Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

The identification of a group of nuclear enzymes called cyclin-dependent kinases (CDK) has profoundly advanced our understanding of cell cycle progression (Garrett and Fattaey, 1999; Johnson and Walker, 1999). CDK activity is primarily stimulated by the binding of G₁ cyclins in response to the effects of both mitogenic growth factors and the extracellular matrix; CDK activity is opposed by the so-called CDK inhibitors, such as p21^Waf1 and p27^Kip1 (Sherr and Roberts, 1999). p21^Waf1 was initially identified as a p53-dependent gene product in cells with wild-type p53 after exposure of the cells to DNA-damaging agents (Harper et al., 1993; Dulic et al., 1994), and it has since been shown that p21^Waf1 can also be induced in a p53-independent manner (Michieli et al., 1994; Macleod et al., 1995). In contrast, p27^Kip1 was first discovered in Mv1Lu mink lung epithelial cells arrested at G₁ phase of the cell cycle by contact inhibition or by treatment with transforming growth factor- (Toyoshima and Hunter, 1994; Polyak et al., 1994).

We and others have recently reported that, in contrast to p27^Kip1, which arrests cell cycle traversal solely at G₁ phase (Coats et al., 1996; Rivard et al., 1996), p21^Waf1 arrests the cell cycle at both G₁ phase and G₂ phase (Cayrol et al., 1998; Medema et al., 1998; Schmidt et al., 2000). The latter arrest at G₂ phase is caused by p21^Waf1-mediated inhibition of CDC2 activity, which is required for cells to enter M phase of the cell cycle. The differential modulation of cell cycle traversal by p21^Waf1 and p27^Kip1 was correlated with their effects on the cytotoxicity of paclitaxel, an M phase-dependent antineoplastic drug used to treat a variety of human cancers (Rowinsky, 1997; Aisner and Cortes-Funes, 1997). When ectopically expressed, both p21^Waf1 and p27^Kip1 conferred resistance to paclitaxel-mediated apoptosis on the cells. However, induction of p21^Waf1 after exposure to paclitaxel produced much greater resistance to the drug than did expression of p27^Kip1 (Schmidt et al., 2000). This result was attributed to the additional G₂ phase block by p21^Waf1, which prevented cells from entering M phase of the cell cycle and becoming committed to apoptosis after paclitaxel treatment (Schmidt et al., 2000).

Given the remarkable chemoresistance toward paclitaxel in p21^Waf1- or p27^Kip1-arrested cells, we examined a panel of chemotherapeutic agents that were categorized for their cell cycle-dependent or -independent antineoplastic activities. Cell cycle dependence in this context refers to the fact that active cell cycling is required for an antineoplastic agent to exert a cytotoxic effect (Myers and Chabner, 1990; Perry, 1992). DNA-intercalating, cross-linking, or alkylating agents (such as cisplatin and melphalan) do not necessarily require their targeted cells to be actively cycling; in contrast, antimetabolites and topoisomerase inhibitors (such as 5-fluorouracil and camptothecin) may be toxic to the cells only during the S phase of a cell cycle, and agents that interfere with the mitotic spindle (such as paclitaxel) may kill the cells primarily during M phase. We observed a substantially increased resistance to several known cell cycle-dependent antineoplastic agents in p21^Waf1- or p27^Kip1-induced cells, compared with uninduced cells, whereas only a desensitation of the arrested cells was observed upon treatment with several known cell cycle-independent agents. Additionally, we found that p21^Waf1 and p27^Kip1, when induced after drug exposure, conferred differential chemoresistance to the cytotoxicity of D-24851, a novel M phase-dependent compound (Bacher et al., 2001) in a manner very similar to what we reported of their effects on the cytotoxicity of paclitaxel, a well-known antineoplastic agent with M phase-dependent action (Schmidt et al., 2000). Our results suggest that this p21^Waf1- and p27^Kip1-inducible expression system might be a useful tool for determining not only the cell cycle dependence but also the M phase dependence of candidate compounds after the antineoplastic activity of the candidate compounds has been established during a primary high-throughput screening assay.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Stable RKO p21^Waf1- and p27^Kip1-Inducible Expression Clones. The RKO human colon carcinoma cells containing an ecdysone-inducible expression vector of p21^Waf1 or p27^Kip1 (RKO-p21 or RKO-p27) were described previously (Schmidt et al., 2000). Briefly, human waf1 and kip1 cDNAs were amplified by polymerase chain reaction techniques using MCF10A human nonmalignant mammary epithelial cell cDNA as a template and subcloned into pIND (Invitrogen, Carlsbad, CA) via restriction sites included in the waf1- and kip1-specific oligonucleotide primers. The pINDp21^Waf1 or pINDp27^Kip1 vector was transfected with the Fugene-6 kit (Roche Diagnostic Corp., Indianapolis, IN) into an RKO cell clone that was previously transfected with the pIND-regulatory vector pVgRXR (Invitrogen); the latter contains a transgene carrying an ecdysone-responsive regulatory sequence. Stable pINDp21^Waf1-transfected or pINDp27^Kip1-transfected clones (RKO-p21 or RKO-p27) were kept in double-selection culture medium containing 200 µg/ml Zeocin (Invitrogen, Carlsbad, CA) and 500 µg/ml neomycin (G418). Expression of p21^Waf1 or p27^Kip1 was induced by exposure to 3 µM muristerone A for 24 h and examined by Western blot analysis with specific antibodies.

Antineoplastic Agents. The antineoplastic agents used in this study were as follows: cisplatin (Platinol-AQ) and etoposide (VePesid) were from Bristol Laboratories (Princeton, NJ); doxorubicin (Adriamycin) was from Gensia Sicor Pharmaceuticals (Irvine, CA); melphalan (Alkeran) was from GlaxoSmithKline (Research Triangle Park, NC); 5-fluorouracil (Adrucil) was from Pharmacia (Kalamazoo, MI); camptothecin was from Sigma Chemical Co. (St. Louis, MO); and D-24851 (N-[pyridin-4-yl]-[1-[4-chlorbenzyl]-indol-3-yl]-glyoxyl-amid) was from ASTA Medica AG (Frankfurt, Germany) (Bacher et al., 2001).

Western Blot Analysis. Western blot analysis was performed as described previously (Fan et al., 1995). Briefly, RKO-p21 or RKO-p27 cells were lysed in a buffer containing 50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.5% Nonidet P-40, 50 mM NaF, 1 mM Na₃VO₄, 1 mM phenylmethylsulfonyl fluoride, 25 µg/ml leupeptin, and 25 µg/ml aprotinin, and the lysates were sonicated at 4°C. Equal amounts of lysate were separated on SDS-polyacrylamide gels and subsequently blotted onto nitrocellulose membranes for incubation with antibodies against p21^Waf1 (Neo Markers Biotechnology Inc., Union City, CA), p27^Kip1, Rb, or the Rb-related p130 (Santa Cruz Biotechnology Inc. Santa Cruz, CA). Specific signals were visualized using the enhanced chemiluminescence detection kit (ECL; Amersham Pharmacia Biotech Inc., Piscataway, NJ).

CDK Assays. The CDK assay with GST-Rb (Santa Cruz Biotechnology) as a substrate was performed as reported previously (Fan et al., 1995; Wu et al., 1996). Briefly, cells were lysed as described above. CDK2 was immunoprecipitated from sonicated lysates with corresponding antibodies (Santa Cruz Biotechnology) and subjected to an in vitro kinase reaction in the presence of the CDK substrate GST-Rb and [-³²P]ATP, followed by separation with SDS-polyacrylamide gel electrophoresis and autoradiography.

Flow Cytometric Analysis. After completion of the desired treatment, RKO-p21 or RKO-p27 cells were harvested by trypsinization, and an aliquot of 1 × 10⁶ cells was washed once with cold phosphate-buffered saline and then fixed with cold 70% ethanol. The DNA was stained with a solution containing 25 µg/ml propidium iodide and 10 µg/ml RNase A (Sigma Chemical Co) in phosphate-buffered saline for 6 h. Cell cycle distribution was analyzed with a FACScan flow cytometer (BD Biosciences, San Jose, CA) at an excitation wavelength of 488 nm (Fan et al., 1995).

Cytotoxicity Analysis (MTT Assay). Cell viability was measured by a 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) assay (Schmidt et al., 2000). The cytotoxic effects of selected drugs on RKO-p21 or RKO-p27 with and without induction of p21^Waf1 or p27^Kip1 were determined after drug treatment, and the results were expressed as a percentage of relative cell numbers of control cells that were not exposed to the drugs. Briefly, after desired treatment, cells were incubated with 1 mg/ml MTT (Sigma) in 0.5 ml of culture medium for 3 h in a 37°C CO₂ incubator, followed by cell lysis with 0.5 ml of lysis buffer containing 20% SDS in dimethyl formamide/H₂O, pH 4.7, at 37°C for more than 6 h. Optical absorbance of the cell lysate was determined at a wavelength of 595 nm.

Determination of Mitotic Index. Upon completion of the desired treatment, cells were harvested by trypsinization. An aliquot (5 × 10³ cells) was spun onto glass slides using a Cytospin and subsequently fixed and stained with the Hema-3 kit (Biochemical Sciences, Inc., Swedesboro, NJ). Two hundred cells were counted each time under a microscope from different areas for a total of five times. Cells in mitotic phase were scored individually, and the numbers were expressed as the percentage of the total cell number counted. Representative areas for each treatment condition were photographed (Schmidt et al., 2000).

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

The p21^Waf1- or p27^Kip1-Inducible Expression RKO Cell Clones. RKO cells are human colon cancer cells that contain wild-type p53 and Rb and express very low levels of endogenous p21^Waf1 or p27^Kip1 (Boyd et al., 1988; Kessis et al., 1993; van Bree et al., 1999; Schmidt et al., 2000). Figure 1A presents Western blot analysis data showing the expression of p21^Waf1 or p27^Kip1 in representative RKO-p21 and RKO-p27 transfectant clones upon gene induction. Expression of p21^Waf1 or p27^Kip1 was detectable approximately 2 h after the gene expression inducer muristerone A was added to the cell culture medium. The expression level reached a plateau approximately 24 h after induction, and expression persisted for at least 72 h without further supplementation with muristerone A (Schmidt et al., 2000). Induction of p21^Waf1 or p27^Kip1 strongly inhibited CDK2 activity (Fig. 1B). This inhibition of CDK2 activity was accompanied by dephosphorylation of Rb and the Rb-related p130 protein (Fig. 1C). Flow cytometric analysis further indicated that induction of p27^Kip1 completely arrested the cells at G₁ phase of the cell cycle within 24 h, whereas induction of p21^Waf1 arrested the cells at both G1 and G₂, although most cells were arrested in G₁ phase. Cell proliferation was completely inhibited upon the expression of p21^Waf1 or p27^Kip1. No signs of cytotoxic effects were observed up to 5 days after the induced expression of p21^Waf1 or p27^Kip1 (data not shown).

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Inducible expression of p21Waf1 and p27Kip1 in RKO cells. A, induction of p21Waf1 and p27Kip1 in representative RKO-p21 and RKO-p27 clones, respectively. The cells were incubated for 24 h in culture medium in the absence () or presence (+) of 3 µM muristerone A. The samples were analyzed by Western blotting with specific antibodies against p21Waf1 or p27Kip1. B, inhibition of cyclin-dependent kinase activity by ectopically expressed p21Waf1 and p27Kip1. RKO-p21 and RKO-p27 cells were incubated as described in A. The samples were subjected to CDK2 kinase assay using GST-Rb as a substrate. C, dephosphorylation of Rb and the Rb-related p130 after inducible expression of p21Waf1 or p27Kip1 in RKO-p21 and RKO-p27 cells. The cells were incubated for 24 h in culture medium in the absence () or presence (+) of 3 µM muristerone A. The samples were analyzed by Western blotting with specific antibodies against Rb or p130. D, cell cycle distribution analysis of RKO cells upon induction of p21Waf1 or p27Kip1. RKO-p21 and RKO-p27 cells were kept for 24 h in culture medium in the absence or presence of 3 µM muristerone A. The cell cycle distribution of the cells was analyzed by flow cytometric analysis after propidium iodide staining of the DNA.

Effect of p21^Waf1- and p27^Kip1-Inducible Expression on the Cytotoxicity of Antineoplastic Agents. Fig. 2A shows the treatment schedule of RKO-p21 and RKO-p27 cells with three representative antineoplastic agents known to be cell cycle-dependent (5-fluorouracil, camptothecin, and etoposide). RKO-p21 and RKO-p27 cells showed a concentration-dependent response to the cytotoxic effects mediated by these three drugs (Fig. 2B). Continuous expression of either p21^Waf1 or p27^Kip1 conferred on these cells nearly complete resistance to cell death induced by 5-fluorouracil, camptothecin, or etoposide. The IC₅₀ values in p21^Waf1- or p27^Kip1-uninduced cells versus p21^Waf1- or p27^Kip1-induced cells shifted from approximately 40 µM to >1,000 µM for 5-fluorouracil, from 160 nM to >10,000 nM for camptothecin, and from 35 µM to >1,000 µM for etoposide. Induction of p21^Waf1 and p27^Kip1 did not produce any noticeable difference in conferring resistance on the RKO-p21 and RKO-p27 cells to cell cycle-dependent antineoplastic agents.

View larger version (43K): [in this window] [in a new window]   Fig. 2.   Differential responses of p21Waf1- and p27Kip1-arrested RKO cells to known cell cycle-dependent and -independent antineoplastic agents. A, schematic illustration of the treatment schedule. RKO-p21 or RKO-p27 cells were seeded in medium in the presence or absence of 3 µM muristerone A for 24 h. After being exposed to various concentrations of chemotherapeutics for 3 h, the cells were cultured for an additional 72 h with or without 3 µM muristerone A in the culture medium. An MTT assay was then performed as described under Experimental Procedures. B, treatment of RKO-p21 or RKO-p27 cells with the cell cycle-dependent antineoplastic agents 5-fluorouracil, camptothecin, and etoposide. Cells were treated as described in A, and the relative cell number after treatment was determined in comparison with untreated cells. , uninduced cells; , induced cells. Values are means ± S.D. for three independent experiments. Each experiment had triplicate wells for each treatment group. C, treatment of RKO-p21 or RKO-p27 with the cell cycle-independent antineoplastic agents cisplatin, doxorubicin, and melphalan. See B for details.

Cell Cycle-Dependent Antineoplastic Activity of D-24851. On the basis of our results with known antineoplastic agents, we next evaluated the cytotoxicity of N-[pyridin-4-yl]-[1-[4-chlorbenzyl]-indol-3-yl]-glyoxyl-amid (D-24851), a novel investigational synthetic compound. D-24851 is a successful candidate for antineoplastic activity, having been selected by a cell-based, high-throughput screening assay by ASTA Medica AG, Germany. The compound is a novel synthetic microtubule inhibitor, and we have shown that D-24851 has potent cytotoxic properties in vitro and in vivo with much less neurotoxicity than occurs with paclitaxel (Bacher et al., 2001).

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Cell cycle-dependent cytotoxicity of the investigational compound D-24851. A, schematic illustration of the treatment schedule, similar to that described in the legend to Fig. 2. B, RKO-p21 or RKO-p27 cells were seeded in medium in the presence or absence of 3 µM muristerone A for 24 h. After being exposed to various concentrations of D-24851 for 3 h, the cells were cultured for an additional 72 h with or without 3 µM muristerone A in culture medium. An MTT assay was performed to determine cytotoxicity. , uninduced cells; , induced cells. Values are means ± S.D. for three independent experiments. Each experiment had triplicate wells for each treatment group. C, Flow cytometric analysis of RKO cells treated with D-24851 with or without induction of p21Waf1 or p27Kip1. RKO-p21 cells or RKO-p27 cells were treated with 1 µM D-24851 for 3 h. Twenty-four hours later, the cells were harvested, stained with propidium iodide, and subjected to cell cycle distribution analysis.

M Phase-Dependent Antineoplastic Activity of D-24851. We took the advantage of the additional G₂ block produced by induction of p21^Waf1 in the RKO-p21 cells (but not by induction of p27^Kip1 in the RKO-p27 cells) to further confirm that the antineoplastic activity of D-24851 depends on arrest of the cells in M phase of the cell cycle. In contrast to the experiments shown in Fig. 3, in which cell cycle arrest was achieved by induction of either p21^Waf1 (a dual G₁ and G₂ arrest) or p27^Kip1 (a complete G₁ arrest) before the exposure of RKO-p21 or RKO-p27 to D-24851 and continued during the 72-h postdrug period, p21^Waf1 and p27^Kip1 were induced only during the 72-h postdrug period (Fig. 4A). In this experimental setting, the cell cycle distribution in both the RKO-p21 and RKO-27 cells was supposed to be identical to that in their parental RKO cells before drug exposure; thus, equal percentages of cells in different phases of the cell cycle would be exposed to D-24851 during the 3-h pulse treatment period. Compared with the results shown in Fig. 3B, induction of p21^Waf1 or p27^Kip1 after exposure to D-24851 produced a less strong resistance of the RKO-p21 and RKO-p-27 cells to D-24851-induced cytotoxicity (Fig. 4B), which is to be expected because in this setting, both the RKO-p21 and RKO-p27 cells were still actively progressing through the cell cycle during the drug exposure. However, this change in the sequence of p21^Waf1 or p27^Kip1 induction with reference to drug exposure produced differential effects between p21^Waf1 and p27^Kip1 on mediating cell resistance to D-24851-mediated cytotoxicity measured 72 h later. Notably, induction of p27^Kip1 resulted in much less chemoresistance to D-24851 than did induction of p21^Waf1 (a >30-fold difference). It took approximately 24 h after D-24851 treatment for the cell cycle distribution pattern to transit from a normal pattern to an M phase-dominant pattern (Fig. 3C). The result shown in Fig. 4B suggested that the additional effect of p21^Waf1 at G₂ phase that impeded the entrance of cells into M phase during the period after D24851 treatment may have contributed to this difference in conferring D-24851 resistance to cells. The data therefore indicate that entry into M phase might be critical for D-24851-induced cytotoxicity (M phase-dependent toxicity).

View larger version (48K): [in this window] [in a new window]   Fig. 4.   M phase-dependent cytotoxicity of D-24815 on RKO cells. A, schematic illustration of the new treatment schedule. B, in contrast to the procedure described in Fig. 3B, RKO-p21 or RKO-p27 cells were seeded in muristerone A-free medium for 24 h. After a 3-h exposure to various concentrations of D-24851, the cells were divided into two groups and cultured for an additional 72 h with or without 3 µM muristerone A in the medium. An MTT assay was performed to determine cytotoxicity. , uninduced cells; , induced cells. Values are means ± S.D. for three independent experiments. Each experiment had triplicate wells for each treatment group. C, cytochemical staining of the RKO-p21 and RKO-p27 cells. Cells were exposed to 1 µM D-24851 for 3 h without (middle) or with (right) subsequent induction of p21Waf1 or p27Kip1 for 24 h. The cells were then fixed and stained with methylene blue and eosin. Control cells (left) were left untreated. Individual representative areas were photographed. D, determination of the mitotic index. Two hundred cells each time were blindly counted from the experiments shown in C, with cells counted for a total of five times in different areas under a microscope. Cells in mitotic phase with condensed chromosomes were counted and are expressed as the percentage of total cells. The experiment was repeated for three times; similar results were obtained.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

Screening for potential antineoplastic compounds from chemical libraries is often hampered by the lack of a system that can distinguish a potential antineoplastic compound from nonspecific toxins, such as poisons of the respiratory chain. In this study, the CDK inhibitors p21^Waf1 and p27^Kip1, which were ectopically expressed in a human colon adenocarcinoma cell line and showed differential effects on cell cycle arrest (a complete G₁ phase arrest of the cell cycle by p27^Kip1 and a dual G₁ and G₂ phase arrest by p21^Waf1), were evaluated for their application in drug discovery studies by investigating their modulation of cell sensitivity to several antineoplastic agents known to be cell cycle-dependent and -independent and to D-24851, a novel investigational compound.

We first showed that the p21^Waf1 and p27^Kip1-inducible expression system could distinguish between cell cycle-dependent and -independent antineoplastic agents. Induction of p21^Waf1 in the RKO-p21 or p27^Kip1 in the RKO-p27 cells conferred marked resistance to cell cycle-dependent drugs but only moderately desensitized the cells to cell cycle-independent drugs. We performed similar studies with another cell line, A431 human epidermoid carcinoma cells. We found that inducible expression of either p21^Waf1 or p27^Kip1 caused markedly increased resistance of the cells to chemotherapeutic agents, but we did not find substantial differences in drug resistance between paclitaxel (cell cycle-dependent) and cisplatin (cell cycle-independent) in A431 cells, presumably because A431 cells contain a mutated p53, whereas RKO cells have wild-type p53 (Schmidt and Fan, 2001).

In Fig. 2C, the relative cell number pattern reflecting the cytotoxicity after exposure of the cells to lower concentrations of doxorubicin and melphalan seems similar to the pattern after exposure of the cells to the cell cycle-dependent compounds. This is probably because doxorubicin inhibits DNA topoisomerase II at lower concentrations and intercalcates with DNA at higher concentrations (Bodley et al., 1989). We speculate that this may also be the case with melphalan.

Our current results lead us to propose the use of this RKO cell-based system for screening natural or synthetic compounds that have cell cycle-dependent antineoplastic activities, after the antineoplastic activities of these candidate compounds have been established during a primary high-throughput screening assay.

We further demonstrated that the differential effects of p21^Waf1 and p27^Kip1 on the cell cycle presented an additional advantage for the identification of candidate compounds that are M phase-dependent. We showed recently that the G₂ block produced by p21^Waf1, but not by p27^Kip1, contributed to their unequal modulation of sensitivity to paclitaxel-mediated apoptosis (Schmidt et al., 2000). In the current study, we used this system to demonstrate that the cytotoxicity of the investigational compound D-24851 is M phase-dependent. Our results suggest that, by comparing the cytotoxicity of candidate compounds from chemical libraries in the RKO-p21 and RKO-p27 cells with and without postdrug induction of p21^Waf1 or p27^Kip1, we may be able to determine whether the antineoplastic activities of those candidate compounds are M phase-dependent. Thus, this cell-based system presented here may be used not only for screening for cell cycle-dependent antineoplastic compounds, but also for determining which candidate compounds are M phase-dependent.

The therapeutic index of a potential antineoplastic compound is in principal based on its preferential activity to rapidly dividing cells. Screening of chemical libraries with the described p21^Waf1- and p27^Kip1-inducible system will not only provide information on the mode of antineoplastic activity but also predict possible adverse effects of candidate compounds on normal cells. Our system compares the antineoplastic activity of candidate compounds on proliferating cells versus quiescent cells of the same type and genetic background, thus providing unbiased information on both the therapeutic index and the mode of action.

Cell cycle synchronization can also be achieved by several known chemical compounds such as aphidicolin or nocodazole; however, these compounds are all very toxic to target cells and are therefore unfit for drug-screening studies. Additionally, serum deprivation is frequently used to synchronize cells; however, as has been observed in many cancerous cells, serum deprivation was not successful in inducing cell cycle synchronization in the RKO cells. In fact, it has been suggested that the mechanism of serum deprivation-induced cell cycle synchronization is mediated primarily by induction of p27^Kip1 (Coats et al., 1996). Induction of p21^Waf1 in RKO-p21 cells or p27^Kip1 in RKO-p27 cells does not cause any notable cytotoxic effects in RKO cells and thus avoids any overlay of cytotoxic effects on a candidate compound; therefore, any observed cytotoxicity could be attributed primarily to the effect of the candidate compound. Other approaches, such as analysis of fluorescence-activated cell sorting, can provide information on the cell cycle distribution caused by candidate compounds but cannot indicate in which phase such compounds may be effective in killing cancer cells or inhibiting cancer cell proliferation.

Although inducible or constitutive expression of p21^Waf1 or p27^Kip1 in cancer cells has been described in the literature during the last several years (St Croix et al., 1996, 1998; Rivard et al., 1996; Niculescu et al., 1998; Ruan et al., 1998; Yamamoto et al., 1999), these previous studies have not demonstrated a complete cell cycle arrest by p21^waf1 or p27^kip1 and a differential cell cycle arrest by p21^waf1 or p27^kip1 as shown in our system (Fig. 1D). Our system can be used in a highly standardized setting, allowing for very reproducible results. This p21^Waf1- and p27^Kip1-inducible expression system may also be used to examine the requirement or correlation of an active cell cycle with reference to the cytotoxic effects of a variety of other known cytotoxic agents, such as cytokines, and other anticancer treatments, such as radiation therapy.