In Vitro Binding of Mutagenic Pyrolyzates to Lactic Acid Bacterial Cells in Human Gastric Juice

In Vitro Binding of Mutagenic Pyrolyzates to Lactic Acid Bacterial Cells in Human Gastric Juice XUE BIN ZHANG and YOSHlYUKl OHTA Laboratory for Microbial Biochemistry Faculty of Applied Biological Science Hiroshima University Higashi-Hiroshima 724, Japan ABSTRACT

The binding of mutagenic pyrolyzates to freeze-dried cells of Streptococcus cremoris 2 2 5 was investigated in human gastric juice and compared with the ability of these pyrolyzates to bind in distilled water @H 6.30). The pH of donated gastric juice ranged from 1.2 to 7.8. Binding that occurred in gastric juice was pH dependent, and binding was affected by NaCl, CaCl2, and MgCl2. Less of TrpP-2 was bound in gastric juice at low pH (1.2, 1.48, and 2.1) than at pH 4 to 8. The strain bound simultaneously Trp-P-l , Trp-P-2, and Glu-P-1. The maximum amounts of Trp-P-2 bound to S. cremoris 2 2 5 were approximately 47.50 pdmg of cells. Freeze-dried cells treated at 120’C for 15 min or 80’C for 3 h showed increased binding of Trp-P-2. However, binding decreased 10% after heating at 120°C for 15 min. Cells nonthermally killed by gastric juice (pH 1.5) had the same Trp-P-2 binding capacity as viable cells. Lactobacillus acidophilus I F 0 13951 and Bifiabbacterium bifdum IF0 14252 representation of the intestinal bacteria nonthermally killed by gastric juice also showed larger binding of TpP-2. (Key words: binding, human gastric juice, mutagenic pyrolyzates) INTRODUCTION

Various pyrolysis products from amino acids, proteins, and proteinaceous foods are highly mutagenic to Salmonella typhimurium TA 98 and TA 100 compared with other typical mutagens (11). The chemical structures of these

Received April 26, 1990. Accepted September 24, 1990. 1991 J Dairy Sci 74:752-757

mutagens were identified as new heterocyclic aromatic amines (4, 8). Some of them, Trp-P-1 (3-amino-l,4-dimethyl-5H-pyrido [4,3-b]indole, Trp-P-2 (3-amino- 1-methyl-5H-pyrido [4,3blindole), and Glu-P-1 (2-amino-6-methyldipyrido[lJ-a:3’, 2’d]imidaZole) are carcinogenic to mice and rats (4, 5). The first compounds of this series to have their pathway of metabolic activation identified were Trp-P-1 and Trp-P-2, and their biological activity has been intensively studied (10). Many studies clearly show that fermented milk products and lactic acid bacteria inhibit the activity of mutagenic chemical compounds. Ad libitum intake of yogurt for 7 consecutive d after tumor implantation inhibited tumor cell growth (11). Shiraishi et al. (9) suggested that oral administration of Srreptococcus pyogenes A3 was of value for immunotherapy of liver tumor. Cell wall, plasma membrane, cytoplasmic fraction, and sum-protein showed antitumor activity, but the effect varied according to the tumor site and route of administration. It is important to control mutagenesis and carcinogenesis by identifying hazardous agents and removing them from the environment. Recently, reports have described the inactivation of mutagenic pyrolyzates by adsorbtion to intestinal bacteria (7). Our previous study (6) reported that freeze-dried cells of lactic acid bacteria effectively bound mutagenic pyrolyzates. Heating cells at 120’C for 15 min or at 80’C for 3 h did not significantly decrease binding of mutagenic pyrolyzates by the cells. There! are no reported studies of the binding of mutagens to lactic acid bacteria in human gastric juice. It is necessary to determine whether lactic acid bacteria can bind mutagenic pyrolyzates in gastric juice and to compare that finding with previous studies (13) to provide data for in vivo testing in the gastrointestinal tract. We reported that the two elements responsible for the binding and antimutagenicity of lactic acid bacteria to mutagens were the

752

753

BINDING OF MUTAGENS IN GASTRIC JUICE

polysaccharide and peptidoglycan of the cell at least 12 h of fasting. Five milligrams of walls (12). In this study, we have investigated freeze-dried cells were suspended in .95 ml of how cells of lactic acid bacteria bind mutagenic gastric juice. Methanolic solutions of Trp-P-1, pyrolyzates and whether thermally and nonther- Trp-P-2, or Glu-P-1 (100 pg each in .05 ml) mally killed cells have higher binding activity were added to the suspensions, and the mixture compared with viable cells in human gastric was incubated at 37'C for 10 min. After incubation, the reaction mixture was centrifuged at juice. 1500 x g for 15 min. The supernatant was removed and ultrafiltered (membrane filter pore MATERIAL AND METHODS size: .45 pn. Toyoyoshi, Ltd, Tokyo, Japan). The amount of unbound mutagen was deterBacteria Culture mined by HPLC (13). Streptococcus cremoris 2-25 was isolated from traditional Chinese cheese Nui Nai Ge Da Heat Treatment in this laboratory. The strain was maintained in Harvested cells were killed by heating at a GPY medium containing glucose 10 g; pep tone, 7 g; yeast extract, 5 g; Na2HPO4,6 g; and 120'C for 15 min or at 80'C for 3 h. Viable cells were not found by a plate count method. NaH2P04, 4 g in 1 L of distilled water. Lactobacillus acidophilus IF0 13951 and Killed cells were freeze-dried Bijihbacterium bifidum IF0 14252 were purchased from the Institute of Fermentation Effect of pH on Survival OSAKA (Osaka, Japan). These strains were Fifty milligrams of lyophilized cells were grown anaerobically in LC broth at 37°C for 24 h. The LC broth contained yeast extract, 5.5 g; suspended in 10 ml of gastric juice or phospeptone, 12.5 g; glucose, 11 g; MgS0~7H20, phate buffer adjusted to pH 1.5. The mixture -1 g; KH2P04, .25 g; K2HPO4, .25 g; sodium was incubated at 37'C for 3 h. Cells were acetate, 10 g ; FeS04.7H20, .005 g ; harvested by centrifugation at 1500 x g for 20 MnS0411H20. .005 g; and 1 L of distilled min, washed twice with 1:15 M phosphate buffwater; the pH was adjusted to 6.8 f .1 with 1N er, and lyophilized. Viable organisms were enumerated by plate count at 0 and 3 h. NaOH. RESULTS

Mutagens Used in the Study

The Trp-P-1 and TrpP-2, derived from a pyrolyzate of tryptophan, and Glu-P-1, derived Binding of Trp-P-2 to Freeze-Dried Cells from glutamic acid, were obtained from Wako Binding of Trp-P-2 to freeze-dried cells of S. Pure Chemical Co. Ltd. (Osaka,Japan). cremoris 2-25 was investigated in gastric juice (Table 1). The pH of freshly aspirated gastric Preparation of Lyophilized Cells juice varied from 1.2 to 7.8 (the subject experienced extreme saliva production due to irritaSterilized GPY medium (300 ml) was inocu- tion by the tube), the gastric juice of numbers lated with l ml of the culture and incubated at 36'C for 12 h. Cells were harvested by centrifugation at 1500 x g for 20 min, washed twice with 1:15 M phosphate buffer (pH 6.8) and 1. In vim bof Trp-P-2 by freeze-dried cells maintained at -20'C for 1 h. The cells were TABLE of Srr roco~cuscremris 2-25 gastric juice from five then lyophilized for 14 h using a freeze-drier donors.T (VD-15 Yamato, Japan). Samples were stored 1 2 3 4 5 at -20'C until used. Age Binding of Pyrolyzates to Cells

PH Binding, 46

28 1.48 11.2

Gastric juice was obtained by aspiration 'Assays were through a nasogastric tube from volunteers after means.

20

2.1 12.9

23 7.8 56.0

27

4.50

53.1

24

1.2 11.0

performed in duplicate. Values are

Journal of Dairy Science Vol. 74, No. 3. 1991

754

ZHANG AND

OHTA

TABLE 2. In vitro binding of mutagens by freeze-dried cells of Streptococcus cremoris 2-25 in distilled water and gastric juice @H 6.30).l

Mutagens

TPP-1

TpP- 2

GlU-P-1

Distilled water Gastric juice

86.6 65.0

83.9 53.6

18.6 13.0

'Assays were performed in duplicate. Values are

means.

Effect of Metal Salts

We investigated the inhibitory effect of metal salts on the binding of Trp-P-2 to cells of PH Strep. cremoris 2-25 (Figure 2). The effect of Figure 1. Effect of pH on the binding of TrpP-2 (la0 NaC1, CaCl2, and MgC12 varied with concentrapgfml) to freeze-dried cells of Streptococcus cremoris 2 2 5 tion. At concentrations of 100 mM, the inhibiin human gastric juice. tion was higher than at Concentrations of 10 and 50 mM. Inhibition of binding by CaCl2 was greater than inhibition by either NaCl or Mg(321, 2, and 5 had a pH of 1.48, 1.2, and 2.1, respectively, and cells bound less Trp-P-2 than Effect of Cell Concentration on the Blnding of cells in the gastric juice of number 3 @H 7.8), Trp-P-2 to Freeze-Dried Cells and 4 (pH 4.5). These results demonstrated that The dose response relationship for binding the binding was pHdependent. of Trp-P-2 to freeze-dried cells of Sfrep. cremEffect of pH

The pH of one sample of donated gastric juices was adjusted to pH 2, 4, 6, 7, and 9 by adding 1%NaOH solution. The cells of Strep. cremoris 2-25 showed maximal Trp-P-2 binding at pH above 6 and less than 9 (Figure 1). At pH 4, 6, and 7, over 50% of the Trp-P-2 bound to the cells.

oris 2-25 was investigated in gastric juice. As shown in Figure 3, the amount of bound T p P -

eo

Simultaneous Blndlng of Yutagenlc Pyrolyzates

Table 2 shows simultaneous binding of mutagenic pyrolyzates by the freeze-dried cells of Strep. cremoris 2-25. When the strain was incubated with gastric juice @H 6.3) containing Trp-P-1, Trp-P-2, and Glu-P-1, more Trp-P-1 than Trp-P-2 and Glu-P-1 was bound to the cells. The binding of Glu-P-1 by cells was lowest. The binding ability of the cells was higher in distilled water than in gastric juice for all of the mutagens. Journal of Dairy Science Vol. 74, No. 3, 1991

Figure 2. Effect NaCl, CaCI2, and MgCl2 on the binding of TrpP-2 to freezedried cells of Srreptococcus cremoris 225.Inhibition rate = (1 - binding %[+salt]/binding %[no salt]). Trp-P-2 concentration = 100 p@. A 10 W, 0: 50 mM, c loo mu.

755

BINDING OF MUTAGENS IN GASTRIC JUICE

Icx3

a N I

e a I

t u u-0 01

ZiVJ .r

m

I 23 0 0

2

4

G

8

10

Freeze-dried c e l l s (mg/ml)

Figure 3. Dose-response binding of TrpP-2 (100 crp/ ml) to freeze-dried cells of Sneprococcu.v cremoris 2 2 . 5in human gastric juice @H 6).

2 increased as the number of cells increased. The maximum amount of Trp-P-2 bound to freeze-dried cells was approximately 47.50 pg/ ml.

Retentlon time l m i n )

Figure 4. Binding of Trp-P-2 to freeze-dried cells of Sweptocuccur cremoris 225 in human gastric juice. The HPLC of Trp-P-2 (100 &ml) solution (A)was compared with that of the filtrate obtained from solution containing freeze-dried cells @freeze-dried I), cells autoclaved at 12072 for 15 min (0, and freeze-driedcells heated at 8o'C for 3 h @).

Effect of Heating on Binding Ability

Figure 4 shows that untreated cells of Strep. cremoris 2-25 bound more Trp-P-2 than treated cells. When cells were autoclaved at 120'C for 15 min binding of T v P - 2 (C) decreased as compared with the untreated cells (B). However, there was little difference in the binding abilities of cells heated at 8o'C for 3 h @), and cells heated at 120'C for 15 min (D) were similar.

DISCUSSION

It is important to reduce the risk of carcinogenesis by preventing the interaction of chemical mutagens and carcinogens with DNA. Lactic acid bacteria could possibly play a vital role in lowering the risk of cancer development by suppressing the activity of mutagens in the intestine. In a previous study (13), we reported Strep. cremoris and Streptococcus lactis inactivated mutagenic pyrolyzates by adsorbing them in vitro. In this study uable l), cells of Strep. Binding Ability of Freeze-Drled Cells cremoris 2-25 also showed high binding ability Exposed to Gastric Juice for Trp-P-2 added to gastric juice. However, Viable counts of all three bacterial strains binding capacity was affected by the pH of were reduced from loglo (7.5 to 9.7), to loglo human gastric juices. The pH of human gastric (1.3 to 3) in phosphate buffer or in gastric juice is usually between pH 1 to 5 (1). At pH 2, juice, respectively. After incubation at 37'C for cells bound approximately 26.23% of the Trp3 h at pH 1.5 (Table 3), Lactobacillus acidophi- P-2. However, at pH 6, about 68.85% of the lus IF0 13951 survived better than the other T p P - 2 W ~ Sbound Binding of T p P - 2 in gasstrains. Binding of Trp-P-2 to cells of bacteria tric juice was reported as negligible at pH 1.1 did not change to a large extent in treated cells and 3.0, and it was rapidly absorbed by the compared with viable cells. Binding was higher small intestine and the large intestine (13). The in phosphate buffer than in gastric juice. pH of the intestine is almost neutral (3). ThereJ o d of Dairy Science Vol. 74, No. 3, 1991

756

ZHANG AND OHTA

TABLE 3. Effect of exposing freeze-driedcells to gastric juice or phosphate buffer at 37'C for 3 h at pH 1.5 on binding of TwP-2.

Phomhate buffer viable counts Cloglo CM) B12 AI3

Gastric juice Viable counts (loglo Wml) BI AI

Streptococcus cremoris 2-25

Bound TrpP-1. % Laciobncillus Bifidobacrerium acidophilus IF0 bifidum IF0 13951 14252

8.3 (1.3)' 98.55 97.89

9.1 (1.6) 99.12 98.78

7.5 (<1) 97.64 98.00

9.7 (2.3) 65.66 64.2 1

8.8 (3) 71.22 72.12

7.9 (Q) 67.34 67.5 1

lViable counts after incubation at 37'C for 3 h. *Before incubation. 3After incubation at 37'C for 3 h.

fore, it may be likely that cells of lactic acid bacteria easily bind mutagens in the intestinal tract. It is very important to test this possibility in vivo. The results presented in Table 2 show that the Strep. cremoris 2-25 cells can simultaneously bind three mutagenic pyrolyzates. The binding of mutagens to cells is different depending on the mutagen. We found that the binding was higher in distilled water than in gastric juice and is probably caused by NaCl, CaC12, and MgCl2, which affected binding in this study. This is consistent with results of the study by Morotomi et al. (7). The binding activity of mutagens by cells seems to be affected by pH, metal salts, and other factors. We have detected that the binding activities of lactic acid bacteria killed nonthermally, or thermally, were not drastically affected compared with those of viable cells (Figure 3 and Table 3). It is well known that because the pH of gastric juice is low @H 1 to 5 ) when lactic acid bacteria pass through the stomach, survival of viable cell counts is reduced (1). Therefore, it is important that dead cells of lactic acid bacteria still have high binding ability. Both B. bifidum and L. acidophilus are intestinal bacteria. B@dobacterium is dominant in intestines. Thus, it is important that cells of B . bifidum killed nonthemally, by exposure to gastric juice, also have high binding ability for mutagens. Although no evidence has been currently found to support this hypothesis, bacterid Journal of Dairy Scieace Vol. 74, No. 3. 1991

strains may be important in preventing the absorption of amines by the intestine. We also found that the binding activities of lactic acid bacteria occu mainly in the peptidoglycan of cell walls (13). Bacterial cell walls also strongly bind N-nitroamines (12). We are continuing our research to determine whether lactic acid bacteria and intestinal bacteria killed thermally and nonthermally would be useful in the binding of mutagenic pyrolyzates or N-nitroamines in vivo. REFERENCES

1 Conwar. P. L., S. L. Gorbach, and B. R. Goldin. 1987. Survival of lactic acid bacteria in the human stomach and adhesion to intestinal cells. J. Dairy Sci. 701. 2Fanner, R. E., K. U Shahani, and G.V. Geddy. 1975. Inhibitory effect of yogurt components. J. Dairy Sci. 58:787. 3 Ishikawa, T.,S. Takayama, T.Kitagawa, T. Kawachi, M. Kiuebuchi, N. Matsukura, E. Uchida, and T. Sugimm. 1979. In vivo experiments on tryptophan pyrolysis products. Page 159 in Naturally occurring carcieogen--mutagens and modalator of carcinogenesis. University Park Press, Baltimore, MD. 4Kasai, H., 2. Yamaizumi, S . Nishimura, K. W-yashi, M. Nagao, T.Sugimura, N. E. SpiDgam, J. H. Weisburger, S. Yokoyama, and T.A. Miyazawa. 1981. A potent mutagen in broiled fish. Part 1. 2-Amino-3-methyl-3H-imidazo[4.5-fl-quinoline, J. Chem. Soc. Perkin Trans.I. 77:2290. 5 Kimura, T.,T.Nakayama, Y. Kurosaki, Y. Suzuki, S. Arimoto and H. Hayatsu. 1985. Absorption of 3amino- 1-methyl-[5H]-pyrido[4.3-b]indole, mutagencarcinogen present in tryptophan pyrolysate from the gastrointestinal tract in the rat. J. Cancer Res. 76:273.

BINDING OF MUTAGENS IN GASTRIC JUICE 6 Matsulolra, N.,T. Kawachi, IC. Morino, H. ohgaki,T. Sugimura, and S. Takayama. 1981. Carcinogenicily in

mice of mutagenic compounds from a tryptophan pyrolyzate. Science 213:346. 7 Morotomi, M.,and M.Mutai. 1986. In vitro binding of potent mutagenic pyrolyzates to intestinal bacteria. J. NaU. Cancer Inst. 77195. SNagao, M., and T. Sugimura. 1978. Environmntal mutagens and carcinogens. Ann. Rev. Genet. 12117. 9 Shiraishi, T., Y.Nai, S. Imai, M.Tsubono, H. Morimoto, K. Obgaki, and T. Tobe. 1988. Page 1098 in Abstracts of papers, 2nd Int Cod. Anticancer Res. 11-15 S~roniP,Greece. (Abstr.) 10 Sugimura, T.. T. Kawachi, M. Nagao, M. Yamada, S. Takayama, N. Mntsukura, and K. Wakabayashi. 1980. Genotoxic carcinogens and co-mutagens in !qptophan

757

pyrolyzate. Page 297 in Biochemical and medical aspects of tryptophan metabolism. 0. Hayaishi, T. 1-m and R. a d o , ed. ElserviWNorth-Holland Biomedical Press, Amsterdam, Neth. 11 Sugimura, T.,and U Nagao, 1982. The use of mutagenicity to evaluate carcinogenic hazards in our daily lives. Page 73 in Mutagenicity: new horizons in genetic toximlogy. J. A. Heddle, ed. Academic Press, New

Yo& NY. 12Zhang, X. B.. and Y. Ohta. 1990. Page 495 in Abstfacts of papers, Annu. Mtg.Agric. Chem.Soc. Japan, Fulruka, JPlL 13zhang, X. B., Y. Ohta, and A. Hosono. 1990. Autimutagenicity and binding of lactic acid bacteria from Chinese cheese to mutagenic pyrolyzates. J. Dairy Sci. 73:2702.

Journal of Dairy Science Vol. 74. No. 3, 1991