Anthranilate N-methyltransferase, a branch-point enzyme of acridone biosynthesis

Acridone alkaloids formed by acridone synthase in Ruta graveolens L. are composed of N-methylanthraniloyl CoA and malonyl CoAs. A 1095 bp cDNA from elicited Ruta cells was expressed in Escherichia coli, and shown to encode S-adenosyl-l-methionine-dependent anthranilate N-methyltransferase. SDS-PAGE of the purified enzyme revealed a mass of 40 +/- 2 kDa, corresponding to 40 059 Da for the translated polypeptide, whereas the catalytic activity was assigned to a homodimer. Alignments revealed closest relationships to catechol or caffeate O-methyltransferases at 56% and 55% identity (73% similarity), respectively, with little similarity ( approximately 20%) to N-methyltransferases for purines, putrescine, glycine, or nicotinic acid substrates. Notably, a single Asn residue replacing Glu that is conserved in caffeate O-methyltransferases determines the catalytic efficiency. The recombinant enzyme showed narrow specificity for anthranilate, and did not methylate catechol, salicylate, caffeate, or 3- and 4-aminobenzoate. Moreover, anthraniloyl CoA was not accepted. As Ruta graveolens acridone synthase also does not accept anthraniloyl CoA as a starter substrate, the anthranilate N-methylation prior to CoA activation is a key step in acridone alkaloid formation, channelling anthranilate from primary into secondary branch pathways, and holds promise for biotechnological applications. RT-PCR amplifications and Western blotting revealed expression of the N-methyltransferase in all organs of Ruta plants, particularly in the flower and root, mainly associated with vascular tissues. This expression correlated with the pattern reported previously for expression of acridone synthase and acridone alkaloid accumulation.     

Specificities of functionally expressed chalcone and acridone synthases from Ruta graveolens.

The common rue, Ruta graveolens L., expresses two types of closely related polyketide synthases that condense three malonyl-CoAs with N-methylanthraniloyl-CoA or 4-coumaroyl-CoA to produce acridone alkaloids and flavonoid pigments, respectively. Two acridone synthase cDNAs (ACS1 and ACS2) have been cloned from Ruta cell cultures, and we report now the cloning of three chalcone synthase cDNAs (CHS1 to CHS3) from immature Ruta flowers. The coding regions of these three cDNAs differ only marginally, and the translated polypeptides show about 90% identity with the CHSs from Citrus sinensis but less than 75% with the Ruta endogeneous ACSs. CHS1 was functionally expressed in Eschericha coli and its substrate specificity compared with those of the recombinant ACS1 and ACS2. 4-Coumaroyl-CoA was the preferred starter substrate for CHS1, but cinnamoyl-CoA and caffeoyl-CoA were also turned over at significant rates. However, N-methylanthraniloyl-CoA was not accepted. In contrast, highly active preparations of recombinant ACS1 or ACS2 showed low, albeit significant, CHS side activities with 4-coumaroyl-CoA, which on average reached 16% (ACS1) and 12% (ACS2) of the maximal activity determined with N-methylanthraniloyl-CoA as the starter substrate, while the conversion of cinnamoyl-CoA was negligible with both ACSs. The condensation mechanism of the acridone ring system differs from that of chalcone/flavanone formation. Nevertheless, our results suggest that very minor changes in the sequences of Ruta CHS genes are sufficient to also accommodate the formation of acridone alkaloids, which will be investigated further by site-directed mutagenesis.     

Transformation of acridone synthase to chalcone synthase.

Acridone synthase (ACS) and chalcone synthase (CHS) catalyse the pivotal reactions in the formation of acridone alkaloids or flavonoids. While acridone alkaloids are confined almost exclusively to the Rutaceae, flavonoids occur abundantly in all seed-bearing plants. ACSs and CHSs had been cloned from Ruta graveolens and shown to be closely related polyketide synthases which use N-methylanthraniloyl-CoA and 4-coumaroyl-CoA, respectively, as the starter substrate to produce the acridone or naringenin chalcone. As proposed for the related 2-pyrone synthase from Gerbera, the differential substrate specificities of ACS and CHS might be attributed to the relative volume of the active site cavities. The primary sequences as well as the immunological cross reactivities and molecular modeling studies suggested an almost identical spatial structure for ACS and CHS. Based on the Ruta ACS2 model the residues Ser132, Ala133 and Val265 were assumed to play a critical role in substrate specificity. Exchange of a single amino acid (Val265Phe) reduced the catalytic activity by about 75% but grossly shifted the specificity towards CHS activity, and site-directed mutagenesis replacing all three residues by the corresponding amino acids present in CHS (Ser132Thr, Ala133Ser and Val265Phe) fully transformed the enzyme to a functional CHS with comparatively marginal ACS activity. The results suggested that ACS divergently has evolved from CHS by very few amino acid exchanges, and it remains to be established why this route of functional diversity has developed in the Rutaceae only.     

An acridone-producing novel multifunctional type III polyketide synthase from Huperzia serrata

A cDNA encoding a novel plant type III polyketide synthase was cloned and sequenced from the Chinese club moss Huperzia serrata (Huperziaceae). The deduced amino acid sequence of Hu. serrata polyketide synthase 1 showed 44-66% identity to those of other chalcone synthase superfamily enzymes of plant origin. Further, phylogenetic tree analysis revealed that Hu. serrata polyketide synthase 1 groups with other nonchalcone-producing type III polyketide synthases. Indeed, a recombinant enzyme expressed in Escherichia coli showed unusually versatile catalytic potential to produce various aromatic tetraketides, including chalcones, benzophenones, phloroglucinols, and acridones. In particular, it is remarkable that the enzyme accepted bulky starter substrates such as 4-methoxycinnamoyl-CoA and N-methylanthraniloyl-CoA, and carried out three condensations with malonyl-CoA to produce 4-methoxy-2',4',6'-trihydroxychalcone and 1,3-dihydroxy-N-methylacridone, respectively. In contrast, regular chalcone synthase does not accept these bulky substrates, suggesting that the enzyme has a larger starter substrate-binding pocket at the active site. Although acridone alkaloids have not been isolated from Hu. serrata, this is the first demonstration of the enzymatic production of acridone by a type III polyketide synthase from a non-Rutaceae plant. Interestingly, Hu. serrata polyketide synthase 1 lacks most of the consensus active site sequences with acridone synthase from Ruta graveolens (Rutaceae).     

Starter substrate specificities of wild-type and mutant polyketide synthases from Rutaceae

Chalcone synthases (CHSs) and acridone synthases (ACSs) belong to the superfamily of type III polyketide synthases (PKSs) and condense the starter substrate 4-coumaroyl-CoA or N-methylanthraniloyl-CoA with three malonyl-CoAs to produce flavonoids and acridone alkaloids, respectively. ACSs which have been cloned exclusively from Ruta graveolens share about 75-85% polypeptide sequence homology with CHSs from other plant families, while 90% similarity was observed with CHSs from Rutaceae, i.e., R. graveolens, Citrus sinensis and Dictamnus albus. CHSs cloned from many plants do not accept N-methylanthraniloyl-CoA as a starter substrate, whereas ACSs were shown to possess some side activity with 4-coumaroyl-CoA. The transformation of an ACS to a functional CHS with 10% residual ACS activity was accomplished previously by substitution of three amino acids through the corresponding residues from Ruta-CHS1 (Ser132Thr, Ala133Ser and Val265Phe). Therefore, the reverse triple mutation of Ruta-CHS1 (mutant R2) was generated, which affected only insignificantly the CHS activity and did not confer ACS activity. However, competitive inhibition of CHS activity by N-methylanthraniloyl-CoA was observed for the mutant in contrast to wild-type CHSs. Homology modeling of ACS2 with docking of 1,3-dihydroxy-N-methylacridone suggested that the starter substrates for CHS or ACS reaction are placed in different topographies in the active site pocket. Additional site specific substitutions (Asp205Pro/Thr206Asp/His207Ala or Arg60Thr and Val100Ala/Gly218Ala, respectively) diminished the CHS activity to 75-50% of the wild-type CHS1 without promoting ACS activity. The results suggest that conformational changes in the periphery beyond the active site cavity volumes determine the product formation by ACSs vs. CHSs in R. graveolens. It is likely that ACS has evolved from CHS, but the sole enlargement of the active site pocket as in CHS1 mutant R2 is insufficient to explain this process.     

Cloning and Expression of Mycobacterium tuberculosis and Mycobacterium leprae Dihydropteroate Synthase in Escherichia coli

The genes for dihydropteroate synthase of Mycobacterium tuberculosis and Mycobacterium leprae were isolated by hybridization with probes amplified from the genomic DNA libraries. DNA sequencing revealed an open reading frame of 840 bp encoding a protein of 280 amino acids for M. tuberculosis dihydropteroate synthase and an open reading frame of 852 bp encoding a protein of 284 amino acids for M. leprae dihydropteroate synthase. The dihydropteroate synthases were expressed under control of the T5 promoter in a dihydropteroate synthase-deficient strain of Escherichia coli. Using three chromatography steps, we purified both M. tuberculosis and M. leprae dihydropteroate synthases to >98% homogeneity. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis revealed molecular masses of 29 kDa for M. tuberculosis dihydropteroate synthase and 30 kDa for M. leprae dihydropteroate synthase. Gel filtration of both enzymes showed a molecular mass of ca. 60 kDa, indicating that the native enzymes exist as dimers of two identical subunits. Steady-state kinetic parameters for dihydropteroate synthases from both M. tuberculosis and M. leprae were determined. Representative sulfonamides and dapsone were potent inhibitors of the mycobacterial dihydropteroate synthases, but the antimycobacterial agent p-aminosalicylate, a putative dihydropteroate synthase inhibitor, was a poor inhibitor of the enzymes.     

A Streptomyces collinus thiolase with novel acetyl-CoA:acyl carrier protein transacylase activity

Acetyl-CoA:acyl carrier protein (ACP) transacylase (ACT) activity has been demonstrated for the 3-ketoacyl-ACP synthase III (KASIII) which initiates fatty acid biosynthesis in the type II dissociable fatty acid synthases of plants and bacteria. Several lines of evidence have indicated the possibility of ACT activity being associated with proteins other than KASIII. Using a crude extract of Streptomyces collinus, we have resolved from KASIII an additional protein with ACT activity and subsequently purified it 85-fold in five chromatographic steps. The 45 kDa protein was shown by gel filtration to have a molecular mass of 185 +/- 35 kDa, consistent with a homotetrameric structure for the native enzyme. The corresponding gene (fadA) was cloned and sequenced and shown to encode a protein with amino acid sequence homology to type II thiolases. The fadA was expressed in Escherichia coli, and the resulting recombinant FadA enzyme purified by metal chelate chromatography was shown to have both ACT and thiolase activities. Kinetic studies revealed that in an ACT assay FadA had a substrate specificity for a two-carbon acetyl-CoA substrate (K(m) 8.7 +/- 1.4 microM) but was able to use ACPs from both type II fatty acid and polyketide synthases (Streptomyces glaucescens FabC ACP, K(m) 10.7 +/- 1.4 microM; E. coli FabC ACP, K(m) 8.8 +/- 2 microM; FrenN ACP, K(m) 44 +/- 12 microM). In the thiolase assay kinetic analyses revealed similar K(m) values for binding of substrates acetoacetyl-CoA (K(m) 9.8 +/- 0.8 microM) and CoA (K(m) 10.9 +/- 1.8 microM). A Cys92Ser mutant of FadA possessed virtually unchanged K(m) values for acetoacetyl-CoA and CoA but had a greater than 99% decrease in k(cat) for the thiolase activity. No detectable ACT activity was observed for the Cys92Ser mutant, demonstrating that both activities are associated with FadA and likely involve formation of the same covalent acetyl-S-Cys enzyme intermediate. An ACT activity with ACP has not previously been observed for thiolases and in the case of the S. collinus FadA is significantly lower (k(cat) 3 min(-1)) than the thiolase activity of FadA (k(cat) 2170 min(-1)). The ACT activity of FadA is comparable to the KAS activity and significantly higher than the ACT activity, reported for a streptomycete KASIII.     

Purification and characterization of photosystem I and photosystem II core complexes from wild-type and phycocyanin-deficient strains of the cyanobacterium Synechocystis PCC 6803

Highly photoactive Photosystem I (PS I) and Photosystem II (PS II) core complexes have been isolated from the cyanobacterium Synechocystis Pasteur Culture Collection (PCC) 6803 and a phycocyanin-deficient mutant, enriched in PS II. Cell breakage using glass beads was followed by sucrose density gradient centrifugation and two high-performance liquid chromatography steps involving anion-exchange and hydroxyapatite. The PS I core complex has an apparent molecular mass of 300 +/- 20 kDa (including a detergent shell of about 50 kDa) and contains subunits of approximately 60, approximately 60, 18.5, 18.5, 16, 15, 10.5, 9.5, and 6.5 kDa as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and immunoblots; its antenna size is 75 +/- 5 chlorophyll/P-700. The PS II core complex has an apparent molecular mass of 310 +/- 20 kDa (including the detergent shell); subunits of 43, 37, 33, 29, and 10-11 kDa were identified by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and immunoblotting. The antenna size of the average PS II complex is 45 +/- 5 chlorophyll/primary quinone electron acceptor (QA). This preparation procedure also yields, as a byproduct, a highly purified cytochrome b6f complex. This complex contains four subunits of 38, 24, 19, and 15 kDa and b- and c-type cytochromes in a ratio of 2:1. Its apparent molecular mass of 180 +/- 20 kDa (including the detergent shell) is consistent with a monomeric complex.     

Molecular cloning and sequencing of the psaD gene encoding subunit II of photosystem I from the cyanobacterium, Synechocystis sp. PCC 6803

Photosystem I reaction center was isolated from the cyanobacterium, Synechocystis sp. PCC 6803, in a form which contains seven different polypeptide subunits. One of the subunits, with a molecular mass of about 16 kDa, was isolated, and protein sequence information was obtained for the amino terminus and several tryptic peptides. Oligonucleotide probes, corresponding to these sequences, were used to probe a genomic library, and the gene, designated psaD, encoding subunit II was cloned and sequenced. The gene encodes a polypeptide with a mass 15,644 Da, which exhibits a high degree of similarity to subunit II from tomato, as well as amino acid sequences reported from barley photosystem I. In addition to this gene, three large open reading frames were identified. Two remain unidentified, and the third is highly homologous to anthranilate synthase, component 1 from Escherichia coli and Saccharomyces cerevisiae.     

Biochemical characterization of the minimal polyketide synthase domains in the lovastatin nonaketide synthase LovB

The biosynthesis of lovastatin in Aspergillus terreus requires two megasynthases. The lovastatin nonaketide synthase, LovB, synthesizes the intermediate dihydromonacolin L using nine malonyl-coenzyme A molecules, and is a reducing, iterative type I polyketide synthase. The iterative type I polyketide synthase is mechanistically different from bacterial type I polyketide synthases and animal fatty acid synthases. We have cloned the minimal polyketide synthase domains of LovB as standalone proteins and assayed their activities and substrate specificities. The didomain proteins ketosynthase-malonyl-coenzyme A:acyl carrier protein acyltransferase (KS-MAT) and acyl carrier protein-condensation (ACP-CON) domain were expressed solubly in Escherichia coli. The monodomains MAT, ACP and CON were also obtained as soluble proteins. The MAT domain can be readily labeled by [1,2-(14)C]malonyl-coenzyme A and can transfer the acyl group to both the cognate LovB ACP and heterologous ACPs from bacterial type I and type II polyketide synthases. Using the LovB ACP-CON didomain as an acyl acceptor, LovB MAT transferred malonyl and acetyl groups with k(cat)/K(m) values of 0.62 min(-1).mum(-1) and 0.032 min(-1).mum(-1), respectively. The LovB MAT domain was able to substitute the Streptomyces coelicolor FabD in supporting product turnover in a bacterial type II minimal polyketide synthase assay. The activity of the KS domain was assayed independently using a KS-MAT (S656A) mutant in which the MAT domain was inactivated. The KS domain displayed no activity towards acetyl groups, but was able to recognize malonyl groups in the absence of cerulenin. The relevance of these finding to the priming mechanism of fungal polyketide synthase is discussed.     

Evolution of vitamin B2 biosynthesis: 6,7-dimethyl-8-ribityllumazine synthases of Brucella

The penultimate step in the biosynthesis of riboflavin (vitamin B2) involves the condensation of 3,4-dihydroxy-2-butanone 4-phosphate with 5-amino-6-ribitylamino-2,4(1H,3H)-pyrimidinedione, which is catalyzed by 6,7-dimethyl-8-ribityllumazine synthase (lumazine synthase). Pathogenic Brucella species adapted to an intracellular lifestyle have two genes involved in riboflavin synthesis, ribH1 and ribH2, which are located on different chromosomes. The ribH2 gene was shown previously to specify a lumazine synthase (type II lumazine synthase) with an unusual decameric structure and a very high Km for 3,4-dihydroxy-2-butanone 4-phosphate. Moreover, the protein was found to be an immunodominant Brucella antigen and was able to generate strong humoral as well as cellular immunity against Brucella abortus in mice. We have now cloned and expressed the ribH1 gene, which is located inside a small riboflavin operon, together with two other putative riboflavin biosynthesis genes and the nusB gene, specifying an antitermination factor. The RibH1 protein (type I lumazine synthase) is a homopentamer catalyzing the formation of 6,7-dimethyl-8-ribityllumazine at a rate of 18 nmol mg(-1) min(-1). Sequence comparison of lumazine synthases from archaea, bacteria, plants, and fungi suggests a family of proteins comprising archaeal lumazine and riboflavin synthases, type I lumazine synthases, and the eubacterial type II lumazine synthases.     

The monomeric polypeptide comprises the functional flavanone 3beta-hydroxylase from Petunia hybrida

Flavanone 3beta-hydroxylase catalyzes the Fe(II)/oxoglutarate-dependent hydroxylation of (2S)-flavanones to (2R,3R)-dihydroflavonols in the biosynthesis of flavonoids, catechins and anthocyanidins. The enzyme had been partially purified from Petunia hybrida and proposed to be active as a dimer of roughly 75 kDa in size. More recently, the Petunia 3beta-hydroxylase was cloned and shown to be encoded in a 41655 Da polypeptide. In order to characterize the molecular composition, the enzyme was expressed in a highly active state in Escherichia coli and purified to apparent homogeneity. Size exclusion chromatographies of the pure, recombinant enzyme revealed that this flavanone 3beta-hydroxylase exists in functional monomeric and oligomeric forms. Protein cross-linking experiments employing a specific homobifunctional sulfhydryl group reagent or the photochemical activation of tryptophan residues confirmed the tendency of the enzyme to aggregate to oligomeric complexes in solution. Thorough equilibrium sedimentation analyses, however, revealed a molecular mass of 39. 2+/-12 kDa for the recombinant flavanone 3beta-hydroxylase. The result implies that the monomeric polypeptide comprises the catalytically active flavanone 3beta-hydroxylase of P. hybrida, which may readily associate in vivo with other proteins.     

Evolution of vitamin B2 biosynthesis. A novel class of riboflavin synthase in Archaea

The open reading frame MJ1184 of Methanococcus jannaschii with similarity to riboflavin synthase of Methanothermobacter thermoautotrophicus was cloned into an expression vector but was poorly expressed in an Escherichia coli host strain. However, a synthetic open reading frame that was optimized for expression in E.coli directed the synthesis of abundant amounts of a protein with an apparent subunit mass of 17.5 kDa. The protein was purified to apparent homogeneity. Hydrodynamic studies indicated a relative mass of 88 kDa suggesting a homopentamer structure. The enzyme was shown to catalyze the formation of riboflavin from 6,7-dimethyl-8-ribityllumazine at a rate of 24 nmol mg(-1) min(-1) at 40 degrees C. Divalent metal ions, preferably manganese or magnesium, are required for maximum activity. In contrast to pentameric archaeal type riboflavin synthases, orthologs from plants, fungi and eubacteria are trimeric proteins characterized by an internal sequence repeat with similar folding patterns. In these organisms the reaction is achieved by binding the two substrate molecules in an antiparallel orientation. With the enzyme of M.jannaschii, 13C NMR spectroscopy with 13C-labeled 6,7-dimethyl-8-ribityllumazine samples as substrates showed that the regiochemistry of the dismutation reaction is the same as observed in eubacteria and eukaryotes, however, in a non-pseudo-c2 symmetric environment. Whereas the riboflavin synthases of M.jannaschii and M.thermoautotrophicus are devoid of similarity with those of eubacteria and eukaryotes, they have significant sequence similarity with 6,7-dimethyl-8-ribityllumazine synthases catalyzing the penultimate step of riboflavin biosynthesis. 6,7-Dimethyl-8-ribityllumazine synthase and the archaeal riboflavin synthase appear to have diverged early in the evolution of Archaea from a common ancestor. Some Archaea have eubacterial type riboflavin synthases which may have been acquired by lateral gene transfer.     

An approach for cloning biosynthetic genes of 2-deoxystreptamine-containing aminocyclitol antibiotics: isolation of a biosynthetic gene cluster of tobramycin from Streptomyces tenebrarius

Genes homologous to 2-deoxystreptamine (DOS) biosynthetic genes were isolated from aminoglycoside producers, Micromonospora and Streptomyces spp., using PCR primers based on the core sequences of 2-deoxy-scyllo-inosose (DOI) synthase and L-glutamine: scyllo-inosose aminotransferase (GIA). Identities of 40-45% were observed for DOI synthases, and 65-75% were observed for GIAs. The gene cluster of tobramycin biosynthesis was isolated from the genomic library of Streptomyces tenebrarius using DOI synthase as a probe. Sequencing of 33.9 kb revealed 24 putative open reading frames including the tobramycin biosynthetic gene cluster (13.8 kb) and a transport protein. This cluster encodes proteins homologous to 2-deoxystreptamine biosynthetic enzymes, glycosyltransferase and other aminocyclitols biosynthetic enzymes. Sequence analysis revealed the evolution of DOI synthases from 3-dehydroquinate (DHQ) synthases in actinomycetes. DOI synthases and GIA are therefore useful for cloning biosynthetic genes of DOS-containing aminocyclitol antibiotics or for screening such metabolites producers.     

Cloning and analysis of a cDNA encoding farnesyl diphosphate synthase from Artemisia annua

A cDNA encoding farnesyl diphosphate (FPP) synthase (FPPS) has been cloned from a cDNA library of Artemisia annua. The sequence analysis showed that the cDNA encoded a protein of 343 amino acid (aa) residues with a calculated molecular weight of 39 420 kDa. The deduced aa sequence of the cDNA was highly similar to FPPS from other plants, yeast and mammals, and contained the two conserved domains found in polyprenyl synthases including FPPS, geranylgeranyl diphosphate synthases and hexaprenyl diphosphate synthases. The expression of the cDNA in Escherichia coli showed enzyme activity for FPPS in vitro.