Contribution of local regeneration of glucocorticoids to tissue steroid pools.

11ß-Hydroxysteroid dehydrogenase 1 (11ßHSD1) is a drug target to attenuate adverse effects of chronic glucocorticoid excess. It catalyses intracellular regeneration of active glucocorticoids in tissues including brain, liver and adipose tissue (coupled to hexose-6-phosphate dehydrogenase, H6PDH). 11ßHSD1 activity in individual tissues is thought to contribute significantly to glucocorticoid levels at those sites, but its local contribution vs glucocorticoid delivery via the circulation is unknown. Here, we hypothesised that hepatic 11ßHSD1 would contribute significantly to the circulating pool. This was studied in mice with Cre-mediated disruption of Hsd11b1 in liver (Alac-Cre) vs adipose tissue (aP2-Cre) or whole-body disruption of H6pdh. Regeneration of [9,12,12-2H3]-cortisol (d3F) from [9,12,12-2H3]-cortisone (d3E), measuring 11ßHSD1 reductase activity was assessed at steady state following infusion of [9,11,12,12-2H4]-cortisol (d4F) in male mice. Concentrations of steroids in plasma and amounts in liver, adipose tissue and brain were measured using mass spectrometry interfaced with matrix-assisted laser desorption ionisation or liquid chromatography. Amounts of d3F were higher in liver, compared with brain and adipose tissue. Rates of appearance of d3F were ~6-fold slower in H6pdh-/- mice, showing the importance for whole-body 11ßHSD1 reductase activity. Disruption of liver 11ßHSD1 reduced the amounts of d3F in liver (by ~36%), without changes elsewhere. In contrast disruption of 11ßHSD1 in adipose tissue reduced rates of appearance of circulating d3F (by ~67%) and also reduced regenerated of d3F in liver and brain (both by ~30%). Thus, the contribution of hepatic 11ßHSD1 to circulating glucocorticoid levels and amounts in other tissues is less than that of adipose tissue.

Quantification of 11ß-hydroxysteroid dehydrogenase 1 kinetics and pharmacodynamic effects of inhibitors in brain using mass spectrometry imaging and stable-isotope tracers in mice.

11ß-Hydroxysteroid dehydrogenase 1 (11ß-HSD1; EC 1.1.1.146) generates active glucocorticoid hormones. Small molecule inhibitors have been developed to target 11ß-HSD1 for the treatment of dementia; these must enter brain subregions, such as the hippocampus, to be effective. We previously reported mass spectrometry imaging measurement of murine tissue steroids, and deuterated steroid tracer infusion quantification of 11ß-HSD1 turnover in humans. Here, these tools are combined to assess tissue pharmacokinetics and pharmacodynamics of an 11ß-HSD1 inhibitor that accesses the brain. [9,11,12,12-2H]4-Cortisol was infused (1.75 mg/day) by minipump for 2 days into C57Bl6 mice (male, age 12 weeks, n = 3/group) after which an 11ß-HSD1 inhibitor (UE2316) was administered (25 mg/kg oral gavage) and animals culled immediately or 1, 2 and 4 h post-dosing. Mice with global genetic disruption of Hsd11B1 were studied similarly. Turnover of d4-cortisol to d3-cortisone (by loss of the 11-deuterium) and regeneration of d3-cortisol (by 11ß-HSD1-mediated reduction) were assessed in plasma, liver and brain using matrix assisted laser desorption ionization coupled to Fourier transform cyclotron resonance mass spectrometry. The tracer d4-cortisol was detected in liver and brain following a two day infusion. Turnover to d3-cortisone and on to d3-cortisol was slower in brain than liver. In contrast, d3-cortisol was not detected in mice lacking 11ß-HSD1. UE2316 impaired d3-cortisol generation measured in whole body (assessed in plasma; 53.1% suppression in rate of appearance in d3-cortisol), liver and brain. Differential inhibition in brain regions was observed; active glucocorticoids were suppressed to a greater in extent hippocampus or cortex than in amygdala. These data confirm that the contribution of 11ß-HSD1 to the tissue glucocorticoid pool, and the consequences of enzyme inhibition on active glucocorticoid concentrations, are substantial, including in the brain. They further demonstrate the value of mass spectrometry imaging in pharmacokinetic and pharmacodynamic studies.

Overexpression of human kynurenine-3-monooxygenase protects against 3-hydroxykynurenine-mediated apoptosis through bidirectional nonlinear feedback.

Kynurenine 3-monooxygenase (KMO) is a critical regulator of inflammation. The preferred KMO substrate, kynurenine, is converted to 3-hydroxykynurenine (3HK), and this product exhibits cytotoxicity through mechanisms that culminate in apoptosis. Here, we report that overexpression of human KMO with orthotopic localisation to mitochondria creates a metabolic environment during which the cell exhibits increased tolerance for exogenous 3HK-mediated cellular injury. Using the selective KMO inhibitor Ro61-8048, we show that KMO enzyme function is essential for cellular protection. Pan-caspase inhibition with Z-VAD-FMK confirmed apoptosis as the mode of cell death. By defining expression of pathway components upstream and downstream of KMO, we observed alterations in other key kynurenine pathway components, particularly tryptophan-2,3-dioxygenase upregulation, through bidirectional nonlinear feedback. KMO overexpression also increased expression of inducible nitric oxide synthase (iNOS). These changes in gene expression are functionally relevant, because siRNA knockdown of the pathway components kynureninase and quinolinate phosphoribosyl transferase caused cells to revert to a state of susceptibility to 3HK-mediated apoptosis. In summary, KMO overexpression, and importantly KMO activity, have metabolic repercussions that fundamentally affect resistance to cell stress.

Bacterial expression of human kynurenine 3-monooxygenase: solubility, activity, purification.

Kynurenine 3-monooxygenase (KMO) is an enzyme central to the kynurenine pathway of tryptophan metabolism. KMO has been implicated as a therapeutic target in several disease states, including Huntington's disease. Recombinant human KMO protein production is challenging due to the presence of transmembrane domains, which localise KMO to the outer mitochondrial membrane and render KMO insoluble in many in vitro expression systems. Efficient bacterial expression of human KMO would accelerate drug development of KMO inhibitors but until now this has not been achieved. Here we report the first successful bacterial (Escherichia coli) expression of active FLAG™-tagged human KMO enzyme expressed in the soluble fraction and progress towards its purification.

Antidiabetic and toxicological evaluations of naringenin in normoglycaemic and NIDDM rat models and its implications on extra-pancreatic glucose regulation.

AIM: The present investigation was designed to determine the in vivo antidiabetic effect of naringenin (NG) in normoglycaemic and diabetic rat models through blood glucose (GLU) measurements following acute and subchronic time periods. Possible modes of action of NG were investigated and its acute toxicity determined. METHODS: Normoglycaemic and non-insulin-dependent diabetes mellitus (NIDDM) rat models were treated for acute and subchronic (5 days) time periods with 50 mg/kg/day of NG. Blood biochemical profiles were determined after 5 days of the treatment in normoglycaemic and NIDDM rats using commercial kits for GLU, triglycerides (TG), total cholesterol (CHOL) and high-density lipoprotein (HDL). In order to elucidate its antidiabetic mode of action, NG was administered intragastrically and an oral glucose tolerance test performed using GLU and sucrose (2 g/kg) as substrates. The inhibitory effect of a single concentration of NG (10 microM) on 11beta-hydroxysteroid dehydrogenase type 1 (11beta-HSD1) activity in vitro was determined. Finally, the preclinical safety and tolerability of NG was determined by toxicological evaluation in mice and rats using Organization for Economic Cooperation and Development (OECD) protocols. RESULTS: Intragastrically administered NG (50 mg/kg) induced a significant decrease in plasma GLU in normoglycaemic and NIDDM rat models (p < 0.05) following acute and subchronic time periods. After 5 days of administration, NG produced significant diminished blood GLU and TG levels in streptozotocin-nicotinamide-induced diabetic rats. The administration of NG to normal rats significantly increased the levels of TG, CHOL and HDL (p < 0.05). NG (5 and 50 mg/kg) induced a total suppression in the increase of plasma GLU levels after administration of substrates (p < 0.01), but NG did not produce inhibition of alpha-glucosidase activity in vitro. However, NG (10 microM) was shown to inhibit 11beta-HSD1 activity by 39.49% in a cellular enzyme assay. Finally, NG showed a Medium Lethal Dose LD(50) > 5000 mg/kg and ranking at level five based on OECD protocols. CONCLUSION: Our findings suggest that NG may exert its antidiabetic effect by extra-pancreatic action and by suppressing carbohydrate absorption from intestine, thereby reducing the postprandial increase in blood GLU levels.

Probing the NADPH-binding site of Escherichia coli flavodoxin oxidoreductase.

The structure of the Escherichia coli flavodoxin NADP(+) oxidoreductase (FLDR) places three arginines (R144, R174 and R184) in the proposed NADPH-binding site. Mutant enzymes produced by site-directed mutagenesis, in which each arginine was replaced by neutral alanine, were characterized. All mutants exhibited decreased NADPH-dependent cytochrome c reductase activity (R144A, 241.6 min(-1); R174A, 132.1 min(-1); R184A, 305.5 min(-1) versus wild type, 338.9 min(-1)) and increased K(m) for NADPH (R144A, 5.3 microM; R174A, 20.2 microM; R184A, 54.4 microM versus wild type, 3.9 microM). The k(cat) value for NADH-dependent cytochrome c reduction was increased for R174A (42.3 min(-1)) and R184A (50.4 min(-1)) compared with the wild type (33.0 min(-1)), consistent with roles for R174 and R184 in discriminating between NADPH/NADH by interaction with the adenosine ribose 2'-phosphate. Stopped-flow studies indicated that affinity (K(d)) for NADPH was markedly reduced in mutants R144A (635 microM) and R184A (2.3 mM) compared with the wild type (<5 microM). Mutant R184A displays the greatest change in pyridine nucleotide preference, with the NADH/NADPH K(d) ratio >175-fold lower than for wild-type FLDR. The rate constant for hydride transfer from NADPH to flavin was lowest for R174A (k(red)=8.82 s(-1) versus 22.63 s(-1) for the wild type), which also exhibited tertiary structure perturbation, as evidenced by alterations in CD and fluorescence spectra. Molecular modelling indicated that movement of the C-terminal tryptophan (W248) of FLDR is necessary to permit close approach of the nicotinamide ring of NADPH to the flavin. The positions of NADPH phosphates in the modelled structure are consistent with the kinetic data, with R174 and R184 located close to the adenosine ribose 2'-phosphate group, and R144 likely to interact with the nicotinamide ribose 5'-phosphate group.

Mechanism of 8-amino-7-oxononanoate synthase: spectroscopic, kinetic, and crystallographic studies.

8-Amino-7-oxononanoate synthase (also known as 7-keto-8-aminopelargonate synthase, EC 2.3.1.47) is a pyridoxal 5'-phosphate-dependent enzyme which catalyzes the decarboxylative condensation of L-alanine with pimeloyl-CoA in a stereospecific manner to form 8(S)-amino-7-oxononanoate. This is the first committed step in biotin biosynthesis. The mechanism of Escherichia coli AONS has been investigated by spectroscopic, kinetic, and crystallographic techniques. The X-ray structure of the holoenzyme has been refined at a resolution of 1.7 A (R = 18.6%, R(free) = 21. 2%) and shows that the plane of the imine bond of the internal aldimine deviates from the pyridine plane. The structure of the enzyme-product external aldimine complex has been refined at a resolution of 2.0 A (R = 21.2%, R(free) = 27.8%) and shows a rotation of the pyridine ring with respect to that in the internal aldimine, together with a significant conformational change of the C-terminal domain and subtle rearrangement of the active site hydrogen bonding. The first step in the reaction, L-alanine external aldimine formation, is rapid (k(1) = 2 x 10(4) M(-)(1) s(-)(1)). Formation of an external aldimine with D-alanine, which is not a substrate, is significantly slower (k(1) = 125 M(-)(1) s(-)(1)). Binding of D-alanine to AONS is enhanced approximately 2-fold in the presence of pimeloyl-CoA. Significant substrate quinonoid formation only occurs upon addition of pimeloyl-CoA to the preformed L-alanine external aldimine complex and is preceded by a distinct lag phase ( approximately 30 ms) which suggests that binding of the pimeloyl-CoA causes a conformational transition of the enzyme external aldimine complex. This transition, which is inferred by modeling to require a rotation around the Calpha-N bond of the external aldimine complex, promotes abstraction of the Calpha proton by Lys236. These results have been combined to form a detailed mechanistic pathway for AONS catalysis which may be applied to the other members of the alpha-oxoamine synthase subfamily.

The crystal structure of 8-amino-7-oxononanoate synthase: a bacterial PLP-dependent, acyl-CoA-condensing enzyme.

8-Amino-7-oxononanoate synthase (or 8-amino-7-ketopelargonate synthase; EC 2.3.1.47; AONS) catalyses the decarboxylative condensation of l-alanine and pimeloyl-CoA in the first committed step of biotin biosynthesis. We have cloned, over-expressed and purified AONS from Escherichia coli and determined the crystal structures of the apo and PLP-bound forms of the enzyme. The protein is a symmetrical homodimer with a tertiary structure and active site organisation similar to, but distinct from, those of other PLP-dependent enzymes whose three-dimensional structures are known. The critical PLP-binding lysine of AONS is located at the end of a deep cleft that allows access of the pantothenate arm of pimeloyl-CoA. A cluster of positively charged residues at the entrance to this cleft forms a putative diphosphate binding site for CoA. The structure of E. coli AONS enables identification of the key residues of the PLP-binding site and thus provides a framework with which to understand the biochemical mechanism, which is similar to that catalysed by 5-aminolevulinate synthase and two other alpha-oxoamine synthases. Although AONS has a low overall sequence similarity with the catalytic domains of other alpha-oxoamine synthases, the structure reveals the regions of significant identity to be functionally important. This suggests that the organisation of the conserved catalytic residues in the active site is similar for all enzymes of this sub-class of PLP-dependent enzymes and they share a common mechanism. Knowledge of the three-dimensional structure of AONS will enable characterisation of the structural features of this enzyme sub-family that are responsible for this important type of reaction.