Isolation, Characterization, and Synthesis of the Barrettides: Disulfide-Containing Peptides from the Marine Sponge Geodia barretti

Article abstract of DOI:10.1021/acs.jnatprod.5b00210

Two disulfide-containing peptides, barrettides A (1) and B (2), from the cold-water marine sponge Geodia barretti are described. Those 31 amino acid residue long peptides were sequenced using mass spectrometry methods and structurally characterized using NMR spectroscopy. The structure of 1 was confirmed by total synthesis using the solid-phase peptide synthesis approach that was developed. The two peptides were found to differ only at a single position in their sequence. The three-dimensional structure of 1 revealed that these peptides possess a unique fold consisting of a long β-hairpin structure that is cross-braced by two disulfide bonds in a ladder-like arrangement. The peptides are amphipathic in nature with the hydrophobic and charged residues clustered on separate faces of the molecule. The barrettides were found not to inhibit the growth of either Escherichia coli or Staphylococcus aureus but displayed antifouling activity against barnacle larvae (Balanus improvisus) without lethal effects in the concentrations tested. (Figure Presented).

Full text of DOI:10.1021/acs.jnatprod.5b00210

Article
pubs.acs.org/jnp
Isolation, Characterization, and Synthesis of the Barrettides:
Disulfide-Containing Peptides from the Marine Sponge Geodia
barretti
Bodil B. Carstens,† K. Johan Rosengren,‡ Sunithi Gunasekera,§ Stefanie Schempp,§ Lars Bohlin,§
Mia Dahlstrom,⊥ Richard J. Clark,‡ and Ulf Goransson*,§
̈
̈
‡
†Institute for Molecular Bioscience and School of Biomedical Sciences, University of Queensland, Brisbane, QLD 4072, Australia
§Division of Pharmacognosy, Department of Medicinal Chemistry, Biomedical Center, Uppsala University, Box 574, SE-751 23
Uppsala, Sweden
⊥Department of Chemistry, Materials and Surfaces, SP Technical Research Institute of Sweden, Arvid Wallgrens Backe 20, SE-413 46
Goteborg, Sweden
̈
S
* Supporting Information
ABSTRACT: Two disulfide-containing peptides, barrettides A
(1) and B (2), from the cold-water marine sponge Geodia barretti
are described. Those 31 amino acid residue long peptides were
sequenced using mass spectrometry methods and structurally
characterized using NMR spectroscopy. The structure of 1 was
confirmed by total synthesis using the solid-phase peptide
synthesis approach that was developed. The two peptides were
found to differ only at a single position in their sequence. The
three-dimensional structure of 1 revealed that these peptides
possess a unique fold consisting of a long β-hairpin structure that
is cross-braced by two disulfide bonds in a ladder-like
arrangement. The peptides are amphipathic in nature with the
hydrophobic and charged residues clustered on separate faces of
the molecule. The barrettides were found not to inhibit the
growth of either Escherichia coli or Staphylococcus aureus but
displayed antifouling activity against barnacle larvae (Balanus improvisus) without lethal effects in the concentrations tested.
Marine sponges have been a rich source of a range of
bioactive natural products including both organic molecules
and peptides.10,11 Interestingly, despite the diverse range of
disulfide-rich peptides from other phyla and classes of marine
organisms, there have been very few disulfide-rich peptides
described from marine sponges. The bacterial sialidase inhibitor
asteropine A was the first disulfide-rich peptide to be described
from a marine sponge, Asteropsus simplex Carter.12 Asteropine
A contains 36 amino acids with six cysteines that form three
disulfide bonds in a knotted topology. This cystine knot is a
common motif in disulfide-rich peptides, including conotoxins,
presumably because it imparts substantial stability to the
molecule.13 Another type of cystine-knot peptide, the
asteropsins, from an Asteropsus species has also recently been
described. The asteropsins have the same cystine framework as
asteropine A, but there is little conservation within the
remainder of the sequence. This is also a common feature
seen in other disulfide-rich peptides such as conotoxins, spider
toxins, and the plant-derived cyclotides, where a conserved
isulfide-rich peptides from marine sources such as the
Dconotoxins from snails of the Conus genus and the
cnidarian peptide toxins are a diverse source of biologically
active molecules for use as both drug leads and pharmacological
tools.1−6 These disulfide-containing peptides typically range
from 1.5 to 5 kDa in molecular weight and contain multiple
cysteines forming a network of disulfide bonds that stabilize the
peptide and define its three-dimensional structure. For example,
the conotoxins exhibit vast variations in the number, spacing,
and connectivity of cysteines, resulting in 26 cysteine
frameworks (defined by the number and spacing of the
cysteine residues) and 12 different classes of folds (defined by
the cystine framework and the disulfide pattern).7,8 Disulfide-
rich peptides interact with a range of molecular targets
including ion channels, membrane transporters, and receptors
with exquisite potency and selectivity for their molecular
target.4,9 One reason for the potency and selectivity of these
peptides is that the disulfide bonds provide a stable framework
on which specific residues within the intercysteine loops can be
incorporated to maximize the interaction with the molecular
target.
Received: March 8, 2015
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.5b00210
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
disulfide framework is used as a template to generate a
“combinatorial library” of peptides, even within the same
species.9,14,15 The neopetrosiamides, isolated from a Neopetrosia
sp. sponge, like asteropine A and the asteropsins also contain
six cysteines, but they form an alternate disulfide arrangement,
CysI-V, CysII-IV, and CysIII-VI.16 These peptides are potent
inhibitors of amoeboid invasion by human tumor cells17 via the
disruption of cell adhesion through loss of cell surface
integrins.18 Chemical synthesis of neopetrosiamide A using
solid-phase peptide synthesis (SPPS) methods and an
orthogonal cysteine protection strategy was subsequently used
to determine the disulfide connectivity of the peptide.16 Being
able to produce neopetrosiamide A synthetically has also
facilitated structure/activity studies, which revealed the key role
that the disulfide bonds play in defining the bioactive
conformation of neopetrosamide A, as alternate disulfide
isomers of this peptide were biologically inactive.19
The marine sponge Geodia barretti Bowerbank (family
Geodiidae, order Tetractinellida, class Demospongiae), a
cold-water marine sponge found on the north Atlantic
continental shelf, has been found to produce a wide range of
secondary metabolites including several cyclic dipeptides
derived from tryptophan and arginine, known as the bare-
ttins.20−22 The barettins have recently been shown to act in
synergy to deter larvae of surface settlers,23 and field trials using
barettins incorporated into paint were shown to prevent surface
fouling by barnacles (Balanus improvisus) and blue mussels
(Mytilis edulis).24 More recently, barettin was shown to be an
antioxidant and also inhibited secretion of several pro-
inflammatory cytokines from LPS-stimulated THP-1 cells.25
In this study we describe the isolation and characterization of
two novel disulfide-containing peptides from G. barretti, which
we have named barrettides A (1) and B (2). We further
characterized 1 by determining its three-dimensional structure
using NMR spectroscopy and also developed an efficient
approach to chemically synthesize these peptides, which can be
used for future structure/activity studies. Finally, we describe
the antibacterial and antifouling properties of both 1 and 2.
Figure 1. RP-HPLC traces of the isolation and separation of peptides
from an extract of G. barretti. (A) The previously described barettins
eluted after ∼28 min, whereas the two novel peptides barrettides A (1)
and B (2) eluted after ∼30 min. The peptide displaying a [M + 2H]2+
ion of 1619 termed Gb3238 eluted after ∼32 min. The peptide with an
[M + 2H]2+ ion of 1648 was not detected in these initial isolation
studies. RP-HPLC purification was performed on a 1.5% buffer B/min
gradient, and absorbance was measured at 215 nm. (B) RP-HPLC
trace after size exclusion chromatography showing the peak
comprising 1 and 2 eluting after 30 min. RP-HPLC purification was
performed on a 0.5% buffer B/min gradient starting at 25% buffer B.
Absorbance was measured at 215 nm.
The amino acid content of 1 and 2 was quantitatively
determined after 24 h hydrolysis with 6 M HCl to assist the
sequencing and indicated the presence of six Asp/Asn, three
Thr, three Ser, three Pro, three Val, two Glu/Gln, two Gly, two
Ala, one Phe, one Lys, and one Arg residue for 1. Amino acid
analysis of 2 revealed it contained the same amino acid content
as 1 except there were only two Val residues and an additional
Leu residue. Furthermore, peptides were treated with
dithiothreitol (DTT) to break disulfide bonds and subsequently
alkylated using iodoacetamide (IAM) to block reactive cysteine
residues. Following reduction and alkylation, an increase in 232
Da was observed for both peptides, corresponding to alkylation
of four cysteines involved in two disulfide bonds. Hence, the
monoisotopic masses can then be calculated to be 3212.2791
and 3226.2948 Da for 1 and 2, which corresponds to deviations
of 33 and 38 ppm, respectively, from their experimentally
determined masses. In our hands, this is within the normal
range of the instrument used (a Waters Qtof Micro)
considering the size of the molecules.
The reduced and alkylated peptides were cleaved with
various enzymes, including trypsin, chymotrypsin, and
endoproteinase Glu-C, in order to elucidate their sequences.
These cleavage mixtures were analyzed using LC-MS-MS for
sequence determinations. The major fragments and their
sequence sections for both peptides are shown in Figure 2.
Fragments discovered from trypsin digestion revealed major
fragments with monoisotopic masses of 1800.4 Da for both
peptides, which were sequenced to be DVSPCFCV-
RESULTS AND DISCUSSION
■
A specimen of the cold-water marine sponge G. barretti,
obtained from the Swedish Kosterfjord, was extracted with
CH3CN/H2O, and compounds were purified extensively by
RP-HPLC and size-exclusion chromatography, as shown in
Figure 1. Mass spectrometry analysis of the extract revealed two
major peaks corresponding to two peptides present in all three
extraction steps, termed barrettides A (1) and B (2). Mass ions
1607.4 and 1072.2 correspond to [M + 2H]2+ and [M + 3H]3+,
respectively, for 1, and mass ions 1614.4 and 1076.6 correspond
to [M + 2H]2+ and [M + 3H]3+, respectively, for 2. After RP-
HPLC (first purification step) and size-exclusion chromatog-
raphy (second purification step) the recovered yield of the
peptides was 2.3% from the extract. The recovered yield for
purified peptides based on RP-HPLC recovery (third
purification step) was 2.2% and 1.0% (from the extract) for 1
and 2, respectively. In total, four novel peptides with [M +
2H]2+ ions of 1607 (1), 1614 (2), 1619, and 1648 were
extracted and isolated. The latter two were present in small
quantities, so further studies focused on barrettides A and B.
The monoisotopic masses (M) for 1 and 2 were determined to
be 3212.3862 and 3226.4175, respectively, by deconvolution of
their [M + 3H]3+ ions.
B
DOI: 10.1021/acs.jnatprod.5b00210
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
NMR data were recorded at 298 K for extracted and purified
1 and 2. Homonuclear spectra were assigned using the
sequential assignment procedure.29 Briefly TOCSY and DQF-
COSY data were used to identify individual spin systems, and
NOESY data were used to link neighboring residues in the
sequence. The fingerprint region of the NOESY spectrum is
shown in Supplementary Figure 1. Sequential αHi−NHi+1
NOEs were observed for the entire peptide chains, except at
Pro4, Pro20, and Pro31, because these residues lack amide
protons. Pro4 and Pro31 were found to be in a trans
configuration, as indicated by intense Hα−Hδ NOEs between
the preceding residue and the Pro, whereas Pro20 was in a cis
configuration, as indicated by intense Hα−Hα NOEs between
residue 19 and Pro20. The NMR spectra recorded showed
single spin systems for each residue, except for residues 29−31,
where a second minor conformation was apparent due to cis−
trans isomerization of Pro31.
Secondary Hα NMR chemical shifts are the difference
between the observed chemical shift and the corresponding
random coil value and are strong indicators of the presence of
secondary structure in peptides.30 Analysis of the measured
chemical shifts for both peptides revealed two series of positive
secondary Hα shifts between residues 6−9 and 16−19, as
shown in Figure 3, which is indicative of β-sheet regions. The
Figure 2. Sequences of peptides isolated from the cold-water sponge
G. barretti. The sequences of barrettides A (1) and B (2) were
determined using MS/MS sequencing of peptide fragments from
enzymatic digests in combination with amino acid analysis. Sequence
fragments identified from the trypsin, Endo-GluC, and chymotrypsin
digests are shown. Barrettides A and B differ only at position 19, where
1 has a Val, whereas 2 has a Leu. Both peptides have four Cys residues
that form two disulfide bonds between residues Cys5−Cys23 and
Cys7−Cys18. Cys residues are highlighted in gray, and the sequence
difference at position 19 between the two peptides is highlighted by a
box.
EDETSGAK. Additionally, two fragments with monoisotopic
masses of 1294.5 and 1308.5 Da were sequenced to be
TCVPDNCDASR and TCLPDNCDASR for 1 and 2,
respectively. Additional fragments from Glu-C digestion were
sequenced to be DVSPCFCVEDE for both peptides and
TSGAKTCVPDNCDASRGTNP for 1 and TSGAKTCLPDN-
CDASRGTNP for 2. Moreover, chymotrypsin digestion
resulted in a fragment with a monoisotopic mass of 1901.4,
which was sequenced to be DVSPCFCVEDETSGAKT for
both peptides. Chymotrypsin cleavage data also helped in
determining the order of the last four residues at the C-
terminus of each peptide to be GTNP.
These sequence fragments in combination with the amino
acid quantification data yielded the complete sequences of the
two peptides to be determined as DVSPCFCVED-
ETSGAKTCVPDNCDASRGTNP and DVSPCFCVED-
ETSGAKTCLPDNCDASRGTNP for 1 and 2, respectively,
as shown in Figure 2. NMR spectroscopy was also used to
confirm the sequences. The sole difference between the
sequences of the two peptides is at position 19, where 1 has
a valine, whereas 2 has a leucine residue. A search to identify
peptides related to the barrettides using both tBLASTN and
BLASTp26 revealed no other identical or similar peptide
sequences in the NCBI databases. A recent study27 reported the
metatranscriptome (complete set of gene transcripts) of G.
barretti. Using Ugene,28 the peptide sequences of 1 and 2 were
matched against the G. barretti transcriptome. Surprisingly, only
one hit, DVSPCFCVEDETSGAKTCVPDLSGLAR, shared
similar sequence sections to the novel peptides found in this
study, which was reported to be 82% and 78% for 1 and 2,
respectively. Its monoisotopic mass does not match one of the
low-abundant peptides that was isolated from the sponge. The
hit matches 1 from Asp1−Asp21, but differs in the C-terminal
region including missing the fourth cysteine, which is required
if two disulfide bonds are to be formed. It is possible that
differences in time and location of specimen collection between
the sponge sample used in this study and the specimen of G.
barretti used for the metatranscriptomics study, which was
collected off the Norwegian west coast,27 could explain the
discrepancy between our study and the transcriptomic data. In
addition to that possibility of variation of gene expression, it
cannot be ruled out that this particular sequence escaped
sequencing.
Figure 3. Secondary Hα chemical shifts for barrettides A (1) and B
(2). Their sequences are indicated at the bottom of the figure, aligned
with their corresponding point in the traces. Barrettide A (1) is
represented in black bars, whereas Barrettide B (2) is represented in
white bars. The trend in secondary Hα shifts is identical, indicating
that their secondary structures will be very similar, although there are
some clear differences at positions 19 and 20, due to the different
amino acids at position 19, where 1 has a Val and 2 has a Leu.
NH of Asp1 of 2 could not be observed, as the free amine of
the N-terminal residue is rapidly exchanging with the solvent.
However, Asp1 Hα and Hβ protons of 1 could be located in
the 1H−13C HSQC spectrum of this peptide. The NMR
secondary shift values for 1 and 2 follow the same pattern with
some small localized variations at positions 19−20 due to the
nature of the different amino acids at position 19. Overall, this
indicates that they have identical three-dimensional structures
and perhaps similar bioactivities.
The NMR spectroscopy data were used to determine the
three-dimensional structure of 1. Structural restraints (Table 1)
included interproton distances generated from NOE intensities,
and backbone dihedral angle restraints (ψ and φ angles) were
obtained from the angles predicted by TALOS-N analysis using
the HN, Hα, N, Cα, and Cβ chemical shifts.31,32 The χ1
dihedral angle conformations and stereospecific assignments of
Hβ methylene pairs were determined from intraresidue NOE
patterns derived from a 100 ms NOESY spectrum together
with 3JHα−Hβ coupling patterns from the DQF-COSY spectrum.
Cys7 and Cys18 were found to have χ1 dihedral angles of −60°,
C
DOI: 10.1021/acs.jnatprod.5b00210
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
data. These eight hydrogen bonds were identified as follows
based on structure calculations: Phe6−Val19 O, Val8−Thr17
O, Thr12−Asp10 OD1/2, Gly14−Asp10 O, Ala15−Asp10 O,
Thr17−Val8 O, Val19−Phe6 O, and Cys23−Pro20 O. NMR
analysis revealed strong NOE cross-peaks between Cys5 and
Cys23 and between Cys7 and Cys18, suggesting the disulfide
connectivity to be 5−23 and 7−18. The Cys7−Cys18 disulfide
bond is situated between two strands in a β-sheet, and a strong
Hα−Hα peak was observed between them. To further confirm
the disulfide connectivity, a set of structures with the two other
possible disulfide connectivities was calculated. The resulting
higher energies and more restraint violations from the
calculated structures with the two other possible disulfide
connectivities, i.e., Cys5−Cys18 and Cys7−Cys23, and Cys5−
Cys7 and Cys18−Cys23, showed that the Cys5−Cys23 and
Cys7−Cys18 connectivity was structurally favorable, which was
further supported by the consistency between experimentally
determined χ1 dihedral angles and the conformation of the
disulfide bonds.
Table 1. Energies and Structural Statistics for the Family of
the 20 Lowest Energy Structures with Highest MolProbity
Scores for Barrettide A (1)
Energies (kcal/mol)
overall
−964.7 13.26
14.31 0.281
38.21 1.00
bonds
angles
improper
van der Waals
NOE
15.45 0.729
−73.21 1.59
0.084 0.004
0.275 0.055
133.7 0.25
cDih
dihedral
electrostatic
−1093.0 13.4
MolProbity Statistics
clashes (0.4 Å/1000 atoms)
poor rotamers
7.248 0.6556
0.185 0.185
1.035 0.440
94.14 0.97
1.728 0.0497
86.90 1.73
0.00 0.00
Ramachandran outliers
Ramachandran favored
a
MolProbity score
A set of 50 structures was calculated, and the best 20
structures based on lowest overall energies and good covalent
geometries judged by MolProbity scores were chosen to
represent the solution structure of 1 (Figure 4). Structural
b
MolProbity score percentile
residues with bad bonds
residues with bad angles
0.00 0.00
c
Atomic rmsd
mean global backbone
mean global heavy
0.92 0.34 Å
1.58 0.35 Å
Distance Restraints
intraresidue (i − j = 0)
sequential (|i − j| = 1)
67
107
medium range (1 < |i − j| < 5)
long range (|i − j| ≥ 5)
hydrogen bonds
39
60
16 (for 8 H-bonds)
289
total
Dihedral Angle Restraints
ϕ
18
10
28
χ1
total
Violations from Experimental Restraints
total NOE violations exceeding 0.2 Å
1 (0.226 Å)
0
total dihedral violations exceeding 3.0°
a
MolProbity score combines the Clashscore, rotamer, and Ramachan-
dran evaluations into a single score, normalized to be on the same
b
scale as X-ray resolution. 100th percentile is the best among
structures of comparable resolutions; 0th percentile is the worst.
c
Atomic rmsd values are for residues 5−23.
Cys5, Asp10, and Asp21 have χ1 dihedral angles of 180°, and
Cys23 has a χ1 dihedral angle of +60°. Additionally, another
four χ1 angles were incorporated into the structure calculations
based on the TALOS-N analysis. Amide protons involved in
hydrogen-bonding interactions were identified by deuterium
exchange experiments and temperature coefficient data. The
slow exchange of the amide protons of Val8, Thr17, and Val19
with deuterated water suggested that these three residues were
involved in hydrogen bonding. Moreover, amide protons with
temperature coefficients > −4.6 ppb/K are considered as being
protected from the solvent and are thus probably engaged in
hydrogen bonds.33 The amide protons of residues Phe6, Val8,
Thr12, Gly14, Ala15, Thr17, Val19, and Cys23 exhibited
temperature insensitivity indicative of hydrogen-bonding
involvement. In total, 16 restraints for eight hydrogen bonds
were included in the NMR structure determination based on
analysis of slow exchanging amides and temperature coefficient
Figure 4. Three-dimensional structure of barrettide A (1) determined
using NMR spectroscopy. (A) Superimposition of the peptide
backbone from Cys5 to Cys23 of the 20 minimum energy structures
of 1 with the disulfide bonds between Cys5−Cys23 and Cys7−Cys18
highlighted in yellow. (B) Ribbon representation of 1 illustrating the
β-hairpin structure of the peptide, which is cross braced by the two
disulfide bonds shown in ball-and-stick format (yellow). The Cys
residues and the termini are labeled. (C) Surface representation of the
lowest energy structure of 1. The coloring of the surface distinguishes
negatively charged (red), positively charged (blue), hydrophobic
(green), polar (white), cysteine (yellow), and glycine (cyan) residues.
The individual amino acids within the sequence are also labeled. The
structures were determined using NMR spectroscopy, and the figures
were created in MOLMOL43 and PyMOL.
D
DOI: 10.1021/acs.jnatprod.5b00210
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
statistics for the structure ensemble are summarized in Table 1
and demonstrate that the structure is of high quality and in
good agreement with both experimental data and covalent
geometry. Analysis of the structure identified two antiparallel β-
sheets comprising residues 6−10 and 15−19 forming a β-
hairpin that is cross braced by two disulfide bonds at the base of
the hairpin (Figure 4B). A tight turn occurs after the second β-
strand, in which Pro20 adopts a cis configuration. This tight
turn is presumably to allow for the ladder-like arrangement of
the disulfide bonds and the difference in spacing between Cys5
and Cys7 (one residue) and Cys18 and Cys23 (four residues).
As evident from the overlay of the ensemble of 1 structures,
both the termini are disordered and appear to be flexible in
solution, consistent with secondary Hα chemical shifts close to
random coil values. The structural topology of 1, a β-hairpin
fold cross braced by two disulfide bonds in a ladder-like
arrangement, is also seen in the analgesic conotoxin MrIA.34
However, MrIA is only a 13-residue peptide, and therefore the
β-hairpin is much shorter than that seen in 1. Analysis of the
surface distribution of residues reveals a clustering of
hydrophobic residues consisting of Val2, Pro4, Phe6, Val8,
V19, and P20 on one face of the molecule (Figure 4C). There
is also a series of residues, Glu9, Glu11, Lys16, and Asp24,
forming a charged region on the opposite face of the molecule
from the hydrophobic surface, which suggests 1 has an
amphipathic nature.
Peptide Synthesis and Oxidative Folding. In our initial
attempts to synthesize 1 by manual 9H-fluoren-9-ylmethox-
ycarbonyl (Fmoc)-SPPS, peptide purity rapidly decreased after
assembling the first half of the peptide sequence. Analysis of the
crude peptide demonstrated poor coupling efficiencies after
reaching the third cysteine (Cys18). Several factors may explain
the poor efficiency: the length of the peptide chain, the
presence of amino acids with β-branched side chains, and the
tendency of peptides to undergo aggregation or β-sheet
formation during synthesis. With the structure in hand, the
poor yield is likely connected to the propensity of β-sheet
formation.
In our case, we changed to microwave-assisted Fmoc-SPPS
to facilitate coupling, and we also incorporated a pseudoproline
dipeptide at the Glu11−Thr12 position to reduce peptide β-
sheet formation during synthesis. Using this method, the
peptide was synthesized on chlorotrityl chloride resin using
microwave-heated coupling and deprotection steps at a
temperature of 60 °C. This temperature is 10 °C lower than
the one used in our standard microwave−SPPS protocol, to
prevent premature cleavage of the peptide from the resin. N,N′-
Diisopropylcarbodiimide (DIC) was used as the coupling agent
together with the base ethyl (hydroxyimino)cyanoacetate
(OxymaPure) in place of hydroxybenzotriazole (HOBT),
which has been commonly used in the past.35 The combination
of DIC and OxymaPure as the coupling agent assisted by
microwave heating gave a crude yield of 1 of 77.6%. The main
side product was a peptide with an additional +56 Da mass that
most likely resulted from an uncleaved trityl protecting group.
The purified reduced 1 peptide was then oxidized under
optimized folding conditions for 48 h and repurified on RP-
HPLC. Folded 1 was identified by MS, and the oxidized
peptide coeluted with native peptide on analytical RP-HPLC,
confirming that the native disulfide connectivity was maintained
in the synthetic peptide (Figure 5). A comparison of the Hα
chemical shift values of synthetic and native 1 also confirmed
that the correct disulfide isomer was obtained (Figure 5D).
Figure 5. Comparison of native and synthetic barrettide A (1). (A)
RP-HPLC chromatogram of synthetic 1 and mass spectrometric data
(inset) confirming the expected molecular weight. (B) RP-HPLC
chromatogram of native 1 including mass spectrum (inset). (C) RP-
HPLC of a 1:1 mixture of synthetic and native 1 showing coelution of
the two peptides. (D) Comparison of Hα chemical shifts of synthetic
(solid squares) and native (open circles) of 1 in H2O/D2O (90:10) at
298 K.
Together, the total synthesis and NMR studies of barrettide A
demonstrate unambiguous support of the sequence as
determined by quantitative amino acid analysis, MS, MS-MS,
and transcriptome analyses.
Biological Activity. As the barettins from G. barretti have
previously been shown to possess antifouling properties, both 1
and 2 were tested for their effect on the settlement of barnacle
larvae of B. improvisus. Both peptides were tested at three
concentrations (0.06, 0.6, and 6 μM), and the number of
settled and metamorphosed cyprids, living cyprids, and dead
cyprids were counted after 4 days (Figure 6). Barretide A (1)
significantly inhibited the settlement of larvae at both 0.6 and 6
μM (Figure 6A), whereas 2 inhibited settlement only at 6 μM
(Figure 6B). These potencies of the barrettides are comparable
to both barettin (EC50 = 0.9 μM) and 8,9-dihydrobarettin
(EC50 = 7.9 μM)21 but are lower than bromobenzisoxazolone
barettin (EC50 = 15 nM)22 and the cyclic disulfide-rich peptide
cycloviolacin O2 (EC50 = 0.25 μM).36 It would be interesting to
assess the antifouling effect of the barrettides in combination
with the barettins, as it has previously been demonstrated that a
E
DOI: 10.1021/acs.jnatprod.5b00210
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
EXPERIMENTAL SECTION
■
General Experimental Procedures. RP-HPLC was performed
using an Agilent 1100 binary HPLC system (Santa Clara, CA, USA)
with UV detection at 215 or 230 nm for analytical and preparative
HPLC, using Phenomenex (Torrance, CA, USA) Jupiter C18 columns
[250 × 4.6 (i.d.), 5 μm, 300 Å, and 250 × 21.2 (i.d.) mm, 10 μm, 300
Å] respectively. Direct infusion MS was done on a Thermo-Finnigan
LCQ Deca ion trap MS (San Jose, CA, USA) in the positive ESI mode.
LC-MS and LC-MS-MS for sequencing were done in positive mode
on a Waters QTof Micro (Waters, Milford, MA, USA) fed by a Waters
NanoAcquity UPLC. Solution NMR data were recorded on a Bruker
Avance 600 MHz spectrometer equipped with a cryoprobe. Microwave
SPPS was performed on a CEM Liberty One automated microwave
synthesizer (Matthews, NC, USA). Quantitative determinations of the
amino acid content of the peptides were performed at the Amino Acid
Analysis Centre, Department of Biochemistry and Organic Chemistry,
Uppsala University. Resins, amino acids, and reagents for SPPS were
obtained from Iris Biotech GmbH (Marktredwitz, Germany).
Sponge Material. Barrettides were isolated from a specimen of
Geodia barretti Bowerbank collected in the Swedish Kosterfjord
(58°52′ N, 11°5′ E) followed by homogenization and lyophilization.
́
Dr. Paco Cardenas identified the sample, and vouchers with the
reference code MO20GB are kept at the Division of Pharmacognosy
and deposited at the Museum of Evolution Herbarium, Uppsala
University.
Figure 6. Effects of barrettide A and barrettide B on the settlement of
cyprid larvae of Balanus improvisus. In the control, 64
8% of
barnacles settled (black bars), 19 4% were alive but had not settled
(gray bars), and 17 5% were found dead (white bars), after 4 days of
incubation in seawater. (A) Results of incubation of B. improvisus
larvae including barrettide A in the seawater (0.06, 0.6, or 6 μM): a
slight but statistically significant inhibition of settlement can be seen
already at 0.6 μM, and at a concentration of 6 μM only 15 2% of
larvae settled. For barrettide B (B) statistically significant inhibition
was demonstrated at 6 μM, at which 8 2% had settled. Notably, the
percentage of dead larvae decreased at higher concentrations of
barrettides. Data are given as means SEM (n = 5, *p < 0.05, ***p <
0.001 versus control).
Extraction and Isolation of Peptides. A three-step extraction
protocol with different ratios of CH3CN in H2O was employed. The
first step involved incubation of the sponge in 60% CH3CN/40%
H2O/0.1% formic acid at a concentration of 100 g/L overnight at
room temperature (rt). The second step used 30% CH3CN/70%
H2O/0.1% formic acid, and the third and last extraction step was with
H2O/0.1% formic acid. Supernatants were diluted to a final
concentration of 10% CH3CN, filtered, and then purified by RP-
HPLC (first purification step) on a C18 column using a gradient of 0−
60% B (buffer A: H2O/0.05% TFA; buffer B: 60% CH3CN/40%
H2O/0.045% TFA) in 60 min at a flow rate of 10 mL/min.
Electrospray ionization-mass spectroscopy (ESIMS) confirmed the
presence and molecular mass of peptides in each extraction step.
Fractions containing peptides were subject to size-exclusion
chromatography using Sephadex G-15 (second purification step) to
remove low molecular weight compounds, including the barettins on a
mobile phase of 60% CH3CN/40% H2O/0.05% TFA. Lyophilized
fractions were dissolved in ∼3 mL of the mobile phase and injected
into the HPLC operated at a flow rate of 0.6 mL/min. The high
molecular weight fractions containing peptide(s) were repurified by
RP-HPLC (third purification step) using a gradient of 25−50% buffer
B in 55 min at a flow rate of 1 mL/min. Analytical RP-HPLC and
ESIMS confirmed the purity and molecular masses of the peptides.
Sequence Analysis. For amino acid sequence analysis, the
peptides were reduced with dithiothreitol in 0.25 M Tris-HCl
containing 1 mM EDTA and 6 M guanidine-HCl (pH 8.5) for 2−3
h at 37 °C and subsequently alkylated by the addition of 50 mg of
iodoacetamide. The reaction was stopped after 2−10 min by adding
0.5 M citric acid and subsequently desalted using solid-phase
extraction on C18 columns. Samples were diluted to a final
concentration of 10% CH3CN before LC-MS-MS analysis. The
reduced and alkylated peptides were cleaved by several enzymes,
including trypsin, chymotrypsin, and endoproteinase Glu-C in 50 mM
NH4HCO3 by incubating overnight at 37 °C. The sequence analysis
was performed using nanospray MS-MS sequencing based on the
presence of b and y-series ions (N- and C-terminal fragments).
Together with the amino acid quantification data, the sequence orders
for the peptides were confirmed.
mixture of barettin and 8,9-dihydrobarettin acts synergistically
to inhibit larval settlement.23 Importantly, both barrettides were
not toxic, as the number of dead larvae in the peptide
treatments was the same as for controls after 24 h.
Many antimicrobial peptides are amphipathic in nature.
However, despite having clustered regions of hydrophobic and
charged residues, the barrettides showed no antibacterial
activity against the Gram negative Escherichia coli or the
Gram positive Staphylococcus aureus bacteria. No inhibition was
detected up to a concentration of 80 μM, which is a typical
cutoff value for antimicrobial peptides classifying them as
inactive. In comparison, the human antimicrobial peptide LL-37
used as a control showed normal inhibitory concentrations (1−
2 μM). Despite having amphipathic character, the charged
residues in 1 and 2 are predominantly negative, resulting in a
−4 net charge. The bacterial cell wall membrane has a high
density of anionic groups and, hence, interacts better with
cationic peptides, which explains why most antimicrobial
peptides are positively charged.
In conclusion, we have characterized two disulfide-containing
peptides, barrettides A (1) and B (2), from the cold-water
marine sponge G. barretti. They have a unique three-
dimensional structure when compared to previously described
disulfide-containing peptides and expand on the diversity of this
class of peptides described from marine sponges. Furthermore,
we have demonstrated that the barrettides are amenable to
peptide synthesis, which will facilitate future work on
understanding the structure/activity relationships of this
interesting family of peptides.
NMR Spectroscopy. Samples were dissolved in 90% H2O/10%
D2O (pH 5.15) or 100% D2O to a final concentration of 1.2 mM and
0.62 mM for 1 and 2, respectively. 2,2-Dimethyl-2-silapentane-5-
sulfonate (DSS) was added as an internal chemical shift reference (0.0
ppm). Two-dimensional NMR spectra were recorded at 298 K, and at
283−313 K to collect temperature coefficient data, and processed by
F
DOI: 10.1021/acs.jnatprod.5b00210
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Topspin 2.1 (Bruker). Data sets included homonuclear DQF-COSY
(16 scans), TOCSY (4 scans) with a mixing time of 80 ms, and
NOESY (48 scans or 64 scans) with mixing times of 100−200 ms.
Spectra were recorded with a sweep width of 11.97 ppm and a
resolution of 4k data points in the f2 dimension and 512 increments in
C
18 column using a gradient of 0−80% B (buffer A: H2O/0.05% TFA;
buffer B: 90% CH3CN/10% H2O/0.045% TFA) over 80 min. ESIMS
was used to confirm the molecular mass of the synthesized peptide.
Formation of the disulfide bonds was achieved by incubation of the
reduced peptide (0.2 mg/mL) in a buffer consisting of 0.1 M
NH4HCO3, 20 mM reduced glutathione, and 4 mM oxidized
glutathione (pH 8.5) for 48 h at 23 °C with constant stirring. The
folded peptide was purified by RP-HPLC as described previously, and
its molecular weight confirmed by ESIMS. Co-injection on RP-HPLC
and NMR data analysis were used to compare synthetic and isolated 1.
Antimicrobial Testing. E. coli (ATCC 25922) and S. aureus
(ATCC 29213) were grown to mid logarithmic phase in 3% tryptic
soy broth at 37 °C and then washed twice by centrifugation and
resuspension in 10 mM Tris buffer, pH 7.4. Serial dilutions of the two
peptides (0.16 to 80 μM) were prepared in 96-well tissue culture
plates to which 50 000 bacteria/well (assessed by OD600) were added,
giving a total volume of 100 μL. After 5 h incubation at 37 °C, 5 μL of
20% tryptic soy broth was added to each well. Plates were reincubated
for 5 h or longer to generate observable microbial growth. Negative
controls lacked peptide or bacteria, and the benchmark antimicrobial
peptide LL-37 was used as a positive control, showing its typical 1−2
μM minimum inhibitory concentration. This MIC assay is designed to
minimize interference by polyelectrolytes and amphiphilic compounds
in the growth media on antimicrobial peptide activity.45 Both peptides
were tested three times (n = 3).
the f1 dimension. In addition H−15N and H−13C HSQC data sets
were recorded at natural abundance using 256 scans/128 increments
and 120 scans/256 increments, respectively. For detection of slow-
exchanging amides, a series of one-dimensional and TOCSY spectra
were recorded after immediately dissolving the peptide in D2O.
Secondary shifts were calculated using the random coil values reported
by Wishart et al.37
1
1
Structure Determination. Spectroscopic data were analyzed
using CARA38 and CCPNMR.39 Spectra were assigned using the
sequential assignment protocol,29 and HSQC data were assigned based
1
on the H chemical shifts. NOESY spectra recorded at 298 K with a
mixing time of 200 ms in both 90% H2O/10%D2O and 100% D2O
were used to derive interproton distance restraints. All peaks were
integrated, assigned, and calibrated automatically into distance
restraints using CYANA.40 Backbone dihedral angle restraints (phi
and psi) were derived from a TALOS-N analysis31 of HN, Hα, N, Cα,
and Cβ chemical shifts. The χ1 dihedral angle conformations and
stereospecific assignments of HB methylene pairs were determined
from intraresidue NOE patterns derived from a 100 ms NOESY
3
spectrum together with JHα−Hβ coupling patterns from the DQF-
Antifouling Testing. The settlement propensity and mortality
COSY spectrum. Initial structures were generated using Cyana 3.0,
and the final structures were calculated and refined in explicit water
within CNS 1.2.41 From the final round of calculation, 20 structures of
a total of 50 were chosen to represent the solution structure of 1,
based on the lowest overall energies and good covalent geometries.
Structures were analyzed using MolProbity,42 and figures prepared
using MOLMOL.43
were used to evaluate the biological activity of 1 and 2, as described by
Sjogren et al.21,46 Statistical analysis was performed using an unpaired
̈
Student’s t-test, and differences were considered significant if p < 0.05.
ASSOCIATED CONTENT
■
S
* Supporting Information
Peptide Synthesis and Folding. Barettide A (1) was assembled
on 2-Cl-trityl chloride (1.6 mmol/g, Iris Biotech) resin using a
combination of manual SPPS44 and microwave-assisted synthesis. The
first Nα-Fmoc-protected amino acid, Pro31 (1.2 equiv), was coupled
to the resin (312.5 mg, 1.6 mmol/g) in the presence of N,N-
diisopropylethylamine (DIPEA; 4 equiv) and CH2Cl2 for (2 × 1.5 h at
rt). The resin was then washed with CH2Cl2 (2 × 10 mL), and the
coupling step repeated. The resin was successively washed with a
mixture of CH2Cl2/MeOH/DIPEA, 17:2:1 (3 × 5 mL), CH2Cl2 (3 ×
5 mL), dimethylformamide (DMF; 2 × 5 mL), and CH2Cl2 (2 × 5
mL) and then dried overnight under vacuum. The loading of the resin
(0.41 mmol/g) was determined by absorbance measurement at 304
nm for the dibenzofulvene−piperidine adduct formed during a
standard Fmoc test. Barettide A (1) was synthesized on a 0.1 mM
scale with Fmoc chemistry. Solutions of 1 M DIC and 2 M
OxymaPure in DMF (filtered) were prepared separately and used for
in situ coupling. The Fmoc deprotection was carried out using 20%
piperidine (20 min), and the next three C-terminal amino acids Asn30,
Thr29, and Gly28 (5 equiv) were coupled using Oxymapure/DIC (10
equiv:10 equiv) at 23 °C (30 min × 2). Then the synthesis was
continued until Ser13 was on the synthesizer using previously specified
microwave-assisted deprotection (30 s at 60 °C, 3 min at 60 °C) and
standard coupling (60 °C, 5 min) parameters.35 Fmoc-Arg was
coupled using modified conditions (25 min at 23 °C × 1, 5 min at 60
°C × 1). To facilitate assembly of the peptide chain, the
psesudoproline dipeptide Fmoc-Glu(OtBu)-Thr(ΨMe,Me pro)-OH (5
equiv) was incorporated at positions 11 and 12 using DIC/
OxymaPure (10 equiv:10 equiv) as coupling agent (30 min × 2 at
23 °C). The remaining amino acids were then coupled by microwave
assistance using double-coupling cycles (5 min at 60 °C × 2). Final
crude peptide was obtained in 43% yield based on the increase in resin
weight.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.5b00210.
AUTHOR INFORMATION
■
Corresponding Author
*Tel: +46 18 4715031. Fax: +46 18 509101. E-mail: ulf.
goransson@fkog.uu.se.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
B.B.C. is a holder of a University of Queensland Graduate
School International Travel Award. R.J.C. is an ARC Future
Fellow (FT100100476). Research at the Division of
Pharmacognosy is supported by EU FP7 BlueGenics
(311848; L.B.) and the Swedish Research Council (2012-
́
5063; 2011-3403; U.G.). We thank Dr. P. Cardenas for
identification of Geodia barretti and support with bioinfor-
matics; Mr. E. Jacobsson for support with bioinformatics; and
Dr. A. A. Stromstedt for help with antibacterial assays.
̈
REFERENCES
■
(1) Lazcano-Perez, F.; Roman-Gonzalez, S. A.; Sanchez-Puig, N.;
Arreguin-Espinosa, R. Protein Pept. Lett. 2012, 19, 700−707.
(2) Vetter, I.; Lewis, R. J. Curr. Top. Med. Chem. 2012, 12, 1546−
1552.
(3) 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.; Cotter, P. D.; Craik, D. J.; Dawson, M.; Dittmann, E.;
Donadio, S.; Dorrestein, P. C.; Entian, K. D.; Fischbach, M. A.;
Garavelli, J. S.; Goransson, U.; Gruber, C. W.; Haft, D. H.;
Hemscheidt, T. K.; Hertweck, C.; Hill, C.; Horswill, A. R.; Jaspars,
M.; Kelly, W. L.; Klinman, J. P.; Kuipers, O. P.; Link, A. J.; Liu, W.;
The peptides were cleaved from resin using trifluoroacetic acid
(TFA) with triisopropylsilane (TIPS) and H2O as scavengers
(9.5:2.5:2.5 TFA/TIPS/H2O) at rt for 2 h. TFA was dried down to
0.5 mL with nitrogen, and the peptide was precipitated in ice-cold
ether, filtered, and dissolved in 50% CH3CN containing 0.05% TFA,
then lyophilized. The crude peptides were purified by RP-HPLC on a
G
DOI: 10.1021/acs.jnatprod.5b00210
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Marahiel, M. A.; Mitchell, D. A.; Moll, G. N.; Moore, B. S.; Muller, R.;
Nair, S. K.; Nes, I. F.; Norris, G. E.; Olivera, B. M.; Onaka, H.;
Patchett, M. L.; Piel, J.; Reaney, M. J.; Rebuffat, S.; Ross, R. P.; Sahl, H.
G.; Schmidt, E. W.; Selsted, M. E.; Severinov, K.; Shen, B.; Sivonen,
K.; Smith, L.; Stein, T.; Sussmuth, R. D.; Tagg, J. R.; Tang, G. L.;
Truman, A. W.; Vederas, J. C.; Walsh, C. T.; Walton, J. D.; Wenzel, S.
C.; Willey, J. M.; van der Donk, W. A. Nat. Prod. Rep. 2013, 30, 108−
160.
(34) Nilsson, K. P.; Lovelace, E. S.; Caesar, C. E.; Tynngard, N.;
Alewood, P. F.; Johansson, H. M.; Sharpe, I. A.; Lewis, R. J.; Daly, N.
L.; Craik, D. J. Biopolymers 2005, 80, 815−823.
(35) Friligou, I.; Papadimitriou, E.; Gatos, D.; Matsoukas, J.; Tselios,
T. Amino Acids 2011, 40, 1431−1440.
(36) Goransson, U.; Sjogren, M.; Svangard, E.; Claeson, P.; Bohlin, L.
̈
̈
̊
J. Nat. Prod. 2004, 67, 1287−1290.
(37) Wishart, D. S.; Sykes, B. D.; Richards, F. M. Biochemistry 1992,
31, 1647−1651.
(4) Castaneda, O.; Harvey, A. L. Toxicon 2009, 54, 1119−1124.
(38) Keller, R. L. J. The Computer Aided Resonance Assignment
Tutorial, 1st ed.; Cantina Verlag: Goldau, 2004.
(5) Shiomi, K. Toxicon 2009, 54, 1112−1118.
(6) Norton, R. S. Toxicon 2009, 54, 1075−1088.
(39) Vranken, W. F.; Boucher, W.; Stevens, T. J.; Fogh, R. H.; Pajon,
A.; Llinas, M.; Ulrich, E. L.; Markley, J. L.; Ionides, J.; Laue, E. D.
Proteins: Struct., Funct., Genet. 2005, 59, 687−696.
(7) Akondi, K. B.; Muttenthaler, M.; Dutertre, S.; Kaas, Q.; Craik, D.
J.; Lewis, R. J.; Alewood, P. F. Chem. Rev. 2014, 114, 5815−5847.
(8) Kaas, Q.; Westermann, J. C.; Halai, R.; Wang, C. K. L.; Craik, D.
J. Bioinformatics 2008, 24, 445−446.
(40) Guntert, P. In Protein NMR Techniques; Downing, A. K., Ed.;
̈
Humana Press: 2004; Vol. 278, Chapter 17, pp. 353−378.
(41) Brunger, A. T.; Adams, P. D.; Clore, G. M.; DeLano, W. L.;
Gros, P.; Grosse-Kunstleve, R. W.; Jiang, J. S.; Kuszewski, J.; Nilges,
M.; Pannu, N. S.; Read, R. J.; Rice, L. M.; Simonson, T.; Warren, G. L.
Acta Crystallogr., Sect. D: Biol. Crystallogr. 1998, 54, 905−921.
(42) Chen, V. B.; Arendall, W. B., 3rd; Headd, J. J.; Keedy, D. A.;
Immormino, R. M.; Kapral, G. J.; Murray, L. W.; Richardson, J. S.;
Richardson, D. C. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2010, 66,
12−21.
(9) King, G. F. Expert Opin. Biol. Ther. 2011, 11, 1469−1484.
(10) Blunt, J. W.; Copp, B. R.; Keyzers, R. A.; Munro, M. H.; Prinsep,
M. R. Nat. Prod. Rep. 2014, 31, 160−258.
(11) Mehbub, M. F.; Lei, J.; Franco, C.; Zhang, W. Mar. Drugs 2014,
12, 4539−4577.
(12) Takada, K.; Hamada, T.; Hirota, H.; Nakao, Y.; Matsunaga, S.;
van Soest, R. W.; Fusetani, N. Chem. Biol. 2006, 13, 569−574.
(13) Kolmar, H. Curr. Opin. Pharmacol. 2009, 9, 608−614.
(14) Olivera, B. M. J. Biol. Chem. 2006, 281, 31173−31177.
(43) Koradi, R.; Billeter, M.; Wuthrich, K. J. Mol. Graphics 1996, 14,
29−32.
(15) Burman, R.; Gunasekera, S.; Stromstedt, A. A.; Goransson, U. J.
̈
̈
Nat. Prod. 2014, 77, 724−736.
(44) Merrifield, R. B. J. Am. Chem. Soc. 1963, 85, 2149−2154.
(16) Liu, H.; Boudreau, M. A.; Zheng, J.; Whittal, R. M.; Austin, P.;
Roskelley, C. D.; Roberge, M.; Andersen, R. J.; Vederas, J. C. J. Am.
Chem. Soc. 2010, 132, 1486−1487.
(45) Stromstedt, A. A.; Felth, J.; Bohlin, L. Phytochem. Anal. 2014, 25,
̈
13−28.
(46) Sjogren, M.; Johnson, A. L.; Hedner, E.; Dahlstrom, M.;
Goransson, U.; Shirani, H.; Bergman, J.; Jonsson, P. R.; Bohlin, L.
Peptides 2006, 27, 2058−2064.
̈
̈
(17) Williams, D. E.; Austin, P.; Diaz-Marrero, A. R.; Soest, R. V.;
Matainaho, T.; Roskelley, C. D.; Roberge, M.; Andersen, R. J. Org.
Lett. 2005, 7, 4173−4176.
̈
(18) Austin, P.; Heller, M.; Williams, D. E.; McIntosh, L. P.; Vogl, A.
W.; Foster, L. J.; Andersen, R. J.; Roberge, M.; Roskelley, C. D. PLoS
One 2010, 5, e10836.
(19) Towle, K. M.; Chaytor, J. L.; Liu, H.; Austin, P.; Roberge, M.;
Roskelley, C. D.; Vederas, J. C. Org. Biomol. Chem. 2013, 11, 1476−
1481.
(20) Lidgren, G.; Bohlin, L. Tetrahedron Lett. 1986, 27, 3283−3284.
(21) Sjogren, M.; Goransson, U.; Johnson, A. L.; Dahlstrom, M.;
̈
̈
̈
Andersson, R.; Bergman, J.; Jonsson, P. R.; Bohlin, L. J. Nat. Prod.
2004, 67, 368−372.
(22) Hedner, E.; Sjogren, M.; Hodzic, S.; Andersson, R.; Goransson,
U.; Jonsson, P. R.; Bohlin, L. J. Nat. Prod. 2008, 71, 330−333.
(23) Sjogren, M.; Jonsson, P. R.; Dahlstrom, M.; Lundalv, T.;
̈
̈
̈
̈
̈
Burman, R.; Goransson, U.; Bohlin, L. J. Nat. Prod. 2011, 74, 449−
̈
454.
(24) Sjogren, M.; Dahlstrom, M.; Goransson, U.; Jonsson, P. R.;
̈
̈
̈
Bohlin, L. Biofouling 2004, 20, 291−297.
(25) Lind, K. F.; Hansen, E.; Osterud, B.; Eilertsen, K. E.; Bayer, A.;
Engqvist, M.; Leszczak, K.; Jorgensen, T. O.; Andersen, J. H. Mar.
Drugs 2013, 11, 2655−2666.
(26) Altschul, S. F.; Madden, T. L.; Schaffer, A. A.; Zhang, J.; Zhang,
Z.; Miller, W.; Lipman, D. J. Nucleic Acids Res. 1997, 25, 3389−3402.
(27) Radax, R.; Rattei, T.; Lanzen, A.; Bayer, C.; Rapp, H. T.; Urich,
T.; Schleper, C. Environ. Microbiol. 2012, 14, 1308−1324.
(28) Okonechnikov, K.; Golosova, O.; Fursov, M. Bioinformatics
2012, 28, 1166−1167.
(29) Wuthrich, K. NMR of Proteins and Nucleic Acids; John Wiley &
̈
Sons: New York, 1986.
(30) Wishart, D. S.; Bigam, C. G.; Yao, J.; Abildgaard, F.; Dyson, H.
J.; Oldfield, E.; Markley, J. L.; Sykes, B. D. J. Biomol. NMR 1995, 6,
135−140.
(31) Shen, Y.; Delaglio, F.; Cornilescu, G.; Bax, A. J. Biomol. NMR
2009, 44, 213−223.
(32) Cornilescu, G.; Delaglio, F.; Bax, A. J. Biomol. NMR 1999, 13,
289−302.
(33) Cierpicki, T.; Otlewski, J. J. Biomol. NMR 2001, 21, 249−261.
H
DOI: 10.1021/acs.jnatprod.5b00210
J. Nat. Prod. XXXX, XXX, XXX−XXX