De novo Biosynthesis of “Non-Natural” Thaxtomin Phytotoxins

Article abstract of DOI:10.1002/anie.201801525

Thaxtomins are diketopiperazine phytotoxins produced by Streptomyces scabies and other actinobacterial plant pathogens that inhibit cellulose biosynthesis in plants. Due to their potent bioactivity and novel mode of action there has been considerable interest in developing thaxtomins as herbicides for crop protection. To address the need for more stable derivatives, we have developed a new approach for structural diversification of thaxtomins. Genes encoding the thaxtomin NRPS from S. scabies, along with genes encoding a promiscuous tryptophan synthase (TrpS) from Salmonella typhimurium, were assembled in a heterologous host Streptomyces albus. Upon feeding indole derivatives to the engineered S. albus strain, tryptophan intermediates with alternative substituents are biosynthesized and incorporated by the NRPS to deliver a series of thaxtomins with different functionalities in place of the nitro group. The approach described herein, demonstrates how genes from different pathways and different bacterial origins can be combined in a heterologous host to create a de novo biosynthetic pathway to “non-natural” product target compounds.

Full text of DOI:10.1002/anie.201801525

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
International Edition Chemie
www.angewandte.org
Accepted Article
Title: De novo Biosynthesis of 'Non-Natural' Thaxtomin Phytotoxins
Authors: Michael Winn, Daniel Francis, and Jason Micklefield
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.201801525
Angew. Chem. 10.1002/ange.201801525
Link to VoR: http://dx.doi.org/10.1002/anie.201801525
http://dx.doi.org/10.1002/ange.201801525

Angewandte Chemie International Edition
10.1002/anie.201801525
COMMUNICATION
De novo Biosynthesis of ‘Non-Natural’ Thaxtomin Phytotoxins
Michael Winn, Daniel Francis and Jason Micklefield*
Abstract: Thaxtomins are diketopiperazine phytotoxins produced by
Streptomyces scabies and other actinobacterial plant pathogens that
inhibit cellulose biosynthesis in plants. Due to their potent bioactivity
and novel mode of action there has been considerable interest in
developing thaxtomins as herbicides for crop protection. To address
the need for more stable derivatives, we have developed a new
approach for structural diversification of thaxtomins. Genes encoding
the thaxtomin NRPS from S. scabies, along with genes encoding a
promiscuous tryptophan synthase (TrpS) from Salmonella
typhimurium, were assembled in a heterologous host Streptomyces
albus. Upon feeding indole derivatives to the engineered S. albus
strain, tryptophan intermediates with alternative substituents are
biosynthesized and incorporated by the NRPS to deliver a series of
thaxtomins with different functionalities in place of the nitro group. The
approach described herein, demonstrates how genes from different
pathways and different bacterial origins can be combined in a
heterologous host to create a de novo biosynthetic pathway to ‘non-
natural’ product target compounds.
Many valuable therapeutic agents and agrochemicals are
derived from natural products. However, many of these
compounds do not have the necessary bioactivity or
physicochemical properties at the outset, and further
derivatization or analogue synthesis is required to develop
optimized, effective compounds. This optimization is traditionally
achieved through semi-synthesis or total synthesis. However,
Figure 1. Thaxtomin biosynthesis. (A) Thaxtomin congeners from
Streptomyces scabies. (B) The proposed biosynthesis of thaxtomins.
multistep chemical synthesis often requires deleterious reagents
and can be costly, which is particularly problematic for the
development of agrochemicals that need to be produced in large
quantities. Biosynthetic engineering is an alternative approach to
produce new and potentially improved natural product variants
directly by fermentation.[ Although biosynthetic engineering has
been used to generate novel ‘non-natural’ products[1-6] most
approaches have been limited to manipulating biosynthetic
pathways in the well characterized native producers. However
there are many microorganisms that produce useful secondary
metabolites which are not amenable to such genetic
manipulation. The rise of synthetic biology has provided access
Figure 1B).[20-23] TxtD is a nitric oxide (NO) synthase and TxtE, a
P450 enzyme, then utilizes this NO with O , to catalyze nitration
Nonribosomal peptide
synthetases (NRPS) TxtA and TxtB activate the L-phenylalanine
and L-4-nitrotryptophan precursors, catalyzing subsequent
condensation, N-methylation and cyclisation to give
(
2
[21,24]
at the 4-position of L-tryptophan.
1,2]
diketopiperazines including thaxtomin
D
(3).
A
P450
monooxygenase, TxtC, then hydroxylates the L-phenylalanine
residue at the -position and at either 2-, 3-, or 4-positions of the
aryl group resulting in a family of compounds (4-11) with
[7-9]
to larger synthetic DNA constructs, rapid DNA capture,
-
thaxtomin A (6) being the major metabolite produced by S.
editing,[
10-11]
assembly
[12-13]
and other advances.
[14,15]
The
scabies.[25-28] Their potent herbicidal activity and novel mode of
prospect of using these new tools to assemble de novo
biosynthetic pathways in well characterized heterologous host
strains[16,17] for diversification and optimization of natural products,
derived from less tractable microorganisms, is an attractive goal.
In this paper we describe a new approach for the de novo
biosynthesis of ‘non-natural’ variants of thaxtomin phytotoxins
[29-
action has driven the development of several synthetic routes.
36]
However, these routes are all multi-step, low yielding, and
produce diastereoisomeric mixtures. We therefore sought to
establish an alternative biosynthetic route for the production of
thaxtomin derivatives. Previous studies have established that the
4-nitro substituent of the L-tryptophan is important for bioactivity.
produced by Streptomyces scabies (Figure 1A), which inhibit
However this moiety is photolabile and can result in harmful
degradation products.[37,38] Consequently, thaxtomin derivatives
cellulose biosynthesis in plants.[
18,19]
Thaxtomins (1-11) are
biosynthesized from L-phenylalanine and the unusual L-4-
nitrotryptophan, which is produced by TxtD and TxtE enzymes
with alternative substituents at the C4 position of the indole ring
were identified as target compounds for synthetic biology.
In addition to being difficult to genetically manipulate, S.
scabies and other thaxtomin producers are plant pathogens with
tight regulations controlling their use and genetic modification. It
was thus desirable to establish heterologous expression of
thaxtomin biosynthetic genes in a more tractable strain. To this
Dr M. Winn, Dr D. Francis, Prof. J. Micklefield
School of Chemistry, Manchester Institute of Biotechnology, The University
of Manchester, 131 Princess Street, Manchester M1 7DN, UK,
E-mail: jason.micklefield@manchester.ac.uk
Supporting information is given via a link at the end of the document.
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.201801525
COMMUNICATION
(
27) was also detected in each of the cultures, including those that
were not supplemented with exogenous precursors. There is no
evidence in the literature of desnitro thaxtomins being produced
by S. scabies, and none were detected in cultures of S. scabies
grown for this study. Presumably in the wild type S. scabies L-4-
nitrotryptophan is present in abundance during thaxtomin
production which prevents TxtB loading L-tryptophan. Another
strain, Streptomyces ipomoeae, is reported to make desnitro
thaxtomin(s), this strain may be less efficient in producing L-4-
nitrotryptophan and/or may possess a more promiscuous NRPS
assembly line.[43]
While thaxtomin D (3) is known to possess some herbicidal
activity, the hydroxylated thaxtomin A (6) is a more potent
herbicide, mostly due to a gain in solubility rather than any direct
increase in biological activity.[28] To this end a third DNA part, with
optimized ribosomal binding and spacer sequence, was
constructed by amplifying the thaxtomin hydroxylase txtC from S.
scabies genomic DNA. All three DNA parts (txtA, txtB and txtC)
were then assembled together into a synthetic operon, within the
non-integrative vector, under the control of the ermE* promoter
and introduced into the S. albus host. The resulting strain S.
albus::txtABC (Figure 3, ii) was cultivated and supplemented with
Figure 2. Enzymatic synthesis of tryptophan derivatives using
tryptophan synthase from Salmonella typhimurium (StTrpS).
Reactions were carried out in 50 mL with 15 mM indole analogue, 30
mM L-Ser, 0.25 mM PLP, 100 mM KPi pH 7.6, 5 mL StTrpS cell lysate
and incubated at 37 °C for 72 hours.
end, Streptomyces albus was selected as a suitable host. Initially
a two-step process for production of thaxtomin derivatives was
envisaged; a biotransformation step utilizing tryptophan synthase
(
TrpS) to produce 4-substituted L-tryptophans (Figure 2) which
are then fed to an S. albus strain possessing the NRPS TxtA and
TxtB (Figure 3). Previous studies have shown that the TrpS from
Salmonella typhimurium (ATCC 37845) is promiscuous and will
condense a range of indole derivatives with L-serine or L-
threonine to yield L-Trp derivatives.[39-42] To explore if this enzyme
4-nitrotryptophan, which resulted in a major product with m/z
439.1616, which has an identical HPLC retention time and MS-
MS to a sample of thaxtomin A (6) (Δ 0.4 ppm) which was isolated
from S. scabies (Table S2, Figure S2). Two other minor products
could be used to generate a library of C4-substituted tryptophan
analogues, StTrpS was overproduced in E. coli BL21 and the
resulting cell lysate incubated with a variety of 4-substituted
indoles. The resulting tryptophan analogues were isolated with
yields ranging from 5-49% (Figure 2). For the second step in the
process (Figure 3), NRPS-encoding genes txtA and txtB were
amplified from S. scabies genomic DNA as two separate DNA
parts, with optimized ribosomal binding sites and spacer
sequences replacing the native RBS and spacer sequences of
each gene. These two DNA parts were then assembled into a
non-integrative vector under the control of the constitutive ermE*
promoter. This vector was conjugated into S. albus and the
resulting strain (Figure 3, i) was supplemented with L-4-
nitrotryptophan (12). After 3 days of fermentation the cells were
removed and the culture supernatant was purified by C18 RP-
HPLC affording thaxtomin D (3).
(
<10%) were also evident with m/z 439.1623 and 439.1621 (Δ 1
and 0.8 ppm), which are most likely regioisomers that have
undergone hydroxylation at the -position and at C2 or C4 of the
L-phenylalanine aryl group. These additional two peaks are also
present in samples of thaxtomin A isolated from S. scabies with
identical retention times and MS fragmentation patterns (Figure
S2). It has previously been shown that whilst thaxtomin A is the
main metabolite produced by S. scabies minor congeners that
have undergone hydroxylation at C2 or C4 (5 & 7, Figure 1A) can
also be isolated.48 Taken together with the results presented here,
this suggest that the P450 hydroxylase TxtC has relaxed
regioselectivity. In addition, S. albus::txtABC supplemented with
4
-nitrotryptophan also produced minor amounts of a metabolite
This strain was then supplemented with the tryptophan
derivatives 13–20 which resulted in new thaxtomin D analogues
possessing Cl, Br, F, or Me substituents at C4 of the indole moiety
(
23–26) (Figure 3, Table S1 and Figure S1A). Interestingly, other
than 4-nitrotryptophan, none of the tryptophan analogues
possessing more polar substituents were incorporated. This could
be due to substrate selectivity of the thaxtomin NRPS and/or less
efficient cellular uptake of the more polar L-Trp derivatives. To
explore if the structural diversity of thaxtomin derivatives could be
further expanded, 4-chloro- and 4-bromo-7-azatryptophans (21 &
22) were prepared (Figure 2) and fed to S. albus::txtAB. Whilst, 4-
chloro- and 4-bromotryptophan (13 & 14) are accepted, the
corresponding azatryptophans (21 & 22), which are the same size
but more polar, are not incorporated which further suggests that
hydrophobicity rather than steric factors appear to govern
selectivity of the NRPS. In addition to the five thaxtomin D
analogues, a new thaxtomin containing unmodified tryptophan
Figure 3. Heterologous biosynthesis of thaxtomin derivatives
from modified tryptophans. (i) The two NRPS genes required for
thaxtomin biosynthesis in S. scabies were assembled into a
synthetic operon and heterologously expressed in S. albus.
Feeding of tryptophan analogues to this strain resulted in new
thaxtomin D analogues. (ii) The inclusion of the P450 TxtC into the
pathway resulted in hydroxylated thaxtomin A analogues.
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.201801525
COMMUNICATION
Figure 4. A Synthetic Pathway for Biosynthesis of Thaxtomin D Derivatives Directly from Indole. The two NRPS genes required for
thaxtomin biosynthesis in S. scabies were optimized into a synthetic operon with codon optimized TrpS genes from Salmonella and
heterologously expressed in S. albus. Feeding of indole analogues to this strain resulted in thaxtomin D analogues being produced in vivo.
with m/z 407.1710, which was identical to thaxtomin D (3)(Δ 0.3
ppm), and a metabolite m/z 423.1667, consistent with the mono-
hydroxylated derivative 4 (Δ 0.4 ppm) (Figure S3). It has
previously been postulated that TxtC first hydroxylates the α-
carbon of the phenylalanine, before introducing a second hydroxyl
group on the aromatic ring as only thaxtomins singly hydroxylated
at the α position have been isolated from the wild type strain
analogues. Following several days incubation, thaxtomin D (3)
and analogues possessing Br, Cl, F, and Me substituents (23-26)
were again produced. In contrast, control experiments, where
indole analogues are fed to S. albus::txtAB or S. albus::txtABC
strains lacking the heterologous trpAB, produced only trace
amounts of thaxtomin analogues (Figure S6 and S7), which
illustrates the success of this de novo pathway (Figure S5). The
fact that S. albus::txtAB,trpAB failed to produce thaxtomin
derivatives from the indoles with polar substituents, despite the
increased lipophilicity of the indole versus tryptophan, indicates
that cellular uptake may not be the reason why these are not
accepted as substrates by the pathway, instead A-domain
specificity may be limiting analogue incorporation.
In summary, we have assembled a novel artificial pathway that
allows thaxtomin analogues to be generated in vivo in a single
fermentation step from indole precursors. TrpS has been widely
used to generate tryptophan analogues in vitro, however the work
described here shows how TrpS can also be exploited to generate
tryptophan derivatives in vivo providing alternative precursors for
incorporation into natural products. Given that there are many
natural products derived from tryptophan, this approach could be
adopted to create a wide range of new biosynthetic assembly
lines, in vivo, generating novel natural product variants.
Consequently, in addition to providing new ways of producing
more effective herbicides, this approach could also be applied to
diversify a wide range of other bioactive natural product scaffolds
including important antibiotics. One can envisage expanding this
concept, of combining multiple enzymes from different pathways
and origins in heterologous host strains, to create many novel
“non-natural” variants for agrochemical, therapeutic and other
applications.
(
compound 4 and 9, Figure 1A).[44]
Having established that S. albus::txtABC (Figure 3, ii) can
produce hydroxylated thaxtomin A, this strain was then
supplemented with tryptophan derivatives 13-16. This resulted in
new hydroxylated thaxtomin analogues with either Cl, Br, F or Me
group at the 4-position of the indole ring. As with the natural
pathway, the cultures fed with the three halogenated tryptophan
analogues (13-15) also resulted in several dihydroxylated
thaxtomins variants, with
a
single regioisomer forming
preferentially (>75%, Figure S4 i-iv, Table S2). Some residual non
hydroxylated and singly hydroxylated intermediates could also be
detected in each case. Methylated and desnitro thaxtomins
(
compounds 26 and 27) were also converted by TxtC into
hydroxylated analogues, however these two substrates resulted
in the formation of a singly hydroxylated compound predominantly
with around 25% conversion to the major doubly hydroxylated
product (Figure S4 v,vi).
It was then envisaged that a third strain could be developed that
was capable of producing thaxtomin analogues directly from
simple, commercially available indole starting materials by
introducing the tryptophan synthase genes, trpA and trpB, from S.
typhimurium, along with genes encoding the thaxtomin NRPS,
txtA and txtB, into the S. albus host (Figure 4). Using 4-substituted
indole as the starting material would also provide more lipophilic
substrates that may promote better cellular uptake than the more
polar tryptophan precursors. The trpA and trpB genes from S.
typhimurium were optimized for codon usage in Streptomyces
and used to create two further DNA parts with ribosome binding
and spacer sequences optimized which were then assembled
together with the S. scabies txtA and txtB into the non-integrative
ermE*p containing vector used before and conjugated into the S.
albus host. Cultures of this exconjugant (S. albus::txtAB,trpAB,
Figure 4) were supplemented with commercially available indole
Acknowledgements: We thank BBSRC (grant BB/K002341/1)
and Syngenta for funding. We also acknowledge Gregory Challis
(University of Warwick), Jason Vincent and Nick Mulholland
(Syngenta) for helpful discussion.
Keywords: Nonribosomal peptide • biosynthesis • biocatalysis •
biosynthetic engineering • synthetic biology
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.201801525
COMMUNICATION
[1]
M. Winn, J. K. Fyans, Y. Zhuo, J. Micklefield. Nat. Prod. Rep., 2016, 33,
17-347.
[41] D. Francis, M. Winn, J. Latham, M. F. Greaney, J. Micklefield.
Chembiochem, 2017, 18, 382-386.
3
[
2]
3]
K. J. Weissman. Nat. Prod. Rep., 2016, 33, 203-230.
[42] M. Winn, A. D. Roy, S. Grüschow, R. S. Parameswaran, R. J. M. Goss.
Bioorganic Med. Chem. Lett. 2008, 18, 4508-4510.
[
K. T. Nguyen, D. Ritz, J-Q. Gu, D. Alexander, M. Chu, V. Miao, P. Brian,
R. H. Baltz. Proc. Natl. Acad. Sci. U S A, 2006, 103, 17462-17467.
M. J. Calcott, J. G. Owen, I. L. Lamont, D. F. Ackerley. Appl. Environ.
Microb., 2014, 80, 5723-5731.
[43] R. R. King, C. H. Lawrence, L. A. Calhoun, J. B. Ristaino. J. Agric. Food
Chem. 1994, 42, 1791-1794.
[4]
[44] R. R. King, L. A. Calhoun. Phytochemistry, 2009, 70, 833-841.
[
5]
6]
A. Wlodek et al. Nat. Commun., 2017, 8.
[
J. Thirlway, R. Lewis, L. Nunns, M. Al Nakeeb, M. Styles, A. W. Struck,
C. P. Smith, J. Micklefield. Angew. Chem. Int. Ed., 2012, 51, 7181-7184.
J. Fu et al. Nat. Biotechnol., 2012, 30, 440-446.
[7]
[8]
[9]
D. Du, L. Wang, Y. Tian, H. Liu, H. Tan, G. Biu. Sci. Rep., 2015, 5, 8740.
K. Yamanaka, K. A. Reynold, R. D. Kersten, K. S. Ryan, D. J. Gonzalez,
V. Nizet, P. C. Dorrestein, B. S. Moore. Proc. Natl. Acad. Sci. USA, 2014,
111, 1957-1962.
[
[
[
10] R. E. Cobb, Y. Wang, H. Zhao. ACS Synth. Biol., 2014, 4, 723-728.
11] B. Zetsche et al. Nat. Biotechnol., 2017, 35, 31-34.
12] D. G. Gibson, L. Young, R-Y. Chuang, J. C. Venter, C. A. Hutchison III,
H. O. Smith. Nat. Methods, 2009, 6, 343-U341.
[13] D. Montiel, H. S. Kang, F. Y. Chang, Z. Charlop-Powers, S. F. Brady.
Proc. Natl. Acad. Sci. USA, 2015, 112, 8953-8958.
[
14] Y. Luo, B. Enghiad, H. Zhao. Nat. Prod. Rep., 2016, 33, 174-182.
15] M. J. Smanski, H. Zhou, J. Claesen, B. Shen, M. A. Fischbach, C. A.
Voigt. Nat. Rev. Microbiol., 2016, 14, 135-149.
[
[
[
[
[
[
16] M. Komatsu, T. Uchiyama, S. Omura, D. E. Cane, H. Ikeda. Proc. Natl.
Acad. Sci. U S A, 2010, 107, 2646-2651.
17] K. Zhou, K. J. Qiao, S. Edgar, G. Stephanopoulos. Nat. Biotechnol.,
2
015, 33, 377-U157.
18] V. Bischoff, S. J. Cookson, S. Wu, W. R. Scheible. J Exp. Bot., 2009, 60,
55-965.
9
19] W. R. Scheible, B. Fry, A. Kockevenko, D. Schindelasch, L. Zimmerli, S.
Somerville, R. Loria, C. R. Somerville. Plant Cell, 2003, 15, 1781-1794.
20] Kers, J. A.; Cameron, K.D.; Joshi, M.V.; Bukhalid, R.A.; Morello, J.E.;
Wach, M.J.; Gibson, D.M.; Loria, R. Mol. Microbiol., 2005, 55, 1025-
1033.
[21] J. A. Kers, M. J. Wach, S. B. Krasnoff, J. Widom, K. D. Cameron, R. A.
Bukhalid, D. M. Gibson, B. R. Crane, R. Loria. Nature, 2004, 429, 79-82.
[22] E. G. Johnson, S. B. Krasnoff, D. R. Bignell, W. C. Chung, T. Tao, R. J.
Parry, R. Loria, D. M. Gibson. Mol. Microbiol., 2009, 73, 409-418.
[23] R. R. King, C. H. Lawrence, M. C. Clark, M. L. A. Calhoun. J. Chem. Soc.,
Chem. Commun., 1989, 849-850.
[24] S. M. Barry et al. Nat. Chem. Biol., 2012, 8, 814-816.
[25] F. G. Healy, S. B. Krasnoff, M. Wach, D. M. Gibson, R. Loria. J.
Bacteriol., 2002, 184, 2019-2029.
[
26] R. R. King, C. H. Lawrence. J. Agric. Food Chem. 1996, 44, 1108-1110.
27] R. R. King, C. H. Lawrence, L. A. Calhoun. J. Agric. Food Chem. 1992,
[
4
0, 834-837.
28] R. R. King, C. H. Lawrence, J. A. Gray. J. Agric. Food Chem. 2001, 49,
298-2301.
29] D. E. Nitecki, B. Halpern, J. W. Westley. J. Org. Chem., 1968, 33, 864-
66.
[
[
2
8
[30] R. R. King. Can. J. Chem., 1997, 75, 1172-1173.
[31] J. Gelin, J. Mortier, J. Moyroud, A. Chene. J. Org. Chem., 1993, 58, 3473-
3
475.
32] J. Moyroud, J. Gelin, A. Chene, J. Mortier. Tetrahedron, 1996, 52, 8525-
534.
[
[
[
8
33] P. P. Molesworth, M. G. Gardiner, R. C. Jones, J. A. Smith, R. S. Tegg,
C. Wilson. Aust. J. Chem., 2010, 63, 813-820.
34] J. P. Bourgault, A. R. Maddirala, P. R. Andreana. Org. Biomol. Chem.,
2014, 12, 8125-8127.
[
[
[
[
[
[
35] H. Zhang et al. Org. Lett., 2013, 15, 5670-5673.
36] H. Zhang et al. J. Agric. Food Chem, 2015, 63, 3734-3741.
37] M. J. Strauss. Ind. Eng. Chem. Prod. RD. 1979, 18, 158-166.
38] K. S. Ju, R. E. Parales. Microbiol. Mol. Biol. Rev., 2010, 74, 250-272.
39] R. S. Phillips. Tetrahedron: Asymmetry, 2004, 15, 2787-2792.
40] R. J. M. Goss, P. L. A. Newill. Chem. Commun., 2006, 4924-4925.
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.201801525
COMMUNICATION
Entry for the Table of Contents (Please choose one layout)
Layout 1:
COMMUNICATION
Thaxtomins are phytotoxins with
potent bioactivity and a novel mode of
action. The thaxtomin NRPS from S.
scabies, along with genes encoding
tryptophan synthase (TrpS), from
Salmonella typhimurium, were
assembled in a S. albus heterologous
host. Feeding indole analogues to this
host allowed the biosynthesis of a
series of thaxtomin derivatives in a
single fermentation from commercially
available indole starting materials.
Michael Winn, Daniel Francis and Jason
Micklefield*
Page No. – Page No.
De novo Biosynthesis of ‘Non-
Natural’ Thaxtomin Phytotoxins.
.
This article is protected by copyright. All rights reserved.