Promiscuity of a modular polyketide synthase towards natural and non-natural extender units

Article abstract of DOI:10.1039/c3ob40633d

Combinatorial biosynthesis approaches that involve modular type I polyketide synthases (PKSs) are proven strategies for the synthesis of polyketides. In general however, such strategies are usually limited in scope and utility due to the restricted substrate specificity of polyketide biosynthetic machinery. Herein, a panel of chemo-enzymatically synthesized acyl-CoA's was used to probe the promiscuity of a polyketide synthase. Promiscuity determinants were dissected, revealing that the KS is remarkably tolerant to a diverse array of extender units, while the AT likely discriminates between extender units that are native to the producing organism. Our data provides a clear blueprint for future enzyme engineering efforts, and sets the stage for harnessing extender unit promiscuity by employing various in vivo polyketide diversification strategies. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c3ob40633d

Organic&
BiomolecularChemistry
www.rsc.org/obc
Volume 11 | Number 27 | 21 July 2013 | Pages 4421–4560
ISSN 1477-0520
PAPER
Gavin J. Williams et al.
Promiscuity of a modular polyketide synthase towards natural and non-natural
extender units

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Promiscuity of a modular polyketide synthase towards
natural and non-natural extender units†
Cite this: Org. Biomol. Chem., 2013, 11,
4449
Irina Koryakina, John B. McArthur, Matthew M. Draelos and Gavin J. Williams*
Combinatorial biosynthesis approaches that involve modular type I polyketide synthases (PKSs) are
proven strategies for the synthesis of polyketides. In general however, such strategies are usually limited
in scope and utility due to the restricted substrate specificity of polyketide biosynthetic machinery.
Herein, a panel of chemo-enzymatically synthesized acyl-CoA’s was used to probe the promiscuity of a
polyketide synthase. Promiscuity determinants were dissected, revealing that the KS is remarkably toler-
ant to a diverse array of extender units, while the AT likely discriminates between extender units that are
native to the producing organism. Our data provides a clear blueprint for future enzyme engineering
efforts, and sets the stage for harnessing extender unit promiscuity by employing various in vivo poly-
ketide diversification strategies.
Received 29th March 2013,
Accepted 8th May 2013
DOI: 10.1039/c3ob40633d
www.rsc.org/obc
of polyketide synthases (PKSs), which catalyze iterative Claisen
condensations between activated malonate-derived extender
Introduction
Polyketides are a large class of secondary metabolites that units and other acyl thioesters.18 Cumulatively, only a modest
display a broad range of potent biological activities. There is variety of extender units are available to polyketide bio-
significant interest in developing synthetic strategies that synthetic pathways,19,20 and polyketide producing organisms
could be used to improve the pharmacological properties of usually only provide biosynthetic routes to a small number of
polyketide based drugs or to synthesize polyketide analogs to extender units. For example, the most common PKS extender
probe, interrogate and manipulate biological processes. units include malonyl-Coenzyme A (CoA), methylmalonyl-CoA,
However, notable exceptions aside,1–4 total synthesis of poly- and ethylmalonyl-CoA (2a–c, respectively, Scheme 2).19 Less
ketides remains challenging. In addition, semi-synthesis prevalent are extender units which are directly functionalized
strategies that rely on diversifying the structures of polyketide on standalone acyl carrier proteins (ACPs).21 Several extender
scaffolds, for example by ‘click’ chemistry or other chemoselec- units, including chloroethyl-, propyl-, and hexylmalonyl-CoA
tive reactions,5–8 are limited in scope and utility, due to the are available to a subset of PKSs via the reductive carboxylation
necessity to selectively introduce suitable chemical ‘handles’ of α,β-unsaturated acyl-CoA precursors by crotonyl-CoA carb-
into polyketides that are otherwise not available in naturally oxylase/reductase (CCR) homologs.22–24 Thus, the range of
produced polyketide structures.9,10
chemical diversity derived from extender unit selection is
Alternatively, biosynthetic strategies for polyketide diversifi- modest and includes little opportunity for further diversifica-
cation offer the potential for regioselective modification and tion by semi-synthesis.
combinatorial exploration of polyketide chemical space.11,12
The vast majority of PKS specificity studies have been
For example, inactivated acyltransferase (AT) domains can be limited almost entirely to 2a–c,17,25–28 likely due in part to the
supplemented with trans-ATs that have orthogonal specificity lack of easily accessible biosynthetic routes to other acyl-CoA’s
to the target AT13 (Scheme 1). Site-directed mutagenesis has or N-acetylcysteamine (SNAc) thioesters. Nevertheless, in vitro
also been used to relax extender unit specificity of a PKS AT assays using modules from erythromycin29,30 and pikromy-
domain,14–16 while substitution of entire AT domains has also cin27 biosynthesis have revealed tolerance towards one or two
led to production of the corresponding polyketide analogs non-native extender units (e.g. 2c) that are not available to the
(Scheme 1).12,17 Yet, the scope and utility of these biosynthetic natural producer host strain. Accordingly, the synthetic poten-
strategies is ultimately determined by the substrate specificity tial of PKSs remains largely unexplored, and we set out to
determine whether the inherent promiscuity of AT and keto-
synthase (KS) domains of PKSs was sufficient to support the
Department of Chemistry, North Carolina State University, Raleigh, NC 27695-8204,
selection and installation of a wide variety of natural and non-
USA. E-mail: gavin_williams@ncsu.edu
natural extender units into polyketide synthase products. We
reasoned that the discovery of new extender units could (i)
†Electronic supplementary information (ESI) available: HPLC data, MS data, and
NMR characterization of 3b/3e. See DOI: 10.1039/c3ob40633d
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 4449–4458 | 4449

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 1 Strategies for regioselective modification of polyketides. (A) AT domains of wild-type PKSs select a cognate extender unit for loading onto the ACP; (B)
Selected AT-domains are inactivated by site-directed mutagenesis and supplemented with exogenous trans-ATs; (C) Site-directed mutagenesis affords mutant ATs
with relaxed substrate specificity; and (D) Entire AT domains can be substituted with those of different acyl-CoA specificity. The majority of these strategies have
been employed with a very limited set of extender units (e.g. where R2 = H, Me, or Et). See main text for references.
Scheme 2 Probing the extender unit specificity of the erythronolide B synthase terminal module, Mod6TE. Shown boxed are those acyl-CoA’s successfully utilized
by holo-Mod6TE. SNAc = N-acetylcysteamine. Domains in red are likely required for the indicated transformation. Wavy line on Mod6TE represents the phosphopan-
tetheine prosthetic arm.
4450 | Org. Biomol. Chem., 2013, 11, 4449–4458
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
expand the potential scope and utility of polyketide biosyn-
thesis strategies, (ii) provide a blueprint for future enzyme
engineering, and (iii) potentially lead to development of high-
throughput screens and selections for altering PKS substrate
specificity.
Results and discussion
Extender unit generation
To provide a panel of potential PKS extender units that offer
chemical functionality beyond that provided by natural bio-
synthetic systems, we recently created mutants of the Rhizobium
trifolii malonyl-CoA synthetase MatB which are able to utilize
commercially available and synthesized malonic acid ana-
logues substituted at the C2 position (Scheme 2) to furnish the
desired mono-thioesters in a single step to ∼90% yield.31,32
Conveniently, MatB-catalyzed synthesis of acyl-CoA’s can be
coupled to downstream biosynthetic systems without purifi-
cation. For this study, 2-phenylethyl malonic acid (1i) was
identified as an additional substrate for a previously described
MatB variant, T207G/M306I,31 which was then used for the
synthesis of 2i (see Experimental).
Fig. 1 HPLC and High Res LC-MS analysis of holo-Mod6TE-catalyzed triketide
lactone formation with natural and non-natural extender units. Compound
numbers for the acyl-CoA used are shown to the left of each chromatogram.
See Scheme 2 for reaction. See Fig. S1/S2 and Tables S1/S2 in the ESI† for full
series of negative controls and LC-MS data. *indicates contaminants present in
the 2g–i acyl-CoA preparations which were detected under the specialized con-
ditions required for HPLC analysis of the corresponding lactones.
Table 1 Extender unit specificity of Mod6TE variants
Acyl-CoA extender unit promiscuity of Mod6TE
Acyl-CoA/
lactonea
holo-
holo-AT
Sfp-acylated AT
°-Mod6TEd
Next, we set out to use this panel of acyl-CoAs (2a–l) to probe
the extender unit specificity of the terminal module and
thioesterase (TE) from the 6-deoxyerythronolide B synthase
(DEBS), Mod6TE.33 This system was chosen because it rep-
resents one of the best-studied PKSs, and a wealth of data29,30
is available to benchmark any promiscuity revealed in this
study. Moreover, this minimal system could be easily dissected
and probed in order to identify specificity determinants, and
much is known regarding the ability to use DEBS modules in a
combinatorial fashion.11,12,34–36 Thus, we first set out to probe
the ability of holo-Mod6TE to catalyze the formation of tri-
ketide lactones from condensation of the diketide-SNAc 4
(Scheme 2) and each potential extender unit, 2a–l. In order to
generate the expected lactone 3a–l using this simplified PKS
system, the AT domain presumably must select and charge the
ACP with a suitable extender unit, and the KS is required to
catalyze the Claisen condensation with electrophile 4. Sub-
sequently, purified holo-Mod6TE was incubated with 4 and
each acyl-CoA (2a–l), and the extender unit specificity was then
determined using an HPLC-based end-point assay (Fig. 1 and
Table 1) (see Experimental). As expected, no product was
detected when 2a was used as substrate, which is a known
poor extender for DEBS and Mod6TE.37 In contrast, and also
as expected,30 acyl-CoAs 2b and 2c each resulted in production
of the corresponding triketide lactone product peak (Fig. 1 and
Table 1; Fig. S1 and Table S1†). This result confirms that acyl-
CoA’s synthesized by the engineered MatB variant are cata-
lytically competent with the erythronolide PKS machinery,
which accepts only the (2S) stereoisomer of 2b.38 Remarkably,
putative product peaks were also identified when propargyl 2d,
allyl 2e, methoxy 2k, and azidoethyl 2l were included in the
Mod6TEb
°-Mod6TEc
2a/3a
2b/3b
2c/3c
2d/3d
2e/3e
2f/3f
2g/3g
2h/3h
2i/3i
N.D.e
100
44
N.D.e
N.D.e
N.D.e
N.D.e
N.D.e
N.D.e
Tracef
Trace f
Tracef
N.D.e
N.D.e
N.D.e
58
100
103
45
15
36
82
N.D.e
1
N.D.e
58
Trace f
2
Trace f
33
2j/3j
2k/3k
2l/3l
N.D.e
9
N.D.e
21
31
64
a See Scheme 2 for structures of acyl-CoAs and lactones. b Reaction
included holo-Mod6TE, 4, and each acyl-CoA (Scheme 2). Numbers
indicate relative product yield with activity toward 2b set to 100.
Standard error of assay is 23% of the conversion value (determined
with 2b, n = 3). c Reaction included holo-AT°-Mod6TE, 4, and each acyl-
CoA. d Reaction included Sfp, apo-AT°-Mod6TE, 4, and each acyl-CoA
(Scheme 3). Numbers indicate relative product yield with activity
toward 2b set to 100. Standard error of assay is 28% of the conversion
value (determined with 2g, n = 3). e N.D., non-detected. f Product could
not be confidently quantified by HPLC (minimal detection limit 1.6%)
but was detected by MS analysis.
reaction, with synthetic conversions ranging from 9–44%, com-
pared to that with 2b. In addition to these robust substrates,
butyl 2g, phenyl 2h, and phenylethyl 2i resulted in very low
(<2%) or trace conversions (Fig. 1 and Table 1; Fig. S1 and
Table S1†). Analysis of control experiments that lacked either
acyl-CoA, holo-Mod6TE, or 4 demonstrated that inclusion of
each component was required for product formation in every
case except when native extender 2b was used (see Experi-
mental, Fig. S2 and Table S2†), which provided the triketide
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 4449–4458 | 4451

View Article Online
Paper
Organic & Biomolecular Chemistry
lactone 3b in the absence of 4 due to ‘stuttering’.39,40 In every AT°-Mod6TE with each extender unit 2a–l and 4 led to trace
successful case, product identity was confirmed by high quantities of lactone when extenders 2g–i (Fig. S7†) were
resolution LC-MS analysis of the product mixtures (Fig. 1; included in the reaction mixture. In addition, crude extract
Fig. S1 and Table S1†). Conversion of 2b and 2e were selected prepared from the host E. coli strain used to express Mod6TE
as representative examples, scaled up, and the lactones 3b and failed to support formation of 3a–l from extender units 2a–l,
3e each purified by semi-preparative HPLC (see Experimental). (data not shown). Taken together, this set of experiments
Subsequent structural analysis by 1H NMR (Fig. S3 and S4†) demonstrates that in the case of 2g–i, neither a catalytically
was totally consistent with the expected structures. To the best competent AT or ACP are absolutely required for lactone for-
of our knowledge, direct utilization of acyl-CoAs 2e, 2g–i, 2k, mation. Presumably, the KS of Mod6TE is able to directly
and 2l by a PKS is unprecedented and extends the known sub- utilize 2g–i, whereby the CoA moiety mimics the ACP-displayed
strate tolerance of Mod6TE to include substrates with hetero- substrate, albeit inefficiently.
atoms (e.g. 2k, 2l), large bulky substituents (2i), and sp/sp2
hybridized carbon atoms (e.g. 2d, 2e).
Directly probing promiscuity of the KS domain
Next, it was established whether the KS domain of holo-
Mod6TE limited utilization of the acyl-CoA’s that were not
Role of AT domain in lactone formation
Intriguingly, self-acylation has been reported in some PKSs, detectable substrates for Mod6TE (2a, 2f, 2j). It was envisioned
and has been attributed to activity catalyzed by the ACP that Sfp44 could be used to transfer each extender unit onto
itself.41,42 Accordingly, in order to determine whether the AT the ACP of the apo-form of AT°-Mod6TE (Scheme 3). In this
domain of Mod6TE was absolutely required for the observed way, AT-specificity is bypassed, and the KS specificity directly
Mod6TE promiscuity, the AT active site mutation Ser672Ala probed.45–47 The acyl-CoA specificity of Sfp was first deter-
was introduced into the Mod6TE gene affording the AT-null mined using an end-point MS conversion assay using 2a–l and
mutant AT°-Mod6TE (see Experimental and Scheme 3). The apo-ACP6 from DEBS (see Experimental). In each case, an
AT-null mutant was then subjected to in vivo phospho- increase in mass of the ACP was detected that corresponds to
pantetheinylation, purified, and incubated with each extender the covalent attachment of the acylphosphopantetheine
unit (2a–l) and 4 (see Experimental). Subsequent analysis of the moiety (Table S3 and Fig. S8†). Next, Sfp and the apo-form of
reaction mixtures by HPLC and LC-MS revealed that as AT°-Mod6TE were incubated with each extender unit and 4
expected,13 introduction of the AT active site mutation failed to and the product mixtures analyzed by HPLC and LC-MS
support triketide lactone formation when extender units 2a–f (Table 1, Fig. S10 and Table S4†). Conversion of apo-AT°-
and 2j–l were used (Table 1 and Fig. S6†). However, low abun- Mod6TE to the corresponding lactone (Scheme 3) was easily
dance molecular ions corresponding to the expected lactone identified with the established Mod6TE substrates 2b–e and
were observed when 2g–i were used in this assay (Table 1 and 2k–l (as expected from the wild-type holo-Mod6TE data in
Fig. S6†). While the presence of a contaminant trans-AT could Fig. 1), while 2h produced the phenyl-substituted lactone in
in principle explain these results, the specificity of known trace quantities. Notably, three extender units, 2a (in agree-
Escherichia coli (E. coli) trans-ATs are not consistent with the ment with an earlier study)45 and 2g/2i, (used very poorly by
activities observed here.43 To further probe the source of 2g–i the wild-type holo-Mod6TE), were revealed as robust substrates
formation via the AT-null mutant, AT°-Mod6TE was expressed (33–64% conversion, relative to that of 3b) for the AT-null
and purified from an E. coli host that lacked Sfp and there- mutant/Sfp system, illustrating that the KS domain is able to
fore is not expected to be phosphopantetheinylated (see utilize fully 9 out of 12 extender units. Although we cannot
Experimental). Interestingly, incubation of the apo form of completely rule out failure of Sfp to load apo-AT°-Mod6TE in
Scheme 3 Bypassing AT specificity of Mod6TE using the broad specificity phosphopantetheinyl transferase Sfp to load extender units onto the inactivated AT-null
(AT°) mutant of Mod6TE. Following successful condensation, the subsequent holo-AT°-Mod6TE cannot be recharged by Sfp, resulting in single turnover to the
lactone. SNAc = N-acetylcysteamine. Domains in red are those likely required for the indicated transformation. Wavy line on Mod6TE represents the phosphopan-
tetheine prosthetic arm.
4452 | Org. Biomol. Chem., 2013, 11, 4449–4458
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
units. Interestingly, although some evidence is emerging that
PKSs can tolerate at least one or two non-native extender
units,27 the full synthetic scope of PKSs remains poorly
described, and AT domains are usually described as being
highly selective, at least in terms of discrimination between
endogenous extender units. Promiscuous polyketide biosyn-
thetic machinery that can tolerate diverse extender units could
enable expansion of diversification strategies to include a
broad range of chemical functionality. Here, the promiscuity
of the terminal module and thioesterase domain from DEBS
was probed using a panel of 12 diverse extender unit acyl-
CoA’s. Results described in this study demonstrate PKS
module turnover of acyl-CoA extender units that (i) are non-
native for DEBS (2c–e, 2g–i, 2k, 2l), (ii) are not found in natural
biosynthetic systems (2d, 2i, 2k, 2l) and (iii) contain functional
handles (2d, 2e, 2l) that could enable downstream diversifica-
tion of polyketide structure via chemoselective ligation. Cumu-
latively, these results considerably expand the known extender
unit promiscuity of PKSs.
Fig. 2 Mod6TE-catalyzed triketide lactone formation in the presence of 2b and
competing extender unit. The yield of each of the two possible lactones in each
reaction is displayed as a percentage of the total product. Each yield was deter-
mined in triplicate (standard deviation <10% of the mean). See Experimental for
assay details and detection methods.
In an effort to determine whether the substrate tolerance of
the KS domain of Mod6TE was limiting utilization of the non-
substrate extender unit acyl-CoA’s, the AT catalytic residue Ser-
672 was mutated to alanine by site-directed mutagenesis, and
the AT-null Mod6TE loaded directly using the broad specificity
phosphopantetheinyl transferase, Sfp. If the KS domain is
broadly tolerant to these extenders, then lactones should be
detected in this assay system given the AT domain is inacti-
vated and this is the usual site of extender unit hydrolysis. Our
data indicated that the KS domain could not process 2f/2j, but
triketide lactone was detected when 2a was used. Thus, the KS
domain will likely require engineering in the case of extender
units that resemble 2f/2j. In addition, this result reveals at
least some role of the AT-domain in limiting the promiscuity
of Mod6TE, likely through hydrolysis of non-native extender
units.27 Presumably, the Ser672Ala mutation reduces or elim-
inates any editing function normally displayed the AT domain
of Mod6TE, and allows loading of 2a (and perhaps 2g/2i) via
Sfp and subsequent condensation by the KS. Increasing the
effective active site concentration of 2a/2g/2i via the Sfp-based
strategy (Scheme 3) could also play a role in ‘rescuing’ activity
of the AT-null mutant in comparison to the wild-type holo-
Mod6TE. Although we have yet to directly quantify AT-catalyzed
loading and/or subsequent hydrolysis, clearly the intact holo-
Mod6TE system is sufficiently robust to generate the expected
lactone products using most of the extender unit panel. Our
results concur with a previous proposal that combinatorial bio-
synthesis efforts could begin to focus on non-native extenders
that are not hydrolyzed by the AT,27 and now further suggest
that this strategy could be expanded to include non-natural
extender units. It is intriguing to speculate that given Mod6TE
utilizes acyl-CoA extender units with the largest C2 substituent
(Me, 2b) available to the erythromycin producing host, this
enzyme does not require discrimination against extender units
the case of 2f and 2j, it seems likely that the KS of Mod6TE
cannot tolerate β-branched extender units or those possessing
a free hydroxyl.
Competition experiments
In an effort to describe the relative level of discrimination
between the natural extender unit for Mod6TE (2b) and the
successfully utilized non-native and non-natural extenders,
Mod6TE-catalyzed triketide lactone formation was assayed in
the presence of acyl-CoA mixtures that included the native
extender 2b and either 2c, 2e, or 2l as competing acyl-CoA’s.
Several concentrations of 2b, 2c, 2e and 2l were used in this
analysis and are likely below the reported KM for the 2b. In
this way, the resulting product distribution is likely to at least
approximately reflect relative catalytic efficiencies (kcat/KM) for
each substrate. Gratifyingly, when 2b was present at 0.3 mM
and each non-native extender was included at just 5-fold
higher concentration, the resulting product mixture contained
the lactone derived from the non-native extender unit as the
major product, as judged by HPLC analysis (Fig. 2). This frac-
tion was further increased by lowering the concentration of
both 2b and each non-native extender to 0.12 mM and
0.6 mM, respectively, or by using a 20-fold excess of 2c, 2e, or
2l, compared to 2b (Fig. 2).
Clearly, modulating the extender unit concentrations
in vitro has some capacity to alter the product distribution to
favor production of triketide lactones derived from non-native
extender units when the natural extender 2b is also present.
Discussion
Several strategies are available for the regioselective modifi- with larger side-chains (e.g. 2c–i, and 2k–l), and thereby lacks
cation of polyketides, yet the vast majority of these methods the hydrolytic editing mechanism that would normally remove
are restricted to only a very small number of unique extender the smaller side-chain of 2a. Consequently, other PKSs that
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 4449–4458 | 4453

View Article Online
Paper
Organic & Biomolecular Chemistry
display promiscuity might be revealed by probing the extender tolerant of a broad range of extender unit acyl-CoA’s. By dis-
unit specificity of PKS modules that transfer malonyl-CoA secting and probing Mod6TE in vitro by site-directed muta-
derivatives with C2 substituents that are the largest provided genesis and complementation via Sfp, we determined that the KS
by the host organism. The conservation of AT and KS active is remarkably promiscuous towards diverse extender units,
site amino acid sequences among PKSs48–50 indicates that while the AT domain may only play a role in substrate discrimi-
such extender unit promiscuity might not be a feature unique nation when native extender units are employed. The vast
to the erythronolide PKS.
majority of polyketide biosynthetic diversification strategies
Synthetic conversions of the Mod6TE-catalyzed reactions have focused on only a very small number of extender units
using non-native and non-natural extenders are comparable to that include limited chemical diversity. The remarkable prom-
previously reported synthesis of triketide lactones using DEBS iscuity described here sets the stage for significantly expanding
modules and native extenders, and could be scaled even the potential scope and utility of such strategies, particularly
further to yield ∼100 mg lactone.51 Nevertheless, in vitro bio- given the ease with which non-native and non-natural acyl-
synthesis of complete polyketide scaffolds using type
I
CoA’s can be generated using engineered MatB variants.
PKSs52,53 has yet to match the scale and efficiency of those that Future efforts will now focus on harnessing extender unit
involve type II and type III PKSs.54,55 Clearly, extender unit promiscuity using in vitro and in vivo methods. In particular,
promiscuity is likely to be better harnessed via in vivo poly- the KS promiscuity discovered here could be harnessed by
ketide diversification strategies. This study therefore provides a various precursor directed approaches, and by coupling with
platform for expanding the scope and utility of such strategies. trans-ATs that display inherent or engineered acyl-CoA promis-
In particular, KS promiscuity could be coupled with inherent31 cuity. Further, a complete description of PKS extender unit
or engineered acyl-CoA promiscuity of trans-ATs to affect promiscuity now provides a guide for future engineering
regioselective polyketide modification. Moreover, emerging efforts which could include rational redesign of selected AT
methods15 to shift extender unit specificity of a given AT domain specificity and directed evolution that could for
domain towards non-native or non-natural extenders could example utilize ‘click’ handles for high-throughput screens
afford PKS modules tailored towards specific extender units, and selections.
including those described here. In lieu of detailed kinetic ana-
lysis of extender unit specificity, we demonstrated that when
Mod6TE was supplied with a mixture of 2b and non-native
Experimental
extender unit, the major triketide lactone produced was that
General
derived from the non-native substrate when the concentration
Unless otherwise stated, all materials and reagents were of the
highest grade possible and purchased from Sigma (St. Louis,
MO). Isopropyl β-D-thiogalactoside (IPTG) was from Calbio-
chem (Gibbstown, NJ). Bacterial strain Escherichia coli BL21
(DE3) pLysS competent cells was from Promega (Madison, WI).
Bacterial strain E. coli K207-359 was a gift from Prof. Keatinge-
Clay. Primers were ordered from Integrated DNA Technologies
(Coralville, IA). Plasmid pET28a-MatB was as previously
described. Analytical HPLC was performed on a Varian ProStar
system. Nuclear magnetic resonance spectra were acquired on
a Varian Mercury-VX NMR instrument operating at 300 MHz.
Chemical shifts (δ) in 1H NMR spectra are expressed in ppm
downfield of tetramethylsilane and were referenced to the
residual solvent peak.
of the non-native substrate was present at just five-fold excess
over the native substrate, 2b (Fig. 2). This result indicates that
specificity between 2b and other successful extender units
might not be too high for alteration by enzyme engineering.
Additionally, the modulation of extender unit concentration
should prove very helpful for strategies that harness stringent
AT/KS domains that have been substituted with promiscuous
AT/KS domains, such as those described here. The scope and
utility of these in vivo strategies could further be expanded by
generation of polyketide analogs modified with non-natural
handles for chemoselective ligation chemistry (e.g. from 2d,
2e, 2l). Incorporation of such handles could enable rapid
downstream diversification of polyketides via semi-synthesis.
Interestingly, promiscuous activities such as those described
here often provide suitable starting points for successful
directed evolution campaigns or rational redesign. Successful
Synthesis of diketide-SNAc 4
utilization of extender units that include azido and alkynyl Synthesis of
4
was largely as previously described
handles by Mod6TE or other PKSs suggest various (Scheme S1†).60 Meldrum’s acid (2.88 g, 20 mmol), recrystal-
strategies56–58 for developing high-throughput screens and lized from toluene, and dry pyridine (3.25 mL, 40 mmol) were
selections that could be used to identify PKS variants with added to 40 mL distilled DCM and the solution was stirred at
altered substrate specificities.
4 °C. A solution of propionyl chloride (1.75 mL, 20 mmol) in
10 mL distilled DCM was added to the reaction over
10 minutes. The reaction was stirred at 0 °C for 1 hour, then
allowed to warm to room temperature and stirred for an
additional 5 hours. The reaction was washed with 0.1 M HCl
Conclusions
The results presented here demonstrate that the AT and KS (3 × 50 mL), and the aqueous fractions were extracted with
domains of Mod6TE, and likely those of other PKSs, are highly DCM. The combined organic extracts were washed with brine,
4454 | Org. Biomol. Chem., 2013, 11, 4449–4458
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
dried over MgSO4, and filtered. The solvent was removed MatB-catalyzed synthesis of extender unit acyl-CoA’s 2a–l
under vacuum to yield a red solid, which was recrystallized
Extender units 2a–h and 2j–l were chemo-enzymatically syn-
from petroleum ether to yield the acylated meldrum’s acid (4b)
thesized as previously described.31,32 Briefly, reactions were
1
as a yellow crystalline solid (1.65 g, 41%). H-NMR (300 MHz,
performed in 50 μL reaction mixture containing 100 mM
CDCl3): 3.1 (q, J = 7.5 Hz, 2H), 1.7 (s, 6H), 1.25 (t, J = 7.5 Hz,
sodium phosphate (pH 7), MgCl2 (2 mM), ATP (4 mM), co-
enzyme A (8 mM), malonate or analog 1a–h, 1j–l (16 mM) and
3H).
4b (1.0 g, 5 mmol) was dissolved in 5 mL dry toluene.
wild-type or mutant MatB (10 μg) at 25 °C. The optimal MatB
N-Acetylcysteamine (0.60 g, 5 mmol) was added, and the reac-
mutant for each conversion was chosen on the basis of steady
state kinetic data: WT MatB (1a, 1b, 1j), T207S/M306I (1c, 1e),
T207A/M306I (1k), T207G/M306I (1d, 1f, 1h), T207A (1g), and
T207G/M306V (1l). To ensure >90% conversion to the corres-
ponding acyl-CoA, aliquots were removed after overnight incu-
bation, and quenched with an equal volume of ice-cold
methanol, centrifuged at 10 000g for 10 min, and cleared
supernatants used for HPLC analysis on a Varian ProStar
tion was stirred at 80 °C under N2 for 5 hours. The solvent was
removed under vacuum to yield S-(2-acetamidoethyl) 3-oxo-
pentanethioate (4a) as a yellow crystalline solid (1.02 g, 94%).
1H-NMR (300 MHz, CDCl3): 6.0 (s, broad, 1H), 3.7 (s, 2H), 3.45
(q, J = 6.0 Hz, 2H), 3.1 (t, J = 6.0 Hz, 2H), 2.56 (q, J = 7.2 Hz,
2H), 1.96 (s, 3H), 1.1 (t, J = 7.2 Hz, 3H).
4a (0.500 g, 2.3 mmol) and potassium tert-butoxide (0.30 g,
2.67 mmol) were added to 10 mL distilled THF at 0 °C. Iodo-
HPLC system. A series of linear gradients was developed from
methane (0.85 mL, 13.65 mmol) was added, and the solution
0.1% TFA in water (A) to methanol (HPLC grade, B) using the
following protocol: 0–32 min, 80% B; 32–35 min, 100% A. The
was stirred overnight at 0 °C to room temperature. The reac-
tion was quenched with 0.1 M HCl (50 mL) and extracted with
flow rate was 1 mL min−1, and the absorbance was monitored
ethyl acetate (3 × 50 mL). The combined organic extracts were
at 254 nm using Pursuit XRs C18 column (250 × 4.6 mm,
washed with brine, dried over MgSO4, and filtered. The solvent
Varian Inc.). Product identity was confirmed by LC-MS as
was removed under vacuum to yield a light yellow oil, which
described below and as reported earlier. Synthesis of the
was purified by flash column chromatography (EtOAc) to give
2-phenylethyl analog 2i proceeded with the commercially avail-
able 1i. MatB mutant T207G/M306I was used to convert 1i to
the diketide-SNAc (4) as a light yellow oil (0.36 g, 68%).
1H-NMR (300 MHz, CDCl3): 6.0 (s, broad, 1H), 3.7 (q, 1H), 3.4
2i and product identity confirmed by LC-MS (Calculated mass
(m, 2H), 3.0 (m, 2H), 2.5 (m, 2H), 1.9 (s, 3H), 1.4 (d, J = 7.2 Hz,
3H), 1.1 (t, J = 7.2 Hz, 3H).
958.1855, observed 958.186 ([M + H]1+)).
Mass spectrometry analysis of MatB-synthesized acyl-CoAs
Samples were subjected to negative-ESI LC/MS on a Thermo
TSQ Quantum Discovery MAX connected to a UV/Vis diode
array detector with a Waters BEH C18, 2.1 × 50 mm, 1.7 μm
particle column. A series of linear gradients was developed
from water/1 mM ammonium formate (pH 5.3) (A) to metha-
nol (B) using the following protocol: 0–10 min, 3–80% B;
10–11 min, 80–95% B; 11–13 min, 95% B; 13–14 min, 95%–5%
B; 14–17.5 min, 5% B.
Expression and purification of wild-type and mutant MatB
E. coli BL21(DE3) pLysS competent cells were transformed with
the suitable plasmid and positive transformants were selected
on LB agar supplemented with 30 μg mL−1 kanamycin. A
single colony was transferred to LB (3 mL) supplemented with
kanamycin (30 μg mL−1) and grown at 37 °C and 250 rpm over-
night. The culture was used to inoculate LB media (1 L) sup-
plemented with kanamycin (30 μg mL−1). One liter culture was
incubated at 37 °C and 250 rpm to an OD600 of 0.6, at which
time protein synthesis was induced by the addition of IPTG to
a final concentration of 1 mM. After incubation at 18 °C and
200 rpm for 18 h, cells were collected by centrifugation at
5000g for 20 min, and resuspended in 100 mM Tris-HCl pH
8.0 (20 mL) containing NaCl (300 mM) and then lysed by soni-
cation. Following centrifugation at 10 000g, the soluble extract
was loaded onto a 1 mL HisTrap HP column (GE Healthcare,
Piscataway, NJ) and purified by fast protein liquid chromato-
graphy using the following buffers: wash buffer [20 mM phos-
phate (pH 7.4) containing 0.5 M NaCl and 20 mM imidazole]
and elution buffer [20 mM phosphate (pH 7.4) containing 0.5
DEBS holo-Mod6TE reactions
DEBS holo-Mod6TE reactions were set up by adding 90 μg
DEBS holo-Mod6TE to 35 μL of 50 mM Tris-HCI (pH 7) con-
taining 2 mM MgCl2, 5 mM 4, and 4 mM each acyl-CoA (pro-
vided by suitable mutant MatB enzymes, see above). Reactions
were incubated overnight at room temperature and analyzed
by RP-HPLC and mass spectrometry as described below. A
series of negative controls that lacked either Mod6TE, each
acyl-CoA or diketide-SNAc 4 were also set up and analyzed in
the same way (Fig. S2 and Table S2†).
RP-HPLC analysis of DEBS holo-Mod6TE reactions
M NaCl and 200 mM imidazole]. The purified protein was con- Each reaction sample was quenched with an equal volume of
centrated using an Amicon Ultra 10000 MWCO centrifugal methanol, centrifuged at 10 000g for 10 min, and 25 μL used
filter (Millipore Corp., Billerica, MA) and stored as 10% gly- for HPLC analysis. HPLC analysis was performed on a Varian
cerol stocks at −80 °C. Protein purity was verified by ProStar HPLC system. A series of linear gradients was develo-
SDS-PAGE. Protein quantification was carried out using the ped from 0.1% TFA in water (A) to 0.1% TFA in acetonitrile
Bradford Protein Assay Kit from Bio-Rad.
(HPLC grade, B) using the following protocol: 0–40 min,
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 4449–4458 | 4455

View Article Online
Paper
Organic & Biomolecular Chemistry
10–30% B; 40–42 min, 100% B; 42–47 min, 10% B. The flow Preparation of the Mod6TE S672A mutant
rate was 1 mL min−1, and the absorbance was monitored at
Mod6TE S672A plasmid was constructed by using the
290 nm using Pursuit XRs C18 column (250 × 4.6 mm, Varian
Inc.). For detection of 3g–3i, the following HPLC protocol was
used: 0–40 min, 0–100% B; 40–42 min, 100% B; 42–45 min,
100% A.
Stratagene QuikChange II Site-Directed Mutagenesis Kit, as
described by the manufacturer using the pET28b/Mod6TE
as template and the oligonucleotides Mod6TE-S672A-FOR
(5′-TCAGCCGTTATCGGTCATG
Mod6TE-S672A-REV (5′-GCAATTTCGCCCTGA̲G̲C̲
AACGGCTGA-3′) (altered codons underlined). Construct was
̲C̲T̲
Mass spectrometry analysis of holo-Mod6TE reactions
confirmed to carry the correct mutation by DNA sequencing.
For High Res LC-MS analysis of holo-Mod6TE reaction pro-
ducts, reaction mixtures were analyzed by positive-ESI LC/MS
on a Thermo TSQ Quantum Discovery MAX connected to a UV/
Vis diode array detector with a 2.1 mm × 50 mm Agilent XDB
C-18 1.8 μm column (Agilent, Santa Clara CA), using a gradient
of 25–95% MeOH in 0.1% formic acid/H2O for 8 min at 1 mL
min−1, with detection at 235 nm (thioester bond) and 290 nm
(triketide pyrone). For Low Res LC/MS analysis of the Mod6TE
reaction products, reaction mixtures were analyzed by positive-
ESI LC/MS on a Shimadzu Prominence LC-20 connected to a
UV/Vis diode array detector with a 2.1 mm × 50 mm Kinetex
C-18 2.6 μm column (Phenomenex, Torrance CA), using the fol-
lowing gradient: 0–4.4 min, 5–99% B; 4–4.9 min, 99% B;
4.9–6 min, 5% B (A – 0.1% formic acid/H2O, B – 0.1% formic
acid/acetonitrile).
Expression and purification of DEBS holo-AT°-Mod6TE
DEBS holo-AT°-Mod6TE was over-expressed in E. coli K207-3 as
an N-terminally His6-tagged fusion protein as previously
described.29
DEBS holo-AT°-Mod6TE reactions
DEBS holo-AT°-Mod6TE reactions were set up by adding 90 μg
DEBS holo-Mod6TE to 35 μL of 50 mM Tris-HCI (pH 7) con-
taining 2 mM MgCI2, 5 mM 4, and 4 mM each acyl-CoA (pro-
vided by suitable mutant MatB enzymes, see above). Reactions
were incubated overnight at room temperature and analyzed
by RP-HPLC and mass spectrometry as described for the holo-
Mod6TE reactions (Fig. S6†).
Expression and purification of DEBS apo-AT°-Mod6TE
Scale up and purification of 3b and 3e
DEBS apo-AT°-Mod6TE was over-expressed from vector
pET28b/Mod6TE in E. coli BL21(DE3) as an N-terminally His6-
tagged fusion protein as previously described.29
DEBS holo-Mod6TE reactions were set up by adding 60 mg
DEBS holo-Mod6TE to 14 mL of 50 mM Tris-HCI (pH 7) con-
taining 2 mM MgCl2, 10 mM 4, and 4 mM each acyl-CoA (pro-
vided by suitable mutant MatB enzymes, see above). Reactions
were incubated for 2 days at room temperature and analyzed
by RP-HPLC. The reactions were quenched with an equal
volume of ice-cold methanol, centrifuged, and decanted from
the precipitated protein. The reactions were concentrated by
lyophilization and then extracted 3 times with dichloro-
methane. The solvent was removed under vacuum, and the
residue was dissolved in 2 mL 50% methanol and HPLC puri-
fied using solvent A (0.1% TFA in H2O) and solvent B (0.1%
TFA in acetonitrile) and the following protocol: 0–40 min,
10–30% B; 40–42 min, 100% B; 42–47 min, 10% B. The col-
lected fractions were pooled, concentrated by lyophilization,
and extracted 3 times with dichloromethane. The solvent was
removed under vacuum. The remaining residue was dissolved
in CDCl3, and the 1H-NMR spectrum was taken (Fig. S3 and
S4†). 1H-NMR 3b (300 MHz, CDCl3): 2.5 (q, J = 7.2 Hz, 2H),
1.97 (s, 3H), 1.95 (s, 3H), 1.2 (t, J = 7.2 Hz, 3H). 1H-NMR 3e
(300 MHz, CDCl3): 5.6 (m, 1H), 5.2 (m, 2H), 3.1 (m, 2H), 2.5 (q,
J = 7.2 Hz, 2H), 1.8 (2, 3H). With purified 3b in hand, a HPLC
calibration curve was constructed (Fig. S5†) and used to deter-
mine the % conversion efficiency of the scaled-up Mod6TE
reactions, using the same RP-HPLC conditions as described
above for analysis of the Mod6TE-catalyzed reactions. Sub-
sequently, the % conversion (from 4) of the large scale holo-
Mod6TE catalyzed synthesis of 3b and 3e was 4.1 and 2.3,
DEBS apo-AT°-Mod6TE reactions
DEBS apo-AT°-Mod6TE reactions were set up by adding 1.3 mg
DEBS apo-AT°-Mod6TE to 40 μL of 50 mM Tris-HCI (pH 8.8)
containing 10 mM MgCI2, 5 mM 4, and 1.6 mM each acyl-CoA
(provided by suitable mutant MatB enzymes, see above). Reac-
tions were incubated for 3 h at room temperature and analyzed
by RP-HPLC and mass spectrometry as described for the holo-
Mod6TE reactions (Fig. S7†).
Cloning, expression and purification of DEBS apo-ACP6
DEBS apo-ACP6 from the Saccharopolyspora erythraea erythro-
mycin biosynthetic gene cluster was cloned in pET28a as
described61 and over-expressed as the apo-ACP in E. coli BL21
(DE3) as an N-terminally His6-tagged fusion protein as pre-
viously described.43 The apo-ACP6 was purified as described
for MatB.
Expression and purification of Sfp
The phosphopantetheinyl transferase Sfp for in vitro acyl-CoA
specificity studies was over-expressed in E. coli BL21(DE3) as
an N-terminally His6-tagged fusion protein as previously
described.62 Sfp was purified as described for MatB.
Sfp reactions with apo-ACP6 from DEBS
respectively, and is similar to previously reported optimized Reactions containing 100 mM sodium phosphate (pH 7),
yields for Mod6TE-catalyzed syntheses.51
MgCl2 (2 mM), ATP (4 mM), coenzyme A (4 mM), 1a–l (16 mM)
4456 | Org. Biomol. Chem., 2013, 11, 4449–4458
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
and suitable MatB mutant (10 μg) at a final volume of 50 μL
were incubated at 25 °C for 24 h, or until conversion to the
acyl-CoA was complete (as judged by HPLC, see above). The
optimal MatB mutant used was as described for the MatB-cata-
lyzed acyl-CoA syntheses (see above). The reaction product
mixture was added directly to 100 μL of 50 mM Tris-HCI (pH
8.8) containing 5 mM DTT, 10 mM MgCl2, 100 μg apo-ACP6,
and 50 μg Sfp, and incubated at 25 °C for 5 h. For LC-MS ana-
lysis of acylated ACPs, reaction mixtures were analyzed by posi-
tive-ESI LC/MS on a Thermo TSQ Quantum Discovery MAX
connected to a UV/Vis diode array detector with a 2.1 mm ×
75 mm Poroshell 300SB-C18 5 μM column (Agilent, Santa
Clara CA), using a gradient of 25–100% MeOH in 0.1% formic
acid/H2O for 5 min at 1 mL min−1 (Fig. S8/S9 and Table S3†).
2 E. M. Stang and M. C. White, Nat. Chem., 2009, 1, 547–551.
3 B. A. DeChristopher, B. A. Loy, M. D. Marsden, A. J. Schrier,
J. A. Zack and P. A. Wender, Nat. Chem., 2012, 4, 705–710.
4 P. A. Wender and A. J. Schrier, J. Am. Chem. Soc., 2011, 133,
9228–9231.
5 J. M. Langenhan, N. R. Peters, I. A. Guzei, F. M. Hoffmann
and J. S. Thorson, Proc. Natl. Acad. Sci. U. S. A., 2005, 102,
12305–12310.
6 X. Fu, C. Albermann, J. Jiang, J. Liao, C. Zhang and
J. S. Thorson, Nat. Biotechnol., 2003, 21, 1467–1469.
7 S. N. Dupuis, A. W. Robertson, T. Veinot, S. M. A. Monro,
S. E. Douglas, R. T. Syvitski, K. B. Goralski, S. A. McFarland
and D. L. Jakeman, Chem. Sci., 2012, 3, 1640–1644.
8 B. Ruan, K. Pong, F. Jow, M. Bowlby, R. A. Crozier, D. Liu,
S. Liang, Y. Chen, M. L. Mercado, X. Feng, F. Bennett,
D. von Schack, L. McDonald, M. M. Zaleska, A. Wood,
P. H. Reinhart, R. L. Magolda, J. Skotnicki, M. N. Pangalos,
F. E. Koehn, G. T. Carter, M. Abou-Gharbia and
E. I. Graziani, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 33–38.
9 J. Kennedy, Nat. Prod. Rep., 2008, 25, 25–34.
10 A. Kirschning and F. Hahn, Angew. Chem., Int. Ed., 2012,
51, 4012–4022.
11 H. G. Menzella, R. Reid, J. R. Carney, S. S. Chandran,
S. J. Reisinger, K. G. Patel, D. A. Hopwood and D. V. Santi,
Nat. Biotechnol., 2005, 23, 1171–1176.
12 R. McDaniel, A. Thamchaipenet, C. Gustafsson, H. Fu,
M. Betlach and G. Ashley, Proc. Natl. Acad. Sci. U. S. A.,
1999, 96, 1846–1851.
13 P. Kumar, A. T. Koppisch, D. E. Cane and C. Khosla, J. Am.
Chem. Soc., 2003, 125, 14307–14312.
Sfp-catalyzed acylation of DEBS apo-AT°-Mod6TE and triketide
lactone formation assay
Reactions containing 100 mM sodium phosphate (pH 7),
MgCl2 (2 mM), ATP (4 mM), coenzyme A (4 mM), 1a–l (16 mM)
and suitable MatB mutant (10 μg) at a final volume of 50 μL
were incubated at 25 °C for 24 h, or until conversion to the
acyl-CoA was complete (as judged by HPLC, see above). The
optimal MatB mutant used was as described for the MatB-cata-
lyzed acyl-CoA syntheses (see above). 20 μL of the reaction
product mixture was added directly to 80 μL of 50 mM Tris-
HCI (pH 8.8) containing, 10 mM MgCl2, 1.3 mg apo-AT
°-Mod6TE (see above for expression and purification), and
20 μg Sfp, and incubated at 25 °C for 3 h. Reactions were ana-
lyzed by RP-HPLC and LC-MS analysis as described for the
holo-Mod6TE reactions (Fig. S10 and Table S4†), except 50 μL
of the reaction mixture was used for HPLC analysis instead of
25 μL.
14 C. D. Reeves, S. Murli, G. W. Ashley, M. Piagentini,
C. R. Hutchinson and R. McDaniel, Biochemistry, 2001, 40,
15464–15470.
Competition experiments
15 U. Sundermann, K. Bravo-Rodriguez, S. Klopries,
S. Kushnir, H. Gomez, E. Sanchez-Garcia and F. Schulz,
ACS Chem. Biol., 2013, 8, 443–450.
16 F. Del Vecchio, H. Petkovic, S. G. Kendrew, L. Low,
B. Wilkinson, R. Lill, J. Cortes, B. A. M. Rudd, J. Staunton
and P. F. Leadlay, J. Ind. Microbiol. Biotechnol., 2003, 30,
489–494.
17 D. L. Stassi, S. J. Kakavas, K. A. Reynolds, G. Gunawardana,
S. Swanson, D. Zeidner, M. Jackson, H. Liu, A. Buko and
L. Katz, Proc. Natl. Acad. Sci. U. S. A., 1998, 95, 7305–7309.
18 C. Hertweck, Angew. Chem., Int. Ed., 2009, 48, 4688–4716.
19 Y. A. Chan, A. M. Podevels, B. M. Kevany and
M. G. Thomas, Nat. Prod. Rep., 2009, 26, 90–114.
20 M. C. Wilson and B. S. Moore, Nat. Prod. Rep., 2011, 29, 72–
86.
Triketide lactone formation catalyzed by holo-Mod6TE was
assayed in the presence of varying concentrations of 2b and
each 2c, 2e, and 2l. Concentrations of 2b/non-native extender
unit were 0.3/1.5 mM, 0.12/0.6 mM, and 0.03/0.6 mM, respect-
ively. Aside from extender unit concentrations, assay con-
ditions were the same as that for “DEBS holo-Mod6TE
reactions”, described above, while detection was the same as
that for “Sfp-catalyzed acylation of DEBS apo-AT°-Mod6TE and
triketide lactone formation assay”, described above.
Acknowledgements
This study was supported in part by an NSF CAREER Award to
G.J.W. (CHE-1151299). The authors would like to thank the
Mass Spectrometry Facility at NC State, and Daniel Santi and
Dr Adrian Keatinge-Clay for the kind gift of Mod6TE.
21 Y. A. Chan and M. G. Thomas, Methods Enzymol., 2009,
459, 143–163.
22 A. S. Eustaquio, R. P. McGlinchey, Y. Liu, C. Hazzard,
L. L. Beer, G. Florova, M. M. Alhamadsheh, A. Lechner,
A. J. Kale, Y. Kobayashi, K. A. Reynolds and B. S. Moore,
Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 12295–12300.
23 Y. Liu, C. Hazzard, A. S. Eustaquio, K. A. Reynolds and
B. S. Moore, J. Am. Chem. Soc., 2009, 131, 10376–10377.
Notes and references
1 B. M. Trost and G. B. Dong, Nature, 2008, 456, 485–488.
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 4449–4458 | 4457

View Article Online
Paper
Organic & Biomolecular Chemistry
24 M. C. Wilson, S. J. Nam, T. A. Gulder, C. A. Kauffman, 42 A. Misra, S. K. Sharma, N. Surolia and A. Surolia, Chem.
P. R. Jensen, W. Fenical and B. S. Moore, J. Am. Chem. Soc.,
2011, 133, 1971–1977.
25 C. D. Reeves, L. M. Chung, Y. Liu, Q. Xue, J. R. Carney,
Biol., 2007, 14, 775–783.
43 A. T. Koppisch and C. Khosla, Biochemistry, 2003, 42,
11057–11064.
W. P. Revill and L. Katz, J. Biol. Chem., 2002, 277, 9155– 44 L. E. Quadri, P. H. Weinreb, M. Lei, M. M. Nakano,
9159.
P. Zuber and C. T. Walsh, Biochemistry, 1998, 37, 1585–
26 X. Ruan, A. Pereda, D. L. Stassi, D. Zeidner, R. G. Summers,
1595.
M. Jackson, A. Shivakumar, S. Kakavas, M. J. Staver, 45 M. Hans, A. Hornung, A. Dziarnowski, D. E. Cane and
S. Donadio and L. Katz, J. Bacteriol., 1997, 179, 6416–6425. C. Khosla, J. Am. Chem. Soc., 2003, 125, 5366–5374.
27 S. A. Bonnett, C. M. Rath, A. R. Shareef, J. R. Joels, 46 P. Kumar, C. Khosla and Y. Tang, Methods Enzymol., 2004,
J. A. Chemler, K. Hakansson, K. Reynolds and
D. H. Sherman, Chem. Biol., 2011, 18, 1075–1081.
28 A. Y. Chen, N. A. Schnarr, C. Y. Kim, D. E. Cane and
C. Khosla, J. Am. Chem. Soc., 2006, 128, 3067–3074.
388, 269–293.
47 E. M. Musiol, T. Hartner, A. Kulik, J. Moldenhauer, J. Piel,
W. Wohlleben and T. Weber, Chem. Biol., 2011, 18, 438–
444.
29 A. J. Hughes and A. Keatinge-Clay, Chem. Biol., 2011, 18, 48 G. Yadav, R. S. Gokhale and D. Mohanty, J. Mol. Biol., 2003,
165–176. 328, 335–363.
30 N. L. Pohl, M. Hans, H. Y. Lee, Y. S. Kim, D. E. Cane and 49 S. Donadio and L. Katz, Gene, 1992, 111, 51–60.
C. Khosla, J. Am. Chem. Soc., 2001, 123, 5822–5823.
31 I. Koryakina, J. McArthur, S. Randall, M. M. Draelos,
E. M. Musiol, D. C. Muddiman, T. Weber and
G. J. Williams, ACS Chem. Biol., 2013, 8, 200–208.
32 I. Koryakina and G. J. Williams, ChemBioChem, 2011, 12,
2289–2293.
33 R. S. Gokhale, S. Y. Tsuji, D. E. Cane and C. Khosla,
Science, 1999, 284, 482–485.
34 A. Ranganathan, M. Timoney, M. Bycroft, J. Cortes,
50 J. F. Aparicio, I. Molnar, T. Schwecke, A. Konig,
S. F. Haydock, L. E. Khaw, J. Staunton and P. F. Leadlay,
Gene, 1996, 169, 9–16.
51 A. D. Harper, C. B. Bailey, A. D. Edwards, J. F. Detelich and
A. T. Keatinge-Clay, ChemBioChem, 2012, 13, 2200–2203.
52 R. Pieper, G. Luo, D. E. Cane and C. Khosla, Nature, 1995,
378, 263–266.
53 J. D. Mortison, J. D. Kittendorf and D. H. Sherman, J. Am.
Chem. Soc., 2009, 131, 15784–15793.
I. P. Thomas, B. Wilkinson, L. Kellenberger, U. Hanefeld, 54 M. I. Kim, S. J. Kwon and J. S. Dordick, Org. Lett., 2009, 11,
I. S. Galloway, J. Staunton and P. F. Leadlay, Chem. Biol.,
1999, 6, 731–741.
35 B. S. Kim, T. A. Cropp, G. Florova, Y. Lindsay,
3806–3809.
55 Q. Cheng, L. Xiang, M. Izumikawa, D. Meluzzi and
B. S. Moore, Nat. Chem. Biol., 2007, 3, 557–558.
D. H. Sherman and K. A. Reynolds, Biochemistry, 2002, 41, 56 A. J. Link, M. K. Vink, N. J. Agard, J. A. Prescher,
10827–10833.
C. R. Bertozzi and D. A. Tirrell, Proc. Natl. Acad. Sci. U. S. A.,
2006, 103, 10180–10185.
57 I. Chen, B. M. Dorr and D. R. Liu, Proc. Natl. Acad. Sci. U. S.
A., 2011, 108, 11399–11404.
58 Y. Zou and J. Yin, Bioorg. Med. Chem. Lett., 2008, 18, 5664–
5667.
59 S. Murli, J. Kennedy, L. C. Dayem, J. R. Carney and
J. T. Kealey, J. Ind. Microbiol. Biotechnol., 2003, 30, 500–
509.
36 K. Watanabe, C. C. C. Wang, C. N. Boddy, D. E. Cane and
C. Khosla, J. Biol. Chem., 2003, 278, 42020–42026.
37 G. F. Liou, J. Lau, D. E. Cane and C. Khosla, Biochemistry,
2003, 42, 200–207.
38 A. F. Marsden, P. Caffrey, J. F. Aparicio, M. S. Loughran,
J. Staunton and P. F. Leadlay, Science, 1994, 263, 378–380.
39 S. J. Moss, C. J. Martin and B. Wilkinson, Nat. Prod. Rep.,
2004, 21, 575–593.
40 B. Wilkinson, G. Foster, B. A. Rudd, N. L. Taylor, 60 S. K. Piasecki, C. A. Taylor, J. F. Detelich, J. Liu, J. Zheng,
A. P. Blackaby, P. J. Sidebottom, D. J. Cooper, M. J. Dawson,
A. D. Buss, S. Gaisser, I. U. Bohm, C. J. Rowe, J. Cortes,
A. Komsoukaniants, D. R. Siegel and A. T. Keatinge-Clay,
Chem. Biol., 2011, 18, 1331–1340.
P. F. Leadlay and J. Staunton, Chem. Biol., 2000, 7, 111–117. 61 Z. Ye, M. Bair, H. Desai and G. J. Williams, Mol. BioSyst.,
41 C. J. Arthur, A. E. Szafranska, J. Long, J. Mills, R. J. Cox, 2011, 7, 3152–3156.
S. C. Findlow, T. J. Simpson, M. P. Crump and J. Crosby, 62 J. Yin, A. J. Lin, D. E. Golan and C. T. Walsh, Nat. Protoc.,
Chem. Biol., 2006, 13, 587–596.
2006, 1, 280–285.
4458 | Org. Biomol. Chem., 2013, 11, 4449–4458
This journal is © The Royal Society of Chemistry 2013