Transcriptomic changes during the replicative senescence of human articular chondrocytes

doi:10.21203/rs.3.rs-4876006/v1

Download PDF

Research Article

Transcriptomic changes during the replicative senescence of human articular chondrocytes

https://doi.org/10.21203/rs.3.rs-4876006/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Objective

Aging is a major risk factor for OA, but the specific mechanisms underlying this connection remain unclear. Although chondrocytes rarely divide in adult articular cartilage, they undergo replicative senescence in vitro which provides an opportunity to study changes related to aging under controlled laboratory conditions.

Methods

Cartilage was recovered from two knees with OA and one normal knee. Chondrocyte cultures were established and sub-cultured until their Hayflick limit. Bulk RNA sequencing on early- and late-passage human articular chondrocytes allowed identification of transcriptomic changes associated with cellular aging.

Results

One male (80 years old) and one female (72 years old) patient provided OA cartilage. The donor of normal chondrocytes was a 26-year old male. Early passage chondrocytes from the two OA samples already had the phenotype of senescing cells, unlike normal chondrocytes. Nevertheless, all three chondrocyte cultures underwent 30 population doublings before replicative exhaustion, by which point all cells displayed a senescent phenotype. During this process, the cells lost their ability to form cartilaginous pellets. Differentially expressed genes (DEGs) analysis confirmed distinct transcriptomic profiles between early- and late-passage chondrocytes, as well as between cells isolated from normal and OA cartilage. Various changes in expression of genes related to cartilage matrix synthesis, degradation, inflammation and the senescence-associated secretory phenotype (SASP) were noted.

Conclusions

Although only a small pilot study, its data suggest that a larger and deeper study of the molecular and metabolic events accompanying the senescence of chondrocytes could provide important insights into the pathobiology of OA.

osteoarthritis

aging

chondrocyte replicative senescence

Hayflick limit

transcriptomics

Age is a major risk factor for osteoarthritis (OA), with post-menopausal females being disproportionately affected.¹ Although various different articular tissues undergo collaborative pathological changes in OA, much attention has focused on articular cartilage whose degeneration is a hallmark of OA. Multiple studies of articular cartilage have documented changes resulting from age and OA, including enhanced degradation of components of the extracellular matrix (ECM), particularly aggrecan, the major proteoglycan of cartilage, and declines in chondrocyte cell yields.^2,3 Such changes contribute to cartilage thinning, mechanical weakening and ultimately loss of cartilage function.⁴ Chondrocytes derived from older individuals exhibit characteristic features of cellular senescence, including shortened telomeres, heightened cell mortality, elevated senescence-associated β-galactosidase (SA-β-Gal) activity and increased expression of key regulators such as p53, p21^CIP1, and p16^INK4A.^5,6 Moreover, they produce increased amounts of reactive oxygen species (ROS) known to trigger the activation of genes associated with chondrocyte dedifferentiation and senescence.⁷

Senescent chondrocytes adopt a senescence-associated secretory phenotype (SASP).^8,9 This involves the secretion of pro-inflammatory cytokines¹⁰ including interleukins, oncostatin M (OSM), granulocyte macrophage-colony stimulating factor (GM-CSF), and tumor necrosis factor-alpha (TNFα), as well as matrix metalloproteinases (MMPs)^11,12 and other proteinases that degrade the ECM of cartilage, including members of the “a disintegrin and a metalloproteinase with thrombospondin motifs” (ADAMTS) family. There is also altered activity and expression of growth factors^13–16 including transforming growth factor-beta (TGF-β) and insulin-like growth factor-1 (IGF-1). Notably, there is evidence of elevation in SASP factor levels with age in OA patients.^17,18 These age-related alterations in cartilage physiology are likely contributors to the onset and progression of OA.

As well as undergoing aging in situ, chondrocytes undergo replicative senescence when serially passaged in vitro. The irreversible cell cycle arrest that occurs after a finite number of cell divisions in vitro was first described by Hayflick and Moorhead ^19,20. Primary human fibroblasts in culture underwent a progressive loss of mitotic capacity with repeated passaging, leading to complete mitotic arrest after approximately 30–40 population doublings. The inability of primary cells to proliferate beyond a finite number of doublings is now known as the "Hayflick limit". Evans and Georgescu²¹ first demonstrated that cultures of articular chondrocytes from rabbits, dogs, and humans undergo replicative senescence and suggested that chondrocyte senescence explains the association of OA with age. Because chondrocytes in adult cartilage rarely divide, even in OA, it was suggested that aging in situ reflected extended periods in the G₀ phase of the cell cycle.²²

In this pilot study, we studied chondrocytes from the knees of three individuals: two older individuals (72 and 80 years old) with OA, and one young adult (26 years old) without OA. Chondrocytes from all donors were serially passaged to replicative senescence²¹ and bulk RNA-sequencing was performed to identify passage-related transcriptomic changes.

Study Design

Chondrocytes recovered from knee articular cartilage were serially passaged until they reached their Hayflick limit. RNA was extracted from first passage and terminal passage cultures and subjected to bulk sequencing and bioinformatic analysis. The abundance of certain transcripts was confirmed by quantitative RT-PCR. The quality of cartilage formed in pellet culture was assessed at passage 1 and passage 15 using chondrocytes from the female donor.

Biospecimen Processing

Cartilage collection: Human osteoarthritic cartilage biopsies (n = 2, patient 1: male aged 80, patient 2: female aged 72) were collected from individuals undergoing total knee arthroplasty at Mayo Clinic, Department of Orthopedic Surgery with approval by the Institutional Review Board (IRB number: 13-005619).

Chondrocyte Isolation

Full-thickness pieces of articular cartilage were removed aseptically from the femoral condyles and tibial plateaux using a scalpel. The cartilage was cut into 1–2 mm³ pieces and washed with phosphate buffered saline (PBS), (Gibco, Gaithersburg, MD, USA). The cartilage fragments were digested for 2 hours at 37°C in 0.2% pronase (Millipore Sigma, St. Louis, MO, USA) in Dulbecco's Modified Eagle Medium/Nutrient Mixture F-12 (DMEM/F-12; Sigma-Aldrich, St. Louis, MO, USA) with a mixture of penicillin (100 U/ml) and streptomycin (100 µg/ml) (1% P/S, Gibco). After pronase digestion, the cartilage fragments were further digested for 16 hours at 37°C in 0.05% type II collagenase (Gibco) in DMEM/F-12 media supplemented with 5% fetal bovine serum (FBS; Gibco) and 1% P/S. The digest was filtered through a 70 µm cell strainer and cells recovered and washed. Cells were counted, seeded into 75 cm² culture flasks at a density of 1.0 x 10⁴ cells/cm² in 10 mL of DMEM/F12 supplemented with 1% P/S and 10% FBS, and incubated at 37^o in 5% CO₂. Human articular chondrocytes from a normal knee joint were purchased from Lonza (Basel, Switzerland), (n = 1, male aged 26).

Serial passage of chondrocytes

Cells were recovered from confluent cultures by trypsinization and re-seeded into 75 cm² flasks at a 1:4 split ratio such that each passage produced 2 population doublings. Briefly, cells were treated with 4 mL of TrypLE™ (Thermo Fisher Scientific, Waltham, MA, USA) and incubated at 37^o in 5% CO₂ for 5 minutes to detach the cells. Subsequently, 8 mL of medium containing 10% FBS were added, followed by centrifugation at 400 g for 5 minutes. The supernatant was discarded, and the pellet reconstituted in 1 mL of medium. Cells were counted and 250 µL of the cell suspension seeded into a 75 cm² flask with 10 mL medium. This process was repeated until cells could no longer achieve confluence, at which point they were considered to have reached their Hayflick limit. Doubling time was calculated from the following formula: Doubling time=[Duration⋅ln(2)] / [ln(Final concentration/Initial concentration)].²³

3D Pellet culture

First passage or terminal passage chondrocytes from the female donor were recovered and re-cultured as pellets as previously described.²⁴ Briefly, cells were plated at a density of 2 × 10⁵ cells per well in polypropylene, v-bottom, 96-well plates (Corning, Corning, NY, USA). Plates were centrifuged at 400 g for 5 minutes; supernatants were removed and replaced with chondrogenic medium (high glucose DMEM (Sigma-Aldrich), 1% insulin-transferrin-selenium premix (BD Biosciences, Franklin Lakes, NJ, USA), 100 nM dexamethasone (Sigma-Aldrich), 50 µg/mL ascorbate-2-phosphate (Sigma-Aldrich), 1 mM sodium pyruvate (Sigma-Aldrich), 40 µg/mL L-proline (Sigma-Aldrich), and 10 ng/mL TGFβ1 (PeproTech, Cranbury, NJ, USA). During the initial 24 hours of incubation, cells formed free-floating aggregates. Media were changed the following morning and every other day thereafter.

Histology

Pellet cultures

Pellets were fixed for 1 hour in 4% paraformaldehyde, centrifuged at 400 rpm for 5 minutes, washed with 95% alcohol and resuspended in 100 µL of 0.8% low melting point agarose solution (Invitrogen Carlsbad, CA, USA), followed by re-centrifugation at 1500 rpm for 5 minutes. The resulting agarose-embedded cell pellet was allowed to solidify for 30 minutes at 4°C, then processed as per routine histology; 5-µm-thick sections were cut using an automatic microtome (HM 355S, Thermo Fischer Scientific) and mounted onto positively charged slides (Superfrost™ Plus Microscope Slides, Thermo Fisher Scientific).

Toluidine blue staining

Sections were deparaffinized and rehydrated through a series of xylenes and graded alcohols to distilled water, followed by staining with 0.1% toluidine blue pH 2.3 for 1.5 minutes, rinsing in distilled water for 1 minute, then quickly dehydrating in a graded ethyl alcohol series, cleared with xylene, and mounted with xylene-based mounting medium (Richard-Allan Scientific™, Kalamazoo, MI, USA). Bright field images were acquired using an automated inverted microscope (Olympus IX83) at 4X and 20X magnifications using the cellSens Olympus imaging software.

Bulk RNA sequencing and data analysis

RNA isolation

A two-step RNA isolation protocol was followed. Organic phase isolation was performed with TriReagent®/ bromo–3–chloropropane (BCP) (Thermo Fisher Scientific/Molecular Research Center, Inc., Cincinnati, OH, USA) and then used for total RNA isolation using a RNeasy Plus Mini Kit (Qiagen, Germantown, MD, USA) according to the manufacturer’s instructions. RNA concentration was measured using a nanophotometer® (Implen, Inc., Westlake Village, CA, USA). RNA purity was confirmed spectrophotometry (A_260/280 and A_260/230 ˃ 2.0). RNA integrity was assessed using an Agilent 2100 bioanalyzer (Agilent, Santa Clara, CA, USA) to confirm that samples had a RNA integrity number (RIN) ˃ 8.

Illumina Stranded mRNA library preparation and Illlumina NextSeq2000 Sequencing

RNA libraries were prepared using 200 ng of total RNA in a volume of 20 µL according to the manufacturer’s instructions for the Illumina Stranded mRNA Ligation Sample Prep Kit (Illumina, San Diego, CA). The concentration and size distribution of the completed libraries were determined using an Agilent Bioanalyzer DNA 1000 chip and Qubit fluorometry (Invitrogen). Libraries were sequenced using Illumina’s standard protocol for the NovaSeq 6000. The NovaSeq SP PE100 flow cell was sequenced as 100 X 2 paired end reads using NovaSeq Control Software v1.8.0. Base-calling was performed using Illumina’s RTA version 3.4.4.

Bioinformatics and data analysis: Fastq files for all samples were processed using the Mayo Clinic RNA-Seq bioinformatics pipeline, MAP-RSeq version 3.1.4.²⁵ Gene and exon expression quantification was performed using the Subread²⁶ package to obtain both raw gene counts and normalized Fragments Per Kilobase per Million mapped reads (FPKM) values. Finally, FASTQC²⁷ and MultiQC²⁸ tools were used for comprehensive quality control assessment of the aligned reads. Principal component analysis (PCA) and hierarchical clustering were performed for quality control, exploratory data analysis, and to identify sample outliers. Using the raw gene counts generated from the MAP-RSeq pipeline, genes with significant differential expression between different samples were assessed using edgeR.²⁹ Genes with counts per million ˂ 0.5 were filtered out prior to differentially expressed genes (DEGs) analysis. Three comparisons were performed including i) Late passage vs. early passage chondrocytes from male with OA; ii) Late passage vs early passage chondrocytes from female with OA and iii) Late passage vs. early passage chondrocytes from the healthy male. DEGs were visually represented using volcano plots and heatmaps generated with the R package ggplot2. DEGs were reported along with their magnitude of change (log2 fold change ≥ 1 or ≤ -1) and level of significance (False Discovery Rate, FDR < 0.05). Pathway analysis was conducted with R package enrichR³⁰ or Metascape³¹ (https://metascape.org) using human pathway settings.

Quantitative Real Time Polymerase Chain Reaction (qRT-PCR)

Total RNA was isolated from samples snap-frozen in liquid nitrogen and homogenized in an Eppendorf tube using 3-mm Tungsten carbide beads and a TissueLyser II system (Qiagen) set to 4 minutes at 30 Hz; 1 µg of cDNA was synthesized using a high-capacity cDNA reverse transcription kit (Applied Biosystems, Waltham, MA, USA). PCR experiments were performed using 20 ng of cDNA per reaction, qRT-PCR Brilliant III ultra-fast master mix kit (Agilent) and TaqMan gene expression probes (Thermo Fisher Scientific) as follows: IL-1α (Hs00174092_m1), IL-1β (Hs01555410_m1), IL-6 (Hs00174131_m1), p16^INK4A (Hs00923894_m1), p21^CIP1 (Hs00355782_m1), p53 (Hs01034249_m1). The PCR assay was conducted using the AriaMx qRT-PCR detection system (Agilent). Gene expression values were normalized to the GAPDH housekeeping gene to calculate the Δ^Ct value.

Statistical Analysis

Statistical analyses for qRT-PCR used GraphPad Prism 9.2.0 (GraphPad, Boston, MA, USA). The experimental design involved 3 biological replicates (n = 3). Data distribution was assessed using either the Kolmogorov–Smirnov test or Shapiro–Wilk tests. A two tailed unpaired t-test or two-way ANOVA was employed for normally distributed data, while the Mann–Whitney U-test was utilized for non-normally distributed data. The results were presented as mean ± standard deviation (SD). Statistical significance was set at a p-value < 0.05.

Replicative senescence of chondrocytes

Chondrocytes from all three subjects reached their Hayflick limit after 30 doublings (15 passages). Early passage chondrocytes from osteoarthritic donors displayed various cell morphologies, including stellate, elongate and fibroblastic (Fig. 1A). Early passage cultures of chondrocytes from the healthy young donor were smaller and comprised a mixture of polygonal and rounded cells. Terminal passage cells were much larger and displayed a mixture of stellate and flattened phenotypes. Passage 7 cells consist of a blend of both early and late passage cell types. Although chondrocytes from the healthy donor exhibited morphological differences from those of individuals with OA during early and middle passages, in late passage morphologies were similar in cells from OA and healthy donors. The doubling time of chondrocytes from OA and healthy donors was similar from passage 1 to passage 15 (Fig. 1B). Following passage 13, a significant and abrupt increase in doubling time was observed in all cultures regardless of sex or whether the chondrocytes were from normal joints or joints with OA. After reaching passage 15, the cells were left in a T175 flask for a year, during which neither proliferation nor cell death was observed.

Early and late passage chondrocytes showed clear differences in their ability to form cartilaginous nodules in pellet culture (Fig. 1C). Pellets formed from early passage chondrocytes were larger and displayed the characteristic white, shiny appearance of cartilage. Those formed from late passage cells were smaller with a slight yellow discoloration. The ECM of pellets formed from late passage chondrocytes lacked the metachromatic staining of pellets formed by early passage cells (Fig. 1D).

DEGs between early and late passage chondrocytes

Transcriptomic profiling was performed on early and late passage chondrocyte cultures from all three donors. PCA of all genes showed clear separation between the late passage and early passage chondrocytes, indicating substantial differences in gene expression (data not shown). DEGs in early and late passage cells were then analyzed for each of the three chondrocyte cultures.

Sample 1 – Early and late passage chondrocytes from male with OA

Distinct clusters were observed when comparing the transcriptomes of early and late passage chondrocytes obtained from an 80-year old male with OA (Fig. 2A), confirmed by a heat of gene expression patterns (Fig. 2B). A volcano plot revealed the most significantly altered genes, with Endothelial cell Specific Molecule 1 (ESM1) exhibiting the greatest upregulation and Faciogenital Dysplasia 5 (FGD5) showing the greatest downregulation (Fig. 2C). The analysis identified 2,309 DEGs, with 983 genes upregulated and 1,326 downregulated in terminal passage chondrocytes (Fig. 2D). Pathway analysis revealed downregulation of ECM organization, collagen fibril organization, glycosaminoglycan catabolic process, skeletal and nervous system development, and upregulation of focal adhesion, axon guidance and, paradoxically, DNA replication genes in late passage cells (Fig. 2E).

Sample 2 – Early and late passage chondrocytes from female with OA

Distinct clusters were also observed when comparing the transcriptomes of early and late passage chondrocytes obtained from a 72-year-old female with OA (Fig. 3A). This sample displayed a comparable heatmap pattern to that of sample 1 (Fig. 3B). The largest gene expression changes were the upregulation of tissue-type plasminogen activator (PLAT) and the downregulation of Solute Carrier Family 7 Member 2 (SLC7A2) (Fig. 3C). A total of 3,281 DEGs were identified when comparing early and late passage chondrocytes (Fig. 3D) with 1,577 genes upregulated and 1,704 genes downregulated. Similar to sample 1, replicative senescence was associated with the downregulation of genes related to ECM organization and collagen, with upregulation of genes associated with neuron development and, paradoxically, the cell cycle and DNA replication (Fig. 3E).

Sample 3 - Early vs late passage chondrocytes from a male without OA

Hierarchical clustering and heatmap patterns differed from those of samples 1 and 2 (Fig. 4A, B). The volcano plot highlighted SLC7A2 as the most downregulated gene, while Chloride Channel Accessory 2 (CLCA2) appeared as the most upregulated gene (Fig. 4C). Fewer DEGs (1,298) were identified in this sample than samples 1 and 2 (Fig. 4D). In contrast to the data from the OA samples, cell cycle and DNA replication genes were downregulated in late passage cells, while prostaglandin genes were upregulated (Fig. 4E).

Analysis of ECM components, enzymes that degrade ECM and SASP factors

A focused analysis was undertaken on components of the ECM, the enzymes that degrade them and SASP factors within each sample (Fig. 5). Early passage cells from all three donors expressed high levels of Col1A1 as well as cartilage specific Col2A1 and ACAN. The latter two transcripts fell to extremely low levels in all terminal passage cultures, regardless of donor. Expression of collagen type I α1 chain (Col1A1) was more variable, with expression levels dropping to extremely low levels in terminal passage cells from the healthy donor, but with expression levels reduced to a lesser degree in cells from osteoarthritic joints (Fig. 5A).

Of the enzymes that degrade the ECM of cartilage, expression of MMP19 was increased in the late passage chondrocytes of all samples. Expression of ADAMTS4 and ADAMTS8 showed significant upregulation in late passage chondrocytes obtained from OA samples, but not in those obtained from healthy cartilage.

Expression of transcripts associated with the SASP (p16^INK4A, p21^CIP1, p53,) was variable. Expression of p16^INK4A did not differ between early and late passage cells derived from normal cartilage. Expression in chondrocytes derived from joints with OA was lower than that in chondrocytes from young, healthy cartilage but increased considerably in late passage cells. Expression of p21^CIP1 showed the opposite trend, being higher in early passage chondrocytes derived from joints with OA than in early passage cells from young, healthy cartilage, but lower in late passage OA samples than in late passage samples from young, healthy joints. Expression of p53 decreased with passage to varying degrees in all cultures (Fig. 5B).

Among the cytokine components of SASP, IL-1α expression increased in all cultures with passage. Expression of IL-1β did not increase with passage of chondrocytes derived from young, healthy joints but increased in chondrocytes derived from OA cartilage. Whereas IL-6 expression was roughly similar in early passage cells of all sources, it increased considerably with passage of chondrocytes derived from joints with OA, but fell in chondrocytes derived from healthy joints. IL-7 expression increased in all late passage samples. TNFα expression was barely detectable in any culture. C-C motif chemokine ligand 2 (CCL2) was upregulated in the late passage of chondrocytes derived from patients with OA, but not those derived from healthy cartilage.

Pathway analysis comparing chondrocytes from male and female with OA

Pathway analysis was conducted to compare the DEGs profiles between chondrocytes derived from the male and female samples with OA, utilizing data from samples 1 and 2. This analysis revealed 611 genes unique to males and 1,583 genes unique to females, with 1,698 genes common between the sexes (Fig. 6A). The top 20 pathways common between males and females with OA included ECM organization, collagen formation, musculoskeletal development and nervous system development (Fig. 6B). The top 20 pathways unique to males with OA encompassed ECM organization, mitotic cell cycle, and regulation of neuron differentiation (Fig. 6C). The top 20 pathways unique to females with OA featured in regulation of cell cycle process, DNA metabolic process, and PID-PLK1 pathway (Fig. 6D).

Quantitative RT-PCR

Certain findings from bulk RNA sequencing were validated by qRT-PCR (Fig. 7). Consistent with the RNA sequencing data, significant upregulation of p16^INK4A, IL-1α, IL-1β, and IL-6 was observed in late passage chondrocytes (Fig. 7).

In this study we demonstrated that serial passaging of chondrocytes to replicative senescence induced characteristic morphological changes and reduced their chondrogenic potential. Chondrocytes reached replicative senescence at passage 15 after 30 population doublings, which is consistent with earlier data ²⁰. It is remarkable that chondrocytes derived from old (72 years, 80 years) and young (26 years) donors had the same Hayflick limit, as diploid cells from young individuals typically have a greater mitotic potential than those from old individuals. Indeed, the pattern of replicative senescence was remarkably similar between the three cultures (Fig. 1B). It is possible that the quiescent state of articular chondrocytes in vivo preserves their mitotic potential as the person ages chronologically. Consistent with this, Rose et al. ³² demonstrated that although DNA damage occurs in osteoarthritic chondrocytes, there is no evidence of telomere shortening. Nevertheless, when plated into monolayer culture, chondrocytes from the two old donors already had the morphology of late passage cells. Because this is a small pilot study involving only 3 donors it is not possible to draw strong, general conclusions, but these observations are nevertheless intriguing. We also showed that late passage chondrocytes have a reduced capacity to form cartilaginous tissues, as evidenced by smaller pellet sizes and decreased proteoglycan content. This confirms that replicative senescence impairs the chondrogenic potential of cultured chondrocytes.

Bulk RNA sequencing revealed clear separation between early and late passage cells, highlighting substantial differences in gene expression. In several cases the changes in gene expression were common to all cultures. The uniform decline in expression of type II collagen and aggrecan core protein with passage, for instance, confirms loss of the chondrocyte phenotype as the cells undergo replicative senescence. Other changes common to all cultures included the upregulation of IL-1α, IL-7, and MMP19 with passage, as well as the near absence of TNF-α in any culture. It is intriguing, however, that the changes in gene expression as a result of replicative senescence were not identical between chondrocytes recovered from normal joints and those recovered from joints with OA. Indeed, changes with passage in the expression of several genes, including p16^INK4A, p21^CIP1, IL-1β, IL-6, ADAMTS4 and ADMTS8, while qualitatively similar between the two samples from OA cartilage, differed from changes with passage in cells derived from healthy cartilage (Figs. 6, 7). Moreover, pathway analysis revealed that expression of genes associated with DNA replication were paradoxically upregulated in late passage chondrocytes derived from OA cartilage but, as might be expected, downregulated in late passage cells derived from healthy cartilage. Collectively, these data are consistent with the induction of epigenetic changes occurring in situ as a result of aging and OA that survive passaging.

The preliminary nature of this study does not justify a deep analysis of the totality of the bulk RNA seq data, but several findings are worth mentioning. For instance, volcano plots revealed that FGD5 was the most downregulated gene in late passage OA chondrocytes. While the precise involvement of FGD5 in OA development is not known, Yang et al. ³³ demonstrated that FGD5-AS1 regulates OA progression via the miR-302d-3p/TGFBR2 axis, protecting chondrocytes from inflammation-induced injury and reducing the ECM degradation. Phospholipase A2 Group IIA (PLA2G2A) expression was also downregulated in late passages of both male and female OA chondrocytes. In agreement with this finding, Tsolis et al. ³⁴ demonstrated that PLA2G2A exhibits higher expression in healthy chondrocytes compared to chondrocytes from OA cartilage. Likewise, adenylyl cyclase type 2 (ADCY2) was also downregulated in late passage cultures derived from both male and female subjects with OA. Notably, ADCY2 was undetectable in healthy chondrocytes ³².

Concerning possible sex-differences in gene expression with replicative senescence, DIO2 was downregulated in late passage, male OA chondrocytes. This gene is recognized as one of several OA susceptibility genes that play a crucial role in pathways involved in both pre- and post-natal joint development, ultimately contributing to endochondral ossification.³⁵ Similarly, genes located in the male-specific region of the Y chromosome, namely DDX3Y, USP9Y and NLGN4Y, were found to be significantly downregulated in a late passage, male OA sample, an observation consistent with the data of Liu et al. ³⁶ Also of possible relevance to sex-related differences in the incidence of OA, PLAT, an estrogen regulated gene (Thakkar et al.³⁷), was upregulated in the late passage, female OA sample.

Basic pathway analyses of late passages were performed to identify sex specific DEGs in OA. The top 20 pathways common to both males and females with OA were generally related to ECM organization, collagen formation, musculoskeletal development and nervous system development. However, females (1583 genes) showed distinct pathways compared to males (611 genes), including regulation of cell cycle processes, DNA metabolism, and the PLK1 signaling pathway. PLK1 is a key regulator of mitotic cell division and has been shown to be overexpressed in several human cancers, including breast cancer. ³⁸ Driscoll et al. ³⁹ showed that inhibiting PLK1 causes post-mitotic DNA damage and senescence in tumor cell lines. Kim et al. ⁴⁰ found that decreasing PLK1 levels stimulates senescence through a p53-dependent pathway. Additionally, Wierer et al. ⁴¹ showed that PLK1 mediates estrogen receptor-regulated genes in human breast cancer cells suggesting PLK1 could play a role in the sex differences seen in the incidence of OA.

Small sample size is the main limitation of this pilot study. Additional studies with more subjects from both sexes are important to fully characterize chondrocyte senescence and elucidate sex differences in relation to its pathophysiologic role in OA. The limited availability of normal cartilage specimens is a major obstacle to achieving this expeditiously. Another consideration is collecting samples from individuals with idiopathic, rather than post-traumatic, OA where chondrocyte senescence may be the key underlying pathophysiologic process. Overall, while preliminary, these data provide insights into transcriptomic changes underlying chondrocyte aging and OA development. They also question whether replicative senescence of chondrocytes in vitro is an appropriate surrogate for in situ chondrocyte senescence when studying its relationship to the etiopathophysiology of OA.

In this small (n = 3) pilot study, distinct DEG differences were observed between early and late passage chondrocytes, as well as between males and females and those with or without OA. These findings indicate possible important age- and sex-related differences in gene expression associated with OA. The upregulation of IL-1α, IL-1β, IL-6, IL-7 and p16^INK4a, in late passage chondrocytes from OA samples highlights the link between cellular senescence and inflammation. This was associated with significant downregulation of key ECM genes encoding Col2A1 and aggrecan core protein, along with upregulation of matrix-degrading enzymes in late passage chondrocytes derived from joints with OA, consistent with a pivotal role of chondrocyte senescence in OA pathogenesis. Additionally, the sample derived from a female patient exhibited DEG patterns that differed from those of samples derived from males, although more samples are needed to confirm a sex-related difference in DEG. In several instances transcriptomic changes with passage of chondrocytes from joints with OA were different from those occurring with passage of normal chondrocytes.

Acknowledgements

RNA sequencing was conducted in collaboration with the Mayo Clinic Genome Analysis Core. Bioinformatics support was provided by the Department of Quantitative Health Sciences. Graphical abstract was created with BioRender.com.

Author Contributions

AAZ; study design, data analysis and interpretation, writing original draft. GPH; data acquisition. CVN: review & editing. CLDP: histology, review & editing. MPA: resources, review & editing. CHE: Conceptualization, investigation, review & editing, supervision. All authors have read and agreed to the published version of the manuscript.

Funding

AAZ received support from the Building Interdisciplinary Research Careers in Women's Health (BIRCWH) program through grant number K12AR084222. CHE’s research is supported, in part, by the John and Posy Krehbiel Professorship in Orthopedics.

Conflict of Interest Statement

The authors declare no conflicts of interest.

Data availability

All computational analysis pipelines can be found on NCBI Gene Expression Omnibus (GEO) (https://www.ncbi.nlm.nih.gov/geo/) under accession numbers GSE246425.

Hernandez, P.A., Bradford, J.C., Brahmachary, P., Ulman, S., Robinson, J.L., June, R.K., and Cucchiarini, M. (2024). Unraveling sex-specific risks of knee osteoarthritis before menopause: Do sex differences start early in life? Osteoarthritis Cartilage. 10.1016/j.joca.2024.04.015.
Toh, W.S., Brittberg, M., Farr, J., Foldager, C.B., Gomoll, A.H., Hui, J.H., Richardson, J.B., Roberts, S., and Spector, M. (2016). Cellular senescence in aging and osteoarthritis. Acta Orthop 87, 6-14. 10.1080/17453674.2016.1235087.
Loeser, R.F. (2011). Aging and osteoarthritis. Curr Opin Rheumatol 23, 492-496. 10.1097/BOR.0b013e3283494005.
Temple, M.M., Bae, W.C., Chen, M.Q., Lotz, M., Amiel, D., Coutts, R.D., and Sah, R.L. (2007). Age- and site-associated biomechanical weakening of human articular cartilage of the femoral condyle. Osteoarthritis Cartilage 15, 1042-1052. 10.1016/j.joca.2007.03.005.
Philipot, D., Guérit, D., Platano, D., Chuchana, P., Olivotto, E., Espinoza, F., Dorandeu, A., Pers, Y.M., Piette, J., Borzi, R.M., Jorgensen, C., et al. (2014). p16INK4a and its regulator miR-24 link senescence and chondrocyte terminal differentiation-associated matrix remodeling in osteoarthritis. Arthritis Res Ther 16, R58. 10.1186/ar4494.
Ashraf, S., Cha, B.H., Kim, J.S., Ahn, J., Han, I., Park, H., and Lee, S.H. (2016). Regulation of senescence associated signaling mechanisms in chondrocytes for cartilage tissue regeneration. Osteoarthritis Cartilage 24, 196-205. 10.1016/j.joca.2015.07.008.
Hui, W., Young, D.A., Rowan, A.D., Xu, X., Cawston, T.E., and Proctor, C.J. (2016). Oxidative changes and signalling pathways are pivotal in initiating age-related changes in articular cartilage. Ann Rheum Dis 75, 449-458. 10.1136/annrheumdis-2014-206295.
Millerand, M., Berenbaum, F., and Jacques, C. (2019). Danger signals and inflammaging in osteoarthritis. Clin Exp Rheumatol 37 Suppl 120, 48-56.
Loeser, R.F., Collins, J.A., and Diekman, B.O. (2016). Ageing and the pathogenesis of osteoarthritis. Nat Rev Rheumatol 12, 412-420. 10.1038/nrrheum.2016.65.
Coryell, P.R., Diekman, B.O., and Loeser, R.F. (2021). Mechanisms and therapeutic implications of cellular senescence in osteoarthritis. Nat Rev Rheumatol 17, 47-57. 10.1038/s41584-020-00533-7.
Forsyth, C.B., Cole, A., Murphy, G., Bienias, J.L., Im, H.J., and Loeser, R.F., Jr. (2005). Increased matrix metalloproteinase-13 production with aging by human articular chondrocytes in response to catabolic stimuli. J Gerontol A Biol Sci Med Sci 60, 1118-1124. 10.1093/gerona/60.9.1118.
Loeser, R.F., Goldring, S.R., Scanzello, C.R., and Goldring, M.B. (2012). Osteoarthritis: a disease of the joint as an organ. Arthritis Rheum 64, 1697-1707. 10.1002/art.34453.
van der Kraan, P.M., Blaney Davidson, E.N., and van den Berg, W.B. (2010). A role for age-related changes in TGFbeta signaling in aberrant chondrocyte differentiation and osteoarthritis. Arthritis Res Ther 12, 201. 10.1186/ar2896.
Blaney Davidson, E.N., Remst, D.F., Vitters, E.L., van Beuningen, H.M., Blom, A.B., Goumans, M.J., van den Berg, W.B., and van der Kraan, P.M. (2009). Increase in ALK1/ALK5 ratio as a cause for elevated MMP-13 expression in osteoarthritis in humans and mice. J Immunol 182, 7937-7945. 10.4049/jimmunol.0803991.
Loeser, R.F., Gandhi, U., Long, D.L., Yin, W., and Chubinskaya, S. (2014). Aging and oxidative stress reduce the response of human articular chondrocytes to insulin-like growth factor 1 and osteogenic protein 1. Arthritis Rheumatol 66, 2201-2209. 10.1002/art.38641.
Wei, F.Y., Lee, J.K., Wei, L., Qu, F., and Zhang, J.Z. (2017). Correlation of insulin-like growth factor 1 and osteoarthritic cartilage degradation: a spontaneous osteoarthritis in guinea-pig. Eur Rev Med Pharmacol Sci 21, 4493-4500.
Wakale, S., Wu, X., Sonar, Y., Sun, A., Fan, X., Crawford, R., and Prasadam, I. (2023). How are Aging and Osteoarthritis Related? Aging Dis 14, 592-604. 10.14336/ad.2022.0831.
Motta, F., Barone, E., Sica, A., and Selmi, C. (2023). Inflammaging and Osteoarthritis. Clin Rev Allergy Immunol 64, 222-238. 10.1007/s12016-022-08941-1.
Hayflick, L., and Moorhead, P.S. (1961). The serial cultivation of human diploid cell strains. Exp Cell Res 25, 585-621. 10.1016/0014-4827(61)90192-6.
Hayflick, L. (1965). THE LIMITED IN VITRO LIFETIME OF HUMAN DIPLOID CELL STRAINS. Exp Cell Res 37, 614-636. 10.1016/0014-4827(65)90211-9.
Evans, C.H., and Georgescu, H.I. (1983). Observations on the senescence of cells derived from articular cartilage. Mech Ageing Dev 22, 179-191. 10.1016/0047-6374(83)90111-2.
CH, E. (1979). Is the A (Go) state time independent? J Theor Biol 79, 259-262.
Hirsch, H.R., and Engelberg, J. (1965). Determination of the cell doubling-time distribution from culture growth-rate data. J Theor Biol 9, 297-302. 10.1016/0022-5193(65)90114-1.
De la Vega, R.E., Scheu, M., Brown, L.A., Evans, C.H., Ferreira, E., and Porter, R.M. (2019). Specific, Sensitive, and Stable Reporting of Human Mesenchymal Stromal Cell Chondrogenesis. Tissue Eng Part C Methods 25, 176-190. 10.1089/ten.TEC.2018.0295.
Kalari, K.R., Nair, A.A., Bhavsar, J.D., O'Brien, D.R., Davila, J.I., Bockol, M.A., Nie, J., Tang, X., Baheti, S., Doughty, J.B., Middha, S., et al. (2014). MAP-RSeq: Mayo Analysis Pipeline for RNA sequencing. BMC Bioinformatics 15, 224. 10.1186/1471-2105-15-224.
Liao, Y., Smyth, G.K., and Shi, W. (2013). The Subread aligner: fast, accurate and scalable read mapping by seed-and-vote. Nucleic Acids Res 41, e108. 10.1093/nar/gkt214.
Andrews, S. (2010). FastQC: A Quality Control Tool for High Throughput Sequence Data [Online].
Ewels, P., Magnusson, M., Lundin, S., and Käller, M. (2016). MultiQC: summarize analysis results for multiple tools and samples in a single report. Bioinformatics 32, 3047-3048. 10.1093/bioinformatics/btw354.
Robinson, M.D., McCarthy, D.J., and Smyth, G.K. (2010). edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 26, 139-140. 10.1093/bioinformatics/btp616.
Xie, Z., Bailey, A., Kuleshov, M.V., Clarke, D.J.B., Evangelista, J.E., Jenkins, S.L., Lachmann, A., Wojciechowicz, M.L., Kropiwnicki, E., Jagodnik, K.M., Jeon, M., et al. (2021). Gene Set Knowledge Discovery with Enrichr. Curr Protoc 1, e90. 10.1002/cpz1.90.
Zhou, Y., Zhou, B., Pache, L., Chang, M., Khodabakhshi, A.H., Tanaseichuk, O., Benner, C., and Chanda, S.K. (2019). Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat Commun 10, 1523. 10.1038/s41467-019-09234-6.
Rose, J., Söder, S., Skhirtladze, C., Schmitz, N., Gebhard, P.M., Sesselmann, S., and Aigner, T. (2012). DNA damage, discoordinated gene expression and cellular senescence in osteoarthritic chondrocytes. Osteoarthritis Cartilage 20, 1020-1028. 10.1016/j.joca.2012.05.009.
Yang, Y., Sun, Z., Liu, F., Bai, Y., and Wu, F. (2021). FGD5-AS1 Inhibits Osteoarthritis Development by Modulating miR-302d-3p/TGFBR2 Axis. Cartilage 13, 1412s-1420s. 10.1177/19476035211003324.
Tsolis, K.C., Bei, E.S., Papathanasiou, I., Kostopoulou, F., Gkretsi, V., Kalantzaki, K., Malizos, K., Zervakis, M., Tsezou, A., and Economou, A. (2015). Comparative proteomic analysis of hypertrophic chondrocytes in osteoarthritis. Clin Proteomics 12, 12. 10.1186/s12014-015-9085-6.
Bomer, N., den Hollander, W., Ramos, Y.F., Bos, S.D., van der Breggen, R., Lakenberg, N., Pepers, B.A., van Eeden, A.E., Darvishan, A., Tobi, E.W., Duijnisveld, B.J., et al. (2015). Underlying molecular mechanisms of DIO2 susceptibility in symptomatic osteoarthritis. Ann Rheum Dis 74, 1571-1579. 10.1136/annrheumdis-2013-204739.
Liu, W., Chen, Y., Zeng, G., Yang, S., Yang, T., Ma, M., and Song, W. (2021). Single-Cell Profiles of Age-Related Osteoarthritis Uncover Underlying Heterogeneity Associated With Disease Progression. Front Mol Biosci 8, 748360. 10.3389/fmolb.2021.748360.
Thakkar, A.D., Raj, H., Chakrabarti, D., Ravishankar, Saravanan, N., Muthuvelan, B., Balakrishnan, A., and Padigaru, M. (2010). Identification of gene expression signature in estrogen receptor positive breast carcinoma. Biomark Cancer 2, 1-15. 10.4137/bic.S3793.
Bousset, L., and Gil, J. (2022). Targeting senescence as an anticancer therapy. Mol Oncol 16, 3855-3880. 10.1002/1878-0261.13312.
Driscoll, D.L., Chakravarty, A., Bowman, D., Shinde, V., Lasky, K., Shi, J., Vos, T., Stringer, B., Amidon, B., D'Amore, N., and Hyer, M.L. (2014). Plk1 inhibition causes post-mitotic DNA damage and senescence in a range of human tumor cell lines. PLoS One 9, e111060. 10.1371/journal.pone.0111060.
Kim, H.J., Cho, J.H., and Kim, J.R. (2013). Downregulation of Polo-like kinase 1 induces cellular senescence in human primary cells through a p53-dependent pathway. J Gerontol A Biol Sci Med Sci 68, 1145-1156. 10.1093/gerona/glt017.
Wierer, M., Verde, G., Pisano, P., Molina, H., Font-Mateu, J., Di Croce, L., and Beato, M. (2013). PLK1 signaling in breast cancer cells cooperates with estrogen receptor-dependent gene transcription. Cell Rep 3, 2021-2032. 10.1016/j.celrep.2013.05.024.

No competing interests reported.

Graphicalabstract.jpeg

Download PDF

Editor assigned by journal
12 Aug, 2024
Submission checks completed at journal
11 Aug, 2024
First submitted to journal
07 Aug, 2024

You are reading this latest preprint version

Transcriptomic changes during the replicative senescence of human articular chondrocytes

Status:

Version 1

Abstract

Objective

Methods

Results

Conclusions

Figures

Introduction

Materials and Methods

Study Design

Biospecimen Processing

Serial passage of chondrocytes

3D Pellet culture

Histology

Bulk RNA sequencing and data analysis

Quantitative Real Time Polymerase Chain Reaction (qRT-PCR)

Statistical Analysis

Results

Replicative senescence of chondrocytes

DEGs between early and late passage chondrocytes

Pathway analysis comparing chondrocytes from male and female with OA

Quantitative RT-PCR

Discussion

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1