Materials and methods

Cell lines and cell culture

Colorectal cancer cell lines SW480 and DLD-1, normal colon fibroblast cell line CCD-18Co, and prostate cancer cell lines DU145 and LNCaP were obtained from the American Type Culture Collection (Rockville, MD). Colon cancer cell lines HCT116p53 (+ / +) and HCT116p53 (- / -) were gifts from Dr. Bert Vogelstein (Johns Hopkins Medical Oncology Center, Baltimore, MD). All cell lines were grown in RPMI 1640 medium, except SW480 and CCD-18Co, which were maintained in Dulbecco's modified Eagle's medium and in modified essential medium in Earle's balanced salt solution with nonessential amino acids, respectively. The growth medium was supplemented with 10% fetal bovine serum, antibiotics, and L-glutamine (Gibco -BRL, New York, NY).

Construction of recombinant adenoviral vector

Construction and production of the recombinant adenoviral vectors carrying PTEN (Ad-PTEN) or the luciferase gene (Ad-Luc) have been described elsewhere.³⁰

XTT assay

Inhibition of tumor cell growth by treatment with combinations of Ad-PTEN, Ad-Luc, or IR and caffeine was analyzed by quantitatively determining cell viability using an improved XTT assay (Roche Molecular Biochemicals, Indianapolis, IN).³¹ Cells were plated in 96-well microtiter plates at densities of 2 10³–6 10³ cells per well in 100 l of medium. At 24 h after the cells were plated, 50-l aliquots of medium containing varying concentrations of Ad-PTEN or Ad-Luc were added to each well, or cells were exposed to varying doses of IR after 30 minutes of exposure to 100-l aliquots of medium containing varying concentrations of caffeine. After 3 hours of infection with Ad-PTEN or Ad-Luc at various multiplicity of infection (MOI) units (i.e., viral particles [vp]/cell), 50-l aliquots of medium containing varying concentrations of caffeine were added into each well. Cells were then incubated at 37°C in a humidified atmosphere containing 5% CO₂. At 3 days after incubation, cell growth and viability were quantified by XTT assay. Briefly, the culture medium was removed, and 150 l of XTT reaction mixture was added into each well with fresh medium at a final concentration of 0.3 mg/ml/well. Cells were then incubated for 2 hours at 37°C. The absorbance was measured at a wavelength of 450 nm against a reference wavelength of 630 nm in a microplate reader (Model ELX808; Bio-Tek Instruments, Inc., Winooski, VT). Percentage cell viability was calculated in terms of the absorbancy in treated cells relative to the absorbancy in untreated control cells. Experiments were repeated at least three times for each treatment in each individual experiment.

Isobologram analysis

The combination effects were analyzed by a modified isobologram method.³² Briefly, three isoeffect curves (modes I, IIA, and IIB), which were derived from each growth-inhibition curve, were drawn; the total area enclosed by these three lines represented an "envelope of additivity." Actual IC₅₀ values were obtained and plotted on the envelope. If the experimentally observed IC₅₀ was plotted on the left side of the envelope, the combination was considered to show a supra-additive (synergistic) interaction. If it was plotted within the envelope, the combination was regarded as additive, and if it was plotted on the right of the envelope and within the dotted-line square, the combination was considered to be subadditive. If the observed IC₅₀ was plotted outside the square, the combination was considered to be protective.

Gene transduction

Preliminary experiments using an adenoviral vector carrying the green fluorescent protein (Ad-GFP) showed that, at an MOI of 5000 vp/cell, the adenovirus can infect 99.3% of HCT116 p53 (+ / +) and 81.9% of CCD-18Co cells and more than 80% of other cells (data not shown) by 24 hours after infection. On the basis of these results, we used Ad-PTEN or Ad-Luc at an MOI of 5000 vp/cell in all subsequent experiments.

Cell-cycle analysis

Cells were seeded in 10-cm culture dishes (5 10⁵–10 10⁵ cells per dish) and treated with 2 mM caffeine 3 hours after infection with Ad-PTEN or Ad-Luc, or treated with 20 M of LY294002 (PI3K inhibitor) or 10 M U0126 (MEK1/2 inhibitor) (Cell Signaling Technology, Beverly, MA). To produce DNA damage, cells were exposed to 2.0 Gy IR 30 minutes after treatment with caffeine. At specified times after treatment, cells were harvested by trypsinization, washed once with ice-cold phosphate-buffered saline (PBS), fixed with 70% ethanol, and stored at -20°C. Cells were then washed twice with ice-cold PBS and treated with RNase (30 minutes at 37°C, 500 U/ml) (Sigma Chemicals, St Louis, MO), and DNA was stained with 50 g/ml propidium iodide (Boehringer-Mannheim, Indianapolis, IN). DNA contents and cell-cycle phases were analyzed on a fluorescence-activated cell sorter (FACScan, EPICS XL-MCL; Beckman Coulter, Fullerton, CA).

Apoptotic staining

Cells were seeded in six-well tissue culture dishes at a density of 1 10⁵ cells per well and treated with 2 mM caffeine or PBS as a mock control 3 hours after infection of Ad-PTEN. At 72 hours after infection, cells were analyzed for apoptosis using Hoechst 33258 staining (Sigma Chemicals). Apoptotic cells were identified via apoptotic body and/or chromosome condensation.

Western blot analysis

Cells were plated at densities of 5 10⁵–10 10⁵ in 10-cm dishes overnight. They were treated with 20 M LY294002 or 10 M U0126 alone, with 2 mM caffeine 3 hours after infection with Ad-PTEN or Ad-Luc, or exposed to 2.0 Gy IR at 30 minutes after treatment with caffeine. Cells were incubated for the indicated times at 37°C and then collected, and whole-cell lysates prepared. The cell lysates were analyzed for the expression of various proteins by Western blot analysis. The blots were reprobed using antibodies against -actin (Sigma Chemicals) to ensure equal loading and transfer of proteins. The following primary antibodies were purchased and used at 1:500 or 1:1000 dilution: PTEN, Cdc2, p53 (Bp53-12), and p27^Kip1, (Santa Cruz Biotechnology, Santa Cruz, CA); Cdc25C, phospho-Akt, and phospho-p44/42MAPK (Cell Signaling Technology); cyclin B1 (Neo Markers, Fremont, CA); and p21^WAF1 (Oncogene Research Products, Boston, MA).

Results

Combined treatment with Ad-PTEN and caffeine produces a synergistic effect in colorectal cancer cells

To observe the effect of the combination of Ad-PTEN and caffeine, four colorectal cancer cell lines (HCT116 p53 (+ / +), HCT116 p53 (- / -), SW480, DLD-1) and one normal colon cell line (CCD-18Co) were analyzed 72 hours after treatment (Fig 1a, b). All of the experiments were repeated at least three times for each cell line. At the IC₅₀ concentration for each cell line, combination of Ad-PTEN and caffeine produced a complete synergistic effect in HCT116 p53 (+ / +) and SW480 cells but a boundary effect between addition and synergy in HCT116 p53 (- / -) and DU145 cells. The observed data points were distributed either within the additive envelope or scattered around the boundary between the additive and the synergistic areas (Fig 1a). However, a protective effect was observed in CCD-18Co cells under these treatment conditions (Fig 1b). An additive effect was also observed in LNCaP cells treated with Ad-PTEN and caffeine (data not shown). To further confirm that the observed synergistic effect was specific, we tested the effects of a combination of Ad-Luc and caffeine in HCT116 p53 (+ / +) cells (Fig 1c). Unlike Ad-PTEN plus caffeine, Ad-Luc plus caffeine produced a boundary effect between additive and protective effects. These results indicate that treatment with a combination of Ad-PTEN and caffeine specifically induces a synergistic effect in tumor cells but not in normal cells (Table 1).

Figure 1.

Isobologram analysis of the effects of treatment with two-agent combinations on colorectal cell lines. Cells seeded in 96-well plates were treated with various concentrations of Ad-PTEN, caffeine, Ad-PTEN, or Ad-Luc and caffeine. At 72 hours after treatment, isobolograms at IC₅₀ levels were generated in colorectal cancer cell lines (HCT116p53 (+ / +), HCT116p53 (- / -), SW480, DLD-1) (a) and (c), and normal colorectal fibroblast cells (CCD-18Co), (b), treated with combinations of Ad-PTEN and caffeine (a) and (b), or Ad-Luc and caffeine (c).

Full figure and legend (129K)

Table 1 - Therapeutic effect produced in various cell lines when treated with a combination of Ad-PTEN and caffeine is independent of PTEN and p53 status.

Full table

Treatment with a combination of Ad-PTEN and caffeine induces apoptosis in colorectal cancer cells

To determine whether treatment with Ad-PTEN and caffeine induced apoptosis, HCT116 p53 (+ / +) and CCD-18Co cells were analyzed 72 hours after treatment for apoptotic changes by using fluorescence-activated cell sorting (FACS) and Hoechst 33258 staining. A significant increase (P<.01) in the number of cells in sub-G₀/G₁ phase, an indicator of apoptotic changes, was observed by FACS analysis in HCT116 p53 (+ / +) cells treated with Ad-PTEN alone (7.96%) or with a combination of Ad-PTEN and caffeine (13.05%) (Fig 2a). Cells treated with caffeine, Ad-Luc, or a combination of Ad-Luc and caffeine demonstrated no significantly higher number of apoptotic cells than PBS-treated control cells. In CCD-18Co cells, treatment with Ad-PTEN or Ad-PTEN and caffeine yielded no significantly higher number of apoptotic cells than control cells (Fig 2a). No increase in apoptotic cells was observed in CCD-18Co, even on day 5 after combined treatment, suggesting that normal cells are resistant to such treatments (data not shown). To confirm these results, cells were stained with Hoechst 33258. Tumor cells (HCT116 p53 [+/+]) but not normal cells (CCD-18Co) revealed condensed and fragmented nuclei, an indicator of cells undergoing apoptosis, when treated with Ad-PTEN, or with Ad-PTEN and caffeine (Fig 2b). Cells treated with PBS or caffeine showed no apoptotic morphology.

Figure 2.

Induction of apoptosis by treatment with combinations of Ad-PTEN and caffeine. (a) The numbers of cells at phase sub-G₀/G₁ (apoptotic cells) in colorectal cancer cells HCT116p53 (+ / +) and normal colorectal fibroblast cells CCD-18Co were analyzed by flow cytometry 72 hours after treatment with PBS, Ad-Luc, Ad-PTEN, caffeine, or combinations of caffeine with Ad-Luc or Ad-PTEN. Data represent the means of two experiments. Error bars denote standard error (SE). (b) Apoptotic analysis of HCT116p53 (+ / +) cancer cells and CCD-18Co normal cells by Hoechst 33258 staining was performed 72 hours after treatment with PBS, caffeine, or combinations of Ad-PTEN and caffeine. Treatment with combinations of Ad-PTEN and caffeine induced apoptosis in cancer cells but not in normal cells. Arrows indicate apoptotic cells (magnification 400).

Full figure and legend (232K)

Caffeine abrogates Ad-PTEN-induced G₂/M cell-cycle arrest

We next determined the effects of Ad-PTEN and caffeine treatment on cell-cycle phases in HCT116 p53 (+ / +) and CCD-18Co cells by using FACS analysis. Cell-cycle analysis demonstrated a significantly greater G₂/M population (P<.05) in 28.9% of HCT116 p53 (+ / +) cells and in 17.85% of CCD18-Co cells 72 hours after treatment with Ad-PTEN alone (Fig 3a, b) than in control cells treated with Ad-Luc, caffeine, PBS, or a combination of Ad-PTEN or Ad-Luc and caffeine. However, HCT116 p53 (+ / +) and CCD-18Co cells treated with LY294002 were arrested at G₁ phase, since LY294002 is an PI3K inhibitor of phosphatidyl inositol-3-kinase or PI3K (data not shown). Although one of the PTEN functions is PI3K inhibition, Ad-PTEN, unlike LY294002, did not induce G₁ arrest in HCT116 p53 (+ / +) and CCD-18Co cells carrying wt-PTEN.^9,11,33 These results, which suggest that Ad-PTEN induces G₂/M arrest in cells that express wt-PTEN, are in agreement with the results of our recent studies.^12,30 In contrast, treatment with a combination of Ad-PTEN or Ad-Luc and caffeine yielded dramatically lower numbers of HCT116 p53 (+ / +) and CCD18-Co cells that underwent G₂/M-phase arrest than treatment with Ad-PTEN or Ad-Luc alone. Treatment with caffeine alone yielded no significant decrease in the number of cells in the G₂/M phase when compared to control cells treated with PBS. However, the number of cells in G₂/M phase was significantly higher (P<.01) in cells treated with Ad-PTEN and caffeine than in cells treated with Ad-Luc and caffeine.

Figure 3.

Induction and abrogation of G₂/M cell-cycle arrest due to overexpression of PTEN and treatment with caffeine. (a) HCT116p53 (+ / +) colorectal cancer cells and (b) CCD-18Co normal colorectal fibroblast cells were treated with PBS, Ad-Luc, Ad-PTEN, caffeine, combinations of caffeine with Ad-Luc or Ad-PTEN, or 20 M LY294002. Cells were harvested 72 hours after treatment and cell-cycle analysis was performed by using flow cytometry. In total 20,000 events were captured for each treatment, and the data are shown as histograms. The cell-cycle phase is represented on the X-axis. Data were generated in duplicate; the average values are shown. Bars denote standard error (SE).

Full figure and legend (247K)

Signaling pathways regulated by PTEN overexpression and caffeine treatment in colorectal cancer cells and normal cells

To investigate the mechanism through which PTEN-induced G₂/M arrest is abrogated by caffeine, proteins related to the G₂/M and G₁ cell-cycle checkpoints were evaluated. In this analysis, HCT116 p53 (+ / +) and CCD18-Co cells were treated with 2 mM caffeine, PBS, Ad-PTEN or Ad-Luc alone, or a combination of Ad-PTEN or Ad-Luc and caffeine. Total cell lysates were prepared 48 hours after treatment and analyzed by Western blot analysis. Cells treated with Ad-PTEN, or Ad-PTEN and caffeine, demonstrated exogenous PTEN protein expression that resulted in inhibition of pAKT in both cell types (Fig 4a). However, the PTEN protein expression was significantly increased in HCT116 p53 (+ / +) cells, but not in CCD18-Co cells, when treated with the combination of Ad-PTEN and caffeine. Treatment with Ad-PTEN alone increased expression of p53, p21 and p27 and decreased expression of both phosphorylated and nonphosphorylated Cdc25C when compared to cells treated with PBS or Ad-Luc. However, no change in the expression of phospho-Chk1 and phospho-Chk2 was observed (data not shown). HCT116 p53 (+ / +) cells treated with caffeine alone demonstrated decreased expression of p53, p21, and cyclin B1 and increased levels of both phosphorylated and nonphosphorylated Cdc25C. In contrast, CCD18-Co cells treated with caffeine demonstrated decreased expression of p21, Cdc25C, and cyclin B1 with no change in p53. Cells treated with combination of Ad-PTEN and caffeine demonstrated higher expression of p53 and p27 than cells treated with PBS. Increased expression of p53 and p27 was due to PTEN since the expression of these proteins did not increase in cells treated with PBS alone. However, expression of p21, both phosphorylated and nonphosphorylatedCdc25C, cyclin B1, and Cdc2 decreased after treatment with a combination of Ad-PTEN and caffeine. Treatment with LY294002 decreased expression of cyclin B1, but did not change the expression level of Cdc25C, when compared with control cells (data not shown).

Figure 4.

Signaling pathways regulated by PTEN overexpression and caffeine. HCT116p53 (+ / +) colorectal cancer cells and CC18-Co normal cells were treated with PBS, Ad-Luc, Ad-PTEN, caffeine, combinations of caffeine with Ad-Luc or Ad-PTEN, or 10 M U0126. At 48 hours after treatment, cells were harvested and examined by Western blot analysis. (a) G₁ and G₂/M phase-associated proteins. (b) Phosphorylation status of p44/42MAPK. The corresponding -actin levels are shown as a loading control. (c) The numbers of cells at sub-G₀/G₁ (apoptotic cells) in HCT116p53 (+ / +) colorectal cancer cells were analyzed by flow cytometry 72 hours after treatment with PBS, Ad-Luc, Ad-PTEN, caffeine, or combinations of caffeine with Ad-Luc, Ad-PTEN, or Ad-PTEN and10 M U0126. Data represent the means of duplicate experiments. Error bars denote standard error (SE).

Full figure and legend (229K)

To further determine the underlying mechanism through which treatment with a combination of Ad-PTEN and caffeine produced a synergistic effect, we examined the phosphorylation status of p44/42MAPK and FAK, which PTEN inhibits mainly through dephosphorylation.^34,35 Although phospho-p44/42MAPK expression was not changed by Ad-PTEN treatment alone, it was significantly increased in HCT116 p53 (+ / +) but not in by treatment with the combination of Ad-PTEN and caffeine (Fig 4b). No significant change in phospho-p44/42MAPK was observed in CCD18-Co cells when treated with Ad-PTEN and caffeine. Therefore, to determine whether the synergistic effect of Ad-PTEN and caffeine in HCT116 p53 (+ / +) was due to increased phosphorylation of p44/42MAPK, we used FACS analysis to investigate phosphorylation status of p44/42MAPK and apoptotic ratio by treating cells with 10 M U0126, which is a MEK1/2 inhibitor. Treatment with Ad-PTEN, caffeine, and U0126 significantly increased the phosphorylation status of p44/42MAPK; however, expression of p44/42MAPK was less than in cells treated with Ad-PTEN and caffeine only (Fig 4b). In contrast, FACS analysis showed a significantly greater apoptotic ratio (39.35%) (P<.01) in cells treated with a combination of Ad-PTEN, caffeine, and U0126 than in cells treated with Ad-PTEN and caffeine only (13.05%; Fig 4c). No significant change in the expression of phospho-FAK was observed among the treatment groups 48 hours after treatment (data not shown), although phospho-FAK was decreased at later time points in HCT116 p53 (+ / +) cells treated with caffeine alone or with Ad-PTEN and caffeine (data not shown).

Comparison of the combined effects of Ad-PTEN and caffeine treatment and the combined effects of radiation and caffeine in colorectal cancer cells

To test whether caffeine can abrogate G₂/M phase arrest induced by other agents such as IR, we evaluated the combined effects of radiation treatment and caffeine and compared those effects to the effects of PTEN and caffeine in HCT116 p53 (+ / +) cells. Cells were treated with Ad-PTEN (5000 vp/cell) and caffeine or IR (2 Gy) and caffeine and subjected to cell-cycle analysis 48 hours after treatment. Cells treated with Ad-PTEN (51.4%) or IR (33.4%) alone were arrested at the G₂/M phase (Fig 5a, b). When these treatments were combined with caffeine, however, the numbers of cells in the G₂/M phase in the Ad-PTEN-treated (27%) and IR-treated (23.6%) groups were significantly decreased (P<.01), with concomitant increases in the number of cells in G₁ phase (Fig 5a, b). Note, although a shift in the number of cells from G₂/M phase to G₁ phase was observed when treated with Ad-PTEN or IR plus caffeine, there was no significant increase in the number of cells in the sub-G₁ phase. This is probably due to the early time point tested since an increase in the number of apoptotic cells (sub-G₁) was observed at later time points (>72 hours; data not shown). These results indicate that caffeine abrogates G₂/M arrest induced by various therapeutic agents.

Figure 5.

Abrogation of G₂/M cell-cycle arrest due to overexpression of PTEN or ionized radiation combined with caffeine. HCT116p53 (+ / +) colorectal cancer cells were (a) treated with Ad-PTEN or combination of Ad-PTEN and caffeine, or (b) exposed to 2 Gy IR or a combination of IR and caffeine. Cells were harvested 24 hours after treatment and cell-cycle analysis was performed by using flow cytometry. In total, 20,000 events were captured for each treatment, and the data are shown as histograms. The cell-cycle phase is represented on the X-axis. Data were generated in duplicate; the average values are shown. Bars denote standard error (SE).

Full figure and legend (44K)

To further investigate whether treatment with the combination of IR and caffeine produced a synergistic effect similar to that observed with Ad-PTEN and caffeine, HCT116 p53 (+ / +) cells were treated with PBS, Ad-PTEN, IR, or a combination of Ad-PTEN or IR and caffeine. At 72 hours after treatment, cells were analyzed by XTT assay for synergism using isobologram analysis and by FACS to determine the apoptotic ratio. Combination treatment with IR and caffeine produced an additive effect (Fig 6a). FACS analysis 72 hours after treatment revealed no significant difference in the apoptotic ratio in cells treated with a combination of Ad-PTEN and caffeine and in cells treated with a combination of IR and caffeine (Fig 6b). After 120 hours, however, the apoptotic ratio was significantly higher in cells treated with a combination of Ad-PTEN and caffeine (99%), than in cells treated with IR and caffeine (15.45%). Furthermore, we observed induction of phosphorylated p44/42MAPK expression when cells were subjected to IR treatment alone or to a combination of IR and caffeine (Fig 6c). Induction of p44/42MAPK was higher in cells treated with IR and caffeine and was similar to that seen when cells were treated with Ad-PTEN and caffeine.

Figure 6.

Effect of treatment with combinations of IR and caffeine. HCT116p53 (+ / +) colorectal cancer cells were treated with PBS, Ad-PTEN, a combination of Ad-PTEN and caffeine, 2 Gy IR, a combination of IR and caffeine, or 10 M U0126. (a) At 72 hours after treatment, isobolograms at IC₅₀ levels were generated for cells treated with the combination of IR and caffeine. (b) The numbers of cells at sub-G₀/G₁ (apoptotic cells) were analyzed by flow cytometry on days 3 and 5 after treatment. Data represent the means of duplicate experiments. Error bars denote standard error (SE). (c) At 48 hours after treatment, cells were harvested and phosphorylation status of p44/42MAPK examined by Western blot analysis. The corresponding -actin levels are shown as a loading control.

Full figure and legend (162K)

Discussion

In the present study, we investigated the effects of treatment with a combination of Ad-PTEN and caffeine in colorectal cancer cells that express wt-PTEN. Treatment with this combination successfully produced complete synergistic suppression and induced apoptosis to a significantly higher extent in colorectal cancer cells (HCT116 p53 (+ / +)) than in normal colorectal cells (CCD-18Co). To further compare the effects of this combination in cells that express wild-type p53 (wt-p53) and cells that express mutant p53 (mt-p53), we used HCT116 p53 (+ / +) and HCT116 p53 (- / -) cells. To compare the effects of this combination in wt-PTEN cells and in mutant-type PTEN cells, we used prostate cancer cell lines LNCaP, expressing mutated PTEN, and DU145, expressing wt-PTEN. We showed not only that overexpression of PTEN significantly suppressed growth and induced apoptosis in tumor cells compared to normal cells, but also that treatment with a combination of overexpression of PTEN and caffeine produces a synergistic inhibitory effect in cancer cells expressing wt-PTEN (Table 1). Although in the present study we have tested two prostate cell lines that differ in their PTEN status to support our findings on the ability of Ad-PTEN plus caffeine treatment to induce a synergistic effect in colorectal cancer cells that are wt-PTEN, a note of caution is that these cells may have additional differences apart for their differences in the PTEN status. Furthermore, this synergistic inhibitory effect was independent of endogenous p53 status. This contrasts with previous reports demonstrating that p53-null or -mutant cells were more sensitive to caffeine-induced radiosensitization than wt-p53 cells.^36,37,38 It is possible that the results of previous studies are supported by the observation that p53 plays an important role in DNA damage-induced cell-cycle checkpoints and contributes to the prevention of polyploidy formation.^{30,31,32,33,34,35,36,37,38,39,40,41} More recent studies have demonstrated that PTEN protects p53 from Mdm2, allows cells to respond to damage or mutation with an apoptotic response, and sensitizes cancer cells to chemotherapy.^41,43 In fact, Tsuchiya et al⁴³ showed that the synergistic antitumor effect of caffeine and cisplatin was enhanced by reintroduction of wild-type p53 into human osteosarcoma cells. Thus, the role of p53 remains controversial.

The mechanism by which the synergistic inhibitory effect was produced was determined by cell-cycle analysis. Treatment with Ad-PTEN alone resulted in G₂/M-phase arrest of both tumor cells and normal cells, while treatment with caffeine alone resulted in G₁ arrest. Induction of G₂/M arrest by Ad-PTEN in cells that express wt-PTEN is not surprising and is in agreement with results of our recent studies.^12,30 However, PTEN induction of G₂/M phase was abrogated by caffeine in both tumor and normal cells. Caffeine-induced G₁ was significantly more common in normal CCD-18Co cells than in HCT116 p53 (+ / +) cells, suggesting that caffeine had a stronger inhibitory effect on proliferation of CCD-18Co cells proliferation than on that of HCT116 p53 (+ / +) cells. Interestingly, the increased inhibitory effect of caffeine in CCD-18Co cells did not result in any increase in the number of apoptotic cells, even on day 5 after treatment (data not shown).

To further understand the mechanism by which the G₂/M checkpoint can be altered by Ad-PTEN, we analyzed the level of Cdc25C protein expression. In G₂/M phase, active Cdc25C phosphatase dephosphorylates Cdc2 on both threonine-14 and tyrosine-15 and then activates Cdc2/cyclin B1 complexes, leading to progression of the cells to M phase. In response to DNA damage or inhibition of DNA replication, however, inhibitory phosphorylation of cdc2 remains, and the cells are arrested at G₂ phase.^44,45,46,47 Our data demonstrate that Cdc25C expression was downregulated following Ad-PTEN treatment in both tumor cells and normal cells. ATM and ATR are proximal components of DNA damage-induced cell-cycle checkpoint pathways, and activated ATM and ATR phosphorylate Chk2 and Chk1, respectively. Therefore, DNA damage leads to activation of two related protein kinases, Chkl and Chk2, which then phosphorylate the Cdc25C phosphatase on serine-216.⁴² p53 is also required for G₂ arrest in response to DNA damage;⁴⁸ therefore, we assayed 14-3-3, GADD45, and p21^WAF1, which reside downstream of p53 and are associated with modulation of Cdc2 phosphorylation and G₂/M-phase arrest.⁴⁹ Ad-PTEN increased p53 and 14-3-3 expression, but not GADD45 (data not shown) or p21^WAF1, in HCT116 p53 (+ / +) colorectal cancer cells compared with controls. We next evaluated the effect of caffeine, an inhibitor of ATM and ATR that abrogates the G₂ checkpoint as described, on G₂ arrest induced by PTEN. Caffeine dramatically abolished G₂ arrest in HCT116p53 (+ / +) cells and modulated G₂ checkpoint-associated proteins (p53, p21^WAF1, p27 ^Kip1, both phosphorylated and nonphosphorylated Cdc25C, cyclin B1, and Cdc2), but we were not able to detect dephosphorylation of Chkl and Chk2 induced by caffeine. These results suggest that Ad-PTEN induced G2/M arrest in cancer cells carrying wt-PTEN occurs by inhibition of Cdc25C and support the findings of our recent studies.^12,33 Thus, the main targets of PTEN in cells carrying wt -PTEN may be different from those in cells carrying mutant- or deletion-type PTEN. We suggest that, on the basis of the status of endogenous PTEN, cells could be arrested at either G₁ or G₂ phase. Similarly, the downstream pathways may be different. However, the exact mechanism by which PTEN causes this difference in cell-cycle arrest is unclear and remains to be investigated.

Further analysis of the underlying mechanism of the observed synergistic inhibitory effect of Ad-PTEN and caffeine treatment in tumor cells demonstrated inhibition of the Akt pathway and upregulation of the p44/42MAPK pathway. However, inhibition of Akt was observed in both tumor and normal cells. In contrast, activation of p44/42MAPK was observed only in tumor cells but not in normal cells. However, activation of p44/42MAPK appeared to be associated with cytoprotective functions as previously reported,⁵⁰ since treatment with a combination of Ad-PTEN, caffeine, and MEK1 inhibitor UO126 resulted in lower expression of p44/42MAPK and a significantly higher apoptotic rate than treatment with Ad-PTEN and caffeine only. Alternatively, the increased PTEN protein expression observed in tumor cells, but not in normal cells, when treated with Ad-PTEN and caffeine could be responsible for the observed synergistic inhibitory effect. Whatever the underlying mechanism maybe, the present study demonstrates that the combination treatment of Ad-PTEN and caffeine exerts synergistic inhibitory effect in tumor cells but not in normal cells. To further test whether the effect and the function of PTEN as an growth inhibitor and apoptotic inducer are similar to other agents, HCT116 p53 (+ / +) cells were treated with Ad-PTEN or IR. Cell-cycle analysis showed that IR treatment induced G₂/M arrest and caffeine abrogated the G2/M arrest similar to that observed with Ad-PTEN and caffeine treatment. However, associated with the abrogation of the G2/M arrest, an increase in the number of apoptotic cells was not observed. This is due to the fact that the cell-cycle analysis was performed at early time points (48 hours). Analysis at later time points demonstrated increased number of apoptotic cells (data not shown). A major difference between PTEN and IR treatment, however, was in the expression of p44/42MAPK. IR induced p44/42MAPK expression but Ad-PTEN did not, while caffeine enhanced its expression. These results suggest that PTEN has an inhibitory effect on phospho-p44/42MAPK, but IR does not, and caffeine may enhance this kinase activity because of cytoprotective functions and/or mitogenic or differentiation-related stimuli.^51,52 Furthermore, combined treatment with IR and caffeine produced an additive effect, but not synergy, and induced significantly fewer apoptotic cells than treatment with Ad-PTEN and caffeine on day 5 after treatment. These results suggest that PTEN is a more effective therapeutic agent when combined with caffeine. In summary, the present study demonstrates that treatment with a combination of Ad-PTEN and caffeine produces a synergistic therapeutic effect and selectively induces apoptosis in colorectal cancer cells, especially in cancer cells that express wt-PTEN, but not in normal cells. Thus, combined treatment with Ad-PTEN and caffeine may offer a therapeutic strategy for patients with colorectal cancer.

References

1. Li J, Yen C, Liaw D, et al. PTEN, a putative protein tyrosine phosphatase gene mutated in human brain, breast, and prostate cancer. Science. 1997;275:1943–1947. | Article | PubMed | ISI | ChemPort |
2. Cairns P, Okami K, Halachmi S, et al. Frequent inactivation of PTEN/MMAC1 in primary prostate cancer. Cancer Res. 1997;57:4997–5000. | PubMed | ISI | ChemPort |
3. Tashiro H, Blazes MS, Wu R, et al. Mutations in PTEN are frequent in endometrial carcinoma but rare in other common gynecological malignancies. Cancer Res. 1997;57:3935–3940. | PubMed | ISI | ChemPort |
4. Wang SI, Puc J, Li J, et al. Somatic mutations of PTEN in glioblastoma multiforme. Cancer Res. 1997;57:4183–4186. | PubMed | ISI | ChemPort |
5. Yokomizo A, Tindall DJ, Drabkin H, et al. PTEN/MMAC1 mutations identified in small cell, but not in non-small cell lung cancers. Oncogene. 1998;17:475–479. | Article | PubMed | ChemPort |
6. Cheney W, Johnson D, Vaillancourt M-T, et al. Suppression of tumorigenicity of glioblastoma cells by adenovirus-mediated MMAC1/PTEN gene transfer. Cancer Res. 1998;58:2331–2334. | PubMed | ISI | ChemPort |
7. Davies MA, Lu Y, Sano T, et al. Adenoviral transgene expression of MMAC/PTEN in human glioma cells inhibits Akt activation and induces anoikis. Cancer Res. 1998;58:5285–5290. | PubMed | ISI | ChemPort |
8. Furnari FB, Huang HJ, Cavenee WK. The phosphoinositol phosphatase activity of PTEN mediates a serum-sensitive G₁ growth arrest in glioma cells. Cancer Res. 1998;58:5002–5008. | PubMed | ISI | ChemPort |
9. Li DM, Sun H. PTEN/MMAC1/TEP1 suppresses the tumorigenicity and induces G1 cell cycle arrest in human glioblastoma cells. Proc Natl Acad Sci USA. 1998;95:15406–15411. | Article | PubMed | ChemPort |
10. Minaguchi T, Mori T, Kanamori Y, et al. Growth suppression of human ovarian cancer cells by adenovirus-mediated transfer of the PTEN gene. Cancer Res. 1999;59:6063–6067. | PubMed | ISI | ChemPort |
11. Weng LP, Gimm O, Kum JB, et al. Transient ectopic expression of PTEN in thyroid cancer cell lines induces cell cycle arrest and cell type-dependent cell death. Hum Mol Genet. 2001;10:251–258. | PubMed |
12. Saito Y, Swanson X, Mhashilkar A, et al. Adenovirus-mediated transfer of the PTEN gene inhibits human colorectal cancer growth in vitro and in vivo. Gene Therapy. (in press).
13. Elledge SJ. Cell cycle checkpoints: preventing an identity crisis. Science. 1996;274:1664–1672. | Article | PubMed | ISI | ChemPort |
14. O'Connor PM. Mammalian G₁ and G₂ phase checkpoints. Cancer Surv. 1997;29:151–182. | PubMed |
15. Hartwell LH, Kastan MB. Cell cycle control and cancer (review). Science. 1994;266:1821–1828. | Article | PubMed | ISI | ChemPort |
16. IARC. IARC monograph evaluative carcinogen risks humans. 1991.
17. Mirvish SS, Cardesa A, Wallcave L, Shubik P. Induction of mouse lung adenomas by amines or ureas plus nitrite and by N-nitroso compounds: effect of ascorbate, gallic acid, thiocyanate, and caffeine. Nat Cancer Inst. 1975;55:633–636.
18. Nomura T. Diminution of tumorigenesis initiated by 4-nitroquinoline-1-oxide by post-treatment with caffeine in mice. Nature. 1976;260:547–549. | Article | PubMed |
19. Nomura T. Comparative inhibiting effects of methylxanthines on urethane-induced tumors, malformations, and presumed somatic mutations in mice. Cancer Res. 1983;43:1342–1346. | PubMed |
20. Theiss JC, Shimkin MB. Inhibiting effect of caffeine on spontaneous and urethane-induced lung tumors in strain A mice. Cancer Res. 1978;38:1757–1761. | PubMed |
21. Welsch CW. Caffeine and the development of the normal and neoplastic mammary gland. Proc Soc Exp Biol Med. 1994;207:1–12. | PubMed |
22. Nishikawa A, Furukawa F, Imazawa T, Ikezaki S, Hasegawa T, Takahashi M. Effects of caffeine on glandular stomach carcinogenesis induced in rats by N-methyl-N'-nitro-N-nitrosoguanidine and sodium chloride. Food Chem Toxicol. 1995;33:21–26. | PubMed |
23. Nishikawa A, Furukawa F, Imazawa T, Yoshimura H, Mitsumori K, Takahashi M. Effects of caffeine, nicotine, ethanol and sodium selenite on pancreatic carcinogenesis in hamsters after initiation with N-nitrosobis(2-oxopropyl)amine. Carcinogenesis. 1992;13:1379–1382. | PubMed |
24. Huang MT, Xie JG, Wang ZY, et al. Effects of tea, decaffeinated tea, and caffeine on UVB light-induced complete carcinogenesis in SKH-1 mice: demonstration of caffeine as a biologically important constituent of tea. Cancer Res. 1997;57:2623–2629. | PubMed | ISI | ChemPort |
25. Zajdela F, Latarjet R. Inhibition of skin carcinogenesis in vivo by caffeine and other agents. Natl Cancer Inst Monogr. 1978;133–140.
26. Lau CC, Pardee AB. Mechanism by which caffeine potentiates lethality of nitrogen mustard. Proc Natl Acad Sci USA. 1982;79:2942–2946. | PubMed | ChemPort |
27. Matsuoka S, Huang M, Elledge SJ. Linkage of ATM to cell cycle regulation by the Chk2 protein kinase. Science. 1998;282:1893–1897. | Article | PubMed | ISI | ChemPort |
28. Sarkaria JN, Busby EC, Tibbetts RS, et al. Inhibition of ATM and ATR kinase activities by the radiosensitizing agent, caffeine. Cancer Res. 1999;59:4375–4382. | PubMed | ISI | ChemPort |
29. Schlegel R, Pardee AB. Caffeine-induced uncoupling of mitosis from the completion of DNA replication in mammalian cells. Science. 1986;232:1264–1266. | Article | PubMed | ISI | ChemPort |
30. Stewart AL, Mhashilkar AM, Yang XH, et al. PI3K blockade by Ad-PTEN inhibits invasion and induces apoptosis in radial growth phase and metastatic melanoma cells. Mol Med. 2002;8:451–461. | PubMed | ISI | ChemPort |
31. Roehm NW, Rodgers GH, Hatfield SM, Glasebrook AL. An improved colorimetric assay for cell proliferation and viability utilizing the tetrazolium salt XTT. J Immunol Methods. 1991;142:257–265. | Article | PubMed | ISI | ChemPort |
32. Nishizaki M, Meyn RE, Levy LB, et al. Synergistic inhibition of human lung cancer cell growth by adenovirus-mediated wild-type p53 gene transfer in combination with docetaxel and radiation therapeutics in vitro and in vivo. Clin Cancer Res. 2001;7:2887–2897. | PubMed | ISI | ChemPort |
33. Taylor V, Wong M, Brandts C, et al. 5' Phospholipid phosphatase SHIP-2 causes protein kinase B inactivation and cell cycle arrest in glioblastoma cells. Mol Cell Biol. 2000;20:6860–6871. | Article | PubMed | ISI | ChemPort |
34. Gu J, Tamura M, Yamada KM. Tumor suppressor PTEN inhibits integrin- and growth factor-mediated mitogen-activated protein (MAP) kinase signaling pathways. J Cell Biol. 1998;143:1375–1383. | Article | PubMed | ISI | ChemPort |
35. Tamura M, Gu J, Matsumoto K, Aota S, Parsons R, Yamada KM. Inhibition of cell migration, spreading, and focal adhesions by tumor suppressor PTEN. Science. 1998;280:1614–1617. | Article | PubMed | ISI | ChemPort |
36. DeFrank JS, Tang W, Powell SN. p53-null cells are more sensitive to ultraviolet light only in the presence of caffeine. Cancer Res. 1996;56:5365–5368. | PubMed |
37. Powell SN, DeFrank JS, Connell P, et al. Differential sensitivity of p53(-) and p53(+) cells to caffeine-induced radiosensitization and override of G₂ delay. Cancer Res. 1995;55:1643–1648. | PubMed | ISI | ChemPort |
38. Yao SL, Akhtar AJ, McKenna KA, et al. Selective radiosensitization of p53-deficient cells by caffeine-mediated activation of p34cdc2 kinase. Nat Med. 1996;2:1140–1143. | Article | PubMed | ISI | ChemPort |
39. Cross SM, Sanchez CA, Morgan CA, et al. A p53-dependent mouse spindle checkpoint. Science. 1995;267:1353–1356. | Article | PubMed | ISI | ChemPort |
40. Kuerbitz SJ, Plunkett BS, Walsh WV, Kastan MB. Wildtype p53 is a cell cycle checkpoint determinant following irradiation. Proc Natl Acad Sci USA. 1992;89:7491–7495. | Article | PubMed | ChemPort |
41. Tsuiki H, Nitta M, Tada M, Inagaki M, Ushio Y, Saya H. Mechanism of hyperploid cell formation induced by microtubule-inhibiting drug in glioma cell lines. Oncogene. 2001;20:420–429. | Article | PubMed | ChemPort |
42. Mayo LD, Dixon JE, Durden DL, Tonks NK, Donner DB. PTEN protects p53 from Mdm2 and sensitizes cancer cells to chemotherapy. J Biol Chem. 2002;277:5484–5489. | Article | PubMed | ISI | ChemPort |
43. Tsuchiya H, Mori Y, Ueda Y, Okada G, Tomita K. Sensitization and caffeine potentiation of cisplatin cytotoxicity resulting from introduction of wild-type p53 gene in human osteosarcoma. Anticancer Res. 2000;20:235–242. | PubMed | ChemPort |
44. Dunphy WG, Kumagai A. The cdc25 protein contains an intrinsic phosphatase activity. Cell. 1991;67:189–196. | Article | PubMed | ISI | ChemPort |
45. Kastan MB, Onyekwere O, Sidransky D, Vogelstein B, Craig R. Participation of p53 protein in the cellular response to DNA damage. Cancer Res. 1991;51:6304–6311. | PubMed | ISI | ChemPort |
46. Peng CY, Graves PR, Thoma RS, Wu Z, Shaw AS, Piwnica-Worms H. Mitotic and G₂ checkpoint control: regulation of 14-3-3 protein binding by phosphorylation of Cdc25C on serine-216. Science. 1997;277:1501–1505. | Article | PubMed | ISI | ChemPort |
47. Sebastian B, Kakizuka A, Hunter T. Cdc25M2 activation of cyclin-dependent kinases by dephosphorylation of threonine-14 and tyrosine-15. Proc Natl Acad Sci USA. 1993;90:3521–3524. | Article | PubMed | ChemPort |
48. Thompson DA, Belinsky G, Chang TH, Jones DL, Schlegel R, Munger K. The human papillomavirus-16 E6 oncoprotein decreases the vigilance of mitotic checkpoints. Oncogene. 1997;15:3025–3035. | Article | PubMed | ISI | ChemPort |
49. Taylor WR, Stark GR. Regulation of the G₂/M transition by p53. Oncogene. 2001;20:1803–1815. | Article | PubMed | ISI | ChemPort |
50. Tsuneoka M, Mekada E. Ras/MEK signaling suppresses Myc-dependent apoptosis in cells transformed by c-myc and activated ras. Oncogene. 2000;19:115–123. | Article | PubMed | ISI | ChemPort |
51. Cross TG, Scheel-Toellner D, Henriquez NV, Deacon E, Salmon M, Lord JM. Serine/threonine protein kinases and apoptosis. Exp Cell Res. 2000;256:34–41. | Article | PubMed | ISI | ChemPort |
52. Seger R, Krebs EG. The MAPK signaling cascade. FASEB J. 1995;9:726–735. | PubMed | ISI | ChemPort |

Acknowledgements

We thank Kathryn Hale for editorial assistance and Alma Vega for help in the preparation of this manuscript. This study was partially supported by the Texas Higher Education Coordinating Board ATP/ARP Grant 003657-0078-2001 (RR), by Public Health Service Grant PO1-CA 78778-01A1 (JAR), by a Career Development award from The University of Texas SPORE in Lung Cancer P50CA70907-5 (RR), by M.D. Anderson Cancer Center Institutional Research Grant (RR), by M.D. Anderson Cancer Center Support Grant CA16672, by the W.M. Keck Foundation Fund for Human Cancer Gene Prevention and Therapy (RR), by a BESCT Lung Cancer Program Grant (DAMD17-01-1-0689), by a TARGET Lung Cancer Grant (DAMD17-02-1-0706), and by a sponsored research agreement with Introgen Therapeutics, Inc.