Purpose: Cellular senescence is a consequence of aging, which is associated with cell cycle arrest and has recently been identified as a major factor in the pathogenesis of osteoarthritis. Senescence could contribute to OA by several mechanisms including joint tissue increase in pro-inflammatory senescence-associated secretory proteins (SASPs). Our goal was to quantify the extent and characteristics of cellular senescence of human OA cartilage. To this end, we developed a short enzymatic digestion protocol for primary chondrocyte isolation and compared intact and damaged regions of cartilage within each patient to minimize potential confounding by cell stress that could induce senescence phenotypes during the cell isolation and manipulation procedures. We hypothesized that differences between damaged and intact regions of cartilage from each patient will more reliably identify disease associated senescence than evaluation of damaged regions alone. We characterized senescence based on β-galactosidase (β-gal) activity and p16 INK4a protein expression (two accepted markers for senescent cell identification), quantification of MMP3 protein expression (a SASP) and characterization of cell cycle phases. Methods: Human knee joints were obtained as surgical waste tissue under IRB approval from 4 OA patients during total knee arthroplasty. To isolate primary chondrocytes from macroscopically damaged vs intact regions, human articular cartilage from selected regions was finely diced and digested in 0.3% type II collagenase solution for three hours at 37 °C. Chondrocytes were isolated by centrifugation (300G) and directly stored as cell pellets in -80oC for β-gal activity (fluorimetric quantification) or fixed in 80% methanol and stored in -20oC for flow-cytometric analysis of p16 INK4a(ROCHE #9517 ), MMP3 (Thermo #701285) and cell cycle phase (by propidium iodide, Miltenyi Biotec). β-gal was assessed by SensoLyte ® MUG β - Galactosidase Assay Kit (Anaspec AS-72132) and normalized to DNA content (Biotium #31008). Results: Our protocol yielded a mean of 2.78 x 106 (range: 1.94-3.65x106) chondrocytes from 1 gram of human cartilage. Compared to intact cartilage, β-gal activity and p16INK4a protein expression were higher in damaged cartilage (Figure 1A, N=4, P=0.034; Figure 1B, N=4, P=0.031). The proportion of p16INK4a+/MMP3+ cells was 2.8-fold higher in damaged cartilage compared with intact cartilage (Figure 1C, N=4, P=0.082). Overall, β-gal activity was positively correlated with p16INK4aprotein expression. (Figure 1D, N=8, r=0.57, p=0.072). Compared with chondrocytes from intact cartilage, a higher proportion of chondrocytes from damaged cartilage were in the tetraploid G2 cell cycle phase (P=0.04) and a lower proportion were in the diploid G1 phase (P=0.02) (Figure 2A, N=4). Compared to diploid (G1 phase) chondrocytes,p16INK4a protein was much more highly expressed in tetraploid (G2 phase) chondrocytes (Figure 2B, N=8, P=0.001). Conclusions: We successfully isolated primary human chondrocytes using a short enzymatic digestion protocol to minimize phenotypic changes from cell manipulation, a stressor that might artifactually increase senescence cell markers in chondrocytes. β-gal activity was positively correlated with p16 INK4a expression and both markers were higher in damaged cartilage, which suggested an association of senescent chondrocytes with OA. Moreover, MMP3 expression was positively correlated with p16 INK4a expression. These results suggest at least one means, namely increased cartilage degradation by MMP3, by which senescent chondrocytes could contribute to the pathogenesis of OA. A previous study reported a higher percentage of tetraploid chondrocytes within damaged cartilage; we further identified increased p16INK4a expression in tetraploid chondrocytes. In conclusion, this study provides supportive evidence for the role of cellular senescence in the pathogenesis of human OA.View Large Image Figure ViewerDownload Hi-res image Download (PPT)
Read full abstract