Abstract
Elevated telomerase levels are found in many malignancies, offering an attractive target for therapeutic intervention and diagnostic or prognostic purposes. Here we describe the use of a novel nanosensor developed for rapid screens of telomerase activity in biological samples. The technique utilizes magnetic nanoparticles that, on annealing with telomerase synthesized TTAGGG repeats, switch their magnet state, a phenomenon readily detectable by magnetic readers. We tested the efficacy of different telomerase inhibitors in crude human and murine samples and show that phosphorylation of telomerase regulates its activity. High-throughput adaptation of the technique by magnetic resonance imaging allowed processing of hundreds of samples within tens of minutes at ultrahigh sensitivities. Together, these studies establish and validate a novel and powerful tool for rapidly sensing telomerase activity and provide the rationale for developing analogous magnetic nanoparticles for in vivo sensing.
INTRODUCTION
Telomerase is a specialized reverse transcriptase, which is composed of an essential catalytic subunit and an RNA component (1, 2, 3) that, together with telomere-associated proteins, maintains telomere length and function (4, 5). Most somatic human tissues and primary cells possess low or undetectable telomerase activity, which results in telomere shortening with each cell division. In normal cells a critical telomere length is eventually reached, resulting in cellular senescence and finally apoptosis.
Elevated levels of telomerase activity are found in the majority of malignancies and are believed to play a critical role in tumorigenesis (6, 7, 8, 9, 10, 11). Telomere dysfunction also results in genetic instability with complex cellular and molecular responses involving the retinoblastoma gene/p53 gene checkpoints and apoptosis pathways (12, 13, 14, 15, 16, 17, 18). Telomerase inhibitors have shown preclinical efficacy, and a myriad of new agents are under development (19, 20). The ability to detect telomerase activity rapidly, quantitatively, and repeatedly would have significant value in cancer research, the development of more efficient telomerase inhibitors, titrating treatment efficacy, and in the detection of enzyme levels for prognostic purposes (6). Most commercially available telomerase assays use PCR-based amplification (6, 21, 22, 23), which is highly sensitive but also susceptible to amplification related errors (24, 25). Semiquantitative versions of the assay have been described previously (23, 26, 27) but can be technically challenging (24). Thus there is a need for an inherently quantitative assay that does not rely on PCR amplification and can be performed in crude cell or tissue extracts rather than highly purified samples.
Previously, we developed biocompatible superparamagnetic iron oxide nanoparticles (28) and used them to detect mRNA, protein (29), and protease activities (30) in high throughput, magnetic relaxation switch (MRS)-based assays. Here we describe the development of a new method to measure telomerase activity based on hybridization of magnetic nanoparticles (consisting of caged iron oxide crystals) to telomerase-associated TTAGGG repeats. Due to nanoparticle assembly formation, the relaxation time (T2) of surrounding water changes significantly, which can be readily measured by benchtop magnetic resonance (MR) relaxometers or imaging systems (Fig. 1). Here we describe the developed assay, validate it, and show that it can be used to rapidly measure telomerase activity in biological samples, to quantify therapeutic inhibition, and to investigate telomerase regulation.
MATERIALS AND METHODS
Synthesis of Nanosensors.
Amidated cross-linked iron oxide nanoparticles (CLIO-NH2) were obtained as described previously (29, 31) and conjugated to oligonucleotides (Tufts University Core Facility, Somerville, MA) via a stable thioether linkage with succinimidyl iodoacetate. Remaining free oligonucleotides were removed by magnetic purification of the Oligonucleotide-CLIO-NH2 (Miltenyi Biotec, Auburn, CA). Probe Telo-1 had the DNA sequence 5′-CCC-TAA-CCC-TAA-CCC-TAA-3′, and Telo-2 had the DNA sequence of 5′-CCC-TAA-CCC-TAA-3′. Each CLIO particle had an average of four oligonucleotides bound. This number was determined by adding to Telo-1 and Telo-2 a Cy-5-labeled complementary oligonucleotide (5′-TTA-GGG-TTA-GGG-3′; Tufts University Core Facility) and calculating the molar amount of fluorochrome bound to the particles using the extinction coefficient of Cy-5 at 648 nm. Telo-1 and Telo-2 were designed to bind only to the telomeric repeats and not to primer Ts [5′-AAT-CCG-TCG-AGC-AGA-GTT (6)].
Cell Extracts.
Cell extracts were prepared from various cell lines (B16 melanoma, CaPan-2 pancreas carcinoma, MCF-7 breast cancer, PC3 prostate cancer, ovcar-5 ovarian carcinoma, H-4-II-E hepatocellular carcinoma, Lewis lung carcinoma, 9L glioblastoma, HeLa, E6-1 Jurkat lymphoma, and rin 5F rat insulinoma) by incubating approximately 1 × 106 cells in 200 μl of lysis buffer for 30 min on ice and then centrifuging cells at 12,000 × g for 30 min (6). Lysis buffer consisted of 10 mm Tris-HCl, 1 mm MgCl2, 1 mm EGTA, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acid (0.5%), and 0.1 mm phenylmethylsulfonyl fluoride (all from Sigma-Aldrich, St. Louis, MO). Aliquots of the supernatant were stored at −80°C. The protein concentration (BCA; Pierce, Rockford, IL) varied between 0.5 and 4 mg/ml. For each telomerase assay, 1 μg of protein was used. Normal human melanocytes were used as negative control (32) and were a kind gift from Dr. Mark Eller (Department of Dermatology, Boston University School of Medicine). Tumor tissue samples of liver metastasis were mechanically homogenized before incubation in lysis buffer.
Assessing Telomerase Inhibition and Telomerase Regulation.
To study telomerase inhibition, Jurkat lymphoma cell extracts were incubated for 30 min at 37°C with either 200 μm azidothymidine (33, 34), 30 μm dideoxyguanosine (33, 34), 30 μm 2′o-methyl RNA (35), 10 μm antisense oligonucleotide (35), 10 μm phosphorothioate oligonucleotide (36), 10 μm of a hexameric phosphorothioate oligonucleotide in which a pair of three bases are separated by a nine-carbon phosphoramidite spacer (37), or 10 μm diaminoanthraquinone derivate (38). Inhibitors were obtained from Sigma-Aldrich (azidothymidine and dideoxyguanosine) and Calbiochem (San Diego, CA). To study telomerase regulation by phosphorylation, cultures of 9L glioblastoma cells were incubated with 0.1 μm 12-O-tetradecanoylphorbol-13-acetate (TPA) in DMSO to activate protein kinase C (39). Cells were harvested after 1.5, 3, 6, or 24 h, and cell extracts were tested for telomerase activity.
To evaluate the state of telomerase activation by phosphorylation, extracts from melanoma, ovarian cancer, and glioblastoma cell lines were incubated at 37°C for 30 min in the presence or absence of 80 milliunits of protein phosphatase 2A (PPA; Calbiochem). Telomerase activity was then determined as described before.
Telomerase-Mediated Primer Elongation.
The reaction buffer was modified from Ref. 6, consisting of 20 mm Tris-HCl, 1.5 mm MgCl2, 63 mm KCl, 1 mm EGTA, 0.05% Tween, and 50 μm nucleotides (all from Sigma-Aldrich). To the reaction buffer, 0.15 nmol of Ts primer and 5 μg of cell extract (protein) were added to generate the final reaction mixture. This reaction mixture was then diluted to a final volume of 50 μl with distilled water and incubated for 1 h at 37°C in a thermal cycler (MJ Research Inc., Waltham, MA). Thereafter, the temperature was increased to 94°C for 5 min to deactivate telomerase and to terminate the elongation process.
Relaxation Time (T2) Measurements.
T2 relaxation time measurements were carried out at 0.47 T and 40°C (Bruker NMR Minispec, Billerica, MA) in a total volume of 200–500 μl and with an iron concentration of 10 μg/ml. δT2 values were obtained after adding varying amounts of synthetic telomeric repeats to determine the sensitivity of the assay.
For MR plate imaging, 50 μl of the assay were transferred into a 384-well plate with an iron concentration of 10 or 5 μg/ml (1:1 dilution). Plates were imaged simultaneously at 1.5 T (Signa; GE Medical Systems, Milwaukee, WI) using T2-weighted spin echo sequences with varying echo times to obtain a T2 map [repetition time (TR), 2000 ms; echo time (TE), 25–200 ms] and a T1-weighted (TR, 600 ms; TE, 30 ms) spin echo sequence. Image T2 analysis was performed using CMIR-Image, an image analysis package developed in our laboratory in the Interactive Data Language (Research Systems Inc., Boulder, CO). A three-dimensional T2 image was constructed by fitting of a standard exponential transverse relaxation model (Moexp(−TE/T2)) to stacks of spin echo MR image slices acquired at a TR of 2000 ms and varying TE of 25–200 ms. Renderings were performed at multiple angles to highlight the resolution and three-dimensional nature of the calculated T2 maps. Data are shown as δT2 (T2 of the blank minus T2 of sample), unless noted otherwise.
Atomic Force Microscopy and Magnetic Force Microscopy.
Atomic force microscopy using a Dimension 3100 AF microscope (Digital Instruments, Santa Barbara, CA) was performed to obtain images of Telo-1 and Telo-2 in native and hybridized states. The nanoparticles were allowed to bind to the surface of mica for 20 min, and the surface was then washed with buffer, and images were acquired. A silicon nitride cantilever (20–40-nm tip size) was used in tapping mode for atomic force microscopy, and a magnetically coated tip (magnetic enough to magnetize assemblies) was used in tapping mode for magnetic force microscopy.
Conventional Telomerase Assays.
The reaction buffer for the telomeric repeat amplification protocol (TRAP) assay contained 0.1 nmol of the reverse primer Cxa (Ref. 22; Tufts University, Somerville, MA) and 2 units of Taq polymerase (Roche, Mannheim, Germany). After completion of telomerase inactivation, 31 cycles were performed, consisting of 94°C for 30 s, 50°C for 30 s, and 72°C for 90 s. The amplification products were visualized using PAGE, staining the DNA with SYBR Green (Molecular Probes, Eugene, OR). To test for correlation with commercially available assays, the Telomerase PCR ELISA-plus (Roche) and TRAPeze XL (Intergen, Purchase, NY) were used. Controls consisted of heat-inactivated cell extracts (94°C, 10 min), RNase treatment of cell extract (37°C, 30 min), omitting or replacing primer Ts with an unrelated primer, omitting the nucleotides, and cell extracts from telomerase-negative primary tissue melanocytes.
RESULTS
Thiolated oligonucleotides, chosen to recognize 30-bp telomeric repeats in complementary fashion, were attached to biocompatible, caged magnetic nanoparticles (40). The resultant sensors were stable in solution over months, were monodisperse when examined by atomic force microscopy, and had an overall size of 45 ± 4 nm by laser light scattering (Fig. 1, A and B). When sensors were incubated with telomeric repeats, they assembled in a linear fashion along the repeats, a feature that was not observed with control nonsense sequences (Fig. 1, A–C). To measure the effect of hybridization on magnetic spin-spin relaxation times (T2), we performed serial relaxometry studies (Fig. 1,B). Magnetic switching (δT2) was fast and pronounced, with half-maximum changes occurring in 30 s (P = 0.001) and plateauing after 40–60 min (Fig. 1,B). Primer oligonucleotides not hybridizing to telomeric repeats produced no measurable δT2. Further confirmation of magnetic switching was obtained by magnetic force microscopy (Fig. 2,A) and by correlating δT2 with the size of forming nanoassemblies (Fig. 1 C).
To determine the sensitivity of the method, a MR relaxometer-based assay was developed (Fig. 2,B). The detection threshold was approximately 100 attomoles of telomeric repeats in test tube format and 10 attomoles in 384-well plates, within the upper range to that of PCR-dependent ELISA-based photometric assays (Fig. 2,B, RTA, relative telomerase activity). There was a tight correlation between the conventional telomerase assays and the magnetic assay (r2 = 0.99, see the inset in Fig. 2 B); the latter, however, did not require PCR.
Screening Biological Samples.
To determine the status of telomerase activity in biological samples, we investigated a series of human and murine cell lines and tissues. Crude tissue or extracts were prepared, and a telomeric elongation step on a primer was performed. Samples were then incubated with the nanosensors, and relaxometry or MR imaging measurements were performed (Fig. 3,A). All tumor cell lines (breast, prostate, ovarian, pancreatic, lung, and liver carcinoma and glioma, melanoma, lymphoma, and insulinoma) and primary tumor tissue samples (liver metastasis) as well as human skin fibroblasts from newborns tested positive for telomerase, whereas primary skin melanocytes tested negative (Fig. 3, A and B). Control samples that included (a) heat- or RNase-inactivated samples, (b) samples without telomerase primer or nucleotides, and (c) samples with non-sense primers were all negative. There was also a tight correlation between measurements performed with the magnetic and photometric-based assays (r2 = 0.94), similar to that observed previously during assay validation (Fig. 2).
Rapid Throughput Screening.
The studies above demonstrated the feasibility of measuring telomerase in primary tumor samples and cell extracts; however, those studies were performed only at modest throughput due to the use of the single sample MR relaxometer. We therefore decided to adapt the technique to MR imaging of entire 384-well plates as an alternative read-out method. Fig. 3 C represents a three-dimensional reconstruction of two plates acquired by MR imaging within 15 min. The original Hahn multi-echo image data set was reconstructed to yield T2 relaxation times, which are displayed as telomerase units/well (1 unit = 2.94 × δT2). With this particular set-up, it is possible to image up to 7 plates simultaneously (2,688 sample points) in 15 min, resulting in throughputs of over 10,000 samples/hour. Importantly, this format also had an increased detection threshold and lower sample requirements (20 versus 200 μl) compared with MR relaxometer measurements.
Telomerase Inhibition and Regulation.
We next determined whether the developed technology could be used to measure telomerase inhibition. Specifically, we tested 2′o-methyl RNA, an antisense oligonucleotide covering the template region of the RNA integrated in telomerase, two different phosphorothioate oligonucleotides, a G-quadruplex-interacting diaminoanthraquinone derivate, and the reverse transcriptase inhibitors azidothymidine and dideoxyguanosine (Fig. 4 A). After treating the samples, the δT2 was reduced significantly with all inhibitors in dose-dependent fashion. The most efficient inhibition was achieved by the phosphorothioate oligonucleotides and the anthraquinone derivate as indicated by a δT2 of 0 compared with the baseline (the blank value).
Because telomerase activity has been shown to be tightly regulated by phosphorylation (41, 42, 43), we also determined the effects of TPA and of PPA on telomerase activity. TPA, a protein kinase C activator, enhanced telomerase activity in a time-dependent fashion (Fig. 4,B), whereas PPA, a phosphatase, decreased it. Interestingly, we observed differences among ovarian carcinoma, gliosarcoma, and melanoma cell lines (Fig. 4 C), suggesting different native phosphorylation states of telomerase in these cell lines.
DISCUSSION
The developed magnetic nanosensors can be used to measure telomerase activity in a variety of applications. Depending on the read-out, the sensitivity of the method ranges from potential single molecule detection (e.g., magnetic force microscopy) to 10–100-attomole levels using benchtop read-outs (relaxometers). The assay permitted the detection of about 10 attomoles of telomerase-synthesized DNA by MR imaging, which competes well with other PCR-independent assay methodologies. For comparison, a biosensor chip assay was reported to have a threshold of about 3000 attomoles of DNA (44), a chemiluminescent assay with a sensitivity of 100 attomoles (25), and a bioluminescent assay could detect 167 attomoles of telomeric repeats (Ref. 45; Fig. 2 B). Importantly, the developed technique has significant advantages over other methods: (a) the assay is inherently quantitative; (b) the method is simple and fast (approximately 150 min for an entire determination and only minutes for actual measurements); and (c) it requires no solid phase, and the method can be extended to a high-throughput screening format. An additional advantage of the MRS assay method is that it achieves this degree of sensitivity without PCR and therefore avoids PCR-related artifacts and difficulties in quantification.
Screening Telomerase Inhibitors and Telomerase Regulation.
The MRS telomerase assay is ideally suited for testing libraries of potential telomerase inhibitors (46) because of its ready application to high-throughput screening. A number of different approaches to telomerase inhibition are under active investigation (33, 34, 35, 47, 48). In our experiments, we were able to demonstrate a higher potency of phosphorothioate oligonucleotides when compared with antisense RNA at identical concentrations of 10 μm (36) as well as the much lower potency of azidothymidine (34). We were also able to confirm the high potency of the G-quadruplex-interacting anthraquinone derivate (38), which was comparable with that of phosphorothioate oligonucleotides in vitro.
Much of the mechanism of specific telomerase regulation remains unknown. Prior work, however, has shown that phosphorylation of human telomerase reverse transcriptase enhances telomerase activity (41, 42, 43, 49) and that dephosphorylation inhibits it (50, 51). Although the native telomerase activity was similar in the glioma, ovarian carcinoma, and melanoma cell lines tested, there were considerable differences in inhibition by PPA among the cell lines (Fig. 4 C). Differential screens in the native (untreated) and PPA-treated samples could thus be used to investigate the phosphorylation state of telomerase. Similar experiments conducted with protein kinase C activator TPA could complement comprehensive telomerase assessment.
Probing Telomerase-Related Function.
The magnetic nanosensors developed here could easily be modified to probe for other telomere targets. For example, it is conceivable to use the sensors to measure telomere length, human telomerase reverse transcriptase, telomerase mRNA, and/or the human template RNA integrated in telomerase. These and other protein-sensing nanosensors can be synthesized by covalent attachment of different ligands to the caged iron oxide nanoparticles. Proof of principle of such sensors for DNA-DNA, protein-protein, protein-small molecule, and enzyme reactions has been shown previously (29). Such multiligand sensors could then be used to parallel screen different telomerase targets. For example, it has been shown that telomere length does not necessarily correlate with levels of telomerase activity (6, 52). Furthermore, the expression of telomerase is not a predictor of telomerase activity because it is expressed ubiquitously (53). Whereas it has been reported that the telomerase catalytic subunit human telomerase reverse transcriptase mRNA is detected only in telomerase-positive germ line and tumor cells, human telomerase reverse transcriptase mRNA has nevertheless been found in lymphocytes and thymocytes, independent of the status of telomerase activity (54). Because of these emerging observations, multifunctional telomerase/telomere sensors might prove useful to investigate telomerase biology further.
In Vivo Sensing Using MRS Nanoparticles.
A major application of the described nanosensors for telomerase activity would be the ability to measure telomerase activity in living cells in vitro or in tissues and organs in vivo. MRS nanoparticles are biocompatible and nontoxic, and similar base materials are used clinically, e.g., to image lymph node metastases (28). For imaging of telomerase activity, nanoparticles would need to be delivered to the nucleus, which has been achieved recently by attaching membrane translocation signals to identical magnetic particles (55). Interestingly, nuclear localization of these magnetic nanoparticles was nontoxic to the cells tested in vitro (55). Furthermore, independent magnetic sensing of T1 (to determine local concentrations of nanoparticles) and T2 (to determine magnetic switching) should allow bulk sensing of nanoassembly formation in living cells. A number of studies are under way to investigate the feasibility of this approach. Irrespective of these potential uses of the nanoparticles, we believe that the described telomerase switches will be an important tool for telomerase research.
Grant support: J. G. is supported by a grant from the German Research Society (DFG).
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Requests for reprints: Ralph Weissleder, Center for Molecular Imaging Research, Building 149, 13th Street, 5403, Charlestown, Massachusetts 02129. Phone: (617) 726-8226; Fax: (617) 726-5708.
Telomerase magnetic nanosensor. A, telomerase elongates primer ends (blue) by adding TTAGGG telomeric repeats (red). Magnetic nanoparticles with complementary oligonucleotide motifs have low relaxivity in monomeric, dispersed state and high relaxivity in hybridized, assembled state. Atomic force microscopy shows the nanoparticles (40-nm diameter) align along single DNA strands (black bar, 0.2 μm). B, induced magnetic T2 changes as a function of time after adding an oligonucleotide consisting of either a 54-bp telomeric repeat or the primer. C, correlation between induced relaxation and size changes (measured by laser light scattering).
Telomerase magnetic nanosensor. A, telomerase elongates primer ends (blue) by adding TTAGGG telomeric repeats (red). Magnetic nanoparticles with complementary oligonucleotide motifs have low relaxivity in monomeric, dispersed state and high relaxivity in hybridized, assembled state. Atomic force microscopy shows the nanoparticles (40-nm diameter) align along single DNA strands (black bar, 0.2 μm). B, induced magnetic T2 changes as a function of time after adding an oligonucleotide consisting of either a 54-bp telomeric repeat or the primer. C, correlation between induced relaxation and size changes (measured by laser light scattering).
Sensing telomerase. A, magnetic force microscopy of clustered nanoparticles (binding to single telomeric repeats) shows locally detectable increases in magnetic field distortion (black horizontal arrows). Equal-sized but nonclustered nanoparticles have a much lower magnetic effect (white arrows; black bar = 1 μm). B, comparative sensitivities of benchtop magnetic and standard telomeric repeat amplification protocol assay using a 54-mer telomeric repeat. Note the similar sensitivities of both formats and good correlation (r2 = 0.99).
Sensing telomerase. A, magnetic force microscopy of clustered nanoparticles (binding to single telomeric repeats) shows locally detectable increases in magnetic field distortion (black horizontal arrows). Equal-sized but nonclustered nanoparticles have a much lower magnetic effect (white arrows; black bar = 1 μm). B, comparative sensitivities of benchtop magnetic and standard telomeric repeat amplification protocol assay using a 54-mer telomeric repeat. Note the similar sensitivities of both formats and good correlation (r2 = 0.99).
Telomerase magnetic switch assay in tumor samples. A, magnetic switch assay to sense telomerase activity in different cell and tissue samples. Blank, assay without added cell extract; Heat, cell extract that has been heated to 84°C for 8 min to inactivate telomerase; no Primer, the telomerase-specific primer Ts was omitted; nonsense Primer, Ts was exchanged for a primer not recognized by telomerase; no Ncltd, the nucleotides were omitted; RNase, pretreatment of cell extract with RNase to inactivate telomerase; Melanocytes, cell extract from primary melanocytes negative for telomerase (32). B, representative PAGE gel of standard telomeric repeat amplification protocol assay of some examples from A. The bands in the nonsense primer assay are PCR artifacts. Only the assays with tumor cell extracts show a positive result. C, high-throughput screening of telomerase activity by magnetic resonance imaging of well plates. Data are shown as a three-dimensional reconstruction of T2 maps at 1.5 T, derived from magnetic resonance images of two 384-well plates stacked on each other and filled partly with 50 μl of buffer (blue) and controls or telomerase containing samples; units on the color bar are shown as relative units of telomerase activity.
Telomerase magnetic switch assay in tumor samples. A, magnetic switch assay to sense telomerase activity in different cell and tissue samples. Blank, assay without added cell extract; Heat, cell extract that has been heated to 84°C for 8 min to inactivate telomerase; no Primer, the telomerase-specific primer Ts was omitted; nonsense Primer, Ts was exchanged for a primer not recognized by telomerase; no Ncltd, the nucleotides were omitted; RNase, pretreatment of cell extract with RNase to inactivate telomerase; Melanocytes, cell extract from primary melanocytes negative for telomerase (32). B, representative PAGE gel of standard telomeric repeat amplification protocol assay of some examples from A. The bands in the nonsense primer assay are PCR artifacts. Only the assays with tumor cell extracts show a positive result. C, high-throughput screening of telomerase activity by magnetic resonance imaging of well plates. Data are shown as a three-dimensional reconstruction of T2 maps at 1.5 T, derived from magnetic resonance images of two 384-well plates stacked on each other and filled partly with 50 μl of buffer (blue) and controls or telomerase containing samples; units on the color bar are shown as relative units of telomerase activity.
Testing telomerase inhibitors. A, telomerase activity in a lymphoma cell extract and the same extract treated either with 2′o-methyl RNA (2′O me RNA), antisense oligonucleotide, 2 different phosphorotioate (PT) oligonucleotides, an anthraquinone derivate, azidothymidine (AZT), or dideoxyguanosine (ddG). The inset demonstrates the dose-response curve of telomerase inhibition by dideoxyguanosine (r2 = 1.0). B, treatment of 9L cell cultures with the protein kinase C activator TPA results in a significant increase of telomerase activity as compared with the control (DMSO alone). C, treatment of cell extracts with protein phosphatase 2A (PPA) leads to a clear reduction of telomerase activity but with considerable differences among the cell lines, suggesting different phosphorylation states.
Testing telomerase inhibitors. A, telomerase activity in a lymphoma cell extract and the same extract treated either with 2′o-methyl RNA (2′O me RNA), antisense oligonucleotide, 2 different phosphorotioate (PT) oligonucleotides, an anthraquinone derivate, azidothymidine (AZT), or dideoxyguanosine (ddG). The inset demonstrates the dose-response curve of telomerase inhibition by dideoxyguanosine (r2 = 1.0). B, treatment of 9L cell cultures with the protein kinase C activator TPA results in a significant increase of telomerase activity as compared with the control (DMSO alone). C, treatment of cell extracts with protein phosphatase 2A (PPA) leads to a clear reduction of telomerase activity but with considerable differences among the cell lines, suggesting different phosphorylation states.