A trap for in situ cultivation of filamentous actinobacteria

Available online at www.sciencedirect.com Journal of Microbiological Methods 72 (2008) 257 – 262 www.elsevier.com/locate/jmicmeth A trap for in situ cultivation of filamentous actinobacteria Ekaterina Gavrish, Annette Bollmann, Slava Epstein, Kim Lewis ⁎ Department of Biology and Antimicrobial Discovery Center, Northeastern University, Boston, Massachusetts 02115, United States Received 21 August 2007; received in revised form 11 December 2007; accepted 14 December 2007 Available online 5 February 2008 Abstract The approach of growing microorganisms in situ, or in a simulated natural environment is appealing, and different versions of it have been described by several groups. The major difficulties with these approaches are that they are not selective for actinomycetes — a group of grampositive bacteria well known as a rich source of antibiotics. In order to efficiently access actinomycetes, a trap for specifically capturing and cultivating these microorganisms in situ has been developed, based on the ability of these bacteria to form hyphae and penetrate solid environments. The trap is formed by two semi-permeable membranes (0.2–0.6 μm pore-size bottom membrane and 0.03 μm pore-size top membrane) glued to a plastic washer with sterile agar or gellan gum inside. The trap is placed on top of soil, and filamentous microorganisms selectively penetrate into the device and form colonies. Decreasing the size of the pores of the lower membrane to 0.2 μm restricted penetration of fungi. The trap produced more filamentous actinobacteria, and a higher variety of them, as compared to a conventional Petri dish cultivation from the same soil sample. Importantly, the trap cultivation resulted in the isolation of unusual and rare actinomycetes. © 2007 Elsevier B.V. All rights reserved. Keywords: Trap; In situ cultivation; Filamentous actinobacteria 1. Introduction Traditionally, the main source for antibiotics has been natural products derived from bacteria and fungi (Demain and Fang, 2000). However, this source has been steadily drying up, which is likely due to repeated isolation and screening of the same microorganisms. Indeed, the rediscovery of common antibiotics from cultivable species is a major obstacle for antibiotic discovery (Baltz, 2006). Such over-mining is not surprising considering the limited number of culturable species, which make up only a fraction of the total diversity in the environment (Osburne et al., 2000). The majority of known antibiotics come from Actinomycetes (Hopwood, 2006) and developing methods to access new representatives of this group of microorganisms is of particular importance. We previously introduced a method of in situ cultivation that bypassed the difficulties of replicating the natural environment inherent in traditional Petri dish-based approaches (Kaeberlein ⁎ Corresponding author. Department of Biology, 134 Mugar Hall, 360 Huntington Avenue, Boston, MA 02115, United States. Tel.: +1 617 373 8238; fax: +1 617 373 3724. E-mail address: [email protected] (K. Lewis). 0167-7012/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.mimet.2007.12.009 et al., 2002). That method is based on placing bacteria in a diffusion chamber that is then introduced back into the environment from which the sample originated. The diffusion chamber is a mix of agar and diluted environmental sample sandwiched between two semi-permeable membranes glued onto a washer. The chamber allows for a free diffusion of chemicals while restricting the movement of cells, and allowed for the cultivation of up to 40% of bacterial cells from a marine sediment environment, as compared to 0.05% that grew on a Petri dish (Kaeberlein et al., 2002). We found that the diffusion chambers can be used to grow soil microorganisms as well, but only a small portion of the bacteria that grow in the chamber are actinomycetes. Here we describe a novel method for targeted isolation of Actinomycetes. This method employs a diffusion chamber in a very different way: as a trap for filamentous microorganisms. While in the original approach (Kaeberlein et al., 2002), the diffusion chamber is inoculated with target organisms, and is then incubated in the natural environment from which these organisms originate, the trap is filled with sterile agar and placed back in the environment. The expectation is that filamentous actinomycetes would penetrate through the pores of the membrane (Hirsch, Christensen, 1983; Polsinelli, Mazza, 258 E. Gavrish et al. / Journal of Microbiological Methods 72 (2008) 257–262 1984) and grow in the unoccupied space within the trap. The conditions inside the trap will closely mimic those outside of the trap, leading to growth of filamentous species. This study describes the successful application of the trap method to cultivate novel Actinomycetes. 2. Materials and methods 2.1. Soil sampling Garden soil samples were collected in Verrill Farm (Massachusetts, USA) in October 2005 and Pasadena (California, USA) in February 2006. The pine soil sample was collected in Dover (Massachusetts, USA) in September 2006. Samples were transported to the laboratory in Boston (Massachusetts, USA) and stored at room temperature before use. 2.2. Media used for isolation of bacteria For all traps 1% agar or 1.2% gellan gum supplemented with a 1% vitamin supplement (ATCC) was used as a cultivation medium. For pine soil traps the pH of the medium was adjusted to 4 by adding a few drops of HCl. For standard plating isolation CN agar medium (0.1% Casamino acids, 0.1% Nutrient broth (Difco), 1% Bacto agar), CN gellan gum medium (CN medium with 1.2% gellan gum) and Actinomycete Isolation (AI) Agar (Difco) were used. For isolation from the pine soil sample the pH of the media was adjusted to 4 by adding a few drops of HCl. Nystatin (50 μg/ml) and cycloheximide (100 μg/ml) were added to the media in order to prevent the growth of fungi. 2.3. The trap design and in situ cultivation To form the trap, a 0.2- to 0.6-μm pore-size polycarbonate membrane (47 mm diameter, Isopore, Millipore, USA) was glued to the bottom of a nylon washer (56 mm outer diameter, 35 mm inner diameter, 3 mm thickness, # SPS-2209-1375-0125, from WashersUSA.com), and 3 ml of sterile 1% agar or 1.2% gellan gum with vitamin supplement was poured onto the filter. After the medium solidified, a top polycarbonate membrane (0.03-μm pore size, 47 mm in diameter, Osmonics Inc., USA) was glued to the washer sealing the trap (Fig. 1). The membranes and washer were autoclaved prior to use. A diffusion chamber formed by two 0.03-μm pore-size filters with sterile medium inside was used as a negative control. All manipulations were made aseptically to prevent contamination. After the glue dried, the traps were placed on top of moist soil, insuring that the bottom filter was in good contact with soil. Large Petri dishes (15 cm diameter) were used as a soil reservoir. The plates were sealed with parafilm to prevent evaporation and were incubated for 14–21 days at room temperature in the dark. 2.4. Isolation of bacteria with traps After incubation the traps were opened, the solid agar or gellan content was removed in one piece, inverted and placed Fig. 1. Image and diagram of the trap. 1, plastic or metal washer; 2, top membrane filter (0.03-pm pore size); 3, bottom membrane filter (0.2- to 0.6-μm pore size); 4, agar or gellan gum. into a sterile Petri dish. The solid disks were examined under a stereomicroscope at 20–100× magnification (Discovery V12, Zeiss, Germany). Visible microcolonies were picked with sterile needles and streaked out on plates with CN agar or CN gellan gum medium. Subcultivation was repeated to obtain pure cultures. An additional incubation of the medium from the trap for 5–7 days allowed for some actinomycetes to form an aerial mycelium, which made it easier to pick and subcultivate microcolonies. To prevent drying several drops of water were added to the Petri dish that contained trap medium. 2.5. Isolation of bacteria by a standard isolation method One gram of air-dry soil was mixed with 9 ml sterilized water and vortexed. After the soil particles settled, serial dilutions were made and a 100 μl aliquot from each dilution was plated on CN agar, CN gellan gum, and Actinomycetes agar plates. After two weeks of incubation individual colonies were randomly picked and streaked out to obtain pure cultures. 2.6. Sequencing of 16S rRNA genes and phylogenetic analysis For identification purposes, single colonies were suspended in PCR grade water with approximately 0.05 g of zirconium beads (0.1 mm, BioSpec Products, Inc. USA) and vortexed for 5 min at maximum speed. The supernatants were used for PCR amplification with actinobacteria-specific primers Act235F (5′ CGCGGCCTATCAGCTTGTTG) and Act-878R (5′ CCGTACTCCCCAGGCGGG) (modified after (Stach E. Gavrish et al. / Journal of Microbiological Methods 72 (2008) 257–262 259 Fig. 2. Microorganisms growing in the trap. The traps were opened after 2 weeks of cultivation. a. Bacterial colonies and microfungal hyphae, agar-base trap with 0.4-μm pore-size bottom membrane. Bar, 1 mm. b. Actinomycetes-like microcolonies, gellan gum trap with 0.2-μm pore-size bottom membrane. Bar, 0.1 mm. et al., 2003) and GoTaq Green Master Mix (Promega, USA) with an annealing temperature of 60 °C running for 30 cycles. The PCR products were purified and sequenced commercially (Seqwright, Houston Texas, USA) using Fluorescent dye terminator. The sequences were edited using 4Peaks software (http://mekentosj.com/4peaks/), aligned to the ARB database (Ludwig et al., 2004), and added to the tree in the ARB database with the parsimony addition tool. Nucleotide sequences obtained in this study have been deposited in the GenBank database under accession numbers EF599956–EF600030, EF601722, EF601723 and EU407479. 3. Results and discussion Mycelia of filamentous Actinobacteria can grow through solid substrates, and have been shown to penetrate pores as small as 0.22 μm (Hirsch and Christensen (1983) and Polsinelli and Mazza (1984). This ability was used for selective isolation of Fig. 3. Microcolonies of Actinobacteria from the trap. The trap was incubated for an additional week at room temperature after it was opened. Actinobacterial aerial (a, b, c) and substrate (d) mycelia. Gellan gum trap with 0.2-μm pore-size bottom membrane. Bar, 0.05 mm. 260 E. Gavrish et al. / Journal of Microbiological Methods 72 (2008) 257–262 actinomycetes from mixed populations (soil, water, vegetable materials) on the surface of cellulose membranes placed onto nutrient agar (Hirsch and Christensen, 1983; Polsinelli and Mazza, 1984). During the incubation, actinomycetes mycelia penetrated through the filter pores to the underlying agar medium and formed colonies, whereas other bacteria and fungi were restricted from the agar by the membranes. This procedure is simple and suitable for actinomycetes isolation but requires cultivation in Petri plates and does not simulate the natural environment. We reasoned that we could similarly select for actinobacteria and enable growth of species that do not grow in Petri dishes by a reverse use of the diffusion chamber (Kaeberlein et al., 2002). In this way, a diffusion chamber is not inoculated with cells, but rather placed into the environmental sample empty, and then traps filamentous microorganisms that penetrate its membrane. Traps were initially incubated on top of garden soil (collected in California and transported to the home lab). After 14 days of incubation the traps were opened, and the solid agar or gellan content was removed and examined. A great majority of colonies formed well-developed vegetative mycelia that were easily detected under the stereomicroscope (Fig. 2a,b). This showed that the trap could indeed selectively capture filamentous organisms that penetrated the pores of the filters and settled in the unoccupied space of the chamber. The trap with 0.4 μm and larger pore-size filters contained both bacteria and fungi (Fig. 2a). The fungi grew more rapidly than the actinomycetes and filled the entire trap with fungal hyphae, effectively spreading throughout the trap, making isolation of actinomycetes impractical. Actinomycetes form thinner hyphae compared to fungi, and it seemed possible to selectively trap them by using a membrane with smaller pores that would exclude fungi. Decreasing the pore diameter from 0.4 to 0.2 μm indeed excluded fungi, and the trap was found to primarily contain colonies of filamentous actinomycetes (Fig. 2b). The traps with 0.2 μm pore-size bottom filters were used for further experiments. The solid media slabs from traps were incubated for an additional 5–7 days, after which well-separated aerial mycelia were apparent in a number of cases (Fig. 3). Colonies were picked with sterile needles and purified by streaking them out on agar plates with CN medium. The majority of isolates showed good growth after approximately one week of cultivation. In order to survey the actinomycetes obtained by conventional Petri dish plating, an inoculum was prepared from the same garden soil sample collected in California that served for in situ cultivation. Approximately 90 colonies were picked randomly from the traps and the Petri dishes containing either agar or gellan gum. Of the ∼ 90 randomly picked colonies, 69 isolates recovered from the agar trap and 81 isolates from the gellan gum traps were actinobacteria (Table 1). The diversity of isolates obtained from the gellan gum trap was higher compared to the agar trap. Representatives from 11 actinobacterial genera were isolated from gellan, and most of the isolates formed mycelia. Strains of Streptomyces, together with Agromyces, Cellulomonas and Cellulosimicrobium represented the dominant groups among the isolates. Based on 16S rRNA gene sequence data, these isolates are closely related to at least eight streptomycetes Table 1 Number of actinobacterial strains isolated from garden soil sample Genus Agar Gellan trap gum trap Actinoplanes 3 Agromyces Arthrobacter Cellulomonas Cellulosimicrobium Dactylosporangium 1 Frigoribacterium Kineosporia Kitasatospora 9 Kribbella Microbacterium Micromonospora Microsphaera Mycobacterium Nocardioides Nonomuraea Oerskovia Promicromonospora Rhodococcus Streptomyces 56 Streptosporangium Subtercola Terrabacter Williamsia Total number 69 Non-actinobacterial strains Actinobacterial 69 strains 3 11 Actinomycetes CN gellan agar gum medium 3 4 5 CN agar medium 11 19 14 1 1 4 2 3 4 2 2 4 1 1 1 1 4 4 18 1 9 1 1 7 20 8 1 81 1 60 35 87 46 63 38 81 25 41 25 species, two different Agromyces and three different Cellulomonas species (Table 2). All isolates from Cellulosimicrobium were closely related to the same species — C. cellulans with 98.8–100% 16S rRNA gene sequence identity. Additionally, isolates from 8 other genera were obtained, including relatives of Actinoplanes, Kribbella, Nocardioides and Promicromonospora (Tables 1 and 2). Among the isolates from the agar trap, the majority of the isolates belong to the genus Streptomyces (Table 1). In addition, nine Kitasatospora strains, two different Actinoplanes, and Dactylosporangium closely related to D. aurantiacum were isolated (Table 2). The overlap between the two different traps (agar and gellan gum) was small; only relatives of Actinoplanes italicus and several Streptomyces strains were shared between the traps (Table 2), indicating that agar and gellan gum may select for different actinobacterial strains. Within the strains isolated from direct plating the majority were not Actinobacteria. Among the actinobacterial isolates most of the strains belong to Arthrobacter, Microbacterium and Streptomyces, with several additional strains present as single isolates (Table 1). The isolates from the plates appeared to contain relatively more non-filamentous actinobacterial species as compared to the traps, for example Arthrobacter, Microbacterium, Nakamurella, Subtercola, Terrabacter, Williamsia etc. (Tables 1 and 2). Some 261 E. Gavrish et al. / Journal of Microbiological Methods 72 (2008) 257–262 Table 2 List of the strains isolated from traps and plates and their closest relatives based on 16S rRNA gene partial sequences Closest match Accession number % similarity Agar trap Actinoplanes derwentensis Actinoplanes italicus “Actinoplanes pallidoaurantiacus" Agromyces sp. Agromyces ulmi Arthrobacter bergerei Arthrobacter globiformis Arthrobacter histidinolovorans Arthrobacter sp. Cellulomonas fimi Cellulomonas humilata Cellulomonas sp. Cellulosimicrobium cellulans Dactylosporangium aurantiacum Frigoribacterium sp. Kineosporia rhizophila Kitasatospora nipponensis Kribbella sp. Microbacterium imperiale Microbacterium sp. Microbacterium sp. Microbacterium sp. Microbacterium terregens Microbacterium testaceum “Micromonospora lacustris" Micromonospora sp. Nakamurella multipartita Mycobacterium madagascariense Nocardioides fulvus Nocardioides plantarum Nocardioides sp. Nonomuraea turkmeniaca Oerskovia paurometabola Promicromonospora sukumoe Rhodococcus opacus Rhodococcus sp. Streptomyces clavuligerus Streptomyces coelicolor Streptomyces laceyi Streptomyces lateritius Streptomyces neyagawaensis “Streptomyces sacchari" Streptomyces sp. Streptomyces sp. Streptomyces sp. Streptomyces sp. Streptomyces sp. Streptomyces sp. Streptomyces sp. Streptomyces sp. “Streptosporangium brasiliense" Subtercola pratensis Terrabacter tumescens Williamsia muralis Other non-actinobacterial strains AB036999 AB037005 AJ277578 99. 5 99.8–100 99.7 1 2 AJ252586 AY427830 AJ609632 AY167856 AF501358 AY238502 X83803 X82449 AF060791 AB116667 D85480 99.7–100 98.6 98.6–98.8 99.5–99.7 99.8–100 99.8 99.5–99.8 99.5–100 99.3 98.8–100 99.8 AJ297441 AB003933 AY442263 AY253866 AB042081 AY040877 AB042083 AB042070 AB004721 X77445 X92622 AY221486 Y08541 X55600 96.3 99.5 99.0–100 99.8 99.3–99.5 99.5 100 99.7 100 100 100 99.8 99.8 98.3 AF005017 X69973 AY423719 AF277201 AJ314851 AJ272024 AF095715 AB023374 AB045869 AL356612 AY094367 AY277559 AJ399493 AF306660 AY237556 AF101414 AY114179 AJ001433 AF060793 AF112169 AY465202 AY465216 X89937 AJ310412 AF005023 Y17384 99.8 99.8 96.5 99.8 100 99.3–99.5 100 99.3 99.7 100 98.8–99.0 99.1–100 99.5–100 99.7 99.5–99.7 99.7–100 99.8 99.8 99.8 98.5–99.1 98.6 99.8–100 99.8 100 98.6 99.3 gellan gum trap AI agar CN gellan CN agar 1 2 10 1 3 1 3 1 2 2 2 7 1 1 5 13 1 14 1 1 1 9 3 1 1 3 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 4 4 1 6 1 2 2 5 14 1 14 3 1 1 2 4 3 1 1 1 16 1 1 1 3 1 1 1 7 1 8 filamentous actinobacteria such as Streptosporangium, Micromonospora and Nonomuraea strains were only found on plates but not in the traps. The overlap between the plates and 9 2 3 1 1 1 1 1 35 46 Sequenced strain Accession number CATR-197 CATR-14 CATR-7 EF600021 EF600022 EF600020 Act-77 CATR-150 SM-42 SM-8 SM-31 SM-63 CATR-124 CATR-118 CATR-192 CATR-146 CATR-61 EF600006 EF600007 EF601722 EF599994 EF601723 EU407479 EF600010 EF600009 EF599996 EF599997 EF600019 CATR-116 SM-48 CATR-80 CATR-178 G-78 SM-55 G-47 SM-47 Act-60 Act-103 G-71 G-74 G-96 Act-64 EF600008 EF600013 EF599993 EF600025 EF599999 EF600001 EF600000 EF600002 EF600003 EF599998 EF600018 EF600017 EF600014 EF599976 CATR-139 G-51 CATR-114 Act-73 CATR-180 CATR-188 G-50 G-81 SM-23 CATR-54 Act-100 CATR-149 G-22 CATR-9 CATR-60 CATR-4 SM-10 CATR-40 Act-99 CATR-13 CATR-41 CATR-86 Act-98 Act-78 SM-59 Act-84 EF600026 EF600024 EF600023 EF600016 EF600011 EF599995 EF599977 EF599978 EF599990 EF599981 EF599986 EF599992 EF599988 EF599985 EF599987 EF599989 EF599991 EF599979 EF599980 EF599982 EF599984 EF599983 EF600015 EF600005 EF600012 EF599975 38 the traps was small. Only relatives of Agromyces, Microbacterium and several Streptomyces strains were shared between the plates and traps (Table 2). 262 E. Gavrish et al. / Journal of Microbiological Methods 72 (2008) 257–262 Table 3 List of non-streptomycetes mycelium-forming actinomycetes, isolated by traps from soil samples collected in Massachusetts, USA Strain Accession number Closest relative Accession number % similarity Soil sample MS-2 (2) a MS-8 (3) MS-G2 MS-13 MS-E4 MS-18 MS-15 MS-6 (2) MS-5 MS-17 MS-10 (4) PS-1 (3) PS-3-2 (5) PS-37 (4) EF599956 EF599960 EF599974 EF599961 EF599972 EF599965 EF599963 EF599958 EF599957 EF599964 EF599962 EF600027 EF600030 EF600029 Actinoplanes utahensis Actinoplanes italicus “Actinoplanes roseosporangius" Amycolatopsis keratiniphila “Catellatospora coxensis" Kribbella koreensis Lentzea albidocapillata Lentzea flaviverrucosa Lentzea sp. Nocardioides albus Nocardioides luteus Catenulispora sp. Neo15 Kitasatospora kifunense Streptacidiphilus sp. Aac-32 AJ277574 AB037005 AJ277583 AJ508238 AB200232 AY253865 X84321 AF183957 DQ008601 AF005004 AF005007 AJ865860 U93322 AB180775 99–100 98–100 99 96 100 99 100 98–99 100 100 100 100 99–100 99–100 Garden soil Garden soil Garden soil Garden soil Garden soil Garden soil Garden soil Garden soil Garden soil Garden soil Garden soil Pine soil Pine soil Pine soil a The numbers in parentheses are the numbers of sequenced strains. We also tested the ability of the trap to capture various actinomycetes from two additional soil samples. We used a garden soil sample collected in Massachusetts and stored at room temperature for more than one year, and a pine soil sample also collected in Massachusetts (the sample had pH 3.9–4 and very high fungal diversity). The traps were made and maintained the same way as described above. Only traps with 0.2 μm pore-size bottom filter were used in the experiment. The vast majority of the microorganisms captured in the trap were mycelium-forming actinobacteria with a very small number of other nonactinobacterial isolates. Different species of Streptomyces were dominant among the isolates. In addition several rare and unusual actinobacterial strains were isolated, including Streptacidiphilus, Catellatospora, Lentzea and the recently described new genus Catenulispora (Busti et al., 2006), that were not captured in the parallel experiment by Petri plate cultivation (Table 3). From this first evaluation of the trap for in situ cultivation, we conclude that: • The trap allows for specific capture of filamentous actinobacteria; • Decreasing the filter pore size to 0.2 μm effectively excludes fungi and allows for specific capture of actinomycetes, even in the absence of antifungal agents; • The majority of the organisms captured in the trap are actinomycetes, including representatives of unusual and rare groups such as Dactilosporangium, Catellatospora, Catenulispora, Lentzea, and Streptacidiphilus; • Gellan gum favors captured of more diverse actinobacterial groups compared to agar; • The trap favors filamentous actinobacteria compared to the Petri dish, which produces more non-filamentous isolates. Soil is rich in actinomycetes species, and any method, including the trap, probably captures the most abundant ones from a given environment. At the same time, the trap has the potential of selectively enriching actinomycetes, and could probably do this in an environment relatively poor in these organisms, where rare species are likely to be found. Our data show that the trap for in situ actinomycetes cultivation is a promising technique for gaining access to interesting isolates that are not recovered by conventional Petri dish cultivation. Acknowledgement We are grateful to Dr. Eric Stewart for assistance with the manuscript. This work was supported by grants NIH AI059489-01 and DOE DE-FG02-04ER63782. References Baltz, R.H., 2006. Marcel Faber Roundtable: is our antibiotic pipeline unproductive because of starvation, constipation or lack of inspiration? J. Ind. Microbiol. Biotech. 33, 507–513. Busti, E., Cavaletti, L., Monciardini, P., Schumann, P., Rohde, M., Sosio, M., Donadio, S., 2006. Catenulispora acidiphila gen. nov., sp. nov., a novel mycelium-forming actinomycete and proposal of Catenulisporaceae fam. Nov. Int. J. Syst. Evol. Microbiol. 56, 1741–1746. Demain, A.L., Fang, A., 2000. The natural functions of secondary metabolites. In: Fiechter, I.A. (Ed.), History of Modern Biotechnology. Springer, Berlin, pp. 1–39. Hirsch, C.F., Christensen, D.L., 1983. Novel method for selective isolation of actinomycetes. Appl. Environ. Microbiol. 46, 925–929. Hopwood, D.A., 2006. Soil to genomics: the Streptomyces chromosome. Annu. Rev. Genet. 40, 1–23. Kaeberlein, T., Lewis, K., Epstein, S.S., 2002. Isolating “uncultivable” microorganisms in pure culture in a simulated natural environment. Science 296, 1127–1129. Ludwig, W., Strunk, O., Westram, R., Richter, L., Meier, H., Yadhukumar, Buchner, A., Lai, T., Steppi, S., Jobb, G., Forster, W., Brettske, I., Gerber, S., Ginhart, A.W., Gross, O., Grumann, S., Hermann, S., Jost, R., Konig, A., Liss, T., Lussmann, R., May, M., Nonhoff, B., Reichel, B., Strehlow, R., Stamatakis, A., Stuckmann, N., Vilbig, A., Lenke, M., Ludwig, T., Bode, A., Schleifer, K.H., 2004. ARB: a software environment for sequence data. Nucleic Acids Res. 32, 1363–1371. Osburne, M.S., Grossman, T.H., August, P.R., MacNeil, I.A., 2000. Tapping into microbial diversity for natural products drug discovery. ASM News 66, 411–417. Polsinelli, M., Mazza, P.G., 1984. Use of membrane filters for selective isolation of actinomycetes from soil. FEMS Microbiol. Lett. 22, 79–83. Stach, J.E., Maldonado, L.A., Ward, A.C., Goodfellow, M., Bull, A.T., 2003. New primers for the class Actinobacteria: application to marine and terrestrial environments. Environ. Microbiol. 5, 828–841.