Aim of E9-PPCF team:
Our main objective is to demonstrate, understand and prevent the link that has been pointed out by epidemiological studies between red and processed meat and colorectal cancer, by setting up studies from cellular and animal models to Humans.
Colorectal cancer (CRC) is a major health issue in developed countries where it is the first cause of death by cancer in non-smokers when both sexes are considered. Food factors play an important role in the development of CRC: in 2007 and 2011, the World Cancer Research Fund (WCRF) qualified as convincing the increased risk of CRC associated with red and processed meat consumption. As a result, the recommendations are to reduce red meat consumption and to avoid eating processed meat. In the UK, red and processed meat would account for 16 and 25% of CRC incidence, in women and men, respectively (Parkin, 2011, Br J Nutr). Such epidemiological link is the important and solid basis of our research. But epidemiology is a science of statistical associations. Then, our first objective is to demonstrate experimentally the link between CRC and red / processed meat consumption. Our second objective is to understand the origins of the hazardous effect of meat consumption. However, red meat and meat products have a particular nutritional interest, as source of essential amino-acids, well-absorbed heme iron and micronutrients. Accordingly, we work on the development of preventive approaches including the risk/benefit balance in order to find an alternative to the reduction of red and processed meat consumption.
1-Experimental demonstration that cured meat promotes CRC in rodents
We have previously (2003-2008) shown that red meat and heme iron had a promoting effect in a rodent model of CRC, related to an increased lipid peroxidation in colon lumen, supporting the positive association proposed by epidemiology. We had demonstrated that hemoglobin mimicked red meat effect. However, in a previous meta-analysis, we additionally estimated that 1 g of processed meat increased the risk of CRC 6 times more than 1 g of fresh red meat. So, from 2009 to 2014 we further explored experimentally the association between processed/cured meat consumption and CRC. Most human adenocarcinoma would evolve from aberrant crypt foci (ACF) and adenoma, so we firstly focused on studies at a preneoplastic level (ACF and Mucin Depleted Foci (MDF)) in rats injected with azoxymethane (AOM).
As proof of concept, we firstly demonstrated that freeze-dried ham promoted AOM-induced ACF and MDF in rat colon (Pierre, 2010). As ham is a cured, nitrated and cooked processed meat, it leads to the formation of globin liberated-nitrosyl heme in the colon lumen. We proposed that the pro-carcinogenic effect of ham was due to this free nitrosyl heme that was more toxic than native myoglobin. Moreover, we demonstrated that free hemin mimicked ham effect on early biochemical markers associated with carcinogenesis (urinary and fecal biomarkers of peroxidation, cytotoxicity and nitrosation). Furthermore, to improve dietary relevance of experimental diets, we chose to work with fresh meat in subsequent projects. This was an opportunity to create collaborations with meat industry (ANR projects Hemecancer and Securiviande, Coord. F. Pierre).
In a 2x2x2x2 factorial design study (muscle color, processing temperature, added nitrite, packaging) carried out with model cured meats manufactured by meat industry, we showed that only a cooked nitrite-treated and oxidized high-heme meat (similar to aerobically stored-ham) significantly increased the fecal level of Apparent Total N-nitroso Compounds (ATNC, end-products of nitrosation) and promoted preneoplastic lesions, compared with the no-meat control diet (Santarelli, 2010). We also demonstrated that the major worldwide contributor of cured meat intake, namely “hot dog”, promoted CRC in rats, with a significant correlation with the increase of fecal ATNC.
In total, our data support the hypothesis that N-nitroso compounds (NOC) formed by luminal nitrosation catalyzed by heme are major pro-carcinogenic factors in the gut following processed meat intake.
After those experiments at the preneoplastic level, we have confirmed in the ANR project SecuriViande, the promotion of carcinogenesis by fresh red meat and processed meat at the tumoral stage in Apc Min/+ mice mutated on the adenomatous polyposis coli gene (Apc) and considered as a genetically relevant animal model to human CRC.
2-Heme iron is the major CRC promoting agent in meat
The deleterious factors arising from red meat are heme iron, heterocyclic amines in cooked meat and endogenous NOC. We have demonstrated that heme iron is an actor of the promoting effect of red meat (2003-2008) but the relative contribution of each factor is unknown. The relative part of heme iron in the form of hemoglobin, heterocyclic amines and ATNC was determined by a multifactorial design and preneoplastic endpoint (MDF). We demonstrate the strong role of heme iron in the promotion of colon carcinogenesis, independently of dietary heterocyclic amines and ATNC (Bastide, 2014, submitted). Furthermore, the central role of heme was highlighted in vitro, because only fecal water (FW) from rats given dietary hemoglobin was cytotoxic to normal epithelial cells in this study. We demonstrated here the importance of the hypothesis heme/lipoperoxidation for hemoglobin and red meat, whereas our studies on processed meat highlights the role of the hypothesis Heme/NOC (see §A-1).
We also conducted a meta-analysis of prospective cohort studies of colon cancer reporting heme intake and cases of colon cancer (Bastide, 2011). The relative risk of colon cancer was established at 1.18 (95% CI: 1.06–1.32) for subjects in the highest category of heme iron intake compared with those in the lowest category, confirming the experimental demonstration of the prominent role of heme iron.
1-Red meat: heme iron, from induced aldehyde formation to Darwinian tumor promotion model
The development of CRC is a multistep process involving several somatic mutations; among them, Apc mutation is a frequent and early event. This mutation transforms normal colonocytes into pre-neoplastic cells. To understand the mechanisms linking red meat and CRC, we used conditionally immortalized but non cancerous mouse colonic epithelial cells, bearing or not the Apc mutation (Forest, 2003, Nut Cancer). In these models, we previously showed that fecal water (FW) could have a differential cytotoxic effect: preneoplastic cells were more resistant than normal cells to FW from animals fed meat-enriched diets. This differential sensitivity constitutes a relevant biomarker of CRC promotion by red meat (Pierre, 2007, Carcinogenesis). It also lead us to propose a Darwinian tumor promotion hypothesis (Moreno, 2008, Nat Cancer rev) according to which the mutation on Apc gene confers a selective advantage to the cells that are exposed to FW of heme iron fed rats. Our in vivo experiments have pointed out the possible role of lipid peroxidation issued aldehydes in the promoting effect of red meat on CRC, as the TBARS (ThioBarbituric Acid Reactive Substances) index increased in FW of heme iron-fed rats. The pretreatment of FW with an aldehyde trapping resin decreased TBARs content and FW cytotoxicity, showing the importance of aldehydes as the deleterious compounds in such complex matrix (Bastide, 2014, submitted). In order to decipher the mechanisms related to aldehydes and CRC promotion, we focused on 4-hydroxynenal (HNE), one of the major secondary products of lipid peroxidation, which is present in FW from red meat fed rats. HNE is a biologically active compound, because of its ability to form adducts with proteins. We previously showed that pre-neoplastic cells were also more resistant to HNE than normal cells. To explain this differential effect, we used radiolabelled and stable isotope metabolite tracking technique and we showed that pre-neoplastic cells had a higher capacity than normal cells to detoxify HNE, notably leading to a rapid and efficient formation of thiol compounds and extracellular cysteine conjugates. This suggested that the detoxification pathway could have a major role in the resistance of pre-neoplastic cells (Jouanin, 2011; Baradat, 2011). The phase 2 enzymes involved in HNE detoxification were described to be regulated by the transcription factor Nrf2 (Nuclear factor erythroid-2 related factor-2). In pre-neoplastic cells, we demonstrated that Nrf2 was basically more activated, with a higher expression of the target genes (GSTA4, xCT, ALDHs). When cells were treated by HNE, Nrf2 was induced in normal and preneoplastic cells, but the detoxification enzymes were always higher expressed in preneoplastic cells explaining the better resistance. Through the ectopic expression of Nrf2 in normal cells and the invalidation of Nrf2 in preneoplastic cells by SiRNA, we showed that Nrf2 was responsible for the regulation of the apoptosis/survival balance of normal and preneoplastic cells upon HNE exposure (Figure 1; Dalleau, in preparation). This work was granted by the Ligue Régionale contre le Cancer (2011) and supported by ToxAlim-M2C and TRiX plateforms.
Figure 1:Colorectal cancer promotion by HNE. Darwinian tumor promotion hypothesis. HNE is a secondary product of lipid peroxidation that is generated in the colon via heme iron provided by red meat. The resistance of pre-neoplastic cells regarding HNE is due to a better Nrf2-dependent antioxidant defense. The greater expression of detoxification enzymes leads to a stronger biotransformation of HNE that limits its deleterious effects. The positive selection of pre-neoplastic cells to the detriment of normal cells could explain the promoting effect of red meat in vivo.
2-Processed meat: heme iron, from NOC to nitrosylated heme hypothesis
The hypothesis that NOC would be involved in the etiology of CRC has been proposed by different authors, and our data demonstrated that cured meat promotion can be in part due to heme-induced formation of NOC. Firstly we proposed that total nitrosation catalyzed by heme iron was decisive (Santarelli, 2010), with the formation of NOCs measured as ATNC. We refined this hypothesis in the multifactorial study (hemoglobin x heterocyclic amines x ATNC) with the quantification of ATNC sub-categories in fecal water. After this quantification, we have proposed that nitrosyl iron (FeNO) (a small sub-category of ATNC) is a major actor of the promotion of colon cancer by processed meats (Bastide, 2014, accepted for publication nov,26, 2014).
We have demonstrated the central role of heme iron in the promotion of CRC by red and processed meat (Figure 2). Initially (2003-2008), we explored the prevention of preneoplastic lesions by the chelation of heme by calcium in diets enriched in hemin, freeze-dried beef and more recently in a model of cured meat (Pierre, 2013) and in hot-dogs (Santarelli, 2013). But the chelation of heme by calcium limits its biodisponibility and thereby reduces the nutritional value of meat. Limiting heme iron absorption may not represent a pertinent recommendation, especially in the case of anemia prone people. So, rather than heme chelation, we chose to inhibit heme catalyzed reactions such as peroxidation or nitrosylation, by adding directly antioxidants in cured meat during processing (HemeCancer project; Polyphenol project in collaboration with meat industry; Figure 2). We have thus demonstrated that the addition of tocopherol, red wine extract or pomegranate in a model of cured meat is efficient to inhibit the promotion of MDF. Similarly, in the project Biopuntia (ANR Blanc International, Coord. F. Guéraud), we showed that powdered freeze-dried Opuntia cladodes were able to reduce the cytotoxicty of FW of heme-fed rats, although no significant effect was observed on ACF formation. In the non-processed red meat, the addition of antioxidants was done by marinade. Then, grape-olive marinade was shown to reduce preneoplastic lesions (MDF) in rodents, compared to red meat (ANR SecuriViande project).
Figure 2: Colorectal cancer and heme iron. Catalytic effects of heme iron on the neoformation of NOC and aldehydes, consequences for the development of CRC and preventive strategies. Heme iron catalyzes the formation of NOC and lipid peroxidation endproducts, which can explain the promoting effect of processed and red meat on CRC. The catalytic effects of heme iron can be inhibited by trapping (calcium, chlorophyll). The endogenous formation of NOC and/or lipid peroxides can be inhibited by vitamins C, E and polyphenols.
The molecular studies in normal and preneoplastic cells treated with HNE revealed the importance of the anti-oxidant cell defense dependent on Nrf2 in the regulation of apoptosis. Therefore, Nrf2 was a relevant target for chemoprevention. In this purpose, we chose curcumin (CCM) and pterostilbene (PS), which were described to activate Nrf2. The pretreatment of normal cells with CCM or PS activated Nrf2, up-regulated the HNE-detoxification genes and decreased apoptosis induction upon HNE exposure. In contrast, in preneoplastic cells, CCM and PS did not over-activate Nrf2 and also increased their apoptosis after HNE treatment. Such reversal of the differential of sensitivity in normal and preneoplastic cells could be predictive of a chemopreventive strategy based on early targeting of Nrf2, to reinforce antioxidant defenses in normal cells (Dalleau, in preparation).
On the basis of these in vitro results, we added CCM in cured meat during processing and we observed a decrease in MDF (Polyphenol project), indicating the pertinence of targeted preventive strategies.
Transfer the above mechanisms to human volunteers
Recommendations of WCRF to limit red meat intake and to avoid processed meat intake may reduce the colorectal cancer burden (WCRF, 2007; 2011). However, people of lower social status whose processed meat intake is high are not receptive to such nutritional messages and are less likely than affluent people to change risky behaviors (Darmon, 2008, Am J Clin Nutr; Aston, 2013, J Hum Nutr Diet). The resulting consequences on the quality and length of life are dramatic (Smith, 2010, Health Stat Q), and CRC incidence varies between socioeconomic classes and, therefore, contributes to health inequalities (Weiderpass, 2006, BMC Gastroenterol). Conveying information on risks and benefits has almost no effect on food choices in less-educated people and, thus, tends to enlarge the gap. We decided to include these parameters in the construction of our research study and to be effective; we have chosen to focus on the modification of the food, rather than establishment of dietary recommendations. We decided to develop projects in this way through interaction with meat industry.
The promotion of CRC by a model of cured meat in rats was associated with increased fecal ATNC content and TBARS. To transfer those results in human, we firstly assessed whether a cured meat could increase these early biomarkers in human volunteers and whether prevention strategies efficient in rats were also effective in humans. In a crossover study conducted in the HemeCancer project (in collaboration with CRNH in Clermont), we showed that the consumption of cured meat for 4 days was enough to increase ATNC and TBARS concentrations in stools of volunteers. Furthermore, as in rats, adding calcium in the diet or tocopherol in the processed meat was sufficient to inhibit the increase of ATNC and TBARs in stools.
In the SecuriViande project, as in HemeCancer, we used exactly the same model of cured meat in rats and in human volunteers. However, adding tocopherol (or polyphenols) during process of cured meat could alter organoleptic properties. Since the final endpoint of our projects is a mode of production that leads to non-promoting meat, such meat has to be well accepted by the consumer. So, in collaboration with the Center for Taste and Feeding Behaviour of INRA-Dijon, we evaluated if the modifications of meat processing alter or not organoleptic properties and consumer acceptability. Assessed by a panel of consumers, we demonstrated that tocopherol addition in cured meat had no consequences. With this project, from cells up to humans, we have a unifying project for the team that allows us to study this issue from an academic analysis up to an industrial application.