Evolution of quaternary structure in a homotetrameric enzyme.

Dihydrodipicolinate synthase (DHDPS) is an essential enzyme in (S)-lysine biosynthesis and an important antibiotic target. All X-ray crystal structures solved to date reveal a homotetrameric enzyme. In order to explore the role of this quaternary structure, dimeric variants of Escherichia coli DHDPS were engineered and their properties were compared to those of the wild-type tetrameric form. X-ray crystallography reveals that the active site is not disturbed when the quaternary structure is disrupted. However, the activity of the dimeric enzymes in solution is substantially reduced, and a tetrahedral adduct of a substrate analogue is observed to be trapped at the active site in the crystal form. Remarkably, heating the dimeric enzymes increases activity. We propose that the homotetrameric structure of DHDPS reduces dynamic fluctuations present in the dimeric forms and increases specificity for the first substrate, pyruvate. By restricting motion in a key catalytic motif, a competing, non-productive reaction with a substrate analogue is avoided. Small-angle X-ray scattering and mutagenesis data, together with a B-factor analysis of the crystal structures, support this hypothesis and lead to the suggestion that in at least some cases, the evolution of quaternary enzyme structures might serve to optimise the dynamic properties of the protein subunits.

Assessment of catechol induction and glucuronidation in rat liver microsomes.

Catechols are substances with a 1,2-dihydroxybenzene group from natural or synthetic origin. The aim of this study was to determine whether catechols (4-methylcatechol, 4-nitrocatechol, 2,3-dihydroxynaphthalene) and the antiparkinsonian drugs, entacapone and tolcapone, at doses 150 to 300 mg/kg/day, for 3 days, are able to enhance their own glucuronidation. The induction potency of catechols on rat liver UDP-glucuronosyltransferases (UGTs) was compared with that of a standard polychlorinated biphenyl (PCB) inducer, Aroclor 1254. The glucuronidation rate of these catechols was enhanced up to 15-fold in the liver microsomes of PCB-treated rats, whereas treatment with catechols had little effect. Entacapone, tolcapone, 4-methylcatechol, catechol, 2,3-dihydroxynaphthalene, and 4-nitrocatechol were glucuronidated in control microsomes at rates ranging from 0.12 for entacapone to 22.0 nmol/min/mg for 4-nitrocatechol. Using 1-naphthol, entacapone, and 1-hydroxypyrene as substrates, a 5-, 8-, and 16-fold induction was detected in the PCB rats, respectively, whereas the catechol-induced activities were 1.1- to 1.5-fold only. Entacapone was glucuronidated more efficiently by PCB microsomes than by control microsomes (Vmax/Km, 0.0125 and 0.0016 ml/min/mg protein, respectively). Similar kinetic results were obtained for 1-hydroxypyrene. The Eadie-Hofstee plots suggested the contribution of multiple UGTs for the glucuronidation of 1-hydroxypyrene (Km1, Km2, Km3 = 0.8, 9.7, and 63 microM, and Vmax1, Vmax2, Vmax3 = 11, 24, and 55 nmol/min/mg, respectively), whereas only one UGT could be implicated in the glucuronidation of entacapone (Km = 130 microM, Vmax = 1.6 nmol/min/mg). In conclusion, catechols are poor inducers of their own glucuronidation supported by several UGT isoforms. Their administration is unlikely to affect the glucuronidation of other drugs administered concomitantly.

Glucuronidation of catechols by human hepatic, gastric, and intestinal microsomal UDP-glucuronosyltransferases (UGT) and recombinant UGT1A6, UGT1A9, and UGT2B7.

The substrate specificity of human gastric and intestinal UDP-glucuronosyltransferases (UGTs) toward catechols was investigated and compared to that of liver UGTs. Small catechols were efficiently glucuronidated by stomach (0.8-10.2 nmol/mgprotein x min) and intestine (0.9-7.7 nmol/mgprotein x min) with activities in a range similar to those found in liver (2.9-19 nmol/mgprotein x min). Large interindividual variations were observed among the samples. Immunoblot analysis demonstrated the presence of UGT1A6 and UGT2B7 in stomach and throughout the intestine. Recombinant human UGT1A6, 1A9, and 2B7, stably expressed in mammalian cells, all effectively catalyzed catechol glucuronidation. K(m) values (0.09-13.6mM) indicated low affinity for UGTs and V(max) values ranged from 0.51 to 64.0 nmol/mgprotein x min. These results demonstrate for the first time glucuronidation of catechols by gastric and intestinal microsomal UGTs and three human recombinant UGT isoforms.

Comparison of electrospray, atmospheric pressure chemical ionization, and atmospheric pressure photoionization in the identification of apomorphine, dobutamine, and entacapone phase II metabolites in biological samples.

The applicability of different ionization techniques, electrospray ionization (ESI), atmospheric pressure chemical ionization (APCI), and a novel atmospheric pressure photoionization (APPI), were tested for the identification of the phase II metabolites of apomorphine, dobutamine, and entacapone in rat urine and in vitro incubation mixtures (rat hepatocytes and human liver microsomes). ESI proved to be the most suitable ionization method; it enabled detection of 22 conjugates, whereas APCI and APPI showed only 12 and 14 conjugates, respectively. Methyl conjugates were detected with all ionization methods. Glucuronide conjugates were ionized most efficiently with ESI. Only some of the glucuronides detected with ESI were detected with APCI and APPI. Sulfate conjugates were detected only with ESI. MS/MS experiments showed that the site of glucuronidation or sulfation could not be determined, since the primary cleavage was a loss of the conjugate group (glucuronic acid or SO3), and no site-characteristic product ions were formed. However, it may be possible to determine the site of methylation, since methylated products are more stable than glucuronides or sulfates. Furthermore, the loss of CH3 is not necessarily the primary cleavage, and site characteristic products may be formed. Identification and comparison of conjugates formed from the current model drugs were successfully analyzed in different biological specimens of common interest to biomedical research. A fairly good relation was obtained between the data from in vivo and in vitro models of drug metabolism.

Characterization of catechol glucuronidation in rat liver.

Catechols are a class of substances from natural or synthetic origin that contain a 1,2-dihydroxybenzene group. We have characterized the glucuronidation by rat liver microsomes and by the rat liver recombinant UDP-glucuronosyltransferase isoforms UGT1A6 and UGT2B1 of a series of 42 structurally diverse catechols, including neurotransmitters, polyphenols, drugs, and catechol estrogens. Small catechols (4-nitrocatechol, 2,3-dihydroxybenzaldehyde, 4-methylcatechol, and tetrachlorocatechol), tyrphostine A23, and octylgallate were glucuronidated at the highest rate by rat liver microsomes and the recombinant enzymes. By contrast, polyphenols from green tea (catechin and related compounds), 3,5-dinitrocatechol, the catechol-O-methyltransferase inhibitor drugs (entacapone, nitecapone, and tolcapone), the carboxyl catechols (gallic acid and dihydroxybenzoic acid derivatives), and the neurotransmitters and dopaminergic drugs, except dobutamine, were glucuronidated at low rate. Glucuronidation of most catechols was increased upon treatment of rats by 3-methylcholanthrene (3-MC) or Aroclor 1254. No induction was observed after administration of phenobarbital and clofibrate or treatment with catechols. Partial least-squares modeling was carried out to explain the variations of glucuronidation activity by liver microsomes of nontreated and 3-MC-treated rats. The model developed explained 82% and predicted 61% of the variations of glucuronidation activities. Among the 17 electronic and substructure parameters used that characterize the catechols, the hydrophobicity/molar volume ratio of catechols showed a strong positive correlation with the glucuronidation rate. The effect of the pK(a) of the catechol group was modeled to be nonlinear, the optimal pK(a) value for glucuronidation being between 8 and 9. Hydrogen bonding and steric effects also were important to account for to predict the glucuronidation rates.