Potential anti-cancer effects of virgin olive oil phenolson colorectal carcinogenesis models in vitro
Article first published online: 4 MAY 2005
Copyright © 2005 Wiley-Liss, Inc.
International Journal of Cancer
Volume 117, Issue 1, pages 1–7, 20 October 2005
How to Cite
Gill, C. I.R., Boyd, A., McDermott, E., McCann, M., Servili, M., Selvaggini, R., Taticchi, A., Esposto, S., Montedoro, G., McGlynn, H. and Rowland, I. (2005), Potential anti-cancer effects of virgin olive oil phenolson colorectal carcinogenesis models in vitro. Int. J. Cancer, 117: 1–7. doi: 10.1002/ijc.21083
- Issue published online: 2 AUG 2005
- Article first published online: 4 MAY 2005
- Manuscript Accepted: 18 JAN 2005
- Manuscript Received: 15 OCT 2004
- olive oil phenols;
- DNA damage;
- colorectal cancer
The traditional Mediterranean diet is thought to represent a healthy lifestyle; especially given the incidence of several cancers including colorectal cancer is lower in Mediterranean countries compared to Northern Europe. Olive oil, a central component of the Mediterranean diet, is believed to beneficially affect numerous biological processes. We used phenols extracted from virgin olive oil on a series of in vitro systems that model important stages of colon carcinogenesis. The effect the extract on DNA damage induced by hydrogen peroxide was measured in HT29 cells using single cell microgel-electrophoresis. A significant anti-genotoxic linear trend (p = 0.011) was observed when HT29 cells were pre-incubated with olive oil phenols (0, 5, 10, 25, 50, 75, 100 μg/ml) for 24 hr, then challenged with hydrogen peroxide. The olive oil phenols (50, 100 μg/ml) significantly (p = 0.004, p = 0.002) improved barrier function of CACO2 cells after 48 hr as measured by trans-epithelial resistance. Significant inhibition of HT115 invasion (p < 0.01) was observed at olive oil phenols concentrations of 25, 50, 75, 100 μg/ml using the matrigel invasion assay. No effect was observed on HT115 viability over the concentration range 0, 25, 50 75, 100 μg/ml after 24 hr, although 75 and 100 μg/ml olive oil phenols significantly inhibited HT115 cell attachment (p = 0.011, p = 0.006). Olive oil phenols had no significant effect on metastasis-related gene expression in HT115 cells. We have demonstrated that phenols extracted from virgin olive oil are capable of inhibiting several stages in colon carcinogenesis in vitro. © 2005 Wiley-Liss, Inc.
The traditional Mediterranean diet has long been thought of as a healthy lifestyle,1 which is of interest given the fact that the incidence of cancer, especially colon and breast, is lower in Mediterranean countries than in Northern Europe.2 Olive oil is a central component of the Mediterranean diet and is believed to beneficially affect numerous biological processes.3 The oil also usually represents the main source of fat in the diet.4 The consumption of added lipids to the diet was found to be lower in the United States than in Greece,5 but the former conversely suffers a higher incidence rate of CRC, breast and prostate cancer.6 Several studies suggest that the differences might not be related to the amount of fat consumed but rather the type of fat and other components of the diet.7, 8 A study by Franceschi et al.9 on 1,953 Italian subjects with CRC from 6 areas observed that high intakes of polyunsaturated fatty acids, derived chiefly from olive oils and seed oils, were associated negatively with CRC risk.
A paucity of data exists regarding the effect of olive oil or its phenols on carcinogenesis. A recent study by Bartoli et al.10 showed that dietary olive oil reduced the incidence of aberrant crypt foci in azoxymethane-treated rats. This suggests that olive oil may have chemo-protective effects against colon carcinogenesis. A similar outcome was observed for dimethylbenz(α)antracene-induced mammary carcinogenesis (rat), where dietary olive oil supplementation reduced tumour incidence, multiplicity and volume.11 A recent in vitro study by Llor et al.12 demonstrated that olive oil was shown to induce significant levels of apoptosis (HT29 and CACO2) and cell differentiation (CACO2) but did not inhibit proliferation of either cell line. Furthermore, the oil was shown to downregulate the expression of COX-2 and BCL-2 proteins that have a crucial role in colorectal carcinogenesis.
Although it is unclear exactly which components have a role in the protective effect of olive oil, phenols seem a reasonable choice for investigation. First, phenolic compounds are more abundant in olive oil than seed oils.13 Second, lignans and secoiridoids, which form a major component of the phenolic fraction of olive oil,14, 15 have anticancer properties in general.16, 17, 18, 19 The major phenolic compounds identified and quantified in olive oil belong to 3 different classes: simple phenols (hydroxytyrosol, tyrosol), secoiridoids (oleuropein, the aglycone of ligstroside, and their respective decarboxylated dialdehyde derivatives) and the lignans [(+)-1-acetoxypinoresinol and pinoresinol]. All 3 classes have potent antioxidant properties.14 It must be considered, however, that not all phenolic compounds will reach the colon. Vissers et al.20 reported that approximately 55–66 mol/100 mol of olive oil phenols were absorbed in the small intestine. Less than 4 mol/100 mol of ingested tyrosol and hydroxytyrosol will reach the colon, whereas the amount of ingested oleuropein aglycone and ligstroside aglycone reaching the colon remains unknown. Lignans (or their metabolites) are also believed to exert their effect directly on the colonic epithelium possibly meditated by the gut microflora.21
The aim of our in vitro study was to examine the effects of olive oil phenols on a selection of biomarkers biologically relevant to colorectal cancer. The olive oil phenols used in our study were extracted from virgin olive oil22 and have a total phenolic value of 34.19 mg/100 mg crude extract (Table I, Table II). A crude phenolic extract from virgin olive oil was chosen rather than individual phenols to best represent the phenolic mix that would be present in the diet during consumption of virgin olive oil. The use of the crude extract allows for the possibility of synergistic activity, not achievable through study of single compounds or even limited combinations. The biomarkers chosen to test the efficacy of olive oil phenols represent the major stages in carcinogenesis: initiation, promotion and metastasis. One of the major cancer sites thought to be protected by olive oil is the colon, so for our in vitro studies we chose cell lines used widely as models for colorectal cancer including HT29 for anti-genotoxicity/genotoxicity study,23, 24, 25 CACO2 for barrier function26, 27, 28 and HT115 for matrigel invasion.29, 30, 31 The cell-based biomarker assays comprise various well-established colon cell lines selected for their specific biological suitability e.g., HT115 is the most invasive of the cell lines used on our study and CACO2 cells establish a suitable epithelial layer to study barrier function.
|Free acidity (g oleic acid/100 g oil)||0.38|
|Peroxide index (meq O2/Kg oil)||8.5|
|Spectrophotometric index UV|
|Total phenols (mg/Kg)||460|
|Phenolic compounds||mg/100 mgcrude extract|
|3,4-Dihydroxyphenyl ethanol (3,4 DHPEA)||0.80|
|P-Hydroxyphenyl ethanol (p-HPEA)||0.41|
|Dialdehydic form of elenolic acid linked to 3,4 DHPEA (3,4 DHPEA-EDA)||29.81|
|Dialdehydic form of elenolic acid linked to p-HPEA (p-HPEA-EDA)||8.24|
|(+)-1-acetoxipinoresinol and (+)-pinoresinol1||7.23|
|Oleuropein aglycon (3,4-DHPEA-EA)||11.64|
Material and methods
Virgin olive oil
Drupes of Moraiolo cultivar, grown in the region of Umbria and harvested during the first week of November 2001, were used. Virgin olive oil was extracted at an industrial plant by Rapanelli SPA (Foligno, Italy). The olives were crushed using a hammer crusher, malaxed (to soften by stirring with a thinner substance) for 40 min at 27°C and extracted by centrifugation using a ratio of olive paste to water 1:0.2 (w/v). The oil was filtered and stored at 6°C until analysis.
Our procedure is related to the separation and identification of the standard compounds that were used to define the real concentration of selected phenols in the extract, as the secoiridoid derivative is not available commercially. The procedure of separation and NMR characterization of the standard was reported extensively in a previous article.22 The 3,4-(dihydroxyphenyl) ethanol (3,4-DHPEA) was synthesized in the laboratory according to the procedure of Baraldi et al.32 The dialdehydic form of elenolic acid linked to 3,4-DHPEA or p-hydroxyphenyl (p-HPEA) (3,4-DHPEA-EDA and p-HPEA-EDA, respectively) and the isomer of oleuropein aglycon (3,4-DHPEA-EA) were extracted from virgin olive oil using a procedure reported previously.22, 23 The purity of these substances was tested by high performance liquid chromatography (HPLC) and their chemical structures verified by nuclear magnetic resonance.22p-HPEA was obtained from Janssen Chemical Co. (Beerse, Belgium).
Extraction procedure of phenolic compounds
Virgin olive oil (1 Kg) was mixed with 500 g of methanol: water 80:20 v/v for 3 min using an ultra-turrax model T50 (IKA Labortechnik, Staufen, Germany) and centrifuged at 5,000g for 20 min. The extraction was repeated twice. The methanolic extracts were aspirated, pooled and concentrated under vacuum at 35°C in a nitrogen flow until the complete solvent removal. The crude extract was dissolved in acetonitrile (100 ml) and defatted by adding 200 ml of hexane for 3 times. The acetonitrile was evaporated, from the purified extract, under vacuum at 30°C in a nitrogen flow.33 The extract was reconstituted in DMSO to a stock concentration of 100 mg/ml, aliquoted (20 μl) and stored at −80°C under a nitrogen atmosphere until use. Olive oil phenols were fully soluble at the concentration range tested in all analyses. When used in experimental analysis all concentrations of olive oil phenols used (and controls) were normalized to a final concentration of 0.1% DMSO in cell culture media.
Free acidity, peroxide value, K232, K270 and Δ K of virgin olive oil was evaluated according to the Official European Method of analysis (Reg. E.U. 2568/91), The evaluation of phenolic composition of virgin olive oil crude extract was estimated colorimetrically at 765 nm using the Folin-Ciocalteau reagent and expressed as 3,4-(dihydroxyphenyl) ethanol equivalents as reported previously.33 The HPLC system was a Varian 9010 chromatograph with a 155 mm × 4.6 mm C18 Inertisil ODS-3 Column (Alltech Co.) coupled to a Varian Polycrom 9065 UV diode array detector. The sample was dissolved in methanol and a sample loop of 20 μl capacity was used. The mobile phase comprised water acidified with 0.2% of acetic acid (pH = 3.1) and methanol, the flow rate was 1.5 ml/min and the gradient used was the same as reported in Montedoro et al.33
HT29, HT115, CACO2 and MRC5 cells were obtained from the European Collection of Animal Cell Cultures (ECACC), Salisbury, UK. DMEM and MEM were obtained from Gibco Life Technologies Ltd. (Paisley, Scotland). HT29 and HT115 cells were cultured in Roux flasks as monolayers in DMEM containing 10% and 15% FBS respectively, 2 mM glutamine and 100 U/l penicillin/streptomycin. CACO2 and MRC5 cells were cultured in Roux flasks as monolayers in MEM containing 10% FBS, 2 mM glutamine and 100 U/l penicillin/streptomycin and 1% non-essential amino acids (HT115 only). Cells were cultured for 7 days (<75% confluence) at 37°C with 5% CO2 and 95% filtered air. The medium was changed every 2 days. Cells were washed with PBS for 2 min and re-suspended by the addition of trypsin (0.25% trypsin-EDTA) at 37°C for 5 min. Cells were centrifuged at 258g for 3 min and cells re-suspended in the appropriate medium.
The assay was carried out as described by Rieger et al.34 using the well-established HT29 cell model for colonic DNA damage. The HT29 cells were incubated for 24 hr with olive oil phenolic extracts at various concentrations (0, 5, 10, 25, 50, 75, 100 μg/ml) before harvest for use in the comet analysis. To assess the anti-genotoxic potential of the olive oil phenolic extract, the cultures of HT29 cells were treated with hydrogen peroxide (75 μM H2O2) for 5 min on ice, then centrifuged for 5 min at 258g. HT29 cells were treated omitting the oxidative insult to assess the genotoxic potential of the extract cultures. Positive (75 μM H2O2) negative controls (PBS) were included in all experiments (HT29 cells that had not been exposed to the extract). The supernatant was discarded and the cell pellet re-suspended in 75 μl of 0.85% low melting point agarose (LMPA) in PBS and maintained in a water bath at 40°C. The suspension was added to gels prepared previously (1% normal agarose) on frosted slides and coverslips were added. The slides were immersed in lysis buffer (2.5 M NaCl, 100 mM Na2EDTA, 10 mM TRIS) for 1 hr at 4°C and then placed in electrophoresis buffer and allowed to unwind for 20 min before running at 26 V 300 mA for 20 min. After electrophoresis, gels were washed 3 × 5 min in neutralisation buffer (0.4 M Tris, pH = 7.5) at 4°C. All gels were stained with 20 μl of ethidium bromide (2 μg/ml) before scoring. Images were analysed at 400× magnification using a Nikon eclipse 600 epi-fluorescence microscope. % Tail DNA was recorded using Komet 3.0 image analysis software (Kinetic Imaging Ltd, Liverpool, UK). For each slide, 100 cells were scored. Positive (hydrogen peroxide = 75 μM) and negative (PBS) controls were included for all experiments. The mean was calculated from 100 cells/gel (each sample in triplicate) and the experiment repeated independently 4 times. The mean of each set of data was used in the statistical analysis. Differences between means were evaluated by ANOVA, post-hoc LSD (p < 0.05).
Trans-epithelial resistance assay
CACO2 cells are capable of enterocytic differentiation forming atypical brush border membranes and tight junctions35 and provide an appropriate and frequently used model for permeability, barrier function and transport studies.36, 37 The complexity and the number of tight-junction strands correlate with the electrical resistance of the barrier. Briefly, the CACO2 cell suspension was seeded in 6-well plates with Transwell inserts (0.1% rat tail collagen coated polyethyleneterapthalate membranes; BD Bioscience, Bedford, UK) at a density of 2.5 × 105 cells per insert. The culture medium was replaced (apical = 1.5 ml, basal side = 2.5 ml) every other day for 14 days. Cells were maintained at 37°C in an atmosphere of 5% CO2 and 95% relative humidity. From Days 11–14 the integrity of the monolayer was evaluated by measuring the trans-epithelial resistance (TER) (expressed as Ωcm2) using an EVOM epithelial voltohmmeter (World Precision Instruments Ltd., Aston, UK). Once the TER values had stabilised, the inserts were ready for experimentation. The TER of the CACO2 cell monolayers was measured at 0, 24 and 48 hr after the addition of olive oil phenols (10, 50, 100 μg/ml) to the apical compartment. The values were in the range of 280–430 Ωcm2 at baseline. The treatments were carried out in quadruplicate and the experiment repeated independently 3 times. The mean of each set of data was used in the statistical analysis. Differences between means were evaluated by ANOVA, post-hoc LSD (p < 0.05).
Matrigel invasion assay
The assay was modified from the method of Parr et al.29 MRC5 foetal lung cells are used to enhance invasion rates of the HT115 cells through secretion of hepatocyte growth factor, which is a strong chemo-attractant and has been shown to increase cell scattering and motility in HT115 cells.38, 39 In brief, 6-well Biocoat Matrigel invasion chambers (inserts) (BD Bioscience) were rehydrated with the addition of 2 ml warm (37°C) serum-free culture medium for 2 hr. In parallel to, but separate from insert rehydration, 2 ml of MRC5 cell suspension (4 × 105 cells/well) were seeded in the plate and were incubated at 37°C for 2 hr. The media was removed carefully from the plate eliminating any unattached cells in the process and replaced with 2 ml of DMEM containing 10% FBS per well. Upon completion of insert rehydration, the inserts were transferred to the plate containing the MRC5 cells. The media were removed from the inserts and replaced with 2 ml of HT115 cell suspension (2 × 105 cells, serum free DMEM) in the presence or absence of olive oil phenols (0, 5, 25, 50, 75, 100 μg/ml). Inserts containing cells alone served as a control. Plates were incubated for 24 hr. After incubation, medium was removed and cells on both sides of the membrane were fixed in 70% ethanol for 30 min and then stained with hematoxylin. Using a cotton bud, the non-invasive cells were removed from half of the surface of the insert by “scrubbing.” This process was repeated for the invasive cells on the other side of the insert. The number of invasive and non-invasive cells were then counted in 5 random fields of the insert and % invasion was calculated. An estimate of total cell numbers was also taken by adding the mean number of invasive and non-invasive cells counted on the inserts. The treatments were carried out in duplicate and the experiment repeated independently three times. The mean of each set of data was used in the statistical analysis. Differences between means were evaluated by ANOVA, post-hoc LSD (p < 0.05).
TIMP/MMP gene expression
To avoid degradation of RNA by the activity of ribonucleases, the potent RNase inhibitor diethyl pyrocarbonate (DEPC) was used to treat all solutions and plastic ware before RNA isolation (Sigma Chemicals, Poole, Dorset, UK). All solutions were made using water that had been treated with 0.1% DEPC overnight at 37°C and then autoclaved to remove trace amounts of DEPC. Pipette tips and Eppendorfs were also treated with 0.1% DEPC water overnight at 37°C, dried and then autoclaved to remove traces of the inhibitor.
RNA was extracted using the High Pure RNA Isolation kit and according to the manufacturer's protocol (Roche Molecular Biochemicals, Mannheim, Germany). The extracted RNA was quantified by obtaining the optical density at 260 nm on a Pharmacia Genequant UV spectrophotometer.
cDNA was synthesised using the Applied Biosystems TaqMan Multiscribe Reverse Transcription reagent system. The cDNA was stored at −20°C for future use in quantitative real time PCR reactions.
Primers, probes and cycle numbers as detailed in Table III were used to detect the relative abundance of the genes after normalisation to an 18S endogenous control. The 18S endogenous control was detected by Applied Biosystems human 18S endogenous control assay. The relative abundance of each of the target genes was then calculated using the ΔΔCt method as described in Applied Biosystems User Bulletin 2 ABI PRISM 7700 Sequence Detection System “Relative Gene Expression” (179).
|Gene||Forward primer||Probe1||Reverse primer||No. cycles|
cDNA synthesised previously was used in the quantitative real time PCR assay, 2.5 μl of cDNA was added to 12.5 μl of TaqMan Universal PCR master mix, 2.5 μl of forward primer, 2.5 μl of reverse primer, 2.5 μl of probe and 2.5 μl of water in a 96-well optical plate and covered with an optical adhesive cover. The plate was placed in the heating block and the following thermal profile was run: 50°C for 2 min, 95°C for 10 min, 95°C for 15 sec followed by 60°C for 1 min.
Cell viability and cell cycle analysis
HT115 cells were harvested and seeded into 25 cm2 flasks at a density of 3 × 105 cells 48 hr before treatment with olive oil phenolic extract at various concentrations (0, 25, 50, 75, 100 μg/ml) in duplicate. Media containing the olive oil phenolic extract was added and the cells incubated for 24 hr at 37°C/5% CO2. Cells were then washed with PBS and harvested by the addition of 1 ml trypsin EDTA and incubation at 37°C for up to 5 min. Two millilitres of serum-containing medium was added to inactivate the trypsin. At this stage, a cell and viability count was carried out using a haemocytometer and trypan blue dye (Sigma). The cells were transferred to polypropylene tubes (Becton Dickinson) and centrifuged at 258g for 3 min. The supernatant was carefully discarded and the pellet re-suspended in 200 μl of ice cold PBS and 2 ml of 70% ethanol/30% PBS. Cells were incubated on ice for 30 min, centrifuged at 258g for 3 min, the supernatant carefully discarded, and the cells re-suspended in 800 μl of ice cold PBS, 100 μl of RNase A (1 mg/ml) (Sigma Chemicals), and 100 μl of propidium iodide (400 μg/ml) (Sigma Chemicals). The cells were then incubated at 37°C for 30 min before being analysed. Samples were processed on a FACSCalibur flow cytometer (Becton Dickinson) equipped with a laser (excitation wave length = 488 nm), and the fluorescence emission spectra of propidium iodide was collected at 585 nm (designated FL2 on FACSCalibur), using CellQuest Software (Becton Dickinson). A total of 10,000 events were captured. These emission spectra were analysed subsequently for DNA content using ModFit LT Software (Verity Software House, Topsham, ME). Each experiment was run in duplicate and the complete data set is the mean of 3 independent experiments. The complete dataset was analysed using ANOVA carried out in SPSS for Windows.
HT115 cells were harvested and seeded into 25 cm2 flasks at a density of 9 × 105 cells with media containing olive oil phenol extract at various concentrations (0, 25, 50, 75, 100 μg/ml) and then incubated for 24 hr at 37°C/5% CO2. Cells were then harvested and counted as described above. Each experiment was run in duplicate and the complete data set was the mean of 3 independent experiments.
The effects of a range of olive oil phenolic concentrations (0, 5, 10, 25, 50, 75, 100 μg/ml) on DNA (single strand breaks) damage in HT29 cells are shown in Figure 1. No significant genotoxic activity was observed for any concentration of olive oil phenols pre-incubated with HT29 cells (24 hr). No significant anti-genotoxic effect was observed for hydrogen peroxide challenged cells when treated with olive oil phenols. A significant negative, linear trend was observed for anti-genotoxicity with increasing olive oil phenolic concentration for DNA damage (linear regression, p = 0.011; R2 = 0.23). This demonstrates that olive oil phenols exerted a protective effect against oxidative insult. At 50 μM there is a maximal reduction in DNA damage of approximately 25%. The biological significance of this reduction is hard to speculate on, however, we have shown anti-genotoxic activity for various compounds including probiotic fermentations and cruciferous sprouts on HT29 cells ranging from 30–60%. The anti-genotoxic activity observed is approaching ranges recorded previously.24, 40
The percentage response of trans-epithelial resistance (barrier function) to various concentrations of olive oil phenols (10, 50, 100 μg/ml) over time (0, 24, 48 hr) is displayed in Figure 2. The addition of 10 μg/ml olive oil phenolic extract failed to improve barrier function in the CACO2 monolayer with time, the values observed were not significantly different from control values at any time point. With 50 and 100 μg/ml olive oil phenolic extract no significant difference was observed from control values at the 24-hr time point. The addition of 50 and 100 μg/ml olive oil phenols extract to the CACO2 monolayers did significantly increase barrier function compared to controls at the 48-hr time point with mean differences of 23.62% ± 6.57 (p = 0.004) and 26.2% ± 6.57 (p = 0.002), respectively. The improvement in barrier function seems to be a saturation event because 50 and 100 μg/ml extracts elicited approximately the same increase in CACO2 barrier function (38 and 41%, respectively).
The data shown in Figure 3 is the effect of a range of physiological olive oil phenols concentrations on % HT115 cell invasion rates. At the lowest concentration used (5 μg/ml) the addition of olive oil phenols extract did not have any significant effect on invasion rates as compared to control values. All remaining olive oil phenolic concentrations (25, 50, 75, 100 μg/ml) significantly inhibited HT115 invasion levels with significant mean differences (±STD error of mean difference) 10.68 ± 3.89 (p = 0.026), 12.66 ± 3.96 (p = 0.016), 19.75 ± 2.88 (p = 0.0095) and 20.06 ± 2.83(p = 0.0095) respectively compared to control values. An estimate of cell numbers showed a significant decrease in total cell count (attached cells) for concentrations above 50 μg/ml olive oil phenols (p = 0.000). To further confirm the anti-invasive effect of olive oil phenols we did a probit analysis and found that, conditional on cell numbers, the effect of olive oil phenols concentration on invasion was highly significant (coefficient/SE = −9.63, p < 0.001).
Olive oil phenolic extracts (0, 5, 20, 25 μg/ml) failed to affect gene expression (MMP-2,9; TIMP1,2,3) significantly at any of the concentrations tested (data not shown).
The olive oil phenolic extract had no significant effect on HT115 cell cycle kinetics or apoptotic fraction at any concentration (0, 25, 50, 75, 100 μg/ml) after 24 hr exposure.
Cell viability and attachment
The data shown in Figure 4 displays the effect of various olive oil phenolic concentrations (0, 25, 50, 75, 100 μg/ml) on the viability and attachment of HT115 cells after 24 hr exposure. The extract range tested had no significant effect on cytotoxicity as evidenced by the cell viability data, but clearly affected the ability of cells to adhere to the surface of the culture flasks. Significant impairment of attachment was observed at 75 and 100 μg/ml olive oil phenols (p = 0.011, p = 0.006, respectively).
Our study used a series of in vitro systems to model important stages of colon carcinogenesis including: (i) Initiation, events giving rise to DNA damage that commences the cancer process; (ii) Promotion, events that affect the colonic epithelium allow continuation of the cycle from polyp to carcinoma; and (iii) Metastasis, events that allow the carcinoma to invade other tissues ultimately causing mortality. Using this approach we have demonstrated that olive oil phenols extract can modulate in a beneficial manner all 3 critical stages of the carcinogenesis process. Within this in vitro study we have examined the possibility that luminal phenols can beneficially act on the colonic epithelium. This does not discount the possibility that in vivo the effect of olive oil phenols may be mediated by a blood borne mode rather than direct or some combination of both.
From our study, a significant linear trend was evident for anti-genotoxicity with increasing olive oil phenols concentration against a H2O2 challenge in HT29 human colon cells. Because the olive oil phenols extract was removed from the cells before exposure to the H2O2 challenge, the reduction (approximately 25%) in induced-DNA damage indicates an increased cellular capacity to protect against damage. Although no data exists regarding the effects of olive oil phenols on DNA damage, several studies examining anti-genotoxic effects polyphenols from other sources (e.g., tea, red wine) have demonstrated a protective effect.41, 42 In a recent study by Dhawan et al.43 black tea polyphenols prevented in a concentration-dependent manner (0.1–0.5 mg/ml), DNA damage elicited by the heterocyclic amine 3-amino-1-methyl-5H-pyrido (4,3-b) indole (TRP-P-2) in human lymphocytes. A similar in vitro study in human lymphocytes demonstrated that a number of polyphenols including quercetin, resveratrol and curcumin (3.1–25 mM) protected against DNA damage induced by H2O2.44 Quercetin and other dietary polyphenols have also been shown to reduce DNA damage induced by H2O2 in the CACO2 colonocyte cell line.45 Our findings seem to be consistent with the effects reported for polyphenols in general.
Our findings clearly show that olive oil phenols extract (50 μg/ml and 100 μg/ml) significantly increased barrier function after 48 hr exposure (CACO2 monolayer) by approximately 25% compared to untreated cells. Epithelial tissues act as barriers between two fluid compartments; barrier function arises from the epithelial cells and the tight junctions (TJ) that connect them. Mullin et al.46 demonstrated that treatment of a cultured monolayer with a tumour promoter (phorbol ester) induced increased trans-epithelial permeability and consequently decreased barrier function resulting in tumour-like polyps. A later study by Soler et al.47 reported that increased TJ permeability of the colon epithelium and consequently a decrease in epithelial barrier function preceded the development of colon tumours. The ability for olive oil phenols to elevate barrier function suggests that it may be able to exert an anti-promoter effect in the carcinogenesis pathway. No data is available on the effects of polyphenols from any source on barrier function. The ability of dietary compounds to enhance barrier function has been reported, however, the putative anticancer poly unsaturated fatty acid γ linoleic acid was demonstrated to increase trans-endothelial resistance.30
A loss of TJ function is also thought to be an early aspect of cancer metastasis as reviewed by Martin et al.48 Phorbol esters are known tumour promoters but have also been reported to increase invasiveness of human colon cancer cell lines.49
It was apparent from our study that the olive oil phenols extract (25–100 μg/ml) significantly decreased the invasiveness of HT115 colon cancer cells. Only the 25 μg/ml olive oil phenols extract, which decreased invasion level by approximately 60%, did so without significantly affecting total cell numbers. The anti-invasive effect was further confirmed by the probit analysis that found that, conditional on cell numbers, the effect of olive oil phenols concentration on invasion produced a highly significant negative linear trend. Although to the best of our knowledge this is the first study to demonstrate anti-invasive effects of olive oil phenols in vitro, our findings are consistent with the anti-invasive effects observed from polyphenols on a range of cell types. A recent study by Kim et al.50 showed that fermented pomegranate juice polyphenols caused 75% inhibition of invasion in breast cancer cells at a concentration of 10 μg/ml. The green tea polyphenol epigallocatechin-3-gallate (EGCG) has been shown to inhibit invasion of pancreatic carcinoma cells (100 μM/ml), human biliary tract carcinoma cells (100 μM/ml)51, 52 and also human umbilical vein endothelial cells (100–200 μg/ml).53
The inhibition of invasion observed in our study was not related to changes in expression of MMP or TIMP in the HT115 cells as determined by real time PCR. A few studies have demonstrated the inhibitory effects of a small number of green tea polyphenol including epicatechin gallate and EGCG on MMP activity.53, 54, 55, 56 Little information exists on the effect of polyphenols on MMP gene expression. Annabi et al.57 reported a decrease in MT1 MMP transcripts in bone marrow derived stromal cells treated with EGCG. Although a recent animal study by Adhami et al.58 demonstrated inhibition of MMP-2 and MMP-9 in response to green tea phenolic extract. In our study no variation in gene expression was observed. MMP are not subject to posttranslational modification suggesting that our olive oil phenolic extract may exert its effect on invasion via another pathway or through another MMP family member. For example 12-o-tetradecanoylphorbol-13-acetate (TPA) enhanced invasion of human colon adenocarcinoma cells (RCM-1 L-10) without altering metalloproteinase activity.59 The increased invasion apparently being mediated through protein kinase C.60 There still remains the possibility that olive oil phenols may act directly on MMP activity, but this was beyond the scope of our current study. As cancer cells undergo metastasis, invasion, and migration of a new tissue, they penetrate and attach to the target tissue's basal matrix. This allows the cancer cell intravasate into the tissue. The attachment is mediated by cell-surface receptors known as integrins, which bind to components of the extracellular matrix (ECM). Integrins are crucial for cell invasion and migration, not only for physically tethering cells to the matrix, but also for sending and receiving molecular signals that regulate these processes (reviewed by Hood and Cheresh).61 The olive oil phenols extract did not alter cell viability or the apoptotic fraction, but it was apparent that olive oil phenols interfered with the attachment of cells to surfaces in culture flasks and this anti-attachment effect was more pronounced when cells were cultured on an ECM like surface (matrigel). Given that the extracellular portion of integrins binds to various types of ECM proteins such as laminins, collagens, fibronectin (the main components of matrigel) it seems reasonable to speculate that the pronounced detachment effect observed may in part be due to olive oil phenols effecting integrin mediated attachment. Further studies would be needed to determine if this was a direct binding effect or alterations in the expression or distribution of the various integrin proteins.
Olive oil is suggested to be responsible in part for the beneficial nature of the ‘Mediterranean diet’ and our data supports this view and provides some possible mechanisms for its action. We have demonstrated that a phenol mixture extracted from virgin olive oil is capable of inhibiting, at least in vitro, multiple key stages in the colon carcinogenesis pathway including initiation, promotion and metastasis. The next stage would be to assess the effects in a suitable animal model.
A. Boyd is in receipt of a DEL award (N.I.). M. McCann is in receipt of a VCRS award from the University of Ulster.
- 35Enterocyte-like differentiation and polarization of the human colon carcinoma cell line Caco-2 in culture. Biol Cell 1983; 47: 323–30., , , , , , , , , , .
- 39Association of the HGF/SF receptor, c-met, with the cell-surface adhesion molecule, E-cadherin, and catenins in human tumor cells. Biochem Biophys Res Comm 2: 406–11., .
- 40The effect of cruciferous and leguminous sprouts on genotoxicity, in vitro and in vivo. Cancer Epidemiol Biomark Prev 2004; 13: 1999–205., , , , , , , .
- 59TPA-enhanced invasion of Matrigel associated with augmentation of cell motility but not metalloproteinase activity in a highly metastatic variant (L-10) of human rectal adenocarcinoma cell line RCM-1. Int J Cancer 1993; 55: 974–81., , , , , , .
- 60TPA-enhanced motility and invasion in a highly metastatic variant (L-10) of human rectal adenocarcinoma cell line RCM-1: selective role of PKC-alpha and its inhibition by a combination of PDBu-induced PKC downregulation and antisense oligonucleotides treatment. Clin Exp Metastasis 1999; 17: 351–60., , , .