• Open Access

Induction of multinucleated cells and apoptosis in the PC-3 prostate cancer cell line by low concentrations of polyethylene glycol 1000

Authors

  • Katsuhiro Fukuta,

    1. Biochemistry Division, National Cancer Center Research Institute, 5-1-1, Tsukiji, Chuo-ku, Tokyo 104-0045;
    2. Department of Nephro-Urology, Nagoya City University Graduate School of Medical Sciences, 1, Kawasumi, Mizuho-cho, Mizuho-ku, Nagoya 467-8601;
    Search for more papers by this author
  • Kenjiro Kohri,

    1. Department of Nephro-Urology, Nagoya City University Graduate School of Medical Sciences, 1, Kawasumi, Mizuho-cho, Mizuho-ku, Nagoya 467-8601;
    Search for more papers by this author
  • Hirokazu Fukuda,

    1. Biochemistry Division, National Cancer Center Research Institute, 5-1-1, Tsukiji, Chuo-ku, Tokyo 104-0045;
    Search for more papers by this author
  • Masatoshi Watanabe,

    1. Laboratory for Medical Engineering, Division of Materials Science and Chemical Engineering, Graduate School of Engineering, Yokohama National University, 79-1 Tokiwadai, Hodogaya-ku, Yokohama 240-8501, Japan
    Search for more papers by this author
  • Takashi Sugimura,

    1. Biochemistry Division, National Cancer Center Research Institute, 5-1-1, Tsukiji, Chuo-ku, Tokyo 104-0045;
    Search for more papers by this author
  • Hitoshi Nakagama

    Corresponding author
    1. Biochemistry Division, National Cancer Center Research Institute, 5-1-1, Tsukiji, Chuo-ku, Tokyo 104-0045;
    Search for more papers by this author

To whom correspondence should be addressed. E-mail: hnakagam@gan2.res.ncc.go.jp

Abstract

Polyethylene glycol (PEG) has been reported to inhibit the development of colonic lesions in carcinogen-treated rats when administered orally. However, the precise mechanism for the chemopreventive activity of PEG remains largely elusive. Based on a characteristic feature of PEG as a ‘fusogen’, we investigated its potential as a chemotherapeutic agent through the induction of multinucleated cell formation and apoptosis induction in PC-3 prostate cancer cells. When PC-3 cells were treated with 0.5 and 1.0% PEG 1000, multinucleated cells were induced at a frequency of 8.4 and 13%, respectively, 36 h after PEG treatment under high cell density (1 × 106 cells in 100 µL PEG solution) in vitro. Although abnormality of cell cycle progression was not evident in PEG-treated PC-3 cells, multinucleated cells substantially disappeared at around 38 h due to apoptosis. In contrast, no apparent growth suppression was observed when PC-3 cells were exposed to up to 1.0% PEG at a much lower cell density, namely under ordinary culture conditions. Furthermore, injection of 0.5% PEG solution in vivo into PC-3 xenografts implanted in BALB/c-nu/nu male mice significantly suppressed tumor growth compared to phosphate-buffered saline injection. Multinucleated TdT-mediated dUTP-biotin nick end-labeling (TUNEL)-positive cells were observed inside the PEG-injected tumors. PEG was here demonstrated to have anticell proliferation and antitumor effects via induction of apoptosis, possibly by cell fusion. PEG injection therapy could therefore be adopted as an alternative chemotherapeutic strategy for localized prostate cancers, including those that become refractory to androgen-deprivation therapy. (Cancer Sci 2008; 99: 1055–1062)

Polyethylene glycol (PEG) has the chemical structure H-(O-CH2-CH2)n-OH, and is known to be a nonabsorbed, non-toxic, and non-fermentable polymer.(1,2) Because PEG is non-toxic, it is used widely as a base material in, for example, skin creams and laxatives.(3,4) When PEG is bound to a hydrophobic molecule, a non-ionic detergent surfactant is obtained and can be used as an emulsion agent in cosmetics.(5) PEG is also used to encapsulate and solubilize hydrophobic compounds. It has recently been utilized in drug delivery systems, and PEGylation of interferon a (peginterferon a) and granulocyte colony-stimulating factor (pegfilgrastim) are now used for clinical applications.(6,7) The stability and hydrosolubility of various drugs is thereby markedly improved even under in vivo conditions. In the research field of biotechnology, PEG is also used for cell fusion as a ‘fusogen’.(8,9)

In addition to these features, chemopreventive effects of PEG have been reported in rodent models of colon cancer.(10,11) When rats were treated with PEG 8000 after the administration of azoxymethane, a widely utilized colon carcinogen, a substantial decrease was observed in the number of aberrant crypt foci, putative precancerous lesions of the colon, and colon tumors.(12) Although the mechanistic interpretation for the chemopreventive effect of PEG is still controversial, several possibilities have been proposed. First, an increase in the gut content because of the high osmotic pressure in the digestive tract caused by the non-absorbed high molecular weight PEG could be a possibility, similar to dietary fiber.(10,13,14) Another possibility is that PEG may act directly on colon epithelial cells and exert some biological effects, such as induction of apoptosis, through osmotic pressure.(15)

In the present study, we examined whether PEG induces apoptosis in cancer cells using an androgen-independent human prostate cancer cell line, PC-3,(16) and could be used as an alternative therapeutic agent for human prostate cancers. Prostate cancer is a common cancer in men all over the world, and its incidence in Japan is currently increasing.(17,18) As for prostate cancer therapies, androgen-deprivation therapy (hormone therapy) now prevails all over the world, as it is non-invasive and relatively effective in many cases.(19–21) Radiation therapy, such as ‘brachytherapy’, is also used for localized prostate cancer.(22) However, these therapies have some drawbacks. For example, brachytherapy requires specific and well-guarded facilities for radiation. Hormone therapy is still quite costly, and androgen-independent prostate cancers arise frequently after continuous application of hormone therapy for a long period of time.(23) Establishment of safe and affordable therapies is therefore awaited.

Polyethylene glycol (PEG) has long been utilized to fuse different types of cells, as described above. Various cancer cells have been demonstrated to become fused by high concentrations (30–50%) of PEG.(24) We investigated whether the induction of apoptosis in cancer cells could result from such multinucleated cell formation as multinucleated cells induced by various cell-damaging agents, such as irradiation, doxorubicin, and docetaxel, have been shown to cause cell death by apoptosis.(25–27) We treated PC-3 cells under various conditions in vitro with PEG 1000 (refered to hereafter as ‘PEG’). Interestingly, a concentration of PEG as low as 0.5–1.0% efficiently induced multinucleated cells under conditions of high cell density, and induced apoptosis after 34–38 h in vitro. PEG treatment at a low cell density (‘ordinary culture conditions’) was not similarly effective. When 0.5% PEG in phosphate-buffered saline (PBS) was injected directly into PC-3 xenografts in nude mice, tumor growth was suppressed substantially compared to the PBS-injected control group. The potential molecular mechanisms underlying the suppressive effects of PEG on cell proliferation both in vitro and in vivo are discussed.

Materials and Methods

Chemicals.  PEG and dimethylsulfoxide were obtained from Merck (Darmstadt, Germany), propidium iodide (PI), and etoposide were from Sigma (St Louis, MO, USA) and RNase A was from Qiagen (Valencia, CA, USA). DNA ladders of 100 bp (New England Biolabs, Beverly, MA, USA) and 1 kb (Promega, Madison, WI, USA) were used as molecular weight markers.

Cell culture.  PC-3 cells were purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA) and cultured at 37°C in Dulbecco's modified Eagle's medium (DMEM; Invitrogen Life Technologies, Carlsbad, CA, USA) supplemented with 10% heat-inactivated fetal bovine serum (FBS; Invitrogen Life Technologies), penicillin (100 U/mL), and streptomycin (50 U/mL) (Invitrogen Life Technologies) under a humidified atmosphere with 5% CO2. DMEM (without FBS) was also used to make serial dilutions of PEG before addition into cell culture media.

Polyethylene glycol treatment protocol.  PC-3 cells (1 × 106) were suspended in 100 µL of 0.5 and 1.0% PEG in DMEM (without FBS), incubated at room temperature for 1, 5, 15, or 30 min, then diluted 1:10 with DMEM and incubated at room temperature for 10 min. The cells were collected by centrifugation, gently washed twice with PBS, suspended in 10 mL of culture medium with FBS, seeded into 100-mm culture dishes with cover glasses at the bottom, and propagated for 34, 36, and 38 h. Cover glasses were then removed, the adherent cells on the cover glasses were fixed with 100% ethanol for 15 min, stained with PI solution (50 µg PI and 100 µg RNase A in 1.0 mL PBS), and then counted using a microscope. We adopted the criterion that cells having two or more nuclei were multinucleated. More than 500 cells within a randomly selected area on cover glasses were counted, and the cell counting was repeated three times for each experimental group.

Cell-growth assay.  The effect of PEG on PC-3 cell growth was then evaluated under two different conditions. In the first condition, PC-3 cells (5 × 103) were plated in 100-mm culture dishes and incubated overnight (‘ordinary culture conditions’). Aliquots of PEG in DMEM (1.0 mL) were added directly to cell culture dishes to give a final concentration of 0.5 or 1.0%, and the cells were propagated on the plates. As a reference control, 1.0 mL DMEM alone was added to the culture plates. The cells were harvested at 24, 36, 48, and 72 h after the addition of PEG, stained with trypan blue, and viable cell numbers were counted at each time point. Alternatively, 5 × 103 cells were treated following the PEG treatment protocol with 0, 0.5, or 1.0% PEG, seeded and propagated as above. In this case, cells were exposed to PEG for 1, 5, 15, and 30 min. Cells were then seeded in culture plates, propagated and harvested at 24, 36, 48, and 72 h after PEG treatment, including the nonadherent cells.

Detection of nuclear condensation.  PC-3 cells were treated with 1.0% PEG for 30 min following the PEG treatment protocol, and seeded in 100-mm plates. Cells were collected after 34 h, fixed with 1.0% glutaraldehyde in PBS, incubated at 4°C overnight, collected by centrifugation, and resuspended in 50 µL PBS. After mixing thoroughly, 2 µL Hoechst 33258 (Dojindo, Kumamoto, Japan) was added to visualize nuclear condensation.

Poly(ADP-ribose) polymerase-1 cleavage.  PC-3 cells were exposed to 0, 0.5, 1.0, or 2.0 PEG for 30 min, as described above. Cells were collected after 34 h and suspended in lysis buffer (50 mM Tris-HCl [pH 7.4], 150 mM NaCl, 1.0% Triton X-100) containing a protease inhibitor cocktail (Complete Mini; Roche, Indianapolis, IN, USA). Whole-cell lysates were electrophoresed on 5–20% linear gradient Tris-HCl-ready gels (Bio-Rad, Hercules, CA, USA) and fractionated proteins were transferred to polyvinylidene fluoride membranes (Immobilon-P; Millipore, Billerica, MA, USA). Western blot analysis was then carried out, as described elsewhere,(28) with mouse monoclonal antibody (mAb) against poly(ADP-ribose) polymerase (PARP)-1 (C2-10, 1:1000; Oncogene Research Products, San Diego, CA, USA) and mouse mAb against glyceraldehyde 3-phosphatase dehydrogenase (1:5000; Chemicon International, Temecula, CA, USA) as primary antibodies. For the secondary antibodies, horseradish peroxidase-conjugated antibodies against mouse IgG (NA9310V, 1:5000; Amersham Biosciences, Piscataway, NJ, USA) were used. Immunoreactive bands on the blots were visualized with chemiluminescence substrates (Immobilon Western; Millipore, Billerica, MA, USA). As a positive control, lysates were used from cells treated with a topoisomerase II inhibitor etoposide, which is known to induce apoptosis. Etoposide was dissolved in dimethylsulfoxide (25 mg/mL) and added to culture media to give a final concentration of 25 µg/mL.(29,30) Quantification of the cleaved PARP-1 protein was carried out using MultiGauge software (Fujifilm, Tokyo, Japan). Western blot analysis was carried out at least in triplicate.

Detection of DNA fragmentation.  Following the PEG treatment protocol, PC-3 cells were exposed to 0, 0.5, 1.0, or 2.0% PEG for 30 min, seeded in 100-mm culture plates, and propagated for up to 38 h. Cells were collected, and DNA was extracted using a DNeasy Blood & Tissue Kit (Qiagen) following the manufacturer's instructions. Aliquots of extracted DNA samples were fractionated on a 1.5% SeaKem GTG agarose gel (Cambrex, Bio Science Rockland, Rockland, ME, USA). A DNA sample extracted from PC-3 cells treated with etoposide (25 µg/mL) for 48 h was used as a positive control.

Quantification of apoptosis by flow cytometry.  PC-3 cells were treated with 0.5 or 1.0% PEG for 1 min following the PEG treatment protocol, seeded in 100-mm culture dishes and propagated in DMEM media without PEG for 36 h. Cells were harvested, washed gently with PBS, collected by centrifugation, and then stained using a MEBCYTO Apoptosis Kit (MBL, Nagoya, Japan) following the manufacturer's instructions. Cells were doubly stained with Annexin-V and PI, and the fluorescence intensities were measured by flow cytometry (FCM) (FACScan; BD Bioscience, San Jose, CA, USA) using the CellQuest analysis program.(31) We also conducted FCM analyses as above to evaluate the populations of apoptotic cells specifically among the large cells. Briefly, fluorescence intensities of the PI signal in DMEM-treated PC-3 cells were extracted from FCM data sets, expressed using FL-2A (representing DNA content) and FL-2W (representing cell size), and gated by the R1 region (representing the large-cell population), the cut-off value for which was set arbitrarily. The same scale was applied to the 0.5% PEG-treated cells. Cells extracted by the R1 region were replotted by PI signals against Annexin-V.

Cell cycle analyses by FCM.  Following the PEG treatment protocol, PC-3 cells were exposed to 0, 0.5, 1.0, 2.0, or 10% PEG for 1 min, seeded in 100-mm culture plates, and propagated for 24, 34, 36, and 38 h. Cells were collected, fixed with 100% cold ethanol, and kept at –20°C overnight. After centrifugation, cells were resuspended in PI solution, incubated at room temperature for 15 min, and passed through nylon mesh (Becton, Dickinson and Company, Franklin Lakes, NJ, USA). The fluorescence intensities of PI were measured by FCM using a total of 20 000 PC-3 cells, and the number of cells in the sub-G1, G1, S, G2/M, and >G2/M (the population of cells with DNA contents beyond the G2/M peak) fractions was counted at 24–38 h after PEG treatment. We further examined the number of PC-3 cells in large-cell populations and aneuploid populations using the same FCM data sets above at 34 and 36 h after PEG treatment.

Effects of PEG on in vivo tumor growth.  The effect of PEG on PC-3 xenograft tumor growth was evaluated under two different conditions in vivo. With one of these, following the PEG treatment protocol, PC-3 cells (5 × 106 cells) were pretreated with DMEM or 0.5% PEG for 30 min, and then treated cells were implanted subcutaneously into the backs of BALB/c-nu/nu male mice (CLEA, Tokyo, Japan). Alternatively, PC-3 cells (5 × 106) were first implanted subcutaneously into the back of each mouse, and then at day 7 after implantation, 200-µL aliquots of PBS alone or 0.5% PEG solution in PBS were injected directly into the tumors using a 25-gauge injection needle (Terumo Corporation, Tokyo, Japan). Thereafter, tumors were measured twice a week throughout the experimental period of 19 days, and tumor volume was calculated using the following formula: (length [mm]) × (width [mm])2 × 0.52. In these experiments, a total of 16 nude mice was randomly separated into four groups (4 × 4) receiving: DMEM treatment; 0.5% PEG treatment; intratumor PBS injection; and intratumor 0.5% PEG injection. PC-3 cells were implanted at two sites for each mouse. Animal experimental protocols were approved by the Committee for Ethics in Animal Experimentation, and the experiments were conducted in accordance with the guidelines for Animal Experiments of the National Cancer Center (Tokyo, Japan).

Histopathological analysis and detection of apoptosis in PC-3 xenografts.  PC-3 xenografts were extirpated at 36–38 h after the last injection of PBS or 0.5% PEG for fixation in 10% neutralized formalin and embedding in paraffin blocks. Serial sections were prepared at 3.5-µm thickness, stained with hematoxylin and eosin (H&E), and subjected to histopathological analysis by a trained pathologist (M. W.). Apoptotic cells were detected using the TdT-mediated dUTP-biotin nick end-labeling (TUNEL) method,(32) which was carried out in situ using the DeadEnd Colorimetric TUNEL System (Promega) following the manufacturer's protocol.

Statistical analysis.  All statistical analyses were carried out with Wilcoxon signed-ranks test using KaleidaGraph software (Synergy Software, Reading, PA, USA). Differences were considered significant when the P-value was less than 0.05.

Results

Induction of multinucleated PC-3 cells by PEG treatment in vitro.  With the PEG treatment protocol, higher numbers of multinucleated cells were observed after PEG treatment, using 0.5–1.0% PEG (Fig. 1a). The incidence of multinucleated cells peaked at 36 h, being 8.4 and 13% with 0.5 and 1.0% PEG, respectively (Fig. 1b). The incidence only slightly increased with a longer exposure to PEG for 30 min (Fig. 1c).

Figure 1.

Polyethylene glycol (PEG) induces multinucleated cells. (a) Following the PEG treatment protocol, PC-3 cells were treated with 0.5% PEG for 30 min. Cells were stained with propidium iodide (PI) (red). The left panel indicates the morphological features of the PEG-treated cells under microscopic observation with a green filter. The right panel indicates the nuclear staining of cells with PI. Both panels are the same magnification. The arrows indicate typical multinucleated cells. Scale bars =  50 µm. (b,c) Incidences of multinucleated cells in a total of 500 cells treated with PEG for (b) 1 min and (c) 30 min are shown. Values are the mean of fold ± SD. ***Differences of cell number between PEG treatment (0.5 and 1.0%) and the Dulbecco's modified Eagle's medium (DMEM)-treated control (P < 0.001).

Effects of PEG on PC-3 cell growth in vitro.  In order to exclude possible toxic effects of PEG on PC-3 cells, we cultured PC-3 cells under two different culture conditions. When PC-3 cells were propagated under ordinary culture plates, and PEG was added into the media, no significant effect on cell growth was observed up to 72 h in the presence of 0.5 and 1.0% PEG, compared with the non-PEG treated control (Fig. 2a). However, significant differences were observed in cell growth between PEG-treated and DMEM-treated control cells when PC-3 cells were treated with 0.5% PEG (Fig. 2b) and 1.0% PEG (Fig. 2c) at a high cell density for 5–30 min, as in the PEG treatment protocol. A decrease in cell numbers became most prominent at around 36 h after PEG treatment (Fig. 2b,c).

Figure 2.

Polyethylene glycol (PEG) suppresses cell growth at a high cell density. (a) After PEG solution was added into culture media, viable PC-3 cells were counted. (b,c) Following the PEG treatment protocol, PC-3 cells were treated with (b) 0.5% or (c) 1.0% PEG for 1, 5, 15, and 30 min, and viable cells were counted. Values are the mean of fold ± SD. (b) Differences of cell number between 0.5% PEG (30 min) and the Dulbecco's modified Eagle's medium (DMEM)-treated control, and (c) those between 1.0% PEG (30 min) and DMEM-treated control were significant. *P < 0.05.

Biochemical analyses for apoptosis.  We observed nuclear condensation in multinucleated PC-3 cells on staining with Hoechst 33258 (Fig. 3a), suggesting induction of apoptosis.(33) Although DNA ladder formation was not observed in PEG-treated PC-3 cells, DNA samples extracted from the cells treated with PEG were substantially degraded compared to the DMEM-treated control (data not shown). Furthermore, PARP-1 cleavage, another characteristic feature of apoptosis,(34) was demonstrated in a PEG dose-dependent manner (Fig. 3b). The ratio of cleaved PARP-1 was increased by 14–52% in PEG-treated PC-3 cells when compared to the DMEM-treated control cells, although the increase was not so drastic (Fig. 3c).

Figure 3.

DNA condensation and poly(ADP-ribose) polymerase (PARP)-1 cleavage. (a) PC-3 cells were treated with 1.0% polyethylene glycol (PEG) for 30 min following the PEG treatment protocol. Cells were stained with Hoechst 33258 (blue), as detailed in Materials and Methods. The left panel indicates the typical morphological feature of multinucleated cell (arrows). The right panel indicates the nuclear condensation of multinucleated cells. Both panels are the same magnification. Scale bars = 50 µm. (b) PARP-cleavage. Whole-cell lysates after treatment with Dulbecco's modified Eagle's medium (DMEM), PEG, or etoposide were prepared, and full-length (uncleaved) PARP-1 (116 kDa), cleaved PARP-1 (85 kDa), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (37 kDa) were immunoblotted. Lane 1, DMEM; lane 2, 0.5% PEG; lane 3, 1.0% PEG; lane 4, 2.0% PEG; lane 5, etoposide (positive control). (c) Quantification of the cleaved PARP-1 protein by western blot analysis. The density of both non-cleaved and cleaved PARP-1 and of the background levels were measured, and the ratios of cleaved PARP-1 against the total PARP-1 were calculated as follows: (density of cleaved PARP-1) – (background level)/(density of total PARP-1) –(background level).

Quantification of apoptosis by FCM.  FCM analysis by double staining with Annexin-V (horizontal axis) and PI (vertical axis) demonstrated substantial induction of apoptosis, as depicted in Figure 4a. The incidence of cells in the early apoptotic phase (Annexin-V+, PI) was 16.3%, and that in late apoptotic phase (Annexin-V+, PI+) was 9.2% after treatment with 0.5% PEG (Fig. 4a right). An approximate 10-fold increase was observed in the percentage of Annexin-V-positive cells (25.5%) with PEG treatment compared to 2.8% for the DMEM-treated control (Fig. 4a left). Furthermore, multinucleated cells demonstrated positive staining for Annexin-V (arrows in Fig. 4b).

Figure 4.

Quantification of apoptosis using flow cytometry (FCM). (a) PC-3 cells were treated with Dulbecco's modified Eagle's medium (DMEM) alone (left) or 0.5% polyethylene glycol (PEG) (right) for 1 min following the PEG treatment protocol. After double-staining with Annexin-V and propidium iodide (PI), apoptotic cells were analyzed by FCM. (b) Multinucleated giant cells (arrow) visualized by staining with Annexin-V (green) and PI (red). Scale bar = 50 µm.

In order to clarify whether apoptosis was induced mainly in multinucleated cells, we extracted FCM data sets of only large cells (Fig. 5a), and then replotted the PI signals of each cell against the Annexin-V signals using the above data (Fig. 5b). In Figure 5a, the number of cells in the large-cell fraction (R1), the cut-off value for which was set arbitrarily, increased almost three-fold (8.52 vs 24.26% among 10 000 cells analyzed) after 0.5% PEG treatment. In Figure 5b, among the cells in the R1 fraction, 292 and 1571 cells in DMEM-treated control and 0.5% PEG-treated group respectively, showed positive for Annexin V. Namely, an approximate five-fold increase in apoptotic cells was observed within the large-cell fraction of PEG-treated cells compared to DMEM-treated cells. In particular, a more than 20-fold increase was observed in the number of cells in early apoptotic phase (25 vs 656 cells). Similar results were obtained with various cut-off values for R1 to include 3.0% and 5.5% of total populations in DMEM-treated cells (data not shown).

Figure 5.

Induction of apoptosis in large cells induced by polyethylene glycol (PEG) treatment. (a) From the data sets presented in Figure 4a, large cell populations were examined. (b) On the data sets presented in (a), the large cells gating the R1 region were replotted by propidium iodide (PI) signals against Annexin-V.

No evidence of mitotic catastrophe in PEG-treated cells.  By FCM analysis, no marked changes in cell cycle profile were apparent after PEG treatment (Fig. 6a). Although the number of cells in the sub-G1 fraction was slightly increased by PEG treatment, cells in the G1, S, G2/M, and > G2/M fractions were not changed significantly (Fig. 6b). We also noted that the number of cells in > G2/M fractions decreased substantially between 34 and 36 h after PEG treatment, as depicted in Supplemental Fig. 1.

Figure 6.

Cell cycle analysis. (a) Cell cycle profiles after polyethylene glycol (PEG) treatment at 34 h are demonstrated. (b) From the data sets presented in (a), the cell population of each fraction (sub-G1, G1, S, G2/M, and > G2/M fraction) in the cell cycle is shown.

We then extracted aneuploid-cell and large-cell populations from the > G2/M fraction using the FCM data set, as detailed in Supplemental Fig. 2A, and the corresponding number of cells in Supplemental Fig. 2A was counted between 34 and 36 h (Supplemental Fig. 2B). The number of aneuploid cells in the > G2/M fraction did not show a significant change between 34 and 36 h after 0.5 and 1.0% PEG treatment. However, an approximate 20% increase in the numbers of large cells in the > G2/M fraction was observed 34 h after PEG treatment, which substantially decreased at 36 h (Supplemental Fig. 2B). These results further support the view that reduction of cell numbers by PEG treatment occurs in large-cell fractions, possibly by apoptosis, but not by mitotic catastrophe in aneuploid cells.

Induction of apoptosis by PEG in PC-3 xenografts.  Following the PEG treatment protocol, PC-3 cells were pretreated with DMEM or 0.5% PEG and then implanted into mice. At day 7, the tumor volume of the PEG-treated group was significantly reduced compared to the DMEM-treated group. An approximate 50% reduction in tumor volume was observed with PEG treatment (Fig. 7a). To further confirm the growth-suppressive effect of PEG in an in vivo setting, we conducted the following experiment. Non-treated PC-3 cells were implanted into mice, and PBS or 0.5% PEG solution was injected into palpable tumors from day 7, twice a week (Fig. 7b top). The volumes of PC-3 xenografts increased approximately eight-fold from day 7 to day 25 in the PBS-injected group (Fig. 7b bottom). In contrast, substantial suppression of tumor growth was detected after injecting 0.5% PEG solution directly into tumors. The average sizes of tumors in PBS- and 0.5% PEG-injected groups at day 25 were 1344 ± 292 and 812 ± 215 mm3, respectively. Namely, tumor volumes were reduced approximately 30% in the PEG-injected group compared with those in the PBS group at day 25 (Fig. 7b bottom). No significant change in bodyweight was observed between the two groups during the experimental period (data not shown). Histopathological analysis revealed the presence of vacant spaces in tumors at the site of injection (Fig. 8a–d). Multinucleated cells were frequently observed surrounding these spaces in PEG-injected groups, and multinucleated cells were observed in approximately 20% of the total TUNEL-positive cells. In contrast, few TUNEL-positive cells were detected in PBS-injected tumors, although necrotic lesions were similarly observed. Multinucleated cells were rarely apparent in PBS-injected tumors (Fig. 8c,e).

Figure 7.

Growth-suppressive effects of 0.5% polyethylene glycol (PEG) on PC-3 xenografts in nude mice. (a) After PC-3 cells were treated with Dulbecco's modified Eagle's medium (DMEM) or 0.5% PEG for 30 min, the cells were implanted into the backs of nude mice at day 0. Tumor volumes were measured at day 7. (b) Experimental protocols. PC-3 cells were implanted into backs of nude mice at day 0, and then 200 µL aliquots of phosphate-buffered saline (PBS) or 0.5% PEG was injected into palpable tumors twice a week during 19 days (top). Growth curves on PC-3 xenografts injected with PBS (inline image) or 0.5% PEG (Ξ) are demonstrated (bottom). Values are the mean of fold ± SD. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 8.

Induction of multinucleated cells and apoptosis by polyethylene glycol (PEG) injection in PC-3 xenografts. Tumors with similar sizes were selected and allotted randomly into two groups. Aliquots (200 µL) of phosphate-buffered saline (PBS) or 0.5% PEG were injected directly into each group of tumors at days 7, 11, 14, 18, 21, and 25. Representative histological features of tumors in experimental groups of (a,c,e) PBS injection and (b,d,f) PEG injection are presented. The tumor sizes at day 7 before the PBS and 0.5% PEG injections were 180 ± 42 and 140 ± 70 mm3, respectively. (a,b) Hematoxylin and eosin (H&E) staining. (c–f) TdT-mediated dUTP-biotin nick end-labeling (TUNEL) immunohistochemical staining. (e,f) Magnified images of (c) and (d), respectively. The arrows indicate typical multinucleated cells positive for TUNEL staining. Scale bars = 50 µm.

Discussion

The present study demonstrated the antitumor effects of PEG via multinucleated cell formation, presumably though the induction of cell fusion. This is in line with the recent report by Roy et al. that PEG 800 induces apoptosis in two human colonic adenocarcinoma cells, HT29 and CaCo-2, at concentrations of 3.6–4.8%.(35) Although they briefly described PEG-induced cell fusion to be a possible mechanism, they considered this unlikely because a much higher concentration of PEG, around 30–50%, is generally required to induce cell fusion.(35)

In the present study, we were indeed able to demonstrate that low concentrations of PEG could induce apoptosis via multinucleated cell formation. More surprisingly, nuclear condensation was evident in a fraction of the multinucleated cells. As it has been reported that the cell-fusion process occurs within 1–2 h, mitotic catastrophe should be considered as a causative event for the induction of cell death by PEG.(36) However, the number of aneuploid cells in the > G2/M fraction did not show a significant change during this time, and the number of large multinucleated cells decreased between 34 and 36 h. Therefore, an abrupt decrease in cell numbers after PEG treatment of PC-3 cells in vitro is reasonably explained by the induction of apoptosis in multinucleated cells, although clumped cells with aneuploid characteristics may also be a part of the large-cell fraction. Further work is needed to substantiate this possibility, but the positive staining for Annexin-V and PI and PARP-1 cleavage provide some support for this hypothesis.

Taking the results together, it is plausible that multinucleated cell formation is a trigger and causative event for the induction of apoptosis by PEG. Although an increase in osmotic pressure by PEG has been considered to be important for its cytostatic or apoptotic effects,(35) it is unlikely that osmotic pressure played a role with the low concentrations of PEG used here. Indeed, when PC-3 cells were cultured in the presence of PEG under ordinary culture conditions, we did not detect apparent changes in cell-growth properties for up to 72 h (Fig. 2a). This negates the possible implication of osmotic pressure in the induction of apoptosis by the concentrations of PEG used in the present conditions. Another possibility to be considered is that cell-membrane damage caused by PEG led to the induction of apoptosis. It is thought that the cell membrane could be damaged by injection of cells with PEG particles through microcylinders in in vivo models. Further study is warranted to corroborate or refute this point. However, the induction of nuclear condensation, positive staining for Annexin-V in multinucleated cells, and PARP-1 cleavage by PEG treatment all point to a significant role for apoptosis in the growth-suppressive effects of PEG. Although the mechanism of induction of apoptosis by PEG remains unclear, we clearly demonstrated that PEG exhibited a suppressive effect on tumor growth in vivo and induced apoptosis, possibly via the cell-fusion mechanism.

Lastly, we should note that PC-3 prostate cancer cells were used as a model system. The therapies for prostate cancer, for example hormone therapy or radiation therapy, have some drawbacks. Radioactive material such as iodine-125 using brachytherapy may cause chromosomal aberrations.(37) Occurrence of androgen-independent prostate cancers because of frequent or continuous application of hormone therapy is a very serious problem at present. Once cancer becomes androgen independent, metastatic lesions manifest aggressively within 12–18 months, and the average patient survival time is only 2–3 years.(38) Although chemotherapies have been adapted for androgen-independent aggressive cancer cases, the tumor-suppressive effects brought about by those therapies are not satisfactory.(39) Therefore, for cases with localized non-invasive lesions, urologists generally choose radical prostatectomy.(40) At the same time, however, there are a substantial number of cases in which radical therapy cannot be conducted because of the poor compliance of patients and life-threatening side effects.(41,42)

In the present study, we clearly demonstrated that PEG is able to induce apoptosis in PC-3 cells in an autonomous cell-fusion manner, with efficient suppression of tumor growth in vivo. Based on our observations, we propose that direct intratumoral injection of PEG could be a promising therapeutic approach for androgen-independent prostate cancers. Clearly, if PEG is injected in clinical trials, it may induce cell fusion for normal prostate cells, and unwanted side effects may appear. However, even if PEG induces multinucleated cells between normal and tumor cells, it could be effective for diffuse types of prostate cancer. In addition, direct intratumoral injection of PEG could also assist hormone therapy and radiation therapy. Although several problems remain to be solved, the potential of PEG as a novel, non-invasive and non-toxic therapeutic agent clearly warrants further attention.

Acknowledgments

The authors thank Masako Ochiai and Ibuki Kobayashi (Biochemistry Division, National Cancer Center Research Institute) for technical help and discussion throughout this study. This work was supported in part by a Grant-in-Aid for the Third-Term Comprehensive 10-Year Strategy for Cancer Control from the Ministry of Health, Labour, and Welfare of Japan, and in part by a Grant-in-Aid for Cancer Research from the Ministry of Health, Labour, and Welfare of Japan. K. F. was a recipient of a Research Resident Fellowship from the Foundation for Promotion of Cancer Research in Japan during the research.

Ancillary