Advertisement

Modeling the influence of time and temperature on the levels of fatty acids in the liver of Antarctic fish Trematomus bernacchii

  • Matteo Antonucci
  • Ikram Belghit
  • Cristina Truzzi
  • Silvia Illuminati
  • Pedro AraujoEmail author
Open Access
Original Paper

Abstract

Antarctic fish (Trematomus bernacchii) are an ideal group for studying the effect of ocean warming on vital physiological and biochemical mechanisms of adaptation, including changes in the fatty acid composition to higher heat tolerance in the sub-zero waters of the Southern Ocean. Despite the awareness of the impact of ocean warming on marine life, bioclimatic models describing the effect of temperature and time on fatty acid levels in marine species have not been considered yet. The objective of the present study was to investigate changes in the concentrations of fatty acids in liver from T. bernacchii in response to an increase in temperature in the Antarctic region. Changes in the concentrations of fatty acids in liver from T. bernacchii were observed after varying simultaneously and systematically the temperature and time. The fatty acid profiles were determined by gas chromatography prior to acclimation (− 1.8 °C) and after acclimation (0.0, 1.0, and 2.0 °C) at different times (1, 5, and 10 days). The observed changes were graphically visualized by expressing the fatty acid concentration in absolute units (mg g−1) as a function of the temperature and time using polynomial models. Major changes in fatty acid composition were observed at day 1 of exposition at all temperatures. At day 5, the fish seem to tolerate the new temperature condition. The concentrations of saturated fatty acids were almost constant throughout the various conditions. The concentrations of monounsaturated fatty acids (in particular 18:1n − 9) decrease at day 1 for all temperatures. In contrast, there was an increase in the concentrations of polyunsaturated fatty acids (in particular 20:5n − 3 and 22:6n − 3) with increasing temperatures after 1, 5, and 10 days of exposure. The proposed models were in agreement with reported studies on polar and temperate fish, indicating possibly similar adaptation mechanisms for teleost to cope with global warming.

Keywords

Global warming Bioclimatic models Thermal acclimation Gas chromatography Nototheniidae species Lipid composition 

Introduction

Local climate models are representations of the likely effect of anthropogenic sources on climate (Intergovernmental Panel for Global Climate Change (IPCC) 2000). Bioclimatic models have predicted the extinction of Antarctic toothfish in 30 years under strong warming conditions (Cheung et al. 2008). Data from Pleistocene glaciation have indicated that species more often responded to climate change by local adaptation (Parmesan et al. 2000). It has been reported that gradual increase in water temperature from global warming may result in changes in species composition and acclimation to higher heat tolerance (O’Connor et al. 2007; Cheung et al. 2008). It is undeniable that sea temperature is the principal factor determining distributions, abundance, and physiology of fish species (Pörtner 2001; Roessig et al. 2004). The Commission for the Conservation of the Antarctic Marine Living Resources has pointed out that the Antarctic region and more specifically the Ross Sea have not been dramatically affected by commercial fisheries, whaling, and other human activities when compared to its north counterpart (Antarctic and Southern Ocean Coalition (ASOC) 2009). For this particular reason, this protected region without many human population or other anthropomorphic interference provides a unique potential for documenting the biological and chemical effects of global warming on a polar marine ecosystem (Antarctic and Southern Ocean Coalition (ASOC) 2009). Fatty acids (FA) are regarded as ideal biological markers of environmental exposure in marine research, because they are able to objectively assess dietary intake/status and to identify key processes impacting the dynamics of some of the major ecosystems in the world (Dalsgaard et al. 2003; Könneke and Widdel 2003; Alfaro et al. 2006; Sajjadi and Eghtesadi-Araghi 2011; Copeman et al. 2013; Mourente et al. 2015). Recently, the fatty acid levels of T. bernacchii have been experimentally determined in liver, muscle, and gills (Truzzi et al. 2017, 2018a, b; Corsolini and Borghesi 2017; Malekar et al. 2018).

There is a paucity in information regarding the behavior of the fatty acid profiles when the factors time and temperature are varied simultaneously. The majority of published studies are mainly concerned with the effect of the temperature on the concentrations of fatty acids without considering temporal changes (Ramesha and Thompson 1982; Dey et al. 1993; Buda et al. 1996; Lahdes et al. 2000; Brodte et al. 2008). Some authors have investigated the temporal effect of temperature and tissue type (liver and muscle) on the levels of fatty acids in fish (Copeman et al. 2013). However, the main drawback of this study is that influence of the temperature and time were evaluated at two and five levels, respectively, which precluded the modeling of the factors, as reflected by the lack of degrees of freedom to judge the adequacy of the temperature (at two levels) in a basic model of the form y = mx + b (only one distinct line can be drawn through two different points).

The hypothesis of the present research is to investigate whether mathematical modeling of selected environmental variables (time and temperature) can predict changes in the fatty acid concentration of Antarctic fish T. bernacchii. This ubiquitous species of stenothermal Antarctic teleost (family Nototheniidae) is well adapted to the extremely low temperatures of the Antarctic region (− 1.86 °C) and has been regarded as a key indicator for monitoring anthropogenic impact (Regoli et al. 2005; Di Bello et al. 2007; Illuminati et al. 2010). Considering the important role of lipids in polar fish, the aim of the present study was to investigate the warm acclimation response of T. bernacchii exposed to higher temperatures. It is important to mention that the present research covers some of the 80 priority questions for the Southern Polar Regions that were formulated by researchers during the First Scientific Committee on Antarctic Research, more specifically: predictions of the response of the Antarctic to a warming world (Kennicutt et al. 2015). Climate change has raised concerns over the fate of notothenioids. Although after many decades of research on the adaptation of polar fish to climate warming, there are still many questions that remain to be investigated. The present study provides information on the acclimation of the dominant fish sub‐order to a thermal shock, which helps to better understand the response of these Antarctic species to the global warming. In addition, the computed models used in the present study may predict changes in the lipid profile of Antarctic T. bernacchii in relation to the global warming.

Materials and methods

Reagents

Sodium hydroxide, hexane, methanol, boron trifluoride in methanol (20% w v‒1), and chloroform were purchased from Merck (Darmstadt, Germany). Cod liver oil standard was from Peter Møller (Lysaker, Norway), and fatty acid methyl ester (FAME) pure and model mixture standards (Online Resource 1) were purchased from Nu-Chek Prep (Elysian, MN). Nonadecanoic acid methyl ester (19:0) internal standard was from Fluka (Buchs, Switzerland). De-ionized water was purified in a Milli-Q system (Milli-Q system Millipore, Milford, MA).

Sampling

The experimental study with Antarctic T. bernacchii is reported in detail elsewhere (Truzzi et al. 2018a, b). Briefly, the study was performed at the Mario Zucchelli Station at the Terra Nova Bay in the Antarctic region (Fig. 1) during the Austral summer 2014 (a particular period that is characterized by the lack of ice in the Ross Sea), by following the institutional and national guidelines for the care and use of animals and approved by the Italian Ministry of Foreign Affairs. The Code of Conduct of the Scientific Committee on Antarctic Research (SCAR 2011) was followed for the handling and sacrificing of animals. A total of 63 sexually mature Antarctic T. bernacchii species (weight 113–503 g, length 32–22 cm) were caught by a fishing rod at depths of 30 m in the location indicated in Fig. 1. The seawater temperature at the moment of the capture was − 1.87 °C. Three fish were immediately sacrificed by a sharp blow to the head (designated as control seawater at 0 day and − 1.8 ± 0.2 °C in Table 1) and their liver isolated and immediately frozen in liquid N2 and stored at − 80 °C to preserve them from oxidation and ensure their integrity. Following collection, the rest of the fish (60) were transferred into a net to Mario Zucchelli Station and placed into a flow-through circulating seawater tank containing 1000 L of filtered running seawater at − 1.8 ± 0.2 °C for acclimation at habitat temperature for 30 days. The overview of the sampling flow is portrayed in Online Resource 2. The fish were maintained under a natural photoperiod (24 h daylight). After the acclimation period, six fish were killed by a sharp blow on the head, their liver isolated, frozen in liquid N2, and stored at − 80 °C until analysis (designated as control after acclimation in Table 1). The rest of the acclimatized fish (54) were divided into three different groups and transferred to three new flow-through circulating seawater tanks at pre-established temperatures (0.0, + 1.0, and + 2.0 °C) and held at these conditions for 1, 5, and 10 days (Online Resource 2). The choice of temperatures of 0, + 1, and + 2 °C as thermal stress was based on the shelf water warming of + 0.8 to + 1.48 °C predicted by the year 2200 for the Ross Sea region (Timmermann and Hellmer 2013). Six fish were killed, at each time period, their liver isolated, frozen in liquid N2, and stored at − 80 °C until analysis. Fish were not starved before each sampling period. The same diet was provided ad libitum once every second day to all fish during the trials (including acclimation period). The diet, sampled at the same period and area, consisted of chopped cuttlefish (Sepia officinalis) and bivalve molluscs (Adamussium colbecki) and it was kept in the freezer throughout the experimental trial. The same batch of diet was given to the different groups of fish in the same amounts during the acclimation and experimental periods. A comparison of the fatty acid profiles of the caught fish in the Ross Sea (control seawater) and the fish acclimatized for 30 days was performed to check for differences before and after the acclimatization period and establish whether the experimental diet match the natural diet of T. bernacchii in terms of fatty acids.
Fig. 1

Location of the sampling point at the Mario Zucchelli Station (74°42′052″ South, 164°02′267″ East), Antarctic region

Table 1

Absolute concentrations of fatty acids (mg g‒1) in the liver of Antarctic fish Trematomus bernacchii of the caught fish in the Ross Sea (0 day and − 1.87 °C), fish acclimatized for 30 days in a circulating seawater tank control at − 1.87 °C, and at different times (1, 5, 10 days) and temperatures (0.0, 1.0, 2.0 °C) after the acclimation period

Time (days)

0

Acclimatized

1

5

10

1

5

10

1

5

10

Temperature (°C)

− 1.87

− 1.87

0.00

0.00

0.00

1.00

1.00

1.00

2.00

2.00

2.00

14:0

0.96 ± 0.15

0.57 ± 0.04

0.39 ± 0.04

1.03 ± 0.01

0.58 ± 0.07

0.25 ± 0.02

0.51 ± 0.01

0.32 ± 0.09

0.28 ± 0.02

0.85 ± 0.00

0.60 ± 0.04

16:0

5.36 ± 0.39

3.99 ± 0.43

3.82 ± 0.22

5.13 ± 0.16

4.91 ± 0.72

2.75 ± 0.23

4.20 ± 0.13

3.62 ± 0.24

3.30 ± 0.13

4.75 ± 0.03

5.17 ± 0.46

18:0

1.14 ± 0.08

1.00 ± 0.10

0.96 ± 0.07

1.26 ± 0.07

1.18 ± 0.15

0.77 ± 0.03

1.09 ± 0.10

1.06 ± 0.04

0.78 ± 0.03

1.11 ± 0.01

1.24 ± 0.12

18:3n − 3

0.07 ± 0.01

0.06 ± 0.01

0.07 ± 0.01

0.08 ± 0.00

0.09 ± 0.02

0.06 ± 0.00

0.10 ± 0.01

0.05 ± 0.03

0.04 ± 0.00

0.09 ± 0.01

0.06 ± 0.00

20:5n − 3

2.50 ± 0.14

3.20 ± 0.78

4.19 ± 0.16

3.94 ± 0.05

4.51 ± 0.87

2.94 ± 0.14

4.71 ± 0.50

4.76 ± 0.61

3.44 ± 0.10

5.18 ± 0.16

3.19 ± 0.20

22:5n − 3

0.11 ± 0.01

0.17 ± 0.03

0.25 ± 0.02

0.21 ± 0.00

0.43 ± 0.09

0.12 ± 0.00

0.25 ± 0.01

0.21 ± 0.07

0.15 ± 0.01

0.21 ± 0.00

0.18 ± 0.10

22:6n − 3

3.31 ± 0.16

5.01 ± 1.43

6.95 ± 0.04

5.98 ± 0.09

7.35 ± 1.55

4.14 ± 0.22

7.39 ± 0.67

7.25 ± 1.31

4.98 ± 0.16

8.60 ± 0.34

4.79 ± 0.36

18:2n − 6

0.41 ± 0.04

0.32 ± 0.04

0.38 ± 0.04

0.40 ± 0.01

0.40 ± 0.07

0.21 ± 0.02

0.39 ± 0.02

0.22 ± 0.08

0.18 ± 0.01

0.47 ± 0.00

0.33 ± 0.01

20:4n − 6

0.65 ± 0.04

0.64 ± 0.12

0.94 ± 0.04

0.67 ± 0.01

0.86 ± 0.13

0.47 ± 0.01

0.84 ± 0.06

0.79 ± 0.10

0.68 ± 0.03

0.83 ± 0.02

0.64 ± 0.00

16:1n − 9

0.36 ± 0.02

0.41 ± 0.05

0.33 ± 0.04

0.37 ± 0.01

0.42 ± 0.07

0.21 ± 0.00

0.43 ± 0.00

0.34 ± 0.05

0.18 ± 0.01

0.44 ± 0.00

0.51 ± 0.06

18:1n − 9

7.57 ± 0.69

5.92 ± 0.56

4.83 ± 0.60

9.16 ± 0.39

5.87 ± 0.74

2.71 ± 0.20

5.94 ± 0.58

3.57 ± 1.00

2.02 ± 0.13

6.36 ± 0.02

6.79 ± 0.40

20:1n − 9

0.84 ± 0.05

0.73 ± 0.05

0.69 ± 0.05

0.74 ± 0.01

0.74 ± 0.09

0.35 ± 0.02

0.68 ± 0.02

0.65 ± 0.07

0.42 ± 0.01

0.76 ± 0.00

0.72 ± 0.04

22:1n − 9

0.18 ± 0.02

0.17 ± 0.02

0.24 ± 0.00

0.24 ± 0.00

0.28 ± 0.05

0.17 ± 0.01

0.22 ± 0.02

0.18 ± 0.07

0.18 ± 0.00

0.28 ± 0.02

0.22 ± 0.03

16:1n − 7

3.05 ± 0.19

3.12 ± 0.34

2.70 ± 0.29

8.32 ± 0.42

3.88 ± 0.61

1.81 ± 0.12

3.39 ± 0.13

2.23 ± 0.71

1.04 ± 0.05

2.85 ± 0.04

4.11 ± 0.52

18:1n − 7

2.90 ± 0.27

2.33 ± 0.22

2.07 ± 0.16

2.69 ± 0.06

2.65 ± 0.42

1.39 ± 0.08

2.53 ± 0.12

1.88 ± 0.24

1.43 ± 0.08

2.46 ± 0.01

2.76 ± 0.17

Saturated

7.46 ± 0.43

7.46 ± 0.57

5.17 ± 0.24

7.43 ± 0.17

6.67 ± 0.74

3.77 ± 0.23

5.79 ± 0.16

5.00 ± 0.26

4.36 ± 0.13

6.72 ± 0.03

7.01 ± 0.48

Monounsaturated

14.9 ± 0.77

14.84 ± 1.24

10.9 ± 0.69

21.5 ± 0.58

13.8 ± 1.05

6.64 ± 0.25

13.2 ± 0.61

8.86 ± 1.26

5.27 ± 0.16

13.1 ± 0.05

15.1 ± 0.68

Polyunsaturated

7.06 ± 0.22

7.06 ± 2.47

12.8 ± 0.17

11.3 ± 0.1

13.6 ± 1.79

7.95 ± 0.26

13.7 ± 0.84

13.3 ± 1.45

9.48 ± 0.19

15.4 ± 0.38

9.18 ± 0.36

Total lipids

30.4 ± 2.4

30.38 ± 4.28

29.7 ± 1.8

41.4 ± 1.1

35.4 ± 5.8

18.9 ± 0.4

33.6 ± 0.58

27.8 ± 4.9

19.9 ± 0.69

36.3 ± 0.75

32.3 ± 1.5

The values are expressed as average and standard deviation (µ ± σ) for n = 2

Significant differences (p < 0.05) between seawater control (0 day, − 1.87 °C) and a specific condition are indicated as gray cells

Saturated = sum of 14:0, 16:0, 18:0

Monounsaturated = sum of 16:1n − 7, 18:1n − 7, 16:1n − 9, 18:1n − 9, 20:1n − 9, 22:1n − 9

Polyunsaturated = sum of 18:2n − 6, 20:4n − 6, 18:3n − 3, 20:5n − 3, 22:5n − 3, 22:6n − 3

Lipid extraction and fatty acid determination

The sampled livers were pooled together (3 livers for the control seawater, 6 for acclimation, and 6 for every specific time/temperature condition), minced and homogenized by means of a high dispersing device (Ultra-Turrax T25, Janke e Kunkel, IKA-Labortechnik, Staufen, Germany). The pooling of the samples was based on the characterization of biological variations in terms of an effective number, e.g., the average pooled concentration (Gregorius 1991, 2016). The effective number is a familiar notion in ecology, which results from equating an observed dispersion with the dispersion of an ideal community and solving for the number of types in the latter (Gregorius and Gillet 2015). When applied to the present fish community, the concept of concentration effective number requires the definition of an ideal community (corresponding to the control seawater at 0 day and − 1.8 ± 0.2 °C), that is equated to a pooled average value X. Consequently, any other community at different day and temperature with the same X value is equivalent to the control seawater. It is important to mention that not all dispersion measures based on dissimilarities allow for the computation of concentration effective numbers. For example, nutritional models evaluating the impact of different diets and where the notion of the ideal diet is unknown require computing the biological variability of the individuals within the groups (Gregorius and Gillet 2015). The mathematics behind the requirements for asserting the validity of an effective number are described elsewhere (Gregorius 1991, 2016; Jost 2010; Gregorius and Gillet 2015; Daly 2018). The lipids were extracted in duplicate (n = 2) by adding 4 mL of chloroform/methanol (2: 1, v v‒1) and 0.080 mL of internal standard (19:0 methyl ester) to 200 mg of sample. The extracted lipids were filtered, the residue was discarded, and the chloroform/methanol filtrate was evaporated under a stream of nitrogen. The dried residue was saponified according to a method published elsewhere (Torstensen et al. 2004). Briefly, an aliquot (1 mL) of sodium hydroxide in methanol (0.5 M) was added, heated at 100 °C for 15 min, and cooled in cold water. An aliquot (2 mL) of boron trifluoride in methanol (20%) was added, vortex-mixed, heated at 100 °C for 5 min, and cooled in cold water. Subsequent aliquots of hexane (2 mL) and water (2 mL) were added, vortex-mixed, centrifuged at 1207×g for 1 min, and the hexane phase collected. An additional 2 mL aliquot of hexane is added to the remaining water, vortex-mixed, centrifuged at 1207×g for 1 min, and the hexane phase pooled together with the initial collection. Depending on the fat content, the sample is either concentrated under nitrogen or diluted with hexane and subsequently subjected to gas chromatography analysis.

Instrument

A Perkin-Elmer Autosystem XL, equipped with a liquid autosampler and a flame ionization detector was used. The FAME samples were analyzed on a CP-Sil 88 capillary column (50 m, 0.32 mm ID, 0.2 µm film thickness; Varian, Courtaboeuf, France). Data collection was performed by the Perkin−Elmer Chromeleon® version 7. The analysis time was 60 min and the temperature program was as follows: the oven temperature was held at 60 °C for 1 min, ramped to 160 °C at 25 °C per min, held at 160 °C for 25 min, ramped to 220 °C at 3 °C per min, held at 220 °C for 10 min. Direct on−column injection was used. The injector port temperature was ramped instantaneously from 50 to 250 °C and the detector temperature was 250 °C. The carrier gas was ultra-pure helium at a pressure of 80 kPa.

Data analysis

The impact of the acclimation period on the fatty acid profile from liver of T. bernacchii was evaluated by comparing the variance components at 0 and 30 days. After 30 days of acclimation, the effect of time (0, 1, 5, and 10 days) and temperature (− 1.8, 0.0, 1.0 and 2.0 °C) on the levels of fatty acids in liver of T. bernacchii were evaluated by using the experimental arrangement described in Table 1. The variables were modeled according to a quadratic polynomial of the form:
$$y = b_{o} + b_{1} x_{1} + b_{2} x_{2} + b_{12} x_{1} \times x_{2} + b_{1}^{2} x_{1}^{2} + b_{2}^{2} x_{2}^{2}$$
(1)
where y represents the concentration of fatty acid in mg g‒1, the terms x1 and x2 represent the variables time and temperature, respectively, the term bo represents the intercept, b1 and b2 are the linear coefficients, \(b_{12}\) is the first-order interaction effect coefficient, and \(b_{1}^{2}\)\(b_{2}^{2}\) are second-order curvature effect coefficients. A detailed description of the levels of the variables time and temperature is given in Table 1. Before modeling the various fatty acids, their normality was evaluated by comparing the values of the experimental distribution with the corresponding points of the normal distribution using inverse normal plots (Kirkwood and Sterne 2003). Briefly, the inverse normal values (inv) are calculated by arranging the experimental means (µ) in increasing order and the value of the standard normal distribution corresponding to its quantile (aka probit) is multiplied by the corresponding standard deviation (σ) by the expression:
$$inv = \mu + probit \times \sigma$$
(2)

The original µ ± σ values are plotted against the computed inv values and the inverse normal graphs are generated by using Microsoft Excel 2013 (Online Resource 3).

A comprehensive explanation of the different steps involved in the development of the mathematical model for stearic acid (18:0) is given in Fig. 2. The various features of Fig. 2 are described by using roman numerals. The first step consisted of running in random order the experimental matrix, (i) which contains duplicate conditions for time (x1) and temperature (x2), and determining the responses (y18:0) which are expressed in mg g−1 of fatty acid. (ii) Secondly, the modeling process is carried out by constructing a design matrix (iii) with a number of columns (6 columns) equivalent to the number of coefficients in Eq. 1 (6 coefficients). The six columns in (iii) are constructed as follows: the first column (x0) is filled in with a numerical value of 1 and it is used to estimate the intercept (b0). The two subsequent columns (in gray), filled in with the experimental values of time (x1) and temperature (x2), are used to determine the coefficients b1 and b2. The fourth column (x12) is the result of multiplying columns x1 and x2 and allows computing the coefficient b12. The last two columns (\({x}_{1}^{2}\) and \(x_{2}^{2}\)) represent the square of the columns x1 and x2 and they are used to calculate the coefficients \(b_{1}^{2}\)\({b}_{1}^{2}\) and \(b_{2}^{2}\). The vector y18:0 (ii) is regressed on the matrix (iii) and equation (iv) is computed. The validity of (iv) is confirmed by using the ratio between the lack-of-fit mean square and the pure error mean square at a 95% confidence level (Online Resource 4) and described elsewhere (Araujo 2009). After validating model (iv), a matrix of estimated responses (v) is determined within the boundaries of the experimental ranges by substituting in (iv) the variables x1 and x2 (between 0 and 10 days and between − 2 and 2 °C, respectively). Matrix (v), for example, indicates that the concentration of 18:0 at 0-day/2 °C (0.59 mg g−1) is lower than 10-days/2 °C (1.21 mg g−1). To rapidly summarize the large amount of data in matrix (v), a graphical display of (vi) can be used. The graphical display in Fig. 2 is more practical and operable and shows the distribution of data through color change. Every color in (vi) represents areas with similar concentrations. For example, according to (vi), the brown and gray colors indicate concentration ranges of 0.5–1.0 mg g−1 and 1.0–1.5 mg g−1, respectively. In addition, a rapid inspection of (vi) allows concluding that condition 0-day/2 °C exhibits a lower concentration than condition 10-days/2 °C, without the need of a detailed examination of the numerical matrix (iv). The visualization of the computed regressions for the different fatty acids was performed by using Microsoft Excel 2013. Statgraphics Centurion XVI (Version 16.1.11, StatPoint Technologies, Inc., Warrenton, VA, USA) was used to evaluate statistically the differences between the means of the control seawater and the acclimatized groups at different temperatures and times. Dunnett’s t test was employed as correction procedure for comparing the various groups with the seawater control.
Fig. 2

Steps involved in the development of the mathematical model for stearic acid (18:0). The experimental matrix (i) is proposed and the corresponding responses (ii) are obtained. A design matrix (iii) is constructed and the equation (iv) is computed by regressing (ii) on (iii). The validity of (iv) is confirmed statistically (see Online Resource 4) and a matrix of estimated responses (v) is generated within the experimental limits of the variables x1 (from 0.0 to 10.0 days) and x2 (from − 1.9 to 2.0 °C). The large amount of data in (v) is plotted and summarized in (vi). The colors in (vi) represent areas with similar concentrations and they are read as follows: if a black dashed line is drawn at day 1 between − 1.9 and 2.0 °C, then moving up along this black line allows estimating a concentration range of 1.0–1.5 mg g−1 (gray color) between − 1.9 and − 0.4 °C and a concentration range of 0.5–1.0 mg g−1 (brown color) between − 0.4 and 2.0 °C. Similarly, the red dashed line at day 5 and between − 1.9 and 2.0 °C allows estimating only one concentration range of 1.0–1.5 mg g−1 (gray color). Consequently, it is concluded that between 5 and 10 days, the concentration of 18:0 remains constant in the range of 1.0–1.5 mg g1 in the whole range of explored temperatures

Results

The length and weight of the fish were computed throughout the course of the experiments. These parameters for the control seawater at 0 day were similar to those measured at different times and temperatures. Hence, the effect of allometry was negligible (Online Resource 5).

The average and standard deviation of the fatty acid profiles of the caught fish in the Ross sea (control seawater) and the fish acclimatized for 30 days in a circulating seawater tank were determined (Table 1) and the ratios of the between/within variance components for these profiles (aka Fisher ratio or F test) computed and compared against the tabulated F value of 18.51 (p = 0.05) for 1 and 2 degrees of freedom in the numerator and denominator, respectively. The results revealed the absence of statistical significant differences between the two types of controls (Online Resource 6).

The fatty acid profiles from liver of T. bernacchii at different times (1, 5, and 10 days) and temperatures (0, 1, and 2 °C) after the acclimation period are described in Table 1. The oleic acid (18:1n − 9) and docosahexaenoic acid (22:6n − 3) exhibited the highest concentrations in Table 1, followed by palmitic oleic acid (16:1n − 7) and eicosapentaenoic acid (20:5n − 3) (9.16 ± 0.39, 8.60 ± 0.34, 8.32 ± 0.42, and 5.18 ± 0.16 mg g‒1, respectively). The inverse normal plots (obtained by using Online Resource 3) revealed a high degree of linearity for the various fatty acids (Online Resource 7) confirming that the data in Table 1 were normally distributed. Quadratic polynomial models were generated by using Eq. 1 and the data in Table 1. The behavior of the saturated fatty acids (SFA) (14:0, 16:0, 18:0), monounsaturated fatty acids (MUFA) (16:1n − 7, 18:1n − 7, 16:1n − 9, 18:1n − 9, 20:1n − 9, 22:1n − 9), and polyunsaturated fatty acids (PUFA) (18:2n − 6, 20:4n − 6, 18:3n − 3, 20:5n − 3, 22:5n − 3, 22:6n − 3) (mg g‒1) as a function of the variables time (days) and temperature (°C) is visualized in Fig. 3. The results exhibited good prediction capabilities as observed in Tables 1, 2, where the experimental and predicted concentrations did not revealed any statistical difference between them. Consequently, if a model reproduces the concentrations at the experimental times and temperatures, then it must also reproduce any concentration at any point allocated inside the experimental ranges. The main goal of modeling environmental variables is predicting valid information from the proposed models. This very common approach is implemented in worldwide laboratories where calibration curves are prepared using a preselected concentration range and predictions are made in areas between the experimental points. The computed regression equations are described in Table 2. A variation in fatty acid concentration around 15% was indicative of change in stability. This threshold was based on a study of IUPAC (Firestone and Horowitz 1979) and the FDA recommendations (FDA 2013) that acknowledged 15% as the maximum variation for fatty acid analysis.
Fig. 3

Mathematical models describing the variation in mg of fatty acid per g of liver of T. bernacchii as a function of time and temperature (colored areas in mg g‒1)

Table 2

Predicted concentrations in mg g−1 and regression models used for calculations. The bracketed statistical t values indicate no significant differences between predicted and experimental concentrations in Table 1 (p = 0.01)

Days (x1)

0

1

5

10

1

5

10

1

5

10

Regression models

°C (x2)

− 1.87

0

0

0

1

1

1

2

2

2

 

14:0

0.93 (0.19)

0.46 (1.63)

0.84 (35.70)

0.40 (2.39)

0.28 (1.20)

0.76 (24.32)

0.44 (1.36)

0.21 (4.52)

0.79 (14.30)

0.59 (0.37)

\(y = 0.26 + 0.22x_{1} - 0.26x_{2} + 0.02x_{1} x_{2} - 0.02x_{1}^{2} + 0.06x_{2}^{2}\)

16:0

5.29 (0.18)

3.78 (0.19)

4.79 (2.22)

4.08 (1.16)

3.22 (2.07)

4.59 (3.06)

4.39 (3.12)

2.94 (2.83)

4.70 (1.63)

4.98 (0.41)

\(y = 3.32 + 0.50x_{1} - 0.80x_{2} + 0.10x_{1} x_{2} - 0.04x_{1}^{2} + 0.14x_{2}^{2}\)

18:0

1.14 (0.05)

0.95 (0.14)

1.19 (0.97)

1.09 (0.55)

0.84 (1.81)

1.15 (0.64)

1.15 (2.29)

0.73 (1.87)

1.12 (0.56)

1.21 (0.26)

\(y = 0.84 + 0.11x_{1} - 0.14x_{2} + 0.02x_{1} x_{2} - 0.01x_{1}^{2} + 0.01x_{2}^{2}\)

18:3n − 3

0.07 (0.31)

0.07 (0.24)

0.10 (7.19)

0.07 (1.76)

0.06 (0.36)

0.09 (0.54)

0.07 (0.68)

0.05 (1.337)

0.08 (10.19)

0.06 (0.35)

\(y = 0.05 + 0.02x_{1} - 0.01x_{2} + 0.001x_{1} x_{2} - 0.001x_{1}^{2} - 0.001x_{2}^{2}\)

20:5n − 3

2.41 (0.70)

3.66 (3.28)

4.67 (13.65)

4.18 (0.37)

3.68 (5.11)

4.69 (0.04)

4.21 (0.91)

3.39 (0.52)

4.42 (4.65)

3.94 (33.06)

\(y = 3.21 + 0.49x_{1} + 0.16x_{2} + 0.002x_{1} x_{2} - 0.004x_{1}^{2} - 0.15x_{2}^{2}\)

22:5n − 3

0.10 (1.14)

0.22 (1.38)

0.26 (13.76)

0.30 (1.36)

0.20 (26.63)

0.24 (1.61)

0.28 (0.97)

0.13 (2.58)

0.17 (11.23)

0.20 (2.24)

\(y = 0.21 + 0.01x_{1} + 0.01x_{2} - 0.001x_{1} x_{2} - 0.0003x_{1}^{2} + 0.03x_{2}^{2}\)

22:6n − 3

3.25 (0.35)

5.55 (39.00)

7.48 (17.38)

6.60 (0.49)

5.57 (6.61)

7.49 (0.14)

6.58 (0.51)

5.06 (0.55)

6.96 (4.85)

6.03 (3.44)

\(y = 4.70 + 0.92x_{1} + 0.29x_{2} - 0.005x_{1} x_{2} - 0.07x_{1}^{2} - 0.26x_{2}^{2}\)

18:2n − 6

0.42 (0.10)

0.32 (1.80)

0.45 (3.13)

0.30 (1.47)

0.25 (2.06)

0.42 (1.52)

0.31 (1.07)

0.20 (1.41)

0.40 (19.71)

0.33 (0.31)

\(y = 0.25 + 0.08x_{1} - 0.08x_{2} + 0.01x_{1} x_{2} - 0.01x_{1}^{2} + 0.01x_{2}^{2}\)

20:4n − 6

0.66 (0.13)

0.74 (5.46)

0.82 (10.33)

0.78 (0.66)

0.71 (23.35)

0.80 (0.73)

0.77 (0.21)

0.63 (1.86)

0.73 (6.44)

0.72 (23.08)

\(y = 0.71 + 0.04x_{1} - 0.01x_{2} + 0.003x_{1} x_{2} - 0.003x_{1}^{2} - 0.02x_{2}^{2}\)

16:1n − 9

0.38 (0.88)

0.26 (1.65)

0.42 (5.27)

0.38 (0.52)

0.22 (2.95)

0.41 (10.07)

0.41 (1.70)

0.21 (2.55)

0.43 (4.10)

0.48 (0.58)

\(y = 0.20 + 0.07x_{1} - 0.07x_{2} + 0.01x_{1} x_{2} - 0.01x_{1}^{2} + 0.02x_{2}^{2}\)

18:1n − 9

7.74 (0.24)

4.62 (0.36)

7.94 (3.18)

5.07 (1.07)

2.93 (1.10)

7.04 (1.90)

5.19 (1.61)

1.70 (2.48)

6.61 (16.05)

5.77 (2.53)

\(y = 3.01 + 1.76x_{1} - 2.12x_{2} + 0.20x_{1} x_{2} - 0.16x_{1}^{2} + 0.23x_{2}^{2}\)

20:1n − 9

0.84 (0.03)

0.60 (1.90)

0.77 (3.13)

0.65 (0.96)

0.48 (5.38)

0.72 (1.52)

0.69 (0.53)

0.39 (2.42)

0.69 (19.57)

0.74 (0.62)

\(y = 0.52 + 0.09x_{1} - 0.15x_{2} + 0.02x_{1} x_{2} - 0.01x_{1}^{2} + 0.01x_{2}^{2}\)

22:1n − 9

0.17 (0.06)

0.21 (29.88)

0.25 (3.24)

0.22 (1.21)

0.20 (3.21)

0.25 (1.41)

0.23 (0.60)

0.18 (0.25)

0.24 (2.12)

0.22 (0.20)

\(y = 0.19 + 0.02x_{1} - 0.003x_{2} + 0.001x_{1} x_{2} - 0.002x_{1}^{2} - 0.005x_{2}^{2}\)

16:1n − 7

3.28 (1.24)

3.09 (1.35)

5.82 (5.94)

3.75 (0.21)

1.86 (0.44)

5.02 (12.45)

3.47 (1.74)

0.27 (15.86)

3.84 (23.09)

2.81 (2.50)

\(y = 1.80 + 1.42x_{1} - 1.15x_{2} + 0.10x_{1} x_{2} - 0.12x_{1}^{2} - 0.19x_{2}^{2}\)

18:1n − 7

2.88 (0.06)

1.99 (0.54)

2.69 (0.00)

2.24 (0.98)

1.59 (2.45)

2.51 (0.14)

2.35 (1.94)

1.33 (1.26)

2.48 (1.46)

2.60 (0.89)

\(y = 1.66 + 0.35x_{1} - 0.53x_{2} + 0.06x_{1} x_{2} - 0.03x_{1}^{2} + 0.07x_{2}^{2}\)

Saturated fatty acids

The regression models for 14:0, 16:0, and 18:0 (Table 2) indicate that at 0 and 1 days, there were a continuous decrease in the concentrations of SFA between − 1.8 and 2.0 °C (Fig. 3). For example, the concentration of 16:0 decreases by 58% and ~ 49% at 0 and 1 day, respectively, while 18:0 was characterized by a continuous reduction in concentration when the temperature was raised from − 1.8 to 0.0 °C (~ 2 5%), 1.0 °C (~ 40%), and 2.0 °C (~ 50%) at 0 and 1 day (Fig. 3). These percentages of reduction are computed by means of the regression models of SFA in Table 2. The behavior at 5 days for 14:0 between − 1.8 and 0.0 °C was characterized by a decrease in concentration around 35% with an additional reduction of approximately 10% when the temperature was increased from 0.0 to 1.0 °C with no further change in concentration (0.77 ± 0.02 mg g‒1) between 1.0 and 2.0 °C. A similar reduction in concentration (~ 40%) was observed at 10 days between − 1.8 and 1.0 °C. At 5 and 10 days, the concentrations of 16:0 (5.01 ± 0.64 and 4.42 ± 0.37 mg g‒1, respectively) and 18:0 (1.16 ± 0.09 mg g‒1) remain almost constant in the whole range of temperatures as reflected in the contour plot in Fig. 3, where a monochromatic gray region between 5 and 10 days indicates a concentration range that varies between 4.0 and 6.0 mg g‒1 for 16:0 or between 1.0 and 1.5 mg g‒1 for 18:0. Total SFA contents in liver of fish exposed to higher temperatures for 1 day (5.26 ± 0.35, 3.82 ± 0.29, and 5.09 ± 0.36 mg g‒1 at 0.0, 1.0, and 2.0 °C, respectively) were lower than the seawater control group (7.57 ± 0.63 mg g‒1). Table 1 showed that between 5 (~ 6.7 ± 0.29 mg g‒1) and 10 days (6.23 ± 0.64 mg g‒1), the total SFA contents were similar to the control group.

Monounsaturated fatty acids

At 0, 1, and 5 days, there were significant reductions in the concentration of vaccenic acid (18:1n − 7), palmitic oleic acid (16:1n − 7), oleic acid (18:1n − 9), and gondoic acid (20:1n − 9) (Fig. 3). Table 1 and the regression models (Table 2 and Fig. 3) allow estimating a maximum reduction in the concentrations of 18:1n − 7, 16:1n − 7, 18:1n − 9, and 20:1n − 9 of approximately 60, 92, 80, and 60%, respectively, at 1 day and between − 1.8 and 2.0 °C. The behavior of 16:1n − 9 was characterized by a progressive reduction in concentration of approximately 50, 60, and 70% at 0 day and 40, 50, and 55% at 1 day when the temperature was increased at 0.0, 1.0, and 2.0 °C, respectively. Between 5 and 10 days, the concentration of the majority of MUFA remains almost constant in the whole range of temperatures as observed in Fig. 3, where monochromatic regions indicate similar concentrations. For example, a statistical comparison at the 95% confidence level of the lowest (0.34 ± 0.05 mg g−1) and highest (0.51 ± 0.06 mg g−1) recorded concentrations of 16:1n − 9 at 10 days (Table 1) revealed the lack of significant differences for this fatty acid at different temperatures (texperimental = 2.18 vs. ttabulated = 4.30). Total MUFAs showed a significant reduction with respect to the seawater control group after the first day of exposure to higher temperature (2.0 °C) (~15 ± 0.77 and 5.27 ± 0.16 mg g−1, respectively) (Table 1).

The results did not reveal significant changes in the concentration of 22:1n − 9 in the whole ranges of time and temperature (0.21 ± 0.03 mg g‒1) as reflected in Fig. 3 where the response surface displays basically one single color (gray). A statistical comparison of the lowest and highest concentrations of 22:1n − 9 in Table 1 (0.17 ± 0.01 and 0.28 ± 0.05 mg g−1, respectively) unveiled an experimental t value of 2.16 that was much lower than the tabulated 4.30 at 2 degrees of freedom. Total MUFAs showed a significant reduction with respect to the seawater control group after the first day of exposure to higher temperature (2.0 °C) (~ 15 ± 0.77 and 5.27 ± 0.16 mg g−1, respectively) (Table 1).

Polyunsaturated fatty acids

The elevated temperatures led to the observed variable responses of PUFA concentrations (Fig. 3). Linoleic acid (18:2n − 6) revealed a similar reduction in concentrations at 0 and 1 day in the whole range of tested temperatures. The concentration remains constant (0.49 ± 0.06 mg g‒1) between 0 and 5 days at − 1.8 °C with a further decreased in concentration of around 30% at 10 days. It was also observed that at 5 and 10 days, the concentration of 18:2n − 6 was independent of the temperature with average concentration values of 0.46 ± 0.07 and 0.32 ± 0.02 mg g‒1, respectively. The model for arachidonic acid (20:4n − 6) did not show significant changes in the concentration at different days and temperatures as reflected within day and within temperature averages. The former averages for 20:4n − 6 were 0.66 ± 0.05, 0.69 ± 0.05, 0.77 ± 0.05, and 0.73 ± 0.06 mg g‒1 at 0, 1, 5, and 10 days, while the latter were 0.67 ± 0.03, 0.76 ± 0.04, 0.74 ± 0.06, and 0.67 ± 0.07 mg g‒1 at − 1.8, 0.0, 1.0, and 2.0 °C, respectively. In addition, the concentration of linolenic acid (18:3n − 3) was independent of the temperature at 0 day (0.06 ± 0.01 mg g‒1 between − 1.8 and 0.0 °C), 1 day (0.07 ± 0.01 mg g‒1 between − 1.8 and 1.0 °C), 5 days (0.09 ± 0.01 mg g‒1 between 0.0 and 2.0 °C), and 10 days (0.06 ± 0.00 mg g‒1 between − 1.8 and 2.0 °C). Out of the specified temperature ranges for 0 and 1 day, there was a decrease in concentration while for 5 days, there was an increase in the concentration.

Eicosapentaenoic acid (20:5n − 3), docosapentaenoic acid (22:5n − 3), and docosahexaenoic acid (22:6n − 3) were characterized by a general increase in concentration at different temperatures and days. The concentration of 20:5n − 3 increased ~ 30% between 0 and 10 days when the temperature was raised from − 1.8 to 0.0 °C. Between 0.0 and 2.0 °C, the average concentrations at 0, 1, 5, and 10 days were 3.14 ± 0.15, 3.58 ± 0.15, 4.60 ± 0.14, and 4.12 ± 0.13 mg g‒1, respectively. Increases in the concentration of 22:5n − 3 of 18% (between 0 and 1 day), 54% (between 1 and 5 days), and 33% (between 5 and 10 days) were observed at − 1.8 °C. Identical increase patterns in concentration of 6% (0–1 day), 20% (1–5 days), and 15% (5–10 days) were observed at 0.0 and 1.0 °C. At 2.0 °C, the increase pattern was 9% (0–1 day), 29% (1–5 days), and 19% (5–10 days). The response surface for 22:6n − 3 displayed a behavior similar to 20:5n − 3 (Fig. 3). There was an average increase in concentration of 38% in the whole range of time (0–10 days) between − 1.8 and 0.0 °C. Between 0.0 and 2.0 °C, the average concentrations at 0, 1, 5, and 10 days were 4.56 ± 0.26, 5.41 ± 0.26, 7.32 ± 0.28, and 6.41 ± 0.30 mg g‒1, respectively. Total PUFA showed a statistically significant increase after 1 day of exposure for all tested temperatures that persisted until 5 days of exposure to higher temperatures (Table 1).

Discussion

The ubiquitous species of stenothermal are characterized by their evolutionary adaptation to the extremely low temperatures of the Antarctic region, which is associated with a unique physiological adaptation but as well involves limited compensation capacities to the environmental changes. These species cope with the cold environment by several types of adaptations including, a reduction or loss of hemoglobin, higher mitochondrial densities, expression of anti-freeze glycoproteins, lack of heat shock response, variation in lipid class or changes in the FA composition (Eastman 1993; Pörtner and Playle 1998). The present research aims to answer a crucial question of ocean warming and its consequences on ecological interactions, by measuring changes in fatty acid profiles over time with variation in body temperature of Antarctic fish T. bernacchii.

The models for 14:0, 16:0, and 18:0 revealed that at 10 days and between 0.0 and 2.0 °C, there were non-significant increases of approximately 4, 5, and 6% in absolute concentrations, respectively (Fig. 3). These increases in concentrations are equivalent to 14, 16, and 16% in relative units (computed from Online Resource 8), respectively. These results are close to those reported by Brodte et al. (2008) who measured the fatty acid profiles in liver of Antarctic fish (Pachycara brachycephalum) and reported non-significant increases in concentrations of 20, 16, and 25% for 14:0, 16:0, and 18:0 (measured in relative units), respectively, at 120 days of acclimation and between 0.0 and 2.0 °C. Similarly, non-significant increments of 15 and 20% in relative concentration have been observed in temperate fish by Roche and Pérès (1984) and Hazel (1979), respectively. In the present study, the content of SFA in the liver of the control group at − 1.87 °C was 7.46 ± 0.43 mg g‒1 corresponding to 24.0% of total FA. The obtained value for hepatic SFA was comparable to those observed in T. bernacchii in other studies (Corsolini and Borghesi 2017; Malekar et al. 2018). Upon higher temperatures, the relative concentration of SFA remained almost constant in the liver of T. bernacchii, over the studied acclimation period (1–10 days). Similarly, Brodte et al. (2008) found that the content of SFA in the liver of Antarctic fish did not change at higher temperatures (2, 4, and 6 °C) when compared to the control group (− 0 °C). Recently, Malekar et al. (2018) demonstrated also that the SFA composition of phospholipids in the liver of T. bernacchii remained constant at elevated temperatures during 14 days. The same authors demonstrated that increasing the temperature to 6 °C resulted in an increase of saturated fatty acids and a decrease in the content of unsaturated fatty acids in the membrane of T. bernacchii liver. Based on the present and previous studies, the variation of SFA composition in response to changes in temperatures seems to occur differently based on the temperature of exposure (Brodte et al. 2008; Malekar et al. 2018).

The observed decrease in concentration for 16:1n − 7 between 0 and 5 days as the temperature increases has been also reported in different polar and temperate fish species by evaluating the effect of increasing the temperature at specific times (Roche and Pérès 1984; Dey et al. 1993; Brodte et al. 2008). The reduction in concentration for 18:1n − 7 and 18:1n − 9 between 0 and 5 days in the investigated temperature range is in agreement with the results of Dey et al. (1993) and Roche and Pérès (1984), who reported a decreasing in concentration of 18:1n (not distinction was made between the two isomers) in temperate fish after comparing two experimental temperatures. Brodte et al. (2008) reported an astonishing increase in concentration of ~ 3000% for 18:1n − 7 in the liver of Antarctic fish (difference between fish exposed to 0.0 and 4.0 °C during 120 days) and a decrease of 39% in temperate fish (difference between fish exposed to 4.0 °C and 6.0 °C during 120 days). The behavior of 16:1n in livers of different temperate fish has been analyzed after one month of acclimation period at two different temperatures (Roche and Pérès 1984; Dey et al. 1993). The results of these studies did not reveal differences in FA concentrations in seawater fish (Roche and Pérès 1984) but in freshwater fish (Dey et al. 1993). Similar results to those reported for seawater fish have been predicted by the model for 16:1n − 9 (Fig. 3) where the concentration does not change significantly between 5 and 10 days in the range of − 1.8 and 2.0 °C. The predicted constant concentration at 10 days and between 0.0 and 2.0 °C by the 20:1n − 9 model (Fig. 3) for Antarctic fish T. bernacchii are in full agreement with the studies on Antarctic fish Pachycara brachycephalum which did not observe change in concentration after 120 days in the range of 0.0 and 2.0 °C (Brodte et al. 2008).

The contour plots for the six PUFA (18:2n − 6, 20:4n − 6, 18:3n − 3, 20:5n − 3, 22:5n − 3, 22:6n − 3) predict non-significant changes in concentration at 1, 5, and 10 days between 0.0 and 2.0 °C. The predictions are consistent with the report of Brodte et al. (2008) who determined these PUFAs in liver from Antarctic fish without any significant change in concentration in the same range of temperatures (0.0 and 2.0 °C) after 120 days. Studies on temperate fish have not observed change in the concentration of 18:2n − 6, 20:4n − 6, 18:3n − 3, 20:5n − 3, 22:5n − 3, and 22:6n − 3 (Hazel 1979; Roche and Pérès 1984; Dey et al. 1993; Brodte et al. 2008).

In their natural environment, marine organisms adapt their physiological function to cope with different stress factors. There is an inverse correlation between ambient temperature and the content of unsaturated fatty acids (Roche and Pérès 1984; Hazel 1990, 1995). Indeed, the cell membranes of fish adapted to live in cold temperatures contained generally high levels of MUFA, in particular 18:1n − 9, which reflect the ability to maintain a proper fluidity of the biological membrane (Sinensky 1974; Hazel and Carpenter 1985; Tiku et al. 1996). The biochemical response of stenothermal species to lower temperature is an increase in the concentration of unsaturated fatty acids in both the membrane and depot of lipids (Brodte et al. 2008; Malekar et al. 2018; Truzzi et al. 2018a, b). In the present study, when fish were immediately exposed to higher temperature (2.0 °C after 1 day of exposure), the concentration of MUFA decreased in the liver of T. bernacchii, specifically the concentration of oleic acid (18:1n − 9) decreased by 50% compared to the control group (− 1.8 and 2.0 °C after 1 day of exposure). Such effect has been observed previously in many fish species including Pachycara brachycephalum, Zoarces viviparus (Brodte et al. 2008) and Salmo gairdneri (Hazel 1979; Sellner and Hazel 1982). However, those results were observed at different temperatures and exposure times (from 3 days to 3 months) and in organs (liver, muscle, gill, and kidney) of different fish species (Smith and Kemp 1971; Sellner and Hazel 1982; Hazel and Carpenter 1985), whereas for T. bernacchii, the change in hepatic FA was observed immediately after 1 day of exposure to higher temperatures. The decrease in MUFA composition in the liver of T. bernacchii was in line with the inverse relation between unsaturated fatty acids and changes in temperature, as it has been demonstrated in different fish species (Hazel 1979; Sellner and Hazel 1982; Logue et al. 2000). Furthermore, this decrease in MUFA composition might also serve as indicator for protection mechanisms against oxidative stress induced by higher temperature, as previous studies have shown (Crockett 2008; Vinagre et al. 2014; Machado et al. 2014). Conversely, the content of PUFA in the liver of T. bernacchii increased after the first day of exposure to higher temperatures, from 23% of total FA (− 1.87 °C) to 36, 43, and 40% of total FA at 0.0, 1.0, and 2.0 °C, respectively. These results are somewhat surprising, since most of the studies reported a decrease in PUFA composition at higher temperatures in the tissues of different fish species. (Hazel 1979, 1990; Tiku et al. 1996; Pernet et al. 2007). However, data provided by Brodte et al. (2008) demonstrated also an increase in PUFA levels by more than 70% in the liver of Antarctic fish upon higher temperature (from 0.0 to 4.0 °C after 120 days of exposure). In this study, the authors concluded that the polar fish have already optimized their metabolism by using preferentially MUFA over the PUFA, when the temperature changes (Brodte et al. 2008). Moreover, in the present study, the effect of increasing the temperature on the FA profile of the liver was investigated after 1, 5, and 10 days, however, most of the studies reported a decrease in the PUFA composition at higher temperatures after a long period of exposure (Smith and Kemp 1971; Sellner and Hazel 1982; Hazel and Carpenter 1985). Thus, in the present study, the effect seen on the PUFA composition might be not related to changes in temperature, and also a long period of exposure to higher temperatures should be considered.

The statistical comparison (p < 0.05) between the seawater control (0 day, − 1.87 °C) and 1, 5, and 10 days at 0.0, 1.0, and 2.0 °C allows concluding that at 10 days of exposure to 0.0, 1.0, and 2.0 °C, the FA composition in the liver of T. bernacchii were similar to the control group as indicated in Table 1. Indeed, FA composition in the liver of Antarctic fish contained the same concentration as in their natural environment (− 1.87 °C). T. bernacchii evolved during the Palaeocene period, characterized by high fluctuations in temperatures, and only after this period, the temperatures in the Antarctic have decreased until the current stable temperatures (Clarke and Johnston 1996; Seebacher et al. 2005). The plasticity of these species seems to be comparable in magnitude with eurythermal and temperate fishes (Seebacher et al. 2005; Podrabsky and Somero 2006; Bilyk and DeVries 2011; Bilyk et al. 2012). Furthermore, previous research on warm acclimation of notothenioid fish demonstrated that these species have conserved the capacity to compensate for temperatures changes, by different mechanisms including astounding plasticity in cardiovascular response, metabolic control, Na+/K+-ATPase activity, hypo-osmoregulation, induction of the expression of many genes associated with lipid biosynthesis, or increase in their critical thermal maximum (Bilyk et al. 2012; Brauer et al. 2005; Seebacher et al. 2005; Huth and Place 2016). The present results may indicate that T. bernacchii exposed to thermal shock tries to acclimate to the elevated temperatures in order to maintain the functionally to the liver. Furthermore, there were no detrimental health effects on these fish species when exposed to higher temperatures up to 4.0 °C during several weeks (Bilyk and DeVries 2011). Thus, it can be hypothesized that the changes in fatty acids composition observed in the liver of T. bernacchii might reflect a shift towards a normal acclimation response, rather than symptoms of heat stress.

Conclusions

The fact that the computed models for Antarctic T. bernacchii predict changes that are in agreement with well-documented studies not only on polar but also on temperate fish might indicate that polar and temperate teleosts will respond in a similar way to global warming, which in turn might indicate the existence of a common underlying model for teleosts.

To the best of our knowledge, this is the first report modeling the changes in fatty acid concentration as a function of systematic and simultaneous variations in time and temperature.

Notes

Acknowledgements

Financial support from the Italian Programma Nazionale di Ricerche in Antartide, projects “Chemical Contamination” and “Behavior of microcomponents in the Antarctic Continent in relation to climate change” is gratefully acknowledged. Many thanks are due to the technical personnel of ENEA (Ente Nazionale Energia e Ambiente) at Terra Nova Bay and to the scientists of the Antarctic expedition for the sampling activities on site. MA is grateful for financial support from the EU programme Erasmus+Traineeships for education, training, youth, and sport. This study was also supported by the Norwegian Research Council, project AquaFly, Grant Number 238997.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical approval

All procedures performed in studies involving Trematomus bernacchii were in accordance with the ethical standards of the institution or practice at which the studies were conducted.

Supplementary material

300_2019_2577_MOESM1_ESM.docx (13 kb)
Supplementary material 1 (DOCX 12 kb)
300_2019_2577_MOESM2_ESM.pptx (99 kb)
Supplementary material 2 (PPTX 99 kb)
300_2019_2577_MOESM3_ESM.xls (712 kb)
Supplementary material 3 (XLS 711 kb)
300_2019_2577_MOESM4_ESM.docx (18 kb)
Supplementary material 4 (DOCX 18 kb)
300_2019_2577_MOESM5_ESM.docx (89 kb)
Supplementary material 5 (DOCX 89 kb)
300_2019_2577_MOESM6_ESM.docx (39 kb)
Supplementary material 6 (DOCX 39 kb)
300_2019_2577_MOESM7_ESM.docx (159 kb)
Supplementary material 7 (DOCX 159 kb)
300_2019_2577_MOESM8_ESM.docx (15 kb)
Supplementary material 8 (DOCX 14 kb)

References

  1. Alfaro AC, Thomas F, Sergent L, Duxbury M (2006) Identification of trophic interactions within an estuarine food web (northern New Zealand) using fatty acid biomarkers and stable isotopes. Estuar Coast Shelf Sci 70:271–286.  https://doi.org/10.1016/j.ecss.2006.06.017 CrossRefGoogle Scholar
  2. Antarctic and Southern Ocean Coalition (ASOC) (2009) The case for special protection of the Ross Sea. Antarctic and Southern Ocean Coalition. https://www.asoc.org/storage/documents/Meetings/CCAMLR/XXVIII/The_Case_for_Special_Protection_of_the_Ross_Sea_2009.pdf. Accessed 23 Aug 2017
  3. Araujo P (2009) Key aspects of analytical method validation and linearity evaluation. J Chromatogr B 877:2224–2234.  https://doi.org/10.1016/j.jchromb.2008.09.030 CrossRefGoogle Scholar
  4. Bilyk KT, DeVries AL (2011) Heat tolerance and its plasticity in Antarctic fishes. Comp Biochem Physiol A 158:382–390.  https://doi.org/10.1016/j.cbpa.2010.12.010 CrossRefGoogle Scholar
  5. Bilyk KT, Evans CW, DeVries AL (2012) Heat hardening in Antarctic notothenioid fishes. Polar Biol 35:1447–1451.  https://doi.org/10.1007/s00300-012-1189-0 CrossRefGoogle Scholar
  6. Brauer PR, Sanmann JN, Petzel DH (2005) Effects of warm acclimation on Na+, K+-ATPase α-subunit expression in chloride cells of Antarctic fish. Anat Rec A 285A:600–609.  https://doi.org/10.1002/ar.a.20203 CrossRefGoogle Scholar
  7. Brodte E, Graeve M, Jacob U, Knust R, Pörtner HO (2008) Temperature-dependent lipid levels and components in polar and temperate eelpout (Zoarcidae). Fish Physiol Biochem 34:261–274.  https://doi.org/10.1007/s10695-007-9185-y CrossRefGoogle Scholar
  8. Buda C, Dey I, Farkas T (1996) Role of membrane lipids in temperature acclimatization of carp. Arch für Tierernaehrung 49:61–62.  https://doi.org/10.1080/17450399609381864 CrossRefGoogle Scholar
  9. Cheung WWL, Lam VWY, Pauly D (2008) Modelling present and climate-shifted distribution of marine fishes and invertebrates. Vancouver, Canada. Fisheries Centre, University of British Columbia. https://open.library.ubc.ca/cIRcle/collections/facultyresearchandpublications/52383/items/1.0074754. Accessed 24 Aug 2017
  10. Clarke A, Johnston IA (1996) Evolution and adaptive radiation of Antarctic fishes. Trends Ecol Evol 11:212–218.  https://doi.org/10.1016/S0169-5347(00)01896-6 CrossRefGoogle Scholar
  11. Copeman LA, Laurel BJ, Parrish CC (2013) Effect of temperature and tissue type on fatty acid signatures of two species of North Pacific juvenile gadids: a laboratory feeding study. J Exp Mar Biol Ecol 448:188–196.  https://doi.org/10.1016/j.jembe.2013.07.008 CrossRefGoogle Scholar
  12. Corsolini S, Borghesi N (2017) A comparative assessment of fatty acids in Antarctic organisms from the Ross Sea: occurrence and distribution. Chemosphere 174:747–753.  https://doi.org/10.1016/j.chemosphere.2017.02.031 CrossRefGoogle Scholar
  13. Crockett EL (2008) The cold but not hard fats in ectotherms: consequences of lipid restructuring on susceptibility of biological membranes to peroxidation, a review. J Comp Physiol B 178:795–809.  https://doi.org/10.1007/s00360-008-0275-7 CrossRefGoogle Scholar
  14. Dalsgaard J, StJohn M, Kattner G, Müller-Navarra D, Hagen W (2003) Fatty acid trophic markers in the pelagic marine environment. Adv Mar Biol 46:225–340.  https://doi.org/10.1016/S0065-2881(03)46005-7 CrossRefGoogle Scholar
  15. Daly AJ, Baetens JM, De Baets B (2018) Ecological diversity: measuring the unmeasurable. Mathematics.  https://doi.org/10.3390/math6070119 Google Scholar
  16. Dey I, Buda C, Wiik T, Halver JE, Farkas T (1993) Molecular and structural composition of phospholipid membranes in livers of marine and freshwater fish in relation to temperature. Proc Natl Acad Sci 90:7498–7502.  https://doi.org/10.1073/pnas.90.16.7498 CrossRefGoogle Scholar
  17. Di Bello D, Vaccaro E, Longo V, Regoli F, Nigro M, Benedetti M, Gervasi PG, Pretti C (2007) Presence and inducibility by β-naphthoflavone of CYP1A1, CYP1B1 and phase II enzymes in Trematomus bernacchii, an Antarctic fish. Aquat Toxicol 84:19–26.  https://doi.org/10.1016/j.aquatox.2007.05.010 CrossRefGoogle Scholar
  18. Eastman JT (1993) Antarctic fish biology: evolution in a unique environment. Academic Press, New YorkGoogle Scholar
  19. FDA Guidance for Industry Bioanalytical Method Validation (2013) US Food and Drug Administration, Rockville. https://www.fda.gov/media/70858/download. Accessed 30 Apr 2018
  20. Firestone D, Horowitz W (1979) IUPAC gas chromatographic method for determination of fatty acid composition. J Ass Off Anal Chem 62:709–721Google Scholar
  21. Gregorius HR (1991) On the concept of effective number. Theor Popul Biol 40:269–283.  https://doi.org/10.1016/0040-5809(91)90056-L CrossRefGoogle Scholar
  22. Gregorius HR (2016) Effective numbers in the partitioning of biological diversity. J Theor Biol 409:133–147.  https://doi.org/10.1016/j.jtbi.2016.08.037 CrossRefGoogle Scholar
  23. Gregorius HR, Gillet EM (2015) Classifying measures of biological variation. PLoS ONE.  https://doi.org/10.1371/journal.pone.0115312 Google Scholar
  24. Hazel JR (1979) Influence of thermal acclimation on membrane lipid composition of rainbow trout liver. Am J Physiol 236:R91LP–R101.  https://doi.org/10.1152/ajpregu.1979.236.1.R91 Google Scholar
  25. Hazel JR (1990) The role of alterations in membrane lipid composition in enabling physiological adaptation of organisms to their physical environment. Prog Lipid Res 29:167–227.  https://doi.org/10.1016/0163-7827(90)90002-3 CrossRefGoogle Scholar
  26. Hazel JR (1995) Thermal adaptation in biological membranes: is homeoviscous adaptation the explanation? Annu Rev Physiol 57:19–42.  https://doi.org/10.1146/annurev.ph.57.030195.000315 CrossRefGoogle Scholar
  27. Hazel JR, Carpenter R (1985) Rapid changes in the phospholipid composition of gill membranes during thermal acclimation of the rainbow trout, Salmo gairdneri. J Comp Physiol B 155:597–602.  https://doi.org/10.1007/BF00694450 CrossRefGoogle Scholar
  28. Huth TJ, Place SP (2016) Transcriptome wide analyses reveal a sustained cellular stress response in the gill tissue of Trematomus bernacchii after acclimation to multiple stressors. BMC Genomics 17:127.  https://doi.org/10.1186/s12864-016-2454-3 CrossRefGoogle Scholar
  29. Illuminati S, Truzzi C, Annibaldi A, Migliarini B, Carnevali O, Scarponi G (2010) Cadmium bioaccumulation and metallothionein induction in the liver of the Antarctic teleost Trematomus bernacchii during an on − site short-term exposure to the metal via seawater. Toxicol Environ Chem 92:617–640.  https://doi.org/10.1080/02772240902902349 CrossRefGoogle Scholar
  30. Intergovernmental Panel for Global Climate Change (IPCC) (2000) The applicability of regional climate models at the scale of small island states. Intergovernmental Panel for Global Climate Change. https://www.ipcc.ch/graphics/speeches/statement-ipcc-june-2000.pdf. Accessed 23 Aug 2017
  31. Jost L (2010) The relation between evenness and diversity. Diversity 2:207–232.  https://doi.org/10.3390/d2020207 CrossRefGoogle Scholar
  32. Kennicutt MC, Chown SL, Cassano JJ et al (2015) A roadmap for Antarctic and Southern Ocean science for the next two decades and beyond. Antarct Sci 27:3–18.  https://doi.org/10.1017/S0954102014000674 CrossRefGoogle Scholar
  33. Kirkwood BR, Sterne JAC (2003) Essential medical statistics. Blackwell, LondonGoogle Scholar
  34. Könneke M, Widdel F (2003) Effect of growth temperature on cellular fatty acids in sulphate-reducing bacteria. Environ Microbiol 5:1064–1070.  https://doi.org/10.1046/j.1462-2920.2003.00499.x CrossRefGoogle Scholar
  35. Lahdes E, Balogh G, Fodor E, Farkas T (2000) Adaptation of composition and biophysical properties of phospholipids to temperature by the crustacean, Gammarus spp. Lipids 35:1093–1098.  https://doi.org/10.1007/s11745-000-0624-9 CrossRefGoogle Scholar
  36. Logue JA, de Vries AL, Fodor E, Cossins AR (2000) Lipid compositional correlates of temperature-adaptive interspecific differences in membrane physical structure. J Exp Biol 203:2105–2115Google Scholar
  37. Machado C, Zaleski T, Rodrigues E, Carvalho CS, Cadena SMSC, Gozzi GJ, Krebsbach P, Rios FS (2014) Effect of temperature acclimation on the liver antioxidant defence system of the Antarctic nototheniids Notothenia coriiceps and Notothenia rossii. Comp Biochem Physiol B 172–173:21–28.  https://doi.org/10.1016/j.cbpb.2014.02.003 CrossRefGoogle Scholar
  38. Malekar VC, Morton JD, Hider RN, Cruickshank RH, Hodge S, Metcalf VJ (2018) Effect of elevated temperature on membrane lipid saturation in Antarctic notothenioid fish. PeerJ 6:e4765.  https://doi.org/10.7717/peerj.4765 CrossRefGoogle Scholar
  39. Mourente G, Quintero O, Cañavate JP (2015) Trophic links of Atlantic Bluefin tuna (Thunnus thynnus L.) inferred by fatty acid signatures. J Exp Mar Biol Ecol 463:49–56.  https://doi.org/10.1016/j.jembe.2014.11.002 CrossRefGoogle Scholar
  40. O’Connor MI, Bruno JF, Gaines SD, Halpern BS, Lester SE, Kinlan BP, Weiss JM (2007) Temperature control of larval dispersal and the implications for marine ecology, evolution, and conservation. Proc Natl Acad Sci USA 104:1266–1271.  https://doi.org/10.1073/pnas.0603422104 CrossRefGoogle Scholar
  41. Parmesan C, Root TL, Willig MR (2000) Impacts of extreme weather and climate on terrestrial biota. Bull Am Meteorol Soc 81:443–450CrossRefGoogle Scholar
  42. Pernet F, Gauthier-Clerc S, Mayrand É (2007) Change in lipid composition in eastern oyster (Crassostrea virginica Gmelin) exposed to constant or fluctuating temperature regimes. Comp Biochem Physiol B 147:557–565.  https://doi.org/10.1016/j.cbpb.2007.03.009 CrossRefGoogle Scholar
  43. Podrabsky JE, Somero GN (2006) Inducible heat tolerance in Antarctic notothenioid fishes. Polar Biol 30:39–43.  https://doi.org/10.1007/s00300-006-0157-y CrossRefGoogle Scholar
  44. Pörtner H (2001) Climate change and temperature-dependent biogeography: oxygen limitation of thermal tolerance in animals. Naturwissenschaften 88:137–146.  https://doi.org/10.1007/s001140100216 CrossRefGoogle Scholar
  45. Pörtner HO, Playle RC (1998) Cold ocean physiology. Cambridge University Press, CambridgeCrossRefGoogle Scholar
  46. Ramesha CS, Thompson GA (1982) Changes in the lipid composition and physical properties of Tetrahymena ciliary membranes following low-temperature acclimation. Biochemistry 21:3612–3617.  https://doi.org/10.1021/bi00258a013 CrossRefGoogle Scholar
  47. Regoli F, Nigro M, Benedetti M, Gorbi S, Pretti C, Gervasi PG, Fattorini D (2005) Interactions between metabolism of trace metals and xenobiotic agonists of the aryl hydrocarbon receptor in the antarctic fish Trematomus bernacchii: environmental perspectives. Environ Toxicol Chem 24:1475–1482.  https://doi.org/10.1897/04-514R.1 CrossRefGoogle Scholar
  48. Roche H, Pérès G (1984) Influence of acclimatization to different temperatures and of the seasonal factor on the lipid composition of liver, muscle and intestinal tissues of the sea dace (Dicentrarchus labrax, Pisces). Comp Biochem Physiol B 78:755–759.  https://doi.org/10.1016/0305-0491(84)90130-5 CrossRefGoogle Scholar
  49. Roessig JM, Woodley CM, Cech JJ, Hansen LJ (2004) Effects of global climate change on marine and estuarine fishes and fisheries. Rev Fish Biol Fish 14:251–275.  https://doi.org/10.1007/s11160-004-6749-0 CrossRefGoogle Scholar
  50. SCAR Code of Conduct for the Use of Animals for Scientific Purposes in Antarctica (2011), Buenos Aires, https://www.scar.org/scar-library/search/policy/codes-of-conduct/3408-code-of-conduct-for-the-use-of-animals-for-scientific-purposes-in-antarctica/file/. Accessed 29 Apr 2018
  51. Sajjadi N, Eghtesadi-Araghi P (2011) Determination of fatty acid compositions as biomarkers in the diet of turbo coronatus in Chabahar Bay. J Persian Gulf 2:35–42Google Scholar
  52. Seebacher F, Davison W, Lowe CJ, Franklin CE (2005) A falsification of the thermal specialization paradigm: compensation for elevated temperatures in Antarctic fishes. Biol Lett 1:151–154.  https://doi.org/10.1098/rsbl.2004.0280 CrossRefGoogle Scholar
  53. Sellner PA, Hazel JR (1982) Time course of changes in fatty acid composition of gills and liver from rainbow trout (Salmo gairdneri) during thermal acclimation. J Exp Zool 221:159–168.  https://doi.org/10.1002/jez.1402210206 CrossRefGoogle Scholar
  54. Sinensky M (1974) Homeoviscous adaptation—a homeostatic process that regulates the viscosity of membrane lipids in Escherichia coli. Proc Natl Acad Sci USA 71:522–525CrossRefGoogle Scholar
  55. Smith MW, Kemp P (1971) Parallel temperature-induced changes in membrane fatty acids and in the transport of amino acids by the intestine of goldfish (Carassius auratus L.). Comp Biochem Physiol B 39:357–365.  https://doi.org/10.1016/0305-0491(71)90180-5 CrossRefGoogle Scholar
  56. Tiku PE, Gracey AY, Macartney AI, Beynon RJ, Cossins AR (1996) Cold-induced expression of Δ9-desaturase in carp by transcriptional and posttranslational mechanisms. Science 271:815–818.  https://doi.org/10.1126/science.271.5250.815 CrossRefGoogle Scholar
  57. Timmermann R, Hellmer HH (2013) Southern Ocean warming and increased ice shelf basal melting in the twenty-first and twenty-second centuries based on coupled ice-ocean finite-element modelling. Ocean Dyn 63:1011–1026CrossRefGoogle Scholar
  58. Torstensen BE, Frøyland L, Lie Ø (2004) Replacing dietary fish oil with increasing levels of rapeseed oil and olive oil—effects on Atlantic salmon (Salmo salar L.) tissue and lipoprotein lipid composition and lipogenic enzyme activities. Aquac Nutr 10:175–192.  https://doi.org/10.1111/j.1365-2095.2004.00289.x CrossRefGoogle Scholar
  59. Truzzi C, Illuminati S, Annibaldi A, Antonucci M, Scarponi G (2017) Quantification of fatty acids in the muscle of Antarctic fish Trematomus bernacchii by gas chromatography-mass spectrometry: optimization of the analytical methodology. Chemosphere 173:116–123.  https://doi.org/10.1016/j.chemosphere.2016.12.140 CrossRefGoogle Scholar
  60. Truzzi C, Annibaldi A, Antonucci M, Scarponi G, Illuminati S (2018a) Gas chromatography-mass spectrometry analysis on effects of thermal shock on the fatty acid composition of the gills of the Antarctic teleost Trematomus bernacchii. Environ Chem.  https://doi.org/10.1071/EN18130 Google Scholar
  61. Truzzi C, Illuminati S, Antonucci M, Scarponi G, Annibaldi A (2018b) Heat shock influences the fatty acid composition of the muscle of the Antarctic fish Trematomus bernacchii. Mar Environ Res 139:122–128.  https://doi.org/10.1016/J.MARENVRES.2018.03.017 CrossRefGoogle Scholar
  62. Vinagre C, Madeira D, Mendonça V, Dias M, Roma J, Diniz MS (2014) Effect of increasing temperature in the differential activity of oxidative stress biomarkers in various tissues of the Rock goby, Gobius paganellus. Mar Environ Res 97:10–14.  https://doi.org/10.1016/j.marenvres.2014.01.007 CrossRefGoogle Scholar

Copyright information

© The Author(s) 2019

Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Authors and Affiliations

  1. 1.Institute of Marine Research (IMR)BergenNorway
  2. 2.Department of Life and Environmental SciencesUniversità Politecnica Delle MarcheAnconaItaly

Personalised recommendations