APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Jan. 2010, p. 283–293 Vol. 76, No. 10099-2240/10/$12.00 doi:10.1128/AEM.00744-09Copyright © 2010, American Society for Microbiology. All Rights Reserved.

Insights into the Evolution of Macrolactam Biosynthesis throughCloning and Comparative Analysis of the Biosynthetic Gene

Cluster for a Novel Macrocyclic Lactam, ML-449�†Hanne Jørgensen,1 Kristin F. Degnes,2 Alexander Dikiy,1 Espen Fjærvik,1

Geir Klinkenberg,2 and Sergey B. Zotchev1*Department of Biotechnology, Norwegian University of Science and Technology, N-7491 Trondheim, Norway,1 and Department of

Industrial Biotechnology, SINTEF Materials and Chemistry, SINTEF, N-7034 Trondheim, Norway2

Received 1 April 2009/Accepted 20 October 2009

A new compound, designated ML-449, structurally similar to the known 20-membered macrolactam BE-14106, was isolated from a marine sediment-derived Streptomyces sp. Cloning and sequencing of the 83-kbML-449 biosynthetic gene cluster revealed its high level of similarity to the BE-14106 gene cluster. Comparisonof the respective biosynthetic pathways indicated that the difference in the compounds’ structures stems fromthe incorporation of one extra acetate unit during the synthesis of the acyl side chain. A phylogenetic analysisof the �-ketosynthase (KS) domains from polyketide synthases involved in the biosynthesis of macrolactamspointed to a common ancestry for the two clusters. Furthermore, the analysis demonstrated the formation ofa macrolactam-specific subclade for the majority of the KS domains from several macrolactam-biosyntheticgene clusters, indicating a closer relationship between macrolactam clusters than with the macrolactoneclusters included in the analysis. Some KS domains from the ML-449, BE-14106, and salinilactam geneclusters did, however, show a closer relationship with KS domains from the polyene macrolide clusters,suggesting potential acquisition rather than duplication of certain PKS genes. Comparison of the ML-449,BE-14106, vicenistatin, and salinilactam biosynthetic gene clusters indicated an evolutionary relationshipbetween them and provided new insights into the processes governing the evolution of small-ring macrolactambiosynthesis.

Numerous natural products, including clinically importantdrugs such as erythromycin and vancomycin, are polyketides(PKs), nonribosomal peptides (NRPs), or hybrid PK-NRPs.Both PKs and NRPs are synthesized by enzyme complexesthrough condensation of specialized building blocks. For PKsynthesis, malonyl- and methylmalonyl-coenzyme A (CoA)represent the most common building blocks, while nonriboso-mal peptide synthetases (NRPSs) utilize the common L and D

amino acids as well as a wide range of nonproteinogenic aminoacids and some carboxylic acids (9). The bacterial type Ipolyketide synthases (PKSs) and NRPSs have modular com-positions, where each module is responsible for the incorpo-ration of one building block unit. The minimal set of domainspresent in one module is usually the �-ketosynthase (KS),acyltransferase (AT), and acyl carrier protein (ACP) domainsfor PKSs and the condensation (C), adenylation (A), and pep-tidyl carrier protein (PCP) domains for NRPSs. However,there are many examples where all or a few modules may lacksome of these core domains (6, 20, 37). PKS modules may alsoinclude ketoreductase (KR), dehydratase (DH), and enoyl re-ductase (ER) domains, which determine the oxidation state ofthe �-carbon, while accessory NRPS domains include, e.g.,

epimerization, heterocyclization, and methyltransferase do-mains (9).

The modular nature of PKSs and NRPSs has been exploitedby scientists ever since the principle of how they operate wasfirst discovered. Although the possibilities for designing newtypes of structures by combining different types of domains andmodules seem endless, combinatorial biosynthesis in practiceappears to be more complicated than first assumed, and as oftoday, only a few modules can efficiently be combined to pro-duce novel structures (27). Interestingly, nature appears to beplaying the same type of game, as studies have pointed to theexistence of a natural version of biocombinatorics (8, 17, 34).Although phylogenetic analyses of the KS domains in the PKSenzymes point to generation of new modules within one path-way by duplication of a single ancestor module, the AT, KR,and (DH)-KR domains seem to be exchangeable through re-combinational replacements occurring both within and be-tween biosynthetic gene clusters (17). The same phenomenonappears to be true for NRPS adenylation domains (8). Fur-thermore, in addition to gene duplications, recombination, andgene loss, horizontal gene transfer may have played a centralrole in the evolution of bacterial type I PKSs (11, 16, 29).Understanding the processes that lead to the generation ofnew biosynthetic gene clusters in nature may be important forsuccessful application of the same concept in the laboratory.

Several biosynthetic gene clusters and the correspondingpathways for macrolactam biosynthesis have been described inrecent years. Leinamycin (Fig. 1) is synthesized by an AT-lessNRPS-PKS hybrid system, using D-alanine as a starter (6, 39).

* Corresponding author. Mailing address: Department of Biotech-nology, Norwegian University of Science and Technology, N-7491Trondheim, Norway. Phone: 47 73 59 86 79. Fax: 47 73 59 12 83.E-mail: sergey.zotchev@nt.ntnu.no.

† Supplemental material for this article may be found at http://aem.asm.org/.

� Published ahead of print on 23 October 2009.

283

on August 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from

Vicenistatin and salinilactam (Fig. 1) are synthesized by regu-lar type I PKSs, but amino acids are utilized as starter units (30,43). In the biosynthesis of vicenistatin, 3-methylaspartate ispresumed to represent the starter, while lysine was suggestedas the starter for the biosynthesis of salinilactam. We haverecently sequenced the biosynthetic gene cluster for the mac-rolactam BE-14106 (Fig. 1) and proposed a biosynthetic path-

way involving the synthesis of an aminoacyl starter throughcooperation between a PKS and NRPS-related enzymes (19).Some of the enzymes presumed to be involved in the biosyn-thesis of BE-14106 have putative homologs in the other mac-rolactam gene clusters, suggesting common steps in the bio-synthetic pathways. In this study, we present the cloning andsequencing of a new macrolactam biosynthetic gene cluster

FIG. 1. (A) Chemical structure of ML-449. (B) Chemical structures of BE-14106 and similar macrolactams. (C) Comparison of the BE-14106gene cluster with the ML-449 gene cluster. Arrows indicate genes in the corresponding directions of transcription. PKS genes are shown in greenand orange, genes related to synthesis of the starter unit and genes of unknown function are shown in red, genes related to the regulation/effluxmechanism are shown in blue, and the gene encoding a post-PKS modification enzyme are shown in yellow.

284 JØRGENSEN ET AL. APPL. ENVIRON. MICROBIOL.

on August 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from

with high similarity to the BE-14106 gene cluster. Throughphylogenetic analyses and comparison of the different macro-lactam gene clusters, we establish the evolutionary relationshipbetween the ML-449 and BE-14106 biosynthetic pathways andprovide new insights into the evolution of the macrolactamgene clusters in general.

MATERIALS AND METHODS

Screening for antifungal activity and isolation of strain MP39-85. Sedimentsamples were collected from different sites in the Trondheimsfjord and used forisolation of marine actinomycetes as described by Bredholt et al. (3). The Strep-tomyces isolate MP39-85, producing the compound designated ML-449, wasisolated from a sediment sample collected by use of a box corer at a 450-m depthat position 63°29�N, 10°18�E. The medium used for isolation of this strain wasIM7 (3).

The collection of actinomycete isolates obtained from sediment samples wasinvestigated for production of antifungal compounds by using a set of differentsolidified production media. For Streptomyces sp. MP39-85, growth at 25°C onmedium PM4 (3) for 16 days resulted in production of a compound with strongantifungal activity. Dimethyl sulfoxide (DMSO) extracts of Streptomyces sp.MP39-85 were used in a robotic bioassay procedure with Candida albicansCCUG3943 and Candida glabrata CCUG3942 as indicator organisms as de-scribed by Jørgensen et al. (18). The latter strain has a high level of resistanceagainst polyene antibiotics, while the C. albicans strain is sensitive to polyenes.

High-performance liquid chromatography (HPLC) fractionation and subse-quent liquid chromatography-mass spectrometry-time of flight (LC-MS-TOF)characterization of active fractions were performed as described earlier (19).Molecular masses (10-ppm window) were submitted to the online version of theDictionary of Natural Products (DNP) in order to search for previously charac-terized compounds with bioactivity.

Production and purification of ML-449. Cultivation of Streptomyces sp.MP39-85 for production of ML-449 was performed with BPS-3-1 (oatmeal [30g/liter], malt extract [5.0 g/liter], yeast extract [3.0 g/liter], MgSO4-7H2O [0.4g/liter], NaCl [1.0 g/liter], CaCO3 [5.0 g/liter], and glucose [50.0 g/liter]) supple-mented with 3.0 ml/liter trace mineral solution 1 (35). Fermentations wereperformed with Applikon 3-liter fermentors with 1.5 liters of medium. Theproduction cultures were inoculated with 3% (vol/vol) from a 0.5� Trypticasesoy broth (TSB)-glucose preculture cultivated as described previously (19). Fer-mentations were run for 6 days at 25°C with constant aeration (0.25% [vol/vol/min]) and agitation (1,000 rpm). The pH was controlled at 7.3 with 2 M NaOH.Production of ML-449 in shake flasks was performed with 0.3� BPS-3-1 sup-plemented with glucose as described previously (19). ML-449 was extracted fromfermentation broth and purified by preparative HPLC essentially as describedpreviously (19).

Qualitative and quantitative analyses of ML-449. Quantitative and qualitativeLC-MS analyses of ML-449 were performed on methanol extracts from culturepellets by using an Agilent 1100 HPLC system connected both to a diode arraydetector (DAD) and a TOF mass spectrometer. Electrospray ionization wasperformed in the negative (ESI�) mode as described previously (4, 19). Quan-titative analysis of ML-449 in fermentation broth was performed using BE-14106purified by preparative HPLC as a standard, assuming that the extinction coef-ficient of the two compounds is the same.

Determination of antimicrobial activity. Antimicrobial activity of purified ML-449 was determined in a robotic bioassay procedure with the yeast strains C.albicans CCUG3943 and C. glabrata CCUG3942 and the bacterial strain Micro-coccus luteus ATCC 9341. Serial dilutions (6 parallel wells) of the purifiedcompound were prepared in a 384-well plate. Diluted samples were added toplates containing medium AM19B (19) and inoculated with the yeast strainsspecified above for assay of antifungal activity and to plates with medium AM1(6 g/liter peptone [Oxoid], 4 g/liter tryptone [Difco], 3 g/liter yeast extract [Ox-oid], 1.5 g/liter beef extract [Difco], 1 g/liter glucose [BDH], distilled water)inoculated with M. luteus for the assay of antibacterial activity. Growth in thepresence of ML-449 at different concentrations was compared to growth inreference wells where only a solvent (DMSO) was added, and the concentrationresulting in 50% or more reduction in growth was determined. The assay con-ditions were otherwise as described by Jørgensen et al. (18).

Structure determination by NMR. Samples for nuclear magnetic resonance(NMR) spectroscopy were prepared by dissolving the HPLC-purified and freeze-dried ML-449 in deuterated d6-DMSO to give a final concentration of 1 mM.The NMR sample conditions were maintained the same as those previouslyreported for NMR assignment of a similar macrolactam, BE-14106 (22). All

NMR experiments were recorded at 298 K on a Bruker Avance 600-MHzspectrometer equipped with a 5-mm z-gradient TXI (H/C/N) cryogenic probe.Proton and carbon chemical shifts were referenced to the tetramethylsilyl (TMS)signal. To investigate the chemical structure of the investigated compound, bothone-dimensional 1H and two-dimensional correlation spectroscopy (COSY), 1H-13C heteronuclear single quantum coherence (HSQC), and 1H-13C hetero-nuclear multiple bond correlation (HMBC) spectra were recorded.

Strains and media used for genetic work. Streptomyces sp. MP39-85 strainswere grown on ISP2 agar medium (Difco), soy flour mannitol (SFM), mediumand mineral agar Gause 1 (soluble starch [20 g/liter], K2HPO4 [0.5 g/liter],MgSO4 [0.5 g/liter], KNO3 [1.0 g/liter], NaCl [0.5 g/liter], FeSO4 [0.01 g/liter],agar [20.0 g/liter], tap water) and in tryptone soy broth (Oxoid). Escherichia colistrains were grown in Luria-Bertani (LB) broth or on LB agar. DH5� was usedfor general cloning. EZ cells were used for vectors with blue-white selection.XL-Blue MR was used for construction of the genomic library. ET12567(pUZ8002) was used for intergeneric transfer of pSOK201-based constructs toStreptomyces sp. MP39-85. Antibiotics were supplemented to growth medium atthe following concentrations: for ampicillin, 100 or 150 �g/ml; for apramycin, 50or 100 �g/ml; for chloramphenicol, 20 �g/ml; for kanamycin 50 �g/ml; and fornalidixic acid, 30 �g/ml.

Construction of the genomic library and sequencing of the ML-449 biosyn-thetic gene cluster. Genomic DNA was isolated from Streptomyces sp. MP39-85according to the Kirby mix procedure (21). The genomic library for Streptomycessp. MP39-85 was constructed using a SuperCos 1 cosmid vector kit (Stratagene).The genomic DNA was partially digested with MboI and dephosphorylatedbefore ligation with XbaI-, CIAP-, and BamHI-treated SuperCos 1. E. coliXL1-Blue MR (Stratagene) was used as a host for the construction of the library.A probe for the ML-449 gene cluster was generated using degenerate primersdesigned for amplification of the conserved KS domain-coding regions of PKSgenes (15). PCR products were cloned using a Qiagen PCR cloning kit (Qiagen)and sequenced by MWG. Digoxigenin (DIG)-labeled probes were generatedusing a PCR DIG probe synthesis kit or a High Prime DNA labeling kit (bothfrom Roche Applied Science). The genomic library was screened as describedbefore (19), using the DIG-labeled probe. Cosmid DNA was isolated frompositive clones by using the Wizard Plus SV Minipreps DNA purification system(Promega) and end sequenced using primers designed for the cosmid regionsflanking the insert site (SuperCos_forw [5� GGC CGC AAT TAA CCC TCA C3�] and SuperCos_rev [5� GGC CGC ATA ATA CGA CTC AC 3�]). A BigDyeTerminator version 1.1 cycle sequencing kit (Applied Biosystems) was appliedfor the end sequencing. For complete sequencing of the cosmids, cosmid DNAwas isolated using a Genopure Plasmid maxikit (Roche). Sequencing was per-formed by the Centre for Ecological and Evolutionary Synthesis, Department ofBiology, University of Oslo.

Construction of gene disruption mutant and testing of production. A 3.63-kbBamHI-EcoRI fragment from one of the sequenced cosmids was ligated intopGEM3Zf(�) digested with BamHI-EcoRI. A 3.66-kb EcoRI-HindIII fragmentfrom this construct was ligated into pSOK201 digested with EcoRI-HindIII,resulting in the mlaA1 disruption vector. The vector was introduced intoET12567(pUZ8002) and then used for conjugation with Streptomyces sp.MP39-85 by following the procedure described by Flett et al. (10), but with thedonor cells grown to an optical density at 600 nm (OD600) of 0.4 to 0.5. Antibi-otics were added after 27 h of incubation. Resulting transconjugants were sub-jected to a Southern blot analysis to verify correct integration of the vector.Production of ML-449 was tested by LC-MS of fermentation extracts as de-scribed above.

Phylogenetic analyses. KS domains were assigned to the PKS sequences fromthe BE-14106 (GenBank accession number FJ872523) and ML-449 biosyntheticgene clusters as well as PKS sequences from 9 other PKS clusters retrieved fromthe NCBI database (http://www.ncbi.nlm.nih.gov/). PKS clusters for the biosyn-thesis of the following metabolites were chosen (the source organism andGenBank accession number are shown in parentheses): vicenistatin (Strepto-myces halstedii; AB086653), salinilactam (Salinispora tropica; NC_009380),leinamycin (Streptomyces atroolivaceus; AF484556), nystatin (Streptomycesnoursei; AF263912), pimaricin (Streptomyces natalensis; AJ278573), rapamy-cin (Streptomyces hygroscopicus; X86780), erythromycin (Saccharopolysporaerythrea; Y11199), spinosad (Saccharopolyspora spinosa; AY007564), and ri-famycin (Amycolatopsis mediterranei; AF040570).

AT domains were assigned to some of the PKS sequences from the BE-14106,ML-449, and salinilactam biosynthetic gene clusters and aligned with MlaK,BecK, Strop_2773, and VinK.

Mla/BecJ and -L were aligned with A domains from the bleomycin (BlmXA1/A2 and BlmIX), virginiamycin (VirA), nikkomycin (NikP1), novobiocin

VOL. 76, 2010 COMPARATIVE ANALYSIS OF MACROLACTAM CLUSTERS 285

on August 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from

(NovH), and yersiniabactin (HMWP2) biosyntheses as well as Strop_2774,Strop_2775, VinM, VinN, and LnmQ.

All protein sequences were aligned by CLUSTAL W (41) as implemented byMEGA 4.0.2 (38), using a gap opening penalty of 3.0 and a gap extension penaltyof 1.8 as suggested by Hall (13). Alignments were manually edited, and regionsthat could not be unambiguously aligned were deleted. Sequence alignmentswere exported in the FASTA format and converted to the PHYLIP format(input file for PhyML) by using FasToPhyNex (13).

The alignment of KS domains consisted of 151 sequences displaying 349 aminoacid positions. Phylogenetic trees were inferred by the maximum likelihood (ML)method using PhyML version 2.4.5 (12) and the Jones-Taylor-Thornton (JTT)model with invariants and gamma distribution. Tree reliability was estimated bythe approximate likelihood ratio test (aLRT) (1), using the SH-like option.

The alignment of AT domains consisted of 24 sequences displaying 281 aminoacid positions. The phylogenetic tree was inferred by ML, using PhyML with theWhelan and Goldman (WAG) model with invariants and gamma distribution.

The alignment of A domains/AMP-dependent synthetases/ligases consisted of16 sequences displaying 418 amino acid positions. The phylogenetic tree wasinferred by ML using PhyML with the WAG model.

Nucleotide sequence accession numbers. The DNA sequences for the ML-449biosynthetic gene cluster and the 16S rRNA gene of Streptomyces sp. MP39-85were deposited in GenBank under accession numbers FJ872525 and FJ872526,respectively.

RESULTS

Isolation of the ML-449 producer and identification of thecompound. Streptomyces sp. MP39-85 was isolated as a part ofa screening effort aimed at identifying new antifungal com-pounds produced by marine actinomycetes (18). The screeningresulted in isolation of Streptomyces sp. MP39-85, producing acompound with strong antifungal activity. DMSO extract fromStreptomyces sp. MP39-85 was subjected to LC-fractionation,identification of bioactive fractions, and LC-MS-TOF analysis.The LC-MS-TOF analysis of the bioactive fractions revealed asignificant chromatographic peak corresponding to a molecu-lar mass of 449.2924 Da (see Fig. S1 in the supplementalmaterial), and a search in the Dictionary of Natural Products(DNP) using a 10-ppm window returned no relevant hits forthe observed molecular mass. The DAD-UV profile of theunknown compound was, however, remarkably similar to thatof the previously characterized macrolactam BE-14106 (20). Inaddition, the observed molecular mass of the presumably newcompound was within 2 ppm of the expected molecular mass ofBE-14106 (see Fig. S2 in the supplemental material), but withthe addition of one C2H2 group (calculated molecular mass,449.293 Da). Taken together, these observations suggestedthat the identified compound might be represented by a novelmacrocyclic lactam structurally similar to BE-14106. The anti-microbial activity of a purified sample of ML-449 was deter-mined using the M. luteus bacterial strain and the C. albicansand C. glabrata yeast strains. It was found that the minimumconcentrations resulting in at least 50% inhibition of growth(IC50) were approximately 3.0 �g/ml for all three strains.

The chemical structure of ML-449 was determined usingNMR spectroscopy. Full correspondence of the NMR experi-mental conditions used in this work with those previously re-ported for structure elucidation of BE-14106 (22) allowed fordirect comparison of the NMR data acquired for ML-449 andBE-14106 and thus significantly simplified the assignment ofthe 1H and 13C NMR spectra for the former. A comparison ofthe 1H-13C HSQC spectra of ML-449 and BE-14106 (data notshown) revealed their close similarity. Upon comparison, cor-respondence for the vast majority of the observed cross peaks

could easily be found, and the total spectral patterns were thesame for both compounds. As the chemical shifts are verysensitive indicators of the nuclear chemical environment, thiscomparison indicated that the chemical structures of ML-449and BE-14106 macrolactam rings were identical. The only sig-nificant difference observed between the ML-449 andBE-14106 HSQC spectra occurred in the ranges from 5 to 6.8ppm and 122 to 144 ppm for proton and carbon resonances,respectively. The specified spectral regions for both com-pounds are shown in Fig. S3 in the supplemental material. The1H-13C HSQC spectrum of ML-449 clearly shows the presenceof two new cross peaks, while two other cross peaks are slightlyshifted compared to the analogous spectral region for BE-14106 (see Fig. S3, panel A, in the supplemental material).Comparison of the extra cross peaks’ chemical shifts with thoseassigned for BE-14016 indicated that ML-449 has one addi-tional double bond with respect to BE-14106. This, together,with the data on the accurate mass of ML-449 (and, later,information on the biosynthetic gene cluster), suggested thatthe C-19 acyl side chain of ML-449 contains 2 additional car-bon atoms and one additional double bond, compared to whatwas found for BE-14106. To prove this hypothesis, the fullNMR assignment for the C-19 side chain of ML-449 was per-formed using two-dimensional COSY, 1H-13C HSQC, and 1H-13C HMBC experiments. The obtained NMR data allowed forassignment of all the chain’s proton and carbon resonances,including those differing from the BE-14106 NMR signals (seeTable S1 in the supplemental material). The results clearlyshow that the additional double bond is situated in the ML-449side chain between C-23 and C-24 (Fig. 1A).

Cloning, sequencing, and analysis of the ML-449 biosyn-thetic gene cluster. A genomic cosmid library was establishedfor the ML-449 producer Streptomyces sp. MP39-85. A probefor the ML-449 biosynthetic gene cluster was obtained by PCRamplification of conserved KS domain-coding regions by usingdegenerate primers. Cloning and sequencing of these DNAfragments and subsequent BLAST analysis of the translatedsequences revealed KS domains with high similarity to PKSsequences from the pyrrolomycin biosynthetic gene cluster(44) and to VinP1 from the vicenistatin biosynthetic gene clus-ter (30). As a DNA fragment encoding a KS domain with highsimilarity to VinP1 was successfully used as a probe for theBE-14106 biosynthetic gene cluster (19), the latter was chosenas a probe for identification of the ML-449 gene cluster.Screening of the MP39-85 library with this probe, followed bygenome “walking” and partial DNA sequencing, yielded 4 cos-mids that presumably encompassed a complete ML-449 bio-synthetic gene cluster. These cosmids were fully sequenced,and analysis of the final, contiguous 83-kb sequence by use ofFrameplot (14) and BLAST identified 20 open reading frames(Table 1) within the presumed ML-449 biosynthetic gene clus-ter (mla). The organization of the mla cluster appeared to bealmost 100% identical to that of the BE-14106 biosyntheticgene cluster (bec) (Fig. 1C), with every gene in the bec clusterhaving a counterpart in the mla cluster (Table 1). The onlydifference was represented by the mlaA1 and mlaA2 PKSgenes. The latter gene is not present in the BE-14106 genecluster, while mlaA1 seemingly constitutes the complete ver-sion of the truncated becA. The deduced functional domains ofthe BecA, MlaA1, and MlaA2 proteins are shown in Fig. 2.

286 JØRGENSEN ET AL. APPL. ENVIRON. MICROBIOL.

on August 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from

The high level of similarity between the two clusters indicatesthat the biosynthesis must also be similar, and the structure ofML-449 points to the only difference being in the synthesis ofthe C-19-to-C-27 side chain. In the biosynthesis of BE-14106(19), BecA is presumed to supply the C-17-to-C-25 part of themolecule, including the C-19-to-C-25 side chain. The thirdmodule of BecA is incomplete and lacks the terminal ACPdomain, but a second truncated PKS, BecC, is believed toconstitute the missing part of the module and contains a KRdomain and an ACP domain (19). BecA and BecC are thoughtto complete the synthesis of the acyl chain, which is subse-quently modified into an aminoacyl. Such splitting of moduleson separate proteins has been reported in several cases (7, 23,32, 36). The structure of ML-449 indicates incorporation of

one additional acetate unit compared to what is observed inBE-14106 biosynthesis. MlaA1 constitutes a PKS with threecomplete modules, while MlaA2 is truncated and lacks the KRand ACP domains. In analogy to BecA and BecC, MlaA2 andMlaC presumably represent the fourth module, incorporatingthe last acetate unit of the C-27-to-C-17 acyl chain (Fig. 3A). Aconserved domain search for MlaA2 indicated the presence ofa docking domain in the N-terminal region of the enzyme, andthis prompted us to look for a C-terminal docking domain inMlaA1, which would support protein-protein interaction be-tween the two enzymes. Alignments of the C-terminal regionof MlaA1 and the N-terminal region of MlaA2 with dockingdomain sequences as described by Thattai et al. (40) revealedthe presence of highly conserved docking domains belonging tothe groups H1 and T1 (see Fig. S4 in the supplemental mate-rial). C-terminal docking domains from group H1 are assumedto interact with N-terminal docking domains from group T1(40), and the presence of such domains in MlaA1 and MlaA2indicates that there is an interaction between the enzymes. It istherefore reasonable to assume that MlaA2 and MlaC repre-sent the fourth module, which extends the unfinished acylchain synthesized by MlaA1 and provides the C-19 carbonylassumed to be necessary for amination and generation of anaminoacyl starter. Through feeding experiments with 15N la-beled amino acids, the nitrogen atom in the BE-14106 struc-ture was shown to originate from glycine, whereas parallelexperiments with 13C-labeled amino acids demonstrated thatthe carbon atoms from glycine were not incorporated into thestructure to any extent (19). The enzyme most likely to catalyzethe transfer of the amino group from glycine to the acyl chaingenerated by BecA and BecC is the putative glycine oxidaseBecI, although the exact mechanism of transamination remainsunclear (19). The biosynthesis of ML-449 is thought to proceedwith complete analogy to BE-14106 biosynthesis, with releaseand subsequent activation of the resulting �-amino acid

TABLE 1. Description of proteins encoded in genes identified in the ML-449 biosynthetic gene cluster and homologs

Protein Size (no. ofamino acids)

Homolog fromBE-14106

cluster

% Positives/% identity Proposed function Putative homolog(s) from

other macrolactam cluster(s)

MlaH 951 BecH 94/88 LuxR-type transcriptional regulator Strop_2761MlaA1 6383 BecA 91/86 Polyketide synthase type IMlaA2 1043 Polyketide synthase type IMlaI 363 BecI 89/79 Glycine oxidase/FAD-dependent oxidoreductaseMlaC 695 BecC 90/84 Polyketide synthase type IMlaU 187 BecU 95/88 Putative NRPS accessory proteinMlaB 3530 BecB 92/87 Polyketide synthase type IMlaJ 532 BecJ 91/86 AMP-dependent acyl-CoA synthetase/ligase VinN, Strop_2775MlaK 313 BecK 89/83 Acyltransferase VinK, Strop_2773MlaS 78 BecS 94/88 Peptidyl carrier protein VinL, Strop_2776, LnmPMlaL 504 BecL 90/83 NRPS adenylation domain VinM, Strop_2774, LnmQMlaM 198 BecM 89/86 TetR-type transcriptional regulator Strop_2766MlaN 523 BecN 94/87 MFS-type efflux pump Strop_2765MlaO 411 BecO 97/92 P450 monooxygenaseMlaD 3365 BecD 92/88 Polyketide synthase type IMlaP 311 BecP 91/84 Putative L-amino acid amidase/proline iminopeptidase VinJ, Strop_2777MlaG 1992 BecG 90/85 Polyketide synthase type IMlaF 3373 BecF 93/88 Polyketide synthase type IMlaE 1637 BecE 92/87 Polyketide synthase type IMlaT 88 BecT 82/74 Hypothetical protein, SimX2-like proteinMlaQ 256 BecQ 93/86 Thioesterase type II Strop_2763

FIG. 2. Modules and domains in MlaA1, MlaA2, and BecA. Com-parison of protein and nucleotide sequences indicates that the thirdmodule in BecA was generated by a deletion of the last part of a mlaA1ancestor gene and the first part of an mlaA2 ancestor gene. The KS3domain (green) of BecA is presumed to originate from the mlaA1ancestor gene, and the AT3 domain (yellow) is presumed to originatefrom the mlaA2 ancestor gene.

VOL. 76, 2010 COMPARATIVE ANALYSIS OF MACROLACTAM CLUSTERS 287

on August 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from

through ligation with coenzyme A, followed by loading on thefirst ACP domain of MlaB (Fig. 3A). The biosynthesis of themacrolactam then proceeds through all modules of the MlaBand MlaD to -G PKS enzymes, resulting in release and cycli-zation of the lactam by the terminal TE domain of MlaG (Fig.3B). P450 monooxygenase MlaO presumably hydroxylates thereleased macrolactam at C-8.

Gene inactivation experiment. To verify the involvement ofthe sequenced gene cluster in the biosynthesis of ML-449, a

gene inactivation experiment was performed. To construct asuicide vector for gene disruption, a 3.6-kb fragment repre-senting the part of mlaA1 encoding the KR2, ACP2, KS3, andAT3 domains was ligated with the part of the conjugativevector pSOK201 lacking the Streptomyces origin of replication(45) and transferred from Escherichia coli to Streptomyces sp.MP39-85 by conjugation. Resulting transconjugants were ver-ified by Southern blot analysis to contain the right insertion(data not shown), and production of ML-449 was tested by

FIG. 3. Biosynthesis of the aminoacyl starter unit for ML-449 (A) and completion of the macrolactam ring synthesis (B).

288 JØRGENSEN ET AL. APPL. ENVIRON. MICROBIOL.

on August 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from

LC-MS-TOF analysis of the culture extracts. The mutant wasfound to produce less than 1% of the wild-type level (see Fig.S5 in the supplemental material), confirming the involvementof the sequenced gene cluster in the biosynthesis of ML-449.

Comparison of the ML-449 and BE-14106 biosynthesis geneclusters. The organization of the genes in the mla cluster wasidentical to that of the bec cluster, with the exception of mlaA2(Fig. 1C). The similarity at the protein level was quite striking,with percent identities in the range of 80 to 90% for most ofthe deduced protein sequences (Table 1), thus suggesting acommon ancestor for the two clusters. Comparison of the geneclusters at the nucleotide level by use of the Artemis Compar-ison Tool (ACT) (5) showed that the DNA sequence is quiteconserved between the clusters, as most regions displayed apercent identity within the range of 80 to 93% (see Fig. S5 inthe supplemental material). Interestingly, the analysis dis-closed that the 3� end of becA is more similar to mlaA2 than tothe corresponding part of mlaA1. A closer inspection of thesequences revealed that the KS3 domain of BecA is moresimilar to the KS3 domain of MlaA1 than the KS domain ofMlaA2, while the opposite is true for the AT3 domain of BecA.This indicates that the becA gene is most likely a product ofrecombination between the mlaA1 and mlaA2 ancestor genes,so that BecA retained the KS3 of MlaA1 and gained a newAT3 domain from MlaA2 (Fig. 2).

Evolutionary analysis of the KS domains in the macrolac-tam PKS systems. The organization of the mla and bec geneclusters suggests that some of the PKS genes may have differ-ent origins. mla-becD, -E, -F, and -G are all transcribed in thesame direction and may form part of an operon, and theirtranslational products participate in macrolactam ring biosyn-thesis and are thus likely to have the same origin. mla-becB andC, although transcribed in the same direction as mla-becD to-G, are separated from the other PKS genes by genes encodingenzymes presumed to be involved in the activation of glycineand the aminoacyl starter as well as a regulator, an effluxpump, and a P450 monooxygenase. MlaB/BecB apparentlyparticipate in macrolactam ring biosynthesis, while MlaC/BecCare thought to complete the synthesis of the acyl side chain. Itis possible that the ancestor gene for mlaC-becC was originallya part of the mlaA2-becA ancestor gene, since MlaC/BecCseemingly represent the C-terminal parts of MlaA2/BecA. Thesplit module may have originated following a rearrangement ofthe cluster. Another possibility is that mlaC-becC representsthe remnants of a separate, partially deleted PKS gene. mlaA1-mlaA2 and becA, on the other hand, are positioned at the leftflank of their respective gene clusters and are transcribed inthe opposite direction compared to the other PKS genes in thecluster. Moreover, their translational products synthesize theacyl side chain separately from the macrolactam ring, and inthat respect, it does not seem unlikely that mlaA1-mlaA2-becAmay be of a different origin than mla-becB to -G.

To address the origin of the individual PKS modules in thetwo gene clusters, a phylogenetic analysis for the KS domainswas performed. Earlier analyses have demonstrated that KSdomains (from cis-AT PKSs) belonging to a particular biosyn-thetic gene cluster usually form cluster-specific clades, indicat-ing that the KS domains in individual pathways are generatedby duplication of single ancestor modules (11, 17, 25, 42). Fora gene cluster with PKS genes of different origins, one would

thus expect a phylogenetic separation of the respective KSdomains. The amino acid sequences of the mla and bec KSdomains were compared with those from 9 other characterizedPKS clusters, including 3 gene clusters involved in macrolac-tam biosynthesis (vicenistatin, salinilactam, and leinamycin),and a phylogenetic tree was reconstructed using the maximumlikelihood method (Fig. 4). The KS domains from the leina-mycin (lnm) pathway and the KSQ domains (represented bythe “loading” KS domains of BecA, MlaA1, and SpnA) werefound to be quite different from the rest of the KS domains,forming two separate clades. Leinamycin is synthesized by aPKS system where AT domains are not present in the individ-ual modules and the acyl-CoA substrates are supplied in transby a discrete acyltransferase (6). Piel et al. found that KSdomains from such pathways form a distinct clade from PKSwith cis-acting ATs, suggesting a single evolutionary origin forthe AT-less systems (31). The separation of the lnm KS do-mains from the other KS domains is therefore consistent withearlier observations. KSQ domains are specialized loadingmodule domains that lack condensation activity but retain theability to decarboxylate dicarboxylic acid starters (2, 24). Sev-eral studies have found that KSQ domains consistently groupwith other KSQ domains rather than the KS domains fromtheir respective PKS clusters (11, 28), and this was observed forthe KSQ domains included in this study as well. A secondfeature shown by the same studies is the clustering of KSdomains preceded by NRPS modules in hybrid NRPS-PKSassembly lines (11, 28). The formation of a separate clade bysuch KS domains may be explained by the functional constraintimposed by the unusual amino acid/peptidyl substrate (11).The first KS domain of MlaB/BecB formed a group with theKS1 domain from VinP1 and Strop_2768, and since these KSdomains are presumed to perform condensation of an acylextender unit onto an amino acid derivative/aminoacyl, theformation of a separate clade by these KS domains seems tosupport the above-mentioned observation. However, Ginolhacet al. found that the hybrid KS domain clade was separatedfrom the regular KS domains, while the KS domains from thehybrid systems in this study were localized among the regularKS domains (11).

The overall tree topology corresponded well with earlierstudies, as most KS domains grouped within the expected clus-ter-specific clades (17, 42). A few KS domains were found to beseparate from their respective clade (e.g., some KS domainsfrom the oligomycin [olm] and pimaricin [pim] biosyntheticgene clusters). Such a split has been observed for the same KSdomains in an earlier study (17). The KS domains from Mla/BecD to -G formed a subclade within a bigger macrolactam-specific (cis-AT) clade shared with all vicenistatin (vin) KSdomains (except the amino acid-accepting KS1 domain fromVinP1) and some salinilactam (slm) KS domains (fromStrop_2768 and -2778), suggesting common ancestry for themla, bec, and vin clusters and, partially, the slm cluster. Most ofthe remaining slm domains (from Strop_2778, -2780, and-2781) formed a separate group, except for the KS domainfrom Strop_2779, which was embedded into the polyene mac-rolide clade. The MlaA1/A2 and BecA KS domains were alsofound to group with the polyene clusters, forming a separatesubclade together with the KS domain from Strop_2779. Poly-ene macrolide clusters has been shown to form a mixed group

VOL. 76, 2010 COMPARATIVE ANALYSIS OF MACROLACTAM CLUSTERS 289

on August 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from

rather than cluster-specific clades (17, 42), and this analysisdisplays the same type of distribution for the nys and pim KSdomains. The association of the MlaA1/A2 and BecA KS do-mains with the polyene macrolide-specific clade suggests thatthe becA-mlaA genes might have originated from polyene mac-rolide PKS genes.

Concerning the issue of a common ancestry for the mla andbec clusters, the phylogenetic tree reconstruction for the KSdomains leaves little doubt, as the mla KS domains are alwaysdirect neighbors of their bec homologs from the correspondingmodules.

Enzymes presumed to be involved in activation/modificationof the aminoacyl starter. Mla/BecJ, Mla/BecK, Mla/BecS, andMla/BecL represent a discrete AMP-dependent synthetase/ligase, acyltransferase, peptidyl carrier protein, and NRPS ad-enylation domain, respectively, and are presumed to be in-volved in activation of glycine and the aminoacyl starter (19).The four genes encoding these enzymes form a subclusterwithin the ML-449 and BE-14106 biosynthetic gene clusters.Putative homologs of all four genes are found in the vin andslm clusters, while mla-becJ and mla-becK homologs are miss-

ing from the lnm cluster (Fig. 5A). In the vin cluster, the fourgenes encoding these enzymes (vinK, vinL, vinM, and vinN) areall transcribed in the same direction, as are the two putativehomologous genes present in the lnm cluster (lnmP and lnmQ).In the mla-bec clusters and slm cluster, these genes have dif-ferent organizations (Fig. 5A). Another interesting feature isthe location of vinP1, mla-becB, and strop_2768 in close prox-imity to these four genes, as the KS1 domains from VinP1,Mla/BecB, and Strop_2768 apparently accept the amino acidderivative/aminoacyl resulting from modifications/activationaccomplished by these (and some other) enzymes. A phyloge-netic analysis was undertaken for three of the enzymes andtheir putative homologs to establish the relationship betweenthem. Amino acid sequences of the putative NRPS adenylationdomains and AMP-dependent synthetases/ligases were alignedwith a small collection of A domains from other NRPSs. Aphylogenetic tree was reconstructed using the maximum like-lihood method (Fig. 5B). The A domains and the AMP-de-pendent synthetases/ligases are presumed to be homologous,and they all contain the 10 core motifs of the adenylate-form-ing superfamily of enzymes (26, 33). The analysis showed that

FIG. 4. Phylogenetic analysis of ketosynthase (KS) domains. The tree was reconstructed using the maximum likelihood method. aLRT statisticvalues are indicated at each node. The scale bar indicates 0.2 substitutions per amino acid position. The tree is unrooted. Green box, KS domainsinvolved in macrolactam ring synthesis; orange box, KS domains involved in acyl side chain synthesis; blue box, KS domains accepting aminoacid/aminoacyl; yellow box, KSQ domains.

290 JØRGENSEN ET AL. APPL. ENVIRON. MICROBIOL.

on August 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from

the putative adenylation domains Mla/BecL formed a separategroup together with Strop_2774, VinM, and LnmQ, suggestinga closer evolutionary relationship between these enzymes thanwith A domains from other NRPS clusters. The same cluster-ing was observed for Mla/BecJ, Strop_2775, and VinN, indi-cating that the four enzymes are more closely related to eachother than to the A domains included in the analysis. Theputative acyltransferases MlaK and BecK were aligned withVinK, Strop_2773, and a random selection of AT domainsfrom the mla, bec, and slm PKSs (both malonyl-CoA andmethylmalonyl-CoA-specific AT domains were included). Aphylogenetic tree was reconstructed using the maximum like-lihood method (Fig. 5C). As expected, malonyl-CoA-specificAT domains formed a separate clade from the methylmalonyl-CoA specific AT domains (11). The discrete acyltransferasesMla/BecK, VinK, and Strop_2773 formed a separate cladefrom the other AT domains, indicating a closer relationshipbetween these enzymes than with the other AT domains in theclusters. For comparison, a phylogenetic tree was also con-structed for the 16S rRNA gene sequences of Streptomyces sp.DSM 21069 (BE-14106 producer), Streptomyces sp. MP39-85,S. halstedii (3 strains), S. atroolivaceus (1 type strain plus 1additional strain), and S. tropica CNB440 (see Fig. S6 in thesupplemental material). 16S rRNA gene sequences for thevicenistatin producer S. halstedii HC34 and the leinamycinproducer S. atroolivaceus S-140 could not be found in the

Ribosomal Database Project (RDP) or GenBank, and se-quences from 3 strains reported as S. halstedii and 2 strainsassigned as S. atroolivaceus were included instead. The analysisconfirmed the assumption that Streptomyces sp. MP39-85 andStreptomyces sp. DSM 21069 are quite closely related.

DISCUSSION

Judging from the organization of the genes in the mla andbec clusters; the high levels of similarity at both the nucleotideand the protein levels; and the phylogenetic trees recon-structed for the KS, AT, and A domains, there seems to belittle doubt that the two gene clusters are very closely related.The 16S rRNA gene tree reconstruction is in agreement withthe above-mentioned observations, and it is therefore reason-able to assume that Streptomyces sp. DSM 21069 and Strepto-myces sp. MP39-85 share a recent common ancestor. The beccluster is thought to have arisen following a deletion in theancestor gene cluster.

As to the composition of the two gene clusters, the phylo-genetic reconstruction for the KS domains points to a separateancestry for the mlaA1-mlaA2-becA genes and the rest of thePKS genes, indicating a closer relationship with polyene mac-rolide clusters for the former. One possible origin of the mlaA-becA genes might be a separate polyene macrolide clusterpresent in the same strain. Alternatively, they might represent

FIG. 5. (A) Organization of genes involved in the starter unit synthesis for salinilactam, BE-14106, ML-449, vicenistatin, and leinamycin.Putative homologs are indicated by use of the same color: NRPS adenylation domains (yellow), peptidyl/acyl carrier proteins (green), acyltrans-ferases (red), and AMP-dependent synthetases/ligases (orange). (B, C) Phylogenetic trees for NRPS adenylation domains/AMP-dependentsynthetase/ligases (B) and acyltransferases (C). Phylogenetic trees were generated using PhyML (maximum likelihood). aLRT statistic values areindicated at each node. Trees are unrooted.

VOL. 76, 2010 COMPARATIVE ANALYSIS OF MACROLACTAM CLUSTERS 291

on August 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from

parts from a polyene macrolide gene cluster acquired throughhorizontal gene transfer.

The phylogenetic tree reconstructions for the KS, AT, and Adomains also point to a common ancestry for the mla-becclusters and the vin and slm clusters. KS domains from Mla/BecD to -G form a subclade with the KS domains from the vincluster and some KS domains from the slm cluster. The dis-crete AT domains and A domains/AMP-dependent synthetase/ligases from the four clusters display a closer relationship toeach other than to other AT and A domains. The colocaliza-tion of the genes encoding the discrete AT, A, and PCP do-mains and the AMP-dependent synthetases/ligases also pointto a relationship between the clusters, although several rear-rangements must have occurred during the evolution of theclusters. Considering the different amino acid substrates uti-lized in the different biosynthetic pathways, one might envisionthat these four enzymes represent a flexible platform for se-questering amino acids for modification and then subsequentactivation and loading of the amino acid derivative on the PKS.The discrete adenylation domain presumably activates theamino acid and loads it on the PCP domain to facilitate mod-ification by pathway-specific enzymes, such as the glycine oxi-dase from the BE-14106/ML-449 biosyntheses. Upon releasefrom the PCP domain, the AMP-dependent synthetase/ligaseactivates the now modified amino acid (or aminoacyl inthe case of BE-14106/ML-449) by ligation with CoA, makingthe amino acid derivative an acceptable substrate for the dis-crete acyltransferase. The acyltransferase subsequently loadsthe amino acid-CoA on a PKS loading module ACP domain,providing a starter for the biosynthesis of the macrolactamring. The system can be viewed as flexible in the sense thatdifferent types of enzymes could potentially be recruited formodification of the amino acid or the amino acid itself could beexchanged by altering the substrate specificity of the discreteadenylation domain.

In addition to the genes mentioned above, there are alsoother putative homologs in the vin and slm clusters. Table 1summarizes all potential homologs from the vin, slm, and lnmclusters. Particularly, the slm cluster appears to have much incommon with the mla and bec clusters, as a total of 9 possiblehomologs are present in all three clusters (excluding the PKSgenes). However, the overall organizations of the slm clusterand the mla-bec clusters are dissimilar. There are also severalother genes within the vin and slm clusters that are not presentin the other clusters, such as the genes encoding enzymespresumed to synthesize the vicenisamine sugar moiety of vice-nistatin and some enzymes presumed to be involved in modi-fication of the amino acid starter. Assuming a distant commonancestor for all of the four macrolactam gene clusters, severaldeletions, insertions, and rearrangements must be invoked toexplain the differences. The mechanism behind these eventsremains unclear, although homologous recombination be-tween different gene clusters may be the more likely explana-tion, as there is no trace of transfer-related elements, such asinsertion sequence (IS) elements or transposons, in the mla-bec clusters. As domains within PKS and NRPS enzymes seemto be readily exchangeable (8, 17), it does not appear unrea-sonable to think that other parts of PKS/NRPS gene clustersmay be exchangeable as well. All in all, this paints a picture forthe evolution of such gene clusters occurring through a natural

type of biocombinatorics, where individual genes, modules,and even domains can be recruited from other gene clusters orwithin the same cluster.

ACKNOWLEDGMENTS

We thank H. Sletvold for help with searching for IS elements andtransposons and I. Bakke for assistance with the phylogenetic analysisand for helpful comments on the manuscript. We are also thankful toH. Sletta and T. E. Ellingsen for discussions and comments.

This work was supported by the Research Council of Norway andthe Norwegian University of Science and Technology.

REFERENCES

1. Anisimova, M., and O. Gascuel. 2006. Approximate likelihood-ratio test forbranches: a fast, accurate, and powerful alternative. Syst. Biol. 55:539–552.

2. Bisang, C., P. F. Long, J. Cortes, J. Westcott, J. Crosby, A.-L. Matharu, R. J.Cox, T. J. Simpson, J. Stauntonk, and P. F. Leadlay. 1999. A chain initiationfactor common to both modular and aromatic polyketide synthases. Nature401:502–505.

3. Bredholt, H., E. Fjærvik, G. Johnsen, and S. B. Zotchev. 2008. Actinomyce-tes from sediments in the Trondheim fjord, Norway: diversity and biologicalactivity. Mar. Drugs 6:12–24.

4. Bruheim, P., S. E. F. Borgos, P. Tsan, H. Sletta, T. E. Ellingsen, J.-M.Lancelin, and S. B. Zotchev. 2004. Chemical diversity of polyene macrolidesproduced by Streptomyces noursei ATCC 11455 and recombinant strainERD44 with genetically altered polyketide synthase NysC. Antimicrob.Agents Chemother. 48:4120–4129.

5. Carver, T. J., K. M. Rutherford, M. Berriman, M. A. Rajandream, B. G.Barrell, and J. Parkhill. 2005. ACT: the Artemis Comparison Tool. Bioin-formatics 21:3422–3423.

6. Cheng, Y.-Q., G.-L. Tang, and B. Shen. 2003. Type I polyketide synthaserequiring a discrete acyltransferase for polyketide biosynthesis. Proc. Natl.Acad. Sci. U. S. A. 100:3149–3154.

7. Duitman, E. H., L. W. Hamoen, M. Rembold, G. Venema, H. Seitz, W.Saenger, F. Bernhard, R. Reinhardt, M. Schmidt, C. Ullrich, T. Stein, F.Leenders, and J. Vater. 1999. The mycosubtilin synthetase of Bacillus subtilisATCC6633: a multifunctional hybrid between a peptide synthetase, anamino transferase, and a fatty acid synthase. Proc. Natl. Acad. Sci. U. S. A.96:13294–13299.

8. Fewer, D. P., L. Rouhiainen, J. Jokela, M. Wahlsten, K. Laakso, H. Wang,and K. Sivonen. 2007. Recurrent adenylation domain replacement in themicrocystin synthetase gene cluster. BMC Evol. Biol. 7:183.

9. Fischbach, M. A., and C. T. Walsh. 2006. Assembly-line enzymology forpolyketide and nonribosomal peptide antibiotics: logic, machinery, andmechanisms. Chem. Rev. 106:3468–3496.

10. Flett, F., V. Mersinias, and C. P. Smith. 1997. High efficiency intergenericconjugal transfer of plasmid DNA from Escherichia coli to methyl DNA-restricting streptomycetes. FEMS Microbiol. Lett. 155:223–229.

11. Ginolhac, A., C. Jarrin, P. Robe, G. Perriere, T. M. Vogel, P. Simonet, andR. Nalin. 2005. Type I polyketide synthases may have evolved throughhorizontal gene transfer. J. Mol. Evol. 60:716–725.

12. Guindon, S., and O. Gascuel. 2003. A simple, fast, and accurate algorithm toestimate large phylogenies by maximum likelihood. Syst. Biol. 52:696–704.

13. Hall, B. G. 2008. Phylogenetic trees made easy: a how-to manual, 3rd ed.Sinauer Associates, Inc., Sunderland, MA.

14. Ishikawa, J., and K. Hotta. 1999. FramePlot: a new implementation of theframe analysis for predicting protein-coding regions in bacterial DNA with ahigh G�C content. FEMS Microbiol. Lett. 174:251–253.

15. Izumikawa, M., M. Murata, K. Tachibana, Y. Ebizukac, and I. Fujii. 2003.Cloning of modular type I polyketide synthase genes from salinomycin pro-ducing strain of Streptomyces albus. Bioorg. Med. Chem. 11:3401–3405.

16. Jenke-Kodama, H., A. Sandmann, R. Muller, and E. Dittmann. 2005. Evo-lutionary implications of bacterial polyketide synthases. Mol. Biol. Evol.22:2027–2039.

17. Jenke-Kodama, H., T. Borner, and E. Dittmann. 2006. Natural biocombina-torics in the polyketide synthase genes of the actinobacterium Streptomycesavermitilis. PLoS Comput. Biol. 2:e132.

18. Jørgensen, H., E. Fjærvik, S. Hakvåg, P. Bruheim, H. Bredholt, G. Klinken-berg, T. E. Ellingsen, and S. B. Zotchev. 2009. Candicidin biosynthetic genecluster is widely distributed among Streptomyces spp. isolated from the sed-iments and the neuston layer of the Trondheims fjord, Norway. Appl. En-viron. Microbiol. 75:3296–3303.

19. Jørgensen, H., K. F. Degnes, H. Sletta, E. Fjærvik, A. Dikiy, L. Herfindal, P.Bruheim, G. Klinkenberg, H. Bredholt, S. O. Døskeland, and S. B. Zotchev.2009. Biosynthesis of macrolactam BE-14106 involves two distinct PKS sys-tems and amino acid processing enzymes for generation of the aminoacylstarter unit. Chem. Biol. 16:1109–1121.

20. Keating, T. A., Z. Suo, D. E. Ehmann, and C. T. Walsh. 2000. Selectivity of

292 JØRGENSEN ET AL. APPL. ENVIRON. MICROBIOL.

on August 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from

the yersiniabactin synthetase adenylation domain in the two-step process ofamino acid activation and transfer to a holo-carrier protein domain. Bio-chemistry 39:2297–2306.

21. Kieser, T., M. J. Bibb, M. J. Buttner, K. F. Chater, and D. A. Hopwood. 2000.Practical Streptomyces genetics. The John Innes Foundation, Norwich,United Kingdom.

22. Kojiri, K., S. Nakajima, H. Suzuki, H. Kondo, and H. Suda. 1992. A newmacrocyclic lactam antibiotic, BE-14106. I. Taxonomy, isolation, biologicalactivity and structural elucidation. J. Antibiot. (Tokyo) 45:868–874.

23. Kopp, M., H. Irschik, S. Pradella, and R. Muller. 2005. Production of thetubulin destabilizer disorazol in Sorangium cellulosum: biosynthetic machin-ery and regulatory genes. Chembiochem 6:1277–1286.

24. Long, P. F., C. J. Wilkinson, C. P. Bisang, J. Cortes, N. Dunster, M. Oliynyk,E. McCormick, H. McArthur, C. Mendez, J. A. Salas, J. Staunton, and P. F.Leadlay. 2002. Engineering specificity of starter unit selection by the eryth-romycin-producing polyketide synthase. Mol. Microbiol. 43:1215–1225.

25. Lopez, J. V. 2003. Naturally mosaic operons for secondary metabolite bio-synthesis: variability and putative horizontal transfer of discrete catalyticdomains of the epothilone polyketide synthase locus. Mol. Genet. Genomics270:420–431.

26. Marahiel, M. A., T. Stachelhaus, and H. D. Mootz. 1997. Modular peptidesynthetases involved in nonribosomal peptide synthesis. Chem. Rev. 97:2651–2674.

27. Menzella, H. G., and C. D. Reeves. 2007. Combinatorial biosynthesis for drugdevelopment. Curr. Opin. Microbiol. 10:238–245.

28. Moffitt, M. C., and B. A. Neilan. 2003. Evolutionary affiliations within thesuperfamily of ketosynthases reflect complex pathway associations. J. Mol.Evol. 56:446–457.

29. Nguyen, T., K. Ishida, H. Jenke-Kodama, E. Dittmann, C. Gurgui, T. Hoch-muth, S. Taudien, M. Platzer, C. Hertweck, and J. Piel. 2008. Exploiting themosaic structure of trans-acyltransferase polyketide synthases for naturalproduct discovery and pathway dissection. Nat. Biotechnol. 26:225–233.

30. Ogasawara, Y., K. Katayama, A. Minami, M. Otsuka, T. Eguchi, and K.Kakinuma. 2004. Cloning, sequencing, and functional analysis of the biosyn-thetic gene cluster of macrolactam antibiotic vicenistatin in Streptomyceshalstedii. Chem. Biol. 11:79–86.

31. Piel, J., D. Hui, N. Fusetani, and S. Matsunaga. 2004. Targeting modularpolyketide synthases with iteratively acting acyltransferases from metage-nomes of uncultured bacterial consortia. Environ. Microbiol. 6:921–927.

32. Piel, J., G. Wen, M. Platzer, and D. Hui. 2004. Unprecedented diversity ofcatalytic domains in the first four modules of the putative pederin polyketidesynthase. Chembiochem 5:93–98.

33. Reger, A. S., J. M. Carney, and A. M. Gulick. 2007. Biochemical and crys-

tallographic analysis of substrate binding and conformational changes inacetyl-CoA synthetase. Biochemistry 46:6536–6546.

34. Ridley, C. P., H. Y. Lee, and C. Khosla. 2008. Evolution of polyketidesynthases in bacteria. Proc. Natl. Acad. Sci. U. S. A. 105:4595–4600.

35. Sekurova, O., H. Sletta, T. E. Ellingsen, S. Valla, and S. Zotchev. 1999.Molecular cloning and analysis of a pleiotropic regulatory gene locus fromthe nystatin producer Streptomyces noursei ATCC11455. FEMS Microbiol.Lett. 177:297–304.

36. Silakowski, B., G. Nordsiek, B. Kunze, H. Blocker, and R. Muller. 2001.Novel features in a combined polyketide synthase/non-ribosomal peptidesynthetase: the myxalamid biosynthetic gene cluster of the myxobacteriumStigmatella aurantiaca Sga15. Chem. Biol. 8:59–69.

37. Simunovic, V., J. Zapp, S. Rachid, D. Krug, P. Meiser, and R. Muller. 2006.Myxovirescin A biosynthesis is directed by hybrid polyketide synthases/non-ribosomal peptide synthetase, 3-hydroxy-3-methylglutaryl-CoA synthases,and trans-acting acyltransferases. Chembiochem 7:1206–1220.

38. Tamura, K., J. Dudley, M. Nei, and S. Kumar. 2007. MEGA4: MolecularEvolutionary Genetics Analysis (MEGA) software version 4.0. Mol. Biol.Evol. 24:1596–1599.

39. Tang, G.-L., Y.-Q. Cheng, and B. Shen. 2007. Chain initiation in the leina-mycin-producing hybrid nonribosomal peptide/polyketide synthetase fromStreptomyces atroolivaceus S-140. J. Biol. Chem. 282:20273–20282.

40. Thattai, M., Y. Burak, and B. I. Shraiman. 2007. The origins of specificity inpolyketide synthase protein interactions. PLoS Comput. Biol. 3:1827–1835.

41. Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W:improving the sensitivity of progressive multiple sequence alignment throughsequence weighting, position-specific gap penalties and weight matrix choice.Nucleic Acids Res. 22:4673–4680.

42. Traitcheva, N., H. Jenke-Kodama, J. He, E. Dittmann, and C. Hertweck.2007. Non-colinear polyketide biosynthesis in the aureothin and neoau-reothin pathways: an evolutionary perspective. Chembiochem 8:1841–1849.

43. Udwary, D. W., L. Zeigler, R. N. Asolkar, V. Singan, A. Lapidus, W. Fenical,P. R. Jensen, and B. S. Moore. 2007. Genome sequencing reveals complexsecondary metabolome in the marine actinomycete Salinispora tropica. Proc.Natl. Acad. Sci. U. S. A. 104:10376–10381.

44. Zhang, X., and R. J. Parry. 2007. Cloning and characterization of the pyr-rolomycin biosynthetic gene clusters from Actinosporangium vitaminophilumATCC 31673 and Streptomyces sp. strain UC 11065. Antimicrob. AgentsChemother. 51:946–957.

45. Zotchev, S., K. Haugan, O. Sekurova, H. Sletta, T. E. Ellingsen, and S. Valla.2000. Identification of a gene cluster for antibacterial polyketide-derivedantibiotic biosynthesis in the nystatin producer Streptomyces noursei ATCC11455. Microbiology 146:611–619.

VOL. 76, 2010 COMPARATIVE ANALYSIS OF MACROLACTAM CLUSTERS 293

on August 17, 2020 by guest

http://aem.asm

.org/D

ownloaded from