Effect of heat stress on the antioxidant defense system and erythrocyte morphology of Antarctic ﬁ shes

: This study analyzed the effect of thermal stress on erythrocytes of Notothenia rossii and Notothenia coriiceps , abundant notothenioids in Admiralty Bay, Antarctic Peninsula. In both species, the antioxidant defense system enzymes, superoxide dismutase, catalase, glutathione peroxidase, glutathione-S transferase, glutathione reductase were punctually altered (8°C for 1, 3 and 6 days) in erythrocytes, indicating that these markers are not ideal for termal stress. However, under the infl uence of thermal stress, morphological changes in Notothenia coriiceps erythrocytes were observed at all exposure times (1, 3 and 6 days at 8°C), and in Notothenia rossii occurred in 6 days. These results suggest that Notothenia corriceps presents a lower tolerance to thermal stress at 8°C for up to 6 days, since the cellular and nuclear alterations recorded are pathological and may be deleterious to the cells. Among the morphological markers analyzed in this work, we believe that the shape change and nuclear bubble formation may be good stress biomarkers in erythrocytes of Notothenia rossii and Notothenia coriiceps .


INTRODUCTION
Erythrocytes have gas exchange functions (Li et al. 2016), therefore, they are sensitive to oxidative stress because their cytoplasm has high concentrations of iron and oxygen, which can generate reactive oxygen species (ROS) by the Fenton reaction (Li et al. 2013). Furthermore, their membranes have a high polyunsaturated fatty acid content, making them thus more susceptible to ROS (Gwozdzinski et al. 2017). Some studies regarding marine fi sh have shown that short term exposure to high temperature, especially in an acute way, results in oxidative damage (Grim et al. 2013, Mueller et al. 2012, as a result of an interruption of electron transfer within the respiratory chain (Mueller et al. 2014). When the enzymatic and non-enzymatic components of the antioxidant defense system, are not able to eliminate or neutralize the ROS occurs oxidative stress, able to oxidize lipids and proteins and in more serious cases lead the cell to apoptosis (Li et al. 2013).
The Antarctic ichthyofauna has adapted to the Antarctic climate and temperature conditions, and changes in the thermal stability of the environment can be lethal (Bilyk & DeVries 2011, Crockett 2011, Brodte et al. 2008, di Prisco et al. 2007, Somero & DeVries 1967. The phase of Antarctic fi sh acclimation to heat can affect oxygen supply to tissues (Pörtner 2002), and heat stress can cause hyperglycemia, slow cortisol release in plasma and increased hematocrit in notothenioids (Heise & Abele 2008). Furthermore, increasing the temperature at 4°C for 6 days may increase lipoperoxidation in the erythrocytes of these animals (Klein et al. 2017). Studies have investigated the antioxidant defense system, of several tissues, of the Antarctic fish, especialty Notothenia rossii and Notothenia coriiceps, submitted to thermal stress, and were demonstrated that the responses may be species-specific and tissue-specific, making it difficult to establish biomarkers (Mueller et al. 2012, 2014, Machado et al. 2014, Sattin et al. 2015, Forgati et al. 2017, Klein et al. 2017 Other responses may be associated with oxidative stress and thermal stress in erythrocytes, as is the case with cellular alterations, since ROS may affect membrane permeability, cause nuclear changes (Nagasaka et al. 2004, Zafalon-Silva et al. 2017, lead to dephosphorylation of membrane proteins and lipids due to oxidation of hemoglobin, and cause formation of vesicles on the plasma membrane of erythrocytes (Vodyanoy 2015).
N. rossii and N. coriiceps were chosen in the present study, because they are phylogenetic species and environmentally similar (Near & Cheng 2008, Raga et al. 2015. In addition, these two species are abundant in biomass and are important for the Antarctic trophic chain, including the Admiralty Bay, a Special Site of Scientific Interest (SSSI No. 8) and an Antarctic Specially Managed Area (ASMA No. 1) (Arigony- Neto et al. 2004, Braun et al. 2001. N. rossii has slightly compressed head and body, with an obliquely arranged mouth in adulthood, light brown in color, with darker back, with black spots along the body. N. coriiceps has a broad body with a slightly depressed head, adult coloration is approaching dark brown and may have light brown spots (Fisher & Hureau 1985). Studies on climate change in this region have reported observations of accelerated warming. Over the past 60 years, the seawater temperature of that region at depths between 700-1100 m (between 35-65°S) has increased by 0.2°C (Beers & Sidell 2011), and recently, a similar increase has also been recorded in shallow waters (Turner et al. 2014, Oliva et al. 2016.
This paper was carried out to answer a question raised by Machado et al. (2014) who evaluated the effect of temperature acclimation on the liver antioxidant defence system of the N. coriiceps and N. rossii. The authors suggest, in addition to existing literature (Cassini et al. 1993, Ansaldo et al. 2000, erythrocytes may be important cells for antioxidant defense, especially in neutralizing of superoxides. Besides that, the use of fishes erythrocytes as a biomarker becomes interesting mainly because the analysis of the material is possible without the need of animal sacrifice. Thus, this work proposes to evaluate the effect of thermal stress on the antioxidant defence system, changes in cellular and nuclear morphology in erythrocytes of two endemic species of Antarctic fish, N. coriiceps and N. rossii. No progressive acclimation was use, and the animals were place directly into water at the experimental temperatures used (0°C to 8°C). Although the thermal shock exposes the animals to sharply higher temperatures than their natural environment, this approach is important for reveal physiological responses to high temperatures. Exposure at 8°C is close to the thermal maximum tolerated by Antarctic fishes (Strobel et al. 2013, Somero & DeVries 1967, to justify using the 8°C temperature in the study.

Experimental design
Notothenia rossii (n = 60; total length = 37 + 3 cm; weight= 791 ± 145 g) and Notothenia coriiceps (n = 60; total length = 35 + 3 cm; weight= 775 ± 182 g) were caught by hook and line at Admiralty Bay (61ºS -63º30'S, 53º55'W -62º50'W) and transported to Brazilian Antarctica Station Comandante Ferraz (EACF), located on King George Island in the South Shetlands Archipelago. After collection, fish were kept for 3 days in the laboratory under controlled conditions of temperature (-0.5º to 1.0ºC), 12-hours light and 12-hours dark photoperiod (Donatti & Fanta 2002), salinity (35 ± 1.5) and constant aeration, aiming to reduce collection stress. Next, fish were acclimated in the experimental laboratory for 3 days (Machado et al. 2014, Ryan 1995 under the same controlled abiotic conditions, except that the temperature was maintained at 0 ± 0.5ºC. Subsequently, fishes were fish were marked with a coloured line and acutely transferred to experimental tanks (1000 L of seawater) under the same abiotic acclimation conditions at two different temperature conditions, control (0±0.5°C) and experimental (8±0.5°C), for 1, 3 and 6 days using a total of 10 fish per experiment. The water temperature of the tanks was strictly maintain by thermostats coupled to heaters, and tanks were cleaned every two days by exchanging 50% of the water under the same experimental conditions.
Fish were fed individually every other day with muscle of Antarctic fish (1% of their body mass) (Crawford 1979, Egginton et al. 1991, Coggan 1997 hooked to a long rigid wire. The animals were observed to verify no regurgitation of the food (Donatti & Fanta 2002). Fish that failed to were excluded from the experiments. In all experiments, the first day of acclimation was also the first day of feeding, and both feeding and sacrifice occurred at the same time for the experimental groups and their respective controls. The last offer of food occurred 12 hours before the animal was sacrificed.
Fish were anaesthetized with 20 mg.L -1 benzocaine (from a stock solution of 0.1% (w.v -1 ) in 95% ethanol), and blood samples were obtained using heparinized syringes by caudal vein puncture and killed by spinal transection. Erythrocytes were obtain by centrifuging whole blood at 2,000g for 10 minutes at 4°C. After plasma and white cell removal, a portion of the erythrocytes were fixed in modified Karnovsky fixative (2% paraformaldehyde and 2.5% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.2, at 4°C) (Karnovsky 1965) for morphological analysis. A portion of the erythrocytes was placed in 0.9% NaCl, centrifuged at 5000 x g for 10 min at 4°C. Then, 20 mM Tris-HCl buffer pH 8.0 was added (3: 1 v/v) to obtain hemolysate, which was centrifuged at 12,000 x g for 15 min at 4°C, and the supernatant was frozen in liquid nitrogen to perform enzymatic analysis. The Environmental Assessment Group of the Brazilian Ministry of Environment granted the environmental licenses and the Ethics Committee of the Federal University of Paraná (UFPR) under number 496 approved the experiments.

Analytical methods
The total protein concentration was determine by the Bradford method (1976) using bovine serum albumin (BSA) to establish the standard curve, and the absorbance of the samples was determined at 595 nm.
The readings of superoxide dismutase, glutathione peroxidase, glutathione-Stransferase, glutathione reductase, concentration of glutathione and lipid peroxidation were performed in triplicate in an microplate spectrophotometer (Epoch Microplate Spectrophotometer, BioTek, Winooski, VT, USA), with the exception of catalase (CAT), for which a spectrophotometer with quartz cuvette was employed.

Antioxidant enzyme activity assays
Levels of superoxide dismutase (SOD, EC 1.15.1.1) activity were measured by inhibiting nitrobluetetrazolium (NBT) reduction to blue formazan by the O 2 generated by hydroxylamine (Crouchet al. 1981), and the absorbance was determined at 560 nm. The catalase (CAT, EC 1.11.1.6) activity levels were measured by the consumption of hydrogen peroxide (H 2 O 2 ) using a spectrophotometer at 240 nm following the method of Beutler (1975).
Levels of glutathione peroxidase (GPx, EC 1.11.1.9) activity were determine by NADPH oxidation (Wendel 1981). The reaction occurred at 25°C was initiate by the addition of 0.5 mM hydrogen peroxide and was monitored for 3 min at 340 nm.
Glutathione reductase (GR, EC 1.8.1.7) activity levels were evaluate spectrophotometrically by measuring the changes in absorbance at 340 nm due to the oxidation of NADPH concomitantly with the reduction of GSSG according to the method of Carlberg & Mannervik (1985). The reaction was initiate by the addition of GSSG and monitored for 10 min.
The antioxidant enzyme activities are expressed as units (U)/mg of protein. One unit of SOD activity was define as the amount of enzyme that inhibits the NBT oxidation reaction by 50% of the maximal inhibition. One unit of CAT activity is defined as the amount of enzyme consuming one mmol of substrate per minute.
One unit of GPx, GST and GR activity is defined as the amount of enzyme generating one µmol of product per minute. All enzyme activities were determined at 20°C (Jayasundara et al. 2013, Machado et al. 2014, Almeida et al. 2015, Forgati et al. 2017.

Oxidative stress indices
The concentration of glutathione (GSH) and other thiols was determined using the method described by Sedlak & Lindsay (1968). The method is based on protein precipitation and subsequent reaction of non-protein thiols with 5,5'-dithiobis(2-nitrobenzoic acid) (DTNB), generating a product that absorbs light at 415 nm. The GSH values are expresse as nmol of thiols per mg of protein -1 .
The lipid peroxidation (LPO) index was determined via the levels of malondialdehyde (MDA) through the reaction of TBARS in a spectrophotometer at 535 nm using malondialdehyde-(bis)-acetate as a standard (Uchiyama & Mihara 1978). MDA values are express as nmol of MDA per mg of protein -1 .

Morphological analysis
For light microscopy (LM), sample smear glass slides stained with May-Grunwald (Rios 2010) were use. The images were obtained using a Zeiss Imager.Z2 scanner microscope via the MetaSystems/VSViewer program.
For the morphometric analysis of cellular changes obtained in LM, 1,000 erythrocytes per individual were classified as erythrocytes with abnormal cell shape, erythrocytes with vesicle formation in the plasma membrane, erythrocytes with micronuclei, erythrocytes with blebbed nuclei, or erythrocytes with notched nuclei according to methodology adapted from Grisolia et al. (2009) and Souza & Fontanetti (2006).
To confirm the analyzes performed in LM, transmission electron microscopy (TEM) and scanning electron microscopy (SEM) were realized. For TEM analysis, erythrocytes were postfixed in 2% osmium tetroxide, 0.1 M cacodylate buffer, pH 7.2, contrasted en-bloc with 2% uranyl acetate and embedded in Epon-812 resin (Luft 1961). The ultra-thin sections were contrasted in aqueous 2% uranyl acetate (Watson 1958), lead nitrate (Reynolds 1963) and analyzed in a JEOL 1200EX II transmission electron microscope. For SEM, erythrocytes were submitted to critical point drying in a Bal-Tec CPD -030, metalized in gold in a Balzers SCD -030 and analyzed using a JEOL JSM-6360 LV scanning electron microscope. The analyzes were performed at the Electron Microscopy Center of Federal University of Paraná.

Statistical analysis
Morphometric data, oxidative stress enzyme levels (SOD, CAT, GPx, GST and GR) and nonenzymatic parameters (MDA and GSH) were initially tested for normal distribution using the Kolmogorov-Smirnov test and for variance homogeneity by Levene's test. If the data were homogeneous, they were subject to two-factor analysis of variance (two-way ANOVA) (F) in which the exposure time (TI) (1, 3 and 6 days) and temperature (TE) (0 and 8°C) were categorical variables. In the case of non-parametric data, non-parametric Kruskal-Wallis analysis (H) was use. In the case of significant variation the data were subjected to Tukey's post-test and multiple comparisons. To evaluate the relationship between levels of oxidative stress markers and morphological analysis variables were used correlation and linear regressions. All statistical analyzes were performed with P<0.05 as the level of significance.

RESULTS
All animals in this study survived to effect of independent variables, temperature (0 to 8 ° C) and exposure time (1, 3 and 6 days) no death being recorded. The results presented below indicate the differences between temperatures (0-8ºC). The differences between the exposure times at the same temperature can be seen in detail in Figures 1, 2 and 3. To both species and temperatures tested there were no correlation between oxidative stress markers and morphological changes (Table II). Also none of these variables included in the linear regression model were significant (Table I).

Antioxidant enzyme activity
The SOD levels were modulated by increasing temperature, only in N. rossii and displayed an up regulation at 8°C relative to 0°C over 6 days of exposure (Figure 1a and 1c). In N. rossii, CAT activity was modulate positively on day 3 and negatively on day 6 at 8°C. As for N. coriiceps, CAT activity was negatively modulated as the temperature increased on days 1 and 3 ( Figure  1b and 1d).
Regarding the GPx activity, there was no modulation in relation to the temperature in either species (Figure 2a and 2d). In N. rossii, the GST activity was reduced on day 6 at 8°C with respect to temperature of 8°C. As for GST activity at 8°C, in N. coriiceps, there was an up regulation on day 3 at 0°C and an increase in activity after 6 days (Figure 2b and 2e). The GR levels were modulated by increasing temperature, only in N. coriiceps, GR activity showed an up regulation on days 1 and 3 at 8°C (Figure 2c and 2f). All results of twoway ANOVA test are represents at Table II.  (a-b) and N. coriiceps (c-d) erythrocytes exposed to 0°C or 8°C (empty circles and solid squares, respectively) for 1, 3 or 6 days. The vertical lines indicate mean + SE. Different lowercase letters indicate differences among the exposure times (1, 3 and 6 days) at each temperature. Asterisks indicate differences (P≤0.05) between temperatures (0°C and 8°C).  N. rossii (a-b-c) and N. coriiceps (d-e-f) erythrocytes exposed to 0°C or 8°C (empty circles and solid squares, respectively) for 1, 3 or 6 days. The vertical lines indicate mean + SE. Different lowercase letters indicate differences among the exposure times (1, 3 and 6 days) at each temperature. Asterisks indicate differences (P≤0.05) between temperatures (0°C and 8°C).  (a and b) and N. coriiceps (c and d) erythrocytes exposed to 0°C or 8°C (solid squares and empty circles, respectively) for 1, 3 or 6 days. The vertical lines indicate mean + SE. Different lowercase letters indicate differences between the exposure times at each temperature (0°C and 8°C).

Oxidative stress indices
The GSH concentration was not modulated at 8°C for either species (Figure 3a and 3c). But the MDA content was modulated at 8°C in both species. In N. coriiceps, there was down regulation on day 3 at 8°C ( Figure 3d). As for N. rossii, there was up regulation on day 1 followed by negative modulation on day 3 (Figure 3b).

Morphological analysis
The morphological changes found in the erythrocytes of N. coriiceps and N. rossii are show in Figure 4b-n and Table III. In both species, mature erythrocytes showed oval to ellipsoid shapes. The central electrondense nucleus is due to the predominance of heterochromatin that follows the cell shape, and the cytoplasm has compartments delimited by membranes (Figure 4a). The temperature increase influenced the frequency of erythrocytes with altered shapes (Figure 4bc) with cases of cell lysis being observed by EM (Figure 4g-h). At 8°C, N. coriiceps showed more cell shape changes on days 1 and 3, whereas N. rossii changes were observe mainly on day 6. The 8°C, in N. coriiceps at 1 day of exposure the temperature increased the rate of erythrocytes presenting cell membrane vesicle formation (Figure 4d-e-f ). The frequency of micronuclei was unaffected by the increased temperature or the exposure time in both N. coriiceps and N. rossii. The frequency of blebbed nuclei was influence by the 8°C temperature in N. coriiceps on days 3 and 6 and in N. rossii on day 6. The frequency of notched nuclei ( Figure  4i) was influence by temperature and was higher at 8°C only for N. coriiceps on day 6.

DISCUSSION AND CONCLUSION
Studies have found that Antarctic fishes have higher antioxidant defence activity than tropical fishes (Coppe et al. 2013, Chen et al. 2008. This response appears to be related to the adaptation of these organisms to the low temperature conditions of the Antarctic environment, to the high dissolved oxygen amount and to the high concentration of intracellular lipids (Abele & Puntarulo 2004). These characteristics have led to significant specialization of the enzymes related to oxidative stress, thereby preventing them from being elevated as the temperature increases. And in N. rossii and N. coriiceps, the absence or weak modulation of antioxidant defenses is the more widespread behavior seen in the literature, as reported in several tissues      al. 2017, Machado et al. 2014, Mueller et al. 2014, 2012. The response patterns of the activity of antioxidant enzymes and morphology of erythrocyte relative to increased temperature were diferente between N. rossii and N. coriiceps. In both species, it were observed changes in the levels of enzymes of the antioxidant defense during at 0°C, we believe that such variations can be explained by operational reasons intrinsic to the experimental activity such as confinement and social isolation, since these factors can cause stress in fish, leading to physiological changes (Zuberi et al. 2014). However, to ensure that these factors did not influence the markers evaluated, the experiments at 8°C and their respective controls were carried out simultaneously and their experimental conditions, and abiotic factors strictly controlled, ensuring that differences observed between the experimental and control groups were due to temperature, the only variable tested in this study.
SOD is responsible for the conversion of the superoxide anion to H2O2, the latter being decomposed into O2 and H2O by CAT (Vasconcelos et al. 2007). These two enzymes are fundamental in antioxidant defense processes. Machado et al. (2014) analyzing the hepatic cells of N. rossii and N. coriiceps found no alterations in the activities of SOD and CAT and suggest that the modulation of these enzymes under thermal stress can be dependent on other cells, among them the erythrocytes, which have mechanisms for the removal of superoxide anions, since they carry high concentrations of oxygen. In N. rossii collected in the wild, SOD values were approximately 200-fold higher in erythrocytes than in liver (Ansaldo et al. 2000) and corroborate the high levels of SOD found in this work. In this work the temperature of 8°C was not able to modulate the erythrocyte SOD activity in N. coriiceps and similar results were observed by Klein et al. (2017). However, in N. rossii there was an increase in SOD activity in 6 days at 8 °C, this alteration may be a response to an increase in the production of superoxide radicals, resulting in a higher production of hydrogen peroxide, to which erythrocytes can be susceptible, since there was a decrease in CAT activity and absence in GPx modulation. The damage to hydrogen peroxide degradation may also have occurred in N. coriiceps, since there was a decrease in CAT activity in up to 3 days and no modulation of GPx.
The GST, GR and GPx enzymes integrate the antioxidant defense system and are markers used to detect oxidative stress against thermal stress in fish (Bagnyukova et al. 2007). In non-Antarctic marine fishes, the induction of heat stress may lead to an increase in GST activity to remove products of lipid peroxidation (Madeira et al. 2013). This may have occurred with N. coriiceps, since there was a decrease in the LPO levels in 3 days which was accompanied by increased GST at the same time, conversely, the decrease in MDA levels in 3 days may have influenced the reduction of GST levels in 6 days. Madeira et al. (2013) related that exposure to CTMax promotes protein denaturation, reducing antioxidant defenses and increasing LPO, indicating that in the case of strong oxidative stress, the activity of GST can be suppressed (Li et al. 2010).
The GR activity is involved in metabolism and the reduction of GSSG to GSH (Halliwell & Gutteridge 1999). However, studies warn that the action of multiple alternative systems present in the cytosol of erythrocytes, are able to reduce GSSG and maintain sufficiently high levels of GSH against oxidative stress (Morgan et al. 2013). In N. rossii, the absence of modulation of GR did not affect GSH levels, because this levels was not modulate by the temperature increase, thus supporting the theory that these oxidative stress markers may only be modestly affect by increased water temperature (Morgan et al. 2013). The GPx activity was unaffected by the elevated temperature in both species, this can be explained by microarray analysis reveals that the constitutive positive regulation of GPx in tissues of Antarctic notothenioids relative to that of temperate fishes is an evolutionary adaptation to an oxygen-rich environment (Chen et al. 2008).
Changes in erythrocyte cell and nuclear morphology are related to disorders that affect homeostasis (Tavares-Dias & Moraes 2004), including excess free radicals (Cimen 2008) and temperature variations (Qiang et al. 2013, de Pedro et al. 2005. Although the erythrocyte morphological changes are widely used to indicate environmental genotoxicity (Bhattacharjee & Das 2017, Maheshwari & Dua 2016, Grisolia et al. 2009), this methodology was pointed out as an efficient tool in the biomonitoring of thermal stress in notothenids (Zafalon-Silva et al. 2017). And this study attempted to ensure the quality and control of the experimental conditions, for both the animals submitted to the increased temperature (8°C) and control animals (0°C). Then, the increase registered in the number of erythrocytes with altered shapes and nuclear alterations, was due because of heat stress. The frequency of changes in erythrocyte shape was influence by theincrease in temperature, occurring more frequently in the early days for N. coriiceps and later for N. rossii. These changes in erythrocyte shape are related to cellular macromolecules that make up the cytoskeleton, including actin filaments, which are abundant in these cells and may lead to the loss of cytoplasmic contents as well as to apoptosis (Anbumani & Mohankumar 2012). Vesicle formation, occurs by the weakening of certain cytoskeletal areas (Wong et al. 2012), and cytoplasmic flow in these regions would have formed the protrusions (Taylor et al. 2008) observed in this work in N. coriiceps.
Changes in erythrocyte cell and nuclear morphology are related to disorders that affect homeostasis and may indicate an environmental impact on live organisms (Souza & Fontanetti 2006, Tavares-Dias & Moraes 2004, Talykina et al. 2003. It is believe that mutations in nuclear lamina could be responsible for the loss of the oval shape and the stability of the nuclear envelope (Anbumani & Mohankumar 2012, Strunjak-Perovic et al. 2009). In the present study, the temperature of 8°C did not influence the frequency of micronuclei, but it increased the frequency of blebbed e notched nuclei. Studies have indicated that these nuclear changes possess a similar origin to the micronucleus and allow for the removal and inactivation of damaged genetic material from the cell (Souza & Fontanetti 2006, Cavas & Ergene-Gozukara 2003. In addition, excess free radicals are capable of promoting DNA degradation and LPO in fish erythrocytes that damage cell membranes, causing irreversible damage that can lead to apoptosis (Cimen 2008).
Finally, in this study, statistical analysis showed that there was no correlation between oxidative stress markers and morphological changes indicating that it is not possible to relate the occurrence of oxidative stress to morphological changes. Point changes found in oxidative stress, indicating that these markers are not ideal for the thermal stress analysis in N. rossii and N. coriiceps erythrocytes. However, these results in conjunction with the Kandalski et al. (2018,2019) studies, which involved plasma analysis, showed that markers such as cortisol and catalase are good blood indicators of heat stress , and are more suitable for biomonitoring than markers of stress oxidative in erythrocytes.
Among the morphological markers analyzed in this work, the shape change and nuclear bubble formation may be good stress biomarkers in erythrocytes of N. rossii and N. coriiceps. The highest occurrence of cellular and nuclear alterations due to thermal stress was observed in N. coriiceps, being observed at all exposure times (1, 3 and 6 days), while in N. rossii the highest number of alterations occurred in 6 days. These results suggest that N. coriiceps presents a lower tolerance to thermal stress at 8°C for up to 6 days, since the cellular and nuclear alterations recorded are pathological and may be deleterious to the cells. Thus, in order to elucidate which metabolic pathways or cellular components were responsible for the cellular alterations observed in this work, it is necessary to continue investigating the physiology of these species under thermal stress.