Kinetic isotope effects in the oxidation of arachidonic acid by soybean lipoxygenase-1

Article abstract of DOI:10.1016/j.bmcl.2008.08.108

The reaction of soybean lipoxygenase-1 with linoleic acid has been extensively studied and displays very large kinetic isotope effects. In this work, substrate and solvent kinetic isotope effects as well as the viscosity dependence of the oxidation of arachidonic acid were investigated. The hydrogen atom abstraction step was rate-determining at all temperatures, but was partially masked by a viscosity-dependent step at ambient temperatures. The observed KIEs on k_cat were large (～100 at 25 °C).

Full text of DOI:10.1016/j.bmcl.2008.08.108

Bioorganic & Medicinal Chemistry Letters 18 (2008) 5959–5962
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Kinetic isotope effects in the oxidation of arachidonic acid by soybean
lipoxygenase-1
*
Cyril Jacquot, Sheng Peng, Wilfred A. van der Donk
Roger Adams Laboratory, Department of Chemistry, University of Illinois, 600 S. Mathews Avenue, Urbana, IL 61801, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The reaction of soybean lipoxygenase-1 with linoleic acid has been extensively studied and displays very
large kinetic isotope effects. In this work, substrate and solvent kinetic isotope effects as well as the vis-
cosity dependence of the oxidation of arachidonic acid were investigated. The hydrogen atom abstraction
step was rate-determining at all temperatures, but was partially masked by a viscosity-dependent step at
ambient temperatures. The observed KIEs on kcat were large (ꢀ100 at 25 °C).
Received 26 June 2008
Revised 23 August 2008
Accepted 26 August 2008
Available online 31 August 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Lipoxygenases are non-heme iron-dependent proteins that cat-
alyze the oxidation of polyunsaturated fatty acids to hydroperox-
ides.1 These enzymes occur in plants and animals, and have also
been detected in certain bacteria.2,3 Lipoxygenase products are fur-
ther elaborated by downstream enzymes to give a variety of phys-
iological regulators and pain mediators. In plants, lipoxygenases
convert fatty acids into jasmonates and aldehydes, which are in-
volved in signaling, germination and senescence.4 Soybean lipoxy-
genase-1 (sLO-1) exhibits high structural similarity with
mammalian lipoxygenases despite only sharing 25% sequence
identity.5 sLO-1 acts on polyunsaturated fatty acids in which a
1,4-diene unit is located six carbons away from the methyl termi-
than the semi-classical limit. Quantum chemical tunneling in the
rate-determining hydrogen atom transfer step, mediated by pro-
tein dynamics, has been proposed to account for these large val-
ues.15,16 Thus far, KIE studies have focused on LA due to the
availability of its deuterium-labeled analogs. Previously, our labo-
ratory reported the synthesis of site-specifically deuterium-labeled
AA analogs.17–19 With these compounds in hand, the reaction of
sLO-1 with AA was investigated, revealing a large isotope effect
of 97 for kcat and a more modest effect of 8.0 for kcat/Km, compared
to 64 and 25 for the corresponding parameters for the sLO-1-cata-
lyzed oxidation of LA under similar conditions. In addition, an unu-
sual isotope effect on substrate inhibition by AA was discovered.20
In this contribution, the temperature dependence of these isotope
effects is presented as well as studies to assess whether the hydro-
gen atom abstraction step is fully rate-limiting with AA as sub-
strate. Furthermore, the substrate inhibition by AA is shown to
be consistent with a high affinity of this fatty acid for the ferrous
form of the enzyme.
nus (x-6 fatty acids). Its natural substrate is linoleic acid (LA), a
C18 bisunsaturated fatty acid, which it converts into 13-hydro-
peroxyoctadienoic acid (13-HPODE). sLO-1 can also catalyze the
oxidation of arachidonic acid (AA), a C20 tetraunsaturated fatty
acid found in animals, to 15-hydroperoxyeicosatetraenoic acid
(15-HPETE).
Lipoxygenases abstract a hydrogen atom from the substrate to
form a radical, which then reacts with molecular oxygen.6,7 In rest-
ing lipoxygenase, the iron is in the ferrous form and the enzyme is
inactive.8 Before catalysis can occur, the iron must be converted to
the active ferric hydroxide form by autooxidized compounds. A
subsequent proton-coupled electron transfer from the substrate
to the ferric hydroxide forms an intermediate radical and a ferrous
species (Fig. 1).9,10 After stereoselective reaction of the substrate
radical with molecular oxygen, the peroxyl radical oxidizes the
iron back to the active ferric state and the peroxide product is re-
leased from the enzyme.
Previously, the unusual observation was made that substrate
inhibition was alleviated by isotopic substitution at the site of
hydrogen atom abstraction in the reaction of sLO-1 and AA.20
Based on a series of experimental observations including a more
than 10-fold decrease in the Km for oxygen with deuterium-labeled
substrate,20 a model was proposed to account for the unusual
observation. For protiated substrate, the reaction of the substrate
radical with O2 (kox) was proposed to compete with dissociation
of the radical intermediate from the enzyme (kd) (Fig. 2). The inac-
tive ferrous enzyme generated by substrate radical dissociation
could then be sequestered by AA, giving rise to the observed sub-
strate inhibition (Fig. 2). Three different explanations were sug-
gested for the finding that substrate inhibition is not observed
with LA despite KIEs of similar size. HPODE, the oxidation product
with LA, could be more effective than HPETE, the oxidation product
of AA, at reoxidizing the ferrous enzyme back to the active form.
Alternatively, AA might have a higher binding affinity for the
Lipoxygenases have drawn considerable attention due to the
large kinetic isotope effects (KIEs) exhibited in reactions with
LA.11–14 The observed values of 50–100 for kcat are much larger
* Corresponding author. Tel.: +1 217 244 5360; fax: +1 217 244 8024.
E-mail address: vddonk@uiuc.edu (W.A. van der Donk).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.08.108

5960
C. Jacquot et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5959–5962
O
O
OH
OH
O
OOH
13-HPODE
FeII OH
[O]
linoleic acid
O
OH
OH
OOH
FeIII OH
15-HPETE
arachidonic acid
R1
H
H
R1
FeIII OH
FeIII OH
HOO
R2
R2
R1
R1
FeII OH2
FeII OH2
OO
R2
R2
O2
Figure 1. Oxidation of linoleic acid and arachidonic acid to 13-HPODE and 15-HPETE, respectively, catalyzed by soybean lipoxygenase-1.
reactivation
15S-HPETE
0.2
0.15
0.1
OH
15S-HPETE
FeIII
AA
AA
OH2
FeII
kH
AA
OH2
FeII
OH2
FeII
AA-OO
substrate
inhibition
OH2
FeII
kox
O2
kd
0.05
0
dissociation of
R
R'
radical from enzyme
H
Figure 2. Proposed model for substrate inhibition of the sLO-1 reaction with AA.
The relatively slow capture of the substrate radical with O2 kcat=Km;O results in the
2
dissociation of a subset of radicals from the ferrous enzyme. AA can bind this
inactive enzyme. The oxygen capture of radical when using deuterium-labeled
substrate is more efficient and no substrate inhibition is observed.
0
0.5
1
1.5
Time (min)
2
2.5
3
Figure 3. Formation of oxidized product in the reaction of sLO-1 with LA (red), AA
(green), and a combination of the two substrates (blue). In all cases, the reaction
was initiated by addition of enzyme. The assay was performed in oxygen-saturated
sodium borate buffer (100 mM, pH 10.0) without the addition of activator. The
substrate and enzyme concentrations were 60 and 0.75 nM, respectively.
reduced enzyme than LA. Finally, kd might be larger when AA is the
substrate.20 A subsequent independent study showed that the two
hydroperoxides (HPODE and HPETE) are in fact equally efficient at
oxidizing ferrous sLO-1.21 In this work, we set out to investigate
whether AA has a higher affinity for the ferrous enzyme than LA.
Since direct binding studies are complicated by the requirement
of rigorous anaerobic conditions as well as by an allosteric binding
site on sLO-1,22–25 an indirect measurement was used to compare
the relative affinities of LA and AA for the ferrous enzyme. Lag
phases are typically observed in sLO-1 assays as the inactive fer-
rous form is oxidized to the active ferric form. We noticed that
the lag phase with HPLC-purified AA was significantly longer than
that observed with purified LA (Fig. 3), suggesting a higher affinity
of AA for the ferrous enzyme. The longer lag phase with AA is con-
sistent with a previous report,26 but the difference between the
two substrates was more pronounced in this work (see Supporting
information). To rule out that the LA sample contained an oxidized
impurity that resulted in more efficient oxidation of the ferrous
form of sLO-1, AA was first incubated with the enzyme before
the addition of LA. The rate of product formation again indicated
that the presence of AA significantly increased the observed lag
phase. The long lag times were also observed when LA and AA were
presented at the same time (Fig. 3). Hence, these observations sup-
port a higher affinity of AA than LA for the ferrous form of sLO-1.
The current work does not rule out, however, a lower affinity of
the arachidonyl radical for the reduced enzyme compared to the
linoleyl radical (i.e. larger kD for the former).

C. Jacquot et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5959–5962
5961
In the presence of 13-HPODE, the lag phase is effectively sup-
7
6
5
4
3
2
1
pressed, and therefore the temperature dependence of the KIEs
in the reaction of sLO-1 with AA and 13,13-d2-AA was determined
in the presence of this activator. The use of 13,13-d2-AA results in
both primary and secondary KIEs, but the secondary KIE is ex-
pected to be very small compared to the large primary KIEs. Fur-
thermore, the use of dideuterated substrate prevents any erosion
in the stereoselectivity of hydrogen atom abstraction as a result
of the large KIE; such a reduced stereoselectivity was previously
reported for stereospecifically singly deuterium-labeled LA.27 With
D
13,13-d2-AA, kcat decreased over the temperature range studied,
D
from 150 at 5 °C to 82 at 35 °C (Fig. 4). On the other hand, kcat
/
Km was much smaller and exhibited very little temperature depen-
dence, with the values at all temperatures close to within experi-
mental error. The Km values for 13,13-d2-AA are very small
0
(<2 lM) and have relatively large uncertainty due to the limit of
0.0032 0.0033 0.0034 0.0035 0.0036 0.0037
detection using non-competitive methods (see Supporting infor-
mation). As a result the kcat/Km presented here is an upper limit
1 / T (K-1)
D
and could be even smaller. Competitive methods may provide
more accurate values.
Figure 5. Arrhenius plot for the reaction of sLO-1 with AA (red line) and
13,13-d2-AA (blue line) between and 35 °C. The assay was performed in
oxygen-saturated sodium borate buffer (100 mM, pH 10.0) in the presence of
13-HPODE (16 M).
5
Arrhenius plots using the values for kcat provide an apparent
activation energy of 1.4 0.4 kcal/mol for the protiated AA and
4.1 0.6 kcal/mol for the dideuterated analog (Fig. 5). For compar-
ison, the corresponding energies for the reaction with LA were
1.8 0.4 kcal/mol for protiated substrate and 2.2 0.3 kcal/mol
for dideuterated substrate.15 These low energies suggest tunneling
is occurring for both isotopes in the reaction of AA, similar to pre-
vious conclusions for the reaction of LA. Due to the large errors in
the pre-exponential terms, no conclusions can be drawn from the
value of AH/AD.
The temperature dependence of solvent isotope effects (SIEs)
was investigated next to probe if any steps involving solvent or sol-
vent exchangeable positions on the enzyme were kinetically signif-
icant. However, the SIE on kcat remained close to unity over the
temperature range investigated (Fig. S1). The SIE on kcat/Km was
also temperature-independent and close to unity.
l
was varied by the addition of glucose as described previously for
similar studies for the oxidation of LA by sLO-1.13 For a completely
diffusion-controlled reaction, the dependence of the rate on viscos-
ity has a slope of 1.28 At 20 °C, the rate of oxidation of AA by sLO-1
was 56 5% diffusion-controlled (slope of 0.56, Fig. 6). The corre-
sponding values at 5 and 37 °C were 18 4% and 28 5% respec-
tively (Figures S3 and S4). This pronounced viscosity effect at
ambient temperature, which dissipates at higher and lower tem-
peratures, mirrors the behavior observed in the reaction of sLO-1
with LA.13
The kinetic behavior in the reaction of sLO-1 with AA can be
compared to that with LA. For both substrates, a very large sub-
strate KIE is observed on kcat that greatly exceeds the semi-classi-
cal limit. Furthermore, for both substrates the activation energies
are very small for both protiated and deuterated substrates. There-
fore, the enzymes appear to promote similar tunneling contribu-
tions for these fatty acids. These findings are in contrast to
human 15-lipoxygenase-1 that displays very different substrate
KIEs on kcat with LA (ꢀ40)29 compared to AA (ꢀ10).30 Similarly,
In a final set of experiments, the dependence of the reaction rate
on viscosity was determined. The kcat/Km values for the oxidation
of AA by sLO-1 (3.1 Â 107 M sÀ1 at 25 °C) approach diffusion-con-
trolled rates. If the encounter of substrate and enzyme is (partially)
rate-limiting, the rates should decrease in a predictable manner
with increasing viscosity.28 The viscosity of the reaction buffer
200
150
100
50
50
40
30
20
10
0
3.5
3
2.5
2
1.5
1
0
0.5
0.5
0
5
10 15 20 25 30 35 40
T (ºC)
1
1.5
2
2.5
3
3.5
relative viscosity
Figure 4. Temperature dependence of the KIE of the reaction of sLO-1 with AA and
Figure 6. The effect of viscosity on the reaction of sLO-1 and AA at 20 °C (red line).
The blue line indicates the behavior of a fully diffusion-controlled reaction. The
assays were performed in CHES buffer (100 mM, pH 10.0) in the presence of
D
13-d2-AA. kcat is shown in red while D(kcat/Km) is shown in blue. The assays were
performed in oxygen-saturated sodium borate buffer (100 mM, pH 10.0) in the
presence of 13-HPODE (16
lM) to activate the enzyme.
13-HPODE (16 lM) to activate the enzyme.

5962
C. Jacquot et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5959–5962
the KIEs on kcat/Km are also quite different for the reaction of sLO-
1 with AA and LA. For LA, D(kcat/Km) is large (25) at 20 °C with the
rate being limited to a large extent (48% at pH 9) by diffusion.13 The
reaction with AA is also limited largely by diffusion (56%) but a
much more noticeable masking of the KIE is observed (ꢀ8). A sol-
vent isotope sensitive step is not responsible for this more pro-
nounced decrease in the KIE on kcat/Km. We tentatively
conclude that a slower off-rate of AA results in increased commit-
ment to catalysis which results in smaller observed KIEs. The high-
er affinity of AA for the ferrous form of sLO-1 and the lower Km for
References and notes
1. Brash, A. R. J. Biol. Chem. 1999, 274, 23679.
2. Kuehn, H.; Thiele, B. J. FEBS Lett. 1999, 449, 7.
3. Porta, H.; Rocha-Sosa, M. Microbiology 2001, 147, 3199.
4. Gardner, H. W. Biochim. Biophys. Acta 1991, 1084, 221.
5. Knapp, M. J.; Klinman, J. P. Biochemistry 2003, 42, 11466.
6. McGinley, C. M.; van der Donk, W. A. Chem. Commun. 2003, 23, 2843.
7. Nagel, Z. D.; Klinman, J. P. Chem. Rev. 2006, 106, 3095.
8. Schilstra, M. J.; Veldink, G. A.; Vliegenthart, J. F. G. Biochemistry 1994, 33, 3974.
9. Lehnert, N.; Solomon, E. I. J. Biol. Inoorg. Chem. 2003, 8, 294.
10. Nelson, M. J.; Chase, D. B.; Seitz, S. P. Biochemistry 1995, 34, 6159.
11. Glickman, M. H.; Wiseman, J. S.; Klinman, J. P. J. Am. Chem. Soc. 1994, 116, 793.
12. Hwang, C.-C.; Grissom, C. B. J. Am. Chem. Soc. 1994, 116, 795.
13. Glickman, M. H.; Klinman, J. P. Biochemistry 1995, 34, 14077.
14. Lewis, E. R.; Johansen, E.; Holman, T. R. J. Am. Chem. Soc. 1999, 121, 1395.
15. Jonsson, T.; Glickman, M. H.; Sun, S.; Klinman, J. P. J. Am. Chem. Soc. 1996, 118,
10319.
16. Knapp, M. J.; Klinman, J. P. Eur. J. Biochem. 2002, 269, 3113.
17. Peng, S.; Okeley, N. M.; Tsai, A.-L.; Wu, G.; Kulmacz, R. J.; van der Donk, W. A. J.
Am. Chem. Soc. 2001, 123, 3609.
18. Peng, S.; McGinley, C. M.; van der Donk, W. A. Org. Lett. 2004, 6, 349.
19. Peng, S.; Okeley, N. M.; Tsai, A.-L.; Wu, G.; Kulmacz, R. J.; van der Donk, W. A. J.
Am. Chem. Soc. 2002, 124, 10785.
catalysis (15
lM for AA versus 39 lM for LA) are consistent with a
higher affinity of AA for the enzyme.
In summary, the reaction of sLO-1 with AA displays many sim-
ilarities in its kinetic behavior compared to that with LA. In both
cases, the hydrogen abstraction step displays high KIEs on kcat
due to tunneling contributions. The only difference in the reactions
is the lack of solvent-dependent rate-limiting steps at lower tem-
peratures with AA as substrate, a smaller KIE on kcat/Km and a high-
er affinity of the enzyme for AA.
20. Peng, S.; van der Donk, W. A. J. Am. Chem. Soc. 2003, 125, 8988.
21. Ruddat, V. C.; Mogul, R.; Chorny, I.; Chen, C.; Perrin, N.; Whitman, S.; Kenyon,
V.; Jacobson, M. P.; Bernasconi, C. F.; Holman, T. R. Biochemistry 2004, 43,
13063.
Acknowledgments
22. Wang, Z. X.; Killilea, S. D.; Srivastava, D. K. Biochemistry 1993, 32, 1500.
23. Mogul, R.; Johansen, E.; Holman, T. R. Biochemistry 2000, 39, 4801.
24. Mogul, R.; Holman, T. R. Biochemistry 2001, 40, 4391.
25. Ruddat, V. C.; Whitman, S.; Holman, T. R.; Bernasconi, C. F. Biochemistry 2003,
42, 4172.
This work was supported by a pre-doctoral fellowship to C.J.
from the American Heart Association (#0615604Z) and by the
National Institutes of Health (GM44911).
26. Smith, W. L.; Lands, W. E. M. J. Biol. Chem. 1972, 247, 1038.
27. Rickert, K. W.; Klinman, J. P. Biochemistry 1999, 38, 12218.
28. Brouwer, A. C.; Kirsch, J. F. Biochemistry 1982, 21, 1302.
29. Segraves, E. N.; Holman, T. R. Biochemistry 2003, 42, 5236.
30. Jacquot, C.; Wecksler, A. T.; McGinley, C. M.; Segraves, E. N.; Holman, T. R.; van
der Donk, W. A. Biochemistry 2008, 47, 7295.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2008.08.108.