Download The Crystal Structure of Arabidopsis thaliana Allene Oxide Cyclase

This article is published in The Plant Cell Online, The Plant Cell Preview Section, which publishes manuscripts accepted for publication after they
have been edited and the authors have corrected proofs, but before the final, complete issue is published online. Early posting of articles reduces
normal time to publication by several weeks.
The Crystal Structure of Arabidopsis thaliana Allene Oxide
Cyclase: Insights into the Oxylipin Cyclization Reaction
W
Eckhard Hofmann,a,1,2 Philipp Zerbe,b and Florian Schallerb,1,2
a Lehrstuhl
b Lehrstuhl
für Biophysik, Ruhr-Universität Bochum, D-44780 Bochum, Germany
für Pflanzenphysiologie, Ruhr-Universität Bochum, D-44780 Bochum, Germany
We describe the crystallization and structure elucidation of Arabidopsis thaliana allene oxide cyclase 2 (AOC2), a key enzyme
in the biosynthesis of jasmonates. In a coupled reaction with allene oxide synthase, AOC2 releases the first cyclic and biologically active metabolite, 12-oxo-phytodienoic acid (OPDA). AOC2 (AT3G25770) folds into an eight-stranded antiparallel
b-barrel with a C-terminal partial helical extension. The protein forms a hydrophobic binding cavity with two distinct polar
patches. AOC2 is trimeric in crystals, in vitro and in planta. Based on the observed folding pattern, we assigned AOC2 as a low
molecular weight member of the lipocalin family with enzymatic activity in plants. We determined the binding position of the
competitive inhibitor vernolic acid (a substrate analog) in the binding pocket. Based on models for bound substrate 12,13epoxy-9,11,15-octadecatrienoic acid and product OPDA, we propose a reaction scheme that explains the influence of the C15
double bond on reactivity. Reaction is promoted by anchimeric assistance through a conserved Glu residue. The transition
state with a pentadienyl carbocation and an oxyanion is stabilized by a strongly bound water molecule and favorable p–p
interactions with aromatic residues in the cavity. Stereoselectivity results from steric restrictions to the necessary substrate
isomerizations imposed by the protein.
INTRODUCTION
Since the initial discovery of methyl jasmonate (MeJA) as a
secondary metabolite in essential oils of jasmine in 1962 (Demole
et al., 1962), JAs have become accepted as a new class of plant
hormone. In the early 1980s, their widespread occurrence
throughout the plant kingdom (Meyer et al., 1984) and their
growth-inhibitory (Dathe et al., 1981) and senescence-promoting
activities (Ueda and Kato, 1980) were established. It has become
increasingly clear, however, that biological activity is not limited
to JA but extends to, and even differs among, its many metabolites and conjugates as well as its cyclopentenone precursors
(Kramell et al., 1997; Stintzi et al., 2001). Because of the different
biological function of each member of the group of JAs, the
enzymes of JA biosynthesis and metabolism may thus have a
regulatory function in controlling the activity and relative levels of
different signaling molecules.
Research in recent years generally confirmed the Vick and
Zimmerman pathway of JA biosynthesis (the octadecanoid pathway) and made considerable progress with respect to the biochemistry of the enzymes involved as well as the molecular
organization and regulation of the pathway.
1 These
authors contributed equally to this work.
whom correspondence should be addressed. E-mail eckhard.
[email protected] or [email protected]; fax 49-234-32-14748
or 49-234-32-14187.
The authors responsible for distribution of materials integral to the
findings presented in this article in accordance with the policy described
in the Instructions for Authors (www.plantcell.org) are: Eckhard
Hofmann ([email protected]) and Florian Schaller (florian.
[email protected]).
W
Online version contains Web-only data.
www.plantcell.org/cgi/doi/10.1105/tpc.106.043984
2 To
The early elucidation of the JA biosynthetic pathway and the
demonstration of the growth-inhibiting and senescence-promoting
activity were followed by the discovery that JAs are involved in
plant defense reactions, which greatly stimulated the interest in
these compounds as plant signaling molecules. A role in plant
defense was first shown by Farmer and Ryan, who demonstrated
the induction of proteinase inhibitors by MeJA and JA as part of
the defense response against herbivorous insects (Farmer and
Ryan, 1990; Farmer et al., 1991). JAs were then shown to be
active inducers of antimicrobial phytoalexins by Gundlach et al.
(1992), and subsequent work clearly established their defense
gene–inducing activity for induced resistance against insect predators and pathogens (for a review, see Reymond and Farmer,
1998; Weiler et al., 1998; Blée, 2002). Such a role was unequivocally confirmed by the analysis of mutants compromised in
either the synthesis or the perception of JA signals (for a review,
see Schaller et al., 2005).
The pathway of JA biosynthesis is shown in Figure 1. Biosynthesis is believed to start with the oxygenation of free a-linolenic
acid (LA), which is converted to (9Z,11E,15Z,13S)-13-hydroperoxy-9,11,15-octadecatrienoic acid [13(S)-HPOT] in a reaction
catalyzed by 13-lipoxygenase. The presumed release of LA from
membrane lipids through the action of a lipase may be triggered
by local or systemic signals, such as oligogalacturonides, chitosan,
systemin (Farmer and Ryan, 1992; Mueller et al., 1993; NarváezVásquez et al., 1999), or wounding (Conconi et al., 1996;
Narváez-Vásquez et al., 1999). The 13(S)-HPOT serves as a
substrate for several enzymes (Figure 1), such as divinylether
synthase, lipoxygenase, hydroperoxide reductase, epoxyalcoholsynthase, peroxygenase, hydroperoxide lyase, or allene oxide synthase (AOS) (Blée and Joyard, 1996; Vick and
Zimmerman, 1981; 1987). AOS converts 13(S)-HPOT to the
The Plant Cell Preview, www.aspb.org ª 2006 American Society of Plant Biologists
1 of 17
2 of 17
The Plant Cell
Figure 1. The Metabolic Fate of Fatty Acid Hydroperoxides.
Fatty acid hydroperoxides [only the 13(S)-hydroperoxy derivative of
linolenic acid is shown] are channeled into seven distinct pathways for
oxylipin biosynthesis. The committed steps are catalyzed by AOS for JA
biosynthesis, divinylether synthase for the formation of divinyl ethers,
lipoxygenase for the formation of keto polyunsaturated fatty acids
(PUFAs), hydroperoxide reductase in the formation of hydroxy PUFAs,
epoxyalcohol synthase and peroxygenase resulting in epoxy hydroxy
PUFAs, and hydroperoxide lyase for the production of leaf aldehydes
(green leaf volatiles). In the AOS branch, the Vick and Zimmerman
pathway for JA biosynthesis is shown. 13-LOX, 13-lipoxygenase; OPR3,
12-oxophytodienoate reductase 3. Figure adapted from Feussner and
Wasternack (2002) and Wasternack and Hause (2002).
unstable epoxide 12,13(S)-epoxy-9(Z),11,15(Z)-octadecatrienoic
acid (12,13-EOT), which is cyclized by allene oxide cyclase
(AOC) to the first cyclic and biologically active compound of the
pathway, 12-oxo-phytodienoic acid (OPDA). Reduction of the
10,11-double bond by an NADPH-dependent OPDA reductase
then yields 3-oxo-2(29(Z)-pentenyl)-cyclopentane-1-octanoic
acid (OPC-8:0), which is believed to undergo three cycles of
b-oxidation to yield the end product of the pathway [i.e., JA with
(3R,7S)-configuration; (þ)-7-iso-JA] (Vick and Zimmerman,
1984). The biosynthesis of JA appears to involve two different
compartments: the conversion of LA to OPDA is localized in the
chloroplasts (Vick and Zimmerman, 1987; Song et al., 1993;
Laudert et al., 1997), while the reduction of OPDA to OPC-8:0
(Schaller and Weiler, 1997a, 1997b; Schaller et al., 2000;
Strassner et al., 2002) and the three steps of b-oxidation (i.e.,
conversion of OPC-8:0 to JA) occur in peroxisomes (Gerhardt,
1983; Vick and Zimmerman, 1984; Li et al., 2005). Like JA,
OPDA may accumulate to substantial amounts in plant tissue
(Stelmach et al., 1998). By contrast, OPC-8:0 occurs only in trace
amounts.
Vick and Zimmerman (1981) postulated the existence of a
hydroperoxide cyclase involved in the conversion of 13(S)-HPOT
into OPDA. The allene oxide 12,13-EOT was shown to serve as
the immediate precursor of OPDA in plants (Baertschi et al.,
1988; Crombie and Morgan, 1988; Hamberg and Hughes, 1988).
Hamberg and Fahlstadius (1990) showed that this cyclization
reaction is catalyzed by AOC, a soluble enzyme in maize (Zea
mays). AOC was subsequently purified as an apparent dimer of
47 kD from maize kernels (Ziegler et al., 1997) and characterized
with respect to its substrate specificity: the enzyme accepted
12,13(S)-epoxylinolenic acid but not 12,13(S)-epoxylinoleic acid
as a substrate (Ziegler et al., 1999). This is in contrast with AOS,
which produces both allene oxides from the respective 13(S)hydroperoxy fatty acids (18:3 and 18:2, respectively). It thus
appears that AOC confers additional specificity to the octadecanoid biosynthetic pathway.
An interesting aspect of the AOC reaction is the apparent
competition between the spontaneous decomposition of its
substrate, the unstable allene oxide, to form a- and g-ketols and
racemic cis-OPDA, and the enzyme-catalyzed formation of optically pure (9S,13S)-OPDA [i.e., cis(þ)-OPDA], the first pathway
intermediate having the characteristic pentacyclic ring structure
of JAs (Figure 1) and biological activity. The extremely short halflife of allene oxides (half-time <30 s in water; Brash et al., 1988)
and the optical purity of natural OPDA suggest a tight coupling of
the AOS and AOC reactions, possibly in a synthase-cyclase
complex (Stenzel et al., 2003b). Coupling of the two reactions is
also observed in vitro: AOC from potato (Solanum tuberosum)
or recombinant Arabidopsis thaliana AOC2 in combination with
recombinant Arabidopsis AOS resulted in the production of
highly asymmetrical cis-OPDA consisting nearly exclusively of
the (9S,13S)-enantiomer (Laudert et al., 1997). Recently, we demonstrated that a physical interaction of AOS and AOC is not
essential for the cyclization reaction, but the yield of OPDA formation and optical purity of the product is diminished when
separating both enzymes from each other (P. Zerbe, E.W. Weiler,
and F. Schaller, unpublished data).
AOC has been cloned as a single-copy gene from tomato
(Solanum lycopersicum; Ziegler et al., 2000) and barley (Hordeum
vulgare; Maucher et al., 2004) and as a small gene family (AOC1-4)
from Arabidopsis (Stenzel et al., 2003b). Inspection of the N termini
of the cloned AOCs revealed the presence of transit peptides for
plastid targeting, and localization in chloroplasts was confirmed
immunohistochemically (Ziegler et al., 2000; Stenzel et al., 2003a,
2003b) and by transient expression of AOC1-4/green fluorescent
protein fusions (F. Schaller and P. Zerbe, unpublished data).
To gain more information about the molecular mechanisms
underlying the cyclization of oxylipins, we set out to elucidate the
structure of an AOC from Arabidopsis. We decided to focus our
Crystal Structure of AOC2
biochemical and structural studies on AOC2 because this isozyme has been shown to be highly expressed in planta and was
described as being the most active of the four Arabidopsis AOCs
(Stenzel et al., 2003b).
RESULTS
Overexpression, Purification, and Crystallization
of Arabidopsis AOC2
Because of relatively weak expression levels of published constructs (C. Wasternack, personal communication) we cloned
differently truncated versions of AOC2 in pET 21b(þ) (Novagen/
Merck His6-tag, C-terminally) and pQE30 (Qiagen His6-tag,
N-terminally) and analyzed expression levels of insoluble and
soluble AOC2. Constructs beginning at base 132 showed the
highest expression of soluble protein in both vectors and were
subsequently used.
Using SDS-PAGE analysis, we observed a strong influence of
the His6-tag location on the oligomerization state of the protein.
While AOC2 with a C-terminal His6-tag runs as a monomer with
an apparent mass of ;22 kD, the N-terminally tagged protein
shows a dominant band at a molecular mass of ;60 kD, consistent with an SDS-stable trimer formation. The identity of the
bands could be confirmed by immunoblot analyses using antibodies raised against AOC2 or the His6-tag.
A similar pattern to that of the N-terminally tagged protein has
been observed in immunoblot analyses of native AOCs from plant
extracts (P. Zerbe and F. Schaller, unpublished data). This clearly
shows that the trimeric form is present in planta as well and is not
an artifact due to addition of the N-terminal His6-tag. Addition of
a C-terminal His6-tag seems to destabilize the trimers.
Overall Structure
The structure of Arabidopsis AOC2 has been determined by
multiwavelength anomalous dispersion with selenomethionine
(SeMet)-labeled protein crystals of the space group P212121 and
was refined to a resolution of 1.5 Å (SeMet-AOC2). Additionally,
we solved the structure of the unlabeled protein in space group
P21 at a resolution of 1.8 Å (WT-AOC2). In both space groups,
AOC2 forms closely packed trimers. In space group P21, we find
two copies of this AOC2 trimer, which superimpose with a root
mean square deviation (RMSD) of 0.33 Å for all Ca atoms. Both
align well with the single trimer found in the SeMet-AOC2 structure with RMSDs of 0.40 and 0.44 Å, respectively. Data collection
and refinement statistics are presented in Table 1.
Figure 2 shows the structure of SeMet-AOC2 as a ribbon
diagram together with a topology diagram. The prominent structural feature of AOC2 is the eight-stranded antiparallel b-barrel of
;40-Å height formed by residues 17 to 147 (numbered according to the recombinantly expressed protein). Based on the offset
observed in the hydrogen bonding pattern upon closure of the
barrel, we can assign a shear number of 10, which is a measure
for the tilt of the b-strands relative to the barrel axis (Murzin,
1994). The central cross section is slightly elliptical with axes of
;14 and 18 Å (Figure 2B). The 41 C-terminal residues cover the
3 of 17
bottom and one side of the barrel with two one-turn 3/10 helices
and a short a-helix connected by random coil stretches.
The starting residue of the first b-strand (Glu-17, or Glu-82 in
the unprocessed protein) corresponds to the predicted cleavage
site of the transit peptide (ChloroP; Emanuelsson et al., 1999). In
our construct, we included five additional residues, two of which
are well defined in the density. The first 14 residues, including the
N-terminal His6-tag and some vector-specific residues are disordered in the crystal structure.
The barrel forms an elongated cavity, which is lined mostly by
aromatic and hydrophobic residues and reaches ;14 Å into the
protein (Figure 2C). Three areas of the cavity surface are noteworthy. First, the conserved Glu-23 is positioned at the very
bottom of the cavity, introducing a negative charge. Second, an
additional polar patch is formed by Ser-31, Asn-25, and Asn-53
on one side of the cavity. These three residues together with
the main chain nitrogen from Pro-32 coordinate a tightly bound
water molecule that is found in all our AOC2 structures. Third, we
found Cys-71 on the opposing wall of the cavity. These residues
are strictly conserved among all AOC sequences in the EBI
UNIRef100 database (Figure 3).
Quarternary Structure (Trimer)
In complete agreement with our biochemical data, we found
AOC2 to form trimers when crystallized (Figure 4). The barrel
axes of the monomers are tilted ;308 with respect to the trimer
axis and pack closely with their barrel walls. Trimerization buries
;2000 A2 of the barrel surface and is the only interaction in the
crystal that is identical for all copies of the molecule. Crystal
contacts cover a significantly smaller surface (800 to 900 A2 per
monomer) and are not identical for all monomers. Identical
trimerization is observed in both the WT-AOC2 and SeMetAOC2 crystals, which have different space groups and packing
interactions. Residues involved in trimerization are highly conserved in known AOCs (see Supplemental Figure 1 online). The
majority of the C-terminal part of the protein is not involved in the
interaction of the monomers (Figure 4), but the C terminus (Asn188) together with Arg-29 forms a strong salt bridge to Glu-75 of
a neighboring monomer.
Structural Comparison with Other Proteins
According to the rules underlying the SCOP database (Murzin
et al., 1995), the AOC2 structure with an eight-stranded strictly
antiparallel b-barrel with a shear number of 10 would be either
assigned to the streptavidin fold or to a new fold due to the topological differences in the orientation of the barrel. For functional analysis, it is more fruitful to discuss its relation to protein
families. AOC2 certainly belongs to the superfamily of calycins
(Flower et al., 2000), which are characterized by a central
b-barrel forming a hydrophobic cavity for binding of small lipophilic compounds. Major members of this superfamily are the
triabins, metalloproteinase inhibitors, the fatty acid binding proteins, the avidins, and the lipocalins. Of these, the lipocalins most
closely resemble the tertiary structure and fold of AOC2. Indeed,
a DALI (Holm and Sander, 1994) structural similarity search
clearly assigns AOC to the lipocalin fold, finding as the top hits
4 of 17
The Plant Cell
Table 1. Data Collection and Refinement Statisticsa
Inflection
Point
Data Set
Peak
Beamline
Resolution (Å)
Cell parameters (a, b, c; Å)
(a, b, g; 8)
Space group
Wavelength
Completeness (%)
Multiplicity
Average I/sI
Rsym (%)
Rmeas (%)b
Rmrgd-F (%)b
Phasing
FOM (RESOLVE)c
Refinement
Resolution (Å)
ESRF ID13.1
37–1.5 (1.6–1.5)
64.48, 99.95, 106.15
90 90 90
P212121
0.97935
0.97955
95.9 (94.2)
95.9 (94.2)
7.8 (7.8)
7.8 (7.8)
19.3 (5.3)
19.3 (5.3)
7.0 (38.8)
7.0 (39.1)
7.5 (41.6)
7.5 (41.9)
7.0 (26.1)
7.0 (26.1)
Rcryst (%)
Rfree (%)d
Number of atoms
Protein
Inhibitor
Glycerol
Water/ion
Number of side chains with
alternate conformations
Average B-factor protein (inhibitor)
RMSD from ideality (protein atoms)
Bonds (Å)
Angles (8)
Remote
Inhibitor
Wild Type
0.9789
95.9 (94.2)
7.8 (7.8)
18.6 (4.9)
7.4 (42.7)
7.9 (45.8)
7.6 (29.0)
In house
38–1.7 (1.8–1.7)
64.5, 99.8, 105.9
90 90 90
P212121
1.54
93.5 (89.8)
4.2 (4.2)
15.2 (4.2)
7.9 (34.5)
9.1 (39.5)
9.5 (24.2)
SLS PXII
50–1.8 (2.1–1.8)
67.12, 104.59, 86.96
90 95.0 90
P21
0.950
97.7 (93.8)
4.3 (3.4)
11.3 (4.8)
7.4 (22.0)
8.3 (25.8)
8.6 (28.2)
73–1.7 (1.74–1.7)
86–1.8 (1.85–1.8)
19.8 (29.7)
23.3 (36.7)
20.7 (43.4)
24.8 (51.2)
8179
42
556
19
4124
21
30
369
6
13.5
12.8 (48.2)
23.8
0.008
1.224
0.021
1.838
0.022
1.757
0.86
37–1.5
(1.54–1.5)
17.3 (18.8)
19.1 (21.8)
4194
30
660/2
5
a Data
in parentheses represent values in the highest resolution bin.
definition of Rmeas and Rmrgd-F, see Diederichs and Karplus (1997).
c FOM calculated after density modification and threefold averaging in RESOLVE.
dR
free calculated from 5% of data omitted from refinement.
b For
retinol binding protein (RBP; 1AQB) and nitrophorin (1NP1) each
with a Z-score of 3.2.
The lipocalin protein family (Pervaiz and Brew, 1985, 1987;
Grzyb et al., 2006) is a large group of small extracellular proteins
with great diversity at the sequence level but mostly with either
three characteristic short conserved sequence motifs (SCR I-III;
the kernel lipocalins) or else one conserved motif (SCR I; the
outlier lipocalins). However, the crystal structures are highly
conserved and comprise a single eight-stranded continuously
hydrogen-bonded antiparallel b-barrel with an integral ligand
binding site (Figure 5). The structural basis for these SCRs is a
clustering of conserved residues around a central Trp in SCR I at
the beginning of b-strand A, which stabilizes barrel closure. Several members of the lipocalin family bind to specific cell surface
receptors or form complexes with soluble macromolecules. Initially, lipocalins were described as transport proteins; however, it
has become clear that the lipocalins exhibit a multitude of functions, with roles in retinol transport, olfaction, and pheromone
transport, prostaglandin biosynthesis, invertebrate cryptic coloration, and a role in the xanthophyll cycle. In addition to the
central b-barrel, the C-terminal part of the protein is usually appended by a characteristic a-helix. One end of the barrel is
closed by the N terminus that traverses the base of the barrel
(Figure 5). At the opposite end, the b-barrel is typically open to
solvent and provides access to the cavity. These binding sites of
the lipocalins may have very different shapes (e.g., with a ligand
pocket like an extended cave with lobes, a ligand pocket deep
within the hydrophobic core, with loops encapsulating the ligand,
or a binding site with a wide, funnel-like opening to the solvent)
(for a review, see Schlehuber and Skerra, 2005).
Automated superposition of AOC2 with the two lipocalin hits
from the DALI search leads to RMSD values of 2.6 Å for 74
matched Ca for 1AQB and 3.6 Å for 100 matched Ca for 1NP1.
The superposition with RBP as an archetypal representative of
the lipocalin family is shown in Figure 5.
Stretches of superposed residues all fall within the b-strands,
therefore aligning the barrel walls well, while the loop regions
show large differences. AOC2 does not show the classical
C-terminal helix, and the barrel is not closed by the preceding
sequence but by the extended C terminus. Most striking is the
Crystal Structure of AOC2
5 of 17
Figure 2. Structure of the AOC2 Monomer.
(A) Side view of the AOC2 monomer shown in ribbon representation, color changing from blue (N terminus) to red (C terminus).
(B) Top view, rotated by 908 toward the viewer with respect to (A). The same representation is used as in (A).
(C) Molecular surface representation of the proposed binding pocket. The orientation and color scheme are similar to (A). Also shown are the residues
discussed in the text in stick representation. Water 75 is shown as a red sphere.
(D) Topology diagram in the same color coding as in (A). b-Strands are shown as arrows (with numbers indicating the first and last residues) and helices
as small cylinders.
different position of the termini with respect to the barrel orientation. In the orientation shown in Figure 5, the origin of the closed
b-sheet is on the top for the lipocalins, whereas it is on the bottom
side for AOC2. This topological difference could be the result of a
strand exchange in which either the first or last b-strand (strand A
or strand H in lipocalin nomenclature) has been replaced by C- or
N-terminal sequence stretches, respectively. Notably, the sequence DLVPFTNKLY (residues 47 to 56) from AOC2 could be
automatically aligned to SCR I in pig RPB (accession number
1AQB). While we can observe a weak sequence similarity in this
stretch, the conserved Gly and Trp (residues 22 and 24, respectively, in RBP) are missing in the AOC sequence. Interestingly, the
region identified in AOC2 falls within the second b-strand and
superimposes spatially with the location of SCR I in b-strand A of
the lipocalins (Figure 5).
Recently, the structure of a coral AOS domain has been
reported (Oldham et al., 2005). In soft corals, no AOC has been
described so far, so an intriguing possibility would be that coral
AOS catalyzes both epoxidation and cyclization reactions. Indeed, coral AOS has been found to have a catalase-type fold,
which includes an eight-stranded antiparallel barrel. However,
there has been no report that the barrel fulfils a catalytic role in
cyclization in this case, and the interior is tightly packed with
protein side chains, preventing a similar substrate binding as in
lipocalins.
Substrate Specificity
To gain more information about the catalytic mechanism, we
tested 13(S)-HPOT and 13(S)-hydroperoxy-octadecadienoic acid
(13(S)-HPOD) as substrates for Arabidopsis AOC2 and AOS. We
found that with 13(S)-HPOD as substrate, only negligible amounts
of cyclic product were formed (data not shown). Since AOS
makes use of linoleic acid and linolenic acid derivatives (Siedow,
1991; Song and Brash, 1991; Pan et al., 1995; Laudert et al., 1996),
AOC2 is apparently incapable of cyclizing the corresponding
6 of 17
The Plant Cell
Figure 3. Sequence Alignment of Known AOCs.
Representation of secondary structure elements and numbering is based on the SeMet-AOC2 structure. Asterisks mark residues modeled with
alternate conformations. The following AOC2 sequences are listed (accession numbers in parentheses): Arabidopsis, AOC2_AT (Q9LS02), AOC1_AT
(Q9LS03), AOC3_AT (Q9LS01), and AOC4_AT (Q93ZC5); Pisum sativum, AOC_PS (Q3LI84); Medicago truncatula, AOCa_MT (Q711Q9) and AOCb_MT
(Q599T8); Oryza sativa, AOC_OS (Q8L6H4); Nicotiana tabacum, AOC_NT (Q711R1); Hordeum vulgare, AOC_HV (Q711R0); Zea mays, AOC_ZM
(Q6RW09); Humulus lupulus, AOC1_HL (Q68IP7); AOC4_HL (Q68IP6); Bruguiera sexangula, MAN_BS (Q9AYT8); Lycopersicon esculentum, AOC_LE
(Q9LEG5); S. tuberosum, AOC_ST (Q8H1X5); Physcomitrella patens, AOCa _PP (Q8GS38) and AOCb_PP (Q8H0N6). Figure was produced using the
ESPript server (Gouet et al., 1999).
epoxide, which was indicated before by studies of Ziegler et al.
(1997) on AOC from maize seeds. Therefore, the double bond at
position 15 seems to be essential for the enantioselective cyclization by AOC2. This is in agreement with a proposal for the
chemical cyclization reaction, whereby the 15(Z) double bond
facilitates the opening of the oxirane ring of the allene oxide, and
the resulting C-13 carbocation is stabilized by the p electrons of
the 15(Z) double bond (Grechkin, 1994).
Substrate Binding Site and Biochemical Analysis of
Mutated AOC2 Protein
In 1990, a competitive inhibitor of the AOC of maize [(þ/)-cis12,13-epoxy-9(Z)-octadecenoic acid ¼ vernolic acid] was described (Hamberg and Fahlstadius, 1990). Vernolic acid is a
naturally occurring fatty acid enriched especially in the seed oil
(50 to 90% [w/w] of the total fatty acids) of several Asteraceae
genera (Gunstone, 1954; Badami and Patil, 1980) and Euphorbiaceae species (Kleinman et al., 1965; Spitzer et al., 1996) and is
a substrate analog of 12,13-EOT only missing the C11 double
bond. We tested vernolic acid for its inhibiting effects on Arabidopsis AOC2. In our coupled enzymatic assay, total activity was
downregulated by 50%, with some loss of stereoselectivity
(Figure 6).
To determine the location of the active site and to gain insight
into possible substrate binding modes, we soaked SeMet-AOC2
crystals with vernolic acid and solved the structure of the complex at 1.7-Å resolution (for details, see Methods and Table 1). As
expected by the structural similarity to the lipocalins, we found
vernolic acid bound in the hydrophobic barrel cavity (Figure 7A).
In the observed binding mode, the charged carboxylic head
group is located outside the cavity on the protein surface, while
the more hydrophobic part of the molecule with the oxirane is
buried deep in the barrel. The molecule is in the extended conformation, which is not directly compatible with the cyclization
reaction. The dominant molecular interactions with the protein
are shown in Figure 7B. Overall the protein contributes mostly
aromatic and hydrophobic residues to binding. The only polar/
charged contacts are with Glu-23 close to the C15 double bond
and with the tightly bound water molecule toward the oxirane.
Crystal Structure of AOC2
7 of 17
here, total activity was reduced ;50%, and the stereoselectivity
was reduced to a ratio of 1:4. As this mutant was not as highly
purified as the other proteins, the absolute activity values have to
be treated with some caution.
DISCUSSION
Comparison with Other Arabidopsis AOC Structures
Figure 4. Trimeric Form of Arabidopsis AOC2.
Ribbon representation of the AOC2 trimer. Residues 15 to 147 are shown
in gray, and residues 148 to 188 are colored red. The view is along the
trimer axis.
Based on the observed protein–inhibitor interactions, we decided to investigate the influence of single amino acid substitutions
on the activity of AOC2. Phe-85 seems to contribute to binding of
the substrate by forming a hydrophobic slide, which at the same
time restricts the conformational freedom of the substrate during
cyclization and thereby could render the reaction stereoselective. With PCR-based mutagenesis, we exchanged Phe-85 with
Leu (F85L) and Ala (F85A). Both mutants can be purified in similar
amounts as the wild-type protein and show no overall folding
defect as judged by circular dichroism (data not shown). The
F85L mutant was successfully crystallized and showed no
structural deviations from the wild-type protein. In the coupled
AOS/AOC2 activity test, the F85L mutant showed a moderately
reduced total activity and a slight loss of stereoselectivity (Figure
6). For the F85A mutant, this result was even more pronounced:
During the preparation of this manuscript, the Center for Eukaryotic Structural Genomics consortium (Madison, WI) deposited
the coordinates of ligand-free AOC2 (CESG-AOC2, entry 1Z8K)
and ligand-free AOC1 (CESG-AOC1, entry 1ZVC) to the Protein
Data Bank (G.E. Wesenberg, G.N. Phillips Jr., E. Bitto, C.A.
Bingman, and S.T.M. Allard, unpublished data). Although there is
no publication yet on these structures, it is interesting to compare
our results with the available coordinates.
CESG-AOC2 has been solved using SeMet-labeled protein
and was refined at 1.7-Å resolution in the same orthorhombic
space group as our SeMet-AOC2 structure. Both models superimpose very well with an RMSD of only 0.31 Å for 173 Ca atoms.
A stereoview of the superposition is shown in Figure 8. Closer
inspection does not reveal any significant structural differences;
specifically, the trimerization and the tightly bound water molecule
inside the cavity are also found in the CESG-AOC2 structure.
More interesting is the comparison with CESG-AOC1. The
structure has been solved at a resolution of 1.8 Å in a hexagonal
space group. In these crystals, the same mode of trimerization is
observed for the protein, with the trimer axis now corresponding
to a crystallographic symmetry. Again, we find the same tightly
bound water molecule in the proposed active site of the enzyme.
Sequence comparison to AOC2 shows a total of 10 mutations
Figure 5. Structural Superposition of AOC2 and RBP.
Stereo diagram of pig RBP (Protein Data Bank accession code 1AQB) with the bound ligand retinol superposed with AOC2. RBP is drawn as a yellow
ribbon, and the bound retinol is shown in van der Waals representation in purple and red. AOC2 is shown as a gray ribbon. To illustrate the discussed
strand exchange, N-terminal portions of both proteins are colored blue (residues preceding the first b-strand in RBP and in the first b-strand in AOC2).
C-terminal stretches of the proteins are colored maroon (including the last b-strand for RBP) and red (after the last residue in the barrel for AOC2). For
clarity, the last 29 residues of RBP have been omitted, which include the conserved lipocalin helix mentioned in the text.
8 of 17
The Plant Cell
and AOC2, might therefore represent a subtle optimization for
interactions of these proteins with different partners either in
different cellular locations or physiological states of the plant. For
a complete understanding of these regulatory processes, more
data on the expression profiles for all four enzymes in planta are
necessary. Meanwhile, the structure elucidation of AOC3, whose
crystallization has already been indicated by the CESG in the
structural genomics target database (http://targetdb.pdb.org/),
might shed some more light on the structural variations between
the isoforms.
Figure 6. Enzymatic Activity of AOC2 Mutants and AOC2 with the
Inhibitor Vernolic Acid.
For determination of enzymatic activity, 5 mg of AOS and 10 mg of AOC2
were mixed in 1 mL of 10 mM PPi buffer, pH 7.0. After addition of 100 mg
of HPOT as substrate, samples were incubated at room temperature
for 15 min. Synthesized OPDA was extracted with ethyl acetate and
analyzed via gas chromatography–mass spectrometry (GC-MS) as
described in Methods. Shown are the results of the coupled enzymatic
assay for AOS alone (1), AOS and AOC2 (2), AOS and AOC2 in presence
of 160 mM inhibitor vernolic acid (3), and the mutants F85L (4) and F85A
(5). The results are presented as the relative amount of synthesized
OPDA compared with the yield with AOC2. The gray bars show total
synthesized OPDA, whereas the black and white bars represent the yield
of (þ)- and ()-enantiomer, respectively. Error bars represent SD based
on three independent triplicate measurements (1, 2, 4, and 5) or a
duplicate measurement (3).
(depicted in stick model in Figure 8), most of which can be
classified as conservative exchanges, while in three cases a
charge reversal occurs. Introduction of a positive charge at
position 42 (Met to Arg) in the first turn is counterbalanced by the
exchange of Lys-81 to Asn, which is located on the neighboring
turn after the third b-strand. While we can observe a slight
displacement of the main chain in that region, this is likely to be a
result of different crystal packing, as Leu-41 and Met-42 are
involved in crystal contacts in the case of AOC2. The preceding
three residues (corresponding to residues 39 to 41 in AOC2) have
not been modeled in CESG-AOC1, probably due to disorder.
Another exchange involving a reversal in local charge is located
on the opposite side of the protein at residue 178 due to the
exchange of Glu to Lys.
None of the changes seem to be of potential relevance for the
ligand binding site. Indeed, we found no significant differences in
either total activity, stereo, or substrate specificity for all four
AOC isoforms (AOC1-4) in our activity assays (P. Zerbe, V.
Effenberger, and F. Schaller, unpublished data). This is in contrast with the observation of Stenzel et al. (2003b), who reported
AOC2 having the highest activity but without quantifying this.
A differential expression pattern of AOCs is clearly established.
Upon wounding and application of jasmonic acid, expression of
AOC2 is upregulated to a higher degree than that of AOC1
(Stenzel et al., 2003b), whereas during senescence, a stronger
upregulation of expression has been observed for AOC1 than for
AOC2 (He et al., 2002). The other two isoforms, AOC3 and AOC4,
seem not to be as strongly upregulated or sometimes even
downregulated. The small structural differences, which mainly
involve a change in the surface potential pattern between AOC1
Structural Relation to the Lipocalin Family
While a database search with the PROSITE lipocalin signature
motif results in 113 hits from plants, there are to our knowledge
only four examples of plant lipocalins explicitly assigned in the
literature: the small temperature-induced lipocalin TIL (Charron
et al., 2002), the two xanthophyll cycle enzymes violaxanthin
deepoxidase (VDE) and zeaxanthin epoxidase (ZEP) (Burgos
et al., 1998), and the recently identified chloroplast lipocalin CHL
(Charron et al., 2005). The last three proteins are located in the
chloroplast. While all four proteins share homology in the structurally conserved regions, especially in SCR I with the invariant
Gly and Trp present, homology in SCR II and III is much weaker,
suggesting that these plant proteins would fall in the class of
outlier lipocalins (Burgos et al., 1998). The other hits produced by
the automated search are either false positive hits or the classification as lipocalins has not yet been supported by analysis of
functional data.
As AOC2 can be assigned structurally to the lipocalin fold, this
would allow placement in one of several protein families showing
this general folding pattern or even setting AOC2 aside as a new
structural motif on its own. Considering the observed topology of
the barrel and functional similarities, we propose to consider
AOC2 as a distant member of the lipocalin family in plants. The
most relevant differences are discussed below.
The different placement of the termini in the superposition is
striking (Figure 5). The barrel organization can be described rightor left-handed in the following way: if the thumb points upwards
in the direction of the first b-strand, the bent fingers of the respective hand should indicate the sequential organization of the
successive b-strands. By this definition, all lipocalin family members known so far (to our knowledge, this extends to all structures
assigned to the lipocalin fold) form a left-handed barrel, while the
AOC2 calyx is right-handed. In both cases, possible ligand
binding sites are inside the palm and are accessible from above.
As mentioned before, N- or C-terminal strand insertion with
concomitant strand removal on the opposite end would automatically convert right- to left-handed barrels and vice versa.
Along these lines, the first b-strand in AOC2 (blue in Figure 5)
could have been formed by the insertion of the longer N-terminal
part of the classical lipocalins (also colored blue) into the barrel.
This would push b-strand H of the lipocalins (shown in maroon)
out of the continuous b-sheet. The resulting differences in the
structural organization would reduce the central role of the
tryptophane in SCR I, and a loss of similarity in the SCRs would
be obvious. Indeed, the observed weak similarity to SCR I in the
second b-strand could be indicative of such an early process.
Crystal Structure of AOC2
9 of 17
Figure 7. Structure of AOC2 with Bound Inhibitor Vernolic Acid.
(A) In the stereoview, monomer B is shown in gray ribbon representation, vernolic acid (VA) as blue stick model, and Water75 as red sphere. Residues
directly involved in interactions with either the inhibitor or Water75 are depicted in yellow stick representation and labeled. Phe-85 is colored red. Omit
difference density for vernolic acid (see Methods) at 2.4 s is drawn as red chicken wire. The view is rotated ;908 around the vertical with respect to
Figure 2A.
(B) Ligplot sketch showing the molecular interactions between the protein and the bound inhibitor vernolic acid. Contacts to the conserved Water75 are
shown in green dashed lines with distances, and hydrophobic contacts are represented by opposing red spoked arcs. The interaction of Glu-23 is
represented as a yellow arc. Vernolic acid is shown in purple.
10 of 17
The Plant Cell
Figure 8. Structures of AOC2, CESG-AOC2, and CESG-AOC1
Shown is a stereoview of the Ca trace of AOC2 (in gray), CESG-AOC2 (accession code 1Z8K; in red), and CESG-AOC1 (accession code 1ZVC; in blue).
The 10 residues of AOC2 that differ in the AOC1 sequence are depicted in yellow stick representation.
This divergence in structure from the classical fold is not
unexpected. Investigation of the exon-intron structure of lipocalins clearly places the plant lipocalins together with Dictostelium lipocalins into a gene structure–related group distinct from
lipocalins from bacteria, insects, or animals (Sanchez et al.,
2003). Since the early acquisition of the ancestor gene, it has
been duplicated and adapted to different tasks. While TIL still can
be counted as a genuine lipocalin, VDE evolved to be catalytically
active in deepoxidation and acquired additions to the central
barrel structure. With a size of ;350 amino acids, VDEs are
much larger than the typical lipocalin monomer, which consists
of 160 to 180 amino acids (Flower, 1994; Flower et al., 1995;
Burgos et al., 1998). The same is true for ZEP, for which threading
methods don’t even predict a barrel structure. Therefore, the
peculiar architecture of these proteins raised doubt as to whether
they truly belong to the lipocalin family (Ganfornina et al., 2000;
Salier, 2000).
Up to now, not enough representatives of plant lipocalin genes
have been identified to decide whether the clade of plant
lipocalins does indeed show a much higher variability as the
better-characterized lipocalins from bacteria, insects, or animals.
On the other hand, one cannot exclude the possibility that we
in fact do observe an example of convergent evolution. This is an
ongoing debate given the very low sequence homology among
lipocalins and especially for the outlier lipocalins, for which only
sequence similarity in SCR I is detectable anyway (Grzyb et al.,
2006).
physiological relevant oligomerization state of the protein. First,
we can detect trimers of recombinant AOC2 or AOC2 from plant
extracts in SDS page or immunoblots. Secondly, we observe
very large buried surface areas upon trimerization that include
three salt bridges involving the terminal carboxy group and
several strong hydrogen bonds. This is in agreement with the loss
of trimer formation for the construct with the C-terminal His-tag,
which would likely interfere with the formation of the salt bridges.
Taken together, this pattern is indicative of a stable protein–
protein interaction. Importantly, residues involved in these interactions are predominantly conserved in all known AOC sequences.
Additionally, while both wild-type AOC2 and the CESG-AOC1
from Arabidopsis crystallize in different space groups and have
therefore a modified crystal packing organization, in both cases
the same mode of trimerization is observed.
AOC from maize has been shown to have an apparent molecular mass of 47 kD in size exclusion chromatography and was
discussed as a dimer (Ziegler et al., 1997). We found a similar
behavior for AOC2 of Arabidopsis (data not shown), but based
on our additional structural and biochemical data, we attribute
this discrepancy of 25% to the calculated molecular weight of a
trimer to an atypical retardation of this protein on the size
exclusion matrix.
We therefore expect the trimeric form to be the physiological
relevant oligomerization state for AOCs of Arabidopsis and maize.
Nevertheless, this SDS-stable trimer formation seems not to be
absolutely required, since both the C- and N-terminally tagged
proteins show similar enzymatic activities. However, trimer formation might improve overall protein stability.
Oligomerization of AOC2
In former investigations, AOC2 has been characterized to form
dimers based on size exclusion chromatography (Ziegler et al.,
1997), but several lines of evidence point toward a trimer as the
Active Site and Reaction Mechanism
Based on the structural similarity of AOC2 to the lipocalins, we
propose that the active site is located inside the barrel cavity.
Crystal Structure of AOC2
While most lipocalins just bind small hydrophobic substrates in
the interior, very few are known to be enzymatically active. One example is noteworthy: human b-trace or prostaglandinsynthase-D
catalyzes the conversion of prostaglandin H2 to prostaglandin
D2. As in the case of AOC2, the substrate is unstable in water and
catalysis in a protected environment seems to be favorable.
Here, the active site Cys is located in the barrel interior (Nagata
et al., 1991).
Indeed, in AOC2 we find the competitive inhibitor vernolic acid
to be bound inside the barrel of one monomer. We don’t observe
any induced fit mechanism, as the protein scaffold superposes
with an RMSD of 0.33 Å for all 174 Ca atoms with SeMet-AOC2.
Even on the level of side chains only marginal differences are
present both in comparison to SeMet-AOC2 and to the other two
monomers without bound inhibitor.
Surprisingly, the carboxylic end of the inhibitor does not seem
to be tightly bound to the protein, as there is no clear electron
density for the first five carbon units. This is in contrast with the
observation of Ziegler and coworkers that methylation of this
group abolishes the inhibitory effect for AOC from maize (Ziegler
et al., 1997). One possible explanation is the exchange of Leu-27
of AOC2 to Arg in maize AOC, which would be positioned to form
a strong salt bridge to the inhibitor in this protein. Another
possibility would be that the carboxylic moiety is involved in the
interaction with AOS for both proteins. Because we don’t have
data on the influence of methylation on inhibition of AOC2, a
decision between both cases is not possible yet. Work is in
progress to obtain comparable data on both proteins with the
different forms of the inhibitor.
The finding that the location of the epoxy group in the 12,13position of vernolic acid is essential for binding to AOC from
maize (Ziegler et al., 1997) is consistent with the observed binding position in our structure. Any shift of the epoxy group would
inhibit the formation of the hydrogen bond to the conserved
water molecule.
11 of 17
To allow a discussion of possible reaction mechanisms, we
tentatively modeled 12,13-EOT (substrate) and OPDA (product)
molecules into the pocket in the same binding mode as vernolic
acid (Figure 9A). Based on the resulting arrangement of protein
side chains around the bound substrate, we postulate the
following reaction mechanism for the cyclization reaction inside
the protein (Figure 10).
After binding into the pocket, the opening of the oxirane ring
and the subsequent formation of the classical pentadienyl cation
will be controlled by the protein environment. The influence of the
negative charge of Glu-23 leads to a delocalization of the C15
double bond toward the oxirane, which promotes its opening and
the cyclization reaction by the mechanism of anchimeric assistance (Figure 10A). Formation of the oxyanion is stabilized by the
tightly bound water molecule Water75. For the subsequent pericyclic ring closure, a conformational change around the C10-C11
bond from trans to cis geometry is necessary, producing a nonplanar ring-like pentadienyl carbocation (Figure 10B). Phe-51 is
positioned to stabilize this by p–p interaction. Additionally, Cys-71
is in a favorable position to stabilize the positive carbocation. This
trans-cis isomerization around C10-C11 has to be accompanied
by a cis-trans rotation around the C8-C9 bond due to the steric
restrictions in the protein cavity. Additionally, the conformational
change is promoted by the hydrophobic effect, as it buries more of
the hydrocarbon tail of the molecule inside the cavity.
The proposed reaction scheme assumes as a final step a
classical conrotary pericyclic ring closure, according to the
Woodward-Hoffmann rules (Figure 10C). The absolute stereoselectivity of the reaction follows from the steric restrictions by
the protein environment on the formation of the cis intermediate.
Phe-85, Val-45, and Tyr-105 especially seem to form a greasy
slide, which at the same time facilitates and restricts the conformational change of the hydrocarbon tail. Indeed, the F85L and
F85A mutants show a reduced reactivity and stereoselectivity of
the reaction. The latter effect would be explained by the removal
Figure 9. Substrate and Product Modeled into the Binding Pocket of AOC2 and Comparison with PGHS-1.
(A) Shown are 12,13-EOT and OPDA as green and blue sticks, respectively. The protein is shown in a similar representation as in Figure 7. Also drawn is
the molecular protein surface colored according to atom types. Possible hydrogen bonds between substrate and protein are shown as dashed green
lines.
(B) View into the cyclooxygenase active site of PGHS-1 from sheep (Ovis aris) (accession number 1DIY) with bound substrate arachidonic acid (AA). A
similar representation is chosen as compared with (A). Also shown are the residues discussed in the text.
12 of 17
The Plant Cell
Figure 10. Schematic Overview of the Proposed Cyclization Reaction.
The AOC2 catalyzed cyclization reaction involves the opening of the
oxirane ring and a subsequent conrotary pericyclic ring closure (see text
for details). The fatty acid moiety of 12,13-EOT [(CH2)7COOH] is labeled
as R, and the schematic binding pocket of AOC2 is displayed in gray.
of some of the steric restrictions on the trans-cis rotation around
C10-C11. A rotation in the opposite direction would automatically result in the opposite stereoisomer of the product.
Our proposed mechanism explains the importance of the
C15 double bond for reactivity along the lines of the proposal of
Grechkin (1994) by the mechanism of anchimeric assistance. It
also rationalizes the impact of the protein environment on the
stereoselectivity of the cyclization reaction. It is clear by the
packing in the cavity that changes in the position of the epoxide
group will affect binding affinities of possible substrates.
A similar biosynthesis of cyclic hormones can be found in
mammals during formation of prostanoids, which are structurally
similar to JAs. One of the enzymes involved is the Lipocalin-type
prostaglandin D synthase, which has been shown to be an example for an enzymatically active lipocalin (Urade and Hayaishi,
2000). The enzyme responsible for the first committed step in
prostanoid catalysis from the linear precursor is prostaglandin
endoperoxide synthase (PGHS). Due to its medical relevance as
target for anti-inflammatory drugs like aspirin, it has been very
well characterized. PGHS catalyzes both the cyclization of
arachidonic acid to prostaglandin G2 and the successive peroxidation of PGG2 to prostaglandin H2. While PGHS shows a
completely different overall architecture than AOC, the functional
homology makes a comparison of the active sites worthwhile.
PGHS is a dimeric monotopic membrane protein with a
molecular mass of ;70 kD (monomer), which is located in the
endoplasmic reticulum and nuclear envelopes. Besides an
N-terminal EGF domain, the protein is mainly a-helical and has
two distinct active sites for the two catalyzed reaction steps
(Picot et al., 1994; for a review, see Garavito et al., 2002). In
the cyclooxygenase active site, arachidonic acid binds with the
carboxylic end, forming salt bridges to a conserved Arg residue.
The rest of the molecule is tightly packed into an L-shaped cavity
lined mainly with hydrophobic and aromatic residues, as found in
the AOC2 binding pocket (Figure 9B). The catalytically active Tyr
is positioned at the kink of the cavity for abstraction of the 13proS
hydrogen, the initiating step of the reaction. Interestingly, the
only other polar residue is situated on the opposite side of the
polycarbon chain, similar to the position of Cys-71 in the case of
AOC2. This has been shown to be critical in determining the
stereochemistry of the product (Schneider et al., 2002). It has
been found that the packing at the v-end of the substrate
molecule is very important for maintaining the catalytic selectivity
and activity (Rowlinson et al., 1999; Malkowski et al., 2001). While
increasing the size of residues in this region can abolish activity
due to steric inhibition of binding, mutations increasing the volume of the binding pocket mostly diminish activity, as optimal
positioning of substrate is not enforced. It has been proposed
that the carboxylic end stays fixed during catalysis in PGHS,
while the v-end has to move closer to allow for the substantial
conformational changes necessary for ring closure (Garavito
et al., 2002). In the case of AOC2 from Arabidopsis, a movement
of the carboxyl end of the substrate along our proposed reaction
scheme seems to be more likely. While the relative flexibility of
the carboxy group of vernolic acid might not represent the
situation for the substrate 12,13-EOT in planta, there are no
positively charged residues appropriately placed to form such a
strong salt linkage as observed in the case of PGHS.
Here, we find two completely different structures showing the
same binding principles. A hydrophobic mold binds the substrate exactly in the right position to a single catalytically critical
polar site without apparent induced fit during binding. While the
exact reaction schemes are very different, the hydrophobic mold
in both cases also ensures the stereospecificity of the reaction.
As the comparison with PGHS shows, residues in the vicinity of
the substrate binding site are probably highly optimized to control the reaction in the cavity. Structural and functional analyses
of additional point mutants will help to clarify the individual
influence of the key amino acids in the binding pocket on the
proposed mechanism and the stereoselectivity of the reaction.
Crystal Structure of AOC2
METHODS
Protein Expression and Purification
A truncated version of AOC2 from Arabidopsis thaliana has been cloned
into the vector pQE30 (Qiagen) and pET21b (Novagen/Merck) to yield
a fusion protein (20.2 kD) in which the first 77 amino acids (the predicted transit signal) have been replaced by a His6-tag (sequence
MRGSHHHHHHRS) or in which the His6-tag has been coupled with the
C terminus of the truncated protein. The resulting constructs were
transformed into Escherichia coli strain M15. Cells were grown in 2YT
media and induced at an OD of 0.5 to 0.6 with addition of 0.2 mM
isopropylthio-b-galactoside and 0.2% arabinose, respectively. E. coli
with the pQE30 construct were harvested after an induction time of ;5 h
at 378C and 220 rpm. Bacteria with the pET21b construct were induced
for a further 30 to 40 h at 258C. Cells were broken by ultrasonication, and
the fusion protein was purified by affinity chromatography (Ni-NTA;
Qiagen) and concentrated with Centricon concentrators (5000 D cutoff;
Millipore). The yield of purified protein exceeded 10 mg/L culture.
SeMet-labeled N-terminally His6-tagged AOC2 was produced by
growth of M15 cells in M9 minimal medium supplemented with SeMet
as described (Van Duyne et al., 1993). The mutated proteins of AOC2
were expressed in the pQE vector under the same conditions as the
native AOC2.
AOS (pQE30-AOS; Laudert et al., 1996) was expressed and purified in
accordance with Oh and Murofushi (2002). Protein expression was
induced at an OD600 of ;0.5 by addition of isopropylthio-b-galactoside
to a final concentration of 0.4 mM and further incubation at 168C and 150
rpm overnight. Cells were broken by ultrasonication, and recombinant
AOS was purified via Ni-NTA agarose chromatography (Qiagen) by
adding Triton X-100 up to a final concentration of 0.1% (v/v) to the
recommended buffers.
Generation of Mutant Proteins of AOC2
The generation of AOC2 mutants was done via PCR according to the
sequential PCR step method of the current protocols in molecular
biology. Phe at position 85 was exchanged to Ala (primers: 59-GAAAGGTGAAAGAGCCGAAGCTACTTATA-39and 59-TATAAGTAGCTTCGGCTCTTTCACCTTTC-39) and Leu (primers: 59-GAAAGGTGAAAGACTCGAAGCTACTTATA-39and 59-TATAAGTAGCTTCGAGTCTTTCACCTTTC-39).
The N- and C-terminal primers were 59-TATGGATCCCCAAGCAAAGTTCAAGAACTG-39 (BamHI restriction site underlined) and 59-TATGTCGACTTAGTTGGTATAGTTACTTATAAC-39 (SalI restriction site underlined). The mutated cDNA was cloned afterwards into the expression
vector pQE30, and the protein was expressed under standard conditions.
Gel Electrophoresis and Protein Immunoblotting
Denaturing gel electrophoresis was performed according to Laemmli
(1970). The discontinuous systems consisted of 4% stacking gels and
12.5% resolving gels. Protein blotting onto nitrocellulose was done
electrophoretically overnight (48C, 64 mA) as described by Towbin et al.
(1979). Immunodetection followed standard procedures (Parets-Soler
et al., 1990) with goat anti-rabbit immunoglobin-conjugated alkaline
phosphatase and the enzyme substrates 4-nitrotetrazolium blue and
5-bromo-4-chloro-3-indolyl phosphate.
Determination of Total Enzymatic Activity via
Chiral Capillary GC-MS
For activity analysis, 5 to 20 mg of purified AOS and AOCs (weight-based
ratio 1:2) were added to 1 mL of 10 mM PPi buffer, pH 7.0. The reaction
was started by addition of 100 mg of HPOT. After incubation for 15 min at
13 of 17
room temperature, the reaction was stopped by acidifying with 1 M HCl to
pH 3.0. The synthesized OPDA was extracted twice with 2 volumes of
ethyl acetate and taken to dryness. For GC-MS analysis, 1 nmol of 2[H]5OPDA was added before extraction as an internal standard. Dried
samples were incubated in 250 mL of 0.1 M KOH for 45 min at room
temperature to convert the synthesized OPDA into trans-configuration.
After a second extraction step, OPDA was dissolved in methanol and
methylated with ethereal diazomethane (Hamberg and Fahlstadius,
1990). The dried fractions were then redissolved in 50 to 100 mL of
chloroform, and 1 mL of the sample was injected into a Varian GC 3400
gas chromatograph in splitless mode in direct connection to a MAT
Magnum ion trap mass spectrometer (Finnigan) using a chemical ionization mode with methanol as reactant gas. Separations of the (þ)- and the
()-enantiomer of OPDA were made on a b-Dex120 column (30 m 3 0.25
mm 3 0.15 mm stationary-phase thickness) coated with 20% permethylb-cyclodextrin in SPB-35 (Supelco) with He carrier gas at 1 mL/min (gas
prepressure of 80 kPa) using the following temperature program: injector
temperature of 2208C, 1 min isothermally at 508C, with 108C/min up to
1908C, 75 min isothermally at 1908C, with 38C/min up to 2208C, 15 min
isothermally at 2208C, and transfer line temperature 2208C. For details,
see Laudert et al. (1997) and Schaller et al. (1998). Quantitation of AOC
enzymatic activity was based on (þ)-OPDA-to-()-OPDA ratios derived
from integration of total ion current traces.
Crystallization
Initial crystallization screening for AOC2 was performed using vapor
diffusion against the Wizard I and II kits from Emerald BioSystems. Protein
crystals were found with condition WI-28, which was subsequently
optimized. Using 14 to 16% PEG-8000, 6% tert-butanol, and 0.1 M
MES, pH 5.0 to 6.0, as reservoir solution in either hanging- or sitting-drop
setups (4 mg/mL protein concentration), we were able to grow crystals
(space group P21) up to a size of 250 3 150 3 50 mm, which diffract to
;1.7-Å resolution.
AOC2-SeMet crystals (space group P212121) were grown from reservoir solutions containing 10% PEG-3350, 200 mM NaCl, and 100 mM
phosphate-citrate, pH 4.2, with a protein concentration of 13.5 mg/mL.
They typically reach a size of 250 3 200 3 100 mm and diffract to better
than 1.3-Å resolution. Crystals grew at 188C in ;1 week.
For soaking experiments, AOC2-SeMet crystals were transferred for
1 to 2 d into appropriate mother liquor saturated with inhibitor.
X-Ray Diffraction Data Collection
Data collection was performed at 100K on flash-frozen crystals at the
Swiss Light Source (Villigen, Switzerland) at beamline PXII using a MAR
CCD detector. Cryoprotection was achieved by briefly soaking the
crystals in paraffin oil or in mother liquor supplemented with glycerol.
For SeMet-AOC2 crystals, MAD data at the Se edge were collected at the
European Synchroton Radiation Facility (Grenoble, France) on beamline
ID13.1 at 100K using a MAR CCD detector. Data of inhibitor-soaked
SeMet-AOC2 crystals were collected in house at 100K on a Bruker FR591
rotating anode with Montel multilayer optics and a MAR345dtb detector
system.
Data were processed using the XDS package (Kabsch, 1993). Statistics
of the different data sets are shown in Table 1.
Structure Solution and Refinement
SeMet-AOC2
Self-rotation functions indicated the presence of a threefold noncrystallographic symmetry. Search for the Se substructure and initial phasing
were done with the programs of the SHELX package (Sheldrick and
14 of 17
The Plant Cell
Schneider, 1997) using the hkl2map interface (Pape and Schneider, 2004)
with data to 2-Å resolution. SHELXD identified four major peaks, two of
which in retrospect correspond to two copies of SeMet42. The other two
are a double peak corresponding to the third copy of SeMet42. This
solution showed correlation coefficients of 41 for all data and 32 for weak
data. Phasing with SHELXE using data to 3 Å clearly identified the correct
hand and already resulted in a well-interpretable map. Fifty-eight cycles
of density modification using an estimated solvent content of 0.4 resulted
in density contrast of 0.73 (0.38), connectivity of 0.95 (0.91), and a
pseudo-free correlation coefficient of 83.2 (63.3) (values for the wrong
hand in parentheses).
Density modification (using the threefold symmetry), phase extension,
and auto building in RESOLVE (Terwilliger, 2000) could assign 387 of the
expected 564 (3 3 188) residues, including side chains. An additional 44
residues were placed without side chains.
For refinement, the peak wavelength data set was used. Five percent of
the data were randomly assigned to the Rfree test set and omitted from
refinement.
The resulting model was further improved in several cycles of refinement, automated water placement, and manual rebuilding using programs of the CCP4 package (Collaborative Computational Project,
Number 4, 1994) and ARP/wARP (Cohen et al., 2004).
The final model contains three protein monomers composed of residues 15 to 188. The first 14 residues, including the His6-tag, could not be
placed due to missing density. Based on clear density, a total of 19
residues had to be modeled with alternate conformations. Additionally,
three glycerol and 556 water molecules were included. Part of the
experimental electron density map after RESOLVE with the final model
is shown in Supplemental Figure 2 online.
related molecule. Therefore, binding of inhibitor during soaking in this
monomer is prohibited. Weak NCS restraints were used in the initial
stages of the refinement but were released after inclusion of the inhibitor.
For preparation of the omit maps, vernolic acid was removed from the
final model and random positional shifts (average RMSD 0.12 Å) were
applied to the remaining coordinates to remove model bias. After another
round of refinement with REFMAC, an mFo-DFc difference omit map was
calculated.
Figure Preparation
Figures were generated using Pymol (http://www.pymol.org). To allow
coloring of residues by conservation, an alignment of 22 different AOCs
found with a National Center for Biotechnology Information BLAST search
in the UNIREF100 database was analyzed using the ESPript server (Gouet
et al., 1999). In the resulting coordinate file, the similarity score has been
assigned to the B-factor column, ranging from 100 (fully conserved) to 0
(no conservation). This range was assigned to the color range red to blue
for figure generation.
Superposition of molecules was either performed in COOT (Emsley and
Cowtan, 2004) or using the SuperPose server (Maiti et al., 2004).
Accession Numbers
Coordinates and structure factors can be found in the Protein Data Bank
under accession numbers 2BRJ, 2GIN, and 2DIO for the AOC2-SeMet,
WT-AOC2, and AOC2-SeMet inhibitor structure, respectively.
Supplemental Data
WT-AOC2
The following materials are available in the online version of this article.
Supplemental Figure 1. Conservation of the Trimer Interface.
The structure of WT-AOC2 was solved using the refined coordinates of
SeMet-AOC2 after removal of solvent and alternate conformations and
the mutation of SeMet to Met as a search model. Two trimers were found
in the asymmetric unit of the monoclinic unit cell. After several rounds of
refinement in REFMAC (Murshudov et al., 1997) and manual rebuilding
and water picking in COOT (Emsley and Cowtan, 2004), the final model
consists of 1036 amino acids, five glycerol molecules, two sodium ions,
and 660 water molecules. Gln-102 had to be modeled in alternate
conformations in five of the six copies of AOC2 in the asymmetric unit.
Weak NCS restraints were used throughout the refinement.
SeMet-AOC2 Inhibitor
For analysis of the inhibitor data, one copy of the 2BRJ model without
alternate conformations and solvent molecules was placed in the unit cell.
After several rounds of refinement and water picking, extended residual
density in monomer B was fitted with a model of the inhibitor vernolic acid.
The initial atomic coordinates and the necessary geometry definitions
were created using the PRODRG server (Schuettelkopf and van Aalten,
2004). At the end of the refinement, the occupancy of the inhibitor was set
to 0.25, 0.5, 0.75, and 1 and the structure refined using identical protocols. With full occupancy, the best R-factors were reached. While we
observed clear density for the majority of the inhibitor, the carboxylic end
up to C5 did not show clear density due to disorder (Figure 7A). The shape
of the resulting difference density together with the higher flexibility of the
inhibitor in comparison to the surrounding protein atoms also points to
different minor binding modes of the inhibitor (possibly also the second
enantiomer), which were not modeled.
Elongated residual density in monomer C was judged too weak to
safely allow positioning of the inhibitor. In monomer A, the entrance to the
binding site is completely blocked by Leu-41 from a crystallographically
Supplemental Figure 2. MAD-Solvent-Flattened Electron Density of
AOC2.
ACKNOWLEDGMENTS
We thank C. Kötting for helpful discussions, K. Diederichs and D.R.
Madden for comments on the manuscript, and E.W. Weiler for continuing support and interest in the work. Part of this work was performed at
the Swiss Light Source, Beamline X10SA, Paul Scherrer Institute
(Villigen, Switzerland) and at the European Synchroton Radiation Facility, Beamline ID13.1 (Grenoble, France). We would especially like to
acknowledge the help of the respective beamline staff and of our
colleagues from the Max Planck Institutes for Medical Research
(Heidelberg) and Molecular Physiology (Dortmund) during data collection. We also acknowledge financial support from the Deutsche Forschungsgemeinschaft by Grants SFB480-C6 (E.H.), HO2600 (E.H.), and
SCHA939 (F.S.).
Received May 10, 2006; revised August 4, 2006; accepted October 17,
2006; published November 3, 2006.
REFERENCES
Badami, R.C., and Patil, K.B. (1980). Structure and occurrence of
unusual fatty acids in minor seed oils. Prog. Lipid Res. 19, 119–153.
Baertschi, S.W., Ingram, C.D., Harris, T.M., and Brash, A.R. (1988).
Absolute configuration of cis-12-oxophytodienoic acid of flaxseed:
Crystal Structure of AOC2
Implications for the mechanism of biosynthesis from the 13(S)hydroperoxide of linolenic acid. Biochemistry 27, 18–24.
Blée, E. (2002). Impact of phyto-oxylipins in plant defense. Trends Plant
Sci. 7, 315–322.
Blée, E., and Joyard, J. (1996). Envelope membranes from spinach
chloroplasts are a site of metabolism of fatty acid hydroperoxides.
Plant Physiol. 102, 821–828.
Brash, A.R., Baertschi, S.W., Ingram, C.D., and Harris, T.M. (1988).
Isolation and characterization of natural allene oxides: Unstable
intermediates in the metabolism of lipid hydroperoxides. Proc. Natl.
Acad. Sci. USA 85, 3382–3386.
Burgos, R.C., Hieber, A.D., and Yamamoto, Y. (1998). Xanthophyll
cycle enzymes are members of the lipocalin family, the first identified
from plants. J. Biol. Chem. 273, 15321–15324.
Charron, J.-B.F., Breton, G., Badawi, M., and Sarhan, F. (2002). Molecular and structural analyses of a novel temperature stress-induced
lipocalin from wheat and Arabidopsis. FEBS Lett. 517, 129–132.
Charron, J.-B.F., Ouellet, F., Pelletier, M., Danyluk, J., Chauve, C.,
and Sarhan, F. (2005). Identification, expression, and evolutionary
analyses of plant lipocalins. Plant Physiol. 139, 2017–2028.
Cohen, S.X., Morris, R.J., Fernandez, F.J., Ben Jelloul, M., Kakaris,
M., Parthasarathy, V., Lamzin, V.S., Kleywegt, G.J., and Perrakis,
A. (2004). Towards complete validated models in the next generation
of ARP/wARP. Acta Crystallogr. D60, 2222–2229.
Collaborative Computational Project, Number 4 (1994). The CCP4
suite: Programs for protein crystallography. Acta Crystallogr. D Biol.
Crystallogr. D50, 760–763.
Conconi, A., Miquel, M., Browse, J.A., and Ryan, C.A. (1996). Intracellular levels of free linolenic and linoleic acids increase in tomato
leaves in response to wounding. Plant Physiol. 111, 797–803.
Crombie, L., and Morgan, D.O. (1988). Formation of acylic a- and
g-ketols and 12-oxophytodienoic acid from linolenic acid 13-hydroperoxide by a flax enzyme preparation. Evidence for a single enzyme
leading to a common allene epoxide intermediate. J. Chem. Soc.
Perkin Trans. 8, 558–560.
Dathe, W., Rönsch, H., Preiss, A., Schade, W., Sembdner, G., and
Schreiber, K. (1981). Endogenous plant hormones of the broad bean,
Vicia faba L. (-)-Jasmonic acid, a plant growth inhibitor in pericarp.
Planta 153, 530–535.
Demole, E., Lederer, E., and Mercier, D. (1962). Isolement et détermination de la structure du jasmonate de méthyle, constituant odorant charactéristique de l’essence de jasmin. Helv. Chim. Acta 45,
675–685.
Diederichs, K., and Karplus, P.A. (1997). Improved R-factors for
diffraction data analysis in macromolecular crystallography. Nat.
Struct. Biol. 4, 269–275.
Emanuelsson, O., Nielsen, H., and von Heijne, G. (1999). ChloroP, a
neutral network-based method for predicting chloroplast transit peptides and their cleavage sites. Protein Sci. 8, 978–984.
Emsley, P., and Cowtan, K. (2004). Coot: Model-building tools for
molecular graphics. Acta Crystallogr. D60, 2126–2132.
Farmer, E.E., Johnson, R.R., and Ryan, C.A. (1991). Regulation of
expression of proteinase inhibitor genes by methyl jasmonate and
jasmonic acid. Plant Physiol. 98, 995–1002.
Farmer, E.E., and Ryan, C.A. (1990). Interplant communication: Airborne methyl jasmonate induces synthesis of proteinase inhibitors in
plant leaves. Proc. Natl. Acad. Sci. USA 87, 7713–7716.
Farmer, E.E., and Ryan, C.A. (1992). Octadecanoid-derived signals in
plants. Trends Cell Biol. 2, 236–241.
Feussner, I., and Wasternack, C. (2002). The lipoxygenase pathway.
Annu. Rev. Plant Biol. 53, 275–297.
Flower, D.R. (1994). The lipocalin protein family: A role in cell regulation.
FEBS Lett. 354, 7–11.
15 of 17
Flower, D.R., North, A.C., and Sansom, C.E. (2000). The lipocalin
protein family: Structural and sequence overview. Biochim. Biophys.
Acta 1482, 9–24.
Flower, D.R., Sansom, C.E., Beck, M.E., and Altwood, T.K. (1995).
The first prokaryotic lipocalins. Trends Biochem. Sci. 20, 498–499.
Ganfornina, M.D., Guitierrez, G., Bastiani, M., and Sanchez, D.
(2000). A phylogenetic analysis of the lipocalin protein family. Mol.
Biol. Evol. 17, 114–126.
Garavito, R.M., Malkowski, M.G., and de Witt, D.L. (2002). The
structures of prostaglandin endoperoxide H synthases-1 and -2.
Prostaglandins Other Lipid Mediat. 68–69, 129–152.
Gerhardt, B. (1983). Localization of b-oxidation enzymes in peroxisomes isolated from nonfatty plant tissues. Planta 159, 238–246.
Gouet, P., Courcelle, E., Stuart, D.I., and Metoz, F. (1999). ESPript: Multiple sequence alignments in PostScript. Bioinformatics 15, 305–308.
Grechkin, A.N. (1994). Cyclization of natural allene oxide fatty acids.
The anchimeric assistance of beta, gamma-double bond beside the
oxirane and the reaction mechanism. Biochim. Biophys. Acta 14,
199–206.
Grzyb, J., Latowski, D., and Strzalka, K. (2006). Lipocalins – A family
portrait. J. Plant Physiol. 163, 895–915.
Gundlach, H., Müller, M.J., Kutchan, T.M., and Zenk, M.H. (1992).
Jasmonic acid as a signal transducer in elicitor-induced plant cell
cultures. Proc. Natl. Acad. Sci. USA 89, 2389–2393.
Gunstone, F.D. (1954). Fatty acids: Part II: The nature of the oxygenated
acid present in Vernonia anthelmintica (Willd.) seed oil. J. Chem. Soc.
(Lond.) May, 1611–1616.
Hamberg, M., and Fahlstadius, P. (1990). Allene oxide cyclase: A new enzyme in plant lipid metabolism. Arch. Biochem. Biophys. 276, 518–526.
Hamberg, M., and Hughes, M.A. (1988). Fatty acid allene oxides. III.
Albumin-induced cyclization of 12,13(S)-epoxy-9(Z),11-octadecadienoic acid. Lipids 23, 469–475.
He, Y., Fukushige, H., Hildebrand, D.F., and Gan, S. (2002). Evidence
supporting a role of jasmonic acid in Arabidopsis leaf senescence.
Plant Physiol. 128, 876–884.
Holm, L., and Sander, C. (1994). Searching protein structure databases
has come of age. Proteins 19, 165–173.
Kabsch, W. (1993). Automatic processing of rotation diffraction data
from crystals of initially unknown symmetry and cell constants. J.
Appl. Crystallogr. 26, 795–800.
Kleinman, R., Smith, C.R., Jr., and Yates, S.G. (1965). Search for new
industrial oils: Fifty-eight Euphorbiaceae oils, including one rich in
vernolic acid. J. Am. Oil Chem. Soc. 42, 169–172.
Kramell, R., Miersch, O., Hause, B., Ortel, B., Parthier, B., and
Wasternack, B. (1997). Amino acid conjugates of jasmonic acid
induce jasmonate-responsive gene expression in barley (Hordeum
vulgare L.) leaves. FEBS Lett. 414, 197–202.
Laemmli, U.K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685.
Laudert, D., Hennig, P., Stelmach, B.A., Müller, A., Andert, L., and
Weiler, E.W. (1997). Analysis of 12-oxo-phytodienoic acid enantiomers in biological samples by capillary gas chromatography-mass
spectrometry using cyclodextrin stationary phases. Anal. Biochem.
246, 211–217.
Laudert, D., Pfannschmidt, U., Lottspeich, F., Holländer-Czytko, H.,
and Weiler, E.W. (1996). Cloning, molecular and functional characterization of Arabidopsis thaliana allene oxide synthase (CYP74), the
first enzyme of the octadecanoid pathway to jasmonates. Plant Mol.
Biol. 31, 323–335.
Li, C., Schilmiller, A.L., Liu, G., Lee, G.I., Jayanty, S., Sageman, C.,
Vrebalov, J., Giovannoni, J.J., Yagi, K., Kobayashi, Y., and Howe,
G.A. (2005). Role of beta-oxidation in jasmonate biosynthesis and
systemic wound signaling in tomato. Plant Cell 17, 971–986.
16 of 17
The Plant Cell
Maiti, M., van Domselaar, G.H., Zhang, H., and Wishart, D.S. (2004).
SuperPose: A simple server for sophisticated structural superposition.
Nucleic Acids Res. 32, W590–W594.
Malkowski, M.G., Thuresson, E.D., Lakkides, K.M., Rieke, C.J., Micielli,
R., Smith, W.L., and Garavito, R.M. (2001). Structure of eicosapentaenoic and linoleic acids in the cyclooxygenase site of prostaglandin
endoperoxide H synthase-1. J. Biol. Chem. 276, 37547–37555.
Maucher, H., Stenzel, I., Miersch, O., Stein, N., Prasad, M., Zierold,
U., Schweizer, P., Dorer, C., Hause, B., and Wasternack, C. (2004).
The allene oxide cyclase of barley (Hordeum vulgare L.)–Cloning and
organ-specific expression. Phytochemistry 65, 801–811.
Meyer, A., Miersch, O., Büttner, C., Dathe, W., and Sembdner, G.
(1984). Occurrence of the plant growth regulator jasmonic acid in
plants. J. Plant Growth Regul. 3, 1–8.
Mueller, M.J., Brodschelm, W., Spannagl, E., and Zenk, M.H. (1993).
Signaling in the elicitation process is mediated through the octadecanoid pathway leading to jasmonic acid. Proc. Natl. Acad. Sci. USA
90, 7490–7494.
Murshudov, G.N., Vagin, A.A., and Dodson, E.J. (1997). Refinement of
macromolecular structures by the Maximum-Likelihood Method. Acta
Crystallogr. D53, 240–255.
Murzin, A.G. (1994). Principles determining the structure of b-sheet
barrels in proteins. J. Mol. Biol. 236, 1369–1381.
Murzin, A.G., Brenner, S.E., Hubbard, T., and Chothia, C. (1995).
SCOP: A structural classification of proteins database for the investigation of sequences and structures. J. Mol. Biol. 247, 536–540.
Nagata, A., Suzuki, Y., Igarashi, M., Eguchi, N., Toh, H., Urade, Y.,
and Hayaishi, O. (1991). Human brain prostaglandin D synthase has
been evolutionarily differentiated from lipophilic-ligand carrier proteins. Proc. Natl. Acad. Sci. USA 88, 4020–4024.
Narváez-Vásquez, J., Florin-Christensen, J., and Ryan, C.A. (1999).
Positional specificity of a phospholipase A2 activity induced by
wounding, systemin, and oligosaccharide elicitors in tomato leaves.
Plant Cell 11, 2249–2260.
Oh, K., and Murofushi, N. (2002). Design and synthesis of novel
imidazole derivatives as potent inhibitors of allene oxide synthase
(CYP74). Bioorg. Med. Chem. 10, 3707–3711.
Oldham, M.L., Brash, A.R., and Newcomer, M.E. (2005). The structure
of coral allene oxide synthase reveals a catalase adapted for metabolism of a fatty acid hydroperoxide. Proc. Natl. Acad. Sci. USA 102,
297–302.
Pan, Z., Durst, F., Werck-Reichhart, D., Gardner, H.W., Camara, B.,
Cornish, K., and Backhaus, R.A. (1995). The major protein of
guayule rubber particles is a cytochrome P450. Characterization
based on cDNA cloning and spectroscopic analysis of the solubilized
enzyme and its reaction products. J. Biol. Chem. 270, 8487–8494.
Pape, T., and Schneider, T.R. (2004). HKL2MAP: A graphical user
interface for phasing with SHELX programs. J. Appl. Crystallogr. 37,
843–844.
Parets-Soler, A., Pardo, J.M., and Serrano, R. (1990). Immunocytolocalization of plasmamembrane Hþ-ATPase. Plant Physiol. 93,
1654–1658.
Pervaiz, S., and Brew, K. (1985). Homology of b-lactoglobulin, serum
retinol-binding protein, and protein HC. Science 228, 335–337.
Pervaiz, S., and Brew, K. (1987). Homology and structure-function
correlation between alpha 1-acid glycoprotein and serum retinolbinding protein and its relatives. FASEB J. 1, 209–214.
Picot, D., Loll, P.J., and Garavito, R.M. (1994). The X-ray crystal
structure of the membrane protein prostaglandin H2 synthase-1.
Nature 367, 243–249.
Reymond, P., and Farmer, E.E. (1998). Jasmonate and salicylate as
global signals for defense gene expression. Curr. Opin. Plant Biol. 1,
404–411.
Rowlinson, S.W., Crews, B.C., Lanzo, C.A., and Marnett, L.J. (1999).
The binding of arachidonic acid in the cyclooxygenase active site of
mouse prostaglandin endoperoxides synthase-2 (COX-2). J. Biol.
Chem. 274, 23305–23310.
Salier, J.P. (2000). Chromosomal location, exon/intron organization and evolution of lipocalin genes. Biochim. Biophys. Acta 18,
25–34.
Sanchez, D., Ganfornia, M.D., Gutierrez, G., and Marin, A. (2003).
Exon-intron structure and evolution of the lipocalin gene family. Mol.
Biol. Evol. 20, 775–783.
Schaller, F., Biesgen, C., Müssig, C., Altmann, T., and Weiler, E.W.
(2000). 12-Oxophytodienoate reductase 3 (OPR3) is the isoenzyme
involved in jasmonate biosynthesis. Planta 210, 979–984.
Schaller, F., Hennig, P., and Weiler, E.W. (1998). 12-Oxophytodienoate-10,11-reductase: Occurrence of two isoenzymes of different
specificity against stereoisomers of 12-oxophytodienoic acid. Plant
Physiol. 118, 1345–1351.
Schaller, F., Schaller, A., and Stintzi, A. (2005). Biosynthesis and
metabolism of jasmonates. J. Plant Growth Regul. 23, 179–199.
Schaller, F., and Weiler, E.W. (1997a). Enzymes of octadecanoid
biosynthesis in plants. 12-oxo-phytodienoate 10,11-reductase. Eur.
J. Biochem. 245, 294–299.
Schaller, F., and Weiler, E.W. (1997b). Molecular cloning and characterization of 12-oxophytodienoate reductase, an enzyme of the
octadecanoid signaling pathway from Arabidopsis thaliana. Structural
and functional relationship to yeast old yellow enzyme. J. Biol. Chem.
272, 28066–28072.
Schlehuber, S., and Skerra, A. (2005). Lipocalins in drug discovery:
From natural ligand-binding proteins to ‘‘anticalins’’. Drug Discov.
Today 10, 23–33.
Schneider, C., Boeglin, W.E., Prusakiewicz, J.J., Rowlinson, S.W.,
Marnett, L.J., Samel, N., and Brash, A.R. (2002). Control of prostaglandin stereochemistry at the 15-carbon by cyclooxygenase-1 and
-2: A critical role for Serine-530 and Valine-349. J. Biol. Chem. 277,
478–485.
Schuettelkopf, A.W., and van Aalten, D.M.F. (2004). PRODRG - A tool
for high-throughput crystallography of protein-ligand complexes. Acta
Crystallogr. D60, 1355–1363.
Sheldrick, G.M., and Schneider, T.R. (1997). SHELXL: High-resolution
refinement. Methods Enzymol. 277, 319–343.
Siedow, J.N. (1991). Plant lipoxygenase: Structure and function. Annu.
Rev. Plant Physiol. 132, 1065–1076.
Song, W.C., and Brash, A.R. (1991). Purification of an allene oxide
synthase and identification of the enzyme as a cytochrome P-450.
Science 253, 781–784.
Song, W.C., Funk, C.D., and Brash, A.R. (1993). Molecular cloning of
an allene oxide synthase: A cytochrome P450 specialized for the
metabolism of fatty acid hydroperoxides. Proc. Natl. Acad. Sci. USA
90, 8519–8523.
Spitzer, V., Aitzemuller, K., and Vosmann, K. (1996). The seed oil of
Bernardia puchella (Euphorbiaceae) - A rich source of vernolic acid. J.
Am. Oil Chem. Soc. 73, 1733–1735.
Stelmach, B.A., Müller, A., Hennig, P., Laudert, D., Andert, L., and
Weiler, E.W. (1998). Quantitation of the octadecanoid 12-oxo-phytodienoic acid, a signalling compound in plant mechanotransduction.
Phytochemistry 47, 539–546.
Stenzel, I., Hause, B., Maucher, H., Pitzschke, A., Miersch, O., Ziegler,
J., Ryan, C.A., and Wasternack, C. (2003a). Allene oxide cyclase
dependence of the wound response and vascular bundle-specific
generation of jasmonates in tomato - Amplification in wound signalling.
Plant J. 33, 577–589.
Stenzel, I., Hause, B., Miersch, O., Kurz, T., Maucher, H., Weichert,
H., Ziegler, J., Feussner, I., and Wasternack, C. (2003b). Jasmonate
Crystal Structure of AOC2
biosynthesis and the allene oxide cyclase family of Arabidopsis
thaliana. Plant Mol. Biol. 51, 895–911.
Stintzi, A., Weber, H., Reymond, P., Browse, J., and Farmer, E.E.
(2001). Plant defense in the absence of jasmonic acid: The role of
cyclopentenones. Proc. Natl. Acad. Sci. USA 98, 12837–12842.
Strassner, J., Schaller, F., Frick, U., Howe, G.A., Weiler, E.W.,
Amrhein, N., Macheroux, P., and Schaller, A. (2002). Characterization
and cDNA-microarray expression analysis of 12-oxo-phyto-dienoate
reductase reveals differential roles for octadecanoid biosynthesis in the local versus the systemic wound response. Plant J. 32,
585–601.
Terwilliger, T.C. (2000). Maximum likelihood density modification. Acta
Crystallogr. D56, 965–972.
Towbin, H., Staehelin, T., and Gordon, J. (1979). Electrophoretic
transfer of proteins from polyacrylamide gels to nitrocellulose sheets:
Procedure and some applications. Proc. Natl. Acad. Sci. USA 76,
4350–4354.
Ueda, J., and Kato, J. (1980). Isolation and identification of a senescence-promoting substance from wormwood (Artemisia absinthium
L.). Plant Physiol. 66, 246–249.
Urade, Y., and Hayaishi, O. (2000). Biochemical, structural, genetic,
physiological, and pathophysiological features of lipocalin-type prostaglandin D synthase. Biochim. Biophys. Acta 1482, 259–271.
Van Duyne, G.D., Standaert, R.F., Karplus, P.A., Schreiber, S.L., and
Clardy, J. (1993). Atomic structure of the human immunophilin FKBP12 complexes with FK506 and rapamycin. J. Mol. Biol. 229, 105–124.
17 of 17
Vick, B.A., and Zimmerman, D.C. (1981). Lipoxygenase, hydroperoxide isomerase and hydroperoxide cyclase in young cotton seedlings.
Plant Physiol. 67, 92–97.
Vick, B.A., and Zimmerman, D.C. (1984). Biosynthesis of jasmonic acid
by several plant species. Plant Physiol. 75, 458–461.
Vick, B.A., and Zimmerman, D.C. (1987). Pathways of fatty acid
hydroperoxide metabolism in spinach leaf chloroplasts. Plant Physiol.
85, 1073–1078.
Wasternack, C., and Hause, B. (2002). Jasmonates and octadecanoids: Signals in plant stress responses and development. Prog.
Nucleic Acid Res. Mol. Biol. 72, 165–221.
Weiler, E.W., Laudert, D., Schaller, F., Stelmach, B., and Hennig, P.
(1998). Fatty acid-derived signaling molecules in the interaction of
plants with their environment. In Phytochemical Signals and PlantMicrobe Interactions. Recent Advances in Phytochemistry. J.T. Romeo,
K.R. Downum, and R. Verpoorte, eds (New York: Plenum Press),
pp. 179–205.
Ziegler, J., Hamberg, M., Miersch, O., and Parthier, B. (1997).
Purification and characterization of allene oxide cyclase from dry
corn seeds. Plant Physiol. 114, 565–573.
Ziegler, J., Stenzel, I., Hause, B., Maucher, H., Hamberg, M., Grimm,
R., Ganal, M., and Wasternack, C. (2000). Molecular cloning of
allene oxide cyclase. The enzyme establishing the stereochemistry of
octadecanoids and jasmonates. J. Biol. Chem. 275, 19132–19138.
Ziegler, J., Wasternack, C., and Hamberg, M. (1999). On the specificity of allene oxide cyclase. Lipids 34, 1005–1015.
The Crystal Structure of Arabidopsis thaliana Allene Oxide Cyclase: Insights into the Oxylipin
Cyclization Reaction
Eckhard Hofmann, Philipp Zerbe and Florian Schaller
Plant Cell; originally published online November 3, 2006;
DOI 10.1105/tpc.106.043984
This information is current as of June 18, 2017
Supplemental Data
/content/suppl/2006/11/03/tpc.106.043984.DC1.html
Permissions
https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X
eTOCs
Sign up for eTOCs at:
http://www.plantcell.org/cgi/alerts/ctmain
CiteTrack Alerts
Sign up for CiteTrack Alerts at:
http://www.plantcell.org/cgi/alerts/ctmain
Subscription Information
Subscription Information for The Plant Cell and Plant Physiology is available at:
http://www.aspb.org/publications/subscriptions.cfm
© American Society of Plant Biologists
ADVANCING THE SCIENCE OF PLANT BIOLOGY