Identification of an amino acid determinant of pH regiospecificity in a seed lipoxygenase from Momordica charantia

Article abstract of DOI:10.1016/j.phytochem.2008.09.006

Lipoxygenases (LOX) form a heterogeneous family of lipid peroxidizing enzymes, which catalyze specific dioxygenation of polyunsaturated fatty acids. According to their positional specificity of linoleic acid oxygenation plant LOX have been classified into linoleate 9- and linoleate 13-LOX and recent reports identified a critical valine at the active site of 9-LOX. In contrast, more bulky phenylalanine or histidine residues were found at this position in 13-LOX. We have recently cloned a LOX-isoform from Momordica charantia and multiple amino acid alignments indicated the existence of a glutamine (Gln599) at the position were 13-LOX usually carry histidine or phenylalanine residues. Analyzing the pH-dependence of the positional specificity of linoleic acid oxygenation we observed that at pH-values higher than 7.5 this enzyme constitutes a linoleate 13-LOX whereas at lower pH, 9-H(P)ODE was the major reaction product. Site-directed mutagenesis of glutamine 599 to histidine (Gln599His) converted the enzyme to a pure 13-LOX. These data confirm previous observation suggesting that reaction specificity of certain LOX-isoforms is not an absolute enzyme property but may be impacted by reaction conditions such as pH of the reaction mixture. We extended this concept by identifying glutamine 599 as sequence determinant for such pH-dependence of the reaction specificity. Although the biological relevance for this alteration switch remains to be investigated it is of particular interest that it occurs at near physiological conditions in the pH-range between 7 and 8.

Full text of DOI:10.1016/j.phytochem.2008.09.006

Phytochemistry 69 (2008) 2774–2780
Contents lists available at ScienceDirect
Phytochemistry
journal homepage: www.elsevier.com/locate/phytochem
Identification of an amino acid determinant of pH regiospecificity in a seed
lipoxygenase from Momordica charantia q
Ellen Hornung a,b, Susan Kunze a,b, Alena Liavonchanka a, Grit Zimmermann b, Diana Kühn b,
Kathrin Fritsche b, Andreas Renz c, Hartmut Kühn d, Ivo Feussner a,b,
*
a Georg-August-University Göttingen, Albrecht-von-Haller Institute for Plant Sciences, Department for Plant Biochemistry, Justus-von-Liebig-Weg 11, D-37077 Göttingen, Germany
b Institute for Plant Genetics and Crop Plant Research (IPK), Corrensstraße 3, D-06466 Gatersleben, Germany
c BASF Plant Science GmbH, BPS HO1 – LI444, D-67117 Limburgerhof, Germany
d Humboldt-University Berlin, Universitätsklinikum Charité, Institute of Biochemistry, Monbijoustraße 2, D-10117 Berlin, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 23 January 2008
Received in revised form 1 September 2008
Available online 20 October 2008
Lipoxygenases (LOX) form a heterogeneous family of lipid peroxidizing enzymes, which catalyze specific
dioxygenation of polyunsaturated fatty acids. According to their positional specificity of linoleic acid oxy-
genation plant LOX have been classified into linoleate 9- and linoleate 13-LOX and recent reports iden-
tified a critical valine at the active site of 9-LOX. In contrast, more bulky phenylalanine or histidine
residues were found at this position in 13-LOX. We have recently cloned a LOX-isoform from Momordica
charantia and multiple amino acid alignments indicated the existence of a glutamine (Gln599) at the
position were 13-LOX usually carry histidine or phenylalanine residues. Analyzing the pH-dependence
of the positional specificity of linoleic acid oxygenation we observed that at pH-values higher than 7.5
this enzyme constitutes a linoleate 13-LOX whereas at lower pH, 9-H(P)ODE was the major reaction
product. Site-directed mutagenesis of glutamine 599 to histidine (Gln599His) converted the enzyme to
a pure 13-LOX. These data confirm previous observation suggesting that reaction specificity of certain
LOX-isoforms is not an absolute enzyme property but may be impacted by reaction conditions such as
pH of the reaction mixture. We extended this concept by identifying glutamine 599 as sequence deter-
minant for such pH-dependence of the reaction specificity. Although the biological relevance for this
alteration switch remains to be investigated it is of particular interest that it occurs at near physiological
conditions in the pH-range between 7 and 8.
Keywords:
Dioxygenase
Lipid peroxidation
Oxylipins
Positional specificity
Substrate alignment
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
are classified according to their positional specificity of arachidonic
acid (AA) oxygenation (Kühn et al., 2005). Since AA is either not
Lipoxgenases (linoleate:oxygen oxidoreductase; EC 1.13.11.12;
LOX) are widely distributed in plant and animal kingdoms (Brash,
1999). They constitute a family of non-heme iron containing diox-
ygenases, which catalyze the regio- and stereoselective oxygena-
tion of polyenoic fatty acids containing (1Z,4Z)-pentadiene
systems (Liavonchanka and Feussner, 2006). In mammals, LOX
present in higher plants or is only a minor constituent of cellular
lipids, plant LOX may be classified according to their positional
specificity of linoleic acid (LA) oxygenation into 9- and 13-LOX
(Feussner and Wasternack, 2002). A more comprehensive classifi-
cation of plant LOX has been proposed based on the comparison
of their primary structures (Shibata et al., 1994).
LOX are versatile enzymes catalyzing at least three different
types of reactions: (i) Dioxygenation of lipid substrates (dioxygen-
ase reaction), (ii) secondary conversion of hydroperoxy lipids
(hydroperoxidase reaction), and (iii) formation of epoxy leukotri-
enes (leukotriene synthase reaction) (Feussner and Kühn, 2000).
The mechanism of the dioxygenation reaction is rather complex
and involves the formation of enzyme-bound radical intermediates
(Kühn et al., 2005). The reaction is initiated by hydrogen abstrac-
Abbreviations: AA, arachidonic acid; CP-HPLC, chiral phase-HPLC; RP-HPLC,
reverse phase-HPLC; SP-HPLC, straight-phase-HPLC; hydroperoxy arachidonic acid,
HPETE; 13-H(P)OD, (13,9Z,11E)-13-hydro(pero)xy-9,11-octadecadienoic acid; 9-
H(P)OD, (9,10E,12Z)-9-hydro(pero)xy-10,12-octadecadienoic acid; LA, linoleic acid;
LnA, a-linolenic acid; LOX, lipoxygenase(s).
q
Sequence data: The nucleotide sequence reported in this paper has been
submitted to the GenBankTM with Accession No. AM930395.
* Corresponding author. Address: Georg-August-University Göttingen, Albrecht-
von-Haller Institute for Plant Sciences, Department for Plant Biochemistry, Justus-
von-Liebig-Weg 11, D-37077 Göttingen, Germany. Tel.: +49 551 39 5743; fax: +49
551 39 5749.
tion from
a bisallylic methylene (Hamberg and Samuelsson,
1967). The resulting carbon-centred radical intermediate is subse-
quently rearranged ([+2] or [ꢀ2] rearrangement) involving double
bond conjugation (Fig. 1). For the time being there is no unifying
E-mail address: ifeussn@gwdg.de (I. Feussner).
0031-9422/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phytochem.2008.09.006

E. Hornung et al. / Phytochemistry 69 (2008) 2774–2780
2775
H
H
A
COOH
COOH
Hydrogen abstraction
Radical rearrangement
H
C
[+2]
[-2]
3
COOH
COOH
3
COOH
C
C
Oxygen insertion
3
3
COOH
HOO
OOH
B
C
13-LOX
9-LOX
OOH
OOH
Fe
Fe
COO-
-OOC
+
+
H/F
V
Gardner,1989
9-LOX
13-LOX
OOH
OOH
Fe
Fe
-
COO-
COO
H/F
V
Browner et al., 1998
Fig. 1. LOX reaction mechanism and regio-specificity. Panel (A) specificity of LOX reaction with substrates containing one doubly allylic methylene. In this case, the specificity
of the LOX reaction solely depends on the direction of the radical rearrangement. [+2] Radical rearrangement indicates that dioxygen is inserted at the second carbon atom in
the direction of the methyl terminus of the substrate counted from the site of hydrogen removal. [–2] Indicates an inverse direction of radical rearrangement. Panels (B) and
(C) Alternative explanations for the regio-specificity of LOX-isoforms: (B) model of inverse substrate orientation; and (C) space-related hypothesis of positional specificity.
concept explaining the structural basis for the reaction specificity
of all LOX-isoforms. However, a number of critical amino acids
have been identified at the active site of selected LOX-isoforms,
which impact this enzyme property (Schneider et al., 2007). For
mammalian LOX two alternative hypotheses have been developed
to explain the reaction specificity. The space-related hypothesis
(Browner et al., 1998) suggests that fatty acid substrate penetrate
the active site with its methyl end ahead and that the volume of
the substrate-binding pocket is important for the positional speci-
ficity. According to the orientation hypothesis substrate fatty acids
may slide into the active site with its methyl end ahead leading to
linoleate 13-lipoxygenation ([+2] radical rearrangement). How-
ever, when the substrate is forced into an inverse ‘‘head-to tail” ori-
entation (Gardner, 1989) linoleate 9-lipoxygenation becomes
plausible. For the cucumber lipid body 13-LOX (Hornung et al.,
1999) a critical histidine residue was identified at the bottom of
the substrate-binding pocket, which corresponds in amino acids
alignment to a previously identified primary determinant for the
positional specificity of the rabbit arachidonate 12/15-LOX (Sloane
et al., 1991). Its replacement by less space filling residues altered
the positional specificity from a linoleate 13-LOX to a 9-LOX by
We recently cloned a LOX-isoform from Mormordica charantia
seeds and found that it carries a glutamine at the position where
other plant LOX harbour either histidine/phenylalanine (linoleate
13-LOX) or valine (linoleate 9-LOX) residues. Analyzing the reac-
tion selectivity of this enzyme we found that it exhibits a variable
positional specificity of linoleic acid oxygenation depending on the
pH of the reaction mixture. Below pH 7.5; (9S,10E,12Z)-9-hydro-
(pero)xy-10,12-octadecadienoic acid ((9S)-H(P)OD) was the major
reaction product whereas at higher pH (13S,9Z,11E)-13-hydro-
(pero)xy-9,11-octadecadienoic acid ((13S)-H(P)OD) was dominant.
The structural basis for this variable reaction specificity was
explored by site directed mutagenesis studies.
2. Results
2.1. Cloning of M. charantia seed LOX
To clone an unknown LOX-isoform from M. charantia an EST-li-
brary from developing seeds was analyzed. Database searches and
nucleotide alignments revealed a cDNA of 1673 bp with significant
similarity to other plant LOX. The deduced amino acid sequence
exhibited highest degree of similarity to the lipid body LOX from
cucumber (Accession No. X92890). To create a full-length cDNA
clone, 50-RACE was performed and the RACE-fragment was fused
with the EST. The complete cDNA had a length of 2930 bp. The
open reading frame named MoLOX1 (Accession No. AM930395)
demasking
a
positively charged guanidino group. These
volume-dependent charge alterations were suggested to favour
‘‘head-to tail’’ substrate orientation (Hornung et al., 1999) and
thus, for this enzyme a combination of the two hypotheses may
be applicable.

2776
E. Hornung et al. / Phytochemistry 69 (2008) 2774–2780
Table 1
had a size of 2724 bp encoding a protein of 881 amino acids with a
theoretical molecular weight of 99.5 kDa. The sequence showed
the highest degree of similarity to cucumber lipid body linoleate
13-LOX (82%). Additional sequence analysis revealed that MoLOX1
does not contain a putative chloroplast transit peptide and that the
putative iron ligands (three histidines, one asparagine and the C-
terminal isoleucine) are well conserved (Liavonchanka and Feuss-
ner, 2006). Inspection of active site residues critical for the regio-
specificity indicated that this enzyme harbours a glutamine at
the position 599 where phenylalanine or histidine are found in
other plant 13-LOX. In linoleate 9-LOX a valine residue is located
at this position (Table 1). Moreover the enzyme harbours an ala-
nine at the position 584 suggesting an S-LOX (Coffa and Brash,
2004).
Alignment of amino acid residues possibly determining the positional specificity of
plant LOX according to Liavonchanka and Feussner (2006)
Enzyme
Accession No.
Position of amino
acid residues
Amino acid
residues
13-LOX:
Cucumber lipid body LOX
Soybean seed LOX-1
Potato LOX-H1
Arabidopsis LOX-4
Momordica charantia
X92890
P08170
X96405
CAC19364
AM930395
596/597
556/557
614/615
642/643
598/599
Thr/His
Thr/Phe
Ser/Phe
Cys/Phe
Thr/Gln
9-LOX:
Potato tuber LOX
Tobacco elicitor-Induced LOX
Barley grain LOX-A
Arabidopsis LOX-1
P37831
X84040
L35931
Q06327
575/576
580/581
574/575
576/577
Thr/Val
Thr/Val
Thr/Val
Thr/Val
2.2. Enzymatic characterization of M. charantia seed LOX
For enzymatic characterization MoLOX1 was expressed as his-
tagged fusion protein in E. coli. Under standard assay conditions
(0.9 mM fatty acid substrate, room temperature, 30 min incubation
period, pH 8.0) AA was almost exclusively converted to (15S)-
H(P)ETE (hydroperoxy arachidonic acid) and (13S)-H(P)OD was
identified as major product of LA oxygenation (Fig. 2). These data
identified MoLOX1 as linoleate 13-LOX and arachidonate 15-LOX.
The enzyme exhibits a broad substrate specificity metabolizing
various C18 and C20 fatty acids. Substrate preference was deter-
mined by incubating the enzyme for 30 min at pH 8.0 with an equi-
1600
S
1400
molar mixture of LA, AA and a-linolenic acid (LnA). Afterwards the
reaction was stopped and product formation was quantified. Under
these conditions AA oxygenation was twofold higher than the for-
mation of C18-derived products.
R
1200
1000
800
Next, we recorded the pH-profile of linoleic acid oxygenation
and explored the impact of pH-alterations on the reaction specific-
ity. From Fig. 3A it can be seen that the global activity optimum
was around pH 8. In addition, there were two putative local pH op-
tima below pH 7.5 but owing the large errors ranges the activity
differences were not significant. It should be stressed at this point
that except from the pH all other reaction parameters were kept
constant. When the reaction specificity of linoleic acid oxygenation
was analyzed over a broad pH-range, we found that at pH < 7.4
MoLOX1 oxygenated LA preferentially at carbon 9 of the fatty acid
chain. In contrast, above this pH 13-H(P)OD was the major oxygen-
ation product (Fig. 3A). Thus, there was a clear pH-dependence in
the positional specificity of linoleic acid oxygenation and this alter-
ation was observed in the near physiological range. Chiral HPLC of
the major positional isomers indicated that 13-H(P)OD was prefer-
entially formed as S enantiomer (<85%) and that the S/R-ratio did
hardly vary over the entire pH-range (Fig. 3B). In contrast, the
enantiomers composition of 9-H(P)OD was pH-dependent. Below
pH 7.5 a strong preponderance of the S enantiomer was observed.
However, at higher pH-values the relative share of (9R)-H(P)OD be-
came increasingly abundant reaching a 1:1 ratio at pH 9 (Fig. 3C).
Taken together these results indicate that MoLOX1 constitutes a
seed LOX that exhibits a variable reaction specificity depending
on the pH of the reaction mixture. Similar enzymes have been
characterized before in peas, soybean and corn seed as well as in
potato tubers (Egmond et al., 1972; Gardner, 1989; Hughes et al.,
2001a,b).
S
R
600
400
200
0
25
0
5
10
15
20
retentiontime (min)
100
80
60
40
20
0
S
R
S
R
2.3. Site-directed mutagenesis studies affecting regioselectivity
0
5
10
15
retention time (min)
As indicated in Table 1 Gln599 of MoLOX1 aligns with amino
acids of other LOX, which have previously been identified as se-
quence determinants for the positional specificity. To explore
whether this amino acid may be involved in pH-dependent altera-
tions of the positional specificity we mutated this residue to phen-
ylalanine or histidine, which are normally found at this position in
linoleic 13-LOX. The positional specificity of the resulting mutants
was compared with that of the wild-type enzyme over a pH-range
Fig. 2. HPLC analysis of hydroxy fatty acids formed from AA and LA or AA by
MoLOX1. MoLOX was incubated with 0.9 mM LA or AA at room temperature for
30 min. After reduction of lipids with sodium borohydride, the reaction mixture
was acidified and the lipids were extracted. Oxygenated fatty acid derivatives were
isolated by RP-HPLC, and positional isomers were analyzed by SP-HPLC. Ratios of S
and R were determined by CP-HPLC (Insets). In the upper panel the right inset
shows the S/R-ratio for 5-H(P)ETE being representative for all others except the one
of 15-H(P)ETE which is shown as left inset.

E. Hornung et al. / Phytochemistry 69 (2008) 2774–2780
2777
3. Discussion
Composition of positional isomers
A
80
70
60
50
40
30
20
10
0
The structural basis for the high reaction specificity of most
LOX-isoforms has been studied for many years (Veldink and Vlie-
genthart, 1991; Kühn et al., 2005; Schneider et al., 2007), but still
many aspects remain unclear. For various LOX-isoforms sequence
determinants for the reaction specificity have been characterized
but there is no unifying concept for all LOX variants. In principle,
the LOX reaction consists of four elementary reactions (hydrogen
abstraction, radical rearrangement, oxygen insertion, peroxide
reduction) and during LOX-catalyzed fatty acid oxygenation the
stereochemistry of the elementary reactions is tightly controlled.
A major element for the positional specificity of initial hydrogen
abstraction is the distance of the bisallylic methylene from the
reactive non-heme iron centre. The stereochemistry of oxygen
insertion is far more complex and various control elements have
recently been reviewed (Kühn et al., 2005; Schneider et al.,
2007). (i) Shielding: This concept suggests that fatty acid sub-
strates are bound at the active site of LOX in such a way that only
one of reactive carbon atoms of substrate molecules may be acces-
sible for hydrogen abstraction and oxygen insertion. Reactivity of
other potential candidates is impaired by interaction with the en-
zyme molecule. (ii) Oxygen channelling: Atmospheric dioxygen
may selectively be transported from the reaction buffer to the site
of oxygen insertion via preformed oxygen access channels and
thus, the position of oxygen insertion is predetermined by the pro-
tein structure. Mutagenesis studies and molecular dynamic simu-
lations on the arachidonic acid 15-LOX from soybeans (Knapp
and Klinman, 2003) and rabbit reticulocytes (Saam et al., 2007)
supported the concept of targeted intra-enzyme oxygen move-
ment. It should, however, been stressed that for the rabbit enzyme
the oxygen access channel does not constitute a continuously open
channel but rather a chain of cavities which are temporarily inter-
connected by movement of amino acid site chains (Saam et al.,
2007). (iii) Peroxy radical trapping: This concept is based on the
idea that oxygen addition at the carbon-centred radical forming
the oxygen centred peroxy radical is a reversible reaction and oc-
curs randomly during LOX catalysis. Formation of a distinct prod-
uct depends on selective reduction of a certain peroxyl radical to
the corresponding hydroperoxide. This mechanism requires
favourable positioning of an electron donor. (iv) Substrate twist-
ing: In non-enzymatic lipid peroxidation the carbon-centred pen-
tadienyl radical adopts a planar configuration, which allows
oxygenation at multiple positions and chirality. At the active site
of an enzyme the pentadienyl radical may be twisted, the radical
electron is delocalized and oxygenation takes place at a preferred
position and with selective chirality.
13- H( P) OD
9- H( P) OD
5
6
7
8
9
pHvalues
Enantiomers composition of 13-H(P)OD
B 100
80
60
40
20
0
S in (%)
R in (%)
5
6
7
8
9
pHvalues
Enantiomers composition of 9-H(P)OD
C 100
80
60
40
20
0
S in (%)
R in (%)
5
6
7
8
9
pHvalues
Fig. 3. pH-dependence of the reaction specificity of linoleic acid oxygenation
catalyzed by recombinant MoLOX. MoLOX was incubated with 0.9 mM of LA at
room temperature for 30 min at variable pH-values. The reaction was terminated by
addition of a molar excess of sodium borohydride, which reduces the hydroperoxy
fatty acids formed to the corresponding alcohols. After acidification, total lipids
were extracted and the oxygenated fatty acid derivatives were purified by
preparative RP-HPLC. The composition of the positional isomers was subsequently
analyzed by SP-HPLC (panel A, solid triangles – 13-H(P)OD, empty squares – 9-
H(P)OD). The enantiomers composition of both, 13-H(P)OD (panel B, solid triangles
S-isomer, empty triangles – R-isomer) and 9-H(P)OD (panel C, solid squares – S-
isomer, empty squares – R-isomer) was analyzed by chiral phase-HPLC as indicated
in Section 4. The figure represents the mean value and error range of two
experiments. The error bars represent repetitions of the experiments with a single
enzyme preparation. Maximal product formation (pH 8.0) was set to 100%. This
Since positional specificity is an important enzymatic parame-
ter, which should be determined for all LOX-isoforms, we first
characterized this enzyme property for the novel MoLOX1 under
standard conditions (Fig. 2). Next, we recorded the pH-profile of
linoleic acid oxygenation and found that pH-alterations impacted
the positional specificity of linoleate oxygenation (Fig. 3). Interest-
ingly, these changes occurred in a narrow pH-range at near physi-
ological conditions. To explore the mechanistic basis of this
phenomenon we performed site-directed mutagenesis studies
and found that Gln599His and Gln599Phe mutants exhibited re-
duced pH sensitivity (Fig. 4). In fact, the Gln599His mutant oxygen-
ated LA mainly to 13-H(P)ODE independent of the pH of the
reaction mixture, which contrasts the pH-dependent alterations
of the wild-type enzyme. The mechanistic basis for these effects re-
mains to be investigated. In principle, pH changes impact the acid/
base properties of both, fatty acid substrates and amino acid site
chains. However, previous studies suggested that in the water bulk
phase the vast majority of fatty acid molecules are present as
relative level corresponded to the formation of about 10
sample.
lM H(P)OD per incubation
from 5 to 9.5. As indicated in Fig. 4A the wild-type enzyme was a
linoleate 13-LOX above pH 7.5 but below this value it functioned
as a linoleate 9-LOX. The Q599F-mutant lost its 13-LOX activity
at higher pH-values (Fig. 4B). In fact, at pH-values above 7.5 a
rather unspecific product pattern (1:1 ratio of 13-H(P)ODE and 9-
H(P)ODE) was observed with this enzyme mutant. The unspecific
oxygenation reaction was also indicated by the R/S-ratio (4:6) of
the major reaction products. In contrast, the Q599H-mutation con-
verted the enzyme to a linoleate 13-LOX lacking any pH sensitivity
(Fig. 4C). The pH optimum remained unaffected, and the major
product isomers were in the S-configuration.

2778
E. Hornung et al. / Phytochemistry 69 (2008) 2774–2780
WT
Q599F
A 100
B100
80
60
40
20
0
80
60
40
20
0
8
5
6
7
9
9
5
6
8
7
pH values
pH values
Q599H
C 100
80
60
40
20
0
8
5
6
7
9
pH values
Fig. 4. Reaction specificity of various MoLOX1 mutants using LA as substrate. Wildtype and mutant MoLOX1 species expressed in E. coli were incubated with 0.9 mM of LA at
room temperature for 30 min at variable pH-values. After reduction of the primary reaction products with sodium borohydride, the reaction mixture was acidified and the
lipids were extracted. Oxygenated fatty acid derivatives were isolated by RP-HPLC and the mixture of positional isomers was analyzed by SP-HPLC (13-H(P)OD, black
triangles; 9-H(P)OD, white squares). The sum of 9- and 13- H(P)OD formed was set to 100% for each LOX species. (A) wild-type, (B) Q599F, and (C) Q599H. Panel (A) represents
two independent experiments, all other panels three of them. Repetitions were performed with aliquots from a single enzyme preparation.
uncharged molecules owing to the high pKa of the fatty acid car-
boxylate in aqueous solutions (Kanicky and Shah, 2002, 2003).
These data suggest that the degree of dissociation of the fatty acids
in the reaction mixture (water bulk phase) is only slightly altered
between pH 7 and 9. Thus, alterations in the acid/base properties
of the amino acid site chains appear to be more important. It
should, however, been stressed that the physico-chemical state
of the fatty acid molecules in the water bulk phase may not neces-
sarily mirror the situation at the active site. If basic residues are
accessible at the active site they may act as proton acceptors to
force dissociation of the carboxylate regardless of the situation in
the water bulk phase. On the other hand, when acidic residues
are present the substrate’s carboxylate may be protonated and po-
tential salt bridges, which might contribute to substrate-binding,
are broken. It should, however, been stressed that it is rather diffi-
cult to quantify the association/dissociation behaviour of a fatty
acid molecule inside the active site of an enzyme and thus, the dis-
cussion on the physico-chemical state of the substrate molecules is
rather speculative.
If pH-alterations do not impact the physico-chemical state of
the substrate, pH-dependent changes in the structure of the ac-
tive site are more likely the reasons for the alterations in reaction
specificity. Gln599 carries a non-dissociable side chain and thus,
this amino acid itself may not be sensitive for pH-alterations.
However, our studies clearly indicated that mutation of Gln599
to a more bulky amino acid induced reduction of pH sensitivity.
To explain these data one might suggest that Gln599 modifies
the interaction of an acid/base sensitive residue at the active site
with the substrate’s carboxylate. When we introduced a more
bulky site chain (phenylalanine, histidine) at position 599 this
acid/base residue may be shielded so that the interaction with
the fatty acid is weakened. To identify such an acid/base sensitive
amino acid we searched the immediate surrounding of Gln599 in
the 3D-model of MoLOX1, which was structured on the basis of
the X-ray coordinates for the soybean-LOX1, and identified as
candidate amino acid Gln537. This amino acid residue was mu-
tated into leucine, isoleucine, asparagine or glutamic acid. How-
ever, all mutants turned out to be inactive (data not shown). In
addition to these spatial effects local charge alterations need to
be considered for the Gln599His mutant. In contrast to phenylal-
anine, which is insensitive for pH-alterations, histidine is an ami-
no acid that changes its degree of dissociation depending on pH.
At alkaline pH-values the imidazole site chain delivers a proton
losing its positive charge. This alteration may favour a ‘‘head-to
head” substrate orientation (Fig. 1B), in which the fatty acid en-
ters the active site with its methyl end ahead, which is consistent
with our experimental data.
As indicated above a key problem for the understanding of the
reaction specificity of LOX is the alignment of fatty acid substrates
and reaction intermediates at the active site. Unfortunately, for the
time being there is little direct experimental information on this
interaction. In the absence of any structural data the first concepts
for the regiocontrol of the LOX reaction were developed on the
basis of the product structure and our knowledge of the stereo-
chemistry of hydrogen abstraction (Egmond et al., 1972; Gardner,
1989; Veldink et al., 1972). Later on, when X-ray coordinates and
mutagenesis data became available these models were refined
and modified (Kühn et al., 2005; Schneider et al., 2007). Unfortu-
nately, the current models are solely based on static experimental
data and do not consider the high degree of motional flexibility of
both, fatty acid substrates and LOX. Recent molecular dynamics
simulations (Saam et al., 2007), small angle X-ray scattering
measurements (Hammel et al., 2004; Daniese et al., 2005) and
fluorescence life time studies (Mei et al., 2008) have visualized
the conformational flexibility but exact quantification is still pend-
ing. However, we now know that movement of amino acid side
chains induces frequent opening and closure of preformed cavities
in the 3D-structure of LOX and that there are considerable differ-
ences in the motional flexibility of LOX-isoforms. For interpretation
of future experiments dynamic feature of protein–substrate
interaction should increasingly been considered.
4. Experimental
4.1. Materials
The chemicals used were from the following sources: Standards
of chiral and racemic hydroxy fatty acids from Chayman Chem.
(Ann Arbor, MI, USA) methanol, hexane, 2-propanol (all HPLC

E. Hornung et al. / Phytochemistry 69 (2008) 2774–2780
2779
grade) from Baker (Griesheim, Germany). Restriction enzymes
were purchased from MBI Fermentas (St. Leon-Rot, Germany).
transformed in E. coli strain SG13009. Expression was performed
by growing cells at 37 °C to an OD600 of 0.6, chilling the cells on
ice, inducing the cells with 1 mM IPTG and cultivating them at
10 °C for 50–60 h. Cells were harvested by centrifugation at 4 °C
at 3000g, cells from 100 ml culture were resuspended in 10 ml lysis
buffer (50 mM Tris–HCl, pH 7.5, 500 mM NaCl, 10% (v/v) glycerol,
0.1% (v/v) Tween 20) and lysed by sonification. The lysate was
incubated with fatty acids for 30 min at RT, the resulting hydroper-
oxides were reduced with sodium borohydride, acidified with
glacial acetic acid to pH 3 and lipids were extracted as described
(Bligh and Dyer, 1959). Analysis of the fatty acids was performed
on a HPLC.
4.2. Plant growth
M. charantia was grown from seeds in a greenhouse under 16 h
of artificially supplemented illumination at a temperature ranging
between 22 °C and 30 °C. For all experiments developing seeds
from green fruits were used.
4.3. RNA and cDNA isolation and protein expression
For RNA isolation 20 g developing seeds were shelled, ground in
liquid nitrogen, 200 ml of extraction buffer I (100 mM Tris–HCl, pH
7.5, 25 mM EDTA, 2% (w/v) laurylsarcosyl, 4 M guanidinium thio-
cyanate, 5% (w/v) PVPP, 1% (v/v) b-mercaptoethanol) was added,
further homogenized with an Ultraturrax and the homogenate
was shaken for 10 min. After centrifugation at 3000g for 15 min
the floating solid lipid phase and the pellet were discarded and
the remaining liquid phase was extracted with an equal volume
of PCI (phenol:chloroform:isoamyl alcohol, 20:20:1). After centri-
fugation at 3000g for 10 min the hydrophilic phase was re-ex-
tracted with an equal volume of chloroform and the
centrifugation step performed as before. The hydrophilic phase
was loaded on a CsCl cushion (5 M CsCl) of 8 ml and centrifuged
at 18 °C and 100,000g for 18 h. The RNA precipitate was dried,
resuspended and extracted for 15 min in a mixture consisting of
7.5 ml extraction buffer II (100 mM Tris–HCl, pH 8.8, 100 mM NaCl,
5 mM EDTA, 2% (w/v) SDS) and 10 ml PCI. Centrifugation and
washing of the hydrophilic phase with chloroform followed and
the RNA was precipitated from the hydrophilic phase by an equal
volume of 5 M LiCl overnight at 4 °C. After centrifugation for
30 min at 12,000g at 4 °C the precipitate was washed with 70% eth-
anol, dried and dissolved in 1 ml water. From this total RNA frac-
tion poly(A)+ RNA was enriched using the Poly-Attract-Kit
(Promega, Mannheim, Germany) according to the provided manual
and used for all further experiments.
4.4. Site-directed mutagenesis
Site-directed mutagenesis of pQE-MoLOX was carried out by
using the QuikChange site-directed mutagenesis kit (Stratagene,
Amsterdam, The Netherlands). The oligonucleotides used for muta-
genesis contained apart from the appropriate base changes addi-
tional conservative base exchanges which created new restriction
sites for the identification of mutants. The following oligonucleo-
tides were used: MoLOX-Q599Ha 50-AAC GCG GGT GGT CTT ATT
GAA AGT ACT CAC TTT CCG GCA AAG TAT GCT-30; MoLOX-
Q599Hb 50-AGC ATA CTT TGC CGG AAA GTG AGT ACT TTC AAT
AAG ACC ACC CGC GTT-30; MoLOX-Q599Fa 50-AAC GCG GGT GGT
CTT ATT GAA AGT ACT TTC TTT CCG GCA AAG TAT GCT-30;
MoLOX-Q599Fb 50-AGC ATA CTT TGC CGG AAA GAA AGT ACT
TTC AAT AAG ACC ACC CGC GTT-30. In addition all mutations were
sequenced and at least three different bacterial colonies were
expressed and used for analysis of enzymatic parameters.
4.5. Analytics
HPLC analysis was carried out on an Agilent (Waldbronn, Ger-
many) 1100 HPLC system coupled to a diode array detector. RP-
HPLC of the free fatty acid derivatives was carried out on a Nucleosil
C-18 column (Macherey-Nagel, Düren, FRG; 250 ꢁ 2 mm, 5
lm par-
Five microgram poly(A)+ RNA were used for the construction of
a cDNA-library in Lambda ZAPII (Stratagene, Amsterdam, The
Netherlands). The resulting library consisted of cDNA inserts
cloned directionally (50–30) in the EcoRI and XhoI sites of pBlue-
script SK-vector in Lambda ZAP II and were stored as DMSO stocks
at ꢀ80 °C until used for expressed sequence tag (EST) isolation.
EST-sequencing of 1600 clones was performed using this
cDNA-library and an EST-clone with a putative LOX-sequence
was identified. To obtain a full-length cDNA clone a 50-end RACE
was performed using the Marathon cDNA amplification kit
(Clontech, Heidelberg, Germany) and the LOX-specific primer A:
50-CAC TTG AAA CAC AGA CTC TCC AGC G-30. The PCR reaction
was carried out according to the supplier’s instruction. The RACE-
fragment was clone in pGEM-T Easy (Promega, Mannheim, Ger-
many) and sequenced. To isolate the full length cDNA clone by
PCR, specific primers of the expected open reading frame contain-
ing suitable recognition sites for specific restriction endonucleases
were used for amplification: sense primer B: 50-CGG GAT CCA TGT
TTG GGA TTG GGA AGA G-30, antisense primer C: 50-GTG GGT CGA
CTT ATA TGG AGA TAC TGT TGG G-30. The PCR reaction was carried
out with the Expand High Fidelity System (Roche Diagnostics,
Mannheim, Germany) using an amplification program of 2 min
denaturation at 94 °C, followed by 10 cycles of 30 s at 94 °C, 30 s
at 52 °C, 3 min at 72 °C, followed by 18 cycles of 30 s at 94 °C,
30 s at 52 °C, 3 min at 72 °C (time increment 5 s) and terminated
by 5 min extension at 72 °C. The resulting fragment was cloned
into pGEM-T Easy and sequenced. For expression in E. coli the open
reading frame was cloned into pQE30 (Qiagen, Hilden, Germany) as
a BamHI/SalI-fragment to yield pQE-MoLOX and the plasmid was
ticle size) with a solvent system of methanol/water/acetic acid (90/
10/0.1; by vol.) and a flow rate of 0.18 ml/min. The absorbances at
234 nm (conjugated diene system of the hydroxy fatty acids) and
210 nm (polyenoic fatty acids) were recorded simultaneously.
Oxygenated linoleic acid containing triacylglycerides were sepa-
rated on a Nucleosil C-18 column (Macherey-Nagel, Düren, FRG;
250 ꢁ 4 mm, 5
lm particle size) using a flow rate of 1 ml/min and
a binary gradient system; solvent A: methanol/water/acetic acid
(90/10/0.1; by vol.); solvent B: methanol/acetic acid (100/0.1, by
vol.) and the following gradient program: 10 min at 100% solvent
A, then in 20 min with a linear increase of solvent B to 100% solvent
B followed by an isocratic postrun of 50 min at 100% B. The absor-
bance at 234 nm was recorded. Straight phase-HPLC (SP-HPLC) of
hydroxy fatty acid isomers was carried out on a Phenomenex Luna
Silica column (Aschaffenburg, Germany; 50 ꢁ 4.6 mm, 3
lm parti-
cle size) with a solvent system of n-hexane/2-propanol/acetic acid
(100/1/0.1, by vol.) and a flow rate of 1 ml/min. The enantiomer
composition of the hydroxy fatty acids was analyzed by chiral
phase-HPLC on a Chiralcel OD-H column (Diacel Chem. Industries,
distributed by Merck, Darmstadt, Germany; 150 ꢁ 2.1 mm, 5
lm
particle size) with a solvent system of hexane/2-propanol/acetic
acid (100/5/0.1, by vol.) and a flow rate of 0.1 ml/min.
Acknowledgements
The authors would like to thank Andrea Nickel and Theres Rie-
mekasten for excellent technical assistance. This work was sup-
ported by the Fachagentur für Nachwachsende Rohstoffe and by
BASF Plant Science GmbH. A.L. was supported by a Christoph-

2780
E. Hornung et al. / Phytochemistry 69 (2008) 2774–2780
Hornung, E., Walther, M., Kühn, H., Feussner, I., 1999. Conversion of cucumber
linoleate 13-lipoxygenase to 9-lipoxygenating species by site-directed
Georg Lichtenberg stipend from state of Lower Saxonia and the
International Max Planck Research School Molecular Biology
(Goettingen).
a
mutagenesis. Proc. Natl. Acad. Sci. USA 96, 4192–4197.
Hughes, R.K., Lawson, D.M., Hornostaj, A.R., Fairhurst, S.A., Casey, R., 2001a.
Mutagenesis and modelling of linoleate-binding to pea seed lipoxygenase.
Eur. J. Biochem. 268, 1030–1040.
References
Hughes, R.K., West, S.I., Hornostaj, A.R., Lawson, D.M., Fairhurst, S.A., Sanchez, R.O.,
Hough, P., Robinson, B.H., Casey, R., 2001b. Probing a novel potato lipoxygenase
with dual positional specificity reveals primary determinants of substrate-
binding and requirements for a surface hydrophobic loop and has implications
for the role of lipoxygenases in tubers. Biochem. J. 353, 345–355.
Kanicky, J.R., Shah, D.O., 2002. Effect of degree, type, and position of unsaturation on
the pK(a) of long-chain fatty acids. J. Colloid Interf. Sci. 256, 201–207.
Kanicky, J.R., Shah, D.O., 2003. Effect of premicellar aggregation on the pK(a) of fatty
acid soap solutions. Langmuir 19, 2034–2038.
Bligh, E.G., Dyer, W.J., 1959.
A rapid method of total lipid extraction and
purification. Can. J. Biochem. Physiol. 37, 911–917.
Brash, A.R., 1999. Lipoxygenases: occurrence, functions, catalysis, and acquisition of
substrate. J. Biol. Chem. 274, 23679–23682.
Browner, M.F., Gillmor, S.A., Fletterick, R., 1998. Burying a charge. Nat. Struct. Biol. 5,
179.
Coffa, G., Brash, A.R., 2004.
A single active site residue directs oxygenation
stereospecificity in lipoxygenases: stereocontrol is linked to the position of
oxygenation. Proc. Natl. Acad. Sci. USA 101, 15579–15584.
Dainese, E., Sabatucci, A., van Zadelhoff, G., Angelucci, C.B., Vachette, P., Veldink,
G.A., Agro, A.F., Maccarrone, M., 2005. Structural stability of soybean
lipoxygenase-1 in solution as probed by small angle X-ray scattering. J. Mol.
Biol. 349, 143–152.
Egmond, M.R., Vliegenthart, J.F.G., Boldingh, J., 1972. Stereospecificity of the
hydrogen abstraction at carbon atom n-8 in the oxygenation of linoleic acid
by lipoxygenases from corn germs and soyabeans. Biochem. Biophys. Res.
Commun. 48, 1055–1060.
Knapp, M.J., Klinman, J.P., 2003. Kinetic studies of oxygen reactivity in soybean
lipoxygenase-1. Biochemistry 42, 11466–11475.
Kühn, H., Saam, J., Eibach, S., Holzhütter, H.-G., Ivanov, I., Walther, M., 2005. Structural
biology of mammalian lipoxygenases: enzymatic consequences of targeted
alterations of the protein structure. Biochem. Biophys. Res. Commun. 338, 93–101.
Liavonchanka, A., Feussner, I., 2006. Lipoxygenases: occurrence, functions and
catalysis. J. Plant Physiol. 163, 348–357.
Mei, G., Di Venere, A., Nicolai, E., Angelucci, C.B., Ivanov, I., Sabatucci, A., Daniese, E.,
Kühn, H., Maccarone, M., 2008. Structural properties of plant and mammalian
lipoxygenases. Temperature-dependent conformational alterations and
membrane binding ability. Biochemistry 47, 9234–9242.
Feussner, I., Kühn, H., 2000. Application of lipoxygenases and related
enzymes for the preparation of oxygenated lipids. In: Bornscheuer, U.T. (Ed.),
Enzymes in Lipid Modification. Wiley-VCH, Weinheim, Germany, pp. 309–
336.
Saam, J., Ivanov, I., Walther, M., Holzhütter, H.-G., Kühn, H., 2007. Molecular
dioxygen enters the active site of 12/15-lipoxygenase via dynamic oxygen
access channels. Proc. Natl. Acad. Sci. USA 104, 13319–13324.
Feussner, I., Wasternack, C., 2002. The lipoxygenase pathway. Annu. Rev. Plant Biol.
53, 275–297.
Gardner, H.W., 1989. Soybean lipoxygenase-1 enzymatically forms both 9(S)- and
Schneider, C., Pratt, D.A., Porter, N.A., Brash, A.R., 2007. Control of oxygenation in
lipoxygenase and cyclooxygenase catalysis. Chem. Biol. 14, 473–488.
Shibata, D., Slusarenko, A., Casey, R., Hildebrand, D., Bell, E., 1994. Lipoxygenases.
Plant Mol. Biol. Rep. 12, S41–S42.
13(S)-hydroperoxides from linoleic acid by
a pH-dependent mechanism.
Biochim. Biophys. Acta 1001, 274–281.
Sloane, D.L., Leung, R., Craik, C.S., Sigal, E., 1991. A primary determinant for
Hamberg, M., Samuelsson, B., 1967. On the specificity of the oxygenation of
unsaturated fatty acids catalyzed by soybean lipoxidase. J. Biol. Chem. 242,
5329–5335.
Hammel, M., Walther, M., Prassl, R., Kühn, H., 2004. Structural flexibility of the N-
terminal ß-barrel domain of 15-lipoxygenase-1 probed by small angle X-ray
scattering. Functional consequences for activity regulation and membrane
binding. J. Mol. Biol. 343, 917–929.
lipoxygenase positional specificity. Nature 354, 149–152.
Veldink, G., Garssen, G.J., Vliegenthart, J.F.G., Boldingh, J., 1972. Positional specificity
of corn germ lipoxygenase as a function of pH. Biochem. Biophys. Res. Commun.
47, 22–26.
Veldink, G.A., Vliegenthart, J.F.G., 1991. Substrates and products of lipoxygenase
catalysis. In: Atta-ur-Rahman (Ed.), Studies in Natural Products Chemistry, vol.
9. Elsevier Science Publishers B.V, Amsterdam, pp. 559–589.