Variochelins, Lipopeptide Siderophores from Variovorax boronicumulans Discovered by Genome Mining

Article abstract of DOI:10.1021/acs.jnatprod.5b00932

Photoreactive siderophores have a major impact on the growth of planktonic organisms. To date, these molecules have mainly been reported from marine bacteria, although evidence is now accumulating that some terrestrial bacteria also harbor the biosynthetic potential for their production. In this paper, we describe the genomics-driven discovery and characterization of variochelins, lipopeptide siderophores from the bacterium Variovorax boronicumulans, which thrives in soil and freshwater habitats. Variochelins are different from most other lipopeptide siderophores in that their biosynthesis involves a polyketide synthase. We demonstrate that the ferric iron complex of variochelin A possesses photoreactive properties and present the MS-derived structures of two degradation products that emerge upon light exposure.

Full text of DOI:10.1021/acs.jnatprod.5b00932

Article
pubs.acs.org/jnp
Variochelins, Lipopeptide Siderophores from Variovorax
boronicumulans Discovered by Genome Mining
Colette Kurth,† Sebastian Schieferdecker,† Kalliopi Athanasopoulou,† Ivana Seccareccia,†
and Markus Nett*,†,‡
†Leibniz Institute for Natural Product Research and Infection Biology e.V., Hans-Knoll-Institute, Beutenbergstraße 11a, 07745 Jena,
̈
Germany
‡Department of Biochemical and Chemical Engineering, Technical Biology, Technical University Dortmund, Emil-Figge-Straße 66,
44227 Dortmund, Germany
S
* Supporting Information
ABSTRACT: Photoreactive siderophores have a major impact
on the growth of planktonic organisms. To date, these
molecules have mainly been reported from marine bacteria,
although evidence is now accumulating that some terrestrial
bacteria also harbor the biosynthetic potential for their
production. In this paper, we describe the genomics-driven
discovery and characterization of variochelins, lipopeptide
siderophores from the bacterium Variovorax boronicumulans,
which thrives in soil and freshwater habitats. Variochelins are
different from most other lipopeptide siderophores in that their
biosynthesis involves a polyketide synthase. We demonstrate
that the ferric iron complex of variochelin A possesses
photoreactive properties and present the MS-derived structures
of two degradation products that emerge upon light exposure.
ron is an essential nutrient for virtually all forms of life. It
implications and possibly contributes to the occurrence of
Iplays a crucial role in many key biological processes, such as algal blooms.15
The production of photoreactive siderophores has been
photosynthesis, respiration, N2 fixation, methanogenesis, oxy-
gen transport, gene regulation, and DNA biosynthesis.1 Despite
its abundance in Earth’s crust, iron’s biological availability is
severely limited in aerobic environments, in part due to the
formation of insoluble oxides or hydroxides.2 In order to secure
sufficient iron uptake, many bacteria and fungi secrete low
molecular weight compounds, so-called siderophores, which
have an extremely high affinity toward ferric iron.3,4 After
binding of the metal, the resulting complex is actively
transported back into the cell, where the metal is released by
a reductive or hydrolytic mechanism.5
Siderophores not only support the growth of the producing
organism but also play a significant role in the structuring of
microbial communities.6−9 Moreover, some lipopeptide side-
rophores were shown to serve as chemical mediators for
bacteria−algal interactions in the oceans.10 These molecules are
often distinguished by iron-binding α-hydroxycarboxylate
ligand groups.11 The latter absorb photons in the presence of
UV light, thereby triggering a ligand-to-metal charge transfer
reaction.10,12 As a result, ferrous iron is released into the
environment, which is then available for direct uptake by
surrounding microalgae.13 Since the latter provide organic
nutrients in exchange, the bacterial siderophore producers still
benefit from their resource expenditure for siderophore
biosynthesis.14 This mutualism has important ecological
reported from taxonomically distinct genera of marine bacteria,
such as Halomonas, Marinobacter, Ochrobactrum, Synechococcus,
and Vibrio.16−19 The widespread occurrence of these natural
products suggests that the carbon-for-iron exchange is a
common feature of bacteria−algal interactions. Recently,
three classes of photoreactive lipopeptide siderophores were
isolated from nonmarine strains.20−22 While the cupriachelins
are produced by a freshwater bacterium possibly having a
biological function similar to their marine counterparts,20
taiwachelin and serobactins are made by rhizosphere
bacteria.21,22 These findings raise questions concerning not
only the ecological significance of photoreactivity and
amphiphilicity in a soil environment but also whether additional
nonrecognized producers of such siderophores exist.
Here, we report our recent results on the discovery of novel
lipopeptide siderophores from terrestrial bacteria. Using a
genome mining strategy,23−26 we analyzed various strains for
the presence of genes that are involved in the biosynthesis of
such compounds. This approach resulted in the identification of
the genus Variovorax as a potential source of structurally new
siderophores. Subsequent screening efforts, which were guided
Received: October 19, 2015
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.5b00932
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 1. Putative siderophore biosynthesis gene clusters in bacteria of the genus Variovorax.
Figure 2. Structures of variochelin A (A) and variochelin B (B). The dashed lines through the structures show the “y” and “b” fragments obtained in
a tandem MS experiment. The depicted numbers indicate the corresponding m/z values.
by the chrome azurol S (CAS) assay,27 led to the isolation of
variochelins A and B from the bacterium Variovorax
boronicumulans. The structures of the two natural products
were elucidated by NMR and MS measurements as well as
Marfey’s analysis. We present and discuss the gene cluster
involved in variochelin biosynthesis and evaluate the photo-
reactive properties of variochelin A.
the well-studied taurine dioxygenase (TauD).20,30 In order to
identify siderophore gene clusters with these features, BLASTP
homology searches were conducted using an aspartate-
activating NRPS adenylation domain as well as a TauD-like
protein from cupriachelin biosynthesis as probes. Cross-
searches against the fatty acyl-AMP ligase (FAAL) domain
from taiwachelin biosynthesis or, alternatively, the starter
condensation domain from cupriachelin biosynthesis were
used to narrow down the initial results to siderophore loci that
possess genetic hallmarks of fatty acid incorporation.31 The hits
that were retrieved from this combined analysis were validated
by bioinformatic software to confirm the predictions concern-
ing the molecular architecture of the encoded natural
products.32 In this way, we identified a total of 16 nonmarine
strains that were likely to produce acyl peptide siderophores
with a β-hydroxyaspartate motif (Table S1). After excluding
those strains whose biosynthetic potential had already been
confirmed in previous investigations,20−22 13 strains remained,
covering six different genera. Among the newly identified
RESULTS AND DISCUSSION
■
Genome Mining and Siderophore Screening. A
common structural motif of photoreactive acyl peptide
siderophores is the presence of one or more β-hydroxyaspartate
residues. Biosynthetically, these moieties derive from aspartate,
which is incorporated into the respective siderophore by a
nonribosomal peptide synthetase (NRPS) and, subsequently,
subjected to an enzymatic oxidation reaction to introduce the
β-hydroxy moiety.28,29 The oxidation is carried out by an α-
ketoglutarate-dependent dioxygenase, which is very similar to
B
DOI: 10.1021/acs.jnatprod.5b00932
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 1. NMR Spectroscopic Data for Variochelin A in DMSO-d6
a
a
pos.
δC, type
δH (J in Hz)
HMBC
pos.
δC, type
δH (J in Hz)
HMBC
Nδ-acetyl-Nδ-hydroxyornithine
β-hydroxyaspartic acid
C1
C2
C3
173.4, C
51.6, CH
29.0, CH2
C23
C24
C25
C26
N4
168.4, C
55.0, CH
70.1, CH
173.0, C
4.17, dt (7.8, 4.4)
1.70, m
1, 3
4.74, dd (9.1, 2.7)
4.51, d (2.7)
23, 25, 26, 27
23, 24, 26
a:
2, 4, 5
2, 4, 5
2, 3, 5
3, 4
b:
1.56, m
C4
C5
C6
C7
N1
22.9, CH2
46.5, CH2
170.3, C
1.53, m
7.75, d (9.1)
24, 25, 27
3.48, m
4-amino-7-guanidino-3-hydroxy-2-methylheptanoic acid
C27
C28
C29
C30
C31
C32
175.5, C
20.3, CH3
1.95, s
6
8
41.8, CH
73.3, CH
49.5, CH
27.5, CH2
25.1, CH2
2.49, m
3.56, m
3.66, m
1.22, m
1.44, m
1.36, m
3.10, m
3.02, m
1.00, d (6.9)
27, 29, 34
8.08, d (7.8)
27, 28, 30, 31, 34
29, 31, 32, 36
29, 30, 32
Nδ-acetyl-Nδ-hydroxyornithine
C8
171.4, C
C9
51.6, CH
29.8, CH2
4.35, m
8, 11
a:
b:
a:
b:
30, 31, 33
C10
a:
1.61, m
1.50, m
1.53, m
3.48, m
9, 11, 12
9, 11, 12
9, 10, 12
10, 11
30, 31, 33
b:
C33
40.8, CH2
31, 32, 35
C11
C12
C13
C14
N2
22.8, CH2
46.8, CH2
170.3, C
31, 32, 35
C34
C35
N5
11.3, CH3
156.5, C
27, 28, 29
20.3, CH3
1.96, s
13
15
7.56, d (9.2)
7.40, t (5.8)
29, 30, 36
33
7.96, d (8.7)
N6
proline
dodecanoic acid
172.1, C
C15
C16
C17
171.3, C
59.7, CH
29.8, CH2
C36
C37
C38
C39
C40
C41
C42
C43
C44
C45
C46
C47
4.38, dd (8.4, 3.7)
2.05, m
15, 17, 18
15, 16
35.5, CH2
25.4, CH2
28.7, CH2
28.8, CH2
29.0, CH2
29.1, CH2
29.0, CH2
29.0, CH2
31.3, CH2
22.1, CH2
14.0, CH3
2.07, t (7.4)
1.47, m
36, 38, 39
a:
36, 37, 39, 40
b
b:
1.83, m
16
1.23, m
n.r.
C18
C19
24.1, CH2
47.0, CH2
serine
1.49, m
17, 19
1.23, m
n.r.
3.48, m
17, 18, 20
1.23, m
n.r.
1.23, m
n.r.
C20
C21
C22
N3
168.5, C
52.8, CH
61.8, CH2
1.23, m
n.r.
4.65, q (7.4)
3.48, d (7.4)
7.72, d (7.4)
20, 22, 23
20, 21
1.23, m
n.r.
1.23, m
n.r.
21, 23
1.26, m
45, 47
45, 46
0.85, t (7.0)
a
b
HMBC correlations, optimized for 7.7 Hz, are from proton(s) stated to the indicated carbon. n.r., not resolved.
producers, the genus Variovorax appeared to be of particular
interest. Unlike the known acyl peptide siderophore gene
clusters,29 the loci in the three Variovorax paradoxus strains
include distinctive polyketide synthase (PKS) genes (Figure 1).
A thorough inspection of the NRPS and PKS domain
architecture and substrate specificities32−36 unveiled the close
relatedness of the biosynthetic enzymes in V. paradoxus B4 and
V. paradoxus S110, suggesting that they catalyze the production
of structurally identical molecules (Table S2). In contrast, the
gene cluster of V. paradoxus EPS clearly differs from the other
two loci in size and gene organization. A total of 10 NRPS and
PKS modules in the EPS assembly line outnumbers the six
modules from B4 and S110 and indicates the biosynthesis of a
significantly larger siderophore (Table S3).
To test the secretion of iron-chelating metabolites, we
subjected five Variovorax strains available in our laboratory to a
modified CAS assay, in which the siderophore detection is
spatially separated from the growth area of the respective
bacterium.37 All five strains produced an orange halo in the
CAS zone of the agar plate (Figure S1), thereby indicating the
release of iron-chelating agents.27 A comparison of the different
halo sizes in three independently conducted experiments
revealed that V. boronicumulans BAM-48 consistently gave the
strongest assay response when compared to the other strains.
Since the bacterium also grew vigorously under established
siderophore production conditions (data not shown),20,21 we
decided to carry out the following chemical investigations with
this organism.
Isolation and Structure Determination of Varioche-
lins. In order to induce siderophore biosynthesis in
V. boronicumulans BAM-48, the bacterium was cultivated in
H-3 minimal medium under iron starvation conditions.
Secreted metabolites were recovered from the fermentation
broth by adsorption onto XAD-2 resin. After removal of the
supernatant, the resin was eluted with methanol to release the
bound molecules. The resulting extract was concentrated and
subjected to HPLC. Two peaks in the metabolic profile
corresponded to iron-chelating compounds, as evidenced by a
positive response of the respective fractions in the liquid CAS
assay (Figure S2).27 The associated compounds were isolated
and subjected to ESIMS. Distinctive pseudomolecular ions
appeared at m/z 1074 [M + H]+ for compound 1 and m/z
1102 [M + H]+ for compound 2, respectively. A preliminary
inspection of their 1H NMR spectra suggested both metabolites
to be acyl peptides.
High-resolution ESIMS of variochelin A (1) yielded m/z
1074.6040 for the [M + H]+ ion, indicating a molecular formula
of C47H83N11O17. The constituents in the peptidic headgroup
of 1 and their connectivity were initially deduced by tandem
mass spectrometry.38 A sequential loss of 191, 172, 97, 87, and
C
DOI: 10.1021/acs.jnatprod.5b00932
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 3. Organization of the variochelin biosynthesis cluster (for annotations see Table S4) (A). Molecular assembly line deduced from varF−varJ
and proposed biosynthesis of 1 (B). Domain annotation: FAAL, fatty acyl-AMP ligase; ACP, acyl carrier protein; C, condensation; A, adenylation;
PCP, peptidyl carrier protein; KS, ketosynthase; AT, acyltransferase; KR, ketoreductase; TauD, hydroxylase; E, epimerization; TE, thioesterase.
131 mass units during MALDI-TOF/TOF fragmentation was
attributed to an amino acid sequence of Nδ-acetyl-Nδ-
hydroxyornithine, Nδ-acetyl-Nδ-hydroxyornithine, proline, ser-
ine, and β-hydroxyaspartic acid from the carboxylate terminus
(Figure 2A, Figure S3). The configurations of the amino acid
residues were determined by Marfey’s method upon acidic
hydrolysis of 1.39 This analysis established the proline residue
to be in L configuration, whereas the serine was found to be D-
configured (Figures S4, S5). Marfey’s method also revealed the
presence of threo-β-hydroxyaspartic acid. The elution order of
the diastereomeric pairs of L-FDAA-derivatized threo-β-
hydroxyaspartic acid had previously been shown to be D → L
under reversed-phase conditions.40 The variochelin A hydro-
lysate contained only one single peak of the correct mass upon
conversion with L-FDAA, which eluted at the same retention
time as the second peak of L-FDAA-derivatized D,L-threo-β-
hydroxyaspartic acid (Figure S6). Therefore, we concluded that,
out of the four possible stereoisomers of β-hydroxyaspartic acid,
the L-threo form is present in 1. Despite repeated attempts, the
derivatization of the released ornithine units with L-FDAA
failed. To determine the configuration of these amino acid
residues, we treated the hydrolysate of 1 with (1R,2S,5R)-2-
isopropyl-5-methylcyclohexyl carbonochloridate. Co-chroma-
tography of the resulting ornithine carbamate against synthetic
standards eventually established the L configuration for both
Nδ-acetyl-Nδ-hydroxyornithine moieties in 1 (Figure S7).
The amino acid sequence that had been inferred from the
interpretation of the tandem mass spectra was subsequently
confirmed by NMR data (Table 1, Figures S8−S12). To
elucidate the full structure of 1, we then focused on carbon
C35, which had not been assigned yet and was distinguished by
its chemical shift at 156.5 ppm. This value was lower than those
observed for the amide carbonyl groups and could not be
traced to an aromatic moiety. Instead a comparison with
literature data strongly suggested that C35 is part of a
guanidino group.41−43 HMBC interactions then enabled the
identification of a 4-amino-7-guanidino-3-hydroxy-2-methyl-
heptanoate fragment. The relative stereochemistry of the
three chiral centers in this moiety was deduced as
(28S*,29S*,30S*) by selective NOESY experiments. Upon
irradiation at the resonance frequency of H34, an NOE was
observed with H30, but not with H29. Likewise, irradiation on
H28 revealed an NOE with H29, but not with H30. Eventually,
it was shown by long-range correlations from H24 and NH4 to
C27 that the unusual γ-amino acid was connected to the β-
hydroxyaspartate residue of 1. The remaining nonassigned
signals in the 1H and 13C spectra were distinctive of an
unbranched acyl chain. The latter was attributed to a
dodecanoyl residue in consideration of the molecular formula
of 1. HMBC interactions from H30 and NH5 to C36 linked the
acyl moiety with the rest of the molecule and, thereby,
established the complete structure of 1.
The sum formula of variochelin B (2) was calculated as
C49H87N11O17 from its high-resolution mass. Tandem mass
spectrometry revealed the same y fragments as those observed
during the fragmentation of 1. However, the corresponding b
fragments were increased by 28 mass units each (Figure 2B).
MS and NMR data (Figures S13, S14) suggested that both
variochelins differ in their fatty acid tail, with 2 featuring a
tetradecanoic acid residue. A 13C NMR spectrum recorded with
an inverse-gated decoupling pulse sequence44 confirmed the
presence of 14 carbons in the alkyl chain of 2.
Variochelin Gene Cluster and Biosynthetic Model. The
guanidino-containing γ-amino acid that is present in both
D
DOI: 10.1021/acs.jnatprod.5b00932
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 4. Proposed reaction scheme for the photolysis of ferric 1. The depicted cleavage products 3 and 4 were detected by HRESIMS. The
complexed ferric iron is likely to be reduced via ligand-to-metal charge transfer.
variochelins can be biosynthetically rationalized as the product
of a decarboxylative Claisen condensation of arginine and
methylmalonate, which is then subject to a β-keto reduction.
This scenario suggests the involvement of a PKS and could
indicate that the variochelin gene cluster is organizationally
closely related to the siderophore loci that were previously
discovered in the three V. paradoxus strains. To find out
whether this assumption is correct, we sequenced and
annotated the entire genome of V. boronicumulans BAM-48.
Gene clusters with a putative role in secondary metabolism
were identified using antiSMASH 3.0.1.32 Out of the seven loci
detected, only one met the defined criteria for the biosynthesis
of an acyl peptide with a β-hydroxy aspartate motif. The cluster
boundaries that had been predicted by antiSMASH 3.0.1 were
manually refined on the basis of functional annotations, gene
distances, and GC content shifts.25 According to this analysis,
the variochelin (var.) locus (Figure 3A) consists of 18 genes,
covers 43.2 kb of contiguous DNA, and displays a significant
similarity to the V. paradoxus siderophore clusters. However,
substrate specificity predictions suggest different metabolic
products (Tables S2, S3).33−36 The seven NRPS and PKS
modules encoded by varF, varG, varH, varI, and varJ are
assumed to assemble the molecular backbones of 1 and 2
(Figure 3B). The biosynthesis would hence start with the
activation of dodecanonoic (or tetradecanoic) acid by the
FAAL domain of VarF. It then follows the collinear logic of
assembly line enzymology, where the activated substrates are
incorporated into a linear oligomer by successive condensation
steps.25,45 The PKS VarG contains the typical β-ketoacylsyn-
thase (KS), acyl transferase (AT), and acyl carrier protein
(ACP) domains, as well as a ketoreductase (KR) domain. Each
NRPS module harbors the complete set of condensation (C),
adenylation (A), and peptidyl carrier protein (PCP) domains.
The TauD domain of VarG would be responsible for the
hydroxylation of the incorporated aspartic acid residue.
Additionally, VarI harbors an epimerization (E) domain,
which would be required for the stereochemical inversion of
L-serine. The domain architecture is hence consistent with the
experimentally deduced configurations. Although Marfey’s
analysis provided no information on the stereochemistry of
the 4-amino-7-guanidino-3-hydroxy-2-methylheptanoate moi-
ety, the missing E domain in VarF strongly suggests an
L
configuration for the primary arginine building block. We thus
propose an (28S,29S,30S) absolute configuration for 1 and 2.
Once the chain elongation is completed, the terminal
thioesterase (TE) domain in VarJ releases the newly
synthesized lipopeptide.
The var. cluster also features several accessory proteins that
are essential for the proper functioning of the NRPS and PKS
enzymes. Small MbtH-like proteins, such as VarC, are assumed
to influence amino acid activation by NRPS,46 whereas the role
of type II thioesterases, such as VarD, lies in the removal of
aberrant intermediates that may block the NRPS/PKS assembly
line. Furthermore, type II thioesterases are possibly involved in
substrate selection and in product release.47 The phospho-
pantetheinyl transferase VarE is essential to convert the carrier
protein domains of the NRPS and PKS from the inactive apo
into the active, substrate-binding holo forms.48 VarN and VarO
were annotated as L-ornithine 5-monooxygenase and Nδ-
hydroxyornithine acetyltransferase, respectively. Similar to the
homologous IucD and IucB in aerobactin biosynthesis,49 the
two enzymes are assumed to act in a concerted manner to
generate the hydroxamate ligand groups in variochelins. The
remaining genes that are located in the var cluster are likely
involved in siderophore transport. Uptake of ferric iron−
variochelin complexes should occur via the TonB-dependent
receptor VarK and possibly also via the peptide transporter
VarR. Intracellular iron release from the siderophore would
then be mediated by the ferric iron reductase VarP.50 VarB,
VarL, and VarQ are homologues of FecR, a protein responsible
for the regulation of Fe3+-dicitrate uptake in Escherichia coli.51
Together with the encoded sigma-factors VarA and VarM, we
expect these proteins to regulate gene expression within the
cluster depending on iron availability.
E
DOI: 10.1021/acs.jnatprod.5b00932
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Complexing Properties and Photoreactivity of 1. The
variochelins possess three bidentate ligand groups for the
coordination of metal ions, including an α-hydroxycarboxylate
(i.e., the β-hydroxyaspartate residue) and two hydroxamate
functions. To test the complexing properties of 1, the
compound was treated with an equimolar quantity of a metal
salt and directly subjected to HR-MS. This analysis revealed
that 1 is capable of forming monomeric 1:1 complexes with
Fe3+ and Ga3+ (Figure S15). No masses corresponding to Mn2+,
Co2+, Ni2+, Cu2+, or Zn2+ complexes could be detected, and we
also did not observe any complex formation of 1 in the
presence of boron salts. The observed discrimination between
divalent and trivalent metal ions and the iron-responsive
production suggest a siderophore function for 1.
Depending on their coordination state, bidentate ligand
groups are sensitive to light exposure. While hydroxamates are
in general photochemically stable, catechols are prone to
photooxidation in the absence of metal coordination, but stable
once bound to ferric iron.52 In contrast, α-hydroxy acid
moieties are stable in their metal-free form, but undergo light-
induced oxidation after complexation to ferric iron.10,11 In
order to test the photostability of 1, we exposed an aqueous
solution of its ferric iron complex to direct sunlight and
analyzed product formation via LCESIMS. Two photoproducts
exhibited ions at m/z 1026.5862 [M − H]− and m/z 414.3440
[M + H]+, respectively (Figures S16, S17). The predicted sum
formula for the first photoproduct is consistent with an
aldehyde (3) resulting from a decarboxylation of the β-
hydroxyaspartate residue in 1.53 The second photoproduct (4)
is suggested to result from a more extensive cleavage reaction
(Figure 4). Despite thorough analysis we did not detect the
mass of a peptide fragment lacking the fatty acid tail. In a
parallel experiment, we confirmed the light-induced ligand-to-
metal charge transfer reaction, which leads to a reduction of the
coordinated ferric iron. Samples of Fe3+-variochelin A that were
exposed to sunlight gradually turned red in the presence of the
Fe2+-trapping agent bathophenantrolinedisulfonate (BPDS).
During the preparation of this article, an NRPS-PKS
assembly line with a domain architecture almost identical to
that of V. boronicumulans BAM-48 was reported from the
bacterium V. paradoxus P4B.60 A product from this assembly
line named variobactin A possesses the same molecular formula
as variochelin A. However, the structure of variobactin A was
proposed as a cyclic depsipeptide60 with an amino acid
sequence that deviates from that of variochelin A.
EXPERIMENTAL SECTION
■
General Experimental Procedures. LC-MS experiments were
conducted on an Accela UHPLC system equipped with a C18 column
(Betasil C18, 150 × 2.1 mm, 3 μm; Thermo Scientific) coupled to a
Finnigan Surveyor PDA Plus detector (Thermo Scientific). For
metabolic profiling, a gradient of acetonitrile (ACN) in water + 0.1%
formic acid and a flow rate of 0.2 mL/min were used. The ACN
concentration was increased from 5% to 98% within 16 min, was kept
for 3 min at 98%, and was subsequently decreased to 5% within 14
min. High-resolution mass determination was carried out on an
Exactive mass spectrometer (Thermo Scientific). One- and two-
dimensional MALDI-TOF/MS data using postsource decay were
acquired on a Bruker Ultraflex spectrometer (Bruker Daltonics). NMR
spectra were recorded at 300 K on a Bruker Avance III 500 MHz
spectrometer with DMSO-d6 as solvent and internal standard. The
solvent signals were referenced to δH 2.50 ppm and δC 39.5 ppm.
Siderophore Screening. CAS agar plates were prepared as
previously reported.27,37 Half of the CAS agar layer was cut out, and
the gap was filled with iron-free H-3 mineral medium (1.0 g/L aspartic
acid, 2.3 g/L KH2PO, 2.57 g/L Na2HPO4, 1.0 g/L NH4Cl, 0.5 g/L
MgSO4·7H2O, 0.5 g/L NaHCO3, 0.01 g/L CaCl2·2H2O, and 5 mL/L
SL-6 trace element solution).61 V. boronicumulans BAM-48, V. para-
doxus 351, B13-0-1 D, V. paradoxus B4, V. paradoxus S110, and V. soli
GH9-3 were streaked out on the H-3 half of the plates. The secretion
of iron-complexing metabolites was detected by a color change from
blue to orange in the CAS half of the plates after incubation at 30 °C.
Isolation of Variochelins. V. boronicumulans BAM-48 was grown
in a 10 L scale in iron-free H-3 mineral medium.61 The strain was
shaken (130 rpm) at 30 °C for 5 days. The culture supernatant was
then separated from the cells by centrifugation (8000 rpm) and
extracted with 150 g/L XAD-2 (Supelco). The resin was thoroughly
washed with distilled water before the adsorbed metabolites were
eluted with MeOH. The eluate was dried under vacuum, resuspended
in 1 mL of MeOH, and initially fractionated by flash column
chromatography over Polygoprep 60-50 C18 (Macherey-Nagel) using
an increasing concentration of MeOH in water. CAS-positive fractions
were further purified by high-performance liquid chromatography on a
Shimadzu UFLC liquid chromatography system equipped with a
Nucleodur C18 HTec column (VP 250 × 10 mm, 5 μm; Macherey-
Nagel) using a MeOH−H2O gradient from 50% to 100% over 20 min
and keeping 100% MeOH for 10 min, with 0.1% trifluoroacetic acid.
Variochelin A (1): 1H and 13C NMR data (DMSO-d6, 500 and 125
MHz, respectively), see Table 1; HRESIMS m/z 1074.6040 [M + H]+
calcd for C47H84N11O17 1074.6041.
Absorption at 535 nm increased from 0.004
0.001 to
0.0438 0.001 within 4 h of light exposure. The negative
control that was shielded from light remained colorless in the
same time period. Here, the absorption at 535 nm increased
from 0.0027 0.001 to 0.0032 0.001.
In summary, we herein report the discovery of a new class of
acyl peptides from the bacterium V. boronicumulans BAM-48.
Although the variochelins show typical hallmarks of marine
siderophores, the producing strain originates from soil,54 where
it was shown to contribute to plant growth.55 After
taiwachelin21 and the serobactins,22 the variochelins represent
the third class of photoreactive lipopeptide siderophores that
are produced by plant-associated bacteria. Amphiphilic side-
rophores that might be added to this group include corrugatin
and ornicorrugatin from Pseudomonas spp.,56 ornibactins from
Burkholderia spp.,57 and rhizobactin 1021 from Sinorhizobium
meliloti,58 even though the photoreactivity of these compounds
still awaits experimental confirmation. The occurrence of such
siderophores in bacteria thriving in the vicinity of plants raises
two possible conclusions: amphiphilicity and/or photoreactivity
are somewhat beneficial in the rhizosphere or the chemical
properties inherent to these siderophores represent an
evolutionary relic. It is noteworthy in this context that bacteria
of the genus Variovorax are also commonly found in freshwater
habitats,59 where an ecological advantage of amphiphilicity and
photoreactivity would be plausible.3
Variochelin B (2): 1H NMR (DMSO-d6, 500 MHz) δ 8.08 (1H, d, J
= 7.8 Hz, NH-1), 7.96 (1H, d, J = 8.7 Hz, NH-2), 7.75 (1H, d, J = 9.1
Hz, NH-4), 7.72 (1H, d, J = 7.4 Hz, NH-3), 7.56 (1H, d, J = 9.2 Hz,
NH-5), 7.40 (1H, t, J = 5.8 Hz, NH-6), 4.74 (1H, dd, J = 9.1, 2.7 Hz,
H-24), 4.65 (1H, q, J = 7.4 Hz, H-21), 4.51 (1H, d, J = 2.7 Hz, H-25),
4.38 (1H, dd, J = 8.4, 3.7 Hz, H-16), 4.35 (1H, m, H-9), 4.17 (1H, dt, J
= 7.8, 4.4 Hz, H-2), 3.66 (1H, m, H-30), 3.56 (1H, m, H-29), 3.48
(2H, m, H-5), 3.48 (2H, m, H-12), 3.48 (2H, m, H-19), 3.48 (2H, d, J
= 7.4 Hz, H-22), 3.10 (1H, m, H-33a), 3.02 (1H, m, H-33b), 2.49
(1H, m, H-28), 2.07 (2H, t, J = 7.4 Hz, H-37), 2.05 (1H, m, H-17a),
1.96 (3H, s, H-14), 1.95 (3H, s, H-7), 1.83 (1H, m, H-17b), 1.70 (1H,
m, H-3a), 1.61 (1H, m, H-10a), 1.56 (1H, m, H-3b), 1.53 (2H, m, H-
4), 1.53 (2H, m, H-11), 1.50 (1H, m, H-10b), 1.49 (2H, m, H-18),
1.47 (2H, m, H-38), 1.44 (1H, m, H-32a), 1.36 (1H, m, H-32b), 1.26
(2H, m, H-48), 1.23 (2H, m, H-39), 1.23 (2H, m, H-40), 1.23 (2H, m,
H-41), 1.23 (2H, m, H-42), 1.23 (2H, m, H-43), 1.23 (2H, m, H-44),
F
DOI: 10.1021/acs.jnatprod.5b00932
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
1.23 (2H, m, H-45), 1.23 (2H, m, H-46), 1.23 (2H, m, H-47), 1.22
(2H, m, H-31), 1.00 (3H, d, J = 6.9 Hz, H-34), 0.85 (3H, t, J = 7.0 Hz,
H-49); 13C NMR (DMSO-d6, 125 MHz) δ 175.5 (C, C-27), 173.4 (C,
C-1), 173.0 (C, C-26), 172.1 (C, C-36), 171.4 (C, C-8), 171.3 (C, C-
15), 170.3 (C, C-6), 170.3 (C, C-13), 168.5 (C, C-20), 168.4 (C, C-
23), 156.5 (C, C-35), 73.3 (CH, C-29), 70.1 (CH, C-25), 61.8 (CH2,
C-22), 59.7 (CH, C-16), 55.0 (CH, C-24), 52.8 (CH, C-21), 51.6
(CH, C-2), 51.6 (CH, C-9), 49.5 (CH, C-30), 47.0 (CH2, C-19), 46.8
(CH2, C-12), 46.5 (CH2, C-5), 41.8 (CH, C-28), 40.8 (CH2, C-33),
35.5 (CH2, C-37), 31.3 (CH2, C-47), 29.8 (CH2, C-10), 29.8 (CH2, C-
17), 29.1 (CH2, C-42), 29.0 (CH2, C-3), 29.0 (CH2, C-41), 29.0
(CH2, C-43), 29.0 (CH2, C-44), 29.0 (CH2, C-45), 29.0 (CH2, C-46),
28.8 (CH2, C-40), 28.7 (CH2, C-39), 27.5 (CH2, C-31), 25.4 (CH2, C-
38), 25.1 (CH2, C-32), 24.1 (CH2, C-18), 22.9 (CH2, C-4), 22.8
(CH2, C-11), 22.1 (CH2, C-48), 20.3 (CH3, C-7), 20.3 (CH3, C-14),
14.0 (CH3, C-49), 11.3 (CH3, C-34); HRESIMS m/z 1102.6357 [M +
H]+ calcd for C49H88N11O17 1102.6354.
7.5) was exposed to sunlight for 6 h. An identical solution that was
shielded from sunlight served as a negative control. After photo-
exposure, both samples were dried in vacuo. The samples were then
taken up in 100 μL of MeOH and analyzed by HRESIMS.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.5b00932.
Complete annotation of the Variovorax siderophore gene
clusters, including GenBank accession numbers; testing
of Variovorax strains in the CAS assay; HPLC
purification of 1 and 2; MS data and NMR spectra of
1 and 2; amino acid analysis; photoreactivity test of ferric
1 (PDF)
Amino Acid Analysis by Marfey’s Method. Amino acid
configurations were determined following acid hydrolysis and
derivatization with Marfey’s reagent (1-fluoro-2,4-dinitrophenyl-5-L-
alanine amide, L-FDAA, Sigma-Aldrich)39 by coelution experiments
with L-FDAA-derivatized amino acids. For this purpose, 1 mg of
purified 1 was dissolved in 500 μL of concentrated HI and heated at
110 °C for 8 h. The solution was lyophilized, and the dried hydrolysate
was resuspended in 50 μL of water and 20 μL of aqueous NaHCO3 (1
M). Derivatization was carried out with 100 μL of L-FDAA (1% w/v in
acetone) at 40 °C for 1 h. Afterward the reaction was quenched with
20 μL of HCl (1 M). The products were lyophilized and prepared for
LCHRMS analysis by dissolving in MeOH. Standards for co-
chromatography were prepared by reacting 50 μL of an aqueous
amino acid solution (50 mM) with 20 μL of NaHCO3 (1 M) and 100
μL of L-FDAA (1% w/v in acetone) at 40 °C for 1 h. The dried
reaction mixture was dissolved in MeOH and subsequently analyzed
by LCHRMS.
Configurational Analysis of the Released Ornithine Resi-
dues. Triphosgene (0.065 mmol) and pyridine (0.45 mmol) were
added to a solution of (−)-menthol (0.13 mmol) in 3 mL of DCM in
an ice bath. The solution was stirred for 30 min and subsequently
allowed to warm to room temperature. The stirring was continued for
30 min. The resulting (1R,2S,5R)-2-isopropyl-5-methylcyclohexyl
carbonochloridate was used directly for derivatization. For this
purpose, D- and L-ornithine standards (0.2 mmol) or hydrolyzed 1
was dissolved in a 2:1 DMSO−H2O mixture (3 mL) in the presence of
NaHCO3 (0.2 mmol) and added to the (1R,2S,5R)-2-isopropyl-5-
methylcyclohexyl carbonochloridate solution. After stirring for 1 h at
room temperature, the samples were lyophilized, dissolved in MeOH,
and analyzed by HRLCMS.
Genome Sequencing, Assembly, and Annotation. Genomic
DNA of V. boronicumulans BAM-48 was isolated via phenol
chloroform extraction. The purity, quality, and size of the bulk
gDNA preparation were assessed according to DOE-JGI guidelines.44
Sequencing was performed at GATC Biotech AG (Konstanz,
Germany) by means of single-molecule real-time sequencing.62 The
sequences extracted from the resulting data set were assembled using
the hierarchical genome assembly process.63 Variochelin biosynthesis
genes were first identified using antiSMASH 3.0.1.32 Refinement of the
cluster analysis was conducted as previously described.25,45 The
annotated nucleotide sequence for the variochelin gene cluster has
been deposited in GenBank under accession number KT900023.
Photoreactivity Tests. Photoreactivity tests of variochelins were
performed as previously described.20 Reduction of the complexed
ferric iron to ferrous iron was investigated via the bathophenan-
throlinedisulfonic acid assay. Each reaction contained 100 μM
variochelin A, 10 μM FeCl3, and 40 μM BPDS (Sigma-Aldrich) in
PBS buffer (pH 7.5). The reactions were either exposed to sunlight or
kept in the dark for 4 h. The formation of Fe(BPDS)32+ was monitored
before and after exposure to sunlight/darkness by measuring the
absorption at 535 nm using a Genesys 10 UV spectrophotometer
(Thermo). The experiments were run in duplicate. In order to identify
photolysis products, a 2 mM solution of ferric 1 in PBS buffer (pH
AUTHOR INFORMATION
Corresponding Author
*Phone: +49 231 7554742. Fax: +49 7557489. E-mail: markus.
nett@bci.tu-dortmund.de.
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This project was supported by the Collaborative Research
Center ChemBioSys (CRC 1127 ChemBioSys) and funded by
the Deutsche Forschungsgemeinschaft (DFG). We thank A.
Perner and H. Heinecke (Hans-Knoll-Institute Jena, Depart-
ment of Biomolecular Chemistry) for recording high-resolution
ESIMS and NMR spectra. We also thank M. Poetsch (Hans-
̈
Knoll-Institute Jena, Department of Molecular and Applied
Microbiology) for MALDI-TOF/TOF measurements.
̈
REFERENCES
■
(1) Andrews, S. C.; Robinson, A. K.; Rodríguez-Quinones, F. FEMS
̃
Microbiol Rev. 2003, 27, 215−237.
(2) Shaked, Y.; Erel, Y.; Sukenik, A. Environ. Sci. Technol. 2002, 36,
460−467.
(3) Sandy, M.; Butler, A. Chem. Rev. 2009, 109, 4580−4595.
(4) Hider, R. C.; Kong, X. Nat. Prod. Rep. 2010, 27, 637−657.
(5) Hopkinson, B. M.; Morel, F. M. M. BioMetals 2009, 22, 659−669.
(6) Matthijs, S.; Tehrani, K. A.; Laus, G.; Jackson, R. W.; Cooper, R.
M.; Cornelis, P. Environ. Microbiol. 2007, 9, 425−434.
(7) D’Onofrio, A.; Crawford, J. M.; Stewart, E. J.; Witt, K.; Gavrish,
E.; Epstein, S.; Clardy, J.; Lewis, K. Chem. Biol. 2010, 17, 254−264.
(8) Traxler, M. F.; Seyedsayamdost, M. R.; Clardy, J.; Kolter, R. Mol.
Microbiol. 2012, 86, 628−644.
(9) Bottcher, T.; Clardy, J. Angew. Chem., Int. Ed. 2014, 53, 3510−
̈
3513.
(10) Barbeau, K.; Rue, E. L.; Bruland, K. W.; Butler, A. Nature 2001,
413, 409−413.
(11) Butler, A.; Theisen, R. M. Coord. Chem. Rev. 2010, 254, 288−
296.
(12) Kupper, F. C.; Carrano, C. J.; Kuhn, J.-U.; Butler, A. Inorg.
̈
Chem. 2006, 45, 6028−6033.
(13) Yarimizu, K.; Polido, G.; Gardes, A.; Carter, M. L.; Hilbern, M.;
̈
Carrano, C. J. Metallomics 2014, 6, 1156−1163.
(14) Amin, S. A.; Green, D. H.; Hart, M. C.; Kupper, F. C.; Sunda,
̈
W. G.; Carrano, C. J. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 17071−
17076.
(15) Amin, S. A.; Parker, M. S.; Armbrust, E. V. Microbiol. Mol. Biol.
Rev. 2012, 76, 667−684.
(16) Gauglitz, J. M.; Zhou, H.; Butler, A. J. Inorg. Biochem. 2012, 107,
90−95.
G
DOI: 10.1021/acs.jnatprod.5b00932
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
(17) Martinez, J. S.; Zhang, G. P.; Holt, P. D.; Jung, H.-T.; Carrano,
C. J.; Haygood, M. G.; Butler, A. Science 2000, 287, 1245−1247.
(18) Ito, Y.; Butler, A. Limnol. Oceanogr. 2005, 50, 1918−1923.
(19) Homann, V. V.; Sandy, M.; Tincu, J. A.; Templeton, A. S.; Tebo,
B. M.; Butler, A. J. Nat. Prod. 2009, 72, 884−888.
(51) Braun, V.; Killmann, H. Trends Biochem. Sci. 1999, 24, 104−109.
(52) Barbeau, K.; Rue, E. L.; Trick, C. G.; Bruland, K. W.; Butler, A.
Limnol. Oceanogr. 2003, 48, 1069−1078.
(53) Grabo, J. E.; Chrisman, M. A.; Webb, L. M.; Baldwin, M. J.
Inorg. Chem. 2014, 53, 5781−5787.
(54) Miwa, H.; Ahmed, I.; Yoon, J.; Yokota, A.; Fujiwara, T. Int. J.
Syst. Evol. Microbiol. 2008, 58, 286−289.
(20) Kreutzer, M. F.; Kage, H.; Nett, M. J. Am. Chem. Soc. 2012, 134,
5415−5422.
(55) Liu, Z. H.; Cao, Y. M.; Zhou, Q. W.; Guo, K.; Ge, F.; Hou, J. Y.;
Hu, S. Y.; Yuan, S.; Dai, Y. J. Biodegradation 2013, 24, 855−864.
(21) Kreutzer, M. F.; Nett, M. Org. Biomol. Chem. 2012, 10, 9338−
9343.
(22) Rosconi, F.; Davyt, D.; Martínez, V.; Martínez, M.; Abin-
Carriquiry, J. A.; Zane, H.; Butler, A.; de Souza, E. M.; Fabiano, E.
Environ. Microbiol. 2013, 15, 916−927.
(23) Zerikly, M.; Challis, G. L. ChemBioChem 2009, 10, 625−633.
(24) Gross, H. Appl. Microbiol. Biotechnol. 2007, 75, 267−277.
(25) Nett, M. In Progress in Chemistry of Organic Natural Products;
Kinghorn, D. A.; Falk, H.; Kobayashi, J., Eds.; Springer: Switzerland,
2014; Vol. 99, pp 199−246.
(26) Kleigrewe, K.; Almaliti, J.; Tian, I. Y.; Kinnel, R. B.;
Korobeynikov, A.; Monroe, E. A.; Duggan, B. M.; Di Marzo, V.;
Sherman, D. H.; Dorrestein, P. C.; Gerwick, L.; Gerwick, W. H. J. Nat.
Prod. 2015, 78, 1671−1682.
(56) Matthijs, S.; Budzikiewicz, H.; Schafer, M.; Wathelet, B.;
̈
Cornelis, P. Z. Naturforsch., C: J. Biosci. 2008, 63, 8−12.
(57) Stephan, H.; Freund, S.; Beck, W.; Jung, G.; Meyer, J. M.;
Winkelmann, G. BioMetals 1993, 6, 93−100.
(58) Smith, M. J.; Shoolery, J. N.; Schwyn, B.; Holden, I.; Neilands, J.
B. J. Am. Chem. Soc. 1985, 107, 1739−1743.
(59) Satola, B.; Wubbeler, J. H.; Steinbuchel, A. Appl. Microbiol.
̈
̈
Biotechnol. 2013, 97, 541−560.
(60) Johnston, C. W.; Skinnider, M. A.; Wyatt, M. A.; Li, X.; Ranieri,
M. R. M.; Yang, L.; Zechel, D. L.; Ma, B.; Magarvey, N. A. Nat.
Commun. 2015, 6, 8421.
(61) Pfennig, N. Arch. Microbiol. 1974, 100, 197−206.
(62) Eid, J.; Fehr, A.; Gray, J.; Luong, K.; Lyle, J.; Otto, G.; Peluso,
P.; Rank, D.; Baybayan, P.; Bettman, B.; Bibillo, A.; Bjornson, K.;
Chauduri, B.; Christians, F.; Cicero, R.; Clark, S.; Dalal, R.; deWinter,
A.; Dixon, J.; Foquet, M.; Gaertner, A.; Hardenbol, P.; Heiner, C.;
Hester, K.; Holden, D.; Kearns, G.; Kong, X.; Kuse, R.; Lacroix, Y.;
Lin, S.; Lundquist, P.; Ma, C.; Marks, P.; Maxham, M.; Murphy, D.;
Park, I.; Pham, T.; Phillips, M.; Roy, J.; Sebra, R.; Shen, G.; Sorenson,
J.; Tomaney, A.; Travers, K.; Trulson, M.; Vieceli, J.; Wegener, J.; Wu,
D.; Yang, A.; Zaccarin, D.; Zhao, P.; Zhong, F.; Korlach, J.; Turner, S.
Science 2009, 323, 133−138.
(27) Schwyn, B.; Neilands, J. B. Anal. Biochem. 1987, 160, 47−56.
(28) Kem, M. P.; Naka, H.; Iinishi, A.; Haygood, M. G.; Butler, A.
Biochemistry 2015, 54, 744−752.
(29) Kem, M. P.; Butler, A. BioMetals 2015, 28, 445−459.
(30) Singh, G. M.; Fortin, P. D.; Koglin, A.; Walsh, C. T. Biochemistry
2008, 47, 11310−11320.
(31) Zane, H. K.; Naka, H.; Rosconi, F.; Sandy, M.; Haygood, M. G.;
Butler, A. J. Am. Chem. Soc. 2014, 136, 5615−5618.
(32) Weber, T.; Blin, K.; Duddela, S.; Krug, D.; Kim, H. U.;
Bruccoleri, R.; Lee, S. Y.; Fischbach, M. A.; Muller, R.; Wohlleben, W.;
̈
(63) Chin, C. S.; Alexander, D. H.; Marks, P.; Klammer, A. A.; Drake,
J.; Heiner, C.; Clum, A.; Copeland, A.; Huddleston, J.; Eichler, E. E.;
Turner, S. W.; Korlach, J. Nat. Methods 2013, 10, 563−569.
Breitling, R.; Takano, E.; Medema, M. H. Nucleic Acids Res. 2015, 43,
W237−W243.
(33) Bachmann, B. O.; Ravel, J. Methods Enzymol. 2009, 458, 181−
217.
(34) Stachelhaus, T.; Mootz, H. D.; Marahiel, M. A. Chem. Biol.
1999, 6, 493−505.
(35) Challis, G. L.; Ravel, J.; Townsend, C. A. Chem. Biol. 2000, 7,
211−224.
̌ ́
(36) Baranasic, D.; Zucko, J.; Diminic, J.; Gacesa, R.; Long, P. F.;
Cullum, J.; Hranueli, D.; Starcevic, A. J. Ind. Microbiol. Biotechnol.
2014, 41, 461−467.
(37) Milagres, A. M.; Machuca, A.; Napoleao, D. J. Microbiol. Methods
1999, 37, 1−6.
(38) Papayannopoulos, I. A. Mass Spectrom. Rev. 1995, 14, 49−73.
(39) Marfey, P. Carlsberg Res. Commun. 1984, 49, 591−596.
(40) Fujii, K.; Ikai, Y.; Mayumi, T.; Oka, H.; Suzuki, M.; Harada, K.
Anal. Chem. 1997, 69, 3346−3352.
(41) Fuchs, S. W.; Sachs, C. C.; Kegler, C.; Nollmann, F. I.; Karas,
M.; Bode, H. B. Anal. Chem. 2012, 84, 6948−6955.
(42) Bosello, M.; Zeyadi, M.; Kraas, F. I.; Linne, U.; Xie, X.;
Marahiel, M. A. J. Nat. Prod. 2013, 76, 2282−2290.
(43) Johnston, C. W.; Wyatt, M. A.; Li, X.; Ibrahim, A.; Shuster, J.;
Southam, G.; Magarvey, N. A. Nat. Chem. Biol. 2013, 9, 241−243.
(44) Berger, S.; Braun, S. In 200 and More NMR Experiments; Wiley-
VCH: Weinheim, 2004; pp 318−320.
(45) Schieferdecker, S.; Konig, S.; Weigel, C.; Dahse, H.-M.; Werz,
̈
O.; Nett, M. Chem. - Eur. J. 2014, 20, 15933−15940.
(46) Felnagle, E. A.; Barkei, J. J.; Park, H.; Podevels, A. M.;
McMahon, M. D.; Drott, D. W.; Thomas, M. G. Biochemistry 2010, 49,
8815−8817.
(47) Kotowska, M.; Pawlik, K. Appl. Microbiol. Biotechnol. 2014, 98,
7735−7746.
(48) Lambalot, R. H.; Gehring, A. M.; Flugel, R. S.; Zuber, P.;
LaCelle, M.; Marahiel, M. A.; Reid, R.; Khosla, C.; Walsh, C. T. Chem.
Biol. 1996, 3, 923−936.
(49) Challis, G. L. ChemBioChem 2005, 6, 601−611.
(50) Matzanke, B. F.; Anemuller, S.; Schunemann, V.; Trautwein, A.
̈
̈
X.; Hantke, K. Biochemistry 2004, 43, 1386−1392.
H
DOI: 10.1021/acs.jnatprod.5b00932
J. Nat. Prod. XXXX, XXX, XXX−XXX