Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

The role of chondrocyte senescence in osteoarthritis

The role of chondrocyte senescence in osteoarthritis <h1>Introduction</h1> Normal somatic cells do not divide indefinitely; this leads to an eventual arrest of cell division by a process called replicative senescence ( Hayflick, 1965 ). Such cells remain viable, but often exhibit an altered phenotype. Senescence has been studied extensively in cultured cells, particularly fibroblasts, by allowing cells to grow until replication stops. Senescence is extremely stringent in human cells which rarely undergo spontaneous immortalization. A current hypothesis is that the ‘counter’ for replicative senescence is telomere length. Telomeres cannot be completely replicated in primary cells and hence become shorter with each round of cell division. When telomere length becomes critically short, genetic changes are triggered which stop cell division and the senescent cell changes phenotype with an alteration in gene expression ( Bodnar et al ., 1998 ). A significant correlation between telomere length and donor age has been determined ( Lindsey et al ., 1991 ). Furthermore, in some cases, overexpression of the catalytic subunit of telomerase (the enzyme that maintains telomere length) can delay senescence; telomerase is not usually detectable in normal somatic cells, but is frequently expressed in tumours and immortalized cell lines ( Yudoh et al ., 2001 ) Cells cultured from old donors tend to senesce after fewer population doublings than cells from young donors ( Campisi, 2000 ). Thus, cells in renewable tissues may deplete their replicative potential during aging. It is assumed that senescent cells accumulate in vivo , where their altered phenotype may contribute to age-related pathology, e.g. dermal thinning and collagen breakdown are hallmarks of aging skin that may be due to senescent fibroblasts which overexpress collagenase and underexpress collagenase inhibitors ( Khorramizadeh et al ., 1999 ); endothelial cells lining vessels may senesce in response to haemodynamic stress and this may initiate or exacerbate atherosclerosis ( Okuda et al ., 2000 ; Minamino et al ., 2002 ). The most direct evidence for senescent cells in vivo comes from a modified histochemical marker for the enzyme ॆ-galactosidase. When detected at pH 6, ॆ-galactosidase is a biomarker of replicatively senescent cells. Dimri et al . (1995 ) used this marker to show an increase in senescent cells with donor age in skin sections from patients aged 20–90 years old. The SA-ॆ-gal marker has also been used in vivo in other tissues, e.g. arteries ( Minamino et al ., 2002 ) and gastrointestinal tract ( Going et al ., 2002 ). Studies performed in human fibroblasts demonstrate that replicatively senescent, late passage cells display altered patterns of expression of MMPs and TIMPs compared to young, replication-competent cells. Aging fibroblasts express MMP-1 mRNA and secrete MMP-1 at elevated constitutive levels ( Sottile et al ., 1989 ; West et al ., 1989 ; Edwards et al ., 1996 ). This has also been shown for MMP-3 , whereas steady-state levels of TIMP-1 were reduced compared to young fibroblasts ( Millis et al ., 1992 ). Age is a major risk factor for osteoarthritis (OA), with the disease usually starting from the late 40s onwards and causing significant morbidity in the population over 60 years of age. Extracellular matrix (ECM) synthesis by chondrocytes decreases with age and there is an age-dependent decline in the responsiveness of these cells to growth factors and cytokines ( Loeser et al ., 2000 ). The proliferative potential of articular chondrocytes also decreases with age, and in common with other cell types, chondrocytes can only undergo a limited number of cell divisions in vitro ( Evans & Georgescu, 1983 ). Age-related changes in cell phenotype could therefore allow ECM degradation to predominate, resulting in OA-like changes. Type II collagen and aggrecan, the large aggregating proteoglycan, are the two main structural macromolecules of articular cartilage. During OA, there is loss of both of these components and an overall failure of cartilage structural integrity (reviewed in Poole, 1997 ). The principal matrix-degrading enzymes are the matrix metalloproteinases (MMPs); these constitute a family of at least 25 enzymes, which, between them, are capable of degrading all the components of the ECM ( Nagase & Woessner, 2000 ). Activity of the MMPs is controlled in part by the presence of specific natural inhibitors known as tissue inhibitors of metalloproteinases (TIMPs). The current view is that the local balance of MMP and TIMP activities is pivotal in determining the extent of ECM turnover. Disturbance of this balance, resulting in an excess of active MMP over TIMP, underlies pathological cartilage destruction seen in many arthritides ( Dean et al ., 1989 ). The action of collagenases (e.g. MMP-1, -8, and -13) on type II collagen may represent the irreversible step in cartilage degradation. In recent years it has become clear that the major aggrecan-degrading enzymes in cartilage turnover and destruction are not MMPs, but belong to the ADAM-TS family of proteinases. Two aggrecanases, ADAM-TS4 and -TS5, have been described ( Cal et al ., 2002 ). Putting previous evidence together, we know that: (i) age is a major risk factor for OA; (ii) in OA, cartilage matrix is destroyed, the likely candidates for this destruction being MMPs and ADAMs; (iii) senescent cells show altered patterns of MMPs and TIMPs. Hence, we hypothesized that chondrocytes in aging or diseased cartilage may become senescent, with associated phenotypic changes contributing to the development or progression of OA. In this study, senescent cells were identified in OA cartilage using the senescence-associated ॆ-galactosidase (SA-ॆ-gal) biomarker, and this was underpinned using measurement of telomere length by Southern blotting. In addition, we determined whether there were changes in MMPs , ADAMs and TIMPs during OA that could be related to a senescent cell phenotype. <h1>Results</h1> <h2>SA-ॆ-gal as an in vivo marker of replicative senescence in cartilage</h2> To explore the idea that senescent chondrocytes accumulate in vivo during OA or aging, cartilage from both normal and osteoarthritic donors was stained for ॆ-gal activity as detailed in Methods. Staining for lysosomal ॆ-gal at pH 4 was performed as a positive control to show that ॆ-gal activity could be localized by this method in cartilage sections. Lysosomal staining was observed throughout the depth of osteoarthritic cartilage ( Fig. 1a ) and at high magnification this staining can be detected in all chondrocytes ( Fig. 1b ). Similar results were observed in normal cartilage and osteoarthritic cartilage from intact regions of the joint (data not shown). Senescence-associated ॆ-gal was stained at pH 6, and was observed in a subset of the chondrocytes close to the lesion site of both mild ( Fig. 2a ) and severely ( Fig. 2b ) damaged osteoarthritic knee cartilage. Interestingly, this staining appeared to be associated with chondrocyte clusters, which are a classical feature of osteoarthritic cartilage ( Fig. 2c ). No SA-ॆ-gal staining was detected in the intact regions of osteoarthritic knee cartilage that are distal to the lesion site ( Fig. 2d ). A similar pattern of SA-ॆ-gal staining was observed in osteoarthritic hip cartilage but with a lesser intensity, with staining being observed in chondrocytes close to the lesion site of the cartilage ( Fig. 3a ). No staining was observed in intact osteoarthritic hip cartilage ( Fig. 3b ) or in normal hip cartilage ( Fig. 3c ) of any age. <h2>Mean telomere length in cartilage chondrocytes</h2> Mean telomere length determinations were carried out in order to underline our finding of senescent chondrocytes at the lesion site of OA cartilage. Telomere length was decreased in primary chondrocytes derived from the OA lesion compared to chondrocytes from cartilage that was phenotypically intact and distal from the lesion site ( Fig. 4 ). We were unable to collect a large enough age range of ‘normal’ cartilage from non-OA patients to construct any ‘standard curve’ for telomere shortening in chondrocytes with normal aging. <h2>SA-ॆ-gal in cultured human chondrocytes</h2> The above data suggest that senescent chondrocytes are found at the lesion site in osteoarthritic cartilage. The aim of this experiment was to determine if chondrocytes expanded into primary culture from cartilage at the osteoarthritic lesion showed an increased number of senescent cells (as observed by histological staining for SA-ॆ-gal) than chondrocytes expanded from intact cartilage from the same osteoarthritic joint. First passage chondrocytes were cultured as monolayers and stained for both lysosomal and SA-ॆ-gal. As expected, lysosomal staining was observed in a similar pattern in chondrocytes from both the lesion ( Fig. 5a ) and the intact ( Fig. 5c ) osteoarthritic cartilage. There was also no significant difference in the SA-ॆ-gal staining in different chondrocyte populations. A range of 32–52% senescent cells was observed in the lesional chondrocytes ( Fig. 5b ) and 23–46% senescence in the chondrocytes from intact regions ( Fig. 5d ). Normal articular chondrocytes also showed a significant proportion of senescent cells (range 22–38%; data not shown), but again, there was no statistically significant difference compared to the cultures from diseased joints. <h2>Expression of proteinases and inhibitors in cartilage during OA</h2> The phenotypic consequence of cellular senescence includes an alteration in proteinase and inhibitor expression. Hence we examined expression of collagenases ( MMP-1 , -8 , -13 ), aggrecanases ( ADAM-TS4 and -TS5 ) and TIMP-1 and -3 at the mRNA level using TaqMan® RT-PCR. Figure 6 shows a comparison of expression in normal hip cartilage vs. intact cartilage from an OA hip vs. damaged cartilage from an OA lesion. Intact and OA lesion cartilage were from the same hip, and the graphs show data from six OA and six normal hips. A striking induction of both MMP-8 and -13 was observed in OA vs. normal, with MMP-1 showing the reverse pattern of expression. TIMP-1 expression also decreased in OA vs. normal, whilst TIMP-3 was the only gene to show a difference in expression between the intact region and OA lesion (though this did not reach significance in this small sample). Neither ADAM-TS4 nor TS5 showed any significant difference across the three groups (data not shown). <h1>Discussion</h1> We hypothesized that chondrocytes in aging and/or osteoarthritic cartilage might be replicatively senescent with associated phenotypic changes contributing to the development or progression of disease. In this study we demonstrate for the first time that chondrocytes in cartilage stain for the SA-ॆ-gal marker. Thus, senescent cells were observed close to the osteoarthritic lesion in both hip and knee cartilage, but no positive staining was detected in cells in intact cartilage from the same donors or from normal donors. Chondrocyte turnover in normal cartilage is presumed to be limited; however, during disease or insult, mitotic activity may be increased. This might lead to cells in already aging tissue (and thus close to senescence) entering senescence. Support for this possibility comes from our findings that senescent cells were often associated with chondrocyte clusters, which are known areas of high mitotic activity in osteoarthritic cartilage. This would also explain why normal cartilage (even from elderly patients, age range 81–91 years) and intact OA cartilage showed no SA-ॆ-gal staining in vivo . Lysosomal ॆ-gal and SA-ॆ-gal staining appear present in both cytoplasmic and nuclear compartments; this is more clearly seen in vitro , as shown in Fig. 5 . SA-ॆ-gal staining has been criticised as a marker of senescence at least in vitro (see below). In order to confirm that cell division had occurred in the regions of cartilage positive for SA-ॆ-gal, we measured telomere length using Southern blot. Shortening of the telomeres was evident in chondrocytes from the osteoarthritic lesion site compared to the more intact regions of cartilage in the same joint. This demonstrates that chondrocytes in the lesion have undergone more cell division on average and would therefore be expected to be closer to senenscence. In some joints we examined, only a small decrease in telomere length was observed in the lesion, and this may be a result of the severity of disease in these patients. The age of the patient will also be an important factor in determining telomere length and a number of studies have correlated telomere length with donor age ( Okuda et al ., 2000 ; Martin & Buckwalter, 2001 ; Yudoh et al ., 2001 ). Martin & Buckwalter (2001 ) have recently reported that average telomere length in cartilage chondrocytes decreases with age, with lengths ranging from approximately 12 to 8 kilobase pairs in samples from patients across an age range of 1–87 years. These researchers expanded chondrocytes briefly in culture prior to isolating genomic DNA for Southern blot; moreover, no information was given concerning disease status of the donors, and therefore a direct comparison with our data is impossible. However, these authors predict that local variations in average telomere length will exist across the joint with mechanical stress and injury, and this is borne out by our data. Overall, these results suggest that telomere shortening may be responsible for senescence in cartilage chondrocytes. Two recent reports provide data on this point: Piera-Velazquez et al . (2002 ) describe increased survival in vitro of cultured osteoarthritic chondrocytes by exogenous expression of the catalytic subunit of telomerase; Martin et al . (2002 ) argue that both telomere length and telomerase-independent pathways are required for the indefinite extension of chondrocyte survival in vitro . The high percentage of cultured chondrocytes that stain for SA-ॆ-gal when isolated from intact cartilage from osteoarthritic patients or even normal cartilage was unexpected from the in vivo data. However, the specificity of the SA-ॆ-gal stain in culture has been questioned. A recent report in cultured fibroblasts shows variation of SA-ॆ-gal staining with cell density and confluence, and no correlation between donor age and percentage of positive cells ( Severino et al ., 2000 ). In contrast, Martin & Buckwalter (2001 ) do show a correlation in percentage of SA-ॆ-gal-positive chondrocytes in culture with donor age with, e.g. 9% positive cells from an 8-year-old donor, 33% from a 52-year-old donor and 55% positive from a 87-year-old donor. The actual percentage of positive cells in the current study in normal and OA cartilage is in close agreement with the above numbers; moreover, although our data do not reach statistical significance, there is a trend towards increased SA-ॆ-gal staining in the OA lesion vs. distal cartilage and in OA vs. normal cartilage. Hence, our data may be explained in part by changes in expression of SA-ॆ-gal in culture but also by the very focal nature of the in vivo staining which may be lost against a background of other cells isolated from these samples for in vitro culture. Resolution of this question will await the application of microdissection to cartilage. Currently, much of the published data on gene expression in human articular cartilage is based on RNA from chondrocyte cultures. However, we know that chondrocytes in culture will de-differentiate, becoming fibroblast-like, often with alterations in their phenotype ( Lefebvre et al ., 1990 ). Recently, it has been shown that incubation with collagenase to release the chondrocytes from the cartilage matrix induces rapid changes in gene expression ( Martin et al ., 2001 ). This makes it difficult to relate the gene expression changes observed back to the in vivo situation. In this study we concentrate on gene expression in RNA extracted directly from cartilage. MMP-13 has been postulated as the collagenase responsible for collagen destruction in OA cartilage since it is expressed by chondrocytes and has a substrate preference for type II collagen over types I and III ( Knäuper et al ., 1996 ). Our data reinforce this, with raised expression of MMP-13 in OA cartilage compared to normal. MMP-8 , the ‘neutrophil collagenase’ recently shown to be more widely expressed ( Cole et al ., 1996 ), shows the same pattern of expression. However, although we cannot accurately compare absolute gene expression between genes using the current methodology, the ‘threshold cycle’ (Ct) for TaqMan® PCR of MMP-8 is much higher than for MMP-13 , suggesting low expression of the MMP-8 gene overall. Surprisingly, MMP-1 expression decreased in OA tissue; this may reflect a lack of involvement of this enzyme in cartilage collagen destruction in OA, or a repair/turnover role in normal cartilage that is lost in disease. TIMP-1 shows the same pattern of expression, whilst TIMP-3 appears at least slightly raised in the cartilage close to the OA lesion. Whilst all of these data are at the level of mRNA, and are from only n = 6 in each group, they do underline the complex pattern of enzyme and inhibitor expression that occurs in disease. Furthermore, the measurement of mRNA tracks the expression of individual genes, but it does not assess the overall balance of proteinase vs. inhibitor activity; our future studies will therefore seek to measure activity as well as gene expression. Given the focal nature of the SA-ॆ-gal-positive cells in OA cartilage, it is unlikely that change in cell phenotype with senescence is solely responsible for the changes in gene expression we demonstrate. Again, proof will require either microdissection and/or a comparative analysis of gene expression in chondrocytes aged in culture. In conclusion, we present the first demonstration of SA-ॆ-gal-positive cells in cartilage in vivo , and show that these cells occur predominantly associated with the osteoarthritic lesion. Furthermore we show that cells in the lesion have undergone cell division as evidenced by telomere shortening. Extraction of RNA from small (50–100 mg) fragments of cartilage has been optimized and, for the first time, we are now in a position to profile a complete panel of metalloproteinases and inhibitors in normal cartilage, and cartilage from intact and degraded sites in OA. <h1>Experimental procedures</h1> <h2>Human cartilage samples</h2> Human articular osteoarthritic cartilage was obtained from femoral heads of patients undergoing hip replacement surgery ( n = 15, age range 61–83 years) and from the femoral condyles and tibial plateaux of patients undergoing knee replacement surgery ( n = 30, age range 50–92 years) at the Norfolk and Norwich University Hospital. This study was performed with Ethical Committee approval and all patients provided informed consent. Where possible, cartilage was removed from both the site of the osteoarthritic lesion and phenotypically intact regions of the joint. Control cartilage was obtained from the femoral heads ( n = 11, age range 81–93 years) of patients undergoing hip replacement following fracture to the neck of the femur. These control patients had no known history of joint disease. <h2>Detection of ॆ-galactosidase (ॆ-gal) activity</h2> Cartilage removed from osteoarthritic patients and control subjects was immediately snap frozen in n-hexane and stored at −70 °C until use. The cartilage was then embedded in OCT (Merck, Lutterworth, Leicestershire, UK) and 7-µm sections were cut. Sections were fixed in 4% paraformaldehyde (Merck) in phosphate-buffered saline (PBS) for 5 min at room temperature. Sections were washed in PBS (5 min) and then digested with 0.5 mg mL −1 hyaluronidase (Sigma) in PBS for 30 min at 37 °C to permeabilize the cartilage. Residual hyaluronidase was removed by washing in PBS (5 min) and then the sections were immersed in freshly prepared senescence associated ॆ-gal (SA-ॆ-gal) staining solution (1 mg mL −1 of 5-bromo-4-chloro-3-indolyl ॆ- d -galactopyranoside (X-Gal, Melford Laboratories) in 40 m m citric acid/sodium phosphate solution pH 6, 5 m m potassium ferricyanide, 5 m m potassium ferrocyanide, 150 m m NaCl, 2 m m MgCl 2 ) for 12–16 h. A lysosomal form of the ॆ-gal enzyme thought to be present in most cells was detected by staining using the citric acid/sodium phosphate solution at pH 4 instead of pH 6. Sections were counterstained with eosin and mounted with DPX (Merck). Primary chondrocyte cultures grown at 150 000 cells per well in six-well plates were also examined for SA-ॆ-gal and lysosomal-ॆ-gal staining. Cells were fixed for 3–5 min in 2% formaldehyde, 0.2% glutaraldehyde, washed in PBS then immersed in staining solution as described above. <h2>Telomere restriction fragment assay</h2> Fresh human articular cartilage samples were digested overnight in Dulbecco's modified Eagle medium (DMEM; Gibco Life Technologies, Paisley, UK) containing 2 mg mL −1 of collagenase Type 1A (Sigma). The resulting cells were washed with PBS, re-suspended in DMEM containing 10% FCS and then plated at 1 × 10 6 cells in 75-cm 2 flasks. Cells were allowed to reach confluence before harvesting for genomic DNA extraction. The average telomere length in primary chondrocytes was determined by terminal restriction fragment Southern blotting analysis as previously described ( Yudoh et al ., 2001 ). Briefly, chondrocyte cell lysates were digested with 500 µL of DNA extraction buffer [100 m m NaCl, 40 m m Tris-HCl pH 8, 20 m m EDTA, 0.5% sodium dodecyl sulphate (SDS)] containing 0.1 mg mL −1 Proteinase K. Genomic DNA was then extracted using phenol-chloroform. Extracted genomic DNA was digested for 2 h at 37 °C with 10 units each of Rsa I and Hinf I (Gibco). Electrophoresis was then performed in 0.7% agarose gels in 45 m m Tris-borate-EDTA (TBE) buffer at pH 8 at 70–90 V for 4–5 h. Following electrophoresis, gels were depurinated in 0.2 n HCl, denatured in 0.5 m NaOH, 1 m NaCl and neutralized in 0.5 m Tris-HCl pH 7.5, 3 m NaCl. DNA was transferred by capillary action to Hybond-N+ (Amersham Pharmacia Biotech, St Albans, UK) nylon membranes in 20× standard sodium citrate (SSC). The membrane was then air-dried and cross-linked using UV irradiation. Membranes were prehybridized in hybridization buffer (5× Denhardt's solution, 5× SSC, 0.1% SDS), containing 100 µg mL −1 of salmon sperm DNA (Gibco). A synthetic oligonucleotide to the human telomeric repeat sequences (GGGATT) 3 was 5′ end-labelled with े 32 P-ATP using T4 polynucleotide kinase (Gibco) for 1 h at 37 °C and used to probe the membrane overnight at 37 °C. Excess probe was removed by washing the membrane in 3× SSC, 0.1% SDS (3 × 15 min), then in 0.1× SSC, 0.1% SDS (15 min) and the bound probe was detected by autoradiography. <h2>Total RNA extraction and cDNA synthesis</h2> Cartilage samples (50–100 mg) that had been snap frozen in liquid nitrogen were ground for three cycles of 2 min grinding and 2 min cooling at an impact frequency of 10 Hz in a freezer/mill (SPEX CertiPrep 6750; Glen Creston, Stanmore, Middlesex, UK) under liquid nitrogen. RNA was isolated from the powdered cartilage using Trizol® Reagent (Gibco), which is a modification of the single-step acid-phenol guanidinium method developed by Chomczynski & Sacchi (1987 ). After the addition of chloroform and centrifugation, the aqueous phase containing the RNA was mixed with a half volume of 100% ethanol and further purified on silica-gel-based membranes (Rneasy® Plant Mini Kit; Qiagen, Crawley, UK) according to the manufacturer's instructions. The Plant Mini Kit is specifically designed to remove polysaccharides and proteoglycans which interfere with the RNA purification and therefore works particularly well to remove proteoglycans from the cartilage RNA. The RNA was eluted once using 30 µL of RNase-free water, then again using the first eluate and stored at −70 °C until use. cDNA was generated from 1 µg of RNA in a 20-µL reaction using random hexamers and Superscript II reverse transcriptase (Gibco) according to the manufacturer's instructions. <h2>Design of primers and probes for TaqMan® PCR</h2> Oligonucleotide primers and fluorescent labelled TaqMan® probes were designed using Primer Express 1.0 software (Applied Biosystems, Warrington, UK). Sequences for the MMP and TIMP primers and probes are copyright of Applied Biosystems. Sequences for ADAMTS primers and probes are: ADAMTS-4 : Fwd = 5′-CAAGGTCCCATGTGCAACGT-3′; Rev = 5′-CATCTGCCACCACCAGTGTCT-3′; Pr = 5′-FAM-CCGAAGAGCCAAGCGCTTTGCTTC-TAMRA-3′. ADAMTS-5 : Fwd = 5′-TGTCCTGCCAGCGGATGT-3′; Rev = 5′-ACGGAATTACTGTACGGCCTACA-3′; Pr = 5′-FAM-TTCTCCAAAGGTGACCGATGGCACTG-TAMRA-3′. In order to control against amplification of genomic DNA and to ensure that the PCR signal was generated from cDNA, primers were placed within different exons, close to intron/exon boundaries. BLASTN searches were conducted on all primer/probe nucleotide sequences to ensure gene specificity. The 18S ribosomal RNA gene was used as an endogenous control to normalize for differences in the amount of total RNA in each sample. TaqMan® 18S ribosomal RNA primers and a 5′-VIC™-labelled probe were used according to the manufacturer's instructions (Applied Biosystems). <h2>TaqMan® PCR</h2> Relative quantification of gene expression was performed using the Applied Biosystems ABI Prism 7700 sequence detection system (TaqMan®). PCR reactions for all samples were performed in 96-well optical plates using 5 ng of cDNA, 12.5 µL 2× TaqMan® Universal PCR mastermix (Applied Biosystems), 100 n m probe, 200 n m of each primer and water to a 25-µL final volume. Thermocycler conditions comprised an initial holding at 50 °C for 2 min then 95 °C for 10 min. This was followed by a two-step TaqMan® PCR program consisting of 95 °C for 15 s and 60 °C for 60 s for 40 cycles. <h3>Analysis</h3> During PCR the TaqMan® probe emits a fluorescence signal which increases in intensity in direct proportion to the amount of specific amplified product. The ABI Prism 7700 instrument measures the cycle-to-cycle changes in fluorescence in each sample. Data are initially expressed as a threshold cycle (Ct) at which an increase in fluorescence above a baseline signal can first be detected. The fewer cycles it takes to reach the Ct, the greater the initial template copy number. The Ct values generated were used to calculate relative input amounts of template cDNA using the standard curve method as described in the Applied Biosystems User Bulletin #2 (1997). Input cDNA levels measured correspond directly to the levels of RNA reverse transcribed. Thus data are presented graphically as relative levels of mRNA for each primer/probe set with differences ascertained using Student's t -test. Direct comparisons between levels cannot be made between different primer/probe sets. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Aging Cell Wiley

The role of chondrocyte senescence in osteoarthritis

Download PDF
9 pages

Loading next page...
 
/lp/wiley/the-role-of-chondrocyte-senescence-in-osteoarthritis-RPE1vkoT0i

References (32)

Publisher
Wiley
Copyright
Copyright © 2002 Wiley Subscription Services, Inc., A Wiley Company
eISSN
1474-9726
DOI
10.1046/j.1474-9728.2002.00008.x
Publisher site
See Article on Publisher Site

Abstract

<h1>Introduction</h1> Normal somatic cells do not divide indefinitely; this leads to an eventual arrest of cell division by a process called replicative senescence ( Hayflick, 1965 ). Such cells remain viable, but often exhibit an altered phenotype. Senescence has been studied extensively in cultured cells, particularly fibroblasts, by allowing cells to grow until replication stops. Senescence is extremely stringent in human cells which rarely undergo spontaneous immortalization. A current hypothesis is that the ‘counter’ for replicative senescence is telomere length. Telomeres cannot be completely replicated in primary cells and hence become shorter with each round of cell division. When telomere length becomes critically short, genetic changes are triggered which stop cell division and the senescent cell changes phenotype with an alteration in gene expression ( Bodnar et al ., 1998 ). A significant correlation between telomere length and donor age has been determined ( Lindsey et al ., 1991 ). Furthermore, in some cases, overexpression of the catalytic subunit of telomerase (the enzyme that maintains telomere length) can delay senescence; telomerase is not usually detectable in normal somatic cells, but is frequently expressed in tumours and immortalized cell lines ( Yudoh et al ., 2001 ) Cells cultured from old donors tend to senesce after fewer population doublings than cells from young donors ( Campisi, 2000 ). Thus, cells in renewable tissues may deplete their replicative potential during aging. It is assumed that senescent cells accumulate in vivo , where their altered phenotype may contribute to age-related pathology, e.g. dermal thinning and collagen breakdown are hallmarks of aging skin that may be due to senescent fibroblasts which overexpress collagenase and underexpress collagenase inhibitors ( Khorramizadeh et al ., 1999 ); endothelial cells lining vessels may senesce in response to haemodynamic stress and this may initiate or exacerbate atherosclerosis ( Okuda et al ., 2000 ; Minamino et al ., 2002 ). The most direct evidence for senescent cells in vivo comes from a modified histochemical marker for the enzyme ॆ-galactosidase. When detected at pH 6, ॆ-galactosidase is a biomarker of replicatively senescent cells. Dimri et al . (1995 ) used this marker to show an increase in senescent cells with donor age in skin sections from patients aged 20–90 years old. The SA-ॆ-gal marker has also been used in vivo in other tissues, e.g. arteries ( Minamino et al ., 2002 ) and gastrointestinal tract ( Going et al ., 2002 ). Studies performed in human fibroblasts demonstrate that replicatively senescent, late passage cells display altered patterns of expression of MMPs and TIMPs compared to young, replication-competent cells. Aging fibroblasts express MMP-1 mRNA and secrete MMP-1 at elevated constitutive levels ( Sottile et al ., 1989 ; West et al ., 1989 ; Edwards et al ., 1996 ). This has also been shown for MMP-3 , whereas steady-state levels of TIMP-1 were reduced compared to young fibroblasts ( Millis et al ., 1992 ). Age is a major risk factor for osteoarthritis (OA), with the disease usually starting from the late 40s onwards and causing significant morbidity in the population over 60 years of age. Extracellular matrix (ECM) synthesis by chondrocytes decreases with age and there is an age-dependent decline in the responsiveness of these cells to growth factors and cytokines ( Loeser et al ., 2000 ). The proliferative potential of articular chondrocytes also decreases with age, and in common with other cell types, chondrocytes can only undergo a limited number of cell divisions in vitro ( Evans & Georgescu, 1983 ). Age-related changes in cell phenotype could therefore allow ECM degradation to predominate, resulting in OA-like changes. Type II collagen and aggrecan, the large aggregating proteoglycan, are the two main structural macromolecules of articular cartilage. During OA, there is loss of both of these components and an overall failure of cartilage structural integrity (reviewed in Poole, 1997 ). The principal matrix-degrading enzymes are the matrix metalloproteinases (MMPs); these constitute a family of at least 25 enzymes, which, between them, are capable of degrading all the components of the ECM ( Nagase & Woessner, 2000 ). Activity of the MMPs is controlled in part by the presence of specific natural inhibitors known as tissue inhibitors of metalloproteinases (TIMPs). The current view is that the local balance of MMP and TIMP activities is pivotal in determining the extent of ECM turnover. Disturbance of this balance, resulting in an excess of active MMP over TIMP, underlies pathological cartilage destruction seen in many arthritides ( Dean et al ., 1989 ). The action of collagenases (e.g. MMP-1, -8, and -13) on type II collagen may represent the irreversible step in cartilage degradation. In recent years it has become clear that the major aggrecan-degrading enzymes in cartilage turnover and destruction are not MMPs, but belong to the ADAM-TS family of proteinases. Two aggrecanases, ADAM-TS4 and -TS5, have been described ( Cal et al ., 2002 ). Putting previous evidence together, we know that: (i) age is a major risk factor for OA; (ii) in OA, cartilage matrix is destroyed, the likely candidates for this destruction being MMPs and ADAMs; (iii) senescent cells show altered patterns of MMPs and TIMPs. Hence, we hypothesized that chondrocytes in aging or diseased cartilage may become senescent, with associated phenotypic changes contributing to the development or progression of OA. In this study, senescent cells were identified in OA cartilage using the senescence-associated ॆ-galactosidase (SA-ॆ-gal) biomarker, and this was underpinned using measurement of telomere length by Southern blotting. In addition, we determined whether there were changes in MMPs , ADAMs and TIMPs during OA that could be related to a senescent cell phenotype. <h1>Results</h1> <h2>SA-ॆ-gal as an in vivo marker of replicative senescence in cartilage</h2> To explore the idea that senescent chondrocytes accumulate in vivo during OA or aging, cartilage from both normal and osteoarthritic donors was stained for ॆ-gal activity as detailed in Methods. Staining for lysosomal ॆ-gal at pH 4 was performed as a positive control to show that ॆ-gal activity could be localized by this method in cartilage sections. Lysosomal staining was observed throughout the depth of osteoarthritic cartilage ( Fig. 1a ) and at high magnification this staining can be detected in all chondrocytes ( Fig. 1b ). Similar results were observed in normal cartilage and osteoarthritic cartilage from intact regions of the joint (data not shown). Senescence-associated ॆ-gal was stained at pH 6, and was observed in a subset of the chondrocytes close to the lesion site of both mild ( Fig. 2a ) and severely ( Fig. 2b ) damaged osteoarthritic knee cartilage. Interestingly, this staining appeared to be associated with chondrocyte clusters, which are a classical feature of osteoarthritic cartilage ( Fig. 2c ). No SA-ॆ-gal staining was detected in the intact regions of osteoarthritic knee cartilage that are distal to the lesion site ( Fig. 2d ). A similar pattern of SA-ॆ-gal staining was observed in osteoarthritic hip cartilage but with a lesser intensity, with staining being observed in chondrocytes close to the lesion site of the cartilage ( Fig. 3a ). No staining was observed in intact osteoarthritic hip cartilage ( Fig. 3b ) or in normal hip cartilage ( Fig. 3c ) of any age. <h2>Mean telomere length in cartilage chondrocytes</h2> Mean telomere length determinations were carried out in order to underline our finding of senescent chondrocytes at the lesion site of OA cartilage. Telomere length was decreased in primary chondrocytes derived from the OA lesion compared to chondrocytes from cartilage that was phenotypically intact and distal from the lesion site ( Fig. 4 ). We were unable to collect a large enough age range of ‘normal’ cartilage from non-OA patients to construct any ‘standard curve’ for telomere shortening in chondrocytes with normal aging. <h2>SA-ॆ-gal in cultured human chondrocytes</h2> The above data suggest that senescent chondrocytes are found at the lesion site in osteoarthritic cartilage. The aim of this experiment was to determine if chondrocytes expanded into primary culture from cartilage at the osteoarthritic lesion showed an increased number of senescent cells (as observed by histological staining for SA-ॆ-gal) than chondrocytes expanded from intact cartilage from the same osteoarthritic joint. First passage chondrocytes were cultured as monolayers and stained for both lysosomal and SA-ॆ-gal. As expected, lysosomal staining was observed in a similar pattern in chondrocytes from both the lesion ( Fig. 5a ) and the intact ( Fig. 5c ) osteoarthritic cartilage. There was also no significant difference in the SA-ॆ-gal staining in different chondrocyte populations. A range of 32–52% senescent cells was observed in the lesional chondrocytes ( Fig. 5b ) and 23–46% senescence in the chondrocytes from intact regions ( Fig. 5d ). Normal articular chondrocytes also showed a significant proportion of senescent cells (range 22–38%; data not shown), but again, there was no statistically significant difference compared to the cultures from diseased joints. <h2>Expression of proteinases and inhibitors in cartilage during OA</h2> The phenotypic consequence of cellular senescence includes an alteration in proteinase and inhibitor expression. Hence we examined expression of collagenases ( MMP-1 , -8 , -13 ), aggrecanases ( ADAM-TS4 and -TS5 ) and TIMP-1 and -3 at the mRNA level using TaqMan® RT-PCR. Figure 6 shows a comparison of expression in normal hip cartilage vs. intact cartilage from an OA hip vs. damaged cartilage from an OA lesion. Intact and OA lesion cartilage were from the same hip, and the graphs show data from six OA and six normal hips. A striking induction of both MMP-8 and -13 was observed in OA vs. normal, with MMP-1 showing the reverse pattern of expression. TIMP-1 expression also decreased in OA vs. normal, whilst TIMP-3 was the only gene to show a difference in expression between the intact region and OA lesion (though this did not reach significance in this small sample). Neither ADAM-TS4 nor TS5 showed any significant difference across the three groups (data not shown). <h1>Discussion</h1> We hypothesized that chondrocytes in aging and/or osteoarthritic cartilage might be replicatively senescent with associated phenotypic changes contributing to the development or progression of disease. In this study we demonstrate for the first time that chondrocytes in cartilage stain for the SA-ॆ-gal marker. Thus, senescent cells were observed close to the osteoarthritic lesion in both hip and knee cartilage, but no positive staining was detected in cells in intact cartilage from the same donors or from normal donors. Chondrocyte turnover in normal cartilage is presumed to be limited; however, during disease or insult, mitotic activity may be increased. This might lead to cells in already aging tissue (and thus close to senescence) entering senescence. Support for this possibility comes from our findings that senescent cells were often associated with chondrocyte clusters, which are known areas of high mitotic activity in osteoarthritic cartilage. This would also explain why normal cartilage (even from elderly patients, age range 81–91 years) and intact OA cartilage showed no SA-ॆ-gal staining in vivo . Lysosomal ॆ-gal and SA-ॆ-gal staining appear present in both cytoplasmic and nuclear compartments; this is more clearly seen in vitro , as shown in Fig. 5 . SA-ॆ-gal staining has been criticised as a marker of senescence at least in vitro (see below). In order to confirm that cell division had occurred in the regions of cartilage positive for SA-ॆ-gal, we measured telomere length using Southern blot. Shortening of the telomeres was evident in chondrocytes from the osteoarthritic lesion site compared to the more intact regions of cartilage in the same joint. This demonstrates that chondrocytes in the lesion have undergone more cell division on average and would therefore be expected to be closer to senenscence. In some joints we examined, only a small decrease in telomere length was observed in the lesion, and this may be a result of the severity of disease in these patients. The age of the patient will also be an important factor in determining telomere length and a number of studies have correlated telomere length with donor age ( Okuda et al ., 2000 ; Martin & Buckwalter, 2001 ; Yudoh et al ., 2001 ). Martin & Buckwalter (2001 ) have recently reported that average telomere length in cartilage chondrocytes decreases with age, with lengths ranging from approximately 12 to 8 kilobase pairs in samples from patients across an age range of 1–87 years. These researchers expanded chondrocytes briefly in culture prior to isolating genomic DNA for Southern blot; moreover, no information was given concerning disease status of the donors, and therefore a direct comparison with our data is impossible. However, these authors predict that local variations in average telomere length will exist across the joint with mechanical stress and injury, and this is borne out by our data. Overall, these results suggest that telomere shortening may be responsible for senescence in cartilage chondrocytes. Two recent reports provide data on this point: Piera-Velazquez et al . (2002 ) describe increased survival in vitro of cultured osteoarthritic chondrocytes by exogenous expression of the catalytic subunit of telomerase; Martin et al . (2002 ) argue that both telomere length and telomerase-independent pathways are required for the indefinite extension of chondrocyte survival in vitro . The high percentage of cultured chondrocytes that stain for SA-ॆ-gal when isolated from intact cartilage from osteoarthritic patients or even normal cartilage was unexpected from the in vivo data. However, the specificity of the SA-ॆ-gal stain in culture has been questioned. A recent report in cultured fibroblasts shows variation of SA-ॆ-gal staining with cell density and confluence, and no correlation between donor age and percentage of positive cells ( Severino et al ., 2000 ). In contrast, Martin & Buckwalter (2001 ) do show a correlation in percentage of SA-ॆ-gal-positive chondrocytes in culture with donor age with, e.g. 9% positive cells from an 8-year-old donor, 33% from a 52-year-old donor and 55% positive from a 87-year-old donor. The actual percentage of positive cells in the current study in normal and OA cartilage is in close agreement with the above numbers; moreover, although our data do not reach statistical significance, there is a trend towards increased SA-ॆ-gal staining in the OA lesion vs. distal cartilage and in OA vs. normal cartilage. Hence, our data may be explained in part by changes in expression of SA-ॆ-gal in culture but also by the very focal nature of the in vivo staining which may be lost against a background of other cells isolated from these samples for in vitro culture. Resolution of this question will await the application of microdissection to cartilage. Currently, much of the published data on gene expression in human articular cartilage is based on RNA from chondrocyte cultures. However, we know that chondrocytes in culture will de-differentiate, becoming fibroblast-like, often with alterations in their phenotype ( Lefebvre et al ., 1990 ). Recently, it has been shown that incubation with collagenase to release the chondrocytes from the cartilage matrix induces rapid changes in gene expression ( Martin et al ., 2001 ). This makes it difficult to relate the gene expression changes observed back to the in vivo situation. In this study we concentrate on gene expression in RNA extracted directly from cartilage. MMP-13 has been postulated as the collagenase responsible for collagen destruction in OA cartilage since it is expressed by chondrocytes and has a substrate preference for type II collagen over types I and III ( Knäuper et al ., 1996 ). Our data reinforce this, with raised expression of MMP-13 in OA cartilage compared to normal. MMP-8 , the ‘neutrophil collagenase’ recently shown to be more widely expressed ( Cole et al ., 1996 ), shows the same pattern of expression. However, although we cannot accurately compare absolute gene expression between genes using the current methodology, the ‘threshold cycle’ (Ct) for TaqMan® PCR of MMP-8 is much higher than for MMP-13 , suggesting low expression of the MMP-8 gene overall. Surprisingly, MMP-1 expression decreased in OA tissue; this may reflect a lack of involvement of this enzyme in cartilage collagen destruction in OA, or a repair/turnover role in normal cartilage that is lost in disease. TIMP-1 shows the same pattern of expression, whilst TIMP-3 appears at least slightly raised in the cartilage close to the OA lesion. Whilst all of these data are at the level of mRNA, and are from only n = 6 in each group, they do underline the complex pattern of enzyme and inhibitor expression that occurs in disease. Furthermore, the measurement of mRNA tracks the expression of individual genes, but it does not assess the overall balance of proteinase vs. inhibitor activity; our future studies will therefore seek to measure activity as well as gene expression. Given the focal nature of the SA-ॆ-gal-positive cells in OA cartilage, it is unlikely that change in cell phenotype with senescence is solely responsible for the changes in gene expression we demonstrate. Again, proof will require either microdissection and/or a comparative analysis of gene expression in chondrocytes aged in culture. In conclusion, we present the first demonstration of SA-ॆ-gal-positive cells in cartilage in vivo , and show that these cells occur predominantly associated with the osteoarthritic lesion. Furthermore we show that cells in the lesion have undergone cell division as evidenced by telomere shortening. Extraction of RNA from small (50–100 mg) fragments of cartilage has been optimized and, for the first time, we are now in a position to profile a complete panel of metalloproteinases and inhibitors in normal cartilage, and cartilage from intact and degraded sites in OA. <h1>Experimental procedures</h1> <h2>Human cartilage samples</h2> Human articular osteoarthritic cartilage was obtained from femoral heads of patients undergoing hip replacement surgery ( n = 15, age range 61–83 years) and from the femoral condyles and tibial plateaux of patients undergoing knee replacement surgery ( n = 30, age range 50–92 years) at the Norfolk and Norwich University Hospital. This study was performed with Ethical Committee approval and all patients provided informed consent. Where possible, cartilage was removed from both the site of the osteoarthritic lesion and phenotypically intact regions of the joint. Control cartilage was obtained from the femoral heads ( n = 11, age range 81–93 years) of patients undergoing hip replacement following fracture to the neck of the femur. These control patients had no known history of joint disease. <h2>Detection of ॆ-galactosidase (ॆ-gal) activity</h2> Cartilage removed from osteoarthritic patients and control subjects was immediately snap frozen in n-hexane and stored at −70 °C until use. The cartilage was then embedded in OCT (Merck, Lutterworth, Leicestershire, UK) and 7-µm sections were cut. Sections were fixed in 4% paraformaldehyde (Merck) in phosphate-buffered saline (PBS) for 5 min at room temperature. Sections were washed in PBS (5 min) and then digested with 0.5 mg mL −1 hyaluronidase (Sigma) in PBS for 30 min at 37 °C to permeabilize the cartilage. Residual hyaluronidase was removed by washing in PBS (5 min) and then the sections were immersed in freshly prepared senescence associated ॆ-gal (SA-ॆ-gal) staining solution (1 mg mL −1 of 5-bromo-4-chloro-3-indolyl ॆ- d -galactopyranoside (X-Gal, Melford Laboratories) in 40 m m citric acid/sodium phosphate solution pH 6, 5 m m potassium ferricyanide, 5 m m potassium ferrocyanide, 150 m m NaCl, 2 m m MgCl 2 ) for 12–16 h. A lysosomal form of the ॆ-gal enzyme thought to be present in most cells was detected by staining using the citric acid/sodium phosphate solution at pH 4 instead of pH 6. Sections were counterstained with eosin and mounted with DPX (Merck). Primary chondrocyte cultures grown at 150 000 cells per well in six-well plates were also examined for SA-ॆ-gal and lysosomal-ॆ-gal staining. Cells were fixed for 3–5 min in 2% formaldehyde, 0.2% glutaraldehyde, washed in PBS then immersed in staining solution as described above. <h2>Telomere restriction fragment assay</h2> Fresh human articular cartilage samples were digested overnight in Dulbecco's modified Eagle medium (DMEM; Gibco Life Technologies, Paisley, UK) containing 2 mg mL −1 of collagenase Type 1A (Sigma). The resulting cells were washed with PBS, re-suspended in DMEM containing 10% FCS and then plated at 1 × 10 6 cells in 75-cm 2 flasks. Cells were allowed to reach confluence before harvesting for genomic DNA extraction. The average telomere length in primary chondrocytes was determined by terminal restriction fragment Southern blotting analysis as previously described ( Yudoh et al ., 2001 ). Briefly, chondrocyte cell lysates were digested with 500 µL of DNA extraction buffer [100 m m NaCl, 40 m m Tris-HCl pH 8, 20 m m EDTA, 0.5% sodium dodecyl sulphate (SDS)] containing 0.1 mg mL −1 Proteinase K. Genomic DNA was then extracted using phenol-chloroform. Extracted genomic DNA was digested for 2 h at 37 °C with 10 units each of Rsa I and Hinf I (Gibco). Electrophoresis was then performed in 0.7% agarose gels in 45 m m Tris-borate-EDTA (TBE) buffer at pH 8 at 70–90 V for 4–5 h. Following electrophoresis, gels were depurinated in 0.2 n HCl, denatured in 0.5 m NaOH, 1 m NaCl and neutralized in 0.5 m Tris-HCl pH 7.5, 3 m NaCl. DNA was transferred by capillary action to Hybond-N+ (Amersham Pharmacia Biotech, St Albans, UK) nylon membranes in 20× standard sodium citrate (SSC). The membrane was then air-dried and cross-linked using UV irradiation. Membranes were prehybridized in hybridization buffer (5× Denhardt's solution, 5× SSC, 0.1% SDS), containing 100 µg mL −1 of salmon sperm DNA (Gibco). A synthetic oligonucleotide to the human telomeric repeat sequences (GGGATT) 3 was 5′ end-labelled with े 32 P-ATP using T4 polynucleotide kinase (Gibco) for 1 h at 37 °C and used to probe the membrane overnight at 37 °C. Excess probe was removed by washing the membrane in 3× SSC, 0.1% SDS (3 × 15 min), then in 0.1× SSC, 0.1% SDS (15 min) and the bound probe was detected by autoradiography. <h2>Total RNA extraction and cDNA synthesis</h2> Cartilage samples (50–100 mg) that had been snap frozen in liquid nitrogen were ground for three cycles of 2 min grinding and 2 min cooling at an impact frequency of 10 Hz in a freezer/mill (SPEX CertiPrep 6750; Glen Creston, Stanmore, Middlesex, UK) under liquid nitrogen. RNA was isolated from the powdered cartilage using Trizol® Reagent (Gibco), which is a modification of the single-step acid-phenol guanidinium method developed by Chomczynski & Sacchi (1987 ). After the addition of chloroform and centrifugation, the aqueous phase containing the RNA was mixed with a half volume of 100% ethanol and further purified on silica-gel-based membranes (Rneasy® Plant Mini Kit; Qiagen, Crawley, UK) according to the manufacturer's instructions. The Plant Mini Kit is specifically designed to remove polysaccharides and proteoglycans which interfere with the RNA purification and therefore works particularly well to remove proteoglycans from the cartilage RNA. The RNA was eluted once using 30 µL of RNase-free water, then again using the first eluate and stored at −70 °C until use. cDNA was generated from 1 µg of RNA in a 20-µL reaction using random hexamers and Superscript II reverse transcriptase (Gibco) according to the manufacturer's instructions. <h2>Design of primers and probes for TaqMan® PCR</h2> Oligonucleotide primers and fluorescent labelled TaqMan® probes were designed using Primer Express 1.0 software (Applied Biosystems, Warrington, UK). Sequences for the MMP and TIMP primers and probes are copyright of Applied Biosystems. Sequences for ADAMTS primers and probes are: ADAMTS-4 : Fwd = 5′-CAAGGTCCCATGTGCAACGT-3′; Rev = 5′-CATCTGCCACCACCAGTGTCT-3′; Pr = 5′-FAM-CCGAAGAGCCAAGCGCTTTGCTTC-TAMRA-3′. ADAMTS-5 : Fwd = 5′-TGTCCTGCCAGCGGATGT-3′; Rev = 5′-ACGGAATTACTGTACGGCCTACA-3′; Pr = 5′-FAM-TTCTCCAAAGGTGACCGATGGCACTG-TAMRA-3′. In order to control against amplification of genomic DNA and to ensure that the PCR signal was generated from cDNA, primers were placed within different exons, close to intron/exon boundaries. BLASTN searches were conducted on all primer/probe nucleotide sequences to ensure gene specificity. The 18S ribosomal RNA gene was used as an endogenous control to normalize for differences in the amount of total RNA in each sample. TaqMan® 18S ribosomal RNA primers and a 5′-VIC™-labelled probe were used according to the manufacturer's instructions (Applied Biosystems). <h2>TaqMan® PCR</h2> Relative quantification of gene expression was performed using the Applied Biosystems ABI Prism 7700 sequence detection system (TaqMan®). PCR reactions for all samples were performed in 96-well optical plates using 5 ng of cDNA, 12.5 µL 2× TaqMan® Universal PCR mastermix (Applied Biosystems), 100 n m probe, 200 n m of each primer and water to a 25-µL final volume. Thermocycler conditions comprised an initial holding at 50 °C for 2 min then 95 °C for 10 min. This was followed by a two-step TaqMan® PCR program consisting of 95 °C for 15 s and 60 °C for 60 s for 40 cycles. <h3>Analysis</h3> During PCR the TaqMan® probe emits a fluorescence signal which increases in intensity in direct proportion to the amount of specific amplified product. The ABI Prism 7700 instrument measures the cycle-to-cycle changes in fluorescence in each sample. Data are initially expressed as a threshold cycle (Ct) at which an increase in fluorescence above a baseline signal can first be detected. The fewer cycles it takes to reach the Ct, the greater the initial template copy number. The Ct values generated were used to calculate relative input amounts of template cDNA using the standard curve method as described in the Applied Biosystems User Bulletin #2 (1997). Input cDNA levels measured correspond directly to the levels of RNA reverse transcribed. Thus data are presented graphically as relative levels of mRNA for each primer/probe set with differences ascertained using Student's t -test. Direct comparisons between levels cannot be made between different primer/probe sets.

Journal

Aging Cell – Wiley

Published: Oct 1, 2002

Keywords: ; ; ; ;

There are no references for this article.