Mechanistic Studies on CysS - A Vitamin B12-Dependent Radical SAM Methyltransferase Involved in the Biosynthesis of the tert-Butyl Group of Cystobactamid

Article abstract of DOI:10.1021/jacs.9b06454

Cobalamin (Cbl)-dependent radical S-adenosylmethionine (SAM) methyltransferases catalyze methylation reactions at non-nucleophilic centers in a wide range of substrates. CysS is a Cbl-dependent radical SAM methyltransferase involved in cystobactamid biosynthesis. This enzyme catalyzes the sequential methylation of a methoxy group to form ethoxy, i-propoxy, s-butoxy, and t-butoxy groups on a p-aminobenzoate peptidyl carrier protein thioester intermediate. This biosynthetic strategy enables the host myxobacterium to biosynthesize a combinatorial antibiotic library of 25 cystobactamid analogues. In this Article, we describe three experiments to elucidate how CysS uses Cbl, SAM, and a [4Fe-4S] cluster to catalyze iterative methylation reactions: a cyclopropylcarbinyl rearrangement was used to trap the substrate radical and to estimate the rate of the radical substitution reaction involved in the methyl transfer; a bromoethoxy analogue was used to explore the active site topography; and deuterium isotope effects on the hydrogen atom abstraction by the adenosyl radical were used to investigate the kinetic significance of the hydrogen atom abstraction. On the basis of these experiments, a revised mechanism for CysS is proposed.

Full text of DOI:10.1021/jacs.9b06454

Subscriber access provided by the University of Exeter
Article
12
Mechanistic studies on CysS – a Vitamin B-dependent
radical SAM methyltransferase involved in the
biosynthesis of the t-butyl group of cystobactamid
Yuanyou Wang, and Tadhg P. Begley
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.9b06454 • Publication Date (Web): 06 May 2020
Downloaded from pubs.acs.org on May 8, 2020
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 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 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.
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 14
Journal of the American Chemical Society
1
2
3
4
5
6
7
–
Mechanistic studies on CysS a Vitamin B12-dependent radical
SAM methyltransferase involved in the biosynthesis of the t-butyl
group of cystobactamid
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
Yuanyou Wang and Tadhg P. Begley*
Department of Chemistry, Texas A&M University, College Station, Texas 77843, United States
Supporting Information Placeholder
ABSTRACT: Cobalamin (Cbl)-dependent radical SAM methyltransferases catalyze methylation reactions at non-nucle-
ophilic centers in a wide range of substrates. CysS is a Cbl-dependent radical SAM methyltransferase involved in cysto-
bactamid biosynthesis. This enzyme catalyzes the sequential methylation of a methoxy group to form ethoxy, i-
propoxy, s-butoxy, and t-butoxy groups on a p-aminobenzoate peptidyl carrier protein thioester intermediate. This bio-
synthetic strategy enables the host myxobacterium to biosynthesize a combinatorial antibiotic library of 25 cystobac-
tamid analogs. In this paper, we describe three experiments to elucidate how CysS uses Cbl, SAM, and a [4Fe− 4S] clus-
ter to catalyze iterative methylation reactions: a cyclopropylcarbinyl rearrangement was used to trap the substrate radi-
cal and to estimate the rate of the radical substitution reaction involved in the methyl transfer; a bromoethoxy analog
was used to explore the active site topography; and deuterium isotope effects on the hydrogen atom abstraction by the
adenosyl radical were used to investigate the kinetic significance of the hydrogen atom abstraction. Based on these ex-
periments, a revised mechanism for CysS is proposed.
INTRODUCTION
O
SR
O
SR
O
SR
Cobalamin (Cbl)-dependent radical SAM methyltrans-
ferases catalyze methylation reactions at non-nucleophilic
CysS
CysS
H
C
CH3
H
H
CH3
CH3
H
H
O
O
O
H
1
2
3
NH2
NH2
NH2
1-5
centers in a wide range of substrates. Selected examples
of characterized systems are shown in Figure 1. The CysS-
catalyzed methylation of 1 is the subject of this paper.
The cystobactamid biosynthetic gene cluster has been
identified and biochemical studies have demonstrated
that CysS encodes a cobalamin-dependent radical SAM
methyltransferase involved in the iterative methylations
of 3-methoxy-4-aminobenzoic acid attached via a thioe-
ster to the peptidyl carrier protein (CysG1, Figure 1, 1 to
O
SR
O
SR
CysS
CH3
CH3
CH3
CH3
+
CH3
O
O
NH2
4a
NH2
4b
O
O
P
PhpK
P
R
R
H
H3C
O
O
NHAc
NHAc
5
6
H
OH
H3C
HO
OH
O
O
HO
NH2
NH2
NH2
NH2
GenK
H2N
HO
O
H2N
HO
O
6-8
4). Cystobactamids have antibacterial activity in the
O
O
O
O
OH
OH
6,
9
low-µg/mL range against Gram-negative pathogens.
This biosynthetic strategy enables the host myxobacte-
HO
HO
NHMe
NHMe
CH3
7
8
rium to biosynthesize a combinatorial antibiotic library of
O
P
O
P
Fom3
9
RO
OH
RO
OH
10
25 cystobactamid analogs.
O
9
O
While sequence analysis indicates that Cbl-dependent
O
O
OH
NH2
OH
radical SAM methyltransferases are widely used in bio-
NH2
CH3
10
synthesis, it was not until recently that enzymes in this
TsrM
N
H
N
H
class were reconstituted in defined biochemical sys-
11
12
11-14
tems.
Detailed mechanistic studies are still at an early
H
H
N
H
H
N
stage because of substrate complexity and the low yield
ThnK
SR
SR
OH
of pure active enzyme available from most overexpres-
O
O
OH
13
14
15-17
O
O
O
sion systems.
No Cbl-dependent radical SAM methyl-
transferase has yet been structurally characterized.
PoyC
H
H
R
N
R
N
N
R'
N
H
R'
H
15
O
16
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 14
19-21
Figure 1. Selected examples of reconstituted Cbl-dependent
radical SAM methyltransferases. The CysS catalyzed itera-
tive methylation of the methoxy group of 1. PhpK catalyzes
the conversion of 5 to 6 and occurs on the biosynthesis of the
indole methylation (11 to 12).
ThnK catalyzes an iterative
1
2
3
4
C-methylation to assemble an ethyl group during the biosyn-
thesis of the antibiotic thienamycin (13 to 14).
lyzes valine Cβ-Methylation during polytheonamide biosyn-
17, 22
PoyC cata-
15
23
broad-spectrum herbicide phosphinothricin. GenK cata-
thesis (15 to 16).
5
6
7
8
lyzes the C-alkylation of an alcohol (7 to 8) during the bio-
synthesis of the antibiotic gentamicin, and Fom 3 catalyzes a
similar reaction in the biosynthesis of the antibiotic fosfomy-
11-14, 18
cin.
Nosiheptide biosynthesis uses TsrM to catalyze an
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
C
D
A
B
Co(III)
CH3
Co(I)
Co(I)
O
SR
1,e-
Co(III)
CH3
Met
Me
S
H
C
SAM
S
Ade
Ade
H
H
Me
R'
R'
O
O
O
SAH
1
NH2
OH OH
SAH
OH OH
17
S
Ade
R'
[4Fe-4S]1+
18
O
SAM
[4Fe-4S]2+
[4Fe-4S]2+
OH OH
SAM
[4Fe-4S]2+
17
G
F
A
E
O
SR
O
SR
Co(III)
CH3
O
SR
Co(II)
e-
Co(III)
CH3
H
Co(I)
CH3
H
H
C
C
H
H
O
O
H
O
H
1
NH2
2, 5'dA
NH2
21
NH2
CH3
Ade
CH3
O
Ade
2
CH2
Ade
O
O
OH OH
5'dA
[4Fe-4S]2+
OH OH
5'dA 20
[4Fe-4S]2+
[4Fe-4S]2+
[4Fe-4S]2+
OH OH
5'dA
20
19
O
SR
O
SR
O
SR
O
O
N
OH
R:
N
CH3
CH3
CH3
H
H
CH3
OH
CH3
CH3
+
CH3
O
O
O
COOH
R':
H
3
4a
NH2
NH2
NH2
4b
NH2
Figure 2. Mechanistic proposal for the CysS-catalyzed iterative methylation of 1 to form branched alkoxy groups.
2
ACS Paragon Plus Environment

Page 3 of 14
1
2
3
4
5
6
Journal of the American Chemical Society
Our previous mechanistic analysis of CysS is outlined in
3.2 mM NADPH. The reaction was incubated at room tem-
perature and quenched by ultrafiltration through a 10 kDa
cut-off filter (VWR). Components of the reaction were sepa-
rated either on a ZORBAX Eclipse XDB-C18 column (15 cm
x 4.6 mm, 5 µm particles, Agilent Technologies) for HPLC
analysis or on an LC-18-T column (15 cm x 3 mm, 3 µm par-
ticles, Supelco) for LC-MS/MS analysis. Detailed methods
and parameters for HPLC and LC-MS/MS are given in the
supporting information.
7
Figure 2. Initial formation of MeCbl (A to C) is followed by
generation of the 5′-deoxyadenosyl radical by reductive
cleavage of a second molecule of SAM (D to E). This radical
(5′-dA•) 19 then abstracts a hydrogen atom from the me-
thyl group of the substrate and the resulting substrate radical
undergoes a radical substitution (SHi) with MeCbl (F to G) to
7
8
9
give 2. Radical substitutions are widely used in organic syn-
thesis
carbon-based radical has been identified in a Cbl model sys-
24-27
and a methyl group transfer from MeCbl(III) to a
Analysis of the CysS-catalyzed reaction of the cyclopropyl
methoxy substrate analog 26. The cyclopropylmethoxy ana-
log 26 was synthesized as shown in Figure S24. The CysS-
catalyzed reaction with 26 was carried out as described
above using methyl viologen/NADPH as the reducing agent.
The reaction was quenched by ultrafiltration after 1, 2, 4, 6,
and 15 hrs. The filtrates were treated with trifluoroacetic acid
(TFA, 0.1 M in 100 mM phosphate buffer, pH approx. = 2.4)
and analyzed by LC-MS/MS. Compounds 26 and 27 did not
undergo hydrolysis under these reaction conditions and the
vinyl ether hydrolysis reaction reached completion after 200
min. (Figure S2, S3). Reference samples of the cyclopro-
pylethoxy product 27 and aminophenol 28 were synthesized
as shown in Figures S25 and S30 and used to identify the re-
action products by co-elution. Both the cyclopropylethoxy
product 27 and aminophenol 28 were detected in the 4, 6, and
15 hr reactions and the product ratio was determined by cal-
ibrated LC-MS/MS analysis (Figure S1, Table S1).
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
28
tem. Radical substitution mechanisms have been proposed
for the GenK, ThnK, Fom3 and PoyC-catalyzed reactions, but
the proposed radical has not yet been trapped. A non-radical
mechanism has been proposed for TsrM. Reactivation of
CysS by a 1-electron reduction of the Cbl(II), possibly medi-
19
+1
ated by the [4Fe− 4S] cluster, completes the chemistry re-
quired for the first C-C bond formation. Repetition of this se-
quence results in the formation of the i-propoxy, t-butoxy,
and s-butoxy substituted products (3, 4a, 4b).
In this paper, we describe three experiments to further elu-
cidate how CysS uses Cbl, SAM, and a [4Fe−4S]cluster to cat-
alyze a sequence of iterative methylation reactions: a cyclo-
propylcarbinyl rearrangement of 26 was used to trap a sub-
strate radical and to estimate the rate of the radical substitu-
tion reaction involved in the methyl transfer (F to G); a bro-
moethoxy analog 35 was used to explore the active site to-
pography; and deuterium isotope effects on the hydrogen
atom abstraction by the adenosyl radical were used to inves-
tigate the kinetic significance of the hydrogen atom abstrac-
tion in the early steps (A to F) of the reaction scheme out-
lined in Figure 2. Detailed mechanistic studies of cobalamin-
dependent radical SAM methyltransferases are still at an
early stage. While CysS is a relatively well-behaved experi-
mental system, enzyme activity varies between preparations
precluding detailed kinetics studies. Our mechanistic strat-
egy has therefore relied on the use of substrate analogs and
competition reactions in which two substrates are processed
in a single reaction mixture.
Analysis of the CysS-catalyzed reaction of the bromoeth-
oxy substrate analog 35. The bromoethoxy analog 35 was
synthesized as shown in Figure S26. CysS (90 µM), 1 mM
HOCbl, 1 mM SAM, 1.2 mM 35, 1.2 mM methyl viologen and
3.5 mM NADPH were incubated at room temperature in an
anaerobic chamber in the dark for 12-15 hrs. The enzymatic
reaction was then quenched by ultrafiltration and analyzed
by HPLC and LC-MS/MS as described above.
Synthesis of the Cbl+35 adduct 36. Cbl(III) was first reduced
to Cbl(I), under anaerobic conditions, using an excess of
29
NaBH4. This reaction mixture had a UV-vis spectrum con-
sistent with the formation of Cbl(I) (Figure S6). The Cbl(I)
color immediately changed from light gray to pink after ad-
dition of 35, consistent with the formation of the Cbl+35 ad-
duct (36, Figure S6). HPLC and LC-MS/MS analysis of the re-
action mixture confirmed the formation of a new reaction
product with a mass and fragmentation pattern consistent
with the Cbl+35 adduct (36, Figure S7). Detailed procedures
are provided in the supporting information.
EXPERIMENTAL SECTION
CysS overexpression and purification. CysS was cloned into
a modified pET28b vector and co-expressed with a plasmid
encoding the suf operon for in vivo assembly of the [4Fe-4S]
7
cluster in E.coli BL21 (lDE3). CysS was also co-expressed
with the pBAD42-BtuCEDFB (cobalamin uptake system)
Characterization of the Cbl+35 adduct by derivatization.
16
plasmid and the suf plasmid in E. coli. CysS was over-ex-
30-31
For the photochemical derivatization,
the CysS reaction
8
pressed and purified as previously described.
mixture was exposed to ambient light, at room temperature,
General conditions for running and analyzing the CysS-
catalyzed reaction. All CysS enzymatic reactions were car-
ried out in an anaerobic Coy chamber (Grass Lake, MI, USA)
containing 95% nitrogen and 5% hydrogen. A typical enzy-
matic reaction was performed in 100 mM phosphate buffer,
pH 7.5 containing 55 µM of CysS with 0.5 mM MeCbl, 0.5
mM SAM, 0.5 mM substrate, 0.8 mM methyl viologen and
for 4 h and then analyzed by LC-MS/MS (Figure S10). For the
32
cyanide derivatization, the quenched enzymatic reaction
was treated with sodium cyanide (7 mM) for 2 h and then
analyzed by HPLC and LC-MS/MS. (Figure S13).
The Cbl+35 adduct formation is enzyme-catalyzed. For the
non-enzymatic reaction, HOCbl (0.3 mM) in MeOH was re-
duced with 10% v/v 1 M methanolic NaBH4 in an anaerobic
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 14
chamber. (Methyl viologen and NADPH will reduce Cbl(III)
For the intrinsic isotope effect measurement, CD2H-1 was
1
2
3
4
5
6
7
8
to Cbl(II). Further reduction to Cbl(I) requires the iron sulfur
cluster of reduced CysS.) After the color of the solution
turned grey, the excess of NaBH4 was quenched with water
until no additional bubbles were observed. Then, a mixture
of 35 and 1-bromo-2-methoxyethane 41 was added to a final
concentration of 0.5 mM each. The solution turned pink im-
mediately. LC-MS/MS analysis demonstrated the formation
of a mixture of Cbl(I) alkylated by 35 and 41 (Figure S15). The
ratio of the two products was quantitated by LC-MS/MS (Fig-
ure S16).
synthesized as shown in Figure S28. CysS (20 µM) was incu-
bated with 270 µM MeCbl, 540 µM SAM, 540 µM of CD2H-
1, 1 mM methyl viologen and 3 mM NADPH at room tem-
perature in an anaerobic chamber. The reaction was
quenched at different time points (30 min, 1 h, 2 h, 3 h, 4 h)
by ultrafiltration. The reaction mixtures were analyzed by
LC-MS/MS and the ratio of CH3CD2-2 (P1) to CH3CDH-2 (P2)
was measured from the peak intensity after correction for the
natural abundance of the stable isotopes (Figure S20).
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
O
SR
O
SR
For the enzymatic reaction, 100 µM of CysS was incubated
with 0.6 mM HOCbl, 0.5 mM bromoethoxy analog 35, 0.5
mM 1-bromo-2-methoxyethane 41, 0.8 mM methyl viologen,
and 3.2 mM NADPH at room temperature for 6 h and
quenched by ultrafiltration. LC-MS/MS analysis demon-
strated the formation of a mixture of Cbl(I) alkylated by 35
and 41. The ratio of the two products was quantitated by LC-
MS/MS and compared with the ratio of the adducts gener-
ated in the non-enzymatic reaction (Figure S16).
MeCbl
P1
D
D
O
SR
kH
kmt
D
C
D
C
O
O
Me
CH3
Ad
NH2
NH2
CH3CD2-2
D
D
C
O
O
SR
O
SR
H
NH2
kD
CH2
Ad
MeCbl
CD2H-1
D
C
D
P2
H
kmt
O
H
C
O
Me
CH3CDH-2
NH2
NH2
CH2D
Ad
kH
kD
2P1
P2
=
O
O
OH
R:
N
N
H
H
OH
SAM is a weak inhibitor of the CysS-catalyzed Cbl+35 ad-
duct formation. The yield of the Cbl+35 adduct was meas-
ured, by LC-MS/MS, in the presence and absence of SAM (1
Figure 4. Strategy for determining the intrinsic isotope ef-
fect on the CysS-catalyzed hydrogen atom abstraction.
+
mM) using the relative area under the [M + H] : 805.84 peak
CysS reactions with S-adenosyl-[methyl-d3]-methionine
(CD3-SAM) under single turnover conditions. 100 µM CysS
was incubated with 0.8 mM methyl viologen, 3.2 mM
NADPH, 0.2 mM CD3SAM, 10 µM ethyl pantetheinyl sub-
strate 2 and 0.6 mM MeCbl at room temperature. The reac-
tions were quenched by ultrafiltration at various time points
(30 min to overnight) followed by LC-MS/MS analysis. The
ratio of (CH3)2CH-3 to (CD3CH3)CH-3 was obtained by the
ratio of the intensity of the 456.22 Da and 459.23 Da MS peaks
(Figure S21).
(Figure S17).
Formation of 5
′
-deoxyadenosine (5
′
-dA) during the
crosslinking reaction. The CysS-catalyzed reaction of the
bromoethoxy substrate analog 35 was carried out as de-
scribed above. The formation of 5′-dA was measured over
a 7-hour time period by LC-MS/MS (Figure S18). The for-
mation of 5’-dA was much less than that observed for sub-
7
strate 1 (15% uncoupling).
Isotope effect measurements. For the competitive isotope ef-
fect measurement, the CD3-pantetheinyl substrate CD3-1 was
synthesized as shown in Figure S27. CysS (80 µM) was incu-
bated with 1 mM MeCbl, 1 mM SAM, 1 mM CH3-
CysS reactions with deuterium-labeled cobalamin
(CD3Cbl) and deuterium-labeled SAM under single turno-
ver condition. The procedure for the synthesis of CD3Cbl is
given in the supporting information (page S31). 75 µM CysS
was incubated with 1 mM methyl viologen, 4 mM NADPH,
0.75 mM CH3SAM, 10 µM ethyl pantetheinyl substrate 2 and
0.5 mM CD3Cbl at room temperature. The reactions were
quenched by ultrafiltration after 6 h followed by LC-MS/MS
analysis. The ratio of (CH3)2CH-3 to (CD3CH3)CH-3 was ob-
tained by the ratio of the intensity of the 456.21 Da and 459.23
Da MS peaks (Figure S21).
pantetheinyl substrate (CH3-1) and
1 mM of CD3-
pantetheinyl substrate (CD3-1), 1 mM methyl viologen and 3
mM NADPH at room temperature in an anaerobic chamber.
The reaction was quenched at different time points (10 min,
30 min, 1 h, 2.5 h, 4 h, 5.5 h) by ultrafiltration. The reaction
mixtures were analyzed by LC-MS/MS and the ratio of the
products CH3CH2-2 (443.21 Da) and CH3CD2-2 (444.21 Da)
was measured based on the peak intensity after correction
for the natural abundance of the stable isotopes (Figure S19).
Determination of the cobalamin content of the “as iso-
lated” CysS. When the CysS plasmid was co-expressed with
pBAD42-BtuCEDFB (cobalamin uptake system), cobalamin
can be detected with as isolated CysS. The concentration of
16
O
SR
O
SR
O
SR
O
SR
CysS
Me
C
Me
C
D
C
H
C
D
H
H
cobalamin was determined by measuring the absorbance at
D
H
H
O
O
O
O
D
D
1
NH2
NH2
NH2
367 nm of dicyanocobalamin formed after treating the pro-
NH2
CD3-1
CH3CD2-2
PD
2
1:1
16, 33
PH
tein samples with potassium cyanide.
This assay did not
specifically quantitate methylcobalamin because it decom-
poses to cobalamin in the presence of air and ambient light.
Based on the extinction coefficient of dicyanocobalamin
D
PH / PD (kobs
)
H / (kobs
)
=
Figure 3. Strategy for the measurement of the CysS competi-
tive isotope effect.
−1
−1
(e367 = 30,800 M cm ), it was determined that CysS contains
4
ACS Paragon Plus Environment

Page 5 of 14
Journal of the American Chemical Society
0.36 equiv of cobalamin per polypeptide. Protein concentra-
showed the same MS/MS fragmentation pattern, and UV
1
2
3
4
5
6
7
8
tion was measured by the absorbance at 280nm (A280) when
the protein was exposed to air or the [4Fe–4S]2+ cluster was
reduced to [4Fe–4S]1+ by dithionite (to prevent interference
from the [4Fe–4S]2+ absorbance) with an extinction coeffi-
cient calculated using the ProtParam tool of the ExPASy pro-
teomics server.
spectrum (Figures S8, S9). Compound 36 was further charac-
terized by derivatization based on its unique reactivity (Fig-
ure 7). When the enzymatic product 36 was photolyzed (ex-
posed to ambient lighting), aminophenol 37 and the ethoxy
analog 38 were formed (Figure S10-12 and S29). These prod-
ucts were identified by co-migration with authentic synthetic
standards (Figure S11-12). Their formation is consistent with
the well-established photoreactivity of R-Cbl(III) which un-
9
RESULTS
dergoes photochemical fragmentation to give an organic rad-
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
34-35
ical.
When the photolysis reaction is carried out in D2O,
Substrate radical trapping
The M+1 peak of 38 increased from 19% to 65% of the M peak
consistent with a hydrogen atom abstraction from an ex-
changeable site (Figure S12). The R-Cbl(III) C-Co(III) bond
was also cleaved by cyanide treatment to give aminophenol
37 and NC-Cbl(III) as previously demonstrated for other R-
Cbl(III) compounds (Figure 7 and Figure S13). The synthe-
sized sample of 36 showed the same reactivity to light and to
The native substrate for CysS is a modified peptide carrier
protein (CysG1) that is part of a large multicomponent NRPS
system. Since the native substrate is complex, we designed
6, 8
a set of truncated analogs for this study. The cyclopropyl-
methoxy analog 26 was synthesized and tested to probe for
the radical intermediate 21. MS analysis of the CysS reaction
mixture containing CysS, SAM, MeCbl, MV, NADPH and 26
demonstrated the formation of a product with a mass corre-
sponding to the cyclopropylethoxy-containing product 27.
This was confirmed by co-elution with a synthetic standard
(Figure 5). In addition, a Cbl adduct (Cbl+31) was also de-
tected as a minor product (Figures S4 and S5). This product
resulted from the ring-opened radical 31, adding to the metal
or the pi system of Me-Cbl (Figure S22). Since the ring-
opened product 32 (R=H) has the same mass and similar re-
tention time to the substrate 26, it was not possible to directly
detect this product in the reaction mixture by LCMS analysis
because its peak is masked by the large peak resulting from
unreacted 26. All possible ring-opened products contain an
acid labile vinyl ether. Therefore, these products (E/Z iso-
mers of 32 with R=H, Cbl and protein) were quantitated after
selective acid hydrolysis with TFA to the corresponding ami-
nophenol 28 (Figure S4). The yield of this hydrolysis reaction
reached a maximum after 200 min. of acid treatment (Figure
S2). LC-MS/MS analysis gave the expected product mass of
414.1708 Da (Figure 3). Synthetic samples of 27 and 28 were
used to calibrate the MS analysis of the enzymatic reaction
mixture (Figure S1). The ratio of 27 to 28 was determined to
be 12± 0.3 (Table S1). Assuming that kring =k’ring, the rate of the
32
cyanide as the enzymatic product (Figure S14).
Competitive and intrinsic deuterium isotope effects
A 1:1 mixture of CD3-1 and CH3-1 was treated with CysS
to measure the competitive isotope effect. The enzymatic re-
action was quenched after various reaction times (10 min, 30
min, 1 h, 2.5 h, 4 h, 5.5 h) and analyzed by LC-MS/MS. The
product ratio was calculated from the MS peak intensity of
CH3CH2-2 (443.20) and CH3CD2-2 (444.21) (Figure S19). The
10 min ratio (5 % conversion) was used to calculate the com-
petitive isotope effect (17 ± 0.3, Figure 3) because at longer
reaction times the i-propoxy product 3, which selectively de-
pletes the CH3CH2-2 signal, can be detected.
To measure the intrinsic isotope effect (kH/kD) for the hydro-
gen atom abstraction, CD2H-1 was treated with CysS. The re-
action was quenched after 60 min and the product ratio was
determined by LC-MS/MS from the ratio of the correspond-
ing peak intensities for CH3CD2-2 (444.21) and CH3CDH-2
(443.21) (Figure S20). Under these conditions, the i-propoxy
product 3 was first detected after a reaction time of 2 h. As-
suming that the methyl group can rotate rapidly in the active
site, the intrinsic isotope effect was calculated to be 18 ± 0.6
using the equation in Figure 4.
8
−1
methyl transfer reaction can be estimated as 2.4 × 10 s (Fig-
ure 5).
Identification of the substrate form of cobalamin
CysS-catalyzed crosslinking between substrate 35 and
Cbl(I)
Treatment of CysS, containing 0.36 equiv. of cobalamin per
polypeptide, with CH3-SAM and CD3-Cbl (or with CD3SAM
and CH3-Cbl), under single turnover conditions, demon-
strated that the enzyme preferentially used the methyl group
of SAM in each case (Figure S21).
The bromoethoxy analog 35 was synthesized as shown in
Figure S26. LC-MS/MS analysis of the reaction mixture
demonstrated the formation of a new product eluting after
23.8 min. MS analysis of this product was consistent with
structure 36 in which the substrate was crosslinked to the Cbl
cofactor (Figure 6C). SAM is a weak inhibitor of the cross-
linking reaction (6% reduction in the crosslinking yield,
[SAM]=1 mM, [35]=1.2 mM, Figure S17). The expected cross-
linked reaction product was synthesized as shown in Figure
S6 and characterized by MS analysis (Figure S7). The enzy-
matic product co-migrated with the synthetic product 36 and
DISCUSSION
A modified mechanism for the iterative methylation reac-
tions catalyzed by CysS, is shown in Figure 8. In this mecha-
nism, substrate 1 and SAM bind to the enzyme to give com-
plex I. A slow conformational change generates complex J in
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 14
which the SAM methyl group is aligned for Cbl(I) methyla-
radical 30 because 30 has additional stabilization by the aryl
1
2
3
4
5
6
7
8
tion to give complex K. Dissociation of SAH, and binding of
a second SAM gives complex D. Reduction of SAM by the
ring that will retard the reaction. Both of these factors would
reduce the estimated rate for the methyl transfer. Therefore,
+1
36-
8
−1
reduced cluster ([4Fe-4S] ) generates the adenosyl radical
2.4 × 10 s should be viewed as an approximate upper limit
for the methyl transfer rate. However, even with these qual-
ifications, it is clear that the CysS-catalyzed methyl transfer
is the fastest measured rate constant for an enzymatic methyl
transfer reaction (Table 1).
37
38-
which then forms a weak bond with the [4Fe-4S] cluster.
39
Regeneration of the adenosyl radical which abstracts a hy-
drogen atom from the substrate 1 gives complex F. Here we
are assuming that capture of the adenosyl radical by the [4Fe-
4S] cluster is faster than hydrogen atom abstraction. Radical
21 undergoes a radical substitution reaction with Me-Cbl(III)
to give complex G. Product dissociation followed by cobalt
and cluster reduction regenerates the enzyme (Complex H)
for a second and third sequence of methylation reactions.
This mechanism is consistent with the substrate identifica-
tion, the radical trapping, the crosslinking, and the isotope
effects described above.
To confirm the rapid rate of the methyl transfer reaction,
an alternative radical trapping strategy was explored based
on the rapid cleavage of a carbon-bromine bond beta to a rad-
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
8
-1
44-45
ical center (k = 10 s ,Figure S23).
This trapping reaction is
faster than the methyl cyclopropyl ring opening, is catalytic
rather than inactivating, and avoids the possibility of radical
quenching by addition to the protein or the corrinoid. How-
ever, treatment of the enzyme with the bromoethoxy analog
35 resulted in the crosslinking of the substrate to the Cbl(I).
Products from the trapping of the radical intermediate were
not detected in the reaction mixture (Figure S23). The Cbl+35
adduct was characterized by MS, co-migration with an au-
thentic sample and two derivatization reactions. The inacti-
vated enzyme is still capable of binding SAM and producing
low levels of deoxyadenosine in an uncoupled reaction (Fig-
ure S18). This crosslinking reaction was unanticipated be-
cause the mechanism in Figure 2 suggests that Cbl(I) meth-
ylation followed by the generation of the substrate radical
and its quenching by methylation or bromine atom loss was
the likely fate of this substrate analog (Figure S23). The pos-
sibility of non-enzymatic alkylation of Cbl(I) was eliminated
by demonstrating that C-Co bond formation with 35 was 23
times faster than with 1-bromo-2-methoxyethane (41, Figure
S16). The crosslinking experiment establishes the proximity
of the methyl group of the substrate 1 to the methyl group of
MeCbl(III), an active site feature that is essential for the rapid
methyl transfer to the substrate radical.
CysS preferentially uses cobalamin over methylcobalamin
as the enzyme substrate. This demonstrates that during the
catalytic cycle, cobalt methylation (J to K in Figure 8) occurs
preferentially at the enzyme active site (Figure S21).
Our mechanism predicts the formation of a substrate rad-
ical 21 (complex F). To trap this hypothetical radical, we pre-
pared the cyclopropylcarbinyl substrate 26. Ring opening of
cyclopropylcarbinyl radicals is a widely used radical trap
and the rate constant for the ring opening of the (methox-
7
−1 40-41
ymethyl)-cyclopropyl radical is 2 × 10 s .
CysS is an ideal
system for this radical probe because this enzyme binds the
methoxy, ethoxy, isopropoxy, s-butoxy and t-butoxy amino
benzoic acid thioesters (1-4b), suggesting that there is suffi-
cient space at the active site for the cyclopropylcarbinyl sub-
strate analog and its rearrangement product. In the event,
when 26 was treated with CysS, the cyclopropylethoxy sub-
stituted product 27 was the major product detected in the re-
action mixture by LC-MS/MS analysis. We were unable to
detect a major product resulting from cyclopropane ring
opening, most likely because the resulting radical 31 is
formed at low levels and can be quenched in several different
ways giving rise to multiple products (e.g. E/Z isomers of 32
with R=H, Cbl and protein, Figure S4).
A notable feature of the crosslinking reaction is that it is
not strongly inhibited by SAM (1.2 mM 35 + 1.0 mM SAM
Figure 6A, Figure S17). This demonstrates that Cbl(I) is alkyl-
ated faster by the bromoethoxy-analog 35 than by SAM. This
is not consistent with the mechanism shown in Figure 2 in
which Cbl(I) is methylated before substrate binding because
Cbl(I) methylation would then block the crosslinking reac-
tion. We therefore propose that SAM could be bound in two
For the purpose of quantitating the ring opening reaction,
we exploited the fact that all of the possible ring-opened
products contain an acid labile vinyl ether and can be con-
verted to the aminophenol 28 by mild and selective trifluoro-
acetic acid treatment of the CysS reaction mixture (Figures S2
slowly
interconverting
conformations
in
the
CysS/SAM/substrate complex. This conformational change
involves rotation about the two CH2-S bonds with the ade-
nosyl and amino acid moieties fixed by interactions with the
protein and the 4Fe-4S cluster, respectively. It’s possible that
in the first conformation (Figure 8, Structure J) the SAM me-
thyl group is positioned to methylate Cbl(I) while in the se-
cond conformation (Structure D) the SAM methyl group
moves away from the MeCbl(III) to avoid a steric clash be-
tween the two methyl groups. This movement of the SAM
methyl group allows the methyl group of the substrate 1 to
be in close proximity to the methyl group of MeCbl(III) as
and S3). Vinyl ethers are frequently used as acid labile phe-
42-43
nol protecting groups in organic synthesis.
The ratio of 27
to 28 was determined to be 12 ± 0.3 by calibrated MS analysis
(Figure S1). Assuming that kring =k’ring, the rate of the methyl
8
−1
transfer reaction was estimated to be 2.4 × 10 s . This rate
however merits some qualification because it is possible that
the rate of the cyclopropylcarbinyl rearrangement of enzyme
bound 30 could be slower than the non-enzymatic rate for 33
if the active site structure prevented the optimal alignment
of the radical with the sessile bond in the cyclopropane ring.
In addition, the reference rate constant for the ring opening
of 33 is likely to be higher than the rate of ring opening of
6
ACS Paragon Plus Environment

Page 7 of 14
Journal of the American Chemical Society
required for the methyl transfer reaction. This model is con-
sistent with the formation of a CysS/SAM/bromoethoxy sub-
1
2
3
4
5
6
7
8
strate complex with SAM locked in the second conformation
by the bromoethoxy substituent which is then in position to
alkylate the Cbl(I) (Figure 9). This model is also consistent
with previous stereochemical studies on Fom3 that suggest
the substrate is sandwiched between the Cbl cofactor and the
CONCLUSIONS
Radical trapping, a substrate Cbl(I)-crosslinking experi-
ment and isotope effect measurements were carried out to
probe the mechanism of the iterative methylation reactions
catalyzed by CysS. The radical trapping experiment pro-
vided evidence for the intermediacy of the substrate radical
12-14
adenosyl radical.
The isotope effect studies on CysS proved to be quite in-
formative. The intrinsic isotope effect for the hydrogen atom
abstraction from the substrate methyl group by the adenosyl
radical is 18 ± 0.6. This was determined from the intramolec-
ular competition between hydrogen atom and deuterium
atom abstraction from the substrate 1 in which the methyl
group was replaced by a CD2H group (Figure 4). This isotope
effect is larger than the semi-classical limit predicted by tran-
sition state theory. Such large kinetic isotope effects are often
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
8
−1
21, and an estimate of 2.4 × 10 s for the rate of the methyl
transfer reaction. This is the fastest measured rate constant
for an enzymatic methyl transfer reaction. The crosslinking
experiment established the expected proximity of the sub-
strate methyl group to the methyl group of MeCbl(III). The
observation that crosslinking was faster than SAM mediated
methylation suggests a model for the active site in which
SAM binds in two slowly interconverting conformations in-
volving movement of the methyl group towards and away
from the cobalt center. This model is consistent with the for-
mation of a CysS/SAM/bromoethoxy substrate complex with
SAM locked in the second conformation by the bromoethoxy
substituent which is then in position to alkylate the Cbl(I).
The observation that the intrinsic and competitive isotope ef-
fects (18 ± 0.6 and 17 ± 0.3 respectively) are close in magni-
tude demonstrates that there are no slow reactions between
substrate 1 binding to CysS and the hydrogen atom abstrac-
tion. In the context of the mechanism in Figure 8, the isotope
effect measurements suggest that the first irreversible step in
the reaction sequence is the hydrogen atom abstraction and
that substrate binding is reversible up to that point. These
experiments support the revised mechanism for CysS shown
in Figure 8.
interpreted as an indication of quantum mechanical tunnel-
46-47
ing of the hydrogen atom.
Large non-classical isotope ef-
fects in radical SAM enzymes have been previously reported
48-50
for biotin synthase and lipoate synthase.
A competitive
isotope effect was also measured using a competition be-
tween OCH3 and OCD3 substrates and determined to be 17 ±
0.3. This competitive isotope effect monitors C-H cleavage
events, for the substrate 1, in reactions occurring up to and
including the first irreversible step. This step is likely to be
the hydrogen atom abstraction from the substrate (E to F in
Figure 8) because the reverse reaction is estimated to be en-
51
dothermic by 34 kJ/mol (Rate estimate shown in Figure S54).
The observation that the competitive isotope effect is similar
in magnitude to the intrinsic isotope effect is consistent with
reversible binding of substrate 1 to the enzyme in all steps
prior to the hydrogen atom abstraction. This is not surprising
because 1 is a highly truncated analog of the natural sub-
strate. Weak binding of 1 to the enzyme is also consistent
with the formation of a mixture of mono, di, and trimethyl-
ated products and also suggests that the iterative methyla-
tion reactions are not processive. Similar non processive iter-
ative methyl transfer reactions have been demonstrated in
17
complex carbapenem biosynthesis.
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 14
1+
O
R
1
2
3
4
5
6
7
8
482.2319
E
A
B
C
O
R
O
NH2
O
27
27
NH2
1+
483.2360
1+
484.2334
480.2137
480
482
484
486
488 m/z
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
F
O
R
414.1708
OH
D
O
R
28
NH2
OH
28
NH2
412.7828
419.0198
417.5 420.0
410.0
412.5
415.0
m/z
10
15
20
25
30
Retention time (minutes)
G
O
O
S
OH
R:
N
N
H
H
OH
O
R
O
R
O
R
!
"#
CysS
P1
MeCbl
O
O
O
NH2
SAM
5’dA
NH2
NH2
27
30
26
!
$%&'
!′$%&'
2×107 ;<1
O
O
O
R
O
R
O
R
34
33
Acid
P2
(
01
02
O
O
R
)*
OH
=
NH2
NH2
NH2
(
+,-.
28
32
31
Figure 5. Trapping of the substrate radical in the CysS-catalyzed reaction. A) Extracted ion chromatogram (EICs) of a synthe-
sized sample of the cyclopropylethoxy containing product 27 [M + H] (482.23 ± 0.02); (B) EIC of the major CysS reaction
product [M + H] (482.23 ± 0.02), (C) EIC of the new product generated by acid treatment of the reaction mixture [M + H]
(414.17 ± 0.02), (D) EIC of a synthesized sample of aminophenol 28 [M + H] (414.17 ± 0.02), (E) Mass spectrum of the cyclopro-
pylethoxy containing product 27 (Exact mass: 482.2319, Accuracy: <1 ppm); (F) Mass spectrum of the aminophenol containing
product 28 (Exact mass: 414.1693, Accuracy: 3.6 ppm); (G) Strategy for measuring the methyl transfer rate.
+
+
+
+
8
ACS Paragon Plus Environment

Page 9 of 14
Journal of the American Chemical Society
A
B
1
2
3
36
2+ 2+
805.8840868.3504
O
O
S
N
CysS, SAM, HOCbl
H
4
5
6
MV,NADPH
Br
O35
NH2
2+
806.8493
7
Full reaction
8
9
2+
807.3477
36
O
2+
O
S
N
807.8464
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
CysS, HOCbl
MV,NADPH
H
Br
805
806
807
808
O
NH2
35
Control reaction without SAM
C
Control reaction without substrate
Control reaction without CysS
Control reaction without HOCbl
Control reaction without reducing agent
Cbl+35 (36)
10
15
20
25
30
Retention time (minutes)
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 6. CysS-catalyzed reaction with the bromoethoxy analog 35. (A) LC-MS/MS analysis of the CysS reaction mixture. EICs
of the adduct [M + 2H] (805.84 ± 0.02), red trace represents the full reaction; green, black, blue, pink, and black traces repre-
sent the full reaction without SAM, without substrate, without CysS, without HOCbl and without reducing system, respec-
tively. (B) Mass spectrum of the Cbl+35 adduct (36, Exact mass: 805.8464, Accuracy: 9.4 ppm). (C) Proposed structure of the
Cbl+substrate crosslinked species.
2+
O
SR
O
SR
X
H
X
+
Co(II)
Light
O
O
SR
O
H
NH2
NH2
38
39
H
H
C
β‐scission
O
CH2
NH2
36
Co(III)
ethene
O
SR
O
SR
CN
NaCN
+
Co(III)
H
X
X
O
O
H
O
R:
CNCbl
N
H
NH2
NH2
40
37
Figure 7. Derivatization of the crosslinked product to establish the presence of a C-Co bond.
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 14
1
2
3
4
5
6
7
8
9
J
I
H
H
N
1+
Fe
1+
NH
S
1+
Fe
S
Fe
O
S
O
Fe
S
S
Fe
Fe
S
SLOW
O
CH3
OH
O
Fe
S
Fe
S
S
S
H3C
S
(I)Co
Fe
S
(I)Co
(I)Co
S
Fe
Fe
OH
S
H3C
Fe
H3C
OH
O
OH
O
O
O
H2N
A
H2N
A
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
COSR
COSR
1
1
CH3
O
CH3
O
H2N
COSR
H2N
COSR
1
1
K
D
E*
NH
H
S
1+
NH
3+
NH
Fe
O
S
O
1+
Fe
Fe
SAH
S
O
O
Fe
S
Fe
S
S
O
S
O
Fe
S
Fe
S
S
Fe
S
S
(III)Co CH3
H3C
(III)Co CH3
(III)Co CH3
H2N
Fe
S
OH S
OH
Fe
Fe
OH
OH
CH3
S
Fe
CH3
O
O
O
O
OH
OH
SAM
O
O
H2N
1
A
H2N
1
A
A
COSR
COSR
COSR
COSR
1
CH3
O
CH3
O
CH3
O
H2N
H2N
1
COSR
H2N
COSR
1
1
G
F
E
2+
Fe
NH
O
NH
Fe
2+
NH
2+
Fe
O
O
S
S
Fe
O
O
S
Fe
S
S
Fe
S
O
5'dA
S
Met
2
S
S
Fe
S
Fe
S
Fe
S
H
H
O
(II)Co
(III)Co CH3
H2C
(III)Co CH3
H2N
OH
S
OH
20
Fe
S
Fe
OH
S
Fe
H3C
OH
CH3
O
OH
O
OH
19
O
O
O
H2N
2
20
H2N
21
A
A
A
COSR
COSR
COSR
1
O
O
L
OH
R:
N
N
H
H
OH
2+
Fe
O
SR
O
SR
S
O
SR
Repeat
Fe
S
S
Repeat
CH3
CH3
CH3
CH3
CH3
4a
CH3
Fe
S
+
(II)Co
H
CH3
O
O
Fe
O
H
NH2
NH2
3
NH2
4b
2e-
Figure 8. Proposed mechanism for the CysS-catalyzed sequential methylation reactions. The labels for the various intermedi-
ate states retain some of the labels used in our original mechanism shown in Figure 2.
10
ACS Paragon Plus Environment

Page 11 of 14
1
Journal of the American Chemical Society
2
3
Table 1: Representative enzyme-catalyzed methyl transfer rates
-1
Enzyme Methyl donor
Methyl acceptor
Homocysteine
Rate (s )
4
5
6
7
8
9
Methionine Synthase (Cbl independ- Methyltetrahydro-
0.25
52-53
ent)
folate
54
Methionine Synthase (Cbl dependent)
Methyltetrahydro-
folate
Cbl(I)
250
54
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
Methionine Synthase (Cbl dependent)
MeCbl(III)
SAM
Homocysteine
GAATTC
Catechol
140
41
55
EcoRI DNA methyl- transferase
56
Catechol methyl transferase
SAM
0.21
3
57
Glycine N-Methyltransferase
SAM
Glycine
8
CysS
MeCbl(III)
Radical 21
2.4 × 10
H
H
N
1+
Fe
N
1+
Fe
S
O
S
O
Fe
S
S
O
Fe
S
S
O
Fe
S
Br
S
(I)Co
(III)Co
Fe
S
S
OH
A
Fe
OH
A
Fe
O
O
OH
OH
O
H2N
H2N
O
COSR
COSR
35
36
Figure 9. Proposed ternary complex of CysS, SAM and the bromoethoxy substrate 35 consistent with the crosslinking reaction
and the lack of inhibition of this reaction by SAM.
ASSOCIATED CONTENT
Notes
Supporting Information
The authors declare no competing financial interest.
The Supporting Information is available free of charge on the
ACS Publications website.
Experimental methods, additional figures and spectral anal-
ACKNOWLEDGMENT
We thank Prof. Squire J. Booker (Pennsylvania State Univer-
sity) for providing the pBAD42-BtuCEDFB plasmid.
yses. (PDF)
ABBREVIATIONS
AUTHOR INFORMATION
MeCbl, Methylcobalamin; HOCbl, hydroxocobalamin;
CNCbl, cyanocobalamin; SAM, S-Adenosylmethionine; 5′-
dA, 5′-deoxyadenosine; MV, methyl viologen; NADPH,
nicotinamide adenine dinucleotide phosphate hydrate; SAH,
S-adenosylhomocysteine.
Corresponding Author
* begley@chem.tamu.edu
Author Contributions
The manuscript was written through contributions of all
authors.
Funding Sources
REFERENCES
This work was supported by the Robert A. Welch Founda-
tion (A-0034)
1.
Zhang, Q.; van der Donk, W. A.; Liu, W., Radical-
Mediated Enzymatic Methylation: A Tale of Two SAMS.
Accounts Chem Res 2012, 45 (4), 555-564.
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 14
2.
Bauerle, M. R.; Schwalm, E. L.; Booker, S. J.,
16.
B.; Wang, R. X.; Jones, B. S.; Booker, S. J., Enhanced Solubilization
of Class Radical S-Adenosylmethionine Methylases by
Lanz, N. D.; Blaszczyk, A. J.; McCarthy, E. L.; Wang,
1
2
Mechanistic diversity of radical S-adenosylmethionine (SAM)-
dependent methylation. J Biol Chem 2015, 290 (7), 3995-4002.
B
3
4
5
6
7
8
9
3.
Bridwell-Rabb, J.; Grell, T. A. J.; Drennan, C. L., A Rich
Improved Cobalamin Uptake in Escherichia coli. Biochemistry
2018, 57 (9), 1475-1490.
Man, Poor Man Story of S-Adenosylmethionine and Cobalamin
Revisited. Annual Review of Biochemistry, Vol 87 2018, 87, 555-584.
4.
reactions. Curr Opin Chem Biol 2013, 17 (4), 597-604.
5. Wang, S. C., Cobalamin-dependent radical S-
17.
Sinner, E. K.; Lichstrahl, M. S.; Li, R.; Marous, D. R.;
Fujimori, D. G., Radical SAM-mediated methylation
Townsend, C. A., Methylations in complex carbapenem
biosynthesis are catalyzed by a single cobalamin-dependent
radical S-adenosylmethionine enzyme. Chem. Commun.
(Cambridge, U. K.) 2019, 55 (99), 14934-14937.
adenosyl-l-methionine enzymes in natural product biosynthesis.
Nat Prod Rep 2018, 35 (8), 707-720.
18.
Ban, Y. H.; Song, M. C.; Hwang, J.-y.; Shin, H.-l.; Kim,
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
6.
Baumann, S.; Herrmann, J.; Raju, R.; Steinmetz, H.;
H. J.; Hong, S. K.; Lee, N. J.; Park, J. W.; Cha, S.-S.; Liu, H.-w.;
Yoon, Y. J., Complete reconstitution of the diverse pathways of
gentamicin B biosynthesis. Nat. Chem. Biol. 2019, 15 (3), 295-303.
Mohr, K. I.; Huttel, S.; Harmrolfs, K.; Stadler, M.; Muller, R.,
Cystobactamids: myxobacterial topoisomerase inhibitors
exhibiting potent antibacterial activity. Angewandte Chemie 2014,
53 (52), 14605-9.
19.
Blaszczyk, A. J.; Wang, B.; Silakov, A.; Ho, J. V.;
Booker, S. J., Efficient methylation of C2 in L-tryptophan by the
cobalamin-dependent radical S-adenosylmethionine methylase
TsrM requires an unmodified N1 amine. J Biol Chem 2017, 292
(37), 15456-15467.
7.
Wang, Y.; Schnell, B.; Baumann, S.; Muller, R.; Begley,
T. P., Biosynthesis of Branched Alkoxy Groups: Iterative Methyl
Group Alkylation by a Cobalamin-Dependent Radical SAM
Enzyme. Journal of the American Chemical Society 2017, 139 (5),
1742-1745.
8.
Methylations Resulting in the Biosynthesis of the t-Butyl Group
Catalyzed by B12-Dependent Radical SAM Enzyme in
Cystobactamid Biosynthesis. Methods Enzymol 2018, 606, 199-
216.
9.
F.; Moreno, M.; Stadler, M.; Bronstrup, M.; Kirschning, A.;
Muller, R., Discovery and Total Synthesis of Natural
Cystobactamid Derivatives with Superior Activity against
Gram-Negative Pathogens. Angewandte Chemie 2017, 56 (41),
12760-12764.
10.
adenosyl-L-methionine
biosynthesis. Nat. Prod. Rep. 2018, 35 (8), 707-720.
11. Kim, H. J.; Liu, Y. N.; McCarty, R. M.; Liu, H. W.,
Reaction Catalyzed by GenK, a Cobalamin-Dependent Radical
S-Adenosyl-l-methionine Methyltransferase in the Biosynthetic
Pathway of Gentamicin, Proceeds with Retention of
Configuration. Journal of the American Chemical Society 2017, 139
(45), 16084-16087.
20.
Benjdia, A.; Pierre, S.; Gherasim, C.; Guillot, A.;
Carmona, M.; Amara, P.; Banerjee, R.; Berteau, O., The
thiostrepton A tryptophan methyltransferase TsrM catalyses a
cob(II)alamin-dependent methyl transfer reaction. Nature
communications 2015, 6, 8377.
Wang, Y.; Schnell, B.; Muller, R.; Begley, T. P., Iterative
a
21.
Blaszczyk, A. J.; Silakov, A.; Zhang, B.; Maiocco, S. J.;
Lanz, N. D.; Kelly, W. L.; Elliott, S. J.; Krebs, C.; Booker, S. J.,
Spectroscopic and Electrochemical Characterization of the Iron-
Sulfur and Cobalamin Cofactors of TsrM, an Unusual Radical S-
Adenosylmethionine Methylase. Journal of the American Chemical
Society 2016, 138 (10), 3416-26.
Huttel, S.; Testolin, G.; Herrmann, J.; Planke, T.; Gille,
22.
Marous, D. R.; Lloyd, E. P.; Buller, A. R.; Moshos, K.
A.; Grove, T. L.; Blaszczyk, A. J.; Booker, S. J.; Townsend, C. A.,
Consecutive radical S-adenosylmethionine methylations form
the ethyl side chain in thienamycin biosynthesis. P Natl Acad Sci
USA 2015, 112 (33), 10354-10358.
Wang, S. C., Cobalamin-dependent radical S-
enzymes in natural product
23.
Parent, A.; Guillot, A.; Benjdia, A.; Chartier, G.;
Leprince, J.; Berteau, O., The B12-Radical SAM Enzyme PoyC
Catalyzes Valine Cbeta-Methylation during Polytheonamide
Biosynthesis. Journal of the American Chemical Society 2016, 138
(48), 15515-15518.
24.
Denes, F., Heteroatom-centred radicals for the
12.
Sato, S.; Kudo, F.; Kuzuyama, T.; Hammerschmidt, F.;
synthesis of heterocyclic compounds. Top. Heterocycl. Chem.
2018, 54 (Free-Radical Synthesis and Functionalization of
Heterocycles), 151-230.
Eguchi, T., C-Methylation Catalyzed by Fom3, a Cobalamin-
dependent Radical SAM Enzyme in Fosfomycin Biosynthesis,
Proceeds with Inversion of Configuration. Biochemistry 2018, 57
(33), 4963-4966.
25.
menage a trois. Acc. Chem. Res. 1998, 31 (3), 99-107.
26. Kyne, S. H.; Schiesser, C. H., Intramolecular homolytic
Walton, J. C., Homolytic substitution: a molecular
13.
Wang, B.; Blaszczyk, A. J.; Knox, H. L.; Zhou, S.; Blaesi,
E. J.; Krebs, C.; Wang, R. X.; Booker, S. J., Stereochemical and
Mechanistic Investigation of the Reaction Catalyzed by Fom3
from Streptomyces fradiae, a Cobalamin-Dependent Radical S-
Adenosylmethionine Methylase. Biochemistry 2018, 57 (33), 4972-
4984.
substitutions in synthesis. In Encyclopedia of Radicals in Chemistry,
Biology and Materials, Chatgilialoglu, C.; Studer, A., Eds. John
Wiley & Sons: New York, 2012; Vol. 2, pp 629-654.
27.
Branchaud, B. P.; Friestad, G. K., Vitamin B12. In e-
EROS Encyclopedia of Reagents for Organic Synthesis, John Wiley
& Sons: New York, 2001; pp 1-4.
14.
McLaughlin, M. I.; van der Donk, W. A., Stereospecific
Radical-Mediated B12-Dependent Methyl Transfer by the
Fosfomycin Biosynthesis Enzyme Fom3. Biochemistry 2018, 57
(33), 4967-4971.
28.
Mosimann, H.; Kräutler, B., Methylcorrinoids
Methylate Radicals—Their Second Biological Mode of Action?
Angewandte Chemie International Edition 2000, 39 (2), 393-395.
15.
Hu, K. F.; Werner, W. J.; Allen, K. D.; Wang, S. C.,
29.
Dolphin, D. H., Preparation of the reduced forms of
Investigation of enzymatic C-P bond formation using multiple
quantum HCP nuclear magnetic resonance spectroscopy. Magn
Reson Chem 2015, 53 (4), 267-272.
vitamin B12 and of some analogs of the vitamin B12 coenzyme
containing a cobalt-carbon bond. Vitamins and Coenzymes 1971,
18, 34-52.
12
ACS Paragon Plus Environment

Page 13 of 14
Journal of the American Chemical Society
30.
Endicott, J. F.; Netzel, T. L., Early Events and Transient
44.
Wagner, P. J.; Lindstrom, M. J.; Sedon, J. H.; Ward, D.
1
2
3
4
5
6
7
8
Chemistry in the Photohomolysis of Alkylcobalamins. Journal of
the American Chemical Society 1979, 101 (14), 4000-4002.
31.
cobalamins in aqueous solutions and in human blood. J Photoch
Photobio B 2013, 122, 7-14.
32.
for the unexpected associative displacement of adenosyl by
cyanide in coenzyme B-12. Inorg Chem 1997, 36 (15), 3216-3222.
33.
R., Photochemistry of Delta-Halo Ketones - Anchimeric
Assistance in Triplet-State Gamma-Hydrogen Abstraction and
Beta-Elimination of Halogen Atoms from the Resulting
Diradicals. Journal of the American Chemical Society 1981, 103 (13),
3842-3849.
Juzeniene, A.; Nizauskaite, Z., Photodegradation of
Brasch, N. E.; Hamza, M. S. A.; van Eldik, R., Evidence
45.
Joshi, S.; Mahanta, N.; Fedoseyenko, D.; Williams, H.;
Begley, T. P., Aminofutalosine Synthase: Evidence for
Captodative and Aryl Radical Intermediates Using beta-Scission
and SRN1 Trapping Reactions. Journal of the American Chemical
Society 2017, 139 (32), 10952-10955.
Ljungdah.Lg; Legall, J.; Lee, J. P., Isolation of a Protein
9
Containing Tightly Bound 5-Methoxybenzimidazolylcobamide
(Factor Iiim) from Clostridium-Thermoaceticum. Biochemistry
1973, 12 (9), 1802-1808.
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
46.
Kohen, A.; Klinman, J. P., Hydrogen tunneling in
biology. Chem Biol 1999, 6 (7), R191-8.
47.
protein dynamics to enzyme catalysis. Annu. Rev. Biochem. 2013,
82, 471-496.
48.
radical-mediated carbon-sulfur bond formation. Bba-Proteins
Proteom 2012, 1824 (11), 1213-1222.
49.
34.
Lott, W. B.; Chagovetz, A. M.; Grissom, C. B., Alkyl
Klinman, J. P.; Kohen, A., Hydrogen tunneling links
Radical Geometry Controls Geminate Cage Recombination in
Alkylcobalamins. Journal of the American Chemical Society 1995,
117 (49), 12194-12201.
Fugate, C. J.; Jarrett, J. T., Biotin synthase: Insights into
35.
Pratt, J. M., Role of the Protein in the B12-Dependent
Enzymes - Steric Control of a Molecular Switch. Inorg Chim a-
Bioinor 1983, 79 (1-6), 27-28.
36.
T.; Begley, T. P.; Tantillo, D. J.; Britt, R. D., Trapping and electron
paramagnetic resonance characterization of the 5'dAdo.bul.
radical in a radical S-adenosyl methionine enzyme reaction with
a non-native substrate. ACS Cent. Sci. 2019, 5 (11), 1777-1785.
Parry, R. J., Biosynthesis of Lipoic Acid .1.
Sayler, R. I.; Stich, T. A.; Joshi, S.; Cooper, N.; Shaw, J.
Incorporation of Specifically Tritiated Octanoic-Acid into Lipoic
Acid. Journal of the American Chemical Society 1977, 99 (19), 6464-
6466.
50.
Douglas, P.; Kriek, M.; Bryant, P.; Roach, P. L., Lipoyl
synthase inserts sulfur atoms into an octanoyl substrate in a
stepwise manner. Angew Chem Int Edit 2006, 45 (31), 5197-5199.
37.
Yang, H.; McDaniel, E. C.; Impano, S.; Byer, A. S.;
Jodts, R. J.; Yokoyama, K.; Broderick, W. E.; Broderick, J. B.;
Hoffman, B. M., The Elusive 5'-Deoxyadenosyl Radical:
Captured and Characterized by Electron Paramagnetic
Resonance and Electron Nuclear Double Resonance
Spectroscopies. J. Am. Chem. Soc. 2019, 141 (30), 12139-12146.
51.
chemical bond energies. CRC Press: Boca Raton, 2007; p 1655 p.
52. Taurog, R. E.; Jakubowski, H.; Matthews, R. G.,
Synergistic, random sequential binding of substrates in
cobalamin-independent methionine synthase. Biochemistry 2006,
45 (16), 5083-91.
53.
methionine synthase, John Wiley & Sons Ltd.: 2001; pp 1010-1022.
54. Banerjee, R. V.; Frasca, V.; Ballou, D. P.; Matthews, R.
G., Participation of cob(I) alamin in the reaction catalyzed by
methionine synthase from Escherichia coli: a steady-state and
rapid reaction kinetic analysis. Biochemistry 1990, 29 (50), 11101-
9.
Luo, Y.-R.; Luo, Y.-R., Comprehensive handbook of
38.
Broderick, W. E.; Hoffman, B. M.; Broderick, J. B.,
Mechanism of Radical Initiation in the Radical S-Adenosyl-L-
methionine Superfamily. Acc. Chem. Res. 2018, 51 (11), 2611-2619.
39.
Impano, S.; Pagnier, A.; Watts, H.; Denler, C.; Vagstad, A. L.;
Piel, J.; Duschene, K. S.; Shepard, E. M.; Shields, T. P.; Scott, L.
G.; Lilla, E. A.; Yokoyama, K.; Broderick, W. E.; Hoffman, B. M.;
Broderick, J. B., Paradigm Shift for Radical S-Adenosyl-L-
methionine Reactions: The Organometallic Intermediate Ω Is
Central to Catalysis. J. Am. Chem. Soc. 2018, 140 (28), 8634-8638.
Gruber, K.; Kratky, C. In Cobalamin-dependent
Byer, A. S.; Yang, H.; McDaniel, E. C.; Kathiresan, V.;
55.
Reich, N. O.; Mashhoon, N., Presteady State Kinetics
of an S-Adenosylmethionine-Dependent Enzyme - Evidence for
a Unique Binding Orientation Requirement for Ecori DNA
Methyltransferase. J Biol Chem 1993, 268 (13), 9191-9193.
40.
Newcomb, M., Radical Kinetics and Clocks. In
Encyclopedia of Radicals in Chemistry, Biology and Materials
Chatgilialoglu, C.; Studer, A., Eds. John Wiley & Sons: New
York, 2012; Vol. 107-124.
56.
Zhang, J. Y.; Klinman, J. P., Enzymatic Methyl
Transfer: Role of an Active Site Residue in Generating Active
Site Compaction That Correlates with Catalytic Efficiency.
Journal of the American Chemical Society 2011, 133 (43), 17134-
17137.
41.
Martinez, F. N.; Schlegel, H. B.; Newcomb, M., Ab
initio molecular orbital calculations of electronic effects on the
kinetics of cyclopropylcarbinyl radical ring openings. J Org Chem
1998, 63 (11), 3618-3623.
57.
Zhang, J. Y.; Klinman, J. P., Convergent Mechanistic
42.
Olofson, R. A.; Schnur, R. C., Value of the
Features between the Structurally Diverse N- and O-
Methyltransferases: Glycine N-Methyltransferase and Catechol
O-Methyltransferase. Journal of the American Chemical Society
2016, 138 (29), 9158-9165.
vinyloxycarbonyl unit in hydroxyl protection: application to the
synthesis of nalorphine. Tetrahedron Lett. 1977, (18), 1571-4.
43.
based protecting groups. In Handbook of Organopalladium
Chemistry for Organic Synthesis, Negishi, E. i., Ed. John Wiley &
Sons: 2002; Vol. 2, pp 1901-1911.
Lipton, M., Palladium-catalyzed deprotection of allyl-
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 14
1
2
3
For Table of Contents Only
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
14
ACS Paragon Plus Environment