Antitumor Activity and Metabolic Activation of N-Methanocarbathymidine, a Novel Thymidine Analogue with a Pseudosugar Rigidly Fixed in the Northern Conformation, in Murine Colon Cancer Cells Expressing Herpes Simplex Thymidine Kinase

N-Methanocarbathymidine [(N)-MCT], a thymidine analogue incorporating a pseudosugar with a fixed Northern conformation, exhibits antiherpetic activity against both herpes simplex virus (HSV) HSV-1 and HSV-2, with a potency greater than that of the reference standard, ganciclovir (GCV). In the present study, we have assessed the cytotoxic activity in vitro of (N)-MCT in wild-type murine colon cancer cells (MC38) and in cells expressing the herpes simplex thymidine kinase gene (MC38/HSV-tk), and the antitumor activity of (N)-MCT in vivo against HSV-tk transduced and nontransduced MC38 murine tumors. In vitro, when assessed over a 48-h period, the growth-inhibitory activity (IC₅₀) of (N)-MCT toward MC38/HSV-tk cells was 2.9 μm In parallel studies, the cytostatic activity of the reference compound GCV in these tumor lines was 3.0 μm. In studies in vivo, both (N)-MCT and GCV (100 mg/kg) given twice daily for 7 days completely inhibited the growth of HSV-tk-transduced MC38 tumors while exhibiting no effect on nontransduced MC38 tumors in mice. In nontransduced cells both in vitro and in vivo, only low levels of (N)-MCT and its monophosphate could be detected after administration of the parent drug, whereas in HSV-tk-transduced cells (N)-MCT was phosphorylated to its respective mono-, di-, and triphosphates. Furthermore, data showed that (N)-MCT incorporated in high levels into cellular DNA whereas trace levels were measured into RNA. These observations indicate that (N)-MCT may be a useful candidate prodrug for HSV-tk suicide gene therapy of cancer.

tk

encoded by the genome of herpesvirdae type 1 (HSV-tk) has proved to be an excellent target for developmental chemotherapy. As a tool, HSV-tk is responsible for the initial processing and therefore the ultimate usefulness of a variety of nucleoside analogue antiherpetic agents (1–3) and can also be used for tumor suppression using suicide gene therapy (4, 5) and for improving the efficacy of allogeneic bone marrow transplantation (6). The unique feature of viral thymidine kinase (HSV-tk) is its wide range of substrate activity in catalyzing the monophosphorylation of antiherpetic agents, a feature uncharacteristic of the corresponding human cellular enzyme (2, 7, 8). This attribute permits the multifold application of this viral kinase. In most applications, further drug metabolism requires cellular enzymes that continue the phosphorylation process, resulting in the formation of triphosphate, which inhibits viral DNA synthesis and brings about virus eradication. By transfecting tumor cells with HSV-tk, this same activation sequence can be initiated, allowing certain antiherpetic agents, through their triphosphates, to inhibit tumor growth.

Central to the functional disparity between the viral thymidine kinase and its cellular counterpart is the lack of homology in the primary sequence between these two proteins (9). The molecular template for acceptable substrates of HSV-tk is broad, consisting of both purine and pyrimidine nucleoside analogues. Included among these are analogues with modifications in the aglycone or in the sugar moiety. The arrays of active compounds which have been studied include analogues with guanine, uracil, and thymine as nucleobases. Modifications in the sugar are more diverse and include analogues containing natural and substituted ribofuranosides or acyclic and carbocylic pseudosugars (10–12). The range of acceptable substrates for the human enzyme is more limited.

A recently synthesized nucleoside analogue with a pseudosugar rigidly fixed in the Northern conformation, (N)-MCT (Fig. 1), was found to exhibit potent antiherpetic activity against herpes simplex virus types 1 and 2 on evaluation by both the plaque reduction and the cytopathogenic effect methods (13). In view of the widely studied cytotoxic effect of the antiherpetic and anticytomegalovirus agent GCV in human and murine tumor cells expressing HSV-tk and the clinical interest in the GCV/HSV-tk and related systems in cancer gene therapy in human subjects (14–16), we have undertaken the present study to characterize the cytotoxicity and metabolism (N)-MCT in vitro and in vivo toward HSV-tk-expressing murine colon cancer cells tumor cells.

(N)-MCT was synthesized as described previously (13). GCV (Cytovene-IV) was obtained from Hoffman La Roche Laboratories (Nutley, NJ). [Methyl-³H](N)-MCT (1.7 Ci/mmol), [8-³H]GCV (22 Ci/mmol), and [methyl-³H]thymidine (25 Ci/mmol) were obtained from Moravek Biochemicals (Brea, CA). Other nucleoside and nucleotide standards were purchased from Sigma Chemical Co. (St. Louis, MO). The enzymes DNase I [DNase I, type II (bovine pancreas)], phosphodiesterase I, type VII (Crotalus atrox venom), and alkaline phosphatase (Escherichia coli) were purchased from Sigma. All other reagents and chemicals were of the highest quality obtainable.

The 3-methylcholanthrene-induced murine colon adenocarcinoma MC38 was a gift from Dr. Steven Rosenberg (National Cancer Institute, NIH). MC38/HSV-tk cells were generated by transduction of the parental cell lines in supernatant from PA317-STK cells (17) and subsequent G418 selection (Genectin; Life Technologies, Inc., Rockville, MD; 1 mg/ml for 3 weeks), followed by single cell cloning by limiting dilution. The mouse mammary carcinoma cell lines FM3A (wild type), FM3A/tk⁻, and FM3A/tk⁻/HSV-tk (18, 19) were a gift from Prof. Jan Balzarini (Rega Institute, Leuven, Belgium). All cell lines were grown in DMEM supplemented with 10% heat-inactivated FCS, 50 IU/ml penicillin, 50 μg/ml streptomycin, and 2 mm l-glutamine. Cells were in logarithmic growth at the time of use and were maintained at 37°C in a humidified atmosphere of 95% air and 5% CO₂.

For the determination of the cytostatic effect of (N)-MCT and the reference standard GCV, exponentially growing wild-type and HSV-tk-transduced cells (10⁵ cells) were cultured in 24-well plates overnight. Cells were then washed with fresh medium, and various concentrations (0.2–100 μm) of (N)-MCT and GCV were added. After an additional 48 h, cells were harvested by trypsinization, collected, and counted in a Coulter counter. The cell growth rate was expressed as a percentage of the increase in cell number of the untreated control cultures. The IC₅₀s (the drug concentration resulting in 50% inhibition of cell growth) was calculated from the linear portion of the growth inhibition curve.

The experimental protocol was approved by the Animal Care and Use Committee of Ben-Gurion University and is in compliance with the Guide for the Care and Use of Laboratory Animals of NIH. Male C57/BL6 mice, 6–8 weeks of age, were housed six/cage and allowed free access to food and water. Mice were inoculated s.c. with 0.25 × 10⁶ MC38 cells in 0.1 ml of serum-free DMEM. The nontransduced MC38 cell suspensions were injected s.c. into the left flank, and the HSV-tk-transduced cells were injected s.c. into the right flank. Tumors were allowed to grow for 7 days, reaching a size of 50–100 mm³, and mice were then treated i.p. with 100 mg/kg (N)-MCT or GCV or saline (0.5 ml) twice daily for 7 days. Bidimensional tumor measurements were performed with calipers on days 1, 3, 5, and 7 after treatment, and tumor volume was determined using the simplified formula for rotational ellipse (l × w² × 0.5) as described previously (20). Tumor growth rate was calculated as a percentage of the tumor size at the beginning of the treatment. Experiments were terminated at day 7 after treatment because control animals appeared to be in distress from lesions in the nontransduced tumors.

After the appropriate incubations with (N)-MCT, cells were washed three times with PBS, and after trypsinization, collected by centrifugation at 1,500 × g for 10 min, cell pellets were extracted with 0.5 ml of 60% methanol (HPLC grade), and the extracts were heated for 2.5 min at 95°C. After centrifugation at 12,000 × g for 10 min, the clear supernatant fractions were evaporated under nitrogen and redissolved in 250 μl of water, and aliquots of the latter reconstituted samples were subjected to anion-exchange chromatography.

The separations of (N)-MCT and its phosphorylated metabolites were carried out using a Hewlett-Packard 1100 HPLC with a diode-array UV absorption detector. A Partisil-10 SAX column (250 × 4.6 mm) was used, with the following elution program: 0–5 min, 100% buffer A (0.01 m ammonium phosphate, native pH); 5–20 min, linear gradient to 25% buffer B (0.7 m ammonium phosphate with 10% methanol); 20–30 min linear gradient to 100% buffer B; 30–40 min 100% buffer B; 40–55 min, linear gradient to 100% buffer A, and equilibration. The flow rate was 2 ml/min. One-min fractions were collected, and radioactivity was determined by scintillation spectrometry. The retention times of (N)-MCT and its phosphates were as follows: (N)-MCT, 3 min; (N)-MCT-MP, 10 min; (N)-MCT-DP, 20 min; and (N)-MCT-TP, 30 min. Fractions containing radiolabeled (N)-MCT nucleotides were quantitated based on the known specific activity of the parent tritiated nucleoside. To confirm the identity of the radiolabeled (N)-MCT nucleotides, samples were enzymatically degraded to nucleosides and analyzed by anion-exchange and reverse-phase HPLC as described more fully below. Greater than 95% of the total tritium comigrated with authentic standards for (N)-MCT.

Reverse-phase HPLC was used for the separation and identification of the nucleosides (N)-MCT in cellular extracts for the quantitation of (N)-MCT after enzymatic degradation of its respective phosphates and for the DNA and RNA studies. Chromatography was performed using a Beckman Ultrasphere octadecylsilane column (250 × 4.6 mm, 5-μm particle size), with an elution program as follows: 0–25 min, linear gradient to 25% methanol; 25–30 min, isocratic with 25% methanol; 30–40 min, linear gradient to 1% methanol, and equilibration. The flow was 2 ml/min. One-min fractions were collected, and radioactivity was determined by liquid scintillation counting. The retention time for (N)-MCT was 14 min.

The identification of the (N)-MCT metabolites were carried out by selective enzymatic degradation of the extracts of cells exposed to (N)-MCT as described previously (21). Briefly, lyophilized methanolic cell extracts were redissolved in 100 μl of 0.01 m Tris-HCl (pH 9.0), containing 1 mm MgCl₂, and 1.5 units of alkaline phosphatase or 0.03 unit of venom phosphodiesterase was added to the appropriate aliquots. Samples were incubated for 4 h at 37°C, enzymes were inactivated by heating at 95°C for 2.5 min, and aliquots were then analyzed by anion-exchange and C₁₈ reverse-phase HPLC as described above.

DNA and RNA were isolated using the TRI reagent procedure (22, 23). Briefly, 50 × 10⁶ cells were incubated for 24 h with (N)-MCT, 10 μm, 10 μCi/ml; at the end of the incubations, cells were washed three times with cold PBS, harvested by trypsinization, and collected by centrifugation. One ml of TRI reagent (guanidine thiocyanate and phenol in a monophasic solution) was added to the cell pellet, solubilizing the DNA, RNA, and protein. Chloroform was then added, and the mixture was centrifuged. RNA (in the aqueous phase) and DNA (in the interphase) were then separated and quantified according to the TRI reagent protocol.

For DNA hydrolysis, the pH of the DNA solution, in 8 mm sodium hydroxide, was adjusted to pH 7.4, using HEPES buffer (66 μl of 0.1 m HEPES/ml DNA solution), and the DNA was hydrolyzed by overnight incubation at 37°C, with 140 μg of DNase, 0.02 unit of venom phosphodiesterase, and 5 units of alkaline phosphatase, in 1 ml. Aliquots of 100 μl containing ∼25 μg of hydrolyzed DNA were analyzed by reverse-phase HPLC to separate (N)-MCT as described previously (see above). Levels of (N)-MCT in RNA were measured without further hydrolysis, as follows: to 1 ml of isolated RNA in 75% ethanol was added 4 ml of scintillation mixture, and the radioactivity was determined.

The ability of tk to convert (N)-MCT to the monophosphate metabolite by cell lysates was determined by a modification of the methods described previously (19, 24). Briefly, MC38 and MC38/HSV-tk cells (about 100 × 10⁶ cells) were washed three times with cold PBS and resuspended in 1 ml of 50 mm potassium phosphate buffer (pH 7.6) containing 2 mm DTT, 2 mm EDTA, 0.5 mm phenylmethylsulfonyl fluoride, and 5 mm benzamidine. The cell suspension was lysed by three freeze-thaw cycles, sonicated three times for 10 s, and centrifuged at 12,000 × g for 5 min at 4°C to remove cellular debris. To remove any cellular nucleosides and nucleotides, the supernatant was transferred into cellulose membrane dialysis tubing (Sigma) and dialyzed overnight at 4°C against 100 ml of 50 mm Tris-HCl (pH 7.6) containing 5 mm DTT, 5 mm magnesium chloride, and 20% glycerol and stored in aliquots at −70°C. Protein concentration in cell extracts was assayed for tk activity in a standard reaction mixture containing 2.5 mm magnesium chloride, 10 mm DTT, 1 mg/ml BSA, 2.5 mm ATP, 10 mm sodium fluoride, and 10 μl of cell extract (10 μg protein) in a total volume of 100 μl of 50 mm Tris-HCl (pH 8.0). Nine substrate concentrations of [³H]thymidine or [³H](N)-MCT, ranging from 0.5 to 250 μm (10 μCi/ml radioactivity), were used in the kinase assays. The reaction mixtures were incubated at 37°C for 30 min, and the reactions were terminated by addition of 100 μl of absolute methanol and heating at 2.5 min at 95°C. Samples were then dried under nitrogen, redissolved in 100 μl of water, and subjected to HPLC analysis for radiolabeled thymidine, (N)-MCT, and their monophosphate metabolite using an anion-exchange Partisil-10 SAX column as described above.

Six mice bearing wild-type MC38 and MC38/HSV-tk in the left and right flanks, respectively, were treated i.p. with [³H](N)-MCT, 100 mg/kg, containing 400 μCi of labeled drug. MC38 and MC38/HSV-tk tumors were removed 24 h after drug administration from mice under ether anesthesia and immediately frozen in liquid nitrogen and stored at −70°C pending analysis. Tumors (100–200 mg) were homogenized in 1 ml of 60% methanol using a Polytron homogenizer (Kinematica, Lucerne, Switzerland) at 4°C. The homogenate was heated at 95°C for 3 min and centrifuged at 12,000 × g for 10 min at 4°C. The supernatant was collected and evaporated under a nitrogen stream. The residue was redissolved in deionized water and subjected to anion exchange HPLC analysis as described above.

Before initiating in vivo studies, we compared the cytostatic effect of (N)-MCT in wild-type and HSV-transduced murine colon cancer cells with that of the reference compound GCV. As illustrated in Fig. 2, when assessed over a 48-h period, the two compounds were roughly equivalent in their ability to inhibit murine colon cancer cells expressing the HSV-tk gene, with IC₅₀s for both compounds nearly 3 μm. No cytostatic effects were noted in nontransduced cells over the dose range studied (0.2–100 μm; Fig. 2).

Because it is generally accepted that nucleoside analogue antiviral and antitumor agents require intracellular phosphorylation to exert their pharmacological effects, we undertook the characterization of the metabolic products of (N)-MCT in MC38 HSV-tk-transduced and nontransduced tumor cells. Exponentially growing MC38 and MC38/HSV-tk cells (5 × 10⁶) were incubated with [³H](N)-MCT (10 μm; 5μCi/ml) for 6 h, collected by centrifugation, and extracted with methanol (60%), and the extracts were subjected to anion-exchange HPLC (Partisil-10 SAX) as described more fully in “Materials and Methods.” In nontransduced MC38 cells, (N)-MCT yielded a significant (N)-MCT-MP peak in addition to the parent nucleoside, but only trace amounts of the di- and triphosphates were detected (Fig. 3A and Table 1). In the transduced MC38 cells, all three phosphorylated products were detected, with (N)-MCT-TP (elution time, 30 min) being most prominent (Fig. 3B and Table 1).

To confirm the identity of radiolabeled (N)-MCT nucleotides, aliquots of methanolic extracts of [³H](N)-MCT-treated MC38 or MC38/HSV-tk cells were subjected to anion-exchange chromatography after 6-h treatment with either alkaline phosphatase or venom phosphodiesterase, as described in detail in “Materials and Methods.” As shown in Fig. 4, left panel, no significant change was seen in the (N)-MCT-MP peak in extracts from nontransduced cells after phosphodiesterase treatment, whereas all of the detectable radioactivity migrated with the parent nucleoside after phosphatase treatment. In extracts from HSV-tk cells, >95% of the radioactivity migrated with (N)-MCT after phosphatase treatment (Fig. 4, right panel). These results were confirmed by reverse-phase chromatography on an octadecylsilane column (data not shown).

Significant monophosphate formation of (N)-MCT was observed in nontransduced MC38 cells, although only limited amounts of the di- or triphosphate metabolites were detectable (Fig. 3 and Table 1). To determine whether cellular thymidine kinase made an obligatory contribution to the initial phosphorylation step seen in HSV-tk-expressing cells, we examined the phosphorylation profile in mouse mammary carcinoma cells lacking cellular thymidine kinase (FM3A-tk⁻) and in cells lacking cellular tk but transduced with the HSV-tk gene (FM3A/tk⁻/HSV1-tk). As illustrated in Fig. 5, (N)-MCT-MP could not be detected in FM3A/tk⁻ cells but substantial formation of (N)-MCT mono-, di-, and triphosphate was detectable in the FM3A/tk⁻/HSV-tk transduced cells, indicating that although endogenous tk could monophosphorylate (N)-MCT, it was not essential for the initial metabolic activation of the drug.

To assess the overall influence of the mediating factors that contribute to the sustained concentration of (N)-MCT metabolites, a time-course study to evaluate the intracellular accumulation and decay of the phosphorylated products of (N)-MCT was undertaken. MC38 and MC38/HSV-tk cells were incubated with radiolabeled (N)-MCT for 24 h, and the concentrations of the mono-, di-, and triphosphates were determined at timed intervals. At the end of the 24-h incubation period, cells were washed and reincubated in drug-free medium so that the decay of each phosphorylated metabolite could be measured. The results are shown in Fig. 6. In the nontransduced cells, only the (N)-MCT monophosphate was detected, and it increased to achieve an intracellular concentration of ∼ pmol/10⁶ cells at 24 h. After drug removal from the medium, the (N)-MCT-MP decayed with a half-life of 2.4 h (Fig. 6A). In the transduced cell line (MC38/HSV-tk), early accumulation of all three phosphorylated metabolites was apparent; however, after the first 5–6 h, only the di- and triphosphate of (N)-MCT continued to accumulate, with the triphosphate showing the most rapid rate of increase and achieving the highest concentrations at 24 h. The decay of the triphosphate appeared to be monophasic with a half-life of 6.5 h (Fig. 6B).

When native and transduced MC38 tumor cells were exposed to increasing concentrations of (N)-MCT ranging from 0.2 to 100 μm, the cellular concentrations of the mono-, di-, and triphosphate increased in cells with the HSV-tk gene, whereas for the wild-type cells, only levels of the monophosphate increased, with no polyphosphorylated metabolites being detected. The formation of the triphosphate levels in the MC38/HSV-tk cells line did not appear to be saturable, reaching concentrations at least 3-fold higher than the levels of mono- and diphosphates (Fig.7).

To pursue the cause of the cytotoxicity engendered by (N)-MCT, its incorporation into host cell nucleic acids was evaluated. Examination of the DNA and RNA extracts from tumor cells [MC38 (wild-type) and MC38/HSV-tk] after a 24-h exposure to the radiolabeled analogues (10 μm) of [methyl-³H](N)-MCT for 24 h was performed. In the MC38 cells, little or no incorporation of (N)-MCT was found. In the DNA of transduced cells (MC38/HSV-tk), significant quantities of (N)-MCT were found (4.7 pmol/μg DNA; Fig. 8). In the aqueous extracts containing cellular RNA, <10% of the levels in DNA were found in RNA for (N)-MCT.

Previous studies have demonstrated that the cytostatic properties of GCV could be reversed by thymidine. To ascertain whether thymidine could exert a similar effect on the antiproliferative properties of (N)-MCT, transduced tumor cells [MC38/HSV-tk (10⁵ cells/well)] were exposed for 24 h to (N)-MCT (10 μm) alone and in the presence of thymidine. Data obtained represent the mean value of experiments performed in quadruplicate. Compared with the normal growth rate of the transduced cell line, (N)-MCT reduced the growth rate to 33% that of the control. When cells were exposed simultaneously to (N)-MCT (10 μm) and thymidine (10 or 50 μm), cell growth was restored to 89 and 62% that of the control, respectively. In addition, thymidine alone only marginally reduced the rate of cell growth to 80 and 71% of the control when the natural nucleoside was incubated with cells at 10 and 50 μm, respectively (data not shown).

Cellular lysate obtained from MC38 and MC38/HSV-tk cells was used to evaluate the affinity of tk for (N)-MCT. (N)-MCT showed at least a 10-fold greater binding affinity for tk in MC38/HSV-tk cells compared with the nontransduced cells. (N)-MCT did not surpass the natural substrate thymidine in their ability to be monophosphorylated in each cell line (Table 2).

As described more fully in “Materials and Methods,” male C57/BL6 mice were injected with 0.1 × 10⁶ MC38 and MC38/HSV-tk cells in the left and right flank, respectively. Seven days after inoculation, mice were injected i.p. twice daily with saline (0.5 ml), (N)-MCT, or GCV (100 mg/kg). Tumor sizes were measured at 1, 3, 5, and 7 days. As illustrated in Fig. 9B, (N)-MCT and GCV completely inhibited tumor growth in the HSV-tk-expressing tumors while showing no inhibitory effect in the control tumors (Fig. 9A). 

To assess the phosphorylation profile of (N)-MCT in MC38 and MC38/HSV-tk tumors in mice, drug metabolites were measured 24 h after the administration of [³H](N)-MCT, 100 mg/kg, in tumor-bearing mice. As shown in Fig. 10, HPLC analysis of methanolic extracts of HSV-tk-transduced tumors demonstrated that most of the radioactivity was accounted for by four principal peaks, identified as the parent compound and the mono-, di-, and triphosphates of (N)-MCT. In nontransduced tumor extracts, a small amount of (N)-MCT-MP was detected, whereas the levels of di- and triphosphates of (N)-MCT were not detectable (Table 3).

The rational design of nucleosides with restricted conformational options is our current approach to developing new drug leads. One result of this approach has been the synthesis of rigid bicyclo-carbocyclic nucleosides such as (N)-MCT with documented antiherpetic properties. For this new thymidine analogue, the natural ribose is restructured to a Northern configured bicyclo[3.1.0]hexane moiety, resulting in a pseudonucleoside with potent activity against both HSV1 and HSV2 in vitro (EC₅₀, 0.01 and 0.12 μg/ml, respectively; Ref. 13). Biochemical and crystallographic studies of (N)-MCT with HSV-tk demonstrate favorable kinetic and binding properties for this nucleoside analogue with the viral kinase (25). In light of these findings, we sought to investigate further the cellular pharmacology and cytostatic properties of (N)-MCT as a potential gene therapy agent in the HSV/tk transduction test system.

We chose to conduct our initial investigations using GCV as a standard drug for comparison. (N)-MCT exerts antiproliferative activity against a murine colon cancer line expressing HSV-tk gene in vitro, which showed that its antiproliferative properties were wholly dependent on tumor cell transduction with the HSV/tk gene and that the potency of this new drug was equivalent to that of GCV. In addition, this antiproliferative activity proved to be demonstrable in vivo. In the HSV-tk-transduced MC38 xenograft model, (N)-MCT and GCV suppressed tumor growth at comparable doses (100 mg/kg i.p. twice daily × 7 days; Fig. 9).

Metabolic studies to elucidate the pharmacology of the (N)-MCT-induced antitumor effect were undertaken in transduced murine colon (MC38/HSV-tk) and murine mammary (FM3A/HSV-tk) tumor cells. It was found that the phosphorylation of (N)-MCT proceeds efficiently to the triphosphate in transduced cells. The initial phosphorylation step is catalyzed predominately by the viral tk, although a very small nontoxic fraction of total (N)-MCT-MP can apparently be produced by the cellular kinase, as evident from the (N)-MCT metabolism studies carried out in the MC38 wild-type cell. Beyond this initial stage, the cellular kinases successively phosphorylate the (N)-MCT monophosphate to the triphosphate, a process that appears not to be saturable at the highest concentrations of (N)-MCT examined (∼100 μm). More importantly, however, the phosphorylation of (N)-MCT to (N)-MCT-DP is catalyzed efficiently by HSV-tk, a property that we have reported elsewhere (26). This ability of HSV-tk to convert the nucleoside monophosphate to the diphosphate is shared by the BVdU group of antiviral agents but not by GCV (19).

The same general profile is seen for GCV, with the triphosphate achieving the highest intracellular concentration among the phosphorylated metabolites (19, 27). High levels of (N)-MCT triphosphate presumably facilitate the incorporation of (N)-MCT into DNA in lieu of thymidine, as has been reported with certain other nucleoside analogues (28, 29). Incorporation of (N)-MCT leads to (N)-MCT tumoricidal activity as corroborated by reversal of (N)-MCT tumoricidal properties by thymidine.

In the present study, we demonstrated that equivalent doses of (N)-MCT or GCV in mice could significantly inhibit tumor growth of s.c.-implanted colon cancer cells transduced with the HSV-tk gene. (N)-MCT and GCV had no effect, however, on tumor growth of nontransduced tumors. Furthermore, we found that (N)-MCT undergoes significant phosphorylation only in HSV-tk tumors, and that its triphosphate metabolite was generated in high yield. These results are consistent with observations from in vitro studies, where high levels of (N)-MCT-TP in HSV-tk transduced cells were measured. This would suggest that the inhibition of HSV-tk tumor growth in vivo, similar to the inhibition of cell proliferation in vitro, is attributable to (N)-MCT-TP, the active metabolite of (N)-MCT. Other investigators have proposed that the in vitro antitumor activity of GCV in HSV-tk transduced cells is the direct result of HSV-tk expression in tumor cells causing the activation of GCV to its cytotoxic triphosphate derivative, e.g., GCV-TP. GCV-TP is then able to compete with endogenous dGTP pools for incorporation into DNA, interfering with cellular DNA synthesis and leading eventually to cell death (19, 30). In a similar manner, it is likely that (N)-MCT-TP competes with dTTP for incorporation into DNA, because we have demonstrated the incorporation of (N)-MCT-TP into DNA of HSV-tk tumor cells in vitro (Fig. 8), resulting in inhibition of DNA synthesis and cell death.

One of the most significant modulators of antiherpetic activity of pyrimidine nucleosides is their susceptibility to cellular phosphorylases. The use of the antiviral BVdU and its analogues has been hampered attributable in part to their rapid degradation of these compounds by thymidine phosphorylase (31). However, pyrimidine carbocyclic analogues have been shown to be more resistant to pyrimidine phosphorylases than their natural 2′-deoxy nucleoside counterparts (12, 32). The apparent resistance of (N)-MCT to thymidine phosphorylase combined with the long half-life of its triphosphate parallels the overall metabolism of GCV, the half-life of which is 15 h in U251-tk-transduced cells (27).

The in vitro and in vivo tumoricidal properties demonstrated here, together with previously published antiviral data, substantiate the assumption that a carbocyclic pseudosugar in the Northern conformation can satisfy the substrate binding requirements for the nucleoside kinases and DNA polymerases required for drug activation. The common feature for both applications is the viral tk. In other instances where rigid carbocyclic nucleosides have been studied, conformational requirements may vary through the drug activation process, such that while one nucleoside analogue conformation may be suitable to initiate phosphorylation, the alternate conformation may be required by the nonhost DNA polymerase (33).

In conclusion, these studies show that the nucleoside analogue (N)-MCT can be used to exert a significant antiproliferative effect in HSV-tk-transduced murine tumor cells in vitro. This tumoricidal property, which can also be elicited in vivo in tumors transduced with HSV-tk, is effected through the phosphorylation of (N)-MCT and its subsequent incorporation into DNA, a mechanism that parallels that of GCV. The structural features of this pyrimidine pseudonucleoside render possible its activation through a combination of viral and cellular enzymes. Further investigations are warranted to evaluate and exploit the unique features of this new lead compound in suicide gene therapy, ascertaining its bystander effect, pharmacokinetics, and antitumor properties in combined therapy with GCV.