Figures
Abstract
In most environments many microorganisms live in close vicinity and can interact in various ways. Recent studies suggest that bacteria are able to sense and respond to the presence of neighbouring bacteria in the environment and alter their response accordingly. This ability might be an important strategy in complex habitats such as soils, with great implications for shaping the microbial community structure. Here, we used a sand microcosm approach to investigate how Pseudomonas fluorescens Pf0-1 responds to the presence of monocultures or mixtures of two phylogenetically different bacteria, a Gram-negative (Pedobacter sp. V48) and a Gram-positive (Bacillus sp. V102) under two nutrient conditions. Results revealed that under both nutrient poor and nutrient rich conditions confrontation with the Gram-positive Bacillus sp. V102 strain led to significant lower cell numbers of Pseudomonas fluorescens Pf0-1, whereas confrontation with the Gram-negative Pedobacter sp. V48 strain did not affect the growth of Pseudomonas fluorescens Pf0-1. However, when Pseudomonas fluorescens Pf0-1 was confronted with the mixture of both strains, no significant effect on the growth of Pseudomonas fluorescens Pf0-1 was observed. Quantitative real-time PCR data showed up-regulation of genes involved in the production of a broad-spectrum antibiotic in Pseudomonas fluorescens Pf0-1 when confronted with Pedobacter sp. V48, but not in the presence of Bacillus sp. V102. The results provide evidence that the performance of bacteria in soil depends strongly on the identity of neighbouring bacteria and that inter-specific interactions are an important factor in determining microbial community structure.
Citation: Tyc O, Wolf AB, Garbeva P (2015) The Effect of Phylogenetically Different Bacteria on the Fitness of Pseudomonas fluorescens in Sand Microcosms. PLoS ONE 10(3): e0119838. https://doi.org/10.1371/journal.pone.0119838
Academic Editor: Dawn Arnold, University of the West of England, UNITED KINGDOM
Received: October 29, 2014; Accepted: January 19, 2015; Published: March 16, 2015
This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication
Data Availability: All relevant data are within the paper or the supporting information.
Funding: This work is supported by the Netherlands Organization for Scientific Research (NWO) MEERVOUD personal grant issued to Paolina Garbeva (836.09.004).
Competing interests: The authors have declared that no competing interests exist.
Introduction
Culture-independent technologies have given us insight in the tremendous phylogenetic and functional diversity of microbial communities [1,2]. Recently, the role of interactions between the members of microbial communities and how these shape community composition and dynamics is receiving increasing interest [3–6]. Both theoretical models and empirical studies are used to explain the coexistence of competing microbial species and consequently microbial community assembly [5,7].
In soil and in the rhizosphere, many microbial species live in close vicinity and interact with each other in various ways ranging from competition to cooperation [5,8,9]. Bacteria can recognise cues from their environment to modulate behaviour in order to increase their chance of survival.
Using recently developed techniques (NanoDESI and MALDI-TOF imaging mass spectrometry) Traxler and co-authors indicated the importance of interspecific interactions for triggering the production of different secondary metabolites in a single strain [10]. Recent studies in our group also indicate that bacteria are respond differently to the presence of different microbial species [11–13]. Studies on behavior and the transcriptional responses of the soil bacterium Pseudomonas fluorescens Pf0-1 on nutrient-poor agar in confrontation with taxonomically different bacterial species revealed significant differences in the responses of Pseudomonas fluorescens Pf0-1 to different bacteria. In particular, the expression of genes involved in signal transduction and antibiotic production was strongly affected by the identity of the interacting strains [12].
So far the response of Pseudomonas fluorescens Pf0-1 to phylogenetically different bacteria has only been studied during one-to-one confrontations on agar media [12,13]. However, these conditions are very artificial compared to the situation in the natural soil environment, which is a heterogeneous and complex habitat consisting of aggregated particles with huge bacterial diversity [2,14,15]. It is thus plausible that bacteria can sense more easily the presence of neighbours in their vicinity on an agar plate than in soil. Furthermore, in natural environments bacteria are likely to encounter several different competitors at the same time or in sequential events [6]. In the present study, we made a first attempt to study bacterial interactions in soil-like systems. To this end we investigated the interaction between Pseudomonas fluorescens Pf0-1 with monocultures and mixtures of Pedobacter sp. V48 and Bacillus sp. V102 in sand microcosms under two different nutrient conditions. We hypothesised that both nutrient conditions and the identity of the competitor would have an effect on the performance of Pseudomonas fluorescens Pf0-1.
Material and Methods
Bacterial strains and growth conditions
Three different bacterial species, Pseudomonas fluorescens Pf0-1 (γ-Proteobacteria) [16], Pedobacter sp. V48 (Sphingobacteria) and Bacillus sp. V102 (Bacilli) [11] were used in this study (Table 1). The strains were pre-cultured from frozen −80°C glycerol stocks on 1/10th TSB agar (5.0 gL−1 NaCl (Merck), 1.0 gL−1 KH2PO4; 3 gL−1 Tryptic Soy Broth (OXOID); 20 gL−1 Agar (Merck), pH 6.5) [13] for three days at 20°C.
Microcosm setup
Microcosms were established in 100 mL glass vials with a plastic screw cap lid (S2 Fig.) containing sterile acid washed sea sand with pore size fractions ranging from 0.075 to 0.425 mm (Honeywell Specialty Chemicals Seelze GmbH, Germany). The amount of sand was either 25 g (Microcosms supplemented with 1.5 mL 1/10th strength Tryptic Soy Broth (nutrient rich media) (5.0 gL−1 NaCl (Merck), 1.0 gL−1 KH2PO4; 3 gL−1 Tryptic Soy Broth (OXOID)) or 30 g (Microcosms supplemented with 1.5 mL nutrient poor media (5.0 gL−1 NaCl (Merck), 1.0 gL−1 KH2PO4; 0.1 gL− (NH4)2SO4; 0.5 gL−1 Tryptic Soy Broth (OXOID)). The sand was weighed directly into the glass vials and afterwards sterilized by autoclaving for 20 minutes. The sterilized microcosms were dried overnight in a 60°C oven prior to inoculation. All treatments were performed in triplicates over a time of 6 days. A detailed overview of all treatments and controls is given in Table 2.
Microcosm inoculation
Sand microcosms were inoculated with either each strain as monoculture, pairwise combinations, or with all three strains together (Table 2). To inoculate the microcosms a single colony of the respective strain was transferred into 10 ml of 1/10th TSB and grown overnight at 20°C, 220 rpm to an optical density (OD600) of: ∼0.700 (Pseudomonas fluorescens Pf0-1), ∼0.600 (Pedobacter sp. V48) and ∼0.650 (Bacillus sp. V102).The bacterial strains were diluted in an nutrient rich or nutrient poor inoculation master mix to a density of ∼1 * 105 CFU/mL. Each microcosm was inoculated with a volume of 1.5 mL of the respective inoculum master mix in the middle of the sterilized sand and mixed well.
To verify bacterial cell numbers in the inoculum, dilution plating was done in duplicates on selective agar plates (Pseudomonas fluorescens Pf0-1: 1/10th TSBA plates supplemented with 100 μg/mL Ampicillin, Pedobacter sp. V48: 1/10th TSBA plates supplemented with 50 μg/mL Kanamycin, Bacillus sp. V102: samples were pasteurized by heat treatment for 10 min. @ 80°C).
Bacterial enumeration
The growth of the three bacterial strains in the different treatments was tracked by plate counting of all culturable cells (Pseudomonas fluorescens Pf0-1 and Pedobacter sp. V48) or by counting of spores and heat-stable cells (Bacillus sp. V102) (Table 1). The enumeration was performed as follows: after one and six days of incubation a sterilized stainless steel spoon was used for sampling by mingling the sand by a full clock- and one counter- clockwise turn. After mixing 1 g sand was taken from the center of each microcosm and transferred into a 15 mL Greiner tube. A volume of 10 ml 10 mM phosphate buffer (pH 6.5) was added and the tubes were shaken in a rotary shaker at 350 rpm for 30 minutes at 20°C. Subsequently, serial dilutions were prepared and plated in triplicates on selective media (antibiotics used are indicated in Table 1). For the enumeration of the Bacillus sp. V102, samples were pasteurized by heating the tubes to 80°C for 10 min in a pre-warmed heating block. The plates were incubated for two to four days at 20°C and the CFUs of the respective strains were determined.
RNA extraction and quantitative real time PCR
The expression of gene cluster Pfl01_3463-3466, which is involved in the production of a broad-spectrum antibiotic [12] was quantified via quantitative real time PCR. Total RNA was extracted at day 6 from nutrient rich microcosms (1/10th TSB) containing Pseudomonas fluorescens Pf0-1 as monoculture or in interaction as follows: the double volume (2mL) of RNA protect Bacteria Reagent (QIAGEN cat# 76506) was added to 1 g sand sample and centrifuged at 10,000 x g for 10 min (Sigma 3K-14 centrifuge, SIGMA Laborzentrifugen GmbH, Germany). The supernatant was discarded and the pellets were stored at −80°C until further analysis. Total RNA was extracted with the MO-BIO PowerSoil Total RNA Isolation Kit (MO-BIO cat# 12866-25) according to the manufacturer’s protocol. The RNA extracts were treated with the TURBO DNA free Kit from AMBION (cat# 1907) according to the manufacturer’s protocol to remove any remaining DNA. The RNA concentration and quality was checked on a NanoDrop Spectrophotometer (Isogen Life Science, IJssestein, the Netherlands). cDNA was synthesized from the extracted RNA with random hexamer primers from Invitrogen (cat# 48190-011) by using reverse transcriptase of the Fermentas RevertAid Premium First Strand cDNA Synthesis Kit (Fermentas cat#K1651) according to manufacturer’s protocol. The concentration and quality of the cDNA was determined using a NanoDrop spectrophotometer by measuring the A260/A280 ratio and samples were run on a 1.5% agarose gel in 0.5% TBE buffer to check size and integrity of the synthesized cDNA.
The selected gene cluster was targeted with primer combination 3463F835 (5’ ATTTTTACGCGGTCTACGC) and 3463R1036 (5’TGATCAGGTTGCTGTTTCAGG) [12] amplifying 202bp from gene Pfl01_3463 encoding the two branched-chain alpha-keto acid dehydrogenase E1 component. From each treatment, 50 ng cDNA was subjected to quantitative RT- PCR using SYBR Green PCR master mix (Applied Biosystem, Warrington, UK). Quantitative RT- PCR was performed on a Corbett Research Rotor- Gene 3000 thermal cycler (Westburg, Leusden, the Netherlands) with the following settings: initial cycle 95°C for 15min, followed by 40 cycles of 95°C for 15 sec, 56°C for 50 sec and 72°C for 50 sec. All analysis was performed in triplicate. Five standard curves (9.5 ng/μl, 0.95 ng/μl, 0.095 ng/μl, 0.0095 ng/μl and 0.00095 ng/μl) were established. Gene expression data was analysed with a post-hoc LSD- test and differences between the means of data of different Pseudomonas interactions were considered to be statistically different at p ≤ 0.05.
Malthusian parameter
As an estimate for fitness of the Pseudomonas fluorescens Pf0-1 as monoculture or in competition with the two other strains was calculated by applying the Malthusian parameter (M) growth model [17,18]. The Malthusian parameter was calculated for both monocultures and mixed cultures by comparing the number of Pseudomonas fluorescens Pf0-1 individuals at an initial time (t0), N0, to the number of Pseudomonas fluorescens Pf0-1 individuals at a future time (tN): M = ln (Nt/N0) / t.
Statistical analysis
Statistical analyses of the cell counts were performed with IBM SPSS Statistics 20 (IBM, Somers, NY, USA) using one-way ANOVA and post-hoc TUKEY LSD test. Significant differences between the controls (monocultures of the respective bacterial strain) and the treatments are marked with an asterisk (p≤ 0.05).
Results and Discussion
In the present study, we investigated how the interactions between phylogenetically different bacterial strains affect the growth of Pseudomonas fluorescens Pf0-1 in sand microcosms under two different nutrient conditions. Our interests were particularly focused on Pseudomonas fluorescens Pf0-1, as our previous research had shown that Pseudomonas fluorescens Pf0-1 responded differently (behaviour and gene expression profile) to phylogenetically different bacteria on nutrient poor agar [11].
The growth of Pseudomonas fluorescens Pf0-1 in microcosms supplemented with either nutrient rich or nutrient poor growth media are presented in Fig. 1A and 1B. Bacterial enumeration revealed that all tested bacterial strains used in this study were able to grow in the sand microcosms although with lower numbers under nutrient poor conditions (Fig. 2A and B). In microcosms supplemented with nutrient rich media Pseudomonas fluorescens Pf0-1 reached approximately 5.5 * 105 cells/g of sand as a monoculture, while in microcosms supplemented with nutrient poor media reached only 9.2 * 104 cells/g of sand. The Bacillus sp. V102 cell numbers in monocultures reached 3.6 * 104 cells/g of sand in nutrient rich microcosms and 7 * 103 cells/g of sand in nutrient poor microcosms. The cell counts of Pedobacter sp. V48 as monoculture were approximately 2.3 * 106 cells/g of sand in nutrient rich and 2.8 * 104 cells/g of sand in nutrient poor microcosms (Fig. 2A and 2B).
Significant differences between the numbers of Pf0-1 in monoculture and in mixed cultures are indicated with an asterisk (p≤0.05). Error bars are indicating standard deviation (SD) between the triplicates. Abbreviations: Pf0-1: Pseudomonas fluorescens Pf0-1, PB: Pedobacter sp. V48, BAC: Bacillus sp. V102.
Significant differences between the CFUs in monoculture and in mixed cultures are indicated with an asterisk (p≤0.05). Error bars are indicating standard deviation (SD) between the triplicates. Abbreviations: Pf0-1: Pseudomonas fluorescens Pf0-1, BAC: Bacillus sp. V102, PB: Pedobacter sp. V48.
The growth of Pseudomonas fluorescens Pf0-1 was negatively affected when confronted with the Gram-positive Bacillus sp. V102 strain resulting in significantly lower cell counts at day 6 in nutrient rich microcosms (p = 0.012) and at day 1 in nutrient poor microcosms (p = 0.008). When co-cultivated with Bacillus sp. V102 Pseudomonas fluorescens Pf0-1 reached a maximum of approximately 4.8 * 105 cells/g of sand (Fig. 1A and B). Strong reduction of Pseudomonas fluorescens Pf0-1 growth during confrontation with Bacillus sp. V102 was observed previously on nutrient-poor agar even without direct cell-cell contact [12]. When co-cultivated with the Gram-negative Pedobacter sp. V48 strain, no significant effect on the growth of Pseudomonas fluorescens Pf0-1 was observed at day 6 (p = 0.988), whereas there was a significant reduction at day 1 in nutrient-poor microcosms (p = 0.000). Based on the cell enumeration we applied the Malthusian growth model (population growth) as an estimate for fitness (S1 Fig.). This revealed that the population growth of Pseudomonas fluorescens Pf0-1 was significantly negative affected only during co-cultivation with Bacillus sp. V102 on both day 1 and day 6 (p = 0.026 and p = 0.014).
The observed difference in response of strain Pseudomonas fluorescens Pf0-1 to co-cultivated bacteria was not due to the difference in bacterial growth as both Pedobacter sp. V48 and Bacillus sp. V102 were growing in the sand microcosms with Pedobacter sp. V48 reaching higher cell counts per gram of sand than Bacillus sp. V102 (Fig. 2A and 2B). However, when co-cultivated with both Bacillus sp. V102 and Pedobacter sp. V48 simultaneously, there was no significant effect on the growth of Pseudomonas fluorescens Pf0-1 in both nutrient rich (p = 0.650) and nutrient poor microcosms (p = 0.995) (Fig. 1A, 1B). From the inter-specific interactions investigated in the present study, it is clear that Bacillus sp. V102 acts as “bad” neighbour that can negatively affect the fitness of Pseudomonas fluorescens Pf0-1 as compared to the “good” neighbour Pedobacter sp. V48 that did not show any negative effect on the growth of Pseudomonas fluorescens Pf0-1. However, when co-cultivated with both strains simultaneously, Pseudomonas fluorescens Pf0-1 growth was better than when confronted only with Bacillus sp. V102 (Fig. 3).
The effects of the respective interaction on the fitness of Pseudomonas fluorescens Pf0-1 are indicated by the number of the colored circles. Each full circle represents 1.0 * 105 CFU/mL (nutrient rich media) or 1.0 * 104 CFU/mL (nutrient poor media).
From previous studies in our group it is known that Pseudomonas fluorescens Pf0-1 can be triggered to produce a broad-spectrum antibiotic when co-cultivated with Pedobacter sp. V48, but not in co-cultivation with Bacillus sp. V102 [12,13]. It was hypothesized that this facultative- rather than the constitutive production of antibiotic compound represent a cost-effective strategy, as the antibiotic compound is only produced in situation when it is needed [19]. It is plausible that in a more complex habitat, the production of a broad-spectrum antibiotic triggered by Pedobacter sp. V48 gives Pseudomonas fluorescens Pf0-1a advantage when confronted with phylogenetically different strains simultaneously. To confirm that the observed results are related to antibiotic production, we performed quantitative RT-PCR with primers targeting genes Pfl01_3463 known to be involved in the production of a broad-spectrum antimicrobial compound [12]. Results revealed that indeed genes Pfl01_3463 were highly expressed at day 6 in the microcosms where Pseudomonas fluorescens Pf0-1 was interacting with Pedobacter sp. V48 (p = 0.014). Gene expression was slightly higher in treatments were Pseudomonas fluorescens Pf0-1 was confronted with both Pedobacter sp. V48 and Bacillus sp. V102, although not significantly (p = 0.750) (Fig. 4). Unfortunately, due to the low cell number in the microcosms supplemented with nutrient poor growth media, our attempts to extract good quality and quantity of RNA for cDNA synthesis and quantitative RT-PCR failed.
Inter-specific interactions may trigger the production of antimicrobial compounds in complex microbial communities where this so-called chemical warfare may offer comparative advantage for the producing strains [5,6]. A recent study showed that interspecific interactions between soil bacteria can have a major impact on antimicrobial compound production with effects in both directions, i.e. induction or suppression of antimicrobial compound production [20].
Error bars are indicating standard deviation (SD) between the triplicates. Significant differences between the qRT-PCR based gene expression by Pf0-1 in monocultures and mixed cultures is indicated by an asterisk (p≤0.05).
In soil and in the rhizosphere environment Pseudomonas species coexist with many other bacterial species and compete for the same nutrient resources [15,21–23]. The ability to cope with the presence of a range of competing microbial species is essential for growth and survival in soil ecosystems and the performance of soil bacteria may strongly depend on the neighbouring competitors.
Overall, our data suggests that the performance of Pseudomonas fluorescens Pf0-1 in sand microcosms depends greatly on the presence and identity of neighbouring microorganisms. Although Pseudomonas fluorescens Pf0-1 cell counts were lower in the nutrient poor sand microcosms than in the nutrient rich microcosms, similar growth patterns were observed in both experiments. This indicates that, contrary to our initial hypothesis, nutrient levels did not have a strong effect on multispecies interactions and on the ability of Pseudomonas fluorescens Pf0-1 to respond to different bacteria. It is well known that under different nutrient conditions bacteria often produce different secondary metabolites [24–26] and hence influence microbial interactions in different ways.
This work demonstrates that interspecific interactions can play an important role in soil and may influence microbial performance and consequently shape the composition of microbial communities.
Supporting Information
S1 Fig. Malthusian parameter calculated for the time interval from day 0 to day 1 and for the time interval from day 0 to day 6 (nutrient rich media) representing the fitness of Pseudomonas fluorescens Pf0-1 in the four different microcosm treatments.
Error bars are indicating standard deviation (SD) between the triplicates. Significant differences are indicated by an asterisk (p≤0.05).
https://doi.org/10.1371/journal.pone.0119838.s001
(TIF)
S2 Fig. Example of a sand micorocosm used in this study.
https://doi.org/10.1371/journal.pone.0119838.s002
(TIF)
Acknowledgments
This work is supported by the Netherlands Organization for Scientific Research (NWO) MEERVOUD personal grant issued to Paolina Garbeva (836.09.004). The authors want to thank Cristina Martinez Romera for her help during the experiment. We thank Professor Wietse de Boer for critical reading of this manuscript and his valuable and constructive comments. This is publication 5773 of the NIOO-KNAW.
Author Contributions
Conceived and designed the experiments: OT ABW PG. Performed the experiments: OT ABW. Analyzed the data: OT ABW. Contributed reagents/materials/analysis tools: OT ABW. Wrote the paper: OT ABW PG.
References
- 1. Uroz S, Buee M, Murat C, Frey-Klett P, Martin F (2010) Pyrosequencing reveals a contrasted bacterial diversity between oak rhizosphere and surrounding soil. Environmental Microbiology Reports 2: 281–288. pmid:23766079
- 2. Gans J, Wolinsky M, Dunbar J (2005) Computational improvements reveal great bacterial diversity and high metal toxicity in soil. Science 309: 1387–1390. pmid:16123304
- 3. Cornforth DM, Foster KR (2013) Competition sensing: the social side of bacterial stress responses. Nature Reviews Microbiology 11: 285–293. pmid:23456045
- 4. Mitri S, Foster KR (2013) The genotypic view of social interactions in microbial communities. Annual Review of Genetics 47: 247–273. pmid:24016192
- 5. Foster Kevin R, Bell T (2012) Competition, not cooperation, dominates interactions among culturable microbial species. Current biology 22: 1845–1850. pmid:22959348
- 6. Hibbing ME, Fuqua C, Parsek MR, Peterson SB (2010) Bacterial competition: surviving and thriving in the microbial jungle. Nature Reviews Microbiology 8: 15–25. pmid:19946288
- 7. D'Onofrio A, Crawford JM, Stewart EJ, Witt K, Gavrish E, Epstein S, et al. (2010) Siderophores from neighboring organisms promote the growth of uncultured bacteria. Chemistry & Biology 17: 254–264.
- 8. Czaran T, Hoekstra RF (2009) Microbial communication, cooperation and cheating: quorum sensing drives the evolution of cooperation in bacteria. Plos One 4: e6655. pmid:19684853
- 9. Allen B, Nowak MA (2013) Cooperation and the fate of microbial societies. PLoS Biol 11: e1001549. pmid:23637573
- 10. Traxler MF, Watrous JD, Alexandrov T, Dorrestein PC, Kolter R (2013) Interspecies Interactions Stimulate Diversification of the Streptomyces coelicolor Secreted Metabolome. Mbio 4.
- 11. de Boer W, Wagenaar A-M, Gunnewiek PJAK, van Veen JA (2007) In vitro suppression of fungi caused by combinations of apparently non-antagonistic soil bacteria. Fems Microbiology Ecology 59: 177–185. pmid:17233750
- 12. Garbeva P, Silby MW, Raaijmakers JM, Levy SB, de Boer W (2011) Transcriptional and antagonistic responses of Pseudomonas fluorescens Pf0-1 to phylogenetically different bacterial competitors. ISME Journal 5: 973–985. pmid:21228890
- 13. Garbeva P, de Boer W (2009) Inter-specific Interactions Between Carbon-limited Soil Bacteria Affect Behavior and Gene Expression. Microbial Ecology 58: 36–46. pmid:19267150
- 14. Torsvik V, Goksoyr J, Daae FL (1990) High diversity in DNA of soil bacteria. Appl Environ Microbiol 56: 782–787. pmid:2317046
- 15. Demoling F, Figueroa D, Baath E (2007) Comparison of factors limiting bacterial growth in different soils. Soil Biology & Biochemistry 39: 2485–2495.
- 16. Compeau G, Alachi BJ, Platsouka E, Levy SB (1988) Survival of rifampin-resistant mutants of Pseudomonas fluorescens and Pseudomonas putida in soil systems. Applied and Environmental Microbiology 54: 2432–2438. pmid:3144244
- 17. van den Berg HA, Orlov MV, Kiselëv YN (2008) The Malthusian parameter in microbial ecology and evolution: An optimal control treatment. Computational Mathematics and Modeling 19: 406–428.
- 18. Vasi F, Travisano M, Lenski RE (1994) Long-Term Experimental Evolution in Escherichia coli. II. Changes in Life-History Traits During Adaptation to a Seasonal Environment. The American Naturalist 144: 432–456.
- 19. Garbeva P, Tyc O, Remus-Emsermann MNP, van der Wal A, Vos M, Silby M, et al. (2011) No Apparent Costs for Facultative Antibiotic Production by the Soil Bacterium Pseudomonas fluorescens Pf0-1. Plos One 6.
- 20. Tyc O, van den Berg M, Gerards S, van Veen JA, Raaijmakers JM, de Boer W, et al. (2014) Impact of interspecific interactions on antimicrobial activity among soil bacteria. Frontiers in Microbiology 5: 567. pmid:25389421
- 21. Jousset A, Schmid B, Scheu S, Eisenhauer N (2011) Genotypic richness and dissimilarity opposingly affect ecosystem functioning. Ecology Letters 14: 537–545. pmid:21435139
- 22. Becker J, Eisenhauer N, Scheu S, Jousset A (2012) Increasing antagonistic interactions cause bacterial communities to collapse at high diversity. Ecology Letters 15: 468–474. pmid:22394557
- 23. Inglis RF, Brown SP, Buckling A (2012) Spite Versus Cheats: Competition among Social Strategies Shapes Virulence in Pseudomonas Aeruginosa. Evolution 66: 3472–3484. pmid:23106711
- 24. Sanchez S, Chavez A, Forero A, Garcia-Huante Y, Romero A, Sanchez M, et al. (2010) Carbon source regulation of antibiotic production. Journal of Antibiotics 63: 442–459. pmid:20664603
- 25. van Wezel GP, McDowall KJ (2011) The regulation of the secondary metabolism of Streptomyces: new links and experimental advances. Natural Product Reports 28: 1311–1333. pmid:21611665
- 26. Garbeva P, Hordijk C, Gerards S, de Boer W (2014) Volatiles produced by the mycophagous soil bacterium Collimonas. Fems Microbiology Ecology 87: 639–649. pmid:24329759