Bio-Medical Materials and Engineering 24 (2014) S17–S25 DOI 10.3233/BME-140970 IOS Press
S17
Synovial fluid levels of adipokines in osteoarthritis: Association with local factors of inflammation and cartilage maintenance Jean-Baptiste Gross a , Cécile Guillaume b , Pascale Gégout-Pottie b , Didier Mainard a,b and Nathalie Presle b,∗ a
Orthopaedics and Trauma Surgery Department, CHU Nancy, Hôpital Central, Nancy, France UMR 7365 CNRS-UL, Ingénierie Moléculaire et Physiopathologie Articulaire, Campus Biologie-Santé, Faculté de Médecine, Biopôle de l’Université de Lorraine, Vandœuvre-lès-Nancy, France
b
Abstract. The role of body weight in the pathogenesis of osteoarthritis (OA) – previously considered the sole factor in the association between obesity and OA – is being re-evaluated as the contribution of adiposity to the cartilage degenerative process becomes clearer. The current study has been undertaken to better understand the role of adipose-derived proteins, namely adipokines, in OA. For this purpose, we investigated in patients with OA the relationships between the joint levels of leptin, adiponectin and resistin and those of factors involved in inflammation and cartilage maintenance. The sandwich enzyme-linked immunosorbent assays were used to determine in the synovial fluid (SF) from 35 OA patients, the concentrations of adipokines, interleukin-6 (IL-6) and transforming growth factor-β (TGF-β). The soluble form of leptin receptor (sOb-R) was also examined to evaluate the biological active free form of leptin. Correlation analysis indicate that IL-6 levels are positively related to the levels of resistin and adiponectin. Surprisingly, the free form of leptin, but not the total leptin, is negatively associated with IL-6. Beside, adiponectin is the single adipokine that is correlated with TGF-β. Interestingly, a sexual dimorphism is observed in the study as correlations between adipokines and IL-6 or TGF-β are found only with female OA patients. Taken together, these findings suggest that only adiponectin may contribute to the metabolic changes associated with OA. The three adipokines may also be involved in inflammation, but with opposite effects. Both resistin and adiponectin may exhibit pro-inflammatory activity while the free form of leptin may down-regulate the inflammation. Keywords: Adipokines, obesity, osteoarthritis
1. Introduction Osteoarthritis (OA) is a painful and disabling joint disease that affects millions of patients. Because of the ageing of the populations, the prevalence of OA is expected to increase. Its pathology involves the whole joint in a degenerative process that includes focal and progressive cartilage loss, from fibrillations in the early stages of the disease to full-thickness matrix depletion. The chondrocyte, the only cell type residing in the adult cartilage matrix, is responsible for remodelling and maintaining the structural and functional integrity of the extracellular matrix. During OA, chondrocyte exhibits an aberrant behavior *
Address for correspondence: Nathalie Presle, UMR 7365 CNRS-UL, Ingénierie Moléculaire et Physiopathologie Articulaire (IMoPA), Campus Biologie-Santé, Faculté de Médecine, Biopôle de l’Université de Lorraine, Avenue de la forêt de Haye, CS 50184, 54505 Vandœuvre-lès-Nancy cedex, France. Tel.: +33 383 68 54 00; Fax: +33 383 68 54 09; E-mail: [email protected]. 0959-2989/14/$27.50 © 2014 – IOS Press and the authors. All rights reserved
S18
J.-B. Gross et al. / Synovial fluid levels of adipokines in osteoarthritis
with increased matrix degradation and reduced anabolic processes. These pathological events result in a disruption of cartilage homeostasis and involve inflammatory cytokines, metalloproteases and growth factors. Although most key contributors to cartilage destruction are now identified, no proven diseasemodifying therapy has shown any beneficial effect in OA. Insight into the different underlying mechanisms leading to the clinical and pathologic outcomes of OA is crucial in the search of novel drugs. A new hypothesis based on the link between OA and obesity has emerged for the last decade [1]. Indeed, obesity is a well-recognized risk factor for the development of OA. However, the sole role of biomechanical loading cannot explain the increased risk for OA in non-weight bearing joints among overweight individuals, and recent studies indicate that adiposity rather than simply excess in body mass is detrimental for the joint [2–6]. During the last few years, many progress have been made to understand the worldwide increasing prevalence of obesity. The adipose tissue is now considered as an active endocrine organ which plays a critical role through the release of various adipokines in an integrated network that maintains interactions between fat and other organs [7]. These adipose-derived proteins are required for normal physiological homeostasis, but impaired production may be involved in obesity-related disorders, and more especially in the mild chronic inflammation found in obese individuals. As OA develops in a highly metabolic and inflammatory environment of adiposity, adipokines are expected to contribute to the pathological changes observed in OA. We and others have shown actually that among adipokines, leptin, adiponectin, and resistin are detected in the synovial fluid (SF) obtained from OA patients [8–11]. In humans, circulating levels of both leptin and resistin are positively correlated with the BMI and fat mass [12,13] while adiponectin plasma concentration is decreased in obese individuals [14,15]. The role of resistin in OA has not been established, but Schäffler et al. have shown that the SF levels of resistin are positively correlated with systemic markers of inflammation [8]. Beside, the contribution of leptin or adiponectin in OA-affected joints remains unclear despite numerous studies demonstrating their effects in chondrocytes. Adiponectin, leptin and their receptors are strongly up-regulated in human OA cartilage [16–18] but they exhibit dualistic activities. On one hand, they can stimulate chondrocyte proliferation and differentiation, and may have beneficial effect on cartilage through the stimulation of matrix synthesis and the induction of growth factors or Tissue Inhibitors of Metalloproteinases (TIMPs) [16,19–21]. On the other hand, both adipokines may trigger cartilage destruction through the up-regulation of degradative metalloproteinases (MMPs) and pro-inflammatory mediators [17,18,22–24]. The current study was therefore undertaken to further understand the contribution of adipokines to articular changes associated with OA. For this purpose, we investigated the relationship between resistin, leptin and adiponectin and factors involved in inflammation, namely interleukin-6 (IL-6) or in cartilage maintenance, namely the transforming growth factor-β1 (TGF-β1). The concentrations of leptin, resistin, adiponectin, IL-6 and TGF-β1 were thus measured in SF obtained from OA patients. As elevated levels of free leptin, the presumed biologically active form of the adipokine, were found within OA joints [9], we also determined the concentration of the soluble form of leptin receptor (sOb-R). 2. Patients and methods 2.1. Patients and samples Thirty-five patients (20 women and 15 men, ages 50–83 years [mean age 68.4 years]) who required total knee arthroplasty were included in the present study. Knee OA was diagnosed from clinical and
J.-B. Gross et al. / Synovial fluid levels of adipokines in osteoarthritis
S19
radiologic evaluation based on the American College of Rheumatology criteria [25]. SF samples were taken during the knee replacement surgeries, centrifuged at 4000 g for 15 minutes and stored at −80◦ C prior to use in subsequent assays. This human study was conducted in conformity with the declaration of Helsinki principles, and was approved by the local Research Institution (Commission de la Recherche Clinique; registration number UF9607). Written informed consents have been obtained from all participants. 2.2. Assays The concentrations of adipokines (leptin, resistin and adiponectin), sOb-R, IL-6 and TGF-β in SF samples were determined in duplicate by a sandwich enzyme-linked immunosorbent assay (ELISA) using commercially available kits. The leptin, resistin, adiponectin, IL-6 and TGF-β ELISA kits were purchased from R&DSystems (Lille, France) while that for sOb-R was obtained from BioVendor (Heidelberg, Germany). Dilution and spiking experiments were performed to validate the use of commercially ELISA kits for SF samples. Samples were diluted as appropriate and paired samples were assayed in the same run. According to the manufacturers, the detection limits for leptin, resistin, adiponectin, sOb-R, IL-6 and TGF-β assays were 7.8 pg/ml, 26 pg/ml, 250 pg/ml, 800 pg/ml, 0.039 pg/ml and 7 pg/ml respectively. The interassay coefficients of variation for the six assays were 4.4%, 8.4%, 7.8%, 3.6%, 7.8% and 11% respectively. Intraassay coefficients of variation were less than 5%, except for IL-6 assays for which it was 7.4%. 2.3. Statistical analysis The concentrations are expressed as ng/ml for leptin, resistin and sOb-R, as µg/ml for adiponectin and as pg/ml for IL-6 and TGF-β. The results are shown as mean values (SEM). The leptin/sOb-R ratio was used as an index of free leptin, and was determined after converting concentrations to µmol/ml. Statistical analysis was conducted with StatView for Windows, version 5.0 (SAS, Cary, NC). Differences between women and men were analyzed using the nonparametric Mann–Whitney U test and statistical correlations were determined by the Spearman test of rank correlation. A P value less than 0.05 was considered significant for differences and correlations. 3. Results 3.1. Adipokines and sOb-R determination in SF samples obtained from OA patients SF samples obtained from 35 OA patients were analysed for adipokines, i.e. leptin, adiponectin and resistin, and for the soluble form of leptin receptor sOb-R. There was no significant difference between the female and male patients with respect to age or BMI (Table 1). The adipokines were detected in each SF tested with concentrations ranging from 0.80 to 72.31 ng/ml for leptin (mean [SEM], 15.94 [2.42] ng/ml), from 0.62 to 7.01 µg/ml for adiponectin (mean [SEM], 2.32 [0.29] µg/ml), and from 1.62 to 26.41 ng/ml for resistin (mean [SEM], 7.26 [1.21] ng/ml). When gender-specific differences in these levels were examined, only SF level of leptin in the female group was found to be significantly higher than that in the male group (mean [SEM], 20.77 [3.61] ng/ml versus 9.49 [2.14] ng/ml; P < 0.05). No significant gender difference with regard to resistin or adiponectin levels was observed (Table 1).
S20
J.-B. Gross et al. / Synovial fluid levels of adipokines in osteoarthritis
Table 1 Concentrations of adipokines (leptin, adiponectin and resistin), soluble leptin receptor (sOb-R), interleukin-6 (IL-6) and transforming growth-β (TGF-β) in synovial fluid obtained from patients with osteoarthritis n Age (yrs) BMI (kg/m2 ) Leptin (ng/ml) Adiponectin (µg/ml) Resistin (ng/ml) sOb-R (ng/ml) TGF-β (pg/ml) IL-6 (pg/ml)
Female 20 71.05 (7.41) 28.20 (4.28) 20.77 (3.61) 2.34 (0.37) 8.47 (1.75) 7.33 (1.15) 799.30 (98.03) 486.52 (146.82)
Male 15 65.80 (7.74) 30.13 (5.09) 9.49 (2.14)∗ 2.31 (0.47) 5.73 (1.58) 4.85 (1.00) 678.00 (44.16) 74.17 (46.12)∗
Notes: Age and Body Mass Index (BMI) are shown as mean values (SD). Concentrations values are expressed as mean values (SEM). Comparisons between genders were performed using the Mann–Whitney U test and P < 0.05 was considered significant (∗ ).
Beside leptin, sOb-R could be determined in each SF sample. Mean [SEM] level of the soluble form of leptin receptor was higher in women compared with men (7.33 [1.15] ng/ml versus 4.85 [1.00] ng/ml), but the difference did not reach statistical significance (Table 1). 3.2. Relationship between TGF-β and adipokines in the SF obtained from OA patients A significant amount of TGF-β was present in each SF specimen examined with concentrations ranging from 400.29 to 1684.71 pg/ml. Although female patients with OA exhibited a higher SF TGF-β level than did male OA patients, the mean [SEM] values did not differ significantly (799.30 [98.03] pg/ml for women, 678 [44.16] pg/ml for men) (Table 1). A significant positive correlation between the levels of TGF-β and adiponectin was apparent in SF collected from female OA patients (r = 0.682, P = 0.0139) (Table 2), but not in samples from male OA patients. No relationship was found between SF levels of TGF-β and the other adipokines tested. 3.3. Relationship between IL-6 and adipokines in the SF obtained from OA patients IL-6 concentrations could be determined in each SF sample. However, a large variation was observed between OA patients (range from 3.77 pg/ml to 2498 pg/ml). In addition, a strong gender-specific difference was found since IL-6 level was shown to be almost 7-fold higher in SF obtained from female patients than in samples from male patients (mean [SEM], 486.52 [146.82] pg/ml versus 74.17 [46.12] pg/ml). When the relationships between IL-6 levels and the adipokines levels were examined, no significant correlation was found for the male group. By contrast, SF levels of IL-6 from the female group showed significant positive correlations with the levels of resistin and adiponectin (r = 0.461, P = 0.0052; r = 0.689, P = 0.0027, respectively). Interestingly, we failed to demonstrate any significant correlation between leptin and IL-6 levels in SF from female OA patients (P = 0.2104), but Spearman’s rank correlation analysis indicated that IL-6 levels were negatively related to the levels of free leptin (r = −0.797, P = 0.0082) (Table 2).
J.-B. Gross et al. / Synovial fluid levels of adipokines in osteoarthritis
S21
Table 2 Correlations between synovial fluid levels of interleukin-6 (IL-6) or transforming growth factor-β (TGF-β) and the adipokines (resistin, adiponectin and free leptin) in patients with osteoarthritis (20 female and 15 male) Resistin TGF-β IL-6
Adiponectin
Free leptin
Female r = 0.258 (NS)
Male r = 0.271 (NS)
Female r = 0.682 (P = 0.0139)
Male r = −0.227 (NS)
Female r = −0.042 (NS)
Male r = 0.251 (NS)
r = 0.641 (P = 0.0052)
r = 0.460 (NS)
r = 0.689 (P = 0.0027)
r = 0.507 (NS)
r = −0.797 (P = 0.0082)
r = 0.067 (NS)
Notes: Because of a lack of any correlation between synovial fluid levels of IL-6 or TGF-β and leptin, free leptin derived from the ratio leptin/sOb-R was used. Correlations were calculated using the Spearman’s rank correlation test (r, Spearman’s correlation coefficient) and a P value
S17
Synovial fluid levels of adipokines in osteoarthritis: Association with local factors of inflammation and cartilage maintenance Jean-Baptiste Gross a , Cécile Guillaume b , Pascale Gégout-Pottie b , Didier Mainard a,b and Nathalie Presle b,∗ a
Orthopaedics and Trauma Surgery Department, CHU Nancy, Hôpital Central, Nancy, France UMR 7365 CNRS-UL, Ingénierie Moléculaire et Physiopathologie Articulaire, Campus Biologie-Santé, Faculté de Médecine, Biopôle de l’Université de Lorraine, Vandœuvre-lès-Nancy, France
b
Abstract. The role of body weight in the pathogenesis of osteoarthritis (OA) – previously considered the sole factor in the association between obesity and OA – is being re-evaluated as the contribution of adiposity to the cartilage degenerative process becomes clearer. The current study has been undertaken to better understand the role of adipose-derived proteins, namely adipokines, in OA. For this purpose, we investigated in patients with OA the relationships between the joint levels of leptin, adiponectin and resistin and those of factors involved in inflammation and cartilage maintenance. The sandwich enzyme-linked immunosorbent assays were used to determine in the synovial fluid (SF) from 35 OA patients, the concentrations of adipokines, interleukin-6 (IL-6) and transforming growth factor-β (TGF-β). The soluble form of leptin receptor (sOb-R) was also examined to evaluate the biological active free form of leptin. Correlation analysis indicate that IL-6 levels are positively related to the levels of resistin and adiponectin. Surprisingly, the free form of leptin, but not the total leptin, is negatively associated with IL-6. Beside, adiponectin is the single adipokine that is correlated with TGF-β. Interestingly, a sexual dimorphism is observed in the study as correlations between adipokines and IL-6 or TGF-β are found only with female OA patients. Taken together, these findings suggest that only adiponectin may contribute to the metabolic changes associated with OA. The three adipokines may also be involved in inflammation, but with opposite effects. Both resistin and adiponectin may exhibit pro-inflammatory activity while the free form of leptin may down-regulate the inflammation. Keywords: Adipokines, obesity, osteoarthritis
1. Introduction Osteoarthritis (OA) is a painful and disabling joint disease that affects millions of patients. Because of the ageing of the populations, the prevalence of OA is expected to increase. Its pathology involves the whole joint in a degenerative process that includes focal and progressive cartilage loss, from fibrillations in the early stages of the disease to full-thickness matrix depletion. The chondrocyte, the only cell type residing in the adult cartilage matrix, is responsible for remodelling and maintaining the structural and functional integrity of the extracellular matrix. During OA, chondrocyte exhibits an aberrant behavior *
Address for correspondence: Nathalie Presle, UMR 7365 CNRS-UL, Ingénierie Moléculaire et Physiopathologie Articulaire (IMoPA), Campus Biologie-Santé, Faculté de Médecine, Biopôle de l’Université de Lorraine, Avenue de la forêt de Haye, CS 50184, 54505 Vandœuvre-lès-Nancy cedex, France. Tel.: +33 383 68 54 00; Fax: +33 383 68 54 09; E-mail: [email protected]. 0959-2989/14/$27.50 © 2014 – IOS Press and the authors. All rights reserved
S18
J.-B. Gross et al. / Synovial fluid levels of adipokines in osteoarthritis
with increased matrix degradation and reduced anabolic processes. These pathological events result in a disruption of cartilage homeostasis and involve inflammatory cytokines, metalloproteases and growth factors. Although most key contributors to cartilage destruction are now identified, no proven diseasemodifying therapy has shown any beneficial effect in OA. Insight into the different underlying mechanisms leading to the clinical and pathologic outcomes of OA is crucial in the search of novel drugs. A new hypothesis based on the link between OA and obesity has emerged for the last decade [1]. Indeed, obesity is a well-recognized risk factor for the development of OA. However, the sole role of biomechanical loading cannot explain the increased risk for OA in non-weight bearing joints among overweight individuals, and recent studies indicate that adiposity rather than simply excess in body mass is detrimental for the joint [2–6]. During the last few years, many progress have been made to understand the worldwide increasing prevalence of obesity. The adipose tissue is now considered as an active endocrine organ which plays a critical role through the release of various adipokines in an integrated network that maintains interactions between fat and other organs [7]. These adipose-derived proteins are required for normal physiological homeostasis, but impaired production may be involved in obesity-related disorders, and more especially in the mild chronic inflammation found in obese individuals. As OA develops in a highly metabolic and inflammatory environment of adiposity, adipokines are expected to contribute to the pathological changes observed in OA. We and others have shown actually that among adipokines, leptin, adiponectin, and resistin are detected in the synovial fluid (SF) obtained from OA patients [8–11]. In humans, circulating levels of both leptin and resistin are positively correlated with the BMI and fat mass [12,13] while adiponectin plasma concentration is decreased in obese individuals [14,15]. The role of resistin in OA has not been established, but Schäffler et al. have shown that the SF levels of resistin are positively correlated with systemic markers of inflammation [8]. Beside, the contribution of leptin or adiponectin in OA-affected joints remains unclear despite numerous studies demonstrating their effects in chondrocytes. Adiponectin, leptin and their receptors are strongly up-regulated in human OA cartilage [16–18] but they exhibit dualistic activities. On one hand, they can stimulate chondrocyte proliferation and differentiation, and may have beneficial effect on cartilage through the stimulation of matrix synthesis and the induction of growth factors or Tissue Inhibitors of Metalloproteinases (TIMPs) [16,19–21]. On the other hand, both adipokines may trigger cartilage destruction through the up-regulation of degradative metalloproteinases (MMPs) and pro-inflammatory mediators [17,18,22–24]. The current study was therefore undertaken to further understand the contribution of adipokines to articular changes associated with OA. For this purpose, we investigated the relationship between resistin, leptin and adiponectin and factors involved in inflammation, namely interleukin-6 (IL-6) or in cartilage maintenance, namely the transforming growth factor-β1 (TGF-β1). The concentrations of leptin, resistin, adiponectin, IL-6 and TGF-β1 were thus measured in SF obtained from OA patients. As elevated levels of free leptin, the presumed biologically active form of the adipokine, were found within OA joints [9], we also determined the concentration of the soluble form of leptin receptor (sOb-R). 2. Patients and methods 2.1. Patients and samples Thirty-five patients (20 women and 15 men, ages 50–83 years [mean age 68.4 years]) who required total knee arthroplasty were included in the present study. Knee OA was diagnosed from clinical and
J.-B. Gross et al. / Synovial fluid levels of adipokines in osteoarthritis
S19
radiologic evaluation based on the American College of Rheumatology criteria [25]. SF samples were taken during the knee replacement surgeries, centrifuged at 4000 g for 15 minutes and stored at −80◦ C prior to use in subsequent assays. This human study was conducted in conformity with the declaration of Helsinki principles, and was approved by the local Research Institution (Commission de la Recherche Clinique; registration number UF9607). Written informed consents have been obtained from all participants. 2.2. Assays The concentrations of adipokines (leptin, resistin and adiponectin), sOb-R, IL-6 and TGF-β in SF samples were determined in duplicate by a sandwich enzyme-linked immunosorbent assay (ELISA) using commercially available kits. The leptin, resistin, adiponectin, IL-6 and TGF-β ELISA kits were purchased from R&DSystems (Lille, France) while that for sOb-R was obtained from BioVendor (Heidelberg, Germany). Dilution and spiking experiments were performed to validate the use of commercially ELISA kits for SF samples. Samples were diluted as appropriate and paired samples were assayed in the same run. According to the manufacturers, the detection limits for leptin, resistin, adiponectin, sOb-R, IL-6 and TGF-β assays were 7.8 pg/ml, 26 pg/ml, 250 pg/ml, 800 pg/ml, 0.039 pg/ml and 7 pg/ml respectively. The interassay coefficients of variation for the six assays were 4.4%, 8.4%, 7.8%, 3.6%, 7.8% and 11% respectively. Intraassay coefficients of variation were less than 5%, except for IL-6 assays for which it was 7.4%. 2.3. Statistical analysis The concentrations are expressed as ng/ml for leptin, resistin and sOb-R, as µg/ml for adiponectin and as pg/ml for IL-6 and TGF-β. The results are shown as mean values (SEM). The leptin/sOb-R ratio was used as an index of free leptin, and was determined after converting concentrations to µmol/ml. Statistical analysis was conducted with StatView for Windows, version 5.0 (SAS, Cary, NC). Differences between women and men were analyzed using the nonparametric Mann–Whitney U test and statistical correlations were determined by the Spearman test of rank correlation. A P value less than 0.05 was considered significant for differences and correlations. 3. Results 3.1. Adipokines and sOb-R determination in SF samples obtained from OA patients SF samples obtained from 35 OA patients were analysed for adipokines, i.e. leptin, adiponectin and resistin, and for the soluble form of leptin receptor sOb-R. There was no significant difference between the female and male patients with respect to age or BMI (Table 1). The adipokines were detected in each SF tested with concentrations ranging from 0.80 to 72.31 ng/ml for leptin (mean [SEM], 15.94 [2.42] ng/ml), from 0.62 to 7.01 µg/ml for adiponectin (mean [SEM], 2.32 [0.29] µg/ml), and from 1.62 to 26.41 ng/ml for resistin (mean [SEM], 7.26 [1.21] ng/ml). When gender-specific differences in these levels were examined, only SF level of leptin in the female group was found to be significantly higher than that in the male group (mean [SEM], 20.77 [3.61] ng/ml versus 9.49 [2.14] ng/ml; P < 0.05). No significant gender difference with regard to resistin or adiponectin levels was observed (Table 1).
S20
J.-B. Gross et al. / Synovial fluid levels of adipokines in osteoarthritis
Table 1 Concentrations of adipokines (leptin, adiponectin and resistin), soluble leptin receptor (sOb-R), interleukin-6 (IL-6) and transforming growth-β (TGF-β) in synovial fluid obtained from patients with osteoarthritis n Age (yrs) BMI (kg/m2 ) Leptin (ng/ml) Adiponectin (µg/ml) Resistin (ng/ml) sOb-R (ng/ml) TGF-β (pg/ml) IL-6 (pg/ml)
Female 20 71.05 (7.41) 28.20 (4.28) 20.77 (3.61) 2.34 (0.37) 8.47 (1.75) 7.33 (1.15) 799.30 (98.03) 486.52 (146.82)
Male 15 65.80 (7.74) 30.13 (5.09) 9.49 (2.14)∗ 2.31 (0.47) 5.73 (1.58) 4.85 (1.00) 678.00 (44.16) 74.17 (46.12)∗
Notes: Age and Body Mass Index (BMI) are shown as mean values (SD). Concentrations values are expressed as mean values (SEM). Comparisons between genders were performed using the Mann–Whitney U test and P < 0.05 was considered significant (∗ ).
Beside leptin, sOb-R could be determined in each SF sample. Mean [SEM] level of the soluble form of leptin receptor was higher in women compared with men (7.33 [1.15] ng/ml versus 4.85 [1.00] ng/ml), but the difference did not reach statistical significance (Table 1). 3.2. Relationship between TGF-β and adipokines in the SF obtained from OA patients A significant amount of TGF-β was present in each SF specimen examined with concentrations ranging from 400.29 to 1684.71 pg/ml. Although female patients with OA exhibited a higher SF TGF-β level than did male OA patients, the mean [SEM] values did not differ significantly (799.30 [98.03] pg/ml for women, 678 [44.16] pg/ml for men) (Table 1). A significant positive correlation between the levels of TGF-β and adiponectin was apparent in SF collected from female OA patients (r = 0.682, P = 0.0139) (Table 2), but not in samples from male OA patients. No relationship was found between SF levels of TGF-β and the other adipokines tested. 3.3. Relationship between IL-6 and adipokines in the SF obtained from OA patients IL-6 concentrations could be determined in each SF sample. However, a large variation was observed between OA patients (range from 3.77 pg/ml to 2498 pg/ml). In addition, a strong gender-specific difference was found since IL-6 level was shown to be almost 7-fold higher in SF obtained from female patients than in samples from male patients (mean [SEM], 486.52 [146.82] pg/ml versus 74.17 [46.12] pg/ml). When the relationships between IL-6 levels and the adipokines levels were examined, no significant correlation was found for the male group. By contrast, SF levels of IL-6 from the female group showed significant positive correlations with the levels of resistin and adiponectin (r = 0.461, P = 0.0052; r = 0.689, P = 0.0027, respectively). Interestingly, we failed to demonstrate any significant correlation between leptin and IL-6 levels in SF from female OA patients (P = 0.2104), but Spearman’s rank correlation analysis indicated that IL-6 levels were negatively related to the levels of free leptin (r = −0.797, P = 0.0082) (Table 2).
J.-B. Gross et al. / Synovial fluid levels of adipokines in osteoarthritis
S21
Table 2 Correlations between synovial fluid levels of interleukin-6 (IL-6) or transforming growth factor-β (TGF-β) and the adipokines (resistin, adiponectin and free leptin) in patients with osteoarthritis (20 female and 15 male) Resistin TGF-β IL-6
Adiponectin
Free leptin
Female r = 0.258 (NS)
Male r = 0.271 (NS)
Female r = 0.682 (P = 0.0139)
Male r = −0.227 (NS)
Female r = −0.042 (NS)
Male r = 0.251 (NS)
r = 0.641 (P = 0.0052)
r = 0.460 (NS)
r = 0.689 (P = 0.0027)
r = 0.507 (NS)
r = −0.797 (P = 0.0082)
r = 0.067 (NS)
Notes: Because of a lack of any correlation between synovial fluid levels of IL-6 or TGF-β and leptin, free leptin derived from the ratio leptin/sOb-R was used. Correlations were calculated using the Spearman’s rank correlation test (r, Spearman’s correlation coefficient) and a P value
