European Bilberry Extract Ameliorates Dietary Advanced Glycation End Products-Induced Non-Alcoholic Steatohepatitis in Rats via Gut Microbiota and Its Metabolites

The EBE (ACNN15^®, supplied by By-Health Co., Ltd. Guangzhou, China) comprised a mixture of fifteen anthocyanins and five anthocyanidins. The complete chemical profile of this compound was reported previously [22].

Thirty female Sprague–Dawley rats (12 weeks old) were supplied by Vital River Laboratory Animal Center (Beijing, China). The animals were housed under specific pathogen-free conditions with a controlled temperature and a 12 h:12 h light-dark cycle. Food and water were available ad libitum.

The study was approved by the Institutional Animal Care and Use Committee at Tongji Medical College, Huazhong University of Science and Technology, and performed in accordance with the National Research Council’s Guide for the Care and Use of Laboratory Animals (Permission ID:2519, date on 15 July 2020).

The high-AGE diet was produced by baking standard chow (AIN-93M, Medison Biopharmaceutical Co., Ltd. Yangzhou, China) at 150 °C for 30 min, following an established protocol [22]. The nutritional and caloric composition of the standard and high-AGE diets were available in our previous study [22].

After a one-month acclimation period, rats were randomly allocated into three groups (n = 10 each), namely control, high-AGE diet, and EBE groups. The EBE group received the high-AGE diet and a daily oral gavage of 150 mg/kg EBE. The AGE group received the high-AGE diet and a daily oral gavage of vehicle (normal saline). The control group received the standard chow diet and a daily oral gavage of vehicle. Following 80 weeks of intervention, feces were collected for 16S rRNA analysis and SCFA detection. Rats then underwent oral glucose tolerance test (OGTT) and intraperitoneal insulin tolerance test (iPITT). After an overnight fast, the rats were anesthetized with isoflurane (5% v/v) and sacrificed via descending aorta blood collection. Serum was obtained to measure SCFA, AGE, and liver function biomarkers. Sections of liver tissue were either fixed in 4% paraformaldehyde for pathological examination or snap-frozen for subsequent downstream analysis. Throughout the study, body weight was measured twice weekly, and food intake was recorded daily. Four rats in the control group, five in the high-AGE diet group, and five in the EBE group were excluded due to mortality from causes such as tumors, intestinal obstruction, or chronic infection, as assessed by veterinarians.

Following an overnight fast, blood glucose levels were measured at baseline. The rats were then orally gavaged with a 2 g/kg dose of glucose (20% solution; Sigma-Aldrich, St. Louis, MO, USA). Blood glucose levels were subsequently monitored at 15, 30, 45, 60, 90, and 120 min post-administration. Glucose tolerance was assessed by calculating the area under the curve (AUC) value.

Following a 6 h fast, blood glucose levels were measured at baseline. Then, rats were intraperitoneal injected with 0.75 U/kg insulin (Novo Nordisk, Copenhagen, Denmark). Blood glucose levels were subsequently detected at 15, 30, 45, 60, 90, and 120 min post-administration. Insulin sensitivity was assessed by calculating the AUC value.

The concentrations of two major AGEs, N^ε-carboxymethyllysine (CML) and N^ε-carboxyethyllysine (CEL), in diet, serum, and liver samples were quantified using ultra-high performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS) according to our established methodology [22]. These AGEs were analyzed in both free and bound forms.

For the free forms, serum was mixed with the internal standard and deproteinized using methanol and acetonitrile. After centrifugation, the supernatant was derivatized with Nonafluorovaleric Acid aqueous solution for UHPLC-MS/MS analysis. For the bound forms, serum was reduced with sodium borohydride and precipitated with trichloroacetic acid. After centrifugation, the collected precipitates were combined with the internal standard and hydrolyzed in hydrochloric acid. The hydrolysate was then concentrated under a nitrogen stream, reconstituted in pure water, and centrifuged. The resulting supernatant was finally derivatized with Nonafluorovaleric Acid aqueous solution prior to UHPLC-MS/MS analysis. Diet and liver samples were homogenized in water and processed following the same procedure as serum.

Hepatic total cholesterol (TC) and triglyceride (TG) levels, along with serum alanine aminotransferase (ALT) and aspartate transaminase (AST) activities, were quantified using commercial assay kits (Elabscience, Wuhan, China) according to the manufacturers’ protocols. The levels of tumor necrosis factor (TNF)-α, interleukin (IL)-1β, IL-6, and IL-10 in the liver were measured using Elisa kits (R&D Systems, Minneapolis, MN, USA) as per the provided protocols.

Fecal microbial DNA was extracted using a commercial kit (TIANGEN Biotech Co. Ltd., Beijing, China) and quantified with NanoDrop™ ND-2000 (Thermo Fisher Scientific Ltd., Waltham, MA, USA). The V3-V4 hypervariable region of the 16S rRNA gene was amplified with primers V3F and V4R, and the resulting amplicons were sequenced on an Illumina MiSeq platform (Illumina, San Diego, CA, USA). Sequencing reads were processed in QIIME (v.1.9.1), where they were clustered into operational taxonomic units (OTUs) at a 97% similarity threshold for downstream analysis.

The β diversity was performed by non-metric multidimensional scaling (NMDS) analysis based on the weighted Unifrac distance algorithm. Intergroup differences in microbial composition at various taxonomic levels were evaluated based on relative abundance, with a false discovery rate (FDR) < 0.05 considered significant. The heatmap was generated by the ggplot2 R package (version 4.3.0, RStudio, Boston, MA, USA).

The levels of SCFA in feces and serum were measured by gas chromatography–mass spectrometry/mass spectrometry analysis (GC-MS/MS, Agilent Technologies, Santa Clara, CA, USA) as previously described [23]. Briefly, fecal samples were homogenized in ultrapure water and centrifuged to collect the supernatant. Aliquots of both the fecal supernatant and serum were separately mixed with acetone. After centrifugation, the resulting supernatant was derivatized by mixing with acetone, 100 μL 100 mM pentafluorobenzyl bromide solution, 2-ethylbutyric acid, and phosphate-buffered saline at 60 °C for 2 h, followed by the addition of hexane. The mixture was centrifuged, and the resulting supernatant was collected for subsequent GC-MS/MS analysis.

Histopathological evaluation was conducted on paraffin-embedded liver sections (7–8 μm). These sections were routinely stained with hematoxylin and eosin (H&E) for the assessment of general tissue architecture and morphological changes. To specifically evaluate hepatic fibrosis, consecutive paraffin sections were stained with Sirius red according to standard protocols. Meanwhile, lipid accumulation within the liver was assessed by staining fresh frozen sections (10 μm thick) with Oil red O. All stained sections from the various procedures were systematically examined and imaged under a bright-field microscope (Olympus IX71, Tokyo, Japan). Finally, the resulting images were subjected to quantitative analysis using ImageJ (version 1.54p, Bethesda, MD, USA) software to obtain objective measurements of the pathological features.

The paraffin-embedded liver sections (7–8 µm) were first deparaffinized, rehydrated, and subjected to antigen retrieval. The sections were then incubated with a primary antibody against α-smooth muscle actin (α-SMA), followed by a biotin-labeled secondary antibody. The antigen–antibody complexes were visualized using a DAB kit (Zs bio, Beijing, China). Stained images were captured under an Olympus IX71 microscope and quantitatively analyzed with ImageJ (version 1.54p, Bethesda, MD, USA) software.

Protein extraction from liver tissues was performed using RIPA lysis buffer (Beyotime Biotechnology Co., Ltd. Shanghai, China) containing a protease inhibitor cocktail. Following centrifugation, the supernatant was collected for protein quantification with a BCA kit (Beyotime Biotechnology Co., Ltd. Shanghai, China). Proteins were separated via SDS-PAGE, transferred to nitrocellulose membranes, and blocked with 5% milk. The membranes were then incubated with specific primary antibodies, followed by HRP-conjugated secondary antibodies. The primary antibodies were as follows: GPR43 (Cat #19952-1-AP, 1:1000, Proteintech (Rosemont, IL, USA)), HDAC3 (Cat# ab32369, 1:1000, Abcam (Cambridge, UK)), HMGB1 (Cat# ab79823, 1:1000, Thermo Scientific (Waltham, MA, USA)), RAGE (Cat# ab216329,1:1000, Abcam (Cambridge, UK)), p-NF-κB (Cat# 3033, 1:1000, CST (Danvers, MA, USA)), NF-κB (Cat# 8242,1:1000, CST (Danvers, MA, USA)), and GADPH (Cat# 3670S, 1:1000, CST (Danvers, MA, USA)). Quantitative analysis was performed with the Image J (version 1.54p, Bethesda, MD, USA) software.

Prior to statistical analysis, data were assessed for normality and homogeneity of variance. For parametric data, comparisons between two groups were performed using a two-tailed unpaired Student’s t-test; multi-group comparisons employed one-way ANOVA with Tukey’s post hoc test. Non-parametric data were analyzed by the Kruskal–Wallis test followed by Dunn’s post hoc analysis. R p < 0.05 was regarded as significant. Data were shown as mean ± SEM.

As shown in Figure 1A, high temperature baking significantly increased the levels of free-/bound-CML and free-/bound-CEL in diets (p < 0.01, Figure 1A). Consuming this high-AGE diet led to a marked elevation of free-/bound-CML and free-/bound-CEL in the circulation, and of free-/bound-CEL in the liver (p < 0.001, Figure 1A,B). Notably, EBE intervention attenuated the elevated levels of circulating free-/bound-CML and free-/bound-CEL caused by the high-AGE diet. (p < 0.001, Figure 1B). Similarly, EBE intervention also alleviated the elevated levels of free-/bound-CML in the liver caused by the high-AGE diet (p < 0.001, Figure 1C). There were no significant differences in the levels of free-/bound-CEL in the liver among the three groups (Figure 1C). Overall, EBE mitigated the accumulation of AGEs in the circulation and liver caused by a long-term high-AGE diet.

To evaluate the effect of EBE and a high-AGE diet on glucose homeostasis, we performed OGTT and IPITT. Compared with the control group, rats in the AGEs group exhibited increases in blood glucose levels and corresponding AUC values following glucose administration and insulin injection (p < 0.05, Figure 2A–D), indicating impaired glucose tolerance and insulin sensitivity. Meanwhile, compared with the AGEs group, blood glucose levels and corresponding AUC values decreased in the EBE group after glucose administration and insulin injection (p < 0.05, Figure 2A–D). These findings demonstrated that EBE intervention mitigated the impairment of glucose tolerance and insulin sensitivity caused by a high-AGE diet.

To explore the effect of a high-AGE diet on the liver, we conducted liver pathological examination. Compared with the control group, evident lipid droplets and inflammatory cell infiltration were observed in the livers of high-AGE diet-fed rats (Figure 3A,B). However, EBE reduced liver inflammation and steatosis in high-AGE diet-fed rats (Figure 3A,B). According to Sirius red staining and α-SMA staining, increased α-SMA expression and fibrogenesis were induced by a high-AGE diet, whereas EBE inhibited liver fibrosis in high-AGE diet-fed rats (Figure 3C,D). Liver steatosis is directly associated with the accumulation of TC and TG in the liver. Furthermore, liver inflammation and fibrosis are reflected by elevations of ALT and AST. Corresponding with the pathological results, the levels of TC and TG in the liver and the levels of ALT and AST in serum increased in the high-AGE diet group when compared with the control group (p < 0.01, Figure 3E). However, high-AGE diet-induced increases in the levels of TC and TG in the liver and the levels of ALT and AST in serum were reduced by EBE (p < 0.05, Figure 3E). Overall, EBE alleviated high-AGE diet-induced liver inflammation, steatosis, fibrosis, and dysfunction in rats.

16S rRNA analysis was used to investigate the effect of EBE on the gut microbiota. NMDS analysis showed that each sample was clustered within the group, and separated from other groups, which supported that the gut microbiota was obviously different among the three groups (p < 0.001, Figure 4A). At the phylum level, a high-AGE diet-induced increased abundance of the Firmicutes phylum and decreased abundance of the Bacteroidetes phylum were altered by EBE (p < 0.01, Figure 4B). At the genus level, beneficial bacteria, such as Bacteroides, Leuconostoc, Bifidobacterium, Eubacterium, Parabacteroides, Akkermansia, and Lactococcus, were reduced in the high-AGE diet group compared with the control group (p < 0.05, Figure 4C). Meanwhile, deleterious bacteria, such as Eubacterium_hallii_group, Collinsella, and Eubacterium_nodatum_group were elevated in the high-AGE diet group compared with the control group (p < 0.05, Figure 4C). However, the high-AGE diet-induced decreases in the abundance of Bacteroides, Leuconostoc, Bifidobacterium, Eubacterium, Lactococcus and increases in the abundance of Eubacterium_hallii_group, Collinsella, Coprococcus, and Eubacterium_nodatum_group were ameliorated by EBE (p < 0.05, Figure 4C). Collectively, EBE intervention modulated gut dysbiosis in high-AGE diet-fed rats.

We then explored the impacts of EBE on the common microbial metabolites—SCFA. The levels of acetate, propionate, and butyrate in feces and serum were reduced in the high-AGE diet group compared to those in the control group (p < 0.05, Figure 5A,B). High-AGE diet-induced decreases in acetate and propionate levels in feces and serum were alleviated by EBE (p < 0.05, Figure 5A,B). EBE had no effect on the level of butyrate in feces and serum (p > 0.05, Figure 5A,B). These results demonstrated that EBE intervention promoted the production of acetate and propionate in high-AGE diet-fed rats.

After detecting alterations in SCFA levels, we further investigated changes in SCFA receptors and their downstream signaling pathways in the liver. The expression level of GPR43 was downregulated, and the expression level of HDAC3 was upregulated in the high-AGE diet group compared with the control group (p < 0.001, Figure 6A,B). The high-AGE diet induced the downregulation of GPR43, and the upregulation of HDAC3 was significantly altered by EBE (p < 0.01, Figure 6A,B). The expression levels of HMGB1, RAGE, p-NF-κB, and NF-κB were elevated in the high-AGE diet group compared with the control group (p < 0.001, Figure 6A,B). A high-AGE diet induced the upregulation of HMGB1, RAGE, p-NF-κB, and NF-κB, and was also reduced by EBE (p < 0.01, Figure 6A,B). In addition, the levels of TNF-α, IL-1β, and IL-6 increased, while the level of IL-10 decreased in the high-AGE diet group compared with the control group (p < 0.0001, Figure 6C). The high-AGE diet-induced changes in inflammatory factor levels were mitigated by EBE (p < 0.05, Figure 6C). Overall, EBE modulated SCFA receptor and inhibited the activation of HMGB1/RAGE/NF-κB signaling pathway in the liver caused by the high-AGE diet.