2020. 2. 20. 09:20ㆍ카테고리 없음
. 1.2k Downloads.AbstractThe rise of antimicrobial resistance necessitates the discovery and/or production of novel antibiotics. Isolated strains of Paenibacillus alvei were previously shown to exhibit antimicrobial activity against a number of pathogens, such as E. Coli, Salmonella, and methicillin-resistant Staphylococcus aureus (MRSA). The responsible antimicrobial compounds were isolated from these Paenibacillus strains and a combination of low and high resolution mass spectrometry with multiple-stage tandem mass spectrometry was used for identification.
A group of closely related cyclic lipopeptides was identified, differing primarily by fatty acid chain length and one of two possible amino acid substitutions. Variation in the fatty acid length resulted in mass differences of 14 Da and yielded groups of related MS n spectra. Despite the inherent complexity of MS/MS spectra of cyclic compounds, straightforward analysis of these spectra was accomplished by determining differences in complementary product ion series between compounds that differ in molecular weight by 14 Da. The primary peptide sequence assignment was confirmed through genome mining; the combination of these analytical tools represents a workflow that can be used for the identification of complex antibiotics.
The compounds also share amino acid sequence similarity to a previously identified broad-spectrum antibiotic isolated from Paenibacillus. The presence of such a wide distribution of related compounds produced by the same organism represents a novel class of broad-spectrum antibiotic compounds. As the rate of antibiotic resistance increases, there is a growing need for the discovery of novel classes of antibiotics ,. Of particular interest is biocontrol of crop production for food safety applications, where pathogens, including fungal and bacterial growth, can be minimized or prevented by using a bacterial strain or compounds produced by bacteria that are native to the environment and exhibit antibiotic properties ,.
Previously, two naturally-occurring gram-positive bacteria identified as Paenibacillus alvei were isolated from plants native to the Virginia Eastern Shore tomato growing region ,. Both strains showed broad-spectrum antimicrobial activity against gram-negative and gram-positive foodborne pathogens. However, the products responsible for this activity are unknown.Identification of bacterially-produced antibiotics can be challenging because of a range of molecular diversity. Paenibacillus strains produce several classes of antibiotic compounds. For example, P. Polymyxa produces polymyxins, which are cyclic, contain the amino acid diaminobutyric acid, and have an acylated side chain; these are typically used as a last resort against gram-negative bacteria ,.
Another class of compounds is lantibiotics; these antibiotics are ribosomally synthesized and can have a number of post-translational modifications ,. In another investigation, P. Alvei co-produced two different peptide antibiotics that were antagonistic against gram-positive or -negative pathogens. Fusaricidins are cyclic lipopeptides that inhibit fungal growth. Recently, a broad-spectrum antibiotic was discovered from Paenibacillus that is also a cyclic lipopeptide. Furthermore, non-ribosomal peptides can contain diverse amino acids , such as ornithine, which is present in antifungal compounds and cyclic antibiotics. To further complicate classification of these antibiotic compounds, more than one class of compounds can be co-produced in a single strain of Paenibacillus, and the presence of multiple closely related structures can be common within a given class of compounds.
Such complexity makes isolation and identification of the bioactive compound or compounds challenging.Tandem mass spectrometry offers the capability to detect and sequence a wide range of compounds as well as differentiate between closely related forms of the same compound within a given bacterial strain. However, interpretation of the fragmentation spectra of cyclic peptides is nontrivial because of the initial ring opening event occurring at a distribution of sites within a given ion population, resulting in different fragment ion series within the same spectrum. A number of studies have aimed to reduce or handle the complexity of tandem mass spectra of cyclic peptides. Recently, this has been accomplished through database searching , , mass spectral interpretation tools , , or the development of automated data analysis approaches. Searching against genome databases is promising but cannot be implemented when genome data is unavailable.
The development of data analysis approaches to analyze complicated MS data is beneficial and will hopefully lead to the discovery and identification of novel compounds. Other approaches have included using specific MS techniques to aid in the identification of primary sequence assignments either by using a combination of different types of fragmentation data or by sequential fragmentation to reduce the spectral complexity produced from multiple ring openings.Here, we present the detection and identification of broad-spectrum antimicrobial compounds produced by an environmentally isolated Paenibacillus alvei strain from plants native in the U.S. To the Virginia Eastern Shore tomato growing region ,. Compounds were first fractionated by off-line ultra-performance liquid chromatography (UPLC), and the bioactivity of each fraction was assessed against gram-negative and -positive bacteria. Multiple analytical tools were used to identify these purified antibiotic compounds. A combination of matrix-assisted laser desorption/ionization (MALDI) mass spectrometry (MS) and high resolution MS was used to evaluate the complexity of molecular species in the bioactive fractions. A distribution of closely related cyclic lipopeptide compounds was identified with tandem MS, differing primarily by fatty acid chain length and one of two possible amino acid substitutions; the primary peptide sequence was confirmed through gene mining.
With the developed workflow, a complex mixture of multiple related molecular species was identified from a single strain of Paenibacillus, which represents a novel class of antibiotic compounds. Alvei bacterial strains A6-6i and TS-15 were previously isolated from the environment and showed antibiotic properties against a number of pathogens such as Salmonella, E. Coli, and methicillin-resistant Staphylococcus aureus (MRSA) ; however, the responsible antibiotic compounds were unknown.
Assessment of the class and mechanism of antibiotic(s) produced by these Paenibacillus strains required identification of the bioactive compound(s). In this work, we used a combination of fraction collection, tandem mass spectrometry, and genome mining to characterize a series of related compounds that comprise a novel class of cyclic antibiotic peptides with fatty acids attached to the N-terminus of the amino acid ornithine. These peptides primarily differed by the length of the fatty acid and amino acid substitutions at two possible sites. The combination of these analytical tools enabled the identification of compounds illustrated in Scheme. To isolate bioactive compounds, bacterial cells were pooled, extracted, and separated by fraction collection. Seven of the 1-min fractions collected during a 60-min gradient exhibited bioactivity against both gram-negative E.
Coli and gram-positive MRSA (Figure ). Subsequent MALDI-TOF MS analysis revealed a number of compounds that were present in each of the seven bioactive fractions, as shown in Figure. The MALDI spectra provided a view of the full complement of compounds from each bioactive fraction within a single spectrum where clusters of these compounds differed by 14 Da, indicated by red arrows. The compound with a molecular weight of 1623 will be referred to as the primary compound (ion indicated with an asterisk in Figure ), although multiple variants of similar abundances are present (Figure ).
Figure 1Strain TS-15 1-minute bioactive fractions against gram-positive and -negative bacteria and resulting MALDI-TOF MS spectra from the bioactive fractions. ( a) A 10 μL volume from the 1-min fractions of P. Alvei strain TS-15 was spotted on a lawn of 10 6 cells of Escherichia coli O157:H7 strain EDL933 (left) and methicillin-resistant Staphylococcus aureus strain #12 (right). After incubation at 35 ± 2°C for 24 h, the antimicrobial activity exhibited by the 1-min fractions was observed as a clear zone of inhibition. (b) MALDI-TOF MS results of fractions with observed bioactivity. Red arrows indicate a mass difference of 14 Da, which correspond to a molecular difference of CH 2; an asterisk indicates MW 1623, which is designated as the primary sequence; blue arrows show two examples of a mass difference of 2 and 16 Da, which correspond to the other molecular variants.
The MALDI-TOF MS/MS analyses of these molecular species revealed similar fragmentation patterns, which confirmed that the compounds that differ by 14 Da were related. As illustrated in Figure, comparison of MS/MS spectra of the primary compound (MW 1623) and the compound of MW 1637, which differ in molecular weight by 14 Da, revealed a series of product ions that shared the same mass and a second product ion series that differed by 14 Da, suggesting that the mass discrepancy between compounds was localized to one region of the molecule. This enabled identification of complementary product ion pairs, with one direction corresponding to the product ion series retaining the region of the molecule that contained the 14 Da mass-shift and the other direction corresponding to product ions that were identical between the two peptides (Figure ). Manual de novo sequencing resulted in a partial amino acid sequence, yielding a putative sequence assignment. Figure 2MALDI-TOF MS/MS characterization of antibiotic compounds. ( a ) MALDI-TOF MS/MS comparison of two compounds that differ by 14 Da in molecular weight. Mass-to-charge ratios in red with an asterisk indicate an observed mass difference of 14 Da in the comparison between the two spectra.
(b) Partial sequence information from MALDI-TOF MS/MS spectrum of MW 1623 elucidated by manual de novo sequencingAnalysis of MALDI-TOF MS/MS spectra only revealed partial sequence tags. To improve sequence coverage, individual fractions were infused and analyzed with the Orbitrap Elite, allowing for the collection of MS n data. Again, pairs of MS n spectra were analyzed and complementary ion pairs were identified based on the presence of 14 Da mass differences. The combined MS n data allowed a complete amino acid sequence to be determined, which is illustrated in Scheme.
This amino acid sequence was similar to a previously identified cyclic compound isolated from a different Paenibacillus strain, which also showed broad-spectrum activity against MRSA and E. Coli , ; the structure of that compound, designated as paenibacterin, is shown in Supplementary Figure S. The location of the molecular differences between this compound and the compounds described herein are highlighted by asterisks in Scheme.While linearizing the molecule through alkaline hydrolysis simplified de novo sequencing of paenibacterin , Guo et al. Benefited from a strain of Paenibacillus that produced a single dominant cyclic antibiotic.
However, the TS-15 strain presented here produces a wide distribution of related cyclic compounds. Efforts to linearize these compounds further complicated analysis by the additional presence of ions with 18 and 22 Da mass differences, attributable to the linearization of the molecule and presence of sodium adducts, respectively. This necessitated the development of a workflow for the identification of cyclic peptides without linearization.
By determining which series of product ions did or did not contain the molecular component that results in the 14 Da difference, de novo sequencing by MS n analysis was more straightforward. This was particularly critical because the compounds were cyclic and resulting MS n spectra can be difficult to interpret because of multiple ring opening events occurring at a distribution of sites. An example is shown in the MS 3 spectrum in Figure where the ring opens at different amino acid positions, yielding a number of different sequence series within the same spectrum; thus, de novo sequencing of the primary sequence of cyclic peptides can be challenging. However, by using the described approach, the assignment of product ions and the identification of the primary sequence and sequence variants were accomplished without linearizing the molecule. The cumulative ion assignments for the primary amino acid sequence can be found in in Figure.
Subsequent analysis with UPLC coupled to high resolution MS provided accurate mass data, which confirmed that there were actually three predominant compound series, with each series containing groups of compounds that differ by 14.02 Da. The most pronounced differences between the three observed series were either a decrease of 15.99 Da or an increase of 1.98 Da in mass compared with the primary compound series; examples are designated with blue arrows in Figure. A comprehensive list of these compounds and their accurate mass molecular weights can be found in Table. These are designated as F, Y, and Y, –CH 2+O in the table and throughout the figures; F and Y correspond to phenylalanine or tyrosine at position 6 in Scheme and –CH 2+O corresponds to a molecular difference in the fatty acid chain. Two compound series in the bioactive fractions differed from each other by 15.99 Da and exhibited MS 2 spectra with all the compounds in one series yielding a product ion at m/z 657 2+ whereas the other series generates a product at m/z 649 2+ (Figure ). Figure illustrates the MS n spectra of three precursor ions that differed by 14 Da, all of which generated an MS 2 product ion at m/z 657 2+. Because these were conserved product ions within a precursor series that included compounds that differed in molecular weight by 14 Da, we were able to conclude that the region of the compound that yields product ions 657 2+ or 649 2+ does not contain the fatty acid.
The MS 3 spectra of 657 2+ were consistent with one another, confirming that the region of the peptide that generated this sequence was conserved between these compounds (Figure ). The MS 3 spectra from the ion series that generated an MS 2 product ion at 649 2+ were similar to the 657 2+ MS 3 spectra, except for a series of product ions that differed by 16 Da (red masses with asterisks in Figure ). This enabled the distinction between tyrosine and phenylalanine at position 6 (Scheme ), designated as Y and F in the tables and figures, respectively; these amino acids differ in molecular weight by 16 Da. Assignments for the MS/MS spectrum for a peptide containing Phe are illustrated in Figure, where a direct comparison can be observed between the peptides containing Phe and Tyr at position 6. Figure 5MS 2 and resulting MS 3 spectra. (a) Three series of compounds within the class of antibiotics differ by an amino acid or a difference in their attached fatty acid. Compounds that contain a Tyr at position 6 have m/z 657 2+ as a consistent product ion in their resulting MS/MS spectra, whereas Phe at position 6 results in m/z 649 2+.
Red m/z values with an asterisk indicate a mass difference of 16 Da between the MS 3 spectra, which is the mass difference between Phe and Tyr. The mass difference between MW 1623 and MW 1625 corresponds to one less CH 2 group and an additional oxygen in the attached fatty acid (–CH 2+O). (b) Three representative MS 3 spectra demonstrate sequence similarity between compounds of molecular weights that differ by 14 Da and their corresponding complementary ion pair MS 3 spectra. MS 2 spectra are outlined in blue and MS 3 spectra are outlined in orange, with product ions selected for MS 3 highlighted in orange Identification of the Attached Fatty Acid. The MS 2 spectra of the series of compounds that contain a Tyr at position 6 were dominated by product ion 657 2+ and its complementary ion pair (Figure ). While 657 2+ was conserved within the Tyr compound series, its complementary ion contained the same 14 Da mass shift as its precursor (i.e., m/z 311, 325, and 339 in Figure ).
The same trend was also present for the Phe ion series. A list of the multiple complementary ions for 657 2+ and 649 2+ are listed in Table. When these complementary ions were dissociated (examples shown in Figure ), a loss of ornithine was observed.
Subtracting the cyclized peptide sequence mass from the mass of the entire compound yields the mass attributed to an attached fatty acid; molecular formula generation of these masses yield the molecular formulae of the different fatty acid variants (Table and Scheme ). These fatty acids are similar to what was observed in Guo et al., although the lengths of the carbon chains differ and both the TS-15 and A6-6i strains presented in the current work exhibit a greater variability in chain length and composition. Table 2Selected Corresponding Complementary Ion Pairs and Their Resulting MS n Analysis. (A) The Compounds Shared a Similar Product Ion ( m/z 649 2+ or 657 2+) and its Corresponding Complementary Ion Pair Increased by 14 Da with a Molecular Weight Increase of 14 Da. (B) Selected Product Ions from (A) were Dissociated Further (MS 3), which Show the Loss of Ornithine. MS 2 Product Ions Marked with an Asterisk also Exhibited Product Ion m/z 129.102 in MS 3 Spectra, Indicating the Presence of Lys. Subtracting the Mass of the Cyclized Peptide Without the Fatty Acid from the Mass of the Entire Compound Yielded the Molecular Formula of the Attached Fatty Acid.
As mentioned previously, three major series of compounds were determined: two series containing a tyrosine at position 6 and one containing phenylalanine. The two compound series containing Tyr differed by a molecular weight of 1.979 Da. Similar to the MS spectral analysis methodology shown in Figure, these compound series also had similar MS 2 fragmentation patterns, where some product ion masses are conserved and others differ by 1.979 Da (Supplementary Figure S ). This mass difference corresponded to one less CH 2 and an additional oxygen (labeled as –CH 2+O) in the attached fatty acid compared to the tyrosine molecular series. Multiple Compounds with the Same Molecular Weight. It was also observed that more than one of the complementary ion pairs were occasionally present within the same MS 2 spectrum in the infusion experiment data.
This corresponded to two compounds of the same precursor mass being fragmented within the same isolation window. There were multiple examples in the UPLC and NanoLC/MS data that showed several eluting peaks for entities with the same mass (example shown in Figure ). The MS 2 spectra of the ions in each of these chromatographic peaks showed small differences in the fragmentation pattern and, thus, the primary sequences of these respective peptides. Nearly identical amino acid sequences were confirmed for compounds with an identical molecular weight (1579): a compound with Lys at position 7 and a compound with ornithine at position 7 with an additional CH 2 in the attached fatty acid (Lys and ornithine differ by a CH 2 in their side chains). Specifically, the product ion at m/z 649 2+ corresponds to Lys at position 7 and m/z 642 2+ corresponds to ornithine at position 7. The two chromatographic peaks with Lys at position 7 may indicate a structural difference in the attached fatty acid resulting in the observed difference in retention time (Figure ).
Figure 6Different MS/MS spectra (highlighted in blue) for three compounds of the same molecular weight indicate Lys or Orn at Position 7 in Scheme and potential diversity in the structure of the attached fatty acid attributable to distinct chromatographic peaks shown in the extracted ion chromatogram (EIC)A similar substitution was also found at position 1. Some of the subsequent MS 3 analyses of the fatty acid-containing fragment ions ( m/z 325 1+ and 339 1+ in Figure ) indicated that Lys can also be present at position 1 rather than ornithine ( m/z 129 1+ in MS 3 spectra in Figure ). A Lys at position 1 and a decrease of CH 2 in the attached fatty acid resulted in identical molecular weights for each compound. Genome Mining to Confirm Compound Identification. The sequence shown in Scheme was confirmed by genome mining for nonribosomal peptide synthesis. Many pharmacologically important peptides in bacteria are synthesized by nonribosomal peptide synthetases (NRPS). NRPS machinery is composed of modular multi-domain enzymes that act as an assembly line to incorporate each amino acid monomer by one module.
A typical module in an NRPS contains an adenylation (A) domain, which possesses a conserved binding pocket for the recruitment of amino acid monomers that are to be incorporated into the final peptide product. A single contig of 6784766 bp (G + C content, 46.69%) representing the complete chromosome for P. Alvei strain TS-15 was generated.
A total of six NRPSs were identified from the annotated genome, with one corresponding to the compounds characterized in this work. The responsible gene cluster, defined here as Pa-NRPS1, was present on a 49-kb DNA region and contained 13 modules (Table ) corresponding to the 13 amino acids in the microbially-synthesized compounds. A combination of different analytical techniques enabled the detection and identification of a class of cyclic antibiotic compounds.
By analyzing similar MS/MS spectra of compounds that differ by 14 Da, sequence tags can be determined despite the difficulties inherent to cyclic MS/MS interpretation. The sequences elucidated through MS/MS and MS n data are supported by gene mining data, confirming the sequences of the bacterially produced chemical products. This analysis strategy could be readily applied for the identification of novel antibiotics and antibiotic mixtures.
Molecular diversity is apparent in this newly identified compound class; the identified compounds appear to be cyclic with an attached fatty acid of varying lengths (indicated by the molecular weights that differ by 14 Da) and contain different amino acids compared with a previously published similar compound. The molecular weights of the theoretical formulas from the deduced sequence and fatty acid variants were calculated (Scheme and Table ) and correspond to observed values (Table ). Further characterization of these compounds will yield a better understanding of the mechanism responsible for antimicrobial activity. Whatever the mechanism, these compounds potentially represent important future weapons in the control of multi-drug-resistant bacterial pathogens, and understanding their chemical and biological properties is essential to this end.
Q: WHAT IS THE FOCUS OF YOUR LAB’S RESEARCH?A: Almost exclusively high throughput lipid profiling of samples from large epidemiological studies, where we study gene-lifestyle interactions. The method works with plasma samples as well as dried blood spots. The method is also applied to small-scale studies of specific disease groups, dietary interventions and model systems such as yeast. The TriVersa NanoMate is also used to study the lipid composition of tissues analyzed by LESA. Q: WHY DID YOU INCORPORATE THE TriVersaNanoMate INTO YOUR LABORATORY?A: The TriVersa NanoMate is essential to efficiently analyze large-scale studies. It offers a really robust method for high throughput studies with minimal carryover. Q: WHO WOULD YOU RECOMMEND TO PURCHASE THE TriVersa NanoMate?A: I recommend the TriVersa NanoMate to laboratories with a large number of samples requiring a robust and reliable delivery system.
The TriVersa NanoMate eliminates typical nanoelectrospray ionization challenges. Q: DO YOU HAVE ANY PUBLICATIONS OR PRESENTATIONS USING THE TriVersa NanoMate?A: Koulman, A.; Prentice, P.; Wong, M.C.Y.; Matthews, L.; Bond, N.J.; Eiden, M.; Griffin, J.L.; Dunger, D.B., The development and validation of a fast and robust dried blood spot based lipid profiling method to study infant metabolism, Metabolomics 2014. Gadegaard, W.
Advion Triversa Nanomate Manual 2016
Carruthers & K. BurgessAuthentic historic manuscripts fetch high sums, but establishing their authenticity is challenging, relies on a host of stylistic clues and requires expert knowledge. High resolution mass spectrometry has not, until now, been applied to guide the authentication of historic manuscripts. Robert Burns is a well-known Scottish poet, whose fame, and the eponymous ‘Burns Night’ are celebrated world-wide. Authenticity of his works is complicated by the ‘industrial’ production of fakes by Alexander Smith in the 1890s, many of which were of good quality and capable of fooling experts.
This study represents the first analysis of the inks and paper used in Burns poetry, in a minimally destructive manner that could find application in many areas. Applying direct infusion mass spectrometry to a panel of selected authenticated Burns and Smith manuscripts, we have produced a Support Vector Machine classifier that distinguishes Burns from Smith with a 0.77 AUC. Using contemporary recipes for inks, we were also able to match features of each to the inks used to produce some of Burns’ original manuscripts. We anticipate the method and classifier having broad application in authentication of manuscripts, and our analysis of contemporary inks to provide insights into the production of written works of art. Ryan, David Nei, Boone M. Prentice, Kristie L. Rose, Richard M.
Caprioli, Jeffrey M. SpragginsRobotic liquid surface extractions can be used to interrogate discrete regions of tissue to provide protein identifications with high throughput, accuracy, and robustness. The direct coupling of tissue surface extractions and liquid chromatography, offers a new and effective approach to provide spatial proteomics data in an imaging experiment.Tissue extractions were completed using the ® (Advion, Inc., Ithaca, NY, USA) modified to include a glass capillary (LESA Plus) for improved spatial resolution and online integration with LC-based experiments.
36 Scanned images of thaw-mounted samples were uploaded to the ChipSoft Software (Advion, Inc.) to allow histological regions of interest to be selected for analysis. Lieke Lamont, Mark Baumert, Nina Ogrinc Potočnik, Mark Allen, Rob Vreefken, Ron M. Heeren, and Tiffany PortaDirect analysis by mass spectrometry (imaging) has become increasingly deployed in preclinical and clinical research due to its rapid and accurate readouts.
However, when it comes to biomarker discovery or histopathological diagnostics, more sensitive and in-depth profi ling from localized areas is required. We developed a comprehensive, fully automated online platform for high-resolution liquid extraction surface analysis (HR-LESA) followed by micro−liquid chromatography (LC) separation and a data-independent acquisition strategy for untargeted and low abundant analyte identifi cation directly from tissue sections. Applied to tissue sections of rat pituitary, the platform demonstrated improved spatial resolution, allowing sample areas as small as 400 μ m to be studied, a major advantage over conventional LESA.The LESA extraction was performed using the automated ® The LESA extraction was controlled by a beta version of the LESA Plus software (Advion, UK). ABSTRACT: Large-scale metabolic profiling requires the development of novel economical high-throughput analytical methods to facilitate characterization of systemic metabolic variation in population phenotypes. We report a fit-forpurpose direct infusion nanoelectrospray high-resolution mass spectrometry (DI-nESI-HRMS) method with time-of-flight detection for rapid targeted parallel analysis of over 40 urinary metabolites.
The newly developed 2 min infusion method requires.