abiowell) to obtain total protein extracts. After separation by sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE), proteins were transferred onto the nitrocellulose (NC) membranes. After blocking with 5% skimmed milk (AWB0004, Abiowell) for 1.5 h, the NC membranes were incubated with primary antibodies at 4 °C overnight separately. Primary antibodies were as followed: CD8a (1:1000, ab33786, Abcam, Cambridge, UK), CD4 (1:1000, 19068-1-AP, Proteintech, Chicago, IL, USA), CD11b (1:1000, ab133357, Abcam), glyceraldehyde-3-phosphate dehydrogenase (GAPDH, 1:5000, 10494-1-AP, Proteintech), toll-like receptor 4 (TLR4, 1:1000, 19811-1-AP, Proteintech), inhibitor of nuclear factor $\kappa$B$\alpha$ (I$\kappa$B$\alpha$, 1:2000, 10268-1-AP, Proteintech, Chicago, IL, USA), p-I$\kappa$B$\alpha$ (1:10000, ab133462, Abcam), p-p65 (1:1000, ab76302, Abcam), p65 (1:5000, ab16502, Abcam), and $\beta$-actin (1:5000, 66009-1-Ig, Proteintech). The NC membranes were subsequently incubated with horseradish peroxidase (HRP)-goat anti-mouse IgG (1:5000, SA00001-1, Proteintech) or HRP-goat anti-rabbit IgG (1:6000, SA00001-2, Proteintech) for 1.5 h. Finally, the NC membranes were incubated with ECL reagent (AWB0005, Abiowell) and followed by chemiluminescent imaging. The expression levels of proteins were analyzed by Quantity One 4.6.6 software (Bio-Rad Inc., Hercules, CA, USA) using GAPDH or $\beta$-actin as the reference.

For the detection of IL-6 (KE10007, Proteintech), IL-10 (KE10008, Proteintech), interferon-$\gamma$ (IFN-$\gamma$, KE10001, Proteintech), and IL-4 (KE10010, Proteintech). Briefly, whole blood was placed for 2 h at room temperature followed by centrifugation at 1000 g for 15 min at 4 °C, and the supernatant was taken for analysis. For the detection of granulocyte-macrophage colony-stimulating factor (GM-CSF, KE10015, Proteintech, Chicago, IL, USA) and granulocyte colony-stimulating factor (G-CSF, KE10025, Proteintech), liver tissue was washed with PBS and homogenized, then placed at –20 °C overnight. After freeze-thaw treatment, the homogenate was centrifuged at 5000 g for 5 min at 4 °C, and then the supernatant was taken for analysis.

The liver slices were dewaxed by soaking in xylene for 20 min. Subsequently, dehydration was carried out using an ethanol gradient (75–100%). The slices were then boiled in citrate buffer (0.01 M, pH 6.0) (AWI0206a, Abiowell) for antigen retrieval. 1% periodic acid was employed to inactivate the endogenous enzymes. The slices were incubated with antibodies of PD-L1 (1:200, ab238697, Abcam) overnight at 4 °C, and then incubated with 100 µL of HRP-anti-Rabbit-IgG for 30 min at 37 °C. Following these steps, 100 µL of diaminobenzidine (DAB) was added to slices and incubated for 5 min. The slices were counterstained with hematoxylin for 5 min, rinsed with distilled water, and returned to blue in PBS. Dehydration was carried out with an ethanol gradient (60–100%), 5 min for each level. Finally, the slices were clarified in xylene for 10 min and sealed with neutral gum for observation.

DNA was extracted from fecal samples using a TIANamp Stool DNA Kit (#DP328-02, Tiangen, Beijing, China). PCR amplification and library construction were conducted using Phusion enzyme (K1031, APExBIO, Houston, TX, USA) and the V3-V4 region primers (341F 5′-CCTACGGGNGGCWGCAG-3′ and 805R 5′-GACTACHVGGGTATCTAATCC-3′) of the 16S rRNA gene. An Illumina NovaSeq6000 instrument was used for paired-end (PE250) sequencing to collect raw data. Qiime 2 (2020.2) was used to conduct data quality control, calculate the $\alpha$ diversity index, and measure relative abundance. Each Amplicon Sequence Variant/Operating Taxonomic Units (ASV/OUT) sequence was annotated by referring to the silva-132-99 database, and the corresponding species information and abundance distribution were obtained. R software 4.2.3 (University of Auckland, Auckland, New Zealand) (Venn Diagram package) and the Jvenn web page (http://www.bioinformatics.com.cn/static/others/jvenn/example.html) were used to analyze common and unique ASVs in groups. Linear Discriminant Analysis (LDA) Effect Size analysis (LefSe, https://github.com/SegataLab/lefse) was applied to evaluate the differential microbiota.

GraphPad Prism 8.0 (GraphPad Software Inc., San Diego, CA, USA) was applied for data analysis. All experimental data were expressed as mean $\pm$ standard deviation (SD). One-way analysis of variance (one-way ANOVA) was applied to compare groups, and p 0.05 was considered statistically significant.

To assess the impact of combination therapy on the progression of HCC, mice injected with the fourth generation of Hepal-6 cells were treated with quercetin both with and without coordinate injection of anti-PD-1 antibody. The liver tissues in the Model group exhibited obvious tumors, confirming the successful construction of the orthotopically transplanted HCC model. Tumors in the liver tissues of mice were reduced when mice were treated with either quercetin or anti-PD-1 antibody alone. Combination therapy more effectively inhibited the growth of HCC than either monotreatment (Fig. 1A,B). Hematoxylin and eosin (H&E) staining of the liver tissues taken from different groups displayed that the liver cells of mice in the Control group were uniformly sized, neatly arranged, and without inflammatory infiltration. In the Model group, the volume of liver cells increased, nuclear staining deepened, and cell necrosis and fibrosis were observed. However, upon receiving combination therapy, the appearance of hepatocellular lesions was improved, and cell necrosis and inflammatory infiltration were significantly reduced (Fig. 1C). Western blot exhibited that the expressions of CD8a, CD4, and CD11b in liver tissues of the Model group were markedly downregulated compared to Control. Contrarily, combination therapy resulted in markedly increased expression of CD8a, CD4, and CD11b in liver tissues (Fig. 1D). Taken together, these findings indicated that combination therapy inhibited the progression of HCC.

Fig. 1.

The combination therapy inhibited the progression of hepatocellular carcinoma (HCC). (A) The distribution of tumors in liver tissues; (B) The total mass of liver tissues; (C) Hematoxylin and eosin (H&E) staining analysis of the pathological changes in liver tissues; (D) Western blot analysis of the expressions of CD8a, CD4, and CD11b in liver tissues. * p 0.05 vs. Control, # p 0.05 vs. Model, & p 0.05 vs. Quercetin or anti-PD-1.

To explore the impact of combination therapy on the diversity of GM in the orthotopically transplanted HCC model mice, 16S rRNA sequencing was carried out on mouse fecal samples. A rank-abundance curve demonstrated increased richness and evenness of GM in the Model group, but this decreased following combination therapy (Fig. 2A). As shown in the Venn diagram, the number of unique microorganisms in Control, Model, Quercetin, anti-PD-1, and Quercetin+anti-PD-1 groups was 290, 266, 195, 195, and 242, respectively (Fig. 2B). Analysis of inter-group difference showed that the $\alpha$ diversity index of the Model group was higher than that in the Control group, but the difference was not statistically significant. However, the $\alpha$ diversity index was considerably reduced in mice that received combination therapy (Fig. 2C). At the phylum level, the GM principally consisted of Bacteroidota, Firmicutes, Actinobacteria, Verrucomicrobiota, Proteobacteria, Campilobacterota, Desulfobacterota, Patescibacteria, Cyanobacteria, Deferribacterota, and Chloroflexi. The abundance of Firmicutes was decreased but that of Bacteroidota increased in GM of mice in the Model group when compared to the Control group. However, the GM of mice in the Quercetin+anti-PD-1 antibody group displayed that the relative abundance of Bacteroidota decreased, while that of Firmicutes, Actinobacteria, and Verrucomicrobiota increased (Fig. 2D). At the genus level, the GM primarily included Muribaculaceae, Lactobacillus, Lachnospiraceae_NK4A136_group, Dubosiella, Bacteroides, Akkermansia, Enterorhabdus, Alloprevotella, Blautia, Helicobacter, Alistipes, Lachnoclostridium, Bifidobacterium, Erysipelatoclostridium, Parabacteroides, Bacilli, Oscillospiraceae, Anaerotruncus, and Rikenellaceae. The abundance of Lactobacillus and Lachnospiraceae_NK4A136_group decreased, but that of Bacteroides increased in the GM of mice in the Model group compared with that measured in the Control group. However, the GM of mice in the Quercetin+anti-PD-1 antibody group displayed that the abundance of Muribaculaceae decreased, while that of Dubosiella and Akkermansia increased (Fig. 2E). These results proved that the combination therapy influenced the diversity of mouse GM in the orthotopically transplanted HCC model.

Fig. 2.

The combination therapy regulated the diversity of gut microbiota (GM) in orthotopically transplanted HCC model mice. (A) Rank-abundance curve of the richness and evenness of GM. (B) Venn diagram of the number of GM. (C) Analysis of $\alpha$ diversity index. (D,E) Analysis of relative species abundance at phylum and genus level.

We next investigated the effect of combination therapy on the difference in GM in the HCC model. Difference analysis between groups indicated that Clostridia_UCG-014, Lachnospiraceae_UCG-006, Monoglobus, and Akkermania were enriched in the intestine of mice in the Model group. Bifidobacterium, Atopobiaceae, and Faecalibalum were enriched in the intestine of mice in the Quercetin group (Fig. 3A). Subsequently, the phylum-level analysis indicated that Campilobacterota, Deferribacterota, Patescibacteria, Actinobaciota, Proteobacteria, and Verrucomicrobiota were the key differential GM. Campilobacterota and Patescibacteria were enriched in the intestine of mice in both Control and Model groups. Deferribacterota, Actinobaciota, Proteobacteria, and Verrucomicrobiota were enriched in the intestine of mice in the Quercetin+anti-PD-1 group (Fig. 3B). These results proved that the combination therapy regulated GM variation in HCC model.

Fig. 3.

The combination therapy regulated the difference of GM in orthotopically transplanted HCC model mice. (A) Lefse analysis of differential GM between groups. (B) Heatmap of the differential GM abundance between groups at the phylum level.

To explore the impact of combination therapy on the immunity of the orthotopically transplanted HCC model mice, cytokine concentrations were examined using enzyme-linked immunosorbent assay (ELISA) and Western blot. In the blood of Model mice, the levels of IL-6 and IL-4 increased, while those of IL-10 and IFN-$\gamma$ decreased. In mice receiving combination therapy, the levels of IL-6 and IL-4 were lower, but those of IL-10 and IFN-$\gamma$ were higher than that in the Quercetin or anti-PD-1 group (Fig. 4A). The levels of GM-CSF and G-CSF were increased in the liver tissues of the Model group when compared to the Control group. However, both GM-CSF and G-CSF markedly decreased in the Quercetin or anti-PD-1 group. In addition, compared with the therapy of Quercetin or anti-PD-1, combination therapy produced a better inhibitory effect on the levels of GM-CSF and G-CSF (Fig. 4B). Next, we discovered that TLR4 expression was markedly upregulated in the Model group compared to the Control group, and the phosphorylation levels of both I$\kappa$B$\alpha$ and p65 were increased. However, following combination therapy, TLR4 expression was downregulated, and the phosphorylation of I$\kappa$B$\alpha$ and p65 was reduced (Fig. 4C). Additionally, the immunotherapy marker PD-L1 was highly expressed in the liver tissues of the Model group as determined by immunohistochemistry (IHC) staining analysis, however, its expression was suppressed in the Quercetin+anti-PD-1 group (Fig. 4D). These results confirmed that combination therapy regulated the immunity of the orthotopically transplanted HCC model mice.

Fig. 4.

The combination therapy regulated the immunity of orthotopically transplanted HCC model mice. (A) The levels of interleukin (IL)-6, IL-10, interferon (IFN)-$\gamma$, and IL-4 in blood were determined by enzyme-linked immunosorbent assay (ELISA). (B) The levels of granulocyte-macrophage colony-stimulating factor (GM-CSF) and granulocyte colony-stimulating factor (G-CSF) in liver tissues were assessed by ELISA. (C) The expressions of TLR4, I$\kappa$B$\alpha$, p-I$\kappa$B$\alpha$, p65, and p-p65 in liver tissues were analyzed by Western blot. (D) programmed cell death 1 ligand 1 (PD-L1) expression in liver tissues was characterized by IHC staining. * p 0.05 vs. Control, # p 0.05 vs. Model, & p 0.05 vs. Quercetin or anti-PD-1.

Finally, to explore the underlying mechanism of combination therapy in regulating macrophage immunity in orthotopically transplanted HCC model mice, RT-qPCR was conducted to analyze macrophage-related gene expressions. M1 macrophage-related genes, including IL-6, IL-12a, IL-1$\beta$, and TNF-$\alpha$, were highly expressed in the liver tissues of the Model group, whereas their expression levels were suppressed in those of the Quercetin+anti-PD-1 group (Fig. 5A). Moreover, the expressions of M2 macrophage-related genes, including Arg-1, IL-10, TGF-$\beta$, and MMP-9, were inhibited in the liver tissues of the Model group. However, combination therapy overcame this inhibitory effect to a large extent, resulting in the increased expression of Arg-1, IL-10, TGF-$\beta$, and MMP-9 (Fig. 5B). These findings indicated that the combination therapy regulated macrophage immunity in orthotopically transplanted HCC model mice.

Fig. 5.

The combination therapy regulated macrophage immunity in orthotopically transplanted HCC model mice. (A) The expressions of M1 macrophage-related genes IL-6, IL-12a, IL-1$\beta$, and TNF-$\alpha$ in liver tissues were quantified by RT-qPCR. (B) The expressions of M2 macrophage-related genes Arg-1, IL-10, TGF-$\beta$, and MMP-9 in liver tissues were detected by RT-qPCR. * p 0.05 vs. Control, # p 0.05 vs. Model, & p 0.05 vs. Quercetin or anti-PD-1.