Biosynthesis of the N–N-Bond-Containing Compound l-Alanosine

Article abstract of DOI:10.1002/anie.201913458

The formation of a N?N bond is a unique biochemical transformation, and nature employs diverse biosynthetic strategies to activate nitrogen for bond formation. Among molecules that contain a N?N bond, biosynthetic routes to diazeniumdiolates remain enigma

Full text of DOI:10.1002/anie.201913458

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Biosynthesis of the nitrogen-nitrogen bond-containing L-
alanosine
Authors: Menghua Wang, Haruka Niikura, Hai-Yan He, Phillip Daniel-
Ivad, and Katherine Ryan
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201913458
Angew. Chem. 10.1002/ange.201913458
Link to VoR: http://dx.doi.org/10.1002/anie.201913458
http://dx.doi.org/10.1002/ange.201913458

10.1002/anie.201913458
Angewandte Chemie International Edition
Biosynthesis of the nitrogen-nitrogen bond-containing L-alanosine
Menghua Wang,# Haruka Niikura,# Hai-Yan He, Phillip Daniel-Ivad, Katherine S. Ryan*
Department of Chemistry, The University of British Columbia, Vancouver, British Columbia,
Canada
#These authors contributed equally to this work
*Email: ksryan@chem.ubc.ca
Abstract
Formation of a nitrogen-nitrogen bond is a unique biochemical transformation, and nature employs
diverse biosynthetic strategies to activate nitrogen for bond formation. Among nitrogen-nitrogen
bond-containing molecules, biosynthetic routes to diazeniumdiolates remain enigmatic. Here we
report the biosynthetic pathway for the diazeniumdiolate-containing amino acid L-alanosine. Our
work reveals that the two nitrogens in the diazeniumdiolate of L-alanosine arise from glutamic acid
and aspartic acid, and we clarify the early steps of the biosynthetic pathway using both in vitro and
in vivo approaches. Our work demonstrates a peptidyl carrier protein-based mechanism for
activation of the precursor L-diaminopropionate, and we also show that nitric oxide can participate
in non-enzymatic diazeniumdiolate formation. Furthermore, we demonstrate that the gene alnA,
encoding a fusion protein with an N-terminal cupin domain and a C-terminal AraC-like DNA-
binding domain, is required for alanosine biosynthesis.
Main Text
Natural products with nitrogen-nitrogen (N-N) bonds are a group of rarely isolated yet
highly bioactive molecules, including drugs and drug candidates.[1–3] Unlike many natural
products, the biosynthetic routes to N-N bond-containing molecules remained enigmatic until
recently.[4–6] From a fundamental perspective, formation of an N-N bond is a challenging
transformation because of the intrinsic difficulty of joining two typically nucleophilic atoms.[7] To
address this fundamental challenge, diverse biosynthetic pathways have evolved for the activation
of a nitrogen as an electrophile. Key strategies to activate nitrogen include oxidation to
hydroxylamine[8,9] or to nitrous acid.[4] From here, the N-N bond formation in many pathways
proceeds through the action of a dedicated N-N bond-forming enzyme. To date, only two
biosynthetic enzymes, KtzT and SznF (also called StzF[10]), have been demonstrated to form an N-
N bond in vitro.[5,6] KtzT, a piperazate synthase, forms a cyclic hydrazine through a heme-
This article is protected by copyright. All rights reserved.

10.1002/anie.201913458
Angewandte Chemie International Edition
dependent mechanism.[5] By contrast, the streptozocin biosynthetic enzyme SznF has been shown
to employ a diiron domain and a cupin domain to form an N-nitroso amino acid from methyl-
arginine and oxygen,[6] giving a product that may be on-pathway to streptozocin.
The use of a cupin domain for N-N bond formation, as in the case of SznF, is emerging as
potentially widespread strategy toward diverse N-N bonds, as well as other heteroatom-heteroatom
linkages, like sulfonamides.[11] In addition to the streptozocin pathway, cupin domain-encoding
genes have now been linked to N-N bond-containing natural products including pyrazomycin,[12]
hydrazinoacetic acid,[13] triacsins,[14] fragin,[15] and gramibactin,[16] even though the exact
mechanistic details are not fully understood (Fig. 1a). Among the diazeniumdiolates, which
tautomerize to N-hydroxy-N-nitroso compounds (Fig. 1b),[17,18] both fragin and gramibactin have
been linked to cupin domain-containing genes through gene inactivation studies.[15,16] The amino
acid L-alanosine (Fig. 1c) is another important diazeniumdiolate and is representative of a unique
group of small diazeniumdiolates (Fig. 1d). L-Alanosine, also known as SDX-102, inhibits the de
novo synthesis of adenine and is a promising molecule for chemoselective treatment of
methylthioadenosine phosphorylase (MTAP)-deficient tumors. It was brought through a Stage II
trial on patients with MTAP-deficient tumors from 2003-2009.[19,20]
To elucidate the biosynthetic pathway to L-alanosine, we sequenced the genome of the L-
alanosine producer Streptomyces alanosinicus DSM 40606[21] and identified a candidate gene
cluster (Fig. 1e, GenBank Accession Number MN626712) encoding AlnD and AlnE, which are
homologs of L-2,3-diaminopropionic acid (L-Dap) forming enzymes SbnA and SbnB from
staphyloferrin B biosynthesis.[22] We also observed genes encoding AlnM and AlnN, homologs of
CreD/CreE and FzmM/FzmL pairs, which have been shown to produce nitrous acid for
biosynthesis of N-N bond-containing compounds cremeomycin and fosfazinomycin,
respectively.[4,23] The gene cluster encodes a candidate amino acid adenylation domain AlnF, a
peptidyl carrier protein AlnI, and a thioesterase AlnJ. There are also genes encoding an NADPH-
dependent FMN reductase AlnL, and a putative flavin-dependent acyl-CoA dehydrogenase AlnG.
While acyl-CoA dehydrogenases normally operate to catalyze reduction of double-bonds, it is
conceivable that AlnG could operate in the oxidative direction with reduced flavin provided by
AlnL, which could then activate oxygen for formation of a hydroxylamine.[24] Finally, we noted a
cupin domain at the N-terminus of AlnA, connected to an AraC-like DNA-binding C-terminal
domain.[25]
This article is protected by copyright. All rights reserved.

10.1002/anie.201913458
Angewandte Chemie International Edition
We searched by Blast-P to look for other potential L-alanosine biosynthetic gene clusters,
which yielded similar clusters from 13 other Streptomyces and one Saccharothrixs species (Fig.
S1). Twelve genes alnA-L were in most of the 14 other candidate gene clusters, suggesting an
intact biosynthetic gene cluster that includes alnA. By contrast, alnM/alnN are not found in any of
the other 14 gene clusters. Instead, in place of alnM/alnN, we often observed an adjacent nitrate
transporter, nitrate reductase, and nitrite reductases genes, suggesting there might be an alternate
biosynthetic route for the production of oxidized nitrogen in these other strains (Table S3).
alnA alnB
alnC
alnD alnE
alnF
alnG alnH alnJ
alnI
alnL
alnM
alnN
1 kb
alnK
Cupin domain
AraC-type DNA-binding domain
L-Dap synthetic proteins (SbnA/B homologs)
Adenylation domain, PCP and Thiosterase
Putative N-hydroxylation proteins
Regulatory and resistance elements
NOx producing proteins (CreD/E homologs)
Figure 1. N-N bond-containing molecules and the alanosine biosynthetic gene cluster. a) N-N
bond-containing natural products experimentally linked to cupin-domain containing proteins,
shown in red boxes. b) Diazeniumdiolates tautomerize to N-nitroso-N-hydroxylamines. c) L-
Alanosine. d) Molecules related to L-alanosine. e) L-Alanosine gene cluster and annotation of
biosynthetic genes.
To further substantiate the link between this gene cluster and L-alanosine production, we
made gene inactivation mutants ∆alnA, ∆alnL, ∆alnM, and ∆alnN (Fig. S2). The ∆alnM mutant
abrogates production of L-alanosine, thus linking the aln cluster to L-alanosine biosynthesis (Fig.
2a). We also observed only a trace amount of L-alanosine from the ∆alnA mutant. For the ∆alnL
and ∆alnN mutants, we still detected high production of L-alanosine (~60% and ~45% of the wild-
type strain, respectively). In the S. alanosinicus genome, we observed two genes homologous to
alnN, with identities of 61% (orf08239) and 51% (orf00178) respectively, and one gene
This article is protected by copyright. All rights reserved.

10.1002/anie.201913458
Angewandte Chemie International Edition
homologous to alnL with an identity of 50% (orf04442) (Table S4). These genes may complement
alnL and alnN for the production of L-alanosine.
The genes alnM/N are homologous to the gene pair creD/E, and CreD is an amine
−
monooxygenase that reacts with L-aspartate, and CreE is thought to release nitrite (NO2 ) from
nitrosuccinic acid.[4] Because alnM/N are homologous to creD/E, we interrogated whether these
−
inactivation mutants would produce NO2 in culture, which would provide further support for a
role of alnM/N in nitrite production. We observed that wild-type of S. alanosinicus produced high
−
concentrations of NO2 (15 ± 2 µM) compared to a media only control (4.6 µM). For the ∆alnM
−
inactivation mutant, we observed low levels of NO2 production (5.5 ± 0.1 µM), thus providing
−
clear evidence of the involvement of AlnM in NO2 production (Fig. 2b). Both the ∆alnL and
−
∆alnN mutants produced more NO2 (12 ± 2 µM and 11.1 ± 0.2 µM, respectively). This result on
∆alnN is consistent with complementation of this gene by a gene outside of the L-alanosine gene
cluster.
EIC of m/z 150 [M+H]+
a
b
22
18
14
10
6
L-alanosine
WT
∆alnA
∆alnL
∆alnM
∆alnN
2
WT
∆alnL ∆alnM ∆alnN media
1
2
3
4
5 min
c
Unlabeled alanosine
15N-Asp
100
100
80
60
40
20
0
80
60
40
20
0
m/z 121.0
(8%)
m/z 151.0
(7%)
m/z 121.0
(8%)
m/z 151.0
(16%)
110
120
130
140
150
120
130
140
150
110
m/z
m/z
100
80
60
40
20
0
15N-Glu
100
Na15NO2
80
60
40
m/z 121.0
(25%)
m/z 151.0
(22%)
m/z 121.0
(8%)
m/z 151.0
(7%)
20
0
110
120
130
140
150
110
120
130
140
150
m/z
m/z
Figure 2. In vivo studies link the aln gene cluster to L-alanosine production. a) Extracted ion
chromatograms (EIC, m/z 150) of extracts of S. alanosinicus wild-type, ∆alnA, ∆alnL, ∆alnM, and
∆alnN. b) Detection of nitrite by the Griess reagent. Each value is the mean ± SD for three
This article is protected by copyright. All rights reserved.

10.1002/anie.201913458
Angewandte Chemie International Edition
15
15
experiments. c) LC-MS analysis of L-alanosine from cultures supplemented with N-Asp, N-
Glu, or Na15NO2.
To directly interrogate whether aspartate is a precursor to L-alanosine, we fed 15N-labeled
15
aspartic acid to S. alanosinicus. LC-MS analysis showed an enrichment of singly labeled N-
alanosine (m/z 151.0) with a +1 Da signal intensity increasing from 7% to 16% (Fig. 2c).
Additionally, the fragment with m/z 120.0, anticipated to correspond to the loss of the N-nitroso,[26]
showed no significant incorporation above natural isotopic abundance, supporting that the
aspartate-derived nitrogen is specifically incorporated into the distal, N-nitroso nitrogen. No
15
significant +1 Da signal of N-labeled L-alanosine was detected in the Na15NO2 feeding group
(Fig. 2c, Table S5). This results suggests that exogenous nitrite cannot be employed in this
pathway, or that there is a requirement for nitric oxide rather than nitrite.[27] We note that our
results are in contrast to a recent report on alanosine biosynthesis from Ng and co-workers, who
showed ~80% incorporation of labeled nitrogen into this position when using Na15NO2.[28] One
possibility is that the difference may be attributed to the fermentation conditions used (Table S5).
The presence of alnD and alnE in the gene cluster, with homology to sbnA (39% identity
at the amino acid level) and sbnB (27% protein identity at the amino acid level), respectively,
suggests that the non-proteinogenic amino acid L-diaminopropionate (L-Dap) might be the
biosynthetic precursor to L-alanosine. SbnA and SbnB convert glutamic acid and O-phospho-L-
serine to L-Dap, and the nitrogen from glutamic acid is retained in L-Dap. To query the possibility
15
that L-alanosine derives from glutamic acid via L-Dap, we fed N-labeled glutamic acid to the
culture of S. alanosinicus. LC-MS analysis showed that the 15N atom of glutamic acid is
incorporated into L-alanosine, as shown by the increase in the intensity of the +1 Da signal from
7% to 22% when compared with the unlabeled culture (Fig. 2c). Furthermore, we observed a
similar increase in the m/z signal at 121 (8% to 25%) resulting from the presence of the 15N after
cleavage of the N-N bond of the N-nitroso moiety, seeing a similar increase of the +1 Da signal
from 8% to 25% (Fig. 2c), demonstrating that the glutamic acid-derived nitrogen is incorporated
into the N3-hydroxylamine nitrogen. The incorporation of labeled nitrogen to specific nitrogens in
L-alanosine excluded the possibility of nitrogen shuffling in either the 15N-L-glutamic acid or 15N-
L-aspartic acid feeding experiments.
This article is protected by copyright. All rights reserved.

10.1002/anie.201913458
Angewandte Chemie International Edition
To further substantiate the possibility that L-alanosine arises from L-glutamic acid, we
pursued in vitro studies on purified recombinant enzymes AlnD and AlnE. A reaction mixture of
AlnD, AlnE, O-phospho-L-serine, L-glutamic acid, PLP, and NAD+ gave two major peaks, which
correspond to peaks from an authentic sample of L-Dap (Fig. 3a). These peaks both have mass
peaks at m/z 338, corresponding to the [M+H]+ ion for L-Dap-DNS and were absent in a control
using heat denatured AlnD and AlnE. Using 15N-labeled glutamic acid resulted in the same peaks,
except with a +1 Da ion (m/z 339) (Fig. 3a).
Following formation of L-Dap, the free amino acid L-Dap could be adenylated by the
enzyme AlnF, or be further modified prior to adenylation. To interrogate these two possibilities,
we used the colorimetric detection of free pyrophosphate assay[29] to measure the adenylation
activity of AlnF. The highest levels of pyrophosphate production occur when using L-Dap as a
substrate (Fig. 3b). Furthermore, we synthesized L-N3-OH-Dap (Fig. S3) and tested it as a
substrate. While production of pyrophosphate was substantially lower, our results suggest that
adenylation can also occur on this amino acid. Similarly, we tested L-alanosine, but only observed
negligible levels of pyrophosophate production. No activity was observed for any other tested
amino acid (Fig. S6). Our adenylation domain assays suggest that L-Dap is the preferred substrate
for AlnF. From here, the adenylated L-Dap should be activated for transfer to the candidate
peptidyl carrier protein (PCP) AlnI. We purified holo-AlnI from E. coli by using plasmid
sfp/pCDF-Duet for co-expression of AlnI and Sfp, a phosphopantetheinyl transferase necessary
for formation of the holo-form PCPs.[30] A reaction mixture with AlnF, holo-AlnI, ATP, MgCl2,
and L-Dap was analyzed by HPLC and LC-TOFMS. We found the holo-AlnI was converted
completely to L-Dap tethered holo-AlnI (Fig. 3c). Based on this result, it is likely that further
modification of L-Dap occurs when tethered onto the peptidyl carrier protein AlnI.
This article is protected by copyright. All rights reserved.

10.1002/anie.201913458
Angewandte Chemie International Edition
O
b
a
O
100
90
80
70
60
50
40
30
20
10
0
R
O
AlnF
A
R
Cl
S
OH
AMP
O
O
OH
H2NOH
+
O
O
HO
P
O
H215N
OH
OH
NH2
OH
OH
O
NH2
hydroxyamine
NH
substrate
NH2
DNS
O-phospho-L-serine
ATP
N
O
PPi
2-
+
O
O
O
AMP
AlnD, AlnE
DNS-Cl
H15
N
MoO4
ascorbic acid
H215N
OH
HO
BTPPACl
DNS
NH2
O
15NH2
NH2
L-Dap
R
L-Dap-DNS
(MW 337)
OH
L-glutamic acid
α-ketoglutarate
N
H
measured at 620 nm
NH2
hydroxamate derivative
L-Dap-DNS
Whole reaction with
unlabeled L-Glu
L-Dap
L-N3-OH-Dap L-alanosine
Gly
L-Ala D-Ala
AlnI
AlnF
EIC (m/z 338)
c
ATP
PPi
PCP
A
O
AMP
S
O
Whole reaction with
15N labeled L-Glu
OH
H2
N
O
O
NH2
NH2
NH2
NH2
NH2
EIC (m/z 339)
EIC (m/z 338)
L-Dap
10705.72 (Calc M.W.=10704.92)
10619.63 (Calc M.W.=10618.82)
Heat denatured
AlnD and AlnE
Whole reaction
with L-dap
Authentic L-Dap
Reaction with
no substrate
EIC (m/z 338)
4
8
12
16
ꢀꢁꢂ
34
38
42 ꢀꢁꢂ 10500 10600 10700 10800 10900
m/z
Figure 3. In vitro studies of L-alanosine biosynthetic enzymes. a) Formation of L-Dap by AlnD
and AlnE. The peak at ~12 min with m/z 339 is L-serine-DNS, where L-serine is generated from
O-phospho-L-serine in the DNS modification conditions. b) Relative pyrophosphate production
by AlnF and tested amino acids. The method for this assay is shown. Each value is the mean ± SD
for three experiments. BTPPACl: bis(triphenylphosphoranylidene) ammonium chloride. c)
Molecular mass shift of AlnI reactions. The chromatogram and deconvoluted mass spectra are
shown on the left and right, respectively.
We next interrogated how the N-N bond forming reaction might proceed. The encoding of
alnM/alnN, homologs of creD/creE, in the gene cluster, and the biogenesis of the N-nitroso
−
nitrogen of L-alanosine from aspartate, suggested that NO2 should be generated from aspartate
and then employed biosynthetically. Furthermore, we observed that the ΔalnM variant showed a
−
decrease in accumulation of NO2 to levels observed in the media. However, the lack of
15
−
−
appreciable incorporation of NO2 argued against direct incorporation of NO2 itself. We note
−
that nitric oxide would be expected to spontaneously convert to NO2 in oxidative conditions.[31]
Furthermore, other alnA-containing gene clusters lack alnM/alnN homologs and instead contain a
candidate nitrate transporter, nitrate reductase, and nitrite reductase genes, which would
collectively generate NO.[32,33] Previously, the hydroxylamine precursor of fragin was shown to
This article is protected by copyright. All rights reserved.

10.1002/anie.201913458
Angewandte Chemie International Edition
non-enzymatically react with NaNO2 to give fragin.[15] Inspired by this result, we incubated NaNO2
with synthetic L-N3-OH-Dap and observed non-enzymatic formation of L-alanosine. This activity
was pH dependent, with higher production of L-alanosine at pH 6 and no production of L-alanosine
above pH 8 (Fig. S7). This reaction is likely to proceed via the nitrosonium ion, whose production
is thought to require acidic conditions. However, prior work shows that L-alanosine can be
synthesized with a higher yield using butyl nitrite as an organic nitrosylating reagent instead of
NaNO2 in acidic conditions.[34,35] However, we then tested the reaction of NO-donor diethylamine
NONOate with L-N3-OH-Dap, and we observed significantly higher production of L-alanosine.
With this reagent, the pH maximum is at ~6, with production even at pH 9 (Fig. S7). While the
relevance of this non-enzymatic reaction to the biogenesis of L-alanosine in S. alanosinicus
remains to be seen, our results suggest that the radical nitric oxide, rather than nitrous acid or
nitrosonium, can provide a more viable path to diazeniumdiolates from hydroxylamine precursors.
AlnI
AlnI
AlnI
PCP
PCP
OH
OH
PCP
OH
O
O
O
O
OH
NH2
S
S
AlnJ
NH2
NH2
S
AlnG, AlnL
AlnF
AlnD, AlnE
NH2
O
O
O
O
+
NH2
NH2
NH2
NH
NH2
O
NH2
HO
P
N
N
O
N
N
α-ketoglutarate
OH
HO
HO
O
HO
L-Dap
HO
O
L-glutamic acid
O-phospho-L-serine
L-alanosine
non-enzymatic
AlnM, AlnN
AlnA?
O
HO
OH
NOx
O
NH2
L-aspartic acid
fumaric acid
Figure 4. Putative biosynthetic pathway to L-alanosine.
We also show by a gene inactivation experiment that the alanosine biosynthetic gene cluster
includes alnA, a protein encoding an N-terminal cupin domain and a C-terminal AraC-like DNA-
binding domain fusion. AraC-containing enzymes are transcriptional regulators that activate
operons, and employ an effector domain that responds to small molecules to initiate
transcription.[36] One possibility is that AlnA operates in this manner, with the N-terminal cupin
domain binding to a yet-unknown metabolite to initiate transcription of the alanosine gene cluster.
Another intriguing possibility is that the cupin domain of AlnA might participate in N-N bond
formation, consistent with cupin domains now linked to streptozocin, pyrazomycin, s56-p1,
triacsin, fragin, and gramibactin biosynthesis. We note that AlnA appears to have conserved metal-
This article is protected by copyright. All rights reserved.

10.1002/anie.201913458
Angewandte Chemie International Edition
binding residues with these enzymes (Fig. S8). However, further interrogation of the role of AlnA
as either an N-N bond-forming enzyme or a regulator awaits further study.
Nature uses diverse biosynthetic pathways to construct N-N bonds. In this work, we clarify
the early steps of the biosynthesis of the diazeniumdiolate compound L-alanosine through gene
cluster identification, feeding studies, and in vitro work. We demonstrate that L-alanosine derives
from glutamic acid and aspartic acid via L-Dap, and we show activation of L-Dap and subsequent
transfer gives holo-AlnI-L-Dap (Fig. 4). Furthermore, our studies on N-N bond formation suggest
that nitric oxide, rather than nitrite, is useful for construction of the diazeniumdiolate. Our work
sets the stage for consideration of enzymes that may further enhance this non-enzymatic reaction,
with AlnA as one candidate. Finally, our elucidation of the early steps of L-alanosine biosynthesis
will facilitate targeted identification of new diazeniumdiolate-containing molecules by genome
mining.
Acknowledgements
We are grateful to the Natural Sciences and Engineering Research Council of Canada (RGPIN-
2016-03778), the Canadian Institutes of Health Research (FDN-148381 and 201312MSH-322191-
209186), the Michael Smith Foundation for Health Research (Grant 16776), and the JSPS
Overseas Challenge Program for Young Researchers and a Grant-in-Aid for JSPS Fellows
(18J15184) for financial support to H.N.
Conflict of interest
The authors declare no conflict of interest.
References
[1] G. Le Goff, J. Ouazzani, Bioorg. Med. Chem. 2014, 22, 6529–6544.
[2] L. M. Blair, J. Sperry, J. Nat. Prod. 2013, 76, 794–812.
[3] K. D. Morgan, R. J. Andersen, K. S. Ryan, Nat. Prod. Rep., DOI: 10.1039/c8np00076j.
[4] Y. Sugai, Y. Katsuyama, Y. Ohnishi, Nat. Chem. Biol. 2016, 12, 73–75.
[5] Y.-L. Du, H.-Y. He, M. A. Higgins, K. S. Ryan, Nat. Chem. Biol. 2017, 13, 836–838.
[6] T. L. Ng, R. Rohac, A. J. Mitchell, A. K. Boal, E. P. Balskus, Nature 2019, 566, 94–99.
[7] A. J. Waldman, T. L. Ng, P. Wang, E. P. Balskus, Chem. Rev. 2017, 117, 5784–5863.
[8] C. S. Neumann, W. Jiang, J. R. Heemstra, E. A. Gontang, R. Kolter, C. T. Walsh,
ChemBioChem 2012, 13, 972–976.
[9] R. J. Parry, W. Li, Arch. Biochem. Biophys. 1997, 339, 47–54.
[10] H.-Y. He, A. C. Henderson, Y.-L. Du, K. S. Ryan, J. Am. Chem. Soc. 2019, 141, 4026–
4033.
[11] Z. Hu, T. Awakawa, Z. Ma, I. Abe, Nat. Commun. 2019, 10, 184.
This article is protected by copyright. All rights reserved.

10.1002/anie.201913458
Angewandte Chemie International Edition
[12] G. Zhao, S. Yao, K. W. Rothchild, T. Liu, Y. Liu, J. Lian, H.-Y. He, K. S. Ryan, Y.-L. Du,
ChemBioChem. DOI: 10.1002/cbic.201900449.
[13] K. Matsuda, T. Tomita, K. Shin-ya, T. Wakimoto, T. Kuzuyama, M. Nishiyama, J. Am.
Chem. Soc. 2018, 140, 9083–9086.
[14] F. F. Twigg, W. Cai, W. Huang, J. Liu, M. Sato, T. J. Perez, J. Geng, M. J. Dror, I.
Montanez, T. L. Tong, et al., ChemBioChem 2019, 20, 1145–1149.
[15] C. Jenul, S. Sieber, C. Daeppen, A. Mathew, M. Lardi, G. Pessi, D. Hoepfner, M.
Neuburger, A. Linden, K. Gademann, et al., Nat. Commun. 2018, 9, 1297.
[16] R. Hermenau, J. L. Mehl, K. Ishida, B. Dose, S. J. Pidot, T. P. Stinear, C. Hertweck, Angew.
Chem. Int. Ed Engl. 2019, 58, 13024-13029.
[17] R. Hermenau, K. Ishida, S. Gama, B. Hoffmann, M. Pfeifer-Leeg, W. Plass, J. F. Mohr, T.
Wichard, H.-P. Saluz, C. Hertweck, Nat. Chem. Biol. 2018, 14, 841–843.
[18] J. A. Hrabie, L. K. Keefer, Chem. Rev. 2002, 102, 1135–1154.
[19] M. Schmid, M. Sen, M. D. Rosenbach, C. J. Carrera, H. Friedman, D. A. Carson, Oncogene
2000, 19, 5747–5754.
[20] T. Efferth, Pers. Med. 2009, 6, 373–375.
[21] Y. K. S. Murthy, J. E. Thiemann, C. Coronelli, P. Sensi, Nature 1966, 211, 1198–1199.
[22] M. J. Kobylarz, J. C. Grigg, S. J. Takayama, D. K. Rai, D. E. Heinrichs, M. E. P. Murphy,
Chem. Biol. 2014, 21, 379–388.
[23] Z. Huang, K.-K. A. Wang, W. A. van der Donk, Chem. Sci. 2016, 7, 5219–5223.
[24] M. Hetzel, M. Brock, T. Selmer, A. J. Pierik, B. T. Golding, W. Buckel, Eur. J. Biochem.
2003, 270, 902–910.
[25] S. Rhee, R. G. Martin, J. L. Rosner, D. R. Davies, Proc. Natl. Acad. Sci. U. S. A. 1998, 95,
10413–10418.
[26] E. Rivera-Tirado, M. López-Casillas, C. Wesdemiotis, Rapid Commun. Mass Spectrom.
RCM 2011, 25, 3581–3586.
[27] J. D. Caranto, Curr. Opin. Chem. Biol. 2019, 49, 130–138.
[28] T. L. Ng, M. E. McCallum, C. R. Zheng, J. X. Wang, K. Wu, E. P. Balskus, ChemBioChem.
DOI: 10.1002/cbic.201900565.
[29] C. Maruyama, H. Niikura, M. Takakuwa, H. Katano, Y. Hamano, Methods Mol. Biol. 2016,
1401, 77–84.
[30] J. Fang, Y. Zhang, L. Huang, X. Jia, Q. Zhang, X. Zhang, G. Tang, W. Liu, J. Bacteriol.
2008, 190, 6014–6025.
[31] M. A. Marletta, P. S. Yoon, R. Iyengar, C. D. Leaf, J. S. Wishnok, Biochemistry 1988, 27,
8706–8711.
[32] T. Hino, S. Nagano, H. Sugimoto, T. Tosha, Y. Shiro, Biochim. Biophys. Acta BBA -
Bioenerg. 2012, 1817, 680–687.
[33] M. M. M. Kuypers, H. K. Marchant, B. Kartal, Nat. Rev. Microbiol. 2018, 16, 263–276.
[34] P. Strazzolini, M. G. Dall’Arche, M. Zossi, A. Pavsler, Eur. J. Org. Chem. 2004, 2004,
4710–4716.
[35] Y. Isowa, H. Kurita, M. Ohmori, M. Sato, K. Mori, Bull. Chem. Soc. Jpn. 1973, 46, 1847–
1850.
[36] J. Yang, M. Tauschek, R. M. Robins-Browne, Trends Microbiol. 2011, 19, 128–135.
This article is protected by copyright. All rights reserved.

10.1002/anie.201913458
Angewandte Chemie International Edition
TOC Graphic
This article is protected by copyright. All rights reserved.