Increase of cell surface vimentin is associated with vimentin network disruption and subsequent stress-induced premature senescence in human chondrocytes
- Division for Biochemistry of Joint and Connective Tissue Diseases, Department of Orthopedics, University of Ulm, Germany
Figures
Figure 1
Association between chondrosenescence and the OARSI grade as well as mechanical injury.
Safranin-O staining of (a) macroscopic intact (OARSI grade ≤1) and (b) highly degenerated (OARSI grade ≥3) cartilage tissue; scale bar: 200 µm. Cell clusters are indicated by black arrows. (c) Gene expression of CDKN1A, CDKN2A, and SIRT1 in highly degenerated cartilage (grade ≥3) relative to macroscopic intact (grade ≤1) tissue (n≥5). (d–g) Human articular chondrocytes (hAC) were isolated from macroscopic intact (grade ≤1) and highly degenerated (grade ≥3) cartilage tissue of same donors (matched pairs). (d) Senescence-associated-β-galactosidase (SA-β-gal) activity assay and (e) flow cytometric analysis of cell surface vimentin (CSV) (f) including an exemplary histogram demonstrating the shift between cells derived from grade ≤1 (green curve) and grade ≥3 (red curve) tissue. (g) Exemplary images of anti-CSV staining (immunocytochemistry) on isolated hAC derived from grade ≤1 and grade ≥3 tissue; scale bar: 50 µm. (h) Corresponding Pearson correlation analysis between percentage of CSV-positive cells and SA-β-gal activity. (i) Safranin-O staining of macroscopic intact (OARSI grade ≤1) cartilage 7 days after ex vivo traumatization. (j–l) Gene expression of (j) CDKN1A, (k) CDKN2A, and (l) SIRT1 in cartilage tissue (OARSI grade ≤1) 24 hr, 7 days, and 14 days after impact (n≥6); normalized to unimpacted cartilage. (m) CSV-positivity of hAC isolated from impacted (Trauma) or unimpacted (Ctrl) cartilage explants from the same donor at day 7 and analyzed after short in vitro cultivation (7 days, passage 0). Data are presented as box plots with median, whiskers min to max or scatter plots; or column bars with mean, scanning electron microscopy (SEM). Statistics: (c) was analyzed by a multiple t test; (d), (e), and (m) were analyzed by means of paired two-tailed t test; in case of (j–l) Kruskal-Wallis with a Dunn’s post hoc test was used and (h) was analyzed by means of a Pearson correlation analysis. Each data point represents an independent biological replicate (n).
Figure 2
Doxorubicine (Doxo) stimulation leads to stress-induced premature senescence (SIPS) and enhanced cell surface vimentin (CSV)-positivity in human articular chondrocyte (hAC).
hAC were stimulated with 0.1 µM Doxo for 10 days. (a) mRNA levels of senescence-associated (SA) markers, including CDKN1A, CDKN2A, CXCL1, IGFBP7, IL1ß, IL6, IL8, and MMP13 (n=5). (b) Activity of senescence-associated-β-galactosidase (SA-β-gal) in untreated and Doxo-stimulated hAC (n=5). (c) Presence of CSV was determined by means of cytometric analysis; (d) exemplary histograms of untreated (green) and Doxo-stimulated (red) hAC. Exemplary distribution of (e) untreated (green) and (f) Doxo-stimulated (red) hAC during cytometric analysis and (g) corresponding statistical analysis (n=4). (h) Immunofluorescence staining of CSV [84-1] and vimentin [EPR3776] on the surface of unpermeabilized untreated and Doxo-stimulated hAC; white arrow tips indicate double-positive regions (appearing yellow); scale bar: 50 µm. (i) Exemplary scanning electron microscopy (SEM) image of Doxo-treated hAC; positive immunostaining of CSV appears in red. (j) Correlation analysis of VIM mRNA levels and CSV-positivity in hAC from the Doxo experiment, including untreated and Doxo-treated cells. (k) Immunofluorescence staining of the intracellular vimentin network in hAC 10 days after Doxo stimulation; yellow arrow tips indicate vimentin bundles; white arrow tips indicate (nearly) complete loss of vimentin; scale bar: 25 µm. Data are presented as scatter plot; box plots, median, whiskers min to max; or column bars with mean, SEM. Statistics: (a) and (g) were analyzed by a multiple t test; (b) and (c) were analyzed by means of paired two-tailed t test; (j) was analyzed by a Pearson correlation analysis. Each data point represents an independent biological replicate (n).
Figure 3
Influence of CDKN2A knockdown on senescence-associated (SA) markers and cell surface vimentin (CSV).
(a) Outline of the transfection regime. (b–d) siRNA-mediated knockdown of CDKN2A was confirmed by means of (b) quantitative real-time PCR (qRT-PCR) (n=8) and (c) by means of immunofluorescence; scale bar: 25 µm, (d) including quantification of the corrected total cell fluorescence (CTCF) values relative to siNC (n=4). (e) mRNA levels of SA markers, including CDKN1A, CXCL1, MMP13, IL6, and IL8 normalized on untransfected controls (n≥5). (f) Mitotic activity and proliferation, respectively, was determined by means of an alamarBlue assay performed at different time points (3 days, 5 days, 7 days). (g) Exemplary senescence-associated-β-galactosidase (SA-β-gal) staining and (h) corresponding quantification of the SA-β-gal activity in human articular chondrocyte (hAC) after transfection (n=6); scale bar: 50 µm. (i) CSV-positivity of hAC was assessed by means of cytometric analysis (n=8). (j) Correlation analysis of CDKN2A mRNA levels and CSV-positivity in hAC from different experiments. Data are presented as box plots with median, whiskers min to max; or column bars with mean, scanning electron microscopy (SEM). Statistics: (b), (d), (h), and (i) were analyzed by a paired two-tailed t test; (e) and (f) were analyzed by means of a multiple t test; (j) was analyzed by a Pearson correlation analysis. Each data point represents an independent biological replicate (n).
Figure 4
Cellular consequences of siRNA-mediated knockdown of vimentin in human articular chondrocyte (hAC).
(a–c) siRNA-mediated knockdown of VIM was confirmed by means of (a) quantitative real-time PCR (qRT-PCR) (n=8), (b) flow cytometry (n=4), and (d) exemplary co-staining of vimentin and actin cytoskeleton; white arrow tips indicate aberrant nuclei (smaller size and elliptical shape); scale bar: 50 µm. (d) Exemplary images of DCFDA assay and (e) corresponding quantification of the corrected total cell fluorescence (CTCF); scale bar: 50 µm. Senescent phenotype of hAC after knockdown of VIM was determined by (f) exemplary senescence-associated-β-galactosidase (SA-β-gal) staining and (g) quantification of the SA-β-gal activity, (h) mRNA levels of SA markers, including CDKN1A, CDKN2A, CXCL1, MMP13, IL6, and IL8 (all n≥5) and (i) exemplary staining of p53. Mitotic activity and proliferation, respectively, were determined by means of (j) an alamarBlue assay performed at different time points (3 days, 5 days, 7 days; n≥5) as well as (k) a flow cytometry-based cell cycle assay. (l) Flow cytometric analysis of cell surface vimentin (CSV)-positivity on transfected hAC; (m) exemplary histograms of hAC transfected with siNC (black) or siVIM (red); (n≥5). (n) Correlation analysis of VIM mRNA levels and CSV-positivity in hAC from the transfection experiment, including untransfected, siNC, and siVim cells. Data are presented as box plots with median, whiskers min to max; column bars with mean and scanning electron microscopy (SEM); or points and connecting line with mean and SEM. Statistics: (a), (b), (e), (g), and (l) were analyzed by a paired two-tailed t test; (h), (j), and (k) were analyzed by means of a multiple t test; (n) was analyzed by a Pearson correlation analysis. Each data point represents an independent biological replicate (n). Each data point represents an independent biological replicate (n). Abbreviations: siNC = cells transfected with scrambled siRNA; siVim = cells transfected with vimentin-targeting siRNA.
Figure 5
Consequences of simvastatin (Sim)-mediated collapse of the vimentin network in human articular chondrocyte (hAC).
(a) Immunofluorescence staining of the intracellular vimentin and actin network in hAC 20 hr after Sim treatment (10 µM). (b) Assessment of cell surface vimentin (CSV) on hAC treated with Sim by means of immunofluorescence staining and (c) corresponding quantification of CSV-positivity using flow cytometry; white arrow tips indicate aberrant nuclei (smaller size and elliptical shape); scale bar: 50 µm. (d) Fluorescence-based analysis of cytoplasmatic (DCFDA) and mitochondrial (MitoSOX) reactive oxygen species (ROS) in hAC; (e+f) including corresponding quantification of the corrected total cell fluorescence (CTCF). (g) Senescence-associated-β-galactosidase (SA-β-gal) staining and (h) activity assay in Sim-treated hAC. Data are presented as box plots with median, whiskers min to max; n=4 each. Statistics: (c), (e), (f), and (h) were analyzed by a paired two-tailed t test. Each data point represents an independent biological replicate (n).
Figure 6
Association between cell surface vimentin (CSV) and chondrocyte plasticity.
(a) Correlation analysis between passage number and percentage of CSV-positive cells (determined by means of flow cytometry). Analysis includes n=18 different donors, some donors were measured at various passages. (b) Exemplary images of immunocytochemistry staining against CSV on human articular chondrocyte (hAC) at low (P1) and high (P4) passage. (c) Correlation analysis between percentage of CSV-positive and CD105 (green; n=26) or CD166-positive (purple; n=16) hAC; surface markers were determined by means of flow cytometry. (d) hAC were cultured in basal, osteogenic (ODM; turquoise), or chondrogenic (CDM; pink) medium for 28 days. Afterward, one half of the cells were further cultured for 14 days in basal medium at sub-confluence (=ex-ODM and ex-CDM). Percentage of CSV-positive cells was determined via flow cytometry before differentiation (t0), directly after differentiation at day 28 (basal, ODM, CDM), or after further expansion at basal conditions at day 42 (ex-ODM, ex-CDM); n≥5. (e) Percentage of CSV-positive cells in isolated hAC (n=3 and n=6), bm-MSC (n=3), and chondrogenic stem progenitor cell (CSPC) (n=6) was determined by flow cytometry; bm-MSC and CSPC, respectively, were compared to passage-matching hAC. (f) Matrix calcification (Ca2+ deposition) was quantified via Alizarin red staining 28 days after osteogenic differentiation of hAC (n=7). Alizarin red concentrations were correlated with the percentage of CSV-positive cells in the corresponding donor, which was determined by flow cytometry. (g) Exemplary images of safranin-O-stained hAC pellets after chondrogenic differentiation for 28 days; images of three independent donors with increasing CSV-positivity (from left to right) are shown. Glycosaminoglycans appear in red and imply production of hyaline cartilage. Size of the pellet is considered as indicator of matrix production and successful chondrogenesis. Statistics: (a,c,f) were analyzed by a Pearson correlation analysis; (d,e) were analyzed by means of a one-way ANOVA, including a Sidak post hoc test.
Figure 7
Potential role of cell surface vimentin (CSV) in cell adhesion and outline of hypothesis.
(a) Relative amount of CSV-positivity in human articular chondrocyte (hAC) detached after 20 min of EDTA exposure (referred to as ‘early detached cells’=early) and hAC from the same culture plate, which required mechanically detachment using a cell scraper (referred to as ‘late detached cells’=late); n=4. (b) Adhesion strength in hAC with a relatively low percentage of CSV-positive cells (<14%; range: 3.5–13.7%) was compared to hAC with a relatively high percentage of CSV-positive cells (>16%; range: 16.2–35.4%) by means of a cell adhesion assay (each group n=4). (c) Correlation analysis between the percentage of CSV-positive cells and the percentage of remaining (adherent) cells on the culture plate after exposure to EDTA for 9 min of the corresponding donor (n=11). (d,e) Correlation analysis between the percentage of CSV-positive cells and the adherence capacity to different extracellular matrix (ECM) components of the corresponding donor (n=7); attachment of cells on different substrates was determined by means of a commercial ECM cell adhesion array and given as ‘absorbance at OD560’. (f) mRNA levels of integrin subunits, including ITGAV (αV), ITGA1 (α1), ITGA2 (α2), ITGA3 (α3), ITGA4 (α4), ITGA5 (α5), ITGB1 (β1), and ITGB2 (β2) (all n≥5). (g) Outline of our hypothesis: Loss of the chondrogenic phenotype is associated with alterations of the intracellular vimentin network and consequent expression of CSV. While mild stress or in vitro passaging promotes chondrocyte dedifferentiation, excessive stress and irreversible damage eventually results in cellular senescence and enhanced CSV levels. Statistics: (a) was analyzed by a paired two-tailed t test; (b,f) was analyzed by means of a multiple t test; (c,d,e) were analyzed by a Pearson correlation analysis.
Additional files
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Increase of cell surface vimentin is associated with vimentin network disruption and subsequent stress-induced premature senescence in human chondrocytes
eLife 12:e91453.
https://doi.org/10.7554/eLife.91453
