Telomerase activity in HeLa cervical carcinoma cell ... - Springer Link

2 downloads 33 Views 312KB Size Report
immortal HeLa cell line express telomerase, an enzyme that prevents telomere shortening. Although immortal, the existence of non-dividing cells that do not ...
Biogerontology (2007) 8:163–172 DOI 10.1007/s10522-006-9043-9

R E S E A R C H A RT I C L E

Telomerase activity in HeLa cervical carcinoma cell line proliferation Milena Ivankovic´ Æ Andrea C´ukusˇic´ Æ Ivana Gotic´ Æ Nikolina Sˇkrobot Æ Mario Matijasˇic´ Æ Denis Polancˇec Æ Ivica Rubelj

Received: 9 June 2006 / Accepted: 7 August 2006 / Published online: 6 September 2006  Springer Science+Business Media B.V. 2006

Abstract Normal human somatic cells in culture have a limited dividing potential. This is due to DNA end replication problem, whereby telomeres shorten with each subsequent cell division. When a critical telomere length is reached cells enter senescence. To overcome this problem, immortal HeLa cell line express telomerase, an enzyme that prevents telomere shortening. Although immortal, the existence of non-dividing cells that do not incorporate 3H-thymidine over 24 h of growth has been well documented in this cell line. Using DiI labeling and high-speed cell sorting, we have separated and analyzed fractions of HeLa cells that divided vigorously as well as those that cease divisions over several days in culture. We also analyzed telomerase activity in separated fractions and surprisingly, found that the fraction of cells that divided 0–1 time over 6 days in culture have several times higher endogenous telomerase activity than the fastest dividing fraction. Additionally, the non-growing ´ ukusˇic´ Æ I. Gotic´ Æ N. Sˇkrobot Æ M. Ivankovic´ Æ A. C I. Rubelj (&) Department of Molecular Biology, Ru der  Bosˇkovic´ Institute, Bijenicˇka 54, 10000 Zagreb, Croatia e-mail: [email protected] M. Matijasˇic´ Æ D. Polancˇec Biology, Flow Cytometry and Cell Sorting Lab, PLIVA—Research Institute Ltd, Prilaz Baruna Filipovic´a 29, 10000 Zagreb, Croatia

fraction regains an overall high labeling index and low SA-b-Gal activity when subcultured again. This phenomenon should be considered if telomerase inhibition is to be used as an approach to cancer therapy. In this paper we also discuss possible molecular mechanisms that underlie the observed results. Keywords DiI Æ Flow cytometry Æ HeLa Æ Senescence Æ Telomerase

Introduction Normal human somatic cells undergo a finite number of divisions in culture after which they enter a non-replicative state called cellular senescence (Hayflick and Moorhead 1961). Cell senescence is broadly defined as the physiological program of terminal growth arrest, which can be triggered by alterations of telomeres or by different forms of stress. It is thought to provide a barrier against the unlimited proliferation and occurrence of cancer (Campisi 2000). Senescent cells differ from proliferating ones in morphology and biochemical characteristics. They are enlarged, flattened, more granulated but viable and metabolically active. They are arrested in the G1 phase of the cell cycle and cannot be stimulated into additional division by growth factors (Cristofalo et al. 1992). They also expose activity

123

164

of endogenous b-galactosidase at pH 6 known as SA-b-Gal activity (senescence associated b-Gal activity) (Dimri et al. 1995), and accumulate lipofuscin (Von Zglinicki et al. 1995). It is also well established that cell senescence is determined by number of population doublings (PD), not time in culture, and to date the best described counting mechanism for replicative senescence involves the telomere-shortening hypothesis. Due to the DNA end replication problem and additional exonuclease degradation of the 5¢ strand at telomere ends, telomeres shorten with each subsequent cell division (Olovnikov 1973; Harley et al. 1990; Makarov et al. 1997). In addition, telomere shortening can be accelerated by external factors such as damage by free radicals or oxidative stress (Touissaint et al. 2000; Saretzki and Von Zglinicki 2002). When at least one telomere reaches a critical length in normal cells it triggers irreversible growth arrest (Hemann et al. 2001). Unlike most somatic cells, cancer cells efficiently overcome this evolution-determined life barrier and continue to divide indefinitely. Most of them (~85%) have active telomerase, an enzyme that extends telomeres and subsequently prevents cell senescence (Counter et al. 1992; Shay and Bacchetti 1997). Other cancer derived cell lines, however, do not show telomerase activity but use recombination based alternative lengthening of telomeres, or the ALT mechanism (Bryan et al. 1995; Neumann and Reddel 2002). However, tumor cells can be readily induced to undergo senescence by genetic manipulations or by treatment with chemotherapeutic drugs, radiation or differentiating agents (Roninson 2003). Nevertheless, spontaneous appearance of senescent cells has been documented in some tumor populations (Pereira-Smith and Smith 1981; Rubelj et al. 1997). To examine this phenomenon in HeLa cell line, fractions of cells with high or low dividing capacities were separated from cycling cultures by DiI staining and subsequent separation on a high-speed flow cytometer–cell sorter. Isolated fractions were subcultured again and analyzed for growth capacity and telomerase activity. The obtained results suggest that variations in the level of telomerase activity among individual cells may play an important role

123

Biogerontology (2007) 8:163–172

in spontaneous appearance of senescent cells in HeLa cell line.

Materials and methods Cell lines and culture conditions The human cervical carcinoma epithelioid cell line HeLa was obtained from the American Type Culture Collection (ATCC). Cells were grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) at 37C in a humidified incubator with 5% CO2. Depending on experimental design cells were subcultured at a ratio between 1:2 and 1:10. SA-b-galactosidase staining and 3H-thymidine labeling index To determine the 3H-thymidine labeling index, ~104 cells were seeded into 30 mm Petri dish and incubated at 37C, 5% CO2. After ~24 h, 3 H-thymidine was added at a concentration of 10 lCi/ml and incubation continued for another 24 h. Following this treatment, cells were fixed in 1% glutaraldehyde for 10 min, washed in PBS 2 · 5 min and stained for SA-b-gal activity at pH 6.0 over 16–18 h as described (Dimri et al. 1995). For autoradiography, cells were washed 2 · 5 min in PBS, 2 · 5 min in 70% ethanol and dried at room temperature for several hours. In a dark room, fixed cells were overlaid with liquid photographic emulsion (Ilford Scientific Product), wrapped in aluminum foil and stored at 4C for 48 h. Preparations were processed in standard Kodak developer and universal fixer. More than 1,500 cells were scored for statistical analysis. DiI staining and flow cytometry For fluorescent staining, an appropriate number of cells (1–3 · 106) were trypsinized and spun down at 1,000 rpm in a swing-rotor laboratory centrifuge. Cells were than washed twice in PBS, resuspended in 10 ml of serum free DMEM medium containing 5 lM solution of DiI

Biogerontology (2007) 8:163–172

(1,1¢-dioctadecyl-3,3,3¢,3¢ tetramethylindocarbocyanineperchlorate (‘‘DiI’’; DiIC18(3)), Molecular Probes, catalog no. D282) and incubated for 20 min at 37C. To remove excess of dye, cells were washed twice and cultured in the dark, wrapped in aluminum foil at 37C, 5% CO2. After 6 days in culture at ~80% confluence cells were trypsinized, washed twice in freshly prepared washing/staining buffer (PBS (Sigma) containing 2% FBS (Sigma) and 0.02% EDTA (Sigma) and resuspended in the same buffer at final concentration of 2 · 107 ml–1. Cells were kept on ice until sorting. Positive control cells were stained and washed the same way on the day of sorting and placed on ice until sorting. Negative control cells were trypsinized and without staining resuspended in washing/staining buffer. All cells were analyzed and sorted on a MoFlo high-speed cell sorter (DakoCytomation) using Summit software (DakoCytomation), 488 nm argon-ion laser tuned on 125 mW of power was used as a source of excitation. Three measured parameters, FSC (forward-scattered light) and SSC (side-scattered light) were detected with linear signal amplification (Fig. 3a) while DiI specific fluorescence emission was detected on the FL2 channel using a 570/30 dichroic emission filter and logarithmic signal amplification. Two separate cell populations (L fraction and R fraction) were sorted simultaneously into 14 ml Falcon tubes filled with washing/staining buffer. Collected cells were centrifuged and resuspended in complete culture medium for further analysis. Telomerase activity Telomerase activity was assayed using the TRAPeze Telomerase Detection Kit (Intergen). Appropriate number of cells (~106) was lysed in CHAPS lysis buffer and primer mix along with TRAPeze reaction mixture was added. This mixture was incubated for 30 min at 30C to allow telomerase-dependent elongation of TS primer. PCR amplification was performed under following conditions: 90C/90 s following 30 cycles of 94C/30 s, 50C/30 s, 72C/30 s. Elongated and amplified telomerase products were resolved on 12.5% polyacrylamide gel. Intensities of internal

165

standard and telomerase product bands were quantitated by Image Master VDS Software, version 2.0. Telomerase activity in TPG (total product generated) units was calculated by comparing the ratio of telomerase products to an internal standard for each lysate, as instructed by manufacturer. TPG is defined as the number of TS primers extended by at least four telomeric repeats through telomerase activity in an extract, during 10-min incubation at 30C. The amount of telomerase product was quantified using the following formula: TPGðunitsÞ ¼ ðx  x0 Þ=cjðr  r0 Þ=cR  100; where x is non-heat-treated sample extracts, x0 is heat-treated sample extracts, r is quantity control, r0 is negative control, c is internal standard of non-heat-treated sample extracts and cR is internal standard of quantity control.

Results 3

H-thymidine labeling index and SA-b-Gal activity Most normal human cell cultures gradually accumulate non-dividing senescent cells over time, until ultimately the entire culture ceases further divisions (Hayflick 1965). Consequently, even young cultures contain a fraction of senescent cells. Among the most obvious properties of senescent cells are their inability to incorporate 3 H-thymidine, cell size enlargement and a strong SA-b-Gal activity (Hayflick and Moorhead 1961; Dimri et al. 1995; Rubelj et al. 2002). Senescence differs from non-dividing quiescent state, in that quiescent cells retain their ability to divide upon subcultivation, they have a normal young phenotype and lack of SA-b-Gal activity (Campisi 1992). It has been described that a small fraction of cells demonstrating typical senescent phenotype can spontaneously appear in some immortal cell lines including HeLa (Pereira-Smith and Smith 1981; Te Poele et al. 2002). Although intriguing, this phenomenon has not been extensively studied. Using 3H-thymidine labeling index and SA-b-Gal activity along with the presence of

123

166

Biogerontology (2007) 8:163–172

senescence specific morphology, we followed the appearance of senescent cells in HeLa cell line. Four morphologies of cells were observed (Fig. 1a, b). (1) As expected most cells showed a typical immortal phenotype [3H]dT+/SA-b-Gal– and were present at 90.11%. (2) The second largest group with a phenotype [3H]dT–/SA-bGal– included 9.47% cells and they most likely represent cycling cells with delayed entrance into S phase over the period of labeling as was observed previously with normal and precrisis human fibroblasts (Rubelj et al. 2002). (3) The smallest fraction of cells with [3H]dT+/SA-b-Gal+ phenotype were present at 0.15%. Although dividing, these cells showed significant SA-b-Gal activity which could be explained either as sensitivity to stress during subcultivation or cells that have entered spontaneous senescence after their last division (see Discussion). (4) A fraction of

Fig. 1 Tritiated thymidine labeling index and SA-b-Gal activity in HeLa cells. Four different phenotypes were observed (a): [3H]dT +/ SA-b-Gal–; enlarged [3H]dT–/SA-b-Gal+ (arrow); [3H]dT+/SA-bGal+; [3H]dT–/SA-b-Gal– (arrow). At least 1,500 cells per sample were counted for statistical analysis (b)

cells with a typical senescent phenotype [3H]dT–/ SA-b-Gal+ accompanied by significant cell size enlargement was present at 0.28%. Although small, this fraction represented an interesting phenomenon of spontaneous senescence of immortal cells which may have significant scientific importance. In order to separate and analyze these non-dividing/senescent cells, as well as the fastest growing fraction of the same culture, we employed a recently developed technique that employs DiI staining and high-speed cell sorting (Ferenac et al. 2005). Cell sorting and isolation of fractions In order to separate fractions of fast growing and non-dividing/senescent cells from HeLa culture, 1–3 · 106 cells were stained with DiI (Ferenac et al 2005; Ledley et al. 1992). Cells were then

(a)

120

(b) 90,11

percentage of cells

100 80 60 40 20

9,47 0,28

0,15

0 3

123

H-timidin + SA-β-Gal —

3

H-timidin — SA-β -Gal —

3

H-timidin — SA -β -Gal +

3

H-timidin + SA -β -Gal +

Biogerontology (2007) 8:163–172

167

(R fraction in Fig. 3c) and fastest growing cells (L fraction in Fig. 3c). For each fraction gating was restricted to 10% of total number of sorted cells (Fig. 3c). Control cells were cultured in parallel with DiI labeled cells and on the day of sorting were divided into two samples. One was stained and used as positive control and the other remained unstained and used as negative control (Fig. 3b). Isolated fractions demonstrated high efficiency and purity (>97%) of separation. 3

H-thymidine labeling index and SA-b-Gal activity of separated fractions Fig. 2 HeLa cells stained with DiI and DRAQ5 visualized by confocal microscopy. DiI is incorporated into membranes (pseudo green) and DRAQ5 is incorporated into nucleus (red). Reduced DiI intensity in divided cells compared to undivided cell is visible. Cells were optically sectioned and are displayed as maximal projection image

cultured for 6 days until ~80% confluency and submitted to flow cytometry analysis and separation as described in Materials and methods. DiI integrates into the outer and inner cell membranes and while incorporated, fluoresces orangered under green light (Fig. 2). It has an absorption and fluorescence emission maximum separated by about 65 nm, facilitating fluorescent detection efficiency. DiI labeling does not affect cell viability or basic physiological properties, nor transfer to neighboring cells or release to surrounding media. Therefore, as labeled cells divide, each daughter cell inherits approximately 50% of the dye (Fig. 2). This phenomenon allows flow cytometry analysis of different labeling intensities of cultured cells and differential sorting of individual cells that have undergone various numbers of divisions over the period of growth. Depending on variation in size of a particular cell type, fluorescent signals detected from successive generations of DiI labeled cells were broad and overlapped to some extent. However, HeLa cells are satisfactorily uniform in size so that it was possible to clearly distinguish non-labeled controls from maximum intensity stained cells (Fig. 3b). These properties of HeLa cells and high quality sorting resulted in clear separation of two distinct fractions of non-dividing/senescent

Following separation of fast and non-growing fractions we examined their viability, dividing capacity and presence of cells with a senescent phenotype. For this purpose a sample of 5 · 104 cells from each fraction was further subcultured to determine their 3H-thymidine labeling index and SA-b-Gal activity (Fig. 4). It was expected that a subpopulation of intensively dividing cells (L fraction in Fig. 3c) would demonstrate a high percentage of typical dividing phenotype [3H]dT+/SA-b-Gal– but instead this percentage decreased to 71.78% compared to total cell culture (90.11%). At the same time [3H]dT–/SA-bGal– phenotype significantly increased from 9.47% in total culture before separation to 28.22% in this fraction (Fig. 4). These cells most probably have a delayed entrance into S phase due to stress caused by high-speed cell sorting treatment during which they were exposed to intense aeration, high frequency vibrations and a strong electro-magnetic field (see Discussion). It is significant that there were no cells with SA-bGal+ phenotype in this fraction. In contrast, the other subpopulation of non-dividing/senescent cells (R fraction in Fig. 3c) showed a surprisingly high overall 3H-thymidine incorporation, over 66%. Although this is significant decrease compared with whole culture (90.11%), we should include the fact that these cells divided only 0–1 times over 6 days in culture before sorting. Most of the cells (66.44%) had typical dividing phenotype [3H]dT+/SA-b-Gal– but there was also a small fraction (0.77%) with a [3H]dT+/ SA-b-Gal+ phenotype. However, like in the case of the L fraction, [3H]dT–/SA-b-Gal– phenotype

123

168

Biogerontology (2007) 8:163–172

Fig. 3 Gating strategy and sorted cells analysis. a Polygonal region was drawn around cells according to their morphological properties on a FSC/SCC dot plot. Live cells are gated while dead cells, cell debris and clusters were excluded from further analysis and sorting. b Fluorescence intensity of cells stained on the day of experiment as positive control (shaded region) and unstained cells as negative control (unshaded region). c

90 80

71,78

L-fraction R-fraction

66,44

percentage of cells

Fig. 4 Tritiated thymidine labeling index and SA-b-Gal activity of L- and R-sorted fractions. Four different phenotypes were observed: [3H]dT+/ SA-b-Gal–; enlarged [3H]dT–/SA-b-Gal+; [3H]dT+/SA-b-Gal+; [3H]dT–/SA-b-Gal–. At least 1,500 cells per sample were counted for statistical analysis

Sort decisions were made combining live cell gate (see a in this figure) with L region (most dividing cells; blue) for one sort direction and R region (non-divided cells; red) for another sort direction. d Intensity of DiI fluorescence of R fraction after additional 3 days of growth in culture: shaded region R fraction; broken line unstained HeLa cells (negative control); continuous line DiI stained HeLa cells (positive control)

70 60 50 40

28,22 29,48

30 20 10

3,31 0

0

0,77

0 3

123

H-timidin+ SA-β-Gal—

3

H-timidin— SA-β-Gal—

3

H-timidin— SA-β-Gal+

3

H-timidin+ SA-β-Gal+

Biogerontology (2007) 8:163–172

Telomerase activity in separated fractions Telomere shortening is a well known mechanism for cell growth control. In order to continue indefinite divisions all immortal cells, single cell eukaryotes and stem cells must maintain their telomeres above a critical length, mostly through telomerase expression (Kim et al. 1994; Wright et al. 1996). Like most immortal cell lines, HeLa cells also express an active telomerase (Bryan et al. 1998; Morin 1989). Since previous results demonstrated that normal human fibroblasts immortalized by ectopic expression hTERT contain a fraction of senescent cells (Gorbunova et al. 2003) which surprisingly, have increased telomerase activity and similar effects were observed with ectopic expression of hTERT in HeLa cells as well (Goodwin and DiMaio 2001), we decided to examine telomerase activity in both sorted fastand slow-dividing fractions. Telomerase activity was measured by PCR based TRAP method (Fig. 5a, see Materials and methods). We found that telomerase activity was about four times higher in R fraction compared to L fraction or control whole HeLa population; 1600.3 TPG units versus 378.0 TPG units in L fraction and 395.7 TPG units in control culture (Fig. 5b). Therefore, we speculate that, similar to normal fibroblasts expressing catalytic subunit of telomerase, variations in telomerase activity among individual HeLa cells could contribute to their temporary cell cycle arrest or even entering spontaneous senescence.

M

(a)

L

Lhi

R

Rhi

H

Hhi -C TSR8

Trap-products (50 pb) Internal control (36 pb)

2500

(b)

2250 1600,3

2000 1750

TPG - units

increased to 29.48% in this group as well. An expected increase in typical senescent phenotype [3H]dT–/SA-b-Gal+ in the R fraction reached only 3.31%. This is more than 12 times greater percentage than that observed in control culture (0.28%) but nonetheless not as abundant as one would expect. However, these results demonstrate that a great majority of ‘‘non-dividing’’ cells retained their ability to divide when subcultured again after separation. Therefore most of the nondividing cells in HeLa culture, detected by 3Hthymidine labeling, actually represent cells with delayed entrance into S phase, even after 6 days in culture, rather than genuine senescent cells. Only a small portion of them (~0.3%) enter true spontaneous senescence.

169

1500 1250 1000 750 500

378,0

395,7

L

HeLa

250 0 R

Fig. 5 TRAP assay. a Polyacrilamide gel of TRAP reaction: L reaction from fast-dividing fraction, R reaction from non-dividing fraction, H reaction from whole HeLa population; Lhi heat inactivated protein extract from fastdividing fraction, Rhi heat inactivated protein extract from non-dividing fraction, Hhi heat inactivated protein extract from whole HeLa population; –C negative control, TSR8 positive control, M pBR322 HaeIII, Roche marker V. b TPG-unit values: L fast-dividing fraction, HeLa whole HeLa population, R non-dividing fraction. TPG is defined as the number of TS primers extended by at least four telomeric repeats through telomerase activity in an extract, during 10-min incubation at 30C

Additional growth and sorting of R fraction cells Since initial sorting demonstrated that among non-dividing cells there are total ~33% of cells that do not incorporate radioactive thymidine, we decided to prolong growth of this fraction in order to see if they could be separated from the rest of the cycling cells. Therefore, a sample of cells from R fraction was cultured for additional 3 days and analyzed by flow cytometry. We expected to see two distinctive profiles: namely cells that continue to divide would form a peak close to the negative control, whereas cells that permanently ceased

123

170

divisions would form a peak close to the positive control. Instead, analyzed cells did not show the expected bimodal distribution of fluorescence signals which means that virtually all cells ([3H]dT+/SA-b-Gal– and [3H]dT–/SA-b-Gal– together represent ~96% of cells) continue to divide upon subcultivation. It should be noted that the population of senescent cells is too small to be distinguished if all remaining cells divide three more times during an additional 3 days in culture and overgrow these cells, such that the percentage of completely senescent cells drop way below 3.31% (to ~0.41%) (Fig. 3d). These results also confirm that the [3H]dT–/SA-b-Gal– phenotype represent cells with delayed entrance into S phase rather than genuine senescent cells, which support similar findings in human diploid cell strain WI-38 during in vitro aging (Matsumura et al. 1979). DiI fluorescence intensity was clearly shifted further toward lower values demonstrating additional divisions.

Discussion It is well known that tumor cell cultures and human SV-40 immortalized cell lines contain a fraction of cells with reduced dividing capacity. This phenomenon is considered to be a consequence of genomic instability which activates mechanisms of cell cycle arrest (Pereira-Smith and Smith 1981). In a study of HeLa cells it was demonstrated that this cell line also contains a non-proliferating cell population (Martinez et al. 1998). When seeded at clonal density, these nonproliferating cells went through few cell divisions before developing characteristic senescent phenotypes. To determine the percentage of non-proliferating cells in HeLa cell line we used 3H-thymidine labeling. Dividing cells incorporate radioactive thymidine and those that did not divide would be unlabeled. The labeling index of HeLa cells observed in this study was similar to earlier reports; ~90% of cells divided and ~10% did not synthesize DNA during a typical labeling period of 24 h (Dimri et al. 1995). The absence of DNA synthesis does not distinguish senescent cells from those in temporary growth arrest or terminally

123

Biogerontology (2007) 8:163–172

differentiated cells; therefore we used SA-b-Gal activity, a reliable biomarker for senescent phenotype to further analyze these non-dividing HeLa cells. The percentage of cells with significant SA-b-Gal activity was 0.28% of total culture, which is close to previously observed percentage (~0.1%) for this cell line (Dimri et al. 1995). Therefore, not all non-proliferating cells represent a senescent population and the period of 24 h may not be long enough to distinguish between temporary and permanent cell growth arrest. In order to specifically analyze these non-dividing cells we used DiI labeling and high-speed FACS sorting to separate them from the rest of the culture. We assume that ~6 days of growth is a period long enough to clearly distinguish fast growing from non-growing fractions of cells. The presence of a small fraction (0.15%) of cells positive for both radioactive labeling and SA-b-Gal activity is probably caused by stress due to manipulation during the experimental procedure although we cannot rule out possibility that these cells just completed their last division before entering senescence. Following cell sorting, we compared 3H-thymidine labeling index and SA-b-Gal activity in both fast growing and non-dividing fraction of cells. The fast growing L fraction showed an overall lower percentage of cells with incorporated 3H-thymidine (~72%) over 24 h which might be explained by stress caused during the sorting procedure and indicates that for full recovery of growth the cells need a few days. It should be emphasized that this fraction did not contain SA-b-Gal positive cells, thus the experimental procedure did not induce cell damage that might cause endogenous b-galactosidase reaction. On the other hand the R fraction showed higher overall percentage of dividing cells than expected (~67%) and nearly 30% of all cells in this fraction were 3H-thymidine–/SA-b-Gal–. Cells with 3H-thymidine–/SA-b-Gal+ phenotype increased about 12 times which still represent only 3.31% (vs. 0.28% in untreated culture), indicating that all cells with senescent phenotype were collected in this fraction. Consequently, it appears that the vast majority of cells in R fraction (~96.7%) were in temporary growth arrest rather than in pre/senescent stage although they

Biogerontology (2007) 8:163–172

achieved only 0–1 divisions during 6 days in culture. It should be noticed that the SA-b-Gal+ phenotype present in this fraction is not consequence of stress induced by experimental procedure since this phenotype was not found in fast growing L fraction. To determine if the observed ~33% of nondividing cells in R fraction are in permanent growth arrest, we prolonged growth of this fraction for three more days which should result in a more obvious separation if two distinct subpopulations are present regarding their dividing capacity. Additional growth and sorting revealed that virtually all the cells continued to divide and that there was no significant fraction of permanently arrested cells in this ‘‘non-dividing’’ fraction (Fig. 3d). The fact that 3.31% of 3Hthymidine–/SA-b-Gal+ cells in this fraction were not visible upon this additional sorting indicates that they were overgrown by dividing cells so that their percentage dropped to ~0.4%. One conclusion therefore is that the vast majority of ‘‘nonproliferating’’ cells in HeLa cell culture represent temporarily arrested cells rather than genuine spontaneous senescence. Since there are many serious attempts to develop therapy against cancer based on telomerase inhibition (Corey 2002), this behavior of telomerase positive immortal cells should be taken under consideration. This also could contribute to explanations of delayed effects of telomerase inhibitors on growth of various cancer cells, including HeLa: these cells often undergo more than 20 divisions before they cease proliferation in culture (Feng et al. 1995; Corey 2002). Our results suggest that subpopulation of cells can avoid influence of telomerase inhibition over longer period of time through absence of cell divisions and increase of telomerase activity. These cells can resume divisions when conditions changes which could make therapy through telomerase inhibition very difficult if not impossible. Unlike previous reports (Martinez et al. 1978; Pereira-Smith and Smith 1981) when individual subclones of senescent cells were observed in HeLa cell line, we used mass cultures for growth and separation which provided more dense culture conditions for subsequent subpopulations as well as enabled telomerase activity analysis, an

171

important clue which helps to explain observed temporary or permanent growth arrest. The most common mechanism through which cells maintain permanent divisions is activation of telomerase. Since HeLa cells also express telomerase and individual variations in telomerase activity and telomere lengths among individual subclones has been observed (Bryan et al. 1998), we determined telomerase activity in both fast growing and non-growing fractions as well as in untreated HeLa population using a PCR based TRAP assay. We found that non-growing fraction had about four times higher telomerase activity than the fast growing fraction or untreated culture. Similar results were obtained in normal fibroblasts as well as HeLa cells expressing ectopic telomerase catalytic subunit (Goodwin and DiMaio 2001; Gorbunova et al. 2003). After expression of hTERT and prolonged lifespan of three normal human fibroblast cell lines, it was observed that they contain 3–20% of senescent cells. Telomerase activity was elevated in these cells, but telomere lengths were not changed compared with dividing cells of the same culture. These results demonstrate that high telomerase activity can cause a senescent like phenotype, but the mechanisms underlining this phenomenon are still unclear. As our results demonstrate, in HeLa cells this growth arrest is temporary and can last for at least 6 days in culture after which cells can restore their full dividing capacity when subcultured again. What still remains to be explained, is what causes the observed drastic variations in telomerase activity among HeLa cells and why cells with high telomerase activity arrest cell division. Acknowledgments We thank Dr. Olivia Pereira-Smith, Sam and Ann Barshop Center for Longevity and Aging Studies, University of Texas Health Science Center, San Antonio for reviewing the manuscript. This work was supported by Croatian Ministry of Science, Education and Sports grant 0098077.

References Bryan TM, Englezou A, Gupta J, Bacchetti S, Reddel RR (1995) Telomere elongation in immortal human cells without detectable telomerase activity. EMBO J 14:4240–4248

123

172 Bryan TM, Englezou A, Dunham M, Reddel RR (1998) Telomere length dynamics in telomerase-positive immortal human cell populations. Exp Cell Res 239:370–378 Campisi J (1992) Gene expression in quiescent and senescent fibroblasts. Ann NY Acad Sci 21:195–201 Campisi J (2000) Cancer, aging and cellular senescence. In Vivo 14:183–188 Corey RD (2002) Telomerase inhibition, oligonucleotides, and clinical trials. Oncogene 21:631–637 Counter CM, Avilion AA, LeFeuvre CE, Stewart NG, Greider CW, Harley CB, Bacchetti S (1992) Telomere shortening associated with chromosome instability is arrested in immortal cell which express telomerase activity. EMBO J 11:1921–1929 Cristofalo VJ, Pignolo RJ, Rotenberg MO (1992) Molecular changes with in vitro cellular senescence. Ann NY Acad Sci 663:187–194 Dimri GP, Lee X, Basile G, Acosta M, Scott G, Roskelley C, Medrano EE, Linskens M, Rubelj I, Pereira-Smith OM (1995) A biomarker that identifies senescent human cells in culture and aging skin in vivo. Proc Natl Acad Sci USA 92:9363–9367 Feng J, Funk WD, Wang SS, Weinrich SL, Avilion AA, Chiu CP, Adams RR, Chang E, Allsopp RC, Yu J, Le S, West MD, Harley CB, Andrews WH, Greider CW, Villeponteau B (1995) The RNA component of human telomerase. Science 269:1236–1241 Ferenac M, Polancec D, Huzak M, Pereira-Smith OM, Rubelj I (2005) Early-senescing human skin fibroblasts do not demonstrate accelerated telomere shortening. J Gerontol A Biol Sci Med Sci 60:820–829 Goodwin EC, DiMaio D (2001) Induced senescence in HeLa cervical carcinoma cells containing elevated telomerase activity and extended telomeres. Cell Growth Differ 11:525–534 Gorbunova V, Seluanov A, Pereira-Smith OM (2003) Evidence that high telomerase activity may induce a senescent-like growth arrest in human fibroblasts. J Biol Chem 287:7692–7698 Harley CB, Futcher AB, Greider CW (1990) Telomeres shorten during ageing of human fibroblast. Nature 345:458–460 Hayflick L (1965) The limited in vitro lifetime of human diploid cell strains. Exp Cell Res 37:614–636 Hayflick L, Moorhead PS (1961) The limited in vitro lifetime of human diploid cell strains. Exp Cell Res 25:585–621 Hemann MT, Strong MA, Hao LY, Greider CW (2001) The shortest telomere, not average telomere length is critical for cell viability and chromosome stability. Cell 107:67–77 Kim NW, Piatyszek MA, Prowse KR, Harley CB, West MD, Ho PL, Coviello GM, Wright WE, Weinrich SL, Shay JW (1994) Specific association of human telomerase activity in immortal cells and cancer. Science 266:2011–2015 Ledley FD, Soriano HE, O’Malley BW Jr, Lewis D, Darlington GJ, Finegold M (1992) DiI as a marker for cellular transplantation into solid organs. Biotechniques 13:580–587

123

Biogerontology (2007) 8:163–172 Makarov VL, Hirose Y, Langmore JP (1997) Long G tails at both ends of human chromosomes suggest a C strand degradation mechanism for telomere shortening. Cell 88:657–666 Martinez AO, Norwood TH, Prothero JW, Martin GM (1978) Evidence for clonal attenuation of growth potential in HeLa cells. In Vitro 14:996–1002 Matsumura T, Pfendt EA, Hayflick L (1979) DNA synthesis in the human diploid cell strain WI-38 during in vitro aging: an autoradiography study. J Gerontol 34:323–327 Morin GB (1989) The human telomere terminal transferase enzyme is a ribonucleoprotein that synthesizes TTAGGG repeats. Cell 59:521–529 Neumann AA, Reddel RR (2002) Telomere maintenance and cancer—look, no telomerase. Nat Rev Cancer 2:879–884 Olovnikov AM (1973) The incomplete copying of template margin in enzymatic synthesis of polynucleotides and biological significance of the phenomenon. J Theor Biol 41:181–190 Pereira-Smith OM, Smith JR (1981) Expression of SV40 T antigen in finite life-span hybrids of normal and SV40transformed fibroblasts. Somatic Cell Genet 7:411– 421 Roninson IB (2003) Tumor cell senescence in cancer treatment. Can Res 63:2705–2715 Rubelj I, Venable SF, Lednicky J, Butel JS, Bilyeu T, Darlington G, Surmacz E, Campisi J, Pereira-Smith OM (1997) Loss of T-antigen sequences allows SV40transformed human cells in crisis to acquire a senescent-like phenotype. J Gerontol A Biol Sci Med Sci 52:229–234 Rubelj I, Huzak M, Brdar B, Pereira-Smith OM (2002) A single stage mechanism controls replicative senescence through Sudden Senescence Syndrome. Biogerontology 3:213–222 Saretzki G, Von Zglinicki T (2002) Replicative aging, telomeres and oxidative stress. Ann NY Acad Sci 959:24–29 Shay JW, Bacchetti S (1997) A survey of telomerase activity in human cancer. Eur J Cancer 33:787–791 Te Poele RH, Ohorkov AL, Jardine L, Cummings J, Joel SP (2002) DNA damage is able to induce senescence in tumor cells in vitro and in vivo. Cancer Res 62:1876–1883 Touissaint O, Medrano EE, Von Zglinicki T (2000) Cellular and molecular mechanisms of stress-induced premature senescence (SIPS) of human diploid fibroblasts and melanocytes. Exp Gerontol 35:927–945 Von Zglinicki T, Nilsson E, Docke WD, Brunk UT (1995) Lipofuscin accumulation and ageing of fibroblasts. Gerontology 41:95–108 Wright WE, Piatyszek MA, Rainey WE, Byrd W, Shay JW (1996) Telomerase activity in human germline and embryonic tissues and cells. Dev Genet 18:173–179