Abstract

Citation:Seipke RF, Barke J, Brearley C, Hill L, Yu DW, Goss RJM, et al. (2011) A SingleStreptomyces Symbiont Makes Multiple Antifungals to Support the Fungus Farming AntAcromyrmex octospinosus. PLoS ONE 6(8): e22028. https://doi.org/10.1371/journal.pone.0022028

Editor:Jae-Hyuk Yu, University of Wisconsin – Madison, United States of America

Received:April 21, 2011;Accepted:June 13, 2011;Published: August 3, 2011

Copyright: © 2011 Seipke et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding:This work was funded by the Medical Research Council Milstein Award (grant G0801721) to MIH, RJMG and DWY. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Insect fungiculture has been best studied in the attine ants (subfamily Myrmicinae, tribe Attini) whose common ancestor is estimated to have evolved agriculture around 50 million years ago[1]. Attine ants are so dependent on the cultivation of fungus that when a daughter queen leaves to found a new nest she carries a piece of the cultivar fungus in her mouth in order to establish a culture of that fungus in her new nest[1]. Intriguingly, attine ants have also evolved a mutualism with Actinobacteria that produce antibiotics that the ants use as weedkillers to keep their fungal gardens free of other microbes[2],[3],[4]. The relationship between Actinobacteria, fungal cultivar and attine ant has been intensely studied in the branch of higher attines known as the leaf-cutting ants (generaAtta andAcromyrmex) which harvest fresh vegetation to feed to their highly specialised fungal cultivar,Leucoagaricus gongylophorus[1]. The fungus has evolved lipid and carbohydrate rich hyphae known as gongylidia which the ants harvest and use as food[5]. Pathogens of the fungal garden, most notably fungi of the genusEscovopsis, if left unchecked, can destroy a fungal garden and lead to the collapse of the colony within weeks[6],[7].

Two overlapping but conflicting theories have been put forward to explain the evolution of mutualism between attine ants and Actinobacteria. The first suggests co-evolution of attine ants and Actinobacteria belonging to the genusPseudonocardia. This theory suggests that the fungus garden pathogenEscovopsis has also co-evolved and thatPseudonocardia andEscovopsis are engaged in an evolutionary arms race in which the bacteria evolve compounds that specifically targetEscovopsis but do not inhibit the growth of the fungal cultivar[5]. The second model suggests that attine ants select antifungal-producing Actinobacteria from the environment and is consistent with the identification of additional Actinobacterial genera on leaf-cutting ants, includingStreptomyces andAmycolatopsis species[3],[4],[8],[9]. However, these theories are not mutually exclusive and evidence suggests attine ants co-evolve withPseudonocardia bacteria and still select other antifungal producing bacteria from the soil, perhaps to prevent evolution of resistance in the fungal pathogens[9],[10]. Indeed, there is good evidence, both direct and indirect, that leaf-cutting ants use multiple antifungals produced by multiple Actinobacteria during the cultivation of their fungal gardens suggesting that both models do apply, at least in the higher attine genusAcromyrmex[3],[11]. Recent work demonstrated thatAcromyrmex octospinosus are associated with aPseudonocardia strain that may have co-evolved and aStreptomyces strain that was most likely acquired from the environment relatively recently[9].

One intriguing question that still needs to be addressed in the environmental recruitment model concerns how the ants select beneficial bacteria. Actinobacteria are well known producers of useful secondary metabolites, including around 60% of all known antibiotics[12] and it seems likely that this production capability is key to their success as mutualists. Clearly the production of antifungals makes them useful to the ants and we hypothesise that production of multiple antifungals with different targets by single Actinobacterial species would make them more attractive to the ants as mutualists. In this work we carry out a more in depth analysis of the antifungals made by one of the strains associated with the leaf-cutting antA. octospinosus, a species ofStreptomyces which has been proposed to support fungus growing ants through production of the polyene antifungal candicidin[4],[9]. We report the genome sequence and analysis of this strain indicating its capacity to make numerous antibiotics, including at least three antifungal compounds. Curiously, this strain makes both of the antifungals that have been reported in theStreptomyces attine ant mutualists but neither the candicidin or antimycins are obligatory for the inhibition of the co-evolved nest pathogenE. weberi. We propose that additional and as yet unknown antifungal(s) made by this strain specifically targetEscovopsis and that candicidin and antimycins offer protection against other microfungal weeds. We propose that the ability of this single species to make multiple antifungal compounds makes it an attractive acquisition for the ants and their fungiculture.

Results

Genome sequencing and analysis

To determine the antibiotic biosynthetic capability ofStreptomyces S4 a combination of shotgun, 3 kbp and 8 kbp paired end libraries were constructed and 454 pyrosequenced to generate >335 Mbp of sequence that was assembled into 12 scaffolds containing 211 large contigs. The genome consists of one ∼7.5 Mbp linear chromosome, which is within the size range reported for genomes of other sequenced streptomycetes, as well as one linear plasmid (∼180 kbp) and one circular plasmid (∼2 kbp)[13]. TheStreptomyces S4 genome was annotated by a combination of manual and automated methods and multiple biosynthetic gene clusters predicted to produce secondary metabolites were identified.Table 1 summarizes characteristics of predicted natural product biosynthetic gene clusters inStreptomyces S4. As expected based on our previous work, theStreptomyces S4 genome contained a candicidin biosynthetic gene cluster that shares 98% nucleotide sequence identity with the candicidin biosynthetic gene cluster fromStreptomyces sp. FR-008[14]. Other biosynthetic gene clusters of note are a non-ribosomal peptide synthetase (NRPS) biosynthetic gene cluster that is predicted to direct the biosynthesis of an antibacterial similar to mannopeptimycin, as well as a NRPS biosynthetic gene cluster whose predicted product is a gramicidin-like antibacterial. We note these antibacterials could be useful in eliminating competition during colonisation of the ant cuticle.Streptomyces S4 also contains a type II polyketide synthase (PKS) biosynthetic gene cluster that shares 100% nucleotide identity to that of the fredericamycin biosynthetic gene cluster characterized inS. griseus[15]. Fredericamycin is mostly known for its antitumor properties[16]. Additionally, there are six functionally unassigned biosynthetic gene clusters (three NRPS and three hybrid NRPS/PKS) that are not similar to gene clusters with known products. It is likely that at least some these clusters encode secondary metabolites with antibacterial or antifungal activity.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Putative secondary metabolites encoded byStreptomyces S4.

https://doi.org/10.1371/journal.pone.0022028.t001

Mutagenesis of the candicidin biosynthetic gene cluster does not abolish antifungal activity

Evidence that candicidin is not the sole antifungal generated byStreptomyces S4 is demonstrated by the retained antifungal activity of a mutant strain deficient in the biosynthesis of this compound. The candicidin biosynthetic gene cluster was disrupted by deletion of the polyketide synthase gene,fscC, which encodes the candicidin biosynthetic modules 6–10[14]. LC-MS analysis of butanol-extracted culture supernatants from the wild-type strain revealed a molecular ion (m/z 1109.6) consistent with that of candicidin D and showed characteristic polyene absorption bands in its UV spectrum, with absorbance maxima at 360, 380, 403 nm (Fig. 1). As predicted, the molecular ionm/z 1109.6 was not detected in the ΔfscC mutant, indicating that candicidin production is abolished in this strain (Fig. 1). Bioassays of the isogenic wild-type and ΔfscC strains againstC. albicans and the nest pathogenEscovopsis weberi demonstrated that loss of candicidin has no effect on the antifungal bioactivity ofStreptomyces S4 (Fig. 2). This result suggests thatStreptomyces S4 makes at least one additional antifungal compound that has not been identified previously, most likely encoded by one of the other biosynthetic gene clusters identified in this work.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Deletion of the candicidin biosynthetic gene,fscC abolishes production of candicidin.

LC-MS was used to analyze supernatant fromStreptomyces S4 wild-type and S4 ΔfscC. The extracted ion chromatogram for candicidin (m/z 1109.6) is shown and confirmed that only S4 wild-type and not the ΔfscC mutant produced candicidin. The UV visible spectra for the peak at RT 5.15 min displays absorption characteristics consistent with polyene compounds is also shown (bottom).

https://doi.org/10.1371/journal.pone.0022028.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 2. Antifungal bioactivity of the non-antifungal-producing strainStreptomyces lividans,Streptomyces S4 wild-type and mutant strains.

Bioassays withStreptomyces lividans, S4 wild-type, S4 ΔfscC, S4 ΔantC and S4 ΔfscC ΔantC againstEscovopsis weberi (top panel) andC. albicans (bottom panel) demonstrate that deletion offscC does not abolish antifungal activity and that deletion ofantC only reduces antifungal activity againstC. albicans and notE. weberi. The S4 ΔfscC ΔantC double mutant does not display reduced antifungal activity againstE. weberi suggesting the presence of an additional antifungal compound that is responsible for the phenotype observed during in vitro bioassays.

https://doi.org/10.1371/journal.pone.0022028.g002

Identification of the antimycin biosynthetic gene cluster

While this work was in progress another group reported that antimycins are produced by a number of the otherStreptomyces strains associated with attine ant nests[17]. Antimycins inhibit the respiratory chain and are known to have antifungal activity. We investigated whetherStreptomyces S4 is making antimycin compounds in addition to candicidin and hypothesised that antimycins could potentially account for the retained bioactivity againstEscovopsis observed for theStreptomyces S4fscC mutant. LC-MS analysis of culture supernatants of the wild-type strain identified eight compounds withm/z that match those reported for antimycins A1–A4 (Fig. 3). To determine if any of the eight compounds could be antimycins, we co-injected commercially available antimycin standards A1–A4 with our wild-type extract, which revealed that four of the eight compounds possess the same retention time as the antimycin A1–A4 standards (Fig. 3). Four of the eight S4 compounds were identical to the commercially available antimycin standards A1–A4 both in terms of UV absorbance profile and LC retention time. Whilst the remaining four S4 compounds possess the same UV absorbance characteristics as the antimycin standards (Fig. S1), and the samem/z parent ions as those of the standards, they exhibit different retention times (Fig. 3); further experiments are being carried out to identify and characterize these four previously unreported compounds.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 3. LC/MS analysis ofStreptomyces S4 wild-type and mutant strains compared to antimycin standards.

The extracted ion chromatograms for antimcyins A1–A4 are shown. Eight compounds consistent with the mass of antimycin A1–A4 were produced by S4 wild-type and S4 ΔfscC, but were not produced by the ΔantC mutant. Co-injection of antimycin A1–A4 with the S4 wild-type extract demonstrated that antimycin A1–A4 have the same retention time as four of the eight compounds produced by S4 wild-type. The UV visible spectra and ESI positive mode mass spectra for antimycin A1–A4 and the eight antimycin compounds produced by S4 wild-type are shown inFig. S1 andFig. S2, respectively. Antimycin A1: R₁ = CH(CH₃)CH₂CH₃, R₂ = (CH₂)₅CH₃. Antimycin A2: R₁ = CH(CH₃)₂, R₂ = (CH₂)₅CH₃. Antimycin A3: R₁ = CH(CH₃)CH₂CH₃, R₂ = (CH₂)₃CH₃. Antimycin A4: R1 = CH(CH₃)₂ R₂ = (CH₂)₃CH₃.

https://doi.org/10.1371/journal.pone.0022028.g003

To our knowledge the gene cluster that encodes the antimycin biosynthetic pathway has not been identified, despite these compounds first being isolated over 60 years ago[18]. The structure of antimycin suggests that it may be synthesized, at least in part, by an NRPS, and that threonine may be utilized as a substrate (Fig. 3). The Basic Local Alignment Search Tool[19] revealed a region of theStreptomyces S4 genome with 57% amino acid sequence identity to the threonine adenylation domain from the daptomycin biosynthetic protein, DptA[20]. This region of homology enabled us to identify a hybrid NRPS/PKS biosynthetic gene cluster that displays significant amino acid identity to a hybrid NRPS/PKS biosynthetic gene cluster present in bothS. albus andS. ambofaciens and potentially encodes for the biosynthesis of antimycins (Fig. 4).Table 2 displays the proposed functions of proteins present in the hybrid NRPS/PKS cluster. In order to determine if this biosynthetic gene cluster can direct the production of antimycin, we disrupted the hybrid NRPS/PKS gene,antC and assessed antifungal activity againstC. albicans in a plate bioassay. The ΔantC mutant displayed dramatically reduced antifungal activity againstC. albicans compared to that of the wild-type strain (Fig. 2). This strongly suggested that the product of this cluster was an antifungal compound and is consistent with the hypothesis that this cluster could potentially mediate the biosynthesis of antimycins, compounds known to possess strong antifungal activity againstC. albicans (Fig. 2). Confirmation that the hybrid NRPS/PKS encoded byStreptomyces S4 directs the biosynthesis of antimycins was obtained by comparing the LC-MS profiles of the wild-type, ΔfscC and ΔantC mutant strains. Extracted ion chromatograms revealed that the ΔantC mutant does not produce the eight antimycins and that the ΔfscC mutant retained the ability to produce these antimycin compounds (Fig. 3).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 4. Gene schematic of theStreptomyces S4 antimycin biosynthetic gene cluster and comparison to putative antimycin clusters inStreptomyces albus andStreptomyces ambofaciens.

The percent amino acid homology shared between S4 proteins and proteins inS. albus andS. ambofaciens is indicated in the shaded boxes. The draft genomic sequence ofS. albus is incomplete and the sequence for the putative antimycin biosynthetic gene cluster is split over contig 11 and contig 12 with an estimated gap of ∼820 bp in theantC gene. TheStreptomyces S4 antimycin biosynthetic gene cluster is located on scaffold06 at coordinates 81953–106578. The partial genome sequences ofStreptomyces S4,S. albus J1074, andS. ambofaciens ATCC 23877 are available under accession numbers CADY00000000, ABYC00000000 and AM238663, respectively. The gene names forS. albus have been shortened to eliminate the first nine numbers of the gene name (e.g. SalbJ_290 = SalbJ_010100000290).

https://doi.org/10.1371/journal.pone.0022028.g004

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Proposed functions of proteins encoded by the antimycin biosynthetic gene cluster.

https://doi.org/10.1371/journal.pone.0022028.t002

Discussion

Although the attine ant-fungal mutualism has been studied for more than a century, the antibiotic-producing Actinobacterial mutualists were only discovered ∼15 years ago and it is only very recently that scientists have started to address the nature of the antibiotics being produced by these bacteria[2],[4],[9],[17]. It has been hypothesised, although not proven, that these antibiotics are used by the ants to kill off contaminated parts of the garden and / or to suppress the growth of fungal pathogens including co-evolved pathogens in the genusEscovopsis and many other microfungal weeds[6],[7]. Recent studies have shown that strains belonging to two key genera are typically associated with attine ants, species ofPseudonocardia, which have been suggested to have co-evolved with the ants and to be transmitted vertically by the queens, and species ofStreptomyces which have been suggested to be more recently acquired from the environment[3],[4],[8],[9]. Two antifungals have been identified from proposed mutualist species in each genus and both inhibit the nest pathogenEscovopsis in vitro.Pseudonocardia associated with the lower attineA. dentigerum makes dentigerumycin andPseudonocardia associated with the higher attineA. octospinosus makes nystatin P1[2],[9].Streptomyces mutualists associated with higher attines of the genusAcromyrmex are known to produce the well-known antifungals candicidin and antimycins and it has been suggested that these compounds account for the bioactivity of thisStreptomyces strain againstEscovopsis[4],[9],[17]. To date almost all of this work has been carried out through isolation and mass spectrometry analysis of the antifungal compounds, although we used genome scanning of aPseudonocardia mutualist to identify a nystatin-like biosynthetic gene cluster and its product which we named nystatin P1[9].

In this work we have undertaken the first in-depth genome sequence analysis of a proposed attine ant mutualist, in this case a candicidin-producingStreptomyces strain isolated fromA. octospinosus garden worker ants collected in Trinidad[9]. Genome sequencing and analysis identified 17 gene clusters that are predicted to encode for known or unknown secondary metabolites, including the known gene cluster for candicidin biosynthesis. Following the discovery of antimycin production byStreptomyces strains isolated from attine ants in a separate study[17] we identified a gene cluster encoding a pathway that is consistent with antimycin biosynthesis. Surprisingly, despite antimycins first being isolated and characterised >60 years ago the antimycin biosynthetic pathway was not known. We identified eight compounds which we assigned as antimycins and then identified and disrupted the hybrid NRPS/PKS gene cluster which we predicted to encode antimycin biosynthesis. The production of the eight antimycin compounds was abolished in the mutant strain providing strong evidence that this gene cluster does indeed encode the antimycin biosynthetic pathway.

In bioassays of the wild-typeStreptomyces S4 strain alongside strains which cannot make candicidin, antimycins or either of these antifungal compounds we found that whilst antimycin- and candicidin-deficient strains had reduced activity against the human pathogenC. albicans their activity against the nest pathogenE. weberi was unaltered. This is curious as theStreptomyces S4 strain andE. weberi are thought to have co-evolved in this mutualism[21]. This suggests that despite previous research demonstrating that purified candicidin and antimycin preparations inhibit the growth ofEscovopsis in vitro neither compound is responsible for the activity observed in bioassays whereStreptomyces S4 is challenged withE. weberi. We conclude that these compounds potentially inhibit the growth of other microfungal weeds found in the ant-fungus gardens while additional and currently unknown antifungal(s) produced byStreptomyces S4 have stronger activity againstEscovopsis. We also propose that the combination of antifungals produced by this singleStreptomyces strain coupled with the antifungal(s) produced by aPseudonocardia strain isolated from the same nest provides a broad spectrum of antifungal activity that is used by the ants to farm their fungus. Furthermore, the antibacterials made byStreptomyces S4 potentially help it to outcompete other bacteria for the ant host.

Our data suggest that theStreptomyces strains isolated by other researchers from attine ants are likely to make additional antifungals since they appear to be closely related toStreptomyces S4. It will be important to re-examine the biosynthetic capability of these strains in order to fully understand the chemical basis of their interactions with attine ants and their fungal cultivar. This reflects a common problem in the field of natural product antibiotic discovery, in which the reisolation of known compounds hampers the discovery of new antibiotics. The approach we have outlined here is time consuming and technically challenging, but it is perhaps the only way to determine the entire biosynthetic capability of an antibiotic-producing strain particularly if some of the antibiotics being made, and their biosynthetic gene clusters, are new to science. Future work will be aimed at determining the products of the five unassigned biosynthetic gene clusters inStreptomyces S4 and identifying the additional antifungal compound(s) made by this strain. This is likely to involve significant challenges if, as we predict, these are novel secondary metabolites.

In conclusion, although good progress has been made recently we are still a long way from understanding the chemical basis of the symbioses between antibiotic-producing Actinobacteria and their attine ant hosts. We hope that our study will stimulate further research in this area and the identification of additional antifungal and antibacterial compounds in this system.

Materials and Methods

Growth media and strains

Streptomyces strains were routinely grown on soya flour mannitol (SFM) agar plates or in liquid TSB/YEME whileE. coli strains were grown on Lysogeny both- Lennox (LB)[22]. Media was supplemented with antibiotics as required at the following concentrations: carbenicillin (100 µg/ml), hygromycin B (50 µg/ml), nalidixic acid (25 µg/ml), apramycin (50 µg/ml). S4 was isolated and identified by 16S rDNA sequencing in a previous study (GenBank accession HM179229). Antifungal bioassays withC. albicans andE. weberi were carried out as described previously[9]. Strains and plasmids are described inTable 3.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Strains and plasmids used in this study.

https://doi.org/10.1371/journal.pone.0022028.t003

Supporting Information

Acknowledgments

We thank Sabine Grüschow and Mark Philo for assistance with LC-MS and useful discussions, Mark Buttner and Maureen Bibb for providing pKC1132 and The Genome Analysis Centre for sequencing and assembly of theStreptomyces S4 genome.

Author Contributions

Conceived and designed the experiments: RFS JB MIH. Performed the experiments: RFS JB LH. Analyzed the data: RFS JB LH. Contributed reagents/materials/analysis tools: CB DWY RJMG MIH. Wrote the paper: RFS RJMG MIH.

References

1. 1.Schultz TR, Brady SG (2008) Major evolutionary transitions in ant agriculture. Proc Natl Acad U S A 105: 5435–5440.
   • View Article
   • Google Scholar
2. 2.Oh DC, Poulsen M, Currie CR, Clardy J (2009) Dentigerumycin: a bacterial mediator of an ant-fungus symbiosis. Nat Chem Biol 5: 391–393.
   • View Article
   • Google Scholar
3. 3.Sen R, Ishak HD, Estrada E, Dowd SE, Hong E, et al. (2009) Generalized antifungal activity and 454-screening ofPseudonocardia andAmycolatopsis bacteria in nests of fungus-growing ants. Proc Natl Acad U S A 106: 17805–17810.
   • View Article
   • Google Scholar
4. 4.Haeder S, Wirth R, Herz H, Spiteller D (2009) Candicidin-producingStreptomyces support leaf-cutting ants to protect their fungus garden against the pathogenic fungusEscovopsis. Proc Natl Acad U S A 106: 4742–4746.
   • View Article
   • Google Scholar
5. 5.Currie CR (2001) A community of ants, fungi, and bacteria: a multilateral approach to studying symbiosis. Annu Rev Microbiol 55: 357–380.
   • View Article
   • Google Scholar
6. 6.Reynolds HT, Currie CR (2004) Pathogenicity ofEscovopsis weberi: the parasite of the attine ant-microbe symbiosis direclty consumes the ant-cultivated fungus. Mycologia 96: 955–959.
   • View Article
   • Google Scholar
7. 7.Rodrigues A, Bacci M, Mueller UG, Ortiz A, Pagnocca FC (2008) Microfungal ‘weeds’ in the leafcutter ant symbiosis. Microb Ecol 56: 604–614.
   • View Article
   • Google Scholar
8. 8.Kost C, Lakatos T, Bottcher I, Arendholz W-R, Redenbach M, et al. (2007) Non-specific association between filamentous bacteria and fungus-growing ants. Naturwissenschaften 94: 821–828.
   • View Article
   • Google Scholar
9. 9.Barke J, Seipke RF, Grüschow S, Heavens D, Drou N, et al. (2010) A mixed community of actinomycetes produce multiple antibiotics for the fungus farming antAcromyrmex octospinosus. BMC Biol 8: 109.
   • View Article
   • Google Scholar
10. 10.Barke J, Seipke RF, Yu DW, Hutchings MI (2011) A mutualistic microbiome: how do fungus-growing ants select their antibiotic-producing bacteria. Commun Integr Biol 4: 41–43.
    • View Article
    • Google Scholar
11. 11.Mueller UG, Dash D, Rabeling C, Rodrigues A (2008) Coevolution between attine ants and actinomycete bacteria: a reevaluation. Evolution 62: 2894–2912.
    • View Article
    • Google Scholar
12. 12.Challis GL, Hopwood DA (2003) Synergy and contingency as driving forces for the evolution of multiple secondary metabolite production byStreptomyces species. Proc Natl Acad U S A 100: 14555–14561.
    • View Article
    • Google Scholar
13. 13.Seipke RF, Crossman L, Drou D, Heavens D, Bibb MJ, et al. (2011) Draft genome sequence ofStreptomyces S4, a symbiont of the leafcutter antAcromyrmex octospinosus. J Bacteriol.
    • View Article
    • Google Scholar
14. 14.Chen S, Huang X, Zhou X, Bai L, He J, et al. (2003) Organizational and mutational analysis of a complete FR-008/candicidin gene cluster encoding a structurally related polyene complex. Chem Biol 10: 1065–1076.
    • View Article
    • Google Scholar
15. 15.Wendt-Pienkowski E, Huang Y, Zhang J, Li B, Jiang H, et al. (2005) Cloning, sequencing, analysis, and heterologous expression of the fredericamycin biosynthetic gene cluster fromStreptomyces griseus. J Am Chem Soc 127: 16442–16452.
    • View Article
    • Google Scholar
16. 16.Pandey RC, Toussaint MW, Stroshane RM, Kalita CC, Aszalos AA, et al. (1981) Fredericamycin A, a new antitumor antibiotic I. production, isolation and physicochemical properties. J Antibiot 34: 1389–1401.
    • View Article
    • Google Scholar
17. 17.Schoenian I, Spiteller M, Ghaste M, Wirth R, Herz H, et al. (2011) Chemical basis of the synergism and antagonism in microbial communities in the nests of leaf-cutting ants. Proc Natl Acad U S A 108: 1955–1960.
    • View Article
    • Google Scholar
18. 18.Dunshee BR, Leben C, Keitt GW, Strong FM (1949) The isolation and properties of antimycin A. J Am Chem Soc 71: 2436–2437.
    • View Article
    • Google Scholar
19. 19.Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic local alignment search tool. J Mol Biol 215: 403–410.
    • View Article
    • Google Scholar
20. 20.Miao V, Coeffet-LeGal M-F, Brian P, Brost R, Penn J, et al. (2005) Daptomycin biosynthesis inStreptomyces roseosporus: cloning and analysis of the gene cluster and revision of peptide stereochemistry. Microbiol 151: 1507–1523.
    • View Article
    • Google Scholar
21. 21.Currie CR, Wong B, Stuart AE, Schultz TR, Rehner A, et al. (2003) Ancient tripartite coevolution in the attine ant-microbe symbiosis. Science 299: 386–388.
    • View Article
    • Google Scholar
22. 22.Kieser T, Bibb MJ, Buttner MJ, Chater KF, Hopwood DA (2000) PracticalStreptomyces Genetics. Norwich, UK: The John Innes Foundation. 613 p.
23. 23.Bierman M, Logan R, O'Brien K, Seno ET, Rao RN, et al. (1992) Plasmid cloning vectors for the conjugal transfer of DNA fromEscherichia coli toStreptomyces spp. Gene 116: 43–49.
    • View Article
    • Google Scholar
24. 24.Gust B, Chandra G, Jakimowicz D, Yuqing T, Bruton CJ, et al. (2004) Lambda Red-mediated genetic manipulation of antibiotic-producingStreptomyces. Adv Appl Microbiol 54: 107–128.
    • View Article
    • Google Scholar
25. 25.MacNeil DJ, Gewain KM, Ruby CL, Dezeny G, Gibbons PH, et al. (1992) Analysis ofStreptomyces avermitilis genes required for avermectin biosynthesis utilizing a novel integrative vector. Gene 111: 61–68.
    • View Article
    • Google Scholar
26. 26.Maconi P, Bistoni F, Boncio A, Boncio L, Bersiani A, et al. (1976) Utilizzazione di una soluzione salina ipertonica di cloruro di potassio (3M KCl) per l'estrazione di antigeni solubili da Candida albicalns. Ann Sclavo 18: 61–66.
    • View Article
    • Google Scholar
27. 27.Gust B, Challis GL, Fowler K, Kieser T, Chater KF (2003) PCR-targetedStreptomyces gene replacement identifies a protein domain needed for biosynthesis of the sesquiterpene soil odor geosmin. Proc Natl Acad U S A 100: 1541–1546.
    • View Article
    • Google Scholar