- Email: [email protected]
Colloidal bismuth subcitrate and omeprazole inhibit alcohol dehydrogenase mediated acetaldehyde production by Helicobacter pylori
Life Sciences, Vol. 51, pp. PL 195-200 Printed in the USA
Pergamon Press
PHARMACOLOGY LETTERS Accelerated Communication
COLLOIDAL BISMUTH SUBCITRATE AND OMEPRAZOLE INHIBIT ALCOHOL DEHYDROGENASE MEDIATED ACETALDEHYDE PRODUCTION BY HEL1COBACTER PYLOR1 Risto P. Roine 1, Katja S. Salmela 1, Johanna H66k-Nikanne2, Timo U. Kosunen2, and Mikko Salaspuro 1. Research Unit of Alcohol Diseases I and Department of Bacteriology and Immunology2, University of Helsinki, Finland. (Submitted June 30, 1992; accepted August 25, 1992; received in final form September 15, 1992)
Abstract. We have recently shown that Helicobacter pylori possesses marked alcohol dehydrogenase (ADH) activity and is capable - when incubated with an ethanol containing solution in vitro - of producing large amounts of acetaldehyde. In the present study we report that some drugs commonly used for the eradication of H. pylori and for the treatment of gastroduodenal diseases are potent ADH inhibitors and, consequently, effectively prevent bacterial oxidation of ethanol to acetaldehyde. Colloidal bismuth subcitrate (CBS), already at a concentration of 0.01 mM, inhibited H. pylori ADH by 93% at 0.5 M ethanol and decreased oxidation of 22 mM ethanol to acetaldehyde to 82% of control. At concentrations above 5 mM, CBS almost totally inhibited acetaldehyde formation. Omeprazole, a drug also known to suppress growth ofH. pylori, also inhibited H. pylori ADH and suppressed bacterial acetaldehyde formation significantly to 69 % of control at a drug concentration of 0.1 mM. By contrast, the H2-receptor antagonists ranitidine and famotidine showed only modest effect on bacterial ADH and acetaldehyde production. We suggest that inhibition of bacterial ADH and a consequent suppression of acetaldehyde production from endogenous or exogenous ethanol may be a novel mechanism by which CBS and omeprazole exert their effect both on the growth of H. pylori as well as on H. pylori associated gastric injury.
Introduction
Helicobacter pylori is at present considered to be the major cause of type B gastritis, and moreover, contributes significantly to the pathogenesis of peptic ulcer disease (1). H. pylori infection has a large prevalence in healthy asymptomatic populations (2), and has recently also been linked with gastric cancer (3). The pathogenetic mechanisms behind H. pylori associated morbidity have, however, so far not been fully elucidated. Alcohol dehydrogenase (ADH) catalyzed fermentation of sugars to ethanol provides a source of anaerobic energy for ethanologenic bacteria. Such bacterial ethanol production has been shown Address correspondence to: Dr. Mikko Salaspuro, Research Unit of Alcohol Diseases, Tukholmankatu 8 F, 00290 Helsinki, Finland
0024-3205/92 $5.00 + .00 Copyright © 1992 Pergamon Press Ltd All rights reserved.
PL-196
ADH Inhibition
in H. pylori by CBS
Vol.
51, No.
21,
1992
to occur in the intestine of both humans (4) and experimental animals (5). The ADH catalyzed reaction that in bacteria runs from acetaldehyde to ethanol is, however, reversible, and under some conditions bacteria may oxidize ethanol to acetaldehyde, which is a highly reactive and toxic substance. We have recently shown that H. pylori possesses ADH activity which results in acetaldehyde formation both when bacterial cytosol is incubated with ethanol in vitro (6) as well as when the organism is cultured under microaerobic conditions with ethanol (7). This acetaldehyde production - if it also occurs in vivo - could be responsible for H. pylori associated gastric injury as the bacteria may be unable to effectively remove acetaldehyde due to lack of aldehyde dehydrogenase activity (8). The aim of the present study was to determine whether some drugs known to be effective in the eradication ofH. pylori and treatment ofH. pylori associated gastric injury, affect bacterial ADH activity and consequent acetaldehyde production. Based on our findings we propose a novel mechanism of action for these drugs. Materials and methods Colloidal bismuth subcitrate (CBS) in the form of a powder was obtained from Brocades Pharma B.V. (Delft, The Netherlands), omeprazole sodium salt from Astra H~issle AB (Mrlndal, Sweden), and ranitidine hydrochloride from Huhtam~iki Oy Leiras (Turku, Finland) or Sigma Chemical Co. (St. Louis, Mo, USA). Due to solubility problems with the pure compound, we used, when testing famotidine, a dry substance intended for intravenous use from Merck Sharp & Dohme B.V. (Haarlem, The Netherlands). Baker's yeast alcohol dehydrogenase was from Sigma Chemical Co. (St. Louis, Mo, USA). Since CBS is a polymeric molecule forming colloidal particles its concentrations were based on bismuth oxide equivalents (300 mg CBS equivalent to 120 mg Bi203). H. pylori were grown on brucella agar plates (BBL, Cockeysville, MD, USA) supplemented with whole horse blood (7%) in an atmosphere of 5% 02, 10% CO~, and 85% N2 at 35 °C for 48 h. The cultured bacteria (strains NCTC 11637 and NCTC 11638) were first washed twice in 100 mM potassium phosphate buffer and then sonicated in the same buffer solution. This was followed by centrifugation of the sonicate at 100 000 g at 5 °C for 60 min to obtain cytosol. Cytosolic ADH activity was then determined spectrophotometrically by measuring, after addition of ethanol, the reduction of nicotinamide adenine dinucleotide (NAD, final concentration in the reaction mixture 2.5 mM) at 25 °C in 0.1 M glycine buffer with a pH of 9.6. When testing the effect of famotidine (which was not soluble at pH 9.6), however, a Tris-HC1 buffer, pH 7.2 was used. Ethanol concentrations ranging from 12.5 mM to 2.5 M were used. The effect of drugs on ADH activity was tested by adding increasing drug concentrations to the buffer used. Protein concentration of the bacterial cytosols was determined by the Lowry method, and ADH activity was calculated as nmoles of NADH produced by mg bacterial cytosolic protein per minute. For comparison with bacterial ADH, we also tested the effect of CBS on commercially available yeast alcohol dehydrogenase. The effect of the different drugs on the ability of H. pylori to produce acetaldehyde in vitro was determined by incubating 0.1 ml of the cytosol (prepared as described above) in closed vials with 0.4 ml saline solution containing ethanol (final concentration in the vials 22 raM), NAD (final concentration 3.0 mM) and increasing drug concentrations (0 - 10 mM) for 2 h at 37 °C. The pH
Vol.
51, No.
21,
1992
ADH Inhibition
in H. pylori by CBS
PL-197
of the solutions after addition of the drugs varied between 3.5 and 9.2 depending on the drug tested (CBS pH 6.2, omeprazole pH 9.2, ranitidine pH 3.5, famotidine pH 4.1). It was, however, always adjusted so that when studying the effect of a certain drug the pH remained constant over the whole concentration range tested. After the incubation period the vials were rapidly refrigerated and kept at + 4 °C until ethanol and acetaldehyde concentrations were determined using head space gas chromatography by heating the vials to a temperature of 37 °C as suggested earlier (9). Conditions for analysis were following: Column 60/80 Carbopack C/0.2 % Carbowax 1500, 2 m x 1/8" ID (Supelco Inc, Bellefonte, PA, USA); oven temperature, 75 °C; dosing line detector, 200°C; carrier gas (N2), 20 ml/min. Results are expressed as mean + SE. Statistical significance of the results was tested by analysis of variance for repeated measures followed by Bonferroni's t test. P values under 0.05 were considered statistically significant. The type of ADH inhibition observed, was determined graphically employing Lineweaver-Burk and Dixon plots. Results Alcohol dehydrogenase activity of both H. pylori strains was significantly inhibited by CBS already at a drug concentration of 0.01 mM (at 0.5 M ethanol ADH activity 6.9 + 3.1% of control, Fig. 1). In a similar fashion, omeprazole showed significant in vitro inhibition of ADH activity but first at much higher drug levels (at 0.5 M ethanol ADH activity at 0.5 mM omeprazole 54.1 + 10.0% of control, Fig. 1). Both with CBS and omperazole the inhibition appeared to be of the competitive type. The inhibitory effect of ranitidine was comparable to that of omeprazole (at 0.5 M ethanol ADH activity at 0.25 mM ranitidine 45.8 -t- 12.0% of control) whereas famotidine decreased ADH activity only slightly and only at high drug concentrations (Fig. 1). o
oo
>..
100
[~
CBS
I~I
Fm~otkil~
"d
~ 50 -io o
0
0.01
0.1
0.5
Drug concenlraflon (ml~)
Figure 1. Effects of colloidal bismuth subcitrate (CBS), omeprazole, ranitidine, and famotidine on alcohol dehydrogenase activity of two H. pylori strains (results are means ___SE of at least three different determinations with each strain and are expressed as percentage of control without the drug). The determinations shown here were performed at an ethanol concentration of 0.5 M. Due to turbidity problems CBS and ranitidine were tested only up to a concentration of 0.5 m M .
Similar ADH inhibition was also observed with commercial baker's yeast ADH. A 0.01 mM concentration of CBS decreased yeast ADH activity to 22% of control at 0.5 M ethanol.
PL-198
ADH Inhibition
in H. pylori by CBS
Vol.
51, No. 21,
1992
In line with the ADH inhibition, CBS, at a concentration of 0.01 raM, significantly decreased the conversion of ethanol to acetaldehyde by H. pylon to 81.6 + 4.0% of control (Fig. 2). Higher drug concentrations had an even more pronounced inhibitory effect and a CBS concentration of 5 mM or more led to an almost total inhibition of bacterial acetaldehyde formation (Fig. 2). Omeprazole had a similar significant, but slightly less marked effect on acetaldehyde production by H. pylori already at a concentration of 0.1 mM (68.6 + 7.1% of control, Fig. 2). Contrasting to its clear inhibitory effect on ADH activity, ranitidine showed significant inhibition of bacterial acetaldehyde formation first at a concentration of 5 mM (81.9 ___ 3.6% of control, Fig. 2) as did also famotidine (at 5 mM famotidine, acetaldehyde production 55.1 _ 12.7% of control, Fig. 2). [~]
CB$
~ Ranl~no I ~ 1 gomotrdlno
--'~
g 0
;;'u L'o °o; 'o g
ce
ot, o
(m
,o
Figure 2. Effects of colloidal bismuth subcitrate, omeprazole, ranitidine, and famotidine on acetaldehyde production from 22 mM ethanol by two H. pylori strains (results are means ___SE of at least three different determinations with each strain and are expressed as percentage of control without the drug).
Discussion Earlier studies have incriminated ammonia produced by H. pylori urease activity as a possible pathogenetic factor behind H. pylori associated gastric injury (10). Inhibition of urease activity does not, however, affect the growth of the organism (11). Other suggested pathogenic mechanism of H. pylori include its mucolytic effect (12) and its interactions with local mucosal immune systems. Our recent finding that H. pylori possesses ADH (but not aldehyde dehydrogenase) activity and produces significant acetaldehyde concentrations from ethanol in vitro (6-8), could provide an alternative explanation for H. pylori related gastric injury. Acetaldehyde is known to be a highly reactive and toxic compound and forms readily adducts with different cellular proteins and may also induce lipid peroxidation, both of which can lead to tissue injury (13). Acetaldehyde has recently also been incriminated as a possible pathogenetic factor in alcohol associated rectal and bronchopulmonary carcinogenesis (14,15). Local acetaldehyde production in the gastric mucus layer and consequent adduct formation with mucous glycoproteins could alter the protective characteristics of the mucus barrier. Acetaldehyde might also furnish an irritative stimulus responsible for the increased occurrence of gastric cancer in patients infected with H. pylori (3). Although the value of CBS in the treatment of H. pylori induced gastric injury is well established less is known about the mechanisms by which the drug exerts its effect. Suggested actions include
Vol.
51, No.
21,
1992
ADH Inhibition
in H. pylori by CBS
PL-199
formation of a glycoprotein-bismuth complex providing a protective coating for the gastric mucosa against the digestive activity of acid and pepsin (16). Other factors possibly involved are e.g. bismuth stimulated prostaglandin synthesis and direct inhibitory effect of bismuth on H. pylori (16) but no definite conclusion about their respective roles have so far been reached. Similarly, omeprazole is known to suppress H. pylori (17), but the mechanism remains speculative. In our hands, CBS at a drug concentration known to inhibit growth of H. pylori, significantly decreased ADH activity of the organism. Based on our present results we, therefore, suggest that inhibition of H. pylori ADH activity and consequent interference with bacterial energy metabolism could be a novel mechanism by which both CBS and omeprazole inhibit the growth of the bacteria. A similar mechanism, i.e. inactivation of enzymes such as triosephosphate dehydrogenase and ADH has earlier been suggested as a potential mechanism by which bismuth compounds limit fermentation by colonic bacteria (18). On the other hand, interference with bacterial ADH results in a remarkable decrease in bacterial acetaldehyde formation. This could explain the rapid recovery of gastric mucosa during treatment. Several compounds have been reported to inhibit ADH activity (19). These include pyrazoles and their 4-substituted derivatives of which 4-methylpyrazole has been recommended for the management of methanol and ethylenglycol poisoning as well as for the disulfiram-alcohol reaction. Little is, however, known about the role of CBS as an inhibitor of alcohol metabolism. In our hands it turned out to be a highly effective inhibitor of bacterial as well as of yeast ADH. This warrants for further studies as some - although small - amounts of bismuth are absorbed during CBS treatment (20) and may theoretically interfere also with systemic ethanol metabolism. According to our preliminary results, 4-methylpyrazole also inhibits H. pylori ADH and suppresses growth of the organism but only at significantly greater concentrations than CBS. This finding is in line with our proposed theory about the importance of ADH activity for H. pylori. Although ranitidine inhibited H. pylori ADH, it as well as famotidine were unable to prevent bacterial acetaldehyde production until at such high concentrations that are unlikely to occur in the stomach after the intake of a suggested dose of the drug. Further studies with these and other H2-receptor antagonists are, however still needed to reach final conclusions about their roles in the inhibition of bacterial ADH and acetaldehyde production. In summary, both CBS and omeprazole inhibit H. pylori ADH and suppress the formation of acetaldehyde from ethanol. This interference with bacterial alcohol dehydrogenase could provide a novel explanation for both the fact that the drugs inhibit growth of the organism as well for the fact that gastric mucosal injury is alleviated during therapy with these drugs. Acknowledgements The technical assistance of Ms. Eila Kelo, Ms. Tuula Heiskanen, Ms. Riitta Hurtig and Ms. Katri Salmensuu is gratefully acknowledged. Colloidal bismuth subcitrate for the studies was kindly provided by Brocades Pharma B.V., Delft, The Netherlands, omeprazole by Astra H~issle AB (M61ndal, Sweden), and ranitidine hydrochloride by Huhtam~iki Oy Leiras (Turku, Finland). The work was financially supported by a grant from the Yrj6 Jahnsson Foundation, Helsinki and by the Finnish State Alcohol Company (Alko Ltd.), Finland.
PL-200
ADH
Inhibition
in H. pylori by CBS
Vol.
51, No.
21,
1992
References 1. 2. 3.
4. 5. 6.
7. 8. 9. 10. 11. 12.
13.
14.
15. 16. 17. 18. 19. 20.
D.Y. GRAHAM, Gastroenterology 96 (Suppl.) 615-625 (1989). T.U. KOSUNEN, J. HOOK, H. RAUTELIN and G. MYLLYL.~, Scand J Gastroenterol 24 110-114 (1989). A. NOMURA, G.N. STEMMERMAN, P.-H. CHYOU, I. KATO, G.I. PEREZ-PEREZ and M.J. BLASER, N Engl J Med 325 1132-1136 (1991). J.C. BODE, S. RUST and C. BODE, Scand J Gastroenterol 19 853-856 (1984). E. BARAONA, R. JULKUNEN, L. TANNENBAUM AND C.S. LIEBER, Gastroenterology 90 103-110 (1986). R. ROINE, K. SALMELA, J. HOOK-NIKANNE, T.U. KOSUNEN and M. SALASPURO, Life Sci (1992, in press). K.S. SALMELA, R. ROINE, J. H(9OK-NIKANNE, T.U. KOSUNEN and M. SALASPURO, Scand J Gastroenterol 27 (Suppl. 190) 43 (1992). R. ROINE, K.S. SALMELA, J. HOIDK-NIKANNE, T.U. KOSUNEN and M. SALASPURO, Scand J Gastroenterol 27 (Suppl. 190) 44 (1992). P. PIKKARAINEN, M. SALASPURO and C.S. LIEBER, Alcoholism: Clin Exp Res 3 259261 (1979). H.L.T. MOBLEY, L.-T. HU and P.A. FOXALL, Scand J Gastroenterol 2_fi(Suppl. 187) 39-46 (1991). J. GOLDIE, S.J.O. VELDHUYZEN VAN ZANTEN, S. JALALI, H. RICHARDSON and R.H. HUNT, J Clin Pathol 44 695-697 (1991). J. SAROSIEK, D.A. PEURA, R.L. GUERRANT, B.J. MARSHALL, W. LASZEWICZ, A. GABRYELEWICZ and R.W. McCALLUM, Scand J Gastroenterol 26 (Suppl. 187) 47-55 (1991). M. SALASPURO and K. LINDROS, Alcohol-Related Diseases in Gastroenterology, H.K. Seitz and B. Kommerell (eds), 106-123, Springer Verlag, Berlin, Heidelberg, New York, Tokyo (1985). H.K. SEITZ, U.A. SIMANOWSKI, F.T. GARZON, J.M. RIDEOUT, T.J. PETERS, A. KOCH, M.R. BERGER, H. EINECKE and M. MAIWALD, Gastroenterology 98 406-413 (1990). H. MIYAKAWA, E. BARAONA, J.C. CHANG, M.D. LESSER and C.S. LIEBER, Alcohol: Clin Exp Res 10 517-520 (1986). S.L. GORBACH, Gastroenterology 99 863-875 (1990). P. MAINGUET, M. DELMEE and J.-C. DEBONGNIE, Lancet ii 389-390 (1989). R. LEON-BARUA, R. TELLO, M. MORANTE, M. ALVAREZ, R.H. GILMAN, and W.M. SPIRA, Rev Infect Dis 12 $24-$29 (1990). M. SALASPURO, Acta Med Scand 70____33(Suppl.) 219-224 (1985). L.Z. BENET, Scand J Gastroenterol 26 (Suppl. 185) 29-35 (1991).
Pergamon Press
PHARMACOLOGY LETTERS Accelerated Communication
COLLOIDAL BISMUTH SUBCITRATE AND OMEPRAZOLE INHIBIT ALCOHOL DEHYDROGENASE MEDIATED ACETALDEHYDE PRODUCTION BY HEL1COBACTER PYLOR1 Risto P. Roine 1, Katja S. Salmela 1, Johanna H66k-Nikanne2, Timo U. Kosunen2, and Mikko Salaspuro 1. Research Unit of Alcohol Diseases I and Department of Bacteriology and Immunology2, University of Helsinki, Finland. (Submitted June 30, 1992; accepted August 25, 1992; received in final form September 15, 1992)
Abstract. We have recently shown that Helicobacter pylori possesses marked alcohol dehydrogenase (ADH) activity and is capable - when incubated with an ethanol containing solution in vitro - of producing large amounts of acetaldehyde. In the present study we report that some drugs commonly used for the eradication of H. pylori and for the treatment of gastroduodenal diseases are potent ADH inhibitors and, consequently, effectively prevent bacterial oxidation of ethanol to acetaldehyde. Colloidal bismuth subcitrate (CBS), already at a concentration of 0.01 mM, inhibited H. pylori ADH by 93% at 0.5 M ethanol and decreased oxidation of 22 mM ethanol to acetaldehyde to 82% of control. At concentrations above 5 mM, CBS almost totally inhibited acetaldehyde formation. Omeprazole, a drug also known to suppress growth ofH. pylori, also inhibited H. pylori ADH and suppressed bacterial acetaldehyde formation significantly to 69 % of control at a drug concentration of 0.1 mM. By contrast, the H2-receptor antagonists ranitidine and famotidine showed only modest effect on bacterial ADH and acetaldehyde production. We suggest that inhibition of bacterial ADH and a consequent suppression of acetaldehyde production from endogenous or exogenous ethanol may be a novel mechanism by which CBS and omeprazole exert their effect both on the growth of H. pylori as well as on H. pylori associated gastric injury.
Introduction
Helicobacter pylori is at present considered to be the major cause of type B gastritis, and moreover, contributes significantly to the pathogenesis of peptic ulcer disease (1). H. pylori infection has a large prevalence in healthy asymptomatic populations (2), and has recently also been linked with gastric cancer (3). The pathogenetic mechanisms behind H. pylori associated morbidity have, however, so far not been fully elucidated. Alcohol dehydrogenase (ADH) catalyzed fermentation of sugars to ethanol provides a source of anaerobic energy for ethanologenic bacteria. Such bacterial ethanol production has been shown Address correspondence to: Dr. Mikko Salaspuro, Research Unit of Alcohol Diseases, Tukholmankatu 8 F, 00290 Helsinki, Finland
0024-3205/92 $5.00 + .00 Copyright © 1992 Pergamon Press Ltd All rights reserved.
PL-196
ADH Inhibition
in H. pylori by CBS
Vol.
51, No.
21,
1992
to occur in the intestine of both humans (4) and experimental animals (5). The ADH catalyzed reaction that in bacteria runs from acetaldehyde to ethanol is, however, reversible, and under some conditions bacteria may oxidize ethanol to acetaldehyde, which is a highly reactive and toxic substance. We have recently shown that H. pylori possesses ADH activity which results in acetaldehyde formation both when bacterial cytosol is incubated with ethanol in vitro (6) as well as when the organism is cultured under microaerobic conditions with ethanol (7). This acetaldehyde production - if it also occurs in vivo - could be responsible for H. pylori associated gastric injury as the bacteria may be unable to effectively remove acetaldehyde due to lack of aldehyde dehydrogenase activity (8). The aim of the present study was to determine whether some drugs known to be effective in the eradication ofH. pylori and treatment ofH. pylori associated gastric injury, affect bacterial ADH activity and consequent acetaldehyde production. Based on our findings we propose a novel mechanism of action for these drugs. Materials and methods Colloidal bismuth subcitrate (CBS) in the form of a powder was obtained from Brocades Pharma B.V. (Delft, The Netherlands), omeprazole sodium salt from Astra H~issle AB (Mrlndal, Sweden), and ranitidine hydrochloride from Huhtam~iki Oy Leiras (Turku, Finland) or Sigma Chemical Co. (St. Louis, Mo, USA). Due to solubility problems with the pure compound, we used, when testing famotidine, a dry substance intended for intravenous use from Merck Sharp & Dohme B.V. (Haarlem, The Netherlands). Baker's yeast alcohol dehydrogenase was from Sigma Chemical Co. (St. Louis, Mo, USA). Since CBS is a polymeric molecule forming colloidal particles its concentrations were based on bismuth oxide equivalents (300 mg CBS equivalent to 120 mg Bi203). H. pylori were grown on brucella agar plates (BBL, Cockeysville, MD, USA) supplemented with whole horse blood (7%) in an atmosphere of 5% 02, 10% CO~, and 85% N2 at 35 °C for 48 h. The cultured bacteria (strains NCTC 11637 and NCTC 11638) were first washed twice in 100 mM potassium phosphate buffer and then sonicated in the same buffer solution. This was followed by centrifugation of the sonicate at 100 000 g at 5 °C for 60 min to obtain cytosol. Cytosolic ADH activity was then determined spectrophotometrically by measuring, after addition of ethanol, the reduction of nicotinamide adenine dinucleotide (NAD, final concentration in the reaction mixture 2.5 mM) at 25 °C in 0.1 M glycine buffer with a pH of 9.6. When testing the effect of famotidine (which was not soluble at pH 9.6), however, a Tris-HC1 buffer, pH 7.2 was used. Ethanol concentrations ranging from 12.5 mM to 2.5 M were used. The effect of drugs on ADH activity was tested by adding increasing drug concentrations to the buffer used. Protein concentration of the bacterial cytosols was determined by the Lowry method, and ADH activity was calculated as nmoles of NADH produced by mg bacterial cytosolic protein per minute. For comparison with bacterial ADH, we also tested the effect of CBS on commercially available yeast alcohol dehydrogenase. The effect of the different drugs on the ability of H. pylori to produce acetaldehyde in vitro was determined by incubating 0.1 ml of the cytosol (prepared as described above) in closed vials with 0.4 ml saline solution containing ethanol (final concentration in the vials 22 raM), NAD (final concentration 3.0 mM) and increasing drug concentrations (0 - 10 mM) for 2 h at 37 °C. The pH
Vol.
51, No.
21,
1992
ADH Inhibition
in H. pylori by CBS
PL-197
of the solutions after addition of the drugs varied between 3.5 and 9.2 depending on the drug tested (CBS pH 6.2, omeprazole pH 9.2, ranitidine pH 3.5, famotidine pH 4.1). It was, however, always adjusted so that when studying the effect of a certain drug the pH remained constant over the whole concentration range tested. After the incubation period the vials were rapidly refrigerated and kept at + 4 °C until ethanol and acetaldehyde concentrations were determined using head space gas chromatography by heating the vials to a temperature of 37 °C as suggested earlier (9). Conditions for analysis were following: Column 60/80 Carbopack C/0.2 % Carbowax 1500, 2 m x 1/8" ID (Supelco Inc, Bellefonte, PA, USA); oven temperature, 75 °C; dosing line detector, 200°C; carrier gas (N2), 20 ml/min. Results are expressed as mean + SE. Statistical significance of the results was tested by analysis of variance for repeated measures followed by Bonferroni's t test. P values under 0.05 were considered statistically significant. The type of ADH inhibition observed, was determined graphically employing Lineweaver-Burk and Dixon plots. Results Alcohol dehydrogenase activity of both H. pylori strains was significantly inhibited by CBS already at a drug concentration of 0.01 mM (at 0.5 M ethanol ADH activity 6.9 + 3.1% of control, Fig. 1). In a similar fashion, omeprazole showed significant in vitro inhibition of ADH activity but first at much higher drug levels (at 0.5 M ethanol ADH activity at 0.5 mM omeprazole 54.1 + 10.0% of control, Fig. 1). Both with CBS and omperazole the inhibition appeared to be of the competitive type. The inhibitory effect of ranitidine was comparable to that of omeprazole (at 0.5 M ethanol ADH activity at 0.25 mM ranitidine 45.8 -t- 12.0% of control) whereas famotidine decreased ADH activity only slightly and only at high drug concentrations (Fig. 1). o
oo
>..
100
[~
CBS
I~I
Fm~otkil~
"d
~ 50 -io o
0
0.01
0.1
0.5
Drug concenlraflon (ml~)
Figure 1. Effects of colloidal bismuth subcitrate (CBS), omeprazole, ranitidine, and famotidine on alcohol dehydrogenase activity of two H. pylori strains (results are means ___SE of at least three different determinations with each strain and are expressed as percentage of control without the drug). The determinations shown here were performed at an ethanol concentration of 0.5 M. Due to turbidity problems CBS and ranitidine were tested only up to a concentration of 0.5 m M .
Similar ADH inhibition was also observed with commercial baker's yeast ADH. A 0.01 mM concentration of CBS decreased yeast ADH activity to 22% of control at 0.5 M ethanol.
PL-198
ADH Inhibition
in H. pylori by CBS
Vol.
51, No. 21,
1992
In line with the ADH inhibition, CBS, at a concentration of 0.01 raM, significantly decreased the conversion of ethanol to acetaldehyde by H. pylon to 81.6 + 4.0% of control (Fig. 2). Higher drug concentrations had an even more pronounced inhibitory effect and a CBS concentration of 5 mM or more led to an almost total inhibition of bacterial acetaldehyde formation (Fig. 2). Omeprazole had a similar significant, but slightly less marked effect on acetaldehyde production by H. pylori already at a concentration of 0.1 mM (68.6 + 7.1% of control, Fig. 2). Contrasting to its clear inhibitory effect on ADH activity, ranitidine showed significant inhibition of bacterial acetaldehyde formation first at a concentration of 5 mM (81.9 ___ 3.6% of control, Fig. 2) as did also famotidine (at 5 mM famotidine, acetaldehyde production 55.1 _ 12.7% of control, Fig. 2). [~]
CB$
~ Ranl~no I ~ 1 gomotrdlno
--'~
g 0
;;'u L'o °o; 'o g
ce
ot, o
(m
,o
Figure 2. Effects of colloidal bismuth subcitrate, omeprazole, ranitidine, and famotidine on acetaldehyde production from 22 mM ethanol by two H. pylori strains (results are means ___SE of at least three different determinations with each strain and are expressed as percentage of control without the drug).
Discussion Earlier studies have incriminated ammonia produced by H. pylori urease activity as a possible pathogenetic factor behind H. pylori associated gastric injury (10). Inhibition of urease activity does not, however, affect the growth of the organism (11). Other suggested pathogenic mechanism of H. pylori include its mucolytic effect (12) and its interactions with local mucosal immune systems. Our recent finding that H. pylori possesses ADH (but not aldehyde dehydrogenase) activity and produces significant acetaldehyde concentrations from ethanol in vitro (6-8), could provide an alternative explanation for H. pylori related gastric injury. Acetaldehyde is known to be a highly reactive and toxic compound and forms readily adducts with different cellular proteins and may also induce lipid peroxidation, both of which can lead to tissue injury (13). Acetaldehyde has recently also been incriminated as a possible pathogenetic factor in alcohol associated rectal and bronchopulmonary carcinogenesis (14,15). Local acetaldehyde production in the gastric mucus layer and consequent adduct formation with mucous glycoproteins could alter the protective characteristics of the mucus barrier. Acetaldehyde might also furnish an irritative stimulus responsible for the increased occurrence of gastric cancer in patients infected with H. pylori (3). Although the value of CBS in the treatment of H. pylori induced gastric injury is well established less is known about the mechanisms by which the drug exerts its effect. Suggested actions include
Vol.
51, No.
21,
1992
ADH Inhibition
in H. pylori by CBS
PL-199
formation of a glycoprotein-bismuth complex providing a protective coating for the gastric mucosa against the digestive activity of acid and pepsin (16). Other factors possibly involved are e.g. bismuth stimulated prostaglandin synthesis and direct inhibitory effect of bismuth on H. pylori (16) but no definite conclusion about their respective roles have so far been reached. Similarly, omeprazole is known to suppress H. pylori (17), but the mechanism remains speculative. In our hands, CBS at a drug concentration known to inhibit growth of H. pylori, significantly decreased ADH activity of the organism. Based on our present results we, therefore, suggest that inhibition of H. pylori ADH activity and consequent interference with bacterial energy metabolism could be a novel mechanism by which both CBS and omeprazole inhibit the growth of the bacteria. A similar mechanism, i.e. inactivation of enzymes such as triosephosphate dehydrogenase and ADH has earlier been suggested as a potential mechanism by which bismuth compounds limit fermentation by colonic bacteria (18). On the other hand, interference with bacterial ADH results in a remarkable decrease in bacterial acetaldehyde formation. This could explain the rapid recovery of gastric mucosa during treatment. Several compounds have been reported to inhibit ADH activity (19). These include pyrazoles and their 4-substituted derivatives of which 4-methylpyrazole has been recommended for the management of methanol and ethylenglycol poisoning as well as for the disulfiram-alcohol reaction. Little is, however, known about the role of CBS as an inhibitor of alcohol metabolism. In our hands it turned out to be a highly effective inhibitor of bacterial as well as of yeast ADH. This warrants for further studies as some - although small - amounts of bismuth are absorbed during CBS treatment (20) and may theoretically interfere also with systemic ethanol metabolism. According to our preliminary results, 4-methylpyrazole also inhibits H. pylori ADH and suppresses growth of the organism but only at significantly greater concentrations than CBS. This finding is in line with our proposed theory about the importance of ADH activity for H. pylori. Although ranitidine inhibited H. pylori ADH, it as well as famotidine were unable to prevent bacterial acetaldehyde production until at such high concentrations that are unlikely to occur in the stomach after the intake of a suggested dose of the drug. Further studies with these and other H2-receptor antagonists are, however still needed to reach final conclusions about their roles in the inhibition of bacterial ADH and acetaldehyde production. In summary, both CBS and omeprazole inhibit H. pylori ADH and suppress the formation of acetaldehyde from ethanol. This interference with bacterial alcohol dehydrogenase could provide a novel explanation for both the fact that the drugs inhibit growth of the organism as well for the fact that gastric mucosal injury is alleviated during therapy with these drugs. Acknowledgements The technical assistance of Ms. Eila Kelo, Ms. Tuula Heiskanen, Ms. Riitta Hurtig and Ms. Katri Salmensuu is gratefully acknowledged. Colloidal bismuth subcitrate for the studies was kindly provided by Brocades Pharma B.V., Delft, The Netherlands, omeprazole by Astra H~issle AB (M61ndal, Sweden), and ranitidine hydrochloride by Huhtam~iki Oy Leiras (Turku, Finland). The work was financially supported by a grant from the Yrj6 Jahnsson Foundation, Helsinki and by the Finnish State Alcohol Company (Alko Ltd.), Finland.
PL-200
ADH
Inhibition
in H. pylori by CBS
Vol.
51, No.
21,
1992
References 1. 2. 3.
4. 5. 6.
7. 8. 9. 10. 11. 12.
13.
14.
15. 16. 17. 18. 19. 20.
D.Y. GRAHAM, Gastroenterology 96 (Suppl.) 615-625 (1989). T.U. KOSUNEN, J. HOOK, H. RAUTELIN and G. MYLLYL.~, Scand J Gastroenterol 24 110-114 (1989). A. NOMURA, G.N. STEMMERMAN, P.-H. CHYOU, I. KATO, G.I. PEREZ-PEREZ and M.J. BLASER, N Engl J Med 325 1132-1136 (1991). J.C. BODE, S. RUST and C. BODE, Scand J Gastroenterol 19 853-856 (1984). E. BARAONA, R. JULKUNEN, L. TANNENBAUM AND C.S. LIEBER, Gastroenterology 90 103-110 (1986). R. ROINE, K. SALMELA, J. HOOK-NIKANNE, T.U. KOSUNEN and M. SALASPURO, Life Sci (1992, in press). K.S. SALMELA, R. ROINE, J. H(9OK-NIKANNE, T.U. KOSUNEN and M. SALASPURO, Scand J Gastroenterol 27 (Suppl. 190) 43 (1992). R. ROINE, K.S. SALMELA, J. HOIDK-NIKANNE, T.U. KOSUNEN and M. SALASPURO, Scand J Gastroenterol 27 (Suppl. 190) 44 (1992). P. PIKKARAINEN, M. SALASPURO and C.S. LIEBER, Alcoholism: Clin Exp Res 3 259261 (1979). H.L.T. MOBLEY, L.-T. HU and P.A. FOXALL, Scand J Gastroenterol 2_fi(Suppl. 187) 39-46 (1991). J. GOLDIE, S.J.O. VELDHUYZEN VAN ZANTEN, S. JALALI, H. RICHARDSON and R.H. HUNT, J Clin Pathol 44 695-697 (1991). J. SAROSIEK, D.A. PEURA, R.L. GUERRANT, B.J. MARSHALL, W. LASZEWICZ, A. GABRYELEWICZ and R.W. McCALLUM, Scand J Gastroenterol 26 (Suppl. 187) 47-55 (1991). M. SALASPURO and K. LINDROS, Alcohol-Related Diseases in Gastroenterology, H.K. Seitz and B. Kommerell (eds), 106-123, Springer Verlag, Berlin, Heidelberg, New York, Tokyo (1985). H.K. SEITZ, U.A. SIMANOWSKI, F.T. GARZON, J.M. RIDEOUT, T.J. PETERS, A. KOCH, M.R. BERGER, H. EINECKE and M. MAIWALD, Gastroenterology 98 406-413 (1990). H. MIYAKAWA, E. BARAONA, J.C. CHANG, M.D. LESSER and C.S. LIEBER, Alcohol: Clin Exp Res 10 517-520 (1986). S.L. GORBACH, Gastroenterology 99 863-875 (1990). P. MAINGUET, M. DELMEE and J.-C. DEBONGNIE, Lancet ii 389-390 (1989). R. LEON-BARUA, R. TELLO, M. MORANTE, M. ALVAREZ, R.H. GILMAN, and W.M. SPIRA, Rev Infect Dis 12 $24-$29 (1990). M. SALASPURO, Acta Med Scand 70____33(Suppl.) 219-224 (1985). L.Z. BENET, Scand J Gastroenterol 26 (Suppl. 185) 29-35 (1991).