Alcohol and Breast Cancer: Update 2012
Alcohol and Breast Cancer: Update 2012
The mechanisms by which alcohol stimulates breast carcinogenesis are still not understood. Major pathophysiologic research on the carcinogenic effect of chronic alcohol consumption has focused on the upper alimentary tract, the liver and the colorectum (Baan et al., 2007; Seitz and Stickel, 2007; IARC, 2010). Studies investigating mechanisms of ethanol-mediated breast cancer are rare and thus information is limited. Since a carcinogenic effect of estrogens on breast tissue has been observed, and since alcohol ingestion results in an elevation of serum estrogen concentrations, it is speculated that the carcinogenic effect of ethanol is mediated, at least in part, by estrogens (Fernandez, 2011). In addition, some research has concentrated on acetaldehyde (AA), the first and most toxic metabolite of ethanol oxidation which is by itself carcinogenic (Secretan et al., 2009; Seitz and Stickel, 2010) and on oxidative stress (Seitz and Stickel, 2006). The ethanol effect on epigenetic modifications (Stickel et al., 2006) and on interaction with retinoids (Wang and Seitz, 2004) has not been well studied in the breast when compared with other tissues. Alcohol-related carcinogenesis may also interact with other factors such as smoking, diets, comorbidities and depends on genetic susceptibility.
Only one study demonstrated that ethanol by itself (without the additional administration of a carcinogen) resulted in breast cancer. When 10 and 15% ethanol in the drinking water was given to female ICR mice for 25 months, 45% of the animals developed papillary and medullary adenocarcinomas of the breast (P = 0.0012), while no tumours were found in the control group (Watabiki et al., 2000). In addition, a number of animal studies have been performed with alcohol as a modifier of chemically induced mammary carcinogenesis by using either methylnitrosourea (MNU) or dimethylbenzanthracene (DMBA). When MNU (30 mg/kg) or DMBA (5 mg/rat) was used to induce mammary tumours in Sprague Dawley (SD) rats, ethanol as 10–30% of calories in the daily diet significantly increased the incidence of adenocarcinomas (Singletary et al., 1991, 1995), while with higher doses of DMBA (39 mg/kg) no effect on mammary tumourigenesis was seen (Singletary et al., 1991). In addition, other studies with 10 and 15 mg of DMBA and ethanol in the drinking water showed either an increased number of mammary tumours per rat (P < 0.006) (Rogers and Conner, 1990) or an increased tumour incidence (P < 0.001) (Hilakivi-Clarke et al., 2004) Furthermore, in utero exposure to ethanol-increased mammary tumourigenesis induced with DMBA in rats (McDermott et al., 1992). Recently, foetal alcohol exposure (6.7% ethanol, equivalent to 3–5 drinks in 2 h for a women) in SD rats from Day 11–21 of gestation decreased mammary tumour latency in adulthood induced by NMU (Polanco et al., 2010).
Another study investigated the interaction of estrogens and ethanol in ovariectomized mice on mammary tumours. While estrogens alone inhibited tumour growth in this model, the addition of ethanol increased insulin sensitivity and increased tumour growth in these obese mice (Hong et al., 2010).
Long-term exposure to estrogens increases the risk of developing breast cancer in women (Colditz, 1998). In rats, the continuous administration of supra-physiological doses of estrogens resulted in mammary adenocarcinomas, while low doses of estrogens administered over a long-time period lead to fibroadenomas (Russo and Russo, 1996).
The mechanisms by which estrogens induce breast cancer are still not completely clear, although it is well accepted that the binding of estrogens to its nuclear receptor, ER alpha (ERα), initiate a complex intracellular signal sequence, finally stimulating cell proliferation (Suga et al., 2007).
Alcohol-increased plasma estrogen levels had been significantly demonstrated in controlled feeding studies, with human female volunteers (Reichman et al., 1993; Ginsburg et al., 1996; Purohit, 1998; Sarkola et al., 1999, 2000, 2001; Coutelle et al., 2004; Mahabir et al., 2004, Sierksma et al., 2004; Seitz and Maurer, 2007). In postmenopausal women with hormone replacement therapy, 15 or 30 g of ethanol daily consumed over 8 weeks resulted in significantly elevated serum concentrations of oestrone sulphate and dehydroepiandrosterone sulphate (Dorgan et al., 2001). In premenopausal women, not only an acute high dose of ethanol (0.7 g/kg b.wt.) (Mendelson et al., 1988), but also a small dose (0.225 g/kg b.wt.) equivalent to one drink (10–12 g) led to significantly elevated plasma oestradiol levels. When 0.7 g ethanol/kg b.wt. was given, the observed effect was most pronounced during the ovulatory phase of the menstrual cycle and in women using oral contraceptives. With the low dose of ethanol, the elevation has been observed in all phases of the menstrual cycle, but seems to be especially relevant at mid-cycle where oestradiol concentrations are already high, being further stimulated by ethanol by ~27–38% (Coutelle et al., 2004). There is some evidence that alcohol consumption during adolescence and early adulthood may stimulate breast carcinogenesis more than drinking later in life (Longnecker et al., 1995).
In addition, alcohol at a concentration of 0.06% stimulates the expression of ERα and the oestradiol biosynthesis enzyme aromatase in human breast cancer cell lines (Gavaler and Van Thiel, 1992; Singletary and Gapstur, 2001; Etique et al., 2004). Alcohol stimulates ER signalling in human breast cancer cells (Fan et al., 2000). Alcohol-mediated ER-dependent gene expression may result in cell hyperproliferation in ER-positive Michigan Cancer Foundation-7 (MCF-7) human breast cancer cells (Ginsburg et al., 1996). It has also been reported that crosstalk between adenosine receptor (A2A isoform) and ERα mediates ethanol action in MCF-7 breast cancer cells, which may open new therapy strategies in oestrogen-dependent breast cancer (Etique et al., 2009).
The fact that several studies support the hypothesis that alcohol is more strongly related to ER positive than to ER negative breast tumours underlines the pathogenic role of estrogens in alcohol-mediated breast cancer.
Estrogens may also act directly as tumour initiators. They may be genotoxic both in vitro and in vivo and they may result in DNA damage including single-strand breaks and DNA adducts (Chakravarti et al., 2001; Liu and Lin, 2004). Catechol estrogen 3,4-quinones (an estrogen metabolite) generates DNA mutations initiating breast cancer (Cavalieri et al., 2004; Yager and Davidson, 2006). The fact that 'ERα knockout mice expressing the Wnt-1 oncogene (ERKO/Wnt-1) develop breast cancer when they receive estrogens' support the direct genotoxic effect of estrogens without ERα-mediated pathways (Bocchinfuso et al., 1999). Furthermore, the administration of estrogens to ovariectomized mice results in more tumours in a shorter period of time when compared with control animals even when anti-estrogens are given (Fernandez, 2011). Estrogen treatment of human breast epithelial cells induced gene expression alterations, epigenetic modifications and finally phenotypical changes indicating neoplastic transformation (Fernandez et al., 2005, 2006, 2010; Russo et al., 2006; Huang et al., 2007; Fernandez and Russo, 2010). Since these cells are ER negative, direct genotoxic effects of estrogens must be responsible for the changes observed.
Several mechanisms by which ethanol affects the levels of sex hormones in women have been suggested. These include an ethanol-mediated increase in the hepatic redox state resulting in a decrease of steroid metabolism (Sarkola et al., 1999, 2001). Another explanation for the increased estrogen concentrations after alcohol is the increased aromatase activity following chronic ethanol consumption, which leads to an enhanced conversion of testosterone to oestrogens with reduced testosterone and increased oestrogen concentrations (Gavaler and van Thiel, 1992). Alcohol also inhibits the activity of sulfotransferase and 2-hydroxylase, two enzymes important in estrogen degradation (Eagon, 2010). Alcohol may also decrease melatonin secretion which inhibits estrogen production (Stevens et al., 2000). Furthermore, alcohol may interact with the production of luteinizing hormone from the pituitary gland, which favours estradiol release from the ovaries (Rettori and McCann, 1997). An important study in which rats were exposed to ethanol in utero showed an increased number of terminal end buds, which are targets for malignant transformation, suggesting that alcohol when administered during foetal life may modify signals controlling mammary gland development (Hilakivi-Clarke et al., 2004).
Increased estrogen concentrations following ethanol ingestion may not only increase breast cancer risk, it may also explain at least in part the increased susceptibility of the female liver towards alcohol in the pathogenesis of alcoholic liver disease since estrogens may result in enhanced hepatic mitochondrial damage resulting in enhanced fat deposition, apoptosis and the generation of reactive oxygen species (ROS) (Eagon, 2010).
Alcohol is metabolized via alcoholdehydrogenase (ADH) to AA and further via acetaldehydedehydrogenase (ALDH) to acetate (Fig. 3). AA is toxic, mutagenic and carcinogenic in animal experiments. It binds rapidly to proteins and DNA, forming stable carcinogenic adducts. AA also decreases the anti-oxidative defence system and modifies indirectly epigenetic histone and DNA methylation by decreasing the availability of S-adenosinemethionine, the major methyl donator (for review, Seitz and Stickel, 2007). AA has therefore been classified as a carcinogen by the IARC (Secretan et al., 2009). Thus, an increased burden of AA by either enhanced generation or slow degradation may stimulate carcinogenesis. Indeed, this has been shown for other ethanol-mediated cancer sites (Seitz and Stickel, 2007). Approximately 40% of Japanese have a heterozygote mutation at the ALDH2 gene with a striking reduction of the ALDH2 enzyme activity resulting in a significant elevation AA level in blood and saliva (Vakevainen et al., 2000). Since this increase in AA is associated with side effects such as flushing, tachycardia, nausea, vomiting and sweating, alcohol consumption is reduced below three drinks in heterozygots or close to zero in homozygots. Only a few Japanese men still consume alcohol even when side effects occur. Owing to AA accumulation, these individuals have an increased risk for upper aerodigestive tract cancer (oropharynx, larynx and oesophagus) and for colorectal cancer (Yokoyama et al., 1998). AA may act as a carcinogen only when its concentration exerts mutagenic levels, e.g. only when sufficient alcohol is consumed to reach that level. Owing to the fact that Japanese women drink only small amounts of alcohol, and that Japanese women who are ALDH2-deficient tend not drink alcohol (due to the side effects), data for breast cancer in ALDH2 deficient women who consume sufficient alcohol are not available, and no data are available on AA levels in the breast.
(Enlarge Image)
Figure 3.
Ethanol metabolism and its possible role in breast carcinogenesis. Ethanol is metabolized to AA by ADH and further to acetate by ALDH. ADH1B and ADH1C are polymorphic and may generate various amounts of AA. In addition, 50% of Asians have a mutation in the ALDH2 gene, resulting in low ALDH activity with accumulation of AA when they drink alcohol. Either increased generation or decreased degradation of AA may result in its accumulation. AA reacts with DNA, forming DNA adducts; inhibits the antioxidative defence—and the nuclear repair system and also inhibits DNA methylation. Ethanol is additionally metabolized via CYP2E1 to AA. During this process ROS are formed. ROS binds either directly to DNA or via lipidperoxidation products such as 4-hydroxynonenal (4-HNE), increases the expression of the AP-1 gene (cellular hyperproliferation) and stimulates metalloproteinases (increase invasiveness and metastasis). CYP2E1 also converts various procarcinogens, including those present in cigarette smoke to their ultimative carcinogens and degrades retinoic acid to polar metabolites, resulting in a loss of retinoic acid associated with increased cell dedifferentiation and cellular hyperproliferation.
In addition to a decreased metabolism of AA by ALDH deficiency, an increased production also favours its accumulation. ADH1C and ADH1B reveal polymorphism. The ADH1B*2 allele codes for an enzyme 40 times more active compared with that encoded by the ADH1B*1 allele. Thus, ADH1B*2 homozygotes produce high AA concentrations, which are associated with severe side effects (flushing syndrome) when these individuals consume ethanol. As a result, these individuals (almost exclusively Asians) do not drink alcohol at all and are protected against alcoholism (Edenberg, 2007). On the other hand, the ADH1C*1 allele codes for an ADH enzyme 2.5-fold more active compared with the enzyme encoded by the ADH1C*2 allele. Individuals homozygote for ADH1C*1 seem to have also an increased risk for upper gastrointestinal tract cancer, for the liver and for the colorectum (Visapaa et al., 2004; Homann et al., 2006, 2009). With respect to the breast, contradictory results have been reported. Some studies report an increased breast cancer risk in ADH1C*1 homozygous individuals who consume alcohol chronically (Freudenheim et al., 1999; Coutelle et al., 2004) while others do not (Hines et al., 2000; Breast Cancer Association Consortium, 2006; Terry et al., 2006; Visvanathan et al., 2007). Again, it may be the amount of alcohol leading to higher AA concentrations that determines the risk for breast cancer in alcohol-consuming women. Indeed, in the two positive studies, regular alcohol consumption was found to be higher when compared with the other studies. A recent meta-analysis came to the conclusion that there is not sufficient evidence that women who are ADH1C*1 homozygote have an increased breast cancer risk when they consume alcohol (Ding et al., 2012)
Experimental work has shown that AA accumulates in mammary tissue following a single dose of oral ethanol (Castro et al., 2008). It was therefore hypothesized that AA can be produced from ethanol in mammary tissue but its further detoxification to acetate is lacking (Fanelli et al., 2010).
Xanthine oxidoreductase and aldehyde oxidase are present in breast tissue (Wright et al., 1999). Thus, AA metabolism may generate ROS, including the superoxide anion free radical, the neutral hydroxide free radical and hydrogen peroxide. It has also been reported that microsomal ethanol oxidation occurs in mammary tissue which could lead to the generation of ROS (Fanelli et al., 2010).
ROS can lead to DNA mutation, base deletion, single- and double-strand breaks. ROS also results in the activation of the AP-1 gene (c-jun and c-fos) with consecutive changes in cell cycle behaviour (Wang and Seitz, 2004). Furthermore, ROS may lead to lipidperoxidation, resulting in the generation of lipidperoxidation products such as 4-hydoxynonenal, which may bind to adenosine or cytosine bases, forming highly carcinogenic exocyclic etheno DNA adducts (Sodum and Chung, 1988). This has been shown for the liver (Wang et al., 2009) and for the oesophagus (Millonig et al., 2011). ROS can also activate metalloproteases and can lead to an increased expression and secretion of metalloprotease 2 and 9 (Etique et al., 2006; Ke et al., 2006; Luo, 2006), resulting in enhanced invasiveness and metastasis.
Alcohol results in the activation of various intracellular signalling pathways. For example, alcohol leads to an increased expression of the c-fos transcription factor and results in an increased phosphorylation of the c-jun-terminal protein kinase (JNK), the p38 mitogen-activated protein kinase and the phosphatridylinositol 3-kinase (PI3K) (Ma et al., 2003; Ke et al., 2006). Whether JNK activation is due to an increase in oxidative stress, a decrease in retinoic acid or both still remains to be determined.
Ethanol also stimulates epidermal growth factor receptor (EGFR) signalling via cyclic adenosine monophosphate (c-AMP)-dependent stimulation of amphiregulin transcription (Mill et al., 2009). EGFR signalling modulates ER-dependent signalling via the PI3K/Akt pathway and the IKK phosphorylation of ER (Biswas and Iglehart, 2006). EGFR signalling may also be responsible for the decrease in E-cadherin expression noted after ethanol ingestion (Meng et al., 2000). E-cadherin is a tumour suppressor (Jeanes et al., 2008).
The effect of ethanol on the invasion of breast cancer cells correlated significantly with the expression levels of ErbB2, a receptor tyrosine kinase and the ethanol-mediated overexpression of ErbB2 is associated with an enhanced adhesion of breast cancer cells to fibronectin, an extracellular matrix protein mediated by the focal adhesion kinase-1 pathway (Ma et al., 2003; Ke et al., 2006; Xu et al., 2010). Such an adhesion is an important initial step for cancer cell invasion. Furthermore, epithelial-mesenchymal transition (EMT) plays an important role in cancer progression and metastasis and alcohol stimulates EMT via an EGFR-Snail mediated pathway (Forsyth et al., 2010). Alcohol also stimulates the expression and secretion of metalloproteases (possibly via ROS), leading to the degradation of extracellular matrix resulting in an enhanced tumour cell invasiveness and metastasis (Duffy et al., 2000; Meng et al., 2000; Etique et al., 2006; Luo, 2006). All these mechanisms may explain, at least in part, that ethanol enhances metastasis of breast cancer.
Finally, ethanol may affect epigenetics by changing methylation of DNA and/or histones. This has been shown in other tissues such as the liver and is due to a decrease in the major methyldonor S-adenosylmethionine. Chronic ethanol ingestion is associated with folate deficiency, which is important in the generation of methionine, and AA bind and inactivates various enzymes in the generation of S-adenosylmethionine (Stickel et al., 2006). DNA methyltransferase is also inhibited by AA. However, these pathomechanisms have been identified for the liver, but data for the breast are not available.
Possible Mechanisms of Ethanol-Mediated Breast Cancer Development
The mechanisms by which alcohol stimulates breast carcinogenesis are still not understood. Major pathophysiologic research on the carcinogenic effect of chronic alcohol consumption has focused on the upper alimentary tract, the liver and the colorectum (Baan et al., 2007; Seitz and Stickel, 2007; IARC, 2010). Studies investigating mechanisms of ethanol-mediated breast cancer are rare and thus information is limited. Since a carcinogenic effect of estrogens on breast tissue has been observed, and since alcohol ingestion results in an elevation of serum estrogen concentrations, it is speculated that the carcinogenic effect of ethanol is mediated, at least in part, by estrogens (Fernandez, 2011). In addition, some research has concentrated on acetaldehyde (AA), the first and most toxic metabolite of ethanol oxidation which is by itself carcinogenic (Secretan et al., 2009; Seitz and Stickel, 2010) and on oxidative stress (Seitz and Stickel, 2006). The ethanol effect on epigenetic modifications (Stickel et al., 2006) and on interaction with retinoids (Wang and Seitz, 2004) has not been well studied in the breast when compared with other tissues. Alcohol-related carcinogenesis may also interact with other factors such as smoking, diets, comorbidities and depends on genetic susceptibility.
Animal Studies
Only one study demonstrated that ethanol by itself (without the additional administration of a carcinogen) resulted in breast cancer. When 10 and 15% ethanol in the drinking water was given to female ICR mice for 25 months, 45% of the animals developed papillary and medullary adenocarcinomas of the breast (P = 0.0012), while no tumours were found in the control group (Watabiki et al., 2000). In addition, a number of animal studies have been performed with alcohol as a modifier of chemically induced mammary carcinogenesis by using either methylnitrosourea (MNU) or dimethylbenzanthracene (DMBA). When MNU (30 mg/kg) or DMBA (5 mg/rat) was used to induce mammary tumours in Sprague Dawley (SD) rats, ethanol as 10–30% of calories in the daily diet significantly increased the incidence of adenocarcinomas (Singletary et al., 1991, 1995), while with higher doses of DMBA (39 mg/kg) no effect on mammary tumourigenesis was seen (Singletary et al., 1991). In addition, other studies with 10 and 15 mg of DMBA and ethanol in the drinking water showed either an increased number of mammary tumours per rat (P < 0.006) (Rogers and Conner, 1990) or an increased tumour incidence (P < 0.001) (Hilakivi-Clarke et al., 2004) Furthermore, in utero exposure to ethanol-increased mammary tumourigenesis induced with DMBA in rats (McDermott et al., 1992). Recently, foetal alcohol exposure (6.7% ethanol, equivalent to 3–5 drinks in 2 h for a women) in SD rats from Day 11–21 of gestation decreased mammary tumour latency in adulthood induced by NMU (Polanco et al., 2010).
Another study investigated the interaction of estrogens and ethanol in ovariectomized mice on mammary tumours. While estrogens alone inhibited tumour growth in this model, the addition of ethanol increased insulin sensitivity and increased tumour growth in these obese mice (Hong et al., 2010).
The Role of Estrogens in Ethanol-mediated Breast Cancer
Long-term exposure to estrogens increases the risk of developing breast cancer in women (Colditz, 1998). In rats, the continuous administration of supra-physiological doses of estrogens resulted in mammary adenocarcinomas, while low doses of estrogens administered over a long-time period lead to fibroadenomas (Russo and Russo, 1996).
The mechanisms by which estrogens induce breast cancer are still not completely clear, although it is well accepted that the binding of estrogens to its nuclear receptor, ER alpha (ERα), initiate a complex intracellular signal sequence, finally stimulating cell proliferation (Suga et al., 2007).
Alcohol-increased plasma estrogen levels had been significantly demonstrated in controlled feeding studies, with human female volunteers (Reichman et al., 1993; Ginsburg et al., 1996; Purohit, 1998; Sarkola et al., 1999, 2000, 2001; Coutelle et al., 2004; Mahabir et al., 2004, Sierksma et al., 2004; Seitz and Maurer, 2007). In postmenopausal women with hormone replacement therapy, 15 or 30 g of ethanol daily consumed over 8 weeks resulted in significantly elevated serum concentrations of oestrone sulphate and dehydroepiandrosterone sulphate (Dorgan et al., 2001). In premenopausal women, not only an acute high dose of ethanol (0.7 g/kg b.wt.) (Mendelson et al., 1988), but also a small dose (0.225 g/kg b.wt.) equivalent to one drink (10–12 g) led to significantly elevated plasma oestradiol levels. When 0.7 g ethanol/kg b.wt. was given, the observed effect was most pronounced during the ovulatory phase of the menstrual cycle and in women using oral contraceptives. With the low dose of ethanol, the elevation has been observed in all phases of the menstrual cycle, but seems to be especially relevant at mid-cycle where oestradiol concentrations are already high, being further stimulated by ethanol by ~27–38% (Coutelle et al., 2004). There is some evidence that alcohol consumption during adolescence and early adulthood may stimulate breast carcinogenesis more than drinking later in life (Longnecker et al., 1995).
In addition, alcohol at a concentration of 0.06% stimulates the expression of ERα and the oestradiol biosynthesis enzyme aromatase in human breast cancer cell lines (Gavaler and Van Thiel, 1992; Singletary and Gapstur, 2001; Etique et al., 2004). Alcohol stimulates ER signalling in human breast cancer cells (Fan et al., 2000). Alcohol-mediated ER-dependent gene expression may result in cell hyperproliferation in ER-positive Michigan Cancer Foundation-7 (MCF-7) human breast cancer cells (Ginsburg et al., 1996). It has also been reported that crosstalk between adenosine receptor (A2A isoform) and ERα mediates ethanol action in MCF-7 breast cancer cells, which may open new therapy strategies in oestrogen-dependent breast cancer (Etique et al., 2009).
The fact that several studies support the hypothesis that alcohol is more strongly related to ER positive than to ER negative breast tumours underlines the pathogenic role of estrogens in alcohol-mediated breast cancer.
Estrogens may also act directly as tumour initiators. They may be genotoxic both in vitro and in vivo and they may result in DNA damage including single-strand breaks and DNA adducts (Chakravarti et al., 2001; Liu and Lin, 2004). Catechol estrogen 3,4-quinones (an estrogen metabolite) generates DNA mutations initiating breast cancer (Cavalieri et al., 2004; Yager and Davidson, 2006). The fact that 'ERα knockout mice expressing the Wnt-1 oncogene (ERKO/Wnt-1) develop breast cancer when they receive estrogens' support the direct genotoxic effect of estrogens without ERα-mediated pathways (Bocchinfuso et al., 1999). Furthermore, the administration of estrogens to ovariectomized mice results in more tumours in a shorter period of time when compared with control animals even when anti-estrogens are given (Fernandez, 2011). Estrogen treatment of human breast epithelial cells induced gene expression alterations, epigenetic modifications and finally phenotypical changes indicating neoplastic transformation (Fernandez et al., 2005, 2006, 2010; Russo et al., 2006; Huang et al., 2007; Fernandez and Russo, 2010). Since these cells are ER negative, direct genotoxic effects of estrogens must be responsible for the changes observed.
Several mechanisms by which ethanol affects the levels of sex hormones in women have been suggested. These include an ethanol-mediated increase in the hepatic redox state resulting in a decrease of steroid metabolism (Sarkola et al., 1999, 2001). Another explanation for the increased estrogen concentrations after alcohol is the increased aromatase activity following chronic ethanol consumption, which leads to an enhanced conversion of testosterone to oestrogens with reduced testosterone and increased oestrogen concentrations (Gavaler and van Thiel, 1992). Alcohol also inhibits the activity of sulfotransferase and 2-hydroxylase, two enzymes important in estrogen degradation (Eagon, 2010). Alcohol may also decrease melatonin secretion which inhibits estrogen production (Stevens et al., 2000). Furthermore, alcohol may interact with the production of luteinizing hormone from the pituitary gland, which favours estradiol release from the ovaries (Rettori and McCann, 1997). An important study in which rats were exposed to ethanol in utero showed an increased number of terminal end buds, which are targets for malignant transformation, suggesting that alcohol when administered during foetal life may modify signals controlling mammary gland development (Hilakivi-Clarke et al., 2004).
Increased estrogen concentrations following ethanol ingestion may not only increase breast cancer risk, it may also explain at least in part the increased susceptibility of the female liver towards alcohol in the pathogenesis of alcoholic liver disease since estrogens may result in enhanced hepatic mitochondrial damage resulting in enhanced fat deposition, apoptosis and the generation of reactive oxygen species (ROS) (Eagon, 2010).
AA, A Possible Carcinogen for the Breast
Alcohol is metabolized via alcoholdehydrogenase (ADH) to AA and further via acetaldehydedehydrogenase (ALDH) to acetate (Fig. 3). AA is toxic, mutagenic and carcinogenic in animal experiments. It binds rapidly to proteins and DNA, forming stable carcinogenic adducts. AA also decreases the anti-oxidative defence system and modifies indirectly epigenetic histone and DNA methylation by decreasing the availability of S-adenosinemethionine, the major methyl donator (for review, Seitz and Stickel, 2007). AA has therefore been classified as a carcinogen by the IARC (Secretan et al., 2009). Thus, an increased burden of AA by either enhanced generation or slow degradation may stimulate carcinogenesis. Indeed, this has been shown for other ethanol-mediated cancer sites (Seitz and Stickel, 2007). Approximately 40% of Japanese have a heterozygote mutation at the ALDH2 gene with a striking reduction of the ALDH2 enzyme activity resulting in a significant elevation AA level in blood and saliva (Vakevainen et al., 2000). Since this increase in AA is associated with side effects such as flushing, tachycardia, nausea, vomiting and sweating, alcohol consumption is reduced below three drinks in heterozygots or close to zero in homozygots. Only a few Japanese men still consume alcohol even when side effects occur. Owing to AA accumulation, these individuals have an increased risk for upper aerodigestive tract cancer (oropharynx, larynx and oesophagus) and for colorectal cancer (Yokoyama et al., 1998). AA may act as a carcinogen only when its concentration exerts mutagenic levels, e.g. only when sufficient alcohol is consumed to reach that level. Owing to the fact that Japanese women drink only small amounts of alcohol, and that Japanese women who are ALDH2-deficient tend not drink alcohol (due to the side effects), data for breast cancer in ALDH2 deficient women who consume sufficient alcohol are not available, and no data are available on AA levels in the breast.
(Enlarge Image)
Figure 3.
Ethanol metabolism and its possible role in breast carcinogenesis. Ethanol is metabolized to AA by ADH and further to acetate by ALDH. ADH1B and ADH1C are polymorphic and may generate various amounts of AA. In addition, 50% of Asians have a mutation in the ALDH2 gene, resulting in low ALDH activity with accumulation of AA when they drink alcohol. Either increased generation or decreased degradation of AA may result in its accumulation. AA reacts with DNA, forming DNA adducts; inhibits the antioxidative defence—and the nuclear repair system and also inhibits DNA methylation. Ethanol is additionally metabolized via CYP2E1 to AA. During this process ROS are formed. ROS binds either directly to DNA or via lipidperoxidation products such as 4-hydroxynonenal (4-HNE), increases the expression of the AP-1 gene (cellular hyperproliferation) and stimulates metalloproteinases (increase invasiveness and metastasis). CYP2E1 also converts various procarcinogens, including those present in cigarette smoke to their ultimative carcinogens and degrades retinoic acid to polar metabolites, resulting in a loss of retinoic acid associated with increased cell dedifferentiation and cellular hyperproliferation.
In addition to a decreased metabolism of AA by ALDH deficiency, an increased production also favours its accumulation. ADH1C and ADH1B reveal polymorphism. The ADH1B*2 allele codes for an enzyme 40 times more active compared with that encoded by the ADH1B*1 allele. Thus, ADH1B*2 homozygotes produce high AA concentrations, which are associated with severe side effects (flushing syndrome) when these individuals consume ethanol. As a result, these individuals (almost exclusively Asians) do not drink alcohol at all and are protected against alcoholism (Edenberg, 2007). On the other hand, the ADH1C*1 allele codes for an ADH enzyme 2.5-fold more active compared with the enzyme encoded by the ADH1C*2 allele. Individuals homozygote for ADH1C*1 seem to have also an increased risk for upper gastrointestinal tract cancer, for the liver and for the colorectum (Visapaa et al., 2004; Homann et al., 2006, 2009). With respect to the breast, contradictory results have been reported. Some studies report an increased breast cancer risk in ADH1C*1 homozygous individuals who consume alcohol chronically (Freudenheim et al., 1999; Coutelle et al., 2004) while others do not (Hines et al., 2000; Breast Cancer Association Consortium, 2006; Terry et al., 2006; Visvanathan et al., 2007). Again, it may be the amount of alcohol leading to higher AA concentrations that determines the risk for breast cancer in alcohol-consuming women. Indeed, in the two positive studies, regular alcohol consumption was found to be higher when compared with the other studies. A recent meta-analysis came to the conclusion that there is not sufficient evidence that women who are ADH1C*1 homozygote have an increased breast cancer risk when they consume alcohol (Ding et al., 2012)
Experimental work has shown that AA accumulates in mammary tissue following a single dose of oral ethanol (Castro et al., 2008). It was therefore hypothesized that AA can be produced from ethanol in mammary tissue but its further detoxification to acetate is lacking (Fanelli et al., 2010).
Ethanol, Oxidative Stress and Breast Cancer
Xanthine oxidoreductase and aldehyde oxidase are present in breast tissue (Wright et al., 1999). Thus, AA metabolism may generate ROS, including the superoxide anion free radical, the neutral hydroxide free radical and hydrogen peroxide. It has also been reported that microsomal ethanol oxidation occurs in mammary tissue which could lead to the generation of ROS (Fanelli et al., 2010).
ROS can lead to DNA mutation, base deletion, single- and double-strand breaks. ROS also results in the activation of the AP-1 gene (c-jun and c-fos) with consecutive changes in cell cycle behaviour (Wang and Seitz, 2004). Furthermore, ROS may lead to lipidperoxidation, resulting in the generation of lipidperoxidation products such as 4-hydoxynonenal, which may bind to adenosine or cytosine bases, forming highly carcinogenic exocyclic etheno DNA adducts (Sodum and Chung, 1988). This has been shown for the liver (Wang et al., 2009) and for the oesophagus (Millonig et al., 2011). ROS can also activate metalloproteases and can lead to an increased expression and secretion of metalloprotease 2 and 9 (Etique et al., 2006; Ke et al., 2006; Luo, 2006), resulting in enhanced invasiveness and metastasis.
Other Mechanisms
Alcohol results in the activation of various intracellular signalling pathways. For example, alcohol leads to an increased expression of the c-fos transcription factor and results in an increased phosphorylation of the c-jun-terminal protein kinase (JNK), the p38 mitogen-activated protein kinase and the phosphatridylinositol 3-kinase (PI3K) (Ma et al., 2003; Ke et al., 2006). Whether JNK activation is due to an increase in oxidative stress, a decrease in retinoic acid or both still remains to be determined.
Ethanol also stimulates epidermal growth factor receptor (EGFR) signalling via cyclic adenosine monophosphate (c-AMP)-dependent stimulation of amphiregulin transcription (Mill et al., 2009). EGFR signalling modulates ER-dependent signalling via the PI3K/Akt pathway and the IKK phosphorylation of ER (Biswas and Iglehart, 2006). EGFR signalling may also be responsible for the decrease in E-cadherin expression noted after ethanol ingestion (Meng et al., 2000). E-cadherin is a tumour suppressor (Jeanes et al., 2008).
The effect of ethanol on the invasion of breast cancer cells correlated significantly with the expression levels of ErbB2, a receptor tyrosine kinase and the ethanol-mediated overexpression of ErbB2 is associated with an enhanced adhesion of breast cancer cells to fibronectin, an extracellular matrix protein mediated by the focal adhesion kinase-1 pathway (Ma et al., 2003; Ke et al., 2006; Xu et al., 2010). Such an adhesion is an important initial step for cancer cell invasion. Furthermore, epithelial-mesenchymal transition (EMT) plays an important role in cancer progression and metastasis and alcohol stimulates EMT via an EGFR-Snail mediated pathway (Forsyth et al., 2010). Alcohol also stimulates the expression and secretion of metalloproteases (possibly via ROS), leading to the degradation of extracellular matrix resulting in an enhanced tumour cell invasiveness and metastasis (Duffy et al., 2000; Meng et al., 2000; Etique et al., 2006; Luo, 2006). All these mechanisms may explain, at least in part, that ethanol enhances metastasis of breast cancer.
Finally, ethanol may affect epigenetics by changing methylation of DNA and/or histones. This has been shown in other tissues such as the liver and is due to a decrease in the major methyldonor S-adenosylmethionine. Chronic ethanol ingestion is associated with folate deficiency, which is important in the generation of methionine, and AA bind and inactivates various enzymes in the generation of S-adenosylmethionine (Stickel et al., 2006). DNA methyltransferase is also inhibited by AA. However, these pathomechanisms have been identified for the liver, but data for the breast are not available.
Source...