Proton pump inhibitor use and risk of spontaneous bacterial peritonitis in cirrhotic patients: a systematic review and meta-analysis

We used a meta-analysis approach to investigate the association between proton pump inhibitor (PPI) use and risk of spontaneous bacterial peritonitis (SBP) in cirrhotic patients. We searched Ovid Medline, Embase, and the Cochrane Library to identify eligible studies. We included studies that compared cirrhotic patients who did or did not use PPIs. The primary outcome was SBP, and the secondary outcome was overall bacterial infection. Results were pooled using random-effect models. This process led to identification of 12 journal articles and 5 conference abstracts. The pooled data showed that PPI use in patients with cirrhosis and ascites was significantly associated with an increased risk of SBP [odds ratio (OR) = 2.17; 95% confidence interval (CI) = 1.46-3.23; P < 0.05; I2 = 85.6%] and overall risk of bacterial infection (OR = 1.98; 95%CI = 1.36-2.87; P < 0.05; I2 = 0). Subgroup analysis revealed that journal articles and studies reporting adjusted effect estimates demonstrated that PPI users had a significantly increased risk of SBP (OR = 2.13; 95%CI = 1.61-2.82; P < 0.05; I2 = 29.4%; and OR = 1.98; 95%CI = 1.42-2.77; P < 0.05; I2 = 67%, respectively). In conclusion, PPI use increased the risk of SBP and overall bacterial infection in patients with cirrhosis and ascites. PPIs should be administered after careful assessment of the indications in cirrhotic patients. Future well-designed prospective studies are warranted to clarify the dose relationships and to compare infection risks associated with different classes of PPIs.

Effects of N-terminal modification of recombinant human cytochrome P450 1A2 on catalytic activity

1. The high-level expression of mammalian cytochrome P450 in bacteria usually requires modification of the amino-terminal region of the enzyme. The effect of altering amino acids in the N-terminus of human recombinant CYP1A2 on its catalytic activity was investigated herein. 2. Rates of 7-ethoxyresorufin O-deethylation by CYP1A2a (a form made by altering the amino acids LLL of CYP1A2 to RER at positions 3-5) in reconstituted systems were significantly low compared with those of other CYP1A2 N-terminal variants at a low ratio of cytochrome P450 to NADPH-cytochrome P450 reductase, but not at higher reductase concentrations. 3. CYP1A2a-dependent ethoxyresorufin O-deethylase activity in a cumene hydroperoxide-supported system was approximately 2-fold higher than other CYP1A2 N-terminal variants. 4. Our results suggest that modification of three N-terminal amino acids in CYP1A2 alters the interaction between CYP1A2 and the reductase in reconstituted phospholipid vesicles and in the bicistronic membranes.

Characterization of the selectivity and mechanism of cytochrome P450 inhibition by dimethyl-4,4'-dimethoxy-5,6,5',6'-dimethylenedioxybiphenyl-2,2'-dicarboxylate

In vitro studies with human liver microsomes and cytochrome P450 (P450) prototype substrates were performed to characterize the selectivity and mechanism of inhibition of P450 by dimethyl-4,4'-dimethoxy-5,6,5',6'-dimethylenedioxybiphenyl-2,2'-dicarboxylate (DDB). DDB was found to be a strong inhibitor of testosterone 6beta-hydroxylation activity (CYP3A4) with a K(i) value of 0.27 +/- 0.21 microM. At higher concentrations, DDB marginally inhibited caffeine N(3)-demethylation (CYP1A2), diclofenac 4'-hydroxylation (CYP2C9), and dextromethorphan O-demethylation (CYP2D6) activities, but this compound had no effect on CYP2A6-, CYP2C19-, and CYP2E1-mediated reactions. Spectral analysis indicated that the formation of metabolite-P450 complex having absorbance at 456 nm was concentration-dependent; 5 to 33% of the total P450 was complexed in rat and human liver microsomes after a 5-min incubation with DDB. In addition, microsomal incubations with DDB in the presence of NADPH resulted in a loss of spectral P450 content, which was restored after adding K(3)Fe(CN)(6). This complex formation resulted in a time-dependent loss of CYP3A-catalyzed marker activity (testosterone 6beta-hydroxylation) in human liver microsomes. The inhibition was only partially restored upon dialysis. These results collectively suggest that formation of a metabolite-CYP3A complex with DDB was responsible for the CYP3A-selective time-dependent loss of catalytic function of CYP3A.

Glutathione S-transferase mu modulates the stress-activated signals by suppressing apoptosis signal-regulating kinase 1

Apoptosis signal-regulating kinase 1 (ASK1) is a mitogen-activated protein kinase kinase kinase that can activate the c-Jun N-terminal kinase and the p38 signaling pathways. It plays a critical role in cytokine- and stress-induced apoptosis. To further characterize the mechanism of the regulation of the ASK1 signal, we searched for ASK1-interacting proteins employing the yeast two-hybrid method. The yeast two-hybrid assay indicated that mouse glutathione S-transferase Mu 1-1 (mGSTM1-1), an enzyme involved in the metabolism of drugs and xenobiotics, interacted with ASK1. We subsequently confirmed that mGSTM1-1 physically associated with ASK1 both in vivo and in vitro. The in vitro binding assay indicated that the C-terminal portion of mGSTM1-1 and the N-terminal region of ASK1 were crucial for binding one another. Furthermore, mGSTM1-1 suppressed stress-stimulated ASK1 activity in cultured cells. mGSTM1-1 also blocked ASK1 oligomerization. The ASK1 inhibition by mGSTM1-1 occurred independently of the glutathione-conjugating activity of mGSTM1-1. Moreover, mGSTM1-1 repressed ASK1-dependent apoptotic cell death. Taken together, our findings suggest that mGSTM1-1 functions as an endogenous inhibitor of ASK1. This highlights a novel function for mGSTM1-1 insofar as mGSTM1-1 may modulate stress-mediated signals by repressing ASK1, and this activity occurs independently of its well-known catalytic activity in intracellular glutathione metabolism.

Influence of ligand binding to human cytochrome P-450 1A2: conformational activation and stabilization by alpha-naphthoflavone

Human cytochrome P-450 (P-450) 1A2 expressed in Escherichia coli is readily converted into non-native cytochrome P-420 (P-420) in the presence of detergents. alpha-Naphthoflavone (ANF) has been used to prevent P-450 1A2 inactivation to P-420 during purification. However, the mechanism by which ANF modulates P-450 1A2 is not clearly understood. We observed that recombinant human P-450 1A2 prepared in the absence of ANF has an approx. 5 times higher maximum catalytic activity in the O-deethylation of 7-ethoxycoumarin than that in the presence of ANF, with the same K(m) values. The results revealed that the enzyme purified with ANF is not catalytically fully active, indicating that ANF tightly binds to the enzyme, only to be dissociated by heat denaturation. Furthermore, the inactive P-420 form of the enzyme could be reconverted to P-450 by ANF in high concentrations of detergents. The reconversion was concentration-dependent, confirming ANF-induced regeneration of active P-450 1A2. The reconversion coincided with the conformational change of the enzyme including increased alpha-helix content. The conformation of P-450 1A2 was also stabilized by ANF, resulting in an approx. 5 degrees C increase in thermal stability.

Phenotypes of flavin-containing monooxygenase activity determined by ranitidine N-oxidation are positively correlated with genotypes of linked FM03 gene mutations in a Korean population

A non-invasive urine analysis method to determine the in-vivo flavin-containing mono-oxygenase (FMO) activity catalysing N-oxidation of ranitidine (RA) was developed and used to phenotype a Korean population. FMO activity was assessed by the molar concentration ratio of RA and RANO in the bulked 8 h urine. This method was used to determine the FMO phenotypes of 210 Korean volunteers (173 men and 37 women, 110 nonsmokers and 100 smokers). Urinary RA/RANO ratio, representing the metabolic ratio and the reciprocal index of FMO activity, ranged from 5.67-27.20 (4.8-fold difference) and was not different between men and women (P = 0.76) or between smokers and nonsmokers (P = 0.50). The frequencies of RA/RANO ratios were distributed in a trimodal fashion. Among the 210 Korean subjects, 93 (44.3%) were fast metabolizers, 104 (49.5%) were intermediate metabolizers and 13 (6.2%) were slow metabolizers. Subsequently, the relationship between the ranitidine N-oxidation phenotypes and FMO3 genotypes, determined by the presence of two previously identified mutant alleles (Glu158Lys: FMO3/Lys158 and Glu308Gly: FMO3/Gly308 alleles) commonly found in our Korean population was examined. The results showed that subjects who were homozygous and heterozygous for either one or both of the FMO3/Lys158 and FMO3/Gly308 mutant alleles had significantly lower in-vivo FMO activities than those with homozygous wild-type alleles (FMO3/Glu158 and FMO3/Glu308) (P < 0.001, Mann-Whitney U-test). Furthermore, the FMO activities of subjects with either FMO3/Lys158 or FMO3/Gly308 mutant alleles were almost identical to those having both FMO3 mutant alleles (FMO3/Lys158 and FMO3/Gly308). These two mutant alleles located, respectively, at exons 4 and 7 in the FMO3 gene appeared to be strongly linked by cis-configuration in Koreans. Therefore, we concluded that presence of FMO3/Lys158 and FMO3/Gly308 mutant alleles in FMO3 gene is responsible for the low ranitidine N-oxidation (FMO3 activity) in our Korean population.

Reconstitution of recombinant cytochrome P450 2C10(2C9) and comparison with cytochrome P450 3A4 and other forms: effects of cytochrome P450-P450 and cytochrome P450-b5 interactions

Tolbutamide methyl hydroxylation and S-warfarin 7-hydroxylation activities were reconstituted in systems containing recombinant human cytochrome P450 (P450 or CYP) 2C10(2C9) and the optimal conditions for the systems were compared with those of bufuralol 1'-hydroxylation by CYP1A1, theophylline 8-hydroxylation by CYP1A2, bufuralol 1'-hydroxylation by CYP2D6, chlorzoxazone 6-hydroxylation by CYP2E1, and testosterone 6 beta-hydroxylation by CYP3A4. CYP2C10 required cytochrome b5 (b5) for optimal rates of tolbutamide and S-warfarin oxidations and b5 could be replaced by apo-b5; apo-b5 and b5 effects on the reconstituted systems have already been reported in systems containing CYP3A4 for the oxidation of testosterone and nifedipine and for the rapid reduction of CYP3A4 by NADPH-P450 reductase (H. Yamazaki et al., 1996, J. Biol. Chem. 271, 27438-27444). Stopped-flow studies, however, suggested that apo-b5 as well as b5 did not cause stimulation of the reduction of CYP2C10 by NADPH-P450 reductase, while the reduction rates were dependent on the substrates in reconstituted systems. Chlorzoxazone 6-hydroxylation by CYP2E1 was stimulated by b5, but not by apo-b5, in reconstituted systems. Neither apo- nor holo-b5 increased bufuralol 1'-hydroxylation activity by CYP1A1 or 2D6 or theophylline 8-hydroxylation by CYP1A2. Interestingly, we found that testosterone 6 beta-hydroxylation by CYP3A4 was stimulated by CYP1A2 (and also by a modified form in which the first 36 residues of the native human protein were removed) and CYP1A1 as well as by b5, and such stimulations were not seen when other P450 proteins (e.g., CYP2C10, 2D6, or 2E1) were added to the reconstituted systems. In contrast, substrate oxidations by CYP2C10 and CYP2E1 were not stimulated by other P450 proteins. The present results suggest that there are differences in optimal conditions for reconstitution of substrate oxidations by various forms of human P450 enzymes, and in some P450-catalyzed reactions protein-protein interactions between P450 and b5 and other P450 proteins are very important in some oxidations catalyzed by CYP2C10, 2E1, and 3A4.

Identification of retained N-formylmethionine in bacterial recombinant mammalian cytochrome P450 proteins with the N-terminal sequence MALLLAVFL...: roles of residues 3-5 in retention and membrane topology

An N-terminal block to Edman degradation was observed when any of five different mammalian cytochrome P450 (P450) proteins was expressed in Escherichia coli using the N-terminal sequence MALLLAVFL... This block was also seen in Salmonella typhimurium. With all proteins examined, the block could be removed by mild acid hydrolysis (0.6--6 N HCl, 23 degrees C) to expose Met as the N-terminus, suggesting N-formylMet retention. The N-terminal peptide of a modified P450 1A2 ("mutant 1", containing a thrombin-sensitive site inserted at residue 25) was released with thrombin and analyzed by electrospray mass spectrometry and found to yield the M(r) expected for the N-formyl derivative (+/- 0.8 amu). The region of positions 3--5 was altered by random mutagenesis, and three P450 1A2-expressing clones were analyzed for nucleotide and amino acid sequences. The changes from LLL were to RER (P450 1A2a), VDS (P450 1A2b), and WRH (P450 1A2c); these all show slightly dissimilar hydropathy plots compared to the MALLLAVFL... sequence. Mutant P450 1A2a had the N-terminal Met removed to yield N-terminal Ala; P450 1A2b contained an unmodified Met at the N-terminus; P450 1A2c had an approximately 80% block of the N-terminal Met. Experiments with bacterial membranes containing expressed P450 1A2 mutant 1 and P450 1A2 mutant 2 (thrombin-sensitive site inserted at residue 46) suggest that thrombin site 2, but not 1, is sequestered in the membrane. Spheroplasts of bacteria expressing P450 1A2 and the mutants at positions 3--5 were treated with proteinase K; amino acid analysis indicated that no cleavage occurred. These results are interpreted in a model in which most of the mammalian P450 expressed in the bacterium is located in the cytosol, the region near residue 46 is in the inner membrane, the region near residue 25 is in the cytosol, and the N-terminus is either imbedded in the membrane or free in the cytosolic space, depending upon the sequence. However, the possibility that the differences in N-terminal processing are the result of direct changes in interactions with the deformylase and Met aminopeptidase cannot be excluded.

Recombinant human cytochrome P450 1A2 and an N-terminal-truncated form: construction, purification, aggregation properties, and interactions with flavodoxin, ferredoxin, and NADPH-cytochrome P450 reductase

Previous work from this laboratory indicated that the N-terminus of recombinant human cytochrome P450 (P450) 1A2 expressed in Escherichia coli is blocked (P. Sandhu, Z. Guo, T. Baba, M. V. Martin, R. H. Tukey, and F. P. Guengerich, (1994) Arch. Biochem. Biophys. 30, 168 -177). A modification of this construct was done to insert an extra 12 residues containing a thrombin-sensitive site just beyond the most N-terminal hydrophobic segment, and the protein was expressed, purified, and cut with thrombin. Treatment of E. coli membranes in which the P450 1A2 with 12 extra residues was present with thrombin did not release the truncated form, suggesting that the added thrombin site may be imbedded in the membrane. The N-terminal of the recombinant proteins were blocked but mild acid hydrolysis generated the expected Met residues as analyzed by Edman degradation. Laser light scattering studies indicated that purified thrombin-cleaved P450 1A2 (devoid of the usual N-terminal 25 residues or the first 36 residues of the wild-type protein) was still aggregated in the absence of detergent and that some nondenaturing detergents could reduce the apparent size to that of a tetramer. The N-terminal truncated protein was as catalytically active as full-length P450 1A2 but required a higher concentration of NADPH-P450 reductase. P450 1A2 exhibited catalytic activity in E. coli cells, and activity of the purified enzyme could be supported by E. coli flavodoxin and NADPH-flavodoxin reductase. Spinach ferredoxin and NADPH-ferredoxin reductase could also substitute for NADPH-P450 reductase. These artificial electron donors did not require phospholipid for oxidation reactions; however, phospholipid was required for optimal activity when either P450 1A2 or the truncated form was used with NADPH-P450 reductase. Rates of oxidation of 7-ethoxyresorufin were considerably higher for both P450 1A2 and the truncated form when NADPH-P450 reductase was replaced with the "oxygen surrogate" iodosylbenzene, indicating that P450 reduction and oxygen activation are normally limiting in this P450 1A2 reaction.

Bioactivation of aromatic amines by recombinant human cytochrome P4501A2 expressed in Ames tester strain bacteria: a substitute for activation by mammalian tissue preparations

The most widely used bioassay in genetic toxicology is the Ames test, which combines a bacterial mutagenicity assay (reversion of Salmonella typhimurium histidine-auxotrophic tester strains) with an exogenous bioactivation system (hepatic postmitochondrial supernatant or "S9"). The enzymatic activities of S9 prepared from the tissues of experimental animals are difficult to control. We show that the requirement for S9 can be obviated by the engineered expression of enzymes of bioactivation within the bacterial cell. With this strategy, reactive metabolites are produced inside the bacterial cell, proximate to the genetic target. Species boundaries can be crossed, and chimeric or mutant enzymes can be studied. We have constructed an Ames tester strain, expressing both aromatic amine N-acetyltransferase and human cytochrome P4501A2, which detects aromatic amine mutagenicity in the absence of S9.

Induction of physical dependence by chronic alcohol treatment and enhancement of hepatic metabolism of 7-ethoxycoumarin and subsequent conjugation of its metabolite in perfused rat livers

Physical dependence to alcohol was induced in rats and the rates of hepatic oxidation of 7-ethoxycoumarin to 7-hydroxycoumarin and of the subsequent conjugation of 7-hydroxycoumarin were studied using isolated perfused livers. Seven to 11.2 g of ethanol per kg body weight per day were given to rats in 2 divided doses at 10 a.m. and at 8 p.m. for 7, 9, 11, and 13 consecutive days. All rats surviving these courses of ethanol administration have demonstrated moderate degrees of withdrawal signs. In the isolated perfused livers obtained from these alcohol dependent rats, 100 microM 7-ethoxycoumarin was infused. The rate of 7-hydroxycoumarin formation was increased from 36 to 50 nmoles per min per g of liver tissue. This increase has occurred without significant changes of the liver weights. With such increase in the rate of mixed function oxidation, greater amounts of the 7-hydroxycoumarin metabolite were conjugated not only to glucuronide (from 4.9 to 6.7 nmoles), but also to sulfate ester (from 27.6 to 42.1 nmoles). Furthermore, a greater percentage of the produced metabolite was conjugated in the ethanol treated rat liver. A high concentration of 7-hydroxycoumarin (100 microM) was also infused in order to determine the effect of chronic alcohol administration on the maximal available rates of conjugation. Results indicated that the maximal rates of conjugation were significantly decreased. However, these decreased conjugation rates were still much greater than the maximal production rate for 7-hydroxycoumarin from the infused 7-ethoxycoumarin, for which the rate of oxidation has been elevated by the chronic ethanol treatment. Thus, the livers obtained from rats chronically treated with alcohol have increased rates for both the initial drug oxidation and the subsequent conjugation, even though the maximal rates of conjugation, per se, are reduced.

Functional relationship between initial oxidation of 7-ethoxycoumarin and subsequent conjugation of 7-hydroxycoumarin in isolated perfused rat livers

Functional relationship between the initial mixed function oxidation of 7-ethoxycoumarin (EC) to 7-hydroxycoumarin (HC) and the subsequent conjugation of this metabolite to sulfate ester and glucuronide has been studied using isolated perfused rat livers. When increasing concentrations of EC (from 25 to 200 microM) were infused, perfused liver can oxidize only up to about 60 nmol of the infused EC to HC per min/g liver tissue. Most of this HC metabolite was released as sulfate ester, but there was a dose dependent shift to a more significant glucuronidation at the expense of the sulfate form. The dose dependent shift observed upon infusions with increasing dose of EC was not extensive so that the major portion of metabolite released was always the sulfate ester. However, the shift observed with HC was extensive and the major portion released was the glucuronide conjugate. Upon infusions with increasing concentrations of HC, the maximal rates of sulfation and glucuronidation were found to be 60 nmol and 120 nmol of HC conjugated per min/g liver tissue, respectively. Furthermore, the ranges in the rates of conjugation for the infused HC were divided into a sulfate ester 'zone' (less than 20 nmol), a dose-dependent shift 'zone' (between 20 and 180 nmol) with the 'cross-over' occurring at 80 nmol/min/g liver, and reaching the maximal conjugation 'capacity' rate (180 nmol), above which the unconjugated free form of HC was released. Under conditions when EC was infused into normal rat livers, the calculated maximal oxidation rate was only 60 nmol of HC produced/min/g liver. Consequently, under such a condition, the oxidation rate may never reach the 'cross-over' rate and this explains the lack of extensive dose-dependent shift and further indicates that there remained a large reserve conjugation capacity (120 nmol/min/g).