Molecular Characterization and mRNA Expression of ISP2 and ISP4 in the Large Yellow Croaker (Larimichthys Crocea) Under Acute Cold Stress

Ice structure proteins (ISPs), also known as antifreeze proteins, can lower the point of freezing by inhibiting the growth of ice crystals and protect organisms from freezing temperatures. The large yellow croaker (Larimichthys crocea) is an important warm-temperate marine sh in Chinese aquaculture. Only a few ISP studies have been reported in this sh to date. In this study, the cDNA of ISP2 were cloned and characterized, and mRNA expression of ISP2 and ISP4 was assessed in different tissues of the large yellow croaker under different periods of acute cold stress (0, 6, 12, 24, 48 and 72 h, rewarming after 12 and 24 h). We found that ISP2 cDNA is 861 bases in length, encoding a protein of 168 amino acid residues. The mRNA expression of ISP2 and ISP4 in tissues of large yellow croaker under different periods of acute cold stress changed signicantly. In comparison with the control group, ISP2 expression increased dramatically in the heart (1,976 fold) and intestine (26 fold) after 3 h of acute cold stress and increased 43 fold in the spleen after 6 h. ISP4 expression was up-regulated signicantly in the brain (43 fold) and gill (376 fold) at 1 h acute cold stress, and increased 2,774 fold in the intestine at 3 h, 64 fold in muscle and 141 fold in the spleen after rewarming for 1 h after 12 h acute cold stress. These results indicate that ISP2 and ISP4 may play an important role in the response of large yellow croaker to acute cold stress.


Introduction
Ice structure proteins (ISPs), also known as antifreeze protein, play a role in inhibiting the growth of ice crystals, modifying ice morphology and inhibiting the recrystallization of ice via adsorption-inhibition (DeVries and Wohlschlag, 1969; Lee and Kim, 2016). ISPs lower the point of freezing non-colligatively but do not change the point of melting through adsorption on the surface of ice crystals and then inhibit ice growth, which increases the temperature gap between melting point and freezing point, and this temperature gap is termed thermal hysteresis (TH) (Barrett, 2001). The larger the TH activity, the stronger the antifreeze activity of the ISPs. ISPs are vitally important for polar sh to prevent blood from freezing in subzero temperatures of seawater through depressing the freezing point from 1 to 1.2°C in the blood, allowing sh blood to ow during winter; simultaneously, other internal uids are prevented from freezing by modifying the growth of ice crystals, which also protects cell membranes from ice crystal damage (DeVries, 1971; Hew and Yang, 1992;Fletcher et al., 2001;Harding et al., 2003). According to the composition of amino acids and structural characteristics, ISPs in sh are divided into ve types, ISPI, ISP2, ISP3, ISP4 and antifreeze glycoproteins, and no homology exists between these types of ISPs (Zhong and Fan, 2002;Li and Ma, 2012).
ISP4 was rst isolated from the serum of the north-temperate coastal water sh longhorn sculpin (Myoxocephalus octodecimspinosis) . Subsequently, researchers found that the physiological concentration of ISP4 in the blood of adult longhorn sculpins and shorthorn sculpins was lower than any other type of ISP, and could not conduct antifreeze activity Zhao et al., 1998;Gauthier et al., 2008). Based on these ndings, Gauthier et al. (2008) speculated that ISP4 may not be a key antifreeze protein in the blood of polar sh when other types of antifreeze protein are present. Surprisingly, in addition to polar sh, ISP4 is also found in warm-temperate sh and freshwater sh (Nishimiya et al., 2008;Zhang et al., 2009;Kim, 2015;Lee and Kim, 2016). These reports indicate that ISP4 may have other effects as well as antifreeze activity in warm-temperate sh.
Large yellow croaker is a warm-temperate marine sh that inhabits the north of the South China Sea and the East China Sea coastal waters of China. Since 1985, with the success of arti cial breeding techniques in this species, it is now mainly farmed in single net cages (Hong and Zhang, 2003). The temperature of water is a key factor for sh and is closely related to growth, reproduction and immune metabolism.
Extreme changes in water temperature are damaging for sh especially net cage farmed sh for they cannot escape from harmful regions of the sea. Large yellow croaker usually grow in water temperature from 10 ~ 32°C, rapid growth occurs in water temperatures 18 ~ 25°C, whereas survival may be threatened in water temperature higher than 34°C or lower than 7°C (Chen and Wu, 2011). On the basis of comparative transcriptome analysis in the liver of large yellow croaker under acute cold stress after 12 h conducted in our laboratory (data available at SRA http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi? acc=GSE67756), we found that the expression of ISP2 was signi cantly up-regulated and the expression of ISP4 increased but with insigni cance (Qian and Xue, 2016). In this study, we conducted a molecular characterization of ISP2 and ISP4, mRNA expression of these two genes was observed in liver, muscle, gill, heart, spleen, intestine, brain and kidney of 1-year-old large yellow croaker under acute cold stress for 1, 3, 6, 12, 24, 48 and 72 h, rewarming 1 and 3 h after 12 h cold stress (12f1 h and 12f3 h) and rewarming 1 h after 24 h cold stress (24f1 h). Our objectives were (1) to analyze the molecular properties of type-II ISP in large yellow croaker and (2) investigate the spatiotemporal expression of ISP2 and ISP4 in large yellow croaker under acute cold stress.

Animals and acute cold stress
Large yellow croaker (mean weight 80 ± 0.7 g) were purchased from mariculture in Xiangshan Bay (Zhejiang, China) and maintained in a laboratory at the Ningbo Ocean and Fishery Science Technology Innovation Base. Fish were randomly divided into eight groups and acclimated in 500 L plastic aerated tanks with ow-through seawater at 28°C under 14 h light/10 dark photoperiod for 7 days (30 sh in each tank and eight tanks in total). They were fed with granulated feed for large yellow croakers twice per day until 1 day before the experiment. A total of 150 sh in ve tanks were treated with acute cold stress using a seawater chiller until the seawater temperature dropped to 14°C in 2 h (cold stress group). The other 90 sh with no treatment were cultured in another three tanks (control group). After 12 h and 24 h acute cold stress, 30 sh were immediately transferred to a tank with the water temperature at 28°C for the rewarming experiment. Tissues including liver, liver, muscle, gill, heart, spleen, intestine, brain and kidney were harvested from three sh at each time point (1, 3, 6, 12, 24, 48, 72, 12f1, 12f3 and 24f1 h) in the acute cold stress and control groups. All sampled tissues harvested in this experiment were snapfrozen in liquid nitrogen and stored at − 80°C. The protocols of all experiments meet the "Zhejiang Laboratory Animal Management" guideline established by the Zhejiang Provincial Department of Science and Technology on the Use and Care of Animals.

Total RNA extraction and cDNA synthesis
Total RNA was extracted from the tissues harvested in each time point (including three control and acute cold stressed sh) using a Tissue RNA Kit (Omega, Georgia, USA) following the manufacturer's instructions. The total RNA was quanti ed with NanoDrop ND-1000 (Nanodrop Technologies) and RNA integrity (RIN) was assessed with an Agilent 2100 Bioanalyzer. The RIN values of all RNA samples were ≥ 8. All extracted RNA was stored at − 80°C. cDNA was synthesized using a PrimeScript RT reagent Kit with a gDNA Eraser (TaKaRa, Tokyo, Japan) according to the manufacturer's instruction, and cDNA was stored at − 20°C for further experimental analysis.

Cloning of the full cDNA length of ISP2 and ISP4
A partial cDNA length of ISP2 was obtained from the liver transcriptome data of large yellow croaker, which was investigated earlier in our laboratory, and the detailed sequence information can be found at http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE67756 (Qian and Xue, 2016). Gene-speci c primers for 5′-and 3′-RACE were designed based on the aforementioned partial sequences (Table 1). 5′and 3′-RACE was used to obtain sequences of the 5′-and 3′-untranslated terminal regions of ISP2 using a SMARTer RACE 5′/3′ Kit (TaKaRa, Tokyo, Japan) according to the kit manufacturer's instructions. The fulllength cDNA of ISP2 was assembled using overlapped fragments.

Spatiotemporal expression analysis of ISP2 and ISP4
Quantitative real-time PCR (qRT-PCR) was used to investigate the spatial and temporal expression of ISP2 and ISP4. The rst-stand cDNA from different tissues of the control sh (n = 3) as well as acute cold stressed sh were diluted 1:5 with sterile and DNase/RNase free distilled water and used as qRT-PCR templates. Primers were designed for the qRT-PCR of ISP2, ISP4 and β-actin (Table 1)

Molecular characterization of ISP2
The full-length cDNA of ISP2 (GenBank accession number: 2370265) was 861 bp, including a 77 bp 5′untranslated region (UTR), a 301 bp 3′-UTR with an AATAAA signal and a poly (A) tail, and a 507 bp open reading frame (ORF) encoding a putative protein with 168 amino acid residues, which contains a putative signal peptide with 18 amino acid residues (Fig. 1).
By using SMART analysis, we deduced that the ISP2 amino acid sequence contains a C-type lectin (CLECT or CTL) domain or carbohydrate-recognition domain (CRD), which starts at position 40 and ends at position 165 aa. The ISP2 of the large yellow croaker has low homology in multiple sequence alignment with the ISP2 of 11 other sh species, and the highest homology has 76% identity (Amphiprion ocellaris) (Fig. 1). The phylogenetic tree constructed in this study was based on the ISP2 amino acid sequences of 12 sh species (Fig. 2). The ISP2 of the large yellow croaker was the furthest distance from that of other sh species.

ISP2 mRNA spatiotemporal expression during the acute cold stress
The spatiotemporal expression levels of ISP2 in large yellow croaker under acute cold stress are shown in Fig. 3. In liver tissue, ISP2 mRNA expression increased approximately at 3, 6, 48, 72, 12f3 and 24f1 h after acute cold treatment in comparison to the control (P < 0.01). Especially at 3 h after cold treatment, the expression of ISP2 increased to 8.23-fold in comparison to the control but decreased signi cantly at 12 h (P < 0.01). In muscle tissue, the initial expression of ISP2 decreased signi cantly at 1 h after acute cold treatment and then was up-regulated remarkably at 3, 6, 12 h, and down-regulated again at 24 and 72 h (P < 0.01). In brain tissue, ISP2 mRNA expression increased signi cantly at 1 h and 72 h in the cold stress group with 9.95 and 55.57-fold changes, respectively, compared to the control group (P < 0.01). ISP2 mRNA expression decreased remarkably at 48 h and 12f1 h (P < 0.01), and no changes occurred throughout the other sampling time points in comparison to control (P ≥ 0.05). In heart tissue, the expression of ISP2 was up-regulated signi cantly at 1, 3, 6, 48, 72, 12f3 and 24f1 h (P < 0.01). In particular, at 3 h after cold treatment, the increased 1,976.29-fold compared to the control group. ISP2 mRNA expression decreased at 12f1 h (0.01 ≤ P < 0.05), and no expression occurred at 12 h (0.01-fold). In spleen tissue, at 6, 12f1 and 12f3 h after acute cold stress, the expression of ISP2 increased signi cantly by 43.5, 8.41 and 10.99-fold (P < 0.01), respectively, but was down-regulated remarkably at 3 and 24 h (P < 0.01). In kidney tissue, there were no changes in the expression of ISP2 from 1 to 72 h including 12f1 and 12f3 h after cold treatment (P ≥ 0.05), but at 24f1, ISP2 mRNA expression signi cantly increased 5.38-fold relative to the control group (P < 0.01). In gill tissue, the initial expression of ISP2 was decreased signi cantly during 1 to 12 h, then increased remarkably during 24 to 72 h, and decreased signi cantly after rewarming (12f1, 12f3 and 24f1 h) (P < 0.05). In intestinal tissue, ISP2 mRNA expression increased highly signi cantly by 26.62 and 6.09-fold at 3 and 12 h, respectively, and was highly signi cantly decreased at 6 and 24 h (P < 0.01). The expression of ISP2 was increased signi cantly at 72, 12f1 and 12f3 (0.01 ≤ P < 0.05), and no changes occurred throughout the other sampling times relative to the control sh (P < 0.05).

ISP4 mRNA spatiotemporal expression during acute cold stress
The spatiotemporal expression levels of ISP4 in large yellow croaker under acute cold stress are shown in Fig. 4. In liver tissue, ISP4 mRNA expression increased signi cantly at 48 and 24f1 h and decreased signi cantly at 6 and 72 h after acute cold stress (P < 0.05). There were no changes in the expression of ISP4 at other sampling time points in comparison to the control (P ≥ 0.05). In muscle tissue, there was no ISP4 mRNA expression at 1, 24 and 24f1 h, but expression increased dramatically at 3, 48 and 12f1 h (0.01 ≤ P < 0.05), particularly at 12f1 h after acute cold treatment, the expression of ISP4 was up to 64.02fold relative to the control. In brain tissue, the expression of ISP4 was up-or down-regulated signi cantly throughout sampling time points except at 12f3 h there was no change compared to the control (P < 0.05). At 1 h after acute cold treatment, ISP4 mRNA expression was increased by 42.81-fold relative to the control. In heart tissue, ISP4 mRNA expression decreased signi cantly during the rst 1 to 3 h compared to the control but increased signi cantly at 48 to 72 h, including the sampling time point 12f1 and 12f3 h (P < 0.01). In spleen tissue, the expression of ISP4 signi cantly changed at sampling time point 6 and 12f3 h (P < 0.05), particularly at 12f1 h, the expression of ISP4 mRNA was up to 141.27-fold that of the control group. In kidney tissue, ISP4 mRNA expression was signi cantly increased at 6, 12, 48 and 72 h, and decreased signi cantly at 3, 24, 12f3 and 24f1 h in comparison to the control sh (P < 0.05). In gill tissue, ISP4 mRNA expression increased by 375.84-fold that of the control, and increased signi cantly at 12, 24, 72 and 24f1 h, but decreased signi cantly at 3, 48 and 12f3 h (P < 0.05). In intestinal tissue, the expression of ISP4 decreased signi cantly at 6, 12, 24, 12f1 and 12f3 h, and increased signi cantly at 1, 3, 48 and 24f1 h, especially at the sampling time point 3 h, ISP4 mRNA expression was up to 2,774.02fold that of the control group (P < 0.05).

Discussion
Cold stress, especially in winter, is a key limiting factor in warm-temperate marine sh, including the large yellow croaker, which is cultured in net cages. In recent years, researchers have sought to understand the molecular response to cold stress in the large yellow croaker, in studies including the gene expression of CIRP (Miao et al, 2017), HSP27, HSP30, HSP47, HSP90, caspase-1 and caspase-7 (Yang, 2011). Our previous study found that ISP2 mRNA expression was signi cantly increased after 12 h cold stress in the liver of large yellow croaker, but there was no signi cant change in ISP4 compared to the control group (Qian and Xue, 2016). In this study, mRNA expression of ISP2 and ISP4 were investigated in eight tissues of the large yellow croaker during periods of cold stress (1, 3, 6, 12, 24, 48 and 72 h) and rewarming (12f1, 12f3 and 24f1 h). Results showed that the expression changes of these two genes during stress are tissue and time-dependent.
As mentioned earlier, type-II ISPs are only found in a few sh species (Yang, 2016), and almost no studies focus on ISP2 in warm-temperate sh. The brain, neural centre, is the tissue in sh that responds to changes in temperature the fastest and is an important tissue in the regulation of temperature (Crawshaw et al, 1985;Xu, 2011). In this study, encephalic mRNA levels of ISP2 increased signi cantly at 1 h cold stress, while no changes in ISP2 expression occurred in other tissues such as liver, heart, kidney and intestine. This seems to con rm that compared to other tissues the brain in sh responds rst to cold stress. At 3 h of cold stress, ISP2 mRNA expression does not change in the brain but increased dramatically in the liver (8.23-fold), muscle (21.6-fold), and intestine (26.62-fold), especially in the heart, mRNA levels of ISP2 increased by 1796.29-fold compared to the control group. A possible reason for this result could be that cardiomyocytes receive certain signals from central nerves, which promote ISP2 expression dramatically to alleviate the damage of 3 h acute cold stress on cardiomyocytes, and a similar response mechanism occurs in other tissues. It should be noted that, in this study, prolonged cold stress has no effect on the expression of ISP2 in the brain of large yellow croaker until 48 h, and on the contrary, signi cant changes occurred in ISP2 mRNA expression from 3 to 48 h in other tissues including liver, muscle, spleen, heart, gill and intestine. It is possible that ISP2 synthesized in the brain of large yellow croaker at 1 h cold stress is enough to protect nerve cells from subsequent cold stress, and in other tissues, ISP2 must be synthesized constantly to achieve protection in prolonged cold stress. In our study, a strange phenomenon occurred after rewarming, the expression of ISP2 in some tissues including liver (12f3 and 24f1), muscle (12f3 and 24f1), heart (12f3 and 24f1), spleen (12f1 and 12f3), kidney (24f1) and intestine (12f1 and 12f3) were signi cantly increased. It is not clear why rewarming should have such an effect in these tissues. Maybe it is a compensation mechanism in response to the warmer water temperature.
ISP4 (also known as AFP IV) is different from ISP2 and has been detected in warm-water shes as well as cold-water species, its function is controversial (Mao et al, 2018). In addition, the mRNA expression of ISP4 has only been investigated in the liver of large yellow croaker, and the protein of ISP4 has no remarkable antifreeze activity (Zhang et al, 2009 In our study, the mRNA expression of ISP4 was highly up-regulated (2774.02-fold) in the intestine at 3 h, but there were no signi cant effects or it was signi cantly decreased in other tissue including liver, brain, spleen, gill, kidney and heart. It is not clear why changes in ISP4 mRNA expression are different in these tissues in the same sh treated with acute cold stress after 3 h. One possibility could be that ISP4 functions as an apolipoprotein in the intestine after 3 h of cold stress.
In summary, the expression of ISP2 and ISP4 mRNA changed in many tissues of the large yellow croaker after acute cold stress and was time and tissue-dependent. The brain, a high-level central neural system, was a sensitive tissue which responded to acute cold stress within a short time (1 h). The expression of ISP2 and ISP4 were both signi cantly increased at 1 h in the brain, which may indicate that ISP2 and ISP4 proteins could be synthesized in large yellow croaker while the sh is undergoing acute cold stress and protect the central neural system from cold stress. It should be noted that some differences occurred in the level of mRNA expression of these two genes at the same stress time in the same tissue. We do not know why cold stress and rewarming should have such effects in mRNA expression of ISP2 and ISP4 within the same tissue at the same duration of cold stress. Could it mean that ISP2 and ISP4 have a synergistic effect to protect sh from cold stress of extreme differences in temperature? Further studies should be conducted to fully understand the function of ISP2 and ISP4 in this warm temperature sh. Declarations