Effect of Long-Term Day/Night Temperature Oscillations on the Overall Performance of Gilthead Seabream (Sparus aurata) Juveniles

Abstract

Water temperature variations affect fish growth and health, often leading to huge losses in fish production, especially during the cold season. To alleviate this constraint, fish farmers can use a water heating system driven by solar energy during daytime. This action will cause a water temperature drop during the night period, making it important to understand the physiological response of fish exposed to the resulting day/night temperature oscillations. To investigate this scenario, gilthead seabream juveniles (96.3 ± 1.0 g) were exposed to different thermal regimes for 67 days: Tconstant and Tdaily cycles. The latter group was exposed to daily water temperature oscillations between ~19 and 13 °C compared with a constant temperature of ~19 °C for the other experimental group. Temperature fluctuations compromised fish growth efficiency and reduced the proportion of fatty acids in several tissues, with implications for the whole proximate composition. Moreover, temperature oscillations influenced several blood parameters. These results favor the usage of a constant water temperature of ~19 °C for optimal gilthead seabream juvenile production instead of a day/night water temperature oscillating regime. Nevertheless, the type of energy used to warm the water will depend on the operational conditions and/or business strategy of fish farmers.

1. Introduction

Water temperature is critical for ectothermic animals, such as fish, because they cannot regulate their body temperature. Consequently, all physiological processes are affected by this parameter. Numerous studies have shown that temperature has a huge impact on fish growth rate and food consumption, especially for farmed fish which cannot escape from unfavorable environmental conditions, such as low winter temperatures [1,2,3].

Gilthead seabream (Sparus aurata) is one of the most valuable farmed fish species in the Mediterranean Sea. Wild gilthead seabream is exposed to wide fluctuations in water temperature (11 to 30 °C), presenting a broad range of thermal tolerance [4]. Growth is compromised when the water temperature drops below 13 °C, which provokes fasting, growth arrest, general metabolic depression, and many other physiological alterations [5]. Winter syndrome is a pathology that may come from this complex biological response, often leading to acute mortality episodes and thus, considerable economic losses [6]. To overcome these issues in fish farms, rearing water temperature should be controlled, according to each species’ optimal requirements, throughout the overall fish production period, independently of temperature weather variations. However, this scenario is economically inviable for fish farmers since it involves huge electricity and fuel costs, mainly in winter. Photovoltaic technology may provide a sustainable and affordable energetic solution to overcome this obstacle by warming the rearing water only during daytime, causing a water temperature drop during the night period [7]. To test this hypothesis, we evaluated the effect of day/night water temperature fluctuations on the overall performance of gilthead seabream. In this trial, a group of fish was maintained in warm water during daytime to simulate water warming through solar energy and allow the normal feeding routine of the fish. Overnight, to simulate the natural temperature drop associated with the night period without the use of any water heating source, the water temperature was cooled down to ~13 °C, which corresponds to the average minimum temperature occurring over the year in the western Mediterranean Sea [8]. Although several studies have tested the physiological response of gilthead seabream exposed to water temperature oscillations [2,3,9,10], to our knowledge, this is the first study in which gilthead seabream is exposed to repeated day/night water temperature fluctuations where the cold period corresponds to the night period in which fish do not feed and overall fish metabolism is decreased. The decreased metabolism associated with cold temperatures was showed by [11].

Temperature variations can invoke biochemical and physiological changes at the organismal and cellular levels. At the organismal level, these changes are mediated by the neuroendocrine system and are generally characterized by changes in plasma metabolites, hematological features, and the concentration of circulating stress hormones, such as cortisol [12,13]. At the cellular level, thermal stress can induce the expression of heat shock proteins (Hsps) such as Hsp70, which is an abundant and well conserved protein [13,14]. When cells are exposed to a stressor, a rapid increase in Hsp70 levels occurs to protect the cells from the harmful effects of the stressor, such as temperature variations. Due to the greater need for molecular chaperones under stressful conditions (as the rate of cellular protein damage or refolding increases), the expression levels of Hsp70 in fish in both experimental groups were evaluated in the gills, a key tissue that is in direct contact with water temperature, and the liver, a central tissue in overall fish metabolism.

As metabolic functions are affected by temperature, the digestion process of fish can also be disturbed, partly due to alterations in the activity of digestive enzymes [15]. Since fish growth and health are dependent on the proper functioning of these enzymes and the consequent absorption of nutrients in the intestine, digestive enzymes constitute an important indicator to address the effect of long-term temperature fluctuations in fish. Similarly, the digestibility of lipids can be affected by temperature variations, resulting in changes in fatty acid composition [16] with possible effects on the fluidity of the cell membranes of fish. A reduction in temperature results in the accumulation of unsaturated fatty acids in the membranes during acclimation to low temperatures with concomitant decrease of membrane fluidity [17,18]. All these temperature-related effects have an impact on the quality of the final product, such as whole-body composition or flesh texture, as they reflect the metabolic changes that are occurring in the fish body [15,19,20].

Better control of fish rearing conditions such as temperature will lead to the improvement of aquaculture practices and the consequent optimization of production and associated costs. By precisely monitoring temperature, systems can be adjusted to ensure optimal growth conditions of aquatic organisms for various locations, be they warm or cold. Since daily temperature variations can induce changes in the physiological status of fish, we analyzed the effects of day/night temperature fluctuations on gilthead seabream juveniles’ overall performance by monitoring growth and feed utilization parameters, blood and plasma biochemistry, digestive enzyme activity, cellular stress, tissue structure, and fatty acid composition.

2. Materials and Methods

2.1. Fish and Experimental Conditions

The fish trial was performed at the Aquaculture Research Station (EPPO) of the Portuguese Institute of Sea and Atmosphere (IPMA) (Olhão, Portugal), certified by the Direção Geral da Alimentação e Veterinária to execute animal experimentation under the authorization, 2018/12/17-025516. The experiment was directed by trained scientists (following category C FELASA recommendations) and conducted according to the European guidelines on the protection of animals used for scientific purposes (Directive 2010/63/UE of the European Parliament and the European Union Council) and the related Portuguese legislation guidelines (Decreto-Lei 113/2013) on animal experimentation and welfare.

Fish juveniles used in this trial (8 months old) derived from a single spawn of gilthead seabream broodstock maintained at the IPMA’s Aquaculture Research Station (EPPO). A total of 720 gilthead seabream with a mean initial body weight of 96.3 ± 1.0 g were distributed in six 1.5 m³ circular fiberglass tanks, resulting in a density of 7.7 ± 0.08 kg m⁻³. Three tanks were subjected to water temperature variations, between ~13 °C (night) and ~19 °C (day), the Tdaily cycles group, and three tanks were kept at a constant temperature during the whole experimental period (~19 °C) for 67 days, the Tconstant group. The fish were fed three times a day (10.30 am, 2 pm and 4 pm) by hand to apparent satiety. Efforts were made to limit overfeeding, and a continuous record of feed intake was maintained during the entire trial. Rearing tanks were supplied with flow-through gravel-filtered, aerated seawater (salinity: 34 part per thousand; dissolved oxygen 65–85%) at 600 L h⁻¹. The photoperiod was set at 14 L:10 D, with lights on at 9 a.m. Ammonia and nitrite levels were kept around zero mg L⁻¹. At the end of the trial, the fish from each tank were weighed and biomass calculated. From each tank, 5 fish were used for blood collection, tissue sampling and histology, 5 more were used for texture analysis, and another 6 for proximal composition analysis.

2.2. Experimental Protocol—Day/Night Temperature Oscillations

The temperature variations for both groups, the Tconstant and Tdaily cycles, are represented in Figure 1. The water temperature of both experimental groups was kept warm during daytime for 7 h (warm period). The fish from the Tconstant group were kept at 18.8 ± 0.1 °C throughout the whole experimental period, while fish from the Tdaily cycles group were subjected to a temperature variation pattern during the night period. The water temperature of the Tdaily cycles group was programmed to decrease after 5 p.m. to initiate the cooling period. At this stage, the running warm water was replaced by cold water, and after 5 h, the water temperature reached 12.3 ± 0.2 °C. This was the time when the maximum difference between temperatures was attained (6.5 ± 0.1 °C). This cold temperature was maintained for 7 h, characterizing the cold period. At 5 a.m., the water temperature started to increase, taking approximately 5 h to reach the temperature of the warm water (~19 °C), signaling the end of the warming period. The water temperature oscillations were managed using a heat pump system, as described in [21].

2.3. Blood Collection and Tissue Sampling

Blood and tissue sampling occurred at 10 a.m. Blood collection was performed as described in [21]. After blood collection, the fish were immediately sacrificed, weighed, and measured and tissue samples (liver, gills, mid intestine, and white muscle) were collected. The dissection procedure of the tissues was performed on a cold platform. Samples were flash frozen in liquid nitrogen and stored at −80 °C until analysis. Liver samples for histology were collected from the upper part of the longer lobule. Liver and gill samples for histology were fixed with Bouin solution. Liver and visceral fat were weighed.

2.4. Liver Lipid Peroxidation

Lipid peroxidation is higher in well-irrigated tissues with lipid storage capacity, such as the liver, which is therefore considered a good marker of oxidative damage in fish. Lipid peroxidation levels were determined using a commercial kit (Bioxytech MDA-586, OXIS International, Portland, OR, USA, Cat. No. 21044), which is based on the reaction of a chromogenic reagent with MDA, a compound that arises from the oxidation of polyunsaturated fatty acids. Briefly, the liver samples were homogenized in 5 mL g⁻¹ of cold PBS (137 mM NaCl, 2.5 mM KCl, 10 mM Na₂HPO₄, 1.8 mM KH₂PO₄, pH 7.4) using an Ultra-Turrax disperser. A standard curve was constructed with an MDA standard ranging from 0 to 4 µM. Tissue samples and standards were processed in parallel and at the end, 200 µL of each sample was added to a 96-well plate in triplicate. The formation of MDA was measured on a microplate reader at 586 nm (Multiskan GO, Thermo Scientific, Waltham, MA, USA). The results were presented as nmol MDA per mg of total protein. Protein concentrations were determined using the Bradford protein assay (5000006, BioRad, Hercules, CA, USA) [22].

2.5. Fatty Acid Profile Determination in Muscle and Liver Tissues

The fatty acid methyl esters, also known as FAME, were determined by acid-catalyzed transesterification in fish tissue samples, as described in [23]. The results were expressed as a percentage of its relative content.

2.6. Histology Assessment

Liver and gill samples were fixed with Bouin solution for 72 h, washed in tap water, and stored in ethanol 70% (v/v) until paraffin embedding. Sections of 6–7 µm, obtained using a microtome Leica RM-2155 (Leyca, Vienna, Austria), were stained with hematoxylin and eosin and examined under a light microscope. Microphotographs were examined under a binocular microscope, Nikon Eclipse Ci, and photographed using a Nikon DS-Fi2 camera. All sections were examined by the same researcher. Liver sections were evaluated for the integrity of the whole tissue and the hepatocyte cytoplasmic area was used to extrapolate the amount of energy reserves. The cytoplasmic area was determined in a defined image area (n = 3 per section; 15 fish per treatment), chosen to best represent the whole tissue section, considering the distinct dyed structures: dark stained nuclei and light stained cytoplasm and extracellular matrix. Cellular areas were determined using the ImageJ program v1.6.0. Gill sections were evaluated according to the length of the secondary lamellae. Three secondary lamellae were randomly selected from each fish section (15 fish per treatment). In total, 45 measurements per treatment were applied to calculate average values using ImageJ program v1.6.0.

2.7. Protein Extraction and Western Blot Analysis

Proteins were extracted based on the protocol adapted from [24,25]. Equivalent amounts of extracted proteins (75 µg) were separated on 12% (w/v) TGX stain-free polyacrylamide gels (BioRad, Hercules, CA, USA, Cat. No. 1610185) and transferred onto PVDF membranes (Amersham, UK, 0.45 µm) with prior exposure of stain-free gels to UV light for 1 min using the Chemidoc XRS+ imager (BioRad, Hercules, CA, USA, Cat No. 1708265). Immediately after transfer, the stain-free blots were again exposed to UV light for total protein band detection. Western blot analysis was executed as explained in [25] and the detected bands were quantified using Imagelab 6.0 software and normalized to total protein.

2.8. Hematology and Plasmatic Determinations

Before obtaining the plasma, a sub-sample of blood was removed from the initial blood sample drawn from each fish for hematocrit and hemoglobin analysis as described in [21,26]. The remaining blood was centrifuged at 2500× g for 10 min to obtain plasma, which was stored at −80 °C until analysis. Cortisol, glucose, total protein, and triglyceride levels were determined using commercial kits as described in [21]. Each sample underwent duplicate examination.

2.9. Digestive Enzymes

Mid intestine samples were collected and cut up on an ice surface according to [26]. The samples were then homogenized (IKA homogenizer) in 30 volumes (w/v) of ice-cold 2 mM Tris-HCl buffer at pH 7.0 containing 50 mM manitol for periods of 2 × 30 s. Two aliquots of homogenate were obtained. The first aliquot was used for the determination of digestive enzymes in the homogenate, which was centrifuged at 3300× g for 15 min at 4 °C, and the supernatant transferred to a clean tube. The second aliquot was used for brush border membrane purification, where 100 μL of 0.1 M calcium chloride was added to the homogenate, followed by centrifugation at 9000× g at 4 °C for 10 min. The supernatant was centrifuged again at 26,000× g at 4 °C for 40 min. The pellet was re-suspended in 5 mM Tris-Hepes buffer at pH 7.5 containing potassium chloride and 1 mM DTT, and was used for enzyme activity determination. Amylase activity was determined using starch as a substrate, dissolved in a pH 7.4 sodium phosphate buffer [27]. Alkaline phosphatase activity was assayed using 5 mM p-nitrophenyl-phosphate (pNPP) as a substrate in a 30 mM carbonate buffer at pH 9.8 [28]. Aminopeptidase activity was assayed using 0.1 M L-leucine p-nitroanilide as a substrate in an 80 mM solution of phosphate buffer at pH 7.0 [29]. Alkaline protease activity was determined according to the method described by [30], modified by [31]. Total soluble protein in enzyme extracts was determined in line with the Bradford technique, using bovine serum albumin as the standard protein [22]. Enzymatic activities were expressed as mU per mg of total protein.

2.10. Whole-Fish Proximate Composition Analysis

Proximate composition of the whole fish was analyzed to evaluate the impact of temperature variations on body carcasses. At the end of the experiment, 6 fish per tank were collected and frozen at −20 °C. The proximate composition of the whole fish was performed according to the reference AOAC methods [32]: dry matter by drying at 105 °C for 24 h; ash by combustion at 550 °C for 12 h; crude protein (Nx6.25) using a flash combustion technique, followed by gas chromatographic separation and thermal conductivity detection (LECO FP528); and crude fat after ethyl ether extraction using the Soxhlet method.

2.11. Whole-Fish Texture Analysis

Texture analysis was performed on the whole fish, without scaling, by compression, using a TA.XTPlus texture analyzer (Stable Micro Systems, Godalming, UK), equipped with a load cell of 30 kg, applying the single compression test. Whole-fish firmness (hardness) was determined using a 5 mm diameter spherical probe (P/5S) to simulate human finger (thumb) compression, which moved down 4 mm on fish at a constant speed of 1 mm s⁻¹. A total of 4 measurements was made on each fish in the zone between the operculum and the first ray of the dorsal fin (two readings on each side, right and left). Fifteen fish (5 fish from each replicate) from both treatments were analyzed.

2.12. Statistics

All data were tested for homogeneity using Levene’s test, and the assumption of normal distribution was checked using the Shapiro–Wilk test. The Mann–Whitney U test was applied when the normality assumption was not verified. Values were reported as the mean ± SD, with the exception of Hsp70 expression, digestive enzymes, and hematologic parameters, which were expressed as mean ± SEM. The feed conversion ratio (FCR), specific growth rate (SGR), daily feed intake (DFI), and Hsp70 expression parameters were expressed as means of three replicates considering each tank as an experimental unit. The remaining parameters were expressed as means of 15 or 30 replicates, considering each fish an experimental unit per treatment. Student’s t-test was used to check the differences between the Tconstant and the Tdaily cycle treatments. The results were considered significant at the 95% confidence level (p < 0.05). All statistical tests were performed using IBM-SPSS Statistics v25 software. Graphs were created using GraphPad Prism 5 software.

3. Results

3.1. Production Parameters

No fish mortality occurred throughout the experimental trial. The behavior of the fish varied in response to water temperature. The Tconstant group showed higher swimming activity and appetite than fish subjected to thermal fluctuations. The latter feature was supported by a higher daily feed intake (DFI) (

Table 1. Morphometric feed utilization parameters and liver oxidized lipid content of gilthead seabream from the Tconstant and Tdaily cycle groups.

Parameters	Tconstant	Tdaily Cycles	p-Value
IBW 1	97.3 ± 1.9	96.1 ± 0.3	0.382
FBW 2	173.2 ± 8.1	129.5 ± 1.4	0.001 *
WG 3	8866 ± 789	4012 ± 200	<0.001 *
SGR 4	0.86 ± 0.05	0.45 ± 0.02	<0.001 *
CF 5	1.68 ± 0.13	1.59 ± 0.09	0.003 *
DFI 6	1.36 ± 0.05	0.91 ± 0.01	<0.001 *
FCR 7	1.65 ± 0.11	2.07 ± 0.08	0.005 *
HSI 8	1.57 ± 0.21	1.90 ± 0.18	<0.001 *
VFSI 9	2.06 ± 0.64	1.32 ± 0.47	0.001 *
Liver oxidized lipids 10	164.9 ± 55.0	168.1 ± 61.2	0.884

Values represent the mean ± SD. * Indicates statistically significant differences between treatments (t-test, p < 0.05). The statistical value of p is indicated. ¹ Initial body weight (IBW) (g): initial biomass (g)/ initial fish number. ² Final body weight (FBW) (g): final biomass (g)/final fish number. ³ Weight gain (WG) (g): Final biomass (g)—Initial biomass (g). ⁴ Specific growth rate (SGR) (% day⁻¹): [(Ln final body weight (g)—Ln initial body weight (g))/Time (days)] × 100. ⁵ Condition factor (CF) (g cm⁻³): body weight (g)/body length³ (cm) × 100. ⁶ Daily feed intake (DFI) (% day⁻¹): 100 × total amount consumed feed/[Time days × (final weight + initial weight)/2]. ⁷ Feed conversion ratio (FCR): total amount of consumed feed (g)/weight gain (g). ⁸ Hepatosomatic index (HSI) (%): liver weight (g)/body weight (g) × 100. ⁹ Visceral fat somatic index (VFSI) (%): fat weight (g)/body weight (g) × 100. ¹⁰ Liver oxidized lipids (nmol MDA g⁻¹ total protein).

). Fish final body weight (FBW) was greatest in fish maintained at constant temperature compared with fish exposed to temperature variations. Similarly, weight gain (WG), specific growth rate (SGR), and condition factor (CF) were greater in the Tconstant fish group. The feed conversion ratio (FCR) was lower for the Tconstant fish group, indicating significantly better feed utilization than fish exposed to water temperature fluctuations. The hepatosomatic index (HSI) and visceral fat somatic index (VFSI) were affected by water temperature variations, showing that the Tconstant group had higher values than the Tdaily cycles group for the latter parameter. The level of oxidized lipids was unaffected by water temperature variations (Table 1).

3.2. Fatty Acid Composition

Table 2. Fatty acid composition of the liver and white muscle fillets of gilthead seabream exposed to different treatments, Tconstant and Tdaily cycles.

	Liver		Muscle
% of Total Fatty Acids	Tconstant	Tdaily Cycles	Tconstant	Tdaily Cycles
C14:0	1.94 ± 0.16	1.88 ± 0.22	2.04 ± 0.26	2.14 ± 0.18
C16:0	17.31 ± 0.68 *	15.05 ± 0.73	17.11 ± 0.32 *	16.37 ± 0.29
C18:0	6.07 ± 0.86	6.19 ± 0.44	4.79 ± 0.39 *	4.52 ± 0.21
Other SFA	2.34 ± 0.53 *	3.08 ± 1.16	1.81 ± 0.11	1.81 ± 0.13
Total SFA	27.69 ± 1.32 *	26.19 ± 1.24	25.74 ± 0.43 *	24.81 ± 0.38
C16:1 (ω7 + ω9)	5.26 ± 1.23	5.71 ± 0.32	5.89 ± 0.61	6.16 ± 0.34
C18:1 (ω5 + ω7 + ω9)	31.19 ± 2.58	31.23 ± 2.86	29.49 ± 2.05	29.36 ± 1.11
C20:1 (ω7 + ω9)	1.47 ± 0.11 *	1.25 ± 0.26	1.50 ± 0.33	1.38 ± 0.12
Other MUFA	0.89 ± 0.10 *	0.81 ± 0.06	0.91 ± 0.10	0.87 ± 0.09
Total MUFA	38.80 ± 2.45	38.99 ± 3.17	37.79 ± 2.82	37.79 ± 1.49
C18:2 (ω6)	11.37 ± 0.87	11.36 ± 0.90	12.47 ± 0.41 *	12.84 ± 0.34
C18:3 (ω3)	0.84 ± 0.13	0.77 ± 0.12	0.99 ± 0.08	1.01 ± 0.05
C20:4 (ω6)	1.40 ± 0.27	1.61 ± 0.33	1.34 ± 0.30	1.32 ± 0.18
C20:5 (ω3) EPA	3.18 ± 0.49	3.31 ± 0.49	3.94 ± 0.31	4.01 ± 0.19
C22:6 (ω3) DHA	9.53 ± 2.38	10.16 ± 2.14	11.11 ± 2.30	11.27 ± 1.14
Other PUFA	4.30 ± 0.36	4.50 ± 0.30	4.72 ± 0.63	4.71 ± 0.20
Total PUFA	30.59 ± 3.74	31.77 ± 3.64	34.59 ± 2.50	35.19 ± 1.58
DHA/EPA	2.99 ± 0.48	3.03 ± 0.26	2.80 ± 0.38	2.81 ± 0.21
Total ω3	16.26 ± 3.04	17.08 ± 2.75	19.21 ± 2.47	19.39 ± 1.32
Total ω6	13.91 ± 0.91	14.21 ± 1.11	14.99 ± 0.26 *	15.34 ± 0.43
Ratio ω3/ω6	1.16 ± 0.17	1.19 ± 0.14	1.27 ± 0.16	1.26 ± 0.06
Total (SFA + MUFA + PUFA)	97.06 ± 1.09	96.96 ± 0.33	98.10 ± 0.66	97.81 ± 0.17
Total NI	2.94 ± 1.09	3.04 ± 0.33	1.90 ± 0.66	2.19 ± 0.17

Values are mean ± SD (n = 15). * Indicates statistically significant differences between each tissue from the experimental treatments (t-test, p < 0.05). SFA—Saturated fatty acids; MUFA—Monounsaturated fatty acids; PUFA—Polyunsaturated fatty acids; EPA—Eicosapentaenoic acid; DHA—Docosahexaenoic acid; Total NI—Non-identified fatty acids (100—Total SFA + MUFA + PUFA).

shows the fatty acid composition of the liver and white muscle fillets of gilthead seabream exposed to different water temperature regimes. Total fatty acids were unaffected by temperature variations in both tissues (p = 0.745, liver; p = 0.119, muscle). However, the total saturated fatty acid (SFA) content was significantly affected in the group of fish exposed to daily temperature oscillations, with a reduction of C16:0 fatty acids in both tissues (p < 0.001, liver; p < 0.001, muscle) and C18:0 fatty acids only in muscle fillets (p = 0.663, liver; p = 0.032, muscle). The levels of monounsaturated fatty acids (MUFA, p = 0.864, liver; p = 0.401, muscle) and polyunsaturated fatty acids (PUFA, p = 0.403, liver; p = 0.460, muscle) were unaffected by water temperature changes. Total ω3 levels were unaffected in both tissues (p = 0.461, liver; p = 0.227, muscle); however, total ω6 levels increased in the muscle of fish exposed to water temperature oscillations (p = 0.016). Nevertheless, these temperature variations showed no effect on the ratio between these two PUFAs (p = 0.632, liver; p = 0.635, muscle). Eicosapentaenoic acid (EPA, p = 0.473, liver; p = 0.465, muscle) and docosahexaenoic acid (DHA, p = 0.469, liver; p = 0.246, muscle) levels were unchanged by water temperature oscillations in both fish tissues.

3.3. Liver and Gill Histology

No histomorphological changes were found in the liver or gill secondary lamellae structures of fish exposed to both experimental treatments (Figure 2). Moreover, there were no significant differences between secondary lamellae length of gills from fish exposed to water temperature fluctuations and constant temperature (

Table 3. Measurements of gill and liver sections of gilthead seabream exposed to different water temperature treatments, Tconstant and Tdaily cycles.

Parameter	Tconstant	Tdaily Cycles	p-Value
HCA (%)	57.3 ± 4.81	60.6 ± 6.37	0.126
VL (μm g−1)	0.35 ± 0.05	0.34 ± 0.03	0.646

Values are means ± SD (n = 15). HCA—Hepatocyte cytoplasmic area: 100—nuclear area (%); VL—Secondary lamellae length (μm)/fish weight (g). The statistical value of p is indicated (t-test, p < 0.05).

). Similarly, thermal fluctuations had no significant impact on the energy reserves of fish hepatocytes, demonstrated by similar cytoplasmic areas.

3.4. Hsp70 Expression Levels

The relative expression of Hsp70 was monitored in the liver and gills of gilthead seabream (Figure 3). Either in the liver (p = 0.882) or in the gills (p = 0.159), Hsp70 expression levels were unaffected by water temperature variations.

3.5. Blood Parameters and Plasma Metabolites

Fish exposed to temperature variations exhibited greater hemoglobin values than those subjected to constant temperature during the whole experimental period (

Table 4. Blood parameters and plasma metabolites of gilthead seabream from Tconstant and Tdaily cycles groups.

Parameters	Tconstant	Tdaily Cycles	p-Value
Hemoglobin (g dL−1)	3.74 ± 0.46	6.43 ± 0.42	<0.001 *
Hematocrit (%)	32.9 ± 1.4	33.3 ± 0.9	0.841
Glucose (mg L−1)	54.3 ± 3.0	56.1 ± 3.6	0.714
Cortisol (μg dL−1)	35.9 ± 9.5	14.4 ± 4.4	0.031 *
Total proteins (g L−1)	30.9 ± 1.4	33.8 ± 1.4	0.026 *
Triglycerides (mg mL−1)	1.07 ± 0.04	1.15 ± 0.02	0.152

Values represent the mean ± SEM. * Indicates statistically significant differences between treatments (t-test, p < 0.05). The statistical value of p is indicated.

). No significant changes in hematocrit, plasmatic glucose and triglycerides were found between the different thermal treatments. Total protein levels were superior in fish exposed to temperature variations and cortisol levels rose in the group of fish exposed to constant temperature.

3.6. Activity of Intestinal Digestive Enzymes

Higher levels of alkaline phosphatase activity were found in the intestinal mucosa of gilthead seabream exposed to water temperature oscillations when compared with the group of fish subjected to constant temperature. The activity of aminopeptidase, amylase, and alkaline protease was unaffected by water temperature oscillations. Similar observations were registered for the analyzed brush border enzymes (aminopeptidase and alkaline phosphatase) (

Table 5. Intestinal brush border (BB) and digestive enzymes (mU mg⁻¹) in gilthead seabream exposed to different water temperature regimes, Tconstant and Tdaily cycles.

Digestive Enzymes	Tconstant	Tdaily Cycles	p-Value
Aminopeptidase	42.0 ± 6.9	46.1 ± 7.3	0.689
Amylase	2500 ± 300	2300 ± 300	0.641
Alkaline Phosphatase	96.8 ± 18.7	248.3 ± 52.8	0.015 *
Alkaline proteases	5.6 ± 1.8	4.6 ± 1.3	0.867
Aminopeptidase (BB)	244.1 ± 70.5	235.5 ± 54.2	0.436
Alkaline Phosphatase (BB)	586 ± 136	1157 ± 263	0.098

Values represent the mean ± SEM (n = 15). * Indicates statistically significant differences between treatments (t-test, p < 0.05). The statistical value of p is indicated.

).

3.7. Whole-Fish Proximate Composition and Texture

Daily temperature variations significantly affected the dry matter, crude protein, and fat content of fish (

Table 6. Proximate composition analysis and texture of whole fish from Tconstant and Tdaily cycles gilthead seabream groups.

Parameters	Tconstant	Tdaily Cycles	p-Value
Crude protein (%)	45.8 ± 1.7	49.0 ± 1.9	<0.001 *
Crude fat (%)	37.8 ± 2.5	32.9 ± 2.5	<0.001 *
Dry matter (%)	36.1 ± 1.6	34.1 ± 1.4	<0.001 *
Ash (%)	17.5 ± 3.1	19.1 ± 2.5	0.099
Texture	0.99 ± 0.17	1.10 ± 0.15	0.088

Values represent the mean ± SD (n = 15). * Indicates statistically significant differences between treatments (t-test, p < 0.05). The statistical value of p is indicated.

). Dry matter and crude fat content decreased with temperature oscillations while crude protein content increased. In contrast, ash content was unaffected by this variable. Moreover, water temperature variations had no effect on fish texture (Table 6).

4. Discussion

Repeated day/night water temperature oscillations induced several features associated with the winter syndrome, such as a reduction in growth, food ingestion, and increased HSI [6]. These observations may be due to the overall metabolic depression associated with the repeated cold periods [3]. Depressed metabolism may lead to alterations in the hepatic lipid metabolism of fish, causing an increase in liver mass due to the accumulation of lipids [33,34]. Despite a higher HSI, histological analysis showed no differences in the hepatocyte cytoplasmic area of fish exposed to both experimental treatments, suggesting similar energy reserves. In fact, gilthead seabream exposed to 5 cycles of thermal changes showed no lipid deposition in the liver [3]. Our results may indicate a moderate effect of thermal oscillations on the hepatic metabolism, perhaps caused by the continued feeding activity during daytime, which may compensate for any harsh hepatic damage caused by cold night periods. Indeed, liver fatty acid composition only differed in terms of total saturated fatty acid (SFA) content, which was significantly lower in the group of fish exposed to water temperature oscillations than in the Tconstant group, mainly due to the reduction of C16:0 fatty acid (palmitic acid). In muscle, thermal fluctuations significantly decreased the amount of SFA, primarily due to a reduction in C16:0 and C18:0 (stearic acid) fatty acids. This SFA decrease was accompanied by a significant increase in the levels of total ω6 polyunsaturated fatty acids (PUFA), mostly due to an increase in C18:2 fatty acid (linoleic acid). Low temperatures often lead to an increase in unsaturated fatty acids of phospholipids to maintain appropriate fluidity of cell membranes [34]. SFA in phospholipid tails is relatively straight. In contrast, unsaturated fatty acids contain some double bounds between carbon atoms, resulting in torsions of carbon chains. These twists in their tails push adjacent phospholipid molecules away, maintaining some space between the phospholipid molecules and resulting in a more fluid membrane [35]. For this reason, as the proportion of phospholipids with unsaturated fatty acids increases, the fluidity of membranes also increases [36]. The significant decrease in the proportion of SFA in the liver and muscle of gilthead seabream in response to thermal oscillations showed mild adaptation to the repeated cold periods. Our findings correspond to the response of steelhead trout to temperature drops, in which the concentration of unsaturated fatty acids of phospholipids increased in several tissues, including muscle [37]. Ibarz and colleagues reported no homeoviscous adaptation of liver membranes after exposing gilthead seabream to both short-term cold stress at two different rates and 5-cycle repetitive temperature fluctuations. However, they did report increased levels of unsaturated fatty acids in muscle tissue [20]. These contradictory effects on the liver of gilthead seabream in response to the cold may be due to the different experimental conditions, such as different temperature ranges, feeding regimes, and/or cold periods, causing different physiological responses in the fish.

Oxidative stress is an important component of the stress response in marine organisms, which are exposed to environmental stressors, such as temperature variations [38,39]. Cold water environments contain higher levels of dissolved oxygen, which may facilitate any oxidative injury. The higher proportion of unsaturated fatty acids found in the liver membranes of fish from the Tdaily cycles group can accentuate this scenario, since they are more susceptible to being attacked by oxygen [40]. However, these thermal oscillations caused no greater oxidative damage in the liver of fish compared with the group of fish exposed to constant temperature, as demonstrated by similar levels of the lipid peroxidation product, malondialdehyde (MDA). Studies on European sea bass demonstrated oxidative stress as the temperature moved away from thermal optimum [38]. Although daily temperature fluctuations can cause repetitive withdrawals when compared with the Tconstant group, fish are able to recover by remaining without significant oxidative stress injuries for some hours (7 h) of each cycle at ~19 °C. Indeed, the expression levels of Hsp70 showed no significant differences in the liver and gills of either experimental group of fish. Although cold temperatures can stimulate the expression of Hsp70 in the liver of gilthead seabream [41], the tested thermal oscillations did not induce any cellular thermal stress response, possibly because the levels of damaged or misfolded proteins were not high enough to trigger their action. Moreover, the thermal variations caused no histomorphological alterations to the secondary lamellae of the gills, with both groups exhibiting similar villi length. Although gills may adapt their structure to changes in the environment, such as temperature variations [42], our experimental thermal oscillations were not able to cause any change in their structure.

Previous observations of our group showed that 2 cycles of water temperature fluctuations decreased the metabolic activity of the gilthead seabream, with concomitant variation in the levels of several blood and plasmatic parameters, essentially observed close to the influence of the cold water period or during the water cooling period [21]. After a longer period of daily temperature fluctuations, gilthead seabream showed a significant increase in hemoglobin concentration, with no variation in hematocrit levels compared with the group of fish maintained at a constant temperature. The long-term repetitive water temperature oscillations might act as a stressor, forcing the fish to make daily adjustments in their oxygen carrying capacity. These adjustments may be induced either by higher swimming activity or lower dissolved oxygen levels during the warmer water periods or the opposite scenario during the colder phases [43]. Since blood sampling was performed at the beginning of the warm period (10 a.m.), the fish were coming from the warming period where they were adapting to the new warm temperature environment. Therefore, the higher hemoglobin values determined for the Tdaily cycles group of fish may be reflecting this shift in water temperature. Moreover, this group of fish demonstrated a reduction in cortisol levels, which could be associated with the different feeding activity exhibited by both experimental groups [21,44], mainly evidenced by the lower DFI in the Tdaily cycles group compared with the constant temperature group. Cortisol levels can influence glucose metabolism in fish [12]; however, under our experimental conditions, plasma glucose levels were unaltered in both groups. Triglyceride levels were also unchanged in both thermal regimes. This may indicate that these two blood-circulating metabolites are influenced by feeding since, by the time of sampling, both experimental groups were at similar post-prandial status. Drops in plasmatic total protein in fish are associated with fasting and low temperature scenarios [9]; however, in our study, we saw an increase in plasmatic total protein in the fish exposed to temperature fluctuations.

Temperature has a great effect on the structure and catalytic function of enzymes within metabolic pathways. For this reason, it is important to ensure an adequate rearing temperature level to guarantee that all metabolic pathways are working properly to produce high-quality, marketable-sized animals. Studies on the activity of digestive enzymes in fish can elucidate some aspects of the whole digestive capacity of fish. To evaluate the impact of day/night temperature oscillations on the digestive capacity of fish, the activity of pancreatic (amylase) and intestinal enzymes (aminopeptidase, alkaline phosphatase, and alkaline protease) was determined. Additionally, the activity of brush border membrane enzymes (aminopeptidase and alkaline phosphatase) was evaluated since these enzymes are responsible for the final stages of degradation and assimilation of nutrients by the intestinal cells. Overall, the long-term temperature fluctuations had no effect on the activity of brush border and digestive enzymes except for the cytosolic alkaline phosphatase, whose activity was higher in the fish exposed to temperature oscillations. A gradual decrease in water temperature from 18 to 2 °C caused a reduction in the intestinal alkaline phosphatase activity in Carassius carassius [45]. That study disagrees with our results perhaps because of the different experimental thermal oscillation conditions and the physiological response of the investigated species. In fact, the metabolism of the Tdaily cycles group exhibited higher plasmatic total protein levels and higher whole-fish protein content. Since the intestinal alkaline phosphatase dephosphorylates phosphorylated dietary moieties (e.g., proteins), the long-term day/night temperature variations could promote a higher protein metabolism to obtain energy to maintain physiological processes rather than growth. However, the fish reared at a constant temperature may have redirected energy from digested food to growth and fat storage, as directly evidenced by higher somatic growth (superior FBW, WG, SGR, CF), better feed efficiency (lower FCR), and increased visceral fat (VFSI) indicators [15,46]. Curiously, these differences in the whole-fish proximate composition analysis were not reflected in the fish texture. This parameter can be affected by several factors, such as lipid oxidation [47] and/or protein degradation [48]. Despite the differences in the whole-body fat and protein content between groups, the previously mentioned biochemical reactions had little or no effect on fish texture.

5. Conclusions

The overall aim of this study was to simulate the usage of a solar water heating system during daytime and investigate its effects on the whole performance of gilthead seabream juveniles. Long-term water temperature fluctuations reduced fish growth and affected several other physiological processes. Therefore, these findings favor the usage of a constant water temperature of ~19 °C for optimal gilthead seabream juvenile production instead of a day/night water temperature oscillating regime. Solar energy can be used to warm the rearing water 24/7. This decision can be done by fish farmers according to their operational conditions and/or business strategy.