Characterization of an epoxide hydrolase from the Florida red tide dinoflagellate, Karenia brevis

Article abstract of DOI:10.1016/j.phytochem.2015.11.002

Epoxide hydrolases (EH, EC 3.3.2.3) have been proposed to be key enzymes in the biosynthesis of polyether (PE) ladder compounds such as the brevetoxins which are produced by the dinoflagellate Karenia brevis. These enzymes have the potential to catalyze kinetically disfavored endo-tet cyclization reactions. Data mining of K. brevis transcriptome libraries revealed two classes of epoxide hydrolases: microsomal and leukotriene A₄ (LTA₄) hydrolases. A microsomal EH was cloned and expressed for characterization. The enzyme is a monomeric protein with molecular weight 44 kDa. Kinetic parameters were evaluated using a variety of epoxide substrates to assess substrate selectivity and enantioselectivity, as well as its potential to catalyze the critical endo-tet cyclization of epoxy alcohols. Monitoring of EH activity in high and low toxin producing cultures of K. brevis over a three week period showed consistently higher activity in the high toxin producing culture implicating the involvement of one or more EH in brevetoxin biosynthesis.

Full text of DOI:10.1016/j.phytochem.2015.11.002

Phytochemistry xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Phytochemistry
journal homepage: www.elsevier.com/locate/phytochem
Characterization of an epoxide hydrolase from the Florida red tide
dinoflagellate, Karenia brevis
Pengfei Sun a, Cristian Leeson a, Xiaoduo Zhi a, Fenfei Leng a, Richard H. Pierce b, Michael S. Henry b,
Kathleen S. Rein a,
⇑
a Department of Chemistry and Biochemistry, Florida International University, Miami, FL 33199, USA
b Mote Marine Laboratory, 1600 Ken Thompson Parkway, Sarasota, FL 34236, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 20 June 2015
Received in revised form 19 October 2015
Accepted 5 November 2015
Available online xxxx
Epoxide hydrolases (EH, EC 3.3.2.3) have been proposed to be key enzymes in the biosynthesis of poly-
ether (PE) ladder compounds such as the brevetoxins which are produced by the dinoflagellate Karenia
brevis. These enzymes have the potential to catalyze kinetically disfavored endo-tet cyclization reactions.
Data mining of K. brevis transcriptome libraries revealed two classes of epoxide hydrolases: microsomal
and leukotriene A4 (LTA4) hydrolases. A microsomal EH was cloned and expressed for characterization.
The enzyme is a monomeric protein with molecular weight 44 kDa. Kinetic parameters were evaluated
using a variety of epoxide substrates to assess substrate selectivity and enantioselectivity, as well as
its potential to catalyze the critical endo-tet cyclization of epoxy alcohols. Monitoring of EH activity in
high and low toxin producing cultures of K. brevis over a three week period showed consistently higher
activity in the high toxin producing culture implicating the involvement of one or more EH in brevetoxin
biosynthesis.
Keywords:
Dinoflagellate
Karenia brevis
Prorocentrum hoffmanianum
Dinophyceae
Protein biochemistry
Brevetoxin
Epoxide hydrolase
Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction
1987; Lee et al., 1986, 1989). Since then, similar patterns have been
observed in other PE ladder compounds (Bourdelais et al., 2005;
The brevetoxins (e.g. brevetoxin B (1)) (Scheme 1) are produced
by Karenia brevis, the principal HAB organism in the Gulf of Mexico
also known as the Florida red tide dinoflagellate. These polyether
ladder compounds are responsible for massive fish kills, marine
mammal mortalities and, in humans, neurotoxic shellfish poison-
ing (NSP) and asthma-like symptoms through ingestion or inhala-
tion exposures, respectively (Fleming et al., 2005; Flewelling et al.,
2005; Van Dolah et al., 2003). The brevetoxins are representative of
a larger group of marine algal toxins known as polyether (PE) lad-
ders which includes ciguatoxins and yessotoxins. These molecules
share common structural features which include a series of
trans-fused polyether rings with oxygen atoms alternating across
the ‘‘top” and ‘‘bottom” of the molecules. Shortly after their struc-
tures were determined in the 1980s, efforts to understand the
biosynthesis of these complex molecules began. Stable isotope
incorporation experiments using 13C-labeled acetate established
the polyketide origins of the brevetoxins, yet some anomalies in
the labeling patterns, were also observed (Chou and Shimizu,
Gallimore and Spencer, 2006). As a class of compounds, polyke-
tides are structurally diverse yet share common biogenic
a
pathway which is highly analogous to fatty acid biosynthesis.
The carbon skeletons are constructed by the sequential condensa-
tion of small carboxylic acid subunits in a series of reactions cat-
alyzed by a polyketide synthase (PKS). While transcripts coding
for functional domains of PKSs including ketosynthases (KS),
ketoreductase (KR), and an acyl carrier proteins (ACP) have been
identified, K. brevis appears to have a PKS architecture which is
distinct from those of bacteria, plants or fungi (Monroe and Van
Dolah, 2008; Snyder et al., 2003, 2005).
While the origins of the carbon atoms in the brevetoxins have
been determined, the remainder of the polyether ladder pathway
has yet to be determined. The prevailing hypothesis, put forth
nearly 30 years ago by Nakanishi, suggests that PKS-mediated syn-
thesis of the polyene is followed by epoxidation to afford a polye-
poxide which then undergoes an epoxide-opening cascade,
catalyzed by an epoxide hydrolase (EH), to provide the polyether
(brevetoxin B, 1 in the example shown in Scheme 1) (Lee et al.,
1989; Nakanishi, 1985). This hypothesis, which remains uncon-
firmed, accounts for the oxygen–carbon–carbon (O–C–C) connec-
tivity pattern and trans-syn topography in the natural products
⇑
Corresponding author at: 11200 SW 8th St., Miami, FL 33199, USA.
E-mail addresses: psun001@fiu.edu (P. Sun), lengf@fiu.edu (F. Leng), rich@mote.
org (R.H. Pierce), mhenry@mote.org (M.S. Henry), reink@fiu.edu (K.S. Rein).
http://dx.doi.org/10.1016/j.phytochem.2015.11.002
0031-9422/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Sun, P., et al. Characterization of an epoxide hydrolase from the Florida red tide dinoflagellate, Karenia brevis. Phytochem-
istry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.11.002

2
P. Sun et al. / Phytochemistry xxx (2015) xxx–xxx
Scheme 1. Proposed polyether ladder (left) and confirmed polyether (right) biosynthesis.
and is supported by the recent identification of molecular oxygen
(O2) as the source of the ether oxygen atoms in the structurally
similar PE ladder yessotoxin, produced by the dinoflagellate Proto-
ceratium reticulatum (Yamazaki et al., 2012).
The scheme proposed by Nakanishi for brevetoxin biosynthesis
is conceptually analogous to the biosynthesis of monensin (2) and
other bacterial, non-ladder (linked) polyethers, wherein an EH is
the key enzyme responsible for the construction of the polyether
rings (Gallimore and Spencer, 2006; Matsuura et al., 2010;
Minami et al., 2011; Shichijo et al., 2008). An alternative proposal
for PE ladder biosynthesis suggests that chain extension, epoxida-
tion and cyclization may be performed in an iterative process
(Gallimore and Spencer, 2006). At this time, no gene or enzyme
produces an intermediate diol which lactonizes to release the flu-
orescent reporter 4 (Scheme 2) after hydrolysis of the intermediate
cyanohydrin. It was reasoned that a substrate containing aliphatic
rather than aromatic substituents would be more representative of
the native substrate if it were involved in brevetoxin biosynthesis.
Thus a second probe (5b) was prepared from trans-3-hexenoic acid
according to the procedure described for the synthesis of 5a (Jones
et al., 2005). Activity was detected in K. brevis CFE using both
probes, but probe 5b proved to be the better substrate as antici-
pated (Fig. 1A). The ratio of initial rates (Vo) for substrates 5a:5b
was 1:3. In CFE, probes 5a and 5b will detect both EH and esterase
activity. However, the alkene precursors can be used to assess for
esterase activity alone. Higher levels of activity were observed
when using the epoxide probes 5a/b when compared to the alkene
probes 3a/3b, giving rise to the possibility that both esterase and
EH activity were being detected by the epoxide probes. Therefore,
all assays of CFE were corrected for esterase activity by subtracting
the absorbance values obtained with the corresponding esterase
has been definitively linked to polyketide biosynthesis in
a
dinoflagellate. This is in part due to the size and complexity of
the dinoflagellate genome, and the lack of a transformation
system.
In the work reported herein, a study was carried out to identify
and characterize one or more epoxide hydrolases from K. brevis and
to investigate their potential role in brevetoxin biosynthesis.
probes. The ratios of initial rates were comparable (2.02,
r = 0.17,
n = 5 and 2.16, = 0.19, n = 8: represents average of n experiments,
r
each experiment performed in duplicate or triplicate) for the 5a/3a
and 5b/3b pairs when assayed under identical conditions. Finally,
the EH and esterase activities of CFE were evaluated from the
dinoflagellate Prorocentrum hoffmanianum, producer of the linear
polyether okadaic acid 6 (Fig. 2) which consists of only two fused
ether rings. Because the proposed biosynthesis of okadaic acid 6
would involve a single epoxide ring opening reaction, it is interest-
ing and perhaps noteworthy that the initial rates of EH activity
(normalized to protein) for K. brevis CFE was more than two-fold
greater than that of P. hoffmanianum CFE (2.3:1). Furthermore, in
P. hoffmanianum, esterase activity was higher than EH activity
(Fig. 1D).
2. Results and discussion
2.1. Epoxide hydrolase activity in cell free extracts
Efforts to identify and characterize epoxide hydrolases from K.
brevis began by examining a cell free extract (CFE) using a sensitive
fluorescent assay. Substrate (5a) (see Scheme 2) was prepared from
trans-styrylacetic acid and 7-methoxy-2-napthaldehyde, 4 as pre-
viously described (Jones et al., 2005). Typically, an epoxide hydro-
lase will catalyze epoxide ring opening to yield the corresponding
1,2-diol. An EH catalyzed hydrolysis of the epoxide substrate, 5a
Please cite this article in press as: Sun, P., et al. Characterization of an epoxide hydrolase from the Florida red tide dinoflagellate, Karenia brevis. Phytochem-
istry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.11.002

P. Sun et al. / Phytochemistry xxx (2015) xxx–xxx
3
Scheme 2. Preparation and use of EH and esterase probes.
B. EH activity in K. brevis CFEin
A. EH and esterase activity in K.
brevis CFE
the presence of admantyle urea 7
and valpromide 8
7
6
5
4
3
2
1
0
5a
5b
3a
3b
3
2
1
0
5a
5a + 7
5a + 8
0
50
100
Time (min)
150
200
0
50
100
Time (min)
150
200
C. Inhibition of esterase activity of K.
brevis CFE
D. EH and Esterase activity in P.
hoffmanianumCFE
3
3
2
1
0
3a 3a + 7 3a + 8
3b
5b
2
1
0
0
50
100
Time (min)
150
200
0
50
100
150
200
Time (min)
Fig. 1. EH activity in K. brevis and P. hoffmanianum CFE and sensitivity to inhibitors. (A) EH (5a/5b) and esterase (3a/3b) activity. (B) Inhibition of EH activity. (C) Inhibition of
esterase activity. (D) EH (5b) and esterase (3b) activity in CFE of P. hoffmanianum.
Fig. 2. Structure of okadaic acid (6).
Please cite this article in press as: Sun, P., et al. Characterization of an epoxide hydrolase from the Florida red tide dinoflagellate, Karenia brevis. Phytochem-
istry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.11.002

4
P. Sun et al. / Phytochemistry xxx (2015) xxx–xxx
Scheme 3. Mechanisms of epoxide hydrolases. (A) The two step mechanism of the
a,b-hydrolase fold type. (B) Leukotriene-A4-hydrolase. (C) Limonene epoxide hydrolases.
Three types of EH, which operate through distinct catalytic
mechanisms, have been identified (Morisseau and Hammock,
2005). The most studied of these belong to the larger family of
the LEH type. Genes coding for LEHs have been identified in the
biosynthetic gene clusters for the bacterial polyethers monensin
(2) (Oliynyk et al., 2003), lasalocid (Smith et al., 2008), nan-
changmycin (Sun et al., 2003), tetronomycin (Demydchuk et al.,
2008) and nigericin (Harvey et al., 2007). In the case of monensin
(2) and lasalocid, EH encoding genes have been expressed and
the EH activity of the enzymes confirmed (Minami et al., 2011,
2013).
hydrolases containing an
a,b-fold and are often referred to as a,
b-hydrolases. Two conserved tyrosine residues and a conserved
Asp-His-Asp catalytic triad make up the active site. The tyrosines
are believed to serve as general acids for activation of the epoxide
by forming two hydrogen bonds with the oxygen (Yamada et al.,
2000). The carboxylate group of one of the Asp residues is activated
by the His and second Asp in the active site and is the nucleophile
that attacks the epoxide, typically at the least substituted carbon.
In a second step, a water molecule, activated by the His-Asp charge
relay, cleaves the ester that is formed, liberating the diol and
regenerating the enzyme (Scheme 3A; Tzeng et al., 1996). A second
type of EH is the leukotriene-A4 (LTA4) hydrolase (Rudberg et al.,
2004). The LTA4 hydrolase is highly substrate specific, converting
LTA4 into leukotriene-B4 (LTB4). Only two other substrates, the
double bond isomers LTA3 and LTA5, have been identified for this
EH. In this enzyme active site, two histidines and a glutamate coor-
dinate a Zn2+ ion, which in turn coordinates the epoxide oxygen.
Upon activation of the 5,6-epoxide, an intermediate carbocation,
whose charge is delocalized over a conjugated triene system, is
quenched at C-12 by a water molecule which is activated and
delivered by a conserved Asp residue, to produce a 5, 12-diol rather
than a 1,2-diol (Scheme 3B). A covalent adduct between the
enzyme and substrate is not formed in this process making it dis-
The EH activity of K. brevis CFE was examined in the presence of
known inhibitors of two of the three classes of EH: adamantly urea
7, inhibitor of mouse (Ki of 0.33
hydrolase type (Kim et al., 2004) and inhibitors of the Rhodococcus
erythropolis LEH: valpromide 8 (Ki of 100 M) (Arand et al., 2003).
lM) and human (Ki of 6.2 lM) a,b-
l
Due to the substrate specificity and unique mechanism of the LTA4
hydrolase, it is unlikely that our probes would detect such activity.
Among the two EH inhibitors tested, only adamantly urea 7 inhib-
ited the total activity in the K. brevis CFE (Fig. 1B) when using probe
5a as the substrate. At equal concentrations (100 lM) of substrate
5a and inhibitor, a 44% inhibition was observed in the presence of
adamantyl urea 7. Esterase activity was not inhibited by either of
the two EH inhibitors. These data suggest that the observed EH
activity in K. brevis CFE can be attributed to an epoxide hydrolases
of the a,b-hydrolase type.
2.2. Identification, cloning and expression of an EH from K. brevis
tinct from the
a
,b-hydrolase type. The third EH is the limonene-
tBLASTn (Altschul et al., 1990) analysis of a K. brevis EST library
consisting of 22,000 predicted transcripts (Lidie et al., 2005)
against Rhodococcus erythropolis LEH and nine other known or
putative LEHs which are associated with biosynthetic gene clusters
of bacterial polyethers: monenesin (2) (MonB1 and MonB2), lasa-
locid (Lsd19A and Lsd19B), nanchangmycin (NanIA and NanIB),
tetronomycin (TmnB), nigericin (NigB1 and NigB3) was performed.
1,2-epoxide hydrolase (LEH). In this case, the epoxide is activated
by hydrogen bonds to an Asp residue. However, as in the case of
LTA4 hydrolase, a water molecule is activated and delivered by
Asn, Asp and Tyr residues in a one-step mechanism and a covalent
intermediate is not formed (Arand et al., 2003). EHs which are
involved in the biosynthesis of non-ladder type polyethers are of
Please cite this article in press as: Sun, P., et al. Characterization of an epoxide hydrolase from the Florida red tide dinoflagellate, Karenia brevis. Phytochem-
istry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.11.002

P. Sun et al. / Phytochemistry xxx (2015) xxx–xxx
5
All failed to return any significant homologies with the smallest
Expect values on the order of 10À5 tBLASTn analysis against
Emeliania huxleii LTA4 hydrolase (NCBI Accession number:
XP_005787867) returned a single transcript having an Expect value
smaller than 10À5 (3 Â 10À6). However, BLASTx analysis of the sin-
gle transcript did not return significant homologies to any other
LTA4 hydrolase. tBLASTn analysis of the K. brevis EST library against
brevis CFE, the purified EH showed a preference for epoxide 5b over
5a. The ratio of initial rates (Vo) for substrates 5a:5b was 1:4. Epox-
ide hydrolases of the a,b-hydrolase type share significant homolo-
gies with other hydrolases, including esterases. As done with the
CFE, EH activity was distinguished from esterase activity using
the substrates 3a and 5a. No activity was observed using the ester-
ase probes indicating that the expressed protein was indeed an EH.
Substrate and enantioselectivity of the K. brevis EH was assessed
by kinetic studies. Kinetic parameters of the enzyme using sub-
strates 5a and 5b were performed using the sensitive fluorescent
assay. Other substrates having no fluorescent reporter group were
evaluated by GC–MS analysis of the reaction mixtures using 1-
naphthol as an internal standard. The data shown in Table 1 indi-
cates that the K. brevis EH has a strong preference for terminal
epoxides over internal epoxides (Entries 3–9 vs. Entries 1–2). Ali-
phatic substituents are preferred over aromatic substituents by
approximately 5:1 (Entries 7–9 vs. Entries 3–6) and the enzyme
shows only a slight preference for R over S enantiomers (Entries
3 and 4). The reaction rates using 1,2-disubstituted epoxides as
substrates are too slow to be determined using the GC–MS method.
However, disubstituted epoxides proved to be inhibitors of the K.
brevis EH only when an aromatic substituent is present (Entries
10–12 vs. Entries 13–14) and cis-disubstituted epoxides are more
potent inhibitors than those that are trans-disubstituted (Entries
10–12).
A noteworthy distinction between the monensin (2) pathway
and the proposed brevetoxin (1) pathway is the site of nucleophilic
attack on the epoxide substrate (Scheme 1). In the case of non-lad-
der polyethers, the majority of epoxide ring opening reactions pro-
ceed via attack at the proximal carbon (exo-tet cyclization) while
polyether ladder formation requires attack at the distal carbon
(endo-tet cyclization). Both the empirical Baldwin’s rules
(Baldwin, 1976) and molecular orbital calculations (Gruber et al.,
1999) indicate that the endo-tet cyclization pathway is kinetically
a variety of prokaryotic and eukaryotic EHs of the
type revealed transcript (MGID2034061) with significant
homologies having Expect values ranging from 10À126 to 10À103
a,b-hydrolase
a
.
The identified EST coded for a 393 amino acid protein (EH1) with
a calculated molecular weight of 44.6 kDa. BLASTp analysis of this
protein against the NCBI non-redundant protein database returned
a putative hydrolase with Expect value of 10À126, from the polar,
eukaryotic microalga Coccomyxa subellipsoidea (Blanc et al., 2012)
as the closest match. Fig. 3A shows the alignment of the translated
protein with the a,b-hydrolase type EH from Aspergillus niger (NCBI
Accession Number: CAB 49813.1; (Zou et al., 2000)). The two
conserved catalytic tyrosine residues can be identified in the
translated protein. However, the Asp-His-Asp catalytic triad is
replaced with Asp-His-Glu. After optimization of the codon usage
for Escherichia coli expression, the putative K. brevis EH gene was
synthesized, cloned into pUC 57 and subcloned into pET30a+ for
expression with the incorporation of six C-terminal histidine
codons to facilitate purification of the expressed protein. As shown
in Fig. 3B, the purified protein has the anticipated molecular
weight of 44 kDa.
2.3. Characterization of K. brevis EH
Fluorescent probe sets 3b/5b and 3a/5a were used for initial
characterization of the expressed K. brevis EH (Fig. 3C). The EH
was optimally active at pH 6.8–7.1, with half-maximal activity at
pH 6.5 and 8.3. Consistent with the selectivity observed in the K.
A
K. brevis EH
A. niger EH
----------MPLTDLKPFKLTVPDQQLDDLKYRISATRWPDQLDSPANANWEYGTEITYLKSLAEYWRTTFSWRKQEELLNAMPQFTAKLNGVTVHFVHQRCLSSPTAPALLICHGWPG 110
MSAPFAKFPSSASISPNPFTVSIPDEQLDDLKTLVRLSKIAPPTYESLQADGRFGITSEWLTTMREKWLSEFDWRPFEARLNSFPQFTTEIEGLTIHFAALFSERE-DAVPIALLHGWPG 119
.
. :**.:::**:******
:
:: .
.. :*: .:*
:*.:: * * : *.**
*
**::****::::*:*:**.
.
.
* .: : *****
▼
K. brevis EH
A. niger EH
SIFEFHKVLGPLAEN-------FHVVAPSIPGYGFSEAP-HKRGFNAPEAARLFHALMLELGY-QEYVAQGGDWGSVITQCIGKLFPKQCKAIHINMPTAPRPKD-FDMSKLNPAEVEGL 220
SFVEFYPILQLFREEYTPETLPFHLVVPSLPGYTFSSGPPLDKDFGLMDNARVVDQLMKDLGFGSGYIIQGGDIGSFVGRLLGVGF-DACKAVHLNLCAMRAPPEGPSIESLSAAEKEGI 238
*:.**: :* : *:
**:*.**:*** **..* .:.*. : **:.. ** :**: . *: **** **.: : :* * . ***:*:*: :
* : .:..*..** **:
▼
▼
▼
K. brevis EH
A. niger EH
RRTMAFNTYGTGYQKISGTKPQTLSYGLNDSPIGLLAWMTEKFRDWSDCNGTLENAFTQDEILTNAMIYWVSQSIGSSMRFYYENIAVMPGTSKEMIELAAVKVSVPTAVAMFPKEIFFC 340
ARMEKFMTDGLAYAMEHSTRPSTIGHVLSSSPIALLAWIGEKYLQWVDKP------LPSETILEMVSLYWLTESFPRAIHTYRETTPTASAPNGATMLQKELYIHKPFGFSFFPKDLCPV 352
*
* * * .*
.*:*.*:.: *..***.****: **: :* *
:..: ** . :**:::*: ::: * *. .. ....
:
: : * ..::***::
▼
K. brevis EH
A. niger EH
PRAWCEQAYNLQQYTRFEAGGHFAALEKPEELVKDIKDFFLRRLDLDKNQSRL 393
PRSWIATTGNLVFFRDHAEGGHFAALERPRELKTDLTAFVEQVWQK------- 398
**:*
: **
:
.
********:*.** .*:. *. :
:
B
C
14
12
10
8
Purified K. brevisEH
3b
5b
6
4
2
0
0
50
100
Time (min)
150
200
Fig. 3. (A) Alignment of putative K. brevis EH1 with Aspergillus niger EH. Catalytic residues are indicated by .. (B) SDS PAGE of expressed protein after purification and dialysis.
Lanes M: Precision Plus Protein Unstained Standards (Biorad 161-0363) lane 1: lysate; lanes 2–3: column flow-through; lanes 4–5: column washes; lane 6: elution. C. EH (5b)
and esterase (3b) assay of purified K. brevis EH.
Please cite this article in press as: Sun, P., et al. Characterization of an epoxide hydrolase from the Florida red tide dinoflagellate, Karenia brevis. Phytochem-
istry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.11.002

6
P. Sun et al. / Phytochemistry xxx (2015) xxx–xxx
Table 1
Kinetic parameters of the K. brevis EH with various substrates and inhibitors.
Entry
Substrate
Km
(
l
M)
13.6
161
kcat (sÀ1
)
kcat (sÀ1)/Km (M)
IC50 (mM)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Probe 5a (R = Ph)
Probe 5b (R = Et)
(R) styrene oxide
(S) styrene oxide
(R/S) styrene oxide
7.81EÀ4
4.46EÀ3
9.01EÀ1
8.32EÀ1
8.58EÀ1
3.31EÀ1
6.15
5.36
7.12
–
–
57.4
27.7
144
129
131
62.3
238
241
1483
–
–
–
–
–
–
–
–
–
6240
6430
6570
5310
25,800
22,200
4,800
–
–
–
–
–
a
-Methylstyrene oxide
(R) 1-butane oxide
(R/S) 1-butane oxide
1,2,7,8-Diepoxyoctane
cis-b-Methylstyrene oxide
trans-b-Methylstyrene oxide-
b-Dimethylstyrene oxide
cis-2,3-Epoxybutane
–
3.05
14.28
6.99
–
–
–
–
–
–
–
–
2,3-Dimethyl-2,3-epoxybutane
–
which are not understood. Cellular brevetoxin (1) content in both
cultures was determined by LC–MS with the higher toxin culture
having 5–6.8 pg/cell, whereas the low toxin culture had only
0.002–0.05 pg/cell. EH activity was assessed using the EH/esterase
probe set 3b/5b. Esterase activity, determined using probe 3b, was
subtracted from total activity, determined using probe 5b. Cultures
were diluted on day 0 to equal cell counts and monitored over a
21 day period. As shown in Fig. 4A, EH activity for the high toxin
culture was consistently higher than that of the low toxin culture.
On average, EH activity for the high toxin culture was 18% and 60%
higher than the low-toxin culture when normalized to protein con-
tent or to cell counts, respectively. A two tailed t-test confirms that
there is a significance difference in EH activity when normalized to
protein (p = 8 Â 10À14) or normalized to cell counts (p = 2 Â 10À6).
Over the three weeks that the cultures were monitored, both
showed a slight drop in EH activity when normalized to protein
content and an increase in activity when normalized to cell counts.
Expression of the EH mRNA was also monitored over the same per-
iod by Q-PCR. Fig. 4B shows the ratio of cycle thresholds (Ct) for the
EH mRNA normalized to RuBisCO mRNA. EH mRNA levels of the
low toxin culture were either lower (Ct higher) or equal to (day
19 only) that of the high toxin culture. Even though gene expres-
sion in dinoflagellates is believed to occur principally at the trans-
lational level (Greenbaum et al., 2003), a significant difference in
the average ratio of Ct values was observed over the life of the cul-
ture (p = 0.02).
Scheme 4. Regioisomeric cyclizations of trans-4,5-epoxy-hexanol (10).
disfavored. The exception among the bacterial polyketides is lasa-
locid. In this polyether, the pyran ring arises from the endo-tet
cyclization of an intermediate epoxide. This reaction is catalyzed
by the EH Lsd19. Lsd19 is a bifunctional enzyme consisting of
two domains (A and B). Each domain is responsible for the cycliza-
tion of one of the rings with Lsd19B catalyzing the endo-tet cycliza-
tion of the terminal ring (Minami et al., 2011).
The ability of the K. brevis EH to perform cyclization was also
evaluated. The substrate trans-4,5-epoxy-hexanol 9 was prepared
as previously described (Coxon et al., 1973).This substrate may
undergo endo-tet cyclization to produce pyran 10, or exo-tet
cyclization to produce furan 11 (Scheme 4).The substrate was incu-
bated with the expressed EH in reaction buffer and the reaction
was monitored by GC–MS. Both possible products were identified
by GC–MS, by comparison with authentic samples however, nei-
ther the percent conversion nor the product ratio differed from
that of the buffer alone.
2.4. EH activity/expression and toxin content over the life of a culture
2.5. Additional library searches
Two cultures of K. brevis whose toxin content differs by ten-fold
were evaluated over a period of three weeks for EH activity and EH
mRNA levels. Both cultures are of the Wilson strain. However,
toxin production of one culture dropped significantly for reasons
Subsequent to these studies, three K. brevis reference transcrip-
tome libraries containing 86,580 (Wilson strain), 93,668 (strain
SP1) and 84,309 (strain SP3) predicted transcripts (Ryan et al.,
Average
Average
A. EH activity of two strains
B. Ratio of Ct forEH/RuBisCO
of K. brevis over life cycle
0.25
0.24
0.23
0.22
0.21
0.2
1.25
toxic
low toxic
toxic
low toxic
1.2
0.19
0.18
0.17
0.16
0.15
1.15
5
10
15
20
toxic low toxic
5
10
15
Time (days)
20
toxic
low
toxic
Time (days)
Fig. 4. (A) EH activity of low and high toxin K. brevis CFEs over a 21 day period normalized to protein content. (B) Ratio of Ct for EH/RuBisCO for low and high toxin K. brevis.
Please cite this article in press as: Sun, P., et al. Characterization of an epoxide hydrolase from the Florida red tide dinoflagellate, Karenia brevis. Phytochem-
istry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.11.002

P. Sun et al. / Phytochemistry xxx (2015) xxx–xxx
7
K. brevis EH1
K. brevis EH2
-------------------------------------------------MPLTDLKPFKLTVPDQQLDDLKYRISATR-WPDQLDSPANANWEYGTEITYLKSLAEYWRTTFSWRKQEEL 70
MLKIAGGICCILFVAFLSALPIIIDKLFIPTPPRSPEALLRSETEWGNEPEMSDIQPYKIAYDLKHLTDLRRRLLSWRPWSDTVPMAHSKNSSYGVNPKYMQSLISYWRDTYNWRSYEQK 120
::*::*:*::
::* **: *: : * *.* : . . * .**.: .*::** .*** *:.**. *:
▼
K. brevis EH1
K. brevis EH2
K. brevis EH2
LNAMPQFTAKLNGVTVHFVHQRCLSSPTAPALLICHGWPGSIFEFHKVLGPLAEN---FHVVAPSIPGYGFSEAPHKRGFNAPEAARLFHALMLELGYQEYVAQGGDWGSVITQCIGKLF 187
INSLGLKSTTIDGLRLVFHEVNGGKQ---KAILLLHGWPGSIHEFSKVLLKINASTQEYSIVCPSLPGFGFSSAPTEEGYSSVEMALTLLRLMAKLKYRTFVVQGGDWGSVVARAMAYID 237
:*::
::.::*: : * . . ..
*:*: *******.** ***
:
.
▼
: :*.**:**:***.** :.*:.: * *
:
** :* *: :*.********:::.:. :
K. brevis EH1
K. brevis EH2
PKQCKAIHINMPTAPRPKD---FDMSKLNPAEVEG—LRRTMAFNTYG----TGYQKISGTKPQTLSYGLNDSPIGLLAWMTEKFRDWSDCNGTLENAFTQDEILTNAMIYWVSQSIGSS 298
PVSVIGLHLNFFPVLPPPGTYVFDMFKVWAFDDENSKSRAKKLLDVNGLFQITGYFHQQATMPETLGFALSDSPAGLAAWILEKFQGWSDPDRPIEDKFRRDDLLTNIMLYWLTNTITSS 357
* . .:*:*: .. * .
▼
*** *: . : *.
* . ::. *
*** : ..* *:**.:.*.*** ** **: ***:.*** : .:*: * :*::*** *:**::::* **
▼
▼
K. brevis EH1
K. brevis EH2
MRFYYENIAVMPGTSKEMIELAAVKVSVPTAVAMFPKEIFFCPRAWCEQAYN-LQQYTRFEAGGHFAALEKPEELVKDIKDFFLRRLDLDKNQSRL 393
MRMYKEVVSLEA------ISVLYLPVTVPVGISLFPHEVLSPPDTVIKKHFTNIVQFTKHKQGGHFAALEEPKAFTDDLLAFLAKLPKSVPGTDTN 447
**:* * ::: .
*.: : *:**..:::**:*:: * : :: :. : *:*:.: ********:*: :..*: *: :
.
. .
Fig. 5. Alignment of two K. brevis mEHs. Catalytic residues are indicated by . and NC-loop and cap-loop are highlighted in grey.
2014) became available. tBLASTn analysis of these transcriptome
libraries against the 10 LEHs and the Emeliania huxleii LTA4 hydro-
lase were performed. Against the LEHs, a single transcript from the
SP1 transcriptome returned an Expect values of less than 10À5
(4 Â 10À6) against NanIB. However, BLASTx analysis against the
NCBI non-redundant protein database did not return any
significant homologies (10À6 or less). Against the E. huxleii LTA4
hydrolase, numerous transcripts with expect values ranging from
10À12 to10À141 were identified from all three libraries.
niger, the NC loop is located from Leu215 to Leu249 (35 residues)
whereas the cap-loop is located from Trp284 to Ser291 (8 resi-
dues). The lengths of the NC loops of the K. brevis EHs are 36 and
43 residues, whereas both cap-loops are 14 residues. mEHs are
characterized by having long to very long NC-loops (33 or more
residues) and short to medium cap-loops (8–36). The K. brevis
NC-loops and cap-loops would be considered very long and med-
ium in length, respectively. The length of these domains is believed
to govern substrate selectivity of the EH and this combination of
long/med NC-loop/cap-loop would suggest selectivity for aromatic
epoxides. The K. brevis EH1 appears to deviate from this generaliza-
tion as selectivity was observed for aliphatic over aromatic substi-
tuted epoxides. Like the K. brevis EH1, the mEH from A. niger is
unable to catalyze the hydrolysis of non-terminal epoxides
(Pedragosa-Moreau et al., 1996a). However, the mEH from A. niger
has a more pronounced enantioselectivity with relative rates of
hydrolysis of R:S p-nitrostyrene of 55:1 (Morisseau et al., 1999).
The cDNA sequence of the
a,b-epoxide hydrolase (EH1) was
used as a query for a nucleotide BLAST analysis of the three refer-
ence transcriptomes. Each of these transcriptome libraries
returned sequences (EH2) which were identical to each other
(Locus 28231 from SP1; Locus 35217 from SP3; Locus 51025 from
Wilson) having Expect values ranging from 10À79 to 10À84. Addi-
tionally, a single transcript, which was identical to the query
sequence, was identified in the SP3 library (SP3 Locus 73930).
When the newly identified sequence (EH2, Locus 28231 from
SP1) was used as a query to search the original EST library, a single
50 bp sequence was returned which exactly matched the query.
3. Concluding remarks
The alignment of these two
Fig. 5. These two hydrolases share 35% identity and 57% similarity.
,b Epoxide hydrolases exhibit two active site variations: the iden-
a,b-epoxide hydrolases is shown in
Data mining of four transcriptome libraries indicates that K. bre-
vis expresses two of the three main classes of epoxide hydrolases:
a
tity and location of the charge relay acid which may be either Asp
or Glu and may be located either after a central b-strand (b6) or
after b7: and the presence or absence of the second Tyr (Barth
et al., 2004). The two K. brevis EHs are similar in that each has
two active site Tyrs as well as a Glu located after b7. Within the
LTA4 hydrolase and the a,b fold type of epoxide hydrolases. Repre-
sentatives of the limonene type of epoxide hydrolase were not
identified in these libraries which is consistent with inhibition
assays performed on CFE. The involvement of limonene epoxide
hydrolases in the biosynthesis of bacterial polyethers has been
firmly established, and it has long been believed that the dinoflag-
ellate derived polyether ladders share a common biogenic origin
with bacterial polyethers including involvement of one or more
epoxide hydrolases. Thus the absence of LEH comes as somewhat
of a surprise. Given the high substrate specificity of the LTA4
hydrolases, their involvement in polyether ladder biosynthesis
a,b fold epoxide hydrolases, there are three classes which may be
broadly characterized by their highly variable N-terminal domains
(Barth et al., 2004): the soluble (or cytosolic) EH (sEH): the micro-
somal EH (mEH) and the plant EHs which lack this N-terminal
domain entirely. The mEHs are believed to be involved in the meta-
bolism of xenobiotics whereas the sEHs are thought to be involved
in metabolism of fatty acids (Morisseau and Hammock, 2005).
However, only mammalian genomes encode both sEH and mEH,
whereas yeast, fungi, insects and plants encode only one type
and the activity of the absent EH type may be assumed by the
seems unlikely. Two EHs of the
the transcriptome libraries.
a,b fold type were identified in
The selectivity of the expressed K. brevis EH for aliphatic epox-
ides coupled with the consistently higher level of EH activity in the
high toxin vs low toxin cultures of K. brevis is intriguing and sug-
gests the involvement of this EH in brevetoxin biosynthesis. The
expressed EH did not catalyze the cyclization of an epoxy alcohol
to either a furan or pyran. However, epoxy alcohol 9 may not suf-
ficiently resemble the native substrate. Lsd19 was also not capable
of catalyzing the cyclization of monoepoxy alcohol which was sig-
nificantly smaller than pre-lasalocid (Matsuura et al., 2010). A key
other. Both of the
a,b fold type EHs identified in K. brevis belong
to the mEH class (Barth et al., 2004; Pleiss et al., 2000). The N-ter-
minal domains of mEHs are characterized by a microsomal domain
which may or may not be preceded by an N-terminal membrane
anchor. The membrane anchor is found in mammalian and insect
mEHs, but not in EHs from yeast, fungi or bacteria and is also
absent in the two K. brevis mEHs. EHs may be further categorized
according to the length of two loops: the NC-loop which links
the N-terminal catalytic domain and the cap domain (the cap
domain contains the two catalytic Tyr) and the cap-loop, which
difference between the LEH and the
a,b fold type of EH is in the
mechanism. The LEH delivers a nucleophile, either a water mole-
cule, or in the case of the bacterial polyethers, an alcohol, whereas
links two
a
-helices (
a
6 and
a7) within the cap domain (Barth
the a,b fold type EH initially forms a covalent adduct with the
et al., 2004). The length of these loops may be determined from
the pair wise alignment with the EH from A. niger (also an mEH)
or by inspection of a homology model constructed using the A.
niger EH crystal structure as template (Arand et al., 1999). In A.
nucleophilic Asp residue, which must ultimately be cleaved. It is
therefore somewhat difficult to rationalize the involvement of
the
a,b fold type EH in PE ladder biosynthesis. Vilotijevic
demonstrated that non-enzymatic endo selective epoxide-opening
Please cite this article in press as: Sun, P., et al. Characterization of an epoxide hydrolase from the Florida red tide dinoflagellate, Karenia brevis. Phytochem-
istry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.11.002

8
P. Sun et al. / Phytochemistry xxx (2015) xxx–xxx
cascade reactions were promoted in water when templated by a
single tetrahydropyran ring (Vilotijevic and Jamison, 2007). A vic-
inal diol formed by the enzymatic hydrolysis of an epoxide precur-
Inc.) after incubation for 1 h at room temperature. Blank control
and boiled homogenate were also tested; typical activity of homo-
genate was more than twice that of boiled homogenate.
sor by an
a,b fold type EH, followed by cyclization to the
brevetoxin lactone ring, could provide the template for a water
promoted epoxide-opening cascade. It has further been suggested
that an epoxidase could catalyze both epoxidation and cyclization
for PE ladder formation without need for EH involvement
(Gallimore and Spencer, 2006).
4.6. Probe synthesis
The fluorescent probe 5a was prepared as previously described
(Jones et al., 2005). A second probe 5b was prepared in a procedure
identical to that of 5a except that (E)-4-phenylbut-3-enoic acid
was replaced with (E)-pent-2-enoic acid. Both alkene and epoxide
were purified by radial chromatography on a chromatotron (Har-
rison Research Corp.) using plates coated with silica gel 60 PF254
containing gypsum, using 5:1 hexane/EtOAc. Spectroscopic data
for probe 5a were consistent with literature data (Jones et al.,
2005). 1H NMR and 13C NMR data were acquired on a Bruker
Avance spectrophotometer (400 MHz for proton). 1H NMR spectra
were referenced internally to the residual proton resonance in
CDCl3 (d = 7.26 ppm), or with tetramethylsilane (TMS,
d = 0.00 ppm) as internal standard. Chemical shifts are reported
as parts per million (ppm) in the d scale downfield from TMS. 13C
NMR spectra were recorded with continuous proton decoupling.
Chemical shifts are reported in ppm from TMS with the solvent
as the internal reference (CDCl3, d = 77.0 ppm). Accurate mass
spectra were acquired using a Bruker Daltonics – ultrOTOF-Q,
+ESI-Q-q-TOF.
4. Experimental
4.1. Culture methods
K. brevis culture (Wilson strains) were obtained from Mote Mar-
ine Laboratory (Sarasota, Florida) and maintained in L1-Si medium,
with the exception that the NH 15 vitamin supplement (Gates and
Wilson, 1960) replaced the L-1 supplement, in a growth chamber
at ꢀ20 °C under 35
l
mol photons mÀ2 sÀ1. Growth was monitored
by counting a 1:10 dilution of culture in Z pak reagent using a
Beckman Z-series Coulter Counter with aperture size between 10
and 30 lm according to the manufacturer’s instructions.
4.2. HPLC-MS/MS analysis of brevetoxins
Brevetoxins were extracted from K. brevis culture using a previ-
ously described method (Pierce and Henry, 2008). Quantitative
analysis of brevetoxins and conjugates were conducted using a
TSQ Quantum Access/ESI/Accela UHPLC for LC–MS/MS. The LC sys-
tem consisted of a Thermo Electron Accela UHPLC pumping sys-
tem, coupled with the Accela Autosampler and Degasser. The
analytical column of consisted of a Kinetex C-18, reversed phase,
2.6um particle size with dimensions of 100 mm  2.1 mm (Phe-
nomenex, Torrance, CA). Mass Spectrometry was obtained using
a TSQ Quantum Access MS/MS system, calibrated with brevetoxin
standards from Marbionic (Wilmington, NC).
4.6.1. Alkene probe 3b
(65%, colorless oil): 1H NMR (CDCl3, 400 MHz): d 7.94 (s, 1H),
7.80 (t, 2H, J = 8.4 Hz), 7.53 (m, 1H), 7.15–7.23 (m, 2H), 6.57 (s,
1H), 5.47–5.66 (m, 2H), 3.94 (s, 3H), 3.13 (m, 2H), 2.05 (m, 2H),
0.98 (t, 3H, J = 7.6 Hz). 13C NMR (CDCl3, 100 MHz): d 170.5, 159.1,
137.8, 135.5, 130.0, 128.4, 128.3, 128.0, 126.8, 125.1, 123.8,
120.1, 119.1, 116.42, 105.9. HRMS (+ESI): calculated for
C
19H19NO3 [M + Na]+ 332.1257, found 332.1250.
4.6.2. Epoxide probe 5b
4.3. Preparation of K. brevis homogenate
(11%, yield yellow oil): 1H NMR (CDCl3, 400 MHz): d 8.26 (s, 1H),
7.80 (t, 2H, J = 8.8 Hz), 7.52 (m, 1H), 7.15–7.26 (m, 2H), 6.60 (s, 1H),
3.94 (s, 3H), 3.08 (m, 1H), 2.75 (m, 1H), 2.66 (m, 2H), 1.59 (m, 2H),
0.97 (t, 3H, J = 5.6 Hz). 13C NMR (CDCl3, 100 MHz): d 168.8, 159.0,
135.4, 129.9, 128.2, 128.0, 127.8, 126.3, 125.0, 120.0, 116.0,
105.8, 63.4, 59.5, 55.4, 52.9, 37.3, 24.7, 9.6. HRMS (+ESI): calculated
for C19H19NO4 [M + Na]+ 348.1167, found 348.1206.
K. brevis culture (200 ml) was concentrated by centrifugation
(5 min at 450Âg) and the supernatant discarded. The cells were
resuspended in phosphate buffer (800 ll, 50 mM pH 7.0) and vor-
texed for 1 min. This suspension was centrifuged (14,000Âg for
10 min) and the pellet was discarded. Homogenates were analyzed
for protein concentration and glucose-6-phosphate dehydrogenase
activity as described below. Typical protein concentrations ranged
from (0.25 to 0.5 mg/ml).
4.6.3. Synthesis of trans-4,5-epoxy-hexanol (9), 2-methyltetrahydro-
2H-pyran-3-ol (10) and 1-(tetrahydrofuran-2-yl)ethanol (11)
Syntheses were performed as previously described (Coxon et al.,
4.4. Bradford assay
1973). Trans-4-hexen-1-ol (500 ll, 5.88 mmol) was mixed with
Five concentrations of BSA (0 mg/ml, 1 mg/ml, 0.5 mg/ml,
0.25 mg/ml, 0.0625 mg/ml) in phosphate buffer (50 mM, pH 7.0)
were prepared as standards. Commassie Protein Assay Reagent
NaHCO3 (1.775 g), acetone (5.6 ml) and EtOAc (20 ml). Oxone
(2.6 g, 17.1 mmol) was dissolved in H2O (18 ml) and added drop-
wise to the trans-4-hexen-1-ol solution over a period of 40 min.
The reaction was stirred for another hour. The organic solvents
were evaporated in vacuuo. The residue was extracted with EtOAc
(20 ml). The organic layer was washed twice with an equal volume
of brine, dried over anhydrous Na2SO4, filtered and the solvent
evaporated in vacuuo to yield trans-4,5-epoxy-hexanol 9 which
was used in the next step without further purification. Spectro-
scopic data for 9 were consistent with literature data. Trans-4,5-
epoxy-hexanol 9 was treated with a solution of boron trifluoride
etherate in Et2O as described previously to produce a mixture of
10 and 11. Analysis by GC–MS showed two products with the same
m/z (116) in a ratio of 5:1 which were assigned as 10 and 11
respectively, based on previous reports (Coxon et al., 1973).
(250 ll) was added to microplate wells and 20 ll of standard or
unknown sample were added. Absorbance was measured at
595 nm (SynergyTM 2, Bioteck Instrument, Inc.) after incubation
for 5 min at room temperature. Each protein solution was assayed
in duplicate.
4.5. Glucose-6-phosphate dehydrogenase activity
In order to ensure activity of K. brevis cell free extract (150
ll)
was mixed with assay buffer (50 l, 31 mM phosphate buffer,
l
7.5 mM glucose-6-phophate, 2.5 mM NADP, 38 mM MgCl2). Absor-
bance at 340 nm was determined (SynergyTM 2, Bioteck Instrument,
Please cite this article in press as: Sun, P., et al. Characterization of an epoxide hydrolase from the Florida red tide dinoflagellate, Karenia brevis. Phytochem-
istry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.11.002

P. Sun et al. / Phytochemistry xxx (2015) xxx–xxx
9
4.7. Cloning and expression of K. brevis EH1
0.25 lm initial temperature 80 °C for 1 min/increase to 300 °C at
10 °C/minute/hold for 7 min) and compared to standard curve of
either the diol product or starting epoxide with the internal
standard. Each reaction was performed in duplicate.
Contig MGID2034061 of a K. brevis EST library (Lidie et al., 2005)
was synthesized (after optimization of codon usage for E. coli
expression) and cloned into pUC 57 by Genescript. The insert was
subcloned into pET30a+ vector for protein expression: Specific
primers (Forward: CACAATCATATGCCGCTGACGGA, Reverse:
AATAAGCTTTAATGATGATGATGATGATGCAGACGGCTCTGATTTTTAT)
were designed to amplify the codon optimized K. brevis EH. Primers
incorporated HindIII (reverse primer) and NdeI (forward primer)
restriction sites for directional cloning and 6 histidine codons
(reverse primer). The PCR product was ligated with pET30a+ using
T4 DNA ligase (Invitrogen) according to the manufacturer’s instruc-
tions. The newly constructed plasmid was transformed into expres-
sion strain BL21(DE3) by electroporation. Recombinant cells were
grown following manufacturer’s guide (Invitrogen). White colonies
4.8.3. Cyclization reaction
Trans-4,5-epoxy-hexanol 9 (0.4 M stock solution in DMSO
diluted to a final concentration of 4 mM) was incubated with
EH1 protein (0.1 mg/ml in dialysis buffer) or dialysis buffer alone
at 30 °C for 1–3 h. The reaction was extracted using an equal vol-
ume of CH2Cl2. The organic extracts were evaporated to dryness
and reconstituted in CH2Cl2 to ꢀ20 ppm and analyzed by GC–MS
as previously described.
4.9. Determination of Km and kcat
were picked and grownin LB brothwith kanamycin(50 lg/ml). Plas-
mids were extracted using QIAprep Spin Miniprep Kit (Qiagen)
according to the manufacturer’s instruction and sent to Eurofins
Genomics for sequencing to insure fidelity of insert sequence. The
recombinant E. coli was inoculated into LB broth with kanamycin
Specific enzyme activity was determined by fluorescent assay
(5a and 5b) or by GC–MS. EH1 protein (1.926 lM) was incubated
with various substrate concentrations. For fluorescent probes (5a,
5b), the final concentrations of substrate were 0, 25, 50, 75 and
(50
(5 ml) was diluted into fresh LB broth (500 ml) with kanamycin
(50 g/ml), shaken at 37 °C (250 rpm for 2–3 h). IPTG (1 mM final
lg/ml) and shaken at 37 °C overnight. The overnight culture
100 lM. For other substrates, the final concentrations of substrate
were 4, 6, 8, and 10 mM. The assays were performed as previous
described. Lineweaver–Burk plots were obtained using initial
velocities of substrate and their corresponding concentration, and
Km, Vmax and Vmax/Km were determined from the generated
equation. These experiments were performed in duplicate.
l
concentration) was added when the OD600 reached to 0.6–0.8. After
shaking another 4–6 h, the culture was centrifuged (5000Âg for
10 min). Lysozyme (2 ll, 50 mg/ml) and DNAse I (2 ll, 2500 unit/
ml) as well as Bacterial Protein Extraction reagent (B-PER, Thermal
Scientific, 4 ml/ gram of cells) were added to resuspend the pellets.
After incubationat room temperaturefor 10 min, thelysate was cen-
trifuged (15000Âg for 5 min) to separate insoluble proteins. The
supernatant was loaded to a column of 1 ml Ni–NTA agarose (Qia-
gen). The column was washed (20 mM phosphate, 0.5 M NaCl,
20 mM imidazole, 10% glycerol, pH 7.9) until no more protein was
eluted (as determined by A280) in the wash (typically 10 ml). This
was followed by elution buffer (10 ml, 20 mM phosphate, 0.5 M
NaCl, 250 mM imidazole, 10% glycerol, pH 7.9). The purified EH
was dialyzed (MW cutoff 12–14 kDa) in phosphate buffer (50 mM
sodium phosphate, pH 7.0) at 0–4 °C, overnight, flash frozen in liquid
N2 and stored at À80 °C until use. Final protein concentrations ran-
ged from 0.2 to 0.3 mg/ml.
4.10. Inhibition assay
IC50 values could be obtained by either using fixed substrate
concentration with different concentration of inhibitors or with
several different substrate concentrations with a fixed inhibitor
concentration. IC50 values were determined using fluorescent sub-
strate 5b (100 lM). EH (0.1 mg/ml) was incubated with inhibitors
for 2 min in phosphate buffer (50 mM, pH 7.0) at 30 °C prior to
addition of substrate. EH activity was measure as previously
described by monitoring fluorescence. Assays were performed in
duplicate. By definition, the IC50 value is concentration of inhibitor
that reduces EH activity by 50%. Six concentrations of inhibitors
(0, 2, 4, 8, 12, 16 mM) were assayed to obtain an inhibition curve
of inhibition/inhibitor concentration.
4.8. EH assay
4.8.1. Fluorescent assay
Fluorescent probe 3a/b, 5a/b (2
100 M) was added to K. brevis CFE (198
(198
l
l, 10 mM, final concentration
4.11. Q-PCR
l
l
l
l) or purified EH
l) in duplicate. Fluorescence was monitored at 330 nm
4.11.1. RNA isolation
EX/465 nm EM (SynergyTM 2, Bioteck Instrument, Inc.) every
5 min at 30 °C. The concentration of reporter 4 was calculated from
a calibration curve prepared by measuring fluorescence of 7-meth-
K. brevis culture (100 ml) was centrifuged (450Âg for 5 min) at
20 °C. The pellet was resuspended in H2O (100 ll). Denaturing
solution (600 ll, 5 M guanidium thiocyanate, 50 mM Tris pH 8,
oxy-2-naphthaldehyde (0, 3.1, 12, 25, 50
lM in phosphate buffer
25 mM sodium citrate, 0.5% w/v Sarkosyl, 2% PEG, 0.1 M DTT) were
(50 mM phosphate, pH 7.0). Fluorescence readings were normal-
added. After standing at room temperature for 10 min, NaOAc
ized to protein concentration or cell number.
(60 ll, 3 M, pH 5.2) was added and the mixture was vortexed. An
equal volume of phenol: CHCl3: isoamyl alcohol (125:24:1, pH
4.3), was added and the mixture centrifuged at (14,000Âg for
15 min at 4 °C). The top (aqueous) phase was transferred to a
new tube followed by addition of an equal volume CHCl3: isoamyl
alcohol (24:1) and the mixture was vortexed. After centrifugation
(14,000Âg for 15 min) at 4 °C, the top (aqueous) phase was trans-
ferred to new tube which contains equal volume of cold EtOH. RNA
was allowed to precipitate at À20° overnight. The mixture was
centrifuged at 4 °C, for (30 min at 14,000 Â g) and the pellet was
4.8.2. GC–MS assay
Substrate (4 mM to 10 mM final concentrations), internal stan-
dard, 1-naphthol (concentrations ranging from 0.4 to 4 mM; the
concentration was optimized for each substrate to a relative
response ratio of 1:5–5:1 for substrate/internal standard in the
GC–MS), and purified EH (ꢀ0.1 mg) was combined in phosphate
buffer (50 mM phosphate, pH 7.0). Aliquots were removed every
5 min and extracted using an equal volume of CH2Cl2. Samples
were evaporated to dryness and reconstituted in CH2Cl2 to
ꢀ20 ppm. Samples were analyzed by GC–MS (Hewlett Packard
6890 with Alltech 30 m, DB5, capillary column, 0.25 mm (ID)
washed with cold EtOH: H2O (700
the pellet was reconstituted in H2O (50
using the Nanodrop 2000 (Thermo Scientific).
l
l, 70:30, v/v). After air-drying,
ll). RNA was quantitated
Please cite this article in press as: Sun, P., et al. Characterization of an epoxide hydrolase from the Florida red tide dinoflagellate, Karenia brevis. Phytochem-
istry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.11.002

10
P. Sun et al. / Phytochemistry xxx (2015) xxx–xxx
Greenbaum, D., Colangelo, C., Williams, K., Gerstein, M., 2003. Comparing protein
4.11.2. RT-PCR
abundance and mRNA expression levels on a genomic scale. Genome Biol. 4, 117.
Gruber, K., Zhou, B., Houk, K.N., Lerner, R.A., Shevlin, C.G., Wilson, I.A., 1999.
Structural basis for antibody catalysis of a disfavored ring closure reaction.
Biochemistry 38, 7062–7074.
Using the isolated RNA (ꢀ1
lg) as template and oligo dT as pri-
mer (1 ll, 50 mM), cDNA was synthesized using the AMV First
Strand cDNA Synthesis kit (Life Technologies) according to the
manufacturer’s instructions. Reverse transcriptase reactions were
diluted 1:5 in H2O. PCR reactions were carried out using the newly
synthesized cDNA (1
reaction), PCR mix (25
Harvey, B.M., Mironenko, T., Sun, Y., Hong, H., Deng, Z., Leadlay, P.F., Weissman, K.J.,
Haydock, S.F., 2007. Insights into polyether biosynthesis from analysis of the
nigericin biosynthetic gene cluster in Streptomyces sp. DSM4137. Chem. Biol. 14,
703–714.
Jones, P.D., Wolf, N.M., Morisseau, C., Whetstone, P., Hock, B., Hammock, B.D., 2005.
Fluorescent substrates for soluble epoxide hydrolase and application to
inhibition studies. Anal. Biochem. 343, 66–75.
l
l
of the diluted reverse transcriptase
l, iQ SYBR Green Supermix, BioRad),
of each primer (50 mM). EH primer set: forward;
l
1
l
l
CAAGGTGCTGGGTCCTTTGG. Reverse; TTCCGGAGCGTTGAAACCAC.
RuBisCO primers set: forward; GCAGCCTTTTATGCGGTATCG.
Reverse; CACCAGCCATTCGCATCCATTT (Yoon et al., 2002). PCR
reactions were monitored using the DNA Engine Opticon (MJ
Research) which can detect Sybr green (excitation: 497 nm, emis-
sion: 520 nm) fluorescence. Cycling conditions: 94 °C denaturation
for 1 min, annealing for 45 s at 60 °C, extension 72 °C (RuBisCO pri-
mer set only, no extension time for EH primers), repeat 40 cycles.
The efficiency of PCR mix was determined by titrating the template
cDNA concentrations by dilution of the RT reactions (1:5, 1:25 and
1:125) in H2O. Efficiencies were calculated to be ꢀ90–95%. Thresh-
old was set as 10 standard deviations above mean fluorescence
generated during 3–15 baseline cycles.
Kim, I.-H., Morisseau, C., Watanabe, T., Hammock, B.D., 2004. Design, synthesis, and
biological activity of 1, 3-disubstituted ureas as potent inhibitors of the soluble
epoxide hydrolase of increased water solubility. J. Med. Chem. 47, 2110–2122.
Lee, M.S., Repeta, D.J., Nakanishi, K., Zagorski, M.G., 1986. Biosynthetic origins and
assignments of carbon 13 NMR peaks of brevetoxin B. J. Am. Chem. Soc. 108,
7855–7856.
Lee, M.S., Qin, G., Nakanishi, K., Zagorski, M.G., 1989. Biosynthetic studies of
brevetoxins, potent neurotoxins produced by the dinoflagellate Gymnodinium
breve. J. Am. Chem. Soc. 111, 6234–6241.
Lidie, K.B., Ryan, J.C., Barbier, M., Van Dolah, F.M., 2005. Gene expression in Florida
red tide dinoflagellate Karenia brevis: analysis of an expressed sequence tag
library and development of DNA microarray. Mar. Biotechnol. 7, 481–493.
Matsuura, Y., Shichijo, Y., Minami, A., Migita, A., Oguri, H., Watanabe, M., Tokiwano,
T., Watanabe, K., Oikawa, H., 2010. Intriguing substrate tolerance of epoxide
hydrolase Lsd19 involved in biosynthesis of the ionophore antibiotic lasalocid
A. Org. Lett. 12, 2226–2229.
Minami, A., Migita, A., Inada, D., Hotta, K., Watanabe, K., Oguri, H., Oikawa, H., 2011.
Enzymatic epoxide-opening cascades catalyzed by a pair of epoxide hydrolases
in the ionophore polyether biosynthesis. Org. Lett. 13, 1638–1641.
Minami, A., Oguri, H., Watanabe, K., Oikawa, H., 2013. Biosynthetic machinery of
ionophore polyether lasalocid: enzymatic construction of polyether skeleton.
Curr. Opin. Chem. Biol. 17, 555–561.
Acknowledgments
This work was supported by the National Institute of Environ-
mental Health Sciences (NIEHS) Grant S11 ES11181 (to K.R.);
bridge funding from the FIU Division of Research and the College
of Arts and Sciences (to K.R.); NIH grants 5SC1HD063059-04 and
1R15GM109254-01A1 (to F. L.) and a graduate assistantship from
the FIU College of Arts and Sciences (to P.S.).
Monroe, E.A., Van Dolah, F.M., 2008. The toxic dinoflagellate Karenia brevis encodes
novel type I-like polyketide synthases containing discrete catalytic domains.
Protist 159, 471–482.
Morisseau, C., Hammock, B.D., 2005. Epoxide hydrolases: mechanisms, inhibitor
designs, and biological roles. Annu. Rev. Pharmacol. Toxicol. 45, 311–333.
Morisseau, C., Archelas, A., Guitton, C., Faucher, D., Furstoss, R., Baratti, J.C., 1999.
Purification and characterization of a highly enantioselective epoxide hydrolase
from Aspergillus niger. Eur. J. Biochem. 263, 386–395.
Nakanishi, K., 1985. The chemistry of brevetoxins: a review. Toxicon 23, 473–479.
Oliynyk, M., Stark, C.B.W., Bhatt, A., Jones, M.A., Hughes-Thomas, Z.A., Wilkinson, C.,
Oliynyk, Z., Demydchuk, Y., Staunton, J., Leadlay, P.F., 2003. Analysis of the
biosynthetic gene cluster for the polyether antibiotic monensin in Streptomyces
cinnamonensis and evidence for the role of monB and monC genes in oxidative
cyclization. Mol. Microbiol. 49, 1179–1190.
Pedragosa-Moreau, S., Archelas, A., Furstoss, R., 1996. Microbiological
transformation 32: use of epoxide hydrolase mediated biohydrolysis as a way
to enantiopure epoxides and vicinal diols: application to substituted styrene
oxide derivatives. Tetrahedron 52, 4593–4606.
Pierce, R.H., Henry, M.S., 2008. Harmful algal toxins of the Florida red tide (Karenia
brevis): natural chemical stressors in South Florida coastal ecosystems.
Ecotoxicology 17 (7), 623–631.
Pleiss, J., Fischer, M., Peiker, M., Thiele, C., Schmid, R.D., 2000. Lipase engineering
database: understanding and exploiting sequence–structure–function
relationships. J. Mol. Catal. B: Enzym. 10, 491–508.
Rudberg, P.C., Tholander, F., Andberg, M., Thunnissen, M.M., Haeggström, J.Z., 2004.
Leukotriene A4 hydrolase identification of a common carboxylate recognition
site for the epoxide hydrolase and aminopeptidase substrates. J. Biol Chem. 279,
27376–27382.
Ryan, D.E., Pepper, A.E., Campbell, L., 2014. De novo assembly and characterization of
the transcriptome of the toxic dinoflagellateKareniabrevis. BMC Genomics 15, 888.
Shichijo, Y., Migita, A., Oguri, H., Watanabe, M., Tokiwano, T., Watanabe, K., Oikawa,
H., 2008. Epoxide hydrolase Lsd19 for polyether formation in the biosynthesis
of lasalocid A: direct experimental evidence on polyene-polyepoxide
hypothesis in polyether biosynthesis. J. Am. Chem. Soc. 130, 12230–12231.
Smith, L., Hong, H., Spencer, J.B., Leadlay, P.F., 2008. Analysis of specific mutants in
the lasalocid gene cluster: evidence for enzymatic catalysis of a disfavoured
polyether ring closure. ChemBioChem 9, 2967–2975.
Snyder, R., Gibbs, P., Palacios, A., Abiy, L., Dickey, R., Lopez, J.V., Rein, K., 2003.
Polyketide synthase genes from marine dinoflagellates. Mar. Biotechnol. 5, 1–12.
Snyder, R.V., Guerrero, M.A., Sinigalliano, C.D., Winshell, J., Perez, R., Lopez, J.V., Rein,
K.S., 2005. Localization of polyketide synthase encoding genes to the toxic
dinoflagellate Karenia brevis. Phytochemistry 66, 1767–1780.
Sun, Y., Zhou, X., Dong, H., Tu, G., Wang, M., Wang, B., Deng, Z., 2003. A complete
gene cluster from Streptomyces nanchangensis NS3226 encoding biosynthesis of
the polyether ionophore nanchangmycin. Chem. Biol. 10, 431–441.
Tzeng, H.-F., Laughlin, L.T., Lin, S., Armstrong, R.N., 1996. The catalytic mechanism of
microsomal epoxide hydrolase involves reversible formation and rate-limiting
hydrolysis of the alkyl-enzyme intermediate. J. Am. Chem. Soc. 118, 9436–9437.
Van Dolah, F.M., Doucette, G.J., Gulland, F.M., Rowles, T.L., Bossart, G.D., 2003. 10
impacts of algal toxins on marine mammals. Toxicol. Mar. Mammal 3, 247–269.
Vilotijevic, I., Jamison, T.F., 2007. Epoxide-opening cascades promoted by water.
Science 317 (5842), 1189–1192.
References
Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic local
alignment search tool. J. Mol. Biol. 215, 403–410.
Arand, M., Hemmer, H., Durk, H., Baratti, J., Archelas, A., Furstoss, R., Oesch, F., 1999.
Cloning and molecular characterization of a soluble epoxide hydrolase from
Aspergillus niger that is related to mammalian microsomal epoxide hydrolase.
Biochem. J. 344, 273–280.
Arand, M., Hallberg, B.M., Zou, J., Bergfors, T., Oesch, F., van der Werf, M.J., de Bont, J.
A., Jones, T.A., Mowbray, S.L., 2003. Structure of Rhodococcus erythropolis
limonene-1,2-epoxide hydrolase reveals
2583–2592.
a novel active site. EMBO J. 22,
Baldwin, J.E., 1976. Rules for ring closure. J. Chem. Soc. Chem. Commun., 734–736
Barth, S., Fischer, M., Schmid, R.D., Pleiss, J., 2004. Sequence and structure of epoxide
hydrolases: a systematic analysis. Proteins: Struct. Funct. Bioinf. 55, 846–855.
Blanc, G., Agarkova, I., Grimwood, J., Kuo, A., Brueggeman, A., Dunigan, D.D., Gurnon,
J., Ladunga, I., Lindquist, E., Lucas, S., 2012. The genome of the polar eukaryotic
microalga Coccomyxa subellipsoidea reveals traits of cold adaptation. Genome
Biol. 13, R39.
Bourdelais, A.J., Jacocks, H.M., Wright, J.L., Bigwarfe, P.M., Baden, D.G., 2005. A new
polyether ladder compound produced by the dinoflagellate Karenia brevis. J.
Nat. Prod. 68, 2–6.
Chou, H.N., Shimizu, Y., 1987. Biosynthesis of brevetoxins. Evidence for the mixed
origin of the backbone carbon chain and possible involvement of dicarboxylic
acids. J. Am. Chem. Soc. 109, 2184–2185.
Coxon, J.M., Hartshorn, M.P., Swallow, W., 1973. Study of hydroxyl participation in
acyclic epoxide systems-acid-catalyzed rearrangements of trans-3,4-
epoxypentan-1ols, and cis-3,4-epoxypentan-1-ols, 4,5-epoxyhexan-1-ols, and
5,6-epoxyheptan-1-ols. Aust. J. Chem. 26, 2521–2526.
Demydchuk, Y., Sun, Y., Hong, H., Staunton, J., Spencer, J.B., Leadlay, P.F., 2008.
Analysis of the tetronomycin gene cluster: insights into the biosynthesis of a
polyether tetronate antibiotic. ChemBioChem 9, 1136–1145.
Fleming, L.E., Kirkpatrick, B., Backer, L.C., Bean, J.A., Wanner, A., Dalpra, D., Tamer, R.,
Zaias, J., Cheng, Y.S., Pierce, R., 2005. Initial evaluation of the effects of
aerosolized Florida red tide toxins (brevetoxins) in persons with asthma.
Environ. Health Perspect., 650–657
Flewelling, L.J., Naar, J.P., Abbott, J.P., Baden, D.G., Barros, N.B., Bossart, G.D., Bottein,
M.-Y.D., Hammond, D.G., Haubold, E.M., Heil, C.A., 2005. Brevetoxicosis: red
tides and marine mammal mortalities. Nature 435, 755–756.
Gallimore, A.R., Spencer, J.B., 2006. Stereochemical uniformity in marine polyether
ladders—implications for the biosynthesis and structure of maitotoxin. Angew.
Chem. Int. Ed. 45, 4406–4413.
Gates, J.A., Wilson, W.B., 1960. The toxicity of Gonyaulax monilata Howell to Mugil
cephalus. Limnol. Oceanogr. 5, 171–174.
Please cite this article in press as: Sun, P., et al. Characterization of an epoxide hydrolase from the Florida red tide dinoflagellate, Karenia brevis. Phytochem-
istry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.11.002

P. Sun et al. / Phytochemistry xxx (2015) xxx–xxx
11
Yamada, T., Morisseau, C., Maxwell, J.E., Argiriadi, M.A., Christianson, D.W.,
Hammock, B.D., 2000. Biochemical evidence for the involvement of tyrosine
in epoxide activation during the catalytic cycle of epoxide hydrolase. J. Biol.
Chem. 275, 23082–23088.
Yamazaki, M., Izumikawa, M., Tachibana, K., Satake, M., Itoh, Y., Hashimoto, M.,
2012. Origins of oxygen atoms in a marine ladder-frame polyether: evidence of
monooxygenation by 18O-labeling and using tandem mass spectrometry. J. Org.
Chem. 77, 4902–4906.
Yoon, H.S., Hackett, J.D., Bhattacharya, D., 2002. A single origin of the peridinin- and
fucoxanthin-containing plastids in dinoflagellates through tertiary
endosymbiosis. Proc. Natl. Acad. Sci. 99, 11724–11729.
Zou, J., Hallberg, B.M., Bergfors, T., Oesch, F., Arand, M., Mowbray, S.L., Jones, T.A.,
2000. Structure of Aspergillus niger epoxide hydrolase at 1.8 Å resolution:
implications for the structure and function of the mammalian microsomal class
of epoxide hydrolases. Structure 8, 111–122.
Please cite this article in press as: Sun, P., et al. Characterization of an epoxide hydrolase from the Florida red tide dinoflagellate, Karenia brevis. Phytochem-
istry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.11.002