Research Article, J Food Nutr Disor Vol: 4 Issue: 3
Role of pH in the Mutagenicity of Aflatoxin B1 in Maize Tortillas during In Vitro Human Digestion Model
| María de Guadalupe Moctezuma-Zárate1, Magda Carvajal-Moreno1*, Javier J. Espinosa-Aguirre2, María Eugenia Gonsebatt-Bonaparte2, Francisco Rojo-Callejas3, Pável Castillo-Urueta1 Israel Pérez-López4 and Silvia Ruiz-Velasco5 | |
| 1Laboratorio de Micotoxinas, Departamento de Botánica, Instituto de Biología, Universidad Nacional Autónoma de México (UNAM). 04510 México.(Mycotoxins Laboratory, Botany Department, Institute of Biology, National Autonomous University of Mexico (UNAM). 04510, México) | |
| 2Laboratorio de Mutagénesis, Instituto de Investigaciones Biomédicas, UNAM.04510 México.(Mutagenesis Laboratory, Biomedical Research Institute, UNAM, 04510 Mexico) | |
| 3Departamento de Química Analítica, Facultad de Química, UNAM. 04510 México.(Department of Analytical Chemistry, Faculty of Chemistry, UNAM. 04510, Mexico) | |
| 4Instituto Nacional de Pediatría, Secretaría de Salud. México.(National Institute of Pediatrics, Health Ministry, Mexico) | |
| 5Departamento de Probabilidad y Estadística, Instituto de Investigaciones en Matemáticas Aplicadas y en Sistemas, 04510 UNAM.(Probability and Statistics Department, Applied Mathematics and Systems Research Institute, UNAM. 04510 Mexico) | |
|
Corresponding author :Dr. Magda Carvajal-Moreno |
|
| Received: December 23, 2014 Accepted: May 18, 2015 Published: May 22, 2015 | |
| Citation: Moctezuma-Zárate MG, Carvajal-Moreno M, Espinosa-Aguirre JJ, Gonsebatt-Bonaparte ME, Rojo-Callejas F, et al. (2015) Role of pH in the Mutagenicity of Aflatoxin B1 in Maize Tortillas during In Vitro Human Digestion Model. J Food Nutr Disor 4:3. doi:10.4172/2324-9323.1000174 |
Abstract
Role of pH in the Mutagenicity of Aflatoxin B1 in Maize Tortillas during In Vitro Human Digestion Model
Aflatoxins (AFs) are potent mutagens, carcinogens and teratogens for humans; thus their presence in food is of great concern. A maize tortilla survey in Mexico revealed that 17% are contaminated with AFs. The initial process to prepare tortillas includes boiling maize grains with lime and to assumpt that AFs in maize tortillas are destroyed, protecting consumers from their mutagenic effects.The purpose of this study was to investigate the mutagenicity of aflatoxin B1 (AFB1) in tortillas during digestion.
Keywords: Human digestion; Aflatoxins; Mutagenicity; Carcinogenesis; pH
Keywords |
|
| Human digestion; Aflatoxins; Mutagenicity; Carcinogenesis; pH | |
Introduction |
|
| Aflatoxins (AFs) are inevitable food mutagens, carcinogens and teratogens, it is almost impossible to avoid the ingestion of them, so the research about the changes in the mutagenicity of AFs during the digestion process is of upmost importance. On the other hand, cancer is the group of diseases that causes the highest mortality in humans from just born to old people and all the studies that tend to control these diseases are welcome for human survival. | |
| Aflatoxins (AFs) are bisdihydro-furancoumarins produced from some strains of maize-contaminating molds Aspergillus flavus [1], A. parasiticus [2], A. nomius [3], A. pseudotamarii [4] and A. bombycis [5]. The properties of AFs, including their biosynthesis, production conditions and toxic effects, are well-characterized [6]. They are teratogens [7] that cause chromosomal aberrations and DNA breakage [8], and act as biological carcinogens [9] in animal and human organs such as the liver, colon, pancreas, rectum [10], lung [11] and cervix [12]. Aflatoxin B1 (AFB1) requires bioactivation to achieve mutagenicity [13]. | |
| Surveys have demonstrated the presence of AFs in maize tortillas, the staple food of Mexico [14]. Maize grains are cooked with lime, and it is thought that this treatment eliminates AFs. Therefore, this has been proposed as a method of detoxification [15]. In fact, maize tortillas appear to contain smaller amounts of AFB1 than expected for their maize content, possibly because the AFB1 is lost from the maize pericarp and released into residual water during boiling and washing, and because the alkaline conditions present in tortillas open the lactone ring of AFs and inhibit their fluorescence and detection [16,17]. Additionally, lime treatment is not effective for detoxification, as acidification in the stomach reactivates AFs [18]. | |
| During digestion the mutagenicity of AFB1 in maize tortillas changes in relation to exposure to saliva or gastric and pancreatic fluids, either alone or in combination. Saliva contains the enzyme ptyalin, an amylase that maintains a pH of 6.5 to 7.0 in the mouth [19]. Saliva can transport hepatitis B virus passing also AFs, that can be potential causative agents for hepatitis and liver cancer [20]. There are contradictory reports regarding the role of saliva; it has been suggested to be mutagenic and teratogenic [21], to inactivate the mutagenicity of AFB1 [22] and provide protective antibacterial activities [23] through antioxidant enzymes [24]. Gastric fluid contains pepsin, a protease that degrades food proteins into peptides [25]. Chloride creates the acidic environment (pH 1.5 to 3.0) necessary for pepsin activity, and the lowest pH recorded for this secreted acid is 0.8 [26]. Pancreatic fluid contains the enzymes trypsin and chymotrypsin, which digest proteins into peptides and amino acids in the duodenum at a pH of 7.5 to 8.5 at 37°C [27]. The effect of AFB1 mutagenicity on human digestion has not been clarified; our aim in this study was to determine the changes in AFB1 mutagenicity due to pH using an in vitro model of human digestion. | |
Materials and Methods |
|
| In vitro model of human digestion. Ingredients, enzymes and the pH naturally present in the digestive tract were used. Two maize sources were used to make tortillas, one was naturally contaminated with AFB1 (Figure 1A), and the other was uncontaminated ground maize but spiked with AFB1 standard (Sigma-Aldrich, Mexico) at dough stage (32 μg AFB1 + 390 mL MeOH/H2O, 80:20 v/v) (Figure 1B), homogenizing the mixture with care. Tortillas from both sources were ground and subjected to a simplified in vitro model of human digestion following the protocols of physicians from the Institute of Pediatrics Health Ministry of Mexico: | |
 |
Figure 1: Model of human digestion of tortillas in vitro processed from: A) Maize naturally contaminated by AFB1. B) Tortilla prepared from 1000 g of clean maize dough spiked with 32 µg of AFB1 and treated in a model of human digestion. IAC = immunoaffinity column; DMSO = Dimethylsulfoxide; PBS = Phosphate buffered saline; MeOH = Methanol. |
| a) Control tortilla with AFB1. | |
| b) To resemble the model of human digestion in the mouth, 250 mL of natural saliva with ptyalin was collected from two persons fasted for 12 hr, adjusted to pH 7.0, added to the mash that were 3 replications of 75 g of ground tortilla (Figure 1), and the mixture was incubated for 2 min. Complete artificial saliva with α-amylase is not recommended in in vitro toxicology research [28], so we used the natural saliva. Natural saliva is watery substance located in the mouth, secreted by salivary glands and composed 99.5% by water, and 0.5% consists from electrolytes, mucus, glypoproteins, enzymes and antibacterial compounds such as secretory IgA and lysozymes [26]. | |
| c) For a model of digestion in the stomach, the mash was used plus 250 mL of simulated gastric juice with 0.032 % pepsin solution and 2.0 g of NaCl (Merck) and 3.2 g of pepsin (Sigma-Aldrich, Mexico) in 7 mL of HCl at 37.4% (J.T. Baker). The volume was increased to one L with distilled water (H2Od), and adjusted to pH 1.2. This mixture was incubated for 2 hr in the dark to resemble the digestive process in the stomach. | |
| d) To simulate digestion in the intestines, tortilla with AFB1 plus pancreatic fluid (250 mL) was prepared by diluting 6.8 g of KH2O4 (Merck) in 250 mL of H2Od. Then, 190 mL of 0.2 N NaOH (Merck), 400 mL of H2Od and 10.0 g of pancreatin (Sigma-Aldrich, Mexico) were added and mixed, and the solution was adjusted to pH 7.5 ± 0.2 with NaOH and incubated for 5 hr in darkness (Figure 1). | |
| These same digestive fluids were tested in several combinations: | |
| e) To simulate sequential digestion in the mouth and stomach, the mash was mixed with 83 mL of saliva (pH 7.0) incubated for 2 min, then supplemented with 83 mL gastric juice (pH 1.2) for 2 hr in the dark at a final pH of 1.8. | |
| f) To simulate the passage of mash from the stomach to the intestine, a tortilla with AFB1 was mixed with 83 mL of gastric juice (pH 1.2) incubated for 2 hr, then blended with 83 mL of pancreatic fluid (pH 7.5), and incubated for 5 hr at a final pH of 2.4. | |
| g) To simulate complete digestion of the maize tortillas from mouth to gut, the spiked tortilla was incubated for 2 min with 83 mL of saliva (pH 7.0), mixed with 83 mL of gastric juice (pH 1.2) and incubated for 2 hr, then mixed with 83 mL of pancreatic fluid (pH 7.5) for 5 hr in the dark at room temperature and a final pH of 5.8. | |
| The method for AFB1 extraction has been described previously [29]. Quantification of AFB1 was carried out by high-performance liquid chromatography (HPLC) [30], and its mutagenicity was determined by the Ames test [31-34]. | |
| Maize naturally contaminated with AFB1. Maize from the warehouse `Las Yescas´ in Valle Hermoso, State of Tamaulipas, Mexico, was sieved and used to make tortillas. Of the maize used, 98% contained ≥ 20 μg/kg AFB1, with an average of 66 μg/kg [35]. In vitro models of human digestion were applied. AFB1 was quantified by HPLC, and the mutagenic effect of one of ten tortillas made with naturally AFB1-contaminated maize was tested in triplicate by the Ames test (Figure 1A). | |
| Maize dough spiked with AFB1. As described in Figure 1B, because it is not normally present during digestion, methanol (MeOH) was evaporated under an extraction chamber for 24 hr. Tortillas were then prepared, dried, ground and digested by the in vitro human digestion models (Figure 1B). To counteract the pH effect two different extraction assays and column washes were performed, with either H2Od or PBS. Distilled water maintains the natural pH of the digestive fluid solutions used, which ensures that the in vitro digestion model approximates human digestion. In the second assay, ground dry tortilla was blended with 50 mL MeOH/ PBS at 0.25 M (80:20 v/v) to compare the role of the pH in the human digestion model. PBS is not normally present and neutral pH can affect the mutagenicity of AFB1. The two assays were independently filtered and evaporated to 8 mL. Later, the AFB1-spiked tortilla, or mash from both extraction assays were subjected to the in vitro models of human digestion. | |
| Chemical extraction | |
| Two extraction assays were done by blending dry ground tortilla (25 g) + 1.25 g NaCl in two independent diluting solvents: a) 50 mL MeOH/ H2O (80:20 v/v) and b) 50 mL MeOH/PBS at 0.25 M (80:20 v/v) (Figure 1). | |
| Both digested mixtures were later filtrated and monitored for the presence of AFB1 by applying each treatment to separate immunoaffinity columns (IAC) previously equilibrated with 20 mL of 0.25 M PBS at pH 7.4. The IACs were washed with one of two different solvents: a) H2Od, or b) 10 mL of 5% Tween 20-PBS at and later with 10 mL PBS (pH 7.4). | |
| AFB1 was eluted with 1.5 mL of HPLC MeOH and 1.5 mL of H2Od, and the resulting 3 mL of eluate was then dried at 40°C in a Lab-Line Ambi HI-LO Chamber (Novatech, USA). Dry eluate was resuspended in 600 μL of MeOH and divided equally into two vials. One vial was used for HPLC chemical analysis. From this sample, 100 μL were derivatized as previously described [36], and triplicates of 30 μL were quantified by HPLC analysis [30]. | |
| The second vial was dried at 40°C, then resuspended in 100 μL dimethyl sulfoxide (DMSO) for to the Ames test with Salmonella typhimurium strain TA-98 (Figure 1B). | |
| HPLC analysis was validated and the conditions were a Series 400 pump, LC-10 Fluorescence detector (Exc362 nm, Emis450 nm), LCI-100 Laboratory computing integrator, Lambda 3A UV/Vis spectrophotometer (Perkin Elmer) and HS-10 C18 Phenomenex column 0258-0172. The mobile phase was H2Od /ACN/ MeOH (65:15:20 v/v/v) with a flow rate of one mL/min for 20 min. AFB1 standard (Sigma-Aldrich, Mexico) was included, and all samples were analyzed in triplicate. | |
| Ames test for mutagenicity: The Ames test was used to determine the effect of pH on the mutagenicity of AFB1 by microsomal S9 activation. Salmonella typhimurium TA98 [hisD 3052, gal, (chl, uvrB, bio) rfa, pKM101 (MucA/B Amp)] was donated by Bruce Ames to the Experimental Oncology laboratory at the National Institute of Pediatrics, Ministry of Health, Mexico. | |
| Spontaneous reversion test: Top agar (2 mL) mixed with 100 μL of wild type (Wt) or S. typhimurium TA98 bacteria in minimum glucose medium (MGM) was incubated for 48 hr. Reverted colonies (rc) were counted with an electronic colony counter (New Brunswick Scientific). Reference values for spontaneous reversion of the control strain TA98 without S9 were 20 to 50 rc [34,37]. Three colonies that showed the genetic markers were chosen as stocks kept on plates at 4 °C and in cryovials at -80°C. | |
| Cryovial stock: TA98 (200 μL) from an overnight culture was gently homogenized in sterile cryovials with 1800 μL DMSO (Nalgen NY EUA) at 4°C for one hr, frozen at -20°C and stored at -80°C prior to use. To start an overnight culture, a frozen sample from a cryovial stock was inoculated in Oxoid 2 nutritive medium [38]. | |
| Master stock plate S. typhimurium TA98: Bacteria from an overnight culture were plated on agar enriched with ampicillin and incubated at 37°C for 48 hr. Following incubation, the plates were allowed to cool, flipped upside down, wrapped in aluminum foil and then frozen. | |
| Verification tests of genetic markers of S. typhimurium TA98 were performed in comparison with those of the Wt strain [37]. The tests performed were: spontaneous reversion, dependence on histidine, dependence on biotin, dependence on histidine/biotin; sensitivity to crystal violet; presence of the rfa mutation marker; sensitivity to UV light (deletion of the gene uvr-B-bio) and verification of the presence of the plasmid pKM101 [31, 39-41]. | |
| Verification of S. typhimurium TA98 with mutagenicity controls: TA98 bacteria (100 μL) from surface agar was mixed with 100 μL AFB1, diluted in 500 μL DMSO and S9 microsomal extract, vortex homogenized and plated in triplicate on MGM. Plates were incubated inverted for 48 hr at 37°C, and rc were counted. S. typhimurium TA98 was verified with negative controls for mutagenicity that included surface agar (1.9 mL) with TA98 night culture (100 μL), and positive controls for mutagenicity with and without metabolic activation: surface agar (1.3 and 1.8 mL) + 2-aminofluoren, benzo (α) pyrene (100 L) + TA98 night culture (100 L) + S9 (100 and 500 μL). The direct treatments without metabolic activation were surface agar (1.8 mL) + 100 μL of AFB1 (10, 100 and 500 μg/mL) + TA98 (100 μL) and direct treatments with metabolic activation of surface agar (1.3 mL) + 100 μL of AFB1 (10, 100 and 500 μg/mL) + TA98 (100 μL) + S9 (500 μL). | |
| Controls for in vitro human digestion model: Controls were AFB1 standard (150 μL) incubated in 500 μL of the in vitro human digestion model solutions described above; lime (2% Ca(OH)2, pH 12.0, incubated for 40 min and boiled for 15 min), and regulating solution (potassium acetate/acetic acid, pH 5.0, incubated for 5 hr). All samples were dried and processed six times in triplicate for HPLC analysis, and three replicates were dissolved in 100 μL DMSO for the Ames test. | |
| AFB1 standards with enzyme-free digestive solutions: AFB1 was tested as follows: AFB1 in H2Od; AFB1 in pepsin-free gastric solution, pH 1.2, for 2 hr; AFB1 in pancreatin-free pancreatic fluid, pH 7.5, for 5 hr; and AFB1 with the pepsin–free gastric solution for 2 hr followed by pancreatin-free pancreatic fluid for 5 hr at pH 2.4, solutions were then dried. A second assay was performed in which H2Od was replaced with PBS, and samples were dried before treatment with the digestive solutions to maintain the pH of digestion. | |
| Testing of digested tortilla samples by Ames test: The digested samples were dried, resuspended in 100 μL DMSO and analyzed in triplicate to determine the level of mutagenicity. AFB1 (1.6 to 150 ng/mL) in DMSO was used as a positive control with S9 induced by Aroclor 1254 and NADP. Negative controls, applied alone, were: H2Od, PBS, S-9 microsomal rat concentrate, saliva, gastric fluid, pancreatic fluid, saliva + gastric fluid, saliva + pancreatic fluid, saliva + gastric + pancreatic fluids, AFB1 + NaCl + HCl, AFB1 + KH2PO4 + NaOH and AFB1 + NaCl + KH2PO4+ NaOH. | |
| Statistical analysis | |
| Standard deviations (SD) and percentage of relative standard deviation (% RSD) were calculated in revertant colonies from maize tortillas naturally contaminated with AFB1 (μg/kg), in the presence and absence of digestive fluids, controls without tortillas and from tortillas made from dough spiked with AFB1. | |
| A two way analysis of variance was performed using the different treatments (and controls) as one factor and the blended of the tortillas, as the second one, plus the interaction. After fitting the model test differences of means with Bonferroni corrections were used in all the pairs of treatment by blended. | |
Results |
|
| Validation of HPLC method for AFB1 quantification (12). The limits of detection (LOD) and quantification (LOQ) were LOD=0.5 μg/kg and LOQ=1.4 μg/kg, the coefficient of variation repeatability percentage was 1.8%, the equation of regression of the calibration curve (lineality) was y=4.367x – 2.874 with R2=0.999. | |
| The recovery percentage of AFB1 to determine the exactitude of the method was done with three spiked amounts of AFB1 (9.9, 19.9 and 29.7 μg/kg), the recovered AFB1 corresponding to each amount were 8.6, 17.2 and 28.4 μg/kg, the corresponding recovered percentages were 86.4, 86.5 and 95.4 %, the standard deviations were ± 1.5, ± 0.8 and ± 1.2, the percentages of the coefficients of variation were 1.8, 0.9 and 1.2 % and the confidence intervals were 82.6 to 90.3, 84.6 to 88.4 and 92.5 to 98.3. | |
| Spontaneous reversion of S. typhimurium TA98 | |
| The strain TA98 (INP B), with an average of spontaneous reversion of 24.33 rc was chosen because it provided better results in the verification tests of genetic markers than the other strains. Because a three-fold increase in rc over controls constituted a positive Ames test result, 73 rc was the lowest limit of detected mutations that could be considered positive. | |
| Naturally AFB1-contaminated maize in the model of human digestion | |
| Most of the AFs were eliminated in the residual limed water, and although 11 μg/kg of AFB1 were found in the dough by HPLC, no mutagenic effect was observed, Figure 2. A cytotoxic effect was present in only some replicates, shown by cell deformations or ruptures; thus it is possible that proteins or lipids from the maize inhibited AFB1 mutagenicity, because the CYPs are involved in converting AFB1 to electrophiles, e.g. diol-epoxide, responsible for forming DNA adducts and causing mutagenesis, although lipid oxidation cannot be considered in the stomach separately. The average concentration of AFB1 in tortillas made with naturally contaminated maize is given in Table 1. No AFB1 was detected in ground, limed maize because the lactone ring of the AF structure opened as a result of alkaline hydrolysis, which in turn caused a loss of fluorescence [17]. When this ground limed maize was kneaded into dough, 1.2 ng/g of AFB1 appeared. | |
|
Figure 2: Lack of mutagenic effect of tortillas (pH 11.6) made from AFB1 naturally contaminated maize by Ames test. Extract from one tortilla digested with pancreatic fluid at 7.5 pH by HPLC (A= 12.8 ng AFB1 and 43 revertant colonies) or ten tortillas (B= 128 ng AFB1 and 49 revertant colonies) without mutagenic effect. |
|
Table 1: Table 1: Revertant colonies from maize tortillas naturally contaminated with AFB1 (µg/kg), in the presence and absence of digestive fluids. |
| The amount of AFB1 (5.3 ng/g) detected in cooked tortillas increased in the presence of saliva (11.6 ng/g) or pancreatic fluid (12.8 ng/g). Similar amounts of AFB1 (5.3 ng/g) were found with tortillas alone at pH 11.6 and with gastric fluid at pH 1.2 (5.7 ng/g); therefore, both alkaline and acidic pH inhibit AFB1. Saliva (pH 7.0) activated and increased AFB1 to twice the amount found in tortilla alone (11.6 ng/g). Tortillas made with naturally AFB1-contaminated maize lacked mutagenicity in one and the addition of 10 tortillas extracts, and only their digestion with pancreatic fluid (pH 7.5) recovered AFB1 (12.8 μg/kg) (Table 1). | |
| Mutagenic effect of controls, without tortillas. In the AFB1 experiment without tortillas (Table 2), we observed no mutagenicity in the controls with AFB1 standard alone or enzymatic solutions at an alkaline or acidic pH without AFB1. | |
|
Table 2: Effect of controls, enzymatic solutions and treatment with AFB1 standard, without tortillas, in the human digestion model. |
| Treatments in a neutral pH with AFB1 standard in the presence of saliva (pH 7.0), pancreatic fluid (pH 7.5) and saliva + gastric juice + pancreatic fluid (pH 5.8) caused a mutagenic effect, (Table 3). Therefore, pH can be considered a key factor along with the enzyme itself in producing mutagenicity. We found a mutagenic effect of AFB1 with saliva (254 rc, pH 7.0) and with pancreatic fluid (115 rc, pH 7.5). The gastric juice (25 rc, pH 1.2) inhibited AFB1 mutagenicity and acidified saliva (pH 1.8) and pancreatic fluid (pH 2.4). | |
|
Table 3: Mutagenic changes in AFB1 owing to pH in tortillas treated in the human digestion model, made from dough spiked with AFB1. |
| Mutagenic potential of tortillas made from AFB1-spiked dough, in the model of human digestion. Positive results only occurred with tortillas made from dough spiked with AFB1, shown in Table 3. | |
| The dilution liquid used, either H2Od or PBS, influenced both pH and mutagenicity. Saliva with a neutral pH of 7.0 with H2Od gave a more AFB1-based mutagenic response (254 rc) than saliva with PBS buffer (pH 6.9, 112 rc) (Table 3), a result consistent with a previous study [17]. However, saliva did not protect, detoxify or attenuate the mutagenicity of AFB1 as previously reported [22]. Complete artificial saliva (CAS) with α-amylase is not recommended because artificial saliva alters expression of proinflammatory cytokines in human dermal fibroblasts, so it is important to carefully evaluate the “vehicle effects” of CAS and its components in in vitro toxicology research [28]. The mutagenicity of AFB1 was inhibited when saliva was mixed with gastric fluid in H2Od (pH 1.8, 20 rc). When the pH increased to 6.1, due to the use of PBS, the activity of AFB1 recovered (63 ng/g), and the mutagenic effect was restored (99 rc). These results again demonstrate that the role of pH on the mutagenicity of AFB1 is important. | |
| The greatest recovery of, and increase in, AFB1 mutagenicity was with saliva (pH 7.0), followed by pancreatic fluid at pH 7.5, which resembles the pH present in the duodenum, (Figure 3). Similar results were observed for AFB1 with the combination of saliva, gastric and pancreatic fluids at pH 5.8, whereas each of these digestive fluids alone did not render the maize mutagenic. The mutagenicity of AFB1 did not recover at pH of 1.2, as found in the gastric fluid dissolved in H2Od, but it was restored by pH 6.0 when the gastric juice was dissolved in PBS. | |
|
Figure 3: Effect of pH, from the in vitro human digestion model, over the AFB1 mutagenicity in tortilla made from AFB1 spiked dough, shown by revertant-mutated colonies (rc) of Salmonella typhimurium: A) Tortilla without AFB1 as negative control pH 11.6 with 25 rc; B) Tortilla with lime and AFB1 pH 11.6 with 53 rc; C) Tortilla with saliva pH 7 with 254 rc; D) Tortilla with gastric fluid pH 1.2 with 25 rc; E) With pancreatic fluid pH 7.5 with 115 rc; F) With saliva and gastric fluid pH 1.8 and 20 rc; G) With gastric and pancreatic fluids pH 2.4 and 51 rc; H)With saliva, gastric and pancreatic fluids pH 5.8 with 184 rc. |
| The effect of pH in human digestion with water, which is the normal solvent, increased mutagenicity. A neutral pH (6-7) favored the mutagenicity of AFB1 while acidic or basic conditions inhibited it (Figure 4). The exception was with the pancreatic fluid with H2Od (pH 7.5), which produced a positive mutagenic result (115 rc), and with PBS buffer where although a pH 7.0 was obtained, only 66 rc were produced. These results demonstrate that pancreatic fluid itself, and not only pH, plays a role in mutagenicity. | |
| When PBS was used as the dilution solvent, the resulting pH of 6.0 was in the neutral range and the mutagenicity of AFB1 was restored to 121 rc in the gastric juice step. PBS diminished the effect of saliva on mutagenicity, from 254 rc to 112 rc. However, saliva dissolved in both H2Od and PBS was mutagenic (Table 3). A pH dose-response curve is displayed in Figure 4. | |
|
Figure 4: Dose response curve of the relationship between the pH in the human digestion model and the mutagenicity of AFB1 (revertant colonies of Salmonella typhimurium) in the Ames test. |
| Statistical analysis | |
| Table 3 included the statistical analysis of the count of the reverted colonies, there was no influence of the solution fluid shown when tortillas were blended with methanol or distilled water that were not statistical significant different from the control made from AFB1 spiked dough (c), and treatments with gastric + pancreatic fluids (pylori, duodenum) (c), as well as for the tortillas digested with the gastric fluid (stomach) (d) and the control (without AFB1) (d). Whereas the case of tortillas blended with methanol/PBS there were not significant differences between control tortillas made from AFB1 spiked dough (f), with pancreatic fluid (duodenum) (f) and with gastric + pancreatic fluids (pylori, duodenum) (f). All the others differences of means were statistically different (p < 0.05). | |
Discussion |
|
| Although the absorption of the AF mainly occurs in the intestine, AFB1 mutagenicity of the chewed maize tortilla bolus, can change in the different organs where the contaminated food passes or distributes and the accumulation of the AFB1–FAPY adducts in the same place can initiate a cancer [10,12]. Cytochrome P450 CYP1A2 is responsible for the metabolism of various xenobiotics, and is related to the mutagenicity of AFB1 [42]. Cytochrome P450 CYP3A4 is involved in the oxidation of wide range of substrates, and cytochrome P450 CYP2B1 converts AFB1 into mutagenic metabolites [43]. These enzymes convert AFB1 into a genotoxic epoxide that forms N7- guanine DNA adducts that were studied with the Ames test [44]. Low hepatic glutathione S-transferase (GST) and increased hepatic AFB1-DNA adduction correlate with hepatocyte proliferation and contribute to increased tumorigenicity of AFB1 in newborn and partially hepatectomized mice [45]. | |
| The human digestion model may provide means to assess physical and chemical changes of the food and AFB1 when they pass through the digestive tract. However, metabolic activation or mutation occur only when the toxin is absorbed into the body and reached to the site of action. About the susceptibility to AFB1 adduct formation, there are different organ responses. Liver and kidney might be the probable target organs for AFB1 with the highest formation and persistence of DNA adducts [46]. There is a extrahepatic bioactivation of AFB1 in fetal, infant and adult rats in the mucosas of the nasopharynx, some glands in the nose, and of the glandular stomach trachea, bronchioles, colon and caecum, but not in the small intestine, oesophagus or Harderian gland [47]. The nasal olfactory mucosa had a much higher capacity than the liver to form AFB1-metabolites which bound to DNA and protein [47]. Several of the extrahepatic tissues able to bioactivate AFB1 have been reported to be targets for the carcinogenicity of the substance. The extrahepatic carcinogenicity of AFB1 is correlated to a local bioactivation in the sensitive tissues [47]. Liver is by far the AF target organ and the induction of hepatocellular carcinoma [48]. AFB1-DNA binding was observed in both liver cell and was 3 to 5 fold higher in parenchymal cells than in non-parenchymal cells [48]. In contrast no tumours arising from the sinusoidal cell population have been reported after exposure to AFB1. AF activate mainly in the liver, but also in other organs such as lungs [49,50], small intestine [51], nasal mucosa [47], etc. Donnelly et al. [50] investigated AFB1 bioactivation, the role of enzymes, detoxification and pulmonary susceptibility to AFB1 in human lung tissue obtained from patients undergoing clinically indicated lobectomy. Appropriate pharmacokinetic model (including the rate of absorption and metabolism) to assess the mutagenic potential of the AFB1 absorbed at each organ, discussing the role of pH in the mutagenicity of the toxin in each part of the digestive system is a necessary research to clarify AFB1 role in the carcinogenesis. The space in the digestive system in the in vitro model is considered outside, not inside of the body and there might be some changes in the toxicological standpoint. Although stomach and intestine were tested separately in the model a sequential digestion would be plausible, therefore the mixture of enzymes and pH of the different steps (saliva with pepsin pHs, gastric and pancreatic fluids pH’s and all of them) shows a closer approach to real human digestion. Microsomal cytochrome P450 has a minor role in the bio activation of AFB1 in human lung. AFB1 is a pulmonary carcinogen in experimental animals, and epidemiological studies have shown an association between AFB1 exposure and lung cancer in humans [50]. | |
| Intracellular AFB1 adducts are formed in the small intestine, and this reflects, at least in part, the catalytic activity of cytochrome P450 (CYP) 3A enzymes. The small intestine may not play a significant role in the metabolism of AFB1 [51]. Because these AFB1 adducts should ultimately pass in stool, enterocyte CYP3A may represent a regulatable barrier to dietary AF [52]. | |
| Short-term tracheal explant cultures from the rabbit were used to study the metabolism of AFB1 and to determine the cell types that are susceptible to damage by AFB1 and their relative contents of monooxygenase enzymes [53]. The conclusions of these studies were: a) rabbit tracheal explants are able to metabolize AFB1; (b) the nonciliated secretory cell population in this tissue is the target cell for cytotoxicity of this carcinogen; and (c) as is the case in the more distal airways, the nonciliated epithelial cells appear to have a high content of components of the pulmonary cytochrome P450 monooxygenase system, which may be an important factor in the susceptibility of these cells and this region of the airways to suspected airborne carcinogens [53]. | |
| AFB1-DNA adducts in human placenta (58%) and (9%) of adducts in cord blood, readily available specimens that respond to maternal environmental insult and are being used to investigate metabolism, bioactivation, and transplacental transfer of procarcinogens [54]. Thus, monitoring adduct levels in human specimens may provide information not only on carcinogen exposure but also on the relationship among infection with hepatitis B/C virus, dietary exposure to AFB1, and liver cancer. | |
| Dietary and caloric restriction has been recommended to reduce cancer risk [55-57]. Sixty percent dietary restriction reduced AFB1 metabolizing enzyme activity and decreased the AFB1-DNA adduct formation in young rats treated with AFB1. A protective effect due to caloric restriction reduced the metabolic activation of AFB1, this fact brought a decrease of AFB1-DNA binding by more than 50% [57]. Thus, the contributions of caloric restriction are the lower initial AFB1-DNA binding and less DNA damage, presumably by the less apurinic sites formed during the depurination process of AFB1-DNA adducts. Species and tissue specificities exist regarding the metabolic activation of AFB1 [57]. The effects of food restriction on the metabolic activation of AFB1 in rats and mice, which are AFB1-sensitive and -resistant rodent species, respectively has been reported [55]. In a comparison of food restriction and ad libitum food consumption treatments there was a reduction of metabolic activation of AFB1 in both rats and mice, causing formation of hepatic AFB1-DNA adducts to be 43% and 31% lower, respectively. | |
| Human gastric juice is an excellent medium for the oxidation of lipids. Cholesterol sulfate exhibits gastroprotective activity [16,58,59]; in the Ames test of the present study, its low pH inhibited the mutagenicity of AFB1, producing a negative result with 25 rc. However, when gastric juice was dissolved in PBS, the pH increased to 6.0, and the mutagenicity was restored (121 rc). | |
| The present study demostrated that AFs from tortillas, although not detectible by fluorescence, are still present in gastric and pancreatic fluids and their chemical structures and toxicity can be restored. The action of HCl (pH 1.2) alone on maize following lime processing of tortillas did not yield greater levels of AF. Our results demonstrated that there was a 60-76% recovery of AFB1 with digestive solutions in the pH range of 5.8-7.5, data found in agreement with Price and Jorgensen [17] but not with an earlier study [60]. | |
| Traditional lime treatment does not appear to be an adequate process for AF detoxification. AFB1 can produce a mutagenic effect in both the mouth and duodenum, as assays with and without tortillas demonstrated (Tables 3 and 4). Lime treatment blocked the mutagenic effect but did not prevent risk of AFB1 ingestion. A neutral pH, but not an acidic pH, restored the mutagenic effect, as was previously reported [60]. Incubation with pancreatin or pepsin did not affect the amount of AF associated with casein [61]. Pancreatin causes a reduction in the digestibility of maize and an increase in AF content [62]. | |
| The acidity of gastric fluid was suitable for forming nitrosamines, a potent carcinogen for many animal species [63]. However, acidic pH inhibited the mutagenic effect of AFB1. It seems that stomach cancer is more closely associated with nitrosamines, and AFs with colorectal cancer [10], but the etiological role of AFB1 in stomach cancer remains unclear because there are multiple changes in pH during digestion. Further studies on AFB1-DNA adducts in stomach cancer DNA are needed to clarify this issue. Pepsin has no effect in mutagenicity, requiring acid pH to act properly. In contrast, chloride lowers the pH and inhibits the genotoxic properties of AFB1, in agreement with previous findings [64]. The ingestion of maize may cause several mutations, but the human body has protective mechanisms to limit them. In susceptible individuals eating tortillas as a staple food, the risk for cancer increases. | |
| Many phytochemical compounds from vegetables and fruits can block mutagenic and carcinogenic processes [65-67]. Natural substances in maize tend to protect humans against AFB1 mutagenicity and anti-mutagenic activity of extracts from natural maize have been reported [68]. The linoleic acid (9,12 acid– octadecanoic) found in maize and tortillas, as well as a the compound that could contain linoleic acid in its structure, have been reported as anti-mutagenic and antioxidant compounds in AF-contaminated maize [69]. These compounds might protect humans exposed to high amounts of AFs in their diet by blocking the mutagenicity of AFs. Tortillas can contain these compounds and therefore continue acting as anti-mutagens when AFs are extracted. Other reports recommend the control of toxigenic Aspergillus flavus and aflatoxin accummulation in peanuts using a native atoxigenic A. flavus strain based on competitive exclusion of the toxigenic strains [70]. | |
Conclusion |
|
| The pH of each digestion step plays an important role in inducing or inhibiting the mutagenic effect of AFB1. Tortillas prepared from AFB1-spiked dough had a mutagenic effect in the Ames test, but this effect was not observed in tortillas prepared with maize naturally contaminated by AFs. The alkaline pH (11.6 to 12.0) of the lime treatment of tortillas inhibited the mutagenic effect of AFB1, whereas the neutral pH of both saliva (7.0) and pancreatic fluid (7.5) reactivated AFB1 mutagenicity. The acidic pH 1.2 of gastric juice inhibited AFB1 activity, but using PBS to increase the pH to 6.0 restored the mutagenicity of AFB1. This research shows also that the pancreatic fluid has a role in activating the mutagenicity of AFB1, not only the pH. | |
Acknowledgments |
|
| The authors thank the Instituto de Biología, Universidad Nacional Autónoma de México, which supported this work. We also thank the Universidad Autónoma de San Luis Potosí for the PhD scholarship Supera from Promep given to María de Guadalupe Moctezuma. Thanks also to IBUNAM´s personnel: Noemí Chávez from the Secretaría Técnica, Carmen Loyola, Joel Villavicencio, Jorge López, Alfredo Wong, Diana Martínez and Julio César Montero provided valuable assistance with imaging, computer analysis and design. Additionally, we thank Georgina Ortega Leite and Gerardo Arévalo for library information. | |
References |
|
|
|
