Association between serum tumor necrosis factor-α and sarcopenia in liver cirrhosis

Abstract

Background/aims: Sarcopenia is an independent prognostic factor of liver cirrhosis (LC). However, the association between LC-related systemic inflammation and sarcopenia is unclear.

Methods: Sprague-Dawley rats were treated with thioacetamide (TAA) or saline as a control. Rifaximin was administered to TAA-induced LC rats. Enzyme-linked immunosorbent assay was performed to measure inflammatory mediators in rat serum. RT-PCR was performed to measure the molecular expression in tissues. Hematoxylin and eosin (H&E) staining and immunohistochemistry were performed to investigate tissue pathology. Serum tumor necrosis factor-α levels, liver stiffness (LS), and the L3 skeletal muscle index (L3SMI) were measured in 60 patients with chronic liver disease.

Results: LC and sarcopenia were successfully induced by TAA. Serum TNF-α levels were increased in LC rats and correlated with myostatin expression, muscle weight, and myofiber diameter. The expression of intestinal occludin and zona occludens-1 was reduced in LC rats and associated with serum TNF-α levels and sarcopenia. In patients with LS ≥7 kPa or sarcopenia, serum TNF-α levels were significantly increased, which was also confirmed when we raised the LS cutoff to 10 kPa. The L3SMI was inversely correlated with serum TNF-α levels in patients with LS ≥7 kPa. TNF-α was reduced by rifaximin, which might have resulted in reduced expression of muscular MuRF1 and myostatin and improvements in myofiber diameters within muscle tissues.

Conclusion: These results suggest that serum TNF-α is associated with LC-related sarcopenia. Rifaximin might be effective in reducing serum TNF-α levels and improving sarcopenia in LC, but these results need to be validated in future studies.

Keywords: Liver cirrhosis; Rifaximin; Sarcopenia; Tumor necrosis factor-alpha.

Figure 1.

Rat model of liver cirrhosis (LC) and sarcopenia. (A) Summary of the rat model of LC. Six-week-old Sprague-Dawley rats were intraperitoneally administered thioacetamide (200 mg/kg) or saline control three times per week for 10 weeks, and tissues from each organ of interest and blood were collected at 13 weeks. (B) Examples of gross, hematoxylin and eosin (H&E)-stained, and Sirius red-stained livers from the control and LC groups. (C) Comparison of calf muscle weights in the control (n=6) and LC (n=4) groups. (D, E) Comparison of myofiber diameters in the control (n=6) and LC (n=4) groups. H&E staining (D) and the graph (E). (F) Comparison of myostatin expression in the calf muscles of control (n=6) and LC (n=4) rats were measured by RT-PCR. (G, H) Comparison of myostatin staining in the calf muscles of control (n=6) and LC (n=4) rats were measured by immunohistochemistry (IHC). Histological findings (G) and the graph (H). TAA, thioacetamide; IP, intraperitoneal. ^*P<0.05. ^**P<0.01.

Figure 2.

Inflammatory mediators of liver cirrhosis (LC) and sarcopenia. (A, B) Enzyme-linked immunosorbent assay was performed to determine the serum levels of lipopolysaccharides (LPS), interleukin-6 (IL-6), and tumor necrosis factor-α (TNF-α). (A) Serum LPS levels were compared between the control (n=6) and LC (n=4) groups. Serum IL-6 levels were compared between the control (n=4) and LC (n=4) groups. (B) Serum TNF-α levels were compared between the control (n=4) and LC (n=4) groups. (C-E) Correlation analyses were performed between serum TNF level and myostatin expression (C, n=8), muscle weight (D, n=8), and myofiber diameter (E, n=8). n.s, not significant. ^*P<0.05.

Figure 3.

Association of tumor necrosis factor-α (TNF-α) and sarcopenia in human subjects. TNF-α levels in the serum of patients with chronic liver disease were determined by Luminex assays. (A) Comparison of serum TNF-α levels between patients with liver stiffness (LS) <7 (n=17) and patients with LS ≥7 (n=43). (B) Comparison of serum TNF-α levels between patients without sarcopenia (n=42) and patients with sarcopenia (n=18). (C) Comparison of serum TNF-α levels among the LS <7/sarcopenia (-) (n=13), LS <7/sarcopenia (+) (n=4), LS ≥7/sarcopenia (-) (n=29), and LS ≥7/sarcopenia (+) groups (n=14). (D) Patients with LS ≥7 were analyzed. The L3SMI was calculated using the BMI_CT program. Correlation between serum TNF-α levels and the L3SMI in male subjects (n=31) (left) and correlation between serum TNF-α levels and the L3SMI in female subjects (n=12) (right). L3SMI, L3 skeletal muscle index. ^*P<0.05. ^**P<0.01.*^**P<0.001.

Figure 4.

Intestinal tight junctional molecules, tumor necrosis factor-α (TNF-α), and sarcopenia in liver cirrhosis (LC). (A) The expression of occludin and zona occludens (ZO)-1 in the intestines (terminal ileum) of control (n=6) and LC (n=4) rats was measured by RT-PCR. (B) Correlation analyses between serum TNF-α levels and occludin or ZO-1 expression (n=8). (C) Correlation analysis between calf muscle weight and occludin or ZO-1 expression (n=10). (D) Correlation analysis between calf muscle myofiber diameter and occludin or ZO-1 expression (n=10). (E, F) The expression of occludin and ZO-1 in the intestines (terminal ileum) of control (n=6) and LC (n=4) rats was measured by immunohistochemistry (IHC). Histological findings (E) and the graphs (F). RT-PCR, real-time quantitative polymerase chain reaction. ^*P<0.05. ^**P<0.01.

Figure 5.

Effects of rifaximin on tumor necrosis factor-α (TNF-α) and sarcopenia. (A, B) Enzyme-linked immunosorbent assay was performed to analyze the serum of liver cirrhosis (LC) (n=4) and rifaximin-treated LC (n=6) rats. Serum ammonia (A) and TNF-α (B) levels were compared between the groups (A). (C, D) RT-PCR was performed to analyze the calf muscle tissues of LC (n=4) and rifaximin-treated LC (n=6) rats. MuRF1 (C) and myostatin (D) expression in calf muscles was compared between the groups, as measured by RT-PCR. (E, F) Immunohistochemistry (IHC) was performed to analyze myostatin in the calf muscles of LC (n=4) and rifaximin-treated LC (n=6) rats. Histological findings (E) and the graph (F). (G, H) Hematoxylin and eosin (H&E) analysis of calf muscles in LC (n=4) and rifaximin-treated LC (n=6) rats (G) and graphs (H) comparing myofiber diameter (left) and ratio of myofiber diameter: total body weight (right). n.s, not significant; RT-PCR, real-time quantitative polymerase chain reaction. ^*P<0.05.