Toxins and stress in fish: Proteomic analyses and response network

Toxicon 57 (2011) 959–969

Contents lists available at ScienceDirect

Toxicon journal homepage: www.elsevier.com/locate/toxicon

Review

Toxins and stress in fish: Proteomic analyses and response network Mezhoud Karim a, b, Simone Puiseux-Dao a, Marc Edery a, * a

UMR 7245 CNRS-USM 0505 Molécules de communication et adaptation des micro-organismes, Muséum National d’Histoire Naturelle, 12 rue Buffon, F-75231 Paris cedex 05, France Laboratoire de Microbiologie et de Biologie Moléculaire, Centre National des Sciences et Techniques Nucléaires, Sidi Thabet, 2020 Tunis, Tunisia

b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 28 October 2010 Received in revised form 15 March 2011 Accepted 21 March 2011 Available online 30 March 2011

Fish models are increasingly used in toxicological studies in the laboratory as well as in the field. In addition to contributing to the analysis of toxicity mechanisms, one major aim is to select biomarkers from among the metabolic responses to toxic agents observed that could be useful for surveying the aquatic environment. Since proteomics is a developing field in toxicological research, it seems opportune to explore the data obtained using this approach. This article proposes an overview of proteomic studies of fish exposed to environmental stressors comprising a cyanotoxin and the response networks observed. We tend to take a broad view of how proteins communicate and function within the cell, often encompassing large numbers of proteins that operate in pathways. We start by presenting and discussing the data from four experiments in which the medaka fish was treated under the same conditions with the cyanotoxin, microcystin-LR (MC-LR). Liver proteins were analyzed using two techniques: 2D electrophoresis and LCMSMS. In the second and main part of our paper, the proteomic data obtained from fish contaminated with chemicals, including those reported above concerning the medaka fish intoxicated with MC-LR, are considered in the round in order to identify fish responses to chemical stress. A tentative general overview of how groups of proteins work together depending on exposure and/or subcellular location is proposed, with the inclusion of MC-LR data obtained in mice for comparison. Ó 2011 Elsevier Ltd. All rights reserved.

Keywords: Environmental stressors Fish and proteomics

1. Introduction The environment is currently contaminated by myriad chemical and biological pollutants. As a result, organisms are exposed to a diverse array of chemical mixtures which may be simple mixes of just a few identifiable compounds, or may be more complex, containing several hundred related congeners, and/or unrelated compounds. This complexity makes it difficult to characterize the mode or mechanisms of action involved, which are generally based on empirical observations of the toxicities of single chemicals in animal studies. The aquatic environment is a major repository for most of the chemicals generated by human activities. Thus, although estuaries and coastal areas are * Corresponding author. Tel.: þ33 140 793 126; fax: þ33 140 793 594. E-mail address: [email protected] (M. Edery). 0041-0101/$ – see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.toxicon.2011.03.018

important sites of living resources, such as fisheries, they are also those most at risk of toxic contamination. It is difficult to assess this risk, partly because of the complexity of the environment, and partly because of the lack of suitable methods. Chemical analysis can be useful in determining body burdens, unless of course the xenobiotic is bio-transformed, but it does not provide any information about its effects (Katagi, 2010). Ecological monitoring of the local aquatic fauna provides information about the disturbance of homeostatic conditions in natural systems, which is mainly based on the use of biomarkers from sentinel aquatic animals. These biomarkers consist of behavioral, histological and biochemical responses, as well as patterns of protein levels (Livingstone, 1993). Recently, small fish, such as the medaka fish (Oryzias latipes), zebrafish (Danio rerio), mosquito fish (Gambusia affinis), and fathead minnow (Pimephales promelas), have been

960

M. Karim et al. / Toxicon 57 (2011) 959–969

among the most commonly-used models for ecotoxicology and biomedical research (Forne et al., 2010). Using a variety of experimental designs, these aquatic models have been exposed to several of the stressors found in aquatic ecosystems, such as micro algal toxins or xenobiotics, in order to elucidate their toxicity characteristics. Our aim was to explore the data obtained using proteomics in order to point out cellular regulatory network set up after environmental stress. The proteins from proteomic studies were listed and grouped according to their biological function, then they were associated within pathways. All the pathways deduced from proteomic studies were superimposed and common pathways were merged. The distinct pathways sharing common proteins were then connected by their edges to establish pathways networks involved in fish responses to environmental stress. In the conventional biomarker approach, most of the biochemical biomarkers selected are quantitatively modified by chemical stress, since they are relatively easy to handle in laboratory or field experiments. For example, enzymatic activities, hormones such as cortisol or growth factor have been assayed in the blood of Onchorhynchus mykiss after stressful episodes (Fernandes-de-Castilho et al., 2008; Gravel and Vijayan, 2007; Miller et al., 2007). Components of the energy metabolism, such as lactate hydrogenase, glycogen, glucose, lipid peroxidase, aspartate transmitase, and alanine transmitase, were monitored in fish after exposure to salicylate or cadmium (Almeida et al., 2002; Gravel and Vijayan, 2007). Anti-oxidant proteins, such as catalase, glutathione peroxidase and superoxide dismutase, have also been used as stress-related parameters in fish after selenium exposure, as well as the thyroid hormones, T3 and T4 (Miller et al., 2007). This physiological approach has been criticized as demonstrating physiological changes that are not necessarily associated with toxic events. The same objection can be raised to the metallothioneins, which are closely related to metal pollution, but also to sexual maturity (Olsson et al., 1987). Moreover, only a few studies have focused on biomarkers using Western blotting, probably because of the lack of available specific fish antibodies. The Omic approaches, using methods such as genomics and transcriptomics, have made it possible to carry out simultaneous assessments of the expression profiles of numerous genes that respond to a toxic compound within a particular cell type, tissue, or organism (Suter et al., 2004). The data provided by microarrays help us to decipher the signature profile of the modulated genes involved in a stress response (Craig et al., 2009; Iwahashi et al., 2009; Momoda et al., 2007). Thus, signal pathways can be inferred from functional genes. Comparing proteomic data obtained with specific techniques, either 2D-PAGE, DIGE or LCMSMS, with transcriptomic data can sometimes be useful in lower organisms, but is particularly important in higher organisms. In the higher organisms, protein expression is highly regulated, and changes in mRNA and protein levels may not be correlated. The different half-lives of mRNA, as well as protein accumulation, post-translational modifications, and degradation kinetics, all contribute to these differences (Celis et al., 2000). Thus, transcriptomic and proteomic data are complementary.

Toxins induce changes at the level of RNA/protein expression. Decrypting the mechanism of action or molecular targets in environmental relevant organisms via omic techniques is an emerging approach. It provides a lot of data, which is still difficult to compare, but that permits promising synthetic overviews of cellular response networks. The first part of this article deals with the medaka fish model, and proteomic studies of the effects of the cyanotoxin, microcystin-LR (MC-LR). The next section, which constitutes the main part of the article, reports a comprehensive network of proteomic responses deduced from data observed with various different fish and pollutants, including the responses of the exposure of medaka fish to MC-LR, and a toxicological study of MC-LR in mouse liver. 2. Proteomic studies of the effects of the cyanotoxin microcystin-LR on the liver of the medaka fish Aquatic organisms are exposed to microcystins, the most frequent cyanotoxins, by direct ingestion or through trophic chain. Microcystins are known to be hydrophobic and to have relatively high molecular weights (900– 1100Da), and therefore need specific membrane transporters to reach cells such as hepatocytes and renal cells (Xie et al., 2005).Their inhibitory effects on protein phosphatases PP1 and PP2A (Runnegar et al., 1995), ATP synthase (Mikhailov et al., 2003) and aldehyde dehydrogenase (Chen et al., 2006) are dose and time-dependent, and lead to effects including disassembly of the hepatic cytoskeleton, oxidative stress as well as possible DNA damage (Ding et al., 1999; Zegura et al., 2004), apoptosis or necrosis (Malbrouck and Kestemont, 2006). Medaka fish were treated with microcystin-LR (MC-LR) in four successive experiments in which the proteomic analyses focused on the liver. The first treatment provided the following data (Mezhoud et al., 2008a), and was used to design the protocol for the other three exposures: 1) the cyanotoxin was concentrated in the liver, peaking 3h after the toxin had been added to the medium in which the fish were swimming; 2) in order to obtain rigorous levels of contamination, balneation had to be replaced by gavage with a specified quantity of toxin (5 mg MC-LR in 5 mL water) introduced into the stomach via a blunt-tip syringe (Hamilton); 3) in this context of acute toxicity leading to 20% mortality after 2h, metabolic changes could be detected in the livers of the surviving fish by a proteomic approach. Using this protocol, three experiments were performed, each involving triplicates of 12 fish: 6 treated and 6 controls (which were given 5 mL of pure water under the same conditions). In the first two, 2D electrophoresis was used to analyze the proteome and the phosphoproteome of the cytosolic fraction (Mezhoud et al., 2008b), and of the organelle and membranes fraction (Malécot et al., 2009). In the third experiment, gel-free and iTraq procedures were applied to the entire liver (Malécot et al., 2011). Many proteins have been shown to be either up- or downregulated by the treatment compared to controls (Table 1). Clearly the following conclusions can be drawn from the above experiments: 1) as expected, gel-free protein separation with iTraq treatment revealed a higher number of significantly modified proteins; 2) the repeatability of the

M. Karim et al. / Toxicon 57 (2011) 959–969

961

Table 1 Proteomic approach and fish. Fish/Organ

Stressor Acute/chronic

Oryzias latipes Liver

Protein

Accessiona

Change Cellular Function

Methyltransferase 3

Gi:13428840 Q566V3 Q14UD0

U

–

U D

Respiratory electron transport Prevent misfolding

U

Iron storage

U

Cell growth and proliferation

D D

Cell structure Cell structure

U

MC-LR

Cytochrome b

Acute

Chaperonin TCP1

Adult

Ferritin H

2D

Acidic ribosomal phosphoprotein P23 Tubulin beta

1 mg/L

Apolipoprotein Peroxiredoxin GDP dissociation inhibitor NKEF Soul heme binding protein Peptidase M20 APPLE factor XI like Ribosomal protein 10 Oryzias latipes

MC-LR

Phenylalanine hydroxylase

Liver

Selenium binding protein

Acute

Keratin 18

Adult

DJ-1

2D

RKIP

1 mg/g body weight, orally

NKEF Proteasome

U D U U

Iron binding

D

Hydrolase activity

U

Blood coagulation

D

BJ911084 Q6PHI7 BJ007475 Q6PHD9 CAA74664 Q7ZTS4 54792718 Q5W7N7 17354196 Q7KQK1 BJ714211

U U

Structural constituent of ribosome L-phenylalanine catabolic process Selenium binding

U

Cell structure

D

Cell death

D

Kinase activity

D

Cell redox homeostasis

D

Protease

D

Metabolism

D

Cell Structure

U

Glycolysis

(Mezhoud et al., 2008b)

D

Protein kinase C inhibitor

Glucose regulated protein Phenylalanine hydroxylase Aldehyde dehydrogenase ATP synthase Transferrin Beta tubulin

U U D D U D

metabolism metabolism Oxidative stress response Oxidative stress response – Cytoskeleton

40S ribosomal prot

124300851

D

Thiosulfate sulfurtransferase

66706524 D0ZG48 187670465 17357005 Q6NY41 24977713 Q90473 112332544 Q803S0

D

Protein translation & maturation Cytoskeleton

U D

Cholesterol/lipid Metabolism Detoxification

U

Protein translation and maturation Metabolism

Enolase 14-3-3 protein

MC-LR Acute Adult 2D 1 mg/g body weight, orally

17354458 17354922 9935681 Q643S2 AU168347 Q9W6E5 BJ895308 Q8IYS1 BJ708999 P03951 AV671036

Post traductional modification Antioxidant GTPase activity Cell redox homeostasis

(Mezhoud et al., 2008a)

190447 Q6DG91 BJ729370 Q6IQG6 BJ729321 Q6IQN9 BJ722994 Q6TH14 BJ727482 Q5PRD0 110226520 32442452 GI:157278379 47218629 171544935 10242160

Hypoxanthine guanine phophoribosyl transferase Actin capping protein

Oryzias latipes Liver

66696400 Q7ZYX4 AM149448 A7YYB6 81491024 Q9PV90 9936419 66695132 Q8UUK8 66692832

Reference

Apolipoprotein A1 Cytochrome b5 Heat shock cognate 71 Fumarylacetoacetase

U

(Malécot et al., 2009)

(continued on next page)

962

M. Karim et al. / Toxicon 57 (2011) 959–969

Table 1 (continued ) Fish/Organ

Oryzias latipes Liver

Danio rerio Liver

Stressor Acute/chronic

MC-LR Acute Adult LCMSMS 1 mg/g body weight, orally

MC-LR Subchronic Adult 2D 2 or 20 mg/L

Danio rerio Brain

Ethanol Chronic Adult 2D-DIGE 0.5 % (v/v)

Danio rerio Liver

Protein

Accessiona

Change Cellular Function

Prohibitin Protein disulfide A4

41152028 66753090 Q7ZVH2 157311655 O42363

U U

DNA synthesis Cell redox homeostasis

U U

Immunity Lipid metabolism

Q8AYH1

D

Q4V8X4 Q6AXL2 Q803F6 Q6DBT3 Q9PUC1 Q6NVU3

U U D U D U

Q9PUC1 Q5XJ54 B0V1J9 Q32LT1

D U U D

Glutathione metabolic process

Q6DG85 Q5XJ10

D D

Glycolysis

Q9DG46 B0S8J5

D U

Q80360 P15947 Q6NYS4

D U U

Blood Protease inhibitor

P00920 Q8JIP9

D U

Carbonic anhydrase Exocytosis

O93444 Q29R95 Q6PHH4 Q6P5J0 Q68EK1

U U U D U

A8KC20 Q6NY15 Q5U403

D U D

Q504D4 Q6NWD1 Q7SYE2 Q7ZVJ0 Q6P0S2

D D D D U

Q8AWD0

U

O42363 Q7ZUW8

U D

Q9IAC1

U

Q6PC35 O42248

D U

Q6TH14 Q803D7 Q3YMK9 Q6P0H2

U U D D

Response to many stressj ATP binding ATP synthesis WD domain G protein Glycolysis Metabolism/reductase Iron metabolism Antioxidant Defense

Q6DHT4 XP_700749.1 AAP59458 AAH59671 Q6TGZ5

D U U U D

Metabolism/reductase Aminoacid metabolism NADPH metabolism Antioxidant Defense Phenylalanine catabolism

Complement C3-1 Apolipoprotein A-1 17-beta-hydroxysteroid dehydrogenase 4 like acyl-Coenzyme A binding protein Heat-responsive protein 12 40S ribosomal protein S2 Thioredoxin domain containing 2 Calreticulin Heat shock 70 kDa-protein 9 (mortalin) Calreticulin Glutaredoxin Glutathione S-transferase theta 1 Cytochrome P450, family 3, subfamily A Uricase Glyceraldehyde-3-phosphate dehydrogenase Homogentisate 1,2-dioxygenase Aminocarboxymuconate semialdehyde decarboxylase Fumarylacetocetate hydrolase Kininogen like Phosphatidylethanolamine binding protein 1 Carbon dioxide transport Warm-temperature-acclimationrelated-65 kDa-protein Annexin max 1 Zgc:136933 Suclg2 Amylase 6-phosphogluconolactonase Pyruvate kinase 2-hydroxyacyl-CoA Diphosphomevalonate decarboxylase Elastase 3 like Cpb1 Actinin Profilin Voltage dependent anion channel 1 Voltage dependent anion channel 2 Apolipoprotein-A1 Glutamic-oxaloacetic acid transaminase-1 Heat shock protein 70

Hþ transporting ATPase Guanine nucleotide binding protein BFRs(textile) Enolase Acute Sb-cb825 met/reductase Adult Iron Regulatory Protein 2D NADHP(H) dehydrogenase 10 and 100 nmol/g, quinine 1 orally Zgc-92082 Formiminotransferase Transketolase Peroxiredoxin 6 4-Hydroxyphenylpyruvate dioxygenase

Reference

(Malécot et al., 2011)

Translation, maturation and degradation of proteins

Amino acid metabolism

Methyltransferase ATP binding/ligase activity Catalytic Activity Phosphogluconolactonase activity Glycolysis Fatty acid oxydation Lipid biosynthesis

(Wang et al., 2009)

Serine endopeptidase activity Metallocarboxypeptidase Actin Binding Actin cytoskeleton (Damodaran et al., 2006)

Biosynthetic process

(Kling et al., 2008)

M. Karim et al. / Toxicon 57 (2011) 959–969

963

Table 1 (continued ) Fish/Organ

Oncorhynchus mykiss Liver

Gadus morhua Plasma

Stressor Acute/chronic

Sewage Chronic Juvenle 2D 10 ng/L 17aethinylestradiol Crude sea oil (PAHs) Chronic Juvenile 2D 10 mg/L 17boestradiol

Gadus morhua

Produced water (oestradiol) Chronic Embryos /larves 2D 0.06–1 mg oil/kg

Oreochromis mossambicus Serum

Osmotic Acute Adult 2D 25 ng/L NaCl

Rainbow trout gill

Carassius carassius Brain

X-ray acute/bystander adult 2D Bystander effect

Anoxia Acute Adult 2D <0.1 mg O2/L

Protein

Accessiona

Change Cellular Function

Betaine-Homocysteine Aldehyde dehydrogenase Transketolase Bardet-Biedl syndrome 2 protein homolog Betaine aldehyde dehydrogenase Lactate dehydrogenase Carbonyl reductase/20-bhydroxysteroid Mitochondrial ATP synthase alpha Fibrinogen gamma polypeptide Macroglobulin 1

Q5PSM1 Q66121 AAP59458 Q98SP7

D D D

P56533 Q9YGL2 Q9PT36

D D D

Oxidoreductase Pyruvate to lactate Oxydoreductase

Q910C4

U

Produce ATP from ADP

Q7ZVG7 Q9PVU5

D/U D

Protein binding Endopeptidase inhibitor activity

Pentraxin Apolipoprotein B

Q90YD1 Q91480

U U

Serotransferrin Orla C3-1

Q92079 Q9IBH1

D D

Prothrombin Fibrinogen Antiproteinase

Q5NKF9 Q6NYE1 Q7ZZW1

D U U

Psp protein Apolipoprotein A

XP_001341390 D AAU87042.1 U

Myosin ATP synthase Krt4 Myosin Keratin k8b Alpha actinin Heat shock cognate Actin Myosin Rhotekin-2 Semaphorin 3aa ATP synthase NADH dehydrogenase C3

XP_708916.1 Q9PTY0 Q6NY60 XP_708916.1 Q91219 Q8AX99 P47773 P49055 Q6SNT2 Q5XIZ9 Q9W7J1 P00847 Q6P613 Q98TS6

U D D U D U/D D U U U U U U U

Mg dependent neutral sphingomyelinase Caspase 3 RGD-CAP Kappa light chain variable region Trypsin precursor Immunoglobulin D Junction DNA helicase RUVB Cytochrome C oxidase Beta-tubulin Alcohol dehyrogenase

Q4LEU0

U

Q1KZF6 O42390 AAK61529 P00761 AAF72565 AAK61256 Gi:825700 Gi:56603670 Gi:929845

U U U U D D D D D

Catalase Annexin II

Gi:46909301 Q804G9

U U

Hemopexin-like RhoGDI

Q9DFF1 Q6P3J2

U U

PDH

P08559

U

Chromosome 1 SCAF Creatine kinase Fructose-bisphosphate aldolase c Glyceraldehyde-3-phosphate dehydro Triosephophate isomerase Lactate dehydrogenase A

Q4SQV2 Q8AY63 Q8JH70 Q4VSK0

U D D D

Post traductional modification Iron metal binding Endopeptidase inhibitor activity Protease calcium ion binding Platelet activation Endopeptidase inhibitor activity – Post traductional modification Cell motility ATP synthesis Structural molecule activity Cell motility Structural molecule activity Actin binding Stress response/ATP-binding Cell motility Cell motility – Receptor activity ATP synthesis oxidoreductase Endopeptidase inhibitor activity Ceramide biosynthesis process Protease – – Protease – Helicase activity Respiration chain Cytoskeleton Oxidation od EtOH to acetaldehyde Convert the peroxide to water Calcium binding cytoskeletal protein dinding – Rho GDP-dissociation inhibitor activity Pyruvate dehydrogenase complex Nucleic acid binding Creatine kinase activity Glycolysis Glycolysis

Q70I40 Q068V9

D D

Gluconeogenesis Glycolysis

Reference

Aminoacid metabolism Cellular defense NADPH metabolism Microtubule motor activity

(Albertsson et al., 2007)

(Bohne-Kjersem et al., 2009a)

(Bohne-Kjersem et al., 2009b)

(Kumar et al., 2009)

(Smith et al., 2007)

(Smith et al., 2009)

(continued on next page)

964

M. Karim et al. / Toxicon 57 (2011) 959–969

Table 1 (continued ) Fish/Organ

Stressor Acute/chronic

Carassius auratus Effluent Liver wastewater Acute Adult 2D In situ dose from lake

Protein

Accessiona

Change Cellular Function

Voltage-dependent ion channel Gefiltin Dihydropyrimidinase VAT-1 Ependymin-1 Annexin 4

Q6NWC1 Q07962 Q52PJ5 Q4KME8 P13506 Q804G7

U D D D U U

Superoxide dismutase Peptidyl-proly Betaine homocysteine methyltransferase Proteasome Glutathione Peroxidase Ferritin H

O73872 Q499A7 Q32LQ4

U U D

Q6DHI9 Q802G3 Q9W6X5

14-3-3

Q6UFZ5

D D U/D PTM D

Liver basic fatty acid binding

P80856

U/D

Anion Transporter Structural Molecule Activity Signaling by semaphorin Oxidoreductase activity Cell-matrix adhesion Calcium/phospholipid binding Antioxidant defense Protein folding Amine and polyamine degradation Protease Oxidoreductase Ferric iron binding Oxidoreduction Protein domain specific binding Lipid binding transporter activity

Reference

(Wang et al., 2008)

BFRs: brominated flame retardants. D, Down regulated, U, Up regulated. a Uniprot or NCBI accession.

data is fairly limited, since the proteins identified as having been modified were usually only found to be changed in 2 out of each triplicate of samples, especially in the case of the 2D electrophoreses. However, what is clear is that the various proteins identified could belong to a stress response domain in fish, which is clearly demonstrated in the following global analysis of fish responses to chemical stress. 3. Data mining from stressed fish proteomes The data were collected from publications available in peer-reviewed reviews. Tables 1-3 summarize all the data on which this analysis is based. In Table 1, proteomic studies based on 8 fish models exposed to 8 environmental stressors are shown. Studies concerning acute exposure are reported for O. latipes (liver), D. rerio (liver), Oreochromis mossambicus (serum), Rainbow trout (gill), Carassius carassius (brain), Carassius auratus (liver) exposed to microcystin-LR, a mixture of brominates flame retardants (effluent of textile industry), osmotic stress, X-ray, anoxia and wastewater effluent, respectively. Studies related to chronic exposure concern D. rerio (liver and brain) was exposed to microcystins-LR and ethanol, Gadus morhua (plasma and embryos) exposed to crude sea oil (PAHs) or estradiol, and Oncorhynchus mykiss (liver) exposed to sewage. Table 2 outlines a mammalian (Mus musculus, liver) response to acute exposure to MC-LR. Table 3 summarizes studies based on fishes and non-proteomic analysis. In acute treatment, O. mykiss (blood and liver) was exposed to salicylate. In the chronic tests, Oreochromis niloticus (muscle and liver) and O. mykiss (blood) were exposed to cadmium and social stress, respectively. In fact the effects of chronic and acute exposure could be considered together since in the conditions used, response differences between the two types of intoxications were low. The proteins were classified in terms of the biological process in which they are involved, using uniprot databases (Consortium, 2009; Jain et al., 2009), and localized on

pathways using Wikipathways databases (Kelder et al., 2009; Pico et al., 2008), accessible through a plugin cytoscape software used for network visualization and analysis (Cline et al., 2007; Shannon et al., 2003). The corresponding pathways were determined using complementary plugins as super pathways, which made it possible search and select multiple pathways from the Wikipathways database (Kelder et al., 2009). The search also included data reported with MC-LR on D. rerio and on M. musculus for comparison. The proteins common to the pathways found were then identified using BridgeDb framework database (version Dr_Derby_20100601 for D. rerio and Mm_Derby_20100601 for M. musculus) (van Iersel et al., 2010). The resulting network is shown in Fig. 1. The nodes (pathways) colored in yellow are those common to the fish and the mammalian models, and those in green and red are those specific to fish or mice respectively. The network was positioned using the Cytoscape organic layout style. The interactome revealed diverged into two main groups. The pathways on the left are involved in metabolism, and those on the right concern physiological processes, such as oxidative stress and cellular processes, such as division. The pathways are connected to various other nodes by several shared proteins, the number of which is indicated on the connection lines. 4. Metabolic pathways are the first target of nonspecific, environmental stress In most cases, toxic molecules reach the blood either through the gills (mainly hydrophilic molecules) or through the intestine (mainly hydrophobic molecules). Once in the blood, toxic substances reach the filter organs, such as the liver and kidney, which are both essential organs in the management of the energy metabolism. As described by Barton (Barton, 2002), low doses and chronic exposure clearly produce primary responses, which involve the initial neuro-endocrine responses, including the release

Table 2 Toxins - Proteomic approach with a mammalian model. Stressor Acute/chronic

Protein

Accession

Change

Biological process

Reference

Mus musculus Liver

MC-LR Acute Adult 2D 50–70 mg/kg injected

Heat shock protein 1 (chaperonin 10) Beta1Globin Hemoglobin NL minor chain Ribonuclease UK114 Hemoglobin alpha Hemoglobin beta Profilin 1 Destrin Ferritin Heavy Chain Peroxiredoxin 2

Q64433 P02088 P02089 P52760 Q9CY10 Gij229301 P62962 Q9R0P5 P09528 Q61171

U U U U U U U U U U

(Chen et al., 2005)

Ferritin light chain NADH dehydrogenase Fe-S protein 8 Autophagy-related protein 16-1 Glutathione S-transferase kappa 1 Proteasome subunit alpha type-5

P29391 Q8VC72 Q8C0J2 Q9DCM2

U U U U

Protein folding/response to stress Oxygen transport/heme binding Oxygen transport Endonuclease Oxygen transport Oxygen transport Actin Cytoskeleton organization Actin-depolymerizing protein Iron storage : homeostasis Anti-apoptosis Cell redox homeostasis Iron storage Iron ion binding Autophagic vacuole assembly Glutathione metabolic process

P28066

U

Annexin A5

P48036

U

Chloride intracellular channel protein 1 Triosephosphate isomerase Proteasome activator complex subunit 2

Q9Z1Q5

U

P17751 P97372

U U

Beta-lactamase-like protein 2 Carbonic anhydrase 2 Glyceraldehyde-3-phosphate dehydrogenase Thiosulfate sulfurtransferase Actin L-lactate dehydrogenase

Q99KR3 P00920 P16858 P52196 P60710 Q99K20

U U U U U U

Heat shock protein 90, beta (Grp94), member 1 Phosphatidylethanolamine-binding protein 1 Tetratricopeptide repeat protein 36 Crinkled Peroxiredoxin-1 Glutathione S-transferase P 1 Lactoylglutathione Lyase

Q91V38

U

Fatty acid biosynthesis Antigen processing and presentation of exogenous antigen Hydrolase activity/metal binding Carbon dioxide transport Glycolysis Sulfate transport Cell structure Cellular reponse to extracellular stimulus Glycolysis Protein folding

P70296 Q8VBW8 Q5D0F1 P35700 P19157 Q9CPU0

D D D D D D

Protease inhibitor – Receptor Cell proliferation/redox homeostasis Glutathione Metabolic Process Anti-Apoptosis

Threonine-type endopeptidase activity Anticoagulant protein Calcium binding Ion transport

M. Karim et al. / Toxicon 57 (2011) 959–969

Animal/Organ

965

966

M. Karim et al. / Toxicon 57 (2011) 959–969

Table 3 Non proteomic approach (Fish/Environmental contaminant). Fish/Organ

Stressor Acute/chronic

Protein

Change Change Liver

Oreochromis niloticus Muscle/Liver

Oncorhynchus mykiss

Oncorhynchus mykiss Blood/Liver

Oncorhynchus mykiss Blood/Liver

Cadmium Chronic sublethal Juvenile Enzymatic tests 320–2560 mg/L

Social stress Blood Chronic Adult 16 fishes/1500 L Salicylate Acute Adult 100 mg/kg body weight, orally

Selenium Subchronic Juvenile 0.16 mg/L

– U –

Red Muscle – U U

D

U

U

–

–

U

U

D

U

U

D U D

–

Glycogen Glucose Lactate dehydrogenase LDH-E.C.1.1.1.27 Creatine kinase CK-E.C.2.7.3.2 Alanine transminase ALT-E.C.2.6.1.2. Aspartate transminase AST-E.C.2.6.1.1 Superoxide dismutase SOD Glutathione peroxidase GSH-Px Lipoperoxide Cortisol Growth hormone Glucose

–

Cortisol Glucose Lactate StAR Peripheral-type benzodiazepine receptor P450 Liver glycongen Liver Aspartate aminotransferase Liver Alanine aminotransferase Glucocorticoid receptor Cortisol Glucose T3 T4 Gill Na/K ATPase Liver glutathione Liver Glutathione peroxidase Liver Lipid peroxidation

U – – D –

of catecholamines from chromaffin tissue (Reid et al., 1998; Rotllant et al., 2000), and the stimulation of the hypothalamic-pituitary-interrenal (HPI) axis, culminating in the release of corticosteroid hormones into the circulation (Pankhurst, 2010). In the interactome pathways (Fig. 1), cortisol could be involved in the nodes of the prostaglandin (WP374) and glucocorticoid (WP495) metabolisms. Its primary functions result in an increase in blood sugar, through gluconeogenesis, a slowdown of the immune system, and an activation of the lipid, protein and carbohydrate metabolisms. These metabolic pathways are connected to the biotransformation metapathway (WP1251) that comprises several defense pathways, such as cytochrome P450 (WP1274), glutathione transferase, glutathione peroxidase, glutathione reductase, glutathione synthetase and sulfotransferase (all of which are included in WP730). They are directly involved in the physiological response to xenobiotic exposure. The position of the biotransformation metapathway in the interactome (Fig. 1) and the large number of its connections (15) indicate its importance during stress episodes in fish. The pathways of

– D D D D U U – – – D D –

–

Cellular function

Reference (Almeida et al., 2002)

(Fernandes-de-Castilho et al., 2008)

(Gravel and Vijayan, 2007)

(Miller et al., 2007)

metabolism (left), and of physiological and cell processes (right), are driven from this node. On the left, the amino acid metabolism (WP662), urea cycle and amino acid metabolism (WP426) ensure the synthesis of amino acids and urea recycling, which are essential to complete the active physiological events induced by oxidative stress pathways (WP412), and for muscle contraction (WP216). These pathways require energy resources obtained by glycolysis and gluconeogenesis (WP157) and by fatty acid biosynthesis (WP336). From these nodes, the remaining metabolic pathways are indirectly involved in a non-specific response to stress. It is not surprising that each of these pathways plays an important role in the energy metabolism, since one of the first aspects of fish adaptation to stress involves the activation of liver glycogenolysis and gluconeogenesis, in order to compensate for the consumption of glucose. Several studies have confirmed an increase in the glucose export capacity, for example in the liver of fish reared under high density conditions, resulting in a higher blood sugar level (Arends et al., 1999; Rotllant et al., 2001; Sangiao-Alvarellos et al., 2005).

M. Karim et al. / Toxicon 57 (2011) 959–969 Fig. 1. The merge interactome of pathways involved in environmental stress using fish and mammalian models. The nodes were disposed in an organic layout style. Each node corresponds to a pathway. The number indicated on each edge corresponds to the number of common proteins. The pathways colored yellow were shared by the fish and mammalian models, the green and red ones were specific to the fish and mammalian models, respectively. M, Metabolism; EM, Energetic Metabolism; Cell. P, cellular process; Physio. P, Physiological process; Mol. Fun, Molecular Function.

967

968

M. Karim et al. / Toxicon 57 (2011) 959–969

5. Activation of metabolic processes leads to cell activation and physiological responses Several proteins in the energy metabolism pathways (WP63, WP435, WP434, and WP157) that are induced after environmental stress are also involved in the redox reaction, such as the mitochondrial electronic transport chain (WP295), and fatty acid biosynthesis (WP336, WP401). ATP synthase from the WP336 pathway is frequently listed as being modified in the fish proteome exposed to environmental stress. It synthesizes ATP from ADP using the electrochemical gradient of the mitochondrial membrane. In the muscles, the redox reaction cascade activates the calcium regulation pathways in the cardiocytes (WP553), and the myometrial relaxation and contraction pathways (WP385), which are regulated by the voltage-dependent, anion channel protein. Over-induction of these physiological processes related to respiration, energy production and muscle contraction inevitably leads to reactive oxygen stress (ROS), considered to be the main engine of cellular stress (WP412). High levels of oxidative stress proteins (thioredoxin, aldehyde dehydrogenase, superoxide dismutase, DJ-1 and peroxiredoxin) are often associated with cytoskeletal disruption. As the Fas pathways and stressrelated induction of HSP regulation (WP571 are the first biochemical tools involved in reducing ROS, it is hardly a surprise to find that their modulation has the effect of stabilizing cytoskeletal elements, such as actin and P23 (WP523) and also the tubulins. The biotransformation (WP1251) pathway, comprising the cytochrome P450 (WP1274) and oxidative stress (WP412) pathways, which are both markedly involved in cell defense, are interconnected; the WP412 pathway being related to MAPK signaling (WP493). This node is implicated in the regulation of the actin cytoskeleton (WP523; 27 common proteins), and is strongly connected to IL-2 signaling pathways (WP450), suggesting a possible immune response. MAPK signaling (WP493) is also related with G1 to S cell cycle control (WP413). A package arising from this node involves “DNA replication (WP150) and cell cycle (WP190)”. Remarkably the cellular events associated with IL-2 signaling and the cell cycle indicating cellular activation, have been described in fish but not in the mouse, where, in contrast, senescence and autophagy (WP1267) have been reported. These findings suggest that the experimental conditions were probably less drastic in the fish experiments than when the mice were treated IP with MC-LR. However, the overall stress response in teleost fish has many similarities to that in a terrestrial vertebrate. 6. Conclusion This review clearly indicates that fish use a set of biochemical tools to ensure their safety, as do other vertebrates. If we refer to the experiments on fish responses reported in O. niloticus or O. mykiss (Table 3) using classical techniques, the main biochemical changes induced by contamination with cadmium, salicylate or selenium involve a hormonal component especially related to corticosteroids, an increase in the glucide and amino acid metabolism as well as the implication of oxidative stress.

Although relatively few modifications were observed compared to long set of modulated proteins obtained using the proteomic approach, the response domains fit very well with Fig. 1. Listing of proteins detected with proteomics approach does not permit deciphering physiological response or mechanism of action. These processes require determining how these proteins interact between them and also as a function of time. This aim implies to integrate data from in vivo, in vitro and in silico methods with computational approaches. This work shows how data in their contextual scenario based on wikipathway database can be integrated. The resulting Fig. 1 provides an overview of the physiological functions progressing in environmental stress conditions. It appears that higher-throughput of protein screening in toxicology could be considered with automated technology that would integrate data from different areas such as interactomics, pathwayomics or physiology response modeling. Conflict of interest The authors declare no conflict of interest. Acknowledgments This work was supported by grants from the ANR 07 SEST CYANOTOX 005, the AFSSET/APR EST 2007 10 and the ANSES EST 2010/2/002, to Dr. Marc Edery. This paper is a contribution to the IRD research group LOPB UMR6535. References Albertsson, E., Kling, P., Gunnarsson, L., Larsson, D.G., Forlin, L., 2007. Proteomic analyses indicate induction of hepatic carbonyl reductase/ 20beta-hydroxysteroid dehydrogenase B in rainbow trout exposed to sewage effluent. Ecotoxicol Environ. Saf. 68, 33–39. Almeida, J.A., Diniz, Y.S., Marques, S.F., Faine, L.A., Ribas, B.O., Burneiko, R. C., Novelli, E.L., 2002. The use of the oxidative stress responses as biomarkers in Nile tilapia (Oreochromis niloticus) exposed to in vivo cadmium contamination. Environ. Int. 27, 673–679. Arends, R.J., Mancera, J.M., Munoz, J.L., Wendelaar Bonga, S.E., Flik, G., 1999. The stress response of the gilthead sea bream (Sparus aurata L.) to air exposure and confinement. J. Endocrinol. 163, 149–157. Barton, B.A., 2002. Stress in fishes: a Diversity of responses with particular Reference to changes in circulating corticosteroids. Integr. Comp. Biol. 42, 517–525. Bohne-Kjersem, A., Bache, N., Meier, S., Nyhammer, G., Roepstorff, P., Saele, O., Goksoyr, A., Grosvik, B.E., 2009a. Biomarker candidate discovery in Atlantic cod (Gadus morhua) continuously exposed to North Sea produced water from egg to fry. Aquat Toxicol. 96, 280–289. Bohne-Kjersem, A., Skadsheim, A., Goksoyr, A., Grosvik, B.E., 2009b. Candidate biomarker discovery in plasma of juvenile cod (Gadus morhua) exposed to crude North Sea oil, alkyl phenols and polycyclic aromatic hydrocarbons (PAHs). Mar Environ. Res. 68, 268–277. Celis, J.E., Kruhoffer, M., Gromova, I., Frederiksen, C., Ostergaard, M., Thykjaer, T., Gromov, P., Yu, J., Palsdottir, H., Magnusson, N., Orntoft, T. F., 2000. Gene expression profiling: monitoring transcription and translation products using DNA microarrays and proteomics. FEBS Lett. 480, 2–16. Chen, T., Cui, J., Liang, Y., Xin, X., Owen Young, D., Chen, C., Shen, P., 2006. Identification of human liver mitochondrial aldehyde dehydrogenase as a potential target for microcystin-LR. Toxicology 220, 71–80. Chen, T., Wang, Q., Cui, J., Yang, W., Shi, Q., Hua, Z., Ji, J., Shen, P., 2005. Induction of apoptosis in mouse liver by microcystin-LR: a combined transcriptomic, proteomic, and simulation strategy. Mol. Cell Proteomics 4, 958–974. Cline, M.S., Smoot, M., Cerami, E., Kuchinsky, A., Landys, N., Workman, C., Christmas, R., Avila-Campilo, I., Creech, M., Gross, B., Hanspers, K., Isserlin, R., Kelley, R., Killcoyne, S., Lotia, S., Maere, S., Morris, J., Ono, K., Pavlovic, V., Pico, A.R., Vailaya, A., Wang, P.L., Adler, A.,

M. Karim et al. / Toxicon 57 (2011) 959–969 Conklin, B.R., Hood, L., Kuiper, M., Sander, C., Schmulevich, I., Schwikowski, B., Warner, G.J., Ideker, T., Bader, G.D., 2007. Integration of biological networks and gene expression data using Cytoscape. Nat. Protoc. 2, 2366–2382. Consortium, T.U., 2009. The Universal protein resource (UniProt) 2009. Nucleic Acids Res. 37, D169–174. Craig, P.M., Hogstrand, C., Wood, C.M., McClelland, G.B., 2009. Gene expression endpoints following chronic waterborne copper exposure in a genomic model organism, the zebrafish, Danio rerio. Physiol. Genomics 40, 23–33. Damodaran, S., Dlugos, C.A., Wood, T.D., Rabin, R.A., 2006. Effects of chronic ethanol administration on brain protein levels: a proteomic investigation using 2-D DIGE system. Eur. J. Pharmacol. 547, 75–82. Ding, W.X., Shen, H.M., Zhu, H.G., Lee, B.L., Ong, C.N., 1999. Genotoxicity of microcystic cyanobacteria extract of a water source in China. Mutat. Res. 442, 69–77. Fernandes-de-Castilho, M., Pottinger, T.G., Volpato, G.L., 2008. Chronic social stress in rainbow trout: does it promote physiological habituation? Gen. Comp. Endocrinol. 155, 141–147. Forne, I., Abian, J., Cerda, J., 2010. Fish proteome analysis: model organisms and non-sequenced species. Proteomics 10, 858–872. Gravel, A., Vijayan, M.M., 2007. Salicylate impacts the physiological responses to an acute handling disturbance in rainbow trout. Aquat Toxicol. 85, 87–95. Iwahashi, H., Kishi, K., Kitagawa, E., Suzuki, K., Hayashi, Y., 2009. Evaluation of the physiology of Medaka as a model animal for standardized toxicity tests of chemicals by using mRNA expression profiling. Environ. Sci. Technol. 43, 3913–3918. Jain, E., Bairoch, A., Duvaud, S., Phan, I., Redaschi, N., Suzek, B.E., Martin, M.J., McGarvey, P., Gasteiger, E., 2009. Infrastructure for the life sciences: design and implementation of the UniProt website. BMC Bioinform. 10, 136. Katagi, T., 2010. Bioconcentration, bioaccumulation, and metabolism of pesticides in aquatic organisms. Rev. Environ. Contam. Toxicol. 204, 1–132. Kelder, T., Pico, A.R., Hanspers, K., van Iersel, M.P., Evelo, C., Conklin, B.R., 2009. Mining biological pathways using WikiPathways web services. PLoS One 4, e6447. Kling, P., Norman, A., Andersson, P.L., Norrgren, L., Forlin, L., 2008. Genderspecific proteomic responses in zebrafish liver following exposure to a selected mixture of brominated flame retardants. Ecotoxicol Environ. Saf. 71, 319–327. Kumar, V.B., Jiang, I.F., Yang, H.H., Weng, C.F., 2009. Effects of serum on phagocytic activity and proteomic analysis of tilapia (Oreochromis mossambicus) serum after acute osmotic stress. Fish Shellfish Immunol. 26, 760–767. Livingstone, D.R., 1993. Biotechnology and pollution monitoring: use of molecular biomarkers in the aquatic environment. J. Chemical Technology Biotechnology 57, 195–211. Malbrouck, C., Kestemont, P., 2006. Effects of microcystins on fish. Environ. Toxicol. Chem. 25, 72–86. Malécot, M., Marie, A., Puiseux-Dao, S., Edery, M., 2011. iTRAQ-based proteomic study of the effects of microcystin-LR on medaka fish liver. Proteomics. doi:10.1002/pmic.201000512. Malécot, M., Mezhoud, K., Marie, A., Praseuth, D., Puiseux-Dao, S., Edery, M., 2009. Proteomic study of the effects of microcystin-LR on organelle and membrane proteins in medaka fish liver. Aquat Toxicol. 94, 153–161. Mezhoud, K., Bauchet, A.L., Chateau-Joubert, S., Praseuth, D., Marie, A., Francois, J.C., Fontaine, J.J., Jaeg, J.P., Cravedi, J.P., Puiseux-Dao, S., Edery, M., 2008a. Proteomic and phosphoproteomic analysis of cellular responses in medaka fish (Oryzias latipes) following oral gavage with microcystin-LR. Toxicon 51, 1431–1439. Mezhoud, K., Praseuth, D., Puiseux-Dao, S., Francois, J.C., Bernard, C., Edery, M., 2008b. Global quantitative analysis of protein expression and phosphorylation status in the liver of the medaka fish (Oryzias latipes) exposed to microcystin-LR I. Balneation study. Aquat Toxicol. 86, 166–175.

969

Mikhailov, A., Harmala-Brasken, A.S., Hellman, J., Meriluoto, J., Eriksson, J. E., 2003. Identification of ATP-synthase as a novel intracellular target for microcystin-LR. Chem. Biol. Interact 142, 223–237. Miller, L.L., Wang, F., Palace, V.P., Hontela, A., 2007. Effects of acute and subchronic exposures to waterborne selenite on the physiological stress response and oxidative stress indicators in juvenile rainbow trout. Aquat Toxicol. 83, 263–271. Momoda, T.S., Schwindt, A.R., Feist, G.W., Gerwick, L., Bayne, C.J., Schreck, C.B., 2007. Gene expression in the liver of rainbow trout, Oncorhynchus mykiss, during the stress response. Comp. Biochem. Physiol. Part D Genomics Proteomics 2, 303–315. Olsson, P.E., Haux, C., Förlin, L., 1987. Variations in hepatic metallothionen, zinc and copper levels during an annual reproductive cycle in rainbow trout,Salmo gairdneri. Fish Physiollogy Biochem. 3, 39–47. Pankhurst, N.W., 2010. The endocrinology of stress in fish: an environmental perspective. Gen. Comp. Endocrinol.. Pico, A.R., Kelder, T., van Iersel, M.P., Hanspers, K., Conklin, B.R., Evelo, C., 2008. WikiPathways: pathway editing for the people. PLoS Biol. 6, e184. Reid, S.G., Bernier, N.J., Perry, S.F., 1998. The adrenergic stress response in fish: control of catecholamine storage and release. Comp. Biochem. Physiol. C Pharmacol. Toxicol. Endocrinol. 120, 1–27. Rotllant, J., Balm, P.H., Perez-Sanchez, J., Wendelaar-Bonga, S.E., Tort, L., 2001. Pituitary and interrenal function in gilthead sea bream (Sparus aurata L., Teleostei) after handling and confinement stress. Gen. Comp. Endocrinol. 121, 333–342. Rotllant, J., Balm, P.H., Ruane, N.M., Perez-Sanchez, J., Wendelaar-Bonga, S. E., Tort, L., 2000. Pituitary proopiomelanocortin-derived peptides and hypothalamus-pituitary-interrenal axis activity in gilthead sea bream (Sparus aurata) during prolonged crowding stress: differential regulation of adrenocorticotropin hormone and alpha-melanocytestimulating hormone release by corticotropin-releasing hormone and thyrotropin-releasing hormone. Gen. Comp. Endocrinol. 119, 152–163. Runnegar, M., Berndt, N., Kong, S.M., Lee, E.Y., Zhang, L., 1995. In vivo and in vitro binding of microcystin to protein phosphatases 1 and 2A. Biochem. Biophys. Res. Commun. 216, 162–169. Sangiao-Alvarellos, S., Guzman, J.M., Laiz-Carrion, R., Miguez, J.M., Martin Del Rio, M.P., Mancera, J.M., Soengas, J.L., 2005. Interactive effects of high stocking density and food deprivation on carbohydrate metabolism in several tissues of gilthead sea bream Sparus auratus. J. Exp. Zool. A Comp. Exp. Biol. 303, 761–775. Shannon, P., Markiel, A., Ozier, O., Baliga, N.S., Wang, J.T., Ramage, D., Amin, N., Schwikowski, B., Ideker, T., 2003. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 13, 2498–2504. Smith, R.W., Cash, P., Ellefsen, S., Nilsson, G.E., 2009. Proteomic changes in the crucian carp brain during exposure to anoxia. Proteomics 9, 2217–2229. Smith, R.W., Wang, J., Bucking, C.P., Mothersill, C.E., Seymour, C.B., 2007. Evidence for a protective response by the gill proteome of rainbow trout exposed to X-ray induced bystander signals. Proteomics 7, 4171–4180. Suter, L., Babiss, L.E., Wheeldon, E.B., 2004. Toxicogenomics in predictive toxicology in drug development. Chem. Biol. 11, 161–171. van Iersel, M.P., Pico, A.R., Kelder, T., Gao, J., Ho, I., Hanspers, K., Conklin, B.R., Evelo, C.T., 2010. The BridgeDb framework: standardized access to gene, protein and metabolite identifier mapping services. BMC Bioinform.11, 5. Wang, J., Wei, Y., Wang, D., Chan, L.L., Dai, J., 2008. Proteomic study of the effects of complex environmental stresses in the livers of goldfish (Carassius auratus) that inhabit Gaobeidian Lake in Beijing, China. Ecotoxicology 17, 213–220. Wang, M., Chan, L.L., Si, M., Hong, H., Wang, D., 2009. Proteomic analysis of hepatic tissue of zebrafish (Danio rerio) experimentally exposed to chronic microcystin-LR. Toxicol. Sci. 113, 60–69. Xie, L., Xie, P., Guo, L., Li, L., Miyabara, Y., Park, H.D., 2005. Organ distribution and bioaccumulation of microcystins in freshwater fish at different trophic levels from the eutrophic Lake Chaohu, China. Environ. Toxicol. 20, 293–300. Zegura, B., Lah, T.T., Filipic, M., 2004. The role of reactive oxygen species in microcystin-LR-induced DNA damage. Toxicology 200, 59–68.