Hemoglobin System of Sparus Aurata: Changes in Fishes Farmed Under Extreme Conditions

SCIENCE OF THE TOTAL ENVIRONMENT 403 (2008) 148– 153

available at www.sciencedirect.com

www.elsevier.com/locate/scitotenv

Hemoglobin system of Sparus aurata: Changes in fishes farmed under extreme conditions

Salvatore Campoa, Giancarlo Nastasia, Angela D'Ascolaa, Giuseppe M. Campoa, Angela Avenosoa, Paola Trainaa, Alberto Calatronia, Emanuele Burrascanob, Alida Ferlazzob, Giulio Lupidic, Rosita Gabbianellic, Giancarlo Falcionic,⁎

aDepartment of Biochemical, Physiological and Nutritional Sciences, School of Medicine, University of Messina, Policlinico Universitario, Torre Biologica, 5° piano, Via C. Valeria, 98125 Messina, Italy bDepartment of Morphology, Biochemistry, Physiology and Animal Production, School of Veterinary Medicine, University of Messina, Polo Universitario Annunziata, 98168 Messina, Italy cDepartment of MCA Biology, University of Camerino, Via Gentile III da Varano, 62032 Camerino MC, Italy

ARTICLE INFO ABSTRACT

Article history: In order to gain more knowledge on the stress responses of gilhead seabream (Sparus aurata) Received 29 February 2008 under extreme conditions, this study investigated the functional properties of the Received in revised form 19 May 2008 hemoglobin system and globin gene expression under hypoxia and low salinity. The Accepted 20 May 2008 oxygen affinity for the two hemoglobin components present inside the S. aurata erythrocyte Available online 27 June 2008 was practically identical as was the influence of protons and organic phosphates (Root effect). The quantification of S. aurata hemoglobin fractions performed by HPLC and the data Keywords: on gene expression of globin chains assayed by PCR indicate that under hypoxia and low Hemoglobins salinity there is a change in the ratio between the two different hemoglobin components. Root effect The result indicating that the distinct hemoglobins present in S. aurata erythrocyte have Globins almost identical functional properties, does not explain the adaptive response (expression Gene expression change) following exposure of the animal to hypoxia or low salinity on the basis of their Salinity function as oxygen transporter. We hypothesize that other parallel biological functions that Hypoxia the hemoglobin molecule is known to display within the erythrocyte are involved in adaptive molecular mechanisms. The autoxidation–reduction cycle of hemoglobin could be involved in the response to particular living conditions. © 2008 Elsevier B.V. All rights reserved.

1. Introduction influenced by different effectors such as chlorides, protons

(Bohr effect), CO2 and organic phosphates. All these effectors The functional properties of hemoglobin are characterized by bind preferentially to the T state of hemoglobin lowering the

the presence of both homotropic and heterotropic interaction overall O2 affinity of the molecule. phenomena which have been studied over the years. Many fish species have multiple Hb components and this Cooperativity between subunits is achieved through the Hb multiplicity is related to the greater variability in oxygen conformational transition between the deoxy-low-affinity regimes to which animals are subjected to. The presence of state (or T state) and the oxy-high-affinity state (or R state), different hemoglobin components correlates also with the role

which accounts for the sigmoidal shape of the oxygen binding of O2 release against high hydrostatic pressure inside the curve. In addition, the oxygen affinity of hemoglobin is swim bladder. Hemoglobins involved in this function display a

⁎ Corresponding author. Tel.: +39 0737 403211; fax: +39 0737 403290. E-mail address: [email protected] (G. Falcioni).

0048-9697/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.scitotenv.2008.05.027 SCIENCE OF THE TOTAL ENVIRONMENT 403 (2008) 148– 153 149 very marked Bohr effect (called Root effect). The oxygen Three adult globin genes have been found in S. aurata; two binding properties of Hbs that exhibit Root effect are different α-globin genes and one β-globin gene have been characterized by a very strong dependence of both affinity identified and their nucleotide and aminoacid sequences have and cooperativity on pH and organic phosphates. The already been submitted to the GeneBank database (accession essential feature of this phenomenon appears to be a proton numbers DQ520839; DQ520840 and DQ452379 respectively). induced stabilization of the low affinity state of the molecule The oxygen binding properties of the two purified hemoglobin (Brunori, 1975). At the structural level, the Root effect has been components as a function of pH and organic phosphates were correlated (Perutz and Brunori, 1982) with two additional H- also investigated. bonds involving the protonated form of His HC3 (146β) and a serine in position F9 (93β) which, in Root effect hemoglobins, substitutes the cysteinyl residue normally found in mammals. 2. Materials and methods Comparative studies on the structural and functional properties of hemoglobins permit to speculate on the adaptive 2.1. Sample collection molecular mechanisms developed to satisfy particular needs of oxygen demand due to different environmental conditions Six specimens of S. aurata obtained from an Ippocampus fish- to which animals may be subjected to. Environmental factors farmer (Villafranca Tirrena, ME, Italy) were stabulated into a can include variations in temperature, pH, salinity, oxygen 300 L aquarium under standard conditions, at the temperature tension, etc. of 20 °C, and pH 8.0. The concentration of oxygen was kept In intensive fish culture systems, a reduced availability of at 6.5–7.0 mg/L (Titolation kit — Hanna, USA), by a pump dissolved oxygen in water is often observed. This is ascribed to oxygenator, whereas the nitrates/nitrites were monitored and a high fish density and to the feeding practices; algal blooms maintained below 0.05 mg/L (Ascol test, Read Sea, and and elevated temperatures can contribute as well. Nitrification Active Bacteria kit — Aquaristica, Italy). After This lack of oxygen can induce responses and the typical 10 days stabulation, blood was collected from the caudal vein metabolic adjustments caused by the hypoxic stress are to of each specimen, using MS-222 40 mg/L as anesthetic, for no maintain oxygen supplies in the critical organs and to reduce longer than 4 min, and 50 mM Na2EDTA as anticoagulant. consumption of oxygen. Blood samples were immediately processed. In order to cope with periods of hypoxia, many fish species Subsequently, the fishes were stabulated for an additional have evolved new molecular responses (Nikinmaa, 2002; 20 days during which oxygen concentration was progressively Nilsson and Renshaw, 2004; Cossins and Crawford, 2005; reduced to 2.5 mg/L towards the end. A new blood sample was Nikinmaa and Rees, 2005). Some adaptive mechanisms can then collected under the same conditions. change fish gene expression with the aim of saving oxygen Finally, the fishes were stabulated for further 20 days at (Gracey et al., 2001; Ton et al., 2003; Van der Meer et al., 2005). normal oxygen concentration, while progressively reducing the The patterns of hypoxic differential gene expression are not salt concentration from normal 38 ppm to 28 ppm. The third well known. Understanding the tissue-specific and temporal blood sample was then collected under the same conditions. changes in gene expression in fishes exposed to hypoxia could reveal new mechanisms of tolerance to the lack of oxygen and 2.2. Separation of homogeneous hemoglobin components they could shed light on the evolution of this adaptive by preparative column chromatography response in vertebrates. Hemoglobin and myoglobin are respiratory proteins that link and store oxygen. Under hypoxic To obtain purified Hb components, a total of about 4 mL of conditions gene expression of globins could be modified. blood was collected from different animals. Blood from sea Salinity can also contribute to increase environmental bream S. aurata was withdrawn from the caudal vein into an stress. Most fish live in an osmotic disequilibrium between the isotonic medium constituted by 0.1 M phosphate buffer, 0.1 M environment and their blood plasma and body tissues, there- NaCl, 0.2% citrate, 1 mM EDTA, pH 7.8. The red blood cells were fore they must be able to maintain an internal osmotic separated from the plasma by centrifugation for 10 min at environment across a wide range of salinity. Alterations in 1000 g and were washed four times with the same isotonic the environmental amount of salt may affect electrolyte levels buffer. Hemolysis was accomplished by adding 3 volumes of − (i.e. Na+ and Cl ) and, as a consequence, the acid–base balance distilled water to the red cells; after 30 min the solution was (Junsen et al., 1998). centrifuged for 20 min at 10,000 g and the supernatant was There is enormous variety in the types of environmental dialysed against 0.1 M Tris–HCl buffer at pH 9.1 containing − stressors; it is generally agreed that stress in fish evokes an EDTA 5 ×10 4 M. The solution of hemoglobin previously ordered sequence of well-defined physiological changes. dialysed was applied to a DEAE (diethylaminoethylcellulose) The Sparidae Family occupies an important position Sephadex A-50 column (dimensions 40×2 cm) equilibrated among the Teleostei of the Mediterranean basin. In fact, with the same buffer. A pH gradient elution was carried out some fishes of this Family have a special role in the using two containers, one with 250 mL of 0.1 M Tris–HCl − gastronomic, industrial and economic fields. Amongst the buffer +EDTA 5×10 4 M at pH 9.1 under stirring and the other different species, Sparus aurata is the most capable of adapting with 1 L of 0.1 M KH2PO4, the former being connected directly to different environmental conditions, like those of fish farms. to the column. The flow rate was maintained at about 30– The aim of this study was to investigate the stress responses 40 mL/h; 3–4 mL fractions were collected. Representative of S. aurata under hypoxia and low salinity, in particular with fractions were examined by cellulose acetate electrophoresis regards to the hemoglobin system and globin gene expression. and stained with a solution of red poinceau (Fig. 1). 150 SCIENCE OF THE TOTAL ENVIRONMENT 403 (2008) 148– 153

2.6. S. aurata globin gene expression by PCR real time

10 ng of S. aurata total RNA was retrotranscribed by using hexameric random primers and the High Capacity cDNA Archive kit (Applied Biosystems, USA), according to the manufacturer's instruction, to estimate the variation of the globin Fig. 1 – Cellulose acetate electrophoresis of Sparus aurata mRNAs under normal and extreme condition fish-farming. hemoglobin components isolated by column chromatography; Primers and probes for PCR real time were designed and a) hemoglobin preparation applied to the column, b) compo- supplied by Proligo (France). All probes, containing LNA bases nent I, c) component II. (Linking Nucleic Acid), were labeled with 5′-FAM reporter dye and 3′ Black Hole Quencher 1 (BHQ1) (Table 2). For each globin and for β-actin (as endogenous control) the reactions were carried out in 2.3. Spectrophotometric measurements triplicate and in monoplex on the mod. 7500 RT-PCR real time System (Applied Biosystems, USA) by using the TaqMan Universal Oxygen equilibrium curves were determined spectrophoto- PCR Mastermix kit (Applied Biosystems, USA) as suggested by the metrically by the method of Rossi Fanelli and Antonini (1958). manufacturer, and a standard curve prepared side by side with The fractional saturation with oxygen of hemoglobin compo- scalar dilutions of the previously obtained clone. After normal- nents in air (pO2 =155 mm Hg) as a function of pH was followed ization, the results were expressed as relative amounts vs. normal with a Cary 219 spectrophotometer in the visible region. controls (normal stabulated fish samples).

2.4. Identification and quantification of S. aurata hemoglobin fractions 3. Results

An aliquot of each collected blood sample was diluted in an Fig. 2 shows the separation of the components of S. aurata appropriate volume of distilled water (DDW) and cells lysed by hemoglobin as obtained by pH gradient elution from DEAE freezing–thawing. The solution was then centrifuged (3000 g,4°C) Sephadex A-50 column. Homogeneous components tested by to remove the membrane debris and diluted to a final hemoglobin cellulose acetate electrophoresis of S. aurata hemoglobin were concentration of 1 g/dL, spectrophotometrically (Biomate 3 — used for studies on their functional properties. Thermo Electron Corporation, USA) evaluated at 412 nm. The hemoglobin fractions were separated by HPLC with a 3.1. Oxygen binding properties gradient of increasing ionic strength, from 10 mM Bis–Tris/HCl, pH 5.9 to 270 mM Bis–Tris/HCl, pH 5.6 in 6 min. The detection was The oxygen equilibrium curve for the two hemoglobin performed spectrophotometrically at 412 nm and acquired on PC components at pH 7.0 was practically identical (Fig. 3); it is by a dedicated software (mod. Binary Pump 1525, Waters, USA), cooperative (the Hill coefficient n is close to 1.75), implying by ToSo cation exchange resin (5 μm particle diameter, 10 μm). positive interactions between the oxygen binding sites. Fig. 4 shows the effect of pH on the fractional saturation

2.5. Preparation of S. aurata globin cDNA clones for the of S. aurata Hbs in air, i.e., at constant O2 activity. This re- PCR real time standard curves presentation of the effect of pH is particularly useful to illustrate the influence of solvent composition on the Root For each S. aurata globin gene that was identified and for the β- effect. When the experiments were performed in the absence actin (GenBank accession number X89920), a couple of primers of organic phosphate, the midpoint of the transition for both were designed on the 5′ and 3′ mRNAs respectively (Table 1), proteins was pH ~5.0 (at pO2 ~155 mmHg and 25 °C). The with the aim to obtain specific clones for PCR real time addition of saturating amounts of ATP (1 mM) shifted the standard curve preparation. RT-PCRs were carried out using region where the Root effect was present towards higher pH 100 ng of previously extracted total RNA and Super Scrypt One- values: the midpoint was at pH ~6.2. Step RT-PCR Platinum Taq HiFi (Invitrogen, USA), according to the manufacturer's instruction. Each RT-PCR product was 3.2. Quantification of S. aurata hemoglobin fractions by HPLC cloned by using the TOPO TA cloning kit (Invitrogen, USA) as suggested by the manufacturer; the clones were analyzed by High performance liquid chromatography analysis of hemo- nucleotide sequencing (v.1.1 BigDye terminators ready to use lyzate from S. aurata shows two distinct hemoglobin compo- mix and the mod. 310 Genetic Analyzer — Applied Biosystems, nents (called on the basis of their relative retention times, HbI USA) and spectrophotometrically quantitatively evaluated. and HbII). HbII, under standard fish-farming conditions, is the

Table 1 – Primers used to obtain Sparus aurata globins and β-actin cDNA clones for the PCR real time standard curves

α1 globin F(5) 5′-gcagcatcttcttgatccattttc-3′ R(878) 5′-aagggtttggtgacgttgct-3′ (DQ520839) α2 globin F(36) 5′-gaagaaaagggcagtcatgag-3′ R(829) 5′-ttatggttggcgacgtcatg-3′ (DQ520840) β globin F(2) 5′-agaagggttgcgatcaacgt-3′ R(770) 5′-cacttgtagagctgcttcac-3′ (DQ452379) β actin F(234) 5′-ctgggatgacatggagaaga-3′ R(900) 5′-agacagcacagtgttggcat-3′ (X89920) SCIENCE OF THE TOTAL ENVIRONMENT 403 (2008) 148– 153 151

Table 2 – Primers and probes for PCR real time of each globin and β actin mRNAs of Sparus aurata Forward Reverse Probe

α1 globin (693) 5′-caacctgatggtagttgatcg-3′ (770) 5′-aggaacttgtcgaatgagac-3′ (720) 5′-6FAM-cttCccCgcCgaCttc-3′BHQ1 α2 globin (667) 5′-gtcattgccatgtactacc-3′ (787) 5′-tttggatcagtggagtttagc-3′ (692) 5′-6FAM-actTcactGcGgaGgtc-3′BHQ1 β globin (115) 5′-agatcgatgtgggtgaaatc-3′ (311) 5′-tgagaagtgtctttgagtcc-3′ (141) 5′-6FAM-cagGctTtgTccAggc-3′BHQ1 β actin (778) 5′-gccctcttccagccatcc-3′ (899) 5′-gacagcacagtgttggcatac-3′ (800) 5′-6FAM-tcgGtatGgagTcCtgc-3′BHQ1

The oligos were designed on the basis of the GenBank database sequences (accession numbers: DQ520839 for the α1 globin, DQ520840 for the α2 globin, DQ452379 for the β globin and X89920 for the β actin). Capital letters in the probe sequences are LNA bases.

main fraction with a mean value of 65%; while the mean value Under low-oxygen concentration, the levels of α2 globin for HbI under the same conditions was 35% (Fig. 5A). mRNA decreased (0.35 fold) while those of α1 globin increased After fish-farming at low-oxygen concentration, the levels (2.4 fold) with respect to normal conditions. Under low salt of HbII decreased to the mean value of 49%, whereas the mean concentration, the values of α2 and α1 globin mRNAs still level of HbI increased to 51% (Fig. 5B). In a similar way, low salt decreased (0.7 fold) and increased (1.7 fold) respectively. concentration fish-farming induced a reduction in the HbII Instead, the values of the beta globin mRNA remained fraction to 52% and an increase in the HbI mean amount (48%) unchanged. (Fig. 5C). A small amount of ferric hemoglobin which slightly The result in the variation of the S. aurata globin gene increased when the animals were submitted to extreme expression, under the described fishI fraction is probably α –β conditions (hypoxia) was also present (data not shown). composed of the tetramerous 12 2, while the slowest HbII α –β fraction is a 22 2 molecule. 3.3. Globin gene expression

The gene expression of globin chains assayed by PCR real time 4. Discussion was in agreement with the data obtained by HPLC analysis here reported on hemoglobin fractions (Fig. 6). Our results indicate that the two major components of S. aurata hemoglobin are, at least under our experimental conditions, functionally very similar and both display Root effect (a similar behaviour has been reported for carp, where there are three structurally distinct hemoglobins whose functional properties are almost identical to one another) (Gillen and Riggs, 1972; Tan et al., 1973). In other teleost fish (salmon, eel, loach and trout) the oxygen equilibrium of at least one hemoglobin present inside the erythrocyte is characterized by homotropic interactions and by the complete insensitivity to protons and organic phosphates and they have different oxygen affinity (Weber and Jensen, 1988). In fish inhabiting low-oxygen water areas, the hemoglobin–oxygen

Fig. 2 – pH gradient elution from DEAE Sephadex A-50 column. Linear gradient: Tris/HCl 50 mM pH 9.1+EDTA

0.5 mM and KH2PO4 50 mM.

– Fig. 4 Oxygen fractional saturation (Y) in air (pO2 =155 mmHg) of HbI and HbII as a function of pH in the presence and absence Fig. 3 – Hill plot of the oxygen binding curves of HbI (○) and HbII of 1 mM ATP. ■ HbI in Bis–Tris or MES buffer 0.05 M, ▲HbII in (□). Conditions: potassium phosphate buffer 0.1 M pH 7.0; Bis–Tris or MES, X HbI+ATP, ○ HbII+ATP. Other conditions: Bis– T=20 °C. Tris or MES 0.05 M; T=20°C. 152 SCIENCE OF THE TOTAL ENVIRONMENT 403 (2008) 148– 153

Fig. 5 – Cation exchange HPLC fractionation of Sparus aurata hemoglobins under: A) normal conditions (HbI 35% HbII 65%). B) low-oxygen concentration conditions (HbI 51% HbII 49%). C) low salt concentration conditions (HbI 48% HbII 52%).

affinity is higher than that of fish in other regions (Nikinmaa, sence of different hemoglobin components in the same 2001; Powers et al., 1979). Changes in hemoglobin patterns erythrocyte may be of importance for the stability, and thus have been reported due to hypoxia conditions (Marinsky et al., the life-span, of the cell itself and may help to prevent 1990). The result indicating that the distinct hemoglobins precipitation or crystallization inside the red cell. present in the S. aurata erythrocyte have functional properties For trout hemoglobins, we previously reported (Fedeli et al., almost identical to one another, does not explain the adaptive 2001) a peroxidase activity which is greater for HbIV response (expression change) following exposure of the (the hemoglobin component with Root effect) than for trout animal to hypoxia or low salinity on the basis of their function HbI (it binds oxygen cooperatively but is insensitive to pH and of oxygen transport. The physiological significance of other allosteric effectors). It is also necessary to point out this change is unclear. The expression change of the two that hemoglobins with Root effect are often only partially hemoglobin components does not alter the function of oxygen saturated with oxygen and thus more easily oxidizable with transport even in the presence of a modified acid–base respect to the fully oxygenated derivative. balance (eventually due to changes in external salinity) in Comparative evaluation has shown that respiratory pig- plasma. However, responses for fish living in adverse osmotic ments in fish are less resistant to oxidation than in higher conditions have to take into account demand required to vertebrates, specifically in mammals (Jensen, 2001). However, balance the inevitable loss or gain of water by osmosis (Kidder in most fish species, the met-Hb concentration in blood is in et al., 2006). the range of 3–13% and regular changes in met-Hb concentra- The level of nucleotide triphosphates in fish red cells plays tion in fish blood are present during the annual cycle (Soldatov a vital role in adaptation to hypoxia. A variety of fish acquire and Maslova, 1989). During the autoxidation of human an increase in red cell oxygen affinity when placed in hypoxic hemoglobin, the alpha chains are preferentially oxidized water (Weber et al., 1975; Weber and Lykkeboe, 1978). The level of red blood cell nucleotide triphosphates falls within a few days after fish are exposed to hypoxic water. Conversely, when fish are displaced from their hypoxic deep-water habitat to normoxic water, the level of organic phosphate in their red cells rises (Wood et al., 1975). Erythrocytes of teleost fish use adenosine triphosphate (ATP), guanosine triphosphate (GTP) or inositol pentaphosphate and probably also lactate as allosteric effectors (Gillen and Riggs, 1977; Isaacks et al., 1977). GTP has been found to lower oxygen affinity of carp hemoglobins twice as much as ATP (Weber and Lykkeboe, 1978); in trout IV hemoglobin the influence of the allosteric effectors on oxygen affinity resulted the same (Gronenborn et al., 1984), even though the only difference in the lining of the allosteric effector sites lies in the replacement of glu NA2β in carp by Asp in trout IV hemoglobin. Probably, other parallel Fig. 6 – Sparus aurata alpha and beta globin gene expression biological functions that the hemoglobin molecule is known to variations under extreme farming conditions vs. normal display within the erythrocyte (Giardina et al., 1995) are farming conditions; (□) low-oxygen concentration, (■) low salt involved in the adaptive molecular mechanisms. The pre- concentration. SCIENCE OF THE TOTAL ENVIRONMENT 403 (2008) 148– 153 153

(Mansouri and Winterhalter, 1973; Tomoda et al., 1981). A Junsen MK, Madausen SS, Kristiansen K. Osmoregulation and typical feature of fish hemoglobins is represented by their salinity effects on the expression and activity of Na+,K(+)-ATPase in the gills of European sea bass, Dicentrarchus labrax (L.). J Exp stability as tetramers and therefore, all the possible oxidation Zool 1998;282(3):290–300. intermediates (Colosimo et al., 1982), i.e., tetramers containing Kidder III GW, Petersen CW, Preston RL. Energetics of osmoregulation: ferrous and ferric hemes, could originate from the autoxida- I. Oxygen consumption by Fundulus heteroclitus.JExpZool.AComp tion of the hemes in the circulating blood. These oxidized Exp Biol 2006;305(4):309–17. intermediates have lower cooperativity and higher affinity, Mansouri A, Winterhalter KH. Nonequivalence of chains in associated with a reduction in the total oxygen binding hemoglobin oxidation. Biochemistry 1973;12:4946–9. capacity per mass of protein. The autoxidation–reduction Marinsky CA, Houston AH, Murad A. Effect of hypoxia on hemoglobin isomorph abundances in rainbow trout Salmo cycle of hemoglobin could be involved in the response to gairdneri. Can J Zool 1990;68:884–8. particular living conditions (i.e. low-oxygen availability) even Nikinmaa M. Haemoglobin function in vertebrates: evolutionary if the response to abnormal conditions has been attributed to changes in cellular regulation in hypoxia. Res Physiol modifications in the intracellular concentration of modulator 2001;128:317–29. compounds like protons and ATP. Nikinmaa M. Oxygen-dependent cellular functions — why fishes Certainly, information on the heterotropic effect of the and their aquatic environment are a prime choice of study. – different organic phosphates versus the hemoglobin compo- Comp Biochem Physiol 2002;133:1 16. Nilsson GE, Renshaw GMC. Hypoxic survival strategies in two nents, on their intracellular levels and on the fraction of met- fishes: extreme anoxia tolerance in the North European crucian Hb in the circulating blood could contribute to better under- carp and natural hypoxic preconditioning in a coral-reef shark. stand the molecular mechanisms involved in adaptive J Exp Biol 2004;207:3131–9. processes. Nikinmaa M, Rees RB. Oxygen-dependent gene expression in fishes. Am J Physiol 2005;288:R1079–90. Perutz MF, Brunori M. Stereochemistry of cooperative effect in fish REFERENCES and amphibium haemoglobins. Nature 1982;299:421–6. Powers DA, Fyhn HJ, Fyhn UEH, Martin JP, Garlick RI, Wood SC. A comparative study of the oxygen equilibria of blood from 40 Brunori M. Molecular adaptation to physiological requirements: genera of amazonian fishes. Comp Biochem Phys the hemoglobin system of trout. Curr Top Cell Regul 1979;62A:884–8. 1975;9:1–39. Rossi Fanelli A, Antonini E. Studies on the oxygen and carbon Colosimo A, Coletta M, Falcioni G, Giardinà B, Gill SJ, Brunori M. monoxide equilibria of human myoglobin. Arch Biochem Thermodynamics of oxygen binding to trout haemoglobin I Biopyhs 1958;77:478–92. and its oxidation intermediates. J Mol Biol 1982;160:531–43. Soldatov AA, Maslova MN. Concentration of methemoglobin in Cossins AR, Crawford DL. Fish as models for environmental fish blood during annual cycle. Zh Evol Biokhim Fiziol genomics. Nat Rev Genet 2005;6:324–33. 1989;25:454–9. Fedeli D, Tiano L, Gabbianelli R, Caulini GC, Wozniak M, Falcioni G. Tan AL, Noble RW, Gibson QH. Conditions restricting allosteric Hemoglobin components from trout (Salmo irideus): determination transitions in carp hemoglobin. J Biol Chem 1973;248:2880–8. of their peroxidative activity. Comp Biochem Physiol Part B Tomoda A, Yoneyama Y, Tsuji A. Changes in intermediate 2001;130:554–9. hemoglobins during autoxidation of haemoglobin. Biochem J Giardina B, Messana I, Scatena R, Castagnola M. The multiple 1981;195:485–92. functions of hemoglobin. Crit Rev Biochem Molec Biol 1995; Ton C, Stamatiou D, Liew C. Gene expression profile of zebrafish 30(3):165–96. exposed to hypoxia during development. Physiol Genomics Gillen RG, Riggs A. Structure and function of hemoglobins of the 2003;13:97–106. carp, Cyprinus carpio. J Biol Chem 1972;247:6039–46. Van der Meer DL, Van den Thillart GE, Witte F, de Bakker MA, Gillen RG, Riggs A. The enhancement of the alkaline Bohr effect of Besser J, Richardson MK, et al. Gene expression profiling of the some fish hemoglobins with adenosine triphosphate. Arch long-term adaptive response to hypoxia in the gills of adult Biochem Biophys 1977;183:678–85. zebrafish. Am J Physiol 2005;289:R1512–9. Gracey AY, Troll JV, Somero GN. Hypoxia-induced gene expression Weber RE, Lykkeboe G. Respiratory adaptations in carp blood: profiling in the euryoxic fish Gillichthys mirabilis. Proc Natl Acad influences of hypoxia, red cell organic blood phosphate – Sci U S A 2001;98:1993–8. divalent cations and CO2 on hemoglobin oxygen affinity. Gronenborn AM, Clore GM, Brunori M, Giardina B, Falcioni G, Perutz J Comp Physiol 1978;128:127–37. MF. Stereochemistry of ATP and GTP bound to fish haemoglobins. Weber RE, Jensen FB. Functional adaptations in hemoglobins from A transferred nuclear overhauser enhancement, 31P-nuclear ectothermic vertebrates. Annu Rev Physiol 1988;50:161–79. magnetic resonance, oxygen equilibrium and molecular Weber RE, Lykkeboe G, Johansen K. Biochemical aspects of the modelling study. J Mol Biol 1984;178:731–42. adaptation of hemoglobin-oxygen affinity of eels to hypoxia. Isaacks RE, Kim HD, Bartlett GR, Harkness DR. Inositol Life Sci 1975;17:1345–50.

pentaphosphate in erythrocytes of a freshwater fish, Piraracú Wood SC, Johansen K, Weber RE. Effects of ambient pO2 on Arapaima gigas. Life Sci 1977;20:987–90. hemoglobin–oxygen affinity and red cell ATP concentrations in Jensen FB. Comparative analysis of autoxidation of haemoglobin. benthic fish, Pleuronectes platessa. Respir Physiol J Exp Biol 2001;204:2029–33. 1975;25:259–67.