Essential role of the donor acyl carrier protein in stereoselective chain translocation to a fully reducing module of the nanchangmycin polyketide synthase

Article abstract of DOI:10.1021/bi201768v

Incubation of recombinant module 2 of the polyether nanchangmycin synthase (NANS), carrying an appended thioesterase domain, with the ACP-bound substrate (2RS)-2-methyl-3-ketobutyryl-NANS-ACP1 (2-ACP1) and methylmalonyl-CoA in the presence of NADPH gave diastereomerically pure (2S,4R)-2,4-dimethyl-5- ketohexanoic acid (4a). These results contrast with the previously reported weak discrimination by NANS module 2+TE between the enantiomers of the corresponding N-acetylcysteamine-conjugated substrate analogue (±)-2-methyl-3- ketobutyryl-SNAC (2-SNAC), which resulted in formation of a 5:3 mixture of 4a and its (2S,4S)-diastereomer 4b. Incubation of NANS module 2+TE with 2-ACP1 in the absence of NADPH gave unreduced 3,5,6-trimethyl-4-hydroxypyrone (3) with a k_cat of 4.4 ± 0.9 min^-1 and a k_cat/ K_m of 67 min^-1 mM^-1, corresponding to a ～2300-fold increase compared to the k_cat/K_m for the diffusive substrate 2-SNAC. Covalent tethering of the 2-methyl-3-ketobutyryl thioester substrate to the NANS ACP1 domain derived from the natural upstream PKS module of the nanchangmycin synthase significantly enhanced both the stereospecificity and the kinetic efficiency of the sequential polyketide chain translocation and condensation reactions catalyzed by the ketosynthase domain of NANS module 2.

Full text of DOI:10.1021/bi201768v

Article
pubs.acs.org/biochemistry
Essential Role of the Donor Acyl Carrier Protein in Stereoselective
Chain Translocation to a Fully Reducing Module of the
Nanchangmycin Polyketide Synthase
†
‡
‡,§
,†
Xun Guo, Tiangang Liu, Zixin Deng, and David E. Cane*
†
Department of Chemistry, Box H, Brown University, Providence, Rhode Island 02912-9108, United States
‡
Key Laboratory of Combinatorial Biosynthesis and Drug Discovery (Wuhan University), Ministry of Education, and Wuhan
University School of Pharmaceutical Sciences, Wuhan 430071, P. R. China
§
Key Laboratory of Microbial Metabolism and School of Life Sciences & Biotechnology, Shanghai Jiaotong University, Shanghai
00030, P. R. China
2
*
S Supporting Information
ABSTRACT: Incubation of recombinant module 2 of the polyether
nanchangmycin synthase (NANS), carrying an appended thioesterase
domain, with the ACP-bound substrate (2RS)-2-methyl-3-ketobutyryl-
NANS_ACP1 (2-ACP1) and methylmalonyl-CoA in the presence of
NADPH gave diastereomerically pure (2S,4R)-2,4-dimethyl-5-ketohexanoic
acid (4a). These results contrast with the previously reported weak
discrimination by NANS module 2+TE between the enantiomers of the
corresponding N-acetylcysteamine-conjugated substrate analogue (±)-2-methyl-3-ketobutyryl-SNAC (2-SNAC), which resulted
in formation of a 5:3 mixture of 4a and its (2S,4S)-diastereomer 4b. Incubation of NANS module 2+TE with 2-ACP1 in the
absence of NADPH gave unreduced 3,5,6-trimethyl-4-hydroxypyrone (3) with a k of 4.4 ± 0.9 min− and a k /K of 67 min
1
−1
cat
cat
m
−1
mM , corresponding to a ∼2300-fold increase compared to the k /K for the diffusive substrate 2-SNAC. Covalent tethering
cat
m
of the 2-methyl-3-ketobutyryl thioester substrate to the NANS ACP1 domain derived from the natural upstream PKS module of
the nanchangmycin synthase significantly enhanced both the stereospecificity and the kinetic efficiency of the sequential
polyketide chain translocation and condensation reactions catalyzed by the ketosynthase domain of NANS module 2.
odular polyketide synthases (PKSs) are multifunctional
megaproteins that are responsible for the assembly line
upstream module, by transesterification to the active site
cysteine residue of the ketosynthase, followed by KS-catalyzed
decarboxylative condensation with the nucleophilic malonyl- or
methylmalonyl chain extender that is attached to the ACP
domain of the paired subunit of the homodimeric protein
M
biosynthesis of an enormous variety of highly complex natural
1
,2
products, including many important medicinals with anti-
biotic, anticancer, antiparasitic, or immunosuppressive activity
as well as numerous agriculturally important agents. In a typical
modular PKS, each module is responsible for a single round of
polyketide chain elongation and functional group modification,
with the number of modules directly corresponding to the
number of incorporated ketide building blocks. Each
homodimeric module contains a minimum of three core
domains (Figure 1A): (1) an acyl carrier protein (ACP) with a
phosphopantetheine prosthetic group that carries each of the
covalently tethered polyketide biosynthetic intermediates from
one catalytic domain to the next within each module and then
delivers the mature polyketide chain extension product to the
ketosynthase (KS) domain of the downstream module; (2) an
acyltransferase (AT) domain that specifically catalyzes trans-
thioesterification of the appropriate malonyl or methylmalonyl-
CoA chain elongation substrate onto the terminal cysteamine
thiol of the ACP phosphopantetheinyl arm; and (3) a β-
ketoacyl-ACP synthase (ketosynthase or KS) domain that is
responsible for the central polyketide chain-building reaction,
consisting of initial translocation of the electrophilic polyketide
acyl thioester substrate donated by the ACP domain of the
(
(
Figure 1A). Varying combinations of optional ketoreductase
KR), dehydratase (DH), and enoylacyl-ACP reductase (ER)
domains are responsible for the characteristic β-ketoacyl-
modifying reactions of each chain-elongation cycle.
2
For many years, the ∼100−150-amino acid (∼12−18 kDa)
ACP domains of polyketide and fatty acid synthases had been
thought to be more or less inert carriers of the various reaction
intermediates, with the flexible 20 Å phosphopantetheinate
group serving to shuttle intermediates from the active site of
3
one catalytic domain to another. A wealth of recent kinetic and
protein structural findings have considerably altered this
picture. For example, ACP-bound electrophilic diketide
substrates have k /K values for KS-catalyzed chain elongation
cat
m
that are 3 orders of magnitude greater than those of the
corresponding N-acetylcysteamine (SNAC) analogues, largely
but not exclusively due to 1000-fold reductions in the K for
m
Received: November 30, 2011
Revised: January 6, 2012
Published: January 9, 2012
©
2012 American Chemical Society
879
dx.doi.org/10.1021/bi201768v | Biochemistry 2012, 51, 879−887

Biochemistry
Article
1
3,14
elongation.
In addition, ACP domains have also been
shown to play a critical role in the maintaining the observed
stereospecificity of KR-catalyzed reductions while significantly
augmenting both the chemical and configurational stability of
15,16
the ACP-bound α-methyl-β-ketoacyl thioester substrates.
Finally, ACP-bound substrates have also been found to be
strongly preferred over their SNAC analogues as substrates for
1
7,18
recombinant DH domains.
We recently described the first recombinant expression and
biochemical characterization of a discrete, fully saturating PKS
1
9
module. Nanchangmycin synthase (NANS) module 2,
encoded by the NanA2 gene of Streptomyces nanchangensis
and harboring the full suite of β-carbon processing KR, DH,
and ER domains, catalyzes the second round of polyketide
chain elongation and reduction in the biosynthesis of the
2
0,21
anticoccidial polyether nanchangmycin (1)
(Figure 1).
Incubation of the racemic N-acetylcysteamine substrate
analogue (±)-2-methyl-3-ketobutyryl-SNAC (2-methylaceto-
acetyl-SNAC, 2-SNAC) and methylmalonyl-CoA with re-
combinant NANS module 2+TE carrying an appended DEBS
thioesterase (TE) domain yielded 3,5,6-trimethyl-4-hydroxy-
pyrone (3), resulting from enol lactonization of the initially
19
formed 3,5-diketo-2,4-dimethylhexanoyl-ACP intermediate
Figure 2A-a). Inclusion of NADPH in the incubation mixture
(
Figure 1. Early steps of nanchangmycin biosynthesis. (A) Polyketide
chain elongation and keto reduction−dehydration−enoyl reduction
catalyzed by nanchangmycin synthase module 2 (NANS module 2).
The acyltransferase (AT2) domain loads the (2S)-methylmalonyl
chain extension unit onto the ACP of NANS module 2 while the 2-
methyl-3-ketobutyryl unit donated by NANS module 1 is translocated
by ketosynthase catalyzed self-acylation of the active site cysteine of
KS2. Following keto reduction (KR2), syn dehydration (DH2), and
enoyl reduction (ER2), the resulting (2S,4R)-2,4-dimethyl-5-ketohex-
anoyl product is again translocated to the KS3 domain of NANS
module 3 to initiate the next round of polyketide chain elongation and
functional group modification. Substructures in red are derived from
the electrophilic (2S)-2-methyl-3-ketobutyryl unit while substructures
in blue are derived from (2S)-methylmalonyl-CoA. The squiggly bond
from the ACP domain to the thioester represents the attached
phosphopantetheinyl prosthetic group. B. Polyether nanchangmycin
Figure 2. Incubation of diffusive 2-SNAC and ACP-bound 2-ACP1
substrates with NANS module 2+TE.
gave, in addition to hydroxypyrone 3, a 5:3 mixture of (2S,4R)-
2
,4-dimethylhexanoic acid (4a), the predicted triketide product
(1).
of the full keto reduction−dehydration−enoyl reduction
sequence, unexpectedly accompanied by the corresponding
(2S,4S)-diastereomer 4b. The identity and stereochemistry of
both 4a and 4b were established by a combination of LC-MS
and chiral GC-MS on the derived methyl esters 4a-Me and 4b-
Me, including direct comparison with authentic synthetic
standards of each methyl ester (Figure 2A-b). The NANS KS2
domain is thus apparently unable to discriminate cleanly
between (2S)-2-SNAC, the diffusive SNAC analogue of its
native (2S)-2-methyl-3-ketobutyryl-ACP1 substrate, and its
unnatural enantiomer (2R)-2-SNAC. Notably, NANS ER2,
which sets the exclusively observed (2S)-2-methyl configuration
of the saturated triketide product (Figure 1), stereospecifically
reduced the 2-methylenoyl double bond of both the natural and
the unnatural 4-methyl epimers of the transiently generated
(E)-2,4-dimethyl-5-ketohex-2-enoyl-ACP2 intermediate. In sep-
arate incubations with recombinant NANS DH2, we also
established that this dehydratase catalyzes the stereospecific syn
dehydration exclusively of (2R,3R)-2-methyl-3-hydroxyacyl-
ACP substrates to the corresponding (E)-2-methyl-2-enoyl-
4
,5
the reaction. The observed k /K is itself dominated by the
cat
m
second-order rate constant for KS-catalyzed acylation of its
active site cysteine by the electrophilic substrate. The crystal
structures of two KS−AT didomains from the 6-deoxy-
erythonolide B synthase (DEBS)
6
7
−9
as well as those of a
1
0−12
mammalian fatty acid synthase
have all revealed that the
active sites of the respective KS and AT domains are rigidly
separated by >80 Å. A static ACP domain cannot therefore
shuttle intermediates between the KS and AT active sites solely
by pivoting only the 20 Å pantetheine side chain, thus requiring
considerable additional segmental motion of the ACP domain
itself. Most recently, extensive mutational dissection of ACP
domains from the 6-deoxyerythronolide B synthase (DEBS)
has established that specific loop regions of the ACP four-helix
bundle are involved in highly specific protein−protein
interactions with specific docking surfaces of the paired KS
domains and that distinct regions of the ACP domain are
responsible for the specificity of intermodular chain trans-
location compared to intramodular condensative chain
8
80
dx.doi.org/10.1021/bi201768v | Biochemistry 2012, 51, 879−887

Biochemistry
Article
Figure 3. Chemoenzymatic preparation and LC-ESI(+)-MS characterization of 2-ACP1. (A) Preparation of 2-SCoA and 2-ACP1. (B) LC-ESI(+)-
2
MS-MS-MS analysis of 2-methyl-3-ketobutyryl-CoA (2-SCoA). i, MS fragmentation of 2-SCoA showing the Ppant ejection ion at m/z 359.26 and
3
the corresponding phosphorylated pantetheine ejection fragment at m/z 457.21; ii, MS fragmentation of Ppant ejection ion at m/z 359.26, showing
collisionally induced fragment ions at m/z 261.27 due to loss of 2-methyl-3-ketobutyryl moiety and m/z 257.17, due to cleavage of the pantoate
2
3+
2
23+
moiety. (C) LC-ESI(+)-MS-MS-MS analysis of 2-ACP1. i, ESI(+)-MS, showing [M] ion m/z 831.41; ii, MS of [M] ion m/z 831.40, showing
3
Ppant ejection ion at m/z 359.26; iii, MS of Ppant ejection ion, showing collisionally induced fragment ions at m/z 261.27 and 257.18.
ACP products, thereby revealing the previously cryptic
stereochemistry of both the reduction product that normally
is generated by the upstream NANS KR2 domain as well as the
geometry of the dehydration product that must subsequently
possibilities and to better understand the molecular basis for
the requisite chain elongation stereospecificity, we have now
used a chemoenzymatically prepared sample of (±)-(2RS)-2-
methyl-3-ketobutyryl-ACP1 (2-ACP1) to reveal the critical role
of the ACP partner domain in KS-mediated recognition of the
electrophilic substrate, resulting in the complete control of the
stereospecificity of the NANS KS2-catalyzed polyketide chain
translocation and chain elongation reactions.
19
serve as the substrate for the downstream ER2 domain
Figure 1).
The failure of recombinant NANS module 2+TE to
(
distinguish between the enantiomers of the diffusive (2RS)-2-
methyl-3-ketobutyryl-SNAC substrate analogue (2-SNAC) was
unexpected, since the nanchangmycin synthase normally
produces a single diastereomer of the natural polyether in
vivo. In principle, NANS module 1 might only produce a single
enantiomer of the 2-methyl-3-keto diketide substrate, and the
observed in vitro lack of discrimination between the two
enantiomers of 2-SNAC by NANS module 2 might simply have
reflected an intrinsic lack of stereoselectivity by the downstream
NANS KS2 domain. On the other hand, it is far more likely that
for NANS KS2 to exercise its intrinsic stereospecificity, it is
necessary that the electrophilic substrate for chain translocation
be tethered to an ACP domain, as it is during in vivo
biosynthesis, so as to ensure proper orientation and processing
of the substrate. In order to distinguish between these
EXPERIMENTAL PROCEDURES
■
Methods. All DNA manipulations were performed
22
following standard procedures. DNA sequencing was carried
out at the U.C. Davis Sequencing Facility, Davis, CA. All
proteins were handled at 4 °C unless otherwise stated. Protein
concentrations were determined according to the method of
Bradford, using a Hewlett-Packard 8452A diode array UV/vis
spectrophotometer with bovine serum albumin as the stand-
23
ard. Protein purity was estimated using SDS PAGE gel
electrophoresis and visualized using Coomassie Blue stain
24
according to the method of Laemmli. Phosphorimaging was
carried out using Bio-Rad Kodak series screens and a Bio-Rad
FX-Pro Molecular Imager, and data were analyzed using the
8
81
dx.doi.org/10.1021/bi201768v | Biochemistry 2012, 51, 879−887

Biochemistry
software Quantity One. H NMR (400 MHz) and 13C NMR
Article
1
(−SCH CH ), 33.29 (−NHCH ), 33.31 (−SCH CH ), 60.58
(CHCH ), 66.84 (POCH ), 68.85 (OHCH), 69.04
2
2
2
2
2
(
75 and 100 MHz) spectra were obtained with Bruker Avance
3
2
AM 300 and AM 400 spectrometers. Chemical shifts are
(POCH CH), 69.26 (H O POCH), 78.74 (POCH CH),
2 2 3 2
referenced to CDCl at room temperature. High-resolution LC-
78.86 (HOCHCO), 81.38 (NCHO), 113.51, 134.80, 144.30,
3
ESI-MS and ESI-MS-MS spectra were recorded on a Thermo
LXQ LC-ESI-MS equipped with an Agilent Eclipse XDB-C18
column (2.1 mm × 150 mm, 3.5 μm). Optical rotations were
recorded using a Jasco P1010 polarimeter. Kinetic data were
analyzed by direct fitting to the Michaelis−Menten equation
using Kaleidagraph software. Reported standard deviations in
the steady-state kinetic parameters represent the calculated
statistical errors in the nonlinear, least-squares regression
analysis. High-resolution FAB MS were determined on a
JEOL JMS 600 (EI, positive ion mode). GC-MS analysis was
performed on a GC-MS Hewlett-Packard Series 2 GC-MSD,
He at 1 mL/min, 70 eV EI in positive ion mode with a Varian
CP-Chiralsil-DEX CB capillary column, 25 m × 0.32 mm. GC-
MS and LC-MS assays of incubations of NANS module 2+TE
with (2RS)-2-methyl-3-ketobutyryl-SNAC ((±)-2-SNAC) both
with and without added NADPH were carried out as previously
147.80, 150.49 (adenine), 168.93 (HNCOCH ), 169.67
2
(HNCOCH), 195.06 (−SCO), 203.88 (COCH ).
3
Expression and Purification of apo-NANS ACP1.
Plasmid pXG_NANS_ACP1_5 for expression of NANS
ACP1 corresponding to the region from T2747 to N2902 of
20
NANS module 1 was constructed by subcloning the synthetic
gene optimized for expression in E. coli into pET-28a using the
appended NdeI and XhoI restriction sites. Plasmid pXG_NAN-
S_ACP1_5 was transformed by electroporation into E. coli
BL21(DE3), and the resulting transformant, E. coli BL21-
(DE3)/pXG_NANS_ACP1_5, was grown in LB/kanamycin
medium to an OD600 0.6. Protein expression was induced by
adding 0.4 mM IPTG and incubating for an additional 20 h at
18 °C. The cells from 2 L of culture were harvested and lysed in
30 mL of lysis buffer (50 mM sodium phosphate, 300 mM
NaCl, 1 mM benzamidine, 2 mg/L leupepsin, 2 mg/L
pepstatin, 5 mM β-mercaptoethanol, pH 8.0) by two passages
through a French pressure cell at 10 000 psi. The resulting cell
1
9
described.
Materials. Reagents purchased from Sigma-Aldrich were of
the highest quality available and were used without further
lysate containing N-terminal His -tagged apo-NANS ACP1 was
6
14
purification. [2- C]Methylmalonyl-CoA (50 mCi/mmol) was
purchased from American Radiolabeled Chemicals. Amicon
Ultra cellulose centrifugal filters were from Millipore.
Aluminum-backed thin-layer chromatography plates (250
μm) were purchased from Whatman. SDS-PAGE gradient
gels (8−16% acrylamide) were from Bio-Rad. All endonu-
cleases were obtained from New England Biolabs. The
synthetic gene encoding NANS ACP1 with codons optimized
for expression in Escherichia coli was ordered from DNA2.0 Inc.
Recombinant Sfp was expressed and purified as previously
centrifuged at 53000g to remove cell debris, and the
supernatant was mixed with 5 mL of Ni-NTA resin. After
gentle shaking at 4 °C for 1 h, the slurry was transferred to a
fritted column (i.d. 1.5 cm), washed with 5 col. vol. of wash
buffer (50 mM sodium phosphate, 300 mM NaCl, 30 mM
imidazole, 1 mM β-mercaptoethanol, pH 8.0), and eluted with
5 col. vol. of elution buffer (50 mM sodium phosphate, 300
mM NaCl, 250 mM imidazole, 1 mM β-mercaptoethanol, pH
8.0). The protein-containing eluant was concentrated with 10K
MWCO centrifugal filters and further purified by FPLC gel
filtration chromatography with a HiLoad 16/60 Superdex 200
column from Amersham Biosciences (50 mM sodium
phosphate, 350 mM NaCl, 10% glycerol, pH 6.5). The purified
protein fractions were pooled and concentrated, and aliquots
were flash-frozen in liquid nitrogen and stored at −80 °C until
2
5
described. NANS module 2+TE was expressed and purified as
19
previously reported. Synthetic (±)-2-methyl-3-ketobutyryl-N-
26
acetylcysteamine thioester (1) as well as the authentic
reference standards (2S,4R)-2,4-dimethyl-5-ketohexanoic acid
1
9
(
(
4a), (2S,4S)-2,4-dimethyl-5-ketohexanoic acid (4b),
27
2RS,4RS)-2,4-dimethyl-5-ketohexanoic acid (4), and 3,5,6-
use. The purity and MW of His -tag-apo-ACP1 were confirmed
6
28
trimethyl-4-hydroxypyrone (3) were each prepared according
to the cited literature methods and fully characterized by NMR
and GC-MS.
by SDS PAGE and LC-ESI(+)-MS (Figure S3).
Chemoenzymatic Preparation of 2-Methyl-3-keto-
butyryl-ACP1 (2-ACP1). In 5 mL sodium phosphate buffer
(50 mM sodium phosphate, 350 mM NaCl, 300 μM TCEP, 10
(
2RS)-2-Methyl-3-ketobutyryl-CoA (2-SCoA). The
(
2RS)-2-methyl-3-ketobutyryl-CoA (2-SCoA) was synthesized
as previously described. The structure was characterized by
H and C NMR, MALDI-TOF (Supporting Information
Figure S2), and LC-ESI(+)MS (Figure 3B). The LC-MS
mM MgCl , pH 6.40), apo-NANS ACP1 (46 μM) was
2
29
incubated with Sfp PPTase (5.8 μM) and 2-methyl-3-
1
13
ketobutyryl-CoA (2-SCoA) (138 μM) for 45 min in a 30 °C
3
3
25,30
water bath.
The mixture was concentrated to 2 mL in a 10
analysis was performed using an Agilent Eclipse XDB-C18
column (2.1 mm × 150 mm, 3.5 μm). Gradient program: 0−
kDa centrifugal filter and purified by FPLC gel filtration
chromatography with a HiLoad 16/60 Superdex 200 column
from Amersham Biosciences (buffer: 50 mM sodium
phosphate, 350 mM NaCl, pH 6.40). The pooled fractions
were then concentrated to 250 μL, and the concentration of 2-
ACP1 was determined to be 450 μM by Bradford assay. The
1
.5 min, isocratic 95% water + 5% methanol + 0.1% formic
acid; 1.5−14 min, methanol concentration linearly increased to
6
MeOH column rinse; flow rate 200 μL/min. H NMR (CDCl ,
5%; 14−15 min, 65−100% methanol; 15−20 min, 100%
1
3
3
4
00 MHz): δ 0.60 (s, 3H, −CCH ), 0.73 (s, 3H, −CCH ), 1.17
sample was analyzed by LC-ESI(+)MS (Figure 3C).
3
3
(
s, 3H, −CHCH ), 2.13 (s, 3H, −COCH ), 2.28 (t, 2H, J = 6.8
NANS Module 2+TE Activity Assays with 2-Methyl-3-
ketobutyryl-ACP1 (2-ACP1). The assays used with ACP1-
bound substrate 2-ACP1 were based on those previously
3
3
Hz, HNCOCH ), 2.90 (t, 2H, J = 6.0 Hz, −SCH CH ), 3.20 (t,
2
−
2
2
2
H, J = 6.0 Hz, −NHCH ), 3.30 (t, 2H, J = 6.4 Hz,
2
19
SCH CH ), 3.39 (m, 1H, OHCH), 3.68 (m, 1H,
described for incubations with 2-SNAC, which were also
repeated in order to allow direct comparisons of reaction
product distributions.
2
2
H O POCH), 3.87 (s, 1H, OCH), 4.10 (s, 2H, OCH ), 4.45
2
3
2
(
s, 1H, HOCHCO), 6.02 (d, 1H, J = 6.8 Hz, NCH), 8.10 (s,
1
3
1
1
2
H, NH CNCHN), 8.41 (s, 1H, NCHN) C NMR (CDCl3,
Radio TLC-Phosphorimaging. In a total volume of 50 μL
of assay buffer (250 mM sodium phosphate, 1 mM EDTA, 2.5
mM TCEP, 5% glycerol, pH 7.2), NANS module 2+TE (2
2
00 MHz): δ 8.04 (CCH ), 13.07 (CCH ), 15.78 (CHCH ),
3
3
3
3.44 (−COCH ), 23.51 (CCH ), 30.27 (HNCOCH ), 30.35
3
3
2
8
82
dx.doi.org/10.1021/bi201768v | Biochemistry 2012, 51, 879−887

Biochemistry
Article
μM) was preincubated with NADPH (5 mM) for 30 min. Then
off and extracted with 3 × 3 mL of EtOAc. After evaporation of
solvent by rotary evaporator, the residue was dissolved in 100
μL of MeOH and treated with 15 μL of trimethylsilyldiazo-
methane (2 M in hexane) at rt for 30 min to generate the
corresponding methyl esters, followed by AcOH quench. The
methanolic solution was then directly analyzed by chiral GC-
EI(+)-MS (Figure 4).
2
-methyl-3-ketobutyryl-ACP1 (2-ACP1, 60 μM, from 450 μM
14
stock) and [2- C]methylmalonyl-CoA (0.74 mCi/mmol, 0.2
mM, from a 2 mM stock in 10 mM HCl) were added in three
aliquots every 15 min. The reaction was allowed to stand 2 h
and then quenched by addition of 15 μL of 2 M HCl, followed
by extraction with EtOAc (3 × 250 μL). The organic solvent
was removed by Speed-Vac, and the crude organic extract was
resuspended in 3 × 20 μL of EtOAc and analyzed by TLC
Steady-State Kinetics for Incubation of 2-Methyl-3-
ketobutyryl-ACP1 (2-ACP1) with NANS Module 2+TE.
The steady-state parameters for the NANS module 2+TE-
catalyzed formation of pyrone 3 were determined in the
(
silica gel, 1:1 Et O/CH Cl + 0.1% AcOH). Radio-TLC
2 2 2
results were visualized by phosphorimaging (Figure 4A).
19
absence of NADPH, as previously described. The method-
ology was based on that previously used to characterize other
31
recombinant PKS modules. Preliminary experiments estab-
lished that the enzyme-catalyzed reaction was linear over a
period of 90 min. All single-point kinetic assays were therefore
based on an incubation time of 45 min at 30 °C. For
determination of the steady-state kinetic parameters for
methylmalonyl-CoA, 10.8 μg of recombinant NANS module
2
+TE (0.044 nmol, 4 μL of a solution of 2.7 μg/μL, final
concentration 0.88 μM) was mixed with 29.9 μL of assay buffer
250 mM sodium phosphate, 1 mM EDTA, 2.5 mM TCEP, 5%
(
glycerol, pH 7.2). Then 11.1 μL of ACP-bound substrate 2-
ACP1 (450 μM, final concentration 2, 100 μM) was added.
The concentration of 2-ACP1 was not corrected for the <20%
holo-ACP1 that is present in the sample, as determined by LC-
1
4
ESI(+)-MS (Figure 3C.i). Variable concentrations of [2- C]-
methylmalonyl-CoA (0.74 mCi/mmol, 5 μL, final concen-
trations of 0.0625, 0.125, 0.25, 0.5, and 1 mM) were then added
to bring the total volume to 50 μL, and the incubation was
continued for 45 min. The reaction was quenched with 15 μL
of 2 M HCl and extracted with 4 × 300 μL of EtOAc. The
organic layer was then concentrated by Speedvac and spotted
onto a flexible silica gel TLC plate along with reference samples
Figure 4. Incubation of NANS module 2+TE with 2-ACP1 and
14
[
2- C]methylmalonyl-CoA. (A) TLC-phosphorimaging, Lanes: 1,
without NANS module 2+TE; 2, without 2-ACP1; 3, without
NADPH; 4, with NADPH. (B) Chiral GC-EI(+)-MS, SIM mode,
m/z 88. i, Synthetic mixture of 4 diastereomers of 4-Me, ret. time 4.78,
4.99, 5.34, and 5.51 min; ii, Synthetic 4b-Me (ret. time 5.00 min) and
4
a-Me (ret. time 5.35 min); iii, Enzymatically generated 4a-Me (ret.
time 5.37 min). See Figure S3C for mass spectra of synthetic and
enzyme-generated 4a-Me.
of unlabeled 3. The TLC plate was developed with 1:1 Et O/
2
CH Cl + 0.1% acetic acid and then analyzed by phosphor-
2
2
imaging. The system was precalibrated by applying serial
dilutions of [2- C]methylmalonyl-CoA to a reference plate.
Formation of the expected products was confirmed by
comparison with synthetic 3,5,6-trimethyl-4-hydroxypyrone
14
After exposure, the plate was stained with iodine to visualize the
products and locate the pyrone product 3. Kinetic constants
were calculated using the program Kaleidagraph 4.0 to fit the
data to the Michaelis−Menten equation. Reported standard
deviations in the steady-state kinetic parameters represent the
calculated statistical errors in the nonlinear, least-squares
regression analysis. For determination of the steady-state
kinetic parameters for 2-methyl-3-ketobutyryl-ACP1 (2-
ACP1), the procedure was the same, with the methylmalon-
yl-CoA concentration fixed at 1 mM and variable 2-ACP1
(6.25, 12.5, 25, 50, and 100 μM). Since each set of kinetic
measurements was performed at subsaturating concentrations
(3) and (2S,4R)-2,4-dimethyl-5-ketohexanoic acid (4a). (The
TLC conditions cannot resolve the diastereomers 4a and 4b.)
2
LC-ESI(+)MS Analysis of Enzyme-Generated 3 and
Chiral GC-EI(+)MS Analysis of Enzyme-Generated 4a.
The above phosphorimaging analysis indicated pyrone 3 was
also formed along with 4a in the incubation with NADPH.
Thus, the enzyme-generated pyrone 3 and the saturated
triketide acid 4a could each be isolated from the same TLC
plate. A 500 μL scale incubation of NANS module 2+TE (2
μM) was carried out as described above with ACP-bound
1
4
substrate, 2-ACP1, and [2- C]methylmalonyl-CoA (0.25
mCi/mmol, 0.2 mM) in the presence of 5 mM NADPH.
The reaction was run for 2 h and then quenched by addition of
14
of the fixed 2-ACP1 or [2- C]methylmalonyl-CoA cosubstrate,
the k (apparent) values were corrected to give an extrapolated
cat
1
50 μL of 2 M HCl followed by extraction with EtOAc (7 × 1
to k (actual), using the Michaelis−Menten relationship
cat
mL). The organic solvent was evaporated, and the crude
organic extract was resuspended in 3 × 20 μL of EtOAc and
separated by TLC (1/1 Et O/CH Cl + 0.1% AcOH). Radio-
TLC results were visualized by phosphorimaging to locate the
band corresponding to the enzyme-generated pyrone 3 and the
enzyme-generated triketide acid 4a. The silica gel containing
pyrone 3 was scraped off and extracted with 3 × 3 mL EtOAc,
and the solvent was removed by Speedvac. The residue was
dissolved in 200 μL of acetonitrile, and 3 was analyzed by LC-
k (app) = k (actual){[A]/(K + [A])}, where [A] is the
cat
cat
m
concentration of the fixed cosubstrate and K is the Michaelis
m
constant determined for the fixed cosubstrate in the
complementary kinetic experiments when the other cosubstrate
concentration was held constant.
2
2
2
RESULTS AND DISCUSSION
■
Preparation of (2RS)-2-Methyl-3-ketobutyryl-NANS
ACP1 (2-ACP1). To obtain recombinant NANS ACP1, a
synthetic gene optimized for expression in E. coli and
2
ESI(+)MS (Figure S3). Enzyme-generated 4a was also scraped
8
83
dx.doi.org/10.1021/bi201768v | Biochemistry 2012, 51, 879−887

Biochemistry
Article
corresponding to the region from T2747 to N2902 of NanA1
was first subcloned into pET-28a using appended NdeI and
XhoI restriction sites. The resulting plasmid pXG_NAN-
S_ACP1_5 was transformed into E. coli BL21(DE3), and the
preference for the ACP-bound substrate 2-ACP1 over the
diffusive thioester substrate analogue 2-SNAC. Since the
measured k /Km of NANS module 2+TE is intrinsically a
cat
function of only those kinetic events up to and including the
first irreversible biochemical step, this kinetic parameter directly
resulting transformant was used for production of His -tag-apo-
6
6
ACP1 which was obtained in a yield of 14 mg/L culture and
reflects the catalytic specificity of the KS2 domain.
9
5% purity after Ni-NTA immobilized metal ion affinity
Stereospecific Formation of (2S,4R)-2,4-Dimethyl-5-
ketohexanoate (4a) Using ACP-Bound Substrate. In-
cubation of NANS module 2+TE with (2RS)-2-methyl-3-
ketobutyryl-ACP1 (2-ACP1) in the presence of 0.25 mM
[2- C]methylmalonyl-CoA and 5 mM NADPH followed by
analysis of the crude organic extract by radio-TLC phosphor-
chromatography and gel filtration, The MW was determined by
LC-ESI(+)-MS as 18 663 Da (calcd, 18 659, Figure S3). The
requisite (2RS)-2-methyl-3-ketobutyryl-CoA (2-SCoA) was
prepared from ethyl 2-methylacetoacetate as previously
described (Figure 3A), and the structure was verified by
MALDI-TOF MS (Figure S2) and LC-ESI(+)-MS-MS (Figure
14
2
9
imaging demonstrated the formation of triketide acid 4 (R 0.6)
f
1
13
3
B) as well as H and C NMR. Using Sfp phosphopante-
accompanied by pyrone 3, the latter formed by premature TE-
catalyzed release of the initially formed chain elongation
intermediate, as previously observed for the corresponding 2-
2
5,30
theinyl transferase (PPTase),
the phosphopantetheinyl unit
of (2RS)-2-methyl-3-ketobutyryl-CoA (2-SCoA) was fused to
apo-NANS ACP1 to afford (2RS)-2-methyl-3-ketobutyryl-
ACP1 (2-ACP1). LC-ESI(+)-MS-MS-MS analysis confirmed
the formation of 2-methyl-3-ketobutyryl-ACP1 (2-ACP1)
19
SNAC analogue (Figures 2 and 4A). The enzyme-generated
sample of 4 was derivatized with TMSCHN and analyzed by
2
1
9
chiral GC-EI(+)-MS. The enzymatic reaction product
corresponded exclusively to (2S,4R)-4a-Me, the methyl ester
of the native (2S,4R)-2,4-dimethyl-5-ketohexanoic acid (4a),
ret. time 5.37 min, identical by direct comparison in TIC, XIC,
and SIM modes (m/z 88) with synthetic methyl ester 4a-Me in
both ret. time and MS fragmentation pattern (Figures 2B-b and
4 and Figure S4C). Neither the diastereomeric (2S,4S)-4b-Me
nor either of the remaining two diastereomers of 4-Me could be
detected, in contrast to the 5:3 mixture of 4a-Me and 4b-Me
that was generated using the diffusive SNAC substrate analogue
2-SNAC. The KS2 domain of NANS module 2 is therefore
completely stereospecific with respect to translocation of the
correct epimer of 2-ACP1, the NANS ACP1-bound mixture of
the natural (2S)-2-methyl-3-ketobutyryl substrate, and the
epimeric (2R)-diastereomer (Figure 2B-b).
78%, [M]2 m/z 831.41; calcd [M] 19 100; [predicted
3+
+
(
2
MW 19 101]) (Figure 3C). The MS spectrum of 2-ACP1
exhibited the characteristic phosphopantetheinate (Ppant)
ejection fragment (m/z 359.26) carrying the 2-methyl-3-
ketobutyryl moiety, while the corresponding MS spectrum
displayed characteristic fragments at m/z 257.18 and 261.27,
resulting from further collision-induced cleavage of the
pantoate moiety and the thioester bond, respectively
3
3
2,33
2
(Figure 3C), in close agreement with the corresponding MS
3
and MS spectra of 2-methyl-3-ketobutyryl-CoA (2-SCoA)
(
Figure 3B).
Enzymatic Formation of 3,5,6-Trimethyl-4-hydroxy-
pyrone (3). Incubation of NANS module 2+TE with (2RS)-2-
1
4
methyl-3-ketobutyryl-ACP1 (2-ACP1) and [2- C]-
methylmalonyl-CoA in the absence of NADPH for 2 h and
analysis of the resulting concentrated crude organic extract by
radio-TLC phosphorimaging gave as the major product pyrone
The NanA1 protein that normally supplies the substrate for
Q
NANS module 2 is comprised of both a loading KS -AT-ACP
tridomain and the first polyketide chain extension module,
20
3
(R 0.5), identical to synthetic 3 (Figures 2B-a and 4A). The
f
NANS module 1. The loading tridomain provides the acetyl-
result was confirmed by LC-ESI(+)-MS-MS analysis of TLC-
ACP starter unit, derived from malonyl-CoA, which then
L
purified 3 obtained from a 500 μL scale incubation of NANS
serves as the primer for the KS1 domain of NANS module 1.
NANS KS1-catalyzed decarboxylative condensation reaction
with (2S)-methylmalonyl-ACP1 is expected to proceed with
inversion of configuration to afford (2R)-2-methyl-3-ketobu-
1
4
module 2+TE with 2-ACP1 and [2- C]methylmalonyl-CoA
Figure S4). The MS spectrum (ret. time 5.08 min, m/z 155)
(
2
and MS spectrum (m/z 81, 109 and 127) of enzyme-generated
were identical to those of synthetic 3,5,6-trimethyl-4-
7
,16
3
tyryl-SACP1;
NANS module 1 also harbors a reductively
0
hydroxypyrone (3). Steady-state kinetic analysis (45 min
incubation) carried out at a fixed concentration of 2-ACP1
inactive KR domain, analogous to that found in the ketone-
forming modules of numerous macrolide synthases including
DEBS module 3.
1
,34−36
(
100 μM) and variable concentrations of methylmalonyl-CoA
−1
gave a k (app) of 2.4 ± 0.3 min and a K (MM-CoA) of
0
substrate and the methylmalonyl-CoA concentration was fixed
at 1 mM, the k (app) was 3.5 ± 0.7 min and K (2-ACP1)
was 66 ± 26 μM (Figure S5). Correction for the subsaturating
The initially formed (2R)-2-methyl-3-ketobutyryl-SACP1
diastereomer must be converted at some stage to the epimeric
(2S)-2-methyl-3-ketobutyryl stereoisomer that serves as the
electrophilic substrate for NANS KS2. Simple buffer-catalyzed
interconversion of the (2R)- and (2S)-2-methyl-3-ketobutyryl
epimers can be ruled out since the established background
epimerization rate of <0.01 min− for ACP-bound substrate is
considerably slower than the typical rate of KS-catalyzed
cat
m
.25 ± 0.08 mM (Figure S5). When 2-ACP1 was the variable
−1
cat
m
concentrations of each fixed cosubstrate using the Michaelis−
1
Menten equation gave a calculated k (actual) of 4.4 ± 0.9
cat
−1
min . Comparison with the corresponding steady-state kinetic
−
1
parameters of the diffusive substrate (2RS)-2-methyl-3-
condensation (>1−4 min ), which is itself the rate-
1
9
7,16
ketobutyryl-SNAC (2-SNAC) obtained under identical
determining step for each module.
The KS2 domain thus
conditions indicated that while k (actual) for ACP1-bound
must cleanly discriminate between the two epimers of the
ACP1-bound ketodiketide substrate, resulting in priming of
KS2 by only (2S)-2-methyl-3-ketobutyryl-ACP1 (2-ACP1).
Epimerization of the 2-methyl group must therefore precede
covalent attachment of the electrophilic 2-methyl-3-ketoacyl
substrate to the active site Cys of the downstream KS domain.
The requisite epimerase activity may in principle reside in
cat
substrate 2-ACP1 had increased a modest 8-fold from 0.55
−
1
min (2-SNAC), the observed K for 2-ACP1 was ∼300-fold
m
lower than the K of 19 mM for 2-SNAC measured under the
m
same conditions, resulting in a net ∼2300-fold increase in k /
cat
−1
−1
4
−1
−1
K from 29 min
M (2-SNAC) to 6.7 × 10 min M (2-
m
ACP1). NANS module 2+TE thus has a strong kinetic
8
84
dx.doi.org/10.1021/bi201768v | Biochemistry 2012, 51, 879−887

Biochemistry
Article
NANS module 1, possibly associated with the reductively
not alter the observed substrate stereospecificity of the DEBS
KS domains.
37
4,5
inactive KR1° domain, or it may be associated with the KS2
domain of NANS module 2. In the latter case, KS2 would have
to interconvert the epimers of the KS2-bound 2-methyl-3-
ketobutyryl substrate prior to the condensation reaction.
Translocation of the 2-methyl-3-ketobutyryl displaces the
original thioester, be it the natural ACP1-SH or the N-
acetylcysteamine analogue. Once the transesterification of the
active site cysteine of the KS2 domain has occurred, there is
thus no longer any distinction between the origin of respective
electrophilic 2-methyl-3-ketobutyryl units. Epimerization after
acylation of the active site cysteine by the incorrect (2R)-
epimer is firmly excluded by the previously observed lack of
discrimination between the two epimers of 2-methyl-3-
ketobutyryl-SNAC that results in formation of a 5:3 mixture
of the diastereomeric triketide acids 4a and 4b from the
In addition to nanchangmycin (also known as dianemycin),
many polyether natural products, including lenoremycin,
monensin, laidlomycin, nigericin, grisorixin, mutalomycin,
septamycin, and carriomycin, have identical dimethyltetra-
hydropyranyl moieties distal to their carboxy-terminal ends,
each of which is derived from a common (2S,4R)-2,4-dimethyl-
38
5
-ketohexanoyl triketide precursor (Figure S1). Indeed, the
first nine modules of the nanchangmycin, monensin, and
nigericin PKS clusters show a striking similarity in both their
overall modular organization and the specific domain content,
0
particularly the presence of an inactive KR domain in module
39
1
, as well as a discrete, fully reductive module 2. It is therefore
evident that the observed stereospecific interaction of NANS
module 2 with the ACP1-bound 2-methyl-3-ketobutyryl
diketide is directly relevant to the biochemical action of a
wide range of polyether synthases. These results can also
account for the stereochemistry of formation of the (2S,4R)-
19
diffusive substrate (±)-2-SNAC. The KS2 domain therefore
serves as a gatekeeper that translocates only the natural ACP1-
bound (2S)-2-methyl-3-ketoacylbutyryl substrate and rejects
the corresponding (2R)-2-methyl-3-ketoacyl diastereomer that
would be initially generated by the KS1 domain of the upstream
module. The observed stereospecificity for diketide trans-
location by the NANS KS2 domain presumably reflects specific
interactions with the partner ACP1 domain carrying the
diketide substrate. These essential interactions are apparently
absent in the diffusive substrate SNAC substrate analogue 2-
SNAC.
2
,4-dimethyl-5-ketohexanoate moiety found at the carboxyl
termini of both nanchangmycin and lenoremycin. Finally, these
results may also account for the generation of the common
(
6S,8S)-6,8-dimethyl-9-keto substitution pattern that is present
in erythromycin and numerous other macrolides, as summar-
40,41
ized by the well-known Celmer’s rules
and exemplified by
the macrolide aglycones 10-deoxymethynolide, 6-deoxyerythro-
nolide B, and tylactone (Figure S1).
Interestingly, modules 3 of both DEBS and picromycin
synthase (PICS) also harbor a reductively inactive KR domain
0
ASSOCIATED CONTENT
■
that has been suggested to have an intrinsic methylepimerase
*
S
Supporting Information
3
7
activity. This epimerase activity would facilitate the
conversion of ACP3-bound (2R)-2-methyl-3-ketoacyl tetra-
ketide intermediates, initially generated by the corresponding
DEBS and PICS KS3-mediated condensation reactions, to the
epimeric ACP3-bound (2S)-2-methyl-3-ketoacyl tetraketides,
prior to translocation of the electrophilic product to the
respective downstream PICS or DEBS KS4 domains. Notably,
both DEBS and PICS module 4 also mediate the full set of
reductive β-ketoacyl-ACP KR-, DH-, and ER-catalyzed chain
processing reactions so as to generate a fully reduced polyketide
chain elongation product, analogous to the catalytic action of
NANS module 2 (Figure S1).
Significance. The ability of the NANS KS2 domain to
distinguish between the (2S)- and (2R)-epimers of its natural 2-
methyl-3-ketobutyryl-ACP1 substrate depends critically on the
tethering of the substrate to an ACP domain. This is most likely
due to specific KS−ACP interactions, since there is only a
modest 5:3 discrimination in favor of the diffusive (2S)-2-
methyl-3-ketobutyryl-SNAC analogue. It is unlikely that the
pantetheinate residue alone could account for the dramatic
increase in stereospecificity in NANS KS2-catalyzed chain
translocation as well as the observed 2300-fold increase in kcat/
Km observed for 2-ACP1. The requisite epimerization of the
endogenously generated, ACP-bound (2R)-2-methyl-3-keto-
diketide must occur prior to chain translocation and processing
by the downstream KS2 domain, which is itself completely
stereoselective for the correct epimer of the ACP-bound chain
elongation substrate. Although it has previously been reported
that the catalytic efficiency of individual DEBS modules, as
Polyether and macrolide structures, SDS-PAGE, ESI-MS, LC-
MS, and GC-MS data, and kinetic plots. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*
Tel: +1-401-863-3588; e-mail: david_cane@brown.edu.
Funding
This work was supported by National Institutes of Health
Grant GM22172 to D.E.C., by the 973 Project from the
Ministry of Science and Technology of China
(2012CB721001) and the National Science Foundation of
China (31170096 and 31040086) to T.L., and by grants 973
and 863 from the Ministry of Science and Technology of China
to Z.D.
Notes
None of the authors have any conflict of interest.
ACKNOWLEDGMENTS
■
We thank Dr. Dongqing Zhu for her help and advice with the
cloning experiments, Dr. Tun-Li Shen for assistance with mass
spectrometry, and Dr. Russell Hopson for assistance with
NMR. We also thank Prof. Chaitan Khosla of Stanford
University for critical comments on the manuscript.
REFERENCES
■
(
1) Hopwood, D. A. (2009) Methods in Enzymology, Complex
measured by the observed k /K , is enhanced 1000-fold by
cat
m
Enzymes in Microbial Natural Product Biosynthesis, Part B: Polyketides,
Aminocoumarins and Carbohydrates, Vol. 459, Academic Press, New
York.
using ACP-bound diketide substrates in place of the SNAC
thioester analogues, the use of such ACP-bound diketides did
8
85
dx.doi.org/10.1021/bi201768v | Biochemistry 2012, 51, 879−887

Biochemistry
Article
(
2) Smith, S., and Tsai, S. C. (2007) The type I fatty acid and
polyketide synthases: a tale of two megasynthases. Nat. Prod. Rep. 24,
041−1072.
3) Berg, J. M., Tymoczko, J. L., and Stryer, L. (2007) Biochemistry,
th ed., W.H. Freeman, New York.
4) Wu, N., Cane, D. E., and Khosla, C. (2002) Quantitative analysis
(22) Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular
Cloning, A Laboratory Manual, 2nd ed., Cold Spring Harbor
Laboratory Press, Cold Spring Harbor, NY.
(23) Bradford, M. (1976) A rapid and sensitive method for the
quantitation of microgram quantitites of protein utilizing the principle
of protein-dye binding. Anal. Biochem. 72, 248−254.
1
(
6
(
of the relative contributions of donor acyl carrier proteins, acceptor
ketosynthases, and linker regions to intermodular transfer of
intermediates in hybrid polyketide synthases. Biochemistry 41, 5056−
(24) Laemmli, U. K. (1970) Cleavage of structural proteins during
the assembly of the head of bacteriophage T4. Nature 227, 680−685.
(25) Quadri, L. E., Weinreb, P. H., Lei, M., Nakano, M. M., Zuber, P.,
and Walsh, C. T. (1998) Characterization of Sfp, a Bacillus subtilis
phosphopantetheinyl transferase for peptidyl carrier protein domains
in peptide synthetases. Biochemistry 37, 1585−1595.
5066.
(
5) Wu, N., Tsuji, S. Y., Cane, D. E., and Khosla, C. (2001) Assessing
the balance between protein-protein interactions and enzyme-
substrate interactions in the channeling of intermediates between
polyketide synthase modules. J. Am. Chem. Soc. 123, 6465−6474.
(26) Weissman, K. J., Bycroft, M., Cutter, A. L., Hanefeld, U., Frost,
E. J., Timoney, M. C., Harris, R., Handa, S., Roddis, M., Staunton, J.,
and Leadlay, P. F. (1998) Evaluating precursor-directed biosynthesis
towards novel erythromycins through in vitro studies on a bimodular
polyketide synthase. Chem. Biol. 5, 743−754.
(
6) Wu, J., Kinoshita, K., Khosla, C., and Cane, D. E. (2004)
Biochemical analysis of the substrate specificity of the beta-ketoacyl-
acyl carrier protein synthase domain of module 2 of the erythromycin
polyketide synthase. Biochemistry 43, 16301−16310.
(
27) Mukherjee, S., and Nasipuri, D. (1984) Heterocyclic steroids:
Part IV - Synthesis of (±)-3-methoxy-11α±-methyl-6,15-diazaestra-
,3,5,7,9,14-hexaene and some related hydrophenanthridinones. Indian
J. Chem., Sect. B 23B, 193−198.
28) Shimamura, H., Sunazuka, T., Izuhara, T., Hirose, T., Shiomi, K.,
(
7) Khosla, C., Tang, Y., Chen, A. Y., Schnarr, N. A., and Cane, D. E.
2007) Structure and mechanism of the 6-deoxyerythronolide B
synthase. Annu. Rev. Biochem. 76, 195−221.
8) Tang, Y., Chen, A. Y., Kim, C. Y., Cane, D. E., and Khosla, C.
2007) Structural and mechanistic analysis of protein interactions in
module 3 of the 6-deoxyerythronolide B synthase. Chem. Biol. 14,
31−943.
9) Tang, Y., Kim, C. Y., Mathews, I. I., Cane, D. E., and Khosla, C.
2006) The 2.7-Å crystal structure of a 194-kDa homodimeric
1
(
(
(
and Omura, S. (2007) Total synthesis and biological evaluation of
(
verticipyrone and analogues. Org. Lett. 9, 65−67.
(29) Haapalainen, A. M., Merilainen, G., Pirila, P. L., Kondo, N.,
9
(
(
Fukao, T., and Wierenga, R. K. (2007) Crystallographic and kinetic
studies of human mitochondrial acetoacetyl-CoA thiolase: The
importance of potassium and chloride ions for its structure and
function. Biochemistry 46, 4305−4321.
fragment of the 6-deoxyerythronolide B synthase. Proc. Natl. Acad. Sci.
U. S. A. 103, 11124−11129.
(30) Weinreb, P. H., Quadri, L. E., Walsh, C. T., and Zuber, P.
(
10) Maier, T., Jenni, S., and Ban, N. (2006) Architecture of
(1998) Stoichiometry and specificity of in vitro phosphopantetheiny-
mammalian fatty acid synthase at 4.5 Å resolution. Science 311, 1258−
262.
11) Leibundgut, M., Maier, T., Jenni, S., and Ban, N. (2008) The
lation and aminoacylation of the valine-activating module of surfactin
1
(
synthetase. Biochemistry 37, 1575−1584.
(31) Wu, N., Kudo, F., Cane, D. E., and Khosla, C. (2000) Analysis of
multienzyme architecture of eukaryotic fatty acid synthases. Curr.
Opin. Struct. Biol. 18, 714−725.
the molecular recognition features of individual modules derived from
the erythromycin polyketide synthase. J. Am. Chem. Soc. 122, 4847−
(
12) Maier, T., Leibundgut, M., and Ban, N. (2008) The crystal
4
852.
structure of a mammalian fatty acid synthase. Science 321, 1315−1322.
(32) Dorrestein, P. C., Bumpus, S. B., Calderone, C. T., Garneau-
(
13) Kapur, S., Chen, A. Y., Cane, D. E., and Khosla, C. (2010)
Tsodikova, S., Aron, Z. D., Straight, P. D., Kolter, R., Walsh, C. T., and
Kelleher, N. L. (2006) Facile detection of acyl and peptidyl
intermediates on thiotemplate carrier domains via phosphopante-
theinyl elimination reactions during tandem mass spectrometry.
Biochemistry 45, 12756−12766.
Molecular recognition between ketosynthase and acyl carrier protein
domains of the 6-deoxyerythronolide B synthase. Proc. Natl. Acad. Sci.
U. S. A. 107, 22066−22071.
(
14) Charkoudian, L. K., Liu, C. W., Capone, S., Kapur, S., Cane, D.
E., Togni, A., Seebach, D., and Khosla, C. (2011) Probing the
interactions of an acyl carrier protein domain from the 6-
deoxyerythronolide B synthase. Protein Sci. 20, 1244−1255.
(33) Meluzzi, D., Zheng, W. H., Hensler, M., Nizet, V., and
Dorrestein, P. C. (2008) Top-down mass spectrometry on low-
resolution instruments: characterization of phosphopantetheinylated
carrier domains in polyketide and non-ribosomal biosynthetic
pathways. Bioorg. Med. Chem. Lett. 18, 3107−3111.
(
15) Castonguay, R., Valenzano, C. R., Chen, A. Y., Keatinge-Clay,
A., Khosla, C., and Cane, D. E. (2008) Stereospecificity of
ketoreductase domains 1 and 2 of the tylactone modular polyketide
synthase. J. Am. Chem. Soc. 130, 11598−11599.
(34) Donadio, S., Staver, M. J., Mcalpine, J. B., Swanson, S. J., and
Katz, L. (1991) Modular organization of genes required for complex
(
16) Valenzano, C. R., Lawson, R. J., Chen, A. Y., Khosla, C., and
polyketide biosynthesis. Science 252, 675−679.
Cane, D. E. (2009) The biochemical basis for stereochemical control
(35) Donadio, S., and Katz, L. (1992) Organization of the enzymatic
in polyketide biosynthesis. J. Am. Chem. Soc. 131, 18501−18511.
domains in the multifunctional polyketide synthase involved in
erythromycin formation in Saccharopolyspora-erythraea. Gene 111,
(
17) Valenzano, C. R., You, Y. O., Garg, A., Keatinge-Clay, A.,
Khosla, C., and Cane, D. E. (2010) Stereospecificity of the dehydratase
domain of the erythromycin polyketide synthase. J. Am. Chem. Soc.
5
(
1−60.
36) Cortes, J., Haydock, S. F., Roberts, G. A., Bevitt, D. J., and
1
(
32, 14697−14699.
18) Keatinge-Clay, A. (2008) Crystal structure of the erythromycin
polyketide synthase dehydratase. J. Mol. Biol. 384, 941−953.
19) Guo, X., Liu, T., Valenzano, C. R., Deng, Z., and Cane, D. E.
2010) Mechanism and stereospecificity of a fully saturating polyketide
Leadlay, P. F. (1990) An unusually large multifunctional polypeptide
in the erythromycin-producing polyketide synthase of Saccharopoly-
spora erythraea. Nature 348, 176−178.
(
(
37) Zheng, J., and Keatinge-Clay, A. (2011) Structural and
(
functional analysis of C2-type ketoreductases from modular polyketide
synthases. J. Mol. Biol. 410, 105−117.
(38) Cane, D. E., Celmer, W. D., and Westley, J. W. (1983) A unified
stereochemical model of polyether structure and biogenesis. J. Am.
Chem. Soc. 105, 3594−3600.
(39) Harvey, B. M., Mironenko, T., Sun, Y., Hong, H., Deng, Z.,
Leadlay, P. F., Weissman, K. J., and Haydock, S. F. (2007) Insights into
polyether biosynthesis from analysis of the nigericin biosynthetic gene
cluster in Streptomyces sp. DSM4137. Chem. Biol. 14, 703−714.
synthase module: nanchangmycin synthase module 2 and its
dehydratase domain. J. Am. Chem. Soc. 132, 14694−14696.
(
20) Sun, Y., Zhou, X., Dong, H., Tu, G., Wang, M., Wang, B., and
Deng, Z. (2003) A complete gene cluster from Streptomyces
nanchangensis NS3226 encoding biosynthesis of the polyether
ionophore nanchangmycin. Chem. Biol. 10, 431−441.
(
21) Liu, T., Cane, D. E., and Deng, Z. (2009) The enzymology of
polyether biosynthesis. Methods Enzymol. 459, 187−214.
8
86
dx.doi.org/10.1021/bi201768v | Biochemistry 2012, 51, 879−887

Biochemistry
Article
(
40) Celmer, W. D. (1965) A configurational model for macrolide
antibiotics. J. Am. Chem. Soc. 87, 1801−1802.
41) Celmer, W. D. (1971) Stereochemical problems in macrolide
antibiotics. Pure Appl. Chem. 28, 413−453.
(
8
87
dx.doi.org/10.1021/bi201768v | Biochemistry 2012, 51, 879−887