A mechanism-based fluorescence transfer assay for examining ketosynthase selectivity

Article abstract of DOI:10.1039/c2ob26008e

Since their discovery, polyketide synthases have received massive attention from researchers hoping to harness their potential as a platform for generating new and improved therapeutics. Despite significant strides toward this end, inherent specificities within the enzymes responsible for polyketide production have severely limited these efforts. We have developed a mechanism-based, fluorescence transfer assay for a key enzyme component of all polyketide synthases, the ketosynthase domain. As demonstrated, this method can be used with both ketosynthase-containing didomains and full modules. As proof of principle, the ketosynthase domain from module 6 of the 6-deoxyerythronolide synthase is examined for its ability to accept a variety of simple thioester substrates. Consistent with its natural hexaketide substrate, we find that this ketosynthase prefers longer, α-branched thioesters and its ability to distinguish these structural features is quite remarkable. Substrate electronics are also tested via a variety of p-substituted aromatic groups. In all, we expect this technique to find considerable use in the field of polyketide biosynthesis and engineering due to its extraordinary simplicity and very distinct visible readout.

Full text of DOI:10.1039/c2ob26008e

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 6717
www.rsc.org/obc
PAPER
A mechanism-based fluorescence transfer assay for examining ketosynthase
selectivity†
Gitanjeli Prasad, Lawrence S. Borketey, Tsung-Yi Lin and Nathan A. Schnarr*
Received 23rd May 2012, Accepted 26th June 2012
DOI: 10.1039/c2ob26008e
Since their discovery, polyketide synthases have received massive attention from researchers hoping to
harness their potential as a platform for generating new and improved therapeutics. Despite significant
strides toward this end, inherent specificities within the enzymes responsible for polyketide production
have severely limited these efforts. We have developed a mechanism-based, fluorescence transfer assay
for a key enzyme component of all polyketide synthases, the ketosynthase domain. As demonstrated,
this method can be used with both ketosynthase-containing didomains and full modules. As proof of
principle, the ketosynthase domain from module 6 of the 6-deoxyerythronolide synthase is examined for
its ability to accept a variety of simple thioester substrates. Consistent with its natural hexaketide
substrate, we find that this ketosynthase prefers longer, α-branched thioesters and its ability to distinguish
these structural features is quite remarkable. Substrate electronics are also tested via a variety of
p-substituted aromatic groups. In all, we expect this technique to find considerable use in the field of
polyketide biosynthesis and engineering due to its extraordinary simplicity and very distinct visible
readout.
Introduction
Polyketide natural products are biosynthesized via successive
Claisen-like condensation of malonate-derived units governed by
multi-component enzyme systems termed polyketide synthases
(PKSs).1–6 Over the past several decades, the field of polyketide
engineering has faced a constant struggle to overcome often
strict substrate selectivities of the critical biosynthetic
enzymes.7–13 For modular PKS systems, the shuttling of inter-
Fig. 1 Schematic diagram of intermodular transfer in modular poly-
mediates between the acyl carrier protein (ACP) and keto-
synthase (KS) domains dictates the overall flow of polyketide
maturation along the assembly line (Fig. 1).14–17 As a result, KS
substrate selectivity has been the focus of numerous studies with
the hopes of both establishing a better understanding of natural
systems and improving our ability to manipulate them.18–21 In its
simplest form, this requires a means of accurately differentiating
between substrates that acylate the KS active site-cysteine and
those that do not. Unfortunately, the methods traditionally used
to examine KS active site loading have significant drawbacks,
including limited scope, reproducibility, and cost that can
severely constrain the scope of these analyses. It is abundantly
clear that techniques capable of providing easy and reliable
ketide synthases. A polyketide intermediate is passed from an upstream
ACP to a downstream KS via the phosphopantetheine arm of holo-ACP.
Often the substrate selectivity of the acceptor KS domain determines the
efficiency of this process and controls the flow of intermediates down
the assembly line. KS = ketosynthase, ACP = acyl carrier protein, AT =
acyltransferase, DH = dehydratase, ER = enoyl reductase, KR =
ketoreductase.
access to KS selectivity profiles are essential for continued pro-
gress in this field.
Early determinations of KS substrate selectivity relied heavily
on radiolabeling of elongation products or, less commonly, the
substrates themselves.15–17,21 Small molecules were generally
delivered to the KS active site via a compatible ACP or as an N-
acetylcysteamine (SNAc) thioester, a more synthetically accessi-
ble mimic of the phosphopantetheine (Ppant) arm of holo-ACPs.
Subsequent addition of radiolabeled extender units resulted in a
single elongation event with the KS-bound substrate. The
amount of labeled product excised from the protein provided an
Department of Chemistry, University of Massachusetts, 710 N. Pleasant
Street, Amherst, Massachusetts 01003, USA.
E-mail: schnarr@chem.umass.edu
†Electronic supplementary information (ESI) available: compound
characterisation. See DOI: 10.1039/c2ob26008e
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6717–6723 | 6717

View Article Online
indirect measure of KS loading with each substrate. While very
sensitive, the combination of high reagent cost and limited sub-
strate availability has severely restricted the use of this method
for medium and high-throughput applications.
More recently, a number of groups have developed mass spec-
trometry-based techniques for examining KS active site acyl-
ation.19,20,22 Many of these strategies utilize tandem proteolysis/
LC-MS to tease out differences in loading propensities for
various substrates. Despite the many advantages compared to
radiolabeling, MS-based methods still suffer from a number of
important shortcomings. Firstly, the extremely large size of intact
PKS modules generally prohibits their use in these analyses.
Smaller subunits such as KSAT didomains are often employed
as substitutes for the full module and currently it is unclear what,
if any, effect this has on substrate selectivity. In addition, the
influence of various substrates on ionization potentials,
especially when proteolysis is employed prior to analysis, has
not been fully explored and may significantly impact data
interpretation and reproducibility. Finally, when particular hydro-
phobic groups (e.g. octyl, napthyl, cyclohexyl) are utilized as a
substantial part of a given substrate, loss of signal for proteolysis
fragments associated with KS domains is often observed, likely
due to solubility issues.
As advances in genomic mining continue to uncover hosts of
new biosynthetic systems, the need for simple, cost-effective
means of characterizing them has never been greater. Conse-
quently, we were compelled to develop an alternative means of
directly determining KS substrate selectivity using readily acces-
sible, inexpensive reagents and equipment. Such a method
would greatly facilitate the process of high-throughput PKS
investigation and allow for broadened exploration of the rules of
substrate recognition that govern small molecule output.
Fig. 3 Schematic diagram of a mechanism-based, fluorescence transfer
assay for KS-selectivity. A given KS domain is incubated with acyl-
SNAc followed by addition of fluorescently labeled ACP. The amount of
fluorescence transferred from the ACP to the KS active site provides a
direct readout of the substrate tolerance for each acyl-SNAc examined.
KS = ketosynthase, ACP = acyl carrier protein, AT = acyltransferase,
SNAc = N-acetylcysteamine.
means of examining KS selectivity in whole PKS modules.
More specifically, the degree of KS acylation for a given SNAc
substrate may be assayed by introducing a fluorescent ACP after
incubation of KS and substrate and monitoring fluorescence
transfer via gel electrophoresis. The amount of fluorescence
transfer would therefore indicate to what extent the SNAc com-
pound was accepted by the KS domain (Fig. 3). High
KS-loading would block subsequent fluorescence transfer from
ACP, and thus produce little to no fluorescent signal in the gel.
Alternatively, insufficient KS-acylation with a SNAc compound
would leave most KS active sites free to accept the fluorescent
substrate from ACP resulting in a bright band. This technique
would afford a direct measure of KS selectivity while circum-
venting the limitations of mass spectrometry methods described
above. Most importantly, the use of mechanism-based fluor-
escence transfer from acyl-ACPs ensures that any fluorescence
observed on the acceptor module results solely from the KS
active site and not from competing, nucleophilic surface residues
within the complex protein system.
We expect the straightforward nature of this method to permit
all research labs to explore key features of polyketide construc-
tion. Studies ranging in scope from basic KS selectivity to more
complex ACP/KS complementarity for designing hybrid systems
will be greatly simplified, ultimately leading to significant
advances in PKS engineering. Herein, we describe our initial
work toward designing a mechanism-based, fluorescence transfer
method for examining substrate selectivity in PKS assemblies.
Recently, our lab has developed a series of β-lactone and
β-lactam probes capable of directly acylating the Ppant-thiol of
holo-ACPs.23,24 Incorporation of an alkyne functionality at
various ring positions provides a chemical handle for conju-
gation with fluorophores (Fig. 2). The thioester-ACP resulting
from incubation of holo-ACP with a propargyloxycarbonyl
(Poc)-activated β-lactam was shown to be competitive with tra-
ditional N-acetylcysteamine (SNAc) thioesters for KS acylation.
Based on these results, it occurred to us that fluorescent thio-
ester-ACPs, derived from an alkyne-bearing β-lactam, coupled
with acyl-SNAc compounds, could provide a straightforward
Results and discussion
We knew from our previous work that the 6-deoxyerythronolide
B synthase (DEBS) module 2 ACP (ACP2) could efficiently
transfer a β-lactam-derived fluorescent probe to the KS active
site of the DEBS module 6 KSAT25 didomain (KSAT6).24 As a
result, we decided to begin with this ACP/KS pair for proof of
principle studies. It should be noted that careful choice of the
Fig. 2 Schematic representation of our method for site-selective ACP-
acylation using N-activated β-lactams. Subsequent conjugation to a
fluorophore (red star) uses standard click chemistry. The resulting fluore-
scent ACP is used as the fluorescence donor in the current study.
6718 | Org. Biomol. Chem., 2012, 10, 6717–6723
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 6 SDS-PAGE analysis of fluorescence transfer from ACP2 to
KSAT6 pretreated with compounds 1–6. Lack of fluorescence indicates
near quantitative acylation of the KS active site with compound 1–6.
Bright bands indicated little to no KS loading. Lane 1 corresponds to the
positive control where no acyl-SNAc is added. Numbers at the top indi-
cate which compound KS was treated with prior to addition of fluor-
escent ACP.
Fig. 4 Panel of SNAc thioesters used to probe chain-length selectivity.
Fig. 5 SDS-PAGE analysis of fluorescence transfer from ACP2 to
KSAT6 pretreated with variable concentrations of compound 1. Lack of
fluorescence indicates near quantitative acylation of the KS active site
with compound 1. Bright bands indicated little to no KS loading. Lane 1
corresponds to the positive control where no compound 1 is added.
Numbers at the top indicate the concentration of compound 1 added to
KSAT6.
Fig. 7 Panel of SNAc thioesters used to probe α- and β-branching
selectivity.
donor ACP is critical for success using this technique. We had
previously determined that fluorescence transfer from ACP2 to
KSAT6 was complete in as little as 30 min and no transfer
occurred in the presence of the KS active site inhibitor cerulenin.
The speed and specificity of this process prompted us to begin
development of a mechanism-based fluorescence transfer assay
capable of directly probing the influence of substrate structure
(i.e. chain length, sterics, and electronics) on KS loading.
The variable chain length SNAc compounds (1–6) were sub-
jected to the same analysis as above using 2.5 mM of each. To
our satisfaction, a clear preference for longer chain substrates
was observed (Fig. 6). Given that DEBS module 6 KS normally
accepts a hexaketide intermediate from module 5, these results
are consistent with what might be expected of modules further
down the biosynthetic assembly line. Most importantly, the
observed trend was reproducible over multiple, separate runs
(see ESI†). It should be noted that previous attempts on our part
to monitor KS-loading with compounds 5 and 6 via tandem pro-
teolysis/mass spectrometry were unsuccessful most likely due to
precipitation of the acylated KS active site fragment.
With a simple yet powerful method in hand, we were excited
to examine the affects of more complex structural components
on KS-loading. Specifically, we were interested in uncovering
any substrate preferences for branched alkyl-SNAc compounds.
A small panel of these compounds was prepared with variable
branching at both the α- and β-positions of the carbonyl (Fig. 7).
Since the natural polyketide intermediate accepted by DEBS
module 6 KS harbors an α-methyl group, it is not surprising that
compounds 9, 11, and 12 appear to load KSAT6 quite efficiently
while compounds 7 and 8 are not as effective (Fig. 8). The
cyclopropyl-containing compound 10, however, is very interest-
ing. Despite the presence of an α-branch, this substrate clearly
behaves more like those without α-substituents. It is tempting to
speculate that this may arise from the extreme cyclopropane
bond angles which might alter the side chain geometry to the
point where the KS active site does not recognize any branching.
Regardless of its origins, this is a rather remarkable response to
very subtle differences in structure between isopropyl and cyclo-
propyl groups.
Based on the structure for the natural hexaketide recognized
by module 6 KS, we hypothesized that longer chain substrates
would be preferred.2 In addition, module 5 incorporates a
methylmalonyl-extender unit which, following elongation, is
passed to module 6. As a result, we further hypothesized that
α-branched thioesters may be particularly well-suited for acylat-
ing the KS domain. Our fluorescence transfer method would
allow us to readily observe these tendencies and, more impor-
tantly, provide high-resolution information as to their limits.
To examine chain length preferences for KSAT6, a panel of
SNAc thioesters derived from straight-chain carboxylates was
prepared (Fig. 4). Before we could begin, however, it was
necessary to determine an appropriate acyl-SNAc concentration
where subtle differences in KS-loading could be readily
observed. Fluorescently labeled ACP was prepared as previously
described by reacting holo-ACP2 with N-propargyloxycarbonyl-
β-lactam followed by Cu-catalyzed [3 + 2] cycloaddition with
rhodamine azide (see Fig. 2).26–28 Excess click reagents were
removed via filtration to avoid any complications during the
assay. To determine the optimal substrate concentration for KS-
selectivity studies, compound 1 was introduced to KSAT6 at
concentrations ranging from 630 μM to 10 mM. After 1 h incu-
bation, a slight excess, relative to KSAT6, of fluorescent ACP2
was added for 30 min and the protein components were sub-
sequently separated and analyzed by SDS-PAGE. From these
data, it was determined that the inflection point between fully
loaded and unloaded KS corresponded to a substrate concen-
tration of roughly 2.5 mM for compound 1 (Fig. 5).
At this point the next logical step, in our minds, was to
examine electronic aspects of KS-acylation using substrates
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6717–6723 | 6719

View Article Online
Fig. 8 SDS-PAGE analysis of fluorescence transfer from ACP2 to
KSAT6 pretreated with compounds 7–12. Lack of fluorescence indicates
near quantitative acylation of the KS active site with compound 1.
Bright bands indicated little to no KS loading. Lane 1 corresponds to the
positive control where no acyl-SNAc is added. Numbers at the top indi-
cate which compound KS was treated with prior to addition of fluor-
escent ACP.
Fig. 11 SDS-PAGE analysis of fluorescence transfer from ACP2 to
apo-module 6 pretreated with compounds 1–18. Lack of fluorescence
indicates near quantitative acylation of the KS active site with compound
1. Bright bands indicated little to no KS loading. Lane 1 corresponds to
the positive control where no acyl-SNAc is added. Numbers at the top of
each gel indicate which compound module 6 was treated with prior to
addition of fluorescent ACP. Overall trends in each gel are consistent
with what was observed for KSAT6 above indicating that the presence of
DH, ER, KR, and ACP domains does not significantly alter
KS-selectivity.
Fig. 9 Panel of SNAc thioesters used to probe KS-selectivity for
p-substituted aryl groups.
For comparison to the previous data sets, we reasoned that
DEBS module 6 would be the most appropriate target for this
study. This module contains a ketoreductase, acyl carrier protein,
and thioesterase domain in addition to the KS and AT present in
the didomain construct. To avoid any competition with the
native ACP of this module, we opted to express the apo form
which lacks the reactive Ppant arm. In addition, the potential for
fluorescence transfer from ACP2 to the thioesterase domain,
positioned at the C-terminus of DEBS module 6, prompted us to
omit this entity from the construct used in this study. With the
optimized full module in hand, the same 18 probes used to
probe KSAT6 selectivity were applied.
Fig. 10 SDS-PAGE analysis of fluorescence transfer from ACP2 to
KSAT6 pretreated with compounds 13–18. Lack of fluorescence indi-
cates near quantitative acylation of the KS active site with compound 1.
Bright bands indicated little to no KS loading. Lane 1 corresponds to the
positive control where no acyl-SNAc is added. Numbers at the top indi-
cate which compound KS was treated with prior to addition of fluor-
escent ACP.
To our satisfaction, all of the trends observed for KS-selectiv-
ity in the didomain precisely tracked those for the intact module
although the contrast between “good” and “bad” substrates is
less obvious (Fig. 11). The preference for both longer chain and
α-branched substrates is again evident and the remarkable dis-
tinction between isopropyl and cyclopropyl groups appears con-
sistent. The observed preference for the nitrophenyl-containing
compound 16 over all other aryl-SNAc substrates provides
further credence to the role of electronics in KS-acylation. Taken
together, these results offer a clear picture of the substrate scope
tolerated by this KS in two unique environments. The fact that
the observed trends are consistent between the didomain and
intact module is encouraging as “broken modules” become
increasingly popular for modern mechanistic studies of PKS
systems.
bearing various electron withdrawing and donating aromatic
groups. Therefore, a set of acyl-SNAc compounds was prepared
with an assortment of p-substituents for comparison to the parent
phenyl ring of compound 13 (Fig. 9). From the data, it appears
that all p-substituents confer at least a slight reactive advantage
relative to the unsubstituted phenyl ring (Fig. 10). The nitro-
phenyl compound 16 is most reactive as might be expected from
the increased carbonyl electrophilicity offered by the electron
withdrawing nature of the nitro group. Clearly, the electronics of
the thioester play a significant role in KS acylation, but without
structural information regarding the active site binding inter-
actions of each of these substrates, it is difficult to speculate
further. Taken together with the results from the chain length and
branching studies, this method is capable of generating a wealth
of information regarding KS selectivity in short order with
minimal cost and synthetic manipulation. However, to showcase
its maximum value for examining substrate tolerance in poly-
ketide biosynthesis, we felt it appropriate to move away from
didomains in favor of full PKS modules.
Conclusion
With this work, we have successfully developed a mechanism-
based, fluorescence transfer assay for substrate selectivity in KS
6720 | Org. Biomol. Chem., 2012, 10, 6717–6723
This journal is © The Royal Society of Chemistry 2012

View Article Online
Protein purification and isolation
domains. The technique, in its current form, provides a simple,
qualitative output without the need for expensive reagents or
equipment. The experiments described above barely scratch the
surface of what is possible with this method. We envision a
number of diverse applications including: (1) further exam-
ination of substrate attributes (i.e. stereochemistry, heteroatom
incorporation, etc.), (2) KS/ACP complementarity, (3) deter-
mining the impact of substrate binding pocket mutations within
KS domains on substrate tolerance, and (4) assessing the
influence of exogenous domains on KS selectivity in “chimeric”
modules. The work described in this manuscript properly lays
the necessary groundwork toward each of these endeavors. In
all, we expect this method to significantly aid in the study of
new and existing PKS systems leading to improved understand-
ing of how these extraordinary biosynthetic machines function.
Bacteria were grown in 1 L shake cultures of LB-antibiotic
media at 37 °C in an incubating shaker until the OD600 was
between 0.6 and 0.8. Over expression was induced with 200 μL
of 1 M IPTG (per liter of culture) and carried out at 18 °C for
18 h. After this point all work was carried out at 4 °C. Cells
were pelleted by spinning at 5000 RPM for 10 min and resus-
pended in 50 mL of lysis buffer (20 mM Tris-HCl, 150 mM
NaCl, 1 mM Na2EDTA, 1 mM EGTA, 1% Triton, 2.5 mM
sodium pyrophosphate, 1 mM β-glycerophosphate, 1 mM sodium
Na3VO4, 1 μg mL−1 leupeptin, pH 7.5). Cells were lysed
using an ultrasonic converter. The lysate was pelleted at
10 000 rpm for 60 min. The supernatant was equilibrated with
3 mL of Ni-NTA slurry (per liter of culture) for 60 min by stir-
ring with a PTFE-coated stir bar at minimal speed. The mixture
was then poured into a fritted column and the supernatant eluted.
The resin bed was washed with two 15 mL portions of wash
buffer (50 mM phosphate, 300 mM NaCl, 20 mM imidazole,
pH 8.0), and eluted with 3 mL (per liter of culture) of elution
buffer (50 mM phosphate, 300 mM NaCl, 250 mM imidazole,
pH 8.0). The purified protein was loaded into a 3 kDa NMW
centrifugal concentrator and diluted to 15 mL with storage buffer
(100 mM Tris, 1 mM EDTA, 1 mM dithioerythritol, 10% glycerol
pH 8) and spun until the final volume was ≤500 μL. Dilution
and filtration was repeated a total of three times. Protein concen-
tration was determined by Bradford assay with a BSA standard
curve. Proteins were flash frozen and stored at −80 °C until use.
Methods
Materials
All moisture sensitive reactions were performed under argon in
flame-dried glassware with Teflon-coated stir bars. Reagents and
solvents for organic synthesis were used as received unless other-
wise noted. Dichloromethane was distilled from calcium hydride
and stored over activated molecular sieves. Thin-layer chroma-
tography (TLC) was conducted on glass-backed silica plates
visualized with either UV illumination or basic permanganate
stain. Flash chromatography was carried out with 60 Å silica gel.
1H and 13C NMR was performed on a 400 MHz Bruker NMR.
Fluorescence transfer assay
Unless otherwise stated, phosphate buffer refers to 100 mM, pH
7.0 phosphate. ACP2 (25 μM total protein) was equilibrated at
ambient temperature for 15 min in phosphate buffer containing
2.5 mM TCEP. Poc β-lactam (10× with respect to total ACP con-
centration) was added and the mixture was equilibrated at
ambient temperature for 45 min. Rhodamine Azide (2× with
respect to Poc β-lactam), sodium ascorbate (1 mM), and copper(II)
sulfate (1 mM) were added to samples which had been labeled
with 6. The reaction was performed at ambient temperature for
45 min. To remove excess Poc β-lactam and Rhodamine Azide,
the mixture was loaded into a 3 kDa NMW concentrator and the
volume reduced to 100 μL by spinning in a centrifuge cooled to
4 °C. The mixture was diluted to 500 μL with phosphate buffer,
and then concentrated to 100 μL again. This process was
repeated a total of 5 times. Protein was removed from the con-
centrator by inverting it and spinning. For transfer of the fluor-
escent product, [KSAT6] or the full module (25 μM) was pre-
treated with respective SNAc (2.5 mM) derivatives for 1 h. This
mixture was introduced to filtered acyl-ACP2 and incubated for
30 min. The [KS6AT6] and apo-Mod6 control was executed
under the same acylation and click conditions as in other
samples but without the SNAc derivative and therefore rep-
resents full acyl transfer.
Protein expression
ACP2 was expressed from pNW06,14 ACP3 was expressed from
pVYA05.25 [KS6AT6] was expressed from pAYC11.18 apo-
ACPs were harvested from E. coli BL-21 and holo-ACPs were
harvested from E. coli BAP-1.31 pNW06 and pVYA05 contain
kanamycin resistant vectors.
Cloning (pTL-DM6)
Cloning of DEBS-Mod 6 (without TE domain) was done by
PCR on pRSG54 (a gift from Professor Khosla, which encodes
for DEBS Mod 6+TE). Using the primer DEBS-M6-NdeI-F1:
TTTTTTCATATGGACCCGATCGCGATCGTCGGCATG, and
DEBS-M6-HindIII-R1: AAAAAAAAGCTTGAGCTGCTGTC-
CTATGTGGTCGGC, PCR was performed in a 50 μL reaction
mixture containing: 1 × Phusion GC buffer, 0.2 mM dNTP,
0.2 mM MgCl2, 15% Glycerol, 1 μM of each primer, 1 U
Phusion Hot Start II DNA polymerase (Thermo scientific), and
approximately 300 ng of pRSG54 DNA. PCR amplicon of 4215
bp was obtained after 35 cycles PCR amplification. The DNA
amplicons were excised and purified by using QIAquick Gel
Extraction Kit (QIAGEN, Germany) and then directly subjected
to double restriction enzyme digestion of NdeI and HindIII. The
digested product was ligated to pre-digested pET-21b to obtain
pTL-DM6.
Chromophore attachment
The reaction was carried out at 25 μL, final concentrations
reported. Rhodamine Azide, (2× with respect to Poc β-lactam),
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6717–6723 | 6721

View Article Online
sodium ascorbate (1 mM), and copper(II) sulfate (1 mM) were
added to samples which had been labeled with 6. The reaction
was performed at ambient temperature for 45 min.
under vacuum, and purified by flash column to provide the final
product in pure form.
Acknowledgements
Gel assay
This work was supported by start-up funding from the University
of Massachusetts, Amherst (to N.A.S.). The authors would like
to thank Professor Chaitan Khosla for the generous gifts of
BAP-1, pNW06, pRSG64 and pAYC11. We also thank Pro-
fessors Alan Kennan and James Chambers for helpful
discussions.
Labeled samples were diluted to 20 μL with gel-loading buffer.
Proteins were separated using a 4–20% gradient HEPES-PAGE
gel (100 V, 50 mA, 70 min). Gels were developed in 10% acetic
acid to visualize Rhodamine Azide. Labeled proteins were
imaged on a UV-transilluminator. Total protein was stained using
Coomassie stain.
References
Prop-2-yn-1-yl 2-oxoazetidine-1-carboxylate
1 D. O’Hagan and J. N. Collie, The Polyketide Metabolites, Ellis Horwood,
Chichester, UK, 1991.
2 C. Khosla, Y. Tang, A. Y. Chen, N. A. Schnarr and D. E. Cane, Structure
and Mechanism of the 6-Deoxyerythronolide B Synthase, Annu. Rev.
Biochem., 2007, 76, 195–221.
2-Azetidinone (107 mg, 1.5 mmol) was dissolved in THF
(7 mL) at −78 °C. LiHDMS (330 mg, 2 mmol) dissolved in
THF (2 mL) was added in portions over 10 min and the reaction
mixture was stirred for 30 min. Propargyl chloroformate
(181 mg, 1.5 mmol) dissolved in THF (2 mL) was added in por-
tions over 10 min and the reaction mixture was stirred at −78 °C
for 2 h then allowed to warm to ambient temperature. The reac-
tion mixture was diluted with water and extracted with DCM
(50 mL × 3). The combined organic layers were washed with
brine and dried over Na2SO4. Solvent was removed by rotary
evaporation. Purification of crude product by flash chromato-
graphy with 95 : 5 DCM–Methanol yields (106 mg, 0.69 mmol)
46% of product as pale yellow solid.
3 J. Staunton and K. J. Weissman, Polyketide biosynthesis: a millennium
review, Nat. Prod. Rep., 2001, 18, 380–416.
4 S. Donadio and L. Katz, Organization of the enzymatic domains in the
multifunctional polyketide synthase involved in erythromycin formation
in Saccharopolyspora erythraea, Gene, 1992, 111, 51–60.
5 S. Donadio, M. J. Staver, J. B. McAlpine, S. J. Swanson and L. Katz,
Modular organization of genes required for complex polyketide biosyn-
thesis, Science, 1991, 252, 675.
6 J. Cortes, S. F. Haydock, G. A. Roberts, D. J. Bevitt and P. F. Leadlay, An
unusually large multifunctional polypeptide in the erythromycin-produ-
cing polyketide synthase of Saccharopolyspora erythraea, Nature, 1990,
348, 176.
7 R. McDaniel, A. Thamchaipenet, C. Gustafsson, H. Fu, M. Betlach and
G. Ashley, Multiple genetic modifications of the erythromycin polyketide
synthase to produce a library of novel “unnatural” natural products, Proc.
Natl. Acad. Sci. U. S. A., 1999, 96, 1846–1851.
8 Q. Xue, G. Ashley, C. Hutchinson and D. Santi, A multiplasmid approach
to preparing large libraries of polyketides, Proc. Natl. Acad. Sci. U. S. A.,
1999, 96, 11740–11745.
9 N. A. Schnarr and C. Khosla, Combinatorial biosynthesis of polyketides
and nonribosomal peptides, in Chemical Biology: From Small Molecules
to Systems Biology and Drug Design, Wiley-VCH, Weinheim, 2006.
10 S. Donadio and M. Sosio, Strategies for combinatorial biosynthesis with
modular polyketide synthases, Comb. Chem. High Throughput Screening,
2003, 6, 489–500.
11 C. T. Walsh, Combinatorial biosynthesis of antibiotics: challenges and
opportunities, ChemBioChem, 2003, 3, 124–134.
12 J. Staunton and B. Wilkinson, Combinatorial biosynthesis of
polyketides and nonribosomal peptides, Curr. Opin. Chem. Biol., 2001,
5, 159–164.
Sulforhodamine-B azide
3-Azido-1-aminopropane29 (22 mg, 0.22 mmol) and triethyl-
amine (51 mg, 0.5 mmol) were dissolved in 1 mL of 5 : 1
DCM–DMF and cooled to 0 °C. Sulforhodamine B sulfonyl
chloride [mixture of ortho- and para-sulfonyl chloride isomers]
(115 mg, 0.20 mmol) was added portion-wise over ca. 30 min
and stirred overnight at room temperature. Solvent was removed
in vacuo, and the product was purified by flash chromatography
over silica gel with 90 : 5 : 5 DCM–Acetonitrile–Methanol
mobile phase. The ortho isomer was distinguished by its rever-
sible color change in pH 9.0 buffer.30 Para isomer: (22 mg
0.034 mmol) 17%, ortho isomer (18 mg, 0.028 mmol) 14%.
Both obtained as a red solid with a metallic green luster. The
ortho isomer was found to perform best in the click reaction and
was utilized for labeling experiments.
13 D. E. Cane, C. T. Walsh and C. Khosla, Harnessing the biosynthetic
code: combinations, permutations, and mutations, Science, 1998, 282,
63–68.
14 N. Wu, D. E. Cane and C. Khosla, Quantitative analysis of the relative
contributions of donor acyl carrier proteins, acceptor ketosynthases, and
linker regions to intermodular transfer of intermediates in hybrid poly-
ketide synthases, Biochemistry, 2002, 41, 5056–5066.
15 R. S. Gokhale, S. Y. Tsuji, D. E. Cane and C. Khosla, Dissecting and
exploiting intermodular communication in polyketide synthases, Science,
1999, 284, 482–485.
General method for preparation of all N-acetylcysteamine
(SNAc) thioester derivatives
16 S. Y. Tsuji, D. E. Cane and C. Khosla, Selective protein–protein inter-
actions direct channeling of intermediates between polyketide synthase
modules, Biochemistry, 2001, 40, 2326–2331.
17 N. Wu, S. Y. Tsuji, D. E. Cane and C. Khosla, Assessing the balance
between protein–protein interactions and enzyme–substrate interactions in
the channeling of intermediates between polyketide synthase modules,
J. Am. Chem. Soc., 2001, 123, 6465–6474.
18 A. Y. Chen, N. A. Schnarr, C-Y. Kim, D. E. Cane and C. Khosla, Exten-
der unit and acyl carrier protein specificity of ketosynthase domains of
the 6-deoxyerythronolide B synthase, J. Am. Chem. Soc., 2006, 128,
3067–3074.
To a solution of triethylamine (2.80 mmol) in dichloromethane
(10 mL) was added the appropriate acid (1.40 mmol),
(3-dimethylaminopropyl)-3-ethylcarbodiimide
hydrochloride
(1.40 mmol), 1-hydroxybenzotriazole (HOBt) (1.40 mmol) and
N-acetylcysteamine (SNAc) (1.35 mmol) under argon. The reac-
tion mixture was stirred overnight. The organic layer was washed
with saturated NaHCO3 solution, 0.1 N HCl solution and brine.
It was then dried over anhydrous sodium sulfate, concentrated
6722 | Org. Biomol. Chem., 2012, 10, 6717–6723
This journal is © The Royal Society of Chemistry 2012

View Article Online
19 N. A. Schnarr, A. Y. Chen, D. E. Cane and C. Khosla, Analysis of co-
valently bound polyketide intermediates on 6-deoxyerythronolide
B synthase by tandem proteolysis-mass spectrometry, Biochemistry, 2005,
44, 11836–11842.
20 H. Hong, A. N. Appleyard, A. P. Siskos, J. Garcia-Bernardo, J. Staunton
and P. F. Leadlay, Chain initiation on type I modular polyketide synthases
revealed by limited proteolysis and ion-trap mass spectrometry, FEBS J.,
2005, 272, 2373–2387.
21 J. Wu, K. Kinoshita, C. Khosla and D. E. Cane, Biochemical analysis of
the substrate specificity of the beta-ketoacyl-acyl carrier protein synthase
domain of module 2 of the erythromycin polyketide synthase, Biochemis-
try, 2004, 43, 16301–16310.
22 L. M. Hicks, M. T. Mazur, L. M. Miller, P. C. Dorrestein, N. A. Schnarr,
C. Khosla and N. L. Kelleher, Investigating nonribosomal peptide and
polyketide biosynthesis by direct detection of intermediates on >70 kDa
polypeptides by using Fourier-transform mass spectrometry, ChemBio-
Chem, 2006, 7, 904–907.
25 C. Y. Kim, V. Y. Alekseyev, A. Y. Chen, Y. Tang, D. E. Cane and
C. Khosla, Reconstituting modular activity from separated domains of
6-deoxyerythronolide B synthase, Biochemistry, 2004, 43, 13892–13898.
26 V. V. Rostovtsev, L. G. Green, V. V. Fokin and K. B. Sharpless, A step-
wise huisgen cycloaddition process: copper(I)-catalyzed regioselective
“ligation” of azides and terminal alkynes, Angew. Chem., Int. Ed., 2002,
41, 2596–2599.
27 H. C. Kolb, M. G. Finn and K. B. Sharpless, Click chemistry: diverse
chemical function from a few good reactions, Angew. Chem., Int. Ed.,
2001, 40, 2004–2021.
28 W. G. Lewis, L. G. Green, F. Grynszpan, Z. Radić, P. R. Carlier,
P. Taylor, M. G. Finn and K. B. Sharpless, Click chemistry in situ: acetyl-
cholinesterase as a reaction vessel for the selective assembly of a femto-
molar inhibitor from an array of building blocks, Angew. Chem., Int. Ed.,
2002, 41, 1053–1057.
29 B. Carboni, A. Benalil and M. Vaultier, Aliphatic amino azides as key
building blocks for efficient polyamine syntheses, J. Org. Chem., 1993,
58, 3736–3741.
23 J. W. Amoroso, L. S. Borketey, G. Prasad and N. A. Schnarr, Direct
acylation of carrier proteins with functionalized β-lactones, Org. Lett.,
2010, 12, 2330–2333.
24 G. Prasad, J. W. Amoroso, L. S. Borketey and N. A. Schnarr, N-Activated
β-lactams as versatile reagents for acyl carrier protein labeling, Org.
Biomol. Chem., 2012, 10, 1992–2002.
30 J. E. T. Corrie, C. T. Davis and J. F. Eccleston, Chemistry of sulforhoda-
mine-amine conjugates, Bioconjugate Chem., 2001, 12, 186–194.
31 B. A. Pfeifer, S. J. Admiraal, H. Gramajo, D. E. Cane and C. Khosla,
Biosynthesis of complex polyketides in a metabolically engineered strain
of E. Coli, Science, 2001, 291, 1790.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6717–6723 | 6723