Silatecan DB-67 is a novel DNA topoisomerase I–targeted radiation sensitizer Free

The silatecan 7-tert-butyldimethylsilyl-10-hydroxy-camptothecin (DB-67) represents a new generation of camptothecin derivatives that exhibits a potent in vitro DNA topoisomerase I (TOP1)–mediated DNA-damaging activity, improved blood stability, and holds significant promise for the treatment of human cancers. In this study, we characterize the role of TOP1 in mediating the radiosensitization activity of DB-67. As examined by clonogenic survival assay, DB-67 exhibited potent radiosensitization activity at a concentration 10-fold lower than camptothecin in the human glioma D54-MG and T-98G cells, which harbor wild-type and mutant p53, respectively. Analyzed by the single-hit multitarget model, DB-67 induced radiosensitization by obliterating the “shoulder” of the radiation survival curve in the D54-MG cells. The in vivo targeting of TOP1 by DB-67 was investigated by immunoblot analysis. In a dose-dependent manner, DB-67 specifically stimulates covalent linking of TOP1 to chromosomal DNA at concentrations 10-fold lower than camptothecin in the D54-MG cells. The potency of in vivo targeting of TOP1 by DB-67 correlates well with its cytotoxicity and radiosensitization activity. Furthermore, DB-67 exhibited substantially less cytotoxicity and radiosensitization activity in the TOP1 mutant Chinese hamster lung fibroblast DC3F/C-10 cells than in their parental DC3F cells. Together, our data show that DB-67 exhibits potent cytotoxicity and radiosensitization activity by targeting TOP1 in mammalian cells and has great potential for being developed to treat human cancers.

Despite recent advances in anticancer therapy, the treatment outcome for many human malignant diseases, including high-grade gliomas, remains manifestly inadequate. Lately, the combined use of chemotherapeutic drugs with radiotherapy has gained increasing importance in treating human cancers (1, 2). As one added advantage of such a combination, in addition to having their individual cytotoxic effects, certain chemotherapeutic drugs have shown to enhance radiocytotoxicity by inducing radiosensitization (3, 4). Nevertheless, the efficacy of most chemoradiation regimens remains largely limited by the cumulative normal tissue toxicities from combining two modalities. A better understanding of the mechanism of cytotoxic interactions between radiation and chemotherapeutic agents is urgently needed to guide their clinical use. Particularly, the development of potent radiation sensitizers that can selectively enhance radiation cytotoxicity toward cancer cells but not normal tissue cells may hold great promise to improve treatment outcome of various human cancers (4).

The catalytic activity of DNA topoisomerase I (TOP1) is important for many aspects of nucleic acid metabolism in cells, including DNA replication, transcription, and regulation of DNA supercoiling (reviewed in ref. 5). Mammalian TOP1 has also been identified as the major cellular target of a unique class of anticancer drugs (reviewed in refs. 6, 7), including camptothecin derivatives (8), DNA minor groove–binding drugs (9), and indolocarbazole derivatives (10, 11). Instead of direct inhibition of the catalytic activity of TOP1, TOP1-targeted drugs kill cells by trapping the TOP1-DNA cleavable complexes, which damage DNA through interactions with cellular processes, such as replication of DNA (12–14). The presence of elevated TOP1 levels in both proliferating and quiescent tumor cells has provided a molecular basis for targeting TOP1 in cancer therapy (15–17).

Recently, camptothecin derivatives have been shown to enhance radiation cytotoxicity (18–20). Our previous work showed that a stereospecific interaction between the drug molecule and TOP1 is a prerequisite for the induction of radiosensitization by camptothecin in mammalian cells (21). Our finding suggests that compounds capable of targeting TOP1 in a manner similar to camptothecin may exhibit radiosensitization activity (21, 22). Indeed, we have recently shown that some TOP1-targeted indolocarbazole derivatives can induce radiosensitization in mammalian cells (23). The 7-silyl-modified camptothecins (silatecans) represent a new generation of camptothecin derivatives (24–26). 7-Tert-butyldimethylsilyl-10-hydroxy-camptothecin (DB-67) is one of the most active silatecans that has shown significant promise for the treatment of human cancers, including high-grade gliomas in animal models (25, 26). DB-67 exhibits a potent in vitro TOP1-mediated DNA-damaging activity, an improved blood stability, and sufficient lipophilicity to favor blood-brain barrier transit (25, 26). DB-67 has also been shown to inhibit tumor growth in vitro with an ED₅₀ ranging between 2 and 40 ng/mL, which is at least a 10-fold increase in potency compared with the effects observed with topotecan, and is at least comparable with those of SN-38, the active metabolite of camptothecin-11 (26). In the current study, we characterize the radiosensitization activity of DB-67 in human glioma cells harboring either wild-type or mutant p53. In addition, we investigate the in vivo role of TOP1 in mediating the cytotoxic and radiosensitizing activities of DB-67 in mammalian cells. Our data show that DB-67 is a potent cytotoxic agent, as well as an excellent radiation sensitizer that induces these effects through TOP1.

DB-67 (see Fig. 1A for its chemical structure) was synthesized as previously described (27, 28). Camptothecin lactone (NSC 94600) and DMSO were purchased from Sigma Chemical Co. (St. Louis, MO). All drugs were dissolved in DMSO at a concentration of 10 μmol/L and kept frozen in aliquots at −20°C. Except for fetal bovine serum, which was obtained from Inovar Biologicals (Gaithersburg, MD), media and other reagents for tissue culture work were purchased from Life Technologies (Grand Island, NY) and HyClone (Logan, UT).

The Chinese hamster lung fibroblast DC3F cell line and its camptothecin-resistant DC3F/C-10 subline were kindly provided by Dr. Yves Pommier (National Cancer Institute, Bethesda, MD). The DC3F/C-10 cell line was established by selecting the DC3F cell line with camptothecin (27). The amino acid mutation from Gly⁵⁰⁵ to Ser⁵⁰⁵ of the mutant TOP1 has been shown to be responsible for the camptothecin-resistant phenotype of the DC3F/C-10 cell line (28). These cell lines were grown in Dulbecco's minimal essential medium supplemented with 10% heat-inactivated fetal bovine serum, 2 μmol/L glutamine, and 0.1 μmol/L nonessential amino acids, as previously described (27). The wild-type p53-containing human glioma D54-MG cells (29, 30) were kindly provided by Dr. D. Bigner (Duke University Medical Center, Durham, NC) and grown in Dulbecco's minimal essential medium/F-12 Ham's medium supplemented with 10% heat-inactivated fetal bovine serum. The glioma T98G cell line, which harbors a homozygous p53 mutation (31, 32), was purchased from the American Type Culture Collection (Rockville, MD) and grown in RPMI 1640 supplemented with 10% heat-inactivated fetal bovine serum. All cell lines were grown as stock cultures maintained at 37°C in a humidified atmosphere of 5% CO₂ and 95% air at pH 7.3. Under these conditions, the doubling times and plating efficiencies for DC3F, DC3F/C-10, and human glioma cells were 10 hours, 40% to 50%; 15 hours, 45% to 55%; and 20 to 24 hours, 30% to 40%.

For a typical clonogenic survival experiment, stock cultures of exponentially growing cells were trypsinized, plated (5 × 10⁵ cells per dish) into 100 mm Petri dishes, and incubated at 37°C 18 hours before experimental studies. As shown by our previous work, camptothecin radiosensitized log-phased mammalian cells in a schedule-dependent manner, which requires drug treatment before, but not following, radiation (21). Therefore, all experiments were conducted by drug treatment for 30 minutes before exposing cells to graded doses of radiation. Cells were then trypsinized, counted, and plated for microscopic colony formation. Depending on the anticipated survival level, 10² to 10⁵ cells were plated per dose point. When higher cell numbers were plated, larger Petri dishes were used to avoid possible cell density effect (100 mm instead of 60 mm Petri dishes for cell number >2.5 × 10³). Depending on the treatment protocols, an equivalent amount of DMSO was added into each of the control radiation alone dishes (final DMSO concentration <0.1%). Following 7 to 14 days of incubation, colonies were fixed with methanol/acetic acid (3:1) and stained with crystal violet. Colonies consisting of >50 cells were counted. All survival points were measured in triplicate, and experiments were conducted at least twice. Error bars shown in the figures represent SD and are shown when larger than the symbol.

Tissue culture dishes or flasks containing drug-treated and control cells in medium were irradiated on a rotating stand using a cesium-137 irradiator (Mark I, JL Shepherd, Glendale, CA) at a dose rate between 180 and 200 cGy/min. The dose rate of this specific system was periodically monitored by thermoluminescent dosimetry.

Survival curves were corrected for cytotoxicity induced by drug-alone treatment. Sensitization enhancement ratios (SER) and SDs for survival curves were determined at 10% cell survival for DC3F and DC3F/C-10 cells and at 20% for various human glioma cells. Each SER was calculated by dividing the radiation dose in the absence of radiosensitizer by the radiation dose in the presence of radiosensitizer. The updated programs developed by Dr. N. Albright (Department of Radiation Oncology, University of California, San Francisco, CA; ref. 33) were used for curve fitting for D54-MG cells with the linear quadratic and the single-hit multitarget (SHMT) models. Both the linear quadratic and SHMT models gave qualitatively good fits, the fit with the SHMT model being used for the present study.

The TOP1-targeted drugs exert their biological effects by stimulating covalent linkage of TOP1 to chromosomal DNA (reviewed in refs. 6, 7). One way to show TOP1's involvement in this drug-stimulated process is to measure the residual level of free (not cross-linked to DNA) TOP1 in drug-treated cells by immunoblot analysis (8, 9). This measurement has been used to show in vivo targeting of TOP1 by camptothecin and the minor groove–binder Hoechst 33342 (8, 9). Briefly, equivalent numbers of D54-MG cells were plated overnight before being treated with no drug (control with 1% DMSO), 50 μmol/L camptothecin, or 5 or 10 μmol/L DB-67 for 30 minutes. Cell samples were then directly lysed with 2× SDS sample buffer. Proteins from cell lysates (5 × 10⁵ cells/lane) were separated by electrophoresis in a 10% SDS-polyacrylamide gel and transferred onto a nitrocellulose membrane. The membrane was stained with Ponceau to confirm protein loading. TOP1, TOP2, and β-actin bands were detected by immunoblotting the membrane with human anti-TOP1 scleroderma, polyclonal rabbit anti-TOP2α (TopoGene, Columbus, Ohio), and anti-β-actin antibodies (BioDesign International, Saco, ME).

The radiosensitization activity of DB-67 was evaluated by clonogenic survival assay in the human glioma D54-MG cells, which were known to harbor wild-type p53 (29, 30). As shown in Fig. 1B, a significant level of radiosensitization (illustrated by the shifting of the radiation survival curves clockwise) was induced by either 0.2 or 1.0 μmol/L DB-67 in D54-MG cells. Similar to our prior observation showing no dose dependence in the level of radiosensitization induction between 1 and 8 μmol/L camptothecin (21), there was no dose dependence between 0.2 and 1.0 μmol/L DB-67 (Fig. 1B). Based on the radiation dose at 20% survival, a SER of 1.4 could be induced by 30 minutes of either 0.1 or 0.2 μmol/L DB-67 in the D54-MG cells (Table 1). Therefore, the drug concentrations required to induce similar levels of radiosensitization in the D54-MG cells were 10-fold lower for DB-67 (0.1 and 0.2 μmol/L) than camptothecin (Table 1). Consistently, the drug concentrations required to induce similar levels of radiosensitization in the Chinese hamster lung fibroblast DC3F cells were 10-fold lower for DB-67 than camptothecin (Table 2). Our finding provides the first evidence that DB-67 is a good radiation sensitizer in mammalian cells.

The chemoradiation survival curves of D54-MG cells were analyzed by the updated programs developed by Dr. N. Albright (33) for curve fitting with the SHMT model. Good graphical fits were obtained for curves generated by treating cells with no drug (Fig. 2A) or 0.2 μmol/L DB-67 (Fig. 2B) followed by graded doses of radiation. As shown in Fig. 2, a 30-minute treatment with 0.2 μmol/L DB-67 obliterates the shoulder of the radiation survival curve of D54-MG cells. Figure 2C shows the respective radiation survival curve parameters. D₀, which typically denotes the radiation sensitivity of the cells, is defined as the radiation dose required for reducing the survival fraction to 37% of its previous value (34). Extrapolation of the terminal straight line (of slope −1 / D₀) of the survival curve onto the ordinate axis defines a value N, called the extrapolation number. The point at which this line crosses the abscissa for a survival fraction of 100% is called the quasi–threshold dose D_q. Both N and D_q are measures of the width of the shoulder (34). As shown in Fig. 2C, 0.2 μmol/L DB-67 induced no change in D₀ of D54-MG cells (D₀ = 1.2 Gy for both no drug and 0.2 μmol/L DB-67). In contrast, the values of D_q and N were considerably reduced in cells treated with 0.2 μmol/L DB-67 (D_q = 1.1 Gy, n = 2.6 for no drug compared with D_q = 0.0 Gy, n = 1.0 for 0.2 μmol/L DB-67). Therefore, the analysis indicates that obliteration of the shoulder, but not alteration of the terminal slope (1/D₀), of the radiation survival curve is responsible for the observed radiosensitization induction by DB-67 in D54-MG cells (34). These findings are consistent with the current notion that TOP1-targeted drugs induce radiosensitization by inhibiting cellular repair for “sublethal damage” generated by radiation (21–23).

It has been estimated that more than half of all human malignant diseases, including high-grade gliomas, are associated with alterations of the p53 tumor suppressor gene (35–37). In addition, our recent study also indicated that p53 might modulate the induction of radiosensitization by camptothecin derivatives in human carcinoma cells (38). In view of the potential importance of p53 status in affecting the clinical application of DB-67, the radiosensitization activity of DB-67 was evaluated in the human glioma T98G cells, which harbor a homozygous p53 mutation (31, 32). As shown in Table 1, a 30-minute treatment of 0.1 μmol/L DB-67 induced a similar cytotoxic effect in the T98G cells (cell survival = 40 ± 6%) as did a 30-minute treatment of 0.1 or 0.2 μmol/L DB-67 in the D54-MG cells (cell survival = 59 ± 22% and 46 ± 12 %, respectively). Therefore, at equally cytotoxic conditions, DB-67 exhibits radiosensitization activity in the mutant p53-containing T98G cells (SER = 1.2), albeit to a lesser degree than in the wild-type p53-containing D54-MG cells.

DB-67 has been shown to be a potent inducer of TOP1-mediated DNA damage in a purified enzyme system (25, 26). However, the in vivo targeting of cellular TOP1 by DB-67 has not been fully characterized. The D54-MG cells were treated with no drug, 50 μmol/L camptothecin, or 5 or 10 μmol/L DB-67 for 30 minutes, and total cell lysates underwent immunoblot analysis as described in Materials and Methods. As shown in Fig. 3, DB-67 is 10-fold more potent than camptothecin in depleting cellular TOP1 (5 μmol/L DB-67 induces a higher amount of band depletion than 50 μmol/L camptothecin). The depletion of TOP1 by DB-67 seems to be specific because neither TOP2 nor β-actin protein band from the D54-MG cells was affected (Fig. 3). Consistent with the notion that TOP1 is the in vivo target of DB-67 in mediating its cytotoxicity and radiosensitization activity, the potency of in vivo TOP1 band depletion by DB-67 correlates well with its cytotoxicity and radiosensitization activity in the D54-MG cells (Fig. 1B; Table 1).

The cytotoxic effects of camptothecin and DB-67 were examined in the Chinese hamster lung fibroblast DC3F and its camptothecin-resistant TOP1 mutant DC3F/C-10 cells (27, 28). Cells were treated with DB-67 or camptothecin for 30 minutes before plating cells for clonogenic survival analysis as described in Materials and Methods. As shown in Table 3, consistent with our observation in the human glioma cells, based on the LD₅₀, DB-67 (0.05 ± 0.01 μmol/L) is 16-fold more cytotoxic than camptothecin (0.8 ± 0.1 μmol/L) in the DC3F cells. Also, the mutant TOP1-containing DC3F/C-10 cell line exhibited more than a 10-fold resistance to camptothecin and a 35-fold cross-resistance to DB-67 than the DC3F cell line (Table 3). This finding indicates that TOP1 is an in vivo cytotoxic target of DB-67 in mammalian cells.

The radiosensitization activities of camptothecin and DB-67 were also examined in the DC3F cell line and its camptothecin-resistant TOP1 mutant DC3F/C-10 cell line. Cells were treated with drugs (≤0.1% DMSO for no drug control) for 30 minutes followed by irradiation. As shown in Table 2, both camptothecin and DB-67 induced a significant enhancement of radiocytotoxicity in DC3F cells. Based on the radiation dose at 10% survival, the SERs induced by DB-67 (range, 1.3–1.5) were similar to those induced by camptothecin (range, 1.3–1.4) in the DC3F cells (Table 2). Consistent with our findings in the D54-MG cells (Table 1), the drug concentrations required to induce radiosensitization in the DC3F cells were 10-fold lower for DB-67 than camptothecin (Table 2). The camptothecin-resistant DC3F/C-10 cell line exhibited similar radiation sensitivity as the DC3F cell line (Table 2); however, the radiosensitization activities of camptothecin and DB-67 were both significantly reduced in the TOP1 mutant DC3F/C-10 cells (Table 2). Indeed, camptothecin and DB-67 exhibited considerably reduced radiosensitization activity in the DC3F/C-10 cells even at higher drug concentrations (SER = 1.2 at 2.5 μmol/L DB-67; SER = 1.1 at 10 μmol/L camptothecin). Our findings indicate that mammalian TOP1 is the in vivo radiosensitization target for both camptothecin and DB-67.

Low solubility, as well as difficulty in obtaining sufficient quantity from plant materials, have hindered early clinical development of camptothecin (39). Recently, hydrolysis of the lactone ring, which converts the active “ring-closed” lactone form into the inactive “ring-open” carboxylate form of camptothecins, has also been recognized to be a factor that may compromise the clinical efficacy of camptothecin derivatives (40). To overcome these impediments, extensive efforts have been made toward fully synthesizing new camptothecin derivatives that can readily reach a tumor in the active lactone form (40). Toward this aim, DB-67, one of the most active silatecans, can be fully synthesized using a cascade radical annulation approach (24, 25). In addition, DB-67 contains the key α-hydroxy-δ-lactone pharmacophore and is more stable than other clinically available camptothecin derivatives with a physiologic pH (25, 26, 40). The dual substitution at the 7 and 10 positions, by abolishing the high-affinity binding of the carboxylate form to human serum albumin, further stabilizes DB-67 at its active lactone form in human blood (26, 40). Due to its sufficient lipophilicity, which favors blood-brain barrier transit, DB-67 has shown its antiproliferative activity in a nude mouse intracranial model and has shown significant promise for the treatment of human cancers, including high-grade gliomas of the central nervous system (26).

Mammalian TOP1 is the major cytotoxic target of camptothecin derivatives (reviewed in refs. 6, 7). Although previous studies have shown that DB-67 can stimulate TOP1-medicated DNA cleavage in purified enzyme systems (24–26), the possibility exists that DB-67, secondary to the structural modifications on the 7 and 10 positions of the camptothecin pharmacophore, may interact with and exert its biological effect(s) via other intracellular targets in addition to TOP1. Thus, our current data confirm the previous observation by demonstrating >10-fold potency in the cytotoxicity of DB-67 compared with camptothecin in mammalian cells (Tables 1 and 2). In addition, our results on the specific stimulation of covalent linking of TOP1 by DB-67 (Fig. 3), as well as the high-level cross-resistance of the mutant TOP1-containing DC3F/C-10 cells to DB-67 (Table 3), provide direct evidence that TOP1 is the major cytotoxic target of DB-67 in vivo. Our finding indicates that the 7-tert-butyldimethylsilyl-10-hydroxy modification on the parental camptothecin pharmacophore does not alter the drug-TOP1 interaction and, as a result, preserves its in vivo DNA damaging activity. This is consistent with the general understanding of the interaction between TOP1 and camptothecin derivatives in trapping TOP1 cleavable complexes (see review in ref. 41).

The current study is the first to show that DB-67 is a novel radiation sensitizer for mammalian cells. In both the Chinese hamster DC3F cells and the human glioma D54-MG cells, DB-67 induced a good SER value of 1.4 at drug concentrations that were 10-fold lower than camptothecin (Tables 1 and 2). This finding indicates that the drug levels required for DB-67 to exert its potent radiosensitizing effect are more clinically achievable than camptothecin, particularly in the central nervous system, which is guarded by the blood-brain barrier. We have previously shown that camptothecin derivatives induced radiosensitization in mammalian cells through interaction with TOP1 (21). One prediction based on this finding is that cells containing mutant TOP1 may be less sensitive to the radiosensitizing effect of DB-67 (21, 22). Indeed, DB-67 exhibited significantly lower radiosensitizing effects in the TOP1 mutant DC3F/C-10 cells (Table 2). This finding confirms our prediction and supports the notion that TOP1 mediates the radiosensitizing effect of DB-67 in mammalian cells.

As with other TOP1-targeted drugs (reviewed in ref. 42), obliteration of the shoulder of the survival curve but not its terminal slope (1 / D₀) seems to account for the observed radiosensitization effect induced by DB-67 (Fig. 2). The shoulder of the radiation survival curve generally represents the sublethal damage repair capacity of the irradiated cells for ionizing radiation (1, 37). TOP1-targeted drugs are known to induce various types of DNA damages by trapping the TOP1-DNA cleavable complexes in actively proliferating cells (6–10). Among them, double-strand DNA breaks, TOP1-linked DNA breaks, and the arrest of DNA replication forks have been observed in a cell-free SV40-based in vitro replication system (43). It is plausible that cellular machinery responsible for repairing sublethal DNA damages caused by ionizing radiation may be “saturated” by the DNA damages induced by DB-67, and therefore accounts for the “shoulder-obliterating” radiosensitization (44).

More than half of all human cancers, including high-grade gliomas, are associated with alterations of the p53 tumor suppressor gene (35–37). Fulci et al. (45) reported that the p53 pathway is disrupted in >80% of malignant gliomas (WHO grades III and IV), either by mutation of the p53 gene (∼30%) or by p14arf deletion (∼50%). The importance of p53 in cancer therapy is also reflected by the fact that p53 can be activated by various DNA damaging agents, including camptothecin treatment (46, 47). Our recent work has shown that camptothecin induces a reduced level of TOP1-mediated radiosensitization in cultured human cancer cells deficient in p53 (38)⁵.

In an agreement with this observation, DB-67 exhibited a moderate radiosensitizing effect in the mutant p53-containing human glioma T98G cells albeit to a lesser degree than in the wild type p53-containing human glioma D54-MG cells. Due to the relatively lower levels of TOP1 in nonmalignant cells (15–17), our data suggest that a clinical benefit can still be achieved with combination chemoradiation with DB-67 in patients suffering from malignant gliomas, which harbor an altered p53 pathway. In addition, through careful radiation treatment planning and by using advanced technologies, radiation oncologists can further avoid the enhanced normal tissue toxicity due to the DB-67-induced radiosensitization effect by minimizing radiation dose to normal tissue. Preclinical studies, including human cancer xenografts in nude mice studies, are needed to confirm the benefit of treating human cancers by combining radiation and DB-67.

In summary, our data showed that DB-67 synergistically enhanced radiation cytotoxicity in mammalian cells at a >10-fold lower drug concentration than camptothecin. We provide direct evidence that TOP1 is the in vivo mediator for both the cytotoxic and radiosensitizing effects of DB-67 in mammalian cells. Shoulder obliteration seems to account for the mechanism of DB-67–induced radiosensitization in the wild-type p53-containing human glioma D54-MG cells. Albeit to a lesser degree, DB-67 also exerted a moderate radiosensitizing effect in the mutant p53-containing human glioma T98G cells. The silatecan DB-67, a novel TOP1-targeted drug generated through rational drug development, may hold significant promise for use in combination with radiotherapy for the treatment of human cancers, including high-grade gliomas of the central nervous system. Our study also provides a rationale for proceeding with further preclinical evaluation of the efficacy and safety of chemoradiation with DB-67.

This article is dedicated to the memory of Dr. Thomas G. Burke, a devoted cancer pharmacologist and a driving force in the development of DB-67, who passed away in 2003.

We thank Dr. Julian Perks for his helpful comments during preparation of the manuscript and Philip Boerner for critical reading of the manuscript.