Biochemical characterization and selective inhibition of β-carotene cis-trans isomerase D27 and carotenoid cleavage dioxygenase CCD8 on the strigolactone biosynthetic pathway

Article abstract of DOI:10.1111/febs.13400

The first three enzymatic steps of the strigolactone biosynthetic pathway catalysed by β-carotene cis-trans isomerase Dwarf27 (D27) from Oryza sativa and carotenoid cleavage dioxygenases CCD7 and CCD8 from Arabidopsis thaliana have been reconstituted in vitro, and kinetic assays have been developed for each enzyme, in order to develop selective enzyme inhibitors. Recombinant OsD27 shows a UV-visible λ_max at 422 nm and is inactivated by silver(I) acetate, consistent with the presence of an iron-sulfur cluster that is used in catalysis. OsD27 and AtCCD7 are not inhibited by hydroxamic acids that cause shoot branching in planta, but OsD27 is partially inhibited by terpene-like hydroxamic acids. The reaction catalysed by AtCCD8 is shown to be a two-step kinetic mechanism using pre-steady-state kinetic analysis. Kinetic evidence is presented for acid-base catalysis in the CCD8 catalytic cycle and the existence of an essential cysteine residue in the CCD8 active site. AtCCD8 is inhibited in a time-dependent fashion by hydroxamic acids D2, D4, D5 and D6 (> 95% inhibition at 100 μm) that cause a shoot branching phenotype in A. thaliana, and selective inhibition of CCD8 is observed using hydroxamic acids D13H and D15 (82%, 71% inhibition at 10 μm). The enzyme inhibition data imply that the biochemical basis of the shoot branching phenotype is due to inhibition of CCD8.

Full text of DOI:10.1111/febs.13400

Biochemical characterization and selective inhibition
of b-carotene cis–trans isomerase D27 and carotenoid
cleavage dioxygenase CCD8 on the strigolactone
biosynthetic pathway
Peter J. Harrison1, Sophie A. Newgas1, Flora Descombes1, Sarah A. Shepherd1,
Andrew J. Thompson2 and Timothy D. H. Bugg1
1 Department of Chemistry, University of Warwick, Coventry, UK
2 Cranfield Soil and Agrifood Institute, Cranfield University, Cranfield, UK
Keywords
The first three enzymatic steps of the strigolactone biosynthetic pathway
catalysed by b-carotene cis–trans isomerase Dwarf27 (D27) from
Oryza sativa and carotenoid cleavage dioxygenases CCD7 and CCD8 from
carotenoid cleavage dioxygenase; CCD7;
CCD8; D27; enzyme inhibition; hydroxamic
acid; strigolactone biosynthesis
Arabidopsis thaliana have been reconstituted in vitro, and kinetic assays
have been developed for each enzyme, in order to develop selective enzyme
Correspondence
inhibitors. Recombinant OsD27 shows a UV-visible kmax at 422 nm and is
T. D. H. Bugg, Department of Chemistry,
inactivated by silver(I) acetate, consistent with the presence of an iron–sul-
fur cluster that is used in catalysis. OsD27 and AtCCD7 are not inhibited
University of Warwick, Coventry CV4 7AL,
UK
Fax: +44 02476 524112
by hydroxamic acids that cause shoot branching in planta, but OsD27 is
partially inhibited by terpene-like hydroxamic acids. The reaction catalysed
Tel: +44 02476 573018
E-mail: T.D.Bugg@warwick.ac.uk
by AtCCD8 is shown to be a two-step kinetic mechanism using pre-steady-
state kinetic analysis. Kinetic evidence is presented for acid–base catalysis
(Received 29 April 2015, revised 21 July
in the CCD8 catalytic cycle and the existence of an essential cysteine resi-
due in the CCD8 active site. AtCCD8 is inhibited in a time-dependent
fashion by hydroxamic acids D2, D4, D5 and D6 (> 95% inhibition at
100 lM) that cause a shoot branching phenotype in A. thaliana, and selec-
tive inhibition of CCD8 is observed using hydroxamic acids D13H and
D15 (82%, 71% inhibition at 10 lM). The enzyme inhibition data imply
that the biochemical basis of the shoot branching phenotype is due to inhi-
bition of CCD8.
2015, accepted 4 August 2015)
doi:10.1111/febs.13400
Introduction
Carotenoids, long chain C40 polyene terpenoids pro-
duced by phototrophic organisms, are broken down to
apocarotenoids by a family of enzymes known as the
carotenoid cleavage dioxygenases (CCDs) [1,2]. CCDs
are non-heme-iron dependent enzymes that catalyse
the oxidative cleavage of the polyene backbone of a
carotenoid or apocarotenoid substrate. Vp14 (NCED1)
from Zea mays was the first biochemically character-
ized CCD enzyme [3] and is responsible for the 11,12
cleavage of 90-cis-neoxanthin and/or 90-cis-violaxanthin
in the biosynthesis of the plant hormone abscisic acid,
an inhibitor of seed germination and mediator of plant
responses to drought [4–6].
Since the discovery of Vp14, CCDs have been iden-
tified across all taxa and have been shown to be
involved in the biosynthesis of
a
range of
apocarotenoids. In mammals, the enzyme b-carotene
oxygenase I is required for the biosynthesis of visual
pigment 11-cis-retinal [7,8]. In plants, strigolactones
have been identified as hormones involved in
Abbreviations
CCD, carotenoid cleavage dioxygenase; D27, Dwarf27; GST, glutathione-S-transferase.
3986
FEBS Journal 282 (2015) 3986–4000 ª 2015 FEBS

P. J. Harrison et al.
Inhibition of isomerase D27 and dioxygenase CCD8
rhizosphere signalling and as inhibitors of shoot
branching [9–11]. Biosynthesis of strigolactones
requires the action of two CCD enzymes, CCD7 and
CCD8, which act sequentially on 9-cis-b-carotene, as
shown in Fig. 1 [12]. The biosynthetic pathway begins
with the isomerization of all-trans-b-carotene to 9-cis-
b-carotene catalysed by Dwarf27 (D27) [12,13]. Oxida-
tive cleavage of 9-cis-b-carotene at the 9ʹ,10ʹ position
is then catalysed by CCD7 to produce 9-cis-b-apo-100-
carotenal [12]. CCD8 catalyses an unusual double
oxygenation of 9-cis-b-apo-10ʹ-carotenal to produce
an intermediate known as carlactone [12]. Carlactone
is then oxidized by consecutive P450 enzyme oxida-
tions to give ent-2ʹ-epi-5-deoxystrigol: recombinant
Arabidopsis Max1 has been shown to form car-
lactonoic acid [14], whereas a recombinant Max1
homologue from Oryza sativa (called carlactone oxi-
dase) has been shown to form ent-2ʹ-epi-5-deoxystrigol
directly [15].
compounds, the most effective compound showing a
Ki of 93 lM [21].
A collection of substituted hydroxamic acid com-
pounds have been shown to inhibit CCD enzymes
NCED1 and CCD1 from Solanum lycopersicum [22].
Against NCED1, the most effective compound, D8,
showed 40% inhibition at 100 lM, whilst against
CCD1 IC50 values of < 1 lM were reported [22].
In planta phenotypic effects have been observed in
Arabidopsis with a number of different hydroxamic
acid inhibitors. Compounds D4 and D6 were shown to
cause an increase in the number of shoot branches,
suggesting inhibition of a CCD enzyme involved in
strigolactone biosynthesis [22]. Additionally, smaller
hydroxamic acids F1 and F2 have been shown to
cause a bleaching phenotype, due partly to inhibition
of the enzyme p-hydroxyphenylpyruvate dioxygenase,
but also a second unknown metalloenzyme target [23].
A phenotypic effect on lateral root branching has been
observed with the compound D15, believed to result
from inhibition of the biosynthesis of an unknown
apocarotenoid signalling molecule [24].
A number of CCD enzymes have been structurally
characterized, revealing a conserved seven-bladed b-
propeller motif within the protein structure [16,17].
The catalytic iron(II) centre is coordinated by four his-
tidine residues, an unusual ligand set for non-heme-
iron dependent dioxygenase enzymes [2,18].
To date, no selective inhibitors of the CCD enzymes
CCD7 and CCD8 have been reported. Here we report
further biochemical characterization of the first three
enzymes on the strigolactone biosynthesis pathway,
screening of these enzymes against previously reported
and new hydroxamic acid based compounds, and the
identification of inhibitors for D27 and CCD8.
In view of the physiological effects of abscisic acid
and strigolactones, there is interest in the development
of selective inhibitors for the CCDs [2]. The first com-
pounds shown to inhibit CCDs were the tertiary amine
compounds abamine and abamineSG, which inhibit
NCED on the abscisic acid biosynthesis pathway
[19,20]. AbamineSG was shown to inhibit Vigna un-
guiculata NCED1 with a Ki of 18.5 lM [20]. In planta,
both abamine and abamineSG inhibited abscisic acid
biosynthesis [19,20]. Inhibition of Arabidopsis thaliana
NCED3 has been reported by sesquiterpene-like
Results
Preparation of novel hydroxamic acids
In addition to the collection of hydroxamic acids
described previously [22–24], eight novel hydroxamic
Fig. 1. The strigolactone biosynthesis pathway from all-trans-b-carotene to ent-20-epi-5-deoxystrigol.
FEBS Journal 282 (2015) 3986–4000 ª 2015 FEBS
3987

Inhibition of isomerase D27 and dioxygenase CCD8
P. J. Harrison et al.
acids were synthesized, using a modification of the
synthetic route previously described [22]. In order to
mimic the terpenoid backbone of the strigolactone
biosynthetic pathway intermediates, three terpene-like
hydroxamic acids were prepared: compound D30 con-
tains an N-substituted C10 geranyl unit, while com-
pounds D31 and D32 contain C-substituted C10 and
C12 terpenoid units respectively. Since CCD8 catalyses
oxidative cleavages more distant from the carbocyclic
ring of its substrate b-apo-100-carotenal, hydroxamic
acid B2 containing a five-atom spacer between the
oxygenated aromatic ring and the hydroxamic acid
functional group and compounds D12H and D13H
containing a 2-carbon spacer were available from
another study of metallo-oxygenase inhibition [25].
Finally, two compounds D20 and D21 containing a
more water-soluble carboxylic sidechain were also
prepared. The chemical structures are shown in
Table 1.
can generate a second reaction product in the reverse
reaction, with kmax 443 nm. Incubation of purified
9-cis-b-carotene under the enzyme assay conditions
gave rise to no measurable increase in all-trans-b-caro-
tene, indicating that the background rate of non-enzy-
matic isomerization is low. The KM for 9-cis-b-
carotene in the reverse reaction was determined to be
0.26 lM, with a kcat value of 34 sꢀ1
.
OsD27 has been previously reported to contain iron
[15], but the precise nature of the iron cofactor is
uncertain. The purified recombinant OsD27 from our
protocol was found to show a UV-visible absorbance
maximum at 422 nm (Fig. 3), in agreement with that
observed by Lin et al. [15] and consistent with kmax
values observed for proteins containing [2Fe2S] clus-
ters [29,30]. In order to probe the identity of this spe-
cies, recombinant OsD27 was treated with mild
oxidizing and reducing agents. In contrast to the
observations of Lin et al., no change in the absorbance
spectra was observed on addition of 100 lM sodium
dithionite, and no loss of catalytic activity was
observed upon addition of either 100 lM sodium
dithionite or 100 lM hydrogen peroxide. However, a
decrease in intensity of the peak at 422 nm was
observed on addition of 100 lM silver(I) acetate
(Fig. 3), a reagent that is reported to inactivate pro-
teins containing [4Fe4S] clusters [31]. Addition of sil-
ver acetate was also found to completely inhibit the
isomerization of 9-cis-b-carotene by OsD27 via HPLC
assay (Fig. 4B).
Biochemical characterization of OsD27
OsD27 was expressed in Escherichia coli as a glu-
tathione-S-transferase (GST) fusion protein and puri-
fied using affinity chromatography. The D27-catalysed
reaction was monitored in vitro using C30 reverse
phase HPLC, observing the change in ratio between
the all-trans (retention time 23.9 min) and 9-cis (reten-
tion time 24.5 min) isomers of b-carotene. Isomers of
b-carotene were identified by their UV-visible spectra
with reference to published values [12,26,27]. Commer-
cially available b-carotene was found to contain
approximately 2% 9-cis-b-carotene (Fig. 2C). Addition
of recombinant D27 to the commercially available b-
carotene resulted in little change in the ratio of the
two isomers (Fig. 2D). In order to test the D27 reac-
tion in vivo, the pGEX-OsD27 expression vector was
transformed into genetically engineered E. coli strain
pAC-BETA, which accumulates b-carotene [28], and
the carotenoid products were extracted for HPLC
analysis. As shown in Fig. 2B, introduction of the
expression vector containing OsD27 leads to an
increase of 31% in the amount of 9-cis-b-carotene
compared to pAC-BETA alone.
In order to probe further the mechanism of isomer-
ization catalysed by OsD27, a pH-rate profile of
OsD27 was determined, which showed < 2-fold varia-
tion in the rate of isomerization of 9-cis to all-trans-b-
carotene over the pH range 4–10 (Fig. 4A); hence
there is no apparent acid–base catalysis. Additionally,
the reaction was carried out anaerobically under an
atmosphere of nitrogen, which resulted in the same
ratio of 9-trans : 9-cis products (Fig. 4B), indicating
that OsD27 does not require dioxygen for activity.
Inhibition of OsD27 by hydroxamic inhibitors
Hydroxamic acid inhibitors were tested against OsD27
at a final concentration of 100 lM (Table 1) by moni-
toring the isomerization of 9-cis-b-carotene to all-
trans-b-carotene. At 100 lM only partial inhibition was
observed: most inhibition was observed by the terpene-
like compounds D30, D31 and D32 (41%, 38% and
33% inhibition respectively) and by compounds B2
(40% inhibition) and D12H (40% inhibition). Com-
pounds D2, D4, D5 and D6 that show a shoot branch-
ing phenotype in planta [22] gave no inhibition at all,
The reverse reaction catalysed by OsD27 was then
examined by HPLC purification of 9-cis-b-carotene
(Fig. 2E), followed by addition of recombinant
OsD27, which resulted in the formation of 2 : 1 all-
trans-b-carotene : 9-cis-b-carotene with
a calculated
equilibrium constant of 0.53 between the isomers
(Fig. 2F). A smaller additional peak at retention time
21.6 min was observed in the HPLC trace for the
D27-catalysed reaction (Fig. 2F), indicating that D27
3988
FEBS Journal 282 (2015) 3986–4000 ª 2015 FEBS

P. J. Harrison et al.
Inhibition of isomerase D27 and dioxygenase CCD8
Table 1. Inhibition data for OsD27 at 100 lM and for AtCCD8 at 100 lM and 10 lM. The chemical structures of hydroxamic acids are shown
below.
Inhibitor
CCD8
D27
Inhibition at
100 lM (%)
Inhibition at
Inhibition at
Class
Name
X
Y
100 lM (%)
10 lM (%)
Aryl-CON(OH)-aryl
F1
4-OMe
H
10
16
9
> 95
> 95
> 95
> 95
> 95
> 95
> 95
> 95
> 95
> 95
> 95
> 95
70
< 5
< 5
< 5
< 5
< 5
21
< 5
< 5
17
9
F2
4-OMe
F
F3
3,4-(OMe)2
3,4-(OMe)2
3-Cl
H
F4
F
25
21
7
F5
H
F6
3-NH2
H
Aryl-C1-CON(OH)-aryl
D1
4-OH
H
0
D2
4-OH
F
0
D4
4-OMe
F
0
D5
3,4-(OMe)2
3,4-(OMe)2
3,4-OCH2O-
4-OMe
H
0
D6
F
0
53
25
–
D7
F
0
Aryl-C1-CON(OH)-alkyl
Aryl-C2-CON(OH)-Y
D30
D20
D21
D9
C10H17
41
26
33
0
4-OMe
(CH2)2CO2Na
47
–
Naphthyl
4-OMe
(CH2)2CO2Na
92
–
CH2Ph
CH2PhF
C8H17
C8H17
H
> 95
> 95
> 95
> 95
> 95
78
54
71
< 5
< 5
56
–
D15
D10
D11
D12
D12H
D13
D13H
B1
4-OMe
0
3,4-(OMe)2
4-OMe
0
0
3,4-(OMe)2
3,4-(OMe)2
4-OMe
0
H
40
0
H
> 95
> 95
94
47
82
< 5
–
4-OMe
H
0
Ring-C5-CON(OH)-aryl
Alkyl-CON(OH)-Y
Cyclohexyl
4-MeOPh
C9H15
H
0
B2
CH2PhF
CH2Ph
CH2Ph
–
40
38
33
0
0
D31
D32
Abamine
52
–
C11H17
47
–
–
> 95
< 5
FEBS Journal 282 (2015) 3986–4000 ª 2015 FEBS
3989

Inhibition of isomerase D27 and dioxygenase CCD8
P. J. Harrison et al.
1
190
90
A
B
1.4
1.2
1
2
3
–10
95
0.8
0.6
0.4
0.2
0
1
45
–5
2
3
3
300
350
400
450
500
550
600
Wavelength (nm)
55
35
15
–5
C
OsD27.2 (+)
Series1
OsD27.2
Series2
1
silver acetate
2
Fig. 3. UV-visible spectrum of OsD27 (blue line), showing an
absorbance peak at kmax 422 nm, and spectrum of OsD27
following incubation with silver acetate (red line).
55
35
15
–5
D
E
F
1
mid pBB541 expressing GroES/EL. Recombinant
AtCCD7 was purified by GST affinity chromatogra-
phy. The CCD7 cleavage reaction was monitored
in vitro using C30 reverse HPLC (Fig. 5). The presence
of product b-apo-10ʹ-carotenal was confirmed by high
resolution mass spectrometry and was assigned as the
9-cis isomer by comparison of the kmax (444 nm) to
published values [9]. A KM value of 8.7 lM was deter-
mined for CCD7 using the HPLC-purified 9-cis-b-car-
otene substrate.
3
2
140
90
2
40
1
–10
50
30
10
Hydroxamic acid compounds and abamine were
tested as inhibitors of AtCCD7 at a final concentration
of 100 lM. Surprisingly, none of the compounds tested
was found to show any inhibition of the AtCCD7-
catalysed reaction, nor was any inhibition observed
after pre-incubation with enzyme. Hence AtCCD7 is
much less susceptible to inhibition by hydroxamic
acids than other CCD enzymes studied [22].
1
2
3
–10
30
0
10
20
Time (min)
Fig. 2. HPLC analysis of the reaction catalysed by OsD27. (A) All-
trans-b-carotene extracted from pAC-BETA E. coli (empty vector
control). (B) b-carotene extracted from pAC-BETA pGEX-OsD27
E. coli incubation (in vivo transformation by D27). (C) Authentic all-
trans-b-carotene (95%, Sigma-Aldrich) control. (D) Reaction of all-
trans-b-carotene with recombinant OsD27 in 100 mM MOPS buffer
pH 6.4. (E) Purified 9-cis-b-carotene. (F) Reaction of 9-cis-b-
carotene with recombinant OsD27 in 100 mM MOPS buffer pH
6.4. Peak assignments: 1, all-trans-b-carotene; 2, 9-cis-b-carotene;
3, isomer of b-carotene.
Biochemical characterization of AtCCD8
AtCCD8 lacking the first 168 bp (corresponding to a
chloroplast transit peptide) was expressed with an
N-His6 affinity tag and purified by N-His6 affinity
chromatography. Substrate b-apo-100-carotenal was
generated by the preceding enzyme CCD7. Monitoring
of the CCD8 reaction by C30 reverse phase HPLC
analysis showed the disappearance of substrate b-apo-
10ʹ-carotenal, and monitoring by UV-visible spec-
troscopy was found to give a time-dependent decrease
in absorbance at 430 nm, which was dependent upon
the concentration of enzyme added, with no change
observed using deactivated enzyme (Fig. 6A,B). A new
product peak was observed via HPLC analysis on a
C18 reverse phase column at retention time 15.2 min,
while low levels of inhibition were observed with the
shorter compounds F1–6.
Biochemical characterization of AtCCD7
AtCCD7 was expressed with a GST fusion tag and
showed improved expression in E. coli containing plas-
3990
FEBS Journal 282 (2015) 3986–4000 ª 2015 FEBS

P. J. Harrison et al.
Inhibition of isomerase D27 and dioxygenase CCD8
2.4
A
2.2
2
1.8
1.6
1.4
1.2
1
4.5
5
5.5
6
6.5
7
7.5
pH
8
8.5
9
9.5 10 10.5
4
B
2.5
Fig. 5. HPLC analysis of the CCD7-catalysed reaction of 9-cis-b-
carotene. (A) 9-cis-b-carotene. (B) 9-cis-b-carotene plus CCD7. Peak
assignments: 1, 9-cis-b-carotene; 2, all-trans-b-carotene; 3, 9-cis-b-
apo-10ʹ-carotenal.
2
1.5
1
tion. Kinetic data were modelled using the analysis
software GRAPHPAD, and a KM value of 9.2 lM and
specific activity of 2.48 9 10ꢀ1 lmolꢁmgꢀ1ꢁsꢀ1 were
determined for the 9-cis-b-apo-10ʹ-carotenal substrate,
giving an estimated kcat value of 0.18 sꢀ1. The pre-
steady-state kinetic behaviour of AtCCD8 was investi-
gated using UV-visible stopped flow kinetic analysis.
Upon mixing of a 1 : 1 molar ratio of AtCCD8 to sub-
strate 9-cis-b-apo-10ʹ-carotenal, a decrease in absor-
bance at 440 nm was observed over 40 s (Fig. 7).
Analysis of the data showed that the decay at 440 nm
could be fitted accurately by a biphasic exponential
0.5
0
Fig. 4. (A) pH-rate profile for the OsD27-catalysed isomerization of
9-cis-b-carotene to all-trans-b-carotene. (B) Effect of additives on
the ratio of all-trans-b-carotene: 9-cis-b-carotene following the
reaction of 9-cis-b-carotene with recombinant OsD27. Reactions are
shown with the respective control for each experiment (black bars).
function,
with
kinetic
parameters
A = 0.139,
k1 = 0.43 sꢀ1 and k2 = 0.06 sꢀ1. While it is of some
concern that k2 is slower than the estimated kcat value,
calculation of kcat involves the protein concentration
estimated by Bradford assay using bovine serum albu-
min as standard, which could over- or under-estimate
the true protein concentration, and also involves the
extinction coefficient for 9-cis-b-apo-10ʹ-carotenal,
which may vary in magnitude between different buffers
and pH values.
which showed kmax 267 nm, matching the data of
Alder et al. for the product carlactone [12]. Attempts
to characterize this peak by electrospray mass spec-
trometry were unsuccessful, suggesting that the pro-
duct is chemically unstable. Although Alder et al.
obtained mass spectral data for carlactone, the enol
ether functional group present in carlactone is suscep-
tible to hydrolysis in aqueous solution [32]. Treatment
of the CCD8 organic extract with alkaline sodium
borohydride gave new species at m/z 209.2 and 143.1
by electrospray mass spectrometry, consistent with the
expected structures of the reduced C14 portion of car-
lactone and the reduced C8 by-product of the CCD8
reaction, respectively, as shown in Fig. 6C. Our data
are therefore consistent with the CCD8-dependent con-
version of b-apo-10ʹ-carotenal to carlactone [12], which
in our hands was chemically unstable.
A pH-rate profile of CCD8-catalysed consumption
of 9-cis-b-apo-10ʹ-carotenal shows inflexions at pH 6.0
and pH 8.0 (Fig. 8), indicating the presence of a cat-
alytic base of pKa 6.0 and a catalytic acid of pKa 8.0.
The identity of active site groups was probed via incu-
bation of AtCCD8 with a range of group-specific
reagents at 1 mM concentration. Complete loss of
activity was observed upon treatment of AtCCD8 with
cysteine-directed reagents N-ethyl maleimide or iodoac-
etamide, consistent with the presence of an active site
cysteine residue [33,34]. Loss of activity was also
observed using 1-ethyl-3-(3-dimethylaminopropyl) car-
bodiimide, or using diethyl pyrocarbonate, but no inhi-
bition was observed using succinic anhydride (Fig. 8).
Kinetic experiments revealed that substrate inhibi-
tion was observed at > 15 lM substrate concentra-
FEBS Journal 282 (2015) 3986–4000 ª 2015 FEBS
3991

Inhibition of isomerase D27 and dioxygenase CCD8
P. J. Harrison et al.
0.11
0.1
A
A
0.01 s
1.95 s
4.8 s
0.13
0.125
0.12
0.115
0.09
0.08
0.07
0.06
0.05
0.04
34.6 s
73.4 s
111.4 s
149.8 s
0
200
1 x CCD8
400
2 x CCD8
B
0.53
0.52
0.51
0.5
188.2 s
375 400 425 450 475 500 525 550
Wavelength (nm)
0.49
0.48
0.47
B
0.14
0.135
0.13
0
100
200
300
400
500
Time (s)
0.125
0.12
C
0.115
0.11
0
10
20
30
40
Time (s)
440 nm
Fit
Fig. 7. (A) UV-visible scanning of the CCD8-catalysed reaction of 9-
cis-b-apo-10ʹ-carotenal at 0.01–200 s. (B) Decrease in absorbance
at 440 nm (dashed line) observed from stopped flow analysis of
the CCD8 cleavage reaction at 0.01–40 s, with observed fit to
second-order exponential (dotted line).
reaction was seen by many hydroxamic acids in the D
and F series at 100 lM concentration. Inhibition was
found to be strongly time-dependent, giving only a
small amount of inhibition with no pre-incubation but
> 95% inhibition after 10 min pre-incubation. Inhibi-
tion was also observed using abamine, but no inhibi-
tion was observed using the larger abamineSG.
Inhibition of AtCCD8 was only observed with a smal-
ler group of compounds at 10 lM concentration. Most
effective inhibition was observed with compounds
D13H (82%), D15 (71%), D12 (56%) and D9 (54%)
containing a 2-carbon unit between the aryl group and
the hydroxamic acid functional group. Inhibition was
observed with compounds D6 (53%), D4 (17%) and
D5 (9%) containing a 1-carbon spacer that show a
shoot branching phenotype in planta [22], but less inhi-
bition was observed with compound D2 and with the
shorter F series in which the hydroxamic acid is linked
directly to the hydroxamic acid. Compounds with
Fig. 6. Monitoring of CCD8-catalysed cleavage of 9-cis-b-apo-10ʹ-
carotenal by UV-visible spectroscopy and mass spectrometry. (A)
Decrease in absorbance at 430 nm observed for one- and two-fold
concentrations of CCD8. (B) Absorbance change observed for
boiled enzyme control. (C) Derivatives of carlactone and C8 product
detected by electrospray mass spectrometry.
Inhibition of AtCCD8 by hydroxamic acids
AtCCD8 was screened against the collection of
hydroxamic inhibitors initially at a concentration of
100 lM and then at 10 lM for the most effective inhi-
bitors (Table 1). Inhibition of the CCD8 cleavage
3992
FEBS Journal 282 (2015) 3986–4000 ª 2015 FEBS

P. J. Harrison et al.
Inhibition of isomerase D27 and dioxygenase CCD8
–9
lead to product inhibition in the forward direction
and hence the low conversion in the trans to cis
direction.
A
–9.5
–10
The UV-visible spectrum of purified OsD27 was
found to contain an absorbance peak at kmax 422 nm,
agreeing with the value reported by Lin et al. [13] and
similar to kmax values shown by proteins containing
[2Fe2S] iron–sulfur clusters. This absorbance peak was
removed by addition of silver(I) acetate, which is
known to cause the breakdown of iron–sulfur clusters
[31]. Catalytic activity was also lost by treatment with
silver(I) acetate, implying that this cofactor is required
for catalysis, although the resistance to treatment with
oxidizing and reducing agents implies that it is not
readily accessible by solvent. These two observations
suggest that OsD27 contains a [2Fe2S] iron–sulfur
cluster that is required for catalysis.
–10.5
–11
–11.5
4
5
6
7
pH
8
9
10
100
90
80
70
60
50
40
30
20
10
0
B
In order to probe the mechanism of isomerization
catalysed by OsD27, a pH-rate profile was carried
out, but < 2-fold variation in the rate of the reaction
was observed over the pH range 4–10. If the mecha-
nism of isomerization involved protonation of the
polyene chain to form a carbocation, followed by
deprotonation, then one would expect to see evi-
dence of acid–base catalysis in the pH-rate profile,
which was not observed. Furthermore, it was found
that OsD27 retains catalytic activity under anaerobic
conditions; therefore it does not require dioxygen for
activity. Hence we propose a catalytic mechanism
for OsD27 involving a 1-electron transfer from the
polyene p-system of b-carotene to a [2Fe2S] cluster,
generating a radical cation, which is able to rotate
about the C–C single bond. Electron transfer back
from the reduced [2Fe2S] cluster would then gener-
Reagent
Fig. 8. (A) pH-rate profile of CCD8 cleavage reaction of 9-cis-b-apo-
10ʹ-carotenal. (B) Percentage activity of CCD8 cleavage reaction in
the presence of different additives, relative to control cleavage
reaction.
aryl-N lengths above three (such as compounds B1,
B2, D30, D31 and D32) were not active at 10 lM con-
centration.
ate the 9-cis-b-carotene product.
involving single electron transfer from
A
mechanism
[2Fe2S]
Discussion
a
cluster to form a radical anion intermediate is also
possible (Fig. 9).
Biochemical characterization of OsD27
D27 was first identified by Lin et al. in 2009 as an
iron-containing enzyme of unknown function [13] but
was later identified as an isomerase enzyme responsi-
ble for the isomerization of all-trans-b-carotene to 9-
cis-b-carotene [12]. Our studies have confirmed this
isomerase activity, which gives only a small change in
the ratio of all-trans : 9-cis isomers from all-trans-b-
carotene, but a much greater change is evident when
assayed in the reverse direction, using 9-cis-b-carotene
as substrate. Interestingly, a small amount of a sec-
ond product is formed using 9-cis-b-carotene as sub-
strate, which may perhaps be the 9-cis, 9ʹ-cis isomer
or another isomer of b-carotene. The KM for 9-cis-b-
carotene was found to be 0.26 lM, implying that D27
binds the 9-cis isomer with a high affinity, which may
OsD27 is not strongly inhibited by the collection of
hydroxamic acid metallo-oxygenase inhibitors, as
would be expected if there is no mononuclear iron cen-
tre; however, up to 40% enzyme inhibition was
observed, notably by the terpene-like hydroxamic acids
whose structure most closely resembles that of the
substrates for OsD27.
Biochemical characterization of AtCCD7
CCD7 was first identified by Schwartz et al. in 2004
and was initially believed to act on all-trans-b-caro-
tene, performing a 9,10 cleavage reaction to generate
b-apo-10-carotenal [34,35]. However, the identification
of OsD27 as a b-carotene isomerase by Alder et al. led
FEBS Journal 282 (2015) 3986–4000 ª 2015 FEBS
3993

Inhibition of isomerase D27 and dioxygenase CCD8
P. J. Harrison et al.
them to propose that CCD7 in fact cleaves the 9-cis
isomer of b-carotene [12]. Our assays of AtCCD7 in
the presence of 9-cis-b-carotene confirmed the forma-
tion of a new peak at 10.5 min (Fig. 5) whose mass
spectrum matches that expected for 9-cis-b-apo-10ʹ-
carotenal and whose kmax of 444 nm is consistent with
the value of 445 nm reported by Alder et al. [12]. An
additional small peak was also detected at 9.8 min
(kmax 457 nm) in the AtCCD7 reaction. It was found
that the peak at 9.8 min could be generated from 9-
cis-b-apo-10ʹ-carotenal by heat or UV irradiation;
therefore this appears to be an isomeric product.
Schwartz et al. originally reported that AtCCD7 can
cleave all-trans-b-carotene [35], but it was later
reported that AtCCD7 preferentially cleaves 9-cis-b-
carotene [12,36]. Hence this peak could be a small
amount of the all-trans isomer of b-apo-10ʹ-carotenal
(reported kmax 452 nm) resulting from cleavage of
some all-trans-b-carotene impurity in the reaction;
however, Bruno et al. have reported that CCD7 is
selective for the 9-cis isomer of b-carotene [37]. Sur-
prisingly, no inhibition of CCD7 was observed by any
of the hydroxamic acid compounds. We note that the
amino acid sequence of CCD7 is divergent from other
members of the CCD enzyme family; hence CCD7
may have a somewhat different active site structure
that renders it resistant to inhibition.
catalyses a double oxygenation reaction on the 9-cis-
b-apo-10ʹ-carotenal substrate, forming an enol ether
lactone; hence the catalytic mechanism of CCD8 is
likely to be different from that of other CCDs. A
continuous UV-visible assay was developed for
AtCCD8, which was used to study the kinetic and
catalytic properties of this enzyme. Pre-steady-state
kinetic analysis of CCD8 using UV-visible stopped
flow experiments gave data matching
a two-step
kinetic mechanism, with rate constants k1 = 0.43 sꢀ1
,
k2 = 0.06 sꢀ1. These data imply two kinetic steps in
the CCD8 catalytic cycle, which may correspond to
the two consecutive oxygenation steps but could
potentially correspond to physical steps such as sub-
strate binding.
In order to probe further the catalytic mechanism
of AtCCD8, a pH-rate profile was performed, which
revealed two inflexions at pH 6.0 and 8.0, implying
the presence of a catalytic base of pKa 6.0 and a
catalytic acid of pKa 8.0. Catalytic activity was com-
pletely lost upon treatment with cysteine-specific
reagents iodoacetamide and N-ethyl maleimide,
implying the presence of an active site cysteine resi-
due [32,33]. We suggest a possible catalytic mecha-
nism in Fig. 10 that may rationalize these data. The
mechanism involves two Criegee rearrangements,
known to occur in the catalytic mechanisms of the
non-heme-iron dependent catechol dioxygenases [18].
Such rearrangements are known to require acid
catalysis [38], which is consistent with the observed
pH-rate profile for CCD8, involving an acidic group
of pKa 8.0. In both Criegee rearrangements the more
electron-rich substituent would undergo a 1,2-shift
Biochemical characterization of AtCCD8
The third enzyme on the strigolactone biosynthesis
pathway, CCD8, is responsible for the biosynthesis of
the intermediate carlactone [12]. Interestingly, CCD8
Fig. 9. Proposed catalytic mechanism for carotenoid cis–trans isomerization catalysed by OsD27.
3994
FEBS Journal 282 (2015) 3986–4000 ª 2015 FEBS

P. J. Harrison et al.
Inhibition of isomerase D27 and dioxygenase CCD8
onto the electron-deficient hydroperoxide oxygen,
and it is known that electron-rich substituents
increase the rate of migration in 1,2-rearrangement
reactions [39]. The first oxidative cleavage step
releases the 8-carbon by-product of the CCD8 reac-
tion (for which a reduced dialcohol derivative was
observed by electrospray mass spectrometry) and
generates a covalent intermediate.
branching phenotype caused by these compounds
in planta is due to inhibition of CCD8, leading to
disruption of the strigolactone biosynthetic pathway.
Some inhibition of AtCCD8 was also observed with
abamine, a tertiary amine known to inhibit NCED
[19]. We have previously suggested that abamine,
when protonated at neutral pH, may mimic a ter-
tiary carbocation in the NCED catalytic mechanism
[4,22]. The proposed catalytic mechanism for CCD8
also contains carbocation intermediates (Fig. 10),
which might explain the observed enzyme inhibition.
The potent inhibition of AtCCD8 by compound
D15 is noteworthy, since this compound has also been
found to inhibit lateral root branching in A. thaliana
[24]. Since a deficiency of strigolactones resulting from
null mutations in the max3 and max4 mutants of Ara-
bidopsis was reported either to cause formation of
more lateral roots [40] or to have no effect [24], it was
concluded that D15 was not acting via a block in
strigolactone biosynthesis [24], so D15 may have other
targets in addition to CCD8.
The second half of the catalytic cycle involves
homolytic cleavage of the carbon–X bond, requiring
an active site X group able to form a stable radical
species, for which a cysteine residue would be ideally
suited. The identification of an active site cysteine
residue in CCD8 by cysteine-directed reagents is
therefore consistent with such a mechanism. The
final step of the mechanism would involve oxidation
of a lactol intermediate and regeneration of the
active site nucleophile X and iron(II) centre. An
alternative mechanism for CCD8 has been proposed
by Alder et al. [12], involving two cycloaddition
reactions of dioxygen with an isomerized substrate,
forming endoperoxide intermediates.
As with other plant hormones, strigolactones have
diverse biological functions, from regulating plant
architecture to stimulating the germination of
parasitic plant seeds, to enhancing growth and
branching in arbuscular mycorrhizal fungi [41], and
more recent work has shown wider roles of strigo-
lactones in regulation of cell division and meristem
dormancy [42]. Application of CCD8 inhibitors to
diverse plant species and organs, particularly where
genetic studies are not feasible, may reveal other
biological functions. Inhibitors of strigolactone
biosynthesis may also have applications as agro-
chemical agents affecting shoot branching and plant
development, analogous to the use of gibberellin
biosynthesis inhibitors to affect plant structure and
yield in cotton [43].
CCD8 was found to be inhibited in a time-depen-
dent fashion by a number of hydroxamic acid inhibi-
tors in the D and F series at 100 lM concentration.
Inhibition of AtCCD8 was only observed with a
smaller group of compounds at 10 lM concentration.
Most effective inhibition was observed with com-
pounds D13H (82%), D15 (71%), D12 (56%) and
D9 (54%), which each contain
a 2-carbon unit
between the aryl group and the hydroxamic acid
functional group. We have previously hypothesized
that selectivity for hydroxamic acid CCD enzyme
inhibitors could be designed by varying the length of
spacer between an aromatic ring and a hydroxamic
acid metal-chelating group [22]. The observation that
most effective inhibition of CCD8 is observed with a
C2 spacer, compared with optimum inhibition of
CCD1 by compounds containing a C1 or no spacer
[22], matches this hypothesis, since the initial oxida-
tive cleavage by CCD8 is at the 13,14 position, com-
pared with 9,10 (9ʹ,10ʹ) cleavage by CCD1.
Compounds D2–D7 have previously been shown to
cause a shoot branching phenotype in Arabidopsis at
100 lM concentration (compounds D8–D32 were not
tested in the previous study), with compound D6
showing highest activity in planta [22]. Of these com-
pounds, inhibition of CCD8 in vitro was observed at
10 lM concentration in this study by compounds D6
(53%), D4 (17%) and D5 (9%). Since these com-
pounds show no inhibition of OsD27 or AtCCD7,
the inhibition data strongly imply that the shoot
Experimental procedures
Inhibitor design and synthesis
Hydroxamic inhibitors and abamine were designed and
synthesized as previously published [22–24]. New hydrox-
amic acids B2, D20, D21, D30, D31 and D32 were pre-
pared using the same synthetic strategy as previously
published [22], involving alkylation of N-Boc-O-benzylhy-
droxylamine with an appropriate alkyl halide and NaH;
Boc deprotection using trifluoroacetic acid; DCC coupling
with an appropriate carboxylic acid; and debenzylation
using H2/Pd. Synthetic procedures and analytical data can
be found in the Supporting information.
FEBS Journal 282 (2015) 3986–4000 ª 2015 FEBS
3995

Inhibition of isomerase D27 and dioxygenase CCD8
P. J. Harrison et al.
Fig. 10. Proposed catalytic mechanism for
CCD8, involving an active site nucleophile
X, probably a cysteine residue, and an
acidic residue A of pKa 8.0.
Preparation of 9-cis-b-carotene
Retention time of all-trans-b-carotene 23.9 min. kMAX 452,
476 nm.
All-trans-b-carotene (99%, Sigma-Aldrich, Gillingham,
UK) in hexane (1% w/v, 20 mL) was added to iodine in
hexane (1% w/v, 10 mL) and incubated in the presence of
UV light for 30 min at room temperature. Isomers were
separated by reverse phase HPLC using a ThermoScientific
Acclaim C30 column (5 lm, 4.6 9 250 mm) maintained at
4 °C at a flow rate of 1 mLꢁminꢀ1. The column was eluted
with 75 : 25 methanol : acetonitrile, shifting to 25 : 35 : 50
acetonitrile : methanol : tert-butyl methyl ether (TBME)
over 20 min. The gradient was held for a further 5 min
before returning to 75 : 25 methanol : acetonitrile for a
final 5 min. Retention time of 9-cis-b-carotene 2ꢀ41.5 min.
Gene cloning and expression
OsD27 cDNA was purchased from Genscript and sub-
cloned into the pGEX-4T-1 plasmid. The pGEX-4T-1-
OsD27 plasmid was used to transform E. coli containing
the pBB541 plasmid (Addgene) containing the GroES and
GroEL chaperones [44]. The pGEX-4T-1-AtCCD7 plasmid
(lacking the first 93 bp of AtCCD7, corresponding to a
putative chloroplast transit peptide) was used to transform
E. coli containing the pBB541 plasmid. The pET-151-
AtCCD8 plasmid (lacking the first 168 bp of AtCCD8, cor-
kMAX 448, 474 nm. e at 475 nm, 6.55 9 104
M
ꢁcmꢀ1
.
3996
FEBS Journal 282 (2015) 3986–4000 ª 2015 FEBS

P. J. Harrison et al.
Inhibition of isomerase D27 and dioxygenase CCD8
responding to a putative chloroplast transit peptide) was
used to transform E. coli BL21.
mined by integration compared to control assays. pH-rate
profile experiments were performed as described above,
except that the assay buffer was replaced by 100 mM
sodium acetate buffer (pH 4.1–5.6), 100 mM PBS buffer
(pH 6.7) or 100 mM Tris/HCl (pH 7.7–10.3). For oxygen
dependence assays, assays were set up as described above
in a two-necked glass apparatus, which was thoroughly
degassed and flushed three times with nitrogen gas, prior to
incubation of enzyme and substrate under an atmosphere
of nitrogen. UV assays were performed in a total volume
of 50 lL in an Eppendorf cuvette containing approximately
50 lg of protein in 100 mM bis-tris buffer pH 6.4 plus 10%
glycerol (v/v) and 0.1% Triton X-100 (v/v). Hydrogen per-
oxide, 2-mercaptoethanol, sodium dithionite, sodium
dithionite/pyridine, lead acetate and silver acetate were all
used at a final concentration of 100 lM.
E. coli (500 mL) containing the desired plasmids was
grown to A600
= 0.6 AU and induced with 1 mM IPTG.
Cells were harvested at 4220 g at 4 °C for 20 min. The cell
pellet was resuspended in PBS buffer (Sigma-Aldrich) pH
7.3 plus 0.05% Triton X-100 (Sigma-Aldrich), 1 mM PMSF
(Sigma-Aldrich) and 5 mM DTT (Sigma-Aldrich). Cells
were lysed by cell disruption at 20.1 kpsi before centrifuga-
tion at 20 300 g at 4 °C for 20 min. Cell free extract solu-
tion containing overexpressed OsD27 was added to a 2 mL
GST-Gravitrap column (GE Life Sciences) for OsD27 and
AtCCD7 or a 1 mL HisTrap-Gravitrap column (GE Life
Sciences) for AtCCD8 prepared according to the manufac-
turer’s instructions. Cell free extract was passed through
the column a total of three times. For GST purification the
column was washed with PBS buffer pH 7.3 (25 mL) and
eluted with PBS buffer pH 7.3 (10 mL) plus 20 mM reduced
glutathione. For N-His6 purification the column was
washed with 50 mM sodium hydrogen phosphate, 300 mM
sodium chloride, 20 mM imidazole, pH 8.0 (25 mL), before
elution into 50 mM sodium hydrogen phosphate, 300 mM
sodium chloride, 250 mM imidazole, pH 8.0 (10 mL). Pro-
tein was then concentrated using a 10 kDa centrifugal filter
unit before buffer exchange using a PD10 gel filtration col-
umn (GE Life Sciences) used according to the manufac-
turer’s instructions. Protein was eluted into 100 mM bis-tris
buffer pH 6.7 (3.5 mL) plus 10% glycerol (v/v) and concen-
trated further in a 10 kDa centrifugal filter. Protein concen-
tration was determined using the Bio-Rad protein assay
reagent, using bovine serum albumin as standard.
Protocol for in vivo D27 assay
pGEX-4T-1-OsD27 was used to transform competent pAC-
BETA BL21 E. coli. Luria-Bertani medium (10 mL) contain-
ing appropriate antibiotics 35 lgꢁmLꢀ1 chloramphenicol and
100 lgꢁmLꢀ1 ampicillin was inoculated with a single colony
of pGEX-4T-1-OsD27 pAC-BETA E. coli BL21 and grown
overnight at 37 °C, 180 r.p.m. LB medium (200 mL)
containing appropriate antibiotics was inoculated with the
overnight culture and grown at 37 °C, 180 r.p.m., until A at
600 nm = 0.6 AU. IPTG was added to a final concentration
of 1 mM. Cultures were incubated overnight at 20 °C,
180 r.p.m., harvested at 4220 g at 4 °C for 15 min and then
resuspended in water (5 mL). Lysozyme was added to give a
final concentration of 25 lgꢁmLꢀ1. Following incubation at
room temperature for 10 min, cells were lysed by sonication
at 50 Hz. n-Butanol (1 mL) was added and the organic phase
was collected following centrifugation at 4220 g at 4 °C for
10 min. The butanol was added to a 100 mg 1 mL Agilnet
Bond Elut C18 solid phase extraction column and cen-
trifuged at 2000 g at 4 °C for 2 min. The column was washed
with 10 9 1 mL water before the b-carotene was eluted into
TBME (1 mL). TBME was removed under reduced pressure
and the residue was resuspended into 200 lL methanol.
Samples were analysed via HPLC using a ThermoScientific
Acclaim C30 reversed phase column using the same method
as described for the purification of 9-cis-b-carotene.
Biochemical analysis and assays of OsD27
In vitro assays were based on methods previously published
[12] and were conducted in a final volume of 200 lL.
OsD27 enzyme (10 lL, 40 lg) was incubated with 100 mM
MOPS, 0.2 (v/v) Triton X-100, 1 mgꢁmLꢀ1 catalase, pH
6.4, with 1 lM FeSO4 and 1 lM 2-mercaptoethanol for
10 min. 9-cis-b-carotene in ethanol was added to a final
concentration of 20 lM and assays were incubated in the
dark for 20 min at 25 °C with shaking at 180 r.p.m. Assays
were stopped with the addition of 400 lL 4 : 1 : 4 ace-
tone : petroleum ether : diethyl ether and the resulting
solution was centrifuged at 4220 g for 5 min. The organic
phase was evaporated under reduced pressure. The residue
was resuspended in methanol (200 lL) and analysed by
HPLC using the method detailed for the purification of
9-cis-b-carotene.
Assays of AtCCD7
In vitro assays of AtCCD7 were performed in a total volume
of 200 lL based on previously published methods [12].
AtCCD7 (10 lL, 30 lg) was added to 100 mM HEPES,
1 mM tris-(2-carboxyethyl)phosphine (TCEP) (Sigma-
Aldrich), 1 mgꢁmLꢀ1 catalase, 0.05% Triton X-100, pH 7.8,
with 1 lM FeSO4 and 1 lM sodium ascorbate and incubated
at room temperature for 10 min. 9-cis-b-carotene was added
to a final concentration of 20 lM and assays were incubated
For inhibition assays inhibitors were dissolved in either
DMSO or ethanol (except silver acetate which was dis-
solved in water) and a total of 2 lL was added to assays to
give a final concentration of 100 lM. Inhibitors were pre-in-
cubated with enzyme for 10 min prior to the addition of
substrate. Inhibition was calculated from peak areas deter-
FEBS Journal 282 (2015) 3986–4000 ª 2015 FEBS
3997

Inhibition of isomerase D27 and dioxygenase CCD8
P. J. Harrison et al.
at 25 °C in the dark for 20 min with shaking at 180 r.p.m.
Assays were stopped with the addition of 200 lL 4 M NaCl
and 200 lL ethyl acetate. The resulting solution was cen-
trifuged at 4220 g for 5 min and the organic phase was
desalted with 100 mg 1 mL C18 solid phase extraction car-
tridges (Agilent Bond Elut), washing with 10 9 1 mL water
and eluting into 400 lL TBME. Organic solvent was
removed under reduced pressure and resuspended in metha-
nol. Analysis was performed via HPLC using the method
described above for the purification of 9-cis-b-carotene (re-
tention time 10.5 min; kmax 444 nm; high resolution mass
spectrometry 377.2839, calculated 377.2840, for C27H37O+
(MH+)). Inhibition assays were performed at 100 lM inhibi-
tor concentration, as described for OsD27.
(Syngenta, Jealott’s Hill, Bracknell, UK) for helpful
advice and practical assistance, and Dr Jaimie van Nor-
man and Dr Philip Benfey (Duke University) for help-
ful discussions. This work was supported by a BBSRC/
Syngenta CASE funded PhD studentship to PJH.
Author contributions
PJH, SN, FD and SAS carried out the research, AJT
and TDHB supervised the research, and PJH, TDHB
and AJT wrote the paper.
References
1 Walter MH & Strack D (2011) Carotenoids and their
cleavage products: biosynthesis and functions. Nat Prod
Rep 28, 663–692.
2 Harrison PJ & Bugg TDH (2014) Enzymology of the
carotenoid cleavage dioxygenases: reaction mechanisms,
inhibition and biochemical roles. Arch Biochem Biophys
544, 105–111.
3 Schwartz SH, Tan BC, Gage DA, Zeevaart JAD &
McCarty DR (1997) Specific oxidative cleavage of
carotenoids by VP14 of maize. Science 276, 1872–1874.
4 Taylor IB, Sonneveld T, Bugg TDH & Thompson AJ
(2005) Regulation and manipulation of the biosynthesis
of abscisic acid, including the supply of xanthophyll
precursors. J Plant Growth Regul 24, 253–273.
5 Iuchi S, Kobayashi M, Taji T, Naramoto M, Seki M,
Kato T, Tabata S, Kakubari Y, Yamaguchi-Shinozaki
K & Shinozaki K (2001) Regulation of drought
tolerance by gene manipulation of 9-cis-
Biochemical analysis and assays of AtCCD8
The substrate 9-cis-b-apo-10ʹ-carotenal was prepared from
assays of AtCCD7 with 9-cis-b-carotene as described
above. Collected 9-cis-b-apo-10ʹ-carotenal was concentrated
under reduced pressure and resuspended in ethanol. In vitro
assays were performed in a total volume of 200 lL using a
96-well microtitre plate reader, recording at 430 nm.
AtCCD8 (10 lL, 40 lg) was added to 100 mM HEPES,
1 mM TCEP, pH 7.8, containing 10 lM 9-cis-b-apo-10ʹ-
carotenal, 1 lM FeSO4 and 1 lM sodium ascorbate. Inhibi-
tion assays were performed at 100 lM and 10 lM inhibitor
concentration, as described for OsD27.
Stopped flow experiments were performed at the Univer-
sity of Manchester Institute of Biotechnology using an
Applied Photophysics SX.18MV stopped flow spectrometer
with a 05169 pbp monochromator. 9-cis-b-apo-100-carote-
nal substrate was prepared as described, with the exception
that prior to the removal of 4 : 1 : 4 acetone: petroleum
ether: diethyl ether under vacuum, 0.04% (v/v) Triton X-
100 was added and the residue was resuspended into water.
Reactions were performed in a 1 : 1 molar ratio of 9-cis-b-
apo-10ʹ-carotenal : AtCCD8 at a concentration of 12 lM
substrate and 12 lM AtCCD8 in 100 mM HEPES buffer
pH 7.8 containing 1 mM TCEP. For determination of the
kinetic parameters k1 and k2, analysis was performed using
Applied Photophysics PRO-DATA SX software and the
enzyme was diluted relative to the substrate.
epoxycarotenoid dioxygenase, a key enzyme in abscisic
acid biosynthesis in Arabidopsis. Plant J 27, 325–333.
6 Nambara E & Marion-Poll A (2005) Abscisic acid
biosynthesis and catabolism. Annu Rev Plant Biol 56,
165–185.
7 Yan W, Jang G-F, Haeseleer F, Esumi N, Chang J,
Kerrigan M, Campochiaro M, Campochiaro P,
Palczewski K & Zack DJ (2001) Cloning and
characterization of a human b, b-carotene-15, 150-
dioxygenase that is highly expressed in the retinal
pigment epithelium. Genomics 72, 193–202.
8 Redmond TM, Gentleman S, Duncan T, Yu S, Wiggert
B, Gnatt E & Cunningham FX (2001) Identification,
expression, and substrate specificity of a mammalian b-
carotene 15,150-dioxygenase. J Biol Chem 276, 6560–6565.
9 Akiyama K, Matsuazaki K-I & Hayashi H (2005) Plant
sesquiterpenes induce hyphal branching in arbuscular
mycorrhizal fungi. Nature 435, 824–827.
10 Gomez-Roldan V, Fermas S, Brewer PB, Puech-Pages
V, Dun EA, Pillot J-P, Letisse F, Matusova R, Danoun
S, Portais S-C et al. (2008) Strigolactone inhibition of
shoot branching. Nature 455, 189.
pH-rate profile experiments were performed as described
for OsD27. Assays with group-specific reagents were per-
formed by pre-incubation of AtCCD8 with group-specific
reagents at a final concentration of 1 mM for 10 min before
addition of substrate, then monitoring at 430 nm as
described above.
Acknowledgements
We thank Joshua Cartwright, Samuel Lowe and
Mohsina Kahn (University of Warwick) for additional
synthetic work. We thank Dr David Brocklehurst
3998
FEBS Journal 282 (2015) 3986–4000 ª 2015 FEBS

P. J. Harrison et al.
Inhibition of isomerase D27 and dioxygenase CCD8
11 Umehara M, Hanada A, Yoshidam S, Akiyama K,
Arite T, Takeda-Kamiya N, Magome H, Kamiya Y,
Shirasu K, Yoneyama K et al. (2008) Inhibition of
shoot branching by new terpenoids plant hormones.
Nature 455, 195–200.
12 Alder A, Jamil M, Marzorati M, Bruno M, Vermathen
M, Bigler P, Ghisla S, Bouwmeester H, Beyer P & Al-
Babili S (2012) The path from b-carotene to carlactone,
a strigolactone-like plant hormone. Science 335, 1348–
1351.
22 Sergeant MJ, Li JJ, Fox C, Brookbank N, Rea D,
Bugg TDH & Thompson AJ (2009) Selective inhibition
of carotenoid cleavage dioxygenases: phenotypic effects
on shoot branching. J Biol Chem 284, 5257–5264.
23 Sergeant MJ, Harrison PJ, Jenkins R, Moran G, Bugg
TDH & Thompson AJ (2013) Phytotoxic effects of
selected N-benzyl-benzoylhydroxamic acid metallo-
oxygenase inhibitors: investigation into mechanism of
action. New J Chem 37, 3461–3465.
24 Van Norman J, Zhang J, Cazzonelli C, Pogson B,
Harrison PJ, Bugg TDH, Chan K-X, Thompson AJ &
Benfrey PN (2014) Periodic root branching in
Arabidopsis requires synthesis of an uncharacterized
carotenoid derivative. Proc Natl Acad Sci USA 111,
E1300–E1309.
13 Lin H, Wang R, Qian Q, Yan M, Meng X, Fun Z,
Yan C, Jiang B, Su Z, Li J et al. (2009) DWARF27, an
iron-containing protein required for the biosynthesis of
strigolactones, regulates rice tiller bud outgrowth. Plant
Cell 21, 1512–1525.
14 Abe S, Sado A, Tanaka K, Kisugi T, Asami K, Ota S,
Kim HI, Yoneyama K, Xie X, Ohnishi T et al. (2014)
Carlactone is converted to carlactonoic acid by MAX1
in Arabidopsis and its methyl ester can directly interact
with AtD14 in vitro. Proc Natl Acad Sci USA 111,
18084–18089.
15 Zhang Y, van Dijk AD, Scaffidi A, Flematti GR,
Hofmann M, Charnikhova T, Verstappen F, Hepworth
J, van der Krol S, Leyser O et al. (2014) Rice
cytochrome P450 MAX1 homologs catalyze distinct
steps in strigolactone biosynthesis. Nat Chem Biol 10,
1028–1033.
16 Kloer DP, Ruch S, Al-Babili S, Beyer P & Schulz GE
(2005) The structure of a retinal-forming carotenoid
oxygenase. Science 308, 267–269.
17 Messing S, Gabelli S, Echeverria I, Vogel JT, Guan J-
C, Tan B-C, Klee HJ, McCarty CR & Amzel L-M
(2010) Structural insights into maize Viviparous14, a
key enzyme in the biosynthesis of the phytohormone
abscisic acid. Plant Cell 22, 2970–2980.
25 Sainsbury PD, Mineyeva Y, Mycroft Z & Bugg TDH
(2015) Chemical intervention in bacterial lignin
degradation pathways: development of selective
inhibitors for intradiol and extradiol catechol
dioxygenases. Bioorg Chem 60, 102–109.
26 Tsukida K, Saiki K, Takii T & Koyama Y (1982)
Separation and determination of cis/trans-b-carotenes
by high-performance liquid chromatography.
J Chromatogr A 3, 359–364.
27 Emenhiser C, Sander LC & Schwartz SJ (1995)
Capability of a polymeric C30 stationary phase to
resolve cis-trans carotenoid isomers in reversed-phase
liquid chromatography. J Chromatogr A 707, 205–216.
28 Cunningham FX, Pogson B, Sun ZR, McDonald KA,
DellaPenna D & Gantt E (1996) Functional analysis of
the beta and epsilon lycopene cyclase enzymes of
Arabidopsis reveals a mechanism for control of cyclic
carotenoid formation. Plant Cell 8, 1613–1626.
29 Fee JA, Findling KL, Yoshida T, Hille R, Tarr GE,
Hearshen DO, Dunham WR, Day EP, Kent TA &
€
Munck E (1983) Purification and characterization of
18 Bugg TDH (2003) Dioxygenase enzymes: catalytic
mechanisms and chemical models. Tetrahedron 59,
7075–7101.
19 Han SY, Kitahata N, Sekimata K, Saito T, Kobayashi M,
Nakashima K, Yamaguchi-Shinozaki K, Yoshida S &
Asami T (2004) A novel inhibitor of 9-cis-epoxycarotenoid
dioxygenase in abscisic acid biosynthesis in higher
plants. Plant Physiol 135, 1574–1582.
20 Kitahata N, Han S-Y, Noji N, Saito T, Kobayashi M,
Nakano T, Kuchitsu K, Shinozaki K, Yoshida S,
Matsumoto S et al. (2006) A 9-cis-epoxycarotenoid
dioxygenase inhibitor for use in the elucidation of
abscisic acid action mechanisms. Bioorg Med Chem 14,
5555–5561.
the Rieske iron-sulfur protein from Thermus
thermophilus. Evidence for a [2Fe-2S] cluster having
non-cysteine ligands. J Biol Chem 259, 124–133.
30 Tagawa K & Arnon DI (1968) Oxidation-reduction
potentials and stoichiometry of electron transfer in
ferredoxins. Biochem Biophys Acta 153, 602–613.
31 Xu FF & Imlay JA (2012) Silver (I), mercury (II),
cadmium (II), and zinc (II) target exposed enzymic
iron-sulfur clusters when they toxify Escherichia coli.
Appl Environ Microbiol 78, 3614–3621.
32 Reymond J-L, Jahangiri GK, Stoudt C & Lerner RA
(1993) Antibody catalysed hydrolysis of enol ethers.
J Am Chem Soc 115, 3909–3917.
21 Boyd J, Gai Y, Nelson KM, Lukiwski E, Talbot J,
Loewen MK, Owen S, Zaharia LI, Cutler AJ, Abrams
SR et al. (2009) Sesquiterpene-like inhibitors of a 9-cis-
epoxycarotenoid dioxygenase regulating abscisic acid
biosynthesis in higher plants. Bioorg Med Chem 17,
2902–2912.
33 Gadda G, Banerjee A, Dangott LJ & Fitzpatrick PF
(2000) Identification of a cysteine residue in the active
site of nitroalkane oxidase by modification with
N-ethylmaleimide. J Biol Chem 275, 31891–31895.
34 Fox BS & Walsh CT (1983) Mercuric reductase:
homology to glutathione reductase and lipoamide
FEBS Journal 282 (2015) 3986–4000 ª 2015 FEBS
3999

Inhibition of isomerase D27 and dioxygenase CCD8
P. J. Harrison et al.
dehydrogenase. Iodoacetamide alkylation and sequence
of the active site peptide. Biochemistry 22, 4082–4088.
35 Schwartz SH, Qin XQ & Loewen MC (2004) The
biochemical characterization of two carotenoid cleavage
enzymes from Arabidopsis indicates that a carotenoid-
derived compound inhibits lateral branching. J Biol
Chem 279, 46940–46945.
dihydroxyphenylpropionate 1,2-dioxygenase (MhpB).
Biochemistry 43, 13390–13396.
39 Lowry TH & Richardson KS (1987) Mechanism and
Theory in Organic Chemistry, 3rd edn. pp. 501–504.
Harper Collins, New York.
40 Koltai H (2011) Strigolactones are regulators of root
development. New Phytol 190, 545–549.
36 Alder A, Holderman I, Beyer P & Al-Babili S (2008)
Carotenoid oxygenases involved in plant branching
catalyse a highly specific conserved apocarotenoid
cleavage reaction. Biochem J 416, 289–296.
41 Xie X, Yoneyama K & Yoneyama K (2010) The
strigolactone story. Annu Rev Phytopathol 48, 93–117.
42 Mason MG (2013) Emerging trends in strigolactone
research. New Phytol 198, 975–977.
37 Bruno M, Hofmann M, Vermathen M, Alder A, Beyer
P & Al-Babili S (2014) On the substrate- and
stereospecificity of the plant carotenoid cleavage
dioxygenase 7. FEBS Lett 588, 1802–1807.
43 Ren X, Zhang L, Du M, Evers JB, van der Werf W,
Tian X & Li Z (2013) Managing mepiquat chloride and
plant density for optimal yield and quality of cotton.
Field Crops Res 149, 1–10.
38 Mendel S, Arndt A & Bugg TDH (2004) Acid-base
catalysis in the extradiol catechol dioxygenase reaction
mechanism: site-directed mutagenesis of His-115 and
His-179 in Escherichia coli 2,3-
44 de Marco A (2007) Protocol for preparing proteins with
improved solubility by co-expressing with molecular
chaperones in Escherichia coli. Nat Protoc 2, 2632–2639.
4000
FEBS Journal 282 (2015) 3986–4000 ª 2015 FEBS