http://orcid.org/0000-0003-1518-2727
Figures
Abstract
The oxyvinylglycine 4-formylaminooxyvinylglycine (FVG) arrests the germination of weedy grasses and inhibits the growth of the bacterial plant pathogen Erwinia amylovora. Both biological and analytical methods have previously been used to detect the presence of FVG in crude and extracted culture filtrates of several Pseudomonas fluorescens strains. Although a combination of these techniques is adequate to detect FVG, none is amenable to high-throughput analysis. Likewise, filtrates often contain complex metabolite mixtures that prevent the detection of FVG using established chromatographic techniques. Here, we report the development of a new method that directly detects FVG in crude filtrates using laser ablation electrospray ionization-mass spectrometry (LAESI-MS). This approach overcomes limitations with our existing methodology and allows for the rapid analysis of complex crude culture filtrates. To validate the utility of the LAESI-MS method, we examined crude filtrates from Pantoea ananatis BRT175 and found that this strain also produces FVG. These findings are consistent with the antimicrobial activity of P. ananatis BRT175 and indicate that the spectrum of bacteria that produce FVG stretches beyond rhizosphere-associated Pseudomonas fluorescens.
Citation: Okrent RA, Trippe KM, Manning VA, Walsh CM (2018) Detection of 4-formylaminooxyvinylglycine in culture filtrates of Pseudomonas fluorescens WH6 and Pantoea ananatis BRT175 by laser ablation electrospray ionization-mass spectrometry. PLoS ONE 13(7): e0200481. https://doi.org/10.1371/journal.pone.0200481
Editor: Vijai Gupta, Tallinn University of Technology, ESTONIA
Received: April 7, 2017; Accepted: June 27, 2018; Published: July 10, 2018
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 and its Supporting Information files.
Funding: Research in the laboratory of K.T. is supported by US Department of Agriculture Agricultural Research Service (https://www.ars.usda.gov) Project # 2072-21410-004-00D to KT. This work was also supported by the Agriculture and Food Research Initiative Competitive Grant No. 2012-67012-19868 from the USDA National Institute of Food and Agriculture (https://nifa.usda.gov/) to R.O. During data collection, C.W. was employed by Protea BioSciences, Inc. Protea Biosciences, Inc. provided support in the form of salaries for C.W. but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.
Competing interests: The authors have read the journal’s policy and the authors of this manuscript have the following competing interests: CW was a paid scientist for Protea Biosciences, Inc. This does not alter our adherence to PLOS ONE policies on sharing data and materials.
Introduction
Oxyvinylglycines are a small class of biologically-active molecules produced by certain strains of bacteria [1]. Examples of oxyvinylglycines include aminoethoxyvinylglycine from Streptomyces sp. [2], 4-methoxyvinylglycine from Pseudomonas aeruginosa [3], rhizobitoxine from Bradyrhizobium elkanii [4], and 4-formylaminooxyvinylglycine (FVG) from P. fluorescens [5]. FVG is a potent secondary metabolite that arrests the germination of weedy grasses [6] and demonstrates antibiotic activity against the bacterial plant pathogen Erwinia amylovora [7].
The lability and hydrophilicity of oxyvinylglycines complicate efforts to precisely detect FVG or FVG-like compounds in crude culture filtrates. The routine determination of an FVG phenotype therefore requires an ethanol-based extraction of crude culture filtrate [8]. The presence or absence of FVG is subsequently determined by simultaneously examining partially purified filtrates with thin-layer chromatography [9], and two biological assays [6,7]. Simultaneous analyses are required because each method in itself cannot unequivocally identify FVG as the active molecule. For example, we have observed that thin-layer chromatograms of other compounds are remarkably similar to thin-layer chromatograms of FVG (data not shown). In addition, the biological activities of other vinylglycines, including aminoethoxyvinylglycine and 4-methoxyvinylglycine, overlap significantly with FVG [7,10]. Collectively, the extraction and analysis of crude culture filtrate using current methodologies is time-consuming and requires large sample volumes. Because of these challenges, we have sought a high-throughput and precise method to directly detect FVG in crude culture filtrates.
Laser ablation electrospray ionization (LAESI) is a sample introduction technique for mass spectrometry. In the LAESI process, a mid-infrared laser ablates samples to introduce molecules into the gas phase, which then collide with an intersecting electrospray, leading to ionization and detection by a mass spectrometer [11]. LAESI-MS is suitable for analyzing small volumes of aqueous samples, is tolerant to salts, and is very rapid (15 s or less per sample). It can be used for high-throughput screening of complex, native biological samples, as has been demonstrated recently for detection of domoic acid toxin in shellfish homogenates [12]. LAESI-MS has been successfully used to detect metabolites from bacterial colonies on an agar surface [13,14] and in a biofilm [15], but has not yet been reported for detecting metabolites in crude culture filtrate.
To date, FVG has only been detected in strains of P. fluorescens [5]. However, additional bacterial strains are predicted to produce FVG or FVG-like compounds. Pantoea ananatis BRT175 produces a compound that, like P. fluorescens WH6, has activity against E. amylovora. This compound is linked with a biosynthetic gene cluster similar to one associated with FVG production in WH6 (Fig 1) [16–19]. The aims of the current study were to: 1) develop methods for LAESI-MS-based detection of FVG directly in bacterial culture filtrates, and, 2) validate this method by directly analyzing culture filtrates from BRT175 to determine if FVG is produced by this non-pseudomonad.
Gene arrows are colored based on the class of the encoded protein. Gene designations in bold indicate genes that, when mutated, result in a null-FVG phenotype and/or antibiotic activity against Erwinia amylovora. Gray shading indicates homologous regions of the gvg cluster in each strain. TM, transmembrane.
Materials and methods
Preparation of culture filtrates
Bacterial strains are listed in Table 1. Pseudomonas fluorescens and Pantoea ananatis strains were inoculated into either 60 ml or 5 ml of Pseudomonas Minimal Salts Medium (PMS) [6] in a 125-ml Wheaton bottle or 20-ml test tube, respectively. The cultures were grown at 28 °C with shaking for either seven (60-ml cultures) or two days (5-ml cultures). Subsequently, cells were removed from the filtrate by centrifugation (3000 x g for 15 min) and filter sterilized (Millipore GP express Steritop, 0.22-μm pore size (EMD Millipore, Billerica, MA, USA) or 25-mm Pall Acrodisc with 0.22-μm pore size Supor membrane (Pall Corporation, Port Washington, NY, USA), respectively). The dilution series was prepared by concentrating the culture filtrate in a Savant SPD1010 SpeedVac Concentrator (Thermo Scientific, Waltham, MA, USA) without heat (2X), or by diluting (0.3X, 0.1X, 0.03X, 0.01X) culture filtrate with PMS media. As a positive control, a sample enriched in FVG was prepared by extraction with 90% ethanol as described in [9].
Biological assays and thin layer chromatography
The agar diffusion bioassay for anti-microbial activity against E. amylovora was performed as previously described [7]. The areas of the zones of inhibition were measured on triplicate plates for each sample using Able Image Analyser (Mu Labs). Culture filtrates were extracted with 85% ethanol and analyzed by TLC as described in [9]. The semi-quantitative bioassay for germination arrest activity was performed using the standard protocol and scoring system described in [6] for annual bluegrass (Poa annua L.) seeds. Briefly, each well of a 48-well culture plate (Corning Costar 3548) received a 200-uL aliquot of extracted or crude culture filtrate and three surface- sterilized annual bluegrass seeds. For each treatment, three replicate wells were prepared. The plates were sealed with parafilm and incubated for 7 d at 20°C (8 h light, 16 h dark d-1). Seed germination was subsequently scored using the system described by Banowetz [6]. A score between 0 and 4 is assigned to each seed, and nine seeds per treatment are scored. A score of 0 indicates that no visible signs germination are evident under 10X magnification. A score of 4 indicates that the first true leaf is longer than the coleoptile and that roots have elongated. The details of the scoring system between 1 and 4 are described in S1 Fig.
LAESI-MS analysis
LAESI-MS analysis was performed by Protea Biosciences, Inc. Samples of crude culture filtrate, non-inoculated filtrate, or extracted culture filtrate (20 μl) were aliquoted into individual wells of a 96-well plate and analyzed with a LAESI DP-1000 (Protea Biosciences, Morgantown, WV, USA) coupled to a Q Exactive Orbitrap mass spectrometer (Thermo Scientific). In the initial experiments with extracted and crude culture filtrate from WH6 and mutant strains, settings for the LAESI DP-1000 were electrospray (+ 4000 V, 1.25 μl/min flow, 50% methanol, 0.1% acetic acid), 20 pulses per well at 1 Hz laser repetition rate, and laser energy 1000 μJ. A full scan from m/z 50 to 750 was performed for detection of FVG [M+H]+ m/z 161.0556 and [M+Na]+ m/z 183.0374 with an error of 1.1 ppm. MS/MS was performed on the precursor ion m/z 183.037 [M+Na]+ of the sodium adduct of FVG using a Q Exactive Orbitrap mass spectrometer. Conditions were slightly altered for analysis of a dilution series of culture filtrates from WH6 and the comparison of BRT175 to WH6 samples. The conditions of ablation and scanning were optimized for WH6 culture filtrate and a media-only control. In subsequent runs, settings for the LAESI DP-1000 were electrospray (+ 4000 V, 1.0 μl/min flow, 50% methanol, 0.1% acetic acid), 100 pulses per well at 20 Hz laser repetition rate, and laser energy ~600 μJ. The samples were analyzed in selected-ion monitoring mode at a resolution of 140,000 with scans centered on the sodium adduct of FVG, [M+Na]+ m/z 183.037. The average scan signal intensities were extracted on each well using LAESI Bridge software. Measurements were also performed on a second set of biological replicates for comparison.
Results and discussion
FVG-null mutant strains
The genetic basis of FVG production has previously been investigated in P. fluorescens WH6 [17,22]. Recently, Okrent et al. comprehensively examined the importance of each gene within the gvg biosynthetic gene cluster [16], which encodes proteins with regulatory, biosynthetic and transport functions (Fig 1). In that study, deletions in the biosynthetic genes encoding a carbamoyltransferase (gvgF), aminotransferase (gvgH) and formyltransferase (gvgI), led to loss of FVG production [16]. These mutant strains were used as controls in the development of methods for LAESI-MS-based detection of FVG in the current study.
Feasibility of LAESI-MS for detection of FVG
In initial trials of LAESI-MS, the sodium adduct of FVG (m/z 183.0372 with 1.1 ppm mass error) was detected in P. fluorescens WH6 culture filtrates that had been extracted with 90% ethanol. Fragmentation of the precursor ion by MS/MS resulted in characteristic fragments similar to those previously obtained from a low resolution MS/MS of the m/z 183 [M+Na]+ peak (S2 Fig) [5].
The ability of LAESI-MS to detect FVG in crude culture filtrates of P. fluorescens WH6 was subsequently examined. In this experiment, culture filtrates from both wild-type and mutant strains were directly analyzed by LAESI-MS in a 96-well plate format. Using a full scan analysis, a peak at m/z 183.037, the sodium adduct, was detected in filtrates collected from wild-type WH6 (Fig 2). This peak was not detected in a media-only control or in culture filtrates from the mutant strains previously shown to display a null-FVG phenotype [16], WH6-30G (ΔgvgH) and WH6-31G (ΔgvgI) (Fig 3). These results collectively indicate that FVG can be specifically detected in complex samples that have been minimally processed.
An MS spectrum from m/z 50 to 750 is shown with inset (A.) showing the peaks within the range m/z 183.00–183.09 only. The chemical structure of 4-formylaminooxyvinylglycine (FVG) is shown in inset B.
Laser ablation electrospray ionization-mass spectrometry was used to scan crude culture filtrates for the ion corresponding to 4-formylaminooxyvinylglycine (FVG). Each duplicate well of a 96-well plate contained crude culture filtrate from wild-type Pseudomonas fluorescens WH6, a null FVG-mutant strain [WH6-30G (ΔgvgH) or WH6-31G (ΔgvgI)], or non-inoculated filtrate. Data were collected from 20 laser pulses per sample well, and the peak trace corresponds to the extracted ion chromatogram for sodiated FVG, m/z /183.0372.
The sensitivity of LAESI-MS for detection of FVG was investigated by analyzing a dilution series of culture filtrate from WH6 (2X, 1X, 0.30X, 0.10X, 0.03X and 0.01X) by LAESI-MS. These analyses were performed in selected-ion monitoring mode with scans centered on the sodium adduct of FVG as above. FVG was consistently detected by LAESI-MS in dilutions of filtrates to 0.30X (Table 2). While less sensitive than the bioassays, this method is sufficient to detect FVG in bacterial strains that produce FVG. Under the current instrument settings, the signal intensity is not linear with FVG concentration (Table 2). Development of a quantitative assay would require further optimization, such as including an internal standard, adjusting for detection of FVG in both the high and low range, and addressing potential causes of well-to-well variability.
Detection of FVG in Pantoea ananatis BRT175 by LAESI-MS
Pantoea ananatis BRT175 is predicted to produce FVG based on the similarity of its PNP-1 gene cluster to the gvg cluster of P. fluorescens WH6, and due to the common activity of their products (Fig 1) [18,19]. However, analyses of extracted culture filtrate from Pantoea ananatis BRT175 with TLC were ambiguous because several ninhydrin-reactive compounds co-migrated, and obscured the FVG-associated band (S3 Fig). Despite several modifications to our existing protocol (data not shown), it was not possible to distinguish a FVG-associated band from overlapping bands in the chromatogram. Therefore, we analyzed crude culture filtrates from Pantoea ananatis BRT175 by LAESI-MS to determine if P. ananatis BRT175 produces FVG and if that production requires the PNP-1 cluster. Culture filtrates from wild-type BRT175, wild-type WH6 and strains with mutations in the carbamoyltransferase-encoding gene (pnp1D::Tn5 and ΔgvgF) were assayed by LAESI-MS in selected-ion monitoring mode with scans centered on the sodium adduct of FVG. Analysis of the spectra indicated that FVG was detected in the filtrates from both wild-type strains but not in the mutant strains (Table 3). These results are consistent with biological assays performed on these filtrates (Table 3), and confirm both that BRT175 produces FVG and that production of FVG is linked to the PNP-1 cluster.
The capability of BRT175 to produce FVG is unusual within its genus. Walterson et al. conducted a PCR-based survey of 117 strains of Pantoea and failed to detect pnp orthologs in any strain other than BRT175 [18]. Similarly, of the 81 sequenced Pantoea strains in the IMG database of the Joint Genome Institute (as of September 2016), no others had orthologous gene clusters. As noted previously [18], the genome of Pantoea agglomerans Eh318 does contain homologs of a few of the genes within the gvg cluster. However, that strain does not contain homologs of all the genes required for production of FVG, and thus would not be expected to produce FVG.
Based on sequence comparisons, we predict that there are differences in the regulation, transport, and metabolism of FVG in BRT175 compared to WH6. For example, BRT175 lacks orthologs of the ECF sigma factor/anti-sigma factor pair prtIR, which have been shown to regulate FVG production in WH6 [23]. FVG production in BRT175 may instead be regulated through other networks. Similarly, export of FVG may also differ, as BRT175 contains only one of the LysE family exporters known to export FVG in WH6 (the ortholog of gvgJ). Simultaneous mutation of both gvgJ and gvgK is lethal in WH6 [16]. Mutation of pnp1G in BRT175 may also be lethal, as it was the only gene in the PNP-1 cluster in which a Tn5 insertion was not recovered [18]. In addition, the PNP-1 cluster of BRT175 lacks orthologs of gvgD and gvgE, which encode an amidinotransferase and LysE family exporter, respectively. Although gvgD and gvgE are not required for production of FVG in WH6, they are likely involved in the synthesis of an additional product [16] which would not be produced by BRT175. Additional comparative genomic analyses between BRT175 and WH6 may identify other similarities and differences between these bacteria to shed light on oxyvinylglycine production, regulation, and metabolism. The use of LAESI-MS will facilitate that effort and other on-going studies.
Conclusions
In the current study, we demonstrate that LAESI-MS can be used to detect FVG in both extracted and crude culture filtrate and differentiates between samples known to contain FVG from those that lack it. Overall, LAESI-MS based detection of FVG is more rapid and requires substantially smaller sample volumes than previous methodologies, enabling potential use in high-throughput analyses. However, detection of relatively low amounts or quantitative analysis of FVG will require further optimization. We demonstrate the utility of this technique by successfully identifying FVG in crude culture filtrates from Pantoea ananatis BRT175, the first example of FVG production by a non-pseudomonad.
Supporting information
S1 Fig. Visual key to germination-arrest scores for annual bluegrass seeds.
In this semi-quantitative assay, complete germination corresponds to a score of 4.0 and complete arrest corresponds to a score of 0, as described in Banowetz [6]. Figure adapted from Okrent et al. (2017).
https://doi.org/10.1371/journal.pone.0200481.s001
(PDF)
S2 Fig. MS/MS spectrum of fragments from precursor ion m/z 183.037 [M+Na] of 4-formylaminooxyvinylglycine (FVG).
Peaks noted with asterisks correspond to known fragments from the low resolution ESI-MS/MS spectrum from purified FVG [5].
https://doi.org/10.1371/journal.pone.0200481.s002
(PDF)
S3 Fig. Thin layer chromatograms showing separation of metabolites in extracted culture filtrates from Pseudomonas fluorescens WH6 and Pantoea ananatis BRT175.
Culture filtrates were extracted with 85% ethanol, and extracts were applied to silica chromatographic plate, and visualized by staining with ninhydrin.
https://doi.org/10.1371/journal.pone.0200481.s003
(PDF)
Acknowledgments
Disclaimer: The use of trade, firm, or corporation names in this publication is for the information and convenience of the reader. Such use does not constitute an official endorsement or approval by the United States Department of Agriculture or the Agricultural Research Service of any product or service to the exclusion of others that may be suitable.
We thank John Stavrinides for providing the wild-type and mutant strains of Pantoea ananatis BRT175 and Kerry McPhail for helpful discussion.
References
- 1. Berkowitz DB, Charette BD, Karukurichi KR, McFadden JM. α-Vinylic amino acids: occurrence, asymmetric synthesis, and biochemical mechanisms. Tetrahedron Asymmetry. 3;17: 869–882.
- 2. Pruess DL, Scannell JP, Kellett M, Ax HA, Janecek J, Williams TH, et al. Antimetabolites produced by microorganisms. X. L-2-amino-4-(2-aminoethoxy)-trans-3-butenoic acid. J Antibiot (Tokyo). 1974;27: 229–33.
- 3. Scannell JP, Pruess D, Blount JF, Ax HA, Kellett M, Weiss F, et al. Antimetabolites produced by microorganisms. XII. (S)-alanyl-3-[alpha-(S)-chloro-3-(S)-hydroxy 2-oxo-3-azetidinylmethyl]-(S)-alanine, a new beta-lactam containing natural product. J Antibiot (Tokyo). 1975;28: 1–6.
- 4. Owens LD, Thompson JF, Pitcher R, Williams T. Structure of rhizobitoxine, an antimetabolic enol-ether amino-acid from Rhizobium japonicum. J Chem Soc Chem Commun. 1972; 714.
- 5. McPhail KL, Armstrong DJ, Azevedo MD, Banowetz GM, Mills DI. 4-Formylaminooxyvinylglycine, an herbicidal germination-arrest factor from Pseudomonas rhizosphere bacteria. J Nat Prod. 2010;73: 1853–1857. pmid:20979386
- 6. Banowetz G, Azevedo MD, Armstrong DJ, Halgren AB, Mills DI. Germination-Arrest Factor (GAF): Biological properties of a novel, naturally-occurring herbicide produced by selected isolates of rhizosphere bacteria. Biol Control. 2008;46: 380–390.
- 7. Halgren A, Azevedo M, Mills D, Armstrong D, Thimmaiah M, McPhail K, et al. Selective inhibition of Erwinia amylovora by the herbicidally active germination-arrest factor (GAF) produced by Pseudomonas bacteria. J Appl Microbiol. 2011;111: 949–959. pmid:21726360
- 8. Banowetz GM, Azevedo MD, Armstrong DJ, Mills DI. Germination arrest factor (GAF): Part 2. Physical and chemical properties of a novel, naturally occurring herbicide produced by Pseudomonas fluorescens strain WH6. Biol Control. 2009;50: 103–110.
- 9. Armstrong DJ, Azevedo M, Mills D, Bailey B, Russell B, Groenig A, et al. Germination-Arrest Factor (GAF): 3. Determination that the herbicidal activity of GAF is associated with a ninhydrin-reactive compound and counteracted by selected amino acids. Biol Control. 2009;51: 181–190.
- 10. Lee X, Azevedo MD, Armstrong DJ, Banowetz GM, Reimmann C. The Pseudomonas aeruginosa antimetabolite L-2-amino-4-methoxy-trans-3-butenoic acid inhibits growth of Erwinia amylovora and acts as a seed germination-arrest factor. Environ Microbiol Rep. 2013;5: 83–89. pmid:23757135
- 11. Nemes P, Vertes A. Laser Ablation Electrospray Ionization for atmospheric pressure, in vivo, and imaging mass spectrometry. Anal Chem. 2007;79: 8098–8106. pmid:17900146
- 12. Beach DG, Walsh CM, McCarron P. High-throughput quantitative analysis of domoic acid directly from mussel tissue using Laser Ablation Electrospray Ionization—tandem mass spectrometry. Toxicon. 2014;92: 75–80. pmid:25449096
- 13. Parsiegla G, Shrestha B, Carrière F, Vertes A. Direct analysis of phycobilisomal antenna proteins and metabolites in small cyanobacterial populations by Laser Ablation Electrospray Ionization Mass Spectrometry. Anal Chem. 2012;84: 34–38. pmid:22141353
- 14. Li H, Balan P, Vertes A. Molecular imaging of growth, metabolism, and antibiotic inhibition in bacterial colonies by Laser Ablation Electrospray Ionization Mass Spectrometry. Angew Chem. 2016;128: 15259–15263.
- 15. Dean SN, Walsh C, Goodman H, van Hoek ML. Analysis of mixed biofilm (Staphylococcus aureus and Pseudomonas aeruginosa) by laser ablation electrospray ionization mass spectrometry. Biofouling. 2015;31: 151–161. pmid:25672229
- 16. Okrent RA, Trippe K, Maselko M, Manning V. Functional analysis of a biosynthetic cluster essential for production of 4-formylaminooxyvinylglycine, a germination-arrest factor from Pseudomonas fluorescens WH6. Microbiology. 2017;163: 207–217. pmid:28270265
- 17. Halgren A, Maselko M, Azevedo M, Mills D, Armstrong D, Banowetz G. Genetics of germination-arrest factor (GAF) production by Pseudomonas fluorescens WH6: identification of a gene cluster essential for GAF biosynthesis. Microbiology. 2013;159: 36–45. pmid:23125119
- 18. Walterson AM, Smith DDN, Stavrinides J. Identification of a Pantoea biosynthetic cluster that directs the synthesis of an antimicrobial natural product. PLOS ONE. 2014;9: e96208. pmid:24796857
- 19. Walterson AM, Stavrinides J. Pantoea: insights into a highly versatile and diverse genus within the Enterobacteriaceae. FEMS Microbiol Rev. 2015; pmid:26109597
- 20. Elliott L, Azevedo M, Mueller-Warrant G, Horwath W. Weed control with rhizobacteria. Soil Sci Agrochem Ecol. 1998;33: 3–7.
- 21. Johnson K, Stockwell V, McLaughlin R, Sugar D, Loper J, Roberts R. Effect of antagonistic bacteria on establishment of honey bee-dispersed Erwinia amylovora in pear blossoms and on fire blight control. Phytopathology. 1993;83: 995–1002.
- 22. Kimbrel JA, Givan SA, Halgren AB, Creason AL, Mills DI, Banowetz GM, et al. An improved, high-quality draft genome sequence of the Germination-Arrest Factor-producing Pseudomonas fluorescens WH6. BMC Genomics. 2010;11: 522. pmid:20920191
- 23. Okrent RA, Halgren AB, Azevedo MD, Chang JH, Mills DI, Maselko M, et al. Negative regulation of germination-arrest factor production in Pseudomonas fluorescens WH6 by a putative extracytoplasmic function sigma factor. Microbiology. 2014;160: 2432–2442. pmid:25165126