A remarkable activity of human leukotriene A4 hydrolase (LTA4H) toward unnatural amino acids

Article abstract of DOI:10.1007/s00726-014-1694-2

Leukotriene A₄ hydrolase (LTA4H - EC 3.3.2.6) is a bifunctional zinc metalloenzyme, which processes LTA₄ through an epoxide hydrolase activity and is also able to trim one amino acid at a time from N-terminal peptidic substrates via its aminopeptidase activity. In this report, we have utilized a library of 130 individual proteinogenic and unnatural amino acid fluorogenic substrates to determine the aminopeptidase specificity of this enzyme. We have found that the best proteinogenic amino acid recognized by LTA4H is arginine. However, we have also observed several unnatural amino acids, which were significantly better in terms of cleavage rate (k _cat/K _m values). Among them, the benzyl ester of aspartic acid exhibited a k _cat/K _m value that was more than two orders of magnitude higher (1.75 × 10⁵ M^-1 s^-1) as compared to l-Arg (1.5 × 10³ M^-1 s^-1). This information can be used for design of potent inhibitors of this enzyme, but may also suggest yet undiscovered functions or specificities of LTA4H.

Full text of DOI:10.1007/s00726-014-1694-2

Amino Acids
DOI 10.1007/s00726-014-1694-2
ORIGINAL ARTICLE
A remarkable activity of human leukotriene A4 hydrolase
(LTA4H) toward unnatural amino acids
•
Anna Byzia Jesper Z. Haeggstrom
•
¨
•
Guy S. Salvesen Marcin Drag
Received: 3 September 2013 / Accepted: 6 February 2014
Ó The Author(s) 2014. This article is published with open access at Springerlink.com
Abstract Leukotriene A4 hydrolase (LTA4H––EC 3.3.2.6)
is a bifunctional zinc metalloenzyme, which processes
LTA4 through an epoxide hydrolase activity and is also
able to trim one amino acid at a time from N-terminal
peptidic substrates via its aminopeptidase activity. In this
report, we have utilized a library of 130 individual pro-
teinogenic and unnatural amino acid fluorogenic substrates
to determine the aminopeptidase specificity of this enzyme.
We have found that the best proteinogenic amino acid
recognized by LTA4H is arginine. However, we have also
observed several unnatural amino acids, which were sig-
nificantly better in terms of cleavage rate (kcat/Km values).
Among them, the benzyl ester of aspartic acid exhibited a
Keywords Aminopeptidase ꢀ Unnatural amino acid ꢀ
Fluorogenic substrate ꢀ Protease ꢀ Substrate specificity
Introduction
Leukotriene A4 hydrolase (LTA4H) is an enzyme with dual
activities. On the one hand, it has an epoxide hydrolase
activity that processes LTA4 into LTB4 to generate a
chemotactic agent for human neutrophils, eosinophils,
monocytes, and T cells (Haeggstrom and Funk 2011). On
the other hand, LTA4H is an aminopeptidase with a cata-
lytic zinc(II) bound to a canonical HEXXH-(X)18-E motif
(Orning et al. 1991). The neutrophil chemoattractant pro-
line-glycine-proline (PGP), which is a biomarker for
chronic obstructive pulmonary disease and implicated in
neutrophil persistence in pulmonary infections, was
recently identified as an endogenous peptide substrate for
LTA4H (Snelgrove et al. 2010). LTA4H can process very
efficiently tripeptides with arginine at the N terminus,
while dipeptidic, tetrapeptidic and pentapeptidic substrates
were processed with lower efficiency by more than one
order of magnitude in terms of kcat/Km value (Orning et al.
1994). Using appropriate substrates, both the epoxide
hydrolase and the aminopeptidase activity of LTA4H
k
cat/Km value that was more than two orders of magnitude
higher (1.75 9 105 M-1 s-1) as compared to L-Arg (1.5 9
103 M-1 s-1). This information can be used for design of
potent inhibitors of this enzyme, but may also suggest yet
undiscovered functions or specificities of LTA4H.
Electronic supplementary material The online version of this
article (doi:10.1007/s00726-014-1694-2) contains supplementary
material, which is available to authorized users.
A. Byzia ꢀ M. Drag (&)
Division of Bioorganic Chemistry, Faculty of Chemistry,
Wroclaw University of Technology, Wybrzeze Wyspianskiego
27, 50-370 Wrocław, Poland
proceed with similar catalytic rate constants (kcat
/
Km & 106 M-1 s-1), suggesting that both the functions of
this enzyme are biologically important (Tholander et al.
2008). In addition, LTA4H can cleave, albeit much less
efficiently, synthetic substrates based on single amino acids
coupled with a chromophore in P10 position, with Arg
preferred over Ala (Tholander et al. 2008).
e-mail: marcin.drag@pwr.wroc.pl
¨
J. Z. Haeggstrom
Division of Chemistry 2, Department of Medical Biochemistry
and Biophysics, Karolinska Institutet, 171 77 Stockholm,
Sweden
The crystal structure of LTA4H provides insight into
possible substrate–enzyme interactions within the active
site. It was demonstrated that tripeptides are optimal
G. S. Salvesen
Program in Apoptosis and Cell Death Research, Sanford
Burnham Medical Research Institute, La Jolla, CA 92037, USA
123

A. Byzia et al.
substrates for LTA4H due to their tight binding along a
conserved GXMEN motif. Selectivity for Arg at the N
terminus of a putative substrate was explained by a narrow
S1 pocket capped with Asp-375 forming hydrogen bonds
with water molecules and the guanidine group of Arg
(Tholander et al. 2008). The bulk size of the S1 pocket
suggests that LTA4H should recognize not only substrates
with and N-terminal Arg, but also other amino acids.
However, not much is known about the full aminopeptidase
substrate specificity repertoire of LTA4H toward natural
amino acids.
preparative Waters Spherisorb S10ODS2 or analytical
Waters Spherisorb S5ODS2 columns. Preparative HPLC
was carried out with a Waters M600 solvent delivery
module equipped with a Waters M2489 Detector system
and a C18 column. Solvent composition: system A: water/
0.1 % TFA and system B: acetonitrile 80 %/water 20 %
with 0.1 % TFA. To determine turnover rates of individual
substrates a Spectra MAX Gemini EM fluorimeter
(Molecular Devices) operating in the kinetic mode in
96-well plates was used. The excitation/emission wave-
lengths were established at 355 nm/460 nm.
To gain insight into LTA4H preferences in recognition
of substrates in the P1 pocket, we used a library of amino
acids connected to a fluorophore group, 7-amino-4-carba-
moylmethylcoumarin (ACC). This approach was previ-
ously used successfully to probe substrate specificity of
several human, animal, plant and bacterial aminopeptidases
(Poreba et al. 2012a, b; Gajda et al. 2012; Veillard et al.
2012; Zervoudi et al. 2011; Drag et al. 2010; Weglarz-
Tomczak et al. 2013). In this approach, in addition to
natural amino acids (proteinogenic amino acids), we also
employed more than one hundred unnatural amino acids
(D-amino acids, L-amino acids with different unnatural side
chains). Our study uncovers a set of unexpected observa-
tions regarding the substrate specificity of LTA4H, and
predicts that this enzyme may have selectivity for post-
translationally modified N-terminal amino acids.
Enzyme isolation
Human recombinant LTA4H was expressed in E. coli and
purified to apparent homogeneity as previously described
(Rudberg et al. 2002).
Library synthesis
An ACC fluorophore with Fmoc protecting group (2.5 eq.)
was coupled to deprotected Rink Amide Resin (1 eq.)
using Hydroxybenzotriazole hydrate (HOBt––2 eq.) and
HBTU as coupling reagent in dimethylformamide (DMF)
solvent for 24 h. To increase the level of ACC substitution
to the resin, this procedure was repeated one more time
with half amount of reagents. Next, the protecting group
was removed by 20 % of piperidine/DMF for two times
5 min, and once 20 min. The resin was washed with DMF
(about 5–6 times) and three times with dichloromethane
and three times with methanol, and left in vacuum for 24 h
to be dried. The resin was divided equally into separate
wells of 48-well cartridge and swelled with dichloro-
methane for 2 h, then washed with DMF three times. The
coupling reaction of Fmoc-protected amino acids (2.5 eq.)
to ACC was carried out in the presence of HATU (2.5 eq.)
and 2,4,6-collidine (2.5 eq.) in DMF for 24 h. Also here, to
increase the level of amino acid substitution to the resin,
this procedure was repeated one more time with half
amount of reagents. Next, the resin was washed five times
with DMF and the protecting group was removed by 20 %
of piperidine/DMF for 5, 5 and 20 min. The resin was
washed with DMF (about 5–6 times), three times with
dichloromethane and three times with methanol, and left in
vacuum for 24 h. Substrate cleavage from the resin was
performed in cold 95 % TFA, 2.5 % H2O, 2.5 % triiso-
propyl silane for 1.5 h. Finally, the substrates were pre-
cipitated in diethyl ether for 30 min at 4 °C and
centrifuged. Obtained substrates were purified and ana-
lyzed using HPLC. After lyophilization compounds were
analyzed by mass spectrometry and dissolved in anhydrous
DMSO to a final concentration of 50 mM.
Materials and methods
General
All chemicals were obtained from commercial sources and
used without further purification. Rink amide (amino-
methyl) polystyrene (100–200 mesh, extent of labeling
0.46 mmol g-1 N-1 loading, 1 % cross-linked), piperidine
and trifluoroacetic acid (TFA) were from Iris Biotech
GmbH, Fmoc protected amino acids were from Iris Biotech
GmbH, Sigma Aldrich, Fluka and Novabiochem.
1-Hydroxybenzotriazole hydrate (HOBt), O-(7-azabenzo-
triazol-1-yl)-N,N,N0N0-tetramethyluronium hexafluoropho-
sphonate (HATU), N,N-diisopropylethylamine (DIPEA),
and triisopropylsilane (TIPS) were purchased from Sigma
Aldrich. 7-Fmoc-aminocoumarin-4-acetic acid (Fmoc-
ACC) was synthesized according the procedure by (Maly
et al. 2002). LC–MS data were recorded with the Shimadzu
LCMS-2010EV system at the and the Department of Bio-
technology, University of Wroclaw. Analytical high per-
formance liquid chromatography (HPLC) analysis was
performed with a Waters M600 solvent delivery module
equipped with a Waters M2489 detector system using
123

A remarkable activity of human leukotriene A4 hydrolase
Substrate library assay
Table 1 Comparison of the
catalytic efficiency (kcat/Km) of
the selected natural amino acid
based substrates
Natural
amino acid
kcat/Km
(M-1 s-1
)
A library of 130 amino acids coupled to the ACC fluoro-
phore was used to screen the substrate specificity of
LTA4H. It consists of 19 of the 20 natural amino acids (not
cysteine, which is prone to oxidation), 18 D-amino acids,
and the others are L-derivatives of unnatural amino acids.
The enzyme was assayed in 250 mM tris buffer, pH 7.5,
containing 500 mM KCl and 0.1 % of bovine serum
albumin, both of which stimulate the aminopeptidase
activity (Wetterholm and Haeggstrom 1992; Orning and
Fitzpatrick 1992). The buffer was prepared at room tem-
perature. Enzyme was incubated for 30 min at 37 °C and
added to the substrates in the wells of a 96-well plate. The
fluorescence increase (Relative Fluorescence Unit per
second) was monitored using Spectra MAX Gemini EM
fluorimeter operating in kinetic mode (Molecular Devices,
the excitation wavelength was 355 nm and emission was
460 nm). The time of each assay was 30 min, but only the
linear portion of the curve was used to calculate velocity.
The final enzyme concentration was in the range of
30–40 nM. Individual substrate concentrations were 2 lM,
which is low enough below estimated KM values to obtain
velocity values proportional to the initial velocity for each
substrate. The DMSO concentration in each assay was
below 2 %. Enzymatic hydrolysis of each substrate was
measured at least three times. Results are presented as
relative activity of each substrate (the efficiency of fluo-
rophore release) in comparison to the best one (100 %).
Ala
Arg
Leu
Lys
Met
Phe
Pro
470.6 91.5
1,496.7 102.8
343.4 57.7
617.4 80.1
297.8 92.3
747.4 104.9
449.1 41.9
Results
The substrate library
To determine the ability of LTA4H to recognize and
hydrolyze different substrates we used a library of natural
and unnatural amino acids coupled to the ACC fluorophore
used in previous studies (Drag et al. 2010; Zervoudi et al.
2011). In addition, we extended this library by several new
derivatives of unnatural amino acids using a previously
described methodology (Drag et al. 2010; Zervoudi et al.
2011). The library consists of compounds with side chains
of unnatural amino acids characterized by significant
structural diversity. Small or bulky, branched or unbran-
ched, acidic, neutral or basic, hydrophobic or hydrophilic
side chains guarantee detailed probing of the S1 pocket of
LTA4H. All structures and information about fluorogenic
substrates are in Online Resource 1. To verify the
requirements of LTA4H for substrates with unblocked a-
amino group (a vital feature for aminopeptidases) we
checked the ability of this enzyme to recognize substrates
with secondary amine derivatives (Pro, Tic, Oic) and with
modified main chain (Apns (2S,3S), b-Ala, 6-Ahx). Ste-
reoselectivity of the enzyme was investigated with D-
derivatives of selected amino acids.
Determination of kinetic parameters for best substrates
(kcat, Km, kcat/Km)
The best substrates were selected to determine kinetic
parameters of hydrolysis by LTA4H. The assay condition
was the same as described above. The ACC concentration
and corresponding Relative Fluorescence Unit was calcu-
lated by total hydrolysis of five independent substrates at
the same concentration. The average value was used for
further calculations. To determine kinetic parameters,
velocities for eight different concentrations of each sub-
strate were measured. Each experiment was repeated at
least three times and results are presented as mean SD.
The enzyme concentration was 30 nM and the substrate
concentrations for kcat/Km determination ranged from 1 to
500 lM. The total time for each experiment was between
15 and 30 min and only the linear portions of the curves
were used for calculations. For natural amino acids, with
exception of L-Arg, activity was low and did not obey
saturation kinetics within the substrate range tested, sug-
gesting increased Km and reduced kcat. Only a rough esti-
mation of kcat/Km was possible (Table 1). The final DMSO
concentration was below 2 %.
Substrate specificity screening assay
To establish the optimal condition for LTA4H catalysis we
performed a preliminary screen of the library. The best
substrate containing a natural amino acid, L-Arg-ACC, had
a Km value of 301 lM, whereas the lowest measured Km
value was for the substrate L-Bpa-ACC with an unnatural
amino acid was 2.55 lM. Subsequently, to evaluate sub-
strate specificity at conditions where substrate cleavage
velocities are proportional to kcat/Km, we performed
screening at a final substrate concentration of 1 lM, which
was more than two times lower than the lowest established
Km value. To validate this approach, we performed an
additional screen at substrate concentrations of 5 lM. This
did not affect the observed overall data, revealing that the
123

A. Byzia et al.
Fig. 1 Preferred natural amino acid substrates for LTA4H. Initial
screening of the 19-membered natural amino acid substrate library
and 18 D-amino acid substrates. Enzyme activity was monitored using
an fMax multi-well fluorescence plate reader (Molecular Devices) at
excitation wavelength of 355 nm and an emission wavelength of
460 nm. The x-axis represents the abbreviated amino acid names (for
full name and structure see Fig. S1). The y-axis represents the average
relative activity expressed as a percent of the best amino acid
substrate. All structures and information about fluorogenic substrates
are in Online Resource 1
Km values for these substrates were higher than 5 lM, thus
guaranteeing a proportional correlation between fluores-
cence signal and kcat/Km.
aminopeptidases. One common feature of all the preferred
unnatural amino acid substrates is a very large, long, non-
charged and hydrophobic character of the side chain, dem-
onstrating that the S1 pocket of this enzyme is very spacious
and able to accommodate much bigger moieties compared to
the biggest natural amino acids.
Specificity toward natural and D-amino acids
The best substrate for LTA4H with a natural amino acid
was L-Arg, which was cleaved more than two times more
rapidly compared to the second best natural amino acid L-
Ala (Fig. 1). Preferred substrates generally had hydropho-
bic or basic properties, and negligible hydrolysis of L-Gln,
L-His, L-Ser and L-Trp was observed. Extremely high ste-
reospecificity of LTA4H in the S1 pocket was demon-
strated by lack of processing of D-amino acids.
To gain better insight into the unusual activity of
LTA4H toward unnatural amino acid substrates, we mea-
sured detailed kinetic parameters (kcat, Km, and kcat/Km) for
the nine best (Table 2). The obtained data confirmed the
results from the library screening. The highest catalytic
efficiency was observed for L-AspBzl (kcat/Km = 1.75 9
105 M-1 s-1), which was more than one hundred times
higher than the value obtained for the best natural amino
acid conjugate, L-Arg. The catalytic efficiencies of the
other substrates matched well the rates obtained during
library screening. Interestingly, extension of the alkyl chain
of L-Arg by one methylene group to L-hArg dramatically
improved the processing of this substrate by LTA4H.
Similarly to other previously tested aminopeptidases,
LTA4H had a very strong preference for amino acids with
an amino group in the a position and was inactive toward
substrates with a hydroxyl group instead of an amine
(Apns) or amino acids without amino groups present in
this position (6-Ahx or b-Ala) (Drag et al. 2010; Zervoudi
et al. 2011). This confirms that the aminopeptidase
activity of LTA4H is consistent with other enzymes from
this family.
Subsequent calculation of kcat/Km parameters for the
best seven natural amino acids matched very well with
library screening data. The highest observed kcat/Km for L-
Arg was 1,496.7 M-1 s-1, while L-Phe, the next well-
cleaved substrate was about two times lower (Table 1).
Other amino acids were processed by LTA4H proportion-
ally lower. Interestingly, L-Pro is quite efficiently hydro-
lyzed by LTA4H, while in previous studies this substrate
was not recognized by aminopeptidase N (CD13), which
belongs to the same family (Drag et al. 2010; Tholander
et al. 2008).
Specificity toward substrates containing unnatural
amino acids
Finally, LTA4H was incubated with a range of previ-
ously studied general aminopeptidase phosphonate inhibi-
tors (Arg-PO3H2, Nle-PO3H2 and hPhe-PO3H2) at
concentration of 200 lM for 30 min (Drag et al. 2010;
Kannan Sivaraman et al. 2013; Weglarz-Tomczak et al.
2013). No cleavage of the library substrates in the presence
of these compounds confirms that the observed fluores-
cence signal in incubations without inhibitors is the result
of LTA4H aminopeptidase activity.
Interestingly, LTA4H demonstrated enhanced activity
toward a significant number of substrates containing
unnatural amino acids. The most striking preference was
observed for O-benzyl esters of aspartic acid (AspBzl) and
for homoarginine (hArg) (Fig. 2). The most preferred amino
acid conjugates were processed by LTA4H at rates never
observed before in direct comparison of natural and unnat-
ural amino acid preferences by previously investigated
123

A remarkable activity of human leukotriene A4 hydrolase
Fig. 2 Individual preferences
in the S1 pocket of LTA4H
toward unnatural amino acid
substrates compared to the best
natural amino acid conjugate, L-
Arg. Enzyme activity was
monitored using an fmax multi-
well fluorescence plate reader
(molecular devices) at
excitation wavelength of
355 nm and an emission
wavelength of 460 nm. The x-
axis represents the abbreviated
amino acid names. The y-axis
represents the average relative
activity expressed as a percent
of the best amino acid. All
structures and information about
fluorogenic substrates are in
Online Resource 1
Discussion
analysis of the crystal structure of LTA4H (Tholander et al.
2008). Most likely, elongation of the alkyl chain of L-Arg
results in even better alignment of the L-hArg substrate and
thus much more efficient hydrolysis.
Leukotriene A4 hydrolase/aminopeptidase is a bifunctional
zinc metalloenzyme that recently attracted considerable
attention due to its crucial role in limiting chronic pul-
monary neutrophilic inflammation (Snelgrove et al. 2010).
However, despite a recent strong interest in development of
LTA4H inhibitors, the detailed substrate specificity of this
enzyme had not previously been investigated. Especially,
intriguing was the fact that the structures published so far
of inhibitors docked to this enzyme reveal a spacious S1
pocket, suggesting that the size of this pocket is large
enough to accept molecules that are much bulkier than L-
Arg, the natural amino acid best recognized by LTA4H
(Thunnissen et al. 2001). To test this hypothesis, we sys-
tematically investigated the substrate specificity of this
enzyme using a large library of fluorogenic substrates. The
results that we obtained strongly confirmed our hypothesis.
The best substrates in our study were derivatives of
natural amino acids, L-Asp, L-Glu, L-Cys and L-Ser that
were protected with bulky and spacious groups. Interest-
ingly, these amino acids (cysteine was not tested) in their
unprotected native form were poorly cleaved (L-Ser) or not
cleaved at all (L-Asp and L-Glu) by LTA4H. A separate
example here is L-hArg, which is only one methylene group
longer than the proteinogenic amino acid L-Arg, but is
processed by LTA4H at a markedly higher rate. A likely
explanation to this may be the electronegative nature of the
bottom of the S1 pocket, with the dominant Asp-375
forming hydrogen bonds with L-Arg, as demonstrated in an
Our studies also demonstrate that substrates based on
very bulky and particularly hydrophobic unnatural amino
acids such as L-hPhe, L-hCha or L-Bip were cleaved by
LTA4H albeit at a significantly lower rate compared to the
best substrates identified in our study. This suggests that the
enzyme prefers very bulky substrates, which in structure
possesses hydrophilic atom(s) like oxygen, sulfur or nitro-
gen, to achieve optimal binding in the S1 pocket. An
example that confirms this conclusion is L-Bpa, whose side
chain has two hydrophobic phenyl rings and is quite similar
to L-Bip, but has an additional oxygen atom in the side
chain. As a result, L-Bpa is much more efficiently cleaved
compared to L-Bip. Recent studies with b-naphthylamide
substrates demonstrated also an advantage of L-Bpa over L-
Bip (Poras et al. 2013). On the other hand, benzyl protected
tyrosine is quite weakly processed by LTA4H. It seems that
despite the presence of a big hydrophobic moiety and
oxygen in the side chain structure, this derivative is already
very big and demonstrates space limitations within the S1
pocket. Interestingly, despite a high similarity with the best
cleaved L-AspBzl, cyclohexyl esters of aspartic acid (L-
Asp(OcHx) or glutamic acid (L-Glu(OcHx) are hydrolyzed
at significantly lower rates by LTA4H. This suggests that
for optimal binding in the S1 pocket a hydrophobic group is
also required, but of aromatic character, like a phenyl ring
present in the best processed substrates in our study.
123

A. Byzia et al.
Table 2 Comparison of the
catalytic efficiency (kcat/Km) of
selected unnatural amino acid
based substrates
Unnatural
amino acid
kcat/Km × 103 M-1 s-1
Structure
kcat (s-1)
Km (µM)
SerBzl
0.426 0.05
42.20 8.2
10.1 0.9
23.5 5.5
Cys(MeBzl)
Nle(6-O-Bzl)
0.309 0.04
0.599 0.25
0.530 0.04
13.16 5.2
19.94 9.4
16.43 2.9
30.02 4.51
Cys(Bzl)
32.29 4.82
GluBzl
1.133 0.07
2.754 0.2
0.579 0.02
24.18 2.1
59.76 3.6
8.86 2.2
46.88 3.54
46.07 9.51
65.29 17.01
hSerBzl
Cys(4-MeOBzl)
AspBzl
hArg
4.079 0.32
23.33 4.1
174.85 22.73
7.22 1.08 85.66 7.68
84.17 5.35
Analysis of previously published crystal structures of
LTA4H in complex with inhibitors confirms our data. For
example, (Kirkland et al. 2008) synthesized a series of
arylamide derivatives of glutamic acid, which were potent
inhibitors of LTA4H. These compounds were based on
glutamic acid, the side chain of which was protected with a
123

A remarkable activity of human leukotriene A4 hydrolase
hydrophobic phenyl ring derivatized with different sub-
stituents. In another study, (Enomoto et al. 2009) obtained
S-(4-cyclohexyl)benzyl-cysteine derivatives as potent and
selective LTA4H inhibitors, which contained a sulfhydryl
group of cysteine protected with very bulky and hydro-
phobic phenyl and additional cyclohexyl rings. Moreover,
a recent comprehensive review on LTA4H inhibitors
demonstrates several structures with features in common
with the above discussed structural elements preferred in
the S1 pocket of LTA4H (Caliskan and Banoglu 2013).
Finally, analysis of the crystal structure of LTA4H dem-
onstrates that besides the electronegative properties at the
bottom of the S1 pocket, its upper parts are more neutral
and facilitate binding of hydrophobic groups (Tholander
et al. 2008).
In summary, we have performed a detailed analysis of
the S1 pocket specificity of LTA4H aminopeptidase. Using
substrate library screening we demonstrated key features of
this enzyme, disclosing particular preferences in shape and
size of potential substrates and inhibitors, which can
facilitate rational design of small molecular-mass selective
markers that can be used in studies of LTA4H.
Acknowledgments We thank Marcin Poreba, Zofia Pilch, Paulina
Piatek, and Anders Wetterholm for help with synthesis of fluorogenic
substrates and preparation of purified LTA4H. The Drag laboratory is
supported by the State for Scientific Research Grant N N401 042,838
¨
and the Foundation for Polish Science in Poland. The Haeggstrom
laboratory is supported by the Swedish Research Council (10,350,
2011-5003). The Salvesen laboratory is supported by NIH Grants
CA163743 and GM09040.
In addition to providing information about substrate
specificity, a very intriguing observation in our study was
the dramatic increase of the catalytic efficiency of sub-
strates based on unnatural amino acids compared to those
containing natural ones. Usually, in our previous studies,
the maximal increase in efficiency was about 2–3-fold
(Drag et al. 2010; Poreba et al. 2012b; Zervoudi et al.
2011). To the best of our knowledge, a gain in catalytic
efficiency exceeding two orders of magnitude has never
been observed with synthetic substrates for any type of
exopeptidase. One could perhaps expect this result with tri-
or tetrapeptidic substrates for endoproteases, but with
conjugates of single amino acids for exopeptidases this
result is highly unusual. It provides evidence for the
potential of unnatural amino acids to be applied in substrate
activity screenings of proteolytic enzymes, but also raises
the question of the molecular basis for this effect. Recently,
we speculated that substrate specificity of proteases could
be dictated by posttranslational modifications, an issue that
has not been investigated in detail before (Kasperkiewicz
et al. 2012). Our observation of increased catalytic effi-
ciency against side-chain modified versions of L-Asp, L-
Glu, L-Cys and L-Ser seems to fit perfectly with this sce-
nario. However, it is difficult to speculate about the nature
of putative natural amino acid modifications based on data
with only single amino acids and this issue will certainly
require further studies, most probably using proteomic
methods on endogenous substrates.
Conflict of interest The authors declare that they have no conflict
of interest.
Open Access This article is distributed under the terms of the
Creative Commons Attribution License which permits any use, dis-
tribution, and reproduction in any medium, provided the original
author(s) and the source are credited.
References
Caliskan B, Banoglu E (2013) Overview of recent drug discovery
approaches for new generation leukotriene A(4) hydrolase
inhibitors. Expert Opin Drug Discov 8(1):49–63. doi:10.1517/
17460441.2013.735228
Drag M, Bogyo M, Ellman JA, Salvesen GS (2010) Aminopeptidase
fingerprints, an integrated approach for identification of good
substrates and optimal inhibitors. J Biol Chem 285(5):3310–
3318. doi:10.1074/jbc.M109.060418
Enomoto H, Morikawa Y, Miyake Y, Tsuji F, Mizuchi M, Suhara H,
Fujimura K, Horiuchi M, Ban M (2009) Synthesis and biological
evaluation of N-mercaptoacylcysteine derivatives as leukotriene
A4 hydrolase inhibitors. Bioorg Med Chem Lett 19(2):442–446.
doi:10.1016/j.bmcl.2008.11.042
Gajda AD, Pawelczak M, Drag M (2012) Substrate specificity
screening of oat (Avena sativa) seeds aminopeptidase demon-
strate unusually broad tolerance in S1 pocket. Plant Physiol
Biochem 54C:6–9. doi:10.1016/j.plaphy.2012.02.006
Haeggstrom JZ, Funk CD (2011) Lipoxygenase and leukotriene
pathways: biochemistry, biology, and roles in disease. Chem Rev
111(10):5866–5898. doi:10.1021/cr200246d
Kannan Sivaraman K, Paiardini A, Sienczyk M, Ruggeri C, Oellig
CA, Dalton JP, Scammells PJ, Drag M, McGowan S (2013)
Synthesis and Structure-Activity Relationships of Phosphonic
Arginine Mimetics as Inhibitors of the M1 and M17 Amino-
peptidases from Plasmodium falciparum. J Med Chem 56(12):
5213–5217. doi:10.1021/jm4005972
Kasperkiewicz P, Gajda AD, Drag M (2012) Current and prospective
applications of non-proteinogenic amino acids in profiling of
proteases substrate specificity. Biol Chem 393(9):843–851.
doi:10.1515/hsz-2012-0167
Kirkland TA, Adler M, Bauman JG, Chen M, Haeggstrom JZ, King
B, Kochanny MJ, Liang AM, Mendoza L, Phillips GB,
Thunnissen M, Trinh L, Whitlow M, Ye B, Ye H, Parkinson J,
Guilford WJ (2008) Synthesis of glutamic acid analogs as potent
Our study also confirms the unusual LTA4H activity
toward substrates containing L-Pro. The secondary amine at
its N terminus is normally not well recognized by enzymes
of the M1 family of metallopeptidases with the GXMEN
motif, which is responsible for interactions with the free
amine in a position (Tholander et al. 2008). However,
processing of L-Pro substrates in our studies is consistent
with the ability of LTA4H to hydrolyze the endogenous
substrate Proline-Glycine-Proline (PGP), a chemoattractant
for leukocytes (Snelgrove et al. 2010).
123

A. Byzia et al.
inhibitors of leukotriene A4 hydrolase. Bioorg Med Chem
16(9):4963–4983. doi:10.1016/j.bmc.2008.03.042
MalyDJ, Leonetti F, Backes BJ, Dauber DS, Harris JL, Craik CS, Ellman
JA (2002) Expedient solid-phase synthesis of fluorogenic protease
substrates using the 7-amino-4-carbamoylmethylcoumarin (ACC)
fluorophore. J Org Chem 67(3):910–915. doi:jo016140o
Orning L, Fitzpatrick FA (1992) Albumins activate peptide hydrolysis
by the bifunctional enzyme LTA4 hydrolase/aminopeptidase.
Biochemistry 31(17):4218–4223
Orning L, Krivi G, Fitzpatrick FA (1991) Leukotriene A4 hydrolase.
Inhibition by bestatin and intrinsic aminopeptidase activity
establish its functional resemblance to metallohydrolase enzymes.
J Biol Chem 266(3):1375–1378
Orning L, Gierse JK, Fitzpatrick FA (1994) The bifunctional enzyme
leukotriene-A4 hydrolase is an arginine aminopeptidase of high
efficiency and specificity. J Biol Chem 269(15):11269–11273
Poras H, Duquesnoy S, Fournie-Zaluski MC, Ratinaud-Giraud C,
Roques BP, Ouimet T (2013) A sensitive fluorigenic substrate
for selective in vitro and in vivo assay of Leukotriene A4
hydrolase activity. Anal Biochem. doi:10.1016/j.ab.2013.06.016
Poreba M, Gajda A, Picha J, Jiracek J, Marschner A, Klein CD,
Salvesen GS, Drag M (2012a) S1 pocket fingerprints of human
and bacterial methionine aminopeptidases determined using
fluorogenic libraries of substrates and phosphorus based inhibi-
tors. Biochimie 94(3):704–710. doi:10.1016/j.biochi.2011.10.014
Poreba M, McGowan S, Skinner-Adams TS, Trenholme KR, Gardiner
DL, Whisstock JC, To J, Salvesen GS, Dalton JP, Drag M
(2012b) Fingerprinting the Substrate Specificity of M1 and M17
Aminopeptidases of Human Malaria. PLoS One 7(2):e31938.
doi:10.1371/journal.pone.0031938
catalytic residue with specific roles in two distinct enzyme
mechanisms. J Biol Chem 277(2):1398–1404. doi:10.1074/jbc.
M106577200
Snelgrove RJ, Jackson PL, Hardison MT, Noerager BD, Kinloch A,
Gaggar A, Shastry S, Rowe SM, Shim YM, Hussell T, Blalock
JE (2010) A critical role for LTA4H in limiting chronic
pulmonary neutrophilic inflammation. Science 330(6000):
90–94. doi:10.1126/science.1190594
Tholander F, Muroya A, Roques BP, Fournie-Zaluski MC, Thunnis-
sen MMGM, Haeggstrom JZ (2008) Structure-based Dissection
of the Active Site Chemistry of Leukotriene A4 Hydrolase:
implications for M1 Aminopeptidases and Inhibitor Design.
Chem Biol 15(9):920–929. doi:10.1016/J.Chembiol.07.018
Thunnissen MM, Nordlund P, Haeggstrom JZ (2001) Crystal structure
of human leukotriene A(4) hydrolase, a bifunctional enzyme in
inflammation. Nat Struct Biol 8(2):131–135. doi:10.1038/84117
Veillard F, Potempa B, Poreba M, Drag M, Potempa J (2012)
Gingipain aminopeptidase activities in Porphyromonas gingiva-
lis. Biol Chem. doi:10.1515/hsz-2012-0222
Weglarz-Tomczak E, Poreba M, Byzia A, Berlicki L, Nocek B,
Mulligan R, Joachimiak A, Drag M, Mucha A (2013) An
integrated approach to the ligand binding specificity of Neisseria
meningitidis M1 alanine aminopeptidase by fluorogenic sub-
strate profiling, inhibitory studies and molecular modeling.
Biochimie 95(2):419–428. doi:10.1016/j.biochi.2012.10.018
Wetterholm A, Haeggstrom JZ (1992) Leukotriene A4 hydrolase: an
anion activated peptidase. Biochim Biophys Acta 1123(3):275–281
Zervoudi E, Papakyriakou A, Georgiadou D, Evnouchidou I, Gajda A,
Poreba M, Salvesen GS, Drag M, Hattori A, Swevers L,
Vourloumis D, Stratikos E (2011) Probing the S1 specificity
pocket of the aminopeptidases that generate antigenic peptides.
Biochem J 435(2):411–420. doi:10.1042/BJ20102049
Rudberg PC, Tholander F, Thunnissen MM, Haeggstrom JZ (2002)
Leukotriene A4 hydrolase/aminopeptidase. Glutamate 271 is a
123