Biochemical characterization of a novel indole prenyltransferase from Streptomyces sp. SN-593

Abstract

Genome sequencing of Streptomyces species has highlighted numerous potential genes of secondary metabolite biosynthesis. The mining of cryptic genes is important for exploring chemical diversity. Here we report the metabolite-guided genome mining and functional characterization of a cryptic gene by biochemical studies. Based on systematic purification of metabolites from Streptomyces sp. SN-593, we isolated a novel compound, 6-dimethylallylindole (DMAI)-3-carbaldehyde. Although many 6-DMAI compounds have been isolated from a variety of organisms, an enzyme catalyzing the transfer of a dimethylallyl group to the C-6 indole ring has not been reported so far. A homology search using known prenyltransferase sequences against the draft sequence of the Streptomyces sp. SN-593 genome revealed the iptA gene. The IptA protein showed 27% amino acid identity to cyanobacterial LtxC, which catalyzes the transfer of a geranyl group to (-)-indolactam V. A BLAST search against IptA revealed much-more-similar homologs at the amino acid level than LtxC, namely, SAML0654 (60%) from Streptomyces ambofaciens ATCC 23877 and SCO7467 (58%) from S. coelicolor A3(2). Phylogenetic analysis showed that IptA was distinct from bacterial aromatic prenyltransferases and fungal indole prenyltransferases. Detailed kinetic analyses of IptA showed the highest catalytic efficiency (6.13 min(-1) microM(-1)) for L-Trp in the presence of dimethylallyl pyrophosphate (DMAPP), suggesting that the enzyme is a 6-dimethylallyl-L-Trp synthase (6-DMATS). Substrate specificity analyses of IptA revealed promiscuity for indole derivatives, and its reaction products were identified as novel 6-DMAI compounds. Moreover, DeltaiptA mutants abolished the production of 6-DMAI-3-carbaldehyde as well as 6-dimethylallyl-L-Trp, suggesting that the iptA gene is involved in the production of 6-DMAI-3-carbaldehyde.

FIG. 1.

Structures of 6-DMAI-3-carbaldehyde (A) and 6-dimethylallyl-l-Trp (B) with key 2D NMR correlations. The bold lines and arrows show DQF-COSY and HMBC correlations, respectively.

FIG. 2.

Distributions of IptA homologs in Streptomyces spp. The genetic organization harboring orf6 (iptA) from Streptomyces sp. SN-593 is compared with the genetic organizations for S. ambofaciens ATCC 23877 and S. coelocolor A3(2). Numbers of amino acid residues are shown below the arrows.

FIG. 3.

Multiple sequence alignment of IptA homologs and phylogenetic analysis of microbial prenyltransferases. (A) The amino acids conserved in all four proteins are indicated by asterisks. Dashes indicate gaps introduced for optimization of the alignment. IptA is from Streptomyces sp. SN-593 (GenBank accession no. AB512764), SAML0654 is from Streptomyces ambofaciens ATCC 23877 (CAJ89640), SCO7467 is from Streptomyces coelicolor A3(2) (NP_631515), and LtxC is from a cyanobacterium (AAT12285). Bold letters show the identical amino acids among IptA, SAML0654, and SCO7467. (B) Phylogenetic tree of microbial prenyltransferases. CLUSTALW (http://clustalw.ddbj.nig.ac.jp/top-j.html) was used for alignment. An unrooted phylogenetic tree was produced using the TREE VIEW program (http://taxonomy.zoology.gla.ac.uk/). The scale bar corresponds to a genetic distance of 0.1 substitution per position. The Streptomyces prenyltransferases (GenBank accession no.) used were Fnq26 (CAL34104) from Streptomyces cinnamonensis, NphB (BAE00106) from Streptomyces sp. CL 190, and CloQ (AAN65239) from Streptomyces roseochromogenes. The fungal indole prenyltransferases (GenBank accession no.) used were DmaW (Q6X2E0) from Claviceps purpurea; FtmPT1 (AAX56314), FtmPT2 (EU622826), FgaPT1 (XP_756136), FgaPT2 (AAX08549), CdpNPT (ABR14712), and 7-DMATS (ABS89001) from Aspergillus fumigatus; TdiB (ABU51603) from Aspergillus nidulans; MaPT (ABZ80612) from Malbranchea aurantiaca; and AnaPT (EAW16181) from Neosartorya fischeri.

FIG. 4.

SDS-PAGE and HPLC chromatograms of IptA assays. (A) SDS-12.5% PAGE analysis of purified His-tag-free IptA. Lane 1, molecular mass markers; lane 2, purified enzyme (arrow). The enzyme purification was described in Materials and Methods. Protein was stained with Coomassie brilliant blue R-250. (B) IptA (60 pmol enzyme) was incubated with l-Trp (0.5 mM) and DMAPP (0.2 mM). The enzyme reaction product was analyzed by LC-MS under the conditions described in Materials and Methods. Dotted and solid lines indicate product profiles after 1- and 10-min reactions, respectively. (C) IptA (100 pmol enzyme) was incubated with l-Trp (0.5 mM) and GPP (0.1 mM) for 120 min.

FIG. 5.

HPLC chromatograms of IptA reaction product from indole-3-carbaldehyde. (A) LC-MS analysis of purified authentic 6-DMAI-3-carbaldehyde was performed as described in Materials and Methods. (B) LC-MS analysis of the enzyme reaction product. IptA (7.4 μmol enzyme) reacted with indole-3-carbaldehyde (0.1 mM) and DMAPP (0.1 mM) under the assay conditions described in Materials and Methods. (C) Authentic 6-DMAI-3-carbaldehyde and the enzyme reaction product were mixed and analyzed by LC-MS.

FIG. 6.

Effects of metal and pH on 6-DMATS activity. (A) The concentration of 8 divalent metal ions was varied from 1 to 10 mM in the standard assay solution. (B) The assay was conducted at a pH range of 6.0 to 11, using 50 mM NaH₂PO₄ (solid squares), 50 mM Tris-HCl (solid circles), and 50 mM glycine-NaOH buffer (open triangles).

FIG. 7.

Heat stability (A) and effects of temperature (B) on 6-DMATS activity.

FIG. 8.

Gene disruption of iptA and Southern blot analysis. (A) Scheme of iptA disruption and restriction map of wild-type and ΔiptA mutant strains. The bar shows the expected fragment size (bp) after SalI digestion. A 4,550-bp probe was amplified by oligonucleotide primers iptA-SalI-F and iptA-SalI-R (Table 1). (B) Southern blot analysis of the wild type (lanes 2 and 7) and of ΔiptA mutants (lanes 3 to 6) in individual isolations. Genomic DNAs digested with SalI were applied to a 0.9% agarose gel and stained by ethidium bromide. Southern blot analysis indicated the expected sizes of the DNA fragments from wild-type and ΔiptA mutant strains. The arrows indicate the expected sizes of the DNA fragments from the wild type (solid) and the ΔiptA mutants (open), respectively.

FIG. 9.

LC-MS analysis of 6-DMAI-3-carbaldehyde and 6-dimethylallyl-l-Trp from culture extracts of the wild-type strain and the ΔiptA mutant strain. The culture, extraction, and LC-MS analysis were performed as described in Materials and Methods. Authentic standards of 6-DMAI-3-carbaldehyde (A) and 6-dimethylallyl-l-Trp (D) were analyzed. The production of 6-DMAI-3-carbaldehyde from the MPLC fraction of the wild type (B) or the ΔiptA mutant strain (C) was analyzed. The production of 6-dimethylallyl-l-Trp from n-butanol extracts of the wild type (E) and the ΔiptA mutant strain (F) was analyzed. The chromatograms shown in panels A, B, and C are the extracted ion chromatograms of m/z 212 [M − H]⁻. The chromatograms shown in panels D, E, and F are the extracted ion chromatograms of m/z 273 [M + H]⁺.