Expanding the chemical space of polyketides through structure-guided mutagenesis of Vitis vinifera stilbene synthase

Article abstract of DOI:10.1016/j.biochi.2015.05.019

Several natural polyketides (PKs) have been associated with important pharmaceutical properties. Type III polyketide synthases (PKS) that generate aromatic PK polyketides have been studied extensively for their substrate promiscuity and product diversity. Stilbene synthase-like (STS) enzymes are unique in the type III PKS class as they possess a hydrogen bonding network, furnishing them with thioesterase-like properties, resulting in aldol condensation of the polyketide intermediates formed. Chalcone synthases (CHS) in contrast, lack this hydrogen-bonding network, resulting primarily in the Claisen condensation of the polyketide intermediates formed. We have attempted to expand the chemical space of this interesting class of compounds generated by creating structure-guided mutants of Vitis vinifera STS. Further, we have utilized a previously established workflow to quickly compare the wild-type reaction products to those generated by the mutants and identify novel PKs formed by using XCMS analysis of LC-MS and LC-MS/MS data. Based on this approach, we were able to generate 15 previously unreported PK molecules by exploring the substrate promiscuity of the wild-type enzyme and all mutants using unnatural substrates. These structures were specific to STSs and cannot be formed by their closely related CHS-like counterparts.

Full text of DOI:10.1016/j.biochi.2015.05.019

Biochimie 115 (2015) 136e143
Contents lists available at ScienceDirect
Biochimie
journal homepage: www.elsevier.com/locate/biochi
Research paper
Expanding the chemical space of polyketides through structure-
guided mutagenesis of Vitis vinifera stilbene synthase
,
,
b,
c
,
, c, *
Namita Bhan a 1, Brady F. Cress a, Robert J. Linhardt a
d, Mattheos Koffas a
a Department of Chemical and Biological Engineering, Rensselaer Polytechnic Institute, Center for Biotechnology and Interdisciplinary Studies, Troy, NY, USA
b Department of Chemistry and Chemical Biology, Rensselaer Polytechnic Institute, Center for Biotechnology and Interdisciplinary Studies, Troy, NY, USA
c Department of Biological Sciences, Rensselaer Polytechnic Institute, Center for Biotechnology and Interdisciplinary Studies, Troy, NY, USA
d Department of Biomedical Engineering, Rensselaer Polytechnic Institute, Center for Biotechnology and Interdisciplinary Studies, Troy, NY, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Several natural polyketides (PKs) have been associated with important pharmaceutical properties. Type
III polyketide synthases (PKS) that generate aromatic PK polyketides have been studied extensively for
their substrate promiscuity and product diversity. Stilbene synthase-like (STS) enzymes are unique in the
type III PKS class as they possess a hydrogen bonding network, furnishing them with thioesterase-like
properties, resulting in aldol condensation of the polyketide intermediates formed. Chalcone syn-
thases (CHS) in contrast, lack this hydrogen-bonding network, resulting primarily in the Claisen
condensation of the polyketide intermediates formed. We have attempted to expand the chemical space
of this interesting class of compounds generated by creating structure-guided mutants of Vitis vinifera
STS. Further, we have utilized a previously established workflow to quickly compare the wild-type
reaction products to those generated by the mutants and identify novel PKs formed by using XCMS
analysis of LC-MS and LC-MS/MS data. Based on this approach, we were able to generate 15 previously
unreported PK molecules by exploring the substrate promiscuity of the wild-type enzyme and all mu-
tants using unnatural substrates. These structures were specific to STSs and cannot be formed by their
closely related CHS-like counterparts.
Received 9 April 2015
Accepted 22 May 2015
Available online 3 June 2015
Keywords:
Stilbene synthase
Aromatic polyketides
Type III polyketide synthases
Flavonoids
Resveratrol
ꢀ ꢀ
ꢀ
© 2015 Elsevier B.V. and Societe Française de Biochimie et Biologie Moleculaire (SFBBM). All rights
reserved.
1. Introduction
diverse type III PKSs arise due to variable substitutions in non-
catalytic residues present in the three catalytically important
Polyketides (PKs) are a chemically important class of com-
pounds with several beneficial pharmaceutical properties [1,2].
Natural PKs generated by type III polyketide synthases (PKSs)
have been associated with the slowing of the aging process in
model organisms [3,4], anti-inflammatory and anti-cancer prop-
erties [5e10] and have shown potential to ameliorate diabetes
and nervous system disorders related complications [11,12]. Type
III PKSs are found in several plants, bacteria and fungi [13e15].
They are homodimeric enzymes that catalyze iterative conden-
sation of repeating units to a CoA-tethered starter substrate
through a conserved Cys-His-Asn catalytic triad. Functionally
cavities: the substrate binding pocket, composed of the important
residues S133, Q192, T194, T197, S338; the CoA binding tunnel,
composed of L55, R58, L62; and the cyclization pocket, composed
of T132, M137, F215, I254, G256, F265, P375 (residues numbered
according to the Vitis vinifera stilbene synthase (VvSTS)). Altering
these residues results in diversity of preference for starter sub-
strates, number of extender substrates incorporated through
iterative condensations and mechanism of cyclization of the poly-
b
-keto intermediate formed through Claisen condensation, aldol
condensation or lactonization. Moreover, type III PKSs possess
unusually broad substrate promiscuity and can accept several
non-natural substrates to form novel non-natural PKs [2,16].
Apart from the naturally occurring type III PKSs, several intuitive
and structure-guided mutations have also been carried out to
alter their enzymatic activity, so as to expand the chemical
diversity of PKs [17,18]. Some of the novel non-natural PKs formed
through these processes has also been demonstrated to possess
important biological activities [16].
Abbreviations: PKS, polyketide synthase; PK, polyketide; Vv, Vitis vinifera; STS,
stilbene synthase; CHS, chalcone synthase; Wt, wild-type; HBN, hydrogen-bonding
network; BNY, bisnoryangonin; CATL, p-coumaroyltriacetic acid lactone.
* Corresponding author.
E-mail address: koffam@rpi.edu (M. Koffas).
Present address: Northwestern University, Evanston, Illinois, USA.
1
http://dx.doi.org/10.1016/j.biochi.2015.05.019
ꢀ ꢀ
ꢀ
0300-9084/© 2015 Elsevier B.V. and Societe Française de Biochimie et Biologie Moleculaire (SFBBM). All rights reserved.

N. Bhan et al. / Biochimie 115 (2015) 136e143
137
Fig. 1. General reaction catalyzed by wild type STS and CHS. STS and CHS react with their natural substrates, p-coumaroyl-CoA and malonyl-CoA to form primarily resveratrol and
naringenin chalcone respectively. Bisnoryangonin (BYN) and p-coumaroyltriaceticacid lactone are natural derailment products of the reaction.
Stilbene synthase (STS) belongs to the type III PKS super family
and is unique due to the presence of a hydrogen-bonding networks
(HBN) [19], which is absent in the closely related chalcone
synthase-like (CHS) enzymes (70e90% sequence similarity to STS).
Due to the presence of this HBN, STS-like type III PKSs cyclize their
polyketide intermediate through an aldol condensation reaction.
CHS-like type III PKSs, in contrast, cyclize their intermediate pri-
marily through a Claisen condensation (Fig. 1). While both STS and
CHS form a tetraketide intermediate with p-coumaroyl-CoA and
three molecules of malonyl-CoA, STS cyclizes the intermediate into
resveratrol through a C2eC7 aldol condensation in contrast to CHS,
which cyclizes the intermediate into naringenin chalcone through a
C1eC6 Claisen condensation. Both enzymes form bisnoryangonin
(BNY) and p-coumaroyltriacetic lactone (CTAL) as derailment
products during the reaction [20].
We attempted to exploit the novel thioesterase-like property of
STSs to further diversify the chemical space of PKs. Along these
lines we created 6 mutants of Vitis vinifera stilbene synthase
(VvSTS) and challenged these with non-natural substrates (Table 1).
2. Materials and methods
2.1. Site-directed mutagenesis
The plasmids expressing the his-tagged VvSTS and VvSTS
T197G mutant were obtained from our previous studies [21].
Table 1
Table summerizing the results obtained in this study. List of PKs formed by the mutants created in this study. Cavity 1 is size of the cycliztion
pocket calculated from CASTp. Cavity 2 is the size of the substrate binding pocket calculated from CASTp. The novel previously unreported PKs are
highlighed in bold. “e” denotes reactions not tested.
Enzyme
Cavity 1 (Å3)
Cavity 2 (Å3)
Propionyl-CoA
Myristoyl-CoA
Octanoyl-CoA
Methylmalonyl-CoA
Wt
L214I
T197A
T197G
T197IG256L
T197M
T197GG265L
721.4
705.7
743.1
1134.3a
612.4
635
265.2
291.4
293
1,2,3
1.2.3
2,8,9
1,14
1,2,8,9
e
4
e
5
e
6,7
e
10,11
4
e
e
e
5,12
15,16
e
6,13
6,7
6,7
7
265.2
265.2
e
1016.6a
e
5,12,16
e
a
The T197G and the T197GG256L mutations result in merging of the two cavities, we thus mention the size of the entire cavity in these cases.

138
N. Bhan et al. / Biochimie 115 (2015) 136e143
The rest of the VvSTS mutants were constructed for this study
with a QuickChange Site-Directed Mutagenesis Kit (Stratagene)
according to the manufacturer's protocol and using the primers in
Supplementary Table S1. Each construct was sequence analyzed to
confirm the point mutation. The double mutants were created
sequentially.
2.4. Substrates
4-Coumaroyl-CoA was chemically synthesized as previously
described [22]. 13C3-malonyl-CoA, malonyl-CoA, propionyl-CoA,
octanoyl-CoA, methylmalonyl-CoA, myristoyl-CoA were purchased
from Sigma.
2.5. XCMS workflow for analyzing online LCMS data
2.2. Protein expression and purification
The workflow established included creating the structure-
based mutants and carrying out the in vitro enzymatic re-
actions. The in vitro reactions were run with and without the
13C3-malonyl-CoA for easier identification of novel PK structure
from the HR-LC-MS/MS analysis. The reaction extracts were then
run on the LC-MS to obtain online HR-LCMS ( 5 ppm accuracy)
which was finally compared using XCMS to directly calculate the
fold change in production of PKs in the mutants compared to
the wild-type and to identify peaks unique to the mutant. Raw
data was converted to XML format using R processor (Script S1,
obtained from http://www.metabolomics.strath.ac.uk/showPage.
php?page¼processingscripts), and analyzed using XCMS online
[23]. XCMS was utilized for calculating fold changes in CTAL
and BNY and all other peaks common to the wild type and
mutant proteins (Fig. 6). All calculations were based on tripli-
cates. This workflow obviated thin-layered chromatography
(TLC) analysis of the reaction products and utilization of radio-
active substrates for identification of products unique to the
mutants.
After confirmation of the sequence, the plasmid was trans-
formed into Escherichia coli BL21* (DE3). The cells harboring the
plasmid were cultured to an OD600 of 0.6 in LB medium containing
chloramphinecol (30
-galactopyranoside (IPTG) (1.0 mM) was added to induce
m
g/ml) at 37 ꢀC. Subsequently, isopropylthio-
b-D
protein expression, and the cells were further cultured at 30 ꢀC for
4 h. All of the following procedures were performed at 4 ꢀC. E. coli
cells were harvested by centrifugation at 4000 ꢁ g and frozen
at ꢂ80 ꢀC until further processing. The cells were disrupted by
incubating with lysis buffer (50 mM NaH2PO4, 300 mM NaCl,
10 mM imidazole, pH 8.0), lysozyme (10 mg/ml) and 1 culture
volume of Benzonase® nuclease (3 units/ml) at 4 ꢀC for 30 min.
The lysate was then centrifuged at 12,000 ꢁ g for 30 min. The
supernatant was loaded onto a Ni-NTA spin column (Qiagen) pre-
equilibrated with lysis buffer. The column was then washed with
wash buffer (50 mM NaH2PO4, 300 mM NaCl, 20 mM imidazole,
pH 8.0). The protein was eluted from the column using elution
buffer (50 mM NaH2PO4, 300 mM NaCl, 300 mM imidazole, pH
8.0). The protein concentration was determined by the Bradford
method (Protein Assay, Bio-Rad) with bovine serum albumin as
the standard.
2.6. Homology modeling and cavity size analysis
The models of the Wt VvSTS and the all the mutants were
generated by the SWISS-MODEL package (http://expasy.ch/spdbv/
) provided by the Swiss-PDB-Viewer program [24] based on the
crystal structure of wild-type STS from Acharis hypogaea (PDB
code: 1Z1E). The model quality was assessed using PROCHECK
[25]. In the Ramachandran plot calculated for the model, most of
the amino acid residues were present in the energetically allowed
regions with only a few exceptions, primarily Gly residues that
can adopt phi/psi angles in all four quadrants. The cavity volume
was calculated by the program CASTP (http://cast.engr.uic.edu/
cast/) [26].
2.3. Enzymatic reaction
The reaction mixture contained starter substrate [4-coumaroyl-
CoA (54
methylmalonyl-CoA (54
substrate [malonyl-CoA (108
and the purified enzyme (20
(500 l, 100 mM, pH 7.0). The purified enzyme was added to the
reaction mixture last. Incubations were performed at 30 ꢀC for
30 min and were stopped by the addition of ethyl acetate (500 l)
m
M), octanoyl-CoA (54
M), or myristoyl-CoA (54
M) or 13C3-malonyl-CoA (108
g) in potassium phosphate buffer
m
M), propionyl-CoA (54
M)], extender
M)],
mM),
m
m
m
m
m
m
m
with 1% HCl. The extracted products (ethyl acetate extracts) were
then concentrated in a speed vacuum and re-suspended in ethyl
3. Results
acetate (10
running on a column. The products were separated by reverse-
phase HPLC (Agilent 1260) on Zorbax C18 column
(4.6 ꢁ 150 mm, 5 m, at a flow rate of 0.7 ml/min). Gradient elution
m
l), and centrifuged at 14000 rpm for 10 min before
3.1. Substrate promiscuity of wild-type VvSTS
a
Several non-natural substrates have been supplied to wild-type
Acharis hypogea STSs, resulting in the formation of mainly CATL and
BYN-type pyrones [27,28]. We first investigated the substrate pro-
miscuity of wild-type VvSTS (Wt VvSTS) with non-natural sub-
strates that have not been tested previously. Specifically, we
supplied Wt VvSTS with starter substrates of varying size:
propionyl-CoA, myristoyl-CoA, octanoyl-CoA and methylmalonyl-
CoA; malonyl-CoA was used primarily as the extender substrate
for all of the reactions (Fig. 2).
Malonyl-CoA and methylmalonyl-CoA or propionyl-CoA acted as
extender substrates in case of compounds 2, 3 & 7, as is clearly
indicated in the color-coding in Fig. 3. The Wt VvSTS accepted all of
these aliphatic compounds as substrates to afford non-natural PKs
(Fig. 3).
m
was performed with H2O and acetonitrile (ACN), both containing
0.2% trifluoroacetic acid: 0e7 min, 20% ACN; 7e15 min, linear
gradient from 20% to 60% ACN; 15e30 min, linear gradient from
60% to 70% ACN; 30e36 min, linear gradient from 70% to 30% ACN.
Three reactions (technical replicates) were pooled into ethyl ace-
tate (10 ml) for LC-MS analysis. Online HReESI-LCMS spectra were
measured with an Agilent Technologies HPLC 1200 series HPLC
coupled to a Thermo Scientific LTQ Orbitrap XLTM mass spec-
trometer fitted with an electrospray ionization (ESI) source. The ESI
capillary temperature and the capillary voltage were 320 ꢀC and
4.0 V, respectively. The tube lens offset was set at 20.0 V. All spectra
were obtained in the negative and positive mode, over a mass range
of m/z 150e700, and at a range of one scan every 0.2 s. The collision
gas was helium, and the relative collision energy scale was set at
30.0% (1.5 eV). Dependent MS/MS scans were acquired for the first
four most abundant parent ions.
Propionyl-CoA and malonyl-CoA resulted in the formation of 3
major products. Compound 1 with the parent ion peak [M þ FA-H]ꢂ
at m/z 241.0700, a resorcinol derivative. Compound 2 with parent
ion peak [MꢂH]ꢂ at m/z 263.1290 and 3 with m/z 323.1392, a

N. Bhan et al. / Biochimie 115 (2015) 136e143
139
Fig. 2. Substrates utilized in this study. Color-coded substrates; for subsequent ease of understanding of PK structures generated by the wild-type and the mutants enzymes.
Fig. 3. Wt VvSTS accepts all the tested non-natural substrates to form PKs. Non-natural PKs formed by Wt VvSTS when supplied with non-natural substrates like propionyl-CoA
(compounds 1e3), myristoyl-CoA (compound 4), octanoyl-CoA (compound 5), or methylmalonyl-CoA (compounds 6 & 7) as starter substrates, coupled with malonyl-CoA as the
extender substrate.
[M þ COOꢂ]ꢂ ion (Fig. 3, Supplementary Table S2 for MS details).
Myristoyl-CoA and malonyl-CoA formed carboxylic acid 4, which
displayed a parent ion peak [M þ FA-H]ꢂ at m/z 293.1756 (Fig. 3,
Supplementary Table S2). Octanoyl-CoA and malonyl-CoA resulted
in the formation of primarily a phlorophenone (5) with the parent
ion [MꢂH]ꢂ at m/z 209.1185 (Fig. 3, Supplementary Table S2).
Methylmalonyl-CoA and malonyl-CoA resulted in the formation of
two non-natural PKs, a pentanone 6 at m/z 263.1274 [M ꢂ H]ꢂ and a
pyranone 7 at m/z 307.1292 [M þ COOꢂ]ꢂ (Fig. 3, Supplementary
Table S2).

140
N. Bhan et al. / Biochimie 115 (2015) 136e143
3.2. Substrate promiscuity of structure-based mutants of VvSTS
cyclization pocket from 721.4 Å3 to 705.7 Å3 explains the absence of
formation of any longer PKs.
Next we tested the above substrates with the structure-based
mutants of VvSTS. As hypothesized, perturbation of residues in
the catalytically relevant pockets of VvSTS altered its substrate
promiscuity. We supplied relevant substrates to each mutant based
on predicted change in the properties of the catalytically relevant
pockets; so all substrates were not supplied to all mutants. The
mutants created led to the formation of novel, non-natural PKs that
have not been reported previously (Figs. 4 and 5).
The leucine at position 214 is important for the structure of the
cyclization pocket. Not surprisingly, replacing it with isoleucine
(L214I) did not significantly alter the product profile. Compared to
Wt VvSTS, mutant L214I led to a 14.4-fold, 27-fold, and 9.4-fold
increase in production of compounds 1, 2, and 3, respectively,
when propionyl-CoA was used as a starter substrate (Fig. 3,
Supplementary Fig. S1). This could be explained by the increase in
the starter substrate binding pocket from 265.2 Å3 for Wt VvSTS to
291.4 Å3 in the L214I mutant, making it easier for the mutant to
utilize propionyl-CoA as the substrate. The decrease in the
Upon increasing the size of the cyclization pocket from 721.4 Å3
to 743.1 Å3 and the substrate-binding pocket from 265.2 Å3 to
293 Å3 by replacing the tyrosine at position 197 with an alanine,
VvSTS T197A catalyzed formation of 5 new PKs unique to the
mutant. Propionyl-CoA and malonyl-CoA resulted in the formation
of two novel compounds, a benzoic acid with [MꢂH]ꢂ m/z of
279.0858 (8) and a phlorophenone with m/z 233.0811 [Mꢂ2H]ꢂ (9)
(Fig. 4). In both 8 & 9 propionyl-CoA was utilized as an extender
substrate as well. A 1.2-fold increase in the production of the
2263.1290 with m/z 331.2014 [M þ CH3COOꢂ]ꢂ was observed
(Supplementary Fig. S2). Upon supplying myristoyl-CoA and
malonyl-CoA as the starter substrate and the extender substrate,
respectively, two unique, novel phlorophenones were formed; 10
with a parent ion peak [MꢂH]ꢂ at 335.20, and 11 with a parent ion
peak of [M þ FA-H]ꢂ at 337.2016 (Fig. 4). Octanoyl-CoA and
malonyl-CoA were also utilized by T197A to form 5 but 30-fold
lower than the amount formed by the wild-type (Supplementary
Fig. S2). Instead, the novel phlorophenone 12 with m/z [MꢂH]ꢂ of
Fig. 4. Non-natural PKs (8e13) formed by mutant VvSTS. Novel PKs formed by the mutants when supplied with propionyl-CoA (compounds 8 and 9), myristoyl-CoA (compounds
10 and 11), octanoyl-CoA (compound 12) and methylmalonyl-CoA (compound 13) as starter substrates, coupled with malonyl-CoA primarily as the extender substrate.

N. Bhan et al. / Biochimie 115 (2015) 136e143
141
Next we created a double mutant, replacing the threonine at
position 197 to an isoleucine and the glycine at position 256 to a
leucine (T197IG256L). This resulted in a decrease of the cyclization
pocket from 721.4 Å3 to 612.4 Å3 without significant change in the
substrate binding pocket size. We did not carry out the in vitro
reaction with myristoyl-CoA or octanoyl-CoA as we did not expect
the mutant to form any novel PKs with these bulkier substrates.
Propionyl-CoA and malonyl-CoA were utilized as substrates by
T197IG256L to form 1 and 2 at 6-fold and 2-fold lower levels,
respectively, than the wild-type (Supplementary Fig. S4). Moreover,
compounds 8 and 9, formed by T197A, were also formed by
T197IG256L. The mutant did not form any new PKs with
methylmalonyl-CoA and malonyl-CoA as substrates.
Replacing the threonine at position 197 with the bulkier
methionine (T197M) resulted in a reduction of the substrate-
binding pocket from 721.4 Å3 to 635 Å3. When we supplied
methylmalonyl-CoA and malonyl-CoA as substrates, compound 7
was formed, however a 2.3-fold decrease in comparison to the Wt
VvSTS was observed (Supplementary Fig. S4).
Upon creating the double mutant by replacing threonine at
position 197 with a glycine and the glycine at position 256 with a
leucine (T197GG256L), the homology model again predicted the
merging of the cyclization and substrate binding pockets, except
this time with a decreased volume of 1016.6 Å3 as compared to
1029.2 Å3 for Wt VvSTS. Surprisingly supplementation with
propionyl-CoA and malonyl-CoA did not result in the formation of
any significant products. Octanoyl-CoA and malonyl-CoA resulted
in the formation of 12 and 5, and 16, however 15 was not formed in
this case. Both 12 and 15 are formed from the same 18-carbon in-
termediate. Compound 12 is a derailment product, thus it is a
pyrone, while formation of 15 requires cyclization by the enzyme.
The absence of 15 can be explained by the possible incapability of
the T197G256L mutant to successfully cyclize the longer PK inter-
mediate (18 carbons). This is further supported by the significant
increase in the amount of the derailment product 12 formed.
3.3. XCMS work-flow for analyzing the in vitro enzymatic reactions
Fig. 5. Non-natural PKs (14e16) formed by mutant VvSTS. Novel PKs formed by the
mutants created with methylmalonyl-CoA (compound 14) and octanoyl-CoA (com-
pounds 15 and 16) as starter substrates, coupled with malonyl-CoA as extender
substrate.
We utilized a previously established workflow [21] for quick
identification of non-natural PKs by the 6 mutants of VvSTS created
in this study (Fig. 6). Briefly, enzymatic reactions were analyzed by
high resolution LC-MS (HR-LCMS), and novel reaction products were
automatically tabulated by comparison against the product profile of
a control reaction using an online LC-MS data comparison analysis
software known as XCMS [23,29]. Structural elucidation of novel
reaction products was further aided by performing paired reactions,
where one reaction used unlabeled malonyl-CoA and the other uti-
lized stable-isotope labeled 13C3-malonyl-CoA extender units. Novel
parent ions were subjected to HR-LC-MS/MS, and comparison of
fragmentation patterns between paired reactions enabled identifi-
cation of novel products by tracking the additional mass conferred by
incorporation of 13C3-malonyl-CoA extender units.
335.1488 was formed as the major product (Fig. 4). Methylmalonyl-
CoA and malonyl-CoA as substrates resulted in the formation of the
novel isochromanone 13 with the parent ion peak of [MꢂH]ꢂ
279.0858 (Fig. 4), where methylmalonyl-CoA was utilized as both
extender and starter substrate. A 7.3-fold increase in the production
of 6 was observed (Supplementary Fig. S2). The increase in size of
both cyclization and substrate binding pocket of the mutant
accommodated the formation of longer PKs.
Mutating the threonine at position 197 to glycine (T197G), as
previously reported [21], results in the joining of the cyclization and
substrate binding pockets, with a total cavity size of 1134.3 Å3 as
compared to 1029.2 Å3 for Wt VvSTS. Propionyl-CoA and malonyl-
CoA formed novel PK 14 (Fig. 5) with a parent ion [M þ Cl]ꢂ m/z of
291.0994, again propionyl-CoA acted as both starter and extender
substrate. A 4.1 fold decrease in formation of 1 was observed
(Supplementary Fig. S3). Myristoyl-CoA and malonyl-CoA resulted
in the formation of 4 with a 1.1-fold decrease as compared to Wt
VvSTS (Supplementary Fig. S3). Octanoyl-CoA and malonyl-CoA
resulted in the formation of two new PKs 15 and 16 with parent
ion peaks [MꢂH]ꢂ of m/z 315.1261 and 273.1162 respectively
(Fig. 5). Methylmalonyl-CoA and malonyl-CoA resulted in a 6.4-fold
increase in formation of 7 and 7.1-fold increase in formation of 6 in
comparison to the Wt VvSTS (Supplementary Fig. S3).
4. Discussion
We have successfully diversified the PK space by creating
structure-guided mutants of VvSTS, and supplying it with non-
natural substrates. Specifically we altered the size of the substrate
binding pocket (T197), the cyclization pocket (L214) and both (T197
& G256).
The data obtained demonstrated that wild-type VvSTS could
utilize methylmalonyl-CoA, propionyl-CoA, octanoyl-CoA and
myristoyl-CoA as starter substrates coupled with malonyl-CoA
primarily as the extender substrate; of the 7 non-natural PKs
generated by wild-type VvSTS, 6 have not previously been reported.

142
N. Bhan et al. / Biochimie 115 (2015) 136e143
Fig. 6. XCMS based workflow for quick identification of novel PKs formed. (1) Create structure-based mutants. (2) Carry out in vitro reaction with and without 13C labeled
substrate. (3) Carry out LC-MS analysis. (4) XCMS analysis of online XCMS data. (5) MS/MS analysis of interesting peaks identified via XCMS analysis.
Several other possible combinations of substrates, with a range of
different sizes and polarities, could be tested, to afford a compre-
hensive expansion of the PK space, although testing all of these
possible combinations is beyond the scope of this work. Upon
feeding non-natural substrates to the different mutants we were
able to further diversify the PK space by generating 9 additional
non-natural PKs that have not been previously reported. All of the
PKs characterized in this study are unique to STS-like enzymes,
establishing STS as a candidate enzyme for future protein engi-
neering efforts and as a tool for generation of libraries of novel PKs
with potential therapeutic value. Finally, XCMS analysis was uti-
lized for quick identification of PKs that were formed only by the
mutants, and the established workflow aided in simple and rapid
identification of all 15 novel, non-natural PKs directly from the
online HR-LCMS data.
funding provided by the RPI Biocatalysis Constellation. The authors
acknowledge no conflict of interest.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.biochi.2015.05.019.
References
[1] C. Stewart Jr., C.R. Vickery, M.D. Burkart, J.P. Noel, Confluence of structural and
chemical biology: plant polyketide synthases as biocatalysts for a bio-based
future, Curr. Opin. Plant Biol. 16 (2013) 365e372.
[2] I. Abe, Novel applications of plant polyketide synthases, Curr. Opin. Chem.
Biol. 16 (2012) 179e185.
We propose that this quick identification workflow can be uti-
lized for generation of many more novel PKs. In the future, sup-
plementation with nitrogen-containing substrates could help
generate novel non-natural alkaloid analogs. Further optimization
of the biosynthetic production of these novel PKs can be carried
out, as has been previously reported [30e35].
[3] B. Agarwal, J.A. Baur, Resveratrol and life extension, Resveratrol Health 1215
(2011) 138e143.
[4] Q. Xu, L.-Y. Si, Resveratrol role in cardiovascular and metabolic health and
potential mechanisms of action, Nutr. Res. 32 (2012) 648e658.
[5] N. Jang, J. Cai, G.O. Udeani, K.V. Slowing, C.F. Thomas, C.W. Beecher, H.H. Fong,
N.R. Farnsworth, A.D. Kinghorn, R.G. Mehta, R.C. Moon, J.M. Pezzuto, Cancer
chemopreventive activity of resveratrol,
a natural product derived from
grapes. Science 275 (1997) 218e220.
€
[6] M. Bohm, S. Rosenkranz, U. Laufs, Alcohol and red wine: impact on cardio-
vascular risk, Nephrol. Dial. Transpl. 19 (2004) 11e16.
Conflict of interest
[7] G. Chen, W. Shan, Y. Wu, L. Ren, J. Dong, Z. Ji, Synthesis and anti-inflammatory
activity of resveratrol analogs, Chem. Pharm. Bull. (Tokyo) 53 (2005)
1587e1590.
[8] M.A. Birrell, K. McCluskie, S. Wong, L.E. Donnelly, P.J. Barnes, M.G. Belvisi,
Resveratrol, an extract of red wine, inhibits lipopolysaccharide induced
airway neutrophilia and inflammatory mediators through an NF-kappaB-
independent mechanism. FASEB 19 (2005) 840e841.
None.
Acknowledgments
[9] S.L. Wu, C.E. Pan, L. Yu, K.W. Meng, Immunosuppression by combined use of
cyclosporine and resveratrol in a rat liver transplantation model. Transpl.
Proceed. 37 (2005) 2354e2359.
The authors would like to thank Dr. Dmitri Zagorevski for help
with LC-MS analysis. This work was partially supported using

N. Bhan et al. / Biochimie 115 (2015) 136e143
143
[10] S.L. Wu, L. Yu, K.W. Meng, Z.H. Ma, C.E. Pan, Resveratrol prolongs allograft
survival after liver transplantation in rats. World J. Gastroenterol. 11 (2005)
4745e4749.
[23] R. Tautenhahn, G.J. Patti, D. Rinehart, G. Siuzdak, XCMS online: a web-based
platform to process untargeted metabolomic data,, Anal. Chem. 84 (2012)
5035e5039.
[11] D. Vauzour, Dietary polyphenols as modulators of brain functions: biological
actions and molecular mechanisms underpinning their beneficial effects,
Oxid. Med. Cell. Longev. 2012 (2012), 914273e914273.
[12] P. Maher, R. Dargusch, J.L. Ehren, S. Okada, K. Sharma, D. Schubert, Fisetin
lowers methylglyoxal dependent protein glycation and limits the complica-
tions of diabetes, Plos One 6 (2011).
[13] N. Funa, Y. Ohnishi, I. Fujii, M. Shibuya, Y. Ebizuka, S. Horinouchi, A new
pathway for polyketide synthesis in microorganisms,, Nature 400 (1999)
897e899.
[14] Y. Katsuyama, Y. Ohnishi, Type III polyketide synthases in microorganisms, in:
D.A. Hopwood (Ed.), Nat. Prod. Biosynth. Microorg. Plant, Pt A, Elsevier Aca-
demic Press Inc, San Diego, 2012, pp. 359e377.
[15] D. Yu, F. Xu, J. Zeng, J. Zhan, Type III polyketide synthases in natural product
biosynthesis, Iubmb Life 64 (2012) 285e295.
[16] H. Morita, M. Yamashita, S.-P. Shi, T. Wakimoto, S. Kondo, R. Kato, S. Sugio,
T. Kohno, I. Abe, Synthesis of unnatural alkaloid scaffolds by exploiting
plant polyketide synthase, Proc. Natl. Acad. Sci. U. S. A. 108 (2011)
13504e13509.
[24] N. Guex, M.C. Peitsch, SWISS-MODEL and the Swiss-PdbViewer: an environ-
ment for comparative protein modeling, Electrophor. 18 (1997) 2714e2723.
[25] R.A. Laskowski, J.C. Rullmann, M.W. MacArthur, R. Kaptein, J.M. Thornton,
AQUA and PROCHECK-NMR: programs for checking the quality of protein
structures solved by NMR, J. Biomol. NMR 8 (1998) 477e486.
[26] J. Liang, H. Edelsbrunner, C. Woodward, Anatomy of protein pockets and
cavities: measurement of binding site geometry and implications for ligand
design, Prot. Sci. 7 (1998) 1884e1897.
[27] H. Morita, H. Noguchi, J. Schroder, I. Abe, Novel polyketides synthesized with a
higher plant stilbene synthase, Euro, J. Biochem. 268 (2001) 3759e3766.
[28] I. Abe, T. Watanabe, H. Noguchi, Enzymatic formation of long-chain polyketid
pyrones by plant type III polyketide synthases, Phytochem. 65 (2004)
2447e2453.
[29] C.A. Smith, E.J. Want, G. O'Maille, R. Abagyan, G. Siuzdak, XCMS: processing
mass spectrometry data for metabolite profiling using nonlinear peak align-
ment, matching, and identification, Anal. Chem. 78 (2006) 779e787.
[30] C.G. Lim, Z.L. Fowler, T. Hueller, S. Schaffer, M.A. Koffas, High-yield resveratrol
production in engineered Escherichia coli, Appl. Environ. Microbiol. 77 (2011)
3451e3460.
[17] I. Abe, Engineering of plant polyketide biosynthesis, Chem. Pharm. Bull. 56
(2008) 1505e1514.
[31] N. Bhan, P. Xu, O. Khalidi, M.A.G. Koffas, Redirecting carbon flux into malonyl-
CoA to improve resveratrol titers: proof of concept for genetic interventions
predicted by OptForce computational framework, Chem. Eng. Sci. 105 (2012).
[32] N. Bhan, P. Xu, M.A.G. Koffas, Pathway and protein engineering approaches to
produce novel and commodity small molecules, Curr. Opin. Biotech. 24 (2013)
1137e1143.
[18] I. Abe, Engineered biosynthesis of plant polyketides: structure-based and
precursor-directed approach, Top. Curr. Chem. 297 (2009) 45e66.
[19] M.B. Austin, M.E. Bowman, J.L. Ferrer, J. Schroder, J.P. Noel, An aldol switch
discovered in stilbene synthases mediates cyclization specificity of type III
polyketide synthases, Chem. Biol. 11 (2004) 1179e1194.
[20] G. Hrazdina, F. Kreuzaler, K. Hahlbrock, H. Grisebach, Substrate specificity of
flavanone synthase from cell suspension cultures of parsley and structure of
release products in vitro, Arch. Biochem. Biophys. 175 (1976) 392e399.
[21] N. Bhan, L. Lin, C. Cai, P. Xu, R.J. Linhardt, M.A.G. Koffas, Enzymatic formation
of a resorcylic acid by creating a structure-guided single-point mutation in
stilbene synthase, Prot. Sci. 24 (2014) 167e173.
€
[33] B.F. Cress, O.D. Toparlak, S. Guleria, M. Lebovich, J.T. Stieglitz, J.A. Englaender,
J.A. Jones, R.J. Linhardt, M.A.G. Koffas, CRISPathBrick: modular combinatorial
assembly of type II-a CRISPR arrays for dCas9-mediated multiplex transcrip-
tional repression in E. coli. ACS Synth. Biol. (2015) [Epub ahead of print].
[34] S. Zhao, J.A. Jones, D.M. Lachance, N. Bhan, O. Khalidi, S. Venkataraman,
Z. Wang, M.A.G. Koffas, Improvement of catechin production in Escherichia coli
through combinatorial metabolic engineering, Metab. Eng. 28 (2015) 43e53.
[35] P. Xu, N. Bhan, M.A.G. Koffas, Engineering plant metabolism into microbes: from
systems biology to synthetic biology, Curr. Opin. Biotech. 24 (2013) 291e299.
[22] J. Stockigt, M.H. Zenk, Chemical synthesis and properties of hydrox-
ycinnamoyl coenzyme
A deriatives. Zeitschrift Fur Naturforschung C-a J.
Biosci. 30 (1975) 352e358.