Patients with knee osteoarthritis (OA) are characterized by increased muscle inflammation and altered gait. We investigated the association between proinflammatory mediators in the vastus lateralis and physical function and gait in patients with knee OA.
Nineteen patients with knee OA underwent gait analysis, assessment of self-reported pain and physical function (Western Ontario and McMaster Universities Osteoarthritis Index [WOMAC]), and a muscle biopsy that was taken during their knee replacement surgery. Muscle was analyzed for cellular protein inflammatory mediators, interleukin-6, monocyte chemotactic protein 1 (MCP-1), p65 NF-κB, signal transducer and activator of transcription 3 (STAT-3), and JNK-1. Sagittal plane knee function, including early stance knee range of motion (ROM) and knee sagittal plane impulse, was measured using a motion analysis system. Pearson's correlation was used to assess relationships between selected variables.
Significant positive correlations were found between MCP-1 and self-perceived stiffness, physical function, and the total WOMAC score (P < 0.05). MCP-1 was also negatively correlated with early stance knee ROM (r = −0.52, P = 0.023). Reduced velocity was associated with elevated levels of p65 NF-κB and STAT-3 (P < 0.05). Knee sagittal plane impulse was negatively correlated with JNK-1 (P = 0.02), indicating reduction in knee impulse with an increased level of JNK-1.
Increased levels of several proinflammatory mediators were correlated with altered knee function during walking as well as greater physical disability and slower gait velocity. Identification of the cellular and molecular mechanisms associated with muscle inflammation is important to better understand the underlying mechanism responsible for altered gait and function in patients with knee OA.
Knee osteoarthritis (OA) is a major cause of pain and functional disability among older adults (1, 2). It is estimated that one-half of people ages >80 years have some form of arthritis (3). It appears that the disease process affects the entire joint structure, with increased inflammation within the synovial membrane and synovial tissues (4–6). Recently, we have reported an increase in inflammatory markers in the vastus lateralis in people with knee OA, suggesting that increased inflammatory responses may not necessarily be localized to the knee joint but may also be present in the skeletal muscle surrounding the knee (7).
It has been shown that the presence of proinflammatory markers such as interleukin-1β (IL-1β), tumor necrosis factor α, suppressor of cytokine signaling 3, and the p65 subunit of NF-κB is implicated in muscle atrophy and impaired muscle regeneration (8, 9). However, they may also lead to the generation and maintenance of pain by stimulating nociceptors (10–13). Studies have also reported that increased systemic IL-6 is associated with lower self-reported physical function (using the Western Ontario and McMaster Universities Osteoarthritis Index [WOMAC]) and slower walking speed (14) as well as imbalance of the knee joint flexor and extensor muscles (15). Increased proinflammatory markers in skeletal muscle (vastus lateralis) were also associated with lower quadriceps strength in people with severe knee OA (7).
The presence of inflammatory markers within the vastus lateralis muscle adversely influences skeletal muscle mass maintenance and, in combination with leg disuse due to pain and disease symptoms, may result in further deterioration of muscle mass and strength (16, 17). The vastus lateralis, being the largest component muscle of the quadriceps femoris, plays a major role in knee extension and is important for daily functioning such as walking. Consequently, the reduction in muscle strength and function, particularly of the quadriceps, can have a detrimental impact on physical function and disability in people with knee OA (18, 19). However, the effects of increased proinflammatory markers in the quadriceps muscle on the performance of daily life activities such as walking are unclear and require further investigation.
Altered gait pattern is often observed in people with knee OA (20, 21). Pain associated with knee OA has been thought to result in gait modification as an attempt to reduce the load on the affected joint, with studies reporting changes in joint loading as a response to pain relief (22–24), or pain induction in experimental trials (25, 26). Given that an increase in inflammatory markers can promote pain, it may also affect muscle function during gait, leading to different gait strategies to reduce joint load. However, to date no study has examined whether an increase in muscle inflammatory markers is related to gait characteristics, particularly knee function, during walking in patients with knee OA. Such information has the potential to inform clinicians about the importance of intramuscular inflammation in knee OA; thus, treatments can be tailored to reduce inflammation and improve the functional capacity of patients with knee OA. Therefore, the aim of this study was to examine the association between proinflammatory transcription factors, JNK-1, p65 NF-κB, and signal transducer and activator of transcription 3 (STAT-3) and inflammatory intracellular molecules IL-6 and monocyte chemotactic protein-1 (MCP-1) in the vastus lateralis and pain, physical function, and knee biomechanics in people with severe knee OA. We hypothesized that an increase in inflammatory mediators will be correlated with impaired gait in people with knee OA.
Significance & Innovations
An increased intramuscular abundance of several proinflammatory mediators was associated with impaired knee function during walking.
An increased intramuscular abundance of several proinflammatory mediators was associated with greater physical disability and slower gait velocity.
Gait velocity may be an important indicator for the inflammatory status of the muscle in patients with knee osteoarthritis.
MATERIALS AND METHODS
Nineteen patients (9 men and 10 women) with diagnosed knee OA were recruited to participate in the study. To be eligible, patients had to be scheduled for knee replacement surgery and be able to walk at least 45 meters independently (without the use of walking aids). Patients were excluded if they had uncontrolled systemic disease (nonmusculoskeletal conditions that would make testing difficult and uncomfortable for the participants, such as chronic obstructive airway disease and congestive heart failure) and a preexisting neurologic or other orthopedic condition affecting walking. Participants were also excluded if they had the following foot conditions: partial foot amputation or ulceration, foot fractures, any neurologic condition that affects lower leg strength (e.g., stroke, polio), any other musculoskeletal conditions that may affect the lower leg (e.g., rheumatoid arthritis, gout), and previous surgeries to the ankle, foot, or knees (excluding knee arthroscopy). Patients were screened over the phone to identify their eligibility to participate. Seven patients were excluded: 5 due to previous surgeries to foot and ankle, 1 due to inability to walk without a walking stick, and 1 due to diagnosed rheumatoid arthritis in all joints in the lower leg. Patients were receiving a range of medications, including diuretics (n = 3), antidiabetic agents (n = 2), anticholesterol agents (n = 3), beta-blockers (n = 3), angiotensin II receptor antagonists (n = 5), calcium-channel blockers (n = 2), glucosamine (n = 6), angiotensin-converting enzyme inhibitors (n = 6), and aspirin (n = 3). Patients with knee OA (n = 9) receiving antiinflammatory medications were taken off of the medications a week prior to the knee replacement surgery. Patients were recruited from the La Trobe University Medical Centre and the Warringal Private Medical Centre. The study protocol was approved by the Human Research Ethics Committees of Victoria University, La Trobe University, and Warringal Private Hospital. All of the participants were informed about the nature of the study and signed a consent form prior to participation.
Patients underwent gait analysis and completed a questionnaire to assess pain, physical function, and joint stiffness approximately 7–14 days prior to their knee replacement surgery. A muscle sample of the vastus lateralis was taken during the knee replacement surgery.
Knee pain, function, and stiffness.
Physical function, pain, and stiffness were assessed using the WOMAC (27). This index assesses the severity of knee pain during 5 daily activities (range 0–500), stiffness (range 0–200), and the severity of impairment of lower extremity function during 17 activities (range 0–1,700). The items were scored with the use of a 10-mm visual analog scale, where 0 represents no pain or no difficulty with physical function and higher scores represent worse functional health. All 3 subcategories are summed to give a global WOMAC score (range 0–2,400).
Participants attended the gait laboratory at La Trobe University. To assess the motion of the knee in the sagittal plane, retroreflective markers were attached on anatomic landmarks over the lower legs in accordance with the Oxford foot model marker set and Plug-In-Gait as described by Stebbins et al (28). The following measurements were taken for the calculation of the knee joint center: height, weight, leg length, and knee and ankle width. The 3-dimensional (3-D) trajectories of the markers during gait were acquired using a 3-D motion analysis system (Vicon, Oxford Metrics) with 10 (MX3 and MX40) cameras operating at a sampling rate of 100 Hz. Two force plates (Kistler, type 9865B, and AMTI, OR6; 1,000 Hz) were used to capture ground reaction forces and identify gait cycle events. Knee kinematics and knee kinetics in the sagittal plane during level walking were computed using the Plug-In-Gait biomechanical model (29).
The following variables of the affected knee (scheduled for knee replacement surgery) were assessed: early stance knee range of motion (ROM) and knee impulse. Early stance knee ROM was calculated as the knee excursion (total motion) from initial contact to peak flexion in the sagittal plane during the loading phase (Figure 1). Since one of the important biomechanical functions of the knee is to act as a shock absorber during the loading phase (0–12% of gait cycle) by controlling knee flexion, the early stance knee ROM measure provides information about any restriction in the knee flexion motion (30). Knee flexion/extension impulse was calculated as the cumulative magnitude of the time-integral external knee flexion and extension moments (normalized to % of body weight × height) throughout the stance phase (i.e., the total area under the curve, as indicated in Figure 2) (20). Several studies have used the knee impulse previously as a useful gait parameter (31, 32). Moreover, knee moments in the sagittal plane provide an indirect indication of muscle force production (30).
Participants were asked to walk at a comfortable walking pace along a 12-meter walkway. Five suitable trials (when the participant's foot landed on the center of the force plate without any interference to their gait) were collected for the affected leg. The parameters of interest of each trial were extracted and the average of the 5 trials was used in the analysis.
Resting muscle samples were taken from the vastus lateralis as previously described (7). In brief, the muscle sample from knee OA was collected during their knee replacement surgery approximately 5 cm proximal to the suprapatellar pouch. The biopsies were taken after the skin was incised and prior to knee joint capsule incision with no trauma to the muscle or the joint at that time (7).
Protein extraction and Western blot analysis.
Skeletal muscle (30 mg) was homogenized in cell lysis buffer (Bio-Rad Laboratories). Protein homogenates were separated by 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred onto a nitrocellulose membrane. Blots were blocked and primary antibodies were applied and incubated overnight at 4°C, including STAT-3, p65, JNK (Cell Signaling Technology), and actin (Sigma-Aldrich). Proteins were visualized by enhanced chemiluminescence (Western Lighting Chemiluminescence Reagent Plus, Perkin-Elmer) and densitometry was performed (Kodak Imaging software, Kodak ID 3.5, Perkin-Elmer). Membranes were stripped with restore Western blot stripping buffer (Quantum Scientific) for 60 minutes before being reprobed for actin.
Multiplex suspension array system.
A Bio-Plex suspension array for IL-6 and MCP-1 was conducted following the manufacturer's instructions (Bio-Rad Laboratories) and using reagents from the cytokine reagent kit (Bio-Rad Laboratories). Protein lysates prepared for the Western blot were added to a 96-well filtration plate containing premixed beads coated with target antibodies and incubated for 30 minutes. Plates were washed in wash buffer, and 25 μl of premixed detection antibodies were added and allowed to incubate for a further 30 minutes. Wells were washed and 50 μl of streptavidin–phycoerythrin was added and incubated for 10 minutes. Wells were washed a final time in wash buffer and resuspended in 125 μl of assay buffer. The plate was read on the Bio-Plex Suspension Array System (Bio-Rad Laboratories). All of the samples were run in triplicate and the coefficient of variation (CV) was calculated; the mean CVs were between 9% and 11%. Protein concentrations were calculated based on standard curve data with a weighted 5-parameter logistic fit analysis using Bio-Plex Manager Software, version 5.0 (Bio-Rad Laboratories). The muscle data are reported elsewhere (7).
The associations between several proinflammatory markers, including JNK-1, IL-6, MCP-1, p65 NF-κB, STAT-3, and knee kinematics (early stance knee ROM) and kinetics (knee impulse), during the stance phase of walking, gait velocity, pain, function, and stiffness were assessed using partial Pearson's correlation coefficient. Data were adjusted for body mass index and age.
The main characteristics of the sample population, including self-reported pain, stiffness, and function, as well as the gait parameters, are shown in Table 1. Significant positive correlations were found between MCP-1 protein abundance and each of self-perceived stiffness (P = 0.04), physical function (P = 0.02), and the total WOMAC score (P = 0.02) (Figure 3). Moreover, p65 NF-κB and STAT-3 were negatively correlated with gait velocity (Figure 4). No significant correlations were found between the other inflammatory markers and self-reported pain and function (r range 0.02–0.31, P > 0.05) or between gait velocity and MCP-1, JNK-1, or IL-6 (r range −0.29 to 0.23, P > 0.05). Knee sagittal plane impulse was negatively correlated with JNK-1 (r = −0.49, P = 0.02), and MCP-1 was negatively correlated with early stance knee ROM (Figure 5). No significant correlations were found between STAT-3 and p65 NF-κB and early stance knee ROM or knee impulse (r range −0.21 to 0.01, P > 0.05). Also, IL-6 did not correlate with functional or gait parameters in the current cohort (r range −0.29 to 0.02, P > 0.05).
WOMAC = Western Ontario and McMaster Universities Osteoarthritis Index; ROM = range of motion; %BW × Ht = % of body weight × height.
69.89 ± 6.49
1.66 ± 0.82
82.40 ± 11.77
Body mass index, kg/m2
29.85 ± 3.83
Self-reported pain, physical function, and stiffness
191.26 ± 96.33
91.26 ± 41.09
610.95 ± 288.53
893.47 ± 357.52
Gait velocity, meters/second
1.13 ± 0.18
Knee early stance ROM, degrees
7.02 ± 3.74
Knee impulse, %BW × Ht
0.71 ± 0.21
The knee OA disease process affects the entire joint structure and is characterized by increased inflammation in the joint (4–6) as well as within the musculature (7). The present study demonstrated a novel relationship between elevated levels of proinflammatory markers in the muscle and self-reported greater physical disability and joint stiffness, in addition to impaired knee function and reduction in gait velocity during walking. These findings point for the first time to the potential detrimental role of proinflammatory mediators, present within the muscle, to impaired muscle function and gait in knee OA.
Poor muscle strength and dysfunction in the affected legs is a common feature and likely contributes to impaired quality of life through deterioration of functional capacity in people with knee OA (18, 19, 33). We previously demonstrated a correlation with increased protein abundance of inflammatory chemokine MCP-1 and impaired quadriceps muscle strength (7). In the present study we show that MCP-1 was also correlated with greater self-reported disability, knee stiffness, and the overall WOMAC score, indicating that the presence of proinflammatory molecules within the muscle may have a negative effect on physical function.
To our knowledge, this is the first study to explore a relationship between MCP-1 and muscle function, with findings pointing to an as yet undescribed role between this inflammatory mediator and physical disability. Due to the crosstalk of MCP-1 with a number of inflammatory molecules, this chemokine has divergent roles in skeletal muscle. MCP-1 is secreted by human muscle cells and macrophages in response to inflammatory stimuli and is elevated in inflammatory myopathies (34, 35). In healthy skeletal muscle, MCP-1 stimulates myoblast proliferation following muscular injury promoting myofiber repair and reducing lipid infiltration (36), suggesting a positive role of MCP-1 in regulating effective muscle regeneration. However, as the primary role of circulating MCP-1 is in the recruitment of monocytes, neutrophils, and other inflammatory cells (37), a role in muscle atrophy during states of chronic inflammatory stress may occur (37), leading to impaired regeneration and associated loss of muscle mass and strength.
Pain is a common symptom in OA and a major cause of disability and functional decline. Increases in the presence of proinflammatory markers in the joint may lead to the generation and maintenance of pain by acting on nociceptive nerve cells (10–13). A previous study reported that high concentrations of inflammatory markers in the serum were associated with poor physical function and increased pain and stiffness (14). Synovial inflammation was also associated with worse knee pain and function scores in people with a meniscal tear (12, 38). However, in the present study, no correlations were found between self-reported pain and proinflammatory cytokines in the vastus lateralis. The underlying mechanism for pain in knee OA has been attributed to both peripheral and central sensitization (39, 40). It is unclear if the pain reported by the current knee OA patients is related to joint or muscle pain, particularly as referred, and radiating pain is often common in this group (39). This may possibly explain the lack of correlation, as it is unclear if the pain reported in the current sample may have been originated from the muscle. Therefore, to better understand the mechanism of pain and inflammation associated with OA, simultaneous investigation of the inflammatory response in the muscles and in the joint is needed.
Gait alterations are often seen in people with knee OA during walking, which have been suggested to be an attempt to reduce pain and joint load (21, 41). People with knee OA typically walk more slowly and with a reduced knee ROM during early stance, as well as a reduced total knee ROM, compared to asymptomatic control subjects (20, 21). Changes in knee joint moments in the sagittal plane have also been reported in people with knee OA compared to asymptomatic controls (42). An interesting finding in the present study was the significant negative correlation between MCP-1 and early stance knee ROM. This suggests that an increased level of MCP-1 abundance was associated with reduced knee ROM during early stance (loading phase). Moreover, an increase in the level of JNK-1 was correlated with reduced knee sagittal plane impulse. During the loading phase (approximately 10–12% of the gait cycle) (Figures 1 and 2), the knee is flexed and the quadriceps muscles are required to work eccentrically to enable weight acceptance as part of the knee biomechanical function of shock attenuation (30). Reduction in knee excursion motion during the loading phase, together with reduced knee impulse, may be an attempt to minimize knee joint loading through restriction of knee motion during stance. Given the important role of the quadriceps muscle to attenuate shock during the loading phase of the gait cycle, the presence of OA in the joint and subsequent joint inflammation may lead to worsened function partly due to the atrophic effects of this inflammation on the surrounding musculature. However, further research is needed to ascertain the causal relationship between the level of proinflammatory markers in the muscle and joints and gait in people with knee OA.
Gait velocity is an important indicator for physical function and disability in older people (43), and a reduction in gait velocity is common in people with knee OA (20, 21). It is essential to understand the factors underlying reduced gait velocity in OA. We show for the first time that elevated p65 NF-κB and STAT-3 abundance in muscle were correlated with reduced gait velocity. This is consistent with the finding of Penninx et al, who reported that an increase in serum concentration of IL-6 in the blood was also associated with a slower walking speed (14). However, this study greatly extends this by showing a direct relationship between p65 NF-κB and STAT-3 in the quadriceps muscle and gait velocity during walking. Both p65 NF-κB and STAT-3 are stress-responsive kinases regulated by a wide array of signaling cascades, particularly involving IL-6, and are activated by an array of possible upstream ligands and perturbations to cellular homeostasis (44, 45). They are, however, critical in mediating the transcription of many genes necessary for cell survival, adaptation, and further cellular inflammatory mediation. We have previously reported a significant increase in the level of p65 NF-κB for the current knee OA group compared to age-matched controls (7). Given these results, gait velocity may not only be an important indicator for health status of older people, but also an important indicator for the inflammatory status of the muscle in patients with knee OA. This finding has important implications, since management of intramuscular inflammation may therefore be influential in improving gait and mobility.
This study was a cross-sectional investigation and, as such, casual relationships cannot be determined. Despite this, the novel findings presented in this study raise important questions regarding the role of muscle inflammation in the development and progression of knee OA, as well as the functional implications of elevated inflammatory markers. The data suggest that the inflammatory response in the muscle is associated with greater physical disability and an altered gait pattern. Further studies are needed to investigate potential nonpharmacologic (such as exercise intervention) and pharmacologic therapeutic strategies to reduce skeletal muscle inflammation and improve functional capacity in people with knee OA. Since impaired muscular function is a major determinant of functional performance such as walking (18, 46), understanding the underlying mechanisms of the loss of muscle mass and function in knee OA and the effects of muscle inflammation on muscle function is important and warrants further investigation.
An increased intramuscular abundance of several proinflammatory mediators was associated with impaired knee function during walking, as well as with greater physical disability and slower gait velocity. Identification of the cellular and molecular mechanisms associated with muscle inflammation is important to better understand the underlying mechanisms responsible for the impairments in gait and function in patients with knee OA.
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Pazit Levinger had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. Pazit Levinger, Feller, McKenna, Cameron-Smith, Itamar Levinger.
Acquisition of data. Pazit Levinger, Caldow, Bartlett, Bergman, Itamar Levinger.
Analysis and interpretation of data. Pazit Levinger, Caldow, Bartlett, McKenna, Cameron-Smith, Itamar Levinger.