The seahorse of H. kuda (body length 4.96 ± 0.34 cm, weight 0.276 ± 0.031 g) was provided by Xiangshan Base of Ningbo Yonghe Biotechnology Company, and temporarily raised for 2 weeks at the pilot base of Meishan campus, Ningbo University. Individuals with clean body and good vitality were randomly selected for the experiment. O. melatigma (body length 3.12 ± 0.46 cm, weight 1.730 ± 0.243 g) was bred in our laboratory in Meishan campus, also randomly selected for the experiment. Both fishes had similar ages of about 45 days old.

The temporary breeding conditions were: water temperature of 25°C ± 0.5°C, salinity of 25, dissolved oxygen (DO) >5 mg/L, light intensity of 1000lx, and photoperiod of 14L:10D. During the periods of temporary breeding and experiment, H. kuda was fed frozen Mysis (Tianjin Fengnian Aquaculture Company, China) with a total length of 1–1.5 cm as bait, and O. melatigma was fed with artificial formula feed (Ningbo Tianbang Group, China). Fed once a day at 8:00 Am, and removed any remained bait and feces after 2 h of feeding. The seawater was exchanged approximately 50% before feeding.

The experiment consisted of two tests.

Test 1: The set temperature changing amplitude was ±2°C. There were 3 groups of breeding temperature in the test, included the control group (Ck) with the temperature of 25°C, the heating group (HG) with an increase of 2°C (27°C) and the cooling group (CG) with a decrease of 2°C (23°C). There were 3 ways to reach the set temperatures (from 25°C to 23 or 27°C), that was 3 types of changing speeds: direct input (Marked as 0H2±), increasing or decreasing 2°C within 2 h (2H2±), where the changing speed was 1°C/h, and changing 2°C within 4 h (4H2±, 0.5°C/h).

Test 2: The set temperature changing amplitude was ±4°C. There were 3 groups of breeding temperature (25°C, 29°C, 21°C) and 4 speed treatments of 0H4± (direct input), 2 h (2H4±, 2°C/h), 4 h (4H4±, 1°C/h), and 8 h (8H4±, 0.5°C/h).

The temperature changing process was carried out in controlled-temperature illuminated incubators (Jiangnan GXM-508A, Ningbo, China). By conducting pilot experiments, appropriate temperature changed conditions were identified to ensure that the temperature changing amplitude was achieved within the required time (not sure at a uniform rate). The air temperature inside the incubator was controlled within the variable range of ±0.5°C, and the water temperature fluctuation was less than ±0.2°C in the incubator. After the temperature changes completed, it would be maintained the target temperature for 7 days before ending the experiment. In the experiment, the salinity of each treatment remained at 25, with three replicates for each treatment.

Statistical analysis of experimental data was conducted using SPSS 22.0 (IBM Co., NY, United States). Two-way ANOVA was used to examine the effects of temperature changes in amplitude and in speed on the growth, feeding behavior, gene expression levels, and enzyme activity of both fish. T-test was used to analyze the effects of different temperature changes and compare the differences in average values within and between 2 species fish. The data were expressed as mean ± standard error, with P < 0.05 indicating significant differences. Use GraphPad Prism 8.0 to plot data and statistical results.

The variations of RNA/DNA ratios in their muscles of H. kuda and O. melastigma under different changing amplitudes and speeds of temperature were shown in Table 2.

Under 2°C amplitude, there was a mild difference on the RNA/DNA ratios among 3 groups with different speeds in H. kuda, and no difference on O. melastigma at all, compared with their CK, respectively. The ratios of H. kuda in 3 treatments of 0H2± and 2H2+ had significantly decreased on the 4th day (P < 0.05). And 0H2- and 2H2+ had recovered on the 7th day, but 0H2+ became worse (Table 2). Under 4°C amplitude, there was a significant effect on the RNA/DNA ratios among 3 groups with different speeds in both H. kuda and O. melastigma. For O. melastigma, the 0H4- treatment on the 4th day was affected by a −4°C cooling (P < 0.05), and recovered on the 7th day. 2H4+ and 0H4+ were also affected by the heating (P < 0.05), and 2H4+ had recovered while 0H4+ was still suffered on the 7th day. For H. kuda, all treatments showed a significant decrease on the 4th day (P < 0.05). And 4H4± and 8H4± had recovered, while 2H4± and 0H4± became even worse on the 7th day. Comparing both fish, the impact of temperature changes on the growth of H. kuda was greater than that on O. melastigma. At the same time the impact of the same speed was different under different amplitudes, reflecting the synergistic effect of two factors on growth. Roughly summarized, the critical safe speed of O. melastigma was quite big, while that of H. kuda was around 1°C/h with 2°C amplitude.

Three parameters of feeding behavior, included feeding response time, feeding rate, and food intake, were measured for both H. kuda and O. melastigma in each treatment on the 4th and 7th day. The variations of food intake were shown in Table 3.

Overall, the variations in food intake of both fish were basically consistent with their growth. Under 2°C amplitude, different speeds of temperature changes had almost no effect on O. melastigma (P> 0.05), whether in heating or in cooling groups. For H. kuda, the food intake decreased obviously in 2 treatments of 0H2± on the 4th day (P < 0.05), and only 0H2- had recovered on the 7th day (Table 3). Under 4°C amplitude, the decreases in food intake of O. melastigma only occurred in 0H4± and 2H4+ treatments (P < 0.05). And all treatments (except for 0H4+) had recovered on the 7th day. However, the food intake of all H. kuda treatments had significantly decreased on the 4th day (P < 0.05). At the end of the experiment, only 8H4± had recovered, 2H4- had improved slightly, other 3 treatments of 0H4± and 2H4+ were even worse (Table 3). Comparing the treatments of the same speed under different amplitudes, it was evident that larger amplitudes had a greater impact on H. kuda.

By analysis, the other 2 feeding parameters also showed a similar characteritic, with the feeding response time prolonging and the feeding rate decreasing in H. kuda.

The activities or content of six enzymes (SOD, MDA, LPS, AMS, ACP, AKP) were analyzed in the experiment. The variations of SOD, LPS, and AKP in H. kuda and O. melastigma were shown in Figure 1.

Under both amplitudes, various enzymes showed their significant fluctuations in response to the different speeds of temperature changes, and their response trends were basically similar in H. kuda and O. melastigma (Figure 1). The bigger the amplitude, the greater the enzymes’ response was. And the faster the speed, the greater the enzymes’ response also was. At the same time, all enzymes also showed similar responses to temperature increasing/heating or deseasing/cooling, the impact of increasing temperature was greater. Certainly, there was some difference between H. kuda and O. melastigma at different speed with same amplitude, the main performance was that each enzyme had a difference in response time and response magnitude (Figure 1). Under 2°C amplitude, There was not much difference between 2 fish, except in 2H2- and 4H2+, where the enzyme expressions were slightly differed between 2H2- and 4H2+ in H. kuda (P < 0.05), which did not occur in O. melastigma. Under 4°C amplitude, the difference was also reflected in the rapid response at 12 h, as well as the increasing number of treatments involved in the continuous increase in enzyme activities in H. kuda (Figure 1).

The growth is the process of synthesizing and accumulating proteins, thereby achieving an increase in indicators such as body weight and body length. The expression of growth-related genes can directly or indirectly reflect protein synthesis in organisms (Luo et al., 2023, p.e81858). MSTN (myostatin) is a muscle growth inhibitor produced by the expression of Mstn gene, and is an important regulatory protein that inhibits muscle growth and development (Luo et al., 2023, p.e81858). GH (growth hormone) produced by the expression of Gh gene, which can accelerate cell division, and promote protein synthesis (Luo et al., 2023, p.e81858). Both Mstn and Gh play important regulatory roles in the growth and development of fish (Pan et al., 2022, p.1084). In this study, under the stresses of rapid temperature changes, Gh was downregulated and Mstn was upregulated in both fish, which indicated the growth of both fish had been affected. There was a good correlation between both genes expressions and the speeds of temperature changes. This indicated that the fish slowed down its cell division and reduced protein synthesis as the changing speed increased (Pan et al., 2022, p.1084), diverting some energy originally used for growth to alleviate the stress, leading to a slowdown in growth (Rasmussen and McLean, 2004, p.397; Pan et al., 2022, p.1084).

Many studies had used the ratio of RNA/DNA as an indicator to evaluate the growth status of fish, such as Sardina pilchardus (Ramirez et al., 2004, p.57), Labeo rohita (Siddiqua and Khan, 2022, p.e738114), and so on. Generally, RNA is highly expressed in rapidly growing fish, and DNA is relatively stable (Siddiqua and Khan, 2022, p.e738114). A high RNA/DNA ratio indicated there was a fast growth of fish (Siddiqua and Khan, 2022, p.e738114). In addition, the ratio was also highly consistent with the relative expressions of growth-related genes (Gh and Mstn) (Pan et al., 2020, p.1421). According the ratio and the expressions of both genes, we found the growth of O. melastigma was rapid recovery in the experiment, but the growth of H. kuda was severely inhibited, and the degree was increased with increasing amplitude or speed of temperature changes. The rapid recovery/adaptation ability of O. melastigma had also been reported in other species, such as O. sinensis, whose growth was not significant difference under the stress of temperature changes (±8°C, for 15 days) (Fang, 2021). In this study, the growth inhibition of O. melastigma only appeared in 0H4+ under 4°C amplitude. However, there was a critical line of growth difference observed in H. kuda, which was 2°C amplitude +1°C/h speed with 7 day, or 4°C amplitude +1°C/h speed with 4 day. Obviously, O. melastigma was more adaptable to the stresses caused by rapid temperature changes than H. kuda.

In fact, growth was a comprehensive performance that was related to all internal and external conditions, e.g., food. Usually, low food intake inhibited Gh and promoted Mstn to slow down the growth (Jackson et al., 2020, p.113319). So that, the fish not only required more energy to resist stresses (Portz et al., 2006, p.125), their poor physiological and health state in turn affected their appetite and led to a decrease in food intake, resulting in a decrease in growth rate.

Many studies had delved into the impact of temperature on fish’s feeding, indicating that changes in temperature could affect its digestive tract homeostasis, thereby affecting feeding quantity and feeding behavior (Sharma et al., 2017, p.134; Folkedal et al., 2012, p.350). Therefore, the changes of feeding quantity or behavior can be used to evaluate the impact of temperature stresses in fish. For example, temperature stress had a significant impact on the feeding rate and feed conversion rate of Cyprinus carpio (Chen et al., 2022, p.737636), Seriola lalandi (Stone et al., 2022, p.738567), and Scophthalmus maximus (Aydn et al., 2021, p.736891). Our previous researches also observed an increase in feeding response time and a gradual decrease in feeding rate and frequency for H. kuda with increasing temperature changes (Pan et al., 2020, p.1084). In this study, both the amplitude and speed of temperature changes had an impact on the feeding behavior of both fish, included prolonged the feeding response time, decreased the feeding rate and the food intake. Actually, the impact was relatively limited on O. melastigma, while obviously decreased the food intake in almost treatments of H. kuda (Table 3). It indicated that rapid temperature changes had a greater impact on the feeding of H. kuda.

In fact, a decrease in food intake was an external manifestation of internal physiology, which included the enzyme activities and gene expressions related to nutrition and metabolism. Under rapid temperature changes, the homeostasis in fish was affected to a certain extent. In order to adapt to new environmental conditions, the fish had to expend high energy costs for adaptive regulation (Portz et al., 2006, p.125). Carbohydrate and lipid were the two main energy substrates in fish (Portz et al., 2006, p.125). Therefore, it was of great significance that exploring the impact of environmental changes on the activities of key enzymes mainly involved in carbohydrate and lipid metabolisms and the relative expressions of genes encoded the key enzymes. This study analyzed two key enzyme genes encoding carbohydrate metabolism, Idh3b and Mdh1, and two key enzyme genes encoding lipid metabolism, Cpt1 and Fas, as well as two endogenous digestive enzymes, AMS and LPS, to compare the effects of temperature stresses on energy metabolism of O. melastigma and H. kuda.

Endogenous digestive enzymes were important indicators for studying the digestion and utilization of nutrients. The enhancement of their activities can promote the absorption and utilization of nutrients in fish (Jiang et al., 2021, p.57). It was generally believed that sugar was short-lived energy substances that underwent rapid decomposition through the action of AMS enzyme. Lipids, on the other hand, were long-lived energy substances that decomposed by LPS enzyme. In this study, the activities of AMS and LPS in different treatments had a significant change. Based on data of food intake, it was speculated that both fish may have increased their digestive enzyme activities, thereby increasing the assimilation efficiency of foodborne sugars and lipids to cope with excessive energy consumption (Tan et al., 2016, p.637). After a prolonged period of 4 or 7 days with severe stress, energy supplementation from carbohydrates became insufficient, and the consumption of lipids increased, showed by an increase in LPS activity (Figure 1). However, the LPS activity in O. melastigma and H. kuda was significantly different, which had been consistently increasing in H. kuda and showed up and down oscillations in O. melastigma. It indicated that O. melastigma may have already adapted to the temperature stress, with a relatively balanced assimilation alienation effect in the body. However, H. kuda was constantly consuming energy to resist the stress, thus fat-decomposition continued to increase. It was speculated that H. kuda will require a longer period to adapt to rapid temperature changes under the same stress.

Glucose was the primary energy substrate preferred by organisms (Portz et al., 2006, p.125). Malic acid dehydrogenase (MDH), generated by the expression of Mdh1 gene, was a key limiting-rate enzyme in glucose metabolism. Fish regulated carbohydrate metabolism by enhancing MDH activity in the liver to reduce the energy burden (Sharma et al., 2017, p.134; Chang et al., 2020, p.110744). Researches had found that the expression of Mdh1 was also easily affected in temperature-sensitive organisms. For example, Cyprinus carpio who was sensitive to low temperature was significantly lower of Mdh1 expression than that in cold-tolerant individuals (Liu et al., 2018, p.168). In this study, after 12 h stress of rapid temperature changes, the Mdh1 expression was rapidly upregulated in both fish, indicating that the fish increased their sugar metabolism levels in order to maintain a normal internal physiological state (Sharma et al., 2017, p.134). Clearly, O. melastigma had a faster response and earlier recovered at the following 4th or 7th day, while its expression was still increasing or oscillating in H. kuda at the same time (Figure 3). Similar results had also been reported in other fish (Chang et al., 2020, p.110744). Based on the results, it was speculated that there was some difference in the nutritional metabolism mechanism between O. melastigma and H. kuda in response to rapid temperature changes. Firstly, H. kuda faced a larger energy gap under the stress, secondly, aerobic carbohydrate metabolism was probablly the main energy pathway for H. kuda to resist stress (Sharma et al., 2017, p.134), so did O. melastigma, but O. melastigma was relatively stable and less affected under the same conditions. In a word, O. melastigma had a stronger adaptability and its more sources of energy than H. kuda to deal with rapid temperature changes (Chang et al., 2020, p.110744).

Lipid was also important energy storage substances in fish, and the decomposition of fatty acids can be accompanied by a large amount of energy release. Similarly, the mobilization and decomposition of stored lipids were strictly regulated by various internal and external factors (Cheng et al., 2017, p.110744; Sharma et al., 2017, p.134). The liver also was the main site of lipid metabolism in fish. Before fatty acids entered its decomposition pathway, they needed to be catalyzed by carnitine palmitoyltransferase 1 (CPT1) to form acyl CoA, which then entered the mitochondria β-oxidation reaction. CPT1, generated by the expression of Cpt1 gene, was the main limiting-rate enzyme in this process (Sharma et al., 2017, p.134). In this study, the expression of Cpt1 in both fish was affected by temperature changes. H. kuda showed an upregulated response in all treatments under 2°C ampitude and in the low-speed treatments under 4°C ampitude, and downregualted response (expression inhibition) in the high-speed treatments. Moreover, Cpt1 in O. melastigma showed little or no response under 2°C ampitude, and quickly upregulated and recovered under 4°C ampitude (Figure 4). It can be seen that Cpt1 gene in H. kuda was obviously upregulated in the treatments with slight temperature changes, which indicated that H. kuda increased the level of lipid metabolism in the liver, providing more energy for maintaining normal physiological activities (Sharma et al., 2017, p.134). In evidence, it also exceeded its regulatory capacity with large temperature changes. In O. melastigma, carbohydrate metabolism could ensure energy supply under slight changes. The lipid metabolism was only called on under large changes of temperature. Based on the above analysis, it indicated that the fish provided more additional energy for further adaptation to stress by upregulating the level of fatty acid decomposition (Volkoff et al., 2005, p.3). It also indicated that in the response to rapid temperature changes, the utilization mechanism of fatty acids in both fish was the same, but the utilization efficiency was significant difference.

Generally, environmental stresses made the production of intracellular reactive oxygen species (ROS) increasing, and organisms had to consume a large amount of energy from sugar and lipid decomposition to improve their antioxidant defense system. This system included antioxidant enzymes and antioxidant active substances, which were used to clear excess intracellular ROS and maintain normal physiology (Pan et al., 2022, p.1421). As an important member, SOD participated in the catalytic decomposition reaction of ROS, and its activity was an important indicator for evaluating the antioxidant capacity and health state of organisms (Pan et al., 2022, p.1421). MDA was a product of lipid peroxidation in cells and widely used to evaluate the degree of oxidative stress in organisms (Pan et al., 2022, p.1421). In addition, oxidative stress and immunity were closely related, jointly maintaining normal physiology (Fan et al., 2019). In this study, the antioxidant and immune responses of O. melastigma and H. kuda under rapid temperature changes were analysed and compared. Obviously, rapid temperature changes had a significant impact on the SOD activities and MDA contents in the 2 fish. Many similar results had been reported (Tan et al., 2016, p.637; Shi et al., 2022, p.1), indicating that the stress of rapid temperature changes can also induce antioxidant responses in both O. melastigma and H. kuda in this study. However, the timing of inducing antioxidant responses was different. In H. kuda, there was a significant increase in SOD activity and MDA content in the treatments with >1°C/h of changing speeds of temperautre, indicating that the stresses in the high-speed treatments may cause more serious oxidative damage to H. kuda. For the same reason, the critical safe speed of temperature changes in O. melastigma was relatively much higher.

Phosphatases, included ACP and AKP, were important metabolic and immune regulatory enzymes in organisms. They can enhance non-specific immune ability in fish (Fan et al., 2019, p.58). Temperature changes can affect their activity fluctuations (Ning et al., 2016, p.117). In this study, AKP activity did not change significantly in the 2 fish under small ampitude. However, it had significantly decreased with increasing changing speeds in H. kuda under 4°C ampitude of temperature, making it difficult to recover healthy, indicating that rapid temperature changes possibly inhibited the non-specific immune function of H. kuda, also happened in the 0H4± treatments of O. melastigma. In fact, temperature changes and duration of stress had a significant impact on AKP activity in fish (Meng et al., 2020, p.122). So, the non-specific immune abilities produced by O. melastigma and H. kuda in response to different amplitudes and speeds of temperature changes were different, and their antioxidant abilities were also different, resulting in different outcomes.

The enzymes involved in antioxidation and immunity were synthesized through the expressions of related genes. There were also some differences in the expression of Sod gene between O. melastigma and H. kuda. H. kuda had a significant expression response to slight temperature changes, while showed an inhibition and downregulation to large temperature changes. On the contrary, O. melastigma had little response to slight stress, and had strong regulatory ability against large stress. The expression of Gst gene, for synthetizing glutathione thiol transferase (GST) (Hooper et al., 2014, p.26), had a similar trend to that of Sod. The rapid upregulation of Gst was observed in O. melastigma treatments with 2°C amplitude and in H. kuda treatments with 4°C amplitude. All results indicated that H. kuda was more sensitive and had limited regulatory ability to rapid temperature changes in this study. This had been confirmed in terms of food supplementation, energy metabolism, and growth performance. In this study, the expression of heat shock proteins (HSPs) and its related genes (Hsp70, Hsp90) also were analyzed and compared. HSPs can repair cell damage caused by oxidative stress (Iwama et al., 1998, p.35). When the fish felt abnormal stress signals, its heat shock response was activated, and the expression of Hsp gene was upregulated, starting to produce HSPs to repair cell damage (Iwama et al., 1998, p.35; Chu et al., 2022, p.528). Therefore, the Hsp expression level can reflect the degree of stress response in organisms (Iwama et al., 1998, p.35; Chu et al., 2022, p.528). This mechanism was widely present in aquatic organisms, such as Chlamys farreri (Liu et al., 2013b, p.462), Procambarus clarkii (Xu et al., 2017, p.9), and Pampus argenteus (Shi et al., 2022, p.1). Temperature fluctuations, caused Hsp genes expression, and their expression levels were related to the stress-regulation mechanism of organisms under temperature stress (Cui et al., 2013, p.919; Shi et al., 2022, p.1). In this study, the expression level of hsp70 in H. kuda gradually increased with the increase of speed of temperature changes, but without significant changes in O. melastigma, further indicating that rapid temperature changes caused significant oxidative damage to H. kuda.

In this study, the variations in cell apoptosis signaling pathways were also analyzed and compared in O. melastigma and H. kuda. Apoptosis was the main pathway for maintaining the stability of the internal environment in organisms, and was controlled by genes for autonomous and orderly cell death (Zhou et al., 2020, p.735681). It was one of the important forms of non-specific immune system response and regulation to environmental stress (Zhou et al., 2020, p.735681). When a fish was subjected to stress, abnormal stress signals would activate its cell apoptosis signaling factors, inducing the activation of cell apoptosis signaling pathways (Zhou et al., 2020, p.735681). Among them, P53 (tumor suppressor protein) was an important component of the cell cycle signaling process (Siikavuopio and Sather, 2006, p.351). Caspase3 (CASP3), a key enzyme, was widely involved in processes such as the P53 signaling pathway and cell apoptosis pathway (Siikavuopio and Sather, 2006, p.351). In this study, rapid temperature changes had a significant impact on the relative expression levels of P53 and Casp3 genes in the liver of H. kuda, with significantly increased expression levels and gradually upregulated with increasing amplitude and speed of temperature changes, indicating that the process of apoptosis in liver cells of H. kuda had been activated (Liu et al., 2019, p.1407; Zhou et al., 2020, p.735681). The expressions of these 2 genes were also upregulated in O. melastigma, mainly showed by up and down oscillations and quick recovery. Overall, the expression levels of P53 and Casp3 in H. kuda were higher than those in O. melastigma, indicating that the transmission of apoptotic signals in liver cells of H. kuda was more responsive to rapid temperature changes than that in O. melastigma, which was possibly related to the fact that H. kuda was more sensitive to temperature changes (Dawood et al., 2022, p.397).

Interleukin-10 (Il-10) was an important inhibitory factor, and played an important role in the inflammation and immune regulation of organisms (Hossain et al., 2022, p.101038). In this study, the Il-10 expression in O. melastigma was regular, but chaotic in H. kuda. (Pan et al., 2020, p.1421) speculated that the main reason was the disruption of immune system in H. kuda caused by rapid temperature changes and its stress. Because of thermal stimulation, the expression of Il-10 as an anti-inflammatory factor fluctuated rapidly and disorderly, which probably was an expression of transitioning from normal physiological stress into pathological stress in H. kuda, represented a weakened response to anti-inflammation and a significant decrease in non-specific immune disease resistance (Liu et al., 2013a, p.462). It was generally believed that the intestine was the central organ of biological stress response. Pathological stress induced pathological changes of intestinal tissues, mainly manifested as mucosal erosion, ulcers, bleeding, and structural damage to mesenteric lymph nodes, leading to a decrease in immune function (Liu et al., 2013b, p.462), and ultimately an increase in disease incidence. This speculation was consistent with the actual situation in production (Wang et al., 2020, p.730). Under rapid temperature changes, H. kuda was likely to experience anorexia, then developed into enteritis and death.