Quinolactacin Biosynthesis Involves Non-Ribosomal-Peptide-Synthetase-Catalyzed Dieckmann Condensation to Form the Quinolone-γ-lactam Hybrid

Article abstract of DOI:10.1002/anie.202005770

Quinolactacins are novel fungal alkaloids that feature a quinolone-γ-lactam hybrid, which is a potential pharmacophore for the treatment of cancer and Alzheimer's disease. Herein, we report the identification of the quinolactacin A2 biosynthetic gene cluster and elucidate the enzymatic basis for the formation of the quinolone-γ-lactam structure. We reveal an unusual β-keto acid (N-methyl-2-aminobenzoylacetate) precursor that is derived from the primary metabolite l-kynurenine via methylation, oxidative decarboxylation, and amide hydrolysis reactions. In vitro assays reveal two single-module non-ribosomal peptide synthetases (NRPs) that incorporate the β-keto acid and l-isoleucine, followed by Dieckmann condensation, to form the quinolone-γ-lactam. Notably, the bioconversion from l-kynurenine to the β-keto acid is a unique strategy employed by nature to decouple R*-domain-containing NRPS from the polyketide synthase (PKS) machinery, expanding the paradigm for the biosynthesis of quinolone-γ-lactam natural products via Dieckmann condensation.

Full text of DOI:10.1002/anie.202005770

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Quinolactacin Biosynthesis Involves NRPSs Catalyzed
Dieckmann Condensation to Form the Quinolone-γ-lactam Hybrid
Authors: Fanglong Zhao, Zhiwen Liu, Shuyuan Yang, Ning Ding, and
Xue Gao
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.202005770
Link to VoR: https://doi.org/10.1002/anie.202005770

10.1002/anie.202005770
Angewandte Chemie International Edition
RESEARCH ARTICLE
Quinolactacin Biosynthesis Involves NRPSs Catalyzed Dieckmann
Condensation to Form the Quinolone-γ-lactam Hybrid
Fanglong Zhaoa, Zhiwen Liua, Shuyuan Yanga, Ning Dinga, Xue Gaoa,b*
[a]
[b]
Dr. F. L. Zhao, Dr. Z. W. Liu, S. Y. Yang, N. Ding, Dr. X. Gao
Department of Chemical and Biomolecular Engineering
Rice University
Houston, Texas 77005, USA.
Dr. X. Gao
Department of Bioengineering
Rice University
Houston, Texas 77005, USA.
Corresponding Author
*E-mail: xue.gao@rice.edu
Abstract: Quinolactacins are novel fungal alkaloids that feature a
quinolone-γ-lactam hybrid, which is a potential pharmacophore for
the treatment of cancer and Alzheimer’s disease. Here, we report the
identification of the quinolactacin A2 biosynthetic gene cluster and
elucidate the enzymatic basis for the formation of the quinolone-γ-
lactam structure. We reveal an unusual β-keto acid (N-methyl-2-
aminobenzoylacetate) precursor that is derived from the primary
metabolite L-kynurenine via methylation, oxidative decarboxylation,
and amide hydrolysis reactions. In vitro assays reveal two single-
module NRPSs that incorporate the β-keto acid and L-isoleucine,
followed by Dieckmann condensation, to form the quinolone-γ-
lactam. Notably, the bioconversion from L-kynurenine to the β-keto
acid is a unique strategy employed by Nature to decouple R* domain-
containing NRPS from PKS machinery, which expands the paradigm
for the biosynthesis of quinolone-γ-lactam natural products via
Dieckmann condensation.
Figure 1. Structures of quinolactacin A2 (1) and other quinolone-γ-lactam
hybrids containing natural products.
Notably, 1 shows 14-fold higher AChE inhibitory activity than its
epimer QUL A1.[4d] Several synthetic strategies have been
developed to synthesize 1 chemically.[5] The biomimetic synthesis
of 1 from tryptamine required eight-step reactions with a yield of
~8%.[5a] An improved method achieved a yield of ~30% from
isatoic anhydride; however, its diastereomer QUL A1 was also
produced with a ratio of 8:1 (QUL A2: QUL A1).[5b] Understanding
Introduction
Quinolone and γ-lactam are both important pharmacophores in
many clinical drugs and have been extensively studied.[1] For
example, the quinolone antibiotics ciprofloxacin, ofloxacin, and
sparfloxacin are currently most commonly used to treat
tuberculosis cases,[2] while the γ-lactam compound levetiracetam
is one of the main anticonvulsant drugs for epilepsy treatment.[3]
The scaffold of quinolone-γ-lactam hybrid represents a new
pharmacophore structure and has the potential to combat the
increasing crisis of multidrug resistance. In recent years, the
how Nature builds such
a
unique quinolone-γ-lactam
pharmacophore in a highly selective and efficient manner can
reveal useful enzymatic mechanisms and lead to the discovery of
new bioactive molecules.
The biosynthesis of 1 has only been explored by feeding
experiments using 13C single-labeled substrates in Penicillium sp.
EPF-6.[6] By analyzing the distribution patterns of 13C atoms, it
was proposed that the quinolone-γ-lactam hybrid scaffold is
synthesized from the combination of L-isoleucine (L-Ile) and L-
kynurenine (2), followed by quinolone cyclization, oxidative
decarboxylation, and lactam formation. Additionally, the N-methyl
group is derived from methionine,[6] which might be catalyzed by
an S-adenosylmethionine (SAM)-dependent methyltransferase.
The nonproteinogenic amino acid 2 is proposed to be an
intermediate metabolite in L-tryptophan catabolism and also the
precursor leading to the production of the important cofactor
nicotinamide adenine dinucleotide.[7] Bioconversion of L-
quinolone-γ-lactam
containing
natural
products,
e.g.,
quinolactacins (QULs) and quinolactacides, have been isolated
and characterized (Figure 1).[4] As the main representatives of the
unique quinolone-γ-lactam hybrid products, QULs were isolated
from the fermentation broth of mulberry pyralid derived Penicillium
sp. EPF-6, and subsequently were demonstrated to possess
diverse bioactivities.[4a, 4c, 4d] Among them, QUL A2 (1) exhibits
inhibitory activity against tumor necrosis factor production
induced by macrophages, making it a viable therapeutic target for
cancer treatment.[4c] Furthermore, 1 and its C-3 epimer QUL A1
(Figure 1) exhibit inhibitory activity against acetylcholinesterase
(AChE) and could potentially be developed as anti-Alzheimer’s
drug candidate.[4d] In addition, the quinolone-γ-lactam core in 1
has been proven to be chemically unstable and could undergo C-
3 epimerization to generate QUL A1 in mild reaction conditions.[4b]
tryptophan to
2 could be catalyzed by indoleamine-2,3-
dioxygenase (IDO) or tryptophan-2,3-dioxygenase (TDO) to
produce an unstable product, N-formyl-L-kynurenine, followed by
kynurenine formamidase catalyzed hydrolysis.[7a] These studies
have provided some clues to investigate the biosynthesis of
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202005770
Angewandte Chemie International Edition
RESEARCH ARTICLE
quinolone-γ-lactam hybrid. However, the enzymatic basis of the
biosynthetic steps remains elusive.
search of QulI showed another two IDOs (Table S7 and Figure
S5), which might complement the function of QulI and account for
the remaining production of 1 in ΔqulI mutant strain.
Herein, we describe the enzymatic formation of quinolone-γ-
lactam starting from the identification of a non-ribosomal peptide
synthetase (NRPS) gene cluster in Penicillium citrinum (P.
citrinum) ATCC 9849 for 1 biosynthesis. The detailed biosynthetic
pathway of 1 was proposed by gene knockout analyses and
enzymatic in vitro assays. We found that 2 underwent further
modification by a dedicated methyltransferase (QulM), an FMN-
dependent dehydrogenase (QulF), and an amidase (located
outside of the qul gene cluster) to produce an unstable β-keto acid
precursor 9, which could be spontaneously dehydrated to form 7.
Intriguingly, the NRPS QulB was able to incorporate the unstable
β-keto acid 9 and efficiently compete with the spontaneous
reaction. By further extending the β-keto acid 9 with L-Ile, the R*
domain of QulA could perform a Dieckmann condensation to form
the γ-lactam ring and release a 4-ketopyrrolidinone intermediate
from the assembly line. This intermediate could plausibly further
undergo a spontaneous cyclization to yield the final quinolone-γ-
lactam hybrid structure.
Figure 2. Genetic characterization of qul gene cluster. (A) Putative cluster
for 1 biosynthesis in P. citrinum ATCC 9849. (B) HPLC analysis (UV 254
nm) of 1 production in P. citrinum wild-type (WT) and qul mutants.
Results and Discussion
Cascaded modification of L-kynurenine (2) by QulM and QulF.
To elucidate the biosynthetic pathway of 1, we further conducted
metabolite analyses and enzymatic in vitro assays. Detailed LC-
MS analysis showed that a new compound 3 accumulated in
ΔqulM (Figure 3A, trace iv) while it was absent in the WT.
Bioinformatics analysis and genetic manipulations of 1
biosynthetic gene cluster. We first set out to identify and
characterize the qul biosynthetic gene cluster. The production of
1 in P. citrinum was confirmed by comparative NMR analysis
(Table S3 and Figures S10.1). We reasoned that the incorporation
of 2 and L-Ile is likely to be catalyzed by a dedicated two-module
NRPS machinery, and a methyltransferase should be required to
install the methyl group at the N-4 position of 1. We then searched
the P. citrinum genome using antiSMASH[8] and identified 12
possible NRPS or polyketide synthase-nonribosomal peptide
synthetase (PKS–NRPS) clusters. One of the identified clusters
encodes two separate single-module NRPSs, QulA and QulB, as
well as a methyltransferase QulM, thereby fulfilling the minimum
requirement for the biosynthesis of 1 (Figure 2A). A putative IDO
(QulI) required for 2 biosynthesis from L-tryptophan is also
included in this gene cluster, which further supports this gene
cluster as a putative target. Some additional enzymes are also
encoded in this gene cluster; among these are QulF (a putative
FMN-dependent dehydrogenase) and QulP (a putative
monocarboxylate permease), etc. (Figure 2A and Table S6). Two
highly similar gene clusters were also found in the genomes of
Aspergillus novofumigatus IBT 16806[9] and Penicillium steckii,[10]
in which the homologous genes of qulA, qulB, and qulF are
present, while the qulM homologs are absent (Figure S1).
1
Comparison of UV and H NMR spectra (Figure S6.1 and Table
S3) of 1 and 3 suggested that they are very similar, but the N-
methyl signal (δH = 3.85) in the spectrum of 1 is missing in that of
3 (Table S3, Figure S10.1, and S10.2). This result evidenced that
QulM should be responsible for the methylation at the N-4 position.
However, when 3 and S-adenosylmethionine (SAM) were
incubated with QulM expressed from E. coli (Figure S7C), no new
product was detected (Figure S8A, trace ii), indicating that 3 is not
the real substrate for QulM. Meanwhile, inactivation of qulF led to
the accumulation of compounds 4 and 5 (Figure 3A, trace v;
Figure S6.2B and S6.3B); NMR analysis confirmed the chemical
structure of 4 as N-methyl-kynurenine (Table S4, Figure S10.3-
S10.5), suggesting that QulM may utilize 2 as the substrate.
Indeed, incubation of QulM with either L- and D-kynurenine led to
the installation of the N-methyl group in both cases, with a
turnover rate of 197.6 min-1 and 1.5 min-1, respectively (Figure 3B,
trace ii, Figure S9A and S9B), which confirmed the function of
QulM as a kynurenine methyltransferase and also suggested that
the methylation at the N-4 position happens prior to the formation
of the quinolone-γ-lactam core structure. Furthermore, structural
characterization of 5 was elucidated as N-methyl-4-carboxyl-2-
quinolone (Table S5, Figure S10.6-S10.8). It was reported that
superfluous intracellular 2 would be transformed into a 4-(2-
aminophenyl)-2,4-dioxobutanoate intermediate by kynurenine
aminotransferase (KAT), followed by a spontaneous dehydration
to form kynurenic acid (6)[7a] (Figure S9C and S9E). We
speculated that the endogenous P. citrinum KAT (PcKAT) might
catalyze the deamination of 4 to form 5. When we expressed the
PcKAT in E. coli (Figure S7F) and incubated it with 4, 5 was
indeed produced (Figure 3B, trace iv, and Figure S9D). Our
results here confirmed 5 was derived from 4 in ΔqulF as a shunt
product, while 4 could be an intermediate that would be further
modified by QulF.
To unequivocally link the qul gene cluster to the biosynthesis of 1,
we performed a set of genetic knockouts using a split-marker
recombination approach (Figure S2 and S3).[11] The metabolic
profiling showed that ΔqulA, ΔqulB, ΔqulF, and ΔqulM mutants
were no longer able to produce 1, while deletion of qulP, orf1, orf2,
and orf3 did not influence 1 production (Figure 2B and S4). This
result confirmed that this gene cluster is responsible for 1
biosynthesis. Deletion of qulI did not completely abolish the
production of 1, but led to substantial attenuation of the titer to
~5% of that in the wild-type (WT) strain (Figure 2B, trace vi). QulI
is predicted to be a putative IDO and possibly involved in the
biosynthesis of the primary metabolite 2.[7b] The remaining low-
level production of 1 in ΔqulI suggests that there could be other
endogenous IDO genes in the P. citrinum genome. A local BLAST
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202005770
Angewandte Chemie International Edition
RESEARCH ARTICLE
Knocking out the NRPS gene (ΔqulA or ΔqulB) revealed a high-
level accumulation of compound 7 compared with the WT strain
(Figure 3A, trace ii and iii, and Figure S6.4B), which was
characterized by NMR as N-methyl-4-hydroxy-2-quinolone (Table
S5, Figure S10.9-S10.11). Comparing the structures of 7 with 5
(the shunt product accumulated in ΔqulF), the lack of a C-10
carboxyl group in 7 (Figure 3C) suggests QulF might function as
an oxidative decarboxylase to further modify a precursor
substrate for the downstream NRPSs.
spectrometry (HRMS) and the HRMS-MS/MS analysis (Figure
S13), suggesting that the structure of
8
is 2-
aminobenzoylacetamide (Figure 3C). When we performed the
single amino acid substitution of His289 to Ala (the conserved
His289 was proposed to be part of the active sites and involved
in the initial deprotonation step),[12a] the resulting QulFH289A did not
show any activity in the formation of 8 (Figure 3B, trace vi), which
confirmed the catalytic function of QulF. This result indicates that
QulF could catalyze two tandem reactions, including
Figure 3. Metabolite analysis in qul knockout mutants and enzymatic in vitro assays. (A) Mass filter analyses of 3 (m/z=257.1 [M+H]+), 4 (m/z=223.1 [M+H]+),
5 (m/z=204.1 [M+H]+), and 7 (m/z=176.1 [M+H]+) in P. citrinum WT and qul knockout mutants. (i) WT, (ii) △qulA, (iii) △qulB, (iv) △qulM, (v) △qulF, (vi) △qulI,
(vii) △qulP, (viii) standards of 3, 4, 5, and 7. (B) HPLC analysis of in vitro assays (UV 254 nm). (i) QulM in vitro assay control, 2 + SAM, (ii) QulM + 2 + SAM,
(iii) PcKAT in vitro assay control, 4 + P5P + PYR, (iv) 4 + P5P + PYR + PcKAT, (v) 4 + QulF. (vi) 4 + QulFH289A, (vii) QulF in vitro assay control, 4 + buffer, (viii)
2, 4, 5, and 7 standards. (C) Proposed pathway for 4, 5, 7, and 8 biosynthesis. (D) Proposed catalytic mechanism of QulF. Abbreviations: SAM, S-adenosyl
methionine; P5P, pyridoxal-5-phosphate; PYR, pyruvate.
dehydrogenation and decarboxylation, in which the His289
BLAST analysis revealed that QulF belongs to an FMN-
initiates the reaction by deprotonating the amino group of 4 and
transferring a hydride to FMN to give imino-4 and FMNH2. FMNH2
could then be oxidized by O2 to generate peroxy-flavin anion
(FMN-4a-OO-), followed by the nucleophilic attack on the imine
group and subsequent decarboxylation, resulting in the
production of 8 (Figure 3D). Therefore, QulF could function as an
FMN-dependent redox enzyme catalyzing the oxidative
decarboxylation of the amino acid N-methyl-kynurenine (4).
dependent α-hydroxy acid dehydrogenase family (Table S9 and
Figure S11), with the highest similarity (39.29%) to lactate 2-
monooxygenase (Mycobacterium smegmatis), which was well
characterized
to
catalyze
the
dehydrogenation
and
decarboxylation of lactate to form acetate.[12] However, no study
has reported that this type of enzyme family can function on amino
acid substrates. To test the function of QulF and probe the product
to be utilized by QulA or QulB NRPS, we expressed and purified
QulF (Figure S7D) from E. coli. The LC-MS analysis of denatured
QulF confirmed that FMN is the cofactor of QulF (Figure S12).
Incubation of 4 with QulF yielded a new product, 8, with
m/z=193.1 [M+H]+ (Figure 3B, trace v, and S6.2C), while QulF
failed to produce any new compounds when 2 was used as the
substrate (Figure S8B), suggesting that the methylated 4 is a
more favored substrate for QulF. The molecular formula of 8 was
determined to be C10H12N2O2 based on high-resolution mass
In addition to 8, a small amount of 7 was also detected from the
QulF assays based on LC-MS analysis (Figure 3B, trace v, and
Figure S14). Because 7 was also evident in the ΔqulA and ΔqulB
metabolism, we speculated that the spontaneous hydrolysis of the
amide group of 8 might also occur to give 7 through the unstable
intermediate 9 (N-methyl-2-aminobenzoylacetate). However, the
sluggish non-enzymatic hydrolysis of the amide group is unlikely
to maintain the highly efficient production of 7 as shown in the
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202005770
Angewandte Chemie International Edition
RESEARCH ARTICLE
ΔqulA or ΔqulB mutant, and there might be an endogenous
amidase outside of the qul gene cluster that catalyzes the amide
hydrolysis and accelerates the hydrolysis process.
Next, we sought to find the actual precursor substrates for both
QulA and QulB. Compound 7 was accumulated in ΔqulA and
ΔqulB knockout strain metabolites, and also observed in the QulF
and amidase in vitro assays; however, 7 does not possess a
carboxylic group to be utilized by the NRPS A domain. Thus, we
proposed that the unstable compound 9 might be the actual
precursor substrate of QulA or QulB for the biosynthesis of 1.
Although multiple strategies were employed to synthesize 9, the
methylation step of the highly reactive primary amine group was
Amide hydrolysis of 8 by amidase. To verify our hypothesis, we
then aimed to identify the amidase that might locate outside of the
qul biosynthetic gene cluster and catalyze the amide hydrolysis of
8 to 9. BLAST analysis against the P. citrinum genome using an
acetamidase from Aspergillus oryzae[13] showed eleven hits
containing amidase signature sequence (Table S8, Figure
S15).[14] Next, we cloned and overexpressed six amidases (#2, #4,
#7, #9, #10, and #11) from E. coli for further characterization
(Figure S16A). Five of these amidases were successfully
expressed, except for amidase #9 (Figure S16B). Interestingly,
incubation of QulF and compound 4 with amidase #7 (9 hours) led
to complete conversion of 8 to 7, while the other amidases failed
to further facilitate the reaction (Figure 4A). In addition, a new
peak with m/z=194.1 [M+H]+ was also detected in the reaction
with amidase #7 (Figure 4A, trace iii). According to the UV and
mass data (Figure S6.2E), we proposed this new intermediate
could be the unstable precursor 9, which might be spontaneously
cyclized to 7 (Figure 3C). When a set of time-course assays using
amidase #7 were performed, the complete conversion of 8 to 7
was observed within 20 min (Figure 4B). These results indicate
that the intracellular amidase (amidase #7) could hydrolyze 8 to
9, followed by spontaneous cyclization to afford 7.
not successful. Considering that ΔqulM produced
3
(desmethylated 1, Figure 3A, trace iv), we speculated that QulA
and QulB could also react with 2-aminobenzoylacetate (10, the
desmethylated 9) and L-Ile to produce 3. To verify our hypothesis,
we chemically synthesized compound 10 (Figure S6.2D) and
verified its structure by NMR (Figure S10.12). We also tried to
methylate 10 to produce 9 through chemical synthesis, but doing
so proved challenging because 10 was also unstable in the
reaction solution and quickly converted to compound 11.[17] We
purified and characterized the structure of 11 via 1D and 2D NMR
1
(Figure S6.4A, Table S5, and Figure S10.13-S10.15). H NMR
and 13C NMR spectra matched the reported data of 4-hydroxy-2-
quinolone,[18] which is a desmethylated version of 7. The observed
spontaneous dehydration of 10 to 11 supported our hypothesis
that compound 7 should be the spontaneously dehydrated
product of compound 9. Notably, when 10 was incubated with
QulM and SAM, 9 was not produced, indicating that 10 was not a
substrate of QulM (Figure S8C) and thus could not be synthesized
enzymatically.
We then heterologously expressed and purified QulA and QulB
(Figure S7) from E. coli BAP1.[19] To measure the adenylation
activity of the A domains, the colorimetric detection of free
pyrophosphate assay was performed.[20] The high level of
pyrophosphate production was only observed when L-Ile and 10
were used as substrates for QulA and QulB, respectively (Figure
5B). These results suggested that L-Ile is the preferred substrate
for QulA, and that β-keto acid 10 is the preferred substrate for
QulB. It is worth noting that, although numerous nonproteinogenic
amino acids participated in NRPS biosynthesis, an NRPS A
domain recognizing a β-keto acid has been rarely reported. In
earlier reports, ten amino acid residues (A1-A10) have been
proposed to be involved in the substrate recognition, based on
information from the adenylation domain PheA structure in
GrsA.[21] Alignment of selective NRPS A domains showed the
highly conserved acidic amino acid Asp (A1-position, interacts
with the α-amino group of the substrate amino acids) in QulB is
changed into the basic amino acid His423 (Figure S18A). Docking
analysis of the QulB-A domain with 9 showed a favorable
interaction between the His423 (A1-position) and the β-ketone
group of 9 (Figure S18B). We thus proposed the His423 in A1-
position is crucial to accommodate the β-keto group, which
enables the NRPS QulB to recognize and incorporate β-keto acid.
Figure 4. Amide hydrolysis of 8 by amidase #7. (A) Amidase assays for 8
hydrolysis to 9. (i)-(vi) enzymatic assays of QulF, 4, and amidase #2, #4,
#7, #9, #10, and #11, respectively. (vii) standards of 4 and 7. (B) Time-
course assays of amidase #7. (i)-(iii) amidase #7 reaction for 0 min, 5 min,
and 20 min, respectively. (iv) control, incubation of 4 and amidase #7 for 20
min. (v) standards of 4 and 7.
QulA and QulB are involved in the formation of quinolone-γ-
lactam hybrid. NRPS module analysis shows that QulB consists
of condensation (C), adenylation (A), and thiolation (T) domains
(Figure 5A). A close examination showed that the conserved
HHxxxDG motif is completely missing in the C domain of QulB.[15]
This type of inactivated C domain was named the C* domain and
proposed to play a structural role for protein stability.[15] QulA
consists of C-, A-, T-, and reductase (R) domains (Figure 5A). The
phylogenetic analysis showed that the C-terminal domain of QulA
falls into the same group as the previously reported NAD(P)H-
independent R* domains found in the fungal PKS-NRPS hybrids,
e.g., CpaA, ApdA, and TenS (Figure S17), which usually catalyze
the Dieckmann condensation between a β-keto intermediate and
an amino acid.[16] Furthermore, like other R* domains, the C-
terminal domain of QulA lacks the critical Ser-Tyr-Lys catalytic
triad and the conserved NAD(P)H binding sites in typical
reductase domain protein (Figure S17B).
When QulA and QulB were incubated with 10 and L-Ile in the
presence of ATP, the formation of 3 was observed (Figure 5C,
trace ii, time-course analysis shown in Figure S19A), which
confirmed that QulA and QulB could catalyze the formation of the
tricyclic quinolone-γ-lactam hybrid. In the absence of QulA and
QulB, compound 10 spontaneously converted to compound 11
(desmethylated version of 7) (Figure 5C, trace i). The in vitro
enzymatic synthesis of 3 in the absence of NAD(P)H strongly
supported the supposition that the C-terminal domain of QulA is
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202005770
Angewandte Chemie International Edition
RESEARCH ARTICLE
an R* domain that could catalyze the non-redox Dieckmann
condensation.
group (C-3a) and undergo spontaneous cyclization to form the
final structure of the quinolone-γ-lactam hybrid structure (Figure
5E). Such consecutive dehydrocyclizations were also observed in
other CT or R domain that catalyzed fungal alkaloids biosynthesis,
such as those observed in Tryptoquialanine, Asperlicins, and the
recently discovered Peramine.[23]
Conclusion
The R* domain catalyzed Dieckmann condensation was only
reported in the machinery of PKS-NRPS hybrids in which the key
intermediate β-keto intermediate was synthesized through PKS
machinery.[24] Herein, our study unraveled a new pathway to
generate the β-keto acid via an FMN-dependent redox enzyme
QulF catalyzed dehydrogenation and decarboxylation, as well as
the amidase #7 mediated amide hydrolysis reaction. Based on
gene inactivation studies and enzymatic assays, reaction
mechanisms that rationalize the formation of the quinolone-γ-
lactam was proposed. Of note, the mutation of a highly conserved
residue in the A1 position could enable QulB to incorporate a β-
keto acid and decouple R* domain-containing NRPS from a PKS
machinery, which expands the paradigm for the biosynthesis of
quinolone-γ-lactam natural products via Dieckmann condensation.
Acknowledgments
Figure 5. Characterization of the functions of QulA and QulB. (A) QulA and
QulB functional domain predication. (B) A domain adenylation activity
assays. (C) HPLC analysis (UV 254 nm) of in vitro QulA and QulB assays.
(i) Control, 10 + Ile + ATP + Mg2+; (ii) 10 + Ile + ATP + Mg2+ + QulA + QulB;
(iii) 3 standard; (iv) synthesized and purified 10. Compound 10 is unstable
once purified and it is converted into 2,4-dihydroxyquinoline (11) in
aqueous solution. (D) One-pot assay for production of 1. (i) 4 + QulA +
QulB + QulF + amidase #7 + Ile + Mg2+, (ii) 4 + QulA + QulB + QulF +
amidase #7 + Ile + ATP + Mg2+. (iii) standards. (E) Proposed reaction
mechanism for quinolone-γ-lactam ring formation. Abbreviation: Ile, L-
isoleucine.
We thank Dr. Christopher T. Walsh for helpful discussion of the
QulF mechanism and proofreading the manuscript.
Funding Sources
This work was supported by a Robert A. Welch Foundation grant
(C-1952) and a Hamill Innovation Award (Hamill Foundation) to X.
G.
Finally, we tried to incubate 4, L-Ile, ATP, Mg2+, QulF, QulA, QulB,
and amidase #7 in a one-pot reaction to determine if 1 could be
produced. After 9 hrs incubation, we indeed detected the
production of 1 (Figure 5D, trace ii), while only 7 was observed
when ATP was omitted in the reaction (Figure 5D, trace I, and
Figure S19C). This result suggested that, although 9 is unstable,
it could be efficiently incorporated into the downstream NRPS
assembly line, which competes with the rapid and spontaneous
self-cyclization to form 7. We thus proposed the mechanism for
the quinolone-γ-lactam ring formation as shown in Figure 5E. The
NRPSs QulA and QulB should incorporate L-Ile and β-keto acid 9
(or 10), respectively. Subsequently, the C domain of QulA
catalyzes a condensation step to assemble the β-keto amide
Notes
The authors declare no competing financial interest.
Keywords: Quinolone-lactam biosynthesis
incorporation • NRPS • Dieckmann condensation
•
β-keto acid
References
[1]
a) J. Caruano, G. Muccioli, R. Robiette, Org. Biomol.
Chem. 2016, 14, 10134-10156; b) A. M. Emmerson, A. M.
Jones, J. Antimicrob. Chemother. 2003, 51, 13-20.
S. E. Berning, Drugs 2001, 61, 9-18.
intermediate, which further undergoes
a
Dieckmann
[2]
[3]
[4]
condensation catalyzed by the R* domain of QulA to form a
ketopyrrolidinone intermediate 12 (or 13) and gets released from
the NRPS assembly lines. In the one-pot assay for the
biosynthesis of 1, we indeed detected a small peak with
m/z=289.1 [M+H]+, corresponding to the mass of the intermediate
12 (Figure S19C). Similarly, a peak with m/z=275.1 [M+H]+, which
matches the mass of 13, was also detected when QulA and QulB
were incubated with 10 and Ile (Figure S19B). In an earlier study
for quinolactacin B2 total synthesis, an intermediate 3-
acyltetramic acid, an analogue of 12 and 13, was also detected.[22]
It was reported that 3-acyltetramic acid cyclized spontaneously on
silica gel to form the quinolone-γ-lactam structure.[22] Therefore, it
is likely that the amine N-4 in 12 (or 13) could attack the keto-
M. Dooley, G. L. Plosker, Drugs 2000, 60, 871-893.
a) G. da Siliva, R. Moreira, A. M. S. Silva, in Studies in
Natural Products Chemistry, Vol. 62 (Ed.: R. Atta ur),
Elsevier, 2019, pp. 433-453; b) B. Clark, R. Capon, E.
Lacey, S. Tennant, J. Gill, Org. Biomol. Chem. 2006, 4,
1512-1519; c) N. Kakinuma, H. Iwai, S. Takahashi, K.
Hamano, T. Yanagisawa, K. Nagai, K. Tanaka, K. Suzuki,
F. Kirikae, T. Kirikae, A. Nakagawa, J. Antibiot. 2000, 53,
1247-1251; d) W. Kim, N. Song, I.-D. Yoo, J. Antibiot.
2001, 54, 831-835.
[5]
[6]
a) X. Zhang, W. Jiang, Z. Sui, J. Org. Chem. 2003, 68,
4523-4526; b) S.-J. Park, K.-N. Cho, W.-G. Kim, K.-I. Lee,
Tetrahedron Lett. 2004, 45, 8793-8795.
T. Sasaki, S. Takahashi, K. Uchida, S. Funayama, M.
Kainosho, A. Nakagawa, J. Antibiot. 2006, 59, 418-427.
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202005770
Angewandte Chemie International Edition
RESEARCH ARTICLE
[7]
[8]
[9]
a) A. Badawy, Int. J. Tryptophan Res. 2017, 10,
1178646917691938; b) J. Savitz, Mol. Psychiatry 2019,
25, 131-147.
K. Blin, S. Shaw, K. Steinke, R. Villebro, N. Ziemert, S. Y.
Lee, M. Medema, T. Weber, Nucleic. Acids. Res. 2019,
47, W81-W87.
I. Kjærbølling, T. Vesth, J. Frisvad, J. Nybo, S. Theobald,
A. Kuo, P. Bowyer, Y. Matsuda, S. Mondo, E. Lyhne, M.
Kogle, A. Clum, A. Lipzen, A. Salamov, C. Ngan, C.
Daum, J. Chiniquy, K. Barry, K. Labutti, M. Andersen,
Proc. Natl. Acad. Sci. U. S. A. 2018, 115, E753-E761.
J. C. Nielsen, S. Grijseels, S. Prigent, B. Ji, J. Dainat, K.
Nielsen, J. Frisvad, M. Workman, J. Nielsen, Nat.
Microbiol. 2017, 2, 17044.
[10]
[11]
[12]
R. S. Goswami, Methods Mol. Biol. 2012, 835, 255-269.
a) C. J. Murphy, M. G. Shepherd, P. A. Sullivan,
Biochemistry 1983, 22, 1665-1669; b) C. Walsh, O.
Lockridge, V. Massey, R. Abeles, J. Biol. Chem. 1973,
248, 7049-7054.
[13]
[14]
[15]
[16]
K. Gomi, K. Kitamoto, C. Kumagai, Gene 1992, 108, 91-
98.
M. P. Patricelli, M. A. Lovato, B. F. Cravatt, Biochemistry
1999, 38, 9804-9812.
B. D. Ames, C. T. Walsh, Biochemistry 2010, 49, 3351-
3365.
a) X. Liu, C. T. Walsh, Biochemistry 2009, 48, 8746-8757;
b) J. W. Sims, E. W. Schmidt, J. Am. Chem. Soc. 2008,
130, 11149-11155; c) W. Xu, X. Cai, M. E. Jung, Y. Tang,
J. Am. Chem. Soc. 2010, 132, 13604-13607.
A. K. Bera, V. Atanasova, H. Robinson, E. Eisenstein, J.
P. Coleman, E. C. Pesci, J. F. Parsons, Biochemistry
2009, 48, 8644-8655.
[17]
[18]
[19]
[20]
A. G. Osborne, J. M. Buley, H. Clarke, R. C. H. Dakin, P. I.
Price, J. Chem. Soc., Perkin Trans. 1 1993, 2747-2755.
B. A. Pfeifer, S. J. Admiraal, H. Gramajo, D. E. Cane, C.
Khosla, Science 2001, 291, 1790-1792.
C. Maruyama, H. Niikura, M. Takakuwa, H. Katano, Y.
Hamano, in Nonribosomal Peptide and Polyketide
Biosynthesis, Vol. 1401, Humana Press, New York, NY,
2016, pp. 77-84.
[21]
[22]
[23]
G. L. Challis, J. Ravel, C. A. Townsend, Chem. Biol. 2000,
7, 211-224.
K. Tatsuta, H. Misawa, K. Chikauchi, J. Antibiot. 2001, 54,
109-112.
a) X. Gao, S. W. Haynes, B. D. Ames, P. Wang, L. P.
Vien, C. T. Walsh, Y. Tang, Nat. Chem. Biol. 2012, 8, 823-
830; b) S. W. Haynes, X. Gao, Y. Tang, C. T. Walsh, J.
Am. Chem. Soc. 2012, 134, 17444-17447; c) D. Berry, W.
Mace, K. Grage, F. Wesche, S. Gore, C. L. Schardl, C. A.
Young, P. P. Dijkwel, A. Leuchtmann, H. B. Bode, B.
Scott, Proc. Natl. Acad. Sci. U. S. A. 2019, 116, 25614-
25623.
[24]
a) X. Liu, C. T. Walsh, Biochemistry 2009, 48, 8746-8757;
b) J. W. Sims, E. W. Schmidt, J. Am. Chem. Soc. 2008,
130, 11149-11155; c) W. Xu, X. Cai, M. E. Jung, Y. Tang,
J. Am. Chem. Soc. 2010, 132, 13604-13607; d) M. N.
Heneghan, A. A. Yakasai, K. Williams, K. A. Kadir, Z.
Wasil, W. I. M. Bakeer, K. M. Fisch, A. M. Bailey, T. J.
Simpson, R. J. Cox, C. M. Lazarus, Chem. Sci. 2011, 2,
972-979; e) T. B. Kakule, D. Sardar, Z. Lin, E. W. Schmidt,
ACS Chem. Biol. 2013, 8, 1549-1557; f) B. J. C. Law, Y.
Zhuo, M. Winn, D. Francis, Y. Zhang, M. Samborskyy, A.
Murphy, L. Ren, P. F. Leadlay, J. Micklefield, Nat. Catal.
2018, 1, 977-984; g) T. Awakawa, X.-L. Yang, T.
Wakimoto, I. Abe, Chembiochem 2013, 14, 2095-2099.
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202005770
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
We reveal that the novel pharmacophore quinolone-γ-lactam hybrid is constructed from the NRPSs catalyzed condensation of an
unnatural and unstable β-keto acid with isoleucine. Most intriguingly, this β-keto acid was biosynthesized from tryptophan via extensive
modifications catalyzed by a dedicated set of enzymes, which represents a new strategy used by Nature to generate structural
complexities.
7
This article is protected by copyright. All rights reserved.