The pentacyclic palmarumycins are structurally unique natural products with both antifungal and antibacterial activities but their antineoplastic effects are not well established. We have examined their antiproliferative actions against tumor cells using a temperature-sensitive tsFT210 mouse mammary carcinoma cell line and found that a novel palmarumycin analog, [8-(furan-3-ylmethoxy)-1-oxo-1,4-dihydronaphthalene-4-spiro-2′-naphtho[1",8"-de][1′,3′][dioxin] or SR-7, prominently blocked mammalian cell cycle transition in G2/M but not in G1 phase. We found no evidence for inhibition of the critical mitosis-controlling cyclin-dependent kinase Cdk1, or its regulator, the dual specificity phosphatase Cdc25. Moreover, Cdk1 was hypophosphorylated and not directly inhibited by SR-7. SR-7 also failed in vitro to hypernucleate bovine tubulin, did not compete with colchicine for tubulin binding, and only modestly blocked GTP-induced assembly. In addition, SR-7 caused almost equal inhibition of paclitaxel-sensitive and -resistant cell growth. Moreover, unlike benchmark tubulin-disrupting agents, SR-7 did not cause hyperphosphorylation of the antiapoptotic protein Bcl-2. Thus, SR-7 represents a novel chemical structure that can inhibit G2/M transition by a mechanism that appears to be independent of marked tubulin disruption.
Cell cycle checkpoints are important regulators of eukaryotic cell growth. Because loss of the G1 checkpoint often occurs in malignant cells, synthetic, small molecule, modulators of the G2/M checkpoint are of particular interest. Several clinically important anticancer compounds, such as the vinca alkaloids and taxanes, act as prominent inhibitors of G2/M phase transition, thus validating the concept that disruption of G2/M transition can be an effective mechanism for controlling some cancers. Nonetheless, all of these compounds disrupt tubulin directly and they have complex chemical structures that restrict chemical modification. In addition, several prominent disrupters of tubulin, such as nocodazole and colchicine, lack antitumor efficacy. Therefore, novel chemical structures that block G2/M phase transition are valuable as pharmacological probes and represent possible lead structures for future therapeutic agents.
The naphthoquinone acetals, palmarumycins, diepoxins, and deoxypreussomerins (Fig. 1) are structurally unique fungal metabolites with both antifungal and antibacterial activities (Schlingmann et al., 1993; Krohn et al., 1994; Wipf and Jung, 1998) but their antiproliferative activity against malignant mammalian cells has not been extensively studied. Biological studies have been limited partly due to the extraordinary synthetic challenges associated with the extensive levels of oxygenation and the highly electrophilic functionality present in these spiroketal natural products. We have developed an efficient synthetic approach toward palmarumycins, diepoxins, and deoxypreussomerins (Wipf and Jung, 1999) and have generated a small focused library of analogs (Wipf et al., 2001). In a preliminary report we found that a number of these palmarumycins analogs had antiproliferative activity against human cancer cells (Wipf et al., 2001). We have now extended these studies by focusing on one of the most active members of the library, 8-(furan-3-ylmethoxy)-1-oxo-1,4-dihydronaphthalene-4-spiro-2′-naphtho[1",8"-de][1′,3′]dioxin, or SR-7, which retained the antiproliferative activity of palmarumycin CP1 and diepoxin against the estrogen-receptor positive, p53 replete human MCF-7 breast cancer cells, the estrogen-receptor negative, p53 deficient human MDA-MB-231 breast cancer cells and virally transformed mouse fibroblasts.
Materials and Chemicals.
The synthesis and initial antiproliferative evaluation of the palmarumycin/diepoxin analogs have been described elsewhere (Wipf et al., 2001). The synthesis, and biochemical and cellular properties of the Cdc25 inhibitor SC-ααδ9 have also previously been published (Rice et al., 1997;Tamura et al., 1999). Curacin A was prepared as described previously (Wipf and Xu, 1996). tsFT210 cells, which contain a temperature-sensitive mutant form of Cdk1 allowing for convenient cell cycle synchronization, were a gift from Dr. Chris Norbury (Oxford University, Oxford, UK) and were maintained for no longer than 30 passages (Th'ng et al., 1990). Paclitaxel-resistant (1A9/PTX10, 1A9/PTX22) and parental 1A9 human ovarian carcinoma cells were gifts from Drs. Paraskevi Giannakakou and Tito Fojo of the National Cancer Institute (Bethesda, MD). The SV40 large T antigen transformed cells have been previously characterized (Vogt et al., 2000). Anti-Cdk1 (sc-54), anti-Cdc25, and anti-Bcl-2 (sc-509) antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). An agarose conjugate of anti-Cdk1 was used for immunoprecipitation. Histone H1 was obtained from Boehringer-Mannheim (Indianapolis, IN) and [γ-32P]ATP (10 mCi/mmol) was from Amersham Life Science, Inc. (Arlington Heights, IL). Colchicine [ring C, methoxy-3H] (61.4 Ci/mmol, 2.3 TBq/mmol) was from NEN (Boston, MA). Paclitaxel was obtained from the Drug Synthesis Branch of the National Cancer Institute. All other reagents were from Sigma (St. Louis, MO) unless indicated otherwise.
The proliferation of MCF-7, MDA-MB-231 and SV40 transformed mouse embryonic fibroblasts was measured by a previously described colorimetric assay (Vogt et al., 1998; Wipf et al., 2001). Briefly, we seeded 4000 to 6000 cells/well in microtiter plates. Cells were allowed to attach overnight and treated with vehicle or compounds for 72 h, after which the medium was replaced with serum-free medium containing 0.1% 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide. Plates were incubated for 3 h in the dark and total cell number was determined spectrophotometrically at 540 nm as previously described (Vogt et al., 1998). The growth inhibition of 1A9, 1A9/PTX10, and 1A9/PTX22 cells was also measured with a slightly different colorimetric assay. Cells were maintained in RPMI 1640 medium containing 10% fetal bovine serum and the paclitaxel-resistant cells also contained 17 nM paclitaxel and 10 μM verapamil. Cells were plated (2000/well) in 96-well plates and allowed to attach and grow for 72 h (paclitaxel and verapamil were removed from resistant cell medium 2 weeks before this plating). They were then treated with 0.5% DMSO vehicle control or 0.08 to 10 μM compound. The number of cells was determined spectrophotometrically at 490 nm minus absorbance at 630 nm after exposure to 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium and N-methylphenazine methyl sulfate.
Flow Cytometry Analysis.
tsFT210 cells were seeded at 2 × 105 cells/ml and maintained at 32.0°C as previously described (Th'ng et al., 1990; Tamura et al., 1999). Cell proliferation was blocked at G2 phase by incubation at 39.4°C for 17 h. The synchronized cells were then released by reincubating at 32.0°C and treated immediately with 0 to 10 μM SR-7, 1 μM nocodazole, or 100 μM SC-ααδ9, respectively, to probe for G2/M arrest. Cells were treated 6 h after G2/M release to determine G1 arrest. We used 100 μM SC-ααδ9 and 50 μM roscovitine as positive control compounds for G1 arrest. A final concentration of 0.5% DMSO was used for all compounds and as a negative control. For both G2/M and G1 blockage studies, treated cells were incubated at 32.0°C for an additional 6 h after each drug exposure, and then harvested with phosphate-buffered saline at 5 × 105cells/ml. The harvested cells were stained with a solution containing 50 μg/ml propidium iodide and 250 μg/ml RNase A. Flow cytometry analysis was conducted with a Becton Dickinson FACS Star (Franklin Lakes, NJ). Cell cycle distribution was statistically analyzed using a one-tailed Student's t test assuming unequal variances.
Western Blotting and Cdk1 Assays.
tsFT210 cells were harvested using the same procedure for cell synchronizing and drug exposure as described above for the G2/M flow cytometric analysis. The protein lysates were analyzed by Western blotting for Cdk1 as described previously (Tamura et al., 2000). We used Cdk1 isolated from human MCF-7 cells to assay for in vitro inhibition of enzyme activity because of available antibodies and convenience. Asynchronous cells grown at 37°C in 5% CO2 in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum were treated with lysis buffer and harvested as previously described (Vogt et al., 1998). Cdk1 kinase activity assay was performed as previously described (Yu et al., 1998). Briefly, 2 mg of the protein lysates was incubated with anti-Cdk1 antibody agarose conjugate for 2 h at 4°C. The immunoprecipitates were treated in vitro with DMSO vehicle, 300 nM flavopiridol, or 10 μM SR-7 for 20 min at 30°C. The treated immunoprecipitates were reincubated in 20 μl of kinase reaction buffer (Yu et al., 1998) for an additional 20 min at 30°C with 3 μg of histone H1, 20 mM Tris-HCl, 10 mM MgCl2, 5 μM cold ATP, and 10 μCi of [γ-32P]ATP. Histone H1 was separated from other proteins by SDS-PAGE and analyzed for incorporation of radioactive phosphate with a Molecular Dynamics (Sunnyvale, CA) STORM 860 PhosphoImager.
Tubulin without microtubule-associated proteins was isolated from fresh bovine brains (Hamel and Lin, 1984). Inhibition of assembly reactions was carried out as described previously (Verdier-Pinard et al., 1998). Tubulin (1 mg/ml) was preincubated for 15 min at 30°C with compounds, 4% DMSO, and 0.8 M monosodium glutamate. The reaction mixtures were cooled to 0°C, adjusted to 0.4 mM GTP, and transferred to cuvettes in a Beckman-Coulter 7400 spectrophotometer with a six-cuvette holder on a motorized stage reading absorbance at 340 nm. Baselines were established at 0°C and the temperature was raised to 30°C in approximately 1 min with an electronically controlled Peltier temperature controller. The change in absorbance 20 min after samples reached 30°C was used to calculate the extent of polymerization. The change in absorbance for vehicle without GTP was considered 100% assembly inhibition, whereas the change in absorbance for GTP-containing reaction mixtures with DMSO vehicle was used for 0% inhibition. Each series of determinations included positive and negative control samples.
For microtubule-stabilization/hypernucleation reactions, baselines were established with reaction mixtures containing glutamate and tubulin held at 0°C by the temperature controller. Compound (1–50 μM) or an equivalent volume of DMSO was added, the reactants were quickly mixed, and temperature was changed to 30°C as described above. In parallel reactions the temperature was raised to 37°C. Paclitaxel (10 μM) was the positive control and DMSO was the negative control.
Inhibition of [3H]Colchicine Binding.
Using methods described previously (Verdier-Pinard et al., 1998), we incubated 5 μM [3H]colchicine with either 5% DMSO vehicle or compound (5 or 50 μM) at 37°C for 15 min with 1 μM tubulin in the presence of 1 M monosodium glutamate, 0.1 M glucose-1-phosphate, 1 mM MgCl2, 1 mM GTP, and 0.5 mg/ml bovine serum albumin. The solutions were filtered through two stacks of DEAE-cellulose filters and the radioactivity in the filtrate was determined by liquid scintillation spectrometry. Each series of determinations included positive controls of 5 and 50 μM curacin A.
We measured the activities of the GST-fusion proteins Cdc25B2, vaccinia H1-related phosphatase, and human recombinant PTP1B with the assay conditions that have previously been described (Rice et al., 1997). Fluorescence emission from the product of the substrate, O-methyl fluorescein phosphate (Molecular Probes, Inc., Eugene, OR), was measured over a 20- to 60-min reaction period at ambient temperature with a multiwell plate reader (PerSeptive Biosystems Cytofluor II; Framingham, MA; excitation filter/band width, 485/20; emission filter/band width, 530/30). All enzyme reactions were linear over the incubation time and were directly proportional to both the enzyme and substrate concentration.
Attempts to determine the phosphorylation status of Bcl-2 in tsFT210 cells using two different antibodies were unsuccessful due to the antibodies' inability to detect mouse Bcl-2 or the lack of specificity. Therefore, phosphorylated and nonphosphorylated Bcl-2 was detected in lysates from human MCF-7 cells (American Type Culture Collection, Manassas, VA) treated with microtubule-perturbing agents. Equal amounts of protein were separated by electrophoresis on 15% SDS-PAGE followed by immunoblotting with an anti-human Bcl-2 antibody (sc-509; Santa Cruz Biotechnology). Positive antibody reactions were visualized using peroxidase-conjugated secondary antibodies (Jackson ImmunoResearch, West Grove, PA) and an enhanced chemiluminescence detection system (Renaissance; NEN) according to manufacturer's instructions.
Cytotoxicity and G2/M Phase Inhibition by SR-7.
MCF-7 cells showed similar sensitivity to the antiproliferative actions of palmarumycin CP1, diepoxin ς, and SR-7 with IC50 values of 0.96, 1.64, and 1.13 μM, respectively (Fig. 2A). In contrast, the close structural analog SR-4 had considerably less antiproliferative activity against these cells with an IC50 of 11.2 μM (Fig. 2A). Because of the potential importance of the tumor suppressor gene p53 and the estrogen receptor in controlling the cellular response to cytotoxic agents, we also examined the sensitivity of MDA-MB-231 cells, which lack functional p53 and estrogen receptors. As indicated in Fig. 2B, MDA-MB-231 cells were equally sensitive to palmarumycin CP1, diepoxin ς, and SR-7, with IC50 values of 2.61, 2.01, and 2.44 μM, respectively. In contrast, the IC50 for SR-4 was 10.3 μM, revealing the importance of the 2-furyl moiety in the C8 position of SR-7. As might be expected, these p53 and estrogen receptor-deficient cells were generally less sensitive to both natural products and analogs compared with MCF-7 cells. The differential cytotoxicity of the SR-7:SR-4 pair was confirmed when we tested mouse embryonic fibroblasts transformed with SV40 large T antigen; we found 3 μM SR-7, palmarumycin CP1, and diepoxin ς were required to inhibit growth by 50%, whereas no significant inhibition was seen with 10 μM SR-4, the highest concentration tested (data not shown).
Because SR-7 shared with its fellow palmarumycin-related compounds potent growth inhibition against several malignant cell types, because it was readily available, and because a minor structural modification in the C8 position of SR-7 resulted in marked differences in compound activity, we examined the SR-7 and SR-4 pair in greater detail. We probed the cell cycle specificity of SR-7 growth inhibition using murine tsFT210 mammary carcinoma cells, because they can be readily synchronized without exogenous compounds due to a temperature-sensitive Cdk1 (Th'ng et al., 1990). When incubated at the permissive temperature of 32.0°C, tsFT210 cells had a normal cell cycle distribution (Fig. 3A). Culturing tsFT210 cells at the restrictive temperature (39.4°C) for 17 h resulted in a significantly increased percentage of G2/M phase cells from 37.8 to 59.0% (p = 0.038), and a decrease in G0/G1 cells from 40.6 to 14.8% (p = 0.004) (Fig. 3, A and B; Table1) due to Cdk1 inactivation (Th'ng et al., 1990). When G2/M arrested cells were cultured at the permissive temperature for 6 h with DMSO vehicle alone, we saw clear evidence of entry into G1(2C) (Fig. 3C) with a significant proportion (31.4%, p= 0.015) accumulating in G1, although 52.5% of the cells remained in the G2/M phase. This G2/M retention at 4C is probably due to the extended cell cycle blockage at 39.4°C (Osada et al., 1997). The proportion of G1 phase cells was significantly lower when cells were released in the presence of either nocodazole (8.5%, p = 0.011) or SR-7 (16.9%, p = 0.019), compared with DMSO (Table 1). Treatment with 1 μM nocodazole blocked cell passage through G2/M (Fig. 3D). To determine the effect of SR-7 on G2/M cell cycle transition, we treated cells with 2.5 to 10 μM SR-7 for 6 h after releasing cells at 32.0°C. As indicated in Fig. 3, E–H, SR-7 caused a concentration-dependent arrest in the G2/M phase, with obvious blockage even with 2.5 μM SR-7. The percentage of G2/M cells treated with nocodazole (77%) or SR-7 (67.2%) was not significantly different than that in cells arrested at the restrictive temperature (p = 0.074 and 0.159, respectively). Thus, SR-7 (10 μM) was as active as nocodazole (1 μM) in preventing cells from progressing to G1. The G2/M inhibition was similar to that seen with the previously reported and structurally unrelated compound SC-ααδ9 (Fig. 3I). SC-ααδ9 is an inhibitor of all three of the Cdc25 phosphatases, which control cell cycle checkpoints (Rice et al., 1997).
We next examined G1 transition in tsFT210 cells after SR-7 treatment. We again arrested tsFT210 cells at G2/M by shifting to the nonpermissive temperature and then released cells into G1 phase by returning to the permissive temperature. In these experiments, however, we added DMSO vehicle, roscovitine, SR-7, or SC-ααδ9 6 h after G2/M phase release. Cells that were treated with the DMSO vehicle passed through G1 phase as expected and produced the predicted broad S-phase peak between diploid (2C) and tetraploid (4C) states (Fig.4D), whereas cells exposed continuously to 50 μM roscovitine were blocked and did not pass through G1 (Fig. 4E). As illustrated in Fig. 4, F and G, cells treated with 5 or 10 μM SR-7 were not delayed at G1. The results of two independent experiments are quantified in Table 2. Thus, 6 h after the G2/M block release, the percentage of S-phase cells in the DMSO control samples increased approximately 3-fold (from 12.8 to 36.8%, DMSO, 12 h), indicating cell cycle progression through the G1/S checkpoint. An S-phase accumulation of similar magnitude was observed in cells treated with SR-7 (33.1%). In contrast, cells treated with roscovitine, an inhibitor of cyclin-dependent kinases, prevented cell cycle progression through G1/S and had S-phase cell numbers (16.9%) comparable to the cells at the initiation of treatment (12.8%). As expected from our previous studies (Tamura et al., 2000), the dual phase-specific inhibitor SC-ααδ9 caused a prominent G1 block and also prevented cells that were at the G2/M interphase from progressing, which resulted in two prominent cell cycle peaks (Fig. 4H).
Cdk1 Dephosphorylation in the Presence of SR-7.
The major controlling molecule for G2/M transition is the cyclin-dependent kinase Cdk1, whose cellular activity is tightly regulated by phosphorylation (Hunter and Pines, 1994; Pines, 1999). Therefore, we performed Western blotting on tsFT210 cell extracts to determine the Cdk1 phosphorylation level in the presence or absence of SR-7. Protein lysates of tsFT210 cells arrested at the G2/M boundary were harvested and analyzed by SDS-PAGE. Similar to our previous observations (Tamura et al., 1999), approximately 50% of Cdk1 was in the mitotic-inactive hyperphosphorylated form as reflected by a slower migrating Cdk1 (Fig.5, lane 1). The phosphorylation of Cdk1 decreased gradually after cells were released from G2/M block, and most of the Cdk1 was dephosphorylated, and thus activated, 6 h after G2/M release, even in the presence of the DMSO vehicle (Fig. 5, lanes 2–4). When we incubated cells with 1 μM nocodazole, which caused a G2/M arrest, no hyperphosphorylation of Cdk1 was seen, consistent with its proposed inhibitory activity after Cdk1 activation (Fig. 5, lane 5). Similarly, Cdk1 was completely dephosphorylated in the presence of either 10 or 20 μM SR-7 (Fig. 5, lanes 6 and 7). In contrast, treatment with 100 μM SC-ααδ9, which also causes G2/M block (Tamura et al., 1999), yielded a hyperphosphorylated Cdk1 (Fig. 5, lane 8). In vitro studies confirmed that SR-7 and SR-4 at 30 μM caused no inhibition of recombinant Cdc25B2, vaccinia H1-related phosphatase, or PTP1B activity; even at 100 μM we found ≤16% inhibition (data not shown). We also examined the ability of SR-7 to directly inhibit Cdk1 kinase activity. Cellular Cdk1 was immunoprecipitated with an anti-Cdk1 antibody and the resulting protein treated with DMSO vehicle, 0.3 μM flavopiridol, or 30 μM SR-7 for 20 min. Exposure to 0.3 μM flavopiridol, a known Cdk1 inhibitor, completely blocked the ability of the immunoprecipitated Cdk1 to phosphorylate histone H1, whereas treatment of the immunoprecipitate with 30 μM SR-7 did not inhibit activity (Fig.6). We are uncertain whether the apparent increase in Cdk1 activity seen in Fig. 6 was biologically important. Nonetheless, the G2/M blockage produced by SR-7 was clearly not associated with hyperphosphorylation or direct inhibition of Cdk1 or inhibition of Cdc25 enzyme activity.
SR-7 Effects on Tubulin.
We next examined the ability of SR-7 to alter tubulin polymerization or depolymerization in vitro. Addition of 0.4 mM GTP to isolated bovine brain tubulin produced robust polymerization that began to plateau approximately 20 min after microtubule assembly commenced (Fig. 7A). Inclusion of 5 μM curacin A completely inhibited GTP-induced tubulin assembly, whereas 1 μM curacin A caused a 50% inhibition. In contrast, SR-7 even at 40 μM caused only moderate inhibition of tubulin assembly (Fig. 7A). Furthermore, we found no evidence that SR-7 could bind to the colchicine site of tubulin. As indicated in Table3, at 5 μM SR-7 failed to significantly inhibit colchicine binding, whereas the positive control curacin A caused almost 90% inhibition at 5 μM. At 50 μM, SR-7 actually enhanced [3H]colchicine binding by an unknown mechanism. Thus, we concluded SR-7 did not compete for the most common small-molecule target on tubulin, the colchicine binding site, although we have not formally excluded the possibility that SR-7 could bind to some other site, such as that used by the vinca alkaloids.
We next examined drug-induced polymerization of isolated bovine brain tubulin to test whether SR-7 had paclitaxel-like activities. At 30°C paclitaxel (10 μM) appeared to be equivalent to 0.4 mM GTP in causing tubulin assembly, whereas SR-7 at concentrations ranging from 5 to 40 μM caused no hypernucleation of isolated tubulin (Fig. 7B). Incubation at 37°C also showed SR-7 to be without paclitaxel-like activity. We also examined the sensitivity of three human A2780 ovarian carcinoma cell lines, the parental clonal line 1A9 (Behrens et al., 1987) and two derived lines made paclitaxel-resistant by incubating 1A9 cells with increasing concentrations of paclitaxel in the presence of verapamil: 1A9/PTX10 and 1A9/PTX22, which are 32- and 43-fold resistant to paclitaxel, respectively (Giannakakou et al., 1997). The median growth inhibitory concentrations for SR-7 were 1A9 parental cells, 0.81 ± 0.1 μM; 1A9/PTX10 cells, 1.1 ± 0.1 μM; and 1A9/PTX22 cells, 1.3 ± 0.1 μM (mean ± S.E.M.,N = 9). Thus, SR-7 was a potent inhibitor of proliferation in these human tumor cell lines and there was no marked cross-resistance toward SR-7 in the paclitaxel-resistant cells.
Effect of SR-7 on Bcl-2 Phosphorylation.
All known microtubule-disrupting compounds cause hyperphosphorylation of the antiapoptotic protein Bcl-2 (Haldar et al., 1995; Basu and Haldar, 1998). Thus, we examined the phosphorylation status of Bcl-2 in cells treated with various compounds, including SR-7. Although paclitaxel, nocodazole, and vincristine caused prominent phosphorylation of Bcl-2, SR-7 at either 3 or 10 μM failed to generate obvious hyperphosphorylated Bcl-2 (Fig. 8). Densitometric evaluation revealed 40% of the Bcl-2 was phosphorylated after nocodazole treatment, whereas <2% of the Bcl-2 was phosphorylated in control or SR-7-treated cells. These results strongly suggested SR-7 did not directly disrupt tubulin in whole cells.
The unique chemical structure of the natural products palmarumycins and diepoxins and their potential for functionalization suggest their biological activity should be examined in greater detail. Our cell culture results demonstrated compounds of this general class are potent inhibitors of cell division. Interestingly, the preservation of antiproliferative activity against murine and human malignant cells with different p53 and estrogen receptor status by both palmarumycin CP1 and SR-7 indicated that the epoxide moieties of diepoxin ς are not essential for cell growth inhibition. Furthermore, the retention of antiproliferative activity by SR-7 revealed that the C8 position of the palmarumycin/diepoxin structure could be modified to form unnatural analogs but the loss of activity with SR-4 demonstrated that the nature of this substitution was important. Based on this study and others (Wipf et al., 2001), we have concluded that allylic π-systems present in heterocyclic or aromatic ethers lead to a significant increase mammalian cell growth inhibition.
When we examined SR-7 in greater detail, we found it blocked tsFT210 cells at a G2/M checkpoint. Cell cycle checkpoints are critical regulators of genome integrity and faithful cell replication. One of the main abnormalities in human tumor cells is the loss of the G1-phase checkpoint, which not only permits cellular replication but also encourages genomic instability. Consequently, enforcement of the G2/M checkpoint represents an attractive mode of action for new antineoplastic agents. G2/M progression is tightly regulated by several cellular macromolecules, including tubulins, and microtubule-associated proteins and motor proteins, such as kinesins and dynesins. An additional essential regulator is the maturation/M-phase promoting factor comprising Cdk1/cyclin B. Cdk1/cyclin B itself is regulated by a complex group of positive and negative regulating kinases. In mammalian cells, these include wee1, myt1, cyclin-activating kinase, Chk1, and cds1 kinases. In addition, Cdc25 phosphatases, which are also regulated by other kinases and phosphatases, are responsible for the activation of Cdk1.
Inhibitors of tubulin polymerization or depolymerization are widely available but only a few disrupters of other regulators of G2/M progression have been identified. For example, we have identified several novel small-molecule inhibitors of Cdc25 that block G2/M progression but they also affect G1 transition (Tamura et al., 1999, 2000). Others have recently isolated a novel mitotic blocker that appears to act as a specific inhibitor of a mitotic kinesin (Mayer et al., 1999). Nonetheless, there continues to be a great need for pharmacologically distinct agents both to investigate the G2/M progression process and as new pharmacophores for drug design strategies.
One of the most attractive cellular systems for identifying novel agents that inhibit cell cycle progression at the G2/M phase boundary has been the mouse Cdk1 mutant cell line tsFT210. This cell line has two point mutations in Cdk1, an Ile52Val change in the PSTAIR region and Pro272Ser change in the C-terminal domain of the protein (Th'ng et al., 1990). These mutations produce a thermolabile Cdk1 that is inactive when cells are incubated at 39.4°C, leading to arrest in the mid-to-late G2 phase. These cells are not only useful for identifying inhibitors of Cdk1 but also for highlighting agents that disrupt any of the other factors essential for G2/M progression (Osada et al., 1997). In the current study, we have used tsFT210 cells to identify a novel small molecule that effectively blocks the G2/M transition without functioning through any well established mechanism. Although the cell cycle block has some biochemical features that are similar to those seen with nocodazole, our in vitro studies do not support a direct role for SR-7 with tubulin. Furthermore, cells resistant to paclitaxel are not markedly resistant to SR-7. Important also was our failure to see hyperphosphorylation of Bcl-2, which is a hallmark of tubulin interactive agents. Thus, SR-7 appears to exert its antimitotic effects by a novel mechanism that does not primarily involve microtubule perturbations; possible candidate targets for SR-7 action, such as microtubule-associated proteins or motor proteins, warrant future investigation.
We thank Angela Wang and Eileen Southwick for excellent technical support.
- Received July 12, 2000.
- Accepted October 10, 2000.
Send reprint requests to: Dr. John S. Lazo, Department of Pharmacology, Biomedical Science Tower E-1340, University of Pittsburgh, Pittsburgh, PA 15261. E-mail:
This work was supported in part by grants from the U.S. Public Health Service, National Institutes of Health CA 78039, the Department of Defense DAMD17-1-7229 and PC970414, and the Fiske Drug Discovery Fund.
- cyclin-dependent kinase
- simian virus 40
- 4-(benzyl-(2-[(2,5-diphenyl-oxazole-4-carbonyl)-amino]-ethyl)-carbamoyl)-2-decanoylamino butyric acid
- dimethyl sulfoxide
- polyacrylamide gel electrophoresis
- The American Society for Pharmacology and Experimental Therapeutics