Increased life span of human osteoarthritic chondrocytes by exogenous expression of telomerase.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 46, No. 3, March 2002, pp 683–693 DOI 10.1002/art.10116 © 2002, American College of Rheumatology Increased Life Span of Human Osteoarthritic Chondrocytes by Exogenous Expression of Telomerase Sonsoles Piera-Velazquez, Sergio A. Jimenez, and David G. Stokes Objective. To extend the life span of human osteoarthritic (OA) articular chondrocytes by introduction of the catalytic component of human telomerase while preserving the chondrocyte-specific phenotype. Methods. Human articular chondrocytes were isolated from the femoral head and tibial plateau of patients undergoing knee joint replacement for OA. The chondrocytes were cultured as monolayers and infected with a retroviral telomerase expression construct followed by selection with G418 for 10–14 days. Telomericrepeat amplification protocol assays and telomere terminal restriction fragment length assays were performed on pools of transduced cells in order to measure telomerase activity and telomere length. Growth kinetics and population doubling capacity were assessed by passaging the cells in monolayer culture. Redifferentiation of the monolayer chondrocyte cultures was induced by transfer to suspension culture on poly(2-hydroxyethyl-methacrylate) (polyHEMA)–coated dishes. Induction of the chondrocyte-specific phenotype was monitored by analysis of gene expression utilizing reverse transcription–polymerase chain reaction. Results. OA chondrocytes isolated from 3 different donors (ages 41, 69, and 75 years) were transduced with a retroviral construct expressing telomerase. After selection, pooled populations of cells from all donors and a clonal cell line from 1 donor expressed telomerase activity and exhibited lengthening of telomeres. Chondrocytes expressing telomerase showed an increase of 5–9 population doublings over 234 days of culture in monolayer. The telomerase-transduced cells recovered a chondrocyte-specific gene expression pattern following culture on polyHEMA-coated dishes. Conclusion. The exogenous expression of telomerase may represent a way to expand human OA chondrocytes while allowing maintenance of the chondrocytespecific phenotype. These cells have the potential to be used for restoration of the articular cartilage defects occurring in this disease. A number of current therapeutic protocols for correction of focal cartilage defects and possible future treatment of degenerative joint diseases such as osteoarthritis (OA) involve the expansion of autologous chondrocytes followed by reimplantation of these cells into cartilage defects, their injection into affected joints, or their utilization for engineering of replacement tissue ex vivo (1,2). All of these protocols require an expansion phase of chondrocytes in culture in order to obtain sufficient cells to implement treatment. However, it is well known that primary mammalian cells in culture have a finite replicative life span and eventually enter a state of senescence in which they remain metabolically active but cease to proliferate (3). Furthermore, the mitotic potential of primary cells in culture is dependent on the age of the donor, with cells from older individuals exhibiting a lower proliferative life span (3). Normal adult chondrocytes also possess a limited mitotic potential and inevitably enter a state of replicative senescence in which cellular proliferation ceases (4–6). It has been demonstrated that chondrocytes from various species show a relationship between the number of population doublings achieved in vitro and the life span of the organism (5). Recent work has shown that both the proliferative activity and the telomere length of human articular chondrocytes decrease with the age of the donor (6). Indeed, chondrocytes from articular cartilage exhibit a number of age-related changes in their Supported in part by a grant from Geron Corporation. Dr. Stokes’ work was supported by NIH grant AG-19962-01. Dr. PieraVelazquez is the recipient of an Arthritis Foundation postdoctoral fellowship. Sonsoles Piera-Velazquez, PhD, Sergio A. Jimenez, MD, David G. Stokes, PhD: Thomas Jefferson University, Philadelphia, Pennsylvania. Address correspondence and reprint requests to David G. Stokes, PhD, Thomas Jefferson University, Division of Rheumatology, 233 S. 10th Street, Room 509 BLSB, Philadelphia, PA 19107-5541. E-mail: David.G.Stokes@mail.tju.edu. Submitted for publication August 7, 2001; accepted in revised form October 12, 2001. 683 684 phenotype. Among these changes are decreased response to growth factors such as transforming growth factor ␤ (TGF␤) and insulin-like growth factor 1, increased apoptosis, and decreased extracellular matrix production (7,8). Compounding the latter events are the phenotypic changes in chondrocytes that occur during OA, a disease that has a high correlation with age (9,10). These alterations, collectively, place a limit on the usefulness of autologous chondrocytes isolated from aged OA joints in the therapeutic strategies mentioned above. Numerous studies have shown that shortening of telomere length, which occurs during each cell division, is probably a signal for cellular senescence, since cells in which telomere length has been shortened to a critical level fail to undergo further mitotic events (3,11–13). The enzyme telomerase is a reverse transcriptase that solves the problem of DNA end-replication, and therefore affects telomere shortening due to the inability of eukaryotic DNA polymerases to completely replicate the ends of linear chromosomes (14,15). Telomerase catalyzes the synthesis of telomeres which, in the case of humans, are repeat sequences of ⬃1,000 copies of the short guanine-rich sequence motif TTAGGG, located at the ends of chromosomes (for review, see ref. 16). Human telomerase is a ribonucleoprotein consisting of a reverse transcriptase (hTERT or hTRT) protein subunit and an RNA template subunit known as TER or hTR (17,18). The hTR RNA subunit encodes ⬃1.5 telomeric repeats, and thereby acts as a template for the hTERT reverse transcriptase to add new telomeric sequences onto the chromosome ends. However, most human somatic cells no longer express telomerase and it is believed that the absence of this activity is a major contributing factor to their finite replicative life span (19,20). Recently, the life span of certain human somatic cell types (skin fibroblasts, endothelial cells, and retinal pigment epithelial cells) has been extended in culture by the ectopic expression of the human telomerase reverse transcriptase protein (21–25). The immortalized cells maintained their differentiated phenotype and did not acquire transformed or malignant characteristics (23,24). Interestingly, the extension of the life span of skin fibroblasts and retinal pigment epithelial cells by introduction of telomerase did not require any of the other components of the telomere-lengthening machinery such as telomere binding proteins or the RNA template component (21,22,25). These results imply that the life span of a wide variety of somatic cells can be extended simply by the ectopic expression of telomerase, PIERA-VELAZQUEZ ET AL and support the telomere hypothesis of cellular aging which proposes that most somatic cells become senescent because of progressive shortening of their telomeres with each cell division. Although the effects of expression of exogenous telomerase have been investigated in a number of human somatic cell types, there are no studies to date that have examined these effects in human articular chondrocytes. Given the importance of extending the proliferative capacity and delaying the occurrence of senescence in articular chondrocytes, we undertook studies to express the human telomerase protein subunit in OA articular cartilage chondrocytes by utilizing a retroviral complementary DNA (cDNA) expression construct. Herein, we report the successful transfer of this cDNA into human OA articular chondrocytes, and demonstrate that expression of the enzyme in these cells results in an extension of their telomere lengths and an increase in their replicative capacity in culture, without altering their chondrocyte-specific phenotype. PATIENTS AND METHODS Isolation and culture of chondrocytes. OA articular cartilage was obtained from patients undergoing knee replacement surgery in the Department of Orthopedic Surgery at Thomas Jefferson University Hospital. All tissues were procured following protocols reviewed and approved by the Institutional Review Committee in accordance with the National Organ Transplant Act and the Pennsylvania Organ Transplant Act. OA chondrocytes were obtained from 5 patients, a 60-year-old man (C203), a 57-year-old woman (C208), a 75-year-old woman (C209), a 69-year-old man (C223), and a 41-year-old man (C222). Articular cartilage was removed by careful dissection and precautions were taken to avoid inclusion of the subchondral bone. Chondrocytes were isolated from these tissues with the use of previously described procedures (26). Briefly, chondrocytes from OA human cartilage were obtained by excising the remaining macroscopically healthy cartilage and then mincing the tissue into small pieces with a razor blade. To remove adherent fibrous tissues, the cartilage was incubated in Dulbecco’s minimum essential medium (DMEM) containing 2 mg/ml of trypsin and bacterial collagenase (Worthington Biochemical, Lakewood, NJ) for 1 hour at 37°C. The medium was discarded and the tissue fragments were digested overnight at 37°C in DMEM containing 10% fetal bovine serum (FBS) and 0.5 mg/ml of bacterial collagenase. The cells released by the enzymatic digestion were filtered through a 70 nylon filter and collected by centrifugation at 250g for 5 minutes, and then resuspended and washed 4 times with collagenase-free medium. Chondrocytes were then cultured in plastic tissueculture flasks to allow their expansion as dedifferentiated cells. The cultures were maintained at 37°C in a 5% CO2 humidified atmosphere in DMEM containing 10% FBS, 2 mM EXOGENOUS EXPRESSION OF TELOMERASE FOR EXPANSION OF OA CHONDROCYTES 100 units/ml penicillin, 100 g/ml streptomycin, 1% (volume/volume) vitamin supplement, 2.5 g/ml Fungizone (Life Technologies, Carlsbad, CA), and 50 g/ml ascorbic acid. The medium was replaced every 3–4 days and the cells were passaged at 7–10-day intervals. The monolayers were subjected to enzymatic dissociation with trypsin at each passage and the cell numbers were determined by counting with a hemocytometer. To allow redifferentiation, the monolayers were dissociated as described above and aliquots of the cell suspensions were cultured for 7 days at a density of 5–10 ⫻ 106 cells in 60-mm petri dishes that had been previously coated with a 0.9% solution of poly-(2-hydroxyethyl-methacrylate) (polyHEMA) prepared as previously described (26). The experiments concerning the effect of basic fibroblast growth factor (bFGF) on chondrocyte proliferation were performed using 3 monolayer cultures of the C203 cell line cultured in the absence of bFGF or with either 5 ng/ml or 10 ng/ml bFGF (final concentrations). The medium was replaced every 3–4 days and the cells were passaged at 7-day intervals. Construction of a human telomerase retroviral vector and infection and selection of chondrocytes. The human telomerase cDNA was provided by Geron Corporation (Menlo Park, CA). The telomerase retroviral expression plasmid was constructed by inserting the human telomerase cDNA into the pLNCX vector from Clontech (Palo Alto, CA). To obtain retroviral vector particles containing the telomerase construct or control vector (pLNCX alone), standard protocols were used (27). Briefly, 15 g of control vector plasmid or telomerase construct was transfected into the amphotropic PT67 packaging cell line (Clontech) using the Profection calcium phosphate kit from Promega (Madison, WI). The cells were then selected by culturing them for 10 days in the presence of 350 g/ml of G418. Retrovirus was pooled from producing cells derived from the stably transfected PT67 cells. The supernatants were filtered through a 0.45-m filter and were mixed with DMEM and 8 g/ml polybrene (Sigma, St. Louis, MO) to infect chondrocytes cultured as monolayers. The chondrocytes were infected for 1 day, selected for 10 days with 350 g/ml G418, and then expanded and assayed for telomerase activity. In each case, chondrocytes were infected with pLNCX-telomerase retrovirus (referred to as hTERT chondrocytes) or with pLNCX retrovirus as a control (referred to as pLNCX chondrocytes). One hTERT- and 1 pLNCX-transduced clonal cell line was established from sample C222 by limiting dilution to 0.5–1 cells/well in 96-well plates containing 200 l medium/well. Medium was changed every 3–4 days until confluence, at which time the cell clones were expanded. Detection of telomerase activity by telomeric-repeat amplification protocol (TRAP) assay. Telomerase activity was detected with the use of the standard TRAPeze protocol (Intergen, Purchase, NY) and the assay was performed according to the manufacturer’s instructions. Briefly, using aliquots corresponding to 1 ⫻ 103 cells, polymerase chain reaction (PCR) was carried out with specific primers for the telomerase-extended product, using Taq polymerase. An aliquot equal to one-half volume (500 cells) of the PCR product was electrophoresed on 10% polyacrylamide gels and the PCR products were detected by staining with SYBR Green (BioWhittaker Molecular Applications, Rockland, ME) and a FluorImager (Molecular Dynamics, Sunnyvale, CA). Relative L-glutamine, 685 levels of activity were obtained by comparison with the activity in control HeLa cells. Assessment of telomere length by Southern analysis. Genomic DNA was extracted from the cells transfected with either the telomerase retroviral vector or with the empty vector using the DNeasy tissue system (Qiagen, Valencia, CA). Genomic DNA (3 g) was digested with Hinf I and Rsa I and electrophoresed on 0.6% or 0.7% agarose gels and transferred onto a nylon membrane by capillary blotting. The membrane was then hybridized with a 32P end-labeled telomere probe, (CCCTAA)3, utilizing a standard protocol, and exposed to a PhosphorImager screen (Molecular Dynamics) for 24 hours. Images were scanned by using ImageQuant software (Molecular Dynamics) and mean telomere lengths were calculated using the method of Harley et al (11). Assessment of chondrocyte proliferation and morphology. Every 7–10 days, the cultures were subjected to enzymatic dissociation with trypsin and the cell numbers were determined by counting an aliquot of the resulting suspension in a hemocytometer. The population doublings were calculated with the following equation: PD ⫽ Log10 (N/N0) ⫻ 3.33, where N is the number of cells at the end and N0 is the number of cells at the beginning of the experiment (28). The morphology of pLNCX and hTERT chondrocytes was examined by phase-contrast microscopy. Photomicrographs were obtained of cells that were cultured as monolayers for various passages or following their redifferentiation by culture on polyHEMA-coated dishes for 7 days. Analysis of chondrocyte gene expression patterns. The pattern of expression of genes encoding types I and II collagen and aggrecan was examined by reverse transcription (RT)– PCR analysis. Cells from either monolayer or polyHEMA cultures were harvested and total RNA extracted using the TRIzol reagent (Life Technologies). RT-PCR amplification of target messenger RNA (mRNA) was performed with 1 g of RNA by using the Titan One-tube RT-PCR system (Roche, Indianapolis, IN). PCR was performed using specific primers for human GAPDH (5⬘-GCCAGTGGACTCCACGACGT-3⬘ and 5⬘-ATGGGGAAGGTGAAGGTCGG-3⬘), type I collagen (5⬘-TAAAGGGTCACCGTGGCT-3⬘ and 5⬘-CGAACCACATTGGCATCA-3⬘), type II collagen (5⬘-AACAACCAGATTGAGAGC-3⬘ and 5⬘-TTACAAGAAGCAGACCGG-3⬘), and aggrecan (5⬘-AGCATCCACCCAGGTCT-3⬘ and 5⬘-AGAGAAGATTCTGGGTC-3⬘). The products were electrophoresed on 1% agarose gels stained with SYBR Green I. The reaction products were visualized with a FluorImager (Molecular Dynamics). Images were quantitated by using ImageQuant software (Molecular Dynamics) and the signals were normalized to those of GAPDH. RESULTS Establishment of culture conditions for optimal proliferation of human adult OA chondrocytes. To establish the optimal culture conditions for chondrocyte growth that would permit an appropriate level of retroviral infection and allow for subsequent selection, freshly isolated OA chondrocytes were cultured as monolayers in tissue-culture flasks. However, since human OA chon- 686 Figure 1. Effect of basic fibroblast growth factor (bFGF) on osteoarthritic (OA) chondrocyte proliferation. Growth curves are shown for OA chondrocytes (cell line C203 from a 60-year-old man) in monolayer that were left untreated (open squares) or treated with 5 ng/ml bFGF (open circles) or 10 ng/ml bFGF (solid triangles). drocytes do not readily proliferate, conditions that would stimulate them to divide in culture were examined. It has been previously demonstrated that both bFGF and TGF␤ are mitogenic for chondrocytes grown as monolayers (29). Also, the expansion of dedifferentiated bovine chondrocytes in the presence of bFGF enhances their ability to redifferentiate in 3-dimensional culture systems (30). Therefore, in order to determine whether bFGF stimulates the cell division of human OA chondrocytes, a requirement for retroviral infection, OA chondrocytes isolated from a 60-year-old patient (C203) were treated with varying doses of bFGF. The results, shown in Figure 1, demonstrate that bFGF can stimulate human OA chondrocytes to undergo increased cell division as compared with the untreated cells. However, the difference in cell proliferation was observed only after weekly subpassages conducted once per week for 4 weeks. The cells treated with 5 ng/ml bFGF increased 181-fold (5 million to 1.8 billion), for a total of 8.52 population doublings, and the 10 ng/ml bFGF–treated cells increased 758-fold (5 million to 7.58 billion), for a total of 10.5 population doublings. Since the effect of bFGF was only evident at later time points, it was decided not to use bFGF to stimulate growth of the chondrocytes unless the starting cell numbers were limiting. With respect to the 3 cell lines studied (C209, C223, and C222), C209 required bFGF to stimulate cell growth during the infection, selection, and early culture phases PIERA-VELAZQUEZ ET AL only. No bFGF was used to expand C223 or C222, the other cell lines studied. Generation of a human telomerase retroviral expression vector and determination of telomerase expression in chondrocytes. The human telomerase cDNA was provided by Geron Corporation. A telomerase expression plasmid was constructed by inserting the human telomerase cDNA into the pLNCX vector, which places it under control of the cytomegalovirus promoter (pLNCX-telomerase) (Figure 2). Packaging cell lines were produced by transfecting amphotropic PT67 cells (provided by Clontech) with either the pLNCX vector or the pLNCX-telomerase construct. PT67 cells were transfected using the calcium phosphate method, and the cells were selected by culturing them for 10 days in the presence of 350 g/ml of G418 (27). The endogenous level of telomerase expression was determined in uninfected OA chondrocytes, HeLa cells, and in the transfected PT67 packaging cells by utilizing a PCR-based TRAP assay (see Patients and Methods). Figure 3A shows that both the positive control HeLa cell extract (lane 1) and the extract from the PT67 cells stably transfected with the pLNCX-telomerase construct (lane 2) displayed telomerase activity. However, no telomerase activity was detected in OA chondrocytes (C208) that were grown as monolayers in either the absence (lane 3) or presence (lane 4) of bFGF. Furthermore, no telomerase activity was found in the C209 OA chondrocytes cultured either as monolayer (dedifferentiated) or on polyHEMAcoated dishes (differentiated) (lanes 5 and 6, respectively). TRAP assays were then performed on uninfected Figure 2. Schematic diagram of the human pLNCX-telomerase construct (pLNCXtelom). LTR ⫽ long terminal repeat; Amp(R) ⫽ ampicillin resistance gene; PSI ⫽ viral packaging signal; NEO(R) ⫽ neomycin resistance gene; CMV ⫽ cytomegalovirus. EXOGENOUS EXPRESSION OF TELOMERASE FOR EXPANSION OF OA CHONDROCYTES 687 Figure 3. Telomeric-repeat amplification protocol (TRAP) assay performed on human osteoarthritic (OA) chondrocytes and various cell lines. A, TRAP assays were performed on HeLa cells, PT67 retroviral packaging cells stably transfected with the pLNCX-hTERT construct, human OA chondrocytes isolated from a 57-year-old woman (C208) and treated with 0 ng/ml basic fibroblast growth factor (bFGF) (lane 3) or 5 ng/ml bFGF (lane 4) for 10 days, and OA chondrocytes isolated from a 75-year-old woman (C209) and grown as monolayer for 10 days (p1; lane 5) or cultured on poly-(2-hydroxyethyl-methacrylate)–coated dishes for 10 days (pH; lane 6). B, TRAP assays were performed on OA chondrocytes isolated from a 75-year-old woman (C209) and a 69-year-old man (C223) that were uninfected (U), infected with the pLNCX control retroviral vector (V), or infected with the pLNCX-hTERT retroviral vector (T). C, TRAP assays were performed on OA chondrocytes isolated from a 41-year-old man (C222) to serve as a control clonal line (C) and hTERT clonal line (T). IC ⫽ internal control; AS ⫽ after selection; P ⫽ passage number. C209 and C223 OA chondrocytes or C209 and C223 OA chondrocytes that had been transduced with the pLNCX vector or the pLNCX-hTERT construct (Figure 3B). No detectable telomerase activity was observed in either the uninfected (lanes 1 and 6) or pLNCX-transduced (lanes 2 and 7) chondrocyte cell lines. In contrast, both hTERT-transduced chondrocyte cell lines showed substantial levels of telomerase activity (lanes 3 and 8). Furthermore, it was found that the telomerase activity in these cells increased with subsequent passages (lanes 4, 5, and 9). Since the pooled cell lines that have been characterized thus far were most likely composed of cells with different expression levels of telomerase, clones of hTERT and control chondrocyte cell lines were established. At passage 6, chondrocytes isolated from the 41-year-old OA patient that had been infected with either the pLNCX-telomerase construct or the pLNCX control vector were plated at a density of 0.5–1 cell/well in 96-well plates. One hTERT and 1 control clonal cell line was expanded. Figure 3C shows that the C222 clonal hTERT cell line exhibited telomerase activity, whereas the control line was negative. Assessment of telomere lengths. Southern analyses were performed with digested genomic DNA from the OA chondrocyte cell lines to determine the effects of telomerase expression on telomere length. Analyses were performed with genomic DNA isolated from uninfected, pLNCX-transduced, and hTERT-transduced C209 and C223 chondrocytes immediately after selection with G418. The results, shown in Figure 4, indicate that in the C209 cell line, telomerase expression caused a detectable increase in telomere length. In contrast, the same analysis performed on the initial pool of the transduced C223 cell line failed to show any changes in telomere length as a result of telomerase expression. However, subsequent passaging of hTERT-transduced C223 cells resulted in longer telomeres than that seen in pLNCX-transduced cells (Table 1). The data shown in Table 1 represent the average from at least 2 different genomic DNA digestions analyzed by different hybridizations. Table 1 also shows the increase in telomere length of the hTERT C222 clonal line as compared with the control pLNCX C222 clonal cell line. Effects of telomerase expression on OA chondrocyte proliferation and population doubling levels. The effects of telomerase expression on chondrocyte proliferation and population doubling levels were examined in the 2 cell lines and in a clonal cell line during culture periods ranging from 80 days to 250 days. Cell line C209 688 PIERA-VELAZQUEZ ET AL Figure 4. Terminal restriction fragment lengths. Mean telomere lengths in the C209 and C223 cell lines were determined for uninfected osteoarthritic (OA) chondrocytes (U), OA chondrocytes infected with the pLNCX retroviral vector (V), or OA chondrocytes infected with the pLNCX-hTERT retroviral vector (T). AS ⫽ after selection; P ⫽ passage number. transduced with pLNCX alone reached senescence and failed to increase in cell number after 40 days in culture (Figure 5A). In contrast, C209 hTERT chondrocytes displayed a 3-fold increase in cell number by day 20, followed by a reduction in the cell number that reached baseline levels at 60 days. However, a remarkable increase in cell number was observed in subsequent periods so that at 80 days, a nearly 4-fold increase was observed. Cell growth continued so that by day 103, the cell number had increased 20-fold and by 201 days, by 40-fold (Figure 5A). Similarly, the behavior of the transduced C223 cell line showed that C223 pLNCXtransduced cells reached a state of senescence after 77 Table 1. Mean telomere lengths of osteoarthritic chondrocyte cell lines Mean telomere length, kb Treatment C209 C223 C222 clones Uninfected Early passage pLNCX hTERT Later passage pLNCX hTERT 8.8 10.4 10.5 8.1 9.9 9.5 9.5 11.5 9.6 10.5 10 13.3 Figure 5. Growth curves of the osteoarthritic (OA) chondrocyte cell lines A, C209, B, C223, and C, clonal C222, infected with pLNCX vector or pLNCX-hTERT. OA chondrocytes infected with pLNCX vector (open circles) or with pLNCX-hTERT (solid squares) were grown in monolayer culture, trypsinized, counted, and replated at lower density. At the end of the culture periods, the population doublings (PD) were calculated using the equation, PD ⫽ Log10 (N/N0) ⫻ 3.33 (see Patients and Methods). days in culture, while C223 hTERT-transduced cells still continued growing after 80 days and 16 passages. Population doubling levels were then calculated for the 2 cell lines C209 and C223, and the clones EXOGENOUS EXPRESSION OF TELOMERASE FOR EXPANSION OF OA CHONDROCYTES 689 Figure 6. Morphology of osteoarthritic (OA) chondrocyte cell lines cultured as monolayer or in poly-(2hydroxyethyl-methacrylate) (polyHEMA)–coated dishes. OA chondrocytes were infected with the pLNCX vector (A and B) or the pLNCX-hTERT construct (C and D) and were cultured as monolayer or redifferentiated on polyHEMA-coated dishes for 4 days (E and F). derived from the C222 chondrocytes. In contrast to the C209 pLNCX cells, which showed no detectable population doubling levels, the C209 hTERT cells exhibited 9 population doublings by the time the experiment was terminated at 234 days. Similar results were observed with the C223 hTERT cells, which exhibited 8 population doublings by 84 days of culture (Figure 5B), compared with C223 pLNCX cells, which exhibited only 3 population doublings. Interestingly, the C222 clonal cell lines exhibited a higher population doubling level than did the pooled cell lines. The C222 control clonal cells senesced after ⬃20 population doublings, whereas the hTERT clone continues to grow with ⬎28 population doublings to date (Figure 5C). Morphology and gene expression of telomeraseexpressing chondrocytes. Figure 6 shows the morphology of the C209 and C223 cell lines cultured either as dedifferentiated chondrocytes in monolayer or as redifferentiated chondrocytes in polyHEMA-coated dishes. The morphology of the hTERT chondrocytes was similar to that of the control cells (pLNCX chondrocytes) under both culture conditions, and reacquisition of the chondrocyte cell morphology was evident when the cells were redifferentiated in polyHEMA-coated dishes for up to 4 days (Figures 6E and F). In order to further verify that the expression of hTERT had not altered the phenotype of the chondrocytes, gene expression analysis of control cells and of cells expressing telomerase cultured as either monolayer (dedifferentiated) or in polyHEMA-coated dishes (redifferentiated) was performed. RT-PCR of total mRNA using human type I collagen–specific primers showed that both C209 hTERT cells and C223 hTERT cells that had redifferentiated in polyHEMA culture for 7 days expressed less type I collagen message (C209 at the ninth and nineteenth passage and C223 at the sixth and twelfth passage) than in plastic culture (Figure 7). We also detected a higher level of aggrecan expression in both lines of redifferentiated hTERT-infected chondrocytes; the difference observed by RT-PCR was 8.7-fold in C209 at passage 9 and 3.7-fold in C223 at passage 6 which increased to 5.2-fold in C223 at passage 12 (Figure 8). The level of type II collagen mRNA, as measured by RT-PCR, remained low in redifferentiated hTERTinfected chondrocytes (C223) after 6 or 12 passages on plastic followed by 7 days on polyHEMA-coated dishes (Figure 9). The low level of type II collagen mRNA observed could be due to the short period of time in which the cells were allowed to redifferentiate (7 days). However, as shown in Table 2, the ratio of type II 690 Figure 7. Reverse transcription–polymerase chain reaction (RTPCR) analysis of RNA isolated from osteoarthritic (OA) chondrocyte cell lines for COL1A1 mRNA expression. RT-PCR analysis was performed on RNA isolated from the C209 (A) and C223 (B) OA chondrocyte cell lines as described in Patients and Methods, with primers specific for human COL1A1 and GAPDH mRNA. Upper panels indicate the COL1A1 and GAPDH (as a control) RT-PCR products, and lower panels show the corresponding quantitation. Quantitation was performed with a FluorImager and the pixel volumes (vol) of the COL1A1 bands were normalized to those of GADPH, which were obtained in a separate RT-PCR reaction from the same RNA samples. P ⫽ passage number; pl ⫽ pLNCX-hTERT–infected chondrocytes as monolayer; pH ⫽ pLNCX-hTERT–infected chondrocytes as monolayer and then redifferentiated on poly-(2-hydroxyethylmethacrylate)–coated dishes for 7 days. collagen to type I collagen increased ⬃2.5-fold when C223 hTERT cells were transferred to polyHEMAcoated dishes for 7 days. PIERA-VELAZQUEZ ET AL Figure 9. RT-PCR analysis of RNA isolated from the C223 cell line for COL2A1 mRNA expression. RT-PCR analysis was performed on RNA isolated from the C223 OA chondrocyte cell line as described in Patients and Methods, with primers specific for human COL2A1 and GAPDH mRNA. Upper panels show the COL2A1 and GAPDH (as a control) RT-PCR products, and lower panels show the corresponding quantitation. Quantitation was performed with a FluorImager and the pixel volumes of the RT-PCR bands for COL2A1 were normalized to those of GADPH. pH (alone) ⫽ uninfected chondrocytes cultured on polyHEMA; P ⫽ passage number; pl ⫽ pLNCX-hTERT–infected chondrocytes as monolayer; pH (with passage number) ⫽ pLNCXhTERT–infected chondrocytes as monolayer and then redifferentiated on polyHEMA-coated dishes for 7 days. See Figure 7 for other definitions. DISCUSSION Figure 8. RT-PCR analysis of RNA isolated from OA chondrocyte cell lines for aggrecan mRNA expression. RT-PCR analysis was performed on RNA isolated from the C209 (A) and C223 (B) OA chondrocyte cell lines as described in Patients and Methods, with primers specific for human aggrecan and GAPDH mRNA. Upper panels show the aggrecan and GAPDH (as a control) RT-PCR products, and lower panels show the corresponding quantitation. Quantitation was performed with a FluorImager and the pixel volumes of the aggrecan bands were normalized to those of GAPDH, which were obtained in a separate RT-PCR reaction from the same RNA samples. pH (alone) ⫽ uninfected chondrocytes cultured on polyHEMA; P ⫽ passage number; pl ⫽ pLNCX-hTERT–infected chondrocytes as monolayer; pH (with passage number) ⫽ pLNCX-hTERT– infected chondrocytes as monolayer and then redifferentiated on polyHEMA-coated dishes for 7 days. See Figure 7 for other definitions. One basis for cellular aging in vitro and in vivo is the telomere length hypothesis. The telomere length hypothesis proposes that as primary somatic cells divide, their telomeric sequences shorten due to the lack of telomerase activity (13,15,31,32). When telomeres become shortened to a certain threshold length, the cells Table 2. Ratio of type II collagen (CII) to type I collagen (CI) mRNA expression in transduced C223 cells cultured as monolayer in plastic and following culture on polyHEMA for 7 days* CII:CI P6 monolayer P6 f pH† P12 monolayer P12 f pH† 0.48 1.18 1.29 3.48 Fold increase in CII:CI 2.5 2.7 * P ⫽ passage number. † Redifferentiated 7 days in poly-(2-hydroxyethyl-methacrylate) (polyHEMA)–coated dishes. EXOGENOUS EXPRESSION OF TELOMERASE FOR EXPANSION OF OA CHONDROCYTES respond by entering a metabolically active but nondividing state of senescence. Indeed, telomere length has been shown to correlate with replicative life span in vitro and somatic cells from older individuals possess a lower replicative potential and shorter telomeres (12). In vivo experiments also support the telomere length hypothesis. Mice deficient in telomerase activity (mTR⫺/⫺), due to deletion of the RNA component, show decreased viability after ⬃4–6 generations, depending on the genetic background (33). Additionally, mTR⫺/⫺ mice exhibit a reduced capacity to respond to stresses such as wound healing and ablation of the hematopoietic system (34). These results imply that loss of telomere length contributes to the changes observed during aging and age-related disease. Chondrocytes, similar to other primary somatic cell types, possess a limited replicative life span in vitro and invariably enter into the nondividing, metabolically active state of senescence (4,5,26). Studies have shown that the population doubling capacity of normal human articular chondrocytes is in the range of 35–40 and that lapine articular chondrocytes exhibit a relationship between the age of the animal and the replicative life span in vitro (4,5). During the aging process and during the development and progression of OA, articular cartilage loses both thickness and cellularity along with the development of changes in the cartilage matrix and metabolic function of the chondrocytes (9,10). The loss of cellularity has been postulated to be due to a decreased ability of the chondrocytes to proliferate or respond to mitogenic signals coupled with an increase in apoptosis (7,8). Indeed, Martin and Buckwalter (6) have recently shown that both mitotic potential and telomere length are diminished in human articular chondrocytes isolated from older individuals. We report here, for the first time, the effect of ectopic expression of the human telomerase reverse transcriptase in human articular chondrocytes. Articular chondrocytes isolated from patients undergoing knee replacement surgery for OA were cultured and infected with a human telomerase cDNA retroviral expression construct. Uninfected chondrocytes or chondrocytes infected with the retroviral vector alone did not display any detectable telomerase activity, whereas chondrocytes infected with the telomerase retroviral construct exhibited telomerase activity comparable with that of HeLa cells. Interestingly, the telomerase activity increased as the cells were passaged in culture, possibly indicating a selective growth advantage. Indeed, articular chondrocytes from a 75-year-old patient (C209) that were transduced with the hTERT expression vector were 691 able to grow for an additional 9 population doublings and exhibited increased telomere lengths as compared with control cells from the same donor. A second hTERT chondrocyte cell line, C223 (from a 69-year-old patient), grew for an additional 5 population doublings as compared with control cells, and this line still continues to grow. We also established clonal hTERT and control cell lines from the chondrocytes of a 41-year-old OA patient. The clonal control cell line ceased dividing after 20 population doublings, whereas the hTERT clonal line has undergone 28 population doublings and currently exhibits a population doubling rate of once every 3 days. The difference in population doubling levels achieved by the pooled cell lines in comparison with the C222 clonal lines is probably due to the heterogeneity of telomerase expression levels in the pooled lines. Furthermore, the expression of telomerase in these chondrocytes did not impede the ability of the cells to undergo the first stages of redifferentiation, as was evidenced by a reduction in type I collagen gene expression, an increase in aggrecan gene expression, and reappearance of a chondrocytic cell morphology after transfer of the monolayer hTERT chondrocytes to polyHEMA-coated dishes. Recently, several laboratories have reported the establishment of immortalized human primary cell lines that retain their differentiated phenotype and normal responses to external stimuli by engineering them to express the hTERT protein. Bodnar et al (21) first demonstrated that human foreskin fibroblasts and retinal pigment epithelial cells transfected with an hTERT cDNA expression vector acquired an increased population doubling capacity in comparison with their hTERTnegative counterparts. Later it was shown that the hTERT⫹ cell lines still retained their normal differentiated response to signals such as serum starvation, contact growth inhibition, and anchorage-dependent growth (23). Subsequently, normal human endothelial cells were immortalized by ectopic telomerase expression also without acquiring malignant characteristics (25). However, not all human primary cell types appear to have the ability to be immortalized only by the expression of telomerase. Human mammary epithelial cells and human keratinocytes also required the downregulation or inactivation of the p16INK4a tumor suppressor gene (35,36). Also, expression of hTERT in human epithelial cells resulted in cells that expressed higher levels of the cMyc protooncogene (37). Interestingly, telomerase expression has been shown to provide a beneficial effect on age-related 692 PIERA-VELAZQUEZ ET AL cellular processes such as apoptosis and skin regeneration. In a recent study, telomerase-expressing skin fibroblasts were compared with young or senescent fibroblasts in an in vivo model of skin growth by implantation into SCID mice (38). The senescent fibroblasts gave rise to skin that exhibited blistering and was more fragile than the skin derived from either young or telomeraseexpressing fibroblasts. In other studies, the inhibition of telomerase activity in pheochromocytoma cells and in primary embryonic hippocampal neurons was found to sensitize them to apoptosis-inducing insults such as oxidative stress and the effect of the cytotoxic amyloid ␤-peptide associated with Alzheimer’s disease (39,40). Finally, hTERT expression by human endothelial cells also induced resistance to conditions that normally result in apoptosis (25). These studies suggest that expression of telomerase in aged, diseased, or senescent cells may restore certain beneficial biologic functions. However, the long-term effect of ectopic telomerase expression by somatic cells in vitro and in vivo is unknown. Telomerase expression is associated with the progression of cancer and the endogenous hTERT gene has been shown to be up-regulated by the protooncogene cMyc (41). Therefore, aged or diseased somatic cells engineered to continually express telomerase for cell and gene therapy may have potential drawbacks, such as the occurrence of malignant transformation. Nevertheless, alternative strategies such as the transient introduction of telomerase or the use of inducible or tissue-specific promoters to drive expression of telomerase may allow for the controlled expansion of aged somatic cells ex vivo which can then be used for therapeutic purposes and for tissue engineering. In conclusion, the engineering of aged human OA chondrocytes to re-express telomerase may eventually represent a novel way to expand these cells ex vivo for use in the treatment of defects associated with the degeneration of cartilage during OA. ACKNOWLEDGMENT We are grateful to David Hawkins for excellent technical assistance. REFERENCES 1. Buckwalter JA, Mankin HJ. Articular cartilage repair and transplantation. Arthritis Rheum 1998;41:1331–42. 2. Brittberg M, Lindahl A, Nilsson A, Ohlsson C, Isaksson D, Peterson L, et al. Treatment of deep cartilage defects in the knee with autologous chondrocyte transplantation. N Engl J Med 1994;331:889–95. 3. Hayflick L. The limited in vitro lifetime of human diploid cell strains. Exp Cell Res 1965;37:614–36. 4. Evans C, Georgescu HI. Observations of the senescence of cells derived from articular cartilage. Mech Ageing Dev 1983;22: 179–91. 5. Adolphe M, Ronot X, Jaffray P, Hecquet C, Fontagne J, Lechat P. Effects of donor’s age on growth kinetics of rabbit articular chondrocytes in culture. Mech Ageing Dev 1983;23:191–8. 6. Martin JA, Buckwalter JA. Telomere erosion and senescence in human articular cartilage chondrocytes. J Gerontol A Biol Sci Med Sci 2001;56:B172–9. 7. Guerne PA, Blanco F, Kaelin A, Desgeorges A, Lotz M. Growth factor responsiveness of human articular chondrocytes in aging and development. Arthritis Rheum 1995;38:960–8. 8. Adams CS, Horton WE. Chondrocyte apoptosis increases with age in the articular cartilage of adult animals. Anat Rec 1998;250: 418–25. 9. Hamerman D. Aging and osteoarthritis: basic mechanisms. J Am Geriatr Soc 1993;41:760–70. 10. Abyad A, Boyer JT. Arthritis and aging. Curr Opin Rheumatol 1992;4:153–9. 11. Harley CB, Futcher AB, Greider CW. Telomeres shorten during aging of human fibroblasts. Nature 1990;345:459–60. 12. Allsopp RC, Vaziri H, Patterson C, Goldstein S, Younglai EV, Futcher AB, et al. Telomere length predicts replicative capacity of human fibroblasts. Proc Natl Acad Sci U S A 1992;89:10114–8. 13. Greider CW. Telomeres, telomerase and senescence. Bioessays 1990;12:363–9. 14. Bryan TM, Cech TR. Telomerase and the maintenance of chromosome ends. Curr Opin Cell Biol 1999;11:318-24. 15. Meyerson M. Role of telomerase in normal and cancer cells. J Clin Oncol 2000;18:2626–34. 16. Collins K. Mammalian telomeres and telomerase. Curr Opin Cell Biol 2000;12:378–83. 17. Lingner J, Hughes TR, Shevchenko A, Mann M, Lundblad V, Cech TR. Reverse transcriptase motifs in the catalytic subunit of telomerase. Science 1997;276:561–7. 18. Feng J, Funk WD, Wang SS, Weinrich SL, Avilion AA, Chiu CP, et al. The RNA component of human telomerase. Science 1995; 269:1236–41. 19. Kim NW, Piatyszek MA, Prowse KR, Harley CB, West MD, Ho PL, et al. Specific association of human telomerase activity with immortal cells and cancer. Science 1994;266:2011–5. 20. Shay JW, Bacchetti S. A survey of telomerase activity in human cancer. Eur J Cancer 1997;33:787–91. 21. Bodnar AG, Ouellette M, Frolkis M, Holt SE, Chiu CP, Morin GB, et al. Extension of life-span by introduction of telomerase into normal human cells. Science 1998;279:349–52. 22. Steinert S, Shay JW, Wright WE. Transient expression of human telomerase extends the life span of normal human fibroblasts. Biochem Biophys Res Commun 2000;273:1095–8. 23. Morales CP, Holt SE, Ouellette M, Kaur KJ, Yan Y, Wilson KS, et al. Absence of cancer-associated changes in human fibroblasts immortalized with telomerase. Nat Genet 1999;21:115–8. 24. Jiang X, Jimenez G, Chang E, Frolkis M, Kusler B, Sage M, et al. Telomerase expression in human somatic cells does not induce changes associated with a transformed phenotype. Nat Genet 1999;21:111–4. 25. Yang J, Chang E, Cherry AM, Bangs CD, Oei Y, Bodnar A, et al. Human endothelial cell life extension by telomerase expression. J Biol Chem 1999;274:26141–8. 26. Reginato AM, Iozzo RV, Jimenez SA. Formation of nodular structures resembling mature articular cartilage in long-term primary cultures of human fetal epiphyseal chondrocytes on a hydrogel substrate. Arthritis Rheum 1994;37:1338–49. 27. Cepko C, Pear W. Transduction of genes using retrovirus vectors. In: Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, EXOGENOUS EXPRESSION OF TELOMERASE FOR EXPANSION OF OA CHONDROCYTES 28. 29. 30. 31. 32. 33. 34. Smith JA, et al, editors. Current protocols in molecular biology. New York: John Wiley & Sons; 1997. p. 9.9.1–9.14.5. Cristofalo VJ, Allen RG, Pignolo RJ, Martin BG, Beck JC. Relationship between donor age and the replicative lifespan of human cells in culture: a reevaluation. Proc Natl Acad Sci U S A. 1998;95:10614–9. Bradham DM, Horton WE. In vivo cartilage formation from growth factor modulated articular chondrocytes. Clin Orthop 1998;352:239–49. Martin I, Vunjak-Novakovic G, Yang J, Langer R, Freed LE. Mammalian chondrocytes expanded in the presence of fibroblast growth factor 2 maintain the ability to differentiate and regenerate three-dimensional cartilaginous tissue. Exp Cell Res 1999;253:681–8. Autexier C, Greider CW. Telomerase and cancer: revisiting the telomere hypothesis. Trends Biochem Sci 1996;10:387–91. Harley CB. Telomere loss: mitotic clock or genetic time bomb? Mutat Res 1991;256:271–82. Herrera E, Samper E, Martin-Caballero J, Flores JM, Lee HW, Blasco MA. Disease states associated with telomerase deficiency appear earlier in mice with short telomeres. EMBO J 1999;18: 2950–60. Rudolph KL, Chang S, Lee HW, Blasco M, Gottlieb GJ, Greider C, et al. Longevity, stress response, and cancer in aging telomerase-deficient mice. Cell 1999;96:701–12. 693 35. Kiyono T, Foster SA, Koop JI, McDougall JK, Galloway DA, Klingelhutz AJ. Both Rb/p16INK4a inactivation and telomerase activity are required to immortalize human epithelial cells. Nature 1998;396:84–8. 36. Dickson MA, Hahn WC, Ino Y, Ronfard V, Wu JY, Weinberg RA, et al. Human keratinocytes that express hTERT and also bypass a p16INK4a-enforced mechanism that limits life span become immortal yet retain normal growth and differentiation characteristics. Mol Cell Biol 2000;20:1436–47. 37. Wang J, Hannon GJ, Beach DH. Risky immortalization by telomerase. Nature 2000;405:755–6. 38. Funk WD, Wag CK, Shelton DN, Harley CB, Pagon GD, Hoeffler WK. Telomerase expression restores dermal integrity to in vitroaged fibroblasts in a reconstituted skin model. Exp Cell Res 2000;258:270–8. 39. Fu W, Begley JG, Killen MW, Mattson MP. Anti-apoptotic role of telomerase in pheochromocytoma cells. J Biol Chem 1999;274: 7264–71. 40. Zhu H, Fu W, Mattson MP. The catalytic subunit of telomerase protects neurons against amyloid ␤-peptide induced apoptosis. J Neurochem 2000;75:117–24. 41. Cerni C. Telomeres, telomerase, and myc: an update. Mutat Res 2000;462:31–47.