The Mediterranean diet includes a variety of food patterns from different regions of the Mediterranean, and it is characterised by the consumption of an abundance of plant foods (fruit, vegetables, cereals, grains and nuts), dairy products, fish and olive oil as the principal source of fat(Reference Willett, Sacks and Trichopoulou1). This diet has been traditionally linked to the longevity of populations from the Mediterranean countries, and it has been associated with a significant improvement in health status, as seen by a reduction in overall mortality, mortality from CVD, incidence of or mortality from cancer, and incidence of other chronic diseases(Reference Sofi, Cesari and Abbate2). Olive oil, as a major energy source in the Mediterranean diet, is being investigated for its contribution to the reported health benefits of this diet. The main active components of olive oil include monounsaturated lipids (especially oleic acid), phenolic constituents (such as hydroxytyrosol, tyrosol and oleuropein) and squalene(Reference Hashim, Eng and Gill3, Reference Owen, Giacosa and Hull4).
Cancer is a major health problem and a leading cause of mortality worldwide. The most commonly diagnosed cancers in men are lung (16·5 % of the total), prostate (13·8 %) and colorectum (10·0 %), whereas in women they are breast (22·9 %), colorectum (9·4 %) and cervix uteri (8·8 %)(Reference Ferlay, Shin and Bray5). Geographical differences in cancer incidence rates indicate a role of environmental factors in the aetiology of this disease, nutrition being one of the most relevant. Epidemiological and especially experimental studies have found a relationship between dietary lipids and some cancers, such as breast and colorectum(6, Reference Escrich, Solanas and Moral7). Human data have also shown that women in the Mediterranean European countries have a lower incidence of breast cancer than women in other Western countries(Reference Trichopoulou, Lagiou and Kuper8), and recent prospective studies have associated the Mediterranean diet with a reduction of breast cancer risk(Reference Cottet, Touvier and Fournier9, Reference Trichopoulou, Bamia and Lagiou10). In fact, it has been proposed that up to 25 % of colorectal, 15 % of breast and 10 % of prostate, pancreas and endometrial cancers could be prevented by adhering to this diet(Reference Trichopoulou, Lagiou and Kuper8).
Thus, because strong evidence suggests that consuming the Mediterranean diet may have a protective effect, the aim of the present paper is to analyse the influence of the main source of fat of this diet (olive oil) on breast cancer and the mechanisms involved in this effect.
Effects of extra virgin olive oil on experimental breast cancer: clinical parameters
The relationship between dietetic factors, especially dietary lipids, and breast cancer has long been analyzsd. Although experimental studies in animals have clearly shown an influence of dietary lipids on breast cancer(Reference Escrich, Solanas and Moral7, Reference Wynder, Cohen and Muscat11–Reference Bartsch, Nair and Owen13), heterogeneous human epidemiological results have been reported. While cohort studies have generated conflicting results, case–control analyses and new prospective studies have shown that high fat intake is an important modulator of breast cancer risk(Reference Bartsch, Nair and Owen13–Reference Schulz, Hoffmann and Weikert16). The main source of fat in the Mediterranean diet is olive oil, but the effects of this oil on breast cancer risk are not well elucidated. Moreover, although dietary habits in the Mediterranean countries have been traditionally linked to the high consumption of olive oil, in recent years the consumption of seed oils rich in n-6 PUFA – such as sunflower or corn oils – and other oils has increased. Thus, more scientific evidence is needed in relation to the effects of olive oil and high-n-6 PUFA oils on breast cancer. Owing to the difficulty in obtaining data with controlled variables in humans, animal models are widely used to gain insight into the influence of dietetic factors on health. For more than 20 years, we have been using the experimental model of breast cancer chemically induced with 7,12-dimethylbenz(α)anthracene (DMBA) in the female Sprague-Dawley rat(Reference Escrich17). The suitability of this model for the possible application of the results in human breast cancer was previously validated, and it is used extensively in breast cancer studies(Reference Escrich18, Reference Russo, Gusterson and Rogers19). For the investigation of the different effects that olive oil and n-6 lipids may have on the initiation and promotion of mammary adenocarcinomas, a low-fat diet (with 3 % of corn oil), a high-fat n-6 diet (with 20 % of corn oil) and high-fat n-9 (with a 17 % of extra virgin olive oil (EVOO) and 3 % of corn oil) diets have been designed and validated(Reference Escrich, Solanas and Segura20–Reference Solanas, Hurtado and Costa22).
To study the effects of dietary lipids on experimental breast cancer, several clinical parameters have been defined, in addition to suitable statistical analysis(Reference Escrich, Ribalta and Muntané23–Reference Goncalves, Cabral and Ruiz de Villa26). In sixteen experimental series developed in our laboratory, results have always been conclusive: diets with a high content of n-6 PUFA (corn oil) exert a clear stimulating effect of the experimental mammary carcinogenesis in all the clinical parameters studied(Reference Solanas, Hurtado and Costa22, Reference Solanas, Moral and Escrich27–Reference Moral, Escrich and Solanas31) (Fig. 1). This stimulating influence has been mainly observed in the promotion stage of carcinogenesis, although an influence on the initiation stage cannot be ruled out. In contrast, the diet high in EVOO, with the same total content in fat as the high-corn oil (HCO) diet, has noticeably shown a different effect (Fig. 1). Thus, depending on the parameter studied, the high-olive oil diet exerted a weak stimulatory effect to a protective effect on mammary carcinogenesis. In all the experimental studies developed in our laboratory, the high-EVOO diet, in comparison with the HCO diet, consistently increased tumour latency (retarding the appearance of tumours) and decreased tumour incidence, multiplicity and volume(Reference Escrich, Solanas and Moral7, Reference Solanas, Hurtado and Costa22, Reference Moral, Solanas and García30–Reference Escrich, Solanas and Moral33). Moreover, in some of the experimental series, we have observed that the olive oil diet reduced tumour content and volume, even in comparison with a control low-fat diet, which suggested a protective effect of the olive oil on mammary carcinogenesis(Reference Solanas, Hurtado and Costa22, Reference Escrich, Solanas and Moral33). The study of tumour regression, as a proposed protective effect of this high-olive oil diet, showed a slow progression of tumours in the animals fed that diet, rather than a real regression(Reference Solanas, Hurtado and Costa22). The differences observed in the experimental series can be related to the different timing of the dietary intervention, as well as to distinct varieties of oil used, as it has been reported that the influence of olive oil on carcinogenesis depends on the proportions of oleic acid and minor components of the oil(Reference Cohen, Epstein and Pittman34). Such findings are of interest considering the high percentage of olive oil used in this experimental diet, as it should be taken into consideration that all high-fat diets may have an unspecific stimulatory effect on carcinogenesis(Reference Guthrie and Carroll35). In this sense, a positive association between the dietary energy supply and cancer mortality rates has been shown, and energy restriction has an indiscriminate inhibitory effect on carcinogenesis(Reference Kritchevsky36). In the literature, experimental studies on the effects of olive oil in the diet have reported inconsistent results(Reference Escrich, Solanas and Moral7). Administration of diets with a lower percentage of oil (7 % and 15 %) during prenatal life has significantly increased the number of tumour-free rats(Reference Kossoy, Stark and Tendler37).
Fig. 1 Effect of an HEVOO diet and an HCO diet on body weight and mass of female rats at 235–254 d of age; (a) body weight (g); (b) BMI (g/cm2). HEVOO, high-extra virgin olive oil group; HCO, high-corn oil group; LF, low-fat group. Values are mean and se. *P < 0·05 compared to control LF group (Tukey's test; n 20/group)
Effects of extra virgin olive oil on experimental breast cancer: histopathological characteristics
Our group has also characterised morphologically, for the first time according to our knowledge, the effect that dietary lipids may exert on experimental breast cancer. Previously, a histological grading system adapted to rat mammary carcinomas was developed(Reference Costa, Esquius and Solanas38, Reference Costa, Solanas and Escrich39). These studies first showed the correspondence between the histological pattern and the clinical characteristics of each tumour, finding a higher degree of biological aggressiveness in tumours with higher histopathological degrees, desmoplastic reaction, limfoplasmocitic infiltrate, tumour necrosis and prevalent cribiform architectural patterns(Reference Costa, Esquius and Solanas38, Reference Costa, Solanas and Escrich39). Furthermore, the analysis of the effects of different diets showed that the HCO diet promoted adenocarcinomas with a high histological degree, stromal invasion, more prominent tumour necrosis and a frequent cribiform pattern in comparison with the controls and high-olive oil diet. In contrast, animals fed a diet rich in EVOO developed adenocarcinomas with a low histopathological grade, few invasive and necrotic areas (similar to the tumours from the control group) and extensive papillary areas(Reference Costa, Moral and Solanas40). In conclusion, tumours from the animals fed the high-EVOO diet clearly showed a lower degree of tumour malignancy than those from the animals fed the HCO diet, being more similar to control tumours. All these results are compatible with the lower degree of clinical malignancy and with the non-promoting effect of a diet rich in EVOO(Reference Costa, Moral and Solanas40).
Mechanisms of action of extra virgin olive oil on breast cancer: growth and sexual maturation
The mammary gland has the particularity that, unlike other organs, after birth it remains highly undifferentiated until the onset of puberty. Reproductive events increasing the number of menstrual cycles and therefore lifetime exposure to oestrogens, such as early menarche, have been associated with increased risk of breast cancer(6). Thus, modifications in growth and sexual maturation due to the effect of nutrition may modify the susceptibility to mammary transformation, advancing, retarding or expanding the windows of this susceptibility.
We observed that a diet high in corn oil increased the body weight of the rats as well as the BMI. In contrast, the high-olive oil diet, with the same percentage of fat as the HCO diet, did not modify body weight or mass in relation to the low-fat diet(Reference Moral, Escrich and Solanas31) (Fig. 2). Several studies have also reported an effect of the high-n-6 PUFA diet on body weight, but there are few experimental data regarding the effects of the olive oil diet. In this sense, a 30 % EVOO diet has also been shown to produce lower bodyweight gain in rats when compared to a 30 % corn oil diet(Reference Oi-Kano, Kawada and Watanabe41). On the other hand, there are human epidemiological data suggesting that the Mediterranean diet may have a protective effect on obesity(Reference Buckland, Bach and Serra-Majem42–Reference Perez-Lopez, Chedraui and Haya45).
Fig. 2 Effects of an HEVOO diet and an HCO diet on rat DMBA-induced mammary carcinogenesis in two different experimental series. Tumour-free animals: percentage of rats without palpable tumours. Experimental series 1: animals were fed a low-fat diet (3 % corn oil; w/w), an HCO diet (20 % corn oil) or an HEVOO (3 % corn oil + 17 % olive oil) diet after the induction with the carcinogen (promotion phase of carcinogenesis (P)). Experimental series 2: animals were fed the low fat, HCO or HEVOO diet from weaning (initiation and promotion phase of the carcinogenesis (IP)) or after the induction with the carcinogen (only promotion phase P). HEVOO, high-extra virgin olive oil group; HCO, high-corn oil group; DMBA, 7,12-dimethylbenz(α)anthracene; LF, groups fed the low-fat diet; HCOIP, group fed the HCO diet from weaning (initiation + promotion); HCOP, groups fed the HCO diet from induction (promotion); HEVOOIP, group fed the HEVOO diet from weaning (initiation + promotion); HEVOOP, groups fed the HEVOO diet from induction (promotion)
Sexual maturation has also been studied in rats fed these high-fat diets. Although a high amount of fat in the diet generally advances the onset of puberty, we observed different effects depending on the type of oil consumed. The onset of puberty in animals fed the HCO diet was significantly advanced, followed by animals fed a high-EVOO diet, and the control animals(Reference Moral, Escrich and Solanas31) (Table 1). No differences were found either in body weight or mass when arriving at each stage, thus suggesting that the reproductive morphological changes are related to the acquisition of a threshold level of body mass, as has been already reported in humans(Reference Vizmanos and Martí-Henneberg46). This threshold weight would appear earlier in the animals of the HCO group, followed by the high-EVOO group and the controls. These results were in accordance with the study of a marker of puberty, the hypothalamic expression of kisspeptin(Reference Castellano, Roa and Luque47). The mRNA levels of kisspeptin in the hypothalamus of rats fed the HCO diet suggested higher expression of this marker at around puberty, while no differences were observed in the animals fed the EVOO diet in relation to the controls. No differences were observed in other parameters of maturation due to the effect of diets (uterine weight, time span among stages, time from vaginal opening to first oestrus or cycle), which also suggested that high-fat diets may advance the process of maturation but did not change its evolution(Reference Moral, Escrich and Solanas31).
Table 1 Effect of an LF, HCO and an HEVOO diet on day of arrival at different states of maturation in female rats
LF, low fat; HCO, high-corn oil; HEVOO, high-extra virgin olive oil.
*P < 0·05 compared to control LF group.
†P < 0·05 compared to HEVOO group (parametric Tukey's test for mean, non-parametric Mann-Withnney's U test for median).
Mechanisms of action of extra virgin olive oil on breast cancer: hormonal status
Breast cancer is an oestrogen-dependent neoplasia, and thus may be influenced by modifications in hormonal status(6). Thus, the possible effect of dietary lipids on serum concentrations of luteinizing hormone, follicle-stimulating hormone, oestradiol, progesterone, prolactin, insulin and corticosterone has been analysed, but no significant modifications due to the effect of high-olive oil or HCO diets(Reference Escrich48) (E Escrich, unpublished results) have been found. In the literature, there are discordant results regarding the influence of dietary lipids on hormone levels. For example, pregnant rats fed with a high-n-6 PUFA diet have been shown to either increase(Reference Hilakivi-Clarke, Clarke and Onojafe49) or not modify(Reference de Assis, Khan and Hilakivi-Clarke50) serum oestradiol levels. Other authors have found lower concentrations of oestradiol in lactating dams fed a 7 % olive oil diet in comparison to those fed a 7 % corn oil diet(Reference Stark, Kossoy and Zusman51). The discrepancy among results may be related to the cycling nature of hormones, as their serum concentrations are highly dependent on the hour of the day and the phase of the oestrous cycle(Reference Sylvester, Russell and Ip52).
To better understand the effects that dietary lipids may have on hormonal status, we also analysed the expression levels of hormone receptors in the mammary gland and in the experimental tumours, finding little difference due to the effect of the high-fat diets. Thus, analyses of the expression of estrogen receptors (ERα, ERβ1 and ERβ2) showed no significant differences due to the effect of diets. We observed higher levels of progesterone receptors in the mammary gland of rats fed the EVOO diet in relation to the animals fed the HCO diet at the age around puberty, which may be related to the development of lobulo-alveolar structures(Reference Moral, Escrich and Solanas31). The few changes observed suggest that the modulation of the expression levels of these hormone receptors are not a major mechanism by which these high-fat diets may influence mammary gland biology.
Mechanisms of action of extra virgin olive oil on breast cancer: morphology and differentiation of the mammary gland
The mammary gland is a target of sexual maturity. Various studies on animals have shown that the risk of malignant transformation of this tissue can be highly dependent on the influence of early life events, and highlight the importance of the degree of differentiation at the time of a carcinogenic insult(Reference Russo and Russo53). Thus, we characterised the differentiation degree of the mammary gland by quantifying the different epithelial structures in the periphery of the gland. Other authors have suggested that high-fat diets, such as HCO diet, may influence mammary carcinogenesis by increasing the number of the epithelial structures that are target of the carcinogen (the undifferentiated terminal end buds (TEB))(Reference Hilakivi-Clarke, Clarke and Onojafe49, Reference Russo and Russo53). However, we found little influence from either of the high-fat diets on the number of TEB, terminal buds, alveolar buds or lobular structures of the breast tissue(Reference Moral, Escrich and Solanas31).
To further characterise the degree of differentiation of the mammary gland, mRNA expression of β-casein, classically considered a molecular differentiation marker of this gland, has also been analysed. β-Casein expression increased in the mammary tissue over time, but no significant differences were found due to the effect of dietary lipids. Although β-casein mRNA levels inversely correlated with the number of undifferentiated terminal ducts, we did not find a correlation with TEB, the main target of carcinogens(Reference Russo and Russo53), or, therefore, with the susceptibility of the mammary gland to transformation.
Mechanisms of action of extra virgin olive oil on breast cancer: molecular changes in tumours
The molecular mechanisms by which dietary lipids may exert their effects on the development of mammary cancer are not well understood. Experimental studies support the hypothesis of the existence of a specific effect aside from the unspecific one derived from the energy supply(Reference Guthrie and Carroll35). Some of the proposed mechanisms, to which we have contributed with experimental data, include modifications of the structure and function of cell membranes, modulation of gene expression or influence on cell signalling pathways(Reference Escrich, Moral and Grau54). Moreover, several studies have reported beneficial effects of olive oil through other mechanisms, such as modifications in oxidative stress and changes in immune system function(Reference Escrich, Solanas and Moral7, Reference Escrich, Moral and Grau54, Reference Ortega55). Lipids could act through several of these and other mechanisms in an integrated, simultaneous and/or sequential way.
Modifications of cell membranes
Lipids are an essential component of cell membranes and thus regulate their biological activity. Membrane lipid composition may be altered by dietary lipids, mainly by n-3 PUFA and the n-3/n-6 ratio of the diet(Reference Hulbert, Turner and Storlien56). Few studies have been carried out regarding the influence of olive oil on membrane composition, but there is evidence that adherence to the Mediterranean diet may affect the structural properties of the erythrocyte cell membrane of hypertensive patients. This influence has been related to the beneficial effect of the Mediterranean diet on hypertension(Reference Barceló, Perona and Prades57). A high content of polyunsaturated lipids in membranes increases the fluidity and the susceptibility to peroxidation(Reference Hulbert, Turner and Storlien56). We have studied the membrane composition of experimental tumours by analysing fourteen different fatty acids in six lipidic fractions. A high-n-6 PUFA diet changed the tumour lipid profile, increasing the 18 : 2n-6 relative content and decreasing that of the 18 : 1n-9 (significantly in three lipidic fractions: phosphatidylcholine, free fatty acids and TAG)(Reference Escrich, Solanas and Soler28). Taking into consideration that those tumours had more aggressive clinical and histopathological behaviour, these results would be in accordance with the lower effect of MUFA on membrane fluidity and the lower susceptibility of these lipids to peroxidation, thus decreasing the oxidative damage in membranes.
Effects on gene expression
It has long been known that different dietary components can modulate the expression of specific genes. Although data regarding the effect of dietary lipids in genes with a role in cancer are scarce, there is a wealth of information reporting the effects of such dietary compounds on genes of the metabolism(Reference Jump58). Moreover, the effects of dietary lipids could be modified by the disease, as we have observed changes in the normal regulation of the expression of CPT-I, HMGCoA synthase and PPARα by high-fat diets in the liver of tumour-bearing rats(Reference Moral, Solanas and Manzanares59).
The healthy benefits of the Mediterranean diet have been partly associated with changes in the expression of atherosclerosis-related genes through olive oil polyphenols(Reference Konstantinidou, Covas and Muñoz-Aguayo60). Thus, olive oil may exert some of its influence on breast cancer through the modulation of expression of genes involved in the carcinogenesis process. We found in experimental mammary tumours that the high-EVOO diet and the HCO diet had different influences on the modulation of the expression of the ErbB family of membrane receptors, especially c-erbB1. Hence, the HCO diet increased the ratio between the 9·5 kb mRNA of EGFR (coding the functional full-length receptor) and the 2·7 kb mRNA (coding an inactive truncated receptor), whereas the high-EVOO diet decreased this ratio in the mammary tumours(Reference Moral, Solanas and García30). On the other hand, the expression levels of p21Ras, a key transductor of ErbB proliferative signalling that is frequently involved in human mammary carcinogenesis(Reference Malaney and Daly61), were not modified by these dietary lipids(Reference Solanas, Moral and Escrich29). High-fat diets did not modify either the expression levels of HMG-CoA reductase or squalene synthase, two genes that codify enzymes of the mevalonate pathway(Reference Solanas, Grau and Moral62). Mevalonate is the source of the prenyl groups needed for the post-translational modification of p21Ras and its subsequent activation(Reference Singh, Kumar and Kapur63).
Cell dedifferentiation is part of the alterations linked to cancer, and thus our group studied whether a high-EVOO diet and a HCO diet changed the degree of molecular differentiation of DMBA-induced mammary tumours(Reference Moral, Solanas and Garcia64). We analysed the expression of known mammary differentiation markers, α-casein, β-casein and transferrin, in addition to β-actin and its transporter protein ZBP1. In the mammary adenocarcinomas, the expression levels of α-casein, β-casein and transferrin were not related to the degree of morphological differentiation or to the clinical behaviour of the tumours, which suggested that these genes were not good biomarkers of the modifications that the experimental diets conferred to the adenocarcinomas. The HCO diet, but not the high-EVOO diet, increased the β-actin mRNA levels but not those of the protein. This last result, in addition to the increase observed in the transporter ZBP1 as a result of the HCO diet, suggested a deregulated transport and translation of β-actin that was associated with the most malignant phenotype of the tumours from animals fed a diet rich in corn oil, but not with those of the animals fed the high-EVOO diet(Reference Moral, Solanas and Garcia64).
New screening technologies allowed different experimental approaches in order to gain insight into the modulation of gene expression by dietary lipids. cDNA microarrays analyses were used to discover four genes significantly downregulated due to the effect of the HCO diet in tumours(Reference Escrich, Moral and García65) but not modified by the high-EVOO diet(Reference Solanas, Moral and Garcia66). The genes differentially modulated in the adenocarcinomas of animals fed a diet rich in n-6 PUFA were α2u-globulin, VDUP1, the imprinted gene H19 and an unknown gene (Fig. 3). Experimental evidence suggests that these known genes are potentially related to differentiation and cell proliferation, thus being involved in the tumour-stimulating effect of the HCO diet(Reference Escrich, Moral and García65). Moreover, the IGFII gene, related to proliferation and reciprocally imprinted with H19, was upregulated by the HCO diet and downregulated by the high-EVOO diet. Furthermore, both high-fat diets had an opposing effect on the activity of thioredoxin, an oxidoreductase inhibited by VDUP1 that promotes cell growth and has an antiapoptotic action. This result was in accordance with the opposing effect of the HCO and EVOO diets on mammary carcinogenesis(Reference Escrich, Moral and García65, Reference Solanas, Moral and Garcia66).
Fig. 3 (colour online) Molecular mechanisms of the effects of an HEVOO diet and an HCO diet on DMBA-induced tumour malignancy. Animals fed the HCO diet displayed tumours with a higher histopathological degree of tumour aggressiveness, while animals fed the HEVOO diet displayed tumours similar to those of the animals fed a low-fat diet (LF). The HCO diet downregulated genes with a role in cell differentiation. The HEVOO diet downmodulated the Ras/Akt signalling pathway, which may elicit lower proliferation and higher apoptosis. Moreover, caspase 3 (casp 3) was upregulated in tumours from animals fed the HEVOO, thus suggesting a proliferation--apoptosis balance shifted to apoptosis. Furthermore, the tumours from the HEVOO group showed lower levels of ubiquitinated PCNA, which is related to DNA damage, suggesting lower levels of this damage due to the effect of the EVOO diet (HEVOO, high-extra virgin olive oil group; HCO, high-corn oil group; DMBA, 7,12-dimethylbenz(α)anthracene; proliferating cell nuclear antigen (PCNA))
Influence on signalling pathways
Our investigations have also focused on the effects that olive oil diet may have on signalling pathways with an important role in mammary development and human breast cancer, such as the ErbB-Ras and their main effectors (Erk1/2, PI3K/Akt, RalGDS/Ral)(Reference Malaney and Daly61). The protein levels and activity of the ErbB1, ErbB2 and ErbB3 were not altered due to the HCO diet or the high-olive oil diet, but we observed a significantly decreased expression in the 80 kDa ErbB4-truncated protein due to the effect of the high-EVOO diet(Reference Solanas, Grau and Moral62).
We next investigated the effect of the olive oil diet on p21Ras expression and activity. Animals fed this diet developed tumours with significantly higher levels of p21Ras protein, but with a significantly decreased activity (Fig. 3). In contrast, no changes were observed in p21Ras function as a result of the HCO diet, suggesting that p21Ras may have a role in the negative modulatory effect of olive oil on breast cancer, whereas it is not involved in the stimulatory effect of the HCO diet(Reference Solanas, Grau and Moral62). Several mechanisms of action for this decreased activity have been addressed: the tumour mutation status of Ha-ras1 and the expression of two key enzymes of the mevalonate pathway, the HMG-CoA reductase and squalene synthase (SQS). The experimental diets did not modify the rate of the activating point mutation in the codon 61 of the gene c-Ha-ras1, or the mRNA expression levels of HMG-CoA and SQS in the mammary adenocarcinomas (as mentioned above). Other experimental studies have suggested that an olive oil diet may have a protective effect on colon cancer through the effect of squalene, a minor compound of olive oil, inhibiting the HMG-CoA reductase(Reference Rao, Newmark and Reddy67). However, such effects have been observed at very high concentrations of squalene (1 % of diet), and our results did not support the hypothesis of a chemopreventive effect of olive oil diet through squalene on HMG-CoA reductase inhibition, at least at the expression level(Reference Solanas, Grau and Moral62).
The analysis of the main effectors of p21Ras showed increased ERK1/2 activity and reduced ERK1/2 expression due to the effect of both high-fat diets. These results suggested an unspecific effect of such diets upregulating the ERK1/2 proliferating pathway, which would not be elicited by ErbB receptors and could be mediated, at least partly, by a p21Ras-independent mechanism. No differences were found in the expression or activity of RalA and RalB proteins due to the effect of the experimental diets. On the other hand, the EVOO diet decreased both Akt expression and activation, whereas the HCO diet only decreased Akt protein levels(Reference Solanas, Grau and Moral62) (Fig. 3). A reduced activity of Akt has been associated with induction of apoptosis(Reference Franke, Hornik and Segev68); thus, these results suggested that the high-olive oil diet may downregulate the pro-survival p21/Akt pathway in the mammary tumours. Few data have been published regarding the effects of olive oil on these pathways, but it has been shown that in breast cancer cells oleic acid both increased(Reference Hardy, St-Onge and Joly69) and decreased(Reference Menendez, Vellon and Colomer70) active Akt, whereas hydroxytirosol abolished Akt phosphorylation and induced transient Erk1/2 phosphorylation-dephosphorylation in colon cancer cells(Reference Guichard, Pedruzzi and Fay71).
The influence of the high-olive oil diet on mammary tumours may thus result in upregulation of ERK1/2 pathway concomitantly with a downregulation in Akt signalling. There are experimental data in the literature suggesting that this relative balance between ERK1/2 and Akt pathways may result in pro-apoptotic signalling(Reference Chang, Hsu and Tsai72, Reference Karmakar, Banik and Ray73). To address the hypothesis that the EVOO diet induced cell apoptosis, we analysed the levels of the activated caspase-3, which is considered the main executioner of the caspase cascade involved in both extrinsic and intrinsic apoptosis(Reference Jin and El-Deiry74). The tumours from the animals fed the olive oil diet had significantly higher levels of this protein, which was concordant with the proposed pro-apoptotic effect of the olive oil(Reference Solanas, Grau and Moral62) (Fig. 3). These results are in line with other studies that have reported that olive oil, oleic acid and minor compounds of olive oil can modulate apoptosis(Reference Stark, Kossoy and Zusman51, Reference Han, Talorete and Yamada75–Reference Llor, Pons and Roca77).
Apoptosis is a crucial process in cancer development, and it has been suggested that tumour growth is partly determined by the balance between cell proliferation and apoptosis. The histopathological analysis of the mitotic activity and the protein expression levels of the mitotic marker proliferating cell nuclear antigen (PCNA) indicated that the high-EVOO diet did not exert a significant effect on tumour cell proliferation, in contrast to the HCO diet, which significantly increased tumour mitotic activity(Reference Costa, Moral and Solanas40, Reference Solanas, Grau and Moral62). This last result was consistent with an influence of HCO diet on the survival and proliferation signalling pathways and with its stimulatory effect on mammary carcinogenesis(Reference Escrich, Solanas and Moral7, Reference Solanas, Hurtado and Costa22, Reference Solanas, Moral and Escrich27–Reference Escrich, Solanas and Moral33). Moreover, the high-EVOO diet decreased the levels of monoubiquitylated PCNA, which is associated with DNA damage, in the mammary tumours(Reference Solanas, Grau and Moral62) (Fig. 3). The effect that olive oil may have on the proliferation of tumour cells is still not well elucidated. In vitro experiments have found evidence that some minor compounds of olive oil, such as hydroxytyrosol and oleuropein, inhibit proliferation of breast cancer cells(Reference Han, Talorete and Yamada75), while oleate has been reported to have a stimulatory effect on cell proliferation(Reference Hardy, Langelier and Prentki76).
Conclusions
There is increasing epidemiological and experimental evidence indicating the beneficial effect of the Mediterranean diet, and in particular of EVOO, on the progression of some cancers, especially breast cancer. This effect could be the result of its MUFA (oleic acid) and its minor compounds.
In an experimental model of mammary cancer, we have observed that a high-EVOO diet clearly has a different effect on breast carcinogenesis if compared with the effect of other diets (an n-6 PUFA enriched diet) with the same amount of fat, showing the importance of the type of oil consumed beyond its energy content. Some of the parameters that we have studied in relation to the clinical and histopathological behaviour of tumours, in addition to experimental data published in the literature using lower percentages of olive oil in the diet, suggest that EVOO may have a beneficial effect on breast cancer risk if its consumption is moderated. The mechanisms by which this type of oil may exert its effects are diverse, such as a protective effect on obesity or molecular influences eliciting a balance between proliferation and apoptosis shifted in favour of apoptosis or lower levels of DNA damage in tumours. These data may contribute to the published evidence of the healthy effect of the Mediterranean diet and highlight that the use of EVOO as the source of fat from childhood may be considered as a healthy choice.
Acknowledgements
The present study was supported by grants from the ‘Plan Nacional de I+D+I 2004–2007’ (AGL2006-07691/ALI); ‘Fundación Patrimonio Comunal Olivarero 2008–2012’ (FPCO2008-165·396); ‘Agencia para el Aceite de Oliva del Ministerio de Medio Ambiente y de Medio Rural y Marino 2008–2012’ (AAO2008-165·471); ‘Organización Interprofesional del Aceite de Oliva Español 2009–2013’ (OIP2009-165·646) and the ‘Departaments d'Agricultura, Alimentació i Acció Rural, i de Salut de la Generalitat de Catalunya’ (GC2010-165·000). The sponsors had no role in the study design, data collection and analysis, interpretation of results, the preparation of the manuscript, the decision to submit the manuscript for publication or the writing of the manuscript. E.E. conceived the study, its design, and coordinated the project. All authors are involved with the study design and review of the literature, data analysis and interpretation, made critical revisions of the paper, and provided approval for its publication. The authors thank Raquel Escrich and Irmgard Costa for their collaboration on these studies.
