1 INTRODUCTION

The large yellow croaker Larimichthys crocea is one of the most important fish species in Chinese mariculture. The annual output of this fish has exceeded any other single net, cage-farmed marine species in China (Li et al., 2013; Shen and Heino, 2014).

Temperature is one of the most important ecological factors that significantly affects the growth, metabolism, development, and other life activities of fish (Brett, 1971; Pankhurst and King, 2010; Quinn et al., 2011). For L. crocea, the adapted temperature range is 10-32℃, and the optimum growth temperature is 18-25℃ (Xue et al., 2014). The current aquaculture methods for L. crocea are still dominated by shallow sea cages at a depth of 4-6 m. In summer, owing to the shallow depth, L. crocea are forced to live in high temperature seawater that is near to, or higher than, its adapted endurable temperatures for several days, resulting in a weakened body, susceptibility to disease, and even death. With rising global temperatures (Gleckler et al., 2012), it is necessary to develop thermal-tolerant lines of L. crocea using marker-assisted selection breeding. Currently, many markers associated with thermal tolerance have been found in fish, such as Paralichthys olivaceus (Lu et al., 2007) and Scophthalmus maximus (Ma et al., 2011). However, there are no reports of these in L. crocea.

In this study, three microsatellite markers related to thermal tolerance were screened and identified using a Bulked Segregation Analysis (BSA) method (Michelmore et al., 1991). These markers may assist in breeding thermal-tolerant L. crocea in the future.

2 MATERIAL AND METHOD 2.1 High temperature stress experiment

A high temperature stress experiment was conducted from May to June 2014 in Hatcheries of Jinling Fisheries Ltd., the city of Ningde, Fujian Province, China. Before starting the experiment, 700 2-month-old healthy fish were randomly selected and placed into indoor concrete ponds (2 m³) to acclimate for one week. During that time, the water was maintained at ambient temperature (25.0±0.3℃) and changed once a day. The fish were fed commercially manufactured feed. After holding for 7 d, the water temperature was increased 1℃ per day until the experimental fish began to die. We then stopped heating the water and maintained the temperature for 1 d. The temperature was then increased at a rate of 0.5℃ daily until all of the fish died. During the experiment, the fish were fed with commercially manufactured feed and the water was changed once a day using preheated water. Although dissolved oxygen measurements were not recorded in the test, water was aerated and vigorously circulated with compressed air released through a submerged air stone to provide sufficient oxygen. The number of deaths, time of death and corresponding temperature for each fish were recorded, and the fins were collected for DNA extraction. In this study, a dynamic heating method was employed to estimate thermal tolerance for the 2-month-old fish. Fish whose opercula had stopped beating for 2 min were defined as dead. The critical thermal maximum (CTMax) was determined by exposing all individuals to water with a constant increasing temperature until all fish were dead (see Bennetti and Judd, 1992; Kita et al., 1996). The temperature at which half the population reached the end-point was reported as the 50% critical thermal maximum (50% CTMax) when the water temperature was increased (Jian et al., 2003; Cheng et al., 2013). The temperature at which the fish started to die was recorded as the survival thermal maximum (STMax) when temperature was increased. This experiment was approved by the Animal Care and Use committee of Fisheries College of Jimei University, Xiamen, China.

2.2 DNA extraction and SSR marker analysis

Genomic DNA was extracted from each fish using a standard phenol chloroform protocol (Sambrook et al., 1989). DNA quality and quantity were detected using a UV spectrophotometer. Each DNA concentration was adjusted to 30 ng/μL. Microsatellite primers were synthesized by Shanghai Sangon Biological Engineering Co. Ltd. (Shanghai, China). A total volume of 10 μL of reaction mixture was composed of template DNA, 1 μL (30 ng); 10×PCR buffer, 1.0 μL; 15 mmol/L MgCl 2, 1.0 μL; 10 mmol/L dNTPs, 0.2 μL, 10 mmol/L of primers, each 0.2 μL; 5 U/μL Taq enzyme, 0.1 μL; and water, 7.3 μL. The PCR cycle procedure included an initial denaturation at 94℃ for 5 min, followed by 30 cycles of denaturation at 94℃ for 30 s, an annealing temperature for 30 s, an extension at 72℃ for 30 s, and a 10 min final extension at 72℃. The amplification products were resolved by electrophoresis on 6% polyacrylamide denaturing gels at 1 500 V for 1.5 h, and the marker bands were revealed using a silverstaining protocol and recorded for analysis by photography.

2.3 Marker-phenotype association analysis

Phenotypic extreme bulks (R bulk and S bulk) were made for the marker analysis. The R bulk contained DNA from 15 thermally tolerant fish, and the S bulk contained DNA from 15 thermally sensitive fish. One hundred and sixteen SSR primer pairs (developed by Ye et al.) distributed across the L. crocea genome were used for PCR amplification (Ye et al., 2014). The SSR primer pairs, which generated polymorphic markers between the R bulk and S bulk, were surveyed on 60 independent extreme fish to evaluate the association between the microsatellite markers and thermal tolerance using a Chi-squared test.

2.4 Cloning and sequencing of the associated markers

The associated makers detected from 60 independent extreme fish were cloned and sequenced. The steps were as follows: First, the target strips were cut from the polyacrylamide gel, added to ddH₂O and put into water at 95℃ for 5 min. The samples were then preserved at room temperature overnight. Second, the mixed solutions were centrifuged (13 800 × g) for 2 min, and the supernatant fluids were taken as the template for another PCR reaction. Third, the PCR products were gel-purified and cloned into a PMD-19T vector (TaKaRa, Dalian). They were then transformed into competent Escherichia coli DH5α cells and sequenced by Shanghai Sangon Biological Engineering Co. Ltd. (Shanghai, China).

3 RESULT 3.1 Thermal tolerance evaluation of large yellow croaker

In the high temperature stress experiments, the number of deaths, time and corresponding temperature of L. crocea were recorded (Table 1). The survival thermal maximum (STMax) was 33.0℃, 50% critical thermal maximum (50% CTMax) was 35.5℃ and critical thermal maximum (CTMax) was 36.0℃.

3.2 Marker-phenotype association analysis

Seven amplification fragments showed frequency differences between the R bulk and S Bulk (Table 2, ID 1-15 DNA were mixed as S Bulk, ID 46-60 DNA were mixed as R bulk).

Four fragments appeared in the S Bulk, and three fragments in the R Bulk (Table 3). These fragments were verified in 60 single extreme fish (ID 1-60 in Table 2), and the results are shown in Table 4. The fragments were amplified at LYC0148 (181 bp) and LYC0200 (197 bp) and showed a frequency difference between the thermally tolerant fish and thermally sensitive fish (Table 4, Figs. 1, 2). The frequencies in the thermally tolerant fish were higher than the thermally sensitive fish (P < 0.01). By contrast, the fragments amplified at LYC0435 (112 bp and 100 bp) in the thermally sensitive fish were significantly higher than in the thermally tolerant fish (P < 0.01) (Table 4, Fig. 3).

3.3 Cloning and sequencing of the associated makers

To verify the associated markers were the existing microsatellites, and obtain the microsatellite sequences, the associated makers were gel-purified, cloned and sequenced. The sequencing results are shown in Table 5. For these loci, the core motif was "AC" repeats. Online BLAST analyses with the sequences was undertaken (Table 5) to align to the genome of the large yellow croaker (GenBank assembly accession: GCA_000972845.1) (http://blast.ncbi.nlm.nih.gov/Blast.cgi?PAGE_TYPE=BlastSearch&PROG_DEF=blastn&BLAST_SPEC=Assembly&ASSEMBLY_NAME=GCA_000972845.1). The locus LYC0148 had significant alignments with "the Larimichthys crocea isolate SSNF unplaced genomic scaffold scaffold761, whole genome shotgun sequence" (accession No.: KQ041183.1). The locus LYC0200 had significant alignments with "the Larimichthys crocea isolate SSNF unplaced genomic scaffold scaffold45, whole genome shotgun sequence" (accession No.: KQ041981.1). The locus LYC0435 had significant alignments with "the Larimichthys crocea isolate SSNF unplaced genomic scaffold scaffold33, whole genome shotgun sequence" (accession No.: KQ041581.1).

4 DISCUSSION

Within a certain temperature range, fish can acclimate to a change of ambient temperature by adjusting their physiological activity and metabolism. However, if the temperature exceeds a fish's tolerance limits, it will cause internal environmental disorders and even death (Feng and Wang, 1984). Currently, there are two main methods for evaluating the temperature tolerance of fish, acute heating and slow heating (Bevelhimer and Bennett, 2000; Rajaguru and Ramachandran, 2001; Mora and Maya, 2006; Eme and Bennett, 2009). Fish are ectotherms. Therefore, slow heating allows the fish to have a sufficient time to adapt to a wide range of temperature shifts (Carveth et al., 2007; Ndong et al., 2007). Based on this, the present study used the slow heating method. We found that the STMax for L. crocea was 33.0℃, 50% CTMax was 35.5℃, and CTMax was 36.0℃. The STMax in this study is slightly higher than that of our former study (Li et al., 2015), and this may be related to experimental subjects. In the present study, 2-month-old fish were used, whereas in the previous study, 12-month-old fish were used. The thermal-tolerance of fry is higher than adult fish which is also observed in other fish such as redband trout Oncorhynchus mykiss gairdneri (Rodnick et al., 2004) and rohu Labeo rohita (Das et al., 2005).

In this study, three markers (four fragments) associated with high temperature were found. However, these markers are only associated markers and not specific markers (i.e., only appeared in thermally tolerant fish or thermally sensitive fish), suggesting that these markers may be closely linked with the thermal tolerance gene. In addition, thermal tolerance is a quantitative trait that may be determined by many genes. According to Ye et al. (2014), the locus LYC0148 was assigned to linkage group LG20, the loci LYC0200 and LYC0435 were assigned to the same linkage group LG12 and the genetic map distance between the two markers is 9.1 cm. This gives an indication that there is one gene in this region of the genome that is associated with thermal tolerance. The associated markers found in this study can only partly explain the thermal tolerance of L. crocea. Therefore, it will be necessary to undertake a more thorough study and screen additional markers or genes associated with high temperature.

5 CONCLUSION

In this study, thermal tolerance of L. crocea was evaluated and three associated microsatellite markers were screened and identified using a Bulked Segregation Analysis (BSA) method. Our results provide reference data for breeding and domestication of thermally tolerant L. crocea.