In vitro reconstitution of the biosynthetic pathway of 3-hydroxypicolinic acid

Article abstract of DOI:10.1039/c8ob02972e

3-Hydroxypicolinic acid (3-HPA) is an important pyridine building block of bacterial secondary metabolites. Although the main biosynthetic pathways of these metabolites have been identified and well characterized, the enzymatic mechanism underlying the biosynthesis of 3-HPA has yet to be elucidated. In this work, we successfully reconstituted the complete biosynthetic pathway of 3-HPA in vitro. We showed that an l-lysine 2-aminotransferase, a two-component monooxygenase, and a FAD-dependent dehydrogenase are required to convert l-lysine to 3-HPA. We further demonstrated that 3-HPA does not derive from the direct hydroxylation of the picolinic acid at C-3, but from a successive process of C-3 hydroxylation of the piperideine-2-carboxylic acid and tautomerization of the produced 3-hydroxyl dihydropicolinic acid. Therefore, this study unveils the unusual assembly logic of 3-HPA and sheds light on the potential of engineering the 3-HPA pathway for generating novel pyridine-based building blocks.

Full text of DOI:10.1039/c8ob02972e

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
COMMUNICATION
In vitro reconstitution of the biosynthetic pathway
of 3-hydroxypicolinic acid†
Cite this: Org. Biomol. Chem., 2019,
7, 454
1
a
a
a
b
a
a,c
Xuan Yun,‡ Qian Zhang,‡ Meinan Lv,‡ Hai Deng,
Zixin Deng* and Yi Yu
*
Received 28th November 2018,
Accepted 10th December 2018
DOI: 10.1039/c8ob02972e
rsc.li/obc
4
,5
3
-Hydroxypicolinic acid (3-HPA) is an important pyridine building pound in the area of analytical chemistry. 3-HPA is also a
block of bacterial secondary metabolites. Although the main bio- key building unit of several bacterial secondary metabolites,
synthetic pathways of these metabolites have been identified and including the depsipeptide antibiotics viridogrisein (etamy-
6
7
8
9
well characterized, the enzymatic mechanism underlying the bio- cin), virginiamycin S, pristinamycin I, pyridomycin, and
1
0
synthesis of 3-HPA has yet to be elucidated. In this work, we suc- the benzoxazole antibiotics A33853 (Fig. 1A). The main bio-
cessfully reconstituted the complete biosynthetic pathway of synthetic gene clusters of these compounds have been identi-
-HPA in vitro. We showed that an L-lysine 2-aminotransferase, a fied recently.
two-component monooxygenase, and a FAD-dependent dehydro- is, however, only partially proposed.
genase are required to convert L-lysine to 3-HPA. We further Labelling studies have shown that L-lysine is the biosyn-
demonstrated that 3-HPA does not derive from the direct hydroxy- thetic precursor of 3-HPA in both etamycin and virginiamycin
1
1–15
3
The chemical logic of the assembly of 3-HPA
1
1
1
6–18
lation of the picolinic acid at C-3, but from a successive process of S producing strains.
Namwat et al. then demonstrated
C-3 hydroxylation of the piperideine-2-carboxylic acid and tauto- that an L-lysine 2-aminotransferase (L2AT), the gene product
merization of the produced 3-hydroxyl dihydropicolinic acid. of visA, is responsible for the L-lysine cyclization reaction in
1
2
Therefore, this study unveils the unusual assembly logic of 3-HPA virginiamycin
S
biosynthesis.
Moreover, the external
and sheds light on the potential of engineering the 3-HPA pathway addition of 3-HPA into the ΔvisA mutant could recover the
for generating novel pyridine-based building blocks.
production of virginiamycin S, supporting the idea that the
cyclization of L-lysine is the initial reaction in 3-HPA biosyn-
1
2
thesis. In general, L2AT catalyzes the α-transamination of
L-lysine to form an intermediate, 2-keto-6-aminocaproic acid,
followed by spontaneous cyclization and dehydration to give
the piperideine-2-carboxylic acid (P2C), which is probably the
Introduction
Pipecolic acid and its derived moieties are important constitu-
1
,18
1
–3
first intermediate in the 3-HPA pathway (Fig. 1B).
ents of many natural and synthetic bioactive molecules. As a
pipecolic acid derivative, 3-hydroxypicolinic acid (3-HPA, 1)
has proved useful as an excellent matrix material of matrix-
assisted laser desorption/ionization (MALDI) for the analysis
of nucleic acids, resulting in intensive research on this com-
Bioinformatics analysis has revealed highly homologous
open reading frames (orfs) of visA in each of the biosynthetic
pathways of the 3-HPA-containing antibiotics (Fig. 1A), imply-
ing that the L2AT catalyzed reaction is a key step in 3-HPA
biosynthesis.
An L2AT homolog was also found in the biosynthetic
pathway of the peptidyl nucleoside antibiotic nikkomycin D
a
Key Laboratory of Combinatorial Biosynthesis and Drug Discovery (Ministry of
Education), School of Pharmaceutical Sciences, Wuhan University, 185 East Lake
Road, Wuhan 430071, P. R. China. E-mail: yu_yi@whu.edu.cn, zxdeng@sjtu.edu.cn
that contains
a 5-hydroxypicolinic acid (5-HPA) moiety
b
(Fig. 1A). 5-HPA and 3-HPA have the same pyridyl scaffold,
implying that the assembly of these two precursors may share
common enzymatic steps. The biosynthesis of the pyridyl
residue in nikkomycin D has been extensively studied, in
which the L2AT encoded by nikC converts L-lysine to P2C first,
followed by a four-electron oxidation to yield the picolinic acid
Marine Biodiscovery Centre, Department of Chemistry, University of Aberdeen,
Aberdeen AB24 3UE, Scotland, UK
c
Hubei Key Laboratory of Industrial Microbiology, Hubei Provincial Cooperative
Innovation Center of Industrial Fermentation, Key Laboratory of Fermentation
Engineering (Ministry of Education), Hubei University of Technology, Wuhan 430068,
P. R. China
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/
c8ob02972e
These authors contributed equally to this work.
1
9,20
(
PA, 2) catalyzed by the FAD-dependent oxidase NikD.
The
‡
homologs of NikD were also found in the biosynthetic path-
454 | Org. Biomol. Chem., 2019, 17, 454–460
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Communication
Fig. 1 Structures of the natural products containing 3-HPA and the genes putatively involved in 3-HPA biosynthesis. (A) Homologous genes of
bomA, bomB, bomD, and bomE, which were proposed to be responsible for the biosynthesis of the 3-HPA unit in A33853, are conserved in different
strains producing 3-HPA moiety-containing natural products. These strains are marked with different colors that are consistent with the corres-
ponding natural products they produce. (B) Proposed biosynthetic route of the 3-HPA moiety in A33853.
ways of the 3-HPA-containing antibiotics (Fig. 1A), suggesting
Recently, we have identified the gene cluster of the benzox-
that PA may be the next intermediate following P2C in the bio- azole antibiotics A33853 and assigned the genes putatively
1
1
synthetic route of 3-HPA (Fig. 1B, route a).
involved in the biosynthesis of 3-HPA (Fig. S1†). In this
However, Okumura and colleagues found that the addition work, we set out to solve the puzzle of 3-HPA biosynthesis by
of PA into the culture of the etamycin producing strain reconstitution of this compound’s assembly line in vitro. We
Streptomyces griseoviridus did not affect the production of eta- showed that four enzymes, including the pyridoxal phosphate
2
1
mycin. Exogenous supplementation of a series of 3-HPA ana- (PLP)-dependent L2AT BomE, the two-component monooxy-
logues including PA to the mutant strain of S. griseoviridus, of genase system BomA/D, and the flavin adenine dinucleotide
which the nikD homologous gene sgvS was inactivated, (FAD)-dependent dehydrogenase BomB, could convert
2
2
resulted in no recovery of the etamycin production. These L-lysine to 3-HPA in a one-pot reaction. We further demon-
results together demonstrated that PA is not the immediate strated that a hydroxylation reaction catalyzed by BomA/D
precursor of 3-HPA in etamycin biosynthesis. It thus raises an occurs at the C-3 of P2C instead of PA (Fig. 1B, route b), thus
interesting question that how and when the hydroxyl group at establishing the sequence of reactions involved in the assem-
C-3 of 3-HPA is introduced.
bly of 3-HPA.
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 454–460 | 455

View Article Online
Communication
Organic & Biomolecular Chemistry
Streptomyces host as N-terminal hexa-His-tagged fusion recom-
Results and discussion
binant proteins, followed by purification to near homogeneity
2
+
We have proposed a biosynthetic route of the 3-HPA moiety in by Ni -affinity chromatography (Fig. S2†).
1
1
A33853. As shown in Fig. 1B, the L2AT BomE first converts
As the first step of 3-HPA biosynthesis, the transamination
L-lysine into P2C, then the FAD-dependent dehydrogenase of L-lysine could involve the 2- or the 6-amino group, yielding
BomB aromatizes P2C to give PA, followed by P2C or piperideine 6-carboxylate, respectively. BomE shares
C-3 hydroxylation of PA to generate 3-HPA catalyzed by the high sequence homology (59% identity) with the well-charac-
two-component monooxygenase complex BomA/D (route a). terized L2AT NikC, indicating that BomE is responsible for the
Although bioinformatics analysis and the gene knockout conversion of L-lysine into P2P like NikC does. Our previous
1
1
experiments provided some evidence to support route a, bio- study has demonstrated this hypothesis through detecting the
chemical investigation is, however, required to delineate the reaction product of BomE using the derivatizing reagent ami-
exact timing of the actions of the enzymes involved, particu- nobenzaldehyde, which could react with P2C to form a dihy-
larly the enzymatic sequences of aromatization and hydroxy- droquinazolinium adduct with an absorption maximum at
1
1,23
lation where hydroxylation on C-3 could potentially occur 450 nm.
To gain further insight into the BomE catalyzed
before the aromatization of P2C (route b). Pieces of evidence reaction, we optimized the initial reaction components by sub-
supporting route b came from the studies on the etamycin bio- stituting 2-ketoglutarate with 2-oxobutyrate as the amino
synthesis, as mentioned above, that PA was found not be the acceptor, and reinvestigated the reaction by performing the
2
1,22
precursor of 3-HPA.
Our previous data also showed that liquid chromatography-electrospray ionization high-resolution
exogenous supplementation of PA into the bomE in-frame del- mass spectrometry (LC-ESI-HRMS) analysis directly. Two new
etion strain could not support the production of A33853, but compounds (3 and 4) with the same m/z as P2C (128.0726
+
yielded its analog in which the 3-HPA moiety was substituted [M + H] ) were then detected in the reaction mixture, while no
1
1
with PA. Thus, to elucidate the exact timing of the enzymatic transformation of L-lysine was observed when heat-inactivated
steps in the 3-HPA biosynthesis, we carried out the biochemi- BomE was used in the control reaction (Fig. 2A, trace iii and
cal investigation of these four enzymes. To this end, these four iv). The production of two isomers of P2C is consistent with
1
enzymes were heterologously overexpressed separately in the the previous report that P2C can exist in both imine (Δ -piperi-
+
Fig. 2 LC-ESI-HRMS analysis of in vitro enzymatic 3-HPA formation. (A) Extracted ion chromatograms (EICs) of m/z 140.0342 [M + H] (corres-
+
+
ponding to 3-HPA, 1), m/z 124.0393 [M + H] (corresponding to PA, 2), m/z 128.0726 [M + H] (corresponding to 3 and 4), and m/z 144.0655
[
+
M + H] (corresponding to 5, 6 and 7) are shown for (i) reference compound 1 (1 mM), (ii) reference compound 2 (1 mM), (iii) reaction consisting of
BomE and L-lysine, (iv) reaction consisting of boiled BomE and L-lysine, (v) reaction consisting of BomB/E and L-lysine, (vi) reaction consisting of
BomA/D and 2, (vii) reaction consisting of boiled BomA/D and 2, (viii) reaction consisting of BomE/A/D and L-lysine, (ix) one-pot reaction consisting
+
of BomE/A/D/B and L-lysine (two non-specific peaks of m/z 124.0393 [M + H] are also shown), (x) one-pot reaction control consisting of boiled
BomE/A/D/B and L-lysine. All reactions were carried out at 30 °C for 3 h and quenched with an equal volume of methanol. (B) MS/MS analysis of the
fragmentation patterns of compounds 1, 2, 3, 4, 5, 6, and 7.
456 | Org. Biomol. Chem., 2019, 17, 454–460
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Communication
2
deine-2-carboxylic acid) and enamine (Δ -piperideine-2-car-
2
4
boxylic acid) tautomeric forms. MS/MS analysis of both ions’
fragmentation patterns further confirmed that the structures
1
2
of 3 and 4 correspond to that of Δ - or Δ -P2C (Fig. 2B). These
findings demonstrate that BomE is an L2AT responsible for
the conversion of L-lysine into P2C.
In the following steps, P2C will undergo a series of oxi-
dation tailoring steps to give 3-HPA (Fig. 1B). Previous studies
on the biosynthesis of the pyridyl residue in nikkomycins
demonstrated that the FAD-dependent dehydrogenase NikD
could catalyze two successive steps of 2-electron oxidation of
2
0,25
P2C to generate PA.
with NikD, and it contains the conserved FAD binding site
C321) and three key catalytic residues (K358, D27, and R53) as
BomB shares 40% sequence similarity
(
NikD has (Fig. S3†), suggesting that BomB may catalyze the
similar reaction in 3-HPA biosynthesis as NikD does. Purified
BomB had a bright yellow color and major absorption peaks at
Fig. 3 Determination of the flavin substrate of BomA. UV-Vis spectra of
BomA catalyzed reactions in the presence of (A) NADH and FAD (for
3
92 nm and 456 nm (Fig. S2†), indicating the presence of a
3
h), (B) NADH and riboflavin (for 3 h), (C) NADH and FMN (for 3 h), and
flavin prosthetic group. To test the activity of BomB, incu-
(D) NADH and FMN (for a time gradient up to 24 h). The arrows indicate
bation of purified BomB together with BomE, PLP, FAD, and the characteristic UV absorption wavelength (450 nm) of riboflavin, FAD,
and FMN. The boiled BomA was used in all the control reactions.
L-lysine under aerobic conditions was carried out.
LC-ESI-HRMS analysis of the reaction showed that a new com-
+
pound (m/z 124.0393 [M + H] ) was produced at the same
retention time as the standard PA (Fig. 2A, trace v), the identity with the above results, BomA contains the conserved residues
of which was further confirmed to be PA through the MS/MS for binding FMN (F79) and NADH (H123, W145, and R148)
analysis (Fig. 2B). At this stage, these results supported the according to the structure data of its homologous protein
3
0
hypothesis that PA might be the precursor for 3-HPA biosyn- HpaC (26% identity, Fig. S5†). Therefore, BomA is an FMN
thesis (Fig. 1B, route a). We then continued to verify the reductase that can supply FMNH to BomD, for the sub-
2
remaining reaction steps towards 3-HPA.
sequent oxidation reaction.
BomD shows around 30% sequence identity with SsuD that
Bioinformatics analysis revealed that the gene bomA
encodes a putative flavin mononucleotide (FMN) reductase is involved in the oxidation of alkanesulfonate and with TgaE
and the gene bomD encodes a putative flavin-dependent that is responsible for C-17 hydroxylation of Soce-thuggacin
1
1
31,32
hydroxylase. BomA and BomD may constitute a two-com- B.
These enzymes belong to the small family of bacterial
for
ponent flavin-dependent monooxygenase that is responsible monooxygenases that use reduced flavins to activate O
2
for the installation of the C-3 hydroxyl group in 3-HPA. substrate oxidation, and flavins usually function as a co-sub-
Generally, flavin reductases (or named NAD(P)H:flavin oxido- strate rather than a prosthetic group. Sequence alignment of
reductases) are enzymes defined by their ability to catalyze the BomD with the secondary structural elements of the bacterial
reduction of free riboflavin, FMN, or FAD, by using reduced luciferase LuxA (42% similarity) showed that BomD contains
2
6
pyridine nucleotides, such as NADPH or NADH. The other the conserved residues Y208 and A239 that constitute the
3
3
flavin-dependent monooxygenase(s) component will then use active site of LuxA, further supporting the monooxygenation
the reduced flavins to catalyze the subsequent oxidation of function of BomD (Fig. S6†). Purified BomD from Streptomyces
2
7,28
various substrates.
We thus set out to investigate the flavin was colorless, like SsuD, indicating the absence of any bound
substrate of BomA before we tested the activity of the whole flavin cofactors. We then tested the activity of BomA/D by incu-
BomA/D system. Purified BomA was brown with major absorp- bating these two enzymes with PLP, FMN, and PA under
tion peaks at 365 nm and 453 nm (Fig. S2†), indicating the aerobic conditions. LC-ESI-HRMS monitoring of the reaction
presence of a flavin cofactor. Since enzymes belonging to the did not find the appearance of any new compound (Fig. 2A,
flavin reductase family are not flavoproteins but instead use trace vi and vii), suggesting that PA is not the substrate of
2
9
flavin as the substrate, we then set up the BomA reaction BomD. We then tested the reaction sequence shown in route b
under anaerobic conditions by incubating BomA (2.0 μM) with by incubating BomA/D with BomE together in the presence of
an excess amount of substrates (1.0 mM), which consist of PLP, FMN, and L-lysine under aerobic conditions. Three new
+
NADH and riboflavin, or FMN, or FAD. UV-visible (Vis) spec- compounds with the same m/z (144.0655 [M + H] ) were
troscopy monitoring of the different reaction mixtures showed detected in the reaction mixture (Fig. 2A, trace viii), and their
that only FMN could be reduced by BomA using NADH as the identities were further confirmed through the MS/MS analysis
electron donor (Fig. 3). We performed the same experiments to be three putative tautomers of hydroxylated P2C (5, 6, and
by replacing NADH with NADPH, and found that BomA could 7) (Fig. 2B). These data suggest that the two-component mono-
not use the latter as a substrate (Fig. S4†). In full agreement oxygenase BomA/D accepted P2C, not PA, as the substrate for
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 454–460 | 457

View Article Online
Communication
Organic & Biomolecular Chemistry
the C-3 hydroxylation reaction, thus ruling out route a in of system can be replaced by flavin reductases from other
3
4,35
which PA is proposed to be the precursor for 3-HPA origins to accomplish the same reaction.
In contrast to the
biosynthesis. biosynthetic gene clusters of the 3-HPA containing natural pro-
As mentioned above, BomB could accept P2C as the sub- ducts, only homologous genes of bomB and bomE were found
strate for the oxidation reaction. If the biosynthesis of 3-HPA in the nikkomycin gene cluster. The lack of homologous genes
follows route b, BomB could also aromatize the hydroxylated of bomA and bomD in the genome of the nikkomycin produ-
P2C. To verify this speculation, we reconstituted the biosyn- cing strain S. tendae (Fig. 1A) thus raises another interesting
thesis of 3-HPA in vitro by incubating all four enzymes BomA/ question about the chemistry logic underlying the
B/D/E together with exogenously supplied PLP, FMN, FAD, and C-5 hydroxylation in 5-HPA biosynthesis, which may be
L-lysine under aerobic conditions. LC-ESI-HRMS monitoring of different from the one(s) involved in 3-HPA biosynthesis.
the reaction mixture revealed a new compound with m/z
Based on the above data, we proposed the corresponding
40.0342 ([M + H] ) at the same retention time as the standard enzymatic mechanism that directs the biosynthesis of 3-HPA
-HPA (Fig. 2A, trace ix and x). MS/MS analysis of the com- (Scheme 1). BomE, the PLP-dependent L2AT, initiates the
+
1
3
pound’s fragmentation patterns further confirmed its identity whole process by catalyzing the α-transamination reaction of
as 3-HPA (Fig. 2B). We thus established the biosynthetic route L-lysine to generate the intermediate 6-amino-2-oxoadipate, fol-
of 3-HPA starting from L-lysine for the first time.
The homologous genes of bomA/B/D/E were also found in has two tautomeric forms (imine and enamine). BomA is a
the biosynthetic gene clusters of other 3-HPA containing reductase that uses NADH to reduce FMN to give FMNH
lowed by spontaneous cyclization to yield P2C, which generally
2
,
natural products (Fig. 1A), suggesting that the reactions cata- which subsequently reacts with molecular oxygen to yield a
lyzed by these four enzymes represent a common strategy for C4a-hydroperoxyflavin intermediate (Fl-OOH) in the monooxy-
3
-HPA biosynthesis. The only exception is the biosynthetic genase component BomD (the conversion process of flavin is
gene cluster of etamycin, in which no homologous gene of shown in Scheme 1). It is proposed that BomD catalyzes the
2
bomA was found (Fig. 1A). One reasonable explanation is that hydroxylation reaction using the enamine Δ -P2C as the
other putative FMN reductase(s) encoded outside of the etamy- nucleophile and Fl-4-OOH as the electrophile to give
cin gene cluster may associate with SgvF (the BomD homolog) 3-hydroxyl P2C, which is subsequently tautomerized into the
to constitute the two-component monooxygenase system. It more stable intermediate 3-keto-pyridine-2 carboxylic acid (5),
has been observed that there is no direct interaction between as evidenced that 5 is the dominant compound in the
the two proteins in most two-component monooxygenase LC-ESI-HRMS analysis (Fig. 2A, trace viii). The final stage of
systems reported, and the oxidoreductase partner of such kind 3-HPA biosynthesis is that BomB, a FAD-dependent dehydro-
Scheme 1 Putative reaction mechanism underlying 3-HPA biosynthesis.
458 | Org. Biomol. Chem., 2019, 17, 454–460
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Communication
genase, catalyzes the aromatization of 5 via two successive
steps of 4-electron oxidation. The first two-electron oxidation
occurs by abstracting the acidic proton at the C-4 of 5 to yield
5 S. N. Jackson and A. S. Woods, in The Encyclopedia of Mass
Spectrometry, ed. M. L. Gross and R. M. Caprioli, Elsevier
Ltd, 2016, vol. 9, pp. 124–131.
8
, which can tautomerize to 9. In the last step, BomB will use 9
6 Q. R. Bartz, E. John, G. P. Park, M. P. Knudsen and
R. M. Smith, USA: U. S. Pat, 3023204, 1962.
as the substrate for the second two-electron oxidation using
the nucleophilicity of the nitrogen of the piperidine ring as the
driving force to yield the final 3-HPA. Taken together, this
study not only confirms our previous hypothesis that four
enzymes BomA/B/D/E constitute the whole assembly line of
7 P. Crooy and R. Deneys, J. Antibiot., 1972, 25, 371.
8 D. Vazquez, in Mechanism of Action of Antimicrobial and
Antitumor Agents. Antibiotics, vol 3, ed. J. W. Corcoran,
F. E. Hahn, J. F. Snell and K. L. Arora, Springer, Berlin,
Heidelberg, 1975.
3
-HPA, but also reveals an unusual assembly logic underlying
the biosynthesis of 3-HPA.
9 K. Maeda, H. Kosaka, Y. Okami and H. Umezawa,
J. Antibiot., 1953, 6, 140.
1
1
1
1
0 K. H. Michel, L. D. Boeck, M. M. Hoehn, N. D. Jones and
M. O. Chaney, J. Antibiot., 1984, 37, 441–445.
1 M. A. Lv, J. F. Zhao, Z. X. Deng and Y. Yu, Chem. Biol., 2015,
Conclusion
Substituted PAs have wide applications in the pharmaceutical
industry as the building blocks for the synthesis of various bio-
logically active compounds. However, the regiospecific
functionalization of PA is difficult to achieve through chemical
strategies, thus making biocatalysis an attractive and appli-
cable supplementary technology. In this study, we have
established the biosynthetic pathway of 3-HPA by using an in
vitro reconstitution approach. Four enzymes, including an
L-lysine 2-aminotransferase, a two-component FMN-dependent
monooxygenase system, and a FAD-dependent dehydrogenase,
were demonstrated to convert L-lysine to 3-HPA successively.
Interestingly, the C-3 hydroxylation reaction occurs specifically
on P2C, thus revealing a novel biocatalyst that is different from
22, 1313–1324.
2 W. Namwat, Y. Kamioka, H. Kinoshita, Y. Yamada and
T. Nihira, Gene, 2002, 286, 283–290.
3 V. Blanc, P. Gil, N. BamasJacques, S. Lorenzon, M. Zagorec,
J. Schleuniger, D. Bisch, F. Blanche, L. Debussche,
J. Crouzet and D. Thibaut, Mol. Microbiol., 1997, 23, 191–
3
6
202.
14 T. T. Huang, Y. M. Wang, J. Yin, Y. H. Du, M. F. Tao, J. Xu,
W. Q. Chen, S. J. Lin and Z. X. Deng, J. Biol. Chem., 2011,
286, 20648–20657.
1
1
1
1
1
2
2
2
2
2
2
2
2
5 Y. C. Xie, B. Wang, J. Liu, J. C. Zhou, J. Y. Ma, H. B. Huang
and J. H. Ju, ChemBioChem, 2012, 13, 2745–2757.
6 D. J. Hook and L. C. Vining, J. Chem. Soc., Chem. Commun.,
3
7
the extensively studied bacterial pyridine hydroxylases. In
summary, this newly discovered 3-HPA biosynthetic pathway
may broaden our knowledge about L-lysine catabolism, and
pave the way for the future engineering of the corresponding
enzymes to create pyridine-based building blocks.
1973, 185–186, DOI: 10.1039/c39730000185.
7 A. A. Molinero, D. G. I. Kingston and J. W. Reed, J. Nat.
Prod., 1989, 52, 99–108.
8 J. W. Reed, M. B. Purvis, D. G. I. Kingston, A. Biot and
F. Gossele, J. Org. Chem., 1989, 54, 1161–1165.
9 C. Bruntner and C. Bormann, Eur. J. Biochem., 1998, 254,
3
47–355.
0 D. Venci, G. Zhao and M. S. Jorns, Biochemistry, 2002, 41,
5795–15802.
Conflicts of interest
1
There are no conflicts of interest to declare.
1 N. Nagato, Y. Okumura, R. Okamoto and T. Ishikura, Agric.
Biol. Chem., 1984, 48, 3135–3136.
2 Y. Xie and J. Ju, Weishengwu Xuebao, 2013, 53, 1179–
Acknowledgements
1
188.
3 K. Soda, H. Misono and T. Yamamoto, Biochemistry, 1968,
, 4110–4119.
4 Y. Nishina, K. Sato and K. Shiga, J. Biochem., 1991, 109,
05–710.
This work was supported by grants from the National Natural
Science Foundation of China (31570033, 31811530299, and
7
3
1870035 to Y. Y.), the Fundamental Research Funds for the
7
Central Universities (2042017kf0191 to Y. Y.), and the Newton
Mobility Grant Award (IEC\NSFC\170617 to Y. Y. and H. D.).
5 R. C. Bruckner, G. Zhao, D. Venci and M. S. Jorns,
Biochemistry, 2004, 43, 9160–9167.
6 M. Fontecave, J. Coves and J. L. Pierre, Biometals, 1994, 7,
3–8.
References
7 H. R. Ellis, Arch. Biochem. Biophys., 2010, 497, 1–12.
1
2
3
M. He, J. Ind. Microbiol. Biotechnol., 2006, 33, 401–407.
D. O’Hagan, Nat. Prod. Rep., 2000, 17, 435–446.
V. Vranova, L. Lojkova, K. Rejsek and P. Formanek, 29 V. Niviere, M. A. Vanoni, G. Zanetti and M. Fontecave,
Chirality, 2013, 25, 823–831. Biochemistry, 1998, 37, 11879–11887.
K. J. Wu, A. Steding and C. H. Becker, Rapid Commun. Mass 30 M. Okai, N. Kudo, W. C. Lee, M. Kamo, K. Nagata and
Spectrom., 1993, 7, 142–146. M. Tanokura, Biochemistry, 2006, 45, 5103–5110.
28 J. Sucharitakul, R. Tinikul and P. Chaiyen, Arch. Biochem.
Biophys., 2014, 555–556, 33–46.
4
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 454–460 | 459

View Article Online
Communication
Organic & Biomolecular Chemistry
3
1 K. Buntin, H. Irschik, K. J. Weissman, E. Luxenburger, 34 T. M. Louie, X. S. Xie and L. Y. Xun, Biochemistry, 2003, 42,
H. Blocker and R. Muller, Chem. Biol., 2010, 17,
42–356.
2 E. Eichhorn, C. A. Davey, D. F. Sargent, T. Leisinger and
7509–7517.
3
35 C. E. Jeffers, J. C. Nichols and S. C. Tu, Biochemistry, 2003,
42, 529–534.
3
3
T. J. Richmond, J. Mol. Biol., 2002, 324, 457–468.
3 A. J. Fisher, T. B. Thompson, J. B. Thoden, T. O. Baldwin 37 J. Q. Sun, L. Xu, Y. Q. Tang, F. M. Chen, J. J. Zhao and
36 M. Petersen and A. Kiener, Green Chem., 1999, 1, 99–106.
and I. Rayment, J. Biol. Chem., 1996, 271, 21956–21968.
X. L. Wu, Appl. Microbiol. Biotechnol., 2014, 98, 455–464.
460 | Org. Biomol. Chem., 2019, 17, 454–460
This journal is © The Royal Society of Chemistry 2019