1Departments of Obstetrics and Gynecology

__________________________________________________________________________________

Key words: telomerase, ovarian cancer, cisplatin, drug resistance

Abbreviations: Cisplatin, (CP); DNA fragmentation factor 45, (DFF45); Dulbecco's phosphate buffered saline, (DPBS); Green fluorescent protein, (GFP); Oligonucleotide, (ODN); Polymerase chain reaction—enzyme linked immunosorbant assay, (PCR-ELISA); reverse transcriptase-polymerase chain reaction, (RT-PCR); Sodium dodecyl sulfate-polyacrylamide gel electrophoresis, (SDS-PAGE)

Ovarian cancer is the fifth most common cancer in women with an estimated 23,100 new cases in the U.S. annually (Greenlee et al, 2000). Ovarian cancer has the highest mortality rate among gynecological cancers (5-year survival is approximately 35%). Optimal cytoreduction followed by platinum based chemotherapy is the gold standard of therapy. While the response rate to primary chemotherapy can be as high as 76%, response rate is dramatically reduced after relapse of disease (Kigawa et al, 1993). Platinum resistance, defined as disease recurrence less than six months from completion of therapy is an important prognostic predictor. Patients with platinum-resistant tumors have a response rate of less than 10% when retreated with platinum compounds (Blackledge et al, 1989). Alternative options also have poor response rates of 18-30% (Kohn et al, 1994; Thigpen et al, 1994; Creemers et al, 1996; Gordon et al, 1997; Kauffman and von Minckwitz, 1997; Muggia et al, 1997; Ozols, 1997; ten Bokkel Huinink et al, 1997; Bookman et al, 1998; Rose et al, 1998).

Telomerase is a multimeric ribonucleoprotein that adds telomeric repeats to chromosome ends (Morin, 1989). Thus, it stabilizes chromosomal DNA and is thought to confer immortality to cells (Shay and Bacchetti, 1999). Telomerase activity is generally limited to stem cell populations and tumor cells and is suppressed in normal somatic cells (Kim et al, 1994). Telomerase consists of a RNA component (hTR) that serves as an internal telomeric template, telomerase associated proteins and hTERT, a reverse transcriptase (Kim et al 1994; Weinrich et al, 1997). While both the protein catalytic hTERT subunit and the hTR RNA component have been identified and cloned, reports suggest that the reverse transcriptase catalytic subunit is the limiting determinant of telomerase activity (Feng et al, 1995; Meyerson et al, 1997; Weinrich et al, 1997; Counter et al, 1998). We have previously shown that a number of exogenous stresses can enhance telomerase activity in ovarian cancer cells as well as activate de novo telomerase activity in non-malignant ovarian surface epithelial cells (Alfonso-De Matte et al, 2001). Further, stress-induced telomerase activity is regulated in a phosphatidylinositol triphosphate kinase/c-Jun NH₂-terminal kinase-dependent manner (Alfsono-De Matte et al, 2001, 2002a,b, 2004).

Over 90% of tumors examined to date, including ovarian cancers, express telomerase activity (Hastie et al, 1990; Counter et al, 1994; Kyo et al, 1996, 1998a,b; Summerfeld et al, 1996; Hoos et al, 1998). In the ovary, studies have shown an absence of telomerase activity in normal ovarian surface epithelium and pre-malignant lesions, while tumor cells from both acites fluid and ovarian carcinomas express telomerase activity (Counter et al, 1994; Kyo et al, 1996, 1999; Murakami et al, 1997, 1998a,b; Duggan et al, 1998; Datar et al, 1999; Oishi et al, 1999; Park et al, 1999). Telomerase activity has been directly linked to ovarian tumor stage and aggressiveness (Murakami et al, 1998a,b; Hahn et al, 1999; Yan et al, 1999).

While the primary function of telomerase is the maintenance of telomeric integrity, several recent reports suggest an association between telomerase with reduced apoptosis and increased chemotherapeutic resistance (Asai et al, 1998; Kondo et al, 1998a,b; Herbert et al, 1999; Tian et al, 1999; Zhang et al 1999; Kiyozuka et al, 2000; Villa et al, 2000). Iida et al, (2000) showed that telomerase activity was significantly higher in colorectal cancers concurrently expressing the anti-apoptotic protein, Bcl-2. A similar relationship was demonstrated in cervical and colorectal cancer cell lines (Mandal et al, 1997). Likewise, reduced apoptosis in pancreatic cancer cells after exposure to etoposide was associated with elevated telomerase activity (Sato et al, 2000). Lastly, recent reports suggested that hTERT suppressed apoptosis prior to mitochondrial dysfunction and caspase activation in cultured cells (Holt et al, 1999; Fu et al, 2000).

Although there have been very few similar studies in the ovary, Faraoni et al, (1999) showed that telomerase activity in primary cultures of ovarian cancer cells was inversely related to chemosensitivity. In a pilot study, Takahashi et al, (2000) showed that ovarian cancer patients responding to platinum therapy had low levels of telomerase whereas 50% of non-responders demonstrated elevated telomerase activity.

Given its clinical importance, telomerase has been investigated as a potential diagnostic or prognostic tool and especially as a therapeutic target (Murakami et al, 1998b; Hahn et al, 1999; Yan et al, 1999). Telomerase inhibitors include both synthetic and naturally occurring compounds (Naasani et al, 1998; Perry et al, 1998; Togashi et al, 1998; Lian et al, 2001; Lin et al, 2003; Kraveka et al, 2003; Yokoyama et al, 2004) and fall into two general categories. The first involves compounds that bind telomeric DNA, preventing telomerase access to the telomere and predominately include small linear molecules that intercalate the G-quadruplex telomeric structure (Perry et al, 1998; Gavathiotis et al, 2003; Shammas et al, 2003, 2004; Yin et al, 2003). This group of inhibitors also includes modified oligonucleotides with sequence homology to telomeric DNA (Gavathiotis et al, 2003; Mata et al, 1997). The second group of telomerase inhibitors abrogate the enzymatic function of telomerase. These include reverse transcriptase inhibitors, most notably AZT (Kitagawa et al, 2000; Brown et al, 2003; Yamahuchi et al, 2003), RNA interference directed towards the hTERT catalytic component (Kosciolek et al, 2003), expression of dominant-negative hTERT mutants (Delhommeau et al, 2002), peptide nucleic acids (Herbert et al, 1999), as well as anti-sense or modified oligonucleotides complementary to the telomerase hTR RNA template (Kushner et al, 2000). Accessibility and easy design of oligonucleotides makes hTR an ideal target for telomerase inhibition. In addition, oligonucleotide modification, often with oligomer methylation or phosphorothioate linkages between bases (Levin, 1999; Chen et al, 2003), enhances oligomer stability from nuclease digestion to dramatically increase oligomer half-life (Geary et al, 2001).

To date, there have been many promising and favorable reports regarding anti-telomerase chemotherapeutic strategies. Anti-sense inhibition of telomerase increased the susceptibility of glioblastoma cells to cisplatin-induced apoptosis (Kondo et al, 1998a). Inhibition of telomerase with oligonucleotides resulted in enhanced pheochromocytoma cell apoptosis (Fu et al, 1999). Hahn et al, (1999) showed that complete inhibition of telomerase activity using a mutant form of human telomerase resulted in the death of tumor cells and in vivo expression of mutant telomerase eliminated tumorigenicity.

Clearly, chemosensitivity is a crucial prognostic factor for ovarian cancer and levels of or changes in telomerase may predict therapeutic outcome. Because of the presence of telomerase in ovarian tumor cells and the propensity for therapeutic failure from the development of drug resistance, we sought to determine whether inhibition of telomerase improved sensitivity to cisplatin in vitro in ovarian cancer cell lines OV2008 and C13, parent and daughter, CP-sensitive and CP-resistant cell lines, respectively (Asselin et al, 2001). Two telomerase inhibitors were chosen, a hexameric phosphorothioate oligonucleotide with sequence homology identical to the repeat sequence of the mammalian telomere, 5'-d(TTAGG)-3' (Mata et al, 1997) and a 2'-O-(2-methoxyethyl) RNA oligomer complementary to the telomerase hTR RNA component (Pitts and Corey, 1998, Herbert et al, 1999). In the present study, we show that inhibition of telomerase activity dramatically improved sensitivity to cisplatin in CP-resistant ovarian cancer cells.

Two ovarian cancer cell lines, OV2008 and C13 were used. The cell lines were selected because OV2008 is CP sensitive and its daughter cell line, C13 is resistant to CP (Asselin et al, 2001). Cells were maintained in medium 199/MDCB 105 (1:1) (Sigma, St. Louis, MO) supplemented with 10% fetal bovine serum (Hyclone, Logan, UT), and 10 m g/ml gentamicin (GIBCO BRL, Grand Island, NY) in a humidified 5% CO₂/ 95% air atmosphere. Cell counts were performed using a hemocytometer and cellular viability determined by Trypan blue exclusion assay.

Cell growth was determined by the MTS colorimetric assay (Promega, Madison, WI). The assay was performed in 96 well microtiter plates according to manufacturer's instructions and is based on soluble formazan production by dehydrogenase enzymes found in metabolically active cells. Samples were seeded six at 5x10³ cells per well. Absorbance was determined at 490nm using a Dynex MRX plate reader (Dynex Technologies, Chantilly, VA).

To quantitatively detect changes in telomerase levels, cells were assayed for telomerase activity using the telomerase polymerase chain reaction-enzyme-linked immunosorbent assay (PCR-ELISA) (Roche Molecular Biochemicals, Indianapolis, IN) according to the manufacturer's directions and as performed previously (Alfonso-De Matte et al, 2001). This assay has been demonstrated to be as sensitive as the radioactive, telomere repeat amplification protocol (TRAP) assay. Briefly, cells were washed with Dulbecco's phosphate buffered saline (DPBS), lysed in CHAPS lysis buffer and then assayed for protein using the Bio-Rad DC Protein Assay (Bio-Rad, Hercules, CA) according to the manufacturer's instructions. C13 and OV2008 cell extracts equivalent to 3 m g of protein were used. Following PCR-ELISA, telomerase activity was detected using a Dynex-MRX plate reader and recorded as absorbance units.

To examine the contribution of transcriptional control for telomerase activity in OV2008 and C13 cells, cells were incubated ± 5 m g/ml Actinomycin D and RT-PCR for hTERT mRNA was performed as described previously (Alfonso-De Matte et al, 2001). Total RNA was collected using Trizol reagent (GIBCO BRL). One microgram total RNA, oligo(dT) and reverse transcriptase were used to generate single-stranded cDNA. The cDNA samples were amplified using the Perkin-Elymer GeneAmp kit (Palo Alto, CA). b -actin was used as an internal control. The amplified products were separated by electrophoresis on a 9% polyacrylamide gel, stained with 1X SyberGreen (FMC Bioproducts, Rockland, MD) and analyzed with the Kodak EDAS 120 Digital Analysis System. Net hTERT mRNA intensities were normalized to their corresponding b -actin mRNA levels and expressed as a percentage of b -actin mRNA expression.

To examine the effect of interfering with telomerase activity on drug sensitivity, exponentially growing cells were treated with phosphorothioate DNA oligonucleotides (ODNs) complementary to the human telomeric repeat sequence or composed of a scramble sequence (Oncogene, Boston, MA) as described previously (Mata et al, 1997) at 2.5 m M in serum free media for 24 hours. Parallel cultures were also treated with 25 m M CP for two hours after incubation with the ODNs. Cells were then collected and assayed for telomerase activity, cell growth and caspase 3 levels at various time intervals up to 72 hours. Controls were untreated cells in serum free media.

To examine the effect of telomerase inhibition using the 2'-O-(2-methoxyethyl) RNA oligomer targeting the hTR template (Oncogene, Boston, MA) on CP sensitivity, cells were plated at 7.5x10⁴ cells per 60 mm² dish in serum free media for 24 hours. C13 cells were transfected with 34 m l Lipofectamine (Life Technologies, Inc., Grand Island, NY), and 8 m l RNA [100nMol dissolved in 469 m l MW 4691.8]. OV2008 cells were transfected in the same manner except 17 m l Lipofectamine was used after optimization with Green Fluorescent Protein (GFP) cDNA to establish transfection conditions favoring the most oligomer uptake. In all transfection experiments, parallel cultures transfected with GFP were used as a control for transfection efficiency. After 24 hours the transfection media was removed ± 25 m M CP was added for 2 hours followed by serum containing media. Cells were then collected and assayed for telomerase activity, cell viability and levels of DNA fragmentation factor 45 (DFF45) at various time intervals up to 72 hours.

In order to determine the extent to which apoptosis was affected by telomerase inhibition, cell lysates were assayed for DFF45 cleavage or caspase-3 activation. Fifteen micrograms of protein were added to 4x loading buffer (250 mM Tris pH 6.8, 8% SDS, 20% glycerol, 0.012% bromophenol blue, 4% b -mercaptoethanol), heated to 95 ° C for 5 minutes, electrophoresed in 12.5% SDS-polyacrylamide gels and transferred to nitrocellulose membranes (Amersham, Piscataway, NJ) via semi-dry transfer. Membranes were blocked with 3% nonfat milk in DPBS or 5% nonfat milk in Tris buffered saline plus 0.1% Tween-20 for DFF45 and caspase 3, respectively. All membranes were incubated overnight at 4° C in primary antibody.

Polyclonal antibodies for caspase-3 and DFF45 were purchased from Cell Signaling (Beverly, MA) and BD-PharMingen (San Diego, CA), respectively. Membranes were incubated and developed according to the Enhanced Chemiluminescent Protocol, according to manufacturer's instructions (Amersham). After initial blotting, membranes were reprobed for actin to ensure even loading.

Immunoblots were scanned and analyzed with ImagQuant software. Values reported for target proteins were normalized to the immunoblots' respective actin levels.

Samples for telomerase activity, MTS and trypan blue viability counts were run at least in triplicate and the data subjected to Student t test analysis for determination of statistical significance for telomerase activity or inhibition between treated and untreated samples. The results were expressed as mean ± standard error.

In contrast to primary cultures of normal ovarian surface epithelial cells (MF10 and XD cells) that are telomerase negative (Kruk et al, 1999), both OV2008 and C13 ovarian cancer cells expressed telomerase activity (Figure 1A). However, telomerase activity in the CP-sensitive OV2008 cells was consistently lower by approximately 25% than in C13 cells. RT-PCR analysis revealed at least three times more of the 145-bp hTERT RNA product in C13 cells compared with OV2008 cells when normalized to their respective 98-bp b -actin product (Figure 1B) confirming increased telomerase in C13 cells. In addition, hTERT mRNA stability was at least 2.5-fold greater in C13 cells compared to OV2008 cells with 40%

Figure 1. Enhanced telomerase expression in CP-resistant ovarian cancer cells. (A) Telomerase activity was determined by PCR-ELISA from lysates of C13 and OV2008 ovarian cancer cells as well as from lysates of primary cultures of normal ovarian surface epithelial cells (MF10 and XD). Results are expressed as the average absorbance at 450nm ± SE. (B) C13 (grey bars) and OV2008 (open bars) cells were treated with 5 m g/ml Actinomycin D for 2 hours. The 145-bp hTERT product seen following RT-PCR and PAGE was normalized to the respective 98-bp b -actin product..

loss of transcript in OV2008 cells compared to 15% transcript loss in C13 cells following incubation with Actinomycin D (Figure 1B).

B. Telomerase activity is inhibited by 2'-O'methyl-RNA, but not by 5'-d(TTAGG)-3'

OV2008 and C13 cell lines were assayed for telomerase activity by PCR-ELISA after treatment with 5'd(TTAGG)-3' ODN or 2'-O'methyl-RNA oligomers. As expected, telomerase activity was not significantly lower at 24 (p<0.8) or 48 hours (p<0.67) in OV2008 cells when treated with the 5'd(TTAGG)-3' (T) ODN alone (Figure 2A) since 5'd(TTAGG)-3' blocks access of telomerase to the telomere and does not directly inhibit telomerase activity, but contributes to increased telomeric instability. Similarly, the negative control antisense scramble oligonucleotide (TN) did not cause a significant change in telomerase activity (Figure 2A). While a 2 hour treatment with CP did not immediately affect telomerase activity, a 46% decrease in telomerase activity was observed 48 hours after the 2 hour CP exposure in OV2008 cells regardless of ODN treatment (*p<0.02 and **p<0.01 when compared to controls or to 5'd(TTAGG)-3' alone). This reduction in telomerase activity appeared to correlate with increased cell mortality as determined by morphologic examination. Likewise, ODNs did not cause a significant decrease in the telomerase activity in C13 cells (Figure 2B). Further, CP failed to have any significant effect telomerase activity in these cells.

Cells were transfected with 2'-O-methyl-RNA after optimization with GFP. Transfection efficiency was 70% in C13 cells and 50% in OV2008. There was no significant change in telomerase activity in OV2008 cells by 24 hours, however, telomerase activity was reduced by 40% at 48 hours following transfection (Figure 2C) (*p£ 0.04 and **p£ 0.02 2'-O-methyl-RNA or 2'-O-methyl-RNA + CP compared to controls, respectively). In contrast, in the faster growing C13 cells, telomerase activity was reduced by up to 90% at 24 hours following transfection with 2'-O-methyl-RNA ± CP (*p<0.004 and **p<0.005 compared to control values, respectively) (Figure 2D). However, telomerase inhibition appeared to be reversible in C13 cells as demonstrated by the restoration to baseline telomerase activity at 48 hours following transfection.

In order to determine the effect of 5'-d(TTAGG)-3'-mediated telomerase inhibition on ovarian cancer cell growth, MTS assays were performed. We noted that OV2008 cells grew slower than the C13 cells with an average doubling time approximately half that of C13 cells. Though 5'-d(TTAGG)-3' (T) and its scramble negative control oligomer (TN) had no effect on CP-sensitive OV2008 cell growth, CP, in the absence or presence of either T or TN ODNs, abrogated all cell growth and reduced cell numbers to well below starting values (*p£ 0.0005) as would be expected of CP-sensitive ovarian cancer cells (Figure 3A). ODNs or CP treatments alone also had little effect on C13 cell growth, reducing cellular proliferation less than 20% (*p£ 0.001) (Figure 3B). However, combined treatment of 5'-d(TTAGG)-3' with CP completely abolished C13 cell growth, indicating that such a combined treatment improved sensitivity to CP (**p£ 0.00004 and ***p£ 0.0003 compared to controls and CP-treated cells) (Figure 3B).

To determine the role of apoptosis for improved chemosensitivity by 5'-d(TTAGG)3', cell lysates were assayed for cleaved caspase-3 using SDS-PAGE and western immunoblot. The levels of active caspase 3 among controls, samples collected immediately after a 2 hr treatment with 25 m M CP or 5'-d(TTAGG)3' oligomer-treated samples were negligible in both OV2008 and C13 cells (Figure 3C,D). However, densitometric analysis revealed at least an 85x and 44x increase of caspase 3 activity in OV2008 and C13 samples, respectively, collected 48 hours after treatment with both 5'-

Figure 2. 2'-O-(methyl)RNA oligomers inhibit telomerase activity in OV2008 and C13 cells. Telomerase activity was determined from triplicate samples of OV2008 (A,C) and C13 (B,D) cells following treatment with ODNs (A,B) or transfection with 2'-O-(methyl)RNA (C,D). A,B) Cells were treated with 5'-d(TTAGG)3' (T) or a scramble sequence (TN) ODN. Parallel cultures were treated for 2 hours with 25 m M CP after which time the medium was refreshed and the cultures maintained for up to 48 hours while controls were untreated samples. C,D) Cells were transfected with 2'-O-(methyl)RNA (T) and similarly treated ± CP as in (A,B). Controls consisted of untreated samples collected at 24 hours. Results are expressed as fold increase or decrease in telomerase activity compared to control samples ± SE .

d(TTAGG)3' and 2 hours CP (Figure 3C,D).

(Figure 4A,B). In agreement with our results presented above, sensitivity to CP in OV2008 cells was apparent. Treatment of OV2008 cells with CP for 2 hours, with or without 2'-O-(2-methoxyethyl) RNA, caused a 99% decrease in the number of living cells by 72 hours following treatment (*p£ 0.001) (Figure 4A). 2'-O-(2-methoxyethyl) RNA alone did not affect OV2008 cell viability (Figure 4A).

As expected, C13 cells were only modestly sensitive to CP with only a 34% decrease in adherent living cells by 72 hours when treated with CP alone (*p£ 0.05) (Figure 4B). Similarly, 2'-O-(2-methoxyethyl) RNA did not significantly affect the number of viable C13 cells (Figure 4B). However, the combined treatment of CP and 2'-O-(2-methoxyethyl) RNA caused a 61% reduction in adherent, living cells as compared to the control group (**p£ 0.006) and further increased CP-mediated cytotoxicity by an additional 40% (***p£ 0.04) (Figure 4B).

To determine whether telomerase inhibition by 2'-0-(2-methoxyethyl) RNA also enhanced apoptosis, SDS-PAGE was performed on cell lysates and western immunoblots were probed for DFF45, a known cleavage substrate of caspase-3 (Widlak, 2000). While treatment with 2'-O-(2-methoxyethyl) RNA alone did not affect DFF45 expression in either OV2008 or C13 cells, combined treatment of 2'-O-(2-methoxyethyl) RNA with CP greatly enhanced DFF45 degradation (Figure 4C,D). Densitometric analysis revealed approximately 2x and 3x increased DFF45 cleavage in OV2008 and C13 cells, respectively, when CP was combined with RNA treatment

Figure 3. 5'-d(TTAGG)-3' enhances CP-mediated cytotoxicity and caspase 3 activation in C13 cells. Cell growth was measured by MTS assays and expressed as average absorbance at 490 nm ± SE in OV2008 (A) and C13 (B) cells after treatment with 5'-d(TTAGG)-3' (T) or the scrambled negative control (TN) ± 2 hour treatment with CP while controls consisted of untreated samples. Parallel cultures of OV2008 (C) and C13 (D) cells were subjected to SDS-PAGE. Following transfer, blots were probed for (active) cleaved caspase-3 (17-20kDa). Actin served as a loading control.

Figure 4. 2'-O-(methyl) RNA enhances CP-mediated cytotoxicity and DFF45 cleavage in C13. Cell viability was determined by trypan blue exclusion counts and expressed as the number of cells x 104 ± SE in OV2008 (A) and C13 (B) cells after transfection with the 2'-O-(methyl) RNA inhibitor (I) ± 2 hour treatment with CP while controls consisted of untreated samples. Parallel cultures of OV2008 (C) and C13 (D) cells were subjected to SDS-PAGE. Following transfer, blots were probed for the caspase 3 cleavage substrate, DFF45. Actin served as a loading control.

Since telomerase is expressed in >90% of human tumors and absent from most normal somatic cells and its expression correlates with tumor aggressiveness, telomerase is an attractive target for therapeutic intervention. Telomerase inhibition has been successfully achieved by several mechanisms including antisense technology, ribozymes directed against telomerase, and G-quadruplex binding agents that focus on either preventing access of telomerase to telomeric ends or abrogating hTERT and/or hTR sites (Rowley and Tabler, 2000). Inhibiting the addition of telomeric repeats to chromosome ends does not necessarily cause immediate cell death and can require substantial time to induce telomeric instability with subsequent growth arrest because the rate of telomeric erosion is approximately 50-100 bases per cell population doubling (Hastie et al, 1990; Shay, 1995; Shammas et al, 2004). In contrast, inhibition of telomerase by directly abrogating hTERT or hTR function may disrupt telomerase associated anti-apoptotic properties as well as promote telomeric instability. Nonetheless, disruption of telomerase activity appears to increase sensitivity to DNA damaging chemotherapeutic agents in breast cancer cells (Ludwig et al, 2001), glioblastoma cells (Kondo et al, 1998), and in neoplastic cells derived from telomerase RNA null mice (Lee et al, 2001). In agreement, we found that platinum resistance in C13 ovarian cancer cells was associated with increased telomerase activity that could contribute to enhanced cancer cell survival. In contrast, prior to the development of drug resistance, CP is an extremely effective chemothapeutic agent as demonstrated by the exceptional cytotoxic capacity of CP for OV2008 cells. Therefore, since chemoresistance, especially to platinum agents, is well documented in ovarian cancer and presents a serious therapeutic challenge, we sought to improve chemotherapeutic response in CP-resistant ovarian cancer cells by employing telomerase inhibition as a novel adjuvant therapy.

5'-d(TTAGG)3' has been shown to inhibit telomerase activity in cell lysates, lengthen cell doubling time and induce apoptosis in Burkitts lymphoma cells (Mata et al, 1997). In the present study, 5'-d(TTAGG)3' did not significantly lower the telomerase activity though it did dramatically improve chemosensitivity to CP in C13 cells. A scrambled negative control did not affect telomerase activity or improve sensitivity to CP though partial activity of the nonsense negative controls was described in Mata et al, (1997). While Lee et al, (2001) suggested that telomere dysfunction rather than telomerase itself is the principle determinant governing chemosensitivity, unlike previous studies, we did not see changes in MTL likely because of the short time frame of our experiments.

2-O-methyl RNA was an effective inhibitor of telomerase in both OV2008 and C13 cells. This inhibitor has already been characterized in animal and early human clinical trials (Pitts and Corey, 1998). Our investigation demonstrated a significant reduction in cell viability in cisplatin resistant cells when treated with both the 2-O-methyl RNA and cisplatin.

In the classical model of telomerase activity, inhibition of the enzyme alone would require multiple cell doublings to cause significant telomere shortening to signal cell death. Our investigation, as well as others (Kondo et al, 1998; Fu et al, 1999, Kushner et al, 2000), suggests that it is unlikely that telomerase inhibition-mediated cell death is related to telomere erosion alone although telomeric and, hence, chromosomal distortions due to inhibition of telomerase function could signal DNA damage responses, including increased apoptosis. Our results support an anti-apoptotic contribution by telomerase, since inhibition of telomerase resulted in increased caspase 3-dependent apoptosis. This is in agreement with studies reporting an association between telomerase activity and enhanced cancer cell survival consistent with poor prognosis in colon, breast, gastric, cervical, uterine and lymphoid cancers (Asai et al, 1998; Kiyozuka et al, 2000; Villa et al, 2000) in which telomerase appears to mediate its protective effect by conferring resistance to apoptosis (Herbert et al, 1999; Iida et al, 2000; Kondo et al, 1998a; Mandel and Kumar, 1997; Tian et al, 1999; Zhang et al, 1999).

Oligonucleotides as telomerase inhibitors have many benefits. These oligonucleotides are commercially available or can be readily synthesized which makes them attractive experimental agents. They bind the hTR template and prevent the addition of telomeric repeats and mismatch negative controls can be used to delineate proof of the mechanism of inhibition. There are methods for large-scale oligo synthesis, their pharmacokinetic properties are well characterized and the toxicity is low (Geary et al, 2001). There are also indications that adequate oral bioavailability can be achieved (Khatsenko et al, 2000). However, telomerase as a therapeutic target still raises many questions since not all tumors exhibit telomerase activity (Shay, 1995). Further, some normal cells, such as germ cells and hematopoietic cells, express low levels of telomerase activity and an effective inhibitor may impair the function of these cells (Morin, 1995). Subsequently, telomere loss resulting from prolonged telomerase inhibition in normal cells could result in chromosomal instability that can promote mutations required for cellular transformation (de Lange and Jacks, 1999).

Nevertheless, our results are promising in that both telomerase inhibitors employed significantly increased CP-mediated cytotoxicity in CP-resistant ovarian cancer cells, possibly in a caspase 3-dependent manner. The possibility of improving the sensitivity of such cells to current chemotherapeutic treatments is exciting and, clearly, warrants further study as an effective adjuvant treatment for recurrent platinum resistant ovarian cancer. Therefore, by understanding the relationship between telomerase activity and apoptotic pathways in ovarian tumor progression, abrogation of telomerase in conjunction with conventional platinum regimes could improve current therapeutic efficacy and clinical outcome by increasing cellular sensitivity to chemotherapeutic agents.

We thank Dr. Nicole C. Johnson and Ms. Yira Bermudez for help with densitometric analysis. This work was supported, in part, by a United States Army Department of Defense New Investigator Award DAMD 17-00-1-0565 (P.A.K.).