If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Postgraduate Program in Physical Education and Sport, School of Physical Education and Sport of Ribeirão Preto, University of São Paulo, Ribeirão Preto, Brazil
Address correspondence and reprint requests to: A.S.R.da Silva, School of Physical Education and Sport of Ribeirão Preto, University of São Paulo Av Bandeirantes 3900, Monte Alegre, Ribeirão Preto, SP, Brazil 14040-907 Tel.: 55-16-3315-0522.
Postgraduate Program in Physical Education and Sport, School of Physical Education and Sport of Ribeirão Preto, University of São Paulo, Ribeirão Preto, BrazilPostgraduate Program in Rehabilitation and Functional Performance, Ribeirão Preto Medical School, University of São Paulo, Ribeirão Preto, Brazil
Address correspondence and reprint requests to: Walter Herzog, Faculty of Kinesiology, University of Calgary, Address: Faculty of Kinesiology, University of Calgary, 2500 University Drive NW, Calgary, Alberta, T2N 1N4, Canada. Tel.: 1-403-220-8525.
Faculty of Kinesiology, University of Calgary, Calgary, Alberta, CanadaBiomechanics Laboratory, School of Sports, Federal University of Santa Catarina, Florianopolis, SC, Brazil
Increased levels of pro-inflammatory cytokines are associated with the release of degradative enzymes leading to osteoarthritis (OA) development. Although physical exercise (PE) is generally recognized as beneficial for OA symptoms, excessive training workload and eccentric muscular exercise have increased OA risk. Here, we investigated the effects of excessive exercise workload and exercise type on systemic inflammation and knee joint OA.
Methods
Mice were divided into five groups: sedentary (SED), uphill training (TRU), downhill training (TRD), excessive uphill training (ETU), and excessive downhill training (ETD) for an 8-week training intervention protocol.
Results
ETD group had increased pro-inflammatory cytokines in serum, vastus lateralis (VL), and vastus medialis (VM) muscles, while ETU group mice had increased cytokine levels in the VL and VM. Total knee joint OARSI score were more significant in ETD group compared to SED and TRU groups. They were also more meaningful for the medial tibial plateau of ETD group compared to SED group. MMP-3 and cleaved Caspase-3 were higher in the ETD group than the SED and TRU group, while Adamts-5 was higher in the ETD group than the SED group. TRU group had increased PRG-4 levels compared to ETU and ETD group. ETD group had decreased total bone volume, trabecular bone volume, and cortical thickness compared to SED group.
Conclusion
Excessive downhill training induced a chronic pro-inflammatory state in mice and was associated with early signs of cartilage and bone degeneration that are clinical indicators of knee OA.
Matrix metalloproteinase and pro-inflammatory cytokine production by chondrocytes of human osteoarthritic cartilage: associations with degenerative changes.
. Moreover, IL-1beta promotes the release of MMP-3 (Matrix Metalloproteinase-3) and the aggrecanase Adamts-5 (a disintegrin and metalloproteinase with thrombospondin motifs-5)1, enzymes involved in articular cartilage matrix degradation, a hallmark of OA
Proteoglycan 4: from mere lubricant to regulator of tissue homeostasis and inflammation: does proteoglycan 4 have the ability to buffer the inflammatory response?.
Risk of osteoarthritis associated with long-term weight-bearing sports: a radiologic survey of the hips and knees in female ex-athletes and population controls.
Arthritis Rheum: Off J Am Coll Rheumatol.1996; 39: 988-995
, may increase the risk of OA. Overtraining (OT), usually used to improve athletic performance, may lead to a nonfunctional overreaching (NFOR) state characterized by decreased physical performance
Prevention, diagnosis and treatment of the overtraining syndrome: joint consensus statement of the European College of sport science (ECSS) and the American College of sports medicine (ACSM).
. Mice submitted to excessive downhill/uphill running had increased serum levels of IL-6, IL-1beta, and TNF-alpha, developed muscular insulin resistance, hypothalamic inflammation, and hepatic fat accumulation
Excessive eccentric exercise leads to transitory hypothalamic inflammation, which may contribute to the low body weight gain and food intake in overtrained mice.
. However, the effects of chronic systemic inflammation caused by the NFOR state on cartilage degradation have not been investigated.
Aside from the metabolic implications of OT, eccentric exercise (EE), as occurs in downhill running, has been associated with increased cartilage degeneration and chondrocyte death compared to concentric exercise (uphill running)
. Mice subjected to excessive downhill running have been found to have increased IL-6 and TNF-alpha and pro-inflammatory proteins IkappaB kinase (IKK) and SOCS3 (Suppressor of cytokine signaling 3) in skeletal muscle
Therefore, this study aimed to determine the impact of moderate eccentric and concentric training, and excessive eccentric and concentric training leading to an NFOR state, on systemic inflammation and knee joint OA. To our best knowledge, this is the first study to document the early onset of OA associated with systemic inflammation caused by overtraining.
Methods
Animals
Eight-week-old C57BL/6 mice were divided randomly into five groups: SED (sedentary n = 14), TRU (moderate uphill training n = 14), TRD (moderate downhill training n = 14), ETU (excessive uphill training n = 13), ETD (excessive downhill training n = 12). All experimental procedures followed the Brazilian College of Animal Experimentation and were approved by the Ethics Committee of the University of São Paulo (I.D. 2016.5.84.90.0). Fig. 1(A) illustrates the experimental protocol schematically.
Fig. 1(A) Experimental design of the study. (B–D) Serum cytokines in pg.ml−1. (B) serum IL-1β; (C) serum IL-6; (D) serum TNF-alpha. n = 12–14 per group. ∗significant different from SED. ∗∗significant different from TRU. +significant different from TRD. #significant different from ETU. ++significant different for ETD. P < 0.05.
The running intensities for the moderate and overtraining protocols were prescribed based on each animal's aerobic fitness and maximum velocity (MV), which was determined during a performance test preceding the training protocols
. The moderate uphill and downhill training protocols lasted for 8 weeks, and the animals ran 3 days a week (Mon/Wed/Fri) for a total of 1080 min of running. The uphill and downhill overtraining protocols consisted of five consecutive days of training (Mon–Fri) for a total volume of 2475 min
After the 8-week exercise-intervention protocol, mice were fasted for 12 h and anesthetized by intraperitoneal administration of xylazine (10 mg/kg body weight) and ketamine (100 mg/kg body weight). Once deeply anesthetised
, mice were decapitated, total blood was collected and centrifuged at 1100 G for 15 min at 4°C and stored at −80°C for subsequent cytokine analysis using Luminex™ (Millipore, ST Charles, MO). Vastus lateralis (VL) and vastus medialis (VM) muscles were harvested and stored at −80°C. Knee joints were trimmed, cutting the femur and tibia/fibula approximately 0.5 cm above and below the joint line.
Micro-computed tomography (μCT)
μCT scans were obtained using a vivaCT 40 – in vivo μCT (Scanco Medical AG, Bassersdorf, Switzerland). Joints were fixed with low-density material, scanned at an isotropic resolution of 15 μm with 55 kVp tube voltage, an integration time of 200 ms, and a tube current of 145 μA. Image Processing Language (IPL V5.08b, Scanco Medical, Brüttisellen, Switzerland) was used to analyze the scans via segmentation
. Regions of interest (ROI) were generated by cropping the femur and tibia/fibula at the metaphyseal growth plate and obtaining 120 slices. Using the growth plate as a landmark, changes of the bone near the cartilage were quantified. Trabecular bone morphology consisting of total volume (TV), bone volume (BV), bone volume fraction (BV/TV), trabecular spacing (TS), trabecular thickness (TBT), connectivity density (CD), bone mineral density (BMD), and cortical bone morphology consisting of mean cortical thickness (CT) were calculated from the ROI
Histological and Osteoarthritis Research Society International (OARSI) scoring
After μCT analysis, the intact knee joints were stored in formalin for 2 weeks. Samples were decalcified using Cal-Ex II decalcifying solution (Fisher Scientific, Hampton, NH) for 14 days, with the solution changed daily. Samples were washed with purified water, underwent tissue processing and paraffin embedding. Joints were sectioned on a Leica RM2255 (Wetzlar, Germany) microtome and stained with Hematoxylin, Fast green, and Safranin-O using a Leica ST5010 Autostainer XL (Wetzlar, Germany) to visualize cell nuclei, collagen, and proteoglycans. Sections were imaged with an Olympus BX53 (Tokyo, Japan) using cell-sens standard 1.18 (Olympus, Tokyo, Japan). Joint sections were scored using the Osteoarthritis Research Society International (OARSI) scoring system
. Two blinded assessors scored six regions of the knee from 6 to 8 animals/group: medial and lateral tibial plateaus, medial and lateral femoral condyles, patella, and femoral groove, totalizing 296 sections. The total OARSI scores reflect the sum of the six regional scores.
Immunohistochemistry
Non-stained joint sections were separated and placed in a histological bath for deparaffinization and rehydration. Afterward, antigen retrieval was performed with a 10 mM sodium citrate solution (70°C for 1 h). Slides were then blocked in goat serum (1:500 dilution, 1 h) and were washed for 5 min in a TBS-T (tris buffered saline with tween 20) solution. The specific primary antibodies anti-Adamts-5 (PA5-27165), anti-cleaved-Caspase-3 (PA5-23921), anti-PRG-4 (MABT401), and anti-MMP-3 (AB_2566077) were conjugated with fluorophores using a DyLight fast conjugation kit [AB_201803 (650); AB_201800 (550), Abcam, MA, USA]. Slides were labeled, and 20 μl of antibody solution was placed in each section. In control sections, 20ul of TBS-T was used. Slides were placed in a humidity chamber and stored at −4°C overnight. Slides were then washed three times for 10 min with TBS-T. EverBrite Mounting media with DAPI (Biotium) was placed on the slides, covered with a coverslip, and left to dry in the dark at room temperature. Slides were scanned with an Axion Scan.Z1 (Zeiss, Germany) microscope and analyzed on the Zen 2.5 lite software (Zeiss). An ROI was selected, and cell staining was quantified. Quantification was normalized based on background autofluorescence of the negative control, and those data were used to quantify positive and negative cells. The total score was obtained as the sum of the values from the lateral and medial knee joint compartments. Antibodies were obtained from Thermo Fisher Scientific (MA, USA), EMD Millipore (MA, USA), and BioLegend Way (CA, USA).
Immunoblotting
After homogenizing the skeletal muscle samples in an extraction buffer at 4°C, the extracts were centrifuged (9900g) for 40 min at 4°C to remove insoluble material, and the supernatants were used for quantification of proteins using the method of Bradford
. Proteins were run on an SDS-PAGE gel and transferred to nitrocellulose membranes (GE Healthcare Hybond ECL, RPN303D). Membranes were blocked with Tris-buffered saline (TBS) with 0.1% Tween-20 containing 5% BSA (bovine serum albumin) or 5% milk (BioRad) for 1 h at 4°C, then incubated overnight with a specific antibody [anti-IL-6 (AB9324), anti-IL-10 (sc-JES5-2S5), anti-TNF-alpha (60,291), anti-IL1-beta (OAPA 00159), anti-IL-15 (sc-7889), anti-IKK (sc-34674), phospho-IKK (sc-23470), anti-SOCS3 (sc-518020), and anti-GAPDH (14C10)]. After washing with TBS containing 0.1% Tween-20, membranes were incubated for 1 h at room temperature with anti-mouse or anti-rabbit secondary antibody (1:10,000 dilution) (GE Healthcare). The specific immunoreactive bands were detected through chemiluminescence (GE Healthcare, ECL Plus Western Blotting Detection System, RPN2132). Images were acquired by C-DiGitTM Blot Scanner (LI-CORR, Lincoln, Nebraska) and quantified using Image Studio software for C-digit Blot Scanner. Antibodies were acquired from Cell Signaling Technology (MA, USA), Abcam (MA, USA), Aviva Systems Biology Corporation (CA, USA), Santa Cruz Biotechnology (CA, USA), and Proteintech Group (IL, USA) according to availability. Routine chemical reagents were purchased from Sigma Chemical Corporation (St. Louis, MO, USA).
Blood cytokine analysis
Serum concentrations of IL-1beta, IL-6, and TNF-alpha were evaluated using Luminex TM multiplex reagents according to the manufacturer's instructions (Millipore, St Charles, MO). A MILIPLEX MAP Mouse Cytokine Panel – 3 Plex (Millipore, cat. number MPXMCYTO-70K) was used for cytokine measurements. Samples were collected on the Luminex MAP200 instrument and were analyzed using the 3.1 xPONENT System.
Statistical analysis
Analysis was performed using the GraphPad Prism software version 8 (San Diego, CA, USA). Data are expressed as means ± standard errors of the mean. Non-parametric Kruskal–Wallis with Dunn's post-hoc test was performed. Spearman correlation was used to quantify associations between the total knee OARSI scores and all μCT-derived bone variables. The level of significance was set at P < 0.05. No prior power/sample size calculations were performed. We based the sample size on the previous investigations
Excessive eccentric exercise leads to transitory hypothalamic inflammation, which may contribute to the low body weight gain and food intake in overtrained mice.
that analyzed the same overtraining protocols as used in the current study.
Results
Body mass
SED = 24.5g ± 0.7; TRU = 23.7g ± 1.3; TRD = 24.6g ± 1.3; ETU = 23.3g ± 1.2; ETD = 22.5g ± 1.7. There were no statistically significant differences between groups.
Serum cytokines
TRU (0.65 ± 0.34), TRD (0.56 ± 0.23), and ETU (0.58 ± 0.23) group had decreased IL-1β levels compared to the SED group (1.2 ± 0.34) (P = 0.04, P = 0.001, and P = 0.005, respectively), while the ETD group (1.28 ± 0.58) had increased IL-1β levels compared to TRD (P = 0.01) and ETU (P = 0.03) groups [Fig. 1(B)]. IL-6 levels for the ETD group (2.5 ± 1.24) were increased compared to the SED (1.5 ± 0.28) (P = 0.003), TRU (1.62 ± 0.39) (P = 0.01), and TRD (1.54 ± 0.41) (P = 0.006) groups [Fig. 1(C)]. TNF-alpha levels were the same for all groups [Fig. 1(D)].
Immunoblotting
Vastus lateralis (VL): There were no statistically significant differences between groups for IKKβ and IL-15 [Fig. 2(A) and (E)]. ETU group (4.22 ± 1) had increased IL-1β levels compared to SED group (1.38 ± 0.53) (P = 0.04) [Fig. 2(B)], increased IL-6 levels (5.3 ± 2.1) compared to SED (2.1 ± 0.7) (P = 0.01), TRU (2.23 ± 0.67) (P = 0.01), and TRD (2.29 ± 0.37) (P = 0.02) groups [Fig. 2(C)], and increased TNF-α levels (4.16 ± 2.15) compared to SED (0.3 ± 0.26) (P = 0.001), TRU (0.43 ± 2.15) (P = 0.001), and TRD (0.79 ± 0.35) (P = 0.004) groups [Fig. 2(G)]. ETD group had increased IL-1β (4.63 ± 0.31) compared to SED (P = 0.04 and P = 0.003) group [Fig. 2(B)] and decreased SOCS3 levels (2.34 ± 0.73) compared to SED group (4.85 ± 1.5) (P = 0.04) [Fig. 2(F)]. TRU group increased IL-10 levels (5.7 ± 1.1) compared to ETD group (3.41 ± 0.61) (P = 0.004) [Fig. 2(D)].
Fig. 2Vastus lateralis muscle immunoblotting proteins. n = 4–6 per group. (A) IKKβ; (B) IL-1β; (C) IL-6; (D) IL-10; (E) IL-15; (F) SOCS3; (G) TNF-α. ∗significant different from SED. ∗∗significant different from TRU. +significant different from TRD. #significant different from ETU. ++significant different for ETD. P < 0.05.
VM: There were no statistically significant differences between groups for IKKβ, IL-10, IL-15, and SOCS3 levels (Fig. 3(A), (D), 3(E), and 3(F)). ETU group (4.3 ± 0.74) had increased IL-1β levels compared to SED (3.08 ± 0.31) (P = 0.01) [Fig. 3(B)], and increased TNF-α levels (2.8 ± 1.2) compared to SED (0.31 ± 0.25) (P = 0.03) and TRU (0.32 ± 0.3) (P = 0.04) groups [Fig. 3(G)]. ETD group (4.37 ± 0.42) had increased IL-1β levels compared to SED (P = 0.01) [Fig. 3(B)], increased IL-6 levels (6.3 ± 2.37) compared to SED (2.76 ± 1.5) (P = 0.02) and TRU (2.74 ± 0.28) (P = 0.04) [Fig. 3(C)], and increased TNF-α levels (4.62 ± 2.9) compared to SED (P = 0.02) and TRU (P = 0.03) groups [Fig. 3(G)].
Fig. 3Vastus medialis muscle immunoblotting protein. n = 4–6 per group. (A) IKKβ; (B) IL-1β; (C) IL-6; (D) IL-10; (E) IL-15; (F) SOCS3; (G) TNF-α. ∗significant different from SED. ∗∗significant different from TRU. +significant different from TRD. #significant different from ETU. ++significant different for ETD. P < 0.05.
There were no significant differences in OARSI scoring for the femur, patella, trochlear groove, and lateral tibial plateau of the knee between groups (data not shown). Also, we did not observe osteophytes or abnormalities in the synovium. However, OARSI scores were higher (and thereby knee OA more pronounced) in the medial tibial plateau (P = 0.008) of ETD group (2.2 ± 0.37) compared to SED group (0.12 ± 0.35) [Fig. 4(B)]. Also, the total OARSI score was higher for ETD group (8.02 ± 1.8) compared to SED (3.43 ± 2.12) and TRU groups (3.53 ± 2.35) (P = 0.02) [Fig. 4(A)]. In general, defects were contained within the articular cartilage and did not reach the calcified cartilage. Most lesions were located on the posterior aspect of the tibia. SED and TRU knee joints showed no disruptions of the cartilage and had robust Safranin O staining, indicating the presence of proteoglycans [Fig. 4(C) and (D)]. TRD and ETU groups frequently had frayed cartilage zones and disruption between the superficial and mid-zone layers (arrows in Fig. 4(E), (F), (J), and (K)). ETD group had the most severe lesions presenting as cracked surfaces and deep erosions (arrows in Fig. 4(G) and (L)).
Fig. 4Histological cartilage analysis. n = 6–8 per group (A) Total OARSI score; (B) Medial Tibia OARSI score. Representative medial Tibia compartments: (C) SED; (D) TRU; (E) TRD; (F) ETU; (G) ETD. ∗significant different from SED. ∗∗significant different from TRU. +significant different from TRD. #significant different from ETU. ++significant different for ETD. P < 0.05.
There was a reduction of TV (6.26 ± 1.64) (P = 0.006) BV (2.07 ± 0.55) (P = 0.003), and cortical thickness (0.06 ± 0.002) (P = 0.03) for ETD group compared to SED group (10.1 ± 1.75; 3.24 ± 0.19; 0.081 ± 0.003 respectively) [Fig. 5(A), (B), and (E)]. There were no statistically significant differences in BV/TV, TS, TBT, CD, and BMD between groups. There was a significant negative correlation between OARSI scores and TV (r = −0.349; P = 0.047) and BV (r = 0.374; P = 0.031).
Fig. 5Micro-computed tomography scans of mice knee joint. n = 6–8 per group (A) Total volume; (B) Bone volume; (C) BV/TV %; (D) Connective density; (E) Cortical thickness; (F) Trabecular spacing; (G) Trabecular thickness; (H) Bone mineral density. ∗significant different from SED. ∗∗significant different from TRU. +significant different from TRD. #significant different from ETU. ++significant different for ETD. P < 0.05.
There was an increase of MMP-3 for ETD group (23.2 ± 10.7) compared to SED (9.7 ± 7.9) (P = 0.02) and TRU (9.3 ± 5.1) (P = 0.03) groups [Fig. 6(B)]. The ETD group (14.9 ± 3.3) had increased Adamts-5 positive chondrocytes than SED group (5 ± 4.7) (P = 0.01) [Fig. 6(A)]. Furthermore, there were increased levels of cleaved Caspase-3 positive chondrocytes in ETD group (23.8 ± 9.2) compared to SED (7.2 ± 4.2) (P = 0.002) and TRU (11.4 ± 3.9) (P = 0.03) groups [Fig. 7(A)]. TRU group (8.3 ± 1.3) had increased PRG-4 positive chondrocytes compared to ETD (2.5 ± 1.7) and ETU (2.2 ± 2.7) groups (P = 0.008; P = 0.01) [Fig. 7(B)].
Fig. 6Immunohistochemical analysis of articular cartilage and representative images of medial and lateral cartilage compartments of each group. (A) Total Adamts-5; (B) Total MMP-3. n = 7 per group. ∗significant different from SED. ∗∗significant different from TRU. +significant different from TRD. #significant different from ETU. ++significant different for ETD. P < 0.05.
Fig. 7Immunohistochemical analysis of articular cartilage and representative images of medial and lateral cartilage compartments of each group. (A) Total cleaved Caspase-3; (B) Total PRG-4. n = 7 per group. ∗significant different from SED. ∗∗significant different from TRU. +significant different from TRD. #significant different from ETU. ++significant different for ETD. P < 0.05.
The main findings of this study were: (i) ETD group had increased serum and skeletal muscle pro-inflammatory cytokines, while ETU group had increased skeletal muscle pro-inflammatory cytokines. TRU group had increased anti-inflammatory cytokine IL-10 levels in skeletal muscle; (ii) ETD group showed distinct signs of early knee OA; (iii) ETD group had increased MMP-3, cleaved Caspase-3 and Adamts-5 levels in chondrocytes, while TRU group had increased PRG-4 levels; (iv) ETD group had decreased BV, TV, and cortical thickness. These findings support our hypothesis that excessive downhill training led to OA's premature onset in this model, which may be caused by increased chronic systemic and local inflammation.
. The lower serum IL-1beta in TRU, TRD, and ETU group compared to the SED group may be explained by the timing of the blood sample collection which occurred 36h after the last training session. The timing may also explain the inflammatory state in the ETD group, as we would expect IL1-beta to be lower 36h after exercise in these animals. The inflammatory condition is reinforced by the increased levels of IL-6 in the ETD group, in agreement with earlier findings using the same excessive training protocol
. In the ETU group, we observed local inflammation, which was characterized by increased IL-1beta and TNF-alpha levels in VM and VL, as well as IL-6 in VL. These results corroborate previous findings of increased IL-6 and TNF-alpha levels in soleus and gastrocnemius after excessive uphill training
. However, these local events were insufficient to produce a rise in circulating pro-inflammatory markers.
The ETD group also had local inflammation characterized by increased levels of IL-1beta in VL and VM samples, as well as increased IL-6 and TNF-alpha levels in VL samples, contributing to the increased systemic IL-1beta and IL-6 levels. It has been shown previously that TNF-alpha, IL-1beta, and IL-6 are increased in gastrocnemius, and IL-1beta and IL-6 are increased in extensor digitorum longus (EDL) and soleus after excessive downhill training
. Also, IL-10 is an anti-inflammatory cytokine that can increase muscle insulin sensitivity and protect against inflammation. It is decreased systemically in obese individuals with insulin resistance
. Here, we measured smaller levels of IL-10 in VL in the ETD group than the TRU group, suggesting increased inflammation in the first and decreased inflammation in the latter.
observed micro-injuries with macrophage infiltration in EDL and soleus with neutrophil infiltration in EDL after the ETD protocol, corroborating the previous findings of Fielding et al.
, who determined that the muscle damage caused by eccentric contraction was associated with IL-1beta accumulation and neutrophil infiltration. These findings suggest that eccentric exercise, such as downhill running, may cause more damage and heightened pro-inflammatory insult than concentric exercise (uphill running). Therefore, the heightened inflammatory state of the ETD group mice, caused by the eccentric exercise, may contribute to the early onset of OA.
Early-onset of OA and mechanical loading
Mechanical loading of joints is essential and can elicit anabolic or catabolic responses, depending on the duration and intensity of loading, the type of muscular contraction, and age
observed a rise of 54% for downhill (9° slope) and a decrease of 22% for uphill (6°) in peak ground reaction forces compared to level running. We observed a premature onset of knee OA in the ETD group, corroborating findings that suggest that excessive running coupled with eccentric exercise may elicit a degenerative/OA response
. Even though the ETU group also ran excessively, the predominance of eccentric muscle action in the ETD group may have been responsible for the more significant damage of the knee joints in the ETD group than the SED and TRU group. These findings suggest that the NFOR state, caused by excessive eccentric exercise, causes a chronic inflammatory state, leading to the onset of knee OA in this pre-clinical mouse model.
Inflammation leads to the early onset of OA through degradative enzyme activation and apoptosis
There is increasing evidence demonstrating a close relationship between pro-inflammatory cytokines and OA
. Chondrocytes in OA cartilage express and synthesize IL-1 in concentrations capable of inducing MMP expressions and co-localize with MMP-3 in regions of ECM degeneration
observed increased MMP-3 and OA development levels after excessive exercise. We also measured increased MMP-3 staining in chondrocytes of the ETD group compared to the SED group. More chondrocytes in ETD than SED stained for Adamts-5, a major aggrecan-degrading enzyme. Adamts-5 has been implicated in cartilage degeneration through enhanced proteolysis
observed increased Syndecan-4, an Adamts-5 inductor, in the cartilage of animals subjected to forced exercise, with data suggesting Adamts-5 activation was mediated by MMP-3 release. Thus, the increased Adamts-5 observed in the ETD group may be related to the increased systemic inflammation and the high levels of TNF-alpha in muscles and MMP-3 in cartilage. Therefore, the early knee OA observed here may be associated with an increase in Adamts-5 levels in response to excessive downhill training.
Pro-inflammatory cytokines can induce chondrocyte apoptosis through Caspase-3 signaling mediation
observed an intensity-dependent activation of PRG-4 in animals performing a low and moderate-intensity exercise for 8 weeks, while animals performing intense exercise displayed lower levels of PRG-4. Our PRG-4 data corroborate these findings: moderate uphill training enhanced PRG-4 compared to excessive downhill and uphill running. The decrease of PRG-4 for the ETU group may be related to the observed increase in skeletal muscle pro-inflammatory cytokines. Furthermore, the cleaved Caspase-3 increase in ETD group mice may be mediated by the enhanced levels of pro-inflammatory cytokines in serum and muscle, thereby decreasing PRG-4 levels.
Subchondral bone adaptations
There was a decrease in BV, TV, and cortical thickness for the ETD group. Although BV/TV is often observed to increase in OA
, BV and TV were reduced proportionally here. BV/TV reflects trabecular bone connectivity; however, it is not a good indicator of bone formation and resorption
. Here, we observed that increasing OARSI scores were associated with decreased BV and TV values. Cortical bone mass is considered a better determinant of bone strength than trabecular bone connectivity (BV/TV)
. We observed lower cortical thickness in ETD than SED group, which may indicate a weakened bone. Furthermore, pro-inflammatory cytokines, such as IL-1beta and IL-6, are found in osteoblasts during OA progression, and osteoblast compression increases IL-6 and MMP-3 release, indicating a relationship between IL-6 and OA
, which may have caused the lower cortical thickness in our ETD compared to SED group.
Conclusions
In contrast to the early signs of OA observed after excessive downhill running, moderate uphill training displayed a healthy joint with no disruptions in the cartilage and robust Safranin-O staining, indicating the presence of proteoglycans. Also, moderate uphill training increased the levels of IL-10 (an anti-inflammatory cytokine) in VL and PRG-4 in chondrocytes. PRG-4 improves joint lubrification and plays a role in buffering inflammation
Proteoglycan 4: from mere lubricant to regulator of tissue homeostasis and inflammation: does proteoglycan 4 have the ability to buffer the inflammatory response?.
, highlighting the benefits of this training type. These findings corroborate previous studies showing the beneficial role of moderate training in pre-clinical models of metabolic OA
. Moderate downhill running and excessive uphill running did not present early stages of OA; however, these groups did not display a healthy joint, which was highlighted by the presence of frayed zones and disruption in the superficial layer of cartilage. Although the eccentric component in the TRD group and exhaustion characteristic in the ETU group may have played a role in the lack of positive outcomes, future investigations should test these assumptions.
In summary, we suggest that excessive downhill training led to the premature onset of knee OA. The following scenario seems feasible and should be further explored. The excessive exercise led to an NFOR state associated with chronic systemic and local muscular inflammation, similar to that observed in patients with metabolic syndrome. Combined with the muscular damage known to occur with eccentric exercise and the high external loading, cartilage and bone started to degenerate. We suggest that further studies be conducted to investigate if athletes who experience overtraining in sports that rely heavily on eccentric muscle action (i.e., Alpine skiing) present excessive inflammation and develop early onset of OA.
Limitations
No prior power/sample size calculations were performed. Therefore, it is possible that we may have been underpowered to detect moderate associations. While the alpha level for each set of pairwise comparisons performed for each outcome was controlled using Dunn's post-hoc test, we did not adjust for the overall number of outcomes (n= >30 outcomes).
Authors' contributions
G.P.M., A.S.R.S., and W.H. conceived and designed research; G.P.M., and A.L.R., performed chronic protocols; R.S., and C.C. performed histological experiments; G.P.M., and C.C. performed bone experiments; G.P.M., and A.M. performed immunohistochemical analyses; G.P.M. performed blood experiments and immunoblotting experiments; G.P.M., A.S.R.S., and W.H. analyzed data; G.P.M., R.K., A.S.R.S., and W.H. interpreted experiment results; G.P.M. prepared figures; G.P.M. drafted the manuscript; G.P.M., C.C., A.M., R.K., A.S.R.S., and W.H. edited and revised manuscript; G.P.M., C.C., A.M., R.S., A.L.R., R.K., A.S.R.S., and W.H. approved the final version of the manuscript.
Competing interests
The authors declare that they have no competing interests.
Acknowledgments
The present work received financial support from the São Paulo Research Foundation (FAPESP; process numbers 2016/25766-4, and 2017/13251-2), National Council for Scientific and Technological Development (CNPq; process number 301279/2019-5), the Coordination for the Improvement of Higher Education Personnel (CAPES; finance code 001), The Canada Research Chair Program, The Killam Foundation, The Canadian Institutes for Health Research, Alberta Innovates and UCalgary Eyes High Ph.D. Studentships, Canada Research Chair in Bone and Joint Stem Cell Biology and NSERC Discovery Grant.
References
Tetlow L.C.
Adlam D.J.
Woolley D.E.
Matrix metalloproteinase and pro-inflammatory cytokine production by chondrocytes of human osteoarthritic cartilage: associations with degenerative changes.
Proteoglycan 4: from mere lubricant to regulator of tissue homeostasis and inflammation: does proteoglycan 4 have the ability to buffer the inflammatory response?.
Risk of osteoarthritis associated with long-term weight-bearing sports: a radiologic survey of the hips and knees in female ex-athletes and population controls.
Arthritis Rheum: Off J Am Coll Rheumatol.1996; 39: 988-995
Prevention, diagnosis and treatment of the overtraining syndrome: joint consensus statement of the European College of sport science (ECSS) and the American College of sports medicine (ACSM).
Excessive eccentric exercise leads to transitory hypothalamic inflammation, which may contribute to the low body weight gain and food intake in overtrained mice.
00672-5&id=gr1.jpg){kind=link}
00672-5&id=gr2.jpg){kind=link}
00672-5&id=gr3.jpg){kind=link}
00672-5&id=gr4.jpg){kind=link}
00672-5&id=gr5.jpg){kind=link}
00672-5&id=gr6.jpg){kind=link}
00672-5&id=gr7.jpg){kind=link}