Introduction

Sarcopenia, defined as “age-related loss of muscle mass along with decreased muscle strength and/or poor physical performance,” is a syndrome commonly associated with the aging process in the elderly population1. The prevalence of sarcopenia in the elderly population can range from 5–50%2. Studies have shown that abnormal lipid metabolism is closely related to skeletal muscle function in the elderly3. Skeletal muscle fat infiltration was found in the tissue sections of the elderly with sarcopenia4. Statins are low density lipoprotein receptor (LDLR) agonists that can regulate lipid metabolism and reduce blood lipids by increasing the use of low density lipoprotein (LDL) in tissues5. However, studies have found that statins have damaging effects on skeletal muscles, such as muscle soreness6,7. Nevertheless, the role of LDLR on the skeletal muscle mass and muscle strength of aging remains further elucidated.

Studies have found that aging can change the relative abundance of gut microbiota and subsequently affect skeletal muscle mass8. Supplementation with Lactobacillus as a probiotic improved age-related sarcopenia by improving skeletal muscle mass and function in mice9,10. In addition, gut microbiome mediated bile acid metabolism has been found to affect skeletal muscle mass in an aging mice model11. Unfortunately, this article lacks muscle strength evidence11. Under normal circumstances, conjugated bile acids are actively transported by the apical sodium-dependent bile acid transporter (ASBT) and then bound to the farnesoid X receptor (FXR), which activates FXR to produce fibroblast growth factor 15 (FGF15)12. FGF15 in the circulation was reported to regulate skeletal muscle protein synthesis by binding with fibroblast growth factor receptor 4 (FGFR4) on the surface of skeletal muscle cells13.

Therefore, this study aimed to explore the effect of gut microbiota on skeletal muscle strength of LDLR knockdown aging mice. We also focused on the mediating role of bile acid metabolism in gut microbiota regulation and skeletal muscle strength improvement of LDLR knockdown aging mice.

Materials and methods

Mice and intervention

All animal experiments have been approved after an animal ethics review by the Experimental Animal Ethics Committee of Youjiang Medical University for Nationalities (Approval number: 2022100801). The licensing committee approving the experiments, including any relevant details. The study was performed with regulations and by the ARRIVE guidelines.

Eight-week-old C57BL/6J male mice were purchased from Changsha Tianqin Biotechnology Company Limited and maintained in the Experimental Animal Center under pathogen-free settings and specified barrier systems at Youjiang Medical University for Nationalities. Mice have unrestricted access to standard food and clean water, raised at a constant temperature (22 ± 2 °C) and moderate humidity (50 ± 5%) while maintaining a light/ darkness cycle (12 h/12 h). Mice aged twelve weeks (adult, N = 24) and seventy-six weeks (old, N = 24) were used in the experiments. Adult mice were randomly divided into two subgroups, the Acon (young adult control mice, N = 12) group, and the ALKd (young adult LDLR knockdown mice, N = 12) group. Old mice were then randomly divided into two subgroups, the Ocon (old control mice, N = 12) group, and the OLKd (old LDLR knockdown mice, N = 12) group. At the baseline of the experiment, the Acon group and Ocon group were treated with a 1 × 1012 VP empty vector (VectorBuilder, Guangzhou, China) via the caudal vein administration. ALKd group and OLKd group were treated with 1 × 1012 VP LDLR knockdown (VectorBuilder, Guangzhou, China) via the caudal vein administration. After a single injection, mice were housed under the same conditions as previously mentioned. All animals were sacrificed 8 weeks after administration.

Ethology test

Skeletal muscle strength tests were done on the day before the mice were sacrificed. The grip strength and limb grip were measured by using a grip strength tester (Science Biological Technology, SA415). Gently lift the tail of the mouse, make the mouse’s forelimbs or limbs fully grasp the steel grab rod, and then pull the mouse’s forelimbs or limbs back horizontally at a constant speed until the mouse’s forelimbs are pulled away from the steel grab rod. Read out the maximum value during this process and record the data. The test was repeated three times for each mouse, and the average value of the three times was taken as the grasping power of the mouse’s grip strength or limb grip (g). Place the mouse’s forelimb on the suspended wire, make sure that the hind limb and tail are not entangled in the wire, start the timing until the mouse falls off the wire, read the time that the mouse remains suspended, representing the limb suspension time (s).

Sample collection

The feces of each mouse were collected on the day before the mice were sacrificed. The sterile forceps were used for feces collection. After being placed in sterilized 1.5EP tubes, the feces were frozen at minus 80 degrees Celsius refrigerator. The sample completed the testing of various indicators within one week. Independent sterile forceps were used for each mouse during sampling to prevent bacterial contamination.

The blood of each mouse was collected from the orbital venous plexus after 12 h fasting. Serums were collected following a 15-minute centrifugation at 3000 rpm and 4 °C. The gastrocnemius of each mouse was collected into a separate 1.5 ml EP tube, while the ileum was rinsed with a cold saline solution to remove feces from the intestine. All the samples were quick-frozen by liquid nitrogen and stored in the refrigerator at minus 80 degrees celsius until parameter analysis.

Serum parameters analyses

Blood glucose, triglycerides (TG), total cholesterol (TC), low density lipoprotein cholesterol (LDL-C), and high density lipoprotein cholesterol (HDL-C) levels were used the commercial colorimetric kits (Elabscience, Wuhan, China) to measure. Blood glucose, TG, TC, LDL-C, and HDL-C were determined by the optical density (OD) value at 505 nm, 510 nm, 545 nm, and 546 nm wavelength, respectively. The final value was obtained according to the instructions of the kits.

Feces 16 S rRNA sequencing analysis

In this study, the genomic DNA of feces was extracted by E.Z.N.A™ Mag-Bind soil DNA kit (Omega, Shanghai, China). A suitable quantity of DNA was diluted with sterile water to 1 ng/mL. The template for PCR amplification was then diluted with genomic DNA. The 16s rRNA primer of the V3-V4 variable region was selected according to a previous report (341 F CCTACGGGNGGCWGCAG, 805R GACTACHVGGGTATCTAATCC)14. The Illumina MiSeq platform was used for PCR product sequencing, after the purification, quantification, and quality inspection process. Then the sequencing data were clustered into operational taxonomic units (OTUs, 97% similarity) by using QIIME2 software15. The functional prediction of gut microbiota was analyzed in the Kyoto Encyclopedia of Genes and Genomes (KEGG) database16. This experiment was commissioned by Sangon Biotech Shanghai Company.

Feces high resolution Non target metabonomics analysis

In this study, take an appropriate amount of sample, add precooled methanol/acetonitrile/water solution, vortex mixing, low-temperature ultrasound for 30 min, stand at -20℃ for 10 min, and centrifuge at 14,000 g 4℃ for 20 min. During mass spectrometry analysis, add 100 µL acetonitrile aqueous solution for re dissolution, vortex, and centrifuge at 14,000 g 4℃ for 15 min. Take the supernatant and inject it into LC-MS/MS mass spectrometry for qualitative and quantitative analysis of metabolites.

Quantitative real-time PCR (qPCR) analysis

The mRNA expression levels of LDLR, ASBT, FXR, FGF15 of the ileum, and LDLR, FGFR4 of the gastrocnemius were detected. High-quality RNA was extracted from frozen intestinal/gastrocnemius tissue by using an RNA extraction kit for tissue (Tiangen Biotech, Beijing, China). The UV spectrophotometry at 260 and 280 nm was employed to determine the concentration as well as the purity of the RNA solution. 1000 ng of RNA was reverse-transcribed to cDNA by using the HiScript III RT SuperMix (Vazyme, Nanjing, China) for qPCR(+ gDNA wiper). The cDNA was mixed with the 2× Color SYBR Green qPCR Master MIX and amplified by using the LightCycler ® 96 system13. GAPDH or α-actin was used as a housekeeping gene17,18. The relative expression levels of mRNA show the content of the target gene compared to the content of the housekeeping gene. Primers were listed in Supplementary Table 1.

Western blotting

Proteins of ileum and gastrocnemius were extracted with RIPA buffer containing protease inhibitors and quantified with Bradford BCA protein assay kit (Beyotime, Shanghai, China). Denatured proteins were subjected to SDS-PAGE, transferred to the PVDF membranes, and blocked by the 5% non-fat dry milk (NFDM) solution for 2 h at room temperature (RT). The membranes were then incubated with the primary antibodies overnight at 4 °C, respectively. Including ASBT (ab203205, 1:1000; Abcam), FXR (sc-25309, 1:1000; Santa Cruz), FGF15 (sc-398338, 1:1000, Santa Cruz), HSP90 (sc-13119, 1:2000; Santa Cruz), FGFR4 (sc-136998, 1:1000, Santa Cruz), LDLR (ab52818, 1:1000, Abcam), α-actin (M4276, 1:2000, Sigma-Aldrich). Secondary antibodies conjugated to anti-rabbit (BS0295G, 1:2000, Bioss) or anti-mouse (BS0296G, 1:2000, Bioss) horseradish peroxidase were incubated on the membranes for 1 h at room temperature. The immunoreactive bands were visualized by using ECL prime western blotting detection reagent (Monad, Suzhou, China), and the band intensities were measured with the multi-automatic chemiluminescence analysis system (Tanon, 5200). The signal intensities of each antibody were quantified by Image J19. HSP90 or α-actin was used as a housekeeping gene20,21. The quantitative results of proteins show the ratio of the target protein level to the housekeeping protein level. Fold change represents the quantitative difference fold between the target protein and the housekeeping protein. It is obtained by comparing the grayscale value of the target protein with the grayscale value of the housekeeping protein22.

Statistical analysis

The experimental data were expressed by mean ± standard error of the mean (SEM). The GraphPad Prism (version 8.0.2) was used for statistical analyses. The two-tailed Student’s t-test or the Mann–Whitney U-test was used to assess the differences between the two groups. The one-way ANOVA followed by Fisher’s least significant difference post hoc test was used to assess the multiple groups. A p-value less than 0.05 was considered statistically significant.

Results

Identification of specific skeletal muscle LDLR knockdown mouse models

At the level of protein, the content of LDLR of the Ocon group was increased compared with the Acon (Fig. 1A, B). In addition, the content of LDLR in the OLKd group was lower than that in the Ocon group, and in the ALKd group was also lower than that in the Acon group (Fig. 1A, B). At the mRNA level, LDLR expression of gastrocnemius in the OLKd group scored 60.2% lower than the Ocon group, and the ALKd group was significantly lower than that in the Acon group (Fig. 1C). Furthermore, there was no difference in the expression level of LDLR in the ileum among the four groups (Fig. 1D).

Fig. 1
figure 1

The relative mRNA and protein levels of LDLR. GAPDH or α-actin served as a loading control. (A) Protein image of the gastrocnemius LDLR, (B) Protein quantitative results of gastrocnemius LDLR, (C) mRNA expression level of LDLR in the gastrocnemius, (D) mRNA expression level of LDLR in the ileum. Data are reported as mean ± SEM. Two-tailed Student’s t-test or Mann–Whitney U test was used for the comparisons between two groups, and one-way analysis of variance (ANOVA) was used for the comparisons among multiple groups. *p < 0.05 vs. Acon group, ǂp < 0.05 vs. ALKd group, †p < 0.05 vs. Ocon group. Acon adult mice with empty vector, ALKd adult mice with LDLR knockdown, Ocon old mice with empty vector, OLKd old mice with LDLR knockdown (Each group ≥ 3).

Full size image

Effect of LDLR knockdown on skeletal muscle strength in mice

Grip strength and limb suspension time of the Ocon group declined compared with the Acon and the ALKd groups, respectively. After LDLR knockdown, grip strength in the OLKd group was significantly enhanced compared with the Ocon groups. However, the limb suspension time in the OLKd group was decreased when compared with the Acon and the ALKd groups. Meanwhile, the limb suspension time in the ALKd group was prolonged by 1.43 times compared with the Acon group. In addition, the limb grip was no significant difference among the four groups (Table 1).

Table 1 The effects of LDLR knockdown on the body in aging mice.
Full size table

Effect of LDLR knockdown on serum indexes in mice

Lower levels of blood glucose and LDL-C were observed in the Ocon group than in the Acon group. After LDLR knockdown, TC and TG were found to be decreased in the OLKd group than in the Ocon group. Furthermore, decreased blood glucose, TC, TG, LDL-C, and HDL-C levels were observed in the OLKd group when compared with the ALKd group. Compared with the Acon group, the blood glucose and blood glucose/LDL-C were increased in the ALKd group (Table 1).

Effect of LDLR knockdown on gut microbiota in mice

According to the gut microbiota analysis, Chao 1, Ace, Shannon, and Simpson index showed no difference among groups at the OTU levels (Fig. 2A-D). However, principal component analysis (PCOA) was separated among groups (Fig. 2E). In addition, similar results were seen for weighted non-metric multidimensional scaling (NMDS) (Fig. 2F).

Fig. 2
figure 2

Alpha and beta diversity of the gut microbiota at OUT levels. Alpha and beta diversity of the gut microbiota at OUT levels. (A) Chao 1 index, (B) Ace index, (C) Shannon index, (D) Simpson index, (E) Principal component analysis (PCA), (F) Weighted non-metric multidimensional scaling (NMDS). Acon adult mice with empty vector, ALKd adult mice with LDLR knockdown, Ocon old mice with empty vector, OLKd old mice with LDLR knockdown (Each group = 6).

Full size image

Venn diagram shows that, at the genus level, 425 strains were found in the Acon group, 415 in the Ocon group, and 427 strains were found in both the ALKd and OLKd groups (Fig. 3A). Among the four groups, 325 common strains were observed. In addition, there were 35 independent strains in the Acon group, 27 in the Ocon group, 26 in the ALKd group, and 36 in the OLKd group. Then, through community composition analysis, we found that the proportion of dominant species at the genus level changed (Fig. 3B). The increased proportion of unclassified_Muribaculaceae, unclassied_Prevotellaceae, unclassied_Bacteroidales, Paramuribiaculum, Limosilactobacillus, Lactococcus, Faecalibaculum, Ruminococcus and decreased proportion of the unclassied_Clostrididales, Ligilactobacillus, unclassied_Erysipelotrichaceae, Bacteroides, Akkermansia, Spartobacteria_genera_incertae_sedis, Lachnospiracea_genera_incertae_sedis were found in the Ocon group than the Acon group. We observed that the proportion of Duncaniella, unclassied_Clostrididales, unclassied_Erysipelotrichaceae, Akkermansia, Ileibacterium, Faecalibaculum, Lachnospiracea_genera_incertae_sedis were increased in the OLKd group compared with the Ocon group which was caused by LDLR knockdown. Meanwhile, the decreased proportion of unclassified_Muribaculaceae, unclassied_Prevotellaceae, unclassied_Bacteroidales, Limosilactobacillus, Phocaeicola, Lactococcus, and Ruminococcus were observed in the OLKd group than in the Ocon group. In addition, unclassied_Muribaculum, Allobaculum, unclassied_Lachnospiraceae, unclassied_Bacteroidales, Paramuribiaculum, Limosilactobacillus, Ileibacterium, Lachnospiracea_ genera_ incertae_sedis were increased in the ALKd group when compared with the Acon group. Moreover, decreased proportion of unclassied_Clostrididales, Ligilactobacillus, unclassied_Erysipelotrichaceae, Bacteroides, Akkermansia, Spartobacteria_genera_incertae_sedis, Phocaeicola, Lactococcus, Faecalibaculum, and Ruminococcus were found in the ALKd group contrasted with the Acon group.

Fig. 3
figure 3

The effects of LDLR knockdown on the gut microbiota species composition in aging mice. (A) Venn: Different colors represent different groups, overlapping numbers represent the number of species common to multiple groups, and non-overlapping numbers represent the number of species unique to the corresponding groups, (B) Circos groups and species relationship diagram: The left half circle represents the species composition in the groups; The right half circle shows the proportion of species in different samples at the taxonomic level. Acon adult mice with empty vector, ALKd adult mice with LDLR knockdown, Ocon old mice with empty vector, OLKd old mice with LDLR knockdown (Each group = 6).

Full size image

LEfSe linear discriminant analysis (LDA > 3, Fig. 4A, B) showed that Akkermansia which belongs to Verrucomicrobia was the taxonomic biomarker in the Acon group and Muribaculum in the Ocon group. Moreover, the Duncaniella, Lachnospiracea_genera_incertae_sedis, Faecalibaculum, Ileibacterium in the OLKd group and the Limosilactobacillus, Allobaculum in the ALKd group might as the taxonomic biomarker of the group, respectively.

Fig. 4
figure 4

The species difference analysis in LDLR knockdown aging mice. (A) LEFSe Cladogram: Different colors represent different groups, and different color nodes in the branches represent the microbial groups that play an important role in the corresponding color groups, (B) LDA discriminant histogram (LDA > 3). Acon adult mice with empty vector, ALKd adult mice with LDLR knockdown, Ocon old mice with empty vector, OLKd old mice with LDLR knockdown (Each group = 6).

Full size image

Seven increased strains and one decreased strain of the genus were observed in the Ocon group than in the Acon group from the difference analysis results (Fig. 5A). The increased gut microbiotas were Turicibacter, Muribaculum, Romboutsia, Clostridium_sensu_stricto, unclassified_Desulfovibrionaceae, unclassified_Sutterellaceae, Lactococcus and the decreased one was unclassified_Coriobacteriia. When comes to knocking down the LDLR in old mice, three strains of the genus were observed to be increased and five to be decreased in the OLKd group (Fig. 5B). Ileibacterium, Staphylococcus, Lachnospiracea_incertae_sedis were found to be increased and unclassified_Desulfovibrionaceae, Streptococcus, Clostridium_sensu_stricto, Parabacteroides, Clostridium_XVIII were found to be decreased.

Fig. 5
figure 5

The DESeq2 analysis of gut microbiota in LDLR knockdown aging mice. The horizontal axis is the normalized abundance mean after the Log2 conversion, and the vertical axis is the difference multiple after the Log2 conversion. According to the absolute value of Log2FC > 1 is up-regulated, and padj < 0.05 is down-regulated. (A) Acon vs. Ocon, (B) Ocon vs. OLKd. Acon adult mice with empty vector, ALKd adult mice with LDLR knockdown, Ocon old mice with empty vector, OLKd old mice with LDLR knockdown (Each group = 6).

Full size image

KEGG function prediction found thirteen functional pathways of gut microbiota that Selenocompound metabolism, D-Arginine and D-ornithine metabolism, Sulfur metabolism, Glutamatergic synapse, Biosynthesis of unsaturated fatty acids, Steroid biosynthesis, Fatty acid elongation in mitochondria, Meiosis – yeast, Fluorobenzoate degradation, Chagas disease (American trypanosomiasis), Circadian rhythm – plant, Systemic lupus erythematosus and Caffeine metabolism in level 3 (Fig. 6). Moreover, at level 2, the focus was on Metabolism of Other Amino Acids, Energy Metabolism, Nervous System, Lipid Metabolism, Cell Growth and Death, Xenobiotics Biodegradation and Metabolism, Infectious Diseases, Environmental Adaptation, Immune System Diseases; Biosynthesis of Other Secondary Metabolites.

Fig. 6
figure 6

The gut microbiota function prediction in LDLR knockdown aging mice. KEGG functional abundance heat map, drawn with a functional abundance matrix, in which each column represents a sample, rows represent functions, and color blocks represent functional abundance values. The redder the color, the higher the abundance, the bluer the color, and vice versa. Acon adult mice with empty vector, ALKd adult mice with LDLR knockdown, Ocon old mice with empty vector, OLKd old mice with LDLR knockdown (Each group = 6).

Full size image

Effect of LDLR knockdown on bile acid metabolites in mice

A total of 979 metabolites were detected in feces, of which 26 were bile acid metabolites, and there were 6 bile acid metabolites with statistically significant differences (Fig. 7). There were 9 bile acid metabolites related to behavior, serum indicators, and FGF15, of which 4 were secondary bile acids and 5 were primary bile acids.

Fig. 7
figure 7

The bile acid metabolites function prediction in LDLR knockdown aging mice. Heat map of accumulation of bile acid metabolites, drawn with a functional abundance matrix, in which each column represents a sample, rows represent functions, and color blocks represent functional abundance values. The redder the color, the higher the abundance, the bluer the color, and vice versa. Acon adult mice with empty vector, ALKd adult mice with LDLR knockdown, Ocon old mice with empty vector, OLKd old mice with LDLR knockdown (Each group = 6).

Full size image

Effect of LDLR knockdown on mRNA gene expression of bile acid metabolism

A reduced mRNA gene expression level of FGF15 was found in the Ocon compared to the Acon group (Fig. 8C). After LDLR knockdown, FGF15, and FGFR4 were observed to be increased in the OLKd group compared with the Ocon group (Fig. 8C, D). Furthermore, FGF15 expression was increased in the OLKd group with the ALKd group, but reduced in the ALKd group than the Acon group (Fig. 8C). There was no significant difference in ASBT and FXR expression among the four groups (Fig. 8A, B).

Fig. 8
figure 8

The relative mRNA levels in LDLR knockdown aging mice. GAPDH or α-actin served as a loading control. (A) mRNA expression level of LDLR in gastrocnemius, (B) mRNA expression level of LDLR in the ileum, (C) mRNA expression level of ASBT in the ileum, (D) mRNA expression level of FXR in the ileum, (E) mRNA expression level of FGF15 in the ileum, (F) mRNA expression level of FGFR4 in gastrocnemius. Data are reported as mean ± SEM. Two-tailed Student’s t-test or Mann–Whitney U test was used for the comparisons between two groups, and one-way analysis of variance (ANOVA) was used for the comparisons among multiple groups. *p < 0.05 vs. Acon group, ǂp < 0.05 vs. ALKd group, †p < 0.05 vs. Ocon group. Acon: Adult mice with empty vector; ALKd: Adult mice with LDLR knockdown; Ocon: Old mice with empty vector; OLKd: Old mice with LDLR knockdown (Each group ≥ 3).

Full size image

Effect of LDLR knockdown on the protein level of bile acid metabolism in mice

Consistent with mRNA levels, there were reduced protein levels of FXR and FGF15 in the Ocon group compared to the Acon group (Fig. 9C-F). In addition, the protein levels of FGF15 and FGFR4 were found to be increased in the OLKd group compared with the Ocon group (Fig. 9E-H). Moreover, FXR and FGF15 were increased in the OLKd group than in the ALKd group. On the other hand, those parameters were reduced in the ALKd group than in the Acon group (Fig. 9C-F). The protein level of ASBT did not change among the four groups (Fig. 9A, B).

Fig. 9
figure 9

The protein levels in the ileum and gastrocnemius in LDLR knockdown aging mice. HSP90 or α-actin served as a loading control. (A) Protein image of ileum ASBT, (B) Protein quantitative results of ileum ASBT, (C) Protein image of ileum FXR, (D) Protein quantitative results of ileum FXR, (E) Protein image of ileum FGF15, (F) Protein quantitative results of ileum FGF15, (G) Protein image of the gastrocnemius FGFR4, (H) Protein quantitative results of gastrocnemius FGFR4. Data are reported as mean ± SEM. Two-tailed Student’s t-test or Mann–Whitney U test was used for the comparisons between two groups, and one-way analysis of variance (ANOVA) was used for the comparisons among multiple groups. * p < 0.05 vs. Acon group, ǂ p < 0.05 vs. ALKd group, † p < 0.05 vs. Ocon group. Acon: Adult mice with empty vector; ALKd: Adult mice with LDLR knockdown; Ocon: Old mice with empty vector; OLKd: Old mice with LDLR knockdown (Each group ≥ 3).

Full size image

Correlation analysis between muscle-related parameters and gut microbiota in mice

As shown in the heat map, the grip strength was negatively correlated with unclassified_Muribaculaceae (Fig. 10). The limb suspension time was negatively correlated with Faecalibaculum, Akkermansia, and Lactococcus but positively correlated with Saccharibacteria_genera_incertae_sedis. The blood glucose was positively correlated with the Allobaculum. TC was negatively correlated with Faecalibaculum and Ileibacterium but positively correlated with Saccharibacteria_genera_incertae_sedis. Moreover, TG was negatively correlated with Lachnospiracea_incertae_sedis. HDL-C was negatively correlated with Akkermansia and Lachnospiracea_incertae_sedis but positively correlated with Streptococcus. In addition, blood glucose/LDL-C was negatively correlated with Clostridium_sensu_stricto. Furthermore, FGF15 was positively correlated with Akkermansia and Ileibacterium but negatively correlated with Streptococcus. The correlation results of bile acid metabolites are as follows : limb suspension time was positively correlated with cholic acid, 3.alpha.,7.alpha.-dihydroxy-12-oxocholanoic acid and 3.alpha.-hydroxy-7-oxo-5.beta.-cholanic acid. TC was positively correlated with 3-oxocholic acid. HDL-C was positively correlated with deoxycholic acid and 3-oxocholic acid. TG was negatively correlated with ursocholanic acid. Furthermore, FGF15 was positively correlated with chenodeoxycholate and trihydroxycholestanoic acid but negatively correlated with glycocholic acid.

Fig. 10
figure 10

Correlations between muscle-related parameters and gut microbiota in LDLR knockdown aging mice. Data are analyzed by using Spearman’s rank correlation analysis. Red indicates a positive correlation and blue indicates a negative correlation. * p < 0.05. TC total cholesterol, TG triglycerides, LDL-C low density lipoprotein cholesterol, HDL-C high density lipoprotein cholesterol.

Full size image

Discussion

Sarcopenia is a syndrome of skeletal muscle abnormalities associated with age progression23. Statins are often used clinically to increase the expression of LDLR to reduce blood lipids24. However, one of the side effects of statin use is muscle damage25. In addition, there are lipid droplets formed by fat in the skeletal muscle of elderly patients with sarcopenia, which affect the quality and function of the skeletal muscle of patients26. Therefore, we focused on whether LDLR knockdown can improve skeletal muscle strength in aging mice.

A study has found that skeletal muscle atrophy in germ-free mice, and bacterial metabolites can affect skeletal muscle mass27. Antibiotic treatment caused skeletal muscle mass loss in mice, which was reversed after gut microbiota transplantation. However, all of these studies were done with young mice28. Besides, increased skeletal muscle function and mass were found in probiotic-treated C57BL/6 young or old mice29,30. Gut microbiota and bile acid metabolism of aging C57BL/6J mice affected skeletal muscle mass under a normal diet20. However, they did not focus on the effect of LDLR knockdown on skeletal muscle. In addition, accelerated-aging mice can be used to study age-related diseases, but they show characteristics not seen in normal older mice31. Therefore, naturally aging C57BL/6J mice are more suitable for confirming the role of gut microbiota on skeletal muscle function in LDLR knockdown mice.

We observed reduced grip strength and shortened limb suspension time in aging mice compared with adult mice (Fig. 11). Contrary to previous studies32, aging mice exhibited lower levels of blood glucose and LDL-C, indicative of reduced grip strength and short limb suspension time. At the same time, the decreased blood glucose and LDL-C may be attributed to impaired ileal reabsorption in aging mice. One of the pieces of evidence was an alteration of gut microbiota. Despite finding that there was no difference in the alpha diversity, the beta diversity indicated that the gut microbiota of adult and aging mice was distinct. We consider that gut microbiota breakdown results in carbohydrate loss and decreased circulating glucose33. In addition, FGF15 showed a marked decline in aging mice compared to adult mice. However, the ASBT, FXR, and FGFR4 showed no difference in mRNA or protein levels between adult mice and aging mice. Suggested that the decrease of FGF15 production caused by gut microbiota disorder could be one of the causes of weakened skeletal muscle strength in aging mice20.

Fig. 11
figure 11

Disordered gut microbiota with decreased FGF15 and muscle strength were found in aging mice. After LDLR knockdown, remodeled gut microbiota with increased FGF15, FGFR4 and muscle strength were observed.

Full size image

Our results showed that LDLR knockdown can significantly improve grip strength in aging mice. When compared with the control group, a respectively increased limb grip was observed in OLKd and ALKd groups, but there was no statistical significance. However, LDLR knockdown does not increase limb suspension time. At the same time, TC, TG, and LDL-C at the circulating level were not increased in LDLR knockdown group. Suggested that glycolipid metabolism in skeletal muscle was improved by LDLR knockdown, but there was no significant increase in ATP production of skeletal muscle cells. However, subsequent further research is required to confirm this conjecture.

Notably, after LDLR knockdown, the substantially altered gut microbiota of aging mice was found. In detail, the proportion of Akkermansia, Ileibacterium, Faecalibaculum, Lachnospiracea_genera_incertae_sedis, Duncaniella, unclassied_Clostrididales, unclassied_Erysipelotrichaceae were increased. Meanwhile, the decreased proportion of unclassified_Muribaculaceae and Streptococcus in LDLR knockdown aging mice were observed. Previous studies showed that supplementation with Akkermansia improved muscle mass and strength in aging mice which were consistent with our results29,34. We also found that Akkermansia was the dominant genus in adult mice from the LEFse analysis. At the same time, we found a positive correlation between Akkermansia and FGF15 of ileum tissue. The FGF15 of ileum and its receptor FGFR4 located on the skeletal muscle cell surface were increased in aging mice after LDLR knockdown. FGF15 enhances muscle mass and function in aging mice as reported by the previous study20. These results suggest that LDLR knockdown remodeled gut microbiota and may improve muscle strength in aging mice by enhancing the FGF15-FGFR4 pathway. In addition, Ileibacterium has been found to play an important role in predicting skeletal muscle mass35. In our work, we found that Ileibacterium increased in LDLR knockdown aging mice and has become the dominant genus, which is consistent with the reference. Besides, Ileibacterium has a positive correlation with TC and FGF15. This is consistent with previous research showing that Ileibacterium can decrease blood lipids36. These results suggest that Ileibacterium may improve muscle mass through bile acid metabolism. Faecalibaculum was found to increase the glycolipid metabolism and AMPK signaling pathway in skeletal muscle37. Our results showed an increased proportion in Faecalibaculum, suggesting that aging mice with LDLR knockdown had improved skeletal muscle metabolism. Lachnospiracea, Duncaniella, unclassied_Clostrididales, and unclassied_Erysipelotrichaceae in high abundance have all been reported to prevent obesity38,39,40. These indicated that the four genera may change the lipid metabolism to improve the skeletal muscle. Furthermore, the unclassified_Muribaculaceae were known as gram-negative bacteria41. The major component of gram-negative bacteria cell walls, lipopolysaccharide, has been linked to inflammation42. The decreased unclassified_Muribaculaceae may indicate that intestinal inflammation in the aging mice was alleviated after LDLR knockdown. In addition, gram-positive bacteria Ileibacterium, Faecalibaculum and anti-inflammatory bacteria Lachnospiracea incertae sedis, Duncaniella were the dominant genera after LDLR knockdown, which together proved that intestinal inflammation was alleviated43,44,45,46. One study found that Streptococcus was negatively associated with muscle improvement outcomes47. Our results showed a reduction in Streptococcus and negatively associated with FGF15, suggesting that aging mice with LDLR knockdown had recovered skeletal muscle damage.

KEGG results showed that gut microbiota mainly through Metabolism of Other Amino Acids, Energy Metabolism, Nervous System, Lipid Metabolism, Cell Growth and Death, Xenobiotics Biodegradation and Metabolism, Infectious Diseases, Environmental Adaptation, Immune System Diseases, Biosynthesis of Other Secondary Metabolites. However, the secondary classification with the highest proportion were Metabolism of Other Amino Acids and Lipid Metabolism which may plays the important role of the gut microbiota in mice.

There were statistically significant secondary bile acids deoxycholic acid, 3-oxocholic acid, Ursocholanic acid and 3.alpha.-hydroxy-7-oxo-5.beta.-cholanic acid in the LDLR knockdown group. It is generally believed that secondary bile acids are formed after primary bile acids are metabolized by intestinal flora. These secondary bile acids are correlated with behavior, serum indicators, and FGF15, indicating that the intestinal flora affects the muscle strength of LDLR knockdown mice by affecting bile acid metabolism.

Limitations of the study

In this study, the bile acids concentration of feces and blood have not been measured, and there was no direct evidence of the role of bile acids on skeletal muscle strength. This will necessitate additional research in the future. Additionally, there was no explicit evidence that LDLR knockdown directly affected the FGF15-FGFR4 pathway of skeletal muscle. This should be the focus of the following research. The amount of serum obtained was small, and other biomarkers could not be verified.

Conclusions

Our results suggest that LDLR knockdown partially improved skeletal muscle strength in aging mice possibly by altering gut microbiota. The underlying mechanism of gut microbiota alteration on skeletal muscle strength improvement may be attributed to the effect of bile acid metabolism in LDLR knockdown aging mice.