Phytotoxic effects of selected N-benzyl-benzoylhydroxamic acid metallo-oxygenase inhibitors: Investigation into mechanism of action

Article abstract of DOI:10.1039/c3nj00491k

Treatment of Arabidopsis thaliana with 100 μM hydroxamic acids F1 and F2, found previously to inhibit carotenoid cleavage dioxygenase enzyme CCD1, was found to cause chlorophyll bleaching and phytotoxicity. A further set of hydroxamic acid analogues was synthesised, and these compounds were found to be phytotoxic towards A. thaliana at 16-400 μM, and to show some phytoxicity towards broad-leaved weeds C. album and S. media at 100 μM. Compound F1 was found to inhibit p-hydroxy-phenylpyruvate dioxygenase (HPPD), a known herbicide target (IC₅₀ 30 μM), but compounds F5 and F8 showed no inhibition of HPPD, despite F8 showing higher levels of phytotoxicity. Plants grown in the presence of F1 or F5 that were treated with 50 μM homogentisic acid showed partial recovery of growth, indicating some inhibition of HPPD in planta. These are the first hydroxamic acid inhibitors reported for HPPD, but the results indicate that inhibition of HPPD is only partly responsible for the observed phytotoxicity, and that another unknown metalloenzyme is also targeted by these compounds.

Full text of DOI:10.1039/c3nj00491k

NJC
View Article Online
View Journal | View Issue
PAPER
Phytotoxic effects of selected N-benzyl-
benzoylhydroxamic acid metallo-oxygenase inhibitors:
investigation into mechanism of action
Cite this: NewJ.Chem., 2013,
37, 3461
Martin J. Sergeant,a Peter J. Harrison,b Robert Jenkins,b Graham R. Moran,c
Timothy D. H. Bugg*b and Andrew J. Thompson*ad
Treatment of Arabidopsis thaliana with 100 mM hydroxamic acids F1 and F2, found previously to inhibit
carotenoid cleavage dioxygenase enzyme CCD1, was found to cause chlorophyll bleaching and phytotoxicity.
A further set of hydroxamic acid analogues was synthesised, and these compounds were found to be
phytotoxic towards A. thaliana at 16–400 mM, and to show some phytoxicity towards broad-leaved weeds
C. album and S. media at 100 mM. Compound F1 was found to inhibit p-hydroxy-phenylpyruvate dioxygenase
(HPPD), a known herbicide target (IC50 30 mM), but compounds F5 and F8 showed no inhibition of HPPD,
despite F8 showing higher levels of phytotoxicity. Plants grown in the presence of F1 or F5 that were treated
with 50 mM homogentisic acid showed partial recovery of growth, indicating some inhibition of HPPD
in planta. These are the first hydroxamic acid inhibitors reported for HPPD, but the results indicate that
inhibition of HPPD is only partly responsible for the observed phytotoxicity, and that another unknown
metalloenzyme is also targeted by these compounds.
Received (in Montpellier, France)
9th May 2013,
Accepted 29th June 2013
DOI: 10.1039/c3nj00491k
www.rsc.org/njc
Introduction
The carotenoid cleavage dioxygenases (CCDs) catalyse the oxidative
cleavage of a range of carotenoid substrates in plants, mammals
and bacteria (see Fig. 1A), to form apocarotenoid cleavage
products.1,2 In plants, which possess multiple CCD genes,
carotenoid oxidative cleavage is a key step in the biosynthesis
of abscisic acid,3 and in the biosynthesis of the recently-
discovered strigolactone shoot branching hormone,4,5 and
apocarotenoid natural products such as b-ionone and geranyl-
acetone plant volatiles.6
In order to study the function of members of the CCD family
using a chemical genetics approach, we have previously developed
a class of hydroxamic acid inhibitors of tomato CCD1, in which
varying the aryl–N distance in the inhibitor was found to
a Warwick HRI, University of Warwick, Wellesbourne CV35 9EF, Warwickshire, UK
b Department of Chemistry, University of Warwick, Coventry CV4 7AL, UK.
E-mail: T.D.Bugg@warwick.ac.uk; Fax: +44 (0)2476 524112;
Tel: +44 (0)2476 573018
Fig. 1 (A) Carotenoid oxidative cleavage reactions catalysed by CCD enzymes;
(B) design of CCD hydroxamic acid inhibitors.
c Department of Chemistry and Biochemistry, University of Wisconsin-Milwaukee,
3210 Cramer St., Milwaukee, WI 53201-0413, USA. E-mail: moran@uwm.edu;
Fax: +1 414 229 5530; Tel: +1 414 229 5031
d Cranfield Health, Cranfield University, Cranfield, Bedfordshire, MK43 0AL, UK.
E-mail: A.J.Thompson@cranfield.ac.uk; Fax: +44 (0)1234 758380;
Tel: +44 (0)755 7012423
influence the specificity of enzyme inhibition.7 Hydroxamic
acids in the D and F series showed inhibition of CCD1, a
9,10-cleavage enzyme (see Fig. 1B), and some compounds also
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem., 2013, 37, 3461--3465 3461

View Article Online
Paper
NJC
showed inhibition of 9-cis-epoxycarotenoid dioxygenase (NCED),
an 11,12-cleavage enzyme, and some compounds also showed a
shoot branching phenotype in planta.7 In the course of this
study, we observed that 2 out of a collection of 20 hydroxamic
acids, compounds F1 and F2, caused loss of chlorophyll (bleaching)
from leaves of A. thaliana when grown on agar plates. Since new
mechanisms for selective plant phytotoxicity are of potential
interest for crop protection, we have investigated this observa-
tion further, and we report structure–activity data and investiga-
tion of the mechanism of action of these compounds.
Fig. 3 Synthetic route for hydroxamic acids.
Results and discussion
Phytotoxicity of F1 & F2
Table 1 Phytotoxicity and inhibition of A. thaliana HPPD and LeCCD1 by
In our previous study, we found that hydroxamic acids in the D
and F series (see Fig. 1) inhibited the 9,10 cleavage enzyme
CCD1 in vitro.7 Unexpectedly, compounds F1 and F2 caused
bleaching and death of A. thaliana plants when incorporated at
100 mM into agar plates (see Fig. 2). None of the other Inhibitor
compounds exhibited this property, even compounds which
contained just an extra methoxy group at the 30 position on the
aryl ring. Since the D compounds showed similar or greater
activity against CCD1, yet did not show the bleaching pheno-
type, it seems unlikely that CCD1 is the target for phytotoxic
activity. Therefore, the molecular target for F1 and F2 may be a
non-CCD oxidative enzyme with a non-heme iron or other
metal ion co-factor.
hydroxamic acids
% inhibition
of A. thaliana % inhibition
Phyto- HPPD @
toxicity 50 mm
of LeCCD1
@ 100 mm
+
+
85
70
>95
>95
+
+
+
0
90
0
0
>95
30
Structure–activity studies
Further hydroxamic acids F5, F6 and F8 containing chlorine,
methoxy and/or amino substituents on the same carbon skeleton
were synthesized. The synthetic route, as described previously,7
involves alkylation of O-benzyl, N-Boc-hydroxylamine, followed
by Boc deprotection and coupling to the required substituted
benzoic acid (see Fig. 3).
The new analogues were found to be phytotoxic at 16–210 mM
against A. thaliana in agar media (see Tables 1 and 2), whereas
no bleaching was observed using hydroxamic acids D2 and D4,
containing a phenylacetyl rather than benzoyl substituent.
Compounds F6 showed strong inhibition of CCD1, comparable
with F1 and F2. The hydroxamic acids and a recently commer-
cialised HPPD-inhibiting herbicide tembotrione,8 were tested
ꢀ
ꢀ
90
60
>95
>95
for phytotoxicity against A. thaliana and broad-leaved weeds.
Table 2 LC50 values (in mM) for each inhibitor against A. thaliana and two broad-
leaved weeds Chenopodium album and Stellaria media, grown in agar. No
bleaching or phytotoxicity was observed with compounds D2 and D4
F1
16
740
670
F5
38
310
90
F6
F8
Tembotrione
Arabidopsis thaliana
Chenopodium album
Stellaria media
70
210
40
370
0.012
0.036
0.036
>1000
820
When applied in agar media, F1, F2, F5, and F6 showed activity
against A. thaliana at 100 mM concentration, with F8 showing
slightly lower activity (see Table 2). Compound F8 also showed some
herbicidal activity against broad-leaved weed Chenopodium album
Fig. 2 A. thaliana grown on agar media until flower initiation in the presence of
100 mM hydroxamic acid F1. Bleaching was not observed in the absence of F1.
c
3462 New J. Chem., 2013, 37, 3461--3465
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
NJC
Paper
at 100 mM concentration, and compound F5 was active against
Stellaria media, whereas compounds F1 and F2 were inactive
against these plants. Different phenotypic effects were seen:
compound F8 stunted the growth of the C. album seedlings and
turned them completely white; whereas compound F5 prevented
germination of S. media (data not shown).
Inhibition of A. thaliana p-hydroxyphenylpyruvate dioxygenase
p-Hydroxyphenylpyruvate dioxygenase (HPPD, see Fig. 4), a metallo-
oxygenase enzyme involved in ubiquinone biosynthesis in
plants, is a known herbicide target enzyme.9 Given the struc-
tural resemblance of F1 & F2 to the substrate p-hydroxyphenyl-
pyruvic acid, it seemed quite likely that inhibition of HPPD
might be the mechanism of action of these compounds.
Herbicide inhibitors of HPPD are known to act via chelation
of the iron(II) cofactor,9–11 which is the mechanism of inhibition
of CCD enzymes by these compounds, though no hydroxamic
acid inhibitors have been reported for HPPD.
Fig. 6 Growth of A. thaliana in the presence of 50 mM homogentisic acid (HGA)
and 100 mM hydroxamic acid F1 for 20 days.
increasing concentrations of F1 gave rise to enzyme inhibition
using this assay method (see Fig. 5A). Enzyme inhibition was
also observed using an HPLC assay method (see Fig. 5B).14 The
two methods gave somewhat different IC50 values of 30 mM and
100 mM respectively for compound F1. The level of inhibition
increased upon pre-incubation with enzyme over 10–20 min (data
not shown), indicating a time-dependent onset of inhibition.
Hydroxamic acids F6 and F7 also showed inhibition of A. thaliana
HPPD using the enolborate method, giving 90% and 85% inhibition
respectively at 50 mM concentration, however, compounds F5
and F8 showed no measurable inhibition of HPPD at 50 mM
concentration, despite F8 showing higher phytotoxicity (see
Table 1). Compounds D2 and D4 prepared previously,7 based
on substituted phenylacetic acids rather than benzoic acid, also
showed 90% and 60% inhibition of HPPD at 50 mM concen-
tration, but showed no phytotoxicity phenotype.
Recombinant A. thaliana HPPD12,13 was expressed in Escherichia
coli, and the enzyme was purified to homogeneity using a literature
protocol.13 The HPPD reaction can be monitored by complexation
of the enol form of p-hydroxyphenylpyruvic acid with borate,
giving rise to an absorbance at 320 nm.14 Incubation with
Fig. 4 Reaction catalysed by p-hydroxyphenylpyruvate dioxygenase (HPPD).
‘‘Metabolite rescue’’ experiment
Given the lack of correlation between HPPD inhibition and
phytotoxicity, a further experiment was carried out in which
A. thaliana plants grown in presence of 100 mM F1 or F5 were
treated with homogentisic acid (HGA), the product of the
HPPD-catalysed reaction. If HPPD inhibition was occurring in
the plant, then addition of HGA should restore the ability to
biosynthesise ubiquinone, and hence restore growth. As shown
in Fig. 6, treatment with HGA was found to partially alleviate
the toxicity caused by F1; mean plant fresh weight with F1 alone
(1.8 ꢁ 0.6 mg) was increased to 3.3 ꢁ 0.3 mg by the F1 + HGA
treatment (P o 0.05, n = three plates, 9 seeds per plate),
suggesting that some inhibition of HPPD is taking place in
the plant. Similar results were obtained with compound F5.
Conclusions
We have observed a selective phytotoxicity phenotype in a small
group of hydroxamic acids based upon substituted benzoic acid
skeletons. Compounds F1 and F2, as well as analogue F6,
inhibit the known herbicide target enzyme HPPD with IC50
values of o50 mM, and are the first hydroxamic acid inhibitors
reported for HPPD. Triketone herbicide inhibitors of HPPD
Fig. 5 Inhibition of A. thaliana p-hydroxyphenylpyruvate dioxygenase by hydroxamic
acid F1 using (A) HPLC assay, compared with a known HPPD inhibitor sulcotrione;9
(B) enol-borate assay. Controls lacked inhibitor and were with (+Con) or without
(ꢀCon) addition of enzyme.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem., 2013, 37, 3461--3465 3463

View Article Online
Paper
NJC
F8. dH (300 MHz, d6-DMSO) 7.8 (1H, s), 7.71 (1H, d, J = 8 Hz),
7.34 (5H, m), 6.95 (1H, d, J = 8 Hz), 4.83 (2H, s), 3.90 (3H, s);
dC (75 MHz, d6-DMSO) 156.0, 131.0, 130.8, 129.8, 128.8, 128.7,
128.5, 128.0, 120.0, 113.7, 56.7, 53.2, carbonyl not seen; HRMS
obs 292.0739, calc 292.0735 for C15H14ClNO3H+.
Fig. 7 Structures of metalloenzyme inhibitors that show herbicide activity.
Enzyme inhibition assays
A. thaliana p-hydroxyphenylpyruvate dioxygenase was over-
expressed and purified as previously described.11 The enol-borate
assay was performed essentially as described.10 Briefly, 4-hydroxy-
phenylpyruvic acid (pHPP, Sigma) was prepared in keto form by
dissolving in ethanol at 100 mM and then diluting to 10 mM
with 0.2 M TrisꢂHCl pH 7.5. A solution of 1.5 M sodium
ascorbate, 10 mM FeSO4, 25 mg mlꢀ1 catalase (Sigma C9322,
2000–5000 units mgꢀ1) and 10 mM pHPP was then prepared
and then diluted 100-fold with 0.8 M K2HPO4, 0.4 M H3BO3,
0.2 M TrisꢂHCl pH 7.5. This reaction mixture was allowed
to equilibrate at room temperature for 10 min to form the
enol-borate complex. To initiate the reaction, 0.9 ml of this
reaction mix was added to 100 ml of crude lysate from the
HPPD-expressing E. coli strain (containing 10 mM HEPES
pH 7.0 and 1 mM FeSO4). The O.D. at 320 nm was then followed
for 5 minutes. The HPLC assay for HPPD activity was performed
as previously reported12 except that purification of the enzyme
reactions was by precipitation in the presence of 50% methanol
followed by centrifugation at 20 000 g for 10 min; supernatant
was used for HPLC analysis.
have been shown to bind to the active Fe(II) form of the enzyme.10,11
The time-dependent inhibition observed with these hydroxamic
acids might be due to oxidation to a more tightly-bound Fe(III) form
of the enzyme after ligand binding.
Surprisingly, analogues F5 and F8, which show the phyto-
toxicity phenotype, do not inhibit HPPD. Furthermore, phenyl-
acetyl analogues D2 and D4, which do inhibit HPPD, are not
phytotoxic. Hence there is not a strong correlation between
HPPD inhibition and plant phytotoxicity, and these structure–
activity studies raise doubts about whether HPPD is the primary
target enzyme for the observed phytotoxicity.
A metabolite rescue experiment using 50 mM homogentisic
acid showed partial recovery of growth, implying that some
inhibition of HPPD is occurring in the plant. Therefore our
conclusion is that these compounds do inhibit HPPD in
the plant, but that there must be a second unknown metallo-
enzyme whose inhibition also leads to the observed phytotoxi-
city phenotype.
Other metalloenzyme herbicide targets have been described in
recent years: glutamine phosphoribosyl pyrophosphoryl amido-
transferase, involved in adenine biosynthesis, is the target for
bleaching herbicide DAS 734 (see Fig. 7);15 1-deoxyxylulose 5-phos-
phate reductoisomerase is inhibited by the hydroxamic acid
fosmidomycin (see Fig. 7);16 and imidazole glycerol-phosphate
dehydrogenase is involved in histidine biosynthesis.17 It is there-
fore possible that one of these enzymes is the alternative enzyme
target for phytotoxicity of these compounds.
Phytotoxicity testing
Plant agar-based growth media and environmental growth condi-
tions for Arabidopsis and for weed species were as described
previously.7 Agar media were supplemented with the indicated
inhibitors and/or HGA. Weed seeds were obtained from Herbiseed,
Berkshire, UK.
Acknowledgements
This work was supported by a research grant (BB/FOF/273)
from the Biotechnological and Biological Sciences Research
Council. We thank Nathanael Hsueh and Nicola Brookbank for
additional synthetic work.
Experimental section
Synthesis of hydroxamic acid compounds
Hydroxamic acids F1, F2, F5, F6, and F8 were synthesised using
the synthetic route shown in Fig. 3, previously described
in ref. 7, from starting materials 3-chlorobenzoic acid (F5),
3-nitrobenzoic acid (F6) and 3-chloro-4-methoxybenzoic acid
(F8). Compounds F1 and F2 were described in ref. 7. Data for
compounds F5, F6 and F8:
References
1 M. E. Aldridge, D. R. McCarty and H. J. Klee, Curr. Opin.
Plant Biol., 2006, 9, 315–321.
F5. dH (400 MHz, CDCl3) 7.51 (1H, s), 7.50 (1H, d, J = 8 Hz),
7.40 (1H, d, J = 8 Hz), 7.36 (1H, t, J = 8 Hz), 7.29 (5H, m), 4.8 (2H,
s); dC (75 MHz, CDCl3) 172.9, 134.8, 130.8, 130.4, 128.5, 128.4,
127.9, 127.8, 127.1, 126.6, 125.8, 53.75; HRMS obs 263.0661,
calc 263.0663 for C14H13ClNO2H+.
F6. dH (300 MHz, CDCl3) 7.35 (5H, m), 7.28 (1H, d, J = 8 Hz),
7.15 (1H, t, J = 8 Hz), 6.86 (1H, s), 6.77 (1H, d, J = 8 Hz), 4.84 (2H,
s); dC (75 MHz, CDCl3) 172.1, 142.9, 129.0, 128.7, 128.6, 128.5,
128.5, 128.4, 128.0, 116.2, 113.8, 53.8; HRMS obs 243.1131, calc
243.1128 for C14H15N2O2H+.
2 G. Giuliano, S. Al-Babili and J. von Lintig, Trends Plant Sci.,
2003, 8, 145–149.
3 I. B. Taylor, T. Sonneveld, T. D. H. Bugg and A. J. Thompson,
J. Plant Growth Regul., 2005, 24, 253–273.
4 M. Umehara, A. Hanada, S. Yoshida, K. Akiyama, T. Arite,
N. Takeda-Kamiya, H. Magome, Y. Kamiya, K. Shirasu,
K. Yoneyama, J. Kyozuka and S. Yamaguchi, Nature, 2008,
455, 195–201.
5 V. Gomez-Roldan, S. Fermas, P. B. Brewer, V. Puech-Pages,
E. A. Dun, J.-P. Pillot, F. Letisse, R. Matusova, S. Danoun,
c
3464 New J. Chem., 2013, 37, 3461--3465
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
NJC
Paper
´
J.-C. Portais, H. Bouwmeester, G. Becard, C. A. Beveridge, 12 I. M. Fritze, L. Linden, J. Freigang, G. Auerbach, R. Huber
C. Rameau and S. F. Rochange, Nature, 2008, 455, 189–195. and S. Steinbacher, Plant Physiol., 2004, 134, 1388–1400.
6 M. H. Walter and D. Strack, Nat. Prod. Rep., 2011, 28, 663–692. 13 V. M. Purpero and G. R. Moran, Biochemistry, 2006, 45,
7 M. J. Sergeant, J.-J. Li, C. Fox, N. Brookbank, D. Rea, 6044–6055.
T. D. H. Bugg and A. J. Thompson, J. Biol. Chem., 2009, 14 K. Grossman and T. Ehrhardt, Pest Manage. Sci., 2007, 63,
284, 5257–5264. 429–439.
8 W. Schulte and H. Kocher, Bayer CropSci. J., 2009, 62, 35–52. 15 T. A. Walsh, T. Bauer, R. Neal, A. O. Merlo, P. R. Schmitzer,
¨
9 D. L. Lee, M. P. Prisbylla, T. H. Cromartie, D. P. Dagarin,
S. W. Howard, W. M. Provan, M. K. Ellis, T. Fraser and
L. C. Mutter, Weed Sci., 1997, 45, 601–609.
G. R. Hicks, M. Honma, W. Matsumura, K. Wolff and
J. P. Davies, Plant Physiol., 2007, 144, 1292–1304.
´
16 C. Zingle, L. Kuntz, D. Tritsch, C. Grosdemange-Billiard and
10 M. Kavana and G. R. Moran, Biochemistry, 2003, 42,
10238–10245.
11 J. Brownlee, K. Johnson-Winters, D. H. T. Harrison and
G. R. Moran, Biochemistry, 2004, 43, 6370–6377.
M. Rohmer, Bioorg. Med. Chem. Lett., 2012, 22, 6563–6567.
17 S. C. Sinha, B. N. Chaudhuri, J. W. Burgher, G. Yakovleva,
V. J. Davisson and J. L. Smith, J. Biol. Chem., 2004, 279,
15491–15498.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem., 2013, 37, 3461--3465 3465