Transcriptional Responses of Copper-Transport-Related Genes ctr1, ctr2 and atox1 and Their Roles in the Regulation of Cu Homeostasis in Yellow Catfish Pelteobagrus fulvidraco

Here, we characterized the function of ctr1, ctr2 and atox1 promoters in yellow catfish Pelteobagrus fulvidraco, a common freshwater teleost in Asian countries. We obtained 1359 bp, 1842 bp and 1825 bp sequences of ctr1, ctr2 and atox1 promoters, and predicted key transcription factor binding sites on their promoters, including MRE, SREBP1, NRF2, KLF4 and STAT3. Cu differentially influenced the activities of ctr1, ctr2 and atox1 promoters from different regions. We found that the −326/−334 bp and −1232/−1240 bp locus in the atox1 promoter were functional NRF2 binding sites, which negatively controlled the activity of the atox1 promoter. The −91/−100 bp locus in the ctr1 promoter and −232/−241 bp and −699/−708 bp locus in the atox1 promoter were functional SREBP1 binding sites, which positively controlled the activities of ctr1 and atox1 promoters. Cu inhibited the NRF2 binding ability to the atox1 promoter, but promoted the SREBP1 binding ability to the ctr1 and atox1 promoters. Dietary Cu excess significantly down-regulated hepatic mRNA and total protein expression of CTR1, CTR2 and ATOX1 of yellow catfish, compared to the adequate dietary Cu group. The subcellular localization showed that CTR1 was mainly localized on the cell membrane, CTR2 in the cell membrane and the lysosome, and ATOX1 in the cytoplasm. In conclusion, we demonstrated the regulatory mechanism of three Cu transporters at the transcription levels, and found the functional NRF2 and SREBP1 response elements in ctr1, ctr2 and atox1 promoters, which provided new insights into their roles in the regulation of Cu homeostasis in fish.


Introduction
Copper (Cu) is an essential mineral in vertebrates, including fish, and plays important roles in different metabolic pathways, including cellular respiration, transcriptional regulation, ion uptake and signal recognition [1,2]. However, when the body ingests excessive Cu, it will be potentially toxic [2]. Therefore, Cu homeostasis must be regulated tightly, which helps to maintain intracellular Cu levels within a reasonable range and prevent Cu toxicity.
The regulatory mechanism for Cu homeostasis is highly conserved from unicellular yeast to mammals, which was mainly involved in Cu uptake, distribution, storage and export [3,4]. Cu homeostasis is primarily controlled by several Cu-uptake-related proteins, such as copper-transport-related proteins (CTR1 and CTR2), two Cu-ATPase proteins (ATP7A and ATP7B) and three copper chaperone proteins (ATOX1, CCS and COX17) [5]. Among these members, CTR1, CTR2 and ATOX1 play important function in the regulation of Cu homeostasis. CTR1 is a highly active Cu transporter and mainly localized in the cytoplasmic membrane, which can transport Cu into cells in a low-Cu environment [3,6]. CTR2 is a low-affinity Cu transporter and more localized on the membrane of intracellular organelles, which can transport Cu when the environmental Cu concentration is high [7]. ATOX1 is one metallochaperone that delivers Cu to ATP7A and ATP7B transporters [8]. Currently, the functions of these Cu transporter proteins have been studied [8][9][10], but their transcriptionally regulatory mechanisms in vertebrates, including fish, are unclear.
In eukaryotes, studies suggested that the most effective regulation for gene expression occurs at the transcriptional level [11]. Therefore, it is very crucial to study how the genes initiate their transcription. Nuclear factor erythroid 2-related factor 2 (NRF2) is an important regulator of the body's antioxidative stress, which can bind to antioxidant response elements (AREs, 5'-TGACNNGC-3') to activate the transcription of downstream genes [12]. The sterol regulatory element-binding protein 1 (SREBP1) coordinates the gene transcription of enzymes involved in the lipogenic pathway [13,14]. Studies suggested that Cu enhanced NRF2 signaling [15,16] and increased the srebp1 mRNA expression [14,17]. At present, although the function and its potential targets of NRF2 and SREBP1 have been studied [18,19], it remains unknown whether these transcriptional factors mediated the regulation of these Cu transporters. As a matter of fact, it is of great significance to study whether NRF2 and SREBP1 can directly target Cu-transport-related genes to regulate Cu homeostasis.
In the present study, we used yellow catfish Pelteobagrus fulvidraco, a common freshwater teleost in China and other Asian countries, as the experimental fish because its genome information is available openly, and because yellow catfish had physiological processes similar to other vertebrates [20]. Recently, we have isolated and characterized the ctr1, ctr2 and atox1 full-length cDNA sequences in yellow catfish [5]. To further investigate their functions and the regulatory mechanism of Cu transporters, this study characterized ctr1, ctr2 and atox1 promoters in yellow catfish, and demonstrated the transcriptionally regulatory mechanism of these three Cu-transport-related proteins in response to Cu, and determined their intracellular localization.

Sequence Analysis of the ctr1, ctr2 and atox1 Promoters
We cloned the 1359 bp ctr1 promoter ( Figure S1), the 1842 bp ctr2 promoter ( Figure S2) and the 1825 bp atox1 promoter ( Figure S3) in yellow catfish. We also identified the transcription start sites (TSS) of ctr1, ctr2 and atox1 promoters. The first nucleotide of each TSS was designated as +1. Some core promoter elements were predicted, including TATA-box (

Analysis of the 5 -Sequence Deletion of the ctr1, ctr2 and atox1 Promoters
We constructed the 5 -sequence deletion plasmids and transfected them into the HEK293T cells to identify the core regions of three promoters ( Figure 1). For the ctr1 promoter, the sequence deletion from −796 bp to −1039 bp significantly reduced the relative luciferase activity, but the sequence deletion from −1039 bp to −1359 bp significantly increased the relative luciferase activity ( Figure 1A). For the ctr2 promoter, the sequence deletion from −522 bp to −1019 bp and from −1557 bp to −1842 bp significantly reduced the luciferase activities ( Figure 1B). For the atox1 promoter, the luciferase activity of the sequence deletion from −1592 bp to −1825 bp was significantly reduced ( Figure 1C).

Analysis of the 5′-Sequence Deletion of the ctr1, ctr2 and atox1 Promoters
We constructed the 5′-sequence deletion plasmids and transfected them into the HEK293T cells to identify the core regions of three promoters ( Figure 1). For the ctr1 promoter, the sequence deletion from −796 bp to −1039 bp significantly reduced the relative luciferase activity, but the sequence deletion from −1039 bp to −1359 bp significantly increased the relative luciferase activity ( Figure 1A). For the ctr2 promoter, the sequence deletion from −522 bp to −1019 bp and from −1557 bp to −1842 bp significantly reduced the luciferase activities ( Figure 1B). For the atox1 promoter, the luciferase activity of the sequence deletion from −1592 bp to −1825 bp was significantly reduced ( Figure 1C). Values mean the ratio of activities of Firefly to Renilla luciferase. Results are presented as means ± SEM (n = 3). Hash symbol (#) means significant differences between two groupss (p < 0.05).
We also used 5'-sequence deletion analysis to explore the response of these three promoters to Cu ( Figure 2). For the ctr1 promoter, compared with the control, Cu treatment increased the relative luciferase activity of the −470/+57 bp sequence, indicating the regulatory effect of Cu on the ctr1 promoter. For different sequence deletion plasmids treated Values mean the ratio of activities of Firefly to Renilla luciferase. Results are presented as means ± SEM (n = 3). Hash symbol (#) means significant differences between two groupss (p < 0.05).
We also used 5'-sequence deletion analysis to explore the response of these three promoters to Cu ( Figure 2). For the ctr1 promoter, compared with the control, Cu treatment increased the relative luciferase activity of the −470/+57 bp sequence, indicating the regulatory effect of Cu on the ctr1 promoter. For different sequence deletion plasmids treated with Cu, the luciferase activity of the −1039/+57 bp region was higher than that of −1359/+57 bp, suggesting that the region from −1039 bp to −1359 bp has a negatively regulatory element (Figure 2A). For the ctr2 promoter, compared with the control, Cu treatment reduced the relative luciferase activity of the −522/+43 bp sequence but increased the luciferase activities of −1557/+43 bp and −1842/+43 bp sequences, indicating both positive and negative regulatory elements in response to Cu on the ctr2 promoter ( Figure 2B). For the atox1 promoter, the activity of the sequence deletion from −649 bp to −1088 bp was significantly up-regulated after Cu treatment, indicating the positively regulatory element in this region ( Figure 2C). with Cu, the luciferase activity of the −1039/+57 bp region was higher than that −1359/+57 bp, suggesting that the region from −1039 bp to −1359 bp has a negatively re ulatory element ( Figure 2A). For the ctr2 promoter, compared with the control, Cu trea ment reduced the relative luciferase activity of the −522/+43 bp sequence but increased t luciferase activities of −1557/+43 bp and −1842/+43 bp sequences, indicating both positi and negative regulatory elements in response to Cu on the ctr2 promoter ( Figure 2B). F the atox1 promoter, the activity of the sequence deletion from −649 bp to −1088 bp w significantly up-regulated after Cu treatment, indicating the positively regulatory eleme in this region ( Figure 2C).  . Asterisk (*) indicates significant differences in relative luciferase activities between the Cu-treated group and the control in the same fragment (p < 0.05); hash symbol (#) indicates significant differences in relative luciferase activities between the two fragmentss under the same treatment (p < 0.05).
For the ctr2 promoter, no potential NRF2 binding sites were predicted in the present study. Compared with the control, with the Cu treatment, the mutation of the −630/−639 bp SREBP1 site (Mut-ctr2-SREBP1(1)) and the −1058/−1067 bp SREBP1 site (Mut-ctr2-SREBP1(2)) increased the luciferase activities. Therefore, the plasmid (Mut-ctr2-SREBP1(12)) containing these two mutation sites also up-regulated luciferase activity under the Cu treatment, indicating that the SREBP1 binding sites were important for Cu-induced transcriptional regulation of the ctr2 promoter ( Figure 3B).

Analysis of the Functional Binding Sites Based on EMSA
Based on the above site mutation analysis, the −820/−830 bp and −1050/−1059 bp sites for NRF2 binding and the −91/−100 bp site for SREBP1 binding in the ctr1 promoter, as well as the −326/−334 bp and −1232/−1240 bp sites for NRF2 binding and the −232/−241 bp and −699/−708 bp sites for SREBP1 binding in the atox1 promoter, were considered functional. Thus, EMSA was used to further verify these potential binding sites ( Figure 4). For the ctr1 promoter, the 100-fold unlabeled NRF2 binding sequences (−820/−830 bp and −1050/−1059 bp) and the sequences (Mut-ctr1-NRF2(1) and Mut-ctr1-NRF2(2)) mutated from the 100-fold unlabeled NRF2 binding sites did not compete for the binding of the added protein (Lane 3-4), suggesting that NRF2 could not combine with these two regions. Com-pared with the control, Cu treatment did not significantly influence the band brightness, indicating that the NRF2 site of the ctr1 promoter did not interact with Cu ( Figure 4A

The Effects of Dietary Cu Levels on ctr1, ctr2 and atox1 mRNA and Protein Expressions in Yellow Catfish Liver Tissues
In order to study the effects of Cu on ctr1, ctr2 and atox1 mRNA and protein expressions, we designed in vivo experiments ( Figure 5). The original western blot figures are given in File S1. Compared with the AC (adequate Cu) group, Cu excess down-regulated the ctr1 and ctr2 mRNA levels, but did not significantly influence the atox1 mRNA level ( Figure 5A). Cu expression also significantly down-regulated the CTR1, CTR2 and ATOX1 protein expressions ( Figure 5B,C).  (2) and Mut-atox1-NRF2(3)) did not compete for the nuclear protein with the labeled probe (Lane 4). After Cu incubation, the brightness of the bands was weaker than that of the bands in the control, indicating that the −326/−334 bp and −1232/−1240 bp NRF2 binding sites on the atox1 promoter could bind with the nuclear protein, and Cu incubation weakened the binding ( Figure 4C,D).
Similarly, we used the SREBP1 binding sequences as the probes. For the ctr1 promoter, the 100-fold unlabeled SREBP1 binding sequence (−91/−100 bp) competed for the nuclear protein with the labeled probe (Lane 3-4). Compared with the control, the brightness of the last band was significantly higher than the band brightness in the control, indicating that the −91/−100 bp SREBP1 sequence was a functional binding site and that Cu was involved in the transcriptional regulation of the ctr1 promoter ( Figure 4E). For the atox1 promoter, the 100-fold unlabeled −232/−241 bp and −699/−708 bp SREBP1 binding sequences competed for the nuclear protein with the labeled probe (Lane 3), while the 100-fold unlabeled SREBP1 binding sites' mutated sequences (Mut-atox1-SREBP1(1) and Mut-atox1-SREBP1(2)) did not compete for the binding, suggesting that SREBP1 could bind with these regions of the atox1 promoter. Compared with the control, Cu increased the band brightness, suggesting that Cu mediated the transcriptional regulation of atox1 promoter between the −232/−241 bp and −699/−708 bp SREBP1 binding sites ( Figure 4F,G). In order to study the effects of Cu on ctr1, ctr2 and atox1 mRNA and protein expressions, we designed in vivo experiments ( Figure 5). The original western blot figures are given in File S1. Compared with the AC (adequate Cu) group, Cu excess down-regulated the ctr1 and ctr2 mRNA levels, but did not significantly influence the atox1 mRNA level ( Figure 5A). Cu expression also significantly down-regulated the CTR1, CTR2 and ATOX1 protein expressions ( Figure 5B,C).

Subcellular Localization of CTR1, CTR2 and ATOX1 in HEK293T Cells
To determine the intracellular localization of CTR1, CTR2 and ATOX1 in yellow cat-

Subcellular Localization of CTR1, CTR2 and ATOX1 in HEK293T Cells
To determine the intracellular localization of CTR1, CTR2 and ATOX1 in yellow catfish, we transfected fusion CTR1-EGFP, CTR2-EGFP and ATOX1-EGFP plasmids into HEK293T cells ( Figure 6). The original subcellular localization images are given in File S1. The results showed that CTR1 was mainly located in the cell membrane ( Figure 6A), CTR2 in the cell membrane and the lysosome ( Figure 6B). ATOX1 was widely expressed in the cytoplasm rather than in the nucleus ( Figure 6C).

Discussion
CTR1, CTR2 and ATOX1 play important roles in controlling intracellular Cu homeostasis and are regulated by Cu [5,9], but the mechanism remains unclear. Here, we characterized the function and transcriptional regulation of ctr1, ctr2 and atox1 genes in response to Cu in yellow catfish, which provided new insights into their roles in the control of Cu homeostasis.

Discussion
CTR1, CTR2 and ATOX1 play important roles in controlling intracellular Cu homeostasis and are regulated by Cu [5,9], but the mechanism remains unclear. Here, we characterized the function and transcriptional regulation of ctr1, ctr2 and atox1 genes in response to Cu in yellow catfish, which provided new insights into their roles in the control of Cu homeostasis.
The identification of core promoters is the first step in investigating the mechanism of transcription initiation [21]. Here, we predicted core elements on the ctr1, ctr2 and atox1 promoters of yellow catfish, including a GC box (SP1), TATA-box and CCAAT-box (NF-Y) in the ctr1 promoter, similar to other reports [22,23], and a TATA-box on the ctr2 promoter, similar to this report by Beneš et al. [24]. We also predicted two SP1s in the ctr2 and atox1 promoters, respectively, but it was unclear whether SP1 was involved in the control of the ctr2 and atox1 promoters [25,26]. However, we found neither CCAAT-box nor TATA-box in the core region of the atox1 promoter. Similarly, Roy and Singer [27] reported that only about 5-7% of eukaryotic promoters had the TATA-box. We also predicted a cluster of the TFBSs, such as NRF2, SREBP1, MAC1, STAT3, STAT4, KLF4, PPARγ, AP1 and CREB1, in the three gene promoters. Similar TFBSs were also found in other promoters [15,28]. However, it is unclear whether these functional TFBSs exist in the ctr1, ctr2 and atox1 promoters of other species.
Since NRF2 binding sites existed on ctr1 and atox1 promoters, and SREBP1 binding sites existed on all three promoters, we explored whether they were functional sites. In the present study, site mutation analysis showed that NRF2 significantly reduced the luciferase activities of ctr1 and atox1 promoters. NRF2 is a transcription factor that potently transduces chemical signals to regulate a range of cytoprotective genes [18]. Our result suggested that NRF2 repressed the expression of downstream ctr1 and atox1 genes. Furthermore, we found that Cu incubation exacerbated the down-regulation of relative luciferase activities by NRF2 on the ctr1 and atox1 promoters, suggesting that NRF2 inhibited the expressions of the downstream ctr1 and atox1 genes more obviously under Cu incubation. Similarly, Zeng et al. [16] found that in the anterior and mid-intestines of large yellow croaker, Cu stress up-regulated the mRNA expressions of nrf2 and down-regulated the mRNA expressions of ctr1, indicating that there might be an antagonistic effect between NRF2 and CTR1 under Cu stress. SREBP1 is an important transcriptional regulator of lipogenesis [29]. Our study indicated that SREBP1 at the −91/−100 bp binding site of the ctr1 promoter and SREBP1 at the −232/−241 bp and −699/−708 bp binding sites of the atox1 promoter could significantly up-regulate the relative luciferase activities; Cu incubation further aggravated the up-regulation of SREBP1, suggesting that SREBP1 could promote the expressions of the downstream ctr1 and atox1 genes. Similarly, Dong et al. [30] found that the protein expressions of SREBP1 were enhanced in obese spontaneously hypertensive (SHROB) rats, and the induction of atox1 mRNA was also higher in the liver of male rats, indicating the potentially synergistic effect between SREBP1 and ATOX1. We also predicted two SREBP1 binding sites (−114/−123 bp and −664/−673 bp) on the ctr1 promoter and two SREBP1 binding sites (−630/−639 bp and −1058/−1067 bp) on the ctr2 promoter, but the relative luciferase activities did not change significantly after these SREBP1 binding sites were mutated, suggesting that these four sites were not functional. Thus, our results suggested that NRF2 and SREBP1 had distinct regulatory roles by targeting the ctr1, ctr2 and atox1 promoters in yellow catfish. Zhong et al. [31] demonstrated that in the liver of yellow catfish, the mRNA levels of srebp1 did not change significantly when compared with the adequate Cu group, while the NRF2 protein expression increased significantly in the Cu excess group. Similarly, Yu et al. [32] also found that juvenile blunt snout bream Megalobrama amblycephala had lower hepatic srebp1 mRNA expressions but higher mRNA levels of nrf2 after feeding 0.04% fenugreek seed extracts (FSE) diets.
In our study, although site mutation analysis indicated that NRF2 caused changes in the luciferase activities of the ctr1 promoter under Cu treatment, EMSA analysis showed that the 100-fold unlabeled NRF2 binding sequences (−820/−830 bp and −1050/−1059 bp) did not compete for the binding of the added protein. Thus, we speculated that these two NRF2 binding sites were also not functional. Battino et al. [33] showed that after NRF2 is translocated to the nucleus, it forms complexes with coactivators and binds to promoter regions (AREs) to activate the expressions of target genes. Therefore, we thought that NRF2 regulated the ctr1 promoter by combining with other coactivators to form a complex. In addition, our study indicated that the −326/−334 bp and −1230/−1240 bp NRF2 binding sequences of atox1 promoter competed for the binding, suggesting that the positions from −326 bp to −334 bp and −1230 bp to −1240 bp were functional binding sites and Cu inhibited the binding of nuclear protein to the sites. At present, no research proved the direct regulatory effects of NRF2 on the ctr1 and atox1 promoters. We also found that the −91/−100 bp SREBP1 binding sequence of ctr1 promoter and −232/−241 bp and −699/−708 bp SREBP1 binding sequences of atox1 promoter competed for binding, indicating that these three sites were functional sites, and Cu facilitated the binding of nuclear protein to these sites. Pan et al. [14] showed that Cu significantly increased the gene expression of srebp1 in the hepatocytes of zebrafish. Huang et al. [34] pointed out that hepatic srebp1 mRNA levels were significantly increased in Synechogobius hasta when exposed to waterborne Cu for 30 days, whereas srebp1 mRNA levels were significantly decreased after 60 days of exposure. Our research also suggested that NRF2 indirectly mediated transcriptional activity of ctr1 but directly mediated transcriptional activity of atox1, and that SREBP1 directly mediated transcriptional activities of ctr1 and atox1. NRF2 are the member of the basic leucine zipper family [15]. At present, whether NRF2 directly associates with Cu-transport-related proteins to regulate Cu homeostasis remains unknown in fish. SREBP1 is a transcription factor that regulates the expressions of enzymes required for endogenous cholesterol [14]. Furthermore, Cu-induced changes in SREBP1 levels have been reported in several studies. For instance, Tang et al. [35] suggested that Cu deficiency stimulated hepatic lipogenic gene expressions by increasing the hepatic translocation of mature SREBP1. Other studies indicated that the Cu-induced effects on srebp1 mRNA levels were fish species-and tissuedependent [14,17,36]. However, to our knowledge, the present study is the first report on the presence of SREBP1 sites on ctr1, ctr2 and atox1 promoters. Our results indicated that Cu had different regulatory effects on ctr1, ctr2 and atox1 promoters of different lengths through NRF2 and SREBP1 transcription factors, thereby maintaining Cu homeostasis.
Cu-transport-related proteins function in modulating Cu homoeostasis, but the mechanisms by which these proteins regulate Cu levels remain to be further investigated [9]. Therefore, we cultured yellow catfish with two dietary Cu levels to explore the regulation of Cu levels and Cu-transport-related proteins expressions. Our study found that mRNA and protein levels of ctr1 and ctr2 in the liver of yellow catfish were significantly down-regulated in the CE (Cu excess) group, indicating that the expression of ctr1 and ctr2 was inhibited under high dietary Cu levels. Our previous study [5] showed that the ctr1 mRNA levels were significantly down-regulated in the anterior and mid-intestine of yellow catfish when Cu was excessive, while the ctr2 mRNA levels were significantly up-regulated in the anterior intestine but down-regulated in the mid-intestine. Similarly, another study reported that Cu reduced mRNA levels of ctr1 in the gill of guppy Poecilia vivipara [37]. Therefore, these results suggested that the expression levels of ctr1 and ctr2 would be reduced in the presence of Cu excess to maintain Cu homeostasis and to guard against Cu toxicity. Our study also found that dietary Cu levels did not influence the mRNA levels of atox1 in the liver of yellow catfish but Cu excess reduced its protein levels, indicating that the ATOX1 was regulated by Cu at the protein levels. In contrast, Cheng et al. [5] found that the atox1 mRNA levels were significantly down-regulated in the anterior intestine and mid-intestine of yellow catfish under Cu excess. Zhao et al. [38] found that during Cu 2+ exposure, red swamp crayfish inhibited intracellular Cu transport by suppressing the expression levels of ctr1 and atox1 in the hepatopancreas. The results showed that ATOX1 acts as a Cu chaperone, and its expression was tightly regulated by intracellular Cu concentration, thereby preventing excessive Cu-induced high cytotoxicity and maintaining intracellular Cu homeostasis. Cu homeostasis is ensured by the activities of metal transporters and intracellular chaperones [9,39]. Loss of ATOX1 has been reported to result in impaired intracellular Cu efflux, significantly increased intracellular Cu content, and abnormal intracellular Cu distribution [40,41]. However, the mechanism by which Cu transporters in fish were affected by Cu concentration was still unclear and needed to be further explored.
In this study, CTR1 was mainly located in the cell membrane. Ohrvik and Thiele [4] suggested that the localization of CTR1 on the cell membrane can form a highly selective ion channel structure to transport extracellular cuprous ions (Cu + ) into the cells. We found that CTR2 was located in the cell membrane and the lysosome. Studies had demonstrated that CTR2 was more localized on the membrane of intracellular organelles, such as vacuole, membrane vesicle, endosome and lysosome, in addition to its localization on the cell membrane [7]. CTR2 located on lysosomes can transport Cu ions from organelles to the cytosols, which may have a role in recycling Cu from intracellular reservoirs. CTR2 localized on the cell membrane plays the same role as CTR1, participating in Cu uptake. Bertinato et al. [42] showed that after over-expression of CTR2 in Cu-deficient COS-7 cells, culture under Cu-rich conditions resulted in excessive accumulation of cellular Cu. Moreover, our study found that ATOX1 was localized in the cytoplasm. Chen et al. [8] also found that ATOX1 was mainly localized in the cytoplasm when it was not functioning. ATOX1 plays a crucial role in cellular Cu homeostasis. ATOX1 captures Cu in the cytosols for subsequent transfer to Cu pumps in the anti-Golgi network, thereby facilitating the supply of Cu to various Cu-dependent oxidoreductases that mature in secretory vesicles [43]. In addition, ATOX1 functioned as a Cu-chaperone for CRIP2, a nuclear copper-dependent autophagy activator [43]. Therefore, these results could reflect the roles of CTR1, CTR2 and ATOX1 in maintaining Cu homeostasis, respectively.

Ethical Statement
The experimental protocols performed in yellow catfish followed the guideline of the Ethics Committee of Huazhong Agricultural University (HZAU) for the use of experimental animals and were approved by the Ethics Committee of HZAU.

Experimental Animals, Cells and Reagents
In order to clone the ctr1, ctr2 and atox1 promoters and characterize their function, yellow catfish were purchased from a local farm (Wuhan, China).

Analysis of the 5 -Sequence Deletion of ctr1, ctr2 and atox1 Promoters
According to the method of Xu et al. [44], we transfected the plasmids into HEK293T cells and measured their relative luciferase activities. Briefly, the HEK293T cells were cultured in the DMEM medium containing 10% (v/v) heat-inactivated FBS (Gibco, CA, USA) in an incubator at 37 • C and 5% CO 2 . Lipofectamine 293 (Beyotime Biotechnology, Shanghai, China) was used as the transfection reagent. Before transfection, the HEK293T cells were cultured in a 24-well plate at 1.2 × 10 5 cells/well. They were cultured for 24 h to 70-80% confluence. Following the manufacturer's protocols and our previous study [45], we co-transfected 400 ng reporter plasmids and 20 ng pRL-TK into HEK293T cells. This transfection reagent did not require medium replacement after transfection. Based on our recent studies [46], two Cu concentrations, namely the control (without extra Cu addition) and Cu-treated group (10 µM Cu), were used in this experiment, and Cu was added in the form of CuSO 4 ·5H 2 O. After the incubation for 24 h, the cells were lysed and collected for analyzing the relative luciferase activities by the Dual-Luciferase Reporter Assay System (Promega, Madison, WI, USA). The relative luciferase activities were obtained via the calculation of the ratio of Firefly to Renilla luciferase activity. We performed each cell transfection in triplicate, and the experiments were repeated in triplicate independently.

Primary Hepatocyte Culture and Treatments
According to Wu et al. [47], we isolated and cultured yellow catfish primary hepatocytes. Based on Chen et al. [46], we designed the control (without extra Cu addition) and Cu-treated group (10 µM Cu). After 48 h incubation, the cells were collected for electrophoretic mobility shift assay (EMSA).

EMSA
The EMSA experiment was performed to identify the functional NRF2 and SREBP1 binding sites of ctr1, ctr2 and atox1 promoters. After the incubation with or without 10 µM Cu for 48 h, the nucleus proteins were extracted from primary hepatocytes of yellow catfish. Then, we used the BCA method to determine the protein concentrations. The Lightshift Chemiluminescent EMSA kit (Beyotime Biotechnology, Shanghai, China) was used in our study. Briefly, we first incubated the oligonucleotide duplexes of NRF2 and SREBP1 with 10 µg nucleus proteins at room temperature for 10 min. The biotin probe was added and incubated at room temperature for another 20 min. Subsequently, the loading buffer was added, and the electrophoresis was performed on a 6.5% native polyacrylamide gel. After the electrophoresis, the bands were transferred to nitrocellulose films. By blocking and washing together with HRP (1:2000), we used the Vilber Fusion FX6 Spectral imaging system (Vilber Lourmat) to visualize the binding bands. We also used 100-fold excess of unlabeled double-stranded oligonucleotides, in combination with or without the NRF2 and SREBP1 mutation, for the competitive analyses. The experimental protocols for feed formulation, animal culture and feeding were based on Zhao et al. [48] and Zhong et al. [31]. Dietary Cu was added in the form of CuSO 4 ·5H 2 O at the levels of 0.008 (adequate Cu, AC) and 0.4 (Cu excess, CE) g/kg diet, respectively. Referring to Tan et al. [49], the Cu content in the AC diet meets the dietary Cu requirements of yellow catfish. 180 juvenile yellow catfish (3.5 ± 0.01 g, mean ± SD) were stocked in six tanks (300 L water volume), with 30 fish per tank. They were fed two full meals a day. The feeding experiment lasted for 10 weeks.
At the end of the feeding experiment, yellow catfish were fasted for 24 h to avoid postprandial effects, euthanized with MS-222 (Sinopharm Chemical Reagent Co., Ltd., AE1052101) solution, and then sampled. Nine fish were randomly selected from each tank, and their liver samples were taken, quickly frozen in liquid nitrogen, and stored at −80 • C for real-time quantitative PCR (qPCR) and Western blot analysis, respectively.

Exp. 3: Intracellular Co-Location Analysis of Ctr1, Ctr2 and Atox1 Immunofluorescence
Based on Pang et al.'s study [51], we constructed the open reading frame of CTR1, CTR2 and ATOX1 with the deletion of stop codons into vector pcDNA3.1-EGFP, respectively. Then, fusion plasmids were transfected into HEK293T cells. After 24 h incubation, the cells were fixed in the 4% paraformaldehyde. Subsequently, Dil (Beyotime Biotechnology, Shanghai, China) was used as a red fluorescent dye for cell membrane staining, Hochest (Beyotime Biotechnology, Shanghai, China) was used as a blue fluorescent dye for cell nucleus staining, and Lyso-Tracker Red (Beyotime Biotechnology, Shanghai, China) was used as a red fluorescent dye for lysosome staining, respectively. We visualized these images using laser confocal microscopy (Leica, Carl Zeiss, Jena, Germany).

Statistical Analysis
We used the SPSS 22.0 software (Armonk, NY, USA) to perform the statistical analysis. All the data were presented as means ± standard errors of means (SEM). The Student's t-test was used to analyze the data between two treatments, and one-factor ANOVA and Duncan's multiple range test were used to analyze the data among ≥ 3 treatments. The p < 0.05 was considered to be statistically significant.

Conclusions
In conclusion, we identified the promoters of the Cu-transport-related genes ctr1, ctr2 and atox1 in yellow catfish, and identified functional NRF2 and SREBP1 binding sites in these three promoters. Cu could regulate the activities of these three promoters through NRF2 and SREBP1. Dietary Cu addition influenced the ctr1, ctr2 and atox1 mRNA and total protein levels, thereby maintaining Cu homeostasis. We also determined the intracellular localization of CTR1, CTR2 and ATOX1. Taken together, our study provided new evidence for the transcriptional response mechanism of ctr1, ctr2 and atox1 promoters to NRF2 and SREBP1 response elements, and provided new ideas for the role of Cu transporter-related proteins CTR1, CTR2 and ATOX1 in controlling Cu homeostasis in vertebrates.

Institutional Review Board Statement:
The protocols for animal experiments followed the institutional ethical guidelines of Huazhong Agricultural University for the care and use of laboratory animals and were approved by the university's ethics committee (I.D. code: Fish-2020-09-24).
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data used to support the findings of this study are available from the corresponding author upon reasonable request.