Characterizing amosamine biosynthesis in amicetin reveals amig as a reversible retaining glycosyltransferase

Article abstract of DOI:10.1021/ja401016e

The antibacterial and antiviral agent amicetin is a disaccharide nucleoside antibiotic featuring a unique α-(1→4)-glycoside bond between amosamine and amicetose, characteristic of a retaining glycosylation. In this study, two key steps for amosamine biosynthesis were investigated: the N-methyltransferase AmiH was demonstrated to be requisite for the dimethylation in amosamine, and the glycosyltransferase AmiG was shown to be necessary for amosaminylation. Biochemical and kinetic characterization of AmiG revealed for the first time the catalytic reversibility of a retaining glycosyltransferase involved in secondary metabolite biosynthesis. AmiG displayed substrate flexibility by utilizing five additional sugar nucleotides as surrogate donors. AmiG was also amenable to sugar and aglycon exchange reactions. This study indicates that AmiG is a potential catalyst for diversifying nucleoside antibiotics and paves the way for mechanistic studies of a natural-product retaining glycosyltransferase.

Full text of DOI:10.1021/ja401016e

Communication
pubs.acs.org/JACS
Characterizing Amosamine Biosynthesis in Amicetin Reveals AmiG as
a Reversible Retaining Glycosyltransferase
Ruidong Chen,†,‡,⊥ Haibo Zhang,†,⊥ Gaiyun Zhang,§ Sumei Li,† Guangtao Zhang,† Yiguang Zhu,†
Jinsong Liu,∥ and Changsheng Zhang*,†
†CAS Key Laboratory of Tropical Marine Bio-resources and Ecology, RNAM Center for Marine Microbiology, Guangdong Key
Laboratory of Marine Materia Medica, South China Sea Institute of Oceanology, Chinese Academy of Sciences, 164 West Xingang
Road, Guangzhou 510301, China
‡Graduate University of the Chinese Academy of Sciences, Beijing 100049, China
§Key Laboratory of Marine Biogenetic Resources, Third Institute of Oceanography, State Oceanic Administration, No. 184 Daxue
Road, Xiamen 361005, China
∥State Key Laboratory of Respiratory Disease, Guangzhou Institutes of Biomedicine and Health, Chinese Academy of Sciences, 190
Kaiyuan Avenue, Guangzhou 510530, China
S
* Supporting Information
Amicetin (1a), a disaccharide pyrimidine nucleoside antibiotic
with antimicrobial and antiviral activities,8 was first isolated from
ABSTRACT: The antibacterial and antiviral agent
Streptomyces vinaceusdrappus NRRL 2363,9 and years later,
amicetin is a disaccharide nucleoside antibiotic featuring
a unique α-(1→4)-glycoside bond between amosamine
and amicetose, characteristic of a retaining glycosylation.
In this study, two key steps for amosamine biosynthesis
were investigated: the N-methyltransferase AmiH was
demonstrated to be requisite for the dimethylation in
amosamine, and the glycosyltransferase AmiG was shown
to be necessary for amosaminylation. Biochemical and
kinetic characterization of AmiG revealed for the first time
the catalytic reversibility of a retaining glycosyltransferase
involved in secondary metabolite biosynthesis. AmiG
displayed substrate flexibility by utilizing five additional
sugar nucleotides as surrogate donors. AmiG was also
amenable to sugar and aglycon exchange reactions. This
study indicates that AmiG is a potential catalyst for
diversifying nucleoside antibiotics and paves the way for
mechanistic studies of a natural-product retaining glyco-
syltransferase.
amicetin analogoues, including bamicetin (1b), plicacetin (2a),
and norplicacetin (2b) were also discovered (Scheme 1).10 The
Scheme 1. Proposed Amosamine Biosynthesis Pathway and
Structures of Amicetin (1a) and Related Compounds
any medicinal natural products are appended with a wide
Mvariety of sugar moieties, which are often critical for their
biological activity, specificity, and pharmacological properties.1
Attachments of sugars to natural-product scaffolds are usually
catalyzed by glycosyltransferases (GTs), which utilize an
activated donor sugar substrate (in most cases a nucleotide
sugar) to form a glycosidic bond with either inversion or
retention of configuration at the anomeric carbon of the donor
substrate.1 Biochemical characterizations of a number of natural-
product inverting GTs have revealed their reaction reversibility
and substrate flexibility and have expanded their synthetic utility
in glycodiversification of natural products by performing sugar
and aglycon exchange reactions.2 Despite the implication of
many retaining GTs in biosynthesis of natural products such as
spirotetronates,3 ansamycins,4 aminoglycosides,5 orthosomy-
cins,6 and nucleoside antibiotics,7 biochemically well-studied
retaining GTs for natural products remain scarce.5
most characteristic feature of the amicetin group of antibiotics is
the presence of an α-(1→4)-glycoside bond between amosamine
and amicetose with retention of configuration at the anomeric
carbon of amosamine. This structural feature sets amicetin as a
model for studying the biosynthetic mechanism of a retaining
GT. We recently cloned and characterized the 1a biosynthetic
gene cluster from S. vinaceusdrappus NRRL 2363 and proposed
five ami gene products for amosamine biosynthesis (Scheme 1),7
including the functionally characterized nucleotidylyltransferase
Received: January 29, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja401016e | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
AmiE,11 the thymidine diphosphate (TDP)-hexose-4,6-dehy-
dratase AmiU, the aminotransferase AmiB, the methyltransferase
AmiH, and the GT AmiG. Herein we describe the validation of
AmiH as an amosamine N-dimethyltransferase and AmiG as a
retaining GT tailoring amicetin biosynthesis. Notably, this study
demonstrates for the first time the reaction reversibility of a
retaining GT involved in secondary metabolite biosynthesis.
BLAST analysis showed that AmiH displays the highest
identity (34%) to a bacterial type-12 methyltransferase
(GenBank accession number ZP_08874093). AmiH also
distantly resembles the N-dimethyltransferase EryCVI for
TDP-desosamine biosynthesis in erythromycin A,12 indicating
that AmiH functions to dimethylate TDP-D-viosamine (Scheme
1). To confirm the function of AmiH, the amiH gene was
inactivated to afford the mutant AM1005 (Tables S1 and S2 and
Figure S1 in the Supporting Information). HPLC−MS analysis
revealed that the AM1005 mutant produced a new metabolite, 1c
(Figure 1). Subsequently, a 10 L-scale fermentation of AM1005
AM1004 (Figure S2) produced a new product distinct from
amicetin (Figure 1). This product, 1, was isolated from a 9 L-
scale fermentation and was determined by HR-ESI-MS to have
the chemical formula C21H27N5O6 (m/z 468.1869 [M + Na]+,
calcd 468.1859). Compound 1 differs from 1a only by the
absence of NMR signals for the amosaminyl moiety (Tables S3
and S4 and Figure S9), and the upfield shift of 3.9 ppm for C4′
confirmed the structure of 1 as deamosaminylamicetin (Scheme
1). A minor compound with the chemical formula C10H15N3O3
(HR-ESI-MS m/z 226.1193 [M + H]+, calcd 226.1186) was
isolated and elucidated as 3 (Scheme 1) on the basis of the
absence of amosamine signals that are present in 3a [e.g., δH 4.97
(1H, d, J = 3.5 Hz, C1″H)] (Table S5 and Figure S10). These
data suggested that AmiG functions as the amosaminyltransfer-
ase in 1a biosynthesis.
Next, we tried to probe the AmiG activity in vitro. The full-
length amiG gene was amplified by PCR and cloned from the
genomic DNA of S. vinaceusdrappus NRRL 2363 to yield the
expression plasmid pCSG3247 (Table S1). N-His6-tagged AmiG
was overproduced as a soluble protein in Escherichia coli
BL21(DE3)/pCSG3247 upon IPTG induction and was purified
to near homogeneity by Ni-NTA affinity chromatography
(Figure S11). Given the lack of availability of the native donor
TDP-D-amosamine, we first tested TDP-D-glucose as a surrogate
donor substrate (Scheme 2A, i). To our delight, a new product
Scheme 2. Representative AmiG-Catalyzed Reactions: (A)
Forward Reaction (i), Reverse Reaction (ii), Sugar Exchange
Reaction (iii); (B) Aglycon Exchange Reaction
Figure 1. HPLC analysis of metabolite profiles of (i) the ΔamiH mutant
AM1005, (ii) the ΔamiG mutant AM1004, and (iii) the wild-type strain
S. vinaceusdrappus NRRL 2363.
led to the isolation of three compounds: 1c, 2c, and 3c (Scheme
1). The chemical formula of the main product 1c was determined
to be C27H38N6O9 by high-resolution electrospray ionization
mass spectrometry (HR-ESI-MS) (m/z 591.2773 [M + H]+,
calcd 591.2778). On the basis of their 1D and 2D NMR spectra,
the only difference between 1c and 1a is that the singlet for the
two N-methyl groups of the dimethylamino substituent at C4″ in
1a [δH 2.64 (6H, s), δC 42.4] are absent in 1c (Tables S3 and S4
and Figures S3 and S4). The chemical formula of the minor
compound 2c was determined to be C23H31N5O7 by HR-ESI-MS
(m/z 512.2104 [M + Na]+, calcd 512.2121). Comparison of 1H
NMR data revealed that 2c differs from 1c by the absence of
NMR signals for the (+)-α-methylserine moiety in 1c [δH 3.83,
4.13 (2H, ABq, J = 11.5 Hz, C19H2); δH 1.64 (3H, s, C20H3)]
(Tables S3, S4; Figure S5). The other minor compound 3c has
the chemical formula C16H26N4O6 (HR-ESI-MS m/z 371.1936,
[M + H]+, calcd 371.1930) and was elucidated by NMR analysis
(Figure S6 and Table S5) to be didemethylcytosamine (Scheme
1) on the basis of the absence of the two N-methyl groups that
are present in cytosamine (3a) [δH 2.77 (6H, s)].7,13 Generally,
N-methylations occurred on the nucleoside diphosphate
(NDP)-sugar level prior to glycosyl transfer.1b The lack of the
two N-methyl groups in 1c, 2c, and 3c (Scheme 1) and the failure
of the ΔamiI mutant to biotransform 1c into 1a (Figure S7)7
established AmiH as the requisite N-dimethyltransferase to
convert TDP-D-viosamine to TDP-D-amosamine in 1a biosyn-
thesis.
was detected upon HPLC−MS analysis (Figure 2, traces i and ii).
The new product had a chemical formula of C27H37N5O11 as
determined by HR-ESI-MS (m/z 608.2554 [M + H]+, calcd
608.2568; Figure S12), consistent with the expected product 1d
(Scheme 2A). We then examined the AmiG activity at various
pH and temperatures and in the presence of different divalent
cations (Figure S13). AmiG functioned at pH 6−9, with the best
activity at pH 6.5 in 50 mM MOPS buffer, in contrast to
methymycin GT DesVII, which worked best under alkaline
conditions.14 AmiG exhibited activity at 20−40 °C, with the
highest turnover at 35 °C. AmiG displayed weak activity without
any divalent cations and was still active in the presence of 10 mM
EDTA. The AmiG activity was enhanced by the presence of
Mg2+, Ca2+, and Mn2+, with optimal activity at >10 mM MgCl2,
but was inhibited by other divalent cations (e.g., Co2+, Cu2+, Fe2+,
AmiG is closely related to a number of GTs belonging to the
GT1 family but is unique in having a longer C-terminus of about
100 amino acids (Figure S8). The amiG inactivation mutant
B
dx.doi.org/10.1021/ja401016e | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
reaction of 1a was driven by TDP to provide more aglycon 1 for
glucosylation as a result of an excess of TDP-D-glucose over TDP
in the assay. AmiG could also catalyze the deglycosylation of 2b
to yield 2 (Figure 2, traces xi and xii). UDP was also capable of
mediating AmiG reverse catalysis at lower turnover relative to
TDP (Figure 2, trace xiii). However, 2b remained unchanged in
AmiG assays with ADP, CDP, and GDP as cosubstrates (data not
shown). In the presence of TDP, AmiG could also catalyze the
deglycosylation of 2a, 1c, and 2c (Figure S16). However, 3a and
3c appeared to be unchanged in AmiG reverse assays (data not
shown). Taking advantage of AmiG reverse catalysis, we were
able to prepare 2 from 2a by TDP-mediated deglycosylation. The
purified product 2 displayed the chemical formula C17H20N4O4
as determined by HR-ESI-MS (m/z 345.1557 [M + H]+, calcd
345.1563, Figure S17). The structure of 2 was deduced as
deamosaminylplicacetin (Scheme 2) on the basis of its 1H NMR
data, which were different from those of 2a only by the absence of
signals for the amosaminyl moiety [δH 2.64 (6H, s); δH 4.97 (1H,
d, J = 3.5 Hz, C1″H)] (Figure S17 and Table S3), and was
supported by 2D NMR analyses (Figure S17). AmiG was also
able to convert 2 to 2c−f (Figure 2, traces xv−xviii) in the
presence of the corresponding sugar donors (Figure S14). The
products were characterized to have the expected molecular
masses by LC−MS analyses (Figure S15), demonstrating that 2
is another AmiG acceptor.
Subsequently, AmiG was shown to catalyze aglycon exchange
reactions (Scheme 2B). For example, by upon coincubation of
2b, 1, TDP, and AmiG, the TDP-demethylamosamine generated
in situ from 2b by AmiG reverse catalysis could be transferred to
an alternative aglycon 1 to afford 1b (m/z 605.2942 [M + H]+,
calcd for C28H40N6O9+ 605.2935; Figure S15) with concomitant
formation of the deglycosylated product 2 (Figure 2, traces xix
and xx). Similarly, the TDP-D-amosamine generated from 2a and
TDP-D-viosamine generated from 2c in situ could also be
transferred to 1 by AmiG to afford 1a and 1c, respectively, in one-
pot aglycon exchange assays (Figure S18). In separate sugar
exchange assays, 2a and 2c could also be converted to 2d in one-
pot sugar exchange reactions, which were enhanced by addition
of TDP (Figure S19). These data confirmed that 2 is an
alternative acceptor for AmiG. The isolation of 3a from the
ΔamiF mutant,7 where the formation of the amide bond between
cytosine and p-aminobenzoate was blocked, hinted that the
amosaminylation of 3 to produce 3a should happen in vivo.
However, in an aglycon exchange reaction carried out by
coincubating 1a, 3, and TDP with AmiG, the in situ-generated
TDP-D-amosamine was not transferred to 3 to yield 3a (Figure
S20). Incubation of 3 and TDP-D-viosamine with AmiG failed to
yield detectable product. These data indicate that 3 is not an in
vitro acceptor for AmiG. Although we realized that the labile
breakdown of the amide bond between cytosine and p-
aminobenzoate in amicetin analogues might produce 3c and 3a
during the isolation process (Figure S21), the production of 3a
by the ΔamiF mutant might suggest a substrate discrepancy of
AmiG in vivo and in vitro.
Figure 2. HPLC analysis of representative AmiG assays: (i) 1 + TDP-D-
glucose; (ii) 1 + TDP-D-glucose + AmiG; (iii) 1 + UDP-D-glucose (5
mM) + AmiG; (iv) 1 + TDP-D-viosamine + AmiG; (v) 1 + TDP-4,6-
dideoxy-4-keto-D-glucose + AmiG; (vi) 1 + TDP-2-deoxy-D-glucose +
AmiG; (vii) 1a + TDP; (viii) 1a + TDP + AmiG; (ix) 1a + TDP-D-
glucose (5 mM) + AmiG; (x) 1a + TDP-D-glucose (5 mM) + TDP +
AmiG; (xi) 2b + TDP; (xii) 2b + TDP (5 mM) + AmiG; (xiii) 2b + UDP
(5 mM) + AmiG; (xiv) 2 + TDP-D-glucose; (xv) 2 + TDP-D-glucose +
AmiG; (xvi) 2 + TDP-D-viosamine + AmiG; (xvii) 2 + TDP-4,6-
dideoxy-4-keto-D-glucose + AmiG; (xviii) 2 + TDP-2-deoxy-D-glucose +
AmiG; (xix) 1 + 2b + TDP; (xx) 1 + 2b + TDP + AmiG. Assays were
performed in 50 mM MOPS buffer (pH 6.5) at 35 °C for 6 h using 100
μM 1 analogue, 1 mM NDP or NDP-sugar, and 3.3 μM AmiG, unless
otherwise stated.
Ni2+, and Zn2+). Under the optimized conditions [50 mM MOPS
buffer (pH 6.5) with 10 mM MgCl2 at 35 °C], the sugar
nucleotide specificity for AmiG was subsequently probed with 11
other commercially available nucleotide sugars in addition to
TDP-D-glucose (Figure S14). AmiG could recognize TDP-D-
viosamine, UDP-D-glucose, TDP-4,6-dideoxy-4-keto-D-glucose,
and TDP-2-deoxy-D-glucose as donor substrates, converting 1 to
1d−1f, respectively, with the expected molecular masses (Figure
2, traces iii−vi; Figure S15), while conversion of 1 was not
observed with TDP-L-rhamnose, ADP-D-glucose, GDP-D-
glucose, GDP-mannose, UDP-D-N-acetylglucosamine, UDP-D-
galactose, or UDP-D-glucuronic acid (Figure S14). Unexpect-
edly, no conversion of 3 by AmiG was detected using TDP-D-
viosamine or TDP-D-glucose as the donor substrate (data not
shown), indicating that 3 is not an in vitro acceptor of AmiG.
Finally, 1.0 mg of 1d was isolated, and its structure was
characterized by 1H NMR analysis (Figure S12). The presence of
an α-(1→4)-glycoside bond between glucose and amicetose in
1d was self-evident from the coupling constant between the
protons on C1″ and C2″ (δH 5.05, d, J = 4.0 Hz).
Following established protocols for reverse assays of inverting-
GT-catalyzed reactions by the Thorson lab,2 we found that AmiG
could deglycosylate 1a to afford 1 in the presence of TDP (Figure
2, traces vii and viii), confirming the reaction reversibility of
AmiG (Scheme 2A, ii). In a one-pot AmiG-catalyzed sugar
exchange reaction carried out by incubating 1a with TDP-D-
glucose (Scheme 2A, iii), the amosamine in 1a could be directly
replaced by glucose to afford 1d and a minor amount of 1 (Figure
2, trace ix). This sugar exchange reaction was enhanced by the
addition of TDP (Figure 2, trace x), probably because the reverse
Finally, we performed steady-state kinetic characterizations of
AmiG-catalyzed reactions. AmiG displayed KM values of 15.8 μM
and 1.99 mM for 1 and TDP-D-viosamine, respectively, and the
corresponding kcat values of 1.38 and 1.59 min−1 were in close
agreement (Table 1 and Figure S22). AmiG showed a kcat/KM
value of 87.4 mM−1 min−1 toward 1, a catalytic efficiency
comparable to those of teicoplanin GTs.15 Although AmiG
exhibited similar KM values for 1 and 2 (Table 1 and Figure S22),
it appeared that 1 was a favored acceptor over 2 for AmiG on the
C
dx.doi.org/10.1021/ja401016e | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Author Contributions
Table 1. AmiG Kinetic Parameters for Different Substrates
⊥R.C. and H.Z. contributed equally.
a
Toward 1 and 2 as Acceptors
Notes
acceptor
KM (μM)
kcat (min−1
)
kcat/KM (mM−1 min−1
)
The authors declare no competing financial interest.
1
2
15.8 1.9
16.7 2.5
1.38 0.08
0.84 0.04
87.4
50.6
ACKNOWLEDGMENTS
■
b
Toward TDP-D-viosamine and TDP-D-glucose as Donors
Financial support was provided in part by NNSFC (31125001,
31170708, 30870060), MOST (2010CB833805), and CAS
(KZCX2-YW-JC202, KSCX2-EW-G-12). C.Z. is a scholar of the
“100 Talents Project” of CAS (08SL111002). We are grateful for
the analytical facilities at SCSIO.
donor
KM (mM)
kcat (min−1
)
kcat/KM (mM−1 min−1
)
TDP-D-viosamine
TDP-D-glucose
1.99 0.29
2.67 0.18
1.59 0.08
0.56 0.03
0.80
0.21
a
b
With saturating TDP-D-viosamine (10 mM) as the donor. With
saturating 1 (100 μM) as the acceptor.
REFERENCES
■
(1) (a) Lairson, L. L.; Henrissat, B.; Davies, G. J.; Withers, S. G. Annu.
Rev. Biochem. 2008, 77, 521. (b) Thibodeaux, C. J.; Melanco
basis of the kcat/KM values (Table 1), indicating that the
attachment of the terminal (+)-α-methylserine might occur prior
to amosamine transfer in 1a biosynthesis. We also determined
the catalytic parameters of AmiG toward the surrogate sugar
donor TDP-D-glucose using 1 as a saturating acceptor (Table 1
and Figure S22). Interestingly, AmiG displayed very close KM
values for TDP-D-viosamine (1.99 mM) and TDP-D-glucose
(2.67 mM). However, TDP-D-viosamine was apparently a better
donor substrate than TDP-D-glucose on the basis of their kcat/KM
values (Table 1). Following established methods,16 we also
determined the equilibrium constant for the AmiG-catalyzed
reaction of 1 and TDP-D-viosamine to give 1c to be Keq = 120
(Figure S23), which differs from the reported values for GtfE
(4.5)2b and OleD (156).16 Although GT reverse catalysis
displays apparent beauty for sugar nucleotide synthesis,2g,h we
realize that the making of sugar nucleotides by GT reverse
catalysis is yet not practical in some cases, especially when the GT
displays a relative large Keq. It turned out to be difficult for us to
harvest a large amount of TDP-D-amosamine via the TDP-
mediated AmiG reverse reaction using 2a as a substrate.
In summary, we have validated AmiH as the N-methyltransfer-
ase and AmiG as the amosaminyltransferase in amicetin
biosynthesis. Notably, we have shown that the retaining GT
AmiG is capable of reverse catalysis and is amenable to sugar
exchange and aglycon exchange reactions in a manner analogous
to inverting GTs,2 despite the distinct catalytic mechanisms of
retaining and inverting GTs. Unlike the clearly established SN2
replacement mechanism of inverting GTs, the catalytic
mechanism of retaining GTs remains unclear, with proposals
of a double displacement or SNi mechanism.1 In addition, AmiG
was demonstrated to have a certain substrate flexibility by
utilizing TDP-D-amosamine, TDP-D-demethylamosamine,
TDP-D-viosamine, T(U)DP-D-glucose, TDP-4,6-dideoxy-4-
keto-D-glucose, and TDP-2-deoxy-D-glucose as sugar donors.
Given the kinetic advantage of 1 over 2 in AmiG catalysis, we
conclude that amosaminylation by AmiG might tailor 1a
biosynthesis. This study extends the reaction reversibility to a
retaining GT and warrants further mechanistic investigations.
̧
n, C. E., III;
Liu, H.-w. Angew. Chem., Int. Ed. 2008, 47, 9814. (c) Williams, G. J.;
Thorson, J. S. Adv. Enzymol. Relat. Areas Mol. Biol. 2009, 76, 55.
(2) (a) Minami, A.; Kakinuma, K.; Eguchi, T. Tetrahedron Lett. 2005,
46, 6187. (b) Zhang, C.; Griffith, B. R.; Fu, Q.; Albermann, C.; Fu, X.;
Lee, I. K.; Li, L.; Thorson, J. S. Science 2006, 313, 1291. (c) Zhang, C.;
Albermann, C.; Fu, X.; Thorson, J. S. J. Am. Chem. Soc. 2006, 128, 16420.
(d) Zhang, C.; Fu, Q.; Albermann, C.; Li, L. J.; Thorson, J. S.
ChemBioChem 2007, 8, 385. (e) Zhang, C.; Bitto, E.; Goff, R. D.; Singh,
S.; Bingman, C. A.; Griffith, B. R.; Albermann, C.; Phillips, G. N.;
Thorson, J. S. Chem. Biol. 2008, 15, 842. (f) Zhang, C.; Moretti, R.; Jiang,
J. Q.; Thorson, J. S. ChemBioChem 2008, 9, 2506. (g) Gantt, R. W.;
Peltier-Pain, P.; Cournoyer, W. J.; Thorson, J. S. Nat. Chem. Biol. 2011,
7, 685. (h) Gantt, R. W.; Peltier-Pain, P.; Singh, S.; Zhou, M.; Thorson, J.
S. Proc. Natl. Acad. Sci. U.S.A. 2013, 110, 7648.
(3) (a) Zhang, H.; White-Phillip, J. A.; Melanco̧ n, C. E., III; Kwon, H.-
j.; Yu, W.-l.; Liu, H.-w. J. Am. Chem. Soc. 2007, 129, 14670. (b) Fang, J.;
Zhang, Y.; Huang, L.; Jia, X.; Zhang, Q.; Zhang, X.; Tang, G.; Liu, W. J.
Bacteriol. 2008, 190, 6014. (c) Li, S.; Xiao, J.; Zhu, Y.; Zhang, G.; Yang,
C.; Zhang, H.; Ma, L.; Zhang, C. Org. Lett. 2013, 15, 1374.
(4) Kim, C. G.; Lamichhane, J.; Song, K. I.; Nguyen, V. D.; Kim, D. H.;
Jeong, T. S.; Kang, S. H.; Kim, K. W.; Maharjan, J.; Hong, Y. S.; Kang, J.
S.; Yoo, J. C.; Lee, J. J.; Oh, T. J.; Liou, K.; Sohng, J. K. Arch. Microbiol.
2008, 189, 463.
(5) (a) Yokoyama, K.; Yamamoto, Y.; Kudo, F.; Eguchi, T.
ChemBioChem 2008, 9, 865. (b) Fan, Q.; Huang, F.; Leadlay, P. F.;
Spencer, J. B. Org. Biomol. Chem. 2008, 6, 3306.
(6) Hofmann, C.; Boll, R.; Heitmann, B.; Hauser, G.; Durr, C.; Frerich,
A.; Weitnauer, G.; Glaser, S. J.; Bechthold, A. Chem. Biol. 2005, 12, 1137.
(7) Zhang, G.; Zhang, H.; Li, S.; Xiao, J.; Zhang, G.; Zhu, Y.; Niu, S.; Ju,
J.; Zhang, C. Appl. Environ. Microbiol. 2012, 78, 2393.
(8) Carrasco, L.; Vazquez, D. Med. Res. Rev. 1984, 4, 471.
(9) Flynn, E. H.; Hinman, J. W.; Caron, E. L.; Woolf, D. O. J. Am. Chem.
Soc. 1953, 75, 5867.
(10) (a) Haskell, T. H.; Ryder, A.; Frohardt, R. P.; Fusari, S. A.;
Jakubowski, Z. L.; Bartz, Q. R. J. Am. Chem. Soc. 1958, 80, 743.
(b) Evans, J. R.; Weare, G. J. Antibiot. 1977, 30, 604.
(11) Hu, T.; Zhang, G.; Zhu, Y.; Li, S.; Zhang, H.; Zhang, G.; Yang, X.;
Ju, J.; Zhang, C. Acta Microbiol. Sin. 2012, 52, 214.
(12) Summers, R. G.; Donadio, S.; Staver, M. J.; Wendt-Pienkowski, E.;
Hutchinson, C. R.; Katz, L. Microbiology 1997, 143, 3251.
(13) Konishi, M.; Naruishi, M.; Tsuno, T.; Tsukiura, H.; Kawaguchi,
H. J. Antibiot. 1973, 26, 757.
ASSOCIATED CONTENT
* Supporting Information
(14) Borisova, S. A.; Zhao, L.; Melanco
w. J. Am. Chem. Soc. 2004, 126, 6534.
(15) Howard-Jones, A. R.; Kruger, R. G.; Lu, W.; Tao, J.; Leimkuhler,
C.; Kahne, D.; Walsh, C. T. J. Am. Chem. Soc. 2007, 129, 1008.
(16) Quiros, L. M.; Carbajo, R. J.; Brana, A. F.; Salas, J. A. J. Biol. Chem.
2000, 275, 11713.
̧ n, C. E., III; Kao, C.-L.; Liu, H.-
■
S
Experimental procedures, characterization data for new com-
pounds, and determination of AmiG kinetic parameters. This
material is available free of charge via the Internet at http://pubs.
acs.org.
AUTHOR INFORMATION
Corresponding Author
czhang2006@gmail.com
■
D
dx.doi.org/10.1021/ja401016e | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX