Enzymatic assay for perfluoro-tagged metabolites of l-DOPA using crude lysate from E. coli transformed with pKKAADCII

Isomers of l-DOPA and dopamine with a nine-atom ¹⁹F atom tag were exposed to aromatic acid decarboxylase (AADC) in the lysate of Escherichia coli JM109 that had been transformed with the plasmid pKKAADCII. The resulting samples were analyzed with HPLC. The first study investigated the conversion of the tagged l-DOPA into tagged dopamine, using the tagged dopamine as a standard. A second study was undertaken to identify the source of peaks seen in the enzymatic assays. l-DOPA with the tag bonded at position 5 served as the best substrate for AADC. Isomers that fit into the active site of AADC are likely to follow the biosynthetic path for dopamine in vivo and are potentially useful in magnetic resonance studies. The enzymatic assay described here provides an efficient and cost-effective tool for screening new compounds for use in the fluorine imaging of neural pathways.     

Direct and decarboxylation-dependent effects of neurotransmitter precursors on firing of isolated monoaminergic neurons

To elucidate mechanisms that underlie the profound physiological effects of the monoamine precursors 5-hydroxy-l-tryptophan (5-HTP) and l-3,4-dihydroxyphenylalanine (l-DOPA), we examined their action on single monoaminergic neurons isolated from the ganglia of the gastropod snail Lymnaea stagnalis. In isolated serotonergic PeA motoneurons, 5-HTP produced excitation. The effect was mimicked by serotonin at 0.5–1 μM, masked by pretreatment with serotonin at higher concentrations, and abolished by the inhibitor of aromatic amino acid decarboxylase (AAAD) m-hydroxybenzylhydrazine (NSD-1015), the inhibitor of the vesicular monoamine transporter reserpine or the serotonin receptor antagonist mianserin. Exposure of the dopaminergic interneurons RPeD1 to l-DOPA caused a biphasic effect composed of a depolarization followed by a hyperpolarization. AAAD inactivation with NSD-1015, as well as the blockade of dopamine receptors with sulpiride, resulted in the enhancement of the excitatory effect, and the abolition of the inhibitory effect. Dopamine caused hyperpolarization and masked the inhibitory phase of l-DOPA action. The results show that precursors affect the rate of firing of isolated monoaminergic neurons and that their effect is completely or partially mediated by the enhanced synthesis of the respective neurotransmitter, followed by extrasynaptic release of the latter and activation of extrasynaptic autoreceptors.     

Purification and characterization of a novel l-2-amino-Δ2-thiazoline-4-carboxylic acid hydrolase from Pseudomonas sp. strain ON-4a expressed in E. coli

l-2-Amino-Δ²-thiazoline-4-carboxylic acid hydrolase (ATC hydrolase) was purified and characterized from the crude extract of Escherichia coli, in which the gene for ATC hydrolase of Pseudomonas sp. strain ON-4a was expressed. The results of SDS–polyacrylamide gel electrophoresis and gel filtration on Sephacryl S-200 suggested that the ATC hydrolase was a tetrameric enzyme consisted of identical 25-kDa subunits. The optimum pH and temperature of the enzyme activity were pH 7.0 and 30–35°C, respectively. The enzyme did not require divalent cations for the expression of the activity, and Cu²⁺ and Mn²⁺ ions strongly inhibited the enzyme activity. An inhibition experiment by diethylpyrocarbonic acid, 2-hydroxy-5-nitrobenzyl bromide, and N-bromosuccinimide suggested that tryptophan, cysteine, or/and histidine residues may be involved in the catalytic site of this enzyme. The enzyme was strictly specific for the l-form of d,l-ATC and exhibited high activity for the hydrolysis of l-ATC with the values of K _m (0.35 mM) and V _max (69.0 U/mg protein). This enzyme could not cleave the ring structure of derivatives of thiazole, thiazoline, and thiazolidine tested, except for d,l- and l-ATC. These results show that the ATC hydrolase is a novel enzyme cleaving the carbon–sulfur bond in a ring structure of l-ATC to produce N-carbamoyl-l-cysteine.     

Effect of administration of 5-hydroxytryptophan and an inhibitor ofl-aromatic amino acid decarboxylase on glucose metabolism in rat brain

The effect of 5-hydroxytryptophan (5-HTP)—the precursor of serotonin (5-hydroxytryptamine, 5-HT)—and of an inhibitor,N-(dl-seryl)-N-(2,3,4-trihydroxybenzyl)hydrazine (Ro4-4602), ofl-aromatic amino acid decarboxylase on the metabolism of glucose to amino acids in brain tissue was investigated. Labeled glucose (20 Ci, 0.24 mg in 0.2 ml 0.9% saline) was injected intravenously into fed rats pretreated with Ro4-4602 (50 mg/kg intraperitoneally) either alone or in combination with 5-HTP (30 mg/kg intravenously) or with the appropriate vehicle. After the injection of Ro4-4602 plus 5-HTP, the concentrations of 5-HT and 5-HTP in brain were increased, but the increase of 5-HTP that Ro4-4602 slightly inhibits the reaction of decarboxylation in the brain, although at the dose used the drug is usually considered to act only peripherally. After administration of Ro4-4602 alone or combined with 5-HTP, the concentration of glucose in plasma was not significantly increased. However, the concentration of glucose in brain was markedly increased with such treatments. The administration of Ro4-4602 alone or combined with 5-HTP reduced the flux of¹⁴C from labeled glucose to amino acids in brain. The concentrations of amino acids in brain were little changed by these treatments.     

The Roles of Striatal Serotonin and <Emphasis Type="SmallCaps">l</Emphasis>-Amino-acid Decarboxylase on <Emphasis Type="SmallCaps">l</Emphasis>-DOPA-induced Dyskinesia in a Hemiparkinsonian Rat Model

The administration of l-DOPA is the standard treatment for Parkinson’s disease (PD). However, the symptomatic relief provided by long-term administration may be compromised by l-DOPA-induced dyskinesia (LID) that presents as adverse fluctuations in motor responsiveness and progressive loss of motor control. In the later stages of PD, raphestriatal serotonin neurons compensate for the loss of nigrostriatal dopamine (DA) neurons by converting and releasing DA derived from exogenous l-DOPA. Since the serotonin system does not have an autoregulatory mechanism for DA, raphe-mediated striatal DA release may fluctuate dramatically and precede the development of LID. The 6-hydroxydopamine lesioned rats were treated with l-DOPA (6 mg/kg) and benserazide (15 mg/kg) daily for 3 weeks to allow for the development of abnormal involuntary movement score (AIMs). In rats with LID, chronic treatment with l-DOPA increased striatal DA levels compared with control rats. We also observed a relative increase in the expression of striatal l-amino-acid decarboxylase (AADC) in LID rats, even though tyrosine hydroxylase (TH) expression did not increase. The administration of l-DOPA also increased striatal serotonin immunoreactivity in LID rats compared to control rats. Striatal DA and 5-hydroxytryptamine (5-HT) levels were negatively correlated in l-DOPA-treated rats. These results of this study reveal that 5-HT contributes to LID. Striatal DA positively influences LID, while 5-HT is negatively associated with LID. Finally, we suggest that by strategic modification of the serotonin system it may be possible to attenuate the adverse effects of chronic l-DOPA therapy in PD patients.     

Release of serotonin from nerve endings byl-5-hydroxytryptophan,α-methyl-m-tyramine,and elevated potassium ions

The release of [³H]5-hydroxytryptamine ([³H]5-HT) byl-5-hydroxytryptophan (L-5-HTP),-methyl-m-tyramine (-MMTA), and elevated levels of K⁺ was studied using crude synaptosomal preparations (P₂) isolated from the telencephalon of the rat and pigeon. Studies were conducted in vitro in the presence of either 2×10^–5 M tranylcypromine, which inhibited the MAO activity of both the extrasynaptosomal mitochondria and the mitochondria contained within the nerve endings (intrasynaptosomal mitochondria), or 2×10^–5 M nialamide, which inhibited the MAO activity of the extrasynaptosomal mitochondria under the experimental conditions used. In the P₂ fraction isolated from the rat, either 55 mM K⁺, 0.10 mMl-5-HTP, or 0.03 mM-MMTA significantly increased the release of [³H]5-HT above control levels, regardless of which MAO inhibitor was present in the medium. In the presence of tranylcypromine, this increased release by 55 mM K⁺ or 0.10 mMl-5-HTP was partially suppressed if Ca²⁺ was omitted from the medium. In the presence of nialamide, the release by 55 mM K⁺ was completely prevented if Ca²⁺ was omitted; the release byl-5-HTP was only partially affected. The release of [³H]5-HT by-MMTA did not appear to be markedly affected by removal of Ca²⁺, regardless of which MAO inhibitor was present. Very similar data were obtained in the presence of nialamide using the P₂ fraction isolated from the telencephalon of the pigeon, with the exception that 0.10 mMl-5-HTP caused an increase in the release of [³H]5-HIAA (which was not calcium-dependent) instead of [³H]5-HT. The data are discussed in     

Production of GDP-l-fucose, l-fucose donor for fucosyloligosaccharide synthesis, in recombinant Escherichia coli

A recombinant Escherichia coli strain was developed to produce guanosine 5′-diphosphate (GDP)-l-fucose, donor of l-fucose, which is an essential substrate for the synthesis of fucosyloligosaccharides. GDP-d-mannose-4, 6-dehydratase (GMD) and GDP-4-keto-6-deoxymannose 3, 5-epimerase 4-reductase (WcaG), the two crucial enzymes for the de novo GDP-l-fucose biosynthesis, were overexpressed in recombinant E. coli by constructing inducible overexpression vectors. Optimum expression conditions for GMD and WcaG in recombinant E. coli BL21(DE3) were 25°C and 0.1 mM isopropyl-β-d-thioglucopyranoside. Maximum GDP-l-fucose concentration of 38.9 ± 0.6 mg l⁻¹ was obtained in a glucose-limited fed-batch cultivation, and it was enhanced further by co-expression of NADPH-regenerating glucose-6-phosphate dehydrogenase encoded by the zwf gene to achieve 55.2 ± 0.5 mg l⁻¹ GDP-l-fucose under the same cultivation condition.     

Production of tyrosine from sucrose or glucose achieved by rapid genetic changes to phenylalanine-producing <Emphasis Type="Italic">Escherichia coli</Emphasis> strains

Escherichia coli K12 strains producing l-phenylalanine were converted to l-tyrosine-producing strains using a novel genetic method for gene replacement. We deleted a region of the E. coli K12 chromosome including the pheA gene encoding chorismate mutase/prephenate dehydratase, its leader peptide (pheL), and its promoter using a new polymerase chain reaction-based method that does not leave a chromosomal scar. For high level expression of tyrA, encoding chorismate mutase/prephenate dehydrogenase, its native promoter was replaced with the strong trc promoter. The linked ΔpheLA and Ptrc-tyrA::Kan^R genetic modifications were moved into l-phenylalanine producing strains by generalized transduction to convert l-phenylalanine-producing strains to l-tyrosine-producing strains. Moreover, introduction of a plasmid carrying genes responsible for sucrose degradation into these strains enabled l-tyrosine-production from sucrose.     

l-Ribulose production by an Escherichia coli harboring l-arabinose isomerase from Bacillus licheniformis

Recombinant Escherichia coli harboring the l-arabinose isomerase (BLAI) from Bacillus licheniformis was used as a biocatalyst to produce l-ribulose in the presence of borate. Effects of substrate concentration, the borate to l-arabinose ratio, pH, and temperature on the conversion of l-arabinose to l-ribulose were investigated. l-Ribulose production was efficient when pH was higher than 9 and temperature was higher than 50 °C. Borate addition to the reaction mixture was essential for high conversion of l-arabinose to l-ribulose as it resulted in an equilibrium shift in favor of the product. Under the optimal conditions determined by response surface methodology, the E. coli harboring BLAI produced 375 g l⁻¹ L-ribulose from 500 g l⁻¹ l-arabinose at a reaction time of 60 min, corresponding to a conversion yield of 75% and productivity of 375 g l⁻¹ h⁻¹. When the resting recombinant E. coli cells were recycled, 85% of the yield was obtained even after seven cycles of reuse. The productivity and final concentration of l-ribulose obtained in the present study were the highest yet reported.     

Enhanced production of GDP-<Emphasis Type="SmallCaps">l</Emphasis>-fucose by overexpression of NADPH regenerator in recombinant <Emphasis Type="Italic">Escherichia coli</Emphasis>

Biosynthesis of guanosine 5′-diphosphate-l-fucose (GDP-l-fucose) requires NADPH as a reducing cofactor. In this study, endogenous NADPH regenerating enzymes such as glucose-6-phosphate dehydrogenase (G6PDH), isocitrate dehydrogenase (Icd), and NADP⁺-dependent malate dehydrogenase (MaeB) were overexpressed to increase GDP-l-fucose production in recombinant Escherichia coli. The effects of overexpression of each NADPH regenerating enzyme on GDP-l-fucose production were investigated in a series of batch and fed-batch fermentations. Batch fermentations showed that overexpression of G6PDH was the most effective for GDP-l-fucose production. However, GDP-l-fucose production was not enhanced by overexpression of G6PDH in the glucose-limited fed-batch fermentation. Hence, a glucose feeding strategy was optimized to enhance GDP-l-fucose production. Fed-batch fermentation with a pH-stat feeding mode for sufficient supply of glucose significantly enhanced GDP-l-fucose production compared with glucose-limited fed-batch fermentation. A maximum GDP-l-fucose concentration of 235.2 ± 3.3 mg l⁻¹, corresponding to a 21% enhancement in the GDP-l-fucose production compared with the control strain overexpressing GDP-l-fucose biosynthetic enzymes only, was achieved in the pH-stat fed-batch fermentation of the recombinant E. coli overexpressing G6PDH. It was concluded that sufficient glucose supply and efficient NADPH regeneration are crucial for NADPH-dependent GDP-l-fucose production in recombinant E. coli.     

Biotechnological production of <Emphasis Type="SmallCaps">l</Emphasis>-ribose from <Emphasis Type="SmallCaps">l</Emphasis>-arabinose

l-Ribose is a rare and expensive sugar that can be used as a precursor for the production of l-nucleoside analogues, which are used as antiviral drugs. In this work, we describe a novel way of producing l-ribose from the readily available raw material l-arabinose. This was achieved by introducing l-ribose isomerase activity into l-ribulokinase-deficient Escherichia coli UP1110 and Lactobacillus plantarum BPT197 strains. The process for l-ribose production by resting cells was investigated. The initial l-ribose production rates at 39°C and pH 8 were 0.46 ± 0.01 g g⁻¹ h⁻¹ (1.84 ± 0.03 g l⁻¹ h⁻¹) and 0.27 ± 0.01 g g⁻¹ h⁻¹ (1.91 ± 0.1 g l⁻¹ h⁻¹) for E. coli and for L. plantarum, respectively. Conversions were around 20% at their highest in the experiments. Also partially purified protein precipitates having both l-arabinose isomerase and l-ribose isomerase activity were successfully used for converting l-arabinose to l-ribose.     

Pyridoxal 5′-phosphate levels in brain after treatments which impair cerebral glucose metabolism

Pyridoxal 5-phosphate (PLP) concentrations were measured in brains of rats to determine whether a deficiency of this coenzyme was a common feature in hepatic coma, ethanol intoxication, and in animals treated withl-dopa or with 5-hydroxytryptophan (5-HTP) alone or with inhibitors of MAO or ofl-aromatic amino acid decarboxylase. These treatments have been shown previously to be associated with reduced conversion of glucose to amino acids in brain. Cerebral PLP concentrations were reduced after some of these treatments, notably injection of ethanol, orl-dopa alone or with -phenylisopropylhydrazine, an inhibitor of MAO, or of 5-HTP together withN-[-(chlorophenoxy)ethyl]cyclopropylamine hydrochloride, Lilly 51641, another MAO inhibitor. However, in other circumstances where inhibition of conversion of glucose to amino acids has been shown {treatment with 5-HTP, or with Lilly 51641 or with [N-(d,l-seryl)-N-2,3,4-trihydroxybenzyl]hydrazine, an inhibitor ofl-aromatic amino acid decarboxylase, together withl-dopa or with 5-HTP}, PLP levels in brain were unchanged, or were increased (in hepatectomized rats).     

Modulation of guanosine nucleotides biosynthetic pathways enhanced GDP-<Emphasis Type="SmallCaps">l</Emphasis>-fucose production in recombinant <Emphasis Type="Italic">Escherichia coli</Emphasis>

Guanosine 5′-triphosphate (GTP) is the key substrate for biosynthesis of guanosine 5′-diphosphate (GDP)-l-fucose. In this study, improvement of GDP-l-fucose production was attempted by manipulating the biosynthetic pathway for guanosine nucleotides in recombinant Escherichia coli-producing GDP-l-fucose. The effects of overexpression of inosine 5′-monophosphate (IMP) dehydrogenase, guanosine 5′-monophosphate (GMP) synthetase (GuaB and GuaA), GMP reductase (GuaC) and guanosine–inosine kinase (Gsk) on GDP-l-fucose production were investigated in a series of fed-batch fermentations. Among the enzymes tested, overexpression of Gsk led to a significant improvement of GDP-l-fucose production. Maximum GDP-l-fucose concentration of 305.5 ± 5.3 mg l⁻¹ was obtained in the pH-stat fed-batch fermentation of recombinant E. coli-overexpressing Gsk, which corresponds to a 58% enhancement in the GDP-l-fucose production compared with the control strain overexpressing GDP-l-fucose biosynthetic enzymes. Such an enhancement of GDP-l-fucose production could be due to the increase in the intracellular level of GMP.     

Amplification of the tryptophan operon gene in Escherichia coli chromosome to increase l-tryptophan biosynthesis

Classical mutagenesis could desensitize the feedback inhibition of l-tryptophan (l-Trp) biosynthesis. Among the mutants, a5-fluorotryptophan-resistant strain, Escherichia coli EMS4-C25 produced 3 g/l of l-Trp within 18 h. The feedback-resistant l-Trp operon gene (trp) prepared from E. coli EMS4-C25 was inserted into pUC19 and pHSG576 to generate pTC701 and pTC576, respectively. When pHSG576 and pTC701 were introduced into E. coli EMS4-C25, chromosomal integration occured through homologous recombination. By using Souther hybridization, we demostrated that the integrated plasmids existed as multicopies. The strains with integrated foreign trp operon gene had higher activities of anthranilate synthase and Trp synthase than those found for the host strain and produced 9.2 g/l of l-Trp with 13% conversion yield from d-glucose. The integration and implification of the trp-operon-beraing plasmid avoided the plasmid instability and increased l-TRp production. Correspondence to: E.-C. Chan     

Fluorescence microscopy of the 5-HTP turnover in the exocrine pancreas of mice and rats

Summary The turnover ofl-5-HTP,d-5-HTP and 5-HT in the exocrine pancreas have been studied by means of the fluorescence method ofFalck andHillarp. l- andd-5-HTP are easily taken up by the acinar cells, whereas 5-HT seems to pass into the cells only to a minor extent. After the administration ofl-5-HTP (and in some cases after 5-HT administration), specific fluorescence is seen in the form of apically located granules (probably identical with the zymogen granules) for a short period, which is prolonged, if the animals are pretreated with a MAO inhibitor. Decarboxylase inhibition prevents the appearance of these fluorescent granules. Administration ofd-5-HTP does not give rise to this granular fluorescence but to a diffuse fluorescence throughout the cells. Thus, there are reasons to assume that the granular fluorescence derives from 5-HT. The results obtained in this work correspond well with those from a similar study withl-DOPA and some of its analogues.abbreviations DOPA 3,4-dihydroxyphenylalanine - DA dopamine - NA noradrenaline - A adrenaline - 5-HTP 5-hydroxytryptophan - 5-HT 5-hydroxytryptamine - MAO monoamine oxidase This work was supported by grants from the Swedish Medical Research Council (B68-12X-712-03B and B68-14X-56-04B), the United States Public Health Service (06701-02) and the Faculty of Medicine, University of Lund, Lund, Sweden.     

The outer membrane TolC is involved in cysteine tolerance and overproduction in <Emphasis Type="Italic">Escherichia coli</Emphasis>

l-Cysteine is an important amino acid in terms of its industrial applications. We previously found marked production of l-cysteine directly from glucose in recombinant Escherichia coli cells by the combination of enhancing biosynthetic activity and weakening the degradation pathway. Further improvements in l-cysteine production are expected to use the amino acid efflux system. Here, we identified a novel gene involved in l-cysteine export using a systematic and comprehensive collection of gene-disrupted E. coli K-12 mutants (the Keio collection). Among the 3,985 nonessential gene mutants, tolC-disrupted cells showed hypersensitivity to l-cysteine relative to wild-type cells. Gene expression analysis revealed that the tolC gene encoding the outer membrane channel is essential for l-cysteine tolerance in E. coli cells. However, l-cysteine tolerance is not mediated by TolC-dependent drug efflux systems such as AcrA and AcrB. It also appears that other outer membrane porins including OmpA and OmpF do not participate in TolC-dependent l-cysteine tolerance. When a low-copy-number plasmid carrying the tolC gene was introduced into E. coli cells with enhanced biosynthesis, weakened degradation, and improved export of l-cysteine, the transformants exhibited more l-cysteine tolerance and production than cells carrying the vector only. We concluded that TolC plays an important role in l-cysteine tolerance probably due to its export ability and that TolC overexpression is effective for l-cysteine production in E. coli. Natthawut Wiriyathanawudhiwong and Iwao Ohtsu contributed equally to this work.     

Metabolic engineering of <Emphasis Type="Italic">Escherichia coli</Emphasis> for improving <Emphasis Type="SmallCaps">l</Emphasis>-3,4-dihydroxyphenylalanine (<Emphasis Type="SmallCaps">l</Emphasis>-DOPA) synthesis from glucose

l-3,4-dihydroxyphenylalanine (l-DOPA) is an aromatic compound employed for the treatment of Parkinson's disease. Metabolic engineering was applied to generate Escherichia coli strains for the production of l-DOPA from glucose by modifying the phosphoenolpyruvate:sugar phosphotransferase system (PTS) and aromatic biosynthetic pathways. Carbon flow was directed to the biosynthesis of l-tyrosine (l-Tyr), an l-DOPA precursor, by transforming strains with compatible plasmids carrying genes encoding a feedback-inhibition resistant version of 3-deoxy-d-arabino-heptulosonate-7-phosphate synthase, transketolase, the chorismate mutase domain from chorismate mutase-prephenate dehydratase from E. coli and cyclohexadienyl dehydrogenase from Zymomonas mobilis. The effects on l-Tyr production of PTS inactivation (PTS⁻ gluc⁺ phenotype), as well as inactivation of the regulatory protein TyrR, were evaluated. PTS inactivation caused a threefold increase in the specific rate of l-Tyr production (q _l-Tyr), whereas inactivation of TyrR caused 1.7- and 1.9-fold increases in q _l-Tyr in the PTS⁺ and the PTS⁻ gluc⁺ strains, respectively. An 8.6-fold increase in l-Tyr yield from glucose was observed in the PTS⁻ gluc⁺ tyrR ⁻ strain. Expression of hpaBC genes encoding the enzyme 4-hydroxyphenylacetate 3-hydroxylase from E. coli W in the strains modified for l-Tyr production caused the synthesis of l-DOPA. One of such strains, having the PTS⁻ gluc⁺ tyrR ⁻ phenotype, displayed the best production parameters in minimal medium, with a specific rate of l-DOPA production of 13.6 mg/g/h, l-DOPA yield from glucose of 51.7 mg/g and a final l-DOPA titer of 320 mg/l. In a batch fermentor culture in rich medium this strain produced 1.51 g/l of l-DOPA in 50 h.     

Production of diprenylated indole derivatives by tandem incubation of two recombinant dimethylallyltryptophan synthases

Two dimethylallyltryptophan synthases, FgaPT2 and 7-DMATS, which catalysed the prenylation of l-tryptophan at positions C4 and C7, respectively, have been recently identified in Aspergillus fumigatus and proven biochemically. These enzymes were successfully used for the production of monoprenylated indole derivatives. In this study, we showed that C4,C7-diprenylated indole derivatives, e.g. 4,7-di-(dimethylallyl)-l-tryptophan, 4,7-di-(dimethylallyl)-l-abrine and 4,7-di-(dimethylallyl)-11-methyltryptophan, could be conveniently produced by tandem incubation of both enzymes. The structures of the isolated enzymatic products were elucidated by NMR and MS analyses. High conversion yields of up to 93% were achieved by an incubation sequence of FgaPT2 followed by 7-DMATS. The results reported in this study demonstrated the potential of secondary metabolite enzymes as promising tools for the production of designed compounds.     

Hyper-production of l-trytophan via fermentation with crystallization

A stable and fast l-tryptophan producer, AGX1757, was isolated from Escherichia coli W3110 trpAE1 trpR tnaA, which carried pSC101-trpI15·14. Cells of AGX1757 did not lose the composite plasmid during fermentation. Whenever a fed-batch culture of AGX1757 attained an l-tryptophan concentration of about 30 g/l, indole began to appear in the broth. The emergence of indole was caused by inhibition of tryptophan synthase due to accumulated l-tryptophan. Hence, the production rate of l-tryptophan sharply decreased. A higher solubility of l-tryptophan in the supernatant of culture broth (about 32 g/l) than that in the initial medium (about 22 g/l) was attributed to some unknown interaction between l-tryptophan and certain macromolecular material(s) coming from the bacterial cells. An addition of non-ionic detergents into the supernatant was effective for decreasing the solubility of l-tryptophan, hence causing crystallization of l-tryptophan. Pluronic L-61 was supplied from outside to an extent of 0.5% in terms of wt% concentration at around 45 h of fermentation when the l-tryptophan accumulated reached about 25 g/l. This addition actually caused crystallization of l-tryptophan and, as a result, the inhibitory effect of tryptophan synthase by l-tryptophan accumulated in the broth could be alleviated. Thus far, further fermentation became possible. l-Tryptophan of more than 50 g/l was finally produced by feeding solutions of both glucose and anthranilic acid. Correspondence to: H. Tsunekawa     

A novel N-carbamoyl-L-amino acid amidohydrolase of Pseudomonas sp. strain ON-4a: purification and characterization of N-carbamoyl-L-cysteine amidohydrolase expressed in Escherichia coli

N-carbamoyl-l-cysteine amidohydrolase (NCC amidohydrolase) was purified and characterized from the crude extract of Escherichia coli in which the gene for NCC amidohydrolase of Pseudomonas sp. strain ON-4a was expressed. The enzyme was purified 58-fold to homogeneity with a yield of 16.1% by three steps of column chromatography. The results of gel filtration on Sephacryl S-300 and SDS-polyacrylamide gel electrophoresis suggested that the enzyme was a tetramer protein of identical 45-kDa subunits. The optimum pH and temperature of the enzyme activity were pH 9.0 and 50°C, respectively. The enzyme required Mn²⁺ ion for activity expression and was inhibited by EDTA, Hg²⁺ and sulfhydryl reagents. The enzyme was strictly specific for the l-form of N-carbamoyl-amino acids as substrates and exhibited high activity in the hydrolysis of N-carbamoyl-l-cysteine as substrate. These results suggested that the NCC amidohydrolase is a novel l-carbamoylase, different from the known l-carbamoylases.