Evidence for activation of mutated p53 by apigenin in human pancreatic cancer.
Journal: 2012/May - Biochimica et Biophysica Acta - General Subjects
ISSN: 0006-3002
Abstract:
Pancreatic cancer is an exceedingly lethal disease with a five-year survival that ranks among the lowest of gastrointestinal malignancies. Part of its lethality is attributable to a generally poor response to existing chemotherapeutic regimens. New therapeutic approaches are urgently needed. We aimed to elucidate the anti-neoplastic mechanisms of apigenin-an abundant, naturally-occurring plant flavonoid-with a particular focus on p53 function. Pancreatic cancer cells (BxPC-3, MiaPaCa-2) experienced dose and time-dependent growth inhibition and increased apoptosis with apigenin treatment. p53 post-translational modification, nuclear translocation, DNA binding, and upregulation of p21 and PUMA were all enhanced by apigenin treatment despite mutated p53 in both cell lines. Transcription-dependent p53 activity was reversed by pifithrin-α, a specific DNA binding inhibitor of p53, but not growth inhibition or apoptosis suggesting transcription-independent p53 activity. This was supported by immunoprecipitation assays which demonstrated disassociation of p53/BclXL and PUMA/BclXL and formation of complexes with Bak followed by cytochrome c release. Treated animals grew smaller tumors with increased cellular apoptosis than those fed control diet. These results suggest that despite deactivating mutation, p53 retains some of its function which is augmented following treatment with apigenin. Cell cycle arrest and apoptosis induction may be mediated by transcription-independent p53 function via interactions with BclXL and PUMA. Further study of flavonoids as chemotherapeutics is warranted.
Relations:
Content
Citations
(9)
References
(29)
Diseases
(2)
Chemicals
(7)
Genes
(1)
Organisms
(4)
Processes
(3)
Anatomy
(1)
Affiliates
(1)
Similar articles
Articles by the same authors
Discussion board
Biochim Biophys Acta 1823(2): 593-604

Evidence for activation of mutated p53 by apigenin in human pancreatic cancer

1. INTRODUCTION

Pancreatic cancer is a dismal diagnosis with a 5-year survival of only 5%. The lethality of pancreatic cancer is attributable to its exceedingly malignant biology with early metastatic spread to lymph nodes and distant organs such as the liver and lungs as well as to its aggressive local invasiveness. Furthermore, patients tend to present late in the disease progression at a time when extirpative surgery-currently the only potentially curative therapy-is not feasible. Chemotherapeutic regimens based on gemcitabine offer little potential for improved survival with randomized controlled studies demonstrating only a five-week survival benefit versus control patients.[1] New therapeutics capable of extending survival and down-staging otherwise inoperable tumors in the neoadjuvant setting are urgently needed in order to improve both survival and quality of life for pancreatic cancer patients.

In response to this need recent basic science and translational research has identified naturally-occurring plant polyphenols such as apigenin as attractive candidate chemotherapeutics. Apigenin is inexpensive to administer and is non-toxic at pharmacologic doses making it possible to administer concurrently with other chemotherapeutics whilst avoiding additive toxicity.[2] Apigenin has been shown to inhibit pancreatic cancer cell growth in vitro by induction of cell cycle arrest[3] and leads to synergistic effects when combined with gemcitabine by inhibiting pro-survival pathways involving pAKT which can contribute to gemcitabine resistance.[4] Breast and colon cancer cells respond similarly to apigenin treatment with cell cycle arrest and apoptosis induction which is linked to increased phosphorylation of p53 in both wild type and mutant p53-expressing cell lines.[5,6] Modulation of p53 activity appears to be central to the effectiveness of apigenin: mechanistic studies utilizing osteosarcoma cells identified phosphorylation of p53 at threonine 55 as a stimulus for nuclear export-and thus deactivation of wild type p53 which was specifically reversed by apigenin treatment.[7]

However, the role of mutant p53 in cancer cells is multi-fold and exceedingly complex. In its native state, p53 is known to act as a transcription factor responsible for regulating the expression of cell-cycle regulatory genes such as p21/waf, and cell-death inducing proteins.[8] Classically, p53 induces cell-cycle arrest following DNA damage from UV exposure or chemical insult and initiates either DNA-repair or cell death machinery through its transcription-dependent regulation of gene transcription. More recently alternative roles of p53 have been discovered involving p53-Upregulated Modulator of Apoptosis (PUMA) and the mitochondrial pathway of programmed cell death.[9] Interactions between PUMA, p53, and anti-apoptotic Bcl-2-family proteins such as Bcl-XL have been described in which p53 acts outside of the nucleus in a transcription-independent fashion. There is a poor understanding of the degree to which mutant p53 retains its transcription-dependent and -independent functions in pancreatic cancer cells.[10] Furthermore, in addition to loss of tumor suppressor functions, recent studies have identified gain of function properties of mutant p53 involving activation of proliferative and anti-apoptotic pathways and genes responsible for drug resistance[11].

Given the complexity of the roles p53 plays in tumorogenesis and aggressive tumor biology, the use of apigenin to re-establish or augment the growth-inhibitory and proapoptotic activity of mutant p53 could be a powerful strategy to check the growth of pancreatic cancer and improve the effectiveness of existing chemotherapeutics. We hypothesized that apigenin inhibits pancreatic cancer growth through reconstitution of mutated p53 function in pancreatic cancer cells, thereby inhibiting cancer cell proliferation and inducing apoptosis.

2. METHODS

2.1. Cell Culture and Reagents

Human pancreatic cancer cell lines (MiaPaCa-2, BxPC-3) were obtained from American Type Culture Collection (ATCC) and maintained in DMEM (MiaPaCa-2) or RPMI culture medium (BxPC-3) (Invitrogen; Carlsbad, CA) supplemented with 10% fetal bovine serum and 1% penicillin-streptamycin. Cells were cultured under standard culture conditions. All solvents and regents for HPLC were HPLC grade (Fisher Scientific, Fairlawn, NJ). Luteolin, quercetin and β-glucuronidase/sulfatase (type H-5 from Helix Pomatia) were purchased from Sigma-Aldrich (St. Louis, MO). Apigenin (Purity ≥98%; Cayman Chemical, Ann Arbor, MI) and Pifithrin-α (Purity ≥95%; Sigma-Aldrich, St. Louis, MO) were dissolved in DMSO as stock solutions and stored at −20° C for in vitro experiments. Antibodies for total p53, phosphorylated p53 (ser-392), topoisomerase I, and Protein A-agarose beads were purchased from Santa Cruz Biotechnology (Santa Cruz, CA); antibodies for acetylated p53 (lys-382), PUMA, and p21 were obtained from AbCam (Cambridge, MA); and antibodies for GAPDH, Cox IV, Bcl-XL, Bax, and Bak were purchased from Cell Signaling Technology (Boston, MA).

2.2. Growth Inhibition Assay

Cells were seeded onto 96-well plates at 10 cells/well and allowed to attach overnight at 37° C. Cells were serum starved using FBS-free DMEM or RPMI and allowed to grow for another 24 hr at 37° C after which varying concentrations of Apigenin ± pifithrin-α were incubated for 24 or 48 hours. For growth inhibition, MTT (3-[4,5-Dimethyl-2-thiazolyl]-2,5-diphenyl-2H-tetrazolium bromide; Sigma-Aldrich, St. Louis, MO), was added for 3 h, cells solublized in DMSO, and plates were read with an ELISA reader at 560nm.

2.3. Apoptosis Detection Assay

Cells were plated to 6-well plates at 5 × 10 cells/well and allowed to attach overnight at 37° C. Cells were serum starved as above and allowed to grow for 24 hr at 37° C followed by treatment as for growth inhibition assays. Ultraviolet (UV) irradiation was performed by exposing cells without medium with a standard UV lamp (Sylvania G30T8, 30W; Danvers, MA) for a total dose of 50 mJ/cm. Medium was replaced following exposure and cells harvested as for apigenin and control-treated cells. A commercially available cell death detection ELISA kit (Roche Diagnostics, Mannheim, Germany) detecting nucleosomes in the cytoplasm was used according to instructions. Briefly, cells were harvested by trypsinization, counted, and 10 cells were subjected to lysis. Lysate was added to microplate wells coated with streptavidin along with anti-histone antibody conjugated with biotin and anti-DNA antibody conjugated with horseradish peroxidase (HRP) and incubated. Substrate was added and plates were read with an ELISA reader at 405nm. Enrichment Factor (EF) was calculated as a ratio of absorbance of treated cells versus that of cells treated with vehicle alone (DMSO). EF = absorbance of sample / absorbance of vehicle control.

2.4. Whole Cell Lysate and Western Blotting

Cells were grown on 10 cm dishes in standard cell culture conditions to 80% confluence and were serum starved overnight. Cells were treated with apigenin ± pifithrin-α or UV irradiation and were incubated for 30 min to 48 h. Cells were washed twice with ice-cold PBS and then underwent lysis with ice cold RIPA buffer containing Protease and Phosphatase Inhibitor Cocktail (Thermo Scientific, Rockford, IL). Cells were disrupted by repeated aspiration through a 23G needle prior to centrifugation at 14,000 × g for 10 min at 4° C. Pellet was discarded and protein concentration determined by DC Protein Assay (BioRad, Hercules, CA). Proteins (20 μg/well) were separated on 10–15% SDS-polyacrylamide gel and transferred to nitrocellulose membranes. The membranes were blocked with 5% nonfat dry milk and probed with primary antibody overnight at 4° C followed by secondary antibody. Bands were visualized by chemiluminescence substrate (Thermo Scientific, Rockford, IL) and autoradiography. Digital image analysis was performed with QuantityOne software (BioRad, Hercules, CA).

2.5. Nuclear Extraction

Cells were treated as above for western blotting. Cells were washed twice with ice-cold PBS and then incubated with cold lysis buffer (10 mM HEPES [pH 7.9], 1.5 mM MgCl2, 10 mM KCl, 0.5 mM DTT, 0.1% NP-40, 2 mM activated Na3VO4, 3 μg/mL Aprotinin, 25 μg/mL Leupeptin, 25 μg/mL Pepstatin, 25 μg/mLChymostatin, 0.2 mM PMSF, and 5 mM NaF) for 10 min on ice. Cells were collected and solubilized by pipetting up and down followed by centrifugation at 16,000 × g for 5 min at 4° C. Cytosolic supernatant was aspirated and stored at −80° C. Nuclear pellet was solubilized in nuclear lysis buffer (20 mM HEPES [pH 7.9], 1.5 mM MgCl2, 420 mM NaCl, 0.5 mM DTT, 25% glycerol, 2 mM activated Na3VO4, 3 μg/mL Aprotinin, 25 μg/mL Leupeptin, 25 μg/mL Pepstatin, 25 μg/mLChymostatin, 0.2 mM PMSF, and 5 mM NaF) and incubated for 20 min on ice followed by centrifugation at 16,000 × g for 5 min at 4° C. Protein concentration was determined and samples were stored at −80° C prior to either western blotting or p53 binding activity assay.

2.6. Mitochondrial Extraction

Cells were treated as above for western blotting. Cells were trypsinized, collected, and centrifuged at 370 × g for 5 min. Cells were washed with ice-cold NKM buffer (1 mM Tris-HCl [pH 7.4], 0.13 M NaCl, 5 mM KCl, 7.5 mM MgCl2) and re-centrifuged at 370 × g for 5 min. Wash step was repeated once and then cell pellet was resuspended in homogenization buffer (10 mM Tris-HCl [pH 6.7], 10 mM KCl, 0.15 mM MgCl2, 1 mM PMSF, 1 mM DTT) and incubated for 10 min on ice. Cells were broken by 20 strokes of a Dounce homogenizer and incubated with 2 M sucrose for 10 min on ice followed by centrifugation at 1,200 × g for 5 min at 4° C. The pellet was discarded and supernatant was combined with 2 M sucrose and samples spun at 7,000 × g for 5 min at 4° C. Cytoplasmic supernatant was aspirated and stored at −80° C. The pellet was re-suspended in mitochondrial suspension buffer (10 mM Tris-HCl [pH 6.7], 0.15 mM MgCl2, 0.25 M sucrose, 1 mM PMSF, 1 mM DTT) and centrifuged at 9,500 × g for 5 min at 4° C to pellet mitochondria. Pellet was resuspended in mitochondrial suspension buffer, protein quantified with DC protein assay, and stored at −80° C.

2.7. p53 Binding Activity Assay

Binding Activity of nuclear extracts was assayed with Human Active p53 Activity Assay (R&D Systems, Minneapolis, MN) according to manufacturer instructions. Briefly, nuclear extracts (20 μg of total protein/well) were incubated with a biotin-labeled oligonucleotide corresponding to a consensus p53-binding sequence on ice. Samples were then added to microplate wells coated with a p53 capture antibody. Wells were treated with Streptavidin-HRP and the plate was read with an ELISA reader at 450nm.

2.8. Co-Immunoprecipitation

Cells were treated as above for western blotting. Cells were washed twice with ice-cold PBS and then incubated with cold lysis buffer (10 mM HEPES [pH 7.9], 1.5 mM MgCl2, 10 mM KCl, 0.5 mM DTT, 0.1% NP-40, 2 mM activated Na3VO4, 3 μg/mL Aprotinin, 25 μg/mL Leupeptin, 25 μg/mL Pepstatin, 25 μg/mLChymostatin, 0.2 mM PMSF, and 5 mM NaF) for 10 min on ice. Cells were collected and solubilized by pipetting up and down followed by centrifugation at 16,000 × g for 5 min at 4° C. Lysate was pre-cleared with normal rabbit IgG (Santa Cruz Biotechnology, Santa Cruz, CA) and protein A-Agarose beads (Santa Cruz Biotechnology, Santa Cruz, CA). Beads were pelleted by centrifugation at 1,000 × g for 5 min at 4° C and protein quantified in the supernatant using DC protein assay. Lysate (250 μg total protein) was incubated with primary antibody for one hr on ice followed by addition of protein A-Agarose beads and incubation at 4° C overnight on a rocker platform. Beads were pelleted by centrifugation at 1,000 × g for 5 min at 4° C and pellet was washed with PBS + 1% Tween-20 four times, pelleting beads by centrifugation each time. Resulting pellet was resuspended in Laemmli buffer and boiled at 95° C for 10 min prior to centrifugation at 1,000 × g for 10 min to pellet beads. Proteins were separated and visualized by SDS-PAGE as above for western blotting.

2.9. Cytochrome C Immunofluoresence

Cells were treated as above for western blotting. Cells were trypsinized, collected, and centrifuged at 1,000 × g for 10 min. A commercially available kit was used to perform immunofluoresence for cytochrome c (EMD Chemicals, Darmstadt, Germany). Briefly, 10 cells were resuspended in permeabilization buffer and incubated for 10 min on ice. Cells were fixed with 8% paraformaldehyde for 5 min and pelleted by centrifugation at 1,000 × g for 5 min. Cells were washed × 3 with wash buffer, repeating centrifuge step each time, and then incubated with blocking buffer for one hr. Samples were incubated with primary antibody followed by secondary antibody and then pelleted as above. Cells were resuspended in wash buffer and mounted on a microscope slide with Vectashield HardSet mounting medium + DAPI (Vector Labs, Burlingame, CA) and visualized with immunofluorescence microscopy.

2.10. Experimental Animals

All animal experiments were performed according to protocols approved by the Chancellors Animal Research Committee at UCLA. Twelve four week-old male nude mice (Nu/Nu) were obtained from Charles River Laboratories (Wilmington, MA) and provided standard γ-irradiated rodent chow and autoclaved water ad libitum prior to orthotopic xenograft experiments.

2.11. Orthotopic Xenograft Model of Pancreatic Cancer

Human pancreatic cancer cells (MiaPaCa-2) were grown to 80% confluence, trypsinized, and counted. Cells were resuspended in serum free DMEM supplemented with 30% growth factor reduced Matrigel (BD Biosciences, San Jose, CA). Mice (n=6 treatment; n=6 control) were anesthetized with inhaled isoflurane (Phoenix Pharmaceutical, St. Joseph, MO), skin was prepared with Betadine (Purdue Products, Stamford, CT), and a one cm paramedian laparotomy in the left abdomen created. The spleen and tail of the pancreas was exteriorized and 5 × 10 cells injected into the tail of the pancreas. The viscera were reoriented within the abdomen and the wound closed in two layers with sutures.

2.12. Dietary Supplementation with Apigenin and Tissue Procurement

Apigenin (Purity ≥98%) was mixed with powdered AIN-76A rodent diet (Dyets; Bethlehem, PA) for in vivo feeding experiments. Animals were provided with either control diet (six animals; AIN-76A reconstituted with autoclaved water alone) or 0.2% apigenin-supplemented diet (six animals; AIN-76A + 0.2% apigenin [w/w] reconstituted with autoclaved water) ad libitum for 6 weeks following orthotopic xenograft tumor implantation. Apigenin dose was calculated based on pilot feeding experiments in which we assessed feeding tolerance and potential toxicity as well as apigenin levels in the serum and tissues. Feed supplemented with 0.2% apigenin was tolerated well by mice with no noted toxicity and tissue levels were detectable in the tens-of micromoles per liter-a level which showed cytotoxic effects in in vitro experiments (data not shown). Feed was replaced every-other day and intake of feed was measured daily. Mice were weighed weekly. At the end of the six week study period mice were sacrificed and tissues collected. Plasma was collected by exsanguinating cardiac puncture. Ten μL heparin (1,000 U/mL) was added to 1 mL of whole blood and centrifuged at 500 × g for 10 min at 4° C. Plasma was mixed with 10% (v/v) 0.58 M acetic acid and stored at −80° C. Liver, pancreas, and tumor specimens were snap frozen in liquid nitrogen and stored at −80° C. Additional liver, pancreas, and tumor specimens were fixed in 10% paraformaldehyde, paraffin embedded, and sectioned for immunohistochemical analysis.

2.13. Immunohistochemistry

Paraffin-embedded tissue sections (4 micron) were dewaxed and rehydrated in xylene and graded ethanol washes followed by deionized water and stained with the TUNEL method. Apoptosis was quantified by counting the number of TUNEL positive cells per 100 cells in 10 random microscopic fields (400×; 1000 cells counted/sample; control [n=6] and 0.2% apigenin-fed [n=6]).

2.14. HPLC for Detection of Apigenin in Plasma and Tumor Samples

Plasma samples were acidified with 10% (v/v) 0.58M acetic acid to limit flavonoid losses. Acidified plasma samples were further prepared with 0.2 M sodium acetate buffer (pH 5.0) with 1% ascorbic acid, β-glucuronidase, and sulfatase. The resulting mixture was vortexed and then incubated at 37°C for 2 hours. After incubation, ethyl acetate was added followed by 10 μL quercetin in MeOH as an internal standard. Apigenin and its metabolites were extracted with 800 μL ethyl acetate three times. In each extraction mixture was vortexed for 1 min and then centrifuged at 600 rpm for 5 min. Supernatant was transferred to a glass test tube and combined. Solvent was removed in a SpeedVac at room temp until completely dry. Residue was reconstituted in 100 μL of methanol/acetone/water (3:3:4 by volume), vortexed and then sonicated for 1 min. An aliquot of 50 μL supernatant was injected into the HPLC column.

For tissue sample analysis, frozen tissue was weighed and homogenized in sodium acetate buffer with a tissue grinder and internal standard was added. The mixture was then hydrolyzed and extracted similar to plasma samples. HPLC analysis was performed with a RP-18 Luna column (150 × 4.6 mm, 3 μm, Phenomenex, Torrance, CA) on an Agilent 1100 HPLC system comprising of an autosampler, a quaternary pump and a photodiode array detector (Agilent Technologies, Palo Alto, CA). The mobile phase consisted of a binary gradient of 0.1% (v/v) ortho-phosphoric acid in water (eluent A) and methanol (eluent B), used with flow rate of 0.5 mL/min in the following conditions: 50–80% B (0–12 min); 80% B (12–17 min); and 80–50% B (17–20 min). Column temperature was held on 28°C. The chromatograms were recorded at 337 nm and 255 nm for apigenin and its metabolites.

Data were analyzed with the Hewlett Packard Chemstation® software. Mouse plasma and tissue concentrations were determined by internal calibration. Concentration of the stock solutions was determined spectrophotometrically using Beer's Law. Calibration standards were prepared from the stock solutions by series dilution. The calibration curves generated from standard solutions of apigenin showed a linear relationship between peak area and concentration in the range of 22.07 ng/mL to 5.65 μg/mL. The limit of quantitation for apigenin was 11.0 ng/mL. Stability of apigenin in mouse diet was measured at 0, 1, 2, and 3 days following preparation by HPLC following excessive extraction with methanol.

2.15. Statistical Analysis

Descriptive statistics such as mean and standard error (SE) were summarized for all variables using bar plots. P values were calculated with two-tailed t-test and statistical significance reached at the p≤0.05 level. For analysis of animal studies, logarithm transformation was applied to pancreas total apigenin and plasma free apigenin to improve normality. Two sample t-tests were performed to study the treatment effect of 0.2% apigenin relative to control for tumor weight. Scatter plots were used to illustrate patterns of tumor weight with respect to log-transformed pancreas total apigenin and log-transformed plasma free apigenin respectively, as well as the pattern of log-transformed pancreas total apigenin with respect to log-transformed plasma free apigenin. Pearson's correlation coefficients and p-values for testing the null hypothesis of no correlation were reported.

2.1. Cell Culture and Reagents

Human pancreatic cancer cell lines (MiaPaCa-2, BxPC-3) were obtained from American Type Culture Collection (ATCC) and maintained in DMEM (MiaPaCa-2) or RPMI culture medium (BxPC-3) (Invitrogen; Carlsbad, CA) supplemented with 10% fetal bovine serum and 1% penicillin-streptamycin. Cells were cultured under standard culture conditions. All solvents and regents for HPLC were HPLC grade (Fisher Scientific, Fairlawn, NJ). Luteolin, quercetin and β-glucuronidase/sulfatase (type H-5 from Helix Pomatia) were purchased from Sigma-Aldrich (St. Louis, MO). Apigenin (Purity ≥98%; Cayman Chemical, Ann Arbor, MI) and Pifithrin-α (Purity ≥95%; Sigma-Aldrich, St. Louis, MO) were dissolved in DMSO as stock solutions and stored at −20° C for in vitro experiments. Antibodies for total p53, phosphorylated p53 (ser-392), topoisomerase I, and Protein A-agarose beads were purchased from Santa Cruz Biotechnology (Santa Cruz, CA); antibodies for acetylated p53 (lys-382), PUMA, and p21 were obtained from AbCam (Cambridge, MA); and antibodies for GAPDH, Cox IV, Bcl-XL, Bax, and Bak were purchased from Cell Signaling Technology (Boston, MA).

2.2. Growth Inhibition Assay

Cells were seeded onto 96-well plates at 10 cells/well and allowed to attach overnight at 37° C. Cells were serum starved using FBS-free DMEM or RPMI and allowed to grow for another 24 hr at 37° C after which varying concentrations of Apigenin ± pifithrin-α were incubated for 24 or 48 hours. For growth inhibition, MTT (3-[4,5-Dimethyl-2-thiazolyl]-2,5-diphenyl-2H-tetrazolium bromide; Sigma-Aldrich, St. Louis, MO), was added for 3 h, cells solublized in DMSO, and plates were read with an ELISA reader at 560nm.

2.3. Apoptosis Detection Assay

Cells were plated to 6-well plates at 5 × 10 cells/well and allowed to attach overnight at 37° C. Cells were serum starved as above and allowed to grow for 24 hr at 37° C followed by treatment as for growth inhibition assays. Ultraviolet (UV) irradiation was performed by exposing cells without medium with a standard UV lamp (Sylvania G30T8, 30W; Danvers, MA) for a total dose of 50 mJ/cm. Medium was replaced following exposure and cells harvested as for apigenin and control-treated cells. A commercially available cell death detection ELISA kit (Roche Diagnostics, Mannheim, Germany) detecting nucleosomes in the cytoplasm was used according to instructions. Briefly, cells were harvested by trypsinization, counted, and 10 cells were subjected to lysis. Lysate was added to microplate wells coated with streptavidin along with anti-histone antibody conjugated with biotin and anti-DNA antibody conjugated with horseradish peroxidase (HRP) and incubated. Substrate was added and plates were read with an ELISA reader at 405nm. Enrichment Factor (EF) was calculated as a ratio of absorbance of treated cells versus that of cells treated with vehicle alone (DMSO). EF = absorbance of sample / absorbance of vehicle control.

2.4. Whole Cell Lysate and Western Blotting

Cells were grown on 10 cm dishes in standard cell culture conditions to 80% confluence and were serum starved overnight. Cells were treated with apigenin ± pifithrin-α or UV irradiation and were incubated for 30 min to 48 h. Cells were washed twice with ice-cold PBS and then underwent lysis with ice cold RIPA buffer containing Protease and Phosphatase Inhibitor Cocktail (Thermo Scientific, Rockford, IL). Cells were disrupted by repeated aspiration through a 23G needle prior to centrifugation at 14,000 × g for 10 min at 4° C. Pellet was discarded and protein concentration determined by DC Protein Assay (BioRad, Hercules, CA). Proteins (20 μg/well) were separated on 10–15% SDS-polyacrylamide gel and transferred to nitrocellulose membranes. The membranes were blocked with 5% nonfat dry milk and probed with primary antibody overnight at 4° C followed by secondary antibody. Bands were visualized by chemiluminescence substrate (Thermo Scientific, Rockford, IL) and autoradiography. Digital image analysis was performed with QuantityOne software (BioRad, Hercules, CA).

2.5. Nuclear Extraction

Cells were treated as above for western blotting. Cells were washed twice with ice-cold PBS and then incubated with cold lysis buffer (10 mM HEPES [pH 7.9], 1.5 mM MgCl2, 10 mM KCl, 0.5 mM DTT, 0.1% NP-40, 2 mM activated Na3VO4, 3 μg/mL Aprotinin, 25 μg/mL Leupeptin, 25 μg/mL Pepstatin, 25 μg/mLChymostatin, 0.2 mM PMSF, and 5 mM NaF) for 10 min on ice. Cells were collected and solubilized by pipetting up and down followed by centrifugation at 16,000 × g for 5 min at 4° C. Cytosolic supernatant was aspirated and stored at −80° C. Nuclear pellet was solubilized in nuclear lysis buffer (20 mM HEPES [pH 7.9], 1.5 mM MgCl2, 420 mM NaCl, 0.5 mM DTT, 25% glycerol, 2 mM activated Na3VO4, 3 μg/mL Aprotinin, 25 μg/mL Leupeptin, 25 μg/mL Pepstatin, 25 μg/mLChymostatin, 0.2 mM PMSF, and 5 mM NaF) and incubated for 20 min on ice followed by centrifugation at 16,000 × g for 5 min at 4° C. Protein concentration was determined and samples were stored at −80° C prior to either western blotting or p53 binding activity assay.

2.6. Mitochondrial Extraction

Cells were treated as above for western blotting. Cells were trypsinized, collected, and centrifuged at 370 × g for 5 min. Cells were washed with ice-cold NKM buffer (1 mM Tris-HCl [pH 7.4], 0.13 M NaCl, 5 mM KCl, 7.5 mM MgCl2) and re-centrifuged at 370 × g for 5 min. Wash step was repeated once and then cell pellet was resuspended in homogenization buffer (10 mM Tris-HCl [pH 6.7], 10 mM KCl, 0.15 mM MgCl2, 1 mM PMSF, 1 mM DTT) and incubated for 10 min on ice. Cells were broken by 20 strokes of a Dounce homogenizer and incubated with 2 M sucrose for 10 min on ice followed by centrifugation at 1,200 × g for 5 min at 4° C. The pellet was discarded and supernatant was combined with 2 M sucrose and samples spun at 7,000 × g for 5 min at 4° C. Cytoplasmic supernatant was aspirated and stored at −80° C. The pellet was re-suspended in mitochondrial suspension buffer (10 mM Tris-HCl [pH 6.7], 0.15 mM MgCl2, 0.25 M sucrose, 1 mM PMSF, 1 mM DTT) and centrifuged at 9,500 × g for 5 min at 4° C to pellet mitochondria. Pellet was resuspended in mitochondrial suspension buffer, protein quantified with DC protein assay, and stored at −80° C.

2.7. p53 Binding Activity Assay

Binding Activity of nuclear extracts was assayed with Human Active p53 Activity Assay (R&D Systems, Minneapolis, MN) according to manufacturer instructions. Briefly, nuclear extracts (20 μg of total protein/well) were incubated with a biotin-labeled oligonucleotide corresponding to a consensus p53-binding sequence on ice. Samples were then added to microplate wells coated with a p53 capture antibody. Wells were treated with Streptavidin-HRP and the plate was read with an ELISA reader at 450nm.

2.8. Co-Immunoprecipitation

Cells were treated as above for western blotting. Cells were washed twice with ice-cold PBS and then incubated with cold lysis buffer (10 mM HEPES [pH 7.9], 1.5 mM MgCl2, 10 mM KCl, 0.5 mM DTT, 0.1% NP-40, 2 mM activated Na3VO4, 3 μg/mL Aprotinin, 25 μg/mL Leupeptin, 25 μg/mL Pepstatin, 25 μg/mLChymostatin, 0.2 mM PMSF, and 5 mM NaF) for 10 min on ice. Cells were collected and solubilized by pipetting up and down followed by centrifugation at 16,000 × g for 5 min at 4° C. Lysate was pre-cleared with normal rabbit IgG (Santa Cruz Biotechnology, Santa Cruz, CA) and protein A-Agarose beads (Santa Cruz Biotechnology, Santa Cruz, CA). Beads were pelleted by centrifugation at 1,000 × g for 5 min at 4° C and protein quantified in the supernatant using DC protein assay. Lysate (250 μg total protein) was incubated with primary antibody for one hr on ice followed by addition of protein A-Agarose beads and incubation at 4° C overnight on a rocker platform. Beads were pelleted by centrifugation at 1,000 × g for 5 min at 4° C and pellet was washed with PBS + 1% Tween-20 four times, pelleting beads by centrifugation each time. Resulting pellet was resuspended in Laemmli buffer and boiled at 95° C for 10 min prior to centrifugation at 1,000 × g for 10 min to pellet beads. Proteins were separated and visualized by SDS-PAGE as above for western blotting.

2.9. Cytochrome C Immunofluoresence

Cells were treated as above for western blotting. Cells were trypsinized, collected, and centrifuged at 1,000 × g for 10 min. A commercially available kit was used to perform immunofluoresence for cytochrome c (EMD Chemicals, Darmstadt, Germany). Briefly, 10 cells were resuspended in permeabilization buffer and incubated for 10 min on ice. Cells were fixed with 8% paraformaldehyde for 5 min and pelleted by centrifugation at 1,000 × g for 5 min. Cells were washed × 3 with wash buffer, repeating centrifuge step each time, and then incubated with blocking buffer for one hr. Samples were incubated with primary antibody followed by secondary antibody and then pelleted as above. Cells were resuspended in wash buffer and mounted on a microscope slide with Vectashield HardSet mounting medium + DAPI (Vector Labs, Burlingame, CA) and visualized with immunofluorescence microscopy.

2.10. Experimental Animals

All animal experiments were performed according to protocols approved by the Chancellors Animal Research Committee at UCLA. Twelve four week-old male nude mice (Nu/Nu) were obtained from Charles River Laboratories (Wilmington, MA) and provided standard γ-irradiated rodent chow and autoclaved water ad libitum prior to orthotopic xenograft experiments.

2.11. Orthotopic Xenograft Model of Pancreatic Cancer

Human pancreatic cancer cells (MiaPaCa-2) were grown to 80% confluence, trypsinized, and counted. Cells were resuspended in serum free DMEM supplemented with 30% growth factor reduced Matrigel (BD Biosciences, San Jose, CA). Mice (n=6 treatment; n=6 control) were anesthetized with inhaled isoflurane (Phoenix Pharmaceutical, St. Joseph, MO), skin was prepared with Betadine (Purdue Products, Stamford, CT), and a one cm paramedian laparotomy in the left abdomen created. The spleen and tail of the pancreas was exteriorized and 5 × 10 cells injected into the tail of the pancreas. The viscera were reoriented within the abdomen and the wound closed in two layers with sutures.

2.12. Dietary Supplementation with Apigenin and Tissue Procurement

Apigenin (Purity ≥98%) was mixed with powdered AIN-76A rodent diet (Dyets; Bethlehem, PA) for in vivo feeding experiments. Animals were provided with either control diet (six animals; AIN-76A reconstituted with autoclaved water alone) or 0.2% apigenin-supplemented diet (six animals; AIN-76A + 0.2% apigenin [w/w] reconstituted with autoclaved water) ad libitum for 6 weeks following orthotopic xenograft tumor implantation. Apigenin dose was calculated based on pilot feeding experiments in which we assessed feeding tolerance and potential toxicity as well as apigenin levels in the serum and tissues. Feed supplemented with 0.2% apigenin was tolerated well by mice with no noted toxicity and tissue levels were detectable in the tens-of micromoles per liter-a level which showed cytotoxic effects in in vitro experiments (data not shown). Feed was replaced every-other day and intake of feed was measured daily. Mice were weighed weekly. At the end of the six week study period mice were sacrificed and tissues collected. Plasma was collected by exsanguinating cardiac puncture. Ten μL heparin (1,000 U/mL) was added to 1 mL of whole blood and centrifuged at 500 × g for 10 min at 4° C. Plasma was mixed with 10% (v/v) 0.58 M acetic acid and stored at −80° C. Liver, pancreas, and tumor specimens were snap frozen in liquid nitrogen and stored at −80° C. Additional liver, pancreas, and tumor specimens were fixed in 10% paraformaldehyde, paraffin embedded, and sectioned for immunohistochemical analysis.

2.13. Immunohistochemistry

Paraffin-embedded tissue sections (4 micron) were dewaxed and rehydrated in xylene and graded ethanol washes followed by deionized water and stained with the TUNEL method. Apoptosis was quantified by counting the number of TUNEL positive cells per 100 cells in 10 random microscopic fields (400×; 1000 cells counted/sample; control [n=6] and 0.2% apigenin-fed [n=6]).

2.14. HPLC for Detection of Apigenin in Plasma and Tumor Samples

Plasma samples were acidified with 10% (v/v) 0.58M acetic acid to limit flavonoid losses. Acidified plasma samples were further prepared with 0.2 M sodium acetate buffer (pH 5.0) with 1% ascorbic acid, β-glucuronidase, and sulfatase. The resulting mixture was vortexed and then incubated at 37°C for 2 hours. After incubation, ethyl acetate was added followed by 10 μL quercetin in MeOH as an internal standard. Apigenin and its metabolites were extracted with 800 μL ethyl acetate three times. In each extraction mixture was vortexed for 1 min and then centrifuged at 600 rpm for 5 min. Supernatant was transferred to a glass test tube and combined. Solvent was removed in a SpeedVac at room temp until completely dry. Residue was reconstituted in 100 μL of methanol/acetone/water (3:3:4 by volume), vortexed and then sonicated for 1 min. An aliquot of 50 μL supernatant was injected into the HPLC column.

For tissue sample analysis, frozen tissue was weighed and homogenized in sodium acetate buffer with a tissue grinder and internal standard was added. The mixture was then hydrolyzed and extracted similar to plasma samples. HPLC analysis was performed with a RP-18 Luna column (150 × 4.6 mm, 3 μm, Phenomenex, Torrance, CA) on an Agilent 1100 HPLC system comprising of an autosampler, a quaternary pump and a photodiode array detector (Agilent Technologies, Palo Alto, CA). The mobile phase consisted of a binary gradient of 0.1% (v/v) ortho-phosphoric acid in water (eluent A) and methanol (eluent B), used with flow rate of 0.5 mL/min in the following conditions: 50–80% B (0–12 min); 80% B (12–17 min); and 80–50% B (17–20 min). Column temperature was held on 28°C. The chromatograms were recorded at 337 nm and 255 nm for apigenin and its metabolites.

Data were analyzed with the Hewlett Packard Chemstation® software. Mouse plasma and tissue concentrations were determined by internal calibration. Concentration of the stock solutions was determined spectrophotometrically using Beer's Law. Calibration standards were prepared from the stock solutions by series dilution. The calibration curves generated from standard solutions of apigenin showed a linear relationship between peak area and concentration in the range of 22.07 ng/mL to 5.65 μg/mL. The limit of quantitation for apigenin was 11.0 ng/mL. Stability of apigenin in mouse diet was measured at 0, 1, 2, and 3 days following preparation by HPLC following excessive extraction with methanol.

2.15. Statistical Analysis

Descriptive statistics such as mean and standard error (SE) were summarized for all variables using bar plots. P values were calculated with two-tailed t-test and statistical significance reached at the p≤0.05 level. For analysis of animal studies, logarithm transformation was applied to pancreas total apigenin and plasma free apigenin to improve normality. Two sample t-tests were performed to study the treatment effect of 0.2% apigenin relative to control for tumor weight. Scatter plots were used to illustrate patterns of tumor weight with respect to log-transformed pancreas total apigenin and log-transformed plasma free apigenin respectively, as well as the pattern of log-transformed pancreas total apigenin with respect to log-transformed plasma free apigenin. Pearson's correlation coefficients and p-values for testing the null hypothesis of no correlation were reported.

3. RESULTS

3.1. Apigenin Inhibits Pancreatic Cancer Cell Growth and Induces Apoptosis in vitro

MiaPaCa-2 and BxPC-3 cells treated with apigenin at concentrations of 1–100 uM resulted in dose-dependent decrease in cell proliferation as detected by MTT with an IC50 of ~25–50 uM for MiaPaCa-2 and ~1–10 uM for BxPC-3 at 24 hours following treatment. Inhibition of proliferation was time-dependent as well (Figure 1A and B). These data were confirmed by cell counts with trypan blue staining and 5-Bromo-2'-deoxy-uridine (BrdU) incorporation ELISA (Supplemental Figure 1A and B). There was a 7.7±2.9-fold increase in apoptosis for MiaPaCa-2 cells treated with 25 uM apigenin for 24 hours and 16.5±5.0-fold increase for similarly treated BxPC-3 cells (P<0.05; Figure 1C). This exceeded the amount of apoptosis for cells treated with UV irradiation (50mJ/cm).

An external file that holds a picture, illustration, etc.
Object name is nihms-351716-f0001.jpg

Apigenin inhibited pancreatic cancer cell growth dose- and time- dependently for MiaPaCa (A) and BxPC-3 (B) cells. Apigenin dose-dependently induced apoptosis for MiaPaCa and BxPC-3 cells (C). *P≤0.05; **P<0.01, #P<0.01 versus control

3.2. Apigenin Modulates Post-Translational Modification, Nuclear Translocation, and DNA Binding of p53

Apigenin treatment led to increased acetylation of lysine 382 (75.8±12.5% at 8hr, P<0.05; Figure 2A)-a signal associated with nuclear translocation and stabilization of the p53 tetramer. Increased acetylation followed a time course that showed a peak at 4 to 8 hours for BxPC-3 and MiaPaCa-2 cells, respectively-though a measurable increase in acetylation was observed as early as 30 minutes following treatment with 25 uM apigenin (25.6±9.3% increase). Accordingly, nuclear extracts showed increased levels of p53 protein relative to cytoplasmic extracts peaking at 2 hours following apigenin treatment (52.4±8.9% increase, P<0.01; Figure 2B), indicating nuclear translocation of p53. Nuclear translocation was also accompanied by increased p53 binding to a consensus promoter sequence detected by ELISA despite the presence of mutated p53 in both cell lines tested (MiaPaCa-2 65.8±0.6% increase and BxPC-3 48.9±0.3% increase, P<0.01; Figure 2C). These events were all in the context of a slight decrease in p53 expression observed in whole cell lysates (24.3±3.5% decrease at 12hr, P<0.01; Figure 2D). These data indicate that treatment with apigenin may increase the ability of p53 to bind to promoter regions of targeted genes and thus partially restore the function of the mutant p53 protein. Interestingly the magnitude of increased p53 DNA binding was similar for both cell lines indicating the possibility of additional mechanisms responsible for the anti-proliferative effects of apigenin.

An external file that holds a picture, illustration, etc.
Object name is nihms-351716-f0002.jpg

Apigenin induced acetylation of p53 at lysine 382 (AcK382) time-dependently in pancreatic cancer cells (A), and was associated with nuclear translocation which peaked at 2 hours (B), and increased DNA binding of p53 at 4 hours (C) despite overall reduced expression of p53 (D). *P≤0.05 versus control; **P<0.01 versus control.

3.3. Apigenin-Induced Apoptosis is Not Dependent on p53-Mediated Transcriptional Regulation

In non-neoplastic cells, p53 activation has been shown to cause cell cycle arrest and cell death via activating expression of numerous cell cycle and/or apoptosis mediators such as p21 and PUMA (p53 Upregulated Mediator of Apoptosis). These proteins then mediate downstream events such as cyclin-dependent kinase inhibition (p21) and caspase activation (PUMA). Apigenin treatment (25 uM) induced the expression of p21 and PUMA, consistent with our data showing nuclear translocation and increased DNA binding of p53 (Figure 3A). This upregulation was blocked by the addition of 25 uM pifithrin-α (PFT), a specific inhibitor of p53 DNA binding (Figure 3B). However, the anti-proliferative effect of apigenin was not reversed by treatment with pifithrin-α nor was increased apoptosis (Figure 3C,D) indicating that apigenin may activate p53 through a parallel, transcriptionally independent pathway of programmed cell death.

An external file that holds a picture, illustration, etc.
Object name is nihms-351716-f0003.jpg

Apigenin induced expression of p21 and PUMA (A) and co-treatment with pifithrin-α reversed p53-mediated expression of p21 and PUMA (B) but was not able to reverse growth inhibition (C) or apoptosis (D) following apigenin treatment. **P<0.01 versus control.

3.4. Apigenin Induces Mitochondrial Translocation and Modulates Protein-Protein Interaction of p53 and PUMA with Bcl-XL

Consistent with a model of transcriptionally-independent p53 activation we found there was increased translocation of p53 to the mitochondria in apigenin-treated cells (31% increase; Figure 4A). Immunopreciptiation of the anti-apoptotic Bcl-2 protein Bcl-XL showed there was dissolution of complexes between Bcl-XL/p53 and Bcl-XL/PUMA following apigenin treatment. Binding of p53 and PUMA with Bcl-XL effectively sequestered these molecules in the unstressed cells (Figure 4B). Following liberation from Bcl-XL, p53 and PUMA both formed stable complexes with Bak leading to mitochondrial membrane permeabilization and cytochrome c release seen by immunofluoresence (Figure 4C).

An external file that holds a picture, illustration, etc.
Object name is nihms-351716-f0004.jpg

Apigenin induced mitochondrial translocation of p53 (A) and influenced interaction of p53 and PUMA with the anti-apoptotic BH-3 domain protein Bcl XL and induced binding between p53 and Bak (B). Immunofluorescence of pancreatic cancer cells stained for cytochrome c. Release of cytochrome c is marked by dimunition of fluorescence signal from mitochondria (arrows) (C). In 4A, GAPDH loading controls for lanes 1, 3 and 5 represent protein loading for cytosolic fractions of the mitochondrial isolation experiment and are roughly equal. A GAPDH signal is missing in lanes 2, 4 and 6 and demonstrate purity of the mitochondrial fraction (i.e.: minimal cytosolic contamination). Conclusions discussed in the Results and Discussion section are made from the relative increase in p53 signal between lanes 2 and 4 where the loading control signal (Cox IV) is also roughly equal.

3.5. Dietary Apigenin Reduces the Volume of Orthotopic Pancreatic Cancer Xenografts in Nude Mice

Using a well-established nude mouse model of human pancreatic cancer we tested the hypothesis that dietary apigenin supplementation would yield smaller pancreatic tumors compared to mice fed standard rodent diet. All mice fed and gained weight appropriately with no gross evidence of toxicity. Serum chemistries were performed at the conclusion of the feeding period and there were no significantly elevated indicies among a panel testing for pancreatic, renal, and hepatic function. Finally, feed samples were tested for the stability of apigenin by HPLC and there was no significant degradation of apigenin over 72 hours at ambient humidity and temperature (data not shown).

Apigenin-fed mice had significantly elevated levels of apigenin detected in both plasma (0.3±0.04 vs. 0.2±0.01 μg/mL, P<0.05; Figure 5A) and tumor (1.8±0.7 vs 0.2±0.1 mg/g tissue, P<0.05; Figure 5B). Tumors tended to be smaller in apigenin-fed mice (0.7±0.1 vs. 0.9±0.1 grams, P=0.058, Figure 5C) and there was a strong correlation between tumor apigenin level and tumor size (Pearson correlation coefficient −0.65, P<0.05; Figure 5D) indicating mice that ingested the most apigenin had the smallest tumors. TUNEL staining performed on tumor sections demonstrated an increased proportion of apoptotic cells for the treated tumors (56.3±6.2 vs. 33.2±8.3 TUNEL positive cells/HPF; P<0.05; Figure 6).

An external file that holds a picture, illustration, etc.
Object name is nihms-351716-f0005.jpg

Animals fed a diet supplemented with 0.2% apigenin (w/w) had higher levels of apigenin detected in their plasma (A) and pancreas (B) and trended towards smaller tumors compared to control animals (C). There was a correlation between tumor apigenin and tumor weight (Pearson's Correlation Coefficient = −0.65, P=0.04; D); (n=6 0.2% apigenin-fed; n=6 control-fed).

An external file that holds a picture, illustration, etc.
Object name is nihms-351716-f0006.jpg

TUNEL staining of tumor sections revealed more tumor cell apoptosis in apigenin-treated animals compared with control (A,B). *P<0.05 versus control (n=6 0.2% apigenin-fed; n=6 control-fed).

3.1. Apigenin Inhibits Pancreatic Cancer Cell Growth and Induces Apoptosis in vitro

MiaPaCa-2 and BxPC-3 cells treated with apigenin at concentrations of 1–100 uM resulted in dose-dependent decrease in cell proliferation as detected by MTT with an IC50 of ~25–50 uM for MiaPaCa-2 and ~1–10 uM for BxPC-3 at 24 hours following treatment. Inhibition of proliferation was time-dependent as well (Figure 1A and B). These data were confirmed by cell counts with trypan blue staining and 5-Bromo-2'-deoxy-uridine (BrdU) incorporation ELISA (Supplemental Figure 1A and B). There was a 7.7±2.9-fold increase in apoptosis for MiaPaCa-2 cells treated with 25 uM apigenin for 24 hours and 16.5±5.0-fold increase for similarly treated BxPC-3 cells (P<0.05; Figure 1C). This exceeded the amount of apoptosis for cells treated with UV irradiation (50mJ/cm).

An external file that holds a picture, illustration, etc.
Object name is nihms-351716-f0001.jpg

Apigenin inhibited pancreatic cancer cell growth dose- and time- dependently for MiaPaCa (A) and BxPC-3 (B) cells. Apigenin dose-dependently induced apoptosis for MiaPaCa and BxPC-3 cells (C). *P≤0.05; **P<0.01, #P<0.01 versus control

3.2. Apigenin Modulates Post-Translational Modification, Nuclear Translocation, and DNA Binding of p53

Apigenin treatment led to increased acetylation of lysine 382 (75.8±12.5% at 8hr, P<0.05; Figure 2A)-a signal associated with nuclear translocation and stabilization of the p53 tetramer. Increased acetylation followed a time course that showed a peak at 4 to 8 hours for BxPC-3 and MiaPaCa-2 cells, respectively-though a measurable increase in acetylation was observed as early as 30 minutes following treatment with 25 uM apigenin (25.6±9.3% increase). Accordingly, nuclear extracts showed increased levels of p53 protein relative to cytoplasmic extracts peaking at 2 hours following apigenin treatment (52.4±8.9% increase, P<0.01; Figure 2B), indicating nuclear translocation of p53. Nuclear translocation was also accompanied by increased p53 binding to a consensus promoter sequence detected by ELISA despite the presence of mutated p53 in both cell lines tested (MiaPaCa-2 65.8±0.6% increase and BxPC-3 48.9±0.3% increase, P<0.01; Figure 2C). These events were all in the context of a slight decrease in p53 expression observed in whole cell lysates (24.3±3.5% decrease at 12hr, P<0.01; Figure 2D). These data indicate that treatment with apigenin may increase the ability of p53 to bind to promoter regions of targeted genes and thus partially restore the function of the mutant p53 protein. Interestingly the magnitude of increased p53 DNA binding was similar for both cell lines indicating the possibility of additional mechanisms responsible for the anti-proliferative effects of apigenin.

An external file that holds a picture, illustration, etc.
Object name is nihms-351716-f0002.jpg

Apigenin induced acetylation of p53 at lysine 382 (AcK382) time-dependently in pancreatic cancer cells (A), and was associated with nuclear translocation which peaked at 2 hours (B), and increased DNA binding of p53 at 4 hours (C) despite overall reduced expression of p53 (D). *P≤0.05 versus control; **P<0.01 versus control.

3.3. Apigenin-Induced Apoptosis is Not Dependent on p53-Mediated Transcriptional Regulation

In non-neoplastic cells, p53 activation has been shown to cause cell cycle arrest and cell death via activating expression of numerous cell cycle and/or apoptosis mediators such as p21 and PUMA (p53 Upregulated Mediator of Apoptosis). These proteins then mediate downstream events such as cyclin-dependent kinase inhibition (p21) and caspase activation (PUMA). Apigenin treatment (25 uM) induced the expression of p21 and PUMA, consistent with our data showing nuclear translocation and increased DNA binding of p53 (Figure 3A). This upregulation was blocked by the addition of 25 uM pifithrin-α (PFT), a specific inhibitor of p53 DNA binding (Figure 3B). However, the anti-proliferative effect of apigenin was not reversed by treatment with pifithrin-α nor was increased apoptosis (Figure 3C,D) indicating that apigenin may activate p53 through a parallel, transcriptionally independent pathway of programmed cell death.

An external file that holds a picture, illustration, etc.
Object name is nihms-351716-f0003.jpg

Apigenin induced expression of p21 and PUMA (A) and co-treatment with pifithrin-α reversed p53-mediated expression of p21 and PUMA (B) but was not able to reverse growth inhibition (C) or apoptosis (D) following apigenin treatment. **P<0.01 versus control.

3.4. Apigenin Induces Mitochondrial Translocation and Modulates Protein-Protein Interaction of p53 and PUMA with Bcl-XL

Consistent with a model of transcriptionally-independent p53 activation we found there was increased translocation of p53 to the mitochondria in apigenin-treated cells (31% increase; Figure 4A). Immunopreciptiation of the anti-apoptotic Bcl-2 protein Bcl-XL showed there was dissolution of complexes between Bcl-XL/p53 and Bcl-XL/PUMA following apigenin treatment. Binding of p53 and PUMA with Bcl-XL effectively sequestered these molecules in the unstressed cells (Figure 4B). Following liberation from Bcl-XL, p53 and PUMA both formed stable complexes with Bak leading to mitochondrial membrane permeabilization and cytochrome c release seen by immunofluoresence (Figure 4C).

An external file that holds a picture, illustration, etc.
Object name is nihms-351716-f0004.jpg

Apigenin induced mitochondrial translocation of p53 (A) and influenced interaction of p53 and PUMA with the anti-apoptotic BH-3 domain protein Bcl XL and induced binding between p53 and Bak (B). Immunofluorescence of pancreatic cancer cells stained for cytochrome c. Release of cytochrome c is marked by dimunition of fluorescence signal from mitochondria (arrows) (C). In 4A, GAPDH loading controls for lanes 1, 3 and 5 represent protein loading for cytosolic fractions of the mitochondrial isolation experiment and are roughly equal. A GAPDH signal is missing in lanes 2, 4 and 6 and demonstrate purity of the mitochondrial fraction (i.e.: minimal cytosolic contamination). Conclusions discussed in the Results and Discussion section are made from the relative increase in p53 signal between lanes 2 and 4 where the loading control signal (Cox IV) is also roughly equal.

3.5. Dietary Apigenin Reduces the Volume of Orthotopic Pancreatic Cancer Xenografts in Nude Mice

Using a well-established nude mouse model of human pancreatic cancer we tested the hypothesis that dietary apigenin supplementation would yield smaller pancreatic tumors compared to mice fed standard rodent diet. All mice fed and gained weight appropriately with no gross evidence of toxicity. Serum chemistries were performed at the conclusion of the feeding period and there were no significantly elevated indicies among a panel testing for pancreatic, renal, and hepatic function. Finally, feed samples were tested for the stability of apigenin by HPLC and there was no significant degradation of apigenin over 72 hours at ambient humidity and temperature (data not shown).

Apigenin-fed mice had significantly elevated levels of apigenin detected in both plasma (0.3±0.04 vs. 0.2±0.01 μg/mL, P<0.05; Figure 5A) and tumor (1.8±0.7 vs 0.2±0.1 mg/g tissue, P<0.05; Figure 5B). Tumors tended to be smaller in apigenin-fed mice (0.7±0.1 vs. 0.9±0.1 grams, P=0.058, Figure 5C) and there was a strong correlation between tumor apigenin level and tumor size (Pearson correlation coefficient −0.65, P<0.05; Figure 5D) indicating mice that ingested the most apigenin had the smallest tumors. TUNEL staining performed on tumor sections demonstrated an increased proportion of apoptotic cells for the treated tumors (56.3±6.2 vs. 33.2±8.3 TUNEL positive cells/HPF; P<0.05; Figure 6).

An external file that holds a picture, illustration, etc.
Object name is nihms-351716-f0005.jpg

Animals fed a diet supplemented with 0.2% apigenin (w/w) had higher levels of apigenin detected in their plasma (A) and pancreas (B) and trended towards smaller tumors compared to control animals (C). There was a correlation between tumor apigenin and tumor weight (Pearson's Correlation Coefficient = −0.65, P=0.04; D); (n=6 0.2% apigenin-fed; n=6 control-fed).

An external file that holds a picture, illustration, etc.
Object name is nihms-351716-f0006.jpg

TUNEL staining of tumor sections revealed more tumor cell apoptosis in apigenin-treated animals compared with control (A,B). *P<0.05 versus control (n=6 0.2% apigenin-fed; n=6 control-fed).

4. DISCUSSION

Our initial growth inhibition and apoptosis experiments indicated that apigenin exhibited potent anti-proliferative and pro-apoptotic effects on pancreatic adenocarcinoma cells in vitro. These results are consistent with prior reported work with pancreatic cancer cells treated with apigenin. The concentrations of apigenin required to inhibit growth and induce apoptosis were somewhat higher than those attained in in vivo murine experiments. Several factors may contribute to this phenomenon: cell culture experiments were performed in serum-free medium where the solubility of apigenin is limited, effectively decreasing its bioavailability in vitro.[12] Additionally, pharmacologic studies of apigenin have shown that apigenin sequesters in the lipid fraction of organs; the resulting high volume of distribution makes the serum concentration of apigenin a relatively small component of the total apigenin level in vivo.[13] These variables may increase the apparent IC50 in vitro by necessitating a large amount of apigenin dissolved in the medium to attain the same total apigenin concentration seen in vivo.

p53 has long been recognized to be a critical factor in cell cycle progression and apoptosis in untransformed cells following genotoxic insults such as UV or gamma irradiation.[14] Furthermore mutation of the Tp53 gene has been implicated as a transformative factor in numerous human malignancies, including pancreatic adenocarcinoma where it is mutated in up to 76% of resected tumors and 67% of established cell lines.[15,16] MiaPaCa-2 and BxPC-3 cells have Tp53 mutations in codons corresponding to the DNA-binding domain of the p53 protein (codon 248 and 273 [M.P.] and 220 and 265 [Bx],[17]). Mutations in these codons are the most commonly seen alterations of the Tp53 gene for pancreatic cancers as well as many other types of human malignancy. There is evidence that these mutations may have characteristic functional consequences affecting protein half-life, conformation, and DNA sequence-specificity.[18] Tp53 mutation, along with activating mutation of Kras, is thought to be a key event in pancreatic carcinogenesis. Importantly, the MiaPaCa-2 cell line has mutated Kras while BxPC-3 is unusual in that it has a wild-type gene. Given both cell lines ultimately respond to apigenin treatment we surmised that apigenin may act through a mechanism independent of Kras. For this reason, our choice of both Kras mutant and wild-type cell lines was helpful in directing our investigation away from the RAS/RAF/MAPK pathway. We chose to study p53 because it is a key molecule uniquely situated in cellular machinery responsible for both cell cycle inhibition and apoptosis induction. Furthermore, we hypothesized that loss-of-function mutations in Tp53 seen in pancreatic cancer may be partially circumvented by treatment with apigenin leading to cell cycle arrest and apoptosis seen in in vitro growth assays.

In the current study, in vitro experiments demonstrated transcription-dependent functionality of p53 despite the presence of deactivating mutations in the Tp53 gene in both cell lines tested. This finding is in concordance with previous reports in the literature studying the effect of apigenin in widely disparate cancer cell lines, including carcinoma and sarcoma cell lines.[57] Apigenin treatment was accompanied by increases in acetylation of lysine-392. This modification is carried out by the acetyltransferases CBP/p300 and is known to affect p53 sequence-specific DNA binding and p53 half-life.[19] Consequently, binding of p53 to DNA was also promoted by apigenin treatment. p53 fulfilled its classically described role as an activator of transcription of the cell cycle regulatory gene p21 and the pro-apoptotic gene PUMA correlating to our findings of decreased cellular proliferation and increased apoptosis. Importantly, this transcriptional activation was promoted by apigenin treatment and completely abrogated by co-treatment with pifithrin-α-a specific inhibitor of p53/DNA binding-indicating p53 transcriptional activation was causally related to apigenin treatment.

Based upon these results apigenin appeared to restore the transcription-dependent activation of p53. However, we did not find a correlation between the relative sensitivity of the two cell lines to apigenin's anti-proliferative and pro-apoptotic effects to the increase in p53 activation (i.e.: BxPC-3 IC50 was two-to-five fold lower than MiaPaCa-2 while p53-DNA binding was similar). Furthermore, we found there was very little reversal of the anti-proliferative and pro-apoptotic effects of apigenin when transcription-dependent p53 activation was blocked with pifithrin-α. These date indicate the possibility of parallel pathway(s) involving the [co-] activation of PUMA as has been demonstrated in neoplastic and non-neoplastic cells with intact p53 function.[20,21]

Current models of the p53/PUMA apoptotic pathway describe dissolution of the p53/Bcl-XL complex accompanied by creation of a PUMA/Bcl-XL pair which is required for `freeing' p53 prior to its activation of the downstream BH123 proteins such as Bax or Bak.[20] These findings were reported by Chipuk et al using HCT116 colorectal carcinoma cells and mouse embryonic fibroblasts which express wild-type p53. Both nuclear p53, which incites transcriptional activation of PUMA, as well as cytosolic p53, which binds to and activates Bax, are necessary for induction of apoptosis in this model as shown by the lack of PUMA-mediated apoptosis in p53 null cells. This model has been disputed by Callus et al who observed continued PUMA-mediated apoptosis in p53-null mouse embryonic fibroblasts and murine myeloid cell lines undergoing forced overexpression of PUMA (thus obviating the need for p53-mediated transcriptional upregulation of PUMA) and contended that the role of p53 in PUMA-mediated apoptosis was limited to transcriptional upregulation of PUMA.[22]

Our results indicate that there may be a combination of these two models of transcription-independent p53 function in pancreatic cancer cell lines. Apigenin treatment resulted in increased expression of PUMA (via p53 transcriptional activation) and coimmunoprecipitation demonstrated dissociation of stable complexes between both p53/Bcl-XL and PUMA/Bcl-XL, indicating that p53 and PUMA may act in parallel to activate Bax following dissociation from Bcl- XL. In the case of PUMA/Bcl-XL, there was no detectable complex following treatment with apigenin as opposed p53/Bcl-XL where there was a partial dissolution. This finding is supported by structure/function and affinity binding studies which have shown that p53-though it lacks a true BH3 domain-binds to Bcl-XL through its DNA binding domain to the `bottom side' of Bcl-XL-at residues distinct from its BH3 domain.[23] The dissociation constant for the p53/Bcl-XL complex was 164±54 nM KD versus 10±4 nM KD for PUMA/Bcl-XL.[20]

Once p53 and PUMA are freed from sequestration by Bcl-XL both formed complexes with pro-apoptotic Bak, suggesting both proteins act as BH3-only molecules in this situation. These events appear to occur in parallel in contrast to the findings of Chipuk et al in which p53 and PUMA `flip-flop' their association with Bcl-XL so that p53 can then activate pro-apoptotoic Bax.[20]

Further support for this model comes from studies in which another polyphenolic flavonoid, rottlerin, was found to alter interactions of Bcl-XL with the BH3-only proteins PUMA and Bim in a manner similar to a known Bcl-XL inhibitor, BH3I-2' which preferentially binds to the hydrophobic pocket of Bcl-XL, preventing its binding with BH3-only proteins.[24] These experiments were performed in the MiaPaCa-2 and Panc-1 pancreatic cancer cell lines, making their findings directly applicable to this study. Based on these data we propose that apigenin induces dissociation of p53 and PUMA from sequestration by Bcl-XL allowing activation of Bak by both p53 and PUMA at the mitochondrial membrane leading to cytochrome c release and initiation of the caspase-driven mitochondrial cell death cascade as has been shown by others.[25] Release of p53 from Bcl-XL appears to have the added effect of nuclear translocation of p53 leading to increased transcription of p53 target genes, including PUMA. However, the transcriptionally-dependent function of p53 does not appear to be necessary for induction of apoptosis (Figure 7). It is not clear whether apigenin's effect is direct through its binding to p53, PUMA, or Bcl-XL or indirect via its known ability to inhibit the activity of numerous kinases including PKC, PI3-kinase, and the JAK/STAT pathway.[26,27] Additional experiments utilizing a p53 null cell line, or `knock-down' of p53 may help to further elucidate the role of p53 in the mechanism of apigenin.

An external file that holds a picture, illustration, etc.
Object name is nihms-351716-f0007.jpg

Proposed mechanism of apigenin-induced apoptosis in pancreatic cancer cells. The shaded area indicates activity inhibited by pifithrin-α.

Our in vivo studies were designed to demonstrate the efficacy of apigenin as an orally administered chemotherapeutic. Our primary outcome measure was tumor size following 6 weeks of dietary supplementation of apigenin. While the tumors in treated animals were smaller than control diet-fed mice this difference was not statistically significant (P=0.058). Correlation of apigenin levels with tumor size showed a significant reduction in tumor volume in mice with the highest tissue levels of apigenin. Because apigenin was administered as a dietary supplement ad libitum (as opposed to gavage feed), mice that ate more feed were exposed to greater doses of apigenin and may have achieved higher tissue levels. This along with individual differences in absorption and elimination of apigenin and the dose-dependency of apigenin's anti-neoplastic properties contributed to the inhomogeneity of the data, and the apparent lack of treatment effect when examining tumor volume alone. The small sample sizes (n=6 treatment, n=6 control mice) also may have contributed to the lack of significance through type II error (accepting null hypothesis when treatment effect truly exists due to insufficient sample size). While the administration method employed may be seen as a weakness in our study design it does allow us to conclude that apigenin supplementation at a level known to be non-toxic and well-tolerated (as opposed to mega-doses that may be intolerable or otherwise unachievable in human subjects) can have a positive effect on tumor size and lead to increased tumor apoptosis as seen with TUNEL staining.

Further study is warranted, particularly in the area of apigenin as a supplemental therapy in combination with standard chemotherapeutics such as gemcitabine. Studies have indicated a facultative role of apigenin in increasing the efficacy of gemcitabine's anti-tumor effects.[4,28] This polypharmacy strategy may also help to improve the tolerability of chemotherapeutic regimens by reducing the dose necessary to realize a treatment effect. It is not known whether apigenin would have a similar effect when combined with newer chemotherapy combinations. At least one study has indicated that apigenin may actually decrease the effectiveness of standard chemotherapeutics necessitating a deeper understanding of the exact mechanism of apigenin's effects and a more targeted approach to combination with existing drugs.[27] Finally, apigenin may have chemopreventative properties which would halt the progression of pre-neoplastic PanIN lesions to invasive carcinoma.[29] Apigenin's anti-neoplastic properties as they relate to the timing of p53 mutation, which is thought to be a late event in the dysplasia-carcinoma continuum, would be of particular interest in light of our findings here.

Although these data do not definitively establish a causal mechanism between Bcl-XL, PUMA, and p53, they do indicate that the roles of p53 and PUMA may be parallel and facultative, not sequential and inexorably linked as has been previously proposed. Furthermore, our data show that even mutated p53 retains some of its cell cycle regulatory and pro-apoptotic function(s) in the presence of growth-inhibitory stimuli such as apigenin treatment. Most importantly from a translational perspective apigenin reduced the size of xenograft tumors in a dose-dependent fashion with greater cellular apoptosis, thus correlating in vitro observations with in vivo data. These findings contribute additional support to a growing body of evidence indicating apigenin may hold promise as a chemotherapeutic for the treatment of pancreatic cancer.

Supplementary Material

01

01

Click here to view.(25K, doc)
Department of Surgery, David Geffen School of Medicine at UCLA 72-107 CHS, 10833 LeConte Ave, Los Angeles, CA 90095
Hirshberg Laboratories for Pancreatic Cancer Research 2535 MRL, 675 Charles E Young Drive, Los Angeles, CA 90095
Center for Human Nutrition, Department of Medicine, David Geffen School of Medicine at UCLA 14-155 Warren Hall, 900 Veteran Ave, Los Angeles, CA 90095
Department of Biostatistics, School of Public Health at UCLA 26-070 CHS, 10833 LeConte Ave, Los Angeles, CA 90095
Center for Excellence in Pancreatic Diseases, David Geffen School of Medicine at UCLA 13-146 Warren Hall, 900 Veteran Ave, Los Angeles, CA 90095
To whom correspondence and reprint request should be addressed at Department of Surgery, David Geffen School of Medicine at UCLA, Los Angeles, CA. Phone: 310-206-0441, Fax: 310-206-2472, ude.alcu.tendem@seniheoj
Publisher's Disclaimer

Abstract

Pancreatic cancer is an exceedingly lethal disease with a five-year survival that ranks among the lowest of gastrointestinal malignancies. Part of its lethality is attributable to a generally poor response to existing chemotherapeutic regimens. New therapeutic approaches are urgently needed. We aimed to elucidate the anti-neoplastic mechanisms of apigenin-an abundant, naturally-occurring plant flavonoid-with a particular focus on p53 function. Pancreatic cancer cells (BxPC-3, MiaPaCa-2) experienced dose and time-dependent growth inhibition and increased apoptosis with apigenin treatment. p53 post-translational modification, nuclear translocation, DNA binding, and upregulation of p21 and PUMA were all enhanced by apigenin treatment despite mutated p53 in both cell lines. Transcription-dependent p53 activity was reversed by pifithrin-α, a specific DNA binding inhibitor of p53, but not growth inhibition or apoptosis suggesting transcription-independent p53 activity. This was supported by immunoprecipitation assays which demonstrated disassociation of p53/BclXL and PUMA/BclXL and formation of complexes with Bak followed by Cytochrome c release. Treated animals grew smaller tumors with increased cellular apoptosis than those fed control diet. These results suggest that despite deactivating mutation, p53 retains some of its function which is augmented following treatment with apigenin. Cell cycle arrest and apoptosis induction may be mediated by transcription-independent p53 function via interactions with BclXL and PUMA. Further study of flavonoids as chemotherapeutics is warranted

Abstract

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Footnotes

REFERENCES

REFERENCES

References

  • 1. Storniolo AM, Enas NH, Brown CA, Voi M, Rothenberg ML, Schilsky RAn investigational new drug treatment program for patients with gemcitabine: Results for over 3000 patients with pancreatic carcinoma. Cancer. 1999;85:1261–1268.[PubMed][Google Scholar]
  • 2. Meyer H, Bolarinwa A, Wolfram G, Linseisen JBioavailability of apigenin from apiin-rich parsley in humans. Ann Nutr Metab. 2006;50:167–172.[PubMed][Google Scholar]
  • 3. Ujiki MB, Ding XZ, Salabat MR, Bentrem DJ, Golker L, Milam B, Talamonti MS, Bell RH, Jr., Iwamura, Adrian TEApigenin inhibits pancreatic cancer cell proliferation through G2/M cell cycle arrest. Mol Cancer. 2006;5:76–83.[Google Scholar]
  • 4. Strouch MJ, Milam BM, Melstrom LG, McGill JJ, Salabat MR, Ujiki MB, Ding XZ, Bentrem DJThe flavonoid apigenin potentiates the growth inhibitory effects of gemcitabine and abrogates gemcitabine resistance in human pancreatic cancer cells. Pancreas. 2009;38:409–415.[PubMed][Google Scholar]
  • 5. Seo HS, Ju JH, Jang K, Shin IInduction of apoptotic cell death by phytoestrogens by up-regulating the levels of phospho-p53 and p21 in normal and malignant estrogen receptor α-negative breast cells. Nutr Res. 2011;31:139–146.[PubMed][Google Scholar]
  • 6. Zhong Y, Krisanapun C, Lee SH, Nualsanit T, Sams C, Peungvicha P, Baek SJMolecular targets of apigenin in colorectal cancer cells: Involvement of p21, NAG-1 and p53. Eur J Cancer. 2010;46:3365–74.[Google Scholar]
  • 7. Cai X, Liu XInhibition of Thr-55 phosphorylation restores p53 nuclear localization and sensitizes cancer cells to DNA damage. Proc Natl Acad Sci USA. 2008;105:16958–16963.[Google Scholar]
  • 8. Janus F, Albrechtsen N, Dornreiter I, Wiesmüller L, Grosse F, Deppert WThe dual role model for p53 in maintaining genomic integrity. Cell Mol Life Sci. 1999;55:12–27.[PubMed][Google Scholar]
  • 9. Nakano K, Vousden KHPUMA, a novel proapoptotic gene, is induced by p53. Mol Cell. 2001;7:683–694.[PubMed][Google Scholar]
  • 10. Brosh R, Rotter BWhen mutants gain new powers: News from the mutant p53 field. Nat Rev. 2009;9:701–713.[PubMed][Google Scholar]
  • 11. Blandino G, Levine AJ, Oren MMutant p53 gain of function: differential effects of different p53 mutants on resistance of cultured cells to chemotherapy. Oncogene. 1999;18:477–485.[PubMed][Google Scholar]
  • 12. Boyong L, Robinson DH, Birt DFEvaluation of properties of apigenin and [G-3H] apigenin and analytic method development. J Pharm Sci. 1997;86:721–725.[PubMed][Google Scholar]
  • 13. Gradolatto A, Basly JP, Berges R, Teyssier C, Chagnon MC, Siess MH, Canivenc-Lavier MCPharmacokinetics and metabolism of apigenin in female and male rats after a single oral administration. Drug Metab Dispos. 2005;33:49–54.[PubMed][Google Scholar]
  • 14. Kastan MB, Onyekwere O, Sidransky D, Vogelstein B, Craig RWParticipation of p53 protein in the cellular response to DNA damage. Cancer Res. 1991;51:6304–6311.[PubMed][Google Scholar]
  • 15. Suwa H, Yoshimura T, Yamaguchi N, Kanehira K, Manabe T, Imamura M, Hiai H, Fukumoto MK-ras and p53 alterations in genomic DNA and transcripts of human pancreatic adenocarcinoma cell lines. Jpn J Cancer Res. 1994;85:1005–1014.[Google Scholar]
  • 16. Rozenblum E, Schutte M, Goggins M, Hahn SA, Panzer S, Zahurak M, Goodman SN, Sohn TA, Hruban RH, Yeo CJ, Kern SETumor-suppressive pathways in pancreatic carcinoma. Cancer Res. 1997;57:1731–1734.[PubMed][Google Scholar]
  • 17. Ruggeri B, Zhang SY, Caamano J, DiRado M, Flynn SD, Klein-Szanto AJHuman pancreatic adenocarcinomas and cell lines reveal frequent and multiple alterations in the p53 and Rb-1 tumor-suppressor genes. Oncogene. 1992;7:1503–1511.[PubMed][Google Scholar]
  • 18. Hussain SP, Harris CCp53 mutation spectrum and load: the generation of hypotheses linking the exposure of endogenous or exogenous carcinogens to human cancer. Mutation Research. 1999;428:23–32.[PubMed][Google Scholar]
  • 19. Liu L, Scolnick DM, Trievel RC, Zhang HB, Marmorstein R, Halazonetis TD, Berger SLp53 sites acetylated in vitro by PCAF and p300 are acetylated in vivo in response to DNA damage. Mol Cell Biol. 1999;19:1202–1209.[Google Scholar]
  • 20. Chipuk JE, Bouchier-Hayes L, Kuwana T, Newmeyer DD, Green DRPUMA couples the nuclear and cytoplasmic proapoptotic function of p53. Science. 2005;309:1732–1735.[PubMed][Google Scholar]
  • 21. Yu J, Wang Z, Kinzler KW, Vogelstein B, Zhang LPUMA mediates the apoptotic response to p53 in colorectal cancer cells. Proc Natl Acad Sci USA. 2003;100:1931–1936.[Google Scholar]
  • 22. Callus BA, Ekert PG, Heraud JE, Jabbour AM, Kotevski A, Vince JE, Silke J, Vaux DLCytoplasmic p53 is not required for PUMA-induced apoptosis. Cell Death Diff. 2008;15:213–215.[PubMed][Google Scholar]
  • 23. Mihara M, Erster S, Zaika A, Petrenko O, Chittenden T, Pancoska P, Moll UMp53 has a direct apoptogenic role at the mitochondria. Mol Cell. 2003;11:577–590.[PubMed][Google Scholar]
  • 24. Ohno I, Eibl G, Odinokova I, Edderkaoui M, Damoiseaux RD, Yazbec M, Abrol R, Goddard WA, III, Yokosuka O, Pandol SJ, Gukovskaya ASRottlerin stimulates apoptosis in pancreatic cancer cells through interactions with proteins of the Bcl-2 family. Am J Physiol Gastrointest Liver Physiol. 2010;298:G63–73.[Google Scholar]
  • 25. Schuler M, Bossy-Wetzl E, Goldstein JC, Fitzgerald P, Green DRp53 induces apoptosis by caspace activation through mitochondrial cytochrome c release. J Biol Chem. 2000;275:7337–7342.[PubMed][Google Scholar]
  • 26. Agullo G, Gamet-Payrastre L, Manenti S, Viala C, Remesy C, Chap H, Payrastre BRelationship between flavonoid structure and inhibition of phosphatidylinositol 3-kinase: a comparison with tyrosine kinase and protein kinase c inhibition. Biochem Pharmacol. 1997;53:1649–1657.[PubMed][Google Scholar]
  • 27. Ruela-de-Sousa RR, Fuhler GM, Blom N, Ferrerra CV, Aoyama H, Peppelenbosch MPCytotoxicity of apigenin on leukemia cell lines: Implications for prevention and therapy. Cell Death Dis. 2010;1:1–12.[Google Scholar]
  • 28. Lee SH, Ryu JK, Lee KY, Woo SM, Park JK, Yoo JW, Kim YT, Yoon YBEnhanced anti-tumor effect of combination therapy with gemcitabine and apigenin in pancreatic cancer. Cancer Letters. 2005;259:39–49.[PubMed][Google Scholar]
  • 29. Patel D, Shukla S, Gupta SApigenin and cancer chemoprevention: Progress, potential and promise (a review) Int J Oncol. 2007;30:233–245.[PubMed][Google Scholar]
Collaboration tool especially designed for Life Science professionals.Drag-and-drop any entity to your messages.