Myostatin serum concentrations are correlated with the severity of knee osteoarthritis

Abstract

Objective

Myostatin, a member of the transforming growth factor‐β family, contributes to joint deterioration in mice. Thus, we aimed to assess the correlation of myostatin concentrations with the presence and severity of knee osteoarthritis (OA).

Material and Methods

We determined serum and synovial fluid (SF) myostatin concentrations in a population of 184 patients with knee OA and 109 healthy controls.

Results

The knee OA group presented with higher serum myostatin concentrations than the controls. Knee OA patients with KL grade 4 showed higher serum and SF myostatin concentrations compared with those with KL grade 2 and 3. Knee OA patients with KL grade 3 had higher serum and SF myostatin concentrations compared with those with KL grade 2. Serum and SF myostatin concentrations were significantly correlated with KL grading.

Conclusion

Serum and SF myostatin concentrations were correlated with the presence and severity of knee OA.

1. Introduction

Osteoarthritis (OA) is a common degenerative joint disease whose typical characteristics include articular cartilage degradation, subchondral bone damage, and synovitis.1 OA contributes to healthcare burden because of its adverse effect on ambulation and mobility.2, 3 Inflammation is considered to be a risk factor of developing OA.4 However, the underlying causes of OA remain unknown.

Myostatin is a myokine protein belonging to the transforming growth factor‐β (TGF‐β) superfamily. It is mainly expressed in skeletal muscles but is also found in cardiac muscles, blood, and adipose tissues.5 Myostatin negatively regulates skeletal muscle growth.6 Inactivation of the myostatin gene expression in mice accelerates muscle growth.7 Myostatin deficiency significantly reduces fat accumulation in mice.8 Recently, Dankbar et al.9 reported that myostatin deficiency or antibody‐mediated inhibition decreases bone destruction in an arthritis mouse model. Therefore, myostatin is likely implicated in the pathogenesis of OA.

This study aimed to examine the correlation of myostatin concentrations with the presence and severity of knee OA.

2. Materials and Methods

2.1. Patients

We enrolled a population of 184 patients with knee OA and 109 healthy subjects. Knee OA diagnosis was confirmed on the basis of the criteria of the American College of Rheumatology.10 The exclusion criteria were knee surgery history, osteochondritis dissecans, fracture in or adjacent to the knee, septic arthritis, osteonecrosis, and serious systemic diseases. Control subjects were recruited from healthy subjects who underwent medical check‐up in our hospital, and they were confirmed with normal knee radiographs using radiological examination. Subjects with systemic diseases (e.g., heart failure, renal failure, hematologic, cardiologic, and metabolic disorders), inflammatory disease, and arthritis history were excluded from the control group. This study was approved by the ethics committee of our hospital, and informed consent was obtained from all participants.

Disease severity was evaluated in terms of the Kellgren and Lawrence (KL) grading system.11 The patients with OA showed a radiographic knee OA of KL grade≥2 in at least one knee, whereas the control subjects manifested KL grades of 0.

2.2. Laboratory methods

SF was collected while the patients were taking hyaluronic acid treatment. Serum and SF myostatin concentrations were measured using a commercial enzyme‐linked immunosorbent assay kit from R&D Company [coefficient of variations (CVs) for intra assay: 1.8%‐5.4%; CVs for inter assay: 3.1%‐6%; sensitivity: 5.32 pg/mL, detection limit range: 31.3‐2000 pg/mL].

2.3. Statistical analysis

Sample size was calculated through power analysis using preliminary data obtained in our laboratory under the following assumptions: α=0.05 (two‐tailed) and power=90%. A minimum of 25 subjects in the case and control groups were required to detect the differences in serum myostatin concentrations. Data were statistically analyzed using the statistical package for social sciences (SPSS) software, version 16.0 for Windows. The data were displayed as a form of means±standard errors (quantitative data normally distributed) or median (interquartile range) (quantitative data not normally distributed). Data normality was analyzed using the Kolmogorov‐Smirnov test. Differences in the variables between the cases and control groups were determined via chi‐square tests, unpaired t test (quantitative data normally distributed), or Mann‐Whitney U‐test (quantitative data not normally distributed). Differences in myostatin concentrations between the knee OA groups were obtained via Kruskal‐Wallis test. The myostatin differences between different knee OA groups were compared through Dunn's post‐test after Kruskal‐Wallis test was conducted. The correlation of myostatin concentrations with KL grades was assessed through Spearman correlation analysis and multinomial logistic regression analysis. Statistical significance was accepted at two‐tailed P‐value less than .05.

3. Results

3.1. Clinical parameters between the two groups

No age, gender, and body mass index (BMI) differences were found between the case and control groups (Table 1).

Table 1.

The characteristics between patients with knee OA and healthy controls

Characteristics	Knee OA patients (n=184)	Healthy controls (n=109)	P‐value
Age (years)	60.77±9.08	61.48±7.93	.501
Gender (male/female)	77/107	42/67	.576
BMI (kg/m2)	24.43±3.76	24.25±3.52	.677
Serum myostatin (ng/mL)	2.71 (2.12‐3.36)	1.97 (1.55‐2.36)	<.001
SF myostatin (ng/mL)	1.50 (1.15‐1.77)

3.2. Serum myostatin concentrations in the case and control groups

The serum myostatin concentrations in the case group were higher than those in the control group (P<.001) (Table 1).

3.3. Myostatin concentrations in knee OA group

As shown in Figures 1 and 2, the serum and SF myostatin concentrations in the patients with knee OA of KL grade 4 were higher than those in the patients with knee OA of KL grade 2 and 3 The serum and SF myostatin concentrations in the patients with knee OA of KL grade 3 were higher than those in the patients with knee OA of KL grade 2.

Figure 1.

Serum myostatin concentrations in OA patients with different KL grades. ^a P<.05 vs. patients with grade 2 OA patients; ^b P<.05 vs. grade 3 OA patients

Figure 2.

SF myostatin concentrations in OA patients with different KL grades. ^a P<.05 vs. patients with grade 2 OA patients; ^b P<.05 vs. grade 3 OA patients

3.4. Correlation of myostatin concentrations with other variables

The correlation of myostatin concentrations with age, gender, and BMI were determined. The serum myostatin concentrations were negatively correlated with BMI in the control group and in the OA patient group (r=−.249, P=.009, and r=−.205, P=.005, respectively). However, the correlation between SF myostatin concentrations and BMI was not significant (r=−.095, P=.202).

3.5. Correlation of KL grades with other variables

Spearman correlation analysis revealed that the serum and SF myostatin concentrations were correlated with KL grades in OA patients (r=.600, P<.001 and r=.630, P<.001). After factors were adjusted for age, gender, and BMI, the serum and SF myostatin concentrations remained correlated with KL grades (r=.636, P<.001 and r=.639, P<.001). Multinomial logistic regression analysis indicated that serum and SF myostatin concentrations were significantly associated with KL grades (P<.001 and P<.001).

4. Discussion

Our study revealed that the serum myostatin concentrations in OA patients were higher than those in the control group. Serum and SF myostatin concentrations were significantly correlated with KL grading. This study is the first to report the association of serum and SF myostatin concentrations with the presence and severity of OA.

Although myostatin is expressed in the synovial fibroblasts and synovial tissues of OA patients, the differences in myostatin expression or concentrations between OA patients and healthy controls have not yet to be examined. Our study showed that the serum myostatin concentrations in OA patients were higher than those in the controls. This finding supports the possibility of using myostatin as a biomarker to assess the risk of OA. Serum and SF myostatin concentrations correlated with the radiographic severity of OA. Myostatin is possibly involved in the mechanism of OA progression. However, the exact mechanism has been elucidated through further study of basic research.

Chondrocyte degeneration and cartilage destruction are important processes in OA development and progression. Myostatin is a member of the TGF‐β family and correlated closely with bone turnover.12 The key role of myostatin in OA etiology has been extensively investigated. Myostatin expression is detected in synovial fibroblasts and synovial tissues from patients with rheumatoid arthritis and OA.9 Osteoclastogenesis is involved in OA mechanism.13 Myostatin‐deficient mice exhibit a markedly decreased receptor activator for the nuclear factor‐κB ligand (RANKL)‐induced osteoclast formation and the reduced number of osteoclast.9 No clinical signs of arthritis are found in myostatin‐deficient mice.9 Myostatin‐specific antibodies treatment led to fewer clinical signs and reduced bone erosion compared to the controls.9 Therefore, myostatin may induce OA development by promoting osteoclast formation and joint destruction.

Inflammation is involved in the pathological process of OA.14 Mild to moderate severity of synovial inflammation could be found in approximately 33% of patients with OA after synovial biopsies are obtained.15 OA synovial fibroblasts incubated with inflammatory cytokines, including tumor necrosis factor‐α (TNF‐α), interleukin‐1 (IL‐1), and IL‐17 induce the upregulation of myostatin expression.9 TNF‐α stimulation enhanced the myostatin expression via the NF‐κB pathway. By contrast, myostatin stimulates muscle cells releasing IL‐6.16 TNF‐α can activate myostatin through the NF‐κB pathway in myotubes,17 but inhibition of the NF‐κB pathway can inactivate myostatin.17 Moreover, silencing myostatin with myostatin antibody decreases TNF‐α and IL‐6 levels.18 Thus, myostatin likely interacts with inflammatory cytokines and contributes to the presence and progression of OA.

Several potential limitations should be considered in this study. First, this study applies a cross‐sectional approach. The causative relation found in this study should be confirmed through future longitudinal studies. Second, we did not examine SF myostatin concentrations of healthy controls because of ethical concerns. Last, We did not registered this clinical study.

In summary, serum and SF myostatin concentrations were correlated with the presence and severity of knee OA. However, the potential mechanism of myostatin in OA pathogenesis should be explained through further investigations.

Source of Funding

No funding.