Mechanism of a Class C Radical S-Adenosyl- l -methionine Thiazole Methyl Transferase

Article abstract of DOI:10.1021/jacs.7b10203

The past decade has seen the discovery of four different classes of radical S-adenosylmethionine (rSAM) methyltransferases that methylate unactivated carbon centers. Whereas the mechanism of class A is well understood, the molecular details of methylation by classes B-D are not. In this study, we present detailed mechanistic investigations of the class C rSAM methyltransferase TbtI involved in the biosynthesis of the potent thiopeptide antibiotic thiomuracin. TbtI C-methylates a Cys-derived thiazole during posttranslational maturation. Product analysis demonstrates that two SAM molecules are required for methylation and that one SAM (SAM1) is converted to 5′-deoxyadenosine and the second SAM (SAM2) is converted to S-adenosyl-l-homocysteine (SAH). Isotope labeling studies show that a hydrogen is transferred from the methyl group of SAM2 to the 5′-deoxyadenosine of SAM1 and the other two hydrogens of the methyl group of SAM2 appear in the methylated product. In addition, a hydrogen appears to be transferred from the β-position of the thiazole to the methyl group in the product. We also show that the methyl protons in the product can exchange with solvent. A mechanism consistent with these observations is presented that differs from other characterized radical SAM methyltransferases.

Full text of DOI:10.1021/jacs.7b10203

Subscriber access provided by Gothenburg University Library
Article
Mechanism of a Class C Radical SAM Thiazole Methyl Transferase
Zhengan Zhang, Nilkamal Mahanta, Graham A. Hudson, Douglas A. Mitchell, and Wilfred A. van der Donk
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.7b10203 • Publication Date (Web): 30 Nov 2017
Downloaded from http://pubs.acs.org on November 30, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society 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 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
1
Mechanism of a Class C Radical SAM Thiazole Methyl Transferase
Zhengan Zhang,1,2 Nilkamal Mahanta,1,3 Graham A. Hudson,1 Douglas A. Mitchell1,3*
and Wilfred A. van der Donk1ꢀ3*
2
3
4
1 Department of Chemistry, University of Illinois at UrbanaꢀChampaign, Urbana, Illinois 61801,
USA. 2 Howard Hughes Medical Institute, University of Illinois at UrbanaꢀChampaign, Urbana, Illinois
14
15
5
16
17
6
7
8
9
61801, 3 Institute for Genomic Biology, University of Illinois at UrbanaꢀChampaign, 1206 West Gregory
Drive, Urbana, IL, 61801, USA.
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
* To whom correspondence should be addressed:
10
11
12
13
14
15
Douglas A. Mitchell
600 S. Mathews Avenue
Urbana, Illinois 61801
douglasm@illinois.edu
phone: (217) 333ꢀ1345
fax: (217) 333ꢀ0508
Wilfred A. van der Donk
600 S. Mathews Avenue
Urbana, Illinois 61801
vddonk@illinois.edu
phone: (217) 244ꢀ5360
fax: (217) 244ꢀ8533
16
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 29
1
2
3
4
5
6
7
1
2
Abstract
The past decade has seen the discovery of four different classes of radical Sꢀadenosylmethionine
(rSAM) methyltransferases that methylate unactivated carbon centers. Whereas the mechanism of class
A is well understood, the molecular details of methylation by class BꢀD are not. In this study, we present
detailed mechanistic investigations of the class C rSAM methyltransferase TbtI involved in the
biosynthesis of the potent thiopeptide antibiotic thiomuracin. TbtI Cꢀmethylates a Cysꢀderived thiazole
during posttranslational maturation. Product analysis demonstrates that two SAM molecules are required
for methylation and that one SAM (SAM1) is converted to 5’ꢀdeoxyadenosine and the second SAM
(SAM2) is converted to SꢀadenosylꢀLꢀhomocysteine (SAH). Isotope labeling studies show that a
hydrogen is transferred from the methyl group of SAM2 to the 5’ꢀdeoxyadenosine of SAM1 and the
other two hydrogens of the methyl group of SAM2 appear in the methylated product. In addition, a
hydrogen appears to be transferred from the βꢀposition of the thiazole to the methyl group in the product.
We also show that the methyl protons in the product can exchange with solvent. A mechanism consistent
with these observations is presented that differs from other characterized radical SAM
methyltransferases.
8
9
3
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
4
5
6
7
8
9
10
11
12
13
14
15
2
ACS Paragon Plus Environment

Page 3 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
1
2
Introduction
For many decades, methyl transfer in nature utilizing SAM as the methyl donor was thought to
occur exclusively by SN2ꢀlike processes.1 However, early metabolic labeling studies suggested that
methyl groups derived from methionine are also transferred to nonꢀnucleophilic carbons.2ꢀ5 Subsequent
characterization of the biosynthetic gene clusters6ꢀ9 and bioinformatic and biochemical analyses of the
proteins involved in these reactions10,11 showed that they are part of the radical SAM (rSAM)
superfamily of enzymes,12 which comprise the largest enzyme superfamily known.13,14 In the years since
this groundbreaking discovery, four distinct classes of rSAM methyltransferases (MTs) have been
identified that differ in their cofactor/coꢀsubstrate requirements and their proposed reaction
mechanisms.15ꢀ18 The mechanism of RNA base methylases (class A) has been wellꢀstudied (Figure
S1),19ꢀ23 but the mechanisms of the other three classes are considerably less understood. Class B consists
of cobalaminꢀdependent methyltransferases, including TsrM,24,25 Fom3,10,26,27 GenK,28 CysS,29 ThnK,11
Sven0516,30 and PoyC.31 Class C are cobalaminꢀindependent and are involved in the biosynthesis of
several natural products,32ꢀ36 whereas class D enzymes use methylenetetrahydrofolate as the methyl
group donor.37 In this study, we undertook a detailed mechanistic investigation of a class C rSAM MT
(TbtI) involved in the biosynthesis of the thiopeptide thiomuracin (Figure 1a), and show that it uses a
fundamentally different mechanism from other characterized classes of rSAM MTs.
7
8
9
3
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
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19 Results
20
21
TbtI contains one iron-sulfur cluster and does not require additional cysteines
We recently reported the in vitro reconstitution of TbtI, which regioselectively Cꢀmethylates one of
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 29
1
2
3
4
5
6
7
8
9
1
2
six thiazoles during the biosynthesis of thiomuracin (Figure 1b).36 Although sequence analysis does not
place TbtI among class A rSAM MTs, we nevertheless first assessed whether TbtI might employ a
similar mechanism. Class A rSAM MTs utilize an unusual mechanism in which a methyl group from
SAM is first transferred to a Cys on the protein. Then, reductive cleavage of a second SAM molecule
occurs, mediated by a reduced ironꢀsulfur [4Feꢀ4S] cluster, which generates a 5’ꢀdeoxyadenosyl radical
(5’ꢀdA•). The 5’ꢀdA• species then abstracts a hydrogen atom from the methylcysteine to initiate the
methyl transfer process (Figure S1). We first analyzed the sequence of TbtI and its orthologs from
various thiopeptide gene clusters36 and identified five conserved Cys residues (Cys14, Cys18, Cys21,
Cys254 and Cys343; Figure S2). Three of these appear in a canonical motif (CxxxCxxC), which
typically supply the ligands for a characteristic rSAM [4Feꢀ4S] cluster.12 TbtI variants in which these
Cys residues were replaced by Ala (C14A, C18A and C21A) were inactive (Figure S3), but substitution
of the remaining Cys residues (C89A, C254A, and C343A) did not adversely affect the activity of TbtI
(Figure S3), suggesting that they are functionally dispensable.
3
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
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Thus, these data show that unlike class A rSAM MTs, which require at least two additional Cys
other than the three used to bind the [4Feꢀ4S] cluster,22 TbtI does not require additional Cys residues for
catalysis. This conclusion is further supported by the fact that a TbtI homolog involved in nocathiacin
biosynthesis (NocN) contains only one additional Cys other than the three involved in [4Feꢀ4S] cluster
coordination.15 We considered that a methyl group from SAM could be transferred to an amino acid
14
residue other than Cys. To test this possibility, the enzyme was treated with CH3ꢀSAM and the peptide
substrate without addition of reducing agents. The enzyme was isolated and analyzed by sodium dodecyl
4
ACS Paragon Plus Environment

Page 5 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
2
sulfate polyacrylamide gel electrophoresis (SDSꢀPAGE) and visualized by Coomassie staining or
autoradiographic imaging. We did not observe a radioactive band for the protein, suggesting that the
methyl group is not initially transferred to the enzyme. This finding could also be explained by a methyl
group already being present on the enzyme upon isolation from Escherichia coli; however, single
turnover experiments presented below argue against this possibility. We cannot completely rule out that
transfer of the methyl group from SAM to an amino acid residue only occurs in the presence of all
reaction components including reductants. TbtI lacks the cobalaminꢀbinding motif characteristic of class
B rSAM MTs,11,12,24,29,31,38ꢀ41 as well as the tetrahydrofolateꢀbinding domain observed in a recently
discovered class D rSAM MT.37 All these factors point towards a distinct mechanism for TbtI.
3
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
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Product analysis
To gain insight into the mechanism of methyl transfer, we first determined the products of the
reaction that are derived from SAM. Out of a concern that the strong reductant dithionite might react
with the initial products of the enzymatic reaction, we replaced dithionite in our original assay36 with
flavodoxin/flavodoxin reductase and NADPH to deliver the reducing equivalents to TbtI (see
Experimental Section). As anticipated based on other rSAM proteins,42 5’ꢀdeoxyadenosine (5’ꢀdA) was
detected in slightly more than one equivalent compared to the methylated TbtA product (Figure 1c).
Additionally, S-adenosylꢀLꢀhomocysteine (SAH) was detected in near stoichiometric quantities, and
5’ꢀdA and SAH were formed at very similar rates (Figure 1c). When the reaction was conducted with
CD3ꢀSAM (96% CD3 by ESIꢀMS; see Supporting Information, Figure S4), ~90% of the 5’ꢀdA contained
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 29
1
2
3
4
5
6
7
8
9
1
2
a single deuterium (Figure S4b), suggesting that the 5’ꢀdA• generated from SAM1 may abstract a
hydrogen atom from another SAM molecule (SAM2) during catalysis. It also suggests that the small
amount of unlabeled 5’ꢀdA produced was the result of uncoupled SAM cleavage, as has been observed
in other rSAM enzymes.42 Control experiments showed that production of SAH and 5’ꢀdA was
dependent on the presence of both TbtI and the reducing system (Figure S5). It was recently reported
that another class C rSAM MT involved in nosiheptide biosynthesis (NosN) produces thioadenosine in
its reaction.34 We therefore investigated if TbtI generated thioadenosine; however, we were unable to
observe its production by LCꢀMS whereas an authentic standard was readily detected at the
concentrations expected to be produced (Figure S5). Another recent study on NosN also demonstrated
that thioadenosine is not a product of the reaction.35 Based on the products we observed, we formulated
a series of working hypotheses for the enzymatic mechanism of TbtI (Figure 2).
3
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
4
5
6
7
8
9
10
11
12
13
14
Isotope labeling studies involving the methyl group of SAM
To evaluate the proposed mechanisms, we first investigated whether thiazole methylation involves
15
16
17
transfer of the intact methyl group of SAM or a methylene group. Therefore, reactions with TbtI were
carried out using CD3ꢀSAM. Owing to the large molecular weight of the TbtA hexazole core substrate
even after removal of most of the leader peptide (three residues remain, ValꢀGlyꢀAla) with
endoproteinase GluC (C71H85N19O19S6, [M+H]+ = 1700 Da), the numerous isotopologues visible by
mass spectrometry rendered isotope labeling studies difficult to interpret. Therefore, we prepared the
TbtA hexazole substrate from E. coli grown in minimal media using 13Cꢀdepleted glucose and
47 18
48
49
50
51
52
53
54
19
20
21
15Nꢀdepleted (NH4)2SO4 as the sole carbon and nitrogen sources, and again removed the majority of the
6
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 7 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
2
leader peptide by treatment with endoproteinase GluC. The resulting TbtA hexazole core peptide 1
(Figure 1b) was analyzed by electrospray ionization mass spectrometry (ESIꢀMS) and exhibited a
muchꢀsimplified spectrum (Figure S6). Use of CD3ꢀSAM in a 16 h methylation assay with TbtI and
peptide 1 resulted in an unexpectedly complicated mass spectrum for the product in which ions were
observed consistent with the addition of CH3, CH2D, and CHD2 groups (Figure 3a). These findings
indicated that there was either an intermediate that underwent considerable solvent exchange or the
product exchanged with solvent after initial formation (or both). To test the latter possibility, we repeated
the experiment with CD3ꢀSAM and quenched the reaction after only 1 h. The product was analyzed by
ESIꢀMS and an ion was observed that was consistent with the transfer of a CD2H group (Figure 3a, blue
spectrum). This product was purified by high performance liquid chromatography (HPLC), exposed to
the assay conditions for the extended reaction time (16 h), reꢀisolated, and then again analyzed by
ESIꢀMS. In agreement with the original findings, a complicated mixture of product peptide was
observed that corresponded to apparent transfer of CH3, CH2D, and CHD2 (Figure S7). This exchange
was dependent on the presence of TbtI and SAM; replacing SAM with SAH did not result in exchange.
Thus, these experiments show that the initial product of the reaction of TbtI with CD3ꢀSAM and 1
mainly contains a CD2H group, but that the enzyme can exchange the deuterium labels with protium
from solvent. Unfortunately, the large amounts of 5’ꢀdA produced from the abortive cleavage of SAM
during the 16 h reactions precluded our ability to determine if deuterium from the methyl group is
transferred to the 5’ꢀdA during the exchange process.
3
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
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Because of the solvent exchange observed at extended reaction times, all subsequent assays were
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 29
1
2
3
4
5
6
7
8
9
1
2
performed for 1 h. We first investigated whether the generation of the product containing a CHD2 group
when using CD3ꢀSAM in H2O involves the incorporation of a proton from solvent. Thus, the reaction
was performed using CD3ꢀSAM in D2O. Analysis of the product by ESIꢀMS (Figure 3b) showed that the
major product still contained CHD2, although the relative amount of CD3 transfer increased. This
observation suggests that a hydrogen in an initially nonꢀexchangeable position migrates to the methyl
group, possibly with the intermediacy of a partially solvent exchangeable intermediate. The mechanisms
in Figure 2 provide three possible sources for this hydrogen. First, the thiazole proton on the βꢀcarbon of
1 that must be removed during the methylation process could migrate via an active site base to the
methyl group of the product (pathway a). Alternatively, a hydrogen that originates from the protein that
does not exchange could be transferred to the methyl group (pathway b). Based on literature precedent,
an example of the latter could be the αꢀprotons of Gly.28,29 Lastly, the hydrogen could originate from the
methyl group of 5’ꢀdA (pathway c).
3
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
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Isotope labeling studies involving the 5’-position of SAM and the substrate thiazole
We first considered 5’ꢀdA as the source for the additional hydrogen in the methyl group of the
product. Abstraction of a hydrogen atom from 5’ꢀdA by radical 4 (pathway c, Figure 2) would explain
transfer of CHD2 when employing CD3ꢀSAM. We therefore prepared SAM from MetꢀCD3 and
5’,5’,4’,3’ꢀD4ꢀATP (~ 92% CD3ꢀ5’,5’,4’,3’ꢀD4ꢀSAM by ESIꢀMS, Fig. S8). The resulting
CD3ꢀ5’,5’,4’,3’ꢀD4ꢀSAM was then reacted with TbtI and the TbtA hexazole 1 in D2O. The reaction
products were then analyzed by ESIꢀTOF MS. The amount of product that contained CHD2 was similar
8
ACS Paragon Plus Environment

Page 9 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
1
2
to that observed when the 5’ position of SAM was unlabeled (Figure 3b). Hence, the hydrogen in this
product does not originate from the 5’ position of the adenosine, ruling out pathway c.
8
9
3
We next considered pathway b to account for the product hydrogen, in which a nonꢀexchangeable
hydrogen on the protein is transferred to the product methyl group in the first turnover; this hydrogen
would be replaced by a solventꢀderived hydrogen in subsequent turnovers. This behavior has previously
been observed, such as the transfer of Cα hydrogens from a Gly residue.43,44 In principle, hydrogen
transfer could also originate from other amino acids or very tightly hydrogen bonded positions that
would only exchange upon reaction. Because TbtI contains several conserved Gly residues (Figure S2),
we reacted a 20ꢀfold excess of substrate 1 and CD3ꢀSAM with TbtI in buffered D2O and analyzed the
product by ESIꢀMS (Figure S9). The major product again contained a CHD2 group, arguing against a
proteinꢀderived hydrogen that is transferred to the methyl group.
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
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Having ruled out pathways b and c, we next replaced the Cβ proton of the thiazole with deuterium
to interrogate pathway a (Figure 2). We prepared TbtA hexazole using an E. coli strain that is
13
auxotrophic for Cys and provided βꢀD2ꢀCys to the C/15Nꢀdepleted minimal media. Unfortunately, the
expression level of the peptide under these conditions was very low. During optimization of the
expression conditions, we noted that improved production could be achieved upon addition of a limited
number of unlabeled amino acids to the culture media (see Experimental Section for details). Analysis
by tandem ESIꢀMS confirmed the desired deuterium labeling of the six Cys residues in the TbtA
precursor (Figure S10). The resulting peptide (~ 95% D2 at Cys4 by ESIꢀMSꢀMS) was then converted to
the corresponding hexazoleꢀcontaining core peptide 2 in vitro as previously reported.45 Owing to the
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 29
1
2
3
4
5
6
7
8
9
13
15
1
2
need to add additional amino acids (which were not Cꢀ and Nꢀdepleted), the monoisotopic purity of
the hexazole 2 was lower than that of the substrate 1. Nevertheless, an ion consistent with the
hexaꢀdeuterated TbtA hexazole core peptide was the major product observed by ESIꢀMS.
3
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
4
With peptide 2 in hand, we conducted a series of single turnover experiments. If the proton in the
transferred CHD2 was proteinꢀderived in Figure 3b, then a single turnover experiment in D2O with
substrate 2 and CD3ꢀSAM should still result in CHD2 transfer. However, when the experiment was
performed with a threeꢀfold excess of TbtI over substrate 2, CD3 transfer was observed as the major
product (Figure 4a). This result provided additional data against pathway b: a nonꢀexchangeable
hydrogen from the enzyme was not the source of hydrogen in the methyl group of the product. We
tentatively propose that the small amount of CHD2 transfer observed may originate in part because of
incomplete deuteration of the reagents. We then repeated the single turnover experiment with CD3ꢀSAM
in D2O but employed substrate 1. A considerable amount of CD3 transfer was observed, indicating that
solvent exchange occurred in the single turnover experiments with excess enzyme (Figure 4b), but
importantly, the relative amount of CD3 transfer declined considerably compared to experiments
conducted with substrate 2. Collectively, these data provide strong support for pathway a being
operative during TbtI catalysis. These experiments also argue against TbtI already containing a
covalently bound methyl group introduced during expression in E. coli; this methyl group would have
been unlabeled and therefore, we would have observed a considerable amount of product containing
CH3 and CH2D.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
To identify residues that may mediate the transfer process, we substituted conserved residues Tyr8
10
ACS Paragon Plus Environment

Page 11 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
and Glu105 of TbtI with Ala (Figure S2). The corresponding residues are near the SAMꢀbinding site in
HemN.46,47 a rSAM decarboxylase from the heme biosynthetic pathway that is homologous to TbtI.
Replacement of either residue resulted in TbtI variants with severely decreased activity (Figure S11).
Two additional conserved residues (Trp23 and Tyr200) were targeted for Ala substitution given that they
could participate in electron or hydrogen atom transfer; however, the activity of the resulting TbtI
variants was not decreased, suggesting that they are not involved in catalysis (Figure S3C).
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
7
8
Discussion
9
A combination of product analysis and isotope labeling studies has allowed us to formulate a
mechanism for TbtI that is consistent with all of our available data. The observation of 5’ꢀdA as a
product suggests that the TbtIꢀcatalyzed reaction is initiated in a manner similar to canonical rSAM
proteins: electron transfer from the reduced [4Feꢀ4S] cluster initiates the reductive cleavage of SAM1 to
form methionine and 5’ꢀdA• (Figure 2).42 We propose that the 5’ꢀdA• then abstracts a hydrogen atom
from the methyl group of SAM2. This hypothesis is supported by the observation that use of CD3ꢀSAM
resulted in the incorporation of one deuterium into 5’ꢀdA. This reaction appears irreversible since we
never observed multiple deuteriums in 5’ꢀdA (Figure S4). Use of two molecules of SAM is also
consistent with (i) the observed stoichiometry of 5’ꢀdA and SAH to the methylated product, (ii) the
previously reported crystal structure of HemN with two bound SAM molecules,46,47 and (iii) recent
results reported for the class C rSAM MT NosN.35 Although HemN and TbtI catalyze different reactions
(decarboxylation and methylation, respectively), the use two molecules of SAM might be a general
10
11
12
13
14
15
16
17
18
19
20
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 29
1
2
3
4
5
6
7
1
2
property of class C rSAM enzymes.
Upon hydrogen atom abstraction from the methyl group of SAM2, we propose that the resulting
methylene radical adds to the electrophilic C=C bond of the thiazole, generating resonanceꢀstabilized
radical 3 (Figure 2). This may trigger deprotonation from the βꢀposition by an active site base B to
eliminate SAH and form resonanceꢀstabilized radical 4 that partially restores thiazole aromaticity
(Figure 2). This proton is adjacent to an enoxyradical and previous studies on 2ꢀhydroxyglutarylꢀCoA
dehydratase48 have suggested that the pKa of such protons are on the order of ~14. One candidate residue
for the active site base is Glu105, which upon substitution with Ala severely decreases TbtI activity
(Figure S11). The equivalent Glu in HemN (Glu145) is near the SAMꢀbinding site;46,47 thus, an
alternative role for Glu105 could be to bind SAM. Reduction of radical 4, possibly by the [4Feꢀ4S]
cluster, would produce resonanceꢀstabilized anion 5, which can then be protonated by the conjugate acid
of the active site base, possibly Glu105. This pathway would explain the observed deuterium transfer
from the β position of the thiazole to the methyl group in the product (Figure 4).
8
9
3
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
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
A variation of this mechanism features abstraction by radical 4 of a hydrogen atom from a redox
active amino acid in the active site (e.g. Tyr/Trp/Gly; Cys can be ruled out since their replacement with
Ala had no effect on activity, Figure S3). The amino acid radical would then be reduced and protonated
to reset the enzyme for another turnover (Figure 2, pathway b). The single turnover experiments argue
against the possibility that this hydrogen is in a nonꢀexchangeable position in the first turnover but
becomes a solventꢀderived hydrogen in subsequent turnovers. Hence, the involvement of Gly radicals is
unlikely, but other redox active amino acids cannot yet be ruled out. For instance, a tyrosinate could
12
ACS Paragon Plus Environment

Page 13 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
2
serve as the base that deprotonates radical 3 and the resulting phenol moiety could then provide a
hydrogen atom to radical 4. Tyr8 is conserved in the class C rSAM proteins (Figure S2). In HemN, the
equivalent Tyr56 πꢀstacks with the adenine of SAM2. Substitution of TbtI Tyr8 with Ala severely
decreased activity (Figure S11), and hence Tyr8 might be a hydrogen atom donor that generates the final
product. We ruled out the possibility, at least in vitro, that the final hydrogen atom donor is 5’ꢀdA by
conducting experiments with SAM that was deuterated at both the methyl group and the 5’ position
(Figure 2, pathway c). Use of this deuterated SAM analogue did not result in an increase in the
deuterium content of the final product. However, from an efficiency perspective, hydrogen atom
abstraction from 5’ꢀdA would be attractive as it would directly regenerate the 5’ꢀdA• and would not
require the additional electron required for pathways a and b. Furthermore, as shown by the
SAMꢀdependent solvent exchange of the methyl group in the product, 5’ꢀdA• may abstract a hydrogen
atom from the methyl group, which would be the microscopic reverse of the last step of pathway c.
Hence, while pathway c clearly is not consistent with our in vitro findings, perhaps under the right
conditions pathway c might be operational in vivo.
3
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
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
The proposed mechanism accounts for the observed solvent exchange upon prolonged reaction
times by realization that the methylated peptide product, SAM1, and SAH can likely still bind to the
active site since these three molecules can occupy nearly identical positions as the substrate peptide,
SAM1, and SAM2 at the start of catalysis (the only difference is that the methyl group of SAM2 is
transferred to the peptide). Reductive cleavage of SAM1 would generate 5’ꢀdA• and the methyl group of
the product peptide would be in a very similar position as the methyl group of SAM2 during productive
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 29
1
2
3
4
5
6
7
8
9
1
2
turnover. Thus, hydrogen atom abstraction could again generate radical 4 and either pathway a or b
would result in hydrogen transfer by an active site residue, thus affecting the observed exchange (Figure
S12a). Alternatively, the exchange could occur by a heterolytic deprotonation/reprotonation sequence
that requires bound SAM (Figure S12b).
3
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
4
5
In summary, our study suggests that methyl transfer catalyzed by TbtI involves two SAM
molecules. Our data are consistent with a mechanism in which one SAM is reductively cleaved to 5’dA•
as in other rSAM enzymes. The 5’dA• then abstracts a hydrogen atom from a second SAM, and the
resulting carbonꢀbased methylene radical attacks the thiazole at position 4 of the peptide substrate. This
SAMꢀderived radical is unprecedented in rSAM enzymology. We note that like the class A rSAM MTs
and a recent report on the class C MT NosN,35 TbtI formally transfers a methylene group from SAM to
the substrate and hence might not be considered a true methyl transferase. Regardless, the manner by
which this is accomplished adds a fundamentally different use of SAM in this class of enzymes, and
highlights their versatility and catalytic diversity in natural product biosynthesis. Additional biochemical
and structural studies on TbtI are underway to gain further insights to the mechanism of this remarkable
transformation.
6
7
8
9
10
11
12
13
14
15
16
17
18
Experimental Section
19
20
21
22
General methods
All polymerase chain reactions (PCR) were carried out on a C1000™ thermal cycler (BioꢀRad). DNA sequencing
was performed by ACGT, Inc. MALDIꢀTOF MS was carried out on a Bruker Daltonics UltrafleXtreme MALDI
TOF/TOF instrument (Bruker). ESIꢀMS/MS analyses were performed using a SYNAPT ESI quadrupole TOF
14
ACS Paragon Plus Environment

Page 15 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
1
2
Mass Spectrometry System (Waters) equipped with an ACQUITY Ultra Performance Liquid Chromatography
(UPLC) system (Waters). HiTrap columns for NiꢀNTA affinity chromatography were purchased from GE
Healthcare.
3
9
4
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
5
Materials
6
Reagents used for molecular biology experiments were purchased from New England BioLabs (Ipswich, MA),
Thermo Fisher Scientific (Waltham, MA), or Gold Biotechnology Inc. (St. Louis, MO). Other chemicals were
purchased from SigmaꢀAldrich (St. Louis, MO). Plasmid inserts were sequenced at ACGT Inc. (Wheeling, IL).
Oligonucleotides were purchased from Integrated DNA Technologies Inc. (Coralville, IA). Deuteriumꢀlabeled
reagents were purchased from Cambridge Isotope Laboratories.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Strains and plasmids
E. coli DH5α and E. coli BL21 (DE3) cells were used as host for cloning and plasmid propagation, and host for
protein expression, respectively. Coꢀexpression vector pRSFDuet-1 was obtained from Novagen. E. coli
JW3582ꢀ2 that is auxotrophic for cysteine was purchased from the Coli Genetic Stock Center at Yale University,
https://cgsc2.biology.yale.edu/Strain.php?ID=108920.
Construction of TbtI variants
Site directed mutagenesis (SDM) was performed using the QuikChange method (Agilent) as per the
manufacturer’s instructions using PfuTurbo DNA polymerase. The primers for each mutant are listed in Table S1.
The mutations were verified by sequencing using a custom MBP forward primer or the T7 reverse primer (Table
S1).
22
23
24
Expression, purification and assays of TbtI
25
26
27
Expression, purification and in vitro reconstitution of the [4Feꢀ4S] cluster of TbtI were carried out as previously
described.36 Briefly, the in vitro methylation reaction mixtures generally included the following components:
TbtA hexazole substrate (50–100 ꢁM), purified TbtI from E. coli heterologous expression after in vitro
15
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 29
1
2
3
4
5
6
7
8
9
1
2
reconstitution of the [4Feꢀ4S] cluster (10 ꢁM), SꢀadenosylꢀLꢀmethionine (SAM, 1 mM), flavodoxin (10 ꢁM),
flavodoxin reductase (10 ꢁM), reduced nicotinamide adenine dinucleotide phosphate (NAPDH, 2 mM) in reaction
buffer (50 mM TrisꢀHCl, pH 7.5). The reaction was allowed to proceed for 1 h at 25 °C in an anaerobic chamber
(Coy laboratory).
3
4
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
5
For reactions performed in D2O, all reagents that participate in the reaction were first dissolved in D2O,
lyophilized and reꢀdissolved in D2O followed by adjusting of the pD. The buffer containing the proteins was
exchanged with D2O storage buffer (50 mM Tris, 300 mM NaCl, pD 7.5) by repeated concentration by
ultrafiltration and dilution (four times) before addition into the reaction. The reaction was allowed to proceed in
D2O reaction buffer (50 mM Tris, pD 7.5) for 1 h at 25 °C in the anaerobic chamber.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
For single turnover reactions, a reaction mixture generally included the following components: TbtA hexazole
substrate (20 ꢁM), purified TbtI after in vitro reconstitution of the [4Feꢀ4S] cluster (60 ꢁM),
SꢀadenosylꢀLꢀmethionine (SAM, 1 mM), flavodoxin (30 ꢁM), flavodoxin reductase (30 ꢁM), reduced
nicotinamide adenine dinucleotide phosphate (NAPDH, 2 mM) in reaction buffer (50 mM TrisꢀHCl, pH 7.5). The
reaction was allowed to proceed for 10 min at 25 °C in the anaerobic chamber.
To determine the specific activity of the TbtI used, the enzyme (10 ꢁM) was incubated with the hexazole substrate
peptide (100 ꢁM), SꢀadenosylꢀLꢀmethionine (SAM, 1 mM), flavodoxin (20 ꢁM), flavodoxin reductase (20 ꢁM),
reduced nicotinamide adenine dinucleotide phosphate (NAPDH, 2 mM) in reaction buffer (50 mM TrisꢀHCl, pH
7.5). Aliquots were removed at set time points (30, 60, 90, and 120 s) and the reaction was allowed to proceed at
25 °C in an anaerobic chamber. At 120 s the conversion was 15%. Product formation over time was linear with a
velocity of 15.6 × 10ꢀ3 ꢁmol/min/mg of TbtI (i.e. 0.72 minꢀ1). To assure that under these conditions the reaction
catalyzed by TbtI was rate limiting, the amounts of flavodoxin and flavodoxin reductase were doubled, and no
change in velocity was observed. Similarly, to assure that all substrates were saturating, the concentrations of SAM,
NADPH and peptide were doubled without observing a change in velocity.
Generation of 13C- and 15N-depleted TbtA hexazole
Isotopically depleted TbtA hexazole was produced in E. coli cultured in modified M9 minimal media.49 E. coli
BL21(DE3) cells containing a pTrc99 plasmid encoding maltose binding protein (MBP)ꢀtagged TbtF,
16
ACS Paragon Plus Environment

Page 17 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
2
MBPꢀtagged TbtG, and TbtE and a pRSFDuet-1 plasmid encoding the His6ꢀTbtA were grown overnight at 37 ºC
in lysogeny broth (LB) containing 25 ꢁg/mL kanamycin and 50 ꢁg/mL ampicillin. After harvesting the cells, the
supernatant was discarded and the cells were washed with 5 mL of wash buffer (22 mM KH2PO4, 42 mM
Na2HPO4, 8.5 mM NaCl, pH 7.4). Following washing, the cells were resuspended in wash buffer and used to
inoculate (1:100) the modified M9 minimal media, with the following composition per 100 mL:10 mL of a 10×
stock of minimal media (220 mM KH2PO4, 420 mM Na2HPO4, 85 mM NaCl, pH 7.4), 0.3 mL of 40% aqueous
(14NH4)2SO4 (SigmaꢀAldrich), 2 mL of 20% aqueous 13Cꢀdepleted glucose (Cambridge Isotope Laboratories), 0.1
mg of FeSO4, 10 ꢁg of thiamine, 200 ꢁL of 1 M MgSO4, 10 ꢁL of 1 M CaCl2, and 75 ꢁL of a trace element
solution (5 mM CaCl2, 1.25 mM ZnCl2, 260 ꢁM CuCl2•H2O, 252 ꢁM CoCl2•6H2O, 250 ꢁM Na2MoO4•2H2O, pH
7.4). The media also contained 50 ꢁg/mL ampicillin and 25 ꢁg/mL kanamycin. The cells were grown at 37 °C and
induced at OD600 = 0.6ꢀ0.8 by the addition of isopropyl βꢀDꢀ1ꢀthiogalactopyranoside (IPTG) to a final
concentration of 0.5 mM and grown for an additional 16ꢀ18 h at 18 °C before harvesting. Purification of TbtA
hexazole was carried out as previously described.50
3
4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
5
6
7
8
9
10
11
12
13
14
15
Generation of TbtA hexazole deuterium labeled at each thiazole
33 16
βꢀD2ꢀcysteine labeled MBPꢀtagged TbtA was generated using E. coli strain JW3582ꢀ2. A lysogenization step was
performed to JW3582ꢀ2 before production of MBPꢀtagged TbtA so that the host strain could be used to express
target genes cloned in T7 expression vectors. Lysogenization was performed using the λDE3 Lysogenization Kit
(Novagen) as per the manufacturer’s instructions. The lysogenized JW3582ꢀ2 was transformed with a pET28
plasmid containing the ampicillin resistance gene and encoding MBPꢀtagged TbtA.
34
17
18
19
20
21
22
23
24
25
26
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
Expression of βꢀD2ꢀcysteine labeled MBPꢀTbtA was performed in modified M9 minimal media as described
above, except that amino acids were added into the culture (final concentrations: 0.5 mM Arg, 1 mM His, 1 mM
Leu, 1 mM Lys, 0.5 mM Met, 2 mM Pro, 3 mM Thr, 0.4 mM Trp and 1 mM of βꢀD2ꢀCys). Purification and in
vitro cyclodehydration of the βꢀD2ꢀcysteineꢀlabeled MBPꢀtagged TbtA was carried out as previously described.45
Leader peptide removal
17
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 29
1
2
3
4
5
6
1
2
Proteolytic removal of the leader peptide to generate TbtAꢀhexazole core peptides for TbtI reactions were carried
out as previously described.50
7
8
3
9
4
Preparation of SAM
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
5
E. coli strain DM22ꢀ(pK8), which overexpresses E. coli SAM synthetase, was constructed by G. D. Markham et
al. (Fox Chase Cancer Center, Philadelphia, PA).51 DM22ꢀ(pK8) was grown in LB containing 30 ꢁg/mL
tetracycline at 37 °C while being shaken and harvested by centrifugation 12 h after inoculation. The cells (∼12 g)
were resuspended in 36 mL of 0.1 M TrisꢀHCl (pH 8.0) containing 1 mM ethylenediaminetetraacetic acid (EDTA)
and 50 ꢁg/mL lysozyme and gently stirred at room temperature for 30 min. Phenylmethylsulfonyl fluoride (PMSF)
was added to a final concentration of 1 mM, and cells were lysed with a MultiFlex C3 Homogenizer (Avestin).
The cell lysate was centrifuged for 30 min at 26,500 × g at 4 °C. The cleared lysate was dialyzed against 4 L of
100 mM TrisꢀHCl (pH 8.0), 1 mM EDTA, in 2 × 30 mL 20,000 molecular weight cutꢀoff (MWCO) SlideꢀAꢀLyzer
dialysis cassettes (Pierce) at 4 °C overnight, and was divided into three portions, frozen in liquid nitrogen, and
stored at ꢀ80°C.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
SAM was generated enzymatically by incubating an aliquot of the lysate in a buffered solution (0.1 L) containing
0.1 M TrisꢀHCl (pH 8.0), 50 mM KCl, 1 mM EDTA, 20% acetonitrile, 26 mM MgCl2, 13 mM adenosine
triphosphate (ATP), and 10 mM methionine. After the mixture had been gently stirred for 5 h at room temperature,
the reaction was quenched by adjusting the pH to 5.0 with HCl. The mixture was placed on ice for 15 min and
centrifuged at 4,000 × g for 30 min at 4 °C to remove precipitated material. The supernatant was diluted to 1 L
with 1 mM sodium acetate (pH 5.0) and loaded on a CMꢀ52 cation exchange resin (Whatman) that had been
charged with 0.2 M sodium acetate (pH 5.0) and equilibrated in 1 mM sodium acetate (pH 5.0). The column was
rinsed with 1 L of 1 mM sodium acetate (pH 5.0), and SAM was eluted with 0.5 L of 40 mM H2SO4. The presence
of SAM was monitored by UVꢀVis absorption at 260 nm, and fractions containing SAM were combined. The
solution was concentrated by rotovap to a volume of 4 mL, and was divided into several fractions and stored at
−80 °C. Deuteriumꢀlabeled SAM was prepared using the same procedure but with labelled methionine or ATP.
Radioactive assay with 14CH3-SAM. The assay to determine whether the SAM methyl group is transferred onto
18
ACS Paragon Plus Environment

Page 19 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
2
TbtI was performed with 14CH3ꢀSAM (0.02 mCi/mL, 300 ꢁM). The in vitro reconstitution of the [4Feꢀ4S] cluster
of MBPꢀTbtI was performed as previously described,36 followed by addition of TbtA hexazole inside an anaerobic
chamber. After incubation at room temperature for 1 h, the reaction mixture was taken out of the anaerobic
chamber and moved into an anaerobic bag filled with nitrogen gas, and 1:1 14CH3ꢀSAM/unlabeled CH3ꢀSAM
mixture was then added into the reaction inside the anaerobic bag. The reaction mixture included the following
3
4
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
5
14
6
components: MBPꢀTbtI (10 ꢁM), TbtA hexazole (100 ꢁM), 1:1 CH3ꢀSAM/unlabeled CH3ꢀSAM mixture (300
7
ꢁM) in a final volume of 100 ꢁL. The reaction was incubated at room temperature for 1 h, and 30 ꢁL of the
reaction mixture was loaded on a 4ꢀ20% polyacrylamide gel. Electrophoresis was performed for 45 min at 150 V
(constant voltage), and then the gel was stained with Coomassie blue staining solution (40% methanol, 10% acetic
acid, 0.1 % Coomassie blue R250 in H2O) for 30 min and destained in destain buffer (20% methanol, 10% acetic
acid in H2O) for 3 h. After drying the gel overnight with a gelꢀdrying kit (Promega) in accordance with the
manufacturer’s procedure, the dried gel was exposed on a phosphorimager cassette for 4 days. The film was
finally visualized with a STORM840 Phosphorimager Scanner (Amersham Biosciences). Whereas the Coomassie
stained gel clearly showed the presence of the protein, no radioactivity was observed in the gel.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Chitin binding domain (CBD)-fused flavodoxin (Flv)/flavodoxin reductase (Flx) overexpression and purification.
E. coli BL21(DE3) cells were transformed with an intein based expression vector pTYB1 plasmid encoding the CBD
fused Flv/Flx.52 Cells were grown for 24 h on LB agar plates containing 100 ꢁg/mL ampicillin at 37 °C. Single colonies
were picked to inoculate 10 mL of LB containing the same concentration of antibiotic and grown at 37 °C for 16–18 h.
This culture was used to inoculate 1 L of LB supplemented with the same concentration of antibiotic and grown to an
OD600 of 0.6 before being placed on ice for 15 min. Protein expression was then induced with the addition of 0.5 mM
IPTG and supplemented with flavin adenine mononucleotide (FMN) for overexpression of Flv and flavin adenine
dinucleotide (FAD) for overexpression of Flx to a final concentration of 10 ꢁM respectively. Expression was allowed to
proceed for 12 h at 18 °C. Cells were harvested by centrifugation at 4,000 × g for 20 min, washed with
phosphateꢀbuffered saline (PBS; 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, and 1.8 mM KH2PO4), and harvested
by centrifugation. The cells were flashꢀfrozen with liquid nitrogen and stored at ꢀ80 °C for a maximum of one week
before use.
19
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 29
1
2
3
4
5
6
7
8
9
1
2
Cells were resuspended in cold lysis buffer (20 mM TrisꢀHCl pH 8.5, 500 mM NaCl, 1 mM TCEP, and 0.1% Triton
Xꢀ100 (v/v)) containing 4 mg/mL lysozyme, 2 ꢁM leupeptin, 2 ꢁM benzamidine, and 2 ꢁM E64. Cells were further
lysed by sonication (3 × 45 s with 10 min agitation periods at 4 °C). Insoluble debris was removed by centrifugation at
20,000 × g for 40 min at 4 °C. The supernatant was then applied to a column containing lysisꢀbuffer preꢀequilibrated
chitin resin (NEB; 15 mL of resin per 1L of initial cell culture). The column was washed with 10ꢀ15 column volumes of
wash buffer (20 mM TrisꢀHCl pH 8.5, 1 M NaCl, 1mM TCEP and either 10 ꢁM FMN for Flv or 10 ꢁM FAD for Flx
respectively). The column was flushed quickly with 20ꢀ25 mL of cleavage buffer (20 mM TrisꢀHCl pH 8.5, 500 mM
NaCl, 1mM EDTA, 25 mM DTT) and the column was incubated at 4 °C for 20ꢀ24h. The CBP fused proteins were
eluted using elution buffer (20 mM TrisꢀHCl pH 8.5, 500 mM NaCl), and collected in 1mL fractions. The yellow
colored fractions were pooled and concentrated using a 30 kDa MWCO Amicon Ultra centrifugal filter (EMD
3
4
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
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Millipore).
A
buffer exchange step was performed with protein storage buffer (50 mM
4ꢀ(2ꢀhydroxyethyl)ꢀ1ꢀpiperazineethanesulfonic acid (HEPES) pH 7.5, 300 mM NaCl, 1mM TCEP and 5% glycerol
(v/v)) prior to final concentration, freezing in liquid nitrogen, and storage at ꢀ80 °C.
UPLC MS and ESI-MS/MS. The sample was prepared as described previously.45 LCꢀMS was performed using a
Waters SYNAPT mass spectrometer outfitted with an ACQUITY UPLC, an ACQUITY Bridged Ethyl Hybrid C8
column (2.1 × 50 mm, 1.7 ꢁm particle size, 200 Å; Waters), an ESI ion source, and a quadruple TOF detector. A
gradient of 2–100% aq. MeCN with 0.1% formic acid (v/v) over 20 min was used. Fragmentation of the sample
was performed using a collisionꢀinduced dissociation (CID) method. A ramping of cone voltage of 15–18 kV
during the scan was performed to generate peptide fragments for MS/MS analysis.
20
ACS Paragon Plus Environment

Page 21 of 29
Journal of the American Chemical Society
1
2
3
4
1
Figures and legends
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
2
3
4
Figure 1 | Product analysis for the methyl transfer reaction catalyzed by TbtI. (a) Structure of thiomuracin
A1. The methyl group that is introduced by TbtI is shown in red. (b) Regioselective methylation of the thiazole at
position 4 of the TbtA core peptide by TbtI. The ValꢀGlyꢀAla sequence at the Nꢀterminus of the peptide 1
originates from the leader peptide.36 (c) Timeꢀdependent formation of the methylated TbtA peptide, 5’ꢀdA, and
SAH. Reaction mixtures included the following components: TbtA hexazole (50 ꢁM), [4Feꢀ4S] cluster
reconstituted TbtI (10 ꢁM), SAM (1 mM), flavodoxin (10 ꢁM), flavodoxin reductase (10 ꢁM), NAPDH (2 mM)
in reaction buffer (50 mM TrisꢀHCl, pH 7.5).
5
6
7
8
9
10
21
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 29
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
2
3
4
5
Figure 2 | Proposed potential reaction mechanisms for TbtI. 5’ꢀdA• is generated from SAM1 in a process
mediated by the reduced [4Feꢀ4S] cluster. The 5’ꢀdA• then abstracts a hydrogen atom from the methyl group of
SAM2, and the resulting radical adds to the thiazole. An active site base deprotonates radical 3 leading to the
22
ACS Paragon Plus Environment

Page 23 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
2
3
4
5
elimination of SAH, yielding radical 4. Three different pathways (a-c) can provide the product and reset the
enzyme. In pathway a, the radical is reduced to the anion 5 and protonated. In pathway b, radical 4 abstracts a
hydrogen atom from an active site amino acid (XꢀH). XꢀH could be the protonated base BꢀH. In pathway c,
radical 4 abstracts a hydrogen atom from the methyl group of 5’ꢀdA. Ade = adenine.
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
7
23
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 29
1
2
3
4
5
6
7
8
9
13
15
1
2
3
4
5
6
Figure 3 | ESI mass spectra to investigate the origin of the methyl protons. Cꢀ and Nꢀdepleted substrate
peptides were obtained as described (see Experimental Section). (a) Spectra showing the doubly charged ion for
the hexazoleꢀcontaining core peptide substrate 1 (red) and the methylated product obtained with CD3ꢀSAM in
H2O after 1 h (blue) or 16 h (black) reactions. (b) Spectra showing substrate (red) and product obtained with
CD3ꢀSAM in D2O (blue) and CD3ꢀ5’,5’,4’,3’ꢀD4ꢀSAM in D2O (black) after 1 h. See Table S2 for calculated and
observed masses.
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
24
ACS Paragon Plus Environment

Page 25 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
1
31
32
33
2
3
4
5
6
7
8
Figure 4 | A solvent exchangeable proton migrates from the β-position of thiazole 4 to the methyl group in
the product. (a) Spectra showing the doubly charged core peptide of the TbtA hexazole substrate 2 that is
deuterated at the βꢀcarbon of each thiazole (red) and the corresponding product obtained with CD3ꢀSAM in D2O
(blue) in a single turnover reaction. (b) Spectra showing the doubly charged core peptide of the TbtA hexazole
substrate 1 (red) and the corresponding product obtained with CD3ꢀSAM in D2O (blue) in a TbtIꢀcatalyzed single
turnover reaction. See Table S2 for calculated and observed masses.
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
9
25
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 26 of 29
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
Acknowledgments
This work was supported by the Howard Hughes Medical Institute (to W.A.V.), the National
Institutes of Health (GM097142 to D.A.M.), and a Seemon Pines Fellowship from the Department of
Chemistry at the University of Illinois at UrbanaꢀChampaign (to G.A.H.). Constructs for expression of
flavodoxin/flavodoxin reductase were generously provided by the laboratory of Squire J. Booker (Penn
State University), and constructs for SAM synthetase were provided by Vahe Bandarian (University of
Utah).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
8
9
ASSOCIATED CONTENT
Supporting Information
10
11
12
Experimental details and supporting figures. This material is available free of charge via the Internet at
http://pubs.acs.org.
34 13
35
Notes The authors declare no competing financial interest
36
37
14
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
26
ACS Paragon Plus Environment

Page 27 of 29
Journal of the American Chemical Society
1
2
3
4
1
References
5
2
3
6
7
8
9
(1) Takusagawa, F.; Fujioka, M.; Spies, A.; Schowen, R. L. In Comprehensive Biological Catalysis; Sinnott, M., Ed.; Academic
Press: New York, 1998, p 1.
4
5
(2) Kamigiri, K.; Hidaka, T.; Imai, S.; Murakami, T.; Seto, H. J. Antibiot. 1992, 45, 781.
(3) Kuzuyama, T.; Hidaka, T.; Kamigiri, K.; Imai, S.; Seto, H. J. Antibiot. 1992, 45, 1812.
(4) Williamson, J. M.; Inamine, E.; Wilson, K. E.; Douglas, A. W.; Liesch, J. M.; Albers-Schönberg, G. J. Biol. Chem. 1985, 260,
4637.
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
7
8
9
(5) Houck, D. R.; Kobayashi, K.; Williamson, J. M.; Floss, H. G. J. Am. Chem. Soc. 1986, 108, 5365.
(6) Hidaka, T.; Goda, M.; Kuzuyama, T.; Takei, N.; Kidaka, M.; Seto, H. Mol. Gen. Genet. 1995, 249, 274.
(7) Hidaka, T.; Hidaka, M.; Kuzuyama, T.; Seto, H. Gene 1995, 158, 149.
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
(8) Kuzuyama, T.; Seki, T.; Dairi, T.; Hidaka, T.; Seto, H. J. Antibiot. 1995, 48, 1191.
(9) Nunez, L. E.; Mendez, C.; Brana, A. F.; Blanco, G.; Salas, J. A. Chem. Biol. 2003, 10, 301.
(10) Sato, S.; Kudo, F.; Kim, S. Y.; Kuzuyama, T.; Eguchi, T. Biochemistry 2017, 56, 3519.
(11) Marous, D. R.; Lloyd, E. P.; Buller, A. R.; Moshos, K. A.; Grove, T. L.; Blaszczyk, A. J.; Booker, S. J.; Townsend, C. A. Proc.
Natl. Acad. Sci. USA 2015, 112, 10354.
(12) Sofia, H. J.; Chen, G.; Hetzler, B. G.; Reyes-Spindola, J. F.; Miller, N. E. Nucleic Acids Res. 2001, 29, 1097.
(13) Akiva, E.; Brown, S.; Almonacid, D. E.; Barber, A. E., 2nd; Custer, A. F.; Hicks, M. A.; Huang, C. C.; Lauck, F.; Mashiyama,
S. T.; Meng, E. C.; Mischel, D.; Morris, J. H.; Ojha, S.; Schnoes, A. M.; Stryke, D.; Yunes, J. M.; Ferrin, T. E.; Holliday, G. L.;
Babbitt, P. C. Nucleic Acids Res. 2014, 42, D521.
(14) Landgraf, B. J.; McCarthy, E. L.; Booker, S. J. Annu. Rev. Biochem. 2016, 85, 485.
(15) Zhang, Q.; van der Donk, W. A.; Liu, W. Acc. Chem. Res. 2012, 45, 555.
(16) Fujimori, D. G. Curr. Opin. Chem. Biol. 2013, 17, 597.
(17) Bauerle, M. R.; Schwalm, E. L.; Booker, S. J. J. Biol. Chem. 2015, 290, 3995.
(18) Ding, Y.; Yu, Y.; Pan, H.; Guo, H.; Li, Y.; Liu, W. Mol. Biosyst. 2010, 6, 1180.
(19) Yan, F.; LaMarre, J. M.; Rohrich, R.; Wiesner, J.; Jomaa, H.; Mankin, A. S.; Fujimori, D. G. J. Am. Chem. Soc. 2010, 132,
3953.
(20) Boal, A. K.; Grove, T. L.; McLaughlin, M. I.; Yennawar, N. H.; Booker, S. J.; Rosenzweig, A. C. Science 2011, 332, 1089.
(21) Yan, F.; Fujimori, D. G. Proc. Natl. Acad. Sci. USA 2011, 108, 3930.
(22) Grove, T. L.; Benner, J. S.; Radle, M. I.; Ahlum, J. H.; Landgraf, B. J.; Krebs, C.; Booker, S. J. Science 2011, 332, 604.
(23) Schwalm, E. L.; Grove, T. L.; Booker, S. J.; Boal, A. K. Science 2016, 352, 309.
(24) Pierre, S.; Guillot, A.; Benjdia, A.; Sandstrom, C.; Langella, P.; Berteau, O. Nat. Chem. Biol. 2012, 8, 957.
(25) Blaszczyk, A. J.; Silakov, A.; Zhang, B.; Maiocco, S. J.; Lanz, N. D.; Kelly, W. L.; Elliott, S. J.; Krebs, C.; Booker, S. J. J. Am.
Chem. Soc. 2016, 138, 3416.
(26) Woodyer, R. D.; Li, G.; Zhao, H.; van der Donk, W. A. Chem. Commun. 2007, 359.
(27) Allen, K. D.; Wang, S. C. Arch. Biochem. Biophys. 2014, 543, 67.
(28) Kim, H. J.; McCarty, R. M.; Ogasawara, Y.; Liu, Y. N.; Mansoorabadi, S. O.; LeVieux, J.; Liu, H. W. J. Am. Chem. Soc. 2013,
135, 8093.
(29) Wang, Y.; Schnell, B.; Baumann, S.; Muller, R.; Begley, T. P. J. Am. Chem. Soc. 2017, 139, 1742.
27
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 28 of 29
1
2
3
4
5
6
7
8
9
1
2
(30) Inahashi, Y.; Zhou, S.; Bibb, M. J.; Song, L.; Al-Bassam, M. M.; Challis, G. L. Chem. Sci. 2017, 8, 2823.
(31) Parent, A.; Guillot, A.; Benjdia, A.; Chartier, G.; Leprince, J.; Berteau, O. J. Am. Chem. Soc. 2016, 138, 15515.
(32) Huang, W.; Xu, H.; Li, Y.; Zhang, F.; Chen, X. Y.; He, Q. L.; Igarashi, Y.; Tang, G. L. J. Am. Chem. Soc. 2012, 134, 8831.
(33) Hiratsuka, T.; Suzuki, H.; Kariya, R.; Seo, T.; Minami, A.; Oikawa, H. Angew. Chem. Int. Ed. 2014, 53, 5423.
3
4
5
(34) Ding, W.; Li, Y.; Zhao, J.; Ji, X.; Mo, T.; Qianzhu, H.; Tu, T.; Deng, Z.; Yu, Y.; Chen, F.; Zhang, Q. Angew. Chem. Int. Ed.
2017, 56, 3857.
6
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
7
(35) LaMattina, J. W.; Wang, B.; Badding, E. D.; Gadsby, L. K.; Grove, T.; Booker, S. J. J. Am. Chem. Soc. 2017, doi:
10.1021/jacs.7b08492.
8
9
(36) Mahanta, N.; Zhang, Z.; Hudson, G. A.; van der Donk, W. A.; Mitchell, D. A. J. Am. Chem. Soc. 2017, 139, 4310.
(37) Allen, K. D.; Xu, H.; White, R. H. J. Bacteriol. 2014, 196, 3315.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(38) van der Donk, W. A. J. Org. Chem. 2006, 71, 9561.
(39) Werner, W. J.; Allen, K. D.; Hu, K.; Helms, G. L.; Chen, B. S.; Wang, S. C. Biochemistry 2011, 50, 8986.
(40) Zhou, S.; Alkhalaf, L. M.; de Los Santos, E. L.; Challis, G. L. Curr. Opin. Chem. Biol. 2016, 35, 73.
(41) Bridwell-Rabb, J.; Zhong, A.; Sun, H. G.; Drennan, C. L.; Liu, H. W. Nature 2017, 544, 322.
(42) Broderick, J. B.; Duffus, B. R.; Duschene, K. S.; Shepard, E. M. Chem. Rev. 2014, 114, 4229.
(43) Shisler, K. A.; Broderick, J. B. Arch. Biochem. Biophys. 2014, 546, 64.
(44) Kamat, S. S.; Raushel, F. M. Perspectives in Science 2015, 4, 32.
(45) Hudson, G. A.; Zhang, Z.; Tietz, J. I.; Mitchell, D. A.; van der Donk, W. A. J. Am. Chem. Soc. 2015, 137, 16012.
(46) Layer, G.; Moser, J.; Heinz, D. W.; Jahn, D.; Schubert, W. D. EMBO J. 2003, 22, 6214.
(47) Layer, G.; Grage, K.; Teschner, T.; Schunemann, V.; Breckau, D.; Masoumi, A.; Jahn, M.; Heathcote, P.; Trautwein, A. X.;
Jahn, D. J. Biol. Chem. 2005, 280, 29038.
(48) Smith, D. M.; Buckel, W.; Zipse, H. Angew. Chem. Int. Ed. Engl. 2003, 42, 1867.
(49) Miller, L. M.; Chatterjee, C.; van der Donk, W. A.; Kelleher, N. L. J. Am. Chem. Soc. 2006, 128, 1420.
(50) Zhang, Z.; Hudson, G. A.; Mahanta, N.; Tietz, J. I.; van der Donk, W. A.; Mitchell, D. A. J. Am. Chem. Soc. 2016, 138,
15511.
(51) Markham, G. D.; DeParasis, J.; Gatmaitan, J. J. Biol. Chem. 1984, 259, 14505.
(52) Lanz, N. D.; Grove, T. L.; Gogonea, C. B.; Lee, K. H.; Krebs, C.; Booker, S. J. Methods Enzymol. 2012, 516, 125.
28
ACS Paragon Plus Environment

Page 29 of 29
Journal of the American Chemical Society
1
2
3
4
5
1
2
TOC Graphic
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
3
29
ACS Paragon Plus Environment