 |
Intragastric formation and modulation of N-nitrosodimethylamine in a dynamic in vitro gastrointestinal model under human physiological conditions
Cyrille A. M. Krul ,  , a, Marco J. Zeilmakerb, Ronald C. Schothorstb and Robert Havenaara
a TNO Nutrition and Food Research, PO Box 360, 3700 AJ, Zeist, The Netherlands
b National Institute of Public Health and the Environment (RIVM), PO Box 1, 3720 BA, Bilthoven, The Netherlands
Received 27 May 2003; accepted 11 August 2003. ; Available online 30 September 2003.
Abstract
Human exposure to carcinogenic N-alkylnitrosamines can occur exogenously via food consumption or endogenously by formation of these compounds through nitrosation of amine precursors. Information on the intragastric formation of NDMA from complex mixtures of precursors and inhibitors in humans is not available. In this study the formation of N-nitrosodimethylamine (NDMA) has been quantitatively analysed in a dynamic in vitro gastrointestinal model, in which gastric conditions can be modulated and closely simulates the physiological situation in humans. Substantial amounts of NDMA were produced when nitrite and dimethylamine or codfish were simultaneously introduced into the model. However, humans are gradually exposed to nitrite by the intake of nitrate-containing food. Nitrate secreted in saliva is converted to nitrite by oral bacteria. To mimic the human exposure to nitrite in a realistic way, nitrite was gradually added into the gastric compartment, simulating the swallowing of nitrite containing oral fluid after the intake of nitrate at the level of 0.110 times the ADI. Under these conditions, the cumulative amounts of NDMA formed were 2.3422  g NDMA and 1.842.7 g NDMA at a rapid and slow gastric pH decrease, respectively. Beside codfish, various fish species and batches in combination with nitrite, simulating the intake of for times the ADI of nitrate, were investigated. Herring, pollack and plaice were also able to induce NDMA formation. Mackerel, salmon and pike perch did not result in increased NDMA formation. Furthermore, the effect of nitrosation modulators on NDMA formation was investigated. Thiocyanate (2 m ) increased NDMA formation, but the increase was not statistically significant. In contrast, orange jus and tea effectively, but not totally, reduced the amount of NDMA formed in the gastric compartment. These experiments show that (1) the dynamic in vitro gastrointestinal model is an appropriate tool for mechanistic studies on the intragastric formation of nitrosamines, and (2) that the results obtained with this model are helpful in evaluating human cancer risk for the combined intake of codfish-like fish species and nitrate-containing vegetables.
Author Keywords: Gastrointestinal model; N-nitrosodimethylamine; Nitrite; Codfish; Ascorbic acid; Toxicokinetic model
1. Introduction
N-alkylnitrosamines are potent carcinogens. They induce tumours in a variety of animal species, such as rat, mouse, rabbit, fish and birds and in different organs such as the bladder, kidney, liver, oesophagus or stomach ([Peto et al., 1984 and IARC, 1987]). For the following reasons, it is assumed that N-alkylnitrosamines can induce tumours in humans as well. First because human exposure to N-alkylnitrosamines has been associated with increased risk for e.g. gastric, bladder and colon cancer ( [Bartsch et al., 1990, Knekt et al., 1999 and Mirvish, 1995]). Second because biotransformation and biological activity of these nitrosocompounds in animals appear to be similar to those in humans ( [Peuessman, 1990]). The IARC evaluated various N-nitrosamines as Group 2A (probably carcinogenic to humans), such as N-nitrosodimethyl-amine and N-nitrosodiethylamine, and as Group 2B (possibly carcinogenic to humans), such as N-nitrosodiethanolamine ( [IARC, 1978 and IARC, 2000]). However, epidemiological studies on N-alkylnitrosamines and cancer risk have yielded inconsistent findings (reviewed by [Eicholzer & Gutzwiller, 1998]).
Human exposure to N-alkylnitrosamines can occur via two pathways, namely via exogenous and endogenous routes (extensively reviewed by [Tricker 1997]; [Walker, 1990 and Shepard et al., 1987]). Exogenous exposure may result from consumption of food or beverages (e.g. beer), inhalation of tobacco smoke, the use of rubber articles (e.g. teats and soothes) and cosmetics, or from occupational exposure in the rubber industry ( [Lijinsky, 1999, Tricker & Preussmann, 1991, Gray & Stachiw, 1987 and Straif et al., 2000]). Endogenous exposure results from the nitrosation of precursors of nitrosamines in the human body. The nitroso moiety is derived from nitrite, which reacts with ingested secondary amines to form N-nitrosamines under acidic conditions. These conditions prevail in the stomach ( [Mirvish, 1995 and Leaf et al., 1989]).
A minor part of nitrite exposure occurs via nitrite-containing food, such as preserved meat and certain vegetables, but the concentration is generally very low ([Gangolli et al., 1994 and MAFF, 1992]). The main route of exposure is from the reduction of nitrate into nitrite by oral bacteria ( [Stephany & Schuller, 1980 and Walters & Smith, 1981]). Nitrate in food or drinking water is absorbed in the small intestine. Roughly 25% of the intake is actively secreted by the salivary glands into the oral cavity. Of this fraction approximately 20% (and thus 5% of dietary nitrate) is converted into nitrite by oral bacteria ( [Spiegelhalder et al., 1976]). This nitrite enters gradually the stomach after swallowing of oral fluid. At low pH, nitrite will be converted to nitrous acid and subsequently to N2O3, which reacts with secondary amines in its deprotonated form (e.g. from food sources such as fish), to give nitrosamines ([Mirvish, 1983]). The optimum pH for the nitrosation reaction, as determined in in vitro studies, is between 2.0 and 3.4, depending on the type of amine ( [Mirvish, 1975]). In addition, nitrosation has also been reported at neutral pH by nitrosating bacterial species ( [Mirvish, 1995 and Leaf et al., 1989]).
In order to assess the daily exposure to exogenous N-alkylnitrosamines, sensitive and precise gas chromatography (GC-) methods are currently available for the detection of N-alkyl-nitrosamines. In the Netherlands the dietary intake of the volatile N-nitrosodimethyl-amine (NDMA) is low, approximately 0.1 g per person per day ([Ellen et al., 1990]). The accuracy of the quantification of endogenous formation of N-alkylnitrosamines in the human body and the contribution of endogenously formed N-alkylnitrosamines to the total exposure to N-nitrosamines is still a matter of debate ( [Gangolli et al., 1994]). Matechematic models ( [Licht & Deen, 1998]) and in vivo studies with N-nitrosoproline (NPRO), formed out of  -proline and nitrite, have indicated that the endogenous exposure is possibly much less important than exogenous exposure ([Bartsch & Spiegelhalder, 1996]). However, NPRO is a non-carcinogenic nitrosamine, it is not metabolized and excreted almost completely in the urine ( [Oshima et al., 1981 and Shapiro et al., 1991]). It is not clear how to extrapolate this finding to the intragastric formation of carcinogenic N-nitrosamines.
Vermeer and colleagues demonstrated that the intake of nitrate at the level of the acceptable daily intake (ADI; 3.65 mg/kg bw) via drinking water or vegetables in combination with a meal containing different types of amine-rich fish, led to significantly increased urinary NDMA excretion (0.640.87 g/24 h) by human volunteers ([Vermeer et al., 1998 and Van Maanen et al., 1998]). [Spiegelhalder & Preussmann, 1985] assumed that at least 0.5% of the NDMA is excreted in the urine, which means that substantial amounts of NDMA are produced in the stomach.
Although it is indicated that NDMA can be formed in the stomach, only a crude estimate can be given of the actual amount. It could be argued that a more accurate estimate might be obtained by in vivo measurement of NDMA formation. However, for practical and ethical reasons a study of NDMA formation in the stomach of human volunteers is very difficult. We therefore used an alternative way to estimate the formation of NDMA in the human stomach, i.e. with a dynamic in vitro gastrointestinal model, consisting of four compartments that represent the stomach, duodenum, jejunum and ileum ([Minekus et al., 1995]). The model closely simulates the human physiological situation after the intake of food. In a previous publication, it has been demonstrated that this gastrointestinal model is a useful tool to study the availability and interaction of food mutagens (e.g. heterocyclic amines) and antimutagens, such as black and green tea extracts ( [Krul et al., 2000]).
In the gastrointestinal model the nitrosation reaction was investigated between nitrite and dimethylamine (DMA) and codfish under different gastric pH conditions: a slow and a more rapid pH decrease. Nitrosation of DMA was studied because DMA is often present at high concentrations in food (e.g. codfish) and the nitrosation product NDMA is one of the most potent carcinogenic N-alkylnitrosamines in rodents.
To simulate realistically the swallowing of nitrite-containing oral fluid that has been formed from nitrate by the microflora in the oral cavity, the formation of nitrite was quantified with the aid of a toxicokinetic model. With this model the flow of nitrite-containing oral fluid into the stomach was calculated for the intake of different levels of nitrate (0.110 times ADI) and incorporated in the gastrointestinal model. Beside codfish, we determined possible differences in NDMA formation for a variety of frequently consumed fish species, such as herring, mackerel, plaice, pollack and salmon.
Thiocyanate catalyzes the formation of nitrosamines, especially under acidic conditions ([Boyland & Walker, 1974]). Thiocyanate might be secreted by salivary glands or directly into the gastric juice ( [Dougall et al., 1995]). It has been established that endogenous NOC formation occurs at a higher level in smokers ( [Hoffman & Brunneman, 1983]). An explanation might be the higher level of thiocyanate present in saliva of smokers (5.5±0.2 m) compared to non-smokers (1.8±0.2 m) ([Walters et al., 1979]). We investigated the influence of thiocyanate (2 m) on the formation of NOC after the consumption of codfish, spinach and nitrite at the level of two times the acceptable daily intake (ADI) of nitrate.
On the other hand various antioxidants, such as ascorbic acid and polyphenols, showed inhibitory effects on the formation of nitrosamines ([Bartsch et al., 1988]). The intragastric ascorbic acid concentration is determined by consumption of fresh fruit and vegetables, and by active secretion of ascorbic acid by the gastric mucosa. Ascorbic acid inhibits NOC formation by a rapid reduction of nitrosating species, such as N2O3 into nitric oxide (NO). The ability of ascorbic acid to remove nitrite is however markedly reduced by the presence of oxygen, because NO can then react with oxygen to reform nitrite.
Experiments with respect to the inhibition of NDMA formation were performed with two inhibitors, viz. orange juice (as a source of ascorbic acid) and black tea (which contains large amounts of polyphenols).
2. Materials and methods
2.1. Chemicals and food samples
The digestive juices and enzymes used in the in vitro digestion model were lipase, (150 U/mg; Rhizopus lipases F-AP 15, Amano Pharmaceuticals) and pepsine (2100 U/mg; Sigma, Zwijndrecht, the Netherlands). N-nitrosodiisopropylamine (NDiPA) in methanol was supplied by Schmidt (Prochem). Nitrosodimethylamine (NDMA), dimethylamine and sodium nitrite were purchased from Sigma (Zwijndrecht, the Netherlands). All other chemicals were purchased from Merck (Darmstadt, Germany). Freshly pasteurised codfish fillet from the North Sea was obtained from a Dutch fish supplier and stored at -20 °C.
Lyophilized green tea and black tea were kindly provided by Unilever Research Vlaardingen (the Netherlands). Deep-frozen, chopped spinach (Iglo, the Netherlands) was obtained from a local supermarket. The various species of frozen fish, codfish, salmon, pollack, plaice and pike perch were obtained from fish suppliers and stored at below -18 °C. Smoked mackerel, and salted herring were not frozen, but stored overnight at 4 °C.
2.2. The dynamic in vitro gastrointestinal model
The TNO in vitro gastrointestinal model is a dynamic, multi-compartmental, computer-controlled system that mimics the physiological processes in the human stomach and small intestine ([Minekus et al., 1995 and Krul et al., 2000]). In the model the body temperature, pH, peristaltic movements and secretion of digestive enzymes, are simulated. These parameters and variables can be handled in a controlled and standardised way.
Artificial gastric juice containing 3.1 g/l NaCl, 1.1 g/l KCl, 0.6 g/l NaHCO3 and 0.15 g/l CaCl2.2H2O is added into the model at a rate of 0.5 ml/min. Lipase (500 mg/l) and pepsin (150 mg/l) are separately added into the gastric compartment at a rate of 1.0 ml/min. The model is made of glass jackets with flexible walls inside, which are surrounded by water of 37 °C. The water pressure squeezes the walls; its relaxation and compression ensure mixing and simulate the peristaltic movements. Valves mimicking the pyloric sphincter control the transit of the gastric content to the duodenal compartment. For the experiments described, the gastric compartment of the model was used with some minor modifications (Fig. 1).
 (9K)
Fig. 1. Schematic view of the gastric model: (1) gastric compartment, made of glass jackets, (2) flexible wall, surrounded by 37 °C water, to ensure peristaltic movements, (3) continuous measurement of the pH and addition of HCl to follow the preset pH curve, (4) addition of lipase and pepsin, (5) addition of nitrite simulating the swallowing of oral fluid, (6) valve system that mimics the pyloric sphincter, (7) compartment for the termination of the nitrosation reaction, (8) addition of NaOH, 9) collection of the samples on ice.
The pH in the gastric compartment of the model was continuously measured and accurately adjusted by the addition of 1 M HCl. This offered the opportunity to define every desirable pH curve, in combination with a programmed gastric emptying rate. In the first set of experiments two different physiological conditions were simulated. Namely, the condition after intake of a homogenised, low-caloric solid meal, with gastric pH set to decrease slowly and the condition after intake of a semi-liquid meal, in which the pH in the gastric compartment decreased more rapidly. The gastric emptying rate was the same in all experiments and simulated the gastric delivery of a low-caloric meal ([Moore et al., 1984]) ( Fig. 2).
 (5K)
Fig. 2. The pre-set pH curves in the gastric compartment simulating a rapid (squares), moderate (triangles) and slow (circles) gastric pH decrease. The gastric delivery rate (dotted curve) was the same in all experiments.
Three hours after the addition of food the emptying of the gastric compartment was complete for more than 95% (Fig. 2). During this period gastric delivery in 1h aliquots (01 h, 12 h, and 23 h) was collected on ice. The further reaction of the precursors nitrite and DMA to NDMA after collection of fractions from the gastric compartment was prevented by the immediate addition of 10M NaOH to reach a pH >10. The volumes of the 1 h fractions were measured and the samples were stored at <-70 °C until analysis of NDMA concentrations. Except in experiments described in Section 2.4.5, the fractions collected between 0 and 1 h and between 1 and 2 h were pooled. The pooled sample (02 h) and the 23 h fraction were analysed. The model was protected from UV light by UV-absorbing foil. All experiments were performed in duplicate, unless otherwise indicated.
2.3. Analysis of NDMA
NDMA was analysed in the gastric samples by gas chromatography in combination with a Thermal Energy Analyser detector (GC-TEA) as described by Pensabene, with some minor modifications ([Pensabene & Fiddler, 1994 and Dallinga et al., 2001]). Quantification of the NDMA content was based on the peak areas of NDMA and the internal standard (N-nitrosodiisopropylamine) using linear regression. The detection limit of the method was approximately 80 pg injected, which would correspond to an NDMA concentration of 1.6 g/l gastric juice. The cumulative amounts of NDMA formed, were calculated based on concentration in the samples and the volume of the delivered gastric content. For confirmation purposes a selection of extracts in which NDMA was detected, were irradiated for three hours with UV-light (254 nm). On subsequent GC-TEA re-analysis all peaks due to the presence of NDMA has disappeared. This means that the initial peak was indeed NDMA.
2.4. Experiments
2.4.1. NDMA formation measured with DMA and a fixed dose of nitrite
To investigate whether the formation of NDMA can occur under the experimental conditions of the in vitro model, the gastric compartment was loaded with 300 ml of 0.1M sodium citrate buffer (pH 6.8) containing 5 m DMA and 5 m sodium nitrite. In control experiments the model was loaded with 0.1 M sodium citrate buffer (pH 6.8) or 5 m DMA dissolved in 0.1 M sodium citrate buffer (pH 6.8), both without nitrite. These experiments were performed to verify that NDMA, possibly released from the rubber components of the gastric compartment or as contamination of the DMA solution, was not detected after passage through the model.
2.4.2. NDMA formation measured with codfish and a fixed dose of nitrite
Codfish was used to investigate whether amines such as DMA are released from a food matrix under the physiological gastrointestinal conditions as simulated in the in vitro model, and whether these amines can act as nitrosation precursors. The fish was pasteurised (20 min at 70 °C), cooled down to 10 °C within 1 h, and stored at -20 °C. Before use, 100 g of frozen codfish was thawed and thereafter heated in a microwave oven (55 s, 600 W) and then crushed and mixed with 5 m sodium nitrite in 0.1 m citrate buffer with a blender. The mixture was loaded into the gastric compartment. Endogenous NDMA was measured in a control experiment in which the model was loaded with codfish without nitrite.
2.4.3. NDMA formation measured with codfish and gradual addition of nitrite
To simulate the nitrosation of DMA under more physiological conditions, the nitrite solution was gradually added to the gastric compartment at a rate which mimics the swallowing of nitrite-containing oral fluid, after uptake of nitrate from food (0.110 times the ADI of nitrate). The rates were calculated with the aid of a toxicokinetic model, based on human data (see below). Experiments were performed with 5 m DMA and nitrite secretion simulating the intake of nitrate at 10 times its acceptable daily intake (ADI, 3.65 mg nitrate/kg body weight). Subsequently, experiments were performed with 100 g of codfish and transport of nitrite corresponding to the intake of 0.1, 1, 5 and 10 times the ADI of nitrate at a slow and rapid gastric pH decrease (except five times the ADI of nitrate, which was only performed under the conditions of a slow gastric pH decrease). This range represents the human exposure to nitrate, the median adult exposure being 1 mg/kg/day with a range of 0.120 mg/kg/day (10 times ADI corresponds to 36 mg/kg/day) ([Slob et al., 1995]). DMA and codfish, respectively, were loaded into the model at once at the start of the experiment.
2.4.4. NDMA formation measured with various fish species and batches
Frequently consumed fish species (89% of the total variants) were selected based on the Dutch food consumption bearing (19971998), to investigate the NDMA formation. Five batches each of codfish and herring, four batches each of salmon, mackerel, plaice, and pollack and one batch of pike perch were analysed. Before use, 100 g of fish was thawed, heated in a microwave oven (3 min, 600 W) and thereafter crushed and mixed with artificial gastric juice. Salted herring and smoked mackerel were not heated, because generally they are consumed cold.
Nitrite was gradually added into the gastric compartment, simulating the swallowing of oral fluid containing nitrite after the intake of five times the acceptable daily intake of nitrate. Control experiments were performed in which either one of the precursors for NDMA formation was omitted: 0.1 M sodium citrate buffer pH 6.8 with nitrite only, and herring without the addition of nitrite. In addition, one experiment was performed with pike perch, which is supposed not to contain DMA. Nitrite was added simulating the intake of 10 times the ADI of nitrate.
2.4.5. Modulation of NDMA formation, influence of thiocyanate and antioxidants
First 70 g codfish (prepared as described above) was introduced into the gastric compartment. Thereafter, to simulate the nitrosation of amines in codfish under more physiological conditions defrosted spinach (50 g) was mixed with the homogenised codfish and introduced into the model.
Secondly, in order to determine the effect of salivary thiocyanate on the formation of NDMA, experiments were performed in which codfish and spinach were added to the gastric compartment, with or without thiocyanate. Thiocyanate (2 m) was continuously added to the gastric compartment (6 ml/h) simulating the swallowing of oral fluid (saliva) containing thiocyanate. In total six experiments were performed with two batches of codfish.
Thiocyanate is normally present in saliva, therefore further experiments were performed in the presence of 2 m thiocyanate. The inhibition of NDMA formation was investigated by introducing codfish and spinach into the gastric model together with a putative nitrosation inhibitor. Codfish (70 g) and spinach (50 g) mixed with artificial saliva and gastric juice were added to the model with or without 80 ml of orange juice (containing 35 mg ascorbic acid). The same experiment was performed in which orange juice was replaced by 80 ml black tea (0.5 g lyophilized tea extract dissolved in boiling water). Furthermore, experiments were performed in which orange juice or black tea, was loaded into the model 1 h after the start, simulating the effect of the intake of inhibitors after the meal.
Nitrite was gradually added, simulating the intake of two times the ADI of nitrate.
2.4.6. Flow of nitrite-containing oral fluid, quantification based on toxicokinetic model
A toxicokinetic model of nitrate and nitrite has been developed and will be described elsewhere ([Zeilmaker et al.]). This model is based and verified on various studies in humans, including the time-course of the concentrations of nitrate and nitrite in plasma and oral fluid after the intake of nitrate-containing vegetables and drinking water.
The concentration curves of nitrite in oral fluid were calculated by the model simulating the intake of 0.1, 1, 2, 5 and 10 times the ADI of nitrate. The oral fluid concentration of nitrite increased approximately linearly, almost immediately after the intake of nitrate. A maximum was reached at around 1 h; thereafter the concentration of nitrite declines linearly (example 10 times ADI of nitrate, Fig. 3).
 (2K)
Fig. 3. Model simulation of the time-course of the concentration of nitrite in human oral fluid after exposure to 10 times the Acceptable Daily Intake of nitrate, i.e. 36.5 mg nitrate/kg body weight. X-axis: T=time in hours; Y-axis: Csano2=concentration of nitrite in oral fluid (m).
The swallowing of nitrite-containing oral fluid according to above calculated curves was mimicked by continuously infusing a solution of nitrite (32 m) into the gastric compartment of the gastrointestinal model during the 3 h experiment, at the following flow-rates: 1 ml/h for the first 15 min, 3.5 ml/h between 15 and 30 min, 6.9 ml/h between 30 and 45 min, 10.5 ml/h between 45 and 60 min, 11.7 ml/h between 60 and 120 min, and 10.7 ml/h between 120 and 180 min. Similar procedures were performed to simulate the intake of 0.1, 1, 2 and 5 times the ADI of nitrate, resulting in total nitrite flows during 3 h of 0.028, 0.072, 0.105, and 0.45 m (Fig. 4). In this way we simulated the swallowing of nitrite-containing oral fluid in humans after the intake of 0.110 times the ADI of nitrate.
 (3K)
Fig. 4. Model simulation of the total amount of nitrite swallowed with oral fluid after the exposure to 0.1 to 10 times the Acceptable Daily Intake of nitrate (0.3636.5 mg nitrate/kg body weight).
2.5. Statistics
To describe the doseresponse curve between the amount of nitrite added to the model and the amount of NDMA formed in the gastric compartment, the log likelihood ratio test was used to determine the parameters for the best fit of the data (P<0.05; Advanced Continuous Simulation Language, Edition 10.1, Mitchell and Gauthier Associates, Concord, MA 01742, USA). The Student t-test was used to evaluate the statistical significance of the observed differences between experiments with or without thiocyanate (P-value of 0.05).
3. Results
3.1. NDMA formation measured with DMA and a fixed dose of nitrite
In general, more NDMA was formed during the 3-h gastric passage when the gastric pH decreased rapidly, compared to the NDMA formation during the slow gastric pH decrease (Fig. 5). Under the conditions of a rapid gastric pH decrease the majority of NDMA was produced between 60 and 120 min after the start of the experiment. During this time interval the pH in the gastric compartment decreased from 2.5 to 1.7. When the pH decreased more slowly, most of the NDMA was formed between 120 and 180 min, during which time the pH decreased from 3 to 1.7. Relative high cumulative amounts of NDMA were produced (mean 128 g; range 113143 g) after the intake of 5 m nitrite and 5 m DMA, under the conditions of a rapid gastric pH decrease. Lower cumulative amounts of NDMA were formed (mean 39 g; range 3542 g) under the conditions of a slow gastric pH decrease. Under the conditions of a rapid gastric pH decrease, the formation of NDMA was only weakly affected by the concentration of its precursor DMA. A lower concentration of DMA (2.5 m) did not significantly change the cumulative amount of the NDMA formed compared to 5 m DMA (data not shown).
 (4K)
Fig. 5. The cumulative amount (mean±range; n=2) of NDMA formed in the gastric compartment after the intake of 5 m DMA and 5 m nitrite, performed under conditions of a rapid gastric pH decrease (white columns) and a slow gastric pH decrease (black columns). The dotted curves show the pre-set pH in the gastric compartment.
In the control experiments with 5 m DMA without nitrite, cumulative amounts of 7.8 and 14.3 g NDMA were measured, under conditions of slow and rapid gastric pH decrease, respectively. The explanation is probably contamination of the DMA solution with a small amount of NDMA. In the control experiments without any of the precursors 1.21.4 g NDMA was detected.
In conclusion, this first set of experiments showed that NDMA was produced in the gastric compartment when nitrite and DMA are introduced into the model as pure compounds.
3.2. NDMA formation measured with codfish and a fixed dose of nitrite
The addition of codfish as a source of amines appeared to influence the pH in the gastric compartment (Fig. 6). Consequently, the pH did not exactly follow the pre-set points of the curve for rapid gastric pH decrease (compare Fig. 6 and Fig. 2).
 (4K)
Fig. 6. The cumulative amount (mean±range; n=2) of NDMA formed in the gastric compartment after the intake of 100 g of codfish and 5 m nitrite, performed under conditions of a rapid gastric pH decrease (white columns) and a slow gastric pH decrease (black columns). The curves show the pH measured in the gastric compartment.
Higher amounts of NDMA were formed when codfish was used instead of 5 m DMA together with 5 m nitrite. The cumulative amount of NDMA produced was 677 g (range 669684 g) and 109 g (range 90128 g), under conditions of a rapid and slow gastric pH decrease, respectively. Apparently DMA from codfish became available later in time, at the time the pH reached a level that favours the nitrosation reaction, and thus resulted in higher amounts of NDMA (compared to the DMA solution).
In the control experiments in which codfish was added to the model without nitrite 1.6±1 g (n=3; range 0.92.8 g) NDMA was produced (Table 1). Apart from NDMA, a very small amount of nitrosodiethylamine (NDEA) was found in the GC-TEA chromatogram of some of the samples derived from the in vitro model.
Table 1. The cumulative amount of N-nitrosodimethylamine (NDMA) formed in the gastric compartment of the gastrointestinal model after the intake of nitrosation precursors (n=2; except were no range is given)
 (<1K)
3.3. NDMA formation measured with codfish and gradual addition of nitrite
The gastric compartment was loaded with 5 m DMA while the addition of nitrite simulated the swallowing of nitrite-containing oral fluid after the intake of 10 times the ADI of nitrate. The mean cumulative amount of NDMA formed under the conditions of a rapid and slow gastric pH decrease was 106 g (range 65146 g) and 39 g (range 1958 g), respectively (Fig. 7). The cumulative amount of NDMA was equal to the amount formed after the intake of a mixture of 5 m DMA and 5 m nitrite at once (for both the rapid and slow gastric pH decrease). However, differences were seen in the time course of NDMA formation. Most of the NDMA was produced between 120 and 180 min. This is in contrast to the results of the experiments in which nitrite was added as a fixed dose, where NDMA formation occurred largely between 60 and 120 min (compare Fig. 5 and Fig. 7).
 (4K)
Fig. 7. The cumulative amount (mean±range; n=2) of NDMA formed in the gastric compartment after the intake of 5 m DMA and the gradual addition of nitrite (continuously from 0 to 3 h), simulating the swallowing of nitrite-containing oral fluid after the intake of 10 times the Acceptable Daily Intake of nitrate. The experiments were performed under conditions of a rapid gastric pH decrease (white columns) and a slow gastric pH decrease (black columns). The curves show the pH measured in the gastric compartment.
In the experiments with nitrite at the level of 10 times the ADI of nitrate in combination with 100 g of codfish, the mean cumulative amount of NDMA was 423 g (range 292554 g) and 43 g (range 4144 g) under the conditions of a rapid and slow gastric pH decrease, respectively (Fig. 8). These amounts of NDMA were less than produced in the experiments in which codfish and 5 m nitrite were given at once (Table 1).
 (4K)
Fig. 8. The cumulative amount (mean±range; n=2) of NDMA formed in the gastric compartment after the intake of 100 g of codfish and the gradual addition of nitrite (continuously from 0 to 3 h), simulating the swallowing of nitrite containing oral fluid after the intake of 10 times the Acceptable Daily Intake of nitrate. The experiments were performed under the conditions of a rapid gastric pH decrease (white columns) and a slow gastric pH decrease (black columns). The curves show the pH measured in the gastric compartment.
In addition, experiments were performed with 100 g of codfish and the addition of nitrite, simulating the intake of 0.1, 1 and 5 times the ADI of nitrate. All experiments were performed under the conditions of a rapid and slow gastric pH decrease (except for five times the ADI of nitrate, which was only performed under the conditions of a slow gastric pH decrease). At 0.1 and 1 time the ADI of nitrate, 2.3 and 16.5 g NDMA were formed, under the conditions of a rapidly decreasing gastric pH (Table 1). At 0.1, 1 and 5 times the ADI of nitrate, 1.8, 5.1 and 21.1 g NDMA were formed under the conditions of a slowly decreasing gastric pH ( Table 1).
3.4. NDMA formation measured with various fish species and batches
The cumulative formation of NDMA (mean±S.D.) in the gastric compartment during 3 h after the intake of various species and batches of fish (100 g portions) and the addition of nitrite simulating the intake of five times the ADI of nitrate, ranged from 0 to 235 g (Table 2). The mean amount of NDMA produced was the highest for codfish (194±32 g), followed by herring (76±29 g), pollack (32±21 g), and plaice (28±8 g). The cumulative amount of NDMA formed after introduction of samples derived from mackerel, salmon and pike perch was around the background level, i.e. the level of NDMA formed or present in the model without the introduction of one or both of the precursors (Table 2).
Table 2. The cumulative amount of N-nitrosodimethylamine (NDMA) formed in the gastric compartment of the gastrointestinal model after the intake of various fish species (100 g) combined with the addition of nitrite simulating the intake of five times the ADI of nitrate
(<1K)
The results obtained with codfish and the addition of nitrite simulating the intake of five times the ADI of nitrate, together with the results obtained with codfish and nitrite simulating the intake of 0.1, 1 and 10 times the ADI of nitrate, were used to analyse the doseresponse relation between the amount of nitrite added to the model and the amount of NDMA formed in the gastric compartment. Under the condition of a rapid gastric pH decrease, a non-linear model is to be preferred over a linear model in describing the doseresponse relation (log likelihood ratio test, P<0.05; Fig. 9).
 (2K)
Fig. 9. Doseresponse curve (y=0.72 x1.82+1.47) between the amount of nitrite added to the model and the amount of NDMA formed in the gastric compartment, simulating a rapid gastric pH decrease.
3.5. Modulation of NDMA formation, influence of thiocyanate and antioxidants
The mean cumulative formation of NDMA after the intake of 70 g of codfish and the addition of nitrite, simulating of the intake of 2 times the ADI of nitrate, was 63 g (range 5374 g). Simulating a more realistic meal intake with spinach (50 g) and codfish (70 g) in combination with nitrite simulating two times the ADI of nitrate, substantially lowered the NDMA formation (7.2±3.4 g; range 3.713 g, n=6). Previously, in a pilot experiment, we observed that spinach (75 g) added to codfish (100 g) and nitrite simulating the intake of 10 times the ADI of nitrate, substantially lowered the cumulative amount of NDMA from 423 g (range 291555 g) to 159 g (range 125189 g).
The role of thiocyanate was determined in an experiment performed under the same conditions as described above; 70 g of codfish, 50 g spinach and nitrite simulating the intake of two times the ADI of nitrate. When thiocyanate (2 m) was continuously added to the gastric compartment, simulating the swallowing of oral fluid, the total formation of NDMA was increased from 7.2±3.4 g (without thiocyanate) to 10.6±5.2 g (range 5.519.5 g, n=6). Although, the mean NDMA formation was increased with 47%, the difference was not statistically significant (Table 3).
Table 3. The cumulative amount of N-nitrosodimethylamine (NDMA) formed in the gastric compartment of the gastrointestinal model after the intake of nitrosation precursors and modulators (n=2)
(10K)
In contrast to the increase of NDMA observed by the addition of thiocyanate, introducing antioxidants into the meal with codfish, spinach and nitrite (simulating two times the ADI of nitrate) inhibited the NDMA formation. Eighty milliliters of orange juice (containing 35 mg ascorbic acid) added to the meal intake before introduction into the model, resulted in the cumulative formation of 4.5 g NDMA (range 3.95.1 g). Orange juice introduced into the gastric compartment 1 h after the intake of the meal with spinach and codfish, resulted in 4.6 g NMDA (range 4.34.8 g).
Similar results were obtained for black tea (0.5 g dissolved in boiling water): 3.2 g NDMA (range 3.03.4 g) and 3.0 g NDMA (range 2.93.0 g) was formed, when black tea was given together with the meal or 1 h after the intake of the meal, respectively. Thus, the addition of antioxidants simultaneously or one hour after the meal intake (codfish and spinach) did result in a similar inhibition of NDMA formation; 4.5 g and 3.0 g for orange juice and tea, respectively, compared to 10.6 g NDMA under the same conditions without these inhibitors (Table 3).
4. Discussion
The present study shows that substantial amounts of NDMA are formed in the gastric compartment of the dynamic in vitro gastrointestinal model in the presence of its precursors, a variety of fish species in combination with nitrite, under defined conditions simulating the human physiological situation in the stomach.
For the risk evaluation of NDMA exposure after consumption of fish and nitrate-containing vegetables, it would be useful to predict the level of NDMA formation based on the concentrations of amines present in fish. However, this seems to be hardly possible. No strict correlation was found between the individual concentrations of amines, such as dimethylamine (DMA), trimethylamine and trimethylamine oxide, measured in batches of codfish and herring used in this study and the formation of NDMA (M. Kotterman, personal communication, RIVO). Nevertheless, it is clear that to some extent DMA or enzymes responsible for the conversion of amines into DMA are necessary precursors for NDMA formation. Pike perch, as an example of a non-gadoid fish species, and grilled steak tartar, as an example of meat (data not shown), did not result in NDMA formation above the background level.
The formation of NDMA from precursors with in vitro models was studied earlier under specific static conditions, such as a fixed gastric pH and unrealistically high concentrations of nitrite ([Walker, 1990]). A few studies were performed with food products, among which different types of fish, mimicking the human situation by addition of artificial saliva and gastric juice ( [Groenen et al., 1982 and Sen et al., 1985]). These static in vitro experiments do not reflect the kinetic conditions in the human stomach. In our study the pH profile in the dynamic in vitro gastrointestinal model was set to simulate the human physiological situation in the stomach: an initial incubation at pH 5 (immediately after ingestion) and then a gradual decrease to pH 1.7.
Direct comparison between static models and our dynamic model is difficult. In our experiment under conditions of a rapid gastric pH decrease, the amount of NDMA formed was 16.5 g (range 14.718.2 g) simulating the intake of 100 g of codfish and one times the ADI of nitrate (total mass flow of 0.105 m nitrite). This is in the same range as observed by [Groenen et al., 1980 and Groenen et al., 1982]. They found 618 g NDMA after 2 h of incubation at pH 2 with 250 g of codfish and 0.15 m nitrite. In our experiments, however, the concentrations of the precursors in the gastric compartment at the time the gastric pH was two, were much lower than in the studies of [Groenen et al., 1982]. Thus, quantitative comparison showed that in our in vitro model approximately the same amounts of NDMA were formed using lower concentrations of the precursors.
Apart from the lack of pH changes in the static systems, another important shortcoming of the static models is that nitrite is added directly to the incubation mixture, whereas the actual uptake of nitrite via food is usually negligible. The nitrite concentration starts to increase by active secretion of nitrate in the saliva, approximately 1 h after the intake of nitrate. To predict the nitrosation more precisely, the nitrite concentration in our model was programmed to increase slowly after the intake of fish, reaching a peak value after 1 h. Higher concentrations of nitrite were reached in the gastric compartment when the rate of gastric emptying decreases, thereby increasing the efficiency of the reaction of nitrite with DMA. In conclusion, the dynamic in vitro gastrointestinal model simulating human digestion mimics the in vivo nitrosation probably better than static models.
In contrast to in vivo studies, the dynamic in vitro gastrointestinal model presented here offers the opportunity to investigate the formation of carcinogenic nitrosamines, such as NDMA, over time in the gastric compartment under realistic conditions. Research on the process of nitrosation in the human stomach with the use of food matrices is difficult and not ethical. Therefore human studies are usually limited to analysis of excretion products in the urine, e.g. performed with the N-nitrosoproline (NPRO) assay. However, in human studies no correlation was found between N-nitrosoproline excretion and dietary nitrate intake ([Tannenbaum, 1987]).
On the basis of NDMA excretion in the urine, [Vermeer et al., 1998] estimated the amount of NDMA produced in the stomach at 174 g per day. In our study 16.5 g NDMA were formed, simulating approximately the same conditions as in the human study of Vermeer et al. However, urinary NDMA excretion in the human study reflects the repeated daily exposure to a fish meal and nitrate at the level of one times the ADI, while in our experiment the amount of NDMA was formed after a single exposure (viz. the intake of 100 g codfish and nitrite at the level corresponding to one times the ADI of nitrate). Besides, N-nitrosamines are rapidly and almost completely metabolised, estimations of the NDMA formation rates in the human stomach on the basis of urinary excretion may not be accurate. Another explanation for the difference in NDMA formation could be that part of the NDMA was formed at other sites in the body by non-acid-catalysed mechanisms, as described by [Leaf et al., 1989 and Leaf et al., 1990]. This extra-gastric NDMA formation is obviously not detected in our gastric model.
It should be mentioned that the dynamic in vitro model does not take into account the gastric absorption of nitrite. [Licht et al., 1986 and Licht & Deen, 1998] argued that nitrite absorption in the human stomach, rather than gastric emptying and dilution by food consumption is the main determining factor for NDMA formation. This argument was based on observations in the stomach of fasting dogs, i.e. circumstances that may facilitate nitrite absorption. In humans, NDMA formation occurs in a stomach that contains food, i.e. under circumstances that should inhibit absorption. This means that the permeability rate of the human stomach for nitrite may be quite different from the rate that is prevalent in the empty stomach of dogs. Mirvish et al., indeed showed that nitrite loss in the stomach of rats is much slower when rats were given food mixed with nitrite, compared to a liquid nitrite solution ( [Mirvish et al., 1975]). Unfortunately there are no human data available that can be used to evaluate the effect of the food matrix on the absorption of nitrite in the stomach. On the other hand, nitrite excretion by the stomach mucosa can neither be excluded.
Furthermore, the gastric nitrite concentration might have been lowered in humans by the presence of antioxidants, such as ascorbic acid or tea polyphenols in the gastric juice. The results obtained in our gastrointestinal model showed that food compounds, such as orange juice and black tea effectively inhibited, but not totally prevented the nitrosation reaction under physiological conditions. Nitrosation inhibitors usually react with nitrosating agents. Therefore they act as competitors for amines, which are substrates for the nitrosating agents. [Vermeer et al., 1999] presented the effects of ascorbic acid on urinary excretion of NDMA in human volunteers. They showed that the urinary NDMA excretion during the intervention with fish, nitrate and 250 mg ascorbic acid was decreased to the background levels. A four times higher dose of ascorbic acid did not show an additional effect on the inhibition of urinary NDMA excretion. Other studies in humans also reported no effect of ascorbic acid on the basal NDMA and NPRO excretion ( [Leaf et al., 1987, Garland et al., 1986 and Wagner et al., 1985]). Likewise, in our model, at higher levels of NDMA formation (by the introduction of 100 g of codfish and nitrite mimicking the intake of 10 times the ADI of nitrate), the reduction of nitrosation by 250 mg ascorbic acid was approximately the same as observed in the experiments with orange juice, namely 60% (data not shown). In addition, under the same conditions the NDMA formation could not totally be reduced at higher dose levels of ascorbic acid (up to 1000 mg).
The results presented in this study also showed that spinach reduced the formation of NDMA considerably. The preventive mechanism could be related to the reduction of nitrite into NO, which is not a directly reactive compound, by ferridoxin nitrite reductase or ascorbic acid present in spinach ([Mikami & Ida, 1989 and Gill et al., 1999]). Our results suggest that relatively small amounts of nitrosation inhibitors present in fruit and vegetables are sufficient to considerably reduce, but not completely prevent, the NDMA formation. A human intervention study indeed showed that the amount of ascorbic acid and other inhibitors present in the daily intake of vegetables and fruit was not sufficient to prevent the endogenous formation of NOC ( [Vermeer et al., 1998]).
The formation of NDMA was also effectively inhibited by black tea extracts (0.5 g) added to the food intake of the model. Similar to the results obtained with ascorbic acid, NDMA formation could not totally be reduced, even not at higher dose levels (up to 4 g of green or black tea; data not shown). In the human study by Vermeer et al., 4 g of green tea resulted in even increased NDMA excretion ([Vermeer et al., 1999]). Those results suggest that increasing the ratio between nitrite and nitrite-reducers does not always result in a progressively stronger inhibition of NDMA formation. The doseeffect relationships and the effectiveness of NOC inhibitors need further investigation.
Several studies have indicated that dietary nitrate exposure increases the nitrite concentration in the saliva. The reaction between salivary nitrite and ascorbic acid into NO is extremely rapid, but NO is also rapidly recycled to nitrite in the presence of oxygen ([Mowat et al., 1999]). Swallowing is the major source of delivering oxygen into the stomach. Therefore, oxygen will enhance ascorbic acid consumption and increase nitrosation at the side where salivary nitrite enters acidic gastric juice ( [Moriya et al., 2002]). The pH in gastro oesophageal junction and the cardia of the stomach favours nitrosation, because it escapes the buffering effect of food ( [Fletcher et al., 2001]). The anatomical site where acid nitrosation is maximal corresponds with the increasing incidence of mutagenesis and carcinogenesis at the gastro oesophageal junction and cardia ( [Hansson et al., 1993, Hansen et al., 1997 and Mayne et al., 2001]).
It is evident that the endogenous formation of NDMA depends on many factors, including the pH in the gastric compartment, the concentrations of secondary amines, nitrate, nitrite, thiocyanate and nitrosation inhibitors. Information on the intragastric formation of NDMA from complex mixtures of precursors and inhibitors in humans is not available. It is important to investigate NDMA formation under various physiological conditions present in the human gastrointestinal tract. The dynamic in vitro gastrointestinal model offers the possibility to follow the NDMA formation in time at the location where it is formed. In this model the cumulative amount of N-alkylnitrosamines formed can be quantified, in contrast to the human body in which N-alkylnitrosamines are quickly distributed and metabolised. The present study clearly demonstrates that rather high amounts of NDMA are formed under physiological gastric conditions, if both appropriate precursors are present. The results are supported in experiments using more complex food matrices, such as nitrate-containing vegetables, various species of fish and food-related nitrosation inhibitors. The results show that the model seems to be a useful tool to investigate mechanistically the NDMA formation, under human conditions without ethical and practical limitations.
IARC has classified various N-alkylnitrosamines as probably or possibly carcinogenic to humans. Results obtained with this model are helpful in evaluating possible human cancer risks for long-term exposure (e.g. to frequent low levels) or to incidental exposure (e.g. to high levels) to nitrate containing vegetables in combination with gadoid (codfish-like) fish. Exposure to endogenous NDMA formation turned out to be much higher than exposure to exogenous NDMA. This indicates that risk assessment for nitrate intake should be focused not only on the toxicity of nitrate itself, but also on its role as substrate for the formation of one of the NDMA precursors. In the absence of extensive quantitative risk estimation so far, one may be well-advised to reduce exposure to exogenous N-nitrosamines and to be careful with combined intake of the precursors of N-nitrosamines.
Acknowledgements
The work described here was financially supported by the Platform Alternatives to Animal Experiments (PAD) and the Ministry of Health, Welfare and Sport, the Netherlands. The authors thank Dr. Ingrid Vermeer, Dr. Jan Dallinga and the late Dr. Jan van Maanen for the fruitful collaboration and discussions during the study. This work is dedicated to Dr. Jan van Maanen, who suddenly passed away in November 2002.
References
Bartsch et al., 1988. H. Bartsch, H. Ohshima and B. Pignatelli, Inhibitors of endogenous nitrosation. Mechanisms and implications in human cancer prevention. Mut. Research 202 (1988), pp. 307324. Abstract | Abstract + References | PDF (1381 K)
Bartsch et al., 1990. H. Bartsch, H. Ohishima, D.E.G. Shuker, B. Pignatelli and S. Calmels, Exposure of humans to endogenous N-nitroso compounds: implications in cancer etiology. Mutation Research 238 (1990), pp. 255267. Abstract | Abstract + References | PDF (1091 K)
Bartsch & Spiegelhalder, 1996. H. Bartsch and B. Spiegelhalder, Environmental exposure to N-nitroso compounds (NOC) and precursors: an overview. Eur. J. Cancer Prev. 5 (1996), pp. 1118. Abstract-EMBASE
Boyland & Walker, 1974. E. Boyland and S.A. Walker, Effect of thiocyanate on nitrosation of amines. Nature 248 (1974), pp. 601602. Abstract-EMBASE | Abstract-MEDLINE
Dallinga et al., 2001. J.W. Dallinga, C.A.M. Krul, A. Tenfelde, E.J.C. Moonen, I.T.M. Vermeer, D. van Doorn and R. Schothorst, Determination of N-nitrosodimethylamine in artificial gastric juice by gas chromatography-mass spectrometry and by gas-chromatography-termal energy analysis. European Journal of Cancer Prevention 10 (2001), pp. 265268. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Elsevier BIOBASE | Full Text via CrossRef
Dougall et al., 1995. H.T. Dougall, L. Smith, C. Duncan and N. Benjamin, The effect of amoxycillin on salivary nitrite concentrations: an important mechanism of adverse reactions?. Br. J. Clin. Pharmacology 21 (1995), pp. 1822.
Eicholzer & Gutzwiller, 1998. M. Eichholzer and F. Gutzwiller, Dietary nitrates, nitrites, and N-nitroso compounds and cancer risk: a review of the epidemiologic evidence. Nutr. Rev. 56 (1998), pp. 95105. Abstract-Elsevier BIOBASE | Abstract-MEDLINE | Abstract-EMBASE
Ellen et al., 1990. G. Ellen, E. Egmond, J.W. van Loon, E.T. Sahertian and K. Tolsma, Dietary intake of some essential metals and non-essential trace elements, nitrate, nitrite and N-nitrosamines by Dutch adults: estimated via 24-hour duplicate portion study. Food Add. Contam. 7 (1990), pp. 207221. Abstract-MEDLINE | Abstract-EMBASE
Fletcher et al., 2001. J. Fletcher, A. Wirtz, J. Young, R. Vallence and K.E.L. McColl, Unbuffered highly acidic gastric juice exists at the gastroesophageal junction after a meal. Gastroenterology 121 (2001), pp. 775783. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Elsevier BIOBASE
Gangolli et al., 1994. S.D. Gangolli, P.A. van den Brandt and V.J. Feron, Janzowsky, C, Koeman, J. H., Speijers, G.J.A., Spiegelhalder, B., Walker R., Wishnok, J.S. Nitrate, nitrite and N-nitroso compounds. Eur. J. Pharmacol. Environ. Toxicol. Pharmacol. Section 292 (1994), pp. 138. Abstract | Abstract + References | PDF (3507 K)
Garland et al., 1986. W.A. Garland, W. Kuenzig, F. Tubio, H. Kornychuk, E.P. Norkus and A.H. Conney, Urinary excretion of nitrosodimethylamine and nitrosoproline in humans: interindividual and intraindividual differences and the effect of administered ascorbic acid and alpha tocopherol. Cancer Res. 46 (1986), pp. 53925400. Abstract-EMBASE | Abstract-MEDLINE
Gill et al., 1999. M.I. Gill, F. Ferreres and F.A. Tomas-Barberan, Effect of postharvest storage and processing on the antioxidant consistents (flavonoids and vitamin C) of fresh-cut spinach. J. Agric. Food Chem. 47 (1999), pp. 22132217.
Gray & Stachiw, 1987. J.I. Gray, Stachiw, A, Gas chromatographic-thermal energy analysis method for determination of volatile n-nitrosamines in baby bottle rubber nipples; collaborative study. J. AOAC 70 (1987), pp. 6468. Abstract-MEDLINE
Groenen et al., 1980. P.J. Groenen, M.W. de Cock-Bethbeder, J. Bouwmna and J.H. Dhont, Formation of N-nitrosamines and N-nitrosamino acids from food products and nitrite under simulated gastric conditions. IARC Sci. Publ. 31 (1980), pp. 215229. Abstract-MEDLINE
Groenen et al., 1982. P.J. Groenen, J.B. Luten, J.H. Dhont, M.W. de Cock-Bethbeder, L.A. Prins and J.W. Vreeken, Formation of volatile N-nitrosamines from food products, especially fish, under simulated gastric conditions. IARC Sci. Publ. 41 (1982), pp. 99112. Abstract-MEDLINE
Hansen et al., 1997. S. Hansen, J.N. Wijg, K.E. Giercksky and S. Tretli, Esophageal and gastric carcinoma in Norway 19581992: incidence time trend variability according to morphological subtypes and organ subsites. Int. J. Cancer 71 (1997), pp. 340344. Abstract-MEDLINE | Abstract-EMBASE | Full Text via CrossRef
Hansson et al., 1993. L.E. Hansson, P. Sparen and O. Nyren, Increasing incidence of carcinoma of the gastric cardia in Sweden from 1970 to 1985. Br. J. Surgery 80 (1993), pp. 374377. Abstract-MEDLINE | Abstract-EMBASE
Hoffman & Brunneman, 1983. D. Hoffmann and K.D. Brunnemann, Endogenous formation of N-nitroproline in cigarette smokers. Cancer Research 43 (1983), pp. 55705574. Abstract-EMBASE | Abstract-MEDLINE
IARC, 1978. IARC, IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, Vol.17. Some N-Nitroso Compounds. , IARC Press, Lyon (1978).
IARC, 1987. IARC, IARC Monograph on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, Vol. 37. Tobacco Habits Other Than Smoking: Betel-quid and Area-nut Chewing, and Some Related Nitrosamines. , IARC Press, Lyon (1987).
IARC, 2000. IARC, IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol. 77, Some Industrial Chemicals. , IARC Press, Lyon (2000).
Knekt et al., 1999. P. Knekt, R. Jarvinen, J. Dich and T. Hakulinen, Risk of colorectal and other gastrointestinal cancers after exposure to nitrate, nitrite and N-nitroso compounds: a follow-up study. Int. J. Cancer 80 (1999), pp. 852856. Abstract-MEDLINE | Abstract-EMBASE | Full Text via CrossRef
Krul et al., 2000. C.A.M. Krul, A. Luiten-Schuite, R. Baan, H. Verhagen, G. Mohn, V. Feron and R. Havenaar, Application of a dynamic in vitro gastrointestinal tract model to study the availability of food mutagens, using heterocyclic amines as model compounds. Food Chem. Toxicol. 38 (2000), pp. 783792. SummaryPlus | Full Text + Links | PDF (304 K)
Leaf et al., 1987. C.D. Leaf, A.J. Vecchio, D.A. Roe and J.H. Hotchkiss, Influence of ascorbic acid dose on N-nitrosoproline formation in humans. Carcinogenesis 6 (1987), pp. 791795. Abstract-MEDLINE | Abstract-EMBASE
Leaf et al., 1989. C.D. Leaf, J.S. Wishnok and S.R. Tannenbaum, Mechanism of endogenous nitrosation. Cancer Surv. 8 (1989), pp. 323334. Abstract-MEDLINE | Abstract-EMBASE
Leaf et al., 1990. C.D. Leaf, J.S. Wishnok, J.P. Hurley, W.D. Rosenblad, J.G. Fox and S.R. Tannenbaum, Nitrate biosynthesis in rats, ferrets and humans. Precursor studies with L-arginine. Carcinogenesis 11 (1990), pp. 855858. Abstract-MEDLINE | Abstract-EMBASE
Licht et al., 1986. W.R. Licht, D.S. Schultz, J.G. Fox, S.R. Tannenbaum and W.M. Deen, Mechanisms for nitrite loss from the stomach. Carcinogenesis 7 (1986), pp. 16811687. Abstract-MEDLINE
Licht & Deen, 1998. W.R. Licht and W.M. Deen, Theoretical model for predicting rates of nitrosamine and nitrosamide formation in the human stomach. Carcinogenesis 9 (1988), pp. 22272237. Abstract-MEDLINE | Abstract-EMBASE
Lijinsky, 1999. W. Lijinsky, N-Nitroso compounds in the diet. Mutat. Res. 443 (1999), pp. 129138. SummaryPlus | Full Text + Links | PDF (95 K)
MAFF, 1992. MAFF, 1992. Nitrate, nitrite and N-nitroso compounds in food. Second report. Food Surveillance, paper no. 32.
Mayne et al., 2001. S.T. Mayne, H.A. Risch, R. Dubrow, Chow Wong-Ho, M.D. Gammon, T.L. Vaughan, D.C. Farrow, J.B. Schoenberg, J.L. Stanford, Ahsan Habibul, A. Brian West, H. Rotterdam, W.J. Blot and J.F. Fraumeni, Jr, Nutrient intake and risk of subtypes of esophageal and gastric cancer. Cancer Epidemiology, Biomarkers and Prevention 10 (2001), pp. 10551062. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Elsevier BIOBASE
Mikami & Ida, 1989. B. Mikami and S. Ida, Spinach ferredoxin-nitrite reductase: charecterization of catalytic activity and interaction of the enzyme with substrates. J. Biochem. 105 (1989), pp. 4750. Abstract-MEDLINE
Minekus et al., 1995. M. Minekus, P. Marteau, R. Havenaar and J.H.J. Huis in `t Veld, A multicompartimental dynamic computer-controlled model simulating the stomach and small intestine. ATLA 23 (1995), pp. 197209.
Mirvish, 1975. S.S. Mirvish, Formation of N-nitroso compounds: chemistry, kinetics, and in vivo occurrence. Tox. Appl. Pharm. 31 (1975), pp. 325351. Abstract-EMBASE | Abstract-MEDLINE
Mirvish et al., 1975. S.S. Mirvish, K. Patil, P. Ghadirian and V.R.C. Kommineni, Disappearance of nitrite from the rat stomach: contribution of emptying and other factors. J. Natl. Cancer Inst. 54 (1975), pp. 869875. Abstract-EMBASE | Abstract-MEDLINE
Mirvish, 1983. S.S. Mirvish, The etiology of gastric cancer. Intragastric nitrosamide formation and other theories. JNCI 71 (1983), pp. 631647.
Mirvish, 1995. S.S. Mirvish, Role of N-nitroso compounds (NOC) and N-nitrosation in etiology of gastric, esophageal, nasopharyngeal and bladder cancer and contribution to cancer of known exposures to NOC. Cancer Lett. 93 (1995), pp. 1748. Abstract | PDF (2262 K)
Moore et al., 1984. J.G. Moore, P.E. Christian, J.A. Brown, C. Brophy, F. Datz, A. Taylor and N. Alazraki, Influence of meal weight and caloric content on gastric emptying of meals in man. Dig. Dis. Sci. 29 (1984), pp. 513519. Abstract-EMBASE | Abstract-MEDLINE
Moriya et al., 2002. A. Moriya, J. Grant, C. Mowat, C. Williams, A. Carswell, T. Preston, S. Anderson, K. Iijima and K.E.L. McColl, In vitro studies indicate that acid catalysed generation of N-nitrosocompounds from dietary nitrate will be maximal at the gastro-oesophageal junction and cardia. Scand. J. Gastroenterology 3 (2002), pp. 253261. Abstract-MEDLINE | Abstract-Elsevier BIOBASE | Abstract-EMBASE | Full Text via CrossRef
Mowat et al., 1999. C. Mowat, A. Carswell, A. Wirz and K.E.L. McColl, Omeprazole and dietary nitrate independently affect levels of Vitamin C and nitrite in gastric juice. Gastroenterology 116 (1999), pp. 813822. Abstract-EMBASE | Abstract-MEDLINE
Oshima et al., 1981. H. Ohshima and H. Bartsch, Quantitative estimation of endogenous nitrosation in humans by monitoring N-nitrosoproline excreted in the urine. Cancer Res. 41 (1981), pp. 36583662. Abstract-EMBASE | Abstract-MEDLINE
Pensabene & Fiddler, 1994. J.W. Pensabene and W. Fiddler, Gas chromatography-thermal energy analyzer method for N-nitrosodibenzylamine in hams processed in elastic rubber netting. J. AOAC Int. 77 (1994), pp. 981984. Abstract-MEDLINE
Peto et al., 1984. R. Peto, R. Gray, P. Brantom and P. Grasso, Nitrosamine carcinogenesis in 5210 rodents: chronic administration of sixteen different concentrations of NDEA, NDMA, NPYR and NPIP in the water of 4440 inbred rats, with parallel studies on NDEA alone of the effect of age of starting (3,6 or 20 weeks) and on species (rats, mice or hamsters). In: I.K. O'Neill, R.C. von Borstel, C.T. Miller, J. Long and H. Bartsch, Editors, N-nitroso Compounds: Occurrence, Biological Effects and Relevance to Human Cancer, IARC Sci. Publ. No 84, Lyon (1984), pp. 627665. Abstract-MEDLINE
Peuessman, 1990. R. Preussman, Carcinogenicity and structure-activity relationships of N-nitroso compounds. Rev. Drug. Dev. Eval. 16 (1990), p. 3 -18 .
Sen et al., 1985. N.P. Sen, L. Tessier, S.W. Seaman and P.A. Baddoo, Volatile and non-volatile nitrosamines in fish and the effect of deliberate nitrosation under simulated gastric conditions. J. Agric. Food Chem. 33 (1985), pp. 264268. Abstract-EMBASE
Shapiro et al., 1991. K.B. Shapiro, J.H. Hotchkiss and D.A. Roe, Quantitative relationship between oral nitrate reducing activity and the endogenous formation of N-nitroamine acids in humans. Food Chem. Toxicol. 29 (1991), pp. 751755. Abstract | Abstract + References | PDF (423 K)
Shepard et al., 1987. S.E. Shephard, Ch. Schlatter and W.K. Lutz, Assessment of the risk of formation of carcinogenic N-nitroso compounds from dietary precursors in the stomach. Food Chem. Toxicol. 25 (1987), pp. 91110.
Slob et al., 1995. Slob, W., van den Berg, R., van Veen, M.P., 1995. A statistical exposure model applied to nitrate intake in the Dutch population. In: Health Aspects of Nitrate and Its Metabolites (Particularly Nitrite), Council of Europe Press, Strasbourg, France.
Spiegelhalder et al., 1976. B. Spiegelhalder, G. Eisenbrand and R. Preussmann, Influence of dietary nitrate on nitrite content of human saliva: possible relevance to in vivo formation of N-nitroso compounds. Food Cosmet. Toxicol. 14 (1976), pp. 545548. Abstract-EMBASE | Abstract-MEDLINE
Spiegelhalder & Preussmann, 1985. B. Spiegelhalder and R. Preussmann, In vivo nitrosation of aminopyrine in humans: use of `ethanol effect' for biological monitoring of N-nitrosodimethylamine in urine. Carcinogenesis 6 (1985), pp. 545548. Abstract-EMBASE | Abstract-MEDLINE
Stephany & Schuller, 1980. R.W. Stephany and P.L. Schuller, Daily dietary intakes of nitrate, nitrite and volatile N-nitrosamines in the Netherlands using the duplicate portion sampling technique. Oncology 37 (1980), pp. 203210. Abstract-EMBASE | Abstract-MEDLINE
Straif et al., 2000. K. Straif, S.K. Weiland, M. Bungers, D. Holthenrich, D. Taeger, S. Yi and U. Keil, Exposure to high concentrations of nitrosamines and cancer mortality among a cohort of rubber workers. Occup. Environ. Med. 57 (2000), pp. 180187. Abstract-Elsevier BIOBASE | Abstract-EMBASE | Abstract-MEDLINE | Full Text via CrossRef
Tannenbaum, 1987. S.R. Tannenbaum, Endogenous formation of N-nitroso compounds: a current perspective. IARC Sci. Publ. 84 (1987), pp. 292296. Abstract-MEDLINE
Tricker 1997. A.R. Tricker, N-nitroso compounds and man: sources of exposure, endogenous formation and occurence in body fluids. Eur. J. Cancer Prev. 6 (1997), pp. 226268. Abstract-MEDLINE | Abstract-EMBASE
Tricker & Preussmann, 1991. A.R. Tricker and R. Preussmann, Carcinogenic N-nitrosamines in the diet: occurrence, formation, mechanisms and carcinogenic potencial. Mutat. Res. 259 (1991), pp. 277289. Abstract | Abstract + References | PDF (909 K)
Van Maanen et al., 1998. J.M.S. Van Maanen, D.M.F.A. Pachen, J.W. Dallinga and J.C.S. Kleinjans, Formation of nitrosamines during consumption of nitrate- and amine-rich food and the influence of mounthwashes. Cancer Detect. Prev. 22 (1998), pp. 204212. Abstract-Elsevier BIOBASE | Abstract-MEDLINE | Full Text via CrossRef
Vermeer et al., 1998. I.T.M. Vermeer, D.M.F.A. Pachen, J.W. Dallinga, J.C.S. Kleinjans and J.M.S. van Maanen, Volatile N-nitrosamine formation after intake of nitrate at the ADI level in combination with an amine-rich diet. Environ. Health Perspect. 106 (1998), pp. 459463. Abstract-MEDLINE | Abstract-EMBASE
Vermeer et al., 1999. I.T.M. Vermeer, E.J.C. Moonen, J.W. Dallinga, J.C.S. Kleinjans and J.M.S. van Maanen, Effect of ascorbic acid and green tea on endogenous formation of N-nitrosodimethylamine and N-nitrosopiperidine in humans. Mutat. Res. 428 (1999), pp. 353361. SummaryPlus | Full Text + Links | PDF (192 K)
Wagner et al., 1985. D.A. Wagner, D.E.G. Shuker, C. Bilmazes and M. Obedzinski, Baker, I, Young, V. R., Tannenbaum, S.R. Effects of vitamins C and E on endogenous synthesis of N-nitrosamino acids in humans: precursor-product studies with [15N]nitrate. Cancer Res. 45 (1985), pp. 65196522. Abstract-EMBASE | Abstract-MEDLINE
Walker, 1990. R. Walker, Nitrates, nitrites and N-nitroso compounds: a review of the occurrence in food and diet and the toxicological implications. Food Add. Contam. 5 (1990), pp. 717768. Abstract-MEDLINE | Abstract-EMBASE
Walters & Smith, 1981. C.L. Walters and P.L.R. Smith, The effect of water-borne nitrate on salivary nitrite. Food Chem. Toxicol. 19 (1981), pp. 297302. Abstract-EMBASE | Abstract-MEDLINE
Walters et al., 1979. C.L. Walters, F.P.A. Carr, C.S. Dyke, M.J. Saxby, P.L.R. Smith and R. Walker, Nitrite sources and nitrosamine formation in vitro and in vivo. Fd. Cosmet. Toxicol. 17 (1979), pp. 473479. Abstract | Full Text + Links | PDF (817 K)
Zeilmaker et al. Zeilmaker, M.J., Meulenbelt, J., Slob, W. A doseresponse model for nitrate induced methemoglobin in humans. In preparation.
|
