Enzymatic Thioamide Formation in a Bacterial Antimetabolite Pathway

Article abstract of DOI:10.1002/anie.201804158

6-Thioguanine (6TG) is a DNA-targeting therapeutic used in the treatment of various cancers. While 6TG was rationally designed as a proof of concept for antimetabolite therapy, it is also a rare thioamide-bearing bacterial natural product and critical virulence factor of Erwinia amylovorans, plant pathogens that cause fire blight. Through gene expression, biochemical assays, and mutational analyses, we identified a specialized bipartite enzyme system, consisting of an ATP-dependent sulfur transferase (YcfA) and a sulfur-mobilizing enzyme (YcfC), that is responsible for the peculiar oxygen-by-sulfur substitution found in the biosynthesis of 6TG. Mechanistic and phylogenetic studies revealed that YcfA-mediated 6TG biosynthesis evolved from ancient tRNA modifications that support translational fidelity. The successful in vitro reconstitution of 6TG thioamidation showed that YcfA employs a specialized sulfur shuttle that markedly differs from universal RNA-related systems. This study sheds light on underexplored enzymatic C?S bond formation in natural product biosynthesis.

Full text of DOI:10.1002/anie.201804158

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Enzymatic Thioamide Formation in Bacterial Antimetabolite
Pathway
Authors: Christian Hertweck
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201804158
Angew. Chem. 10.1002/ange.201804158
Link to VoR: http://dx.doi.org/10.1002/anie.201804158
http://dx.doi.org/10.1002/ange.201804158

10.1002/anie.201804158
Angewandte Chemie International Edition
DOI: 10.1002/anie.200((will be filled in by the editorial staff))
Biosynthesis
Enzymatic Thioamide Formation in Bacterial Antimetabolite Pathway
Agnieszka Litomska, Keishi Ishida, Kyle L. Dunbar, Marco Boettger, Sébastien Coyne, and Christian
Hertweck*
Abstract: 6-Thioguanine (6TG) is a DNA-targeting therapeutic used
in the treatment of various cancers. While 6TG was rationally
designed as a proof of concept for antimetabolite therapy, it is also a
rare thioamide-bearing bacterial natural product and critical
virulence factor of Erwinia amylovorans, plant pathogens causing
fire blight. Through gene expression, biochemical assays and
mutational analyses we identified a specialized bipartite enzyme
system, consisting of an ATP-dependent sulfur transferase (YcfA)
and a sulfur-mobilizing enzyme (YcfC), responsible for the peculiar
oxygen-by-sulfur substitution. Mechanistic and phylogenetic studies
revealed that YcfA-mediated 6TG biosynthesis evolved from
ancient tRNA modifications that support translational fidelity. The
successful in vitro reconstitution of 6TG thioamidation showed that
YcfA employs a specialized sulfur shuttle that markedly differs
from universal RNA-related systems. This study sheds light on
underexplored enzymatic C–S bond formations in natural product
biosynthesis.
clinical use for the treatment of various severe diseases including
psoriasis, inflammatory bowel disease, as well as chronic and acute
leukemias.[5]
Only recently, 6TG was identified as critical virulence factor of the
pathogenic bacterium Erwinia amylovora, the causative agent of fire
blight.[6] This highly contagious plant disease is known to rapidly
destroy apple and pear trees, causing huge economic damage.[7]
Functional gene analyses revealed that 6TG biosynthesis is encoded
by a small gene locus (ycfA–D) (Figure 1B) that is unique to the fire
blight pathogen and closely related Erwinia species.[6] Yet, the
thioamide-forming enzyme(s) and the timing of thioamide formation
remained enigmatic. Here we report the successful in vitro
reconstitution of 6TG thioamidation by a bipartite enzyme system
consisting of an adenylating enzyme (YcfA) and a designated sulfur
shuttle enzyme (YcfC), which operates independently of the general
sulfur mobilization pathways.
By targeted gene inactivation in E. amylovora we previously found
that YcfA is a key enzyme in the biosynthetic pathway.[6a] YcfA is
homologous to proteins of the adenine nucleotide alpha hydrolase-
like (AANH-like) superfamily (Figure 1C), a diverse protein family
with a conserved ATP-binding alpha/beta/alpha fold. Among the
many subfamilies of the AANH-like enzymes are the enzymes
responsible for sulfur incorporation on tRNA nucleobases (MnmA,
TtcA, and ThiI).[8] Although YcfA is highly divergent from these
characterized thiolation enzymes, the critical role that YcfA plays in
6TG biosynthesis suggested that this protein could be responsible
for thioamide formation. Therefore, we produced YcfA in
Escherichia coli and purified the protein by affinity chromatography
(Figure S2). Acquiring soluble and pure YcfA was highly
challenging due to issues with aggregation and stability. Only after
optimization of the expression and purification conditions were we
able to obtain sufficient amounts of soluble protein to investigate its
function.
Sulfur is an element that has played a key role in the development
of life and is essential to all known living beings. Oxygen-by-sulfur
substitutions in biomolecules emerged early in evolution with the
thiolation of tRNA nucleobases.[1] These thio modifications stabilize
the RNA structure and confer fidelity in ribosomal peptide synthesis
across all domains of life (Figure 1A).[2] Beyond these biopolymers,
the replacement of oxygen with sulfur may dramatically alter the
functions of small molecules.[3] Thioamides, sulfur-containing
peptide bond surrogates, are markedly different from amides, and
their chemical properties are often appreciated in therapeutics.[4] An
important example of a thioamide-containing drug is 6-thioguanine
(6TG, 1), a rationally designed RNA- and DNA-targeting cytotoxin.
Originally synthesized as proof of concept for antimetabolite
therapy by Nobel Prize winners Elion and Hitchings, it is now in
As the timing of thioamide formation in the 6TG pathway was
unknown, all enzyme assays were performed in parallel with a
number of potential precursors: guanine (2), guanosine (3),
guanosine monophosphate (GMP, 4), guanosine diphosphate (GDP,
5), and guanosine triphosphate (GTP, 6) (Figure 2A). Since
inorganic sulfide has previously been shown to function as a sulfur
source for AANH enzymes in vitro,[9] we added lithium sulfide to
the reaction mixtures and product formation was monitored by
HPLC. However, none of the substrates were converted to the
corresponding thioamide; a representative trace with 4 is displayed
in Figure 2B.
[∗]
A. Litomska, Dr. K. Ishida, Dr. K. L. Dunbar, M. Boettger, Dr. S.
Coyne, Prof. Dr. C. Hertweck
Department of Biomolecular Chemistry
Leibniz Institute for Natural Product Chemistry and Infection
Biology (HKI)
Beutenbergstr. 11a, 07745 Jena, Germany
Fax: (+) 49 3641 5320804
E-mail: christian.hertweck@leibniz-hki.de
Homepage ((optional)): www.leibniz-hki.de
For characterized RNA-thiolation enzymes, the sulfur nucleophile
often originates from L-cysteine through the action of the cysteine
desulfurase IscS in vivo (Figure 1).[1] Thus, we expressed the IscS
homolog from E. amylovora (Ea-IscS) in E. coli and purified the
protein by affinity chromatography (Figure S2). Reactions were
performed with YcfA, Ea-IscS, L-cysteine, ATP, PLP and the full
range of guanine substrates (compounds 2–6). While control
experiments demonstrated that Ea-IscS catalyzed sulfide release
from L-cysteine (Figure S3), no thiolated products were detected
under the conditions tested; a representative trace with 4 is
displayed in Figure 2B.
Prof. Dr. C. Hertweck
Natural Product Chemistry
Friedrich Schiller University, Jena, Germany
[∗∗] Financial support by the Humboldt foundation for a postdoctoral
fellowship (to K. D.) and by the DFG (Leibniz prize to C. H.) is
gratefully acknowledged.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
1
This article is protected by copyright. All rights reserved.

10.1002/anie.201804158
Angewandte Chemie International Edition
was comparable to the strain containing the ycfABC expression
plasmid. Thus, YcfA and YcfC are necessary and sufficient for
heterologous 6TG biosynthesis (Figure 2C).
Figure 1. Biosynthesis of 6TG evolved from ancient tRNA
modifications. A) The key enzyme in 1 biosynthesis (YcfA) is
homologous to AANH-like enzymes responsible for thiolation of tRNA
nucleobases that support translational fidelity and efficiency. The
sulfur source for such RNA modifications originates from L-cysteine
via cysteine desulfurase (IscS) activity. The sulfur source for the
thioamide modification in 1 is unknown. B) Organisation of the 6TG
biosynthesis gene cluster. C) Neighbor-joining tree of diverse AANH
family members with 1,000 bootstrap replicates. The cladogram
includes sulfur transferases (YcfA, MnmA, TtcA, ThiI, and
homologues) and other AANH enzymes from primary metabolism
(GMP synthase, NAD+ synthase, TilS, and homologues). Structures
of the enzymatic products of each AANH subclass are displayed with
the linkage installed by AANH colored red. Red asterisks indicate
proteins from E. amylovora. For a more detailed phylogenetic tree,
see Supplementary.
Figure 2. In vitro reconstitution of the thioamide formation via
YcfA-YcfC bipartite enzyme system. A) Substrates used in this
manuscript along with the predicted thioamide-containing products
are displayed. B) Representative HPLC traces (λ = 340 nm) from in
vitro assays for thioamide formation by YcfA and different sulfur
sources (Li2S, Ea-IscS + L-Cys, or YcfC + L-Cys) with 4 as a substrate
are shown. C) HPLC profiles of extracts from liquid cultures of E. coli
heterologously expressing fragments of the 6TG biosynthetic gene
cluster. D) Determination of sulfide production by YcfC using the
methylene blue method. Error bars represent the standard deviation
from the mean (n = 3). E) General scheme depicting the thioamide-
formation assay with YcfA and YcfC. F) HPLC traces (λ = 340 nm)
from in vitro assays with different substrates are shown. All new
peaks were identified by comparison to synthetic reference
compounds. G) HPLC-UV traces of YcfA/YcfC reactions with
substrate 4 (λ = 260 nm) are shown with the pertinent reference
compounds. H) HPLC traces (λ = 340 nm) of YcfA/YcfC reactions
lacking each component demonstrate that all components are
required for robust activity. I) Rates of thiobase formation in
YcfA/YcfC enzyme reactions with different concentrations of 4, 5, and
6. Error bars represent the standard deviation from the mean (n = 3).
The inability to identify an appropriate sulfur donor for YcfA
prompted us to revisit the genetic basis of 6TG biosynthesis. We
previously demonstrated that high levels of 6TG production were
achieved in E. coli using an expression plasmid containing the
ycfABC gene cassette.[6a] In contrast, only trace amounts of 6TG are
produced in the absence of ycfC. Since YcfB is homologous to
members of the NUDIX hydrolase protein family, we reasoned that
this protein is likely not involved in thioamide formation. Thus, we
constructed an E. coli strain expressing ycfA and ycfC. HPLC-
UV/MS analysis of this strain demonstrated that 6TG production
2
This article is protected by copyright. All rights reserved.

10.1002/anie.201804158
Angewandte Chemie International Edition
amount of
8
produced in the YcfA-YcfC reaction to the
concentrations of phosphate and pyrophosphate formed (correlating
to ADP and AMP formation, respectively; Figure 3B). Although
low levels of phosphate were produced, the stoichiometry of
pyrophosphate to 8 was approximately one-to-one, providing strong
evidence for the adenylation of 4 by YcfA during thiolation.
Although the reason for the accumulation of ADP in reactions
containing YcfA is unknown, we suspect that it arises from either a
side reaction of YcfA or is attributable to contaminating proteins in
the enzyme preparations. To provide further support for the
adenylation activity of YcfA we generated variants of YcfA with
point mutations in the predicted ATP-binding pocket, YcfAD19A and
YcfAS20A (Figure 3C). These YcfA mutants were deficient in
thioamide formation in vivo (Figure 3D). To confirm these results,
we aimed at testing the mutated enzymes in vitro. Due to the low
Although the gene product of ycfC is annotated as a hypothetical
protein, in-depth in silico analysis using I-TASSER and HHpred
suggested that YcfC is distantly related to pyridoxal phosphate
(PLP)-dependent transferases including cysteine desulfurases and
carbon-sulfur lyases. As some of these enzymes are involved in
sulfur activation and incorporation in other biosynthetic pathways,[1,
3, 10]
we postulated that YcfC could provide the sulfur required for
YcfA activity.
To test this model, YcfC was heterologously produced in E. coli and
purified by affinity chromatography (Figure S2). As with YcfA,
YcfC could only be obtained in small quantities and was susceptible
to aggregation. Therefore, freshly purified YcfC was incubated with
ʟ-cysteine, and sulfide release was determined by the methylene
blue method. As YcfC co-purifies with only trace amounts of PLP
(Figure S2), the reaction was supplemented with additional PLP.
Consistent with the bioinformatically predicted function of YcfC,
robust sulfide release was only observed in PLP-supplemented YcfC
reactions (Figure 2D); only minute levels of sulfide were detected in
reactions lacking exogenous PLP.
solubility of protein we were not able to obtain mutant YcfAS20A
.
The variant with the D19A mutation, however, could be produced in
soluble form and was purified by affinity chromatography (Figure
S2). The in vitro assay performed with YcfAD19A did not result in
thioamide formation (Figure 3E). These results showed that YcfA
requires ATP for substrate activation.
With YcfC in hand, we returned to the reconstitution of thioamide
formation by YcfA. Reactions were performed with freshly purified
enzymes, PLP, ATP, the aforementioned panel of guanine
derivatives (compounds 2–6), and L-cysteine as a sulfur source
(Figure 2E). Although product peaks were not observed in the
HPLC profiles of reactions with guanine and guanosine, new peaks
with a UV absorbance at 340 nm, which is characteristic for 6TG,
were formed in assays with each of the three tested guanine
nucleotides (Figure 2F). By comparison with synthetic references
we unequivocally identified these new compounds as the
corresponding thioamides, 6-thio GMP (6TGMP, 8, Figure 2G), 6-
thio GDP (6TGDP, 9) and 6-thio GTP (6TGTP, 10) (Figure S4).
Additionally using LC-HMRS and by comparison to reference
compounds we confirmed that new compounds are thioamides. Data
for 8 formation is presented in Figures S5-6. Notably, products 8-10
were not observed in assays lacking any of the reaction components
(YcfA, YcfC, Mg2+, ATP, or L-cysteine; Figure 2H). Consistent
with the low level of sulfide production observed for YcfC reactions
performed without PLP supplementation (Figure 2D), only minute
amounts of 8 could be detected when PLP was not present in the
assay. In addition to the expected product, 2 h reactions with GDP
and GTP produced low amounts of 6TGTP and 6TGDP,
respectively. These side products are likely due to impurities in the
enzyme preparations.
These results clearly show that thioamide formation in 6TG
biosynthesis occurs prior to cleavage of the glycosidic bond and that
it requires both YcfA and YcfC. In order to gain further insight into
the substrate specificity of YcfA, we monitored the rate of
thioamidation with increasing concentrations of compounds 4-6.
Due to the high concentration of enzyme required to observe a
measurable rate of product formation in our enzyme assays, we were
unable to obtain Michaelis-Menten parameters for the substrates.
Nonetheless, the kinetic curves indicate that all three phosphorylated
substrates are biotransformed by YcfA into thiolated products with
similar efficiency in vitro (Figure 2I).
As previously characterized AANH-like thiolating enzymes
adenylate their substrates prior to sulfide addition and because ATP
was required for thioamide formation in the YcfA-YcfC assay, we
reasoned that YcfA adenylates the guanine base prior to sulfide
attack, which would result in AMP production following
elimination. To test this hypothesis, enzyme reactions containing
YcfA, YcfC or both were performed with 4, and the ATP hydrolysis
products were monitored by HPLC. Consistent with the
bioinformatic predictions, only reactions containing YcfA showed
robust ATP hydrolysis (Figure 3A). Unexpectedly, detectable levels
of ADP were produced in addition to AMP. Therefore, in an attempt
to clarify the mechanism of substrate activation, we compared the
Figure 3. YcfA is an ATP-dependent sulfur transferase. A) HPLC
traces (λ = 260 nm) of the ATP hydrolysis products and 8 formation in
reactions with YcfA. B) Quantification of enzyme reaction products.
The concentrations of phosphate and pyrophosphate produced in
enzyme reactions were determined using a malachite green assay.
The concentration of 8 was calculated from HPLC chromatograms of
enzyme reactions by comparison to a standard curve. Error bars
represent the standard deviation from the mean (n = 3). C) An
alignment of the conserved ATP-binding domains of YcfA and diverse
AANH superfamily members. Apart from YcfA, all sequences are from
3
This article is protected by copyright. All rights reserved.

10.1002/anie.201804158
Angewandte Chemie International Edition
Received: ((will be filled in by the editorial staff))
Published online on ((will be filled in by the editorial staff))
E. coli. The ATP-binding PP-loop is colored red. Mutants made in this
study are highlighted. D) HPLC profiles (λ = 340 nm) of extracts from
liquid cultures of E. coli heterologously expressing: ycfABCD,
ycfABCD with the D19A point mutation in YcfA, and ycfABCD with the
S20A point mutation in YcfA. E) HPLC traces (λ = 340 nm) of in vitro
assays with YcfA and YcfAD19A. F) A model for sulfur mobilization and
transfer in 6TG thioamide formation mediated by the YcfA-YcfC
bipartite enzyme system.
Keywords: Biosynthesis · Enzymology · DNA Antimetabolite ·
Natural Product · Thioamide
[1]
[2]
[3]
[4]
K. A. Black, P. C. Dos Santos, Biochim. Biophys. Acta 2015,
1853, 1470-1480.
a) N. Shigi, Front. Genet. 2014, 5, 67; b) C. Zheng, K. A. Black,
P. C. Dos Santos, Biomolecules 2017, 7, 33.
K. L. Dunbar, D. H. Scharf, A. Litomska, C. Hertweck, Chem.
Rev. 2017, 117, 5521-5577.
a) D. S. Cooper, New Engl. J. Med. 2005, 352, 905-917; b) S.
Thee, A. Garcia-Prats, P. Donald, A. Hesseling, H. Schaaf,
Tuberculosis 2016, 97, 126-136; c) F. Wang, R. Langley, G.
Gulten, L. G. Dover, G. S. Besra, W. R. Jacobs, J. C. Sacchettini,
J. Exp. Med. 2007, 204, 73-78; d) P. Karran, N. Attard, Nat. Rev.
Cancer 2008, 8, 24-36.
In conjunction with our previous in vivo analyses and mutagenesis
studies,[6a] the successful in vitro reconstitution corroborates and
refines the model for thioamide formation in 6TG biosynthesis
(Figure 3F). Several lines of experimental evidence demonstrate that
YcfA activates guanosine nucleotides by adenylating the carbonyl
oxygen. Then, a cysteine-derived sulfur nucleophile is provided by
YcfC and transferred onto YcfA. Previously displayed mutagenesis
data suggests that the transfer of the sulfur nucleophile from YcfA
[5]
[6]
a) G. B. Elion, Biosci. Rep. 1989, 9, 509-529; b) M. Stanulla, H. J.
Schünemann, Lancet 2006, 368, 1304-1306.
to the activated substrate likely employs
a reactive sulfur
a) S. Coyne, C. Chizzali, M. N. Khalil, A. Litomska, K. Richter,
L. Beerhues, C. Hertweck, Angew. Chem. Int. Ed. 2013, 52,
10564-10568; b) S. Coyne, A. Litomska, C. Chizzali, M. N.
Khalil, K. Richter, L. Beerhues, C. Hertweck, ChemBioChem
2014, 15, 373-376.
nucleophile bound to one of the cysteine residues in the YcfA active
site.[6a] Following sulfur insertion, AMP would be lost to afford the
thioamide.
[7]
a) M. Malnoy, S. Martens, J. L. Norelli, M.-A. Barny, G. W.
Sundin, T. H. Smits, B. Duffy, Annu. Rev. Phytopathol. 2012, 50,
475-494; b) J. Mansfield, S. Genin, S. Magori, V. Citovsky, M.
Sriariyanum, P. Ronald, M. Dow, V. Verdier, S. V. Beer, M. A.
Machado, Mol. Plant Pathol. 2012, 13, 614-629; c) J. L.
Vanneste, Fire blight: the disease and its causative agent,
Erwinia amylovora, CABI, London, 2000.
a) D. Bouvier, N. Labessan, M. Clémancey, J. M. Latour, J. L.
Ravanat, M. Fontecave, M. Atta, Nucleic Acids Res. 2014, 42,
7960-7970; b) R. Kambampati, C. T. Lauhon, Biochemistry 2003,
42, 1109-1117; c) E. G. Mueller, P. M. Palenchar, C. J. Buck, J.
Biol. Chem. 2001, 276, 33588-33595.
M. Chen, S.-i. Asai, S. Narai, S. Nambu, N. Omura, Y. Sakaguchi,
T. Suzuki, M. Ikeda-Saito, K. Watanabe, M. Yao, N. Shigi, Y.
Tanaka, Proc. Natl. Acad. Sci. U. S. A. 2017, 114, 4954-4959.
D. Kessler, FEMS Microbiol. Rev. 2006, 30, 825-840.
a) S. Banala, R. D. Süssmuth, ChemBioChem 2010, 11, 1335-
1337; b) S. Behnken, T. Lincke, F. Kloss, K. Ishida, C. Hertweck,
Angew. Chem. Int. Ed. 2012, 51, 2425-2428; c) B. J. Burkhart, C.
J. Schwalen, G. Mann, J. H. Naismith, D. A. Mitchell, Chem. Rev.
2017, 117, 5389-5456; d) G. E. Kenney, A. C. Rosenzweig, ACS
Chem. Biol. 2012, 7, 260-268; e) T. Lincke, S. Behnken, K.
Ishida, M. Roth, C. Hertweck, Angew. Chem. Int. Ed. 2010, 49,
2011-2013.
F. Kloss, S. Pidot, H. Goerls, T. Friedrich, C. Hertweck, Angew.
Chem. Int. Ed. 2013, 52, 10745-10748.
a) A. B. Charette, M. Grenon, J. Org. Chem. 2003, 68, 5792-
5794; b) T. Ozturk, E. Ertas, O. Mert, Chem. Rev. 2007, 107,
5210-5278; c) D. L. Priebbenow, C. Bolm, Chem. Soc. Rev. 2013,
42, 7870-7880.
a) N. Mahanta, A. Liu, S. Dong, S. K. Nair, D. A. Mitchell, Proc.
Natl. Acad. Sci. U. S. A. 2018, 115, 3030-3035; b) D. D. Nayak,
N. Mahanta, D. A. Mitchell, W. W. Metcalf, eLife 2017, 6.
G. E. Kenney, L. M. K. Dassama, M. E. Pandelia, A. S. Gizzi, R.
J. Martinie, P. Gao, C. J. DeHart, L. F. Schachner, O. S. Skinner,
S. Y. Ro, X. Zhu, M. Sadek, P. M. Thomas, S. C. Almo, J. M. J.
Bollinger, C. Krebs, N. L. Kelleher, A. C. Rosenzweig, Science
2018, 359, 1411-1416.
In recent years, a growing number of thioamide-bearing natural
products have been discovered.[11] In cases where it was
investigated, the thioamide functionality in these compounds has
been shown to be an essential pharmacophore or chelating
residue.[11d, 12] Whereas the chemical synthesis of thioamides with
phosphorus sulfides (Lawesson’s reagent) and other reagents is
already well established,[13] the biochemical basis of enzymatic
amide thionation has remained obscure. Only recently, the
posttranslational thioamidation of a protein, methyl-coenzyme M
reductase,[14] and a ribosomal peptide, methanobactin,[15] have been
uncovered. In the case of methyl-coenzyme M reductase maturation,
an ATP-dependent YcaO enzyme catalyzes thionation, likely
through activation of the carbonyl oxygen with ATP. In
[8]
[9]
[10]
[11]
methanobactin biosynthesis,
a
metalloenzyme from an
uncharacterized protein family catalyzes thioamide formation in an
ATP-independent manner. While the finer details of these
mechanisms of thioamide formation remain to be deciphered, it is
notable that an AANH-like protein is not involved in either instance,
highlighting the diversity of enzyme classes used in nature for the
synthesis of thioamides. Our report on the first in vitro
reconstitution of thioamide formation in a non-peptidic natural
product expands our limited understanding of enzymatic thioamide
formation and sets the stage for an in-depth investigation of the
sulfur transfer mechanism.
[12]
[13]
Phylogenetic and functional studies revealed that the YcfA-
mediated enzymatic C-S bond formation on the guanine nucleobase
is related to the ancient tRNA modifications.[2] This finding is
particularly intriguing from an evolutionary point of view, as the
life-sustaining biosynthetic machinery in these plant-pathogenic
bacteria has evolved to produce a natural product toxin. Illuminating
the key biosynthetic steps in the biosynthesis of the clinically and
ecologically important antimetabolite has further practical
implications. Insight into the functions of the thiolation enzymes
enables the genomics-driven discovery of new thioamide-containing
natural products and may inspire biocatalytic oxygen-by-sulfur
substitutions. Finally, since both YcfA and YcfC are specialized
enzymes required for the biosynthesis of a pivotal virulence factor,
our work also sets a starting point for controlling one of the top 10
plant pathogens.[7b, 7c]
[14]
[15]
Experimental Section
See Supporting Material.
4
This article is protected by copyright. All rights reserved.

10.1002/anie.201804158
Angewandte Chemie International Edition
Entry for the Table of Contents
Biosynthesis
Primordial Pathway. The biosynthesis
of the thioamide moiety of the potent
DNA antimetabolite 6-thioguanine was
successfully reconstituted in vitro.
Thioamidation involves an ATP-
A. Litomska, K. Ishida, K. L. Dunbar, M.
Boettger, S. Coyne, and C.
Hertweck*__________ Page – Page
YcfA YcfC
dependent biocatalyst that evolved from
ancient RNA-modifying enzymes, and a
designated sulfur shuttle.
Bipartite
enzyme system
Enzymatic Thioamide Formation in
Bacterial Antimetabolite Pathway
5
This article is protected by copyright. All rights reserved.