Subscriber access provided by READING UNIV
Article
Characterization of the polyspecific transferase of murine type I fatty acid synthase (FAS) and implications for polyketide synthase (PKS) engineering Alexander Rittner, Karthik S. Paithankar, Khanh Vu Huu, and Martin Grininger ACS Chem. Biol., Just Accepted Manuscript • DOI: 10.1021/acschembio.7b00718 • Publication Date (Web): 12 Jan 2018 Downloaded from http://pubs.acs.org on January 13, 2018
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Chemical Biology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Biology
This study characterizes the transferase domain (MAT) of the murine fatty acid synthase (FAS) and explains its substrate promiscuity on a molecular basis. By engineering FAS based constructs, among other generating bimodular constructs as shown in the Figure, it is demonstrated that properties of MAT can be harnessed for polyketide synthase (PKS) engineering. 39x28mm (300 x 300 DPI)
ACS Paragon Plus Environment
ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 34
2
Characterization of the polyspecific transferase of murine type I fatty acid synthase (FAS) and implications for
3
polyketide synthase (PKS) engineering
1
4 5
Alexander Rittner, Karthik S. Paithankar, Khanh Vu Huu and Martin
6
Grininger*
7 8
Author affiliation
9 10 11 12 13
Institute of Organic Chemistry and Chemical Biology, Buchmann Institute for Molecular Life Sciences, Cluster of Excellence for Macromolecular Complexes, Goethe University Frankfurt, Max-von-Laue-Str. 15, 60438 Frankfurt am Main, Germany.
14
Correspondence
15
*
[email protected] 16 17
Abstract
18
Fatty acid synthases (FASs) and polyketide synthases (PKSs) condense acyl
19
compounds to fatty acids and polyketides, respectively. Both, FASs and PKSs,
20
harbor acyltransferases (ATs), which select substrates for condensation by β-
21
ketoacyl synthases (KSs). Here, we present the structural and functional
22
characterization of the polyspecific malonyl/acetyltransferase (MAT) of murine FAS.
23
We assign kinetic constants for the transacylation of the native substrates, acetyl-
24
and malonyl-CoA, and demonstrate the promiscuity of FAS to accept structurally and
25
chemically diverse CoA-esters. X-ray structural data of the KS-MAT didomain in a
26
malonyl-loaded state suggests a MAT-specific role of an active site arginine in
27
transacylation. Owing to its enzymatic properties and its accessibility as a separate
28
domain, MAT of murine FAS may serve as versatile tool for engineering PKSs to
29
provide custom-tailored access to new polyketides that can be applied in antibiotic
30
and antineoplastic therapy.
31
ACS Paragon Plus Environment
1
Page 3 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Biology
32
Introduction
33
Fatty acids are key compounds of cell membranes, metabolism and signaling,
34
and are obtained either directly from the diet, or synthesized de novo in a repeating
35
cyclic reaction catalyzed by fatty acid synthases (FASs). The chemistry of fatty acid
36
synthesis is largely conserved from bacteria to eukaryotes. Acetyl- and malonyl-CoA
37
compounds are first loaded by acyltransferases (ATs) and then condensed by a
38
β-ketoacyl synthase (KS) in a decarboxylative Claisen condensation. Subsequently,
39
the β-ketoester is modified by three active sites in the ketoreductase (KR),
40
dehydratase (DH) and enoylreductase (ER) domains. A first cycle delivers a fully
41
reduced ACP-bound butyryl group, which is elongated with a malonyl moiety in
42
further cycles to a select chain length (typically C16). During this multi-step catalytic
43
process, the acyl carrier protein (ACP) domain shuttles the fatty acid cargo between
44
the different enzymatic domains (Figure 1A).
45
The enzymatic functions of fatty acid synthesis are contained in remarkably
46
different structural frames. In general, type II FAS systems comprised of separate
47
enzymes are found mostly in plants, bacteria and in mitochondria. In case of
48
eukaryotes, cytosolic de novo fatty acid synthesis is performed by large multidomain
49
type I FAS.1, 2 For example, fungal FASs exhibit large barrel-shaped complexes of
50
about 2.6 MDa, in which the synthesis is enclosed in reaction chambers. In contrast,
51
animal FASs show an open, X-shaped architecture of about 540 kDa.3-7
52
The synthetic concept of fatty acid synthesis is exploited in polyketide
53
synthases (PKSs) for the production of a large set of bioactive compounds.8 Similarly
54
to FASs, PKSs do occur in different structural arrangements with type II and III
55
consisting of individual proteins. However, type I PKSs are contained within
56
multidomain polypeptides creating complex assembly lines of up to several MDa in
57
size.9 Structural information, recently collected on PKS modules, illustrates structural
58
similarity of PKSs and animal FASs.10-13 Contrary to FASs, PKSs vary the degree of
ACS Paragon Plus Environment
2
ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 34
59
β-carbon processing from module to module and utilize a broad set of acyl-CoA
60
priming and elongation compounds, which leads to a gigantic structural and chemical
61
versatility of polyketide natural compounds.
62
Animal FASs employ an AT domain with the dual function of priming fatty acid
63
synthesis with acetyl-CoA and elongating fatty acid chains with malonyl-CoA. The
64
domain is termed malonyl/acetyltransferase (MAT) (see Figure 1A).14 Earlier studies
65
revealed that MAT also transfers unusual CoA-esters, as for example priming
66
substrates like phenylacetyl-CoA,15 and extender substrates like methylmalonyl-
67
CoA.16, 17 The substrate promiscuity of MAT is not apparent in the narrow product
68
spectrum of mainly C16 fatty acids of animal FAS. This may not surprise, as previous
69
reports have shown that product specificity in fatty acid synthesis is determined by
70
multiple factors, including substrate availability and the substrate specificity of the KS
71
domain.18, 19
72
ATs of modular PKSs have a more diversified function as MAT of animal
73
FAS. As part of each elongation module, ATs are responsible for the selection of the
74
appropriate acyl-substrate in each elongation cycle, which is typically a malonyl- or a
75
methylmalonyl-CoA. As part of loading modules, ATs prime the synthesis of
76
polyketides. Both of these transferases are primary targets for engineering, leading
77
to the regiospecific modification of polyketides. The impact of the transferase
78
specificity on the PKS product spectra has been demonstrates on the loading-AT of
79
the avermectin PKS and the elongation-AT of the monensin PKS (module 5).
80
Insertion of these domains in heterologous PKS systems as well as the modulation of
81
these domains in their substrate specificity can induce the production of novel
82
compounds.20-25 Despite its occurrence in animal FAS only, MAT can be an attractive
83
target for custom polyketide synthesis, when its polyspecificity is harnessed for PKS
84
engineering. To this end, we recently started to characterize structure and function of
85
MAT. We aimed at revealing the molecular basis of MAT substrate polyspecificity to
86
understand MAT substrate promiscuity, as well as at making MAT available for PKS
ACS Paragon Plus Environment
3
Page 5 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Biology
87
engineering.
88
Here, we report the in vitro characterization of the MAT domain of murine
89
FAS. To the best of our knowledge, this is the first time an extensive enzyme kinetic
90
analysis has been performed on MAT with both native substrates and multiple non-
91
canonical CoA-esters. Utilizing a continuous enzyme-coupled assay, this study
92
provides a profound quantitative description of the polyspecificity in MAT-mediated
93
transacylation.15, 17, 26, 27 Kinetic analysis further describes the distinctive features of
94
remarkably low hydrolysis rates and fast transfer rates. We further report the 2.9 Å
95
resolution structure of the KS-MAT didomain trapped in its malonyl-loaded state,
96
which helps to reveal the molecular foundation for the polyspecific character of
97
transacylation. Moreover, in a rational engineering approach, we express and purify
98
MAT as stand-alone domain in Escherichia coli and demonstrate its functionality. We
99
express and purify FAS-based bimodular constructs, which mimic loading-module–
100
module 1 PKS assemblies, to underline the direct usability of MAT in PKS
101
engineering. Murine MAT is a valuable new tool in the toolkit for PKS engineering
102
that may enlarge the spectrum of accessibly molecules in tailored polyketide
103
synthetic pathways.
104
ACS Paragon Plus Environment
4
ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
105
Page 6 of 34
RESULTS AND DISCUSSION
106
Expression and purification of the KS-MAT didomain and ACP. Murine
107
FAS was chosen after extensive screening of multiple human and animal FAS
108
constructs in recombinant expressions, as it gave soluble and relatively large
109
quantities of protein (Figure S1). E. coli expression yielded about 2 mg of purified
110
protein after tandem affinity purification from 1 L of culture. Comprehensive enzyme
111
kinetic analysis of MAT and structural data were collected on the didomain
112
subconstruct KS-MAT (2–853), bearing a KS knockout (C161G) when used for
113
functional studies. Both subconstructs as well as constructs extending the murine
114
FAS fold have finally been characterized. Murine FAS (UniProt accession code
115
P19096) is highly homologous to human, porcine and rat FAS (with a sequence
116
identity of 81 %, 78 % and 94 %, respectively to murine FAS when aligned using the
117
program BLAST).28
118
Using recombinant expression in E. coli, ACP was not post-translationally
119
modified and remained in the apo state. For full activation, different co-expression
120
strategies with the 4'-phosphopantetheinyl transferase from Bacillus subtilis (Sfp)
121
were tested. As observed in this study, only bicistronic co-expression reproducibly
122
delivered protein at high yield and quality. For construct design and expression yields
123
in E. coli, see Supporting Information Note S1 and Figure S2. For more information
124
on protein preparation, see Experimental Procedures.
125
Characterization of transacylation polyspecificity. MAT catalyzes the acyl
126
transfer via a ping-pong bi-bi mechanism with an acylated serine being the central
127
intermediate of the reaction (Figure 1B).29 All kinetic investigations of MAT were
128
performed with an enzyme-coupled, continuous assay using α-ketoglutarate
129
dehydrogenase (αKGDH) to couple the release of free coenzyme A (CoA) to the
130
reduction of nicotinamide adenine dinucleotide (NAD+), which was monitored
131
fluorometrically (Figure S3).26, 27 In the established assay we ensured that initial rates
ACS Paragon Plus Environment
5
Page 7 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Biology
132
of transferase reactions indeed depend on MAT activity and not on the coupled
133
downstream reaction. The mutant protein S581A was inactive, confirming the
134
specificity of MAT-mediated transacylation. The assay was further able to reproduce
135
the phenomenon of a compromised transfer of malonyl-CoA when the active site
136
arginine R606 is mutated to alanine (Figure S3B and Table 1, for more information
137
see also Experimental Procedures).30
138
In the next step, we determined MAT-mediated substrate hydrolysis rates to
139
enable appropriate background subtraction for transacylation (see Figure 1B).
140
Hydrolysis was slow with turnover rates of 9.8 ± 1.7 × 10–3 s–1 and 9.3 ± 0.8 × 10–3 s–1
141
for malonyl- and acetyl-CoA, respectively. These values agree well with previously
142
reported data for rat FAS (4.1 ± 0.8 nmol min–1 mg–1 protein (3.2 × 10–3 s–1)).30
143
Transacylation was about 3-4 magnitudes faster than hydrolysis, and allowed
144
ignoring MAT-mediated hydrolysis for background correction. Depending on the acyl-
145
CoA, significant non-enzymatically catalyzed hydrolysis or self-acylation of murine
146
ACP was observed,31 which required background corrections (see Experimental
147
procedure).
148
We selected 12 acyl-CoA substrates for analyzing the specificity of
149
transacylation, by determining the apparent kcat/KM values (Table 1). Substrates were
150
tested in a concentration range of 0.1 to 100 µM at fixed ACP concentrations. While
151
being most specific for malonyl-CoA, MAT was capable of transferring all tested
152
CoA-esters, except palmitoyl-CoA, with specificity constants in the same order of
153
magnitude. The high efficiency of MAT in the transacylation of a set of chemically
154
and structurally diverse compounds illustrates that MAT is not just tolerant in its
155
transacylation reaction, but indeed polyspecific (Table 1 and Figure S4). To ensure
156
that rates for non-native CoA-esters originate from transacylation and not from a
157
possible increased hydrolysis rate, the MAT-mediated hydrolysis rate was also
158
determined for every substrate. Although many non-native CoA-esters, for example,
ACS Paragon Plus Environment
6
ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 34
159
crotonyl- and succinyl-CoA, were hydrolyzed up to three times faster than acetyl- or
160
malonyl-CoA, the apparent kcat of the transacylation reaction remained 102-103 orders
161
of magnitude faster. Hence, we can rule out a “proof reading” function of MAT by
162
hydrolysis. The differences in hydrolysis rates rather imply slightly different transition
163
states for non-native substrates, which are less protected from hydrolysis. In this
164
aspect, MAT may differ from special ATs from PKSs, where hydrolysis was
165
suggested to determine specificity 32, 33
166
This set of acyl-CoA substrates was carefully chosen both for experimentally
167
exploring MAT transacylation promiscuity and for their commercial availability.
168
However, several of the tested compounds are also physiologically relevant, and
169
such data can provide interesting insight into animal fatty acid synthesis. (This is
170
outlined in Supporting Information Note S2).
171
Crystal structure of KS-MAT with bound malonyl moiety. The purified
172
murine KS-MAT didomain was crystallized, and crystals were soaked with malonyl-
173
CoA. X-ray diffraction data were collected to a resolution of 2.9 Å, and the resulting
174
structural model refined to R/Rfree of 0.18/0.22 (Table S1). The asymmetric unit
175
contains four molecules (referred to chains A, B, C and D), arranged as two
176
biological dimers wedged in the cleft between the KS and the linker domain. The
177
individual KS-MAT dimers are observed in a conformation that superimposes well
178
with the KS-MAT architectures of the natively purified porcine FAS,34 and the
179
corresponding didomain structure of human FAS (average r.m.s.d (Cα) of biological
180
dimers to porcine FAS of 1.6 and 1.7 Å; human KS-MAT superimposes to porcine
181
FAS with an average r.m.s.d 1.3 and 1.4 Å) (Figure 2A).35, 36 The four molecules in
182
the asymmetric unit show overall similar structure and superimpose with an r.m.s.d
183
(Cα) of 1, 0.8, and 0.9 Å (for chains A, B and C vs. chain D, respectively).
184
As observed in KS-MAT structures before,34, 35 the subdomains undergo rigid-
185
body movement (Figure 2B). A KS domain-based superposition (average r.m.s.d
ACS Paragon Plus Environment
7
Page 9 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Biology
186
(Cα) 0.1 Å over residue range 1–407) indicates the variability in the relative
187
positioning of KS vs. MAT. A superposition of Cα atoms of the MAT α/β-hydrolase
188
subdomain (MAT without the ferredoxin-like fold, average r.m.s.d (Cα) is less than
189
0.4 Å over residue ranges 488–615 and 685–806) shows pronounced conformational
190
flexibility for the ferredoxin-like fold, represented by a rotational movement of the
191
subdomain of 10.6, 6.3 and 10.9 degrees (values given for chains A, B and C vs.
192
chain D). The MAT entry tunnel and the active site are comprised by residues of the
193
α/β-hydrolase and the ferredoxin-like subdomain (see Figure 2B).34, 37
194
Although the four molecules in the asymmetric unit superimpose well onto
195
one another, we found striking differences in substrate binding. Upon soaking KS-
196
MAT crystals with malonyl-CoA, the malonyl group was transferred onto the MAT
197
domain resulting in a covalent malonyl-enzyme complex in two of the four molecules
198
(chain A and D). Malonyl loading confirms MAT activity in the crystallized
199
conformation. In chain C, one molecule malonyl-CoA is non-covalently bound at the
200
entrance of the active site tunnel, whereas no substrate density could be detected in
201
chain B (Figure 2C).
202
Mechanistic insight into MAT-mediated substrate transfer. The active site
203
is characterized by a narrow shape and a hydrophobic environment (Figure 3A). The
204
active site serine (S581) is positioned in a strand-turn-helix motif (identified by the
205
consensus sequence GHSXG), termed the ‘nucleophile elbow’,38 where its
206
nucleophilicity is enhanced by a helix dipole moment.39 The tightness of the strand-
207
turn-helix motif induces the nucleophilic residue to adopt energetically unfavorable
208
main chain torsion angles, and imposes steric restrictions on residues located in its
209
proximity. This is very important for transition state stabilization by the oxyanion
210
hole.38 The catalytic dyad is accomplished by the presence of H683 at short distance,
211
as the general acid/base catalyst of the transferase reaction. For catalysis, Nπ of the
212
imidazole ring accepts the S581 hydroxyl group proton, whereas Nτ is in H-bond
ACS Paragon Plus Environment
8
ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 34
213
distance to the backbone carbonyl groups of residue N738, L739 and S741. After
214
nucleophilic attack of S581 at the thioester carbonyl group a tetrahedral intermediate
215
state is formed. The X-ray structural data on MAT in its malonyl-loaded state allows
216
localizing the oxyanion hole that stabilizes the transition state by charge-dipol
217
interactions. Here, the carbonyl oxygen is in H-bond distance to the backbone amide
218
of M499 in chain A and D (2.9 Å).37 The distance to the backbone amide of L582 is
219
3.6 Å in both malonyl-loaded active sites (chains A and D), which exceeds a typical
220
H-bond length. This may indicate that in MAT the trapped acyl-enzyme
221
conformational state has already moved away from the transition state conformation
222
and the oxyanion hole is disassembling. In the E. coli malonyl transferase FabD,
223
structural data revealed a slightly different embedment of malonyl in the oxyanion
224
hole (carbonyl oxygen at distances of 3.0 and 3.1 Å; distance of nitrogen atoms
225
5.2 Å) (Figure S5A).40
226
The active site arginine R606 is located at the bottom of the active site tunnel.
227
Earlier data have identified R606 to be important in malonyl transfer,41 as interacting
228
with free carboxyl group of the malonyl moiety.33,
229
densities of the non-substrate-bound active sites (chain B and C) are less well
230
defined in this region, R606 is observed pointing towards S581. Such an orientation
231
has also been observed in the unloaded active site of human KS-MAT, which was
232
solved at a higher resolution of 2.15 Å (Figure 3B and S5B).35 In the malonyl-loaded
233
active sites (chains A and D), the guanidinium group of R606 is in bidendate
234
interaction with the carboxyl of the malonyl moiety. Contrary to structural data on the
235
malonyl transferases FabD,40 and DynE8,33 where an end-on coordination is
236
observed, the active site arginine R606 in murine KS-MAT forms a side-on bidentate
237
interaction (Figure 3C). In addition, R606 of MAT is embedded in an H-bond network
238
that is less extended than observed in FabD and DynE. A strongly positionally
239
conserved phenylalanine (F553, murine FAS numbering), which replaces glutamine
240
of FabD and DynE, is largely responsible for the reduced H-bonding of R606
40
Although the overall electron
ACS Paragon Plus Environment
9
Page 11 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Biology
241
(Figure 3C).
242
We assume that a positionally rather variable guanidinium group of R606 is
243
able to coordinate polar groups and may even swing out for the accommodation of
244
bulky residues (see Figure S5C). Herein, the side-on coordination of the malonyl
245
moiety by the guanidinium group may well reflect a unique active site geometry of
246
MAT that provides the necessary conformational variability of R606.
247
Transacylation of malonyl vs. acetyl moieties. To aid in interpreting
248
structural data, we expanded the enzyme kinetic analysis of both substrates.
249
Particularly, we were interested in refining the understanding of R606 in
250
transacylation of malonyl vs. acetyl groups. The assay in hand allowed addressing
251
absolute steady-state kinetic parameters by double reciprocal fits. We determined
252
apparent Km and kcat constants for malonyl- and acetyl-CoA at five different ACP
253
concentrations by fitting the raw data with the Hanes-Woolf equation (see Figure S4).
254
Individual reciprocal plotting of these constants against the reciprocal ACP
255
concentration resulted in the absolute Km of 8.8 µM and the absolute kcat of 119 s–1
256
for malonyl-CoA, and 12 µM (Km) and 99.2 s–1 (kcat) for acetyl-CoA (Figure 4). The
257
difference in transition state energies between acetyl and malonyl transfer ∆∆GES‡
258
can be calculated to 1.1 kJ mol–1. Molecular dynamics (MD) simulation recently
259
yielded
260
∆∆GES‡ = +14.2 ± 1.1 kJ mol–1 for the mutation of arginine to lysine in the
261
malonyl/palmitoyltransferase (MPT) of the yeast FAS I.42 To gain further insight into
262
the role of R606, we determined apparent kinetic constants for the R606A mutant at
263
a fixed ACP concentration (Table 1). Whereas hydrolysis rates remained nearly
264
unaltered, the specificity constant for malonyl-CoA decreased significantly being forty
265
fold smaller than the wildtype constant (kcat/Km of 0.3 ± 0.1 M–1 s–1). This corresponds
266
to
267
(2.3 kcal mol-1). Intriguingly, enzyme kinetic data did disclose a lower impact of
268
carboxyl coordination in malonyl binding and transacylation of murine FAS MAT, than
a
an
difference
estimated
in
transition
weakened
state
free
malonyl-CoA
energy
ACS Paragon Plus Environment
of
binding
of
∆∆GES‡ = 9.2 kJ mol–1
10
ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 12 of 34
269
estimated from MD simulations on the yeast FAS MPT domain and visual inspection
270
of structural data on malonyl-loaded type II AT (see Figure 3C).33, 40 Data indicate
271
that transacylation of the malonyl vs. the acetyl moiety does not differ in the plain
272
contribution of binding energy by bidentate coordination of the free carboxyl group of
273
the malonyl moiety, but includes a more global reshaping of the active site. A
274
possible explanation for the role of R606 may lie in the compensation for the
275
desolvation of substrates in the hydrophobic environment of the active site (see
276
Figure 3A).43 Such a role may be embedded in a molecular mode of substrate
277
accommodation that is based on the clamping of acyl-CoA compounds between the
278
positionally variable α/β-hydrolase and ferredoxin-like fold rather than their diffusion
279
via a substrate binding channel. The different occupancy of the active sites, observed
280
in the four molecules of the asymmetric unit (see Figure 2C), could then be
281
interpreted by the sensitivity of substrate accommodation towards the relative
282
positioning of the α/β-hydrolase and ferredoxin-like fold (see Figure 2B).
283
Engineering of MAT and murine FAS. To make MAT transacylation
284
properties available for PKS engineering, we assessed whether murine FAS MAT
285
can be used as a discrete, KS-independent (stand-alone) domain. As such, it may be
286
employed in trans-AT systems,44 or embedded in AT-ACP didomain structures of
287
loading-modules. Guided by our structural data and by comparison with homologous
288
proteins, two constructs were engineered, in which the MAT domain is liberated from
289
its structural embedment in the rigid KS-MAT fold. A LD-MAT construct was
290
designed that contained the MAT-domain and the linker domain (LD). It resembles a
291
construct of the human FAS reported in a previous study (Figure 5A).37 Another MAT
292
construct (486–808) was designed by mainly comparing structural data on the
293
avermectin synthase loading-AT (see Figure 5A; for construct design see
294
Figure S6).21 While MAT expressed at low concentrations, a fusion construct with a
295
maltose binding protein (MBP) attached as domain to the C-terminal helix
ACS Paragon Plus Environment
11
Page 13 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Biology
296
significantly increased yields.
297
As a next step, we constructed two bimodular systems that resemble loading-
298
module–module1 assemblies as found in modular PKS. With these constructs, we
299
sought to collect initial information whether such bimodular fusions can, in principle,
300
deliver intact proteins. As a “module1”, we worked with full-length murine FAS. Both
301
the LD-MAT and the MAT domain, the latter without the MBP fusion, were N-
302
terminally fused to full-length FAS. Additionally, the ACP domain was integrated in
303
both constructs, as found in loading-modules (see Figure 5A and Figure S7). We
304
would like to note that such bi- and multimodular FAS constructs may be interesting
305
for FAS-mediated compound synthesis. They outline the perspective of FAS-
306
mediated custom synthesis of small molecules.42, 45, 46
307
The structural integrity of truncated and bimodular constructs was checked by
308
SDS-PAGE, phosphopantetheinylation with fluorescent CoA (FAS and bimodular
309
constructs),
310
(Figure 5B-G; see also Supporting Information Note S1 and Figure S8). Although
311
fusion of loading didomains did not reduce the general protein quality of both
312
bimodular constructs, the overall activity of the inherent FAS fold was decreased
313
compared to the wildtype FAS as judged by NADPH consumption. To evaluate
314
whether the additional loading module interferes structurally or catalytically with fatty
315
acid synthesis, we tested three controls of the LD-MAT-ACP-FAS construct providing
316
either loading module domain MAT or ACP or both as functional knockouts
317
(Figure S8). All three controls regained activity of wildtype FAS, as read out by
318
NADPH consumption rates. These data are interesting, since they imply the
319
structural integrity of module 1 (the integral FAS fold) in the bimodular constructs.
320
The comprised activity of the non-mutated wildtype bimodular construct (no knock-
321
out in MAT and ACP domains) further demonstrates a functional communication of
322
the loading module with the integral FAS fold.
323
size
exclusion
chromatography
(SEC)
and
enzymatic
activity
Implications for PKS design. Murine FAS MAT is relevant for PKS
ACS Paragon Plus Environment
12
ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 14 of 34
324
engineering as it has unique catalytic properties. (i) MAT fulfills its physiological
325
function in loading murine FAS with acetyl- and malonyl moieties by lose substrate
326
specificity. As shown in this study, murine MAT is truly polyspecific in its
327
transacylation properties holding out to prospect to load PKS with a broad set of acyl
328
compounds in chimeric systems. (ii) Murine MAT is fast in transacylation, as
329
demonstrated by absolute turnover rates (kcat) of 119 s–1 for malonyl and 99.2 s–1 for
330
acetyl groups. Absolute kinetic values for transacylation rates have not been
331
determined for other ATs yet. (iii) Murine MAT shows remarkable low hydrolytic
332
activity. Rates are more than an order of magnitude lower as compared to ATs of
333
PKS ((9.8 ± 1.7 × 10–3 s–1 and 9.3 ± 0.8 × 10–3 s–1 for malonyl- and acetyl-CoA,
334
respectively; DEBS AT3 of 3.3 × 10–2 s–1 for methylmalonyl-CoA and AveAT0 of
335
2.5 × 10–1 s–1 for 2-methylbutyryl-SNAC).21,
336
relevance of murine MAT for engineering, paving the way to an energetically less
337
costly incorporation of designer acyl-CoA substrates. (iv) Murine MAT is available in
338
the heterologous host E. coli and, as such, easily amenable to in vitro engineering
339
and mutational studies.
27
Low hydrolysis rates increase the
340
Murine MAT may be harnessed for PKS engineering in two aspects. The
341
provided substrate-bound structure may serve as a template to broaden the
342
substrate spectrum of a very specific PKS AT by introducing prominent active site
343
residues, like F553. Or, on the other hand, this particular polyspecific domain can be
344
introduced in a chimeric FAS/PKS system. The latter has been shown in this study
345
by the construction of bimodular FAS constructs. If the structural independence of
346
MAT from the KS-fold also allows an exchange of PKS extender-ATs remains
347
unknown.
348
Broadly substrate-accepting ATs are valuable in PKSs as they can pave the
349
way to regiospecific modification of polyketide lead compounds with non-canonical
350
acyl substrates. To incorporate designer acyl substrates, as e.g. provided by the Erb
351
lab,47 a promiscuous enzyme as MAT may be better suited for adaptive protein
ACS Paragon Plus Environment
13
Page 15 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Biology
352
engineering than highly evolved, substrate specific ATs from existing modular
353
PKSs.48
354 355
ACS Paragon Plus Environment
14
ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 16 of 34
356
Methods
357
Cloning and construct expression. Fasn gene from Mus musculus was
358
purchased from Source BioScience (cDNA clone: IRAVp968A0187D). We essentially
359
performed ligation independent cloning with the In-Fusion HD Cloning Kit (Clonetech)
360
(Table S2). The sequence of all plasmids was confirmed with the “dye terminator”
361
method. All constructs were expressed in E. coli BL21gold(DE3) cells at 20 °C and
362
180 rpm, after induction with 0.25 mM IPTG. Proteins were essentially purified by
363
tandem affinity (Ni-chelating and Strep-tactin) chromatography. Active FAS and
364
active bimodular constructs were obtained by co-expressing with Bacillus subtils Sfp.
365
α-Ketoglutarate dehydrogenase coupled activity assay and fatty acid
366
synthase activity assay. The enzyme-coupled assay performed in this study was
367
adapted from references.26, 27 NADH fluorescence was monitored in a 96-well set-up.
368
Eight data points were collected that cover substrate concentrations (Sol 3) of
369
0.2 × Km; 0.3 × Km; 0.5 × Km; 0.75 × Km; 1.25 × Km; 2 × Km; 3 × Km; 5 × Km. Overall
370
fatty acid synthase activity was measured fluorometrically or by the decrease of
371
absorbance at 340 nm following the oxidation of NADPH at 25 °C.
372
Crystallization. Crystals were obtained at 0.2 M potassium-sodium tartrate,
373
25 % (w/v) PEG 3350 at 20 °C growing to sizes of about 75 × 75 × 75 µm3. Drops
374
with the crystals were supplemented with 0.5 µl of 10 mM malonyl-CoA (Sigma-
375
Aldrich) for up to 2 minutes and subsequently treated with a cryosolution containing
376
20 % (v/v) glycerol in the mother liquor.
377
Data collection and processing. Crystals were exposed to single
378
wavelength X-radiation at the Swiss Light Source (X06SA), and maintained at 100 K,
379
while data were recorded onto a detector (DECTRIS EIGER 16M). All diffraction data
380
are publicly available at https://zenodo.org/record/. The phase problem was solved
381
using molecular replacement method with the program Molrep,49 using the structural
382
model of a monomer from the human FAS KS-MAT didomain (pdb accession code
ACS Paragon Plus Environment
15
Page 17 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Biology
383
3hhd).35 The model was built using Coot.50 Data collection and refinement statistics
384
are given in Table S1.
385
Accession codes. Atomic coordinates and structure factors have been deposited
386
with the Protein Data Bank under the accession code 5my0.
387 388 389
Author contribution A.R.
performed
molecular
cloning,
protein
expression,
purification
390
experiments, enzymatic assays and analyzed corresponding data. A.R. conceived
391
the project. M.G. designed the research. K.V. helped establishing enzyme kinetics
392
and crystallization. Crystallization was performed by A.R., K.V. and K.S.P. Crystal
393
structure was solved by K.S.P. The authors A.R., K.S.P. and M.G. analyzed data and
394
wrote the manuscript.
395 396
Acknowledgements
397
We thank F. Bourdeaux for measurements with HPLC-MS and the Swiss
398
Light Source for the beamtime. We are also grateful to S. Müller-Knapp for carefully
399
reading the manuscript.
400 401
Supporting Information
402
Supporting Information supplies detailed description of Methods, 8 Figures
403
and 2 Tables, and Supplementary Notes on the construct design and expression
404
yields as well as on the implication of MAT polyspecificity for mammalian fatty acid
405
synthesis.
406
This material is available free of charge via the internet.
407
ACS Paragon Plus Environment
16
ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
408
Page 18 of 34
Funding sources
409
This work was supported by a Lichtenberg grant of the Volkswagen
410
Foundation to M.G. (grant number 85701). This project was further supported by the
411
LOEWE program (Landes-Offensive zur Entwicklung wissenschaftlich-ökonomischer
412
Exzellenz) of the state of Hesse and was conducted within the framework of the
413
MegaSyn Research Cluster.
414 415
References
416 417
[1] White, S. W., Zheng, J., Zhang, Y.-M., and Rock, C. O. (2005) The Structural Biology of Type II Fatty Acid Biosynthesis, Annu Rev Biochem. 74, 791-831.
418 419
[2] Maier, T., Leibundgut, M., Boehringer, D., and Ban, N. (2010) Structure and function of eukaryotic fatty acid synthases, Q. Rev. Biophys. 43, 373-422.
420 421 422
[3] Jenni, S., Leibundgut, M., Boehringer, D., Frick, C., Mikolasek, B., and Ban, N. (2007) Structure of fungal fatty acid synthase and implications for iterative substrate shuttling, Science 316, 254-261.
423 424 425
[4] Leibundgut, M., Jenni, S., Frick, C., and Ban, N. (2007) Structural Basis for Substrate Delivery by Acyl Carrier Protein in the Yeast Fatty Acid Synthase, Science 316, 288-290.
426 427 428
[5] Lomakin, I. B., Xiong, Y., and Steitz, T. A. (2007) The Crystal Structure of Yeast Fatty Acid Synthase, a Cellular Machine with Eight Active Sites Working Together, Cell 129, 319-332.
429 430 431
[6] Johansson, P., Wiltschi, B., Kumari, P., Kessler, B., Vonrhein, C., Vonck, J., Oesterhelt, D., and Grininger, M. (2008) Inhibition of the fungal fatty acid synthase type I multienzyme complex, Proc Natl Acad Sci U S A. 105, 12803-12808.
432 433
[7] Wakil, S. J. (1989) Fatty Acid Synthase, A Proficient Multifunctional Enzyme, 28, 4523-4530.
434 435
[8] Hertweck, C. (2009) The Biosynthetic Logic of Polyketide Diversity, Angew. Chem. Int. Ed. 48, 4688–4716.
436 437
[9] Weissman, K. J. (2015) The structural biology of biosynthetic megaenzymes, Nat. Chem. Biol. 11, 660-670.
438 439 440
[10] Tang, Y., Kim, C. Y., Mathews, I. I., Cane, D. E., and Khosla, C. (2006) The 2.7Å crystal structure of a 194-kDa homodimeric fragment of the 6-deoxyerythronolide B synthase, Proc. Natl. Acad. Sci. U.S.A. 103, 11124-11129.
441 442 443
[11] Herbst, D. A., Jakob, R. P., Zähringer, F., and Maier, T. (2016) Mycocerosic acid synthase exemplifies the architecture of reducing polyketide synthases, Nature 531, 533-537.
ACS Paragon Plus Environment
17
Page 19 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Biology
444 445 446 447
[12] Keatinge-Clay, A. T., Shelat, A. A., Savage, D. F., Tsai, S.-C., Miercke, L. J. W., O'Connell III, J. D., Khosla, C., and Stroud, R. M. (2003) Catalysis, Specificity, and ACP Docking Site of Streptomyces coelicolor Malonyl-CoA:ACP Transacylase., Structure 11, 147-154.
448 449 450
[13] Dutta, S., Whicher, J. R., Hansen, D. A., Hale, W. A., Chemler, J. A., Congdon, G. R., Narayan, A. R., Håkansson, K., Sherman, D. H., Smith, J. L., and Skiniotis, G. (2014) Structure of a modular polyketide synthase., Nature 510, 512-517.
451 452
[14] Smith, S., and Tsai, S.-C. (2007) The type I fatty acid and polyketide synthases: a tale of two megasynthases., Nat. Prod. Rep. 24, 1041-1072.
453 454 455
[15] Smith, S., and Stern, A. (1983) The Effect of Aromatic CoA Esters on Fatty Acid Synthetase: Biosynthesis of ω-Phenyl Fatty Acids, Arch. Biochem. Biophys. 222, 259-265.
456 457 458 459
[16] Buckner, J. S., Kolattukudy, P. E., and Rogers, L. (1978) Synthesis of Multimethyl-Branched Fatty Acids by Avian and Mammalian Fatty Acid Synthetase and Its Regulation by Malonyl-CoA Decarboxylase in the Uropygial Gland., Arch. Biochem. Biophys. 186, 152-163.
460 461 462
[17] Seyama, Y., Otsuka, H., Kawaguchi, A., and Yamakawa, T. (1981) Fatty Acid Synthetase from the Harderian Gland of Guinea Pig: Biosynthesis of MethylBranched Fatty Acids, J. Biochem. 90, 789-798.
463 464 465
[18] Pirson, W., Schuhmann, L., and Lynen, F. (1973) The Specificity of Yeast Fatty Acid Synthetase with Respect to the "Priming" Substrate. Decanoyl-CoA and Derivatives as "Primers" of Fatty Acid Synthesis in vitro, Eur. J. Biochem. 36, 16-24.
466 467 468
[19] Sumper, M., Oesterhelt, D., Riepertinger, C., and Lynen, F. (1969) Synthesis of various carboxylic acids by the multienzyme complex of fatty acid synthesis from yeast, and clarification of their structure, Eur. J. Biochem. 10, 377-387.
469 470 471
[20] Dutton, C. J., Gibson, S. P., Goudie, A. C., Holdom, K. S., Pacey, M. S., Ruddock, J. C., Bu'Lock, J. D., and Richards, M. K. (1991) Novel avermectins produced by mutational biosynthesis., The Journal of Antibiotics 44, 357-365.
472 473 474
[21] Wang, F., Wang, Y., Ji, J., Zhou, Z., Yu, J., Zhu, H., Su, Z., Zhang, L., and Zheng, J. (2015) Structural and Functional Analysis of the Loading Acyltransferase from Avermectin Modular Polyketide Synthase, ACS Chem. Biol. 10, 1017-1025.
475 476 477 478
[22] Bhatt, A., Molle, V., Besra, G. S., Jacobs, W. R. J., and Kremer, L. (2007) The Mycobacterium tuberculosis FAS-II condensing enzymes: their role in mycolic acid biosynthesis, acid-fastness, pathogenesis and in future drug development., Mol. Microbiol. 64, 1442-1454.
479 480 481 482
[23] Sundermann, U., Bravo-Rodriguez, K., Klopries, S., Kushnir, S., Gomez, H., Sanchez-Garcia, E., and Schulz, F. (2013) Enzyme-Directed Mutasynthesis: A Combined Experimental and Theoretical Approach to Substrate Recognition of a Polyketide Synthase, ACS Chem. Biol. 8, 443-450.
483 484 485 486
[24] Bravo-Rodriguez, K., Ismail-Ali, A. F., Klopries, S., Kushnir, S., Ismail, S., Fansa, E. K., Wittinghofer, A., Schulz, F., and Sanchez-Garcia, E. (2014) Predicted Incorporation of Non-native Substrates by a Polyketide Synthase Yields Bioactive Natural Product Derivatives, ChemBioChem 15, 1991-1997.
ACS Paragon Plus Environment
18
ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 20 of 34
487 488 489 490
[25] Gregory, M. A., Petkovic, H., Lill, R. E., Moss, S. J., Wilkinson, B., Gaisser, S., Leadlay, P. F., and Sheridan, R. M. (2005) Mutasynthesis of Rapamycin Analogues through the Manipulation of a Gene Governing Starter Unit Biosynthesis, Angew. Chem. Int. Ed. 44, 4757-4760.
491 492 493
[26] Molnos, J., Gardiner, R., Dale, G. E., and Lange, R. (2003) A continuous coupled enzyme assay for bacterial malonyl–CoA:acyl carrier protein transacylase (FabD), Anal. Biochem. 319, 171-176.
494 495 496
[27] Dunn, B. J., Cane, D. E., and Khosla, C. (2013) Mechanism and Specificity of an Acyltransferase Domain from a Modular Polyketide Synthase, Biochemistry 52, 1839-1841.
497 498
[28] Altschul, S. F., Gish, W., Miller, W., Myers, E. W., and Lipman, D. J. (1990) Basic local alignment search tool., J. Mol. Biol. 215, 403-410.
499 500 501
[29] Cleland, W. W. (1963) The kinetics of enzyme-catalyzed reactions with two or more substrates or products, Biochimica et Biophysica Acta (BBA) - Specialized Section on Enzymological Subjects 67, 188-196.
502 503 504
[30] Rangan, V. S., and Smith, S. (1996) Expression in Escherichia coli and Refolding of the Malonyl-/Acetyltransferase Domain of the Multifunctional Animal Fatty Acid Synthase, J. Biol. Chem. 271, 31749-31755.
505 506 507
[31] Misra, A., Sharma, S. K., Surolia, N., and Surolia, A. (2007) Self-Acylation Properties of Type II Fatty Acid Biosynthesis Acyl Carrier Protein, Chem. Biol. 14, 775-783.
508 509 510 511
[32] Wang, Y.-Y., Bai, L.-F., Ran, X.-X., Jiang, X.-H., Wu, H., Zhang, W., Jin, M.-Y., Li, Y.-Q., and Jiang, H. (2015) Biochemical Characterization of a Malonyl-Specific Acyltransferase Domain of FK506 Biosynthetic Polyketide Synthase., Protein Pept. Lett. 22, 2-7.
512 513 514
[33] Liew, C. W., Nilsson, M., Chen, M. W., Sun, H., Cornvik, T., Liang, Z. X., and Lescar, J. (2012) Crystal Structure of the Acyltransferase Domain of the Iterative Polyketide Synthase in Enediyne Biosynthesis, J. Biol. Chem. 287, 23203-23215.
515 516
[34] Maier, T., Leibundgut, M., and Ban, N. (2008) The Crystal Structure of a Mammalian Fatty Acid Synthase, Science 321, 1315-1322.
517 518 519
[35] Pappenberger, G., Benz, J., Gsell, B., Hennig, M., Ruf, A., Stihle, M., Thoma, R., and Rudolph, M. G. (2010) Structure of the Human Fatty Acid Synthase KS–MAT Didomain as a Framework for Inhibitor Design, J. Mol. Biol. 397, 508-519.
520 521 522
[36] Krissinel, E., and Henrick, K. (2004) Secondary-structure matching (SSM), a new tool for fast protein structure alignment in three dimensions, Acta Crystallogr. D Biol. Crystallogr. 60, 2256-2268.
523 524 525 526
[37] Bunkoczi, G., Misquitta, S., Wu, X., Lee, W. H., Rojkova, A., Kochan, G., Kavanagh, K. L., Oppermann, U., and Smith, S. (2009) Structural Basis for Different Specificities of Acyltransferases Associated with the Human Cytosolic and Mitochondrial Fatty Acid Synthases, Chem. Biol. 16, 667-675.
527 528
[38] Nardini, M., and Dijkstra, B. W. (1999) α/β Hydrolase fold enzymes: the family keeps growing, Curr. Opin. Struct. Biol. 9, 732-737.
ACS Paragon Plus Environment
19
Page 21 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Biology
529 530
[39] Hol, W. G. (1985) Effects of the α-helix dipole upon the functioning and structure of proteins and peptides., Adv. Biophys. 19, 133-165.
531 532 533
[40] Oefner, C., Schulz, H., D'Arcy, A., and Dale, G. E. (2006) Mapping the active site of Escherichia coli malonyl-CoA-acyl carrier protein transacylase (FabD) by protein crystallography, Acta Crystallogr. D Biol. Crystallogr. 62, 613-618.
534 535 536 537
[41] Rangan, V. S., and Smith, S. (1997) Alteration of the Substrate Specificity of the Malonyl-CoA/Acetyl-CoA: Acyl Carrier Protein S-Acyltransferase Domain of the Multifunctional Fatty Acid Synthase by Mutation of a Single Arginine Residue, J. Biol. Chem. 272, 11975-11978.
538 539 540
[42] Gajewski, J., Buelens, F., Serdjukow, S., Janßen, M., Cortina, N., Grubmüller, H., and Grininger, M. (2017) Engineering fatty acid synthases for directed polyketide production, Nat. Chem. Biol. 13, 363–365.
541 542
[43] Warshel, A., Aqvist, J., and Creighton, S. (1989) Enzymes work by solvation substitution rather than by desolvation, Proc. Natl. Acad. Sci. U. S. A. 86, 5820-5824.
543 544
[44] Helfrich, E. J. N., and Piel, J. (2016) Biosynthesis of polyketides by trans-AT polyketide synthases, Nat. Prod. Rep. 33, 231-316.
545 546 547
[45] Zhu, Z., Zhou, Y. J., Krivoruchko, A., Grininger, M., Zhao, Z. K., and Nielsen, J. (2017) Expanding the product portfolio of fungal type I fatty acid synthases, Nat. Chem. Biol. 13, 360-362.
548 549 550
[46] Robbins, T., Liu, Y.-C., Cane, D. E., and Khosla, C. (2016) Structure and mechanism of assembly line polyketide synthases, Curr. Opin. Struct. Biol. 41, 1018.
551 552 553 554
[47] Peter, D. M., Schada von Borzyskowski, L., Kiefer, P., Christen, P., Vorholt, J. A., and Erb, T. J. (2015) Screening and Engineering the Synthetic Potential of Carboxylating Reductases from Central Metabolism and Polyketide Biosynthesis, Angew. Chem. Int. Ed. Engl. 54, 13457-13461.
555 556 557
[48] Mohamed, M. F., and Hollfelder, F. (2013) Efficient, crosswise catalytic promiscuity among enzymes that catalyze phosphoryl transfer, Biochim. Biophys. Acta 1834, 417-424.
558 559
[49] Vagin, A., and Teplyakov, A. (1997) MOLREP: an Automated Program for Molecular Replacement, J. Appl. Crystallogr. 30, 1022-1025.
560 561
[50] Emsley, P., Lohkamp, B., Scott, W. G., and Cowtan, K. (2010) Features and development of Coot, Acta Crystallogr. D Biol. Crystallogr. 66, 486-501.
562 563 564
[51] Grininger, M. (2014) Perspectives on the evolution, assembly and conformational dynamics of fatty acid synthase type I (FAS I) systems, Curr. Opin. Struct. Biol. 25, 49-56.
565 566 567 568
[52] Afonine, P. V., Moriarty, N. W., Mustyakimov, M., Sobolev, O. V., Terwilliger, T. C., Turk, D., Urzhumtsev, A., and Adams, P. D. (2015) FEM: feature-enhanced map, Acta Crystallogr. D Biol. Crystallogr. 71, 646-666.
569
ACS Paragon Plus Environment
20
ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 22 of 34
570 571 572
Figure 1: Animal fatty acid synthesis and MAT-mediated transacylation.
573
(A) Compartmentalized synthesis by FAS centered around the loading function of the
574
MAT domain. The elongation cycle is divided into a condensation reaction,
575
performed by KS, and modification of the β-carbon, catalyzed by the domains KR,
576
DH and ER. The flexible attachment of ACP via linkers is abstracted by the zigzag
577
lines for the elongation cycle. Abbreviations as introduced in the text. Adapted from
578
51
579
CoA-ester is transferred to the enzyme to form an acyl-enzyme intermediate (MAT-X;
580
active site serine-acylated), with the concomitant release of the free CoA (ping). The
581
ACP then interacts with the loaded enzyme intermediate forming the acylated ACP
582
and regenerating the native enzyme (pong). The left branch represents the common
583
side reaction of hydrolysis, which can occur when water is utilized as a nucleophile
584
attacking the acyl-enzyme intermediate.
. (B) Scheme of the ping-pong bi-bi mechanism of MAT. The acyl-moiety of the
585
ACS Paragon Plus Environment
21
Page 23 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Biology
586 587
Figure 2. Overall structure of murine KS-MAT didomain.
588
(A) Superposition of the two biological dimers of KS-MAT onto the porcine FAS KS-
589
MAT dimer. For clarity, the full X-ray structural model of porcine FAS, excluding the
590
domains ACP and TE that were not traceable in electron density,34 is shown as inset.
591
Porcine FAS KS-MAT is shown in grey, and the two biological dimers in orange
592
(chain A and B) and cyan (chain C and D). Subdomains and folds are indicated by
593
labels and coloring. (B) Superposition of the four KS-MAT modules in the asymmetric
594
unit aligned via the KS domains (residue range 1–407; upper panel) and of the MAT
595
(including LD/LD2) aligned via their α/β-hydrolase domain (residue ranges 488–615
596
and 685–806; lower panel). Domains and folds are color coded as depicted in the
597
attached cartoon. Cofactor and residues are shown for chain D (malonyl-loaded
598
state). (C) Binding site of murine MAT in its apo-state (chain B), in complex with
599
malonyl-CoA (chain C) and in its acyl-enzyme state (chain A and D). Electron density
600
of an unbiased feature-enhanced map of the active site regions is contoured at 1σ.52
601
View 70° rotated (y-axis) as compared to Figure 2B.
602
ACS Paragon Plus Environment
22
ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 24 of 34
603 604
Figure 3. MAT binding site architecture.
605
(A) Acyl-CoA binding cavity as comprised by the α/β-hydrolase and the ferredoxin-
606
like domain. The binding cavity is shown in mesh representation (left panel) and with
607
surfaces colored in electrostatic potential (right). The cavity was drawn with default
608
settings of PyMOL (vacuum electrostatics function; The PyMOL Molecular Graphics
609
System, Schrödinger, LLC.). Figures of the LD-MAT substructure in the upper panel
610
are provided for overview. (B) Comparison of the binding site of the acyl-enzyme
611
state (chain D) with the apo-state (chain A, orange) of human MAT. Distances
612
indicate that N738 is the only residue in H-bond distance in the apo state.
613
Superposition has been performed with α/β-hydrolase domains (residue ranges 488–
614
615 and 685–806; pdb: 3hhd).35 Residue numbers are identical for both proteins. H-
615
bonding to the guanidinium group is indicated by dashed lines. (C) Malonyl-loaded
616
enzyme states of MAT (chain D, residues in yellow) in superposition with malonyl
617
transferases FabD (pdb: 2g2z, grey, H-bonds in blue dashed lines),40 and AT of
618
DynE8 (pdb: 4amp, white, H-bonds in grey).33 Superposition indicates side-on vs.
619
end-on bidentate coordination of arginine and the carboxyl group of serine bound
620
malonyl. H-bond networks of active site arginines are different in FabD (right top) and
621
in AT of DynE8 (right bottom) to MAT (see Figure 3B).
622
ACS Paragon Plus Environment
23
Page 25 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Biology
623 624
Figure 4: Double-reciprocal plots of the Michaelis-Menten constants. Reciprocal
625
Km (top panel) and reciprocal initial rates kcat (bottom) are presented as functions of
626
reciprocal ACP concentrations. Data collected on malonyl-CoA are shown in black,
627
acetyl-CoA in grey, and linear fits are accordingly. Error bars reflect the standard
628
deviation between three biological replicates. A biological replicate was defined by
629
three independently expressed MAT preparations and different batches of purified,
630
high quality ACP (see Figure S3), mainly from bicistronic expression with Sfp. To
631
establish the fidelity of the assay, every measurement was performed in technical
632
triplicates. Tables underneath graphs summarize absolute kinetic constants
633
calculated from the double-reciprocal plot. Michaelis-Menten constants for ACP were
634
calculated from both plots and averaged (245 µM and 245 µM for malonyl-transfer;
635
284 µM and 246 µM for acetyl-transfer).
636 637
ACS Paragon Plus Environment
24
ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 26 of 34
638 639 640
Figure 5: MAT as tool in PKS engineering.
641
(A) Domain organization of murine FAS and constructs engineered in this study; FAS
642
(1), KS-MAT (2), LD-MAT (3), MAT-MBP (4), LD-MAT-ACP-FAS (5), MAT-ACP-FAS
643
(6). (B) Coomassie-stained SDS-PAGE gel (NuPAGE 4–12 % Bis-Tris) of purified
644
constructs after tandem affinity chromatography and SEC. Calculated sizes of
645
constructs (1), (2), (3) and (4) are 277, 97, 49 and 81 kDa, respectively. (C) SEC of
646
purified constructs with absorption normalized to the highest peak. Peaks (1), (2)
647
(2m)), (4) and (3) correspond to apparent molecular weights of 601, 231 (monomer:
648
118), 85 and 48 kDa, respectively. Calibration is shown in inset; thyroglobulin (T),
649
aldolase (A), ovalbumin (O) and ribonuclease (R). (D) Partial activity of MAT in the
650
αKGDH-assay monitored at fixed substrate concentrations of 20 µM CoA-ester and
651
60 µM ACP. Malonyl- and acetyl-turnovers are shown as black and grey bars,
652
respectively. All data were collected in technical triplets. Error reflects standard
653
deviation from four biological replicates. (E) In-gel fluorescence for monitoring the
654
degree of phosphopantetheinylation in co-expressions with Sfp (lower panel) and
655
Coomassie-stained SDS-PAGE gel (upper panel). Star (*) indicates activated
656
(phosphopantetheinylated) protein received in co-expressions with Sfp. Calculated
657
sizes of constructs (1), (2), (5) and (6) are 277, 338 and 323 kDa, respectively.
658
Proteins were incubated with 10 mM MgCl2, 10 µM purified Sfp and 1 µM CoA 647
659
for 1 h at 37 °C. SDS-PAGE gel (NuPAGE 4–12 % Bis-Tris) of fluorescently labeled
660
constructs is shown after fluorescence detection and Commassie-staining. For
661
monitoring phosphopantetheinylation, KS-MAT (not containing ACP) was used as a
662
negative, and FAS, expressed as apo-protein and in vitro phosphopantetheinylated,
663
was used as positive control. (F) Preparative SEC of purified constructs after co-
664
expression with Sfp with absorption normalized to the highest peak to isolate the
665
dimeric fraction of each protein. Peaks (1*), (5*) and (6*) correspond to apparent
ACS Paragon Plus Environment
25
Page 27 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Biology
666
molecular weights of 697 kDa, 784 kDa and 727 kDa, respectively. Peak shoulders
667
indicate the presence of monomers in all samples, which are in temperature and
668
buffer dependent equilibrium with the dimer (for analytic SEC, see also Figure S8).
669
(G) FAS activity addressed by the specific oxidation of NADPH in 50 mM potassium
670
phosphate. The consumption of NADPH was monitored fluorometrically. FAS without
671
Sfp co-expression shows no activity.
672
ACS Paragon Plus Environment
26
ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 28 of 34
673
Table 1: Kinetic analysis of the transacylation reaction with unusual CoA-esters at a
674
fixed acceptor concentration of 60 µM ACP. app
(µM)
app
–1
kcat/Km (M 1 × 106)
s
–
Km
Malonyl-CoA
1.28
±
0.13
15.6
±
1.4
12.2
± 1.7
9.8
±
1.7
kcat
(s )
–1
Substrate
Hydrolysis rate 10–3 (s-–1)
Methylmalonyl-CoA
0.62
±
0.07
4.0
±
0.8
6.5
± 1.4
6.0
±
1.2
Acetyl-CoA
1.63
±
0.10
13.7
±
2.1
8.4
± 1.4
9.3
±
0.8
Butyryl-CoA
1.64
±
0.16
10.8
±
0.3
6.6
± 0.7
8.5
±
1.7
Octanoyl-CoA*
0.73
±
0.28
4.1
±
0.3
5.6
± 2.2
6.1
±
1.0
Palmitoyl-CoA**
4.8
±
1.1
2.5
±
0.5
0.5
± 0.2
16.0
±
1.4
Acetoacetyl-CoA*
0.86
±
0.03
7.3
±
0.4
8.5
± 0.5
13.0
±
2.5
Hydroxybutyryl-CoA
1.72
±
0.39
6.1
±
1.1
3.6
± 1.0
20.4
±
2.7
Crotonyl-CoA
1.36
±
0.13
6.0
±
1.1
4.4
± 0.9
24.0
±
2.0
Phenylacetyl-CoA*
0.96
±
0.10
6.0
±
0.2
6.2
± 0.7
21.3
±
2.3
Methylbutyryl-CoA
2.06
±
0.36
7.2
±
0.9
3.5
± 0.7
10.4
±
1.3
Succinyl-CoA
2.10
±
0.17
13.4
0.6
6.4
± 0.6
28.7
±
2.7
R606A mutant
Kmapp
(µM)
kcat
app
± –1
(s )
kcat/Km (M 6 × 10 )
–1
s
–
1
Hydrolysis rate –3 –1 10 (s )
Malonyl-CoA
18
±
5.0
5
±
1.7
0.3
± 0.1
14.7
±
2.3
Acetyl-CoA
3.0
±
0.56
13
±
2.1
4
± 1.1
15.2
±
2.1
675
All parameters were determined in technical and biological (independent expressed
676
and purified MAT preparations) triplets. (*) and (**) MAT at 2 nM and 4 nM,
677
respectively, were used, because of low turnover rates.
678
Hydrolysis rate was determined in 6 repetitions.
679
ACS Paragon Plus Environment
27
Page 29 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Biology
680
ACS Paragon Plus Environment
28
ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Figure 1: Animal fatty acid synthesis and MAT-mediated transacylation. (A) Compartmentalized synthesis by FAS centered around the loading function of the MAT domain. The elongation cycle is divided into a condensation reaction, performed by KS, and modification of the β-carbon, catalyzed by the domains KR, DH and ER. The flexible attachment of ACP via linkers is abstracted by the zigzag lines for the elongation cycle. Abbreviations as introduced in the text. Adapted from 51. (B) Scheme of the ping-pong bi-bi mechanism of MAT. The acyl-moiety of the CoA-ester is transferred to the enzyme to form an acyl-enzyme intermediate (MAT-X; active site serine-acylated), with the concomitant release of the free CoA (ping). The ACP then interacts with the loaded enzyme intermediate forming the acylated ACP and regenerating the native enzyme (pong). The left branch represents the common side reaction of hydrolysis, which can occur when water is utilized as a nucleophile attacking the acyl-enzyme intermediate.
140x54mm (300 x 300 DPI)
ACS Paragon Plus Environment
Page 30 of 34
Page 31 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Biology
Figure 2. Overall structure of murine KS-MAT didomain. (A) Superposition of the two biological dimers of KS-MAT onto the porcine FAS KS-MAT dimer. For clarity, the full X-ray structural model of porcine FAS, excluding the domains ACP and TE that were not traceable in electron density,34 is shown as inset. Porcine FAS KS-MAT is shown in grey, and the two biological dimers in orange (chain A and B) and cyan (chain C and D). Subdomains and folds are indicated by labels and coloring. (B) Superposition of the four KS-MAT modules in the asymmetric unit aligned via the KS domains (residue range 1–407; upper panel) and of the MAT (including LD/LD2) aligned via their α/β-hydrolase domain (residue ranges 488–615 and 685–806; lower panel). Domains and folds are color coded as depicted in the attached cartoon. Cofactor and residues are shown for chain D (malonyl-loaded state). (C) Binding site of murine MAT in its apo-state (chain B), in complex with malonyl-CoA (chain C) and in its acyl-enzyme state (chain A and D). Electron density of an unbiased feature-enhanced map of the active site regions is contoured at 1σ.52 View 70° rotated (y-axis) as compared to Figure 2B.
140x120mm (300 x 300 DPI)
ACS Paragon Plus Environment
ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Figure 3. MAT binding site architecture. (A) Acyl-CoA binding cavity as comprised by the α/β-hydrolase and the ferredoxin-like domain. The binding cavity is shown in mesh representation (left panel) and with surfaces colored in electrostatic potential (right). The cavity was drawn with default settings of PyMOL (vacuum electrostatics function; The PyMOL Molecular Graphics System, Schrödinger, LLC.). Figures of the LD-MAT substructure in the upper panel are provided for overview. (B) Comparison of the binding site of the acyl-enzyme state (chain D) with the apostate (chain A, orange) of human MAT. Distances indicate that N738 is the only residue in H-bond distance in the apo state. Superposition has been performed with α/β-hydrolase domains (residue ranges 488–615 and 685–806; pdb: 3hhd).35 Residue numbers are identical for both proteins. H-bonding to the guanidinium group is indicated by dashed lines. (C) Malonyl-loaded enzyme states of MAT (chain D, residues in yellow) in superposition with malonyl transferases FabD (pdb: 2g2z, grey, H-bonds in blue dashed lines),40 and AT of DynE8 (pdb: 4amp, white, H-bonds in grey).33 Superposition indicates side-on vs. end-on bidentate coordination of arginine and the carboxyl group of serine bound malonyl. H-bond networks of active site arginines are different in FabD (right top) and in AT of DynE8 (right bottom) to MAT (see Figure 3B).
140x141mm (300 x 300 DPI)
ACS Paragon Plus Environment
Page 32 of 34
Page 33 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Biology
Figure 4: Double-reciprocal plots of the Michaelis-Menten constants. Reciprocal Km (top panel) and reciprocal initial rates kcat (bottom) are presented as functions of reciprocal ACP concentrations. Data collected on malonyl-CoA are shown in black, acetyl-CoA in grey, and linear fits are accordingly. Error bars reflect the standard deviation between three biological replicates. A biological replicate was defined by three independently expressed MAT preparations and different batches of purified, high quality ACP (see Figure S3), mainly from bicistronic expression with Sfp. To establish the fidelity of the assay, every measurement was performed in technical triplicates. Tables underneath graphs summarize absolute kinetic constants calculated from the double-reciprocal plot. Michaelis-Menten constants for ACP were calculated from both plots and averaged (245 µM and 245 µM for malonyl-transfer; 284 µM and 246 µM for acetyl-transfer). 119x214mm (300 x 300 DPI)
ACS Paragon Plus Environment
ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Figure 5: MAT as tool in PKS engineering. (A) Domain organization of murine FAS and constructs engineered in this study; FAS (1), KS-MAT (2), LDMAT (3), MAT-MBP (4), LD-MAT-ACP-FAS (5), MAT-ACP-FAS (6). (B) Coomassie-stained SDS-PAGE gel (NuPAGE 4–12 % Bis-Tris) of purified constructs after tandem affinity chromatography and SEC. Calculated sizes of constructs (1), (2), (3) and (4) are 277, 97, 49 and 81 kDa, respectively. (C) SEC of purified constructs with absorption normalized to the highest peak. Peaks (1), (2) (2m)), (4) and (3) correspond to apparent molecular weights of 601, 231 (monomer: 118), 85 and 48 kDa, respectively. Calibration is shown in inset; thyroglobulin (T), aldolase (A), ovalbumin (O) and ribonuclease (R). (D) Partial activity of MAT in the αKGDH-assay monitored at fixed substrate concentrations of 20 µM CoA-ester and 60 µM ACP. Malonyland acetyl-turnovers are shown as black and grey bars, respectively. All data were collected in technical triplets. Error reflects standard deviation from four biological replicates. (E) In-gel fluorescence for monitoring the degree of phosphopantetheinylation in co-expressions with Sfp (lower panel) and Coomassiestained SDS-PAGE gel (upper panel). Star (*) indicates activated (phosphopantetheinylated) protein received in co-expressions with Sfp. Calculated sizes of constructs (1), (2), (5) and (6) are 277, 338 and 323 kDa, respectively. Proteins were incubated with 10 mM MgCl2, 10 µM purified Sfp and 1 µM CoA 647 for
ACS Paragon Plus Environment
Page 34 of 34
Page 35 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Biology
1 h at 37 °C. SDS-PAGE gel (NuPAGE 4–12 % Bis-Tris) of fluorescently labeled constructs is shown after fluorescence detection and Commassie-staining. For monitoring phosphopantetheinylation, KS-MAT (not containing ACP) was used as a negative, and FAS, expressed as apo-protein and in vitro phosphopantetheinylated, was used as positive control. (F) Preparative SEC of purified constructs after coexpression with Sfp with absorption normalized to the highest peak to isolate the dimeric fraction of each protein. Peaks (1*), (5*) and (6*) correspond to apparent molecular weights of 697 kDa, 784 kDa and 727 kDa, respectively. Peak shoulders indicate the presence of monomers in all samples, which are in temperature and buffer dependent equilibrium with the dimer (for analytic SEC, see also Figure S8). (G) FAS activity addressed by the specific oxidation of NADPH in 50 mM potassium phosphate. The consumption of NADPH was monitored fluorometrically. FAS without Sfp co-expression shows no activity.
140x160mm (300 x 300 DPI)
ACS Paragon Plus Environment