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
Department of Molecular and Experimental Medicine, The Scripps Research Institute, La Jolla, CA, USANational Research Institute for Child Health and Development, Tokyo, JapanDepartment of Systems Bio Medicine, Tokyo Medical and Dental University, Tokyo, Japan
Address correspondence and reprint requests to: M.K. Lotz, Department of Molecular and Experimental Medicine, MEM-161, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037, USA.
Department of Molecular and Experimental Medicine, The Scripps Research Institute, La Jolla, CA, USA
Aging is a major risk factor for osteoarthritis (OA). Forkhead-box class O (FoxO) transcription factors regulate mechanisms of cellular aging, including protein quality control, autophagy and defenses against oxidative stress. The objective of this study was to analyze FoxO transcription factors in normal, aging and OA cartilage.
Knee joints from humans ages 23–90 and from mice at the age of 4–24 months and following surgically induced OA were analyzed for expression of FoxO proteins. Regulation of FoxO protein expression and activation was analyzed in cultured chondrocytes.
Human cartilage expressed FOXO1 and FOXO3 but not FOXO4 proteins. FOXO1 and FOXO3 were more strongly expressed the superficial and mid zone as compared to the deep zone and were mainly localized in nuclei. During human joint aging, expression of FOXO1 and FOXO3 was markedly reduced in the superficial zone of cartilage regions exposed to maximal weight bearing. In OA cartilage, chondrocyte clusters showed strong FOXO phosphorylation and cytoplasmic localization. Similar patterns of FOXO expression in normal joints and changes in aging and OA were observed in mouse models. In cultured chondrocytes, IL-1β and TNF-α suppressed FOXO1, while TGF-β and PDGF increased FOXO1 and FOXO3 expression. FOXO1 and FOXO3 phosphorylation was increased by IL-1β, PDGF, bFGF, IGF-1, and the oxidant t-BHP.
Normal articular cartilage has a tissue specific signature of FoxO expression and activation and this is profoundly altered in aging and OA in humans and mice. Changes in FoxO expression and activation may be involved in cartilage aging and OA.
Potential involvement of oxidative stress in cartilage senescence and development of osteoarthritis: oxidative stress induces chondrocyte telomere instability and downregulation of chondrocyte function.
. One of the prominent pathways involved in turnover of cellular constituents is autophagy. Compromised autophagy associated with a reduction and loss of ULK1, Beclin1, and LC3 expression was observed in human OA and age-related and surgically induced OA in mice
One of the major signaling pathways that regulate cellular aging and stress resistance is the Insulin/IGF-1 signaling pathway. The protein kinase Akt is an important upstream signaling component in this pathway that regulates diverse cellular functions related to longevity, cellular senescence, and metabolism. Among of the most evolutionarily conserved targets of Akt are the forkhead-box class O (FOXO) transcription factors
. Activity of the three FOXO molecules is controlled by phosphorylation, which modulates their cellular localization. Akt directly phosphorylates FOXO1, FOXO3, and FOXO4 at three conserved sites, resulting in nuclear export and subsequent degradation
. The objectives of this study were to analyze protein expression and activation of FOXO transcription factors in normal cartilage and to determine changes in aging and OA.
Human knee joints
Human knee joints from individuals ages 23–90 were obtained at autopsy under approval by the Scripps Human Subjects Committee. The entire femoral condyles of young normal knee joints were harvested from six donors (age 23–48 years, mean ± SD = 36.0 ± 9.6, OA grade I, Mankin score = 0) having no history of joint disease. Aged normal knee joints were also obtained at autopsy from four donors having no history of joint diseases or overt OA (age 68 to 76, mean ± SD = 72.5 ± 3.6, OA grade I–II, Mankin score = 1–3). Human OA joints were obtained from four donors (age from 64 to 90, mean ± SD = 81.5 ± 10.3, OA grade III–IV, Mankin score = 7–8). Articular surfaces were graded macroscopically according to a modified Outerbridge scale
. Osteochondral slabs (5 mm thickness) were harvested from the central part of medial femoral condyle for histomorphologic analysis. Subsequently, the slabs were cut into six tissue blocks from the anterior to the posterior condyle [Fig. 1(B)]. Each block was fixed in 10% zinc-buffered formalin for 2 days, decalcified in TBD-2 for 7 days, followed by paraffin embedding. Serial sections (4 μm each) were cut, stained with Safranin O-fast green, and graded according to Mankin scoring system
. For protein isolation directly from cartilage tissue, the cartilage strips obtained from patients undergoing total knee arthroplasty were separated based on the extent of degradation and frozen in liquid nitrogen. Normal cartilage samples were collected from young donors having no history of joint disease. The frozen tissues were crushed and homogenized. Samples were incubated in TRIzol (Invitrogen) at room temperature. After addition of chloroform, samples were vortexed vigorously and centrifuged for 15 min at 12,000× g at 4°C. The interphase and organic phase was collected, followed by addition of 100% ethanol. After centrifugation, the supernatant was collected for protein isolation. Proteins were precipitated by the addition of isopropanol and diluted in 6M Urea, 2% SDS.
Human chondrocyte cultures
The isolated chondrocytes were plated at high density in DMEM with 10% CS and antibiotics and allowed to attach to the culture flasks. The cells were incubated at 37°C in a humidified gas mixture containing 5% of CO2 balanced with air. The chondrocytes were used in the experiments at confluence (2–3 weeks in primary culture).
Human chondrocytes were seeded in six-well plates at a density of 4.0 × 105 cells/well. After 1 day, the cells were washed and incubated in DMEM with 0.5% CS for 24 h. Cytokines, growth factors, tert-Butyl hydroperoxide (t-BHP), and CS were added at the following final concentrations: IL-1β (1 ng/ml), TNF-α (10 ng/ml), IL-6 (10 ng/ml), TGF-β1 (10 ng/ml), BMP-7 (100 ng/ml), bFGF2 (25 ng/ml), PDGF-AA (25 ng/ml), IGF-1 (100 ng/ml), t-BHP (25 μM and 250 μM), and 10% CS. Cells were harvested after 30 min, 60 min, 1 day, 2 days, and 5 days of incubation.
Mouse knee joints
All animal experiments were performed according to protocols approved by the Institutional Animal Care and Use Committee (IACUC) at The Scripps Research Institute. In the spontaneous aging-related OA model, C57BL/6J mice were kept under normal conditions and knee joints were collected at 4, 12 and 24 months of age. The surgical OA model was induced in 4 months old C57BL/6J mice by transection of the medial meniscotibial ligament and the medial collateral ligament (MMTL + MCL) as described
Knee joints from both murine models were resected from both hind legs, fixed in 10% zinc-buffered formalin for 2 days, decalcified in TBD-2 for 24 h. Serial sections (4 μm each) were cut, and expression of FoxO proteins was analyzed by immunohistochemistry.
For antigen unmasking, the tissue sections were incubated with 2.5 mg/ml of hyaluronidase for 60 min at 37°C. After washing with phosphate buffered saline (PBS), sections were blocked with 10% goat serum for 1 h at room temperature. This condition was obtained after optimization and provided specific positive staining without non-specific signals. Anti-FOXO1A (1:50 dilution), Anti-FOXO1A (phospho S256) (1:100 dilution), Anti-FOXO3A (1:100 dilution), Anti-FOXO3A (phospho S253) (1:100 dilution), and negative control rabbit IgG (1 μg/ml) were applied with 0.1% Tween 20 and incubated overnight at 4°C. All primary antibodies were purchased from Abcam. Incubation with secondary antibody, substrate and hematoxylin were performed as described
. The percentages of positive cells for total FOXO and phosphorylated FOXO were determined independent on cellular localization. Cells were counted as positive for cytoplasmic p-FOXO when staining was predominantly cytoplasmic positive staining on sections that were counter-stained with hematoxylin [Fig. 2(A)]. The frequency of positive cells was expressed as a percentage relative to the total number of cells counted in each zone.
Quantification of FoxO immunoreactivity in mouse knee joints
Cartilage cellularity in C57BL/6J mice was quantified by counting the chondrocytes in a microscopic field
. Three pictures were taken under 40× magnification, representing the center of the femoral condyle that is not covered by the meniscus as well as the anterior and posterior femoral condyles covered by the meniscus. Then, the total numbers of cells and FoxO positive cells were counted in each section.
Quantitative western blotting
Quantitative western blotting was performed with the LiCor immunofluorescence detection system (Licor, Lincoln, NE). Primary antibodies from Cell Signaling were used: Anti-FOXO1A (1:1000 dilution), Anti-FOXO1A (phospho S256) (1:1000 dilution), Anti-FOXO3A (1:1000 dilution), Anti-FOXO3A (phospho S253) (1:1000 dilution), Anti-Akt (phospho S473) (1:1000 dilution), and GAPDH (1:5000) in 1/2× Odyssey buffer in PBS with 0.1% Tween 20. After washing in TBST, secondary antibodies goat anti-rabbit – IRDye 800 (1:5000 dilution) for FOXOs and goat anti mouse – IRDye 680 (1:10,000 dilution) for GAPDH, diluted in 1/2× Odyssey buffer in PBS with 0.1% Tween 20 and 0.01% SDS, were applied. Blots were washed in PBS and then water before acquisition on the LiCor Odyssey. In-lane background was removed (Median: Top/Bottom) before analysis with the Odyssey software version 3.0 (LiCor). Integrated intensity values (K counts) for each protein of interest were normalized to those of GAPDH.
Statistically significant differences between three groups were determined with Kruskal Wallis H-test and Friedman test. When a significant differences were found among three groups, Mann–Whitney U test and Wilcoxon signed-rank test were used to analyze the specific sample pairs for significant differences. The results are reported as median and quartile 25%–75%. P values less than 0.05 were considered significant.
FOXO protein expression in young normal human cartilage
Articular cartilage from normal human knee joints expressed predominantly FOXO1 and FOXO3 proteins but not FOXO4 as detected by western blotting [Fig. 1(A)]. Both non-fibrillated and fibrillated OA cartilage showed significant FOXO1 and FOXO3 reduction compared to normal cartilage [Fig. 1(A)]. Immunohistochemistry was used to determine the distribution and phosphorylation of FOXO1 and FOXO3. Locations in the knee joint differ in regard to exposure of articular cartilage to weight bearing and susceptibility to OA. To determine regional changes in FOXO protein expression, adjacent cartilage sections representing the entire femoral condyle (n = 6 per condyle) were analyzed [Fig. 1(B)]. The most proximal and the most distal sections (#1 or 2 and # 5 or 6) are exposed to minimal, while the central sections (#3 or 4) are exposed to maximal weight bearing [Fig. 1(C)]. We also assessed differences in FOXO among the superficial, mid and deep zone [Fig. 2(A)]. FOXO1 and FOXO3 proteins (n = 6 donors each) were more highly expressed in the superficial and middle than the deep zone [Fig. 2(B), (C)]. When determining nuclear/cytoplasmic localization with FOXO by counter-staining of hematoxylin, FOXO1 and FOXO3 were found mainly in the nuclei [Fig. 2(B), (C)]. Comparison of areas of cartilage exposed to minimal vs maximal weight bearing showed that FOXO3 protein in the deep zone was more strongly expressed in areas exposed to minimal weight bearing.
Phosphorylated FOXO1 and FOXO3 (n = 6 donors each) were detected at higher levels in the superficial and middle zones as compared to the deep zone. The levels of total FOXO proteins and phosphorylated FOXOs were significantly greater in the nucleus as compared to cytoplasm [Fig. 2(B), (C)].
In summary, among the FOXO isoforms, FOXO1 and FOXO3 proteins are most strongly expressed in normal cartilage. Overall, their protein expression is higher in the superficial and middle zone as compared to deep zone. In regard to regional differences, only FOXO3 protein in the deep zone was more strongly expressed in areas exposed to minimal weight bearing. Nuclear localization of FOXO proteins indicates that most cells in the superficial and mid zone express activated FOXO1 and FOXO3 proteins.
Aging and OA-associated changes in FOXO protein expression in human cartilage
For the analysis of aging and OA cartilage, we selected representative maximal weight bearing regions [#3 or as shown in Fig. 1(B)]. OA cartilage had no superficial zone in maximal weight bearing areas.
In aged donors, FOXO1 and FOXO3 (n = 4 donors each) were significantly decreased in the superficial zone compared with young normal cartilage (Fig. 3). In contrast, FOXO1 and FOXO3 in OA cartilage (n = 4 donors each) were significantly increased in the middle zone compared with normal cartilage. In terms of nuclear/cytoplasmic localization, FOXO1 and FOXO3 in the middle zone of OA cartilage were stronger in the cytoplasm compared with normal cartilage [Fig. 3(B)]. Moreover, phosphorylated FOXO1 and FOXO3 were significantly stronger in the middle zone compared with normal cartilage. This increase was due to the cell clusters localized in the middle zone in OA cartilage (Fig. 3). These cell clusters showed strong protein expression and phosphorylation of FOXO1 and FOXO3. These results indicate an age-related reduction in FOXO1 and FOXO3 protein expression in the superficial zone and increased phosphorylation and cytoplasmic localization in the OA cluster chondrocytes.
FoxO protein expression in normal, aging and OA mouse cartilage
Two different types of OA models in mice were used, including aging-related OA and mechanical overload induced OA in order to mitigate their limitations and to correlate with aging and OA-related changes in human knee cartilage. Both mouse models showed cartilage degradation with only small variations in OA severity.
Normal joints from skeletally mature 4-month-old C57BL/6J mice showed high levels of FoxO1 and FoxO3 proteins in the superficial and upper middle zones. FoxO1 and FoxO3 were localized mainly in the nucleus. The signals for the phosphorylated FoxOs were less intense as compared to the total FoxOs in 4-month-old mice. At the ages of 12 months and 24 months, there was a significant aging-related reduction in FoxO1- and FoxO3-positive cells compared with 4-month-old mice, as well as a significant reduction in FoxO1 and FoxO3 in 24-month-old mice compared with 12-month-old mice [Fig. 4(B)]. In 12- and 24-month-old mice, the reduction of FoxO protein was more marked in the meniscus non-covered regions compared to the meniscus-covered regions [Fig. 4(A)].
Articular cartilage in joints with surgical OA showed a reduction of FoxO1-positive cells and FoxO3-positive cells compared with non-operated 4-month-old mice. The quantitative analysis of positive cells showed a significant reduction of FoxO1 and FoxO3 10 weeks after surgery. In contrast, the cluster-like chondrocyte aggregates in fibrillated lesions, showed strong expression of FoxO1 and FoxO3 protein and also of the phosphorylated forms of FoxO1 and FoxO3.
Regulation of FOXO protein expression and activation in cultured human chondrocytes
In cultured human chondrocytes, FOXO1 and FOXO3 proteins were basally expressed, whereas FOXO4 was expressed to a lesser extent. IL-1β and TNF-α significantly repressed FOXO1 at 2 and 5 days. FOXO3 protein expression was also reduced by IL-1β at 5 days, but to a lesser extent by TNF-α. On the other hand, stimulation with TGF-β1 significantly increased FOXO1 protein at 2 days. FOXO3 was also increased 2 days and 5 days after the addition of PDGF (Fig. 5). Shorter treatment periods (30 min–48 h) did not significantly affect expression of the FOXO proteins.
Phosphorylation of FOXO1, FOXO3, and FOXO4 increased in chondrocytes stimulated with IL-1β, TNF-α, bFGF, PDGF, t-BHP, and serum (Fig. 6). We used antibodies directed against phospho-serine 256 of FOXO1 (which cross-react with FOXO4) or antibodies against phospho-serine 253 of FOXO3. These conserved sites are phosphorylated by Akt, which was also activated in chondrocytes upon these stimuli, as judged from blots with the Akt anti-phospho-serine 473 antibody (Fig. 6).
The present study is the first to elucidate the characteristics of the protein expression and activation of FOXO transcription factors in articular cartilage. Our results demonstrated that among three FOXOs, FOXO1 and FOXO3 proteins were highly expressed in normal human and mouse cartilage, suggesting that they play important roles in cartilage homeostasis. FOXO protein expression showed differences among cartilage zones with higher expression in superficial and middle zones compared to the deep zone. These findings are consistent with observations that MnSOD, one of the major FOXO target antioxidants, was abundantly expressed in the superficial but not in the deep zone of human cartilage
In aged human cartilage, the protein expression of FOXO transcription factors was significantly decreased in the superficial zone of regions exposed to maximal weight bearing. Similarly, in mouse knee joints there was an age-related reduction of FOXO positive cells especially in the meniscus non-covered regions. In bone, FoxO1 expression also progressively decreased with aging, whereas FoxO3 and FoxO4 levels remained stable
. In the present study, we observed increased FOXO positive cells in the middle zone of human OA cartilage and in fibrillated areas and osteophytes in mouse knees. Importantly, the cellular localization of FOXO in OA cartilage was predominantly in the cytoplasm in contrast to normal cartilage, indicating inactivation of FOXO transcription factors. The chondrocytes in these regions formed clusters containing cells that are abnormally activated and differentiated
Cell density and composition in OA cartilage are different as compared to normal tissue. Cell density is reduced in some areas but in fibrillated areas there are chondrocyte clusters, which represent a different cell phenotype as compared to the normal zone-specific chondrocytes. In non-weight bearing areas of cartilage from OA joints where the superficial zone is still present and there are no fibrillations, there is a reduction in the total number of cells, in the total number of FOXO positive cells and in the percentage of FOXO positive cells.
The activity of FOXOs is tightly controlled by a phosphorylation-dependent shuttling system that modulates their cellular localization
and this is predominantly controlled by the Insulin/IGF-1 pathway acting through PI3K and Akt-mediated phosphorylation of FOXO. Phosphorylation of FOXOs by Akt provokes their nuclear export. In this study, we observed significantly higher levels of phosphorylated FOXOs in the cytoplasm of OA cartilage compared to normal cartilage. High levels of IGF-1 and IGF-1 receptor have been found in human OA cartilage, especially in chondrocytes of the upper zone
. These prior observations suggest a similar pattern that may be mechanistically related to the specific distribution of phosphorylated FOXOs seen in this study. Mechanical load is an alternative mechanism for dysregulated FOXOs in OA cartilage, as shear stress downregulated FOXO DNA-binding activity by Akt activation in muscle
In cultured normal chondrocytes IL-1β and TNF-α reduced FOXOs protein expression, and increased phosphorylation of FOXOs. This observation is similar to synovial cells where FOXO1 and FOXO4 were phosphorylated following stimulation with IL-1β and TNF-α
. Therefore, proinflammatory cytokines appear to inhibit the activity of FOXOs. On the other hand, our results showed that TGF-β up-regulated FOXO1 protein expression but did not affect phosphorylation, whereas PDGF up-regulated FOXO3 and increased phosphorylation of all FOXOs. PDGF was reported to induce the phosphorylation and inactivation of FOXO protein in hepatic stellate cells
. Upon treatment with H2O2, c-Jun-N-terminal kinase (JNK) phosphorylates FOXO at two threonine residues (Thr447 and Thr451 in FOXO4), which are different from the Akt phosphorylation sites, resulting in increased nuclear localization and activation of transcriptional activity
. In contrast, oxidative stress can increase Akt activity and subsequent FOXO phosphorylation, dependent on the cellular context. Our results in cultured chondrocytes stimulated with t-BHP showed FOXO phosphorylation (serine 256 of FOXO1 and serine 253 of FOXO3). The opposing forces of Akt vs JNK signaling are considered to determine whether FOXO will direct a transcriptional response regulating entry into quiescence or senescence
. To address functions of FOXO and phosphorylated FOXO in chondrocytes, further studies with manipulating FOXO expression are needed.
In summary, the present results suggest that attenuated FOXO protein expression or altered FOXO activation may represent a novel mechanism in the development of OA. In view of their central function in aging, FOXO proteins may provide a potential molecular target for the treatment of OA.
ML, YA, HA and YI conceived of the study, and participated in its design and coordination. YA, AH and MS carried out histology and immunohistochemistry experiments and performed quantitative analysis. All authors read and approved the final manuscript. Dr Lotz 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.
This study was conducted with the approval of the Human Subjects Committee and the Institutional Animal Care and Use Committee at The Scripps Research Institute.
Role of funding source
This study was supported by National Institutes of Health grants AG007996 and AR050631 , the Sam and Rose Stein Endowment Fund , a grant from National Center for Child Health and Development (20A-3) and the Strategic Young Researcher Overseas Visits Program for Accelerating Brain Circulation .
Conflict of interest
The authors acknowledge Diana Brinson for technical assistance.
Osteoarthritis: aging of matrix and cells – going for a remedy.
Potential involvement of oxidative stress in cartilage senescence and development of osteoarthritis: oxidative stress induces chondrocyte telomere instability and downregulation of chondrocyte function.