The Potential Sex Determination Genes, Sox9a and Cyp19a, in Walleye Pollock ( Gadus Chalcogrammus ) Are Inﬂuenced by Water Temperature

: Our aim was to study the relationship between the sex-determining genes, sox9a and cyp19a , and water temperature in Gadus chalcogrammus . We assessed the sex ratio based on the expression levels of sox9a and cyp19a at di ﬀ erent water temperatures (5, 8, 11, and 14 ◦ C) and at di ﬀ erent stages of walleye pollock development (embryos, larvae, and juveniles). Next, we used immature walleye pollock to assess sox9a expression in males and cyp19a and vitellogenin (VTG) expression in females at di ﬀ erent water temperatures. Males expressed sox9a in the gonadal tissues, while females expressed cyp19a in the gonadal tissues and VTG in the blood plasma. In the ﬁrst experiment, cyp19a expression was higher at 5 ◦ C and 8 ◦ C, and sox9a expression was higher at 11 and 14 ◦ C. In the second experiment, sox9a expression remained relatively stable, but cyp19a expression decreased with increasing temperature, decreasing signiﬁcantly after 14 ◦ C. Similar patterns were also observed for VTG expression. These results indicate that lower water temperatures increase cyp19a expression, which increases the female ratio. Higher water temperatures increase sox9a expression, which increases the male ratio. Therefore, this study highlights the potential of the sex-determining genes and the inﬂuence of water temperature.


Introduction
Walleye pollock (Gadus chalcogrammus) resides in the coastal waters of Korea and Japan, throughout the Kamchatka Peninsula and the Bering Sea and into the central coast areas of California [1]. The habitat water temperatures (in the Bering Sea) range between 1 and 10 • C, and for fertilized larvae, between 2 and 7 • C [2,3]. Rising water temperatures due to climate change have affected the walleye pollock habitats. Water temperature influences habitat selection, migration, and physiological changes in walleye pollock, including sex determination, reproduction, growth, hormonal changes, and metabolism [4][5][6][7][8][9]. Environmental factors such as the water temperature are important to sex determination [10,11]. Sex determination happens at an early developmental stage, and sex is determined by a combination of genetic-dependent sex determination (GSD), environment-dependent sex determination (ESD) such as temperature, density, pH, and hypoxia, and endocrine factors [12][13][14]. In particular, the water 2 of 13 temperature experienced during embryos development has greater influence on sex determination than the genetic makeup [10,15]. After individual sex has been determined, gonadal differentiation in teleosts can lead directly to sexual differentiation in males and females or to development of both in hermaphroditic species [10,16]. Teleost fish possess numerous sex-determination genes. In male fish, sexual differentiation is achieved by the upregulation of doublesex and mab-3-related transcription factor 1 (dmrt1), which is highly conserved. Dmrt1 acts with transcription factor SRY (Sex determining region Y)-box 9a (sox9a) in testis formation [17,18]. Members of the SRY-box (Sox)-family are important for male determination because of their roles in testis determination and cartilage formation [19,20]. In female fish, sexual differentiation is achieved by a positive feedback loop of cytochrome P450 aromatase (cyp19a), which involves the transcription factor forkhead box protein L2 (foxl2) [21,22]. Cytochrome P450 aromatase (cyp19a), a member of the cytochrome P450 superfamily, is important for female determination due to its role in synthesizing estrogens. Estrogens promote ovarian development and maintain female differentiation. Cyp19a has been identified in females in different animal phyla. Simpson et al. (1994) found cyp19a transcripts in human ovaries, placenta, adipose tissues and the brain [23]. In fish, cyp19a is used for steroidogenesis of the gonadal steroids, which plays a critical role in gonadogenesis.
This study aimed to investigate the expression levels of sox9a and cyp19a and the mechanisms underlying the response of these genes to different water temperatures. Prior to this study, gonadal tissues from adult fish were studied to investigate the induction of sox9a and cyp19a, gonadal histological analysis, and VTG analysis in blood plasma. They were then divided into two experiments. The water temperatures studied were 5, 8, 11, and 14 • C. The first experiment reported the induction levels of sox9a and cyp19a in the embryos, larva, and juveniles (approximately 8-12 cm in length) at the different water temperatures. The second experiment investigated the induction levels of sox9a and cyp19a in immature fish (approximately 20-25 cm in length) to study the effects of water temperature in developed gonadal tissues (testis or ovary). The induction levels of vitellogenin (VTG) in female immature fish were also measured based on the VTG induction levels at different water temperatures, previously reported from this laboratory [24]. VTG is an egg yolk precursor that is synthesized in the liver. It is stimulated by estrogen and is rarely induced in males (though the VTG gene is present in male fish) [25]. As mentioned previously, cyp19a encodes aromatase to control estrogen synthesis, so correlation between cyp19a and VTG expression patterns in different water temperatures were compared.
To date, there have been no studies of sex determination or sex changes in walleye pollock in different water temperatures. Therefore, we suggest a potential mechanism of sexual determination based on water temperature in this report.

Animals and Materials
Walleye pollock (G. chalcogrammus) were obtained from the East Sea Fisheries Research Institute in Korea. Adult fish (3+ years old) were held in a flow-through tank (10-ton, 4 m diameter and 1 m depth), with an ambient photoperiod and a continuous supply of filtered seawater (10 µm PREFLOW II, Cs Technofil Co. Ltd., Gyeonggi-do, Korea) at 7.5 • C. The fish were fed twice per day. Embryos were obtained from the adult fish and were hatched and grown until the immature fish stage. The sizes of the fish used in the experiment at each developmental stage are shown in Figure 1. Approximately 1000 embryos ( Figure 1A) and larvae ( Figure 1B), 100 juveniles ( Figure 1C), and 50 immature fish ( Figure 1D) were grown in 10-ton water tanks at different temperatures (5,8,11, and 14 • C). After the eggs hatched, the fish were fed (in order) rotifers, Artemia nauplius, and frozen copepods. The dissolved oxygen (DO) and pH were maintained at 9-10 mg/L and 7.9-8.1, respectively. In experiment 1, the embryos were reared for 5 days, the larva for 30 days, and the juveniles for 180 days. In experiment 2, the immature fish were reared at 7.5 • C for approximately 180 days and then transferred to the water temperature treatments for~10 months before the experiment ( Figure 1E). The experimental protocol used in this study was reviewed and approved by the National Institute of Fisheries Science-Institutional Animal Care and Use Committee (NIFS-IACUC) per the ethical procedures and scientific care guidelines (approval number: 2018-NIFS-IACUC-16). The fish were sacrificed by an anesthetic overdose (MS-222 (Sigma), 50 ppm), and the heart, brain, kidney, intestine, and liver tissues were collected and stored at −80 • C. protocol used in this study was reviewed and approved by the National Institute of Fisheries Science-Institutional Animal Care and Use Committee (NIFS-IACUC) per the ethical procedures and scientific care guidelines (approval number: 2018-NIFS-IACUC-16). The fish were sacrificed by an anesthetic overdose (MS-222 (Sigma), 50 ppm), and the heart, brain, kidney, intestine, and liver tissues were collected and stored at −80 °C.

Isolation of mRNA Sequence
The reverse transcriptase PCR (RT-PCR) was performed to confirm the mRNA sequences and induction levels of sox9a and cyp19a gonadal tissues (testis or ovary) in adult walleye pollock. The gonadal tissues were homogenized using liquid nitrogen and PYREX ® Dounce Homogenizer Tissue Grinders (Thomas Scientific, Swedesboro, NJ, USA), and total RNA was extracted using TRIzol (Invitrogen). Only high-quality RNA (a ratio of 2.0 or higher for the absorbance at 260 nm and 280 nm) was used in the experiment. RNA integrity was assessed by electrophoresis on 1% agarose-formaldehyde gels after staining with ethidium bromide. The single-strand cDNA was synthesized from 5 µg of the total RNA with an Oligo (dT) primer by Avian Myeloblastosis Virus (AMV) Reverse Transcriptase (Promega). PCR was performed with the obtained cDNA (as a template), and 20 µM of the forward primer, reverse primer, Taq DNA polymerase, and 2.5 µM of each dNTP (Enzynomics, Korea). The conserved regions of the zebrafish (Danio rerio) sox9a (GenBank accession no. NM131643), Atlantic cod (Gadus morhua) sox9a (Genbank accession no. JN802288), and olive flounder (Paralichthys olivaceus) sox9a (GenBank accession no. KY924902) were used to design the primers. The cyp19a primers were designed from the conserved regions of the Atlantic cod (Gadus morhua) cyp19a (GenBank accession no. JJN802290), Sweetfish (Plecoglossus altivelis) cyp19a (Genbank accession no. KF296362), and common carp (Cyprinus carpio) cyp19a (GenBank accession no. DQ534411). All the primers were designed in the 'Primer 3' program, and Table 1 presents the primer sequences used in the experiments. The PCR conditions were 30 cycles at 94 • C for 30 s, 52 • C for 50 s, and at 72 • C for 1 min. Electrophoresis was used to evaluate the PCR products before insertion into a TA vector (Enzynomics, Korea) and cloning. The PCR products were evaluated by electrophoresis, and the nucleotide sequences were determined on a 3730xl DNA analyzer (Cosmogentec, Korea). The sox9a and cyp19a fragments were registered in GenBank (GenBank accession no. MT350342 and MT350341, respectively).

Quantitative Real Time-PCR
The quantitative real time-PCR (RT-qPCR) was used to examine changes in sox9a and cyp19a mRNA expression levels in whole or target tissues. For this experiment, 50 mg of the embryos and larval samples were used. The brain, intestine, liver, kidney, heart, and gonadal tissues from ten juveniles, five immature fish and adult fish from each water tank (chosen randomly) were used for total RNA extraction. First of all, DNase I was treated to remove genomic DNA contamination in total RNA. Each 20 µL reaction, based on the TOPreal™ One-step RT-qPCR Kit (Enzynomics co Ltd., Daejeon, Republic of, Korea), produced a 20 µL reaction that included the SYBR Green with the final concentration of 0.25 µM. Both primers had a concentration of 10 pmol/µL each, along with 50 ng/µL RNA extract. The Chromo4 System (Bio-Rad Inc., Contra Costa, CA, USA) was used to measure the reaction with 30 min at 50 • C, 10 min at 95 • C, 40 cycles of 5 s at 95 • C and 30 s at 60 • C (with endpoint data collection). All samples were tested in triplicate. The primers used for qRT-PCR was designed using 'primer3' program. The primer sequences are shown in Table 1. The 2 −∆Ct method was used for the two independent tests of relative Ct values. Ct values for reference gene (18s rRNA) were used to normalized expression of target genes.

Vitellogenin Assay
The correlation between VTG and cyp19a relative expression in female fish were investigated in differing water temperatures. Fish (>10 months old, approximately 20 cm) were raised from March to October in tanks with different water temperatures (5,8,11, and 14 • C). To distinguish the sex, the fish were dissected. The protocol for the previous experiment was used for VTG detection [24].

Histological Analysis
The gonadal tissues in adult fish were fixed in 10% neutral buffered formalin for 48 h, dehydrated in a graded alcohol series, embedded in paraffin wax, and sectioned with a Reichert microtome (4 µm).
The sections were stained with hematoxylin and eosin (H&E) and examined using an Olympus Provis AX70 microscope equipped with an Olympus Camedia C70/70 camera.

Statistical Analysis
Statistical analyses were performed in GraphPad Prism (version 8.4.3). Differences between the means were tested by Mann-Whitney or Kruskal-Wallis with post-hoc Dunn's multiple comparison tests were used (p < 0.05). All of the data are plotted as mean ± standard deviation (SD), and the appropriate statistical test values are reported. Pearson correlation analysis was used to examine the correlations between VTG and cyp19a.

Sox9a and Cyp19a mRNA Expression in the Gonadal Tissues with Vitellogenin
Prior to experiments 1 and 2, the adult fish were investigated for histological analysis of gonadal tissues, RT-PCR for the fragment mRNA sequence (data not shown; GenBank provided in Section 2.2), RT-qPCR of sox9a and cyp19a in gonadal tissues, and the VTG level in blood plasma, which are shown in Figure 2. Histological analysis of male gonadal tissues showed spermatozoa (Spz) and empty spaces after spermatozoa removal at a mature stage (Figure 2A). The testis is an unrestricted lobule that was spermatozoa-filled with spermatocysts, while the female ovarian follicles is filled with yolk globules and vacuoles and encased in zona radiata at the vitellogenesis stage (pre-spawning condition) ( Figure 2B). Sox9a was only found in male gonadal tissues while cyp19a was only found in female gonadal tissues ( Figure 2B). The male gonadal tissue showed only sox9a with 260 bp PCR product, while the female gonadal tissue showed only cyp19a with 360 bp PCR product. The comparative expression was very high-sox9a was expressed at a high level, and cyp19a was expressed at a very low level ( Figure 2D). In females, an average level of plasma VTG was 434.6 ± 29.18 ng/mL, while in males, the average level of plasma VTG was 1.76 ± 0.43 ng/mL ( Figure 2E).  Figure 3A shows the expression of sox9a and cyp19a in embryos at different water temperatures. Cyp19a expression was higher at 5 and 8 °C, while sox9a expression was higher at 11 and 14 °C . Figure  3B shows the expression of sox9a and cyp19a in larvae at different water temperatures. Similar to the embryos, cyp19a expression was higher at 5 and 8 °C, and sox9a expression was higher at 11 and 14 °C . Figure 3C shows the increment of sox9a induction in the embryos compared to cyp19a; sox9a expression increased with increasing water temperature, and at 14 °C, it was significantly different. Figure 3D shows the increment of sox9a induction in larvae compared to cyp19a. Similarly, sox9a expression increased with increasing water temperature, and at 14 °C, it was significantly different.  Figure 3A shows the expression of sox9a and cyp19a in embryos at different water temperatures. Cyp19a expression was higher at 5 and 8 • C, while sox9a expression was higher at 11 and 14 • C. Figure 3B shows the expression of sox9a and cyp19a in larvae at different water temperatures. Similar to the embryos, cyp19a expression was higher at 5 and 8 • C, and sox9a expression was higher at 11 and 14 • C. Figure 3C shows the increment of sox9a induction in the embryos compared to cyp19a; sox9a expression increased with increasing water temperature, and at 14 • C, it was significantly different. Figure 3D shows the increment of sox9a induction in larvae compared to cyp19a. Similarly, sox9a expression increased with increasing water temperature, and at 14 • C, it was significantly different.

Distribution of Sox9a and Cyp19a mRNA in Juvenile Fish
The distribution of sox9a and cyp19a mRNA expression was observed in the tissues of juvenile fish. These tissues were from the brain, intestine, liver, kidney, heart, and gonads (testis and ovary). The relative abundance of mRNAs in the walleye pollock tissues showed distinct differences ( Figure  4). The cyp19a gene in females was induced at a high level in the brain and gonads and at comparatively lower levels in the intestine, liver, kidney, and heart at 5 °C ( Figure 4A). Similarly, the sox9a gene in males was induced at a high level in the brain and gonads at 14 °C ( Figure 4B). Table 2 shows the number of individuals with high sox9a or cyp19a expression in the different water temperatures. There were more individuals with high cyp19a expression at 5 to 11 °C , whereas at 14 °C, there were more individuals with high sox9a expression.

Distribution of Sox9a and Cyp19a mRNA in Juvenile Fish
The distribution of sox9a and cyp19a mRNA expression was observed in the tissues of juvenile fish. These tissues were from the brain, intestine, liver, kidney, heart, and gonads (testis and ovary). The relative abundance of mRNAs in the walleye pollock tissues showed distinct differences (Figure 4). The cyp19a gene in females was induced at a high level in the brain and gonads and at comparatively lower levels in the intestine, liver, kidney, and heart at 5 • C ( Figure 4A). Similarly, the sox9a gene in males was induced at a high level in the brain and gonads at 14 • C ( Figure 4B). Table 2 shows the number of individuals with high sox9a or cyp19a expression in the different water temperatures. There were more individuals with high cyp19a expression at 5 to 11 • C, whereas at 14 • C, there were more individuals with high sox9a expression. expression was examined by randomly selecting 10 fish from each tank.

Sox9a and Cyp19a mRNA Expression with Vitellogenin in Immature Walleye Pollock
As mentioned previously (Section 3.1), it was confirmed that in gonads of adult fish, sox9a was only induced in testis, while cyp19a was only induced in ovary. Based on this result, the change in the induction levels of sox9a and cyp19a were investigated immature walleye pollock of male testis and female ovary, respectively. In immature fish, the samples were dissected to identify sex via gonadal tissues. Immature walleye pollocks were raised at different temperatures for 10 months, and the induction of sox9a and cyp19a mRNA in the gonadal tissues was studied in relation to the VTG induction level in female blood plasma ( Figure 5). The water temperature had no significant impact in the sox9a induction in testis. For cyp19a, there was little difference between 5 and 8 °C, but at 11 °C , the induction level decreased. The cyp19a induction level at 14 °C was less than half of the value observed at 5 °C in ovary ( Figure 5A). These results indicated that the sox9a induction level was similar at all water temperatures, but cyp19a induction decreased as the water temperature decreased. VTG induction in female blood plasma was 306.63 ± 28.84 ng/mL at 5 °C , 277.81 ± 17.57 ng/mL at 8 °C , 147.23 ± 18.03 ng/mL at 11 °C, and 25.68 ± 6.58 ng/mL at 14 °C ( Figure 5B). The individual distribution between cyp19a and VTG were recorded based on different temperatures ( Figure 5C). The individuals that were raised at 5 and 8 °C can be found at the top right. The individuals that were raised at 11 °C are found in the middle. The individuals raised at 14 °C formed a group at the bottom left ( Figure 5C). The correlation coefficient between cyp19a and VTG at different temperatures were highest in 5 °C groups with R 2 =0.89019 ( Figure 5D). There was a tendency for the coefficients to decrease as the temperature increased, with R 2 =0.8166 at 8 °C ( Figure 5E), R 2 = 0.5257 at 11 °C ( Figure  5F), and R 2 = 0.465 at 14 °C ( Figure 5G). This result showed that VTG was similar patterns to cyp19a. As mentioned previously (Section 3.1), it was confirmed that in gonads of adult fish, sox9a was only induced in testis, while cyp19a was only induced in ovary. Based on this result, the change in the induction levels of sox9a and cyp19a were investigated immature walleye pollock of male testis and female ovary, respectively. In immature fish, the samples were dissected to identify sex via gonadal tissues. Immature walleye pollocks were raised at different temperatures for 10 months, and the induction of sox9a and cyp19a mRNA in the gonadal tissues was studied in relation to the VTG induction level in female blood plasma ( Figure 5). The water temperature had no significant impact in the sox9a induction in testis. For cyp19a, there was little difference between 5 and 8 • C, but at 11 • C, the induction level decreased. The cyp19a induction level at 14 • C was less than half of the value observed at 5 • C in ovary ( Figure 5A). These results indicated that the sox9a induction level was similar at all water temperatures, but cyp19a induction decreased as the water temperature decreased. VTG induction in female blood plasma was 306.63 ± 28.84 ng/mL at 5 • C, 277.81 ± 17.57 ng/mL at 8 • C, 147.23 ± 18.03 ng/mL at 11 • C, and 25.68 ± 6.58 ng/mL at 14 • C ( Figure 5B). The individual distribution between cyp19a and VTG were recorded based on different temperatures ( Figure 5C). The individuals that were raised at 5 and 8 • C can be found at the top right. The individuals that were raised at 11 • C are found in the middle. The individuals raised at 14 • C formed a group at the bottom left ( Figure 5C). The correlation coefficient between cyp19a and VTG at different temperatures were highest in 5 • C groups with R 2 = 0.89019 ( Figure 5D). There was a tendency for the coefficients to decrease as the temperature increased, with R 2 = 0.8166 at 8 • C ( Figure 5E), R 2 = 0.5257 at 11 • C ( Figure 5F), and R 2 = 0.465 at 14 • C ( Figure 5G). This result showed that VTG was similar patterns to cyp19a.

Discussion
In this report, we have described the expression patterns of two sex-related genes, sox9a and cyp19a, in walleye pollock. Consistent with previous experiments, we found sox9a in mature male walleye pollock gonadal tissues and cyp19a in mature female counterparts. Degani (2014) and Adolfi et al. (2015) suggested that fish sox9a is a sex-related gene that is increased in expression in the testes and decreased in the ovaries [26,27]. Other researchers have studied the expression patterns of these genes zebrafish, medaka, and tilapia gonadal tissues during differentiation [28][29][30]. In several studies, cyp19a expression was higher in the ovaries, which was similarly observed in other species, including tilapia, black porgy (Acanthopagrus schlegelii), half-smooth tongue-sole (Cynoglossus semilaevis), goldfish (Carassius auratus), and spotted scat (Scatophagus argus) [31][32][33]. Fish sex determination and differentiation occur during fertilization and in the post-larval stages, and the water temperature plays a vital role in this process (as reported in > 60 different fish species) [10,13,[34][35][36]. In this study, the expression pattern of sox9a and cyp19a were similar during the embryos and larval stages. At 5 • C, cyp19a expression was the highest, and as the temperature increased, cyp19a expression decreased. Conversely, sox9a expression was high between 11-14 • C. This suggests that increases in water temperature are associated with decreased cyp19a expression. Tanaka et al. (2019) raised walleye pollocks in 2, 5, and 8 • C water and reported higher female ratios in the 2 and 5 • C treatments and a higher male ratio in the 8 • C treatment [37]. Fish in our study were raised at 7.5 • C, providing our reasoning for testing at higher temperatures. Despite our use of different water temperatures in juvenile fish, the results were consistent: the female ratio was higher at lower temperatures. To fully investigate the sex ratio, additional studies with more samples are required. Nonetheless, the results of these two independent studies indicate that water temperature is an important factor in walleye pollock sex determination. This is expected, as the sex in walleye pollocks is determined at early developmental stages, and cyp19a affects the sex-determination of females. Sox9a and cyp19a expression in the brain and gonadal tissues of juvenile fish were high. This is significant because the aromatization of androgens into estrogens in the brain is necessary for the regulation of several physiological and behavioral processes. In vitro assays of the brains in rats [38], birds, and fish [39] have also shown aromatase activity in many regions. The gonadal tissue investigation showed that high cyp19a expression was associated with female tissues, while high sox9a expression was associated with male tissues. For most fish, including walleye pollock, the up-regulation of sox9a leads to differentiation into male testis, whereas the down-regulation of sox9a leads to differentiation into female ovaries. The opposite patterns have been observed for cyp19a; up-regulation leads to differentiation into ovaries and down-regulation leads to differentiation into testes [21]. These results also show that walleye pollock sex is determined early in development, at the embryo or larval stages. Jorgensen et al. (2008) found that sox9a in zebrafish is expressed in the undifferentiated gonads, and Nakamoto et al. (2005) and Kobayashi et al. (2008) reported roles for sox9 in the formation of primitive gonadal tissues in medaka and tilapia [31][32][33]. Similar observations have been reported in zebrafish and black porgy [40].
Immature walleye pollocks (approximately 20-25 cm in length) were raised in different water temperatures for 10 months, and there were no differences in sox9a expression in the male gonadal tissues. However, cyp19a expression significantly decreased at 11 • C and even more at 14 • C.
Previously, we studied the expression levels of 17β-estradiol (E2) and VTG at different water temperatures and found that increased water temperature is associated with E2 decreases, which affects the induction of VTG [24]. High temperatures impaired the ovaries in female pejerreyes (Odontesthes bonariensis) by inhibiting the follicle-stimulating hormone receptors (fshr) and plasma E2 levels [41]. Anderson et al. (2012 and2017) found that suppression of plasma E2 and impairment of downstream VTG is common in female Salmo salar exposed to heat [42,43]. Studies have also shown that cyp19a plays an important and conserved role in the regulation of ovarian E2 [21,44,45]. In this study, VTG induction decreased as the water temperature increased and declined sharply at 14 • C (similar to cyp19a). The correlation between cyp19a and VTG is 11 and 14 • C lower than 5 and 8 • C, which may be related to the vitellogenesis stage during feminization, so further study is needed.

Conclusions
This study shows that sex determination and/or differentiation in walleye pollocks occurs in the embryos and post-larval stages. Higher water temperatures tend to result in a higher male ratio, whereas lower water temperatures lead to a higher female ratio. Higher water temperatures also suppress cyp19a expression, which decreases VTG and potentially affects sexual differentiation in female Gadus chalcogrammus.