A novel actinomycete strain de-replication approach based on the diversity of polyketide synthase and nonribosomal peptide synthetase biosynthetic pathways

Please download to get full document.

View again

of 12
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Document Description
The actinomycetes traditionally represent one of the most important sources for the discovery of new metabolites with biological activity; and many of these are described as being produced by polyketide synthases (PKS) and nonribosomal peptide
Document Share
Document Tags
Document Transcript
  Appl Microbiol Biotechnol (2005) 67: 795  –  806DOI 10.1007/s00253-004-1828-7 APPLIED GENETICS AND MOLECULAR BIOTECHNOLOGY Angel Ayuso .Desmond Clark  .Ignacio González .Oscar Salazar .Annaliesa Anderson .Olga Genilloud  A novel actinomycete strain de-replication approach basedon the diversity of polyketide synthase and nonribosomal peptidesynthetase biosynthetic pathways Received: 13 July 2004 / Revised: 25 October 2004 / Accepted: 30 October 2004 / Published online: 13 January 2005 # Springer-Verlag 2005 Abstract  The actinomycetes traditionally represent oneof the most important sources for the discovery of newmetabolites with biological activity; and many of these aredescribedasbeingproducedbypolyketidesynthases(PKS)and nonribosomal peptide synthetases (NRPS). We present a strain characterization system based on the metabolic potentialofmicrobialstrainsbytargetingthesebiosyntheticgenes.Afteraninitialevaluationoftheexistingbiasderivedfrom the PCR detection in a well defined biosynthetic sys-tems,wedevelopedanewfingerprintingapproachbasedonthe restriction analysis of these PKS and NRPS amplifiedsequences. This method was applied to study the distribu-tion of PKS and NRPS biosynthetic systems in a collectionof wild-type actinomycetes isolated from tropical soil sam- ples that were evaluated for the production of antimicrobialactivities. We discuss the application of this tool as an al-ternative characterization approach for actinomycetes andwe comment on the relationship observed between the presence of PKS-I, PKS-II and NRPS sequences and theantimicrobial activities observed in some of the microbialgroups tested. Introduction In recent decades, natural-product screening programs de-voted an enormous effort to discovering new microbialsecondary metabolites with interesting biological activities.One of the most important steps in the drug discovery pro-cess from microbial natural products is the selection of new microorganisms, such as members of the actinobacteriagroup. The selection criteria applied to wild-type actinomy-cete strains evolved in response to these needs and include avast array of morphological, chemo-taxonomic and molec-ular fingerprinting methods (Peláez and Genilloud 2003). Inthe past decade, multiple molecular fingerprinting methodswere developed for   Streptomyces  and other members of theactinomycete group, based on random amplifications of repetitive sequences from microbial genomes (Anderson andWellington 2001). Alternatively, the amplification of specificribosomal gene sequences (Jensen et al. 1993; Welsh and McClelland 1991) or the analysis or restriction patterns of rDNA PCR products (Cook and Meyers 2003) were fre- quently used to evaluate diversity and to distinguish amonglarge actinomycete populations.There are several examples where PCR screens for genesassociated with secondary metabolism are used to evaluatethe biosynthetic potential of actinomycetes. These includenon-ribosomal peptide synthetases (NRPS), modular andaromatic polyketide synthases (PKS-I, PKS-II), hydroxy-methylglutaryl coenzyme A reductases and aminoglyco-side resistance genes (Metsa-Ketela et al. 1999; Anderson et al. 2002; Sigmund et al. 2003; Ritacco et al. 2003; Ayuso and Genilloud 2004). The drawback to these methods is that they are not effective at   “ dereplicating ”  or identifyingunique isolates without extensive sequencing of the PCR  products obtained. It is not currently known whether thedetection of high amplification frequencies in a given ac-tinomycete population reflects or not a potential to producesecondary metabolites with biological activity. If so, com- bining the data derived from traditional molecular finger- printing methods and new fingerprinting approaches basedon the metabolic potential to produce secondary metabo-lites would enable screening efforts to focus on the most talented groups, increasing the chances of finding second-ary metabolites with interesting biological activities.For this purpose, we initially evaluated the potential biasof the PCR amplification approach when targeting con- A. Ayuso .I. González .O. Salazar  .O. Genilloud ( * )Centro de Investigación Básica,Merck Sharp and Dohme de España S.A.,Josefa Valcárcel 38,Madrid, 28027, Spaine-mail: olga_genilloud@merck.comTel.: +34-91-3210568Fax: +34-91-3210614 D. Clark  .A. Anderson ( * )Merck Research Laboratories,770 Sumneytown Pike, WP16-100,P.O. Box 4, West Point, PA 19486, USAe-mail: liesa_anderson@merck.comTel.: +1-215-6123138Fax: +1-215-9931788  served sequences within biosynthetic domains, beforedeveloping a fingerprinting approach based on the restric-tion analysis of amplified metabolic gene sequences. Thismethod was later applied to study the occurrence of these biosynthetic systems in a collection of wild-type actino-mycetes isolated from tropical soil samples where the presence of PKS-I, PKS-II and NRPS sequences could berelated to the antimicrobial activity observed in some of themicrobial groups tested. Here, we discuss the application of this approach as an alternative dereplication method for actinomycetes on the basis of their potential to produce bioactive compounds. Materials and methods Bacterial strains and cosmidsThe wild-type actinomycetes used in this study wereisolated from three tropical soils collected on Martinique(Windward Islands, Central America). Soils were seriallydiluted and plated onto soil extract- and humic acid-basedagar media supplemented with 20 μ  g/ml nalidixic acid andcycloheximide; and colonies were isolated under a stereo-scope after observation of their distinct morphology.The type strains used were  Saccharopolyspora erythraea MA6655(ATCC11635), S.hygroscopicus  NRRL5491(ATCC29253) and  S. avermitilis  MA6847 (ATCC 31267). Cos-mids that contained the avermectin biosynthetic pathway(pVE855,pVE859,pVE923,pVE924)wereobtainedfromD.J. MacNeil (MacNeil et al. 1992). All strains were grown at 28°C on YME agar medium(0.4% yeast extract, 1% malt extract, 0.4% glucose, 0.2%Bacto-agar) and ATCC-2 liquid medium (0.5% yeast ex-tract, 0.3% beef extract, 0.5% peptone, 0.1% dextrose,0.2% potato starch, 0.1% CaCO 3 , 0.5% NZamine E).DNA manipulation  DNA extraction Total genomic DNA from the different actinomycetes usedin this study was recovered and purified as described byInnis et al. (1990).  PCR primers Consensus ketosynthase (KS) PCR primers KS-BEF (5 ′ -CCGCGCGAGGCGCTGGCCGTCGAC- 3 ′ ) andKS-BER (5 ′ -CCGCGCCGGGCGGGGGTCTCGTCGTTCGGCATCAGCGGCACCAACGCG-3 ′ ) were designed to univer-sally detect KS domains from PKS-I systems. DegeneratePCRprimerstargetspecificallyNRPSadenylationdomains(A3, A7R), PKS-I KS domains and methyl malonyl trans-ferase domains (K1F, M6R), as described by Ayuso andGenilloud (2004). A second degenerate KS primer, K2R  (5 ′ -CVTTCGGVVTCAGCGGSACBAA-3 ′ ) derived fromKS_BER, was used in combination with K1F to amplifyspecifically a shorter fragment of KS domains. PKS-II se-quences were amplified specifically using the primers KS α  and KS β  (5 ′ -TSGRCTACRTCAACGGSCACGG-3 ′ , 5 ′ -TACSAGTCSWTCGCCTGGTTC-3 ′ ), designed to target conserved sequences in KS α   and KS β  domains (A. Ayuso,unpublished data). Primers G1 andL1 were used to amplifythe variable ribosomal spacer regions between the 16S and23S rDNAs (Jensen et al. 1993). All primers were supplied  by ECOGEN.  PCR amplifications PCR amplifications were performed in a Peltier PTC-200thermal cycler in a final volume of 50 μ  l containing 10% of extracted DNA, 0.4 μ  mol of each primer, 0.2 mmol of eachof the four dNTPs (Roche), 5  μ  l of extracted DNA, 1 unit of Appligene Taq polymerase (with its recommendedreaction buffer) and 10% of DMSO. PKS-I amplificationswith primers KS-BE F/R were performed according to thefollowing profiles: 5 min denaturation at 95°C and 35 cy-cles of 1 min at 94°C, 1 min at 62°C and 1 min extension at 72°C, followed by 10 min at 72°C, 5 min denaturation at 95°C and 15 cycles of 1 min at 94°C, 1 min at 62°C and 1min extension at 72°C, followed by 10 min at 72°C. NRPS,PKS-I and PKS-II amplifications with degenerate primerswere performed according to the following profile: 5 min at 95°C and 35 cycles of 30 s at 95°C, 2 min at either 55°C(for K1F/M6R, K1F/K2R), 58°C (for KS α  /KS β ) or 59°C(for A3F/A7R) and 4 min at 72°C, followed by 10 min at 72°C. Amplification products were analyzed by electro- phoresis in 1% agarose gels stained with ethidium bromide.G1/L1 PCR amplifications were performed as described by Hirsch and Sigmund (1995). PCR products were analyzed by electrophoresis in 4  –  20% acrylamide gradient mini-gels (Criterion TBE gels, BioRad).RFLP analysisPCR products were cloned using the TOPO TA cloning kit (Invitrogen). Restriction analysis of the clones was per-formed using  Hin fI or   Sau 3A1 according to Sambrook and Russel (2001) and separation on 4  –  20% gradient  polyacrylamide gels (Criterion precast gel 4  –  20% TBE;Bio-Rad).Cosmid library constructionA total of 3.0 g (wet weight) of   Streptomyces  sp. strainDC328 mycelia was used for genomic DNA isolation.Genomic DNA was prepared for cloning by randomshearing and then end-repair was used to generate blunt ends; and sized DNA was then ligated into pWEB::TNC 796  (Epicentre Technologies) as per the manufacturer  ’ s in-structions. The DNA was then packaged using MaxPlack  packaging extracts and the cosmid was used to infect   Escherichia coli  EPI100-T1, according to the instructionsof the manufacturer (Epicentre Technologies). A total of 1,800 cosmid-containing clones were isolated and individ-ually frozen at   − 80°C in microtiter dish wells.DNA sequencingCloned products were sequenced using universal primersM13R-28 and M13F-20 in an ABI Prism dye terminator cycle sequencing kit (Perkin Elmer).Evaluation of antimicrobial activityIn vitro antimicrobial susceptibility tests were performedusing the following panel of strains:  Bacillus subtilis MB964 (ATCC 6633) and clinical isolates  Staphylococcusaureus  MB 5393,  E. coli  MB4926 and  Candida albicans MY1055, all from the Merck Culture Collection. The in-oculum and assay plates for bacteria and yeast were pre- pared as described by Suay et al. (2000). Agar plugs from wild-type actinomycete strains cultivated in YME agar for atleast7dayswereappliedtothesurfaceoftheassayplates,which were incubated at either 28°C (yeast) or 37°C (bac-teria). Growth inhibition zones were measured after 24 hof incubation.Data analysis of G1/L1 PCR amplifications and  HinF1 restriction fingerprintingG1/L1 PCR amplification patterns and restriction finger- printing patterns were analyzed with the BioNumerics ver.2.5 program (Applied Maths). A similar matrix was gen-erated using the Pearson correlation moment; and cluster analysis was performed with the UPGMA algorithm.Sequence database searching was done using BLAST(Altschul et al. 1990). Sequence alignments were done using CLUSTAL W (Thompson et al. 1994) and phyloge- netic analysis was done in PHYLIP (Felsenstein 1981), using the neighbor algorithm. The analysis was boot-strapped using 100 replicates. Results Development of the type I PKS-I KS consensus PCR screenThepreviouslydescribeddegenerateprimersK1FandM6R targeted specifically to PKS-I KS domains and methylmal-onyl transferase domains in actinomycetes and had shownthelimitationsoftheir usetodetectthefrequentlyoccurringmalonyl-transferase domains in PKS-I biosynthetic path-ways(AyusoandGenilloud2004).Tobroadenthedetection range,wefocusedexclusivelyonKSdomainsanddesignedtwo new pairs of KS PCR primers. One pair was non-degenerate (KS-BE F/R) and encompassed a region of 1,014 nt. The second was degenerate (K2R/K1R) and am- Table 1  Detection of KS domains in the four model systems studiedStrain PCR  primersPKS type I pathwaysTarget  pathwayKSdomainsin target  pathwayKS domainclonesscreenedKSdomainsdetectedfromtarget  pathwayKSdomainsdetectedfromother  pathwaysComment  Saccharopolysporaerythraea KS-BE Not known Erythromycin 6 10 1 1 Standard PCR: 1  μ  g/mlDNAtemplate concentration,×35 cycles60 2 5 100 ng/ml DNAtemplate concentration,×35 cycles60 2 9 100 ng/ml DNAtemplate concentration,×15 cycles Streptomycesavermitilis KS-BE 8 Avermectin 12 60 1 10 Chromosomal DNA60 6 0 Avermectin cosmids Streptomyces  spDC328KS-BE Not known Not known Not known30 - 5 Chromosomal DNA250 - 25  Streptomyces  sp DC328cosmids Streptomyceshygroscopicus K1F-K2R  Not known Rapamycin 12 25 2 14 Chromosomal DNA797   plified a 250 bp region. The specificity of both pairs of  primers was evaluated using the erythromycin biosyntheticcluster of   Saccharopolyspora erythraea  MA6655 and therapamicin biosynthetic system of   Streptomyces hygros-copicus  NRRL 5491 respectively as model systems. After amplification of   Saccharopolyspora erythraea  genomicDNA(1 μ  g)withprimers KS-BE F/R,PCRfragmentswerecloned and analyzed by restriction analysis with  Sau 3A1.The erythromycin biosynthetic pathway from  S. erythraea containssixKSdomains,fromwhichonlyonewasdetectedamong the two distinct clones selected according to their restriction profile. Further PCR optimization was done,which included reducing the number of PCR cycles andreducing the DNA template concentration (Table 1). Ul-timately, 11 KS domains were obtained, which included the previously detected KS domains and two additional newones (Fig. 1). Similarly, genomic DNA from  Streptomyceshygroscopicus  was amplified with the degenerate primersKS1F/KS2R and cloned. Twenty-five clones were selectedfor analysis and 16 different KS sequences were identified,of which two were from a rapamycin cluster and the rest were uncharacterized KS domains.To investigate the distribution of PKS-I KS domainswithin actinomycete isolates further, we compared the PCR detection of KS domains from isolates having cosmidlibraries with the KS domains obtained by direct PCR of the source DNA. To do this, we used  Streptomyces aver-mitilis , which has eight documented PKS-I pathwayswhich include avermectin and oligomycin (Omura et al.2001), and a soil isolate,  Streptomyces  sp DC328, whichhad an unknown polyketide production capacity but was positive by PCR using the KS-BE PCR primers. Eleven KSdomains were detected by screening  Streptomyces avermi-tilis  chromosomal DNA, of which one was from the aver-mectin biosynthetic pathway, seven from the oligomycin biosynthetic pathway and an additional three from un-characterized PKS-I pathways in this strain. Analysis of thefour overlapping cosmids that contained the avermectin Fig. 1  Sequence diversity of KS domains isolated from  S.erythraea  by PCR and compar-ison of the KS domains obtainedfrom  S. erythraea  with KSdomains from the erythromycin biosynthetic pathway. A KSdomain from the rapamycin biosynthetic cluster of   Strepto-myces hygroscopicus  wasincluded as a root (  RapA3 ;accession number X86780).The bootstrap values for 100 repli-cates are given.  Asterisks  repre-sent novel KS domains. Theaccession numbers for the  Sac-charopolyspora erythraea  KSdomains are:  S. erythraea  KS1AY678081,  S. erythraea  KS2AY678082,  S. erythraea  KS3AY678080,  S. erythraea  KS4AY678079,  S. erythraea  KS5AY678073,  S. erythraea  KS6AY678078,  S. erythraea  KS7AY678077,  S. erythraea  KS8AY678076,  S. erythraea  KS9AY678075,  S. erythraea  KS10AY678074,  S. erythraea  KS11AY678083798   biosynthetic pathway enabled the detection of six of the 12KS domains (Table 1). Finally, we also tested this approachwith  Streptomyces  sp. DC328. Chromosomal DNA wasscreened by PCR using the KS-BE PCR primers, the PCR  products were cloned and 30 clones were characterized byrestriction digestion with  Sau 3A1. After sequencing, fiveunique KS domains were identified. A systematic approachwas taken to further evaluate this strain for KS domains byscreening a cosmid library that contained 1,800 cosmids;and 49 were positive for KS domains by PCR (Table 1).This represented 2.6% of the library. PCR products werecloned and 250 clones were analyzed by restriction di-gestion. Eleven cosmids were identified that containedmultiple KS domains. Single KS domains were detectedfrom the remaining 38 cosmids that were positive for KSdomains by PCR (Fig. 2a). The different KS domainsidentified by restriction analysis were all sequenced andfound to represent 25 KS domains, including the fivedomains that were also detected by PCR of chromosomalDNA (Fig. 2 b). The diversity of novel KS domains from Streptomyces  sp. DC328 was compared with KS domainsfrom characterized PKS pathways (Fig. 2c).Detection of PKS-I, PKS-II and NRPS sequences inwild-type actinomycetesTaking into account the inherent limitations of the direct PCR method, we focused our study on a wild-type acti-nomycete population of 329 strains isolated from three Fig. 2  Evaluation of PKS-Iscreen using  Streptomyces  sp.DC328.  a  example of multiple-restriction profiles obtainedfrom amplification productsfrom a single cosmid ( i ), com- pared with single-restriction profiles that were obtained for 38 of the 49 positive cosmids( ii ). Each gel represents theresults for a single cosmid.  b Dendrogram of 20 of the uniquerestriction profiles obtainedfrom  Streptomyces  sp. DC328,using  Sau 3A1 digests.  c  Cluster analysis of the KS sequencesobtained for   Streptomyces  sp.DC328 compared with knownPKS-I KS domains.  Streptomy-ces  sp. DC328 sequences are in numbered red boxes ; and all thePKS-I KS domains from char-acterized pathways are  color-coded   with the key given in thefigure799
Similar documents
View more...
Search Related
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks