Content uploaded by Kevin Jace Miranda
Author content
All content in this area was uploaded by Kevin Jace Miranda on Oct 17, 2023
Content may be subject to copyright.
Available via license: CC BY 4.0
Content may be subject to copyright.
Citation: Miranda, K.J.; Jaber, S.;
Atoum, D.; Arjunan, S.; Ebel, R.;
Jaspars, M.; Edrada-Ebel, R.
Pseudomonassin, a New Bioactive
Ribosomally Synthesised and
Post-Translationally Modified
Peptide from Pseudomonas sp. SST3.
Microorganisms 2023,11, 2563.
https://doi.org/10.3390/
microorganisms11102563
Academic Editor: Gary A. Strobel
Received: 21 September 2023
Revised: 9 October 2023
Accepted: 13 October 2023
Published: 15 October 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
microorganisms
Article
Pseudomonassin, a New Bioactive Ribosomally Synthesised
and Post-Translationally Modified Peptide from Pseudomonas
sp. SST3
Kevin Jace Miranda 1, 2, * , Saif Jaber 3, Dana Atoum 3, Subha Arjunan 1, Rainer Ebel 1, Marcel Jaspars 1and
RuAngelie Edrada-Ebel 3
1Marine Biodiscovery Centre, Department of Chemistry, University of Aberdeen, Meston Walk,
Aberdeen AB24 3UE, UK; subha.arjunan@syngenta.com (S.A.); r.ebel@abdn.ac.uk (R.E.);
m.jaspars@abdn.ac.uk (M.J.)
2College of Pharmacy and Graduate School, Adamson University, 900 San Marcelino Street, Ermita,
Manila 1000, Philippines
3Strathclyde Institute of Pharmacy and Biomedical Sciences, University of Strathclyde, John Arbuthnott
Building, 161 Cathedral Street, Glasgow G4 0RE, UK; saif.jaber@strath.ac.uk (S.J.);
dana.atoum@hu.edu.jo (D.A.); ruangelie.edrada-ebel@strath.ac.uk (R.E.-E.)
*Correspondence: kevin.jace.miranda@adamson.edu.ph
Abstract:
Genome mining and metabolomics have become valuable tools in natural products research
to evaluate and identify potential new chemistry from bacteria. In the search for new compounds
from the deep-sea organism, Pseudomonas sp. SST3, from the South Shetland Trough, Antarctica, a
co-cultivation with a second deep-sea Pseudomonas zhaodongensis SST2, was undertaken to isolate
pseudomonassin, a ribosomally synthesised and post-translationally modified peptide (RiPP) that be-
longs to a class of RiPP called lasso peptides. Pseudomonassin was identified using a genome-mining
approach and isolated by means of mass spectrometric guided isolation. Extensive metabolomics
analysis of the co-cultivation of Pseudomonas sp. SST3 and P. zhaodongensis SST2, Pseudomonas sp.
SST3 and Escherichia coli, and P. zhaodongensis SST2 and E. coli were performed using principal
component analysis (PCA) and orthogonal projections to latent structures discriminant analysis
(OPLS-DA), which revealed potential new metabolites in the outlier regions of the co-cultivation,
with other metabolites identified previously from other species of Pseudomonas. The sequence of pseu-
domonassin was completely deduced using high collision dissociation tandem mass spectrometry
(HCD-MS/MS). Preliminary studies on its activity against the pathogenic P. aeruginosa and its biofilm
formation have been assessed and produced a minimum inhibitory concentration (MIC) of 63
µ
g/mL
and 28 µg/mL, respectively.
Keywords:
genome mining; metabolomics; ribosomally synthesised and post-translationally
modified
peptide; Pseudomonas sp. SST3
1. Introduction
Genome mining in natural products research has accelerated the discovery and iso-
lation of new and novel natural products. It has become the core approach for natural
products discovery due to the identification of biosynthetic gene clusters (BGCs) that en-
code for a diverse arrays of secondary metabolites, from polyketide synthases (PKS) and
non-ribosomal peptide synthases (NRPS) to ribosomally synthesised post-translationally
modified peptides (RiPPs) [
1
,
2
]. Genome sequences of a diverse range of organisms such
as bacteria, cyanobacteria, and fungi have become more and more accessible through the
use of public-domain gene sequence repositories and reduced sequencing costs, which
have enabled the investigation of BGCs and hence of the specialised metabolites they
encoded [
3
]. The encoded BGCs in the genome of an organism involved in the biosynthesis
of specialised metabolites are often silent or cryptic under laboratory conditions. This often
Microorganisms 2023,11, 2563. https://doi.org/10.3390/microorganisms11102563 https://www.mdpi.com/journal/microorganisms
Microorganisms 2023,11, 2563 2 of 11
poses difficulties in targeting and isolating new compounds with potential bioactivity [
4
].
A technique that is commonly used to elicit the production of secondary metabolites and
the activation of some gene clusters in bacteria is co-cultivating it with other organisms.
This could be another bacterium or fungus to enhance the production, target metabolites,
express the silent genes, or produce new secondary metabolites from their interactions.
This approach also shows features that are not produced in the single-strain culture, and
it often leads to the elucidation of new metabolic pathways [
5
]. Most of the co-culture
experiments that have revealed new biosynthetic pathways and new secondary metabolites
are between bacteria and fungi due to their competition for nutrients and fermentation
conditions that differ between the strains in the co-culture experiment [
6
]. Bioinformatics
allows us to map the metabolome of organisms from their gene clusters. Several approaches
dealing with the optimisation, evaluation, and isolation of bioactive secondary metabolites
using metabolomics platforms and algorithms have been successfully employed using
different bioinformatics tools [
7
–
9
]. Since metabolites are considered to be the final product
of entire cellular processes, the outcomes of enzymatic processes are also observed in the
study of the metabolome of an organism. Thus, metabolomics became useful in drug
discovery, discovering biomarkers for disease, as well as in the study of plants, bacteria,
and fungi [10,11].
Lasso peptides form a class of RiPPs that have a lariat knot topology in which the tail
is threaded through the macrolactam ring [
12
]. Lasso peptides are classified according to
the disulphide bridges in their topology: Class I has two disulphide bridges; Class II, the
most common type of lasso peptide. has no disulphide bridges; finally, Class III and Class
IV each have one disulphide bridge; the Class III has a disulphide bridge that interlinks
the C-terminus of the tail and the N-terminus of ring and Class IV has a disulphide bridge
on the tail [
13
]. The lasso topology of these peptides confers stability to their structures
and makes them resistant to proteases and thermal degradation [
13
,
14
]. Even though this
type of RiPP is structurally (in terms of the disulphide bridges they have) and functionally
diverse, they still follow a collective biosynthesis in which they are all encoded in the
gene as a precursor peptide, which consists of the N-terminal leader sequence and the C-
terminal core sequence [
15
]. RiPPs have been investigated as a new source of antimicrobial
therapeutics because of their specific activity against pathogens; for example, the antibiotic
microcin J25 from E. coli which is only active against Salmonella and Shigella spp., which
both belong to the Enterobacteriaceae family as E. coli [
16
–
18
]. Their stability in the face of
both thermal and proteolytic degradation also made them promising therapeutics [19,20].
This work aims to elicit and characterise the new RiPP
1
that has been identified
using genome mining and bioinformatics tools, which was identified as a lasso peptide.
The elicitation was carried out by co-cultivating two Pseudomonas spp. isolated from
the same environment in South Shetland Trench. We also performed dereplication and
metabolomics analysis of the extracts using PCA and OPLS-DA and revealed potential new
metabolites, alongside with known tryptophan- or phenylalanine-containing phenolate
siderophores that have been also identified in the genome. The entire sequence of the RiPP
1
predicted from the BGC was confirmed using stepwise HCD-MS/MS, which enables us
to fragment all amino acid residues of the peptide, making this technique an invaluable
tool in RiPPs research.
2. Materials and Methods
2.1. Organism and Fermentation
Deep-sea bacteria P. zhaodongensis SST2 and Pseudomonas sp. SST3, were collected
as part of the PharmaDeep Project in South Shetland Trench (SST), Antarctica [
21
], and
the E. coli K-12 strain was obtained from the laboratory of Dr. Hai Deng. Small-scale
cultures of modified GYM medium containing 4.0 g yeast extract, 10.0 g malt extract, 4.0 g
glucose, 12.0 g of calcium carbonate, distilled water up to 1 L, and pH 7.20 were prepared
by inoculating 50 mL of media with a single colony of the organism and incubating it for
7 days
at 28
◦
C with shaking at 150 rpm. For the monoculture, 50 mL of the medium was
Microorganisms 2023,11, 2563 3 of 11
used. For the co-cultivation, 50 and 25 mL of each medium were used for each organism.
The prepared co-cultivation flasks were as follows: 50 mL Pseudomonas sp. SST3 and 50 mL
P. zhaodongensis SST2, 50 mL Pseudomonas sp. SST3 and 50 mL E. coli, 50 mL P. zhaodongensis
SST2 and 50 mL E. coli, 50 mL Pseudomonas sp. SST3 and 25 mL P. zhaodongensis SST2, and
50 mL
P. zhaodongensis SST2 and 25 mL Pseudomonas sp. SST3. The co-cultivation flasks were
combined on the 6th day and shaken for 24 h prior to extraction. Large-scale fermentation
was conducted on 9 L of medium in which 500 mL of the culture of P. zhaodongensis SST2
culture was placed in twelve 2 L flasks and 250 mL of Pseudomonas sp. SST3 culture was
placed in twelve 1 L flasks and combined together on the 6th day of incubation.
2.2. Extraction and Isolation
Diaion HP-20 (Fisher Scientific UK, Loughborough, UK) (3 g/50 mL) were placed in
each flask to absorb the secondary metabolites produced on the 6th day and shaken for
24 h
.
Culture broths with HP-20 were then filtered using glass wool (Sigma-Aldrich, St. Louis,
MO, USA), washed with water to remove excess salts, and then extracted with methanol
(100 mL
×
3). Successive methanol extracts were combined and concentrated under
reduced pressure. Fractionation was conducted using solid-phase extraction (SPE) chro-
matography C-18 column (Phenomenex Strata) with 25%, 50%, 75%, and 100% methanol
(Fisher Scientific UK). Final purification was conducted using C-18 reverse phase high-
performance liquid chromatography (HPLC) (Agilent, Santa Clara, CA, USA) with 35–100%
95:5 water:methanol (solvent A) and 100% methanol (solvent B) gradient for 35 min.
2.3. HCD-MS/MS Analysis
High-resolution HCD-MS analyses were conducted at the Institute of Medical Sciences,
University of Aberdeen, Forester Hill Campus, using a Q Exactive Plus hybrid quadrupole—
Orbitrap mass spectrometer (Thermo Fisher Scientific, Waltham, MA, USA). The sample
in 0.1% formic acid was infused directly from the syringe pump at 5
µ
L/min to the HESI
source. Instrument settings for MS/MS of 871.46 m/zion (z = 2). Fragmentation NCE was
varied stepwise from 20 to 50. MS/MS was also acquired over a lower scan range (50.0
to 750.0 m/z) for some NCE values. Acquisition time was from 0.2 min to 0.5 min. Direct
infusion positive-mode HCD was applied over the range of 20–50 NCE. In the Thermo
®
Orbitrap instrument, NCE is linearly correlated with the mass-to-charge (m/z) of the
precursor ion. This set-up of the instrument allows for the acquisition of a high-resolution
fragmentation, particularly of peptides below 200 Daltons (Da) and over 2000 Da mass
range, and the data are automatically gathered regardless of the mass of the analyte [22].
2.4. Metabolomics Analysis
Multivariate analysis (MVA) was conducted using SIMCA 15.0.1 (Umetrics, Umeå,
Sweden) software, and we set the parameters for the principal component analysis (PCA)
and orthogonal projections to latent structures discriminant analysis (OPLS-DA) algorithm.
We converted the positive-mode MS data of the extracts using MZMine and Proteowiz-
ard. Data processing from MZMine followed a step-by-step protocol from peak detection,
deconvolution, deisotoping, filtering, alignment, gap filling, the addition of adducts and
complexes for eliminating misassignments of predicted molecular formulae in the algo-
rithm, and the threshold of molecular formula was set at 5 ppm. The processed MZMine
data were then converted into a CSV file and input at SIMCA to generate score plots.
These score plots measure the similarities and differences of the metabolites in the mass
spectrometry data. To further discriminate and characterise the data of the extracts, we go
into the PCA and OPLS-DA analysis and generated graphs.
2.5. Antibacterial Assay
Minimum inhibitory concentration (MIC) was conducted using alamarBlue (Invitro-
gen, Waltham, MA, USA) broth dilution assay [
23
,
24
]. The test organism, P. aeruginosa
(ATCC 27853), was prepared by inoculating a loop of each bacterium in 5 mL of Luria-
Microorganisms 2023,11, 2563 4 of 11
Bertani (LB) broth (GIBCO). The test organism was incubated for 16 h at 37
◦
C to reach the
stationary phase. After 16 h, 100
µ
L of aliquot was transferred to new culture containing
5 mL
LB broth. The new culture was incubated for 6 h at 37
◦
C to reach the log phase. The
assay was performed in 96-well plates (Thermo Fisher UK) by mixing 10
µ
L of prepared
extracts, gentamicin as the positive control, and dimethylsulfoxide (DMSO) (Fisher Scien-
tific UK) as the negative control in the dilution plate with 90
µ
L of LB broth containing
bacteria to give a final concentration of 100
µ
g/mL, and this was incubated at 37
◦
C for
16 h.
A 10
µ
L of 10% alamarBlue solution was added to each well and incubated for another
4 h. Absorbance readings at the 560 nm excitation wavelength and the 590 nm emission
wavelength (Hidex) were conducted to evaluate cell viability [24].
2.6. Planktonic Assay Solution for Biofim Inhibition
Assay plates were prepared in the same manner as the cell viability test with 10
µ
L of
the extract, gentamicin, and 90
µ
L of LB broth containing bacteria, which was incubated
for 16 h at 37
◦
C. After incubation, the wells were emptied and washed twice with
100 µL
of phosphate buffer saline (PBS). Absorbance readings at 600 nm were undertaken to
determine biofilm formation viability and activity by means of turbidity [25,26].
3. Results
3.1. Genome Mining and Prediction of the Structure of Pseusomonassim
A deep-sea bacterium was obtained from South Shetland Trench (SST), Antarctica, as
part of the PharmaDeep Project [
21
]. Whole-genome sequencing revealed that the species
belongs to the Pseudomonas genus, Pseudomonas sp. SST3. Genome mining studies using
antibiotics and Secondary Metabolite Analysis Shell (antiSMASH) [
27
], a bioinformatics
tool that annotates and compares gene clusters from known sequences that encodes for
secondary metaboliutes, showed the annotated lasso peptide biosynthetic gene clusters
(BGCs) consisting of the precursor peptide (pdnA), an ATP-dependent protease similar to
cysteine/transglutamase (pdnB), an ATP-dependent macrolactam synthetase with homol-
ogy to glutamine (Gln) synthase (pdnC), and the ABC transporter (pdnD). The prediction
of the structure of the lasso peptide relies on the sequence of the precursor peptide in the
gene cluster. The precursor peptide allows for the identification of the target sequence of
the mature peptide and its confirmation using mass spectrometry (MS) [
11
]. In our search
for the putative structure of this interesting peptide using a genome mining approach, we
entered the sequence of the precursor peptide of this gene cluster into BLAST [
28
] and
translated the sequence to the RiPPMiner [
29
] platform, which gave three possible putative
structures (Figure S1). Since these kinds of peptides have a motif governing the number
of amino acid residues in both ring and tail [
30
], the structure with the most plausible
number of residues was chosen (Figure 1). The predicted peptide has an accurate mass of
1740.9060 Da.
3.2. Elicitation of the Target RiPP
A single culture of Pseudomonas sp. SST3 and P. zhaodongensis SST2 and the co-
cultivation of Pseudomonas sp. SST3 with P. zhaodongensis SST2, Pseudomonas sp. SST3
with E. coli, and P. zhaodongensis SST2 with E. coli were used for the target RiPP in modified
GYM medium. We evaluated which of the co-cultivations or the monoculture was the best
for the large-scale production of the target RiPP. In order to obtain sufficient yields of the
peptide
1
for potential isolation and full structural characterisation, approaches to stimu-
late increased production of the target compound by the two Pseudomonas strains, SST2
and SST3, co-occurring in the SST, and both containing the respective BCGs for the lasso
peptide, were evaluated. Thus, we developed a co-cultivation protocol of the two strains,
P. zhaodongensis SST2 and Pseudomonas sp. SST3, with each other, as well as the terrestrial
Gamma-proteobacterium E. coli, and also including the relevant controls (Supplementary
Table S1). All cultures were fermented in modified GYM medium as stated above. The
rationale behind this approach was that these types of RiPPs are known to have a narrow
Microorganisms 2023,11, 2563 5 of 11
spectrum of antibacterial activity within the same class. An example for this finding is
microcin J25, isolated from E. coli, which displayed activity against E. coli H157:O7, Shigella,
and Salmonella, all of which belong to the Enterobacteriaceae family [
31
,
32
]. In addition
to the single culture of Pseudomonas sp. SST3 as described above, expression of the target
peptide
1
was detected in the co-cultivation of P. zhaodongensis SST2 and Pseudomonas sp.
SST3 and the co-cultivation of Pseudomonas sp. SST3 and E. coli. Conversely, no production
of the peptide was observed in the single culture of P. zhaodongensis SST2 or in its co-culture
with E. coli. A semiquantitative assessment of the titres was obtained by comparing the
relative intensity of the base peak at m/zof 871.45, corresponding to the doubly charged
ion [M+2H]
2+
of the predicted peptide
1
, and through a metabolomics approach using
SIMCA, multivariate analysis (MVA), such as principal component analysis (PCA), and
orthogonal projections to latent structures discriminant analysis (OPLS-DA).
Microorganisms 2023, 11, x FOR PEER REVIEW 5 of 12
Figure 1. BGC of pseudomonassin (1) showing the four main gene clusters for its biosynthesis and
the sequence of the leader and core peptide.
3.2. Elicitation of the Target RiPP
A single culture of Pseudomonas sp. SST3 and P. zhaodongensis SST2 and the co-culti-
vation of Pseudomonas sp. SST3 with P. zhaodongensis SST2, Pseudomonas sp. SST3 with E.
coli, and P. zhaodongensis SST2 with E. coli were used for the target RiPP in modified GYM
medium. We evaluated which of the co-cultivations or the monoculture was the best for
the large-scale production of the target RiPP. In order to obtain sufficient yields of the
peptide 1 for potential isolation and full structural characterisation, approaches to stimu-
late increased production of the target compound by the two Pseudomonas strains, SST2
and SST3, co-occurring in the SST, and both containing the respective BCGs for the lasso
peptide, were evaluated. Thus, we developed a co-cultivation protocol of the two strains,
P. zhaodongensis SST2 and Pseudomonas sp. SST3, with each other, as well as the terrestrial
Gamma-proteobacterium E. coli, and also including the relevant controls (Supplementary
Table S1). All cultures were fermented in modified GYM medium as stated above. The
rationale behind this approach was that these types of RiPPs are known to have a narrow
spectrum of antibacterial activity within the same class. An example for this finding is
microcin J25, isolated from E.coli, which displayed activity against E. coli H157:O7, Shigella,
and Salmonella, all of which belong to the Enterobacteriaceae family [31,32]. In addition to
the single culture of Pseudomonas sp. SST3 as described above, expression of the target
peptide 1 was detected in the co-cultivation of P. zhaodongensis SST2 and Pseudomonas
sp. SST3 and the co-cultivation of Pseudomonas sp. SST3 and E. coli. Conversely, no pro-
duction of the peptide was observed in the single culture of P. zhaodongensis SST2 or in its
co-culture with E. coli. A semiquantitative assessment of the titres was obtained by com-
paring the relative intensity of the base peak at m/z of 871.45, corresponding to the doubly
charged ion [M+2H]
2+
of the predicted peptide 1, and through a metabolomics approach
using SIMCA, multivariate analysis (MVA), such as principal component analysis (PCA),
and orthogonal projections to latent structures discriminant analysis (OPLS-DA).
3.3. Pseudomonassin Characterisation Using High-Energy Collision Dissociation Tandem Mass
Spectrometry (HDC-MS/MS)
The peptide in the genome of Pseudomonas sp. SST3 was elicited in the co-cultivation
medium. The sequence of the entire peptide was elucidated using high-energy collision
dissociation tandem mass spectrometry (HDC-MS/MS). The precursor mass [M+2H]
2+
Figure 1.
BGC of pseudomonassin (
1
) showing the four main gene clusters for its biosynthesis and
the sequence of the leader and core peptide.
3.3. Pseudomonassin Characterisation Using High-Energy Collision Dissociation Tandem Mass
Spectrometry (HDC-MS/MS)
The peptide in the genome of Pseudomonas sp. SST3 was elicited in the co-cultivation
medium. The sequence of the entire peptide was elucidated using high-energy collision
dissociation tandem mass spectrometry (HDC-MS/MS). The precursor mass [M+2H]
2+
871.45 +/
−
0.02 m/zof the peptide was selected in all HCD analyses. In this experiment,
we observed that there is little or almost no fragmentation of product ions observed at
20 normalised collision energy (NCE) for the doubly charged peptide species, and the
precursor ion [M+2H]
2+
871.45 +/
−
0.02 m/zdominates the spectrum. At this point, the
higher-mass fragments from the tail residues are of the b-ions (b
12
, b
14
, b
15
, and b
16
); these
ions have a mass above 1000 Da and some of the y-ions (y
5
and y
6
), both oh which have less
than 1000 Da in mass also appear on the spectrum. As the energy increases, the precursor
ion is fragmented into product ions more effectively, and more fragments from the lower
mass region are observed [
33
]. At 30 NCE, this energy level was able to show the mass
of all the fragments of the peptide; thus, we deduced its entire sequence from tail down
to the macrolactam ring and the location of the ring opening. Here, we find the 30 NCE
as the optimum collision energy, wherein we observed both the b- and y-ion fragments,
respectively (Figure S2). The opening of the macrolactam ring was observed to be at the
site of the glutamic acid sequentially going to glycine, as we mined from the putative
structure. The precursor mass of the peptide is still observable at a high intensity, and the
product ions of the peptide, from highest to lowest molecular weight of each amino acid
Microorganisms 2023,11, 2563 6 of 11
component of the b-ions, are seen in the spectrum. For the y-ions, y
11
, y
15
, and y
16
cannot
be observed at this collision energy. As we increased the NCE to 40 and 50, the shifting of
the fragmentations from higher masses to lower ones were already appearing [
34
–
36
], the
precursor ion [M+2H]
2+
871.45 +/
−
0.02 m/zwas no longer observable, and we started
to see the smaller products’ ions. At this collision energy, the spectrum became highly
dominated by ions less than 400 Da, and the lowest possible fragmentation of the amino
acids were seen in the spectrum on both b- and y-ions. (Figure S3). The HCD-MS/MS
method was preferred for the complete elucidation and sequencing of pseudomonassin.
ProteinProspector [
37
] readily calculates the theoretical accurate masses expected
upon fragmentation of the ring, can be compared to those observed at 30% collision energy,
and displayed all the b- and y- ions of the fragmented peptide in 30 NCE (Table 1).
Table 1.
Fragmentation pattern observed on the a-,b-, and y-ions: the loss of water from the N-
terminus and C-terminus from ProteinProspector and the actual mass fragments from 30 NCE
HCD-MS/MS.
Amino
Acid
b-ions y-ions
Theoretical
MW
Fragment
Observed
MW
Fragment
Error
Theoretical
MW
Fragment
Observed
MW
Fragment
Error
G - - -
V 157.0972 157.0971 0.64 ppm 1684.8918 - -
P 254.1499 254.1495 1.57 ppm 1585.8234 - -
G 311.1714 311.1710 1.29 ppm 1488.7706 1488.7701 −0.34 ppm
G 368.1928 368.1923 1.36 ppm 1431.7492 1431.7497 0.34 ppm
G 425.2143 425.2137 1.41 ppm 1374.7277 1374.7292 1.09 ppm
I 538.2984 538.2979 0.93 ppm 1317.7062 -
I 651.3824 651.3818 0.92 ppm 1204.6222 1204.6205 −1.41 ppm
E 762.4145 762.4138 0.92 ppm 1091.5381 1091.5361 −1.83 ppm
G 819.4359 819.4351 0.98 ppm 980.5061 980.5026 −3.57 ppm
R 975.5370 975.5354 1.64 ppm 923.4846 923.4833 −1.41 ppm
Y 1138.6004 1138.5991 1.14 ppm 767.3835 767.3825 −1.30 ppm
G 1195.6218 1195.6205 1.09 ppm 604.3202 604.3201 −0.17 ppm
V 1294.6908 1294.6897 0.85 ppm 547.2987 547.2984 −0.55 ppm
W 1480.7696 1480.7712 −1.08 ppm 448.2303 448.23 −0.67 ppm
S 1567.8016 1567.8020 −0.26 ppm 262.151 262.1507 −1.14 ppm
R - - - 175.1190 175.1188 −1.14 ppm
3.4. Metabolomics Analysis Reveals Potential New Metabolites from the Co-Cultivation of Two
Pseudomonas Species
We utilised two multivariate analysis (MVA) methods, principal component analysis
(PCA) and orthogonal partial least squares discriminant analysis (OPLS-DA), to extract
information from the data gathered in the LC-MS on the extracts of the monoculture and
co-cultivation of Pseudomonas sp. SST3, P. zhaodongensis SST2, and E. coli. The total number
of m/zfeatures from the monoculture and co-cultivation extracts is 30,059. These features
were processed in the mzMine to remove the noise and background peaks of the culture
medium. From these mass features, we conducted the dereplication of the extracts using a
Microsoft Excel macro and coupled them with information from the database Dictionary
of Natural Products (DNP). The principle of this approach, which, in part, is based on
proprietary protocols established in the group of Dr. RuAngelie Edrada-Ebel, is described
in a series of recent publications [
38
–
41
]. The hits were filtered according to the bacterial
source set to the genus Pseudomonas, and 22 potential hits were identified on the basis of
matching molecular formulae. Some of these compounds are produced by Pseudomonas
fluorescens and include the family of pseudomonic acid, from which the commercial topical
antibiotic mupirocin is derived [
42
,
43
]. This dereplication is a good indication of the
Microorganisms 2023,11, 2563 7 of 11
diversity of the metabolites being produced by deep-sea organisms in both monoculture
and co-cultivation settings (Table S1).
Prior to multivariate analysis (MVA), the mass spectral data were pre-processed in
mzMine to separate the signals from the noise [
44
]. MVA was initiated with PCA to give an
overview of the observations (samples) in correlation with their respective variables (fea-
tures). In this study, observations represented the samples from the different co-cultivation
parameters and ratios of the organisms employed, while the variables were the observed
LC-MS ion peaks for various metabolites. The PCA of eight extracts gave R2 (goodness
of fit) and Q2 (predictability) values of 0.923 and 0.881, respectively. Good predictability
and fit values indicated a linear correlation between the co-cultivation extracts and the
production of the metabolites. In general, the R2 should not exceed Q2 by more than
3 units
to ensure that a model is not over-fitted [
45
]. In the PCA scores plot (Figure 2A),
the two Pseudomonas spp. co-cultivated with E. coli were separated from the cluster (la-
beled as KM490 and KM492). The cluster consisted of the single and co-cultures of two
Pseudomonas spp. for the loadings plot (Figure 2); the target peptide was found amongst
the discriminating metabolites on the third quadrant (lower left) corresponding to that of
KM491. Dereplication of discriminating metabolites in the PCA loadings plot revealed
compounds of the amonabactin family, a class of tryptophan or phenylalanine-containing
siderophores, namely, amonabactin P693 m/z693.3010 [M+H]
+
, amonabactin P750 m/z
750.3217 [M+H]
+
(4.39), and amonabactin P789 m/z789.3335 [M+H]
+
, respectively. These
dereplicated compounds are of NRPS origin and have been annotated in the Pseudomonas
sp. SST3 genome. Using DNP, NP Atlas, [
46
], and Reaxys [
47
] databases gave no hits for ion
peaks at m/z[M+H]
+
674.3524, 694.3082, 758.4085, and 1124.6382, whereas ion peaks at m/z
[M+H] 530.2971, 576.3515, 652.4073, 671.4139, and 751.3670 afforded hits originating from
plants, fungi, and Gram-positive microorganisms such as Streptomyces and Micromonospora
species, which were disregarded. The unidentified discriminating metabolites for KM490
and KM492 could potentially be new metabolites and warrant further investigation for
targeted isolation in future.
Microorganisms 2023, 11, x FOR PEER REVIEW 8 of 12
(metabolites) responsible for the separation of the assigned groups. In this case, either the
monoculture or the co-culture extracts were considered to determine the production of
the target peptide. Interpretation of the OPLS-DA scores and loadings plots was the same
as the PCA [44]. For the OPLS-DA scores plot (Figure 2), the sample extracts were grouped
according to the type of culturing approach, as in monocultures (right quadrants) versus
co-cultures (left quadrants). The co-cultures were further grouped according to the em-
ployed organisms. In the OPLS-DA loadings plots, discriminating feature m/z [M+H]
+
318.1129 could be assigned to the monocultures, while ion peaks at m/z [M+H]+ 598.4905
and [M+2H]
2+
871.4596 were assigned for the co-cultures. However, due to the separation
of KM492 from the rest of the other co-culture extracts, the variation within the groups
(57.4%) was greater than that between groups (11.7%). Such separation further indicated
that the discriminating ion peak at m/z [M+H]
+
598.4905 was unique to KM492. The ion
peak at m/z [M+2H]
2+
871.4596 was dereplicated as the doubly charged target peptide,
which correlated with the co-culture extracts with E. coli.
Figure 2. PCA scores plot (A) and loading plots (B) and OPLS-DA scores plot (C) and loading plots
(D) of the single culture and co-cultivation of Pseudomonas zhaodongensis SST2, Pseudomonas sp. SST3,
and Escherichia coli Highlighted in blue, red and yellow circles are the featured m/z outliers, which
could be potential new metabolites and the double charged m/z for pseudomonassin.
3.5. Antibacterial and Biofilm Assay
Pseudomonassin was tested against the pathogenic P. aeruginosa with gentamicin as
the standard, as well as for the biofilm formation. The method that has been used for the
biofilm inhibition is the planktonic assay method. This test is used to assess the bacterial
aachment using the staining method for the biomass. It shows moderate activity against
the pathogen with an MIC of 63 µg/mL and a good activity against biofilm formation at
28 µg/mL (Figure 3).
Figure 2.
PCA scores plot (
A
) and loading plots (
B
) and OPLS-DA scores plot (
C
) and loading plots
(
D
) of the single culture and co-cultivation of Pseudomonas zhaodongensis SST2, Pseudomonas sp. SST3,
and Escherichia coli Highlighted in blue, red and yellow circles are the featured m/z outliers, which
could be potential new metabolites and the double charged m/z for pseudomonassin.
Microorganisms 2023,11, 2563 8 of 11
An OPLS-DA of the mass spectral data was also accomplished. It was employed
to construct a model for ease of interpretation of the discriminating spectral features
(metabolites) responsible for the separation of the assigned groups. In this case, either
the monoculture or the co-culture extracts were considered to determine the production
of the target peptide. Interpretation of the OPLS-DA scores and loadings plots was the
same as the PCA [
44
]. For the OPLS-DA scores plot (Figure 2), the sample extracts were
grouped according to the type of culturing approach, as in monocultures (right quadrants)
versus co-cultures (left quadrants). The co-cultures were further grouped according to the
employed organisms. In the OPLS-DA loadings plots, discriminating feature m/z[M+H]
+
318.1129 could be assigned to the monocultures, while ion peaks at m/z[M+H]+ 598.4905
and [M+2H]
2+
871.4596 were assigned for the co-cultures. However, due to the separation
of KM492 from the rest of the other co-culture extracts, the variation within the groups
(57.4%) was greater than that between groups (11.7%). Such separation further indicated
that the discriminating ion peak at m/z[M+H]
+
598.4905 was unique to KM492. The ion
peak at m/z[M+2H]
2+
871.4596 was dereplicated as the doubly charged target peptide,
which correlated with the co-culture extracts with E. coli.
3.5. Antibacterial and Biofilm Assay
Pseudomonassin was tested against the pathogenic P. aeruginosa with gentamicin as
the standard, as well as for the biofilm formation. The method that has been used for the
biofilm inhibition is the planktonic assay method. This test is used to assess the bacterial
attachment using the staining method for the biomass. It shows moderate activity against
the pathogen with an MIC of 63
µ
g/mL and a good activity against biofilm formation at
28 µg/mL (Figure 3).
Microorganisms 2023, 11, x FOR PEER REVIEW 9 of 12
(A) (B)
Figure 3. Bioassay results graph showing the activity of pseudomonassin against Pseudomonas ae-
ruginosa (A) and the inhibition of biofilm formation (B).
0 50 100 150
0.0
0.5
1.0
1.5
2.0
2.5
Planktonic Assay Solution
Concentration (µg/mL)
% log biofilm inhibition
Pseudomonassin
Gentamicin
Figure 3.
Bioassay results graph showing the activity of pseudomonassin against Pseudomonas
aeruginosa (A) and the inhibition of biofilm formation (B).
4. Discussion
The elicitation of the peptide and comparison with the LC-MS/MS revealed that the
co-culture of P. zhaodongensis SST2 and Pseudomonas sp. SST3, in a 2:1 volume ratio of the
culture broths, gave the highest intensity, i.e., twice as much compared to the 1:1 volume
ratio, 3 times higher when compared to the co-culture with E. coli, and up to
40 times
higher
when compared with the monoculture of Pseudomonas sp. SST3 (Figure S4). Obviously,
due to the differences in ionization efficiency in LC-MS/MS, there is no intention to imply
that the intensity of the respective ion accurately reflects the concentration of the target
compound in each extract; nonetheless, this analysis served to optimise fermentation
conditions for the subsequent isolation of 1upon large-scale fermentation.
In the OPLS-DA analysis, potential new metabolites also arising from the co-culture
have been identified. For the ion peaks m/z[M+H]
+
318.1129 and 598.4905, no hits were
found from any of the databases used, suggesting that these discriminating features can be
Microorganisms 2023,11, 2563 9 of 11
potentially new metabolites. Using a metabolomics approach on chemical profiling studies
assisted the prioritisation of the co-cultivation method and predicted the probability of
each of the conditions to elicit the production of the target peptide as well as other potential
new metabolites. The metabolomics approach also lowers the chance of the redundancy of
re-isolating known metabolites for a more efficient targeted isolation work. This is the first
time that this kind of peptide isolated from its native strain has shown activity against this
type of pathogen and inhibition of biofilm formation. Continuous research and trials on
how to combat biofilm formation in clinical isolates of P. aeruginosa is still a difficult task
due to the mutations that lead, once again, to resistance against several drug combinations.
These findings on its bioactivity, as well as its ability to inhibit biofilm formation, present
the potential of this class of RiPPs as a therapeutic model or drug delivery method against
selective pathogens.
5. Conclusions
Using a co-cultivation technique, we have successfully elicited the class of RiPPs called
lasso peptides in Pseudomonas sp. SST3, pseudomonassin. LC-MS analysis highlighted
that the co-culture of P. zhaodongensis SST2 and Pseudomonas sp. SST3 gave the highest
intensity of the target peptide. Subsequent fragmentations of the b- and y-ions of the tail
confirmed the correct sequence, as predicted by RiPPMiner. The metabolomics analysis
revealed several discriminating metabolites between the single and the co-cultures, some
of which are assumed to belong to the amonabactin family of siderophores, as suggested
by their molecular formulae. Furthermore, several potentially new compounds, with no
hits in natural products databases, represent promising targets for future isolation and
structural characterization.
Supplementary Materials:
The following supporting information can be downloaded at https:
//www.mdpi.com/article/10.3390/microorganisms11102563/s1.
Author Contributions:
K.J.M., D.A., S.J., S.A., R.E., M.J. and R.E.-E.: formal analysis and investigation;
K.J.M., M.J., R.E. and R.E.-E.: writing the original draft; M.J., R.E. and R.E.-E.: review and editing;
R.E., M.J. and R.E.-E: supervision and project administration. All authors have read and agreed to the
published version of the manuscript.
Funding:
K.J.M. is grateful to the British Council—Commission on Higher Education Philippines
(CHED) Newton Agham Scholarship Grant for the Ph.D. Fellowship. S.J. is grateful to the Middle
East University, Amman, Jordan for the Ph.D. Fellowship Grant.
Data Availability Statement:
Whole genome sequence of Pseudomonas sp. SST3 is publicly available
in GenBank.
Acknowledgments:
We would like to thank Scott Jarmusch for his contribution in mining and
predicting the RiPPs in our bacterial sample, as well as David A. Stead of the Institute of Medical
Sciences, University of Aberdeen, for the HCD-MS/MS data.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
Ziemert, N.; Alanjary, M.; Weber, T. The evolution of genome mining in microbes-a review. Nat. Prod. Rep.
2016
,33, 988–1005.
[CrossRef]
2.
Trivella, D.B.B.; de Felicio, R. The Tripod for Bacterial Natural Product Discovery: Genome Mining, Silent Pathway Induction,
and Mass Spectrometry-Based Molecular Networking. mSystems 2018,3, e00160-17. [CrossRef]
3.
Sayers, E.W.; Cavanaugh, M.; Clark, K.; Ostell, J.; Pruitt, K.D.; Karsch-Mizrachi, I. GenBank. Nucleic Acids Res.
2019
,47, D94–D99.
[CrossRef]
4.
Baral, B.; Akhgari, A.; Metsä-Ketelä, M. Activation of microbial secondary metabolic pathways: Avenues and challenges. Synth.
Syst. Biotechnol. 2018,3, 163–178. [CrossRef]
5.
Jones, J.A.; Wang, X. Use of bacterial co-cultures for the efficient production of chemicals. Curr. Opin. Biotechnol.
2018
,53, 33–38.
[CrossRef]
6.
Wakefield, J.; Hassan, H.M.; Jaspars, M.; Ebel, R.; Rateb, M.E. Dual induction of new microbial secondary metabolites by fungal
bacterial co-cultivation. Front. Microbiol. 2017,8, 1284. [CrossRef] [PubMed]
Microorganisms 2023,11, 2563 10 of 11
7.
Tawfike, A.F.; Viegelmann, C.; Edrada-Ebel, R. Metabolomics and Dereplication Strategies in Natural Products. Methods Mol. Biol.
2013,1055, 227–244.
8.
Macintyre, L.; Zhang, T.; Viegelmann, C.; Martinez, I.J.; Cheng, C.; Dowdells, C.; Abdelmohsen, U.R.; Gernert, C.; Hentschel, U.;
Edrada-Ebel, R.A. Metabolomic tools for secondary metabolite discovery from marine microbial symbionts. Mar. Drugs
2014
,12,
3416–3448. [CrossRef]
9.
Stuart, K.A.; Welsh, K.; Walker, M.C.; Edrada-Ebel, R.A. Metabolomic tools used in marine natural product drug discovery. Expert
Opin. Drug Discov. 2020,15, 499–522. [CrossRef] [PubMed]
10. Worley, B.; Powers, R. Multivariate Analysis in Metabolomics. Curr. Metabolomics 2013,1, 92–107. [CrossRef] [PubMed]
11.
Martin-Gómez, H.; Tulla-Puche, J. Lasso peptides: Chemical approaches and structural elucidation. Org. Biomol. Chem.
2018
,16,
5065–5080. [CrossRef]
12. de Bary, A. Morphologie und Physiologie der Pilze, Flechten und Myxomyceten; Wilhelm Engelmann: Leipzig, Germany, 1866.
13.
Hegemann, J.D.; Zimmermann, M.; Xie, X.; Marahiel, M.A. Lasso Peptides: An Intriguing Class of Bacterial Natural Products.
Acc. Chem. Res. 2015,48, 1909–1919. [CrossRef]
14.
Tan, S.; Moore, G.; Nodwell, J. Put a bow on it: Knotted antibiotics take center stage. Antibiotics
2019
,8, 117. [CrossRef] [PubMed]
15.
Arnison, P.G.; Bibb, M.J.; Bierbaum, G.; Bowers, A.A.; Bugni, T.S.; Bulaj, G.; Camarero, J.A.; Campopiano, D.J.; Challis, G.L.; Clardy,
J.; et al. Ribosomally synthesized and post-translationally modified peptide natural products: Overview and recommendations
for a universal nomenclature. Nat. Prod. Rep. 2013,30, 108–160. [CrossRef] [PubMed]
16.
Naimi, S.; Zirah, S.; Hammami, R.; Fernandez, B.; Rebuffat, S.; Fliss, I. Fate and biological activity of the antimicrobial lasso
peptide microcin J25 under gastrointestinal tract conditions. Front. Microbiol. 2018,9, 1764. [CrossRef]
17.
Li, Y.; Rebuffat, S. The manifold roles of microbial ribosomal peptide-based natural products in physiology and ecology. J. Biol.
Chem. 2020,295, 34–54. [CrossRef]
18.
Cheung-Lee, W.L.; Parry, M.E.; Cartagena, A.J.; Darst, S.A.; James Link, A. Discovery and structure of the antimicrobial lasso
peptide citrocin. J. Biol. Chem. 2019,294, 6822–6830. [CrossRef]
19.
Zimmermann, M.; Hegemann, J.D.; Xie, X.; Marahiel, M.A. The astexin-1 lasso peptides: Biosynthesis, stability, and structural
studies. Chem. Biol. 2013,20, 558–569. [CrossRef] [PubMed]
20.
Knappe, T.A.; Linne, U.; Robbel, L.; Marahiel, M.A. Insights into the Biosynthesis and Stability of the Lasso Peptide Capistruin.
Chem. Biol. 2009,16, 1290–1298. [CrossRef]
21.
Abdel-Mageed, W.M.; Lehri, B.; Jarmusch, S.A.; Miranda, K.J.; Al-Wahaibi, L.H.; Stewart, H.A.; Jamieson, A.J.; Jaspars, M.;
Karlyshev, A.V. Whole genome sequencing of four bacterial strains from South Shetland Trench revealing biosynthetic and
environmental adaptation gene clusters. Mar. Genomics 2020,54, 100782. [CrossRef] [PubMed]
22.
Zuo, M.Q.; Sun, R.X.; Fang, R.Q.; He, S.M.; Dong, M.Q. Characterization of collision-induced dissociation of deprotonated
peptides of 4–16 amino acids using high-resolution mass spectrometry. Int. J. Mass Spectrom. 2019,445, 116186. [CrossRef]
23.
Kowalska-krochmal, B.; Dudek-wicher, R. The Minimum Inhibitory Concentration of Antibiotics: Methods, Interpretation,
Clinical Relevance. Pathogens 2021,10, 165. [CrossRef] [PubMed]
24.
CLSI. Performance Standards for Antimicrobial Susceptibility 30th ed CLSI supplement M100; Clinical and Laboratory Standards
Institute: Wayne, PA, USA, 2020.
25.
Oliveira, A.; Cunha, M.D.L.R.S. Comparison of methods for the detection of biofilm production in coagulase-negative staphylo-
cocci. BMC Res. Notes 2010,3, 260. [CrossRef]
26.
Macià, M.D.; Rojo-Molinero, E.; Oliver, A. Antimicrobial susceptibility testing in biofilm-growing bacteria. Clin. Microbiol. Infect.
2014,20, 981–990. [CrossRef] [PubMed]
27.
Blin, K.; Pascal Andreu, V.; De Los Santos, E.L.C.; Del Carratore, F.; Lee, S.Y.; Medema, M.H.; Weber, T. The antiSMASH database
version 2: A comprehensive resource on secondary metabolite biosynthetic gene clusters. Nucleic Acids Res.
2019
,47, D625–D630.
[CrossRef] [PubMed]
28.
Altschul, S.F.; Madden, T.L.; Schäffer, A.A.; Zhang, J.; Zhang, Z.; Miller, W.; Lipman, D.J. Gapped BLAST and PSI-BLAST: A new
generation of protein database search programs. Nucleic Acids Res. 1997,25, 3389–3402. [CrossRef]
29.
Agrawal, P.; Khater, S.; Gupta, M.; Sain, N.; Mohanty, D. RiPPMiner: A bioinformatics resource for deciphering chemical
structures of RiPPs based on prediction of cleavage and cross-links. Nucleic Acids Res. 2017,45, W80–W88. [CrossRef]
30.
Tietz, J.I.; Schwalen, C.J.; Patel, P.S.; Maxson, T.; Blair, P.M.; Tai, H.C.; Zakai, U.I.; Mitchell, D.A. A new genome-mining tool
redefines the lasso peptide biosynthetic landscape. Nat. Chem. Biol. 2017,13, 470–478. [CrossRef]
31.
Yu, H.; Li, N.; Zeng, X.; Liu, L.; Wang, Y.; Wang, G.; Cai, S.; Huang, S.; Ding, X.; Song, Q.; et al. A comprehensive antimicrobial
activity evaluation of the recombinant microcin j25 against the foodborne pathogens salmonella and e. Coli O157:H7 by using a
matrix of conditions. Front. Microbiol. 2019,10, 1954. [CrossRef]
32.
Rintoul, M.R.; De Arcuri, B.F.; Salomón, R.A.; Farías, R.N.; Morero, R.D. The antibacterial action of microcin J25: Evidence for
disruption of cytoplasmic membrane energization in Salmonella newport. FEMS Microbiol. Lett. 2001,204, 265–270. [CrossRef]
33.
Diedrich, J.K.; Pinto, A.F.M.; Yates, J.R. Energy dependence of HCD on peptide fragmentation: Stepped collisional energy finds
the sweet spot. J. Am. Soc. Mass Spectrom. 2013,24, 1690–1699. [CrossRef]
34.
Frese, C.K.; Altelaar, A.F.M.; Hennrich, M.L.; Nolting, D.; Zeller, M.; Griep-Raming, J.; Heck, A.J.R.; Mohammed, S. Improved
Peptide Identification by Targeted Fragmentation Using CID, HCD and ETD on an LTQ-Orbitrap Velos. J. Proteome Res.
2011
,10,
2377–2388. [CrossRef]
Microorganisms 2023,11, 2563 11 of 11
35.
Tu, C.; Li, J.; Shen, S.; Sheng, Q.; Shyr, Y.; Qu, J. Performance Investigation of Proteomic Identification by HCD/CID Fragmenta-
tions in Combination with High/Low-Resolution Detectors on a Tribrid, High-Field Orbitrap Instrument. PLoS ONE
2016
,11,
e0160160. [CrossRef] [PubMed]
36.
Stieger, C.E.; Doppler, P.; Mechtler, K. Optimized Fragmentation Improves the Identification of Peptides Cross-Linked by
MS-Cleavable Reagents. J. Proteome Res. 2019,18, 1363–1370. [CrossRef] [PubMed]
37. Clauser, K.; Baker, P. Protein Prospector. 1999. Available online: Http://Prospector.Ucsf.Edu/ (accessed on 15 November 2019).
38.
Mazlan, N.W.; Tate, R.; Yusoff, Y.M.; Clements, C.; Edrada-Ebel, R. Metabolomics-Guided Isolation of Anti-Trypanosomal
Compounds from Endophytic Fungi of the Mangrove plant Avicennia Lanata. Curr. Med. Chem.
2019
,27, 1815–1835. [CrossRef]
[PubMed]
39.
Tawfike, A.F.; Romli, M.; Clements, C.; Abbott, G.; Young, L.; Schumacher, M.; Diederich, M.; Farag, M.; Edrada-Ebel, R.A.
Isolation of anticancer and anti-trypanosome secondary metabolites from the endophytic fungus Aspergillus flocculus via
bioactivity guided isolation and MS based metabolomics. J. Chromatogr. B Anal. Technol. Biomed. Life Sci.
2019
,1006–1007, 71–83.
[CrossRef] [PubMed]
40.
do Nascimento, T.G.; dos Santos Arruda, R.E.; da Cruz Almeida, E.T.; dos Santos Oliveira, J.M.; Basílio-Júnior, I.D.; Celerino
de Moraes Porto, I.C.; Rodrigues Sabino, A.; Tonholo, J.; Gray, A.; Ebel, R.A.E.; et al. Comprehensive multivariate correlations
between climatic effect, metabolite-profile, antioxidant capacity and antibacterial activity of Brazilian red propolis metabolites
during seasonal study. Sci. Rep. 2019,9, 18293. [CrossRef]
41.
Sebak, M.; Saafan, A.E.; AbdelGhani, S.; Bakeer, W.; El-Gendy, A.O.; Espriu, L.C.; Duncan, K.; Edrada-Ebel, R. Bioassay- And
metabolomics-guided screening of bioactive soil actinomycetes from the ancient city of Ihnasia, Egypt. PLoS ONE
2019
,14,
e0226959. [CrossRef]
42.
El-Sayed, A.K.; Hothersall, J.; Thomas, C.M. Quorum-sensing-dependent regulation of biosynthesis of the polyketide antiobiotic
mupirocin in Pseudomonas fluorescens NCIMB 10586. Microbiology 2001,147, 2127–2139. [CrossRef]
43.
Gao, S.S.; Hothersall, J.; Wu, J.; Murphy, A.C.; Song, Z.; Stephens, E.R.; Thomas, C.M.; Crump, M.P.; Cox, R.J.; Simpson, T.J.; et al.
Biosynthesis of mupirocin by pseudomonas fluorescens NCIMB 10586 involves parallel pathways. J. Am. Chem. Soc.
2014
,136,
5501–5507. [CrossRef]
44.
Bartel, J.; Krumsiek, J.; Theis, F.J. Statistical methods for the analysis of high-throughput metabolomics data. Comput. Struct.
Biotechnol. J. 2013,4, e201301009. [CrossRef] [PubMed]
45.
Veerasamy, R.; Rajak, H.; Jain, A.; Sivadasan, S.; Varghese, C.P.; Agrawal, R.K. Validation of QSAR Models—Strategies and
Importance. Int. J. Drug Des. Disocovery 2011,2, 511–519.
46.
Van Santen, J.A.; Jacob, G.; Singh, A.L.; Aniebok, V.; Balunas, M.J.; Bunsko, D.; Neto, F.C.; Castaño-Espriu, L.; Chang, C.; Clark,
T.N.; et al. The Natural Products Atlas: An Open Access Knowledge Base for Microbial Natural Products Discovery. ACS Cent.
Sci. 2019,5, 1824–1833. [CrossRef] [PubMed]
47.
Lawson, A.J.; Swienty-Busch, J.; Géoui, T.; Evans, D. The making of reaxys—Towards unobstructed access to relevant chemistry
information. In The Future of the History of Chemical Information; ACS Symposium Series; ACS Publications: Washington, DC,
USA, 2014.
Disclaimer/Publisher’s Note:
The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.
