The chemistry of escapin: Identification and quantification of the components in the complex mixture generated by an L-amino acid oxidase in the defensive secretion of the sea snail Aplysia Californica

Article abstract of DOI:10.1002/chem.200801696

Escapin is an L-amino acid oxidase in the ink of a marine snail, the sea hare Aplysia californica, which oxidizes L-lysine (1) to produce a mixture of chemicals which is antipredatory and antimicrobial. The goal of our study was to determine the identity

Full text of DOI:10.1002/chem.200801696

FULL PAPER
DOI: 10.1002/chem.200801696
The Chemistry of Escapin: Identification and Quantification of the
Components in the Complex Mixture Generated by an l-Amino Acid
Oxidase in the Defensive Secretion of the Sea Snail Aplysia californica
Michiya Kamio,[a, b, d] Ko-Chun Ko,[a, b, d] Shilong Zheng,[c] Binghe Wang,[c]
Stacy L. Collins,[c] Giovanni Gadda,[a, c] Phang C. Tai,[a, b] and Charles D. Derby*[a, b]
Abstract: Escapin is an l-amino acid
oxidase in the ink of a marine snail,
the sea hare Aplysia californica, which
oxidizes l-lysine (1) to produce a mix-
ture of chemicals which is antipredato-
ry and antimicrobial. The goal of our
study was to determine the identity
and relative abundance of the constitu-
ents of this mixture, using molecules
generated enzymatically with escapin
and also using products of organic syn-
theses. We examined this mixture
under the natural range of pH values
for ink—from ꢀ5 at full strength to
ꢀ8 when fully diluted in sea water.
The enzymatic reaction likely forms an
equilibrium mixture containing the
linear form a-keto-e-aminocaproic acid
(2), the cyclic imine D1-piperidine-2-
carboxylic acid (3), the cyclic enamine
D2-piperidine-2-carboxylic acid (4),
possibly the linear enol 6-amino-2-hy-
droxy-hex-2-enoic acid (7), the a-dihy-
droxy acid 6-amino-2,2-dihydroxy-hex-
anoic acid (8), and the cyclic aminol 2-
equilibrium shifts to produce relatively
more 2, 7, and/or 9. Studies of escapinꢀs
enzyme kinetics demonstrate that be-
cause of the high concentrations of es-
capin and l-lysine in the ink secretion,
millimolar concentrations of 3, H2O2,
and ammonia are produced, and also
lower concentrations of 2, 4, 7, and 9 as
a result. We also show that reactions of
this mixture with H2O2 produce d-ami-
novaleric acid (5) and d-valerolactam
(6), with 6 being the dominant compo-
nent under the naturally acidic condi-
tions of ink. Thus, the product of esca-
pinꢀs action on l-lysine contains an
equilibrium mixture that is more com-
plex than previously known for any l-
amino acid oxidase.
hydroxy-piperidine-2-carboxylic
acid
(9). Using NMR and mass spectrosco-
py, we show that 3 is the major compo-
nent of this enzymatic product at any
pH, but at more basic conditions, the
equilibrium shifts to produce relatively
more 4, and at acidic conditions, the
Keywords: amino acids · enols ·
imino compounds · oxidation
Introduction
l-Amino acid oxidases (LAAOs; E.C. 1.4.3.2) are enzymes
that oxidatively deaminate l-amino acids.[1] LAAOs differ in
their substrate specificities. LAAOs can function as toxins in
defensive or offensive systems. Examples include the ven-
omous enzymes of snakes in which LAAOs function in prey
capture and in defense from predators,[2] secreted enzymes
for allelopathy between bacterial species,[3–5] enzymatic
chemical defenses in marine algae against parasites[6] and in
mucus of fish against microbes,[7,8] and enzymatic antimicro-
bial and antipredator defenses in secretions of land and
marine snails.[9–13] One type of marine snail—the sea hare—
has FAD-containing LAAOs in its ink.[11,12] “Escapin” is the
LAAO in the ink of the sea hare Aplysia californica.[10]
When attacked by predators, A. californica co-releases esca-
pin in ink and escapinꢀs substrate l-lysine in opaline. This
mixture contains escapinꢀs products and acts as a chemical
[a] Dr. M. Kamio, K.-C. Ko, Prof. G. Gadda, Prof. P. C. Tai,
Prof. C. D. Derby
Department of Biology, Georgia State University, P.O. Box 4010
Atlanta, GA 30302-4010 (USA)
[b] Dr. M. Kamio, K.-C. Ko, Prof. P. C. Tai, Prof. C. D. Derby
Neuroscience Institute, Georgia State University, P.O. Box 5030
Atlanta, GA 30302-5030 (USA)
Fax : (+1)404-413-5446
E-mail: cderby@gsu.edu
[c] Dr. S. Zheng, Prof. B. Wang, S. L. Collins, Prof. G. Gadda
Department of Chemistry, Georgia State University, P.O. Box 4098
Atlanta, GA 30302-4098 (USA)
[d] Dr. M. Kamio, K.-C. Ko
These authors contributed equally to this study.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200801696.
Chem. Eur. J. 2009, 15, 1597 – 1603
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1597

defense against predators such as California spiny lobsters
and wrasses.[13,14] The preferred substrates of escapin in vitro
are l-lysine and l-arginine, similar to that of LAAOs in
many other species of sea hares.[10,12,15] But since l-lysine is
ꢀ300 times more concentrated than l-arginine in A. cali-
fornicaꢀs opaline (ꢀ145 vs ꢀ0.5 mm), l-lysine is escapinꢀs
primary substrate in situ.[16] Sea hares also have LAAOs in
their egg masses, which may protect the eggs from mi-
crobes.[10–12]
term “escapin intermediate product”, as well as the subse-
quent non-enzymatic reaction products resulting from inter-
actions among these enzymatic products, which we call “es-
capin end product”. Second, we wanted to identify their rel-
ative abundance under the natural pH conditions of ink.
Third, we wanted to evaluate escapinꢀs enzymatic activity at
different pH values to quantify its production toward pre-
dicting the concentrations of the products of LAAO activity
under natural conditions.
A goal of our work is to identify bioactive compounds
produced by the escapin pathway, and by extension, those of
other LAAOs. Previous work on LAAOs suggested and in
some cases demonstrated the identity of molecules in this
pathway.[17–26] This scheme is summarized in the portion of
Scheme 1 highlighted in yellow. This shows that LAAOs oxi-
dize l-lysine (1) to form a-keto-e-aminocaproic acid (2),
and in the process produce H2O2 and NH3. Acid 2 can cy-
clize to yield an equilibrium mixture also containing the
imine, D1-piperidine-2-carboxylic acid (3), and the enamine,
D2-piperidine-2-carboxylic acid (4). H2O2 may react with 2
to form d-aminovaleric acid (5) and with 3 or 4 to form d-
valerolactam (6). However, the ratio of 5 and 6 is highly var-
iable, possibly due to different conditions of the LAAO re-
actions. Our preliminary work on escapin suggested that the
chemistry of this LAAO, and thus perhaps other LAAOs,
may be more complicated than suggested by published stud-
ies and as depicted in Scheme 1, and that the identity of the
molecular species in the equi-
Results and Discussion
Identification of the dominant molecular species in escapinꢀs
intermediate products at neutral pH: Escapin intermediate
product of lysine (EIP) is an equilibrium mixture; its major
component was identified as 3 by NMR and mass spectros-
copy. EIP was prepared by incubating escapin, l-lysine·HCl,
and catalase (to prevent the reaction of H2O2 with EIP).
The presence of anion peaks at m/z 126 and 144 in ESI-MS
experiments using neutral carrier solvents indicated that
EIP contains at least two major species in the EIP equilibri-
um mixture (Scheme 2b). Molecular formulae for molecules
corresponding to these ion peaks were determined by HR-
ESI-TOF-MS as C6H9NO2 (m/z 126.0555 [MꢁH]ꢁ, D+
0.5 mmu; m/z 128.07165 [M+H]+, D+0.4 mmu) and
C6H11NO3 (m/z 144.0660 [MꢁH]ꢁ, Dꢁ0.1 mmu). On the
librium mixture and which
dominate under natural condi-
tions for these enzymes are un-
known. One factor that may in-
fluence the reactions and equi-
librium in the escapin/l-lysine
pathway in vivo is pH, since the
pH of sea hare ink is ꢀ5 when
secreted and ꢀ8 when diluted
in sea water,[27] and since pH is
known to affect some reactions
in the general LAAO pathway
in vitro.[24,28] This issue of iden-
tifying molecular species in an
equilibrium mixture, quantify-
ing their relative abundances,
and determining their bioactiv-
ity has analogs in other fields,
such as in food sciences, where
sugars exist as an equilibrium
mixture and identifying the mo-
lecular species responsible for
sweetness is important in prod-
uct development.[29]
Scheme 1. Summary of the compounds of the escapin/l-lysine pathway in the ink and opaline secretion of sea
hares. The portion of the pathway highlighted in yellow summarizes the state of knowledge of the generic l-
amino acid oxidase/l-lysine pathway from the literature before our study. l-lysine (1), a-keto-e-aminocaproic
acid (2), D1-piperidine-2-carboxylic acid (3), D2-piperidine-2-carboxylic acid (4), d-aminovaleric acid (5), d-va-
lerolactam (6), 6-amino-2-hydroxy-hex-2-enoic acid (7), 6-amino-2,2-dihydroxy-hexanoic acid (8), and 2-hy-
The specific aims of this
study were threefold. First, we
wanted to identify all of esca-
pinꢀs enzymatic reaction prod-
ucts with l-lysine, which we droxy-piperidine-2-carboxylic acid (9).
1598
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1597 – 1603

Sea Hare Ink
FULL PAPER
1
other hand, H NMR spectra of this mixture reveal that one
of the equilibrium mixtureꢀs molecular species dominates
this aqueous solution (Scheme 2b, second spectrum). Since
NMR spectroscopy observes molecules in aqueous condi-
tions, it gives realistic ratios of the molecular species. In
D2O, the intensity of the mixtureꢀs CH2 signal at 2.98 ppm
became smaller and totally disappeared after 24 h, decou-
pling the CH2 signals at 1.88 and 3.76 ppm, indicating that
the CH2 protons at 2.98 ppm are exchangeable with deuteri-
um in D2O (Scheme 2b, bottom spectrum). This indicates
the existence of equilibrium between 3 and 4, 2 and 7. To
avoid this exchange and to stabilize this compound, further
NMR experiments were performed in 90% H2O/10% D2O
under neutral pH conditions. An absence of olefinic protons
towards molecular species with an anion peak at m/z 144,
which include 2, 7, and 9. These results were obtained by
NMR and mass spectroscopy at different pH values. 2D-
NMR analysis showed that under all pH conditions, 3 was
the major component, although the chemical shifts of the
dominant proton signals in EIP at pH 9 were different from
those at pH 1 and 7 (Scheme 2b). An olefinic proton at d
5.70 ppm was observed as minor components at pH 9
(Scheme 2b, third spectrum), indicating that 4 or 7 was pro-
duced in greater relative amounts under higher pH condi-
tions. HMQC and COSY experiments assigned partial struc-
ture for 4 or 7 (Table 2). However, the likelihood for this
being 7 is low because the cyclic forms dominate in this mix-
ture based on mass spectral data (Scheme 2a, c). An a-keto
carbon at 190–204 ppm,[32] an expected key signal for 2, the
direct product of escapin, was not observed in 13C and
HMBC experiments at any pH. Small proton signals from
1.7–1.9 ppm at pH 1 and pH 7, 2.0–2.1 ppm at pH 9, 3.2–
3.4 ppm at pH 1, 7, and 9 were observed (Scheme 2b,
Table 3). However, these protons could not be assigned be-
cause their signals were low intensity and overlapping. MS
analysis was performed using three different carrier solvents
at different pH values. As pH decreased, the intensity of the
ion peak at 144, corresponding to 2, 7, or the cyclic aminol,
2-hydroxy-piperidine-2-carboxylic acid (9), increased rela-
tive to the intensity of the peak at 126, corresponding to 3
or 4 (Scheme 2a, c). Signals ꢀ3.5 ppm might be methylene
protons next to nitrogen atoms of 2, 7, 9, and/or the a-dihy-
droxy acid, 6-amino-2,2-dihydroxy-hexanoic acid (8). Signals
at 2.3 ppm and 1.65–1.85 ppm might be the other methylene
protons of 2, 7, 8, and/or 9. However, full assignment was
not possible because of overlapping signals and lack of suffi-
cient HMBC cross peaks due to low concentrations of these
molecular species. Although it is not possible to give clear
assignment for these molecular species, 7 might not be pres-
ent under the acidic and neutral conditions because the ole-
finic proton was not observed in 1H NMR spectra under
acidic pH conditions when mass spectra show ions corre-
sponding to 7. Ion signals corresponding to 8 were not ob-
served at any pH in mass spectra, indicating that 8 might
not readily ionize under our mass spectrometric conditions;
however, it might exist in aqueous solutions.
1
in H NMR spectra (Scheme 2b, second spectrum) indicated
that this major compound in EIP is not 4 or the linear enol,
6-amino-2-hydroxy-hex-2-enoic acid (7) (Scheme 2c). 2D
NMR experiments established the connectivity of all proton
and carbon atoms in this compound. The C-H connectivity
was established by HMQC experiments to be the following:
CH2-3 (dH =2.98 ppm; dC =29.4 ppm), CH2-4 (dH =
1.88 ppm; dC =19.0 ppm), CH2-5 (dH =1.91 ppm; dC =
21.5 ppm), CH2-6 (dH =3.76 ppm; dC =47.5 ppm) (Table 1).
The ring system was established starting from C3 on the
basis of COSY correlation from H3 to H4, H4 to H5, and
H5 to H6. Long-range coupling between H3 and H6 (J3,6
=
2.7 Hz) is consistent with a structure that has an imine
group in the six-membered ring structure 3. This spin system
is extended on the basis of HMBC correlations from H3 and
H6 to imine carbon C2 (dC =184.5 ppm) and carboxylic
carbon C7 (dC =164.9 ppm) (Table 1), thus establishing that
3 is the major component in EIP. Assignment of the imine
carbon was confirmed based on a down field shift of the
imine carbon with protonation on the imine nitrogen atom
(174 ppm at pH 10 to 184 ppm at pH 7 and pH 1)[30,31]
(Table 1). A small pH effect on the carboxyl carbon might
occur because of the formation of an intramolecular salt
bridge to inhibit protonation on the carboxyl group under
acidic conditions.
Effect of pH on the equilibrium of escapinꢀs intermediate
products: The effect of pH on the equilibrium mixture of
EIP is summarized in Scheme 2c. Compound 3 is the domi-
nant species in the equilibrium at any pH. Basic pH shifts
the equilibrium towards 4. Acidic pH shifts the equilibrium
Synthesis of putative escapin intermediate products: We
synthesized the putative compounds in EIP to verify our
Table 1. NMR data of imine (3) at three different pH conditions in 90% H2O/10% D2O.[a]
pH 1
dC
pH 7
dC
pH 10
dC
Position dH
HMBC (H to C) COSY dH
(J in Hz)
HMBC (H to C) COSY dH
HMBC (H to C) COSY
AHCTUNGTRENNUNG
2
184.4
184.5
29.4
19.0
21.5
174.0
2.47 29.4
1.71 20.8
1.66 23.4
3.56 50.3
163.7
3
4
5
6
7
2.96 29.5
1.87 19.1
1.89 21.6
3.75 47.6
164.9
2,4,5
2,3,5,6
3,4,6
2,4,5
4,6
3,5
4,6
3,5
2.98
1.88 (2.7)
1.91
2,4,5,7
2,3,5,6
3,4,6
4,6
3,5
4,6
3,5
2,4,5
3,5
4
4,6
3,5
4,6
3,5
3.76 (2.7, 5.8) 47.5
164.9
2,3,4,5,7
2, 4
[a] All proton signals are multiplets. J value in pH 7 was analyzed by decoupling caused by deuterium exchange (Scheme 2b, bottom spectrum).
Chem. Eur. J. 2009, 15, 1597 – 1603
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1599

C. D. Derby et al.
Scheme 2. Identification of the molecular species in the equilibrium mixture of escapin intermediate products, and the effect of pH on the equilibrium
mixture. a) Changes in ESI-TOF mass spectrum in negative mode for escapin intermediate products. Aqueous carrier solvents with different pH values
were used. Top: acidic conditions, using 50% MeOH+0.1% HCOOH. Middle: neutral conditions, using 50% MeOH. Bottom: Basic conditions, using
50% MeOH+0.5% NH3. This shows that basic conditions favor the 126-Dalton form and acidic conditions favor the 144-Dalton form. b) Effect of pH
and D2O on 1H NMR spectrum of escapin intermediate products generated by dissolving 5 mg of freeze-dried EIP in 0.5 mL of a 90% H2O/10% D2O
mixture (except bottom spectrum). Top: acidic conditions with 0.4% TFA (pH 1); second: neutral conditions with double distilled H2O (pH 7); third:
basic conditions with 100 mm Na2CO3 (pH 9–10), 3 is the major molecular species at any pH, and 4 (or 7) is observed only at pH 9. * denotes signals that
could not be assigned; these might be signals for 2 or 8–9. We have no experimental evidence for the existence of 8, but it is theoretically possible.
Bottom: neutral condition with 100% D2O. After this mixture was held for 24 h in D2O at room temperature, the CH2 signal at 2.98 ppm indicated by
solid arrow disappeared, causing decoupling of the CH2 signals at 1.88 and 3.76 ppm and thus indicating that the CH2 protons at 2.98 ppm are exchangea-
ble with deuterium in D2O. Imine–enamine conversion between 3 and 4 and/or keto–enol conversion between 2 and 7 makes this proton of 3 exchangea-
ble with deuterium in the NMR solvent. c) Summary of the effects of pH on the equilibrium of EIP. 2, 7, and 9 correspond to 144 (m/z) and 3 and 4 cor-
respond to 126 (m/z) in mass spectra (see Scheme 2a). 8 corresponds to 162 (m/z), but it was not observed in mass spectra under these experimental con-
ditions.
product of this synthetic EIP and 4 is the minor one, as ex-
pected. Comparison of H NMR spectra of natural EIP and
Table 2. NMR data of a spin system reasonable for enamine (4) or enol
(7) in 90% H20/10% D2O at pH 10.[a]
1
synthetic EIP revealed identical signals corresponding to 3
Position
dH mult (J in Hz)
dC
COSY
(Figure S1, Supporting Information).
2
3
4
5
6
7
N/A
112.7
24.8
24.0
44.5
N/A
5.70 t (4.1)
4
2.16 td (4.1, 6.5)
1.75 m (N/A)
3.06 t (5.55)
3,5
4,6
5
Determination of escapin end products: To determine the
molecules in the escapin end products for l-lysine (EEP), a
solution of EEP prepared at pH 7 was compared with candi-
1
date compounds using H NMR spectroscopy.[18,21] 1H NMR
[a] N/A: Cross peaks were not observed because of low sensitivity or
overlap of signals in HMQC experiment. HMBC data were not available
because of low sensitivity and signal overlap as well.
spectra of EEP (Figure 1a, top) showed two sets of spin sys-
tems, corresponding to 5 and 6 (Figure 1). These two com-
pounds were purified by HPLC from EEP mixture and mo-
lecular formulae were confirmed by HR-ESI-TOF-MS as
C5H11NO2 (m/z 116.0710 [MꢁH]ꢁ, Dꢁ0.2 mmu) for 5 and
C5H9NO (m/z 100.0767 [M+H]+, D+0.5 mmu) for 6. Ex-
periments in which EEP was spiked with standards con-
firmed the identity of these compounds (Figure 1a, second
identifications of their presence in the escapin pathway. We
used a non-enzymatic synthetic pathway starting with pipe-
colinic acid ethyl ester, as described by Lu and Lewin.[24]
1H NMR spectra and mass spectra show that 3 is the major
1600
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1597 – 1603

Sea Hare Ink
FULL PAPER
Table 3. List of candidate NMR signals for 2, 8, and 9 at three different pH conditions in 90% H2O/10% D2O.[a]
pH 1
dC
pH 7
dC
pH 10
dC
Serial number of the signals
dH
COSY
dH
COSY
dH
COSY
1
2
3
4
5
1.7 m
1.8 m
2 m
2.44 t
3.27 m
N/A
19.0
34.5
N/A
43.5
2, 3, 5
1, 3, 5
1, 2
3
1, 2
1.69 m
1.78 m
2.32 t
N/A
21.1
N/A
40.7
3
3
2
1,2
1.76 m
1.86 m
2.32 t
28.50 or 24.13
3
4
1
2
24.0
21.8
44.4
3.26 m
3.29 t
[a] These spin systems are candidates for structures 2, 8, and 9 but could not be assigned in this equilibrium mixture. N/A: Cross peaks were not ob-
served because of low sensitivity or overlap of signals in HMQC experiment. HMBC data were not available because of low sensitivity and signal over-
lap as well.
Effect of pH on conversion of
escapin intermediates to esca-
pin end products: A pH-depen-
dent change in the position of
the equilibrium in the EIP mix-
ture was observed using ESI-
MS and NMR (Scheme 2).
Since the 145-Dalton compo-
nent increased at acidic condi-
tion, 2 and/or 9 were expected
to have stronger signals in the
NMR spectra under acidic con-
ditions. However, we could not
identify the molecular species
as 2 or 9 in the NMR spectra of
EIP because of overlapping of
signals. So, we employed an in-
Figure 1. Identification of escapin end products as d-aminovaleric acid (5) and d-valerolactam (6), and the
effect of pH on the ratio of two forms. EEP was generated by adding H2O2 (50 mmol) to EIP generated from
lysine.HCl (13.7 mmol). D2O was added to make a final concentration of 10%. a) 1H NMR spectra. Top spec-
trum: Escapin end product (EEP); Second spectrum: 5; Third spectrum: 6; Fourth spectrum: mixture of
EEP+5; Bottom spectrum: mixture of EEP+. b) Change in ratio of integral value of 1HNMR signal of 5 to
(5+6) in EEP as pH changes from 6 to 8. Values are means ꢂS.E.M., N=5. 5 and 6 were purchased from
Sigma-Aldrich.
direct method—comparing the
ratio of EEP at different pH
values. This solution was pre-
pared by mixing EIP and H2O2
in phosphate buffered solutions
of pH 6, 7, or 8. The relative
amount of 5 in this mixture in-
to fifth). Thus, the conversion of the components of EIP to
their respective components of EEP involves a reaction of 2
with H2O2 to form 5 and a reaction of 4 with H2O2 to form
6, respectively. Additionally, we believe that 3 can be con-
verted directly to 6 by a mechanism similar to the conver-
sion of 2 to 5.
creased at lower pH values and decreased at higher pH
values (Figure 1). These changes may reflect an increased
relative amount of 2 at lower pH. Scheme 1 provides a final
summary of escapinꢀs reaction with l-lysine.
Effect of pH on escapinꢀs kinetics: We examined the effect
of pH on the kinetics of the reaction of escapin with l-lysine
as the substrate (Table 1). Initial rates of reaction were de-
termined in air-saturated buffer by monitoring the rate of
An equilibrium between 5 and 6 is theoretically possible.
We tested this by incubating of either 5 (1.5 mg) or 6
(1.5 mg) in H2O (50 mL) at pH 2, 7, 9, or 11, for 4 h at room
temperature. We did not find any conversion between the
oxygen consumption in the pH range from 4.5 to 8.5. The
1
app
two forms in H NMR spectra. We conclude that there is no
values of
V
app(Vmax/Km) increased with increasing
and ACHTGNURTENNUGN
max
conversion between 5 and 6 within a biological context.
In previous work on LAAOs, 5 is described as the only or
major product in some reactions and 6 the only or major
product in others.[18–21] Which product occurs or dominates
may depend on experimental conditions, such as purity of
the enzymes, pH, concentrations of enzyme and substrate,
temperature, and other factors that may influence the equi-
librium. In the next sections, we explore some of these fac-
tors, such as pH and chemical background, for escapin.
pH and reached limiting values above pH 8, indicating that
an unprotonated group in the active site of the enzyme is re-
quired for the reaction with l-lysine as substrate catalyzed
app
by escapin. The pKa value in the
V
pH-profile was
max
app(Vmax/Km)
lower than the corresponding pKa value in the ACTHUNGRTENNUGN
pH-profile, consistent with decreases of the appKm values to
limiting values of 0.03 to 0.05 mm with increasing pH
app
(Table 1). Thus, while the
V
values increased with in-
max
creasing pH, the appKm values decreased.
Chem. Eur. J. 2009, 15, 1597 – 1603
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1601

C. D. Derby et al.
With l-lysine as substrate, the appKm value at pH 5 is
ꢀ0.8 mm. Assuming that 1 mg of escapin that is released in
ꢀ1 mL (i.e., the amount released during a bout of inking) is
combined with 145 mm l-lysine (i.e., its concentration in
opaline), then escapin is more than 99% saturated with the
substrate. Under these conditions, enzymatic turnover of es-
capin with l-lysine would take as little as ꢀ5 s to produce
1 mm H2O2 and ꢀ30 s to produce ꢀ5 mm H2O2. At higher
pH values, the reactions are even faster. For example, at
pH 8, it would take less than 1 s to produce 1 mm H2O2 and
ꢀ1 min to produce ꢀ80 mm H2O2. Thus, escapin remains
active at the naturally low pH of ink, and owing to the high
concentrations of escapin and lysine in the defensive secre-
tion, prodigious amounts of its reaction products are pro-
duced.
with the blunt end of a scalpel handle. Escapin was isolated and purified
using an ꢂKTA 100 Automated FPLC. A two-step purification process
involving gel filtration followed by purification using cation exchange
was performed according to Yang et al.[10]
Preparation of the intermediate products of escapinꢀs oxidation of l-
lysine: The method of Meister (1952) was modified to make “escapin in-
termediate product for l-lysine” (EIP). l-Lysine monohydrochloride
(10 mg) was incubated with escapin (3 mg) and catalase (130 mg) in
double distilled H2O (1 mL) at 308C on a shaker for up to 20–24 h until
l-lysine was completely consumed, as determined by thin layer chroma-
tography. This solution was then filtered using an Amicon Ultra-4 Centri-
fugal Filter Device (Millipore Corp., Billerica, MA, USA) to remove es-
capin and catalase, and then stored at ꢁ808C until used later. To make
“escapin end product for l-lysine” (EEP), the same procedure was per-
formed except that catalase was not included in the incubation and con-
centration of escapin (0.6 mg) in H2O mL. All chemicals were purchased
from Sigma–Aldrich (St. Louis, MO, USA).
Spectroscopy: NMR spectra were recorded on a Bruker Avance 400
NMR spectrometer using conventional pulse sequence, and FID data
were processed using MestRe Nova software (Mestrelab Research) on
Windows XP. 10% D2O/90% H2O was used as solvent for all experi-
ments except that in Scheme 2b, bottom spectrum. Chemical shift was
referenced for internal TSP. Mass spectra (ESI) were obtained using a
Waters Q-TOF micro mass spectrometer and Applied Biosystems
QSTAR XL.
Conclusions
The mixture resulting from the escapin/lysine reaction in the
ink secretion of sea hares is complex. It is generated by an
initial enzymatic step that generates an equilibrium mixture
of components, and followed by non-enzymatic reactions be-
tween these components and H2O2. The balance within the
equilibrium mixture is dependent on the mixtureꢀs pH,
which ranges from ꢀ5 in full strength ink to ~8 when dilut-
ed in sea water. The enzymatic step of the escapin pathway
generates D1-piperidine-2-carboxylic acid (3) and H2O2 as
the major components, regardless of the pH value. Minor
forms include a-keto-e-aminocaproic acid (2), 6-amino-2-hy-
droxy-hex-2-enoic acid (7), and aminol 2-hydroxy-piperi-
dine-2-carboxylic acid (9), which increase in abundance at
lower pH, and D2-piperidine-2-carboxylic acid (4), which in-
creases in abundance at higher pH. The non-enzymatic reac-
tion of H2O2 with these intermediate forms results in the
end products d-aminovaleric acid (5) and d-valerolactam
(6), with 6 dominating under all pH conditions but the rela-
tive abundance of the latter increasing under acidic condi-
tions. The complexity of the mixture generated by escapin in
aqueous solution is greater than previously realized for any
l-amino acid oxidase. The pH effect on the equilibrium in
EIP and on the production of end products shows that pH
affects the activity of EIP and EEP. Our results with escapin
may indicate equally complex mixtures are generated by
other l-lysine oxidases. The products of the escapin pathway
can now be tested in biological assays to determine which
are antipredatory chemical defenses or antimicrobial agents.
Enzyme kinetics and pH profiles: Enzyme activity was measured in air-
saturated 50 mm potassium phosphate, 150 mm NaCl, in the pH range
from 4.5 to 8.5, at 258C, by monitoring the rate of oxygen consumption
with a computer-interfaced Oxy-32 oxygen monitoring system (Hansa-
tech Instrument, Ltd.). The reactions were started by the addition of es-
capin to a 1 mL reaction mixture, with the final concentration of enzyme
in the 5 to 20 mgmLꢁ1. At any given pH, the final concentrations of l-
lysine spanned from 0.2- to 5-times the appKm values, which corresponded
to 0.01 to 7 mm depending upon the pH. Data were fit with Kaleida-
Graph software (Synergy Software, Reading, PA). The apparent kinetic
app
parameters
V
max
, ACHTUNGRTENNUNG
app(Vmax/Km), and appKm for l-lysine as substrate for es-
capin in air-saturated buffers were determined by fitting initial rates of
reaction at different concentrations of substrate to the Michaelis–Menten
app
equation for one substrate. The pH dependence of the
V
and
max
app(Vmax/Km) values were determined by fitting initial rate data to equa-
ACHTUNGTRENNUNG
tion 1, which describes a curve with a slope of +1 and a plateau region
at high pH. Y is the pH-independent value of the kinetic parameter of in-
terest, and Ka is the dissociation constant for the ionization of groups
which are relevant for catalysis.
!
Y
10ꢁpH
10ꢁpKa
log Y ¼ log
ꢀ
ꢁ
ð1Þ
1 þ
Conversion of EIP to EEP at different pH values: Potassium phosphate
buffer was used as the incubation solution for the conversion of EIP to
EEP, since the type of buffer affected the results. Freeze-dried EIP
(2.5 mg, 13.7 mmol lysine equivalent) and H2O2 (50 mmol in 5.6 mL of
30% solution) were mixed with 100 mm potassium phosphate buffer and
adjusted pH to 6, 7, and 8 to a final volume of 0.5 mL. Reactions were
performed for 1.5 h at 258C (Thermomixer, Eppendorf), and products
were analyzed using NMR. For high resolution mass spectroscopy, com-
pound
5 and 6 were purified by Beckman HPLC System GOLD
equipped with Nomura Chemical Develosil RPAQUEOUS column.
Water - methanol gradient containing 0.1% trifluoroacetic acid was used
for elution of the compounds.
Experimental Section
Animals: Sea hares Aplysia californica Cooper 1863 were collected in
California by Marinus Scientific (Garden Grove, CA, USA). They were
dissected on the day of arrival in our laboratory. Experiments were per-
formed within the university regulations and national guidelines.
Acknowledgements
Collection of ink and isolation of escapin: The ink glands were dissected
from anesthetized animals and frozen at ꢁ808C until used. Purple ink
was collected by gently squeezing dissected ink glands in a Petri dish
Supported by NSF IBN-0324435 and 0614685 (to C.D.D.), NIH GM-
34766 (to P.C.T.), NSF MCB-0545712 and American Chemical Society
1602
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1597 – 1603

Sea Hare Ink
FULL PAPER
PRF 47636-AC4 (to G.G.), a grant from GSUꢀs Brains & Behavior pro-
gram, Core Facilities of the Center for Biotechnology and Drug Design,
and the Georgia Research Alliance. K.C.K. is a Fellow of GSUꢀs Molecu-
lar Basis of Disease Program. We thank Dr. Masaki Fujita, Dr. Julia Ku-
banek, Dr. Yoichi Nakao, and Dr. Toshiyuki Wakimoto, for valuable dis-
cussions on chemical shift assignments of NMR data and Siming Wang
and David Bostwick for operation of mass spectroscopy.
[15] P. M. Johnson, C. E. Kicklighter, M. Schmidt, M. Kamio, H. Yang,
D. Elkin, W. C. Michel, P. C. Tai, C. D. Derby, J. Exp. Biol. 2006,
209, 78–88.
[16] C. D. Derby, C. E. Kicklighter, P. M. Johnson, X. Zhang, J. Chem.
Ecol. 2007, 33, 1105–1113.
[17] A. Meister, J. Biol. Chem. 1952, 197, 309–317.
[18] A. Meister, J. Biol. Chem. 1954, 206, 577–585.
[19] P. Boulanger, P. R. Osteux, Biochim. Biophys. Acta 1956, 21, 552–
561.
[20] A. Meister, A. N. Radhakrishnan, S. D. Buckley, J. Biol. Chem. 1957,
229, 789–800.
[1] A. Meister, Biochemistry of the Amino Acids, Academic, New York,
1965.
[21] D. B. Hope, K. C. Horncastle, Biochem. J. 1967, 102, 910–916.
[22] A. J. L. Cooper, A. Meister, J. Biol. Chem. 1978, 253, 5407–5410.
[23] H. Kusakabe, K. Kodama, A. Kuninaka, H. Yoshino, H. Misono, K.
Soda, J. Biol. Chem. 1980, 255, 976–981.
[24] S.-P. Lu, A. H. Lewin, Tetrahedron 1998, 54, 15097–15104.
[25] S. N. Murthy, M. K. Janardanasarma, Mol. Cell. Biochem. 1999, 197,
13–23.
[26] E. V. Lukasheva, T. T. Berezov, Biochemistry (Moscow) 2002, 67,
1394–1402.
[27] S. Shabani, S. Yaldiz, L. Vu, C. D. Derby, J. Comp. Physiol. A 2007,
193, 1195–1204.
[2] X.-Y. Du, K. J. Clemetson, Toxicon 2002, 40, 659–665.
[3] D. Gꢃmez, P. Lucas-Elꢄo, A. Sanchez-Amat, F. Solano, Biochim.
Biophys. Acta Proteins Proteomics 2006, 1764, 1577–1585.
[4] D. Gꢃmez, E. Espinosa, M. Bertazzo, P. Lucas-Elꢄo, F. Solano, A.
Sanchez-Amat, Appl. Microbiol. Biotechnol. 2008, 79, 925–930.
[5] A. Mai-Prochnow, P. Lucas-Elꢄo, S. Egan, T. Thomas, J. S. Webb, A.
Sanchez-Amat, S. Kjelleberg, J. Bacteriol. 2008, 190, 5493–5501.
[6] F. Weinberger, Biol. Bull. 2007, 213, 290–302.
[7] S. K. Jung, A. Mai, M. Iwamoto, N. Arizono, D. Fujimoto, K. Saka-
maki, S. Yonehara, J. Immunol. 2000, 165, 1491–1497.
[8] Y. Kitani, N. Kikuchi, G. Zhang, S. Ishizaki, K. Shimakura, K.
Shiomi, Y Nagashima, Comp. Biochem. Physiol. 2008, 149B, 394–
400.
[28] J. Cabello, B. Leon, V. Prajoux, M. Plaza, Arch. Biochem. Biophys.
1964, 107, 51–56.
[29] M. G. Lindley, R. S. Shallenberger, R. L. Whistler, J. Food Sci. 1976,
41, 575–577.
[30] T. Seki, M. Satake, L. Mackenzie, H. F. Kaspar, T. Yasumoto, Tetra-
hedron Lett. 1995, 36, 7093–7096.
[31] D. Uemura, T. Chou, T. Haino, A. Nagatsu, S. Fukuzawa, S.-Z.
Zheng, H.-S. Chen, J. Am. Chem. Soc. 1995, 117, 1155–1156.
[32] W. Bremser, B. Franke, H. Wagner, Chemical Shift Ranges in
Carbon-13 NMR Spectroscopy, VCH, Weinheim, 1982.
[9] N. Kanzawa, S. Shintani, K. Ohta, S. Kitajima, T. Ehara, H. Kobaya-
shi, H. Kizaki, T. Tsuchiya, Arch. Biochem. Biophys. 2004, 422, 103–
109.
[10] H. Yang, P. M. Johnson, K.-C. Ko, M. Kamio, M. W. Germann, C. D.
Derby, P. C. Tai, J. Exp. Biol. 2005, 208, 3609–3622.
[11] H. Kamiya, R. Sakai, M. Jimbo in Molluscs: From Chemo-ecological
Study to Biotechnological Application, Marine Molecular Biotech-
nology, Progress in Molecular and Subcellular Biology (Eds.: G.
Cimino, M. Gavagnin), Springer, Berlin, 2006, pp. 215–239.
[12] C. D. Derby, Biol. Bull. 2007, 213, 274–289.
[13] J. F. Aggio, C. D. Derby, J. Exp. Mar. Biol. Ecol. 2008, 363, 28–34.
[14] M. Kamio, C. Kicklighter, K.-C. Ko, M. Nusnbaum, J. Aggio, M.
Hutchins, C. D. Derby, Chem. Senses 2007, 32, A37.
Received: August 15, 2008
Published online: January 7, 2009
Chem. Eur. J. 2009, 15, 1597 – 1603
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1603