Deciphering Nature's Intricate Way of N,S-Dimethylating l -Cysteine: Sequential Action of Two Bifunctional Adenylation Domains

Article abstract of DOI:10.1021/acs.biochem.7b00980

Dimethylation of amino acids consists of an interesting and puzzling series of events that could be achieved, during nonribosomal peptide biosynthesis, either by a single adenylation (A) domain interrupted by a methyltransferase (M) domain or by the sequential action of two of such independent enzymes. Herein, to establish the method by which Nature N,S-dimethylates l-Cys, we studied its formation during thiochondrilline A biosynthesis by evaluating TioS(A_3aM_3SA_3bT₃) and TioN(A_aM_NA_b). This study not only led to identification of the exact pathway followed in Nature by these two enzymes for N,S-dimethylation of l-Cys, but also revealed that a single interrupted A domain can N,N-dimethylate amino acids, a novel phenomenon in the nonribosomal peptide field. These findings offer important and useful insights for the development and engineering of novel interrupted A domain enzymes to serve, in the future, as tools for combinatorial biosynthesis.

Full text of DOI:10.1021/acs.biochem.7b00980

Article
Cite This: Biochemistry XXXX, XXX, XXX-XXX
pubs.acs.org/biochemistry
Deciphering Nature’s Intricate Way of N,S-Dimethylating L‑Cysteine:
Sequential Action of Two Bifunctional Adenylation Domains
Shogo Mori, Atefeh Garzan, Oleg V. Tsodikov, and Sylvie Garneau-Tsodikova*
Department of Pharmaceutical Sciences, College of Pharmacy, University of Kentucky, Lexington, Kentucky 40536-0596, United
States
*
S Supporting Information
ABSTRACT: Dimethylation of amino acids consists of an
interesting and puzzling series of events that could be achieved,
during nonribosomal peptide biosynthesis, either by a single
adenylation (A) domain interrupted by a methyltransferase (M)
domain or by the sequential action of two of such independent
enzymes. Herein, to establish the method by which Nature N,S-
dimethylates L-Cys, we studied its formation during thiochondril-
line A biosynthesis by evaluating TioS(A M A T ) and
3
a
3S 3b 3
TioN(A M A ). This study not only led to identification of
a
N b
the exact pathway followed in Nature by these two enzymes for
N,S-dimethylation of L-Cys, but also revealed that a single
interrupted A domain can N,N-dimethylate amino acids, a novel
phenomenon in the nonribosomal peptide field. These findings offer important and useful insights for the development and
engineering of novel interrupted A domain enzymes to serve, in the future, as tools for combinatorial biosynthesis.
atural products have been and continue to be a very
important source of drugs and drug leads. Members of
natural product community. Because methylation may
1
5
N
beneficially alter the activity of natural products, engineered
6
one of the largest families of natural products, the non-
ribosomal peptides (NRPs), are biosynthesized on mega-
methyltransferases or ones (naturally occurring or engineered)
7
−10
with promiscuous activity
have been investigated as
2
enzymes termed nonribosomal peptide synthetases (NRPSs).
biocatalysts for the synthesis of novel bioactive compounds.
However, although very important, M domains, which are
commonly found embedded in A domains of NRPSs, remain
NRPSs are separated into modules that work sequentially to
assemble amino acid-like building blocks into short peptides,
which are usually further modified to form complex natural
products. Each basic NRPS module consists of three core
domains: an adenylation (A), a thiolation (T), and a
condensation (C) domain. The role of the A domain is to
activate its specifically selected substrate to its AMP counterpart
by using adenosine triphosphate (ATP). Once adenylated, the
substrate is transferred onto the thiol group of the
phosphopantetheine (Ppant) arm of its holo (active) T domain
partner, which is generated by posttranslational modification of
the apo (inactive) enzyme by a phosphopantetheinyltransferase
11
poorly understood. So far, all that is known about these M
domains embedded into A domains is that (i) the resulting
1
2,13
AaMAb enzymes are bifunctional,
(ii) they can be
14
engineered (by removal or swapping of M domains), and
iii) in some cases, methylation occurs after adenylation. It
13
(
has also been noted that auxiliary domains can be inserted into
11
various areas of A domains, which comprise 10 conserved
15
core signature sequence motifs (a1−a10). The majority of
interrupted A domains have their M domains embedded
16,17
between their a8 and a9 motifs.
So far, only one example of
(
PPTase) and coenzyme A (CoA). Finally, a C domain
an M domain placed within the relatively long amino acid
sequence between the a2 and a3 motifs of an A domain,
TioN(A M A ), has been found in natural product gene
catalyzes condensation of the substrates tethered to its
surrounding T domains to form an amide bond. This sequence
of events is repeated in a successive manner, working from the
most upstream to the most downstream module of the NRPS
assembly-line, to eventually produce an NRP chain. This NRP
chain can be modified during its assembly by auxiliary domains
incorporated into NRPS modules, such as methyltransferase
a
N b
clusters. We previously preliminarily demonstrated that TioN-
A M A ) is capable of adenylating and methylating L-Cys
(
a
N b
without confirming the exact methylated product (N-Me-L-Cys,
1
3
S-Me-L-Cys, or N,S-diMe-L-Cys) that it generated.
Some of the major questions that remain unanswered about
these bifunctional enzymes are (i) whether methylation occurs
(
(
M), oxidase (Ox), epimerase (E), cyclizase (Cy), halogenase
3
Hal), and ketoreductase (KR) domains. These auxiliary
domains, with M being one of the most common, are major
sources of diversification of NRPs.
As methylation is an important and prevalent modification of
4
Received: September 29, 2017
Revised: October 21, 2017
natural products, it has attracted a great deal of attention in the
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.biochem.7b00980
Biochemistry XXXX, XXX, XXX−XXX

Biochemistry
Article
Figure 1. (A) Structures of thiocoraline and thiochondrilline A (a metabolite from Verrucosispora sp. isolated from the sponge Chondrilla caribensis f.
caribensis). (B) Thiocoraline (and thiocondrilline A) gene cluster from which both NRPs are biosynthesized. (C) Structural organization of the
thiocoraline (and thiochondrilline A) NRPS. Abbreviations: A, adenylation; ACP, acyl carrier protein; C, condensation; T, thiolation; E,
epimerization; M, methylation; TE, thioesterase.
(
(
a) pre- or post-adenylation or (b) pre- or post-thiolation, (ii)
a) whether two interrupted A domains sequentially methylate
MATERIALS AND METHODS
■
Bacterial Strains, Plasmids, Materials, and Instrumen-
the backbone and side chain (or vice versa) of an amino acid to
dimethylate it and, if so, (b) the order in which the backbone
versus side-chain methylations occur, and (iii) whether one
interrupted A domain can perform both backbone and side-
chain methylation on a given amino acid residue.
13
29
tation. The TioN(A M A ) and Sfp proteins were
a
N b
overexpressed and purified using previously described proce-
dures. Chemically competent Escherichia coli TOP10 cells were
purchased from Invitrogen (Carlsbad, CA). The E. coli BL21
(DE3)ybdZ::aac(3)IV strain used for co-expression of TioS-
(A M A ) with TioT was a generous gift from M. G. Thomas
As our group has a long-standing interest in understanding
3a
3S 3b
the unique transformations involved in thiocoraline and
(University of WisconsinMadison, Madison, WI). The
pET28a and pACYCDuet-1 overexpression vectors were
bought from Novagen (Gibbstown, NJ). DNA primers for
polymerase chain reaction (PCR) were purchased from Sigma-
Aldrich (St. Louis, MO). Restriction enzymes, Phusion DNA
polymerase, T4 DNA ligase, and all other cloning reagents were
purchased from New England BioLabs (Ipswich, MA). All
chemicals (with the exception of those presented below) and
buffer components were purchased from Sigma-Aldrich or
VWR and used without any further purification. All reagents for
chemical syntheses were purchased from commercial sources
18−23
thiochondrilline A biosyntheses,
as well as in adenylation
2
4,25
12,13,26
domains
and methyltransferase enzymes,
to answer
these questions, we found the N,S-diMe-L-Cys residue of
2
7
28
thiocoraline and thiocondrilline A to be a highly suitable
model (Figure 1A). By comparing the structures of thiocoraline
and thiochondrilline A and by examining their common gene
cluster and NRPS assembly-line (Figure 1B,C), we postulated
that TioS(A M A T ), a part of the TioS protein (C -
3a
3S 3b
3
3
A M A -T -C -A M A -T -TE), and the stand-alone
3
a
3S 3b
3
4
4a 4S 4b
4
TioN(A M A ) could be responsible for backbone and side-
a
N
b
1
and used without any further purification. H nuclear magnetic
chain methylation of L-Cys, respectively, during the formation
of one of the N,S-diMe-L-Cys residues (Figure 1A, highlighted
by a red box) found in the thiochondrilline A structure. On the
basis of this hypothesis, the N,S-diMe-L-Cys covalently tethered
resonance (NMR) spectra were recorded on a 400 MHz NMR
spectrometer (Varian Inova) using CDCl or D O. Chemical
3
2
shifts (δ) are referenced to residual solvent peaks and are
reported in parts per million. All reactions were performed
under a nitrogen atmosphere, and all reported yields represent
to TioS(T ) [N,S-diMe-L-Cys-S-TioS(T )] could be produced
3
3
1
via one of 12 potential unique pathways (Figure 2, pathways
a −a to l −l ). Herein, by deciphering Nature’s clever and
isolated yields. All compounds were characterized by H NMR
1
4
1
4
and were in complete agreement with characterizations
reported in the literature for these compounds. All compounds
intricate way of N,S-dimethylating L-Cys, we answer the major
remaining questions posed above about bifunctional adenylat-
ing/methylating enzymes.
3
are ≥95% pure according to NMR spectra. [methyl- H]SAM
3
32
35
(S-adenosyl-L-methionine), [ H]AcCoA, [ P]PP , and [ S]-L-
i
B
DOI: 10.1021/acs.biochem.7b00980
Biochemistry XXXX, XXX, XXX−XXX

Biochemistry
Article
acidified with HCl (1 N). The reaction mixture was evaporated
to dryness under reduced pressure to afford N-methyl-L-
1
cysteine hydrochloride (0.14 g, 74%) as a light yellow solid: H
30
NMR (400 MHz, D O, which matches the literature ) δ 4.18
2
(
t, J = 4.4 Hz, 1H), 3.24 (dd, J = 15.6, 4.4 Hz, 1H), 3.11 (dd, J
=
15.6, 4.4 Hz, 1H), 2.76 (s, 3H).
Synthesis of N-Boc-S-methyl-L-cysteine. The known N-Boc-
S-methyl-L-cysteine was prepared as follows. (Boc) O (0.89 g,
2
4
.07 mmol) was added to a solution of S-methyl-L-cysteine
(
0.50 g, 3.70 mmol) and NaHCO (0.78 g, 9.25 mmol) in 1,4-
3
dioxane (10 mL) and H O (15 mL). The reaction mixture was
2
stirred at room temperature for 12 h. The reaction mixture was
treated with H O (10 mL) and acidified with HCl (1 N) to pH
2
∼
3. The reaction mixture was extracted with EtOAc (3 × 10
mL), and the organic layer was concentrated under reduced
pressure to afford N-Boc-S-methyl-L-cysteine (0.87 g, 100%) as
1
a colorless oil: H NMR (400 MHz, CDCl , which matches the
3
31
literature ) δ 5.37 (m, 1H), 4.52 (m, 1H), 2.96 (m, 2H), 2.13
(
s, 3H), 1.43 (s, 9H).
Synthesis of N,S-Dimethyl-L-cysteine Hydrochloride. The
known N,S-dimethyl-L-cysteine hydrochloride was prepared as
follows. N-Boc-S-methyl-L-cysteine (0.587 g, 3.70 mmol) was
dissolved in THF (3 mL) and the mixture added to a solution
of NaH (0.36 g, 8.88 mmol, 60% suspension in mineral oil) in
THF (12 mL) at 0 °C dropwise. MeI (0.76 mL, 12.21 mmol)
was added, and the reaction mixture was stirred at room
temperature for 12 h. MeOH and H O were then added to
2
quench the reaction, and the mixture was then acidified with
HCl (1 N) to pH ∼5. The reaction mixture was extracted with
EtOAc, and the organic layer was concentrated under reduced
pressure. The residue was dissolved in HCl (3 mL, 3 N solution
in EtOAc). The reaction mixture was stirred at room
temperature for 2 h. The reaction mixture was treated with
Figure 2. Twelve potential pathways (a−l, with four steps each) for
the formation of N,S-dimethyl-L-Cys-S-TioS(T ) by TioN(A M A )
H O (3 mL) and washed with EtOAc (3 × 3 mL). The reaction
2
3
a
N b
and TioS(A M A ).
mixture was evaporated to dryness under reduced pressure to
3a
3S 3b
afford N,S-dimethyl-L-cysteine hydrochloride (0.27 g, 60%) as a
1
yellow wax: H NMR (400 MHz, D O, which matches the
2
Cys were purchased from PerkinElmer (Waltham, MA).
EcoLume liquid scintillation cocktail was purchased from MP
biomedicals (Santa Ana, CA). Radioactivity was counted by
using a TriCarb 2900TR liquid scintillation analyzer
30
literature ) δ 4.19 (td, J = 5.6, 1.2 Hz, 1H), 3.19 (d, J = 5.6 Hz,
2
H), 2.79 (s, 3H), 2.58 (br s, 1H), 2.16 (s, 3H).
Synthesis of 2-Ethylacrylic Acid. To a mixture of 2-
ethylmalonic acid (1.00 g, 7.57 mmol) and diethylamine
(
PerkinElmer). DNA sequencing was performed at Eurofins
(
2.34 mL, 22.71 mmol) in EtOAc (10 mL) was added
Scientific (Louisville, KY).
paraformaldehyde (0.34 g, 11.36 mmol) at 0 °C. The resulting
mixture was refluxed for 12 h. The reaction mixture was treated
with H O (10 mL) and acidified with HCl (1 N) to pH ∼1.
The reaction mixture was extracted with EtOAc (3 × 10 mL),
and the organic layer was concentrated under reduced pressure
to afford the known 2-ethylacrylic acid (0.49 g, 64%) as a light
yellow oil: H NMR (400 MHz, CDCl , which matches the
literature ) δ 6.27 (s, 1H), 5.64 (d, J = 1.6 Hz, 1H), 2.31 (q, J
Chemical Synthesis and Characterization of Com-
pounds To Be Used as Potential Substrates. Synthesis of
N-Methyl-L-cysteine Hydrochloride. The known N-methyl-L-
cysteine hydrochloride was prepared as follows. Boc-Cys(Trt)-
OH (0.50 g, 1.08 mmol) was dissolved in tetrahydrofuran
2
(
(
(
THF) (1 mL) and the mixture added to a solution of NaH
0.10 g, 2.59 mmol, 60% suspension in mineral oil) in THF
3.5 mL) at 0 °C dropwise. MeI (0.22 mL, 3.56 mmol) was
1
3
32
=
7.6 Hz, 2H), 1.09 (t, J = 7.6 Hz, 3H).
Synthesis of 2-Acetylsulfanylmethylbutyric Acid. A mixture
added, and the reaction mixture was stirred at room
temperature for 12 h. MeOH and H O were then added to
2
of 2-ethylacrylic acid (0.27 g, 2.70 mmol) and thioacetic acid
0.73 mL, 10.26 mmol) in benzene (5 mL) was refluxed for 12
quench the reaction mixture, which was then acidified with HCl
(
(
1 N) to pH ∼5. The reaction mixture was extracted with
h. The solution was concentrated under reduced pressure,
diluted with EtOAc, and washed with HCl (1 N). The organic
layer was concentrated under reduced pressure to afford the
known 2-acetylsulfanylmethylbutyric acid (0.40 g, 83%) as a
yellow oil: H NMR (400 MHz, CDCl
literature ) δ 3.11 (dd, J = 13.6, 5.6 Hz, 1H), 3.01 (dd, J =
13.6, 8.8 Hz, 1H), 2.55 (p, J = 7.6 Hz, 1H), 2.31 (s, 3H), 1.73−
1.62 (m, 2H), 0.97 (t, J = 7.6 Hz, 3H).
EtOAc, and the organic layer was concentrated under reduced
pressure. The residue was dissolved in CH Cl (2 mL) followed
2
2
by addition of triisopropylsilane (1.11 mL, 5.40 mmol) and
TFA (2.15 mL, 28.08 mmol). The reaction mixture was stirred
at room temperature for 2 h. The reaction mixture was
1
, which matches the
3
32
evaporated and treated with H O (3 mL) and extracted with
2
Et O (3 × 3 mL). The reaction mixture was evaporated to
2
dryness under reduced pressure to give a residue, which was
C
DOI: 10.1021/acs.biochem.7b00980
Biochemistry XXXX, XXX, XXX−XXX

Biochemistry
Article
Synthesis of 2-Mercaptomethylbutyric Acid (MMB). 2-
Acetylsulfanylmethylbutyric acid (0.20 g, 1.13 mmol) was
with its MbtH-like protein partner TioT for which the cloning
20
was previously reported. pTioS(A M A )-pET28a was
3a 3S 3b
dissolved in HCl (4 mL, 6 N solution in H O). The resulting
transformed into chemically competent E. coli strain BL21
(DE3)ybdZ::aac(3)IV in which pTioT-pACYCDuet-1 was
previously transformed. This was cultured on LB agar
supplemented with 50 μg/mL kanamycin and 35 μg/mL
chloramphenicol. The protein was overexpressed in 3 × 1 L of
LB medium supplemented with 50 μg/mL kanamycin, 35 μg/
2
mixture was refluxed for 12 h. The reaction mixture was
extracted with EtOAc (3 × 5 mL), and the organic layer was
concentrated under reduced pressure to afford the known 2-
1
mercaptomethylbutyric acid (0.12 g, 80%) as a colorless oil: H
32
NMR (400 MHz, CDCl , which matches the literature ) δ
3
2.82−2.74 (m, 1H), 2.69−2.62 (m, 1H), 2.53 (p, J = 7.6 Hz,
1H), 1.77−1.63 (m, 2H), 1.52 (t, J = 8.8 Hz, 1H), 0.96 (t, J =
7.6 Hz, 3H).
mL chloramphenicol, and 10 mM MgCl . The transformant
was incubated in 3 × 5 mL of LB medium with the antibiotics
overnight. Each 5 mL culture was inoculated into 1 L of LB
2
Synthesis of 3-Mercaptopyruvic Acid (MP). Thioacetic acid
0.32 mL, 4.49 mmol) was added to a mixture of bromopyruvic
medium with the antibiotics and MgCl (10 mM) and grown at
2
(
37 °C and 200 rpm until the OD600 reached 0.6−0.8. Protein
expression was then induced with 0.2 mM isopropyl β-D-1-
thiogalactopyranoside (IPTG), and the bacterial culture was
incubated at 16 °C and 200 rpm for 20 h. The cells were
harvested by centrifugation at 5000 rpm for 10 min at 4 °C.
acid (0.50 g, 2.99 mmol) in THF (9 mL) at 0 °C dropwise.
Et N (0.50 mL, 3.59 mmol) was added, and the reaction
3
mixture was stirred at room temperature for 12 h. H O was
2
then added to quench the reaction, and the mixture was then
acidified with HCl (1 N) to pH ∼1. The reaction mixture was
extracted with EtOAc, and the organic layer was concentrated
under reduced pressure. The residue was dissolved in HCl (11
They were subsequently washed with H O and buffer A [25
2
mM Tris-HCl, 400 mM NaCl, and 10% glycerol (pH 8.0)]
supplemented with 5 mM imidazole followed by resuspension
in 30 mL of buffer A supplemented with 5 mM imidazole, 1
mM dithiothreitol (DTT), and 1 mM phenylmethanesulfonyl
fluoride (PMSF). This was lysed by sonication (four cycles of 2
min alternating with 2 s “on” and 10 s “off”), and the cell debris
was removed by centrifugation at 16000 rpm for 45 min at 4
°C. The supernatant was incubated with 0.75 mL of washed
mL, 6 N solution in H O). The reaction mixture was refluxed
2
for 12 h. The reaction mixture was treated with H O (10 mL)
2
at room temperature and extracted with EtOAc (3 × 10 mL).
The reaction mixture was evaporated to dryness under reduced
pressure to afford the known 3-mercaptopyruvic acid (0.25 g,
1
6
9%) as a yellow oil: H NMR (400 MHz, CDCl , which
3
33
II
matches the literature ) δ 4.41 (s, 2H), 4.17 (br s, 1H).
Ni -NTA agarose resin (Qiagen) at 4 °C for 2 h while being
Preparation of pTioS(A M A )-pET28a and pTioS-
gently shaken. The resin was loaded onto a column and washed
3a
3S 3b
(
T3)-pET28a Overexpression Constructs. The truncated
with 10 × 10 mL of buffer A supplemented with 40 mM
tioS genes, tioS(A M A ) and tioS(T ), were amplified from
imidazole, and TioS(A M A ) and TioT were co-eluted with
3a
3S 3b
3
3a 3S 3b
genomic DNA of Micromonospora sp. ML1 by PCR with the
forward and reverse primers tioS(A M A )-fwd and tioS-
3 × 5 mL of buffer A supplemented with 500 mM imidazole.
After sodium dodecyl sulfate−polyacrylamide gel electro-
phoresis (SDS−PAGE) analysis, the first two elution fractions
that contained the protein of interest were dialyzed in 3 × 2 L
of buffer B [40 mM Tris-HCl, 200 mM NaCl, 2 mM β-
mercaptoethanol, and 10% glycerol (pH 8.0)] at 4 °C for a
total of 20 h. The resulting solution was concentrated by using
Amicon Ultra-15 Centrifugal Filter Units (EMD Millipore,
Billerica, MA) with a 3 kDa molecular weight cutoff (MWCO)
membrane, and the protein concentration was spectrophoto-
metrically determined by using the calculated extinction
coefficient at 280 nm (http://protcalc.sourceforge.net/cgi-
3
a
3S 3b
(
A M A )-rev, respectively, for tioS(A M A ) or tioS(T )-
3a 3S 3b 3a 3S 3b 3
fwd and tioS(T )-rev, respectively, for tioS(T ) (Table S1).
3
3
PCRs were performed in reaction mixtures consisting of 0.25
μL of Phusion High-Fidelity DNA polymerase, 5 μL of Phusion
GC Buffer, 15 ng of Micromonospora sp. ML1 genomic DNA,
0.25 μL each of 50 μM forward and reverse primer, 0.5 μL of
1
0 mM dNTP, 1.5 μL of DMSO, and 16.25 μL of ddH O for a
2
total volume of 25 μL. The PCR conditions were optimized as
follows: initial denaturation for 30 s at 98 °C, 30 cycles of 30 s
at 98 °C, 30 s at 72 °C, and 3.5 min for tioS(A M A ) or 1.5
3
a
3S 3b
min for tioS(T ) at 72 °C, and a final extension for 10 min at 72
bin/protcalc). The yield of TioS(A M A ) with TioT was
3
3a 3S 3b
°
C. The DNA fragments of tioS(A M A ) and tioS(T ) were
1.34 mg/L of culture (Figure S2).
3
a
3S 3b
3
purified by agarose gel extraction using the QIAquick Gel
Extraction Kit (Qiagen, Valencia, CA), and they were digested
by restriction enzymes NdeI and HindIII. After being purified
by agarose gel extraction again, the DNA fragments were
ligated between the NdeI and HindIII multiple cloning sites of
expression plasmid vector pET28a (Novagen), which contains
a His6 tag at the 5′-end. The resulting plasmids pTioS-
Substrate Specificity and Determination of Kinetic
Parameters of the A Domains of TioS(A M A ) and
3
a
3S 3b
3
2
TioN(A M A ) by ATP−[ P]PP Exchange Assays. To
evaluate the substrate specificity and kinetics of the A domains
a
N
b
i
3
2
of TioS(A M A ) and TioN(A M A ), ATP−[ P]PP
3
a
3S 3b
a
N
b
i
exchange assays were performed. For the substrate profiles,
each 100 μL reaction {75 mM Tris-HCl, 5 mM TCEP, 10 mM
MgCl , 5 mM ATP, 1 mM Na P O with ∼400000 cpm of
(
A M A )-pET28a and pTioS(T )-pET28a were trans-
3a 3S 3b 3
2
4
2
7
32
formed into a chemically competent E. coli TOP10 strain and
cultured on Luria-Bertani (LB) agar supplemented with 50 μg/
mL kanamycin overnight. A colony of each construct was
subsequently cultured in 5 mL of LB medium with the
antibiotic overnight for plasmid isolation by the QIAprep Spin
Miniprep Kit (Qiagen). Successful cloning of the genes was
confirmed by sequencing reactions of the purified plasmids,
which match the sequence of UniProtKB entry Q333U7 for
tioS(A M A ) and tioS(T ).
[ P]PP , 5 mM substrate [L-Cys, N-Me-L-Cys, S-Me-L-Cys,
i
N,S-diMe-L-Cys, L-Ser, L-Thr, L-Ala, L-Gly, L-Val, L-Leu, L-Ile, L-
Pro, L-Phe, L-Tyr, L-Trp, L-Met, L-Asn, L-Gln, L-Lys, L-Arg, L-
His, L-Asp, L-Glu, 2-mercaptomethylbutyric acid (MMB), or 3-
mercaptopyruvic acid (MP)], and 1.0 μM TioS(A M A ) or
3a
3S 3b
2.5 μM TioN(A M A ), at pH 7.5} was initiated by adding the
a
N b
substrate and each mixture incubated for 2 h at room
temperature. The reactions were quenched by adding 500 μL
of the quenching solution [1.6% (w/v) activated charcoal, 4.5%
(w/v) Na P O , and 3.5% (v/v) perchloric acid in H O]. The
3a
3S 3b
3
Co-Overexpression and Purification of TioS-
4
2
7
2
(
A M A ) with TioT. TioS(A M A ) was overexpressed
charcoal was pelleted by centrifugation for 7 min at 14000 rpm
3a
3S 3b
3a 3S 3b
D
DOI: 10.1021/acs.biochem.7b00980
Biochemistry XXXX, XXX, XXX−XXX

Biochemistry
Article
Table 1. Steady-State Kinetic Parameters for AMP Derivatization by TioS(A M A ) or TioN(A M A )
3
a
3S 3b
a
N b
protein
substrate
L-Cys
Km (mM)
kcat (min−1)
k /K (mM−1 min−1
)
cat
m
TioS(A M A )
0.140 ± 0.003
2.2 ± 0.2
10.0 ± 0.7
0.037 ± 0.001
1.1 ± 0.2
7.72 ± 0.03
8.7 ± 0.3
6.5 ± 0.2
4.13 ± 0.02
5.3 ± 0.3
3.72 ± 0.05
55.4 ± 1.4
4.0 ± 0.4
0.66 ± 0.05
111 ± 4
5.0 ± 0.9
1.06 ± 0.04
3a
3S 3b
N-Me-L-Cys
S-Me-L-Cys
L-Cys
TioN(A M A )
a
N b
N-Me-L-Cys
S-Me-L-Cys
3.5 ± 0.1
and washed twice with 500 μL of the wash solution [4.5% (w/
onto a column and washed with 10 × 10 mL of buffer A
v) Na P O and 3.5% (v/v) perchloric acid in H O]. Each of
supplemented with 40 mM imidazole, and TioS(T ) was eluted
4
2
7
2
3
these was subsequently resuspended in 500 μL of H O and
with 3 × 5 mL of buffer A supplemented with 500 mM
imidazole. After SDS−PAGE analysis, the first two elution
fractions that contained the protein of interest were dialyzed in
3 × 2 L of buffer B at 4 °C for a total of 20 h. The resulting
solution was concentrated by using Amicon Ultra-15
Centrifugal Filter Units (EMD Millipore) with a 3 kDa
MWCO membrane, and the concentration was spectrophoto-
metrically determined by using the calculated extinction
coefficient at 280 nm (http://protcalc.sourceforge.net/cgi-
2
added to a scintillation vial containing 5 mL of scintillation
cocktail, whose radioactivity was counted by the liquid
scintillation analyzer. The substrate profiles are presented in
Figure 3.
The kinetic parameters of TioS(A M A ) and TioN-
3
a
3S 3b
(
A M A ) were calculated by the reaction rate with each
a N b
substrate (L-Cys, N-Me-L-Cys, and S-Me-L-Cys). The 100 μL
reactions [75 mM Tris-HCl, 5 mM TCEP, 10 mM MgCl , 5
2
32
mM ATP, 1 mM Na P O with ∼400000 cpm of [ P]PP ,
bin/protcalc). The yield of TioS(T ) was 4.29 mg/L of culture
4
2
7
i
3
varied concentrations of substrate (0, 0.05, 0.1, 0.25, 0.5, 1,
.75, 2.5, 5, 10, and ≤15 mM until the substrate starts inhibiting
the reaction), and 1.0 μM TioS(A M A ) or 2.5 μM
(Figure S6).
1
Characterization of TioS(T ) Activity by Trichloro-
3
acetic Acid (TCA) Precipitation Assays. To characterize the
3
a
3S 3b
TioN(A M A ), at pH 7.5] were initiated by adding the
activity of TioS(T ), conversion of the protein from its apo
a
N
b
3
substrate and the mixtures incubated at room temperature for
2 min for TioS(A M A ) and 15 min for TioN(A M A ),
(inactive) to holo (active) form was monitored in the presence
3
1
of Sfp (phosphopantetheinyltransferase) and [ H]AcCoA.
3
a
3S 3b
a
N b
3
where the adenylation rate is in a linear range. The reaction
mixtures were processed and counted by the same method as
that described for the substrate profile assay. Because the
kinetic parameters of TioN(A M A ) with N,S-diMe-L-Cys
Incorporation of [ H]acetyl into the TioS(T ) apo-protein
3
over time was assessed by using a TCA precipitation assay at
13
room temperature as previously described. The total 25 μL
reaction [75 mM Tris-HCl, 10 mM MgCl , 1 mM TCEP, 100
a
N
b
2
3
could not be determined because of enzyme inhibition by high
concentration of the substrate, the activity was measured in a
time course assay (Figure S5). Each 100 μL reaction [75 mM
μM AcCoA containing ∼160000 cpm of [ H]AcCoA, and 25
μM apo TioS(T ), at pH 7.5] was initiated by addition of Sfp
3
(1 μM). The reaction was quenched by adding 100 μL of a 10%
TCA solution at 0, 2, 5, 10, and 30 min. The precipitate was
pelleted by centrifugation at 14000 rpm for 7 min and washed
twice with 100 μL of a 10% TCA solution. After the final wash,
the pellets were dissolved in 100 μL of 88% formic acid. This
was transferred into liquid scintillation vials containing 5 mL of
the scintillation fluid, and the radiation activity was measured
with a liquid scintillation counter (Figure S7).
Tris-HCl, 5 mM TCEP, 10 mM MgCl , 5 mM ATP, 1 mM
2
3
2
Na P O with ∼400000 cpm of [ P]PP , 5 mM substrate, and
4
2
7
i
2
.5 μM TioN(A M A ), at pH 7.5] was initiated by adding the
a N b
substrate and each mixture incubated for 0, 0.5, 1, 2, 4, 8, and
2 h at room temperature. The reactions were quenched and
1
the mixtures processed by the same method as that described
for the substrate profile assay. All experiments were performed
in at least duplicate. The steady-state kinetic parameters are
listed in Table 1 and presented in Figures S3 and S4.
To verify loading of L-Cys or N- or S-Me-L-Cys onto holo
TioS(T ), the apo to holo conversion was first performed for
3
Overexpression and Purification of TioS(T ). pTioS-
T )-pET28a was transformed into chemically competent E.
3
TioS(T ) in a 12.5 μL reaction mixture (mixture A) for 30 min
as described above, but by using CoA instead of [ H]AcCoA.
3
3
3
(
coli BL21 (DE3), which was cultured on LB agar supplemented
with 50 μg/mL kanamycin. The transformant was incubated in
Simultaneously, the activation of L-Cys to L-Cys-AMP was
performed in a separate 12.5 μL reaction (mixture B) [75 mM
3
× 5 mL of LB medium with the antibiotic overnight. Each 5
Tris-HCl, 10 mM MgCl , 1 mM TCEP, 5 mM ATP, 5 μM
2
mL culture was inoculated into 1 L of LB with the antibiotic
TioS(A M A ) or TioN(A M A ), and 1 mM L-Cys
3
a
3S 3b
a
N b
3
5
and grown at 37 °C and 200 rpm until the OD600 reached 0.6−
containing ∼3500000 cpm of [ S]L-Cys per reaction, at pH
7.5]. After substrate activation for 2 h, final concentrations of 5
μM for A domain and 5 mM for SAM or the same volume of
0.8. Protein expression was then induced with 0.2 mM IPTG,
and the bacterial culture was incubated at 16 °C and 200 rpm
for 20 h. The cells were harvested by centrifugation at 5000
rpm for 10 min at 4 °C. They were subsequently washed with
H O was added to reaction mixture B, and the mixture was
2
incubated for an additional 2 h. The loading reaction was
initiated by combining reaction mixtures A and B and quenched
by adding 100 μL of a 10% TCA solution at 0, 2, 5, 10, 30, 60,
and 120 min. This was processed by the same method as
described above (Figure 4). To precisely evaluate the loading of
L-Cys in the presence and absence of SAM, the radioactivity of
H O and buffer A supplemented with 5 mM imidazole followed
2
by resuspension in 30 mL of buffer A supplemented with 5 mM
imidazole, 1 mM DTT, and 1 mM PMSF. This was lysed by
sonication (four cycles of 2 min alternating with 2 s “on” and
10 s “off”), and the cell debris was removed by centrifugation at
1
6000 rpm for 45 min at 4 °C. The supernatant was incubated
the reactions with different concentrations of TioS(T ) was
3
II
with 0.75 mL of washed Ni -NTA agarose resin (Qiagen) at 4
measured to draw V versus [TioS(T )] plots. The assay was
3
°C for 2 h while being gently shaken. The resin was loaded
performed under the same condition as described above, except
E
DOI: 10.1021/acs.biochem.7b00980
Biochemistry XXXX, XXX, XXX−XXX

Biochemistry
Article
for various concentrations of TioS(T ) (0, 25, 50, and 90 μM of
TioS(A M A ) or additional TioN(A M A ). The reaction
3
3a 3S 3b
a
N b
3
5
the final concentration) and the radioactivity of [ S]-L-Cys
∼400000 cpm). The reaction was quenched at 120 min and
the mixture processed by the same method as described above
was quenched at 0, 1, 2, 4, 8, 16, 24, and 36 h by addition of
100 μL of a 10% TCA solution, and the mixture was processed
by the same method as described above in Characterization of
(
(Figure 4C).
TioS(T ) Activity by Trichloroacetic Acid (TCA) Precipitation
3
Characterization of the Methyltransferase (M) Do-
Assays (Figure 6).
main Activity of TioS(A M A ) and TioN(A M A ) by
3a
3S 3b
a
N b
TCA Precipitation Assays. . To evaluate the M domain
RESULTS AND DISCUSSION
■
activity of TioS(A M A ) and TioN(A M A ), time course
3a
3S 3b
a
N b
Cloning, Heterologous (Co)expression, and Purifica-
tion of Enzymes Used in This Study. To evaluate the
proposed N- and S-methylating activity of TioS(A M A T )
TCA precipitation assays were performed by using L-Cys, N-
Me-L-Cys, and S-Me-L-Cys as substrates. The time course was
started after two sets of preincubation of reaction mixtures. The
first reaction mixture, mixture A, which was prepared for
adenylation of the substrate (L-Cys, N-Me-L-Cys, or S-Me-L-
Cys), contained 10 μL of a solution comprised of 93.75 mM
3a
3S 3b 3
and TioN(A M A ), respectively, and to determine if
a
N b
methylation occurs pre- or post-adenylation or pre- or post-
thiolation, we separately cloned, overexpressed, and purified
TioS(A M A ) and TioS(T ) (Figures S2 and S6). We also
3a
3S 3b
3
Tris-HCl, 12.5 mM MgCl , 1.25 mM TCEP, 3.125 mM ATP,
2
13
purified TioN(A M A ) as we previously described. It has
a
N b
3.125 mM substrate, and 2.5 μM A domain, at pH 7.5. This was
previously been shown that some A domains require an MbtH-
like protein partner, which binds to the A domain around its a6
incubated for 2 h for L-Cys, 4 h for N-Me-L-Cys, and 6 h for S-
Me-L-Cys. The second reaction mixture, mixture B, which was
3
4−36
12,20,37−39
and a7 region,
for their soluble expression,
prepared for apo to holo conversion of TioS(T ), contained 10
3
40−43
44
activity,
involved in thiocoraline and thiochondrilline A studied so far
TioK(AT) and TioN(A M A )] required the MbtH-like
and/or substrate specificity. The two A domains
μL of a solution comprised of 93.75 mM Tris-HCl, 12.5 mM
MgCl , 1.25 mM TCEP, 0.625 mM CoA, 125 μM TioS(T ),
2
3
[
a
N b
and 1.25 μM Sfp, at pH 7.5. Mixture B was incubated for 30
min before being mixed with mixture A to create reaction
mixture C for the loading of the activated substrate onto holo
protein partner TioT for their activity and/or soluble
13,20
expression.
We found that TioS(A M A ) was not an
3a 3S 3b
exception to this observation and that it, too, could be
expressed in a soluble and active form only when co-expressed
with TioT (Figure S2).
TioS(T ). Reaction mixture C was further incubated for 4 h
3
after the addition of extra A domain to maintain a
concentration of 2.5 μM in this solution. Five microliters of a
solution containing radioactive SAM (0.5 mM final concen-
Substrate Profile and Kinetic Characterization of the
3
A Domains of TioS(A3aM A3b) and TioN(A M A ). To
3S a N b
tration as in 25 μL with >3500000 cpm of [ H]SAM) was
assess if L-Cys is methylated [irrespective of the position (N or
S)] before or after being adenylated (Figure 2, pathways a−c
and g−i vs d−f and j−l), we first determined the substrate
added to reaction mixture C to start the methylation process.
This reaction was quenched at 0, 5, 10, 20, 30, 45, and 60 min
by adding 100 μL of a 10% TCA solution and the mixture
processed by the same method as described above in
32
specificity of TioS(A M A ) by an ATP−[ P]PP exchange
3a
3S 3b
i
assay by surveying 20 natural L-amino acids, three methylated L-
Cys derivatives (N-Me-, S-Me-, and N,S-diMe-L-Cys), and two
additional L-Cys analogues [2-mercaptomethylbutyric acid
Characterization of TioS(T ) Activity by Trichloroacetic Acid
3
(
TCA) Precipitation Assays (Figure 5).
To verify N-methylation at the secondary position of
(
MMB) and 3-mercaptopyruvic acid (MP)] (Figure 3A). We
nitrogen on L-Cys by TioS(A M A ), the following
3
a
3S 3b
methylation assay was performed. N,S-DiMe-L-Cys was at first
activated by TioN(A M A ) in 6.25 μL of reaction mixture D
a
N b
[
150 mM Tris-HCl, 20 mM MgCl , 2 mM TCEP, 4 mM ATP,
2
4
mM N,S-diMe-L-Cys, and 8 μM TioN(A M A ), at pH 7.5],
a N b
which was incubated overnight at room temperature. To
prepare holo TioS(T ), 11.25 μL of reaction mixture E [83.3
3
mM Tris, 11.1 mM MgCl , 1.1 mM TCEP, 556 μM CoA, 1.1
2
μM Sfp, and 192.2 μM TioS(T ), at pH 7.5] was incubated for
3
15 min at room temperature. Reaction mixtures D and E were
mixed together to create reaction mixture F before the addition
of another portion of TioN(A M A ) (50 μM in 1 μL; final
a
N b
concentration of 8 μM) and incubation for 5 h at room
temperature. SAM (12.5 mM in 1 μL; final concentration of
676 μM) was added in mixture F to diminish the level of
background methylation and incubated for an additional 4 h at
room temperature. During the last 9 h of incubation, small
amounts of TioN(A M A ) were added every hour to achieve a
a
N b
final concentration of 10 μM in the final reaction mixture.
3
[
H]SAM (1.5 μL, 900000 cpm) was added to reaction mixture
F followed by addition of 5 μL of 50 μM TioS(A M A ) or
3
a
3S 3b
Figure 3. Relative substrate specificity of the A domain portion of (A)
TioN(A M A ) (as a negative control) to start the reaction, for
a
N b
TioS(A M A ) and (B) TioN(A M A ) co-expressed with TioT
3
a
3S 3b
a
N b
a mixture that finally contained 75 mM Tris, 10 mM MgCl , 1
32
2
(2.5 μM), as determined by ATP−[ P]PP exchange assays at a 2 h
i
mM TCEP, 1 mM ATP, 1 mM N,S-diMe-L-Cys, 250 μM CoA,
end point. The substrates for which kinetic parameters were
determined (Table 1) are designated by the dark or light blue or
green bars.
3
0
8
.5 μM Sfp, 500 μM SAM (with 900000 cpm of [ H]SAM),
6.5 μM TioS(T ), 10 μM TioN(A M A ), and 10 μM
3
a
N b
F
DOI: 10.1021/acs.biochem.7b00980
Biochemistry XXXX, XXX, XXX−XXX

Biochemistry
Article
found that L-Cys was the best substrate for TioS(A M A )
whether methylation occurred pre- or post-thiolation (our
3
a
3S 3b
and that N-Me-L-Cys and S-Me-L-Cys could also be activated at
8 and 56%, respectively, when compared to L-Cys. However,
N,S-diMe-L-Cys was not activated at all by TioS(A M A ).
question ib). We compared the loading onto TioS(T ) of the
3
35
35
9
activated [ S]-L-Cys (Figure 2, step f or l ), N-Me-[ S]-L-Cys
2 2
produced by TioS(A M A ) (Figure 2, step k ), and S-Me-
3
a
3S 3b
3a 3S 3b
3
35
These data show that N,S-dimethylation cannot happen prior
[ S]-L-Cys generated by TioN(A M A ) (Figure 2, step e ) in
a TCA precipitation time course assay using [ S]-L-Cys as the
starting point in the absence and presence of SAM (Figure
a N b 3
35
to adenylation by TioS(A M A ) and thereby exclude
3
a
3S 3b
pathways a and g. To test if L-Cys was the natural substrate
for TioS(A M A ), we determined Michaelis−Menten
4A,B). In this assay, TioS(T ) was at first posttranslationally
modified by a PPTase and CoA. T domains of NRPSs are
3
a
3S 3b
3
kinetic parameters for each substrate (Table 1 and Figure
S3). Although the maximum turnover rates (k ) were not
expressed in their apo form, and a PPTase is necessary to
cat
significantly different among the three substrates tested, their
affinities for this enzyme varied significantly, with the K of L-
m
Cys being the smallest, followed by that of N-Me-L-Cys (15-
fold higher than that of L-Cys) and then that of S-Me-L-Cys
(
70-fold higher than that of L-Cys). The overall catalytic
efficiency (k /K ) for L-Cys was 14- and 84-fold higher than
cat
m
those for N-Me-L-Cys and S-Me-L-Cys, respectively. These
kinetic data strongly suggest that neither N- nor S-methylation
occurs prior to adenylation, excluding pathways b, c, h, and i in
addition to pathways a and g that have been eliminated by the
substrate profile assay. These TioS(A M A ) results are also
3
a
3S 3b
consistent with our previous preliminary report showing that
methylation of L-Cys by TioN occurs post-adenylation,
providing a convincing answer (to question ia) that methylation
13
post-adenylation is favored over methylation pre-adenylation.
Although a substrate profile of TioN(A M A ) was already
a
N b
1
3
reported, it did not include N-Me-L-Cys or N,S-diMe-L-Cys.
Therefore, we performed ATP−[ P]PP exchange assays to
assess the substrate profile and kinetic parameters of TioN-
3
2
i
(
A M A ) for L-Cys and its methylated derivatives (Figure 3B,
a N b
Table 1, and Figure S4). Even though L-Cys and S-Me-L-Cys
13
had been previously tested, we used them again in this study
to ensure self-consistent experiments with the newly purified
TioN(A M A ) enzyme. Both the substrate profile and kinetic
a
N b
parameters for TioN(A M A ) showed trends similar to those
a
N b
observed with TioS(A M A ). L-Cys remained the substrate
3
a
3S 3b
of choice for TioN(A M A ), followed by N-Me-L-Cys and S-
a
N b
Me-L-Cys, which were activated at 91 and 59%, respectively,
when compared to L-Cys. Interestingly, while adenylation of
N,S-diMe-L-Cys was basically undetectable with TioS-
(
A M A ), adenylation of this compound by TioN(A M A )
3a 3S 3b a N b
could be observed (7% activity compared to L-Cys). Like what
was observed with TioS(A M A ), the rank order of catalytic
3
a
3S 3b
efficiencies of TioN(A M A ) with L-Cys and its methylated
a
N b
derivatives (22- and 104-fold lower for N-Me-L-Cys and S-Me-
L-Cys, respectively, than for L-Cys) was determined by the
affinity of these substrates (29- and 95-fold higher for N-Me-L-
Cys and S-Me-L-Cys, respectively, than for L-Cys) for the
enzyme and did not heavily rely on the kcat values. Although
N,S-diMe-L-Cys was found to be a poor substrate of
TioN(A M A ), the steady-state kinetic parameters for its
a
N b
AMP derivatization could not be determined as the enzyme
function was inhibited by a high concentration of the substrate
(>5 mM), where the plot of Vmax versus [N,S-diMe-L-Cys]
remains linear (Figure S5). Overall, these series of ATP−
3
2
[
(
P]PP exchange assays with TioS(A M A ) and TioN-
i 3a 3S 3b
A M A ) unequivocally narrow down the possible pathway for
a
N b
the preparation of N,S-diMe-L-Cys-S-TioS(T ) to one in which
L-Cys is first converted to L-Cys-AMP by the A domain of
3
Figure 4. (A and B) Time courses and (C) TioS(T ) concentration-
dependent velocity to monitor the loading of [ S]-L-Cys-AMP onto
3
35
TioS(A M A ) (Figure 2, pathways d−f and j−l).
3
a
3S 3b
TioS(T ) in the absence (blue circles) or presence (orange circles) of
3
35
Assay of Loading of [ S]L-Cys onto TioS(T ) in the
35
3
SAM by TCA precipitation assays. [ S]-L-Cys was activated and
Presence or Absence of SAM. We next tested if loading of L-
loaded onto TioS(T ) by using TioS(A M A ) (A and C) or
3
3a 3S 3b
Cys onto TioS(T ) was affected by SAM, which could indicate
TioN(A M A ) (B).
3
a N b
G
DOI: 10.1021/acs.biochem.7b00980
Biochemistry XXXX, XXX, XXX−XXX

Biochemistry
Article
Figure 5. Methylation of L-Cys-S-TioS(T ), N-methyl-L-Cys-S-TioS(T ), and S-methyl-L-Cys-S-TioS(T ) by (A) TioS(A M A ) or (B)
3
3
3
3a 3S 3b
3
TioN(A M A ) with [ H]SAM as a methylating agent, measured by TCA precipitation assays.
a
N b
transfer the flexible Ppant arm from CoA to a conserved serine
residue of T domains to render them holo. We showed that
pathways (f and l), we next focused on determining if L-Cys
could be sequentially dimethylated by two independent
interrupted A domains (our question iia,b). To assess if
TioS(T ) could be converted from its apo to holo form by a
3
3
TCA precipitation assay using [ H]acetyl-CoA and the well-
known promiscuous PPTase Sfp (Figure S7). The separately
methylation of L-Cys-S-TioS(T ) first happens on the nitrogen
3
29
of the L-Cys backbone (Figure 2, step l ) or on the sulfur of its
3
activated holo TioS(T ) was mixed with a reaction solution in
side chain (Figure 2, step f ), methylation of L-Cys-S-TioS(T ),
3
3
3
which SAM (or H O as a control) was added after adenylation
N-Me-L-Cys-S-TioS(T ), and S-Me-L-Cys-S-TioS(T ), which
2
3
3
35
of the [ S]-L-Cys substrate by either TioS(A M A ) or
are the only possible M domain substrates remaining on the
basis of the experiments described above, by TioS(A3aM3SA3b)
or TioN(A M A ) was assessed by TCA precipitation assays
3
a
3S 3b
TioN(A M A ) (Figure 4). We measured the radioactivity of
a
N b
the loaded TioS(T ) in time courses for both enzymes (Figure
3
a
N b
3
4
A,B) or at 120 min for TioS(A M A ) with different
using [ H]SAM (our question iib). After activation by the A
domain portions of TioS(A M A ) or TioN(A M A ), the
3
a
3S 3b
concentrations of the T domain to calculate the velocity of the
3a
3S 3b
a
N b
loading activity at each TioS(T ) concentration (Figure 4C).
substrate AMPs were loaded onto holo TioS(T ). After loading
3
3
35
We observed that [ S]-L-Cys was loaded onto TioS(T ) with
for 4 h, when we assumed that TioS(T ) was fully modified
3
3
similar efficiencies in the presence and absence of SAM. This
experiment suggests that the loading step is fast and likely
proceeds with L-Cys-AMP prior to its methylation (which
cannot be observed directly in this experiment). These data
indicate that pathways f and l are favored over pathways d, e, j,
and k. This order is consistent with the recent crystal structure
of the adenylation−formylation−thiolation tridomain, where
the thiolation domain loads the amino acid residue (by using
with the expected L-Cys, N-Me-L-Cys, or S-Me-L-Cys species,
radioactive [ H]SAM was added to the reaction mixture to
3
measure the methylating activity in time course assays (Figure
5). Not surprisingly, we found that both TioS(A M A )
3
a
3S 3b
(Figure 5A) and TioN(A M A ) (Figure 5B) were capable of
a
N b
methylating their natural substrate L-Cys covalently attached to
TioS(T ) (Figure 5, top reactions). Also, as predicted, we
3
observed that S-Me-L-Cys-S-TioS(T ) was not a substrate for
3
45
the adenylated residue as a substrate) prior to its formylation.
additional methylation by TioN(A M A ), confirming that this
a
N b
Methylation Assay of L-Cys, N-Me-L-Cys, and S-Me-L-
Cys with [ H]SAM. Having narrowed down the biosynthesis
enzyme is an S-methyltransferase (Figure 5B and bottom right
3
reaction). However, when it was reacted with TioS(A M A ),
3
a
3S 3b
of N,S-diMe-L-Cys-S-TioS(T ) from 12 to two possible
we observed that S-Me-L-Cys-S-TioS(T ) could be further N-
3
3
H
DOI: 10.1021/acs.biochem.7b00980
Biochemistry XXXX, XXX, XXX−XXX

Biochemistry
Article
methylated, albeit not as effectively as L-Cys-S-TioS(T3)
(Figure 5A and bottom left reaction). Similarly, we found
that N-Me-L-Cys-S-TioS(T ) could be further S-methylated by
3
TioN(A M A ), but in this case, the additional methylation
a
N b
proceeded as efficiently as that of L-Cys-S-TioS(T ) (Figure 5B
3
and middle right reaction). These data strongly suggest that
TioS(A M A ) and TioN(A M A ) can sequentially and
3a
3S 3b
a
N b
specifically methylate the backbone and then the side chain of
L-Cys while it is tethered to TioS(T ), not only providing
3
answers to our question iia,b, but also pointing to pathway l in
Figure 2 as the favored sequence of events followed by Nature
for the biosynthesis of N,S-diMe-L-Cys-S-TioS(T ). Although
3
this is only speculation (as there are currently no available
structures of A domains interrupted by M domains), the fact
that TioS(A M A ) is physically connected to TioS(T ) in
3a
3S 3b
3
the thiocoraline/thiochondrilline A gene cluster, while TioN is
a stand-alone enzyme, could potentially explain two intriguing
observations. The first is why the M domain of TioS-
(
A M A ), which should be closer in space and more
3a 3S 3b
available, could act first to N-methylate the L-Cys attached to
TioS(T ). The second is why the M domain of TioN-
3
(
A M A ), which is in its naturally occurring form, has an
a N b
activity higher than that of TioS(A M A ), which likely has a
3
a
3S 3b
relatively larger loss of activity because of its excision from the
TioS protein.
Figure 6. Methylation of N,S-dimethyl-L-Cys-S-TioS(T
) by TioS-
3
(
A M A ) (blue circles) or TioN(A M A ) (white circles, as a
While evaluating the ability of TioS(A M A ) to further
3a 3S 3b a N b
3
a
3S 3b
3
negative control) with [ H]SAM as a methylating agent, measured by
TCA precipitation assays.
methylate N-Me-L-Cys-S-TioS(T ), we made the unexpected
3
and puzzling observation that this substrate could be further
methylated by the enzyme (Figure 5A and middle left
reaction). This observation could indicate that TioS-
(
A M A ) is capable either of both backbone and side-
3a 3S 3b
and (iii) one interrupted A domain is not capable of
methylating both the backbone and side chain of a given
amino acid residue. However, we serendipitously discovered
that A domains interrupted between their a8 and a9 motifs by
M domains can perform amino acid backbone dimethylation.
This work provides a deeper understanding of how bifunctional
A/M domains function and lays the foundation for future
engineering of unique bi- and trifunctional interrupted A
domains to serve in the development of novel biologically
active natural product-derived compounds. Such work is
currently underway in our laboratory.
chain methylation to produce N,S-diMe-L-Cys-S-TioS(T ) on
its own or of dimethylation of the backbone to produce N,N-
diMe-L-Cys-S-TioS(T ); neither of these phenomena has ever
been reported.
3
3
3
Methylation Assay of N,S-DiMe-L-Cys with [ H]SAM.
To solve the TioS(A M A ) dimethylation puzzle described
above, we utilized N,S-diMe-L-Cys-S-TioS(T ) in time course
methylation assays (Figure 6). We took advantage of the fact
that N,S-diMe-L-Cys can be activated by TioN(A M A ) at a
very slow rate (Figure S5) and loaded it onto TioS(T ) to
prepare the N,S-diMe-L-Cys-S-TioS(T ) used in these methyl-
3
a
3S 3b
3
a
N b
3
3
ation assays. Prior to studying the dimethylation by TioS-
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.bio-
chem.7b00980.
■
(
A M A ), although we suspected that TioN(A M A )
3a 3S 3b a N b
S
*
would not interfere in the assay because we established that
it does not N-methylate S-Me-L-Cys-S-TioS(T ) (Figure 5B
3
and bottom right reaction), we first confirmed that this was the
case by concomitantly performing the assay with TioN-
A table of primers used to clone the various constructs
for protein overexpression (Table S1), synthetic schemes
used for potential substrate synthesis (Figure S1), SDS−
PAGE gels for TioS(A M A ) (Figure S2) and
(
A M A ) and observing no activity (Figure 6, white circles).
a N b
We then performed the methylation assay with TioS-
A M A ) and N,S-diMe-L-Cys-S-TioS(T ) and found this
(
3
a
3S 3b
3
3
a
3S 3b
substrate to be N-methylated (Figure 6, blue circles). These
results clearly show that the second methylation performed by
TioS(A M A ), which we observed in Figure 5, occurs on the
TioS(T ) (Figure S6), Michaelis−Menten plots for
3
TioS(A M A ) (Figure S3) and TioN(A M A )
3
a
3S 3b
a
N b
3a
3S 3b
(
Figure S4), the time course for adenylation of N,S-
backbone of the amino acid and not on the sulfur atom of its
side chain.
diMe-L-Cys by TioN(A M A ) (Figure S5), and the apo
a
N b
to holo conversion of TioS(T ) (Figure S7) (PDF)
3
CONCLUSION
■
AUTHOR INFORMATION
Corresponding Author
E-mail: sylviegtsodikova@uky.edu.
In summary, this study revealed that (i) adenylation happens
prior to methylations, which themselves likely occur only after
loading of the activated amino acids onto T domains, (ii) two
interrupted A domains can work sequentially to methylate first
the backbone and then the side chain of a single amino acid,
■
*
ORCID
Sylvie Garneau-Tsodikova: 0000-0002-7961-5555
I
DOI: 10.1021/acs.biochem.7b00980
Biochemistry XXXX, XXX, XXX−XXX

Biochemistry
Article
Author Contributions
(17) Stachelhaus, T., Mootz, H. D., and Marahiel, M. A. (1999) The
specificity-conferring code of adenylation domains in nonribosomal
peptide synthetases. Chem. Biol. 6, 493−505.
S.M. and S.G.-T. designed the study. S.M., O.V.T., and S.G.-T.
analyzed the data and wrote the manuscript. S.M. performed all
biochemical experiments. A.G. performed chemical synthesis.
(
18) Mady, A. S., Zolova, O. E., Millan, M. A., Villamizar, G., de la
Calle, F., Lombo, F., and Garneau-Tsodikova, S. (2011) Character-
́
Funding
ization of TioQ, a type II thioesterase from the thiocoraline
biosynthetic cluster. Mol. BioSyst. 7, 1999−2011.
This project was supported by National Science Foundation
CAREER Award MCB-1149427 (to S.G.-T.) and by start-up
funds from the College of Pharmacy at the University of
Kentucky (to S.G.-T.).
(19) Sheoran, A., King, A., Velasco, A., Pero, J. M., and Garneau-
Tsodikova, S. (2008) Characterization of TioF, a tryptophan 2,3-
dioxygenase involved in 3-hydroxyquinaldic acid formation during
thiocoraline biosynthesis. Mol. BioSyst. 4, 622−628.
Notes
(20) Zolova, O. E., and Garneau-Tsodikova, S. (2012) Importance of
the MbtH-like protein TioT for production and activation of the
thiocoraline adenylation domain of TioK. MedChemComm 3, 950−
The authors declare no competing financial interest.
9
55.
REFERENCES
■
(21) Zolova, O. E., Mady, A. S., and Garneau-Tsodikova, S. (2010)
(
1) Newman, D. J., and Cragg, G. M. (2016) Natural products as
Recent developments in bisintercalator natural products. Biopolymers
93, 777−790.
(22) Biswas, T., Zolova, O. E., Lombo, F., de la Calle, F., Salas, J. A.,
́
sources of new drugs from 1981 to 2014. J. Nat. Prod. 79, 629−661.
(
2) Weissman, K. J. (2015) The structural biology of biosynthetic
megaenzymes. Nat. Chem. Biol. 11, 660−670.
3) Fischbach, M. A., and Walsh, C. T. (2006) Assembly-line
Tsodikov, O. V., and Garneau-Tsodikova, S. (2010) A new scaffold of
an old protein fold ensures binding to the bisintercalator thiocoraline.
J. Mol. Biol. 397, 495−507.
(
enzymology for polyketide and nonribosomal peptide antibiotics:
logic, machinery, and mechanisms. Chem. Rev. 106, 3468−3496.
(23) Mori, S., Shrestha, S. K., Fernandez, J., Alvarez San Millan, M.,
(
4) Walsh, C. T. (2016) Insights into the chemical logic and
́
Garzan, A., Al-Mestarihi, A. H., Lombo, F., and Garneau-Tsodikova, S.
(2017) Activation and loading of the starter unit during thiocoraline
biosynthesis. Biochemistry 56, 4457−4467.
enzymatic machinery of NRPS assembly lines. Nat. Prod. Rep. 33,
27−135.
5) Park, H. B., Kim, Y. J., Park, J. S., Yang, H. O., Lee, K. R., and
1
(
(24) McQuade, T. J., Shallop, A. D., Sheoran, A., Delproposto, J. E.,
Tsodikov, O. V., and Garneau-Tsodikova, S. (2009) A nonradioactive
high-throughput assay for screening and characterization of
adenylation domains for nonribosomal peptide combinatorial biosyn-
thesis. Anal. Biochem. 386, 244−250.
(25) Garneau-Tsodikova, S., Dorrestein, P. C., Kelleher, N. L., and
Walsh, C. T. (2006) Protein assembly line components in prodigiosin
biosynthesis: characterization of PigA, G,H, I,J. J. Am. Chem. Soc. 128,
12600−12601.
(26) Tsodikov, O. V., Hou, C., Walsh, C. T., and Garneau-Tsodikova,
S. (2015) Crystal structure of O-methyltransferase CalO6 from the
calicheamicin biosynthetic pathway: a case of challenging structure
determination at low resolution. BMC Struct. Biol. 15, 13.
(27) Perez Baz, J., Canedo, L. M., Fernandez Puentes, J. L., and Silva
Elipe, M. V. (1997) Thiocoraline, a novel depsipeptide with antitumor
activity produced by a marine Micromonospora. II. Physico-chemical
properties and structure determination. J. Antibiot. 50, 738−741.
(28) Wyche, T. P., Hou, Y., Braun, D., Cohen, H. C., Xiong, M. P.,
and Bugni, T. S. (2011) First natural analogs of the cytotoxic
thiodepsipeptide thiocoraline A from a marine Verrucosispora sp. J. Org.
Chem. 76, 6542−6547.
(29) Quadri, L. E., Weinreb, P. H., Lei, M., Nakano, M. M., Zuber, P.,
and Walsh, C. T. (1998) Characterization of Sfp, a Bacillus subtilis
phosphopantetheinyl transferase for peptidyl carrier protein domains
in peptide synthetases. Biochemistry 37, 1585−1595.
(30) Liu, J.-F., Tang, X.-X., and Jiang, B. (2002) A convenient
synthesis of N-fluorenylmethoxycarbonyl-N-methyl-l-cysteine deriva-
tives [Fmoc,Me-Cys(R)-OH]. Synthesis 11, 1499−1501.
(31) Nudelman, A., Marcovici-Mizrahi, D., Nudelman, A., Flint, D.,
and Wittenbach, V. (2004) Inhibitors of biotin biosynthesis as
potential herbicides. Tetrahedron 60, 1731−1748.
(32) Lee, H. S., and Kim, D. H. (2003) Synthesis and evaluation of
alpha,alpha-disubstituted-3-mercaptopropanoic acids as inhibitors for
carboxypeptidase A and implications with respect to enzyme inhibitor
design. Bioorg. Med. Chem. 11, 4685−4691.
Kwon, H. C. (2011) Glionitrin B, a cancer invasion inhibitory
diketopiperazine produced by microbial coculture. J. Nat. Prod. 74,
2
(
309−2312.
6) Bhuiya, M. W., and Liu, C. J. (2010) Engineering monolignol 4-
O-methyltransferases to modulate lignin biosynthesis. J. Biol. Chem.
85, 277−285.
7) Zhang, Q., and van der Donk, W. A. (2012) Catalytic promiscuity
of a bacterial alpha-N-methyltransferase. FEBS Lett. 586, 3391−3397.
8) Itoh, N., Iwata, C., and Toda, H. (2016) Molecular cloning and
2
(
(
characterization of a flavonoid-O-methyltransferase with broad
substrate specificity and regioselectivity from Citrus depressa. BMC
Plant Biol. 16, 180.
(
9) Kim, E., Song, M. C., Kim, M. S., Beom, J. Y., Lee, E. Y., Kim, D.
M., Nam, S. J., and Yoon, Y. J. (2016) Characterization of the two
methylation steps involved in the biosynthesis of mycinose in tylosin. J.
Nat. Prod. 79, 2014−2021.
(
10) Singh, S., Zhang, J., Huber, T. D., Sunkara, M., Hurley, K., Goff,
R. D., Wang, G., Zhang, W., Liu, C., Rohr, J., Van Lanen, S. G., Morris,
A. J., and Thorson, J. S. (2014) Facile chemoenzymatic strategies for
the synthesis and utilization of S-adenosyl-(L)-methionine analogues.
Angew. Chem., Int. Ed. 53, 3965−3969.
(
11) Labby, K. J., Watsula, S. G., and Garneau-Tsodikova, S. (2015)
Interrupted adenylation domains: unique bifunctional enzymes
involved in nonribosomal peptide biosynthesis. Nat. Prod. Rep. 32,
6
(
41−653.
12) Zolova, O. E., and Garneau-Tsodikova, S. (2014) KtzJ-
dependent serine activation and O-methylation by KtzH for
kutznerides biosynthesis. J. Antibiot. 67, 59−64.
(
13) Al-Mestarihi, A. H., Villamizar, G., Fernandez, J., Zolova, O. E.,
́
Lombo, F., and Garneau-Tsodikova, S. (2014) Adenylation and S-
methylation of cysteine by the bifunctional enzyme TioN in
thiocoraline biosynthesis. J. Am. Chem. Soc. 136, 17350−17354.
(
14) Shrestha, S. K., and Garneau-Tsodikova, S. (2016) Expanding
substrate promiscuity by engineering a novel adenylating-methylating
NRPS bifunctional enzyme. ChemBioChem 17, 1328−1332.
(33) de Castro, M. V., Ioca, L. P., Williams, D. E., Costa, B. Z.,
Mizuno, C. M., Santos, M. F., de Jesus, K., Ferreira, E. L., Seleghim, M.
H., Sette, L. D., Pereira Filho, E. R., Ferreira, A. G., Goncalves, N. S.,
Santos, R. A., Patrick, B. O., Andersen, R. J., and Berlinck, R. G. (2016)
Condensation of macrocyclic polyketides produced by Penicillium sp.
DRF2 with mercaptopyruvate represents a new fungal detoxification
pathway. J. Nat. Prod. 79, 1668−1678.
(
15) Schwarzer, D., Finking, R., and Marahiel, M. A. (2003)
Nonribosomal peptides: from genes to products. Nat. Prod. Rep. 20,
75−287.
16) Conti, E., Stachelhaus, T., Marahiel, M. A., and Brick, P. (1997)
2
(
Structural basis for the activation of phenylalanine in the non-
ribosomal biosynthesis of gramicidin S. EMBO J. 16, 4174−4183.
J
DOI: 10.1021/acs.biochem.7b00980
Biochemistry XXXX, XXX, XXX−XXX

Biochemistry
Article
(
34) Herbst, D. A., Boll, B., Zocher, G., Stehle, T., and Heide, L.
(
2013) Structural basis of the interaction of MbtH-like proteins,
putative regulators of nonribosomal peptide biosynthesis, with
adenylating enzymes. J. Biol. Chem. 288, 1991−2003.
(
35) Miller, B. R., Drake, E. J., Shi, C., Aldrich, C. C., and Gulick, A.
M. (2016) Structures of a nonribosomal peptide synthetase module
bound to MbtH-like proteins support a highly dynamic domain
architecture. J. Biol. Chem. 291, 22559−22571.
(
36) Tarry, M. J., Haque, A. S., Bui, K. H., and Schmeing, T. M.
(
2017) X-Ray crystallography and electron microscopy of cross- and
multi-module nonribosomal peptide synthetase proteins reveal a
flexible architecture. Structure 25, 783−793.
(
37) Heemstra, J. R., Jr., Walsh, C. T., and Sattely, E. S. (2009)
Enzymatic tailoring of ornithine in the biosynthesis of the Rhizobium
cyclic trihydroxamate siderophore vicibactin. J. Am. Chem. Soc. 131,
1
(
(
5317−15329.
38) Imker, H. J., Krahn, D., Clerc, J., Kaiser, M., and Walsh, C. T.
2010) N-acylation during glidobactin biosynthesis by the tridomain
nonribosomal peptide synthetase module GlbF. Chem. Biol. 17, 1077−
083.
39) McMahon, M. D., Rush, J. S., and Thomas, M. G. (2012)
1
(
Analyses of MbtB, MbtE, and MbtF suggest revisions to the
mycobactin biosynthesis pathway in Mycobacterium tuberculosis. J.
Bacteriol. 194, 2809−2818.
(
40) Boll, B., Taubitz, T., and Heide, L. (2011) Role of MbtH-like
proteins in the adenylation of tyrosine during aminocoumarin and
vancomycin biosynthesis. J. Biol. Chem. 286, 36281−36290.
(
41) Zhang, C., Kong, L., Liu, Q., Lei, X., Zhu, T., Yin, J., Lin, B.,
Deng, Z., and You, D. (2013) In vitro characterization of echinomycin
biosynthesis: formation and hydroxylation of l-tryptophanyl-S-enzyme
and oxidation of (2S,3S) beta-hydroxytryptophan. PLoS One 8,
e56772.
(
42) Zhang, W., Heemstra, J. R., Jr., Walsh, C. T., and Imker, H. J.
2010) Activation of the pacidamycin PacL adenylation domain by
MbtH-like proteins. Biochemistry 49, 9946−9947.
43) Felnagle, E. A., Barkei, J. J., Park, H., Podevels, A. M., McMahon,
(
(
M. D., Drott, D. W., and Thomas, M. G. (2010) MbtH-like proteins as
integral components of bacterial nonribosomal peptide synthetases.
Biochemistry 49, 8815−8817.
(
44) Davidsen, J. M., Bartley, D. M., and Townsend, C. A. (2013)
Nonribosomal propeptide precursor in nocardicin A biosynthesis
predicted from adenylation domain specificity dependent on the
MbtH family protein NocI. J. Am. Chem. Soc. 135, 1749−1759.
(
45) Reimer, J. M., Aloise, M. N., Harrison, P. M., and Schmeing, T.
M. (2016) Synthetic cycle of the initiation module of a formylating
nonribosomal peptide synthetase. Nature 529, 239−242.
K
DOI: 10.1021/acs.biochem.7b00980
Biochemistry XXXX, XXX, XXX−XXX