INTRODUCTION
Totoaba macdonaldi (Gilbert, 1890) is an endemic species from the Gulf of California protected by a permanent ban since 1975 (NOM-059-SEMARNAT-2001), and listed as Critically Endangered through the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) in 1976 as well as the International Union for Conservation of Nature (IUCN) in 2013. Totoaba macdonaldi is a long-lived fish that can reach 25 years old (Román-Rodríguez & Hammann, Reference Román-Rodríguez and Hammann1997). Their reproductive cycle occurs during low-temperature seasons (late winter to early spring). Spawning occurs in areas near the Colorado River mouth, characterized as nursery areas where juveniles remain for ~2 years (Cisneros-Mata et al., Reference Cisneros-Mata, Montemayor-López and Román-Rodríguez1995, Reference Cisneros-Mata, Botsford and Quinn1997). Totoaba then migrate through the Gulf of California to the great islands (Islas Tiburón and Ángel de la Guarda) region during their juvenile phase. Later, all adult totoaba travel to feeding grounds in the continental region of Sonora and Sinaloa, and finally go back to the high Gulf of California to spawn (Berdegué, Reference Berdegué1955; Cisneros-Mata et al., Reference Cisneros-Mata, Botsford and Quinn1997). This migratory pattern exposes totoaba to important environmental temperature variations.
The physical conditions of aquatic ecosystems, such as dissolved oxygen concentration, temperature, salinity and presence of pollutants, can affect the health of fish (Vasseur & Cossu-Leguille, Reference Vasseur and Cossu-Leguille2003). Fish species that live permanently in estuaries, or that inhabit these areas at least during part of their life cycle, are prone to physiological alterations resulting from the constant variations in environmental parameters (Freire et al., Reference Freire, Welker, Storey, Storey, Hermes-Lima, Abele, Vázquez-Medina and Zenteno-Savín2011; Matoo et al., Reference Matoo, Ivanina, Ullstad, Beniash and Sokolova2013). While T. macdonaldi was long regarded as an estuarine-dependent species (Berdegué, Reference Berdegué1955; Flanagan & Hendrickson, Reference Flanagan and Hendrickson1976), recent evidence indicates that the high Gulf of California, which totoaba use as spawning site, no longer maintains estuarine features (De-Anda-Montañez et al., Reference De-Anda-Montañez, García de León, Zenteno-Savín, Balart, Méndez-Rodríguez, Bocanegra-Castillo, Martínez-Aguilar, Campos-Dávila, Román-Rodríguez, Valenzuela-Quiñónez-Valenzuela, Rodríguez-Jaramillo, Meza-Chávez, Ramírez-Rosas, Saldaña-Hernández, Olguín-Monroy and Martínez-Delgado2013; Valenzuela-Quiñónez et al., Reference Valenzuela-Quiñonez, Garza, De-Anda-Montañez and García-de-León2014, Reference Valenzuela-Quiñonez, Arreguín-Sánchez, Salas-Márquez, García-de-León, Garza, Román-Rodríguez and de-Anda-Montañez2015). Although the oceanographic characteristics of the high Gulf of California have been described recently to be antiestuarine, totoaba continue to spawn in the same area (De-Anda-Montañez et al., Reference De-Anda-Montañez, García de León, Zenteno-Savín, Balart, Méndez-Rodríguez, Bocanegra-Castillo, Martínez-Aguilar, Campos-Dávila, Román-Rodríguez, Valenzuela-Quiñónez-Valenzuela, Rodríguez-Jaramillo, Meza-Chávez, Ramírez-Rosas, Saldaña-Hernández, Olguín-Monroy and Martínez-Delgado2013; Valenzuela-Quiñónez et al., Reference Valenzuela-Quiñonez, Garza, De-Anda-Montañez and García-de-León2014, Reference Valenzuela-Quiñonez, Arreguín-Sánchez, Salas-Márquez, García-de-León, Garza, Román-Rodríguez and de-Anda-Montañez2015).
Water temperature changes can affect metabolic rates and, potentially, the production of reactive oxygen species (ROS) in fish as well as in other aquatic animals (Parihar et al., Reference Parihar, Dubey, Javeri and Prakash1996; Heise et al., Reference Heise, Puntarulo, Nikinmaa, Abele and Pörtner2006; Lushchak & Bagnyukova, Reference Lushchak and Bagnyukova2006; Bagnyukova et al., Reference Bagnyukova, Luzhna, Pogribny and Lushchak2007; Lushchak, Reference Lushchak2011). An imbalance between ROS production and antioxidant defences could cause oxidative stress, with consequent oxidative damage to macromolecules such as lipids, proteins, free amino acids, DNA and carbohydrates, which can result in pathological conditions (Sies, Reference Sies1997; Toyokuni, Reference Toyokuni1999).
Pörtner (Reference Pörtner2002) suggests that marine fishes are key bioindicators of climate change due to their lack of thermal regulation mechanisms. In estuarine fish species, such as Diplodus vulgaris (Geoffroy Saint-Hilaire, 1817), Diplodus sargus (Linnaeus, 1758), Dicentrarchus labrax (Linnaeus, 1758), Gobius niger (Linnaeus, 1758) and Liza ramada (Risso, 1827), oxidative stress indicators, including markers of oxidative damage and antioxidant defences, are significantly altered by thermal stress (Madeira et al., Reference Madeira, Narciso, Cabral, Vinagre and Diniz2013). High levels of lipid peroxidation and catalase (CAT) activity occur when fish (D. labrax) are exposed to temperatures outside their optimal ranges (Vinagre et al., Reference Vinagre, Madeira, Narciso, Cabral and Diniz2012). However, after the peak of elevated antioxidant enzyme activity, protein denaturation and degradation can inactivate enzymes and damage their catalytic function (Kregel, Reference Kregel2002; Abele & Puntarulo, Reference Abele and Puntarulo2004).
Fish in general have high superoxide dismutase (SOD) activity, apparently as a consequence of a protective mechanism against temperature changes (Filho et al., Reference Filho, Giulivi and Boveris1993). This has been reported for both marine and freshwater fish, including species such as Geophagus brasiliensis (Filho et al., Reference Filho, Giulivi and Boveris1993), and dealing with abrupt changes in environmental temperature has been suggested to be the biggest challenge for and energetic cost in ectothermic aquatic animals (Filho et al., Reference Filho, Giulivi and Boveris1993). Sea temperature also plays a determinant role in fish reproduction because, due to its influence on the nervous and endocrine systems that regulate the hypothalamus-hypophysis-gonadal axis (Tyler & Sumpter, Reference Tyler and Sumpter1996), temperature is one of the main factors in the synchronization of gonadal development (Zanuy et al., Reference Zanuy, Prat, Carrillo and Bromage1995; Jalabert, Reference Jalabert2005).
Oxidative stress, and the concomitant oxidative damage to macromolecules, is also considered a physiological consequence of reproduction (Harshman & Zera, Reference Harshman and Zera2007; Costantini et al., Reference Costantini, Dell'Ariccia and Lipp2008; Monaghan et al., Reference Monaghan, Metcalfe and Torres2009; Garratt et al., Reference Garratt, Vasilaki, Stockley, McArdle, Jackson and Hurst2010; Metcalfe & Monaghan, Reference Metcalfe and Monaghan2013). Several studies on vertebrates suggest differences in antioxidants between individuals under reproductive conditions and non-reproductive individuals (Dammann & Burda, Reference Dammann and Burda2006; Garratt et al., Reference Garratt, Vasilaki, Stockley, McArdle, Jackson and Hurst2010; Speakman & Garratt, Reference Speakman and Garratt2010; Schmidt et al., Reference Schmidt, Blount and Bennett2014).
Liver is a multifunctional tissue that has been reported to have an elevated metabolic rate reflecting the physiological processes (including reproduction) in which it participates; it is an energy reservoir (Love, Reference Love1970; Lyons & Dunne, Reference Lyons and Dunne2003) and lipid synthesizer (Mommsen, Reference Mommsen and Evans1998), as well as a detoxifier of potentially harmful elements (Wolf & Wolfe, Reference Wolf and Wolfe2005). Liver participation in reproduction may be due to its function as energy (lipids, proteins, glycogen) storage (Encina & Granado-Lorencio, Reference Encina and Granado-Lorencio1997). These energetic molecules are channelled to the ovaries, where they participate in the synthesis of the yolk that will sustain embryogenesis (Patiño & Sullivan, Reference Patiño and Sullivan2002; Jalabert, Reference Jalabert2005). Therefore, the hypothesis of this study was that oxidative stress indicators in liver samples from T. macdonaldi change with season (SST) and reproductive stages, with differences between sexes.
A specific search for publications on totoaba in Pubmed, EBSCO, Springer Link and ScienceDirect databases was performed on 13 June 2016 using ‘Totoaba macdonaldi, antioxidant enzyme, superficial sea temperature’ as keywords. The displayed results included only one publication on totoaba, which reported the effect of specific diets on intermediate metabolism and antioxidant status in aquaculture totoaba juveniles (Bañuelos-Vargas et al., Reference Bañuelos-Vargas, López, Pérez-Jiménez and Peres2014). However, there are no publications on oxidative stress biomarkers on free-living totoaba. The objective of this study was to evaluate the relationship between season and oxidative stress indicators in liver samples from T. macdonaldi as well as possible variations between sexes and reproductive stages of this species. This study is, to our knowledge, the first effort to describe oxidative stress in wild totoaba.
MATERIALS AND METHODS
Fieldwork
Specimens of T. macdonaldi were captured along the Gulf of California under scientific fishing permits (SGPA/DGVS/02913/10, SGPA/DGVS/05508/11 and SGPA/DGVS/00039/13) (Figure 1). Outboard motor boats were used. Gill net with mesh size of 10″ was set and monitored between 1 and 2 h. Fishing rods with hooks number 5 and 6 were also used, with shrimp, squid and/or grouper heads as bait. Sampling took place from April, May and November 2010, February, March, October, November and December 2011, April 2012, January to February 2013. Samples were grouped by season (spring, autumn and winter), which are characterized mainly by temperature changes. Total length and weight and geographic position of each captured totoaba specimen were recorded. A Studio24 TM® weight hook (±10 g) and a Renegade conventional measuring tape (±1.0 mm) were used. Liver samples were placed in cryovials (Simport®), frozen immediately by immersion in liquid nitrogen and transported to the Oxidative Stress laboratory at CIBNOR, La Paz, Baja California Sur, Mexico, where they were preserved in a Thermo Scientific® ultrafreezer at −80°C until analysed, within 6 months of collection. Sex was identified by histological analysis of gonads; immature females (FI) were differentiated from mature females (FM), and immature males (MI) from mature males (MM). Classification was based on the development of gametes and characteristics of reproductive structures such as follicles, germinal epithelium and interfollicular connective tissue (Tyler & Sumtper, Reference Tyler and Sumpter1996). Fish whose gonads did not have characteristics allowing the differentiation of ovaries or testicles were classified as undifferentiated (UN). Average daily sea surface temperatures (SST) at fishing sites were obtained from the AQUA-MODIS sensor (http://coastwatch.pfel.noaa.gov/erddap).
Fig. 1. Study area. Totoaba macdonaldi fishing zone in the Gulf of California, indicating number of fish by season. (■) spring (
) autumn and (•) winter.
Oxidative stress indicators/sample preparation
Each liver sample was divided in four segments of ~100 mg. Frozen tissue was homogenized (Polytron PT 1300D, Kinematica, Switzerland) with PMSF, 1 mM (phenylmethylsulfonyl fluoride), in an ice-cold phosphate buffer solution (50 mM, ph 7.5). Homogenates were centrifuged (Sorvall Legend RT Germany) at 2124 × g for 15 min at 4°C. Supernatants were recovered and the following techniques, proven and validated for other marine species including mammals, reptiles, crustaceans, elasmobranchs and corals (Zenteno-Savín et al., Reference Zenteno-Savín, Clayton-Hernández and Elsner2002, Reference Zenteno-Savín, Saldierna-Martínez and Ahuejote-Sandoval2006, Reference Zenteno-Savín, St Leger and Ponganis2010; Liñán-Cabello et al., Reference Liñán-Cabello, Flores-Ramírez, Zenteno-Savin, Olguín-Monroy, Sosa-Avalos, Patiño-Barragan and Olivos-Ortiz2010; López-Cruz et al., Reference López-Cruz, Zenteno-Savín and Galván-Magaña2010; Zenteno-Savín et al., Reference Zenteno-Savín, St Leger and Ponganis2010; Labrada-Martagon et al., Reference Labrada-Martagon, Rodríguez, Méndez-Rodríguez and Zenteno-Savin2011), were used to evaluate oxidative stress indicators. All analyses were run in triplicate.
Antioxidant enzyme activity
A spectrophotometer (Beckman DU 800, Felton, CA, USA) was used to calculate antioxidant enzyme activity using colorimetric techniques. Catalase activity (CAT) was calculated from the decrease in hydrogen peroxide (H2O2) concentration at 240 nm (Hermes-Lima & Storey, Reference Hermes-Lima and Storey1993). CAT activity is expressed as U mg−1 of protein. One unit of CAT activity is defined as the quantity of enzyme necessary to consume one H2O2 μmol per minute. Superoxide dismutase (SOD) activity was measured by following the inhibition of the reduction of nitrotetrazolium blue chloride (NBT) by the generation of O2•− using the xanthine/xanthine oxidase system (Suzuki, Reference Suzuki, Taniguchi and Gutteridge2000). SOD activity is expressed in U mg−1 protein. One unit of SOD activity is defined as the amount of enzyme needed to inhibit maximum reaction by 50%. Glutathione peroxidase (GPx) activity was calculated by monitoring the continuous decrease in NADPH concentration using H2O2 as a substrate (Folhé & Günzler, Reference Folhé, Günzler and Packer1984). GPx activity is expressed in U mg−1 protein. One GPx unit is defined as the quantity of enzyme that oxidizes 1 µmol NADPH per minute. Glutathione S-transferase (GST) activity was monitored through the synthesis of thioether glutathione dinitrobenzene as a product of the reaction between glutathione (GSH) and 1-chloro-2,4-dinitrobenzene (CDNB) (Habig & Jakoby, Reference Habig and Jakoby1981). GST activity is expressed in U mg−1 protein. One unit of GST activity is defined as the quantity of enzyme that catalyses the production of 1 µmol CDNB per minute. Glutathione reductase (GR) activity was calculated by monitoring the decrease in absorbance during NADPH oxidation (Goldberg & Spooner, Reference Goldberg, Spooner and Bergmeyer1987). GR activity is expressed in U mg−1 of protein; a unit of GR activity is defined as the quantity of enzyme necessary to reduce 1 µmol oxidized glutathione (GSSG) per minute.
Superoxide radical production rate (O2•−)
Endogenous O2•− production was calculated by following the ferrocytochrome c reduction (Drossos et al., Reference Drossos, Lazou, Panagopoulos and Westaby1995). Liver samples were incubated for 15 min at 37°C in a Krebs buffer solution (NaCl, 110 M; KCl, 4.7 mM; MgSO4, 12 mM; NaH2PO4, 12 mM; NaHCO4, 25 mM; glucose, 1 mM) with cytochrome c (15 µM) in a shaking water bath (Precision USA). The reaction was stopped with N-ethylmaleimide (3 mM) and centrifuged at 2124 × g for 10 min at 4°C. Supernatants were recovered and measured in a spectrophotometer at 550 nm. O2•− production was calculated using the extinction coefficient for the change from ferricytochrome to ferrocytochrome (21 nM cm−1). O2•− radical production was expressed in nmol min−1 mg−1 of protein.
Lipid peroxidation
Lipid peroxidation levels were quantified as indicator of oxidative damage to lipids as the amount of thiobarbituric acid reactive substances (TBARS) present in a sample (Buege & Aust, Reference Buege, Aust and Packer1978). Homogenized liver samples were incubated in a water bath for 15 min at 37°C with continuous agitation. After that, samples were moved into a cold bath and 20% trichloroacetic acid (TCA) and thiobarbituric acid (TBA) were added to stop the reaction. The tissue was incubated again in a water bath for 10 min at 90°C with continuous agitation, followed by a cold bath. Samples were later centrifuged at 2124 × g for 10 min at 4°C. Supernatants were recovered to be measured at a 560 nm. TBARS concentration was calculated from a standard curve of malondialdehyde (MDA), which was run in parallel to samples. TBARS levels are expressed in nmol of TBARS mg−1 protein.
Total proteins
In order to standardize values of oxidative stress indicators, soluble protein concentration was measured using the method proposed by Bradford (Reference Bradford1976) modified for microplate reader (BioRad), using bovine serum albumin (BSA) as the standard, and measuring the absorbance at 590 nm. Results are expressed in mg of protein mL−1.
Statistical analysis
Non-parametric tests for oxidative stress indicators data were run after evaluation of normality and homoscedasticity assumptions (Shapiro-Wilk and Bartlett tests) (Zar, Reference Zar1999). Differences between groups (IF, MF, IM and MM) and seasons (spring, autumn and winter) were estimated using Kruskal–Wallis tests with multiple post-hoc comparisons (Zar, Reference Zar1999). A one-way analysis of variance (ANOVA) was used to analyse SST data. The established significance level for all statistical tests was 0.05. In order to identify the variables with greatest contribution to the variability of oxidative damage in liver of totoaba, five generalized linear models (GLM) focusing on each O2•− production and TBARS levels were created. Each of these models was developed as a function of the activity of each specific enzyme. Different distributions were tested, but the distribution which yielded the best fit was the gamma distribution, because it was not significantly different from the Chi square test (P > 0.05) from the dependent variables (O2•− production and TBARS). Link function used was log; the variables included in the models are season, sex/reproductive maturity status and activity of each of the analysed enzymes. All of these variables are detailed in Table 1. The models were validated by visual analysis of the deviance residual and observed and predicted values. Akaike's information criterion (AIC) was used (Akaike, Reference Akaike, Petrov and Csaki1973) to identify the model with best fit for each biomarker. Statistical analyses were performed using Statistica v8 (StatSoft Inc., Tulsa, OK, USA), Microsoft Excel 2010 and R Project version 3.1.1.
Table 1. Description of possible factors affecting the biomarkers of oxidative stress in the liver of totoaba, Totoaba macdonaldi, captured in the Gulf of California, Mexico.
RESULTS
A total of 173 totoaba were analysed (64 FI, 25 FM, 51 MI, 22 MM and 11 UN). Body mass of totoaba ranged between 1 and 57 kg (11.9 ± 8.75 kg, average ± SD), and size ranged between 0.52 and 1.86 m (1.06 ± 0.25 m), the UN individuals being the smallest ones (7.44 ± 3.93 kg and 0.93 ± 0.18 m) (P < 0.05 compared with the other groups). Totoaba specimens were captured during three seasons with SSTs of 20.2 ± 0.8°C in spring, 23.7 ± 3.1°C in autumn, and 15.1 ± 2.1°C in winter. Totoaba captured in winter were significantly larger (19.2 ± 7.78 kg; 1.26 ± 0.16 m) than those captured in spring and autumn (P < 0.05).
SSTs were significantly different between the seasons (P < 0.05) (Figure 2). Results of oxidative stress indicators by season and for each sex/reproductive maturity stage are shown in Table 2. There were few significant differences in antioxidant enzyme activities; results with significant differences are indicated.
Fig. 2. Sea surface temperature (SST) by season (spring, autumn and winter) in the Gulf of California during the years 2010, 2011 and 2013. Different letters indicate significant differences (P < 0.05).
Table 2. Oxidative stress indicators in the liver of Totoaba macdonaldi by sex/maturity stage groups and season (spring, autumn and winter).
Totoaba groups: FI, immature female; FM, mature female; MI, immature male; MM, mature male; and UN; undifferentiated. Oxidative stress indicators: Superoxide radical production (O2•−), lipid peroxidation (TBARS) levels, and activity of antioxidant enzymes glutathione reductase (GR), catalase (CAT), glutathione S-transferase (GST), glutathione peroxidase (GPx) and superoxide dismutase (SOD).
Antioxidant enzyme activity (U mg−1 protein), superoxide radical production (O2•−, nmol min−1 mg−1 protein), lipid peroxidation levels (TBARS nmol mg−1 protein). Different letters or symbols indicate significant differences (P < 0.05); uppercase letters indicate differences within a group between seasons; lowercase letters denote differences within a season between maturity stages for each sex; symbols denote differences between all totoabas captured in each season.
Values are medians and means ± standard error (SD).
FI had significantly higher TBARS levels in spring and winter than in autumn (P = 0.002), and had significantly higher SOD activity than in autumn and winter (P = 0.02). FM showed significantly higher levels of TBARS in spring compared with winter (P = 0.04). MI had higher production of O2•− in winter and spring than in autumn (P = 0.0003); also TBARS levels were significantly higher in winter (P = 0.0002) and spring compared to autumn (P = 0.04). MM were captured only during spring and winter and no significant differences in any of the quantified indicators of oxidative stress were observed. UN were caught only in autumn; therefore, no comparisons between seasons were possible. In general, totoaba captured in winter had higher O2•− production than those captured in spring (P = 0.04) and autumn (P = 0.006). TBARS levels were significantly lower in autumn than in spring (P = 0.008) or winter (P = 0.02). CAT activity was higher in spring than in winter (P = 0.009).
In spring, FI had higher SOD activity than FM (P = 0.003). There were no significant differences by sex/reproductive maturity stage among males in spring. In autumn, only immature totoaba (FI, MI, UN) were captured. TBARS levels and O2•− production were higher in UN than FI and MI (no significant differences; P > 0.05). In winter, FI had higher TBARS levels than FM (P = 0.02). MI had higher TBARS levels (P = 0.004) and SOD activity (P = 0.0001) than MM.
Table 3 shows the parameters obtained for the GLMs of O2•− production rate and TBARS levels. For both indicators, the best-fit model according to AIC included SOD activity. In general, the seasons were significant in 90% of the applied models. In addition to this, all models were validated using analysis of residuals. Figure 3 shows the analysis of residuals for the best-fit model for each biomarker, which suggests that the variance of the residuals is homogeneous over the groups (independent variables) in the models, and then the models fit the data reasonably well. O2•− production and TBARS levels tended to decrease when SST increased (Figure 4). This supports the high significance of the seasons in the applied models, particularly autumn.
Fig. 3. Results of the residuals analysis for each biomarker, (A–C) superoxide radical production rate and (D–F) lipid peroxidation levels. The best-fit model (shown here) was that which included superoxide dismutase (SOD) activity.
Fig. 4. Variation in the oxidative stress biomarkers in liver samples from Totoaba macdonaldi and their relationship with superficial sea surface temperature fluctuations (SST). (A) Superoxide radical production rate (O2•−, nmol min−1 mg−1 protein); (B) Lipid peroxidation levels (TBARS, nmol mg−1 protein). The shaded area indicates the range of variability of the data.
Table 3. Statistics of goodness of fit, parameter estimates and standard error (SE) of the different generalized linear models to describe the oxidative stress biomarkers in liver samples from Totoaba macdonaldi.
n, number of data; RD, residual deviance; DF, degrees of freedom; AIC, Akaike's information criterion. Totoaba groups: FI, immature female; FM, mature female; MI, immature male; MM, mature male and UN, undifferentiated.
Values marked with * were significantly different (P = 0.05) from the intercept.
DISCUSSION
Superoxide radical production, antioxidant enzyme activities and oxidative damage were quantified in the liver of undifferentiated, immature and mature (both male and female) totoaba. Few significant changes (five of 28 possibilities, 17%) were observed by season for each group of totoaba, which suggests that, with particular exceptions, the oxidative stress indicators in liver of totoaba are generally stable despite seasonal changes in SST. TBARS levels were higher in both FI and MI collected in spring and winter (low and intermediate temperatures) than in autumn. Similar results were reported for the flatfish Paralichthys orbingnyanus (Valenciennes, 1839), in which lipid peroxidation levels were significantly higher in winter and spring than in summer and autumn (Amado et al., Reference Amado, Berteaux, Geracitano, Monserrat and Bianchini2006b). Higher lipid peroxidation levels at low temperatures have been attributed to the fact that ectotherms tend to elevate polyunsaturated fatty acid (PUFA) concentration in cell membranes under such conditions (Chapelle et al., Reference Chapelle, Meister, Brichon and Zwingelstein1997). Higher lipid peroxidation levels have been associated with higher PUFA content in P. orbignyanus liver (Amado et al., Reference Amado, Berteaux, Geracitano, Monserrat and Bianchini2006b). High lipid peroxidation was observed during summer and autumn in the mullet Mugil cephalus (Linnaeus, 1758) in a marine environment in Portugal (Ferreira et al., Reference Ferreira, Moradas-Ferreira and Reis-Henriques2005). Mullet and totoaba have similar biological characteristics, such as their relationship to the benthic environment.
No significant differences in antioxidant enzyme activity were found between seasons, except for higher SOD activity in FI from spring compared with winter. Similar results were reported for the flatfish P. orbingnyanus, species in which the activity of antioxidant enzymes CAT and GST was constant during the four seasons, particularly in unpolluted areas (Amado et al., Reference Amado, Berteaux, Geracitano, Monserrat and Bianchini2006b). It is possible that T. macdonaldi is a thermotolerant species (Pörtner, Reference Pörtner2010), since the antioxidant defences appear to remain unchanged throughout the seasons. In certain fish and mollusc species that live in habitats with extreme conditions antioxidant enzyme activities appear to be quite stable (Abele & Puntarulo, Reference Abele and Puntarulo2004), particularly in polar species, for which antioxidant enzymatic activity is usually high to compensate for the vulnerability at low temperatures (Abele et al., Reference Abele, Heise, Pörtner and Puntarulo2002); for example, Cassini et al. (Reference Cassini, Favero and Albergoni1993) observed that SOD activity in liver, heart and muscle was not significantly different between polar and Mediterranean fishes (Chionodraco hamatus, Lönnberg, 1905, and Pagothenia bernacchii, Boulenger, 1902). In contrast, Micropogonias furnieri (Desmarest, 1823), a species that is taxonomically close to T. macdonaldi (Sciaenidae), has higher CAT and GST activity in summer than in winter (Amado et al., Reference Amado, da Rosa, Meirelles, Moraes, Vaz Pires, Leães, Martinez, Berteaux, Maia, Monserrat, Bianchini, Martinez and Garacitano2006a). It is possible that the differences in antioxidant enzyme activities observed are associated with the physiological state of the totoaba, such as reproduction, feeding and migration rather than seasonal changes in SST.
Totoaba reproduce during winter and spring (low SST seasons) (Cisneros-Mata et al., Reference Cisneros-Mata, Montemayor-López and Román-Rodríguez1995; Valenzuela-Quiñonez et al., Reference Valenzuela-Quiñonez, García-de-León, De Anda Montañez and Balart2011; De Anda-Montañez et al., Reference De-Anda-Montañez, García de León, Zenteno-Savín, Balart, Méndez-Rodríguez, Bocanegra-Castillo, Martínez-Aguilar, Campos-Dávila, Román-Rodríguez, Valenzuela-Quiñónez-Valenzuela, Rodríguez-Jaramillo, Meza-Chávez, Ramírez-Rosas, Saldaña-Hernández, Olguín-Monroy and Martínez-Delgado2013). Liver acts as an energy reservoir, as it is rich in lipids, and synthesizes vitellogenin (van Bohemen et al., Reference van Bohemen, Lambert and Peute1981; Allen & Wootton, Reference Allen and Wootton1982). Lipids in the liver are moved towards the gonads during the reproductive season (Lowerre-Barbieri et al., Reference Lowerre-Barbieri, Ganias, Saborido-Rey, Murua and Hunter2011). Sexual maturity may affect lipid peroxidation levels, because the increase in oestrogen concentration can modify the degree of fatty acid unsaturation, especially in organs such as liver (Tocher, Reference Tocher2003). Both lipid mobilization and increased PUFA concentration in liver during the reproductive season (Aten et al., Reference Aten, Duarte and Behrman1992; Wu, Reference Wu1992; Oakes & Van Der Kraak, Reference Oakes and Van Der Kraak2003) could contribute to the higher TBARS levels observed in totoaba reproduction that occur during winter and spring. However, increased TBARS levels were not higher in the mature (MM, FM) but in the immature (FI, MI) totoaba. PUFAs are more susceptible to the action of pro-oxidant agents such as ROS (De Zwart et al., Reference De Zwart, Meerman, Commandeur and Vermeulen1999). The UN group and, particularly, the MI totoaba in winter, had the highest production of O2•−, which may be affecting TBARS levels. This may be due to the migration of young totoabas from the breeding and spawning grounds in the northernmost region of the Gulf of California towards the large islands area (Berdegué, Reference Berdegué1955; Cisneros-Mata et al., Reference Cisneros-Mata, Botsford and Quinn1997); this migration can be greater than 200 miles. Fish that migrate could change their physiological status in response to exercise, changes in salinity, temperature and diet, among others (Miller et al., Reference Miller, Schulze, Ginther, Li, Patterson, Farrell and Hinch2009). Salmon Oncorhynchus nerka (Walbaum, 1792) migratory movements are associated with changes in metabolic enzyme activities (cathepsin D, carboxypeptidase A and glucose-6-phosphate dehydrogenase), as well as in protein and free amino acid content (Mommsen et al., Reference Mommsen, French and Hochachka1980). FI and MI are characterized by being morphologically (length and weight) smaller and, therefore, are assumed to be younger than FM and MM. Young fish are reported to be more active and to have higher metabolic rates than adults under basal conditions (Ware, Reference Ware1975). In addition, larger fish tend to show lower swimming speeds than smaller ones (Brett, Reference Brett1965). Exercise can generate conditions of oxidative stress by increasing oxygen consumption and energy expenditure (Konigsberg, Reference Konigsberg2008).
Fish gender can affect physiological markers, particularly in marine species (Chung et al., Reference Chung, Lee and Lee2013; Tkachenko et al., Reference Tkachenko, Kurhaluk, Grudniewska and Andriichuk2014). Differences in oxidative stress indicators have been reported between males and females (Rudneva, Reference Rudneva1995). For example, when exposed directly to xenobiotics, males have higher lipid peroxidation and SOD activity than females (Vega-López et al., Reference Vega-López, Galar-Martínez, Jimenez-Orozco, Garcia-Latorre and Dominguez-López2007). Under hypoxic or hyperoxic conditions, males are prone to change their reproductive behaviour because male pheromones are vulnerable to lipid peroxidation (Chung et al., Reference Chung, Galano, Oger, Durand and Chung-Yung2015). Males are susceptible to lipid peroxidation when dietary carotene concentration is reduced, and appear to have an accelerated ageing rate (Pike et al., Reference Pike, Blount, Bjerkeng, Lindström and Metcalfe2007). In T. macdonaldi, sex itself did not appear to have a significant effect on oxidative stress indicators. However, grouping by reproductive stage suggested that reproductively immature (FI, MI) totoaba had higher SOD activity than mature (FM, MM) totoaba. This could also be associated with age; Rudneva et al. (Reference Rudneva, Skuratovskaya, Kuzminova and Kovyrshina2010) suggest three strategies in relation to enzyme activity vs age: (1) enzyme activity does not change with age, (2) it decreases with age, or (3) it increases with age. In the present study, mature (presumably older) totoaba had lower antioxidant enzyme activity; totoaba appear to fit within Rudneva et al. (Reference Rudneva, Skuratovskaya, Kuzminova and Kovyrshina2010)’s option 2. Totoaba reach reproductive maturity at 1.3 m (female) and 1.1 m (male) (De Anda-Montañez et al., Reference De-Anda-Montañez, García de León, Zenteno-Savín, Balart, Méndez-Rodríguez, Bocanegra-Castillo, Martínez-Aguilar, Campos-Dávila, Román-Rodríguez, Valenzuela-Quiñónez-Valenzuela, Rodríguez-Jaramillo, Meza-Chávez, Ramírez-Rosas, Saldaña-Hernández, Olguín-Monroy and Martínez-Delgado2013); thus, in general, immature totoaba may be considered to be <5 years old, while mature totoaba were >5 years old. In vertebrates, ageing is associated with higher ROS production (Sies, Reference Sies1997). However, no significant differences in O2•− production were observed between mature and immature (presumably younger) totoaba.
The variables that best explained the variability of O2•− production and TBARS levels in totoaba liver samples, as determined by the GLMs, included season, and therefore SST, and antioxidant enzyme activity. GLM results indicated that O2•− production and TBARS levels are related to SST with respect to reproduction; however, antioxidant enzymatic activity was associated only with reproduction. Fish and other aquatic species are susceptible to oxidative stress induced by abiotic factors, specifically changes in SST (Buchner et al., Reference Buchner, Abele-Oeschger and Theede1996; Lesser, Reference Lesser2006; Kong et al., Reference Kong, Wang and Li2008; Tremblay et al., Reference Tremblay, Gómez-Gutiérrez, Zenteno-Savín, Robinson and Sánchez-Velasco2010; Vinagre et al., Reference Vinagre, Madeira, Narciso, Cabral and Diniz2012; Madeira et al., Reference Madeira, Narciso, Cabral, Vinagre and Diniz2013). Our results suggest that totoaba has a moderate thermal sensitivity in terms of oxidative stress, and that observed changes may be more related to physiological variations, mainly in immature (potentially young) totoaba.
Possible variations of oxidative stress indicators in response to changes and the influence of abiotic factors under natural conditions have not been previously addressed in T. macdonaldi. Bañuelos-Vargas et al. (Reference Bañuelos-Vargas, López, Pérez-Jiménez and Peres2014) evaluated the influence of different diets made from soy protein and supplemental taurine on juvenile totoaba cultivated in aquaria, and suggest that taurine is a potential antioxidant in liver. Available evidence suggests differences in proximal composition and fatty acid content between wild and cultivated totoaba, in which cultivated totoaba presents higher fatty acid levels (López et al., Reference López, Durazo, Rodríguez-Gómez, True and Viana2006). However, it is necessary to increase research on issues related to health, mainly because this species is considered critically endangered.
ACKNOWLEDGEMENTS
The authors thank L. Campos Dávila, N.O. Monroy Olguín, J.J. Ramírez Rosas, L. Rivera Rodríguez, F. Valenzuela Quiñonez, O. Rodríguez García, H. Bervera León, M. Román Rodríguez, M. Vélez Alavez, J. Isboset Saldaña, F. Valverde, R. Martínez Rincón and O. Lugo Lugo for their invaluable assistance in the field, with sample collection and/or sample and data processing. We also thank the Fishermen Federation from San Felipe and Golfo de Santa Clara for assistance with fieldwork. Samples were collected under permits (SGPA/DGVS/02913/10, SGPA/DGVS/05508/11, SGPA/DGVS/00039/13 and SGPA/DGVS/00230/14) issued by the Mexican government.
FINANCIAL SUPPORT
This study was funded by SEP-CONACYT (2011-01/165376) and CIBNOR (EP). SBHA is a CONACYT postdoctoral fellow (175821).
