Effect of Feeding Nitrite, Ascorbate, Hemin, and Omeprazole on

Feces from individual mice was collected on day 7, dried, and extracted with water. ... The enhancement by hemin may help explain why red meat is a ca...
0 downloads 0 Views 146KB Size
2344

Chem. Res. Toxicol. 2008, 21, 2344–2351

Effect of Feeding Nitrite, Ascorbate, Hemin, and Omeprazole on Excretion of Fecal Total Apparent N-Nitroso Compounds in Mice Sidney S. Mirvish,*,†,‡ Michael E. Davis,† Michal P. Lisowyj,† and Nilesh W. Gaikwad† Eppley Institute for Research in Cancer and Departments of Pharmaceutical Sciences and of Biochemistry and Molecular Biology, UniVersity of Nebraska Medical Center, Omaha, Nebraska 68198 ReceiVed May 27, 2008

It was proposed that colon cancer induced by red and nitrite-preserved meat is due to meat-derived N-nitroso compounds in the colonic contents. To explore this view, we previously showed that feeding beef and hot dogs increased the fecal output of total apparent N-nitroso compounds (ANC) in mice. In the current project, adult Swiss mice were fed a semipurified diet and water containing additives for 7 days. Feces from individual mice was collected on day 7, dried, and extracted with water. Extracts were analyzed for ANC as before. Feeding 2.0 g sodium nitrite (NaNO2)/L drinking water raised fecal ANC levels from 5 to 63 nmol/g feces. In a dose-response study, fecal ANC levels were proportional to the nitrite concentration squared. Even 32 mg NaNO2/L raised fecal ANC levels 2.3-fold (P < 0.05). In other results, 64, 125, and 250 mg hemin/kg diet, fed with 2 g NaNO2/L water, showed 2.3, 2.2, and 4.6 times the ANC level for nitrite alone. Sodium nitrate (12 g/L water) did not affect fecal ANC output. Omeprazole (400 mg/kg diet) and sodium ascorbate (23 g/kg diet), when fed with 1 g NaNO2/L water, lowered fecal ANC levels by 65 and 41%, indicating that, when nitrite was fed, acid-catalyzed reactions in the stomach produced ANC, which passed down the gut to the feces. Tests indicated that nitrosothiols constituted about 20% of fecal and hot dog ANC. The observed effect of NaNO2 is thus far not consistent with the proposed hypothesis. The enhancement by hemin may help explain why red meat is a cause of colon cancer. Introduction In continuation of a previous study (1), we report here additional dietary effects on the concentration of total nonvolatile apparent N-nitroso compounds (ANC)1 in the feces of mice. In the United States, there were 52000 deaths from colon cancer in 2007, with similar rates in men and women (2). This is the third highest death rate in both sexes for cancer at specific sites. Colon cancer has been linked to the consumption of “red meat” (beef, pork, and lamb) but not to that of poultry and fish (3-5). In 2007, an expert international panel concluded that “the evidence that red meat and processed meat are causes of colorectal cancer is convincing...and is stronger than it was in the mid-1990s” (5). Feeding beef as 60% of the diet to dimethylhydrazine-treated rats significantly increased the number of colonic aberrant crypt foci, a probable precancer lesion for colon cancer (6). The risk of colon cancer for “processed,” that is, nitrite-preserved, red meat is greater than that for fresh red meat (7-9). There is evidence that the ANC fraction in foods and feces consists mainly of N-nitroso compounds (NOC) (4, 10, 11). Fecal ANC output in humans rose 3-fold after feeding 420 g beef/day but was not affected by feeding chicken or fish (3, 12, 13). Thus, red meat could be a cause of colon cancer because it elevates carcinogenic NOC levels in the colonic lumen, and these NOC act on the colonic mucosa to induce cancer. Raised levels of O6-carboxymethylguanine were detected * To whom correspondence should be addressed. E-mail: smirvish@ unmc.edu. † Eppley Institute for Research in Cancer. ‡ Departments of Pharmaceutical Sciences and of Biochemistry and Molecular Biology. 1 Abbreviations: NOC, N-nitroso compounds; ANC, total apparent NOC; ANCP, ANC precursors; Exp., experiment; GI, gastrointestinal; and SA, sulfamic acid.

in colonic DNA of humans on a high red-meat diet (14), suggesting alkylation by a nitrosated glycine derivative (14). Since 1978, processed meat has been prepared with 120 mg/ kg of sodium nitrite (NaNO2) and 500 mg/kg of sodium erythorbate (iso-ascorbate), a nonvitamin isomer of ascorbate. This has reduced the volatile nitrosamine level in fried bacon from that in the product prepared before 1978, which contained 150 mg/kg of sodium nitrite (NaNO2) and no erythorbate (15). However, the ANC content of processed meat has remained high, with a mean of 5.5 nmol/g in hot dogs (4). There are no data on the ANC content of processed meat before 1978 because the ANC method was only being developed at that time (16). Because nitrite increases the levels of both volatile nitrosamines and nonvolatile ANC and the volatile nitrosamine level has been reduced, the nonvolatile ANC may be the components of processed meat that increase the risk of colon cancer. In 2001 (4), we determined ANC in a number of foods (4) by modification of a method (17-19) involving the preparation of aqueous extracts, addition of sulfamic acid (SA) to destroy nitrite, liberation of nitric oxide (NO) from the ANC by treatment with hydrobromic acid (HBr), and determination of the NO by thermal energy analysis. Following studies in humans (13), we reported that fecal ANC output in mice increased 2.6-2.9 or 3.7-5.0 times when the diet contained 18% of beef or hot dogs (1). Later (20), we partially purified the ANC precursors (ANCP) in hot dogs, identified one of the ANCP as the secondary amine, 1-deoxy-1-N-glucosyl glycine, and showed that the nitrosated, purified ANCP fraction was directly mutagenic in the Ames test (20). The color of red meat is due to myoglobin, which consists of a globin and heme. In nitrite-preserved meat products, the heme in myoglobin binds NO to give the pink dye, nitrosyl

10.1021/tx8001884 CCC: $40.75  2008 American Chemical Society Published on Web 10/31/2008

Total Apparent N-Nitroso Compounds in Mouse Feces

Chem. Res. Toxicol., Vol. 21, No. 12, 2008 2345

Table 1. ANC Concentration in Feces of Mice Given Various Treatmentsa treatment

ANC [nmol/g feces, mean ( SD (no. of mice)]

cotreatment

group no.

test material

concentration

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

NaNO2 NaNO2 NaNO2 NaNO2 hemin hemin Na NO3 Na ascorbate Na ascorbate Na ascorbate ellagic acid R-tocopherol hot dogs NaNO2 Na ascorbate omeprazole omeprazole omeprazole

1.0 g/L water 2.0 g/L water 1.0 g/L water 2.0 g/L water 125 mg/kg diet 1.0 g/kg diet 12 g/L water 23 g/kg diet 23 g/kg diet 23 g/kg diet 1.0 g/kg diet 5.0 g/kg diet 180 g/kg diet 1.0 g/L water 23 g/kg diet 0.40 g/kg diet 0.40 g/kg diet 0.40 g/kg diet

test material

concentration

commercial diet commercial diet NaNO2

125 mg/L water

NaNO2 NaNO2

1.0 g/L water 2.0 g/L water

NaNO2 NaNO2

1.0 g/L water 125 mg/L water

hot dogs hot dogs NaNO2

180 g/kg diet 180 g/kg diet 1.0 g/L water

hot dogs

180 g/kg diet

for test groups

for control groupsb

mean test/mean control results

21 ( 8 (13)c 63 ( 24 (9)c 198 ( 34 (4)d 261 ( 59 (3) 9 ( 3 (7) 11.9 ( 6.5 (3) 5.1 ( 1.7 (4) 17 ( 7 (4) 22 ( 10 (8)c 3.7 ( 1.8 (9) 24 ( 7 (4) 6 ( 2 (7) 21 ( 10 (12)c 59 ( 11 (7)c 12 ( 3 (8) 6.3 ( 4.1 (8)c 5.6 ( 1.7 (8) 23.3 ( 5.7 (8)

2.8 ( 0.8 (13) 5 ( 3 (9) 8 ( 3 (4) 2.0 ( 0.2 (2) 6 ( 3 (8) 9.8 ( 1.8 (3) 6.6 ( 0.8 (4) 29 ( 7 (4) 96 ( 49 (7) 3.1 ( 1.3 (11) 18 ( 5 (4) 6 ( 3 (8) 3 ( 1 (11) 26 ( 8 (8) 11 ( 2 (7) 17.8 ( 7 (7) 6.2 ( 4.6 (10) 17.3 ( 7.6 (8)

7.5 12.6 25 131 1.5 1.21 0.77 0.58 0.44 1.2 1.33 1.0 7.0 2.3 1.1 0.35 0.90 1.35

a

Test materials were administered in the drinking water or diet as indicated. The basic diets were the semipurified diets listed in the Experimental Procedures except in groups 3 and 4, where a commercial diet was fed. b The control groups were given only the cotreatment or, when no cotreatment is listed, were given the appropriate diet alone. c Results were significantly different from those for the control groups, with P < 0.01. d Results were significantly different from those for the control groups, with P < 0.05.

myoglobin (21). Red meat could induce colon cancer because NO bound to heme in myoglobin is transferred to amines or amides to produce carcinogenic NOC (22, 23). Accordingly, we tested the effect of feeding hemin on fecal ANC levels in mice also given NaNO2. Hemin (ferric protoporphyrin) was used for these tests because it is more readily available and more stable than its ferrous analogue, heme. Wang et al. determined ANC using cuprous chloride instead of HBr to liberate NO from ANC (11). This method determined S-nitrosothiols (RSNO) in addition to NOC. S-Nitroso-Nacetylpenicillamine and S-nitroso glutathione gave responses in the ANC test that were 3 and 32% of that for N-nitrosoproline. Therefore, we checked here whether ANCs in feces and hot dogs include nitrosothiols. To determine nitrosothiols, we mainly used a method that equates the nitrosothiol level to the loss of ANC on adding mercuric chloride (HgCl2), which reacts with nitrosothiols to liberate nitrite, which is destroyed with SA (24). We also tested a method based on the finding that ascorbic acid decomposes nitrosothiols over a wide pH range (25, 26). Some of the work reported here was presented at national meetings (27, 28).

Experimental Procedures Chemicals and Diets. Chemicals (purest grade available) were purchased from Aldrich unless mentioned otherwise. NaNO2 and sodium nitrate (NaNO3) were of ACS grade. Hot dogs of a brand with a relatively high ANC content (4) were bought locally. All diets were purchased from Harlan Teklad (Madison, WI). As before (1), the TD-94045 diet was a slightly modified AIN93G semipurified diet, and the “hot dog diet” was prepared by adding 18% by weight of hot dogs (fresh weight) to the TD-01407 diet, which contained 21.4% casein and no oil. This addition brought the fat and protein content to 6-8 and 17-21% of the diet, similar to that in the TD-94045 diet. For the control groups of the hot dog tests, 26 g of casein and 57 g of soy oil were added to 917 g of theTD-01407 diet. All diets were stored at 4 °C. The hot dog diet was not stored for more than 7 days. Mice and Conduct of Experiments. Male Swiss-Webster mice that were 6-7 weeks old were obtained from Charles River Laboratories (Wilmington, MA). All experiments were approved by the Institutional Animal Use and Care Committee of our Medical Center. The mice were kept 4-5/cage in microisolator cages with

filter tops and were maintained on pelleted Sterilizable Rodent Diet (“commercial diet”) and tap water in a 1.4 m × 2.4 m animal room. All diets were supplied by Harlan Teklad. The test diets were fed in custom-made stainless steel mouse feeders (Ancare, Bellmore, NY). Diets were blended with an Oster 10 cup food processor (Shelton, CT) and added to the feeders every second day. Cage bedding was Bed-o-Cobs (Anderson, Maumee, OH). Each experiment included an untreated control group of four mice. In experiments where the interactions of two factors were examined, four groups were generally set up, each with four mice, for example, groups with heme plus NaNO2, NaNO2 alone, heme alone, and no treatment. Some results in Table 1 are taken from these tests. On day 7 of all experiments, the mice were housed singly over wire grids with absorbent paper under the grids, and the feces was collected for 24 h at room temperature (the wire grids prevented the mice from eating their feces). We could have left the mice on the grids for longer than 1 day to habituate them to the grids, but our Institutional Animal Care and Use Committee did not wish us to leave the mice on the grids for longer than 1 day. Throughout the 7 day experiment, the mice were fed commercial diet (groups 3 and 4 of Table 1), hot dog diet (groups 13-15 of Table 1), or the TD-94045 semipurified diet mixed with additives (all other groups). For drinking, the mice in the designated groups of Table 1 and in the experiments shown in Figures 1 and 2 received NaNO2 or NaNO3 solutions in distilled water. The more dilute NaNO2 solutions were prepared by serial 1:1 dilutions of solutions with 256 mg NaNO2/L water. For drinking, the controls in the NaNO2 and NaNO3 tests received distilled water during the 7 day experiment; all other mice received tap water for drinking. Sodium ascorbate, ellagic acid, R-tocopherol, and omeprazole were mixed with the TD-94045 diet at the levels shown in Table 1. The control mice for these tests were fed the same diet without additives. At the end of the omeprazole experiment, the mice were killed, the stomachs were dissected, and the pH of the glandular stomach content was measured with pH paper (pH range, 1.0-5.5; Micro Essential Laboratory, New York). Determination of ANC and ANCP. All analyses were performed in the authors’ laboratories. The entire 24 h fecal sample was dried to constant weight for 2 days at 0 °C and