Structural Revision of Natural Cyclic Depsipeptide MA026 Established by Total Synthesis and Biosynthetic Gene Cluster Analysis

Article abstract of DOI:10.1002/anie.202015193

A revised structure of natural 14-mer cyclic depsipeptide MA026, isolated from Pseudomonas sp. RtlB026 in 2002 was established by physicochemical analysis with HPLC, MS/MS, and NMR and confirmed by total solid-phase synthesis. The revised structure differs from that previously reported in that two amino acid residues, assigned in error, have been replaced. Synthesized MA026 with the revised structure showed a tight junction (TJ) opening activity like that of the natural one in a cell-based TJ opening assay. Bioinformatic analysis of the putative MA026 biosynthetic gene cluster (BGC) of RtIB026 demonstrated that the stereochemistry of each amino acid residue in the revised structure can be reasonably explained. Phylogenetic analysis with xantholysin BGC indicates an exceptionally high homology (ca. 90 %) between xantholysin and MA026. The TJ opening activity of MA026 when binding to claudin-1 is a key to new avenues for transdermal administration of large hydrophilic biologics.

Full text of DOI:10.1002/anie.202015193

Angewandte
Communications
Chemie
How to cite: Angew. Chem. Int. Ed. 2021, 60, 8792–8797
International Edition:
German Edition:
doi.org/10.1002/anie.202015193
doi.org/10.1002/ange.202015193
Natural Products
Structural Revision of Natural Cyclic Depsipeptide MA026 Established
by Total Synthesis and Biosynthetic Gene Cluster Analysis
Chihiro Uchiyama+, Akane Fukuda+, Minagi Mukaiyama+, Yoshiki Nakazawa,
Yuka Kuramochi, Kyohei Muguruma, Mitsue Arimoto, Akihiro Ninomiya, Koichiro Kako,
Yohei Katsuyama, Sho Konno, Akihiro Taguchi, Kentaro Takayama, Atsuhiko Taniguchi,
Yoko Nagumo, Takeo Usui,* and Yoshio Hayashi*
Abstract: A revised structure of natural 14-mer cyclic depsi-
peptide MA026, isolated from Pseudomonas sp. RtlB026 in
2002 was established by physicochemical analysis with HPLC,
MS/MS, and NMR and confirmed by total solid-phase syn-
thesis. The revised structure differs from that previously
reported in that two amino acid residues, assigned in error,
have been replaced. Synthesized MA026 with the revised
structure showed a tight junction (TJ) opening activity like that
of the natural one in a cell-based TJ opening assay. Bioinfor-
matic analysis of the putative MA026 biosynthetic gene cluster
(BGC) of RtIB026 demonstrated that the stereochemistry of
each amino acid residue in the revised structure can be
reasonably explained. Phylogenetic analysis with xantholysin
BGC indicates an exceptionally high homology (ca. 90%)
between xantholysin and MA026. The TJ opening activity of
MA026 when binding to claudin-1 is a key to new avenues for
transdermal administration of large hydrophilic biologics.
Table S1).[3–7] Among them, a 14-mer cyclic depsipeptide,
xantholysin A (Figure 1), which is an anti-microbial product
I
t is known that the infectious hematopoietic necrosis virus
(IHNV) causes infection and killing of salmon and trout,
resulting in a significant influence on the revenue of the
salmon farming industry.[1] By screening effective anti-IHNV
compounds, MA026 was isolated in 2002 from Pseudomonas
sp. RtlB026 which exists in the digestive tract of rainbow
trout, and its structure was determined by amino acid, MS/MS
and NMR analysis[2] to be a 14-mer cyclic depsipeptide (1)
containing several d-amino acid residues and an acyl tail at
the N-terminus (Figure 1). Pseudomonas spp. have also been
reported to produce many other cyclic lipodepsipeptides with
diverse biological activities (Supporting Information,
Figure 1. Reported (1)[2,9] and revised (2) structures of MA026 and the
structure of xantholysin A.[3,4] *Stereochemistry of a-carbons is
deduced from in silico experiments.[4]
[*] C. Uchiyama,[+] A. Fukuda,[+] Y. Nakazawa, Y. Kuramochi,
Dr. K. Muguruma, Dr. S. Konno, Dr. A. Taguchi, Dr. K. Takayama,
Dr. A. Taniguchi, Prof. Y. Hayashi
The University of Tokyo
1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657 (Japan)
and
Department of Medicinal Chemistry, School of Pharmacy
Tokyo University of Pharmacy and Life Sciences
1432-1 Horinouchi, Hachioji, Tokyo 192-0392 (Japan)
E-mail: yhayashi@toyaku.ac.jp
M. Mukaiyama,[+] M. Arimoto, Dr. A. Ninomiya, Dr. K. Kako,
Dr. Y. Nagumo, Prof. T. Usui
Collaborative Research Institute for Innovative Microbiology
The University of Tokyo
Bunkyo-ku, Tokyo 113-8657 (Japan)
Dr. K. Muguruma
Present address: Department of Chemical Science and Engineering,
School of Materials and Chemical Technology
Tokyo Institute of Technology
Graduate School of Life and Environmental Sciences
University of Tsukuba
Tokyo 152-8552 (Japan)
[+] These authors contributed equally to this work.
1-1-1 Tennodai, Tsukuba, Ibaraki 305-8572 (Japan)
E-mail: usui.takeo.kb@u.tsukuba.ac.jp
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
https://doi.org/10.1002/anie.202015193.
Prof. Y. Katsuyama
Department of Biotechnology
Graduate School of Agricultural and Life Sciences
8792
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 8792 –8797

Angewandte
Communications
Chemie
isolated in 2013 from Pseudomonas putida BW11M1 by Li
dilution conditions occur naturally on the solid support.[14]
et al.[3] and is known to show broad biological activities,[3,4,8]
has a high similarity to MA026 in the amino acid sequence. A
total solution-phase synthesis of MA026 and analysis of its
anti-hepatitis C viral activity by binding to the TJ membrane
protein claudin-1 were reported in 2013 by Shimura et al.[9]
Kanda et al. reported in 2017 that natural MA026 has a tight
junction (TJ) opening activity.[10]
The TJ is a continuous intercellular barrier between
epithelial cells that separates the internal and external
environments of cellular sheets, and controls invasion of
foreign substances and a diffusion of solutes and water across
the epithelium.[11] The TJ is composed of a multi-protein
complex including several transmembrane proteins, such as
claudins, occludin and several junctional adhesion molecules.
Among these bio-molecules, claudins found in at least 23
kinds of proteins in human are key regulators of paracellular
permeability.[11] Claudin-1 in particular, is highly expressed in
skin and plays an essential role in the formation of an
impermeable epidermal barrier.
Oral administration of peptide/protein drugs encounters
several problems.[12] With their high molecular weight and
hydrophilicity, such drugs are unable to diffuse through the
lipid bilayer of the cell membrane, and to solve this problem,
paracellular delivery, particularly transdermal absorption, has
received considerable attention.[13] Control of the TJ opening
by attenuating the function of claudin-1 has proved to be
attractive because it could allow safe and controllable trans-
dermal administration of peptides or proteins and many other
different hydrophilic macromolecular drugs.
In an effort to develop safe and effective paracellular
permeability enhancers leading to improved drug bioavaila-
bility from skin, we studied the TJ opener MA026. Although
we started a structure–activity relationship (SAR) study of
MA026, physicochemical data of the material synthesized
based on the reported structure were different from those of
natural MA026 and failed to show any TJ opening activity.
This led us to reconsider the structure of MA026 with further
synthetic study and physicochemical analyses of 1 and its
analogues, coupled with the latest phylogenetic analysis of
nonribosomal peptide synthetase (NRPS) of microorganisms.
In the new synthesis, the aim of which was the development of
a medical tool from MA026, we developed an efficient solid-
phase synthetic method, in which the complete cyclic peptide
structure was built on the resin. In the phylogenetic analysis,
virtual identification of natural cyclic peptides was used as
a complementary method providing a rationale to the
chemistry-based structure determination. This was also vali-
dated by a gene analysis of putative MA026 biosynthetic gene
cluster (BGC) derived from the whole genome sequencing of
RtIB026.
Our efficient solid-phase synthetic method features construc-
tion of the entire cyclic depsipeptide structure on the solid-
support and subsequent one-shot deprotection. For the
cyclization of the depsipeptide unit on the resin, an orthog-
onal protection strategy with allyl chemistry was adopted by
use of a Pd0-catalyzed deprotection method.[15] The planned
SPPS of MA026 was expected to facilitate the efficient
generation of a variety of derivatives to support an SAR
study.
A retrosynthetic analysis of MA026 based on the standard
Fmoc/tBu-based SPPS was conducted (Supporting Informa-
tion, Scheme S1). A key in this synthesis is construction of the
cyclic depsipeptide unit, that is, a macrolide containing an
ester bond, while a section containing a linear peptide chain
(Leu1 to Glu6) with an N-terminal (R)-3-hydroxydecanoyl
group can be constructed using conventional SPPS. Cycliza-
tion at the ester bond on the resin could reduce total yield and
cause epimerization at Ile14. Because of this, the application
of a dipeptide unit (Fmoc-d-Ser(Alloc-Ile)-OH, 9), in which
Ile14 is esterified with the b-hydroxy group of d-Ser7, was
planned. In this analysis, the amide bond between Ile14 and
d-Gln13 was disconnected and the amino group of Ile14 and
the carboxy group of d-Gln13 were protected by an Alloc
group and an allyl ester, respectively. This allows the loading
of Fmoc-d-Glu-OAll onto the Fmoc-NH-SAL resin via the g-
amide bonding, leading to a conventional and sequential
amide bond formation on the resin starting from d-Gln13.
The synthesis of the authentic structure of 1 and its
analogues is depicted in Scheme 1. Briefly, in the conven-
tional SPPS, Fmoc-deprotection was performed with 20%
piperidine in DMF, and coupling of each Fmoc-protected
amino acid was performed in DMF using the HATU/HOAt/
DIPEA method.[16] The solid-supported d-Gln (10) was
prepared by the coupling of Fmoc-d-Glu-OAll to the Fmoc-
NH-SAL resin after Fmoc-deprotection (Scheme 1). After
constructing the protected hexapeptide-resin (11a) and sub-
sequent deprotection of its Fmoc group, the dipeptide unit
Fmoc-d-Ser(Alloc-Ile)-OH (9; Supporting Information,
Scheme S2) was coupled using the DIPCI/HOBt method[17]
to obtain the resin-based depsipeptide (12a). The Alloc group
and allyl ester of 12a were removed in the presence of
[15]
Pd(PPh3)4 and PhSiH3
(Supporting Information, Fig-
ure S1), and on-resin cyclization was performed using the
DIPCI/HOBt method to obtain the cyclic depsipeptide-resin
(13a). HPLC and MS analysis (Supporting Information,
Figure S2) revealed no significant level of dimerization. In the
subsequent Fmoc deprotection of 13a using 20% piperidine/
DMF, an expected O-N intramolecular acyl migration[18] at d-
Ser7 was effectively suppressed to 5.2% by reducing the
piperidine-treatment time to 5 min (Supporting Information,
Figure S3). Further elongation of the peptide chain from 13a
and addition of the carboxylic acid (14)[9] with the conven-
tional SPPS gave a protected form of 1 on the resin. Final one-
shot deprotection and cleavage of peptides from the resin
with aqueous TFA (95%) afforded the crude peptide (1).
Subsequent final HPLC purification gave 1 (RT= 26.8 min)
in a yield of 15% with purity of > 99%. Peptide 2 was also
synthesized in a yield of 11% from the resin (purity 98%)
The first reported total synthesis of the published
structure of MA026 was performed in the solution phase[2,9]
and its solid-phase peptide synthesis (SPPS), which is an
efficient method for the synthesis of various derivatives, was
never reported. Generally, cyclization reactions in the solu-
tion phase synthesis must be carried out under high dilution
conditions to prevent dimer formation.[14] In solid-phase
synthesis, such concerns can be reduced because pseudo
Angew. Chem. Int. Ed. 2021, 60, 8792 –8797
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org
8793

Angewandte
Communications
Chemie
Scheme 1. Solid-phase synthesis of MA026 and its analogues. HATU: O-(7-azabenzotriazol-1-yl)-N,N,N’,N’-tetramethyluronium hexafluorophos-
phate, HOAt: 1-hydroxy-7-azatriazole, DIPEA: diisopropylethylamine, DIPCI: N,N’-diisopropylcarbodiimide, HOBt: 1-hydroxybenzotriazole.
with a protocol similar to that used for peptide 1. These results
suggest that the proposed on-resin synthetic route effectively
tolerates diversity and could be applicable to other cyclic
depsipeptides in the future. Details of the syntheses are
described in the Supporting Information.
Information, Figure S5). These results suggest that the
structure of MA026 is different from that of the reported
peptide 1. We synthesized a linear peptide (15) based on the
reported structure 1, but without cyclization. Hydrolysis of
natural MA026 at the ester was also performed to produce
a linear peptide (16). The comparison of HPLC RT of these
linear peptides revealed that peptides 15 and 16 are not
identical (Figure 2B). From these results, it was concluded
that the amino acid sequence or stereochemistry of MA026 is
different from that reported previously.
To confirm the amino acid sequence of natural MA026,
we carried out an MS/MS analysis of the hydrolysate (16)
derived from natural MA026 (Supporting Information, Fig-
ure S6). As a result, the detected y4 and b10 ions, corre-
sponding to Leu11-Ile14 and Leu1-Gln10 ion fragments,
respectively, indicated the existence of a Gln10-Leu11
The HPLC retention time (RT) (26.8 min) of 1, whose
synthesis was based on the reported structure was different
from that of natural MA026 (32.6 min; Figure 2A), although
HR-MS analyses indicated that 1 and natural MA026 have
the same molecular weight and a molecular formula of
1
C84H146N18O23. In addition, the H NMR data of 1 were also
different from that of natural MA026 (Supporting Informa-
tion, Figure S4). For example, natural MA026 showed peaks
around 3.8 and 1.1 ppm which were absent from the NMR
spectra of the synthetic 1. Moreover, the synthetic 1 exhibited
no TJ-opening activity in a transport assay (Supporting
8794 www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 8792 –8797

Angewandte
Communications
Chemie
Figure 2. HPLC charts of A) reported MA026 (1), revised MA026 (2)
and natural MA026, and B) synthetic linear peptide 15 and hydrolyzed
natural product 16. HPLC conditions: a linear gradient starting from
30 to 70% CH3CN in aqueous TFA over 40 min at flow rate of
1.0 mLminÀ1, UV: 220 nm.
sequence in 16, which differs from that reported (l-Leu10-d-
Gln11). These results reveal that the amino acid sequence is
reversed at the positions 10 and 11 of MA026.
Therefore, a cyclic depsipeptide (2) with a d-Gln10-l-
Leu11 sequence (Figure 1) was synthesized using the method
described above (Scheme 1). We found that the HPLC RT of
2 (32.5 min) is the same as that of natural MA026 (Figure 2A;
Figure 3. Reported (1)[2,9] and revised (2) structures of MA026 and the
structures of MA026 derivatives (1’, 2’ and 3–8).
1
Supporting Information, Figure S8A). Furthermore, H and
the BGC of MA026 have more than 90% sequence identity
with those encoded in the BGC of xantholysin (Supporting
Information, Table S4). Phylogenetic analysis of NRPS
domains revealed the corresponding A, C, C/E, and TE
domains of MA026 and xantholysin NRPSs are clustered into
same clades (Supporting Information, Figures S9–S11). These
results strongly suggest that MA026 produced by Pseudomo-
nas sp. RtI026 is identical to xantholysin A. Molina-Santiago
et al. however, reported that xantholysin A contains d-Gln6
and l-Gln13,[4] while revised MA026 (2) should have l-Gln6
and d-Gln13 from our analysis. Therefore, it is likely that the
two natural products are identical but impossible to prove at
this moment.
From the position of the C/E domains of NRPSs, there
were two residues whose predicted stereochemistries were
not consistent with the structure of MA026 (2); the stereo-
chemistries of Leu1 and Gln6 were predicted to be d while
our synthetic MA026 (2) has l. Therefore, we further
synthesized 3 diastereomers of MA026; d-Leu1/l-Gln6 (6),
l-Leu1/d-Gln6 (7), and d-Leu1/d-Gln6 (8) (Figure 3) and
compared with natural MA026. However, the RTs of these
compounds were different from that of natural MA026
(Supporting Information, Figure S12). Taken together with
absolute configuration of Leu and Gln/Glu from hydrolysate
of natural MA026 (Supporting Information, Figure S13), we
concluded that MA026 contains l-Leu1/l-Gln6. When we
analyzed HHI/LxxxxGD motif, which is conserved in C/E
domains but not in C domains,[23,24] it is not precisely
conserved in modules 2 and 7 (Supporting Information,
Table S5). This observation indicated that epimerase activ-
ities are probably eliminated in these domains and suggests
13C NMR data of 2 showed good agreement with those of
natural MA026 (Supporting Information, Figure S7,
Tables S2 and S3). Furthermore, we synthesized three ste-
reoisomers (3–5) with sequences l-Gln10-l-Leu11, l-Gln10-
d-Leu11 and d-Gln10-d-Leu11, respectively (Figure 3) using
the same route that was used for peptide 1. In HPLC analysis,
these three peptides showed RTs different from that of
natural MA026 (Supporting Information, Figure S8B), and it
was concluded that the structure of 2 is a correct structure for
natural MA026.
From this structural revision of MA026, it became
apparent that xantholysin A has the same amino acid
sequence as revised MA026. Recently, its d/l configuration
of each amino acid residue has been partially reported,[4] but
some differences to our revised structure remain. To reveal
whether or not MA026 and xantholysin A are the same
compound, we determined the whole genome sequence of
Pseudomonas sp. RtlB026. The putative BGC of MA026
encodes the nonribosomal peptide synthetase (NRPS) which
has 14 modules, and terminal tandem thioesterase (TE)
domains that are frequently observed in BGCs of other cyclic
lipodepsipeptides[19] and has been suggested to be involved in
productivity enhancement[20,21] (Figure 4). All modules except
for modules 1, 12 and 13 contained condensation/epimeriza-
tion (C/E), adenylation (A), and thiolation (T) domains. The
modules 1, 12, and 13 have condensation (C) domains instead
of a C/E domain. The predicted substrate specificity of the A
domains by antiSMASH techniques[22] showed that the amino
acid sequence of MA026 is Glu10-Leu11, identical with that
in the xantholysins.[3,4] Furthermore, the NRPSs encoded in
Angew. Chem. Int. Ed. 2021, 60, 8792 –8797
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org
8795

Angewandte
Communications
Chemie
Figure 4. MA026 biosynthetic gene cluster of Pseudomonas sp. RtIB026. The organization of the genomic region with the MA026 synthetase genes
(A, B, and C), the associated regulatory gene (R) and putative transporter genes (D, E, and F) is shown (GenBank accession numbers:
AP023348).
that the amino acid changes in consensus sequence, His to
Leu of first amino acid residue, His to Tyr of second amino
acid residue, or deletion, lead the loss of epimerase activity.
Interestingly, xantholysin biosynthetic gene clusters still
retain epimerase activity although the consensus sequence
of the C/E domain of module 7 is not conserved (Supporting
Information, Table S5).
Finally, we evaluated the TJ opening activity of revised
MA026 (2) with transport assay using FD4 (FITC-dextran
4 kDa).[10,13] In this assay, the fluorescence intensity from
FITC is measured in the transmembrane part and recognized
as the TJ opening activity. As shown in Figure 5, revised
MA026 (2) efficiently transports FD4 similarly to natural
MA026, but the reported MA026 (1) and its epimer at C3
position of N-terminal acyl tail (1’ in the Supporting
Information, Figures S5 and S14) failed to do so. The revised
MA026 (2) induced the time-dependent increase of cumu-
lative fluorescence intensity and the activity was equivalent to
that of the natural product (Supporting Information, Figur-
es S14 and S15). Furthermore, an epimer (2’) of the revised
MA026 in the N-terminal acyl section had drastically
decreased transport activity, suggesting the importance of
the stereochemistry of the hydroxyl group at the N-terminal
acyl group. On the other hand, two epimers of MA026; d-
Leu1/l-Gln6 (6) and l-Leu1/d-Gln6 (7) showed more potent
transport activity, but the transport activity of d-Leu1/d-Gln6
(8) resembled that of 2 (Supporting Information, Figure S16).
The epimers 6 and 7 could be seed compounds to obtain
a peptide that is more potent than natural MA026.
In conclusion, we successfully achieved an Fmoc-based
SPPS of MA026 in which the whole structure of MA026 was
constructed on the resin. This synthesis allowed the efficient
preparation of various MA026 derivatives. With a synthetic
study, we determined the revised structure (2) of MA026
based on the evidence from HPLC, NMR and MS(/MS)
analyses and a TJ opening activity assay. BGCꢀs genomic
analysis, which serves to identify the structure of this natural
cyclic peptide, worked very well and agreed with the
physicochemical analyses. The revised structure contains d-
Gln10-l-Leu11 which is different from the reported structure
(1). 1 has no transport activity, but diastereomers of 2 (2’ and
6–8) showed transport activity (Figures 5 and S16). These
results suggest that the stereochemistry of each amino acid
residue, at least those in side chains, is not important for
transport activity compared to amino acid sequence. Detailed
SAR study of MA026 is in progress and SAR information will
contribute to the development of efficient drug administra-
tion methods using MA026 analogues.
Acknowledgements
This work was partly supported by JSPS KAKENHI Grant
Numbers JP24658107, 19H02893 and 19H03356, and Japan
Society for Bioscience, Biotechnology, and Agrochemistry
(JSBBA Innovative Research Program Award) to T.U. We
appreciate Professor Kamisuki for providing natural MA026.
Figure 5. Transport assay of revised MA026 (2) and its epimer at C3
position of N-terminal acyl tail (2’), reported MA026 (1), and natural
product (MA026) (each 3 mM). Vehicle: DMSO, the MDCK II mono-
layer was treated with each compound for 30 min, and fluorescence
intensity of transported FD4 was measured. Triplicate. Values were
mean Æ standard deviation (n=3).
8796 www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 8792 –8797

Angewandte
Communications
Chemie
[10] Y. Kanda, Y. Yamasaki, S. Shimura, S. Kamisuki, F. Sugawara, Y.
Conflict of interest
Nagumo, T. Usui, J. Antibiot. 2017, 70, 691 – 694.
[11] a) S. Tsukita, M. Furuse, M. Itoh, Nat. Rev. Mol. Cell Biol. 2001,
2, 285 – 293; b) M. Furuse, S. Tsukita, Trends Cell Biol. 2006, 16,
181 – 188.
[12] K. Sonaje, K. J. Lin, J. J. Wang, F. L. Mi, C. T. Chen, J. H. Juang,
H. W. Sung, Adv. Funct. Mater. 2010, 20, 3695 – 3700.
[13] T. Shiobara, T. Usui, J. Han, H. Isoda, Y. Nagumo, PLoS One
2013, 8, e79954.
[14] C. J. White, A. K. Yudin, Nat. Chem. 2011, 3, 509 – 524.
[15] a) J. Tsuji, H. Takahashi, M. Morikawa, Tetrahedron Lett. 1965,
6, 4387 – 4388; b) H. Kunz, C. Unverzagt, Angew. Chem. Int. Ed.
Engl. 1984, 23, 436 – 437; Angew. Chem. 1984, 96, 426 – 427.
[16] L. A. Carpino, J. Am. Chem. Soc. 1993, 115, 4397 – 4398.
[17] D. Sarantakis, J. Teichman, E. L. Lien, R. L. Fenichel, Biochem.
Biophys. Res. Commun. 1976, 73, 336 – 342.
[18] J. M. Stewart in The Peptide, Vol. 3 (Eds.: E. Gross, J.
Meienhofer), Academic Press, New York, 1981, p. 170.
[19] N. Roongsawang, K. Washio, M. Morikawa, Int. J. Mol. Sci. 2011,
12, 141 – 172.
The authors declare no conflict of interest.
Keywords: bioinformatics · MA026 · natural products ·
structure elucidation · xantholysin A
[1] J. Kim, N. Oseko, T. Nishizawa, M. Yoshimizu, Dis. Aquat. Org.
2009, 83, 105 – 113.
[2] PCT WO 02/062831.
[3] W. Li, H. Rokni-Zadeh, M. D. Vleeschouwer, M. G. K. Ghe-
quire, D. Sinnaeve, G. Xie, J. Rozenski, A. Madder, J. C. Martins,
R. D. Mot, PLoS One 2013, 8, e62946.
[4] C. Molina-Santiago, Z. Udaondo, A. Daddaoua, A. Roca, J.
Martin, I. Perez-Victoria, F. Reyes, J.-L. Ramos, Microb.
Biotechnol. 2015, 8, 716.
[5] F. E. Oni, N. Geudens, A. Adiobo, O. O. Omoboye, E. A. Enow,
J. T. Onyeka, A. E. Salami, R. D. Mot, J. C. Martins, M. Hçfte,
Microorganisms 2020, 8, 1079.
[6] M. Anselmi, T. Eliseo, L. Zanetti-Polzi, M. R. Fullone, V.
Fogliano, A. D. Nola, M. Paci, I. Grgurina, Biochim. Biophys.
Acta Biomembr. 2011, 1808, 2102.
[7] D. D. Nguyen, A. V. Melnik, N. Koyama, X. Lu, M. Schorn, J.
Fang, K. Aguinaldo, T. L. Lincecum, Jr., M. G. K. Ghequire, V. J.
Carrion, T. L. Cheng, B. M. Duggan, J. G. Malone, T. H. Mauch-
line, L. M. Sanchez, A. M. Kilpatrick, J. M. Raaijmakers, R. D.
Mot, B. S. Moore, M. H. Medema, P. C. Dorrestein, Nat. Micro-
biol. 2016, 1, 16197.
[20] J. Hou, L. Robbel, M. A. Marahiel, Chem. Biol. 2011, 18, 655 –
664.
[21] D. Mandalapu, X. Ji, J. Chen, C. Guo, W. Q. Liu, W. Ding, J.
Zhou, Q. Zhang, J. Org. Chem. 2018, 83, 7271 – 7275.
[22] K. Blin, S. Shaw, K. Steinke, R. Villebro, N. Ziemert, S. Y. Lee,
M. H. Medema, T. Weber, Nucleic Acids Res. 2019, 47, W81.
[23] C. J. Balibar, F. H. Vaillancourt, C. T. Walsh, Chem. Biol. 2005,
12, 1189.
[24] K. Bloudoff, T. M. Schmeing, Biochim. Biophys. Acta Proteins
Proteomics 2017, 1865, 1587.
[8] D. J. Lim, S. Y. Yang, M. Y. Noh, C. W. Lee, J. C. Kim, I. S. Kim,
Entomol. Res. 2017, 47, 337.
[9] S. Shimura, M. Ishima, S. Nakajima, T. Fujii, N. Himeno, K.
Ikeda, J. Izaguirre-Carbonell, H. Murata, T. Takeuchi, S.
Kamisuki, T. Suzuki, K. Kuramochi, K. Watashi, S. Kobayashi,
F. Sugawara, J. Am. Chem. Soc. 2013, 135, 18949 – 18956.
Manuscript received: November 13, 2020
Revised manuscript received: January 16, 2021
Accepted manuscript online: February 2, 2021
Version of record online: March 8, 2021
Angew. Chem. Int. Ed. 2021, 60, 8792 –8797
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org 8797