Accepted Manuscript Title: Characterization of Glutathione-S-transferases in zebrafish (Danio rerio) Author: Branka Glisic Ivan Mihaljevic Marta Popovic Roko Zaja Jovica Loncar Karl Fent Radmila Kovacevic Tvrtko Smital PII: DOI: Reference:

S0166-445X(14)00316-6 http://dx.doi.org/doi:10.1016/j.aquatox.2014.10.013 AQTOX 3952

To appear in:

Aquatic Toxicology

Received date: Revised date: Accepted date:

4-7-2014 15-10-2014 21-10-2014

Please cite this article as: Glisic, B., Mihaljevic, I., Popovic, M., Zaja, R., Loncar, J., Fent, K., Kovacevic, R., Smital, T.,Characterization of Glutathione-S-transferases in zebrafish (Danio rerio), Aquatic Toxicology (2014), http://dx.doi.org/10.1016/j.aquatox.2014.10.013 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Characterization of Glutathione-S-transferases in zebrafish (Danio rerio) Authors: BRANKA GLISICa, IVAN MIHALJEVICb, MARTA POPOVICb, ROKO ZAJAb, c, JOVICA LONCARb, KARL FENTd, e, RADMILA KOVACEVICa, TVRTKO SMITALb* a

Laboratory for Ecotoxicology, Department of Biology and Ecology, University of Novi Sad,

ip t

Faculty of Sciences, Novi Sad, Serbia b

Laboratory for Molecular Ecotoxicology, Division for Marine and Environmental Research,

Ruđer Bošković Institute, Zagreb, Croatia c

cr

Sir William Dunn School of Pathology, University Oxford, Oxford, England (current address)

d

University of Applied Sciences Northwestern Switzerland, School of Life Sciences, Muttenz,

us

Switzerland e

Swiss Federal Institute of Technology (ETHZ), Department of Environmental System

an

Sciences, Zürich, Switzerland

M

These authors contributed equally to this work

Keywords: Glutathione-S-transferases; zebrafish; phylogenetic analysis; mRNA expression;

ed

enzyme kinetics; functional characterization

Tvrtko Smital, PhD

pt

*Corresponding author

Laboratory for Molecular Ecotoxicology

Ac ce

Division for Marine and Environmental Research Ruđer Bošković Institute Bijenička 54

10 000 Zagreb, CROATIA Tel.:

**385 1 45 61 039

E-mail: [email protected]

1 Page 1 of 42

Abstract Glutathione-S-transferases (GSTs) are one of the key enzymes that mediate phase II of cellular detoxification. The aim of our study was a comprehensive characterization of GSTs in zebrafish (Danio rerio) as an important vertebrate model species frequently used in environmental research.

ip t

A detailed phylogenetic analysis of GST superfamily revealed 27 zebrafish gst genes. Further insights into the orthology relationships between human and zebrafish GSTs/Gsts were obtained

cr

by the conserved synteny analysis. Expression of gst genes in six tissues (liver, kidney, gills, intestine, brain and gonads) of adult male and female zebrafish was determined using qRT-PCR.

us

Functional characterization was performed on 9 cytosolic Gst enzymes after overexpression in E.

an

coli and subsequent protein purification. Enzyme kinetics was measured for GSH and a series of model substrates. Our data revealed ubiquitously high expression of gstp, gstm (except in liver),

M

gstr1, mgst3a and mgst3b, high expression of gsto2 in gills and ovaries, gsta in intestine and testes, gstt1a in liver, and gstz1 in liver, kidney and brain. All zebrafish Gsts catalyzed the conjugation of GSH to model GST substrates 1-chloro-2,4-dinitrobenzene (CDNB) and

ed

monochlorobimane (MCB), apart from Gsto2 and Gstz1 that catalyzed GSH conjugation to dehydroascorbate (DHA) and dichloroacetic acid (DCA), respectively. Affinity towards CDNB

pt

varied from 0.28 mM (Gstp2) to 3.69 mM (Gstm3), while affinity towards MCB was in the range of 5

Ac ce

µM (Gstt1a) to 250 µM (Gstp1). Affinity toward GSH varied from 0.27 mM (Gstz1) to 4.45 mM (Gstt1a). Turnover number for CDNB varied from 5.25 s-1 (Gstt1a) to 112 s-1 (Gstp2). Only Gst Pi enzymes utilized ethacrynic acid (ETA). We suggest that Gstp1, Gstp2, Gstt1a, Gstz1, Gstr1, Mgst3a and Mgst3b have important role in the biotransformation of xenobiotics, while Gst Alpha, Mu, Pi, Zeta and Rho classes are involved in the crucial physiological processes. In summary, this study provides the first comprehensive analysis of GST superfamily in zebrafish, presents new insight into distinct functions of individual Gsts, and offers methodological protocols that can be used for further verification of interaction of environmental contaminants with fish Gsts.

2 Page 2 of 42

Introduction

The cellular detoxification system includes four temporally and spatially distinct phases. Phase zero (0) includes uptake of xenobiotics by membrane transport proteins; phase I denotes enzymatic bioactivation of parent compounds through oxidation-reduction reactions; phase II

ip t

includes enzyme(s) mediated conjugation of phase I metabolites or parent compounds to the water soluble moieties; and finally, phase III refers to the efflux of parent compounds or metabolites by

cr

membrane transporters (Hodgson, 2010). The glutathione-S-transferases (protein name GSTs in

us

humans, Gsts in all other species; gene name GST in humans, gst in all other species) are one of the key enzymes that mediate phase II of cellular detoxification. GSTs have diverse cellular

an

functions through peroxidase and isomerase activities, biosynthesis of physiologically important biomolecules such as leukotrienes and prostaglandins and isomerization of steroids, as well as

M

through the ligand binding (non-catalytic activity) which makes them integral part of cell signaling pathways (Hayes and Pulford, 1995; Sau et al., 2010; Sheehan et al., 2001). However, the most

ed

recognized function of GSTs is to alleviate the toxicity of diverse endogenous and exogenous compounds (xenobiotics). This primary role is fulfilled through catalysis of the nucleophilic attack of

pt

the sulphur atom of reduced glutathione (GSH) on the electrophilic group of the substrate which is either parent compound or phase I metabolite (Mannervik et al., 1985). This type of reaction

Ac ce

usually results in reduction of the compound reactivity and increase in its water solubility as a major prerequisite for the subsequent elimination through the efflux transporters (Sau et al., 2010). Catalytic activity of GSTs depends on two processes: binding and activation of GSH through the conserved G-site (within the N-terminal domain) and binding of electrophilic reaction partner through highly variable H-site (within the N-terminal and the C-terminal domains) (Zimniak, 2007). Through the described mechanism, GSTs catalyze conjugation of diverse endogenous compounds including products of lipid peroxidation and downstream products of oxidative stress including 4hydroxinonenal (4-HNE) and acrolein, a highly toxic propenal base, product of oxidative DNA degradation (Berhane et al., 1994; Jowsey and Hayes, 2007). Besides endogenous substrates, 3 Page 3 of 42

GSTs catalyze the conjugation of a wide range of foreign compounds such as pharmaceuticals, heavy metals, pesticides, herbicides, persistent organic pollutants (POPs) and polyaromatic hydrocarbons (PAHs) (Higgins and Hayes, 2011). GSTs are classified into three distinct families: cytosolic, mitochondrial and membrane associated (Hayes et al., 2005). Based on the sequence similarities, mammalian cytosolic GSTs

ip t

are further divided into seven classes (Alpha, Mu, Pi, Theta, Sigma, Omega, and Zeta). Mitochondrial GSTs belong to the Kappa (GSTK) class, while membrane integral GSTs belong to

cr

the MGST or MAPEG class (Membrane-Associated Proteins in Eicosanoid and Glutathione

us

metabolism) (Mannervik et al., 1992). Cytosolic GSTs are generally homodimers involved in the metabolism of foreign compounds and detoxification of potentially harmful endogenous compounds

an

(Hayes and Pulford, 1995; Sheehan et al., 2001). Members of this family exhibit isomerization activity, covalent/non-covalent ligand binding ability and modulation of signal transduction

peroxidase

activity.

MAPEG

family

M

pathways (Jowsey and Hayes, 2007). Both, cytosolic and mitochondrial GSTs also exhibit GST members

are

membrane-associated

homotrimers,

ed

predominantly involved in the biosynthesis of leukotrienes and prostanoids (Jakobsson et al., 1999), while some members possess high glutathione peroxidase activity that protects membranes

pt

from the action of lipid hydroperoxides.

GSTs have been extensively studied in mammals in the context of xenobiotic metabolism. In

Ac ce

fish, the detoxification role of Gst enzymes have been recognized since the 1970s. However, only a few enzymes have been functionally characterized, and the role of individual Gst within specific tissues have not been resolved. So far, mRNA expression studies have been performed in river pufferfish (Takifugu obscurus) (Kim et al., 2010), bighead carp (Aristichthys nobilis) (Li et al., 2010) and most recently during the embryonic development of zebrafish (Timme-Laragy et al., 2013). In addition, Espinoza et al. (2012) analyzed gst expression in tissues of coho salmon and its modulation by cadmium exposure. Fish specific Gsts from the Rho class have been recently found and quantified in the olfactory bulb of coho salmon (Espinoza et al., 2013). The majority of kinetic studies measured total Gst activity in fish tissues using nonspecific substrates such as 1-chloro1,4-dinitrobenzen (CDNB). Although this approach is important for determination of the total Gst 4 Page 4 of 42

activity in vivo, it does not offer insights into the function of individual Gst enzymes. So far, Alpha class kinetics have been studied in plaice (Pleuronectes platessa), red seabream (Pagrus major) and mangrove killifish (Rivulus marmoratus) (Martinez-Lara et al., 2002; Konishi et al., 2005; Lee et al., 2006), Pi and Omega in coho salmon (Oncorhynchus kisutch) (Espinoza et al., 2013), Theta in mangrove killifish (Lee et al., 2006), Rho in coho salmon, red seabream and amphioxus

ip t

(Branchiostoma belcheri) (Espinoza et al., 2013; Konishi et al., 2005; Fan et al., 2007). However, comprehensive analyses of gst genes within teleost species are scarce. Therefore, the major goal

cr

of our study was a comprehensive characterization of the GST superfamily in zebrafish (Danio

us

rerio) as an important vertebrate model organism. Using phylogenetic analysis, tissue mRNA expression profiling and kinetic assays on purified enzymes, we provide new data that will

an

contribute to the better understanding of diversity, evolution and toxicological and/or physiological

M

relevance of GSTs.

ed

Materials and Methods

Ethics Statement

pt

This study was carried out in accordance with the directions given in the EU Guide for the Care

Ac ce

and Use of Laboratory Animals, Council Directive (86/609/EEC) and the Federal Act on the Protection of Animals (NN 135/2006). The protocol was approved by the Bioethics Committee of the Ruđer Bošković Institute, Zagreb, Croatia (Permit Number: BP-1504/2-2011).

Phylogenetic analysis

GST/Gst protein sequences were retrieved from following databases: National Center of Biotechnology

Information

(NCBI)

(http://www.ncbi.nlm.nih.gov/),

Ensembl

(http://www.ensembl.org/index.html) and DOE Joint Genome Institute (http://genome.jgi-psf.org/). Previously annotated mammalian GSTs/Gsts were named accordingly, while for non-mammalian Gsts blastx algorithm was used to search for Gst genes by blasting human sequences against the 5 Page 5 of 42

genome of chicken (Gallus gallus), anole lizard (Anolis carolinensis), western clawed frog (Xenopus tropicalis), zebrafish (Danio rerio), stickleback (Gasterosteus aculeatus), green spotted pufferfish (Tetraodon nigroviridis), Japanese pufferfish (Takifugu rubripes), medaka (Oryzias latipes), Florida lancelet (Brachiostoma floridae), and sea squirt (Ciona intestinalis). All previously non-annotated Gsts were annotated according to the phylogenetic analysis. Provisional gene

ip t

names were given in accordance with the new nomenclature adopted by the HUGO Gene Nomenclature Committee (http://www.genenames.org). Sequence alignments were performed with

cr

the Muscle algorithm (Edgar, 2004). Phylogenetic trees were built in PhyML Software using the

us

maximum likelihood method (Guindon and Gascuel, 2003) with UDP glycosyltransferase (UGT) gene superfamily as the outgroup. Robustness of the tree was determined by the approximate

an

likelihood ratio test (aLRT) (Anisimova and Gascuel, 2006). BioEdit Software version 7.0. was used for sequence editing, alignment display and calculation of sequence identities (Hall, 1999).

M

Orthology predictions using syntenic relationships between zebrafish and human genes of interest were made using Genomicus (http://www.genomicus.biologie.ens.fr/genomicus), a conserved

ed

synteny browser synchronized with genomes from the Ensembl database (Louis et al., 2013).

pt

Tissue-specific gene expression analysis

Adult male and female zebrafish, purchased from a local fish supplier, were sacrificed for the

Ac ce

collection of tissues. Brain, gills, liver, intestine and gonads from four specimens of the same gender were pooled together, with one pool representing one sample, in order to get substantial amount of material for RNA isolation. In the case of RNA isolation from kidney, 14 specimens were pooled together due to the small size of zebrafish kidneys. In that way, 3-5 samples for each tissue were collected to conduct tissue- and gender-specific expression analysis. After isolation, tissues were stored in RNA later (Qiagen, Hilden, Germany). For RNA isolation, tissues were homogenized using rotor-stator homogenizer at 10,000 rpm for 20 s. Total RNA isolation from each tissue was carried out with RNeasy Mini Kit (Qiagen, Hilden, Germany), RNA was quantified using Bio-Spec Nano spectrophotometer (Shimadzu Corporation, Kyoto, Japan), and the integrity of RNA 6 Page 6 of 42

was determined by gel electrophoresis. Genomic DNA digestion was carried out using RNase-free DNase Set (Qiagen, Hilden, Germany). Purified total RNA was reversely transcribed (1 μg of total RNA) using High Capacity cDNA Reverse Transcription Kit with RNase Inhibitor (Applied Biosystems, Foster City, CA, USA). For the qRT-PCR analysis, specific primers were manually designed (Table S1), and purchased from Life Technologies (Carlsbad, CA, USA). Target

ip t

amplicons of 95-127 bp were cloned using the pGEM-T Vector System I (Promega, Medison, WI, USA). Plasmids were purified by DNA-SpinTM (iNtRON Biotechnology, Kyungki-Do, Korea), and

cr

amplicons were verified by sequencing at the Ruđer Bošković Institute DNA Service (Zagreb,

us

Croatia). Primer efficiencies were determined for each gene using the recombinant plasmid as a template (Table S1). Quantification of the gst genes was performed using the qRT-PCR relative

an

quantification method (Qgene method) and normalized to the housekeeping gene ef1a (elongation factor 1α), as previously described (Muller et al., 2002; Simon, 2003; Loncar et al., 2010). We

M

previously tested the expressional pattern of several housekeeping genes (gapdh, 18s, b-actin, and ef1a) to all tissues in order to find gene with the highest expression stability among all

ed

investigated tissues and both zebrafish genders. The expressional pattern of ef1α was very stable in all tissues (Fig. S5), which was in accordance with the previously published data (McCurley and

pt

Callard, 2008). Therefore, ef1α was further used as a reference control for expression data.

Ac ce

In the case of Gst Alpha, Mu and Pi classes, specific qRT-PCR primers which would distinguish members within each class could not be designed due to very high sequence identity among transcripts (Fig. S1). Accordingly, quantification of gsta1-3, gstm1-3 and gstp1-2 represent the sum of expression of all genes within each class. However, in order to distinguish among the genes within Gst Alpha, Mu and Pi class, respectively, specific PCR primers within the 5’ UTR and 3’ UTR regions of each gene were designed. Gsta1, gsta2, gsta3, gstm1, gstm2, gstm3 and gstp1 and gstp2 were amplified from zebrafish cDNA of both gender by polymerase chain reaction (PCR) with Taq DNA polymerase (Thermo Scientific, MA, USA), using specific forward and reverse primers (Table S2). All the PCR products were purified from the agarose gel using MinElute Gel Extraction 7 Page 7 of 42

Kit (Qiagen, Hilden, Germany) and sequenced at the Ruđer Bošković Institute DNA Service (Zagreb, Croatia) in order to confirm specificity of gene amplification within each tissue (Fig. 3).

Cloning and protein purification Each gst gene was amplified from zebrafish cDNA by polymerase chain reaction (PCR) with

ip t

high fidelity Phusion DNA polymerase (Thermo Scientific, MA, USA), using specific forward and reverse primers, with introduced cloning sites for NheI and XhoI restriction enzymes (Thermo

cr

Scientific, MA, USA) (Table S3). PCR products were digested with NheI and XhoI restriction

us

enzymes and cloned into the pET-21a(+) vector that contained C-terminal His-tag sequence, (Novagen, Madison, WI, USA), transformed into DH5α competent cells (Invitrogen, Carlsbad, CA,

an

USA) and sequenced at the Ruđer Bošković Institute DNA Service (Zagreb, Croatia). Afterwards, each protein was expressed by transforming BL21 E. coli competent cells. Overnight E.coli

M

cultures, grown at 37°C in Luria-Bertani medium with 50 µg/mL ampicilin, were diluted 40 times, followed by growth to OD 0.6. The induction was performed at 30°C for 6 h using 1 mM IPTG

ed

(isopropyl β-D-thiogalactoside). At the end of the induction period, cells were collected by centrifugation. GST proteins were purified using Ni2+-nitrilotriacetic acid (Ni-NTA) affinity binding of

pt

His-tagged protein according to the manufacturer instructions (Qiagen, Hilden, Germany). Induced cells (100 mL) were centrifuged, followed by pellet resuspension in 10 mL of lysis buffer without

Ac ce

imidazole (25 mM Tris-HCl (pH 7.5), 500 mM NaCl, 1 mM EDTA, 1 mg/mL of lysozyme). After incubation period of 30 min, cells were sonicated with six 15 s low frequency bursts intercepted with 1 min incubations on ice. Cell lysates were centrifuged at 10,000 g for 45 min, followed by the collection of supernatants which were incubated for 1 h with 0.5 mL of Ni-NTA beads in lysis buffer containing 20 mM imidazole. After the incubation, the beads with bound proteins were collected by centrifugation at 800 g and transferred to polypropylene columns. The Ni-NTA resin was washed twice with 5 mL of washing buffer containing 25 mM Tris-HCl (pH 7.5), 500 mM NaCl and 40 mM imidazole. Zebrafish Gsts were eluted with 2 mL of elution buffer containing 25 mM Tris-HCl (pH 7.5), 500 mM NaCl and 500 mM imidazole, followed by dialysis with dialysis tubing (Biorad, 8 Page 8 of 42

Hercules, CA, USA) against dialysis buffer containing 25 mM Tris-HCl (pH 7.5), 50 mM NaCl, 1% glycerol and 0.5 mM dithiothreitol. Recombinant protein(s) solubility (Table S7) in E. coli expression system was analysed using the University of Oklahoma recombinant protein solubility prediction, a server-based predictor for estimating protein solubility (Diaz et al., 2009). Analysis showed that out of 9 purified proteins only Gstt1a is insoluble. However, there were no observable

ip t

insoluble forms (e.g., inclusion bodies) of any recombinant proteins during the purification process. This was confirmed both by the SDS-PAGE of all fractions during the purification process which

cr

showed no major retention of proteins in precipitate fractions (data not shown) and the western blot

us

of purified recombinant proteins.

an

Western blot analysis

Purified His-tagged proteins were analyzed on Western blots. Protein concentration was

M

measured using Bradford assay (Bradford, 1976), and 2.5 µg of protein was loaded per lane and separated on 0.1% sodium dodecyl sulphate polyacrylamide gel (final concentration of

ed

polyacrylamide was 12%). After the semidry transfer to the polyvinylidene difluoride membrane (Millipore, Billerica, MA, US), the membrane was blocked in a blocking solution containing 5% low

pt

fat milk (Roth, Karlsruhe, Germany), 50 mM Tris, 150 mM NaCl and 0.05% Tween 20. The membrane was incubated with anti-His mouse monoclonal primary antibody (Santa Cruz

Ac ce

Biotechnology, CA, USA) for two hours with gentle rocking, and goat anti-mouse IgG horseradish peroxidase conjugate secondary antibody for one hour. The samples were detected using enhanced chemiluminiscence method and visualized by photographic film exposure (Abcam, Cambridge, UK).

GST kinetic assays GST

activities

towards

1-chloro-2,4-dinitro

benzene

(CDNB)

were

performed

spectrophotometricaly according to Habig et al. with slight modifications (Habig et al., 1974). Reaction mixture (250 µL) contained 100 mM phosphate buffer (pH 6.5), 1 mM CDNB, 1 mM reduced glutathione (GSH) and 2 µg/mL of the Gst enzyme. Absorbance was measured 9 Page 9 of 42

continuously for 10 min at 15 s intervals at 340 nm (ε = 9.6 mM-1cm-1). Spectrofluorometric assay for GST activity towards monochlorobimane (MCB) was performed in 250 µL reaction mixture containing 100 mM phosphate buffer, 200 µM MCB and 1 mM GSH with 2 µg/mL of Gst enzyme (Kamencic et al., 2000). Fluorescence intensity was measured in 15 s kinetic intervals for 10 min at 355 nm excitation and 460 nm emission wavelengths. The spectrometric method for determination

ip t

of specific Gstz1 glutathione-dependent oxygenation of dichloroacetic acid (DCA) to the glyoxylic acid was performed according to Tong et al. (1998) with the modification of reaction temperature

cr

from 37°C for the human GSTs in the original protocol to 25°C for zebrafish Gsts. In order to

us

determine specific activities of Gstz1 with DCA as a substrate, the standard curve with glyoxylic acid was used for calibration. Absorbance was measured at 535 nm. Specific activities of Gsto2

an

and Gst Pi class enzymes with dehydroascorbate (DHA) (265 nm) and ethacrynic acid (ETA) (270 nm) were tested with spectrophotometric kinetic assays according to Wells et al. (1995) and Habig

M

et al. (1974), respectively, with a scale down of reaction mixtures from 1 mL to 250 µL. From the linear slope of rising absorbance of CDNB, ETA and DHA glutathione conjugates, formation of

ed

substrate-GSH conjugate was calculated based on Beer-Lambert law and available molar extinction coefficients: 9.6, 5 and 14 mM-1cm-1 for CDNB, ETA and DHA-GSH conjugates,

pt

respectively. All kinetic assays were performed on the monomeric forms of GST proteins. In all cases, blanks (rate of non-enzymatic reaction) were subtracted from the rate of the enzymatic

(Greiner

Ac ce

reaction. All absorbance measurements were performed in 96-well, flat bottom, transparent plates Bio-One

GmbH,

Frickenhausen,

Germany)

apart

from

the

ETA

conjugation

measurements where UV-star 96-well transparent plates (Greiner Bio-One GmbH, Frickenhausen, Germany) were used. Fluorescence was measured in 96-well black plates (Sigma-Aldrich, Taufkirchen, Germany).

Data analysis Tissue expression results are presented as mean values ± SEM from 3-5 independent pools. Each pool was measured in duplicates. Tissue statistical comparisons were performed by one-way 10 Page 10 of 42

analysis of variance (ANOVA) with Tukey’s multiple comparison post-hoc test (Table S5). Sex related statistical comparisons were not performed due to low number of samples needed for nonparametric statistics. p values of 26), at moderate levels when MNE was 1000 – 10,000 (Ct = 22 - 25), at high levels if MNE was 10,000 – 100,000 (Ct = 19 - 21), and at very high levels when

ip t

MNE was above 100,000 (Ct < 18). Observations on possible gender dependent differences in gene expressions were made in some cases. However, as these observations are not based on

cr

statistical analysis, but rather on high differences in MNE values, they can only serve as indications

us

for further investigations that can only be done with higher number of specimens.

Gst Alpha class showed high expression in intestine and gonads, in comparison to other

an

zebrafish Gst classes (Figs. 2A and S4, Table S5). Gst Alpha genes were moderately expressed in other tissues (Fig. 2A). Gsta1 and gsta3 appeared to be predominantly expressed in the female

kidney, intestine and testes (Fig. 3).

M

brain and male intestine, respectively, whereas gsta2 is dominant in female gills, followed by liver,

ed

Gst Mu genes showed high expression in gonads and brain as well as in the kidney and female gills, followed by moderate expression in other tissues. Pronounced sex differences in Gst Mu

pt

class expression in favor of females appeared to be present in the kidney (9.1-fold), gills (5-fold) and brain (2.4-fold) (Fig. 2A). Gstm1 is ubiquitously expressed, as opposed to gstm2 which is

Ac ce

found in the brain, gills, intestine and gonads, and gstm3 that is present in the gills, intestine and gonads of both genders (Fig. 3).

Gst Pi class showed very high expression in gills (male) and ovaries, followed by ubiquitously high expression in all other tissues: kidney, gills (female), intestine, brain, testes and male liver, and moderate expression in female liver. Statistical analysis revealed ovaries-specific transcriptional level of Pi class in female zebrafish (Table S5). Pronounced gender differences appeared to be present in gonads where Gst Pi genes showed 4-fold higher expression in ovaries. The opposite trend was observed in liver and brain (Fig. 2B). Gstp1 showed very high expression throughout zebrafish tissues, unlike low expression of gstp2. (Fig. 3). 14 Page 14 of 42

Expression pattern of Theta class genes showed markedly different distribution throughout the tissues. Gstt1a showed very high liver-specific expression in males, followed by high expression in female liver, kidney of both genders, and testes, following by moderate to low expression in other analyzed tissues. On the contrary, gstt1b was ubiquitously low expressed, whereas gstt2 showed ubiquitously moderate expression with the exception of low expression in ovaries (Figs. 2B and 2C,

ip t

Table S5).

Similar to the Theta class, Omega genes in zebrafish showed pronounced variation in

cr

expression pattern among the co-orthologs. Gsto1 was much lower expressed throughout

us

zebrafish tissues than gsto2. Gsto1 was present at moderate expression levels in intestine, brain and testes, whereas levels were low in other tissues. On the contrary, gsto2 showed high

an

expression in ovaries (8.1-fold higher than in testes), gills of both genders and female kidney, intestine and brain, followed by moderate expression in the male kidney, intestine and brain and

M

low expression in the liver of both genders (Fig. 2D).

Single member of the Gst Zeta class, gstz1, showed comparatively lower expression across

ed

tissues in respect to the other gst genes, with high expression in liver, followed by moderate expression in kidney, intestine, brain of both genders. (Fig. 2E).

pt

The fish specific Gst Rho class gene (gstr1) showed very high expression in male liver and intestine, female kidney, testes and brain of both genders and generally revealed ubiquitously high

Ac ce

expression throughout all tissues (Fig. 2E, Table S5). Expression analysis of Microsomal Gsts (mgsts) showed comparatively lower expression of mgst1a, mgst1b and mgst2 than the other quantified gst genes. Mgst1a was ubiquitously moderately expressed, with the exception of high expression in testes. Mgst1b was ubiquitously low expressed, with the exception of moderate expression in the brain (Fig. 2F, Table S5). Mgst2 was found at moderate expression levels in liver and ovaries, with significantly higher expression in comparison to other tissues, as well as in male gills, intestine and brain, followed by low expression in other analyzed tissues (Fig. 2G, Table S5). On the contrary, both mgst3 co-orthologs were highly expressed. Mgst3a was found at very high expression levels in intestine, followed by 15 Page 15 of 42

ubiquitously high expression in all other tissues, while mgst3b showed ubiquitously high expression in both genders (Fig. 2H). Overall, considering tissue-specific differences in mRNA expression of gst genes in adult zebrafish, dominant gst genes in zebrafish tissues are: gstt1a, gstr1, mgst3a and mgst3b in the liver, closely followed by Gst Pi class genes and gstz1; gstr1 in the kidney, followed by gstp1-2,

ip t

gstt1a, mgst3a and mgst3b; Gst Pi class genes in gills, followed by gsto2, gstr1, mgst3a and mgst3b; gstr1 and mgst3a in intestine, followed by gstp1-2 and mgst3b. In zebrafish brain, gstr1 is

cr

predominantly expressed, closely followed by mgst3a and mgst3b and Gst Mu and Pi class genes.

us

In the ovaries, predominant are Gst Pi and Mu class genes, gsto2 and mgst3a and mgst3b, whereas in testes expression of Rho, Alpha, Mu and Pi class and mgst3a and mgst3b is most

an

pronounced (Fig. S4).

M

Functional characterization of zebrafish cytosolic GSTs

Enzyme kinetics was determined for nine cytosolic Gst proteins (Figure 4). We found that all

substrate (Table 2).

ed

analyzed Gsts catalyze conjugation of reduced glutathione (GSH) (Table 3) and a specific model

pt

Zebrafish Gsta3 was characterized as a representative of the Gst Alpha class. We found that Gsta3 utilizes CDNB in the conjugation reaction with GSH, whereas it showed no activity with MCB

Ac ce

(Table 2A). Affinity towards CDNB was in the milimolar range (1.7 mM) (Fig. 5), while affinity for the GSH was in a micromolar range (750 µM) (Table 2A and Table 3). Turnover number for CDNB was 38 s-1, while the enzyme efficiency (kcat/Km) was 22 mM-1s-1. Members of zebrafish Gst Mu class, Gstm1 and Gstm3, utilized CDNB at the efficiencies (kcat/Km) of 22 mM-1s-1 and 12 mM-1s-1, respectively (Table 2A). MCB showed to be a very potent substrate of both, Gstm1 and Gstm3, with affinities of 0.22 and 0.11 mM, respectively (Table 2A). In reaction with CDNB, Gstm3 showed comparatively lower affinity for GSH than Gstm1 (Table 3). Two members of zebrafish Gst Pi class, Gstp1 and Gstp2, conjugated CDNB with GSH with the highest enzyme efficiencies in respect to other zebrafish Gsts (232 mM-1s-1and 400 mM-1s-1, 16 Page 16 of 42

respectively) (Table 2A). Both tested Gst Pi members showed activities with MCB and ETA (Table 2). Gstp1 and Gstp2 showed similar affinities for ETA of 0.20 mM and 0.24 mM, respectively, with similar specific activities and turnover numbers for both, Gstp1 (Vmax = 0.34 µmol/mg protein/min, kcat = 2.24 s-1) and Gstp2 (Vmax = 0.37 µmol/mg protein/min, kcat = 1.23 s-1) (Table 2B). The affinities for GSH were within a moderate range for both enzymes (in comparison to other zebrafish Gsts)

ip t

(Table 3, Fig. 7).

Gsto2 did not accept CDNB, MCB or ETA as substrates. However, it mediated reduction of DHA

cr

into ascorbate in the presence of GSH. Gsto2 reduced DHA with affinity of 1.08 + 0.05 mM and

us

specific activity of 0.03 µmol/mg protein/min (kcat = 12 s-1), while affinity for GSH was 2.91 + 0.54 mM (Table 2B, Table 3).

an

The only significantly expressed member of Gst Theta class, Gstt1a, conjugated CDNB with the affinity of 0.32 mM and efficiency of 16 mM-1s-1, together with comparatively lower affinity for co-

M

substrate GSH of 4.44 mM (in respect to other zebrafish Gsts) (Table 2A and Table 3). The most notable characteristic of Gstt1a was an exceptionally high affinity for MCB, with Km value of 4.68

ed

µM (Fig. 6, Table 2A).

The only member of Gst Zeta class, Gstz1 did not show activity towards CDNB, MCB and ETA

pt

(Table 2). However, it showed ability to oxygenate DCA to glyoxylic acid in the presence of GSH with affinity of 0.27 + 0.03 mM towards DCA and 0.27 + 0.04 mM for GSH (Table 2B and Table 3).

Ac ce

Despite comparatively higher affinity towards DCA, considering its lower specific activity (15.6 nmol/mg protein/min, kcat = 0.002 s-1) the efficiency of DCA catalysis was in a lower range in respect to other zebrafish Gsts (kcat/Km = 0.0074 mM-1s-1). Fish specific Gst enzyme, Gst Rho (Gstr1) accepted model substrates of other Gst enzymes, with affinities for CDNB and MCB of 0.58 and 0.05 mM, respectively (Tables 2A, Fig. 8). Enzyme efficiency in the CDNB conjugation was kcat/Km = 132 mM-1s-1.

Discussion

17 Page 17 of 42

Glutathione-S-transferases are widely recognized as important parts of cellular detoxification system, thus making Gst(s) gene expression and/or activity frequently utilized as biomarker of exposure to xenobiotics in the environmental studies. Accordingly, numerous studies on fish Gsts are available. However, most of these studies focused only on the gene expression, or measurement of total GST activity in a specific tissue, while only a few studies determined enzyme

ip t

kinetics of specific Gst enzyme(s) after protein purifications (Blanchette et al., 2007). Based on these premises we performed a comprehensive analysis of fish Gsts in zebrafish as an important

cr

vertebrate model species. Firstly, we identified all gst genes in chordates using a genome-wide

us

phylogenetic analysis. As a result, 27 zebrafish gst genes, within 9 classes, were annotated. Orthology relationships with human GSTs were further examined by conserved synteny analysis.

an

Gene identification was followed by gene expression analysis of 20 gsts. Mitochondrial genes from the kappa class and MAPEG members without glutathione transferase activity (flap, ltc4s and

M

ptges) were not quantified. Based on the tissue expression analysis and orthology relationships with human GSTs of potential (eco)toxicological relevance, we have selected 9 cytosolic Gsts for

substrates.

ed

further characterization through the enzyme kinetic measurements using the model GST

pt

Zebrafish and human Gst Alpha clusters are syntenic (Fig. S3). Major difference in tissue expression profile between human and zebrafish Gst Alpha class is present in the liver, where

Ac ce

GSTA1 and GSTA2 are highly expressed (Suzuki et al., 1993), as opposed to the low hepatic expression of gsta genes. Interestingly, gsta genes in other fish species, Japanese pufferfish (Kim et al., 2010), mangrove killifish (Lee et al., 2006) and common carp (Fu and Xie, 2006) showed the highest expression in liver, while similar to zebrafish low expression pattern is reported in the liver of bighead carp (Li et al., 2010). Despite comparatively high hepatic expression in three fish species mentioned above, Gst Alpha is not the dominant gst gene(s) in fish liver, as Mu, Theta, Pi and MAPEG classes are more pronounced (Kim et al., 2010; Li et al., 2010; Espinoza et al., 2012). All these findings suggest differences in the hepatic function of GSTs between mammals and fish. High intestinal expression, as well as moderate expression in kidney and brain of Gst Alpha class in zebrafish is also reported in other fish species: bighead carp (Li et al., 2010), river pufferfish 18 Page 18 of 42

(Kim et al., 2010) and mangrove killifish (Lee et al., 2006). Zebrafish Gsta3 affinity towards the model substrate CDNB is 2.7 – 3.5 times lower than the affinity of human co-orthologs GSTA1-4 (Singhal et al., 1990; Kolobe et al., 2004), while it is similar to the affinities of Gsta enzymes characterized in other teleost species (Martinez-Lara et al., 2002; Konishi et al., 2005; Lee et al., 2006). In respect to the turnover number (kcat), Gsta3 is similar to human co-orthologs, GSTA1 and

ip t

GSTA2 (38 s-1 for Gsta3, 59 s-1 for human GSTA1, and 33 s-1 for GSTA2) (Zhao et al., 1999; Kolobe et al., 2004) which is significantly higher (150 – 220 times) than the human GSTA3 isoform

cr

(Fedulova et al., 2010). Interestingly, comparison of enzyme efficiencies (kcat/Km) revealed that

us

zebrafish Gsta3 is 7 times less efficient than human GSTA1, similarly to GSTA2 and 5 times less efficient than GSTA3 (Zhao et al., 1999; Johansson and Mannervik, 2002). Higher efficiency of

an

human GSTA1 than the zebrafish Gsta3 cannot be due to the difference in the assay temperature because measurements were performed at 25°C in both cases (Zhao et al., 1999), whereas higher

M

efficiency of GSTA3 could be a consequence of increased reaction temperature (measured at 30°C) (Johansson and Mannervik, 2002). Affinity towards GSH is similar to human GSTA3 (440

ed

µM), and 3, 4.7 and 15 times lower than the GSH affinities of GSTA4, GSTA1 and GSTA2, respectively (Johansson and Mannervik, 2002; Pettersson and Mannervik, 2011). Gsta3 affinity

pt

toward GSH is similar to the plaice Gsta (600 µM) and Gsta2 from red sea bream (690 µM) and 2.6 and 5 times lower than plaice Gsta1 and mangrove killifish Gsta, respectively (Martinez-Lara et al.,

Ac ce

2002; Lee et al., 2006). Altogether, comparison of kinetic data showed that zebrafish Gsta3 is most similar to human co-ortholog GSTA2 in respect to CDNB kinetics (affinity and enzyme efficiency), and to GSTA3 in respect to the GSH affinity, while it revealed functional similarities to the Gsta enzymes in other fish species.

Zebrafish and human gstm/GSTM clusters are syntenic (Figure S3). Zebrafish Gst Mu class showed high expression in brain and gonads (Figure 2A), similar to human GSTM1-3 and GSTM5 (Eaton and Bammler, 1999) as well as to the expression of Mu class genes in other fishes, bighead carp and river pufferfish (Kim et al., 2010; Li et al., 2010). Human GSTM1 shows very high expression in liver (Takahashi et al., 1993), as well as Mu class genes in river pufferfish (Kim et al., 2010). On the contrary, lower hepatic expression observed in zebrafish is also found in bighead 19 Page 19 of 42

carp and coho salmon (Li et al., 2010; Espinoza et al., 2012). Interestingly, gstm1 is dominant isoform in zebrafish tissues of both genders, whereas gstm2 appeared to be markedly expressed only in female brain, gills and ovaries and gstm3 is dominant in male gills, intestine and gonads. (Fig. 3). Affinity of zebrafish Gstm1 and Gstm3 enzymes for CDNB is similar to the human GSTM3 (Rowe et al., 1998) and Gstm5 in rodents (Alin et al., 1985; Hu et al., 1996), whereas it is 3.4 – 5.7

ip t

times lower than human GSTM1a (Rowe et al., 1998). The affinity of zebrafish Gstm1 towards GSH (in CDNB reactions) is 3.6 times higher in comparison to Gstm3 (Table 3) and is similar to the

cr

human co-orthologs (Rowe et al., 1998). Enzyme efficiencies of Gstm1 and Gstm3 (22 and 12 mM1

us

sec-1, respectively) are 10 – 20 times lower than efficiencies of human GSTM1a, and 10 - 20 times

higher than efficiencies of human GSTM3 (Rowe et al., 1998). Overall, in respect to the expression

an

pattern and affinity towards CDNB and GSH, both zebrafish Gst mu co-orthologs, Gstm1 and m3, are more similar to the human co-ortholog GSTM3 than to the GSTM1. Given the expression of

M

zebrafish Gstm1 in the brain and kinetic similarities to GSTM3, it is possible, that zebrafish Gstm1 function as prostaglandin E2 synthases, similar to the role of human GSTM3 (Beuckmann et al.,

ed

2000).

Two zebrafish Gst Pi genes, gstp1 and gstp2 are syntenic with the single human ortholog,

pt

GSTP1 (Figure S3). The difference of gene numbers between zebrafish and human can be explained by an evolutionary important additional round of the whole genome duplication called the

Ac ce

teleost-specific genome duplication (TSD). Together with TSD, the elevated number of zebrafish genes could be due to independent gene duplications (Meyer and Schartl, 1999). Gst Pi class in zebrafish is expressed ubiquitously high (Fig. 2B), similar to the expression pattern in another cyprinid species, bighead carp (Li et al., 2010), as well as to the human ortholog GSTP1 (Eaton and Bammler, 1999). Most importantly, from the combination of qRT-PCR (Fig. 2B) and PCR data (Fig. 3), it is clear that Gstp1 is actually the predominant gene. High expression of zebrafish gstp1 in liver, kidney, gills and intestine could suggest a pivotal role in the metabolism of xenobiotics. Indeed, the affinities of Gstp1 and Gstp2 for CDNB were the highest among zebrafish Gsts (excluding Gstt1a) (Table 2A). In comparison to the human co-ortholog, affinity for CDNB of both zebrafish enzymes is 3.5 – 4.6 higher, while at the same time affinity towards GSH is substantially 20 Page 20 of 42

lower (5.4 and 7.9 times lower for Gstp1 and Gstp2, respectively) (Ricci et al., 1995). Zebrafish Pi enzymes are more efficient (in respect to the CDNB conjugation) (kcat/Km = 232 and 400 mM-1s-1 for Gstp1 and Gstp2, respectively) than their human co-ortholog (kcat/Km = 55 mM-1s-1) (Ali-Osman et al., 1997). Unique feature of both zebrafish Gst Pi enzymes is high affinity for ethacrynic acid (ETA) (Table 2B), which is comparable to the human GSTP1 (0.18 mM). Besides the pivotal role in

ip t

the metabolism of xenobiotics, high expression of zebrafish Gst Pi enzymes in nonbiotransformation tissues such as gonads and brain (Fig. 2B) suggests involvement in

cr

physiological processes through the non-enzymatic activity, similar to its mammalian co-orthologs

us

(Castro-Caldas et al., 2009; Castro-Caldas et al., 2012; Keating et al., 2010; Bhattacharya and Keating, 2012).

an

Zebrafish gstt1a and gstt1b are syntenic to human GSTT1, whereas clear one-to-one orthology relationship is present in the case of zebrafish gstt2 and human GSTT2 (Fig. S3). Gstt1a is

M

ubiquitously expressed, with very high expression in liver, kidney and testes, followed by moderate expression in brain, intestine, gills and ovaries, a pattern similar to the human GSTT1 (Eaton and

ed

Bammler, 1999), whereas gstt1b is negligibly expressed in all analyzed tissues. Zebrafish gstt2 shows ubiquitous expression at moderate levels, similar to the mouse Gstt2 (Knight et al., 2007),

pt

and different from human ortholog GSTT2 which is found primarily in the liver (Sherratt and Hayes, 2002). Interestingly, kinetics of CDNB binding significantly differs among zebrafish Gstt1a and its

Ac ce

mammalian co-orthologs, as human, mouse and rat GSTT1/Gstt1 do not accept CDNB as a substrate. On the contrary, zebrafish Gstt1a shows comparatively high affinity and high specific activity towards CDNB (Table 2A). Similar affinity towards CDNB is reported for Gst Theta member in mangrove killifish (Km = 700 nM), whereas specific activity is significantly lower than for mangrove killifish Gst Theta enzyme (15 µmol/mg protein/min) (Lee et al., 2006). Considering high expression of gstt1a in liver and kidney as main detoxification organs, comparatively high activity and affinity towards CDNB and exceptionally high affinity towards MCB (Fig. 7), we suggest that zebrafish Gstt1a might be involved in the metabolism of xenobiotics, similar to its human coortholog GSTT1 (Hiratsuka et al., 1997; Shokeer et al., 2005). 21 Page 21 of 42

Similar to humans, zebrafish has two Omega genes, gsto1 and gsto2 (Whitbread et al., 2005). Despite close orthology relationships confirmed with conserved synteny analysis (Fig. S3), tissue distribution pattern markedly differs between human and zebrafish Omega genes. Gsto2 is highly expressed in ovaries, gills, female kidney, intestine and brain, followed by moderate expression in male kidney, intestine and brain (Fig. 2D). On the contrary, human GSTO1 shows high expression

ip t

in liver, heart, kidney and skeletal muscle, while GSTO2 is predominantly expressed in testis, followed by moderate expression in heart and liver (Whitbread et al., 2005). Unlike human ortholog

cr

(Wang et al., 2005), zebrafish Gsto2 does not show activity with CDNB. Similar to human GSTO1

us

and GSTO2 (Board et al., 2000; Board, 2011) it does have glutathione-dependent dehydroascorbate reductase activity. Pronounced differences in the tissue expression pattern and

an

interaction with CDNB between Gsto2 and GSTO2 suggests functional differences between these two enzymes.

M

Gst Zeta class, with a single gene gstz1, is present in all analyzed vertebrate species (Table 1) with conserved synteny (Fig. S3). Such conservation of one-to-one orthologies suggests important

ed

function in vertebrate physiology. Hepatic expression of zebrafish gstz1 is similar to high expression of gstz1 as reported in liver of river pufferfish (Kim et al., 2010) and juvenile coho

pt

salmon (Espinoza et al., 2012), while moderate to high hepatic and renal expression of gstz1 is similar to the expression of mammalian co-orthologs in human and rodents (Knight et al., 2007;

Ac ce

Board et al., 1997). Gst Zeta shows comparatively high affinity for GSH and DCA with no detectable activity towards typical GST substrates, CDNB and MCB (Table 2 and Table 3), similar to the human ortholog GSTZ1 (Polekhina et al., 2001). Despite mentioned similarities in respect to substrate selectivity, kinetics of DCA conjugation significantly differs between zebrafish Gstz1 and its human ortholog: DrGstz1 showed significantly higher affinity towards DCA (Km = 0.27 mM) than GSTZ1 (Km = 71.4 µM) (Wells et al., 1995), whereas specific activity and turnover number of human enzyme is much higher (Vmax = 1.4 µmol/mg protein/min, kcat = 0.30) (Ricci et al., 2004). In summary, considering the ability of Gstz1 to biotransform DCA, a common drinking water contaminant, along with the moderate to high expression in liver and kidney, we suggest the involvement of zebrafish Gstz1 in the defense against environmental contaminants. 22 Page 22 of 42

Our results revealed presence of single Gst Rho member in zebrafish genome, which is in sharp contrast to the great diversification of the Rho class in other teleosts (3 – 5 genes) (Table 1). Konishi et al. (2005) proposed that Gst Rho class is an evolutionarily distinct branch of GST superfamily, which is in accordance with our phylogenetic analysis (Figs, 1 and S2). Characterization of Gst genes in other aquatic organisms will clarify if the Gst Rho class is fish

ip t

specific. Tissue expression pattern of zebrafish gstr1 revealed high expression in kidney, brain, liver and intestine, which corresponds to the expression pattern in another fish species, bighead

cr

carp (Li et al., 2010). Affinity and specific enzyme activity of Gstr1 for CDNB (Table 2A) are very

us

similar to the CDNB kinetics of GST Rho ortholog in other species, red seabream (Km = 0.59 mM, Vmax = 7.9 µmol/mg protein/min) (Konishi et al., 2005) and amphioxus (Vmax = 3.4 µmol/mg

an

protein/min) (Fan et al., 2007), whereas Gst Rho in coho salmon shows 10 times higher enzyme activity (30 µmol/mg protein/min) than zebrafish Gstr1 (Espinoza et al., 2013). In summary, along

M

with the Pi class Gstr1 is the most dominant gene in zebrafish tissues, which suggests its crucial role in the zebrafish physiology, whereas high expression in detoxification tissues indicates

ed

important role in the xenobiotic metabolism.

MAPEG (Membrane-Associated Proteins in Eicosanoid and Glutathione metabolism) enzymes

pt

are highly conserved from bacteria to mammals (Lee and DeJong, 1999), which is in line with our phylogenetic and conserved synteny analysis (Figs. S2 and S3, Table 1). Our results revealed that,

Ac ce

in contrast to comparatively lower expression of zebrafish mgst1a, mgst1b and mgst2, mgst3a and mgst3b are expressed ubiquitously high (Fig. 2). Similarly, high expression of mouse Mgst3 was reported in intestine (Knight et al., 2007), whereas dominant liver expression of mgst3/MGST3 was found in pufferfish and human (Staffas et al., 1998; Mulder et al., 1999). Considering high expression of both zebrafish mgst3 genes in liver, similar to the mammalian co-ortholog we propose an important role of Mgst3 enzymes in the metabolism of xenobiotics, a hypothesis that should be tested in future studies.

Conclusion

23 Page 23 of 42

Our study provides the first comprehensive analysis of the GST superfamily in zebrafish. Phylogenetic analysis revealed a great diversity of fish Gsts, with 27 members found in zebrafish. Conserved synteny analysis resolved orthology relationship with human genes which offered a new perspective on evolution of Gsts in teleosts and other chordates. Tissue expression profiling provided new insight into the importance of specific gst genes within zebrafish tissues, thus

ip t

implying the distinct functions of individual Gsts in adult zebrafish. Initial functional characterization of 9 cytosolic Gst enzymes, representatives of 7 Gst classes, revealed their specific interactions

cr

with model GST substrates and enabled the comparison of kinetic properties with human and fish

us

orthologs. We suggest involvement of Gst Pi class, Gstt1a, Gstz1, Gstr1, Mgst3a and Mgst3b in the biotransformation of xenobiotics based on their high expression in barrier tissues (liver, kidney,

an

gills and intestine) and/or on their functional similarity to the human orthologs in respect to the xenobiotic metabolism. Likewise, our data indicate that Gst Alpha, Mu, Pi, Zeta and Rho classes

M

may have crucial role in important physiological processes. Based on the data presented and the methodological protocols developed in this study, the specific physiological role of individual

Ac ce

Acknowledgments

pt

be verified in future studies.

ed

zebrafish Gsts, as well as their putative interaction with specific environmental contaminants can

This work was supported by the Ministry for Science, Technology and Sports of the Republic of Croatia by a project led by TS (grant number 098-0982934-2745), Swiss National Science Foundation (SNSF) grant awarded to KF, TS and RK (project number SCOPES-IZ73Z0_128025/1) and grant awarded to RK by the Serbian Ministry of Education, Science and Technological Development (project number 173037).

Supporting information

24 Page 24 of 42

Figure S1. Nucleotide sequence alignments of zebrafish gst genes of classes (A) Alpha, (B) Mu and (C) Pi. Figure S2. Phylogenetic analysis of GST superfamily (Glutathione-S-transferases) in chordates. The tree was built using the maximum likelihood method in PhyML Software with Approximate Likelihood-Ratio Test for Branches (aLRT). Species abbreviations: Hs, Homo sapiens; Gg, Gallus

ip t

gallus; Ac, Anolis carolinensis; Xt, Xenopus tropicalis; Dr, Danio rerio; Tn, Tetraodon nigroviridis; Tr, Takifugu rubripes; Ga, Gasterosteus aculeatus; Ol, Oryzias latipes; Bf, Brachiostoma floridae;

cr

Ci, Ciona intestinalis.

Figure S3. Conserved synteny analysis of zebrafish and human GSTs/Gsts: (A) Alpha and Omega,

us

(B) Mu and Theta, (C) Pi and MGST1 subclass, (D) Zeta, (E) Kappa, (F) MAPEG classes. Numbers next to the gene names represent megabase pair (Mbp) of gene location on the

an

chromosome.

Figure S4. Expression pattern of gst genes in (A) liver, (B) kidney, (C) gills, (D) intestine, (E) brain and (F) gonads of adult zebrafish. Tissue expression results are presented as mean values ± SEM

M

from 3-5 pools of 4-14 specimens. Each pool was measured in duplicates. MNE stands for mean

ed

normalized expression normalized to the housekeeping gene ef1α (elongation factor 1α). Figure S5. Expression pattern of the housekeeping gene ef1α used as a reference control for qRT-

pt

PCR data, showing uniformity of expression in all analyzed tissues. Figure S6. Amino acid sequence alignments of human and zebrafish GST/Gst proteins: classes (A)

Ac ce

Alpha, (B) Mu, (C) Pi, (D) Theta, (E) Zeta, (F) Omega and (G) Microsomal GSTs/Gsts. Table S1. Primer sequences used for relative quantification of zebrafish gst(s) mRNA using qRTPCR analysis, and parameters of related standard curves. Table S2. Specific PCR primer sequences within the 5’ UTR and 3’ UTR regions designed to distinguish tissue expression of each individual member of Gst Alpha, Mu and Pi classes. Table S3. Primer sequences, restriction enzyme sites introduced in the primer sequences (bold letters), and Tm used for amplification of each gst gene. Table S4. Protein annotation and accession numbers of protein sequences used in the phylogenetic analysis of GST superfamily (Glutathione-S-transferases) in chordates. Species 25 Page 25 of 42

abbreviations: Hs, Homo sapiens; Gg, Gallus gallus; Ac, Anolis carolinensis; Xt, Xenopus tropicalis; Dr, Danio rerio; Tn, Tetraodon nigroviridis; Tr, Takifugu rubripes; Ga, Gasterosteus aculeatus; Ol, Oryzias latipes; Bf, Brachiostoma floridae; Ci, Ciona intestinalis. Table S5. Statistical analysis of zebrafish gst(s) transcriptional levels between different tissues (results presented in Figure 2).

ip t

References

cr

Alin P, Jensson H, Guthenberg C, Danielson UH, Tahir MK, Mannervik B (1985) Purification of major basic glutathione transferase isoenzymes from rat liver by use of affinity chromatograohy and fase protein liquid chromatofocusing. Anal Biochem 146: 313-320.

us

Ali-Osman F, Akande O, Antoun G, Mao J-X, Buolamwini J (1997) Molecula cloning, characterization, and expression in Echerichia coli of full-length cDNAs of three human glutathione S-transferase Pi gene variants. J Biol Chem 272: 10004-10012. Anisimova M, Gascuel O (2006) Approximate likelihood-ratio test for branches: A fast, accurate, and powerful alternative. Syst Biol 55: 539–552. doi: 10.1080/10635150600755453

M

an

Berhane K, Widersten M, Engström a, Kozarich JW, Mannervik B (1994) Detoxication of base propenals and other alpha, beta-unsaturated aldehyde products of radical reactions and lipid peroxidation by human glutathione transferases. Proc Natl Acad Sci USA 91: 1480–1484. doi:10.1073/pnas.91.4.1480 Beuckmann CT, Fujimori K, Urade Y, Hayaishi O (2000) Identification of mu-class glutathione transferases M2-2 and M3-3 as cytosolic prostaglandin E synthases in the human brain. Neurochem Res 25: 733–738. doi: 10.1023/A:1007579507804

ed

Bhattacharya P, Keating AF (2012) Protective role for ovarian Glutathione S-transferase isoform pi during 7,12-dimethylbenz[a]anthracene-induced ovotoxicity. Toxicol Appl Pharmacol 260: 201– 208. doi: 10.1016/j.taap.2012.02.014

pt

Blanchette B, Feng X, Singh BR (2007) Marine glutathione S-transferases. Mar Biotechnol (NY) 9: 513–542. doi: 10.1007/s10126-007-9034-0

Ac ce

Board PG (2011) The omega-class glutathione transferases: structure, function, and genetics. Drug Metab Rev 43:226-235. doi: 10.3109/03602532.2011.561353 Board PG, Baker RT, Chelvanayagam G, Jermiin LS (1997) Zeta, a novel class of glutathione transferases in a range of species from plants to humans. Biochem J 328: 929–935. Board PG, Coggan M, Chelvanayagam G, Easteal S, Jermiin LS, et al. (2000) Identification, characterization, and crystal structure of the Omega class glutathione transferases. J Biol Chem 275: 24798–24806. doi: 10.1074/jbc.M001706200 Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248–254. doi: 10.1016/00032697(76)90527-3 Castro-Caldas M, Carvalho A, Rodrigues E, Henderson C, Wolf C, et al. (2012) Glutathione STransferase pi Mediates MPTP-Induced c-Jun N-Terminal Kinase Activation in the Nigrostriatal Pathway. Mol Neurobiol 45: 466–477. doi: 10.1007/s12035-012-8266-9 Castro-Caldas M, Milagre I, Rodrigues E, Gama MJ (2009) Glutathione S-transferase pi regulates UV-induced JNK signaling in SH-SY5Y neuroblastoma cells. Neurosci Lett 451: 241–245. doi: 10.1016/j.neulet.2009.01.017 26 Page 26 of 42

Diaz AA, Tomba E, Lennarson R, Richard R, Bagajewicz MJ, Harrison RG (2009) Prediction of protein solubility in Escherichia coli using logistic regression. Biotechnol Bioeng 105: 374-383. doi: 10.1002/bit.22537 Eaton DL, Bammler TK (1999) Concise Review of the Glutathione S-Transferases and their Significance to Toxicology. Toxicol Sci 49: 156–164. doi: 10.1093/toxsci/49.2.156 Edgar RC (2004) MUSCLE: a multiple sequence alignment method with reduced time and space complexity. BMC Bioinformatics 5: 113. doi: 10.1186/1471-2105-5-113

ip t

EC. 1986. Council Directive 86/609/EEC of 24 November 1986 on the approximation of laws, regulations and administrative provisions of the Member States regarding the protection of animals used for experimental and other scientific purposes. Brussels, Belgium: European Economic Community.

cr

Espinoza HM, Shireman LM, McClain V, Atkins W, Gallagher EP (2013) Cloning, expression and analysis of the olfactory glutathione S-transferases in coho salmon. Biochem Pharmacol 85: 839– 848. doi: 10.1016/j.bcp.2012.11.018

us

Espinoza HM, Williams CR, Gallagher EP (2012) Effect of cadmium on glutathione S-transferase and metallothionein gene expression in coho salmon liver, gill and olfactory tissues. Aquat Toxicol 110-111: 37–44. doi: 10.1016/j.aquatox.2011.12.012

an

Fan C, Zhang S, Liu Z, Li L, Luan J, et al. (2007) Identification and expression of a novel class of glutathione-S-transferase from amphioxus Branchiostoma belcheri with implications to the origin of vertebrate liver. Int J Biochem Cell Biol 39: 450–461. doi: 10.1016/j.biocel.2006.09.013

M

Fedulova N, Raffalli-Mathieu F, Mannervik B (2010) Porcine glutathione transferase Alpha 2-2 is a human GST A3-3 analogue that catalyses steroid double-bond isomerization. Biochem J 431: 159167. doi: 10.1042/BJ20100839

ed

Guindon S, Gascuel O (2003) A Simple, Fast, and Accurate Algorithm to Estimate Large Phylogenies by Maximum Likelihood. Syst Biol 52: 696–704. doi: 10.1080/10635150390235520 Habig WH, Pabst MJ, Jakoby WB (1974) Glutathione S-Transferases: The first enzymatic step in mercapturic acid formation. J Biol 249: 7130–7139.

pt

Hall TA (1999) BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp 41: 95–98.

Ac ce

Hayes JD, Flanagan JU, Jowsey IR (2005) Glutathione transferases. Annu Rev Pharmacol Toxicol 45: 51–88. doi: 10.1146/annurev.pharmtox.45.120403.095857 Hayes JD, Pulford DJ (1995) The glutathione S-transferase supergene family: regulation of GST and the contribution of the isoenzymes to cancer chemoprotection and drug resistance. Crit Rev Biochem Mol Biol 30: 445–600. doi: 10.3109/10409239509083491 Higgins LG, Hayes JD (2011) Mechanisms of induction of cytosolic and microsomal glutathione transferase (GST) genes by xenobiotics and pro-inflammatory agents. Drug Metab Rev 43: 92– 137. doi: 10.3109/03602532.2011.567391 Hiratsuka A, Yamane H, Yamazaki S, Ozawa N, Watabe T (1997) Subunit Ya-specific Glutathione Peroxidase Activity toward Cholesterol 7-Hydroperoxides of Glutathione S -Transferases in Cytosols from Rat Liver and Skin. J Biol Chem 272: 4763–4769. doi: 10.1074/jbc.272.8.4763 Hodgson E, (2010) Introduction to Biotransformation (Metabolism). In: Krieger R, editor. Hayes' Handbook of Pesticide Toxicology. Elsevier Inc., Vol. Volume 1. pp. 865–875. Hu X, Srivastva SK, Xia H, Awasthi YC, Singh SV (1996) An alpha class mouse glutathione Stransferase with exceptional catalytic efficiency in the conjugation of glutathione with 7β,8αDihydroxy-9α,10α-oxy 7,8,9,10-tetrahydrobenzo(a)pyrene J Biol Chem 271: 32684-32688.

27 Page 27 of 42

Jakobsson P, Morgenstern R, Mancini J, Ford-hutchinson A, Persson B (1999) Common structural features of MAPEG — A widespread superfamily of membrane associated proteins with highly divergent functions in eicosanoid and glutathione metabolism. Protein Sci 8: 689–692. Johansson A-S, Mannervik B (2002) Active-site residues governing high steroid isomerase activity in human glutathione transferase A3-3. J Biol Chem 277: 16648–16654. doi: 10.1074/jbc.M201062200 Jowsey, I.R., Hayes JD (2007) Mammalian Glutathione S-Transferase Genes: Structure and Regulation. In: Awasthi YC, editor. Toxicology of glutathione Transferases. New York. pp. 27–46.

ip t

Kamencic H, Lyon A, Paterson PG, Juurlink BHJ (2000) Monochlorobimane Fluorometric Method to Measure. Anal Biochem 286: 35–37. doi: 10.1006/abio.2000.4765

cr

Keating AF, Sen N, Sipes IG, Hoyer PB (2010) Dual protective role for glutathione S-transferase class pi against VCD-induced ovotoxicity in the rat ovary. Toxicol Appl Pharmacol 247: 71–75. doi: 10.1016/j.taap.2010.06.002

us

Kim J, Raisuddin S, Rhee J, Lee Y, Han K, et al. (2009) Molecular cloning, phylogenetic analysis and expression of a MAPEG superfamily gene from the pufferfish Takifugu obscurus. Comp Biochem Physiol Part C 149: 358–362. doi: 10.1016/j.cbpc.2008.09.003

an

Kim J-H, Dahms H-U, Rhee J-S, Lee Y-M, Lee J, et al. (2010) Expression profiles of seven glutathione S-transferase (GST) genes in cadmium-exposed river pufferfish (Takifugu obscurus). Comp Biochem Physiol C Toxicol Pharmacol 151: 99–106. doi: 10.1016/j.cbpc.2009.09.001

M

Knight TR, Choudhuri S, Klaassen CD (2007) Constitutive mRNA expression of various glutathione S-transferase isoforms in different tissues of mice. Toxicol Sci 100: 513-524. doi: 10.1093/toxsci/kfm233

ed

Kolobe D, Sayed Y, Dirr HW (2004) Characterization of bromosulphophthalein binding to human glutathione S-transferase A1-1: thermodynamics and inhibition kinetics. Biochem J 382: 703-709. doi: 10.1042/BJ20040056 Konishi T, Kato K, Araki T, Shiraki K, Takagi M, et al. (2005) A new class of glutathione Stransferase from the hepatopancreas of the red sea bream Pagrus major. Biochem J 388: 299– 307. doi: 10.1042/BJ20041578

pt

Lee SH, DeJong J (1999) Microsomal GST-I: genomic organization, expression, and alternative splicing of the human gene. Biochim Biophys Acta 1446: 389–396. doi: 10.1016/S01674781(99)00112-8

Ac ce

Lee, Y-M, Seo JS, Jung S-O, Kim I-C, Lee J-S (2006) Molecular cloning and characterization of θclass glutathione S-transferase (GST-T) from the hermaphroditic fish Rivulus marmoratus and biochemical comparisons with α-class glutathione S-transferase (GST-A). Biochem Biophys Res Commun 346: 1053-1061. doi: 10.1016/j.bbrc.2006.06.014 Li G, Xie P, Li H, Chen J, Hao L, et al. (2010) Quantitative profiling of mRNA expression of glutathione S-transferase superfamily genes in various tissues of bighead carp (Aristichthys nobilis). J Biochem Mol Toxicol 24: 250–259. doi: 10.1002/jbt.20333 Loncar J, Popovic M, Zaja R, Smital T (2010) Gene expression analysis of the ABC efflux transporters in rainbow trout (Oncorhynchus mykiss). Comp Biochem Physiol C Toxicol Pharmacol 151: 209–215. doi: 10.1016/j.cbpc.2009.10.009 Louis A, Muffato M, Roest Crollius H (2013) Genomicus: five genome browsers for comparative genomics in eukaryota. Nucleic Acids Res 41: D700–D705. doi: 10.1093/nar/gks1156 Mannervik B, Alin P, Guthenberg C, Jensson H, Tahir MK, et al. (1985) Identification of three classes of cytosolic glutathione transferase common to several mammalian species: Correlation between structural data and enzymatic properties. Proc Natl Acad Sci U S A 82: 7202–7206. doi: 10.1073/pnas.82.21.7202 28 Page 28 of 42

Mannervik B, Awasthi YC, Board PG, Hayes JD, Di Ilio C, et al. (1992) Nomenclature for human glutathione transferases. Biochem J 282: 305–308. Martinez-Lara E, Leaver M, George S (2002) Evidence from heterologous expression of glutathione S-transferases A and A1 of the plaice (Pleuronectes platessa) that their endogenous role is in detoxification of lipid peroxidation products. Mar Environ Res 54: 263-266. McCurley AT, Callard GV (2008) Characterization of housekeeping genes in zebrafish: malefemale differences and effects of tissue type, developmental stage and chemical treatment. BMC Mol Biol 9:102. doi:10.1186/1471-2199-9-102

ip t

Meyer A, Schartl M (1999) Gene and genome duplications in vertebrates: the one-to-four (-to-eight in fish) rule and the evolution of novel gene functions. Curr Opin Cell Biol 11: 699-704.

cr

Mulder TPJ, Court DA, Peters WHM (1999) Variability of glutathione s-transferase α in human liver and plasma. Clin chem 45: 355-359. Muller PY, Janovjak H, Miserez AR, Dobbie Z (2002) Processing of Gene Expression Data Generated by Quantitative Real-Time RT-PCR. Biotechniques 32: 1372–1374, 1376, 1378–1379.

us

Pettersson PL, Mannervik B (2011) The role of glutathione in isomerization of Δ5- androstene-3,17dione catalyzed by human glutathione transferase A1-1. J Biol Chem 276: 11698-11704. doi: 10.1074/jbc.M009146200

an

Polekhina G, Board P, Blackburn A, Parker M (2001) Crystal Structure of Maleylacetoacetate Isomerase/Glutathione Transferase Zeta Reveals the Molecular Basis for Its Remarkable Catalytic Promiscuity. Biochemistry 40: 1567–1576. doi: 10.1021/bi002249z

M

Ricci G, Turella P, De Maria F, Antonini G, Nardocci L, Bord PG, Parker MW, Carbonelli MG, Federici G, Caccuri AM (2004) Binding and kinetic mechanisms of the zeta class glutathione transferase. J Biom Chem 279: 33336-33342. doi: 10.1074/jbc.M404631200

ed

Rowe JD, Patskovsky YV, Patskovska LN, Novikova E, Listowsky I (1998) Rationale for reclassification of a distinctive subdivision of mammalian class Mu glutathione S-transferases that are primarily expressed in testis. J Biol Chem 273: 9593-9601.

pt

Sau A, Pellizzari Tregno F, Valentino F, Federici G, Caccuri AM (2010) Glutathione transferases and development of new principles to overcome drug resistance. Arch Biochem Biophys 500: 116– 122. doi: 10.1016/j.abb.2010.05.012

Ac ce

Sheehan D, Meade G, Foley VM, Dowd CA (2001) Structure, function and evolution of glutathione transferases: implications for classification of an ancient enzyme superfamily. Biochem J 360: 1– 16. doi: 10.1042/0264-6021:3600001 Sherratt, PJ. and Hayes JD (2002) Glutathione S-transferases. In: Ioannides C, editor. Enzyme Systems that Metabolise Drugs and Other Xenobiotics. Chichester, UK: John Wiley & Sons, Ltd. pp. 319–352. Shokeer A, Larsson AK, Mannervik B (2005) Residue 234 in glutathione transferase T1-1 plays pivotal role in the catalytic activity and selectivity against alternative substrates. Biochem J 388: 387-392. Simon P (2003) Q-Gene: processing quantitative real-time RT-PCR data. Bioinformatics 19: 1439– 1440. doi: 10.1093/bioinformatics/btg157 Singhal SS, Gupta S, Ahmad H, Sharma R, Awasthi Y (1990) Characterization of a novel α-class anionic glutathione S-transferase isozyme from human liver. Arch Biochem Biophys 279: 45-53. Staffas L, Ellis EM, Hayes JD, Lundgren B, DePierre JW, Mankowitz L (1998) Growth hormoneand testosterone-dependent regulation of glutathione transferase subunit A5 in rat liver. Biochem J 332: 763-768.

29 Page 29 of 42

Staffas L, Mankowitz L, Söderstörm M, Blanck A, Porsch-Hällstörm I, Sundberg C, Mannervik B, Olin B, Rydström J, DePierre JW (1992) Further characterization of hormonal regulation of glutathione transferase in rat liver and adrenal glands. Biochem J 286: 65-72. Suzuki T, Johnston PN, Board PG (1993) Structure and organization of the human alpha class glutathione S-transferase genes and related pseudogenes. Genomics 18: 680–686. doi: 10.1016/S0888-7543(05)80373-8

ip t

Takahashi Y, Campbell EA, Hirata Y, Takayama T, Listowsky I (1993) A basis for differentiating among the multiple human Mu-glutathione S-transferases and molecular cloning of brain GSTM5. J Biol Chem 268: 8893-8898. Timme-Laragy AR, Goldstone J V, Imhoff BR, Stegeman JJ, Hahn ME, et al. (2013) Glutathione redox dynamics and expression of glutathione-related genes in the developing embryo. Free Radic Biol Med 65: 89–101. doi: 10.1016/j.freeradbiomed.2013.06.011

cr

Tong Z, Board PG, Anders MW (1998) Glutathione transferase Zeta catalyses the oxygenation of the carcinogen dichloroacetic acid to glyoxylic acid. Biochem J 331: 371–374.

us

Wang CG, Chen YX, Li Y, Wei W, Yu Q (2005) Effects of low dose tributyltin on activities of hepatic antioxidant and phase II enzymes in Sebastiscus marmoratus. Bull Environ Contam Toxicol 74: 114-119. doi: 10.1007/s00128-004-0556-0

an

Wells WW, Xu DP, Washburn MP (1995) [4] Glutathione: Dehydroascorbate oxidoreductases. Methods Enzymol 252: 30–38. doi: 10.1016/0076-6879(95)52006-6

M

Whitbread AK, Masoumi A, Tetlow N, Schmuck E, Coggan M, et al. (2005) Characterization of the omega class of glutathione transferases. Methods Enzymol 401: 78–99. doi: 10.1016/S00766879(05)01005-0

ed

Zhao T, Singhal SS, Piper JT, Cheng J, Pandya U, Clark-Wronski J, Awasthi S, Awasthi YC (1999) The role of human glutathione S-transferase hGSTA1-1 and hGSTA2-2 in protection against oxidative stress. Arch Biochem Biophys 367: 216-224. Zimniak P, (2007) Substrates and Reaction Mechanisms of Glutathione Transferases. In: Awasthi YC, editor. Toxicology of Glutathione Transferases. New York: CRC Press. pp. 24–46.

pt

ftp://ftp.ncbi.nlm.nih.gov/genomes/; February 2014 http://www.ensembl.org/info/about/species.html; February 2014 http://genome.jgi.doe.gov/Brafl1/Brafl1.download.ftp.html; February 2014

Ac ce

http://biotech.ou.edu/; September 2014

Figure legends

Figure 1. Phylogenetic tree of the GST superfamily in zebrafish and human. The tree is a part of the comprehensive phylogenetic tree of GST superfamily in all analyzed chordate genomes (Figure S2). Species abbreviations: Hs, Homo sapiens; Dr, Danio rerio. 30 Page 30 of 42

Figure 2. qRT-PCR analysis of gst genes in adult zebrafish: (A) gsta1-3 and gstm1-3, (B) gstp1-2 and gstt1a, (C) gstt1b and gstt2, (D) gsto1 and gsto2, (E) gstz1 and gstr1, (F) mgst1a and mgst1b, (G) mgst2, (H) mgst3a and mgst3b Tissue expression results are presented as mean values ± SEM from 3-5 pools. Each pool was measured in duplicates. MNE stands for mean normalized expression normalized to the housekeeping gene ef1α (elongation factor 1α). Figure 3. Tissue expression profile of gsta1, gsta2, gsta3, gstm1, gstm2, gstm3, gstp1, and gstp2

cr

ip t

in adult male and female zebrafish.

Figure 4. Verification of purified recombinant zebrafish Gst enzymes. (A) SDS-PAGE and (B)

an

us

Western blot analysis of the 9 purified cytosolic zebrafish Gsts.

Figure 5. Michaelis-Menten kinetics of GST-CDNB (1-chloro-2,4-dinitro benzene) conjugation for zebrafish Gsta3, Gstm3 and Gstp2 enzymes. The reaction was measured in the presence of 1 mM

M

reduced glutathione (GSH). Data are representative of tree separate experiments (n=3) and are

ed

given as mean ± SD.

pt

Figure 6. Michaelis-Menten kinetics of zebrafish Gstm1, Gstp1 and Gstt1a for monochlorobimane (MCB) in the presence of 1 mM reduced glutathione (GSH). Data are representative of tree

Ac ce

separate experiments (n=3) and are given as mean ± SD.

Figure 7. Michaelis-Menten kinetics of zebrafish Gstm1, Gstp1 and Gstp2 utilizing reduced glutathione (GSH) in the presence of 1 mM CDNB. Data are representative of tree separate experiments (n=3) and are given as mean ± SD.

Figure 8. Michaelis-Menten kinetics of zebrafish Gstr1. (A) Kinetics of GST-CDNB (1-chloro-2,4dinitro benzene) conjugation (1 mM GSH), (B) GST-MCB (monochlorobimane) conjugation and (C) glutathione (GSH) binding (1 mM CDNB). Data are representative of tree separate experiments (n=3) and are given as mean ± SD. 31 Page 31 of 42

ip t

Tables

Table 1. Number of genes present within each class of GST superfamily in chordates: human (Homo sapiens), chicken (Gallus gallus), anole lizard (Anolis carolinensis), western clawed frog

cr

(Xenopus tropicalis), zebrafish (Danio rerio), stickleback (Gasterosteus aculeatus), green spotted pufferfish (Tetraodon nigroviridis), Japanese pufferfish (Takifugu rubripes), medaka (Oryzias

us

latipes), Florida lancelet (Branchiostoma floridae), and sea squirt (Ciona intestinalis).

Mitochondrial

Microsomal

Mu

Pi

Omega

Theta

Zeta

Rho

Kappa

MAPEG

H.sapiens

5

5

1

2

2

1

-

1

3

G.gallus

4

1

-

1

2

1

-

1

3

A.carolinensis

2

1

2

1

X.tropicalis

3

1

3

2

D.rerio

3

3

2

2

G.aculeatus

1

1

-

T.nigroviridis

1

1

T.rubripes

1

2

O.latipes

1

B.floridae

2

C.intestinalis

3

1

-

1

2

2

1

-

2

3

3

1

1

4

8

1

1

1

5

1

2

-

1

2

1

3

1

2

-

1

2

1

3

1

2

2

-

2

3

1

5

1

3

2

-

4

1

1

3

1

2

3

-

4

1

-

-

1

1

Ac ce

pt

2

ed

Alpha

M

an

Cytosolic

Table 2. Specific activities, substrate affinities (Km) and turnover numbers (kcat) of the 9 zebrafish Gst enzymes characterized with model substrates: (A) 1-chloro-2,4-dinitro benzene (CDNB) and monochlorobimane (MCB) and (B) ethacrynic acid (ETA), dichloroacetic acid (DCA) and dehydroascorbate (DHA). Specific activities for MCB are presented in fluorescent units (FU)/mg protein/min.

32 Page 32 of 42

A CDNB

Gsta3 Gstm1 Gstm3 Gstp1 Gstp2 Gsto2 Gstt1a Gstz1 Gstr1

7.12 ± 0.91 4.31 ± 0.35 3.66 ± 0.33 4.61 ± 0.20 5.70 ± 0.24 1.22 ± 0.23 3.52 ± 0.10

K m (mM)

k cat (s -1)

V max

K m (mM)

(FU*1000/mg protein/min)

1.68 ± 0.47 37.45 2.24 ± 0.35 48.93 18.88 ± 1.29 0.22 ± 0.04 3.69 ± 0.52 43.02 11.04 ± 0.83 0.11 ± 0.03 0.39 ± 0.06 90.42 15.09 ± 1.24 0.25 ± 0.05 0.28 ± 0.05 111.67 19.30 ± 1.64 0.18 ± 0.04 0.32 ± 0.17 5.25 15.97 ± 0.46 0.004 ± 0.0005 0.58 ± 0.05 76.48 9.01 ± 0.42 0.05 ± 0.006

ip t

(µmol/mg protein/min)

cr

V max

Protein

MCB

B DCA

V max V max Protein (µmol/mg K m (mM) k cat (s -1) (µmol/mg protein/min) protein/min)

0.02 ± 0.0003 0.27 ± 0.03 0.002 -

M

an

2.24 1.23 -

K m (mM) k cat (s -1 )

ed

Gsta3 Gstm1 Gstm3 Gstp1 0.34 ± 0.40 0.20 ± 0.07 Gstp2 0.37 ± 0.36 0.24 ± 0.06 Gsto2 Gstt1a Gstz1 Gstr1

DHA

us

ETA

V max (µmol/mg protein/min)

K m (mM) k cat (s-1 )

0.03 ± 0.007 1.08 ± 0.05 11.90 -

Table 3. Kinetic parameters of GST-GSH (glutathione) binding in conjugation reaction with model

pt

substrates for cytosolic zebrafish Gsts. 1-chloro-2,4-dinitro benzene (CDNB) was used as a model substrate except for Gsto2 and Gstz1, where substrates were dehydroascorbate (DHA) and

Ac ce

dichloroacetic acid (DCA), respectively.

GSH -1

Protein

Vmax (µmol/mg protein/min)

Km (mM)

kcat(s )

Gsta3

3.16 ± 0.16

0.75 ± 0.12

16.6

Gstm1

3.21 ± 0.14

0.58 ± 0.08

22.7

Gstm3

2.66 ± 0.20

2.05 ± 0.34

57.2

Gstp1

6.56 ± 0.45

0.82 ± 0.16

129

Gstp2

11.1 ± 0.72

1.19 ± 0.20

14.1

Gsto2

1.51 ± 0.15

2.91 ± 0.54

8.69

Gstt1a

2.62 ± 0.42

4.44 ± 1.20

11.2

Gstz1

0.01 ± 0.006

0.27 ± 0.04

0.005

Gstr1

0.97 ± 0.47

0.50 ± 0.08

20.8

33 Page 33 of 42

Ac ce p

te

d

M

an

us

cr

ip t

Figure(s)

Page 34 of 42

Ac ce p

te

d

M

an

us

cr

ip t

Figure(s)

Page 35 of 42

Ac ce p

te

d

M

an

us

cr

ip t

Figure(s)

Page 36 of 42

Ac

ce

pt

ed

M

an

us

cr

i

Figure(s)

Page 37 of 42

Ac

ce

pt

ed

M

an

us

cr

i

Figure(s)

Page 38 of 42

Ac

ce

pt

ed

M

an

us

cr

i

Figure(s)

Page 39 of 42

Ac

ce

pt

ed

M

an

us

cr

i

Figure(s)

Page 40 of 42

Ac

ce

pt

ed

M

an

us

cr

i

Figure(s)

Page 41 of 42

*Highlights (for review)

We report the first comprehensive characterization of GSTs in zebrafish;



27 zebrafish gst genes are identified and orthology relationships determined:



Expression pattern of gst genes was determined in 6 adult zebrafish tissues;



Functional characterization was performed on 9 cytosolic Gst enzymes;



Gsts that may have role in biotransformation of xenobiotics are identified.

Ac

ce pt

ed

M

an

us

cr

ip t



Page 42 of 42